PGP-SAM: Prototype-Guided Prompt Learning for Efficient Few-Shot Medical Image Segmentation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Yan, Zhonghao, Yin, Zijin, Lin, Tianyu, Zeng, Xiangzhu, Liang, Kongming, Ma, Zhanyu
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910780947955712
author Yan, Zhonghao
Yin, Zijin
Lin, Tianyu
Zeng, Xiangzhu
Liang, Kongming
Ma, Zhanyu
author_facet Yan, Zhonghao
Yin, Zijin
Lin, Tianyu
Zeng, Xiangzhu
Liang, Kongming
Ma, Zhanyu
contents The Segment Anything Model (SAM) has demonstrated strong and versatile segmentation capabilities, along with intuitive prompt-based interactions. However, customizing SAM for medical image segmentation requires massive amounts of pixel-level annotations and precise point- or box-based prompt designs. To address these challenges, we introduce PGP-SAM, a novel prototype-based few-shot tuning approach that uses limited samples to replace tedious manual prompts. Our key idea is to leverage inter- and intra-class prototypes to capture class-specific knowledge and relationships. We propose two main components: (1) a plug-and-play contextual modulation module that integrates multi-scale information, and (2) a class-guided cross-attention mechanism that fuses prototypes and features for automatic prompt generation. Experiments on a public multi-organ dataset and a private ventricle dataset demonstrate that PGP-SAM achieves superior mean Dice scores compared with existing prompt-free SAM variants, while using only 10\% of the 2D slices.
format Preprint
id arxiv_https___arxiv_org_abs_2501_06692
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PGP-SAM: Prototype-Guided Prompt Learning for Efficient Few-Shot Medical Image Segmentation
Yan, Zhonghao
Yin, Zijin
Lin, Tianyu
Zeng, Xiangzhu
Liang, Kongming
Ma, Zhanyu
Computer Vision and Pattern Recognition
Artificial Intelligence
The Segment Anything Model (SAM) has demonstrated strong and versatile segmentation capabilities, along with intuitive prompt-based interactions. However, customizing SAM for medical image segmentation requires massive amounts of pixel-level annotations and precise point- or box-based prompt designs. To address these challenges, we introduce PGP-SAM, a novel prototype-based few-shot tuning approach that uses limited samples to replace tedious manual prompts. Our key idea is to leverage inter- and intra-class prototypes to capture class-specific knowledge and relationships. We propose two main components: (1) a plug-and-play contextual modulation module that integrates multi-scale information, and (2) a class-guided cross-attention mechanism that fuses prototypes and features for automatic prompt generation. Experiments on a public multi-organ dataset and a private ventricle dataset demonstrate that PGP-SAM achieves superior mean Dice scores compared with existing prompt-free SAM variants, while using only 10\% of the 2D slices.
title PGP-SAM: Prototype-Guided Prompt Learning for Efficient Few-Shot Medical Image Segmentation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2501.06692