Dec. 3—Confirma's CADstream system's patented "worst curve" algorithm for analyzing kinetics may assist in determining malignancy, according to a study presented by Lilian Wang, MD, of the radiology resident program at the Seattle Cancer Care Alliance and University of Washington in Seattle, at the 93rd scientific assembly and annual meeting of the Radiological Society of North America (RSNA) in Chicago last week.
The study, “MRI Detected Suspicious Breast Findings: Comparison of Kinetic Features Measured by Computer-aided Evaluation in Benign and Malignant Lesions,” focused on the evaluation of kinetics and found that the most suspicious curve as identified by computer-aided detection (CAD) was significantly different between benign and malignant lesions. Confirma highlighted the study because its product, CADstream, was used in the study.
Wang and her colleagues retrospectively examined 125 suspicious lesions (42 malignant, 83 benign) for which CAD and subsequent image-guided biopsy had been performed, out of a pool of 1,538 breast MRI examinations conducted from November 2004 to November 2006.
The study compared three distinct CAD kinetic features of suspicious breast MRI lesions to determine which could best predict benign or malignant outcomes. The study concluded that, of the kinetic features analyzed, only the most suspicious curve type identified by CADstream was significantly different between benign and malignant lesions.
The findings support the recommendation of the ACR’s Breast Imaging Reporting and Database System Atlas for breast MRI to report the “worst looking” curve. The study found that any washout enhancement was associated with malignancy in nearly half of lesions.