Breast tissue segmentation on MR images using KFCM with spatial constraints

Hong Song, Qian Zhang, Feifei Sun, Jiandong Wang, Quansheng Wang, Jingdan Qiu, Deqiang Kou

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

3 Citations (Scopus)

Abstract

Accurate segmentation of breast on MR images is an essential and crucial step for computer-aided breast disease diagnosis and surgical planning. In this paper, an effective approach is proposed for segmenting the breast image into different regions, each corresponding to a different tissue. The segmentation work flow comprises two key steps. Firstly, we use the threshold-based method and morphological operations to determine the breast-air boundary and breast-chest wall so that the breast region can be extracted. Then a kernelled fuzzy C-means algorithm with spatial information (SKFCM) is used to separate the fibroglandular tissues from the fat. The proposed method is used to segment the clinical breast MR images. Experimental results have been shown visually and achieve reasonable consistency. The SKFCM method is appropriate for the problem of breast tissue segmentation.

Original languageEnglish
Title of host publicationProceedings - 2014 IEEE International Conference on Granular Computing, GrC 2014
EditorsYasuo Kudo, Shusaku Tsumoto
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages254-258
Number of pages5
ISBN (Electronic)9781479954643
DOIs
Publication statusPublished - 11 Dec 2014
Event2014 IEEE International Conference on Granular Computing, GrC 2014 - Hokkaido, Japan
Duration: 22 Oct 201424 Oct 2014

Publication series

NameProceedings - 2014 IEEE International Conference on Granular Computing, GrC 2014

Conference

Conference2014 IEEE International Conference on Granular Computing, GrC 2014
Country/TerritoryJapan
CityHokkaido
Period22/10/1424/10/14

Keywords

  • Breast MRI
  • Breast tissue segmentation
  • SKFCM

Fingerprint

Dive into the research topics of 'Breast tissue segmentation on MR images using KFCM with spatial constraints'. Together they form a unique fingerprint.

Cite this