SD-FSMIS: Adapting Stable Diffusion for Few-Shot Medical Image Segmentation

TL;DR

This paper introduces SD-FSMIS, a novel method that adapts the pre-trained Stable Diffusion model for few-shot medical image segmentation, leveraging visual priors for improved data efficiency and robustness.

Contribution

It proposes a new framework with support-query interaction and visual-to-textual translation modules to effectively adapt diffusion models for FSMIS tasks.

Findings

Achieves competitive results on standard FSMIS benchmarks.

Demonstrates strong generalization in cross-domain scenarios.

Highlights the potential of large-scale generative models in medical segmentation.

Abstract

Few-Shot Medical Image Segmentation (FSMIS) aims to segment novel object classes in medical images using only minimal annotated examples, addressing the critical challenges of data scarcity and domain shifts prevalent in medical imaging. While Diffusion Models (DM) excel in visual tasks, their potential for FSMIS remains largely unexplored. We propose that the rich visual priors learned by large-scale DMs offer a powerful foundation for a more robust and data-efficient segmentation approach. In this paper, we introduce SD-FSMIS, a novel framework designed to effectively adapt the powerful pre-trained Stable Diffusion (SD) model for the FSMIS task. Our approach repurposes its conditional generative architecture by introducing two key components: a Support-Query Interaction (SQI) and a Visual-to-Textual Condition Translator (VTCT). Specifically, SQI provides a straightforward yet powerful means of adapting SD to the FSMIS paradigm. The VTCT module translates visual cues from the support set into an implicit textual embedding that guides the diffusion model, enabling precise conditioning of the generation process. Extensive experiments demonstrate that SD-FSMIS achieves competitive results compared to state-of-the-art methods in standard settings. Surprisingly, it also demonstrated excellent generalization ability in more challenging cross-domain scenarios. These findings highlight the immense potential of adapting large-scale generative models to advance data-efficient and robust medical image segmentation.