3D Image Reorientation and De-rotation

Facilitates assisted or automatic reorientation and automatic de-rotation of 3D image stacks

3D image analysis, particularly medical image analysis, often requires reorientation to a consistent reference frame for manual or semi-automatic determination of a 3D volume-of-interest (VOI). However, this reorientation can cause distortion of the original image stack due to interpolation error.
Here, we provide functions which enable 1) reorientation of 3D image stacks and 2) de-rotation of reoriented stacks back to the original reference frame. This enables the following workflow: 1) reorient an image to the desired orientation, 2) perform VOI selection of one or more areas of interest, and capture these as 3D logical masks (code is not provided for this step), and 3) de-rotate the completed logical masks back to the original reference frame so that they align with the original image. Through this methodology, VOI selection can be consistently performed in the desired reference frame without introducing interpolation error into the image analysis.
Two functions are provided in this set:
1) imstack_reorient - facilitates assisted or automatic reorientation of a 3D image stack to the desired reference frame via sequential rotations.
2) imstack_derotate - uses the outputs of imstack_reorient to perform de-rotation of 3D logical image stacks back to the original orientation.