Published October 24, 2015
| Version 0.9c
Software
Open
cognition: DB binary-to-text converter and pseudonymiser for clinical research
Authors/Creators
Description
- Cognition converts binary documents in your relational database (doc, pdf, scanned images, scanned pdfs etc.) into text, and it optionally de-identifies person identifiers in the text by using approximate string matching algorithms (specifically by using the Levenshtein distance between words). It works on any relational database. It was created to process Electronic Health Records (EHR), however you can use it for any kind of data on any relational database. It will use all the available CPU-cores, so if you have a 32-core computer for example, it should quickly process your entire database. Contact me at ismail.e.kartoglu@kcl.ac.uk if you're interested.
Files
cognition-0.9c.zip
Additional details
Related works
- Is supplement to
- https://github.com/iemre/cognition/tree/0.9c (URL)