Thanks for your great works. But I have some question about the performance of the model.
I saw in the paper that SwinUNETR can obtain an average dice score of 0.918 on BTCV, but the model you provided only has a dice score of 0.8225
So I would like to know how to approach the scores in the paper. Thank you!
Thanks for your great works. But I have some question about the performance of the model. I saw in the paper that SwinUNETR can obtain an average dice score of 0.918 on BTCV, but the model you provided only has a dice score of 0.8225 So I would like to know how to approach the scores in the paper. Thank you!