arXiv 2304.078752023-04-16eess.IVcs.CV

Segment Anything 基础模型在 MRI 上实现了良好的脑肿瘤自动分割精度，可支持放疗治疗计划

The Segment Anything foundation model achieves favorable brain tumor autosegmentation accuracy on MRI to support radiotherapy treatment planning

Universitätsklinikum Erlangen（埃尔兰根大学医院）；Friedrich-Alexander-Universität Erlangen-Nürnberg（埃尔朗根-纽伦堡弗里德里希·亚历山大大学）；Comprehensive Cancer Center Erlangen-EMN (CCC ER-EMN)（埃尔兰根-欧洲分子神经科学网络综合癌症中心）

Florian Putz, Johanna Grigo, Thomas Weissmann, Philipp Schubert, Daniel Hoefler, Ahmed Gomaa, Hassen Ben Tkhayat, Amr Hagag, Sebastian Lettmaier, Benjamin Frey,… 展开作者

Florian Putz, Johanna Grigo, Thomas Weissmann, Philipp Schubert, Daniel Hoefler, Ahmed Gomaa, Hassen Ben Tkhayat, Amr Hagag, Sebastian Lettmaier, Benjamin Frey, Udo S. Gaipl, Luitpold V. Distel, Sabine Semrau, Christoph Bert, Rainer Fietkau, Yixing Huang

备注：: 13 pages, 5 figures

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文评估了 Segment Anything 基础模型在 MRI 脑肿瘤自动分割中的零样本性能，在 BraTS 2020 数据集上取得较高精度，证明其有潜力加速和辅助放疗治疗计划。

摘要：

背景：MRI 中的肿瘤分割对于脑肿瘤患者的放疗（RT）治疗计划至关重要。Segment Anything（SA）是一种新型的可提示基础模型，用于自动分割，在多个分割任务中表现出高精度，但尚未在医学数据集上进行评估。方法：在来自 369 个 MRI 数据集（BraTS 2020）的 16744 个横断面切片上，对 SA 在胶质瘤脑肿瘤自动分割的点到掩码任务中进行了评估。每个切片最多放置 9 个点提示。在对比增强 T1w 序列上分割肿瘤核心（增强肿瘤 + 坏死核心）。在 SA 预测的 3 个掩码中，分别对计算出的 IoU 最高的掩码（oracle 掩码）和模型预测 IoU 最高的掩码（suggested 掩码）进行精度评估。除了在完整 MRI 切片上评估 SA 外，还在裁剪到肿瘤区域的图像（最大 3D 范围 + 2 cm）上对 SA 进行了评估。结果：在完整 MRI 切片上使用 oracle 掩码的平均最佳 IoU（mbIoU）为 0.762（IQR 0.713-0.917）。最佳 2D 掩码在平均 6.6 个点提示（IQR 5-9）后获得。与低级别胶质瘤病例相比，高级别胶质瘤的分割精度显著更高（mbIoU 0.789 对 0.668）。使用裁剪到肿瘤区域的 MRI 切片时精度较差（mbIoU 0.759），使用 suggested 掩码时精度更差（完整切片 0.572）。在所有实验中，肿瘤体素较少的周边切片精度较低（mbIoU，<300：0.537 对 >=300：0.841）。将完整轴位 MRI 切片的最佳 oracle 分割结果堆叠后，肿瘤核心的平均 3D DSC 为 0.872，通过结合轴位、矢状位和冠状位掩码可提高到 0.919。结论：Segment Anything 基础模型虽然在照片上训练，但可以在 MRI 切片上实现高零样本精度的胶质瘤脑肿瘤分割。结果表明，当在临床应用中适当集成时，Segment Anything 可以加速并促进放疗治疗计划。

英文摘要：

Background: Tumor segmentation in MRI is crucial in radiotherapy (RT) treatment planning for brain tumor patients. Segment anything (SA), a novel promptable foundation model for autosegmentation, has shown high accuracy for multiple segmentation tasks but was not evaluated on medical datasets yet. Methods: SA was evaluated in a point-to-mask task for glioma brain tumor autosegmentation on 16744 transversal slices from 369 MRI datasets (BraTS 2020). Up to 9 point prompts were placed per slice. Tumor core (enhancing tumor + necrotic core) was segmented on contrast-enhanced T1w sequences. Out of the 3 masks predicted by SA, accuracy was evaluated for the mask with the highest calculated IoU (oracle mask) and with highest model predicted IoU (suggested mask). In addition to assessing SA on whole MRI slices, SA was also evaluated on images cropped to the tumor (max. 3D extent + 2 cm). Results: Mean best IoU (mbIoU) using oracle mask on full MRI slices was 0.762 (IQR 0.713-0.917). Best 2D mask was achieved after a mean of 6.6 point prompts (IQR 5-9). Segmentation accuracy was significantly better for high- compared to low-grade glioma cases (mbIoU 0.789 vs. 0.668). Accuracy was worse using MRI slices cropped to the tumor (mbIoU 0.759) and was much worse using suggested mask (full slices 0.572). For all experiments, accuracy was low on peripheral slices with few tumor voxels (mbIoU, <300: 0.537 vs. >=300: 0.841). Stacking best oracle segmentations from full axial MRI slices, mean 3D DSC for tumor core was 0.872, which was improved to 0.919 by combining axial, sagittal and coronal masks. Conclusions: The Segment Anything foundation model, while trained on photos, can achieve high zero-shot accuracy for glioma brain tumor segmentation on MRI slices. The results suggest that Segment Anything can accelerate and facilitate RT treatment planning, when properly integrated in a clinical application.

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。