Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for radiographers

PACS Image Reconciliation Agent

Make sure every performed exam has its complete images in the right place.

PACS Image Reconciliation Agent: what goes in, what the agent does and what you get

What it does

After a shift, some exams show as performed but have no images in the archive, or the images sit under the wrong patient or study. This agent compares the performed exam list from the information system with what the archive holds. It finds missing, incomplete or misfiled studies. For each, it triggers a resend from the modality and rechecks the archive after a short wait. If still missing, it flags the exam to the lead with the details. The lead approves any fix that changes patient identity or merges studies. Edge case: images exist but under a misspelled name, so the agent proposes a match for the lead to approve instead of merging.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueApprovedNo 1 STARTS WHEN Shift ends 2 USES A TOOL Pull the exams performed and archive study lists 3 DOES Match by accession number and image count 4 DOES List missing, short and misfiled studies 5 USES A TOOL Trigger a resend from the modality 6 CHECKS THE RESULT Do the archived images now match the expected count? If not: wait and resend once more, then flag the exam tothe lead. Back to step 5. 7 DOES Propose identity matches for misfiled studies 8 YOU APPROVE Lead approves fixes that change patient or studyidentity 9 USES A TOOL Apply the approved fix and recheck 10 RESULT Reconciliation report
Read the steps as a list
  1. Shift ends
  2. Pull the exams performed and archive study lists
  3. Match by accession number and image count
  4. List missing, short and misfiled studies
  5. Trigger a resend from the modality
  6. Do the archived images now match the expected count?If not: wait and resend once more, then flag the exam to the lead. Back to step 5.
  7. Propose identity matches for misfiled studies
  8. Lead approves fixes that change patient or study identityThe agent waits here for your OK.
  9. Apply the approved fix and recheck
  10. Reconciliation report

How it decides

It matches by accession number, then by patient and time. An exam is closed only when the archive shows the expected image count.

  • Match first on accession number
  • Resend up to 2 times before flagging
  • Never merge studies without the lead
  • Report exams older than 24 hours without images as urgent

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Resend limit (default 2)
  • Wait between rechecks (default 10 minutes)
  • Modalities covered
  • Report recipient

What keeps you in control

It always asks you first

  • Lead approves identity changes and merges

Hard limits

  • Never delete studies
  • Never change identity without approval

It stops when

  • Done: all exams reconcile
  • Stop: modality offline, tell the lead

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensAt 11 p.m. the list showed 62 exams and the archive held 61. One CT had 180 of 300 images. The agent triggered a resend and rechecked after 10 minutes: still 180. After a second resend it showed 300. A separate MRI sat under a misspelled name, and the agent proposed a match to the lead, who approved.

More agents for radiographers