Machine Learning-Based Approach for Cardiovascular Disease Detection and Classification

Abstract

Cardiovascular disease (CVD) remains the primary cause of mortality worldwide and continues to exhibit an alarming upward trend. Detecting CVD efficiently and accurately in large populations has become an urgent necessity. The proposed framework aims to detect and classify five major cardiovascular disorders in the heart, including Heart Attack, Heart Failure, Heart Valve disease, pericardial disease, and vascular disease (Blood Vessel Disease). Developed within the Matlab environment, the system will undergo comprehensive simulation to ensure its effectiveness. To evaluate the system’s performance, a two-pronged approach will be employed: psycho-visual and parametric analysis. Through psycho-visual analysis, human experts will visually assess the system’s outputs, offering qualitative insights into its accuracy and reliability. Meanwhile, parametric analysis will utilize objective metrics to quantitatively measure the system’s efficiency in detecting and categorizing the cardiovascular conditions. The successful implementation of this proposed system holds promise for early and precise identification of cardiovascular disorders, facilitating timely medical interventions and improving patient outcomes. By incorporating both subjective and objective evaluations, this research seeks to develop a robust and efficient tool for cardiovascular disease screening, ultimately contributing to enhanced healthcare and reduced disease burden.