Single Channel based Demarcation of Yogic and Non-Yogic Sleep Patterns using Observational Sleep EEG

Abstract

Yoga practice brings some of the important physiological and biochemical improvements that lead to better well-being and mental prosperity. Yoga is not just simply helpful in enhancing core stability, adaptability, and levels of anxiety; it also boosts sleep effectiveness, sleep latency, duration of sleep, and quality of sleep by relieving pain, depression, and anxiety, and relaxing the mind. This study aims at demarcating the sleep patterns of yogic and non-yogic subjects. In this work, time domain statistical parameters like mean, maximum, minimum, median along with frequency domain features like dominant frequency and Shannon entropy of the normalized PSD are considered as the discriminating features for classification of EEG (O1A1 Channel) with 0.5-sec window length with 50% overlap. The experimental results show that KNN classifier verify with 95% confidence interval, sensitivity, specificity and accuracy of 99%., 99% and 99.4%., respectively.