Abstract

Heart disease is a major challenge for global health, along with high morbidity and mortality. The earlier it is diagnosed, the better the outcome of the patient given timely intervention. This project employs a form of machine learning to train and create a risk assessment model of heart disease from the user-submitted data. The model employs the Random Forest algorithm, one of the most accurate robust algorithms available. We will use a dataset having patient records, such as age, gender, blood pressure, and others. Other medical indicators. Therefore, data preprocessing and feature selection will be applied to enhance the model performance. The main idea is to design an interactive web application which can provide easy access to predict heart disease risk assessment. Users input health parameters, and the application, developed with Streamlit, indicates an immediate prediction for heart disease risk. The interface can be accessed by anyone without technical backgrounds. After the submission, it will process the data through the machine learning model, and risk assessment will be shown. This immediately outcome gotten from this type of risk assessment may prove to be an early warning for patients as they are encouraged to seek medical evaluation in appropriate times. This project, therefore, aims to bridge the gap that exists between advanced machine learning and an actual setting of practical healthcare where users can be well cared for proactively with their hearts and in an effort to ease the burden of heart diseases through early detection and intervention.