Saige-Dx (3.1.0)

Saige-Dx is artificial intelligence software that analyzes digital breast tomosynthesis (DBT) mammograms to detect soft tissue lesions and calcifications potentially indicative of cancer. It outputs suspicion levels for individual findings and an overall case suspicion level to aid radiologists in screening mammography interpretation, intended for women 35 and older.

Saige-Dx v3.1.0 is a software-only device using machine learning algorithms to analyze mammography images from multiple vendors (GE, Hologic). Compared to predicate v3.0.0, the subject device adds cloud-enabled deployment capability and includes performance upgrades for speed and memory optimization, but maintains identical indications for use and similar technical operation principles.

ISO 14971:2019 (risk management), IEC 62304:2015 (software lifecycle processes), NEMA PS3 (DICOM standard), FDA Guidance on Premarket Submissions for Software in Medical Devices (May 2005), and FDA Guidance on Software as a Medical Device (SAMD) Clinical Evaluation (December 2017).

Both devices share identical indications for use, patient population, and software-based AI/machine learning architecture for detecting and characterizing mammography findings. The minor differences—cloud deployment consideration and performance optimizations—do not alter intended use or clinical function. Verification and validation testing on independent datasets confirmed the subject device meets pre-specified performance criteria on Hologic and GE exams without unintentional differences from the predicate, establishing that design changes do not affect safety or effectiveness.