Print

---

Tom & Ged,

Tom's code did work and gives me what SPM2 did (near enough - neglecting 
spatial norm differences etc). I hadn't appreciated the alternative to 
relative threshold masking to the global mean. This bit of code will 
threshold relative to the user entered vector rather than mean.

In this case I had large scale hypometabolism in frontal areas so the 
mask was cut down using this method as opposed to thresholding to 
global, but as Ged mentioned, its easy enough to create an implicit mask.

Thanks for your help. Problem solved.
Joel


Thomas Nichols wrote:
> Joel,
>
> If you change the lines
>     case 'tmr',
>         % Relative
>         M_T = job.masking.tm.tmr.rthresh*sqrt(-1);
>         % Need to force calculation of globals
>         iGXcalc=3;
> to
>     case 'tmr',
>         % Relative
>         M_T = job.masking.tm.tmr.rthresh*sqrt(-1);
>         % Need to force calculation of globals
>         if iGXcalc~=2, iGXcalc=3; end
> does that solve your problem?
>
> -Tom
>
> On 9/14/07, *Joel Dunn * <[log in to unmask] 
> <mailto:[log in to unmask]>> wrote:
>
>     Using SPM5 I am having trouble entering my own global
>     normalisation factors.
>     I 'm using "flexible factorial" to compare one FDG PET patient
>     with a group
>     of controls.
>
>     Under "Global Calculation" I have chosen "User" and entered my own
>     vector
>     (in this case mean occipital lobe values).
>
>     When I run this the "Global Calculation" changes - this is set in
>     SPM.xGX.iGXcalc = 3. It should be 2.
>
>     I've checked the results: the mean occipital lobe value should be
>     what I set
>     the proportional scaling to be (I checked this to be the case in
>     SPM2), but
>     its not - its normalised to global mean values.
>
>     I think iGXcalc is reset in spm_config_factorial_design.m, line
>     1411, where
>     it is set to = 3 when a relative threshold is set, and user
>     entered values
>     are then ignored.
>
>     I would still like to set a relative threshold to mask out low
>     intensities
>     as I believe I did in SPM2.
>
>     Can I do this? Am I doing something wrong?
>
>     Please help.
>     Regards
>     Joel Dunn
>
>
>
>
> -- 
> -Tom
>
> ____________________________________________
> Thomas Nichols, PhD
> Director, Modelling & Genetics
> GlaxoSmithKline Clinical Imaging Centre
>
> Senior Research Fellow
> Oxford University FMRIB Centre 

-- 
-----------------------------------------------
Joel Dunn
Research Associate
PET Imaging Centre
Guy's, King's and St Thomas' School of Medicine
Lower Ground Floor, Lambeth Wing,
St Thomas' Hospital, Lambeth Palace Road,
London SE1 7EH UK            Tel: 020 7188 7445
Email: [log in to unmask]   Fax: 020 7620 0790
-----------------------------------------------