# How Reliable Are Out-of-Distribution Generalization Methods for Medical Image Segmentation?

arXiv: https://arxiv.org/abs/2109.01668 · HTML: https://mlsanity.com/paper/2109.01668/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 15 references that are in the map.
- First posted on arXiv 3 Sep 2021.

## Builds on (most cited first, 9 of 9)

- [Unsupervised Domain Adaptation by Backpropagation](https://mlsanity.com/paper/1409.7495/) (2014)
- [Invariant Risk Minimization](https://mlsanity.com/paper/1907.02893/) (2019)
- [A Survey on Deep Learning in Medical Image Analysis](https://mlsanity.com/paper/1702.05747/) (2017)
- [Out-of-Distribution Generalization via Risk Extrapolation (REx)](https://mlsanity.com/paper/2003.00688/) (2020)
- [A large annotated medical image dataset for the development and evaluation of segmentation algorithms](https://mlsanity.com/paper/1902.09063/) (2019)
- [nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation](https://mlsanity.com/paper/1809.10486/) (2018)
- [Causality matters in medical imaging](https://mlsanity.com/paper/1912.08142/) (2019)
- [Invariant Risk Minimization Games](https://mlsanity.com/paper/2002.04692/) (2020)
- [Hippocampus Segmentation on Epilepsy and Alzheimer's Disease Studies with Multiple Convolutional Neural Networks](https://mlsanity.com/paper/2001.05058/) (2020)

## Built on by (most cited first, 10 of 10)

- [Continual Hippocampus Segmentation with Transformers](https://mlsanity.com/paper/2204.08043/) (2022)
- [Task-agnostic Continual Hippocampus Segmentation for Smooth Population Shifts](https://mlsanity.com/paper/2208.03206/) (2022)
- [Finding Competence Regions in Domain Generalization](https://mlsanity.com/paper/2303.09989/) (2023)
- [RELand: Risk Estimation of Landmines via Interpretable Invariant Risk Minimization](https://mlsanity.com/paper/2311.03115/) (2023)
- [NCAdapt: Dynamic adaptation with domain-specific Neural Cellular Automata for continual hippocampus segmentation](https://mlsanity.com/paper/2410.23368/) (2024)
- [Picking the Right Specialist: Attentive Neural Process-based Selection of Task-Specialized Models as Tools for Agentic Healthcare Systems](https://mlsanity.com/paper/2602.14901/) (2026)
- [Disentanglement enables cross-domain Hippocampus Segmentation](https://mlsanity.com/paper/2201.05650/) (2022)
- [Human Heterogeneity Invariant Stress Sensing](https://mlsanity.com/paper/2506.02256/) (2025)
- [Out-of-distribution data supervision towards biomedical semantic segmentation](https://mlsanity.com/paper/2507.12105/) (2025)
- [Beyond In-Distribution Metrics: A Systematic Out-of-Distribution Evaluation of Congenital Heart Disease Segmentation](https://mlsanity.com/paper/2609.17068/) (2026)

## Abstract (arXiv)

The recent achievements of Deep Learning rely on the test data being similar in distribution to the training data. In an ideal case, Deep Learning models would achieve Out-of-Distribution (OoD) Generalization, i.e. reliably make predictions on out-of-distribution data. Yet in practice, models usually fail to generalize well when facing a shift in distribution. Several methods were thereby designed to improve the robustness of the features learned by a model through Regularization- or Domain-Prediction-based schemes. Segmenting medical images such as MRIs of the hippocampus is essential for the diagnosis and treatment of neuropsychiatric disorders. But these brain images often suffer from distribution shift due to the patient's age and various pathologies affecting the shape of the organ. In this work, we evaluate OoD Generalization solutions for the problem of hippocampus segmentation in MR data using both fully- and semi-supervised training. We find that no method performs reliably in all experiments. Only the V-REx loss stands out as it remains easy to tune, while it outperforms a standard U-Net in most cases.