From NAMIC Wiki
Jump to: navigation, search
Home < CTSC:ARRA.033010

Back to CTSC:ARRA supplement


  • Implementation of retrieve portion of mi2b2

Harvard Catalyst Medical Informatics group Meeting Minutes March 30, 2010

In attendance:

  • Darren Sack
  • Shawn Murphy
  • Ron Kikinis
  • Charles McGovern
  • Wendy Plesniak
  • Alex Zeitsev
  • Randy Gollub
  • Chris Herrick
  • Bill Telier
  • Steve Pieper
  • Yong Gao
  • Bill Hanlon
  • Dan Marcus (phone)
  • Paul Lamonica (phone)
  • Jesse Wei (phone)

Chris Herrick reported on his team's progress over the past week. They have the STAR-D prototyped on a Windows server. They were able to enter an MRN into a query and retrieve a list of studies. Our next phase is to communicate with the i2b2 workbench. And continue testing to make sure that the cell is robust enough to deploy at the participating sites.

The next step is to purchase or identify existing hardware for each site for hosting a server that would run both an i2b2 cell instance and DICOM (e.g. DCM4che) libraries and other software tools as selected by this team. The software is almost entirely Java based, will be able to use REST or SOAP Web services and will need Apache and JBoss. A simple ~$1,000 laptop running Linux will likely be more than sufficient for this development phase effort. But the team will consider other options. Each site can have it's own solution. In fact, the team thinks that it is optimal at this point to compare/contrast options for next stage of development as well as to learn limits of flexibility of the i2b2 STAR-D software to support different hardware solutions. One option would be to buy space for a Virtual Machine on a shared server. Jesse is going to investigate that option which he thinks might be the most advantageous at BIDMC where a 'permanent' lease option can secure a fixed IP address and additional funds can secure sufficient disk space (we likely need between 300 GB and 1 TB). Darren will investigate options at MGH and report back next week. CHB has existing hardware set up through i2b2 that might be possible to use for this purpose. It is a Windows based server at a site that is convenient for both i2b2 interface and PACS access. Randy/Shawn to follow up with Ramin and Kathy regarding BWH options.

Randy brought up another potential use case for technical development, a multi-site calibration and image acquisition harmonization study under development by the Catalyst Imaging Consortium. They will need a mechanism for sharing and working with the image data from the different sites (CHB, MGH, BWH and BIDMC). Some interest expressed, but need more information to determine feasibility. Randy to find out and report back whether the acquired data will be collected on standard clinical imaging machines, if there will be associated MRNs, accession numbers etc for each subject.

Randy will inquire with Catalyst Central whether there is flexibility to spend Year 2 equipment budget prior to year 2 if the development cycle is moving sufficiently quickly to drive this need.