CRATE: Clinical Records Anonymisation and Text Extraction
Created by Rudolf Cardinal. Currently developed by Rudolf Cardinal and Martin Burchell. See contributors.
CRATE is an open-source software system that:
anonymises or de-identifies arbitrary relational databases;
maps patient identifiers to research identifiers (pseudonyms);
connects relational databases to external tools for Natural Language Processing (NLP);
provides a web front end for research and administrative functions;
supports a means by which patients may consent to be contacted about research;
provides advanced ways to link records for the same person across de-identified database.
Note
To install CRATE, see Installation
Note
If you’re a clinician or a researcher using a CRATE web site to explore a research database, see the Overview for clinicians and researchers.
Contents
- 1. Introduction
- 2. Installing CRATE
- 3. Preprocessing tools
- 4. Anonymisation
- 5. Natural language processing (NLP)
- 6. Configuring the CRATE web interface
- 7. Using the CRATE web interface
- 7.1. Overview for clinicians and researchers using the CRATE web interface
- 7.2. Database structure
- 7.3. Research queries
- 7.4. Standard site queries
- 7.5. SQL helpers
- 7.6. Patient Explorer queries
- 7.7. Privileged functions for clinicians
- 7.8. Archive view
- 7.9. Contacting patients about research studies
- 7.10. Research Database Manager functions
- 7.11. Developer functions
- 8. Ancillary tools
- 9. FAQs and troubleshooting
- 10. Linkage tools
- 11. Additional notes
- 12. Automatic documentation of source code
- 13. Index of CRATE commands
- 14. Licences
- 15. Change log/history
- 16. Abbreviations
- 17. Glossary