Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Region based nnunet scripts #66

Merged
merged 50 commits into from
Nov 10, 2023
Merged

Region based nnunet scripts #66

merged 50 commits into from
Nov 10, 2023

Conversation

plbenveniste
Copy link
Collaborator

@plbenveniste plbenveniste commented Oct 21, 2023 •
edited
Loading

In this PR we detail the scripts used for the training and evaluation of the region-based nnUNet. It contains:

  • README.md : instructions to use the different files
  • convert_BIDS_to_nnunet.py : script to convert data from the BIDS format to the nnUNet format
  • convert_predictions_to_BIDS.py : script to convert the nnUNet predictions to the BIDS format using the conversion dictionnary
  • evaluate_lesion_seg_prediction.py : this script computes the metrics using anima
  • nnUNet_inference_analysis.ipynb : this notebook shows the performance of the nnunet with respect to its different folds
  • convert_BIDS_to_nnunet_with_mul_PSIR.py : this last script was added to convert BIDS to nnunet format and multiplying PSIR images by -1

This list is to be updated.

@plbenveniste
conversion to nnunet region-based format
b77ea9f
@plbenveniste
modified for training or inference
a5f82a1
@plbenveniste
added exclusion based on exclude.yml file
00e437f
@plbenveniste
created file for evaluation of lesion seg pred
ea37122
@plbenveniste
added anima function in script
6548a4d
@plbenveniste
added readme file for anima installation
ca5e645
@plbenveniste
changed folder structure and adding save to csv
e945a7e
@plbenveniste
removed voxel labels above the first disc level
87c3766
@plbenveniste
changed file termination to seg-manual and removed break
f3038d8
@plbenveniste
added dictionnary for inference
fb6c94e
@plbenveniste
changed encoding to np.int16 for masks for nnunet compatibility
24c83bb
@plbenveniste
Created file for conversion from nnunet predictions to BIDS format
2a7e39e
@plbenveniste
modified script to run on entire folder (not single prediction)
d6d870b
@plbenveniste
removed two useless prints
3149d73
@plbenveniste
fixed selection of file in dictionnary problematic because of 000 at…
373b0be 
… end
@plbenveniste
created file for nnunet dataset generation with multiplication by -1 …
d62c7f5 
…for PSIR images
@plbenveniste
fixed site value and added contrast value to dictionnary
c00e7a9
@plbenveniste
create notebook to compare fold results
edc3a50
@plbenveniste
finished analysis of first nnUnet predictions
22965ca
@plbenveniste plbenveniste self-assigned this Oct 21, 2023
@plbenveniste plbenveniste marked this pull request as ready for review November 9, 2023 21:50
@plbenveniste plbenveniste requested a review from valosekj November 9, 2023 21:51
valosekj
valosekj reviewed Nov 10, 2023
View reviewed changes
nnunet/README.md Outdated Show resolved Hide resolved
valosekj
valosekj reviewed Nov 10, 2023
View reviewed changes
nnunet/README.md Outdated Show resolved Hide resolved
valosekj
valosekj reviewed Nov 10, 2023
View reviewed changes
nnunet/README.md Outdated Show resolved Hide resolved
valosekj
valosekj approved these changes Nov 10, 2023
View reviewed changes
Copy link
Member

@valosekj valosekj left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM! Thanks! I left only a few minor comments. Then the PR can be merged!

@plbenveniste plbenveniste merged commit b73709c into main Nov 10, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@valosekj valosekj valosekj approved these changes

Assignees

@plbenveniste plbenveniste

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@plbenveniste @valosekj