PhD

Sep 1, 1996 - Dec 6, 2000

Human Machine Communication System

Smartphone Camera-Based PPG for ECG Parameters Estimation Using Artificial Neural Network

Jan 1, 2025

Journal AIP Conference Proceedings

publisher AIP

DOI 10.1063/5.0258999

The procedure of recording the heart's electrical activity using an electrocardiography instrument is called an electrocardiogram (ECG). ECG is an important examination for cardiac patients and elderly people. Because the ECG test is difficult to conduct without operators' participation, recently the researchers try to rebuild the ECG signal from photoplethysmography (PPG). PPG has a strong link to the cardiovascular system in human. In this study, a low-cost approach using Smartphone camera to estimate the essential ECG parameters (PR interval, QRS interval, QT interval and RR interval) presented. The camera of a smartphone was used to measure PPG signal. The suggested methodology relates each PPG signal pulse to corresponding cardiac cycle (CC) parameters obtained from the ECG test. These PPG pulses were used by Artificial Neural Network (ANN) to generate a training model for the ECG parameters. 40 subjects participated in the study, each of whom had a 20-second video recorded simultaneously with their ECG reading. The experimental results showed good performance accuracy in estimating the ECG parameters, which came out to be 0.15*10-3 Mean Square Error (MSE). The regression between the output and the targets showed an outstanding correlation of 0.998125 for the testing data.