Our healthcare AI

Clinical documents, read and explained

The AI behind our own app, made available to organisations — inside your product, your portal or the communications you already send.

Talk to us How it is built

What it reads

Four kinds of document

Laboratory reports

Every value on the page, matched to a canonical measurand and its aliases — the same test is printed a dozen different ways across laboratories — and explained against the reference range your own laboratory printed beside it.

Prescriptions

What each medicine is for, what to watch for, and what the label itself says. From the drug’s own labelling, never invented.

Meals and nutrition

A photograph of a plate becomes what is on it and what it is made of.

Open questions

The things people will not ask a person, answered against their own documents rather than in general.

How it behaves

The rules it is built to

We publish no reference ranges of our own

Not a gap — a decision. The app holds no clinical thresholds; it reads the range your laboratory printed. A number without a named source does not appear.

Say anything true about the numbers, nothing about the person

It explains what a value is and what the report says. It does not tell somebody what is wrong with them.

Structured data out, not only prose

An explanation a patient can read, and values your own systems can store.

98 languages, including spoken input

Transcription is priced per minute of audio rather than per word, so a question asked in Urdu costs what the same question costs in English.

What it costs

Priced per document, not per token

You know how many reports you issue. You do not know how many tokens that is, and neither does anybody else — so we price in the unit your business already counts, with an annual commitment and a volume band.

Talk to our team