ATRIAL FIBRILLATION RISK PREDICTION FROM ELECTROCARDIOGRAM AND RELATED HEALTH DATA WITH DEEP NEURAL NETWORK

ATRIAL FIBRILLATION RISK PREDICTION FROM ELECTROCARDIOGRAM AND RELATED HEALTH DATA WITH DEEP NEURAL NETWORK

Sharmila R1, R. Mohan Kumar2

1P.G Scholar, Dept. of Biomedical Engineering, Udaya School of Engineering, Vellamodi, Tamil Nadu, India

2Assistant Professor, Dept. of Biomedical Engineering, Udaya School of Engineering, Vellamodi, Tamil Nadu, India

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - Atrial fibrillation (AF) is a common cardiac arrhythmia that can increase the risk of stroke and other complications. Early detection and risk prediction of AF can aid in the prevention of its complications. ECG is an important non-invasive clinical tool for the diagnosis of heart diseases. The detection of cardiac arrhythmias is a challenging task since the small variations in ECG signals cannot be distinguished precisely by the human eye. The objective of this work is to detect cardiac arrhythmias with the highest detection accuracy. This project proposes a method for AF risk prediction using electrocardiogram (ECG) signals and deep neural networks (DNNs) combined with wavelet transformation implemented in MATLAB. The proposed method involves preprocessing the ECG signal using wavelet transformation to extract time-frequency features, followed by feature selection and classification using a DNN algorithm. The proposed model was evaluated on a dataset of ECG signals and achieved an accuracy of 96%, demonstrating its potential for clinical use in AF risk prediction. The proposed approach has the potential to assist clinicians in identifying patients at high risk of AF and providing personalized interventions to prevent its complications. The use of MATLAB makes the proposed approach readily accessible to researchers and clinicians in the field.

Key Words:  Atrial fibrillation, electrocardiogram, wavelet transformation, deep neural networks, ada-boost classifier.