Medical records, organised without the manual work

Takes patient documents and files them where they belong, structured

Reception stops spending the morning scanning and renaming files

In a small clinic a patient's file lives across a program, a cabinet and somebody's inbox. Nobody has the whole picture in under ten minutes.

What you get

Documents reach the right file

Results, referrals and consent forms are read, identified and attached to the right patient.

Searchable, not thumbed through

What was on paper or in a scanned PDF becomes searchable text, with date and document type.

Less time at the front desk

Scanning, renaming and filing leaves the day of someone who should be talking to patients.

You can see who saw what

Every access is logged — a baseline requirement once documents carry health data.

How it works

The flow, step by step

  1. 1. Intake

    Documents arrive by scanner, email or upload, in whatever format the lab sends them.

    • OCR

  2. 2. Identification

    Name, personal code or file number are extracted and checked against existing patients.

    • OCR
    • PostgreSQL

  3. 3. Classification

    The document type — result, referral, consent — decides where it lands and who is notified.

    • GPT-4.1

  4. 4. The human check

    Any uncertain match stops and goes to a person. A document filed to the wrong patient is worse than one not filed at all.

  5. 5. Filing

    The document enters your system, with access logged and your retention rules applied.

    • PostgreSQL

Results

A 70-employee catering group with two event venues

Reading documents and matching them automatically is the same technology already running in production.

Automation workflows
125 63
WhatsApp errors per hour
15–30 0
Security findings
584 383
Weeks to production
— 6

Works with what you already run

  • n8n
  • PostgreSQL

Related solutions

Frequently asked questions

Do you hold security certifications for health data?
We hold no ISO 27001 or SOC 2 and we claim nothing else. If your policy or a contract requires them, tell us at the outset — we will point you to a certified supplier rather than waste your time.
Does patient data leave our premises?
It need not. The system can run entirely on your infrastructure, with locally hosted models reading the documents. We tell you exactly what passes where before we start.
Who is responsible if a document reaches the wrong patient?
You are, as the controller. That is why uncertain matches are never made automatically — they stop at a person. The system is built to refuse rather than guess.
Does it work with our clinic software?
If it has an API we write into it directly. If not, we automate the interface, with more upkeep — we tell you the difference upfront.
How long does it take to set up?
Three to six weeks for one document type, longer if documents come from many labs in differing formats.
What will you not automate here?
Anything amounting to a clinical decision. We organise and move documents; interpreting them stays with the clinician, always.

How much of your front desk's day goes on paperwork?

Claim a free audit