Yeh Joe Chao-Yuan, Yu Wei-Hsiang, Yang Cheng-Kun, Chien Ling-I, Lin Ko-Han, Huang Wen-Sheng, Hsu Po-Kuei
aetherAI, Co., Ltd., Taipei.
Department of Nursing, Taipei Veterans General Hospital, Taipei.
Ann Transl Med. 2021 Jan;9(1):37. doi: 10.21037/atm-20-1419.
The presence of lymphovascular invasion (LVI) and perineural invasion (PNI) are of great prognostic importance in esophageal squamous cell carcinoma. Currently, positron emission tomography (PET) scans are the only means of functional assessment prior to treatment. We aimed to predict the presence of LVI and PNI in esophageal squamous cell carcinoma using PET imaging data by training a three-dimensional convolution neural network (3D-CNN).
Seven hundred and ninety-eight PET scans of patients with esophageal squamous cell carcinoma and 309 PET scans of patients with stage I lung cancer were collected. In the first part of this study, we built a 3D-CNN based on a residual network, ResNet, for a task to classify the scans into esophageal cancer or lung cancer. In the second stage, we collected the PET scans of 278 patients undergoing esophagectomy for a task to classify and predict the presence of LVI/PNI.
In the first part, the model performance attained an area under the receiver operating characteristic curve (AUC) of 0.860. In the second part, we randomly split 80%, 10%, and 10% of our dataset into training set, validation set and testing set, respectively, for a task to classify the scans into the presence of LVI/PNI and evaluated the model performance on the testing set. Our 3D-CNN model attained an AUC of 0.668 in the testing set, which shows a better discriminative ability than random guessing.
A 3D-CNN can be trained, using PET imaging datasets, to predict LNV/PNI in esophageal cancer with acceptable accuracy.
淋巴管浸润（LVI）和神经周围浸润（PNI）的存在在食管鳞状细胞癌中具有重要的预后意义。目前，正电子发射断层扫描（PET）是治疗前唯一的功能评估手段。我们旨在通过训练三维卷积神经网络（3D-CNN），利用PET成像数据预测食管鳞状细胞癌中LVI和PNI的存在情况。
收集了798例食管鳞状细胞癌患者的PET扫描图像和309例I期肺癌患者的PET扫描图像。在本研究的第一部分，我们基于残差网络ResNet构建了一个3D-CNN，用于将扫描图像分类为食管癌或肺癌的任务。在第二阶段，我们收集了278例接受食管切除术患者的PET扫描图像，用于分类和预测LVI/PNI存在情况的任务。
在第一部分中，模型性能在受试者工作特征曲线（AUC）下的面积达到0.860。在第二部分中，我们将数据集的80%、10%和10%分别随机分为训练集、验证集和测试集，用于将扫描图像分类为LVI/PNI存在情况的任务，并在测试集上评估模型性能。我们的3D-CNN模型在测试集中的AUC为0.668，显示出比随机猜测更好的判别能力。
使用PET成像数据集可以训练3D-CNN，以可接受的准确性预测食管癌中的LNV/PNI。
