Autism Spectrum Disorder (ASD) seriously affects social, communication, and behavioral functions, and early accurate diagnosis is crucial to improve the prognosis of patients. Traditional diagnosis methods rely on professional doctors to make subjective diagnosis through scales, the feature extraction of existing machine learning methods is inefficient, and existing deep learning methods have limitations in capturing time-varying features and the joint expression of time–frequency features. To this end, this study proposes a time–frequency synergy network (TFSNet) to improve the accuracy of ASD EEG signal classification. The proposed Dynamic Residual Block (TDRB) was used to enhance time-domain feature extraction; Short-Time Fourier Transform (STFT), convolutional attention mechanism, and transformation technology were combined to capture frequency-domain information; and an adaptive cross-domain attention mechanism (ACDA) was designed to realize efficient fusion of time–frequency features. The experimental results show that the average accuracy of TFSNet on the University of Sheffield (containing 28 ASD patients and 28 healthy controls) and KAU dataset (containing 12 ASD patients and five healthy controls) reaches 98.68%and 97.14%, respectively, yielding significantly better results than the existing machine learning and deep learning methods. In addition, the analysis of model decisions through interpretability analysis techniques enhances its transparency and reliability.

Background: Brain–computer interfaces (BCIs) based on motor imagery (MI) offer promising solutions for motor rehabilitation and communication. However, electroencephalography (EEG) signals are often characterized by low signal-to-noise ratios, strong non-stationarity, and significant inter-subject variability, which pose significant challenges for accurate decoding. Existing methods often struggle to simultaneously model the spatial interactions between EEG channels, the local fine-grained features within signals, and global semantic patterns. Methods: To address this, we propose the graph attention-based hierarchical temporal network (GAH-TNet), which integrates spatial graph attention modeling with hierarchical temporal feature encoding. Specifically, we design the graph attention temporal encoding block (GATE). The graph attention mechanism is used to model spatial dependencies between EEG channels and encode short-term temporal dynamic features. Subsequently, a hierarchical attention-guided deep temporal feature encoding block (HADTE) is introduced, which extracts local fine-grained and global long-term dependency features through two-stage attention and temporal convolution. Finally, a fully connected classifier is used to obtain the classification results. The proposed model is evaluated on two publicly available MI-EEG datasets. Results: Our method outperforms multiple existing state-of-the-art methods in classification accuracy. On the BCI IV 2a dataset, the average classification accuracy reaches 86.84%, and on BCI IV 2b, it reaches 89.15%. Ablation experiments validate the complementary roles of GATE and HADTE in modeling. Additionally, the model exhibits good generalization ability across subjects. Conclusions: This framework effectively captures the spatio-temporal dynamic characteristics and topological structure of MI-EEG signals. This hierarchical and interpretable framework provides a new approach for improving decoding performance in EEG motor imagery tasks.

In the medical field, it is extremely important to use deep learning technology to automatically generate diagnostic reports for chest X-ray images. This technology provides an effective solution to the challenges faced by the medical field in processing large numbers of chest X-ray images. Especially during large-scale outbreaks of epidemics such as the new COVID-19, rapid and accurate screening and diagnosis of cases become important tasks. This study uses deep learning technology to automatically generate diagnostic reports for chest X-ray images, which significantly reduces the workload of doctors, reduces the risk of misdiagnosis and missed diagnosis, and provides technical support for improving public health emergency response capabilities. In this study, we propose an innovative network architecture to address the limitations of traditional image description networks in generating chest X-ray diagnostic reports, especially the large area deviation between abnormal and normal areas, and the lack of effective alignment of the two modalities of image and text. The convolutional block attention module (CBAM) is adopted to effectively alleviate the data bias problem through a sophisticated feature attention mechanism and improve the model’s ability to recognize abnormal image areas. The cross-attention mechanism is adopted to optimize the alignment process between images and texts, ensuring the accuracy and reliability of the diagnosis report.

Functional near-infrared spectroscopy (fNIRS) offers a safe and portable signal source for brain–computer interface (BCI) applications, particularly in motor imagery (MI) decoding. However, its low sampling rate and hemodynamic delay pose challenges for temporal modeling and dynamic brain network analysis. To address these limitations in temporal dynamics, static graph modeling, and feature fusion interpretability, we propose TopoTempNet, an innovative topology-enhanced temporal network for biomedical signal decoding. TopoTempNet integrates multi-level graph features with temporal modeling through three key innovations: (1) multi-level topological feature construction using local and global functional connectivity metrics (e.g., connection strength, density, global efficiency); (2) a graph-modulated attention mechanism combining Transformer and Bi-LSTM to dynamically model key connections; and (3) a multimodal fusion strategy uniting raw signals, graph structures, and temporal representations into a high-dimensional discriminative space. Evaluated on three public fNIRS datasets (MA, WG, UFFT), TopoTempNet achieves superior accuracy (up to 90.04% ± 3.53%) and Kappa scores compared to state-of-the-art models. The ROC curves and t-SNE visualizations confirm its excellent feature discrimination and structural clarity. Furthermore, the statistical analysis of graph features reveals the model’s ability to capture task-specific functional connectivity patterns, enhancing the interpretability of decoding outcomes. TopoTempNet provides a novel pathway for building interpretable and high-performance BCI systems based on fNIRS.

Brain-computer interface (BCI) technology based on motor imagery (MI) represents an advanced human-computer interaction approach that enables us to communicate and exchange directly with the external environment through neural activities. However, the non-stationarity and individual variability of EEG signals present significant challenges to improving decoding accuracy. Existing studies often struggle with feature extraction, dynamic feature selection, and temporal modeling, failing to capture critical EEG patterns effectively. In this study, we propose a multi-scale adaptive fusion network (MSAFNet) to improve the accuracy of MI-EEG classification. The MSAFNet integrates two primary modules: scale-aware channel attention (SCA) and adaptive dynamic feature fusion (ADF). The SCA module strengthens the model’s ability to perceive features of different frequencies and spatio-temporal information through multi-scale feature interactions to extract richer primary feature representations. The ADF module captures short-term dynamic variations in EEG signals while performing key feature selection and cross-layer feature fusion. Additionally, long-term temporal dependence is modeled by a temporal convolution network (TCN) to enhance the extraction capability of temporal features of EEG signals. Ablation experiments and t-SNE visualization analysis verified the effectiveness of the modules. Gradient-weighted Class Activation Mapping generated EEG topography maps visualized the brain regions of interest of the model, and the experimental results show that the decision-making process of the model in the MI-EEG decoding task is in accordance with neurophysiological principles. The within-subject average classification accuracies on the BCI Competition IV-2a and IV-2b public datasets are 88.08% and 90.47%, respectively, and the subject-independent average classification accuracies reach 68.79% and 80.86%, achieving state-of-the-art decoding accuracies and providing new possibilities for practical applications of BCI.