CTSC:ARRA.092110

Back to CTSC:ARRA supplement

Agenda

• Update on Phase 2/3 mi2b2 software
• ACRIN presentation
• XNAT -- mi2b2 integration

Harvard Catalyst Medical Informatics group Meeting Minutes September 21, 2010

In attendance:

• Bill Wang
• Wendy Plesniak
• Chris Herrick
• Shawn Murphy
• Bill Tellier
• Vincent Roch
• Randy Gollub
• Valerie Humblet
• Yong Gao
• Darren Sack
• Steve Pieper
• Bill Hanlon (phone)
• Katie Andriole (phone)
• Mark Anderson (phone)
• Paul Lamonica (phone)

Meeting Minutes

mi2b2 software update

• Problem with laptop based server at MGH is that it keeps going to sleep during use and because of the laws governing laptop encryption, someone needs to go in person to enter the password and wake it up. Discussed need to gather and harmonize requirements for the next generation of server hardware, disk storage and back up at each site. MGH will likely need to be the first site to need to order this. Kathy's site may already be at this level of hardware configuration. She will circulate to this group the specs for the BWH mi2b2 system. Kathy noted that having separate development, +/- test, and production mi2b2 environments is critical. At her site the existing system would be appropriate for a development/test system. In near future will need to invest in a production system for BWH. Storage space needs for dev/test system are minimal, but more resources will be needed for our production systems.

Vincent supplied a list of additional MRNs and associated image data sets to use for testing performance issues at BWH. It is coming along fine. Phase 3 programming continues to progress steadily. Chris expects to have Phase 3 ready for installation at all sites by the end of September. UI programming is on track and continuing to progress.

ACRIN presentation

Discussed the content and format of slides that Chris will present. Shawn noted the importance of linking our mi2b2 work to the parent i2b2 project to clearly delineate the genotype-phenotype connection and how the medical imaging data fit into the overall plan. Went through examples of vignettes that could illustrate the points (e.g. stroke imaging, brain cancer imaging). Discussion triggered great idea from Steve Pieper where MRNs that are tagged so that if they arrive at a site for a clinical imaging study, the technologist will be triggered to collect additional images of a particular variety to be used for research purposes.

XNAT -- mi2b2 integration

• Shawn spoke with Dan regarding the questions raised in last week's meeting. Dan's opinion is that zipping up images and sending them is the best way to transfer images to XNAT. Shawn also clarified the importance of a single sign on to authenticate i2b2 users in XNAT.

• Still need to figure out what XNAT instance should be used for testing purposes. Chris will follow up with XNAT team to determine.