The image quality and lesion characterization of breast using automated whole-breast ultrasound: A comparison with handheld ultrasound

Objective: To prospectively evaluate the image quality of automated whole breast ultrasonography (AWUS) in the characterization of breast lesions compared with handheld breast ultrasonography (HHUS).

Materials and methods: This prospective study included a total of 411 lesions in 209 women. All patients underwent both HHUS and AWUS prior to biopsy. An evaluation of identical image pairs of 411 lesions obtained from both modalities was performed, and the image quality of AWUS was compared with that of HHUS as a reference standard. The overall image quality was evaluated for lesion coverage, lesion conspicuity, and artifact effect using a graded score. Additionally, the factors that correlated with differences in image quality between the two modalities were analyzed.

Results: In 97.1%, the image quality of AWUS was identical or superior to that of HHUS, whereas AWUS was inferior in 2.9%. In only 0.5%, the poor quality of AWUS images caused by incomplete lesion coverage and shadowing due to a contact artifact inhibited precise interpretations. The two main causes resulting in degraded AWUS image quality were blurring of the margin (83.3%) and acoustic shadowing by Cooper's ligament or improper compression pressure of the transducer (66.7%). Among various factors, peripheral location from the nipple (p=0.01), lesion size (p=0.02), shape descriptor (p=0.02), and final American College of Radiology Breast Imaging Reporting and Data System (BI-RADS) category (p=0.001) were correlated with differences in image quality between AWUS and HHUS.

Conclusion: Although the image quality of AWUS was comparable to that of HHUS for lesion interpretation, HHUS was clearly superior to AWUS for analyzing lesions with a peripheral location, an irregular shape, a non-circumscribed margin and BI-RADS category 4 or 5.