Automatic atlas-based segmentation of brain white matter in neonates at risk for neurodevelopmental disorders

Book chapter (2017)

Authors

Lúcia Fonseca Eindhoven University of Technology
C. van Pul Maxima Medical Center, Veldhoven, Eindhoven University of Technology
N. Lori University of Minho, LANEN, INECO, INCYT (Favaloro-CONICET)
R van den Boom Eindhoven University of Technology
P Andriessen Neonatology Maxima Medical Center
J. Buijs Neonatology Maxima Medical Center
A. Vilanova Bartroli Computer Graphics and Visualisation - EEMCS , Eindhoven University of Technology
Research Group
Computer Graphics and Visualisation (EEMCS) (TU Delft)
More Info
expand_more

Published Date

2017

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Electrical Engineering, Mathematics and Computer Science

Department

Intelligent Systems

Research Group

Computer Graphics and Visualisation

Abstract

Very preterm infants, < 32 weeks gestation, are at high risk for brain injury. Cognitive deficits are often diagnosed at a later stage, since there are no available predictive biomarkers in the neonatal period. The maturation of specific white matter (WM) brain structures is considered a promising early-stage biomarker. With Diffusion Tensor Imaging (DTI) tractography, an in vivo and non-invasive evaluation of these anatomical structures is possible. We developed an automatic tractography segmentation pipeline, which allows for maturation assessment of the different segmented WM structures. Our segmentation pipeline is atlas-based, specifically designed for premature neonates at term equivalent age. In order to better make use of global information from tractography, all processing is done in the fiber domain. Segmented fiber bundles are further automatically quantified with respect to volume and anisotropy. Of the 24 automatically segmented neonatal tractographies, only three contained more than 30% mislabeled fibers. Results show no dependency to WM pathology. By automatically segmenting WM, we reduced the user-dependency and bias characteristic of manual methods. This study assesses the structure of the neonatal brain based on an automatic WM segmentation in the fiber domain method using DTI tractography data.