Post mortem mapping of connectional anatomy for the validation of diffusion MRI

Abstract

Despite the impressive advances in diffusion MRI (dMRI) acquisition and analysis that have taken place during the Human Connectome era, dMRI tractography is still an imperfect source of information on the circuitry of the brain. In this review, we discuss methods for post mortem validation of dMRI tractography, fiber orientations, and other microstructural properties of axon bundles that are typically extracted from dMRI data. These methods include anatomic tracer studies, Klingler's dissection, myelin stains, label-free optical imaging techniques, and others. We provide an overview of the basic principles of each technique, its limitations, and what it has taught us so far about the accuracy of different dMRI acquisition and analysis approaches.

High-fidelity approximation of grid- and shell-based sampling schemes from undersampled DSI using compressed sensing: Post mortem validation

“…This suggests that our in-plane angular errors are not biased by the presence of through-plane diffusion. Finally, the 2D angular errors in our PSOCT studies agree with both the 2D and 3D angular errors reported in studies that used histological staining, further supporting the validity of in-plane angular errors as a measure of accuracy (Anastasia Yendiki et al 2021).…”

“…There are several other techniques that have been used to assess the accuracy of dMRI orientation estimates. For a comprehensive discussion of their relative merits and weaknesses, we refer the reader to a recent review (Anastasia Yendiki et al 2021). One approach is to extract orientations from myelin-stained sections (Leergaard et al 2010;Choe et al 2012;K.…”

“…Thus, from a single scan, we can generate data with multiple q-space sampling schemes and perform systematic, side-by-side comparisons, as we did most recently in the IronTract Challenge (Maffei et al 2020, Maffei et al 2021. Given that long acquisition times are one of the main challenges in the post mortem validation of dMRI (Yendiki et al 2021), any opportunity to further reduce the scan time would be beneficial, as it would allow us to scan more samples or to scan the same samples at higher spatial resolution in the same amount of time.…”

“…Specifically, we image human white-matter samples with high-resolution DSI at 9.4T and with polarization-sensitive optical coherence tomography (PSOCT) (De Boer et al 1997). The latter is a technique in the family of label-free 3D optical imaging methods, which provide direct measurements of fiber orientations at microscopic resolution (Wang et al 2018;Wang et al 2011) and have emerged in recent years as a powerful tool for the validation of the mesoscopic-resolution diffusion orientations obtained from dMRI (see Yendiki et al, (2021) for a review). We have recently demonstrated the ability of PSOCT to resolve complex fiber configurations, including interdigitated crossing fibers, distinct (non-interdigitated) crossing fibers, and branching fibers, in human brain tissue (Jones et al 2020).…”

High-fidelity approximation of grid- and shell-based sampling schemes from undersampled DSI using compressed sensing: Post mortem validation

“…Given that long acquisition times are one of the main challenges in the post mortem validation of dMRI ( Yendiki et al, 2021 ), any opportunity to further reduce the scan time would be beneficial, as it would allow us to scan more samples or to scan the same samples at higher spatial resolution in the same amount of time. Compressed sensing (CS) has been proposed as an approach to accelerated DSI (CS-DSI).…”