NAP: Attention-Based Late Fusion for Automatic Sleep Staging
Faculty of Informatics, Università della Svizzera Italiana(信息学院,瑞士意大利大学);Inst. of Digital Tech. for Personalized Healthcare, SUPSI(个性化医疗数字技术研究所,SUPSI);Sleep Wake Epilepsy Center, Department of Neurology, Inselspital University of Bern(睡眠觉醒癫痫中心,神经科,伯恩大学医院);Faculty of Informatics Università della Svizzera Italiana(信息学院,瑞士意大利大学)
Alvise Dei Rossi, Julia van der Meer, Markus H. Schmidt, Claudio L. A. Bassetti, Luigi Fiorillo, Francesca Faraci
多导睡眠图信号具有高度异质性,在不同数据集和临床机构之间,其模态组成(如脑电图、眼电图、心电图)、通道可用性(如额叶、枕叶脑电图)以及采集协议均存在差异。现有大多数处理多导睡眠图数据的模型依赖于固定的模态或通道子集,因此未能充分利用其固有的多模态特性。为解决这一局限,我们提出了NAP(Neural Aggregator of Predictions,预测神经聚合器),这是一种基于注意力的模型,通过三轴注意力机制学习组合多个预测流,该机制能够捕捉时间、空间和预测器级别的依赖关系。NAP经过训练可适应不同的输入维度。通过聚合来自冻结的、预训练的单通道模型的输出,NAP始终优于单个预测器和简单集成方法,在多个数据集上取得了最先进的零样本泛化性能。尽管该方法是在基于多导睡眠图的自动睡眠分期背景下进行验证的,但所提出的方法可以扩展到其他多模态生理应用。
英文摘要:
Polysomnography signals are highly heterogeneous, varying in modality composition (e.g., EEG, EOG, ECG), channel availability (e.g., frontal, occipital EEG), and acquisition protocols across datasets and clinical sites. Most existing models that process polysomnography data rely on a fixed subset of modalities or channels and therefore neglect to fully exploit its inherently multimodal nature. We address this limitation by introducing NAP (Neural Aggregator of Predictions), an attention-based model which learns to combine multiple prediction streams using a tri-axial attention mechanism that captures temporal, spatial, and predictor-level dependencies. NAP is trained to adapt to different input dimensions. By aggregating outputs from frozen, pretrained single-channel models, NAP consistently outperforms individual predictors and simple ensembles, achieving state-of-the-art zero-shot generalization across multiple datasets. While demonstrated in the context of automated sleep staging from polysomnography, the proposed approach could be extended to other multimodal physiological applications.
Sleep Stage Classification across Heterogeneous EEG/EOG Channels with Multi-Channel Differential Transformer
School of Artificial Intelligence and Robotics, Xiamen University Malaysia(厦门大学马来西亚分校人工智能与机器人学院);Department of Biomedical Engineering, Universiti Malaya(马来亚大学生物医学工程系);Department of Biomedical Engineering and Health Sciences, Universiti Teknologi Malaysia(马来西亚理工学院生物医学工程与健康科学系)
Sleep Staging from Airflow Signals Using Fourier Approximations of Persistence Curves
Khoury College of Computer Sciences, Northeastern University(东北大学Khoury计算机科学学院);Courant Institute of Mathematical Sciences, New York University(纽约大学柯朗数学科学研究所);Bouvé College of Health Sciences, Northeastern University(东北大学Bouvé健康科学学院)
A Context-Aware Temporal Modeling through Unified Multi-Scale Temporal Encoding and Hierarchical Sequence Learning for Single-Channel EEG Sleep Staging
Self-supervised Contrastive Learning for EEG-based Sleep Staging
Wuhan University(武汉大学)
EEG信号通常很容易获得,但标记起来很昂贵。虽然监督学习在脑电信号分析领域得到了广泛的应用,但其泛化性能受到标注数据量的限制。自监督学习(Self-supervised learning,SSL)作为计算机视觉(Computer Vision,CV)和自然语言处理(Natural Language Processing,NLP)领域的一种流行的学习范式,可以利用未标记的数据来弥补监督学习的数据不足。在本文中,我们提出了一个自我监督的对比学习方法的脑电信号的睡眠阶段分类。在训练过程中,我们为网络设置了一个任务,以匹配从脑电信号中产生的正确的变换对。通过这种方式,网络通过学习脑电信号的一般特征来提高表示能力。网络的鲁棒性也在处理不同的数据时得到了提高,即从变化的数据中提取恒定的特征。具体而言,网络的性能取决于转换的选择和自监督学习训练过程中使用的未标记数据量。在Sleep-edf数据集上的实验结果表明,该方法在睡眠分期上具有较好的性能(88.16%的准确率和81.96%的F1分数),并验证了SSL策略在有限标记数据制度下对EEG信号分析的有效性。所有代码都在网上公开提供。
SleepPPG-Net: a deep learning algorithm for robust sleep staging from continuous photoplethysmography
Technion-Israel Institute of Technology(以色列理工学院);University of Cambridge(剑桥大学);University of London(伦敦大学);Mines ParisTech - PSL University(巴黎国立高等矿业学校-巴黎文理研究大学)
Zhejiang University(浙江大学);Affiliated Mental Health Center & Hangzhou Seventh People’s Hospital, Zhejiang University School of Medicine(浙江大学医学院附属精神卫生中心&杭州市第七人民医院);The First Affiliated Hospital, College of Medicine, Zhejiang University(浙江大学医学院附属第一医院)
Rethinking Random Transformers as Adaptive Sequence Smoothers for Sleep Staging
School of Biological Science and Medical Engineering(生物科学与医学工程学院);Southeast University(东南大学);University of Bath(巴斯大学);Donders Institute for Brain, Cognition and Behaviour(脑、认知与行为研究所);Radboud University Medical Center(拉德堡德大学医学中心);School of Computer Science & Software Engineering(计算机科学与软件工程学院)
SleepPPG-Net2: Deep learning generalization for sleep staging from photoplethysmography
The Faculty of Data and Decision Sciences, Technion, Israel Institute of Technology(以色列理工学院数据与决策科学学院);Faculty of Biomedical Engineering, Technion, Israel Institute of Technology(以色列理工学院生物医学工程学院)
SleepBand: Single-Source Domain Generalization for Sleep Staging via Physiologically Structured Spectral Modeling
Laboratory of Intelligent Collaborative Computing, University of Electronic Science and Technology of China(电子科技大学智能协同计算实验室);School of Information Science and Technology, University of Science and Technology of China(中国科学技术大学信息科学学院);School of Cyber Science and Technology, University of Science and Technology of China(中国科学技术大学网络科学与技术学院)
ST-USleepNet: A Spatial-Temporal Coupling Prominence Network for Multi-Channel Sleep Staging
Saw Swee Hock School of Public Health, National University of Singapore(新加坡国立大学公共健康学院);Institute of Data Science, National University of Singapore(新加坡国立大学数据科学研究所);Beijing Key Laboratory of Brainnetome and Brain-Computer Interface, Institute of Automation, Chinese Academy of Sciences(北京脑网膜与脑机接口重点实验室,中国科学院自动化研究所);Brainnetome Center, Institute of Automation, Chinese Academy of Sciences(脑网膜中心,中国科学院自动化研究所)
Towards More Accurate Automatic Sleep Staging via Deep Transfer Learning
Queen Mary University of London(伦敦玛丽女王大学);Singapore Institute of Technology(新加坡理工学院);University of Oxford(牛津大学);Peking University(北京大学);University of Lübeck(吕贝克大学);KU Leuven(鲁汶大学)
SleepGMUformer: A gated multimodal temporal neural network for sleep staging
College of Computer Science and Technology, Qingdao University(青岛大学计算机科学与技术学院);School of Information Science and Engineering, University of Jinan(济南大学信息科学与工程学院);School of Computing and Mathematical Sciences, University of Leicester(莱斯特大学计算与数学科学学院)
University of Eastern Finland(东芬兰大学);Kuopio University Hospital(库奥皮奥大学医院);The University of Queensland(昆士兰大学);Reykjavik University(雷克雅未克大学);Landspitali–The National University Hospital of Iceland(冰岛国立大学医院);Princess Alexandra Hospital(亚历山德拉公主医院)
Pediatric Automatic Sleep Staging: A comparative study of state-of-the-art deep learning methods
Queen Mary University of London(伦敦玛丽女王大学);Alan Turing Institute(艾伦·图灵研究所);University of Lübeck(吕贝克大学);German Research Center for Artificial Intelligence (DFKI)(德国人工智能研究中心);University of Adelaide(阿德莱德大学)
ProductGraphSleepNet: Sleep Staging using Product Spatio-Temporal Graph Learning with Attentive Temporal Aggregation
Sharif University of Technology(谢里夫理工大学);Massachusetts General Hospital(麻省总医院);Harvard Medical School(哈佛医学院);Georgia Institute of Technology(佐治亚理工学院);Emory School of Medicine(埃默里大学医学院)
SleepTransformer: Automatic Sleep Staging with Interpretability and Uncertainty Quantification
Queen Mary University of London(伦敦玛丽女王大学);Alan Turing Institute(艾伦·图灵研究所);Aarhus University(奥胡斯大学);University of Bristol(布里斯托大学);University of Lübeck(吕贝克大学);German Research Center for Artificial Intelligence (DFKI)(德国人工智能研究中心(DFKI));KU Leuven(鲁汶大学)
From Sleep Staging to Spindle Detection: A Case Study on End-to-End Automated Sleep Analysis
Department of Medical Engineering and Technomathematics, FH Aachen University of Applied Sciences(医学工程与技术数学系,亚琛应用科学大学);Department of Information and Computing Sciences, Utrecht University(信息与计算科学系,乌得勒支大学);Institute for Data-Driven Technologies, FH Aachen University of Applied Sciences(数据驱动技术研究所,亚琛应用科学大学);Department of Psychiatry and Psychotherapy, University Hospital Carl Gustav Carus, Technische Universität Dresden(精神病学与心理治疗系,卡尔·古斯塔夫·卡尔斯大学医院,德累斯顿技术大学)