Add checks for mask size and dimensions. #223
Merged
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 function to check if ROI is large enough for computation. First checks if label is present in the mask, then checks if the number of dimensions in which the size > 1 is at least the minimum number of dimensions (default: 1). Finally an an optional check for a minimum number of voxels in the tumor region can be performed (switched off by default). This check is performed in `featureextractor` after the loading of the image and any image pre-processing steps (e.g. resampling), but before features are extracted or filters applied.
Add a check that ensures that the ROI defined in the mask occupies a physical space that is contained within the physical space of the image. If this is not the case, log an error and cancel extraction for that combination. Additionally, calculate the resampling grid using only the maskNode. This prevents unexpected behaviour of the defined resampling grid when image and mask with different geometry is used. Standardize use of quote and double quote when defining strings in PyRadiomics (use single quote to identify a string, 3 double quotes for a docstring)
Add FAQ explaining PyRadiomics behavior when using image and mask with different geometry. Additionally, add an example script, usable from the commandline, that resamples a mask to a reference image. This allows users to force a mask to the same geometry as the image, so it can be used in PyRadiomics without resampling.
fedorov
reviewed
Mar 24, 2017
| @@ -74,7 +74,8 @@ def main(): | |||
|
|
|||
| provenance = extractor.getProvenance(imagePath, maskPath, mask) | |||
|
|
|||
| image, mask, bb = imageoperations.cropToTumorMask(image, mask) | |||
There was a problem hiding this comment.
I would avoid using "Tumor". Maybe "cropToMaskBoundingBox", or "cropToMaskROI".
There was a problem hiding this comment.
Sorry, I understand this is not related to the essence of this specific PR
There was a problem hiding this comment.
Sorry, I understand this is not related to the essence of this specific PR
No, it's not, but it's still a good suggestion. I am running over the documentation for errors, will update this then also.
|
LGTM otherwise! |
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.
Add function to check if ROI is large enough for computation.
First checks if label is present in the mask, then checks if the number of dimensions in which the size > 1 is at least the minimum number of dimensions (default: 1). Finally an an optional check for a minimum number of voxels in the tumor region can be performed (switched off by default).
This check is performed in
featureextractorafter the loading of the image and any image pre-processing steps (e.g. resampling), but before features are extracted or filters applied.Addresses #18
cc @Radiomics/developers