Wearable API for Research Studies

Studies that use consumer wearables spend more time on data engineering than on analysis: each vendor has its own export format, its own sampling cadence and its own idea of what a "day" is. WearLink normalises the summaries and exposes the high-resolution series streams with canonical units.

Provenance is preserved on every record — which provider, which device type where exposed, when it was ingested and whether it was revised — so a methods section can state exactly what was measured.

Per-participant export produces the full normalised history for a user in one request, which is the shape most analysis pipelines want.

What usually goes wrong

How WearLink handles it

• High-resolution series streams — Heart rate, HRV, respiratory rate, temperature and more as series events alongside the daily summaries.
• Canonical units, originals preserved — SI and clinical units on the normalised object with the provider’s original values retained for audit.
• Per-participant export — One request returns the full normalised history for a user, including revisions.
• Sandbox with seed data — Build the analysis pipeline against realistic seeded data before the first participant is enrolled.