Disease Diagnosis for Healthcare System using Random Forest Machine Learning

Abstract

Over one billion people worldwide suffer from high blood pressure, which is defined as having a diastolic pressure of 90 or higher and a systolic pressure of 140 or higher. Tobacco usage is estimated to cause about 10% of all cardiac problems, with approximately one billion people currently smoking worldwide. Elevated blood sugar levels, particularly in people with diabetes, significantly increase susceptibility to HD, a leading contributor to more than 60% fatalities among those with diabetes. The presence of elevated cholesterol levels in the bloodstream also increases the risk of developing coronary artery disease and stroke, accounting for 29% of cases of ischemic coronary disease. Despite the availability of tools and methodologies for the prediction of cardiac illnesses, there are still no effective models that can identify the disease. However, the increasing volume of data presents an opportunity for ML to play a pivotal role. As a subset of Artificial Intelligence (AI), ML has gained increasing popularity and is anticipated to become even more prominent. In the current digital era, remote monitoring devices collect vast amounts of patient data, contributing to 30% of global data generated by the healthcare industry annually, with an expected increase of 36% by 2025. This paper is predicting heart disease (HD) using different machine learning (ML) technique. The 10552 HD dataset is used for this paper and all technique is implement python software and calculate precision (P), accuracy (A) and recall (R). In all ML technique, the random forests (RF) provide good result compared to other technique.