Aakanksha Rana

this seems to be in 2D (https://github.com/LIVIAETS/boundary-loss/blob/master/keras_loss.py ). need to checkout its 3D version.

Another Model to add ASD-NET (MICCAI'18): github page https://github.com/ginobilinie/asdnet

Isnt this somewhat realized with intensity transformations available in intensity transformations such as noise addition ?

We don't do any quantize-ing yet in any of our augmentation, for sure. But this can be done and should be easy. Quantize-ing intensity and not using the continous gray-scale...

Hi @shmann to answer your question, losses and metrics in nobrainer are mainly designed to to stay aligned with [tf.keras.metrics](https://www.tensorflow.org/api_docs/python/tf/keras/metrics). style.

@shmann right. Actually a wrapper function can be created to use the thresholded y_pred values which can then estimate dice. Would you like to send a PR for the same?...

Labelled volumes would be helpful for transfer learning a model for animal scans, but these labels correspond to parcellation? and how many volumes are there?

> > To be clear, we don't have labelled volumes - yet; we are still assessing what would be needed on our part. > > But we could generate and...