CuraRad-ICH

K192167

CuraCloud Corp. · cleared 2020-04-13 · product code QAS · Radiology

Premarket evidence — what FDA accepted

Device typesamd
source quote (p.4)
“CuraRad-ICH is software as a medical device (SaMD) that detects intracranial hemorrhage (ICH) condition by analyzing non-contrast CT images.”
Algorithmdeep learning algorithm, end-to-end trainable 3D classification framework
source quote (p.5)
“The core technology of this software is a deep learning algorithm trained on non-contrast head CT scans with ICH ground truth provided by experienced radiologists. The algorithm utilizes an end-to-end trainable 3D classification framework for automatic ICH detection.”
Adaptive (vs locked)No
PCCPNo
Cybersecurity addressedNo

Validation studies (1)

Retrospective clinical

n=388 cases · 296 site(s)

endpoints: identifying ICH findings; system processing time

Reported performance (8 observations)

sensitivity0.906CI 95% CI: 85.9%–94.2%
source quote (p.6)
“The observed ICH detection sensitivity was 90.6% (95% CI: 85.9%–94.2%)”
specificity0.931CI 95% CI: 88.3%–96.4%
source quote (p.6)
“and specificity was 93.1% (95% CI: 88.3%–96.4%)”
ppvas written: “Positive Predictive Value (1% prevalence)”0.118CI 95% CI: 7.2%-18.8%
source quote (p.6)
“At a prevalence rate of 1%, PPV was 11.8% (95% CI: 7.2%-18.8%)”
npvas written: “Negative Predictive Value (1% prevalence)”0.999CI 95% CI: 99.8%-100%)
source quote (p.6)
“and NPV was 99.9% (95% CI: 99.8%-100%).”
ppvas written: “Positive Predictive Value (15% prevalence)”0.7CI 95% CI: 57.4%-80.1%)
source quote (p.6)
“At a prevalence rate of 15%, PPV was 70% (95% CI: 57.4%-80.1%)”
npvas written: “Negative Predictive Value (15% prevalence)”0.983CI 95% CI: 97.4%-98.8%)
source quote (p.6)
“and NPV was 98.3% (95% CI: 97.4%-98.8%).”
ppvas written: “Positive Predictive Value (54.9% prevalence)”0.891CI 95% CI: 84.3%-92.5%)
source quote (p.6)
“At a prevalence rate of 54.9%, PPV was 89.1% (95% CI: 84.3%-92.5%)”
npvas written: “Negative Predictive Value (54.9% prevalence)”0.918CI 88.6%-94.3%)
source quote (p.6)
“and NPV was 91.8% (88.6%-94.3%).”

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
0
drift signals on this device

Recall and MAUDE counts are product-code-level (reports aren't reliably attributable to one device); a recall is shown as device-attributed only when the recall record itself lists this clearance number. 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).

Applicable FDA guidance — what the submission is measured against

FDA guidance documents and guiding principles applicable to 510(k) AI/ML devices in the Radiology panel. A curated reference index, not legal or regulatory advice — each item states its own status, and a draft is never binding.

Applicability is derived from the device's FDA advisory panel and pathway — cross-cutting guidances apply to every AI/ML device; panel-specific ones are flagged. Titles, dates, and links verified against fda.gov as of July 2026.

Constat Precedent · public FDA/CMS data · descriptive decision-support, not regulatory or reimbursement advice. Share this page: constat.dev/precedent/device/K192167