Detection of sleep arousal from STFT-based instantaneous features of single channel EEG signal.

IF 2.3 4区 医学 Q3 BIOPHYSICS
Md Hussain Ali, Md Bashir Uddin
{"title":"Detection of sleep arousal from STFT-based instantaneous features of single channel EEG signal.","authors":"Md Hussain Ali, Md Bashir Uddin","doi":"10.1088/1361-6579/ad7fcb","DOIUrl":null,"url":null,"abstract":"<p><strong>Objective: </strong>Sleep arousal, a frequent interruption in sleep with complete or partial wakefulness from sleep, may indicate a breathing disorder, neurological disorder, or sleep-related disorders. These phenomena necessitate the detection of sleep arousals. Uses of deep learning methods to detect features inhibits the scope to understand the specific distinctive nature of the signals and reduces the interpretability of the model. To evade these inconsistencies and to improve the classification performance of the sleep arousal detection model, a model has been proposed in this study on the prospect of understandable features that are useful in detecting sleep arousals. &#xD;Approach: Time-frequency analysis of the electroencephalogram (EEG) signals was performed using Short-Time Fourier Transform (STFT). From the STFT coefficients, the spectrogram and instantaneous properties (frequency, bandwidth, power spectrum, band energy, local maxima, and band energy ratios) were investigated. From these properties, instantaneous features were generated by statistical analysis. Additive feature sets and reduced feature sets, formed by adding features successively and reducing features using the analysis of variance test respectively, were subjected to a tri-layered neural network classifier to evaluate the capability of the features to detect sleep arousal and normal sleep segments. &#xD;Main results: The reduced feature set (Set 6) has proved to be efficacious in facilitating superior classification performance metrics (accuracy, sensitivity, specificity, and AUC of 89.14%, 83.52%, 89.49%, and 93.84% respectively). &#xD;Significance: This efficient model can be incorporated with an automatic sleep apnea detection system where the estimation of hypopnea requires the detection of sleep arousal.&#xD;&#xD.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":null,"pages":null},"PeriodicalIF":2.3000,"publicationDate":"2024-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Physiological measurement","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1088/1361-6579/ad7fcb","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"BIOPHYSICS","Score":null,"Total":0}
引用次数: 0

Abstract

Objective: Sleep arousal, a frequent interruption in sleep with complete or partial wakefulness from sleep, may indicate a breathing disorder, neurological disorder, or sleep-related disorders. These phenomena necessitate the detection of sleep arousals. Uses of deep learning methods to detect features inhibits the scope to understand the specific distinctive nature of the signals and reduces the interpretability of the model. To evade these inconsistencies and to improve the classification performance of the sleep arousal detection model, a model has been proposed in this study on the prospect of understandable features that are useful in detecting sleep arousals. Approach: Time-frequency analysis of the electroencephalogram (EEG) signals was performed using Short-Time Fourier Transform (STFT). From the STFT coefficients, the spectrogram and instantaneous properties (frequency, bandwidth, power spectrum, band energy, local maxima, and band energy ratios) were investigated. From these properties, instantaneous features were generated by statistical analysis. Additive feature sets and reduced feature sets, formed by adding features successively and reducing features using the analysis of variance test respectively, were subjected to a tri-layered neural network classifier to evaluate the capability of the features to detect sleep arousal and normal sleep segments. Main results: The reduced feature set (Set 6) has proved to be efficacious in facilitating superior classification performance metrics (accuracy, sensitivity, specificity, and AUC of 89.14%, 83.52%, 89.49%, and 93.84% respectively). Significance: This efficient model can be incorporated with an automatic sleep apnea detection system where the estimation of hypopnea requires the detection of sleep arousal. .

从基于 STFT 的单通道脑电信号瞬时特征检测睡眠唤醒。
目的:睡眠唤醒是指睡眠经常中断,并从睡眠中完全或部分醒来,可能预示着呼吸紊乱、神经紊乱或睡眠相关疾病。这些现象都需要对睡眠唤醒进行检测。使用深度学习方法来检测特征,会抑制对信号特殊性的理解,降低模型的可解释性。为了避免这些不一致性,并提高睡眠唤醒检测模型的分类性能,本研究提出了一个模型,有望获得有助于检测睡眠唤醒的可理解特征:使用短时傅里叶变换(STFT)对脑电图(EEG)信号进行时频分析。根据 STFT 系数,研究了频谱图和瞬时特性(频率、带宽、功率谱、频带能量、局部最大值和频带能量比)。根据这些特性,通过统计分析生成瞬时特征。利用三层神经网络分类器对通过连续添加特征和利用方差分析检验减少特征而形成的添加特征集和减少特征集进行了处理,以评估这些特征检测睡眠唤醒和正常睡眠片段的能力:事实证明,减少后的特征集(特征集 6)能有效提高分类性能指标(准确率、灵敏度、特异性和 AUC 分别为 89.14%、83.52%、89.49% 和 93.84%):这一高效模型可用于睡眠呼吸暂停自动检测系统,在该系统中,低通气估计需要检测睡眠唤醒。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Physiological measurement
Physiological measurement 生物-工程:生物医学
CiteScore
5.50
自引率
9.40%
发文量
124
审稿时长
3 months
期刊介绍: Physiological Measurement publishes papers about the quantitative assessment and visualization of physiological function in clinical research and practice, with an emphasis on the development of new methods of measurement and their validation. Papers are published on topics including: applied physiology in illness and health electrical bioimpedance, optical and acoustic measurement techniques advanced methods of time series and other data analysis biomedical and clinical engineering in-patient and ambulatory monitoring point-of-care technologies novel clinical measurements of cardiovascular, neurological, and musculoskeletal systems. measurements in molecular, cellular and organ physiology and electrophysiology physiological modeling and simulation novel biomedical sensors, instruments, devices and systems measurement standards and guidelines.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
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
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
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 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信