Volume 67, Issue 4

Although fetal phonocardiogram (fPCG) signals have become a good indicator for discovered heart disease, they may be contaminated by various noises that reduce the signals quality and the final diagnosis decision. Moreover, the noise may cause the risk of the data to misunderstand the heart signal and to misinterpret it. The main objective of this paper is to effectively remove noise from the fPCG signal to make it clinically feasible. So, we proposed a novel noise reduction method based on Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN), wavelet threshold and Crow Search Algorithm (CSA). This noise reduction method, named ICEEMDAN-DWT-CSA, has three major advantages. They were, (i) A better suppress of mode mixing and a minimized number of IMFs, (ii) A choice of wavelet corresponding to the study signal proven by the literature and (iii) Selection of the optimal threshold value. Firstly, the noisy fPCG signal is decomposed into Intrinsic Mode Functions (IMFs) by the (ICEEMDAN). Each noisy IMFs were decomposed by the Discrete Wavelet Transform (DWT). Then, the optimal threshold value using the (CSA) technique is selected and the thresholding function is carried out in the detail’s coefficients. Secondly, each denoised (IMFs) is reconstructed by applying the Inverse Discrete Wavelet Transform (IDWT). Finally, all these denoised (IMFs) are combined to get the denoised fPCG signal. The performance of the proposed method has been evaluated by Signal to Noise Ratio (SNR), Mean Square Error (MSE) and the Correlation Coefficient (COR). The experiment gave a better result than some standard methods.

Parkinson’s disease (PD), a slow-progressing neurological disease, affects a large percentage of the world’s elderly population, and this population is expected to grow over the next decade. As a result, early detection is crucial for community health and the future of the globe in order to take proper safeguards and have a less arduous treatment procedure. Recent research has begun to focus on the motor system deficits caused by PD. Because practically most of the PD patients suffer from voice abnormalities, researchers working on automated diagnostic systems investigate vocal impairments. In this paper, we undertake extensive experiments with features extracted from voice signals. We propose a layer Recurrent Neural Network (RNN) based diagnosis for PD. To prove the efficiency of the model, different network models are compared. To the best of our knowledge, several neural network topologies, namely RNN, Cascade Forward Neural Networks (CFNN), and Feed Forward Neural Networks (FFNN), are used and compared for voice-based PD detection for the first time. In addition, the impacts of data normalization and feature selection (FS) are thoroughly examined. The findings reveal that normalization increases classifier performance and Laplacian-based FS outperforms. The proposed RNN model with 300 voice features achieves 99.74% accuracy.

Supervised automatic sleep scoring algorithms are usually trained using sleep stage labels manually annotated on 30 s epochs of PSG data. In this study, we investigate the impact of using shorter epochs with various PSG input signals for training and testing a Long Short Term Memory (LSTM) neural network. An LSTM model is evaluated on the provided 30 s epoch sleep stage labels from a publicly available dataset, as well as on 10 s subdivisions. Additionally, three independent scorers re-labeled a subset of the dataset on shorter time windows. The automatic sleep scoring experiments were repeated on the re-annotated subset.The highest performance is achieved on features extracted from 30 s epochs of a single channel frontal EEG. The resulting accuracy, precision and recall were of 92.22%, 67.58% and 66.00% respectively. When using a shorter epoch as input, the performance decreased by approximately 20%. Re-annotating a subset of the dataset on shorter time epochs did not improve the results and further altered the sleep stage detection performance. Our results show that our feature-based LSTM classification algorithm performs better on 30 s PSG epochs when compared to 10 s epochs used as input. Future work could be oriented to determining whether varying the epoch size improves classification outcomes for different types of classification algorithms.

The incidence of vision impairment is rapidly increasing. Diagnosis and classifying retinal abnormalities in ophthalmological applications is a significant challenge. Using Optical Coherence Tomography (OCT), the study aims to develop a computer aided diagnosis system for detecting and classifying retinal disorders. Choroidal neovascularization, diabetic macular edema, drusen, and normal cases are the investigated groups. Both deep learning and machine learning are combined to build the system. The SqueezeNet neural network was modified to extract features. The Support Vector Machine (SVM), K-Nearest Neighbor (K-NN), Decision Tree (DT), and Ensemble Model (EM) algorithms were used for disorder classification. The Bayesian optimization technique was also used to determine the best hyperparameters for each model. The model’ performance was evaluated through nine criteria using 12,000 OCT images. The results have demonstrated accuracies of 97.39, 97.47, 96.98, and 95.25% for the SVM, K-NN, DT, and EM, respectively. When results are compared to relevant studies in terms of accuracy and tested samples, they show superior performance. As a result, a novel computer-aided diagnosis system for detecting and classifying retinal diseases has been developed, reducing human error while also saving time.

In order to prevent failure as well as ensure comfort, patient-specific modelling for prostheses has been gaining interest. However, deterministic analyses have been widely used in the design process without considering any variation/uncertainties related to the design parameters of such prostheses. Therefore, this study aims to compare the performance of patient-specific anatomic Total Knee Arthroplasty (TKA) with off-the-shelf TKA. In the patient-specific model, the femoral condyle curves were considered in the femoral component’s inner and outer surface design. The tibial component was designed to completely cover the tibia cutting surface. In vitro experiments were conducted to compare these two models in terms of loosening of the components. A probabilistic approach based on the finite element method was also used to compute the probability of failure of both models. According to the deterministic analysis results, 103.10 and 21.67 MPa von Mises stress values were obtained for the femoral component and cement in the anatomical model, while these values were 175.86 and 25.76 MPa, respectively, for the conventional model. In order to predict loosening damage due to local osteolysis or stress shield, it was determined that the deformation values in the examined cement structures were 15% lower in the anatomical model. According to probabilistic analysis results, it was observed that the probability of encountering an extreme value for the anatomical model is far less than that of the conventional model. This indicates that the anatomical model is safer than the conventional model, considering the failure scenarios in this study.

The study aimed to evaluate the biomechanical behaviors of different prosthetic materials and posterior implant angles in All-on-4 implant-supported fixed maxillary prostheses with three-dimensional (3D) finite element analysis. The model of complete edentulous maxilla was created using the Rhinoceros and VRMesh Studio programs. Anterior vertical and 17°- and 30°-angled posterior implants were positioned with All-on-4 design. Straigth and angled multi-unit abutments scanned using a 3D scanner. Two different prosthetic superstructures (monolithic zirconia framework and lithium disilicate veneer (ZL) and monolithic zirconia-reinforced lithium silicate (ZLS)) were modeled. Four models designed according to the prosthetic structure and posterior implant angles. Posterior vertical bilateral loading and frontal oblique loading was performed. The principal stresses (bone tissues-Pmax and Pmin) and von Mises equivalent stresses (implant and prosthetic structures) were analyzed. In all models, the highest Pmax stress values were calculated under posterior bilateral loading in cortical bone. The highest von Mises stress levels occured in the posterior implants under posterior bilateral load (260.33 and 219.50 MPa) in the ZL-17 and ZL-30 models, respectively. Under both loads, higher stress levels in prosthetic structures were shown in the ZLS models compared with ZL models. There was no difference between posterior implant angles on stress distribution occurred in implant material and alveolar bone tissue. ZLS and ZL prosthetic structures can be reliably used in maxillary All-on-4 rehabilitation.

Non-woven textile has been largely used as medical implant material over the last decades, especially for scaffold manufacturing purpose. This material presents a large surface area-to-volume ratio, which promotes adequate interaction with biological tissues. However, its strength is limited due to the lack of cohesion between the fibers. The goal of the present work was to investigate if a non-woven substrate can be reinforced by embroidery stitching towards strength increase. Non-woven samples were produced from both melt-blowing and electro-spinning techniques, reinforced with a stitching yarn and tested regarding several performances: ultimate tensile strength, burst strength and strength loss after fatigue stress. Several stitching parameters were considered: distance between stitches, number of stitch lines (1, 2 or 3) and line geometry (horizontal H, vertical L, cross X). The performance values obtained after reinforcement were compared with values obtained for control samples. Results bring out that reinforcement can increase the strength by up to 50% for a melt-blown mat and by up to 100% for an electro-spun mat with an X reinforcement pattern. However, after cyclic loading, the reinforcement yarn tends to degrade the ES mat in particular. Moreover, increasing the number of stitches tends to fragilize the mats.