Unfolding the hippocampus: An intrinsic coordinate system for subfield segmentations and quantitative mapping

Abstract

The hippocampus, like the neocortex, has a morphological structure that is complex and variable in its folding pattern, especially in the hippocampal head. The current study presents a computational method to unfold hippocampal grey matter, with a particular focus on the hippocampal head where complexity is highest due to medial curving of the structure and the variable presence of digitations. This unfolding was performed on segmentations from high-resolution, T2-weighted 7T MRI data from 12 healthy participants and one surgical patient with epilepsy whose resected hippocampal tissue was used for histological validation. We traced a critical image feature composed of the hippocampal sulcus and stratum radiatum lacunosum-moleculare, (SRLM) in these images, then employed user-guided semi-automated techniques to detect and subsequently unfold the surrounding hippocampal grey matter. This unfolding was performed by solving Laplace's equation in three dimensions of interest (long-axis, proximal-distal, and laminar). The resulting 'unfolded coordinate space' provides an intuitive way of mapping the hippocampal subfields in 2D space (long-axis and proximal-distal), such that similar borders can be applied in the head, body, and tail of the hippocampus independently of variability in folding. This unfolded coordinate space was employed to map intracortical myelin and thickness in relation to subfield borders, which revealed intracortical myelin differences that closely follow the subfield borders used here. Examination of a histological resected tissue sample from a patient with epilepsy reveals that our unfolded coordinate system has biological validity, and that subfield segmentations applied in this space are able to capture features not seen in manual tracing protocols.

Hippocampal subfields revealed through unfolding and unsupervised clustering of laminar and morphological features in 3D BigBrain

“…These features are illustrated at the top of the figure, but additional details can be found in the Materials and Methods section. As in related work (Duvernoy et al, 2013;DeKraker et al, 2018), thickness was highest in the subiculum and CA4 and lowest in CA2. Curvature was generally high in subiculum, which reflects its outward curling away from the rest of the hippocampus, but in CA1 vertical bands of positive and negative values can be seen which correspond to the hippocampal gyrifications seen in Figure 1.…”

“…This lack of detail would also lead to overinflated thickness measures, larger overall volumes, and perhaps differences in the proportional sizes of some subfields along the anterior-posterior extent of the hippocampus. Quantitative MR, such as T1 mapping, may additionally provide cues to approximate cytoarchitectonic features and indeed in our previous work we observed higher T1-weights in CA2 and CA3 (DeKraker et al, 2018) which may be driven by higher neuronal cell densities as observed in the current study. Thus the features described here show clear promise for characterizing or segmenting the hippocampus in future MRI work.…”

“…More recently, increasing interest has also emerged concerning graded differences along the anterior-posterior axis based on subfield composition and connectivity (Strange et al, 2014;Poppenk et al, 2013;Plachti et al, 2019). An organizational principle that shapes these dimensions, i.e., subfields and anterior-posterior differences, is the complex topology within the hippocampus that results from its ontological development (Duvernoy et al, 2013;DeKraker et al, 2018). This principle has received only limited investigation to date but requires careful consideration in any effort to characterize the internal architecture of the hippocampus.…”

“…The gyrification seen in the hippocampus is morphologically similar to gyrification in the neocortex (although not necessarily based on the same ontogeny). It has been shown to vary considerably between individuals (DeKraker et al, 2018;Chang et al, 2018) and can be correlated with age (Cai et al, 2019) or disease, such as temporal-lobe epilepsy (Blümcke et al, 2013). This folding is an important aspect of understanding the internal structure of the hippocampus, and for appreciation of the continuity of subfields, particularly in its anterior portion that includes the uncus (Ding & Van Hoesen, 2015).…”

“…This folding is an important aspect of understanding the internal structure of the hippocampus, and for appreciation of the continuity of subfields, particularly in its anterior portion that includes the uncus (Ding & Van Hoesen, 2015). Topological analyses can provide a framework for extracting these continuities, for example through unfolding (DeKraker et al, 2018), and offer the basis for laminar and further morphological characterization of complete hippocampal structure in 3D, including subject-specific gyrification.…”