The mechanisms underlying the impact of metabolic syndrome on cognitive dysfunction in patients with schizophrenia remain unclear. The present study employed a two-factor factorial design to investigate the effects of metabolic syndrome on white matter microstructure in schizophrenia and its association with cognitive function. A total of 187 participants were included and classified into four groups based on the diagnoses of schizophrenia and metabolic syndrome: schizophrenia patients with metabolic syndrome (SZ-wMS), schizophrenia patients without metabolic syndrome, healthy controls with metabolic syndrome, and healthy controls without metabolic syndrome. Diffusion tensor imaging data were acquired. Using diffusion tensor model, fractional anisotropy (FA) was calculated to characterize the microstructural integrity of white matter. Peripheral metabolic indices and multiple domains of cognitive function were also assessed. The SZ-wMS group showed further reduced FA in the right sagittal stratum. Within the two-factor analytical framework, an interaction effect between metabolic syndrome and schizophrenia on FA in the right sagittal stratum was identified. Correlation analyses revealed that reduced FA in the right sagittal stratum was associated with impaired language function in patients with schizophrenia. Moreover, mediation analysis indicated that body mass index might indirectly affect language function by influencing FA in the right sagittal stratum. In summary, reduced integrity of white matter fibers in sagittal stratum may represent a potential neural mechanism underlying the comorbidity of schizophrenia and metabolic syndrome and may be associated with language dysfunction in schizophrenia.
Pilonidal sinus disease (PSD) is a suppurative inflammatory condition characterized by a sinus tract in the sacrococcygeal gluteal cleft and surrounding skin. Surgical resection is an important treatment for PSD; however, postoperative recurrence occurs in some patients, severely affecting their quality of life. Due to the lack of objective clinical indicators for predicting recurrence, developing an artificial intelligence-based prediction model is an effective approach. In this study, postoperative hematoxylin and eosin (H&E) stained pathological sections from PSD patients were used. The horizontal and vertical distance network (HoVer-Net) deep learning model for automated nuclei segmentation and classification was employed to extract nuclear features from lesion regions. These features were then integrated with multidimensional clinical characteristics to select key features. And then a machine learning classifier was incorporated to construct a recursive feature elimination-balanced random forest (RFE-BalancedRF) ensemble model to predict PSD recurrence. Experimental results demonstrate that the proposed PSD recurrence prediction model achieves an area under the curve of receiver operating characteristic (ROC_AUC) of 0.751 ± 0.048 and an accuracy (Acc) of 0.715 ± 0.066. Furthermore, a correlation exists between multimodal features and the risk of PSD recurrence, confirming the effectiveness of the proposed model in predicting PSD recurrence. This study innovatively integrates radiomics features with clinical characteristics from the dual perspectives of pathological image analysis and multimodal feature fusion to construct a PSD recurrence risk prediction model, potentially providing a new pathway with considerable clinical translation potential for artificial intelligence assisted medicine.
The evaluation of disability grades in traffic accidents is a professional forensic clinical appraisal matter, and its results directly affect the fairness of judicial compensation. In the construction of automated disability grade evaluation models, the imbalanced distribution of disability cases leads to low recognition accuracy for minority categories, becoming a key bottleneck restricting the technology's implementation. In response, this paper proposes an imbalanced data classification method based on a hybrid parameter scaling weight optimization mechanism. First, a loss weight calculation model is constructed based on category proportion, category sparsity, and category diversity. Second, the loss weight calculation model is designed by integrating the focal loss function's ability to focus on hard samples with the cross-entropy loss function's global gradient stability advantage. Then, at the early stages of training, the model proposed in this paper aligns sensitivity to imbalanced categories and constructs a low-computational-demand hybrid parameter scaling weight optimization mechanism. Experimental results show that, compared with the best-performing baseline methods, the proposed method significantly improves both accuracy and macro-F1 score on the traffic accident disability grade dataset. It can effectively enhance the classification performance of minority grade categories in imbalanced data and help improve the accuracy of automated appraisal in judicial identification of traffic accident disability grades.
Breast whole slide image (WSI) serves as an essential basis for breast cancer diagnosis and subtype classification. However, its high resolution, multi-scale structural characteristics, and arbitrary orientation pose substantial challenges for automated analysis. Existing deep learning methods typically rely on large amounts of annotated data and struggle to handle the rotational variability and scale differences inherent in pathological images, which limits their generalization ability in cross-center settings. To address these issues, this study proposes a self-supervised learning model, IAM-BYOL, which incorporated geometric priors into the representation learning process. Building upon the BYOL framework, the model introduced an isotropic attention module (IAM). By employing discrete rotation group convolution, IAM enabled weight sharing across different rotated versions of the convolutional kernels, endowing the encoder with structural rotation equivariance. A subsequent group pooling operation converted the equivariant features into rotation-invariant isotropic representations. In addition, a multi-scale attention mechanism adjusted feature weights adaptively according to responses from different receptive fields, allowing the model to capture informative patterns ranging from nuclear-level details to tissue-level organization. Experimental results demonstrated that IAM-BYOL achieved classification accuracies of 98.74%, 99.04%, 99.01%, and 98.63% on the BreakHis dataset at 40×, 100×, 200×, and 400× magnifications, respectively, while attaining an accuracy of 93.02% on the cross-center private clinical dataset BCD. These findings indicate that introducing geometric inductive biases into pathological image representation learning can effectively enhance model robustness and generalization capability.
Auditory neurofeedback (ANF) utilizes sound to map brain activity in real-time, guiding individuals to self-regulate neural states via operant conditioning. Compared to visual neurofeedback, ANF offers distinct advantages in millisecond-level response speed and non-visual dependency, significantly enhancing ecological validity and reducing visual fatigue. This paper reviews the physiological basis, signal processing workflows, and three typical experimental paradigms of ANF: threshold regulation, parameter modulation, and state-dependent triggering. Furthermore, it discusses current applications in neurorehabilitation, psychiatric treatment, sports medicine, and auditory cognitive screening. Finally, the paper analyzes challenges regarding mechanistic evidence and parameter standardization, and prospects future trends such as ear electroencephalography-based portable design and multimodal fusion, providing theoretical insights for clinical translation.
Nanosecond pulsed electric field (nsPEF) exposure can disrupt and disaggregate amyloid-β, indicating its potential to improve symptoms of Alzheimer's disease. However, the propagation and distribution patterns of nsPEF within brain tissue remain insufficiently understood, making related simulation analysis necessary. In this study, a high-resolution three-dimensional human head model incorporating the scalp, skull, cerebrospinal fluid, gray matter, white matter, and hippocampus was constructed. Based on the spectral characteristics of nsPEF, the dielectric properties of human tissues at different frequency ranges were assigned, and a transient finite-element model of nsPEF exposure in the human brain was established. The simulation analysis identified two optimal electrode-pair positions and characterized the spatial distributions of intracranial electric field strength as well as current density. It further elucidated the dependence of the hippocampal electric field response and current density on pulse parameters. In addition, a physical human brain model was constructed to experimentally validate the finite-element simulation results. The results showed that transcranial nsPEF can reach deep brain regions with extremely narrow pulse widths, and pulsed electric fields with kilovolt-level amplitudes and nanosecond-scale pulse widths can generate electric field strengths of approximately 10 3 V/m in the hippocampus. In summary, this work provides a theoretical basis and experimental support for optimizing the electrode configuration and stimulation parameters of transcranial nsPEF, thereby laying a foundation for future research on its application in non-invasive physical interventions for Alzheimer's disease.
Attention-deficit/hyperactivity disorder (ADHD) is a common neurodevelopmental disorder. Adults with ADHD continue to exhibit deficits in attention and executive function, whereas the neural mechanisms underlying visual mismatch negativity (vMMN)-related processing remain unclear. This study investigated the functional brain connectivity characteristics of adults with ADHD by combining standardized low-resolution brain electromagnetic tomography (sLORETA), phase-locking value (PLV) analysis, and a graph convolutional network (GCN) based on a visual Oddball paradigm that elicited vMMN responses. Electroencephalography (EEG) data were collected from 10 adults with ADHD and 10 healthy controls using a 128-channel recording system. Source signals from 68 brain regions defined by the Desikan-Killiany atlas were reconstructed using sLORETA. Functional connectivity networks were constructed using PLV and subsequently classified by the GCN model. The results showed that the accuracy, precision, recall, and F1-score of the GCN model under five-fold cross-validation were (85.13 ± 1.94)%, (80.58 ± 2.08)%, (86.04 ± 1.76)%, and (83.21 ± 1.89)%, respectively. Node feature weights and classification contribution analyses identified the lingual gyrus, calcarine fissure and surrounding cortex, parahippocampal gyrus, and precuneus as highly discriminative brain regions. These findings indicate that adults with ADHD exhibit abnormal functional connectivity patterns during vMMN-related processing and provide evidence for the auxiliary identification and neural mechanism investigation of ADHD.
To address the issues of time-consuming manual tooth alignment, reliance on physician experience, and insufficient structural constraints in automated methods in traditional orthodontic treatment planning, this paper proposes an automated tooth alignment network model that integrates multi-source geometric information learning. First, this study uses tooth mesh data as input to construct a graph convolution-based network for extracting tooth geometric features, fully utilizing the topological connectivity and local geometric features of the tooth model to enhance the extraction of tooth morphological features. Then, an arch line prediction network is designed to guide the teeth to align orderly along the natural arch line by explicitly modeling the tooth alignment trajectory, thereby ensuring the anatomical rationality and aesthetics of the overall structure. Based on this, this study designs collision avoidance loss and arch alignment loss to reduce geometric conflicts between adjacent teeth and constrain the overall posture, making the alignment results more stable and realistic. To verify the effectiveness of the method, validation experiments were conducted on multiple sets of real dental arch data, and comparisons were made with mainstream methods such as tooth alignment network (TaligNet), parameterized spatial transformation network (PSTN), and Transformer-based network for tooth alignment (TANet). The results show that the proposed method improves the evaluation metrics to varying degrees, fully integrates global and local geometric information, improves tooth posture coordination and alignment smoothness, and provides efficient and standardized intelligent support for digital orthodontic treatment planning.
Coronary artery segmentation is a critical step in the clinical diagnosis of coronary heart disease. To tackle segmentation breaks and false segmentation caused by the thin and complex coronary vessels as well as the severe foreground-background imbalance in computed tomography angiography images, this paper proposes MSCNet, a coronary artery segmentation network with multi-scale cascade encoding and dynamic spatial context enhancement. The network constructed a multi-scale cascaded encoder using Swin Transformer and large-kernel convolutions. It sequentially modeled and fused multi-scale features by capturing long-range dependencies and local details, and reparameterized large-kernel convolutions via a spatial frequency matrix to strengthen fine detail capture. Meanwhile, a spatial transformer module was designed to dynamically guide multi-head attention learning and optimize decoding performance. On the ImageCAS dataset, MSCNet achieved an average Dice coefficient of 81.24%, which was 3.57%, 3.78%, and 3.85% higher than 3D UX-Net, SwinUNETR, and SegMamba, respectively. MSCNet effectively improves the accuracy of coronary artery segmentation and provides support for clinical evaluation.
This review systematically examines the mechanisms and recent research progress of magnesium ions in promoting tendon-bone interface repair by regulating stem cell functions. Firstly, the analysis indicates that magnesium ions synergistically regulate stem cell proliferation, migration, and multidirectional differentiation through multiple signaling pathways, while simultaneously promoting angiogenesis and optimizing the immune microenvironment. Subsequently, a summary of existing research findings confirms that magnesium-based biodegradable biomaterials demonstrate favorable tissue repair-promoting effects in animal experiments. Finally, this article proposes that, given the spatiotemporal sequential characteristics of magnesium ion regulation on stem cells and the healing microenvironment, future research should focus on developing novel smart materials capable of precisely controlling magnesium ion release to match the healing process, thereby advancing its clinical translation and precision medicine applications in the field of tendon-bone repair.
For the endoscopic surgical instrument segmentation task, existing methods have failed to address the semantic gap caused by the mismatch between high-frequency spatial details and low-frequency semantic features in the U-shaped network (U-Net) architecture. This study proposes a U-Net algorithm based on frequency-domain adaptive feature decomposition and visual Mamba, namely frequency-domain decoupling Mamba U-Net (FDMUNet), for surgical instrument segmentation. This algorithm embeds a frequency-domain adaptive feature enhancement module into the skip connections, and decomposes features into high-frequency and low-frequency components through the Fourier transform and a learnable filter, followed by channel weighting and fusion, so as to enhance surgical instrument edge information and bridge the semantic gap between encoder and decoder features. On the Endoscopic Vision Challenge 2017 public dataset, FDMUNet achieved Intersection over Union, mean Intersection over Union, and mean class Intersection over Union scores of 70.79%, 74.25%, and 69.50%, respectively. In addition, the ablation experiment further verified the effectiveness of the proposed module. This method not only provides a new solution for instrument segmentation in complex scenes, but also provides a new research idea for the application of frequency-domain information in medical image segmentation.
Existing deep learning models for epileptic electroencephalogram (EEG) signal analysis frequently overlook intrinsic pathological characteristics during feature extraction and exhibit insufficient cross-dataset generalization. To address these limitations, this study proposes an innovative dual-attention epilepsy detection network (EDDANet). The model integrates a multi-band and multi-scale dual-attention module with a dynamic kernel sampling adaptive convolutional module to classify interictal and ictal EEG signals. Extensive experiments conducted on four heterogeneous public datasets demonstrate that EDDANet consistently outperforms state-of-the-art models across key evaluation metrics, including accuracy and recall. Notably, this work is the first to achieve robust generalization across varying lead configurations, sampling rates, and electrode layouts. In conclusion, this study provides a valuable methodological framework for the design and optimization of automated epilepsy detection systems in complex scenarios, providing reference for enhancing the generalizability and clinical utility of deep learning models in real-world environments.
Ultra-high molecular weight polyethylene (UHMWPE) fiber has emerged as a critical material advancing the development of minimally invasive medical devices, owing to its exceptional specific strength, outstanding wear resistance, and inherent bio-inertia. We systematically review the current application landscape of UHMWPE fibers in minimally invasive medicine, highlighting their broad use in orthopedic sutures and fixation systems, reinforcement layers for cardiovascular interventional devices, cables to drive surgical robots, and materials in frontier neural interfaces. These applications underscore the material's core advantages across diverse scenarios. However, its broader clinical translation faces multiple challenges, including surface bio-inertia, long-term dynamic durability, difficulty in processing, and a lack of standardization. To address these challenges, this article delves into comprehensive strategies encompassing surface engineering, composite material development, structural optimization, and intelligent control algorithms. Looking forward, UHMWPE fibers are poised to evolve towards intelligence and functional integration. Through deep convergence with flexible electronics and data-driven research, coupled with the establishment of robust standardization systems, UHMWPE fibers are expected to play an even more pivotal role in the next generation of advanced minimally invasive medical devices, ultimately propelling the field towards greater precision and personalization.
To address the problems of misidentification of similar gaits, excessive feature dimensionality, and computational complexity in gait recognition, this paper proposes a gait recognition method based on feature-level fusion. After validating the complementarity between motion posture signals and surface electromyography (sEMG) signals, parameters in the time, frequency, and time-frequency domains of the two types of signals were extracted. Based on the energy distribution of acceleration, angular velocity, and angle signals from motion posture signals, feature-level fusion was performed. A dual constraint strategy combining Gain-based discriminability filtering and energy-ratio stability filtering was adopted to reduce feature dimensions, yielding the most discriminative feature subset, upon which the XGBoost model was applied for gait recognition. Experimental results showed that the proposed method improved the average recognition accuracy by 8.6% over the baseline model that used only motion posture signals, reaching 95.8%. Specifically, the accuracies for forward, backward, and turning gaits reached 89.8%, 95.2%, and 97.3%, respectively, effectively reducing the misidentification rates for these three similar gaits. Furthermore, the feature-level fusion strategy effectively improved computational efficiency, and the energy distribution-based feature selection strategy reduced the impact of background noise on feature parameter perturbations, thereby enhancing model stability. This method provides strong technical support and engineering application value for gait feature parameter identification and real-time intelligent gait recognition control of exoskeletons.
Electroencephalogram (EEG)-based emotion recognition is an important research area in affective computing and mental health assessment. To address the insufficient modeling of long-term dependencies in EEG signals, this paper proposes an EEG emotion recognition method based on multi-branch convolutional neural networks (CNN) and Transformer (MCT). The proposed method employs a multi-scale CNN to extract local temporal features from EEG signals and constructs a parallel CNN-Transformer architecture to capture both short-term variations and long-term dependencies, thereby enabling temporal feature modeling at different time scales. Furthermore, a dual-branch convolutional structure is utilized to learn both global and local spatial channel features of EEG signals. A convolutional block attention module (CBAM) is then introduced to fuse the spatio-temporal features of EEG signals. Experimental results show that the proposed MCT model achieves a classification accuracy of 83.83% on the Shanghai Jiao Tong University emotion EEG dataset (SEED). On the music emotion EEG dataset (MEEG), it attains accuracies of 90.00% and 92.62% for the arousal and valence dimensions, respectively. On the database for emotion analysis using physiological signals (DEAP), it attains accuracies of 61.30% and 61.04% for the arousal and valence dimensions. The accuracy results on all three datasets outperform those of the best-performing baseline models. These findings indicate that MCT can effectively learn discriminative features associated with emotional states, providing a new perspective for EEG-based emotion recognition research.
In the field of medical image registration, corresponding landmarks can serve as an crucial reference for the quantitative evaluation of registration algorithm errors. However, the corresponding landmarks extracted by existing methods generally lack anatomical correspondence. While some approaches have begun to take bifurcation points of tubular structures as landmarks, the establishment of accurate one-to-one correspondences between these points is hindered by highly similar local features and missing correspondences, thereby impeding effective quantification of registration errors. To address these limitations, this paper proposes a reweighted graph matching-based algorithm for automatic extraction of corresponding landmarks in medical images. First, the bifurcation points of tubular structures with anatomical correspondence are extracted as landmarks. Secondly, to address the issues of highly similar local features and missing correspondences, a reweighted graph matching model is constructed base on a local similarity strategy and a modality independent neighborhood descriptor, which iteratively identifies correct matches and rectifies mismatches to progressively refine the correspondence between landmarks. Finally, the K-nearest neighbor graph of landmarks is dynamically updated, and outlier matches are filtered based on a local similarity strategy, thereby achieving accurate landmark matching under partial correspondence conditions. Experimental results on two public datasets demonstrate that the proposed method can accurately extract corresponding landmarks and effectively evaluate registration error, thereby providing technical support for enhancing the reliability of medical image registration algorithms in clinical applications.
Postoperative pulmonary infection (PPI) after esophageal cancer surgery occurs frequently and severely impairs patients' prognosis. Most existing prediction models cannot realize staged classification of risk factors, which limits targeted risk identification and intervention. Based on machine learning algorithms, this study integrates preoperative baseline characteristics and perioperative indicators to construct a preoperative-perioperative two-stage risk prediction model for postoperative pulmonary infection. Clinical data of 2 200 patients undergoing esophageal cancer surgery admitted to the Cancer Hospital, Chinese Academy of Medical Sciences between October 2022 and August 2024 were retrospectively enrolled. The least absolute shrinkage and selection operator was combined with multivariate logistic regression to screen independent predictive variables for the two stages, and five machine learning models were established accordingly. Six independent predictive variables were identified. The preoperative predictors included gender, American Society of Anesthesiologists (ASA) physical status classification, and colonization of multidrug-resistant bacteria. All models yielded area under the curve values ranging from 0.71 to 0.72 with a specificity higher than 98%, which can be used for preoperative risk stratification of high-risk individuals. On the basis of preoperative variables, the perioperative stage additionally incorporated operation duration, postoperative intensive care unit (ICU) admission, and peak C-reactive protein level within 0-3 days after surgery, leading to a remarkable improvement in predictive performance with all area under the curve values greater than 0.82. The gradient boosting machine (GBM) model achieved a favorable balance between a sensitivity of 69.07% and a specificity of 82.59%, providing support for risk stratification and clinical management decision-making. Further multicenter studies are required to validate the generalization ability of the model.
Aiming at the deficiencies of insufficient cross-domain generalization and poor rhythm sensitivity in atrial fibrillation (AF) detection from multi-lead electrocardiogram (ECG) signals, this paper proposes a novel AF detection algorithm based on multi-scale patch attention fusion. The method segmented ECG signals into overlapping temporal fragments of different scales to capture local waveform details and long-range rhythm patterns respectively; it fused cross-scale feature information through a multi-scale attention mechanism to strengthen the model's ability to perceive local and global rhythm features, and introduced the self-attention mechanism of Transformer to capture long-range rhythm correlations among fragments, thus realizing in-depth mining of ECG features. The algorithm was validated on the public CinC2021 dataset and the self-constructed clinical Clin-ECG dataset. Experimental results showed that the algorithm achieved an accuracy of 94.6% and 92.7% on the two datasets, with F1 scores reaching 0.945 and 0.923, respectively. Compared with baseline models such as ECG-ResNet and CNN-BiLSTM, the proposed algorithm exhibited higher accuracy and better cross-dataset generalization ability, providing an effective method for the automatic detection of AF from multi-lead ECG signals.
Post-stroke cognitive impairment (PSCI) is a common complication of stroke that severely affects patients' quality of life and social functional recovery. Quantitative electroencephalography (QEEG), as a non-invasive neuroelectrophysiological technique with high temporal resolution, can reveal PSCI-related brain dysfunction and plasticity changes from multiple levels and dimensions, including neural oscillations, brain network connectivity, and dynamic optimization of brain function. However, its feature integration and efficacy evaluation system still require further refinement. In recent years, acupuncture has attracted increasing attention in the intervention of PSCI owing to its holistic regulatory effects and favorable safety profile. This review systematically integrates the characteristics of brain rhythms, brain functional networks, and dynamic brain activity in patients with PSCI, analyzes the associations between different electroencephalography (EEG) indicators and cognitive domains, and summarizes the multidimensional changes in EEG features following acupuncture intervention in PSCI. The aim is to provide evidence for a deeper understanding of the neuroelectrophysiological mechanisms underlying PSCI, while also offering references for the objective evaluation and modernization of acupuncture interventions.
In colorectal cancer histopathology, the collaborative perception of microscopic and salient lesions is critical for effective diagnosis and improved patient prognosis. However, existing deep learning methods struggle to simultaneously capture microscopic glandular disorganization and salient tissue lesions. To address this limitation, a colorectal cancer diagnosis network based on dynamic gland-aware and tissue soft-clustering (DGTSNet) is proposed. The method employs dynamic gland-aware convolution to explicitly perceive gland boundaries and dynamically adjust sampling offsets, while incorporating a continuous-domain constraint to prevent out-of-bound sampling, thereby enhancing the perception of microscopic glandular disorders. Meanwhile, a tissue soft-clustering module is utilized to adaptively generate clustering prototypes and guide pixels toward relevant prototype centroids according to semantic similarity, suppressing irrelevant background interference and enhancing responses to significant tissue lesions. Finally, a cascaded sparse coupling module is introduced to collaboratively modulate cross-semantic feature representations along both the spatial and channel dimensions, while constructing differentiable masks to suppress low-contribution semantics, thereby achieving collaborative coupling of cross-semantic features. Experimental results demonstrated that the proposed method achieved superior performance on the Chaoyang, Kather-5K, and EBHI datasets, as well as a real-world clinical validation cohort, outperforming multiple baseline models. The study shows that the proposed method can effectively enhance the joint perception of microscopic glandular disorders and significant tissue lesions, providing an effective solution for colorectal cancer diagnosis.
