CLEWICU System

K233216

Clew Medical Ltd. · cleared 2024-01-13 · product code QNL · Cardiovascular

Premarket evidence — what FDA accepted

Device typesamd
source quote (p.5)
“The CLEWICU System is a stand-alone analytical software product that includes the ClewICUServer and the ClewICUnitor. ClewICUServer and ClewICUnitor are software-only devices that are run on a user-provided server or cloud-infrastructure.”
Algorithmmodels derived from machine learning to calculate the likelihood of occurrence of certain clinically significant events
source quote (p.5)
“It uses models derived from machine learning to calculate the likelihood of occurrence of certain clinically significant events for patients in hospital critical care settings including:”
Adaptive (vs locked)Yes
source quote (p.8)
“This 510(k) implements a PCCP to allow the CLEW models to be trained and validated for new input data sets following the same validation protocol and meeting the same performance criteria as were used to validate the CLEWICU models described in this 510(k).”
PCCPYes
source quote (p.1)
“FDA's substantial equivalence determination also included the review and clearance of your Predetermined Change Control Plan (PCCP).”
Cybersecurity addressedNo

Validation studies (1)

Retrospective clinical

n=11,603 patients · 2 site(s)

endpoints: sensitivity/TPR for the CLEWHI model; specificity/TNR for the CLEWLR model; positive predictive value (PPV) for the CLEWHI model; sensitivity/TPR for the CLEWLR model

standards: IEC 62304:2015-06

Reported performance (10 observations)

sensitivity0.63CI 95% CI: 59-67%
source quote (p.8)
“UMASS: Sensitivity was 63% (95% CI: 59-67%), Specificity was 93% (95% CI: 93%-94%), and PPPV was 12% (95% CI: 11-14%)”
specificity0.93CI 95% CI: 93%-94%
source quote (p.8)
“UMASS: Sensitivity was 63% (95% CI: 59-67%), Specificity was 93% (95% CI: 93%-94%), and PPPV was 12% (95% CI: 11-14%)”
ppvas written: “PPV (CLEWHI UMass)”0.12CI 95% CI: 11-14%
source quote (p.8)
“UMASS: Sensitivity was 63% (95% CI: 59-67%), Specificity was 93% (95% CI: 93%-94%), and PPPV was 12% (95% CI: 11-14%)”
sensitivityas written: “Sensitivity (CLEWHI MIMIC)”0.69CI 95% CI: 66 – 73%
source quote (p.8)
“MIMIC: Sensitivity was 69% (95% CI: 66 – 73%), Specificity was 87% (95% CI: 87%-88%), and PPV was 10% (95% CI: 9 – 11%)”
specificityas written: “Specificity (CLEWHI MIMIC)”0.87CI 95% CI: 87%-88%
source quote (p.8)
“MIMIC: Sensitivity was 69% (95% CI: 66 – 73%), Specificity was 87% (95% CI: 87%-88%), and PPV was 10% (95% CI: 9 – 11%)”
ppvas written: “PPV (CLEWHI MIMIC)”0.1CI 95% CI: 9 – 11%
source quote (p.8)
“MIMIC: Sensitivity was 69% (95% CI: 66 – 73%), Specificity was 87% (95% CI: 87%-88%), and PPV was 10% (95% CI: 9 – 11%)”
sensitivityas written: “Sensitivity (CLEWLR UMass)”0.47CI 95% CI: 46.8 – 47.2)
source quote (p.8)
“UMASS: Sensitivity was 47% (95% CI: 46.8 – 47.2) and Specificity was 90.5% (95% CI: 89.6 - 91.4)”
specificityas written: “Specificity (CLEWLR UMass)”0.905CI 95% CI: 89.6 - 91.4)
source quote (p.8)
“UMASS: Sensitivity was 47% (95% CI: 46.8 – 47.2) and Specificity was 90.5% (95% CI: 89.6 - 91.4)”
sensitivityas written: “Sensitivity (CLEWLR MIMIC)”0.355CI 95% CI 35.3 – 35.7)
source quote (p.8)
“MIMIC: Sensitivity was 35.5% (95% CI 35.3 – 35.7) and Specificity was 90% (95% CI 89.1-80.9)”
specificityas written: “Specificity (CLEWLR MIMIC)”0.9CI 95% CI 89.1-80.9)
source quote (p.8)
“MIMIC: Sensitivity was 35.5% (95% CI 35.3 – 35.7) and Specificity was 90% (95% CI 89.1-80.9)”

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 Cardiovascular 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/K233216