Exploiting variational auto-encoders for tumor subregion segmentation from 3D MRIs

Home

Discover

Bookmarks

Keep Up With Top Trending Papers In CS / AI / ML ...

Source arXiv

Computer Vision

Neurons and Cognition

Exploiting variational auto-encoders for tumor subregion segmentation from 3D MRIs

3D MRI brain tumor segmentation using autoencoder regularization

We describe a semantic segmentation network for tumor subregion segmentation from 3d magnetic resonance images (3d mris) based on encoder-decoder architecture.

A variational auto-encoder branch is added to reconstruct the input image itself in order to regularize the shared decoder and impose additional constraints on its layers.

The current approach won 1st place in the brats 2018 challenge.

Authors

Andriy Myronenko

◐ Recommended Members

◐ Latest Activity