Nadia

you can try softmax_dice2, ,you have to limit the label 0

Thank you for your impressive work! Could you also send me the trained model and data at [email protected]? I am looking forward to your reply. Thank you very much.

thanks for sharing！ could you share your yolov4.h5， thanks alot

why i got a vary poor performance on the model ？（use the brats dataset to segment brain tumor）

Hi! I found the article "Structure Boundary Preserving Segmentation for Medical Image With Ambiguous Boundary", it's wonderful. i Want to rrerun again, can you provide open source code? looking forword...