北京生物医学工程

基于光电容积脉搏波的精神疲劳评估方法

Mental fatigue assessment method based on photoplethysmography signal

作者：余越严良文曹可乐

单位：上海大学机电工程与自动化学院（上海 200444）<br />通信作者：严良文。E-mail: lw_yan@staff.shu.edu.cn

分类号：R318.04

出版年·卷·期（页码）：2023·42·1（45-51）

摘要：

目的针对精神疲劳难于定量评估的问题，本文探索一种非侵入式可穿戴检测方法获取人体生理参数，从而实现对人体精神疲劳的定量评估。方法搭建光电容积脉搏波（photoplethysmography， PPG）采集平台采集20名健康在校生的PPG信号，对PPG信号进行预处理和特征提取，获取时、频域共143维特征。使用机器学习方法建立分类模型，对比特征选择方法皮尔逊相关系数法、F检验和relief-F得到特征权值，选择最优的特征子集，使用降维后的特征子集训练模型，减少复杂度和过拟合概率。结果与实际状态对比，基于该方法的单个体疲劳检测平均准确率为92.48%，多个体疲劳检测准确率最大值为92.2%，可以有效地识别精神疲劳。结论光电容积脉搏波信号经过时频域分析构建的特征能够使用机器学习算法进行准确的精神疲劳状态分类评估。

Objective In view of the difficulty in quantitative assessment of mental fatigue, this paper explores a non-invasive wearable detection method to obtain human physiological parameters, so as to achieve quantitative assessment of human mental fatigue. Methods PPG signals of 20 healthy school students were collected by photoplethysmography (PPG) acquisition platform. PPG signals were preprocessed and features were extracted. A total of 143 dimensional features were obtained in time and frequency domains. The machine learning method was used to establish the classification model, and the feature weights were obtained by comparing the feature selection method Pearson correlation coefficient method, F test and Relief -F. The optimal feature subset was selected, and the feature subset training model after dimension-reduction was used to reduce the complexity and overfitting probability. Results Compared with the actual state, the average accuracy of individual fatigue detection based on this method was 92.48%. The maximum accuracy of multi-individual fatigue detection is 92.2%, which can effectively identify mental fatigue. Conclusions The features constructed by time-frequency domain analysis of photoplethysmography signals can be used for accurate mental fatigue state classification assessment using machine learning algorithms.

参考文献：

[1]苗雨, 刘晓虹, 刘伟志, 等. 多维疲劳量表中文版在军队基层医护人员中的初步修订[J]. 中国心理卫生杂志, 2008, 22(09): 658-660, 668.
Miao Y, Liu XH, Liu ZW, et al. Initial revision of the chinese version of multidimensional fatigue inventory-20 in medical staff of military basic level [J]. Chinese Mental Health Journal, 2008(09): 658-660, 668.
[2] kerstedt T,Gillberg M. Subjective and objective sleepiness in the active individual[J]. Taylor & Francis, 2009, 52(1-2): 29-37.
[3]何金良. 人体疲劳评价标准综述[J]. 西藏医药杂志, 2013, 34(02): 29-32.
He JL. Review of human fatigue evaluation criteria[J]. Tibetan Medicine, 2013, 34(02): 29-32.
[4]赵晓华, 房瑞雪, 荣建, 等. 基于生理信号的驾驶疲劳综合评价方法试验研究[J]. 北京工业大学学报, 2011, 37(10): 1511-1516, 1523.
Zhao XH, Fang RX, Rong J, et al. Experiment study on comprehensive evaluation method of driving fatigue based on physiological signals[J]. Journal of Beijing University of Technology. 2011, 37(10): 1511-1516, 1523.
[5] Lee BG, Park JH, Pu CC, et al. Mobile-based kernel-fuzzy-c-means-wavelet for driver fatigue prediction with cloud computing[C]. Sensors. IEEE, 2014:1236-1239.
[6] 于露, 金龙哲, 徐明伟, 等. 基于光电容积脉搏波的有限空间生理疲劳测量[J]. 工程科学学报, 2018, 40(10): 1215-1222.
Yu L, Jin LZ, Xu MW, et al. Confined space physiological fatigue measurement based on photoplethysmography pulse wave signal[J]. Chinese Journal of Engineering, 2018, 40(10): 1215-1222.
[7]赵小静, 路海月, 王梦悦，等. 基于心率变异性的脑力疲劳检测[J]. 中国医学物理学杂志, 2018, 35(5): 592-597.
Zhao XJ, Lu HY, Wang MY, et al. Mental fatigue detection based on heart rate variability[J]. Chinese Journal of Medical Physics, 2018, 35(5): 592-597.
[8]王大伟, 贾荣丛, 王划一. 基于Matlab的巴特沃斯滤波器设计[J]. 现代电子技术, 2012, 35(21): 71-72, 75.
Wang DW, Jia RC, Wang HY. Design of Butterworth analog filter based on Matlab[J]. Modern electronics technique, 2012, 35(21): 71-72, 75.
[9] 孔令杰. 脉搏图像与心电信号多信息同步采集及其应用研究[D].兰州：兰州理工大学, 2011.
Kong LJ. Synchronous acquisition and application based on multi-information of pulse images and ECG signals[D]. Lanzhou：Lanzhou University of Technology, 2011.
[10] 杨琳, 张松, 杨益民, 等. 利用三次样条差值法抑制脉搏波基线漂移[J]. 北京生物医学工程, 2010, 29(2): 198-200.
Yang L, Zhang S, Yang YM, et al. Removal of pulse waveform baseline drift using cubic spline interpolation[J]. Beijing Biomedical Engineering, 2010, 29(02): 198-200.
[11] 李顶立. 基于脉搏波的无创连续血压测量方法研究[D]. 杭州：浙江大学, 2008.
Li SL. Research on non-invasive continuous blood pressure measurement based on pulse wave[D]. Hangzhou: Zhejiang University, 2008.
[12] Mesa I, Rubio A, Tubia I, et al. Channel and feature selection for a surface electromyographic pattern recognition task[J]. Expert Systems with Applications, 2014, 41(11): 5190-5200.
[13] Chieh TC, Mustafa MM, Hussain A, et al. Development of vehicle driver drowsiness detection system using electrooculogram (EOG)[C]. 2005 1st International Conference on Computers, Communications, & Signal Processing with Special Track on Biomedical Engineering. IEEE. 2005: 165-168.
[14] 黄诗童, 张威强, 张朋柱. 基于HRV分析的可穿戴心电仪精神疲劳检测[J]. 计算机应用研究, 2019, 36(07): 2093-2097, 2103.
Huang ST, Zhang WQ, Zhang PZ. Detection of mental fatigue with wearable ECG devices based on HRV analysis[J]. Application Research of Computers, 2019, 36(07): 2093-2097, 2103.

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