In-vivo quantitative structural imaging of the human midbrain and the superior colliculus at 9.4T
Introduction
In-vivo magnetic resonance imaging (MRI) studies of the midbrain are challenging due to the small size and proximity of its nuclei and structures, and often unknown sensitivity to MR contrast mechanisms. MR imaging of midbrain structures has considerably improved in recent years due to advances in technology, including an increased availability of high field scanners, tailored sequence development, and optimized post-processing methods for brainstem imaging (for a recent review see Sclocco et al., 2018).
Both signal-to-noise ratio and image contrast increase with field strength which consequently enables the acquisition of high resolution images within reasonable acquisition times (Pohmann et al., 2011, 2016). The effects of local differences in magnetic susceptibility are particularly prominent at high magnetic field strengths (Budde et al., 2014; Duyn et al., 2007) and therefore enable a better definition of deep brain nuclei (Deistung et al., 2013a, 2013b; Sclocco et al., 2018; Stüber et al., 2014). Multiple studies reported the delineation of the red nucleus (RN) and substantia nigra (SN) at 7T using magnitude images of T2* weighted sequences (Abduljalil et al., 2003; Bourekas et al., 1999; Eapen et al., 2011). Due to the increase of the susceptibility-effect at high field, phase images also provided enhanced contrast in cortical and deep brain structures revealing details that were not visible in magnitude images (Abduljalil et al., 2003). However, phase and magnitude images represent non-local contrast mechanisms that show orientation-dependence and cannot be easily compared across subjects, thus impeding reproducibility (Li et al., 2011). The use of quantitative relaxometry measures, such as mapping of the longitudinal (R1), the transversal relaxation rates (R2 or R2*), and quantitative susceptibility mapping (QSM) are emerging for the in vivo study of deep brain anatomy potentially yielding more reproducible results. These parameters are influenced by the presence of myelin and iron in the brain and the combination of different maps may reveal local features based on the cyto- and myeloarchitecture of tissues (Deistung et al., 2013b; Deoni, 2010; Stüber et al., 2014).
Deistung et al. (2013a) reported the delineation of the boundary between the SN and the subthalamic nucleus (STN) using QSM at 7T for the first time. Deistung et al. (2013a) also reported multiple fiber tracts in the midbrain as well as SN and RN subdivisions, which were not visible in other quantitative maps. Forstmann et al. (2014), delineated the STN and SN and created probabilistic maps based on these findings. Several studies used a combination of diffusion based, T2 and T2* weighted MRI to automatically delineate several midbrain nuclei and provided probabilistic information on their location and extent (Abduljalil et al., 2003; Bianciardi et al., 2015, 2018; Forstmann et al., 2014; Keuken et al., 2014; Lorio et al., 2016). Several of these in vivo findings are available as MNI-based atlases, thus bridging the gap between in vivo data and previous atlases based on histology (Ewert et al., 2017). Beyond the critical importance of MNI space coregistration for the transfer of anatomical information across studies, it may support the detection of subtle anatomical detail with low contrast to noise ratios, in case these show spatial congruency across individuals and across normalized images from multiple individuals.
While the more prominent midbrain nuclei, particularly the STN and SN, have already received much attention in high field MRI imaging, the superior colliculus (SC) has not been investigated in detail, despite its potential role in several neurological diseases. Alterations of the SC's structural integrity, functional properties, and connections are reported to occur in patients with degenerative diseases and in animal models of degenerative diseases, such as dementia with Lewy bodies (Erskine et al., 2017), Alzheimer's disease and progressive supranuclear palsy (Dugger et al., 2011; Terao et al., 2013), amyotrophic lateral sclerosis (van Zundert et al., 2008), and cervical dystonia (Holmes et al., 2012; Hutchinson et al., 2014). Visual and oculomotor impairments in patients with Parkinson's disease are associated with pathological changes of brainstem networks including the SC (Cubizolle et al., 2014; Diederich et al., 2014; Rolland et al., 2013; Terao et al., 2013).
Beyond a mere delineation of the SC as a whole, objective methods based on high-resolution MRI to detect and describe the intra-collicular structure would be needed to fully understand its functional role in health and disease. A regular pattern characterizing the SC emerges from post-mortem studies describing the histology of the human SC (Laemle, 1980; Laemle, 1983; Tardif and Clarke, 2002; Tardif et al., 2005) as well as in non-human primates and the cat (Kanaseki and Sprague, 1974; Ma et al., 1990). Three main zones that constitute the superficial, the intermediate, and the deep SC with fundamental differences in their histology (May 2006), mirrored by differences in their functional MR signal (Loureiro et al., 2017; Katyal and Ress, 2014; Zhang et al., 2015), have been described. However, even finer subdivisions based on the intra-collicular structure can be made. When going from the surface of the SC towards its deeper parts, as many as seven layers with distinct microstructural features, alternating between myelinated fibers and neuronal cell bodies of different size and shape have been described (Kanaseki and Sprague, 1974; Ma et al., 1990). The use of quantitative MRI methods, performed with high-resolutions sequences and at high magnetic field strength, may be well suited to reveal this pattern, in virtue of their intrinsic sensitivity to cyto- and myeloarchitecture features of the SC (Nieuwenhuys et al., 2008).
In this study, our main objective was therefore to explore the possibility that high field, high-resolution MRI lends itself to the detection of the finer details of the SC, down to the level of single layers. We used relaxometry and QSM at 9.4T to examine anatomical properties of the human midbrain, in particular the SC and its internal structure in 14 subjects. Besides the SC, we also investigated the use of these different MR modalities in neighboring midbrain regions such as the RN, the cerebral aqueduct (CA) and a region covering white matter tracts in the central midbrain (WM), like the medial lemniscus and the ventral tegmental tract. We evaluated the MR-properties (R1, R2* and QSM) in these regions and calculated the coefficient-of-variation observed across subjects, both based on data from each subjects' individual images in native space and after transformation of all individual images into the common MNI brain space. Microstructural features were investigated by depth profiles (DP) across the SC layers, evaluated at different positions along the rostro-caudal axis, enabling the study of MR properties likely reflecting the underlying microarchitecture and the layering pattern of this structure. The combination of R1, R2*, and QSM maps provided a better understanding of the effect of the intrinsic microstructure, in terms of cyto- and myeloarchitecture of the SC tissue on the different MR maps than each single modality on its own. We foresee that the proposed approach can be used to improve our understanding of microstructural features of the SC layers and eventually relate these to their different functions.
Section snippets
Participants
Fourteen healthy volunteers with a mean age of 27 years (range 21–34 years; 3 females) participated in this study. In accordance with local research ethics policies and procedures, the volunteers underwent a physical and psychological check-up by a local physician and provided written informed consent. All investigations were conducted in agreement with the Declaration of Helsinki in its most recent version (WMA, 2013).
Data acquisition and imaging protocol
All measurements were conducted with a 9.4T whole-body MRI scanner (Siemens,
Identification and delineation of midbrain structures across modalities
Several structures could be identified by virtue of their previously described cyto- and myeloarchitecture (Nieuwenhuys et al., 2008). These structures are shown in Fig. 2 (numbers indicate the structure in Fig. 2) with dominant cytoarchitectonic features in SC (1); periaqueductal gray (PAG) (2); oculomotor nucleus (3); parvocellular (5) and magnocellular (18) portion of RN; pars compacta (6) and pars reticulata (7) of the SN; medial geniculate body (8); and with dominant myeloarchitectural
Discussion
In this study we evaluated the use of quantitative MR-properties (magnetic susceptibility, longitudinal and effective transverse relaxation rates) for investigating the tissue microstructure of the SC and neighboring midbrain structures in-vivo at 9.4T. The results presented build on recent advancements in imaging technology and the availability of high field MRI methods, on one hand, and on the close relationship between measured MR-properties and tissue cyto- and myeloarchitecture, on the
Summary and conclusions
From this work, it becomes clear that the study of the anatomy of deep brain structures, highly benefits from the combined evaluation of the different quantitative maps at high-field in individual scans, as different contrasts reveal complementary structural details which are unlikely to be visible in one single modality or at standard fields (3T and below). With this study, the extra structural information that is acquired at 9.4T is translated into a common brain template (MNI) that can be
Acknowledgments
We gratefully acknowledge funding by the Max Planck Society, the ministry of Science, Research and the Arts of Baden-Württemberg (Az: 32-771-8-1504.12/1/1), and a PhD Grant from the Carl-Zeiss Stiftung (1/01/2015–30/06/2017).
References (70)
- et al.
A probabilistic template of human mesopontine tegmental nuclei from in vivo 7T MRI
NeuroImage
(2018) - et al.
Ultra-high resolution imaging of the human brain using acquisition-weighted imaging at 9.4 T
NeuroImage
(2014) - et al.
Parkinson's disease, L-Dopa and “express” saccades: superior colliculus dyskinesias?
Clin. Neurophysiol.
(2014) - et al.
Toward in vivo histology: a comparison of quantitative susceptibility mapping (QSM) with magnitude-, phase-, and R2*-imaging at ultra-high magnetic field strength
NeuroImage
(2013) - et al.
Disease specificity and pathologic progression of tau pathology in brainstem nuclei of Alzheimer's disease and progressive supranuclear palsy
Neurosci. Lett.
(2011) - et al.
Neuronal loss and alpha-synuclein pathology in the superior colliculus and its relationship to visual hallucinations in dementia with Lewy bodies
Am. J. Geriatr. Psychiatry
(2017) - et al.
Imaging iron stores in the brain using magnetic resonance imaging
Magn. Reson. imaging
(2005) - et al.
Whole brain MP2RAGE-based mapping of the longitudinal relaxation time at 9.4 T
NeuroImage
(2017) - et al.
Quantifying inter-individual anatomical variability in the subcortex using 7 T structural MRI
NeuroImage
(2014) - et al.
Surface-based analysis methods for high-resolution functional magnetic resonance imaging
Graph models. 2011
(2011)
Multiparametric brainstem segmentation using a modified multivariate mixture of Gaussians
Neuroimage Clin.
Quantitative susceptibility mapping of human brain reflects spatial variation in tissue composition
NeuroImage
Magnetic susceptibility anisotropy of human brain in vivo and its molecular underpinnings
NeuroImage
A method for estimating and removing streaking artifacts in quantitative susceptibility mapping
NeuroImage
High-field (9.4 T) MRI of brain dysmyelination by quantitative mapping of magnetic susceptibility
NeuroImage
New tissue priors for improved automated classification of subcortical brain structures on MRI
Neuroimage
MP2RAGE, a self bias-field corrected sequence for improved segmentation and T 1-mapping at high field
NeuroImage
The mammalian superior colliculus: laminar structure and connections
Prog. Brain Res.
The use of power images to perform quantitative analysis on low SNR MR images
Magn. Reson. imaging
Enhanced visual responses in the superior colliculus and subthalamic nucleus in an animal model of Parkinson's disease
Neuroscience
Challenges and opportunities for brainstem neuroimaging with ultrahigh field MRI
NeuroImage
Myelin and iron concentration in the human brain: a quantitative study of MRI contrast
NeuroImage
Commissural connections of human superior colliculus
Neuroscience
Deterioration of horizontal saccades in progressive supranuclear palsy
Clin. Neurophysiol.
User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability
NeuroImage
Layer-specific response properties of the human lateral geniculate nucleus and superior colliculus
NeuroImage
Enhanced gray and white matter contrast of phase susceptibility-weighted images in ultra-high-field magnetic resonance imaging
J. Magn. Reson Imaging
An assessment of current brain targets for deep brain stimulation surgery with susceptibility-weighted imaging at 7 tesla
Neurosurgery
The in vivo influence of white matter fiber orientation towards B0 on T2* in the human brain
NMR Biomed.
Toward an in vivo neuroimaging template of human brainstem nuclei of the ascending arousal, autonomic, and motor systems
Brain Connect.
High resolution MRI of the deep gray nuclei at 8 Tesla
J. Comput. Assist. Tomogr.
Characterization of white matter fiber bundles with T 2* relaxometry and diffusion tensor imaging
Magn. Reson. Med.
High-resolution MR imaging of the human brainstem in vivo at 7 Tesla
Front. Hum. Neurosci.
Quantitative relaxometry of the brain
Top. magnetic Reson. imaging TMRI
Are patients with Parkinson's disease blind to blindsight?
Brain
Cited by (11)
Establishing sample-preparation protocols for X-ray phase-contrast CT of rodent spinal cords: Aldehyde fixations and osmium impregnation
2020, Journal of Neuroscience MethodsCitation Excerpt :MRI, unlike histological neuroimaging work, can remain free from staining and labeling, and can produce truly-volumetric nervous tissue maps composed of (possibly) thousands of virtual slices. Ultra-high field MRI scanners are equipped with strong gradients, which permit access to 3D nervous tissue mesoscopic (Sear et al., 2015) bio-scale structure without the need for sample sectioning, and afford high CNS soft-tissue contrast already in-vivo (Loureiro et al., 2018). Post-mortem high-field MRI (PMMR) can efficiently reach, though obviously at the cost of long measurement times, even higher CNS image quality (Calabrese et al., 2018), and obtain e.g. exquisite soft-tissue contrast and ultra-high resolution within large human brain specimens (Sengupta et al., 2018) (isotropic pixel sizes in the order of 50 μm), or even full rodent CNS visualizations (Harrison et al., 2013) (isotropic pixel sizes in the order of 25 μm).
Visual Functions of the Primate Superior Colliculus
2023, Annual Review of Vision SciencePhase-based masking for quantitative susceptibility mapping of the human brain at 9.4T
2022, Magnetic Resonance in MedicineMulti-echo gradient-recalled-echo phase unwrapping using a Nyquist sampled virtual echo train in the presence of high-field gradients
2021, Magnetic Resonance in MedicineQuantitative Susceptibility Mapping of the Basal Ganglia and Thalamus at 9.4 Tesla
2021, Frontiers in Neuroanatomy
- 1
These authors contributed equally to this work.