https://github.com/idso-fa1-pathology/VitaminPScope
VitaminPScope – VitaminPScope — AI Digital Pathology Viewer for Whole-Slide Images (WSI)
https://github.com/idso-fa1-pathology/VitaminPScope
cellsegmentation digital-pathology nuclei-segmentation pathology pathology-platform
Last synced: 25 days ago
JSON representation
VitaminPScope – VitaminPScope — AI Digital Pathology Viewer for Whole-Slide Images (WSI)
- Host: GitHub
- URL: https://github.com/idso-fa1-pathology/VitaminPScope
- Owner: idso-fa1-pathology
- License: mit
- Created: 2026-03-06T18:50:21.000Z (3 months ago)
- Default Branch: main
- Last Pushed: 2026-04-20T20:29:48.000Z (about 2 months ago)
- Last Synced: 2026-04-20T22:32:01.295Z (about 2 months ago)
- Topics: cellsegmentation, digital-pathology, nuclei-segmentation, pathology, pathology-platform
- Language: Python
- Homepage:
- Size: 202 MB
- Stars: 2
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
- Security: SECURITY.md
Awesome Lists containing this project
- awesome-medical-ai - VitaminPScope - fa1-pathology/VitaminPScope?style=flat-square) | ⭐⭐ B- | Local-first digital pathology platform for whole-slide and multi-channel microscopy viewing, with Docker quickstart, OpenSeadragon-based web UI, and AI whole-cell segmentation via the bundled VitaminP service. | (Medical Imaging & Radiology)
README
# VitaminPScope — AI Digital Pathology Viewer for Whole-Slide Images (WSI).
**VitaminPScope** is an AI-powered digital pathology platform for viewing whole-slide images (WSI) and performing whole-cell segmentation using deep learning.
It supports SVS, OME-TIFF, and multi-channel microscopy, and runs locally with Docker in a single command.
🚀 **Run in one command — no setup, no configuration**
[🐳 Docker](#-quick-start) • [🐛 Issues](https://github.com/idso-fa1-pathology/VitaminPScope/issues)
---
---
## ✨ Overview
VitaminPScope is a production-ready, web-based platform for:
- Whole-slide image (WSI) viewing
- Multi-channel microscopy visualization
- Slide comparison and annotation
- AI-powered whole-cell segmentation with **VitaminP**
Built with **React + FastAPI + OpenSeadragon + Viv** — optimized for real-world deployment.
**VitaminPScope is an open-source alternative to commercial digital pathology platforms (e.g., QuPath, HALO) with built-in AI segmentation**
---
## 💡 Why VitaminPScope
- 🧠 Built-in AI — no external pipelines required
- ⚡ Runs locally with Docker in minutes
- 🔬 Designed for real pathology workflows
- 📊 Scales to gigapixel whole-slide images
---
## 🎬 Demo
---
## 🔍 Keywords
digital pathology, WSI viewer, whole-slide image viewer, pathology AI, cell segmentation, histopathology, H&E analysis, multiplex imaging, computational pathology, FastAPI, React viewer
---
# 🚀 Quick Start
### 1. Clone repository
```bash
git clone https://github.com/idso-fa1-pathology/VitaminPScope.git
```
### 2. Move into project
```bash
cd VitaminPScope
```
### 3. Launch the app
## ⚙️ GPU Support (Optional)
VitaminPScope runs out-of-the-box on both CPU and GPU environments.
### 🖥️ Run on CPU (default)
```bash
docker compose -f docker-compose.public.yml up -d
```
---
### 🚀 Run with GPU (Linux + NVIDIA only)
```bash
docker compose -f docker-compose.public.yml --profile gpu up -d
```
---
## ⚡ One-line setup
### 🖥️ CPU (default)
```bash
git clone https://github.com/idso-fa1-pathology/VitaminPScope.git && cd VitaminPScope && docker compose -f docker-compose.public.yml up -d
```
### 🚀 GPU (Linux + NVIDIA)
```bash
git clone https://github.com/idso-fa1-pathology/VitaminPScope.git && cd VitaminPScope && docker compose -f docker-compose.public.yml --profile gpu up -d
```
---
## 🌐 Open the app
Once running, open in your browser:
👉 **http://localhost:3000**
---
## 📂 Data
Place your slides in:
```
./data/
```
**Supported formats:**
- SVS
- NDPI
- TIFF / OME-TIFF
---
## 🤖 Powered by VitaminP
This platform integrates **VitaminP**, a state-of-the-art AI model for whole-cell segmentation.
➡️ https://github.com/idso-fa1-pathology/vitaminp
---
## 🧪 Use Cases
- Histopathology research (H&E analysis)
- Spatial transcriptomics validation
- Multiplex imaging visualization
- AI-assisted cell segmentation workflows
- Digital pathology education and demos
---
## 🏗 Architecture
- **Frontend:** React + OpenSeadragon (WSI rendering)
- **Backend:** FastAPI (API + orchestration)
- **AI Service:** VitaminP inference engine
- **Deployment:** Docker Compose (multi-service)
---
## 🔌 Extensibility
VitaminPScope can be extended via API or integrated into custom pipelines for automated pathology workflows.
---
## 🧠 AI Model
The **VitaminP model is already included** inside the Docker image.
✅ No downloads
✅ No configuration
✅ Ready out of the box
---
## 📄 Associated Paper
VitaminPScope is powered by **VitaminP**, a cross-modal learning framework for whole-cell segmentation directly from routine H&E histology.
📄 Read the preprint: https://arxiv.org/abs/2604.23799
If you use VitaminPScope in your research, please consider citing this work.
---
## 📄 License
MIT License