Abstract: History and Future of Artificial Intelligence Related Health Care Research: A Bibliometric Analysis (Society for Social Work and Research 25th Annual Conference - Social Work Science for Social Change)

All live presentations are in Eastern time zone.

558P History and Future of Artificial Intelligence Related Health Care Research: A Bibliometric Analysis

Schedule:
Tuesday, January 19, 2021
* noted as presenting author
Zhichao Hao, MSW, PhD Candidate, University of Alabama, Tuscaloosa, AL
Yuqi Guo, PhD, Assistant Professor, University of North Carolina at Charlotte, Charlotte, NC
Fan Yang, PhD, Assistant Professor, Dongbei University of Finance and Economics, Dalian, China
Background: As a critical driving power to promote health care, AI has especially become an indispensable component to advance the development and innovation of health care and medical diagnosis. The implementation of AI technologies fosters the prediction, diagnosis, and treatment of diseases, which benefits both patients and health care providers. The development of healthcare-related AI research and the direction for the patterns and trends in the future need to be understood. This study aims to do a dynamic and longitudinal bibliometric analysis on healthcare-related AI publications to explore the characteristic of research activities and research hotspot tendencies.

Methods: The Web of Science (WoS) was searched to retrieve all existing and highly cited Artificial Intelligence related healthcare research papers published in English up to December 2019. Based on bibliometric indicators, a search strategy was developed to screen title for eligibility, using the abstract and full-text where needed. The characteristics of research activities, and research hotspot tendencies were computed by the HistCite software.

Results: The top five health problems are cancer, depression, Alzheimer’s disease, heart failure, and diabetes. The top five AI technologies are machine learning, artificial neural network, deep learning neural network, electronic health record, and support vector machine. The top five functions are case classification, diagnosis, prediction, risk estimate, and chronic condition management. The top five populations focused by healthcare-related AI research are children, adult, women, men, and elders. We presented the major milestones in the development of AI in health care by analyzing the list of keywords that have strong citation bursts between 1996 and 2019. The first milestone keywords in the studies are neural network, logistic regression, and carcinoma. The next milestone is artificial neural network (highly cited in 1998-2002). The most recent milestone is survival analysis, highly cited until 2014.

Implications: Keeping abreast of the fast-growing body of healthcare-related AI research helps practitioners and policymakers to catch the opportunities of applying AI interventions to promote the well-being of patients and their caregivers. AI research on healthcare is accelerating rapidly, with potential applications being demonstrated across various domains of medicine. However, there are currently limited examples of such techniques being successfully deployed into clinical practice. Future AI research should dedicate to fill in the gap between AI healthcare research and clinical application.