MedVision: Medical Image Analysis Platform
📑 Paper | 🌐 Project Page | 💾 Released Resources | 📦 Repo
This is the resource page for our MedVision synthesized dataset collection on Huggingface.
Dataset
| Dataset | Link |
|---|---|
| MedVision-ChestX-Synth | 🤗 |
Please also check the raw data after our processing if you are interested: toolevalxm/MedVision-ChestX-Raw.
Models Used for Synthesis
| Base Model | Task |
|---|---|
| SDXL-Medical | Image augmentation and generation |
| BiomedCLIP | Feature extraction |
Introduction
Medical imaging datasets often suffer from class imbalance and limited samples for rare pathologies. Our approach leverages state-of-the-art diffusion models, specifically SDXL-Medical (based on stabilityai/stable-diffusion-xl-base-1.0), to generate realistic synthetic chest X-ray images for training more robust diagnostic AI systems.
*Due to ethical and compliance requirements, we only release the synthesized subset of our full augmentation pipeline.
License
The license for this dataset is OpenRAIL++.