Demonstration of HL7 Ingest, Extraction, LLMs, and HL7 Dashboarding with Solr for Healthcare Providers using Cloudera DataFlow, Solr, and Machine Learning
Begin by launching AMP in CML to deploy CML related components and spin up components 3 and 4 of Demo. Download the Nifi template from step 1 and deploy in CDF/Nifi. Setup the elements of step 2 to spec below in CDW/Data Viz.
Download/Deploy in CDF/Nifi: hl7 demo
Optional component, this is a great way to show loading medical data directly into CDW and Using Data Viz for visuals.
Note that the below 2 components require completing the Final AMP Setup (Manual) at the bottom of the Readme.
Browse to https://[SOLR SERVER CML APPLICATION]/solr to Solr homepage and view the hl7_messages
collection hl7
Accessing the SOLR collection from BANANA: Go to https://[SOLR SERVER CML APPLICATION]/solr/banana/src/index.html#/dashboard
Leverage the created dashboard json file in 4_solr if not loaded automatically: chronic-symptoms-solr-dashboard.json
Open the Physician Portal in the CML Application to access and interface with this component.
- Setup Solr app instance for unauthenticated app access (IMPORTANT! Must do before next step!)
-
Execute Solr Setup Scripts (can be done in session or by the created Job) - Create collection, generate data, add data, add stopwords, make dashboard
If in session, run script:
4_solr/execute-solr-scripts.py
If by job, run
Execute Solr Scripts
job that was created during AMP setup (preferred) -
Navigate to
https://[CML APPLICATION]/solr/banana/src/index.html#/dashboard
to view Banana UI