MULTI-DISEASE DETECTION USING MACHINE LEARNING

Computer Engineering

The increasing demand for early disease diagnosis calls for innovative, accessible, and efficient healthcare solutions that can cater to a wide range of individuals, especially in areas with limited access to healthcare facilities. This paper presents a machine learning-based system designed to predict multiple diseases, including heart disease, kidney disease, thyroid disorders, Parkinson’s disease, diabetes, and brain tumors. For diseases such as heart disease, kidney disease, and diabetes, Random Forest classifiers are utilized, providing accurate, real-time predictions based on relevant health data. For brain tumor detection, a Convolutional Neural Network (CNN) is employed, taking MRI images as input and delivering highly accurate diagnoses. Additionally, the system incorporates a unique symptom-based disease detection feature, where users can select at least three symptoms from a predefined list. This allows the system to predict the most probable disease based on the selected symptoms and suggest appropriate medication and treatment options. A user-friendly, Streamlit-based web application serves as the interface, augmented with an AI-powered chatbot that provides real-time first aid guidance and personalized healthcare advice. This system improves prediction accuracy compared to existing methods, providing a transformative tool for preliminary healthcare screening and democratizing access to diagnostic technologies. By enabling early intervention, the system aims to empower users to take proactive steps in managing their health, especially in regions with limited access to healthcare services. This solution demonstrates a significant advancement in the use of machine learning and AI for healthcare, offering a cost-effective and scalable way to support early disease detection and healthcare accessibility.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.