Breast MRI CAD zeroes in on malignant lesions

June 3, 2008

Results from two recent studies suggest computer-aided detection for breast MRI may be closing in on true cancer detection.

Results from two recent studies suggest computer-aided detection for breast MRI may be closing in on true cancer detection.

Researchers from Penn Diagnostics, a breast MRI CAD company, and George Washington University tested the ONCAD system for contrast-enhanced MRI. They reviewed images of 102 biopsy-proven breast cancers and 131 cancer-free breasts from six major U.S. research institutions. The investigators found ONCAD could identify up to 96% of true malignancies.

In another study, researchers from New York University and Siemens Medical Solutions presented results of a computer classification system designed to improve breast MRI BI-RADS classification. Forty patients were included in this study. The investigators found the automated classification tool yielded more accurate BI-RADS classifications than three radiologists blinded to results. Both studies were released at the 2008 ARRS meeting.