Allow to specify fixed dimensions for affine/rigid transformations

[Markus92]

Hi all, I'm currently working on intra-patient registration of DWI to T2 images. It is known and I can clearly see in my data that the main geometric distortion occurs in the phase-encoding direction (A-P or y-axis for my images). The movement in the other directions is insignificant and can be disregarded. It'd be great if I could specify this to elastix, such that the other axis can remain fixed. Currently it does seem to be registering stuff to each other that should not be registered and this leads to worse results. Right now a simple translation transform seems to offer best results, but it'd be even better if I could do both translation and scaling in one axis. Am I correct in seeing this is not currently supported by elastix? And what would be the easiest way to implement this? Thanks in advance, Mark

[mstaring]

You can do this via the optimizer setting "Scales":

Scales: the scale factor for each parameter, used by the optimizer. example: (Scales -1 -1 -1 100000000 -1 -1 -1 -1 -1 -1 -1 -1) The scales work a bit different in this transformation, compared with the other transformations like the EulerTransform. By default, if no scales are supplied, the scales are estimated automatically. With the scales parameter, the user can override the automatically determined scales. Values smaller than or equal to 0 are ignored: the automatically estimated value is used for these parameters. Positive values override the automatically estimated scales. So, in the example above, the shear_x parameter is given a very large scale, which effectively turns off the optimization of that parameters. This might be useful in some MR-DTI application.