Dicom to NIfTI conversion issue if the image orientation does not exist

[omaghsoudi]

Describe the bug The dataset, that we have in Dr. Kontos's Lab, has many cases with no ImageOrientationPatient in Dicom header. As you know, ImageOrientationPatient has six components showing the cosine factors for the affine transformation of an image plane to the world coordinate system. If this field in a Dicom header is not available, the CaPTK considers that to be zero for all six components. However, the real orientation can be estimated using Patient_Orientation, another field in the header file. This is a good webpage explaining the details for using Patient_Orientation to find an approximate orientation of an image in the world coordinate system: http://dicomiseasy.blogspot.com/2013/06/getting-oriented-using-image-plane.html If you care about the correct world coordinates, I think that it can be considered. I should mention that the dimensions are correct in the world coordinate system (because the pixel spacing is always available), but the orientation is totally wrong by assuming the values to be zero.

Screenshots An example NIfTI file from CaPTK when ImageOrientationPatient was not available in the header: Case3_mask.nii.gz

CaPTk Version 1.6.1

Desktop (please complete the following information): Linux and Windows

[sarthakpati]

Tagging people to keep them posted about this issue.

[sarthakpati]

@ashishsingh18 have you had a chance to look at this?

[ashishsingh18]

Not yet.