Computer Methods in Biomechanics and Biomedical Engineering最新文献

Atherosclerosis in the carotid arteries increases stroke risk, yet current treatment decisions rely mainly on stenosis degree, which poorly reflects individual vulnerability. We present an ultrasound-based computational fluid dynamics (CFD) framework for patient-specific hemodynamic assessment. Using tracked 2D ultrasound and automated segmentation, we reconstructed carotid geometries for five healthy subjects and three patients with severe stenoses. CFD simulations quantified TAWSS, OSI, RRT, and helicity, visualized through risk maps. Healthy arteries showed localized risk near bifurcations, whereas stenosed geometries exhibited extensive disturbed flow and altered helicity patterns. This approach demonstrates the feasibility of ultrasound-driven CFD for personalized risk mapping and highlights helicity's potential as a diagnostic marker.

The defense against microbial pathogens can be functionalized by leukocytes because via singling immune response to enhance Inflammation. In this manuscript, a dynamical analysis for the concentration of circulating white blood cells is functionalized by fractional differential operators. The mathematical investigations for fractionalized and non-fractionalized concentration of circulating white blood cells have been traced out. The comparative analysis of circulating white blood cells has been discussed for delay between white blood cell productions. Finally, our results suggested that the hemogram reflects blood-clotting disorders and infection on the basis of fractionalized and non-fractionalized concentration of circulating white blood cells.

This study investigated an upper-extremity exoskeleton for machine learning-based discrimination of orthopedic shoulder pathology and identification of discriminative temporal features. Twelve patients with shoulder impairments and thirty healthy controls performed eight standardized tasks. Logistic regression with stratified 5-fold cross-validation was used for classification. Temporal effect sizes were computed using pointwise Cohen's d, and permutation-based phase ablation quantified the contribution of movement phases to AUROC. Classification performance ranged from 0.70 to 1.00, with six tasks achieving AUROC ≥ 0.90. Mid-cycle phases dominated in flexion and abduction tasks, whereas early and late phases were most informative for rotational movements, supporting interpretable, phase-aware ML models.

This paper proposes a modified Quantum Dilated Convolutional neural network (QDCNN) to detect cancer using gene expression data. Primarily, the input gene expression data is taken from a specified dataset. Then, data transformation is done using Adaptive Box-Cox transformation and feature fusion is done by a Deep Neural Network (DNN) with Kulczynski. The refined features are then fed into the modified QDCNN, which effectively predicts cancer. The modified QDCNN attains an accuracy of 90.6%, a True Positive Rate (TPR) of 89.0%, False Negative Rate (FNR) of 0.109, and a Matthews correlation coefficient (MCC) of 89.9% when using the PANCAN dataset.

Purpose: To optimize scoliosis correction strategies by comparing continuous and interval pedicle screw configurations and proposing a dual-geometry screw design.

Methods: A patient-specific T11-L5 scoliotic spine model was reconstructed via finite element analysis (FEA). Continuous and interval screw placements were evaluated for biomechanical performance. A novel dual-geometry screw (tapered-cylindrical transition) was developed.

Results: Continuous configurations achieved a 43.5% reduction in displacement (1.33 mm vs. 2.36 mm) and a 29.7% decrease in screw stress (444.08 MPa vs. 631.35 MPa). The dual-geometry screw lowered drilling stress (16.5%, p < 0.05) and displacement heterogeneity (22.4%).

Conclusion: Continuous screws enhance stability through synergistic load transfer, while dual-geometry screws mitigate interfacial damage. This provides biomechanical criteria for clinical scoliosis correction.

The aim of this study is to examine the mechanical behavior of different types of composite resins (short fiber-reinforced composite, conventional high-fill hybrid composite and bulk-fill composite) used in the restoration of class II MOD cavities of primary molar teeth by the finite element analysis (FEA). Three three-dimensional tooth models were created in a computer environment. Model 1: tooth model without restoration (control group), Model 2: class II MOD cavity tooth model restored using composite resin (incremental technique), and Model 3: class II MOD cavity tooth model restored using composite resin (bulk technique). Subgroups were formed using the properties of different types of composite resins tested in the class II MOD cavity tooth model. To simulate the average bite force in a child with primary dentition, vertical static loading of 245 N was applied to each of the occlusal contact points of the models. The maximum von Mises stress values were calculated for the models. For all models, the von Mises stress values obtained in enamel were higher than those obtained in dentin. Similar von Mises stress values were obtained in all subgroups of Model 2. The lowest von Mises stress values transmitted to the dental tissues were obtained in Model 3.

In this work, a novel framework is proposed which includes Hjorth parameters as features from time and time-frequency domain (Multi-Domain) and attention-enhanced temporal modeling, to classify epileptic seizure stages, namely normal, inter-ictal, and ictal. Three different approaches are compared, i.e. Hjorth parameters in time domain, time-frequency domain, and multi-domain. In time-frequency domain, Hjorth parameters are derived from the wavelet coefficients obtained using Discrete Wavelet Transform (DWT). The extracted features are then fed to a 1D Convolutional Neural Network (CNN), Bidirectional Long Short-Term Memory (BiLSTM), and attention mechanism. The performance of the proposed framework is evaluated on Bonn EEG dataset using different performance evaluation metrics namely precision, recall, F1-score, and accuracy. The binary, three-class, and five-class seizure classification are examined using the proposed framework. The validation of the model is performed through the 10-fold cross-validation with sample level partitioning. Experimental findings show that the proposed framework with multi-domain features has given outstanding performance with 98.40, 98.00, and 85.40% test classification accuracy for binary, three-class, and five-class discrimination, respectively.

This study proposes a human motion measurement model combining molecular chain conformation with a silicone rubber strain sensor embedded with carbon nanotubes to enhance signal response stability. An improved least mean square algorithm is used to optimize signal processing. Experimental results show the model achieves 95.12% measurement accuracy, 92.45% F1 score, 35.14 dB SNR, and 60.45 ms latency. Across different age groups and motion states such as gait, running, and jumping, the average detection error remains below 3%, and physiological monitoring errors for heart rate and oxygen saturation are as low as 0.42. The model operates stably in dynamic conditions.

Machine learning (ML) has been used to predict lower extremity joint torques from joint angles and surface electromyography (sEMG) signals. This study trained three bidirectional Long Short-Term Memory (LSTM) models, which utilize joint angle, sEMG, and combined modalities as inputs, using a publicly accessible dataset to estimate joint torques during normal walking and assessed the performance of models, that used specific inputs independently plus the accuracy of the joint-specific torque prediction. The performance of each model was evaluated using normalized root mean square error (nRMSE) and Pearson correlation coefficient (PCC). Each model's median scores for the PCC and nRMSE values were highly convergent and the bulk of the mean nRMSE values of all joints were less than 10%. The ankle joint torque was the most successfully predicted output, having a mean nRMSE of less than 9% for all models. The knee joint torque prediction has reached the highest accuracy with a mean nRMSE of 11% and the hip joint torque prediction of 10%. The PCC values of each model were significantly high and remarkably comparable for the ankle (∼ 0.98), knee (∼ 0.92), and hip (∼ 0.95) joints. The model obtained significantly close accuracy with single and combined input modalities, indicating that one of either input may be sufficient for predicting the torque of a particular joint, obviating the need for the other in certain contexts.

Alzheimer's disease (AD) is the most prevalent neurodegenerative disease. There are currently no effective interventions to slow down or prevent the occurrence and progression of AD. Neutrophil extracellular traps (NETs) have been proven to be tightly linked to AD. This project attempted to identify hub genes for AD based on NETs. Gene expression profiles of the training set and validation set were downloaded from the Gene Expression Omnibus (GEO) database, including non-demented (ND) controls and AD samples. NET-related genes (NETRGs) were collected from the literature. Differential analysis identified 21 AD differentially expressed NETRGs (AD-DE-NETRGs) majorly linked to functions such as defense response to bacterium as well as pathways including IL-17 signaling pathway, as evidenced by enrichment analyses of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG). Protein-protein interaction (PPI) network, Minutia Cylinder-Code (MCC) algorithm, and molecular complex detection (MCODE) algorithm in the CytoHubba plug-in were employed to identify five hub genes (NFKBIA, SOCS3, CCL2, TIMP1, ACTB). Their diagnostic ability was validated in the validation set using receiver operating characteristic (ROC) curves and gene differential expression analysis. A total of 16 miRNAs and 132 lncRNAs were predicted through the mirDIP and ENCORI databases, and a lncRNA-miRNA-mRNA regulatory network was constructed using Cytoscape software. Small molecular compounds such as Benzo(a)pyrene and Copper Sulfate were predicted to target hub genes using the CTD database. This project successfully identified five hub genes, which may serve as potential biomarkers for AD, proffering clues for new therapeutic targets.