CTSC:ARRA.110210

Back to CTSC:ARRA supplement

Agenda

• Update on Phase 2/3/4 mi2b2 software
• mi2b2 hardware

Harvard Catalyst Medical Informatics group Meeting Minutes November 2, 2010

In attendance:

• Bill Wang
• Chris Herrick
• Randy Gollub
• Shawn Murphy
• Steve Pieper
• Yong Gao
• Darren Sack
• Valerie Humblet
• Wendy Plesniack
• Kathy Andriole (phone)
• Bill Hanlon (phone)
• Mark Anderson (phone)
• Alex Zeitsev

Meeting Minutes

mi2b2 software update

• The parent code for i2b2 just went through an upgrade from version 1.4 to 1.5, so we upgraded our Phase 3 mi2b2 code to be in harmony with that upgrade. Also went through Phase 3 code to ensure Open Source licensing requirements.

• BWH is ready to install Phase 3 this week. CHB is also ready as soon as Phase 3 install package is ready. Bill is going to do an install of Phase 3 on the MGH laptop computer as an interim step prior to the purchase and setting up of the new workstation.

• On Thursday Darren, Vincent and Chris met to query the MGH PACS to find representative images to use for testing using the images identified through Randy's approved IRB.

mi2b2 hardware discussion

• Upgraded the quotes for 2 Dell servers to have 16GB RAM and 1TB storage for upgrading the MGH site to use for the Phase 3 installation. Choice for where to place the dev/test and production environment machines is still under discussion. Important considerations in order of importance are 1) connection bandwidth, 2) ease of support, 3) costs.

• Discussed if should use LINUX or Windows for the base operating system. Considerations for that choice are: 1) support at each site, 2) disk configuration (OS and disk configurations need to be harmonized, e.g. Windows hates LINUX SANS configurations). Both MGH and BWH both are requesting the Windows option for ease of support. Will ask CHB and BIDMC for their preferences. Another consideration is what environment is required for XNAT. Mark reports from a conversation with Tim Olsen that the XNAT development team is steadily working towards being primarily LINUX based and away from Windows.

• This raised the question of whether XNAT must be on the same server as mi2b2. From a management point of view it is likely that the XNAT server will come under the umbrella of research computing and not Radiology IT support going forward. The programming requirements for the integration of XNAT and mi2b2 may also influence what solution is necessary. Also, if placed on separate servers at each site, need to clarify if need separate ones for MGH and BWH or whether a single one would suffice. Chris, Yong, Bill and Alex all have insight and will try to help bring back information to this group for next week.