基于文献计量方法的人工智能在超声心动图中的应用进展研究

Research of the Application Progress of Artificial Intelligence in Echocardiography Based on Bibliometric Method

目的分析人工智能技术在超声心动图检查设备中应用的现状、趋势以及热点。方法在Web of Science的核心数据库中搜索近10年关于人工智能在超声心动图中应用的文献,利用Citespace软件,运用文献计量的方法进行文献分析。从文献的年发文量、主要研究国家及机构、刊文期刊来源及关键词共现、共被引文献聚类等角度进行可视化分析。结果共检索到383篇文献,发文量最多的国家是中国以及美国,发文量最多的研究机构是梅奥医学中心、不列颠哥伦比亚大学、西北大学、加州大学旧金山分校等,发文最多的期刊为Front Cardiovasc Med、JACC-Cardiovasc Imag。研究趋势的早期在于探索如何将人工智能技术应用于超声心动图的识别,即如何将图像的特征与相应的诊断进行关联;后来即通过开发相应的算法对二维超声心动图的各种视图进行识别;最近的热点在将深度学习应用于超声心动图的心肌病及冠心病等的诊断方面。结论应用CiteSpace分析人工智能技术在超声心动图中的应用,展示出了该领域目前的主要研究及彼此合作的国家和机构,揭示了该领域的研究趋势,可为该设备开发及应用人员提供一定参考价值。

Objective To analyze the present situation,trend and hot spots of the application of artificial intelligence technology in echocardiography examination equipment.Methods In the core database of Web of Science,literature about the application of artificial intelligence in echocardiography in recent ten years was searched.And the literature was analyzed by using Citespace software and bibliometric.Visual analysis was carried out from the perspectives of the annual publication volume,the main research countries and institutions,the sources of published journals and the co-occurrence of keywords,and the clustering of co-cited documents.Results A total of 383 articles were retrieved,among which the countries with the largest number of published articles were China and the United States.The research institutions with the largest number of published articles were Mayo Medical Center,University of British Columbia,Northwest University,University of California,San Francisco,etc..The journals with the largest number of published articles were Front Cardiovasc Med,JACC-Cardiovasc Imag.The early research trend lied in exploring how to apply artificial intelligence technology to the identification of echocardiography,that is,how to correlate the image features with the corresponding diagnosis.Later,various views of two-dimensional echocardiography were identified by developing corresponding algorithms.Recently,the hot spot was the application of deep learning to the diagnosis of cardiomyopathy and coronary heart disease of echocardiography.Conclusion CiteSpace is used to analyze the application of artificial intelligence technology in echocardiography,which shows the main research in this field and the countries and institutions cooperating with each other at present,and reveals the research trend in this field,which can provide certain reference value for the equipment development and application personnel.