Dear Paul,
On Tue, 29 Mar 2016 01:14:59 +0100, Paul Chou <[log in to unmask]> wrote:
>Dear Dr Gaser
>
>Thanks for you and your team provide this wonderful toolbox for neuroimage society. I used VBM8 toolbox in the past two years and wish to use the CAT12 in my future research. I face some problems and have some questions about this new toolbox. [I am using CAT 912, in SPM 6685 with MATLAB 2012a under MACI64 - OSX10.11.4]
>
>1. After T1 segmentation , I found the background noise level of bias corrected T1 image was elevated (I attached the corresponded figure of original T1 and bias corrected T1 image in this email). Is it normal or something wrong during the bias correction procedure ? In my past experience, the bias corrected T1 image looks normal in VBM8 toolbox.
The bias corrected image might be looking indeed confusing because the intensity normalization is adapting the values for gray and white matter and CSF, but the background intensity is sometimes also increased depending on your MR sequence. Furthermore, the background is also not de-noised in order to save computation time and therefore looks for some sequences very noisy. I will think about how to change that behavior although this will only affect the appearance of the bias corrected image and not the segmentation itself.
>
>2. For the writing options of the "DARTEL export" of "Bias, noise and intensity corrected T1 image" , whether I choose "Rigid" , "Affine" or "Both" , the CAT 12 toolbox only output the rigid bias corrected T1 image for me. I think it may be a bug for this writing options.
Thanks for pointing to that error which should be solved in the next releases.
>
>3. After segmentation , I found the warning message in the pdf report file. It stated that "cat_main:DiffTissue Peaks". Is it ok for further VBM analysis or do I need to deal with the corresponded problem for this T1 data ?
This warning usually occurs if signal intensity for CSF is close to that of background and can be ignored.
>
>4. How to extract the indices of the "image and Preprocessing Quality" in the report PDF file and XML file (I can't find the exact line of these QC indicies in XML file) ? I want to write up some codes to extract these indices to investigate the data quality for my dataset. In addition, are there any references describe the definition of these QC indices ?
In the "Check Sample Homogeneity" tool some quality measures can be used to plot and compare these measures across a sample. Check cat_stat_check_conv.m for getting some ideas how to read the xml-information:
xml = convert(xmltree(your_xml_file));
QM = [str2double(xml.qualityratings.NCR) str2double(xml.qualityratings.ICR) str2double(xml.qualityratings.IQR)];
NCR is the noise quality, ICR the bias quality and IQR the weighted overall quality measure.
>
>5. It seems CAT12 toolbox could identify the WMH out from T1 image, but after segmentation, I didn't find any output files corresponded to WMH (The input T1 file with some WMH near the ventricle). How to identify the WMH in CAT12 toolbox ? (I think I need to modify the option of "WM Hyperintensity Correction (WMHC) under "Extended options for CAT12 segmentation". Am I right ?)
Please note that this functionality is still experimental and not heavily tested. You can change the parameter "cat.extopts.WMHC" in cat_defaults to
2 (also correct segmentation to WM) or 3 (handle as separate class). The latter option also allows to output the WMH as separate file which is only available in the expert mode or by changing the parameter "cat.output.WMH.*".
>
>6. I also face the error when I want to perform "Surface and thickness estimation" (modify this option from No to Yes). The relevant error message list as follows:
This should be solved with release 914.
Best,
Christian
>
>=========================================================================
>Write result maps: 0s
>Surface and thickness estimation:
>lh:
> Thickness estimation (0.50 mm?):
> WM distance: 67s
> CSF distance: 74s
> PBT2x thickness: 23s
> Final Corrections: 199s
> Create initial surface
>Setting up...
>Starting main process...
>Done with 10 percent.
>Done with 20 percent.
>Done with 30 percent.
>Done with 40 percent.
>Done with 50 percent.
>Done with 60 percent.
>Done with 70 percent.
>Done with 80 percent.
>Done with 90 percent.
>Done with 100 percent.
>Made 533 adjustments.
>Extracting component with largest number of vertices...
>Vertices: 523888 Triangles: 1047804
> 6s
> Reduce surface to 100000 faces: /bin/bash: CAT_RefineMesh: command not found
>
>/bin/bash: CAT_SeparatePolygon: command not found
>
>/bin/bash: CAT_Surf2Sphere: command not found
>
> 0s
> Topology correction and surface refinement /bin/bash: CAT_FixTopology: command not found
>
>/bin/bash: CAT_DeformSurf: command not found
>
>/bin/bash: CAT_RefineMesh: command not found
>
>/bin/bash: CAT_DeformSurf: command not found
>
> 0s
> Spherical mapping with areal smoothing /bin/bash: CAT_Surf2Sphere: command not found
>
> 0s
> Spherical registration /bin/bash: CAT_WarpSurf: command not found
>
> 0s
> Thickness / Depth mapping
>------------------------------------------------------------------------
>CAT Preprocessing error: : ./Analyze_Field/CAT_12/1_Original_Data/ACN_00003_T1_C.nii,1
>------------------------------------------------------------------------
>[GIFTI] Loading of XML file /Users/PaulSmith/Analyze_Field/CAT_12/1_Original_Data/surf/lh.central.ACN_00003_T1_C.gii failed.
>------------------------------------------------------------------------
> 17 - read_gifti_file
> 74 - gifti
> 353 - cat_surf_createCS
> 1622 - cat_main
> 525 - cat_run_job
> 13 - cat_run_newcatch
> 233 - run_job
> 149 - cat_run
> 29 - cfg_run_cm
> 1688 - local_runcj
> 959 - cfg_util
> 703 - MenuFileRun_Callback
> 96 - gui_mainfcn
> 53 - cfg_ui
>------------------------------------------------------------------------
>
>Printing 'Graphics' figure to:
> /Users/PaulSmith/Research_Software/spm12/toolbox/cat12/CAT.maci64/spm_2016Mar27.ps
>Print 'Graphics' figure to:
> /Users/PaulSmith/Analyze_Field/CAT_12/1_Original_Data/report/catreport_ACN_00003_T1_C.pdf
>Failed 'CAT12: Segmentation'
>Error using read_gifti_file (line 17)
>[GIFTI] Loading of XML file /Users/PaulSmith/Analyze_Field/CAT_12/1_Original_Data/surf/lh.central.ACN_00003_T1_C.gii failed.
>In file "[log in to unmask]" (v6404), function "read_gifti_file" at line 17.
>In file "[log in to unmask]" (v6601), function "gifti" at line 74.
>In file "/Users/PaulSmith/Research_Software/spm12/toolbox/cat12/cat_surf_createCS.m" (v912), function "cat_surf_createCS" at line 353.
>In file "/Users/PaulSmith/Research_Software/spm12/toolbox/cat12/cat_main.m" (v912), function "cat_main" at line 1622.
>In file "/Users/PaulSmith/Research_Software/spm12/toolbox/cat12/cat_run_job.m" (v903), function "cat_run_job" at line 525.
>In file "/Users/PaulSmith/Research_Software/spm12/toolbox/cat12/cat_run_newcatch.m" (???), function "cat_run_newcatch" at line 13.
>In file "/Users/PaulSmith/Research_Software/spm12/toolbox/cat12/cat_run.m" (v903), function "run_job" at line 233.
>In file "/Users/PaulSmith/Research_Software/spm12/toolbox/cat12/cat_run.m" (v903), function "cat_run" at line 149.
>
>The following modules did not run:
>Failed: CAT12: Segmentation
>===========================================
>
>Sorry for my poor English and many naive questions.
>
>Best
>
>Paul
|