SOFAonFHIR Prototype Pipeline

This flask prject was build for a master thesis project to automatically calculate the SOFA score from (simulated) FHIR Resources and return it as FHIR Resources.

Installation

• It is recommended to install the fhir client directly from the git repository: pip install git+https://github.com/smart-on-fhir/client-py.git
• Install Requirements: pip install -r requirements.txt
• Adjust your config.py
• You can use an open FHIR Test Server or create your own Firely Server: http://docs.simplifier.net/firelyserver/deployment/deployment.html

Start with

python3 .\run.py

This application can be used via the provided web interface, with REST Api calls e.g. via Postman or with a curl command.

Simulation data

To generate data that can be used with this application: https://github.com/julsas/sofa-on-fhir

REST API Call

<base_url_server>/SOFA_API/?source_p_id=<source_patient_id>&target_p_id=<target_patient_id>&source=<source_server>&target=<target_server>&auth_key=<key>&org=<organisation>&new_patient=<True>

Curl command

curl -H "Content-Type: application/json" -d '{"source_p_id":"source_patient_id","target_p_id":"target_patient_id", "source":"source_server", "target":"target_server","auth_key":"key", "org":"organisation", "new_patient": "True"}' -v <base_url_server>/SOFA_API_CURL/

DOCKER with provided docker file

• Build Image: sudo docker build -t <your_username>/repo .
• Run Container: sudo docker run -p 5000:5000 <your_username>/repo
• Publish Image: sudo docker push <your_username>/repo
• Pull Image: sudo docker pull <your_username>/repo

• A public image is provided on gauxi/sofa_on_fhir_pipeline:latest

Copyright and License

• MIT

Links

• FHIR profiles from the Medical Informatics Initiative (MII)
• SMART on FHIR Python Client