Detection and Classification of Cardiovascular Diseases in ECG images by using Deep Learning

Authors: Akshay Ramesh Navale, Aditya Ravindra Nikam, Tushar Kailas Kusmude, Tejas Anil Lahase, R. N. Muneshwar

Country: India

Full-text Research PDF File:   View   |   Download

Abstract: Cardiovascular diseases (CVDs) continue to be the primary cause of death globally, placing immense pressure on healthcare systems to provide accurate and timely diagnoses for effective intervention. Electrocardiograms (ECGs) are widely used to detect and diagnose heart conditions; however, analyzing ECG images alone may not fully capture a patient’s overall cardiovascular health or the complex risk factors influencing CVD. Traditional ECG-based methods often lack integration with other critical health data, which can result in limited diagnostic precision and missed insights.
To address these limitations, this project proposes a hybrid deep learning model that combines ECG image analysis with patient medical history to create a more comprehensive diagnostic tool for cardiovascular disease classification. This model utilizes convolutional neural network (CNN) architectures, such as Alex Net and Squeeze Net, for ECG feature extraction, allowing the system to identify subtle visual patterns and irregularities in heart activity. The extracted ECG features are then integrated with structured medical data—such as age, blood pressure, cholesterol levels, and medical history—processed through a fully connected neural network (FCNN). By merging visual and non-visual health data, the hybrid model aims to improve classification accuracy and provide a deeper understanding of CVD risk factors.
This integrated approach enables a more holistic and data-rich diagnostic system, capable of delivering enhanced prediction reliability and supporting early intervention strategies. With its ability to analyze diverse health data, the model holds the potential to revolutionize cardiovascular diagnostics, ultimately aiding in personalized treatment and improved patient outcomes.

Keywords: Cardiovascular disease (CVD), Electrocardiogram (ECG), Deep learning, Hybrid model, CNN (Convolutional Neural Network) SqueezeNet, Fully connected neural network (FCNN), Medical history integration, Disease classification, Health data fusion

Paper Id: 232417

Published On: 2025-04-24

Published In: Volume 13, Issue 2, March-April 2025

Share this