Related Experiment Videos
The application of artificial intelligence in systemic lupus erythematosus: a bibliometric analysis of current trends
Ziyi Liu1, Wenqian Yu2, Xinxin Meng1
1Dongzhimen Hospital, Beijing University of Chinese Medicine, Beijing, China.
Background:
Systemic lupus erythematosus (SLE) poses significant clinical challenges due to its heterogeneity and unpredictable relapses. Artificial intelligence (AI) is driving a paradigm shift in SLE management through molecular subtyping and prognostic modeling. However, a comprehensive quantitative analysis of this rapidly growing field is lacking. This study employs bibliometric and visualization methods to systematically map the development, collaboration patterns, and research frontiers of AI in SLE.
Methods:
English-language original articles and reviews on AI in SLE, published between January 1, 2005, and June 10, 2026, were retrieved from the Web of Science Core Collection and Scopus. After rigorous screening, 707 core publications were analyzed using R-Bibliometrix, VOSviewer, and CiteSpace to evaluate publication trends, collaboration networks, keyword co-occurrence, and citation bursts.
Results:
The 707 included publications demonstrated a polynomial accelerated growth trend since 2005. Research has evolved from algorithmic proof-of-concepts to deep learning and multi-omics integration. China and the United States lead global output and collaboration, with institutions like the Karolinska Institute showing significant impact. Although "machine learning" and "lupus nephritis" remain core topics, the research focus has expanded. Current studies increasingly emphasize molecular stratification, automated pathological image classification, and the longitudinal prediction of flares and organ damage. Recent citation bursts for "immunosuppressive agent," "tacrolimus," and "prednisone" highlight personalized treatment efficacy prediction as the latest frontier.
Conclusion:
AI has matured from methodological exploration into a robust clinical decision-support tool for SLE, particularly for lupus nephritis assessment and flare prediction. Future advancements rely on conducting prospective clinical validations in real-world cohorts, predicting personalized drug efficacies, and leveraging multi-omics to decode pathological mechanisms. These steps are essential to transition SLE management from traditional empirical approaches to data-driven precision medicine.
