面向心理压力评估的脑电信号多重分形去趋势波动分析方法研究

发布时间：2018-05-14 01:00

本文选题：脑电信号 + 多重分形去趋势波动分析　；参考：《生物医学工程学杂志》2017年02期

【摘要】：本文基于多重分形去趋势波动分析方法(MFDFA)实现了对受试者心理压力状态的评估。研究针对不同心理压力状态下,脑电信号多重分形去趋势波动分析中最优分形阶数确定问题,重点分析了多重分形去趋势波动分析方法中的奇异指数、Hurst指数等参数与阶数的关系,进而确定最优分形阶数,实现了基于脑电信号多重分形去趋势波动分析的心理压力状态评估。试验采集了14名在校学生有/无心理压力状态下的脑电信号,分别比较了奇异指数、奇异维数、Hurst指数、质量指数与阶数关系,确定了最优分形阶数范围为[—5,5],实现了基于脑电信号β波多重分形去趋势波动分析方法的心理压力状态评估。研究结果表明,心理压力状态下,脑电信号的Hurst指数和质量指数大于无压力状态下,脑电信号的相应参数,随着阶数的变大,Hurst指数减小,趋近于定值,而质量指数增大,奇异值随阶数的变化幅度较明显。本文还比较了有/无心理压力状态下,脑电信号的峰值和奇异谱宽度,结果表明,不同心理压力状态下脑电信号多重分形谱特性不同,心理压力状态下,脑电信号的奇异谱宽度明显大于无压力状态下脑电信号的奇异谱宽度。本文研究结果说明,该方法可以有效地评估心理压力状态,为实现心理压力状态干预,提高心理健康等提供支持与帮助。
[Abstract]:Based on the multifractal trend fluctuation analysis method (MFDFA), this paper realizes the assessment of the psychological stress state of the subjects. The study focuses on the determination of the optimal fractal order in the multi fractal detrending fluctuation analysis of EEG signals under different psychological pressures, and focuses on the analysis of the singular exponents in the multifractal detrending wave analysis method. The relationship between the parameters of Hurst index and the order, and then determine the optimal fractal order, and realize the psychological pressure state assessment based on the analysis of the multi fractal detrending fluctuation of EEG signal. The experiment collected 14 students with / without psychological pressure, and compared the singular index, the singular dimension, the Hurst index, the mass index, respectively. In relation to the order, the optimal fractal scale is determined to be [- 5,5]], and the psychological pressure state evaluation based on the Bo multifractal detrending wave analysis method is realized. The results show that the Hurst index and mass index of EEG signal are larger than that under pressure state, and the corresponding parameters of EEG signals are associated with the stress state. The Hurst exponent decreases and the index decreases. The mass index increases, the mass index increases and the singular value varies with the order. The peak value and the singular spectrum width of EEG signal under / without psychological pressure are compared. The results show that the multi fractal spectrum characteristics of the brain electrical signals in different psychological pressure States are different, and the psychological pressure is different. The singular spectrum width of EEG signals is obviously larger than the singular spectrum width of the EEG in the non pressure state. The results show that the method can effectively evaluate the psychological stress state and provide support and help for the intervention of psychological pressure and the improvement of mental health.

【作者单位】：燕山大学电气工程学院生物医学工程研究所;燕山大学河北省测试计量技术及仪器重点实验室;北京工业大学生命科学与生物工程学院;前景光电技术有限公司;
【基金】：河北省自然科学基金资助项目(F2014203244) 中国博士后科学基金资助项目(2014M550582)
【分类号】：R741.044;TN911.6

【相似文献】