# SIAM: Head and Brain MRI Segmentation from Few High-Quality Templates via Synthetic Training

arXiv: https://arxiv.org/abs/2605.02737 · HTML: https://mlsanity.com/paper/2605.02737/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 50 references that are in the map.
- First posted on arXiv 4 May 2026.

## Builds on (most cited first, 16 of 16)

- [nnU-Net Revisited: A Call for Rigorous Validation in 3D Medical Image Segmentation](https://mlsanity.com/paper/2404.09556/) (2024)
- [SynthSeg: Segmentation of brain MRI scans of any contrast and resolution without retraining](https://mlsanity.com/paper/2107.09559/) (2021)
- [Metrics reloaded: Recommendations for image analysis validation](https://mlsanity.com/paper/2206.01653/) (2022)
- [Automated brain extraction of multi-sequence MRI using artificial neural networks](https://mlsanity.com/paper/1901.11341/) (2019)
- [3D Whole Brain Segmentation using Spatially Localized Atlas Network Tiles](https://mlsanity.com/paper/1903.12152/) (2019)
- [Bayesian QuickNAT: Model Uncertainty in Deep Whole-Brain Segmentation for Structure-wise Quality Control](https://mlsanity.com/paper/1811.09800/) (2018)
- [DeepNAT: Deep Convolutional Neural Network for Segmenting Neuroanatomy](https://mlsanity.com/paper/1702.08192/) (2017)
- [AssemblyNet: A large ensemble of CNNs for 3D Whole Brain MRI Segmentation](https://mlsanity.com/paper/1911.09098/) (2019)
- [FastSurferVINN: Building Resolution-Independence into Deep Learning Segmentation Methods -- A Solution for HighRes Brain MRI](https://mlsanity.com/paper/2112.09654/) (2021)
- [Learning joint segmentation of tissues and brain lesions from task-specific hetero-modal domain-shifted datasets](https://mlsanity.com/paper/2009.04009/) (2020)
- [Comprehensive analysis of synthetic learning applied to neonatal brain MRI segmentation](https://mlsanity.com/paper/2309.05306/) (2023)
- [UltraCortex: Submillimeter Ultra-High Field 9.4 T Brain MR Image Collection and Manual Cortical Segmentations](https://mlsanity.com/paper/2406.18571/) (2024)
- [Fighting the scanner effect in brain MRI segmentation with a progressive level-of-detail network trained on multi-site data](https://mlsanity.com/paper/2211.02400/) (2022)
- [Synthetic Data for Robust Stroke Segmentation](https://mlsanity.com/paper/2404.01946/) (2024)
- [GOUHFI: a novel contrast- and resolution-agnostic segmentation tool for Ultra-High Field MRI](https://mlsanity.com/paper/2505.11445/) (2025)
- [Comprehensive segmentation of deep grey nuclei from structural MRI data](https://mlsanity.com/paper/2503.21955/) (2025)

## Built on by (most cited first, 2 of 2)

- [LISynSeg: Data-Centric Label-to-Image Synthesis for Cross-Modality Whole-Heart Segmentation](https://mlsanity.com/paper/2608.31073/) (2026)
- [Synthetic training for long-tail haemorrhagic lesion segmentation in data-scarce settings](https://mlsanity.com/paper/2610.01542/) (2026)

## Abstract (arXiv)

Synthetic training has recently advanced brain MRI segmentation by enabling contrast-agnostic models trained entirely on generated data. However, most existing approaches rely on hundreds of automatically labeled templates, introducing systematic biases and limiting their flexibility to incorporate new anatomical structures. We present the Segment It All Model (SIAM), a 3D whole-head segmentation framework for 16 anatomical structures, trained using only six high-quality, manually annotated templates. SIAM extends domain randomization to both intensity and shape domains: synthetic image generation ensures contrast variability, while high-resolution spatial transformations model anatomical differences in cortical thickness and deep nuclei morphology. Unlike prior synthetic models, SIAM simultaneously segments brain as well as extra-cerebral tissues, including cerebrospinal fluid, vessels, dura mater, skull, and skin, enabling fully automated, preprocessing-free analysis. Evaluation across eight heterogeneous datasets (N=301), that include multiple contrasts (T1-weighted, T2-weighted, CT) and span a wide range of ages, demonstrates that SIAM matches or outperforms state-of-the-art methods for brain structures, in addition to extending automated segmentation to non-brain structures. The model also exhibits superior consistency across contrasts and repeated acquisitions, together with improved sensitivity to subtle gray matter atrophy. We openly release the model and the label templates at https://github.com/romainVala/SIAM.