Center News

Enhancing the image for tumor detection

New software enhancement of breast MRI scans helps radiologists reduce false positives

Aug. 1, 2007
Dr. Connie Lehman

Dr. Connie Lehman was one of the corresponding authors of a study that reports computer-aided enhancement software has the potential to improve the discrimination of benign and malignant breast MRI lesions.

University of Washington Photo

Using commercially available software to enhance breast scans done by magnetic resonance imaging reduces the number of false-positive identifications of malignant tumors and the subsequent need for biopsies, according to a new study.

Dr. Teresa Williams and colleagues at the Seattle Cancer Care Alliance and the University of Washington Medical Center did a retrospective examination of 154 breast lesions deemed suspicious by radiologists that were only visible on MRI and that had been biopsied under MRI guidance. They compared the findings and recommendations made by radiologists at the time to new findings using computer-aided enhancement software to enhance and evaluate the visible response to contrast agents absorbed by breast tissue.

The study, published in the July edition of the journal Radiology, reports 23 percent fewer false positives with CAE software set to its highest enhancement level. Williams conducted the research while serving as a medical resident in radiology at the Seattle Cancer Care Alliance. She is now a fellow in pediatric radiology at Children's Hospital and Regional Medical Center in Seattle.

"In summary, our findings suggest that CAE has the potential to improve the discrimination of benign and malignant breast MRI lesions," the authors said. "We believe that CAE is useful as a tool to supplement the radiologist's subjective interpretation, but should not be relied upon exclusively to guide management."

"There are challenges associated with breast MRI and one is the time it takes to process and evaluate the many images acquired," said Dr. Connie Lehman, one of the study's co-authors and director of radiology at the SCCA. "Computer software programs such as the one evaluated in our study can assist us in interpreting breast MRI scans more easily. Our study suggests that the information provided may improve our ability to distinguish between benign and malignant lesions."

Currently, radiologists use MRI scans in addition to mammography to get a better view of tissue they suspect may be malignant. MRI as an adjunct to mammography also is standard practice at the SCCA for women who are at high risk for breast cancer and to examine the other, or contralateral, breast of women with a new diagnosis.

One particular challenge in breast MRI is the interpretation of the morphology and kinetic features — the amount of contrast agent absorbed by breast tissue over time — on multiple-imaging series. Typically, a woman will receive one scan without contrast agent and two more after the administration of the contrast. One key analysis function performed by CAE is automatic kinetic assessment.

"The detailed CAE lesion kinetic information differs substantially from information radiologists obtain by conventional manual placement of a region of interest," the authors wrote. This is because CAE generates detailed data for the entire lesion versus only the portion of the lesion that the region-of-placement highlights.

Between 2001-2004, researchers identified and biopsied lesions from 125 women ages 27-86. They used CADstream™ 3.0, a CAE system developed by Confirma, Inc. of Kirkland, Wash for processing. The presence of CAE threshold enhancement was sensitive for malignancy in 38 of the 41 malignant lesions examined using the software, according to the study. However, the software did not perform perfectly; it failed to confirm the malignancy of the three lesions. "Given the presence of three false-negative lesions, a finding deemed suspicious by the radiologist should be further evaluated regardless of the enhancement features determined by CAE," according to the study.

Williams said she advocates the use of CAE software analysis of MRI scans as an aid to radiologists' interpretations. "The software is already commercially available and it has shown it is useful in reducing the false-positive rate of breast MRI," she said.

Fred Hutchinson Cancer Research Center is a world leader in research to prevent, detect and treat cancer and other life-threatening diseases.