Henry Schein One deploys AI X-ray checker to 10,000 dental practices

Image Verify addresses a concrete and costly problem in dental practice: up to 20 percent of insurance claims are initially denied, with missing or low-quality images among the leading causes. Traditionally, a clinician reviews an X-ray hours or days after capture, discovering problems only when a claim is rejected or when a treatment plan stalls. By that time, the patient has left, and a return visit is required—adding cost, delay, and frustration. Image Verify closes this feedback loop by evaluating quality at the moment of capture, while the patient is still in the chair.

The technical achievement lies in meeting five simultaneous demands: latency under three seconds (clinicians won't wait longer), accuracy across multiple quality dimensions without false positives (to maintain trust), scale to tens of thousands of locations concurrently, cost efficiency at GPU inference scale, and consistent global deployment. Henry Schein One achieved this by building on Amazon SageMaker's async inference (handling variable request volumes without over-provisioning) and by benchmarking GPU instance families—ultimately migrating from ml.g4e.4xlarge to ml.g7e.4xlarge, which reduced median latency and consolidated the fleet from 15 to 10 instances, a 33 percent reduction in GPU infrastructure.

The product went from concept in fall 2025 to over 10,000 active locations within months, processing over 11 million X-rays. For dental practices, the practical impact is immediate: fewer patient callbacks, higher-quality claims submissions, and a training tool for new technicians. Henry Schein One's approach also avoided regulatory constraint by positioning Image Verify as a quality assessment tool, not a diagnostic AI, allowing rapid iteration and deployment at scale.