This paper is in the following e-collection/theme issue:

Scientometrics, Infometrics, and Altmetrics (94) Artificial Intelligence (1371) Geriatrics (24) Health Services Research (482) Aging with Chronic Disease (110) Aging in Place (283)

Published on in Vol 25 (2023)

Preprints (earlier versions) of this paper are available at https://preprints.jmir.org/preprint/46014, first published .
Application of Artificial Intelligence in Geriatric Care: Bibliometric Analysis

Application of Artificial Intelligence in Geriatric Care: Bibliometric Analysis

Application of Artificial Intelligence in Geriatric Care: Bibliometric Analysis

Authors of this article:

Jingjing Wang1, 2 Author Orcid Image ;   Yiqing Liang1, 2 Author Orcid Image ;   Songmei Cao1 Author Orcid Image ;   Peixuan Cai2, 3 Author Orcid Image ;   Yimeng Fan1, 2 Author Orcid Image
Article Authors Cited by (8) Tweetations (4) Metrics

    Original Paper

    1Department of Nursing, The Affiliated Hospital of Jiangsu University, Zhenjiang, China

    2Medical College, Jiangsu University, Zhenjiang, China

    3Department of Geriatrics, The Affiliated Huaian No 1 People's Hospital of Nanjing Medical University, Huaian, China

    Corresponding Author:

    Songmei Cao, PhD

    Department of Nursing

    The Affiliated Hospital of Jiangsu University

    No 438 North Jiefang Road, Jingkou District, Jiangsu Province

    Zhenjiang, 212000

    China

    Phone: 86 13815159881

    Email: caosongmei75@126.com


    Background: Artificial intelligence (AI) can improve the health and well-being of older adults and has the potential to assist and improve nursing care. In recent years, research in this area has been increasing. Therefore, it is necessary to understand the status of development and main research hotspots and identify the main contributors and their relationships in the application of AI in geriatric care via bibliometric analysis.

    Objective: Using bibliometric analysis, this study aims to examine the current research hotspots and collaborative networks in the application of AI in geriatric care over the past 23 years.

    Methods: The Web of Science Core Collection database was used as a source. All publications from inception to August 2022 were downloaded. The external characteristics of the publications were summarized through HistCite and the Web of Science. Keywords and collaborative networks were analyzed using VOSviewers and Citespace.

    Results: We obtained a total of 230 publications. The works originated in 499 institutions in 39 countries, were published in 124 journals, and were written by 1216 authors. Publications increased sharply from 2014 to 2022, accounting for 90.87% (209/230) of all publications. The United States and the International Journal of Social Robotics had the highest number of publications on this topic. The 1216 authors were divided into 5 main clusters. Among the 230 publications, 4 clusters were modeled, including Alzheimer disease, aged care, acceptance, and the surveillance and treatment of diseases. Machine learning, deep learning, and rehabilitation had also become recent research hotspots.

    Conclusions: Research on the application of AI in geriatric care has developed rapidly. The development of research and cooperation among countries/regions and institutions are limited. In the future, strengthening the cooperation and communication between different countries/regions and institutions may further drive this field’s development. This study provides researchers with the information necessary to understand the current state, collaborative networks, and main research hotspots of the field. In addition, our results suggest a series of recommendations for future research.

    J Med Internet Res 2023;25:e46014

    doi:10.2196/46014

    Keywords

    artificial intelligence; older adults; geriatric care; bibliometric analysis 


    The older adult population is growing rapidly and posing unprecedented challenges worldwide. The United Nations (UN) data shows that the older adult population aged ≥60 years is expected to reach around 2 billion worldwide by 2050 [1]. The rapid growth of aging is putting enormous pressure on some societies and countries, impacting their economies, workforce structures, and social security systems [2]. As adults age, their physical function declines and they are at high risk for chronic diseases. Fong [3] showed that 85% of older people have at least 1 chronic illness. The demand for medical services for older people is increasing [4]. In addition, many countries are facing a shortage of caregivers, with a global shortage of 17 million health care providers in 2019 [5,6].

    Artificial intelligence (AI) is a science that involves multiple fields of knowledge and encompasses a wide range of techniques and aims to simulate and extend human intelligence through machines, and research fields include expert systems, machine learning, robotics, decision support systems, and pattern recognition [7,8]. AI is developing quickly [8]. It has already been used in geriatric care, alleviating the shortage of caregivers and the uneven distribution of resources [6,9]. A growing number of researchers believe that AI can effectively address the unmet needs of older adults further by reallocating the distribution of nurses and other health care resources. Many researchers have conducted studies on the application of AI in geriatric care, such as the care for patients with Alzheimer disease, geriatric care, disease recognition, and medication reminders [10-16].

    Nevertheless, the field’s research themes and trends are not yet clear. Currently, there are no studies that analyze the current status and hotspots of research on the application of AI in geriatric care in the form of bibliometrics.

    Bibliometrics refers to the quantitative synthesis of publications using mathematical and statistical methods. This method analyzes the development trends, distribution structure, and relationships within a specific field based on the external characteristics of publications, such as the number of publications, authors, and journals. In this study, we used bibliometric methods to conduct a comprehensive analysis of the research related to the application of AI in geriatric care in order to summarize the status of research in the field. This research also analyzed research hotspots and development trends as a reference for future research directions and researchers.


    Data Sources and Search Strategy

    We retrieved the publications used in this study from the Web of Science Core Collection from inception to August 2022. We formulated a search strategy concurrent to reading the relevant literature that had been identified previously. Our search used the following keywords: (“artificial intelligence” OR “machine intelligence” OR “robot*” OR “robot technology” OR “assistant robot” OR “robot-assisted” OR “computational intelligence” OR “computer reasoning” OR “deep learning” OR “computer vision system” OR “sentiment analysis” OR “machine learning“ OR “neural network*” OR “data learning“ OR “expert* system*” OR “natural language processing“ OR “support vector machine*” OR “decision tree*” OR “data mining“ OR “deep learning” OR “neural network*” OR “bayesian network*” OR “intelligent learning” OR “feature* learning” OR “feature* extraction” OR “feature* mining” OR “feature* selection” OR “unsupervised clustering” OR “image* segmentation“ OR “supervised learning” OR “semantic segmentation” OR “deep network*” OR “neural learning” OR “neural nets model“ OR “graph mining“ OR “data clustering“ OR “big data“ OR “knowledge graph”) AND (“senior∗” OR “aged∗” OR “elder∗” OR “geri∗” OR “age∗” OR “older adult∗” OR “geriatric∗” OR “aging” OR “very elderly” OR “frail elderly”) AND (“nurs*” OR “care” OR “psychological care” OR “bibliotherapy”). Inclusion criteria were limited to studies that (1) involved AI technologies, (2) were related to older adults, and (3) were published as papers. Exclusion criteria were also defined with source types, such as conference papers, reviews, data papers, editorial materials, and retraction. After the first search, all the literature was screened and checked separately by 2 researchers to ensure that all the papers used were relevant to the study topic. In the case of a dispute, we consulted a third investigator.

    Analysis Tools

    The following tools were used to analyze the literature:

    • HistCite 12.03.17, the history of citation, is a citation analysis tool developed by Eugene Garfield, the inventor of the Science Citation Index (SCI). HistCite can provide an overview of the history of a field’s development and recognize the most influential literature and authors [17]. It was used in this study to generate all tables.
    • VOSviewer 1.6.18, a free Java software for document mapping, was developed by the Centre for Science and Technology Studies, University of Leiden, the Netherlands. It was used in this study to analyze collaborative networks and the keyword co-occurrence network.
    • We used the Bibliometrix Online Analysis Platform to present the collaborative networks among countries/regions [18,19]. It was used in this study to analyze a network map of scientific cooperation among countries/regions.
    • Citespace 6.2.R2, a scientometric tool developed by Chaomei Chen, was used to report burst keywords in this study [20].

    Data Analysis

    The analysis function in the Web of Science database was used to summarize external characteristics, including the publication number and the average number of citations. Literature records and cited reference data were downloaded in text form. Keywords with similar meanings but different spellings or expressions were merged (eg, “elder people” and “older adults”). Furthermore, general and nonsensical keywords were removed (eg, “men,” “scale”). Once cleaned, the data were imported into the aforementioned software separately for internal structure analysis.

    Ethical Considerations

    The data used in this study were obtained from the Web of Science Core Collection, and no patients or public contributions were involved in this research.


    Analysis of Publication Outputs and the Total Local Citation Score

    A total of 4641 papers were searched from the Web of Science Core Collection, and 230 (4.96%) papers were included in the analysis after the data were cleaned. Figure 1 shows the literature-screening process and research framework, and the annual number of publications is shown in Figure 2. The number of publications on the application of AI in geriatric care has steadily increased, from the first paper in 2000 to 32 papers in 2022. Publications increased sharply from 2014 to 2022, accounting for 90.87% (209/230) of all publications. However, the variation in the total local citation score (TLCS) in this field is not smooth. In contrast, the later the papers are published, the lower their times of being cited. The higher the TLCS, the more important the publication in the field. The TLCS peaked 4 times during the period covered by our study (in 2004, 2013, 2015, and 2017), suggesting that these were pivotal years in which seminal literature may have been published.

    Our results support the correlation between the peaks in the TLCS and the publication of seminal literature. We found that Tamura et al [21] conducted a study of 13 older adults with severe Alzheimer disease to investigate the effectiveness of entertainment robots in occupational therapy in 2004. The results showed that during occupational therapy, entertainment robots effectively increase the social activity of patients with Alzheimer disease. The entertainment robot is an effective rehabilitation tool for the treatment of patients with severe dementia. In addition, the study [21] also proposed that entertainment robots could be substituted for therapy animals in critical care units. Robots are safer than animals, but their price is another issue. Opinions differ as to whether using robots would be worthwhile. We also found that Moyle et al [22] conducted a pilot randomized controlled trial in 2013 to explore the effects of companion robots on the emotional expression of older adults with moderate-to-severe dementia in a residential care setting. The findings indicated that companion robots have a moderate-to-very-positive influence on patients quality of life, resulting in significant improvements in speech function and emotional expression and a reduction in loneliness [22]. In 2015, Sung et al [23] carried out a 1-group before-and-after study to evaluate the effects of robot-assisted therapy on older adults in many long-term care facilities. The participants’ communication, interaction skills, and activity levels significantly improved after receiving 4-week robot-assisted therapy [23]. Robot-assisted therapy could be a daily activity and has the potential to improve the social health of older adults in residential care facilities. In 2017, Hebesberger et al [24] used a mixed methods study to evaluate the social acceptance of and experiences with long-term autonomous robot use by staff and older adults in nursing facilities. This being the first study to focus on the impact of long-term robot exposure on user experience and acceptance, the results showed that older adults and staff felt ambivalent about the robot. On the one hand, they were curious about and engaged with the robot; they enjoyed its presence. On the other hand, the older adults expressed fear and rejected the use of the robot, and the staff was not willing to share their workspace with a robotic aid [24].

    ‎
    Figure 1. Flowchart of the literature-screening process and research framework.
    ‎
    Figure 2. Publication output and the TLCS over time. TLCS: total local citation score.

    Analysis of Countries/Regions and Institutions

    A total of 39 countries/regions and 499 institutions had participated in the publication of the 230 papers. The top 10 countries/regions for publications are shown in Table 1. The United States had the highest number of publications during the study period (n=63, 27.39%), followed by China (n=41, 17.83%) and Japan (n=28, 12.17%). In terms of the number of papers published, the top 10 countries/regions exceeded 90.00% of all the papers, suggesting that research in this area is unevenly developed among countries/regions.

    Table 2 shows the top 10 institutions in terms of publications in the field. Griffith University and the University of Auckland had the highest number of publications during the study period (n=10, 4.35%), followed by the University of Toronto (n=6, 2.61%), Bond University (n=5, 2.17%), and the University of Pennsylvania (n=5, 2.17%).

    Wuchty et al [25] showed that teams often publish more impactful research than individuals and that analysis of the collaborative relationships among different countries/regions, institutions, and authors can also reflect the scholarly exchange in this field. Figure 3 shows the cooperation among countries/regions. The line between countries reflects the cooperation among them, and they are mostly in cooperation with one another. It is worth noting that the more frequent the exchange among countries, the more the output, and this is the case with the United States. The figure shows that the lines crossing with the United States are numerous and thick, indicating that the United States plays an important role in the development of cooperation in this field. Thus, the United States also produces the most output and cooperates closely with China, Canada, Germany, France, and Japan.

    Table 1. Top 10 countries/regions by number of publications.
    RankCountryPublications, n (%)TLCSa
    1United States63 (27.39)39
    2China41 (17.83)17
    3Japan28 (12.17)17
    4Australia24 (10.43)41
    5United Kingdom23 (10.00)36
    6Germany19 (8.26)21
    7Canada14 (6.09)18
    7Italy14 (6.09)4
    8France13 (5.65)10
    8New Zealand13 (5.65)21

    aTLCS: total local citation score.

    Table 2. Top 10 institutions by number of publications.
    RankInstitutionPublications, n (%)TLCSa
    1Griffith University (Australia)10 (4.35)27
    1The University of Auckland (New Zealand)10 (4.35)21
    2University of Toronto (Canada)6 (2.61)6
    3Bond University (Australia)5 (2.17)4
    3University of Pennsylvania (United States)5 (2.17)2
    4Queensland University of Technology (Australia)4 (1.74)16
    5Angers University Hospital (France)3 (1.30)2
    5Columbia University (United States)3 (1.30)0
    5University of Heidelberg (Germany)3 (1.30)0
    5Maastricht University (the Netherlands)3 (1.30)3

    aTLCS: total local citation score.

    ‎
    Figure 3. Network map of scientific cooperation among countries/regions.

    Analysis of Journals and Citations

    Journal analysis can help find the core journals in this field [26,27]. AI has been widely used in various applications, including games, automation, medicine, and process control [28]. Increasingly more researchers and scientists have begun to engage in this theme. The analysis revealed that 124 journals were involved in research on the application of AI in geriatric care during our study period. Table 3 shows the top 10 journals in terms of the number of publications and their respective TLCS. The International Journal of Social Robotics published the largest number of publications (n=22, 9.57%) on this topic. The impact factor of this journal is 3.802. It was followed closely by the Journal of the American Medical Directors Association (n=11, 4.78%), Assistive Technology (n=6, 2.61%), and the Journal of Medical Internet Research (n=6, 2.61%). The journals with the highest TLCS were the International Journal of Social Robotics (TLCS=27) and the Journal of the American Medical Directors Association (TLCS=32), indicating that these 2 journals are likely to be the definitive publications for the application of AI in geriatric care.

    Table 3. Top 10 journals that published papers on the application of AIa in geriatric care.
    RankJournalPublications, n (%)TLCSbIFc
    1International Journal of Social Robotics22 (9.57)273.802
    2Journal of the American Medical Directors Association11(4.78)327.802
    3Assistive Technology6 (2.61)52.170
    3Journal of Medical Internet Research6 (2.61)07.077
    4Journal of Alzheimers Disease5 (2.17)174.160
    5Advanced Robotics4 (1.74)22.057
    5BMC Geriatrics4 (1.74)04.070
    5Frontiers in Aging Neuroscience4 (1.74)05.702
    5Journal of Clinical Nursing4 (1.74)14.423
    5Journal of Gerontological Nursing4 (1.74)171.436

    aAI: artificial intelligence.

    bTLCS: total local citation score.

    cIF: impact factor (Journal Citation Reports 2021).

    Analysis of Publications and Citations

    To some extent, the citation ranking of a paper can also explain the research hotspots in the academic field [29]. The title of a paper states the paper’s subject. Based on the TLCS, Table 4 lists the top 10 most frequently cited publications. Of the top 10 papers, 7 (70%) focused on the impact of robotics on people with Alzheimer disease, indicating that the application of AI to patients with dementia is the primary research question.

    Table 4. Top 10 papers by the TLCSa.
    RankTitle (author)YearJournalTLCS
    1Exploring the Effect of Companion Robots on Emotional Expression in Older Adults With Dementia: A Pilot Randomized Controlled Trial (Moyle et al [22])2013Journal of Gerontological Nursing17
    2The Utilization of Robotic Pets in Dementia Care (Petersen et al [30])2011Journal of Alzhimers Disease16
    3Effects on Symptoms of Agitation and Depression in Persons With Dementia Participating in Robot-Assisted Activity: A Cluster-Randomized Controlled Trial (Joranson et al [31])2015Journal of the American Medical Directors Association12
    4Is an Entertainment Robot Useful in the Care of Elderly People With Severe Dementia? (Tamura et al [21])2004Journal of Gerontology Series A-biological Sciences and Medical Sciences11
    5Attitudes Towards Health-Care Robots in a Retirement Village (Broadbent et al [32])2012American Journal of on Ageing10
    5Use of a Robotic Seal as a Therapeutic Tool to Improve Dementia Symptoms: A Cluster-Randomized Controlled Trial (Moyle et al [33])2017Journal of the American Medical Directors Association10
    6In the Hands of Machines? The Future of Aged Care (Sparrow and Sparrow [34])2006Minds and Machines8
    7Care Staff Perceptions of a Social Robot Called Paro and a Look-Alike Plush Toy: A Descriptive Qualitative Approach (Moyle et al [35])2018Aging & Mental Health7
    8Social Commitment Robots and Dementia (Roger et al [36])2012Canadian Journal on Aging-Revue Canadienne Du Vieillissement6
    9Robot-Assisted Therapy for Improving Social Interactions and Activity Participation Among Institutionalized Older Adults: A Pilot Study (Sung et al [23])2015Asia-Pacific Psychiatry5

    aTLCS: total local citation score.

    Analysis of Authors and Coauthorship Networks

    The high citation count (h-index) was proposed by Jorge E Hirsch of the University of California, San Diego, USA, in 2005. It is a mixed quantitative metric used to assess the scholarly achievement of researchers. The higher the h-index, the greater the academic impact. The h-index indicates that a person has “h” papers, each of which has been cited at least “h” times in a given period [37]. In total, 977 authors took part in related studies in the past. Table 5 presents the top 10 authors in terms of the number of published papers. The top 3 (30%) are Wendy Moyle (Australia), Cindy Jones (Australia), and Elizabeth Broadbent (New Zealand). Moyle has the most publications, and Ngaire has the highest h-index (h-index=46), which means that they are extremely influential in their field. Figure 4 illustrates the collaborative relationships among authors in this field. The different colors stand for different clusters of authors. The analysis of collaboration among all authors showed that only 38 (3.13%) authors formed a collaborative network and were divided into 5 main clusters. The thickness of lines indicates the strength of the relationship between authors relative to others. The largest cluster (in red) involved 11 (28.9%) coauthors, centering on Baishch Stefanice, Knopf Oinks, and Kolling Thorsten. The second cluster (in blue) was dominated by Moyle and Jones. They had the largest nodes and were the most active coauthors in the field. The smallest cluster (in purple) consisted of only 5 authors, but they collaborated with the largest cluster and the authors with the highest number of publications through Barbara Klein. Meanwhile, there was a lack of collaboration among the other top 10 authors involved in research in this field of publication.

    Table 5. Top 10 authors who published research papers.
    RankAuthorPublications, n (%)TLCSaTGCSbh-Index
    1Moyle (Australia)11 (4.78)4443626
    2Jones (Australia)10 (4.35)3433320
    3Broadbent (New Zealand)9 (3.91)2136140
    4Kerse (New Zealand)5 (2.17)1223846
    4MacDonald (New Zealand)5 (2.17)1827226
    5Beattie (Australia)4 (1.74)2322828
    5Johnson (United States)4 (1.74)211216
    5Pu (China)4 (1.74)4437
    5Thalib (Istanbul)4 (1.74)1618637
    6Muramatsu (New Zealand)3 (1.30)153 12

    aTLCS: total local citation score.

    bTGCS: total global citation score.

    ‎
    Figure 4. Visualization of research networks of authors with a minimum of 1 paper [38].

    Co-occurrence of Keywords and Burst Keyword Detection

    Keywords are effective in analyzing the knowledge structure of an academic field from a bibliometric perspective, which can help identify potential research hotspots [39]. Thus, the themes covered in this study could be identified using the keywords. We analyzed the keywords that appeared more than 2 times. This was calculated using a clustering algorithm similar to modularity-based clustering. The titles and abstracts of 230 papers were included in the analysis; we extracted the keywords with the top 140 occurrences and presented them visually. As seen in Figure 5, the colors of the elements stand for the clusters to which they belong and the different clusters are represented by assigned colors. The node size indicates the occurrence of the keyword, and the thickness of the link represents the co-occurrence intensity. The thicker the link between nodes, the greater the co-occurrence between keywords. We clearly divided the keywords into 4 categories using 4 colors (red, green, blue, and yellow) to indicate that these topics demonstrate the mainstream research hotspots and frontier areas.

    Cluster 1 is associated with Alzheimer disease. The primary keywords were “cognitive impairment,” “health care,” “prevention,” “prediction,” and “artificial intelligence.” This cluster explores the application of AI in Alzheimer disease. Cluster 2 is primarily related to aged care. The main keywords were “assistive robotic,” “rehabilitation,” “stroke,” and “socially assistive robots.” The cluster focuses on the application of AI in geriatric care. Cluster 3 is related to the acceptance of AI applications. The keywords were “user acceptance,” “robot acceptance,” and “technology acceptance models.” The cluster explores the acceptance of the applications of AI in geriatric care. Cluster 4 is related to the surveillance and treatment of diseases in older adults. The essential keywords were “machine learning,” “therapy,” “robotic surgery,” “frailty,” and “risk factors.” This cluster explores the application of AI in the monitoring and treatment of diseases in older adults. These 4 themes constitute the mainstream academic literature on the application of AI in geriatric care.

    Keywords that appear suddenly and are cited extensively, or relatively so, for a short period are known as burst keywords [20]. They are found using the default Kleinberg algorithm of CiteSpace. Considered important indicators of frontier research hotspots, burst keywords herald emerging trends. Figure 6 shows the top 19 burst keywords in this field for 2000-2022. The thick red line shows the period of the keyword’s outbreak. Machine learning (4.22) had the highest burst strength, followed by mild cognitive impairment (2.38). In addition, “health service facility” was the keyword with the longest burst (6 years). Recent burst keywords included “machine learning,” “deep learning,” and “rehabilitation.” These burst keywords also reflect the research trends in the field and may become a research hotspot in the future.

    ‎
    Figure 5. Keyword co-occurrence network [38].
    ‎
    Figure 6. Top 19 keywords with strong citation bursts between 2000 and 2022.

    Principal Findings

    Our study shows that research on the application of AI in geriatric care has gradually increased over the past 23 years. Nevertheless, the field has developed relatively slowly and still has a lot of research space. Moreover, the development of research has been uneven among countries/regions, and cooperation among the countries/regions and institutions in which the field has been less developed is more limited. This may be related to the country’s degree of economic development and the size of its aging population. According to the 2019 global ranking of countries with aging populations, Japan, the United Kingdom, Australia, the United States, and China all rank at the top [40]. In addition, most are high-income countries. Meanwhile, the United States has the highest number of publications, which may be related to its state support. On February 11, 2019, the United States issued the executive order “Maintaining American Leadership in Artificial Intelligence,” directing all federal government agencies to implement strategic objectives aimed at accelerating AI research and development [41].

    In this study, we hoped to identify research hotspots and gaps by providing a comprehensive analysis and structured information about the field. Through the analysis of highly cited studies in publications, we found that the application of robotics in the care of patients with Alzheimer disease is the primary focus of the field, and this was validated by the cluster analysis of the keywords.

    Alzheimer disease is a progressive degenerative disease of the nervous system, with an insidious onset. It is a clinical syndrome, characterized by memory impairment, aphasia, agnosia, executive dysfunction,personality, and behavioral changes, etc [42]. Studies have reported that the number of people with Alzheimer disease worldwide will increase from 57 million in 2015 to 152 million in 2050, with a significant increase in its prevalence among older adults [43,44]. Many countries are facing a shortage of caregivers for patients with Alzheimer disease [45]. With the development of information technology, many studies have focused on the application of AI to the treatment of older people with Alzheimer disease. For example, companion robots improve the quality of life, mood, and cognitive abilities of patients with Alzheimer disease [46,47]. Several studies have demonstrated the effectiveness and feasibility of AI in Alzheimer care [48]. However, various factors, including cost and environmental constraints, limit the use of AI in the treatment of patients with Alzheimer disease [49]. Promoting the use of AI in the treatment of patients with Alzheimer disease may be a direction for future research.

    The application of AI in geriatric care is another research hotspot in this area. As aging continues to deepen, geriatric care has become a social issue in urgent need of being solved. Currently, AI is used widely in the daily care of older people. AI can assist older adults with their daily tasks, such as eating, bathing, and dressing [50-52]. It can also help rehabilitate older adults with disabilities, improving their quality of life and increasing their independence [53-56]. Concurrently, many other researchers have explored socially assistive robots. For example, these robots are said to be promising in terms of alleviating feelings of loneliness and social isolation [57-59]. However, there are still 2 problems that we should think about: first, how we can guarantee the safety of AI in the home environment, and second, how older adults face the “digital divide” brought about by AI.

    In recent years, the acceptance of technology has become a hotspot. Needless to say, the population’s acceptance of technology affects the application and development of AI. If the target user is resistant to AI, they will be immune to its advantages. Hence, several recent studies have focused on people’s acceptance of AI. Some scholars have investigated the acceptable behavior and influencing factors of older people for smart geriatric care services, companion robots, daily care robots, and social assistance robots for care and companionship in daily life [60-65]. One study examined the acceptance of home care robots by older people in Finland, Ireland, and Japan. Intriguingly, the results showed that each country has a different view. More older people in Finland have a negative impression of robots compared to their peers in the other 2 countries [66]. These differences may be related to the history, culture, policies, and values associated with the development of AI in each country. We give full consideration to the actual situation of each country throughout the process of promoting the application of AI in the future. However, there are fewer studies on how to increase people’s acceptance of AI, which may be a direction for future research.

    Through an analysis of the literature, we found that AI plays a vital role in monitoring the diseases and treatment of older people. The declining physiological function in older people leads to a significant increase in their morbidity and disability. This phenomenon has led to an increasing demand for medical services by older people [4]. AI can help health care workers not only monitor but also treat the medical conditions of older people. Furthermore, academics have paid a lot of attention to this research hotspot. Health care professionals can monitor the status of older people through telemedicine, for example, wearable smart devices and robots, which can help them assess and improve their patients’ conditions remotely and dynamically [67,68]. In recent years, several researchers have studied the application of machine learning, which is mostly used in the diagnosis and management of diseases. Examples include screening populations with mild cognitive impairment (MCI) or early Alzheimer disease and predicting the incidence of delirium risk after patients’ hip fracture surgeries [69,70]. Future research could try to use machine learning to monitor other disease conditions in older people, and it might be a research direction worth pursuing.

    Limitations

    Inevitably, we need to acknowledge the limitations of this study. First, due to the applicability of the software, the study only searched the Web of Science database, which is the most influential multidisciplinary academic literature abstract index database worldwide. Holding more than 12,400 authoritative and high-impact academic journals worldwide, the core collection of this database has become an important tool for academic analysis and evaluation. However, because we limited our search to just 1 database, we may have missed some important research results [71]. Additionally, our keyword-cleaning process and statistics are self-designed and may be limited by our professional knowledge and experience. In the future, we will further expand our data sources and standardize keywords to help us enhance the overall quality of the paper and the accuracy of our forecasts.

    Conclusion

    In this bibliometric analysis study, we delineated the trajectory of research on AI in geriatric care in the past 23 years. Nevertheless, the development of research and cooperation among countries/regions and institutions are limited in a number of instances. Study suggests that researchers should focus on interinstitutional collaborations, especially international cooperation, to advance AI research [72]. Therefore, strengthening the cooperation and communication between different countries/regions and institutions will contribute to the further development of this field. At present, the field is focused on the care of older adults, the surveillance and treatment of their diseases, the acceptance of AI applications, and Alzheimer disease. This study provides researchers with the information necessary to understand the current state, collaborative networks, and main research hotspots in the field. At the same time, our results suggest a series of recommendations for future research.

    Acknowledgments

    The research was supported by the Sixth “169 Project” Training Subject Scientific Research Project in Zhenjiang, Jiangsu, China (YLJ202103).

    The authors would like to thank all the participants who contributed to this study. The authors would like to also thank TopEdit for the English language editing of this manuscript.

    Data Availability

    The data sets generated and analyzed during this study are available from the corresponding author upon reasonable request.

    Conflicts of Interest

    None declared.

    1. World population prospects: the 2015 revision. Key findings and advance tables. United Nations. 2015. URL: https://www.un.org/en/development/desa/publications/world-population-prospects-2015-revision.html [accessed 2022-09-25]
    2. Skirbekk V, Staudinger U, Cohen J. How to to measure population aging? The answer is less than obvious: a review. Gerontology. Dec 13, 2019;65(2):136-144. [FREE Full text] [CrossRef] [Medline]
    3. Fong JH. Disability incidence and functional decline among older adults with major chronic diseases. BMC Geriatr. Nov 21, 2019;19(1):323. [FREE Full text] [CrossRef] [Medline]
    4. Choi N. Relationship between health service use and health information technology use among older adults: analysis of the US National Health Interview Survey. J Med Internet Res. Apr 20, 2011;13(2):e33. [FREE Full text] [CrossRef] [Medline]
    5. Boumans R, van Meulen F, van Aalst W, Albers J, Janssen M, Peters-Kop M, et al. Quality of care perceived by older patients and caregivers in integrated care pathways with interviewing assistance from a social robot: noninferiority randomized controlled trial. J Med Internet Res. Sep 09, 2020;22(9):e18787. [FREE Full text] [CrossRef] [Medline]
    6. Ho A. Are we ready for artificial intelligence health monitoring in elder care? BMC Geriatr. Sep 21, 2020;20(1):358. [FREE Full text] [CrossRef] [Medline]
    7. Dobrev D. A definition of artificial intelligence. arXiv preprint arXiv:1210.1568. 2012:1568. [CrossRef]
    8. Zhang C, Lu Y. Study on artificial intelligence: the state of the art and future prospects. J Ind Inf Integr. Sep 2021;23:100224. [CrossRef]
    9. Turner-Lee N. Can emerging technologies buffer the cost of in-home care in rural America. Generations: J Am Soc Aging. 2019;43(2):88-93.
    10. Góngora Alonso S, Hamrioui S, de la Torre Díez I, Motta Cruz E, López-Coronado M, Franco M. Social robots for people with aging and dementia: a systematic review of literature. Telemed J E Health. Jul 2019;25(7):533-540. [CrossRef] [Medline]
    11. Chen K, Lou VW, Tan KC, Wai M, Chan L. Effects of a humanoid companion robot on dementia symptoms and caregiver distress for residents in long-term care. J Am Med Dir Assoc. Nov 2020;21(11):1724-1728.e3. [CrossRef] [Medline]
    12. Lv H, Yang G, Zhou H, Huang X, Yang H, Pang Z. Teleoperation of collaborative robot for remote dementia care in home environments. IEEE J Transl Eng Health Med. 2020;8:1-10. [CrossRef] [Medline]
    13. Abdi J, Al-Hindawi A, Ng T, Vizcaychipi MP. Scoping review on the use of socially assistive robot technology in elderly care. BMJ Open. Feb 12, 2018;8(2):e018815. [FREE Full text] [CrossRef] [Medline]
    14. Chang Y, Luo D, Huang T, Goh JO, Yeh S, Fu L. Identifying mild cognitive impairment by using human-robot interactions. JAD. Feb 01, 2022;85(3):1129-1142. [CrossRef] [Medline]
    15. Ghayvat H, Gope P. Smart aging monitoring and early dementia recognition (SAMEDR): uncovering the hidden wellness parameter for preventive well-being monitoring to categorize cognitive impairment and dementia in community-dwelling elderly subjects through AI. Neural Comput Appl. Jun 06, 2021:1-13. [CrossRef]
    16. Su Z, Liang F, Do HM, Bishop A, Carlson B, Sheng W. Conversation-based medication management system for older adults using a companion robot and cloud. IEEE Robot Autom Lett. Apr 2021;6(2):2698-2705. [CrossRef]
    17. Garfield E, Paris SW, Stock WG. HistCite : a software tool for informetric analysis of citation linkage introduction. Inf-Wissenschaft Praxis. 2006;57(8):391-400. [CrossRef]
    18. Perianes-Rodriguez A, Waltman L, van Eck NJ. Constructing bibliometric networks: a comparison between full and fractional counting. J Informetr. Nov 2016;10(4):1178-1195. [CrossRef]
    19. Bibliometric Online Analysis Platform. Bibliometric. URL: https://bibliometric.com/ [accessed 2023-06-02]
    20. Chen C. CiteSpace II: detecting and visualizing emerging trends and transient patterns in scientific literature. J Am Soc Inf Sci. Feb 01, 2006;57(3):359-377. [CrossRef]
    21. Tamura T, Yonemitsu S, Itoh A, Oikawa D, Kawakami A, Higashi Y, et al. Is an entertainment robot useful in the care of elderly people with severe dementia? J Gerontol A Biol Sci Med Sci. Jan 01, 2004;59(1):83-85. [CrossRef] [Medline]
    22. Moyle W, Cooke M, Beattie E, Jones C, Klein B, Cook G, et al. Exploring the effect of companion robots on emotional expression in older adults with dementia: a pilot randomized controlled trial. J Gerontol Nurs. May 2013;39(5):46-53. [CrossRef] [Medline]
    23. Sung H, Chang S, Chin M, Lee W. Robot-assisted therapy for improving social interactions and activity participation among institutionalized older adults: a pilot study. Asia Pac Psychiatry. Mar 01, 2015;7(1):1-6. [CrossRef] [Medline]
    24. Hebesberger D, Koertner T, Gisinger C, Pripfl J. A long-term autonomous robot at a care hospital: a mixed methods study on social acceptance and experiences of staff and older adults. Int J of Soc Robotics. Jan 31, 2017;9(3):417-429. [CrossRef]
    25. Wuchty S, Jones BF, Uzzi B. The increasing dominance of teams in production of knowledge. Science. May 18, 2007;316(5827):1036-1039. [CrossRef] [Medline]
    26. Brookes BC. Bradford's law and the bibliography of science. Nature. Dec 06, 1969;224(5223):953-956. [CrossRef] [Medline]
    27. Desai N, Veras L, Gosain A. Using Bradford's law of scattering to identify the core journals of pediatric surgery. J Surg Res. Sep 2018;229:90-95. [CrossRef] [Medline]
    28. Mohd Ali J, Hussain M, Tade MO, Zhang J. Artificial Intelligence techniques applied as estimator in chemical process systems – a literature survey. Expert Syst Appl. Aug 2015;42(14):5915-5931. [CrossRef]
    29. Wang M, Yu T, Ho Y. A bibliometric analysis of the performance of water research. Scientometrics. Nov 18, 2009;84(3):813-820. [CrossRef]
    30. Petersen S, Houston S, Qin H, Tague C, Studley J. The utilization of robotic pets in dementia care. J Alzheimers Dis. 2017;55(2):569-574. [FREE Full text] [CrossRef] [Medline]
    31. Jøranson N, Pedersen I, Rokstad AMM, Ihlebæk C. Effects on symptoms of agitation and depression in persons with dementia participating in robot-assisted activity: a cluster-randomized controlled trial. J Am Med Dir Assoc. Oct 01, 2015;16(10):867-873. [CrossRef] [Medline]
    32. Broadbent E, Tamagawa R, Patience A, Knock B, Kerse N, Day K, et al. Attitudes towards health-care robots in a retirement village. Australas J Ageing. Jun 2012;31(2):115-120. [CrossRef] [Medline]
    33. Moyle W, Jones CJ, Murfield JE, Thalib L, Beattie ERA, Shum DKH, et al. Use of a robotic seal as a therapeutic tool to improve dementia symptoms: a cluster-randomized controlled trial. J Am Med Dir Assoc. Sep 01, 2017;18(9):766-773. [FREE Full text] [CrossRef] [Medline]
    34. Sparrow R, Sparrow L. In the hands of machines? The future of aged care. Minds Machines. Aug 8, 2006;16(2):141-161. [CrossRef]
    35. Moyle W, Bramble M, Jones C, Murfield J. Care staff perceptions of a social robot called Paro and a look-alike Plush Toy: a descriptive qualitative approach. Aging Ment Health. Mar 2018;22(3):330-335. [CrossRef] [Medline]
    36. Roger K, Guse L, Mordoch E, Osterreicher A. Social commitment robots and dementia. Can J Aging. Mar 2012;31(1):87-94. [CrossRef] [Medline]
    37. Hirsch JE. An index to quantify an individual's scientific research output. Proc Natl Acad Sci U S A. Nov 15, 2005;102(46):16569-16572. [FREE Full text] [CrossRef] [Medline]
    38. VOSviewer Visualizing scientific landscapes. URL: https://www.vosviewer.com/ [accessed 2023-06-21]
    39. Romero L, Portillo-Salido E. Trends in Sigma-1 receptor research: a 25-year bibliometric analysis. Front Pharmacol. May 24, 2019;10:564. [FREE Full text] [CrossRef] [Medline]
    40. World population prospects 2019 volume I: comprehensive tables. UN Department of Economic and Social Affairs. 2019. URL: https://population.un.org/wpp/publications/Files/WPP2019_Volume-I_Comprehensive-Tables.pdf [accessed 2023-04-15]
    41. Executive order No. 13859 of February 11, 2019: maintaining American leadership in artificial intelligence. Federal Register. 2019. URL: https:/​/www.​federalregister.gov/​documents/​2019/​02/​14/​2019-02544/​maintaining-american-leadership-in-arti ficial-intelligence [accessed 2023-04-15]
    42. Se Thoe E, Fauzi A, Tang YQ, Chamyuang S, Chia AYY. A review on advances of treatment modalities for Alzheimer's disease. Life Sci. Jul 01, 2021;276:119129. [CrossRef] [Medline]
    43. GBD 2019 Dementia Forecasting Collaborators. Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: an analysis for the Global Burden of Disease Study 2019. Lancet Public Health. Feb 2022;7(2):e105-e125. [FREE Full text] [CrossRef] [Medline]
    44. Povova J, Ambroz P, Bar M, Pavukova V, Sery O, Tomaskova H, et al. Epidemiological of and risk factors for Alzheimer's disease: a review. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. Jun 1, 2012;156(2):108-114. [FREE Full text] [CrossRef] [Medline]
    45. World Alzheimer report 2015 - the global impact of dementia: an analysis of prevalence, incidence, cost and trends. Alzheimer's Disease International. 2015. URL: https://www.alzint.org/resource/world-alzheimer-report-2015/ [accessed 2023-04-17]
    46. Jøranson N, Pedersen I, Rokstad AMM, Ihlebaek C. Change in quality of life in older people with dementia participating in Paro-activity: a cluster-randomized controlled trial. J Adv Nurs. Dec 16, 2016;72(12):3020-3033. [CrossRef] [Medline]
    47. Koh IS, Kang HS. Effects of intervention using PARO on the cognition, emotion, problem behavior, and social interaction of elderly people with dementia. J Korean Acad Community Health Nurs. 2018;29(3):300. [CrossRef]
    48. Hung L, Liu C, Woldum E, Au-Yeung A, Berndt A, Wallsworth C, et al. The benefits of and barriers to using a social robot PARO in care settings: a scoping review. BMC Geriatr. Aug 23, 2019;19(1):232. [FREE Full text] [CrossRef] [Medline]
    49. Shu S, Woo BK. Use of technology and social media in dementia care: current and future directions. World J Psychiatry. Apr 19, 2021;11(4):109-123. [FREE Full text] [CrossRef] [Medline]
    50. Goher KM, Mansouri N, Fadlallah SO. Assessment of personal care and medical robots from older adults' perspective. Robot Biomim. Sep 20, 2017;4(1):5. [FREE Full text] [CrossRef] [Medline]
    51. Song W, Song W, Kim Y, Kim J. Usability test of KNRC self-feeding robot. IEEE Int Conf Rehabil Robot. Jun 2013:6650501. [CrossRef] [Medline]
    52. Wang RH, Sudhama A, Begum M, Huq R, Mihailidis A. Robots to assist daily activities: views of older adults with Alzheimer's disease and their caregivers. Int Psychogeriatr. Jan 2017;29(1):67-79. [CrossRef] [Medline]
    53. Lopez-Samaniego L, Garcia-Zapirain B. A robot-based tool for physical and cognitive rehabilitation of elderly people using biofeedback. Int J Environ Res Public Health. Nov 24, 2016;13(12):1176. [FREE Full text] [CrossRef] [Medline]
    54. Lee S, Lee H, Chang WH, Choi B, Lee J, Kim J, et al. Gait performance and foot pressure distribution during wearable robot-assisted gait in elderly adults. J Neuroeng Rehabil. Nov 28, 2017;14(1):123. [FREE Full text] [CrossRef] [Medline]
    55. Chen S, Lien W, Wang W, Lee G, Hsu L, Lee K, et al. Assistive control system for upper limb rehabilitation robot. IEEE Trans Neural Syst Rehabil Eng. Nov 2016;24(11):1199-1209. [CrossRef] [Medline]
    56. Lee H, Lee S, Chang WH, Seo K, Shim Y, Choi B, et al. A Wearable Hip Assist Robot Can Improve Gait Function and Cardiopulmonary Metabolic Efficiency in Elderly Adults. IEEE Trans Neural Syst Rehabil Eng. Sep 2017;25(9):1549-1557. [CrossRef] [Medline]
    57. Van Assche M, Moreels T, Petrovic M, Cambier D, Calders P, Van de Velde D. The role of a socially assistive robot in enabling older adults with mild cognitive impairment to cope with the measures of the COVID-19 lockdown: a qualitative study. Scand J Occup Ther. Jan 06, 2023;30(1):42-52. [CrossRef] [Medline]
    58. Lin Y, Fan J, Tate JA, Sarkar N, Mion LC. Use of robots to encourage social engagement between older adults. Geriatr Nurs. Jan 2022;43:97-103. [FREE Full text] [CrossRef] [Medline]
    59. Chen S, Davis BH, Kuo C, Maclagan M, Chien C, Lin M. Can the Paro be my Buddy? Meaningful experiences from the perspectives of older adults. Geriatr Nurs. Jan 2022;43:130-137. [CrossRef] [Medline]
    60. Park Y, Chang HK, Lee MH, Lee SH. Community-dwelling older adults' needs and acceptance regarding the use of robot technology to assist with daily living performance. BMC Geriatr. Aug 05, 2019;19(1):208. [FREE Full text] [CrossRef] [Medline]
    61. Chiu C, Hsieh S, Li C. Needs and preferences of middle-aged and older adults in Taiwan for companion robots and pets: survey study. J Med Internet Res. Jun 11, 2021;23(6):e23471. [FREE Full text] [CrossRef] [Medline]
    62. Gasteiger N, Ahn HS, Fok C, Lim J, Lee C, MacDonald BA, et al. Older adults' experiences and perceptions of living with Bomy, an assistive dailycare robot: a qualitative study. Assist Technol. Jul 04, 2022;34(4):487-497. [CrossRef] [Medline]
    63. Harrington EE, Bishop AJ, Do HM, Sheng W. Perceptions of socially assistive robots: a pilot study exploring older adults’ concerns. Curr Psychol. Mar 18, 2021;42(3):2145-2156. [CrossRef]
    64. Luo J, Meng L. Research on adoption behavior and influencing factors of intelligent pension services for elderly in Shanghai. Front Genet. Jun 15, 2022;13:905887. [FREE Full text] [CrossRef] [Medline]
    65. Lianxin J, Li Z. Research on influencing factors of elderly care institutions’ choice intention based on artificial intelligence and embedded system. J Ambient Intell Human Comput. Apr 04, 2021;12(4):1-12. [CrossRef]
    66. Suwa S, Tsujimura M, Kodate N, Donnelly S, Kitinoja H, Hallila J, et al. Exploring perceptions toward home-care robots for older people in Finland, Ireland, and Japan: a comparative questionnaire study. Arch Gerontol Geriatr. Jul 15, 2020;91:104178. [CrossRef] [Medline]
    67. Adcock AK, Kosiorek H, Parikh P, Chauncey A, Wu Q, Demaerschalk BM. Reliability of robotic telemedicine for assessing critically Ill patients with the full outline of UnResponsiveness Score and Glasgow Coma Scale. Telemed E-health. Jul 2017;23(7):555-560. [FREE Full text] [CrossRef] [Medline]
    68. Veluswamy RR, Whittaker Brown S, Mhango G, Sigel K, Nicastri DG, Smith CB, et al. Comparative effectiveness of robotic-assisted surgery for resectable lung cancer in older patients. Chest. May 2020;157(5):1313-1321. [FREE Full text] [CrossRef] [Medline]
    69. Tsai C, Chen C, Wu EH, Chung C, Huang C, Tsai P, et al. A machine-learning-based assessment method for early-stage neurocognitive impairment by an immersive virtual supermarket. IEEE Trans Neural Syst Rehabil Eng. 2021;29:2124-2132. [CrossRef] [Medline]
    70. Zhao H, You J, Peng Y, Feng Y. Machine learning algorithm using electronic chart-derived data to predict delirium after elderly hip fracture surgeries: a retrospective case-control study. Front Surg. Jul 13, 2021;8:634629. [FREE Full text] [CrossRef] [Medline]
    71. Falagas ME, Pitsouni EI, Malietzis GA, Pappas G. Comparison of PubMed, Scopus, Web of Science, and Google Scholar: strengths and weaknesses. FASEB J. Feb 2008;22(2):338-342. [CrossRef] [Medline]
    72. Niu J, Tang W, Xu F, Zhou X, Song Y. Global research on artificial intelligence from 1990–2014: spatially-explicit bibliometric analysis. IJGI. May 16, 2016;5(5):66. [CrossRef]

    ‎
    AI: artificial intelligence
    TLCS: total local citation score

    Edited by T Leung; submitted 26.01.23; peer-reviewed by L Ivanitskaya, A Ghamgosar; comments to author 14.04.23; revised version received 02.05.23; accepted 25.05.23; published 23.06.23.

    Copyright

    ©Jingjing Wang, Yiqing Liang, Songmei Cao, Peixuan Cai, Yimeng Fan. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 23.06.2023.

    This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included.