首页 > 最新文献

IEEE Journal of Biomedical and Health Informatics最新文献

英文 中文
MCBTNet: Multi-Feature Fusion CNN and Bi- Level Routing Attention Transformer-based Medical Image Segmentation Network.
IF 6.7 2区 医学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS Pub Date : 2025-02-24 DOI: 10.1109/JBHI.2025.3545398
Boheng Zhang, Zelin Zheng, Yanqi Zhao, Yi Shen, Mingjian Sun

Accurate medical image segmentation is crucial for precise diagnosis and treatment in clinical pathology analysis and surgical navigation. While Convolutional Neural Network (CNN)-based approaches excel in capturing and analyzing local features, they often lose key global context. Transformers, utilizing self-attention mechanisms, address this issue but often overlook localized and multi-scale features while also requiring significant computational resources. To integrate the advantages of CNNs and Transformers to achieve efficient and precise medical image segmentation, we propose a segmentation framework based on multi-feature fusion CNN and Bi-level Routing Attention Transformer (MCBTNet). MCBTNet integrates CNNs and Transformers within a U-shaped encoderdecoder architecture. This configuration not only extracts multi-scale features via the U-shaped structure but also efficiently captures global contextual information through the dynamic sparsity of the Bi-Level Routing Attention Transformer. Our novel Frequency-Channel-Spatial multidimensional attention mechanism is implemented on skip connections, enhancing segmentation accuracy and speed by maximizing multi-scale feature utilization. Finally, MCBTNet obtains the segmentation result by fusing the predictions of different scales. Experimental results on five public datasets demonstrate that MCBTNet outperforms state-of-the-art methods in Dice and HD metrics, with lower computational and memory requirements. The code will be available on https://github.com/670768312/MCBTNet.

准确的医学图像分割对于临床病理分析和手术导航中的精确诊断和治疗至关重要。虽然基于卷积神经网络(CNN)的方法在捕捉和分析局部特征方面表现出色,但它们往往会丢失关键的全局背景。变形器利用自我注意机制解决了这一问题,但往往会忽略局部和多尺度特征,同时还需要大量的计算资源。为了整合 CNN 和变换器的优势,实现高效、精确的医学图像分割,我们提出了基于多特征融合 CNN 和双级路由注意变换器(MCBTNet)的分割框架。MCBTNet 在 U 型编码器-解码器架构中集成了 CNN 和变换器。这种配置不仅能通过 U 型结构提取多尺度特征,还能通过双级路由注意力变换器的动态稀疏性有效捕捉全局上下文信息。我们新颖的频率-信道-空间多维注意力机制是在跳过连接上实现的,通过最大限度地利用多尺度特征来提高分割精度和速度。最后,MCBTNet 通过融合不同尺度的预测结果来获得分割结果。在五个公共数据集上的实验结果表明,MCBTNet 在 Dice 和 HD 指标上优于最先进的方法,而且对计算和内存的要求更低。代码将发布在 https://github.com/670768312/MCBTNet 上。
{"title":"MCBTNet: Multi-Feature Fusion CNN and Bi- Level Routing Attention Transformer-based Medical Image Segmentation Network.","authors":"Boheng Zhang, Zelin Zheng, Yanqi Zhao, Yi Shen, Mingjian Sun","doi":"10.1109/JBHI.2025.3545398","DOIUrl":"https://doi.org/10.1109/JBHI.2025.3545398","url":null,"abstract":"<p><p>Accurate medical image segmentation is crucial for precise diagnosis and treatment in clinical pathology analysis and surgical navigation. While Convolutional Neural Network (CNN)-based approaches excel in capturing and analyzing local features, they often lose key global context. Transformers, utilizing self-attention mechanisms, address this issue but often overlook localized and multi-scale features while also requiring significant computational resources. To integrate the advantages of CNNs and Transformers to achieve efficient and precise medical image segmentation, we propose a segmentation framework based on multi-feature fusion CNN and Bi-level Routing Attention Transformer (MCBTNet). MCBTNet integrates CNNs and Transformers within a U-shaped encoderdecoder architecture. This configuration not only extracts multi-scale features via the U-shaped structure but also efficiently captures global contextual information through the dynamic sparsity of the Bi-Level Routing Attention Transformer. Our novel Frequency-Channel-Spatial multidimensional attention mechanism is implemented on skip connections, enhancing segmentation accuracy and speed by maximizing multi-scale feature utilization. Finally, MCBTNet obtains the segmentation result by fusing the predictions of different scales. Experimental results on five public datasets demonstrate that MCBTNet outperforms state-of-the-art methods in Dice and HD metrics, with lower computational and memory requirements. The code will be available on https://github.com/670768312/MCBTNet.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":""},"PeriodicalIF":6.7,"publicationDate":"2025-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143541587","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}
引用次数: 0
Warfarin Dose Management Using Offline Deep Reinforcement Learning.
IF 6.7 2区 医学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS Pub Date : 2025-02-24 DOI: 10.1109/JBHI.2025.3545384
Hannah Ji, Matthew Gill, Evan W Draper, David A Liedl, David O Hodge, Damon E Houghton, Ana I Casanegra

Warfarin is a commonly prescribed anticoagulant with a narrow therapeutic window, requiring frequent and specialized monitoring. This work aims to develop standardized optimal warfarin dose decision support using a machine learning model based on time series anticoagulation data and patient demographic characteristics. We propose an offline reinforcement learning model (RL) using a Batch-Constrained Q-Learning algorithm (BCQ) in the discrete action setting to predict the cumulative warfarin dose for the days until the next INR (International Normalized Ratio) test. Prior approaches utilized time-series supervised learning methods such as regression or Long Short Term Memory (LSTM) neural networks. The key advantage of reinforcement learning is its capacity to learn optimal dosing strategies from suboptimal clinical states in the data. To evaluate the model we compared the predicted warfarin doses with the physician-prescribed doses. Our BCQ model with a prediction accuracy of 98.6% significantly outperformed our baseline Long Short Term Memory (LSTM) model with a prediction accuracy of 71.09%. Further qualitative evaluation for explainability indicated that the model correctly adjusted the warfarin dose at time steps when patients had out-of-range INRs.

{"title":"Warfarin Dose Management Using Offline Deep Reinforcement Learning.","authors":"Hannah Ji, Matthew Gill, Evan W Draper, David A Liedl, David O Hodge, Damon E Houghton, Ana I Casanegra","doi":"10.1109/JBHI.2025.3545384","DOIUrl":"https://doi.org/10.1109/JBHI.2025.3545384","url":null,"abstract":"<p><p>Warfarin is a commonly prescribed anticoagulant with a narrow therapeutic window, requiring frequent and specialized monitoring. This work aims to develop standardized optimal warfarin dose decision support using a machine learning model based on time series anticoagulation data and patient demographic characteristics. We propose an offline reinforcement learning model (RL) using a Batch-Constrained Q-Learning algorithm (BCQ) in the discrete action setting to predict the cumulative warfarin dose for the days until the next INR (International Normalized Ratio) test. Prior approaches utilized time-series supervised learning methods such as regression or Long Short Term Memory (LSTM) neural networks. The key advantage of reinforcement learning is its capacity to learn optimal dosing strategies from suboptimal clinical states in the data. To evaluate the model we compared the predicted warfarin doses with the physician-prescribed doses. Our BCQ model with a prediction accuracy of 98.6% significantly outperformed our baseline Long Short Term Memory (LSTM) model with a prediction accuracy of 71.09%. Further qualitative evaluation for explainability indicated that the model correctly adjusted the warfarin dose at time steps when patients had out-of-range INRs.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":""},"PeriodicalIF":6.7,"publicationDate":"2025-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143541741","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}
引用次数: 0
Temporal and Spectral Impedance Myography: Visualization of Dynamic Electrical Properties in Forearm Tissue during Wrist Movements.
IF 6.7 2区 医学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS Pub Date : 2025-02-21 DOI: 10.1109/JBHI.2025.3541655
Tomiharu Yamaguchi, Akihiko Tsukahara, Yuho Tanaka, Akinori Ueno

The recently proposed short-time impedance spectroscopy (short-time IS) measures impedance spectra with high temporal resolution using a nonsinusoidal mode-switching oscillator. With a view to application for muscular disease diagnosis and human-machine interaction, this study presents the temporal and spectral impedance myography using the short-time IS system. We applied the proposed system to the measurement of human forearm impedance and analyzed impedance changes during wrist motion. The forearm's impedance at 40 frequency points was measured by our developed system. The forearm's rapid impedance response during wrist motion was captured with a high temporal resolution of about 0.5 ms, displaying the impedance magnitude and phase angle change as a color map. Furthermore, we experimentally and theoretically demonstrated that difference and differential Cole-Cole plots reduced skin impedance effects and produced a characteristic curve corresponding to wrist motion, even with two-terminal measurement. We anticipate that the IS system and visualization method will be beneficial for myoelectric control in prosthetic hands and noninvasive muscle disease diagnosis.

{"title":"Temporal and Spectral Impedance Myography: Visualization of Dynamic Electrical Properties in Forearm Tissue during Wrist Movements.","authors":"Tomiharu Yamaguchi, Akihiko Tsukahara, Yuho Tanaka, Akinori Ueno","doi":"10.1109/JBHI.2025.3541655","DOIUrl":"https://doi.org/10.1109/JBHI.2025.3541655","url":null,"abstract":"<p><p>The recently proposed short-time impedance spectroscopy (short-time IS) measures impedance spectra with high temporal resolution using a nonsinusoidal mode-switching oscillator. With a view to application for muscular disease diagnosis and human-machine interaction, this study presents the temporal and spectral impedance myography using the short-time IS system. We applied the proposed system to the measurement of human forearm impedance and analyzed impedance changes during wrist motion. The forearm's impedance at 40 frequency points was measured by our developed system. The forearm's rapid impedance response during wrist motion was captured with a high temporal resolution of about 0.5 ms, displaying the impedance magnitude and phase angle change as a color map. Furthermore, we experimentally and theoretically demonstrated that difference and differential Cole-Cole plots reduced skin impedance effects and produced a characteristic curve corresponding to wrist motion, even with two-terminal measurement. We anticipate that the IS system and visualization method will be beneficial for myoelectric control in prosthetic hands and noninvasive muscle disease diagnosis.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":""},"PeriodicalIF":6.7,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143556728","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}
引用次数: 0
DD-HGNN: Drug-Disease Association Prediction Via General Hypergraph Neural Network With Hierarchical Contrastive Learning and Cross Attention Learning.
IF 6.7 2区 医学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS Pub Date : 2025-02-21 DOI: 10.1109/JBHI.2025.3542784
Zixiao Jin, Xiao Zheng, Hua Zhou, Chengfu Ji, Sen Xiang, Chang Tang

The research on identifying drug-disease associations (DDAs) is widely used in scenarios such as drug development, clinical decision-making, and drug repurposing, holding significant biological and medical significance. Existing methods for drug-disease association prediction have achieved decent performance, they primarily rely on simplistic drug-disease association graphs or similarity graphs. These methods often struggle to capture the high-order correlations of complex multimodal data, limiting their ability to handle the complexity of data associations effectively. In addition, real drug-disease associations are highly sparse, posing a significant challenge to prediction accuracy. To tackle these issues, we propose a general hypergraph neural network framework for drug-disease association prediction based on hierarchical contrastive learning and cross-attention learning. It leverages hypergraph neural networks to learn representations of drugs and diseases carrying high-order correlations and strengthens representation quality using interactive attention learning and hierarchical contrastive learning. Meanwhile, the -weighted loss function is utilized to adapt to the high sparsity property of real drug-disease associations during model training and improve prediction performance. Extensive experiments demonstrate that DD-HGNN surpasses other state-of-the-art methods in predicting drug-disease associations and further validation through case studies on Leukemia and Colorectal Neoplasms underscores its reliability.

{"title":"DD-HGNN: Drug-Disease Association Prediction Via General Hypergraph Neural Network With Hierarchical Contrastive Learning and Cross Attention Learning.","authors":"Zixiao Jin, Xiao Zheng, Hua Zhou, Chengfu Ji, Sen Xiang, Chang Tang","doi":"10.1109/JBHI.2025.3542784","DOIUrl":"https://doi.org/10.1109/JBHI.2025.3542784","url":null,"abstract":"<p><p>The research on identifying drug-disease associations (DDAs) is widely used in scenarios such as drug development, clinical decision-making, and drug repurposing, holding significant biological and medical significance. Existing methods for drug-disease association prediction have achieved decent performance, they primarily rely on simplistic drug-disease association graphs or similarity graphs. These methods often struggle to capture the high-order correlations of complex multimodal data, limiting their ability to handle the complexity of data associations effectively. In addition, real drug-disease associations are highly sparse, posing a significant challenge to prediction accuracy. To tackle these issues, we propose a general hypergraph neural network framework for drug-disease association prediction based on hierarchical contrastive learning and cross-attention learning. It leverages hypergraph neural networks to learn representations of drugs and diseases carrying high-order correlations and strengthens representation quality using interactive attention learning and hierarchical contrastive learning. Meanwhile, the -weighted loss function is utilized to adapt to the high sparsity property of real drug-disease associations during model training and improve prediction performance. Extensive experiments demonstrate that DD-HGNN surpasses other state-of-the-art methods in predicting drug-disease associations and further validation through case studies on Leukemia and Colorectal Neoplasms underscores its reliability.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":""},"PeriodicalIF":6.7,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143556717","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}
引用次数: 0
GPT-Based Automated Induction: Vulnerability Detection in Medical Software.
IF 6.7 2区 医学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS Pub Date : 2025-02-21 DOI: 10.1109/JBHI.2025.3544560
Liangjun Deng, Hang Lei, Fazlullah Khan, Gautam Srivastava, Jingxue Chen, Mainul Haque

Integrating Natural Language Processing (NLP) with Generative Pre-trained Transformer (GPT) models plays a pivotal role in enhancing the accuracy and efficiency of healthcare software, which is essential for patient safety and providing high-quality care. The precision of healthcare software is fundamental to protecting the well-being of the patient. In addition, it can ensure the delivery of superior care, maintain the integrity of healthcare systems, and promote trust and cost-effectiveness. It is necessary to emphasize the importance of software reliability in its development and deployment. Symbolic execution serves as a vital technology in automated vulnerability detection. However, symbolic execution often faces problems such as path explosion, which seriously affects efficiency. Although there have been several studies to reduce the number of computational paths in symbolic execution, this problem remains a major obstacle. Therefore, more efficient solutions are urgently needed to ensure the software security. This paper proposes a large-scale language model(LLM) induction method mitigating path explosion applied to symbolic execution engines. In contrast to traditional symbolic execution engines, which often result in timeout or out-of-memory detection, our approach achieves the task of detecting vulnerabilities in seconds. Furthermore, our proposal improves the scalability of symbolic execution, allowing more extensive and complex programs to be analyzed without significant increases in computational resources or time. This scalability is crucial to tackling modern software systems and improving the efficiency and effectiveness of automated defect verification in healthcare software.

{"title":"GPT-Based Automated Induction: Vulnerability Detection in Medical Software.","authors":"Liangjun Deng, Hang Lei, Fazlullah Khan, Gautam Srivastava, Jingxue Chen, Mainul Haque","doi":"10.1109/JBHI.2025.3544560","DOIUrl":"https://doi.org/10.1109/JBHI.2025.3544560","url":null,"abstract":"<p><p>Integrating Natural Language Processing (NLP) with Generative Pre-trained Transformer (GPT) models plays a pivotal role in enhancing the accuracy and efficiency of healthcare software, which is essential for patient safety and providing high-quality care. The precision of healthcare software is fundamental to protecting the well-being of the patient. In addition, it can ensure the delivery of superior care, maintain the integrity of healthcare systems, and promote trust and cost-effectiveness. It is necessary to emphasize the importance of software reliability in its development and deployment. Symbolic execution serves as a vital technology in automated vulnerability detection. However, symbolic execution often faces problems such as path explosion, which seriously affects efficiency. Although there have been several studies to reduce the number of computational paths in symbolic execution, this problem remains a major obstacle. Therefore, more efficient solutions are urgently needed to ensure the software security. This paper proposes a large-scale language model(LLM) induction method mitigating path explosion applied to symbolic execution engines. In contrast to traditional symbolic execution engines, which often result in timeout or out-of-memory detection, our approach achieves the task of detecting vulnerabilities in seconds. Furthermore, our proposal improves the scalability of symbolic execution, allowing more extensive and complex programs to be analyzed without significant increases in computational resources or time. This scalability is crucial to tackling modern software systems and improving the efficiency and effectiveness of automated defect verification in healthcare software.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":""},"PeriodicalIF":6.7,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143541550","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}
引用次数: 0
Accurate Core Body Temperature Prediction for Infrared Thermography Considering Ambient Temperature and Personal Features.
IF 6.7 2区 医学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS Pub Date : 2025-02-20 DOI: 10.1109/JBHI.2025.3543978
Chengcheng Shan, Jiawen Hu, Tianshu Zhou, Jingsong Li

Accurate and timely core body temperature measurement is essential for identifying and preventing heat-related illnesses. Infrared thermography (IRT) provides a non-invasive, full-scale and efficient temperature path for body temperature screening. However, the complexity of environmental factors and personal features continuously affect the measured skin temperature, resulting in low accuracy and reliability of existing body temperature monitoring by IRT. To address this issue, this study proposed an innovative core temperature prediction model (CTPM) for IRT based on heat transfer mechanism between the human body and the ambient environment. Based on human body thermoregulation, the optimal facial thermal feature that can reflect the impact of ambient temperature on skin temperature is proposed. Combining it with personal features and distributed facial skin temperature features, a CTPM is established based on Random Forest algorithm. The proposed CTPM are evaluated using a publicly available PhysioNet facial and oral temperature dataset. The results demonstrate that the proposed optimal CTPM achieves the best accuracy and consistency in predicting core body temperature. The root-mean-square error of the optimal CTPM is 0.259°C, and the mean lower and upper 95% limits of agreement are -0.505 °C and 0.507°C, respectively. Variable importance analysis indicates that the proposed optimal facial thermal feature makes a dominant contribution to the prediction performance of the optimal CTPM. Our method enables accurate and stable core body temperature prediction in complex ambient environments over a wide range of temperatures, and has the potential to replace traditional contact measurements to meet clinical needs.

{"title":"Accurate Core Body Temperature Prediction for Infrared Thermography Considering Ambient Temperature and Personal Features.","authors":"Chengcheng Shan, Jiawen Hu, Tianshu Zhou, Jingsong Li","doi":"10.1109/JBHI.2025.3543978","DOIUrl":"https://doi.org/10.1109/JBHI.2025.3543978","url":null,"abstract":"<p><p>Accurate and timely core body temperature measurement is essential for identifying and preventing heat-related illnesses. Infrared thermography (IRT) provides a non-invasive, full-scale and efficient temperature path for body temperature screening. However, the complexity of environmental factors and personal features continuously affect the measured skin temperature, resulting in low accuracy and reliability of existing body temperature monitoring by IRT. To address this issue, this study proposed an innovative core temperature prediction model (CTPM) for IRT based on heat transfer mechanism between the human body and the ambient environment. Based on human body thermoregulation, the optimal facial thermal feature that can reflect the impact of ambient temperature on skin temperature is proposed. Combining it with personal features and distributed facial skin temperature features, a CTPM is established based on Random Forest algorithm. The proposed CTPM are evaluated using a publicly available PhysioNet facial and oral temperature dataset. The results demonstrate that the proposed optimal CTPM achieves the best accuracy and consistency in predicting core body temperature. The root-mean-square error of the optimal CTPM is 0.259°C, and the mean lower and upper 95% limits of agreement are -0.505 °C and 0.507°C, respectively. Variable importance analysis indicates that the proposed optimal facial thermal feature makes a dominant contribution to the prediction performance of the optimal CTPM. Our method enables accurate and stable core body temperature prediction in complex ambient environments over a wide range of temperatures, and has the potential to replace traditional contact measurements to meet clinical needs.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":""},"PeriodicalIF":6.7,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143556709","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}
引用次数: 0
BAHBench: A Unified Benchmark for Evaluating Bio-Acoustic Health with Acoustic Foundation Models.
IF 6.7 2区 医学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS Pub Date : 2025-02-20 DOI: 10.1109/JBHI.2025.3543968
Weixiang Xu, Zhongren Dong, Jing Peng, Runming Wang, Zixing Zhang

Acoustic foundation models, through self-supervised learning on large amounts of unlabeled speech data, can acquire rich acoustic representations. In recent years, these models have demonstrated substantial potential in audio-based health-related tasks, remarkably enhancing the efficiency and quality of healthcare services and contributing to the advancement of smart healthcare. However, there is currently a lack of systematic research and exploration on the performance of acoustic foundation models in health-related tasks. Furthermore, inconsistencies in evaluation methods and experimental setups hinder fair comparisons between different methods, severely impeding progress in this field. To address these challenges, we establish a unified Benchmark for evaluating Bio-Acoustic health via acoustic foundation models, namely BAHBench. BAHBench encompasses 6 distinct health-related tasks and evaluates 12 acoustic foundation models within a unified evaluation framework and parameter settings, enabling fair comparisons across different models. Our objective is to explore the effectiveness of current acoustic foundation models in health-related tasks. Thus, we discuss the impact of model size and data diversity on performance, and investigate feature selection and efficient fine-tuning strategy. Experimental results show that different health-related tasks benefit from features from different layers of the foundation model, while LoRA fine-tuning further enhances the model's performance on downstream tasks. Our goal is to provide clear and comprehensive guidance for future researchers. The code related to this study will be available to the research community to promote transparency and reproducibility.

{"title":"BAHBench: A Unified Benchmark for Evaluating Bio-Acoustic Health with Acoustic Foundation Models.","authors":"Weixiang Xu, Zhongren Dong, Jing Peng, Runming Wang, Zixing Zhang","doi":"10.1109/JBHI.2025.3543968","DOIUrl":"https://doi.org/10.1109/JBHI.2025.3543968","url":null,"abstract":"<p><p>Acoustic foundation models, through self-supervised learning on large amounts of unlabeled speech data, can acquire rich acoustic representations. In recent years, these models have demonstrated substantial potential in audio-based health-related tasks, remarkably enhancing the efficiency and quality of healthcare services and contributing to the advancement of smart healthcare. However, there is currently a lack of systematic research and exploration on the performance of acoustic foundation models in health-related tasks. Furthermore, inconsistencies in evaluation methods and experimental setups hinder fair comparisons between different methods, severely impeding progress in this field. To address these challenges, we establish a unified Benchmark for evaluating Bio-Acoustic health via acoustic foundation models, namely BAHBench. BAHBench encompasses 6 distinct health-related tasks and evaluates 12 acoustic foundation models within a unified evaluation framework and parameter settings, enabling fair comparisons across different models. Our objective is to explore the effectiveness of current acoustic foundation models in health-related tasks. Thus, we discuss the impact of model size and data diversity on performance, and investigate feature selection and efficient fine-tuning strategy. Experimental results show that different health-related tasks benefit from features from different layers of the foundation model, while LoRA fine-tuning further enhances the model's performance on downstream tasks. Our goal is to provide clear and comprehensive guidance for future researchers. The code related to this study will be available to the research community to promote transparency and reproducibility.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":""},"PeriodicalIF":6.7,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143556712","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}
引用次数: 0
Enhancing Image Retrieval Performance With Generative Models in Siamese Networks.
IF 6.7 2区 医学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS Pub Date : 2025-02-20 DOI: 10.1109/JBHI.2025.3543907
Alejandro Golfe, Adrian Colomer, Jose Padres, Valery Naranjo

Prostate cancer is a critical healthcare challenge globally and is one of the most prevalent types of cancer in men. Early and accurate diagnosis is essential for effective treatment and improved patient outcomes. In the existing literature, computer-aided diagnosis (CAD) solutions have been developed to assist pathologists in various tasks, including classification, diagnosis, and prostate cancer grading. Content-based image retrieval (CBIR) techniques provide valuable approaches to enhance these computer-aided solutions. This study evaluates how generative deep learning models can improve the quality of retrievals within a CBIR system. Specifically, we propose applying a Siamese Network approach, which enables us to learn how to encode image patches into latent representations for retrieval purposes. We used the ProGleason-GAN framework trained on the SiCAPv2 dataset to create similar pairs of input patches. Our observations indicate that introducing synthetic patches leads to notable improvements in the evaluated metrics, underscoring the utility of generative models within CBIR tasks. Furthermore, this work is the first in the literature where latent representations optimized for CBIR are used to train an attention mechanism for performing Gleason Scoring of a WSI.

{"title":"Enhancing Image Retrieval Performance With Generative Models in Siamese Networks.","authors":"Alejandro Golfe, Adrian Colomer, Jose Padres, Valery Naranjo","doi":"10.1109/JBHI.2025.3543907","DOIUrl":"https://doi.org/10.1109/JBHI.2025.3543907","url":null,"abstract":"<p><p>Prostate cancer is a critical healthcare challenge globally and is one of the most prevalent types of cancer in men. Early and accurate diagnosis is essential for effective treatment and improved patient outcomes. In the existing literature, computer-aided diagnosis (CAD) solutions have been developed to assist pathologists in various tasks, including classification, diagnosis, and prostate cancer grading. Content-based image retrieval (CBIR) techniques provide valuable approaches to enhance these computer-aided solutions. This study evaluates how generative deep learning models can improve the quality of retrievals within a CBIR system. Specifically, we propose applying a Siamese Network approach, which enables us to learn how to encode image patches into latent representations for retrieval purposes. We used the ProGleason-GAN framework trained on the SiCAPv2 dataset to create similar pairs of input patches. Our observations indicate that introducing synthetic patches leads to notable improvements in the evaluated metrics, underscoring the utility of generative models within CBIR tasks. Furthermore, this work is the first in the literature where latent representations optimized for CBIR are used to train an attention mechanism for performing Gleason Scoring of a WSI.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":""},"PeriodicalIF":6.7,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143556722","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}
引用次数: 0
ACEA-Net: Weakly Supervised Prostate 3D MRI Image Segmentation via Advanced Prompt Points.
IF 6.7 2区 医学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS Pub Date : 2025-02-19 DOI: 10.1109/JBHI.2025.3543444
Jie Zou, Mengxing Huang, Yu Zhang, Zhiyuan Zhang, Wenjie Zhou, Uzair Aslam Bhatti, Jing Chen, Zhiming Bai

In prostate 3D MRI image segmentation methods, it is usually necessary to annotate each slice, and these annotations are generally time-consuming and specialized. In this study, we generate pseudo-labels using an annotation method with one foreground seed point and six edge relaxation points. We design a weakly supervised semantic learning segmentation framework, ACEA-Net. This segmentation framework solves the under-expansion problem due to the lack of semantic affinity of the seed point pixels in the pseudo-labeling generation process. We design a Seed Cluster Geodesic Distance Transform (SeedGeo) seed expansion strategy to provide a more complete supervised signal. In the segmentation model training phase, Adaptive Convolutional Normalization (ACN) and Enhanced Simple Parameter-Free Attention Module (SimAM) are utilized to smooth the convolutional layer's output in the U-Net baseline model to suppress noisy labels. The proposed segmentation framework achieves excellent segmentation results on the MSD prostate and PROMISE12 prostate datasets, with Dice similarity coefficients (Dice) of 87.23% and 81.00% for the two segmentation tasks, and Average Symmetry Surface Distances (ASSD) of 1.73mm and 2.02mm, respectively, which are superior to the current state-of-the-art method.

{"title":"ACEA-Net: Weakly Supervised Prostate 3D MRI Image Segmentation via Advanced Prompt Points.","authors":"Jie Zou, Mengxing Huang, Yu Zhang, Zhiyuan Zhang, Wenjie Zhou, Uzair Aslam Bhatti, Jing Chen, Zhiming Bai","doi":"10.1109/JBHI.2025.3543444","DOIUrl":"https://doi.org/10.1109/JBHI.2025.3543444","url":null,"abstract":"<p><p>In prostate 3D MRI image segmentation methods, it is usually necessary to annotate each slice, and these annotations are generally time-consuming and specialized. In this study, we generate pseudo-labels using an annotation method with one foreground seed point and six edge relaxation points. We design a weakly supervised semantic learning segmentation framework, ACEA-Net. This segmentation framework solves the under-expansion problem due to the lack of semantic affinity of the seed point pixels in the pseudo-labeling generation process. We design a Seed Cluster Geodesic Distance Transform (SeedGeo) seed expansion strategy to provide a more complete supervised signal. In the segmentation model training phase, Adaptive Convolutional Normalization (ACN) and Enhanced Simple Parameter-Free Attention Module (SimAM) are utilized to smooth the convolutional layer's output in the U-Net baseline model to suppress noisy labels. The proposed segmentation framework achieves excellent segmentation results on the MSD prostate and PROMISE12 prostate datasets, with Dice similarity coefficients (Dice) of 87.23% and 81.00% for the two segmentation tasks, and Average Symmetry Surface Distances (ASSD) of 1.73mm and 2.02mm, respectively, which are superior to the current state-of-the-art method.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":""},"PeriodicalIF":6.7,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143556710","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}
引用次数: 0
Efficient Breast Lesion Segmentation from Ultrasound Videos Across Multiple Source-limited Platforms.
IF 6.7 2区 医学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS Pub Date : 2025-02-19 DOI: 10.1109/JBHI.2025.3543435
Yan Pang, Yunhao Li, Teng Huang, Jiaming Liang, Ziyu Ding, Hao Chen, Baoliang Zhao, Ying Hu, Zheng Zhang, Qiong Wang

Medical video segmentation is fundamentally important in clinical diagnosis and treatment procedures, offering dynamic tracking of breast lesions across frames in ultrasound videos for improved segmentation performance. However, existing approaches face challenges in striking a balance between segmentation performance and inference speed, hindering real-time application in resource-constrained medical environments. In order to address these limitations, we present BaS, a blazing-fast on-device breast lesion segmentation model. BaS integrates the Stem module and BaSBlock to refine representations through inter- and intra-frame analysis on ultrasound videos. In addition, we release two versions of BaS: the BaS-S for superior segmentation performance and the BaS-L for accelerated inference times. Experimental Results indicate that BaS surpasses the top-performing models in terms of segmenting efficiency and accuracy of predictions on devices with limited resources. This work advances the development of efficient medical video segmentation frameworks applicable to multiple medical platforms. Code: https://github.com/aigzhusmart/BaS.

{"title":"Efficient Breast Lesion Segmentation from Ultrasound Videos Across Multiple Source-limited Platforms.","authors":"Yan Pang, Yunhao Li, Teng Huang, Jiaming Liang, Ziyu Ding, Hao Chen, Baoliang Zhao, Ying Hu, Zheng Zhang, Qiong Wang","doi":"10.1109/JBHI.2025.3543435","DOIUrl":"https://doi.org/10.1109/JBHI.2025.3543435","url":null,"abstract":"<p><p>Medical video segmentation is fundamentally important in clinical diagnosis and treatment procedures, offering dynamic tracking of breast lesions across frames in ultrasound videos for improved segmentation performance. However, existing approaches face challenges in striking a balance between segmentation performance and inference speed, hindering real-time application in resource-constrained medical environments. In order to address these limitations, we present BaS, a blazing-fast on-device breast lesion segmentation model. BaS integrates the Stem module and BaSBlock to refine representations through inter- and intra-frame analysis on ultrasound videos. In addition, we release two versions of BaS: the BaS-S for superior segmentation performance and the BaS-L for accelerated inference times. Experimental Results indicate that BaS surpasses the top-performing models in terms of segmenting efficiency and accuracy of predictions on devices with limited resources. This work advances the development of efficient medical video segmentation frameworks applicable to multiple medical platforms. Code: https://github.com/aigzhusmart/BaS.</p>","PeriodicalId":13073,"journal":{"name":"IEEE Journal of Biomedical and Health Informatics","volume":"PP ","pages":""},"PeriodicalIF":6.7,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143556721","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}
引用次数: 0
期刊
IEEE Journal of Biomedical and Health Informatics
全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1