Artificial intelligence-based approaches for improving the diagnosis, triage, and prioritization of autism spectrum disorder: a systematic review of current trends and open issues
Shahad Sabbar Joudar,A. S. Albahri,Rula A. Hamid,Idrees A. Zahid,M. E. Alqaysi,O. S. Albahri,A. H. Alamoodi
DOI: https://doi.org/10.1007/s10462-023-10536-x
IF: 9.588
2023-06-21
Artificial Intelligence Review
Abstract:The artificial intelligence (AI) trend to embrace Autism Spectrum Disorder (ASD) has dramatically transformed the landscape of medical diagnosis. People often exhibit fear and apprehension towards conditions they lack understanding of, and ASD being a complex affliction, poses challenges in comprehending its intricacies. Researchers have harnessed AI applications to improve the precision of disease diagnosis by utilizing Magnetic Resonance Imaging (MRI), Electroencephalography (EEG), genetic, sociodemographic, and medical data. However, the development of AI systems for early diagnosis and triage in healthcare is still in its nascent stages. In particular, studies have revealed a global increase in the prevalence of ASD, with an estimated 1 in 59 children being diagnosed. However, there is a lack of up-to-date information regarding the current status of ASD. This study aims to provide a systematic review of AI applications in early diagnosis and triage for ASD, supplementing the findings of previous studies and offering a comprehensive overview of the evidence. To achieve this, a rigorous literature search method and selection criteria were employed, resulting in the identification of 46 recent contributions on the applications of AI in ASD from various databases, including ScienceDirect (SD), IEEE Xplore digital library (IEEE), Web of Science (WOS), PubMed, and Scopus. The selected papers were categorized into three main categories: ASD triage levels, clinical diagnosis for ASD, and diagnosis based on telemedicine, with further subcategories under the clinical diagnosis category. Theoretical and practical aspects of AI methods used for ASD diagnosis, as well as the presentation utilizing data analytics, were presented. The paper presents a systematic and comprehensive analysis of previous studies, examining the challenges, motivations, and recommendations, thereby paving the way for potential future research. Additionally, the work provides decisive evidence for the use of AI in ASD healthcare diagnosis and triage, offering nine critical analyses of the current state-of-the-art and addressing relevant research gaps. To the best of our knowledge, this study is innovative in exploring the feasibility of using AI in ASD medical diagnosis and triage. It highlights essential pieces of information, including Explainable AI (XAI), Auto machine learning (AutoML), Internet of Things (IoT)-based AI, robot-assisted therapy-based AI, telemedicine, data fusion techniques, and available ASD datasets with different aspects. The analysis of the revised contributions reveals crucial implications for academics and practitioners. The paper also proposes potential methodological aspects to enhance the triage and prioritization of autistic patients using AI applications in the medical sector, as well as addressing theoretical and practical application aspects and five methodology phases using fuzzy Multi-Criteria Decision Making (MCDM) methods in ASD triage and prioritization.
computer science, artificial intelligence