SnapPatch is an AI-driven web application that takes images of wounds, segments them (currently using a placeholder model), and generates 3D STL files of wound patches. Built for rapid prototyping with future plans to integrate trained ML models and 3D printing hardware.
Transform medical wound treatment by providing:
- Instant STL patch generation from wound images
- Seamless integration with future 3D printers
- Custom-fit, on-demand wound healing solutions
| Layer | Tools Used |
|---|---|
| Frontend | React, Tailwind CSS |
| Backend | FastAPI, Python |
| ML Model | TensorFlow (planned) |
| Image Proc. | OpenCV (planned) |
| 3D Output | STL Generator (custom module) |
| Deployment | Vercel (Frontend), Render (API) |
SnapPatch/
├── frontend/ # React UI for image upload & result view
├── backend/ # FastAPI server
│ └── ml/ # Placeholder ML pipeline
├── stl_generator/ # Converts image masks to STL files
└── docker-compose.yml # (planned for full deployment)
- 🖼️ Upload wound image via frontend.
- 🧪 Backend runs placeholder segmentation (dummy mask).
- 📐 Dummy mask fed to STL generator.
- 📁 STL file returned to user for download.
- Set up frontend & backend scaffolding
- Implement dummy image mask system
- Integrate STL generator
- Collect wound image dataset
- Train TensorFlow-based wound segmentation model
- Real-time patch fitting
- Connect to 3D printer APIs
- Precision medicine with custom healing patches
- Medical training simulators with wound models
- Global health deployment in remote clinics
We’re in early-stage dev! To contribute:
- Fork this repo
- Create a feature branch
- Commit and push changes
- Open a PR
MIT License
Muhammad Rashid
GitHub: @muhammadrashid4587
SnapPatch: Wounds don’t wait. Neither should healing.