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

facebooktwittermailshare

A MULTI-SCALED RECEPTIVE FIELD LEARNING APPROACH FOR MEDICAL IMAGE SEGMENTATION

Abstract: 

Biomedical image segmentation has been widely studied, and lots of methods have been proposed. Among these methods, attention U-Net has achieved a promising performance. However, it has drawbacks of extracting the multi-scaled receptive field features at the high-level feature maps, resulting in the degeneration when dealing with the lesions with apparent scale variations. To solve this problem, this paper integrates an atrous spatial pyramid pooling (ASPP) module in the contracting path of attention U-Net. This module employs multiple dilation rates for the purpose of obtaining several multi-scale receptive fields, which significantly improves the networks' ability of handling both large and small lesions. Evaluation experimental result shows that our approach significantly improves the performance of medical image segmentation and substantially outperforms the representative deep learning models on public datasets.

up
0 users have voted:

Comments

Biomedical image segmentation has been widely studied, and lots of methods have been proposed. Among these methods, attention U-Net has achieved a promising performance. However, it has drawbacks of extracting the multi-scaled receptive field features at the high-level feature maps, resulting in the degeneration when dealing with the lesions with apparent scale variations. To solve this problem, this paper integrates an atrous spatial pyramid pooling (ASPP) module in the contracting path of attention U-Net. This module employs multiple dilation rates for the purpose of obtaining several multi-scale receptive fields, which significantly improves the networks' ability of handling both large and small lesions. Evaluation experimental result
shows that our approach significantly improves the performance of medical image segmentation and substantially outperforms the representative deep learning models on public datasets.

Paper Details

Authors:
Pengcheng Guo Xiangdong Su Haoran Zhang Meng Wang Feilong
Submitted On:
12 February 2020 - 12:23pm
Short Link:
Type:
Poster
Event:
Presenter's Name:
Pengcheng Guo
Paper Code:
5146
Document Year:
2020
Cite

Document Files

Biomedical image segmentation

(8)

Subscribe

[1] Pengcheng Guo Xiangdong Su Haoran Zhang Meng Wang Feilong, "A MULTI-SCALED RECEPTIVE FIELD LEARNING APPROACH FOR MEDICAL IMAGE SEGMENTATION", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/4983. Accessed: Feb. 27, 2020.
@article{4983-20,
url = {http://sigport.org/4983},
author = {Pengcheng Guo Xiangdong Su Haoran Zhang Meng Wang Feilong },
publisher = {IEEE SigPort},
title = {A MULTI-SCALED RECEPTIVE FIELD LEARNING APPROACH FOR MEDICAL IMAGE SEGMENTATION},
year = {2020} }
TY - EJOUR
T1 - A MULTI-SCALED RECEPTIVE FIELD LEARNING APPROACH FOR MEDICAL IMAGE SEGMENTATION
AU - Pengcheng Guo Xiangdong Su Haoran Zhang Meng Wang Feilong
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/4983
ER -
Pengcheng Guo Xiangdong Su Haoran Zhang Meng Wang Feilong. (2020). A MULTI-SCALED RECEPTIVE FIELD LEARNING APPROACH FOR MEDICAL IMAGE SEGMENTATION. IEEE SigPort. http://sigport.org/4983
Pengcheng Guo Xiangdong Su Haoran Zhang Meng Wang Feilong, 2020. A MULTI-SCALED RECEPTIVE FIELD LEARNING APPROACH FOR MEDICAL IMAGE SEGMENTATION. Available at: http://sigport.org/4983.
Pengcheng Guo Xiangdong Su Haoran Zhang Meng Wang Feilong. (2020). "A MULTI-SCALED RECEPTIVE FIELD LEARNING APPROACH FOR MEDICAL IMAGE SEGMENTATION." Web.
1. Pengcheng Guo Xiangdong Su Haoran Zhang Meng Wang Feilong. A MULTI-SCALED RECEPTIVE FIELD LEARNING APPROACH FOR MEDICAL IMAGE SEGMENTATION [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/4983