Automating CBCT delivery: from scanner to referring dentist's inbox in under 60 seconds

Most dental imaging centers still ship CBCT scans the way they did in 2009: burn a CD, label it by hand, mail or hand-deliver it, and hope the referring dentist's computer has a working optical drive. The good ones email a 700 MB DICOM ZIP and pray it doesn't bounce. The math is brutal — that workflow costs hours per day in salary, drives down referrer satisfaction, and gives you exactly zero data about whether the dentist ever opened the file.

This post is for the people who run radiology centers: operations managers, owners, lead techs. We'll walk through what a fully automated CBCT delivery pipeline looks like, the real economics behind it, and a four-phase plan to get there without breaking your daily operation.

The 60-second pipeline

Here's the target end state, described step by step:

1. The scan finishes. Your CBCT machine writes a DICOM folder to a watched directory on the local PACS server, exactly as it does today.
2. A small agent picks it up. A lightweight service running on the PACS computer detects the new study, reads patient metadata, and POSTs a new exam to the CBCTHub API.
3. Files upload directly to cloud storage. The agent receives presigned URLs and streams the DICOM straight to Cloudflare R2 — no proxy, no bottleneck. A 400 MB scan finishes in 15-40 seconds on a typical clinic connection.
4. The exam is confirmed. The agent closes the exam via API. CBCTHub generates a branded share link with your clinic's logo and colors.
5. An email goes out to the referring dentist. The same agent looks up the dentist's email from your scheduling system and triggers the share — or CBCTHub does it automatically based on the referrer field.
6. A webhook updates your PMS. The referring dentist's practice software receives an event, marks the exam as delivered in their record, and notifies the dentist on their phone.

Total elapsed time: well under a minute. Total human effort: zero.

A worked example

Picture a midsize imaging center: 200 studies per day, three techs, an admin assistant who spends most of the afternoon burning CDs and chasing dentists who can't open the files.

Before automation

• 4 hours per day burning CDs, labeling, sealing, and arranging courier or pickup.
• 1 hour per day on the phone with dentists whose viewer software won't open the disc.
• Average delivery time: 24 to 72 hours after the scan.
• No visibility into whether the scan was viewed.
• Estimated cost: 5 hours of admin time per day, plus disc media, courier, and the opportunity cost of slower referral cycles.

After automation

• 0 hours per day on CD logistics. The admin's old job becomes "handle exceptions" — typically two or three a week.
• Average delivery time: under a minute.
• Dashboard shows which dentists opened the scan, when, and how many times.
• Net savings: about 100 hours per month of admin time, plus the soft revenue of dentists referring more because they trust the turnaround.

What you actually get back

Speed is the headline, but the real wins compound over months:

• Higher referrer retention. Dentists pick the center that makes their life easiest. Same-minute delivery beats any discount you could offer.
• Better patient experience. Patients no longer carry CDs from one office to another, and don't get blamed when something doesn't open.
• Tracking data. You finally know which referrers actually look at the scans. That's gold for sales and marketing.
• Less risk. No more lost CDs in the mail, no more PHI sitting on a desk for a week.

How to roll it out in four phases

Phase 1: audit (week 1)

Map your current flow honestly. How long does it take from scan-end to dentist-has-it? How many studies bounce because the dentist can't open them? Which 20% of referrers send 80% of the work? Pick the top three and design the pilot for them.

Phase 2: pilot with one scanner brand (weeks 2-4)

Don't try to roll out to every machine at once. Pick the one CBCT brand that generates the cleanest DICOM, install the upload agent on that PACS server, and run the new pipeline for two weeks while keeping CDs available as a fallback. Measure delivery time, referrer feedback, and exception count daily.

Phase 3: scale to the rest of the floor (weeks 5-8)

Once the pilot is boring, deploy the agent to your other scanners. Train one tech as the in-house expert who fields questions. Most centers we've seen finish this phase with 95%+ of studies going through the new pipeline.

Phase 4: turn off the CDs (week 9 onwards)

Send a one-page letter to every active referrer explaining the new flow. Keep CDs as a paid extra for the small minority who insist — most will drop the request within a quarter. Reinvest the admin time saved into business development or extended hours.

Rough ROI for a 200-studies-per-day center

• Saved time: ~100 admin hours per month at local fully-loaded cost.
• Saved media and courier: $400 to $1,500 per month depending on geography.
• New referrer revenue: conservative 5% volume uplift in 6 months because the center becomes the easiest one to work with.
• API and storage cost: a fraction of the savings, regardless of plan.

Most centers break even on the integration in the first month and net positive every month after.

Wrap up

Manual CBCT delivery is one of the last fully analog workflows in dental imaging, and it's the one your referrers complain about most. Automating it isn't a moonshot project — it's a four-week change that pays for itself, makes you the easiest center to refer to, and gives you the data to defend that reputation. The technology is ready. The only question is when you want to stop burning CDs.

Talk to our team about a high-volume rollout