OHDSI Home | Forums | Wiki | Github

THEMIS Certification F2F Hackathon - Doodle Pool

This is a call out to all the amazing developers we have out there in the community. The THEMIS group is looking for volunteers to sit down and do some old fashion code writing to build a certification program to validate that an OMOP conversion conform with all the THEMIS conventions. It will be held on the beautiful campus of the University of Colorado. We are planning this for sometimes in May. Below is the link to the doodle pool with some dates. If you are interested, please vote so we can pick the best dates. Doodle Pool does not let me go over 2 days so the dates you see in it is for Day 1. Voting will close by Friday, March 1st.

Feel free to ask questions here.

https://doodle.com/poll/w9deq5dh8bcga8qh?utm_campaign=poll_created&utm_medium=email&utm_source=poll_transactional&utm_content=inviteparticipants-cta#email

1 Like

Awesome, @mvanzandt I’m not sure I can attend this precious hackathon. :+1:

It would be great if we can outline which certification programs OHDSI needs.

@SCYou - Do you want to meet and talk about what you guys have on March 1st?

1 Like

@mvanzandt Cool, We’re looking forward to meeting you!

Chan already have my word, if he want.

@mvanzandt, Another suggestion is coordination or harmonization of THEMIS with other working groups, such as Patient-level prediction, Population-level estimation, Gold-standard phenotype library, and Data Quality working group.
I hope THEMIS-qualified CDM data to be fully compatible with existing OHDSI method and phenotype library.

The THEMIS F2F is tomorrow at starting at 9 AM Mountain time (11 AM EST). Below is the agenda. Here is a skype link for those who wants to join the discussion on the first half of Day 1.

Join Skype Meeting

9:00am – Introductions Around the Room/On the Phone

9:30 – 10:30 How to Certify the Ratified THEMIS Standards

10:30 – 10:45 Break

10:45 – 12:00 How to Publish the Certification

12:00 – 1:00 Lunch

1:00 -2:30 Review THEMIS Rules (What Cannot Be Tested, What Can Be Tested, What Needs to be Incorporated into Current Testing)

2:30-3:00 – Break Out Into Groups for Next Steps

3:00-3:15 Break

3:15- 4:30 Break Out Groups

4:30- 5:00 Recap and Prep for Next Day

Day 2

9:00 – 11:30 Break Out Groups

11:30 – 12:00 Recap Morning Progress

12:00- 1:00 Lunch

1:00 – 2:30 Break Out Groups

2:30 – 3:00 Recap Afternoon

3:00 – 4:30 – Break Out Groups

4:30 – 5:00 Final Recap

Can please someone who attended the Themis F2F post a paragraph (any brief summary) of what the discussion was about and possible F2F outputs?

@Vojtech_Huser, I saw @MNairn did this during this week’s community call and someone transcribed it into the Google Doc Minutes for OHDSI Community Calls (https://docs.google.com/document/d/1N7weGlU8GRWFjUAGRjUiSSCPCbcI1BlJM4_Vi0in6eg/edit). Not sure if it was recorded but that is a very good place to start.

Face to Face Themis Hack-a-thon

  1. Tried to define a convention for a column as something that is testable.
  2. Classified tests per a set of definitions proposed in a paper by Michael Khan. See https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5051581/
  3. Group went through existing Themis conventions, identified if they were testable and if so, wrote a query to test the values. See https://docs.google.com/spreadsheets/d/1AUd_yCXnJfy3SjutztDDsN39HW0lj4CbBjKZZWEAlqk/edit#gid=1090520067
t