# Towards Robust In-Context Learning for Medical Image Segmentation via Data Synthesis

arXiv: https://arxiv.org/abs/2509.19711 · HTML: https://mlsanity.com/paper/2509.19711/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 19 references that are in the map.
- First posted on arXiv 24 Sep 2025.

## Builds on (most cited first, 10 of 10)

- [AMOS: A Large-Scale Abdominal Multi-Organ Benchmark for Versatile Medical Image Segmentation](https://mlsanity.com/paper/2206.08023/) (2022)
- [SynthSeg: Segmentation of brain MRI scans of any contrast and resolution without retraining](https://mlsanity.com/paper/2107.09559/) (2021)
- [SegGPT: Segmenting Everything In Context](https://mlsanity.com/paper/2304.03284/) (2023)
- [UniverSeg: Universal Medical Image Segmentation](https://mlsanity.com/paper/2304.06131/) (2023)
- [MAISI: Medical AI for Synthetic Imaging](https://mlsanity.com/paper/2409.11169/) (2024)
- [Robust machine learning segmentation for large-scale analysis of heterogeneous clinical brain MRI datasets](https://mlsanity.com/paper/2209.02032/) (2022)
- [Universal and Extensible Language-Vision Models for Organ Segmentation and Tumor Detection from Abdominal Computed Tomography](https://mlsanity.com/paper/2405.18356/) (2024)
- [Neuralizer: General Neuroimage Analysis without Re-Training](https://mlsanity.com/paper/2305.02644/) (2023)
- [Rethinking Abdominal Organ Segmentation (RAOS) in the clinical scenario: A robustness evaluation benchmark with challenging cases](https://mlsanity.com/paper/2406.13674/) (2024)
- [A Chebyshev Confidence Guided Source-Free Domain Adaptation Framework for Medical Image Segmentation](https://mlsanity.com/paper/2310.18087/) (2023)

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

- [Efficient Universal Models for Medical Image Segmentation via Weakly Supervised In-Context Learning](https://mlsanity.com/paper/2510.05899/) (2025)
- [UniMedSeg: Unified In-Context Learning for Multi-Paradigm 2D/3D Medical Image Segmentation](https://mlsanity.com/paper/2607.12896/) (2026)

## Abstract (arXiv)

The rise of In-Context Learning (ICL) for universal medical image segmentation has introduced an unprecedented demand for large-scale, diverse datasets for training, exacerbating the long-standing problem of data scarcity. While data synthesis offers a promising solution, existing methods often fail to simultaneously achieve both high data diversity and a domain distribution suitable for medical data. To bridge this gap, we propose SynthICL, a novel data synthesis framework built upon domain randomization. SynthICL ensures realism by leveraging anatomical priors from real-world datasets, generates diverse anatomical structures to cover a broad data distribution, and explicitly models inter-subject variations to create data cohorts suitable for ICL. Extensive experiments on four held-out datasets validate our framework's effectiveness, showing that models trained with our data achieve performance gains of up to 63% in average Dice and substantially enhanced generalization to unseen anatomical domains. Our work helps mitigate the data bottleneck for ICL-based segmentation, paving the way for robust models. Our code and the generated dataset are publicly available at https://github.com/jiesihu/Neuroverse3D.