Elsevier

NeuroImage

Volume 30, Issue 1, March 2006, Pages 184-202
NeuroImage

The NIH MRI study of normal brain development

https://doi.org/10.1016/j.neuroimage.2005.09.068Get rights and content

Abstract

MRI is increasingly used to study normal and abnormal brain development, but we lack a clear understanding of “normal”. Previous studies have been limited by small samples, narrow age ranges and few behavioral measures. This multi-center project conducted epidemiologically based recruitment of a large, demographically balanced sample across a wide age range, using strict exclusion factors and comprehensive clinical/behavioral measures.

A mixed cross-sectional and longitudinal design was used to create a MRI/clinical/behavioral database from approximately 500 children aged 7 days to 18 years to be shared with researchers and the clinical medicine community. Using a uniform acquisition protocol, data were collected at six Pediatric Study Centers and consolidated at a Data Coordinating Center. All data were transferred via a web-network into a MYSQL database that allowed (i) secure data transfer, (ii) automated MRI segmentation, (iii) correlation of neuroanatomical and clinical/behavioral variables as 3D statistical maps and (iv) remote interrogation and 3D viewing of database content.

A population-based epidemiologic sampling strategy minimizes bias and enhances generalizability of the results. Target accrual tables reflect the demographics of the U.S. population (2000 Census data). Enrolled subjects underwent a standardized protocol to characterize neurobehavioral and pubertal status. All subjects underwent multi-spectral structural MRI. In a subset, we acquired T1/T2 relaxometry, diffusion tensor imaging, single-voxel proton spectroscopy and spectroscopic imaging. In the first of three cycles, successful structural MRI data were acquired in 392 subjects aged 4:6–18:3 years and in 72 subjects aged 7 days to 4:6 years. We describe the methodologies of MRI data acquisition and analysis, using illustrative results.

This database will provide a basis for characterizing healthy brain maturation in relationship to behavior and serve as a source of control data for studies of childhood disorders. All data described here will be available to the scientific community from July, 2006.

Introduction

Magnetic resonance imaging (MRI) has made it possible to study normal structural and metabolic brain development across age groups. It had been difficult to study infants, children and adolescents with earlier imaging modalities because of safety concerns related to radiation exposure. Hence, relatively little was known about healthy brain development in humans prior to the advent of MRI.

In the 1990s, several research groups demonstrating age-related changes in gray matter volumes, white matter volumes, myelination and subcortical measures with MRI in samples of healthy children aged 4–21 years (Filipek et al., 1994, Jernigan and Tallal, 1990, Jernigan et al., 1991, Pfefferbaum et al., 1994). Subsequently, additional studies have described normal developmental changes in specific brain regions based on samples of children and young adults ranging in size from N = 13 to 176 (Bartzokis et al., 2001, Blanton et al., 2001, Blanton et al., 2004, Blatter et al., 1995, Caviness et al., 1996, Caviness et al., 1999, Courchesne et al., 2000, DeBellis et al., 2001, Giedd et al., 1996, Giedd et al., 1999, Gogtay et al., 2002, Gogtay et al., 2004, Kennedy et al., 1998, Kennedy et al., 2003, Lange et al., 1997, Paus et al., 1999, Reiss et al., 1996, Sowell et al., 1999, Sowell et al., 2002, Sowell et al., 2003, Sowell et al., 2004a). Several recent reviews have summarized this research (Durston et al., 2001, Gogtay et al., 2002, Paus et al., 2001, Sowell et al., 2004b) with respect to maturation and correlation with postmortem findings in infancy and childhood. It is clear that during early childhood and adolescence specific regional brain measures vary widely in healthy populations. Cross-sectional studies have therefore been limited in the conclusions they can reach about healthy brain development. Similarly, studies of pediatric brain disorders have been hampered by the lack of control data for healthy development. Longitudinal studies have noted that individual developmental brain growth trajectories are highly variable, regionally specific and may demonstrate gender-specific patterns (Giedd et al., 1996, Giedd et al., 1999). They highlight the need for large sample sizes in order to obtain reliable conclusions about the normative range for specific regional volumetric data and maturation patterns (Gogtay et al., 2004).

Although a few studies have demonstrated relationships between healthy regional brain structure maturation and specific cognitive abilities, this area of research remains largely untapped (Casey et al., 1997, Sowell et al., 2004b). Significant questions regarding healthy brain development remain. Previously reported studies of self-selected samples, obtained through a variety of media solicitations or samples of convenience, are not representative of U.S. demographics, particularly with respect to race, ethnicity and income. Thus, the generalizability of reported results is limited. Little data have been reported on children younger than age six, a population of particular interest because of rapid brain development during infancy and preschool years. Very few longitudinal studies have been conducted. Most have been limited to analysis of T1-weighted data only. The absence of cross-site collaborations or image acquisition standardization further limits our understanding of brain development and brain–behavior relationships and limits the generalizability or useful application of control data across research laboratories. As yet, no representative, longitudinally acquired database of healthy brain development that combines high quality multi-sequence, multi-modality MR images: anatomical MRI (aMRI), magnetic resonance spectroscopy (MRS) and diffusion tensor imaging (DTI) with comprehensive longitudinal clinical/neurobehavioral assessments is available.

One example of the impact of this knowledge gap can be seen in a recent report of differences in brain morphology between children with attention deficit hyperactivity disorder (ADHD) and healthy controls (Castellanos et al., 2002). This study reported subtle, yet widespread, reductions in regional brain volumes that were generally stable across development with the exception of early reductions in caudate volume which appeared to normalize during adolescence. Despite its utilization of one of the largest normal MRI data sets available (Giedd et al., 1999), this study was limited by a paucity of data for the younger ages, where reliable predictions were not possible. Fewer than 20 healthy male and 10 healthy female children were under 8 years of age in that study. Furthermore, the control data set had restricted socioeconomic representation and (high) IQ range (elevated relative to the ADHD sample). A more complete and representative database of MRI and cognitive/behavioral measures in a demographically diverse sample will greatly increase the power for detecting subtle, yet important, differences in brain developmental trajectories in childhood psychiatric and neurological disorders.

This study was undertaken in order to establish a public database of pediatric aMRI, MRS and DTI brain scans, for several purposes: (1) to elucidate healthy anatomic and metabolic brain development, providing ranges of normal values and defining key developmental periods; (2) to fill a need for a representative and reliable source of healthy control subject data for studies of childhood disorders and brain diseases to be made available to pediatric researchers and clinicians; (3) to provide data for the construction of healthy developmental growth curves for specific brain structures and metabolites; and (4) to aid the development of image analysis methods and diagnostic tools, e.g., the derivation of developmentally sensitive morphometric or metabolic imaging measures not obtainable with methods developed for adult populations.

A major goal was to recruit a demographically representative healthy sample. In order to meet this goal and to minimize ascertainment biases that can be present in samples of convenience, a population-based sampling method was employed. A sample of healthy infants, children and adolescents demographically representative of the U.S. population has been recruited and characterized. Cognitive, neuropsychological and behavioral measures were acquired to screen or exclude subjects as well as to provide a basis for brain–behavior correlational studies with the imaging data.

Collecting neuroanatomical and clinical/behavioral data from children ranging in age from 7 days to 18:3 years is challenging due to the impracticality of applying a common data acquisition protocol across these ages. Many clinical/behavioral measures are only suitable for a limited age range. Similarly, for MR imaging, the optimal scanning protocol depends upon tissue characteristics and practical considerations which vary with age (subject tolerance, motion). Accordingly, the primary goal of the project, the collection of structural MRI and behavior data, was organized as two “Objectives”. The larger Objective 1 includes children in the age range of 4:6–18:3 years, while the smaller Objective 2 includes children from 7 days to 4:6 years. Although the underlying goals of these Objectives are similar, the protocols for recruitment, screening, behavioral and cognitive characterization, MR scanning and sampling frequency differ substantially. Both Objectives employ a longitudinal study design: Objective 1 children are being scanned at 2-year intervals while Objective 2 children are being scanned at approximately quarterly intervals (Fig. 1, Fig. 2).

A comprehensive description of each component of this multi-faceted project within a single report would be prohibitively long. Nevertheless, an accurate representation of the overall project context requires that all elements be presented together. This report therefore outlines the overarching rationale and methodologies of the project, with emphasis on Objective 1. Subsequent reports will provide greater detail regarding other individual components (e.g., Objective 2, MRS, DTI) and their results.

Thus far, the project has enrolled a cohort of 433 Objective 1 subjects and 72 Objective 2 subjects and is following these subjects over a 7-year period. Data are collected at six pediatric study centers (PSCs):

  • Boston—Children's Hospital

  • Cincinnati—Children's Hospital Medical Center

  • Houston—University of Texas Houston Medical School

  • Los Angeles—Neuropsychiatric Institute and Hospital, UCLA

  • Philadelphia—Children's Hospital of Philadelphia (CHOP)

  • St. Louis—Washington University.

Informed consent from parents and adult subjects and child assents were obtained for all subjects enrolled at the PSCs. All protocols and procedures were approved by the relevant Institutional Review Board at each PSC and at each coordinating center.

A Clinical Coordinating Center (CCC) at Washington University St. Louis coordinates the clinical/behavioral aspects of the project including: sampling plan and methods; subject recruitment, inclusion/exclusion criteria and screening/assessment protocols; quality control (QC) for the administration of all clinical and behavioral measures. Structural MRI and clinical/behavioral data are consolidated and analyzed within a purpose-built database at a Data Coordinating Center (DCC) at the Montreal Neurological Institute, McGill University. The DCC coordinates the image acquisition protocols, imaging data quality control and image analysis. Diffusion tensor imaging (DTI) data are analyzed at a DTI Processing Center, National Institute of Child Health and Development (NICHD) NIH (DPC). Spectroscopy data are analyzed at a Spectroscopy Processing Center UCLA (SPC). All data, raw and processed, are eventually consolidated at the DCC.

Three separate repeated study cycles for each child allow both longitudinal and cross-sectional analysis. Imaging and clinical/behavioral data are transferred via a web-based network into a central database that allows for (i) secure, encrypted data transfer and automated quality control, (ii) automated large-scale MRI segmentation, (iii) correlation of neuroanatomical and clinical/behavioral variables as 3D statistical maps and (iv) remote interrogation and 3D viewing of database content.

Imaging data included structural MRI (T1-weighted, T2-weighted, proton-density-weighted). A subset of children had additional data acquisitions (T1/T2 relaxometry, DTI, MRS and MRS imaging (MRSI)). In the following sections, we describe the methodologies for sampling and recruitment, clinical and behavioral assessment, MRI data acquisition, database design, MR segmentation and Data analysis.

Section snippets

Population-based sampling

The population-based sampling method used in this study seeks to minimize biases that can be present in samples of convenience in order to maximize the generalizability of the data collected. The sampling plan for each PSC was developed from the available Census 2000 data, which allowed neighborhood demographic variables to be estimated for the corresponding zip codes (so called geocoding). This allowed targeted recruitment and comparison to the general population by reference to geocoded

Identification, screening and exclusions

(For a complete description of the screening procedures, please see website http://www.bic.mni.mcgill.ca/nihpd/info). A total of 35,429 introductory letters were mailed to families over an 18-month enrollment period, and 28,265 were successfully contacted. Approximately 8% of households contacted did not have a child in the age range of interest living in the area, and 13.5% spontaneously identified one or more exclusion factors prior to structured screening. Approximately 35% were not

Discussion

Structural MRI provides a means to determine healthy growth patterns for regional gray and white matter and various specific brain structures of interest. MRS provides a safe, noninvasive method for establishing a developmental profile of brain metabolites. DTI uses water diffusivity to probe structural and architectural changes occurring in brain tissue during development. The integration of information from these multiple modalities may generate a significant increase in the understanding of

Acknowledgments

This project is supported by the National Institute of Child Health and Human Development (Contract N01-HD02-3343), the National Institute on Drug Abuse, the National Institute of Mental Health (Contract N01-MH9-0002), and the National Institute of Neurological Disorders and Stroke (Contracts N01-NS-9-2314, -2315, -2316, -2317, -2319 and -2320).

We wish to acknowledge the important contribution and remarkable spirit of John Haselgrove, Ph.D. (deceased) who contributed enormously to this project.

References (105)

  • N. Kabani et al.

    Measurement of cortical thickness using an automated 3D algorithm: a validation study

    NeuroImage

    (2001)
  • J.S. Kim et al.

    Automated 3D extraction and evaluation of the outer cortical surface using a Laplacian map and partial volume effect classification

    NeuroImage

    (2005)
  • N.T. Lange et al.

    Variability of human brain structure size: ages 4–20 years

    Psychiatry Res.

    (1997)
  • J. Lerch et al.

    Cortical thickness analysis examined through power analysis and a population simulation

    NeuroImage

    (2005)
  • C. Liao et al.

    Estimating the delay of the hemodynamic response in fmri data

    NeuroImage

    (2002)
  • C.P. Lucas et al.

    The DISC Predictive Scales (DPS): efficiently screening for diagnoses

    J. Am. Acad. Child Adolesc. Psych.

    (2001)
  • T. Paus et al.

    Maturation of white matter in the human brain: a review of magnetic resonance studies

    Brain Res. Bull.

    (2001)
  • D. Shaffer et al.

    NIMH Diagnostic Interview Schedule for Children Version IV (NIMH DISC-IV): description, differences from previous versions, and reliability of some common diagnoses

    J. Am. Acad. Child Adolesc. Psych.

    (2000)
  • E.R. Sowell et al.

    Localizing age-related changes in brain structure between childhood and adolescence using statistical parametric mapping

    NeuroImage

    (1999)
  • J. Tohka et al.

    Fast and robust parameter estimation for statistical partial volume models in brain MRI

    NeuroImage

    (2004)
  • K.J. Worsley et al.

    Analysis of fMRI time-series revisited-again

    NeuroImage

    (1995)
  • K.J. Worsley et al.

    Tests for distributed, non-focal brain activations

    NeuroImage

    (1995)
  • T.M. Achenbach

    Manual for the Young Adult Self-report and Young Adult Behavior Checklist

    (1997)
  • T.M. Achenbach

    Manual for the ASEBA School-Age Forms and Profiles (Child Behavior Checklist for Ages 6–18)

    (2001)
  • T.M. Achenbach et al.

    Manual for the ASEBA Preschool Forms and Profiles (Child Behavior Checklist for Ages 1 1/2–5)

    (2000)
  • D.C. Alexander et al.

    Spatial transformations of diffusion tensor magnetic resonance images

    IEEE Trans. Med. Imaging

    (2001)
  • C.R. Almli

    Measures of Hand Preference and Use Appropriate for Infants and Children

    (1999)
  • G. Bartzokis et al.

    Age-related changes in frontal and temporal lobe volumes in men: a magnetic resonance imaging study

    Arch. Gen. Psychiatry

    (2001)
  • D.D. Blatter et al.

    Quantitative volumetric analysis of brain MR: normative database spanning 5 decades of life

    Am. J. Neuroradiol.

    (1995)
  • J. Cao

    The size of the connected components of excursion sets of χ2, t and f fields

    Adv. Appl. Probab.

    (1999)
  • J. Cao et al.

    The geometry of correlation fields, with an application to functional connectivity of the brain

    Ann. Appl. Probab.

    (1999)
  • B.J. Casey et al.

    The role of the anterior cingulate in automatic and controlled processes: a developmental neuroanatomical study

    Dev. Psychobiol.

    (1997)
  • F.X. Castellanos et al.

    Developmental trajectories of brain volume abnormalities in children and adolescents with attention-deficit/hyperactivity disorder

    J. Am. Med. Assoc.

    (2002)
  • V.S. Caviness et al.

    The human brain age 7–11 years: a volumetric analysis based on magnetic resonance images

    Cereb. Cortex

    (1996)
  • V.S. Caviness et al.

    MRI-based brain volumetrics: emergence of a developmental brain science

    Brain Dev.

    (1999)
  • CeNeS Limited

    Cambridge Neuropsychological Test Automated Battery (CANTAB) for Windows Cambridge

    (1999)
  • L.-C. Chang et al.

    RESTORE: robust estimation of tensors by outlier rejection

  • C.R. Cloninger et al.

    A psychobiological model of temperament and character

    Arch. Gen. Psychiatry

    (1993)
  • C.R. Cloninger et al.

    The Temperament and Character Inventory (TCI): A Guide to Its Development and Use

    (1994)
  • D.L. Collins et al.

    Automatic 3D registration of MR volumetric data in standardized Talairach space

    J. Comput. Assist. Tomogr.

    (1994)
  • D.L. Collins et al.

    Automatic 3D model-based neuroanatomical segmentation

    Hum. Brain Mapp.

    (1995)
  • E. Courchesne et al.

    Normal brain development and aging: quantitative analysis at in vivo imaging in healthy volunteers

    Radiology

    (2000)
  • M.D. DeBellis et al.

    Sex differences in brain maturation during childhood and adolescence

    Cereb. Cortex

    (2001)
  • D.C. Delis et al.

    The California Verbal Learning Test-Children's Version

    (1994)
  • D.C. Delis et al.

    California Verbal Learning Test-II

    (2000)
  • C.D. Elliott

    DAS Administration and Scoring Manual

    (1990)
  • C.D. Elliott

    DAS Introductory and Technical Handbook

    (1990)
  • C.D. Elliott

    Differential Ability Scales

    (1990)
  • P.A. Filipek et al.

    The young adult human brain: an MRI-based morphometric analysis

    Cereb. Cortex

    (1994)
  • R.A. Gardner et al.

    The Purdue Pegboard: normative data on 1334 schoolchildren

    J. Clin. Psychol.

    (1979)
  • Cited by (0)

    Researchers who are interested in using the database resulting from this project are encouraged to request the protocols by e-mailing [email protected].

    1

    See Appendix A for author list.

    View full text