The Current Research Landscape on the Machine Learning Application in Autism Spectrum Disorder: A Bibliometric...

Area of Science:

• Neuroscience and Artificial Intelligence
• Computational Psychiatry
• Bibliometric Analysis

Background:

• Autism spectrum disorder (ASD) is characterized by language deficits, restricted interests, and social difficulties.
• Machine learning and neuroimaging are increasingly combined for ASD research.
• Bibliometric analysis provides insights into the current landscape and emerging trends in machine learning for ASD.

Purpose of the Study:

• To conduct a bibliometric analysis of machine learning applications in ASD.
• To identify research trends, popular topics, and future research directions.
• To map the evolution of machine learning in ASD research from 1999 to 2023.

Main Methods:

• Publications on machine learning and ASD were retrieved from the Web of Science Core Collection (1999-2023).
• Data analysis involved characterizing authors, articles, journals, institutions, and countries using Microsoft Excel and VOSviewer.
• Knowledge networks, collaborative maps, hotspots, and trends were analyzed using VOSviewer and CiteSpace.

Main Results:

• A total of 1357 papers were published between 1999 and 2023, with a significant increase after 2016.
• The United States, China, India, and England were leading contributors, with Stanford University and Harvard Medical School among top institutions.
• Central topics included "autism spectrum disorder", "machine learning", "children", "classification", and "deep learning".

Conclusions:

• Enhanced international collaboration is needed for future research.
• Research focus is shifting towards "artificial intelligence", "deep learning", "electroencephalography", and "pediatrics".
• Future directions include crowdsourcing machine learning and utilizing electroencephalography for ASD diagnosis.