Parascript, LLC, an image analysis and pattern recognition technology provider, announced that its AccuDetect® Galileo computer-aided detection software for digital mammography showed better overall performance in detecting breast cancer in a recent retrospective study against iCad's SecondLook. Findings from the study, performed at Maastricht University Medical Center departments of Radiology and Surgery in the Netherlands, with collaboration from University of Udine's Institute of Diagnostic Radiology in Italy, were presented at the annual meeting of the European Society of Radiology, ECR, on March 3.
In the study, digital mammograms of 326 patients were analyzed (117 patients with biopsy proven breast cancer, 209 negative cases) using AccuDetect Galileo 4.0 and SecondLook version 7.2. AccuDetect Galileo significantly increased true positive fraction (TPF) of cancer cases when compared to SecondLook. It demonstrated a per image increase of 6.9% to 72.2%; per case increase of 4.3% to 84.6%. The University of Maastricht team noted that AccuDetect Galileo had a significant performance improvement in detecting soft tissue densities on extremely dense breasts (BI-RADS breast density class 4) over SecondLook, increasing TPF by15.4% to 69.2%. Dense breast tissue can obscure an underlying cancer, or conversely mimic one that does not exist, thus making accurate detection difficult.
"We are encouraged by the results of this new study," said Yuri Prizemin, director of business development for medical imaging for Parascript. "We believe that Parascript CAD advancements in marking malignant lesions on extremely dense breasts will help radiologists to improve breast cancer detection."
Full findings from the study were presented by the study authors M. Lobbes, K. Keymeulen, M. Smidt, R.G. Beets-Tan, J.E. Wildberger, and C. Boetes from Maastricht University and R. Girometti and C. Zuiani from University of Udine in Retrospective Comparison of the Accuracy of Two Different Computer-aided Detection Systems for Detecting Malignant Lesions on Mammography.
About Parascript, LLC
The Parascript image analysis suite extracts meaningful information from images. Employing patented digital image analysis and pattern recognition technologies, the Parascript image analysis suite improves decision quality in medical imaging, postal and payment automation, fraud detection and forms processing operations. Parascript software processes billions of documents per year. Fortune 500 companies, postal operators, major government and financial institutions rely on Parascript products. Organizations include the U.S. Postal Service, Bell + Howell, Fiserv, Elsag, Lockheed Martin, NCR, Siemens and Burroughs.