license: mit
Med-VAE
Med-VAE is a family of six large-scale, generalizable 2D and 3D variational autoencoders (VAEs) designed for medical imaging. It is trained on over one million medical images across multiple anatomical regions and modalities. Med-VAE autoencoders encode medical images as downsized latent representations and decode latent representations back to high-resolution images. Across diverse tasks obtained from 20 medical image datasets, we demonstrate that utilizing MedVAE latent representations in place of high-resolution images when training downstream models can lead to efficiency benefits (up to 70x improvement in throughput) while simultaneously preserving clinically-relevant features.
Model Description
| Total Compression Factor | Channels | Dimensions | Modalities | Anatomies | Config File | Model File |
|---|---|---|---|---|---|---|
| 16 | 1 | 2D | X-ray | Chest, Breast (FFDM) | medvae_4x1.yaml | vae_4x_1c_2D.ckpt |
| 16 | 3 | 2D | X-ray | Chest, Breast (FFDM) | medvae_4x3.yaml | vae_4x_3c_2D.ckpt |
| 64 | 1 | 2D | X-ray | Chest, Breast (FFDM) | medvae_8x1.yaml | vae_8x_1c_2D.ckpt |
| 64 | 3 | 2D | X-ray | Chest, Breast (FFDM) | medvae_8x4.yaml | vae_8x_4c_2D.ckpt |
| 64 | 1 | 3D | MRI, CT | Whole-Body | medvae_4x1.yaml | vae_4x_1c_3D.ckpt |
| 512 | 1 | 3D | MRI, CT | Whole-Body | medvae_8x1.yaml | vae_8x_1c_3D.ckpt |
Note: Model weights and checkpoints are located in the model_weights folder.
Usage Instructions
Citation
If you use Med-VAE, please cite the original paper:
@article{varma2025medvae,
title = {Med-VAE: --},
author = {Maya Varma, Ashwin Kumar, Rogier van der Sluijs, Sophie Ostmeier, Louis Blankemeier, Pierre Chambon, Christian Bluethgen, Jip Prince, Curtis Langlotz, Akshay Chaudhari},
year = {2025},
publisher = {Arxiv},
journal = {Arvix},
howpublished = {TODO}
}
For questions, please place a Github Issues message.