Dear all,

I tried to use VBM2 by Gaser for longitudinal analysis in SPM2.

I was able to create the customised template, segment the images and specify the model.
However, when I tried to estimate it  I got the following error in Matlab:

****************************************************************
ane 156/156, block   2/2              :                        ...done
Temporal non-sphericity (over voxels)   :             ...REML estimation
Warning: Matrix is close to singular or badly scaled.
         Results may be inaccurate. RCOND = 1.049247e-030
(Type "warning off MATLAB:nearlySingularMatrixUMFPACK" to suppress this warning.)
> In c:\documents and settingpc01831\my documents\spm2\spm2\spm_reml.m at line 77
  In c:\documents and setting1\my documents\spm2\spm2\toolbox\vbm2\spm_spm.m at line 812
Warning: Matrix is close to singular or badly scaled.
         Results may be inaccurate. RCOND = 5.246233e-031.
(Type "warning off MATLAB:nearlySingularMatrix" to suppress this warning.)
> In c:\documents and settings\my documents\spm2\spm2\spm_reml.m at line 82
  In c:\documents and settings\my documents\spm2\spm2\toolbox\vbm2\spm_spm.m at line 812

SPM2: spm_spm (v2.66)                              13:05:41 - 24/07/2009
========================================================================
Initialising parameters                 :                   ...computing??? Index exceeds matrix dimensions.

Error in ==> C:\MATLAB6.5.1\toolbox\matlab\sparfun\spdiags.m
On line 102  ==>       a = [a; i i+d(k) B(i+(m>=n)*d(k),k)];

Error in ==> c:\documents and settings\sc01831\my documents\spm2\spm2\toolbox\vbm2\spm_spm.m
On line 420  ==>  s     = spdiags(1./sqrt(diag(s)),0,nScan,nScan);

Error in ==> c:\documents and settings\s01831\my documents\spm2\spm2\toolbox\vbm2\spm_spm.m
On line 829  ==>  SPM = spm_spm(SPM);

??? Error while evaluating uicontrol Callback.
***************************************************************

I read the previous post  in https://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind06&L=SPM&P=R754340, and tried to estimate the same model in SPM5. However, the suggestion seems not to work in my case as I get a different error message:

***************************************************************************
PM5: spm_spm (v$Rev: 649 $)                       16:02:30 - 24/07/2009
========================================================================
Initialising parameters                 :                   ...computing??? Reference to non-existent field 'dt'.

Error in ==> spm_spm at 494
YNaNrep  = spm_type(VY(1).dt(1),'nanrep');

Error in ==> spm_spm at 299
        spm_spm(SPM);

??? Error using ==> SPM = spm_spm;
Reference to non-existent field 'dt'.

??? Error while evaluating uicontrol Callback

*********************************************************************


Has anyone else come across this? 

Regards,
Stefania



Stefania Benetti
PhD student
Section of Neuroimaging, Box P067
Institute of Psychiatry - King's College
De Crespigny Park
London SE5 8AF

tel. +44(0)20 7848 0284
fax.+44(0)20 7848 0976