Deep Learning-Based Automatic ECG Diagnosis for Arrhythmia Identification

One of the most common medical test used to record and evaluate heart activity is the electrocardiogram (ECG), which enables diagnosis of many common heart problems. Recently, learning techniques have been used to identify patterns in healthcare data sets and have stricken success in diverse tasks t...

Full description

Saved in:
Bibliographic Details
Published in2025 4th International Conference on Computing and Information Technology (ICCIT) pp. 482 - 486
Main Authors Regaeig, Khalil, Amara, Ahmed, Boubaker, Khalil, Nasraoui, Leila
Format Conference Proceeding
LanguageEnglish
Published IEEE 13.04.2025
Subjects
Accuracy
Arrhythmia
Classification
Convolutional Neural Network
Convolutional neural networks
Deep Learning
Electrocardiogram
Electrocardiography
Heart
Intelligent systems
Medical diagnostic imaging
Medical tests
Reliability
Rhythm
Online AccessGet full text
DOI10.1109/ICCIT63348.2025.10989469

Cover

More Information
Summary:One of the most common medical test used to record and evaluate heart activity is the electrocardiogram (ECG), which enables diagnosis of many common heart problems. Recently, learning techniques have been used to identify patterns in healthcare data sets and have stricken success in diverse tasks to improve clinical practice. In this paper, we present an intelligent system designed to assist medical professionals in the early diagnosis of heart arrhythmia by leveraging ECG signals. To achieve this, we employ a convolutional neural network (CNN) to classify ECG signals into five distinct cardiac rhythm categories. The experimental results demonstrate the effectiveness of the proposed approach, achieving an average classification accuracy of 0.95 and a loss of 0.2, highlighting its reliability in arrhythmia detection.
DOI:10.1109/ICCIT63348.2025.10989469