[HZN-87] Incompatible DICOM Formats

[laurencejackson]

Transferred by @LaurenceJackson on Aug 20, 2020, 11:47

:::Issue recreated from Jira issue HZN-87:::

During Acceptance Testing of 3T Siemens Lumina (Fortius Clinic, Wimbledon), it wasn't possible for images exported as 'Enhanced' to be read by Hazen (or MATLAB or pydicom I think). Only 'interoperable' files were readable. In addition, "Create DICOM file system" needed to be UNTICKED. See picture below.

!DICOM options.png|width=200,height=183!

Sample files from Enhanced DICOM and when 'Create DICOM file system' was ticked are here:

!image-2020-08-20-10-46-51-415.png|width=200,height=183!

[laurencejackson]

Transferred by @LaurenceJackson on Aug 20, 2020, 12:00

[Comment originally posted by Caitlin O'Brien 2020-08-20T11:00:21.597+0100

Note: When the Siemens engineers set up a system they should set system default for exporting dicom files to be interoperative. Many PACS systems will complain if given enhanced dicom files. ]

[laurencejackson]

Transferred by @LaurenceJackson on Aug 20, 2020, 11:31

[laurencejackson]

Transferred by @LaurenceJackson on Aug 20, 2020, 11:46

[YassineAzma]

Perhaps the issue I'm having is linked; I'm attempting to load in Philips enhanced DICOM data, which is appearing to be a single image in the DICOM metadata when loaded with the functions used in init.py.

When I replicate this in jupyter, without the following exceptions (is_dicom_file or whether the file is on the list of EXCLUDED_FILES), the PerFrameFunctionalGroupsSequence field appears as expected, which allows you to access the metadata for each image/frame as normal.

[RFranklin-MR]

Relatively low priority for MR Physics, as there is a work around (download Interop images). However, cannot comment on the related issues above.