Abstract: Thyroid disease is a widespread endocrine disorder that often goes undiagnosed due to the limitations of conventional diagnostic methods. This project proposes a machine learning-based web application for automated detection of thyroid disorders using both clinical data and medical images. The system integrates Fuzzy C-Means clustering for analyzing hormone-level data and Convolutional Neural Networks (CNN) for classifying thyroid ultrasound images. It supports role-based access for technicians, doctors, and patients, streamlining diagnosis, prescription, and feedback. The proposed model achieved over 96% accuracy in classification, outperforming traditional algorithms. With modules for appointment booking and real-time doctor-patient chat, the system offers a user-friendly and scalable solution for intelligent healthcare. It aims to reduce human error, improve early detection, and make thyroid diagnosis more accessible, especially in resource-limited settings.
Keywords: Thyroid Disease, CNN, Fuzzy C Means Clustering.
Copyright Form