Convert - FHIR (R4) to HealthLake

POST /convert/v1/fhirr4tohealthlake

This operation converts a FHIR R4 bundle from one of the other Orchestrate FHIR conversions (CDA-to-FHIR, HL7-to-FHIR, or Combine Bundles) into a form compatible with the Amazon HealthLake analysis platform.

Workflow

The FHIR to HealthLake operation requires interaction with both the Orchestrate and HealthLake operations.

  1. Use Combine FHIR Bundles or any of the other Orchestrate “to FHIR” operations to obtain a FHIR bundle with Orchestrate uplifts.
  2. Use the FHIR to HealthLake operation to turn the uplifted bundle into a HealthLake-compatible collection bundle.
  3. Extract the individual batch bundles from the collection and post them, one by one, to the HealthLake API endpoint.

Inputs and Outputs

The input is a FHIR R4 bundle for a single patient from one of the other Orchestrate Convert operations.

The output is a FHIR R4 bundle of type collection, containing individual FHIR bundles compatible with the HealthLake API restrictions. For example, HealthLake expects a batch bundle with no more than 160 resources. If the original bundle is too big, Orchestrate will split it up into multiple batch bundles.

After the conversions, the batch bundles within the collection must be extracted and individually posted to the HealthLake endpoint. See workflow for details.

Error Handling

If the input data is invalid (such as improperly-formatted FHIR data), the request will return an error response (see Using the API). However, in many cases the Convert service can continue processing even when there are issues with the input data, such as missing fields or unexpected data types. In such cases, the invalid data will be omitted from the finished document.

The OperationOutcome in the output collection will also include details if there are missing elements required by the HealthLake API, or any other issues that may cause a failure when posting to HealthLake.

Example

Try It
Try out the operations, with examples, in the Developer Portal Sandbox.

FHIR Bundles are verbose, so this just shows a small sample to illustrate how the data is structured. For a more complete example, try out the API in the Developer Portal Sandbox.

For example, a FHIR response from a Convert operation:

Sample Input
{
  "resourceType": "Bundle",
  "type": "batch-response",
  "entry": [
    {
      "fullUrl": "https://api.rosetta.careevolution.com/Patient/b44082eb-376d-4e93-ac82-a42340edfc7e",
      "resource": {
        "resourceType": "Patient",

would be converted to a FHIR bundle of type collection, with individual batch bundles compatible with HealthLake:

Sample Output
{
    "resourceType": "Bundle",
    "type": "collection",
    "entry": [
        {
            "resource": {
                "resourceType": "Bundle",
                "type": "batch",
                "entry": [
                    {
                        "fullUrl": "Patient/b44082eb-376d-4e93-ac82-a42340edfc7e",
                        "resource": {
                            "resourceType": "Patient",