Autism Detection from Brain Signals: South Asia Perspective

Abstract

Autism Spectrum Disorder (ASD), simply autism, is a complex neurodevelopmental disorder that leads to serious social, communication, and behavioral challenges. The impact of ASD on the family, general well-being, society, and economy is becoming increasingly important due to its high prevalence and the extensive range of clinical treatments required for affected children. Early ASD detection is important because it could help improve access to intervention measures and may help improve developmental outcomes. Autism exposes a critical issue for developing countries, especially in South Asian countries. In the South Asia region, a few institutes are available in the capital or a few big metropolitan cities for diagnosing ASD. Generally, ASD is diagnosed through some conventional methods (e.g., screening). As a neurodevelopmental condition, ASD can be detected through brain signals. On account of easy operation and to maintain low-risk factors, only noninvasive neuroimaging methods, such as electroencephalography (EEG) are considered to measure a child's neural behaviors for classifying ASD. Machine learning methods are used with EEG signals for classifying ASD subjects in different studies than conventional methods. The main concern of this study is to provide some guidelines for machine learning-based automatic ASD detection through EEG signals which might be a prospectus for South Asian countries. Through the advanced system, people from all areas of the country will get proper ASD diagnosis and treatment facilities. Therefore, it is a timely approach of employing brain signals (especially EEG signals) and machine learning-based effective techniques for ASD detection in South Asian countries. Journal of Engineering Science 13(2), 2022, 69-81