1. 2025;28;167-181Research Focus Involving and Trends in Artificial Intelligence for Spinal Pain: A Bibliometric Analysis
   Bibliometric Analysis
   Chaobo Feng, MD, Zhuoxi Zhou, MBBS, Yongen Miao, MBBS, Sheng Yang, MD, Guoxin Fan, MD, and Xiang Liao, MD.

BACKGROUND: Spinal pain is a pervasive global health issue that poses significant challenges because of the disability and economic burden it causes. Despite the availability of various treatments for the condition, a definitive cure for spinal pain remains elusive, underscoring the need for innovative approaches. Artificial intelligence (AI) is considered a potential method for facilitating relief for patients suffering from spinal pain.

OBJECTIVE: This study utilized a bibliometric analysis to explore the impact of AI on spinal pain research, examining publication trends, collaboration patterns, author contributions, and keyword clusters, to analyze research focus and trends in this field.

STUDY DESIGN: Bibliometric analysis.

SETTING: Data were obtained from the Web of Science Core Collection (WoSCC).

METHODS: The literature related to AI-assisted techniques in spinal pain treatment was collected from the WoSCC. The CiteSpace and R Bibliometrix software packages were used in the analysis.

RESULTS: In total, 310 articles were included, with the number of publications and citations increasing progressively. The greatest number of publications and total citations came from the United States. The University of Washington was the institution associated with the most publications. Mork PJ was the byline that appeared most often in association with both publications and total citations. The European Spine Journal was the journal in which the most publications appeared, while Spine had the greatest number of citations. The literature with the most global citations was published by Jamalusin A in the European Spine Journal, while the literature with the most local citations was by Sandal LF on JMIR Research Protocols. The most frequent key words were “machine learning,” “low back pain,” “magnetic resonance imaging,” etc.

LIMITATIONS: Only the English-language articles in the WoSCC database were included, and proceeding papers, meeting abstracts, and book chapters were excluded. Furthermore, we included no research about wearable sensors, virtual reality, and so on. Additionally, the articles from the other databases were not included.

CONCLUSION: The research of applying AI as a treatment for spinal injury has appealed to interdisciplinary efforts, reflecting the potential for self-management, imaging processing, and clinical decision-making. An overall perspective is shown in our study, which facilitates understanding and provides research focuses and trends in this field.

KEY WORDS: Spinal pain, artificial intelligence, bibliometric, CiteSpace