Hi, thanks for your great work and for sharing the code!
I have a question regarding the selection of slices used for training. In my experiments, I extracted the middle 100 slices from each MRI volume. I would like to ask how you and other users typically select slices — for example, do you use a fixed number of central slices, all slices containing the target region, or some other strategy?
Any suggestions or recommended practices would be greatly appreciated.
Thanks!