PeekMed web

K252856

Peek Health, S.A. · cleared 2025-12-22 · product code QIH · Radiology

Premarket evidence — what FDA accepted

Device typesamd
source quote (p.6)
“Being software, it does not interact with any part of the body of the user and/or patient.”
AlgorithmML models for segmentation, landmarking, classification, detection, and reconstruction
source quote (p.7)
“Compared to the predicate device, the subject device includes updates to existing ML variants and the introduction of new ML variants, such as the segmentation and landmarking for the knee region. In addition, certain previously cleared variants have undergone updates to improve performance while maintaining the same intended use. These modifications do not alter the overall intended use of the device”
Adaptive (vs locked)No
source quote (p.17)
“Furthermore, these tests will be repeated and updated when appropriate to ensure that the software is always properly validated, making it possible to understand in which version the problems arise and in which they are solved. Consequently, any problem that may appear in a given PeekMed web version will be identified and can be solved in subsequent versions, as all steps are traceable.”
PCCPNo
Cybersecurity addressedFDA source did not state this

Validation studies (1)

Retrospective clinical

n=672 cases

endpoints: DICE is no less than 90%; HD-95 is no more than 8; STD DICE is between +/- 10%; Precision is more than 85%; Recall is more than 90%; MRE is no more than 7mm; STD MRE is between +/- 5mm; Accuracy is no less than 90%; F1 score is no less than 90%; MAP is no less than 90%

Reported performance (6 observations)

diceas written: “DICE”stated without value
source quote (p.16)
“DICE is no less than 90%”
diceas written: “STD DICE”stated without value
source quote (p.16)
“STD DICE is between +/- 10%”
ppvas written: “Precision”stated without value
source quote (p.16)
“Precision is more than 85%”
sensitivityas written: “Recall”stated without value
source quote (p.16)
“Recall is more than 90%”
accuracyas written: “Accuracy”stated without value
source quote (p.16)
“Accuracy is no less than 90%”
f1as written: “F1 score”stated without value
source quote (p.16)
“F1 score is no less than 90%”

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
3
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/K252856