Computer Methods in Biomechanics and Biomedical Engineering最新文献

Drug discovery relies on the precise prognosis of drug-target interactions (DTI). Due to their ability to learn from raw data, deep learning (DL) methods have displayed outstanding performance over traditional approaches. However, challenges such as imbalanced data, noise, poor generalization, high cost, and time-consuming processes hinder progress in this field. To overcome the above challenges, we propose a DL-based model termed DrugSchizoNet for drug interaction (DI) prediction of Schizophrenia. Our model leverages drug-related data from the DrugBank and repoDB databases, employing three key preprocessing techniques. First, data cleaning eliminates duplicate or incomplete entries to ensure data integrity. Next, normalization is performed to enhance security and reduce costs associated with data acquisition. Finally, feature extraction is applied to improve the quality of input data. The three layers of the DrugSchizoNet model are the input, hidden and output layers. In the hidden layer, we employ dropout regularization to mitigate overfitting and improve generalization. The fully connected (FC) layer extracts relevant features, while the LSTM layer captures the sequential nature of DIs. In the output layer, our model provides confidence scores for potential DIs. To optimize the prediction accuracy, we utilize hyperparameter tuning through OB-MOA optimization. Experimental results demonstrate that DrugSchizoNet achieves a superior accuracy of 98.70%. The existing models, including CNN-RNN, DANN, CKA-MKL, DGAN, and CNN, across various evaluation metrics such as accuracy, recall, specificity, precision, F1 score, AUPR, and AUROC are compared with the proposed model. By effectively addressing the challenges of imbalanced data, noise, poor generalization, high cost and time-consuming processes, DrugSchizoNet offers a promising approach for accurate DTI prediction in Schizophrenia. Its superior performance demonstrates the potential of DL in advancing drug discovery and development processes.

Emotion recognition (ER) plays a crucial role in enabling machines to perceive human emotional and psychological states, thus enhancing human-machine interaction. Recently, there has been a growing interest in ER based on electroencephalogram (EEG) signals. However, due to the noisy, nonlinear, and nonstationary properties of electroencephalography signals, developing an automatic and high-accuracy ER system is still a challenging task. In this study, a pretrained deep residual convolutional neural network model, including 17 convolutional layers and one fully connected layer with transfer learning technique in combination frequency-channel matrices (FCM) of two-dimensional data based on Welch power spectral density estimate from the one-dimensional EEG data has been proposed for improving the ER by automatically learning the underlying intrinsic features of multi-channel EEG data. The experiment result shows a mean accuracy of 93.61 ± 0.84%, a mean precision of 94.70 ± 0.60%, a mean sensitivity of 95.13 ± 1.02%, a mean specificity of 91.04 ± 1.02%, and a mean F1-score of 94.91 ± 0.68%, respectively using 5-fold cross-validation on the DEAP dataset. Meanwhile, to better explore and understand how the proposed model works, we noted that the ranking of clustering effect of FCM for the same category by employing the t-distributed stochastic neighbor embedding strategy is: softmax layer activation is the best, the middle convolutional layer activation is the second, and the early max pooling layer activation is the worst. These findings confirm the promising potential of combining deep learning approaches with transfer learning techniques and FCM for effective ER tasks.

Exercise-induced laryngeal obstruction (EILO) describes paradoxical laryngeal closure during inspiration at high-intensity exercise. It is hypothesised that during intense activity, the air-induced loads on supraglottic walls overcome their internal stiffness, leading to the obstruction. Recent investigations have revealed that the air-induced loads on the supraglottic walls vary nonlinearly with increasing flow rate. It is, however, unclear whether certain geometric configurations of the hypopharynx and larynx may contribute to the predisposition to EILO. This study investigates the influence of hypopharyngeal and laryngeal geometry on upper respiratory tract airflow and air-induced forces. A computational fluid dynamics model is developed to study airflow through larynx. Four real, adult upper respiratory tracts with variable configurations are considered. Two steady, uniform inspiratory flow rates of 60 L/min and 180 L/min are considered. The analysis shows that geometries with a space lateral to the epiglottis (EpiS) and piriform fossae (PF) directs the hypopharyngeal and supraglottic pressure field to remain positive and increase with the flow rate. In geometries with EpiS and PF, pressure differential occurs around the aryepiglottic fold producing a net inward force over the region. The three-fold increase in flow rate induces near ten-fold increases in force over the region which may facilitate the closure. It is concluded that hypopharyngeal anatomy, particularly the piriform fossae, play a significant role in the obstruction of the supraglottic airway and should be considered in research and clinical assessment of EILO.

The mechanical stresses and strains are examined, in ascending thoracic aortic aneurysm (aTAA) models, in a patient-specific aTAA as well as in healthy thoracic aortic models, via Finite Element Analysis. The aneurysms are assumed spherical, 1.5 mm thick, with diameters between 47 mm and 80 mm, eccentrically positioned. The geometry and wall thickness distribution of the aorta along its length are based on open literature data for an average patient age of 66.25 years, accounting for the Body Surface Area (BSA) parameter. The vessel wall material is assumed isotropic and incompressible, with its Young's modulus varying with the aneurysm diameter and the applied intraluminal pressure (120 mmHg to 240 mmHg). In the aTAAs, peak stresses were found to increase nonlinearly with aneurysm diameter (for a given pressure) tending to reach a plateau, appearing at the proximal area of the aneurysm, whereas lower stresses were found at its distal part and even smaller at the aneurysm maximum diameter. Regarding the patient-specific aTAA model, the peak stresses appeared at the distal part of the aneurysm where a tear of the intima layer was detected during surgical intervention. Peak strains exhibited for each pressure a maximum at a certain aneurysm diameter beyond which they dropped so that essentially the vessel wall's distensibility was thus reduced. Examining more than 100 geometry cases and employing a failure stress criterion, the rupture diameter thresholds were estimated to be 65, 52.5, 50 and 47.5 mm for a pressure of 120, 160, 200 and 240 mmHg respectively.

Recently, COVID-19 (coronavirus) has been a huge influence on the socio and economic field. COVID-19 cases are seriously increasing day-day and also don't identified proper vaccine for COVID-19. Hence, COVID-19 is fast spreading virus and it causes more deaths. In order to address this, the work has proposed a machine learning (ML) scheme for the prediction of COVID-19 positive, negative, and deceased instances. Initially, the data is pre-processed by eliminating redundant and missing values. Then, the features are selected using hybrid grey assisted whale optimization algorithm (H-GAWOA). Finally, the classifier ANFIS (adaptive network-based fuzzy inference systems) is used for investigating the confirmed, survival and death rate of COVID-19. The performance is analysed on John Hopkins University dataset and the performances like MSE, RMSE, MAPE, and R² are measured. In all the comparisons, the MSE value is very less for the proposed model. Particularly, in the deceased cases prediction, the MSE value is 0.00 for the proposed H-GAWOA-ANFIS. Finally, it is proved that the suggested model is able to generate the better results when contrast to the other approaches.

A proper and reliable fracture fixation is important for fracture healing. The proximal femoral intramedullary nail (IN), such as proximal femoral nail anti-rotation (PFNA) or Gamma nail, is widely used for intertrochanteric fracture fixation. However, it still suffers considerable stress concentrations, especially at the junction between the nail and the blade or lag screw. In this study, we propose a novel intramedullary nail design to enhance the intramedullary nail integrity by introducing a bolt screw to form a stable triangular structure composed of the nail, the lag screw, and the bolt screw (PFTN, Proximal femoral triangle nail). Systematic finite element numerical simulations were carried out to compare the biomechanical performances of PFTN and PFNA under both static and dynamic loads during the postures of ascending and descending stairs. The simulation results highlight the advantages of the proposed PFTN design with lower stresses, less stress concentration, and higher structure stability.

The classical compartment model is often used to study the spread of an epidemic with one virus. However, there are few types of research on epidemic models with multiple viruses. The article aims to propose two new deterministic and stochastic SIIIRS models with multiple viruses and saturation incidences. We obtain asymptotic properties of disease-free and several endemic equilibria for the deterministic model. In the stochastic case, we prove the existence and uniqueness of positive global solutions. The extinction and persistence of diseases are obtained under different threshold conditions. We analyze the existence of stationary distribution through a suitable Lyapunov function. The results indicate that the extinction or persistence of the two viruses is closely related to the intensity of white noise interference. Specifically, considerable white noise is beneficial for the extinction of diseases, while slight one can lead to long-term epidemics of diseases. Finally, numerical simulations illustrate our theoretical results and the effect of essential parameters.

Analysis of the musculoskeletal movements (gait analysis) is needed in many scenarios. The in vivo method has some difficulties. For example, recruiting human subjects for the gait analysis is challenging due to many issues. In addition, when plenty of subjects are required, the follow-up experiments take a long period and the dropout of subjects always occurs. An efficient and reliable in silico simulation platform for gait analysis has been desired for a long time. Therefore, a technique using three-dimensional (3D) muscle modeling to drive the 3D musculoskeletal model was developed and the application of the technique in the simulation of lower limb movements was demonstrated. A finite element model of the lower limb with anatomically high fidelity was developed from the MRI data, where the main muscles, the bones, the subcutaneous tissues, and the skin were reconstructed. To simulate the active behavior of 3D muscles, an active, fiber-reinforced hyperelastic muscle model was developed using the user-defined material (VUMAT) model. Two typical movements, that is, hip abduction and knee lifting, were simulated by activating the responsible muscles. The results show that it is reasonable to use the improved CFD-FE method proposed in the present study to simulate the active contraction of the muscle, and it is feasible to simulate the movements by activating the relevant muscles. The results from the present technique closely match the physiological scenario and thus the technique developed has a great potential to be used in the in silico human simulation platform for many purposes.

A classification framework for hand gestures using Electromyography (EMG) signals in prosthetic hands is presented. Leveraging the multi-scale characteristics and temporal nature of EMG signals, a Convolutional Neural Network (CNN) is used to extract multi-scale features and classify them with spatial-temporal attention. A multi-scale coarse-grained layer introduced into the input of one-dimensional CNN (1D-CNN) facilitates multi-scale feature extraction. The multi-scale features are fed into the attention layer and subsequently given to the fully connected layer to perform classification. The proposed model achieves classification accuracies of 93.4%, 92.8%, 91.3%, and 94.1% for Ninapro DB1, DB2, DB5, and DB7 respectively, thereby enhancing the confidence of prosthetic hand users.

Cancer-associated fibroblasts (CAFs) are an important component of the tumor microenvironment that contribute toward the development of tumors. This study aimed to establish a new algorithm based on CAF scores to predict the prognosis and immunotherapy response in patients with lung squamous cell carcinoma (LUSC). The RNA-seq data of LUSC patients were obtained from two databases and merged after removing inter-batch differences. The CAF-related data for each sample were obtained through three different algorithms. Consistency cluster analysis was performed to obtain different CAF clusters, which were analyzed to identify differentially expressed genes. These were subjected to uniform cluster analysis to obtain different gene clusters. The Boruta algorithm was used to calculate the CAF score. Three CAF clusters and two gene clusters were obtained, all of which differed in their patient prognoses and the content of infiltrating immune cells. Patients with high CAF scores exhibited worse overall survival, higher expression of biomarkers related to immune checkpoints and immune activity, and lower tumor mutation burden. The CAF score could also predict the immunotherapy response of patients. This study suggests that the CAF score can accurately predict the prognosis and immunotherapy response of LUSC patients.