Velmeni for Dentists (V4D)

K240003

Velmeni Inc. · cleared 2024-08-30 · product code MYN · Radiology

Premarket evidence — what FDA accepted

Device typesamd

source quote (p.9)

“The proposed device is a software-only device.”

AlgorithmMachine Learning (ML) Engine with neural network-based computer algorithms for radiograph type classification, condition detection, tooth numbering, and merging.

source quote (p.5)

“Machine Learning (ML) Engine delivers V4D's core ML capabilities through the radiograph type classifier, condition detection module, tooth numbering module, and merging module. Both devices use neural network-based computer algorithms.”

Adaptive (vs locked)FDA source did not state this

PCCPFDA source did not state this

Cybersecurity addressedFDA source did not state this

Validation studies (2)

Standalone

n=1,797 images

endpoints: sensitivity; specificity; dice coefficient

Reader study (MRMC)

n=1,797 images

endpoints: sensitivity; specificity; weighted alternative free response receiver operating characteristic (wAFROC)

Reported performance (4 observations)

sensitivity0.803CI 95% CI

source quote (p.12)

“Caries Lesion-Level Sensitivity 80.3% (95% CI)”

specificity0.857CI 95% CI

source quote (p.12)

“Caries Case-level Specificity 85.7% (95% CI)”

aurocas written: “auc”0.848

source quote (p.12)

“WAFROC AUC 0.848”

diceas written: “DICE Score Mean”0.8196CI (80.81%, 83.10%)

source quote (p.11)

“Caries 81.96% (80.81%, 83.10%)”

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

recalls in product code, 24mo

MAUDE reports in code, 12mo

-100%

vs code's own 3-yr baseline

drift signals on this device

re_clearance
The FDA AI/ML device list shows a newer 510(k) K250753 (decision 2025-09-02) from Velmeni Inc. for a matching device line ("VELMENI for DENTISTS (V4D)") — a new clearance for the same line is a change event.
first seen 2026-07-08 · k_number:K250753

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).