FHIR GRaphQL

Motivation

My first time seeing a FHIR data resource was very overwhelming. After working with FHIR for the last 8 months, I've grown more familiar with the standard, but it still trips me up form time to time. This project is an experiment to see if a FHIR + Graphql API will make the experience better than a traditional REST API.

Roadmap

• Query a Patient resource from mock data
• Create a connector to call a HAPI-FHIR Java server instead of mocks functions.
• Support Questionnaire
• Support Questionnaire Response
• Support all resources one day

Sample Queries:

All patients with all fields

{
  allPatients {
    resourceType
    id
    meta {
      lastUpdated
      versionId
    }
    gender
    birthDate
    name {
      use
      text
      family
      given
      text
      prefix
      suffix
      period {
        start
        end
      }
    }
    managingOrganization {
      reference
    }
    telecom {
      use
      value
      system
      rank
      period {
        start
        end
      }
    }
    address {
      use
      type
      country
      postalCode
      city
      text
    }
  }
}

Questionnaire Response:

{
  questionnaireResponses {
    resourceType
    id
    meta {
      versionId
    }
    questionnaire{
      reference
    }
    status
    subject{
      reference
    }
    authored
    author {
      reference
    }
    item {
      linkId
      definition
      text
      answer {
        valueCoding {
          display
        }
        valueBoolean
        valueDate
        valueReference {
          reference
        }
      }
    }
  }
}

Running in Docker

A Dockerfile has been made for local containerization.

To build the image:

docker build -t [IMAGE_NAME] .

After building is done, check it exists in your image list

docker image ls

To run the image locally:

docker run -p 3005:3005 -d [IMAGE_NAME]

Note: If you want to map it to a different point, simply replace the port left of the :

eg: 3001:3005

The build.sh and push.sh scripts are for deploying to GCP.