Large Data on the Small Brain: Population-wide Cerebellar Growth Models of Children and Adolescent
Poster No:
1222
Submission Type:
Abstract Submission
Authors:
Carolin Gaiser1, Rick van der Vliet1, Augustijn de Boer2, Opher Donchin3, Pierre Berthet4, Gabriel Devenyi5, Mallar Chakravarty6, Jörn Diedrichsen7, Andre Marquand2, Maarten Frens1, Ryan Muetzel1
Institutions:
1Erasmus Medical Center, Rotterdam, Zuid Holland, 2Radboud University Nijmegen, Nijmegen, Gelderland, 3Ben-Gurion University of the Negev, Be'er Sheva, Be'er Sheva, 4University of Oslo, Oslo, Oslo, 5McGill University, Montreal, Quebec, 6Brain Imaging Centre, Douglas Research Centre, Montreal, Quebec, 7The Brain and Mind Institute, University of Western Ontario, London, Ontario
First Author:
Co-Author(s):
Introduction:
The cerebellum is known to be engaged in a broad spectrum of functions. While its involvement in motor control is best documented, recent efforts have made clear that it is also involved in cognitive function.
In the current study, we describe and provide openly available normative models of anatomical and functional subregions of the cerebellum from a large pediatric population that 1) can be used as reference models to obtain accurate normative ranges, also in smaller datasets, by benefitting from informative hyperpriors based on a large sample, and that 2) can be updated with data from new sites and extended age ranges without the necessity of sharing sensitive patient or participant data. We furthermore illustrate the usefulness and practicality of the current approach by mapping the deviations from typical cerebellar development at the level of the individual in a subpopulation of children with autistic traits (Constantino et al., 2003). These models have the potential to facilitate and maximize the use of cerebellar outcomes in neuroimaging research and as a result aid to better understand the role of the cerebellum in typical as well as atypical neurodevelopment.
In the current study, we describe and provide openly available normative models of anatomical and functional subregions of the cerebellum from a large pediatric population that 1) can be used as reference models to obtain accurate normative ranges, also in smaller datasets, by benefitting from informative hyperpriors based on a large sample, and that 2) can be updated with data from new sites and extended age ranges without the necessity of sharing sensitive patient or participant data. We furthermore illustrate the usefulness and practicality of the current approach by mapping the deviations from typical cerebellar development at the level of the individual in a subpopulation of children with autistic traits (Constantino et al., 2003). These models have the potential to facilitate and maximize the use of cerebellar outcomes in neuroimaging research and as a result aid to better understand the role of the cerebellum in typical as well as atypical neurodevelopment.
Methods:
Anatomical Parcellation
The cerebellum was parcellated in the native space into 35 anatomical subdivisions using the MAGeT pipeline (Chakravarty et al., 2013; Park et al., 2014).
Functional Parcellation
As lobular boundaries of the cerebellum have shown limited correspondence with functional boundaries, we also employed the functional subregions proposed by King and colleagues (King et al., 2019).
Normative Models
To generate normative models for anatomical and functional subregions of the cerebellum, we made use of the PCNtoolkit python package version 0.27 (de Boer et al., 2022; Rutherford et al., 2022b) using Python 3.10.6. Using Hierarchical Bayesian Regression, we estimated normative models of the cerebellum using both the volumes from the anatomical parcellation and the morphological indicators (i.e., grey and white matter densities as well as volumes) of the functional parcellation from age, for each region of interest (ROI) separately. Sex and scanner were modeled as batch-effects.
The cerebellum was parcellated in the native space into 35 anatomical subdivisions using the MAGeT pipeline (Chakravarty et al., 2013; Park et al., 2014).
Functional Parcellation
As lobular boundaries of the cerebellum have shown limited correspondence with functional boundaries, we also employed the functional subregions proposed by King and colleagues (King et al., 2019).
Normative Models
To generate normative models for anatomical and functional subregions of the cerebellum, we made use of the PCNtoolkit python package version 0.27 (de Boer et al., 2022; Rutherford et al., 2022b) using Python 3.10.6. Using Hierarchical Bayesian Regression, we estimated normative models of the cerebellum using both the volumes from the anatomical parcellation and the morphological indicators (i.e., grey and white matter densities as well as volumes) of the functional parcellation from age, for each region of interest (ROI) separately. Sex and scanner were modeled as batch-effects.
Results:
The anatomical and functional regions show similar overall growth trends. As expected in this age range traversing late-childhood into adolescence, we see increasing volumes throughout all ROIs in both parcellations. Consistent with previous findings, we observed an anterior-posterior gradient in cerebellar development likely to reflect and mirror the age-related improvements in underlying functions, with sensorimotor areas predominantly located anteriorly and cognitive areas posteriorly in the cerebellum (King et al., 2019; Klein et al., 2016; Liu et al., 2022). Anterior sensorimotor areas show smaller age-related effects compared to posterior cognitive areas, possibly reflecting protracted growth trajectories for higher-order cognitive compared to sensorimotor regions in the cerebellum.
·Growth in functional cerebellar regions
·Growth in anatomical cerebellar regions
Conclusions:
We present models of cerebellar growth during childhood and adolescence, an important time period for brain development, based on a large, prospective population cohort. We find an anterior-posterior growth gradient mirroring the age-related improvements of underlying behavior and function. The anterior/sensorimotor-posterior/cognitive growth gradient follows a recently proposed functional gradient related to cognitive load as well as cerebral maturation patterns, thus providing evidence for directly related cerebello-cortical developmental trajectories. In recent years, the cerebellum has received increasing attention as a critical node in fundamental cognitive and emotional functions as well as brain development. The current openly accessible growth models will therefore be of great value for uncovering cerebellar deviations and understanding their implications in neuropathology.
Disorders of the Nervous System:
Neurodevelopmental/ Early Life (eg. ADHD, autism)
Lifespan Development:
Early life, Adolescence, Aging 1
Normal Brain Development: Fetus to Adolescence
Modeling and Analysis Methods:
Bayesian Modeling
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Neuroanatomy Other 2
Keywords:
Cerebellum
Development
Modeling
MRI
Open Data
1|2Indicates the priority used for review
Provide references using author date format
Kia, S. M., Huijsdens, H., Rutherford, S., de Boer, A., Dinga, R., Wolfers, T., ... & Marquand, A. F. (2022). Closing the life-cycle of normative modeling using federated hierarchical Bayesian regression. Plos one, 17(12), e0278776.
Constantino, J. N., Davis, S. A., Todd, R. D., Schindler, M. K., Gross, M. M., Brophy, S. L., ... & Reich, W. (2003). Validation of a brief quantitative measure of autistic traits: comparison of the social responsiveness scale with the autism diagnostic interview-revised. Journal of autism and developmental disorders, 33, 427-433.
Kooijman, M. N., Kruithof, C. J., van Duijn, C. M., Duijts, L., Franco, O. H., van IJzendoorn, M. H., ... & Jaddoe, V. W. (2016). The Generation R Study: design and cohort update 2017. European journal of epidemiology, 31, 1243-1264.
Chakravarty, M. M., Steadman, P., Van Eede, M. C., Calcott, R. D., Gu, V., Shaw, P., ... & Lerch, J. P. (2013). Performing label‐fusion‐based segmentation using multiple automatically generated templates. Human brain mapping, 34(10), 2635-2654.
Park, M. T. M., Pipitone, J., Baer, L. H., Winterburn, J. L., Shah, Y., Chavez, S., ... & Chakravarty, M. M. (2014). Derivation of high-resolution MRI atlases of the human cerebellum at 3 T and segmentation using multiple automatically generated templates. Neuroimage, 95, 217-231.
King, M., Hernandez-Castillo, C. R., Poldrack, R. A., Ivry, R. B., & Diedrichsen, J. (2019). Functional boundaries in the human cerebellum revealed by a multi-domain task battery. Nature neuroscience, 22(8), 1371-1378.
Klein, A. P., Ulmer, J. L., Quinet, S. A., Mathews, V., & Mark, L. P. (2016). Nonmotor functions of the cerebellum: an introduction. American Journal of Neuroradiology, 37(6), 1005-1009.
Liu, X., d’Oleire Uquillas, F., Viaene, A. N., Zhen, Z., & Gomez, J. (2022). A multifaceted gradient in human cerebellum of structural and functional development. Nature neuroscience, 25(9), 1129-1133.