Saige-Density (2.5.0)

Saige-Density is a software application that analyzes full-field digital mammography (FFDM) and digital breast tomosynthesis (DBT) images using deep learning to automatically assign an ACR BI-RADS 5th Edition breast density category (A through D). The software processes mammogram studies and outputs two DICOM files: a structured report containing the density category and a secondary capture report for display on mammography workstations. It is intended as an adjunctive tool to aid interpreting physicians in assessing breast tissue composition, not as a diagnostic aid, for female patients age 35 and older.

Saige-Density v2.5.0 uses the same proprietary deep learning algorithm as the predicate device (v2.0.0) with no changes to algorithmic components. The key differences are minor technological upgrades: addition of a new single-execution mode deployment option, support for cloud-enabled deployment environments, and improvements to processing speed and memory efficiency.

Not stated in this summary.

The device is substantially equivalent because it maintains identical indications for use, patient population, intended users, and principles of operation as the predicate Saige-Density v2.0.0. Both devices are designed for the same intended use (breast density assessment via deep learning on FFDM/DBT), produce output in the same format (BI-RADS density categories), and are not diagnostic aids. Since the algorithms are unchanged and only technological deployment aspects and performance optimizations differ, these changes do not alter safety or effectiveness or raise different questions about either. Verification and validation testing confirmed the software meets its requirements.