AUTHOR=Zhang Xiang , Liang Ming , Yang Zehong , Zheng Chushan , Wu Jiayi , Ou Bing , Li Haojiang , Wu Xiaoyan , Luo Baoming , Shen Jun 

TITLE=Deep Learning-Based Radiomics of B-Mode Ultrasonography and Shear-Wave Elastography: Improved Performance in Breast Mass Classification

JOURNAL=Frontiers in Oncology

VOLUME=Volume 10 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2020.01621

DOI=10.3389/fonc.2020.01621

ISSN=2234-943X

ABSTRACT=Objective: Shear-wave elastography (SWE) can improve the diagnostic specificity of the B-model ultrasonography (US) in breast cancer. However, whether deep learning-based radiomics signatures based on the B-mode US (B-US-RS) or SWE (SWE-RS) could further improve the diagnostic performance remains to be investigated. We aim to develop the B-US-RS and SWE-RS and determine their performances in classifying breast masses. Materials and Methods: This retrospective study included 263 women (mean age ± standard deviation, 40.9 years ± 12.3) who had US-visible solid breast masses and underwent biopsy or/and surgical resection between June 2015 and May 2017. B-mode US and SWE images of the masses in 198 patients (training cohort) were segmented respectively to construct B-US-RS and SWE-RS using the least absolute shrinkage and selection operator regression and tested in an independent validation cohort of 65 patients. The performances of B-US-RS and SWE-RS were assessed using receiver operating characteristic (ROC) analysis and compared with that of radiologist assessment (Breast Imaging Reporting and Data System [BI-RADS]) and quantitative SWE parameters (E max , E mean , E ratio , and E SD ) by using the McNemar test.The single best-performing quantitative SWE parameter E max had a higher specificity than BI-RADS assessment in these two cohorts (P < 0.001 for both). The areas under the ROC curves (AUCs) of B-US-RS and SWE-RS both were 0.99 (95% CI: 0.99, 1.00) in the training cohort and 1.00 (95% CI: 1.00, 1.00) in the validation cohort. The specificities of B-US-RS and SWE-RS were higher than that of E max in training (P < 0.001 for both) and validation cohorts (P = 0.02 for both).The B-US-RS and SWE-RS outperformed the quantitative SWE parameters and BI-RADS assessment for classifying breast masses. The integration of the deep learning-based radiomics approach would help improve the classification ability of B-mode US and SWE for breast masses.