The top 100 cited articles on artificial intelligence in Alzheimer's disease and mild cognitive impairment: a...

Alzheimer's disease (AD) and Mild Cognitive Impairment (MCI) pose significant societal and healthcare burden. Artificial intelligence (AI) methods have been widely applied in AD and MCI studies. We conducted a bibliometric analysis of the 100 most cited articles on AI applied to AD and MCI.

We searched the Web of Science database using keywords related to AD, MCI, and AI (e.g., "deep learning," "machine learning," "neural networks"). Citation counts ranked articles, and the top 100 were manually screened. Key parameters such as authors, journals, citation count, countries, institutions, and keywords were automatically extracted. We also manually extracted key information, including publication type, impact factor (IF), Journal Citation Reports (JCR) Category Quartile, AI methods, and clinical data types. Analysis and visualization were conducted using VOSviewer.

Among the 100 articles, 13 were reviews, 2 were basic research papers, and 85 were clinical studies. Seventy seven articles focused on diagnosis and prediction. MRI data was the most frequently used analysis source. Shen Dinggang, the United States, and the University of North Carolina at Chapel Hill were respectively the individual, country, and institution with the highest publication volume. Neuroimage published the most papers (n = 14), and all the top 10 journals belonged to JCR Q1. Emerging keywords included "ensemble learning," "transfer learning," and "structural MRI." Support Vector Machine (SVM) was the most commonly applied AI method (n = 25), closely followed by convolutional neural network (CNN, n = 24).

We analyzed the top 100 cited articles on AI in AD and MCI across authors, journals, countries, institutions, keywords, and AI methods. Diagnosing AD/MCI is the primary research focus, with MRI as the most studied examination. SVM and CNN are the most frequently used AI methods in these studies.