Fuji CR image shown as inverted in WADO - dcm4chee 2.x and 5.x

[Prakash Jayaraman]

DICOM CR Images received from a particular Fuji X-Ray equipment are shown as inverted in the WADO responses of dcm4chee 2.x and dcm4chee-arc-light 5.x.

But the few standalone DICOM workstations (Horos, etc.) are rendering the images as expected (i.e Non-inverted).

When we analysed the DICOM dataset, the default WindowCenter/WindowWidth tag values are not present and the PresentationLUTShape tag value is "INVERSE".

I assume that this has to be handled and rectified in the WADO module (may be using LUT - I'm not sure), since the images are rendered as expected in the DICOM workstation applications. 

Has anyone come across this issue? Are there any solutions for it? Any help would be very much appreciated. I hope Gunter can help in this.

[Gunter Zeilinger]

An anonymized sample image would help...

[Prakash Jayaraman]

Sure. I'm attaching an anonymized sample CR image. 

Note : The attached CR image is compressed in JPEG LS Lossless due to the restriction in attachment size.

---------- Forwarded message ---------

From: Mail Delivery Subsystem <mailer...@googlemail.com>
Date: Wed, Apr 14, 2021 at 2:04 PM
Subject: Delivery Status Notification (Failure)
To: <prakashj...@gmail.com>

Hello Prakash Jayaraman <prakashj...@gmail.com>,

The message you are sending is too large.  All messages must be less than 8.00 MBytes.

If you have questions related to this or any other Google Group, visit the Help Center at https://groups.google.com/support/.

Thanks,

Google Groups  

--
You received this message because you are subscribed to the Google Groups "dcm4che" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dcm4che+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dcm4che/47dab6cc-be51-4c6b-880b-14fbe6c931a5n%40googlegroups.com.

[Gunter Zeilinger]

The image violates the IOD for DX by several attributes. Particularly DICOM PS3.3, C.8.11.3 DX Image Module specifies:

> Presentation LUT Shape (2050,0020)
> Specifies an identity transformation for the Presentation LUT, other than to account for the value of Photometric Interpretation (0028,0004), such that the output of all grayscale transformations defined in the IOD containing this Module are defined to be P-Values.
> Enumerated Values:
> IDENTITY: output is in P-Values - shall be used if Photometric Interpretation (0028,0004) is MONOCHROME2.
> INVERSE: output after inversion is in P-Values - shall be used if Photometric Interpretation (0028,0004) is MONOCHROME1.

so:

1428: (0028,0004) CS #12 [MONOCHROME2] PhotometricInterpretation
:
1730: (2050,0020) CS #8 [INVERSE] PresentationLUTShape

is an inconsistency, and the rendering application has to decide which attribute shall be believed.

Also mandatory attributes:

• Pixel Intensity Relationship (0028,1040)
• Pixel Intensity Relationship Sign (0028,1041)
• Rescale Intercept (0028,1052)
• Rescale Slope (0028,1053)
• Rescale Type (0028,1054)
• Lossy Image Compression (0028,2110)
• Burned In Annotation (0028,0301)
• Window Center (0028,1050)
• Window Width (0028,1051)

are missing and

370: (0008,0008) CS #26 [ORIGINAL\PRIMARY\RT\150000] ImageType

violates DICOM PS 3.3, C.8.11.3.1.1 Image Type

> Value 3 (which is specific to the IOD) shall be present and have zero length (null value).

[Gunter Zeilinger]

Realized, that dcm4che's rendering code already prioritize

(2050,0020) CS #8 [INVERSE] PresentationLUTShape

over

(0028,0004) CS #12 [MONOCHROME2] PhotometricInterpretation

So actually the value for the Presentation LUT Shape in the Fuji CR image is wrong (it should be 

(2050,0020) CS #8 [IDENTITY] PresentationLUTShape

), were

(0028,0004) CS #12 [MONOCHROME2] PhotometricInterpretation

is correct.

[Gunter Zeilinger]

Optionally ignore present (2050,0020) Presentation LUT Shape on retrieve of rendered DICOM images by WADO-RS or WADO-URI services #3155