Date on Master's Thesis/Doctoral Dissertation


Document Type

Master's Thesis

Degree Name

M. Eng.

Department (Legacy)

Department of Biomedical Engineering

Committee Chair

El-Baz, Ayman Sabry


Diagnostic imaging; Autism--Diagnosis


Introduction: Autism is a complex developmental disability that typically appears during the first three years of life, and is the result of a neurological disorder that affects the normal functioning of the brain, impacting development in the areas of social interaction and communication skills. According to the Centers for Disease Control and Prevention (CDC) in 2009, about 1 in 110 American children will fall somewhere in the autistic spectrum. Although the cause of autism is still largely not clear, researchers have suggested that genetic, developmental, and environmental factors may be the cause or the predisposing effects towards developing autism. While shape based statistical analysis methods for autism are still in their early stages, current results show positive outlooks on the ability to detect differences between autistic and normal patients. Methods: The goal of this thesis is to construct a complete package that is capable of taking 2-dimensional images from a standard medical scanner, and be able to construct a three-dimensional representation of the object and examine it through combination of its weighted linear spherical harmonics. The desired outcome is that a distinction can be made between the analysis of autistic and normal brain data. The analysis package created is divided into three distinct components that are capable of performing the complete analysis on a subject. The components included in the package in order of runtime are: volumetric extraction and mesh generation from 2-dimensional medical scanner data, spherical deformation of the constructed mesh, and weighted spherical harmonic representation and analysis. Results: The minimum error for each brain following spherical harmonic reconstruction was calculated along with the fastest iteration at which the brain converged below the error thresholds of 11% and 10%. It was expected that due to the complexity of an Autistic brain these would require more iterations to converge to the same error level as a normal brain. It was also likely that within the number of iterations tested the autistic brains would record a larger final error due to this slower convergence rate. This was confirmed by the data. A global result was examined as well for the autistic and normal data groups. The overall minimum error for normal brain data was significantly lower than the autistic brain data. The average error for autistic brain data was significantly higher in both convergence measurements, but was dramatically higher in the 10% category. Conclusion: Using this method of analyzing data can demonstrate accurate differences in normal and autistic brains. The research that has been generated in this thesis can clearly demonstrate that the normal brain data converged both faster and with a lower rate of error level than the Autistic brain data. This result proves that the autistic brain is a more complex structure, and would be more difficult to reconstruct using this Shape- Based Detection of Cortex Variability process.