Service to matchmake incoming xml files with patients in the ukrdc database.
In testing: the are probably loads of bits that have been missed but the current code can load xml files into a locally generated database spun up using the ukrdc-sqla models.
- Install
pipx pipx install archmdarchmd src/ukrdc_cupid --out "ARCHITECTURE.md" --title="UKRDC-CUPID Architecture"
As of 07/10/24 a test version of cupid can be built with docker. This includes an api and a ukrdc built from the sqlalchemy models with a few modifications that cupid requires.
To fire this beast up run the following:
-
Build image:
docker compose --env-file=.env.docker build -
Start up application:
docker compose --env-file=.env.docker up -dnote that there may be a bunch of errors. -
Build/Rebuild database:
docker compose --env-file=.env.docker run app python scripts/test_deploy/build_db.py
This seperation of concerns allows you to do 2 without doing 3 and visa versa. This is kind of handy if for example you want to build a new version of cupid but want the 48 hours worth of data loading to persist from your last build.
CUPID Universal Patient Identification
CUPID - This is the name of the application, and the application does some recursion
Universal - Matches across both systems within and between hospitals and feeds
Patient Identification - Handles matching files to records, and grouping records into patients
Consolidated Universal Patient Identification
Consildated - Doesn't rely on a separate EMPI database, all matching uses live UKRDC data
Universal - Matches across both systems within and between hospitals and feeds
Patient Identification - Handles matching files to records, and grouping records into patients