A machine learning approach for high blood pressure prediction and music control

Abstract

High blood pressure (HBP) has been one of the major threats to human health. Lack of early detection and control of high blood pressure can cause severe damages to the heart which may lead to death. Most adults suffering from high blood pressure are unaware of the disease because it may have no warning signs or symptoms. This research focuses on the real time prediction of high blood pressure using a machine learning approach and control of high blood pressure using music. The study synchronizes a machine learning technique with a simulator to predict blood pressure and play low beat music if the blood pressure is high. The research was carried out using a large dataset with the following attributes (education, age, body mass index, current smoker and heart rate). Random forest algorithm was the machine learning technique used to construct and validate the prediction model. The prediction accuracy of the model exceeds 97% and the model was able to accurately predict blood pressure and play low beat music when the blood pressure is high.