1247

Probabilistic structural atlas of human lateral parabrachial nucleus, medial parabrachial nucleus and vestibular nuclei complex using in vivo 7 Tesla MRI.
Kavita Singh1, Iole Indovina2,3, Jeffrey P Staab4,5, and Marta Bianciardi1

1Brainstem Imaging Lab, Department of Radiology, Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital and Harvard Medical School, Charlestown, MA, United States, 2Laboratory of Neuromotor Physiology, IRCCS Santa Lucia Foundation, Rome, Italy, 3Centre of Space BioMedicine, University of Rome Tor Vergata, Rome, Italy, 4Department of Psychiatry and Psychology, Mayo Clinic, Rochester, MN, United States, 5Department of Otorhinolaryngology - Head and Neck Surgery, Mayo Clinic, Rochester, MN, United States

Synopsis

Parabrachial and vestibular nuclei are anatomically and functionally connected brainstem gray-matter structures involved in autonomic and vestibular functions. Their assessment in research and clinical investigations is difficult due to limited image-resolution/contrast of clinical scanners and the absence of probabilistic atlas of these structures. We delineated these nuclei in single-subject multi-contrast 1.1mm-resolution 7Tesla MRI of healthy humans and generated a validated in-vivo probabilistic atlas of these nuclei in stereotaxic space. Upon coregistration to clinical-MRI, this atlas might improve the evaluation of lesions and assessment of connectivity-pathways underlying autonomic and vestibular mechanisms in a broad-set of clinical conditions relating to these nuclei.

Introduction:

Parabrachial and vestibular nuclei are anatomically and functionally connected brainstem gray-matter structures, which regulate arousal, sense of balance, spatial orientation and autonomic processes individually or synergistically [1]. Alterations in these nuclei due to disease or injury are manifested in several clinical conditions such as vestibular and balance disorders, sleep disorders, anxiety disorders, and altered autonomic function [2-6]. However, due to the limited image resolution/contrast of clinical (1.5 Tesla, 3 Tesla) MRI-scanners and the lack of a stereotaxic probabilistic atlas of these nuclei in living humans, the assessment of structural and functional alterations of these nuclei is difficult in routine research and clinical investigations.

Purpose:

To create in living humans a stereotaxic probabilistic structural atlas of the right and left lateral parabrachial (LPB-r, LPB-l) nucleus, medial parabrachial (MPB-r, MPB-l) nucleus and vestibular nuclei complex (Ve-r, Ve-l) by using a high-resolution (1.1-mm isotropic) multi-contrast (diffusion fractional-anisotropy-FA and T2-weighted) EPI-approach at 7 Tesla, which provided complementary contrasts for brainstem anatomy with precisely matched geometric distortions and resolution.

Methods:

Data acquisition: Twelve subjects (6m/6f, age 28 ± 1) underwent 7 Tesla MRI under IRB approval. We adopted a common single-shot 2D EPI-readout for 1.1-mm isotropic diffusion tensor (DTI), and T2 weighted (T2w) sagittal images, with matrix-size/GRAPPA-factor/nominal echo-spacing = 180 × 240/3/0.82 ms. This yielded multi-contrast structural images with exactly matched resolution and geometric-distortions. Additional MRI-parameters were: spin-echo EPI, n. slices/diffusion-weighting gradients/echo-time/repetition-time/phase-encoding direction/bandwidth/partial-Fourier/n. diffusion-directions/b-value: 61/unipolar/60.8 ms/5.6 s/“anterior/posterior”/“1488 Hz/pixel”/“6/8”/60/1000 s/mm2, seven interspersed “b0” images (T2-weighted, non-diffusion weighted, b-value = 0 s/mm2), 4 repetitions, acquisition time/repetition 6’43”. Data analysis: DTIs were rotated to standard orientation (“RPI”), motion and distortion corrected (FSL). We then computed the diffusion tensor, tensor-invariants (such as FA) from the preprocessed DTI (FSL, dtifit). Single-subject FA were coregistered to FA-templates [7] in stereotaxic Montreal-Neurological-Institute (MNI) space through high-dimensional non-linear transformations (ANTs [8]). On a single-subject basis, two independent researchers (K.S., M.B.) performed manual segmentation of multi-contrast (FA maps and T2-weighted MRI) images as follows: the lateral parabrachial nucleus was identified as a hypointense thin stripe in the FA-map lateral to the superior cerebellar peduncle, and medial to the cerebrospinal fluid (CSF-the borders with the CSF were visible on the T2-weighted MRI); the medial parabrachial nucleus appeared as a hypointense stripe in the FA-map medial to the superior cerebellar peduncle; the vestibular-nuclei complex was a FA hypointense region in the ponto-medullar junction bounded dorso-medially by the CSF and laterally by the inferior cerebellar peduncle. This segmentation yielded single-subject labels of LPB-r/l, MPB-r/l and Ve-r/l. Only voxels rated by both raters as belonging to a nucleus were included in the final nucleus label. A probabilistic atlas for these nuclei in stereotaxic MNI space was created by computing the overlap of nuclei-labels across subjects (highest probability = 100 % overlap across subjects). Atlas validation: The probabilistic nuclei-atlas was validated by computing for each nucleus and subject: (i) the inter-rater agreement, as the modified Hausdorff-distance [9] between labels delineated by the two raters; (ii) the internal consistency across subjects of the final label, as the modified Hausdorff-distance between each final label and the probabilistic atlas label (thresholded at 35%) generated by averaging the labels across the other 11 subjects (leave-one-out cross-validation). For each nucleus, the modified Hausdorff-distance of (i) and (ii) was then averaged across subjects and displayed.

Results:

The probabilistic neuroimaging structural labels in MNI space of LPB-r/l (Figure1), MPB-r/l (Figure2) and Ve-r/l (Figure3.) are shown. For each nucleus, the average modified Hausdorff-distance assessing the inter-rater agreement and the internal consistency (Figure 4) of nuclei atlas labels was below (p < 0.05, unpaired one-tailed t-test) the linear spatial imaging resolution (1.1 mm), thus validating the generated probabilistic nuclei atlas.

Discussion and Conclusions:

Our findings demonstrated the feasibility of delineating on a single-subject basis tiny lateral and medial parabrachial nuclei along with the vestibular-nuclei complex in high-contrast and high-sensitivity 7 Tesla MRI. Crucially, our work also demonstrated the feasibility of generating a validated in vivo stereotaxic probabilistic atlas of these structures after precise image coregistration to MNI-space. This atlas complements existing in vivo neuroimaging atlases of other brain structures [10-13]. We foresee the use of the generated probabilitic atlas of LPB-r/l, MPB-r/l and Ve-r/l to aid the localization of these nuclei in conventional (e.g. 3T) images in future research studies of autonomic and vestibular functions. Further, this atlas, upon coregistration to clinical MRI, might improve the accuracy of interventions, the evaluation of lesions and the assessment of connectivity pathways underlying autonomic and vestibular mechanisms in a broad set of clinical conditions relating to these nuclei.

Acknowledgements

NIH-NIDCD-R21DC015888; NIH-NIBIB-K01EB019474; NIH-NIBIB-P41-EB015896

References

  1. Olszewski, J., Baxter, D., Karger, S. (Ed.), Cytoarchitecture of the Human Brainstem, JB Lippincott Company, Philadelphia and Montreal North America, 1954.
  2. Staab, J.P., Balaban, C.D., Furman, J.M., 2013. Threat assessment and locomotion: clinical applications of an integrated model of anxiety and postural control. Semin Neurol 33, 297-306.
  3. Balaban, C.D., 2002. Neural substrates linking balance control and anxiety. Physiol Behav 77, 469-475.
  4. Balaban, C.D., Jacob, R.G., Furman, J.M., 2011. Neurologic bases for comorbidity of balance disorders, anxiety disorders and migraine: neurotherapeutic implications. Expert Rev Neurother 11, 379-394.
  5. Agrawal, Y., Carey, J.P., Della Santina, C.C., et al., 2009. Disorders of balance and vestibular function in US adults. Arch Intern Med 169, 938-944.
  6. Brandt, T., Dieterich, M., Strupp, M., 2005. Vertigo and dizziness: common complaints. Springer-Verlag, London.
  7. Varentsova A, Zhang S, Arfanakis K., 2014. Development of a high angular resolution diffusion imaging human brain template. Neuroimage. 91, 177-186.
  8. Avants B.B., Tustison N.J., Song G., et al., 2011. A reproducible evaluation of ANTs similarity metric performance in brain image registration. Neuroimage. 54, 2033-2044.
  9. Dubuisson, M.P., Jain, A.K., 1994. A modified Hausdorff distance for object matching. Proc. 12th International Conference on Pattern recognition, Jerusalem, Israel, 1, 566-568.
  10. Destrieux C., Fischl B., Dale A., et al., 2010. Automatic parcellation of human cortical gyri and sulci using standard anatomical nomenclature. Neuroimage. 53, 1-15.
  11. Desikan R.S., Segonne F., Fischl B., et al., 2006. An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. Neuroimage. 31, 968-980.
  12. Tzourio-Mazoyer N., Landeau B., Papathanassiou D., et al., 2002. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. Neuroimage.15, 273-289.
  13. Bianciardi M., Toschi N., Edlow B.E., et al., 2015. Toward an in vivo neuroimaging template of human brainstem nuclei of the ascending arousal, autonomic, and motor systems. Brain Connect. 5, 597-607.

Figures

Figure1: Upper panel shows probabilistic atlas label (n = 12) in MNI space of the lateral parabrachial nucleus (right: blue-to-cyan; left: red-to-yellow). Lower panel shows FA images with corresponding hypointense label regions. Very good (i.e. up to 100 %) spatial agreement of labels across subjects was observed indicating the feasibility of delineating the probabilistic label of these nuclei involved in arousal and autonomic functions.

Figure2: Upper panel shows probabilistic atlas label (n = 12) in MNI space of the medial parabrachial nucleus (right: blue-to-cyan; left: red-to-yellow). Lower panel shows FA images with corresponding hypointense label regions. Very good (i.e. up to 100 %) spatial agreement of labels across subjects was observed indicating the feasibility of delineating the probabilistic label of these nuclei involved in arousal and autonomic functions.

Figure3: Upper panel shows probabilistic atlas label (n = 12) in MNI space of the vestibular nuclei complex (right: blue-to-cyan; left: red-to-yellow). Lower panel shows FA images with corresponding hypointense label regions. Very good (i.e. up to 100 %) spatial agreement of labels across subjects was observed indicating the feasibility of delineating the probabilistic label of this complex of nuclei involved in vestibular functions (e.g. postural, oculo-motor control).Note that, from a preliminary inspection,we did not have enough resolution/contrast to easily discriminate between individual (e.g. superior, lateral, medial, inferior) vestibular nuclei within this complex.

Figure 4: Atlas validation. We show: (A). The inter-rater agreement of nuclei segmentations (bar/error bar = mean/s.e. modified Hausdorff distance across 12 subjects); (B). the internal consistency of nuclei labels across subjects (bar/errorbar = mean/s.e. modified Hausdorff distance across 12 subjects). The labels of the LPB-r,LPB-l,MPB-r,MPB-l,Ve-r,and Ve-l displayed good inter-rater agreement and internal consistency across raters and subjects (the modified Hausdorff distance was smaller than the spatial imaging resolution), thus validating the probabilistic nuclei atlas.

Proc. Intl. Soc. Mag. Reson. Med. 27 (2019)
1247