A Closer Look at Edema Area Segmentation in SD-OCT Images Using Adversarial Framework

TL;DR

This paper enhances edema area segmentation in SD-OCT images by integrating retinal layer information and test-time adaptation into an adversarial framework, significantly improving accuracy and robustness over existing weakly-supervised methods.

Contribution

It introduces a novel layer-structure-guided post-processing and test-time adaptation to improve weakly-supervised edema segmentation in SD-OCT images.

Findings

Improved segmentation accuracy on public datasets.

Enhanced robustness against variations in edema presentation.

Bridged performance gap between weakly-supervised and fully-supervised models.

Abstract

The development of artificial intelligence models for macular edema (ME) analy-sis always relies on expert-annotated pixel-level image datasets which are expen-sive to collect prospectively. While anomaly-detection-based weakly-supervised methods have shown promise in edema area (EA) segmentation task, their per-formance still lags behind fully-supervised approaches. In this paper, we leverage the strong correlation between EA and retinal layers in spectral-domain optical coherence tomography (SD-OCT) images, along with the update characteristics of weakly-supervised learning, to enhance an off-the-shelf adversarial framework for EA segmentation with a novel layer-structure-guided post-processing step and a test-time-adaptation (TTA) strategy. By incorporating additional retinal lay-er information, our framework reframes the dense EA prediction task as one of confirming intersection points between the EA contour and retinal layers, result-ing in predictions that better align with the shape prior of EA. Besides, the TTA framework further helps address discrepancies in the manifestations and presen-tations of EA between training and test sets. Extensive experiments on two pub-licly available datasets demonstrate that these two proposed ingredients can im-prove the accuracy and robustness of EA segmentation, bridging the gap between weakly-supervised and fully-supervised models.