Source free domain adaptation for medical image segmentation with fourier style mining

Chen Yang, Xiaoqing Guo, Zhen Chen, Yixuan Yuan*

*Corresponding author for this work

Research output: Contribution to journalJournal articlepeer-review

65 Citations (Scopus)

Abstract

Unsupervised domain adaptation (UDA) aims to exploit the knowledge learned from a labeled source dataset to solve similar tasks in a new unlabeled target domain. Existing UDA techniques typically assume that samples from source and target domains are freely accessible during the training. However, it may be impractical to access source images due to privacy concerns, especially in medical imaging scenarios with the patient information. To tackle this issue, we devise a novel source free domain adaptation framework with fourier style mining, where only a well-trained source segmentation model is available for the adaptation to the target domain. Our framework is composed of two stages: a generation stage and an adaptation stage. In the generation stage, we design a Fourier Style Mining (FSM) generator to inverse source-like images through statistic information of the pretrained source model and mutual Fourier Transform. These generated source-like images can provide source data distribution and benefit the domain alignment. In the adaptation stage, we design a Contrastive Domain Distillation (CDD) module to achieve feature-level adaptation, including a domain distillation loss to transfer relation knowledge and a domain contrastive loss to narrow down the domain gap by a self-supervised paradigm. Besides, a Compact-Aware Domain Consistency (CADC) module is proposed to enhance consistency learning by filtering out noisy pseudo labels with shape compactness metric, thus achieving output-level adaptation. Extensive experiments on cross-device and cross-centre datasets are conducted for polyp and prostate segmentation, and our method delivers impressive performance compared with state-of-the-art domain adaptation methods.
Original languageEnglish
Article number102457
Number of pages13
JournalMedical Image Analysis
Volume79
DOIs
Publication statusPublished - Jul 2022

Scopus Subject Areas

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging
  • Computer Vision and Pattern Recognition
  • Health Informatics
  • Computer Graphics and Computer-Aided Design

User-Defined Keywords

  • Source Free Domain Adaptation
  • Fourier Style Mining
  • Contrastive Domain Distillation
  • Consistency Learning

Fingerprint

Dive into the research topics of 'Source free domain adaptation for medical image segmentation with fourier style mining'. Together they form a unique fingerprint.

Cite this