HAPI FHIR Deployment with Synthetic Data

Table of Contents

• Overview
• Features
• Prerequisites
• Installation
• Usage
  • Starting the Server
  • Importing Synthetic Data
• Configuration

Overview

This repository provides a setup for deploying a HAPI FHIR server along with a database and a file server. The file server can be used to import synthetic FHIR data into the HAPI FHIR server.

flowchart LR
    fs(File Server) <-->|Provides synthetic FHIR data files upon request| hf(HAPI FHIR Server)
    hf(HAPI FHIR Server) <-->|Stores FHIR data| db[(PostgreSQL Database)]

Loading

How it works

Here is a little sequence diagram explaining how it works.

sequenceDiagram
    participant HapIngest
    participant HAPI-FHIR-Server
    participant psql-db

    Note over HapIngest: Start Process

    HapIngest->>HAPI-FHIR-Server: Polls server until it is available
    HAPI-FHIR-Server-->HapIngest: Responds when available

    Note over HapIngest: Create Manifest
    HapIngest->>HapIngest: Generates manifests based on resource dependencies

    loop For each manifest
        HapIngest->>HAPI-FHIR-Server: Posts manifest for bulk import
        HAPI-FHIR-Server-->HapIngest: Responds with a job-id for bulk import
        HAPI-FHIR-Server->>HAPI-FHIR-Server: Begins processing the manifest
        HAPI-FHIR-Server->>HapIngest: Fetches resources specified in the manifest
        HapIngest-->HAPI-FHIR-Server: Serves ndjson resources
        HAPI-FHIR-Server-->psql-db: Persists the data into the database

        loop Poll for job status
            HapIngest->>HAPI-FHIR-Server: Polls the job-id status
            HAPI-FHIR-Server-->HapIngest: Responds with job status (In progress/Completed)
        end
    end

    Note over HapIngest: Process Complete

Loading

Features

• HAPI FHIR Server: A fully functional FHIR server using HAPI FHIR jpa-starter.
• Database: Pre-configured database to store FHIR data.
• HAPIngest: A specialized file server for importing synthetic FHIR data into the HAPI FHIR server.

Prerequisites

• Docker and Docker Compose installed on your system.
• Syntetic data files or the ability to run a Java program to generate it

Installation

1. Clone the repository:

   git clone https://github.com/CM1007-FHIR-DATA/deploy.git --recurse-submodules

2. Navigate to the project directory:

   cd deploy

3. Generate synthetic data, (replace 10 with the amount of patients to generate)

   cd synthea && \
   ./generate.sh 10 && \
   cp -R output/fhir/* ../fhir-data/ && \
   cd ..

4. Start the services using Docker Compose:

   docker compose up

Usage

Starting the Server

To start the HAPI FHIR server and hapingest-fileserver, use the following command:

docker compose up

The FHIR server will be accessible at http://localhost:8080/fhir.

Importing Synthetic Data

1. Place your synthetic FHIR data files in the fhir-data directory.
2. Run docker compose up

Configuration

Configuration options for the HAPI FHIR server, db and file server are located in the docker-compose.yml file.