Department of Ultrasound, Harbin Medical University Cancer Hospital, No. 150, Haping Road, Nangang District, Harbin, Heilongjiang Province, China.
School of Computer Science and Technology, Harbin Institute of Technology, No. 92, Xidazhi Street, Nangang District, Harbin, Heilongjiang Province, China.
Eur Radiol. 2022 Oct;32(10):7163-7172. doi: 10.1007/s00330-022-08836-x. Epub 2022 Apr 30.
To develop novel deep learning network (DLN) with the incorporation of the automatic segmentation network (ASN) for morphological analysis and determined the performance for diagnosis breast cancer in automated breast ultrasound (ABUS).
A total of 769 breast tumors were enrolled in this study and were randomly divided into training set and test set at 600 vs. 169. The novel DLNs (Resent v2, ResNet50 v2, ResNet101 v2) added a new ASN to the traditional ResNet networks and extracted morphological information of breast tumors. The accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), area under the receiver operating characteristic (ROC) curve (AUC), and average precision (AP) were calculated. The diagnostic performances of novel DLNs were compared with those of two radiologists with different experience.
The ResNet34 v2 model had higher specificity (76.81%) and PPV (82.22%) than the other two, the ResNet50 v2 model had higher accuracy (78.11%) and NPV (72.86%), and the ResNet101 v2 model had higher sensitivity (85.00%). According to the AUCs and APs, the novel ResNet101 v2 model produced the best result (AUC 0.85 and AP 0.90) compared with the remaining five DLNs. Compared with the novice radiologist, the novel DLNs performed better. The F1 score was increased from 0.77 to 0.78, 0.81, and 0.82 by three novel DLNs. However, their diagnostic performance was worse than that of the experienced radiologist.
The novel DLNs performed better than traditional DLNs and may be helpful for novice radiologists to improve their diagnostic performance of breast cancer in ABUS.
• A novel automatic segmentation network to extract morphological information was successfully developed and implemented with ResNet deep learning networks. • The novel deep learning networks in our research performed better than the traditional deep learning networks in the diagnosis of breast cancer using ABUS images. • The novel deep learning networks in our research may be useful for novice radiologists to improve diagnostic performance.
开发新的深度学习网络（DLN），并结合自动分割网络（ASN），进行形态学分析，并确定在自动乳腺超声（ABUS）中用于诊断乳腺癌的性能。
本研究共纳入 769 个乳腺肿瘤，随机分为训练集和测试集，分别为 600 个和 169 个。新型 DLN（Resent v2、ResNet50 v2、ResNet101 v2）在传统的 ResNet 网络中添加了新的 ASN，提取了乳腺肿瘤的形态信息。计算了准确性、敏感性、特异性、阳性预测值（PPV）、阴性预测值（NPV）、接收器工作特征（ROC）曲线下面积（AUC）和平均精度（AP）。将新型 DLN 的诊断性能与两位具有不同经验的放射科医生进行了比较。
ResNet34 v2 模型的特异性（76.81%）和 PPV（82.22%）均高于其他两个模型，ResNet50 v2 模型的准确性（78.11%）和 NPV（72.86%）较高，ResNet101 v2 模型的敏感性（85.00%）较高。根据 AUC 和 AP，新型 ResNet101 v2 模型的结果（AUC 为 0.85，AP 为 0.90）优于其余五个 DLN。与新手放射科医生相比，新型 DLN 的表现更好。三个新型 DLN 将 F1 评分从 0.77 提高到 0.78、0.81 和 0.82。然而，它们的诊断性能逊于经验丰富的放射科医生。
新型 DLN 的表现优于传统 DLN，可能有助于新手放射科医生提高 ABUS 中乳腺癌的诊断性能。
成功开发并实现了一种新的自动分割网络，用于从 ResNet 深度学习网络中提取形态信息。
与 ABUS 图像乳腺癌诊断相比，我们研究中的新型深度学习网络的性能优于传统深度学习网络。
我们研究中的新型深度学习网络可能有助于新手放射科医生提高诊断性能。
