DicomToNrrdConverter Problem

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

DicomToNrrdConverter Problem

Aditya Gupta
Hello all,

I am trying to use the command line DicomToNrrdConverter for 42 directional DTI data and I am getting the errors attached in the error.txt and error1.txt files. I am sure it is not an issue with the data as I had run the same command on the data few months back and it generated a Nrrd file. I have seen error1.txt before but even with this error the process completes and generates a nrrd file. The error.txt is the new error and this is the issue.

Thank you.

--
Regards,
AGupta

_______________________________________________
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

error1.txt (39K) Download Attachment
error.txt (1K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: DicomToNrrdConverter Problem

Xiaodong Tao-2
Hello Aditya,

For Siemens datasets, you will need to turn on the useBMatrixGradientDirections to have it work correctly. Right now, this option is in advanced. I will make this default for Siemens dataset and check in the code to both Slicer trunk and 3.6 branch today.

Thanks.
Xiaodong

On Jan 27, 2011, at 11:30 AM, Aditya Gupta wrote:

> Hello all,
>
> I am trying to use the command line DicomToNrrdConverter for 42 directional DTI data and I am getting the errors attached in the error.txt and error1.txt files. I am sure it is not an issue with the data as I had run the same command on the data few months back and it generated a Nrrd file. I have seen error1.txt before but even with this error the process completes and generates a nrrd file. The error.txt is the new error and this is the issue.
>
> Thank you.
>
> --
> Regards,
> AGupta
> <error1.txt><error.txt>_______________________________________________
> 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: DicomToNrrdConverter Problem

Magnotta, Vincent A
In reply to this post by Aditya Gupta
Agupta,

Look at the bottom of the error.txt file. It tells you how to fix the problem (useBMatrixGradientDirections). The real problem is that Siemens on occasion places incorrect diffusion directions into the DICOM header.

Vince

-- 
Associate Professor
Department of Radiology
0453-D JCP
200 Hawkins Drive
Iowa City, IA 52242
E-mail: [hidden email]
Phone: 319-356-8255 Fax: 319-353-6275
Website: http://www.healthcare.uiowa.edu/Radiology

From: Aditya Gupta <[hidden email]>
Date: Thu, 27 Jan 2011 11:30:07 -0500
To: <[hidden email]>
Subject: [slicer-users] DicomToNrrdConverter Problem

Hello all,

I am trying to use the command line DicomToNrrdConverter for 42 directional DTI data and I am getting the errors attached in the error.txt and error1.txt files. I am sure it is not an issue with the data as I had run the same command on the data few months back and it generated a Nrrd file. I have seen error1.txt before but even with this error the process completes and generates a nrrd file. The error.txt is the new error and this is the issue.

Thank you.

--
Regards,
AGupta
_______________________________________________ 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



Notice: This UI Health Care e-mail (including attachments) is covered by the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is confidential and may be legally privileged.  If you are not the intended recipient, you are hereby notified that any retention, dissemination, distribution, or copying of this communication is strictly prohibited.  Please reply to the sender that you have received the message in error, then delete it.  Thank you.

_______________________________________________
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: DicomToNrrdConverter Problem

Aditya Gupta
In reply to this post by Xiaodong Tao-2
Hello,

Thanks for your quick reply. I did try this option (useBMatrixGradientDirections) before sending in the error to Slicers mailing list. Attached is the error report with the option turned on. It gives a "segmentation fault" error.

As I analyze it, in the previous version B0 had a direction of 0 0 0. But in the new version (if you see the previous error files), B0 is given a direction of 1 0 0:
  
   Directions 0: 1
   Directions 1: 0
   Directions 2: 0
Image#: 0 BV: 1000 GD: 1 0 0

So I think the program is seeing the B0 image as a gradient image and trying to normalize the next image (the gradient #1) with this B0 image and hence gives the DiffusionVector_magnitude_difference as infinite. Therefore the error with gradient scaling factor of 0:

DiffusionVector_magnitude_difference inf
gradient_scaling_factor 1.21782e-315
DiffusionVector_magnitude 1.00006
ERROR: Gradient vector with unreasonably small magnitude exists.

The previous version of DicomToNrrdConverter works with this dataset and hence this is not a dataset issue.

Thanks you.

Regards,
AGupta

On Thu, Jan 27, 2011 at 11:46 AM, Xiaodong Tao <[hidden email]> wrote:
Hello Aditya,

For Siemens datasets, you will need to turn on the useBMatrixGradientDirections to have it work correctly. Right now, this option is in advanced. I will make this default for Siemens dataset and check in the code to both Slicer trunk and 3.6 branch today.

Thanks.
Xiaodong

On Jan 27, 2011, at 11:30 AM, Aditya Gupta wrote:

> Hello all,
>
> I am trying to use the command line DicomToNrrdConverter for 42 directional DTI data and I am getting the errors attached in the error.txt and error1.txt files. I am sure it is not an issue with the data as I had run the same command on the data few months back and it generated a Nrrd file. I have seen error1.txt before but even with this error the process completes and generates a nrrd file. The error.txt is the new error and this is the issue.
>
> Thank you.
>
> --
> Regards,
> AGupta
> <error1.txt><error.txt>_______________________________________________
> 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




--
Regards,
Aditya Gupta, Ph.D.
UNC School of Medicine
Chapel Hill, NC
M: (407) 451 7735

_______________________________________________
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

error_withbMatrix_flag.txt (1K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: DicomToNrrdConverter Problem

inorton
In case it helps, maybe a related bug - if this data is in a single file. Segfault when only one input file is given. Two distinct or two duplicate files worked fine. backtrace in second note here:

http://www.na-mic.org/Bug/view.php?id=1068
 

On Thu, Jan 27, 2011 at 3:57 PM, Aditya Gupta <[hidden email]> wrote:
Hello,

Thanks for your quick reply. I did try this option (useBMatrixGradientDirections) before sending in the error to Slicers mailing list. Attached is the error report with the option turned on. It gives a "segmentation fault" error.

As I analyze it, in the previous version B0 had a direction of 0 0 0. But in the new version (if you see the previous error files), B0 is given a direction of 1 0 0:
  
   Directions 0: 1
   Directions 1: 0
   Directions 2: 0
Image#: 0 BV: 1000 GD: 1 0 0

So I think the program is seeing the B0 image as a gradient image and trying to normalize the next image (the gradient #1) with this B0 image and hence gives the DiffusionVector_magnitude_difference as infinite. Therefore the error with gradient scaling factor of 0:

DiffusionVector_magnitude_difference inf
gradient_scaling_factor 1.21782e-315
DiffusionVector_magnitude 1.00006
ERROR: Gradient vector with unreasonably small magnitude exists.

The previous version of DicomToNrrdConverter works with this dataset and hence this is not a dataset issue.

Thanks you.

Regards,
AGupta


On Thu, Jan 27, 2011 at 11:46 AM, Xiaodong Tao <[hidden email]> wrote:
Hello Aditya,

For Siemens datasets, you will need to turn on the useBMatrixGradientDirections to have it work correctly. Right now, this option is in advanced. I will make this default for Siemens dataset and check in the code to both Slicer trunk and 3.6 branch today.

Thanks.
Xiaodong

On Jan 27, 2011, at 11:30 AM, Aditya Gupta wrote:

> Hello all,
>
> I am trying to use the command line DicomToNrrdConverter for 42 directional DTI data and I am getting the errors attached in the error.txt and error1.txt files. I am sure it is not an issue with the data as I had run the same command on the data few months back and it generated a Nrrd file. I have seen error1.txt before but even with this error the process completes and generates a nrrd file. The error.txt is the new error and this is the issue.
>
> Thank you.
>
> --
> Regards,
> AGupta
> <error1.txt><error.txt>_______________________________________________
> 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




--
Regards,
Aditya Gupta, Ph.D.
UNC School of Medicine
Chapel Hill, NC
M: (407) 451 7735

_______________________________________________
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: DicomToNrrdConverter Problem

Magnotta, Vincent A
In reply to this post by Aditya Gupta
AGupta,

Would you be willing to share your data? Mark Scully might be able to take a look at this. Do you know what version of Slicer that you are using? I believe that several changes were made to the code at the All Hands Meeting that may have fixed this bug.

Vince


--
Associate Professor
Department of Radiology
0453-D JCP
200 Hawkins Drive
Iowa City, IA 52242
E-mail: [hidden email]
Phone: 319-356-8255 Fax: 319-353-6275
Website: http://www.healthcare.uiowa.edu/Radiology


From: Aditya Gupta [[hidden email]]
Sent: Thursday, January 27, 2011 2:57 PM
To: Xiaodong Tao; [hidden email]; Magnotta, Vincent A
Subject: Re: [slicer-users] DicomToNrrdConverter Problem

Hello,

Thanks for your quick reply. I did try this option (useBMatrixGradientDirections) before sending in the error to Slicers mailing list. Attached is the error report with the option turned on. It gives a "segmentation fault" error.

As I analyze it, in the previous version B0 had a direction of 0 0 0. But in the new version (if you see the previous error files), B0 is given a direction of 1 0 0:
  
   Directions 0: 1
   Directions 1: 0
   Directions 2: 0
Image#: 0 BV: 1000 GD: 1 0 0

So I think the program is seeing the B0 image as a gradient image and trying to normalize the next image (the gradient #1) with this B0 image and hence gives the DiffusionVector_magnitude_difference as infinite. Therefore the error with gradient scaling factor of 0:

DiffusionVector_magnitude_difference inf
gradient_scaling_factor 1.21782e-315
DiffusionVector_magnitude 1.00006
ERROR: Gradient vector with unreasonably small magnitude exists.

The previous version of DicomToNrrdConverter works with this dataset and hence this is not a dataset issue.

Thanks you.

Regards,
AGupta

On Thu, Jan 27, 2011 at 11:46 AM, Xiaodong Tao <[hidden email]> wrote:
Hello Aditya,

For Siemens datasets, you will need to turn on the useBMatrixGradientDirections to have it work correctly. Right now, this option is in advanced. I will make this default for Siemens dataset and check in the code to both Slicer trunk and 3.6 branch today.

Thanks.
Xiaodong

On Jan 27, 2011, at 11:30 AM, Aditya Gupta wrote:

> Hello all,
>
> I am trying to use the command line DicomToNrrdConverter for 42 directional DTI data and I am getting the errors attached in the error.txt and error1.txt files. I am sure it is not an issue with the data as I had run the same command on the data few months back and it generated a Nrrd file. I have seen error1.txt before but even with this error the process completes and generates a nrrd file. The error.txt is the new error and this is the issue.
>
> Thank you.
>
> --
> Regards,
> AGupta
> <error1.txt><error.txt>_______________________________________________
> 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




--
Regards,
Aditya Gupta, Ph.D.
UNC School of Medicine
Chapel Hill, NC
M: (407) 451 7735



Notice: This UI Health Care e-mail (including attachments) is covered by the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is confidential and may be legally privileged.  If you are not the intended recipient, you are hereby notified that any retention, dissemination, distribution, or copying of this communication is strictly prohibited.  Please reply to the sender that you have received the message in error, then delete it.  Thank you.

_______________________________________________
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