# PASS:Test-Time Prompting to Adapt Styles and Semantic Shapes in Medical Image Segmentation

Source: [PASS:Test-Time Prompting to Adapt Styles and Semantic Shapes in Medical Image Segmentation](<https://www.sota2.com/research/paper/68fa75021e209131cea6535a>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-27T06:41:00.437Z

Authors: Chuyan Zhang, Hao Zheng, Xin You, Yefeng Zheng, Yun Gu

Published: 2024-10-02T14:11:26.000Z

## Abstract

Test-time adaptation (TTA) has emerged as a promising paradigm to handle the domain shifts at test time for medical images from different institutions without using extra training data. However, existing TTA solutions for segmentation tasks suffer from (1) dependency on modifying the source training stage and access to source priors or (2) lack of emphasis on shape-related semantic knowledge that is crucial for segmentation tasks.Recent research on visual prompt learning achieves source-relaxed adaptation by extended parameter space but still neglects the full utilization of semantic features, thus motivating our work on knowledge-enriched deep prompt learning. Beyond the general concern of image style shifts, we reveal that shape variability is another crucial factor causing the performance drop. To address this issue, we propose a TTA framework called PASS (Prompting to Adapt Styles and Semantic shapes), which jointly learns two types of prompts: the input-space prompt to reformulate the style of the test image to fit into the pretrained model and the semantic-aware prompts to bridge high-level shape discrepancy across domains. Instead of naively imposing a fixed prompt, we introduce an input decorator to generate the self-regulating visual prompt conditioned on the input data. To retrieve the knowledge representations and customize target-specific shape prompts for each test sample, we propose a cross-attention prompt modulator, which performs interaction between target representations and an enriched shape prompt bank. Extensive experiments demonstrate the superior performance of PASS over state-of-the-art methods on multiple medical image segmentation datasets. The code is available at https://github.com/EndoluminalSurgicalVision-IMR/PASS.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Prostate Segmentation on BMC MRI Prostate (test)](<https://www.sota2.com/research/sota/prostate-segmentation-on-bmc-mri-prostate-test.md>): Prostate Segmentation; BMC MRI Prostate (test); DSC: 80.07.

- [Prostate Segmentation on I2CVB MRI Prostate (test)](<https://www.sota2.com/research/sota/prostate-segmentation-on-i2cvb-mri-prostate-test.md>): Prostate Segmentation; I2CVB MRI Prostate (test); DSC: 71.41.

- [Prostate Segmentation on UCL MRI Prostate (test)](<https://www.sota2.com/research/sota/prostate-segmentation-on-ucl-mri-prostate-test.md>): Prostate Segmentation; UCL MRI Prostate (test); DSC: 84.39.

- [Prostate Segmentation on MRI Prostate Average across sites (test)](<https://www.sota2.com/research/sota/prostate-segmentation-on-mri-prostate-average-across-sites-test.md>): Prostate Segmentation; MRI Prostate Average across sites (test); DSC: 75.6.

- [Multi-sequence MRI Segmentation on BraTS SSA Target 2021](<https://www.sota2.com/research/sota/multi-sequence-mri-segmentation-on-brats-ssa-target-2021.md>): Multi-sequence MRI segmentation; BraTS SSA Target 2021; Dice Coefficient: 79.73.

- [Prostate Segmentation on BIDMC MRI Prostate (test)](<https://www.sota2.com/research/sota/prostate-segmentation-on-bidmc-mri-prostate-test.md>): Prostate Segmentation; BIDMC MRI Prostate (test); DSC: 57.27.

- [Prostate Segmentation on HK (test)](<https://www.sota2.com/research/sota/prostate-segmentation-on-hk-test.md>): Prostate Segmentation; HK (test); DSC: 84.88.

- [Multi-sequence MRI segmentation on BraTS PED Target Domain 2021](<https://www.sota2.com/research/sota/multi-sequence-mri-segmentation-on-brats-ped-target-domain-2021.md>): Multi-sequence MRI segmentation; BraTS PED Target Domain 2021; Dice: 38.68.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2410.01573>)

[PDF](<https://arxiv.org/pdf/2410.01573.pdf>)
