Advanced Hybrid Information Processing. First International Conference, ADHIP 2017, Harbin, China, July 17–18, 2017, Proceedings

Research Article

Automated Segmentation of Carotid Artery Vessel Wall in MRI

  • @INPROCEEDINGS{10.1007/978-3-319-73317-3_33,
        author={Bo Wang and Gang Sha and Pengju Yin and Xia Liu},
        title={Automated Segmentation of Carotid Artery Vessel Wall in MRI},
        proceedings={Advanced Hybrid Information Processing. First International Conference, ADHIP 2017, Harbin, China, July 17--18, 2017, Proceedings},
        proceedings_a={ADHIP},
        year={2018},
        month={2},
        keywords={Medical image segmentation Carotid artery MRI Ellipse fitting Fuzzy C-Means},
        doi={10.1007/978-3-319-73317-3_33}
    }
    
  • Bo Wang
    Gang Sha
    Pengju Yin
    Xia Liu
    Year: 2018
    Automated Segmentation of Carotid Artery Vessel Wall in MRI
    ADHIP
    Springer
    DOI: 10.1007/978-3-319-73317-3_33
Bo Wang1,*, Gang Sha2,*, Pengju Yin,*, Xia Liu1,*
  • 1: Harbin University of Science and Technology
  • 2: Northwestern Polytechnical University
*Contact email: hust_wb@hrbust.edu.cn, shagang@mail.nwpu.edu.cn, pengju.yin@outlook.com, liuxia@hrbust.edu.cn

Abstract

Automatic or semi-automatic segmentation of carotid artery wall in MRI is an important means of early detection of atherosclerosis. In this paper, a new algorithm is proposed for the automated segmentation of the lumen, outer boundary and plaque contours in carotid MR images. It uses the ellipse fitting to detect the outer wall boundaries. By using the outer wall boundaries as the constraint condition, the lumen is detected using an improved fuzzy C-Means (FCM). The plaque is located by obtaining the area changing of lumen. The experimental results show that our method achieves 95.7% of region overlaps when compared to the gold standard results. This new automated method can enhance reproducibility of the quantification of vessel wall dimensions in clinical studies.