CTSC:ARRA.030210

Back to CTSC:ARRA supplement

Agenda

STAR-D technical implementation

Harvard Catalyst Medical Informatics group Meeting Minutes March 2, 2010

In attendance:

• Shawn Murphy
• Valerie Humblet
• Wendy Plesniak
• Alex Zeitsev
• Randy Gollub
• Chris Herrick
• Bill Telier
• Paul Lamonica
• Jesse Wei
• Mark Anderson
• Karl Helmer
• Dan Marcus

1. STAR-D 1st task, retrieve

• STAR-D will be a cell in the i2b2 hive. It comes with 4 requirements:
  • Communication with it needs to be enabled using web services. The services need special header describing who the user is that initiated the search.
  • Username: need to verify that the user has the right privileges.
  • Location of user needs to be defined.
  • Projects: users might have different roles and permission for various projects.
• Levels of permission: there are 3 levels: limited data, de-identified data and PHI data.
• STAR-D will retrieve images from PACS. We need a data structure (table). The different column will be MRN, accession #, UID, time of request, status of workflow, project#, users

2. STAR-D 2nd task, send

• Once we have located the images in the PACS, we don't want to hit the PACS with MOVE and go directly to XNAT. We want to do a GET, put in an intermediate place then move to XNAT in a way that is familiar to XNAT.
• Dan agreed that it is a good idea to have data in a "wait station" where one an do QA, de-identify and then send to XNAT or another dropside.
• All the details about STAR-D will have to be registered in the PACS systems.

Next meeting

• Chris will do a demo of the demo version of i2b2 set up in the amazon cloud.
• Paul or BIll will do a demo of a PACS query-retrieve at CHB.