Genius AI Detection 2.0 is a software device that uses deep learning AI models to analyze digital breast tomosynthesis (DBT) images. It identifies and marks suspicious soft tissue densities and calcifications, providing confidence scores to aid radiologists during breast cancer screening. The AI improves detection specificity and reduces false positives, supporting clinicians in interpreting 3D mammography exams efficiently and accurately.
Computer-aided detection and diagnosis software device intended for use with compatible digital breast tomosynthesis (DBT) systems to identify and mark regions of interest including soft tissue densities (masses, architectural distortions, and asymmetries) and calcifications in DBT exams, providing confidence scores and aiding interpretation concurrently with reading of the exam.
Uses deep learning networks analyzing 1-mm DBT tomosynthesis slices and 6-mm SmartSlices; outputs lesion location, outlines, and confidence scores packaged as DICOM Mammography CAD SR object for display on compliant workstations. Recent version uses a more sophisticated Convolutional Neural Network (CNN) replacing an Artificial Neural Network (ANN) to improve micro-calcification classification. Operates directly on processed tomosynthesis reconstructed slices, eliminating the need for raw projections.
Verification testing confirmed software requirements via unit, integration, and system testing. Validation tested stand-alone performance on a sequestered multi-site dataset of 764 DBT exams including 106 biopsy-proven cancers. Compared to the predicate device, the updated AI algorithm significantly improved specificity by 12% with maintained sensitivity, demonstrating improved detection performance and safety.
No predicate devices specified
Submission
5/18/2022
FDA Approval
10/6/2022
Join hundreds of your peers who rely on RadAI Slice. Get the essential weekly briefing that empowers you to navigate the future of radiology.
We respect your privacy. Unsubscribe at any time.