CAD Software for Digital Mammography Employs Algorithmic Voting for Better Performance

A next-generation computer-aided detection (CAD) software was developed to help radiologists improve breast cancer detection. The system makes significant performance improvements over its predecessor and additionally reduces the possibility for false-positive rates when detecting suspicious lesions on mammograms.

Parascript, LLC (Longmont, CO, USA), an image analysis and pattern recognition technology provider, announced the release of AccuDetect 5.0, which is designed to work with the leading full field digital mammography (FFDM) and computed radiography (CR) systems. The software received CE marking approval and is installed in multiple radiology centers in France, Italy, and The Netherlands. It is also used by the Russian Center for Roentgeno-Radiology Research (RNCRR; Moscow, Russia), Russia’s largest, most recognized breast cancer oncology center.

Available for FFDM vendors interested in reducing false-positive rates of existing CAD systems, AccuDetect is intended to help radiologists in the early detection of breast cancer during mammography screening exams. The technology has been developed using a large database of digital images from leading digital mammography systems. It uses several proprietary complementary algorithms to identify the presence of suspicious lesions on mammogram images. Proprietary voting technology combines the detection results from each algorithm. The result is an innovative CAD product with high sensitivity and low false-positive rates.

“We are proud to leverage our expertise in image analysis and pattern recognition to help radiologists detect suspicious lesions on mammograms. We are especially excited, at Parascript, to report the significant strides and multiple deployments we have accomplished this year with AccuDetect 5.0,” said Yuri Prizemin, director of business development for medical imaging for Parascript.

CAD systems are typically utilized to identify and highlight hard-to-find anomalies on medical images. The technology enables suspicious lesions to be brought to the attention of radiologists. AccuDetect automatically identifies and distinctly marks suspicious areas, enabling a more accurate interpretation of mammograms and increased detection of calcifications and masses.

The Parascript image analysis suite extracts important data 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 over 100 billion imaged documents per year.

Related Links:
Parascript