Henry Schein One has implemented an AI-powered solution using Amazon SageMaker AI to verify dental X-ray quality in real time. The system evaluates images immediately after capture, providing feedback to clinicians before a claim is submitted, thereby reducing delays and improving patient outcomes. This initiative addresses a significant issue in dental insurance claims, where up to 20 percent are initially denied due to poor image quality. The real-time feedback enables technicians to retake images while the patient is still present, avoiding the need for follow-up visits.
The AI system, called Image Verify, was developed to meet stringent requirements for latency, accuracy, scalability, cost efficiency, and global deployment. It processes over 11 million X-rays weekly, with the number growing at a rate of 1.5 million per week. The system operates with a median latency of 1.4 seconds and a P90 of 2.2 seconds, achieving a 0.01 percent error rate across millions of inferences. The solution is built on AWS, using Amazon SageMaker AI for machine learning inference at scale, and is designed to support up to 40,000 locations globally.
The project began as an idea in late 2025, and within months, it was scaled to over 10,000 active locations by late 2026. The system's architecture includes a multi-model inference pipeline hosted on SageMaker AI, which routes images to appropriate quality evaluation models based on type. This approach allows for efficient and accurate assessment of sharpness, alignment, and completeness. The collaboration with AWS enabled the team to optimize GPU instance performance and implement zero-downtime deployments for continuous improvement.
Source: awsml
