# Pretraining Deformable Image Registration Networks with Random Images

arXiv: https://arxiv.org/abs/2505.24167 · HTML: https://mlsanity.com/paper/2505.24167/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 17 references that are in the map.
- First posted on arXiv 30 May 2025.

## Builds on (most cited first, 6 of 6)

- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [SynthMorph: learning contrast-invariant registration without acquired images](https://mlsanity.com/paper/2004.10282/) (2020)
- [A survey on deep learning in medical image registration: new technologies, uncertainty, evaluation metrics, and beyond](https://mlsanity.com/paper/2307.15615/) (2023)
- [On Finite Difference Jacobian Computation in Deformable Image Registration](https://mlsanity.com/paper/2212.06060/) (2022)

## Built on by (most cited first, 3 of 3)

- [Surrogate Supervision for Robust and Generalizable Deformable Image Registration](https://mlsanity.com/paper/2509.09869/) (2025)
- [MorphSeek: Fine-grained Latent Representation-Level Policy Optimization for Deformable Image Registration](https://mlsanity.com/paper/2511.17392/) (2025)
- [Domain-Shift Immunity in Deep Deformable Registration via Local Feature Representations](https://mlsanity.com/paper/2512.23142/) (2025)

## Abstract (arXiv)

Recent advances in deep learning-based medical image registration have shown that training deep neural networks~(DNNs) does not necessarily require medical images. Previous work showed that DNNs trained on randomly generated images with carefully designed noise and contrast properties can still generalize well to unseen medical data. Building on this insight, we propose using registration between random images as a proxy task for pretraining a foundation model for image registration. Empirical results show that our pretraining strategy improves registration accuracy, reduces the amount of domain-specific data needed to achieve competitive performance, and accelerates convergence during downstream training, thereby enhancing computational efficiency.