You could try using MRIConvert to make the
bvecs. If you did not rotate the FOV on the scanner, you should be able to
just use the gradients in the sequence (though you might have to change the
signs of one of the rows, so check that the V1 in lines mode looks like the
anatomy should look in the corpus callosum and corona radiata).
Peace,
Matt.
From:
Sent: Wednesday, March 25, 2009
12:41 PM
To: [log in to unmask]
Subject: Re: [FSL] white FA maps
Hi,
I use "dcm2niiGUI" to transfer the DICOM images to Nifti, and we
could have the bvals and bvecs automatically. I don't know
why all the other data have the right bvecs, but the
three cases have the same bvecs in the 32 directions repeated over
and over. Is there anyway to get the right bvecs?
Best Regards,
Yan
Date: Wed, 25 Mar 2009 12:24:15
-0500
From: [log in to unmask]
Subject: Re: [FSL] white FA maps
To: [log in to unmask]
Your bvecs file contains a b0 and the same
b600 direction repeated over and over. Presumably you scanned different
32 directions, so you will need to list them each in the bvecs file, rather
than listing direction #1 32 times.
Peace,
Matt.
From:
Sent: Wednesday, March 25, 2009
12:18 PM
To: [log in to unmask]
Subject: Re: [FSL] white FA maps
Hi:
Please check the bvecs and bvals in the attachment.
As the problem of interpolation, I could discuss it with our physicists.
Thanks,
Regards,
yan
Date: Wed, 25 Mar 2009 12:08:43
-0500
From: [log in to unmask]
Subject: Re: [FSL] white FA maps
To: [log in to unmask]
Could you post your bvecs and bvals files
also?
In future studies, you will want to avoid
having scanner interpolation on (the cause of your highly anisotropic voxels
0.85X0.85X2.3mm), as this is not ideal for tractography studies. This is
not the cause of your current issue, however.
Peace,
Matt.
From:
Sent: Wednesday, March 25, 2009
12:00 PM
To: [log in to unmask]
Subject: Re: [FSL] white FA maps
Hi,
I have checked by bedpostx_datacheck, but I don't know what it means... Here is
the results:
[yan@localhost ~]$
/opt/fsl/bin/bedpostx_datacheck /mnt/hgfs/Data/Yan/NeonatalDTI/AKBULUT/data/
/mnt/hgfs/Data/Yan/NeonatalDTI/AKBULUT/data//data
data_type FLOAT32
dim1 116
dim2 160
dim3 45
dim4 33
datatype 16
pixdim1 0.8593750000
pixdim2 0.8593750000
pixdim3 2.3000710011
pixdim4 1.0000000000
cal_max 0.0000
cal_min 0.0000
file_type NIFTI-1+
/mnt/hgfs/Data/Yan/NeonatalDTI/AKBULUT/data//nodif_brain_mask
data_type INT16
dim1 116
dim2 160
dim3 45
dim4 1
datatype 4
pixdim1 0.8593750000
pixdim2 0.8593750000
pixdim3 2.3000710011
pixdim4 1.0000000000
cal_max 0.0000
cal_min 0.0000
file_type NIFTI-1+
num lines in
/mnt/hgfs/Data/Yan/NeonatalDTI/AKBULUT/data//bvals
1
num words in /mnt/hgfs/Data/Yan/NeonatalDTI/AKBULUT/data//bvals
33
num lines in /mnt/hgfs/Data/Yan/NeonatalDTI/AKBULUT/data//bvecs
3
num words in /mnt/hgfs/Data/Yan/NeonatalDTI/AKBULUT/data//bvecs
99
Regards,
Yan
Date: Wed, 25 Mar 2009 17:28:20
+0100
From: [log in to unmask]
Subject: Re: [FSL] white FA maps
To: [log in to unmask]
Ok, I see.
So seems that dtifit did not work...
Have you tried
bedpostx_datacheck <dtifit_directory>
?
Regards,
Markus
2009/3/25 cathyliu <[log in to unmask]>
Hi,
The white FA maps means that the FA value is 1.06592
everywhere in the brain mask, so it is quite high anisotropic.
The eddy corrected data look perfect, and the brain mask covers
all the brain. As the FA maps are white, I don't continue to do the bedpostX
yet.
Thank you so much!
Regards,
yan
Date: Wed, 25 Mar 2009 15:56:33 +0100
From: [log in to unmask]
Subject: Re: [FSL] white FA maps
To: [log in to unmask]
Hi!
What do you mean by white FA maps?
Is fslstat FAmap -R [0 0]?
How do the eddy_corrected data look like? How the brain mask?
What does "bedpostx_datacheck" on the dtifit directory return?
Could you give more information, please?
All the best,
Markus
2009/3/25 cathyliu <[log in to unmask]>
Hi,
We have some problems to make good FA maps, and
L1,L2,L3,MD,V1,V2,V3...maps don't work either. We do the image
preparation following your instructions, it works in most of our
cases, but in 3 of them we have the white FA maps. We could view the oraginal
data 32 directions without any problem, and there is not error
information during the processing. We wonder if it is the problem of the
old scanner version, or the wrong gradient table which FSL doesn't
recognize. These three cases were examined in the machine:
Scanner software version: NT 9.1.1.3
scanner model: Gyroscan Intera
number of directions: 32
phase encoding direction: AP
Regards,
Yan
立刻下载 MSN 保护盾,保障 MSN 安全稳定! 现在就下载!
--
Dr. med. Markus Gschwind, M.D.
Laboratory for Neurology and Imaging of Cognition
Dept of Neurosciences
1 Michel-Servet - 1211
Tel 0041 (0) 22 379 5324
Fax 0041 (0) 22 379 5402
email: [log in to unmask]
http://labnic.unige.ch
更多热辣资讯尽在新版MSN首页! 立刻访问!
--
Dr. med. Markus Gschwind, M.D.
Laboratory for Neurology and Imaging of Cognition
Dept of Neurosciences
1 Michel-Servet - 1211
Tel 0041 (0) 22 379 5324
Fax 0041 (0) 22 379 5402
email: [log in to unmask]
http://labnic.unige.ch
更多热辣资讯尽在新版MSN首页! 立刻访问!
立刻下载 MSN 保护盾,保障 MSN 安全稳定! 现在就下载!
更多热辣资讯尽在新版MSN首页! 立刻访问!