Purpose: Autism Spectrum Disorder (ASD) is defined as an idiopathic lifelong neurodevelopmental disorder that develops in early childhood. Its prevalence is broad, yet its nature remains elusive and inconsistent. This research describes a detailed research project related to the construction of computational phantoms, including certain brain structures which can be used in radiological, therapeutic, and diagnostic capacities.
Methods: A collection of three hybrid computational brain phantoms corresponding to two pediatric ASD patients and one control patient were constructed using high quality magnetic resonance imaging (MRI) sets. A total of 12 brain structures, four from each patient, were outlined from radiological images via the 3DSlicer segmentation software and were imported to the three-dimensional (3D) modeling software package Rhinoceros for additional geometric manipulation and analysis. Rhinoceros allowed the conversion of the segmentation into a non-uniform rational basis spline (NURBS) or mesh surface-based phantom.
Results: This set of new phantoms can be used in the future to study the optimization of image quality and radiation dose for patients on the spectrum who have unique neuro-morphology. Given the wide neurological, behavioral, and morphological variations in patients with autism, individualized treatment plans involving an accurate, representative computational model are valuable.
Conclusion: The aim is to develop a repository of data collected from these phantoms into a comprehensive dosimetry database. This novel set of computational brain models present morphological details that are crucial during radiation dose evaluation for ASD patients enduring diagnostic imaging or radiotherapy.