Cardio Vascular Disease Prediction and Classification Report Generation using Data Mining Technique

Abstract

Cardiovascular disease is one of the primary reasons for death in the world today. It has evolved into one of the most challenging illnesses to identify. By a recent WHO research, heart disorders are on the rise. As a result, 17.9 million people die each year. As the population increased, this became increasingly difficult to diagnose and initiate treatment during the initial stages. When it comes to forecasting coronary heart disease, medical analysis of data encounters a huge challenge. Electronic health record systems are currently used to handle the data of patients in hospitals. The huge amount of information created by the medical industry is being misused. A new approach is required to reduce costs and accurately predict heart disease. Hospitals can use appropriate decision support systems to reduce the cost of clinical tests. Several types of research offer barely a glimpse of optimism for employing machine learning approaches for predicting cardiac disease. The proposed study suggests a unique strategy for finding key characteristics via a machine learning approach throughout this work, which would also improve the precision of cardiovascular risk diagnosis. Diverse characteristic correlations and classification algorithms are used to establish the statistical model. Using the Improved random forest with Hyper - parameters tweaking in the classification algorithm for cardiovascular disease, a better reliability with an acceptable accuracy of 94.5% has been obtained. This approach may be valuable to healthcare professionals in their treatment as a decision assistance system.