The application of artificial intelligence in stroke research: A bibliometric analysis

Area of Science:

• Neuroscience
• Medical Informatics
• Rehabilitation Technology

Background:

• Artificial intelligence (AI) is increasingly utilized in stroke prediction, diagnosis, evaluation, and rehabilitation.
• A comprehensive quantitative and qualitative overview of AI in stroke research is currently lacking.

Purpose of the Study:

• To conduct a bibliometric analysis of AI applications in stroke research over the past two decades.
• To elucidate the research status and evolving hotspots in this field.

Main Methods:

• Bibliometric analysis of 4437 publications from the Web of Science Core Collection (2003-2023).
• Analysis of publication trends, country/region cooperation, institution co-occurrence, and keyword co-occurrence using CiteSpace and VOSviewer.

Main Results:

• A consistent upward trend in annual publications was observed.
• The USA leads in publications and top institutions. Key research themes include rehabilitation, machine learning, recovery, and upper limb function.
• Emerging hotspots include machine learning, natural language processing, and atrial fibrillation prediction.

Conclusions:

• The USA is a leading contributor to AI in stroke research.
• Future research is likely to emphasize AI-driven prediction and rapid diagnosis of stroke using machine learning, deep learning, and natural language processing.