Thalamic connectivity gradients analysis reveal atypical thalamic-insula interactions in autism
Poster No:
426
Submission Type:
Abstract Submission
Authors:
Shinwon Park1, Phoebe Thomson1, Han Byul Cho2, Sofie Valk3, Boris Bernhardt4, Michael Milham1,5, Seok Jun Hong6,7, Adriana Di Martino1
Institutions:
1Child Mind Institute, New York, NY, 2Center for Neuroscience Imaging Research, Institute for Basic Science, Suwon-si, Gyeonggi-do, 3Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany, 4Montreal Neurological Institute and Hospital, Montreal, Quebec, 5Nathan S. Kline Institute for Psychiatric Research, Orangeburg, NY, 6Center for Neuroscience Imaging Research, Institute for Basic Science, Suwon, Gyeonggi-do, 7Department of Biomedical Engineering, Sungkyunkwan University, Suwon, Korea, Republic of
First Author:
Co-Author(s):
Han Byul Cho
Center for Neuroscience Imaging Research, Institute for Basic Science
Suwon-si, Gyeonggi-do
Center for Neuroscience Imaging Research, Institute for Basic Science
Suwon-si, Gyeonggi-do
Michael Milham
Child Mind Institute|Nathan S. Kline Institute for Psychiatric Research
New York, NY|Orangeburg, NY
Child Mind Institute|Nathan S. Kline Institute for Psychiatric Research
New York, NY|Orangeburg, NY
Seok Jun Hong
Center for Neuroscience Imaging Research, Institute for Basic Science|Department of Biomedical Engineering, Sungkyunkwan University
Suwon, Gyeonggi-do|Suwon, Korea, Republic of
Center for Neuroscience Imaging Research, Institute for Basic Science|Department of Biomedical Engineering, Sungkyunkwan University
Suwon, Gyeonggi-do|Suwon, Korea, Republic of
Introduction:
Thalamocortical connectivity is crucial for sensory information processing and cognitive integration. Evidence from our work and others indicate that thalamocortical connections with the salience network plays a pivotal role in distinguishing between internally and externally oriented functional networks.[1,2] Extending this framework, we hypothesized that atypical thalamocortical connectivity may underlie the atypical sensory processing and social communication often observed in autism spectrum disorder (ASD). The aim of this study is to examine thalamocortical functional connectivity maps (CMAP) and their neocortical projections (NEOMAP) in ASD compared to age- and sex-matched neurotypicals (NT) and analyze their association with clinical symptoms.
Methods:
Utilizing the Autism Brain Imaging Data Exchange-I (ABIDE-I)[3] dataset from 3 different sites (i.e., NYU, PITT, USM), our study involved 107 ASD and 113 NT individuals, all male, aged 6.4 – 50 years. We used the Brainspace toolbox[4] to derive the CMAPs using diffusion map embedding with a normalized angle kernel, and Procrustes alignment for aligning individual CMAPs to the group mean average of all participants. Next, the NEOMAPs were extracted by multiplying the CMAPs (thalamus voxels x n maps) to the thalamus-cortex correlation matrix (thalamus voxels x cortex vertices). Thus, NEOMAPs can be interpreted as the projections of the CMAPs onto the neocortical surface. CMAPs and NEOMAPs were statistically tested for significant group difference effects, controlling for age and head motion (mean framewise displacement) using surface-based linear models implemented in a MATLAB toolbox, SurfStat.[5] Potential site effects were controlled using ComBat harmonization.[6] To interpret these maps, we profiled them using theYeo-Krienan 7 network parcellation.[7] Finally, we examined associations with the calibrated severity total scores of the Autism Diagnostic Observation Schedule (ADOS CSS).[8]
Results:
Across groups, CMAP patterns resulting from our analyses were consistent with those observed in our prior study of neurotypical thalamocortical connectivity. Group comparisons revealed significant differences between ASD and NT groups in the first CMAP (pFDR<0.05, Figure 1A, 1B). These differences were localized to specific thalamic nuclei: the pulvinar associated with visual processing, the ventral lateral posterior nucleus linked to somatosensory functions, the medial geniculate nucleus integral to auditory processing, and the centromedian nucleus, known for its role in 'gate control' of salient features. The second CMAP did not show significant group differences (Figure 1C). Based on these results, follow up analyses focusing on the first NEOMAP, ranging from salience/somatosensory to default mode networks showed a more compressed gradient in ASD compared to NT (Figure 2A, 2B). A vertexwise analysis revealed significant differences in regions of the right insula, superior temporal, and visual cortices (PRFT-Cluster<0.05; Figure 2C), where ASD showed lower values compared to NT. Network profiling of these regions based on Yeo-Krienan 7 networks,[7] revealed that these differences were most pronounced in the visual, somatosensory, salience, and default mode networks (Figure 2D). Notably, among the regions showing significant group differences (ASD<NT), the right insula, within the salience and somatomotor networks, showed a significant negative correlation with autism symptom severity (r = -0.30, p = 0.002; Figure 2E).
Conclusions:
Our study demonstrates that atypical thalamocortical connectivity, particularly involving the insula, a key node of the salience network, may underlie ASD symptom severity. Further research, however, is necessary to understand the impact of these atypical thalamo-insular connections in ASD, particularly in terms of how they influence the interaction between regions processing internally and externally oriented information.
Disorders of the Nervous System:
Neurodevelopmental/ Early Life (eg. ADHD, autism) 1
Modeling and Analysis Methods:
fMRI Connectivity and Network Modeling
Task-Independent and Resting-State Analysis
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Subcortical Structures 2
Novel Imaging Acquisition Methods:
BOLD fMRI
Keywords:
Autism
Development
FUNCTIONAL MRI
MRI
Sub-Cortical
Thalamus
Other - Functional gradient
1|2Indicates the priority used for review
Provide references using author date format
2. Alves, P. N. et al. An improved neuroanatomical model of the default-mode network reconciles previous neuroimaging and neuropathological findings. Commun Biol 2, 370 (2019).
3. Di Martino, A. et al. The autism brain imaging data exchange: towards a large-scale evaluation of the intrinsic brain architecture in autism. Mol. Psychiatry 19, 659–667 (2014).
4. Vos de Wael, R. et al. BrainSpace: a toolbox for the analysis of macroscale gradients in neuroimaging and connectomics datasets. Commun Biol 3, 103 (2020).
5. Worsley, K. J. et al. A Matlab toolbox for the statistical analysis of univariate and multivariate surface and volumetric data using linear mixed effects models and random field theory. in NeuroImage Organisation for Human Brain Mapping 2009 Annual Meeting vol. 47 S102 (math.mcgill.ca, 2009).
6. Fortin, J.-P. et al. Harmonization of multi-site diffusion tensor imaging data. Neuroimage 161, 149–170 (2017).
7. Yeo, B. T. T. et al. The organization of the human cerebral cortex estimated by intrinsic functional connectivity. J. Neurophysiol. 106, 1125–1165 (2011).
8. Gotham, K., Pickles, A. & Lord, C. Standardizing ADOS scores for a measure of severity in autism spectrum disorders. J. Autism Dev. Disord. 39, 693–705 (2009).