Abstract
Background Deep learning (DL) algorithms are gaining extensive attention for their excellent performance in image recognition tasks. DL models can automatically make a quantitative assessment of complex medical image characteristics and achieve increased accuracy in diagnosis with higher efficiency. Purpose To determine the feasibility of using a DL approach to predict clinically negative axillary lymph node metastasis from US images in patients with primary breast cancer. Materials and Methods A data set of US images in patients with primary breast cancer with clinically negative axillary lymph nodes from Tongji Hospital (974 imaging studies from 2016 to 2018, 756 patients) and an independent test set from Hubei Cancer Hospital (81 imaging studies from 2018 to 2019, 78 patients) were collected. Axillary lymph node status was confirmed with pathologic examination. Three different convolutional neural networks (CNNs) of Inception V3, Inception-ResNet V2, and ResNet-101 architectures were trained on 90% of the Tongji Hospital data set and tested on the remaining 10%, as well as on the independent test set. The performance of the models was compared with that of five radiologists. The models' performance was analyzed in terms of accuracy, sensitivity, specificity, receiver operating characteristic curves, areas under the receiver operating characteristic curve (AUCs), and heat maps. Results The best-performing CNN model, Inception V3, achieved an AUC of 0.89 (95% confidence interval [CI]: 0.83, 0.95) in the prediction of the final clinical diagnosis of axillary lymph node metastasis in the independent test set. The model achieved 85% sensitivity (35 of 41 images; 95% CI: 70%, 94%) and 73% specificity (29 of 40 images; 95% CI: 56%, 85%), and the radiologists achieved 73% sensitivity (30 of 41 images; 95% CI: 57%, 85%; P = .17) and 63% specificity (25 of 40 images; 95% CI: 46%, 77%; P = .34). Conclusion Using US images from patients with primary breast cancer, deep learning models can effectively predict clinically negative axillary lymph node metastasis. Artificial intelligence may provide an early diagnostic strategy for lymph node metastasis in patients with breast cancer with clinically negative lymph nodes. Published under a CC BY 4.0 license. Online supplemental material is available for this article. See also the editorial by Bae in this issue.
Highlights
N The three deep learning models—Inception V3, Inception-ResNet V2, and ResNet-101—achieved 85%, 78%, and 73% sensitivity (P = .40) and 73%, 75%, and 73% specificity (P = .96), respectively, in predicting lymph node metastasis with an independent external test set compared with 73% sensitivity (P = .51) and 63% specificity (P = .62) from a consensus of five radiologists
The clinical US diagnoses were made by 11 radiologists from Tongji Hospital and three radiologists from Hubei Cancer Hospital according to standard protocols [11]
Performance of Deep Learning Models The deep learning models achieved good performance in predicting lymph node metastasis with the use of the primary breast cancer US images of test set A, with AUCs of 0.90 for the Inception V3 model, 0.89 for the Inception-ResNet V2 model, and 0.87 for the ResNet-101 model (P = .44, .33, and .38 for Inception V3 vs Inception-ResNet V2, Inception V3 vs ResNet-101, and Inception-ResNet V2 vs ResNet-101, respectively)
Summary
From October 2018 to April 2019, an independent external test cohort of 78 patients (mean age, 46 years; range, 30–74 years) from Hubei Cancer Hospital, Hubei, China, was screened with the same criteria used for the primary cohort. US images were obtained from the breast imaging databases at Tongji Hospital and Hubei Cancer Hospital. The clinical US diagnoses were made by 11 radiologists from Tongji Hospital and three radiologists from Hubei Cancer Hospital according to standard protocols [11]. One or two of the most representative images were selected by three radiologists from Tongji Hospital (L.Q.Z., X.W.C., and G.G.W., each with 5–6 years of experience) for image quality control based on the pathologic results.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
