# Constrained domain adaptation for Image segmentation

arXiv: https://arxiv.org/abs/1908.02996 · HTML: https://mlsanity.com/paper/1908.02996/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Aug 2019.

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

- [Domain Adaptation for Medical Image Analysis: A Survey](https://mlsanity.com/paper/2102.09508/) (2021)
- [CrossMoDA 2021 challenge: Benchmark of Cross-Modality Domain Adaptation techniques for Vestibular Schwannoma and Cochlea Segmentation](https://mlsanity.com/paper/2201.02831/) (2022)
- [Medical Image Segmentation with Limited Supervision: A Review of Deep Network Models](https://mlsanity.com/paper/2103.00429/) (2021)
- [CellTranspose: Few-shot Domain Adaptation for Cellular Instance Segmentation](https://mlsanity.com/paper/2212.14121/) (2022)
- [Weakly-Supervised 3D Medical Image Segmentation using Geometric Prior and Contrastive Similarity](https://mlsanity.com/paper/2302.02125/) (2023)
- [Negative Evidence Matters in Interpretable Histology Image Classification](https://mlsanity.com/paper/2201.02445/) (2022)
- [Looking at the whole picture: constrained unsupervised anomaly segmentation](https://mlsanity.com/paper/2109.00482/) (2021)
- [Leveraging Anatomical Constraints with Uncertainty for Pneumothorax Segmentation](https://mlsanity.com/paper/2311.15213/) (2023)
- [Medical image segmentation with imperfect 3D bounding boxes](https://mlsanity.com/paper/2108.03300/) (2021)
- [Bladder segmentation based on deep learning approaches: current limitations and lessons](https://mlsanity.com/paper/2101.06498/) (2021)

## Abstract (arXiv)

We propose to adapt segmentation networks with a constrained formulation, which embeds domain-invariant prior knowledge about the segmentation regions. Such knowledge may take the form of simple anatomical information, e.g., structure size or shape, estimated from source samples or known a priori. Our method imposes domain-invariant inequality constraints on the network outputs of unlabeled target samples. It implicitly matches prediction statistics between target and source domains with permitted uncertainty of prior knowledge. We address our constrained problem with a differentiable penalty, fully suited for standard stochastic gradient descent approaches, removing the need for computationally expensive Lagrangian optimization with dual projections. Unlike current two-step adversarial training, our formulation is based on a single loss in a single network, which simplifies adaptation by avoiding extra adversarial steps, while improving convergence and quality of training. The comparison of our approach with state-of-the-art adversarial methods reveals substantially better performance on the challenging task of adapting spine segmentation across different MRI modalities. Our results also show a robustness to imprecision of size priors, approaching the accuracy of a fully supervised model trained directly in a target domain.Our method can be readily used for various constraints and segmentation problems.