Three-Dimensional Digital Reconstruction of the Cerebellar Cortex: Lobule Thickness, Surface Area Measurements, and Layer Architecture

Abstract

The cerebellum is ontogenetically one of the first structures to develop in the central nervous system; nevertheless, it has been only recently reconsidered for its significant neurobiological, functional, and clinical relevance in humans. Thus, it has been a relatively under-studied compared to the cerebrum. Currently, non-invasive imaging modalities can barely reach the necessary resolution to unfold its entire, convoluted surface, while only histological analyses can reveal local information at the micrometer scale. Herein, we used the BigBrain dataset to generate area and point-wise thickness measurements for all layers of the cerebellar cortex and for each lobule in particular. We found that the overall surface area of the cerebellar granular layer (including Purkinje cells) was 1,732 cm² and the molecular layer was 1,945 cm². The average thickness of the granular layer is 0.88 mm (± 0.83) and that of the molecular layer is 0.32 mm (± 0.08). The cerebellum (both granular and molecular layers) is thicker at the depth of the sulci and thinner at the crowns of the gyri. Globally, the granular layer is thicker in the lateral-posterior-inferior region than the medial-superior regions. The characterization of individual layers in the cerebellum achieved herein represents a stepping-stone for investigations interrelating structural and functional connectivity with cerebellar architectonics using neuroimaging, which is a matter of considerable relevance in basic and clinical neuroscience. Furthermore, these data provide templates for the construction of cerebellar topographic maps and the precise localization of structural and functional alterations in diseases affecting the cerebellum.

Keywords: Cerebellar cortical layers measurements; Cerebellar cortical layers morphometry; Cerebellum; Laminar thickness measurements; Lobules; Surface area.

Fig. 1

A comparison of the cerebral gyrus and the cerebellar lobules under the same scale. A. Cerebral folding. B. Cerebellum folding. The width of a typical fold in the cerebrum is 15 mm, while that of a typical fold in the cerebellum (a folium) is 0.9 mm

Fig. 2

The U-net architecture. The U-net architecture comprised convolutional encoding and decoding units that took the cerebellar image as the input and produced the molecular layer surface map

Fig. 3

Segmentation of cerebellar layers. The white matter (WM) and granular layer were segmented by an interactive method [42] and the pial surface was segmented by automatic deep learning-based method [43], see “Materials and Methods” section for detail. A. Coronal, high-resolution histological sections of the BigBrain cerebellum. B. Segmentation of the cerebellum shown in one exemplar slice. The WM penetrates into each folium (arrow a), whereas the WM in the outer folia disappears (arrow b). C. A three-dimensional reconstruction of the BigBrain cerebellar surface. This mesh contains 159,923,340 vertexes and 319,946,592 triangles. D. Local variations in cerebellar thickness. The granular layer is thicker at the crown of the gyrus (arrow d) and thinner in the depth of a sulcus (arrow c). The molecular layer is lost in certain out-facing regions (arrow e), while the folded surface remains complete (arrow f), which induces some thickness calculation artifacts

Fig. 4

The thickness of the cerebellar cortex and cortical layers. In all the three rows, downward points to the posterior or caudal of the subject. A. A superior (left) and inferior (middle) view of the cerebellar cortical thickness mapped onto the pial surface and a histogram of the cortical thickness distribution (right) are depicted. B. The thickness of the granular layer (including the Purkinje cell layer) and a histogram of the thickness distribution are shown. C. The thickness of the molecular layer mapped onto the Purkinje cell surface and a histogram of the thickness distribution are illustrated. The spike on the histogram is close to thickness of 0 mm and corresponds to the dark blue-colored gyri on the inferior side on the left and in the middle. The dark blue-colored regions are due to the absence of the molecular layer in the original BigBrain data, possibly due to tissue damage during sample processing

Fig. 5

Segmentation and corresponding thicknesses of the cerebellar lobules. A. Segmentation of each lobule down to the individual folia resolution. B-D. Thickness histograms of the cerebellar cortex (B), granular (C), and molecular (D) layers of each lobule

Fig. 6

Three-dimensional image of a single lobule and its partial display. A. Thickness maps of the granular (left) and molecular layers (right) of lobule crus II. Arrow a indicates the tip of the folia, where the granular layer is thick, while arrow c indicates where the molecular layer is very thin. B-C. A cross-sectional view of crus II B and the magnified sub-region C. The apparent thickening of the granular layer might be due to the depth of the white matter. The thinning of the molecular layer might be due to erosion during sample preparation

Fig. 7

The local surface with thickness defined on them using the conformal spherical parameterization. A. Conformal mapping of the granular surface to the sphere. B. Conformal mapping of the molecular layer surface to the sphere