 |
 |
Dear Elisabeth, please can you check whether there exists an error-folder? Best, Christian On Thu, 3 Mar 2016 17:13:23 +0100, Elisabeth Kaminski <[log in to unmask]> wrote: >Dear Christian, > >thank you very much for your quick reply. Unfortunately, it didn´t create a report folder, it simply created the rXXXfiles without further notifications. If the program tries to run the segmentation but fails, would it usually give an error report? Otherwise, could it be an installation issue? > >Thank you also for your comments on the bias regularisation. I will change the defaults to a very light bias regularisation even though the MP2Rage images are calculated in a way that should cancel out any field bias. > >Best, >Elisabeth > > > > >----- Original Message ----- >From: "christian" <[log in to unmask]> >To: [log in to unmask], "Elisabeth Kaminski" <[log in to unmask]> >Cc: "christian" <[log in to unmask]> >Sent: Thursday, March 3, 2016 2:52:56 PM >Subject: Re: cat12 longitudinal tlbx processing > >Dear Elisabeth, > >On Wed, 2 Mar 2016 15:30:24 +0000, Elisabeth Kaminski <[log in to unmask]> wrote: > >>Dear SPMers, dear Christian Gaser, >> >>I am currently trying to feed your newly released cat12 toolbox (most recent version) with MP2Rage Images, I acquired. >> >>(1) I am just using the default settings and want to first segment my longitudinal data. However, when starting the batch, only the rigid registration is applied (so rXXX images and the avg image are created) but no segmented images appear. As there is no error message, I assume that the images I put in are fine for doing the analysis. >> >>This is the only output it creates: >> >>Running job #1 >>------------------------------------------------------------------------ >>Running 'Segment longitudinal data' >> >>------------------------------------------------------------------------ >>Running job #2 >>------------------------------------------------------------------------ >>Running 'Longitudinal Rigid Registration' >>Estimated noise sd for ".../XXX_pre.nii,1" = 386.031 >>Estimated noise sd for ".../XXX_post.nii,1" = 382.877 > >Please can you send me the xml-file that is saved in the report-folder. > >> >>(2) Additionally, I need to change the default settings. In the MP2Rage images, you don´t need to apply the bias regularisation as no bias field corrections are necessary. That's why, in the cat_defaults.m file I changed the following parameters: >> >>cat12.opts.biasreg = 0; >>cat12.opts.biasfwhm = inf; >> >>Assuming that this is correct, I ran my defaults file and it said: >> >>Warning: Can not find executable node 'spm.tools.cat.long' - running matlabbatch without default node. >> >>It didn't do anything else than that. Is there any way to avoid this issue? > >I couldn't confirm this error. However, even for data without obvious signal inhomogeineities I would strongly recommend to always use the bias correction that will utilize segmentation also for slight inhomogeneities. Furthermore, the setting to biasreg = 0 seems suboptimal because that would mean to have no regularization which is not recommended. > >Best, > >Christian > >> >> >>Are there any suggestions on that? Thank you very much in advance. Any help is highly appreciated. >>Elisabeth >> >>