Cardiovascular diseases (CVDs) remain the leading cause of mortality worldwide, highlighting the critical need for timely and precise predictive models. Despite recent advancements, significant disparities in healthcare access still exist, prompting the need for innovative solutions in medical diagnostics. This study introduces a novel semi-supervised Grey model using data from the UCI repository, designed to bridge the gap between model interpretability and predictive accuracy—a common trade-off in machine learning. Traditional approaches typically require a choice between White-Box interpretability and Black-Box accuracy. Our proposed Grey-Box model combines the strengths of both, aiming to provide White-Box-like interpretability with nearly Black-Box-level accuracy. This approach offers a promising alternative to conventional models, potentially enhancing the accessibility and effectiveness of medical diagnostics for CVDs.