Cardiovascular Diseases (CVD) is one of the leading causes of mortality worldwide, highlighting the need for early and more accurate identification methods. This project presents a new approach to recognizing cardiovascular disease through analysis of ECG images using enhanced image processing and deep learning techniques. System preparation for ECG images improves signal clarity and extracts features that indicate cardiac damage using a folding network (CNN). By using a robust training pipeline and the status of -ART classification methods, the model identifies important cardiovascular diseases such as arrhythmias and high-precision myocardial infarction. The purpose of the proposed solution is to improve diagnostic efficiency, support relatives of health occupations in the fact that facts are well-discovered decisions, and provide a scalable framework for real applications. This approach not only contributes to early detection and treatment of CVD, but also forms the basis for the provision of AI-controlled health solutions in a variety of clinical settings, facilitating accessible and reliable diagnosis, particularly due to automated evidence of mobile networks and efficient net architecture CVD images. Electrocardiogram (ECG).