Get references to two NLP methods mentioned by Justin G.

[tschaffter]

On Friday March 13, Justin G. mentioned two NLP methods that he would like Sage to deploy first in an EC2, then in an NCATS EC2. According to Justin, both method can be used using a CLI and web interface.

• [x] Find these two methods and list their resources (repos, documentation, papers, etc.)

[tschaffter]

On March 13, Yao shared the following information about two NLP methods.

Method 1: Deidentification

Short description: Automated de-identification of protected health information from electronic health records.

Repository: https://bitbucket.org/MCW_BMI/notes-deidentification/src/master/

Method 2: UCSFPhilter

Short description: Philter is a command-line based clinical text de-identification software that removes protected health information (PHI) from any plain text file.

Repository: https://github.com/BCHSI/philter-ucsf

Resources

• https://informationcommons.ucsf.edu/sites/g/files/tkssra1026/f/wysiwyg/IC-495147816-070120-1723-16802.pdf
• https://ucsf.app.box.com/s/fh1tstj98p7z2sheego6o6as2lx9028x (I don't have access to this resource)

[tschaffter]

@yy6linda Who gave you the Box.com link (see above)?

[yy6linda]

Justin shared it with me. But the documents were removed already by people who created this link.

[jaeddy]

The MCW tool/service also appears to live in GitHub: https://github.com/jayurbain/ctsi-mcw-deid

EDIT Here's the MCW org, with what I assume are more "official" repos: https://github.com/MCW-BMI

[yy6linda]

Here is the link to the review of the two methods: https://docs.google.com/document/d/1xwebCoIt57iCYmmZFLfr96Da69-mD7Zepvsh1B0-Duc/edit

[tschaffter]

Links to resources for these two tools have been successfully compiled in this ticket.