Overjet Periapical Radiolucency Assist

K231678

Overjet, Inc · cleared 2023-09-21 · product code MYN · Radiology

Premarket evidence — what FDA accepted

Device typesamd
source quote (p.4)
“Overjet PARL Assist is a software-only device which operates in three layers: a Network Layer, a Presentation Layer, and a Decision Layer.”
AlgorithmMachine Learning model
source quote (p.4)
“Images are pulled in from a clinic/dental office, and the Machine Learning model creates predictions in the Decision Layer and results are pushed to the dashboard, which are in the Presentation Layer.”
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)

Reader study (MRMC)

n=379 images

endpoints: AUC of the receiver operating characteristic curve (ROC); Image level sensitivity; Image level specificity; instance (polygon) level sensitivity

Standalone

n=763 images

endpoints: Image level standalone sensitivity; Image Level standalone specificity; Polygon (instance) level standalone sensitivity

Reported performance (15 observations)

sensitivity0.88CI 0.847, 0.914
source quote (p.5)
“Image level standalone sensitivity was 88%, 95% CI's (0.847, 0.914).”
specificity0.842CI 0.810, 0.847
source quote (p.5)
“Image Level standalone specificity was 84.2%, 95% CI's (0.810, 0.847).”
aurocas written: “AUC improvement (assisted vs unassisted readers)”0.048CI 0.030, 0.066
source quote (p.5)
“The AUC of the ROC at the image level averaged across all readers showed a 4.8% (95% CI's 0.030, 0.066) improvement in assisted readers compared to unassisted readers.”
sensitivityas written: “Image level sensitivity improvement (assisted vs unassisted readers)”0.136CI 0.110, 0.165
source quote (p.5)
“Average Image level sensitivity across all readers increased by 13.6% (0.110, 0.165) when compared to unassisted readers.”
specificityas written: “Image level specificity difference (assisted vs unassisted readers)”-0.071CI -0.099, -0.042
source quote (p.5)
“The average specificity at the image level decreased slightly from 83.2% to 76.1% (-0.071 difference, CI's (-0.099, -0.042)).”
sensitivityas written: “Instance (polygon) level sensitivity improvement (assisted vs unassisted readers)”0.162CI 0.125, 0.194
source quote (p.5)
“Reader improvement for assisted readers at the instance (polygon) level sensitivity averaged across all readers was 16.2% (0.125, 0.194).”
sensitivityas written: “Polygon (instance) level standalone sensitivity”0.664CI 0.615, 0.711
source quote (p.5)
“Polygon (instance) level standalone sensitivity was 66.4%, 95% CI's (0.615, 0.711).”
sensitivityas written: “Dexis - Sensitivity”0.867CI 0.800, 0.933
source quote (p.5)
“Dexis - Sensitivity: 0.867 (0.800, 0.933)”
specificityas written: “Dexis - Specificity”0.885CI 0.827, 0.942
source quote (p.5)
“Specificity: 0.885 (0.827, 0.942)”
sensitivityas written: “e2v - Sensitivity”0.861CI 0.785, 0.937
source quote (p.5)
“e2v - Sensitivity: 0.861 (0.785, 0.937)”
specificityas written: “e2v - Specificity”0.804CI 0.728, 0.875
source quote (p.5)
“Specificity: 0.804 (0.728, 0.875)”
sensitivityas written: “Gendex - Sensitivity”0.889CI 0.815, 0.951
source quote (p.5)
“Gendex - Sensitivity: 0.889 (0.815, 0.951)”
specificityas written: “Gendex - Specificity”0.793CI 0.712, 0.865
source quote (p.5)
“Specificity: 0.793 (0.712, 0.865)”
sensitivityas written: “Schick - Sensitivity”0.908CI 0.836, 0.961
source quote (p.5)
“Schick - Sensitivity: 0.908 (0.836, 0.961)”
specificityas written: “Schick - Specificity”0.891CI 0.827, 0.945
source quote (p.5)
“Specificity: 0.891 (0.827, 0.945)”

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
-100%
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/K231678