Huang Xiao, Han Fei, Chen Yong-Feng, Sun Qiang, Guo Jian-Wei, Ye Zi, Qi Wei, Zhang Da-Wei
Department of Orthopedics, Xijing Hospital, Air Force Medical University, Xi'an, China.
Lintong Rehabilitation and Convalescent Centre of the Joint Logistics Support Force, Xi'an, China.
Quant Imaging Med Surg. 2025 May 1;15(5):3993-4013. doi: 10.21037/qims-24-1384. Epub 2025 Apr 28.
With the development of artificial intelligence (AI) and the increasing significance of imaging in orthopedics, the application of AI in the field of orthopedic imaging is becoming increasingly extensive. Previous studies show that the application of AI-based orthopedic imaging may break the traditional model of the field. As a result, relevant research has received attention, and numerous articles have been published. Through bibliometric analysis, this study summarized the knowledge structure of AI-based orthopedic imaging and explored its potential research trends and focal points.
In this study, literature on AI in the field of orthopedic imaging available in the Web of Science Core Collection (WoSCC) database from 1 January 2007 to 31 December 2024 was analyzed. In order to identify the main research topics and generate visual charts of countries, institutions, authors, and keyword networks, the search results were imported into VOSviewer and CiteSpace.
A total of 3,147 publications were analyzed, revealing a rapid increase in AI research in orthopedic imaging since 2007, with over 90% of studies published after 2017. The United States (US) and China dominate this field, with the US leading in citations and academic influence, and China demonstrating significant growth in productivity. Institutional analysis highlighted Harvard University and Stanford University as key contributors, reflecting their strong academic influence. Keyword analysis identified three main research focuses: (I) advancements in algorithm development, particularly deep learning (DL) methods such as convolutional neural networks (CNNs); (II) applications in orthopedic disease imaging, including osteoarthritis, osteoporosis, and total knee arthroplasty; and (III) innovations in multimodal fusion and three-dimensional (3D) imaging techniques. Emerging trends emphasize integrating imaging data with clinical biomarkers to improve diagnostic accuracy and therapeutic decision-making. These findings provide a comprehensive overview of AI's role in orthopedic imaging, emphasizing areas of high impact and potential future directions for research.
The research on the application of AI in orthopedic imaging is a hot topic and indicates broad research prospects in the future. However, this study suggests that research teams should strengthen collaboration, especially international cooperation. Based on comprehensive analysis, the development of DL algorithms (especially CNNs), the use of AI in processing image data related to orthopedic diseases (segmentation, classification, and feature map extraction), and the expansion of AI imaging applications in different diseases are expected to become hotspots in future research on the application of AI in orthopedic imaging.
随着人工智能（AI）的发展以及影像学在骨科领域的重要性日益增加，AI在骨科影像学领域的应用越来越广泛。以往研究表明，基于AI的骨科影像学应用可能会打破该领域的传统模式。因此，相关研究受到关注，已有众多文章发表。本研究通过文献计量分析总结了基于AI的骨科影像学的知识结构，并探讨了其潜在的研究趋势和重点。
本研究分析了Web of Science核心合集（WoSCC）数据库中2007年1月1日至2024年12月31日期间骨科影像学领域中关于AI的文献。为了确定主要研究主题并生成国家、机构、作者和关键词网络的可视化图表，将搜索结果导入VOSviewer和CiteSpace。
共分析了3147篇出版物，显示自2007年以来骨科影像学中AI研究迅速增加，超过90%的研究发表于2017年之后。美国和中国在该领域占据主导地位，美国在引文和学术影响力方面领先，而中国在生产力方面呈现显著增长。机构分析突出了哈佛大学和斯坦福大学作为主要贡献者，反映了它们强大的学术影响力。关键词分析确定了三个主要研究重点：（I）算法开发的进展，特别是深度学习（DL）方法，如卷积神经网络（CNN）；（II）在骨科疾病影像学中的应用，包括骨关节炎、骨质疏松症和全膝关节置换术；（III）多模态融合和三维（3D）成像技术的创新。新兴趋势强调将影像数据与临床生物标志物整合以提高诊断准确性和治疗决策。这些发现全面概述了AI在骨科影像学中的作用，强调了高影响力领域以及未来潜在的研究方向。
AI在骨科影像学中的应用研究是一个热门话题，且表明未来具有广阔的研究前景。然而，本研究建议研究团队应加强合作，尤其是国际合作。基于综合分析，DL算法（特别是CNN）的开发、AI在处理与骨科疾病相关的图像数据（分割、分类和特征图提取）中的应用以及AI成像在不同疾病中的应用扩展有望成为未来AI在骨科影像学应用研究的热点。
