MammoScreen 3 is an artificial intelligence software tool designed to assist physicians in interpreting screening mammograms from digital mammography and digital breast tomosynthesis systems. It provides graphical marks indicating suspicious soft-tissue lesions or calcifications, along with a suspicion score and lesion characterization, helping radiologists in their diagnosis and reporting. The software is intended as an aid to, but not a replacement for, the physician's clinical judgment.
MammoScreen 3 is a concurrent reading and reporting aid for physicians interpreting screening mammograms. It is intended for use with compatible full-field digital mammography and digital breast tomosynthesis systems. The device can also use compatible prior examinations in the analysis. Output includes graphical marks of findings as soft-tissue lesions or calcifications with level of suspicion scores and lesion type characterized as mass/asymmetry, distortion, or calcifications. Location of findings is provided as adjunctive information to assist physicians during reporting.
MammoScreen 3 is AI-powered software using deep learning modules trained on large datasets of biopsy-proven breast cancer and normal tissue images for detection and characterization of suspicious findings. It processes 2D (FFDM) and 3D (DBT & 2DSM) mammograms, optionally incorporating prior exams (6-60 months old) in its analysis. The AI outputs suspicion scores, lesion types, and precise lesion locations (quadrant, depth, distance to nipple). It integrates results via a dedicated interface compatible with DICOM viewers and reporting systems.
A multi-reader multi-case clinical study with 23 radiologists and 240 mammogram cases evaluated MammoScreen 3's performance. Results showed significant improvement in radiologist diagnostic performance with the tool, with increased AUROC, sensitivity, and specificity compared to unaided reading. MammoScreen 3 standalone performance was superior to unaided radiologists and non-inferior to aided radiologists. Extensive subgroup analyses and standalone evaluations on over 7,500 exams support safety and effectiveness. Risk management and usability studies confirm that benefits outweigh residual risks.
No predicate devices specified
Submission
2/1/2024
FDA Approval
8/1/2024
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.