April 11, 2017 — Parascript LLC announced a new partnership with Volpara Solutions, creator of Volpara Density automated breast density assessment software. Independent research has associated an increased risk of cancer for patients with high breast density, which can also make cancer harder to detect on conventional mammograms. Together, Parascript AccuDetect CAD (computer-aided detection) and VolparaDensity provide complementary early cancer detection solutions.
Screening patients with dense breasts can pose additional challenges to healthcare providers since dense tissue has been shown to hide tumors and is associated with an increased risk of not only breast cancer, but more aggressive breast cancer. Accurate automated measurement of volumetric fibroglandular breast density assists radiologists in each patient’s personalized screening and decision making for additional diagnostics. Correctly identifying patients at low risk is also important to avoid unnecessary tests, because it can be stressful and costly for patients called back for further assessment. Parascript AccuDetect CAD, powered by deep learning, uses multiple independent cancer detection algorithms and a unique patented voting methodology to combine its findings. Comparing the results of the multiple image recognition processes allows for improved sensitivity and reduced false-positive rates.
In a study by the Karolinska Institute in Sweden of 41,102 women from KARMA (KARolinska MAmmography project for risk prediction of breast cancer) that looked at mammography screenings and clinical mammography at four hospitals in Sweden, VolparaDensity was validated across multiple mammography vendor platforms. This study concluded that automated measurement of volumetric mammographic density using VolparaDensity was a promising tool for breast cancer risk assessment.
In addition to more accurate detection, AccuDetect processes images more rapidly, which can improve response time to patients. AccuDetect CAD processes at 11 seconds per image, the fastest of all available U.S. Food and Drug Administration (FDA)-approved CAD systems, according to the company. The average CAD system takes 30 seconds or more to process a single image. AccuDetect CAD also supports early, more accurate detection, delivering high performance on dense and extremely dense breasts, according to a clinical study reported in Clinical Imaging (M. Lobbes et al., Clinical Imaging 37 (2013) 283-288).
Parascript and Volpara both exhibited at the Society for Breast Imaging (SBI) Breast Imaging Symposium in Los Angeles, held on April 6-8, 2017, and provided product demonstrations.