All work

Blockchain tracking for lab samples

A design that traces every lab sample from patient QR code to on-chain report, against pre-analytical errors.

Stack:
  • Private DLT
  • ERC-6551
  • ERC-7401
  • ERC-721
  • Internet Computer
  • IoT transponders
  • QR codes
  • 46–68%of lab errors happen before analysis, per the health-authority report the design starts from
  • 8steps from the patient's QR code to the on-chain report

A lab sample's trip, written to a ledger

A design, not a prototype: every step of a sample, from the patient's QR code to the report, is signed and written to a permissioned ledger.

Tap a block to see what it does.

  • People
  • Logic
  • Data
  • Output
Patient
Sample journey
Shared ledger
Doctor

People

Patient app

Mobile DApp with two-factor login. Creates the ERC-6551 health passport, which also works as a wallet; the patient grants or refuses access to each record.

  • DApp
  • ERC-6551
  • 2FA

Hands to Sample collection

The problem

In a clinical lab, 46–68% of errors happen before the analysis starts: patient identification, collection, labelling, transport and storage, according to a regional health-authority report. Systems that do not talk to each other and poor sample tracking make those errors hard to catch.

What I built

  1. Proposed a private, permissioned ledger for health operators instead of a public chain, for privacy, performance and a custom consensus.
  2. Designed a patient app with two-factor login that creates a digital health passport; the patient decides who can see each record.
  3. Modelled the record as a token hierarchy: an ERC-6551 token-bound account as the passport, holding ERC-7401 and ERC-721 children for tests, prescriptions and procedures, signed by operators.
  4. Mapped the physical path: the test hash printed as a QR code on the tube, and an IoT transponder container that writes transport and storage conditions to the ledger.

The hard part

Health data on a ledger has to survive a GDPR review. The design answers with a private ledger and patient consent, but it leaves gaps I would close today: hashes on-chain and data off-chain so records can be erased, a clear link between the private ledger and the ICP chain the deck also names, and a way to trust IoT readings before they are written (the oracle problem).

Result

Submitted as my individual final paper, a 16-slide deck, for a blockchain management course. The course's overall grade was B+, not broken down by paper. Concept only: no prototype was built.

My role

Individual work: research, concept, architecture and slides. No AI tools are documented for it.