Format

Send to

Choose Destination
Med Image Anal. 2017 Jan;35:18-31. doi: 10.1016/j.media.2016.05.004. Epub 2016 May 19.

Brain tumor segmentation with Deep Neural Networks.

Author information

1
Université de Sherbrooke, Sherbrooke, Qc, Canada. Electronic address: seyed.mohammad.havaei@usherbrooke.ca.
2
École Normale supérieure, Paris, France.
3
Université de Montréal, Montréal, Canada.
4
Université de Montréal, Montréal, Canada; École polytechnique, Palaiseau, France.
5
Université de Montréal, Montréal, Canada; École Polytechnique de Montréal, Canada.
6
Université de Sherbrooke, Sherbrooke, Qc, Canada.

Abstract

In this paper, we present a fully automatic brain tumor segmentation method based on Deep Neural Networks (DNNs). The proposed networks are tailored to glioblastomas (both low and high grade) pictured in MR images. By their very nature, these tumors can appear anywhere in the brain and have almost any kind of shape, size, and contrast. These reasons motivate our exploration of a machine learning solution that exploits a flexible, high capacity DNN while being extremely efficient. Here, we give a description of different model choices that we've found to be necessary for obtaining competitive performance. We explore in particular different architectures based on Convolutional Neural Networks (CNN), i.e. DNNs specifically adapted to image data. We present a novel CNN architecture which differs from those traditionally used in computer vision. Our CNN exploits both local features as well as more global contextual features simultaneously. Also, different from most traditional uses of CNNs, our networks use a final layer that is a convolutional implementation of a fully connected layer which allows a 40 fold speed up. We also describe a 2-phase training procedure that allows us to tackle difficulties related to the imbalance of tumor labels. Finally, we explore a cascade architecture in which the output of a basic CNN is treated as an additional source of information for a subsequent CNN. Results reported on the 2013 BRATS test data-set reveal that our architecture improves over the currently published state-of-the-art while being over 30 times faster.

KEYWORDS:

Brain tumor segmentation; Cascaded convolutional neural networks; Convolutional neural networks; Deep neural networks

PMID:
27310171
DOI:
10.1016/j.media.2016.05.004
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center