Journal of Engineering Research
Abstract
Abstract- This article provides a comprehensive literature review on technology-based interventions for Autism Spectrum Disorder (ASD). It emphasizes the challenges in early detection and treatment of ASD, highlighting the spectrum nature of the disorder. The review discusses traditional diagnostic strategies such as behavioural observations, developmental screening and medical testing and goes on to explore advanced machine learning and deep learning models, including SVM, k-nearest neighbours, decision tree and LSTM, for predicting ASD characteristics in toddlers and children. Additionally, recent techniques employing more than ten strategies for ASD detection are summarized and various datasets used in early detection are described. The study underscores the utility of artificial intelligence in early ASD detection and recommends integrating advanced techniques, such as eye tracking, EEG, brain imaging, exome sequencing, urine analysis and behaviour coding, with conventional methods for potentially earlier and more effective solutions.
Recommended Citation
Sharabash, Shaimaa Fouad and Elghaish, Hany Ali
(2024)
"Early Autism Detection using Machine Learning Techniques: A Review,"
Journal of Engineering Research: Vol. 8:
Iss.
4, Article 5.
Available at:
https://digitalcommons.aaru.edu.jo/erjeng/vol8/iss4/5