Myocardial infarction remains a leading cause of cardiovascular mortality, making its early detection and localization via electrocardiograms (ECGs) critical for timely intervention. Despite advances in deep learning-based ECG analysis, two key challenges persist: (1) Severe class imbalance biases models toward dominant categories, hindering the learning of discriminative patterns for underrepresented subtypes (e.g., posterior, extensive anterior infarctions). (2) The inability to adequately model complex spatiotemporal dependencies in ECGs limits the effectiveness of current approaches. To address these challenges, we propose SPG-Net, a novel structure-aware spatiotemporal graph network that integrates clinical priors with dynamic spatiotemporal dependencies of ECG for MI localization. The framework extracts high-resolution lead-wise features and constructs anatomically grounded graphs using class-specific lead contribution maps. A dedicated graph attention module integrates static anatomical priors, inter-lead spatial interactions, and temporal dependencies to generate ECG feature representations that preserve physiological consistency while accommodating individual variability. Furthermore, we design a decoupled classification mechanism to reduce interference from overlapping labels, significantly improving minority-class recognition. Comprehensive experiments on PTB and PTB-XL datasets demonstrate SPG-Net's superior overall performance and significantly enhanced minority-class recognition, validating its efficacy in class-imbalanced MI localization.

Electrocardiogram (ECG) classification plays a vital role in the early diagnosis of cardiovascular diseases. Multi-label ECG classification remains challenging despite advances in deep learning. This is mainly due to two factors: (1) the complex spectral characteristics of ECG signals, and (2) the limited ability of conventional models to capture nonlinear physiological patterns. In this work, we propose PSE- KAN et, a novel frequency-aware and spline-adaptive framework tailored for multi-label ECG classification. Central to our design is the Power Spectrum Enhancement (PSE) module, which mitigates the spectral bias commonly observed in deep neural networks, the tendency to overfit low-frequency components through a unified frequency-domain processing strategy. This strategy comprises three coordinated components: spectral reweighting balances across frequency bands, low-frequency anchoring preserves diag-nostic waveform structures, and Gaussian-based band emphasis selectively enhances underrepresented high-frequency features. Together, they enable task aware, frequency selective enhance-ment that enriches class relevant spectral representations. To further improve the modeling of complex label dependencies, we design KANMixLinear (KML), a classification head that integrates linear projection with learnable B-spline interpolation via a data-dependent gating mechanism. This design provides flexible and expressive decision boundaries suited for multi-label scenarios. Experimental results show that PSE-KANet performs better in multi-label ECG classification tasks, outperforming existing state-of-the-art methods across metrics, including AUC, Fl-score, and Recall. The code and data are publicly available at: https://github.com/yang-161224IBIBM.

Myocardial infarction (MI) is one of the most dangerous cardiovascular diseases, typically diagnosed using electrocardiogram (ECG). While many deep learning methods exist for MI detection, they often overlook the correlations and medical knowledge between leads of ECG, resulting in limited interpretability. To address the challenges, we propose a knowledge-guided multi branch dense graph neural network (KGD-GNN) for MI localization. The proposed method incorporates clinical diagnostic knowledge to represent ECG as a graph, capturing inter-lead relationships. A multi-branch graph convolutional module is then used to extract disease-specific features through dense graph convolution. Evaluations on the public PTB-XL dataset demonstrate the superior performance of KGD-GNN across multiple metrics.

This study aims to investigate the key factors affecting the use and interaction of wearable ECG devices from the user's perception. A conceptual model is proposed that combines an expectation-confirmation model with facilitating and inhibiting factors. Besides, health consciousness is set as a moderating variable. A quantitative study is conducted with users who have real-world experience with wearable ECG devices, the findings suggest that the perceived availability and compatibility of wearable ECG devices have a positive effect on confirmation and satisfaction. Technical anxiety and transition costs negatively affect satisfaction but have no impact on confirmation. Health consciousness can mitigate the negative effects of technological anxiety and transition costs while positively moderating the impact of perceived compatibility on satisfaction. The study recommendations focus on optimizing product reliability and real-time visual feedback, while ensuring multi-scenario compatibility and adopting differentiated design strategies. Additionally, incorporating health education and reward mechanisms is suggested.

Effective diabetes management requires regular and accurate blood glucose monitoring; however, traditional invasive methods often cause discomfort and inconvenience. Non-invasive techniques such as photoplethysmography (PPG) have been explored, though single-wavelength PPG systems are limited by the overlapping absorption characteristics between glucose and other biological components, such as water and fat. In this study, a novel multi-wavelength PPG system integrated with temperature and humidity sensors is introduced, coupled with a neural network framework featuring attention mechanisms to enhance glucose prediction. The system employs six optical sensors covering wavelengths from the visible to near-infrared (NIR) spectrum, enabling deeper tissue penetration and enhanced glucose specificity by targeting distinct absorption peaks-especially those above 1000 nm. The system was validated using a robust dataset of 26,063 measurements from 254 participants. The experimental results demonstrate significant improvements, with the model achieving 86.49% compliance with the ISO 15197: 2013 standards and 91.80% of measurements falling within Zone A of the Parkes error grid. The introduction of multiple wavelengths clearly improves performance over single-wavelength systems, and wavelengths above 1000 nm were shown to have a higher contribution in glucose prediction. In addition, the incorporation of temperature and humidity data also enhanced performance by accounting for environmental and physiological factors, and that demographic and meal-related factors significantly impact prediction accuracy, thereby underscoring the potential of this system as a reliable, non-invasive, and personalized glucose monitoring tool.

It is crucial to predict the in-hospital mortality for improving clinical decision-making and optimizing hospital resource allocation. Many recent studies have attempted to integrate multimodal data, such as digital time series and clinical notes from electronic health records (EHRs), to improve the performance of mortality prediction. Although current methods show good performance, it is often difficult to effectively capture fine-grained inter-modal correspondences. In addition, the different distributions and heterogeneous properties of the various modalities lead to modality gaps that severely affect the effectiveness of modal fusion. To overcome these limitations, we propose an Attention-based Multimodal Fusion with Adversarial network (AMFA) for in-hospital mortality prediction. In AMFA, feature extraction is first performed to obtain modality-specific features, and then an attention mechanism is employed to capture inter-modal interrelationships. The main difference with existing methods is that AMFA achieves this by calculating the relative importance of individual features focusing attention on the most relevant features and eliminating irrelevant features, and then reassigning attention between relevant features to obtain finer semantic relevance. Subsequently, we introduce a discriminator network that addresses the modality gap by adjusting the distribution of various modal representations through adversarial training. We evaluate AMFA on two publicly available datasets, and experimental results show that AMFA outperforms several state-of-the-art models in the task of in-hospital mortality prediction.

Myocardial infarction (MI) has the highest mortality of all cardiovascular diseases (CVDs). 12-lead electrocardiogram (ECG) is regarded as an effective noninvasive method for localization of MI. Considering the clinical reality that sometimes patients with heart disease often have multiple MI, and that real-world data faces an imbalance problem for each class of MI. In this study, we propose a data augmentation aided multichannel attention fusion network for multi-label MI localization (DAMAL), fusing both lead correlation and label correlation in a framework to localize MI. Specifically, DAMAL constructs separate channels for each of 12-lead ECG, and builds a matrix channel attention (MCA), to enhance the representativeness of the input features. A spatial & channel fusion module is proposed which includes a channel attention (CA) module to capture the attention levels of different leads, and a spatial attention (SA) module to capture the spatial relationships among the leads. Additionally, data augmentation is performed for multi-label datasets with minority classes to capture the optimal classification thresholds for each class of labels and improve the model performance. Furthermore, DAMAL exhibits good interpretability by the Grad-CAM-based visualization. Experimental results show that DAMAL performs better than other state-of-the-art methods in classifying multi-label MI across various metrics, including F1 score, AUROC, and Hamming Loss.

Hashing technologies have been widely applied for large-scale multimodal retrieval tasks owing to their excellent performance in search and storage tasks. Although some effective hashing methods have been proposed, it is still difficult to handle the intrinsic linkages that exist among different heterogeneous modalities. Moreover, optimizing the discrete constraint problem through a relaxation-based strategy results in a large quantization error and leads to a suboptimal solution. In this article, we present a novel asymmetric supervised fusion-oriented hashing method, named (ASFOH), which investigates three novel schemes to remedy the above issues. Specifically, we first explicitly formulate the problem as matrix decomposition into a common latent representation and a transformation matrix, combined with an adaptive weight scheme and nuclear norm minimization to ensure the information completeness of multimodal data. Then, we associate the common latent representation with the semantic label matrix, thereby increasing the discriminative capability of the model by constructing an asymmetric hash learning framework, thus, making the generated hash codes more compact. Finally, an efficient discrete optimization iterative algorithm based on nuclear norm minimization is proposed to decompose the nonconvex multivariate optimization problem into several subproblems with analytical solutions. Comprehensive experiments on the MIRFlirck, NUS-WIDE, and IARP-TC12 datasets testify that ASFOH outperforms the compared state-of-the-art approaches.

Predicting future health status based on historical patient visits is one of the essential tasks in healthcare. Many existing approaches attempt to enhance the representation learning capability of models by incorporating relevant medical knowledge, but their effectiveness is severely affected by the incompleteness and noise of the knowledge graphs. Moreover, due to the inability to capture temporal features at a fine-grained level, most existing methods also have limitations in learning the temporal development of patients’ health status. To address these issues, we propose a Knowledge-Enhanced Difference-Aware clinical time series representation learning model (KEDA) for diagnosis prediction. In this model, we first combine the medical ontology graph and co-occurrence graph, and use hierarchical graph convolution and contrastive learning methods to enhance the semantic representation of medical entities. After that, a task-specific difference-aware temporal module is designed to improve the accuracy of patient representation, which adds two novel gated units in the original GRU to fuse multi-type clinical information based on the relationship between different types of data and prediction tasks and capture fine-grained temporal evolution of patient health status. We validate our model on two publicly available datasets, and the experimental results demonstrate that KEDA outperforms the state-of-the-art methods.

Early detection of paroxysmal atrial fibrillation (AF) is valuable in determining treatment options and diagnosing complications, and identifying the number and duration of AF episodes in the dynamic electrocardiogram (ECG) may help advance the study of pathological AF mechanisms. Current computer-aided ECG signal diagnosis methods can be divided into two categories: (a) methods based on every single heartbeat split from ECG records, which cannot use information from adjacent heartbeats, and (b) methods based on a segment of ECG records that contain multiple heartbeats, which can only give a classification of the whole segment but cannot identify the categories of each heartbeat in the record at a fine granularity. Few methods can exploit the relationship between heartbeats as well as identify the onset and end of disease episodes. To fill this gap, we designed an improved U-Net model named Percept U-Net to detect AF episodes from the ECG records. In this model, we propose the percept attention to replace skip connections in U-Net to integrate low-level features with high-level features, thereby enhancing the ability to detect and discriminate between normal and abnormal heartbeats. Experimental results show that Percept U-Net achieves higher classification and localization accuracy with less computational overhead and paraments than other comparative methods.

The electrocardiogram sensor network (ECG-SN) is a medical monitoring system based on IoT technology, which can detect heart bioelectric signals in real time. But the ECG signal is vulnerable to human mobility and sensitive to the Age of Information (AoI). In this article, we first analyze the impact of human mobility on channel fading based on a real-world activity trace data set and design a two-state ECG work model based on the tradeoff between the ECG signal quality and energy consumption. Furthermore, an AoI model is proposed to evaluate the data timeliness. Then, an online transmission optimization algorithm is proposed to maximize the system utility by optimizing the sampling rate, transmission power, and data dropping rate. Furthermore, performance analysis presents the bounds for data buffer, battery capacity, and AoI. Numerical results show the dynamics of the system and the impact of the parameters on system performance, which verify that the proposed design has a larger utility and a smaller AoI in comparison with two benchmark schemes.

The body sensor network (BSN), consisting of wearable or implantable devices, is a monitoring system applied to a healthcare environment based on the Internet of Things (IoT) technology. In BSN, prolonging the service cycle of the network is a major challenge due to the limited battery capacity and energy supply for sensors. To this end, improving energy efficiency and harvesting energy are the keys for the network to maintain sustainability. In this paper, we propose a transmission scheduling and energy harvesting strategy to manage energy supply and consumption, and build several dynamic models to capture the stochastic processes in BSN. Besides, a system utility maximization problem is formulated. Since this problem is a multiobjective mixed-integer optimization problem (MMOP) which is difficult to solve directly, we provide a solution framework where MMOP is decomposed into several subproblems by the Lyapunov optimization method. Based on this framework, we propose an online energy sustainability optimization algorithm to solve these subproblems, such as the matching problem and convex optimization problem, and theoretically prove that it can achieve the near-optimal system utility. Additionally, the appropriate sizes of the data buffer and battery capacity are derived, which can give a guidance to determine the sizes of these components. Simulation results show the impact of the system parameter on the utility and data and energy queues, and verify that the proposed strategy and methods can maintain the sustainable operation of BSN effectively.