Transforming Nursing

RESEARCH ARTICLE Open Access

Exploring emerging IoT technologies in smart health research: a knowledge graph analysis Xuejie Yang1†, Xiaoyu Wang2*† , Xingguo Li1†, Dongxiao Gu1,3†, Changyong Liang1, Kang Li1, Gongrang Zhang1 and Jinhong Zhong1

Abstract

Background: At present, Internet of Things technology has been widely used in various fields, and smart health is also one of its important application areas.

Methods: We use the core collection of Web of Science as a data source, using tools such as CiteSpace and bibliometric methods to visually analyze 9561 articles published in the field of smart health research based on the Internet of things (IoT) in 2003–2019, including time distribution, spatial distribution, and literature co-citation analysis and keyword analysis.

Results: The field of smart health research based on IoT has developed rapidly since 2014, but has not yet formed a stable network of authors and institutions. In addition, the knowledge base in this field has been initially formed, and most of the published literatures are multi-theme research.

Conclusions: This study discusses the research status, research hotspots and future development trends in this field, and provides important knowledge support for subsequent research.

Keywords: Internet of things, Disease, Health, Bibliometrics, Visual analysis

Background The Internet of Things (IoT) is an important representative of the new generation of information technology. It is the result of rapid development in the field of wireless commu- nications in recent years, and it is a network that extends on the Internet [1]. It can connect various information sensing devices (such as Radio Frequency Identification, in- frared sensors, laser scanners, etc.) to the Internet to realize the “Internet of Everything” [2]. At present, IoT has been widely used in various fields, such as smart city, smart home, intelligent logistics, intelligent transportation, etc.

Among them, smart health is also one of its important ap- plication areas. There are countless people who lose their lives every year due to various diseases or health problems. In terms of chronic diseases, the number of people dying from chronic diseases accounts for 60% of the total number of deaths worldwide. People are paying more and more at- tention to health issues [3]. Therefore, the use of IoT tech- nology to solve health problems has become one of the research hotspots in the field of smart health. IoT is connecting physical world with virtual world of

Internet. Physical world includes household appliances (such as air purifiers, thermostats, etc.), automobiles, in- dustrial machinery, construction, medical equipment, and human body [4]. Applying IoT technologies to healthcare will help improve the quality of people life, the level of chronic disease management, danger warning and life-

© The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

* Correspondence: mikehfut0551@163.com †Xuejie Yang, Xiaoyu Wang, Xingguo Li and Dongxiao Gu contributed equally to this work. 2The 1st Affiliated Hospital, Anhui University of Traditional Chinese Medicine, 117 Meishan Road, Hefei 230031, Anhui, China Full list of author information is available at the end of the article

Yang et al. BMC Medical Informatics and Decision Making (2020) 20:260 https://doi.org/10.1186/s12911-020-01278-9

saving interventions. There are lots potential applications of healthcare IoT: (1) Health monitoring. Today’s wearable devices can detect basic activities of human body, analyze human behavior, and measure health status. Smart wear- able devices (such as smart watch) can reduce patient anx- iety and reduce waste of resources [5]. This is very different from other sensitive devices for health monitoring in con- ventional hospitals. (2) Health information support for pa- tients. You can remind patients to take medicine on time through some IoT devices, in clinical. Networking devices such as electrocardiogram, blood oxygen, and blood pres- sure can improve the continuous measurement, monitor- ing, and support structure of patients and caregivers, thereby improving clinical outcomes [6, 7]. (3) Service im- provement. IoT can help connect cars to network systems. If a car has an accident, the system can identify the severity of the accident and help traffic administration department and healthcare emergency center via sending the accident location and direction. This will help the injured people ob- tain timely assistance [8]. (4) The collection of information resource for big data analytics. The health IoT can generate massive amounts of health big data. The analysis, mining, and use of health big data can further promote and en- hance the development of health IoT [9]. Since 2003, scholars from all over the world have grad-

ually invested in research in the field of smart health re- search based on the IoT. In response, some scholars have designed smart wearable systems to solve health problems. Li et al. [10] established a model of the acceptance of smart wearable system by the elderly, and pointed out the factors influencing their use of smart wearable technology, includ- ing self-reported health status. Akbulut et al. [11] designed a smart wearable system that monitors cardiovascular disease, which provides continuous medical monitoring. Fraise et al. [12] proposed a multi-agent system (MAS) that uses smart wearable and mobile technology to care for patients in eld- erly care facilities. In recent years, the development of tech- nology has made smart watches and smart bracelets popular. Previously, Lu et al. [13] reviewed the application of smart watches in the field of medical health. Through com- parative experiments, Hataji et al. [14] showed that com- bined treatment could improve the daily physical performance of patients with chronic obstructive pulmonary disease (COPD) under the encouragement of smart watches. Wile et al. [15] used smart watch devices to distinguish be- tween orthostatic recurrent tremor and primary tremor of Parkinson’s disease. Grym et al. [16] pointed out that smart wristbands are a viable continuous monitoring tool during pregnancy. Smart home is also an important application of IoT technology in the field of smart health. Dawadi et al. [17] demonstrated the feasibility of using smart home sensor data and learning-based data analysis to predict clinical scores. Pham et al. [18] proposed a cloud-based smart home environment (CoSHE) for home healthcare. Ghasemi et al.

[19] proposed a smart home medical system that can diag- nose environmental events and health risks quickly and in a timely manner. Alberdi et al. [20] ‘s experiments show that all mobile, cognitive, and depressive symptoms can be pre- dicted by activity-aware smart home data. In addition, re- search on disease and health issues through IoT technology is the focus of research in this field. Zhang et al. [21] pro- posed a medical data fusion algorithm based on IoT for the particularity of medical IoT data. Hossain et al. [22] pro- posed an industrial Internet of Things (IIoT) health moni- toring framework that supports cloud computing. Farahani et al. [23] introduced the overall architecture of the fog- driven IoT e-health ecosystem and discussed the applicabil- ity and challenges of the IoT in the field of healthcare. At present, there are many researches on the application

of smart wearables, smart watches, smart bracelets, smart homes and IoT technologies to the field of smart health. However, there is no research to objectively review and visualize all the literature in this field. In order to analyze the development status and future trends of the intelligent health research field based on the IoT systematically, com- prehensively, and objectively, this study uses bibliometric methods to visualize the analysis from time distribution, spatial distribution, literature co-citation and keywords based on 9561 literature data in this field from 2003 to 2019. This research provides panoramic knowledge sup- port for researchers in related fields to understand the re- search status, future trends and hotspots in the field of smart health research based on IoT.

Methods Data sources The data source for this study was Web of Science (WoS), which selected four core databases of its core collections, including Science Citation Index Expanded, Conference Proceedings Citation Index-Science, and so on. WoS is an important database for obtaining global academic information. It contains more than 13,000 au- thoritative, high-impact academic journals from around the world, covering the fields of natural sciences, engin- eering technology, biomedicine, social sciences, arts and humanities. WoS includes references cited in the paper, with a unique citation index, users can use an article, a patent number, a conference document, a journal or a book as a search term to retrieve their citations and eas- ily trace the origin and history of a research document, or track its latest progress. Although the WoS database cannot include all the literature published in this field, it has some representativeness. We invited 5 experts in the field of health IoT to finalize the database and search strategy through Delphi method. The search strategy we used is as follows: TS = (“#1” AND “#2”), Where “#1” is TS = (“internet of things” OR “smart watch*” OR “smart wristband*” OR “smart home*” OR “wearable device*”

Yang et al. BMC Medical Informatics and Decision Making (2020) 20:260 Page 2 of 12

OR “wearable technolog*” OR “wearable sensor*”), indi- cating the search term related to the IoT; “#2” indicates TS = (“diseas*” OR “health*“OR “hospital*”), which indi- cates a search term related to health. In order to ensure that the retrieved documents are related to the retrieval subject, we organized a panel of evaluation consisting of eight Ph.D. candidates in our research area. After ex- cluding irrelevant articles, we finally got 9561 records. (The search time was August 2020).

Research methods and tools This paper mainly adopts the method of bibliometrics. Bibliometrics refers to the quantitative analysis and man- agement of literature information by mathematical and statistical methods, and then discusses its structure, charac- teristics and laws [24]. This study mainly uses HistCite, CiteSpace and MS Excel to visually analyze the relevant lit- erature in the field of smart health research based on IoT. Because HistCite’s statistical function is relatively powerful, it is mainly used to collect relevant data in this paper, and then use Excel software to draw the chart [25]. CiteSpace is a visualization tool for bibliometrics that focuses on finding key points in the development of a field [26]. Therefore, this article mainly uses CiteSpace to visualize the authors, insti- tutions, literature co-citation and keywords of the IoT- based smart health research field.

Results Time distribution map In order to understand the output of research results in the field of smart health research based on IoT, HistCite was used to statistically analyze the number of scientific literatures in the years from 2003 to 2019, and the trend of annual papers was obtained, as shown in Fig. 1. As

can be seen from the figure, from 2003 to 2019, the an- nual capacity curve presents an overall growth trend, among which the curve of annual capacity from 2003 to 2014 is relatively flat, and the annual capacity basically conforms to the exponential growth trend, or even below the exponential trend line. In 2014–2019, the an- nual capacity curve has grown very rapidly, almost showing a linear upward trend, much higher than the index trend line, with a growth rate of 97.88% in 2014– 2015. This figure suggests that research in this field will continue to increase in the future and that it will remain a hotspot for future research. Then, we explored the input of researchers in the field of

smart health research based on IoT. The same use of Hist- Cite to statistically analyze the number of scientific research participants in the years from 2003 to 2019, and get the trend of the annual author input, as shown in Fig. 2. By comparing Figs. 1 and 2, we can clearly find that their vari- ation trend is roughly the same, in years with a high growth rate of annual capacity, the growth rate of author input is also high. It is also easy to understand that, generally speak- ing, the annual number of articles and the annual amount of authors input is directly proportional to the relationship. Finally, the input-output ratio of scientific researchers in

the field of smart health research based on IoT was under- stood. We calculated the number of participants in a sin- gle paper from 2003 to 2019, and obtained the change trend of the ratio of participants in a single paper, as shown in Fig. 3. The straight line parallel to the abscissa in the figure is the average number of participants in a single document over the years, which is 3.79. Overall, the an- nual input-output ratio fluctuated significantly, especially during 2003–2011. This was because the number of docu- ments in previous years was small and the researchers

Fig. 1 Annual number of published articles

Yang et al. BMC Medical Informatics and Decision Making (2020) 20:260 Page 3 of 12

were not fixed. Then the ratio is roughly stable. In the 17 years, only the authors in 2010 and 2019 had an input- output ratio of more than 4. The small number of partici- pants in a single document indicates that in this research field, the contribution rate of individual authors is high, the cooperation between authors is less, the number of au- thors is not saturated, and there is still much room for improvement.

Space distribution map Author distribution In order to analyze the author’s cooperative network, we import the preprocessed data into CiteSpace to generate the author cooperation network diagram, as shown in

Fig. 4. The figure shows the author’s name and the rela- tionship between the authors with 10 or more articles. The most published is Bonato and Rahmani, which has published 35 articles. In the figure, the size of the node is proportional to the number of articles issued by the author, the thickness of the connection between the nodes is proportional to the number of cooperation be- tween the authors, and different colors indicate the year of cooperation between different authors. As can be seen in the upper left corner of Fig. 4, the number of network nodes is 811, the number of connections between nodes is 745, and the density of the network is only 0.0023. Table 1 specifies the authors of the top 10 articles and

their related information. In the HistCite software system,

Fig. 2 Annual number of authors input

Fig. 3 Annual author input-output ratio

Yang et al. BMC Medical Informatics and Decision Making (2020) 20:260 Page 4 of 12

the citation frequency is divided into LCS and LGS, where LCS (local citation score) refers to the citation frequency of the document in the database of the field, and GCS (global citation score) refers to the frequency of citation in the Web of Science database [27]. It can be seen from Table 1 that Bonato and Rahmani have the largest number of publications, and the total quotation has reached 2024 and 908 respectively. The average single paper has been cited more than 15 times, indicating that his papers are not only issued more, but also have higher recognition. The sparseness of network density indicates that the co- operation between authors is not close enough, the field of smart health research based on IoT has not yet formed a stable core author group. And if the authors strengthen cooperation, they can collide with new sparks, promote research and innovation, and make the field flourish.

Institutional distribution Then we import the pre-processed data into CiteSpace to analyze the organization that publishes the scientific litera- ture, and generate the organization cooperation network diagram, as shown in Fig. 5. Table 2 shows the top 10 orga- nizations and related information. The organization with the largest number of publications is Chinese Academy of Sciences, which has published 134 scientific papers, 52 more than the second-ranked King Saud University. In addition, there are nearly 200 organizations that have pub- lished at least 10 articles, indicating that the research on the field of smart health research based on IoT has received ex- tensive attention from various authoritative academic insti- tutions in the world, showing a hundred schools of contention and a hundred flowers. According to Table 2, Chinese Academy of Sciences not

only has the largest number of publications, but its LCS and GCS are also much higher than the second-ranked King Saud University. Although Georgia Institute of Tech- nology is ranked third in the number of publications, its LCS is not high. The average number of citations in the field is only 1.09, indicating that the published articles are not highly recognized by peers. On the contrary, although the numbers of published literatures are not large by Mas- sachusetts Institute of Technology and Washington State University, their average total number of citations have reached 38 and 33 respectively. At the same time, Chinese Academy of Sciences’ LCS and GCS data ranks first among all institutions. These explain that the quality of the literature published by these institutions is very high and is widely supported and highly recognized by re- searchers and peers.

Fig. 4 Author collaboration network

Table 1 The top 10 authors and their number of published articles

Author Number of published articles LCS GCS

Bonato P 35 291 2024

Rahmani AM 35 260 908

Cook DJ 31 249 1284

Najafi B 30 73 442

Liljeberg P 28 192 674

Rodrigues JJPC 28 49 361

Tenhunen H 24 156 531

Kumar N 24 43 395

Ghasemzadeh H 23 98 517

Guizani M 21 31 417

Yang et al. BMC Medical Informatics and Decision Making (2020) 20:260 Page 5 of 12

Research cooperation between institutions is an import- ant way to enhance the research strength of the organization as a whole, to achieve complementary scien- tific research resources and to share knowledge, and to re- flect one of the important indicators of research status in a certain field [28]. In Fig. 5, the size of the node is propor- tional to the number of documents sent by the organization, the thickness of the connection between the nodes is proportional to the number of cooperative re- searches between the organizations, and different colors in- dicate the year of collaborative research between the institutions. The number of network nodes is 629, the number of connections between nodes is 881, and the density of the network is 0.0045. It can be seen that in the field of smart health research based on IoT, there is not

enough cooperation between institutions, and the relation- ship between them is not close enough, and a stable and mature institutional cooperation relationship has not yet been formed. Institutions should strengthen cooperation and exchanges, give full play to their respective advantages, and make full use of academic resources to promote innovation in research results and promote the vigorous de- velopment of the entire field.

Journal distribution Finally, we analyzed the journals in the field of smart health research based on IoT. We use HistCite to collect statistics on journals in the field, and Table 3 lists the top 10 journals and related information. Table 3 shows that the journal with the largest number of documents is

Fig. 5 Institution collaboration network

Table 2 The top 10 institutions in number of published articles

Institution Number of published articles LCS GCS

Chinese Academy of Sciences 134 562 6399

King Saud University 82 287 1815

Georgia Institute of Technology 68 74 1818

Tsinghua University 65 133 1848

University of California, Los Angeles 64 130 1285

Massachusetts Institute of Technology 59 224 2242

University of Bologna 55 119 860

Imperial College London 53 70 878

Washington State University 51 326 1714

University of Michigan 51 61 894

Yang et al. BMC Medical Informatics and Decision Making (2020) 20:260 Page 6 of 12

Sensors, and its LCS and GCS data are also among the best. Interestingly, four of the top 10 journals are from the IEEE Publishing Group: IEEE Access, IEEE Internet of Things Journal, IEEE Sensors Journal and IEEE Journal of Biomedical and Health Informatics. The average single article of IEEE Journal of Biomedical and Health Inform- atics has been cited as high as 31.7, which is enough to see the journal’s influence in the field of smart health re- search based on IoT. Conversely, although there are many related articles in the journal of IEEE Access, its average times cited is only 14.9, indicating that the quality and in- fluence of these articles are generally not too high. Overall, IEEE is the biggest winner in the field of smart

health research based on IoT. In addition, there are many journals that focus on the IoT, sensors and health, but there are few journals focusing on the intersection of the IoT and health. This shows that the research on the field of smart health research based on IoT has not yet had a great influence, and the major journals have not paid much attention to it.

Knowledge base analysis Co-citation Networks refers to a knowledge network formed by two scientific documents simultaneously cited by the third article [29]. Literature co-citation analysis ex- presses the relationship between documents by the fre- quency cited by other literatures. That is to say, a certain two documents are cited together by several other docu- ments. The higher the frequency of citations, the closer the relationship between the two documents is, which means that the more the subject backgrounds of the two documents are similar [30]. Fundamentally speaking, when certain documents, journals, academic groups or in- dividuals are repeatedly quoted by their peers, the know- ledge carriers that are cited are essentially recognized by the scientific community in which they are located, thus forming a scientific paradigm. This paradigm relationship can be visualized by analysis of the co-citation network of the literature [31]. Therefore, through the literature co-

citation network analysis, the knowledge base of the re- search on the field of smart health research based on IoT can be concretely demonstrated. We import the preprocessed data into CiteSpace,

analyze the co-citation relationship between scientific lit- eratures, and generate a co-citation network diagram of the literature, as shown in Fig. 6. In the figure, each node represents a document that is commonly cited. The size of the node is proportional to the number of times it is cited, the connection between the nodes indicates a co- citation relationship. The thickness of the connection in- dicates the strength of the co-cited, and the different colors indicate the year in which the document was cited. The number of network nodes is 1239, the num- ber of connections between nodes is 2331, and the dens- ity of the network is 0.003. Table 4 lists the top 10 co-citation literature and related

information. In the literature citation network, Atzori’s article published in Computer Networks in 2010 titled “The internet of things: A survey” was cited as the highest frequency, reaching 290 times [1]. So far, this article has been cited as a total of 13,691 times in Google Scholar. Gubbi [32], Islam [33], Patel [34], Al-Fuqaha [35], etc., which are ranked behind, are connected to Atzori [1]. It shows that the correlation between the top citations in the field is very strong, and the topics of the scientific litera- ture are similar. These documents are all about IoT tech- nology and applications. It can be seen that the current research on IoT technology has been initially matured and standardized. Interestingly, five of the top 10 cited articles were published by the IEEE Publishing Group. Combined with the above-mentioned journal analysis, it

can be said that the IEEE Publishing Group has made tre- mendous contributions to the development of this field. In addition, in terms of centrality, Atzori’s performance is also very good, the greater the centrality of a node in the network, indicating that it is more important in the net- work [36]. Therefore, the comprehensiveness of all aspects can reflect the importance of the document Atzori [1]. It

Table 3 The top 10 journals with literature quantity

Name of Journal Literature quantity GCS Average times cited

1 Sensors 424 8391 19.8

2 IEEE Access 237 3531 14.9

3 IEEE Internet of Things Journal 112 2792 24.9

4 IEEE Sensors Journal 110 2596 23.6

5 Future Generation Computer Systems-the International Journal of Escience 85 1812 21.3

6 JMIR Mhealth and Uhealth 77 679 8.8

7 IEEE Journal of Biomedical and Health Informatics 66 2089 31.7

8 Journal of Medical Systems 65 885 13.6

9 Plos One 52 1504 28.9

10 Acs Applied Materials & Interfaces 48 1237 25.8

Yang et al. BMC Medical Informatics and Decision Making (2020) 20:260 Page 7 of 12

can be said that this article lays the foundation for the re- search of smart health research based on IoT. In general, the literature of Fig. 6 has a relatively tightly distributed network, indicating that the knowledge base in this field has been initially formed, which will provide important knowledge support for subsequent research.

Research focus analysis Research hotspots refer to the focus of research in a cer- tain discipline in a certain period of time. Generally speak- ing, there is a large number of scientific literatures, academic thoughts, and research groups emerging on a subject [37]. Kuhn [38] emphasized that the development of science is an alternating appearance of the conventional science and the scientific revolution, which indicates that

the scientific revolution is changing, and that the old and new paradigms are incommensurable. It is precisely be- cause of the existence of incommensurability that the vo- cabulary system between the old and new paradigms will change accordingly. So, whether the scientific revolution occurs can be judged from whether the vocabulary of the period has changed. The number of co-occurrences of dif- ferent keywords in the scientific literature can be counted. The level of the co-occurrence frequency can reflect the correlation between the keywords and the hot issues in the specific field during this period [39]. Therefore, the co-occurrence analysis of keywords can reveal the re- search structure and research hotspots in specific fields. The research results of Callon et al. [40] are the earliest applications of co-word analysis. Subsequently, the co-

Fig. 6 Articles in the co-citation network

Table 4 List of the top 10 co-citation articles with the corresponding frequencies

Author Year Name of Journal Frequency

Atzori L 2010 Computer Networks 290

Gubbi J 2013 Future Generation Computer Systems-The International Journal of eScience 269

Islam SMR 2015 IEEE Access 262

Patel S 2012 Journal of NeuroEngineering and Rehabilitation 177

Al-Fuqaha A 2015 IEEE Communications Surveys and Tutorials 171

Pantelopoulos A 2010 IEEE Transactions on Systems Man and Cybernetics Part C-applications and Reviews 159

Gao W 2016 Nature 150

Zanella A 2014 IEEE Internet of Things Journal 138

Lara OD 2013 IEEE Communications Surveys and Tutorials 126

Miorandi D 2012 Ad Hoc Networks 115

Yang et al. BMC Medical Informatics and Decision Making (2020) 20:260 Page 8 of 12

word analysis method has been widely used in the field of information science. Of course, keyword analysis is also based on certain assumptions [41], including: (1) The se- lection of keywords is cautious; (2) Multiple keywords in the same document are related to each other and are rec- ognized by the author; (3) If enough authors recognize the relationship between the same keyword, then this relation- ship can be considered to have a certain meaning in the field; (4) The keyword can reflect the content of the docu- ment to a certain extent. When the author chooses the keyword, it is usually affected by other research results. The basic principle of co-word analysis is to count the number of times a group of keywords appears in the same group of documents, then the degree of co-occurrence is measured by the number of co-occurrences, the more co- occurrences, the more closely they are related [42]. The key word is a high degree of conciseness and

generalization of an article, which is the core and es- sence of the article, frequently high keywords are often used to identify hot topics in a research field. By analyz- ing keywords, we can intuitively grasp the main research content of a paper, and even the overall research situ- ation in a field [43]. This study extracts the key words from 9561 documents and conducts frequency statistics and frequency co-occurrence analysis to understand the current structural foundations and research hotspots in the field of smart health research based on IoT, and pre- dict the future development direction of the field. Table 5 lists the top 20 keywords of the co-occurrence fre- quency. It can be seen that the keyword with the highest frequency is the internet of things, this is very consistent with the topic of this article. The subject of this paper is the IoT and health. The IoT is also the keyword with the highest centrality, indicating that research in this field is basically carried out around the IoT. These key- words with high frequency of occurrence can be divided into four main categories: (1) Keywords related to the IoT technology, such as internet of thing, sensor, wire- less sensor network, big data, cloud computing, etc.; (2)

Keywords related to health, such as healthcare, health, parkinsons disease, etc.; (3) Keywords related to smart health, such as smart home; (4) Problems arising from research in the field of smart health research based on IoT, such as security. The co-word network refers to an objective knowledge

network that expresses the structure of the scientific knowledge domain, which is composed of co-occurrence between keywords. It can be used to describe the know- ledge structure of a subject domain and can reveal the evolution of a disciplinary structure in combination with time series [44]. We import the pre-processed data into CiteSpace to analyze the keywords of the scientific litera- ture and generate a keyword co-occurrence network dia- gram, as shown in Fig. 7. Each node represents a different keyword, and the size of the node is propor- tional to the frequency of its co-occurrence. The con- nection between nodes indicates the co-occurrence relationship between two keywords in the same docu- ment, and the different colors indicate the years in which different keywords co-occur. The number of net- work nodes is 889, the number of connections between nodes is 3090, and the density of the network is 0.0078. It can be seen from the figure that the network as a whole is relatively dense, and the co-occurrence relation- ship between multiple groups of keywords is relatively close, indicating that the research results in the field of intelligent health research based on the Internet of Things are mostly multi-theme research. In the future research, the IoT is still the center of re-

search in this field. Focusing on the integration of the IoT and the integration of smart health, it is necessary to solve the security and privacy issues in this field and eliminate concerns for users. IoT technologies often appear at the same time as cloud computing, big data, etc. These emer- ging technologies are closely related, and artificial intelligence technology cannot lack them. Therefore, in future research, the IoT, big data and cloud computing will frequently appear in the research in this field. Smart

Table 5 List of the top 20 keywords with the corresponding frequency

Keyword Frequency Keyword Frequency

1 internet of thing 2329 11 technology 351

2 wearable sensor 812 12 security 345

3 system 715 13 accelerometer 313

4 healthcare 670 14 wearable 290

5 sensor 644 15 machine learning 272

6 internet 580 16 wireless sensor network 271

7 wearable device 563 17 cloud computing 268

8 health 430 18 big data 264

9 physical activity 366 19 smart home 263

10 thing 352 20 parkinsons disease 256

Yang et al. BMC Medical Informatics and Decision Making (2020) 20:260 Page 9 of 12

homes and smart cities have received a lot of attention in recent years. Smart home is a platform for housing, using IoT technology to connect various devices in the home to achieve convenience, comfort and intelligence. At present, there are few studies on the use of IoT technology to treat specific diseases (such as parkinsons diseases appearing in the top 20 keywords, etc.), so this may be one of the future research trends.

Discussions Summary of findings Using HistCite, CiteSpace, Excel and other analysis tools, the time distribution, spatial distribution, literature cit- ation and research hotspots of knowledge in this field are deeply analyzed and visualized. (1) In terms of time distribution: the annual load cap-

acity curve and the annual author input curve change trend are roughly the same, and the overall growth trend. In particular, the growth rate since 2014 has been very rapid, almost showing a linear upward trend, much higher than the index trend line. Explain that research in this field will continue to increase in the future and that it will remain a hotspot for future research. (2) In terms of spatial distribution: a) Author distribu-

tion: The author’s cooperation network is sparse, the co- operation between the authors is not close enough, and the stable core author group has not yet formed in this field; b) Institutional distribution: Similarly, the cooper- ation and cooperation between institutions is not close enough, and a stable and mature institutional cooperation

relationship has not yet been formed; c) Journal distribu- tion: There are few journals focusing on the intersection of IoT technology and health. This shows that the re- search in this field has not yet had a great influence, and the major journals have not paid much attention to it. (3) In terms of knowledge base analysis: the literature

has a relatively tightly indexed network, and the know- ledge base in this field has been initially formed, which can provide important knowledge support for subse- quent research. (4) In terms of research hotspot analysis: high-

frequency keywords can be divided into four parts. The Internet of Things is a keyword with the highest co- occurrence frequency and the highest centrality. In addition, most of the research results in this field are multi-theme research. Among them, smart home and the use of IoT technology to assist in the treatment of specific diseases are the future research trends.

Future trends Through the research in this paper, we can find that smart home and smart city are the research hotspots in recent years, and will likely also be the focus of future research. Smart homes can be a “family health consultant”. For

example, the smart home system can realize the “alarm” of the elderly and children, notify the family and locate; The system will automatically start and shut down the air purifier according to the real-time air condition, without manual operation; In addition, the air purifier can be con- trolled based on the humidity level in the house and the

Fig. 7 Keyword co-occurrence network

Yang et al. BMC Medical Informatics and Decision Making (2020) 20:260 Page 10 of 12

PSI level of asthma and allergic rhinitis patients [45]. The smart screen in the kitchen can see the children in the liv- ing room through the security system, and is equipped with detection equipment for harmful gases such as gas to achieve the function of safety protection; The smart cloth- ing care machine in the cloakroom has the functions of steam sterilization, shaping, drying and other functions to protect the health of users. The construction of smart health protection system is

an important part of the construction of smart cities. The construction of “digital health” system is the focus of promotion. The smart city will build a medical health big data platform, realize the data sharing and sharing of medical and health service organizations, and rely on the smart city cloud platform to form a citizen medical health information big data center, and provide support for promoting three-medicine linkage and achieving graded diagnosis and treatment. In addition, electronic health records of residents in the city will be established to realize the networking of health services in hospitals and clinics throughout the city. And promote remote registration, electronic toll collection, online telemedi- cine services, graphic and physical examination diagnos- tic systems, etc., to comprehensively improve the city’s medical and health services.

Future works The following work will be done in the future studies: (1) The first one it to examine how the Internet of Things ac- tually affects medical accessibility; (2) The second one is to evaluate the objectiveness or reproducibility of reported results as well as relationship between prominent authors and industry; (3) The third one is to use more data re- sources rather than only WoS data and further validate our search result; (4) The fourth but not the last one is to eliminate irrelevant search results with technical tools ra- ther than the manual methods used on this study.

Conclusions In order to explore the knowledge base, research hot- spots, development status and future research directions of the research on the field of smart health research based on IoT. We conducted a bibliometric analysis of 9561 articles in the Web of Science core database for the 17 years from 2003 to 2019. The results show that the field of smart health research based on IoT has devel- oped rapidly since 2014, but has not yet formed a stable network of authors and institutions. Our research pro- vides panoramic knowledge support for researchers in related fields to understand the research status, future trends and hotspots in the field of smart health re- search based on IoT.

Abbreviations IoT: Internet of things; MAS: Multi-agent system; COPD: Chronic obstructive pulmonary disease; CoSHE: Cloud-based smart home environment; IIoT: Industrial Internet of Things; WoS: Web of Science; LCS: Local citation score; GCS: Global citation score

Acknowledgements We would like to thank Fenghong Liu and Sofiya Shaanova for their insightful comments on an earlier version of this manuscript.

Authors’ contributions XJY, XYW and DXG contributed to writing the manuscript. KL and XJY collected, analyzed and interpreted the data. CYL, XGL, GRZ and JHZ revised and improved the manuscript. All authors read and approved the final manuscript.

Funding The dataset collection and analysis of this research were partially supported by the National Natural Science Foundation of China (NSFC) under grant Nos. 71771077, 71771075 and 72071063, and Fundamental Research Funds for the central universities under grant No. PA2020GDKC0020.

Availability of data and materials Data materials are available from the lead author upon request.

Ethics approval and consent to participate Not applicable.

Consent for publication Not applicable.

Competing interests The authors declare that they have no competing interests.

Author details 1The School of Management, Hefei University of Technology, 193 Tunxi Road, Hefei 230009, Anhui, China. 2The 1st Affiliated Hospital, Anhui University of Traditional Chinese Medicine, 117 Meishan Road, Hefei 230031, Anhui, China. 3Key Laboratory of Process Optimization and Intelligent Decision-making of Ministry of Education, Hefei 230009, Anhui, China.

Received: 24 September 2018 Accepted: 30 September 2020

References 1. Atzori L, Iera A, Morabito G. The internet of things: a survey. Comput Netw.

2010;54(15):2787–805. 2. Stankovic JA. Research directions for the internet of things. IEEE Internet

Things J. 2014;1(1):3–9. 3. Zhang G, Li K, Gu D, Wang X, Yang X, Zhu K, Liang G. Visualizing knowledge

evolution and hotspots of rural environment and health: a systematic review and research direction. IEEE Access. 2019;7:72538–50.

4. Greengard S. The internet of things: MIT press; 2015. 5. Abawajy JH, Hassan MM. Federated internet of things and cloud computing

pervasive patient health monitoring system. IEEE Commun Mag. 2017;55(1): 48–53.

6. Fernandes CO, De Lucena CJP. A software framework for remote patient monitoring by using multi-agent systems support. JMIR Med Inform. 2017; 5(1):e9.

7. Liang C, Gu D, Tao F, et al. Influence of mechanism of patient-accessible hospital information system implementation on doctor–patient relationships: a service fairness perspective. Inf Manag. 2017;54(1):57–72.

8. Thomas MO, Rad BB. Reliability evaluation metrics for internet of things, car tracking system: a review. Int J Inf Technol Comput Sci (IJITCS). 2017;9(2):1– 10.

9. Mahdavinejad MS, Rezvan M, Barekatain M, Adibi P, Barnaghi P, Sheth AP. Machine learning for internet of things data analysis: a survey. Digit Commun Netw. 2018;4(3):161–75.

10. Li J, Ma Q, Chan AH, Man S. Health monitoring through wearable technologies for older adults: smart wearables acceptance model. Appl Ergon. 2019;75:162–9.

Yang et al. BMC Medical Informatics and Decision Making (2020) 20:260 Page 11 of 12

11. Akbulut FP, Akan A. A smart wearable system for short-term cardiovascular risk assessment with emotional dynamics. Measurement. 2018;128:237–46.

12. Fraile JA, Zato C, Gil O, de la Prieta F. Multiagent Systems and Wearable Devices: Helping People Live Healthier. In: Augusto JC, Corchado JM, Novais P, Analide C, editors. Ambient Intelligence and Future Trends-International Symposium on Ambient Intelligence (ISAmI 2010). Advances in Soft Computing, vol 72. Berlin, Heidelberg: Springer; 2010.. https://doi.org/10. 1007/978-3-642-13268-1_2.

13. Lu T-C, Fu C-M, Ma MH-M, Fang C-C, Turner AM. Healthcare applications of smart watches. Appl Clin Inform. 2016;7(03):850–69.

14. Hataji O, Nishii Y, Ito K, Sakaguchi T, Saiki H, Suzuki Y, D'Alessandro-Gabazza C, Fujimoto H, Kobayashi T, Gabazza EC. Smart watch-based coaching with tiotropium and olodaterol ameliorates physical activity in patients with chronic obstructive pulmonary disease. Exp Therapeutic Med. 2017;14(5): 4061–4.

15. Wile DJ, Ranawaya R, Kiss ZH. Smart watch accelerometry for analysis and diagnosis of tremor. J Neurosci Methods. 2014;230:1–4.

16. Grym K, Niela-Vilén H, Ekholm E, Hamari L, Azimi I, Rahmani A, Liljeberg P, Löyttyniemi E, Axelin A. Feasibility of smart wristbands for continuous monitoring during pregnancy and one month after birth. BMC Pregnancy Childbirth. 2019;19(1):34.

17. Dawadi PN, Cook DJ, Schmitter-Edgecombe M. Automated cognitive health assessment from smart home-based behavior data. IEEE J Biomed Health Inform. 2015;20(4):1188–94.

18. Pham M, Mengistu Y, Do H, Sheng W. Delivering home healthcare through a cloud-based smart home environment (CoSHE). Futur Gener Comput Syst. 2018;81:129–40.

19. Ghasemi F, Rezaee A, Rahmani AM. Structural and behavioral reference model for IoT-based elderly health-care systems in smart home. Int J Commun Syst. 2019;32(12):e4002.

20. Alberdi A, Weakley A, Schmitter-Edgecombe M, Cook DJ, Aztiria A, Basarab A, Barrenechea M. Smart home-based prediction of multidomain symptoms related to Alzheimer's disease. IEEE J Biomed Health Inform. 2018;22(6): 1720–31.

21. Zhang W, Yang J, Su H, Kumar M, Mao Y. Medical data fusion algorithm based on internet of things. Pers Ubiquit Comput. 2018;22(5–6):895–902.

22. Hossain MS, Muhammad G. Cloud-assisted industrial internet of things (iiot)–enabled framework for health monitoring. Comput Netw. 2016;101: 192–202.

23. Farahani B, Firouzi F, Chang V, Badaroglu M, Constant N, Mankodiya K. Towards fog-driven IoT eHealth: promises and challenges of IoT in medicine and healthcare. Futur Gener Comput Syst. 2018;78:659–76.

24. Qiu J-P, Dong K, Yu H-Q. Comparative study on structure and correlation among author co-occurrence networks in bibliometrics. Scientometrics. 2014;101(2):1345–60.

25. Garfield E. From the science of science to Scientometrics visualizing the history of science with HistCite software. J Inf Secur. 2009;3(3):173–9.

26. Chen C. Searching for intellectual turning points: progressive knowledge domain visualization. Proc Natl Acad Sci. 2004;101(suppl 1):5303–10.

27. Emami Z, Hariri N, Khamseh ME, Nooshinfard F. Mapping diabetes research in middle eastern countries during 2007-2013: a scientometric analysis. Med J Islam Repub Iran. 2018;32:84.

28. Gu D, Li J, Li X, Liang C. Visualizing the knowledge structure and evolution of big data research in healthcare informatics. Int J Med Inform. 2017;98:22– 32.

29. Leung XY, Sun J, Bai B. Bibliometrics of social media research: a co-citation and co-word analysis. Int J Hosp Manag. 2017;66:35–45.

30. Holman D, Lynch R, Reeves A. How do health behaviour interventions take account of social context? A literature trend and co-citation analysis. Health. 2018;22(4):389–410.

31. Chen C. Science mapping: a systematic review of the literature. J Data Inf Sci. 2017;2(2):1–40.

32. Gubbi J, Buyya R, Marusic S, Palaniswami M. Internet of things (IoT): a vision, architectural elements, and future directions. Futur Gener Comput Syst. 2013;29(7):1645–60.

33. Islam SMR, Kwak D, Kabir MH, Hossain M, Kwak K. The internet of things for health care: a comprehensive survey. IEEE Access. 2015;3:678–708.

34. Patel S, Park H, Bonato P, Chan L, Rodgers M. A review of wearable sensors and systems with application in rehabilitation. J NeuroEng Rehabil. 2012; 9(1):21.

35. Al-Fuqaha A, Guizani M, Mohammadi M, Aledhari M, Ayyash M. Internet of things: a survey on enabling technologies, protocols, and applications. IEEE Commun Surv Tutorials. 2015;17(4):2347–76.

36. Gu D, Li K, Wang X, Liang C. Visualizing Knowledge Evolution of Emerging Information Technologies in Chronic Diseases Research. In: Chen H, Fang Q, Zeng D, Wu J, editors. Smart Health. ICSH 2018, Wuhan, China. Lecture Notes in Computer Science, vol 10983. Cham: Springer; 2018. https://doi. org/10.1007/978-3-030-03649-2_26.

37. Gu D, Li T, Wang X, Yang X, Yu Z. Visualizing the intellectual structure and evolution of electronic health and telemedicine research. Int J Med Inform. 2019;130:103947.

38. Kuhn Thomas. The Structure of Scientific Revolutions[J]. Philos Books. 1962; 4(3):14–16.

39. Gu D, Li K, Wang X, Li X, Liu F, Jiang L, Zhang F. Discovering and visualizing knowledge evolution of chronic disease research driven by emerging technologies. IEEE Access. 2019;7:72994–3003.

40. Callon M, Courtial J-P, Turner WA, Bauin S. From translations to problematic networks: an introduction to co-word analysis. Information (International Social Science Council). 1983;22(2):191–235.

41. Whittaker J. Creativity and conformity in science: titles, keywords and co- word analysis. Soc Stud Sci. 1989;19(3):473–96.

42. Yan B-N, Lee T-S, Lee T-P. Mapping the intellectual structure of the internet of things (IoT) field (2000–2014): a co-word analysis. Scientometrics. 2015; 105(2):1285–300.

43. Apriliyanti ID, Alon I. Bibliometric analysis of absorptive capacity. Int Bus Rev. 2017;26(5):896–907.

44. Zhao W, Mao J, Lu K. Ranking themes on co-word networks: exploring the relationships among different metrics. Inf Process Manag. 2018;54(2):203–18.

45. Gu D, Deng S, Zheng Q, et al. Impacts of case-based health knowledge system in hospital management: the mediating role of group effectiveness. Inf Manag. 2019;56(8):103162.

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Yang et al. BMC Medical Informatics and Decision Making (2020) 20:260 Page 12 of 12

BioMed Central publishes under the Creative Commons Attribution License (CCAL). Under the CCAL, authors retain copyright to the article but users are allowed to download, reprint, distribute and /or copy articles in BioMed Central journals, as long as the original work is properly cited.