A multimodal computational pipeline for 3D histology of the human brain

Abstract

Ex vivo imaging enables analysis of the human brain at a level of detail that is not possible in vivo with MRI. In particular, histology can be used to study brain tissue at the microscopic level, using a wide array of different stains that highlight different microanatomical features. Complementing MRI with histology has important applications in ex vivo atlas building and in modeling the link between microstructure and macroscopic MR signal. However, histology requires sectioning tissue, hence distorting its 3D structure, particularly in larger human samples. Here, we present an open-source computational pipeline to produce 3D consistent histology reconstructions of the human brain. The pipeline relies on a volumetric MRI scan that serves as undistorted reference, and on an intermediate imaging modality (blockface photography) that bridges the gap between MRI and histology. We present results on 3D histology reconstruction of a whole human hemisphere.1/26 33 1. Cutting and sectioning of tissue for histology introduces distortions (stretching, tearing, 34 folding, cracking, see [6]) that are specific to each section. Therefore, an external volumetric 35 reference (typically an MRI scan) is required to produce an unbiased registration. Without 36 such a reference, naïve pairwise registration leads to accumulation of errors ("z-shift") and 37 spurious straightening of curved structures (the so-called "banana effect", [7]). 382. Large differences in contrast and resolution between MRI and histology, combined with 39 potential inhomogeneous staining and the aforementioned sectioning artifacts, make the 40 alignment of these two modalities a difficult inter-modality registration problem. 413. The large size of the human brain, compared with most animal models, requires cutting 42 the tissue in blocks for whole-brain analyses, with the resulting need to reassemble the 43 blocks [6], which is a part-to-whole registration problem [8,9]. This problem can be solved 44 with whole brain microtomes, although such microtomes are only available in few selected 45 sites around the world. Moreover, they exacerbate the sectioning artifacts, due to the much 46 larger surface area of the sections. 47Even when a reference MRI is available, 3D histology reconstruction is an ill-posed problem, 48 as errors in the -typically linear -registration between the MRI and the histological stack 49 2/26

3D reconstruction of ultra-high resolution neurotransmitter receptor atlases in human and non-human primate brains

“…A recent and particularly innovative approach to 2D alignment of histological and corresponding MRI sections involved using Bayesian methods to simultaneously align the histological section and the MRI section while transforming the pixel intensities of the former to resemble the latter 25,26 .…”