Print

Print
See below.

On Thu, Dec 5, 2013 at 10:56 AM, Doug Merkitch <[log in to unmask]> wrote:
Hello SPMers,

We have a cohort of subjects consisting of both MPRAGE and IRFSPGR T1 structural images. We noticed that the IRFSPGR scans are visibly brighter (higher intensity) than the MPRAGE scans. We are looking at gray matter differences in VBM8 (modulated, non-linear only - i.e. "m0wrp1") between several groups, with all but 1 group having majority MPRAGE scans.

My questions are:

1. Could these intensity differences bias our analyses or does VBM8 somehow account for the intensity differences BETWEEN scan type (i.e. not intensity differences within a single scan)?

Yes. You could add a covariate to try and account for the differences between scan types. There have been numerous papers looking at the effect of scanner and sequence. Here are two:
A population-average MRI-based atlas collection of the rhesus macaque. McLaren DG, Kosmatka KJ, Oakes TR, Kroenke CD, Kohama SG, Matochik JA, Ingram DK, Johnson SC. Neuroimage. 2009 Mar 1;45(1):52-9. doi: 10.1016/j.neuroimage.2008.10.058. Epub 2008 Nov 14. [See Appendix A for intensity differences of scanners]

MRI-derived measurements of human subcortical, ventricular and intracranial brain volumes: Reliability effects of scan sessions, acquisition sequences, data analyses, scanner upgrade, scanner venders and field strengths. Neuroimage. 2009 May 15;46(1):177-92. doi: 10.1016/j.neuroimage.2009.02.010. Epub 2009 Feb 20.


2. If the answer to #1 is no, how could we go about correcting for scan type? Our current practice is to use scan type as a covariate in our SPM analyses.

This is the best way. If the groups are too unbalanced in their ratio of scans, then you might still have issues.
 

Any thoughts or recommendations would be greatly appreciated!

-Doug

Doug Merkitch
Neurological Sciences
Rush University Medical Center
Phone: (312) 563-3853
Fax: (312) 563-4660
Email: [log in to unmask]