brAIn™ Shoulder Positioning

brAIn™ Shoulder Positioning is a cloud-based software tool that assists shoulder surgeons in preoperative planning and visualization of primary total shoulder replacement surgeries. It uses CT-scan imaging to automatically segment shoulder anatomy (scapula and humerus) via machine learning, displays interactive 3D reconstructions, allows surgeons to position anatomical and reverse implants, and generates a planning report with measurements and visual documentation.

The subject device performs automatic segmentation of shoulder anatomy, whereas the predicate device required manual segmentation. brAIn™ includes semi-automatic landmark measurements and implant placement, displays both axial and coronal 2D DICOM views (vs. axial-only in predicate), shows soft tissue visualization in the 3D viewer, and captures patient age, height, and weight in addition to predicate information fields. The underlying principles, web-based architecture, CT-scan input, pre-operative use timing, and implant types planned remain equivalent.

ANSI AAMI IEC 62304:2006/A1:2016 (Medical device software – Software life cycle processes); FDA Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices; FDA Content of Premarket Submission for Management of Cybersecurity in Medical Devices.

The subject device has the same intended use as the predicate (preoperative planning and visualization for total shoulder replacement) and uses identical imaging input (CT-scans) with the same implant types and patient population. Although brAIn™ employs automatic rather than manual segmentation and includes additional measurement and visualization features, validation testing demonstrates that the automatic segmentation meets a Dice Similarity Coefficient of ≥0.95 versus manual segmentation, measurement accuracy matches the predicate's performance, and landmark positioning achieves equivalent accuracy. These technological improvements do not raise new safety or effectiveness concerns—they represent enhancements that maintain or exceed predicate performance without introducing new risks.