生物医学工程学杂志最新文献

Motor imagery electroencephalogram (MI-EEG) decoding algorithms face multiple challenges. These include incomplete feature extraction, susceptibility of attention mechanisms to distraction under low signal-to-noise ratios, and limited capture of long-range temporal dependencies. To address these issues, this paper proposes a multi-branch differential attention temporal network (MDAT-Net). First, the method constructed a multi-branch feature fusion module to extract and fuse diverse spatio-temporal features from different scales. Next, to suppress noise and stabilize attention, a novel multi-head differential attention mechanism was introduced to enhance key signal dynamics by calculating the difference between attention maps. Finally, an adaptive residual separable temporal convolutional network was designed to efficiently capture long-range dependencies within the feature sequence for precise classification. Experimental results showed that the proposed method achieved average classification accuracies of 85.73%, 90.04%, and 96.30% on the public datasets BCI-IV-2a, BCI-IV-2b, and HGD, respectively, significantly outperforming several baseline models. This research provides an effective new solution for developing high-precision motor imagery brain-computer interface systems.

Although transcranial magnetic stimulation (TMS) is widely used in neuromodulation, conventional TMS struggles to achieve both depth and focal specificity. Temporal interference TMS (TI-TMS) offers a promising approach to enhance stimulation depth while reducing the focal area; however, current research remains largely simulation-based, with limited studies on system implementation and experimental validation in rodent deep brain regions. To address this, we developed a TI-TMS system based on a realistic mouse head model using finite element simulation. Electrophysiological recordings of local field potentials (LFPs) in the ventral hippocampal (vHPC) formation were performed to evaluate changes in θ rhythm power spectral density (PSD) and θ-γ phase-amplitude coupling (PAC) following stimulation. The results demonstrated that TI-TMS enhanced θ rhythm power and strengthened θ-γ PAC, indicating effective modulation of deep brain regions. This study establishes a functional TI-TMS system capable of effectively stimulating deep vHPC, providing an experimental basis for its application in precise neuromodulation of subcortical brain areas.

40 Hz light flicker stimulation is deemed to hold considerable promise in the treatment of Alzheimer's disease (AD). However, whether its long-term effect can improve working memory and its related mechanisms remains to be further explored. In this study, 21 adult Wistar rats were randomly divided into the AD light-stimulation group, the AD group and the control group. AD models were established in the first two of these groups, with the light-stimulation group receiving long-term 40 Hz light flicker stimulation. Working memory performance across groups was subsequently evaluated using the T-maze task. To investigate the potential neural mechanisms underlying the effects of 40 Hz light stimulation on working memory, we examined changes in neuronal excitability within the hippocampus (HPC) and medial prefrontal cortex (mPFC), as well as alterations in inter-regional synchronization of neural activity. The findings demonstrated that prolonged 40 Hz light stimulation significantly improved working memory performance in AD model rats. Furthermore, the intervention enhanced the synchronization of neural activity between the hippocampus (HPC) and medial prefrontal cortex (mPFC), as well as the efficiency of information transfer, primarily mediated by theta and low-frequency gamma oscillations. This study provides theoretical support for exploring the mechanisms of 40 Hz light flicker stimulation and its further clinical application in the prevention and treatment of Alzheimer's disease.

The incidence of wounds or skin defects caused by trauma, infection, diabetes, and other factors has been increasing year by year, imposing a substantial burden on global healthcare systems. Metal-organic frameworks (MOFs) are nanomaterials formed by metal ions or metal clusters and organic ligands through coordination bonds, featuring high porosity, large specific surface area, tunable structure, and excellent biocompatibility. MOFs can regulate cellular behaviors and kill bacteria by releasing metal ions during degradation. Additionally, MOFs can act as carriers for delivering bioactive components to exert anti-inflammatory, antioxidant, and cell proliferation-promoting effects. By systematically reviewing relevant domestic and international literature, this paper summarized the synthesis methods, classification, and application strategies of various MOFs in the field of skin repair. On this basis, it also concluded the current challenges in this field and provided an outlook on its future development trends.

Warfarin, a classic oral anticoagulant, is characterized by a narrow therapeutic window and considerable interindividual variability in dosing requirements. This makes precise dose adjustment challenging in clinical practice and increases the risk of bleeding or thrombosis. To improve dose prediction, this study developed a streamlined multilayer perceptron (MLP) model using real-world data from the International Warfarin Pharmacogenomics Consortium (IWPC) database. The LASSO-proj algorithm was applied for high-precision feature selection prior to model construction. The resulting model demonstrated strong predictive performance on the test set, achieving a coefficient of determination ( R ²) of 0.456, a mean absolute error (MAE) of 8.92 mg/week, and 48.522% of its predictions falling within ±20% of the actual stable therapeutic dose. Through SHAP-based interpretation using DeepExplainer, key features influencing warfarin dosing were identified, including the VKORC1 genotype, body weight, age, and ethnicity. The interpretable MLP framework incorporating LASSO-proj not only maintains high predictive accuracy, but also significantly enhances model transparency, providing a valuable tool for guiding warfarin therapy.

Automated detection of myocardial infarction (MI) is crucial for preventing sudden cardiac death and enabling early intervention in cardiovascular diseases. This paper proposes a deep learning framework based on a lightweight convolutional neural network (CNN) combined with one-dimensional gradient-weighted class activation mapping (1D Grad-CAM) for the automated detection of MI and the visualization of key waveform features in single-lead electrocardiograms (ECGs). The proposed method was evaluated using a total of 432 records from the Physikalisch-Technische Bundesanstalt Diagnostic ECG Database (PTBDB) and the Normal Sinus Rhythm Database (NSRDB), comprising 334 MI and 98 normal ECGs. Experimental results demonstrated that the model achieved an accuracy, sensitivity, and specificity of 95.75%, 96.03%, and 95.47%, respectively, in MI detection. Furthermore, the visualization results indicated that the model's decision-making process aligned closely with clinically critical features, including pathological Q waves, ST-segment elevation, and T-wave inversion. This study confirms that the proposed deep learning algorithm combined with explainable technology performs effectively in the intelligent diagnosis of MI and the visualization of critical ECG waveforms, demonstrating its potential as a useful tool for early MI risk assessment and computer-aided diagnosis.

This study proposes an automated neurofibroma detection method for whole-body magnetic resonance imaging (WBMRI) based on radiomics and ensemble learning. A dynamic weighted box fusion mechanism integrating two dimensional (2D) object detection and three dimensional (3D) segmentation is developed, where the fusion weights are dynamically adjusted according to the respective performance of the models in different tasks. The 3D segmentation model leverages spatial structural information to effectively compensate for the limited boundary perception capability of 2D methods. In addition, a radiomics-based false positive reduction strategy is introduced to improve the robustness of the detection system. The proposed method is evaluated on 158 clinical WBMRI cases with a total of 1,380 annotated tumor samples, using five-fold cross-validation. Experimental results show that, compared with the best-performing single model, the proposed approach achieves notable improvements in average precision, sensitivity, and overall performance metrics, while reducing the average number of false positives by 17.68. These findings demonstrate that the proposed method achieves high detection accuracy with enhanced false positive suppression and strong generalization potential.

The human femur is in a relatively complex mechanical environment, subject to the combined effects of multiple factors such as mechanical loads from movement and weight-bearing, as well as changes in the body fluid environment in daily life. In in vitro testing cases of the femur (e.g., testing of distal femoral fractures), changes in load conditions usually significantly affect the mechanical properties of the overall structure. However, there is currently no systematic evaluation standard for in vitro mechanical performance testing of the femur. Therefore, this paper established four human femur models (model A~model D) constructed based on computed tomography (CT) under different load environments, as well as two artificially synthesized femur models (the finite-element model and the experimental model) under the same load environment. Among them, for the human femur models, model A was configured to apply hip joint contact forces together with all muscle forces to approximate the real in vivo mechanical environment, model B was applied with hip joint contact force and abductor muscle force, model C was only applied with hip joint contact force, and model D was subjected to an equivalent resultant force. For the artificially synthesized femur models, both the finite-element model and the experimental model were applied with the same equivalent resultant force as model D. Comparative analyses revealed that model D exhibited femoral head displacement and stress-strain distributions similar to Model A, indicating its suitability as an equivalent in vitro test model. Further comparison between the finite-element and experimental synthetic femur models yielded consistent mechanical responses, thereby validating the equivalent model. In summary, it is hoped that the findings of this study will provide a reference for establishing a systematic, tiered preclinical evaluation system for hip prostheses/implants in the future.

Quantitative magnetic susceptibility imaging (QSM) is an imaging method based on magnetic resonance imaging (MRI) phase signal processing and inversion to obtain tissue magnetic susceptibility distribution, which can generate images reflecting the magnetic characteristics of tissues. QSM reconstruction process is complex, in which dipole inversion stage is the most challenging and decisive link, and traditional methods are easily affected by pathological conditions at this stage, resulting in artifacts and deviations. With the development of deep learning and machine vision technology, using U-network (U-Net) model to improve dipole inversion process can effectively avoid the shortcomings of traditional algorithms. In this paper, the application of the improved model based on U-Net architecture in dipole inversion from 2020 to now is summarized. Firstly, the theoretical concept of QSM is introduced. Secondly, the existing improved models based on U-Net architecture are divided into three categories: improved U-Net based on structural optimization, improved U-Net based on physical constraints and improved U-Net based on improving generalization ability, and their main characteristics and design starting points are sorted out. Finally, the development trend of the future model is prospected and summarized. To sum up, it is expected that the difficulties and challenges of dipole inversion will be solved, the accuracy of QSM images will be improved, and support for disease-aided diagnosis will be provided by summarizing and comparing different improved U-Net models in this paper.

With the intensification of global aging trends and the continuous rise in the incidence of chronic diseases, the demand for health monitoring and early intervention has become increasingly urgent. Owing to their non-invasive nature, portability, and comfort, flexible wearable sensors have emerged as a key technology driving the development of personalized healthcare. Starting from specific application scenarios in health monitoring, this article systematically reviews recent research advances in flexible sensors within the healthcare field. Firstly, it outlines the design fundamentals of flexible sensors. This is followed by a focused analysis of their specific applications in monitoring vital signs, biochemical markers, as well as motion and neural activities, along with an in-depth exploration of the clinical significance, technical challenges, and targeted solutions in different scenarios. Finally, the current technical bottlenecks and clinical challenges are summarized, and an outlook on the future development of health monitoring systems is provided. This review aims to provide a systematic reference for the deep integration of flexible electronics technology and medicine.