Kris & Brad,<br><br>Please see this related bug report for ITK <a href="http://public.kitware.com/Bug/view.php?id=8574">http://public.kitware.com/Bug/view.php?id=8574</a> . My experience when working with diffusion tensors was that the rotation A = R*T*R' was the main use case, and implementing a rotation method directly rather than both pre-/post- multiply methods was easier to use.<br>
<br>Thank you<br clear="all"><br>-- <br>Casey Goodlett<br>