Efficient Brain Tumor Segmentation With Multiscale Two-Pathway-Group Conventional Neural Networks

Manual segmentation of the brain tumors for cancer diagnosis from MRI images is a difficult, tedious, and time-consuming task. The accuracy and the robustness of brain tumor segmentation, therefore, are crucial for the diagnosis, treatment planning, and treatment outcome evaluation. Mostly, the automatic brain tumor segmentation methods use hand designed features. Similarly, traditional methods of deep learning such as convolutional neural networks require a large amount of annotated data to learn from, which is often difficult to obtain in the medical domain. Here, we describe a new model two-pathway-group CNN architecture for brain tumor segmentation, which exploits local features and global contextual features simultaneously. This model enforces equivariance in the two-pathway CNN model to reduce instabilities and overfitting parameter sharing. Finally, we embed the cascade architecture into two-pathway-group CNN in which the output of a basic CNN is treated as an additional source and concatenated at the last layer. Validation of the model on BRATS2013 and BRATS2015 data sets revealed that embedding of a group CNN into a two pathway architecture improved the overall performance over the currently published state-of-the-art while computational complexity remains attractive.

I. Introduction

Cancerous brain tumors present themselves as unnatural, uncontrolled growth and division of cells in the brain. While brain tumors are not very common, they are one of the most lethal cancers. In the US alone for example, approximately 23,000 new cases of brain cancer were diagnosed in 2015. It is an abnormality in the brain tissues that damage the nervous system severely, which result patient death. are the most common brain tumors that are infiltrative in nature, and occur near white matter fibers. They may spread to any part of the brain making it difficult to detect. High-grade gliomas are considered one of the most aggressive tumors with a median survival of 15 months. Gliomas can be measured by MRI using multiple sequences such as T2-weighted uid attenuated inversion recovery (Flair), T1-weighted (T1), T1-weighted contrast-enhanced (T1c), and T2-weighted (T2) using existing automated brain tumor segmentation techniques.