Automated breast ultrasound should not replace handheld ultrasound in a diagnostic breast center but rather should be used in conjunction with it or play a role in screening breast ultrasound, according to a study presented at the 2008 RSNA meeting.
Automated breast ultrasound should not replace handheld ultrasound in a diagnostic breast center but rather should be used in conjunction with it or play a role in screening breast ultrasound, according to a study presented at the 2008 RSNA meeting.
Dr. Linda Hovanessian Larsen, an associate professor of radiology at the University of Southern California, and colleagues retrospectively reviewed 87 women presenting for breast ultrasound as part of their diagnostic imaging workup. One of three mammographers interpreted each study acquired by the ABUS, accompanied by the clinical history and mammogram. The mammographer was blinded to the results of handheld ultrasound. Another mammographer interpreted the handheld ultrasound study with the clinical history and mammogram but was blinded to ABUS.
The researchers found a total of 78 lesions. Of the 78, 91% of the lesions found on handheld ultrasound were also seen on ABUS. Of the 9% of lesions seen only on handheld ultrasound, four were categorized as Breast Imaging and Reporting Data System (BIRADS) 4 and biopsied. Pathology identified two high-risk lesions and two benign lesions. On ABUS, 19 additional lesions not found on handheld ultrasound were identified and 11 were given a BIRADS 4 or 5. When the lesions were evaluated with handheld ultrasound, nine were not reproducible. The researchers biopsied the lesions and found them to be benign.
ABUS should not replace handheld ultrasound in a diagnostic breast center because ABUS did not identify 9% of the lesions seen on handheld ultrasound, two of which were high-risk, according to the researchers. All of the additional lesions found on ABUS were either not found on handheld ultrasound or considered benign.
While ABUS should not be used on its own, it may prove useful in conjunction with handheld ultrasound to identify lesions, Larsen said.
Study: Mammography AI Leads to 29 Percent Increase in Breast Cancer Detection
February 5th 2025Use of the mammography AI software had a nearly equivalent false positive rate as unassisted radiologist interpretation and resulted in a 44 percent reduction in screen reading workload, according to findings from a randomized controlled trial involving over 105,000 women.