Segment Anything Model (SAM) for Digital Pathology: Assess Zero-shot Segmentation on Whole Slide Imaging

TL;DR

This study evaluates the zero-shot segmentation capabilities of the Segment Anything Model (SAM) on digital pathology whole slide images, highlighting its strengths in large object segmentation and limitations in dense object segmentation.

Contribution

The paper provides the first assessment of SAM's zero-shot performance on digital pathology tasks, identifying key limitations and potential avenues for improvement.

Findings

SAM performs well on large connected objects in pathology images.

SAM struggles with dense instance segmentation, even with multiple prompts.

Limitations include image resolution, multi-scale challenges, prompt selection, and the need for fine-tuning.

Abstract

The segment anything model (SAM) was released as a foundation model for image segmentation. The promptable segmentation model was trained by over 1 billion masks on 11M licensed and privacy-respecting images. The model supports zero-shot image segmentation with various segmentation prompts (e.g., points, boxes, masks). It makes the SAM attractive for medical image analysis, especially for digital pathology where the training data are rare. In this study, we evaluate the zero-shot segmentation performance of SAM model on representative segmentation tasks on whole slide imaging (WSI), including (1) tumor segmentation, (2) non-tumor tissue segmentation, (3) cell nuclei segmentation. Core Results: The results suggest that the zero-shot SAM model achieves remarkable segmentation performance for large connected objects. However, it does not consistently achieve satisfying performance for dense instance object segmentation, even with 20 prompts (clicks/boxes) on each image. We also summarized the identified limitations for digital pathology: (1) image resolution, (2) multiple scales, (3) prompt selection, and (4) model fine-tuning. In the future, the few-shot fine-tuning with images from downstream pathological segmentation tasks might help the model to achieve better performance in dense object segmentation.