Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Advancing UWF-SLO Vessel Segmentation with Source-Free Active Domain Adaptation and a Novel Multi-Center Dataset

Hongqiu Wang, Xiangde Luo, Wu Chen, Qingqing Tang, Mei Xin, Qiong Wang, Lei Zhu

TL;DR

Experimental results demonstrate that the approach surpasses existing domain adaptation and active learning methods, considerably reducing the gap between the Upper and Lower bounds with minimal annotations, highlighting the method's practical clinical value.

Abstract

Accurate vessel segmentation in Ultra-Wide-Field Scanning Laser Ophthalmoscopy (UWF-SLO) images is crucial for diagnosing retinal diseases. Although recent techniques have shown encouraging outcomes in vessel segmentation, models trained on one medical dataset often underperform on others due to domain shifts. Meanwhile, manually labeling high-resolution UWF-SLO images is an extremely challenging, time-consuming and expensive task. In response, this study introduces a pioneering framework that leverages a patch-based active domain adaptation approach. By actively recommending a few valuable image patches by the devised Cascade Uncertainty-Predominance (CUP) selection strategy for labeling and model-finetuning, our method significantly improves the accuracy of UWF-SLO vessel segmentation across diverse medical centers. In addition, we annotate and construct the first Multi-center UWF-SLO Vessel Segmentation (MU-VS) dataset to promote this topic research, comprising data from multiple institutions. This dataset serves as a valuable resource for cross-center evaluation, verifying the effectiveness and robustness of our approach. Experimental results demonstrate that our approach surpasses existing domain adaptation and active learning methods, considerably reducing the gap between the Upper and Lower bounds with minimal annotations, highlighting our method's practical clinical value. We will release our dataset and code to facilitate relevant research: https://github.com/whq-xxh/SFADA-UWF-SLO.

Advancing UWF-SLO Vessel Segmentation with Source-Free Active Domain Adaptation and a Novel Multi-Center Dataset

TL;DR

Experimental results demonstrate that the approach surpasses existing domain adaptation and active learning methods, considerably reducing the gap between the Upper and Lower bounds with minimal annotations, highlighting the method's practical clinical value.

Abstract

Accurate vessel segmentation in Ultra-Wide-Field Scanning Laser Ophthalmoscopy (UWF-SLO) images is crucial for diagnosing retinal diseases. Although recent techniques have shown encouraging outcomes in vessel segmentation, models trained on one medical dataset often underperform on others due to domain shifts. Meanwhile, manually labeling high-resolution UWF-SLO images is an extremely challenging, time-consuming and expensive task. In response, this study introduces a pioneering framework that leverages a patch-based active domain adaptation approach. By actively recommending a few valuable image patches by the devised Cascade Uncertainty-Predominance (CUP) selection strategy for labeling and model-finetuning, our method significantly improves the accuracy of UWF-SLO vessel segmentation across diverse medical centers. In addition, we annotate and construct the first Multi-center UWF-SLO Vessel Segmentation (MU-VS) dataset to promote this topic research, comprising data from multiple institutions. This dataset serves as a valuable resource for cross-center evaluation, verifying the effectiveness and robustness of our approach. Experimental results demonstrate that our approach surpasses existing domain adaptation and active learning methods, considerably reducing the gap between the Upper and Lower bounds with minimal annotations, highlighting our method's practical clinical value. We will release our dataset and code to facilitate relevant research: https://github.com/whq-xxh/SFADA-UWF-SLO.
Paper Structure (10 sections, 3 equations, 3 figures, 5 tables)

This paper contains 10 sections, 3 equations, 3 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Methodology
  3. Problem setting
  4. Pipeline of Patch-Based SFADA Framework
  5. Cascade Uncertainty-Predominance selection
  6. Experiments and Results
  7. Data Description
  8. Implementation Details and Evaluation Metrics
  9. Experimental Results
  10. Conclusion

Figures (3)

  • Figure 1: Several data visualization examples from the proposed Multi-center UWF-SLO Vessel Segmentation (MU-VS) dataset, illustrate that various centers encompass different disease categories.
  • Figure 2: Overview of the proposed patch-based SFADA framework including the CUP selection strategy. The first row of gray represents the training of source model; the second row of light blue represents the recommendation and annotation of the valuable image patches based on the CUP strategy; and the third row of green represents the target model's fine-tuning under the supervision of enhanced pseudo-labels.
  • Figure 3: Visual comparisons of our method and other leading domain adaptation and active learning methods. Our method can more accurately segment the vessels (please zoom in for more details).