Breast ultrasound is a highly useful tool in the hands of an expert, but it is notoriously operator-dependent. Systems that offer partial or full automation to minimize performance variability are gaining attention, as trial data accumulate and system designs evolve.
Breast ultrasound is a highly useful tool in the hands of an expert, but it is notoriously operator-dependent. Systems that offer partial or full automation to minimize performance variability are gaining attention, as trial data accumulate and system designs evolve.
An investigational semiautomated screening whole-breast ultrasound system increased the cancer detection rate of screening mammography by 100% in women with dense breasts and known risk factors in a blinded study of nearly 4000 patients. Study results were presented at the RSNA meeting by the system's inventor, Dr. Kevin Kelly.
The SonoCiné unit adjusts some functions normally performed by an operator in handheld ultrasound, such as transducer speed and breast position. The operator still controls pressure on the breast and the angle of incidence of the transducer.
The study group had 34 cancers, including 10 found by both ultrasound and mammography, 16 found by ultrasound alone, and five detected by mammography alone. Three cancers were not found with either technique. Of the total number of cancers, 26 were node negative.
Given the cancer risk factors in the study group, researchers expected a cancer detection rate of 1.5 per 400, but breast ultrasound increased this to about 3 per 400. These results were achieved with just a 2% biopsy rate.
"These additional cancers presumably represent tumors that would have presented one or two years later, either clinically or mammographically," said Kelly, director of breast imaging at the Hill Breast Center in Pasadena, CA.
The results indicate that ultrasound helps detect tumors at earlier stages than mammography. Of 10 cancers smaller than 1 cm, nine were found on ultrasound alone and one was found by both ultrasound and mammography. Of 20 stage I cancers, six were found on both studies, 11 were found on ultrasound alone, and just one cancer was found on mammography alone.
Also at the RSNA meeting, two studies showcased the performance of SomoVu, a fully automated breast ultrasound system commercially available from U-Systems.
In one multicenter trial of 177 breasts in 165 patients, findings on mammography with automated breast ultrasound (ABUS) agreed with mammography plus handheld ultrasound in 94% of breasts analyzed.
Another study performed at the Elizabeth Wende Breast Clinic in Rochester, NY, found that in women with dense breasts, ABUS helps spot additional findings that otherwise would have gone undetected in mammography screening.
With handheld ultrasound, a technologist performs the scan with input and/or monitoring from the physician, or the physician personally performs the ultrasound study. In contrast, with ABUS, parameters are preset, technologists perform the scan, and the radiologist reads the cases on a workstation later, after the patient has left the premises.
Dr. Stamatia Destounis, the study's lead author, noted the potential staffing benefits of the device.
"The bottom line is there are not enough of us to go around. Mammography is not well reimbursed, and not enough young people are entering breast imaging fellowships. Breast imagers are already overworked, and we have to get smarter about how to do things," she said.
Researchers examined use of automated ultrasound as an adjunct to screening mammography in 147 women.
Ultrasound yielded 95 findings, most of which researchers determined were benign, often with the help of prior imaging studies and case histories. Cysts made up the predominant findings. Six patients were recalled, and two were biopsied.
Earlier versions of the SomoVu system imaged the patient while standing, but since then it has become evident that visualization is better when the patient is lying down. There have also been multiple software upgrades.
"If [vendors] keep improving quality, automated breast ultrasound will become a helpful additional tool for screening for breast cancer detection," Destounis said.
FDA Clears AI-Powered Ultrasound Software for Cardiac Amyloidosis Detection
November 20th 2024The AI-enabled EchoGo® Amyloidosis software for echocardiography has reportedly demonstrated an 84.5 percent sensitivity rate for diagnosing cardiac amyloidosis in heart failure patients 65 years of age and older.
FDA Clears Updated AI Platform for Digital Breast Tomosynthesis
November 12th 2024Employing advanced deep learning convolutional neural networks, ProFound Detection Version 4.0 reportedly offers a 50 percent improvement in detecting cancer in dense breasts in comparison to the previous version of the software.