IEEE Journal of Biomedical and Health Informatics最新文献

{"title":"CAM-Interacted Vision GNN for Multi-Label Medical Images.","authors":"Jingchao Wang, Baoyao Yang, Siqi Liu, Xiaoqi Zheng, Wenbin Yao, Junxiang Chen","doi":"10.1109/JBHI.2025.3621455","DOIUrl":"10.1109/JBHI.2025.3621455","url":null,"abstract":"<p><p>Vision Graph Neural Network (ViG) is designed to recognize different objects through graph-level processing. However, ViG constructs graphs with appearance-level neighbors and neglects the category semantic. The oversight results in the unintentional connection of patches that belong to different objects, thus affecting the distinctiveness of categories in multi-label medical image learning. Since the pixel-level annotations for images are not easily available, category-aware graphs can not be directly built. To solve this problem, we consider localizing category-specific regions using Class Activation Maps (CAMs), an effective way to highlight regions belonging to each category without requiring manual annotations. Specifically, we propose a CAM-interacted Vision GNN (CiV-GNN), in which category-aware graphs are formed to perform intra-category graph processing. CIV-GNN includes a Class-activated Patch Division (CAPD) module, which introduces CAMs as guidance for category-aware graph building. Furthermore, we develop a Multi-graph Interactive Processing (MIP) module to model the relations between category-aware graphs, promoting inter-category interaction learning. Experimental results show that CiV-GNN performs well in surgical tool localization and multi-label medical image classification. Specifically, for m2cai16-localization, CiV-GNN exhibits a 1.43% and 7.02% improvement in mAP50 and mAP50-95, respectively, compared to YOLOv8.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":"4175-4185"},"PeriodicalIF":6.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145307838","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Interpretable AI Framework for Secure and Reliable Medical Image Analysis in IoMT Systems.","authors":"Ugochukwu Okwudili Matthew, Renata Lopes Rosa, Muhammad Saadi, Demostenes Zegarra Rodriguez","doi":"10.1109/JBHI.2025.3591737","DOIUrl":"10.1109/JBHI.2025.3591737","url":null,"abstract":"<p><p>The integration of artificial intelligence (AI) into medical image analysis has transformed healthcare, offering unprecedented precision in diagnosis, treatment planning, and disease monitoring. However, its adoption within the Internet of Medical Things (IoMT) raises significant challenges related to transparency, trustworthiness, and security. This paper introduces a novel Explainable AI (XAI) framework tailored for Medical Cyber-Physical Systems (MCPS), addressing these challenges by combining deep neural networks with symbolic knowledge reasoning to deliver clinically interpretable insights. The framework incorporates an Enhanced Dynamic Confidence-Weighted Attention (Enhanced DCWA) mechanism, which improves interpretability and robustness by dynamically refining attention maps through adaptive normalization and multi-level confidence weighting. Additionally, a Resilient Observability and Detection Engine (RODE) leverages sparse observability principles to detect and mitigate adversarial threats, ensuring reliable performance in dynamic IoMT environments. Evaluations conducted on benchmark datasets, including CheXpert, RSNA Pneumonia Detection Challenge, and NIH Chest X-ray Dataset, demonstrate significant advancements in classification accuracy, adversarial robustness, and explainability. The framework achieves a 15% increase in lesion classification accuracy, a 30% reduction in robustness loss, and a 20% improvement in the Explainability Index compared to state-of-the-art methods.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":"3689-3702"},"PeriodicalIF":6.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144698424","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radar HRV Monitoring With Physiological Prior Inspired Deep Neural Networks.","authors":"Haoyu Wang, Jinbo Chen, Dongheng Zhang, Zhi Lu, Yang Hu, Qibin Sun, Yan Chen","doi":"10.1109/JBHI.2025.3628628","DOIUrl":"10.1109/JBHI.2025.3628628","url":null,"abstract":"<p><p>Radar sensing has emerged as a promising solution for the contactless monitoring of Heart Rate Variability (HRV), a crucial indicator of the cardiovascular and autonomic nervous systems. However, due to signal noise and interference that easily obscure heartbeat details, along with variations in heartbeat across different physiological conditions, existing methods remain restricted to laboratory settings with healthy subjects and fail in real-world scenarios involving more complex physiological conditions. In this study, we propose a physiological prior-inspired deep learning framework for robust radar-based HRV monitoring. Specifically, we leverage the prior that internal heartbeats drive movements across the entire torso surface and design a hybrid deep neural network to model the spatio-temporal relationship between full-body radio reflections and heartbeats, effectively mitigating interference. Then, we incorporate the cardiac motion's self-similarity prior to establish a signal augmentation strategy, effectively remodeling the HRV distribution and enhancing performance across diverse physiological conditions. We build and validate our method on a large-scale dataset comprising 7,150 outpatients with complex physiological conditions in real-world scenarios. The experimental results demonstrate that our method achieves a mean IBI error of 19.21 ms, an RMSSD error of 16.23 ms, an SDSD error of 16.70 ms, and a pNN50 error of 7.28%. We further validate the performance by classifying five common cardiac conditions based on HRV results, demonstrating performance comparable to ECG-based methods. These results highlight the great potential of our approach for accurate, contactless HRV monitoring in real-world applications.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":"3784-3797"},"PeriodicalIF":6.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145451827","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Detecting Driver Sleepiness From Physiological Indicators Using a CNN-LSTM Self-Attention Model.","authors":"Yingying Jiao, Yifan Zhang, Wei Liu, Zhuqing Jiao","doi":"10.1109/JBHI.2025.3629974","DOIUrl":"10.1109/JBHI.2025.3629974","url":null,"abstract":"<p><p>Sleepiness at the wheel is an important factor contributing to road traffic accidents. Based on the characteristic changes in Electroencephalography (EEG) and Electrooculography (EOG) signals, a dozing state is refined into three sub-states: the onset, duration, and end state. Each state is characterized by different physiological indicators such as the EEG alpha waves, the rising edge, and falling edge waveforms in EOG signals. To enable real-time detection of these physiological indicators, we propose a framework integrating three Convolutional Neural Network-Long Short-Term Memory-Self-Attention (CLSA) models, which combine CNN-based local feature extraction with self-attention mechanism for global context capture. The framework is evaluated for performance on continuous test data from 12 subjects. Our results demonstrate that by detecting alpha waves and the rising edge waveform, the alpha wave epoch (AWE) at the onset of the dozing state can be identified with high accuracy and precision. Thus, the onset sub-state is calculated as the period from the start time of the rising edge waveform to the time when the AWE is valid. Subsequently, the duration sub-state corresponds to the sustained presence of alpha waves. Furthermore, the falling edge waveform is detected with high accuracy, enabling the classification of the end state into two distinct phenomena: alpha blocking phenomenon or alpha wave attenuation-disappearance phenomenon, representing the sleepiness level-relaxed wakefulness or sleep onset, respectively. Utilizing three-channel signal processing, this framework provides a promising approach for real-time sleepiness detection in real-world driving scenarios.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":"4082-4095"},"PeriodicalIF":6.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145471146","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"EmoSyn: Adaptive Emotion Framework for Sentiment Analysis and Internet of Medical Things.","authors":"Kamran Ahmad Awan, Abdullah M Alqahtani, Korhan Cengiz, Alya Alshammari, Ibrahim Alrashdi","doi":"10.1109/JBHI.2025.3560214","DOIUrl":"10.1109/JBHI.2025.3560214","url":null,"abstract":"<p><p>Sentiment analysis, or more specifically, the integration of IoMT into healthcare systems, requires frameworks that must adapt at runtime with high precision. Most existing methods have several limitations of either latency or accuracy issues, and therefore perform less effectively in dynamic scenarios. This study aims to address these challenges by proposing EmoSyn, a novel framework that incorporates Emotion Wave Modulation (EWM), Neuro-Cognitive Language Dynamics (NCLD), and the Sentient IoMT Interaction Protocol (SIP). EWM generates dynamic emotional waveforms using high-dimensional feature vectors and kernelized mappings, while NCLD employs synthetic neural mappings and adaptive linguistic modeling to capture semantic transitions. SIP facilitates real-time IoMT recalibration through bidirectional sentiment-driven feedback. Implemented using mathematical frameworks and a custom Emotion-Aware Predictive Synthesis Algorithm (EAPSA), EmoSyn ensures precise sentiment interpretation and efficient IoMT interactions. The framework was evaluated using MOSEI and MIMIC-III datasets in a Python-based simulation environment. The results showed EmoSyn achieving 91% precision for MOSEI and 86% for MIMIC-III, with average latencies of 29 ms and 27 ms.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":"3712-3719"},"PeriodicalIF":6.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144003139","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrating LLMs and Knowledge Graphs for Medical AI: Advances, Challenges, and Future Directions.","authors":"Lino Murali, G Gopakumar, Daleesha M Viswanathan, Raghu Raman, Prema Nedungadi","doi":"10.1109/JBHI.2025.3622058","DOIUrl":"10.1109/JBHI.2025.3622058","url":null,"abstract":"<p><p>This review synthesizes how integrating large language models (LLMs) with knowledge graphs (KGs) advances medical AI across methods, applications, and evaluation. While LLMs excel at natural language understanding and contextual reasoning, KGs provide structured factual knowledge, ensuring reliability in critical domains like healthcare AI. This review explores recent advances, emphasizing how LLM-KG synergy enhances knowledge extraction, clinical decision support, and explainability in medical applications. We analyze integration methodologies across three key frameworks: (a) KG-enhanced LLMs, where KGs refine reasoning during pre-training and inference; (b) LLM-augmented KGs, where LLMs improve KG construction, reasoning, and query resolution; and (c) Synergistic LLM-KG systems, which enable bidirectional knowledge exchange for more robust AI-driven decision-making. While these models offer substantial improvements in medical diagnostics, personalized treatment, and automated knowledge discovery, key challenges remain. Issues such as data heterogeneity, reasoning transparency, computational scalability, and ethical considerations surrounding patient data must be addressed to enable real-world clinical adoption. This review outlines future directions, including cross-domain knowledge integration, neurosymbolic AI frameworks, causal reasoning for explainable predictions, and multi-agent ensemble models for adaptive decision-making. We emphasize that scalability, real-time KG updates, and privacy-preserving mechanisms are vital for responsible, high-impact AI deployment in medicine.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":"3833-3848"},"PeriodicalIF":6.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145307902","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AI-Based Localized Latent Neural Representations of Acute and Chronic Pain in Rats.","authors":"Dunyan Yao, David A Lloyd, Yasemin Akay, Masahiro Ohsawa, Metin Akay","doi":"10.1109/JBHI.2025.3628226","DOIUrl":"10.1109/JBHI.2025.3628226","url":null,"abstract":"<p><p>Chronic pain is a widespread phenomenon affecting over 21% of the United States population. Despite the significant impact of pain on a patient's quality of life, the detection and identification of pain relies on subjective methods such as self-reporting. To address the challenges in identifying and treating chronic pain, a quantifiable biomarker for pain is needed. Here we present novel AI-driven method for the identification and isolation of localized pain signals in the brain during both acute and chronic pain. By using Matching Pursuit (MP) to decompose Local Field Potential (LFP) recordings from the Anterior Cingulate Cortex, the Nucleus Accumbens, and the Prelimbic Cortex, we can learn a latent representation with a conditional variational autoencoder (CVAE) and track changes in latent signal components in response to acute, sub-chronic, and chronic pain after injury. This method allows for both the identification of LFP signal components which are the primary drivers of observed aggregate changes in brain activity during pain, as well as for the tracking of said components over time. The model achieves an average per-feature RMSE of 0.130 on validation data and produces functionally separable latent representations of input MP atoms. The combination of MP for feature extraction and CVAE for latent space development allows for the extraction of both generalized and subject-specific pain motifs involved in chronic pain. These AI-driven biomarkers provide a basis for precision identification and quantitative monitoring of pain over time.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":"3934-3945"},"PeriodicalIF":6.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145437961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dual-Branch Attention-Based Frequency Domain Network for Cross-Subject SSVEP-BCIs.","authors":"Yi Yang, Ze Wang, Ziyu Jia, Boyu Wang, Shangen Zhang, Chi Man Wong, Xiaorong Gao, Tzyy-Ping Jung, Feng Wan","doi":"10.1109/JBHI.2025.3630249","DOIUrl":"10.1109/JBHI.2025.3630249","url":null,"abstract":"<p><p>Steady-state visual evoked potential-based brain-computer interfaces (SSVEP-BCIs) hold significant promise for enabling high-speed human-computer interaction in real-world scenarios. However, existing frequency-domain decoding methods treat frequency spectrum features (the real and imaginary spectrum features) as a single feature without considering their unique spatial and spectral characteristics, resulting in insufficient generalizable features and limited classification accuracy in cross-subject scenarios. To address this issue, we propose a Dual-Branch Attention-Based Frequency Domain Network (DB-AFDNet) to independently decode real and imaginary spectral components, aiming to acquire more discriminative and generalizable features for cross-subject applications. Specifically, we construct inter-branch attention similarity constraints to encourage the two branches to have similar attention properties, promoting to learn the consensus characteristics in the dual branches. Furthermore, we propose intra-branch orthogonality constraints to explore branch-specific discriminative features to learn generalizable features. Experimental studies on two public datasets, the Benchmark and Beta datasets, demonstrate that DB-AFDNet outperforms state-of-the-art methods in cross-subject classification, achieving a relative improvement of 1.36$%$ and 1.45$%$, respectively.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":"4096-4107"},"PeriodicalIF":6.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145471120","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unified Online Adaptation Framework for Correlation Analysis-based Spatial Filtering Methods in SSVEP-based BCIs.","authors":"Ze Wang, Lu Shen, Xinran Mi, Leqian Cheng, Yi Yang, Boyu Wang, Tzyy-Ping Jung, Feng Wan","doi":"10.1109/JBHI.2026.3689883","DOIUrl":"https://doi.org/10.1109/JBHI.2026.3689883","url":null,"abstract":"<p><p>Online adaptation is a promising technique for achieving calibration-free recognition in user-friendly brain-computer interfaces (BCIs) but remains underexplored for steady-state visual evoked potential (SSVEP) recognition. In our previous work on online multi-stimulus canonical correlation analysis (OMSCCA), we introduced a state-of-the-art scheme for the online adaptation of SSVEP spatial filters. Despite its effectiveness, this approach can not be directly extended to other advanced spatial filtering methods, thereby seriously limiting the broader development of calibration-free algorithms. To address this limitation, we propose a unified online adaptation frame work for correlation analysis (CA)-based spatial filtering methods, encompassing both spatial filter computation and utilization. Specifically, we extend the least-squares (LS) unified framework originally designed for full calibration with large amounts of training data to the online adaptation scenario without any pre-calibration, thereby enabling continuous updates of spatial filters. Moreover, to sufficiently utilize spatial filters, we introduce a cross-stimulus transfer method for online adaptation of the common impulse response and generation of user-specific templates for all stimuli using limited online unlabeled data. Finally, leveraging the proposed unified framework, we adapt three advanced spatial filtering methods from their calibration based counter parts to online adaptation paradigms and validate their performance through simulation studies. Our results demonstrate the framework's effectiveness in promoting the development ofzero-calibration SSVEP-based BCIs. Compared to the OMSCCA, the proposed online adaptation methods canimprove the recognition performance by more than 12%. This work provides a generalizable approach for transforming existing calibration-based methods into adaptive, user-friendly solutions for practical BCI applications.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":""},"PeriodicalIF":6.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147814429","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An Efficient Intrusion Detection System Using Advanced Machine Learning Techniques in SDN for Healthcare System.","authors":"Muhammad Waseem Asif, Aqsa Aqdus, Rashid Amin, Shehzad Ashraf Chaudhry, Faisal S Alsubaei, Sajid Iqbal","doi":"10.1109/JBHI.2025.3530563","DOIUrl":"10.1109/JBHI.2025.3530563","url":null,"abstract":"<p><p>The quick advancement of healthcare systems necessitates robust and efficient network security keys to defend sensitive patient records and guarantee uninterrupted service delivery. The current IDS has many challenges, such as a high false positive rate, poor accuracy of detection, slow response to threats, and inability to scale well. This paper proposes an efficient and real-time intrusion detection system (IDS) using advanced machine learning techniques within a software-defined networking (SDN) framework specifically tailored for healthcare systems. The proposed architecture implements a Machine Learning (ML) model that combines the SVM and KNN to better identify malicious activities. Full sets of detection and mitigation capabilities are implemented to address different types of traffic in the network with the least interference. Through the different evaluation measures, the efficiency of the proposed model is assured. Network performance is determined by success rate queries, packet losses in each domain path, and the CPU being used by the system. Responsiveness is measured through delay metrics grounded on end-to-end delay, hop-to-hop packet delay, latency rate, and propagation delay. Moreover, model accuracy fidelity is reviewed via precision assessment, alpha ($alpha$) affecting the accuracy of the model, and confusion matrix with different techniques with the proposed hybrid SVM-KNN model. Last of all, a comparison of the security of the models in question strengthens the argument in favor of the proposed model. More specifically, flow and network topology diagrams are included to show how integration may be accomplished in linkage or merger with existing healthcare networks. The results also present a 30% overall advancement in detection and mitigation by presenting the hybrid SVM-KNN model to overcome other traditional models. This proposed model shows significant improvements not less than 20-30% improvement in CPU use, 30-50% reduction in end-to-end delay, 30-40% less latency rate, 20-40% less propagation delay, and 20-30% better prediction accuracy, and outperforms Fuzzy, Logistic Regression and Decision Tree methods.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":"3651-3664"},"PeriodicalIF":6.8,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143541652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}