Open fedorov opened 4 years ago
@fedorov Do you have a link to a study on the sandbox causing these issues?
I believe it is this one: https://idc-sandbox-000.firebaseapp.com/projects/idc-tcia/locations/us/datasets/issues/dicomStores/issue-2-kits/study/1.3.6.1.4.1.14519.5.2.1.3344.2526.267929376932212971281995255548
but it is now failing with a different error
@igoroctaviano What is the status of this?
@sedghi looks like it's still a problem in v2 and also for v3:
@sedghi looks like it's still a problem in v2 and also for v3:
@fedorov do we want to prioritize this issue? (this was created for v2 for looks like it applies to v3)
@sedghi FYI For the first iteration, we should avoid crashing the viewer for this type of error, then later we can address the cause: If ReferencedSeriesSequence of an SR is not part of the current study we should fallback to a series that shares the frame of reference which is not working at the moment.
Actually thisReferencedSeriesSequence
is causing problem in other data in TCIA with SEG that recently got added by SPL. As far as I understand this is a required tag.
Data: https://www.dropbox.com/s/2sbi42khfesmb86/ReMIND-001.zip?dl=0
@fedorov This is the ReMIND data from Tina. Do you happen to know if that ReferencedSeriesSequence
is necessary to be available for SEG? seems like so here https://dicom.innolitics.com/ciods/segmentation/common-instance-reference/00081115
Do you happen to know if that ReferencedSeriesSequence is necessary to be available for SEG?
No, it is not required when segmentation geometry (ImageOrientation/Position/Spacing etc) is described in the SEG instance, which is typically the case.
We just tested this in OHIF demo viewer local mode with the dataset from dropbox above, and the error was different. The yellow circle with exclamation mark had a tooltip that that data type (or something) is not supported.
Yeah I fixed the error that prevented the viewer from showing up, now only the SEG is not loadable. But still not sure what should we do regarding that tag missing (which is required as you mentioned)
Yeah I fixed the error that prevented the viewer from showing up, now only the SEG is not loadable. But still not sure what should we do regarding that tag missing (which is required as you mentioned)
The case here is that the tag is there but references a series not part of the current study. In this case, it should fallback to a frame of reference (it still should be able to render fine by a frame of reference even though the tag references a missing series)
Bug Report
Describe the Bug
Trying to load a study that has a SEG referencing an image, but without corresponding image in that same study, there is a cryptic runtime error:
What steps can we follow to reproduce the bug?
Populate a Google Healthcare DICOM store with a single SEG series that references another series in
ReferencedSeriesSequence
, and try to open the study with that SEG in OHIF Viewer.Observed with 3.11.7.