Search papers, labs, and topics across Lattice.
This paper introduces the Hierarchical Probabilistic Representation (HPR) framework for prompt-free adaptation of the Segment Anything Model (SAM) in medical image segmentation. By leveraging Distributional Anatomical Representation (DAR), Multi-component Anatomical Representation (MAR), and Local Reliability Representation (LRR), HPR-SAM captures complex anatomical features and integrates them through Hierarchical Prediction Fusion (HPF). Experimental results show that HPR-SAM achieves state-of-the-art performance on the Synapse dataset and excels in few-shot scenarios on the LA and PROMISE12 datasets, highlighting its effectiveness in enhancing medical image segmentation without prompts.
HPR-SAM outperforms existing methods by effectively capturing complex anatomical representations, achieving state-of-the-art results in medical image segmentation without the need for prompts.
Prompt-free adaptation of the Segment Anything Model (SAM) has emerged as a promising paradigm for automatic medical image segmentation. Existing methods mainly focus on prompt generation, while overlooking that prompt quality is fundamentally constrained by the expressiveness of anatomical representations. However, deterministic prototypes or semantic tokens are insufficient to jointly capture global anatomical priors, intra-structure diversity, and local structural reliability. To address this limitation, we propose the Hierarchical Probabilistic Representation (HPR) framework, which learns complementary anatomical representations through Distributional Anatomical Representation (DAR), Multi-component Anatomical Representation (MAR), and Local Reliability Representation (LRR), and integrates their predictions via Hierarchical Prediction Fusion (HPF) while remaining compatible with the original SAM decoder. Experiments on the Synapse, LA, and PROMISE12 datasets demonstrate that HPR-SAM achieves state-of-the-art performance on Synapse and the best performance under few-shot settings on LA and PROMISE12, validating the effectiveness of the proposed hierarchical probabilistic representation learning framework for prompt-free medical image segmentation. Code is available at https://anonymous.4open.science/r/HPR-SAM-E4AF.
