A Large Scale Benchmark for Test Time Adaptation Methods in Medical Image Segmentation

TL;DR

MedSeg-TTA introduces a comprehensive benchmark evaluating twenty test time adaptation methods across seven medical imaging modalities, highlighting their strengths, limitations, and the need for careful method selection in clinical settings.

Contribution

This work provides the first systematic cross-modality comparison of test time adaptation methods in medical image segmentation, with standardized datasets and protocols.

Findings

No single method outperforms others across all conditions.

Input-level methods are more stable under mild appearance shifts.

Feature and output-level methods excel in boundary-related metrics.

Abstract

Test time Adaptation is a promising approach for mitigating domain shift in medical image segmentation; however, current evaluations remain limited in terms of modality coverage, task diversity, and methodological consistency. We present MedSeg-TTA, a comprehensive benchmark that examines twenty representative adaptation methods across seven imaging modalities, including MRI, CT, ultrasound, pathology, dermoscopy, OCT, and chest X-ray, under fully unified data preprocessing, backbone configuration, and test time protocols. The benchmark encompasses four significant adaptation paradigms: Input-level Transformation, Feature-level Alignment, Output-level Regularization, and Prior Estimation, enabling the first systematic cross-modality comparison of their reliability and applicability. The results show that no single paradigm performs best in all conditions. Input-level methods are more stable under mild appearance shifts. Feature-level and Output-level methods offer greater advantages in boundary-related metrics, whereas prior-based methods exhibit strong modality dependence. Several methods degrade significantly under large inter-center and inter-device shifts, which highlights the importance of principled method selection for clinical deployment. MedSeg-TTA provides standardized datasets, validated implementations, and a public leaderboard, establishing a rigorous foundation for future research on robust, clinically reliable test-time adaptation. All source codes and open-source datasets are available at https://github.com/wenjing-gg/MedSeg-TTA.