Hi Andrew,
the problem is that every patient is different. For example, case 6 in my 2006 JMRI paper (23:921-32) could not attend to visual stimuli of a speech and language paradigm because the lesion afflicted the dorsal visual stream. With auditory stimulus presentation she did fine. So you will need to decide which stimulation you use based on the patients performance.
Similarily, you need to be able to adjust the speed of your paradigm. Case 11 in the above paper, for example, required a very low task load. If you present your material in a way or speed your patient can not or (according to actual abilities) does not optimally follow, you may get no or no optimal activations.
So prior detailed neuropsychological testing is absolutely mandatory, and your subsequent FMRI should account for the results.These issues seriously limit standardization.
Also, some patients are able to perform multiple paradigms in multiple runs (if so you should record these!), others are not. For clinical decsion-making, you need to get the best out of every one you examine. This requires a strategy that suits the individual.
Motor mapping is dispensable in most of the cases. You can delineate the motor strip by anatomic criteria in the very vast majority of cases. If you have problems, ask your neuroradiologist. He should be able to do it. I hardly do any motor or sensorimotor mapping anymore, and I am not the only one: Susan Bookheimer, for example, has even more so given up on it.
Similarily, there is usually no point in visual or auditory mapping prior to resective surgery. Both are quite easily defined anatomically, and often you'd be willing to sacrifice either because of their bilateral representation: the deficits are usually not devastating. You may, however, use FMRI of the auditory system to demonstrate its principal integrity prior to cochlear implantation, i.e. when auditory activations are evoked by acoustic ("FMRI audiometry") or electrical stimulation (FMRI promontory testing; see http://www.fmrib.ox.ac.uk/fslcourse/physics+apps/bartsch.pdf).
Memory mapping prior to anterior temporal lobectomy, for example, is challenging and the predictive value for postsurgical outcome of a given patient has not been established (even though presurgical FMRI and outcome have shown to somewhat correspond across various patient samples).
Thus, what matters most is speech / language mapping. The problem here is that a single paradigm can not delineate all areas involved equally well (and that the FMRI activations do not tell you which of them may anyway be lesioned without causing deficits, of course). Starting of with a paradigm that reliably activates most areas you are interested in (e.g. "Broca's", "Wernicke's", "Exner's", "Mill's", "Geschwind's"...) usually makes sense. Note that bad aphasias have happended after injury to Exner's (originally implicated to be lesioned in pure motor agraphia) as well as Mill's basotemporal language area. However, other patients seem to cope with lesions in these regions without developing aphasia, so when the neurosurgeon is going to get close to them intra-operative stimulation mapping or pre-operative superselective WADA testing may still be a good idea. We use reading or listening to nonfinal embedded clause sentences vs. consonant strings or non-word "sentences" and alternatively synonym judgements vs. letter matching conditions as complex paradigms to activate the language network at a large scale. These are based on Fernandez et al., Neurology 2003;60:969-975 and Stowe et al., Brain Lang 2000;75:347-58. The latter has been chosen one of the best lowest denominators for the purpose in a German consensus meeting a couple of years ago and translated into a German version by Klemens Gutbrod. I have modified it a bit further but it's only in German. When tolerance and compliance are good, I usually record two runs of it. In further sessions you may want to probe more specific functions, such as description-cued covert naming or word/non-word decisions. Automatic speech (such as repeating the months of the year or phoneme sequences) and letter-cued generation of word trains are widespread but bad speech language paradigms because they tend to test fluency more than speech and yield too poorly lateralized activations. However, letter-cued word generation has still helped me out in some cases which did not activate well in other tasks.
Block designs are usually best for clinical purposes.
Hope that helps a bit;)
Cheers-
Andreas
________________________________
Von: FSL - FMRIB's Software Library im Auftrag von Alexander J. Shackman
Gesendet: Do 12.02.2009 00:51
An: [log in to unmask]
Betreff: Re: [FSL] "standardised" fMRI tasks.
deanna barch at wash u (st louis) has written about this
http://ccpweb.wustl.edu//pdfs/paradigm08.pdf
hth, alex
On Wed, Feb 11, 2009 at 5:32 PM, Andrew Janke <[log in to unmask]> wrote:
Hi,
There is a lot info out there about why we should use standardised tasks (eg):
http://xwiki.nbirn.net:8080/xwiki/bin/view/Function-BIRN/MultisiteFMRIDevelopment#
and other places such as here:
http://www.imagilys.com/clinical-fmri-paradigms/
Can anyone point me in the direction of some papers (with the actual
tasks included) or a set of such tasks on the web somewhere for use in
some research in surgical planning. (ie: motor, sensory, speech, etc).
Yes I realise that there are difficulties in making a one-size-fits
all approach for such things but surely there is a lowest common
denominator for such things.
For example there seems to be a consensus that "Finger, Toe, Tongue"
tasks for the motor cortex are a "good thing". -- [1] But of course
you could deliver this via an auditory system or visual. etc..
FWIW, for now in my current "ghetto-tech" fMRI setup that I bashed
together for a Masters student uses OpenOffice to export the tasks to
PDF which is then "played back" using auto-advance with a pdf viewer.
For simple rArArA type tasks (motor, sensory, visual) it works fine.
Thanks
--
Andrew Janke - [log in to unmask]
Department of Geriatric Medicine, ANU
Canberra->Australia +61 (402) 700 883
1 - Neurol Res. 2008 Nov;30(9):968-73. Epub 2008 Jul 30.
--
Alexander J. Shackman, Ph.D.
Laboratory for Affective Neuroscience
Waisman Laboratory for Brain Imaging & Behavior
University of Wisconsin-Madison
1202 West Johnson Street
Madison, Wisconsin 53706
Telephone: +1 (608) 358-5025
Fax: +1 (608) 265-2875
Email: [log in to unmask]
http://psyphz.psych.wisc.edu/~shackman
|