MammoScreen is an AI-based software that assists physicians in interpreting full-field digital mammograms (FFDM) by identifying suspicious breast lesions such as soft tissue lesions and calcifications. It provides marks on the mammogram images alongside a suspicion score to help detect potential breast cancer, supporting radiologists during their reading process to improve cancer detection without replacing clinical judgment.
MammoScreen is intended for use as a concurrent reading aid for interpreting physicians, to help identify findings on screening FFDM acquired with compatible mammography systems and assess their level of suspicion.
MammoScreen comprises a software-only system with machine learning algorithms, including deep learning modules trained on large biopsy-proven datasets to detect suspicious calcifications and soft tissue lesions. It operates as a DICOM Web compliant node, processing FFDM images from mammography systems, and provides output as marks and a suspicion score on a 1-10 scale via a user interface.
Clinical and non-clinical testing show MammoScreen is safe and effective. A clinical multi-reader multi-case study with 14 radiologists and 240 mammograms demonstrated improved radiologist performance with MammoScreen (AUC increased from 0.77 to 0.80, p=0.035), with significant improvement at breast and lesion levels. Standalone performance was comparable to radiologists. Non-clinical testing included software verification, beta validation, and hazard analysis.
No predicate devices specified
Submission
10/4/2019
FDA Approval
3/25/2020
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.