search

OHDSI / CommonEvidenceModel

Common Evidence Model (CEM) is a structure for standardizing evidence about drug-outcome relationship across disparate evidence sources.
Apache License 2.0
14 stars 9 forks source link

readme

CommonEvidenceModel

For more information about the CommonEvidenceModel, post processing steps like generating negative controls, or learning out to get involved, please visit our wiki.

https://github.com/OHDSI/CommonEvidenceModel/wiki

Run Order:

  1. Information Prep (to load native data needed for CEM)
  2. MeSHTags (preps a MeSH Vocabulary Patch) --Before processing update the config_patient_data.csv --After running, place updated MeSHTags.csv in "evidenceProcessingClean\inst\csv"
  3. Run evidenceProcessingClean (this is a long running process, it may take up to a week)
  4. Run evidenceProcessingTranslated (currently not running Medline_Cooccurrence for extra long processing time)
  5. Run postProcessingEvidenceUnify
  6. Run postProcessingNegativeControlsPrep --This process generates a CSV file on an FTP that needs to be loaded by a separate process
  7. Run postProcessing