MIMIC mapping to OMOP

Data Quality Checks

Google Cloud SDK

curl https://sdk.cloud.google.com | bash
export PATH="$PATH:~/google-cloud-sdk/bin"
gcloud auth login

mimic2omop flow

cd src/MIMIC/
cd vocabulary_refresh
python3 vocabulary_refresh.py -s10
python3 vocabulary_refresh.py -s20
python3 vocabulary_refresh.py -s30
cd ../
python3 scripts/run_workflow.py -e conf/full.etlconf -c conf/workflow_ddl.conf
python3 scripts/run_workflow.py -e conf/full.etlconf -c conf/workflow_staging.conf
python3 scripts/run_workflow.py -e conf/full.etlconf -c conf/workflow_etl.conf
python3 scripts/run_workflow.py -e conf/full.etlconf -c conf/workflow_ut.conf
python3 scripts/run_workflow.py -e conf/full.etlconf -c conf/workflow_metrics.conf
python3 scripts/run_workflow.py -e conf/full.etlconf -c conf/workflow_unload.conf

load MIMIC from Nexcloud volume to Google Cloud Bucket

create script file

nano load_csv_files.sh

paste this code in the script file

#!/bin/bash

# Set your Google Cloud project and dataset
PROJECT_ID="booming-edge-403620"
DATASET="mimiciv_hosp"

# Loop through all CSV files in the current directory
for file in *.csv.gz; do
  # Extract the table name from the filename (assuming filenames are table names)
  table_name=$(basename "$file" .csv.gz)

  # Create the table in BigQuery and upload the data
  bq load --autodetect --source_format=CSV "${PROJECT_ID}:${DATASET}.${table_name}" "$file"
done

give the correct permissions to the script

chmod +x load_csv_files.sh

then run the script

tmux install

sudo apt-get update
sudo apt-get install tmux

Download Waveforms data and upload to Google Cloud Bucket

https://physionet.org/content/mimic4wdb/0.1.0/waves/#files-panel

wget -r -N -c -np https://physionet.org/files/mimic4wdb/0.1.0/ > wget.log 2>&1 &
gsutil cp physionet.org/* gs://shubov-athena/waveforms

Move Results to Google Cloud Bucket

bq ls booming-edge-403620:mimiciv_full_current_cdm

python extract-bq-to-bucket.py

gsutil -m cp -r gs://shubov_mimic-iv/* /data/mimic-iv-2.2/shubov_thesis/OMOP

shubov / mimic2omop