Extract multiple FHIR resources from a document

POST/lang2fhir/document/multi

Extracts text from a document (PDF or image) and converts it into multiple FHIR resources, returned as a transaction Bundle. Combines document text extraction with multi-resource detection. Automatically detects Patient, Condition, MedicationRequest, Observation, and other resource types. Resources are linked with proper references (e.g., Conditions reference the Patient).

Patient identifier handling. US Core requires Patient.identifier (a business identifier such as an MRN). When the source text contains an identifier, it is extracted with an appropriate URI system. When the source text does not contain a detectable identifier, a synthetic one is generated with system: "urn:phenoml:lang2fhir-generated-id" and a UUID value so the bundle remains FHIR-valid and US Core conformant. Callers who need a tenant-specific namespace should rewrite the synthetic system after extraction.

Split classifications (optional). config.split_classifications is a caller-defined list, not a fixed taxonomy. Choose each classification id and write a natural-language description for the per-page classifier. For each page, the classifier assigns the best-matching classification or leaves the page ungrouped. Classifications with operation: "group" keep matching pages and label resources extracted from those pages; classifications with operation: "drop" remove matching pages before extraction. The clinical and admin ids in the example are illustrative, not a fixed set.

RequiresBearerauthentication

Body parameters

versionstringrequired

FHIR version to use

contentstringrequired

Base64 encoded file content. Supported file types: PDF (application/pdf), PNG (image/png), JPEG (image/jpeg), TIFF (image/tiff). File type is auto-detected from content magic bytes.

providerstringoptional

Optional FHIR provider name for provider-specific profiles

implementation_guidestringoptional

Custom Implementation Guide name. When specified, profiles from this IG are included alongside US Core profiles during resource detection. US Core is always the base layer; custom IG profiles are additive.

detection_effortstringoptionaldefault standard

Detection effort. 'standard' runs detection once, 'deep' runs detection multiple times for higher recall.

standarddeep

validation_methodstringoptionaldefault none

FHIR validation method to apply to the generated bundle. 'none' skips validation (default). 'check' runs the bundle through a FHIR structure validator and includes the results in the response. 'fix' runs validation and attempts to auto-correct errors using an LLM (up to 3 validation passes). The response includes results from each pass. Warning: 'fix' can significantly increase latency due to multiple LLM and validation round-trips.

nonecheckfix

configobjectoptional

Optional processing configuration shared across document endpoints.

page_filterobjectoptional

Deprecated. Use split_classifications instead.

split_classificationsobject[]optional

Optional per-page split classifications. Mutually exclusive with page_filter. This is a caller-defined list, not a fixed taxonomy: choose each classification id and write a natural-language description for the per-page classifier. For each page, the classifier assigns the best-matching classification or leaves the page ungrouped. Pages matching operation=drop are removed before extraction. Pages matching operation=group are kept, and extracted resources attributed to those pages include the classification id in response metadata and FHIR meta.tag. Example ids such as clinical and admin are illustrative, not a fixed set.

idstringrequired

Unique classification id. The reserved id "ungrouped" is not allowed.

descriptionstringrequired

Natural-language description of pages that belong to this classification.

operationstringrequired

Operation applied to pages assigned this classification.

groupdrop

resource_reviewobjectoptional

Opt-in, report-only faithfulness audit (honored by /lang2fhir/document/multi). For each selected resource type an LLM checks whether the chosen field kinds are actually supported by the full source document. Resources with an unsupported field are pulled out of the returned bundle and reported under resource_review in the response.

targetsobject[]required

The resource types to audit and which field kinds to check.

Returns

Successfully extracted FHIR resources from document

Response fields

successbooleanoptional

messagestringoptional

bundleobjectoptional

resourceTypestringoptional

typestringoptional

entryarray<object>optional

fullUrlstringoptional

resourceobjectoptional

resourceTypestringoptional

metaobjectoptional

tagarray<object>optional

systemstringoptional

codestringoptional

namearray<object>optional

givenarray<string>optional

familystringoptional

genderstringoptional

birthDatestringoptional

requestobjectoptional

methodstringoptional

urlstringoptional

resourcesarray<object>optional

tempIdstringoptional

resourceTypestringoptional

descriptionstringoptional

originalTextstringoptional

groupstringoptional

sourcePagesarray<integer>optional

page_classificationsarray<object>optional

page_numberintegeroptional

includebooleanoptional

classification_idstringoptional

reasonstringoptional

POSTRequest

# Get authentication token
PHENOML_TOKEN=$(curl -X POST "https://experiment.app.pheno.ml/v2/auth/token" \
  -u "$PHENOML_CLIENT_ID:$PHENOML_CLIENT_SECRET" \
  | jq -r '.access_token')

curl -X POST "https://experiment.app.pheno.ml/lang2fhir/document/multi" \
  -H "Authorization: Bearer $PHENOML_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
  "version": "R4",
  "content": "JVBERi0xLjQKJeLjz9MK...(base64-encoded PDF or image bytes)",
  "provider": "medplum",
  "config": {
    "split_classifications": [
      {
        "id": "clinical",
        "description": "Clinical notes, diagnoses, medications, observations, and patient demographics.",
        "operation": "group"
      },
      {
        "id": "admin",
        "description": "Administrative boilerplate, insurance forms, and cover sheets.",
        "operation": "drop"
      }
    ]
  }
}'

200 OKExample Response

{
  "success": true,
  "message": "Successfully extracted 3 resources",
  "bundle": {
    "resourceType": "Bundle",
    "type": "transaction",
    "entry": [
      {
        "fullUrl": "urn:uuid:patient-001",
        "resource": {
          "resourceType": "Patient",
          "meta": {
            "tag": [
              {
                "system": "https://phenoml.com/fhir/split-classification",
                "code": "clinical"
              }
            ]
          },
          "name": [
            {
              "given": [
                "John"
              ],
              "family": "Doe"
            }
          ],
          "gender": "male",
          "birthDate": "1979-03-15"
        },
        "request": {
          "method": "POST",
          "url": "Patient"
        }
      },
      {
        "fullUrl": "urn:uuid:condition-001",
        "resource": {
          "resourceType": "Condition",
          "meta": {
            "tag": [
              {
                "system": "https://phenoml.com/fhir/split-classification",
                "code": "clinical"
              }
            ]
          },
          "code": {
            "text": "Type 2 Diabetes Mellitus"
          },
          "subject": {
            "reference": "urn:uuid:patient-001"
          }
        },
        "request": {
          "method": "POST",
          "url": "Condition"
        }
      },
      {
        "fullUrl": "urn:uuid:medication-001",
        "resource": {
          "resourceType": "MedicationRequest",
          "meta": {
            "tag": [
              {
                "system": "https://phenoml.com/fhir/split-classification",
                "code": "clinical"
              }
            ]
          },
          "medicationCodeableConcept": {
            "text": "Metformin 500mg"
          },
          "subject": {
            "reference": "urn:uuid:patient-001"
          }
        },
        "request": {
          "method": "POST",
          "url": "MedicationRequest"
        }
      }
    ]
  },
  "resources": [
    {
      "tempId": "urn:uuid:patient-001",
      "resourceType": "Patient",
      "description": "John Doe, born 1979-03-15",
      "originalText": "John Doe, DOB 1979-03-15",
      "group": "clinical",
      "sourcePages": [
        1
      ]
    },
    {
      "tempId": "urn:uuid:condition-001",
      "resourceType": "Condition",
      "description": "Type 2 Diabetes Mellitus diagnosis",
      "originalText": "diagnosed with Type 2 Diabetes",
      "group": "clinical",
      "sourcePages": [
        1
      ]
    },
    {
      "tempId": "urn:uuid:medication-001",
      "resourceType": "MedicationRequest",
      "description": "Metformin 500mg prescription",
      "originalText": "Prescribed Metformin 500mg",
      "group": "clinical",
      "sourcePages": [
        2
      ]
    }
  ],
  "page_classifications": [
    {
      "page_number": 1,
      "include": true,
      "classification_id": "clinical",
      "reason": "clinical demographics and diagnosis"
    },
    {
      "page_number": 2,
      "include": true,
      "classification_id": "clinical",
      "reason": "clinical medication details"
    },
    {
      "page_number": 3,
      "include": false,
      "classification_id": "admin",
      "reason": "administrative cover sheet"
    }
  ]
}