This dataset contains about 1000 3D CTA images, which is considerably larger than the existing public datasets. On the other hand, we also propose a benchmark based on the proposed dataset, in which we have not only implemented several typical existing methods but also proposed a strong baseline method.

An official data split is also prepared for fair comparison.

HIGHLIGHT 20230417: You may find some folders ending with (1), these are duplicated cases. We have updated the dataset to discard these duplicate cases. Please send me emails about this issue to get the link of the latest dataset.

HIGHLIGHT 20231101: We have deployed the dataset on Kaggle https://www.kaggle.com/xiaoweixumedicalai/datasets?scroll=true!

Please send emails to me [email protected] for the link and the password to download the dataset and the benchmark.