Difference between revisions of "CTSC:multicenter101410"

From NAMIC Wiki
Jump to: navigation, search
 
(4 intermediate revisions by the same user not shown)
Line 17: Line 17:
 
<br>
 
<br>
  
Modalities to start with: CT and MRI, protocol without contrast
+
===MRI neuroimaging virtual trial===
Bodypart: neuroimaging, there is already a common neuro protocol for pediatric MS.
+
 
Example: 3 sites, 3 MRI scanners 3T: 1 GE, 1 Philips and 1 Siemens:
+
* It will be done with 3T MRI scanners, without contrast, with a multi-element head coil. Three sequences will be run: T2 weighted, FLAIR and MP RAGE. We will implement a common protocol with human phantoms.
Implement common protocol with human phantoms
+
* The dataset will be transfer on a server (HIPPA compliant) and we will look for reproducibility for 3 outcome measures; total brain volume, CFS and white vs gray matter.  
take the dataset and look for reproducibility for 3 outcome measures; total brain volume, CFS and white/gray matter.  
+
* The goal is to understand where protocols are equivalent and different. Need to understand intra- and inter- variability. We will also test the differences obtained with different versions of same software.
understand where protocols are equivalent and different. Need to understand intra- and inter- variability
+
* We will also do a pediatric version of the trial:
Test difference obtained with different version of same software
+
** Phantoms: we need phantoms appropriate for the age of the child. It is very difficult to scan human phantoms for pediatric studies.
 +
** 2 protocols are needed, one for young children and one for teenager (for example at age 10, the FLAIR sequence needs new settings)
 +
 
 +
=== Extension to other disease and modalities===
 +
 
 +
* We can offer recommendation on the best phantoms for a specific disease. We also offer advices on calibration of theses phantoms.
 +
* Important questions: How often do we do the scanning, how many volunteer does one need to be relevant, what about gender (it is best to cover age and gender but it will depend on the disease).
 +
* Validation: there must be a consensus between segmentation, [http://crl.med.harvard.edu/research/STAPLE/index.php  Staple] is a good tool to provide for validation.
 +
 
 +
===Next steps===
 +
 
 +
* Implement a virtual PET trial
 +
* Look at ACRIN for examples of multi-center CT trials (ex: liver, [http://www.acr.org/MainMenuCategories/media_room/FeaturedCategories/Videos/CTC-Trial.aspx colonography])
 +
* Standardization of dose measurement
 +
* Talk about IRB issues
 +
* Think about funding mechanism

Latest revision as of 19:56, 16 November 2010

Home < CTSC:multicenter101410

Back to Collaboration:Harvard_CTSC

Agenda

  1. establishment of the first virtual trial

Multicenter Clinical Trail Working group Meeting Minutes October 14, 2010

In attendance:

  • Valerie Humblet
  • Bob Lenkinski
  • Clare Tempany
  • Charles Guttmann
  • Simon Warfield


MRI neuroimaging virtual trial

  • It will be done with 3T MRI scanners, without contrast, with a multi-element head coil. Three sequences will be run: T2 weighted, FLAIR and MP RAGE. We will implement a common protocol with human phantoms.
  • The dataset will be transfer on a server (HIPPA compliant) and we will look for reproducibility for 3 outcome measures; total brain volume, CFS and white vs gray matter.
  • The goal is to understand where protocols are equivalent and different. Need to understand intra- and inter- variability. We will also test the differences obtained with different versions of same software.
  • We will also do a pediatric version of the trial:
    • Phantoms: we need phantoms appropriate for the age of the child. It is very difficult to scan human phantoms for pediatric studies.
    • 2 protocols are needed, one for young children and one for teenager (for example at age 10, the FLAIR sequence needs new settings)

Extension to other disease and modalities

  • We can offer recommendation on the best phantoms for a specific disease. We also offer advices on calibration of theses phantoms.
  • Important questions: How often do we do the scanning, how many volunteer does one need to be relevant, what about gender (it is best to cover age and gender but it will depend on the disease).
  • Validation: there must be a consensus between segmentation, Staple is a good tool to provide for validation.

Next steps

  • Implement a virtual PET trial
  • Look at ACRIN for examples of multi-center CT trials (ex: liver, colonography)
  • Standardization of dose measurement
  • Talk about IRB issues
  • Think about funding mechanism