ABSTRACT
Deep Learning (DL) literature is growing as the main engine for encouraging medical care. This study provides a complex and longitudinal bibliometric review of artificial intelligence (AI) publications related to healthcare. Until September 2020, the Web of Science (WOS) was searched to find all current AI research documents written in English. A search strategy based on bibliometric indicators was developed using the Knowledge Matrix Plus and VOSviewer software to screen the eligibility title. The growth rate of publications has trended in science, including years, authors, keywords, and points of interest. The results of the search included 520 publications. Publication performance has increased by 16.90% annually since 2018, but research paper production has increased significantly from 2018 to 2020 to 50.72%. The main health issues examined in DL research are data, algorithms, machine learning, images, electronic health records, neural networks, and deep neural networks. The most significant health effects are artificial neural networks and support vector machines. Nucleosides, neural networks, and tumour markers have remained at the forefront of research topics through 2019. This review offers a detailed analysis of deep-level learning health research, which helps researchers, policymakers, and clinicians understand the evolution of in-depth learning research in healthcare and its implications for practice. Future DL research can address the gaps between DL research in health care and clinical applications.
