Healthcare pack
Lab results CSV import API
Import patient-linked lab results from CSV. LOINC-validated, emits FHIR Observation. Schema-only mode.
curl -X POST https://api.adaptivmapr.com/v1/uploads \
-H "Authorization: Bearer $ADAPTIVMAPR_API_KEY" \
-F "template=lab_results_v1" \
-F "file=@your_data.csv"Canonical columns
The whole schema, printed as it ships.
Every canonical column, the type each row carries, whether it is required, the field-level validators that fire on commit, and the multilingual header hints the cascade resolves against. This is the shipped definition, not a summary of it.
lab_results_v1FHIR · Observation- fields
- 5
- required
- 4
- validated
- 1
- hints
- 13
| Canonical column | Type | Required | Validators | Header hints the cascade matches |
|---|---|---|---|---|
patient_id | string | yes | — | patientpatient_idpid |
loinc_code | string | yes | loinc_code | loinc |
value | number | yes | — | wertvaleurvalor |
unit | string | — | — | einheitunitéunidad |
taken_at | date | yes | — | entnahmeprélèvementfecha |
Read the same definition as JSON at GET /v1/templates/lab_results_v1. A hint match resolves on layer 2 — no LLM call, no token spend, just the flat per-map fee. Hover a validator id to see what it checks.
- 5 canonical fields
- 4 required
- 1 validated
- 13 header hints, 5 languages
- FHIR Observation
Why it exists
Written for the file you actually receive.
The Lab results template is the patient-linked counterpart to lab_result_catalog_v1 — each row is one observation for one patient at one point in time. patient_id links back to the demographics row, loinc_code identifies the test (and runs through the resolver fallback if free-text), value is the numeric result, unit is the human-readable unit, and taken_at is the sample collection timestamp. The template is `medium` risk because the row carries PHI (a result tied to a patient on a date is identifying), so full-data mode requires the AdaptivMapr subscription. Each row emits a FHIR Observation resource — code goes into Observation.code, value goes into Observation.valueQuantity.value, unit goes into Observation.valueQuantity.unit, and effectiveDateTime is taken_at. The cascade picks up "valeur", "valor", "wert" without escalation.
value is required and number-typed; results that come in as strings ("positive", "negative") need either a separate field or a custom enum — the canonical template is quantitative. Reference ranges live on the catalog row, not here, so downstream interpretation joins on loinc_code.
Migration scenarios & the foreign headers they ship
Migration scenarios for Lab results: backfilling a brand-new EMR with historical lab results from a reference-lab archive, building a population-health dashboard that needs years of HbA1c trajectories, feeding a clinical-trial registry with longitudinal lab data for enrolled patients, and migrating between LIS vendors at hospital-system level. Foreign headers seen routinely: "Patient ID / Patientennr / PID / LOINC / Code LOINC / Wert / Valeur / Valor / Valore / Einheit / Unité / Unidad / Unità / Entnahme / Prélèvement / Toma / Prelievo / Datum". Because lab results are PHI under HIPAA, GDPR, and nFADP, the schema-only mode is the default ingress and the cascade has been tuned so the headers alone are enough to make a confident mapping decision without seeing row content.
The cascade
Five layers, and the cheapest one wins.
Layers run in order and stop the moment a column resolves. That is the single biggest cost lever in the system: a column caught on layer 2 never reaches the metered layer 5.
- L1Statisticsno LLM
Auto-accepts a header that past confirmations already resolved the same way, at {minN:100, minRatio:0.95} or {minN:20, minRatio:1.00}.
- L2Heuristicno LLM
Normalises accents, punctuation and whitespace, then compares against the column name, the label, and every registered hint (DE / FR / IT / EN / ES).
- L3Fuzzyno LLM
Token-set ratio plus Levenshtein over the normalised strings. Auto-accepts at 0.80 — it absorbs typos and reordered words.
- L4Semanticcheap, cached
Embedding cosine between the header and the field’s label + hints. Catches the long tail of paraphrases.
- L5LLMmetered
Everything still unresolved goes up in ONE batched, collision-aware call, constrained to this template’s column set so it cannot invent a field.
Try it
One template id, two ways in.
REST for your import pipeline, MCP for your editor. Both run the same cascade and both honour the same schema-only clamp.
REST · POST /v1/uploads
Name the template; the cascade picks up the rest. The canonical definition is read-only at GET /v1/templates/lab_results_v1.
curl -X POST https://api.adaptivmapr.com/v1/uploads \
-H "Authorization: Bearer $ADAPTIVMAPR_API_KEY" \
-F "template=lab_results_v1" \
-F "file=@your_data.csv"MCP · Cursor / Claude Desktop
Drop AdaptivMapr into your editor and call the same cascade as a tool. Schema-only calls leave only column names and up to three clamped sample rows.
// In Cursor or Claude Desktop with the AdaptivMapr MCP server installed:
adaptivmapr.match_headers({
template_id: "lab_results_v1",
headers: ["patient_id", "loinc_code", "value", "unit"]
})The mappings response comes back flagged. PATCH /uploads/:id/mappings returns requires_hitl: true and hitl_status: "pending_review" so you can hold the commit in your own workflow — the flag is a signal, not a queue we run. Schema-only mode (headers plus at most three sample rows, each clamped to 80 characters) is a data-minimization mode enforced at the HTTP edge. Full-data mode routes the metered layer-5 call to phi-cloud in-region under a BAA, costs 20% more on the whole map charge, and stays locked until the workspace accepts the BAA/NDA in Settings → Security & Data.
Questions
Lab results CSV import — FAQ
Ready when you are
Map lab results in production — without shipping raw records.
Schema-only mode leaves only headers and a handful of clamped samples. Add full-data when you need row-level AI, routed in-region under a BAA.