Hello
I think you need to compile SPM5 in order to make it work with your
matlab version. (change the variable MEX_BIN in spm5/src/Makefile file).
Hope this helps
Elise
Chris Watson a écrit :
> Sorry for the bombardment. I tried displaying even the template images
> that come with SPM5 (e.g. EPI.nii), and I get:
> Running "Display Image"
> Image "/occipital/matlab/spm5/templates/EPI.nii" can not be resampled
> Image "/occipital/matlab/spm5/templates/EPI.nii" can not be resampled
>
> I'm all out of ideas, too.
>
> Chris Watson wrote:
>> A little more information.
>> We use Freesurfer to convert our DICOM images (from GE 3T scanner) to
>> Analyze 7.5 format. Should I be converting to nifti instead? I get
>> the same error when manually trying to do, e.g. slice timing using
>> the GUI (that is, bad image handle dimensions). So, as of right now,
>> I guess I can't view previous results or analyze new data.
>>
>> If it helps any, I attached a text file of the output to
>> 'analyze75info('foo.img')', and output of commands per this email :
>> http://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind0710&L=SPM&D=0&P=29198
>>
>> Happy Saturday,
>> Chris
>>
>> Chris Watson wrote:
>>> I downloaded the most recent update, and untarred it. Now when I try
>>> to view an SPM.mat, I get the following error message:
>>> ??? Error using ==> spm_sample_vol
>>> Bad image handle dimensions.
>>>
>>> Error in ==> spm_get_data at 41
>>> Y(i,:) = spm_sample_vol(V(i),XYZ(1,:),XYZ(2,:),XYZ(3,:),0);
>>>
>>> Error in ==> spm_getSPM at 540
>>> Z = min(Z,spm_get_data(xCon(i).Vspm,XYZ));
>>>
>>> Error in ==> spm_results_ui at 274
>>> [SPM,xSPM] = spm_getSPM;
>>>
>>> ??? Error while evaluating uicontrol Callback
>>>
>>> This won't require re-analyzing all the data, will it?
>>> Chris
>>>
>>> Simon Durrant wrote:
>>>> Hi Chris,
>>>>
>>>> The fast and dirty solution is simply to turn the warning off by
>>>> entering the following at the command prompt:-
>>>>
>>>> warning('off','MATLAB:FINITE:obsoleteFunction')
>>>>
>>>> This will stop it slowing down the processing. It's caused by the
>>>> Matlab guys changing the "finite" function name to "isfinite" to
>>>> make it more consistent with their other type-check functions, but
>>>> unfortunately all code using the old version subsequently gets a
>>>> warning. I'm sure the SPM authors know about it and will be
>>>> updating that and many other version-specific issues (mex files
>>>> compiled with older versions of Matlab are another notable example)
>>>> in SPM8.
>>>>
>>>> Best,
>>>>
>>>> Simon
>>>>
>>>>
>>>>
>>>> ------------------------------------------------------------------------
>>>>
>>>>
>>>> > Date: Fri, 8 Aug 2008 12:17:56 -0400
>>>> > From: [log in to unmask]
>>>> > Subject: [SPM] Warning, using Matlab 7.6 (R2008a)
>>>> > To: [log in to unmask]
>>>> >
>>>> > Hello,
>>>> > While doing preprocessing for a dataset, I receive this warning
>>>> (I just
>>>> > upgraded to Matlab 7.6):
>>>> >
>>>> > Warning: FINITE is obsolete and will be removed in future
>>>> versions. Use
>>>> > ISFINITE instead.
>>>> > > In file_array.subsasgn>subfun at 95
>>>> > In file_array.subsasgn at 63
>>>> > In spm_write_plane at 33
>>>> > In spm_slice_timing at 261
>>>> > In spm_config_slice_timing>slicetiming at 170
>>>> > In spm_jobman>run_struct1 at 1540
>>>> > In spm_jobman>run_struct1 at 1548
>>>> > In spm_jobman>run_struct1 at 1548
>>>> > In spm_jobman>run_job at 469
>>>> > In spm_jobman at 85
>>>> >
>>>> > This warning is spit out for just about every slice, and slows
>>>> down the
>>>> > process quite a bit. This is what a tech support guy from The
>>>> Mathworks
>>>> > had to say: (I originally thought the Matlab built-in 'subsasgn'
>>>> was to
>>>> > blame)
>>>> >
>>>> > "It is true that FINITE function has been deprecated in MATLAB
>>>> and we
>>>> > promote using ISFINITE instead of FINITE. As you mentioned that
>>>> spm_*
>>>> > files are from the SPM5 package, you will probably need to
>>>> contact SPM
>>>> > to incorporate the necessary changes about replacing the FINITE
>>>> > functions to INFINITE in the files.
>>>> > From the error messages, it does not seem that SUBASGN is using
>>>> FINITE.
>>>> > Although it is hard to say by looking at the error messages
>>>> alone, but
>>>> > it looks like SUBASGN has been overloaded for the FILE_ARRAY
>>>> object in
>>>> > SPM5 package. Hence, in this case as well, you will probably need to
>>>> > contact SPM for further information."
>>>> >
>>>> > I already changed all instances of "finite" to "isfinite" in the spm
>>>> > files. Is there anything else to be done? (I'm using SPM5 on
>>>> RHEL4, for
>>>> > reference).
>>>> >
>>>> > Thanks,
>>>> > Chris Watson
>>>>
>>>>
>>>> ------------------------------------------------------------------------
>>>>
>>>> Find out how to make Messenger your very own TV! Try it Now!
>>>> <http://clk.atdmt.com/UKM/go/101719648/direct/01/>
>>>
>>>
>>
>> ------------------------------------------------------------------------
>>
>>
>>>> analyze75info('/raid2/fmri5/realsubs_visclos_stroop/CARDIAC/metcalf_b/visclos/vc/bold/image001.img')
>>>>
>>>>
>>
>> ans =
>>
>> Filename:
>> '/raid2/fmri5/realsubs_visclos_stroop/CARDIAC/metcalf_b/visclos/vc/bold/image001.hdr'
>>
>> FileModDate: '08-Aug-2008 14:46:51'
>> HdrFileSize: 348
>> ImgFileSize: 196608
>> Format: 'Analyze'
>> FormatVersion: '7.5'
>> Width: 64
>> Height: 64
>> BitDepth: 16
>> ColorType: 'grayscale'
>> ByteOrder: 'ieee-le'
>> HdrDataType: ''
>> DatabaseName: ''
>> Extents: 0
>> SessionError: 0
>> Regular: 0
>> Dimensions: [64 64 24 1]
>> VoxelUnits: 'mm'
>> CalibrationUnits: ''
>> ImgDataType: 'DT_SIGNED_SHORT'
>> PixelDimensions: [3.7500 3.7500 6 2]
>> VoxelOffset: 0
>> CalibrationMax: 0
>> CalibrationMin: 0
>> Compressed: 0
>> Verified: 0
>> GlobalMax: 8191
>> GlobalMin: 0
>> Descriptor: ''
>> AuxFile: ''
>> Orientation: 'Transverse flipped'
>> Originator: ' ''
>> Generated: ''
>> Scannumber: ''
>> PatientID: ''
>> ExposureDate: ''
>> ExposureTime: ''
>> Views: 0
>> VolumesAdded: 0
>> StartField: 0
>> FieldSkip: 0
>> OMax: 0
>> OMin: 0
>> SMax: 0
>> SMin: 0
>>
>>
>>>> N =
>>>> nifti('/raid2/fmri5/realsubs_visclos_stroop/CARDIAC/metcalf_b/visclos/vc/
>>>>
>>>> bold/image001.img');
>>>> N.dat
>>>>
>>
>> ans =
>>
>> fname: [1x83 char]
>> dim: [64 64 24]
>> dtype: 'INT16-LE'
>> offset: 0
>> scl_slope: 1
>> scl_inter: 0
>>
>>
>>
>>>> size(N.dat)
>>>>
>>
>> ans =
>>
>> 64 64 24
>>
>
--
Elise Metereau, doctorante
Equipe "Neuroimagerie Cognitive: Prise de décisions et récompenses"
CNRS UMR5229 - Centre de Neuroscience Cognitive
67 Bd Pinel, 69675 Bron
tel: 00 33 (0)4 37 91 12 43
fax: 00 33 (0)4 37 91 12 10
http://www.isc.cnrs.fr/dre/
|