Cardiac Irregularity Detection Using Photoplethysmogram Signal

Abstract

In this study, photoplethysmogram (PPG) based heart abnormality detection method was proposed. PPG signals utilized were obtained from MIMIC II Waveform Database, Version 3 Part 1 with sampling frequency of 250 Hz with the duration of 10 seconds each. The feature of the PPG signals were then extracted using MATLAB and the distances between successive minimum troughs as well as the area of Cardioid graphs of PPG signals were calculated and evaluated to differentiate the normal and abnormal PPG signal. Based on the experimentation results, distances between minimum trough and the area of the Cardioid graphs of abnormal PPG signals are larger than the normal segments. Therefore, the results show the proof-of-concept of the proposed heart abnormality detection technique and suggest a better alternative to the current techniques.