Estimation of Mental Fatigue During EEG Based Motor Imagery

Abstract

Mental Fatigue is a cognitive state which is an outcome of labour or protracted exercise finally leading to downgrading of mental performance. This leads to reduction in efficiency and disinclination of motor skills. Analysis of mental fatigue thus becomes momentous for assessing one’s capability. The aim here is to analyse whether motor imagery is fatiguing. There are many parameters to evaluate fatigue. This study analyses parietal alpha and frontal theta in motor imagery tasks. Decrement of arousal level, working memory and information encoding have been proven to be associated with increased theta power. Increase in alpha power indicates increase in mental effort to maintain vigilance level. When a person experiences fatigue, their concentration, attention, focus and vigilance level decreases for which they need to put more attention which leads to increase in alpha power. We exploit these EEG oscillatory rhythm fluctuations to model EEG-fatigue relationships. A statistical classifier is used to model EEG-fatigue relationship accurately. With Kernel Partial Least Square output we track the growth of mental fatigue with time.