Feat/image library entries #77
Conversation
There was a problem hiding this comment.
Thanks @seandoyle. I defer to your judgment of whether and how we should implement modality-specific descriptors. However, that's something we can address with a separate PR.
There was a problem hiding this comment.
I just noticed that the tests fail due to style and type issues (https://travis-ci.org/github/MGHComputationalPathology/highdicom/jobs/773313211). @seandoyle you kindly fix those?
|
Yes - I'll address those. I also need to add some unit tests. |
|
The style issues are now fixed but there are other errors in templates.py which I believe are not related to my changes. Should I look at these as well? BTW: The DICOM SR validator needs a lot of resources. |
There was a problem hiding this comment.
Thanks @seandoyle. We can address the other typing errors in the target branch.
* Integrate new pydicom DS formatting * Fix recording of evidence in structured reports * Fix mypy errors * Add test for report referencing multiple studies * Add properties to content items * Use datetime workaround for python 3.6 support * Add properties and methods on SR templates * Update type of content item property return value * Fix typo in exception message * Fix recording of evidence in structured reports * Update package version * Add image library entries to SR documents (#77) * Fixed enum RelationshipType * Fix handling of library entry descriptors * Fix default value for Pixel Origin Interpretation * Implement indexing by dimension index values * Rename tracking id methods * Add check for correct segment numbers * Fix implementation of content property on SR documents * Make value and unit required for NUM content items * Fix return value type of properties of NUM content item * Add qualifier property to NUM content item * Return uids of type UID instead of str * Simplify reshaping of graphic data array * Add is_root parameter to alternative constructor * Allow constructor of UID to accept a value * Chang Optional[x] -> Union[x, None] in docstrings * Log warning if wrong number of content items * Add property for temporal range type * Move QualitativeEvaluation from content to templates * Allow Concept Name Code Sequence to be absent * Require Concept Name Code Sequence for CONTAINER * Let pydicom handle encapsulated frame decoding * Write error traceback to stderr instead of stdout * Add docstrings for enums * Implementation of Parametric Map IOD (#89) * Changed PatientSexValues to single letters * Expose enum at package root * Remove sphinx duplication for highdicom.enum * Alternative SCImage Constructor (#97) * Only warn if patient name check fails Co-authored-by: hackermd <[email protected]> Co-authored-by: Christopher Bridge <[email protected]> Co-authored-by: Sean Doyle <[email protected]> Co-authored-by: Sean Doyle <[email protected]> Co-authored-by: Markus D. Herrmann <[email protected]> Co-authored-by: Chris Gorman <[email protected]>
Small changes for ImageLibrary entries. The current output passes the pixelmed validator.
TBD - adding modality specific descriptors in Image Library Entries (TID 1602)