The Use of Artificial Intelligence in Screening and Diagnosis of Autism Spectrum Disorder: A Literature Review

Abstract

Objectives:The detection of autism spectrum disorder (ASD) is based on behavioral observations. To build a more objective datadriven method for screening and diagnosing ASD, many studies have attempted to incorporate artificial intelligence (AI) technologies. Therefore, the purpose of this literature review is to summarize the studies that used AI in the assessment process and examine whether other behavioral data could potentially be used to distinguish ASD characteristics. Methods: Based on our search and exclusion criteria, we reviewed 13 studies. Results: To improve the accuracy of outcomes, AI algorithms have been used to identify items in assessment instruments that are most predictive of ASD. Creating a smaller subset and therefore reducing the lengthy evaluation process, studies have tested the efficiency of identifying individuals with ASD from those without. Other studies have examined the feasibility of using other behavioral observational features as potential supportive data. Conclusion: While previous studies have shown high accuracy, sensitivity, and specificity in classifying ASD and non-ASD individuals, there remain many challenges regarding feasibility in the real-world that need to be resolved before AI methods can be fully integrated into the healthcare system as clinical decision support systems.

Data-Driven Automated Detection of Autism Spectrum Disorder Using Activity Analysis: A Review

“…Furthermore, it articulated the content mentioned above and provided a forward path that recommends using machine learning in ASD to win regard to implementation, conceptualization, and mining data. Besides, Song et al [96] dispensed a comprehensive survey on the use of artificial intelligence in screening ASD. It briefly reviewed existing ASD assessment techniques, facial expression, motor movement data analysis, and their results on different learning algorithms.…”

A Systematic Literature Review on the Application of Machine-Learning Models in Behavioral Assessment of Autism Spectrum Disorder

“…Based on the reviewed literature, we sought the answer on whether the recent findings could sufficiently translate to real-life implementation of ML-based ASD screening and diagnostic models. Nonetheless, previous literature reviews assessed the performance of ML models in ASD screening and diagnosis based on the common evaluation metrics of sensitivity, specificity, and accuracy, among others [ 25 , 28 ]. However, none of the existing literature reviews systematically analyzed the subject area and provided enough evidence on the readiness and sufficiency of the models toward real-life implementation of the ML-based systems.…”

“…However, none of the existing literature reviews systematically analyzed the subject area and provided enough evidence on the readiness and sufficiency of the models toward real-life implementation of the ML-based systems. For instance, Song et al [ 28 ] reviewed 13 relevant studies that utilized varying data types and discussed the possibility of achieving effective classification of ASD based on the study findings. Similarly, Thabtah [ 25 ] identified some limitations within the commonly employed research methodologies and proposed intuitive stages toward appending the ML models into ASD screening apps.…”

Applying Artificial Intelligence for Diagnostic Classification of Korean Autism Spectrum Disorder

“…To overcome administrator bias during assessments, many have attempted to use AI technology to create a more concise algorithm for ASD diagnosis. 7 Several studies have attempted to classify items from assessment of ASD diagnostic instruments that are most predictive of the diagnosis to make the process less time-consuming. Wall et al 8 applied 16 ML algorithms with data in multiple ASD data repositories, including the Autism Genetic Resource Exchange (AGRE) and Autism Consortium (AC), and they suggested that ML enables the classification of autism and non-autism with 100% accuracy using tree algorithms such as ADTree.…”

“…To overcome administrator bias during assessments, many have attempted to use AI technology to create a more concise algorithm for ASD diagnosis [ 7 ]. Several studies have attempted to classify items from assessment of ASD diagnostic instruments that are most predictive of the diagnosis to make the process less time-consuming.…”