 |
 |
Dear Russ, >A recent paper by Kastrup et al. (abstract attached below) found that >the absolute size of the fMRI response to visual stimulation did not >vary with global blood flow . This seems to suggest that one should use >additive scaling rather than proportional scaling or ANCOVA to get rid >of global effects in fMRI. Any impressions about this? > Thanks a lot for pointing out the attached abstract. With regard to your question I must admit that I am slightly confused. The ANCOVA model does assume independence between magnitude of gobal CBF and magnitude of rCBF change (and consequently a negative correlation between percentage rCBF change and global CBF), as opposed to proportional scaling which assume independence between percentage rCBF change and global CBF. The attached abstract would therefore to the best of my understanding be consistent with the ANCOVA model. Also I am not really sure what you mean by additive scaling. best regards Jesper > >Relationship between cerebral blood flow changes during visual >stimulation and baseline flow levels investigated with functional MRI >Andreas Kastrup*, Tie-Qiang Li?, Gunnar Krüger?, Gary H. Glover§, >Michael E. Moseley¶ >NeuroReport 10:1751-1756 > >Using fMRI, the relationship between regional cerebral blood flow (rCBF) >changes during visual stimulation and the prevailing baseline global and >regional flow levels was evaluated in 22 volunteers. The absolute >increase in rCBF was not correlated with baseline rCBF values (r = 0.01, >p = 0.8); however, the percentage change in rCBF showed a negative >correlation (r = ?0.78, p < 0.001). Both absolute and relative changes >in rCBF were independent of baseline global CBF >values. These results indicate that caution should be exercised when >comparing relative flow changes during focal brain activation, >especially in functionalneuroimaging studies dealing with altered >baseline flow values. © 1999 Lipponcott Williams & Wilkins > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%