Abstract:

Objective To explore the research status, hotspots and development of artificial intelligence in the field of diagnosis and intervention for children with autism spectrum disorder (ASD).
Methods Literatures about the topic of artificial intelligence for diagnosis and intervention for ASD were retrieved from Web of Science Core Collection till October 28, 2024, and analyzed using CiteSpace 6.3.R1.
Results A total of 219 articles were included, showing a fluctuating upward trend. Peter Washington, Dennis P. Wall, Antonie D. Kline, Haik Kalantarian and Qandeel Tariq were the authors with the most publications. The United States was dominant in the field, and Stanford University, the Egyptian Knowledge Bank, Harvard University and University of London were the core institutions. The researches focused on the application of machine learning, deep learning, virtual reality and other technologies in social communication, emotion recognition and speech rehabilitation. The development of the researches could be classified as three phases: validation (2012-2016), deepening (2017-2020) and integration and innovation (2021-2024).
Conclusion Artificial intelligence has developed significantly in the field of diagnosis and intervention for children with ASD. Algorithms such as machine learning and deep learning can improve diagnostic accuracy, and technologies such as virtual reality and robot-assistance can improve intervention outcomes. The researches are evolving from a single technology validation to a multimodal technology integration, led by the United States with the participation of many countries.

Key words: autism spectrum disorder, children, artificial intelligence, bibliometrics