[Dcmlib] dicom flip

[Luca Picello]

Hi , i have a big trouble and loosing a lot of sleeping time in attempts 
to resolve it :

 

I have a dicom serie of a head :

 

Patient Position is :  L\P

Patient Position is : HFS

Image Orientation Patient is   1,0,0   0,1,0

 

IMAGE POSITION PATIENT for FIRST      SLICE IS :     0 , 0 , 0         

                                          for SECOND SLICE IS :     0, 
 0 , 1

                                          for THIRD SLICE IS     : 
     0,  0 , 2

                                          for 4' SLICE IS            
:      0,  0 , 3

                                          and so on according with slice 
spacing.( 1 mm)

 

1-

   The z axis is increasing toward the head of the patient so *I assume 
the orientation cosines for z should be : 0,0,1*

   *Am I right ?*

   *Considering Image orientation patient (1,0,0  0,1,0);  the axes 
direction cosines became: 1,0,0 ,0,1,0, 0,0,1*

   *Am I right?*

 

2-

    in a different serie *IF* I obtain a z value decreasing  toward the 
head I should assume the orientation cosines for z ( 0,0,-1) ?

    Am I right?

 

3- 

Regarding case 1: even if all seems right with axes orientation cosines 
I discover a strange behaviour after loading the dicom serie ( I 
verified with the real patient left and right sides) :

*The left part of the patient became the right part. Why?*

 

I must flip the dataset along an axe ( flipX for example )  to obstain 
the correct L  and R side according to the real patient

 

I'm using gdcm....

 

Thanks a lot for suggestions

 


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.creatis.insa-lyon.fr/pipermail/dcmlib/attachments/20060415/671d9505/attachment.html>