arXiv 2304.041552023-04-09eess.IVcs.CV

Segment Anything Model (SAM) 在数字病理学中的应用：评估在全切片成像上的零样本分割

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

Vanderbilt University（范德堡大学）；Vanderbilt University Medical Center（范德堡大学医学中心）；Veterans Affairs Tennessee Valley Healthcare System（退伍军人事务部田纳西谷医疗系统）

Ruining Deng, Can Cui, Quan Liu, Tianyuan Yao, Lucas W. Remedios, Shunxing Bao, Bennett A. Landman, Lee E. Wheless, Lori A. Coburn, Keith T. Wilson, Yaohong Wan… 展开作者

Ruining Deng, Can Cui, Quan Liu, Tianyuan Yao, Lucas W. Remedios, Shunxing Bao, Bennett A. Landman, Lee E. Wheless, Lori A. Coburn, Keith T. Wilson, Yaohong Wang, Shilin Zhao, Agnes B. Fogo, Haichun Yang, Yucheng Tang, Yuankai Huo

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本研究评估了 SAM 模型在 WSI 上的零样本分割性能，发现其对大型连通物体效果显著，但对密集实例分割表现不佳，并指出未来少样本微调的潜力。

摘要：

Segment Anything Model (SAM) 作为图像分割的基础模型被发布。该可提示分割模型在1100万张授权且尊重隐私的图像上使用超过10亿个掩码进行了训练。该模型支持使用各种分割提示（例如，点、框、掩码）进行零样本图像分割。这使得 SAM 对医学图像分析具有吸引力，特别是在训练数据稀缺的数字病理学领域。在本研究中，我们评估了 SAM 模型在全切片成像 (WSI) 上代表性分割任务的零样本分割性能，包括：(1) 肿瘤分割，(2) 非肿瘤组织分割，(3) 细胞核分割。核心结果：结果表明，零样本 SAM 模型对大型连通物体实现了显著的分割性能。然而，它对于密集实例物体分割并未持续取得令人满意的性能，即使在每张图像上使用20个提示（点击/框）。我们还总结了其在数字病理学中识别出的局限性：(1) 图像分辨率，(2) 多尺度，(3) 提示选择，以及 (4) 模型微调。未来，使用来自下游病理分割任务的图像进行少样本微调，可能有助于模型在密集物体分割中取得更好的性能。

英文摘要：

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.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。