YOLOv10-Powered Brain Tumor Detection with AI-Based Insights

Abstract

This project presents an AI-powered Brain Tumor Detection and Classification System integrated with an AI Chatbot for medical query processing. The architecture consists of a backend, frontend, and user interface. The backend leverages a YOLOv10-based deep learning model for brain tumor detection, comprising image processing and tumor detection modules. Additionally, an OpenAI multimodal LLM is incorporated to process medical queries and generate captions. A SQLite database is used for data storage and retrieval. The frontend provides API endpoints for image prediction (/predict) and chatbot interactions (/chatbot), serving as a bridge between the backend and user interface. The user interface allows users to upload medical images, view detection results, and interact with the AI chatbot for medical-related inquiries. This architecture enables an automated, efficient, and user-friendly system for early brain tumor detection and medical assistance, improving accessibility and accuracy in diagnosis