RandyGollub at 16:52, 4 January 2011

2011-01-04T16:52:16Z

RandyGollub: Created page with 'Back to CTSC:ARRA supplement
'''Agenda''' * Update on Phase 3 and Phase 4 mi2b2 software * Distribution of mi2b2 software
== Harvard Catalyst Medical Informatic…'

2011-01-04T15:00:51Z

Created page with 'Back to CTSC:ARRA supplement <br> '''Agenda''' * Update on Phase 3 and Phase 4 mi2b2 software * Distribution of mi2b2 software <br> == Harvard Catalyst Medical Informatic…'

New page

Back to [[CTSC:ARRA supplement]]
<br>

'''Agenda'''
* Update on Phase 3 and Phase 4 mi2b2 software
* Distribution of mi2b2 software

<br>
== Harvard Catalyst Medical Informatics group Meeting Minutes January 04, 2011 ==

In attendance:
* Bill Wang
* Chris Herrick
* Randy Gollub
* Steve Pieper
* Darren Sack
* Vincent Roch
* Darren Sack
* Jon Bickle
* Yong Gao
* Alex Zeitsev
* Charles McGow
* Wendy Plesniak (phone)

=== Update on Phase 3 and Phase 4 mi2b2 software/hardware ===

* Bill and Chris used this week to do code clean up and make a nice enough version of Phase 3 to send to Dan Marcus and his team at Wash U to use for development. It is not in its current instance hooked up to any PACS, but it could be set up with a small DMC4chee PACS containing some sample images. This VM version uses a lot of memory. Runs slowly. Version 3 has JBoss running mi2b2.

* The two new MGH computers have finally shipped. One (dev) will be set up in Simches Radiology research space and one (production) will be set up in Needham.

* Bill and Chris are working on supporting image acquisition for the mi2b2 approved IRB on baby brains (Ellen Grant and Randy Gollub, PIs). There is a data set of ~4000 image sessions. This data set will be used to test and develop the image retrieval processes. Steve suggested using a tool called RSYNC for transferring the data sets from the mi2b2 temporary cache to the users disk space.

* We started to discuss ideas for data visualization, e.g. display to show a time line per subject with the dates/ages of images for those individuals with multiple data sets.

* Integration with XNAT (e.g. download directly to XNAT) will have to wait until the Phase 4 version of mi2b2. But this and future versions will retain the capability for transferring the data to an intermediate cache with data analytic support.

=== Distribution of mi2b2 software ===

* Chris is going to try and obtain the domain name www.mi2b2.org and point it to our Amazon cloud server. Then we can use that site to disseminate our software once it is ready for public release. We will make pointers to it from our mi2b2 wiki (need to move it to our hosted site) and from the i2b2 main site.

* Noted the need for a logo. Discussed a few ideas, Wendy will help us collect our thoughts and suggest some designs.

* Noted need for a mailing list and community forum for future support.

@@ Line 4: / Line 4: @@
 '''Agenda'''
 * Update on Phase 3 and Phase 4 mi2b2 software
-* Distribution of mi2b2 software
+* Clinical Utility of mi2b2 functions
 <br>
@@ Line 12: / Line 12: @@
 * Bill Wang
 * Chris Herrick
 * Randy Gollub
 * Steve Pieper
 * Vincent Roch
 * Darren Sack
-* Jon Bickle
+* Charles McGow
 * Yong Gao
 * Alex Zeitsev
-* Charles McGow
+* Jesse Wei
 * Wendy Plesniak (phone)
-=== Update on Phase 3 and Phase 4 mi2b2 software/hardware ===
+* Dave Wang has made great progress on the UI interface including the viewing interface with functitonality for client side filtering, sorting and clean up of other tabs, making the workflow cleaner and more user friendly.  Bill has been workign with XNAT team to clean up the interface between these two software systems.  Need to improve our task management system (setting priorities, sequencing of tasks) with an eye to future expansions.  Will do demo in next meeting or so.
-* Bill and Chris used this week to do code clean up and make a nice enough version of Phase 3 to send to Dan Marcus and his team at Wash U to use for development.  It is not in its current instance hooked up to any PACS, but it could be set up with a small DMC4chee PACS containing some sample images.  This VM version uses a lot of memory.  Runs slowly.  Version 3 has JBoss running mi2b2.
+* Testing on the Mac, has had a few problems, but Dave is plugging away.  Checked in to see if there is any knowledge from Slicer team that would be relevant, but not so unfortunately.
-* The two new MGH computers have finally shipped.  One (dev) will be set up in Simches Radiology research space and one (production) will be set up in Needham.
+* MGH production machine is set up and operational in Needham.  The MGH development machine is not going to be put in Simches, instead it will also be in Needham.  Ticket is in for installation in the near future.
-* Bill and Chris are working on supporting image acquisition for the mi2b2 approved IRB on baby brains (Ellen Grant and Randy Gollub, PIs).  There is a data set of ~4000 image sessions.  This data set will be used to test and develop the image retrieval processes.  Steve suggested using a tool called RSYNC for transferring the data sets from the mi2b2 temporary cache to the users disk space.
+* Jesse is pushing on with BIDMC.  He has downloaded a VM image and is going to try to install it there.  Knows that Bill and Chris are avaialable to assist with the installation as he needs.
-* We started to discuss ideas for data visualization, e.g. display to show a time line per subject with the dates/ages of images for those individuals with multiple data sets.
+* Client side version of programming is 80% developed.  Goal is to have this in the hands of the users by this summer so any final usability issues can be resolved.  Good candidates for this testing is the baby brain project at MGH, CHB and maybe BWH at BIDMC might be more for clinical service mode; for TIMC would be for single studies in real time after acquisition.
-* Integration with XNAT (e.g. download directly to XNAT) will have to wait until the Phase 4 version of mi2b2.  But this and future versions will retain the capability for transferring the data to an intermediate cache with data analytic support.
+* Still missing are: warnings to stay within the IRB approval, building audit log for images accessed and other auditing requirements, managing load on the PACS (rate of request, time of day, date of acquired scan, etc), should controls be per user or per site.  Do we need to consider quotas for number of scans accessed?  Final plans for cache management for managing disk space.  Need proposals for each of these topics to present to the heads of the Radiology Departments in our upcoming External Advisory Board meetings.
-=== Distribution of mi2b2 software ===
+=== Clinical Utility of mi2b2 functions ===
-* Chris is going to try and obtain the domain name www.mi2b2.org and point it to our Amazon cloud server.  Then we can use that site to disseminate our software once it is ready for public release.  We will make pointers to it from our mi2b2 wiki (need to move it to our hosted site) and from the i2b2 main site.
+* Jesse brought up clinical needs (e.g. surgeons who want scans for pre-surgical planning) echoed by Bill Hanlon for what TIMC needs for real time acquisition of scans for 3D analysis.  This is a separate but critical potential value for this work.  Worth future consideration for how and when to develop/implement, where will funding come from.
-* Noted the need for a logo.  Discussed a few ideas, Wendy will help us collect our thoughts and suggest some designs.
+== Randy's To Do List ==
-* Noted need for a mailing list and community forum for future support.
+#Set up Web-Ex meeting (not during regular meeting time slot) for Kathy and ? from BWH to demo their Data Knowledge Warehouse

CTSC:ARRA.010411 - Revision history

RandyGollub at 16:52, 4 January 2011

RandyGollub: Created page with 'Back to CTSC:ARRA supplement '''Agenda''' * Update on Phase 3 and Phase 4 mi2b2 software * Distribution of mi2b2 software == Harvard Catalyst Medical Informatic…'

RandyGollub: Created page with 'Back to CTSC:ARRA supplement
'''Agenda''' * Update on Phase 3 and Phase 4 mi2b2 software * Distribution of mi2b2 software
== Harvard Catalyst Medical Informatic…'