脉搏信号功率谱分析对精神疲劳状态的识别-中国组织工程与临床康复2007年1期目次

脉搏信号功率谱分析对精神疲劳状态的识别***☆

张爱华，豆小玺，王龙

AIM： Based on the internal relationship between mental fatigue and pulse wave, to design a new method for diagnosing and evaluating mental fatigue by using feature parameters extracted from pulse signal, so as to provide scientific evidence for the diagnosis and evaluation of mental fatigue.

METHODS： The experiment was conducted in the laboratory of Electrical Information Institute, Lanzhou University of Technology from April to May 2006. Twenty healthy male college student volunteers were selected for mental fatigue experiment of mentally adding and subtraction operation with three-digit numbers for 3 hours. The pulse signals of radial artery were recorded by using HK-2000C digital integrated pulse transducer and data processing system with the frequency of 200 Hz. Based on the pulse power spectral analysis with modified Welch method, the power spectral peak value and the corresponding frequency at 0－30 Hz were extracted as the feature parameters. Then linear discriminant analysis was applied to classify the extracted features.

RESULT:All 20 subjects were involved in the result analysis. After the mental fatigue experiment, the power spectral peak values of pulse signal were decreased significantly compared with those before experiment [(28.22±1.37), (50.56±1.13) dB, P < 0.01], and the corresponding frequencies were increased obviously [(0.930 7±0.064 4), (0.865 7±0.064 0) Hz, P < 0.05]. The recognition accuracy was up to 85% with the power spectral peak value and the corresponding frequency.

CONCLUSION： The power spectra of pulse signals can reflect the state of mental fatigue. The power spectral peak value and the corresponding frequency could be the objective parameters for evaluation of mental fatigue.

College of Electrical and Information Engineering, Lanzhou University of Technology, Lanzhou 730050, Gansu Province, China

Zhang Ai-hua☆, Doctor, Professor, College of Electrical and Information Engineering, Lanzhou University of Technology, Lanzhou 730050, Gansu Province, China
zhangah@lut.cn

Supported by: the National Natural Science Foundation of China, No. 30670529*, 10572056*; Scientific Research Program Foundation of Gansu Educational Bureau, No. 0503-09*

Received: 2006-08-10
Accepted: 2006-10-07

摘要
目的：根据精神疲劳状态与脉搏信号间的内在联系，设计一种基于脉搏信号特征参数评测精神疲劳状态的方法，为精神疲劳状态的诊断和评测提供科学依据。
方法：实验于2006－04/05在兰州理工大学电子信息研究所实验室进行。选择20名身体健康的男性大学生志愿者，进行连续3 h的3位随机数加减运算的精神疲劳实验。采用HK-2000C集成化数字脉搏传感器和数据采集处理系统获取被试者精神疲劳前后的桡动脉脉搏信号，采样频率为200 Hz，进而用改进后的Welch法对脉搏信号进行功率谱分析，提取0～30 Hz功率谱峰值及峰值频率。应用线性判别式分析对所提取的特征进行分类。
结果：20名受试者全部进入结果分析。精神疲劳实验后，脉搏信号的功率谱峰值显著低于实验前[（28.22±1.37），（50.56±1.13） dB，P < 0.01]；峰值频率明显升高[（0.930 7±0.064 4），（0.865 7±0.064 0） Hz，P < 0.05]。以功率谱峰值和峰值频率作为特征量, 精神疲劳识别正确率达到了85%。
结论：脉搏信号的功率谱分析能够反映人体的精神疲劳状态；脉搏信号的功率谱峰值和峰值频率有望作为精神疲劳状态评测的客观指标。
关键词：精神疲劳；脉搏功率谱；线性判别式

张爱华，豆小玺，王龙.脉搏信号功率谱分析对精神疲劳状态的识别[J].中国组织工程研究与临床康复，2007，11(1):118-120
[www.zglckf.com/zglckf/ejournal/upfiles/07-1/1k-118(ps).pdf]

兰州理工大学电气工程与信息工程学院，甘肃省兰州市 730050

张爱华☆，女，1964年生，河北省永年县人，汉族， 2005年西安交通大学毕业，博士，教授，主要从事生物医学信号检测与处理研究。
zhangah@lut.cn

国家自然科学基金资助项目（30670529*，10572056*）；甘肃省教育厅科研基金项目（0503-09）*

中图分类号:R319.1 文献标识码:A
文章编号:1673-8225
(2007)01-00118-03

收稿日期：2006-08-10
修回日期：2006-10-07
(06-50-8-6097/N·LL)

热点资讯：疲劳是人体的一种主观反应，可表示人体的亚健康状态。但疲劳状态时通常不伴有明显的病理表现,在临床医学检查中难以发现和评判，所以目前对疲劳状态的评估仍缺乏客观的定量测量指标。

创新要点： ①设计一种连续3 h的两个3随机数加减运算的精神疲劳实验。②采用HK－2000C集成化数字脉搏传感器和数据采集处理系统采集被试者精神疲劳前后的桡动脉脉搏信号。③用改进后的Welch法对脉搏信号进行功率谱分析，并提取了可以有效识别精神疲劳状态的脉搏信号特征量。

同行评价：文章采集了被试者精神疲劳前后的桡动脉脉搏信号，提取了可以有效识别精神疲劳状态的脉搏信号特征量，所应用的方法有新意，在疲劳的检测和评估方面具有一定的应用价值。