Mammo-SAM: Adapting Foundation Segment Anything Model for Automatic Breast Mass Segmentation in Whole Mammograms

Xiong, Xinyu; Wang, Churan; Li, Wenxue; Li, Guanbin

doi:10.1007/978-3-031-45673-2_18

Cited by 7 publications

(1 citation statement)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Subsequent research has focused on enhancing SAM's interactive segmentation capability in specific domains through fine-tuning with ground truth prompts and segmentation pairs [3], [32], [33], [34], [35]. Additionally, other studies have explored extending SAM to automatic segmentation, through workaround [36], [37] or through automatic adaptation methods [38], [39], [40], [41], [42], [43], [44], [45]. However, there remains a gap in adapting interactive segmentation models to an automatic segmentation setting while retaining their promptability.…”

Section: Related Workmentioning

confidence: 99%

Vision-Language Contrastive Learning Approach to Robust Automatic Placenta Analysis Using Photographic Images

Pan

Gernand

Goldstein

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Semantic segmentation is a core task in computer vision. Existing methods are generally divided into two categories: automatic and interactive. Interactive approaches, exemplified by the Segment Anything Model (SAM), have shown promise as pre-trained models. However, current adaptation strategies for these models tend to lean towards either automatic or interactive approaches. Interactive methods depend on prompts user input to operate, while automatic ones bypass the interactive promptability entirely. Addressing these limitations, we introduce a novel paradigm and its first model: the Automatic and Interactive Segment Anything Model (AI-SAM). In this paradigm, we conduct a comprehensive analysis of prompt quality and introduce the pioneering Automatic and Interactive Prompter (AI-Prompter) that automatically generates initial point prompts while accepting additional user inputs. Our experimental results demonstrate AI-SAM's effectiveness in the automatic setting, achieving state-of-the-art performance. Significantly, it offers the flexibility to incorporate additional user prompts, thereby further enhancing its performance. The project page is available at https://github.com/ymp5078/AI-SAM.

show abstract