Actual-time dental picture verification with Amazon SageMaker AI at Henry Schein One

In dentistry, picture high quality determines whether or not a declare is paid or denied. As much as 20 % insurance coverage claims are initially denied, with lacking or low-quality photographs among the many main causes. But high quality evaluation has historically been a handbook, after-the-fact course of. A clinician opinions an X-ray hours or days after seize, discovering issues solely when a declare is rejected or a therapy plan can’t proceed. If the picture is blurry, misaligned, or incomplete, the affected person should return for a retake, including price, delay, and frustration for everybody concerned. The elemental hole is timing: high quality suggestions arrives lengthy after the affected person has left and the scientific second has handed.

This put up describes how Henry Schein One closed that hole by constructing Picture Confirm, an AI-powered high quality verification system on Amazon SageMaker AI that evaluates dental X-ray high quality on the level of seize, in actual time, throughout 1000’s of places. The system went from idea to over 10,000 energetic places inside months and has already processed over 11 million X-rays and rising at 1.5 million per week. Henry Schein One is now scaling towards 40,000 places globally throughout 4 areas.

The problem: Actual-time picture high quality at scale

Henry Schein One’s earlier picture verification answer ran on a distinct cloud platform, however it couldn’t ship the latency or price effectivity required for a clean scientific workflow. Rebuilding on AWS wasn’t a migration. It required designing a system that might meet 5 simultaneous necessities. Lacking any considered one of them would make the answer unusable in a scientific setting the place seconds matter and belief is earned incrementally.

1. Latency – Clinicians gained’t wait. High quality evaluation should return in below three seconds to suit naturally into the scientific workflow.
2. Accuracy – A number of machine studying (ML) fashions should consider totally different high quality dimensions, together with sharpness, alignment, and completeness, with out false positives that erode clinician belief.
3. Scale – The system should serve tens of 1000’s of places concurrently, with day by day volumes within the a whole lot of 1000’s.
4. Value effectivity – GPU inference at this scale may be prohibitively costly if not fastidiously optimized.
5. World attain – Healthcare is native, however the platform should deploy throughout a number of areas with constant efficiency.

Henry Schein One’s Picture Confirm

Henry Schein One serves dental practices worldwide by Dentrix and Dentrix Ascend, follow administration platforms utilized by tens of 1000’s of clinicians. Their Platform Companies group recognized a particular and expensive ache level: as much as 20 % of dental insurance coverage claims are initially denied, with lacking or low-quality photographs among the many main causes.

Picture Confirm is an AI-powered high quality evaluation answer constructed natively into the follow administration workflow. When a technician captures an X-ray, Picture Confirm evaluates it instantly and returns a top quality rating on a 1-to-5 scale. If the picture scores low, the technician retakes it whereas the affected person remains to be current, eradicating the necessity for a return go to.

The product went from idea in fall 2025 to manufacturing in a matter of months. Inside weeks of launch, it was stay in 250 practices. By late April 2026, that quantity had grown to over 10,000, a 43-times improve, with greater than 9 million X-rays processed and weekly volumes averaging 1.5 million and rising.

Picture Confirm is a top quality answer, not a diagnostic one. It doesn’t establish pathology. It solutions one query: is that this picture adequate for scientific use? That distinction allowed the group to iterate with out the regulatory constraints related to scientific AI.

For practices, the affect is speedy: fewer affected person callbacks, higher-quality insurance coverage claims, improved coaching for brand new technicians, and a gamification ingredient that drives technician engagement by itself.

“Picture Confirm was solely an concept on the finish of Q3. In 6 months it was created, refined, and now deployed at scale. The design and workflow make adoption quick, intuitive, and scalable.”

— Troy Miller, VP Structure, Henry Schein One

Structure: How AWS and Henry Schein One constructed the answer

Henry Schein One constructed Picture Confirm on AWS from the beginning, utilizing Amazon SageMaker AI for machine studying inference at scale. The appliance layer runs on Amazon Elastic Kubernetes Service (Amazon EKS), which orchestrates requests from the follow administration software to the SageMaker AI inference endpoints and returns the standard rating to the clinician.

Structure diagram:

The inference pipeline

When a picture is captured at a dental follow, it flows by a multi-model machine studying inference pipeline hosted on SageMaker AI. The pipeline operates in sequential levels:

1. Classification – The primary stage identifies the picture kind, similar to bitewing, panoramic, or periapical, and routes it to the suitable high quality analysis fashions.
2. High quality analysis – Specialised fashions assess sharpness, alignment, protection, and completeness for the recognized picture kind.
3. Rating aggregation – Outcomes from the standard fashions are mixed right into a single 1-to-5 high quality rating returned to the follow administration software.

The whole spherical journey, from picture seize to high quality rating displayed on display screen, takes a median of 1.4 seconds with a P90 of two.2 seconds. The system maintains a 0.01 % error charge throughout tens of millions of inferences.

The next structure selections enabled the system to satisfy latency, price, and scale necessities:

• SageMaker AI async inference – Handles variable request volumes with out over-provisioning, with autoscaling based mostly on queue depth slightly than CPU utilization, offering a extra correct sign for GPU workloads.
• GPU occasion choice – The group benchmarked throughout GPU occasion households, in the end migrating from ml.g6e.4xlarge to ml.g7e.4xlarge situations. The newer occasion kind exceeded efficiency expectations. Median latency dropped from 1.687 to 1.432 seconds and P90 from 2.45 to 2.196 seconds. In the meantime, the fleet consolidated from 15 situations all the way down to 10, a 33 % discount in GPU infrastructure with improved response occasions.
• Zero-downtime deployments – An A/B testing framework validates every change in opposition to stay manufacturing site visitors earlier than full rollout, permitting day by day optimization iterations with out threat to the manufacturing setting.
• Multi-Area by AWS Cloud WAN – Community infrastructure gives constant world deployment throughout america, Europe, Canada, and Asia Pacific Areas.

The AWS collaboration

When the engagement started, the AWS Options Structure group reviewed Henry Schein One’s current picture verification workload, which was operating on one other cloud supplier. The group then mapped the structure to equal AWS companies that might ship the identical performance with improved latency and value effectivity at scale. The preliminary focus was on establishing a working baseline on AWS, validating that the multi-model inference pipeline carried out accurately within the new setting. Because the group gained confidence within the practical parity, the collaboration shifted towards efficiency optimization and scale. The structure progressively developed towards a SageMaker AI inference-based answer that might assist tens of 1000’s of places with sub-3-second response occasions.

All through this journey, the AWS group offered ongoing structure opinions, helped establish the best GPU occasion households for the workload profile, and analyzed utilization patterns to uncover bottlenecks. The group additionally developed scaling methods aligned with Henry Schein One’s world rollout goal of 40,000 places. The iterative, hands-on nature of the collaboration allowed each groups to maneuver rapidly, delivery optimizations weekly whereas sustaining manufacturing stability by zero-downtime deployment patterns.

Optimization: Reaching effectivity at scale

One instructive side of Picture Confirm’s journey is the optimization story: how the group recognized infrastructure inefficiencies early, responded rapidly, and arrived at a system that at the moment serves tens of 1000’s of places on a lean, extremely environment friendly GPU fleet.

Figuring out the bottleneck

As Picture Confirm scaled quickly in its first weeks of manufacturing, the group carried out infrastructure profiling to know utilization patterns. Evaluation revealed that the preprocessing pipeline, together with picture decoding, normalization, and resizing, was operating fully on CPU. GPU assets had been being underutilized whereas the general system consumed extra situations than essential to maintain the workload.

This sort of CPU-side bottleneck on GPU situations is a typical pitfall in machine studying inference at scale. The sign from CPU saturation can masks GPU headroom, main groups to provision further situations when the true alternative is pipeline optimization.

The optimization strategy

The group recognized a prioritized set of optimization alternatives throughout the pipeline and commenced delivery them by zero-downtime A/B deployments. The primary enchancment, shifting picture preprocessing from CPU to GPU, delivered speedy and vital good points in occasion effectivity, with no regression in latency or reliability.

A second optimization adopted shortly after, yielding additional enhancements in throughput per occasion. Inside days, the system was serving a quickly rising location footprint on a fraction of the earlier occasion rely.

“Our group hung out figuring out 60+ particular issues that might be optimized. We simply began working down the checklist, usually deploying a number of occasions per day. The three highlights I wish to name out: shifting extra mannequin inference to the GPU immediately (unlocking throughput we will’t get CPU-side); altering to an async inference pipeline; and an A/B testing framework, which lets us safely validate enhancements earlier than we push them to 10,000+ places.”

— Troy Miller, VP Structure, Henry Schein One

Present state

Inside two weeks into the optimization journey, Picture Confirm has processed over 20 million dental X-rays, with weekly volumes averaging 1.5 million and rising. The system serves greater than 10,000 energetic places, a 43x improve from the preliminary 250 practices at launch. It delivers a median latency of 1.4 seconds and a P90 of two.2 seconds with a 0.01 % error charge throughout tens of millions of inferences. The fleet runs at roughly 70 % GPU utilization, with 60 % of the optimization backlog already accomplished and enhancements delivery constantly.

A key affect is the place these numbers translate to affected person outcomes. As much as 20 % of dental insurance coverage claims are initially denied due to poor picture high quality. Picture Confirm catches low-quality photographs on the level of seize, earlier than the affected person leaves, serving to to scale back callbacks, produce cleaner claims, and speed up reimbursement for practices. For clinicians and workplace workers, the suggestions loop is speedy and actionable.

Equally notable is the adoption velocity. The expansion from 250 to over 10,000 places occurred in weeks, not months. The gamification ingredient, the place technicians see high quality scores for his or her captures, drives engagement with out mandates and creates a pure coaching mechanism for newer workers.

Wanting forward, the structure has been validated for a worldwide goal of 40,000 places throughout 4 areas. The present 10,000 places symbolize roughly 26 % of that capability, offering vital runway for continued development with out re-architecture. The group treats infrastructure effectivity as a product function, not a one-time mission. Transport optimizations weekly by zero-downtime A/B deployments with no buyer affect and no scheduled upkeep home windows.

Classes for machine studying inference workloads

The Picture Confirm optimization journey surfaces 4 rules relevant to machine studying inference workloads operating at scale:

1. Profile earlier than scaling – The bottleneck was CPU preprocessing, not GPU compute. Including extra situations would have been costly and ineffective. Instrument the total pipeline earlier than scaling any part.
2. Optimize the pipeline, not solely the mannequin – Inference latency usually hides in preprocessing, postprocessing, and information motion slightly than in mannequin execution. Profile end-to-end, not solely the mannequin ahead cross.
3. Construct for zero-downtime iteration – A/B testing and site visitors shifting assist fast iteration with out manufacturing threat, permitting a day by day deployment cadence at scale.
4. Use the best autoscaling sign – Queue depth and concurrent request rely are extra dependable scaling alerts than CPU utilization when GPU situations carry combined CPU and GPU workloads.

Wanting forward

Picture Confirm demonstrates a sample relevant far past dental imaging: real-time machine studying inference on the level of care, scaled to 1000’s of places, and optimized to run effectively on minimal infrastructure.

The strategy rests on 4 key substances that organizations can apply to comparable workloads:

1. Managed inference – Amazon SageMaker AI handles the operational complexity of GPU fleet administration, autoscaling, and endpoint lifecycle, releasing engineering groups to deal with mannequin and pipeline high quality.
2. Aggressive pipeline optimization – Profiling the total inference pipeline, not solely the mannequin, surfaces the true bottlenecks and delivers effectivity good points with out further infrastructure spend.
3. Zero-downtime deployment patterns – A/B testing and site visitors shifting assist fast iteration with out manufacturing threat, sustaining a day by day deployment cadence at scale.
4. Multi-region structure – AWS Cloud WAN and constant infrastructure patterns throughout 4 areas present the worldwide attain that healthcare organizations require.

Henry Schein One’s subsequent section targets 40,000 places throughout 4 areas, proving that the structure scales not solely technically, however operationally throughout a worldwide healthcare footprint. The group continues to work by its optimization backlog, with every enchancment shipped stay by zero-downtime deployments.

For organizations constructing real-time picture verification or comparable machine studying inference workloads, the sample is: begin with managed inference, instrument every thing, optimize the pipeline end-to-end, and construct deployment practices that assist day by day iteration.

To study extra about operating machine studying inference workloads on AWS, go to Amazon SageMaker AI. To study extra about Picture Confirm and Henry Schein One’s follow administration platform, go to henryscheinone.com.

In regards to the authors