The Role of Structure MRI in Diagnosing Autism.

Mohamed T Ali,Ahmed Shalaby,Ayman El-Baz,Gregory Barnes,Yaser Elnakieb,Ahmed Elnakib,Ali Mahmoud,Jawad Yousaf,Manuel Casanova,Hadil Abu Khalifeh,Mohammed Ghazal

doi:10.3390/diagnostics12010165

Abstract

This study proposes a Computer-Aided Diagnostic (CAD) system to diagnose subjects with autism spectrum disorder (ASD). The CAD system identifies morphological anomalies within the brain regions of ASD subjects. Cortical features are scored according to their contribution in diagnosing a subject to be ASD or typically developed (TD) based on a trained machine-learning (ML) model. This approach opens the hope for developing a new CAD system for early personalized diagnosis of ASD. We propose a framework to extract the cerebral cortex from structural MRI as well as identifying the altered areas in the cerebral cortex. This framework consists of the following five main steps: (i) extraction of cerebral cortex from structural MRI; (ii) cortical parcellation to a standard atlas; (iii) identifying ASD associated cortical markers; (iv) adjusting feature values according to sex and age; (v) building tailored neuro-atlases to identify ASD; and (vi) artificial neural networks (NN) are trained to classify ASD. The system is tested on the Autism Brain Imaging Data Exchange (ABIDE I) sites achieving an average balanced accuracy score of . This paper demonstrates the ability to develop an objective CAD system using structure MRI and tailored neuro-atlases describing specific developmental patterns of the brain in autism.

Highlights

Autism spectrum disorder (ASD) is a heterogeneous neurodevelopmental disorder that has a strong genetic basis, and various clinical presentations
Studies utilized different Magnetic-Resonance Imaging (MRI) modalities in order to capture the effect of ASD on brain from different perspectives i.e., Structural MRI (sMRI), functional MRI (fMRI), and/or diffusiontensor imaging (DTI). sMRI was utilized by studies that cared more about the geometry of the cerebral cortex and morphology of the brain [8,9,10]
We propose a comprehensive ML model to detect imaging markers for autism and utilize these imaging markers to train a set of linear and non-linear classifiers to distinguish between ASD and typically developed (TD)

Summary

Introduction

Autism spectrum disorder (ASD) is a heterogeneous neurodevelopmental disorder that has a strong genetic basis, and various clinical presentations. A fundamental goal of any neurobiological study of autism is a description of brain regions that are of abnormal structure or dysfunctional. Brain studies in-vivo became possible after the invention of Magnetic-Resonance Imaging (MRI). Structural MRI (sMRI) examination is widely used to investigate the brain morphology due to its high contrast sensitivity and spatial resolution and because it entails no exposure to ionizing radiation; the last feature is important for children and adolescents [7]. Studies utilized different MRI modalities in order to capture the effect of ASD on brain from different perspectives i.e., sMRI, fMRI, and/or DTI. SMRI was utilized by studies that cared more about the geometry of the cerebral cortex and morphology of the brain [8,9,10]. Volumetric features usually refer to the volume of the subcortical structures, such as the hippocampus, putamen, thalamus, etc. [12]

Methods

Results

Discussion

Conclusion