[WIP] 141 add dicom loader #200
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
* Add Resize transform (spatial scaling). * Adding tests.
* [DLMED] add UNet example with dict based transforms * [DLMED] temporarily remove constants, will discuss after GTC * [DLMED] adjust default behavior according to comments
* Adding zoom transform and tests. * Fix import error.
* affine transforms * randomised affine and elastic deformation * add unit tests * add a 2D notebook demo * add speed demo
* [DLMED] add LoadNifti transform and common Dataset * [DLMED] update according to comments * [DLMED] fix typo * [DLMED] update example to use latest API * [DLMED] update code to add more features * [DLMED] update to AsChannelFirst * [DLMED] update to use np.moveaxis API instead * [DLMED] update according to comments * update generalized dice to be consistent with the changes in dice loss Co-authored-by: Wenqi Li <wenqil@nvidia.com>
* Add RandomRotate.
Co-authored-by: Kevin Lu <kevlu@nvidia.com> Co-authored-by: Wenqi Li <wenqil@nvidia.com>
Co-authored-by: Nic Ma <nma@nvidia.com>
* [DLMED] add 3D classification inference examples
* [DLMED] change UNet to 1 output
* [DLMED] add 3D classification inference examples
* [DLMED] update and clear all the examples
also added unit tests for ClassificationSaver and Resized
* [DLMED] fix transforms spatial axis issue and update unit tests
* update demos:
- unet: num_classes -> out_channels (signature consistency)
- segmentation demo changed to binary ground truth
- changed segmentation training dict to have validation every n iterations
- segmentation demo image sizes to 128, window size to 96, lr 1e-3 (good results)
- attach classification saver finalize() to Events.COMPLETE
* [DLMED] fix tag_name bug, fix DenseNet3D bug, add comments
Co-authored-by: root <root@apt-sh-ai.nvidia.com>
Co-authored-by: Wenqi Li <wenqil@nvidia.com>
* [DLMED] add DeleteKeys transform * [DLMED] remove unnecessary copy
* [DLMED] add highlight features to README
* [DLMED] add temp images path to highlight features
* documentation updates
- update readmet
- update docs/ structure
- merged highlights.md
- update docs/Makefile commands
* [DLMED] update end-to-end process chart
* revise docstring for transforms/losses
* update docs
* update readthedocs
* 179 add DeleteKeys transform (Project-MONAI#180)
* [DLMED] add DeleteKeys transform
* [DLMED] remove unnecessary copy
* update docs
* fixes github urls
* fixes sys.getsizeof test on mac
Co-authored-by: Nic Ma <nma@nvidia.com>
) * [DLMED] add ScaleIntensityRanged and NormalizeIntensityd transforms * [DLMED] add pytorch example based on MSD spleen dataset * update docs for new transforms, revise typos/styles * [DLMED] add train notebook * [DLMED] update notebook with determinism * [DLMED] add image_threshold Co-authored-by: Wenqi Li <wenqil@nvidia.com>
* added setuptools support updated README * removed relative paths * changed form exchanged raw install by pip install with folder flag
* init. docker * fixes urls * update readme
* [DLMED] add determinism integration test and fix bugs * [DLMED] add determinism support to highlights * [DLMED] add transforms to the test * [DLMED] update to RandUniformPatch * fixes typos * fixes empty indicies issue in balanced sampling Co-authored-by: Wenqi Li <wenqil@nvidia.com>
Nic-Ma
reviewed
Mar 23, 2020
| @@ -34,6 +35,38 @@ def contain_dicom(path): | |||
| return False | |||
|
|
|||
|
|
|||
There was a problem hiding this comment.
Thanks for your update here.
Actually, I think you can treat this converter tool as a temporary reference code and move all the data converter logic into other transforms(or new transforms) after "LoadDICOM".
When all logic is ready in transforms, we can delete this tool and users can use transforms to load DICOM directly.
Thanks.
There was a problem hiding this comment.
Hi @mhubii ,
Do you want to merge this PR first to my previous DICOM PR first and work on that PR directly?
Thanks.
|
this is out of date, please see #856 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Adds a DICOM laoder to #141
Description
masterinto135-add-DICOM-loaderbranchtransforms.LoadDICOM,composables.LoadDICOMdbased on design choices in add a DICOM loader #135 (comment)utils.medical_image_convertertest_load_dicom.pyandtest_load_dicomd.pyStatus
Documentation to be added
Types of changes