ResEncL-OpenMind-MAE Open-Source 3D Medical Image Model - Supports Self-Supervised Learning and Provides Multiple Pre-trained Checkpoints

Resencl OpenMind MAE

Developed by AnonRes

The first comprehensive benchmark study model for self-supervised learning on 3D medical imaging data, providing multiple pre-trained checkpoints

3D Vision #3D Medical Imaging #Self-supervised Learning #Brain MRI

Downloads 20

Release Time : 5/6/2025

Model Overview

This model serves as a benchmark for self-supervised learning on 3D medical imaging (particularly brain MRI), supporting various pre-training methods and architectures, suitable for medical image analysis tasks

Model Features

Comprehensive 3D Medical Imaging Benchmark

The first comprehensive benchmark study for self-supervised learning on 3D medical imaging data

Multiple SSL Method Support

Supports various self-supervised learning methods including VoCo, VF, MG, MAE, and S3D

Dual Architecture Design

Provides both CNN-based and Transformer-based backbone architectures

Standardized Dataset

Trained on the large-scale, standardized public brain MRI dataset OpenMind

Model Capabilities

3D Medical Imaging Feature Extraction

Brain MRI Analysis

Medical Image Segmentation Pre-training

Self-supervised Learning

Use Cases

Medical Image Analysis

Brain MRI Segmentation

Used for segmentation tasks of brain MRI images

Requires fine-tuning before use

Medical Image Classification

Can be used for pre-training in medical image classification tasks

Requires fine-tuning before use

Medical Research

Self-supervised Learning Benchmark

Serves as a benchmark model for self-supervised learning research on 3D medical imaging

🚀 OpenMind Benchmark 3D SSL Models

This repository provides pre - trained checkpoints from the OpenMind benchmark for 3D medical vision self - supervised learning.

🚀 Quick Start

This section offers an overview of the key information about the models in this repository:

Model from the paper: An OpenMind for 3D medical vision self - supervised learning
Pre - training codebase used to create checkpoint: [MIC - DKFZ/nnssl](https://github.com/MIC - DKFZ/nnssl)
Dataset: AnonRes/OpenMind
Downstream (segmentation) fine - tuning: TaWald/nnUNet

OpenMind

📚 Documentation

Overview

This repository hosts pre - trained checkpoints from the OpenMind benchmark: 📄 An OpenMind for 3D medical vision self - supervised learning (Wald, T., Ulrich, C., Suprijadi, J., Ziegler, S., Nohel, M., Peretzke, R., ... & Maier - Hein, K. H. (2024).)
(arXiv:2412.17041) — the first extensive benchmark study for self - supervised learning (SSL) on 3D medical imaging data.

Each model was pre - trained using a particular SSL method on the OpenMind Dataset, a large - scale, standardized collection of public brain MRI datasets.

⚠️ Important Note

These models are not recommended to be used as - is for feature extraction. Instead, we recommend using the downstream fine - tuning frameworks for segmentation and classification adaptation, available in the adaptation repository.

💡 Usage Tip

While manual download is possible, we recommend using the auto - download feature of the fine - tuning repository by providing the repository URL on Hugging Face instead of a local checkpoint path.

Model Variants

We release SSL checkpoints for two backbone architectures:

ResEnc - L: A CNN - based encoder [a, b]
Primus - M: A transformer - based encoder [Primus paper]

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

Property	Details
Volume Contrastive (VoCo)	Contrastive pretraining method for 3D volumes
VolumeFusion (VF)	Spatial volume fusion - based segmentation SSL method
Models Genesis (MG)	Reconstruction and denoising based pretraining method
Masked Autoencoders (MAE)	Default reconstruction based pretraining method
Spark 3D (S3D)	Sparse reconstruction based pretraining method (CNN only)
SimMIM	Simple masked reconstruction based pretraining method (TR only)
SwinUNETR SSL	Rotation, Contrastive and Reconstruction based pre - training method.
SimCLR	Transfer of 2D Contrastive learning baseline method to 3D

📄 License

This project is licensed under the CC - BY - 4.0 license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご