search

BICCN / cell-locator

manually align specimens to annotated 3D spaces
https://cell-locator.readthedocs.io
Other
19 stars 7 forks source link

Slice Offset in certain scenarios differs from expected value. (Slicer 2020) #139

Closed allemangD closed 3 years ago

allemangD commented 4 years ago

Environment

Description

The default slice offset is -6575, but according to the qa checklist is expected to be 6600.

Similarly, performing the Roll/Pitch/Yaw test in the qa checklist produces a slice offset of -9472, but is expected to be 9478.

The numbers are close, except for their sign. I have a hunch this is due to the changed coordinate system, and may be related to issue https://github.com/BICCN/cell-locator/issues/101.

jcfr commented 4 years ago

I have a hunch this is due to the changed coordinate system, and may be related to issue #101.

Agreed with your assessment.

allemangD commented 3 years ago

This is just due to the coordinate system change.

For example, with the Coronal slice offset: the anterior extent offset is -13175um, and the initial offset is -6575um; then the difference from the extent is then +6600um. So when the coordinates were negated, there would have been a +6600um.

The other offset differences are:

Since the relative positions are still the expected values, I'll update the qa checklist with the positions and a note mentioning this comment. If we do want to maintain the old values (and all-positive coordinates), that should be done as part of the work for https://github.com/BICCN/cell-locator/issues/196 by correcting the coordinate system and offsets used in the CCF atlas.