The MOE Key Laboratory of Modern Acoustics, Department of Physics, Nanjing University, Nanjing, 210093, China.
Department of Ultrasound, The First Affiliated Hospital of Nanjing Medical University, Nanjing, 210029, China.
Sci Rep. 2021 Oct 14;11(1):20474. doi: 10.1038/s41598-021-00018-x.
Triple negative (TN) breast cancer is a subtype of breast cancer which is difficult for early detection and the prognosis is poor. In this paper, 910 benign and 934 malignant (110 TN and 824 NTN) B-mode breast ultrasound images were collected. A Resnet50 deep convolutional neural network was fine-tuned. The results showed that the averaged area under the receiver operating characteristic curve (AUC) of discriminating malignant from benign ones were 0.9789 (benign vs. TN), 0.9689 (benign vs. NTN). To discriminate TN from NTN breast cancer, the AUC was 0.9000, the accuracy was 88.89%, the sensitivity was 87.5%, and the specificity was 90.00%. It showed that the computer-aided system based on DCNN is expected to be a promising noninvasive clinical tool for ultrasound diagnosis of TN breast cancer.
三阴性乳腺癌是一种难以早期发现且预后较差的乳腺癌亚型。本文收集了 910 例良性和 934 例恶性（110 例 TN 和 824 例 NTN）B 型乳腺超声图像。经过微调 Resnet50 深度卷积神经网络。结果表明，判别良恶性的受试者工作特征曲线下面积（AUC）平均值分别为 0.9789（良性与 TN）、0.9689（良性与 NTN）。判别 TN 与 NTN 乳腺癌的 AUC 为 0.9000，准确率为 88.89%，灵敏度为 87.5%，特异性为 90.00%。这表明基于 DCNN 的计算机辅助系统有望成为一种有前途的超声诊断 TN 乳腺癌的无创临床工具。
