نمونه متن انگلیسی مقاله
Purpose The objective of this study is to analyse co-authorship and co-citation networks of publications in the field of artificial intelligence in ophthalmology and optometry. As well as, identify the different areas of research and the most cited publication.
Method A search of publications was performed in the Web of Science database for the period from 1977 to December 2021, using the term “Artificial Intelligence AND (Ophthalmol* OR optometry)”. The analysis of the publication was carried out using the Citation Network Explorer, VOSviewer and CiteSpace software.
Results 1086 publications and 2348 citation networks were found. 2020 was the year with the highest number of publications, a total of 351 publications and 115 citation networks. The most cited publication was “Clinically applicable deep learning for diagnosis and referral in retinal disease” published by De Fauw et al. in 2018, with a citation index of 723. Through the clustering function, three groups were found that cover the main research areas in this field: retinal pathology, anterior segment and glaucoma.
Conclusions The citation network analysis offers an in-depth analysis of scientific publications and the adoption of new topics and fields of research. The results of an exhaustive analysis of citation networks in artificial intelligence in the field of ophthalmology and optometry are presented since the publication of the first article in 1977.
Given the anticipated growth in the ageing population in the near future that will result in a higher rate of visual impairment and blindness, healthcare systems all around the world are making considerable efforts to improve eye-care.1
Nowadays, even in developed countries, the provision of ophthalmology consultations and care available is no longer sufficient if we take into account the increasing number of visually impaired patients.2 A study carried out in England found a permanent reduction in visual acuity and visual field in patients due to a 22-week delay in eye care. This could have been avoided if they were intervened earlier.3 This shows the urgent need for solutions to be implemented in order to improve the availability and accessibility of eye care services at primary, secondary, and tertiary level.
This study enabled us to obtain a comprehensive analysis of the available literature on the implementation of AI in the fields of ophthalmology and optometry. For this purpose, we chose to conduct a bibliographic search in the Web of Science database, given that its search range dates back to 1900. However, this database only considers journals of international relevance that have gone through a rigorous selection process.
The CitNetExplorer and CiteSpace software allowed us to determine the connections that exist between the different research fields and groups. By using the clustering function, we were able to group publications according to the relationships that exist between the citations. The drilling down function allowed us to conduct a more in-depth analysis of the bibliography for each group. Additionally, the core publications function helped us to identify the key publications for each group. Finally, scientometric analysis was used to obtain an important quantitative analysis of the existing literature to improve the understanding of this fast-growing field of knowledge.