Better performance of deepseg seg_mice_gm-wm_dwi model if A-P orientation is wrong #3279
Labels
enhancement
category: improves performance/results of an existing feature
sct_deepseg
context: Global entry point for all deep learning segmentation methods
user requested
Raised by user on the SCT forum/email/GitHub. Be sure to notify them when fixed in a release.
Projects
Description
While working on a user's forum issue https://forum.spinalcordmri.org/t/registration-to-t1-ex-vivo-template-error/644/6, I noticed that
sct_deepseg
's modelseg_mice_gm-wm_dwi
performs better if the image is (wrongly) flipped in the A-P direction.See below with the correct orientation, ie: we clearly see the dorsal horns pointing towards "A" whereas it should be "P":
wrong-orientation.mov
However, when fixing the wrong orientation of the image using the following function:
Even though the image is now correctly oriented, the segmentation performance is largely decreased:
correct-orientation.mov
Which makes me wonder if the model may have been trained on flipped data? (ie: A-P were mislabeled). @charleygros
The text was updated successfully, but these errors were encountered: