MixNet: Multi-modality Mix Network for Brain Segmentation

TL;DR

MixNet is a novel multi-modality deep learning network for brain segmentation that achieves state-of-the-art results by effectively combining multi-scale and multi-modal information without pooling layers.

Contribution

This work introduces MixNet, a deep residual CNN with dilated convolutions and pyramid pooling for improved multi-modality brain segmentation.

Findings

Achieved state-of-the-art segmentation performance on MRBrainS dataset.

Won 3rd place in MRBrainS challenge 2018 with limited training data.

Demonstrated effectiveness of multi-modality fusion architectures.

Abstract

Automated brain structure segmentation is important to many clinical quantitative analysis and diagnoses. In this work, we introduce MixNet, a 2D semantic-wise deep convolutional neural network to segment brain structure in multi-modality MRI images. The network is composed of our modified deep residual learning units. In the unit, we replace the traditional convolution layer with the dilated convolutional layer, which avoids the use of pooling layers and deconvolutional layers, reducing the number of network parameters. Final predictions are made by aggregating information from multiple scales and modalities. A pyramid pooling module is used to capture spatial information of the anatomical structures at the output end. In addition, we test three architectures (MixNetv1, MixNetv2 and MixNetv3) which fuse the modalities differently to see the effect on the results. Our network achieves the state-of-the-art performance. MixNetv2 was submitted to the MRBrainS challenge at MICCAI 2018 and won the 3rd place in the 3-label task. On the MRBrainS2018 dataset, which includes subjects with a variety of pathologies, the overall DSC (Dice Coefficient) of 84.7% (gray matter), 87.3% (white matter) and 83.4% (cerebrospinal fluid) were obtained with only 7 subjects as training data.