Hybrid Artificial Intelligence and Psychological Tests Based Autism Spectrum Disorders Identification Using 3D MRI Images

Abstract

Autism Spectrum Disorder (ASD) is a neurodevelopmental condition that affects social
communication and behavior with children, characterized by diverse symptoms and varying
severity levels. This paper explores the application of artificial intelligence (AI) algorithms
combined with efficient psychological tests for autism detection which is one of the most
important and rapidly expanding ailments in the world that are now difficult to diagnose since
there are no exact established diagnostic standards. In order to find biomarkers linked to various
conditions, AI will analyse vast amounts of data, speeding up the diagnosis procedure and
increasing the accuracy of the results. Beside common psychological scales, this study focuses
on developing a three Dimensional (3D) deep learning models based approaches for ASD
identification using structural Magnetic Reasoning Imaging (sMRI) and functional Magnetic
Reasoning Imaging (fMRI) data, integrating models such as 3D Resnet, 3D DenseNet, and 3D
VGG16 to extract spatial patterns associated with ASD. To evaluate the performance of our
proposed system, we conducted several experiments based on multiple parameters that have
been performed using the publicly challenged ABIDE dataset of unconstrained images. The
obtained experimental results proved the effectiveness of the proposed system against other
CNN architectures, as well as with recent state-of-the-art methods.