Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Lightweight V-Net for Liver Segmentation

Abstract: 

The V-Net based 3D fully convolutional neural networks have been widely used in liver volumetric data segmentation. However, due to the large number of parameters of these networks, 3D FCNs suffer from high computational cost and GPU memory usage. To address these issues, we design a lightweight V-Net (LV-Net) for liver segmentation in this paper. The proposed network makes two contributions. The first is that we design an inverted residual bottleneck block (IRB block) and a 3D average pooling block and apply them to the proposed LV-Net. Compared with vanilla convolution, depth-wise convolution and point-wise convolution employed by the IRB block can not only reduce the number of parameters significantly, but also extract features sufficiently well by decoupling cross-channel corrections and spatial correlations. The second is that the LV-Net employs 3D deep supervision to improve the final loss function in training phase, which makes
the proposed LV-Net acquire a more powerful discrimination capability between liver areas and non-liver areas. The proposed
LV-Net is evaluated on public LiTS [1] dataset, and experiments demonstrate that the proposed LV-Net is superior to popular 2D and 3D networks in terms of segmentation performance, parameter quantity and computational cost.

up
0 users have voted:

Paper Details

Authors:
Wenzheng Zhou, Yuxiao Zhang,Risheng Wang, Hongying Meng, asoke K. Nandi
Submitted On:
6 February 2020 - 5:47am
Short Link:
Type:
Poster
Event:
Document Year:
2020
Cite

Document Files

main.pdf

(13)

Keywords

Additional Categories

Subscribe

[1] Wenzheng Zhou, Yuxiao Zhang,Risheng Wang, Hongying Meng, asoke K. Nandi, "Lightweight V-Net for Liver Segmentation", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/4972. Accessed: Feb. 27, 2020.
@article{4972-20,
url = {http://sigport.org/4972},
author = {Wenzheng Zhou; Yuxiao Zhang;Risheng Wang; Hongying Meng; asoke K. Nandi },
publisher = {IEEE SigPort},
title = {Lightweight V-Net for Liver Segmentation},
year = {2020} }
TY - EJOUR
T1 - Lightweight V-Net for Liver Segmentation
AU - Wenzheng Zhou; Yuxiao Zhang;Risheng Wang; Hongying Meng; asoke K. Nandi
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/4972
ER -
Wenzheng Zhou, Yuxiao Zhang,Risheng Wang, Hongying Meng, asoke K. Nandi. (2020). Lightweight V-Net for Liver Segmentation. IEEE SigPort. http://sigport.org/4972
Wenzheng Zhou, Yuxiao Zhang,Risheng Wang, Hongying Meng, asoke K. Nandi, 2020. Lightweight V-Net for Liver Segmentation. Available at: http://sigport.org/4972.
Wenzheng Zhou, Yuxiao Zhang,Risheng Wang, Hongying Meng, asoke K. Nandi. (2020). "Lightweight V-Net for Liver Segmentation." Web.
1. Wenzheng Zhou, Yuxiao Zhang,Risheng Wang, Hongying Meng, asoke K. Nandi. Lightweight V-Net for Liver Segmentation [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/4972