Machine Learning based Mobile App for Heart Disease Prediction

Abstract

The world's leading cause of death is heart disease. A variety of modern technologies are utilized to treat cardiac disease The most common problem in medical centers around the world is that many medical personnel lack equal knowledge and courage to treat their patients, so they develop their own opinions, which leads in bad outcomes and, in some cases, death. Predictions of cardiac illness are employed to overcome these issues. This study has used various criteria to predict cardiac disease. These characteristics are Age, Gender, Cerebral Palsy (CP), Blood Pressure (BP), Fasting blood sugar test (FBS), and so on. The major goal of the research is to create a mobile app that reduces the cost of medical tests while also avoiding human bias. The outcome of the research is to forecast cardiac disease. The research made advantage of the built-in dataset and used PHR data to make predictions. Machine Learning is being used to build the model. This study utilizes a variety of machine learning algorithms, including Logistic Regression, ANN Multi-Layer Perceptron (MLP), and Random Forest (RF). Random Forest (RF) outperforms the other two algorithms in terms of accuracy. As a result, this study employs random forest to forecast heart healthand builds the mobile app with MIT App Inventor and stores the data in the Firebase database. The App could be to maintain personal health records and share our info with doctors. It will forecast heart health when you enter the criteria.