ELSCINT Twin Flash CT - Frame Of Reference

Bartosz Wiklak

unread,

Dec 3, 2009, 9:46:45 AM12/3/09

to

Hi,

I have o problem with ELSCINT Twin Flash CT scans. Images from
different series but the same study, that should be in the same frame
of reference have different Frame Of Reference UID, always in such way
that
images from series 1 Frame Of Reference UID is in form
xxxx.xxx.xx.xx.n.xxx.xxx
and from series 2 have form xxxx.xxx.xx.xx.n+1.xxx.xxx all "x" digits
are the same.

Satellite station "omni pro" shows reference lines but my program,
because of different F.O.R.UID, does not.

Did someone see such situation before?
How can I cope with this?

Bartek

Bartosz Wiklak

unread,

Dec 4, 2009, 3:20:33 AM12/4/09

to

The problem lays in exporting application I suppose - it stores Series
Instance UID in Frame Of Reference UID, Unfortunately changing Frame
Of Reference UID does not help because ImagePosition and Orientation
are not good. The localizer lines does not show on good position and
axis signs are not ok.
The whole mystery seems to hide in (00E3,1000) tag - I even managed to
fine image number among binary data hided in this tag.
Maybe someone knows what there is?
Just for info:
(01F1,0010) "ELSCINT1"
(01F1,1001) "SPIRAL", "HELIX"
(01F1,1002) "STANDARD"
(07A1,1010) "2.7.1"
(07A1,1011) "LOSSLESS RICE" - but the image is converted and has good
PixelData tag

Tag "Image Comments" "$Revision: 1.80 $ of DICOM conversion"

Bartek

Mathieu Malaterre

unread,

Dec 4, 2009, 6:24:12 AM12/4/09

to

Did you try gdcmdump on it ? There are a bunch of private tags
available in gdcmdump.

Ref:
http://sourceforge.net/projects/gdcm/files/gdcm%202.x/GDCM%202.0.14/

HTH

Bartosz Wiklak

unread,

Dec 4, 2009, 7:43:22 AM12/4/09

to

On 4 Gru, 12:24, Mathieu Malaterre <mathieu.malate...@gmail.com>
wrote:

Hi Mathiew,
Yes, I tried, unfortunately only information I quired from gdcm
dictionary is "Tamar Software Version", the other tags did not helped,
but there is hope:

I found that scouts with image type ORIGINAL\PRIMARY\LOCALIZERS\ES
have funny sequence in a private tag (01F3,xx01).
This sequence contains as many items as CT scan has and single item
has form
(01F3,xx02) VR=SS VM=1 - contains reference image number
(01F3,xx03) VR=FL VM=2 - two floats, second always 256 as far as I was
able to check
(01F3,xx02) VR=SS VM=1 - float always 0 in images I managed to create

I made small change to my app and shown a reference line on scout
using (01F3,xx03) first float as a position of line on image in
pixels. This seems to work! I have to make some more research and
images to make myself sure.
I don't know what does two other floats mean.

If I figure out something more, I'll make some notes here.

thanks,
Bartek

Bartosz Wiklak

unread,

Dec 4, 2009, 7:48:51 AM12/4/09

to

correction:
(..) and single item

has form
(01F3,xx02) VR=SS VM=1 - contains reference image number
(01F3,xx03) VR=FL VM=2 - two floats, second always 256

(01F3,xx04) VR=SS VM=1 - float always 0 in images I managed to create

Bartosz Wiklak

unread,

Dec 4, 2009, 8:12:19 AM12/4/09

to

sorry, next typo:
(01F3,xx04) VR=FL VM=1 - float always 0 in images I managed to create

Bartosz Wiklak

unread,

Dec 8, 2009, 4:54:09 AM12/8/09

to

Ok, I recognized tags in the sequence (01F3,xx01) :

(01F3,xx02) VR=SS VM=1 - reference image number
(01F3,xx03) VR=FL VM=2 - Center of localizer line on SCOUT in pixels
(01F3,xx04) VR=FL VM=1 - Tilt of localizer line in radians (sic!)
~=gantry tilt in degrees.