Background: Artificial Intelligence (AI) makes the healthcare industry modern and responsive by enhancing diagnosis, description, and planning for treatment. Machine learning and deep learning algorithms have made it possible to automate diagnosis, improve the quality of images, and make medicine more personalized. But there are still ethical and privacy issues to think about. Objective: This study performs a bibliometric analysis to map key trends, thematic clusters, and author collaborations in AI in healthcare using ScienceDirect and Scopus-indexed literature published between January 2021 and August 2025. Methodology: A systematic review was performed for ScienceDirect with the keywords Artificial Intelligence in Healthcare or AI in Medicine to retrieve research papers. The growth in publications, productive authors, top-cited research articles, resources, and authors’ collaboration publications in AI in healthcare for keyword network analysis figures using bibliometric analysis(specifically using the Bibliometrix Shiny dashboard) were identified. Key Findings / Results: The number of AI publications in the healthcare and medical fields has grown quickly each year with an average growth rate of 62.66% from 2021 to 2024. Yogesh K. Dwivedi, A.S. Albahri, Patrick Mikalef, Rajesh Gupta, and Suddeep Tanwar are the most productive authors identified in this study. A paper by Chanyuan Zhang is one of the most cited manuscripts. It has 2,293 authors and 2,442 author appearances. The keyword co-occurrence network helps find patterns and research trends in this field. Central keywords show well-established research areas, while peripheral nodes show new or niche topics. The network reporting shows that AI in healthcare is a field of study that brings together different areas of research, such as diagnostics, data analytics, medical informatics, and patient monitoring. Conclusion / Implications: By applying bibliometric analysis to the literature identified from ScienceDirect, this study offers valuable insights into emerging trends, thematic patterns, and strategic directions in the field of AI in healthcare. ScienceDirect was selected due to its high-quality peer-reviewed content, institutional access rights, and strong emphasis on technology and applied sciences. However, relying solely on one database is a limitation, as it may not fully represent the breadth of influential research in this area. Future studies should consider incorporating additional databases such as Scopus, Web of Science, and PubMed to improve the comprehensiveness, representativeness, and generalizability of the results. Further research could also examine the integration of AI and the Internet of Things (IoT) in clinical healthcare settings to enable intelligent diagnostics, personalized treatment, and better patient outcomes.
Pujari et al. (Sat,) studied this question.