文献速递肿瘤 PET/CT

卷积神经网络实现肿瘤FDG-PET血池SUV全自动测定以促进SUR计算

A convolutional neural network for fully automated blood SUV determination to facilitate SUR computation in oncological FDG-PET.
PubMed 文献 · Eur J Nucl Med Mol Imaging · 2020-10-01

本研究开发基于U-Net卷积神经网络的主动脉腔全自动勾画方法,用于肿瘤FDG-PET/CT中血池标准化摄取值(BSUV)的测定,从而促进标准摄取值比(SUR)的计算。使用946例来自多个中心的FDG PET/CT扫描(训练366例,测试580例),以手动勾画为参考。测试结果显示自动与手动BSUV高度一致,平均相对差异为-0.5%±2.2%(95%CI [-5.1,3.8]%),4例测试病例的ROI不可用(<1 ml)。结论认为CNN可稳健实现基于图像的BSUV自动测定,集成到PET数据处理流程将显著促进SUR计算而不增加临床工作量。

要点

  • U-Net卷积神经网络可实现主动脉腔全自动勾画以测定血池SUV(BSUV)。
  • 自动与手动BSUV高度一致,平均相对差异-0.5%±2.2%(95%CI [-5.1,3.8]%)。
  • 自动BSUV集成到PET流程可促进SUR计算且不增加临床工作量。

核心结论

  • 卷积神经网络能够稳健地实现基于图像的BSUV全自动测定。
  • 将自动BSUV推导集成到PET数据处理流程将显著促进SUR计算而不增加临床工作量。

研究设计与样本

  • 采用U-Net架构的卷积神经网络进行主动脉腔自动勾画。
  • 使用来自多个中心的946例FDG PET/CT扫描,其中366例用于网络训练、580例用于测试。
  • 所有扫描均进行手动主动脉腔勾画作为参考,勾画时避开运动诱导的衰减伪影区域或邻近FDG高摄取区域的潜在溢出。
  • 通过测试数据中自动与手动BSUV的分数偏差评估网络性能。

主要发现

  • 训练后的U-Net所得BSUV与手动勾画所得高度一致。
  • 自动与手动BSUV的平均相对差异(均值±标准差)为-0.5%±2.2%,95%置信区间为[-5.1,3.8]%,总范围为[-10.0,12.0]%。
  • 4例测试病例所得ROI不可用(<1 ml)。

关键数值与推荐

  • 自动与手动BSUV的平均相对差异为-0.5%±2.2%(95%CI [-5.1,3.8]%),总范围[-10.0,12.0]%。
  • 4例测试病例的ROI不可用,定义为<1 ml。

临床怎么用

  • 可将CNN自动BSUV测定集成到PET数据处理流程中以促进SUR计算。
  • 自动方法可避免手动ROI勾画所需的小心和额外工作量,使SUR方法更适用于临床常规。

局限性

  • 4例测试病例所得ROI不可用(<1 ml)。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

PURPOSE: The standardized uptake value (SUV) is widely used for quantitative evaluation in oncological FDG-PET but has well-known shortcomings as a measure of the tumor's glucose consumption. The standard uptake ratio (SUR) of tumor SUV and arterial blood SUV (BSUV) possesses an increased prognostic value but requires image-based BSUV determination, typically in the aortic lumen. However, accurate manual ROI delineation requires care and imposes an additional workload, which makes the SUR approach less attractive for clinical routine. The goal of the present work was the development of a fully automated method for BSUV determination in whole-body PET/CT. METHODS: Automatic delineation of the aortic lumen was performed with a convolutional neural network (CNN), using the U-Net architecture. A total of 946 FDG PET/CT scans from several sites were used for network training (N = 366) and testing (N = 580). For all scans, the aortic lumen was manually delineated, avoiding areas affected by motion-induced attenuation artifacts or potential spillover from adjacent FDG-avid regions. Performance of the network was assessed using the fractional deviations of automatically and manually derived BSUVs in the test data. RESULTS: The trained U-Net yields BSUVs in close agreement with those obtained from manual delineation. Comparison of manually and automatically derived BSUVs shows excellent concordance: the mean relative BSUV difference was (mean ± SD) = (- 0.5 ± 2.2)% with a 95% confidence interval of [- 5.1,3.8]% and a total range of [- 10.0, 12.0]%. For four test cases, the derived ROIs were unusable (< 1 ml). CONCLUSION: CNNs are capable of performing robust automatic image-based BSUV determination. Integrating automatic BSUV derivation into PET data processing workflows will significantly facilitate SUR computation without increasing the workload in the clinical setting.

来源:Eur J Nucl Med Mol Imaging,查看原文。版权归原作者所有。

引用

  1. Nikulin P, Hofheinz F, Maus J, et al. A convolutional neural network for fully automated blood SUV determination to facilitate SUR computation in oncological FDG-PET. Eur J Nucl Med Mol Imaging. 2020. PMID: 33006022. doi:10.1007/s00259-020-04991-9

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。