An Interpretable Rough Sets-Based Model for Efficient and Accurate Cardiovascular Risk Assessment

Abstract

Cardiovascular disease stands as the foremost cause of mortality globally, responsible for an estimated 17.9 million deaths annually (source: World Health Organization1). Detecting and preventing cardiovascular disease at its onset are imperative steps in alleviating its burdensome impact. However, conventional risk assessment models often exhibit complexity and lack interpretability, posing challenges for practical utilization by clinicians. This master’s thesis project endeavors to forge ahead by crafting an interpretable rough sets-based model tailored for proficient and precise cardiovascular risk assessment. Leveraging rough sets, a machine learning technique adept at distilling insights from data into a comprehensible format, holds promise in this pursuit. The proposed model will be meticulously constructed utilizing The Cleveland Heart Disease Dataset, its efficacy is gauged across an array of metrics encompassing accuracy, interpretability, and efficiency. With an eye towards transformative potential, the envisioned model aims to revolutionize early detection and prevention strategies for cardiovascular disease. By furnishing clinicians with a more interpretable and streamlined tool for risk assessment, the proposed model is poised to catalyze advancements in cardiovascular healthcare delivery, fostering improved patient outcomes and quality of life.