Fuzzy rules-based prediction of heart conditions system

Heart disease is known as the deadliest disease in the world which mostly focus on coronary diseases, cerebrovascular diseases, and ischemic heart disease. The treatment for the diseases is highly costly, and not only that, the monitoring system or devices that are in the market are low in accuracy and not satisfying. This work proposed to develop a prediction system for heart conditions using fuzzy system that is based on essential risk factors: age, gender, body mass index (BMI), blood pressure level (systolic), cholesterol level, heart rate, smoking habit, alcohol intake, eating habit and exercise. The specific fuzzy rules are created and produced in the output category of low, medium, and high risks. The proposed system was later evaluated by comparing the machine learning performance metrics such as accuracy, specificity, sensitivity and F1 score. It is found that the accuracy, sensitivity, specificity and F1 score are calculated as 88.2%, 78.8%, 21.2%, and 80.9%, respectively, which demonstrates a reliable percentage score. It is believed that this work has the potential to be an alternative method in providing as a dependable and cheap means of predicting heart disease.