Feature Extraction and Recognition of Resting EEG in Poststroke Depression Subjects Based on Detrended Fluctuation Analysis
ANG Chun Fang1 ZHANG Li Xin1 LIU Shuang1 SUN Chang Cheng2 WANG Yong Jun2ZHAO Xin1 QI Hong Zhi1 ZHOU Peng1 WAN Bai Kun1 DU Jin Gang2 MING Dong1*
1 Department of Biomedical Engineering, College of Precision Instrument and OptoElectronics Engineering, Tianjin University, Tianjin 300072, China)
2 Tianjin Institute of Rehabilitation Medicine, Tianjin People’s Hospital, Tianjin 300121, China
Abstract：To analyze the specificity of resting EEG signals in poststroke depression (PSD) subjects, the detrended fluctuation analysis (DFA) were used to extract feature parameters of scaling exponent α (the slope of the linear fit in the double logarithmic coordinate relating fluctuation functions F(s) and time sequence length) to 16 channels. The α values of 16 channel EEG signals for three different groups (10 healthy controls, 4 poststroke nondepression subjects and 7 poststroke depression subjects) were conducted independent ttest. Results show that there was a significant difference (P<0.05) of the scaling exponent α between healthy subjects and post stroke subjects in the parietal, temporal and occipital lobe. With the scaling exponents feature α to be 16 dimensional feature space, pattern recognition performed up to 90.9% classification accuracy using support vector machine (SVM) among poststroke subjects (poststroke depression subjects and poststroke nondepression subjects included), which can be expected to provide a new means of clinical aided diagnosis for PSD subjects objectively and effectively.
［1］王文志．中国脑卒中流行病学特征和社区人群干预 ［J］. 中国医学前沿杂志，2009，1(2)：49-53．
［2］丁世飞，齐丙娟，谭红艳．支持向量机理论与算法研究综述 ［J］. 电子科技大学学报，2011,40(1)：2-10．
［3］Angelelli P, Paolucci S, Bivona U, et al. Development of neuropsychiatric symptoms in poststroke patients: a crosssectional study ［J］. Acta Psychiatr Scand, 2004,110(1)：55-63．
［4］何舒，林渝峰，李霞．抑郁症的发病机制研究进展 ［J］. 四川生理科学杂志，2006，28（3）：126-128．
［5］Baskaran A, Milev R, McIntyre RS．The neurobiology of the EEG biomarker as a predictor of treatment response in depression ［J］. Neuropharmacology, 2012, 63(4)：507-513．
［6］Koek RJ, Yerevanian BI, Tachiki KH, et al．Hemispheric asymmetry in depression and mania A longitudinal QEEG study in bipolar disorder ［J］. Journal of Affective Disorders, 1999,53(2)：109-122．
［7］陈爱美．基于神经振荡的抑郁症脑电信号分析 ［D］. 秦皇岛：燕山大学，2010．
［8］胡巧莉．基于相位同步分析方法的抑郁症脑电信号的研究 ［J］. 中国医疗器械杂志，2010,34(4)：246-249．
［9］Davidson RJ．Anterior electrophysiological asymmetries, emotion, and depression: Conceptual and methodological conundrums ［J］. Psychophysiology, 1998, 35(5)：607-614．
［10］Pollock V, Schneider L．Quantitative, waking EEG research on depression ［J］. Biol Psychiatry, 1990, 27(7)：757-780．
［11］Henriques JB, Dividson RJ. Regional brain electrical asymmetries discriminate between previously depressed and healthy control subjects ［J］. J Abnorm Psychol 1990,99(1)：22-31．
［12］Ahmadlou M, Adeli H, Adeli A．Fractality analysis of frontal brain in major depressive disorder ［J］. Int J Psychophysiol, 2012, 85(2)：206-211．
［13］李颖洁，邱意弘，朱贻盛．脑电信号分析方法及其应用 ［M］. 北京:科学出版社，2009：111-117．
［14］梁晓花．基于脑电心电数据融合的睡眠分期 ［D］. 南京：江苏大学，2008．
［15］Watters PA, Martin F. A method for establishing longrange power law correlations from the electroencephalogram［J］. Biol Psychiatry, 2004,66:79-89．
［16］Nikulin VV, Brismar T. Longrange temporal correlations in electroencephalographic oscillations: relation to topography, frequency band, age and gender［J］. Neuroscience, 2005,130:549-58．
［17］Watters PA, Martin F．A method for estimating longrange power law correlations from the electroencephalogram ［J］. Biol Psychiatry, 2004,66(1)：79-89.
［18］Nikulin VV, Brismar T．Longrange temporal correlations in electroencephalographic oscillations: relation to topography, frequency band, age and gender ［J］. Neuroscience, 2005,130(2)：549-558．
［19］Lee JM, Kim DJ, Kim IY, et al．Nonlinearanalysis of human sleep EEG using detrended fluctuation analysis ［J］. Med Eng Phys, 2004,26(9)：773-776．
［20］Gifani P, Rabiee HR, Hashemi MH, et al．Optimal fractalscaling analysis of human EEG dynamic for depth of anesthesia quantification ［J］. J Franklin Inst, 2007,344：212-229．
［21］Lee JS, Yang BH, Lee JH， et al. Detrended fluctuation analysis of resting EEG in depressed outpatients and healthy controls ［J］. Clinical Neurophysiology, 2007,118:2489-2496．
［22］Parish LM, Worrell GA, Cranstoun SD, et al．Longrange temporal correlations in epileptogenic and nonepileptogenic human hippocampus ［J］. Neuroscience, 2004,125(4)：1069-1076．
［23］Bak P, Tang C, Wiesenfeld K．Selforganized criticality: an explanation of the 1/ f noise ［J］. Phys Rev Lett,1987，59(4)：381-384．
