PeekMed web
K251096Peek Health, S.A. · cleared 2025-07-14 · product code QIH · Radiology
Premarket evidence — what FDA accepted
source quote (p.6)
“Being software, it does not interact with any part of the body of the user and/or patient. The subject device includes a new ML model variant for segmentation.”
source quote (p.11)
“The subject device includes a new ML model variant for segmentation. ML models incorporated into the PeekMed web were also developed, trained, tested, and externally validated for their performance according to the internal procedures.”
source quote (p.13)
“A dedicated validation dataset containing different data from the ML development dataset was used. Specifically, the validation dataset was not a sampling of the development dataset, has never been used for the algorithm training or for tuning the algorithm, and leakage between development and validation data sets did not occur. External validation datasets were collected independently of the development data to prevent bias, ensuring the reliability of the results.”
Validation studies (2)
Retrospective clinical
n=402 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%; Precision is no less than 85%; Recall is no less than 90%; F1 score is no less than 90%; MAP is no less than 90%
Bench
sample size not stated
Reported performance (10 observations)
source quote (p.14)
“DICE is no less than 90%”
source quote (p.14)
“STD DICE is between +/- 10%”
source quote (p.14)
“Precision is more than 85%”
source quote (p.14)
“Recall is more than 90%”
source quote (p.14)
“Accuracy is no less than 90%.”
source quote (p.14)
“Precision is no less than 85%”
source quote (p.14)
“Recall is no less than 90%”
source quote (p.14)
“F1 score is no less than 90%”
source quote (p.14)
“Precision is no less than 85%”
source quote (p.14)
“Recall 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
- re_clearance
The FDA AI/ML device list shows a newer 510(k) K252856 (decision 2025-12-22) from Peek Health, S.A. for a matching device line ("PeekMed web") — a new clearance for the same line is a change event.
first seen 2026-07-08 · k_number:K252856
- re_clearance
The FDA AI/ML device list shows a newer 510(k) K252452 (decision 2025-11-12) from Peek Health, S.A. for a matching device line ("PeekMed web") — a new clearance for the same line is a change event.
first seen 2026-07-08 · k_number:K252452
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).