Breast MR CAD: Earlier Detection, Increased Efficiency

David Gruen, MD, of Norwalk Radiology & Mammography Center in Norwalk, Conn., uses CADstream from Confirma for breast MRI on up to 10 patients a day.

“Once we find an abnormality, we’ve got to make a decision on whether it’s something to ignore or worry about.” Two factors help with that determination: the lesion’s kinetics and its morphology. “Before CAD, for breast MR there were proponents of the morphology school of thought,” Gruen says. They believed that only the shape of lesions mattered. Particularly in Europe, the school of thought was more geared to kinetics, such as a lesion’s vascularity, how fast it takes up contrast, and whether contrast washes out. Experts realized both were important and the challenge became how to visualize and evaluate both appropriately.

Clinicians used to look at kinetics by finding each image on the sequence—which could take up to a half-hour per lesion. “CADstream automated that process,” Gruen says. As breast MR has gotten better, with improved spatial and temporal resolution, physicians can find lesions just 2 or 3 millimeters in size. “CAD programs let you quickly evaluate a huge set of data, up to 3,000 images. You can’t be doing high-quality MRI without a CAD package.”

The rate of false-positive biopsies has gone down dramatically over the past couple of years, Gruen says. And, he is finding smaller cancers more readily, particularly ductal carcinoma in situ (DCIS). CADstream lets him detect DCIS without manually performing asymmetric enhancement. Another important gain is that Gruen can’t remember the last time a patient was detected with stage 3 or 4 cancer. While that is partly attributable to a good screening rate and breast cancer awareness, “patients are showing up in our practice with small cancers that are easily treatable. As the technology gets better, we will continue to find smaller cancers.”

Robin Shermis, MD, is one of three radiologists at Northwest Ohio Dedicated Breast MRI Center in Toledo using AuroraCAD from Aurora Imaging Technology. The software is “extremely powerful, very efficient and provides for standardized interpretation,” he says. “It’s been nothing less than tremendous.”

The easy-to-learn software makes it hard to miss even subtle cancers, Shermis says. That, plus educating the referring physicians, has helped the facility increase its referral patterns. “As soon as you’re done scanning, the images in CAD are almost immediately available,” he says. “We’re able to go over the case with the patient, do a second look and even biopsy at that time, if necessary.”

Terri-Ann Gizienski, MD, breast imaging director at a community imaging center that’s part of the University of Pittsburgh Medical Center (UPMC) in Pennsylvania, has been using DynaCAD from Invivo Corp. for almost four years.

She particularly appreciates the software’s biopsy and intervention feature, which helps her better define which lesions require biopsy. It also makes biopsies easier to perform. “It will quickly let you know if the coordinates are not acceptable and you cannot approach the lesion from that angle.”

The software has helped her become more efficient with her reading, spending less time manipulating images. Plus, she can import and export images to and from other facilities. “I can communicate results more quickly and referring doctors can see what I’m talking about.” She also can show patients exactly what she is going to biopsy. For patients being treated for breast cancer, she can show them exactly where the cancer begins and ends. “CAD makes it easy to compare before and after images and show them that their tumor is shrinking.”

Gizienski performs about 10 breast MR studies each week, a figure that has been increasing. “Women are becoming more aware of the capability of breast MRI and referring physicians and surgeons see the utility in it.” 

Beth Walsh,

Editor

Editor Beth earned a bachelor’s degree in journalism and master’s in health communication. She has worked in hospital, academic and publishing settings over the past 20 years. Beth joined TriMed in 2005, as editor of CMIO and Clinical Innovation + Technology. When not covering all things related to health IT, she spends time with her husband and three children.

Trimed Popup
Trimed Popup