Research hotspots and emerging trends of artificial intelligence in the clinical management of mild cognitive impairment: A bibliometric and evidence-based analysis.

OBJECTIVE

This study combines a bibliometric analysis with an umbrella review to delineate the research landscape, hotspots, and emerging trends in the application of artificial intelligence to the clinical diagnosis and treatment of mild cognitive impairment.

METHODS

We searched the Web of Science Core Collection for literature published between 2004 and 2024. Bibliometric analysis of the retrieved publications was performed using CiteSpace and VOSviewer to map publication trends, international collaboration networks, key contributors, and keyword co-occurrence. From this dataset, we systematically identified and selected eligible systematic reviews and meta-analyses to conduct an umbrella review, thereby synthesizing the highest-level evidence on this topic.

RESULTS

A total of 3101 publications were included in the analysis, revealing a consistent upward trend in both publication volume and citation frequency. The United States emerged as the most productive country, while the Chinese Academy of Sciences was identified as the leading contributing institution. Ding-gang Shen ranked as the most frequently cited author, and Frontiers in Aging Neuroscience was the most prolific journal in this field. Twelve eligible systematic reviews and meta-analyses primarily focused on topics such as machine learning-based predictive models, neuroimaging and biomarker applications, and cognitive training or intervention strategies.

CONCLUSION

The application of artificial intelligence in mild cognitive impairment is a rapidly growing field of research. Core research areas include the use of neural networks, electroencephalography, and language processing for diagnostics and monitoring. Emerging research frontiers, such as transfer learning and the investigation of tau pathology pathways, indicate promising directions for future studies.