Cascaded Global Context Convolutional Neural Network for Brain Tumor Segmentation

A cascade of global context convolutional neural networks is proposed to segment multi-modality MR images with brain tumor into three subregions: enhancing tumor, whole tumor and tumor core. Each network is a modification of the 3D U-Net consisting of residual connection, group normalization and deep supervision. In addition, we apply Global Context (GC) block to capture long-range dependency and inter-channel dependency. We use a combination of logarithmic Dice loss and weighted cross entropy loss to focus on less accurate voxels and improve the accuracy. Experiments with BraTS 2019 validation set show the proposed method achieved average Dice scores of 0.77338, 0.90712, 0.83911 for enhancing tumor, whole tumor and tumor core, respectively. The corresponding values for BraTS 2019 testing set were 0.79303, 0.87962, 0.82887 for enhancing tumor, whole tumor and tumor core, respectively.

[1]  Christos Davatzikos,et al.  Advancing The Cancer Genome Atlas glioma MRI collections with expert segmentation labels and radiomic features , 2017, Scientific Data.

[2]  Stephen Lin,et al.  GCNet: Non-Local Networks Meet Squeeze-Excitation Networks and Beyond , 2019, 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW).

[3]  Zhuowen Tu,et al.  Holistically-Nested Edge Detection , 2015, ICCV.

[4]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[5]  Yongdong Zhang,et al.  Deep Cascaded Attention Network for Multi-task Brain Tumor Segmentation , 2019, MICCAI.

[6]  Kaiming He,et al.  Group Normalization , 2018, ECCV.

[7]  Jonathan Tompson,et al.  Joint Training of a Convolutional Network and a Graphical Model for Human Pose Estimation , 2014, NIPS.

[8]  Konstantinos Kamnitsas,et al.  Ensembles of Multiple Models and Architectures for Robust Brain Tumour Segmentation , 2017, BrainLes@MICCAI.

[9]  et al.,et al.  Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge , 2018, ArXiv.

[10]  Tao Mei,et al.  Look Closer to See Better: Recurrent Attention Convolutional Neural Network for Fine-Grained Image Recognition , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[11]  Luca Antiga,et al.  Automatic differentiation in PyTorch , 2017 .

[12]  Shaoting Zhang,et al.  Global and Local Multi-scale Feature Fusion Enhancement for Brain Tumor Segmentation and Pancreas Segmentation , 2019, BrainLes@MICCAI.

[13]  Klaus H. Maier-Hein,et al.  No New-Net , 2018, 1809.10483.

[14]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[15]  Thomas Brox,et al.  3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation , 2016, MICCAI.

[16]  Tanveer F. Syeda-Mahmood,et al.  3D Segmentation with Exponential Logarithmic Loss for Highly Unbalanced Object Sizes , 2018, MICCAI.

[17]  Sébastien Ourselin,et al.  Automatic Brain Tumor Segmentation Based on Cascaded Convolutional Neural Networks With Uncertainty Estimation , 2019, Front. Comput. Neurosci..

[18]  Sébastien Ourselin,et al.  Automatic Brain Tumor Segmentation Using Cascaded Anisotropic Convolutional Neural Networks , 2017, BrainLes@MICCAI.

[19]  Hao Chen,et al.  Robust Multimodal Brain Tumor Segmentation via Feature Disentanglement and Gated Fusion , 2019, MICCAI.

[20]  P. Kleihues,et al.  Population-based studies on incidence, survival rates, and genetic alterations in astrocytic and oligodendroglial gliomas. , 2005, Journal of neuropathology and experimental neurology.

[21]  Andriy Myronenko,et al.  3D MRI brain tumor segmentation using autoencoder regularization , 2018, BrainLes@MICCAI.

[22]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[23]  Abhinav Gupta,et al.  Non-local Neural Networks , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[24]  Brian B. Avants,et al.  The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) , 2015, IEEE Transactions on Medical Imaging.

[25]  Gang Sun,et al.  Squeeze-and-Excitation Networks , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[26]  Konstantinos Kamnitsas,et al.  Efficient multi‐scale 3D CNN with fully connected CRF for accurate brain lesion segmentation , 2016, Medical Image Anal..