Hotspots and Trends in the Application of Artificial Intelligence in Spine Medicine from 2005 to 2024: A Bibliometric...

Artificial intelligence (AI) is emerging as a transformative force in spine medicine, offering support in diagnosis, surgical planning, and postoperative care through advanced algorithms. We aim to perform a comprehensive bibliometric analysis of AI applications in spine medicine to uncover research trends and inform future directions.

We screened English-language publications on AI in spine medicine from 2005 to 2024 in the Web of Science Core Collection. We utilized the CiteSpace, VOSviewer, and Bibliometrix R packages to analyze publication years, countries, institutions, authors, journals, references, and keywords, and to generate social network graphs. Scimago Graphica, Origin, and Microsoft Charticulator were used for additional visualization results.

1344 papers were included. The development of AI in spine medicine can be divided into two phases: slow-growing (2005-2018) and fast-growing (2019-2024). The most productive countries and institutions were the USA and the University of California System, respectively. Galbusera F was the most prolific and influential author. European Spine Journal accepted the most articles, and Spine was cited the most. Analyzing the references and keywords, we found that the current research hotspots are mainly in AI subfields such as machine learning and deep learning, as well as combining AI with imaging technologies such as MRI to assist in diagnosis and classification. Current research frontiers include augmented reality, osteoporosis, and spinal deformities.

AI in spine medicine is evolving rapidly. Author collaboration across teams should be enhanced. The existing studies are more focused on imaging diagnosis and classification. Future studies may increasingly explore spinal intraoperative navigation systems and disease progression prediction.