Electrocardiograph Based Obstructive Sleep Apnea Diagnoses

Abstract

Obstructive sleep apnea (OSA) is a sleeping disorder in which the breathing of a person stops and starts multiple times during sleep. In this process, to automatically diagnose the presence of OSA, a simple, fast, and low cost technique is proposed. In the proposed technique, discrete-Fourier-transform (DFT) is applied to the noise and artifacts free electrocardiograph (ECG) signal. The proposed technique is validated over 260 ECG signals from the PhysioNet database. It is found that compared to polysomnography, which study of sleep profile of a patient and is a long, disturbing, and costly technique, proposed technique performs much better. The accuracy, specificity, and sensitivity of the proposed technique is 98.56%, 97.89%, 99.26% respectively.