[ITK] Registration of multislice images

[Andras Lasso]

Hi Mikael,

You can copy the voxels from each slice into a volume and use that as fixed image, use a mask that has non-zero values only at valid pixels (usually you can use the constructed image itself as a mask), then do a usual volumetric registration. This works quite well, if you have orthogonal images and their resolution is similar (see for example http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=5975205). You can use ITK as is, no modifications are needed.

Alternatively, you can modify a similarity metric to accept multiple images and compute the metric value as sum of the metric value of each image (this is necessary when the image slices are not orthogonal). An IJCARS paper is in press, a high-level overview is given in this abstract: http://perk.cs.queensu.ca/contents/target-localization-mriguided-transperineal-prostate-biopsy-using-multislicetovolume-regist .

Andras


From: Eriksson, Mikael<mailto:mikael.k.eriksson at philips.com>
Sent: ‎Monday‎, ‎September‎ ‎1‎, ‎2014 ‎1‎:‎26‎ ‎AM
To: community at itk.org<mailto:community at itk.org>

Hello,

I have obtained multislice image data for trying to do real-time registration for motion correction purposes. Specifically, every 0.x seconds, one coronal, one sagittal and one axial slice is obtained during the imaging, and I would like to register pairs of such sets with each other, e.g. each set after the first set at time zero should be registered to this time zero set. The slices cover the entire body in the axial plane, and the upper body from the hips to the neck in the sagittal and coronal planes with a size of 336 x 336 and pixel spacing of 1.19 x 1.19.  I have searched for possible implementations in ITK and haven’t found anything. An interesting article on the subject which describes what I want to do can be found here: http://www.researchgate.net/publication/221623301_Intersection_based_registration_of_slice_stacks_to_form_3D_images_of_the_human_fetal_brain

Essentially, what they did was to optimize the (rigid) transform the slices in the moving set until their intensity profiles along the intersections with orthogonal slices match the intensity profiles at the corresponding intersections in the fixed set, within a certain tolerance. Now, this is one way that I would be more than happy to try out on my data, but likewise, if there are other methods already implemented in ITK I’ll try them out as well. If there isn’t anything already implemented, could someone please give me some hints as for how to write a program which does something along the lines of what I described above (and is described in more detail in the link)? Oh, and just knowing whether anything like this already exists and can be used in ITK will be a huge help, since it would allow me to plan my work.

Best regards,
Mikael



………………………………………………………………………………………….
Mikael Eriksson
R&D Physicist Trainee
Philips Medical Systems MR Finland, Feasibility studies
mikael.k.eriksson at philips.com<mailto:mikael.k.eriksson at philips.com>
………………………………………………………………………………………….


________________________________
The information contained in this message may be confidential and legally protected under applicable law. The message is intended solely for the addressee(s). If you are not the intended recipient, you are hereby notified that any use, forwarding, dissemination, or reproduction of this message is strictly prohibited and may be unlawful. If you are not the intended recipient, please contact the sender by return e-mail and destroy all copies of the original message.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/community/attachments/20140901/8b3815a0/attachment-0002.html>