Figure out why some of the images and masks don't line up

[richard-lane]

Some of wahab's input images are stored as stacks of 2D tiffs, which I've converted to 3D with a python script (351, 401, 420, 441 using the new n convention - the 2d tiffs are stored using the old_n convention, the correspondence between them can be found here).

When I do this however, the images don't match up with the masks. It looks like this is partially because the z-axis is inverted, but even after flipping or rotating the images don't quite line up. I haven't been able to find a combination of flips, rotations and transposes that make the images and masks line up, even though it does look like the mask should correspond to the image.

We can consider this issue closed when we can create DICOMs from these images, and have plotted them to show that they line up properly.