Improving Imaging of the Brainstem and Cerebellum in Autistic Children: Transformation-Based High-Resolution Diffusion MRI (TiDi-Fused) in the Human Brainstem

Abstract

Diffusion-weighted magnetic resonance imaging (dMRI) of the brainstem is technically challenging, especially in young autistic children as nearby tissue-air interfaces and motion (voluntary and physiological) can lead to artifacts. This limits the availability of high-resolution images, which are desirable for improving the ability to study brainstem structures. Furthermore, inherently low signal-to-noise ratios, geometric distortions, and sensitivity to motion not related to molecular diffusion have resulted in limited techniques for high-resolution data acquisition compared to other modalities such as T1-weighted imaging. Here, we implement a method for achieving increased apparent spatial resolution in pediatric dMRI that hinges on accurate geometric distortion correction and on high fidelity within subject image registration between dMRI and magnetization prepared rapid acquisition gradient echo (MPnRAGE) images. We call this post-processing pipeline T1 weighted-diffusion fused, or “TiDi-Fused”. Data used in this work consists of dMRI data (2.4 mm resolution, corrected using FSL’s Topup) and T1-weighted (T1w) MPnRAGE anatomical data (1 mm resolution) acquired from 128 autistic and non-autistic children (ages 6–10 years old). Accurate correction of geometric distortion permitted for a further increase in apparent resolution of the dMRI scan via boundary-based registration to the MPnRAGE T1w. Estimation of fiber orientation distributions and further analyses were carried out in the T1w space. Data processed with the TiDi-Fused method were qualitatively and quantitatively compared to data processed with conventional dMRI processing methods. Results show the advantages of the TiDi-Fused pipeline including sharper brainstem gray-white matter tissue contrast, improved inter-subject spatial alignment for group analyses of dMRI based measures, accurate spatial alignment with histology-based imaging of the brainstem, reduced variability in brainstem-cerebellar white matter tracts, and more robust biologically plausible relationships between age and brainstem-cerebellar white matter tracts. Overall, this work identifies a promising pipeline for achieving high-resolution imaging of brainstem structures in pediatric and clinical populations who may not be able to endure long scan times. This pipeline may serve as a gateway for feasibly elucidating brainstem contributions to autism and other conditions.

Brainstem white matter microstructure is associated with hyporesponsiveness and overall sensory features in autistic children

“…This study set out to identify the relationships between sensory features and white matter microstructure of the underexplored brainstem in autistic and non-autistic children. Using a novel DWI protocol that improved the apparent resolution of the brainstem and cerebellum [93], we precisely delineated brainstem and brainstem-cerebellar white matter tracts and examined their associations with total sensory features and specific sensory responses. Consistent with our hypotheses, results revealed that the microstructural properties of brainstem white matter tracts were associated with sensory features, particularly in autistic children.…”

“…In all, the present findings, combined with theoretical work and studies implicating the brainstem in autism [18][19][20], suggest that the brainstem and cerebellum may be integral contributors to the sensory experiences of autistic individuals. Therefore, even though the imaging of the brainstem may require special acquisition and processing procedures [49], including free water elimination, EPI distortion correction, and careful consideration of brainstem masking, these steps are worth taking, as the brainstem and cerebellum are likely key areas to study to better understand the neurobiological basis of the autistic experience.…”

“…DWI data were processed to minimize noise [65,66], Gibbs ringing [67], artifacts caused by motion, eddy current [68][69][70], EPI distortion [62], as well as B0 field inhomogeneities [71,72]. To enhance the apparent spatial resolution, DWI data were then processed in accordance to TiDi-Fused protocol [49]. The mean DWI b = 0 volume was spatially aligned to the T1 weighted image derived from the MPnRAGE using rigid transformations (6 degrees of freedom) implemented with the boundarybased registration (BBR) [73] routine in the FreeSurfer image analysis suite [74].…”

“…Previously, methodological constraints limited the feasibility of collecting high resolution diffusion-weighted imaging (DWI) data (traditionally a time consuming and sensory intensive process) in pediatric populations with sensory features. However, recent advancements in our DWI protocol have allowed us to overcome these limitations, providing high apparent resolution and improved gray-white matter contrast without requiring long acquisition times [ 49 ]. These innovations offer the opportunity to investigate white matter microstructure of brainstem tracts in children with elevated sensory features with a higher degree of precision than ever before.…”

Role of autonomic, nociceptive, and limbic brainstem nuclei in core autism features

“…Therefore, the aim of the present study was to use a brainstem‐optimized image acquisition and post‐processing methods (Guerrero‐Gonzalez et al, 2022) combined with a mapping of the autonomic/pain/emotion brainstem nuclei (Bianciardi et al, 2015, 2018; García‐Gomar et al, 2019, 2022; Singh et al, 2019, 2021) to test if microstructural properties of brainstem relate to individual variation in core social and repetitive‐behavior features in autistic children. Children with no known diagnosis of autism or other neurodevelopmental conditions (i.e., “non‐autistic” children) are included for context and for examining these associations across the broader population, but they are not included for direct group comparisons, as the autonomic literature suggests greater within‐group than between‐group variation.…”