BLOOD PRESSURE PREDICTION AND ANALYSIS USING MACHINE LEARNING

This paper presents a Blood Pressure Prediction and Analysis System that leverages machine learning techniques to
estimate systolic and diastolic blood pressure levels using physiological and lifestyle parameters. The system integrates
data preprocessing, feature engineering, and predictive modeling using algorithms such as Random Forest and Neural
Networks to deliver accurate health insights.
The proposed system collects data including age, body mass index (BMI), heart rate, physical activity, stress level, and
dietary habits from users or IoT-enabled wearable devices. After preprocessing and feature selection, the data is used to
train predictive models capable of identifying patterns and forecasting blood pressure levels.
A user-friendly dashboard visualizes historical trends, predicted values, and risk categories such as normal, elevated,
and hypertensive stages. The system also provides real-time alerts and personalized recommendations for preventive
healthcare.
Experimental evaluation demonstrates high prediction accuracy with reduced error rates (low MSE and RMSE), making
the system suitable for real-world healthcare applications. The proposed solution enhances early detection, reduces
hospital dependency, and supports proactive health management.