A Bibliometric Analysis of 14,822 Researches on Myocardial Reperfusion Injury by Machine Learning

• Main Authors: Chan Li, Zhaoya Liu, Ruizheng Shi
• Format: Article
• Language: English
• Published: MDPI AG 2021-08-01
• Series: International Journal of Environmental Research and Public Health
• Subjects: myocardial reperfusion injury; bibliometric analysis; LDA analysis; machine learning; MeSH term
• Online Access: https://www.mdpi.com/1660-4601/18/15/8231
• ISSN: 1661-7827; 1660-4601

Myocardial ischemia is the major cause of death worldwide, and reperfusion is the standard intervention for myocardial ischemia. However, reperfusion may cause additional damage, known as myocardial reperfusion injury, for which there is still no effective therapy. This study aims to analyze the landscape of researches concerning myocardial reperfusion injury over the past three decades by machine learning. PubMed was searched for publications from 1990 to 2020 indexed under the Medical Subject Headings (MeSH) term “myocardial reperfusion injury” on 13 April 2021. MeSH analysis and Latent Dirichlet allocation (LDA) analyses were applied to reveal research hotspots. In total, 14,822 publications were collected and analyzed in this study. MeSH analyses revealed that time factors and apoptosis were the leading terms of the pathogenesis and treatment of myocardial reperfusion injury, respectively. In LDA analyses, research topics were classified into three clusters. Complex correlations were observed between topics of different clusters, and the prognosis is the most concerned field of the researchers. In conclusion, the number of publications on myocardial reperfusion injury increases during the past three decades, which mainly focused on prognosis, mechanism, and treatment. Prognosis is the most concerned field, whereas studies on mechanism and treatment are relatively lacking.