MammoSightAI

MammoSightAI is software that uses artificial intelligence to analyze full-field digital mammography (FFDM) and digital breast tomosynthesis (DBT) screening mammograms. It generates a priority tag indicating whether a mammogram is suspicious for malignancy, enabling radiologists to prioritize cases in their worklist. The software is intended for passive notification only and does not provide diagnostic information; radiologists remain responsible for final interpretation and clinical decision-making.

Both MammoSightAI and predicate device Saige-Q are software-only devices using AI algorithms that receive DICOM mammogram data, preprocess and analyze images, and output priority tags to PACS/workstation systems. Both operate on FFDM and DBT mammograms in parallel with standard-of-care workflow without removing cases from the worklist or marking specific image regions. Minor differences exist in processing time and output format, but no notable technological differences affect safety and effectiveness.

Not stated in this summary.

MammoSightAI is substantially equivalent to Saige-Q because both devices share identical indications for use, target the same intended users (radiologists) and patient populations, utilize the same imaging modalities (FFDM and DBT), and employ AI algorithms for triage and prioritization. Clinical performance testing demonstrates comparable safety and effectiveness profiles, with MammoSightAI achieving AUC of 0.963 overall (0.975 FFDM, 0.953 DBT) versus predicate's 0.966 FFDM and 0.985 DBT. Minor implementation differences do not introduce new safety concerns nor impact device performance when used as labeled.