LaTIM, INSERM, UMR 1101, University Brest, Brest, France.
GIGA-CRC in vivo Imaging, University of Liège, Liège, Belgium.
Eur J Nucl Med Mol Imaging. 2021 Oct;48(11):3444-3456. doi: 10.1007/s00259-021-05244-z. Epub 2021 Mar 27.
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.
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).
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.
The proposed method could facilitate the deployment of a fully automated radiomics pipeline in such a challenging multicenter context.
在多中心具有异质特征的图像环境中，我们致力于在不依赖其他图像模态或额外先验约束的情况下，从正电子发射断层扫描（PET）图像中全自动确定肿瘤功能摄取。
在宫颈癌中，肿瘤摄取的位置是一个额外的挑战，它靠近甚至紧贴膀胱。利用来自五个机构的 232 名患者的 PET 数据集。为了避免不可靠的手动勾画，使用半自动方法生成真实值：首先手动确定包含肿瘤且不包含膀胱的体积，然后应用经过良好验证的基于模糊局部自适应贝叶斯（FLAB）算法的半自动方法生成真实值。我们基于 U-Net 架构的模型包含带有并发空间挤压和激励模块的残差块，以及可学习的非线性下采样和上采样块。实验依赖于交叉验证（四个机构用于训练和验证，第五个机构用于测试）。
该模型在不同机构之间具有较小的变异性，达到了良好的 Dice 相似系数（DSC）（0.80 ± 0.03），召回率（0.90 ± 0.05）高于精度（0.75 ± 0.05），并且优于标准 U-Net（DSC 0.77 ± 0.05，召回率 0.87 ± 0.02，精度 0.74 ± 0.08）。与 SUVmax 的 40%的固定阈值相比，两者的表现都要好得多（DSC 0.33 ± 0.15，召回率 0.52 ± 0.17，精度 0.30 ± 0.16）。在所有情况下，该模型都可以在不包含膀胱的情况下确定肿瘤摄取。既不需要形状先验，也不需要解剖信息来实现有效的训练。
该方法可以促进在这种具有挑战性的多中心环境中部署全自动放射组学管道。
