Hi - these tracts you sent are in MNI space and look plausible to me.


Saad





On 26 Oct 2016, at 13:42, Noelia Martinez-Molina <[log in to unmask]> wrote:

Hi Saad, 

I followed all your suggestions: I did fslreorient2std and then used BET Robust brain centre estimation (iterates bet2 several times) and provided the centre of the brain in Voxels.

Unfortunately, I'm still having the same issue (tract outside the MNI template). Please find a link to a folder with the fdt_paths.nii.gz output and the T1 after reorientation and BET. 


I'd appreciate if you could tell me what can have gone wrong. 

Thank you for your help and kindness,
Noelia

2016-10-26 14:32 GMT+02:00 Noelia Martinez-Molina <[log in to unmask]>:
Hi Saad, 

I followed all your suggestions: I did fslreorient2std and then use BET Robust brain centre estimation (iterates bet2 several times) and provided the centre of the brain in Voxels.

Unfortunately, I'm still having the same issue. Please find attached the fdt_paths.nii.gz output and a link to the T1 after reorientation and BET. ​
 struct_re_brain_thr0.2.nii
​ I'd appreciate whether you could tell me what went wrong. 

Thank you for your time and help,
Noelia

2016-10-26 13:04 GMT+02:00 Noelia Martinez-Molina <[log in to unmask]com>:
Thank you Saad, I'll do as you say and let you know if this solved the issue.

Best,
Noelia

2016-10-26 12:33 GMT+02:00 Saad Jbabdi <[log in to unmask]>:
Almost:  use fslreorient2std straight away, then use BET (I recommend providing the centre of the brain if you are only dealing with a single brain). If you are dealing with multiple brains and do not want to provide the centre of each brain individually, one trick I use is to average all the non-brain extracted subjects and get an average brain centre that I use for all subjects. If the protocol is similar between subjects this works pretty well.

Cheers
Saad




On 26 Oct 2016, at 11:28, Noelia Martinez-Molina <[log in to unmask]OM> wrote:

Hi Saad,

thanks again for your advice. If I understood it correctly, what I should do now is to repeat BET by providing as input the original T1 (with all non-brain matter) and then choosing a variation of bet2 called Robust brain centre estimation (iterates bet2 several times). Is this what you are suggesting? And then use fslreorient2std before fdt registration. 

Best,
Noelia



2016-10-26 12:07 GMT+02:00 Saad Jbabdi <[log in to unmask]>:
Hi - as I suspected there is a lot of neck left on the brain extracted T1, which is probably why FLIRT failed.

The easiest way to make BET work is to provide it with the centre of the brain (somewhere between the thalami).  

Also you should use fslreorient2std to get your T1 correctly reoriented. 

Cheers
Saad



On 26 Oct 2016, at 10:58, Noelia Martinez-Molina <[log in to unmask]OM> wrote:

Dear Saad, 

Thank you very much for your prompt reply. I did BET on my T1 and reoriented it and the no diffusion image to the ac. Please find a link with the brain extracted T1 that I used for registration. 

Hope this can help.

Best regards,
Noelia​
 struct.nii
​

2016-10-26 11:43 GMT+02:00 Saad Jbabdi <[log in to unmask]>:
Hi - one likely reason why registrations didn’t work is perhaps you did not provide a brain extracted T1, or the brain extraction on the T1 did not work. Please check if that is the case.

Cheers
Saad





On 26 Oct 2016, at 09:10, Noelia Martinez-Molina <[log in to unmask]OM> wrote:

Dear Saad,

Thank you for your email and sorry for the belated reply. In my previous email there was a mistake, I provided as input the transformation matrix standard2diff.mat not diff2standard.mat as I wrote, with the results I commented, i.e. tracts passing through voxels outside the MNI template. 

I was wondering if there is any way to check whether the matrices are correct. For example, I overlayed the no_diff and T1 images in FSL view after the FDT registration and, if I'm not mistaken, they should be in register which is not the case. I interpret this as indicative of an incorrect diff2str.mat which could then lead to the results I observed after probtrackx.

I hope I explained it more clearly now and would really appreciate if you could indicate me what steps to take in order to solve this issue.

Thank you in advance and have a good day,
Noelia 

2016-10-24 18:22 GMT+02:00 Saad Jbabdi <[log in to unmask]>:
Hi -  you need to provide the transformation standard space —> diffusion space (and not diff2standard.mat as you did).  Note that in case you are using nonlinear warping, you need to provide both transforms to probtrackx.

Cheers
Saad






On 24 Oct 2016, at 14:27, Noelia Martinez Molina <[log in to unmask]OM> wrote:

Dear FSL experts,

I'm running probabilistic tractography for the first time and would appreciate some help regarding an issue with FDT registration. For probtrackx, I used a seed in MNI space and so need to include as input the transformation matrix diff2standard.mat that I get after FDT registration. However, after doing this the results show tracts passing through voxels outside the MNI template!

My guess is that the step of registration from diffusion to structural space (with the brain extracted images: no diffusion and  T1) went wrong. This already happened to me when I did the coregister between the T1 and fMRI images in the same dataset and was solved by placing the origin in the anterior commisure (a.c) in both the structural and functional images. So I did the same with the no diffusion, i.e place the origin in the a.c, but still got misplaced voxels with tracts going outside the brain.

I was considering skipping the intermediate step of registration and go straight from the diffusion to the standard space and use this transformation matrix in probtrackx. Is this transformation correct or should I always go from diffusion to structural and then to standard space? If so, could you give me some advice as to how to solve this problem?

Thank your very much in advance for your time and help!
Noelia Martínez-Molina




--

Noelia Martínez-Molina
Cognition and Brain Plasticity Group
Department of Basic Psychology
Universitat de Barcelona
Campus Mundet, Edifici de Ponent
Passeig de la Vall d'Hebron 171
08035 Barcelona, Spain

Office: (34) 93 403 47 68
twitter:@NoeliaMrtnez

---------------------------------------------------------------
Without music, life would be a mistake. 
                                   Friedrich Nietzsche
----------------------------------------------------------------





--

Noelia Martínez-Molina
Cognition and Brain Plasticity Group
Department of Basic Psychology
Universitat de Barcelona
Campus Mundet, Edifici de Ponent
Passeig de la Vall d'Hebron 171
08035 Barcelona, Spain

Office: (34) 93 403 47 68
twitter:@NoeliaMrtnez

---------------------------------------------------------------
Without music, life would be a mistake. 
                                   Friedrich Nietzsche
----------------------------------------------------------------





--

Noelia Martínez-Molina
Cognition and Brain Plasticity Group
Department of Basic Psychology
Universitat de Barcelona
Campus Mundet, Edifici de Ponent
Passeig de la Vall d'Hebron 171
08035 Barcelona, Spain

Office: (34) 93 403 47 68
twitter:@NoeliaMrtnez

---------------------------------------------------------------
Without music, life would be a mistake. 
                                   Friedrich Nietzsche
----------------------------------------------------------------





--

Noelia Martínez-Molina
Cognition and Brain Plasticity Group
Department of Basic Psychology
Universitat de Barcelona
Campus Mundet, Edifici de Ponent
Passeig de la Vall d'Hebron 171
08035 Barcelona, Spain

Office: (34) 93 403 47 68
twitter:@NoeliaMrtnez

---------------------------------------------------------------
Without music, life would be a mistake. 
                                   Friedrich Nietzsche
----------------------------------------------------------------




--

Noelia Martínez-Molina
Cognition and Brain Plasticity Group
Department of Basic Psychology
Universitat de Barcelona
Campus Mundet, Edifici de Ponent
Passeig de la Vall d'Hebron 171
08035 Barcelona, Spain

Office: (34) 93 403 47 68
twitter:@NoeliaMrtnez

---------------------------------------------------------------
Without music, life would be a mistake. 
                                   Friedrich Nietzsche
----------------------------------------------------------------




--

Noelia Martínez-Molina
Cognition and Brain Plasticity Group
Department of Basic Psychology
Universitat de Barcelona
Campus Mundet, Edifici de Ponent
Passeig de la Vall d'Hebron 171
08035 Barcelona, Spain

Office: (34) 93 403 47 68
twitter:@NoeliaMrtnez

---------------------------------------------------------------
Without music, life would be a mistake. 
                                   Friedrich Nietzsche
----------------------------------------------------------------