# Domain-randomized deep learning for neuroimage analysis

arXiv: https://arxiv.org/abs/2507.13458 · HTML: https://mlsanity.com/paper/2507.13458/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 41 references that are in the map.
- First posted on arXiv 17 Jul 2025.

## Builds on (most cited first, 20 of 20)

- [Adversarial Discriminative Domain Adaptation](https://mlsanity.com/paper/1702.05464/) (2017)
- [Domain Generalization: A Survey](https://mlsanity.com/paper/2103.02503/) (2021)
- [VoxelMorph: A Learning Framework for Deformable Medical Image Registration](https://mlsanity.com/paper/1809.05231/) (2018)
- [Deep Visual Domain Adaptation: A Survey](https://mlsanity.com/paper/1802.03601/) (2018)
- [SynthSeg: Segmentation of brain MRI scans of any contrast and resolution without retraining](https://mlsanity.com/paper/2107.09559/) (2021)
- [Correlation Alignment for Unsupervised Domain Adaptation](https://mlsanity.com/paper/1612.01939/) (2016)
- [SynthMorph: learning contrast-invariant registration without acquired images](https://mlsanity.com/paper/2004.10282/) (2020)
- [Domain Adaptation for Visual Applications: A Comprehensive Survey](https://mlsanity.com/paper/1702.05374/) (2017)
- [SynthStrip: Skull-Stripping for Any Brain Image](https://mlsanity.com/paper/2203.09974/) (2022)
- [Diffusion Probabilistic Models beat GANs on Medical Images](https://mlsanity.com/paper/2212.07501/) (2022)
- [Joint super-resolution and synthesis of 1 mm isotropic MP-RAGE volumes from clinical MRI exams with scans of different orientation, resolution and contrast](https://mlsanity.com/paper/2012.13340/) (2020)
- [Learning General-Purpose Biomedical Volume Representations using Randomized Synthesis](https://mlsanity.com/paper/2411.02372/) (2024)
- [Anatomy-aware and acquisition-agnostic joint registration with SynthMorph](https://mlsanity.com/paper/2301.11329/) (2023)
- [AnyStar: Domain randomized universal star-convex 3D instance segmentation](https://mlsanity.com/paper/2307.07044/) (2023)
- [Quantifying white matter hyperintensity and brain volumes in heterogeneous clinical and low-field portable MRI](https://mlsanity.com/paper/2312.05119/) (2023)
- [Boosting Skull-Stripping Performance for Pediatric Brain Images](https://mlsanity.com/paper/2402.16634/) (2024)
- [Improving cross-domain brain tissue segmentation in fetal MRI with synthetic data](https://mlsanity.com/paper/2403.15103/) (2024)
- [Neurovascular Segmentation in sOCT with Deep Learning and Synthetic Training Data](https://mlsanity.com/paper/2407.01419/) (2024)
- [Learn2Synth: Learning Optimal Data Synthesis Using Hypergradients for Brain Image Segmentation](https://mlsanity.com/paper/2411.16719/) (2024)
- [Learning accurate rigid registration for longitudinal brain MRI from synthetic data](https://mlsanity.com/paper/2501.13010/) (2025)

## Built on by (most cited first, 6 of 6)

- [SingleStrip: learning skull-stripping from a single labeled example](https://mlsanity.com/paper/2508.10464/) (2025)
- [Deep infant brain segmentation from multi-contrast MRI](https://mlsanity.com/paper/2512.05114/) (2025)
- [SFL-Net: Source-Factorized Latent Representation Learning for Multi-Contrast MRI to Tau-PET Synthesis](https://mlsanity.com/paper/2602.22545/) (2026)
- [Why Invariance is Not Enough for Biomedical Domain Generalization and How to Fix It](https://mlsanity.com/paper/2604.02564/) (2026)
- [Impact of Synthetic Lesional MR Images in Automated Focal Cortical Dysplasia Detection in Low-Data Scenarios](https://mlsanity.com/paper/2606.07381/) (2026)
- [textttDisMorph: learning to disentangle technical distortions from true biological change](https://mlsanity.com/paper/2608.08173/) (2026)

## Abstract (arXiv)

Deep learning has revolutionized neuroimage analysis by delivering unprecedented speed and accuracy. However, the narrow scope of many training datasets constrains model robustness and generalizability. This challenge is particularly acute in magnetic resonance imaging (MRI), where image appearance varies widely across pulse sequences and scanner hardware. A recent domain-randomization strategy addresses the generalization problem by training deep neural networks on synthetic images with randomized intensities and anatomical content. By generating diverse data from anatomical segmentation maps, the approach enables models to accurately process image types unseen during training, without retraining or fine-tuning. It has demonstrated effectiveness across modalities including MRI, computed tomography, positron emission tomography, and optical coherence tomography, as well as beyond neuroimaging in ultrasound, electron and fluorescence microscopy, and X-ray microtomography. This tutorial paper reviews the principles, implementation, and potential of the synthesis-driven training paradigm. It highlights key benefits, such as improved generalization and resistance to overfitting, while discussing trade-offs such as increased computational demands. Finally, the article explores practical considerations for adopting the technique, aiming to accelerate the development of generalizable tools that make deep learning more accessible to domain experts without extensive computational resources or machine learning knowledge.