Adravision Perio
K232440Adra Corporation · cleared 2023-12-05 · product code QIH · Radiology
Premarket evidence — what FDA accepted
source quote (p.4)
“Adravision Perio is a Software as a Medical Device (SaMD) intended to aid dentists in the measurement of mesial and distal bone levels associated with each tooth on bitewing and periapical dental radiographs of permanent teeth in patients 22 years of age or older.”
source quote (p.6)
“Both devices utilize machine learning technologies to measure interproximal bone levels in dental radiographs and allow users to visualize the radiographs with annotations, add their own annotations and use the information as part of their diagnostic decision.”
Validation studies (2)
Bench
n=340 images
endpoints: Precision; Recall
standards: FDA's Guidance for Industry and FDA Staff, “Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices"
Bench
n=78 images
endpoints: Sensitivity; Specificity; Mean Absolute Error
Reported performance (8 observations)
source quote (p.8)
“In the Length Measurement Assessment, Adravision Perio was evaluated with a dataset of 78 dental radiographs and evaluated the sensitivity and specificity against the consensus CAL length measurements labeled by three (3) ground truthers within those radiographs. Metric: Periapical, Sensitivity: 92.5%”
source quote (p.8)
“In the Length Measurement Assessment, Adravision Perio was evaluated with a dataset of 78 dental radiographs and evaluated the sensitivity and specificity against the consensus CAL length measurements labeled by three (3) ground truthers within those radiographs. Metric: Bitewing, Specificity: 94.3%”
source quote (p.8)
“Metric: Bitewing, Precision: 91.0%”
source quote (p.8)
“Metric: Bitewing, Recall: 94.0%”
source quote (p.8)
“Metric: Periapical, Precision: 84.8%”
source quote (p.8)
“Metric: Periapical, Recall: 89.3%”
source quote (p.8)
“Adravision Perio also obtained an overall tooth number classification accuracy of 91.8% in the OKS study”
source quote (p.8)
“and 93.2% in the Length Assessment demonstrating that it can well locate the CAL measurements on the relevant tooth number and tooth surface.”
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
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).