Department of Radiology, Tsukuba Medical Center, 1-3-1 Amakubo, Tsukuba, Ibaraki, 305-0005, Japan.
Department of Radiology, Faculty of Medicine, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki, 305-8575, Japan.
Eur J Radiol. 2021 Feb;135:109471. doi: 10.1016/j.ejrad.2020.109471. Epub 2020 Dec 5.
To compare deep learning with radiologists when diagnosing uterine cervical cancer on a single T2-weighted image.
This study included 418 patients (age range, 21-91 years; mean, 50.2 years) who underwent magnetic resonance imaging (MRI) between June 2013 and May 2020. We included 177 patients with pathologically confirmed cervical cancer and 241 non-cancer patients. Sagittal T2-weighted images were used for analysis. A deep learning model using convolutional neural networks (DCNN), called Xception architecture, was trained with 50 epochs using 488 images from 117 cancer patients and 509 images from 181 non-cancer patients. It was tested with 60 images for 60 cancer and 60 non-cancer patients. Three blinded experienced radiologists also interpreted these 120 images independently. Sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC) were compared between the DCNN model and radiologists.
The DCNN model and the radiologists had a sensitivity of 0.883 and 0.783-0.867, a specificity of 0.933 and 0.917-0.950, and an accuracy of 0.908 and 0.867-0.892, respectively. The DCNN model had an equal to, or better, diagnostic performance than the radiologists (AUC = 0.932, and p for accuracy = 0.272-0.62).
Deep learning provided diagnostic performance equivalent to experienced radiologists when diagnosing cervical cancer on a single T2-weighted image.
比较深度学习与放射科医生在单张 T2 加权图像上诊断宫颈癌的效能。
本研究纳入了 2013 年 6 月至 2020 年 5 月期间进行磁共振成像（MRI）检查的 418 例患者（年龄 21-91 岁，平均 50.2 岁）。我们纳入了 177 例经病理证实的宫颈癌患者和 241 例非癌症患者。分析矢状 T2 加权图像。使用卷积神经网络（DCNN）的深度学习模型（称为 Xception 架构），使用 117 例癌症患者的 50 张图像和 181 例非癌症患者的 509 张图像进行 50 个 epoch 的训练。用 60 张癌症患者图像和 60 张非癌症患者图像对其进行测试。3 名经验丰富的放射科医生也独立对这 120 张图像进行了解读。比较 DCNN 模型与放射科医生之间的敏感性、特异性、准确性和受试者工作特征曲线下面积（AUC）。
DCNN 模型和放射科医生的敏感性分别为 0.883 和 0.783-0.867，特异性分别为 0.933 和 0.917-0.950，准确性分别为 0.908 和 0.867-0.892。DCNN 模型的诊断性能与放射科医生相当或更好（AUC=0.932，准确性的 p 值为 0.272-0.62）。
在单张 T2 加权图像上诊断宫颈癌时，深度学习提供的诊断性能与经验丰富的放射科医生相当。
