Co-authorship Network Analysis of Medical Images Researchers with Emphasis on Micro and Macro Metrics

 Co-authorship is used to analyze scientific collaboration and identify patterns of collaboration among researchers. Considering the role of medical images in the field of health, it is necessary to identify the cooperation network between authors in order to strengthen research relations between them.

 The purpose of this study is to map and analyze the network structure of all authors of published articles in the field of medical images.

 This research is applied-descriptive and was done with a scientometric approach and social network analysis. The search strategy was implemented in the core collection of the Web of Science database Clarivate Analytics Institute. In this study, 37190 articles in the three time periods of 1991 - 2000, 2001 - 2010, and 2011 - 2020 were reviewed. Data extraction, matrix construction, and mapping of the co-authorship network were performed using Bibexcel, Gephi, and Vosviewer software.

 During the period under review, the pattern of authors’ participation changed from 2 authors to 3 authors and 4 authors. In the years 1991 - 2000, the link strongest with values of 18, 16, and 14 were “Dipaola, R.”, “Frouin, F.” and “Nishikawa, R. M.”, respectively. The co-authorship network consisted of seventy clusters in the years 2001 - 2010, and its strongest members were “Alkadhi, Hatem” and “Leschka, Sebastian” with a total link strength of 100. The co-authorship network in 2011 - 2020 consisted of 60 clusters and the link strongest with values of 58, 55, and 50 belonged to “Van Ginneken, Bram”, “Herrmann, Ken”, and “Ourselin, Sebastien”, respectively. In 2001 - 2010, the network density and clustering coefficient were 0.007 and 0.994, respectively.

 In all 3 decades, the co-authorship network is incoherent. In the 2001 - 2010-decade, 7% of the potential relationships in the co-authorship network were realized. Dispersion in the co-authorship network researchers in the field of medical images is evident. In addition, the amount of density and clustering coefficient of the co-authorship network indicates the greater willingness of authors to collaborate in this decade. The results of this research can be used to expand and strengthen scientific cooperation between researchers.