ML based Approaches for Detection and Development of Autism Spectrum Disorder: A Review

• Published in: 2022 International Conference on Sustainable Computing and Data Communication Systems (ICSCDS) pp. 79 - 84
• Main Authors: Zope, Vidya, Shetty, Tanvi, Dandekar, Maitraiyi, Devnani, Anmol, Meghrajani, Puneet
• Format: Conference Proceeding
• Language: English
• Published: IEEE 07.04.2022
• Subjects: Autism; Autism Spectrum Disorder; Comparative study; Convolutional Neural Network; Data communication; Deep learning; Developmental Disability; Machine Learning; Medical tests; Pediatrics
• DOI: 10.1109/ICSCDS53736.2022.9761040

Autism spectrum disorder (ASD) is considered to be a serious developmental disability that can hamper the ability to communicate and interact since it affects social and communication skills. It is important to diagnose ASD as early as possible but at the same time, it is difficult since there is no particular medical test defined for the same. Early intervention services are key factors that influence and help children develop critical cognitive skills from birth to three years of age. Several approaches have been proposed but none without any drawbacks. It is highly important that the system provides children with a correct and efficient diagnosis that helps them work on their development to become the best version of themselves. It has been observed that the use of different machine learning and deep learning models have provided favourable results as seen in the approaches discussed. The purpose of this paper is to review some of the significant work done in this domain and help identify the lacuna in the current systems in order to overcome the same.