Medical Diagnosis Using Symptoms

Abstract

Healthcare diagnosis often requires timely and accurate identification of diseases based on patient-reported symptoms, but manual diagnosis can be time-consuming and subject to variations in medical expertise. To address this challenge, this paper presents a machine learning–based disease prediction system that utilizes the XGBoost algorithm for efficient classification of diseases from symptom data. The model is trained on a curated dataset containing multiple diseases and their associated symptoms, enabling it to learn complex relationships and provide reliable predictions. The trained model is integrated into a Flask-based web application, offering users an interactive platform to select symptoms and receive predictions of the top three most likely diseases within seconds. This design not only enhances accessibility for individuals seeking preliminary medical insights but also serves as a supportive tool for healthcare professionals to make quicker decisions. By combining advanced machine learning techniques with modern web technologies, the system demonstrates significant potential in improving healthcare accessibility, reducing diagnostic delays, and contributing to better medical outcomes, while emphasizing that it is intended to complement rather than replace professional medical consultation.