Web-Based Disease Prediction Application Using Machine Learning for Symptom Analysis

With increased pressure on healthcare systems and growing demands for accurate and timely medical diagnosis, an urgent need exists for intelligent devices to assist individuals in identifying their health status at an initial stage. In this research, an intelligent disease prediction system is designed to promote easier and more accurate initial medical diagnoses using machine learning algorithms. Ultimately, we aim to give individuals preliminary knowledge of upcoming health conditions based on reliable diagnostic support from self-reported symptoms and personal health indicators. For this purpose, we have created an online system based on several ML algorithms: Decision Tree, Random Forest, XGBoost, and Logistic Regression. Different types of disease categories, from broad symptom-based predictions to disease-specific assessments, have been facilitated in the system. Preprocessing of data, feature selection, and optimum model optimization were carried out for accurate predictions. With its developed system, we demonstrate its reasonable efficacy and potential to enable individuals to make better-informed decisions even before seeking medical professionals. In our future work, our purpose is to expand the diagnostic ability of the system for other types of diseases, expand its explainability using explainability AI methods, and ensure real-time data integration from wearable sensors to enhance predictability and personalization further.