Closed dillonwilliams closed 3 years ago
I think pynetdicom
relies on pydicom
's Dataset
class. pydicom
's dataset is generic enough that I think we can use it in the API.
pydicom
's Datasets are like numpy
's ndarray
class within numerical computing---most numerical programs use it as the common data structure.
@yujanshrestha what do you think about this?
I think it is ok to use pydicom's Dataset
as an API. At least currently, if a Python project deals with DICOM, it probably depends on pydicom.
I think send_report_to_pacs
and send_annotations_to_pacs
should be implemented by the library user. The library should just provide a generic way to push DICOM to a PACS.
I think pacsman should just have a push_dicom_ds(ds)
function.
send_datasets
implements this.
Need ability to store:
pynetdicom
has a general DICOMDataset
, but the abstract methods can't be tied to a particular implementation. Probably better to have separate storage methods for each use-case, likedef send_report_to_pacs(self, patient_id, report_path)
,def send_annotations_to_pacs(self, patient_id, instance_id, json_descriptor_list)
.