Autism Detection Based on Facial Images Using VGG16 and VGG19

Abstract

A mental disability called autism spectrum disorder exhibits specific difficulties with verbal and nonverbal communication, interpersonal skills, and obsessive activities. Around 1% of the total populace is impacted by it, and its side effects frequently show up during the formative stages, or during the initial two years following birth. Autism can be diagnosed at any stage in once life and is said to be a \"behavioural disease\" because in the first two years of life symptoms usually appear. There hasn\'t been a strong diagnosis method, though, because there aren\'t any discernible variations between the facial images of healthy people and those of people with ASD. Machine learning and Deep learning approaches are being used in conjunction with traditional diagnostic procedures to increase the accuracy and turnaround time for diagnoses. In this study, we are looking to build a deep learning model i.e. A Convolution Neural Network that can classify and detect Autism based on facial images. The algorithm involves several key steps, including data collection, pre-processing, model training, and evaluation. This project explores the potential of using deep learning models, specifically VGG16 and VGG19 convolution neural networks (CNNs), for the detection of ASD based on facial images.

Introduction

I. INTRODUCTION

Autism is a mental disorder that is especially common in early childhood. It is a developmental disorder that affects interpersonal communication and relationships. Their names include autism spectrum disorder (ASD). Although the exact cause of autism is unknown, people with autism have behaviours that harm themselves and others. The severity of autism varies from person to person and can be difficult to diagnose. Autism spectrum disorder (ASD) affects more than 1% of children, so it's good to catch the condition early. According to statistics, men are more likely to develop autism spectrum disorder (ASD) than women. Individuals with autism spectrum disorder (ASD) experience abnormalities and social anxiety when viewing images of the external environment. But it is not clear how people see the world from a first-person perspective. One of the most important cognitive processes in humans is the ability to observe important objects in our environment. However, people with autism spectrum disorder have great difficulties with social problems such as face and social environment. ASD affects approximately 5-9% of children. Children with autism face many challenges, including learning disabilities, attention deficits, and psychological problems with thinking, movement, despair, and anxiety. Main symptoms may include decreased vision, unresponsiveness, and poor communication with the caregiver. These symptoms of autism are more evident between 18 and 24 months.

Conclusion

We used both VGG16 and VGG-19 based models in this investigation to divide face image categories into autistic and non-autistic groups. In our study, the VGG16-based model performed better than the VGG19-based model. In this study, we suggested a comprehensive computerized approach for face image-based autism identification. This study used a Convolutional Neural Network with transfer learning to create a deep learning-based online interface for identifying autism. The CNN architecture has the right models to extract the facial landmarks, which can classify faces into autistic and non-autistic types. This is done by producing sequences of facial characteristics and calculating the distance between facial features. Parents and doctors will find it easier to identify autism spectrum disorders in children with the help of this software.

References

[1] Sherkatghanad, Zeinab, et al. \"Automated detection of autism spectrum disorder using a convolutional neural network.\" Frontiers in neuroscience 13 (2020): 1325. [2] T. Eslami, V. Mirjalili, A. Fong, A. R. Laird, and F. Saeed, “ASD-DiagNet: A Hybrid Learning Approach for Detection of Autism Spectrum Disorder Using fMRI Data,” Front. Neuroinform, vol. 13, 2019. [3] Wang, S., Adolphs, R. (2017). Social saliency. In Computational and Cognitive Neuroscience of Vision, pp. 171-193. [4] Paulaetal.-2011- BriefReportPrevalenceofPervasiveDevelopmentalDisorderinBrazi- lAPilotStudy.pdf. [5] E. Fombonne, “Epidemiology of pervasive developmental disorders,” Pediatric Research, vol. 65, no. 6, pp. 591–598, 2009. [6] L. Kanner, “Autistic disturbances of affective contact,” Nerv. Child, vol. 2, pp. 217–250, 1943 [7] M. N. Parikh, H. Li, and L. He, “Enhancing diagnosis of autism with optimized machine learning models and personal characteristic data,” Frontiers in Computational Neuroscience, 2019. [8] Baio, Jon, et al. \"Prevalence of autism spectrum disorder among children aged 8 years—autism and developmental disabilities monitoring network, 11 sites, United States, 2014.\" MMWR Surveillance Summaries 67.6 (2018): 1. [9] Thabtah, Fadi, and David Peebles. \"A new machine learning model based on induction of rules for autism detection.\" Health informatics journal 26.1 (2020): 264-286. [10] Thabtah, Fadi, and David Peebles. \"A new machine learning model based on induction of rules for autism detection.\" Health informatics journal 26.1 (2020): 264-286.

Copyright

Copyright © 2024 Koneru Anitha, Nikhitha Puppala, Harshitha Mantada, Varshith Uddanti, Sri Ram Jagadish Ponnada, Pushya Dhanekula. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.