TY - GEN
T1 - Deep neural network for pancreas segmentation from CT images
AU - Chen, Zhanlan
AU - Zheng, Jiangbin
N1 - Publisher Copyright:
© Springer Nature Switzerland AG 2020.
PY - 2020
Y1 - 2020
N2 - Automatic pancreas segmentation from Computed Tomography (CT) images is a prerequisite of clinical practices such as cancer detection, yet challenging due to the variability in shape. To address this challenge, we propose a Hierarchical Convolutional Neural Network (H-CNN) to fuse multi-scale features, which could remedy the lost image details in progressive convolutional and pooling layers. In our proposed H-CNN, a hierarchical fusion block is designed to fuse low-level and high-level features across different layers. The H-CNN is evaluated on NIH pancreas dataset and outperforms the current state-of-art methods by achieving 86.59% ± 4.33% in terms of DSC. The experimental results confirm the effectiveness of the proposed H-CNN.
AB - Automatic pancreas segmentation from Computed Tomography (CT) images is a prerequisite of clinical practices such as cancer detection, yet challenging due to the variability in shape. To address this challenge, we propose a Hierarchical Convolutional Neural Network (H-CNN) to fuse multi-scale features, which could remedy the lost image details in progressive convolutional and pooling layers. In our proposed H-CNN, a hierarchical fusion block is designed to fuse low-level and high-level features across different layers. The H-CNN is evaluated on NIH pancreas dataset and outperforms the current state-of-art methods by achieving 86.59% ± 4.33% in terms of DSC. The experimental results confirm the effectiveness of the proposed H-CNN.
KW - Convolutional Neural Network
KW - Medical image segmentation
KW - Pancreas segmentation
UR - http://www.scopus.com/inward/record.url?scp=85080956657&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-39431-8_39
DO - 10.1007/978-3-030-39431-8_39
M3 - 会议稿件
AN - SCOPUS:85080956657
SN - 9783030394301
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 406
EP - 413
BT - Advances in Brain Inspired Cognitive Systems - 10th International Conference, BICS 2019, Proceedings
A2 - Ren, Jinchang
A2 - Hussain, Amir
A2 - Zhao, Huimin
A2 - Cai, Jun
A2 - Chen, Rongjun
A2 - Xiao, Yinyin
A2 - Huang, Kaizhu
A2 - Zheng, Jiangbin
PB - Springer
T2 - 10th International Conference on Brain Inspired Cognitive Systems, BICS 2019
Y2 - 13 July 2019 through 14 July 2019
ER -
