Spaces:
Running
Running
| title: AI Liver Segmentation | |
| emoji: 🫀 | |
| colorFrom: red | |
| colorTo: pink | |
| sdk: gradio | |
| sdk_version: 4.8.0 | |
| app_file: app.py | |
| pinned: false | |
| license: mit | |
| # 🫀 AI-Powered Liver Segmentation | |
| Automatic liver segmentation from CT scans using SwinUNETR (Vision Transformer). | |
| ## 📊 Model Performance | |
| - **Architecture**: SwinUNETR with gradient checkpointing | |
| - **Dataset**: Medical Segmentation Decathlon (Task03_Liver) | |
| - **Dice Score**: 95.5% ± 2.0% on test set | |
| ## 🚀 Features | |
| - Upload CT scans in NIfTI format | |
| - Real-time liver segmentation | |
| - Interactive slice visualization | |
| - Download segmentation masks | |
| ## 🛠️ Tech Stack | |
| - PyTorch | |
| - MONAI | |
| - Gradio | |
| - Hugging Face Spaces | |
| ## ⚠️ Disclaimer | |
| This is a research prototype and should not be used for clinical diagnosis. | |