摘要
目的分析中国知网数据库人工智能在肿瘤领域应用的文献,探讨其研究热点和发展趋势,为人工智能在肿瘤领域的进一步研究提供借鉴和参考。方法选择中国知网数据库2010-2020年收录的人工智能在肿瘤领域应用的1000篇文献作为研究对象,采用CiteSpace 5.7 R5软件对作者、机构、关键词进行共现分析,并对关键词进行聚类分析、突现分析。结果发文量排名第一的作者为张旭(41篇),其次为余佩武(25篇)、马鑫(25篇)和孙颖浩(24篇)。发文量居于前3位的机构是中国人民解放军总医院泌尿外科(17篇)、上海交通大学医学院附属瑞金医院泌尿外科(10篇)和第三军医大学西南医院全军普通外科中心微创胃肠外科中心(10篇)。出现频次较高的关键词为机器人(174)、腹腔镜(128)、机器人手术(97)、达芬奇机器人手术系统(74)、达芬奇机器人(67),深度学习(59),中介中心性最大的关键词为肺肿瘤(0.27),其次为诊断(0.22)、胃肿瘤(0.17)、腹腔镜检查(0.17)、达芬奇机器人(0.16)、直肠肿瘤(0.13)、外科手术(0.13)、达芬奇机器人手术系统(0.12)、前列腺肿瘤(0.12)。共产生7个主要聚类,聚类模块值(modularity Q,Q)=0.4997,平均轮廓值(weighted mean silhouette S,S)=0.8214。关键词突现分析结果显示,深度学习突现强度最高(14.52),且深度学习、卷积神经网络、影像组学、迁移学习和图像分割突现结束时间为2020年。结论人工智能在肿瘤领域的研究热点主要集中在诊断、胃肿瘤、腹腔镜检查、达芬奇机器人等方面,深度学习、卷积神经网络、影像组学、迁移学习和图像分割为研究前沿。
Objective To analyze the literature on the application of artificial intelligence in the field of cancer in CNKI database,discusse its research hotspots and development trends,and provide reference for the further research of artificial intelligence in the field of cancer.Methods A total of 1000 literatures on the application of artificial intelligence in cancer field collected in China HowNet database from 2010 to 2020 were selected as the research object,and CiteSpace 5.7 R5 software was used to co-occurrence analysis of authors,institutions and keywords,cluster analysis and emergence analysis of keywords.Results Zhang Xu(41)ranked first in the number of articles published,followed by Yu Pei-wu(25),Ma Xin(25)and Sun Ying-hao(24).The top three institutions with the highest number of articles were the Department of Urology of the General Hospital of the Chinese People’s Liberation Army(17),the Department of Urology of Ruijin Hospital Affiliated to the Medical College of Shanghai Jiaotong University(10)and the minimally invasive gastrointestinal surgery center of the general surgery center of the Southwest Hospital of the Third Military Medical University(10).Keywords with high frequency were robot(174),laparoscope(128),robotic surgery(97),Da Vinci robotic surgery system(74),Da Vinci robot(67),deep learning(59).The keywords with the largest intermediary centrality were lung tumor(0.27),followed by diagnosis(0.22),gastric tumor(0.17),laparoscopy(0.17),Da Vinci robot(0.16),rectal tumor(0.13),surgery(0.13),Da Vinci robot surgery system(0.12)and prostate tumor(0.12).A total of 7 main clusters were generated.The modularity Q was 0.4997 and the weighted mean silhouette S was 0.8214.Keyword emergence analysis showed that deep learning had the highest emergence intensity(14.52),and the end time of deep learning,convolutional neural network,imageomics,transfer learning and image segmentation emergence was in 2020.Conclusions The research hotspots of AI in the field of cancer mainly focus on diagnosis,gastric cancer,laparoscopy,Da Vinci robot and so on.Deep learning,convolutional neural network,imageomics,transfer learning and image segmentation are the research frontiers.
作者
朱新蕾
杜广芬
池晓超
ZHU Xin-lei;DU Guang-fen;CHI Xiao-chao(Department of Imaging,Fifth People's Hospital of Jinan,Jinan 250001,China)
出处
《社区医学杂志》
CAS
2022年第7期407-411,共5页
Journal Of Community Medicine
关键词
人工智能
肿瘤领域
可视化分析
研究热点
发展趋势
artificial intelligence
cancer field
visual analysis
research hotspots
development trend
作者简介
通信作者:朱新蕾,男,山东济南人,主管技师,主要从事医学数字图像处理、机器学习、辐射剂量及防护的研究工作。E-mail:1801313694@qq.com
