Coarse-to-Fine Stacked Fully Convolutional Nets for lymph node segmentation in ultrasound images

Yizhe Zhang, Tin Cheung Ying, Lin Yang, Anil T. Ahuja, Danny Z. Chen

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

27 Citations (Scopus)

Abstract

Ultrasound as a well-established imaging modality is widely used in imaging lymph nodes for clinical diagnosis and disease analysis. Quantitative analysis of lymph node features, morphology, and relations can provide valuable information for diagnosis and immune system studies. For such analysis, it is necessary to first accurately segment the lymph node areas in ultrasound images. In this paper, we develop a new deep learning method, called Coarse-to-Fine Stacked Fully Convolutional Nets (CFS-FCN), for automatically segmenting lymph nodes in ultrasound images. Our method consists of multiple stages of FCN modules. We train the CFS-FCN model to learn the segmentation knowledge from a coarse-to-fine, simple-to-complex manner. A data set of 80 ultrasound images containing both normal and diseased lymph nodes is used in our experiments, which show that our method considerably outperforms the state-of-the-art deep learning methods for lymph node segmentation.
Original languageEnglish
Title of host publicationProceedings - 2016 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2016
PublisherIEEE
Pages443-448
Number of pages6
ISBN (Electronic)9781509016105
DOIs
Publication statusPublished - 17 Jan 2017
Event2016 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2016 - Shenzhen, China
Duration: 15 Dec 201618 Dec 2016

Conference

Conference2016 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2016
CountryChina
CityShenzhen
Period15/12/1618/12/16

ASJC Scopus subject areas

  • Genetics
  • Medicine (miscellaneous)
  • Genetics(clinical)
  • Biochemistry, medical
  • Biochemistry
  • Molecular Medicine
  • Health Informatics

Cite this