Hierarchical Self-Prompting SAM: A Prompt-Free Medical Image Segmentation Framework

TL;DR

This paper introduces Hierarchical Self-Prompting SAM, a prompt-free medical image segmentation framework that learns abstract prompts, significantly improving performance and robustness across various medical imaging tasks and datasets.

Contribution

It presents the first self-prompting framework that learns abstract prompts during segmentation, surpassing previous methods limited to positional prompts, and demonstrates strong generalization in medical imaging.

Findings

Achieves up to 14.04% improvement over state-of-the-art methods.

Performs well across diverse medical imaging modalities.

Exhibits strong generalization to unseen datasets.

Abstract

Although the Segment Anything Model (SAM) is highly effective in natural image segmentation, it requires dependencies on prompts, which limits its applicability to medical imaging where manual prompts are often unavailable. Existing efforts to fine-tune SAM for medical segmentation typically struggle to remove this dependency. We propose Hierarchical Self-Prompting SAM (HSP-SAM), a novel self-prompting framework that enables SAM to achieve strong performance in prompt-free medical image segmentation. Unlike previous self-prompting methods that remain limited to positional prompts similar to vanilla SAM, we are the first to introduce learning abstract prompts during the self-prompting process. This simple and intuitive self-prompting framework achieves superior performance on classic segmentation tasks such as polyp and skin lesion segmentation, while maintaining robustness across diverse medical imaging modalities. Furthermore, it exhibits strong generalization to unseen datasets, achieving improvements of up to 14.04% over previous state-of-the-art methods on some challenging benchmarks. These results suggest that abstract prompts encapsulate richer and higher-dimensional semantic information compared to positional prompts, thereby enhancing the model's robustness and generalization performance. All models and codes will be released upon acceptance.