Saige-Dx (3.1.0)
DeepHealth, Inc.Saige-Dx (3.1.0) is an AI-based software designed to assist radiologists by analyzing digital breast tomosynthesis mammograms to detect suspicious soft tissue lesions and calcifications indicative of cancer. It evaluates 3D and 2D mammogram images, assigning suspicion levels for findings and cases, and produces reports to support concurrent reading. This helps clinicians improve detection accuracy and reduce reading time in breast cancer screening for women aged 35 and older.
Intended Use
Saige-Dx analyzes digital breast tomosynthesis (DBT) mammograms to identify the presence or absence of soft tissue lesions and calcifications that may be indicative of cancer. For a given DBT mammogram, Saige-Dx analyzes the DBT image stacks and the accompanying 2D images, including full field digital mammography and/or synthetic images. The system assigns a Suspicion Level, indicating the strength of suspicion that cancer may be present, for each detected finding and for the entire case. The outputs of Saige-Dx are intended to be used as a concurrent reading aid for interpreting physicians on screening mammograms with compatible DBT hardware.
Technology
Saige-Dx is a software device that processes screening mammograms using artificial intelligence. It takes as input a set of x-ray mammogram DICOM files from a single digital breast tomosynthesis (DBT) study, analyzing both the DBT image stacks and associated 2D images (full-field digital mammography and/or synthetic images). It outputs bounding boxes around detected findings with a Finding Suspicion Level for malignancy and produces a Case Suspicion Level for the overall study. Results are encapsulated in DICOM Structured Report (SR) and Secondary Capture (SC) objects for integration into clinical workflow.
Performance
Verification included software unit, integration, system, and regression testing to ensure the software meets requirements and is substantially equivalent to the predicate device. The algorithm was trained on a diverse dataset from multiple vendors, totaling 141,768 patients and 316,166 studies with racially and geographically diverse populations. Validation included multi-reader multi-case studies and standalone testing on Hologic and GE images including unilateral breasts and breast implants. All tests met pre-specified criteria supporting safety and effectiveness.
Predicate Devices
No predicate devices specified
Device Timeline
- 1
Submission
11/29/2024
20 days - 2
FDA Approval
12/19/2024
Other devices from DeepHealth, Inc.
- TechLive2025 • 510K
- Saige-Density (2.5.0)2024 • 510K
- Saige-Dx2024 • 510K
Ready to Sharpen Your Edge?
Join hundreds of your peers who rely on RadAI Slice. Get the essential weekly briefing that empowers you to navigate the future of radiology.
We respect your privacy. Unsubscribe at any time.