Dear Dr Hamming Zhang -
Thank you for samples of your images. MRIcro and the other image converters
I know of are all designed to convert images that slices that all come from
the same frame of reference - e.g. the images all need to be sliced in a
single plane (whether it is sagittal, coronal, axial or oblique).
The images you sent me are composed of slices where the azimuth changes
with each successive slice (a 'cylindrical' view). These images are nice
for giving a 3D impression of the brain when viewed in quick succession,
but are not directly suitable for conversion to a 3D volume. Essentially,
some slices are coronal, some are sagittal and others are oblique views.
Perhaps other users of the SPM group know of reconstruction software that
can help you. If this helps anyone else, the images report being GE
"Starcam" emission images with an Implementation Version Name of "eNTEGRA
2.0215". With the cylindrical viewing described as "Type of Detector
Motion: STEP AND SHOOT".
It may also be worth contacting the manufacturer. In addition, it is
possible that this is not the best way to acquire data that you plan to
analyze with SPM.
-chris
At 03:07 02/12/2001 +0800, <Hamte zhang> <Hamte zhang> wrote:
>Dear spmers:
> I use MRIcro to open a set of Rcbf images from GE company¡¯s STARcam
> format(one group are patient,another is normal,one scan for each person).
> All the transverse image is normal, but all the coronal images like a
> broad ¡°Z¡± shape. The sagital images are all another strange separated
> shape. The size of the Rcbf image is about 2.0 Mb.
> The header information in MRIcro is
> Dimension size(mm) origin
>X 128 3.24 0
>Y 128 3.24 0
>Z 64 1.00 0
>Offset 4978 Scale 1.000
>
> I continue to use SPM99 to prossess the images . in spatial step of
> normalization, i couldn¡¯t get the correct results(only process one
> person once a time) , Matlab stop iteself . I have submitted the error
> message below:
>
>One kind of error message were:
>--------¡°
>Subject 1: 3D CT Norm:
>iteration 1: FWHM = 67.2 Var = 2034.53
>iteration 2: FWHM = 6.187 Var = 1.63275e-005
>iteration 3: FWHM = 4.043 Var = 2.74566e-006
>iteration 4: FWHM = 2.715 Var = 3.54755e-007
>iteration 5: FWHM = 3.403 Var = 1.78775e-007
>iteration 6: FWHM = 2.027 Var = 4.29714e-008
>iteration 7: FWHM = 0.919 Var = 2.15132e-009
>iteration 8: FWHM = 2.368 Var = 1.10975e-008
>iteration 9: FWHM = 0.5845 Var = 1.10359e-010
>Warning: Matrix is close to singular or badly scaled.
> Results may be inaccurate. RCOND = 1.074173e-016.
> > In D:\ SPMs\spm99\spm99\spm_snbasis.m at line 126
> In D:\ SPMs\spm99\spm99\spm_sn3d.m at line 771
> In D:\ SPMs\spm99\spm99\spm_sn3d.m at line 534
> iteration 10: FWHM = 0.5295 Var = 3.02684e-012
>Warning: Matrix is close to singular or badly scaled.
> Results may be inaccurate. RCOND = 2.860708e-018.
> > In D:\ SPMs\spm99\spm99\spm_snbasis.m at line 126
> In D:\ SPMs\spm99\spm99\spm_sn3d.m at line 771
> In D:\ SPMs\spm99\spm99\spm_sn3d.m at line 534
> iteration 11: FWHM = 0.7659 Var = 1.45939e-009
> iteration 12: FWHM = 1.9 Var = 6.63593e-009
> ??? Error using ==> save
>Variable 'formate\patient\TOMO14_DS_sn3d.mat' not found.
>
>Error in ==> D:\SPMs\spm99\spm99\spm_sn3d.m
>On line 788 ==> eval(['save ' matname ' mgc Affine Dims Transform MF MG
>-v4']);
>Error in ==> D:\SPMs\spm99\spm99\spm_sn3d.m
>On line
>534 ==> spm_sn3d(subj(i).P,subj(i).matname,bb,Vox,...
>??? Error while evaluating uicontrol Callback.
>¡±-------
>
>Another kind of error message(for part of patient) were:
>------- ¡°
>Warning: Matrix is singular to working precision.
> > In D:\ SPMs\spm99\spm99\spm_affsub3.m (spm_affsub1) at line 440
> In D:\ SPMs\spm99\spm99\spm_affsub3.m (spm_affsub2) at line 323
> In D:\ SPMs\spm99\spm99\spm_affsub3.m at line 233
> In D:\ SPMs\spm99\spm99\spm_sn3d.m at line 755
> In D:\ SPMs\spm99\spm99\spm_sn3d.m at line 534
> * - SPM99: spm_affsub3 --------------------------------------------
> There is not enough overlap in the images to obtain a solution.
> Please check that your header information is OK.
> ----------------------------------------- 11:40:52 - 01/12/2001
>??? Error using ==> spm_affsub3 (spm_affsub1)
>There is not enough overlap of the images to obtain a solution
>Error in ==> D:\ SPMs\spm99\spm99\spm_affsub3.m (spm_affsub2)
>On line 323 ==> [alpha_t, beta_t, chi2_t, W(im,:)]
>= ...
>Error in ==> D:\ SPMs\spm99\spm99\spm_affsub3.m
>On line 233 ==> [params] = spm_affsub2(ifun,VG,VF,VW,VW2,
>Hold,samp,params,free,pdesc,gorder,mean0,icovar0);
>Error in ==> D:\ SPMs\spm99\spm99\spm_sn3d.m
>On line 755 ==> p1 = spm_affsub3('affine3',VG,VF,1,6,p1,VW,VW2);
>Error in ==> D:\ SPMs\spm99\spm99\spm_sn3d.m
>On line
>534 ==> spm_sn3d(subj(i).P,subj(i).matname,bb,Vox,...
>??? Error while evaluating uicontrol Callback.
>¡±--------
>i wonder whether the raw data have some problem. But the data provider
>told me her procedure to check the patient and normal is correct. Is there
>anyone coming across this kind of problem? Any suggestion is greatly
>appreciated.
>
> greatly thanks.
> hamming zhang Ph.D CAMS&PUMCH
> EMAIL: [log in to unmask]
__________________________________________
Dr. Chris Rorden
Department of Psychology
University of Nottingham
Nottingham NG7 2RD, UK
Tel: +44 [0]115-951-5294
Fax: +44 [0]115-951-5324
http://www.psychology.nottingham.ac.uk/staff/cr1/
