郭天娇,王敬哲,张志远,龙语诺,张兆南

目的：睡眠分期是睡眠障碍诊断和治疗的重要环节，传统睡眠分期方法主要依赖多导睡眠图（PSG）进行人工判读，存在操作复杂、干扰性强、成本较高等问题，本文旨在提出一种基于呼吸和心电的多模态自动睡眠分期模型，为基于可穿戴、非接触的自动睡眠分期提供技术支持。方法：构建基于深度学习的多模态睡眠分期模型，以呼吸和心电的时域及频域信号作为输入，通过多尺度互信息特征提取（MMIFE）模块、全局关系建模（GRM）模块及交叉域融合（CLF）模块实现多模态、多属性特征的高效融合与互补增强；利用公开的睡眠数据集对模型进行训练与测试，以评估模型的性能。结果：所提模型在SHHS数据集上睡眠分期准确率为70.6%，Kappa系数为0.696，优于经典睡眠分期方法。结论：本文提出一种基于呼吸和心电的多模态自动睡眠分期模型，实现了呼吸、心电信号特征的跨域融合，有效提高了睡眠分期的准确率和可靠性，为可穿戴设备和非接触式监测技术在睡眠健康领域的应用提供了新的思路和方法，具有重要的理论和实际应用价值。

睡眠分期；深度学习；多模态生理信号；特征融合

[1] Maquet P.The role of sleep in learning and memory[J].Science,2001,294(5544):1048-1052. [2] 王俊秀,张衍,李延泽,等.《中国睡眠研究报告(2025)》[M].北京:社会科学文献出版社,2025:127-128. [3] Shimozono K,Adachi H,Nohara S,et al.Sleep-disordered breathing profiles in patients with cardiovascular diseases:Kurume SDB-CVD study[J].Scientific Reports,2025,15(1):16521-16531. [4] Berry,RB.,Budhiraja,Gottlieb,D.J.et al.Rules for scoring respiratory events in sleep:Update of the 2007 AASM manual for the scoring of sleep and associated events(Article)[J].Journal of Clinical Sleep Medicine,2012,8(5):597-619. [5] 郭瑞表,钱小顺,孙宝君,等.多导睡眠图监测的影响因素与干预措施[J].中国综合临床,2013,29(4):392-393. [6] Lee JH,Nam H,Kim DH,et al.Developing a deep learning model for sleep stage prediction in obstructive sleep apnea cohort using 60 GHz frequency-modulated continuous-wave radar.[J].Journal of Sleep Research,2024,33(1):1-9. [7] Long X,Foussier J,Fonseca P,et al.Analyzing respiratory effort amplitude for automated sleep stage classification[J].Biomedical Signal Processing and Control,2014,14:197-205. [8] 黄文汉,张伟,胡立刚,等.基于心电与呼吸信号的睡眠分期算法研究[J].智能计算机与应用,2018,8(1):49-54. [9] 王凤霞.基于迁移学习的非接触式睡眠监测技术研究[D].南京:南京理工大学硕士学位论文,2023. [10] Quan S,Howard BV,Iber C,et al.The Sleep Heart Health Study:design,rationale,and methods[J].Sleep,1997,20(12):1077-1085. [11] Lin Y,Wang M,Hu F,et al.Multimodal Polysomnography-Based Automatic Sleep Stage Classification via Multiview Fusion Network[J]. IEEE Transactions on Instrumentation and Measurement,2024,73:1-12. [12] Supratak A,Dong H,WU C,et al.DeepSleepNet:A model for automatic sleep stage scoring based on rawsingle-channel EEG[J].IEEE Transactions on Neural Systems and Rehabilitation Engineering,2017,25(11):1998-2008. [13] Phan H,Mikkesen K,Chen Y,et al.Sleep-Transformer:Automatic sleep staging with interpretability and uncertainty quantification[J]. IEEE Transactionson Biomedical Engineering,2022,69(8):2456-2467.

点击下载PDF