Health Data & Context
Research • Research Initiative
Health Data & Context is the shared research layer behind Unexpectech's health work — not a consumer app, but an investigation into what health information needs so it stays meaningful over time.
WHY IT EXISTS
Health data is heterogeneous because health is contextual. Some information is structured; some arrives as a lab result, a form, or a spoken observation; some is a summary of something else. A useful system must preserve enough structure to answer basic questions about every important record: where did it come from, when was it true, who contributed it, and what is it connected to?
WHAT IT EXPLORES
Provenance and authorship, longitudinal records that preserve change instead of overwriting it, interoperability (HL7/FHIR, C-CDA, LOINC, SNOMED CT Veterinary Extension — studied as references, not claimed as implementations), and semantic retrieval across authorized context.
THE EVIDENCE-FIRST RULE
Retrieve relevant authorized context first; generate second; preserve traceability. AI-generated language should never silently become medical fact — and "there is not enough information" is a valid answer.
CONNECTIONS
This research shapes SmartMatrix (structured, evolvable matrix entries), Animal Care (veterinary continuity), and rAIn (person-centered contextual memory with permissions and explainability).