A patient’s history as one scrollable line of time

Four familiar disasters

• A 2014 discharge summary sits in a register with 40,000 other files; the doctor asks for it in 30 seconds.
• Repeat tests because the old report could not be found are money the patient should not pay.
• Physical files leak: one unattended rack exposes names, diagnoses and HIV or psychiatric status.
• Medico-legal cases need a documented custody trail, not a statement that the file “was in the room”.

What Chronolipi does instead

1. Scan at the desk. Reception photographs the incoming paper record; the patient ID and date are read from the sheet.

2. Timeline per patient. Every prescription, scan and report for a person, oldest first, with missing visits visible.

3. Ask clinically. “All HbA1c results before the 2022 transplant” retrieves values with the source page.

4. Consent & erasure. Purpose, consent record and retention clock sit next to the document, so deletion is provable.

Extracted as typed fields

• Patient ID
• Encounter date
• Document class
• Reporting doctor / lab
• Values & units
• Consent record
• Retention clock

Anything the model is unsure about lands in a review queue instead of becoming your record.

What this does for your obligations

Clinical Establishments and NABH expectations for record retention, plus the DPDP Act’s explicit-consent-and-purge rules for health data, all want the same three things: encryption, access logging, and a defensible deletion record. This is the sector where “we promised we would not look” is least acceptable to a regulator.