Related papers: Curia-MAE: Multi-Modal Multi-Anatomy MAE Pre-Training for 3D Medical Image Segmentation

Curia-MAE: Multi-Modal Multi-Anatomy MAE Pre-Training for 3D Medical Image Segmentation

URL: http://arxiv.org/abs/2608.05844v2
Date: Fri, 07 Aug 2026 15:11:56 GMT
Title: Curia-MAE: Multi-Modal Multi-Anatomy MAE Pre-Training for 3D Medical Image Segmentation
Abstract summary: We propose Curia-MAE, a multi-modal, multi-anatomy MAE model pre-trained on 300,000 CT and MRI images.<n>On eight anatomy- and lesion-focused segmentation benchmarks, Curia-MAE improves frozen-encoder performance over a strong MAE baseline.<n>These results indicate that a single frozen encoder can be reused across diverse segmentation tasks.
Score: 5.970778891214216
License: http://creativecommons.org/licenses/by-nc-nd/4.0/
Abstract: Radiology foundation models learn transferable representations that can be adapted to new tasks by training only small layers on top of a frozen encoder. Dense prediction tasks such as 3D segmentation are, however, underrepresented in their evaluation, and, with the encoder kept frozen, pre-trained models still fall short of nnU-Net, the state-of-the-art reference trained from scratch. To close this gap we extend convolutional MAE pre-training with a robust reconstruction objective, a feature regularizer, and a local-global similarity objective. Using this method, we propose Curia-MAE, a multi-modal, multi-anatomy MAE model pre-trained on 300,000 CT and MRI images covering a large number of anatomical sites. On eight anatomy- and lesion-focused segmentation benchmarks, Curia-MAE improves frozen-encoder performance over a strong MAE baseline, while remaining competitive under full finetuning and superior on lesion tasks, where labeled data is scarce. These results indicate that a single frozen encoder can be reused across diverse segmentation tasks, reducing the cost of adapting and deploying such models in clinical workflows. Curia-MAE pre-trained model weights are made publicly available at https://huggingface.co/raidium/Curia-MAE.

Related papers

MAE-Based Self-Supervised Pretraining for Data-Efficient Medical Image Segmentation Using nnFormer [0.5437050212139087]
Transformer architectures, including nnFormer, have demonstrated promising results in volumetric medical image segmentation.<n>These models need large quantities of labeled training data and are also likely to overfit and become training unstable.<n>This is a serious practical problem because it is not only time-consuming but also expensive to obtain medical images that are annotated by experts.<n>We have solved these drawbacks by advancing the efficiency of the nnFormer with a self-supervised pretraining framework.
arXiv Detail & Related papers (2026-04-22T11:29:08Z)
MedPruner: Training-Free Hierarchical Token Pruning for Efficient 3D Medical Image Understanding in Vision-Language Models [59.180043227905294]
We propose MedPruner, a training-free and model-agnostic hierarchical token pruning framework for efficient 3D medical image understanding.<n>We show that MedPruner enables models such as MedGemma to maintain or even exceed their original performance while retaining fewer than 5% of visual tokens.
arXiv Detail & Related papers (2026-03-12T07:37:00Z)
Few-Shot Deployment of Pretrained MRI Transformers in Brain Imaging Tasks [2.982793366290863]
We propose a framework for the few-shot deployment of pretrained MRI transformers in diverse brain imaging tasks.<n>By utilizing the Masked Autoencoder (MAE) pretraining strategy, we obtain highly transferable latent representations that generalize well across tasks and datasets.
arXiv Detail & Related papers (2025-08-07T18:53:28Z)
SegBook: A Simple Baseline and Cookbook for Volumetric Medical Image Segmentation [20.026663367994356]
Large amounts of full-body CT images provide the opportunity to pre-train powerful models. It remains unclear in which conditions these pre-trained models can be transferred to various downstream medical segmentation tasks. We collected 87 public datasets varying in modality, target, and sample size to evaluate the transfer ability of full-body CT pre-trained models.
arXiv Detail & Related papers (2024-11-21T19:00:01Z)
ShapeMamba-EM: Fine-Tuning Foundation Model with Local Shape Descriptors and Mamba Blocks for 3D EM Image Segmentation [49.42525661521625]
This paper presents ShapeMamba-EM, a specialized fine-tuning method for 3D EM segmentation. It is tested over a wide range of EM images, covering five segmentation tasks and 10 datasets.
arXiv Detail & Related papers (2024-08-26T08:59:22Z)
MUSCLE: Multi-task Self-supervised Continual Learning to Pre-train Deep Models for X-ray Images of Multiple Body Parts [63.30352394004674]
Multi-task Self-super-vised Continual Learning (MUSCLE) is a novel self-supervised pre-training pipeline for medical imaging tasks. MUSCLE aggregates X-rays collected from multiple body parts for representation learning, and adopts a well-designed continual learning procedure. We evaluate MUSCLE using 9 real-world X-ray datasets with various tasks, including pneumonia classification, skeletal abnormality classification, lung segmentation, and tuberculosis (TB) detection.
arXiv Detail & Related papers (2023-10-03T12:19:19Z)
Disruptive Autoencoders: Leveraging Low-level features for 3D Medical Image Pre-training [51.16994853817024]
This work focuses on designing an effective pre-training framework for 3D radiology images. We introduce Disruptive Autoencoders, a pre-training framework that attempts to reconstruct the original image from disruptions created by a combination of local masking and low-level perturbations. The proposed pre-training framework is tested across multiple downstream tasks and achieves state-of-the-art performance.
arXiv Detail & Related papers (2023-07-31T17:59:42Z)
MIS-FM: 3D Medical Image Segmentation using Foundation Models Pretrained on a Large-Scale Unannotated Dataset [14.823114726604853]
We propose a novel self-supervised learning strategy named Volume Fusion (VF) for pretraining 3D segmentation models. VF forces the model to predict the fusion coefficient of each voxel, which is formulated as a self-supervised segmentation task without manual annotations. experiments with different downstream segmentation targets including head and neck organs, thoracic/abdominal organs showed that our pretrained model largely outperformed training from scratch.
arXiv Detail & Related papers (2023-06-29T13:22:13Z)
LVM-Med: Learning Large-Scale Self-Supervised Vision Models for Medical Imaging via Second-order Graph Matching [59.01894976615714]
We introduce LVM-Med, the first family of deep networks trained on large-scale medical datasets. We have collected approximately 1.3 million medical images from 55 publicly available datasets. LVM-Med empirically outperforms a number of state-of-the-art supervised, self-supervised, and foundation models.
arXiv Detail & Related papers (2023-06-20T22:21:34Z)
Attentive Symmetric Autoencoder for Brain MRI Segmentation [56.02577247523737]
We propose a novel Attentive Symmetric Auto-encoder based on Vision Transformer (ViT) for 3D brain MRI segmentation tasks. In the pre-training stage, the proposed auto-encoder pays more attention to reconstruct the informative patches according to the gradient metrics. Experimental results show that our proposed attentive symmetric auto-encoder outperforms the state-of-the-art self-supervised learning methods and medical image segmentation models.
arXiv Detail & Related papers (2022-09-19T09:43:19Z)
A Multi-Task Cross-Task Learning Architecture for Ad-hoc Uncertainty Estimation in 3D Cardiac MRI Image Segmentation [0.0]
We present a Multi-task Cross-task learning consistency approach to enforce the correlation between the pixel-level (segmentation) and the geometric-level (distance map) tasks. Our study further showcases the potential of our model to flag low-quality segmentation from a given model.
arXiv Detail & Related papers (2021-09-16T03:53:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.