Data Classification for Analyzing Characteristics of Electrocardiogram and Electroencephalogram under Continuous Long-Time Mental Calculation and Rest

Electrocardiogram (EKG) and Electroencephalogram (EEG) are widely used for kinds of disorders detection. In case of EKG, RR interval series is used for heart rate variability (HRV) analysis, which is a reliable reflection of status of autonomic nervous system. HRV is a function of both physical and mental activity. In order to analyze the influence of metal stress on HRV, EKG signals including information of physical activities should be removed. In case of EEG, the major artifacts are induced by electromyogram (EMG) and electrooculogram (EOG). In order to analyze the influence of metal stress on EEG, the signals which include information of body movement should be removed. In this paper, we present a method to classify EEG and EKG signals, based on body movement estimation and artifacts detection. Long time recording are divided into segments and classified. The results indicate that the data classification method purposed in this paper is effective, and body movement is important for analyzing EKG and EEG.