Four hundred daily faxes are OCR'd and classified by AI into referrals, results, records requests, and pharmacy queries, matched to the right patient, and routed to the right department queue automatically. Nobody sorts paper anymore, and urgent items surface in minutes.
Typical results
96%
Classified correctly first pass
400
Faxes sorted daily
8 min
Urgent result to right desk
Figures reflect typical results for this kind of automation.
The challenge
The fax machine never stopped: 400 pages a day of referrals, lab results, records requests, and pharmacy queries, all landing in one tray. Two staffers spent their days reading, stamping, and walking paper to departments. Urgent results waited hours behind routine mail, and misrouted referrals occasionally vanished entirely.
Before and after
Staff hours on fax sorting per day
−85%Arrival to right desk
−97%Misrouted faxes per week
−93%The automation
We digitized every inbound fax line into a single intake pipeline: each document is OCR'd, then an AI classifier reads it and labels the type — referral, result, records request, or pharmacy query — and its urgency. It matches the document to the patient by name, date of birth, and ID against the patient index, files it to the correct department queue, and pings the team on urgent items. Low-confidence classifications route to a human review queue.
How it works
- 01
Fax arrives
Every page from all inbound lines lands in one digital intake — no more paper tray.
- 02
OCR reads it
Text extracts from every document, including low-quality scans and cover sheets.
- 03
AI classifies
The document is labeled referral, result, records request, or pharmacy query — plus urgency.
- 04
Patient matched
Name, date of birth, and ID match the document to the right patient record.
- 05
Routed & flagged
Files land in the right department queue; urgent items ping the team immediately.
Starts from
- 400 daily faxes
- OCR every page
Delivers
- Department queues
- Patient matching
- Urgent item alerts
The impact
Staff time on fax handling typically drops 85% — from twelve hours a day across two people to under two. Urgent results reach the right desk in eight minutes instead of hours, about 96% of documents classify correctly on the first pass, and referrals stop going missing.
Questions about this automation
How does the AI decide what each fax is and how urgent it is?
Every inbound fax line feeds one digital intake, where OCR turns each page into text, including poor scans and cover sheets. An AI classifier then labels the document as a referral, lab result, records request or pharmacy query and assigns an urgency level, so routine paperwork and time-sensitive results are separated on arrival.
How are faxes matched to the right patient?
The automation reads the patient's name, date of birth and ID from the document and checks them against the patient index. A confident match files the fax to that patient and to the correct department queue. Missing details or a low-confidence classification send the document to a human review queue instead of guessing.
What does a medical center need to get started?
The center needs its fax lines routed to a digital fax service or inbox, access to the patient index, and a list of departments with the document types each one handles. Urgent alerts can go to Teams, Slack or email. Simple automations like this are usually live within days.