[Insight-users] discontinuous (cutoff) deformation field from bspline deformable algorithm

[Minjie Wu]

Hello,

We have a problem with discontinuous or cutoff deformation field when we
were trying to register
a T2-weighted ( nonlinearly distorted ) brain image to a T1-weighted (no
distortion) brain image
of the same subject using bspline deformable registration. After the
registration, we got some wierd
results, which frontal brain was somehow cutoff, which led to a
discontinuous warped brain image.
And we found the deformation field was also cutoff.

We also tested the bspline algorithm using different template, such as
T2-weighted template, or proton
density template. The deformation field show the same discontinuity or
cutoff for all the tests.

>From what we have read about the bspline algorithm, it should produce a
smooth and continuous deformation
field, and we are confused with the obtained discontiuous/cutoff deformation
field.

We appreciate if anyone can help us on this. Any suggestion is welcome.

Thanks,
-Minjie
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://public.kitware.com/pipermail/insight-users/attachments/20070201/41fb470f/attachment.htm