CTSC:ARRA.030811

Back to CTSC:ARRA supplement

Harvard Catalyst Medical Informatics group Meeting Minutes March 8, 2011

In attendance:

• Bill Wang
• Darren Sack
• Randy Gollub
• Yong Gao
• Steve Pieper
• Vincent Roch
• Valerie Humblet
• Ron Kikinis
• Steve Pieper
• Mark Anderson (phone)
• Bill Tellier (phone)
• Jesse Wei (phone)
• Bill Hanlon (phone)

mi2b2 software update

• Bill and Chris found that their Why2K workaround (using image number from header file as check for full study) is working nicely. They have been incrementally increasing the pull rate from the MGH PACS for the mi2b2 test project. They are still using a single thread and have decreased the between study interval from 6 minutes to 1 minute. This week with Darren's approval will add the second thread during a slower time period.
• The are working on improving the efficiency of the decompression and zipping it up process for each data set.
• Discussed user performance expectations for mi2b2. Since today there is no solution at any site for retrieving these image files so even if it takes days, weeks even months for large data sets to be delivered, that is still well within an acceptable range for our first preliminary phase of delivery. Not clear what is the upper limit for access at each institution. If demand increases sufficiently may need to find funds for more hardware at the busy sites. This will be determined in the future.
• Bill hopes to have the VMware version for Jesse to try out at BIDMC by the end of the week.
• Dave is making progress with the UI to clean up and update the code including ability to serve multiple node PACS. Likely 2-3 weeks to next video demo.
• Chris continues to develop the IRB auditing capabilities for the mi2b2 STAR-D cell.
• Asynchronous C-moves. Initiator puts in a C-move request to a PACS with a specification for the C-store location and then starts a listener to wait until the PACS delivers the images. Question is whether it is possible to cut the connection between the initiator and the PACS? Folks were not sure, it is being tested at the various sites with disparate results. Chris is staying on this task as well.

American Medical Informatics Association abstract

• Shawn approved this as a good idea and nominated Bill Wang as the first author/presenter if funds can be secured. All in attendance (including Bill) agreed with this plan. Randy will attempt to secure funding and if successful will let Bill know to write the first draft and circulate to the group. See https://www.amia.org/amia2011 for details of the October 22-26, 2011 meeting. Abstract deadline is March 17, 2011.