University of Macau Library

Macau Periodical Index (澳門期刊論文索引)

Author: Chen, Bing Hong；Ng, Hoi Ioi；Ke, Yan；Zhang, Wei；Wang, Gui Qi
Title: Global trends of artificial intelligence-assisted colonoscopy in the last 20 years: a bibliometric analysis
Journal Name: 澳門醫學雜誌
Pub. Info: 2024年1月31日, 第11卷第1期, pp. 3-17
Link: https://www.am.gov.mo/cn/journal/file/v11_1.pdf
Keyword: Global Trends；Artificial Intelligence；Colonoscopy；Bibliometric Analysis；Citespace
Abstract: Objective: With the emergence and rapid development of Artificial Intelligence (AI), AI-assisted applications have yielded exceptionally promising results in medical research, particularly in the field of colonoscopy. Early diagnosis and treatment of cancer have become a public health priority. This article aims to analyze current scientific achievements regarding the role of AI in colonoscopy, providing new ideas for further research. Methods: We retrieved publications from 2000 to 2021 from the Web of Science Core Collection (WoSCC), screening them based on inclusion criteria. We used CiteSpace 5.8.R3c and Microsoft Excel 2019 to evaluate and visualize the results, analyzing countries/regions, institutions, journals, authors, co-cited references, and keywords. Results: A total of 460 relevant papers were included. The number of papers increased steadily over the years, with the USA being the most productive country. The most productive institutions were Showa University and the University of Oslo, and the leading researcher was Yuichi Mori. The result of keyword analysis categorized all studies into five clusters: “Artificial Intelligence”, “polyp detection”, “water exchange”, “colorectal cancer” and “deep learning”. The current research hotspots were “Convolutional Neural Networks”, “Inflammatory Bowel Disease” and “Computer-aided Diagnosis”. Conclusions: Overall, with mounting evidence demonstrating the reliable role of AI in colonoscopy, auxiliary diagnosis of various intestinal diseases may become the primary focus of future research. Given the increasing prevalence of colorectal cancer, AI-assisted colonoscopy represents a potential research hotspot. Paragraph Headings： 1. Introduction 2. Materials and methods 2.1. Data Source and Search 2.2. Data Analysis 3. Results 3.1. Annual publications 3.2. Analysis of journals 3.3. Distribution of countries/regions and institutes 3.4. Analysis of authors 3.5. Analysis of reference co-citation 3.6. Cluster interpretations 3.7. Analysis of keywords 4. Discussion Tables : 1. Top 10 Journals and co-cited Journals 2. Top 10 productive countries/regions and Institutes 3. The 10 most productive authors and the top 10 co-cited authors 4. Top 10 co-cited references 5. The top 20 keywords with the highest frequency Figures : 1. Quantitative analysis process of the role of artificial intelligence in colonoscopy 2. Chronological distribution of publications in artificial intelligence in colonoscopy research from 2000 to 2021 3. A dual-map overlap of journals on AI-assisted colonoscopy research carried out by Citespace 4. CiteSpace network visual map of country/regions and institutions involved in AI-assisted colonoscopyresearch. (A) Collaboration analysis of countries/regions. (B) Collaboration analysis of institutions. Each circlerepresents a country/region/institutions. Size of circle is positively correlated with the number of articles publishedby country/region/institutions, and links between two circles represents a collaboration between two country/region/institutions on the same article. Line thickness is positively correlated with frequency of collaborations 5. CiteSpace network visualization map of co-cited authors of the articles related to AI-assisted colonoscopy research. The node size was positively associated with the cited counts of the authors, and the thickness of the lines between every two nodes represented the frequency of being co-cited between those two authors 6. Analysis of references in AI-assisted colonoscopy research. (A) Network map of co-cited references.Each circle represents a reference. Size of circle is positively correlated with frequency of citations, and links between two circles represent two references that were cited in the same article. (B) Clustered networks of cocitation status of the investigated 460 references via CiteSpace 7. Timeline view of co-cited references related to AI-assisted colonoscopy research. The bold timeline indicates that the clustering topic was a hotspot during this period. Citation tree-rings with different sizes on the timeline represent some key articles with a high citation frequency 8. (A) CiteSpace network visualization map of co-occurring keywords. (B)Top 20 burst keywords in articles related to AI-assisted colonoscopy research