AI BASED DISEASE PREDICTION

Abstract

This project presents the design and implementation of an AI-based disease prediction system that assists healthcare professionals and patients in the early identification of life-threatening conditions using advanced machine learning algorithms. The exponential growth in healthcare data and the increasing availability of patient electronic health records have created a compelling opportunity to apply data-driven intelligence to clinical decision-making. Early and accurate disease diagnosis remains one of the most significant challenges in modern medicine, and this project directly addresses that gap by developing a robust predictive analytics platform. The system analyzes patient health parameters such as symptoms, vital signs, laboratory test results, age, gender, and medical history to predict the likelihood of diseases including diabetes, cardiovascular disease, and other chronic conditions. A machine learning-driven approach is employed, integrating data preprocessing, feature extraction, classification algorithms, and probability scoring to ensure accurate and clinically meaningful predictions. Multiple algorithms were evaluated to assess diagnostic performance, with effectiveness measured through standard clinical metrics including accuracy, precision, recall, F1-score, and ROC-AUC. Among the evaluated models, Random Forest delivered the best overall performance with an accuracy of 92.4% and a ROC-AUC score of 0.9617, identifying key biomarkers, patient age, blood pressure, and glucose levels as the most influential disease predictors. The proposed system highlights key risk factors influencing disease onset and progression, enabling targeted preventive care strategies. A user-friendly web-based interface built with Streamlit enables users to input patient data, visualize prediction results, compare algorithm performance, and receive actionable health insights without requiring advanced technical knowledge. By automating disease screening and enhancing clinical decision-making through predictive analytics, the system aims to improve early diagnosis rates, reduce healthcare costs, and increase the likelihood of successful treatment outcomes. The system represents a significant step towards accessible, scalable, and intelligent healthcare decision support.