Topical Analysis of Telemedicine Studies Using Text Mining Techniques

Abstract

The presence of diverse research in the field of telemedicine necessitates the need to analyze the topics of these studies to have a clear and comprehensive vision of this domain. The present study was conducted with the aim of topic modeling the published articles in the field of telemedicine using the PubMed database. This study employed a descriptive approach using LDA and TF-IDF as text mining techniques. Articles in the field of telemedicine were extracted from the PubMed database using the search formula “Telemedicine”[Majr] without any time limitations. A total of 37,181 records were retrieved. After data cleansing, the abstracts of 31,144 articles were analyzed, and topic modeling was performed. The topic modeling of telemedicine resulted in the identification of eight topical clusters, including e-health, interventions, primary care, remote monitoring, COVID-19, telehealth, cardiovascular disease, and research. The highest publication trend was observed for the COVID-19 topic, followed by primary care. Findings demonstrates the satisfactory performance of the LDA algorithm in topic classification in the field of telemedicine. Also, the results provide a better foundation for developing policies and research programs, as well as increasing awareness and utilization of emphasized topics. Modeling the topics of global telemedicine articles and comparing different algorithms with the current one is recommended for future research. Further studies in this field can lead to improved effectiveness, quality, and accuracy of telemedicine services. © (2024), (International Hellenic University School of Science and Technology). All rights reserved.