Saige-Q

K203517

DeepHealth, Inc. · cleared 2021-04-16 · product code QFM · Radiology

Premarket evidence — what FDA accepted

Device typesamd
source quote (p.4)
“Saige-Q is a software workflow device that processes Digital Breast Tomosynthesis (DBT) and Full-Field Digital Mammography (FFDM) screening mammograms using artificial intelligence to act as a prioritization tool for interpreting radiologists.”
Algorithmartificial intelligence algorithm, deep neural networks
source quote (p.3)
“Saige-Q uses an artificial intelligence algorithm to generate a code for a given mammogram, indicative of the software's suspicion that the mammogram contains at least one suspicious finding. The preprocessed images become the input to the AI algorithm, which generates the Saige-Q code using deep neural networks that have been trained on large numbers of mammograms where cancer status is known.”
Adaptive (vs locked)No
source quote (p.5)
“The preprocessed images become the input to the AI algorithm, which generates the Saige-Q code using deep neural networks that have been trained on large numbers of mammograms where cancer status is known.”
PCCPFDA source did not state this
Cybersecurity addressedFDA source did not state this

Validation studies (2)

Retrospective clinical

n=1,333 cases · 8 site(s)

endpoints: assess the sensitivity and specificity of Saige-Q relative to radiologist performance; assess the processing time performance

Retrospective clinical

n=1,528 cases · 6 site(s)

endpoints: assess the sensitivity and specificity of Saige-Q relative to radiologist performance; assess the processing time performance

Reported performance (14 observations)

sensitivity0.922CI [90.2%, 93.8%]
source quote (p.8)
“The primary endpoint for FFDM was successfully met with Saige-Q demonstrating a specificity at 86.9% sensitivity of 92.2% (95% CI: [90.2%, 93.8%])”
specificity0.912CI [88.4%, 93.4%]
source quote (p.8)
“and a sensitivity at 88.9% specificity of 91.2% (95%: [88.4%, 93.4%]).”
aurocas written: “auc”0.966CI [0.957, 0.975]
source quote (p.8)
“In the FFDM study, Saige-Q achieved an overall area under the receiver operating characteristic curve (AUC) of 0.966 (95% CI: [0.957, 0.975]).”
aurocas written: “AUC for DBT”0.985CI [0.979, 0.990]
source quote (p.8)
“In the DBT study, Saige-Q achieved an overall AUC of 0.985 (95% CI: [0.979, 0.990]) on the DBT data.”
sensitivityas written: “Sensitivity for DBT”0.983CI [97.3%, 99.0%]
source quote (p.8)
“The primary endpoint for DBT was also successfully met with Saige-Q demonstrating a specificity at 86.9% sensitivity: 98.3% (95% CI: [97.3%, 99.0%])”
specificityas written: “Specificity for DBT”0.957CI [93.6%, 97.2%]
source quote (p.8)
“and a sensitivity at 89.9% specificity of 95.7% (95% CI: [93.6%, 97.2%]).”
aurocas written: “AUC for FFDM on soft tissue densities”0.964CI [0.954, 0.974]
source quote (p.8)
“For instance, on FFDM, Saige-Q achieved an AUC of 0.964 (95% CI: [0.954, 0.974]) on soft tissue densities”
aurocas written: “AUC for FFDM on calcifications”0.973CI [0.958, 0.988]
source quote (p.9)
“and an AUC of 0.973 (95% CI: [0.958, 0.988]) on calcifications.”
aurocas written: “AUC for FFDM on dense breasts”0.959CI [0.945, 0.973]
source quote (p.9)
“For breast density, Saige-Q achieved an AUC of 0.959 (95% CI: [0.945, 0.973]) on dense breasts”
aurocas written: “AUC for FFDM on non-dense breasts”0.972CI [0.961, 0.984]
source quote (p.9)
“and an AUC of 0.972 (95% CI: [0.961, 0.984]) on non-dense breasts for FFDM exams.”
aurocas written: “AUC for DBT on soft tissue densities”0.983CI [0.977, 0.990]
source quote (p.9)
“For DBT, Saige-Q achieved an AUC of 0.983 (95% CI: [0.977, 0.990]) on soft tissue densities”
aurocas written: “AUC for DBT on calcifications”0.989CI [0.983, 0.996]
source quote (p.9)
“and an AUC of 0.989 (95% CI: [0.983, 0.996]) on calcifications.”
aurocas written: “AUC for DBT on dense breasts”0.98CI [0.971, 0.988]
source quote (p.9)
“For DBT, Saige-Q achieved an AUC of 0.980 (95% CI: [0.971, 0.988]) on dense breasts”
aurocas written: “AUC for DBT on non-dense breasts”0.988CI [0.981, 0.996]
source quote (p.9)
“and an AUC of 0.988 (95% CI: [0.981, 0.996]) on non-dense breasts.”

Each value carries its own analysis unit and task — never compare or pool across devices. Source: 510(k) summary PDF.

Predicate network

Postmarket — what happened after clearance

0
recalls in product code, 24mo
0
MAUDE reports in code, 12mo
—
vs code's own 3-yr baseline
6
drift signals on this device
  • re_clearance

    The FDA AI/ML device list shows a newer 510(k) K251873 (decision 2025-08-11) from DeepHealth, Inc. for a matching device line ("Saige-Dx") — a new clearance for the same line is a change event.

    first seen 2026-07-08 · k_number:K251873

  • re_clearance

    The FDA AI/ML device list shows a newer 510(k) K243705 (decision 2024-12-19) from DeepHealth, Inc for a matching device line ("Saige-Density (2.5.0)") — a new clearance for the same line is a change event.

    first seen 2026-07-08 · k_number:K243705

  • re_clearance

    The FDA AI/ML device list shows a newer 510(k) K243688 (decision 2024-12-19) from DeepHealth, Inc. for a matching device line ("Saige-Dx (3.1.0)") — a new clearance for the same line is a change event.

    first seen 2026-07-08 · k_number:K243688

  • re_clearance

    The FDA AI/ML device list shows a newer 510(k) K241747 (decision 2024-11-18) from DeepHealth, Inc for a matching device line ("Saige-Dx") — a new clearance for the same line is a change event.

    first seen 2026-07-08 · k_number:K241747

  • re_clearance

    The FDA AI/ML device list shows a newer 510(k) K222275 (decision 2022-12-16) from DeepHealth, Inc. for a matching device line ("Saige-Density") — a new clearance for the same line is a change event.

    first seen 2026-07-08 · k_number:K222275

  • re_clearance

    The FDA AI/ML device list shows a newer 510(k) K220105 (decision 2022-05-12) from DeepHealth, Inc. for a matching device line ("Saige-Dx") — a new clearance for the same line is a change event.

    first seen 2026-07-08 · k_number:K220105

Recall and MAUDE counts are product-code-level (reports aren't reliably attributable to one device). Signals are descriptive observables with sources — never a judgment that the device is unsafe or drifting. Snapshot 2026-07-08.

Reimbursement — how devices like this got paid

Not yet tracked — no payment pathway indexed for this clearance (the reimbursement corpus is a growing seed set).

RIGOR™ Precedent · public FDA/CMS data · descriptive decision-support, not regulatory or reimbursement advice. Share this page: radar.healthai.com/precedent/device/K203517