Medical records, digitised
Every parcha.
Perfectly understood.
India's health records are handwritten, photographed and filed away unreadable. Parcha turns any of them into structured, verified data — in seconds, at a fraction of what it costs to read them by hand.
What you haveA real hospital prescription. Patient, doctor and clinic details removed before publishing.
A real prescription from an Indian hospital — and the actual, unedited record Parcha produced from it.
The hidden cost
Somebody is reading all of it by hand.
At an insurer or a TPA, every claim means a person squinting at a discharge summary, a bill and a stack of prescriptions. It costs ₹30–50 a document, it is slow, and it is where errors quietly enter the system.
Why people trust it
It tells you when it isn't sure.
In the prescription above — four antifungals in all-capital handwriting, on a form nobody designed for a computer — Parcha read every medicine correctly and matched all four to their generic composition. On a Mumbai clinic's parcha it resolved Lanol-ER to Paracetamol and Dompan to Domperidone, straight from the handwriting.
And on a genuinely illegible one, it does the opposite of pretending: it read the medicine, reported just 28% confidence, and sent every field for a human to confirm. See that one →
That is the whole design. A wrong dose is not a software bug — it is a patient-safety event. So nothing uncertain ever slips silently into your data.
Handwriting was hard to read · not found in the Indian medicine database
Indian medicine names, verified — not guessed.
Dolo becomes Paracetamol. Azithral becomes Azithromycin. Zerodol SP becomes Aceclofenac with Paracetamol. Every match is chosen from a real database of Indian brands — the system is not able to invent a medicine that doesn't exist.
What it reads
Four kinds of paper. One clean output.
Prescriptions
Handwritten or printed. Brand names resolved to their generic composition.
Lab reports
Every value with its unit and reference range, flagged high or low.
Discharge summaries
Admission and discharge dates, procedures, and the medicines advised on leaving.
Hospital bills
Line items and totals — with the arithmetic checked against itself.
However it arrives
Photograph it. Upload a thousand. Or just say it out loud.
Photograph
A doctor holds up a phone in the OPD. One tap, and the record exists.
Bulk upload
Send the archive. Hundreds of documents process in the background, with per-document status.
Dictate
Speak the prescription in Hindi or English. Parcha writes it up — Indian brand names and all.
Who it's for
Built for whoever is drowning in paper.
Insurers & TPAs
Your claims team reads discharge summaries, bills and prescriptions by hand. Get structured data your adjudication system can act on — and bills whose totals don’t match flagged before they’re paid.
Diagnostic labs
Reports leave as PDFs and lose all structure. Capture them as data, and become discoverable on India’s ABDM network without writing any code.
Hospitals
Decades of paper in the records room, and rising pressure to digitise. Bulk-process the archive; issue new prescriptions as verifiable digital parchas on your own letterhead.
Health-tech teams
One API instead of an in-house OCR project. Send a document, get FHIR back, with confidence scores you can route on.
Send us your worst document.
The messiest parcha, the longest discharge summary, the bill nobody can reconcile. We'll run it through and show you exactly what comes back — no commitment.
Prefer email? Write to [email protected]