Real-Time Data Integration and Analysis for Heart Disease Prediction Using Machine Learning

Real-Time Data Integration and Analysis for Heart Disease Prediction Using Machine Learning

Ms. Velvizhi R

Department of Computer Science and Engineering

Sathyabama Institute of Science and Technology

Chennai, India velvizhisp@gmail.com

Ms. Manju C Nair

Department of Computer Science and Engineering

Sathyabama Institute of Science and Technology

Chennai, India manjucnair2007@sathyabama.ac.in

Hemanth G

Department of Computer Science and Engineering

Sathyabama Institute of Science and Technology

Chennai, India g.hemanth1403@gmail.com

Sai Harshith G

Department of Computer Science and Engineering

Sathyabama Institute of Science and Technology

Chennai, India harshithgokavarapu1908@gmail.com

Abstract—The " Heart Disease Risk Assessment Web App" is a cutting-edge web application that uses sophisticated machine learning techniques to deliver individualized cardiovascular risk information. The app uses information from the Behavioral Risk Factor Surveillance System (BRFSS) 2022 to determine personalized risk scores and provide useful advice based on user inputs including age, BMI, physical activity, smoking status, and medical history. While SHAP (SHapley Additive exPlanations) improves transparency by clarifying feature relevance, the app uses advanced models, such as Easy Ensemble with LightGBM, to handle class imbalance and guarantee reliable predictions. With the help of Streamlit, the software effectively reduces the risk of heart disease and encourages proactive health management by fusing accuracy, accessibility, and user interaction to provide people with useful information.

Keywords—health evaluation,heart disease, cardiovascular health,SHAP,Simple Ensemble Classifier Model LightGBM,Streamlit,BRFSS dataset,Individualized health advice,Handling class imbalance