Abstract:
"Coronary Artery Disease (CAD) remains a major global health concern, where early detection is vital to improving patient outcomes and reducing mortality rates. This research focuses on developing an efficient CAD prediction system using Electrocardiography (ECG) data and machine learning classification. Motivated by the need for a non-invasive diagnostic tool, the study aims to identify individuals at risk of CAD, enabling timely intervention and personalized healthcare.
The proposed solution utilizes a Convolutional Neural Network (CNN), chosen for its ability to capture spatial dependencies in sequential data such as ECG signals. The CNN architecture was meticulously designed, optimizing parameters such as the number of layers, kernel sizes, and activation functions to enhance predictive performance. Data preprocessing and feature extraction techniques were integrated to improve input data quality, ensuring a robust model training process.
Preliminary results demonstrate the model's potential, achieving a classification accuracy of 99.51% on the test dataset. These findings highlight the efficacy of the approach while emphasizing the need for further evaluation to ensure robustness across diverse datasets. This promising performance underlines the importance of continued optimization and exploration, paving the way for a valuable tool in early CAD detection and prevention.
The study showcases the intersection of machine learning and healthcare, emphasizing the transformative potential of data-driven approaches in addressing critical medical challenges."
