Some cases have common error

[shawnyuen]

Thanks for sharing the large-scale dataset.

I open these nii.gz files via ITK-SNAP, and found the following cases having common error: case_00015, case_00025, case_00061 and case_00117, repeating some slices containing kidney and tumor, but without corresponding mask 'kidney' and mask 'tumor'.

I recommend you remove these slices in the above three cases.

[neheller]

Thanks for helping with quality assurance! We'll look into what's going on here and update the data when we find a fix.

[luvWY]

Thanks for sharing the large-scale dataset.

I open these nii.gz files via ITK-SNAP, and found the following cases having common error: case_00015, case_00025, case_00061 and case_00117, repeating some slices containing kidney and tumor, but without corresponding mask 'kidney' and mask 'tumor'.

I recommend you remove these slices in the above three cases.

I download the dataset 5five days ago and I use np.max to check 'case_00025' segmentation label ... however there is 2 or 1 pixel... How to conclude " without corresponding mask 'kidney' and mask 'tumor'." I just wanna know whether I miss something..

[neheller]

In case_00025 for instance, somehow the imaging taken at a different contrast phase got appended to the end of the phase that we annotated. So there are annotated kidneys and tumor followed by unannotated kidneys and tumor.

[luvWY]

In case_00025 for instance, somehow the imaging taken at a different contrast phase got appended to the end of the phase that we annotated. So there are annotated kidneys and tumor followed by unannotated kidneys and tumor.

OK..I check the code. You are right. I also found "case_00160" has a different shape of img: (512, 796) while others are (512, 512).. I dont know whether it's ok. I just post it here..

[neheller]

It is somewhat odd that case_00160 has a different shape, but it's not an error. When building your model, you should account for the fact that the test cases might not all be 512x512.

[neheller]

Case 15 was replaced entirely for unrelated reasons, and the other three have been cropped accordingly. Thanks again for pointing this out.