A Novel Computational Rough Set Based Feature Extraction For Heart Disease Analysis

Cardiovascular disease is the most difficult disease to diagnose in the medical field. The diagnosis is often contingent on a judgment based on the grouping of vast amounts of clinical and pathological data. As a result of this complication, a number of researchers have based their efforts on determining the most cost-effective and accurate way to predict heart disease.In the case of heart disease, an accurate diagnosis at an early stage is critical, since time is of the essence when heart disease is detected at an inopportune time. Machine learning has evolved in recent years with a plethora of accurate and supporting resources in the medical domain, and it has offered the best support for predicting disease with proper training and research. The main goal of this study is to use a rough computational intelligence approach to find specific heart disease features among a large number of features. The output of the proposed feature selection method outperforms that of conventional feature selection approaches. The rough computation approach's output is evaluated using various heart disease data sets and checked using real-time data sets.