Please use this identifier to cite or link to this item:
http://hdl.handle.net/20.500.11861/6906
Title: | Mu-net: Multi-path upsampling convolution network for medical image segmentation |
Authors: | Chen, Jia He, Zhiqiang Zhu, Dayong Hui, Bei Dr. LI Yi Man, Rita Yue, Xiao-Guang |
Issue Date: | 2022 |
Source: | CMES - Computer Modeling in Engineering and Sciences, 2022, vol. 130(3), pp. 73-95. |
Journal: | CMES - Computer Modeling in Engineering and Sciences |
Abstract: | Medical image segmentation plays an important role in clinical diagnosis, quantitative analysis, and treatment process. Since 2015, U-Net-based approaches have been widely used for medical image segmentation. The purpose of the U-Net expansive path is to map low-resolution encoder feature maps to full input resolution feature maps. However, the consecutive deconvolution and convolutional operations in the expansive path lead to the loss of some high-level information.More high-level information can make the segmentation more accurate. In this paper, we propose MU-Net, a novel, multi-path upsampling convolution network to retain more high-level information. The MU-Net mainly consists of three parts: contracting path, skip connection, and multi-expansive paths. The proposed MU-Net architecture is evaluated based on three different medical imaging datasets. Our experiments show that MU-Net improves the segmentation performance of U-Net-based methods on different datasets. At the same time, the computational efficiency is significantly improved by reducing the number of parameters by more than half. |
Type: | Peer Reviewed Journal Article |
URI: | http://hdl.handle.net/20.500.11861/6906 |
ISSN: | 1526-1492 1526-1506 |
DOI: | 10.32604/cmes.2022.018565 |
Appears in Collections: | Economics and Finance - Publication |
Find@HKSYU Show full item record
SCOPUSTM
Citations
9
checked on Jan 3, 2024
Page view(s)
171
checked on Jan 3, 2024
Google ScholarTM
Impact Indices
Altmetric
PlumX
Metrics
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.