Nick Tustison
Nick Tustison
Sure. Just create a pull request. Thanks.
Since ITK maintains a rigid distinction between "index" and "continuous index" types (and I think that should be continued here), perhaps a separate function should be created for continuous indices.
So what's wrong with having a separate function?
As I mentioned before, I think it's the scale parameter estimator for the optimizer. When the voxel size gets small, the perturbation blows up transforming the image domain outside of...
Okay, so it's the Mattes gradient that is used to estimate the scale parameter which blows up. Note that the measurement seems stable across voxel size. Here's the gradient for...
Sure, give me a sec and I'll make a pull request.
We don't have much existing labeled/prior data (including segmentation priors) for dealing with ICV extraction vs. brain extraction. Perhaps someone has already done that with the MNI template. We have...
Can you post data for me to play with? Assuming that the user would have to supply the initial lesion masks for a given cohort, I don't know if that...
I defer to @stnava for pushing this along but thanks for this @aganders3 .
I agree that the proposed modification would be useful for certain cases. And it's definitely possible. We can certainly leave this as a 'to-do' item but, honestly, the code overhaul...