Quantitative White Matter Analysis of Early Brain Development in Autism

Principal Investigator: Guido Gerig 
Funding Agency: National Alliance for Autism Research 
Agency Number: 1212/GG/01-201-005-00-00

Abstract

There is converging evidence (from post-mortem studies, head circumference measurements and from Magnetic Resonance Imaging (MRI) studies) that early brain growth in autism might follow a different longitudinal trajectory and pattern than in healthy controls. Non-invasive neuroimaging techniques, in particular MRI and Diffusion Tensor Imaging (DTI) seem especially suited to study the morphology of brain growth and brain circuits associated with cognitive and behavioral deficits. UNC Chapel Hill has collected a unique sample of neuroimaging data of young children. This autism study (P.I. Prof. Joseph Piven) includes 60 autistic and 25 control subjects at two-years with follow-up at 4 years (still in progress). Studying this early age group with follow-up presents an excellent database to explore cross-sectional differences and differences in growth trajectories. Image data of this age group is significantly different from adult brain MRI, especially in regard to not fully developed myelination of white matter and the rapid change and large variability of the shape and size of brain structures. The neuroimaging research team at UNC puts a strong focus on new image analysis tools specifically designed for reliable and efficient processing of neuroimage data from birth to 4 years of age. In particular, we are developing novel tools for quantitative analysis of white matter properties and white matter fiber tracts in circuits of interest accessible via DT imaging. The hypothesis of different growth patterns between autistic subjects and typically developing children will be addressed by applying new innovative white matter analysis techniques [4,5,6,7] to DTI data. The methods will be made freely available to the scientific research community via web download. We will also generate pilot data of early white matter development in autism and controls assessed by DTI, data that does not yet exist and that might serve as reference data for future studies. Most relevant for improved understanding of early development in autism, we will study relationships between diffusion properties of specific tracts of interest with major clinical correlates like cognitive ability and behavioral severity.