Segmentation Challenge on the Quantification of Left Atrial Wall Thickness

  • Rashed Karim
  • Marta Varela
  • Pranav Bhagirath
  • Ross Morgan
  • Jonathan M. Behar
  • R. James Housden
  • Ronak Rajani
  • Oleg Aslanidi
  • Kawal S. Rhode
Conference paper

DOI: 10.1007/978-3-319-52718-5_21

Part of the Lecture Notes in Computer Science book series (LNCS, volume 10124)
Cite this paper as:
Karim R. et al. (2017) Segmentation Challenge on the Quantification of Left Atrial Wall Thickness. In: Mansi T., McLeod K., Pop M., Rhode K., Sermesant M., Young A. (eds) Statistical Atlases and Computational Models of the Heart. Imaging and Modelling Challenges. STACOM 2016. Lecture Notes in Computer Science, vol 10124. Springer, Cham

Abstract

This paper presents an image database for the Left Atrial Wall Thickness Quantification challenge at the MICCAI STACOM 2016 workshop along with some preliminary results. The image database consists of both CT (\(n=10\)) and MRI (\(n=10\)) datasets. Expert delineations from two observers were obtained for each image in the CT set and a single-observer segmentation was obtained for each image in the MRI set included in this study. Computer algorithms for segmentation of wall thickness from three research groups contributed to this challenge. The algorithms were evaluated on the basis of wall thickness measurements obtained from the segmentation masks.

Keywords

Image segmentation Left atrium CT Angiography MRI Image quantification 

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Rashed Karim
    • 1
  • Marta Varela
    • 1
  • Pranav Bhagirath
    • 3
  • Ross Morgan
    • 1
  • Jonathan M. Behar
    • 1
    • 2
  • R. James Housden
    • 1
  • Ronak Rajani
    • 2
  • Oleg Aslanidi
    • 1
  • Kawal S. Rhode
    • 1
  1. 1.Division of Imaging Sciences and Biomedical EngineeringKing’s College LondonLondonUK
  2. 2.Department of CardiologyGuy’s and St. Thomas’ NHS Foundation TrustLondonUK
  3. 3.Haga Teaching HospitalThe HagueThe Netherlands

Personalised recommendations