Deep learning model for prediction of elderly stage of sudden cardiac arrest using the interval window of ECG signal

View at publisher Summarise with VIRAT
Authors: Arti Badhoutiya, Manoj Kumar Ojha, Deepak Gupta, Amar Saraswat
Cited by: 0
References: 29

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This paper introduces a deep learning model that is intended to forecast the likelihood of sudden cardiac arrest (SCA) in cardiac patients by examining ECG signal intervals. The model automates the identification of critical points within ECG data and improves efficiency by using binary convolutional neural networks (CNNs). It adopts a 12-layer 1D CNN architecture, with weights and activations binarized during runtime. The model is trained on a labelled dataset annotated by cardiologists, assuring accurate training based on critical ECG features. ECG signals are segmented into two-second intervals for effective processing. The model that was developed demonstrated exceptional performance in the prediction of SCA, with a precision of 98.83%, recall of 98.81%, and F1-score of 98.81%. Additionally, the proposed model has an average accuracy of 99% in the accurate classification of SCA patients and normal controls. It indicates that the model not only accurately predicts the risk of SCA but also significantly outperforms current state-of-the-art methods.

Fields of Study & Subjects

ECG Monitoring and Analysis Cardiac electrophysiology and arrhythmias Artificial Intelligence in Healthcare

Publication details

Journal / source
AIP conference proceedings
Volume
3401
Pages
020069–020069
Publication date
1 January 2026
Language
EN