The read function in diffdrr.data seems to have a bug.

[shenao1994]

When I was using version 0.4.2 of diffdrr, I saved the volume in the subject.

data["image"] = subject.volume.data[0].cpu().numpy().astype(np.float32)

And the oriention of the saved CT is consistent. I visualized the CT slices.

But when I use diffdrr versions ≥ 0.4.4, the saved CT directions are inconsistent. As shown in the following figure.

Can you help me check what the problem is ? Thank U !

[eigenvivek]

hi @shenao1994 , so just to confirm, when you use diffdrr.data.read with v0.4.2 and v0.4.4+, you get different results?

[shenao1994]

hi @shenao1994 , so just to confirm, when you use diffdrr.data.read with v0.4.2 and v0.4.4+, you get different results?

Yes. Can you help me check it?