Meeting Banner
Abstract #1748

Average probabilistic brain atlases for post-mortem newborn and fetal populations and application to tissue segmentation

Eliza Orasanu 1 , Andrew Melbourne 1 , M. Jorge Cardoso 1 , Marc Modat 1 , Andrew M. Taylor 2 , Sudhin Thayyil 3 , and Sebastien Ourselin 1

1 Centre of Medical Image Computing, University College London, London, United Kingdom, 2 Centre for Cardiovascular Imaging, Institute of Cardiovascular Science, University College London, London, United Kingdom, 3 University College Hospital, London, United Kingdom

Segmentation of the fetal and neonatal brain magnetic resonance (MR) imaging is useful for understanding both normal and abnormal brain development, however it is challenging due to post-mortem artefacts and changes in T1 and T2 tissue values after death. In this paper we create average probabilistic brain atlases for newborn and fetus populations and use them for the automatic segmentation of further subjects with similar morphology from the same study. We compare them with manual segmentations, obtaining good agreement. This paper is the first to successfully generate post-mortem brain atlases from MR images of neonates and fetuses fully automatically.

This abstract and the presentation materials are available to members only; a login is required.

Join Here