Hi,
I've used loopcheck and termination mask.I use termination mask option (--stop)for my target mask? I use the probtrackx in command line.how can I use --os2t or --opd?
when I write it (--os2t)simply (without nothing after that) in the command line I get the error message "cmasks does not exist".
many thanks in advance and regards,
Shiva
-----Original Message-----
From: FSL - FMRIB's Software Library on behalf of Andreas Bartsch
Sent: Tue 5/13/2008 14:46
To: [log in to unmask]
Subject: [FSL] AW: [FSL] AW: [FSL] tractography
Hi,
are you running seperate tractographies to compare the waytotal values? I can hardly believe you get identical numbers for these! Also: Have you used loopcheck and termination / exclusion masks to exclude pathways that don't make sense? Have you set the --os2t flag to get the seed_to_(blah-target) image?
Cheers-
Andreas
PS: I guess Tim or Saad will jump in at some point, e.g. once the above issues have been clarified.
________________________________
Von: FSL - FMRIB's Software Library im Auftrag von Keihaninejad, Shiva
Gesendet: Di 13.05.2008 12:02
An: [log in to unmask]
Betreff: Re: [FSL] AW: [FSL] tractography
Hi Andreas,
many thanks for your respond. Sorry I don't know Tim. how can I access to his suggestion about tractography.
the value of waytotal is not accurate at all. for example, I'm checking the connectivity between hippocampus and Parahippocampul gyri.It has the same value for ipsilateral and contra lateral structures(left hippocampus with left Parahippocampul gyri and left hippocampus with right Parahippocampul gyri)
when I check the fdt_path image of these ipsilateral and contra lateral structures, I find some paths that come out from my seed mask but don't reach the target mask and go somewhere else.do you know what the reason is??
Sorry I don't get this sentence " the number or spatial distribution of seed voxels connected to the target and the number of samples that reach the target from a seed voxel found to be connected to the target may contain valuable information for you - so get the seed_to_(blah-target) image and check that out" how can I achieve that?
Many thanks in advance,
Shiva
-----Original Message-----
From: FSL - FMRIB's Software Library on behalf of Andreas Bartsch
Sent: Fri 5/9/2008 13:34
To: [log in to unmask]
Subject: [FSL] AW: [FSL] tractography
Hi,
the waytotal is the number of samples making it to the target from the seed. It's written into a file (waytotal, not a directory).
To some extent, you'd expect the number to increase the larger the seed mask. So if patients differ from controls, for example by atrophy of the seed and / or target region, waytotals may be lowered. But aside from the atrophy, the connectivity may still be of the same strength like in normals. On the other hand, more sample may reach target even from smaller seeds. So per se, the waytotal is hard to interpret if you can't relate it to, for example, the contralateral side or, possibly, another integer tract known to be not affected by the disease.
However, if you know that the connection in question does exist you may relate the connectivity output (fdt_path) to the waytotal number, simply in dividing the former by the latter. Such prior knowledge can be established, for example, by testing motor or visual performance when the pyramidal tract or optic radiation is under investigation. As Tim explain earlier, the normalized connectivity output (fdt_path / waytotal) is, in theory, not or much less dependent on various other variables such as mask size etc. So that may be an image to look at (using fslstats). Also the number or spatial distribution of seed voxels connected to the target and the number of samples that reach the target from a seed voxel found to be connected to the target may contain valuable information for you - so get the seed_to_(blah-target) image and check that out.
Still, it may not be what you want. The question is also if you are after an overall measure or if you want to compare groups across space etc. Unfortunately, there is no general recipe to do these kind of things. But Tim, Heidi or Saad may eloborate further.
Hope it helps a bit,
Cheers-
Andreas
________________________________
Von: FSL - FMRIB's Software Library im Auftrag von Keihaninejad, Shiva
Gesendet: Fr 09.05.2008 13:37
An: [log in to unmask]
Betreff: [FSL] tractography
Dear All,
Could you help me with my problem:
I'm using the FSl tractography. I use single seed mask and termination mask for target. The output is a single image and a directory (waytotal) that has a number of generated tracts. What should I do for checking the connectivity? Shall I call fdt_path image and FA image in fslview?
If I want to study the connectivity for patients and normal subjects and compare them, can I use the number in waytotal?
Many thanks in advance,
Shiva
|