Explainable Tuberculosis Detection from Chest X-Rays using Deep CNN and LLaMA-3 | Data Science

Description Project Title: Explainable Tuberculosis Diagnosis from Chest X-Rays using Deep CNNs and LLaMA-3 Submitted By: Suraj Patil (Roll No: 69) Tejas Musale (Roll No: 70) Uday Patil (Roll No: 71) Institution: GHRCEM, Pune (Savitribai Phule Pune University) Guided By: Prof. Vaishali Kapure & Prof. Sunita Wani Project Overview: This project presents an AI-powered system for early and accurate detection of Tuberculosis (TB) from chest X-ray images using Deep Convolutional Neural Networks (CNNs). The system also integrates LLaMA-3 based chatbot to provide explainable medical insights, helping users understand predictions in a simple way. Features: TB Detection using Deep Learning (CNN) Chest X-ray image classification Explainable AI results (Grad-CAM visualization) AI Chatbot using LLaMA-3 / Groq API Medical guidance and precaution suggestions Full-stack web application Live Project: [https://ai-powered-tb-detection.vercel.app/](https://ai-powered-tb-detection.vercel.app/) Tech Stack: React + Vite (Frontend) Flask (ML Backend) TensorFlow / PyTorch (Deep Learning Models) Node.js + Express (Chatbot Backend) Groq API (LLM) Objective: To assist healthcare professionals in early detection of Tuberculosis using AI and improve diagnostic accuracy with explainable deep learning models.