Atlas‐based segmentation of developing tissues in the human brain with quantitative validation in young fetuses

Abstract

Imaging of the human fetus using magnetic resonance (MR) is an essential tool for quantitative studies of normal as well as abnormal brain development in utero. However, because of fundamental differences in tissue types, tissue properties and tissue distribution between the fetal and adult brain, automated tissue segmentation techniques developed for adult brain anatomy are unsuitable for this data. In this paper, we describe methodology for automatic atlas-based segmentation of individual tissue types in motion-corrected 3D volumes reconstructed from clinical MR scans of the fetal brain. To generate anatomically correct automatic segmentations, we create a set of accurate manual delineations and build an in utero 3D statistical atlas of tissue distribution incorporating developing grey and white matter as well as transient tissue types such as the germinal matrix. The probabilistic atlas is associated with an unbiased average shape and intensity template for registration of new subject images to the space of the atlas. Quantitative whole brain 3D validation of tissue labeling performed on a set of 14 fetal MR scans (20.57–22.86 weeks gestational age) demonstrates that this atlas-based EM segmentation approach achieves consistently high DSC performance for the main tissue types in the fetal brain. This work indicates that reliable measures of brain development can be automatically derived from clinical MR imaging and opens up possibility of further 3D volumetric and morphometric studies with multiple fetal subjects.

Development of the Fetal Cerebral Cortex in the Second Trimester: Assessment with 7T Postmortem MR Imaging

“…11 However, postmortem MR imaging obtained with high magnetic strength is of high quality and is sufficient for segmentation, reconstruction, and quantitative analysis. The consistency in demonstrating the cortical surface between the gross anatomy and the 3D visualization model has been proved.…”

“…9 Currently, prenatal morphometric studies have been focused as early as 17 weeks GA, 10 but only the linear biometric values were obtained on 2D images without 3D reconstruction of the fetal brain. In addition, the existing MR imaging postprocess-ing software with automated tissue segmentation widely used for adults is not suitable for the fetal brain, 11 and thus we still lack 3D reliable and consistent quantitative measurements of the fetal brain at the early developmental phase.…”

A Combined Manifold Learning Analysis of Shape and Appearance to Characterize Neonatal Brain Development

“…Good examples are provided by Habas et al [28] and Murgasova et al [29] for fetal and neonatal data respectively. Adaptations of segmentation techniques to neonatal and fetal data have also been developed in the contexts of longitudinal analysis of neonatal and fetal data [30], [31] and of multi-region atlas-based segmentation [10]. Models for the patterns of MR signal change in neonatal tissue have also been developed [32], [33].…”

Automated injury segmentation to assist in the treatment of children with cerebral palsy

“…Robustness to parameter initialisation can be achieved by interleaving the EM method with Markov Random Field (MRF) models [Kindermann and Snell, 1980] to incorporate spatial information [Ashburner and Friston, 2005;Bricq et al, 2008b;Chai et al, 2015;Habas et al, 2010;Marroquin et al, 2002;Melbourne et al, 2012;Murgasova et al, 2006;Zhang et al, 2001]. Markov modelling has been performed for tissue segmentation without the EM assumptions, instead using non-parametric estimates of neighbourhood tissue distributions [Awate et al, 2006].…”

“…The VBM approach, however, has been found to be susceptible to false positive findings, particularly in the neocortex [Scarpazza et al, 2015]. Hüppi et al, 1998;Inder et al, 1999;Warfield et al, 2000;Cocosco et al, 2003;Vrooman et al, 2007 K- Van Leemput et al, 1999;Wells et al, 1996;Weisenfeld and Warfield, 2009;Prastawa et al, 2005;Xue et al, 2007;Ashburner and Friston, 2005;Bricq et al, 2008;Habas et al, 2010;Marroquin et al, 2002;Melbourne et al, 2012;Murgasova et al, 2006;Zhang et al, 2001;Greenspan et al, 2006;Chai et al, 2015 Djamanakova et al, 2013;Ciofolo et al, 2009;Klein and Hirsch, 2005;Faria et al, 2011;Yoshida et al, 2013 Label propagation Sabuncu et al, 2010;Artaechevarria et al, 2009;Heckermann et al, 2006;Aljabar et al, 2009;Wolz et al, 2010;Lötjönen et al, 2010;Rajchl et al, 2015;Ledig et al, 2015;Oishi et al, 20...…”