The on-ramp for Indian healthcare.

MyParcha is built and operated by MyParcha Innovations, a sole proprietorship based in Lucknow, Uttar Pradesh. Everything below is verifiable against the registrations named in it.

Trade name

MyParcha Innovations

Constitution

Sole Proprietorship (India)

Proprietor

Aditya Kumar Singh

Principal place of business

Plot No. 419, A Block, Dayal ResidencyFaizabad Road, ChinhatLucknow, Uttar Pradesh 226028India

Email

[email protected]

Telephone

+91 63873 33961

Sector

Health technology — clinical document digitisation

GST registration under the proprietor’s PAN is in progress. This page will carry the GSTIN once it is issued; until then no registration number is claimed here, because a number that has not been granted is not evidence of anything.

An Indian medical record is usually a photograph of paper. A prescription written at speed on a pad, a lab report that left the analyser as a PDF and lost every field boundary on the way out, a discharge summary that runs to six scanned pages, a hospital bill nobody can reconcile line by line.

MyParcha takes any of those and returns structured, checked data — medicines resolved to their generic composition, lab values with units and reference ranges, dates and procedures pulled out as fields, bill arithmetic verified against itself. Every field carries a confidence score, and anything the system is not sure about is routed to a human before it is allowed to become a record.

• Prescriptions — handwritten or printed; brand names resolved against a directory of Indian medicines, so “Dolo 650” becomes Paracetamol rather than a guess.
• Lab reports — every analyte 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.

The pipeline is deliberately conservative. A wrong dose is not a software defect — it is a patient safety event — so the design goal is never “read everything”, it is “never be confidently wrong”.

• Classify. Decide what kind of document this is before trying to read it. A bill and a prescription are not parsed the same way.
• Extract. Read the fields, with a confidence score attached to each one.
• Resolve to terminology. Every medicine is matched against a verified directory of Indian brand names and compositions. The system selects from that directory — it is not able to invent a medicine that does not exist.
• Check. Internal consistency — do the bill’s line items sum to its total, is the dosage form compatible with the route.
• Route to a human. Anything below threshold is held for review by clinic staff, field by field, with the original image alongside. Nothing uncertain enters the record silently.

You can watch this happen on a document of your own at /decode, or on worked examples — including one the system deliberately refused to guess at — under /demo.

India has built an impressive digital health stack. ABHA gives a person a health identifier, the facility and professional registries give institutions and clinicians theirs, and the consent framework lets records move between them under the patient’s control.

All of it assumes the record is already structured. In most of the country it is not. The record is a biro scrawl in a patient’s bag, and it never gets onto the network at all — not because the clinic objects, but because typing it in by hand is nobody’s job.

That is the gap MyParcha is built for: the step between paper and the national network. A clinic keeps working the way it already works, and the record becomes digital, structured and portable as a side effect rather than as extra labour.

MyParcha is designed against the Ayushman Bharat Digital Mission specifications — ABHA identity, the Health Facility and Healthcare Professional registries, and the consent-based exchange framework. We are stating our position precisely, because vagueness here helps nobody.

The role we intend to take, once access is granted:

• Health Information Provider (HIP) — MyParcha issues clinical documents on behalf of the clinics that use it, so records created at the point of care become linkable to a patient’s ABHA under the consent framework.
• Health Repository Provider — most small clinics have no infrastructure to store records for the long term. MyParcha holds them, so the clinic can meet that obligation without running servers.

The building blocks we expect that role to touch are ABHA, the Health Facility Registry, the Healthcare Professional Registry, and consent-based exchange for sharing records with a patient or another provider. Our data-residency position and security controls are set out in full at /security, including the one processing step that currently leaves India and what we are doing about it.

• Clinics and hospitals — issue prescriptions as verifiable digital documents on your own letterhead; digitise the archive in the records room.
• Diagnostic labs — release reports as data rather than as flat PDFs.
• Insurers and TPAs — claims documents read by machine first, with only the uncertain minority reaching a human reviewer.
• Patients — read your own prescription in plain words, free and without an account.

MyParcha is early, and we would rather say so than imply a scale we do not have. Working today: document capture and machine reading for the four document types above, the human review queue, the Indian medicine directory, patient access to their own records by a one-time code, and verifiable digital documents issued on a clinic’s letterhead.

Not yet: ABDM network exchange, and the independent security audit required before go-live. Both are described honestly on /security.

Questions about the product, an integration, or this page — [email protected] or +91 63873 33961. Postal address and everything else is on /contact.