Genius AI Detection 2.0

Genius AI Detection 2.0 is a computer-aided detection and diagnosis software device that analyzes digital breast tomosynthesis (DBT) images to identify and mark suspicious lesions including masses, architectural distortions, asymmetries, and calcifications. The device provides confidence scores and a case score to aid radiologists in interpreting DBT exams in real-time concurrent reading, where the physician confirms or dismisses findings during the exam.

The subject device is functionally identical to the predicate (K221449) but adds compatibility with the Envision Mammography Platform in addition to Selenia Dimensions and 3Dimensions systems. New features include support for implant-displaced views on Selenia systems and a CC-MLO correlation algorithm for Envision cases. The Envision platform supports only high-resolution 1-mm and 6-mm smart slices, whereas Selenia supports standard and high-resolution modes.

IEC 62304:2015 (Medical device software – Software Life Cycle Processes) and ISO 14971:2019 (Medical Devices – Application of Risk Management to Medical Devices).

The device uses identical machine learning algorithms and mode of action as the predicate to aid detection and characterization of breast lesions. A standalone comparison study demonstrated non-inferior performance on Envision images versus Selenia images (AUC difference of -0.0017, p=0.87), with equivalent sensitivity, specificity, and false marker rates. Additional validation confirmed the CC-MLO correlation algorithm achieves 90% accuracy on malignant lesions and maintains equivalent clinical workflow, supporting the conclusion that safety and effectiveness are unchanged despite platform expansion.