Project-MONAI / Project-MONAI/tutorials
Data preparation for MAISI tutorial is wrong for non-RAS data
Nobody has claimed this yet.
- Dominant language
- Jupyter Notebook
- Stars
- 2.5k
- Forks
- 803
- Avg merge
- 6d 22h
- Merged PRs (30d)
- 3
Description
In the data preparation, the plain_transforms uses monai.transforms.Orientationd(keys="image", axcodes="RAS"), to reorient the image, but the new dimension is set using plain_transforms({"image": os.path.join(args.data_base_dir, filepath)})["image"].meta["dim"][_i] which has the original dimension, not the new dimension.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start at generation/maisi/scripts/diff_model_create_training_data.py around line 199 and inspect how plain_transforms reorients non-RAS images before the dimension is recorded. Verify that the dimension used for training-data preparation reflects the reoriented image, then run the tutorial's available data-preparation checks to confirm non-RAS inputs produce correct dimensions.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Clearly specified
- Newbie friendliness
- 55/100