Computer Methods in Biomechanics and Biomedical Engineering最新文献

To address the limitation that prevailing cross-domain Parkinson's disease (PD) speech recognition methods mitigate small-sample issues via distribution matching while disregarding substantial feature overlap, we introduce the cross-domain Fisher criterion (CFC). CFC reformulates inter- and intra-class scatter matrices to align with classifier properties: the former enhances separation among heterogeneous target-domain samples, whereas the latter compactly aggregates homologous cross-domain samples around the target centroid and suppresses inter-domain disparities. Numerous experimental results demonstrate that CFC is an effective, efficient, and robust method for PD speech recognition, offering a promising approach for integration into PD speech-based remote rehabilitation and monitoring systems..

Accurate prediction of gestational diabetes mellitus (GDM) is critical for improving maternal and fetal outcomes. This study develops a Transformer-based multimodal fusion model that integrates tabular clinical features and image-encoded electronic health records (EHRs), aiming for accurate end-to-end classification of GDM. Preprocessed EHRs were transformed into grayscale, RGB, and heatmap, with visual features were extracted by a Vision Transformer and tabular features by an MLP. A modality-aware attention mechanism enhances cross-modal fusion. Evaluated on two public datasets, performance gains over the strongest single-modality models reached 3.95% and 0.38% in accuracy.

Renal artery stenosis (RAS) is a major cause of secondary hypertension, requiring accurate hemodynamic evaluation for clinical intervention. This study presents a deep learning framework integrating Mamba-based state-space modeling (SSM) with hierarchical point cloud processing for real-time hemodynamic prediction. A computational dataset was generated from three-dimensional renal artery models using Bessel-curve reconstruction and computational fluid dynamics (CFD) simulations. By combining PointNet++ with Mamba's selective mechanisms, the model effectively captures hemodynamic metrics while preserving local vascular features. The method provides real-time renal hemodynamic predictions with computational efficiency improved by several orders of magnitude while preserving accuracy comparable to CFD.

This study aims to identify mitochondrial-related biomarkers for sarcopenia and Alzheimer's disease (AD). Through the GEO database and machine learning algorithms, three diagnostic biomarkers (FKBP5, PRKAG1, and FBP2) were identified, and the constructed models showed good diagnostic accuracy in both internal and external datasets. Furthermore, multiple analyses such as GSEA, immune infiltration, and drug prediction were conducted. These findings provide new insights into shared pathological mechanisms and clinical diagnosis and treatment of sarcopenia and AD. Further clinical and experimental studies are needed to validate these results.

This study aims to explore the role of necroptosis in osteoporosis and identify potential diagnostic biomarkers. By analyzing the GSE56815 and GSE7429 datasets, we identified 107 differentially expressed genes associated with necroptosis. Enrichment analysis revealed that these genes were significantly enriched in necroptosis, the NOD-like receptor signaling pathway, and the IL-17 signaling pathway. Furthermore, through protein-protein interaction network analysis, multiple algorithms (MCC and MCODE) screening, and LASSO regression modeling, we ultimately established a diagnostic model consisting of 13 key genes. In vitro cell experiments suggest that CASP3 may serve as a potential target for Minocycline in the treatment of osteoporosis.

This study presents a robust deep learning framework for automatic motor imagery detection from raw EEG signals. Six band-power features were extracted using STFT, and dedicated 2D CNN-LSTM models were trained for each band. Their outputs were fused using a Choquet fuzzy integral to enhance decision reliability under noisy EEG conditions. Alpha- and sigma-band models achieved 88% and 87.1% accuracy, respectively. The fused architecture reached 90.4% on BCI IV-2a and 92.21% on BCI IV-1, outperforming existing methods in motor imagery classification.

Towards 120° spiral notch pattern, this paper proposes a finite element model to characterize the nonlinear deformation and designs an inconsistent-configuration device to achieve constant-curvature bending. Effects of the notch geometry on bending stiffness are analyzed using the model. By deriving bending stiffness equations and modeling cable friction, the non-constant curvature bending behavior of the device are characterized. A nonuniform pattern is designed to maintain a constant force-to-stiffness ratio, ensuring uniform curvature. The optimized design reduces tip displacement errors to <1.81 mm under constant-curvature assumptions. This work provides a systematic approach to modeling and optimizing notched devices for neurosurgical applications.

Neck pain is a highly prevalent condition in the general population, but the addition of helmet loads contributes to the neck pain that 97% of fighter pilots report. The mechanism for neck pain in pilots is unknown; however, there is likely nerve root (NR) dysfunction involved. This study implemented a spinal cord, NRs, and novel foraminal ligaments into the previously validated VIVA OpenHBM. After performing sensitivity analyses and validating added tissues, helmet loads and muscle forces were applied. The results indicate that helmet loads generate compression on NRs that temporarily impair blood flow and impulse propagation in a neutral posture.

Bicondylar tibial plateau fractures are often managed with plate fixation clinically, but no unified standard for fixation mechanical stability evaluation exists. This study investigated the mechanical properties of 15 internal fixation schemes via finite element analysis, simulating three postoperative stages to acquire four key indicators: medial/lateral fracture fragment displacements, internal fixation stress, and fracture site stress shielding rate. The coefficient of variation was used to determine the objective weights of these indicators, establishing a multi-indicator integrated mechanical stability evaluation system. Results showed that medial plate fixation exhibited superior stability, providing a valuable reference for clinical selection of fixation schemes.

Ice hockey is a sport with a high incidence of concussion, but growing attention has been placed on the broader exposure to head impacts, regardless of clinical symptoms, due to their potential cumulative effects. While player education, protective equipment, and rule modifications have been introduced to reduce head impact risk, the effect of rule changes on head impact exposure in adult professional ice hockey has not been well quantified. This study compared the frequency and magnitude of head impacts between two NHL seasons of 2003-04 and 2016-17, that bracketed a period of rule changes targeting contact to the head. Twenty regular-season games from each season were analyzed using a combination of video analysis, physical reconstructions, and finite element modeling to estimate brain tissue strain for each head impact. Total head impact frequency per game did not differ significantly between seasons. However, a significant decrease in glove-to-head contacts and fight-related impacts was observed in 2016-17, reflecting the enforcement of stricter penalties for these actions. Additionally, players in the 2016-17 season experienced a significantly higher frequency of impacts classified within the Low brain strain category. These results suggest that while targeted rules may reduce specific types of dangerous contact, they may also shift the nature of impacts without reducing overall exposure. Understanding how such shifts influence cumulative biomechanical load is essential for guiding future rule evaluations and injury prevention strategies in professional ice hockey.