Main Page | Report this Page
 
   
Science Forum Index  »  Image Processing Forum  »  Problem with DICOM Format !
Page 1 of 1    
Author Message
xrayspectrum@googlemail.c
Posted: Mon Dec 11, 2006 7:19 am
Guest
Dear all,

I am a student working with reconstruction of images .. recently we
have got images from a cone beam CT and reconstrcuted them in DICOM
Format.. I donno what went wrong .We could not open the reconstructed
images in treatment planning system like OTP etc. .. we are trying to
figure out the best possiblities and well i believe there must be
something wrong with the tagging ( guess ) .. like for other images
there were many tages .. but for our images there are only few.. it may

look bit silly comparing them directly. I am just working on it ..
trying to add tags to the our image. Hope it would work ..
Please someone help to get some information about DICOM Tags or is
there any other way solving this problem. I am really looking forward
for any precious suggestions and advise ..


thank you so much in advance


regards
Dan
Bill
Posted: Tue Dec 12, 2006 3:42 pm
Guest
xrayspectrum@googlemail.com wrote:
Quote:
Dear all,

I am a student working with reconstruction of images .. recently we
have got images from a cone beam CT and reconstrcuted them in DICOM
Format.. I donno what went wrong .We could not open the reconstructed
images in treatment planning system like OTP etc. .. we are trying to
figure out the best possiblities and well i believe there must be
something wrong with the tagging ( guess ) .. like for other images
there were many tages .. but for our images there are only few.. it may

look bit silly comparing them directly. I am just working on it ..
trying to add tags to the our image. Hope it would work ..
Please someone help to get some information about DICOM Tags or is
there any other way solving this problem. I am really looking forward
for any precious suggestions and advise ..


thank you so much in advance


regards
Dan

Is the problem with a specific image set or all images sent to the
Treatment Planning System (TPS)? How are you transferring the images
to the TPS? Can you transfer successfully? Does the TPS give any
error messages when attempting to open the images?

Are you only sending the reconstructed images to the Treatment Planning
System (TPS)? If sending other images that are not reconstructed, can
they be opened?

I need more answers to help point you into a direction. You are
correct. Reconstructed images do contain many less tags than other
types of images. This is because reconstructed images use a Secondary
Capture SOP Class. This SOP Class contains a minimum of tags to
identify the image.

Bill
xrayspectrum@googlemail.c
Posted: Wed Dec 13, 2006 5:29 am
Guest
Thanks Bill for your reply ,

Well, let me tell you in detail: The reconstructed images, when opened
through an image viewer, the orientation is different than Transverse
so that may be the big reason for treatment planning not able to
identify them. I donno how to change the it. But right now my task is
to transfer the images to TPS . I am under the opinion that if I
change the DICOM Format with an Editor, there is a possibility to open
them in TPS.

Could you please tell which tags are really important for TPS to open
an Reconstructed Image .



Furthur down you will find answer to the questions you have asked me :


1. Is the problem with a specific image set or all images sent to the
Treatment Planning System (TPS) ??

The problem is only with the specific Image data set ( self -
reconstructed Images Only ! )

2.How are you transferring the images to the TPS? Can you transfer
successfully? Does the TPS give any error messages when attempting to
open the images ??

The reconstructed Images were generated in DICOM Format so we are just
transfering them in a regular fashion like the data sets are in the CD
, so copying them in the treatment planning system and trying to open
them. BUT it says there is no image found.

3.Are you only sending the reconstructed images to the Treatment
Planning System (TPS)? If sending other images that are not
reconstructed, can they be opened?

Yeah the images other than the reconstructed ones can be opened easily.
so the problem is only with reconstructed Images . Well I am trying to
change the DICOM tags with an editor .. still there is problem , i mean
TPS does not open them.

Could you please shed some light on this aspects : " Reconstructed
images do contain many less tags than other types of images. This is
because reconstructed images use a Secondary Capture SOP Class. This
SOP Class contains a minimum of tags to identify the image " .

Please I need your precious suggestion.

thanks again !

regards
Dan
Bill
Posted: Wed Dec 13, 2006 12:19 pm
Guest
xrayspectrum@googlemail.com wrote:

Quote:
to transfer the images to TPS . I am under the opinion that if I
change the DICOM Format with an Editor, there is a possibility to open
them in TPS.

I am not sure this opinion is correct. Exactly what Editor are you
using?

Quote:

Could you please tell which tags are really important for TPS to open
an Reconstructed Image .

The SOP Class needs to be the Secondary Capture UID value. The image
must have a Study Instance UID, Series Instance UID, and a SOP Instance
UID. The other critical information is in Group 28. These fields
include Number of Planes, Photometric Interpretation, Rows, Columns,
Bits Allocated, Bits Stored, and High Bit. There are a few other
required fields, but this should at least get the image to display.

Quote:

1. Is the problem with a specific image set or all images sent to the
Treatment Planning System (TPS) ??

The problem is only with the specific Image data set ( self -
reconstructed Images Only ! )

Let me clarify. Do all of your reconstructed images have this problem?

Quote:

2.How are you transferring the images to the TPS? Can you transfer
successfully? Does the TPS give any error messages when attempting to
open the images ??

The reconstructed Images were generated in DICOM Format so we are just
transfering them in a regular fashion like the data sets are in the CD
, so copying them in the treatment planning system and trying to open
them. BUT it says there is no image found.

Actually, the regular fashion is to transfer the images over a network
association using the DICOM Protocol. However, the images can be
transferred via CD. Are there other DICOM images, not reconstructed,
on the CD as well? If so, can those DICOM images be displayed?

You have to be careful when copying DICOM images from one device to
another. Reason. Byte Order Transfer. Intel computers use Little
Endian byte order. Sun, Motorola, and PowerPC use Big Endian byte
order. Copying DICOM images from a Little Endian to a Big Endian, and
visa versa, will cause problems as you are describing here. Plus, the
DICOM image itself states the byte order with its Transfer Syntax.

Quote:

3.Are you only sending the reconstructed images to the Treatment
Planning System (TPS)? If sending other images that are not
reconstructed, can they be opened?

Yeah the images other than the reconstructed ones can be opened easily.
so the problem is only with reconstructed Images . Well I am trying to
change the DICOM tags with an editor .. still there is problem , i mean
TPS does not open them.


Have you tried to view the reconstructed images with some free viewer
software? If so, what were the results?

Bill
xrayspectrum@googlemail.c
Posted: Thu Dec 14, 2006 5:13 am
Guest
Dear Bill,

Thanks again for the reply and time !

Here the answer to the questions

Exactly what Editor are you using?
I am presently using two Dicom Editors: Power Dicom and DCM Editor.

Do all of your reconstructed images have this problem?
All my reconstructed Images has this problem.

Are there other DICOM images, not reconstructed,on the CD as well? If
so, can those DICOM images be displayed?
I don't have any other images on the CD, Exccept my reconstructed
images.

Have you tried to view the reconstructed images with some free viewer
software? If so, what were the results?
I tried to view with free-viewers for example KPAC, Irfan View etc. the
image is displayed. But there are many artifacts.

I wish to send you the image as an attachment , May I send you through
ur yahoo ID ?

thanks once again,

regards,
Dan


Bill wrote:
Quote:
xrayspectrum@googlemail.com wrote:

to transfer the images to TPS . I am under the opinion that if I
change the DICOM Format with an Editor, there is a possibility to open
them in TPS.

I am not sure this opinion is correct. Exactly what Editor are you
using?


Could you please tell which tags are really important for TPS to open
an Reconstructed Image .

The SOP Class needs to be the Secondary Capture UID value. The image
must have a Study Instance UID, Series Instance UID, and a SOP Instance
UID. The other critical information is in Group 28. These fields
include Number of Planes, Photometric Interpretation, Rows, Columns,
Bits Allocated, Bits Stored, and High Bit. There are a few other
required fields, but this should at least get the image to display.


1. Is the problem with a specific image set or all images sent to the
Treatment Planning System (TPS) ??

The problem is only with the specific Image data set ( self -
reconstructed Images Only ! )

Let me clarify. Do all of your reconstructed images have this problem?


2.How are you transferring the images to the TPS? Can you transfer
successfully? Does the TPS give any error messages when attempting to
open the images ??

The reconstructed Images were generated in DICOM Format so we are just
transfering them in a regular fashion like the data sets are in the CD
, so copying them in the treatment planning system and trying to open
them. BUT it says there is no image found.

Actually, the regular fashion is to transfer the images over a network
association using the DICOM Protocol. However, the images can be
transferred via CD. Are there other DICOM images, not reconstructed,
on the CD as well? If so, can those DICOM images be displayed?

You have to be careful when copying DICOM images from one device to
another. Reason. Byte Order Transfer. Intel computers use Little
Endian byte order. Sun, Motorola, and PowerPC use Big Endian byte
order. Copying DICOM images from a Little Endian to a Big Endian, and
visa versa, will cause problems as you are describing here. Plus, the
DICOM image itself states the byte order with its Transfer Syntax.


3.Are you only sending the reconstructed images to the Treatment
Planning System (TPS)? If sending other images that are not
reconstructed, can they be opened?

Yeah the images other than the reconstructed ones can be opened easily.
so the problem is only with reconstructed Images . Well I am trying to
change the DICOM tags with an editor .. still there is problem , i mean
TPS does not open them.


Have you tried to view the reconstructed images with some free viewer
software? If so, what were the results?

Bill
Bill
Posted: Thu Dec 14, 2006 11:12 am
Guest
Quote:
I wish to send you the image as an attachment , May I send you through
ur yahoo ID ?

thanks once again,

regards,
Dan


Yes, you may use my Yahoo ID. If possible could you send me the
original DICOM file, the pixel data file, and the combined
reconstructed DICOM file?

Bill
 
Page 1 of 1       All times are GMT - 5 Hours
The time now is Sat Oct 11, 2008 2:54 am