Lee Jong Jin, Yang Hongye, Franc Benjamin L, Iagaru Andrei, Davidzon Guido A
Division of Nuclear Medicine and Molecular Imaging, Department of Radiology, Stanford University, 300 Pasteur Dr, Stanford, CA, 94305, USA.
Department of Nuclear Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea.
Eur J Nucl Med Mol Imaging. 2020 Dec;47(13):2992-2997. doi: 10.1007/s00259-020-04912-w. Epub 2020 Jun 17.
To evaluate the performance of deep learning (DL) classifiers in discriminating normal and abnormal F-FACBC (fluciclovine, Axumin®) PET scans based on the presence of tumor recurrence and/or metastases in patients with prostate cancer (PC) and biochemical recurrence (BCR).
A total of 251 consecutive F-fluciclovine PET scans were acquired between September 2017 and June 2019 in 233 PC patients with BCR (18 patients had 2 scans). PET images were labeled as normal or abnormal using clinical reports as the ground truth. Convolutional neural network (CNN) models were trained using two different architectures, a 2D-CNN (ResNet-50) using single slices (slice-based approach) and the same 2D-CNN and a 3D-CNN (ResNet-14) using a hundred slices per PET image (case-based approach). Models' performances were evaluated on independent test datasets.
For the 2D-CNN slice-based approach, 6800 and 536 slices were used for training and test datasets, respectively. The sensitivity and specificity of this model were 90.7% and 95.1%, and the area under the curve (AUC) of receiver operating characteristic curve was 0.971 (p < 0.001). For the case-based approaches using both 2D-CNN and 3D-CNN architectures, a training dataset of 100 images and a test dataset of 28 images were randomly allocated. The sensitivity, specificity, and AUC to discriminate abnormal images by the 2D-CNN and 3D-CNN case-based approaches were 85.7%, 71.4%, and 0.750 (p = 0.013) and 71.4%, 71.4%, and 0.699 (p = 0.053), respectively.
DL accurately classifies abnormal F-fluciclovine PET images of the pelvis in patients with BCR of PC. A DL classifier using single slice prediction had superior performance over case-based prediction.
评估深度学习（DL）分类器在基于前列腺癌（PC）伴生化复发（BCR）患者肿瘤复发和/或转移情况鉴别正常与异常F-FACBC（氟代脱氧胸苷，Axumin®）PET扫描中的性能。
2017年9月至2019年6月期间，共对233例BCR的PC患者（18例患者进行了2次扫描）进行了251次连续的F-氟代脱氧胸苷PET扫描。以临床报告作为金标准，将PET图像标记为正常或异常。使用两种不同架构训练卷积神经网络（CNN）模型，一种是使用单层面（基于层面的方法）的二维CNN（ResNet-50），以及相同的二维CNN和使用每个PET图像100个层面的三维CNN（ResNet-14）（基于病例的方法）。在独立测试数据集上评估模型性能。
对于基于二维CNN层面的方法，分别使用6800个和536个层面用于训练和测试数据集。该模型的敏感性和特异性分别为90.7%和95.1%，受试者操作特征曲线下面积（AUC）为0.971（p<0.001）。对于同时使用二维CNN和三维CNN架构的基于病例的方法，随机分配了100幅图像的训练数据集和28幅图像的测试数据集。基于二维CNN和三维CNN病例的方法鉴别异常图像的敏感性、特异性和AUC分别为85.7%、71.4%和0.750（p=0.013）以及71.4%、71.4%和0.699（p=0.053）。
DL能准确分类PC伴BCR患者骨盆的异常F-氟代脱氧胸苷PET图像。使用单层面预测的DL分类器比基于病例的预测具有更好的性能。
