北京生物医学工程

基于多特征融合的睡眠分期

Sleep staging based on multi-feature fusion

作者：熊馨罗剑花王春武易三莉刘瑞湘贺建峰

单位：昆明理工大学信息工程与自动化学院（昆明650500） 吉林师范大学信息与技术学院（吉林四平136000） 云南省第二人民医院临床心理科（昆明650021） 通信作者：贺建峰，教授。E-mails: jfenghe@foxmail.com

关键词：睡眠分期；脑电；多特征融合；样本熵；小波包能量；去趋势波动

分类号：R318.04

出版年·卷·期（页码）：2021·40·5（487-493）

摘要：

目的为了有效实现睡眠自动分期，对睡眠障碍等相关疾病的诊断提供更多依据，本文提出了一种基于多特征融合的睡眠分期方法。方法数据来自ISRUC-Sleep数据库，首先对10名健康受试者和10名睡眠障碍患者的脑电（electroencephalogram ，EEG）信号计算3种特征—-样本熵、小波包能量和去趋势波动。然后采用支持向量机（support vector machine, SVM）构建睡眠分期模型，并验证该模型的准确性。此外，为了进行比较加入心电（electrocardiogram ，ECG）和肌电（electromyogram ，EMG）通道。结果健康受试者和睡眠障碍患者睡眠分期的准确率分别达到87.4%和86.3%。结论基于多特征融合的睡眠分期方法能够有效地提高睡眠分期的准确率。

Objective In order to realize the automatic sleep staging effectively and to provide more evidence for the diagnosis of sleep disorders and other related diseases, a sleep staging method based on multi-feature fusion is proposed in this paper. Methods Electroencephalogram(EEG) signals come from ISRUC-Sleep database, including 10 healthy subjects and 10 patients with sleep disorder problems. Three characteristics that include sample entropy, wavelet packet energy and detrended fluctuations were calculated. Then Support Vector Machine (SVM) was used to construct the sleep staging model and the accuracy of the model was verified. Additionally, electrocardiogram(ECG) and electromyogram(EMG) signals were also involved for comparison. Results The accuracy of sleep staging in two groups reached 87.4% and 86.3% respectively. Conclusions The sleep staging method based on multi-feature fusion can effectively improve the accuracy of sleep staging.

参考文献：

[1] Solomon GE . Electroencephalography—Basic principles, clinical applications and related fields[M].2 ed. Uvbanl:?Schwarzerberg,1987：97-129.

[2]丰彬. 基于多特征提取的癫痫脑电自动分类检测研究[D]. 南宁：广西大学，2018.

Feng B. Automated classification and detection of epilepsy?EEG based on multi- feature extraction[D]. Nanning：Guangxi University,2018.

[3] 王强强. 基于特征筛选脑电睡眠分期算法研究[D]. 重庆：重庆邮电大学,?2019.

Wang QQ.?Study of sleep staging algorithm based on feature selection[D]. Chongqing：Chongqing University of Posts and Telecommunications,2019.

[4] 王春芳,张力新,刘爽等. 基于去趋势波动分析(DFA)的脑卒中后抑郁症静息脑电特征提取与识别[J]. 中国生物医学工程学报,2013,32(5):520-525.

Wang?CF, Zhang LX, Liu S, et al. Feature extraction and recognition of resting eeg in poststroke depression subjects based on detrended fluctuation analysis[J]. Chinese Journal of Biomedical Engineering,2013,32(5):520-525.

[5] 程佳. 基于脑电信号的睡眠分期研究[D]. 北京：北京理工大学，2015.

Cheng J. Research of sleep staging based on EEG signals[D].Beijing： Beijing Institute of Technology,2015.

[6] 唐其彪. 基于脑电信号的自动睡眠分期研究[D]. 广州：广东工业大学，2019.

Tang QB. Research of automatic sleep staging based on EEG?signals[D]. Guangzhou:Guangdong University of Technology,2019.

[7] 郑声涛,郑日荣,曹义成. 小波包能量谱在睡眠分期中的应用[J]. 工业控制计算机，2014,27(4):87-88.

Zheng ST,Zheng RR,Cao YC.Application of wavelet packet energy spectrum in sleep staging[J].?Industrial Control Computer，2014,27(4):87-88.

[8] Boostani?R,?Karimzadeh?F,?Nami?M. A comparative review on sleep stage classification methods in patients and healthy individuals[J]. Computer Methods and Programs in Biomedicine,?2017,140:?77-91.

[9] 吕甜甜,王心醉,俞乾等. 基于多参数特征融合的自动睡眠分期方法[J]. 计算机应用，2017,37(S2):313-317.

Lyu?TT,Wang XZ,YU Q, et al. Automatic sleep staging method based on multi-parameter feature fusion[J]. Journal of Computer Applications,?2017,37(S2):313-317.

[10]?叶仙,胡洁,田畔等. 基于精细复合多尺度熵与支持向量机的睡眠分期[J]. 上海交通大学学报,2019,53(3):321-326.

Ye?X,?Hu?J,?Tian P,?et al. Automatic sleep scoring based on refined composite multi-scale entropy and support vector machine[J]. Journal of Shanghai?Jiaotong Universitye,2019,53(3):321-326.

[11]?何垣谛. 睡眠分期算法研究[D]. 北京：电子科技大学,2019.

He YD.?Research of Sleep Stage Classification Algorithm[D]. Beijing：University of Electronic Science and Technology of China,2019.

[12] Khalighi?S,Sousa?T, Pires?G,?et al.?Automatic sleep staging: a?computer assisted approach for optimal?combination of features and?polysomnographic channels[J].?Expert Systems with Applications,2013,40:7046-7059.

[13] Khalighi?S,Sousa?T,?Santos?JM,?et al. ISRUC-Sleep: a?comprehensive public dataset for sleep researchers[J]. Computer Methods and Programs in Biomedicine,2016,124:?180-192.

[14] Richman JS,?Moorman JR. Physiological time-series analysis using approximate and entropy[J]. American Journal of Physiology-Heart and Circulatory Physiology

,2000,278(06):2039-2049.

[15]周红标,陈若珠,李军红. 基于小波包变换的癫痫脑电信号特征提取[J]. 电子测量技术,2009,32(10):36-39.

Zhou HB,Chen RZ,Li JH.Feature extraction of epileptic EEG signals based on wavelet package transform[J].?Electronic Measurement Technology,2009,32(10):36-39.

[16]宁艳,江朝晖,安滨等. 睡眠生理参数的去趋势波动分析[J].生物医学工程学杂志,2007,24(2):249-252.

Ning Y, Jiang ZH, An B, et al. Detrended fluctuation analysis of physiological parameters during sleep[J].?Journal of Biomedical Engineering,2007,24(2):249-252.

[17]?周静,吴效明. 基于样本熵的睡眠呼吸暂停综合征脑电研究[J]. 中国医学物理学杂志, 2016,33(7):722-725.

Zhou?J, Wu XM. Electroencephalogram of sleep apnea syndrome based on sample entropy[J].?Chinese Journal of Medical Physics,2016,33(7):722-725.

[18]Sen B, Peker M, Cavusoglu A,et al. A comparative study on classification of sleep stage based on EEG signals using feature selection and classification algorithms[J].Journal of Medical Systems,2014,38(3):1-21.

[19]胡叶容. 基于去趋势波动分析的脑电信号睡眠分期[J]. 生物医学工程学进展,2015,36(4):201-204.

Hu YR.Sleep staging of EEG with detrended fluctuation analysis[J]. Progress in Biomedical Engineering,2015,36(4):201-204.

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录