TY - GEN
T1 - Voxel deconvolutional networks for 3D brain image labeling
AU - Chen, Yongjun
AU - Shi, Min
AU - Gao, Hongyang
AU - Shen, Dinggang
AU - Cai, Lei
AU - Ji, Shuiwang
N1 - Publisher Copyright:
© 2018 Association for Computing Machinery.
PY - 2018/7/19
Y1 - 2018/7/19
N2 - Deep learning methods have shown great success in pixel-wise prediction tasks. One of the most popular methods employs an encoder-decoder network in which deconvolutional layers are used for up-sampling feature maps. However, a key limitation of the deconvolutional layer is that it suffers from the checkerboard artifact problem, which harms the prediction accuracy. This is caused by the independency among adjacent pixels on the output feature maps. Previous work only solved the checkerboard artifact issue of deconvolutional layers in the 2D space. Since the number of intermediate feature maps needed to generate a deconvolutional layer grows exponentially with dimensionality, it is more challenging to solve this issue in higher dimensions. In this work, we propose the voxel deconvolutional layer (VoxelDCL) to solve the checkerboard artifact problem of deconvolutional layers in 3D space. We also provide an efficient approach to implement VoxelDCL. To demonstrate the effectiveness of VoxelDCL, we build four variations of voxel deconvolutional networks (VoxelDCN) based on the U-Net architecture with VoxelDCL. We apply our networks to address volumetric brain images labeling tasks using the ADNI and Loni LPBA40 datasets. The experimental results show that the proposed iVoxelDCNa achieves improved performance in all experiments. It reaches 83.34% in terms of dice ratio on the ADNI dataset and 79.12% on the Loni LPBA40 dataset, which increases 1.39% and 2.21% respectively compared with the baseline. In addition, all the variations of VoxelDCN we proposed outperform the baseline methods on the above datasets, which demonstrates the effectiveness of our methods.
AB - Deep learning methods have shown great success in pixel-wise prediction tasks. One of the most popular methods employs an encoder-decoder network in which deconvolutional layers are used for up-sampling feature maps. However, a key limitation of the deconvolutional layer is that it suffers from the checkerboard artifact problem, which harms the prediction accuracy. This is caused by the independency among adjacent pixels on the output feature maps. Previous work only solved the checkerboard artifact issue of deconvolutional layers in the 2D space. Since the number of intermediate feature maps needed to generate a deconvolutional layer grows exponentially with dimensionality, it is more challenging to solve this issue in higher dimensions. In this work, we propose the voxel deconvolutional layer (VoxelDCL) to solve the checkerboard artifact problem of deconvolutional layers in 3D space. We also provide an efficient approach to implement VoxelDCL. To demonstrate the effectiveness of VoxelDCL, we build four variations of voxel deconvolutional networks (VoxelDCN) based on the U-Net architecture with VoxelDCL. We apply our networks to address volumetric brain images labeling tasks using the ADNI and Loni LPBA40 datasets. The experimental results show that the proposed iVoxelDCNa achieves improved performance in all experiments. It reaches 83.34% in terms of dice ratio on the ADNI dataset and 79.12% on the Loni LPBA40 dataset, which increases 1.39% and 2.21% respectively compared with the baseline. In addition, all the variations of VoxelDCN we proposed outperform the baseline methods on the above datasets, which demonstrates the effectiveness of our methods.
KW - Deep learning
KW - Volumetric brain image labeling
KW - Voxel deconvolutional layer
KW - Voxel deconvolutional networks
UR - http://www.scopus.com/inward/record.url?scp=85051533356&partnerID=8YFLogxK
U2 - 10.1145/3219819.3219974
DO - 10.1145/3219819.3219974
M3 - Conference contribution
AN - SCOPUS:85051533356
SN - 9781450355520
T3 - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
SP - 1226
EP - 1234
BT - KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
PB - Association for Computing Machinery
T2 - 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2018
Y2 - 19 August 2018 through 23 August 2018
ER -