Stream: implementers
Topic: Test FHIR Data System for Ventilator Data
Andy Pham (Sep 18 2019 at 04:47):
Hello,
I am a student trying to test the viability of an ICU data system that reads in ventilator data into a FHIR server and then building a web interface to read off the server data using the FHIR API to display the ventilator data in interactive format. I haven't really worked with FHIR before so I'm new to all this.
The framework:
Backend Database: Postgresql
Data Source: Deidentified Ventilation data separated by deidentified patient ID stored in csv files. Includes breath metadata and classifications with a temporal dimension. It’s currently a static data source, but can be generated in real time.
FHIR Server: I have an Apache Tomcat server locally hosted on my computer and one on an Amazon AWS instance both connecting to the same backend
FHIR Resource Configs: I’m using the default hapi-fhir-jpastarter starter package from Github and have modified the configuration files to connect the various endpoints
Code System: Using IEEE Std 11073-10101a-2015 codes for mapping
I’m labelling each ventilator data type as an Observation and connecting it to fake Patient resources.
I wanted to confirm if the framework I have set up is generally correct and if there are improvements I can make to the system. I want to use Django to build a web interface to request data from the FHIR server and display it. The intended end user is for health professionals in the ICU to view the data remotely (of course this is still a long ways away)
Thanks!
Last updated: Apr 12 2022 at 19:14 UTC
