Saige-Density (2.5.0)

Saige-Density is a medical image processing software designed to provide objective breast density categorization. It leverages deep learning to analyze full-field digital mammography and digital breast tomosynthesis images, offering supplementary information for physicians in accordance with the ACR BI-RADS Atlas 5th Edition.

DeepHealth, Inc. has received authorization to market its medical image processing software, Saige-Density (version 2.5.0), concluding that this Class II device is substantially equivalent to a previously marketed version. Saige-Density employs deep learning to analyze full-field digital mammography (FFDM) and digital breast tomosynthesis (DBT) images, providing an ACR BI-RADS Atlas 5th Edition breast density category. This output serves as supplementary information for physicians assessing breast tissue composition and is not intended as a standalone diagnostic aid, thereby enhancing the precision of breast density classification for clinical use. To maintain its market authorization, DeepHealth, Inc. must comply with a range of ongoing regulatory requirements, including general controls such as annual registration, device listing, and good manufacturing practices, as well as specific Class II device obligations like the Quality System regulation and Unique Device Identification (UDI) Rule. The company is advised that any modifications to the device may necessitate further regulatory review and must adhere to all applicable federal statutes, ensuring continuous oversight and safety in the medical device market.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How does the deep learning algorithm in Saige-Density specifically analyze FFDM and DBT images to determine breast density categories?; What validation studies or clinical data support the accuracy and reliability of Saige-Density's breast density assessment compared to human expert interpretation?; What are the limitations or potential biases of Saige-Density's deep learning model, particularly concerning diverse patient populations or image acquisition variations?; How does the supplementary information provided by Saige-Density impact physicians' clinical decision-making regarding breast cancer screening and risk assessment?; What are the specific ongoing data collection and performance monitoring requirements for Saige-Density to ensure its continued efficacy and safety in the market?