---
datasets:
- AnonRes/OpenMind
license: cc-by-4.0
pipeline_tag: image-segmentation
tags:
- medical
---

# OpenMind Benchmark 3D SSL Models (Primus)

This repository hosts pre-trained checkpoints for the **Primus** and **ResEnc-L** backbones from the OpenMind benchmark.

> **Papers**: 
> - [Primus: Enforcing Attention Usage for 3D Medical Image Segmentation](https://huggingface.co/papers/2503.01835)
> - [An OpenMind for 3D medical vision self-supervised learning](https://huggingface.co/papers/2412.17041)
>
> **Code**: [MIC-DKFZ/nnUNet](https://github.com/MIC-DKFZ/nnUNet)  
> **Pre-training framework**: [MIC-DKFZ/nnssl](https://github.com/MIC-DKFZ/nnssl)  
> **Dataset**: [AnonRes/OpenMind](https://huggingface.co/datasets/AnonRes/OpenMind)  

---

![OpenMind](https://huggingface.co/datasets/AnonRes/OpenMind/resolve/main/assets/OpenMindDataset.png)

## Overview

This repository hosts pre-trained checkpoints from the **OpenMind** benchmark, featuring the **Primus** architecture. Primus is a Transformer-centric segmentation architecture that leverages high-resolution tokens and advances in positional embeddings to achieve state-of-the-art results in 3D medical image segmentation.

Each model was pre-trained using a specific self-supervised learning (SSL) method on the [OpenMind Dataset](https://huggingface.co/datasets/AnonRes/OpenMind), a standardized collection of public brain MRI datasets.

**Note**: These models are intended to be used as weights for downstream fine-tuning for **segmentation** tasks using the [nnU-Net framework](https://github.com/MIC-DKFZ/nnUNet).

---

## Model Variants

We release SSL checkpoints for two backbone architectures:

- **Primus-M**: A transformer-based encoder introduced in the [Primus paper](https://huggingface.co/papers/2503.01835).
- **ResEnc-L**: A CNN-based encoder.

Each encoder has been pre-trained using one of the following SSL techniques:

| Method        | Description |
|---------------|-------------|
| [Volume Contrastive (VoCo)](https://arxiv.org/abs/2402.17300) | Contrastive pretraining method for 3D volumes |
| [VolumeFusion (VF)](https://arxiv.org/abs/2306.16925)         | Spatial volume fusion-based segmentation SSL method |
| [Models Genesis (MG)](https://www.sciencedirect.com/science/article/pii/S1361841520302048) | Reconstruction and denoising based pretraining method |
| [Masked Autoencoders (MAE)](https://arxiv.org/abs/2111.06377) | Default reconstruction based pretraining method |
| [Spark 3D (S3D)](https://arxiv.org/abs/2410.23132) | Sparse reconstruction based pretraining method (CNN only) |
| [SimMIM](https://arxiv.org/abs/2111.09886) | Simple masked reconstruction based pretraining method (TR only) |
| [SwinUNETR SSL](https://arxiv.org/abs/2111.14791) | Rotation, Contrastive and Reconstruction based pre-training method. |
| [SimCLR](https://arxiv.org/abs/2002.05709) | Transfer of 2D Contrastive learning baseline method to 3D |

## Citation

```bibtex
@article{wald2025primus,
  title={Primus: Enforcing Attention Usage for 3D Medical Image Segmentation},
  author={Wald, Tassilo and Roy, Saikat and Isensee, Fabian and Ulrich, Constantin and Ziegler, Sebastian and Trofimova, Dasha and Stock, Raphael and Baumgartner, Michael and Köhler, Gregor and Maier-Hein, Klaus},
  journal={arXiv preprint arXiv:2503.01835},
  year={2025}
}

@article{wald2024openmind,
    title={An OpenMind for 3D medical vision self-supervised learning},
    author={Wald, Tassilo and Ulrich, Constantin and Suprijadi, Jonathan and Ziegler, Sebastian and Nohel, Michal and Peretzke, Robin eand Köhler, Gregor and Maier-Hein, Klaus H},
    journal={arXiv preprint arXiv:2412.17041},
    year={2024}
}
```