arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.11210cs.CV
收藏

SAM2在3D CT图像分割中性能的简短回顾与评估

A Short Review and Evaluation of SAM2's Performance in 3D CT Image Segmentation

NVIDIA(英伟达)

Yufan He, Pengfei Guo, Yucheng Tang, Andriy Myronenko, Vishwesh Nath, Ziyue Xu, Dong Yang, Can Zhao, Daguang Xu, Wenqi Li

AI总结:

本文回顾SAM2在3D CT分割中的评估,复现其零样本交互流程,发现直接应用效果不佳,易产生假阳性,仅对肾脏等小物体表现尚可,需更多研究以正确使用。

摘要:

自Segment Anything 2(SAM2)发布以来,医学影像社区一直在积极评估其在3D医学图像分割中的性能。然而,不同研究采用了不同的评估流程,导致结果相互矛盾,使得对SAM2的能力和潜在应用难以形成清晰认识。我们简要回顾了现有基准,并指出SAM2论文明确概述了一个零样本评估流程,该流程模拟用户点击,最多迭代八次。我们在3D CT数据集上复现了这一交互式标注模拟,并提供了结果和代码(https://github.com/Project-MONAI/VISTA)。我们的发现表明,将SAM2直接以零样本方式应用于3D医学影像远非令人满意。当前景物体消失时,它容易产生假阳性,且标注更多切片无法完全抵消这一趋势。对于肾脏和主动脉等较小的单连通物体,SAM2表现尚可,但对于大多数器官,其性能仍远落后于最先进的3D标注方法。3D医学影像社区需要更多的研究和创新,才能正确使用SAM2。

英文摘要:

Since the release of Segment Anything 2 (SAM2), the medical imaging community has been actively evaluating its performance for 3D medical image segmentation. However, different studies have employed varying evaluation pipelines, resulting in conflicting outcomes that obscure a clear understanding of SAM2's capabilities and potential applications. We shortly review existing benchmarks and point out that the SAM2 paper clearly outlines a zero-shot evaluation pipeline, which simulates user clicks iteratively for up to eight iterations. We reproduced this interactive annotation simulation on 3D CT datasets and provided the results and code~\url{https://github.com/Project-MONAI/VISTA}. Our findings reveal that directly applying SAM2 on 3D medical imaging in a zero-shot manner is far from satisfactory. It is prone to generating false positives when foreground objects disappear, and annotating more slices cannot fully offset this tendency. For smaller single-connected objects like kidney and aorta, SAM2 performs reasonably well but for most organs it is still far behind state-of-the-art 3D annotation methods. More research and innovation are needed for 3D medical imaging community to use SAM2 correctly.

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。

↑