Lunit INSIGHT MMG (v1.1.10)

Lunit INSIGHT MMG v1.1.10 is a computer-assisted detection/diagnosis (CADe/x) software device that uses artificial intelligence to identify and characterize suspicious areas for breast cancer on mammograms from compatible Full-Field Digital Mammography (FFDM) systems. The software is intended as an adjunctive tool to be reviewed by interpreting physicians after their initial mammogram read, not as a replacement for physician review or clinical judgment, and is used for screening mammograms in the female population.

Both the subject device (v1.1.10) and predicate device (v1.1.6) are radiological computer-assisted detection/diagnosis software powered by artificial intelligence and deep learning techniques that analyze FFDM images. The primary modifications in v1.1.10 include an updated AI model and removal of the 1-view Secondary Capture output mode, while maintaining the same indications for use, target patient population, intended users, and fundamental technological basis as the predicate.

Testing was conducted in accordance with IEC 62304:2006/A1:2016 (Medical device software – software life-cycle processes) and IEC 62366-1:2015+AMD1:2020 (Medical devices – Part 1: Application of usability engineering to medical devices). Additionally, the device was verified against Lunit's design control processes.

Lunit INSIGHT MMG v1.1.10 is substantially equivalent to predicate device v1.1.6 because both share identical indications for use, regulatory classification (Class II, 21 CFR 892.2090), product code (QDQ), target population, intended users, and fundamental AI/deep learning technology. Standalone performance testing on 2,412 mammograms demonstrated non-inferior safety and effectiveness, with ROC AUC of 0.9104 (95% CI: 0.896–0.925), and the updated AI model does not raise new questions of safety or effectiveness.