How is ITK-Snap (correctly) separating multiple "views" from the same DICOM series?

32 views
Skip to first unread message

Alexander Weston

unread,
Mar 28, 2024, 5:06:18 PMMar 28
to itksnap-users

Hello,

Firstly, huge fan of your work, we have been using it for years and it's a lifesaver!

We have come across an example of a 3D CT Study (saved as a DICOM filestack) that appears to have multiple non-contiguous "views" saved under the same SeriesUID. See in the screenshot attached, ITK-Snap loads a total of 10 "images", with two copies each of Series 3, and Series 103. I have verified that SeriesUID is the same for the duplicated series.

The study is a GE study from 2001 although I have observed similar behavior from other DICOMs.

This is not a “bug”, in fact I think ITK-Snap has the preferred behavior. Not to pick on 3D Slicer, but if you compare Slicer to ITK, (second screenshots), 3D Slicer loads Series 3 as one continuous acquisition but shows two interleaved images (BAD). All other viewers do the same thing, ITK is the only viewer we’ve found that loads Series 3 “correctly”.

Examining the metadata for Series3, there is not much different between the dicom instances, a printout of the limited metadata which differs is below, I wouldn't think any of these could serve to uniquely separate two "views”, since CT can have differing slice spacings (e.g. head CTs which preserve dose to the brainstem).

My question is, what is ITK-Snap “detecting” to correctly separate these two views?

Thank you so much for your help!

 

Alex

 

Pic 1: ITK-Snap (correctly) loads Series 3 as two separate “views” (512x512x155 and 512x512x81)

Pic 2: 3D Slicer loads “series 3” as a single 512x512x236 image (with errors) and then interleaves the two “views” (not what we want)


Some DICOM metadata that is “different” between the two “views” for Series3 (disregarding the ones that I might expect to be different like instance creation time):

0018,0050 Slice Thickness    5.0, 1.25

0018,1100 Reconstruction Diameter   320, 250

0018,1150 Exposure Time  1675, 912

0018,1151 XRayTubeCurrent  190, 340

0018,1152 Exposure  8360, 7140

0018,1170 Generator Power  22800, 47600

0018,1190 Focal Spots  .70, 1.2

0028,0030 Pixel Spacing .625\.625, .488\.488

0028,1050 Window Center  40, 151

0028.1051  Window Width  400,100

Pic1.PNG
Pic2.PNG
Reply all
Reply to author
Forward
0 new messages