AUTOMATED CARDIAC DISEASE PREDICTION AND SEVERITY DETECTION USING IMAGE SEGMENTATION AND DEEP LEARNING

Abstract

Cardiovascular disease remains a leading cause of mortality worldwide, necessitating accurate and early diagnosis. Cardiac imaging, combined with advanced computational techniques, plays a vital role in identifying and assessing heart conditions. This project explores the application of deep learning—particularly Convolutional Neural Networks (CNNs)—in analyzing multimodal cardiac images to improve diagnostic accuracy and efficiency. The proposed system focuses on identifying disease-specific regions in CT images by employing CNN-based image representation and segmentation. A K-Nearest Neighbor (KNN) classifier is used to segment the heart image into three regions based on color, isolating both affected and unaffected areas. By calculating the percentage of affected pixels, the model estimates the severity of the disease, enabling more informed and timely treatment decisions. This approach demonstrates the potential of AI-driven tools to enhance noninvasive diagnostics in cardiology while minimizing procedural risks and costs.

Ähnliche Arbeiten