MIT-LCP / mimic-code

MIMIC Code Repository: Code shared by the research community for the MIMIC family of databases
https://mimic.mit.edu
MIT License
2.51k stars 1.5k forks source link

Update validation scripts for MIMIC-IV-ED v2.2 #1475

Closed ZhipengHe closed 1 year ago

ZhipengHe commented 1 year ago

Prerequisites

Description

MIMIC-IV-ED v2.2 removed a subset of subject_id which will be retained internally as a test set. Future data releases will exclude these patients. It removed 22,625 stay_id from all tables.

The current validation scripts for both mysql and postgres are crafted for MIMIC-IV-ED v2.0. mysql: https://github.com/MIT-LCP/mimic-code/blob/main/mimic-iv-ed/buildmimic/mysql/validate.sql postgres: https://github.com/MIT-LCP/mimic-code/blob/main/mimic-iv-ed/buildmimic/postgres/validate.sql

If I run the v, they do not work for MIMIC-IV-ED v2.2.

psql -U postgres -d mimiciv -f .\validate.sql
Validation results:
    tbl    | expected_count | observed_count | row_count_check
-----------+----------------+----------------+-----------------
 diagnosis |         946692 |         899050 | FAILED
 edstays   |         447712 |         425087 | FAILED
 medrecon  |        3143791 |        2987342 | FAILED
 pyxis     |        1670590 |        1586053 | FAILED
 triage    |         447712 |         425087 | FAILED
 vitalsign |        1646976 |        1564610 | FAILED
(6 rows)

Please update count numbers in validate.sql or create a new validate.sql for new version.

ZhipengHe commented 1 year ago

I have submitted a pull request #1476 for updating validation scripts for MIMIC-IV-ED.

alistairewj commented 1 year ago

Closed by #1476 - thanks!