• Home
  • Search
  • Prototype bank-driven test-time adaptation for medical ultrasound image segmentation.
  • Cite Icon1
  • https://doi.org/10.1002/mp.70280Copy DOI Icon

Prototype bank-driven test-time adaptation for medical ultrasound image segmentation.

Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Deep learning has achieved remarkable success in medical image segmentation, particularly in ultrasound imaging, where deep neural networks have demonstrated near-expert performance. However, these models typically assume that training and test data follow the same distribution-an assumption that often fails in real-world clinical practice due to domain shifts caused by variations in imaging devices, acquisition protocols, and operator techniques. These discrepancies can significantly degrade model performance. Existing solutions-such as supervised fine-tuning, unsupervised domain adaptation, and domain generalization-require either costly labeled data or access to source domain data, limiting their scalability and clinical applicability. To address domain shift in real-world ultrasound image segmentation, this study proposes a test-time adaptation (TTA) framework that eliminates the need for source data or target labels, while ensuring robustness against distributional drift and catastrophic forgetting. We present Prototype Bank-Driven Test-Time Adaptation (PBTTA), a novel TTA framework consisting of two key modules: (1) the Dynamic Statistics Fusion Module (DSFM), which enables domain-level adaptation by dynamically adjusting batch normalization layers using fused statistics from the test sample and source domain; and (2) the Prototype Bank-Guided Semantic Adaptation Module (PBSAM), which maintains a dynamic prototype bank for each semantic class to support semantic-level adaptation. PBTTA employs a dual-classifier strategy that combines a prototype-based classifier for stable semantic guidance and a parameter-based classifier for flexible decision-making. Notably, PBTTA does not require backpropagation to update model parameters during test time adaptation phase, ensuring efficient and stable adaptation. PBTTA achieves state-of-the-art performance across both ultrasound breast and thyroid tumor segmentation tasks. On average, it improves the Dice score by 15.04% (to 64.82%) for breast tumor segmentation and by 8.88% (to 57.45%) for thyroid tumor segmentation, compared to non-adaptive baselines. Moreover, PBTTA exhibits excellent robustness under continuous domain shifts and effectively mitigates catastrophic forgetting.

Similar Papers
  • Research Article
  • Citations66

Deep weakly-supervised breast tumor segmentation in ultrasound images with explicit anatomical constraints

  • Feb 01, 2022
  • Medical Image Analysis
  • Yongshuai Li +4
  • Research Article
  • Citations30

S-CUDA: Self-cleansing unsupervised domain adaptation for medical image segmentation.

  • Dec 01, 2021
  • Medical Image Analysis
  • Luyan Liu +4
  • Abstract

Unsupervised Domain Adaptation of Auto-Segmentation on Multi-Source MRIs

  • Sep 29, 2023
  • International Journal of Radiation Oncology*Biology*Physics
  • W Zhang +4
  • Research Article
  • Citations25

ACT: Semi-supervised Domain-adaptive Medical Image Segmentation with Asymmetric Co-Training.

  • Jan 01, 2022
  • Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
  • Xiaofeng Liu +6
  • Research Article
  • Citations1

Unsupervised domain adaptation with hyperbolic graph convolution network for segmentation of X-ray breast mass

  • Apr 28, 2022
  • Journal of Intelligent & Fuzzy Systems
  • Kai Bi +1
  • Research Article
  • Citations11

O2M-UDA: Unsupervised dynamic domain adaptation for one-to-multiple medical image segmentation

  • Feb 09, 2023
  • Knowledge-Based Systems
  • Ziyue Jiang +9
  • Research Article
  • Citations55

FVP: Fourier Visual Prompting for Source-Free Unsupervised Domain Adaptation of Medical Image Segmentation.

  • Dec 01, 2023
  • IEEE Transactions on Medical Imaging
  • Yan Wang +7
  • Research Article

A3-DualUD: Source-free unsupervised domain adaptation via anatomical anchor alignment and dual-path uncertainty denoising for cross-modality medical image segmentation.

  • Nov 01, 2025
  • Computer methods and programs in biomedicine
  • Jie Zhou +5
  • Research Article
  • Citations70

Breast tumor segmentation in ultrasound images using contextual-information-aware deep adversarial learning framework

  • Aug 14, 2020
  • Expert Systems with Applications
  • Vivek Kumar Singh +7
  • Research Article
  • Citations81

LE-UDA: Label-efficient unsupervised domain adaptation for medical image segmentation.

  • Mar 01, 2023
  • IEEE Transactions on Medical Imaging
  • Ziyuan Zhao +5
  • Preprint Article

Segmentation Cycle-GAN Loss for Structure-Preserving Domain Adaptation of Medical Images

  • Sep 06, 2024
  • Paolo Iacono
  • Preprint Article

Segmentation Cycle-GAN Loss for Structure-Preserving Domain Adaptation of Medical Images

  • Sep 06, 2024
  • Paolo Iacono
  • Research Article
  • Citations4

Syn-Net: A Synchronous Frequency-Perception Fusion Network for Breast Tumor Segmentation in Ultrasound Images.

  • Mar 01, 2025
  • IEEE journal of biomedical and health informatics
  • Guangzhe Zhao +4
  • Book Chapter
  • Citations7

Self-supervised Test-Time Adaptation for Medical Image Segmentation

  • Jan 01, 2022
  • Hao Li +12
  • Conference Article
  • Citations5

Weakly-supervised deep learning for breast tumor segmentation in ultrasound images

  • Sep 11, 2021
  • Yongshuai Li +3
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.