Fabian Laqua

Thank you Nick for your kind and quick reply. And of course I need to thank Brian and you also for spending all that effort into ANTs and ITK. I think this can not be stated too often. I feared, that it would end up in trial and error. Do you have any clue where to start adjusting parameters? Yes you are right, registering every image to end-diastole could also be an option, I should test. Initially i felt like the images( especially systolic) would differ too much and that is why I tried adjacent...

Hi all, I am trying to register a series of 30 time-resolved cyclic cardiac images (original dimension 256x192, cropped to 94x94) as pairs using a 3D approach where Fixed:z=t=(0,...29) and Moving:z=t=(1,...,29,0) where the third dimension represents time. My primary goal is not to transform images, but to use the DisplacementFields to propagate a set of indicatorpoints from image 0 through all images. I have some questions concerning this: 1) As far as I understood it, SyN and its derivates are symmetrical...