A Multicenter Hospital-Based Diagnosis Study of Automated Breast Ultrasound System in Detecting Breast Cancer among Chinese Women

Xi Zhang,Xi Lin,Yanjuan Tan,Ying Zhu,Hui Wang,Ruimei Feng,Guoxue Tang,Xiang Zhou,Anhua Li,Youlin Qiao
DOI: https://doi.org/10.21147/j.issn.1000-9604.2018.02.06
2018-01-01
Abstract:Objective: The automated breast ultrasound system (ABUS) is a potential method for breast cancer detection; however, its diagnostic performance remains unclear. We conducted a hospital-based multicenter diagnostic study to evaluate the clinical performance of the ABUS for breast cancer detection by comparing it to handheld ultrasound (HHUS) and mammography (MG). Methods: Eligible participants underwent HHUS and ABUS testing; women aged 40-69 years additionally underwent MG. Images were interpreted using the Breast Imaging Reporting and Data System (BI-RADS). Women in the BI-RADS categories 1-2 were considered negative. Women classified as BI-RADS 3 underwent magnetic resonance imaging to distinguish true-and false-negative results. Core aspiration or surgical biopsy was performed in women classified as BI-RADS 4-5, followed by a pathological diagnosis. Kappa values and agreement rates were calculated between ABUS, HHUS and MG. Results: A total of 1,973 women were included in the final analysis. Of these, 1,353 (68.6%) and 620 (31.4%) were classified as BI-RADS categories 1-3 and 4-5, respectively. In the older age group, the agreement rate and Kappa value between the ABUS and HHUS were 94.0% and 0.860 (P < 0.001), respectively; they were 89.2% and 0.735 (P < 0.001) between the ABUS and MG, respectively. Regarding consistency between imaging and pathology results, 78.6% of women classified as BI-RADS 4-5 based on the ABUS were diagnosed with precancerous lesions or cancer; which was 7.2% higher than that of women based on HHUS. For BI-RADS 1-2, the false-negative rates of the ABUS and HHUS were almost identical and were much lower than those of MG. Conclusions: We observed a good diagnostic reliability for the ABUS. Considering its performance for breast cancer detection in women with high-density breasts and its lower operator dependence, the ABUS is a promising option for breast cancer detection in China.
What problem does this paper attempt to address?