Composite Artificial Intelligence

12Composite AI for Predictive Analysis of Autism Spectrum Disorder Using Facial Features

S. Usharani^1*, A. Ganesh¹, N. Muralidharan¹ and G. Glorindal²

¹Department of Artificial Intelligence and Machine Learning, IFET College of Engineering,Villupuram, Tamil Nadu, India

²Department of Computer Science, University of EdenBerg, Lusaka, Zambia

Abstract

This chapter presents the composite artificial intelligence (AI) techniques in the face feature analysis for the autism spectrum disorder (ASD) prediction model. This research utilizes various AI techniques like facial feature recognition and deep learning to raise the precision of diagnosis owing to the inherent difficulty of diagnosing ASD. The chapter sets forth key phases like data acquisition, preprocessing, and model construction while concentrating on the identification of relevant face traits showing their link to ASD. Such advanced models consider predictive potential, focusing on Xception and VGG19. Convolutional Neural Networks (CNNs) recognize and categorize facial features. This is done in the chapter where different performance models are assessed to find out how effective they are in discriminating ASD cases from typically developing individuals in terms of accuracy, sensitivity, and specificity. In the end, it would be comprised of Composite AI and facial recognition which would be a very promising step towards the early identification of ASD without any invasive techniques as well as more effective clinical outcomes. ...