FHIR Integration Guide

Build FHIR R4 compatible healthcare applications with HealthCloud Marketplace.

🏥 FHIR R4 Native

HealthCloud Marketplace is built on FHIR R4 (Fast Healthcare Interoperability Resources) standards, ensuring seamless interoperability with existing healthcare systems and EHRs.

Supported FHIR Resources

Patient

Patient demographics and administrative information

Observation

Measurements and clinical findings

Condition

Problems, diagnoses, and health concerns

MedicationRequest

Prescription and medication orders

DiagnosticReport

Diagnostic imaging and lab results

Procedure

Medical procedures and interventions

Encounter

Patient visits and healthcare events

AllergyIntolerance

Allergies and adverse reactions

FHIR Integration Example

Deploying a FHIR-Compatible AI Model

Specify which FHIR resources your model consumes and produces:

{
  "modelId": "radiology-classifier-v2",
  "fhirResources": {
    "inputs": ["DiagnosticReport", "ImagingStudy"],
    "outputs": ["Observation", "DiagnosticReport"]
  },
  "fhirVersion": "R4",
  "deployment": {
    "type": "real-time",
    "endpoint": "https://api.healthcloud.com/fhir/models/radiology-classifier-v2"
  }
}

Creating FHIR Observations

Example of creating a FHIR R4 Observation from wearable device data:

{
  "resourceType": "Observation",
  "id": "heart-rate-123",
  "status": "final",
  "category": [{
    "coding": [{
      "system": "http://terminology.hl7.org/CodeSystem/observation-category",
      "code": "vital-signs",
      "display": "Vital Signs"
    }]
  }],
  "code": {
    "coding": [{
      "system": "http://loinc.org",
      "code": "8867-4",
      "display": "Heart rate"
    }]
  },
  "subject": {
    "reference": "Patient/patient-123"
  },
  "effectiveDateTime": "2025-11-02T10:30:00Z",
  "valueQuantity": {
    "value": 72,
    "unit": "beats/minute",
    "system": "http://unitsofmeasure.org",
    "code": "/min"
  },
  "device": {
    "reference": "Device/apple-watch-series-9",
    "display": "Apple Watch Series 9"
  }
}

EHR System Integration

Supported EHR Systems

HealthCloud provides pre-built connectors for major EHR systems:

✓
Epic FHIR Bridge - OAuth 2.0 integration with Epic's FHIR APIs
✓
Cerner Connector - Real-time data sync with Cerner Millennium
✓
Athena Health Integration - Bi-directional data exchange

Example: Epic FHIR Integration

import { EpicFHIRClient } from '@healthcloud/connectors';

const epicClient = new EpicFHIRClient({
  baseUrl: 'https://fhir.epic.com/interconnect-fhir-oauth',
  clientId: process.env.EPIC_CLIENT_ID,
  clientSecret: process.env.EPIC_CLIENT_SECRET
});

// Fetch patient data
const patient = await epicClient.getPatient('patient-123');

// Retrieve observations
const observations = await epicClient.getObservations({
  patientId: 'patient-123',
  category: 'vital-signs',
  date: '2025-11-02'
});

// Send results back to Epic
await epicClient.createObservation({
  patientId: 'patient-123',
  code: '8867-4', // LOINC code for heart rate
  value: 72,
  unit: 'beats/minute'
});

UMLS Terminology Support

HealthCloud integrates with UMLS (Unified Medical Language System) for standardized clinical terminology:

ICD-10-CM

International classification of diseases

LOINC

Logical observation identifiers

SNOMED CT

Systematized nomenclature of medicine

RxNorm

Normalized drug terminology

Compliance & Security

🔒 HIPAA & GDPR Compliant

• End-to-end encryption for PHI (Protected Health Information)
• Audit trails for all FHIR resource access
• Role-based access control (RBAC)
• Automatic de-identification when required
• Patient consent management

Complete Integration Example

import { HealthCloudClient } from '@healthcloud/sdk';

const client = new HealthCloudClient({
  apiKey: process.env.HEALTHCLOUD_API_KEY
});

// Deploy a FHIR-compatible AI model
const deployment = await client.models.deploy({
  modelId: 'diabetes-risk-predictor',
  fhirResources: {
    inputs: ['Observation', 'Condition', 'MedicationRequest'],
    outputs: ['RiskAssessment']
  },
  compliance: {
    hipaa: true,
    gdpr: true,
    deidentify: false
  }
});

// Process patient data
const riskAssessment = await client.models.predict({
  deploymentId: deployment.id,
  input: {
    resourceType: 'Bundle',
    entry: [
      { resource: glucoseObservation },
      { resource: diabetesCondition },
      { resource: medicationHistory }
    ]
  }
});

// Result is a FHIR RiskAssessment resource
console.log(riskAssessment.prediction.probabilityDecimal); // 0.23
console.log(riskAssessment.outcome.text); // "Low risk of Type 2 Diabetes"