There are several data points that have a wrong physical direction

[dichen9412]

There are several data points that have a wrong physical direction:
(these are my file name but it pretty much maintain the original order)
../Data/dataset2_0003_4_325
../Data/dataset2_0460_5_325
../Data/dataset2_0480_2_324
../Data/dataset2_0616_4_325_hard
../Data/dataset2_0703_4_325
../Data/dataset2_0704_4_325

[EricKani]

Thanks for your issue.

I have checked the sequences you pointed out. But I didn't find any problem in these data and annotations. Can you explain it more clear?

Best,
Eric

[dichen9412]

I mean for all other data points, after orienting them based on the direction information in the file, they all get to the right view.
For these 6 cases, two dimensions are reflected somehow.

[dichen9412]

B.T.W. I was using the simpleITK to load these data and get the direction and spacing information from these data.

[dichen9412]

And I also tried to load these data using the 3D slicer and the sagittal/coronal view of these data points are wrong.

[EricKani]

Is the "reflected wrong" you mean mirror in the sagittal, e.g. liver in the left? (But it's turly in the right, they are just in the prone position)

And I also tried to load these data using the 3D slicer and the sagittal/coronal view of these data points are wrong.

ps: There are truly some weird cases with a "reflected wrong" in the saved array in dicom/nii files, rectifying view in ITKsnap/MITK by "-1" in the direction information. (but not these cases you point out.)

[dichen9412]

Em... I mean, compared to other cases, when you load these cases, their direction is different than others after orientating everyone to the normalized view (X: S->I Y: A->P Z: R->L). It is completely different. Not just a little difference of organs.

Can you try to load these data points to software like "slicer" (which would automatically extract the coronal/sagittal/axial view of the input scan based on the associated affine (direction) information.)

[dichen9412]

To be specific, after orientating every scan to the sagittal view, these cases' Y,Z-axes are reversed compared to other cases.

[dichen9412]

I found this because I saw the right hip mask is on the left side and the left hip is on the right side.