Hi Ce,
I hope you dont mind me replying your question to the list (cc) as well, others might be better able to help you here.
It is very hard to answer such general questions about beta images. IMHO there is no need to 'extract' anything, you have your beta images, and they hold the estimated regression coefficients for your model per factor per voxel. For example, click on a voxel after using 'display' a beta image to see their coordinates and values, or use spm_vol/spm_read_vols to inspect them. What they 'mean' depends on what you model in your GLM analysis, so you really are the only person who can answer that.
NaNs stands for 'not a number' and have no interpretation either. The only thing I can say for sure is that they do not represent a number ;-)
At that place your model couldnt be/wasn't estimated, either because you masked them out or for a number of other conceivable reasons.
Usually people do not look at beta values directly, but at the ensuing t- or f-maps testing some hypothesis about brain function. Only when you need to compute local %signal change they could be of interest. Check the many messages on the list for discussions on %signal change measures.
And how you define an ROI is completely up to you as well. Usually people define it a-priori, which makes a lot of sense to avoid circular reasoning and inflated statistics. You could do it anatomically or based on an (independent) functional 'localizer' session. Just don't draw a spherical ROI around your maximally activated voxel for a contrast and call that your ROI. That would be invalid when you do not have an a priori hypothesis about that region with that center voxel.
Good luck,
Bas
--------------------------------------------------
Dr. S.F.W. Neggers
Division of Brain Research
Rudolf Magnus Institute for Neuroscience
Utrecht University Medical Center
Visiting : Heidelberglaan 100, 3584 CX Utrecht
Room B.01.1.03
Mail : Huispost B01.206, P.O. Box 85500
3508 GA Utrecht, the Netherlands
Tel : +31 (0)88 7559609
Fax : +31 (0)88 7555443
E-mail : [log in to unmask]<mailto:[log in to unmask]>
Web : http://www.neuromri.nl/people/bas-neggers
: http://www.brainsciencetools.com (CEO)
--------------------------------------------------
________________________________
From: [log in to unmask] [[log in to unmask]]
Sent: Tuesday, April 10, 2012 12:27 PM
To: Neggers, S.F.W.
Subject: 回复:Re: [SPM] How to extract voxel-wise beta values
Dear Bas,
Thank you so much for your kind help and advice. I really appreciate it.
In fact, the reason I wanna do this is that I would like to extract the pattern of activation within a certain ROI ( i.e. voxel-by voxel beta values instead of a binary (de)activation/non-activation mapping. To achieve that, two important questions unclear to me:
1) How do I understand or inteprete the NaN (beta) values of certain voxels? If I understand it correctly, these are not negligible because an extracted pattern from an ROI might contain NaNs, which I assume, need to be dealt with. So should I regard it as zeros or some non-sense values or sth else?
2) In many cases, we might need to define an ROI based on high resolution structural images. So is there any difference between the way we define the ROI on functional images and the way we define structural images?
Many thanks and best regards,
Yours sincerely and respectfully
Ce
----- 原始邮件 -----
发件人:Bas Neggers <[log in to unmask]>
收件人:<[log in to unmask]>
抄送人:<[log in to unmask]>
主题:Re: [SPM] How to extract voxel-wise beta values
日期:2012年04月05日 15点28分
[http://service.t.sina.com.cn/widget/qmd/1503902641/1bd88f6e/1.png]<http://t.sina.com.cn/1503902641?s=bW9jaGVjaw==>
------------------------------------------------------------------------------
De informatie opgenomen in dit bericht kan vertrouwelijk zijn en is
uitsluitend bestemd voor de geadresseerde. Indien u dit bericht onterecht
ontvangt, wordt u verzocht de inhoud niet te gebruiken en de afzender direct
te informeren door het bericht te retourneren. Het Universitair Medisch
Centrum Utrecht is een publiekrechtelijke rechtspersoon in de zin van de W.H.W.
(Wet Hoger Onderwijs en Wetenschappelijk Onderzoek) en staat geregistreerd bij
de Kamer van Koophandel voor Midden-Nederland onder nr. 30244197.
Denk s.v.p aan het milieu voor u deze e-mail afdrukt.
------------------------------------------------------------------------------
This message may contain confidential information and is intended exclusively
for the addressee. If you receive this message unintentionally, please do not
use the contents but notify the sender immediately by return e-mail. University
Medical Center Utrecht is a legal person by public law and is registered at
the Chamber of Commerce for Midden-Nederland under no. 30244197.
Please consider the environment before printing this e-mail.
|