DTI fiber tracking

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

DTI fiber tracking

Travis S. Scharr
I am having some problems using the DTI programs in slicer3.6.  When I run the tutorial, everything comes out fine, but when I try to use my own data I get images that are obviously wrong (see attached image).  Could the issue be with my data itself?  I am using the included DICOM to Nrrd converter for my data.  With some other programs I have used it was necessary to enter a gradient table specific to the magnet used.  Is this necessary or am I possibly having other issues?   I can tell when I run the tutorial with my data that things start to appear different very early, during the tensor estimation phase.  I am relatively new to this software and don't really understand enough to troubleshoot what is actually happening.  Can anybody help me?


Travis Scharr
University of Mississippi Medical Center
Class of 2014

Individuals who have received this information in error or are not authorized to receive it must promptly return or dispose of the information and notify the sender. Those individuals are hereby notified that they are strictly prohibited from reviewing, forwarding, printing, copying, distributing or using this information in any way.


_______________________________________________
slicer-users mailing list
[hidden email]
http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users
To unsubscribe: send email to [hidden email] with unsubscribe as the subject

SlicerImageBad_1.png (68K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: DTI fiber tracking

Xiaodong Tao-2
Hello Travis,

It looks like the diffusion weighting directions in your data is not correct. What format is your dataset in?

Xiaodong


On Mar 8, 2011, at 2:36 PM, Travis S. Scharr wrote:

> I am having some problems using the DTI programs in slicer3.6.  When I run the tutorial, everything comes out fine, but when I try to use my own data I get images that are obviously wrong (see attached image).  Could the issue be with my data itself?  I am using the included DICOM to Nrrd converter for my data.  With some other programs I have used it was necessary to enter a gradient table specific to the magnet used.  Is this necessary or am I possibly having other issues?   I can tell when I run the tutorial with my data that things start to appear different very early, during the tensor estimation phase.  I am relatively new to this software and don't really understand enough to troubleshoot what is actually happening.  Can anybody help me?
>
>
> Travis Scharr
> University of Mississippi Medical Center
> Class of 2014
>
> Individuals who have received this information in error or are not authorized to receive it must promptly return or dispose of the information and notify the sender. Those individuals are hereby notified that they are strictly prohibited from reviewing, forwarding, printing, copying, distributing or using this information in any way.
>
> <SlicerImageBad_1.png>_______________________________________________
> slicer-users mailing list
> [hidden email]
> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users
> To unsubscribe: send email to [hidden email] with unsubscribe as the subject

_______________________________________________
slicer-users mailing list
[hidden email]
http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users
To unsubscribe: send email to [hidden email] with unsubscribe as the subject
Reply | Threaded
Open this post in threaded view
|

Re: DTI fiber tracking

inorton
In reply to this post by Travis S. Scharr
Hi Travis,
 
With some other programs I have used it was necessary to enter a gradient table specific to the magnet used.  Is this necessary or am I possibly having other issues?

Manually entering the gradient table should not be necessary IF the gradients are included in the DICOM files.. This gets tricky because there is no standard DICOM field, but DicomToNrrd should handle most GE, Siemens, and Phillips data...

Is your data anonymized? some anonymizers scrub all extra fields, in which case the gradients are no longer available for DicomToNrrd.

A quick check is to look at the file created by DicomToNrrd - the DW image. Does it list gradients in this form (ie):

DWMRI_gradient_0002:= -0.8238094 -0.4178235 -0.3830949

These lines should match the gradients you have manually entered for other programs. That's the place to start...

-Isaiah

 
On Tue, Mar 8, 2011 at 2:36 PM, Travis S. Scharr <[hidden email]> wrote:
I am having some problems using the DTI programs in slicer3.6.  When I run the tutorial, everything comes out fine, but when I try to use my own data I get images that are obviously wrong (see attached image).  Could the issue be with my data itself?  I am using the included DICOM to Nrrd converter for my data.     I can tell when I run the tutorial with my data that things start to appear different very early, during the tensor estimation phase.  I am relatively new to this software and don't really understand enough to troubleshoot what is actually happening.  Can anybody help me?


Travis Scharr
University of Mississippi Medical Center
Class of 2014

Individuals who have received this information in error or are not authorized to receive it must promptly return or dispose of the information and notify the sender. Those individuals are hereby notified that they are strictly prohibited from reviewing, forwarding, printing, copying, distributing or using this information in any way.


_______________________________________________
slicer-users mailing list
[hidden email]
http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users
To unsubscribe: send email to [hidden email] with unsubscribe as the subject


_______________________________________________
slicer-users mailing list
[hidden email]
http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users
To unsubscribe: send email to [hidden email] with unsubscribe as the subject
Reply | Threaded
Open this post in threaded view
|

Re: DTI fiber tracking

Xiaodong Tao-2
In reply to this post by Travis S. Scharr
Hello Travis,

You mentioned that you used DICOM to nrrd converter, what was the header generated by the converter? And what was the gradient table you entered?

Thanks.
Xiaodong

On Mar 8, 2011, at 2:36 PM, Travis S. Scharr wrote:

> I am having some problems using the DTI programs in slicer3.6.  When I run the tutorial, everything comes out fine, but when I try to use my own data I get images that are obviously wrong (see attached image).  Could the issue be with my data itself?  I am using the included DICOM to Nrrd converter for my data.  With some other programs I have used it was necessary to enter a gradient table specific to the magnet used.  Is this necessary or am I possibly having other issues?   I can tell when I run the tutorial with my data that things start to appear different very early, during the tensor estimation phase.  I am relatively new to this software and don't really understand enough to troubleshoot what is actually happening.  Can anybody help me?
>
>
> Travis Scharr
> University of Mississippi Medical Center
> Class of 2014
>
> Individuals who have received this information in error or are not authorized to receive it must promptly return or dispose of the information and notify the sender. Those individuals are hereby notified that they are strictly prohibited from reviewing, forwarding, printing, copying, distributing or using this information in any way.
>
> <SlicerImageBad_1.png>_______________________________________________
> slicer-users mailing list
> [hidden email]
> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users
> To unsubscribe: send email to [hidden email] with unsubscribe as the subject

_______________________________________________
slicer-users mailing list
[hidden email]
http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users
To unsubscribe: send email to [hidden email] with unsubscribe as the subject
Reply | Threaded
Open this post in threaded view
|

Re: DTI fiber tracking

inorton
In reply to this post by inorton
Hi Travis,

Uniform directions like that will make for some very oddly fitted tensors - as you saw. You could try entering the proper gradient directions into the converted DWI file. then re-open and re-estimate in Slicer. This may be easier if you specify the  dicomTonrrd output as a .nhdr file rather than .nrrd. You will end up with a .nhdr and .raw.gz - only edit the .nhdr. once you have a working header you should be able to use that repeatedly unless the sequence changes.

Also, this software may be useful if you are familiar with the command line: http://sourceforge.net/projects/gdcm/
use gdcmdump to check field (0019,100e) in your files.  DicomToNrrd reads that tag for Siemens gradient info. If it is missing due to scrubbing then DicomToNrrd won't work.

Best,
-Isaiah


On Fri, Mar 18, 2011 at 8:34 PM, Travis S. Scharr <[hidden email]> wrote:
The lines do not match.  The following is the table from the converted file:

DWMRI_b-value:=1000
DWMRI_gradient_0000:=0   0   0
DWMRI_gradient_0001:=0   -0.8945   0.4472
DWMRI_gradient_0002:=0   -0.8945   0.4472
DWMRI_gradient_0003:=0   -0.8945   0.4472
DWMRI_gradient_0004:=0   -0.8945   0.4472
DWMRI_gradient_0005:=0   -0.8945   0.4472
DWMRI_gradient_0006:=0   -0.8945   0.4472
DWMRI_gradient_0007:=0   -0.8945   0.4472
DWMRI_gradient_0008:=0   -0.8945   0.4472
DWMRI_gradient_0009:=0   -0.8945   0.4472
DWMRI_gradient_0010:=0   -0.8945   0.4472
DWMRI_gradient_0011:=0   -0.8945   0.4472
DWMRI_gradient_0012:=0   -0.8945   0.4472


And this is the gradient table I have been using:

Another Siemens 12 dir table directly from Siemens (Email dated: 13July2010):USE THIS FOR ESPREE and any s/w level VB13 and after
 0: 0.000, 0.000, 0.000
 1: 1.000, 0.414, -0.414
 2: 1.000, -0.414, -0.414
 3: 1.000, -0.414, 0.414
 4: 1.000, 0.414, 0.414
 5: 0.414, 0.414, 1.000here are some things you can try:
 6: 0.414, 1.000, 0.414
 7: 0.414, 1.000, -0.414
 8: 0.414, 0.414, -1.000
 9: 0.414, -0.414, -1.000
 10: 0.414, -1.000, -0.414
 11: 0.414, -1.000, 0.414
 12: 0.414, -0.414, 1.000


Does this suggest the scrubbing issue you spoke of?

Again, thank you very much for your assistance.

Travis Scharr
Class of 2014
________________________________________
From: Isaiah Norton [[hidden email]]
Sent: Tuesday, March 08, 2011 3:03 PM
To: Travis S. Scharr
Cc: [hidden email]
Subject: Re: [slicer-users] DTI fiber tracking

Hi Travis,

With some other programs I have used it was necessary to enter a gradient table specific to the magnet used.  Is this necessary or am I possibly having other issues?

Manually entering the gradient table should not be necessary IF the gradients are included in the DICOM files.. This gets tricky because there is no standard DICOM field, but DicomToNrrd should handle most GE, Siemens, and Phillips data...

Is your data anonymized? some anonymizers scrub all extra fields, in which case the gradients are no longer available for DicomToNrrd.

A quick check is to look at the file created by DicomToNrrd - the DW image. Does it list gradients in this form (ie):

DWMRI_gradient_0002:= -0.8238094 -0.4178235 -0.3830949

These lines should match the gradients you have manually entered for other programs. That's the place to start...

-Isaiah


On Tue, Mar 8, 2011 at 2:36 PM, Travis S. Scharr <[hidden email]<mailto:[hidden email]>> wrote:
I am having some problems using the DTI programs in slicer3.6.  When I run the tutorial, everything comes out fine, but when I try to use my own data I get images that are obviously wrong (see attached image).  Could the issue be with my data itself?  I am using the included DICOM to Nrrd converter for my data.     I can tell when I run the tutorial with my data that things start to appear different very early, during the tensor estimation phase.  I am relatively new to this software and don't really understand enough to troubleshoot what is actually happening.  Can anybody help me?


Travis Scharr
University of Mississippi Medical Center
Class of 2014

Individuals who have received this information in error or are not authorized to receive it must promptly return or dispose of the information and notify the sender. Those individuals are hereby notified that they are strictly prohibited from reviewing, forwarding, printing, copying, distributing or using this information in any way.


_______________________________________________
slicer-users mailing list
[hidden email]<mailto:[hidden email]>
To unsubscribe: send email to [hidden email]<mailto:[hidden email]> with unsubscribe as the subject


Individuals who have received this information in error or are not authorized to receive it must promptly return or dispose of the information and notify the sender. Those individuals are hereby notified that they are strictly prohibited from reviewing, forwarding, printing, copying, distributing or using this information in any way.




_______________________________________________
slicer-users mailing list
[hidden email]
http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users
To unsubscribe: send email to [hidden email] with unsubscribe as the subject