Effective Feature Learning for 3D Medical Registration via Domain-Specialized DINO Pretraining

• URL: http://arxiv.org/abs/2603.14086v1
• Date: Sat, 14 Mar 2026 19:14:18 GMT
• Title: Effective Feature Learning for 3D Medical Registration via Domain-Specialized DINO Pretraining
• Abstract summary: In this work, we investigate DINO-style self-supervised pretraining directly on 3D medical imaging data.<n>Our domain-specialized pretraining outperforms the DINOv2 model trained on a large-scale collection of natural images.
• Score: 3.1852844552413644
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Medical image registration is a critical component of clinical imaging workflows, enabling accurate longitudinal assessment, multi-modal data fusion, and image-guided interventions. Intensity-based approaches often struggle with interscanner variability and complex anatomical deformations, whereas feature-based methods offer improved robustness by leveraging semantically informed representations. In this work, we investigate DINO-style self-supervised pretraining directly on 3D medical imaging data, aiming to learn dense volumetric features well suited for deformable registration. We assess the resulting representations on challenging interpatient abdominal registration task across both MRI and CT modalities. Our domain-specialized pretraining outperforms the DINOv2 model trained on a large-scale collection of natural images, while requiring substantially lower computational resources at inference time. Moreover, it surpasses established registration models under out-of-domain evaluation, demonstrating the value of task-agnostic yet medical imaging-focused pretraining for robust and efficient 3D image registration.

Related papers

• CoRe: Joint Optimization with Contrastive Learning for Medical Image Registration [2.6926304126225893]
  We propose a novel framework that integrates contrastive learning directly into the registration model.<n>By jointly optimizing the contrastive and registration objectives, we ensure that the learned representations are not only informative but also suitable for the registration task.
  arXiv  Detail & Related papers  (2026-03-24T20:08:34Z)
• Does DINOv3 Set a New Medical Vision Standard? [67.33543059306938]
  This report investigates whether DINOv3 can serve as a powerful unified encoder for medical vision tasks without domain-specific pre-training.<n>We benchmark DINOv3 across common medical vision tasks, including 2D/3D classification and segmentation.<n>Remarkably, it can even outperform medical-specific foundation models like BiomedCLIP and CT-Net on several tasks.
  arXiv  Detail & Related papers  (2025-09-08T09:28:57Z)
• IMPACT: A Generic Semantic Loss for Multimodal Medical Image Registration [0.46904601975060667]
  IMPACT (Image Metric with Pretrained model-Agnostic Comparison for Transmodality registration) is a novel similarity metric designed for robust multimodal image registration.<n>It defines a semantic similarity measure based on the comparison of deep features extracted from large-scale pretrained segmentation models.<n>It was evaluated on five challenging 3D registration tasks involving thoracic CT/CBCT and pelvic MR/CT datasets.
  arXiv  Detail & Related papers  (2025-03-31T14:08:21Z)
• Dual-Attention Frequency Fusion at Multi-Scale for Joint Segmentation and Deformable Medical Image Registration [2.6089354079273512]
  We propose a multi-task learning framework based on dual attention frequency fusion (DAFF-Net) DAFF-Net simultaneously achieves the segmentation masks and dense deformation fields in a single-step estimation. Experiments on three public 3D brain magnetic resonance imaging (MRI) datasets demonstrate that the proposed DAFF-Net and its unsupervised variant outperform state-of-the-art registration methods.
  arXiv  Detail & Related papers  (2024-09-29T11:11:04Z)
• Autoregressive Sequence Modeling for 3D Medical Image Representation [48.706230961589924]
  We introduce a pioneering method for learning 3D medical image representations through an autoregressive sequence pre-training framework.<n>Our approach various 3D medical images based on spatial, contrast, and semantic correlations, treating them as interconnected visual tokens within a token sequence.
  arXiv  Detail & Related papers  (2024-09-13T10:19:10Z)
• 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)
• 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 Deep Discontinuity-Preserving Image Registration Network [73.03885837923599]
  Most deep learning-based registration methods assume that the desired deformation fields are globally smooth and continuous. We propose a weakly-supervised Deep Discontinuity-preserving Image Registration network (DDIR) to obtain better registration performance and realistic deformation fields. We demonstrate that our method achieves significant improvements in registration accuracy and predicts more realistic deformations, in registration experiments on cardiac magnetic resonance (MR) images.
  arXiv  Detail & Related papers  (2021-07-09T13:35:59Z)
• A Meta-Learning Approach for Medical Image Registration [6.518615946009265]
  We propose a novel unsupervised registration model which is integrated with a gradient-based meta learning framework. In our experiments, the proposed model obtained significantly improved performance in terms of accuracy and training time.
  arXiv  Detail & Related papers  (2021-04-21T10:27:05Z)
• Learning Deformable Registration of Medical Images with Anatomical Constraints [4.397224870979238]
  Deformable image registration is a fundamental problem in the field of medical image analysis. We learn global non-linear representations of image anatomy using segmentation masks, and employ them to constraint the registration process. Our experiments show that the proposed anatomically constrained registration model produces more realistic and accurate results than state-of-the-art methods.
  arXiv  Detail & Related papers  (2020-01-20T17:44:11Z)

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

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.