DATASET

[chengengle]

Thank you for your contribution. I would like to ask about the specific data partitioning used for the five-fold cross-validation of topcow2023 in your paper, and whether the training was conducted within the ROI region. Thank you

[ykirchhoff]

Hi @chengengle,

thanks for the questions. I will need to check for the exact data partitioning.
For the experiments in our paper we trained on the whole ROI and only evaluated on the ROI. In later experiments we set the brain outside of the CoW ROI to the ignore label, but that doesn't seem to make a huge difference.

Best,
Yannick