Cardiovascular Disease Prediction based on Electronic Health Records using Machine Learning

Due to a number of fundamental and secondary problems, such as increased risks, illness, and expensive treatment, cardiovascular disease presents a challenging therapeutic practice. In recent times, one of the main causes of death and disability has been heart disease. The machine learning algorithm uses 12 clinical characteristics, including blood pressure, obesity, cholesterol, age, and gender, to identify early stages of cardiac disease and assist medical practitioners. For classification, we employ Naive Bayes, Random Forest, Logistic Regression, and Decision Tree classifiers. Making quick and precise healthcare decisions is facilitated by these classifiers. The program forecasts a person’s risk of developing heart disease using machine learning. The suggested strategy builds a significant knowledge base by identifying interrelationships between medical indicators linked to heart disease. The study enhanced cardiovascular disease prediction accuracy by achieving 97.48 percent classification efficiency and a recall value of 95.5%, surpassing prior studies in the field. The experiment’s unique feature is its ability to predict cardiac disease with 98 percent accuracy in a shorter time, using fewer computational resources and low power.