From NAMIC Wiki
Jump to: navigation, search
Home < CTSC:ARRA.AdvisoryBoard.021511

Back to CTSC:ARRA supplement

Harvard Catalyst Medical Informatics group, Advisory Board Meeting Minutes February 15, 2011

In attendance:

  • Shawn Murphy
  • Randy Gollub
  • Chris Herrick
  • Daniel Nigrin (phone)
  • Diane Keough
  • Valerie Humblet
  • Sabune Winkler
  • Scott Edmiston
  • Kathy Andriole (phone)

1. Aim of mi2b2 project

  • Shawn presented first a recall of the principal aim: The purpose of this Administrative Supplement project is to develop and disseminate additions to the Informatics for Integrating Biology and the Bedside (i2b2)-based projects to hospitals that will allow clinical imaging data from sophisticated medical imaging modalities such as MRI, PET, ultrasound and high-speed CT to be used for secondary research purposes.
  • He reassured the audience that mi2b2 can be implemented at any institution, whether or not there is already i2b2 at that institution.
  • The entire workflow for mi2b2 will be IRB approved prior to opening it up for use by the community. In addition, each individual project will need to be IRB approved.
  • The project, as specified from the initiation, is focused on within institution request for and provision of medical images.
  • There are no plans inter-site distribution of images. Mi2b2 does and will continue to operate within the approved guidelines of the parent RPDR and i2b2 projects.

2. Progress to date

  • mi2b2 is up and running at CHB, BWH and MGH.
  • the mi2b2 cell serves as a filter for medical image request, it knows who is allowed to make a request and on which patients. The cell then contacts the PACS. Images come out of the PACS at a pre-specified buffer rate and are sent into a HIPAA compliant, secure cache directory. The PI specifies where to move the images from this cache to his/her own secure location where he/she will be working on the data. The mi2b2 team will be responsible for verifying that there is sufficient space for the transfer and that the target disk is within the institutional firewall.
  • Shawn demonstrated via slides the different steps involved in an image query/retrieve.

3. Rules for frequency / timing of downloads for studies

  • Each site can customize the image flow parameters from the PACS for each Application Entity (IP address in PACS):
    • Between time#1 and time#2
    • For each Day of the Week
    • Limiting the number of image downloads/hour
    • And the number of threads at a time
    • And the delay in minutes between images
  • Workflow is controlled by researchers, requests are queued and prioritized
    • Achieves scalability for the enterprise
    • Achieves autonomy for the researcher

4. Next step: Audit for Human Subject Privacy Plan

  • Follow common i2b2 plan as for other clinical data types:
    • Query in aggregate
    • Submit IRB proposal for specific hypothesis & detailed data request
    • PI of approved study given identifiable data in secure transfer
    • IRB can audit all data transfers