nii format of the data and annotation

Dear organizers,

The annotation file is hard to parse. Would it be possible for you to show how to conver the annotations to nii format in the notebook? https://drive.google.com/file/d/1DzWLq0mVWeoZ6f6vLXtSKN1h98T9GPUC/view?usp=sharing

It would be highly appreciated if you can provide a csv file for the atherosclerosis diagnosis.

The reason to use an XML format in this challenge is that we use a well-established software[1]: CASCADE, to carry out the manual segmentation and label for carotid vessel wall images, in which the sub-pixel segmentation and smoothness of the segmentation boundaries were all well dealt with and recognized by the vessel wall imaging community. Therefore, we provide the vessel wall segmentation labels based on the format of CASCADE, which is in XML. We totally agree with you that this may be difficult to work with. To solve this problem, we provide the reference Python code https://drive.google.com/file/d/1DzWLq0mVWeoZ6f6vLXtSKN1h98T9GPUC/view?usp=sharing to illustrate how to load the XML files.

And cv2.fillPoly() is avalible to convert the contour to a binary mask.