A Hybrid Approach for Sleep States Detection Using Blood Pressure and EEG Signal

Abstract

AbstractSleep is fundamental to every human being, and it is a physiological phenomenon. The aim of the research is to develop a hybrid approach using time and frequency domain features for automatic sleep states detection. In time domain, statistical and morphological features are extracted from the blood pressure signal, and in frequency domain, a piecewise linear reduction namely Franklin transformation is applied on EEG signal. The Franklin coefficients are used as discriminatory features in frequency domain. The feature vector is constructed by combining 13-time domain features and first 8 Franklin coefficients. Features are classified by using KNN classifier. The novelty of the proposed method is that we considered two cases, the blood pressure signal by itself, and the combination of it with EEG signal. The motivation behind the first one is that in certain cases, e.g., smart personal mobile devices, only the blood pressure signal is available. In both cases, the algorithm is tested on MIT-BIH Polysomnographic database having more than 80 h long-term EEG and blood pressure signals. In both cases, we performed comparison tests with relevant state-of-the-art methods, and our algorithm showed better or equal performance in terms of sensitivity, specificity, accuracy, and false alarm rate.KeywordsSleep statesFaber-SchauderFranklin systemK-nearest neighborADASYNEEGBlood pressureSystolic and diastolic temperatureDicrotic notch