voxel spacing

831 views
Skip to first unread message

松木 圭介

unread,
Sep 29, 2009, 3:43:41 PM9/29/09
to itksna...@googlegroups.com
Hi,

I am creating 3D bone models from CT images (DICOM) using
ITK-snap. They have 0.75mm of voxel depth, and their
reconstructed sagittal and coronal images look stretched
with other DICOM viewers. When I load the images into
ITK-snap, the voxel depth changes into 0.5mm
automatically, and their sagittal and coronal images look
good in proportion. Please anyone tell me why the voxel
depth changes and which value is correct.

Thank you.

Keisuke


--------------------------------------
Yahoo! JAPAN - Internet Security for teenagers and parents.
http://pr.mail.yahoo.co.jp/security/

Tom Schoenemann

unread,
Sep 29, 2009, 5:27:09 PM9/29/09
to itksna...@googlegroups.com
Keisuke,

I'm not sure about your specific images, but there is a distinction between slice thickness and distance between slices (actually: distance between the center plane of each slice).  One can have the scanner produce overlapping slices (e.g., slices are .75 mm thick, but spaced only .5 mm apart).  This is usually done (when it is done) because thicker slices are generally clearer (because they average more tissue, and therefore signal-to-noise ratio increases).  Is it possible that your "voxel depth" is actually slice thickness in other viewers, but distance-between-slices in ITK-snap?  

If this is a correct interpretation of what is going on, it means that .75 is the thickness of the slices, but they are .5 mm apart. ITK-snap is correctly spacing the slices, whereas the other viewers are not.

-Tom
_________________________________________________
P. Thomas Schoenemann

Associate Professor
Department of Anthropology
Indiana University
Bloomington, IN  47405

Open Research Scan Archive at Penn Co-Director
Consulting Scholar
Museum of Archaeology and Anthropology
University of Pennsylvania
Philadelphia, PA 19104-6398


松木 圭介

unread,
Sep 30, 2009, 10:21:54 AM9/30/09
to itksna...@googlegroups.com
Tom,

Thank you very much.
It makes sense to me.

Keisuke

uditha

unread,
Nov 5, 2009, 8:48:57 AM11/5/09
to itksnap-users
sir,
I am working on an automatic liver segmentation project with ITK-Snap.
I encountered that slice spacing in z-direction is not the same when
the image is viewed in ITK-Snap and ImageJ. Due to this I doubt
whether the volumes generated by ITK-Snap is accurate. Can you explain
why this happens....?

Thanks you.

Uditha Lakmal,

Undergraduate,
Dept. of Electronic and Telecommunications Eng.,
Faculty of Engineering,
University of Moratuwa,
Sri Lanka.
Reply all
Reply to author
Forward
0 new messages