Machine learning in glaucoma: a bibliometric analysis comparing computer science and medical fields’ research

Abstract

ABSTRACT The aim of this study was to analyze the current literature on the use of machine learning in glaucoma, comparing the characteristics and citations received by articles on computer science focus or medical focus. We performed a search using the Scopus database on 28th of January 2021 using appropriate keywords for journal articles and conference articles that discussed glaucoma in the context of machine and deep learning. We used Scopus-based field classification and compared different characteristics and citation metrics between articles classified as belonging to the computer science field, medical field, or both fields combined. A total of 858 documents resulted from the search. Upon comparing the mean citation received by publications in the computer science field and medical field, we found a significant difference (p = 0.013). The highest mean citation received was for articles in the combined fields with a mean of 26.2 (SD 41.7), and the least mean citations received by articles in computer science field with a mean of 13.7 (SD 34.6). Most articles related to machine learning assessment of glaucoma were classified as computer science articles. Articles that belong to medical or both fields combined received higher number of citations.