Coronary artery disease remains the leading cause of mortality worldwide, underling the need for accurate tool to support patient-specific prognosis and risk stratification. Traditionally, clinical decision-making processes has relied on statistically derived risk scores, which are fixed and may not fully capture complex non-linear relationship among patient variables. Recent advances in technology have enabled the application of machine learning and deep learning techniques in medicine, offering the opportunity to undercover deeper feature interactions and enhance the understanding and prediction of cardiovascular outcomes. One of the main disadvantages of artificial intelligence algorithms is their dependence on data quality and availability; insufficient, excessive or fragmented data can hamper the development of a robust and generalizable classifier.