Dear Kalina
Bob's advice is the best you can get.
If you still interest in differentiating BOLD vs slice entry effect
(longitudinal sinus), the best may be to use descriptive statistics as the
one based on component analysis (PCA as in Benali (HBM99) or ICA).
If truly problematic they can be avoid using longer TR (and not using
interlace mode for sure), but you have probably more important constraint
on it.
As we are talking about basic acquisition purpose, may I asked who is using
fat saturation with single shot GE EPI ? Some as we do are using it to
avoid Nyquist (or N/2) ghost, but it is never (or rarely) report in the
paper.
We discarded it for time gain. Activation remains, but it is hard to assess
the difference with our previous results. Does any one know about a paper
that assessed fat sat effect on the quality of the results.
Jack
________________________________________________________________
| Jack Foucher Universite Louis Pasteur |
| Institut de Physique Biologique UPRES-A 7004 du CNRS |
| 4 rue Kirschleger Tel: 33 (0)3 88 77 89 90 |
| 67085 STRASBOURG Fax: 33 (0)3 88 37 14 97 |
| France |
| Faster E-mail: [log in to unmask] |
| Other [log in to unmask] |
|_______________________________________________________________ |
-----Message d'origine-----
De: Robert Turner [SMTP:[log in to unmask]]
Date: lundi 29 novembre 1999 10:28
A: Kalina Christoff; [log in to unmask]
Objet: Re: modelling signal from large blood vessels
Dear Kalina,
Further to this issue, it is worth pointing out that simple geometrical
arguments demonstrate that the spatial extent of activation 'leakage' down
larger veins will mostly depend on the spatial extent of the differential
neuronal activation between experimental and control conditions. For
well-designed brain research studies such areas are usually quite small,
5-15 mm across. Thus activation in draining veins should not usually be a
problem in imaging neuroscience.
Without a priori reasons for believing that all long-latency BOLD signals
come from draining veins, one risks removing interesting valid activations
in trying to exclude veins purely by modelling this characteristic.
Hope this helps,
Bob Turner
>Dear SPM list,
>
>Has anybody had experience trying to model explicitly (as a confounding
>variable) the part of the signal in fMRI that may be coming from large
>blood vessels? Since, to my understanding, signal from large blood vessels
>does not reflect local neural activity (and is instead influenced by
>the pooled neural activity over rather large and sometimes distant brain
>regions), it seems that it would be reasonable to try to exclude this
>variance from the activation maps.
>
>Presumably, the signal coming from such large vessels (or at least
>large veins) has a later onset and peak, and it should therefore be
>possible, at least in principle, to model it explicitly and exclude the
>variance due it. Has anybody tried that before and what would an
>appropriate (late) onset be? Anyone's thoughts or experience would be
>greatly appreciated.
>
>Best regards,
>-k.
>
>_______________________________________________________________________
______
>Kalina Christoff Email: [log in to unmask]
> Office: Rm.478; (650) 725-0797
>Department of Psychology Home: (650) 497-7170
>Jordan Hall, Main Quad Fax: (650) 725-5699
>Stanford, CA 94305-2130
>http://www-psych.stanford.edu/~kalina/
>_______________________________________________________________________
______
Professor Robert Turner
Wellcome Principal Research Fellow
Wellcome Department of Cognitive Neurology
Institute of Neurology
12 Queen Square
London WC1N 3BG
Tel: +44 (171) 833-7452
Fax: +44 (171) 813-1445
www: http://www.fil.ion.ucl.ac.uk
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|