Automatic Brain Tumor Segmentation using Cascaded Anisotropic Convolutional Neural Networks

TL;DR

This paper introduces a cascaded anisotropic convolutional neural network approach for multi-region brain tumor segmentation in MR images, achieving high accuracy by decomposing the task into hierarchical binary segmentation steps.

Contribution

The novel cascade framework and multi-view fusion techniques improve brain tumor segmentation accuracy over existing methods.

Findings

Achieved Dice scores of 0.7859, 0.9050, 0.8378 on BraTS 2017 validation set.

Outperformed previous methods in segmentation accuracy.

Demonstrated robustness across multi-modal MR images.

Abstract

A cascade of fully convolutional neural networks is proposed to segment multi-modal Magnetic Resonance (MR) images with brain tumor into background and three hierarchical regions: whole tumor, tumor core and enhancing tumor core. The cascade is designed to decompose the multi-class segmentation problem into a sequence of three binary segmentation problems according to the subregion hierarchy. The whole tumor is segmented in the first step and the bounding box of the result is used for the tumor core segmentation in the second step. The enhancing tumor core is then segmented based on the bounding box of the tumor core segmentation result. Our networks consist of multiple layers of anisotropic and dilated convolution filters, and they are combined with multi-view fusion to reduce false positives. Residual connections and multi-scale predictions are employed in these networks to boost the segmentation performance. Experiments with BraTS 2017 validation set show that the proposed method achieved average Dice scores of 0.7859, 0.9050, 0.8378 for enhancing tumor core, whole tumor and tumor core, respectively. The corresponding values for BraTS 2017 testing set were 0.7831, 0.8739, and 0.7748, respectively.