📄 Abstract
Healthcare systems are increasingly adopting intelligent technologies to support early disease detection and improve healthcare decision-making. Delayed diagnosis caused by limited medical awareness or restricted access to healthcare services can increase the risk of serious health conditions. This research presents an Machine Learning Based Disease Prediction System using Python, developed as a Flask-based web application to provide preliminary disease prediction and health risk assessment. The proposed system consists of two prediction modules: symptom-based disease prediction and health risk assessment. In the first module, users select symptoms to predict the most probable disease, while in the second module, users enter health parameters such as age, blood pressure, blood sugar, cholesterol, body mass index (BMI), and heart rate to obtain a health risk prediction. Machine learning algorithms including Decision Tree, Random Forest, Naïve Bayes, and Support Vector Machine (SVM) were trained and evaluated using standard performance metrics such as accuracy, precision, recall, F1-score, and confusion matrix. Among the evaluated models, Random Forest achieved the best prediction performance and was selected for deployment. The developed web application also includes a responsive dashboard, prediction history, health recommendations, downloadable PDF reports, and input validation to enhance usability. Experimental results demonstrate that the proposed system provides accurate predictions, fast response time, and an intuitive user interface, making it suitable for healthcare awareness, academic research, and educational applications. The system is intended as a decision-support tool and does not replace professional medical diagnosis.
🏷️ Keywords
📚 How to Cite:
Pantam Harshitha, Yelaka Keerthi, D. Aruna Padma, Sunil B , MACHINE LEARNING BASED DISEASE PREDICTION SYSTEM USING PYTHON: A FLASK-BASED WEB APPLICATION FOR INTELLIGENT HEALTHCARE PREDICTION , Volume 12 , Issue 7, July 2026, EPRA International Journal of Multidisciplinary Research (IJMR) , Pages: 570 - 584 ,