Skip to content
Planned — no committed timeline.

Handwritten notes OCR

Reading text from a photo of a note is planned but does not work yet. Today you attach the photo to the appointment and type the content in by hand.

Required permissions:
Practitioner
Reviewed: 2026-08-24

This feature does not work yet. The Settings → AI Features page lists an entry called Handwritten notes OCR, but there is no screen anywhere in the app from which text recognition could be started. Not in the browser, not in the mobile app. Flipping the toggle changes nothing.

We say this plainly, because the entry in the settings list suggests the opposite.

What you can do today

A paper note or a filled-in form goes into the record as an attachment, and you type the content in by hand.

01

Photograph the sheet

With a phone, in good light, square on to the page. The photo has to be legible to a human — because a human is the one who will read it.

02

Attach it to the appointment

In the active session use Attach file. The limit is 10 attachments per session, each up to 20 MB.

03

Type in whatever needs to be searchable

An attachment is an image — it cannot be searched or quoted in a summary. Type the findings, measurements and recommendations into the clinical note. The photo stays as the source evidence.

What is planned

The intention is that a photo of the sheet goes to text recognition, and the recognised text into a draft note, in two steps: Handwritten notes OCR (reading) and Handwritten OCR structuring (splitting the recognised text into sections). Each step has its own credit cost: 2 and 3.

The note image is the only material that cannot be de-identified before it is sent — text recognition needs the original pixels. That is why launching this feature will be preceded by a separate notice in the data-processing agreement, not just by someone flipping a toggle.

Was this article helpful?

Didn't find an answer? Write to us.

Open the contact form