Skip to content

Quick Start

Get up and running with Aether in just a few minutes.

Prerequisites

Before starting, ensure you have:

  • Aether installed - See Installation Guide
  • Go 1.25+ (if building from source)
  • Optional: TORCH server access or DIMP service (for advanced features)

5-Minute Setup

1. Verify Installation

bash
aether --help

You should see the Aether CLI help output.

2. Create a Configuration File

Create aether.yaml in your project directory:

For basic local testing (no external services):

yaml
# aether.yaml
pipeline:
  enabled_steps:
    - local_import

jobs_dir: "./jobs"

For local FHIR data with pseudonymization:

yaml
# aether.yaml
pipeline:
  enabled_steps:
    - local_import
    - dimp

services:
  dimp_url: "http://localhost:8083/fhir"

jobs_dir: "./jobs"

3. Run Your First Pipeline

Option A: Process local FHIR data

Create a test data directory:

bash
mkdir -p test-data

Start a pipeline:

bash
aether pipeline start ./test-data/

Option B: From HTTP URL

bash
aether pipeline start https://fhir.server.org/export/Patient.ndjson

Option C: From TORCH with CRTDL query

First, set TORCH credentials in aether.yaml:

yaml
services:
  torch:
    base_url: "http://torch.hospital.org"
    username: "your-username"
    password: "your-password"

Then run:

bash
aether pipeline start query.crtdl

4. Monitor Progress

bash
# List all jobs
aether job list

# Check specific job status
aether pipeline status <job-id>

# Continue a stopped pipeline
aether pipeline continue <job-id>

Common Use Cases

Processing Local FHIR Data

You have FHIR NDJSON files and want to pseudonymize them:

bash
# Update aether.yaml:
pipeline:
  enabled_steps:
    - local_import
    - dimp

services:
  dimp_url: "http://localhost:8083/fhir"

jobs:
  jobs_dir: "./jobs"

# Run the pipeline
aether pipeline start /path/to/fhir/files/

Extracting from TORCH

You want to extract a patient cohort from TORCH:

yaml
# aether.yaml
services:
  torch:
    base_url: "http://torch.hospital.org"
    username: "researcher"
    password: "secret"

pipeline:
  enabled_steps:
    - torch
    - dimp

Run with CRTDL file:

bash
aether pipeline start query.crtdl

Next Steps

Healthcare data integration made simple