---
license: apache-2.0
library_name: diffusers
pipeline_tag: image-to-image
tags:
  - zoomldm
  - histopathology
  - brca
  - latent-diffusion
  - custom-pipeline
  - arxiv:2411.16969
widget:
  - src: demo_images/input.jpeg
    prompt: BRCA sample conditioned on demo SSL feature (mag=0)
    output:
      url: demo_images/output.jpeg
---

> [!WARNING] we do not have a full checkpoint conversion validation, if you encounter pipeline loading failure and unsidered output, please contact me via bili_sakura@zju.edu.cn

# BiliSakura/ZoomLDM-brca

Diffusers-format **BRCA** variant of ZoomLDM with a bundled custom pipeline and local `ldm` modules.

## Model Description

- **Architecture:** ZoomLDM latent diffusion pipeline (`UNet + VAE + conditioning encoder`)
- **Domain:** Histopathology (BRCA)
- **Conditioning:** UNI-style SSL feature maps + magnification level (`0..4`)
- **Format:** Self-contained local folder for `DiffusionPipeline.from_pretrained(...)`

## Intended Use

Use this model for conditional multi-scale BRCA patch generation when you have compatible pre-extracted SSL features.

## Out-of-Scope Use

- Not intended for diagnosis, treatment planning, or other clinical decisions.
- Not a general-purpose text-to-image model.
- Not validated for data outside the expected acquisition/distribution range.

## Files

- `unet/`, `vae/`, `conditioning_encoder/`, `scheduler/`
- `model_index.json`
- `pipeline_zoomldm.py`
- `ldm/` (bundled dependency modules)

## Usage

```python
import torch
from diffusers import DiffusionPipeline

pipe = DiffusionPipeline.from_pretrained(
    "BiliSakura/ZoomLDM-brca",
    custom_pipeline="pipeline_zoomldm.py",
    trust_remote_code=True,
).to("cuda")

out = pipe(
    ssl_features=ssl_feat_tensor.to("cuda"),      # BRCA UNI-style SSL embeddings
    magnification=torch.tensor([0]).to("cuda"),   # 0..4
    num_inference_steps=50,
    guidance_scale=2.0,
)
images = out.images
```

## Demo Generation (dataset-backed)

This repo includes `run_demo_inference.py`, which uses local repo assets only:

- image: `demo_images/input.jpeg`
- SSL feature: `demo_data/0_ssl_feat.npy`
- magnification label: `0`

Run:

```bash
python run_demo_inference.py
```

## Limitations

- Requires correctly precomputed BRCA conditioning features.
- Magnification conditioning must match expected integer codes.
- Generated content may reflect biases and artifacts from training data.

## Citation

```bibtex
@InProceedings{Yellapragada_2025_CVPR,
  author = {Yellapragada, Srikar and Graikos, Alexandros and Triaridis, Kostas and Prasanna, Prateek and Gupta, Rajarsi and Saltz, Joel and Samaras, Dimitris},
  title = {ZoomLDM: Latent Diffusion Model for Multi-scale Image Generation},
  booktitle = {Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR)},
  month = {June},
  year = {2025},
  pages = {23453-23463}
}
```