Batlas: variational reconstruction of a digital, three-dimensional atlas of the big brown bat (Eptesicus fuscus)
Embargo until
Date
2019-10-24
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Johns Hopkins University
Abstract
Bats are an ideal model for auditory perception due to their highly discriminative and dynamic sound processing capabilities. To understand auditory brain function, neuroscientists must spatially resolve complementary neuroimaging, neurophysiological, and -omic data from studies of bat auditory perception. One way to accomplish this goal is through the use of a standardized, digital, three-dimensional brain atlas. However, there is no such atlas available at present. To address this need, this thesis describes the first digital, three-dimensional, brain atlas of \textit{Eptesicus fuscus}. We present a variational method for three-dimensional reconstruction of surface volumes from sparsely-spaced (150 $\mu$m), serial, two-dimensional atlas segmentations in the absence of an external same-subject volume reference. Using a neurohistological atlas of \textit{Eptesicus fuscus} as an input, brain reconstruction takes place in four stages: (i) segmentation and symmetrization, (ii) restacking of the binary images section-by-section via in-plane rigid transformations, (iii) pair-wise diffeomorphic registration of the sections, conditioned on the restacked slices, and (iv) generation of topologically-correct surface meshes using a restricted Delaunay triangulation scheme. We define Sobolev and total variation priors on image smoothness, which control the derivatives of the images, to regularize (i.e. reduce complexity by removing unreasonable parameter choices from) the high-dimensional parameter space prescribed by the rigid motion dimensions and the diffeomorphism dimensions. We show that the quality of rigid slice alignment brought by introducing a Sobolev prior on the image intensity of a phantom and the bat brain data is superior to that of the total variation priors. We also provide volume and surface area measurements and centroid locations in atlas space of each of the brain structures.
Description
Keywords
Eptesicus fuscus, histology, brain atlas, rigid registration, nonlinear registration, LDDMM