An AI-Driven based Hypertension Disease Prediction Model using ML

Abstract

The worldwide health crisis of hypertension exists because this condition develops without any warning symptoms which results in dangerous complications that include myocardial infarction and stroke and renal failure. The research introduces Predictive Pulse as an artificial intelligence system which uses machine learning to identify hypertension at its beginning stage and monitor disease progression. The system uses real-time physiological signals from wearable devices which track heart rate and activity level and sleep duration to predict blood pressure fluctuations. The evaluation process included five machine learning models which were Logistic Regression and Random Forest and K-Nearest Neighbors KNN and Gradient Boosting and Support Vector Machine SVM. The SVM model achieved its highest accuracy of 99.9% through hyperparameter optimization performed by researchers. The trained model became available as a Flask-based web application, which allowed both people and medical staff to access it easily. The proposed platform demonstrates how AI-based solutions improve preventive healthcare through its automated data-based hypertension monitoring system which operates continuously.

Ähnliche Arbeiten