Volume 64, Issue 5

In this review, scientific investigations of point-of-care testing (POCT) and point-of-care monitoring (POCM) devices are summarized with regard to the measurement accuracy of the hemoglobin concentration. As a common basis, information according to the Bland and Altman principle [bias, limits of agreement (LOA)] as well as the measurement accuracy and precision are considered, so that the comparability can be mapped. These collected data are subdivided according to the manufacturers, devices and procedures (invasive and non-invasive). A total of 31 devices were identified. A comparability of the scientific investigations in particular was given for 23 devices (18 invasive and five non-invasive measuring devices). In terms of measurement accuracy, there is a clear leap between invasive and non-invasive procedures, while no discernible improvement can be derived in the considered time frame from 2010 to 2018. According to the intended use, strict specifications result from the clinical standards, which are insufficiently met by the systems. More stringent requirements can be derived both in the area of blood donation and in the treatment of patients.

Over several years, research had been conducted for the detection of epileptic seizures to support an automatic diagnosis system to comfort the clinicians’ encumbrance. In this regard, a number of research papers have been published for the identification of epileptic seizures. A thorough review of all these papers is required. So, an attempt has been made to review on the pattern detection methods for epilepsy seizure detection from EEG signals. More than 150 research papers have been discussed to determine the techniques for detecting epileptic seizures. Further, the literature review confirms that the pattern recognition techniques required to detect epileptic seizures varies across the electroencephalogram (EEG) datasets of different conditions. This is mostly owing to the fact that EEG detected under different conditions have different characteristics. This consecutively necessitates the identification of the pattern recognition technique to efficiently differentiate EEG epileptic data from the EEG data of various conditions.

To improve the robustness and to reduce the energy consumption of facial expression recognition, this study proposed a facial expression recognition method based on improved deep residual networks (ResNets). Residual learning has solved the degradation problem of deep Convolutional Neural Networks (CNNs); therefore, in theory, a ResNet can consist of infinite number of neural layers. On the one hand, ResNets benefit from better performance on artificial intelligence (AI) tasks, thanks to its deeper network structure; meanwhile, on the other hand, it faces a severe problem of energy consumption, especially on mobile devices. Hence, this study employs a novel activation function, the Noisy Softplus (NSP), to replace rectified linear units (ReLU) to get improved ResNets. NSP is a biologically plausible activation function, which was first proposed in training Spiking Neural Networks (SNNs); thus, NSP-trained models can be directly implemented on ultra-low-power neuromorphic hardware. We built an 18-layered ResNet using NSP to perform facial expression recognition across datasets Cohn-Kanade (CK+), Karolinska Directed Emotional Faces (KDEF) and GENKI-4K. The results achieved better anti-noise ability than ResNet using the activation function ReLU and showed low energy consumption running on neuromorphic hardware. This study not only contributes a solution for robust facial expression recognition, but also consolidates the low energy cost of their implementation on neuromorphic devices, which could pave the way for high-performance, noise-robust and energy-efficient vision applications on mobile hardware.

Event-related mental task information collected from electroencephalography (EEG) signals, which are functionally related to different brain areas, possesses complex and non-stationary signal features. It is essential to be able to classify mental task information through the use in brain-computer interface (BCI) applications. This paper proposes a wavelet packet transform (WPT) technique merged with a specific entropy biomarker as a feature extraction tool to classify six mental tasks. First, the data were collected from a healthy control group and the multi-signal information comprised six mental tasks which were decomposed into a number of subspaces spread over a wide frequency spectrum by projecting six different wavelet basis functions. Later, the decomposed subspaces were subjected to three entropy-type statistical measure functions to extract the feature vectors for each mental task to be fed into a backpropagation time-recurrent neural network (BPTT-RNN) model. Cross-validated classification results demonstrated that the model could classify with 85% accuracy through a discrete Meyer basis function coupled with a Renyi entropy biomarker. The classifier model was finally tested in the Simulink platform to demonstrate the Fourier series representation of periodic signals by tracking the harmonic pattern. In order to boost the model performance, ant colony optimization (ACO)-based feature selection method was employed. The overall accuracy increased to 88.98%. The results underlined that the WPT combined with an entropy uncertainty measure methodology is both effective and versatile to discriminate the features of the signal localized in a time-frequency domain.

Intracranial pressure (ICP) monitoring is crucial in determining the appropriate treatment in traumatic brain injury. Minimally invasive approaches to monitor ICP are subject to ongoing research because they are expected to reduce infections and complications associated with conventional devices. This study aims to develop a wireless ICP monitoring device that is biocompatible, miniature and implantable. Liquid crystal polymer (LCP) was selected to be the main material for the device fabrication. This study considers the design, fabrication and testing of the sensing unit of the proposed wireless ICP monitoring device. A piezoresistive pressure sensor was designed to respond to 0–50 mm Hg applied pressure and fabricated on LCP by standard microelectromechanical systems (MEMS) procedures. The fabricated LCP pressure sensor was studied in a moist environment by means of a hydrostatic pressure test. The results showed a relative change in voltage and pressure from which the sensor’s sensitivity was deduced. This was a proof-of-concept study and based on the results of this study, a number of recommendations for improving the considered sensor performance were made. The limitations are discussed, and future design modifications are proposed that should lead to a complete LCP package with an improved performance for wireless, minimally invasive ICP monitoring.

The aim of this study was to investigate the retention forces of secondary telescopic crowns made of polyetherketoneketone (PEKK) in combination with primary crowns made of four different dental alloys and to determine whether the retention forces change in the course of up to 5000 simulated wear cycles. A total of four groups of telescopic crowns were investigated: group 1: 10 primary crowns made of a gold alloy (NEOCAST ® 3), group 2: 10 primary crowns made of a non-precious metal alloy (Girobond NB), group 3: 10 primary crowns made of zirconium (Cercon ® base) and group 4: 10 primary crowns made from PEKK (Pekkton ® ivory). The corresponding secondary crowns were made from PEKK in all the four groups. Each pair was fixed axially in a wear simulator specifically designed for the study. Overall, 10,000 joining and separating cycles were performed for each group. To simulate intraoral conditions, a saliva substitute served as a lubricant. Force transducers were used to record the retention forces and after completing 10,000 cycles, the surface of each primary crown was examined using a scanning electron microscope (SEM). All groups showed an increase in the retention force for the first 2000 cycles which stayed constant for the remaining 8000 cycles. The Pekkton ® ivory/Pekkton ® ivory and NEOCAST ® 3/Pekkton ® ivory combinations displayed mean retention force values of 16 N after a slight increase in the retention force. The Cercon ® base/Pekkton ® ivory and Girobond NB/Pekkton ® ivory combinations displayed an initially high increase in the retention force and then showed a mean retention force of up to 29 N. All primary crowns displayed surface wear. Zirconium primary crowns showed the least wear compared to PEKK and the gold and non-precious metal alloys.

The aim of this study was to compare the screw-to-bone fixation strength of two insertion techniques: self-tapping screw (STS) and non-self-tapping screw (NSTS). Finite element analysis (FEA) was used for the comparison by featuring three tests (insertion, pull-out and shear) in a human tibia bone model. A non-linear material behavior with ductile damage properties was chosen for the modeling. To validate the numerical models, experimental insertion and pull-out tests were carried out using a synthetic bone. The experimental and numerical results of pull-out tests correlated well. Thread forming was successfully simulated during the insertion process of STS and NSTS. It is demonstrated that the STS generates higher insertion torque, induces a higher amount of stress after the insertion process and relatively more strength under the pull-out and shear tests than the NSTS. However, the NSTS induces more stiffness under the two tests (pull-out and shear) and less damage to the screw-bone interface compared to the STS. It is concluded that the use of STS ensures tighter bony contact and enables higher pull-out strength; however, the use of NSTS improves the stiffness of the fixation and induces less damage to the cortical bone-screw fixation and thus minimum risk is obtained in terms of bone necrosis.

In most contact theories, the most popular of which are the three models of Hertz, Derjaguin, Muller and Toporov (DMT) and Johnson, Kendall and Roberts (JKR), biological cells were considered as an elastic material which is not a proper assumption. The elastic assumption in the case of biological cells could lead to neglecting the loading history as a result of which the stresses and strains applied to the material would not be studied accurately. In this paper, developing the three mentioned elastic models into viscoelastic models, simulating and comparing them with empirical data obtained through the indentation test of the MCF-7 cancer cell showed that the viscoelastic state presents a better prediction of biological cell behavior compared to that of an elastic state. The selection of the suitable creep function for objects in contact is another issue that has a significant importance in the viscoelastic case and this was investigated. Different mechanical models of a cell were studied and simulated for all three named theories among which the creep function obtained from the Kelvin model, a parallel combination of spring-damper, simplified the simulation and gave more precise results for modeling due to the fact that the obtained results from this model are closer to experimental ones and simpler than other models. On the other hand, for a more exact prediction of cell behavior, this model was modified by an equivalent elasticity module which considered cell components instead of the cell cortex only. The results of the simulation confirmed that a new elasticity module can improve the accuracy of cell models. After choosing the suitable mechanical model for the cell, we scrutinized the capability of the developed theories in predicting the results for biological liquid environments. Although the results of the Hertz and DMT viscoelastic models are closer to experimental ones in comparison with viscoelastic JKR, neglecting adhesion makes their prediction in biological liquid environments weak and erroneous. Therefore, it can be concluded that the developed viscoelastic model of JKR is more accurate and has a better performance in different environments than the other mentioned models.

The articular cartilage tissue is an essential component of joints as it reduces the friction between the two bones. Its load-bearing properties depend mostly on proteoglycan distribution, which can be analyzed through the study of the presence of sulfated glycosaminoglycan (sGAG). Currently, sGAG distribution in articular cartilage is not completely known; it is calculated by means of laboratory tests that imply the inherent inaccuracy of a manual procedure. This paper presents an easy-to-use desktop software application for obtaining the sGAG distribution profile in tissue. This app uses color images of stained cartilage tissues taken under a microscope, so researchers at the Trinity Centre for Bioengineering (Dublin, Ireland) can understand the qualitative distribution of sGAG with depth in the studied tissues.

Image denoising has become a crucial task in medical ultrasound (US) imaging due to the presence of speckle or multiplicative noise and additive Gaussian noise. Recently, several denoising techniques such as adaptive wavelet thresholding & joint bilateral (AWT + JB) filter, adaptive fuzzy switching weighted mean (AFSWM) filter and median patch-based locally optimal Wiener (MPBLOW) filter have been proposed to remove the speckle noise. However, these denoising techniques were found to remove noise along with the essential parts of the actual image data which is known as over-filtering. Thereby, it reduces the accuracy of the recognition process. In this paper, a new hybrid filter technique is proposed by combining anisotropic diffusion (AD) with Butterworth band pass filter to overcome over-filtering of the image. In addition, the performance of the proposed hybrid filter and its design parameters are enhanced using the particle swarm optimization (PSO) algorithm. The simulation results show that the proposed filtering technique achieves a better denoising performance when compared with other filtering techniques in terms of peak signal-to-noise ratio (PSNR), signal-to-noise ratio (SNR), structural similarity index (SSIM) and edge preservation index (EPI). Moreover, the results validated that the proposed filtering technique using PSO achieves effective performance than using the harmony search algorithm (HSA) and other filtering techniques.