文献速递肿瘤 PET/CT

卷积神经网络实现PET功能体积全自动分割:多中心开发与验证

Convolutional neural networks for PET functional volume fully automatic segmentation: development and validation in a multi-center setting.
PubMed 文献 · Eur J Nucl Med Mol Imaging · 2021-03-27

本研究针对多中心异质性PET图像,开发基于U-Net的卷积神经网络模型,用于宫颈癌肿瘤功能摄取的全自动分割,无需其他模态或先验约束。模型在五家机构232例患者中交叉验证,DSC为0.80±0.03,优于标准U-Net和固定SUVmax 40%阈值,且能避免膀胱纳入。

要点

  • 模型在多中心宫颈癌PET中DSC达0.80±0.03,跨机构变异性小。
  • 模型召回率0.90±0.05高于精确率0.75±0.05,且均优于标准U-Net。
  • 模型可自动排除膀胱,无需形状先验或解剖信息。

核心结论

  • 基于U-Net的卷积神经网络模型可在多中心异质性PET图像中实现宫颈癌肿瘤功能摄取的全自动分割,且不依赖其他模态或先验约束。

研究设计与样本

  • 研究利用五家机构232例宫颈癌患者的PET数据集,采用交叉验证(四家机构训练与验证,第五家测试)。
  • 参考标准通过半自动方法生成:先手动确定包含肿瘤并排除膀胱的体积,再应用基于模糊局部自适应贝叶斯(FLAB)算法的半自动方法。
  • 模型基于U-Net架构,整合残差块、空间挤压与激发模块以及可学习的非线性下采样和上采样块。

主要发现

  • 模型在各机构间DSC为0.80±0.03,变异性小;召回率0.90±0.05高于精确率0.75±0.05。
  • 标准U-Net的DSC为0.77±0.05,召回率0.87±0.02,精确率0.74±0.08。
  • 固定SUVmax 40%阈值的DSC为0.33±0.15,召回率0.52±0.17,精确率0.30±0.16。
  • 在所有情况下,模型确定肿瘤摄取时均未包含膀胱。

关键数值与推荐

  • 模型DSC 0.80±0.03,召回率0.90±0.05,精确率0.75±0.05。
  • 标准U-Net DSC 0.77±0.05,召回率0.87±0.02,精确率0.74±0.08。
  • 固定SUVmax 40%阈值DSC 0.33±0.15,召回率0.52±0.17,精确率0.30±0.16。

临床怎么用

  • 该全自动分割方法有望促进放射组学流程在多中心挑战性环境中的部署。
  • 模型无需形状先验或解剖信息即可有效训练,且能自动排除膀胱,简化临床工作流。

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: In this work, we addressed fully automatic determination of tumor functional uptake from positron emission tomography (PET) images without relying on other image modalities or additional prior constraints, in the context of multicenter images with heterogeneous characteristics. METHODS: In cervical cancer, an additional challenge is the location of the tumor uptake near or even stuck to the bladder. PET datasets of 232 patients from five institutions were exploited. To avoid unreliable manual delineations, the ground truth was generated with a semi-automated approach: a volume containing the tumor and excluding the bladder was first manually determined, then a well-validated, semi-automated approach relying on the Fuzzy locally Adaptive Bayesian (FLAB) algorithm was applied to generate the ground truth. Our model built on the U-Net architecture incorporates residual blocks with concurrent spatial squeeze and excitation modules, as well as learnable non-linear downsampling and upsampling blocks. Experiments relied on cross-validation (four institutions for training and validation, and the fifth for testing). RESULTS: The model achieved good Dice similarity coefficient (DSC) with little variability across institutions (0.80 ± 0.03), with higher recall (0.90 ± 0.05) than precision (0.75 ± 0.05) and improved results over the standard U-Net (DSC 0.77 ± 0.05, recall 0.87 ± 0.02, precision 0.74 ± 0.08). Both vastly outperformed a fixed threshold at 40% of SUVmax (DSC 0.33 ± 0.15, recall 0.52 ± 0.17, precision 0.30 ± 0.16). In all cases, the model could determine the tumor uptake without including the bladder. Neither shape priors nor anatomical information was required to achieve efficient training. CONCLUSION: The proposed method could facilitate the deployment of a fully automated radiomics pipeline in such a challenging multicenter context.

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

引用

  1. Iantsen A, Ferreira M, Lucia F, et al. Convolutional neural networks for PET functional volume fully automatic segmentation: development and validation in a multi-center setting. Eur J Nucl Med Mol Imaging. 2021. PMID: 33772335. doi:10.1007/s00259-021-05244-z

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。