Brain Tumor Segmentation & Classification with U-Net + CNN

This project provides an end-to-end medical imaging pipeline for analyzing brain MRI scans. It uses:

U-Net for tumor segmentation (detecting and localizing tumors in MRI images).

CNN classifier for tumor type classification (e.g., glioma, meningioma, pituitary, etc.).

Gemini API (Google GenAI) to generate an AI-assisted medical report summarizing findings in natural language.

Features

Segmentation: Pixel-level tumor detection using U-Net.

Classification: CNN model predicts tumor type from segmented regions.

Medical Report Generation: Automatically generates a clinician-style report from results using Gemini.

Interactive App: Streamlit-based interface for uploading MRIs, visualizing masks, classification results, and reports.