Cardiomyopathy, a severe heart condition, necessitates prompt detection for effective treatment. This project aims to automate the identification of cardiomyopathy in cardiac scan reports using Convolutional Neural Networks (CNNs). By categorizing predictions into normal and sick, the system facilitates swift diagnosis and timely intervention. Leveraging CNNs allows for processing complex patterns in medical imaging data, where the model, trained on extensive cardiac scan datasets, identifies subtle features indicative of cardiomyopathy. This reduces the workload on medical professionals, minimizes human error, and enables quicker clinical decision-making—crucial for early intervention. This research aspires to revolutionize cardiac healthcare practices by enhancing the precision and efficiency of diagnosis, potentially improving patient outcomes and democratizing access to high-quality cardiac care, especially in under-resourced areas.