TY - CHAP
T1 - Automatic Brain Tumor Segmentation Using Cascaded Anisotropic Convolutional Neural Networks
AU - Wang, Guotai
AU - Li, Wenqi
AU - Ourselin, Sébastien
AU - Vercauteren, Tom
PY - 2018/2/17
Y1 - 2018/2/17
N2 - 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.
AB - 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.
KW - Brain tumor
KW - Convolutional neural network
KW - Segmentation
UR - http://www.scopus.com/inward/record.url?scp=85042527812&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-75238-9_16
DO - 10.1007/978-3-319-75238-9_16
M3 - Conference paper
AN - SCOPUS:85042527812
SN - 9783319752372
VL - 10670 LNCS
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 178
EP - 190
BT - Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries
PB - Springer Verlag
T2 - 3rd International Workshop on Brainlesion, BrainLes 2017 Held in Conjunction with Medical Image Computing for Computer Assisted Intervention , MICCAI 2017
Y2 - 14 September 2017 through 14 September 2017
ER -