CHICAGO, November 13, 2020 – Newly published clinical evidence in Radiology, a journal by the Radiological Society of North America (RSNA), demonstrates the effectiveness of Qlarity Imaging’s QuantX Diagnostic Artificial Intelligence (AI) to distinguish between benign and cancerous lesions on breast MRI exams. In the De Novo clinical trial conducted by researchers from the Department of Radiology at the University of Chicago, which employed a multireader, multicase design, the diagnostic AI software significantly improved the accuracy of 100% of the readers in the study.
“Many studies have shown the benefits of using dynamic contrast enhanced (DCE) MRI over other imaging modalities in detecting and diagnosing breast cancer, but various factors, including subjective interpretation and complexity of cases, can hinder radiologist performance in evaluating MRI exams,” said Dr. Gillian Newstead, one of the study’s authors and advisor to Qlarity Imaging. “Our goal was to demonstrate how AI can be used as a decision support tool to address challenges with reading breast MRI cases and improve diagnostic accuracy for patients who are anxiously awaiting their results.”
Setting this study apart from previous evaluations of AI systems, the De Novo trial—in support of the first diagnostic AI software approved by the U.S. Food and Drug Administration (FDA)—used a large sample of readers from varying key demographic backgrounds and an enriched dataset, including a mix of lesion types, that was analyzed in a sequential-read protocol. The first read was performed using tools that mimic those available on current breast MRI CAD workstations. The second-read condition featured diagnostic AI insights from QuantX.
Numed, a well established company in business since 1975 provides a wide range of service options including time & material service, PM only contracts, full service contracts, labor only contracts & system relocation. Call 800 96 Numed for more info.
Serving as the primary measurement of accuracy, the average area under the receiver operating characteristic curve (AUC) increased more than 20%, from 0.71 to 0.76 (p=0.05), with the concurrent introduction of the AI system. In addition to improving overall clinical accuracy, the mean sensitivity increased from 90.4% to 94.2% without a statistically significant decrease in specificity and false negatives, i.e. missed cancers, decreased by 39%.
“The findings from this study reinforce our vision of transforming breast imaging with AI technology, starting with our diagnostic AI solution for breast MRI,” said Jon DeVries, chief executive officer of Qlarity Imaging. “By helping radiologists gain clinical clarity, we aim to enhance diagnostic performance and ultimately improve the quality of care for patients in need.”