Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article

Degree

Master of Engineering Science

Program

Biomedical Engineering

Supervisor

Duerden, Emma G.

Abstract

With advances in fetal magnetic resonance imaging (MRI), research in neonatal neuroscience has shifted to identify in utero brain-based biomarkers for outcome prediction in high-risk fetuses, particularly those impacted by growth restriction. Volumetric segmentation of the fetal brain can provide better understanding of the trajectories of brain development and may aid in predicting functional outcomes. The current thesis aimed to develop semi-automatic methods to target deep brain structures in the fetal brain identified on MR images in fetuses with and without growth restriction. In this study, pregnant women (35-39 weeks gestational age [n=9]) with growth appropriate (n=8) and growth restricted fetuses (n=1) were recruited. Fetal MRI was performed on 1.5 Tesla (T) and 3T MRI scanners and 2D stacks of T2-weighted images were acquired. A novel fetal whole brain segmentation algorithm developed for second trimester fetuses was applied to the T2-weighted MR images to reconstruct 2D volumes into 3D images. To segment deep brain structures, an atlas of cortical and subcortical structures was registered to the 3D reconstructed images. Linear and nonlinear registration algorithms, with two types of similarity metrics (mutual information [MI], cross-correlation [CC]), were compared to determine the optimal strategy of segmenting subcortical structures. Dice coefficients were calculated to validate the reliability of automatic methods and to compare the performance between the registration algorithms compared to manual segmentations. Comparing atlas-generated masks against manually segmented masks of the same brain structures, the median Dice-kappa coefficients for linear registration using CC performed optimally. However, post hoc analyses indicated that linear MI and CC performed comparably. Overall, this semi-automatic subcortical segmentation method for third-trimester fetal brain images provides reliable performance. This segmentation pipeline can aid in identifying early predictors of brain dysmaturation to support clinical decision making for antenatal treatment strategies and promote optimal neurodevelopment in fetuses.

Summary for Lay Audience

Each year, over 30 million pregnancies are impacted by growth restriction, which is associated with delayed brain development, and may place newborns at risk for the later development of childhood psychiatric disorders as well as movement disabilities. Brain growth is a key marker for growth restriction as well as other disorders that can influence brain development. Better tools are needed to measure the growth of the fetal brain to aid in diagnosing and predicting developmental outcomes. Magnetic resonance (MR) imaging is an emerging tool to study the developing brain in utero. Presently manual labeling of brain regions in MR images is time consuming and costly, and automated methods are needed to rapidly target brain regions for brain growth study.

This thesis aimed to develop methods to facilitate the targeting of deep brain structures in the fetal brain identified on MR images in fetuses with and without growth restriction. To achieve this aim, a platform of novel algorithms that was designed to rebuild brain images corrupted by fetal motion, was evaluated. To study the brain growth, an atlas that labeled the brain regions was applied to the fetal images using an automated algorithm. To validate the atlas, two brain regions were manually outlined on the images. We compared the manually defined regions with five methods of automated segmentation. The different methods of automated segmentation varied in terms of the methods used to identify and align the features in the atlas with fetal images as well as their computing time. An automated method that uses a dense deformable image registration, where the goal is to identify corresponding areas between the atlas and fetal images, was as appropriate as a more computationally intensive method.

This semi-automated platform can be applied to identify fetuses with delayed brain growth as well as track growth over time during the third trimester to provide an image-based marker of brain health. This method can be further validated in larger samples in fetuses impacted by growth restriction. Studying brain growth in the fetus may aid in informing medical decision making for clinicians as well as improve counseling for families.

Download

Share

COinS