I am not trying to gain more information through re-sampling, I am only trying to reduce the voxel size so that my segmentation contours are smooth. The issue with working at the original MRI resolution is the segmentation quality. I will try to attach images to explain this visually, but because of the scale of the cartilage relative to the voxel size, the segmentation becomes quite jagged and not smooth. The first two images are of the original MRI while the second two are of the resampled one.
I have tried to apply smoothing algorithms or other smoothing methods but those seem to always change the volume of my segmentation in unpredictable ways. Is there a way to improve the smoothness of these segmentations that doesn’t involve me re-sampling the MRI images to unreasonable sizes?
When using the “Fill between slices” tool, is the interpolation along 1 dimension? Or is it in 3 dimensions? Would it only be interpolating between the slices or also within the slices to fill in the gaps? If the interpolation occurs in three dimensions, is there a way to change the interpolation to be restricted to a given dimension?