Volume 7, Issue 2 - Proceedings of the 55th Annual Meeting of the German Society of Biomedical Engineering (5 – 7 October 2021, Hannover)

In Medical Training Therapy (MTT) the exact execution of training exercises developed by the therapist is crucial for the success of the therapy. Currently, a therapist has to treat up to 15 patients at the same time on an outpatient basis. The aim of this paper is to provide an assistance system that can evaluate both the quantity and the quality of the movement performed using a goal-oriented model which gives the patient direct recommendations for action through automated, machine-monitored feedback. To implement the therapeutic measures in a targeted manner, it is advantageous to transfer their knowledge to the patients. This would be possible by transferring the therapist's movement into a patient's movement. Under this assumption a movement sequence of the therapist is recorded and a movement reference frame is extracted with the help of a smart sensor, which is made available to the patient as a target movement. The assistance system records the patient's movement in real time, extracts the its skeleton and compares this movement with the reference. Due to anatomical body differences, the reference skeleton must be adapted to the current patient skeleton, otherwise no movement evaluation can take place. By adjusting the bone lengths and angles in the individual frames of the patient's movement the selected sequence can be compared and any movement deviations that occur can be projected in directly in real time onto the patient's 3D avatar as an error image. The patient tries to reduce the errors and in doing so comes closer and closer to the ideal movement of the therapist which guarantees the best possible therapy success.

Auscultation methods allow the non-invasive diagnosis of pathological conditions (e.g., of the lung, heart or blood vessels) based on sounds that the body produces (e.g., breathing, heartbeat, swallowing or the blood flow). Through regular homebased examinations and Big Data combined with Machine learning techniques like Deep Learning, these could help detect diseases in an early stage, thus preventing serious health conditions and subsequently ensuring optimal therapy through continuous monitoring. This paper presents BODYTUNE, a novel inexpensive multi-auscultation system that aims at providing a tool for establishing a baseline of audio signal derived classification parameters that could be used for the self-monitoring of personal health for everybody through the analysis of deviations from that baseline. In the future, Big Data analysis could additionally lead to prediction and early detection of disease events.

Dry electrodes enable a shorter preparation time for infant EEG. Since infant skin is more sensitive than adult skin, soft electrodes are required to reduce the mechanical stress for this sensitive skin. Thus, soft electrodes are crucial for eventual repetitive and long-term use like in neonatal intensive care units. A biocompatible polyurethane (PU) can be produced in low hardness resulting in a soft and flexible electrode substrate. Silver/silver chloride (Ag/AgCl) electroless plating provides a conductive, electrochemically stable coating but the process may alter the mechanical properties of the electrode substrate. In this study, we assess the hardness of PU material before and after Ag/AgCl plating. The test sample design for Shore hardness measurement is based on ISO 7619-1:2010. Sample production consists of a 3D print master model, silicone molding, PU casting, and finally electroless plating. UPX 8400-1 (Sika AG, Switzerland) is used for the sample substrates. Test samples are produced with 7 different Shore hardness (range A40-A95) and 14 samples (each hardness: 1 uncoated and 1 coated). The hardness measurements are carried out with a lever-operated test stand Shore hardness tester model with a digital hardness tester (TI-AC with HDA 100-1, KERN &SOHN GmbH, Germany).. It is shown that there is a hardness increase (Shore A) due to Ag/AgCl coating with a grand average of 1.1±0.7 (p<0.05). The largest increase of 2.1±0.2 is seen on the initial lowest Shore hardness sample (Shore hardness: 43.4±0.1). The absolute increase of hardness due to the Ag/AgCl coating decreases with increasing substrate hardness. It is concluded that there is no strong hardness increase of PU substrates due to Ag/AgCl plating. Therefore, the material is suitable as a soft electrode for repetitive and long-term use in infant applications.

Heart valve tissue engineering aims at creating living valves through colonization of scaffolds with patient’s own cells. Various cell sources have been explored focusing mainly on endothelialization of the scaffold surface. Endothelial like cells, such as endothelial progenitor cells (EPCs), which can be isolated from peripheral blood or bone marrow could be a suitable option. In this study we investigated cell colonization potential of ovine EPCs (OEPCs) on thermoplastic silicone-based polyurethane (TSPU) polymer scaffolds. TSPU nonwovens with and without vascular endothelial growth factor (VEGF) functionalization were used. SEM images showed that by day 3 the cells were growing as patches on the surface of both polymer groups. The cell patches continued growing and started covering more of the nonwoven surface. On day 7 the cells had almost covered the scaffold surface. The cells were more uniformly distributed as monolayer on the functionalized TSPU compared to the non-functionalized nonwovens. Live/Dead staining provided bright green fluorescence on the samples, indicating metabolically active alive cells. These static cell seeding experiments demonstrated that functionalized TPSU nonwovens support endothelialization. The feasibility of TPSU nonwovens as heart valve prosthesis scaffold could be further explored with animal studies in an ovine model.

Patients with focal ventricular tachycardia are at risk of hemodynamic failure and if no treatment is provided the mortality rate can exceed 30%. Therefore, medical professionals must be adequately trained in the management of these conditions. To achieve the best treatment, the origin of the abnormality should be known, as well as the course of the disease. This study provides an opportunity to visualize various focal ventricular tachycardias using the Offenburg heart rhythm model. Modeling and simulation of focal ventricular tachycardias in the Offenburg heart rhythm model was performed using CST (Computer Simulation Technology) software from Dessault Systèms. A bundle of nerve tissue in different regions in the left and right ventricle was defined as the focus in the already existing heart rhythm model. This ultimately served as the origin of the focal excitation sites. For the simulations, the heart rhythm model was divided into a mesh consisting of 5354516 tetrahedra, which is required to calculate the electric field lines. The simulations in the Offenburg heart rhythm model were able to successfully represent the progression of focal ventricular tachycardia in the heart using measured electrical field lines. The simulation results were realized as an animated sequence of images running in real time at a frame rate of 20 frames per second. By changing the frame rate, these simulations can additionally be produced at different speeds. The Offenburg heart rhythm model allows visualization of focal ventricular arrhythmias using computer simulations. By selecting the frame rate, the speed of the simulation results can be adjusted accordingly to visualize the electric field lines of focal ventricular tachycardias in more detail. The static and dynamic simulation results could be used in the future for teaching and research, including the training of medical professionals.

Localized impedance measurements at the needle tip identifying the present tissue type could aid clinicians in needle procedures. To assess the sensitivity field of a hollow, bipolar needle electrode, a 3D finite element approach using COMSOL Multiphysics was chosen. The simulated bipolar needle electrode consists of two hypodermic needles (17 G and 23 G) with an insulating layer of polytetrafluoroethylene (PTFE) in between. Impedance values were recorded while steadily increasing the insertion depth of the needle electrode in a layered tissue structure of skin (dermis), fat, and blood. Simulation results reveal a highly local sensitivity volume around the needle tip that can be approximated by half a tri-axial ellipsoid with elliptic radii of 0.735 mm, 2.886 mm, and 1.774 mm. A comparison with simulated and measured impedance values shows great correspondence.

A key component of an intensity-based 2D/3D registration is the digitally reconstructed radiograph (DRR) module, which creates 2D projections from pre-operative 3D data, e.g., CT and MRI scans. On average, an intensity-based 2D/3D registration requires ten iterations and the rendering of twelve DRR images per iteration. In a typical DRR implementation, the rendering time is about two seconds, and the registration runtime is four minutes. We present an implementation of the Siddon-Jacobs algorithm that uses a novel pixel-step approach to determine the pixel location of the rendering plane. In addition, we calculate the intensity of each pixel in the rendering plane using a parallel computing approach. The DRR rendering time is reduced to 10ms on average so that the registration runtime can be achieved in an average of 4.8 seconds.

About 50 % of complications during laparoscopy occur when surgical access is created. The Veress needle and proposed technical alternatives do not provide reliable information to support the surgeons in guiding the needle, or the feedback is not clearly perceivable. Based on acoustic emissions, Surgical Audio Guidance (SURAG) proposes a non-invasive and efficient way to enhance the perception of guidance information through acoustic and visual feedback displayed in real-time. This article demonstrates that the developed feedback matches the information about tissue layer crossings provided by force measurements. This indicates that SURAG can provide an effective means to make laparoscopic access more precise and safe, especially in pediatric surgery, where space for placing the needle is minimal.

Reflective Photoplethysmography (PPG) sensors are less obtrusive than transmissive sensors, but they present patient-dependent variations in the so-called “Ratio of Modulation” (R). Thus, the conventionally employed calibration curves for determining peripheral oxygen saturation ( S p O 2 ) may report inaccurate values. In this paper, we study the possibility of overcoming these limitations through Machine Learning (ML). For that, we show the results of applying several algorithms and feature combinations to PPG data from a human hypoxia study. The study was performed on ten healthy subjects. Their target oxygen saturation was reduced in five steps from 98- 100% to 70-77% through an oral mask. Blood Gas Analysis (BGA) was performed five times for each saturation level to measure the arterial oxygen saturation. PPG data were acquired from a reflective pulse oximeter placed in the subjects’ ear canals. PPG signals were pre-processed, and several features in the frequency and temporal domain were calculated. For the ML algorithms’ input, we explored different combinations of the features. We trained and validated the algorithms with the data from seven patients, and we tested them on three. Finally, we performed leaveone- out cross-validation to ensure the universality of the methods. The results show a good agreement of the predictions with the BGA values for Linear Regression, k- Nearest Neighbors, Stochastic Gradient Descent, and Neural Network for all input feature combinations with an average RMSE in the range of 3%. However, the performance of the Linear Regression was not beaten by the Neural Network, even for overfitting with 2000 hidden layers. The combination of R calculated with a Fast-Fourier Transform and ACRMS .red /ACRMS .ir significantly improved the results, reducing the RMSE by 25%. This work demonstrates that a straight-forward Linear Regression is capable of determining S p O 2 with reflective PPG independently of the subject if the ratio ACRMS .red /ACRMS .ir is considered simultaneously with the Ratio of Modulation.

Multispectral imaging devices incorporating up to 256 different spectral channels have recently become available for various healthcare applications, as e.g. laparoscopy, gastroscopy, dermatology or perfusion imaging for wound analysis. Currently, the use of such devices is limited due to very high investment costs and slow capture times. To compensate these shortcomings, single sensors with spectral masking on the pixel level have been proposed. Hence, adequate spectral reconstruction methods are needed. Within this work, two deep convolutional neural networks (DCNN) architectures for spectral image reconstruction from single sensors are compared with each other. Training of the networks is based on a huge collection of different MSI imagestacks, which have been subsampled, simulating 16-channel single sensors with spectral masking. We define a training, validation and test set (‘HITgoC’) resulting in 351 training (631.128 sub-images), 99 validation (163.272 sub-images) and 51 test images. For the application in the field of neurosurgery an additional testing set of 36 image stacks from the Nimbus data collection is used, depicting MSI brain data during open surgery. Two DCNN architectures were compared to bilinear interpolation (BI) and an intensity difference (ID) algorithm. The DCNNs (ResNet-Shinoda) were trained on HITgoC and consist of a preprocessing step using BI or ID and a refinement part using a ResNet structure. Similarity measures used were PSNR, SSIM and MSE between predicted and reference images. We calculated the similarity measures for HitgoC and Nimbus data and determined differences of the mean similarity measure values achieved with the ResNet-ID and baseline algorithms such as BI algorithm and ResNet-Shinoda. The proposed method achieved better results against BI in SSIM (.0644 vs. .0252), PSNR (15.3 dB vs. 9.1 dB) and 1-MSE*100 (.0855 vs. .0273) and compared to ResNet-Shinoda in SSIM (.0103 vs. .0074), PSNR (3.8 dB vs. 3.6 dB) and 1-MSE*100 (.0075 vs. .0047) for HITgoC/Nimbus. In this study, significantly better results for spectral reconstruction in MSI images of open neurosurgery was achieved using a combination of ID-interpolation and ResNet structure compared to standard methods.

The demands on flexible implants for recording of neural signals and electrical stimulating have increased in recent years with regard to their functionality, miniaturization, and spatial resolution. These requirements can be met best by embedding powerful complementary metal oxide semiconductor (CMOS) microchips into thin biocompatible polymer substrates. So-called chip-in-foil systems thus combine mechanical properties of a polymer substrate and performance of CMOS technology. The development of a process for direct transfer of multiple CMOS microchips (edge length <400 μm) simultaneously into thin polyimide (PI) substrates is subject of this study. It allows the use of standard microelectromechanical systems (MEMS) processes for further levelled superficial layer build-up. This is achieved with the help of a silicon carrier wafer equipped with cavities for precise chip placement and a sacrificial layer to facilitate release of the chip-in-foil systems. In a post-processing step all silicon chips are thinned down to 100 μm. With this process a transfer yield of 100 % (n = 34) was achieved for the silicon chips on a die level. Chip rotational error on substrates was determined to be as low as 0.21° ± 0.10°. Die adhesion was examined by shear tests, resulting in shear strength of 58.1 MPa ± 13.7 MPa, which dropped to 15.2 MPa ± 10.5 MPa after accelerated ageing in 60 °C phosphate buffered saline solution (PBS) for 16 days (equivalent to 78 days at 37 °C). This study demonstrated a reliable microchip transfer process with low positioning error into flexible PI substrates with post-processing thinning of the dies. The use of a carrier silicon wafer allowed precise electrical interconnect fabrication with standard MEMS processing techniques and without handling of thin and fragile chips. These results are a prerequisite to meet needs of reliability and structural biocompatibility in implantable flexible bioelectronic devices.

Aortic valve (AV) stenosis is characterized by tissue fibrosis and calcification. Fibrous thickening can result in reduced tissue oxygen supply leading to pathological valvular interstitial cell (VIC) differentiation and calcification. Static 2D VIC cultures and animal models are limited in the ability to reflect human AV calcification. Culturing of VICs in micro-physiological-systems (MPS) in a pulsatile flow and the establishment of a modular AV tissue incubation chamber (TIC) are new approaches to evaluate pathophysiological processes of AV disease. Therefore, a MPS able to adjust hypoxic conditions was applied for VIC culture. A significant increase of mRNA-expression of EGLN1 and HIF1α- regulated LDHA and HIF1α nuclear localisation were proven under hypoxia. AV tissue culture was established within a TIC and viability was monitored by Resazurin-reduction in the incubation medium and visualized by LDH-activity in tissue cryosections. Viability was compared between fluid and static incubated tissues revealing an advantageous effect of the fluidic assay condition. Consecutively, the application of MPS in AV research allows i) the investigation of VIC cultures with efficient oxygen regulation and ii) the culture of porcine or human AV tissues preserving viability and specifically reflecting in vivo parameters. These methods open up new possibilities beyond static 2D culture and facilitate a reduction of animal experiments in AV research.

One promising approach for patients who need parenteral nutrition to live an almost normal life, is to provide extra-hospital therapy. This therapy is supported by an ambulatory care service whose staff connects and disconnects the nutritional fluid day by day. The nutrition therapy takes a long time every day and studies show that this therapy significantly disturbs patients’ daily life. Although home parenteral nutrition is part of the range of services provided by the statutory health insurance (German GKV), it is also expensive in the long run. Therefore, the cost-effectiveness of this form of therapy is constantly reviewed on an individual basis. Aim of this project is to enable a more autonomous life for patients by self-determined in between disconnection and connection of the therapy independently of the ambulatory care service. It has to be taken into account that any interruption of the nutrition process poses health risks to the patient, e.g. if air bubbles or bio contaminants enter the tube system. Flushing with saline solution before and after connecting is therefore an elementary component of the nutrition process. The redesigning of the connecting/disconnecting system and the subsequent development of an automated port flushing system integrated into a mobile feeding pump to realize a safe interruption of the therapy is the overall goal of the project. This paper deals with redesigning the connecting system. After having analysed group discussions, detected impact factors are used as basis of an experimental setup to get the final requirements for redesigning the connector system. The experimental study is done according to the rules of Design of Experimental (DoE). DoE is a statistical method to determine the optimal number of tests by using a complete factorial design. Two repetitions of the tests were carried out to analyse the significance of the impact factors by ANOVA (Analysis of Variance).

Over the years, a constant progress in the development of implantable medical devices (IMD’s) can be observed. On one hand, the advanced implantable electronics enable the implementation of numerous smart functionalities, on the other hand, the variety of electronic components including sensors and a bulky battery severely restrict their degree of miniaturization and reliability. To overcome this limitation, our approach is to realize smart functionalities in leadless and battery-free IMD’s emerging from frugal innovation by exploiting the intrinsic nonlinear properties of the components to be used anyway. The aim of this work is to deepen the understanding of the dynamic behavior of circuit topologies of nonlinear ferroelectric ceramic capacitors and to investigate their potential use for an embedded closed-loop control of the stimulation current. We characterized a selection of 40 commercial ceramic capacitors by measurement and simulation. The degree of nonlinearity resulting from a circuit topology consisting of one, two series and two parallel connected nonlinear capacitors was modeled and evaluated in Mathcad. We present a model with parameterized nonlinear capacitors to simulate the dynamic behavior of an inductively coupled implantable system. The stabilization and amplitude of the stimulation current is controlled by two features. These features are in turn controlled by the circuit topology and the degree of nonlinearity of the capacitors. We found that a high degree of nonlinearity allows the stimulation current to be stabilized within a reasonable range, but it makes the system more prone to instability. However, our model needs to include the dynamic behavior of ferroelectric materials used as dielectric in ceramic capacitors to extend the current investigations and to deepen the understanding of the physics behind the nonlinear properties of ferroelectric capacitors.

Recording of heart rate variability (HRV) is a noninvasive and continuous measurement method that allows investigating the autonomic nervous system (ANS) and its reaction to environmental influences. For a precise measurement of HRV data, a carefully chosen study design and environment is required to minimize secondary influences. One major influence to be avoided is movement. However, in the daily routine and for some scientific questions, movement can often not be avoided. If so, a manual or automated method to differentiate between artifacts caused by body movement and the actual psychophysiological effect is needed to ensure the data quality. In this approach, a chest-worn sensor was developed, that measures the heart rate using a single lead ECG and filters the measured change of the HRV caused by movement. Data from an integrated accelerometer is used to detect upper body movements that affect the resting heart rate. The movementcorresponding time stamps are then used to filter the Interbeat Intervals (IBI) accordingly. Functionality and effectiveness of the sensor system have been tested against state-of-the art sports- or clinical devices in varying scenarios. As our test series showed, motion filtering has a decisive effect when motion occurs, two-thirds of all cases showed a significant effect of motion filtering, with small to medium effect sizes for the parameters SD2, SD2/SD1, and SDNN. Thereby, automatic filtering of motion artifacts can help to significantly reduce the need for costly post-processing of distorted data sets. The results show a better data quality of HRV measurement, a method that is commonly used for the investigation of physiological processes in the field of chronic pain, psychology, psychiatry, or sports medicine.

It is well known that the transition temperatures, e.g. the austenite peak temperature Ap, of NiTi Shape Memory Alloys (SMAs) can be adjusted by changing the alloy composition. This topic recently became more interesting due to the possibilities to produce SMA-parts by additive manufacturing, specifically by Selective Laser Melting (SLM). The potential of new designs and smart structures by so-called 4D-printing with locally adjusted transition temperatures Ap potentially opens up new applications and novel temperature-responsive medical devices. This work focuses on the SLM manufacturing parameters exposure time ET (scanning speed) and laser power P and their impact on the transition temperature Ap beyond the commonly used generic process parameter energy density ED. By systematical variation of process- and scan-parameters, the impact of the P, ET, sample orientation and layer height LH as well as interdependencies between them have been studied. A wide range of transition temperatures Ap between -20°C and 70°C has been reached from one starting material by varying ET. These findings potentially allow the manufacturing of smart devices with multi-stage deformation processes in a single 4Dprinted part.

Targeted stimulation of dorsal nerve roots of the lumbosacral spinal cord (SC) represents a promising way to improve lower extremity functions for people with paraplegia due to SC injury. Stimulation pulses are selectively applied at different segments of the SC by means of an implanted electrode array. For an optimal clinical outcome, correct positioning of the array in rostrocaudal and mediolateral direction of the SC is essential. We developed a neuromonitoring framework to support the neurosurgeon during the implantation procedure by automated stimulation and real-time analysis of evoked muscular responses. For the implementation we developed and combined a new software with several hardware and software modules for intraoperative neuromonitoring. The new software coordinates different interfaces and components including a neurostimulator, a multiplexer module and a biosignal amplifier. In addition, a graphical user interface (GUI) was implemented, allowing control of the system and display of analyzed data. With the framework, stimulation sequences in respect to stimulation amplitudes and selection of contacts of the electrode array can be predefined and executed in an automated manner. Synchronously the evoked electromyographic responses of the target muscles can be recorded, analyzed and displayed together with the stimulation setup for visual evaluation. The framework serves as a fundamental basis for the integration of a clinical knowledge-based algorithm and the realization of a future neuromonitoring system for this purpose.

Intraoperative neurophysiological monitoring (IONM) is an essential tool during numerous surgical interventions to assess and monitor the functional integrity of neural structures at risk. A reliable signal interpretation is of importance to support medical staff by reducing manual evaluation. Deep learning (DL) techniques proved to be a robust tool for the analysis of neurophysiological data. The large amount of required manually labeled data as well as the lack of interpretability of the results however often limit the use of DL in medical scenarios. A possible way to tackle these obstacles is the utilization of Bayesian deep learning (BDL) methods. The modelling of uncertainties in the network parameters and the thereby possible quantification of predictive uncertainties allows both the identification of potential erroneous predictions as well as the targeted selection of informative signals in the context of active learning. To evaluate the applicability of BDL for the analysis of electrophysiological data as well as to increase the training efficiency by active learning, we implemented a multi-task Bayesian Convolutional Neural Network (BCNN) for the simultaneous classification of action potentials and the assessment of relevant signal characteristics (latency, maximum, minimum). We compare the results for electromyographical signals (EMG), containing in total approximately twelve thousand signals from 34 patients, with both a traditional non-Bayesian single-task and multi-task CNN. For all models, including the BCNN, we could achieve similar performances with detection rates over 97% accuracy. Further, we could improve training efficiency of the BCNN using pool-based active learning and therefore significantly reduce the required amount of manual labeling. The evaluated predictive uncertainties of the BCNN prove useful both for the efficient selection of informative signals in the context of active learning as well as the interpretation of the predictive posterior distribution and therefore trustworthiness of the classifications.

Modern industry and multi-discipline projects require highly trained individuals with resilient science and engineering back-grounds. Graduates must be able to agilely apply excellent theoretical knowledge in their subject matter as well as essential practical “hands-on” knowledge of diverse working processes to solve complex problems. To meet these demands, university education follows the concept of Constructive Alignment and thus increasingly adopts the teaching of necessary practical skills to the actual industry requirements and assessment routines. However, a systematic approach to coherently align these three central teaching demands is strangely absent from current university curricula. We demonstrate the feasibility of implementing practical assessments in a regular theory-based examination, thus defining the term “blended assessment”. We assessed a course for natural science and engineering students pursuing a career in biomedical engineering, and evaluated the benefit of blended assessment exams for students and lecturers. Our controlled study assessed the physiological background of electrocardiograms (ECGs), the practical measurement of ECG curves, and their interpretation of basic pathologic alterations. To study on long time effects, students have been assessed on the topic twice with a time lag of 6 months. Our findings suggest a significant improvement in student gain with respect to practical skills and theoretical knowledge. The results of the reassessments support these outcomes. From the lecturers´ point of view, blended assessment complements practical training courses while keeping organizational effort manageable. We consider blended assessment a viable tool for providing an improved student gain, industry-ready education format that should be evaluated and established further to prepare university graduates optimally for their future careers.

With the introduction of novel automated polishing methods, more attention has recently been paid to postprocessing methods of metallic implants. One such method is the polishing process known as DryLyte ® . The most significant difference to previous electropolishing methods is the use of solid organic polymer particles activated with sulfonic acid acting as the electrolyte. The solid particle electrolyte raises new question in terms of polishing results for small features as well as overall polishing quality of metallic surfaces. The aim of this study was to determine the quality of the polishing process for titanium rods with different initial surface roughness and with tapped holes in three different orientations (0°, 45°, 90°) by subjecting them to the DryLyte ® polishing process for 30 min. In addition, the influence of the process parameters voltage and the anodic time T2 during the treatment on the resulting surface quality and the polishing efficiency was determined. In conclusion, the dry electrochemical finishing process has shown great smoothing capabilities for titanium even with small, tapped holes. The Ra values were lowered significantly throughout all titanium samples after 30 min polishing time.

The application of virtual reality (VR) as a supportive tool in psychotherapy has gained great popularity in recent years. Especially for addiction therapy, a combination of virtual exposure and learning or training coping skills by using biofeedback has a high potential to improve conventional therapy. To add value, the new therapy system has to meet the needs of patients and practitioners likewise. Added values consistently named by experts included, but were not limited to, new possibilities of creating individual exposition or coping scenarios, enhanced psychoeducation, a shorter duration of treatment, telemedical aspects, the possibility of measuring and predicting craving and finally an improvement in abstinence. Besides literature research, we evaluated existing technical solutions in the field of virtual addiction treatment, surveyed experts and evolved a concept that led to a first prototype. The prototype consists of a wireless VR headset and a wireless multi-sensor system for measuring the physiological reaction to stimuli or the effectiveness of coping strategies by means of biofeedback. For further studies we developed both, a virtual exposure and a coping scenario and tested the hardware and software in a pilot study in order to elaborate factors that could negatively affect the therapy adherence, the effectiveness of exposition (immersion) and possible hurdles in practical use. Cybersickness and the lack of haptic feedback turned out to be the main limiting factors. Concepts for the next iteration of the therapy system will reflect these points for upcoming clinical studies. In our proof of concept, we demonstrated that virtual therapy can be implemented with a reasonable effort of time and costs. The combination of software and hardware, that supplements the traditional therapeutic approach, lays the foundation for upcoming clinical use and trials to prove the better outcome of VR enhanced addiction therapy.

After the success of virtual reality (VR) applications in supporting exposure-based and psychoeducational cognitive behavioural therapy for various mental illnesses, the technology is also gaining traction in addiction medicine. This paper will describe the concept behind a custom-developed VR simulation for this application with a particular emphasis on alcohol-related addictive disorders. Using Unreal Engine 4, an exposure simulation and a scenario for learning and applying coping strategies were developed for the Oculus Quest. Because of the common association of supermarkets with the procurement of alcohol among addicts, a small shop was implemented as initial scenario. Since the feeling of presence is an important parameter in the quality assessment of a simulation, the degree of immersion was evaluated in initial practical tests after development. In surveys with ten healthy testers, the majority showed a feeling of immersion. Cybersickness and unrealistic collision physics were identified as disruptive factors to the extent of immersion. This work provides the basis for a novel concept for addiction therapy which focusses on a fusion of playful psychoeducation, adjustable exposure, and the internalization of coping strategies with the help of multisensory biofeedback as well as the documentation of the objective physiological measurements for assessment of therapy success. The aim is to best possible address the specific requirements for clinical and home use in terms of self-containment, immersion, and ease of use. To evaluate and exploit the scientific potential of this concept scientific studies and technical improvements to both simulation and hardware are required.

As recently demonstrated, a passive Gough-Stewart platform (a.k.a. hexapod) can be used to create a personalized surgical template to achieve minimally invasive access to the cochlea. The legs of the hexapod are manually adjusted to the desired length, which must be read off an analog scale. Previous experiments have shown that manual length setting of the hexapod’s legs is error-prone because of the imprecise readability of the analog scale. The objective of this study is to determine if integration of a linear encoder and digitally displaying the measured length help reduce the length setting error. Two experiments were conducted where users set the leg length manually. In both experiments, the users were asked to set the leg length to 20 nominal values using the whole setting range from 0 mm to 10 mm. In the first experiment, users had to rely only on the analog scale; in the second experiment, the electronic display additionally showed the user the actual leg length. Results show that the mean length setting error without using the digital display and only relying on the analog scale was (0.036 ± 0.020) mm (max: 0.107 mm) in contrast to (0.001 ± 0.000) mm (max: 0.002 mm) for the experiment with the integrated digital measurement system. The results support integration of digital length measurement systems as a promising tool to increase the accuracy of surgical template fabrication and thereby patients’ safety. Future studies must be conducted to evaluate if integration of a linear encoder in each of the six legs is feasible.

Electric stimulation of the auditory nerve using a cochlear implant (CI) is presumed to be superior when the electrode array (EA) is placed close to the inner wall of the cochlea. Nitinol is investigated as an actuator that enables an intracochlear shape change of the EA from a straight configuration (also necessary for the insertion) to a spiral shape fitting to the inner wall. As shape setting of the thin Nitinol wires is crucial, a method to quantify the accuracy of the shape setting is presented. To measure the trained shape of thin Nitinol wires (ø 100 μm) a contactless, optical method was developed. For each wire, a photomicrograph was captured and processed using a custom Matlab algorithm. Threshold based segmentation followed by morphological operations to remove artefacts were applied to extract the wire’s shape. Utilizing an iterative closest point (ICP) algorithm the actual shape was registered to the desired spiral path. Finally, the root mean squared error describing the deviation between both spirals was calculated as a measure for the “shape error” (εshape). In total 147 Nitinol wires of 16 batches were analyzed to quantify the reliability of the shape setting procedure. The proposed method was successfully applied in all samples. On average εshape was 0.06 ± 0.02 mm. Deviation from the desired shape was < 0.1 mm (< 0.15 mm) in 95% (99%) of the samples. In summary, the presented method is suitable to control the trained shape of thin Nitinol wires. Furthermore, our results confirm a high reliability of the shape setting procedure used for our thin Nitinol actuators intended for future applications in CI EAs.

The electrode array insertion is a critical point during CI surgery and should be performed as gently as possible to preserve residual hearing. In order to measure occurring forces, an insertion tool with an integrated force sensor and an inertial measurement unit (IM U) is being developed. The weight of the electrode holder and the sensor add an unknown offset to the measured forces, depending on the tool orientation. To address this problem, a software which calculates the orientation-induced error and computes a corrected force was developed. The software was written in C++ using the library Qt 5.12.9. For maximization of the computing frequency, the data acquisition of both sensors and the monitoring was parallelized. An algorithm was developed to calculate the error caused by the electrode holder and sensor. For this purpose both weights were determined in a calibration procedure and merged with the provided IM U data. The evaluation was done in two test series (each n=5) with different initial tool orientations. To this end a stepwise 360° rotation around the horizontal axis was performed, while recording the corrected forces. The developed software allows a computing frequency up to 100 Hz with a latency of 10 ms for the online monitoring of the processed data. The evaluation of the corrected force shows a residual error of 0.347 mN ± 0.325 mN for the first and 0.176 mN ± 1.971 mN for the second test series. With the created algorithm, the impact of the extra weight on the sensor can be almost fully equalized. The highly responsive software offers a new possibility to process insertion forces and provide feedback to surgeons. Determining the influence of the tool orientation on the corrected forces is the subject of future researches.

Surgical treatments of arterial occlusive disease with fully absorbable polymeric scaffolds, as a potential alternative to permanent metallic stents, are increasingly penetrating the clinical field. An addition part of the management of patients suffering from vascular diseases is the administration of statins. In this study, absorbable x-ray marked PLLA-based polymer scaffolds and permanent bare-metal stents (BMS) were implanted interventionally into both common carotid arteries (CCA) of 6 healthy female pigs via the left common iliac artery (8F-sheath). The pigs were administered dual antiplatelet drugs oral starting 3 days before the procedure until the end of the study. In Addition, the pigs received atorvastatin orally, beginning 5 days prior to surgery and lasting until the study ended. Stented CCA segments were explanted after 4 weeks, and processed for quantitative histomorphometry, and estimation of vascular inflammation and injury scores. Polymer scaffolds showed a decreased residual lumen area and higher stenosis after 4 weeks (6.41 ± 0.83 mm² and 40.52 ± 5.01%) as compared to the bare-metal reference stent (15.17 ± 0.896 mm² and 7.80 ± 0.88%). After 4 weeks, inflammation score were higher in the polymer group (1.30 ± 0.37) compared to the BMS group (0.42 ± 0.18). In contrast, the BMS showed a slightly elevated vascular injury score (0.85 ± 0.12), as compared to the polymer (0.60 ± 0.23) group. In this preclinical model, the new absorbable polymeric scaffolds showed similar technical feasibility and safety for vascular application as the permanent metal stents. Although no positive trends were observed with oral treatment with atorvastatin, further optimization with a dual-loaded coating is still reasonable. In addition, reduced strut thickness of the polymer scaffolds would have potential to positively impact tissue ingrowth between struts and should be considered in future work on stent design.

Cochlear implants include an electrode array (EA) which needs to be inserted into the cochlea. Insertion tests using artificial cochlear models (ACM) or ex vivo specimens are widely used methods during EA development to characterize EA design properties, including insertion forces. Measured forces are directly linked to the orientation of the cochlear lumen with respect to the insertion axis of the test bench. While desired insertion directions in ACM experiments can be predefined by design, specimens are individually shaped and the cochlear lumen is embedded invisibly. Therefore, a new method for accurate, individual specimen positioning is required. A key element of the proposed method is a customizable pose setting adapter (PSA) used to adjust the specimen’s fine positioning. After rigid fixation of the specimen to a holder featuring spherical registration markers and subsequent cone beam computed tomography the desired insertion direction is planned. The planned data is used to calculate the individual shape of the PSA. Finally, the PSA is 3D printed and mounted between force sensor and specimen holder to correctly align the specimen to the test bench’s insertion axis. All necessary hard- and software have been developed including the specimen holder, a software for registration and trajectory planning, and a custom Matlab script whose output drives a parametric CAD file of the PSA. Positioning accuracy was determined in a first trial using 10 virtual trajectories and was found to be 0.23 ± 0.12 mm and 0.38 ± 0.17°. The presented stereotactic positioning procedure enables high repeatability in future ex vivo insertion experiments due to accurate, image-guided control of the insertion direction.

Atraumatic insertions of electrode arrays (EA) into the cochlea aim to preserve natural structures and residual hearing. However, there is a limit as to how smooth and slow a surgeon can insert an EA. As a potential solution, we recently presented a tool (cochlea hydro drive, CHD) that makes use of an infusion pump to prompt and control the desired, continuous and very slow (< 1 mm/s) forward movement for such insertions. The present work further describes the onset, delay and cessation of the hydraulic actuation in response to different start and stop mechanisms, to better understand the safety of its application for cochlear implant surgery. Methods: Our previously designed tool was used to perform insertions of an EA into an artificial scala tympani model. The prototype is designed to hold an EA, which is then actuated by a standard infusion pump programmed to operate at 0.4 mm/ and 0.1 mm/s. A tubing system between the CHD and the pump includes a three-way valve. Ten insertions were operated using the functions of the pump and ten using the valve. Results: From the programmed start to the actual movement, we observed a larger average delay using the pump’s start function (5 s at 0.4 mm/s; 17 s at 0.1 mm/s) vs. opening the valve (< 0.7 s for both velocities). Moreover, the average cessation of movement with the valve closure was almost immediate (0.7 s for both velocities; this corresponds to < 0.1 mm with the slower tested velocity), as opposed to 60- 80 s delay when using the pump’s stop function. Conclusion: The use of a 3-way valve facilitates motion cessation to the high accuracy level required for cochlear implant surgery. These promising findings support future clinical translation of our tool.

Introduction: Recent investigations focused on the optimization of atraumatic cochlear implant surgery have highlighted the relevance of the electrode array (EA) insertion trajectory. This is particularly studied in the context of minimally-invasive “keyhole” and robotic-assisted approaches, e.g. to avoid injuring structures inside and outside the cochlea. However, little is known about the natural, manual movements and trajectory followed during the insertion process. The present work illustrates the orientation changes within the trajectory a surgeon follows during insertions of EAs into a human cadaveric cochlea. Methods: An EA insertion tool equipped with a gyroscope was developed in our laboratory. During the insertion trials, the gyroscope captures the tool’s spatial orientation. A human head specimen and a single EA were used to perform insertions into a cochlea. A cochlear implant surgeon performed all insertion trials. The recorded orientations were compared to the initial orientation upon cochlea entry to assess the surgeon’s range of motion by calculating the angle between orientation vectors. Results: Fifteen EA insertions were performed with a median maximal deviation from the initial orientation of 7.2° (5.3 -11.1°) across trials. The largest orientation changes were seen towards the last half of each insertion trial. A negative relationship between degree of axis change and number of insertion trial was observed (r = -0.5). Conclusion: Manual EA insertions into a cadaveric cochlea revealed an insertion trajectory with maximum orientation changes of approximately < 10° degrees. The observed trend on decreasing range of motion with increasing number of insertion trials may be attributed to surgeon’s familiarization with the insertion trajectory for this specific specimen but other contributing factors (e.g. EA softening) need to be further elucidated with several EAs. Future evaluations can help determine if this orientation change is influenced by surgeon expertise.

The locomotion of subviral particles of Marburg virus has been shown to be primarily actin based. For this work, a virologist selected 14 subviral particles that show actin tails in fluorescence image sequences. Using the tracked coordinates, examination areas around these subviral particles are defined. The brightness of within the examination area behind the subviral particle is analysed. In addition, the speed of the particle in each frame is calculated to investigate potential correlations between actin activity and particle speed. The results show that actin tracks can be automatically detected and analysed. First hints of a correlation between subviral particle movement and actin activity could be gathered with the presented actin tail quantifier.

The classification of subviral particle motion in fluorescence microscopy video sequences is relevant to drug development. This work introduces a method for estimating parameters for support vector machines (SVMs) with radial basis function (RBF) kernels using grid search with leave-pout cross-validation for classification of subviral particle motion patterns. RBF-SVM was trained and tested with a large number of combinations of expert-evaluated training and test data sets for different RBF-SVM parameters using grid search. For each subtest, the mean and standard deviation of the accuracy of the RBF-SVM were calculated. The RBF-SVM parameters are selected according to the optimal accuracy. For the optimal parameters, the accuracy is 89% +- 13% for N = 100. Using the introduced computer intensive machine learning parameter adjustment method, an RBF-SVM has been successfully trained to classify the motion patterns of subviral particles into chaotic, moderate and linear movements.

Sparse signal modeling often reconstructs a signal with few atoms from a pre-defined dictionary. Hence the choice of wavelet dictionary that represents the sparsity of the target signal is crucial in sparse modeling approach. The challenge of finding an optimal dictionary of different wavelet types using sparse denoising model (SDM) to denoise ECG signal is investigated in this work. A method of finding an optimal wavelet dictionary from a set of orthogonal wavelet sub-dictionaries by the means of the best correlation with ECG signal, is developed. The highly correlated sub-dictionaries from three wavelet dictionaries, namely daubechies, symlets, coiflets and discrete cosine transform are combined to construct an overcomplete dictionary. The weight of Akaike’s information criterion and the signal-to-noise ratio improvement are considered as a criterion to evaluate the performance of the proposed SDM. The results indicate that multi-wavelet dictionary of different types is highly sparse and efficient in denoising the target signal, e.g., ECG.

The paper introduces a method for the automated generation of patient-specific instruments (PSI), here in particular templates, for the implantation of customized implants. The basis is the derivation of data from the morphology of the bony situation and the medical planning. A developed methodological approach based on an Active Shape Model (ASM) is used for the morphological measurement. Determined geometric dimensions are placed on this ASM and automatically adjusted in each case. In addition, specially developed software tools for the planning and design of medical devices will be presented. This includes, among other things, the intuitive control of template parameters by the user when manual adjustments are necessary. The determined data is bundled and applied to previously methodically thought-out and categorized master CAD (Computer Aided Design) models of surgical templates. These master models are fully configurable and designed to be adjusted within defined ranges of values. The templates are printed from the biocompatible material PA12 using selective laser sintering (SLS).

In palliative care, it is important to inform relatives or caring persons about the condition of the affected person, this could also include vital or biomedical data. Nonprofessional caregivers need information around the topic. However, the data should be stored in a backend system and be able to be viewed and edited by multiple caregivers on multiple platforms. Using the example of the backend solution described here, we will show how a FHIR-compliant server for health data, which could be provided from a digital health App (DiGA), can be established, as well as a second server for article database as a "CRUD"-API and user identification.

Modeling the 'digital twin' of a patient's heart has gained traction in the last years and helps to understand the pathogenic mechanisms of cardiovascular disease to pave the way for personalized therapies. Although a 3D patient-specific model (PSM) can be obtained from computed tomography (CT) or magnetic resonance imaging (MRI), the fiber orientation of cardiac muscle, which significantly affects the electrophysiological and mechanical characteristics of the heart, can hardly be obtained in vivo. Several approaches have been suggested to solve this problem. However, most of them require a considerable amount of human interaction, which is both time-consuming and a potential source of error. In this work, a highly automated pipeline based on a Laplace- Dirichlet-rule-based method (LDRBM) for annotating fibers and anatomical regions in both atria is introduced. The calculated fiber arrangement was regionally compared with anatomical observations from literature and faithfully reproduced clinical and experimental data.

We have witnessed impressive advances in preoperative imaging of cancer and the development of dualmodality scanners. However, there is a need for a scanner with functional and anatomical imaging capability suitable for surgical settings and radioguided surgery. The current paper introduces a handheld gamma-ultrasound scanner prototype and illustrates the initial result of testing its very first version. The result of the testing was promising and encouraging in continuing the further development of the prototype.

The photoplethysmography optically measures blood volume changes within micro vascular tissue. Furthermore, photoplethysmographic signals are used within pulse oximeters in order to calculate the oxygen saturation of the blood. This standard measurement technique is performed as a non-invasive spot check method for human health conditions in hospitals or other health care facilities. Usually at least two light sources are used alternating in order to measure photoplethysmograms at different wavelengths. In this paper we will investigate different methods of optically recording photoplethysmographic signals.

Inter-individual differences in the feature distribution of electroencephalograms (EEG) during microsleep (MS) raise questions about the generalizability of the methodology, because methodology should have the same validity for subjects to be included in the analysis in the future as for subjects included so far. We address this question using leave-one-subject- out cross-validation (LOSO CV) to simulate inclusion of test data from future subjects. Investigations are based on EEG of 70 subjects across four studies conducted in our driving simulation lab. 9,297 MS and 10,264 counter-examples of sustained attention (SA) were included in the analysis. Three variants of sample balancing are compared: (1) none, (2) overall balancing, and (3) between-subjects balancing. Resulting classification accuracies range from 80% to 100% for 67 subjects, but for 3 subjects they are at 66.5%, 73.1%, 78.3%. Averaged across subjects, accuracies were (1) 90.2%, (2) 90.2% and (3) 89.1%. Thus, balancing might not be essential if LOSO CV is performed over a sufficiently large number of subjects. Results of standard CV must be regarded as optimistically biased.

The electrocardiogram (ECG) is a well-known technique used to diagnose cardiac diseases. To acquire the spatial signal characteristics from the thorax, multiple electrodes are commonly used. Displacements of electrodes affect the signal morphologies and can lead to incorrect diagnoses. For quantitative analysis of these effects we propose the usage of a numerical computer simulation. In order to create a realistic representation of the human thorax including the heart and lung a three-dimensional model with a simplified geometry is used. The electrical excitation of the heart is modelled on a cellular level via the bidomain approach. To numerically solve the differential equations, describing the signal propagation within the body, we use the finite element method in COMSOL Multiphysics ® . The spatial gradients of the simulated body potentials are calculated to determine placement sensitivity maps. The simulated results show that the sensitivity is different for each considered point in time of each ECG wave. In general, the impact of displacement is increased as an electrode is more closely located to the signal source. However, in some specific regions associated with differential ECG leads the placement sensitivity distribution deviates from this simple circular pattern. The results provide useful information to enhance the understanding of placed specific effects on classical ECG features. By additional consideration of patient-specific characteristics in the future, the used model has the ability to investigate additional body-related aspects such as geometrical body shape or composition of various tissue types.

Health longevity, cost reduction, prevention-based healthcare, personalized medicine, predictive diagnostic, transformed care delivery, de-aging, healthy lifestyle trends, and implementation of intelligent technologies should lead to a more democratized (availability for everyone on this planet) healthcare provision. Forecasting the future of healthcare and health policy certainly is imperfect but nevertheless a precious scientific approach that can lead to novel innovative approaches. In the last decade, the healthcare system faced several challenges, including the significant increase of care costs, demographic changes towards the older population, inability to prevent and overcome pandemics, increased chronic and non-communicable diseases, or resistance to adopting emerging technologies. The need to shift the focus from sickness to health becomes a critical mission. We developed a lecture titled "Healthtech Innovation Design" for graduate students from medicine, biomedical engineering, neuroscience and software engineering. The lecture's goal is to teach innovation methodologies, exponential technologies and methods to achieve healthcare democratization. A critical component is to promote initiatives with global teams focused on the future of health. The educational and initiative programs were impactful in growing interest toward innovation, focusing on disruption and healthcare democratization. Participants awareness towards the main issues and challenges was raised. Interdisciplinary participation was qualitatively processed to generate a holistic vision toward innovation. Through embracing digitalization to a patient-centric approach, affordable care services, and the expansion of precision medicine, the entire healthcare organization and management will likely undergo a worldwide change. Notably, digital technologies, the leverage of artificial intelligence and empathy would satisfy unmet clinical needs. With a future-oriented statement, the forecast of healthcare becomes more imaginable, in which democratization will allow the affordability of services in different countries and economic status.

EMG is an established method to acquire the action potentials of contracted muscles. Although commercial EMG systems are available and it is one of the most researched biosignals, it has never become widely used in rehabilitation or fitness training monitoring. The reasons are technical challenges of wearable EMG systems regarding electrode placement, motion artefacts and the complex connectivity of multi-channel EMG measurements. We address this problem for the lower back and abdominal musculature, through a novel dry electrodes belt, multi-channel high density EMG circuitry and problem-specific signal processing. The subject can easily strap the dry electrodes belt around himself which provides 16 EMG channels. Interferences from the ECG and motion artefacts are reduced by a stationary wavelet decomposition. Afterwards, an inter-channel filter is applied to increase the robustness of the signals. Subject measurements during different kinds of typical abdominal and lower back training exercises were performed wearing the novel dry electrodes belt. The results show the possibility of robust EMG measurements from the lower back and abdominal muscles by utilizing the gathered redundancy, appropriately. The additional information obtained via the multi-channel EMG circuitry and spatial oversampling can be used to address current problems of EMG applications. It combines the advantages of robustness and the capability of using comfortable dry electrodes. Therefore, the proposed measurement method for acquiring spatial information about the muscle contractions from the lower trunk can be used for rehabilitation or fitness training monitoring.

A promising approach to drug delivery applications for chemotherapeutics is the use of drug carriers to reduce the total amount of cytostatics, minimizing side effects. In addition, the carriers, loaded with the drug, can be guided to the tumorous tissue via the vascular system, which enables a local drug release (LDR). In our case, LDR is activated due to the sonosensitive behavior of the nanocapsules by inertial cavitation (IC) caused by focused ultrasound (FUS). Thereby, IC is excited by employing sound pressures within the recommended limit allowed for diagnostic ultrasound. In order to verify this drug delivery approach for its clinical suitability, a tissue-mimicking flow -through phantom, containing a small vessel, is used. Investigations have shown that the drug releasing cavitation effect associated with the sonosensitive and biocompatible nanocapsules also occurs in fine vessel structures, even in the case of moving particles and vessel diameters d c smaller than the wavelength λ.

Ocular current stimulation (CS) exhibits potential for the treatment of neurodegenerative ocular diseases. For a full field electroretinogram (ffERG) we found no CS effect on the characteristic waves (a-wave, b’-wave, and b-wave). To investigate whether the orientation of the generated electric field has an influence on the CS effect, this study repeated the previous ffERG study with changing one CS electrode position and compared the results of both studies. In the first study 15 (8 m, 27.5 ± 4.5 years) and in the second study 17 (7 m, 22.0 ± 1.9 years) healthy volunteers were stimulated with an anodal, cathodal, and sham direct CS of 800 μA for 5 min in three sessions (randomized, different days). For both studies, a cut-sized ring rubber electrode was placed around the eye. A square rubber electrode was placed for the first study at the ipsilateral temple and for the second study at the visual cortex. Before (ERG 1) and during (ERG 2) the CS, the ffERG was measured. For both studies, the difference between the ERG 1 and ERG 2 measurement (CS effect) for the three characteristic waves was analyzed and compared between the studies. For statistical analysis, the Mann-Whitney U test with Bonferroni correction was applied (α = 0.05). The comparison of the data distribution showed only slight differences between the studies. The Mann-Whitney U test found no significant difference of the CS effect between the studies for all amplitudes and CS groups (p ≥ 0.0055). In the mean, the latency differences were smaller than the time resolution, therefore no statements for latency effects were possible. It can be concluded that the retinal cells generating the ffERG are not affected by ocular CS in either electrode montages tested.

Newly developed materials for blood-contacting devices need to undergo hemocompatibility testing to prove compliance with clinical requirements. However, many current in vitro models disregard the influence of flow conditions and blood exchange as it occurs in vivo. Here, we present a flow model which allows testing of blood-surface interactions under more physiological conditions. This modular platform consists of a triple-pump-chip and a microchannel-chip with a customizable surface. Flow conditions can be adjusted individually within the physiological range. A performance test with whole blood confirmed the hemocompatibility of our modular platform. Hemolysis was negligible, inflammation and hemostasis parameters were comparable to those detected in a previously established quasi-static whole blood screening chamber. The steady supply of fresh blood avoids secondary effects by nonphysiological accumulation of activation products. Experiments with three subsequently tested biomaterials showed results similar to literature and our own experience. The reported results suggest that our developed flow model allows the evaluation of blood-contacting materials under physiological flow conditions. By adjusting the occurring wall shear stress, the model can be adapted for selected test conditions.

The detailed characterization of complex forms of atrial flutter relies on the correct interpretation of intra-atrial electrograms. For this, the near field components, which represent the local electrical activity, are decisive. However, far field components arising from distant electrical sources in the atria can obscure the diagnosis. We developed a method to separate and characterize atrial near field and atrial far field components from bipolar intra-atrial electrograms. First, a set of bipolar electrograms was created by simulating different propagation scenarios representing common clinical depolarization patterns. Second, near and far fields were detected as active segments using a non-linear energy operator-based approach. Third, the maximum slope and the spectral power were extracted as features for all active segments. Active segments were grouped accounting for both the timing and the location of their occurrence. In a last step, the active segments were classified in near and far fields by comparing their feature values to a threshold. All active segments were detected correctly. On average, near fields showed 15.1x larger maximum slopes and 40.4x larger spectral powers above 100 Hz than far fields. For 135 active segments detected in 72 bipolar electrograms, 5.2% and 6.7% were misclassified using the maximum slope and the spectral power, respectively. All active segments were classified correctly if only one near field segment was assumed to occur per electrogram. The separation of atrial near and atrial far fields was successfully developed and applied to in silico electrograms. These investigations provide a promising basis for a future clinical study to ultimately facilitate the precise clinical diagnosis of atrial flutter.

In order to achieve the high quality required in medical products, reliable characterization methods and quality management systems are necessary. In the field of musculoskeletal Tissue Engineering (mTE), electrospinning is utilized to manufacture fibre scaffolds as implant material. Depending on the application, in this case the regeneration of tendon-bone junctions, properties like the degree of fibre orientation, homogeneity of fibre throughout the scaffold and reaction to external mechanical load are of particular importance. Currently, destructive methods, like scanning electron microscopy (SEM), are widely used to determine these properties. In addition to the destruction of the samples, these methods often only allow the investigation of very small sections. In this study, we present two new methods for the fast, non-destructive and contactless characterization of electrospun fibre scaffolds for mTE. These methods are based on Transillumination Imaging (TI) and Mueller Matrix Polarimetry (MMP), utilizing low-power laser sources or LED light sources, respectively, to determine the relative homogeneity (TI) and the degree of fibre orientation (MMP) in electrospun fibre scaffolds.

In collaboration with the Institute of Virology, Philipps University, Marburg, a deep-learning-based method that recognizes and classifies cell organelles based on the distribution of subviral particles in fluorescence microscopy images of virus-infected cells has been further developed. In this work a method to recognize cell organelles by means of partial image information is extended. The focus is on investigating loss of accuracy by only providing information about subviral particles and not all cell organelles to an adopted Mask-R convolutional neural network. Our results show that the subviral particle distribution holds information about the cell morphology, thus making it possible to use it for cell organelle-labelling.

Autologous plasma proteins can be used to fabricate patient specific cardiovascular implants but need to be cross-linked to increase their mechanical strength and reduce water solubility. Glutaraldehyde is the state-of-the-art solution but its reaction products have been shown to be cytotoxic and pro-inflammatory. In this work, it has been shown, that cross-linking of plasma proteins with biocompatible alternatives to glutaraldehyde is possible. This was achieved by identifying four candidate substances (thrombin, transglutaminase, genipin, EDC) from current literature and investigating their ability to cross-link porcine plasma proteins in vitro. The degree of crosslinking was examined using calorimetric (DSC) and spectroscopic (FTIR, Raman) methods, mapping the influence of cross-linking on the denaturation temperature and primary amino-group content of the proteins. It could be shown that thrombin, genipin and EDC are able to cross-link plasma proteins to a satisfactory degree and thus represent useful alternatives to glutaraldehyde. Transglutaminase, on the other hand, could not sufficiently cross-link the plasma proteins and was therefore ruled out as an alternative.

This contribution is part of a project concerning the creation of an artificial dataset comprising 3D head scans of craniosynostosis patients for a deep-learning-based classification. To conform to real data, both head and neck are required in the 3D scans. However, during patient recording, the neck is often covered by medical staff. Simply pasting an arbitrary neck leaves large gaps in the 3D mesh. We therefore use a publicly available statistical shape model (SSM) for neck reconstruction. However, most SSMs of the head are constructed using healthy subjects, so the full head reconstruction loses the craniosynostosis-specific head shape. We propose a method to recover the neck while keeping the pathological head shape intact. We propose a Laplace- Beltrami-based refinement step to deform the posterior mean shape of the full head model towards the pathological head. The artificial neck is created using the publicly available Liverpool-York-Model. We apply our method to construct artificial necks for head scans of 50 scaphocephaly patients. Our method reduces mean vertex correspondence error by approximately 1.3 mm compared to the ordinary posterior mean shape, preserves the pathological head shape, and creates a continuous transition between neck and head. The presented method showed good results for reconstructing a plausible neck to craniosynostosis patients. Easily generalized it might also be applicable to other pathological shapes.

Optical Coherence Tomography Angiography (OCTA) is a non-invasive imaging technique that enables the visualization of perfused vasculature in vivo. In ophthalmology, it allows the physician to monitor diseases affecting the vascular networks of the retina such as agerelated macular degeneration or diabetic retinopathy. Due to the complexity of the vasculature in the retina, it is of interest to automatically extract vascular parameters which describe the condition of the vessels. Suitable parameters could improve the diagnosis and the treatment during the course of therapy. We present an automated algorithm to compute the diameters of the vessels in en face OCTA images. After segmenting the images, the vessel centerline was computed using a thinning algorithm. The centerline was refined by detecting invalid pixels such as spurs and by continuing the centerline until the ends of the vessels. Lastly, the diameter was computed by dilating a discrete circle at the position of the centerline or by measuring the distance between both borders of the vessels. The developed algorithms were applied to in vivo images of human eyes. Certainly, no ground truth was available. Hence, a plausibility check was performed by comparing the measured diameters of two different layers of the retina (Superficial Vascular Complex (SVC) and Deep Vascular Complex (DVC)). Each layer exhibits a different characteristic vasculature. The algorithm clearly reflected the differences from both retinal layers. The measured diameters demonstrate that the DVC consists of more capillaries and considerably smaller vessels compared to the SVC. The presented method enables automated analysis of the retinal vasculature and forms thereby the basis for monitoring diseases influencing the vasculature of the retina. The validation of the method using an artificial ground truth is still needed.

Mitral regurgitation alters the flow conditions in the left ventricle. To account for quantitative changes and to investigate the behavior of different flow components, a realistic computational model of the whole human heart was employed in this study. While performing fluid dynamics simulations, a scalar transport equation was solved to analyze vortex formation and ventricular wash-out for different regurgitation severities. Additionally, a particle tracking algorithm was implemented to visualize single components of the blood flow. We confirmed a significantly lowered volume of the direct flow component as well as a higher vorticity in the diseased case.

Facial emotion recognition (FER) is a topic that has gained interest over the years for its role in bridging the gap between Human and Machine interactions. This study explores the potential of real time FER modelling, to be integrated in a closed loop system, to help in treatment of children suffering from Autism Spectrum Disorder (ASD). The aim of this study is to show the differences between implementing Traditional machine learning and Deep learning approaches for FER modelling. Two classification approaches were taken, the first approach was based on classic machine learning techniques using Histogram of Oriented Gradients (HOG) for feature extraction, with a k-Nearest Neighbor and a Support Vector Machine model as classifiers. The second approach uses Transfer Learning based on the popular “Alex Net” Neural Network architecture. The performance of the approaches was based on the accuracy of randomly selected validation sets after training on random training sets of the Oulu-CASIA database. The data analyzed shows that traditional machine learning methods are as effective as deep neural net models and are a good compromise between accuracy, extracted features, computational speed and costs.

Current in vitro and in vivo disease models have been reported to lack sufficient translation to human. Precision-Cut Lung Slices (PCLS) are viable sections of lung tissue and have been described to be a translational model for the ex vivo assessment of pharmacological and toxicological compounds. In most studies PCLS were cultured under static conditions. These lung sections, however, suffer from the limited viability. Here we present a novel modular microphysiological system (MPS) to prolong the cultivation of ex vivo lung tissue. A tailored MPS setup was designed using the PDMS free modular plug&play MPS construction kit. PCLS from mice were cultivated for up to one week under static versus perfused conditions. Using the MPS technology enabled a prolonged culture period with improved viability as shown by lowered lactate dehydrogenase release and improved membrane integrity. Using this technology might allow us to use PCLS for longer culture periods such as e.g. repeated dose toxicity or pharmacology studies.

In robotic telemanipulation for minimally-invasive surgery, lack of haptic sensation and non-congruent movement of input device and manipulator are major drawbacks. Input devices based on cable-driven parallel mechanisms have the potential to be a stiff alternative to input devices based on rigid parallel or serial kinematics by offering low inertia and a scalable workspace. In this paper, the haptic user interface of a cable-driven input device and its technical specifications are presented and assessed. The haptic user interface allows to intuitively control the gripping movement of the manipulator’s end effector by providing a two-finger precision grasp. By design, the interface allows to command input angles between 0° and 45°. Furthermore, interaction forces from the manipulator’s end effector can be displayed to the user’s twofinger grasp in a range from 0 N to 6 N with a frequency bandwidth of 17 Hz.

To assist the insertion of a robot-aided endoscope during colonoscopy, a measuring system is required so that the endoscope tip can align automatically and thus find the curved pathway of the large intestine. To achieve this, a selfexpanding nitinol wire basket is used to sense the contour of the intestine. As the wire basket touches the wall, it is deflected towards the center of the intestine. The relative position of the wire basket within the camera image is captured, which describes the desired direction to follow the organ. To identify the wire basket in the image, the original RGB image stream is converted into the HSV (hue, saturation, value) color space. Thus, a binary image can be created, in which only the neongreen color portion of the wire basket is visible as a cross. The Hough Transformation is used to search for straight lines in the binary image. Once two lines are found, the intersection point can be calculated and thus its position in the image. The evaluation of the execution time of the algorithm on a live stream was 45 ± 31 ms on average. The algorithm robustly recognizes the wire basket even if it was not visible to the human eye in the original RGB image due to deficient lighting.

Irreversible Electroporation (IRE) is a non-thermal tumor ablation treatment applicable for prostate cancer. IRE uses ultra-short but strong electrical pulses to destroy cancer cells nonthermally [1]. Clinically available IRE therapy requires two or more needle electrodes placed around the target tissue to apply the electric fields. A pre-requirement to achieve successful and effective ablation is an accurate and parallel needle placement to cover the tumor zone. Differences in tissue density, organ surface curvature as well as organ and patient motion in combination with long and highly flexible needle electrodes causes’ difficulties to achieve the desired target accuracy during needle placement process. We propose a concept of a shooting mechanism in combination with a grid template support to improve the parallel needle placement process for prostate cancer treatment. Instead of conventionally inserting the needle in the body by hand, it can be placed with high speed using a shooting device setup, that works similar like a biopsy gun.

To overcome obstacles and complexity of decision making in clinical oncology, we propose an integrated clinical decision support approach; the Digital Twin. We analyse the reasons for frustration in applying clinical decision support and provide a multi-levelled approach to implementing a flexible system to support and strengthen clinical decisions. Describing medical patterns and contexts with Resource Description Framework (RDF) allows for standardised way of connecting medical knowledge and processing modules. Having flexible web-based interfaces integrated a multitude of heterogeneous data processing systems to either make clinical data available altogether, or provide calculations and assessments. Transition of the Digital Twin to clinical practice promises effective assistance and safer clinical decisions.

Cone-Beam computed tomography (CBCT) has become the most important component of modern radiotherapy for positioning tumor patients directly before treatment. In this work we investigate alternations to standard acquisition protocol, called preset, for patients with a tumor in the thoracic region. The effects of the changed acquisition parameters on the image quality are evaluated using the Catphan Phantom and the image analysis software Smári. The weighted CT dose index (CTDIW) is determined in each case and the effects of the different acquisition protocols on the patient dose are classified accordingly. Additionally, the clinical suitability of alternative presets is tested by investigating correctness of image registration using the CIRS thorax phantom. The results show that a significant dose reduction can be achieved. It can be reduced by 51% for a full rotation by adjusting the gantry speed. A more patientspecific uptake protocol for patients with laterally located tumor was created which allows a dose reduction of 54%.

Today's healthcare challenges with unmet clinical needs, high regulation and certification standards, and increasing costs demand faster innovation and technical translation. To address this challenge, Stanford released a fellowship called Biodesign, where need-based healthcare innovation is taught with the approach identify, invent and implement. Since then, different European institutions have adopted the Biodesign innovation approach and organized within the Biomedical Engineering- Innovation, Design, and Entrepreneurship Alliance (BMEidea EU). The generation of successful healthcare innovation isn't only based on participating in an Innovation teaching program. It is much more a matter of having the right innovation ecosystem with an open creative mindset, experts, the respective stakeholders, and access to essential resources within reach (close to clinic). Through a qualitative survey, seven Biodesign based teaching programs in the EU were examined. The study from an academic perspective contains information covering Resources, Activities, Academic Performance, and Transfer Performance. The demand for new healthcare innovations, and especially innovation training programs that address challenges, developed collaboratively with the respective stakeholders, is increasing. Additionally, there is a growing expectation that innovation needs to reach the market quickly and be implemented accordingly. A Healthcare Innovation Ecosystem, where different entities function as a productive unit with a shared vision and committed to application-driven research and technology transfer, will increase innovation's success and adaptation.

Introduction : The grade of reperfusion after endovascular treatment of ischemic stroke e.g. mechanical thrombectomy is determined based on the mTICI score. The mTICI score shows significant interrater variability; it is usually biased towards better reperfusion results if selfassessed by the operator. We therefore developed a semiautomated image processing technique for assessing and evaluating the degree of reperfusion independently, resulting in a more objective mTICI score. Methods : Fifty angiography datasets of patients who were treated with mechanical thrombectomy for middle cerebral artery (MCA) occlusion were selected from our database. Image datasets were standardized by adjustment of field of view and orientation. Based on pixel intensity features, the internal carotid artery (ICA) curve was detected automatically and used as a starting point for identifying the target downstream territory (TDT) of the MCA on the DSA series. Furthermore, a grid with predefined dimensions was used to divide the TDT into checkzones and be classified as perfused or unperfused. Results: The algorithm detected the TDT and classified each zone of the grid as perfused or unperfused. Lastly, the percentage of the perfused area in the TDT was calculated for each patient and compared to the grading of experienced clinical users. Conclusion : A semi-automatic image-processing workflow was developed to evaluate perfusion rate based on angiographic images. The approach can be used for the objective calculation of the mTICI score. The semi-automatic grading is currently feasible for MCA occlusion but can be extended for other brain territories. The work shows a starting point for a machine learning approach to achieve a fully automated system that can evaluate and give an accurate mTICI score to become a common AI-based grading standard in the coming near future.

Artificial respiration is saving lives especially in the COVID-19 pandemic, but it also carries the risk to cause ventilator-associated pneumonia (VAP). VAP is one of the most common and severe nosocomial infections, often leading to death and adding a major economic burden to the healthcare system. To prevent a proliferation of microbial pathogens that cause VAP, an endotracheal tube (ETT) equipped with blue LEDs (LED-ETT) was developed. This blue wavelength exhibits antimicrobial properties but may also harm human tracheal cells at higher irradiances. Therefore, the aim of this study was to find the minimal required irradiance for microbial reduction of 1 log level in 24 h by applying LED-ETTs. A LED-ETT with 48 blue LEDs (450 nm) was fixed in a glass tube, which served as a trachea model. The investigation was carried out with irradiations of 4.2, 6.6 and 13.4 mW/cm² at 37 °C for 24 h. The experiments were performed with Acinetobacter kookii as a surrogate of Acinetobacter baumannii, which is classified as critical by the WHO. Samples of A. kookii suspensions were taken every 4 h during irradiation from the trachea model. Bacteria concentrations were quantified by determining colony forming units (CFU)/ml. A homogeneous irradiance of only 4.2 mW/cm² generated by the blue LEDs, at a LED forward current of 3.125 mA, is sufficient to achieve a 1 log reduction of A. kookii within 24 h. The total irradiation dose within this period was 360 J/cm2. Human cells survive this dose without cellular damage. Previous studies revealed that the pathogen A. baumannii is even more sensitive to blue light than A. kookii. Therefore, blue LED-ETTs are expected to reduce A. baumannii without harming human tracheal cells.

The incidence of atrial tachycardia steadily increases in industrial nations. During invasive electrophysiological studies, a catheter measures electrograms within the atrium to assist detailed diagnosis and treatment planning. With unipolar and bipolar electrograms, two different acquisition modes are clinically available. Unipolar electrograms have several advantages over bipolar electrograms. However, unipolar electrograms are more affected by noise and the ventricular far field. Therefore, only bipolar electrograms are typically used in clinical settings. A recently published ventricular far field removal technique models the ventricular far field by a set of dipoles and yielded promising results in a simulation study. However, the method lacks quantitative clinical validation. Therefore, we adapted the technique to clinical needs and applied it to data sets of two patients using four different lengths of the removal window. Results were compared quantitatively by a tailored residual error measure. The used method resulted in a median reduction of the ventricular far field by approximately 89% using a removal window of optimal length for both patients. The results showed that the dipole method provides an alternative to other VFF removal techniques in clinical practice because it can reveal AA originally hidden by VFF without leading to a prolongation of the electrophysiological study.

Optical Coherence Tomography Angiography (OCTA) is an imaging modality that provides threedimensional information of the retinal microvasculature and therefore promises early diagnosis and sufficient monitoring in ophthalmology. However, there is considerable variability between experts analysing this data. Measures for quantitative assessment of the vasculature need to be developed and established, such as fractal dimension. Fractal dimension can be used to assess the complexity of vessels and has been shown to be independently associated with neovascularization, a symptom of diseases such as diabetic retinopathy. This investigation assessed the performance of three fractal dimension algorithms: Box Counting Dimension (BCD), Information Dimension (ID), and Differential Box Counting (DBC). Two of those, BCD and ID, rely on previous vessel segmentation. Assessment of the added value or disturbance regarding the segmentation step is a second aim of this study. The investigation was performed on a data set composed of 9 in vivo human eyes. Since there is no ground truth available, the performance of the methods in differentiating the Superficial Vascular Complex (SVC) and Deep Vascular Complex (DVC) layers apart and the consistency of measurements of the same layer at different time-points were tested. The performance parameters were the ICC and the Mann- Whitney U tests. The three applied methods were suitable to tell the different layers apart and showed consistent values applied in the same slab. Within the consistency test, the non-segmentation-based method, DBC, was found to be less accurate, expressed in a lower ICC value, compared to its segmentation-based counterparts. This result is thought to be due to the DBC’s higher sensitivity when compared to the other methods. This higher sensitivity might help detect changes in the microvasculature, like neovascularization, but is also more likely prone to noise and artefacts.

Today a variety of models describe the physiological behavior of the heart on a cellular level. The intracellular calcium concentration plays an important role, since it is the main driver for the active contraction of the heart. Due to different implementations of the calcium dynamics, simulating cardiac electromechanics can lead to severely different behaviors of the active tension when coupling the same tension model with different electrophysiological models. To handle these variations, we present an optimization tool that adapts the parameters of the most recent, human based tension model. The goal is to generate a physiologically valid tension development when coupled to an electrophysiological cellular model independent of the specifics of that model's calcium transient. In this work, we focus on a ventricular cell model. In order to identify the calcium-sensitive parameters, a sensitivity analysis of the tension model was carried out. In a further step, the cell model was adapted to reproduce the sarcomere length-dependent behavior of troponin C. With a maximum relative deviation of 20.3% per defined characteristic of the tension development, satisfactory results could be obtained for isometric twitch tension. Considering the length-dependent troponin handling, physiological behavior could be reproduced. In conclusion, we propose an algorithm to adapt the tension development model to any calcium transient input to achieve a physiologically valid active contraction on a cellular level. As a proof of concept, the algorithm is successfully applied to one of the most recent human ventricular cell models. This is an important step towards fully coupled electromechanical heart models, which are a valuable tool in personalized health care.

To shorten the production time for PDMS-moulds by additive manufacturing (AM) several 3D printers have been investigated in comparison to standard micro-milling by producing benchmark structures. These are evaluated regarding their shape accuracy, the transparency of the casted PDMS which is linked to the surface quality of the mould, and the production time until the moulds are ready to use. Even though the additively manufactured moulds showed significantly better surface quality and shorter production time, the necessary shape accuracy for non-square-structures or structures with < 250 μm edge length could not be achieved due to limiting factors like nozzle diameter, size of the digital micromirror device or spot size of the LCD-panel.

During the last year, student cooperation has shifted to the digital space. Chats, videoconferencing and online collaboration tools replace physical meetings and face-to-face interactions. While many online tools provide sophisticated technologies to share screens and co-edit manuscripts, the support of collaborative thinking, learning and even fooling around is still underdeveloped, but remains a valuable resource. This study traces whether the given level of instruction details influences the collaboration effectiveness among student working groups in a practical course in Biomedical Engineering. Within the semester’s practical course of Introduction into Medicine, participating students were divided into three groups (A, B and C) of each 4 to 6 students and had to collaborate on a clinical case. Groups A received detailed information and rules on online collaboration, B less detailed information and C only basic technical information. Each group was asked to fulfil an additional pre- and post-task before and after the semester’s practical course. These tasks consisted of solving combinatory puzzles based on isolated information bits and hints that were distributed among the group members. The groups differed only in the instructions they received to fulfil the task. During their work phases, all groups were observed. The study was conducted twice with different student cohorts. School grades were assigned representing 1 for the best up to 6 for the worst performance. The following averaged results were found: Groups B received the best grades (average 1,5 for winter term and average 1,6 for summer term). Second best performance was shown by the groups A (average 2,14 for winter term and average 1,75 for summer term) and finally groups C performed worst (average 2.10 for winter term and average 1,85 for summer term). In the taken times to solve the puzzle tasks, groups B showed up to be best in the summer term and groups A in the winter term. In all categories, groups C delivered the worst grades in average. The results of this study support that it is most reasonable to aid student online collaboration by providing detailed instructions or at least a few hints and instruct them to create their own rules for online collaboration. The results further support that the cooperation gets worse without suggesting any rules or instructions and that student age possesses a negligible influence.

Intraoperative blood flow measurement is an effective way to assess the quality of bypass surgery. Flow quantification from indocyanine green (ICG) angiography promises to be an easy, contact-free method. It shows deviations compared to a reference. These are given as factor k, which depends on the vessel diameter d. The radiation transport within the vessel while recording the ICG passage might cause this. It is analyzed in silico to disclose its impact on k (d). A Fluorescence Monte Carlo Multi Cylinder (FMCMC) model was developed as a static model, assuming homogeneous concentration of ICG. In contrast to published approaches utilizing a Monte Carlo Multi Layer (MCML) model assuming the deepest penetration location within a photon packet’s path to be the fluorescence location, the events are modeled. Fluorescence event modeling, Multi Cylinder geometry and a homogeneous illumination as well as combinations of these were implemented in separate aspect models. Resulting k (d) were compared to k (d) from MCML. Deviations in k (d) derived from FMCMC and MCML in each aspect model were present. The Root Mean Square Error ranges from 6,8% to 36 %, k (d) also varied comparing the aspect models to each other. The model geometry, the modeled fluorescence location and illumination mode show a clear impact on simulated k (d). Therefore, our study shows that simplifications of previous studies are invalid. The developed FMCMC model considers the named aspects, allowing the analysis of radiation transport in ICG angiography. The FMCMC model assumes a homogeneous concentration of ICG which is not true in clinical cases. Obtaining the heterogeneous distribution of ICG is possible via fluid flow models. Coupling the fluid flow model and the developed radiation transport model as well as including a detailed camera optic is the task for future work.

This work examines how a touchless interaction concept contributes to an efficient, direct, and sterile interaction workflow during CT-guided interventions. Two hand gesture sets were designed specifically under consideration of the clinical workflow and the hardware capabilities. These were used to change the position of an X-Ray tube and detector of a CT scanner without breaking sterility and are compared regarding usability and performance in a user study with 10 users. The user study revealed that it is possible to change the angle of the gantry within 10 seconds average in an experimental setup. A straight hand gesture showed higher acceptance than a pistol motivated gesture. Furthermore, the sequences were not optimal and confused the users. It turned out that it feels more natural to activate and confirm the system with the same gesture.

Introduction: Clinical trials currently evaluate the use of mesenchymal stem cells (MSCs) for the treatment of non-union bone fractures. The stem cells are injected directly into the non-union area of a bone via a cannula. During this injection process, pressure and shear forces affect the MSCs which could influence the viability of the cells. One parameter that influences the level of the shear forces is the volume flow. The aim of this study is to show whether the injection process with two different volume flows influences the cell viability. Methods: MSCs were isolated from bone tissue, harvested during arthroplasty. Afterwards, they were diluted to a concentration of 1 million cells/mL and 1 mL of this suspension was injected through a cannula with 200 mm length and 2 mm diameter (14 G) with volume flows of 38 and 100 mL/min. The evaluation was performed by detecting living, apoptotic, and dead cells using flow cytometry. The statistical analysis was performed with a Kruskal-Wallis-test to identify significant differences and with a TOST procedure for significant equivalence. The significance level was set to 5 % and the equivalence margin to 20 %. Results: The cell population of healthy cells was in the control group 85.88±2.98 %. After an injection with 38 mL/min the population of healthy cells was 86.04±2.53 % and with 100 ml/min 85.48±1.64 %. The statistical analysis revealed no significant difference between these groups (p = 0.99), but a significant equivalence between the control group and the two volume flows (38 mL/min: p = 0.002, 100 mL/min: p = 0.001). In addition the results show no increase of apoptotic and dead cells in the population after injection. Conclusion: The results indicate that the injection process through the cannula with these volume flows has no effect on the viability of the MSCs.

COVID-19 induced acute respiratory distress syndrome (ARDS) could have two different phenotypes, which might have different response and outcome to the traditional ARDS positive end-expiration pressure (PEEP) treatment. The identification of the different phenotypes in terms of the PEEP recruitment can help improve the patients’ outcome. In this contribution we reported a COVID-19 patient with seven-day electrical impedance tomography monitoring. From the conductivity distribution difference image analysis of the data, a clear course developing trend can be observed in addition to the phenotype identification. This case might suggest that EIT can be a practical tool to identify phenotypes and to provide progressive information of COVID-19 pneumonia.

Nowadays, rotary blood pumps for the treatment of end-stage heart failure patients are usually fixed-speed controlled. Therefore, without sufficient heart activity, patients may reveal reduced blood flow pulsatility. However, there is evidence that pulsatile flow reduces adverse events. Furthermore, optimized pump flow profiles could be used to achieve certain control objectives. Therefore, a pump flow model is identified and subsequently used to design a control system with an iterative learning controller to generate desired pump flow patterns. Either a flow sensor is required for direct measurement or the flow can be estimated from pressure sensor readings using the proposed pump flow model. For comparison, a PID controller with disturbance feedforward control is also designed. Furthermore, the robustness against respiratory sinus arrhythmia is analyzed.

Polyethylene (PE) inlays of knee endoprostheses are exposed to constant mechanical stress causing particle abrasion and volumetric wear. With the current inlay surveillance strategy, significant change often can be only detected when the implant has already failed. To reduce medical complications arising from implant malposition or failure, early and accurate assessment of the implant condition is of great value. We present a novel concept to monitor PE inlays based on an implanted sensor. Requirements of sensor integration into endoprostheses were discussed and specified for an inlay monitoring concept. A planar eddy current sensor for measuring the gradual descent of the femoral component was proposed where inductive and capacitive sensor elements form a resonant circuit. The potential sensitivity of the proposed measuring method was tested in an experimental setup using an RFID tag as a sensor substitute. A measurement concept, including the sensor and an extracorporal readout coil, was described and the validity was verified using finite element method (FEM) simulation of a simplified knee model. The experiments showed that a significant resonant frequency shift occurs in the sensor with decreasing distance to the femoral component. FEM simulation demonstrated that the sensor could be powered and readout extracorporeally through inductive coupling with an external readout coil. The proposed concept is a promising solution for feasible and accurate reading of the implant status designed to meet medical requirements. It can enable autonomous and routine monitoring as well as early detection of critical inlay deformation with a home-use device.

Gaussian processes are a versatile tool for data processing. Unfortunately, due to storage and runtime requirements, standard Gaussian process (GP) methods are limited to a few thousand data points. Thus, they are infeasible in most biomedical, spatio-temporal problems. The methods treated in this work cover GP inference and hyperparameter optimization, exploiting the Kronecker structure of covariance matrices. To solve regression and source separation problems, two different approaches are presented. The first approach uses efficient matrix-vector-products, whilst the second approach is based on efficient solutions to the eigendecomposition. The latter also enables efficient hyperparameter optimization. In comparison to standard GP methods, the proposed methods can be applied to very large biomedical datasets without any further performance loss and perform substantially faster. The performance is demonstrated on esophageal manometry data, where the cardiac and respiratory signal components are to be inferred by source separation.

Cardiopulmonary coupling (CPC) analysis links heart and respiration rates to assess sleep-related parameters. Typically, the CPC is measured using multi-lead electrocardiography (ECG) and ECG-derived respiration (EDR). Novel textile shirts with embedded ECG sensors offer convenient and continuously monitored sleep at home. We investigate the feasibility of a shirt with textile sensors (Pro- Kit, Hexoskin, Quebec, Canada) for CPC analysis by mobile computing. ECG data is continuously transmitted from the shirt to a smartphone via Bluetooth Low Energy (BLE). We customize a CPC algorithm and use twelve whole-night recordings from four volunteers to perform qualitative and quantitative analysis. We compare EDR with respiratory inductive plethysmography (RIP). In average, EDR and RIP differ 17.22%. After one night, the batteries are reduced to approx. 70% (shirt) and 90% (smartphone). The run time for CPC processing is approx. 3 min. Hence, smart wearables in combination with mobile computing show technical feasibility for CPC analysis. Eventually, this could yield a useful solution for sleep analysis of non-expert users in a private environment.

In vitro wound healing assays are a suitable application to verify the efficiency of pharmaceuticals or growth factors that will be incorporated in or immobilized to e.g. electrospun biomaterials for wound dressings or other biological devices in advance. Thereby, various factors like culture conditions or cell density influence the specific cell proliferation. Hence, to establish a wound healing assay for various cell types, a stepwise adaptation of cell numbers was done for better estimation and comparison of cell density for the validation of the influence of drugs on the wound healing process. Cell proliferation of different tissue relevant cell types was evaluated by impedance measurements and live cell imaging. Cell numbers could be successfully adapted for assay specific cell densities. In general, a universal comparison of biological or chemical materials and agents in vitro may require the creation of appropriate ISO or OECD standards for a consistent and cell specific adaptation or demand of initial cell density.

Auscultation methods allow a non-invasive diagnosis of cardiovascular diseases like atherosclerosis based on blood flow sounds of the carotid arteries. Since this process is highly dependent on the clinician’s experience, it is of great interest to develop automated data processing techniques for objective assessment. We have recently proposed a computerassisted auscultation system that we use to acquire carotid blood flow sounds. In this work, we present an approach for detecting artifacts within the blood flow sound caused by swallowing or coughing events. For this purpose, we first decompose the signal using a discrete wavelet transform (DTW). Then, we compute an energy ratio between the DWT scales associated with the signal information with and without artifacts using a sliding window of 1 s length. Evaluation based on Kruskal-Wallis and Wilcoxon rank-sum tests shows a statistically significant difference (p-value<.0001) between the signal with and without artifact. Therefore, the proposed method allows the identification of the studied signal artifacts.

Deep neural networks have shown effectiveness in many applications, however, in regulated applications like automotive or medicine, quality guarantees are required. Thus, it is important to understand the robustness of the solutions to perturbations in the input space. In order to identify the vulnerability of a trained classification model and evaluate the effect of different perturbations in the input on the output class, two different methods to generate adversarial examples were implemented. The adversarial images created were developed into a robustness index to monitor the training state and safety of a convolutional neural network model. In the future work, some generated adversarial images will be included into the training phase to improve the model robustness.

Features extracted from P waves of the 12-lead electrocardiogram (ECG) have proven valuable for noninvasively estimating the left atrial fibrotic volume fraction associated with the arrhythmogenesis of atrial fibrillation. However, feature extraction in the clinical context is prone to errors and oftentimes yields unreliable results in the presence of noise. This leads to inaccurate input values provided to machine learning algorithms tailored at estimating the amount of atrial fibrosis with clinical ECGs. Another important aspect for clinical translation is the network’s generalization ability regarding new ECGs. To quantify a network’s sensitivity to inaccurately extracted P wave features, we added Gaussian noise to the features extracted from 540,000 simulated ECGs consisting of P wave duration, dispersion, terminal force in lead V1, peak-to-peak amplitudes, and additionally thoracic and atrial volumes. For assessing generalization, we evaluated the network performance for train-validation-test splits divided such that ECGs simulated with the same atria or torso geometry only belonged to either the training and validation or the test set. The root mean squared error (RMSE) of the network increased the most in case of noisy torso volumes and P wave durations. Large generalization errors with a RMSE difference between training and test set of more than 2% fibrotic volume fraction only occurred if very high or low atria and torso volumes were left out during training. Our results suggest that P wave duration and thoracic volume are features that have to be measured accurately if employed for estimating atrial fibrosis with a neural network. Furthermore, our method is capable of generalizing well to ECGs simulated with anatomical models excluded during training and thus meets an important requirement for clinical translation.

Micro-electrode array (MEA) systems are noninvasive platforms for the investigation of electrophysiological properties of cell layers, such as spontaneously active cardiomyocytes. An MEA chip is composed of two-dimensional grids of dot-like electrodes embedded into glass. Here we present a test setup featuring a customised two-chamber silicone overlay. The overlay is designed to be placed on an MEA with two separate electrode fields and enables the seeding of two distinct cell sub-types on the MEA for synchronised drug testing applications while giving the possibility of analysing intersubtype-specific cellular interactions. The overlay has a full size of 10 x 10 x 5 mm (width x length x height), each chamber has a size of 2.5 x 6 x 5 mm (V = 75 mm³). The chambers are separated by a wall with a thickness of 0.3 mm. The overlay was manufactured via silicone-casting, utilising a 3D printed model. The model is 3D printed via high accurate digital light processing (DLP). In addition, a DLP 3D printed cover optimises the attachment of the overlay on an MEA. A proofof- principle of the utilisation of the overlay is demonstrated.

The ongoing coronavirus pandemic spreads through airborne transmission and is therefore difficult to contain. However, coronaviruses are highly sensitive to UVC, so UVC air disinfection systems should be able to inactivate the virus. Unfortunately, so far there are only few possibilities to test the reduction of airborne viruses or other pathogens. A special test rig, which mainly consisted of a nebulizer and an airflow system, was developed to determine the antiviral and antibacterial efficiency of UVC air disinfection systems. In the assessment of such an UVC air disinfection system with nebulized Staphylococcus carnosus and a sampling period of 30 minutes, a mean bactericidal reduction of 3.70 log10 (99.98 %) was determined. For antiviral irradiation of the coronavirus surrogate phi6 a mean viral load reduction of 1.18 log10 (93.40 %) was observed after a sampling period of 10 minutes. Therefore, mobile UVC air disinfection systems could be applied in hospitals, retirement and nursing homes.

Phantoms are essential tools for the development and characterization of Magnetic Particle Imaging (MPI), an imaging technique that can quantitatively map the spatial distribution of magnetic nanoparticles (MNP). The objective of this study was to develop and validate a modular MPI phantom kit with high versatility for platform-independent quality assurance and the assembling of defined geometries in MPI. It was shown that the developed MPI phantom kit can be used for both application scenario testing and quality assurance in MPI which provides the basis for future reference phantoms to directly compare existing scanners within the MPI community.

The COVID-19 is a viral infection that causes respiratory complications. Infected lungs often present ground glass opacities, thus suggesting that medical imaging technologies could provide useful information for the disease diagnosis, treatment, and posterior recovery. The Electrical Impedance Tomography (EIT) is a non-invasive, radiationfree, and continuous technology that generates images by using a sequence of current injections and voltage measurements around the body, making it very appropriate for the study to monitor the regional behaviour of the lung. Moreover, this tool could also be used for a preliminary COVID-19 phenotype classification of the patients. This study is based on the monitoring of lung compliances of two COVID-19-infected patients: the results indicate that one of them could belong to the H-type, while the other is speculated belongs to L-type. It has been concluded that the EIT is a useful tool to obtain information regarding COVID-19 patients and could also be used to classify different phenotypes.

The unique magnetic properties of magnetic nanoparticles (MNP) combined with their small size already led to numerous medical applications. Accurate determination of their magnetic properties is a key requirement enquired by users, that is impeded by the ever-present distribution of MNP sizes. Field flow fractionation (FFF) techniques may help to overcome these limitations by first separating the particles before characterization. In this study, we demonstrate the use of centrifugal FFF coupled to online detectors for fractionation, structural, and magnetic characterization of MNP. The primary goal is to establish a reproducible centrifugal FFF (CF3) method for MNP fractionation We show that CF3 has the same capability as other FFF techniques in resolving the bimodal hydrodynamic size distribution present in the commercial MNP system Resovist® but is faster and more straightforward through its technical approach.

Teeth grinding is, due to its various impacts on the human body, a highly discussed issue in dentistry. It can damage the tooth structure or cause pain due to muscle tension. At the moment, there is neither a satisfactory diagnostic nor a comprehensive treatment option. This paper deals with the development of an app-controlled, small, portable sensor unit that can be used by patients to monitor their teeth grinding in everyday life. It also offers a treatment option due to an implemented biofeedback option. To achieve the most cost-effective device possible, only off-the-shelf electronics and no proprietary software were used. In initial tests, the measuring device showed high level of measurement accuracy when performing measurements without feedback at rest (f_score=-0.025...0).

For the image-based documentation of a colonoscopy procedure, a 3D-reconstuction of the hollow colon structure from endoscopic video streams is desirable. To obtain this reconstruction, 3D information about the colon has to be extracted from monocular colonoscopy image sequences. This information can be provided by estimating depth through shape-from-motion approaches, using the image information from two successive image frames and the exact knowledge of their disparity. Nevertheless, during a standard colonoscopy the spatial offset between successive frames is continuously changing. Thus, in this work deep convolutional neural networks (DCNNs) are applied in order to obtain piecewise depth maps and point clouds of the colon. These pieces can then be fused for a partial 3D reconstruction.

The use of catheters and guide-wires in minimally invasive therapeutic approaches is an important part of clinical practice. In the neurovascular field, the unique nature of cerebral blood vessels necessitates very thin-walled and flexible catheters. The blood vessels in question are highly branched and at the same time can be less than one millimetre in diameter. This results in high demands on micro-catheters and guide-wires for successful endovascular therapy. The interaction of these surgical tools and the vessel wall is of especial interest. Depending on the catheter stiffness, this interaction can be friction, punctual collision or straightening. The work aims to design and implement a test setup for evaluation of these interactions with the vessel wall. For this purpose, a standardized vessel course with representative characteristics is necessary. Furthermore, by implementing suitable measuring equipment, an endovascular intervention can be simulated.

A twisted string actuator (TSA) is a small, strong, lightweight, and low-cost gear, transforming rotation into a linear pulling movement. The TSA consists of two or more strings that are twisted along their common longitudinal axis. The helix formed in this process becomes shorter the further the bundle is twisted. A possible application is a tendon-based endoscopic robot. To control the movement of the endoscope, a precise contraction of the tendon is necessary. Since the strings of the TSA show an elastic behaviour, position feedback is needed to determine the exact movement of the TSA. In this paper, a TSA with a closed-loop position control by a low-cost displacement sensor is presented.

Body weight supported treadmill training is used for patients who are severely limited in their movements. During walking therapy handrails and harnesses ensure the patient’s stability and safety. Although, locomotion training is a common rehabilitation technique, data of forces applied to handrails or overhead lifts are limited. Therefore, the primary objective of this study was to measure the vertical reaction force during treadmill therapy. An additional purpose was the characterization of the loading profiles by defining suitable parameters. To measure the patient’s needs for support during treadmill training, the handrails were instrumented with strain gauges and a force sensor was placed between the harness and the ceiling lift. In order to provide gait phase detection, an inertial sensor was attached to each foot. The results confirm that the combined measuring systems are appropriate to record the total vertical weight relief. Although all patients show different loading profiles, a recurring pattern can be recognized for each patient, which allows the derivation of various parameters to characterize the single profiles. The additional information of the measurements can help the supervising physiotherapist to better understand the specific level of support the patient needs. A subsequent response and the providing of corresponding instructions by the therapist can lead to an even more individualized therapy.

A model of the pulse oximeter, which consists of a transfer function between arterial and peripheral blood oxygen saturation (SpO 2 ) and the noise typical for SpO 2 records, is an important part of a mathematical model of oxygenation in neonates that is designed to test and compare different algorithms of oxygen control. The noise level in the SpO 2 signal is affected by the averaging time setting of the pulse oximeter. This study aimed to characterize the noise level in the SpO 2 signal at the set pulse oximeter averaging times of 2- 4, 8, and 16 seconds. We evaluated SpO 2 records of 17 healthy volunteers who underwent a laboratory experiment in which they evoked different types of artifacts. The noise level in the SpO 2 signal was characterized by two parameters, the deviation of SpO 2 from the true value and the SpO 2 time stability, defined as the interval during which the measured SpO 2 value remained unchanged. Statistical properties of the noise level for the three averaging times were represented by normalized histograms of both the parameters and varied according to the type of artifact. With motion artifacts, the SpO 2 readings deviated from the true value by more than ±2% SpO 2 in 10%, 7%, or 5% of the measurements when the set averaging time was 2-4 s, 8 s, or 16 s. The length of the interval over which the SpO 2 value remained unchanged was most frequently 2 seconds for all set averaging times. Implementation of the noise characteristics into the computer model of oxygenation in neonates will allow more faithful simulations of the output SpO 2 signal that better match clinical observations.

Cartilage constructs produced by SFCTtechnology provide promising opportunities to restore cartilage defects. Mechanical parameters of soft tissues are explicit markers for quantitative tissue characterization. In this study, we present a biphasic 3D-FE-based method to determine the biomechanical properties of SFCT from stress relaxation compression tests (ε = 20 %, t = 3400 s) whereby cartilaginous tissue is modeled as a biphasic material with tension-compression nonlinearity (BMTCN). The FE-model computation was optimized by exploiting the axial symmetry and mesh resolution. The R² of the fit results varies between 0.970 and 0.983. The Young’s and fiber modulus determined from SFCT are 37-times and 5-times lower than from native articular cartilage, respectively. Permeability, on the other hand, is 11-times higher than from native articular cartilage.

Flexible Multi Electrode Arrays (MEAs) for neural interfacing reduce the mechanical mismatch between the soft brain tissue and the electrode arrays allowing accurate signal recordings and neural stimulation while reducing inflammatory responses. Many standard manufacturing processes of MEAs are designed for planar structures and the production of three-dimensional structures is challenging. In the present study, shaft structures with one to two circular gold microelectrodes (10 - 20 μm), each on a base polyimide (PI) substrate, were investigated. We describe a fabrication method, with which shafts made from bi-layer PI flip into the third dimension, which is a first step towards spontaneous assembly of electrodes in flexible 3D MEAs for neuroelectronic applications. A lift-up of the shafts was achieved by the contraction of a second PI layer and a steady nitrogen flow during polycondensation. This shrinking PI was structured in pits with a width of 5 - 600 μm. We achieved liftup angles of up to 42 degrees. The shaft structures can be hardened and later be used for neural implantation experiments.

Mechanical parameters of hard and soft tissues are explicit markers for quantitative tissue characterization. In this study, we present a comparison of biphasic material properties of equine articular cartilage estimated from stress relaxation (ε = 6 %, t = 1000 s) and creep indentation tests (F = 0.1 N, t = 1000 s). A biphasic 3D-FE-based method is used to determine the biomechanical properties of equine articular cartilage. The FE-model computation was optimized by exploiting the axial symmetry and mesh resolution. Parameter identification was executed with the Levenberg- Marquardt-algorithm. Additionally, sensitivity analyses of the calculated biomechanical parameters were performed. Results show that the Young’s modulus E has the largest influence and the Poisson’s ratio of ν ≤ 0.1 is rather insensitive. The R² of the fit results varies between 0.882 and 0.974 (creep model) and between 0.695 and 0.930 (relaxation model). The averaged parameters E and k determined from the creep model yield higher values in comparison to the relaxation model. The differences can be traced back to the experimental settings and to the biphasic material model.

The development of strategies for fibrosis prevention is a perpetual cutting the edge challenge, especially in ophthalmologic fistulating surgery. Stenosis of liquid draining implants but also fibrotic activation of implant-associated tissues are major clinical examples for fibrosis induced implant failure. Implant coating or incorporation of antifibrotic agents offer a promising approach to minimise failure rates. Nintendanib is a drug that has been successfully used for the treatment of ideopathic pulmonary fibrosis since 2015. In this study, we evaluated the suitability of Nintedanib for ophthalmic therapeutic treatment. Therefore, the antifibrotic potential of the active substance was tested with a fibrotic cell culture model on primary fibroblasts of the Tenon (hTF) in vitro. The concentration of 10μM Nintedanib demonstrated a marginal effect on cell viability but coincidently diminished cell proliferation remarkably. Both, immunocytochemical and Western blot analyses revealed a significant inhibitory effect of Nintedanib on the TGF-β1 induced expression of the extracellular matrix (ECM) components fibronectin and collagen. Moreover it supressed the expression and formation of stress fibres of the fibrotic marker protein alpha smooth muscle actin (α-SMA).

Commonly used blood pressure measurement devices have noticeable limitations in accuracy, measuring time, comfort or safety. To overcome these limitations, we developed and tested a surrogate-based, non-invasive blood pressure measurement method using an RGB-camera. Our proposed method employs the relation between the pulse transit time (PTT) and blood pressure. Two remote photoplethysmography (rPPG) signals at different distances from the heart are extracted to calculate the temporal delay of the pulse wave. In order to establish the correlation between the PTT values and the blood pressure, a regression model is trained and evaluated. Tests were performed with five subjects, where each subject was recorded fifteen times for 30 seconds. Since the physiological parameters of the cardiac system are different for each person, an individual calibration is required to obtain the systolic and diastolic blood pressure from the PTT values. The calibration results are limited by the small number of samples and the accuracy of the reference system. However, our results show a strong correlation between the PTT values and the blood pressure and we obtained a mean error of 0.18 +/- 5.50 mmHg for the diastolic blood pressure and 0.01 +/- 7.71 mmHg for the systolic pressure, respectively.

Hearing contact lens (HCL) is a new type of hearing aid devices. One of its main components is a piezoelectric actuator (PEA). In order to evaluate and maximize the HCL’s performance, a model of the HCL coupled to the middle ear was developed using finite element (FE) approach. To validate the model, vibrational measurements on the HCL and temporal bones were performed using a Laser-Doppler-Vibrometer (LDV). The model was validated step by step starting with HCL only. Then a silicone cap was fitted onto the HCL to provide an interface between the HCL and the tympanic membrane. The HCL was placed on the tympanic membrane and additional measurements were performed to validate the coupled model. The model was used to evaluate the sensitivity of geometrical and material parameters with respect to performance measures of the HCL. Moreover, deeper insight was gained into the feedback behavior, which causes whistling sounds, and the contact between the HCL and tympanic membrane.

The combination of transcranial alternating current stimulation (tACS) and electroencephalogram (EEG) for mobile and home-based interventions offers the potential for control and adaptation of stimulation parameters. Yet, during stimulation, the EEG is heavily affected by stimulation artifacts. Spatial filters are often unsuited because too few channels are recorded and hardware capabilities are limited. Due to their speed and as they can be used for single channels, we explore the performance of single-channel weighted comb filters on artifact removal. At any given time point t, the recording r(t) is a superposition of a neurophysiological signal n(t), the stimulation artifact a(t) and noise e(t). Now, we can estimate the artifact a(t) based on the recording from an earlier (or later) time-point shifted by the artifacts period. A weighted estimate based on multiple time points has the potential to improve the signal recovery. Therefore, we explored several approaches and evaluated their performance on simulated and real data. The comb kernel filters were implemented in Matlab (https://github.com/agricolab/ARtACS) and Python (https://github.com/agricolab/pyARtACS), and the code is open access under an X11-license. We found that independent of the weighting function, all comb filters exhibit similarity in their suppression of the DC component, the artifacts frequency, and its harmonics. Yet, different weighting functions exhibit different pass-band performance, evident as ringing and amplification, and their induction of time-domain echoes. Interestingly, we note that a causal uniform filter is comparable to more complex approaches, while offering the option for real-time filtering. Comb filters are able to remove tACS artifacts even if only a single channel is available. As comb filters require no assumptions about the shape of the artifact, they might also be useful for filtering of non-sinusoidal, e.g. pulsed or saw-tooth, transcranial current stimulation.

The documentation of a surgical procedure remains a time-consuming task that surgeons must incorporate into their daily routine. However, since a surgical report should be produced immediately after the operation with all impressions of the procedure in mind, a means of automation assistance should be provided. We, therefore, propose a method that generates surgical reports based on keywords stated during the procedure. Our report generation is based on a sequence-tosequence model that is trained on sentence pairs of two consecutive sentences in a surgical report. The known sentence is augmented with a keyword based on the following surgical action to be documented and is then passed into a language model to generate the next sentence. In this way, the complexity of predicting a vast number of possible surgical report phrasings is reduced to a next sentence prediction task. For the language model, an encoder-decoder structure was used with bidirectional 2-layer Long-Short Term Memory (LSTM) units for both components and an attention layer between input and output sentences. The training data consisted of 50 ear-,nose- and throat surgery (ENT) reports with 1500 sentences. The model training was performed in a k-fold cross-validation study with k = 10 and cross-entropy loss as the objective function. The generated reports were investigated using NIST, ROUGE, and METEOR metrics. Additionally, three medical experts identified the report content regarding plausibility and text errors. The trained models reached an accuracy of 0.82 for the next sentence predictions. The generated reports show consistent sentence structures and keyword correspondence for about 70 % of provided keyword sequences. The NIST, ROUGE, and METEOR metrics reached 0.65, 0.71, and 0.64, respectively. The model underperformed for not yet known keyword sequences and shows signs of overfitting when keyword sequences deviate from the baseline of the training set. Our approach for the keyword-augmented generation of surgical reports shows the potential of reducing the text generation complexity by providing a sequence of anchor words. However, the automated generation of surgical reports remains a difficult task due to individual report phrasings and the high variance in keyword sequences.

We introduce a system that allows the immediate identification and inspection of fat and muscle structures around the lumbar spine as a means of orthopaedic diagnostics before surgical treatment. The system comprises a backend component that accepts MRI data from a web-based interactive frontend as REST requests. The MRI data is passed through a U-net model, fine-tuned on lumbar MRI images, to generate segmentation masks of fat and muscle areas. The result is sent back to the frontend that functions as an inspection tool. For the model training, 4000 MRI images from 108 patients were used in a k-fold cross-validation study with k = 10. The model training was performed over 25-30 epochs. We applied shift, scale, and rotation operations as well as elastic deformation and distortion functions for image augmentation and a combined objective function using Dice and Focal loss. The trained models reached a mean dice score of 0.83 and 0.52 and a mean area error tissue of 0.1 and 0.3 for muscle and fat tissue, respectively. The interactive webbased frontend as an inspection tool was evaluated by clinicians to be suitable for the exploration of patient data as well as the assessment of segmentation results. We developed a system that uses semantic segmentation to identify fat and muscle tissue areas in MRI images of the lumbar spine. Further improvements should focus on the segmentation accuracy of fat tissue, as it is a determining factor in surgical decisionmaking. To our knowledge, this is the first system that automatically provides semantic information of the respective lumbar tissues.

Radial shock wave therapy is used for different therapeutic indications. In order to assess the effect on the treated tissue, it is important to know the sound field parameters. However, it is challenging to measure the pressure curves of ballistic devices, especially at high pulse repetition rates. In the new standard IEC 63045:2020 two possibilities for sound field measurements of non-focusing devices are described, a wet and a dry test bench. The whole sound field can be characterised using a wet test bench, but the process is cumbersome and cavitation is likely to occur at high pulse repetition rates. This effect is avoided using a dry test bench where the measurement position is limited to a single spot. Therefore, a hybrid test bench was developed combining the dry bench’s device mounting and coupling with a small water basin. The ballistic device was coupled to the basin filled with degassed ultrapure water using a latex membrane covered with ultrasound gel and the contact pressure was applied with a spring. A fibre optic probe hydrophone was used for the sound field measurements. The pressure curves of every 10th shot were measured on the beam axis in a distance of 1mm to the membrane. The device was analysed at 3.5 bar with different pulse repetition rates (1 Hz, 10 Hz, 20 Hz). The test setup enables an easy handling and reproducible results at all pulse repetition rates. The ballistic device provides constant peak pressures over different frequencies. The small water basin has the advantage that the water quality is easy to control and the measuring process is fast and uncomplicated. Cavitation suppression requires a clean water basin filled with degassed ultrapure water kept at a constant, low temperature. The hybrid test bench can be used to easily study shock wave parameters of ballistic devices at high repetition rates.

Here, we demonstrate a manufacturing process for individualised, small-sized implant prototypes. Our process is promising for the manufacturing of drug-releasing (micro)implants to be implanted in the round window niche (RWN-I, solid body, free-form-shaped design, 1.1 x 2.7 x 3.1 mm) and for frontal neo-ostium implants (FO-I, tube-like design, length ~ 7 mm, Ø ~ 2-6 mm) for frontal sinus drainage. Implant prototypes are manufactured using micro injection molding (μIM). We use digital light processing (DLP) as a 3D printing technique for rapid tooling of accurate molds for the μIM process. A common acrylate-based photopolymer for stiff and high-detailed modelling but with low head deflection temperature of HDT = 60.5 °C is used for DLP 3D printing of the molds. The molds were 3D printed with a layer height of 50 μm in about 20 min (RWN-I) and 60 min (FO-I). For μIM investigations, we use liquid silicone rubber (LSR) as a biocompatible and medically relevant material. Micro injection molding of LSR was investigated using mold temperatures between Tmold = 110 °C (long tcuring ~ 2 h) up to Tmold = 160 °C (short tcuring ~ 5 min). As a result, small-sized, complex-shaped implant prototypes of LSR can be successfully manufactured via μIM using high Tmold = 160 °C and short curing time. DLP 3D printing material with relative low HDT = 60.5 °C was suitable for μIM. There is no significant wear of the molds, when used for a low number of μIM cycles (n ~ 8). Design of metal mold housing has to be suitable (perfect fit of mold, no cavities facing the molds surface for prevention of thermal expansion of mold into cavities).

Telemedicine promises to increase the quality of emergency treatments. Besides the transfer of speech and video data, medical device and patient data will add additional value to the tele-guided emergency personnel. In this paper, we develop a concept for transmitting device and patient data via the open communication standards SDC in a telemedical context, including data transmission over mobile radio networks while considering the limitations of public networks, and opening new usage scenarios for telemedicine.

3D-printing paves the way for personalized drug therapy via implants individualized for the patient specific anatomy in chronic paranasal sinus diseases. This study brings together the workflow of modeling, manufacturing, and sterilization of 3D-printed individualized frontal neo-ostium implants (FOI) for optimization of Endoscopic Sinus Surgery (ESS) and validates the implantability of the printed devices. The study sample consisted of six adult human cadavers. Digital volume tomography (DVT) images were taken before and after ESS. The FOI models were based on the post-ESS anatomy. The area to be implanted was analysed in the preand post-ESS DVT images for volume and surface area. A specialized surgeon in rhinology judged the implantability of the 3D-printed FOI. The mean values of volume and surface area tended to be larger in the post-operation situs than in preoperation DVT images and we are therefore not yet convinced that pre-operation images will reflect the correct anatomy for the personalization of the FOI. The workflow of manufacturing and implantation of individualized 3D-printed sterile FOI is established but future studies are needed for further improvement.

Implant-associated infections still must be considered as a major risk factor and represent a significant threat to the well-being of patients. The detection of such complications is often late, making therapy at this stage difficult. However, the common route of systemic antibiotic prophylaxis can be associated with numerous side effects on the patient or even lead to resistant strains of bacteria. In this context, local antibiotic drug load is a promising way to protect implants against pathogen adhesion and biofilm formation. This work addresses the incorporation of the antibiotic Doxycycline into a biodegradable nonwoven matrix as potential implant coating. While the nonwoven matrix itself displayed already inhibitory effects on Staphylococcus aureus (S. aureus) biofilm formation in vitro, this effect was even more pronounced by the incorporation of Doxycycline. Antibiotic loaded nonwovens display the possibility for local inhibition of biofilm formation. In the shape of an implant coating, this may further help to avoid implantassociated infections. Nevertheless, as pathogens vary in shape and type, further adjustments have to be performed to improve wide-ranging protective effects

The tecLA project developed a modular system for the area of Ambient Assisted Living (AAL) that enables older people to have simple, intuitive access to medical and technical assistance systems. A commercially available AAL application portal has been subjected to acceptance tests and optimizations and has proven itself in field tests. Furthermore, when comparing the self-assessment values to cognitive measures before and after the intervention, a clear trend towards more positive values in the post-test was recorded. It was also shown that the use of the system has very positive effects on communication behavior and cognitive selfassessment. The nurses and doctors can follow the vital signs online in the patient's file and evaluate them immediately.

Mental stress triggers positive inotropic and chronotropic effects as well as peripheral vasoconstriction. This alters the pulse arrival time (PAT), the duration between electrical excitation of the ventricles and arrival of the pulse wave in the periphery. We conducted a study to examine PAT during five rest blocks and under mental stress utilizing the Mannheim Multicomponent Stress Test. Electrocardiograms as well as finger and earlobe photoplethysmograms were recorded. PAT was calculated for over 135,000 heartbeats from 42 healthy volunteers as the time duration between the R peak in the electrocardiogram and the following pulse onset in the respective photoplethysmogram. To identify the effect of mental stress, block-wise PAT means were statistically analyzed with repeated measures ANOVA. The analyses showed significant differences between the block means for both PAT measures (p < 0.001). Post-hoc tests revealed significantly reduced PAT during the stress block compared to all rest blocks for both PAT measures (p < 0.001). We found no significant differences between the rest blocks. Our results support that PAT is a sensitive vital parameter for the detection of mental stress in healthy volunteers. This holds true for both measurement positions, the finger and the earlobe.

In order to improve the reliability of diagnosis of nasal breathing disorders, aerodynamic properties have to be analyzed through experiments based on 3D models. The surface properties of the prepared respective 3D models using fused deposition modeling (FDM) should match those of native nasal cavities, thus representing their normal state and typical pathologies. In this work, we validated the smoothing of dual extruded 3D printed samples of PLA (polylactide) and PVA (polyvinyl alcohol) using the solvent TFE (trifluoroethanol). The smoothing was conducted in vapour and liquid phases of TFE. Before and after treatment of the samples in liquid and vapour phases of TFE, mass and surface roughness analysis were performed. The results of this work will help to produce and process a representative model of the human paranasal sinuses, which can be created using CT data from a patient.

Due to the fact that a relatively large area can be disinfected in a short time, dry-fog disinfection could have a great potential to improve hygienic measures. However, there are few studies to date on the effectiveness and combinability of this technology with other devices. In this work, the disinfecting potential of the TBT dry fogging technology (TBT Desinfektion GmbH & Co. KG, Germany) and the corresponding disinfectant Defeat AR (Biofluid GmbH, Germany) was examined with regard to microbiological decontamination and potential damaging effects on different devices. The airburden measurement did show a reduction of approx. 66% of the microbiological room contamination. Similarly, decontamination of test surfaces was shown to be effective at all measuring points we selected in the experimental setup. Depending on the accessibility of the surfaces this effect was in the range of several log levels. Nevertheless, a first test run showed that some technical equipment was negatively affected by the process.

The risk of infection from contaminated surfaces has already been shown in several publications. Due to the increased demand for optimized infection control measures during the Corona pandemic, antimicrobial surface technologies have gained more an interest. Apart from many proofs of efficacy, there are only a few studies dealing with the durability of these surface coatings with regard to the material and the reprocessing measures. This work did therefore examine the impact of different materials and surface textures, as well as different detergents and disinfectants, on the durability of antimicrobial surface technologies. Differently structured materials (glass, wood, plastics, metal) and wallpaper bonded to plasterboard were coated with an TiO2Ag based antimicrobial coating (HECOSOL GmbH, Bamberg). These test samples are then used to perform abrasion tests with various cleaning and disinfecting agents and cloth systems (microfiber cloth, cotton cloth, foam cloth). The majority of the test samples in our experimental setup showed at least significant activity. According to our results, both the selection of cleaning and disinfection methods including wiping systems and the surface material have a major impact on the durability of antimicrobial coatings. In order to be able to come to conclusions about the long-term activity of these surface technologies, the effectiveness should be tested not only during the development phase, but also in the finished product and again after several reprocessing cycles in use.

With the new technology of 3D light field (LF) imaging, fundus photography can be expanded to provide depth information. This increases the diagnostic possibilities and additionally improves image quality by digitally refocusing. To provide depth information in the human optic nerve head such as in glaucoma diagnostics, a mydriatic fundus camera was upgraded with an LF imager. The aim of the study presented here was the validation of the technical setup and resulting depth estimations with an appropriate eye model. The technical setup consisted of a mydriatic fundus camera (FF450, Carl Zeiss Meditec AG, Jena, Germany) and an LF imager (R12, Raytrix GmbH, Kiel, Germany). The field of view was set to 30°. The eye model (24.65 mm total length) consisted of a two-lens optical system and interchangeable fundus models with papilla excavations from 0.2 to 1 mm in steps of 0.2 mm. They were coated with red acrylic lacquer and vessels were drawn with a thin brush. 15 images were taken for each papilla depth illuminated with green light (wavelength 520 nm ± 20 nm). Papilla depth was measured from the papilla ground to the surrounding flat region. All 15 measurements for each papilla depth were averaged and compared to the printed depth. It was possible to perform 3D fundus imaging in an eye model by means of a novel LF-based optical setup. All LF images could be digitally refocused subsequently. Depth estimation in the eye model was successfully performed over a 30° field of view. The measured virtual depth and the printed model papilla depth is linear correlated. The presented LF setup allowed high-quality 3D one-shot imaging and depth estimation of the optic nerve head in an eye model.

Passive Acoustic Mapping (PAM) is an ultrasoundbased imaging method developed for monitoring therapeutic ultrasound. By using diagnostic transducers to passively record the acoustic signals that are emitted by cavitation bubbles, the origin of the bubbles can be reconstructed and displayed as intensity maps. In this study, two matrix arrays with different aperture sizes were used for the volumetric reconstruction of simulated and experimental data. In a second step, the number of elements being used for the reconstruction was reduced by more than the factor of eight in order to assess the influence on the imaging quality. In the numerical part of the study, the image quality was greatly improved by increasing the aperture size, while a high number of elements used for the reconstruction merely offers minor improvements. The experimentally obtained results were able to confirm the numerical findings regarding the achievable reconstruction quality.

Extrusion is a common manufacturing process for semi-finished polymer products in various biomedical applications. Extrusion enables processing of a wide range of biomaterials, as well as different cross-sectional geometries. Furthermore, feasibility of drug elution, as it is used for a variety of medical devices, for example microstents for minimally invasive glaucoma therapy, was assessed. The current study deals with manufacturing of polymeric microtubes by extrusion processing. Semi-finished products were made of biodegradable poly-L-lactide (PLLA) and a non-biodegradable polycarbonate-based silicone elastomer (SIL) and covered with a polymer/drug combination, with resveratrol as active ingredient. Three different concentrations of polymer/resveratrol were applied by means of spray-coating. The release behavior of active ingredient was analyzed in vitro at 37°C and showed a correlation between the amount of drug and release time. With higher drug content, a faster release was observed. In addition, the release from SIL was faster compared with PLLA.

Laser bow systems are used for intervention guidance by steering laser rays. Such systems have shown applications in biopsy, nasal surgeries, drainage, pain therapy. In this paper, we introduce an image-based laser guided CT intervention. Instead of relying on traditional laser bow rigidly mounted to the CT scanner, our approach performs registration making use of a patient tracker in image as well as CT data to find the orientation of laser bow with respect to the CT data. The image is acquired by a camera rigidly connected to the laser bow. The proposed approach allows movability to the laser bow system without compromising on accuracy. Using camera allows real-time view of the intervention and can be used for futuristic applications.

Cochlear implants (CI) can restore hearing to people suffering from sensorineural hearing loss. The CI uses an array of electrodes inserted in the scala tympani to stimulate the spiral ganglion neurons (SGN), which are located in the bony axis of the inner ear. The spatial distance between the electrodes and the SGN results in spread of excitation and unfocused stimulation. This distance could be bypassed by a neurite outgrowth towards the CI. In order to establish a culture system for the development of novel strategies for CI optimization, a neurite outgrowth chamber (NOC) allowing clinically relevant electrical stimulation paradigms was developed using fused deposition modelling. It was made out of biocompatible UV-curing silicone. The NOC was equipped with a slot to insert the electrode array into a scala tympani compartment whereas SGN isolated from postnatal rats were cultured in a neighbouring region mimicking the Rosenthal’s canal. Sound delivered through loudspeakers playing a radio program was captured by the CI sound processor, positioned outside the NOC, for 17 hours daily during a four-day period. The NOCs were tightly sealed and the electrode array could be positioned in the scala tympani compartment. The experimental setup allowed cell cultivation and the stimulation resulted in a significantly increased neurite length of around 36% while explant area and neurite number did not differ to the negative control (NC). A new in vitro testing system which considers the anatomy of the cochlea and clinic stimulation conditions has been developed and validated. The validation experiments resulted in increased neurite length. This indicates a progress in bridging the anatomical gap between electrode and stimulated neurons.

Since 2021, the Fraunhofer Institutes ITEM and IIS and the German Society for Biomedical Engineering (VDE DGBMT) have launched a new online seminary. This sustainable format, referred to as "Expert Panel", stimulates scientific interaction by introducing research topics and innovations within sixty minutes. The panel with three experts from different fields presents and discusses perspectives from science, clinical practice, and the industry. The first season of the "Expert Panel" offered comprehensive insights on the latest developments and innovations within "New Interventional Methods for Tomorrow's Hearing". Each topic was presented by a triad of experts from science, clinical practice, as well as the industry. The sessions comprised life-demonstration of planning tools for surgical interventions, the presentation of patient-individual intervention techniques and novel audiological systems for Cochlea Implant (CI) fitting and many more inspiring technologies. Each “Expert Panel” concluded with a moderated Q&A final. The sessions provided condensed updates and unique insights from applied research and had a lively attendance of up to more than 120 participants from the fields of research, clinic, and industry. This kind of format does not only drive interpersonal and interdisciplinary knowledge transfer - especially in times of home office - it also enables establishing a community of network partners, who meet at the “Expert Panel” on a regular basis. In addition, education credits have been offered to the medical community, thus creating additional benefit for the participants.

Existing challenges in surgical education (See one, do one, teach one) as well as the Covid-19 pandemic make it necessary to develop new ways for surgical training. This is also crucial for the dissemination of new technological developments. As today’s live transmissions of surgeries to remote locations always come with high information loss, e.g. stereoscopic depth perception, and limited communication channels. This work describes the implementation of a scalable remote solution for surgical training, called TeleSTAR (Telepresence for Surgical Assistance and Training using Augmented Reality), using immersive, interactive and augmented reality elements with a bi-lateral audio pipeline to foster direct communication. The system uses a full digital surgical microscope with a modular software-based AR interface, which consists of an interactive annotation mode to mark anatomical landmarks using an integrated touch panel as well as an intraoperative image-based stereo-spectral algorithm unit to measure anatomical details and highlight tissue characteristics.We broadcasted three cochlea implant surgeries in the context of otorhinolaryngology. The intervention scaled to five different remote locations in Germany and the Netherlands with lowlatency. In total, more than 150 persons could be reached and included an evaluation of a participant’s questionnaire indicating that annotated AR-based 3D live transmissions add an extra level of surgical transparency and improve the learning outcome.

Magnetic nanoparticles (MNPs) have been used in several medical applications, including targeted hyperthermia, resonance tomography, diagnostic sensors, and localized drug delivery. Further applications of magnetic field manipulation through MNPs in tissue engineering have been described. The current study aims to develop tissue-engineered polymeric scaffolds with incorporated MNPs for applications that require stimulation of the tissues such as nerves, muscles, or heart. Electrospun scaffolds were obtained using 14%w/v polycaprolactone (PCL) in 2,2,2-Trifluoroethanol (TFE) at concentrations of 5% & 7.5%w/v of dispersed MNPs (iron oxide, Fe3O4, or cobalt iron oxide, CoFe2O4). Scaffolds were analyzed using scanning electron microscopy (SEM), energy dispersive x-ray spectroscopy, uniaxial tensile testing, and cell seeding for biocompatibility. Human bone marrow mesenchymal stem cells (bmMSCs) were seeded on the scaffolds. Biocompatibility was assessed by metabolic activity with Resazurin reduction assay on day 1, 3, 7, 10. Cell-cell and cell-scaffold interactions were analyzed by SEM. Electrospun scaffolds containing MNPs showed a decrease in fiber diameter as compared to scaffolds of pure PCL. The maximum force increases with the inclusion of MNPs, with higher values revealed for iron oxide. The metabolic activity decreased with MNPs, especially for cobalt iron oxide at a higher concentration. On the other hand, the cells developed good cell-scaffold and cell-cell interactions, making the proposed scaffolds good prospects for potential use in tissue stimulation.

We present a stereo-multispectral microscope equipped with an additional illumination unit allowing further narrow-band illumination in the spectral range of 400n.m up to 800nm. The combination of the normal microscope illumination with the multispectral light unit allows different illumination modalities to be realized, which enables intraoperative spectral tissue analysis with direct visualization. Two illumination methods were tested in two cholesteatoma surgeries. In addition, two cholesteatom samples were illuminated and analyzed ex vivo. Cholesteatoma showed :fluorescent characteristics in our ex vivo analysis. This behavior could be used intraoperatively using a combination of white light and strong near-UV to blue illumination to highlight cholesteatoma tissue in the microscopic image. Thus, the visual differentiability of different tissue types can be improved and the clinical decision-making process can be accelerated.

Infective endocarditis (IE) is an infection of the endocardium, and the heart valves associated with high morbidity and mortality. Fluorescence in situ Hybridization (FISH) is a molecular imaging technique used for diagnosis of IE based on histological heart valve tissue sections. FISH allows detection and identification of microorganisms and gives information about their quantity and spatial distribution. This information is important to guide appropriate antibiotic treatment. However, as manual FISH image analysis is time- and costexpensive, an automated image analysis pipeline (consisting of tissue segmentation, bacteria detection, and spot detection modules) is proposed to assist locating potential regions with microorganisms. The proposed approach was evaluated in a study, where five observers manually assessed a set of 171 fields-of-view (FoVs) captured in 400-fold magnification from 10 randomly chosen WSI for the presence of microorganisms, morphologically detected by the nucleic acid stain DAPI. The task of the observers was to mark the presented image using a 2-class score (‘positive/questionable’ or ‘negative’). The human assessment was compared to the results suggested by the algorithm. The proposed algorithm locates and ranks potential regions with microorganisms in heart valve sections so that experts can validate them in higher power FoVs for the presence of bacteria and identify their species. The automated system for preselecting and recommending adequate FoVs is thus a starting point to support experts and save human resources. It is now ready to be further developed for the detection of bacteria by FISH.

Needles are key tools to realize minimally invasive interventions. Physicians commonly rely on subjectively perceived insertion forces at the distal end of the needle when advancing the needle tip to the desired target. However, detecting tissue transitions at the distal end of the needle is difficult since the sensed forces are dominated by shaft forces. Disentangling insertion forces has the potential to substantially improve needle placement accuracy.We propose a collaborative system for robotic needle insertion, relaying haptic information sensed directly at the needle tip to the physician by haptic feedback through a light weight robot. We integrate optical fibers into medical needles and use optical coherence tomography to image a moving surface at the tip of the needle. Using a convolutional neural network, we estimate forces acting on the needle tip from the optical coherence tomography data. We feed back forces estimated at the needle tip for real time haptic feedback and robot control. When inserting the needle at constant velocity, the force change estimated at the tip when penetrating tissue layers is up to 94% between deep tissue layers compared to the force change at the needle handle of 2.36 %. Collaborative needle insertion results in more sensible force change at tissue transitions with haptic feedback from the tip (49.79 ± 25.51)% compared to the conventional shaft feedback (15.17 ± 15.92) %. Tissue transitions are more prominent when utilizing forces estimated at the needle tip compared to the forces at the needle shaft, indicating that a more informed advancement of the needle is possible with our system.

Accurate recognition of surgical tools is a crucial component in the development of robust, context-aware systems. Recently, deep learning methods have been increasingly adopted to analyse laparoscopic videos. Existing work mainly leverages the ability of convolutional neural networks (CNNs) to model visual information of laparoscopic images. However, the performance was evaluated only on data belonging to the same dataset used for training. A more comprehensive evaluation of CNN performance on data from other datasets can provide a more rigorous assessment of the approaches. In this work, we investigate the generalisation capability of different CNN architectures to classify surgical tools in laparoscopic images recorded at different institutions. This research highlights the need to determine the effect of using data from different surgical sites on CNN generalisability. Experimental results imply that training a CNN model using data from multiple sites improves generalisability to new surgical locations.

After surgery of the anterior cruciate ligament (ACL), unpredictable effects can occur in the patient’s gait patterns. Thus, patients often experience instability and uncontrolled buckling of the knee joint (“giving way” effect). The “giving way” effect is also difficult to measure due to the uncontrolled and rare occurrence. Within the KneTex project a smart textile-integrated sensor and actuator bandage system is developed to detect instability of the knee joint and actively support the rehabilitation process. Based on the sensor data the calculation of the knee angles as well as a recognition system is implemented. The bandage system can be controlled via a custom app which is developed as part of the project. This paper describes the recent developments and current status of the project. In addition, the results and steps for future work are described as well.

To engage in socio-emotional interactions, children with autism spectrum conditions (ASC) need support to understand and convey emotions. In our approach, a humanoid robot (Pepper, Softbanks Robotics) acts as a tutor for the child within autism care. The robot, equipped with multimodal sensor technology to acquire the emotional feedback of the child, stimulates the child to perform tasks, adapted to its current arousal state. By in-, or decreasing the difficulties of implemented training modules, the child can be given the appropriate task according to its emotional state. The child’s arousal is measured with different techniques implemented in and on the robot: emotion detection based on audio recordings of the speech signal and camera detected facial expressions, or heart rate. To this end, the remote Photoplethysmography (rPPG) signal from camera recordings of the subjects’ face is acquired. While its unintrusive measurement is an advantage, a major drawback for rPPG is its proneness to motion and light artefacts requiring de-noising steps. A wavelet transform based on log-Gabor wavelets and a filter bank with 32 filters was implemented. The signal was filtered with a prior filter and afterwards with a Markov chain in order to extract the underlying pulse rate. Within an initial study, five children were observed watching videos with different co-notated emotions. As reference for the heart rate (HR), a wristband (empatica E4) was used. The captured emotions of all subjects were annotated to identify low and high arousal parts and positive and negative emotions. Extracted HR from rPPG-data indicated a correlation with the annotated emotions.

Primary sclerosis cholangitis (PSC) predisposes individuals to liver failure, but it is challenging for radiologists examining radiologic images to predict which patients with PSC will ultimately develop liver failure. Motivated by algebraic topology, a topological data analysis - inspired framework was adopted in the study of the imaging pattern between the “Early Decompensation” and “Not Early” groups. The results demonstrate that the proposed methodology discriminates “Early Decompensation” and “Not Early” groups. Our study is the first attempt to provide a topological representation-based method into early hepatic decompensation and not early groups.

Fitting the data of an electrochemical impedance spectroscopy (EIS) typically requires manual estimation of the initial values before regression algorithms such as complex nonlinear least squares (CNLS) can be applied. This makes the success rate of the fitting dependent on the user input. Furthermore, the Randles circuit consists of parameters with substantially differing magnitudes (e.g. capacitors and resistors), which can also strongly affect the success rate of the fitting due to numerical effects. The aim of this work is to investigate methods addressing the described limitations of fitting. Therefore, a Python implementation performing a fit for EIS is benchmarked with an equivalent open source library. The examined implementation optionally includes the normalization of the parameter values, the standardization of the impedances and a pre-fit. Applying the same equivalent circuit without additional signal processing steps and with fixed initial values defined by midpoints of the value ranges, both implementations were able to fit 46.50% of the simulated database with different spectra. Applying the normalization of the parameter values (76.25%) or the same method with additional pre-fit (97.75%) lead to a significant improvement of the success rate. The standardization of impedances did not affect the success rate.

Bioimpedance analysis is a label-free and easy approach to obtain information on cellular barrier integrity and cell viability more broadly. In this work, we introduce a small, low-cost, portable in vitro impedance measurement system for studies where a shadow-free exposure of the cells is a requirement. It can be controlled by a user-friendly web interface and can perform measurements automated and autonomously at short intervals. The system can be integrated into an existing IoT network for remote monitoring and indepth analyses. A single-board computer (SBC) serves as the central unit, to control, analyze, store and forward the measurement data from the single-chip impedance analyzer. Various materials and manufacturing methods were used to produce a purpose-built lid on top of a modified 24-well microtiter plate in a “do it yourself” fashion. Furthermore, three different sensor designs were developed utilizing anodic aluminum oxide (AAO) membranes and gold-plated electrodes. Preliminary tests with potassium chloride (KCl) showed first promising results.

Analysing and fusing data from the medical devices of different disciplines (anaesthesiology and surgery) inside the operating rooms may promote awareness during surgical procedures. In this work, the changes of physiological parameters of patients undergoing laparoscopic gynaecology procedures were analysed. The statistical relationship between the intra-abdominal pressure and airway peak pressure was evaluated. Patients ventilated with pressure-controlled ventilation (PCV) and intermittent mandatory ventilation (IMV) were included. The results demonstrated that increasing the intra-abdominal pressure (IAP) resulted in increasing the airway peak pressure and decreasing the lung compliance. The Pearson correlation coefficient between the IAP and the airway peak pressure was 0.910 in PCV-patients and 0.952 in IMV-patients when changes of the ventilation settings were considered. Additionally, major hemodynamic changes included alterations in the mean blood pressure (MBP), where the MBP increased after insufflating the abdomen and decreased after abdomen deflation.

Analysis of the active ingredient distribution of medical devices is typically performed using Raman spectroscopy, a method that is fast and inexpensive [1]. In addition, it offers the advantage of non-destructive analysis without the need for special sample preparation. Assuming that all components are Raman-active and present in sufficient quantities, their distribution can be well represented. The drug distribution in dexamethasone-loaded polymer nonwovens was investigated in order to draw conclusions on the quality of the fleece batches and to make predictions for the release behavior. In the present study, dexamethasone (DMS), a glucocorticoid was used as the active ingredient. Qualitative and quantitative studies of the content of DMS in polymer films by means of Raman spectroscopy have already been carried out in the working group [2]. A representative square section was examined to describe the distribution of active ingredients. The required number of measurement points (spectra) was determined earlier [2].

In the scientific field, data acquisition using commercial or self-developed sensors is a necessity for many research activities. Data security and data privacy are important requirements in all types of IoT applications, especially in the medical context. IoTree42 is a powerful, OpenSource platform that closes the gap between cost-efficiency, the ease of use and digital sovereignty. The flexible and user-friendly design makes the platform ideal for the originally conceived research context. Nevertheless, it can be used for smaller budget-oriented setups and industrial applications with thousands of sensors and actuators likewise. The platform was developed with security in mind by minimizing interfaces and the use of stable software components and transport protocols. IoTree42 utilizes lightweight OpenSource software cast into a flexible backend, allowing the deployment not only on full-fledged servers but also on single board computers like the Raspberry Pi. Data is transmitted with low overhead via MQTT, a robust protocol optimized for machine to machine communication. The transmission technology between all components is freely selectable. Besides data acquisition, IoTree42 enables control of actuators and automation through interfaces with visual programming tools. This allows for remote interventions without requiring prior programming experience. The incoming data can conveniently be visualized in dashboards or easily be exported for further analysis. The network consists of a central server and satellites (gateways) arranged in a star topology. Sensors and actuators connect to the gateway. The initial setup of the server and the gateways is automated, well documented, and can be done within minutes. Thanks to the increasing usage and the contribution of the community, the pool of code examples for sensors is growing and lowers the entry hurdle for users without programming background. IoTree42 is a fast-growing, multi-user, hardware and cloud agnostic, easy to deploy, privacy compliant and competitive solution compared to commercial IoT alternatives.

Background: The study investigated if the disinfecting potential of Hypochlorous acid (HOCl) in suspensions are transferrable to in-air cleaning applications and to what extent aerosolized HOCl solutions can deactivate indoor microbial contaminations in-air at or below legal limits. Material and Method : For the liquid disinfection we used a standard suspension disinfection test protocol. For the in-air tests we conducted several experiments where aerosolized bacterial suspensions were injected into lab chambers preloaded with different HOCl gas concentrations. Results: In suspension experiments we found sufficient efficacies for all studied organisms at minimum concentrations of 200 ppm HOCl. The in-air measurement set-up allows to follow microbe deactivation by HOCl interaction. The deactivation rate increases with the HOCl concentration, and the values are highest for Gram-negative bacteria. Conclusion: We confirmed our hypothesis of the high disinfecting power of HOCl in-air at safe levels for populated indoor places. The investigated bacteria provide a model system for infectious particles, including enveloped viruses (to which Coronavirus belongs). These early results suggest that HOCl should be further evaluated as an air-cleaning method which may complement established concepts.

For the development, training, and validation of AIbased procedures, such as the analysis of clinical data, prediction of critical events, or planning of healthcare procedures, a lot of data is needed. In addition to this data of any origin (image data, bio-signals, health records, machine states, …) adequate supplementary information about the meaning encoded in the data is required. With this additional information - the semantic or knowledge - a tight relation between the raw data and the human-understandable concepts from the real world can be established. Nevertheless, as the amount of data needed to develop robust AI-based methods is strongly increasing, the assessment and acquisition of the related knowledge becomes more and more challenging. Within this work, an overview of currently available concepts of knowledge acquisition are described and evaluated. Four main groups of knowledge acquisition related to AI-based technologies have been identified. For image data mainly iconic annotation methods are used, where experienced users mark or draw depicted entities in the images and label them using predefined sets of classifications. Similarly, bio-signals are manually labelled, whereby important events along the timeline are marked. If no sufficient data is available, augmentation and simulation techniques are applied yielding data and semantics at the same time. In applications, where expensive sensors are replaced by low-cost devices, the high-grade data can be used as semantics. Finally, classic rule-based approaches are used, where human factual and procedural knowledge about the data and its context is translated into machine-understandable procedures. All these methods are depending on the involvement of human experts. To reduce this, more intelligent and hybrid approaches are needed, shifting the focus from the-human-in-the-loop to the-machine- in-the-loop.

Drug delivery systems (DDS) are suitable for controlled local drug release in order to ensure safety and effectiveness of medical treatment. The choice and characterization of biomaterials that can be used as a DDS is a challenging step in the administration of drugs. Novel 3D printing photopolymerization-based techniques create the possibility for designing individual, patient-tailored DDS. Poly(ethylene glycol) diacrylate`s (PEGDA`s) chemical and biological properties make it a suitable photopolymerisable resin for the creation of DDS. This study describes the in vitro release of the antiseptic drug chlorhexidine (CHX) from UV-cured PEGDA and its copolymers (butanediol diacrylate, pentaerythritol triacrylate and pentaerythritol tetraacrylate) samples. A substantial decrease in CHX release with increasing concentration of the copolymers in comparison to pure PEGDA was obtained only for butanediol diacrylate. For pentaerythritol triacrylate and pentaerythritol tetraacrylate only a tendency of decreased CHX release with increasing concentration was detected. Therefore, release profiles of the low molecular drug CHX from PEGDA samples could be modified by the addition of copolymers with a different number of acrylate groups and PEGDA can be considered as a promising candidate for the preparation of novel DDS.

Manufacturer-independent medical device interoperability has been strongly demanded by the clinicians and healthcare delivery organizations (HDOs), however has not been achieved in practice for decades. The ISO/IEEE 11073 Service-oriented Device Connectivity (SDC) family of standards constitutes a paradigm shift. This work supplements SDC with Device Specializations (DevSpecs) or Modular Specifications (ModSpecs): particular interoperability standards for high frequency (HF) surgical equipment and external control devices, such as foot or finger switches. They provide models to describe these devices in a service-oriented medical device system and modes of interaction with other network participants. Additionally, we contribute to the ISO/IEEE 11073-10101 nomenclature standard to provide semantic descriptions of the exchanged information. This is a key enabler for safe and effective medical device interoperability to support the caregivers and improve patient safety as well as clinical outcome.

Hydrophobic surfaces have gained a vast attention in different fields of biomedical applications over the last few years. Laser treatment has been shown as a promising technology for the generation of functional, inter alia, superhydrophobic surfaces. In this study a picosecond Yb-YAG laser was assigned for the generation of superhydrophobic characteristics on a steel alloy with application in surgical instrumentation. Regarding the ablation energy threshold of about 6 μJ for the given pulse width and laser beam characteristics and by assigning a suitable combination of microstructure kinematics and laser processing conditions a persistent hierarchical pattern is mapped over the laser-radiated surfaces. The average measured contact angle on the laserradiated surfaces was about 150°, which indicates their superhydrophobic properties.

Poly(N-isopropylacrylamide) hydrogels are a popular temperature sensitive biomaterial. Bulk hydrogels can be quickly and easily processed into microgels using a ball mill. However, there is no information on whether the mechanical milling process affects critical PNIPAM characteristics. In this work, we compare swelling and thermo-responsive properties for a series of PNIPAM gels in bulk and microgel forms.

For the generation of tailor-made polymer coatings on nanofibrous nonwovens plasma enhanced chemical vapor (PECVD) is a promising process, even for complex geometries. The plasma coatings can greatly improve their suitability for biomedical applications by optimising biocompatibility to the local needs, especially for cardiovascular disease treatments. Therein, wound healing and endothelialisation are important steps which are connected by a complex interaction. The monomers allylamine and hexamethyldisiloxane, as well as different process conditions were studied for the coating of nanofibrous thermoplastic silicone polycarbonate polyurethane (TSPCU) nonwovens. Aim of this study was to investigate the feasibility of plasma polymer coating under preservation of the nanofibrous morphological structure. Beside characterization of the nonwoven, biological evaluation with endothelial and fibroblast cells was performed. The prepared nonwoven samples support the feasibility of plasma coating under preservation of the nanofibrous structure. Also, different effects of the surfaces in contact with fibroblasts and endothelial cells could be observed.

High quality recording of neuronal activities and electrical stimulation require neurotechnical implants with appropriate electrode material. Iridium oxide (IrOx) is an excellent choice of material due to its biocompatibility, low electrochemical impedance, superior charge injection capacity, corrosion resistance, longevity, and electrochemical stability. Plasma enhanced atomic layer deposition (PE-ALD) and a suitable precursor, like (Methylcyclopentadienyl) (1,5- cyclooctadiene) iridium, could be a promising technique to produce highly conformal and performant IrOx-films at low temperatures and low costs. Various studies have reported the deposition of iridium oxide, but usually at very high temperatures. These processes are not suitable for polymer substrates and limit the use of such post-processing together with active implants. In this work the (Methylcyclopentadienyl) (1,5-cyclooctadiene) iridium(I) ((MeCp)Ir(COD)) precursor was used as a promising approach for depositing IrOx-films using low temperature PE-ALD. This precursor is normally used for chemical vapour deposition processes. First experiments were carried out on silicon substrates at deposition temperatures of 110 C°. The precursor was heated up to 75 °C and oxygen plasma was used as coreactant. The deposited films were analysed with EDX and AFM, showing a smooth surface and a promising ratio between the elements iridium and oxygen.

Some optically pumped magnetometer (OPM) sensors available for biomagnetic investigations have a linear range limited to +- 1 nT due to the specific properties of their open loop operation. In a two-layer magnetically shielded room of type Ak3b/Vacoshield Advanced with an external active compensation we studied how much sensor movement is allowed until amplitudes exceed the linearity range. Intentional movements were performed by a subject wearing an OPM-MEG sensor array. It was found that movements of 8 cm did yield non-linear amplitudes, but a reduction of the movement in half already preserves linearity. Despite movements, the heartbeat was found to generate a periodic signal, although the generating mechanism could not be identified so far.

The analysis of the arterial wall properties is vital in the prediction of stroke events and arterial hypertension in humans. Numerous researchers have experimented with several approaches to model arterial vessels and to analyse their biomechanical behaviour for many years now. Our study is focussed on image processing of peripheral arterial cross sections to detect and isolate the distinct layers. These boundaries will enable the creation of FEM models for further analysis of arterial wall properties. In a clinical setting, it facilitates doctors to identify the optimum pressure that can be applied to the artery for the treatment of stenosis without damaging the morphology of the blood vessels. This paper aims at distinguishing the various layers of arterial walls from images by minimizing human intervention. Cross section images of arteries from various sources were collected[10][11]. The boundaries from the image were obtained using image processing techniques of MATLAB(R2021a). The approach identified was to convert the input RGB images to grayscale, thresholding and applying morphological operators to delineate the Intima, Media, and Adventitia. These regions of interests (ROI) were then superimposed to generate an image with differentiated boundaries and void of unnecessary noise and inhomogeneity. This approach gave us an insight of the differences in various methods of boundary detection and to infer the optimum approach for accurate demarcation of boundaries of the three layers of arterial walls. It paves a pathway for forward modelling and to perform detailed FEM analysis in in-vitro diagnostics. In a nutshell, it was observed that the edge detection procedure implemented could be used for healthy and stenotic arteries. Further studies must be conducted to test the efficiency across a wide range of images and hence generalise its usage. Upon satisfactory boundary detection, forward modelling could be performed using the identified geometric forms.

Soft lithography, a tool widely applied in biology and life sciences with numerous applications, uses the soft molding of photolithography-generated master structures by polymers. The central part of a photolithography set-up is a mask-aligner mostly based on a high-pressure mercury lamp as an ultraviolet (UV) light source. This type of light source requires a high level of maintenance and shows a decreasing intensity over its lifetime, influencing the lithography outcome. In this paper, we present a low-cost, bench-top photolithography tool based on ninety-eight 375 nm light-emitting diodes (LEDs). With approx. 10 W, our presented lithography set-up requires only a fraction of the energy of a conventional lamp, the LEDs have a guaranteed lifetime of 1000 h, which becomes noticeable by at least 2.5 to 15 times more exposure cycles compared to a standard light source and with costs less than 850 C it is very affordable. Such a set-up is not only attractive to small academic and industrial fabrication facilities who want to enable work with the technology of photolithography and cannot afford a conventional set-up, but also microfluidic teaching laboratories and microfluidic research and development laboratories, in general, could benefit from this cost-effective alternative. With our self-built photolithography system, we were able to produce structures from 6 μm to 50 μm in height and 10 μm to 200 μm in width. As an optional feature, we present a scaled-down laminar flow hood to enable a dust-free working environment for the photolithography process.

High-Intensity Focused Ultrasound (HIFU) is an alternative tumour therapy with the ability for non-invasive thermal ablation of tissue. For a safe application, the heat deposition needs to be monitored over time, which is currently done with Magnetic Resonance Imaging. Ultrasound (US) based monitoring is a promising alternative, as it is less expensive and allows the use of a single device for both therapy and monitoring. In this work, a method for spatial and temporal US thermometry has been investigated based on simulation studies and in-vitro measurements. The chosen approach is based on the approximately linear dependence between temperature and speed of sound (SoS) in tissue for a given temperature range. By tracking the speckles of successive B-images, the possibility of detecting local changes in SoS and therefore in temperature is given. A speckle tracking algorithm was implemented for 2D and 3D US thermometry using a spatial compounding method to reduce artifacts. The algorithm was experimentally validated in an agar-based phantom and in porcine tissue for temperature rises up to △ 8°C. We used a focusing single element US transducer as therapeutic probe, a linear (/matrix array) transducer with 128 (/32∙32) elements for imaging and thermocouples for validation and calibration. In all experiments, both computational and in-vitro, we succeeded in monitoring the thermal induced SoS changes over time. The in-vitro measurements were in good agreement with the simulation results and the thermocouple measurements (rms temperature difference = 0.53 °C, rms correlation coefficient = 0. 96).

Focal adhesion clusters (FAC) are dynamic and complex structures that help cells to sense physicochemical properties of their environment. Research in biomaterials, cell adhesion or cell migration often involves the visualization of FAC by fluorescence staining and microscopy, which necessitates quantitative analysis of FAC and other cell features in microscopy images using image processing. Fluorescence microscopy images of human umbilical vein endothelial cells (HUVEC) obtained at 63x magnification were quantitatively analysed using ImageJ software. A generalised algorithm for selective segmentation and morphological analysis of FAC, nucleus and cell morphology is implemented. Further, a method for discrimination of FAC near the nucleus and around the periphery is implemented using masks. Our algorithm is able to effectively quantify different morphological characteristics of cell components and shows a high sensitivity and specificity while providing a modular software implementation.

Noise level estimation plays an important role in many applications of signal and image processing, like denoising, compression and detection. Recently, deep neural networks have also been increasingly used for this purpose. In this paper, we develop an effective algorithm of noise level estimation of ECG segments based on trained denoising autoencoder (DAE) with a statistical thresholding method. An important observation is that a well-trained DAE model provides a clean representation of the corrupted training dataset. Two identical cascaded trained DAE models are considered to estimate the statistical properties, e.g., mean and standard deviation, from the trained DAE outputs after applying noise free aligned and jittered training dataset respectively. Two statistical thresholds are calculated from these statistical properties to classify whether the ECG segment is noise-free or jittered or noisy segment. The accuracy of the proposed method is quite promising in classifying and estimating unknow noise level.

Signal processing, pattern recognition as well as modelling and simulation require a large amount of reference data, both for the development of new methods and for their evaluation. Depending on the application, the availability of databases is rather low. In the field of biosignal processing with a focus on the functionalization of furniture for nursing and hospital facilities, a database from a motion capturing system (MoCap), and a method to generate averaged human motion sequences was presented in subsequent works by our research group. Evaluations revealed that the averaged motion sequences partly contain artifacts caused by the averaging and thus are not directly usable. To use the averaged motion sequences e.g., in simulation tasks, this paper presents an extension with kinematic methods to combine averaged motion sequences and to suppress and thus optimize inappropriate motion artifacts by error correction. To check whether the results are usable after the processing steps, four evaluation criteria are proposed. The evaluation of the resulting motion sequences shows that sequences are generated which do not fully correspond to human motion sequences but are well suited for simulation tasks.

The loading situation of the aortic valve is complex, complicating the identification of innovative approaches for heart valve leaflet materials, e.g. for transcatheter aortic valve implantation (TAVI). Materials engineering experiments allow for screening of materials but especially for durability testing, the consideration of physiological loads is vital/critical for the suitability-assessment of innovative leaflet materials. For this reason, a framework structure for the testing of leaflet materials in physiological loading (TAVI-Mimic) was developed. The exemplary use case for the TAVI-Mimic was a test for calcification propensity of pericardium during durability testing. The TAVI-Mimic was designed as a fourparted frame, based on previous work of our group. The leaflet material can be attached between inner and outer shells without sewing. In a second step, the TAVI-Mimic was optimized regarding radial load-deformation in comparison to a commercial TAVI by means of finite element analysis (FEA) and hydrodynamic characterization in a pulse duplicator system. Mechanical properties dependent on water uptake of different materials for 3D-printing of the TAVI-Mimic were investigated. After optimization, TAVI-Mimics were equipped with glutaraldehyde-fixated pericardial tissue and prototypes were calcified by using a heart valve durability tester and a metastable calcification-liquid, developed in earlier studies. The development of the TAVI-Mimic using FEA and experiments was successful, leading to a radial load dependent deformation of 0.6 mm which correlates with commercial TAVI. Two methacrylic photopolymers were identified for 3D-printing of the TAVI-Mimic and prototypes attached with pericardial tissue were manufactured. Pericardium TAVI-Mimics were calcified in vitro for one week and an average calciumphosphate precipitate of 0.34- 0.54 mg/cm² was measured. The optimization of the TAVR-Mimic led to an improved load-dependent behaviour compared to a commercial prosthesis while testing. The calcification method, combining the TAVI-Mimic, the metastable calcification solution and the durability tester enabled a successfully calcification of pericardial tissue, approaching the in vivo situation.

The determination of respiratory parameters via respiration induced surface movements of the upper body has been the subject of research for many years. The displacements of 102 motion capture markers were evaluated in this study in terms of their information content with respect to the tidal volume recorded in parallel using a spirometer. Independent of the breathing types (spontaneous breathing, abdominal breathing, or chest breathing), the number and the location of sensors in a smart shirt to obtain tidal volume information was determined. Only 9 of 102 sensors were sufficient to obtain breathing volume information.

The Urethra is a long tubular structure in the genitourinary tract and serves important functions. Researchers have experimented with some approaches to model the urethra and to analyse its biomechanical properties. However, experiments to model the in-vivo behaviour of urethra with strictures is not thoroughly explored. To analyse the in-vivo urethral properties and specifically for supporting treatment of strictures, a new inflatable sensor-actuator system is being developed. The capabilities of this sensor shall be evaluated in simulations which require appropriate modelling of the human male urethra with strictures. This forms a part of the identification procedure for a variety of urethra conditions and geometries, which in turn forms a basis for inverse modelling. As an initial simplified approach, an axisymmetric Finite Element model was generated that resembled the urethra incorporating a stricture region. An ideal actuator with sensor elements exerting a pressure on inner wall of this urethra was simulated. Three circumference measurement zones within the sensor height (top surface, centre and bottom surface) were implemented. The resulting pressure-extension (circumferential) responses were determined at these measurement zones. The sensor was placed at different lengths within this urethral tube and inflated and the pressure-extension responses were noted. It was found that depending on the position of the sensor-actuator, the extension of tissue can vary. The possible factors for this variation were the finite length of the actuator as well as the influence of tissue properties around the measurement zones. This is important information for the interpretation of sensor data to be gained by the current development. It was possible to generate datasets based on an ideal sensor model, that proved helpful in the evaluation of biomechanical tissue properties in healthy and stricture conditions. This indicates simulations are a versatile and prospective way to test new sensors prior to real experiments.

Being non-invasive, cheap and widely available, the 12-lead electrocardiogram (ECG) is a standard method to assess cardiac function. Still, its reliable interpretation requires specialized knowledge and experience, rendering a second opinion valuable. We evaluated the performance of machine learning based classification of 11,705 healthy and bundle branch block 12-lead ECGs from 3 open databases. For each lead of the ECG signal, a representative QRS-complex template was extracted automatically. Principal component analysis (PCA) was applied to the concatenated, normalized and rescaled QRS signals to reduce their dimensionality. Multilayer perceptron and support-vector machine classifiers were trained using the principal components of weighted and non-weighted QRS template signals as input data. Classifiers achieved F1 scores between 0.92 and 0.96 on the test set for different input configurations. Anomaly based weighting slightly improved the performance of the classifiers. Neither class-wise PCA for feature extraction nor adding information on sex, gender and electrical heart axis to the input data yielded considerable improvement of the F1 scores. The achieved classification accuracy is similar to deep learning classifier performances and should generalize robustly to other ECG datasets. Our results suggest that this simple and well interpretable approach based on morphological signal characteristics is suitable for automatically and non-invasively identifying bundle branch block pathologies in clinical or smart electronics contexts.

Many people are familiar with the feeling of instability, pain, or subsidence in the knee joint after a knee injury. There are many different methods for examining the knee, such as the drawer test or the Lachman test [1], before and after surgery. While these tests can be used in short term and provide useful results, motion capture systems can be used as an alternative measurement method, almost as a substitute in longer term. These include marker-based or mechanica l systems, which achieve good measurement results but are expensive and inflexible. For this reason, this paper presents a mobile, easy-to-use motion and easy expandable capture system using a low-cost IMU-based development system. The modular design of the system allows it to be adapted to each body region with simple adjustments. However, the present work focuses on applications for capturing human motion sequences and deriving three joint angles of the lower extremities to detect malposition.

Arteries are vessel structures that serve vital function of transportation of blood to different parts of the body. Researchers have experimented with some approaches to model the arterial behaviour and to analyse its biomechanical properties. To analyse the in-vivo arterial properties, at Furtwangen University an inflatable sensoractuator system is being developed, which provides the basis for a decision support system for vascular surgeons. The capabilities of this sensor shall be evaluated in simulations which requires appropriate modelling of the arteries. The inverse problem, i.e. how to efficiently identify arterial wall properties from sensor readings is targeted. A histology motivated 3D artery model was implemented in FEM using COMSOL (v5.5). The geometry of one model was based on a cross section of a real artery. The second model was axisymmetric and of equal dimensions with respect to volume, layer thickness etc. A biomechanical pressure-stretch analysis was performed applying an inflating pressure inside the walls of the vessels. Stretch in different areas of the first model was evaluated and the circumferential strain was compared to the axisymmetric model. The results show variation of strains within the segments of the first model of upto 10 percent. In addition, its outer wall circumferential stretch was found to be 10 percent lower compared to the axisymmetric setup. This comparison sheds light upon whether a simplification of arterial models is possible, without loss of accuracy in the context of the novel sensor evaluation. It provides useful information whether e.g. standardizing vessel structures to axisymmetric models will still provide results within allowable tolerance limits. Simulations proved useful to evaluate different vessel model formulations in the context of arterial diagnostics.

A two-step method for investigating the stiffnesses and failure loads of vertebral bodies is presented. It is based on the representation of the spine in a global finite element model of structural elements. A detailed model of two interconnected vertebrae and an intervertebral disc was implemented. Calculations on this model are shown which illustrate the dependencies of structural stiffnesses on vertebral body properties. These structural stiffnesses can then be used to represent the spine in the global model.

Patients undergoing transcatheter aortic valve replacement (TAVR) may suffer severe clinical complications, caused by paravalvular leakage (PVL) which is defined as leakage between TAVR and aortic annulus. PVL is often facilitated by a severely calcified annulus. This limits the expansion of a self-expandable TAVR stent. To assess TAVR performance in terms of leakage, measurement of regurgitation fraction in a pathophysiological annulus is recommended according to ISO 5840. For this purpose, a configuration of a circular annulus with a calcification nodule has been proposed in the recently published ISO 5840. The impact of the proposed pathophysiological annulus model on the expansion of self-expandable TAVR stents and on the regurgitation fraction was investigated in this study. For this purpose, two commercially available selfexpandable TAVRs (Evolut R and Portico) were implanted in a calcified annulus model. Circular expansion of the TAVR stents was investigated based on μCT scans of the implanted TAVR. The calcification-induced area in which retrograde flow can occur during diastole was detected. These results were then compared with the experimentally determined regurgitation fraction obtained from pulse duplicator tests. The results of the μCT scans showed a continuous leakage area in the region of the annulus for the Evolut R compared to a locally larger leakage area of the Portico, which, however, reattaches to the annulus in the distal inflow region. The hydrodynamic measurements confirmed a smaller leakage in the pathological annulus for the Portico. In summary, it can be assumed that a continuous leakage area in the TAVR stent inflow region encourages the PVL of TAVR.

Robot-based service platforms are currently establishing themselves as new and affordable variants for supporting care in elderly, retirement and nursing homes. Many are open multifunctional platforms, which can potentially be integrated into such environments, if the necessary infrastructure is available. Furthermore, many services can be realized on these platforms, which can be used to foster distant interactions between inhabitants and care-providers, while simultaneously keeping up the quality of life of the inhabitants. Open mobile robotic platforms allow the extension with adequate new sensors. To detect infectious diseases of residents and healthcare-professionals, optical sensors can be used for the assessment of vital data such as heartrate and heartrate variability, respiratory rate, SpO2 or temperature. Additionally, you can consider demographic data (age, gender, constitution) of the observed person for the optical assessment, i.e. obtained by facial image analysis. As these mobile platforms are also equipped for telepresence, in case of detecting an infected person, these systems support video conferencing with their built-in cameras and microphones. Finally, the interaction with the electronic care record is necessary to upload all acquired vital data and further relevant information. All the named technologies have been under investigation in the past years and are currently moving from laboratory settings to real-world scenarios. Nevertheless, the smooth integration of all components into one system architecture in combination with (AI-based) data analysis are still open issues.

Mitral regurgitation (MR) is the second most frequent indication for heart valve surgery and catheter interventions. According to European and US-American guidelines, transcatheter mitral valve repair in general and transcatheter edge-to-edge repair (TEER) in particular may be considered as a treatment option for selected high-risk patients. However, the biomechanical impact of TEERdevices on the mitral valve (MV) has not yet been fully understood. To address this problem, a 3D-Fluid-Structure Interaction (FSI) framework utilizing non-linear Finite Element Analysis (FEA) for the MV apparatus and Smoothed Particle Hydrodynamics (SPH) for the pulsatile fluid flow was developed and validated against in vitro data. An artificial MV-model (MVM) with a prolapse in the A2-P2 region and a custom-made TEER device implanted in the A2-P2 region were used for the in vitro investigations. In accordance with ISO 5910, projected mitral orifice areas (PMOA), flow rates as well as atrial and ventricular pressures were measured under pulsatile flow conditions before and after TEER device implantation. For the FSI-model, the MVM geometry was reconstructed by means of microcomputed tomography in a quasi-stress-free configuration. Quasi-static tensile test data was utilized for the development of linear- and hyperelastic material models of the chordae tendineae and leaflets, respectively. The fluid flow was modelled assuming an incompressible, homogenous Newtonian behaviour. Time-varying in vitro transmitral pressure loading was applied as a boundary condition. In vitro investigations show that TEER device implantation in the A2-P2 region effectively reduces the regurgitation fraction (RF) from 55 % to 13 %. Moreover, the comparison of experimental and numerical data yields a deviation of 2.09 % for the RF and a deviation of 0.40 % and 6.47 % for the maximum and minimum PMOA, respectively. The developed FSI-framework is in good agreement with in vitro data and is therefore applicable for the characterization of the biomechanical impact of different TEER devices under pulsatile flow conditions.

Paravalvular leakage (PVL) has a crucial impact on clinical outcomes of transcateheter aortic valve replacements (TAVR), especially the mortality increases dramatically with high-grade PVL. Furthermore, the calcification of the aortic annulus has a decisive influence on the PVL of TAVR. Therefore, we developed a technical model of a calcified aortic annulus and used it for the investigation of PVL in steady-state back-flow conditions. We investigated an Evolut PRO (Medtronic, Minneapolis, MN, USA), implanted the TAVR at different depths in the aortic annulus model ranging between 0 mm and -6 mm and characterized PVL in steady-state retrograde flow from 0 mmHg up to a maximum achievable pressure. The used test bench and detailed test method was described in previous studies. The aortic annulus model exhibits three elevations symmetrically distributed around the circumference. Depending on the degree of calcification the elevations reached 1 mm to 3 mm into the lumen. For the Evolut PRO bioprosthesis, a decreasing PVL was measured with increasing implantation depth. At an implantation depth of 0 mm (inflow of TAVR and annulus model at same height) maximum PVL was measured. Minimum PVL was measured at a height of -6 mm. Furthermore, even a small calcification of 1 mm led to a large increase in PVL. This trend continued with increasing height of the calcification. The maximum regurgitation of (2,025.21 ± 12.47) ml (n = 3 measurements) was measured at a pressure of 6 mmHg in the annulus model with 3 mm calcification. A test method to quantify PVL depending on annular calcification was successfully developed. Additionally, the influence of implantation depth on PVL was characterized. Due to the technical operating principle of the test bench, only a limited increase in pressure was possible when large PVL occurred. In this respect, the test bench must be optimized in the future.

Electrospinning is used for producing nonwovens for medical polymer-based implants, such as prosthetic valves or covered scaffolds. In this study, nonwovens for prosthetic venous valves are investigated regarding their morphology and mechanics in physiological medium. Spinning molds were developed based on previous venous valve leaflet designs, 3D printed in different sizes and covered with electrospun nonwovens. Samples were stored in a physiological 0.9% saline at 37°C to investigate the influence of fiber rearrangement and swelling in medium for several weeks. Two different nonwovens of thermoplastic silicone-based polycarbonaturethane (TSPCU) were compared. Tensile test results show that storage in medium has a relevant influence on the mechanical properties. SEM images of TSPCU show substantially increased fiber diameters after 8 days stored in medium. After detaching the valve leaflet nonwovens from the molds, shrinkage of the material of approximately 12% was detected. A suitable valve size could be identified for joining with the stent structure into an interventional prosthetic venous valve. The results demonstrate the influence of storage conditions on the morphological and mechanical properties of electrospun TSPCU nonwovens. For development and dimensioning of venous valve leaflets, this change in mechanical behavior and possible shrinkage of the material has to be considered.

The assessment of particulate matter generated during a simulated use is requested by international standards for approval of cardiovascular devices. The particle suspension was generated during a simulated use test procedure of a commercially available coronary stent delivery system and analyzed by Dynamic image analysis (DIA) using FlowCAM 8000. Four parameters were analyzed: the diameter, the aspect ratio value, the circularity and the intensity. The results revealed that 53% of the particles were larger than 25 μm. The particle shapes were inhomogeneous with a wide ranging aspect ratio of 0.03 to 0.95. Around 83% had an intensity value of more than 140 grey scales and appeared translucent.

Biomaterial research efforts focus on the development of biomaterials that mimic the natural extracellular environment. In addition, different strategies are applied to render materials for blood-contacting devices nonthrombogenic through surface modifications that would suppress activation of platelets, coagulation and the complement system. A confluent thin layer of endothelial cells lines all blood vessels and produces factors responsible for inhibition of coagulation, thrombosis and neointimal hyperplasia. Thus, the ability to rapidly form a healthy endothelium upon implantation represents a desired property of biomaterials used for cardiovascular devices. In this study we used advanced in vitro methods to investigate the biocompatibility of a biodegradable and a permanent electrospun nanofiber fabric, poly-L-lactic acid and polycarbonate-based silicone elastomer respectively, with the focus on endothelialization and hemocompatibility.

It is commonly accepted that hemodynamic situation is related with cardiovascular diseases as well as clinical post-procedural outcome. In particular, aortic valve stenosis and insufficiency are associated with high shear flow and increased pressure loss. Furthermore, regurgitation, high shear stress and regions of stagnant blood flow are presumed to have an impact on clinical result. Therefore, flow field assessment to characterize the hemodynamic situation is necessary for device evaluation and further design optimization. In-vitro as well as in-silico fluid mechanics methods can be used to investigate the flow through prostheses. In-silico solutions are based on mathematical equitation’s which need to be solved numerically (Computational Fluid Dynamics - CFD). Fundamentally, the flow is physically described by Navier-Stokes. CFD often requires high computational cost resulting in long computation time. Techniques based on deep-learning are under research to overcome this problem. In this study, we applied a deep-learning strategy to estimate fluid flows during peak systolic steady-state blood flows through mechanical aortic valves with varying opening angles in randomly generated aortic root geometries. We used a data driven approach by running 3,500 two dimensional simulations (CFD). The simulation data serves as training data in a supervised deep learning framework based on convolutional neural networks analogous to the U-net architecture. We were able to successfully train the neural network using the supervised data driven approach. The results showing that it is feasible to use a neural network to estimate physiological flow fields in the vicinity of prosthetic heart valves (Validation error below 0.06), by only giving geometry data (Image) into the Network. The neural network generates flow field prediction in real time, which is more than 2500 times faster compared to CFD simulation. Accordingly, there is tremendous potential in the use of AIbased approaches predicting blood flows through heart valves on the basis of geometry data, especially in applications where fast fluid mechanic predictions are desired.

Continuous adaptation of international standards for medical devices requires recurrent modification of test benches. A universal control unit using an open software environment is presented to simplify the maintenance of various test benches. Our developed control unit is equipped with a Raspberry Pi 4, with standard communication interfaces and application-specific electronic assemblies. The software is based on Node-RED, a browser-based editor. A measuring setup was adapted for a flow perfusion system. Our control unit simplifies the handling of the flow perfusion system by controlling a hydraulic pump and all required valves. A software programmed PID algorithm adapts the speed of the pump to adjust the pressure automatically. Actuators like proportional pinch valves are handled to control volumetric flow and pressure within the circulation. Consequently, the user directly observes changes inside the system. The measured data are stored and are available for documentation.

As augmented reality devices become more available, collaborative work in the augmented space is expected to increase. Knowing the position of participants allows for realistic interaction rather than passive participation. The HoloLens 2’s camera and IMUs fuse data to locate the device. In this paper, accuracy and repeatability of the HoloLens2 position finding was analysed to provide a quantitative measure of pose repeatability and deviation from a path while in motion. Deviation from a circular path was found to be below 5 mm per 628 mm travelled.

Polymer materials can be manufactured with high reproducibility and do offer the potential for chemical modification. This enables matrix property modification and fine-tuning of several material characteristics, such as tissue-implant interaction, inflammatory potential or susceptibility to biofilm formation. Whereas manufacturing protocols are crucial for the resulting material properties, also the evaluation in terms of performance and safety has to be considered. Regarding this, both, temperature and composition of test medium may affect the physicochemical properties of implant materials. The present study addresses the influence of test medium compared to dry test conditions, each at two different temperatures, on the mechanical properties of elastomeric film and nonwoven materials.

Titanium represents an important biomaterial for implantable medical devices. During medical device manufacturing by means of welding, implant structures are partially exposed to high temperatures. Additionally, active implants such as pacemakers can heat up during operation. Therefore, numerical studies of heat propagation within titanium structures represent an essential tool to assess functionality and safety of medical devices. The current study focusses on the development of a method for experimental validation of numerical heat transfer analysis of biomaterials such as titanium. Numerical heat transfer analysis was performed using the software Abaqus. A finite-element model was established including material properties such as density, thermal conductivity und specific heat capacity. Temperature distribution among a locally applied thermal load was calculated. Furthermore, effects such as convection were considered. For validation, an experimental setup was implemented according to the numerical calculation using a local heating tool. Heat propagation in the sample was determined, respectively. Radiation-based heat determination was performed using an infrared thermographic camera aligned parallel to the sample surface. Contact-based heat determination was performed using thermocouples fixed to the surface at defined distances from the point of local heat input. For evaluation of numerical and experimental results, temperature- time curves were compared for five distinct measuring points, respectively. While infrared thermography offers the advantage of non-contact measurements, difficulties may arise from the definition of correct emissivity and challenging sample surface characteristics, such as metallic reflectance and surface texture. The thermocouple-based temperature measurement shows a high sensitivity to local temperature changes, but it is not always suitable due to the influence on the sample by thermocouple fixation. Infrared thermography and thermocouple based temperature measurements represent suitable procedures for experimental validation of numerical heat transfer analysis of titanium. An individual decision for the most suitable method must be made considering the specific sample and its further application.

The number of software products in the field of health and medicine increases excessively. Self-tracking, fitness, health advice, dose calculation, and analysis of physiological data - apps are popular and commonly used. Some of the products are affected by the medical device regulation, consciously or unconsciously. For young entrepreneurs, it is difficult to recognize and understand the distinction between software as medical product and health software products. A product-related orientation guide could help start-ups to understand the difference and to find the right strategy for placing their product on the market. All relevant information were collected and analyzed. Input came from a systematic literature research, an evaluation of the CHARISMHA study and a consideration of the medical device regulation as well as the relevant standards (e.g. IEC 62304 and IEC 82304). The guide was adapted to the particular situation of start-ups and microenterprises, inter alia having regard to missing regulatory affairs competences. A product-related orientation guide for young entrepreneurs was developed, which helps to identify whether products are affected by the medical device regulation or not. The guide uses a few simple questions, and it allows the entrepreneurs to estimate the related effort. Furthermore, it provides an overview of relevant standards and needed competences for the chosen path. The guide brings certainty and can help young entrepreneurs to identify the upcoming workload at an early stage. Due to the decision tree concept, it is easy to adapt the product-related strategy and see the consequences. This might help entrepreneurs to find the right strategy for placing their product on the market. Nevertheless, field testing with the guide is necessary to validate its benefit.

The assessment of the coating integrity of cardiovascular implants such as catheters and stent systems is of crucial importance for device approval. Released particles may represent a potential health risk for the patients. Thus, an analysis of the particles released at simulated in-vivo conditions depending on their size and number is required by international standards (ISO, ASTM) as well as national authorities. In this study, an automated test bench for online particle measurement is presented. For software controlled automation, sensor data transmission and solenoid valves were implemented. A user interface was created for setting test parameters and data recording. The setup was validated by investigating standard particles as well as those released during the simulated application of non-industrial coated balloons. The measurement data were compared with results generated using the previous manual test routine. The results show an improvement in the reproducibility of the measurements, which can be attributed to the simplified handling for the user.

Electrospun nonwovens are widely applied in biomedicine and various other fields. For control of the manufacturing process and quality assurance Scanning electron microscopy (SEM) imaging is one standard practice. In this study, statistical datasets of 60 SEM images of three nonwoven samples were evaluated using Gaussian fit to obtain numerical results of their fiber diameter distributions. The question of how much effort is required for acceptable imaging and processing is being discussed. As determined here, for reliable statistics, a minimum surface area of the nonwoven has to be evaluated. The fiber diameter should be in a range of approximately 2 - 3% of the edge length of the square equivalent of the evaluated image area, using sufficiently magnified SEM images, in which the fiber diameter is imaged over at least 30 pixels.

The novel concept of stenting the Eustachian tube was established to provide an effective and safe therapy of Eustachian tube dysfunction. Biodegradable polymer stents are being developed to restore impaired tube function. As the supporting effect may be required for different time periods, PLA-co-PEG copolymers, PLLGA, PDLLA and PDS, having shorter degradation times compared to PLLA, were evaluated as potential stent materials. Since tensile tests and thermal analyses of solvent cast films from PLA-co-PEG copolymers showed comparable properties to PLLA, stent samples were manufactured from these materials. Mechanical stent testing revealed an increase of elastic recoil and slight decrease of collapse pressure compared to PLLA. In a short term accelerated degradation study a considerable percentage molar mass reduction and an increasing degree of crystallinity depending on PEG content was found. Based on the results obtained, the tested polymers offer a promising, faster degradable alternative to the established stent material PLLA.

We present a thermal processing device, equipped for local heating and welding of biomaterials, e.g. electrospun nonwovens or polymer films. Depending on the application and choice of materials, we aim to determine appropriate processing parameters for permanent and non-permanent welds. Due to the modular device structure, applicationspecific heating tools can be used. Process temperature up to 250 °C is feasible. In this concept study, a welding tool with interchangeable small-size heating tips of different geometric shapes were investigated regarding their suitability for welding of polycarbonate urethane based silicone elastomer nonwovens. Generated welds were examined by imaging techniques and tensile testing. Regardless of the tip shape, welds were generated withstanding forces up to 3.0 N and 4.3 N for tensile and shear loading, respectively.

One of the key aspects in the development of novel implants is to find suitable materials and understand the mechanisms that occur when a material is exposed to the tissues of the human body. These mechanisms are commonly referred to as the biocompatibility of the material. A better understanding of the tissue-material interactions becomes more urgent as biomaterials are used in wide-ranging applications like modern medical devices. However, most biomaterials used for implants are opaque, resulting in difficulties for the microscopical evaluation during in vitro biocompatibility testing. Particularly, cell morphology and adhesion capabilities of cells can provide insights in the interactions between implant materials and tissues at the implant site. To improve our capabilities in biocompatibility testing of novel biomaterials we applied a new method to quantitatively assess cellular parameters on opaque samples using fluorescence microscopy and bio-image analysis.

Paravalvular leakage (PVL), defined as the leakage between the aortic annulus and a transcatheter aortic valve replacement (TAVR), is verifiably associated with short- and long-term clinical outcome, especially with increased mortality. Therefore, with the ambition to reduce or even prevent PVL of next generation TAVR, it is necessary to extend the hemodynamic understanding of PVL. This study presents an in vitro flow measurement method to localize PVL during hydrodynamic characterization of TAVR and furthermore presents different design features, socalled outer skirt, to reduce PVL. Particle image velocimetry (PIV) measurements were performed for flow field assessment during hydrodynamic characterization of TAVR. Additionally, two different sealing concepts were developed to reduce PVL. The skirts were manufactured from polymeric-nonwoven and sued to pericardium-based TAVR-prototype. The prepared TAVR-prototypes were then deployed in a pathophysiological model of the aortic root with a calcification nodule of 2 mm according to ISO 5840:2021. To assess PVL, the flow field and the regurgitation volume was measured. The PIV measurements showed a clearly visible leakage jet between the TAVR-prototypes without skirt and the pathophysiological aortic annulus model. Jet velocities of up to 0.5 m/s were measured depending on presence or configuration of a PVL-preventing skirt. When implanted in the physiological annulus model without calcification nodule, PVL was hardly recognizable. The regurgitation volume of a TAVR-prototype without skirt at 5 l/min was 36.26±1.89 ml (n = 10). The developed and manufactured polymeric-nonwoven skirts reduced PVL from 37.67±1.17 ml to 18.36±1.8 ml (n = 10, TAVR-skirt-design1) and from 46.97±1.07 ml to 17.85±1.29 ml (n = 10, TAVR-skirt-design2) at 5 l/min. The localization of PVL during hydrodynamic characterization by means of PIV was successful. The sealing concepts developed in this work were very effective and led to a PVL-reduction of the tested TAVR prototypes of about 50% to 70%.

Sterilization processes ensure sterility of drug delivery systems, but may negatively affect the properties of biomaterials and incorporated drugs by changing their physical, chemical, mechanical properties and drug release behaviour. Therefore, it is important to investigate their influence. In this study, the influence of ethylene oxide (EtO) sterilization on the drug loading and release behaviour of incorporated Diclofenac (DCF) in a Poly-L-lactide (PLLA) coating and Dexamethasone (DMS) in the silicone carrier is presented. Silicone samples containing DMS were coated with PLLA containing DCF varying in layer thickness (5, 10, and 20 μm). Half of the samples underwent EtO sterilization, the other half was not sterilized. All un-/sterilized sample surfaces were in view of the morphology and hydrophilicity examined. Furthermore, in vitro release studies of DMS and DCF were conducted. The sterilized sample surfaces showed no morphological and hydrophilicity changes. The DCF and DMS loadings were similar for the sterile and untreated samples. This also applied to the in vitro DMS release profiles apart from the end of the studies where slight differences were evident. The results indicate that both drugs loaded in the polymer coating and the silicone were not impaired by the sterilization process. Thus, EtO sterilization appears suitable for DMS containing silicone and DCF incorporated PLLA coatings as a dual drug delivery system.

Morphological prior information incorporated with the discrete cosine transformation (DCT) based electrical impedance tomography (EIT) algorithm can improve the interpretability of EIT reconstructions in clinical applications. However, an outdated structural prior can yield a misleading reconstruction compromising the accuracy of the clinical diagnosis and the appropriate treatment decision. In this contribution, we propose a redistribution index scaled between 0 and 1 to quantify the possible error in a DCT-based EIT reconstruction influenced by structural prior information. Two simulation models of different tissue atelectasis and collapsed ratios were investigated. Outdated and updated structural prior information were applied to obtain different EIT reconstructions using this simulated data, with which the redistribution index was calculated and compared. When the difference between prior and reality (the redistribution index) became larger and exceeded a threshold, this was considered as an indicator of an outdated prior information. The evaluation result shows the potential of the redistribution index to detect outdated prior information in a DCT-based EIT algorithm.

Electrospinning (ES) is a suitable and cost effective method to mimic the chemical composition, morphology, and functional surface of natural tissues, for example of the nervous, dermal, vascular, and musculoskeletal systems. This technique is a versatile tool to obtain tailored fibrous scaffolds from various polymer materials. By varying the diameter, porosity, orientation, layering, surface structuring, mechanical properties and biodegradability of the fibers the properties can be adapted for specific applications ranging from implantable medical devices to wound repair and protective clothing. Especially the combination of different polymer types offers a high potential. In this study electrospun two-component nonwoven structures of thermoplastic copolyester elastomer (TPC-ET) and bioresorbable polylactide (PLLA) were fabricated, using different ES setups. A comparative evaluation in terms of porosity, thermal and mechanical properties as well as required fabrication effort, was performed. Nonwovens made from polymer blends and coaxial spun core-sheath fibers showed similar tensile strength, which was higher than dual electrospun fabrics. Porosity was found to be in the range of 80 - 90%. By modifying the polymer solution and process parameters multicomponent nonwoven structures with tailored properties and drug release profiles can be manufactured.

The selection of appropriate test conditions is of critical importance in mechanical testing of biomaterials. In particular the definition of dynamic test conditions is requiring high efforts. In this work, a thermoplastic semicrystalline polymer was characterized regarding mechanical properties by dynamic mechanical analysis (DMA). Timetemperature- superposition (TTS) of dynamic mechanical data provides an efficient method for the experimental design of follow-up studies. Our focus was to provide test conditions for cyclic tests, which correspond to viscoelastic materials. The results show, that the test temperature for dynamic mechanical fatigue tests, as an indicator of mechanical modification for viscoelastic materials, should necessarily remain below the onset temperatures of storage modulus and loss modulus. Moreover, changes in material characteristics due to varying frequencies should be considered when constructing a master curve for the evaluation of test frequencies. Therefore, TTS is particularly beneficial for the rapid determination of test parameters for accelerated material examination.

A micro-sized core-shell drug release device has been prepared by coaxial electrospinning based on a polylactide shell and the inner core phase made of high molecular weight hyaluronan or a cross-linked hyaluronan network. Timolol, which was inserted into the core phase, is widely used as drug for glaucoma treatment to depress the intraocular eye pressure and should be released over a longer time period from this fiber networks. Release studies of drug loaded fiber devices showed an only moderate burst release of timolol in the first days followed by a continuous longlasting drug release period over several weeks. With these findings, coaxial electrospinning represents a promising approach for the development of drug release systems with long-lasting, relatively constant drug delivery.

For the investigation of in vitro drug release, methods have been used in which samples of drug delivery systems are immersed in release medium. The medium is used to measure drug concentration via chromatography or photometry. These systems are suitable to investigate the drug release of different systems or to simulate tissue environments. When considering predominantly humid regions, e.g. for drug release into the cochlea through the round window membrane by a drug delivery system placed at that membrane, reproducible in vitro determination of drug release becomes particularly challenging. In this study the development of a system is reported that allows the investigation of the in vitro drug release simulating such conditions. The presented test system consists of an alginate hydrogel in glass vials simulating the biological membrane, which separates the drug delivery system from the medium filled compartment. Saline is used as release medium and injected under the hydrogel. The samples are placed on top of the hydrogel, which slightly contacts the medium surface. The drug concentration in the release medium was determined by HPLC measurements. This system allows for testing the release of dexamethasone without the samples being completely surrounded by medium. The hydrogel mediates the diffusion of the drug by ensuring the contact with the medium. Release was monitored for more than 23 days. The presented concept was successfully designed and manufactured. The system is inexpensive and can be duplicated easily. In this study, it was used to monitor the drug release of dexamethasone from PEGDA700 derived polymer. One challenge that remains to be considered is the low mechanical stability of the hydrogel, which results in a need for repeated manufacturing during the handling of the system.

Aging is accompanied by changes in the cardiovascular physiology that promote the development of age-related diseases. This paper presents a modern approach to quantify the physiological effects of age on the cardiovascular system by applying modern machine learning techniques to several indicators of autonomic cardiovascular function. In 885 healthy subjects, 33 different indices were calculated on resting state electrocardiogram and continuous blood pressure recordings. Based on those parameters, five different approaches were applied in order to reconstruct the calendar age of healthy subjects, i.e., linear regression (LR), neural network (NN), Gaussian process regression (GPR), support vector regression (SVR), and relevance vector regression (RVR). Hyper parameters of machine learning methods were optimized via grid search. After 20 repetitions of a five-fold cross-validation, the mean absolute error (MAE) was computed between the calendar and estimated age to assess the accuracy of each method. The results show that the lowest error for age estimation was achieved using SVR with a MAE of 5.49 years. GPR performed comparably well to SVR with a MAE of 5.55 years, while NN led to a MAE of 5.72 years. RVR and LR revealed MAE of more than six years (6.21 and 6.34 years). The error of age estimation could be further reduced by applying outlier correction to the input data leading to a minimum MAE of 4.53 years with GPR. In conclusion, our results suggest that machine learning can be used to quantify the effects of healthy aging on cardio-vascular function.

Proximal occlusion of the Fallopian tube is one of the most common causes of female infertility. Due to the occlusion, the passage of the fallopian tubes is no longer given. Basically, there are two options for patients affected by this condition: cost-intensive in vitro fertilization (IVF) or surgery. The pregnancy rates of approximately 50% achieved with current treatment options are not satisfying. In this work, we present a Finite Element Analysis (FEA) model of a previously reported optimized microstent design for minimally invasive therapy of proximal tubal occlusion. Based on experimental investigations, the material model was set up and the simulation was validated. Comparison of the mechanical performance as an application related critical load case was in a good agreement. In this work, the proof of concept for the FEA model and the material model were carried out. In the future, the simulation will be used for further load cases such as the investigation of the bending stiffness and radial force and for the design optimization.

Many catheters and vascular implants are coated to increase biocompatibility or to reduce friction during catheter based implantation. Several regulations require assessment of coating durability over the implant’s life time. An in vitro method for stent testing is presented to measure released particulate matter at defined inspection intervals. The method was validated using polystyrene microspheres with a size of 10, 25 and 50 μm to check for particle recovery (n=6). Two cleaning steps followed. Particle counting was performed by light obscuration method. The recovery rate was 103±5% (10μm), 94±4% (25 μm) and 78±12% (50 μm), respectively, meeting the requirements of FDA guidance documents (i.e. FDA 1545). Less than 3% of the particles were found in the cleaning solutions. The method using a fixed volume during stent loading can be adapted to all durability testers where tubes are used to fix the stents (radial pulsatile, bending, axial compression, torsion).

For the development of combination products, the determination of the release kinetics of polymer-based drug delivery systems (DDS) is a central and often timeconsuming investigation. Classical methods are often unsuitable for large-scale screening of potential polymers for combination products. We present a rapid method based on surface plasmon resonance spectroscopy (SPR), using PEGDA700 (Poly(ethylene glycol) diacrylate, average molecular weight (Mn) 700) as an example. This method is capable of determinating the kinetics of drug association and subsequent release within minutes. The proportion of desorption and diffusion can be determined separately. Surface plasmon resonance spectroscopy (SPR) is a label-free and very sensitive optical technique, which allows for real-time observation of surface interactions. The prepared SPR chips were spin-coated with PEGDA700 and crosslinked via photoinduced polymerization. The association and dissociation kinetics of dexamethasone phosphate in swollen PEGDA700 have been studied at different concentrations. The maximum loading of the surface was also obtained by this method. The study of PEGDA presented here identified the wellestablished SPR-based spectroscopy as a potential tool in the development of combination products.

The coordinated research project “RESPONSE - Partnership for Innovation in Implant Technology”, which is one of ten BMBF-funded research consortia within the program “Twenty20 - Partnership for Innovation”, is active in the field of medical device innovation. The consortium is currently in the process of unfolding its transfer phase. This involves a portfolio of key innovations for novel medical devices, technologies and processes. Also, RESPONSE is aiming at participative technology development, integrating perspectives of patients, developers, as well as systems and innovation researchers. Particular challenges are arising from the current re-prioritization and re-scheduling of project roadmaps in order to manage and alleviate the effects of the corona crisis.

Bioresorbable nanofiber nonwovens with their fascinating properties provide a wide range of potential biomedical applications. Modification of the material enables the adjustment of mechanical and biological characteristics depending on the desired application. Due to the nanosized fiber network, post-production structuring is very challenging. Within this study, we use femtosecond laser technology for structuring permeable and resorbable electrospun poly-L-lactide (PLLA) membranes. We show that this post-production process can be used without disturbing the fiber network near the structured areas. Furthermore, the modification of the water permeability and mechanical characteristics due to the laser structuring was investigated. The results prove femtosecond laser technology to be a promising method for the adjustment of the membrane properties and which in consequence can help to optimize cell adhesion, enable revascularization and open up applications of nanofiber membranes in personalized medicine.

Correct bone screw torque is critical for positive patient outcomes after orthopaedic surgery. Models of the screwing process have been developed to allow a smart screwdriver to optimise the insertion torque. Experimental data is required to test these models, so a test-rig has been developed. Accurate torque measurement is a key part of the test-rig. An FIR filter was designed for this torque signal, implemented on the test-rig, and compared theoretically and experimentally to a mean filter and to no filtering. The FIR and mean filters both performed well, with the FIR achieving better theoretical results, and the mean filter achieving better experimental results. Better understanding of the noise structure and potential signal distortion would be required to improve the FIR filter or to conclusively compare it against the mean filter, however both perform sufficiently well for this application.

Catheters are widely used for therapeutic and diagnostic purposes in various medical applications. Along with frictional properties as well as the catheter profile the catheter stiffness mainly affects the deliverability and thus, the handling properties of the catheter. Within this study the bending stiffness of proximal and distal catheter samples was investigated with a custom made test setup. In particular, the influence of the catheter clamping length on the test results is discussed. Bending stiffness was calculated directly from the measured force, deflection and clamping length considering the test setup compliance. Measurements were performed three times at five positions in circumferential direction. Measured bending stiffness ranged from 629 ± 31 Nmm² to 733 ± 58 Nmm² for the proximal samples and from 30 ± 5 Nmm² to 98 ± 30 Nmm² for the distal samples, respectively. Bending stiffness varied depending on the free catheter length and the reaction force measured. The maximum reaction force decreased with increasing free catheter length leading to a higher measurement uncertainty. However, when considering the same free catheter length quantitative results were similar within the group of proximal and distal samples, respectively.

Mild Cognitive Impairment (MCI) is the asymptomatic, preclinical transitional stage among aging and Alzheimer’s Disease (AD). Detection of MCI can ensure the timely intervention required to manage the disease’s severity. Morphological alterations of Lateral Ventricle (LV) is considered as a significant biomarker for disease diagnosis. This research aims to analyze the shape alterations of the LV region using Fractional Order Jacobi Fourier Moment (FOJFM) features, which are categorized by their generic nature and capabilities to perform time-frequency analysis. T1-weighted transaxial view brain MR images (HC = 92 and MCI = 63) are obtained from publicly available Open Access Series of Imaging Studies (OASIS) database. The LV region is delineated using Weighted Level Set (WLS) segmentation method and results are compared to Ground Truth (GT) images. FOJFM features are employed to characterize the morphometry of LV region. From this segmented region, 200 features are computed by varying the value of order and fractional parameters. Random Forest (RF) and Support Vector Machine (SVM) classifiers are used to differentiate Healthy Control (HC) and MCI subjects. Results show that WLSE is able to delineate the LV structure. The segmented region shows good correlation with the GT area. FOJFM features are observed to be statistically significant in discriminating HC and MCI subjects with p<0.05. For MCI subjects, the feature values show higher variation as compared with HC brain, which might be due to the surface expansion of ventricular area during disease progression. SVM and RF classifiers show high performance F-measure values of 93.14% and 86.24%, respectively, for differentiating MCI conditions. The proposed moment based FOJFM features are able to capture the morphological changes of LV region related to MCI condition. Hence the proposed pipeline of work can be useful for the automated and early diagnosis of diseased conditions

Laser material processing enables precise machining of a wide variety of materials. In order to prevent an excessive heat input, which leads to irreversible material damage, the process parameters have to be adapted to the processed material. To keep the heat load as low as possible, the potential of femtosecond laser technology is exploited. The processing of semi-finished products using femtosecond laser technology is highly depending on the processed material. In this regard, we established a workflow and basic parameters for the processing of polymeric material. The performed parameter study varying cutting speed, cutting gas pressure and pulse energy to optimize the manufacturing process. Scanning electron microscopy (SEM) was utilized to analyze the cutting results, such as cut edge quality or possible melted areas. The established parameter set is also suitable for processing of very filigree material structures as used in innovative medical devices. The SEM analysis of the established parameter set showed that a homogeneous, nonwavy cut edge was created along the kerf.

Protein adsorption on solid surfaces is characterized by multivalence, binding-unit overlap, sequential adsorption, surface allosterics, lateral interactions and pronounced adsorption-desorption hysteresis, giving rise to the sequential, hystallosteric, adsorption model ("SHA model"). Adsorption isotherms of fibrinogen on a titanium miniplates and of the growth factors rhBMP-2 and rhVEGF165 on PDLLA nanofiber fleeces are presented. Controversial Langmuir type isotherms of fibrinogen and rhVEGF165 can be understood on the basis of singular long-lived metastable states central to the SHA-model.

Absorbable stents made of poly-L-lactic acid (PLLA) exhibit inferior mechanical properties compared with metal based devices. Therefore, design optimization by means of finite element analysis (FEA) plays an important role in development of high performance absorbable stents. In the present study, three different meander structures made of PLLA material were compared in terms of their radial resistive force (RRF) using FEA. Material tests were performed to obtain input parameters of PLLA for the simulation. FEA was carried out to simulate the RRF and the length change of the meander structures as a function of the diameter. In addition, the numerical results were compared to the experimental outcome using corresponding meander based stents performing radial testing with a segmented head mechanism. Comparable high stresses occurred in the Ubends of the examined stent designs. In designs with a large radius of curvature less stress was observed while crimping. One meander structure developed comparatively higher structural resistance against radially acting forces, due to its curved strut design. In general, FEA revealed consistent results compared to the experimental investigations.

For patient safety, it is important that a medical device can safely and reliably perform its intended purpose. The challenge in medical technology is that medical devices are heterogeneous systems and thus no widely applicable standard concepts for functional safety exist in medical technology. This is also reflected in the regulatory landscape, with its rather generally applicable standards. Patient safety is currently achieved by performing continuous risk management with an acceptable level of residual risk. Functional safety and its design concepts, as applied in other industries, have so far found little application in the field of medical technology. In this paper, the automotive safety concept "EGAS" is analyzed with regard to its applicability for medical devices. Based on the investigated example of a medical ventilator, important parallels were found between the automotive and the medical device sector, indicating the possibility of successfully applying the EGAS safety concept to medical devices.

Implant devices for micro invasive glaucoma surgery (MIGS) are gaining increasing acceptance in clinical ophthalmic use. The implant requirements are defined in international standards, such as ANSI Z80.27-2014 and the 2015 Guidance for Industry and Food and Drug Administration Staff “Premarket Studies of Implantable Minimally Invasive Glaucoma Surgical (MIGS) Devices”. The exact fluid-mechanical characterization represents a crucial part of the development and approval of innovative implant devices for MIGS. The current work describes the development and preliminary validation of a versatile test facility for pivotal characterization of glaucoma drainage devices. The test setup enables a pressurization of test specimens by means of two water columns. For measurement of pressure and volume flow, a pressure transducer and a total of three liquid flow meters were implemented into the test setup. Validation was conducted by experimental pressureflow characterization of standardized tubes and a comparison to theoretical results according to Hagen Poiseuille's law for stationary laminar flow of a Newtonian fluid in a tube with a circular cross section. Ultrapure water at (35 ± 2) °C was used for the analyses. The developed test setup potentially enables pressure-flow characterization of test specimens in a wide flow range of 0 μl min -1 ≤ Q ≤ 5.000 μl min -1 . The preliminary test facility validation showed a good agreement of measured and theoretical volume flow characteristics as a function of the pressure difference, in the currently investigated flow range of Q < 80 μl min -1 . The developed test facility is suitable for pivotal in vitro characterization of glaucoma drainage devices. Future investigations will focus on the final validation of the whole flow range and on the use of the test facility for fluid-mechanical characterization of self-developed prototypes of glaucoma microstents as well as commercially available glaucoma drainage devices.

Computer-assisted tools can aid in the detection of Alzheimer disease (AD) which is a progressive neurodegenerative disorder that can lead to cognitive impairments and eventually death. The accumulated effects due to AD can cause changes in the appearance of grey matter, white matter and cerebrospinal fluid in brain Magnetic Resonance (MR) images. This study aims to use Kernel Density Estimation (KDE) technique to analyse the textural changes from single slice brain MR images for the detection of AD. The preprocessed, skull stripped T1-weighted MR brain images are obtained from the publicly available OASIS database. A single axial slice per subject is chosen from a volumetric image for further processing to reduce the computational load. Multivariate KDE technique is applied to each pixel, by considering the changes in the neighbourhood based on selected bandwidth to obtain corresponding density estimates. Statistical features quantifying the distribution of density estimates are extracted to characterise textural variations in images. Linear discriminant analysis (LDA) classifier is implemented with ten-fold cross-validation for detecting AD. An optimum bandwidth of 18 for the KDE technique is selected based on the classification performance. Out of seven extracted texture features, three are found to be statistically significant in distinguishing AD. The classification with LDA yields an accuracy of 72.3% with a sensitivity of 80.6% for identifying AD from healthy subjects. The proposed method is efficient in detecting AD by revealing the textural changes within the brain slice without the involvement of any segmentation technique. Thus, the novel KDE-based texture analysis proves to be an effective tool for the automated diagnosis of AD from single slice brain MR images.

Introduction: Pedicle screw fixation systems are widely used for treatment of various spinal pathologies, including spinal stenosis, scoliosis, spinal deformities and fractures. Stress shielding is considered to be a major factor contributing to insufficient fixation strength, leading to screw loosening. In this study, the influence of pedicle screw thread width on the displacement of pedicle screw and stress transfer is analyzed using 2-Dimensional axisymmetric finite element (FE) model. Methods: FE model consisting of cancellous and cortical bone, along with pedicle screw is developed for this study. The pedicle screw thread width is varied between 0.1 mm and 0.6 mm in steps of 0.1 mm, while the other geometric parameters, including screw half-angle, pitch, diameter, and length are kept constant. Three different contact conditions between screw and bone, such as frictionless, frictional, and bonded are considered to simulate hours, days, and months after surgery, respectively. The material properties and boundary conditions are applied based on previous studies. An axial force of 80 N is applied on the screw head to simulate axial pull-out test. Results: Similar patterns of stress distribution are observed for all screw models, with high stress concentration above the first thread. The highest displacement in screw is observed shortly after surgery, while the highest displacement in cancellous and cortical bone is observed few days and months after the surgery, respectively. The average von Mises stress in screw decreases with increase in thread width for all contact conditions. In few hours/days after the surgery, stress transfer parameter increases with increase in thread width, up to a thread width of 0.5 mm and then decreases. The changes in stress transfer parameter are negligible few months after the surgery. Conclusion: This study highlights the influence of thread width on displacement and stress transferred to the bone, at different durations after the surgery. It is observed that a thread width of 0.5 mm exhibits the highest stress transfer, leading to reduced stress shielding and improved bone remodeling. It appears that this study might aid in developing better pedicle screws for the treatment of various spinal pathologies.

Data augmentation is a common method to make deep learning assessible on limited data sets. However, classical image augmentation methods result in highly unrealistic images on ultrasound data. Another approach is to utilize learning-based augmentation methods, e.g. based on variational autoencoders or generative adversarial networks. However, a large amount of data is necessary to train these models, which is typically not available in scenarios where data augmentation is needed. One solution for this problem could be a transfer of augmentation models between different medical imaging data sets. In this work, we present a qualitative study of the cross data set generalization performance of different learning-based augmentation methods for ultrasound image data. We could show that knowledge transfer is possible in ultrasound image augmentation and that the augmentation partially results in semantically meaningful transfers of structures, e.g. vessels, across domains.

Correctly torquing bone screws is an important factor in achieving positive patient outcomes during orthopaedic surgery. A torque-limiting smart screwdriver concept has been proposed, and ongoing work is being undertaken to model the screwing process and allow the concept to work. These models require experimental validation, so a test rig was developed. The magnetostrictive torque sensor in this test rig was affected by magnetic parts of the test rig, which offset the zero-torque point; this raised concerns over the effects on linearity, which were tested here. The torque sensor was tested against a non-magnetostrictive reference under varying external magnetic conditions. While the magnetic field offset the torque, there was no notable change in linearity under the conditions tested, and the linearity was always within the datasheet specifications. Hence, we conclude that in the context of this test rig, there were no negative effects on linearity, although under higher loading or stronger magnetic conditions, this may not hold.

Alzheimer’s Disease (AD) is an irreversible progressive neurodegenerative disorder. Magnetic Resonance (MR) imaging based deep learning models with visualization capabilities are essential for the precise diagnosis of AD. In this study, an attempt has been made to categorize AD and Healthy Controls (HC) using structural MR images and an Inception-Residual Network (ResNet) model. For this, T1- weighted MR brain images are acquired from a public database. These images are pre-processed and are applied to a two-layer Inception-ResNet-A model. Additionally, Gradient weighted Class Activation Mapping (Grad-CAM) is employed to visualize the significant regions in MR images identified by the model for AD classification. The network performance is validated using standard evaluation metrics. Results demonstrate that the proposed Inception-ResNet model differentiates AD from HC using MR brain images. The model achieves an average recall and precision of 69%. The Grad- CAM visualization identified lateral ventricles in the mid-axial slice as the most discriminative brain regions for AD classification. Thus, the computer aided diagnosis study could be useful in the visualization and automated analysis of AD diagnosis with minimal medical expertise.

Electroencephalography (EEG) based emotion recognition is a widely preferred technique due to its noninvasiveness. Also, frontal region-specific EEG signals have been associated with emotional processing. Feature reductionbased optimized machine learning methods can improve the automated analysis of frontal EEG signals. In this work, an attempt is made to classify emotional states using entropybased features and Bayesian optimized random forest. For this, the EEG signals of prefrontal and frontal regions (Fp1, Fp2, Fz, F3, and F4) are obtained from an online public database. The signals are decomposed into five frequency bands, namely delta (1-4 Hz), theta (4-8 Hz), alpha (8-13 Hz), beta (14-30 Hz), and gamma (30-45 Hz). Three entropy features, namely Dispersion Entropy (DE), Sample Entropy (SE), and Permutation Entropy (PE), are extracted and are dimensionally reduced using Principal Component Analysis (PCA). Further, the reduced features are applied to the Bayesian optimized random forest for the classification. The results show that the DE in the gamma band and SE in the alpha band exhibit a statistically significant (p < 0.05) difference for classifying arousal and valence emotional states. The selected features from PCA yield an F-measure of 73.24% for arousal and 46.98% for valence emotional states. Further, the combination of all features yields a higher F-measure of 48.13% for valence emotional states. The proposed method is capable of handling multicomponent variations of frontal region-specific EEG signals. Particularly the combination of selected features could be useful to characterize arousal and valence emotional states.

Triboelectric nanogenerators (TENGs) are energy converters or energy harvesters that convert mechanical motion into electrical energy on the basis of their material properties. A particular advantage of the TENG is its ability to convert small, low-frequency and random mechanical movements that are relevant for body movements and wearable applications. Within the presented study, different Parylene types were analysed as the dielectric material in TENG and found to be promising with respect to providing high output voltages and powers, respectively. Besides the verification of the usability of Parylene for TENG and its superior triboelectric properties, also significant differences were found between the Parylene types.

Inner ear trauma caused by cochlear implantation is a severe clinical problem. It was shown that an electrode alginate coating can reduce the insertion forces in vitro. The grade of the alginate viscosity can be adjusted by using different metal ions for cross-linking the salts of the alginic acid. The aim was to investigate the stability and inner ear biocompatibility of alginate using different in vitro established cross-linkers. Alginate beads were cross-linked in either calcium chloride (CaCl 2 ) or barium chloride (BaCl 2 ) solution. The beads were cultivated in artificial perilymph and stability and swelling were observed for 13 months. Ototoxicity was tested on cochlear whole mount explants from neonatal rats. Neomycin served as positive control to induce hair cell damage and explants without any addition served as negative control. The beads and explants were co-cultured for 48 hours and the hair cell survival was analysed microscopically. Neomycin treatment induced an extensive inner and outer hair cell loss. Neither CaCl 2 nor BaCl 2 cross-linked alginate beads caused any damage to the hair cells. Even though the same volume of alginate and cross-linkers were used, in CaCl 2 cross-linked beads were initially almost double the size of in BaCl 2 cross-linked beads. None of the cross-linked alginate beads had a significant volume change within 3 months being cultured in artificial perilymph. After 3 months the CaCl 2 cross-linked beads swelled massively and dissolved within one week whereas BaCl 2 cross-linked alginate beads remained unchanged until month 13 after culture start. Alginate beads gelled with both cross-linkers are biocompatible with the inner ear sensory epithelium. Both cross-linkers ensure a stable gelation of alginate but a swelling followed by degradation of the in CaCl 2 cross-linked beads occurred after 3 months. For coatings, which need to be long term stable, BaCl 2 should be chosen whereas CaCl 2 may be more suitable for applications where limited stability is needed and the swelling is not affecting the surrounding tissue. Therefore, BaCl 2 cross-linking of alginate may be the best choice for cochlear implant coating.

Understanding the underlying pathology in different tissues and organs is crucial when fighting pandemics like COVID-19. During conventional autopsy, large tissue sample sets of multiple organs can be collected from cadavers. However, direct contact with an infectious corpse is associated with the risk of disease transmission and relatives of the deceased might object to a conventional autopsy. To overcome these drawbacks, we consider minimally invasive autopsies with robotic needle placement as a practical alternative. One challenge in needle based biopsies is avoidance of dense obstacles, including bones or embedded medical devices such as pacemakers. We demonstrate an approach for automated planning and visualising suitable needle insertion points based on computed tomography (CT) scans. Needle paths are modeled by a line between insertion and target point and needle insertion path occlusion from obstacles is determined by using central projections from the biopsy target to the surface of the skin. We project the maximum and minimum CT attenuation, insertion depth, and standard deviation of CT attenuation along the needle path and create two-dimensional intensity-maps projected on the skin. A cost function considering these metrics is introduced and minimized to find an optimal biopsy needle path. Furthermore, we disregard insertion points without sufficient room for needle placement. For visualisation, we display the color-coded cost function so that suitable points for needle insertion become visible. We evaluate our system on 10 post mortem CTs with six biopsy targets in abdomen and thorax annotated by medical experts. For all patients and targets an optimal insertion path is found. The mean distance to the target ranges from (49.9 ± 12.9)mm for the spleen to (90.1 ± 25.8)mm for the pancreas.

Preterm birth (gestational age < 37 weeks) is a serious pregnancy related complication that could lead to fetal morbidity and mortality. Monitoring the activity of uterus is considered to be crucial for the early diagnosis of preterm birth. Uterine Electromyography (uEMG) is a non-invasive technique that provides a quantitative measure of uterine activity from the abdominal surface. In this work, an attempt has been made to characterize preterm uEMG signals using Empirical Mode Decomposition based Detrended Fluctuation Analysis (EMD-DFA). Preterm signals with varied gestational ages are considered from an online database. EMD-DFA is applied on these signals to compute the fluctuation function. The double-logarithmic plot of fluctuation function versus scale is evaluated and Chi-square analysis is performed for identifying linear scaling regions. Five features namely shortterm exponent (Hs), long-term exponent (Hl), inflection point, short-term fractal angle (αHs) and long-term fractal angle (αHl) are extracted and analyzed. Further, Coefficient of Variation (CV) is computed to examine the variations of these features among different subjects. Results show that EMD-DFA is able to characterize the fluctuations of preterm signals. From the double-logarithmic plot, a slow variation of fluctuation function is observed with respect to scale when the time to delivery is more. This indicates the presence of rapid signal fluctuations in the early stages of pregnancy. Based on the feature values, it is observed that the signal fluctuations are more correlated and smoother as the time to delivery approaches. Among the extracted features, CV values of Hs, Hl, αHs and αHl are observed to be low indicating that these features have least inter-subject variations in preterm signals. The EMD-DFA based fractal features show the ability to detect the subtle variations in uEMG signals. As early diagnosis of preterm delivery is imperative for timely medical intervention and treatment, it appears that the proposed approach could aid in determining the changes in uterine contractions in preterm condition.

Neural networks (NN) are a powerful tool to tackle complex problems in hearing aid research, but their use on hearing aid hardware is currently limited by memory and processing power. To enable the training with these constrains, a fixed point analysis and a memory friendly power of two quantization (replacing multiplications with shift operations) scheme has been implemented extending TensorFlow, a standard framework for training neural networks, and the Qkeras package [1, 2]. The implemented fixed point analysis detects quantization issues like overflows, underflows, precision problems and zero gradients. The analysis is done for each layer in every epoch for weights, biases and activations respectively. With this information the quantization can be optimized, e.g. by modifying the bit width, number of integer bits or the quantization scheme to a power of two quantization. To demonstrate the applicability of this method a case study has been conducted. Therefore a CNN has been trained to predict the Ideal Ratio Mask (IRM) for noise reduction in audio signals. The dataset consists of speech samples from the TIMIT dataset mixed with noise from the Urban Sound 8kand VAD-dataset at 0 dB SNR. The CNN was trained in floating point, fixed point and a power of two quantization. The CNN architecture consists of six convolutional layers followed by three dense layers. From initially 1.9 MB memory footprint for 468k float32 weights, the power of two quantized network is reduced to 236 kB, while the Short Term Objective Intelligibility (STOI) Improvement drops only from 0.074 to 0.067. Despite the quantization only a minimal drop in performance was observed, while saving up to 87.5 % of memory, thus being suited for employment in a hearing aid

During a clinical study the AEA (Acoustic Emission Analysis) of arthritic defects in the knee joint was enhanced by the addition of kinetic measurement data. This enhanced AEA based method permits a non-invasive diagnosis and assessment of arthritic joint damage at an early stage. The diagnostic procedure includes three separate measurements that contribute in different ways to an extended diagnosis of the disease pattern [1, 2, 3]. During a series of three knee bends a force plate provides data of the ground force while a video-based gait analysis records the corresponding movement and the angles of hip-, knee-, and ankle joints. At the same time AEA detects the acoustic anomalies of damaged cartilage and the absolute angle of the system. The patterns of the kinetic data were analyzed to define the instants of time to correlate the data of the 3 measurement systems. The analysis of the force data yields a pattern with 8 phases. By means of the stance phase between the knee bends the instants of time are used to synchronize force and video based data. In the second step the synchronization of video based data was done by means of the absolute angle of the AEA system [4, 5]. The superposition of kinetic data and the acoustic emission permits a preliminary graphic representation and assessment of the measurement data. The procedure will be applied for the analysis of patients in a clinical study

: In patients with arthritic knee joint defects the course of movement, the application of muscle forces and the degree of freedom of the joints of the lower limbs differs significantly from the corresponding data of a healthy proband. The enhanced acoustic-kinetic joint analysis based on AEA, ground force reaction measurements and video enhanced gait analysis permits the correlation of force data, joint angles and acoustic emission significant of defective joint cartilage regions. This diagnostic procedure permits a quantifiable and detailed non invasive diagnosis of lesion patterns in the arthritic knee joint by means of a synchronization algorithm. The AEA shows lesion signals in the first and third knee bend of the patient. The lesion signals are assigned to the joint angles and ground reaction force. Comparing the kinematical data there is a shift between the curves of the hip and the knee angle. Overall the force data represent an imbalance between the left and the right leg during the knee bends. The presentation of the three measurement systems, the method, the synchronization of the data sets and their final assessment as well as the occurring difficulties during a case of a clinical study are discussed. This would be helpful regarding further patients of the clinical study