---
license: mit
language:
- en
base_model:
- stabilityai/stable-diffusion-2-1
pipeline_tag: text-to-image
tags:
- medical
- chest-X-ray

extra_gated_prompt: "By agreeing you confirm that you are credentialed and allowed to use [MIMIC-CXR](https://physionet.org/content/mimic-cxr/2.0.0/), and will only share access to the model with people that are also credentialed for MIMIC-CXR. 
Relevant data use agreement: https://physionet.org/content/mimic-cxr/view-dua/2.0.0/"
extra_gated_fields:
  Name: text
  E-mail: text
  Country: country
  Organization or Affiliation: text
  I want to use this model for:
    type: select
    options: 
      - Research
      - Education
      - label: Other
        value: other
  I agree to use this model for non-commercial use ONLY: checkbox
---
Research paper: https://arxiv.org/abs/2508.16783

## 🧨Inference with diffusers

```python
import torch
from diffusers import DiffusionPipeline

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

pipe = DiffusionPipeline.from_pretrained("stanfordmimi/RoentGen-v2")
pipe = pipe.to(device)

prompt = "50 year old female. Normal chest radiograph."
image = pipe(prompt).images[0]
```

More info and instructions for use on [GitHub](https://github.com/StanfordMIMI/RoentGen-v2).

## 🩻 Synthetic CXR Dataset
565k synthetic chest radiographs and associated text prompts: [`stanfordmimi/RoentGen-v2-synthetic-dataset`](https://huggingface.co/datasets/stanfordmimi/RoentGen-v2-synthetic-dataset)

![Visuals](visual_examples.png)

Important: The generated images are for research and educational purposes only and cannot replace real chest x-rays for medical diagnosis.

```bibtex
@misc{moroianu2025improvingperformancerobustnessfairness,
      title={Improving Performance, Robustness, and Fairness of Radiographic AI Models with Finely-Controllable Synthetic Data}, 
      author={Stefania L. Moroianu and Christian Bluethgen and Pierre Chambon and Mehdi Cherti and Jean-Benoit Delbrouck and Magdalini Paschali and Brandon Price and Judy Gichoya and Jenia Jitsev and Curtis P. Langlotz and Akshay S. Chaudhari},
      year={2025},
      eprint={2508.16783},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2508.16783}, 
}
```