license: cc-by-4.0
datasets:
- AnonRes/OpenMind
pipeline_tag: image-feature-extraction
tags:
- medical
OpenMind Benchmark 3D SSL Models
Model from the paper: An OpenMind for 3D medical vision self-supervised learning
Pre-training codebase used to create checkpoint: MIC-DKFZ/nnssl
Dataset: AnonRes/OpenMind
Downstream (segmentation) fine-tuning: TaWald/nnUNet
🔍 Overview
This repository hosts pre-trained checkpoints from the OpenMind benchmark:
📄 "An OpenMind for 3D medical vision self-supervised learning"
(arXiv:2412.17041) — the first extensive benchmark study for self-supervised learning (SSL) on 3D medical imaging data.
The models were pre-trained using various SSL methods on the OpenMind Dataset, a large-scale, standardized collection of public brain MRI datasets.
These models are not recommended to be used as-is. Instead we recommend using the downstream fine-tuning pipelines for segmentation and classification, available in the adaptation repository. While direct download is possible, we recommend using the auto-download of the respective fine-tuning repositories.
🧠 Model Variants
We release SSL checkpoints for two backbone architectures:
- ResEnc-L: A CNN-based encoder [link1, link2]
- Primus-M: A transformer-based encoder [Primus paper]
Each encoder has been pre-trained using the following SSL techniques:
| Method | Description |
|---|---|
| Volume Contrastive (VoCo) | Global contrastive learning in 3D volumes |
| VolumeFusion (VF) | Spatial fusion-based SSL |
| Models Genesis (MG) | Classic 3D self-reconstruction |
| Masked Autoencoders (MAE) | Patch masking and reconstruction |
| Spark 3D (S3D) | 3D adaptation of Spark framework |
| SimMIM | Simple masked reconstruction |
| SwinUNETR SSL | Transformer-based pre-training |
| SimCLR | Contrastive learning baseline |