主要观点总结
本文介绍了关于生物医学信号和AI结合在医疗领域的应用，特别是征稿启事的相关内容。文章还提供了相关的论文参考和顶刊速看系列。
关键观点总结
关键观点1: 人工智能（AI）与生物医学信号的融合为医疗领域带来巨大机会
AI在生物医学信号分析中的应用，如心电图、脑电图和肌电图的分析，有助于提高医疗评估的准确性、效率和可访问性。
关键观点2: 征稿启事关于AI在生物医学信号分析中的特殊议题
邀请原创研究文章、综述和案例研究，涉及机器学习在生物医学信号分析中的应用，包括疾病诊断、可视化改善、自动报告生成、不良事件检测等。
关键观点3: IEEE JBHI期刊的IF值为6.7的相关论文
提到了基于深度学习和脑电图实现扩散性去极化的实时无创检测等论文，这些论文展示了AI在医学中的应用。
关键观点4: 其他相关论文和顶刊速看系列
包括JAMA Network Open、NC（Nature Communications）等顶级期刊上发表的关于AI在医学中的最新研究进展，如基于人工智能的皮瓣监测系统、基于深度学习心电图的先天性心脏病检测等。
正文
• 使用生物医学信号进行疾病诊断的
机器学习
；
•
机器学习辅助
可视化/改善生物医学信号的表示；
•
机器学习
用于自动报告生成；
• 使用生物医学信号进行不良事件检测（例如，猝死）的
机器学习
；
• 机器学习用于个
性化生物医学信号解释
；
•
机器学习用于生成逼真的生物医学信号
；
·
• 用于胎儿、新生儿和儿童生物医学信号分析的机器学习方法；
• 用于从不平衡的生物医学信号数据集中去偏算法的机器学习方法；
•
多模态机器学习
（例如，将ECG和笔记与LLMs结合，交互多个生物医学信号）；
• 精心设计生物医学信号的基础模型，特别关注偏见和公平性；
• 机器学习驱动的生物医学信号分析中的监管和伦理考虑。
客座编辑
Jintai Chen，香港科技大学（广州），[email protected]
Shenda Hong，北京大学，[email protected]
加里·克利福德，埃默里大学和佐治亚理工学院，[email protected]
Jimeng Sun，伊利诺伊大学厄巴纳-香槟分校，[email protected]
关键日期
提交截止日期：2025年2月28日
首次审查截止日期：
2025年5月30日
修订稿截止日期：2025年8月30日
最终决定：
2025年11月31日
2.征稿原文如下
IEEE JOURNAL OF
BIOMEDICAL AND HEALTH INFORMATICS
J-BHI Special Issue on “
Towards Foundation Models of Biomedical Signals for Healthcare
”
The rapid evolution of artificial intelligence (AI) and healthcare presents substantial opportunities
for engineers, computational researchers, and medical experts to develop innovative algorithms for
health monitoring, medical diagnostics, and treatment recommendations, ultimately benefiting
both doctors and patients.
Biomedical signals such as the electrocardiogram (ECG), electroencephalogram (EEG), and
electromyogram (EMG) play a crucial role in the non-invasive monitoring and diagnosis of various
health conditions. These biomedical signals are rich in clinically useful information, reflecting the
underlying physiological and pathological states of the heart, brain, and muscles, respectively. The
integration of AI with these signals has opened new avenues for enhancing the accuracy, efficiency,
and accessibility of medical assessments. For example, in the realm of electrocardiogram, AI-based
algorithms can automatically detect some arrhythmias and abnormalities with close to expert-level
accuracy. However, the development of reliable AI-driven diagnostic tools that use biomedical
signals still face challenges such as noise, interference, artifacts, and the need for robust processing
of very long-term data streams.
Recent advancements in AI, including large language models (LLMs), Mamba neural network, and
Generative AI, have opened new opportunities for developing advanced neural network models to
address biomedical data challenges. As a cornerstone of this interdisciplinary field, foundation
models may serve as sophisticated frameworks, integrating vast biomedical signal data and
enabling the creation of predictive, diagnostic, and therapeutic tools that promise to be more
precise, specific, and personalized, thus potentially revolutionizing the diagnostic and monitoring
landscape. This special issue aims to explore the latest advancements and applications of AI in
biomedical signal analysis for human healthcare, that will pave the way for foundation models.
We invite original research articles, reviews, and case studies that address, but are not limited to,
the following topics:
•
Machine learning using biomedical signals for disease diagnosis;
•
Machine learning assisted visualization/improved representation for biomedical signals;
•
Machine learning for automated report generation;
•
Machine learning for adverse event detection (e.g., sudden death) using biomedical signals;
•
Machine learning for personalized biomedical signal interpretation;
•
Machine learning for realistic biomedical signal generation;
•
Machine learning approaches for fetal, neonatal, and pediatric biomedical signal analysis;
•
Machine learning approaches for debiasing algorithms from imbalanced biomedical signals
datasets;
•
Multimodal machine learning (e.g. combining ECG and notes with LLMs, interacting
multiple biomedical signals);
•
Thoughtful design of foundation models for biomedical signals, particularly with attention
to bias and fairness;
•
Regulatory and ethical considerations in machine-learning-powered biomedical signal
analysis.
Guest Editors
Jintai Chen, Hong Kong University of Science and Technology (Guangzhou),
[email protected]
Shenda Hong, Peking University,
[email protected]
