Journal of Applied Sciences ›› 2014, Vol. 32 ›› Issue (6): 631-638.doi: 10.3969/j.issn.0255-8297.2014.06.013

• Signal and Information Processing • Previous Articles     Next Articles

Entropy Analysis of EEG for Elderly Patients with Mild Cognitive Impairment

WEI Ling1, ZHAO Jian-qiang1, SHI Jun1, XUE Qing2,3, WANG Yu-ping2,3, LI Ying-jie1   

  1. 1. School of Communication and Information Engineering, Shanghai University, Shanghai 200444, China
    2. Xuanwu Hospital, Capital Medical University, Beijing 100053, China
    3. Beijing Key Laboratory of Neuromodulation, Beijing 100053, China
  • Received:2013-07-20 Revised:2014-04-04 Online:2014-11-28 Published:2014-04-04

Abstract:  In this study, eight elderly patients with mild cognitive impairment (MCI) and 17 normal controls
participated in a cognition task. The subjects were asked to judge whether the color of two graphics is matched
or not, and their scalp electroencephalographs (EEG) were recorded. We calculated the relative power, sample
entropy (SEn) and cumulative residual entropy (CREn) of the event-related EEG signals to explore differences
between the MCI patients and normal controls. The following results were obtained. 1) There was no
difference in alpha relative power, while the MCI patients had higher complexity at medial-frontal and medialposterior
regions than the normal controls. 2) The frontal region showed significantly higher complexity than
other regions and the center hemisphere had lower complexity than left and right hemisphere in both groups
throughout the task. 3) Complexity in anterior brain in match task was higher than that in mismatch task
in both groups. Normal controls had more complex EEG activity during the match task in the right-center
region. 4) Relative to SEn, the analysis of CREn found more specific nonlinear characteristic in the EEG of
MCI patients. All these results show the MCI patients have abnormal EEG activity in brain, and CREn is
more suitable for describing the nonlinear characteristic of EEG than SEn.

Key words: cognitive impairment (MCI), electroencephalographs (EEG), relative power, sample entropy, cumulative residual entropy

CLC Number: