AI-Based Disease Prediction System (Diabetes, Heart & Liver Diseases)

Healthcare generates massive datasets, offering significant opportunities for deriving meaningful clinical insights. This paper presents an Artificial Intelligence (AI)-based Disease Prediction System designed to estimate the likelihood of three prevalent chronic conditions: Diabetes, Heart Disease, and Liver Disease, using various medical and lifestyle parameters. By training specialized Machine Learning (ML) models—Logistic Regression for Diabetes, Random Forest for Heart Disease, and Support Vector Machine (SVM) for Liver Disease—on publicly available medical datasets (Kaggle/UCI), the system effectively identifies complex correlations between health indicators and disease onset. Users can input parameters like age, BMI, cholesterol, and glucose level through an interactive web interface, receiving instant disease probability predictions. The system achieved notable accuracies: 83.5% for Diabetes, 90.2% for Heart Disease, and 87.4% for Liver Disease. This application is a crucial tool for early detection, preventive care, and augmenting clinical decision-making, showcasing the potential of multi-disease ML platforms in modern healthcare. The abstract summarizes the objectives, methodology, obtained results, and their significance, remaining under 200 words and containing no numerical references.