Automatic primary gross tumor volume segmentation for nasopharyngeal carcinoma using ResSE-UNet

Zhihao Jin, Xue Chen Li, Linlin Shen, Jinyi Lang, Jie Li, Junxiang Wu, Peng Xu, Jiang Duan

Research output: Chapter in Book/Conference proceedingConference contributionpeer-review

6 Citations (Scopus)

Abstract

Nasopharyngeal carcinoma (NPC) is an endemic disease within specific regions in the world. Radiotherapy is the standard treatment for NPC and accurate segmentation of primary gross tumor volume (GTV) is a critical process of continue therapy. In this paper we proposed a ResSE-UNet network and a Ternary Cross-Entropy (TCE) loss function for delineation of GTV. ResSE-UNet employed ResSE blocks to replace convolutional blocks in the original UNet to extract better features, and reduced the number of down-sampling processing to keep relatively high resolution of the images. TCE combined dice loss and Binary cross-entropy loss for larger gradient and better stability in training. The experimental results showed that among all combinations of networks and loss functions, the ResSE-UNet with TCE loss achieved the best segmentation performance, i.e. about 0.84 DSC can be obtained.

Original languageEnglish
Title of host publicationProceedings - 2020 IEEE 33rd International Symposium on Computer-Based Medical Systems, CBMS 2020
EditorsAlba Garcia Seco de Herrera, Alejandro Rodriguez Gonzalez, KC Santosh, Zelalem Temesgen, Bridget Kane, Paolo Soda
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages585-590
Number of pages6
ISBN (Electronic)9781728194295
DOIs
Publication statusPublished - Jul 2020
Externally publishedYes
Event33rd IEEE International Symposium on Computer-Based Medical Systems, CBMS 2020 - Virtual, Online, United States
Duration: 28 Jul 202030 Jul 2020

Publication series

NameProceedings - IEEE Symposium on Computer-Based Medical Systems
Volume2020-July
ISSN (Print)1063-7125

Conference

Conference33rd IEEE International Symposium on Computer-Based Medical Systems, CBMS 2020
Country/TerritoryUnited States
CityVirtual, Online
Period28/07/2030/07/20

Keywords

  • GTV segmentation
  • Nasopharyngeal carcinoma
  • ResSE-UNet
  • Ternary cross-entropy

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'Automatic primary gross tumor volume segmentation for nasopharyngeal carcinoma using ResSE-UNet'. Together they form a unique fingerprint.

Cite this