Skip to main content
NCBI home page
Neurology logoLink to Neurology
. 2018 Apr 24;90(17):e1493–e1500. doi: 10.1212/WNL.0000000000005352

Cerebellar volume as an imaging marker of development in infants with tuberous sclerosis complex

Siddharth Srivastava 1,*, Anna K Prohl 1,*, Benoit Scherrer 1, Kush Kapur 1, Darcy A Krueger 1, Simon K Warfield 1, Mustafa Sahin 1,✉; On behalf of the TACERN Study Group1
PMCID: PMC5921037  PMID: 29572283

Abstract

Objective

In this cohort analysis, we studied 1-year-old infants with tuberous sclerosis complex (TSC), correlating volumes of cerebellar structures with neurodevelopmental measures.

Methods

We analyzed data from a prospective biomarker study in infants with TSC (ClinicalTrials.gov NCT01780441). We included participants aged 12 months with an identified mutation of TSC1 or TSC2. Using MRI segmentation performed with the PSTAPLE algorithm, we measured relative volumes (structure volume divided by intracranial contents volume) of the following structures: right/left cerebellar white matter, right/left cerebellar exterior, vermal lobules I–V, vermal lobules VI–VII, and vermal lobules VIII–X. We correlated relative volumes to Mullen Scales of Early Learning (MSEL) scores.

Results

There were 70 participants (mean age 1.03 [0.11] years): n = 11 had a TSC1 mutation; n = 59 had a TSC2 mutation. For patients with TSC2 mutation, for every percentage increase in total cerebellar volume, there was an approximate 10-point increase in MSEL composite score (β = 10.47 [95% confidence interval 5.67, 15.27], p < 0.001). For patients with TSC1 mutation, the relationship between cerebellar volume and MSEL composite score was not statistically significant (β = −10.88 [95% confidence interval −22.16, 0.41], p = 0.06). For patients with TSC2 mutation, there were positive slopes when regressing expressive language and visual reception skills with volumes of nearly all cerebellar structures (p ≤ 0.29); there were also positive slopes when regressing receptive language skills, gross motor skills, and fine motor skills with volumes of cerebellar right/left exterior (p ≤ 0.014).

Conclusions

Cerebellar volume loss—perhaps reflecting Purkinje cell degeneration—may predict neurodevelopmental severity in patients with TSC2 mutations.

Tuberous sclerosis complex (TSC) is a genetic disorder caused by defects in TSC1 (hamartin) or TSC2 (tuberin), resulting in hamartomas and other abnormalities in multiple organ systems, including the brain, eyes, heart, lungs, kidneys, and skin.1 In the CNS, gross structural manifestations span cortical dysplasias, subependymal nodules, and subependymal giant cell astrocytomas.2 Microstructural alterations in the brain, such as hypomyelination and axonal loss, are also apparent.3–5

Neurocognition is often negatively affected in TSC, as patients can have variable presentations of epilepsy, intellectual disability (ID), autism spectrum disorder, and other neurodevelopmental impairments. Approximately 30% to 40% of patients have a history of infantile spasms.6,7 In one study regarding cognition, 14% of patients had mild to severe ID, 31% had profound ID, and 55% had normal intelligence,8 constituting a wide spectrum. Autism spectrum disorder is a common comorbidity, affecting close to 40% of patients.9 Finally, other neuropsychiatric disturbances can occur, such as attention-deficit/hyperactivity disorder, anxiety, and depression.10,11 This constellation of cognitive, behavioral, and psychiatric impairments in TSC is collectively known as TAND (TSC-associated neuropsychiatric disorders).10

One of the possible mechanisms of neurodevelopmental dysfunction in TSC is overactivation of the mechanistic target of rapamycin (mTOR) pathway leading to cerebellar Purkinje cell degeneration. TSC1 and TSC2, along with TBC1D7, form a protein complex that negatively regulates mTOR activity through the action of Ras homolog enriched in brain (RHEB).12,13 Therefore, mutations in either TSC1 or TSC2 disinhibit mTOR signaling, which can have downstream repercussions on neuronal and synaptic growth and formation.14,15 Animal models of TSC have suggested that mTOR signaling is necessary for the survival of Purkinje neurons.16–18 Consistent with this notion of decreased Purkinje cell survival in TSC, a prior MRI study on mostly older children and adults with TSC identified reductions in cerebellum volume in patients with TSC compared to controls.19

In this study, we have extended previous findings by not only performing volumetric analysis of the cerebellum using MRI data from infants with TSC, but also comparing these metrics with objective developmental measures captured at the same time as MRI. We hypothesize that degree of cerebellar volume loss in TSC is directly related to severity of neurodevelopmental impairment.

Methods

Standard protocol approvals, registrations, and patient consents

Study patients were part of a multicenter, prospective cohort study designed to evaluate biomarkers in infants with TSC (ClinicalTrials.gov NCT01780441). Participants were eligible if they were 3 to 12 months of age at time of enrollment and if they met genetic or clinical diagnostic criteria for TSC. Five TSC centers throughout the United States (Boston, MA; Birmingham, AL; Cincinnati, OH; Houston, TX; Los Angeles, CA) participated in the trial. Participants underwent longitudinal evaluations (consisting of brain MRI, EEG, and cognitive assessments) at regular intervals through 36 months of age. The protocol for this study was approved by the institutional review board at each institution with Cincinnati Children's Hospital Medical Center being the lead regulatory site. Informed consent was obtained for all participants through their parents or legal guardians. The study was conducted in accordance with Good Clinical Practice guidelines. For this report, we included data for analysis from approximately the 12 months of age time point (range 0.7–1.3 years), and we included only those patients with a pathogenic variant in TSC1 or TSC2.

MRI acquisition

At all sites, patients had brain MRIs on 3-tesla GE, Philips, and Siemens scanners at baseline, 12 months, 24 months, and 36 months of age. The imaging protocol was a consensus, high spatial resolution, clinical imaging protocol that included a 1-mm3, sagittal, T1-weighted magnetization-prepared rapid-acquisition gradient echo and a T2-weighted fast spin echo covering the entire brain. Patients had sedation when clinically indicated. Traveling human phantoms, performed annually at each center, were the basis of intra- and intersite accuracy and reliability of volumetric measurements.

MRI processing

We used the Computational Radiology Kit (crl.med.harvard.edu/) to complete all MRI processing and analysis. For each patient, the T2-weighted image was aligned to the T1-weighted scan using rigid registration with mutual information metric. The intracranial cavity (ICC) was then extracted using a previously validated multispectral ICC segmentation method,20 and the resulting brain image was parcellated in various cerebellar regions of interest (ROIs) for volumetric analysis. To circumvent the bias and inter-/intraexpert variability inherent to manually drawn ROIs, we used a fully automatic, multi-template MRI parcellation approach. It was based on robust, nonrigid registration of a set of templates and on multi-template fusion to identify consensus ROIs in the individual patient. The template library was composed of 15 T1-weighted healthy control MRIs. MRIs were hand-labeled by expert neuroanatomists with anatomical boundary definitions defined by well-established MRI brain labeling protocols21,22 and with test-retest reproducibility quantified. We performed multi-template fusion using the local MAP PSTAPLE (PSTAPLE) algorithm23 which uses both the labels and intensity profiles of the templates to compute probability maps for each target structure, ultimately leading to an automatic consensus labeling of each patient brain. See figure 1 for an illustrative example of cerebellar parcellation on 3 of the participants.

Figure 1. Cerebellar parcellations for some of the participants.

Figure 1

Open in a new tab

Sagittal (A), coronal (B), and axial (C) T1-weighted images demonstrating color-coded, parcellated cerebellar structures in 3 of the participants.

We considered the following cerebellar structures: right and left cerebellar white matter, right and left cerebellar exterior, cerebellar vermal lobules I–V, cerebellar vermal lobules VI–VII, and cerebellar vermal lobules VIII–X. We measured their volumes by counting their number of labeled voxels and then by multiplying by the size of each voxel, thus obtaining the volume in units of cubic millimeters. The volumes were normalized by dividing by the volume of the ICC of each brain, providing relative volume descriptors for each structure.

Cognitive assessment

At each study visit, participants underwent a battery of neuropsychological assessments. For this report, we analyzed data from the Mullen Scales of Early Learning (MSEL) obtained from participants 12 months of age.24 This test evaluates cognition in children 0 to 5 years of age, providing an overall developmental quotient (standard score) and subtest scores (T scores) for receptive language, expressive language, visual reception, gross motor skills, and fine motor skills.

Statistical analysis

We used the Wilcoxon rank sum test to compare quantitative variables between genotypes. We used Pearson χ2 test to compare binomial variables between genotypes. When comparing effect of relative volume on Mullen scores, we used the following linear regression model:

Score = β0 + β1 × volume + β2 × genotype + β3 × genotype × volume

where Inline graphic is the MSEL subdomain or composite score, Inline graphic is the relative volume of the cerebellar structure multiplied by 100, and Inline graphic is a binary variable indicating whether a patient has a pathogenic TSC1 or pathogenic TSC2 variant. We used the Benjamini-Hochberg false discovery rate procedure, which controls for expected proportion of false discoveries relative to total discoveries, to control for multiple comparisons with q = 0.05. Participants who had missing data for a specific regression analysis were not part of that specific analysis.

Results

As of March 2017, 163 individuals were enrolled in the original study. Of these, 93 had brain MRI at approximately 1 year of age (0.7–1.3 years). From these, there were 70 participants with an identified pathogenic TSC1 or TSC2 variant, constituting the cohort we used for analysis. The cohort had a mean age of 1.03 (0.11) years. Approximately half of the entire cohort was male. Eleven participants had a pathogenic TSC1 variant, and 59 participants had a pathogenic TSC2 variant. Six participants with pathogenic TSC2 variants had missing MSEL composite scores, including one participant who also had missing receptive language, visual reception, gross motor, and fine motor scores (but not expressive language score). Approximately 70% of the entire cohort had a history of seizures (either focal or generalized) by the time of the MRI, with a higher frequency in those with a pathogenic TSC2 variant. By the time of the MRI, epileptic spasms occurred in 18.18% of the TSC1 individuals and 57.63% of the TSC2 individuals (table 1).

Table 1.

Demographic characteristics of the cohort stratified by genotype

graphic file with name NEUROLOGY2017844340TT1.jpg

Open in a new tab

Relative volume of the entire cerebellum was not significantly different when comparing patients with pathogenic TSC1 and TSC2 variants. Likewise, there were no statistically significant differences in relative volumes of the cerebellar exterior, white matter, or vermis structures when comparing between genotypes (table 2).

Table 2.

Relative volumes of cerebellar structures stratified by genotype

graphic file with name NEUROLOGY2017844340TT2.jpg

Open in a new tab

In patients with pathogenic TSC2 variants, for every single percentage increase in relative volume of the entire cerebellum, there was a nearly 10-point increase in MSEL composite score (β = 10.47, 95% confidence interval [CI] 5.67, 15.27; p < 0.001). In contrast, for patients with pathogenic TSC1 variants, the linear regression between relative volume of the entire cerebellum and MSEL composite score was not statistically significant (β = −10.88 [95% CI −22.16, 0.41], p = 0.06) (figure 2). In an analysis of patients with pathogenic TSC2 variants, there was a difference in relative volume of the entire cerebellum for those who had seizures by the time of MRI (β = −0.81 [95% CI −1.43, −0.19], p = 0.012), and there was also a difference in MSEL composite score for those who had seizures by the time of MRI (β = −18.61 [95% CI −30.53, −6.69], p = 0.003).

Figure 2. Linear regressions of Mullen composite scores vs relative volumes of cerebellum.

Figure 2

Open in a new tab

Scatterplots and fitted lines of MSEL composite score vs relative volume of entire cerebellum, stratified by genotype. MSEL = Mullen Scales of Early Learning.

In a post hoc subgroup analysis focusing on patients with pathogenic TSC2 variants, we compared MSEL composite and subdomain scores to relative volumes of cerebellar structures, accounting for multiple comparisons. MSEL composite scores increased with higher relative volumes of all cerebellar structures (p ≤ 0.022 for slope in each model). MSEL expressive language scores increased with higher relative volumes of all cerebellar structures (p ≤ 0.029 for slope in each model) except the right cerebellar white matter. Likewise, MSEL visual reception scores increased with increased relative volumes of all cerebellar structures (p ≤ 0.007 for slope in each model) except for cerebellar vermal lobules I–V and VIII–X. Finally, MSEL receptive language, gross motor, and fine motor scores increased with higher relative volumes of the right/left cerebellar exterior and entire cerebellum (p ≤ 0.018 for slope in each model). MSEL fine motor scores also increased with increased relative volume of the cerebellar vermal lobules VI–VII (p = 0.022) (tables e-1 to e-6, links.lww.com/WNL/A384).

Discussion

Our study examined cerebellar volume in relation to developmental measures in TSC. Prior work demonstrated that total cerebellar volume was decreased in patients with TSC compared to controls, particularly for those with TSC2 mutations.19 However, the previous study was limited by the relatively small number and wide age range of study participants (n = 36, ages 1–27 years). Our cohort not only included a larger number of participants but also focused on infants approximately 1 year of age.

In our sample, although relative volume of the entire cerebellum was not significantly different between genotypes, individuals with pathogenic TSC2 variants demonstrated higher scores on all domains of the MSEL with increasing relative volume of the entire cerebellum. Furthermore, these same individuals had better expressive language and visual reception abilities associated with increased relative volumes of nearly all of the cerebellar structures. Better receptive language, gross motor, and fine motor abilities were related to increased relative volumes of the cerebellar exterior. Overall, the data suggest that smaller cerebellar size—perhaps reflecting degree of Purkinje cell degeneration—may be predictive of worse neurodevelopmental outcomes in TSC for patients with pathogenic TSC2 variants.

It is of interest that for patients with pathogenic TSC1 variants, there was an opposite, negative slope in the linear regression between MSEL composite score and relative volume of the entire cerebellum. However, there were only 11 participants in this subgroup, and the regression was not statistically significant. Thus, larger numbers of patients with pathogenic TSC1 variants are needed to better understand how cerebellar volume relates to neurodevelopmental outcomes for this specific genotype. Of note, compared to patients with pathogenic TSC2 variants, patients with pathogenic TSC1 variants had a higher mean MSEL composite score (93.45 vs 82.43) and a higher relative volume of the entire cerebellum (9.09 vs 8.86). Although these differences were not statistically significant, possibly in light of the lower number of TSC1 participants, they are consistent with prior genotype-phenotype correlation studies suggesting that patients with TSC1 are less severely affected compared to their TSC2 counterparts.25 In TSC, involvement of the cerebellum, in particular Purkinje cell degeneration, is an area of intense interest, based on insights from animal model data and human studies. Mice with selective deletion of Tsc2 from Purkinje cells demonstrate not only functional motor deficits but also progressive Purkinje cell death, which improves with rapamycin administration.26 Mice with Tsc1 deletion from Purkinje cells also exhibit decreased Purkinje cell survival, as well as autistic-like behaviors, such as impaired socialization, repetitive behavior, and abnormal vocalizations.17 Humans affected by the disorder have shown corresponding abnormalities in Purkinje cell loss. In a postmortem histologic analysis of 4 human participants with TSC compared to age-matched controls, Purkinje cell density was decreased in 2 of the patients.26

The cerebellum has a number of important roles in neurodevelopment that may explain why cerebellar volume loss is related to language and cognitive deficits in TSC. Although traditionally viewed in relation to motor functioning, the cerebellum is also implicated in language tasks, including verbal fluency and semantic retrieval,27 as well as cognition, including overall intellect, visuospatial performance, executive functioning, working memory, and processing speed.28 Consistent with this notion, children with various cerebellar malformations have reductions in total cerebellar volume that are correlated with impairments in motor abilities, expressive language skills, and cognition.29

In the subgroup of patients with pathogenic TSC2 variants, some of the relationships between volumes of different cerebellar structures and types of developmental outcomes may reflect cerebellar functional topography. For example, based on a meta-analysis of neuroimaging studies, the part of the cerebellum involved in cognition may be the posterior lobe, particularly lobules VI and VII.30 Indeed, in our TSC2 variant subgroup, visual reception scores correlated with volume of vermal lobules VI–VII (but not vermal lobules I–V or VIII–X). The cerebellar hemispheres have a role in lateral motor systems and motor planning31: in addition to receiving projections from the cerebral cortex through corticopontine fibers that synapse with pontocerebellar fibers, the cerebellar hemispheres deliver output to the ventrolateral nucleus of the thalamus, which connects to the primary motor cerebral cortex.32 It is not surprising, then, that in the subgroup analysis, motor scores correlated with volumes of the cerebellar exterior, which includes the cerebellar hemisphere except for the white matter.

Altogether, our data suggest that cerebellar volume may be a potential neuroimaging biomarker associated with TSC. There have been many attempts to elucidate biomarkers of disease severity in TSC. Such biomarkers have valuable roles, not only for use in clinical trials targeting neurocognition in TSC, but also for providing guidance to families of affected children. For example, in the case of a newly diagnosed infant with TSC who has undergone brain MRI but who has not yet had developmental testing, volumetric analysis of the MRI may provide a more refined prognosis than by either tuber burden or genotype alone.

Our study has a few limitations that warrant further consideration. First, a limited number of patients had TSC1 mutations, making it difficult to draw conclusions about neuroimaging findings and cognitive outcomes for this subpopulation of TSC. Second, while the MSEL is a suitable measure of developmental skills in the infant population, it has certain disadvantages. Given the age range of the studied population, floor effects can occur. Likewise, minor increments or decrements in age at testing can lead to disproportionate changes in the outcomes. Third, we did not assess long-term developmental outcomes in this analysis. Developmental trajectories in TSC may change over time33; therefore, correlation between volumetric measures and later cognitive outcomes requires further investigation. Finally, based on our sample size and high prevalence of epilepsy in our cohort, we are unable to further investigate the independent relationship of seizures in the causal pathway of relative volume of entire cerebellum and MSEL composite score. Limitations aside, our study implores further investigation into the role of the cerebellum in the pathogenesis of TSC-related neurodevelopmental dysfunction, particularly for pathogenic TSC2 variants.

Acknowledgment

The authors are sincerely indebted to the families and patients in TSC clinics across the United States who generously contributed their time and effort to this study. The authors thank the Tuberous Sclerosis Alliance for their continued support in TSC research.

Glossary

CI

confidence interval

ICC

intracranial cavity

ID

intellectual disability

MSEL

Mullen Scales of Early Learning

mTOR

mechanistic target of rapamycin

ROI

region of interest

TSC

tuberous sclerosis complex

Contributor Information

Collaborators: TACERN Study Group, M Sahin, D Krueger, M Bebin, Joyce Wu, H Northrup, S Warfield, J Peters, B Scherrer, M Goyal, R Filip-Dhima, K Dies, S Bruns, E Hanson, A Walsh, N2 Bing, B Kent, S O’Kelley, M William, D Pearson, R Mansour, M Valley, R Gerhardt, M Griffith, J Krefting, A Perez, E Salazar, G Cutter, S Roberds, JA Nakagawa, L Mamounas, A Kau, and B Scherrer

Author contributions

Dr. Srivastava: drafted the manuscript, analyzed the data. Ms. Prohl, Dr. Scherrer: collected and analyzed the data, performed critical revision of the manuscript for important intellectual content. Dr. Kapur: provided statistical analysis, performed critical revision of the manuscript for important intellectual content. Dr. Warfield, Dr. Krueger, Dr. Sahin: performed critical revision of the manuscript for important intellectual content, oversaw study concept and design.

Study funding

S.S. was supported by a T32 grant from the NIH (4T32GM007748-38). S.K.W. was supported by the NIH (R01 NS079788). Support to M.S. and D.K. was provided by the NIH (U01-NS082320, U54-HD090255), the Tuberous Sclerosis Alliance, and the Developmental Synaptopathies Consortium (U54NS092090), which is a part of the National Center for Advancing Translational Sciences (NCATS) Rare Diseases Clinical Research Network (RDCRN). RDCRN is an initiative of the Office of Rare Diseases Research, NCATS, funded through collaboration among NCATS, National Institute of Mental Health (NIMH), National Institute of Neurological Disorders and Stroke (NINDS), and National Institute of Child Health and Human Development (NICHD). In addition, this study utilized additional resources supported by NCATS (UL1 TR001425).

Disclosure

S. Srivastava, A. Prohl, B. Scherrer, and K. Kapur report no disclosures relevant to the manuscript. D. Krueger has received consulting fees and research support from Novartis Pharmaceuticals, consulting fees from Mallinckrodt Pharmaceuticals, Advance Medical, and Axis Media. S. Warfield reports no disclosures relevant to the manuscript. M. Sahin has received research support from Roche, Novartis, Pfizer, LAM Therapeutics, Neuren, Ibsen, and Rugen unrelated to this study. Go to Neurology.org/N for full disclosures.

References

  • 1.Curatolo P, Bombardieri R, Jozwiak S. Tuberous sclerosis. Lancet 2008;372:657–668. [DOI] [PubMed] [Google Scholar]
  • 2.Mizuguchi M, Takashima S. Neuropathology of tuberous sclerosis. Brain Dev 2001;23:508–515. [DOI] [PubMed] [Google Scholar]
  • 3.Krishnan ML, Commowick O, Jeste SS, et al. Diffusion features of white matter in tuberous sclerosis with tractography. Pediatr Neurol 2010;42:101–106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Peters JM, Sahin M, Vogel-Farley VK, et al. Loss of white matter microstructural integrity is associated with adverse neurological outcome in tuberous sclerosis complex. Acad Radiol 2012;19:17–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Ruppe V, Dilsiz P, Reiss CS, et al. Developmental brain abnormalities in tuberous sclerosis complex: a comparative tissue analysis of cortical tubers and perituberal cortex. Epilepsia 2014;55:539–550. [DOI] [PubMed] [Google Scholar]
  • 6.Chu-Shore CJ, Major P, Camposano S, Muzykewicz D, Thiele EA. The natural history of epilepsy in tuberous sclerosis complex. Epilepsia 2010;51:1236–1241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Wilbur C, Sanguansermsri C, Chable H, et al. Manifestations of tuberous sclerosis complex: the experience of a provincial clinic. Can J Neurol Sci 2017;44:35–43. [DOI] [PubMed] [Google Scholar]
  • 8.Joinson C, O’Callaghan FJ, Osborne JP, Martyn C, Harris T, Bolton PF. Learning disability and epilepsy in an epidemiological sample of individuals with tuberous sclerosis complex. Psychol Med 2003;33:335–344. [DOI] [PubMed] [Google Scholar]
  • 9.Richards C, Jones C, Groves L, Moss J, Oliver C. Prevalence of autism spectrum disorder phenomenology in genetic disorders: a systematic review and meta-analysis. Lancet Psychiatry 2015;2:909–916. [DOI] [PubMed] [Google Scholar]
  • 10.de Vries PJ, Whittemore VH, Leclezio L, et al. Tuberous sclerosis associated neuropsychiatric disorders (TAND) and the TAND Checklist. Pediatr Neurol 2015;52:25–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Prather P, de Vries PJ. Behavioral and cognitive aspects of tuberous sclerosis complex. J Child Neurol 2004;19:666–674. [DOI] [PubMed] [Google Scholar]
  • 12.Inoki K, Li Y, Xu T, Guan KL. Rheb GTPase is a direct target of TSC2 GAP activity and regulates mTOR signaling. Genes Dev 2003;17:1829–1834. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Dibble CC, Elis W, Menon S, et al. TBC1D7 is a third subunit of the TSC1-TSC2 complex upstream of mTORC1. Mol Cell 2012;47:535–546. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Jaworski J, Sheng M. The growing role of mTOR in neuronal development and plasticity. Mol Neurobiol 2006;34:205–219. [DOI] [PubMed] [Google Scholar]
  • 15.Switon K, Kotulska K, Janusz-Kaminska A, Zmorzynska J, Jaworski J. Molecular neurobiology of mTOR. Neuroscience 2017;341:112–153. [DOI] [PubMed] [Google Scholar]
  • 16.Angliker N, Burri M, Zaichuk M, Fritschy JM, Rüegg MA. mTORC1 and mTORC2 have largely distinct functions in Purkinje cells. Eur J Neurosci 2015;42:2595–2612. [DOI] [PubMed] [Google Scholar]
  • 17.Tsai PT, Hull C, Chu Y, et al. Autistic-like behaviour and cerebellar dysfunction in Purkinje cell Tsc1 mutant mice. Nature 2012;488:647–651. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Reith RM, Way S, McKenna J, Haines K, Gambello MJ. Loss of the tuberous sclerosis complex protein tuberin causes Purkinje cell degeneration. Neurobiol Dis 2011;43:113–122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Weisenfeld NI, Peters JM, Tsai PT, et al. A magnetic resonance imaging study of cerebellar volume in tuberous sclerosis complex. Pediatr Neurol 2013;48:105–110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Grau V, Mewes AUJ, Alcañiz M, Kikinis R, Warfield SK. Improved watershed transform for medical image segmentation using prior information. IEEE Trans Med Imaging 2004;23:447–458. [DOI] [PubMed] [Google Scholar]
  • 21.Strudwick Caviness V, Theodore Lange N, Makris N, Reed Herbert M, Nelson Kennedy D. MRI-based brain volumetrics: emergence of a developmental brain science. Brain Dev 1999;21:289–295. [DOI] [PubMed] [Google Scholar]
  • 22.Klein A, Tourville J. 101 labeled brain images and a consistent human cortical labeling protocol. Front Neurosci 2012;6:171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Akhondi-Asl A, Warfield SK. Simultaneous truth and performance level estimation through fusion of probabilistic segmentations. IEEE Trans Med Imaging 2013;32:1840–1852. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Mullen E. Mullen Scale of Early Learning. Circle Pines, MN: American Guidance Service, Inc.; 1995. [Google Scholar]
  • 25.Au KS, Williams AT, Roach ES, et al. Genotype/phenotype correlation in 325 individuals referred for a diagnosis of tuberous sclerosis complex in the United States. Genet Med 2007;9:88–100. [DOI] [PubMed] [Google Scholar]
  • 26.Reith RM, McKenna J, Wu H, et al. Loss of Tsc2 in Purkinje cells is associated with autistic-like behavior in a mouse model of tuberous sclerosis complex. Neurobiol Dis 2013;51:93–103. [DOI] [PubMed] [Google Scholar]
  • 27.Highnam CL, Bleile KM. Language in the cerebellum. Am J Speech Lang Pathol 2011;20:337–347. [DOI] [PubMed] [Google Scholar]
  • 28.Salman MS, Tsai P. The role of the pediatric cerebellum in motor functions, cognition, and behavior: a clinical perspective. Neuroimaging Clin N Am 2016;26:317–329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Bolduc ME, du Plessis AJ, Sullivan N, et al. Regional cerebellar volumes predict functional outcome in children with cerebellar malformations. Cerebellum 2012;11:531–542. [DOI] [PubMed] [Google Scholar]
  • 30.Stoodley CJ, Schmahmann JD. Functional topography in the human cerebellum: a meta-analysis of neuroimaging studies. Neuroimage 2009;44:489–501. [DOI] [PubMed] [Google Scholar]
  • 31.Proville RD, Spolidoro M, Guyon N, et al. Cerebellum involvement in cortical sensorimotor circuits for the control of voluntary movements. Nat Neurosci 2014;17:1233–1239. [DOI] [PubMed] [Google Scholar]
  • 32.Siegel A, Sapru HN. Essential Neuroscience. Baltimore: Lippincott Williams & Wilkins; 2010. [Google Scholar]
  • 33.Jeste SS, Wu JY, Senturk D, et al. Early developmental trajectories associated with ASD in infants with tuberous sclerosis complex. Neurology 2014;83:160–168. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Neurology are provided here courtesy of American Academy of Neurology

RESOURCES