Few Exemplar-Based General Medical Image Segmentation via Domain-Aware Selective Adaptation

TL;DR

This paper introduces a domain-aware selective adaptation method that enables effective medical image segmentation with only a few exemplars, addressing domain gaps and the lack of precise prior knowledge in clinical scenarios.

Contribution

The paper proposes a novel adaptation approach that transfers general knowledge from large models to medical domains using minimal exemplars, suitable for low-resource settings.

Findings

Significant improvement in segmentation accuracy with fewer exemplars.

Robust performance across different medical imaging modalities.

Enhanced applicability in low- and middle-income countries.

Abstract

Medical image segmentation poses challenges due to domain gaps, data modality variations, and dependency on domain knowledge or experts, especially for low- and middle-income countries (LMICs). Whereas for humans, given a few exemplars (with corresponding labels), we are able to segment different medical images even without exten-sive domain-specific clinical training. In addition, current SAM-based medical segmentation models use fine-grained visual prompts, such as the bounding rectangle generated from manually annotated target segmentation mask, as the bounding box (bbox) prompt during the testing phase. However, in actual clinical scenarios, no such precise prior knowledge is available. Our experimental results also reveal that previous models nearly fail to predict when given coarser bbox prompts. Considering these issues, in this paper, we introduce a domain-aware selective adaptation approach to adapt the general knowledge learned from a large model trained with natural images to the corresponding medical domains/modalities, with access to only a few (e.g. less than 5) exemplars. Our method mitigates the aforementioned limitations, providing an efficient and LMICs-friendly solution. Extensive experimental analysis showcases the effectiveness of our approach, offering potential advancements in healthcare diagnostics and clinical applications in LMICs.