Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A Bibliometric Analysis of the Development of ICD-11 in Medical Informatics

A Bibliometric Analysis of the Development of ICD-11 in Medical Informatics Hindawi Journal of Healthcare Engineering Volume 2019, Article ID 1649363, 12 pages https://doi.org/10.1155/2019/1649363 Research Article A Bibliometric Analysis of the Development of ICD-11 in Medical Informatics 1 1 2 1 Donghua Chen , Runtong Zhang , Hongmei Zhao, and Jiayi Feng Department of Information Management, School of Economics and Management, Beijing Jiaotong University, Beijing 100044, China Peking University People’s Hospital, Beijing 100044, China Correspondence should be addressed to Runtong Zhang; rtzhang@bjtu.edu.cn Received 3 September 2019; Revised 7 November 2019; Accepted 22 November 2019; Published 26 December 2019 Academic Editor: Cesare F. Valenti Copyright © 2019 Donghua Chen et al. +is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +e International Classification of Diseases (ICD), which is used to group and report health conditions and factors, provides a basis for healthcare statistics. +e 11th revision of the ICD (ICD-11) released by the World Health Organization provides stakeholders with novel perspectives on solving the complexity of critical problems in medical informatics. +is study conducts a bibliometric analysis of research published over the period of 1989–2018 to examine the development of ICD-related research and its trends. First, over 4000 ICD-related papers spanning the 30-year period are retrieved from the Web of Science database. +en, based on the meta data of the selected papers, time trend analysis is performed to examine the development of different ICD revisions. Finally, the keywords and topics of these papers are analyzed and visualized using VOSViewer and CiteSpace. Our findings indicate that ICD-11-related research has grown rapidly in recent years compared with studies on ICD-9 and ICD-10. Moreover, the most popular research directions of ICD-11 include the topics psychiatry, psychology, information science, library science, and behavioral science. In terms of perspectives, information system-related research is more common than big data- and knowledge discovery-related work. However, the popularity of big data- and knowledge discovery-related developments has grown in recent years. +e use of ICD-11 facilitates the development of medical informatics from the perspectives of information systems, big data, and knowledge discovery. to monitor epidemics and threats toward public health, 1. Introduction monitors the expenditure burden shouldered by patients, +e International Classification of Diseases (ICD) [1], which evaluates the progress in achieving public health objectives, was developed by the World Health Organization (WHO), determines the obligation of member states of the WHO to plays a crucial role in governments’ reporting, grouping, and provide free or subsidized medical services to their pop- statistical analyses of diseases and other health-related ulations, and develops appropriate healthcare services [3]. factors. +e wide use of ICD makes it a global standard for +erefore, the ICD is key to the sustainable development of medical big data research [4]. diagnostic health information and enables sustainable and systematic recording, analysis, interpretation, and com- ICD standards have been used in medicine and parison of mortality and morbidity rates of different healthcare for over 100 years. +e first ICD standards ini- countries at different time points. +e ICD also covers tially focused on the statistics of the causes of death. In 1946, various signs, symptoms, abnormal findings, complaints, the Interim Commission of the WHO was entrusted to take and social factors suitable for studies on financial aspects, over the revision of the ICD and introduced a method for such as billing or resource allocation, and provides a basis for disease classification [5]. At present, the most widely used big data in personalized healthcare [2]. Moreover, the ICD version of the ICD is its 10th revision (ICD-10), which was provides an information framework that allows stakeholders endorsed by the Forty-third World Health Assembly in 2 Journal of Healthcare Engineering Novel concepts in ICD-11, such as stem codes and post- 1990. After over a decade of revisions by numerous countries based on Internet-based maintenance platforms, the WHO coordination, are proposed to overcome the challenges encountered by ICD-10 in recent years because the latter is released the final version of the 11th revision of ICD (ICD- 11) in June 18, 2018, to provide a new de facto standard of now outdated in the clinical and classification points of view. disease coding for the twenty-first century [6]. +e ICD-11 For example, stem codes containing all pertinent in- was submitted to the 144th Executive Board Meeting in formation in a precombined fashion in ICD-11 are referred January 2019 and then to the 72nd World Health Assembly to as “pre-coordination;” when additional detail pertaining in May 2019 [7]. Following endorsement, the member states to a single condition is described by combining multiple of the WHO are expected to begin reporting on the basis of codes, the code combination is referred to as “post-co- ordination.” ICD-11 also allows stakeholders to operate in ICD-11 on January 1, 2022. +e development of new ICD standards is expected to an electronic environment and capture more information, especially for morbidity use cases. In summary, the newly revolutionize global medical informatics within the next few decades. Over the past 20 years, the ICD-10 has been widely proposed ICD-11 is more suitable for disease coding than past revisions of ICD in the new era of medical informatics. utilized to classify healthcare information. For example, ICD-10-coded hospital big data offer new opportunities for +is study aims to conduct a bibliometric analysis to monitoring flu epidemics [8]. Numerous ICD-10 national examine the development of ICD-11-based studies in modifications have been developed to adapt actual use in healthcare. +e data sources and search strings are first different countries. +e ICD-11 at present is ready for testing determined. Time trend analysis is then performed based on and implementation in accordance with the specific time- the selected papers. Keywords and topics are analyzed and lines and requirements of different countries [9]. +e visualized to summarize the main findings of our study from the perspectives of information systems, big data, and structure and design of the newly proposed ICD-11 are based on clinical practices over the past few decades and knowledge discovery. Finally, we discuss and conclude our work. differ considerably from those of its previous revisions [10]. ICD-11-coded medical records provide the basis of massive health statistics with the latest development of big data- 2. Materials and Methods driven intelligent healthcare using big data analytical plat- forms such as Apache Hadoop and Spark [11, 12]. However, Figure 1 illustrates the flowchart of this bibliometric analysis the increasing use of ICD-11 in medical and health big data of ICD-related research. First, we determine relevant key- reduces the applicability and relevance of past analytical words and conduct search strategies to retrieve ICD-related methods because the ICD-11 features new code schemes and research. Second, the Web of Science database is used to concepts that differ from previous ICDs, such as stem codes retrieve relevant publications. +ird, time trend analysis of representing entities or groupings of high relevance or ICD-related papers is performed. Finally, the analysis results clinical conditions that should always be described as a single are visualized from three perspectives. category. Appropriate utilization of ICD-11 for the analysis of mortality, morbidity, epidemiology, case mixing, quality and safety, primary care, and detailed information from 2.1. Data Sources. +e data sources of ICD-related work medical and health big data are essential to provide the basis published over a certain time period are selected to facilitate for big data research in health informatics [13]. this bibliometric analysis. Many researchers have applied ICD Introduction of the novel concepts of ICD-11 can to their research since 1990, when the ICD-10 was first en- overcome the problems of previous ICD revisions. +e dorsed. Given that the ICD-9 and ICD-10 have played a foundation component and content model are key concepts crucial role in promoting the development of medical in- in ICD-11. +e foundation component is a multidimen- formatics in the last 30 years [17], the trends of relevant sional collection of all ICD entities. +e content model studies in this period should be examined. +e Web of Science describes several specific diseases or disorders and is defined database is used to obtain high-quality papers. However, we by 13 attributes, namely, ICD entity title, classification acknowledge that the database may not contain several properties, textual definitions, terms, body system/body valuable papers in this field. We will synthesize and discuss part, temporal properties, subtype property severity, man- relevant literature. Relevant articles published over the period ifestation properties, causal properties, functioning prop- of 1989–2018 are retrieved from the Web of Science database erties, specific condition properties, treatment properties, by searching the keywords “ICD-9,” “ICD-10,” and “ICD-11” and diagnostic criteria. +e content model also illustrates in the article title field in the core set of the Web of Science background knowledge that provides the basis for the sys- database. +is process can search studies relevant to the ICD tematic definition of each ICD entity to enable computer- national modification because the name of ICD national ization. New disorders, such as gaming disorder, which modifications includes the keyword “ICD.” Table 1 shows the remains controversial, are introduced in ICD-11 [14]. In statistics of the publications selected from the database. contrast to ICD-10, ICD-11 is established on the basis of +e distribution of ICD-related subjects in the selected ontology models [15]. Several value sets in ICD-11 are publications over 30 years is illustrated in Figure 2. Psychiatry derived from external ontologies, such as the Systematized and psychology are the most popular subjects in ICD-related Nomenclature of Medicine–Clinical Terms (SNOMED CT) clinical research. Information science and library science are [16], which has played an important role in healthcare. other popular research fields that may focus on the Journal of Healthcare Engineering 3 big data. Finally, an overlay visualization of the existing 54 Begin publications related to ICD and knowledge discovery is presented to investigate the state of knowledge discovery Determine searching using ICD codes. +e distribution of keywords from these keywords and three perspectives is examined and discussed. strategies Existing literature 2.4. Topic Analysis. +e topics of ICD-related research to Determine data promote healthcare are as follows. sources of publications Web of science database 2.4.1. Information System Perspective. ICD codes provide Examine the trends of the basis of structured medical big data in healthcare. Most ICD-related research From three work used natural language processing and machine- perspectives learning techniques for textual analysis. Without pro- fessional clinical inspection, such as evaluation of the proper coding of the clinical statuses of patients, the collected data Information Big data-driven Knowledge may be imprecise. ICD codes also enable automated clas- systems analysis discovery sification of diagnostic terms, such as application of com- puter-assisted coding in Spanish [18]. +e ICD is useful for solving such problems and produces structured data that Overlay visualization of analyzing improve the reliability of results from big data analysis. results of the selected publications 2.4.2. Big Data Perspective. ICD codes can be related to Discussion different perspectives of big data in healthcare. Analysis of massive individual data from the perspectives of different End sources, dimensions, and time points often reveals trends that traditional medical research approaches cannot Figure 1: Flowchart of the bibliometric analysis in this study. show [19]. However, the contents of existing medical big data are occasionally incorrect, incomplete, and even un- available; few datasets are complete and valuable for research development of ICD standards. Behavioral science and purposes. +e precision and reliability of analyzing ICD- neuroscience neurology are related to clinical research. Health coded results in big data-driven algorithms rely on the science services and science technology are key fields focusing coding quality of ICD when ICD coders encode their on improvements in the practical use of ICD in medical and medical records [20]. health informatics. Research on health science services and science technology is key to intelligent healthcare. 2.4.3. Knowledge Discovery Perspective. +e performance of ICD-related analysis generally relies on changes in the main 2.2. Time Trend Analysis. Time trend analysis aims to ex- diagnosis in the discharge summary of patients or the ac- amine the development of different revisions of ICD over the curacy of techniques for extracting information from patient past 30 years. First, the trends of publications related to ICD- records by medical institutions [21]. Professional and 9, ICD-10, and ICD-11 over the selected period are ex- technical requirements for practitioners, especially for fresh amined. A timeline view of ICD-related research is then used coders, are stringent because ICD coders must establish a to analyze ICD-related topics extracted from the keywords clear disease classification framework in their mind. Disease- of publications retrieved for the period 2009–2018. Finally, related concepts and relations could be retrieved and as- three perspectives, namely, medical information systems, big sociated with other knowledge sources in medical domains data, and knowledge discovery, illustrate the trends of the with the use of ICD to facilitate clinical knowledge discovery number of publications for the past 30 years. from ICD-coded data [22]. 2.3. Keyword Analysis. +ree perspectives of ICD-related research are examined through overlay visualization of the 2.5. Tools for Visualization. VOSViewer [23] and CiteSpace publications. First, an overlay visualization of 234 publi- [24] are used to visualize the search results and examine the cations collected by using the keywords “ICD” and “in- key information and trends of publications on the use of formation systems” from the core set of the Web of Science ICD. database is presented to examine studies related to the During network visualization using VOSViewer, items are represented by their label by default by a circle. +e size implementation of ICD in medical information systems. Second, an overlay visualization of 51 publications related to of the label and circle of an item are determined by the weight of the item. +e higher the weight of an item, the the use of ICD in big data analytics is presented to examine the trends of ICD-related research from the perspective of larger the label and circle of this item. +e color of an item is 4 Journal of Healthcare Engineering Table 1: Main types of publications retrieved for bibliometric analysis. Number of ICD-9-related Publication type Number of ICD-10-related publications Number of ICD-11-related publications publications Article 390 1359 350 Other 178 367 101 Abstract 143 249 112 Meeting 67 89 21 Letter 31 40 31 Editorial 16 83 93 Review 8 116 66 Case report 3 8 1 Clinical trial 3 26 4 News 3 30 9 Reference material 3 8 1 Total 845 2375 789 Number of publications 0 100 200 300 400 500 600 700 Psychiatry Psychology Information Science Library Science Behavioral Sciences Neurosciences Neurology Health Care Sciences Services Science Technology Other Topics Pediatrics Public Environmental Occupational Health Mathematics Geriatrics Gerontology Pathology Demography General Internal Medicine Pharmacology Pharmacy Substance Abuse Toxicology Genetics Heredity Allergy Computer Science Sociology Immunology Mathematical Computational Biology Infectious Diseases Social Sciences Other Topics Figure 2: Research directions related to ICD over the period 1989–2018. determined by the cluster to which the item belongs. Lines titles of publications containing the keywords “ICD-9,” between items represent links. In the overlay visualization “ICD-10,” and “ICD-11” and searching for publication using VOSViewer, a color bar is shown at the bottom right topics containing the strings “ICD-9,” “ICD-10,” and corner of the graphic. +e color bar is shown only if colors “ICD-11.” are determined by item scores, which indicates how scores Figure 3(a) indicates that the number of ICD-11-related are mapped to colors. papers in 2017 exceeds that of ICD-10-related papers in an During timeline visualization using CiteSpace, time is analysis by searching titles of publications. +e numbers of mapped to the horizontal position, and clusters are arranged publications related to ICD-10 gradually increased over the along these horizontal lines. Users can adjust a complex set period of 1989 and 2018, peaked at approximately 160 of parameters to control the analysis process as well as publications in 2014, and then rapidly decreased to 87 in interact and manipulate the visualization of a knowledge 2019. However, the number of ICD-9-related research each domain. year remained stable between 20 and 30. +e figure shows that ICD-11 has become the focus of ICD-related studies. Figure 3(b) indicates that the numbers of publications 3. Results related to ICD-9, ICD-10, and ICD-11 approximately in- 3.1. Trends of ICD-Related Research. Figure 3 shows the creased over each publication year in an analysis by changes in number of publications by publication year searching topics of publications. +e number of publications using different searching strategies, namely, searching related to ICD-11 each year is much smaller than those of Journal of Healthcare Engineering 5 1990 1995 2000 2005 2010 2015 2020 Publication year Title ICD-11 ICD-10 ICD-9 (a) 1990 1995 2000 2005 2010 2015 2020 Publication year Topic ICD-11 ICD-10 ICD-9 (b) Figure 3: Changes in numbers of publications related to ICD-9, ICD-10, and ICD-11 with publication year according to different search strategies. (a) Publication titles containing ICD-9, ICD-10, and ICD-11. (b) Publication topics containing ICD-9, ICD-10, and ICD-11. ICD-9- and ICD-10-related research. +e number of ICD-9- data and knowledge discovery. +e number of publica- related research publications exceeded that of ICD-10-re- tions related to information systems, big data, and lated research between 2012 and 2018. +e number of knowledge discovery increased roughly each year but publications related to ICD-11 continually increased to over studies on information systems were published much 200 in 2019. ICD-11 related research may include keywords earlier and with greater frequency than studies on big data of ICD-9 and ICD-10. Overall, research topics on ICD-11 research and knowledge discovery. +e trends of ICD- and, in turn, the number of relevant publications, began to related research may be expected to play a crucial role in big data analysis based on ICD-coded data and knowl- show an upward trend in 2006. Figure 4 depicts a timeline view of ICD-related research edge-based systems. trends over the period of 2009–2018 by using CiteSpace. +e visualization results in the figure demonstrate the ICD-re- 3.2. Distribution of Keywords. +e distribution of keywords lated topics extracted from the keywords of the retrieved in this bibliometric analysis is summarized in Table 2. +e publications from 2009 to 2018. A larger circle in the figure Total Link Strength (TLS) attribute in the table indicates the indicates a higher popularity of the corresponding topics in number of links of an item with other items and the total the year; conversely, a small circle indicates that the key- strength of the links of an item with other items, respectively. word-related research is less popular. +e topics in the figure +e tabular results show the top 10 keywords involved in the are clustered into seven groups, namely, “ICD-11 defini- three perspectives according to our analysis using VOS- tion,” “relevant specifier,” “obsessive–compulsive disorder,” Viewer. Network visualizations of the keywords in different “healthcare-related harm,” “false positive problem,” “ab- perspectives also follow. normal anxiety,” and “gender incongruence.” Figure 5 illustrates the trends of the number of pub- lications related to the three perspectives discussed earlier. 3.3. Perspective of Information Systems. +e overlay visual- Results indicate that, over the last 30 years, ICD has been ization with respect to the perspective of information sys- more extensively used in information systems than in big tems is shown in Figure 6; Table 2 (Part A) shows the Number of papers Number of papers 6 Journal of Healthcare Engineering Figure 4: Timeline view of ICD-related research over the period of 2009–2018. 30 +e visualization presented in Figure 6 implies that the adaptation process for ICD-11 will be more efficient than that for ICD-10 despite the 10 years required to develop the clinical modification of ICD-10. +e politicking and issues encountered over the past 10 years need not be repeated. 3.4. Perspective of Big Data. Table 2 (Part B) shows the statistics of popular keywords in big data research. +e number of publications related to big data-related re- 0 search is smaller than that related to the use of in- formation systems, as shown in Figure 7. First, big data- 1980 1990 2000 2010 2020 related research (8/51) with ICD is a promising field Publication year because the concept of big data in research only emerged Perspective after 2016. Second, big data research is associated with Information systems data mining, machine-learning algorithms, and man- Big data agement, whereas past research aspects focused on Knowledge clinical research, such as mortality (5/51), breast cancer (3/54), adolescents, and risk. +ird, connections (5/51) Figure 5: Trends of the numbers of publications related to in- formation systems, big data, and knowledge discovery. exist between big data and ICD; these connections in- clude statistics of mortality (5/51) and healthcare (3/51). statistics of popular keywords in medical information sys- Other additional valuable information can be found in tems. First, the network indicates that studies related to Figure 7. information systems are often associated with research on +e visualization in Figure 7 provides the past and administrative data (18/234), classification (20/234), and present trends of research and their connections to the mortality-related data (18/234). Second, ICD-11-related development of different versions of ICDs from different research is becoming a popular trend and often related to perspectives. +e results encourage the use of ICD, especially patient safety, population, mental disorders, and clinical ICD-11, in big data-driven algorithms. Big data-driven al- utilities, although ICD-10 also provides a broad research gorithms can adopt machine-learning-based methods that foundation that covers various needs in medical information enhance the statistics of ICD-11-coded big data for future systems (26/234). +ird, the network visualization illustrates mortality and morbidity research. major research directions that future testing and imple- mentation of ICD-11 should follow. Finally, the visualized results show the trends of the transition of use from ICD-9 3.5. Perspective of Knowledge Discovery. Table 2 (Part C) and ICD-10 to ICD-11. Figure 6 shows that the publication shows the statistics of popular keywords related to years of ICD-9- and ICD-10-related studies (green lines) are knowledge discovery. First, the results in Figure 8 show that between 2008 and 2012 and that the publication years of studies on knowledge discovery are usually related to the ICD-11-related studies (yellow lines) are between 2013 and analysis of ontology-based models for prediction, hospital 2018. management (4/54), and quality of life (4/54). Second, the Number of publications Journal of Healthcare Engineering 7 Table 2: Distribution of top 10 keywords in our bibliometric analysis from the perspectives of information systems, big data, and knowledge discovery. Perspective Keywords Occurrence Total link strength Administrative data 18 82 ICD-10 26 77 Mortality 18 72 Validation 12 67 Care 15 66 (A) Information systems Classification 20 56 Complications 10 52 Quality 9 50 Accuracy 9 47 ICD-9-CM 10 47 Big data 8 14 Care 6 13 ICD 5 13 Big cities 4 10 Mortality 5 10 (B) Big data Racial disparities 4 10 Survival 3 10 Follow-up 3 9 Health 3 9 Breast cancer 3 8 Implantable cardioverter-defibrillator 4 12 Quality of life 4 12 Care 4 9 Classification 8 9 ICD 6 9 (C) Knowledge discovery Knowledge 4 8 Management 4 8 Anxiety 6 7 Sudden cardiac death 4 7 Diagnosis 4 6 Figure 6: Overlay visualization of 320 publications related to the use of ICD in information systems published over the period of 2006–2018. 8 Journal of Healthcare Engineering Figure 7: Overlay visualization of 51 publications related to the use of ICD in big data over the period of 2008–2018. Figure 8: Overlay visualization of 54 publications related to the use of ICD in knowledge discovery. use of ICD-10 in knowledge discovery is currently a hot improving predictions and risk management from the spot; thus, ICD-11 may be expected to become increasingly clinical perspective. popular in the future. +ird, external knowledge sources (4/ +e visualization analysis in Figure 8 indicates that the 54), such as guidelines and databases in the medical field, number of publications related to knowledge discovery is are necessary for knowledge discovery from medical rec- smaller than that related to the implementation of in- formation systems and big data-driven analysis. +e results ords. +ird, most research related to knowledge discovery is disease based. +erefore, ICD-based knowledge dis- provide new perspectives on the use of the newly designed covery in healthcare is insufficient but promising for ICD-11 to enhance knowledge discovery in medical big data. Journal of Healthcare Engineering 9 coding nomenclature should be developed for coders, 4. Discussion medical staff, nurses, and allied health providers (e.g., re- +is study conducts a bibliometric analysis of research spiratory, physical, and occupational therapists). +e WHO published over the period of 1989–2018 to examine the has developed ICD-11 Application Programming Interfaces development of ICD-related research and its trends. +e (ICD-API) and its container version of the ICD-API to time trend analysis indicates that ICD-11 related research support ICD-11 implementation in hospitals [33]. Fourth, has grown rapidly in recent years compared with ICD-9 and physician practices may face financial and operational ICD-10 studies and that the popular research directions of burdens from ICD-10 implementation and other techno- ICD-11 include the topics psychiatry, psychology, in- logical requirements. Finally, the move from the diagnosis formation science, library science, and behavioral science. In and procedural codes of ICD-9 and ICD-10 to those of ICD- terms of perspective, information system-related research is 11 may raise concerns about protected health information more common than big data- and knowledge discovery- security and privacy risks. related work. Information system research is associated with Automated ICD coding for medical records based on keywords including “administrative data,” “ICD-10,” diagnostic information is the most popular research di- “mortality,” and “validation.” However, trends also show rection taken by ICD coding experts to improve the effi- that big data- and knowledge discovery-related research has ciency and accuracy of ICD coding in hospitals. Automated become more popular in recent years. Big data-related re- coding is a complicated computer-aided process involving search is associated with keywords such as “healthcare” and numerous task-oriented algorithms, such as natural lan- “mortality,” while knowledge discovery-based research is guage processing techniques [34] and semantic web tech- related to keywords such as “quality-of-life,” “management,” nology [35], that allow utilization of ICD coding rules to and “anxiety.” +e use of ICD-11 has facilitated the de- support coding of medical records. +e WHO provides us velopment of medical informatics from the perspectives of with a simple ICD-11 coding tool online to demonstrate the information systems, big data, and knowledge discovery. use of ICD-11 coding [36]. +e accuracy of ICD coding tasks +e release of ICD-11 affects the future implementation mainly relies on the abilities of coders from the medical of other standards in the medical field [25, 26]. For example, record department. +ese coders can code the summary of ICD is the most important reference for categorizing dis- diagnosis in electronic health records after patient discharge. ease-related groups in disease-based payment in a hospital. If the coders have questions related to a patient’s records, In this case, the necessity and accuracy of ICD coding is they will ask doctors to clarify the information to maintain highly important. +e use of artificial intelligent techniques medical record quality. However, inexperienced ICD coders also frees clinical coders from the burden of coding records produce medical records with poor coding quality that, in [27]. +erefore, developing big data-driven intelligent al- turn, provide poor-quality big data for analysis [37]. Oc- gorithms that automatically learn massive information from casionally, several semantic web-based approaches for medical record pathological sections and image data to harvesting multilingual textual definitions in ICD-11 must provide guidance for diagnosis and disease treatment and be used among different countries [38]. +e use of the establish different disease models have become more crucial MapReduce model and proper expert knowledge in auto- in the era of intelligent healthcare than in the past [28–30]. mated ICD coding provides high-accuracy and efficient By extracting and structuring ICD-11-coded data and uti- statistics for electronic health records [39]. +us, novel lizing expert knowledge, such as ICD-11 and SNOMED CT, coding algorithms for ICD-11 are necessary to revise and the algorithm with the use of ICD-11 could hold potential adapt actual scenes in big data-driven healthcare. value for solving critical healthcare problems that cannot be Given that ontology modeling can represent knowledge solved by traditional ICD-10. and support knowledge reasoning in specific fields, an +e ICD was initially established to provide mortality ontology-based ICD is naturally suitable for providing a and morbidity statistics [31]. +e future use of ICD-11 in knowledge base for decision-making in healthcare [40, 41]. healthcare expands the utility of ICD-11-coded big data in Researchers can also develop a customized version of healthcare. Facilitating statistical analysis by using ICD-11 WebProteg ´ e´ [42] to support the collaborative development for decision-making based on big data is a key concern of of ICD-11 content. Other ontology-based algorithms may governments and the WHO because these institutions may also be useful to enhance ICD-11-based decision-making require an overview of healthcare data to improve national [43]. +erefore, an important use of ICD-11 is the imple- healthcare policies and provide early warnings for diseases mentation of knowledge discovery in healthcare by trans- and risks [26, 32]. +e information systems of hospitals will forming big data into healthcare knowledge [44]. Big data- likely entail multiple upgrades to support the transition from driven algorithms for knowledge discovery using ICD-11 are ICD-9 and ICD-10 to ICD-11. Numerous reasons for these related to numerous application scenes. For example, re- upgrades may be cited. First, ICD-11 adoption may require searchers may search for a specific diagnostic item in the considerable technological modifications, such as ontology context of disease classification or establish a conceptual modeling for IT vendors, trading partners, external knowledge network from the narratives of electronic health reporting entities, and third-party payors. Second, pro- records by using the entities and their relationships found in ductivity loss is anticipated in functional areas that routinely ICD-11. +rough involvement with other big data sources use ICD-9 and ICD-10 codes. +ird, training programs for related to healthcare, such as massive medical records, ICD- new/revised clinical documentation requirements and 11-based big data analysis algorithms for knowledge 10 Journal of Healthcare Engineering discovery can provide considerable insights into the po- ICD-11-based research is revolutionizing medical in- tential value of big data in healthcare [45]. In these cases, the formatics. +e potential value of the general features, con- algorithm usually requires external knowledge sources for cepts, and code structures of ICD-11 to naturally support big ICD-11-based applications. data-driven medical informatics is examined, and findings According to the current analysis, the ICD has been illustrate the potential uses of ICD-11 in statistical analysis, proven to be the most important component of healthcare automated ICD-11 coding, and knowledge discovery in big information systems for clinical research, medical moni- data in healthcare. +e results further suggest that stake- toring, and public health management on a global scale [46]. holders should be aware of the future use of ICD-11 to ICD-based statistics from the big data perspective [47] in- overcome the challenges encountered in earlier imple- cludes the causes of death, diseases, injuries, and symptoms, mentations of ICD-11 in medical informatics. Substantial as well as diagnostic and external disease factors. Hospitals time and money are required to hire staff in the fields of in different regions can use ICD codes to share and compare medical research, information technology, and adminis- equivalent medical data and promote medical and financial tration to complete the transition from ICD-10 to ICD-11. information management [48]. However, the transition from ICD-10 to ICD-11 complicates the further develop- Data Availability ment of support tools for medical information systems [49–51]. Although the ICD is valuable to research on +e data used to support the findings of this study are healthcare-related diseases, implementing this system in available from the corresponding author upon request. each member state of the WHO is difficult. An inevitable tension exists between the incorporation of locally relevant Conflicts of Interest material and the essential purpose of ICD-11, which is to reliably convey clinical information across diverse bound- +e authors declare that there are no conflicts of interest aries [52]. For example, over 20 years since ICD-10 was first regarding the publication of this paper. released, only about 100 countries have reached ICD-10 standards because the number of codes has increased. In Acknowledgments addition, doctors’ workloads have increased after adoption of the ICD-11 because patients’ diseases, diagnoses, and +is work was supported in part by a key project of National treatment must be recorded as accurately and as precisely as Natural Science Foundation of China (71532002), a major possible. Medical institutions have had to upgrade their project of the National Social Science Foundation of China healthcare information systems to adapt to the needs of (18ZDA086), a key project of Beijing Social Science Foun- ICD-11 coding. Substantial time and money are required to dation Research Base (18JDGLA017), and the Fundamental hire staff in the fields of medical research, information Research Funds for the Central Universities of China technology, and administration to complete the transition (2018YJS072). from ICD-10 to ICD-11 [53]. +e limitations of this study are as follows. First, the References publication data used in this study mainly come from Web of Science, one of the most reputed indexing databases for [1] World Health Organization, “ICD-11,” 2019, https://icd.who. publications. +e ICD is used worldwide for different int. [2] M. Viceconti, P. Hunter, and R. Hose, “Big data, big purposes, and the outcomes are not always documented as knowledge: big data for personalized healthcare,” IEEE scientific published data. Although several other relevant Journal of Biomedical and Health Informatics, vol. 19, no. 4, studies are discussed, other work outside the database may pp. 1209–1215, 2015. not be fully considered in this study. Second, the main [3] L. K. Jones, R. Pulk, M. R. Gionfriddo, M. A. Evans, and contents of the selected publications were not retrieved for D. Parry, “Utilizing big data to provide better health at lower analysis; instead, meta data of the publications were con- cost,” American Journal of Health-System Pharmacy, vol. 75, sidered in keyword and topic analyses. +ird, we focused on no. 7, pp. 427–435, 2018. studies related to the perspective of medical informatics, [4] G. B. Mitropoulos, “+e DSM-ICD diagnostic approach as an rather than clinical perspectives; a thorough review of such essential bridge between the patient and the “big data”,” medical research may be needed to further analyze findings Psychiatriki, vol. 29, no. 3, pp. 249–256, 2018. in the main contents. Finally, the three perspectives used as a [5] T. Lancet, “ICD-11: a brave attempt at classifying a new world,” 3e Lancet, vol. 391, no. 10139, p. 2476, 2018. basis for analysis were selected subjectively; ICD-11 can [6] T. Lancet, “ICD-11,” 3e Lancet, vol. 393, no. 10188, p. 2275, provide other implicit uses for medical informatics in the future. [7] B. Pocai, “+e ICD-11 has been adopted by the world health assembly,” World Psychiatry, vol. 18, no. 3, pp. 371-372, 2019. 5. Conclusions [8] G. Bouzille, ´ C. Poirier, B. Campillo-Gimenez et al., “Leveraging hospital big data to monitor flu epidemics,” +is study conducts a bibliometric analysis on the devel- Computer Methods and Programs in Biomedicine, vol. 154, opment of ICD-related research from three perspectives. +e pp. 153–160, 2018. analysis results indicate that ICD-11-related studies have [9] T. Lancet, “Revising the ICD: explaining the WHO approach,” rapidly developed in recent years. Further development of 3e Lancet, vol. 388, no. 10059, pp. 2476-2477, 2016. Journal of Healthcare Engineering 11 [10] World Health Organization, “ICD-11 reference guide,” 2019, Engineering Conference, pp. 156-157, Shreveport, LA, USA, https://icd.who.int/icd11refguide/en/index.html. 2016. [11] M. G. Weiner, “Point: is ICD-10 diagnosis coding important [26] V. Poznyak, G. M. Reed, and M. E. Medina-Mora, “Aligning in the era of big data? Yes,” Chest, vol. 153, no. 5, the ICD-11 classification of disorders due to substance use pp. 1093–1095, 2018. with global service needs,” Epidemiology and Psychiatric [12] F. Farhangmehr, “Statistical approaches for big data analytics Sciences, vol. 27, no. 3, pp. 212–218, 2017. and machine learning: data-driven network reconstruction [27] J. Simeone, T. D. Bhagnai, T. Rhodes, and M. Reynolds, and predictive modeling of time series biological systems,” “Comparison of ICD-9 to ICD-10 crosswalks derived by physician and clinical coder vs. automated methods,” Value in ProQuest Dissertations Publishing, University of California, San Diego, CA, USA, 2014. Health, vol. 20, no. 8, p. A741, 2017. [28] M. Wang, A. Cyhaniuk, D. L. Cooper, and N. Iyer, “Identi- [13] D. Gu, J. Li, X. Li, and C. Liang, “Visualizing the knowledge structure and evolution of big data research in healthcare fication of people with acquired hemophilia in a large elec- informatics,” International Journal of Medical Informatics, tronic health record database,” Journal of Blood Medicine, vol. 98, pp. 22–32, 2017. vol. 8, no. 8, pp. 89–97, 2017. [14] C. H. Ko, H. C. Lin, P. C. Lin, and J. Y. Yen, “Validity, [29] T. Lancet, “ICD-11: in praise of good data,” 3e Lancet In- functional impairment and complications related to Internet fectious Diseases, vol. 18, no. 8, p. 813, 2018. gaming disorder in the DSM-5 and gaming disorder in the [30] C. C. Yang and P. Veltri, “Intelligent healthcare informatics in big data era,” Artificial Intelligence in Medicine, vol. 65, no. 2, ICD-11,” Australian & New Zealand Journal of Psychiatry, Article ID 0004867419881499, 2019. pp. 75–77, 2015. [31] S. Brocco, P. Vercellion, Ca. A. Goldoni, N. Alba, M. Gatti, [15] J. M. Rodrigues, S. Schulz, A. Rector, K. Spackman, J. Millar, and J. J. Campbell, “ICD-11 and SNOMED CT common and D. Agostini, “‘Bridge coding’ ICD-9, ICD-10 and effects ontology: circulatory system,” Studies in Health Technology on mortality statistics,” Epidemiologia & Prevenzione, vol. 34, and Informatics, vol. 205, pp. 1043–1047, 2017. no. 3, pp. 109–119, 2010. [16] G. Jiang, H. R. Solbrig, R. J. G. Chalmers, K. Spackman, [32] F. Mesle, J. Vallin, and G. Rogers, “+e effect of ICD-10 on A. L. Rector, and C. G. Chute, “A case study of ICD-11 continuity in cause-of-death statistics. +e example of anatomy value set extraction from SNOMED CT,” in Pro- France,” Population, vol. 63, no. 2, pp. 347–359, 2008. [33] World Health Organization, “ICD-API,” 2019, https://icd. ceedings of the International Conference on Biomedical On- tology, pp. 122–138, Buffalo, NY, USA, July 2011. who.int/icdapi. [34] D. Chen, R. Zhang, and R. G. Qiu, “Leveraging semantics in [17] K. Bushra, J. Nathalie, M. Amy et al., “Systematic review of validated case definitions for diabetes in ICD-9-coded and WordNet to facilitate the computer-assisted coding of ICD- 11,” IEEE Journal of Biomedical and Health Informatics, p. 1, ICD-10-coded data in adult populations,” BMJ Open, vol. 6, no. 8, Article ID e009952, 2016. 2019. [18] A. P´erez, K. Gojenola, A. Casillas, M. Oronoz, and A. D´ıaz de [35] G. Jiang, H. R. Solbrig, and C. G. Chute, “Using semantic web Ilarraza, “Computer aided classification of diagnostic terms in technology to support ICD-11 textual definitions authoring,” Spanish,” Expert Systems with Applications, vol. 42, no. 6, Journal of Biomedical Semantics, vol. 4, no. 1, p. 11, 2013. [36] World Health Organization, “ICD-11 coding tool,” 2019, pp. 2949–2958, 2015. [19] C. Austin and F. Kusumoto, “+e application of big data in https://icd.who.int/ct11/icd11_mms/en/release. [37] A. Nishio, E. Kimura, R. Seto, Y. Sato, and H. Mizushima, medicine: current implications and future directions,” Journal of Interventional Cardiac Electrophysiology, vol. 47, no. 1, “+e survey for determining knowledge-related problems in the dissemination of ICD-11,” Studies in Health Technology pp. 51–59, 2016. [20] D. A. Southern, M. Hall, D. E. White et al., “Opportunities and and Informatics, vol. 264, pp. 1865-1866, 2019. [38] G. Jiang, H. R. Solbrig, and C. G. Chute, “A semantic web- challenges for quality and safety applications in ICD-11: an international survey of users of coded health data,” In- based approach for harvesting multilingual textual definitions ternational Journal for Quality in Health Care, vol. 28, no. 1, from Wikipedia to support ICD-11 revision,” in Proceedings of the ICBO 2013 Workshops: International Workshop on pp. 129–135, 2016. [21] M. Halper, H. Gu, Y. Perl, and C. Ochs, “Abstraction networks Vaccine and Drug Ontology Studies, VDOS 2013, International Workshop on Definitions in Ontologies, DO 2013, Part of the for terminologies: supporting management of ‘Big knowl- edge’,” Artificial Intelligence in Medicine, vol. 64, no. 4, Semantic Trilogy 2013, CEUR-WS, vol. 1061, p. 1061, Annecy, pp. 1–16, 2015. France, October 2013. [22] C. Sideris, S. Shaikh, H. Kalantarian, and M. A. Sarrafzadeh, [39] K. Grolinger, M. Hayes, W. A. Hlgashino, A. L’Heureux, “A big-data platform for medical knowledge extraction from D. S. Allison, and M. A. M. Capretz, “Challenges for Map- electronic health records: automatic assignment of ICD-9 Reduce in big data,” in Proceedings of the 2014 IEEE 10th codes,” in Proceedings of the ACM International Conference on World Congress on Services, pp. 182–189, Anchorage, AK, USA, June 2014. Pervasive Technologies Related to Assistive Environments, p. 77, Island, Greece, November 2016. [40] M. B. First, G. M. Reed, S. E. Hyman, and S. Saxena, “+e development of the ICD-11 clinical descriptions and di- [23] A. Perianes-Rodriguez, L. Waltman, and N. J. van Eck, “Constructing bibliometric networks: a comparison between agnostic guidelines for mental and behavioural disorders,” full and fractional counting,” Journal of Informetrics, vol. 10, World Psychiatry, vol. 14, no. 1, pp. 82–90, 2015. no. 4, pp. 1178–1195, 2016. [41] S. Nitsuwat and W. Paoin, “Development of ICD-10-TM [24] C. Chen, “Science mapping: a systematic review of the lit- ontology for a semi-automated morbidity coding system in erature,” Journal of Data and Information Science, vol. 2, no. 2, +ailand,” Methods of Information in Medicine, vol. 51, no. 6, pp. 1–40, 2017. pp. 519–528, 2012. [25] S. M. Krishnan, “Application of analytics to big data in [42] T. Tudorache, N. Csongor, N. F. Noy, and M. A. Musen, healthcare,” in Proceedings of the 32nd Southern Biomedical “WebProteg´ ´ e: a collaborative ontology editor and knowledge 12 Journal of Healthcare Engineering acquisition tool for the web,” Semantic Web, vol. 4, no. 1, pp. 89–99, 2013. [43] Y. Lou, S. W. Tu, C. Nyulas, T. Tudorache, R. J. G. Chalmers, and M. A. Musen, “Use of ontology structure and Bayesian models to aid the crowdsourcing of ICD-11 sanctioning rules,” Journal of Biomedical Informatics, vol. 68, pp. 20–34, [44] H. Yang, E. Kundakcioglu, J. Li et al., “Healthcare intelligence: turning data into knowledge,” IEEE Intelligent Systems, vol. 29, no. 3, pp. 54–68, 2014. [45] M. Rotmensch, Y. Halpern, A. Tlimat, S. Horng, and D. Sontag, “Learning a health knowledge graph from elec- tronic medical records,” Scientific Reports, vol. 7, no. 1, p. 5994, 2017. [46] P. Mirhaji, D. Allemang, R. Coyne, and S. W. Casscells, “Improving the public health information network through semantic modeling,” IEEE Intelligent Systems, vol. 22, no. 3, pp. 13–17, 2007. [47] G. Rey, K. Bounebache, and C. Rondet, “Causes of deaths data, linkages and big data perspectives,” Journal of Forensic and Legal Medicine, vol. 57, pp. 37–40, 2018. [48] V. Sundararajan, P. S. Romano, H. Quan et al., “Capturing diagnosis-timing in ICD-coded hospital data: recommenda- tions from the WHO ICD-11 topic advisory group on quality and safety,” International Journal for Quality in Health Care, vol. 27, no. 4, pp. 328–333, 2015. [49] J. Seare, J. Yang, S. Yu, and V. Zarotsky, “Building a bridge: ICD-9-CM to ICD-10-CM mapping challenges and solu- tions,” Value in Health, vol. 17, no. 3, p. A187, 2014. [50] K. W. Fung, R. Richesson, M. Smerek et al., “Preparing for the ICD-10-CM transition: automated methods for translating ICD codes in clinical phenotype definitions,” eGEMs, vol. 4, no. 1, p. 1211, 2016. [51] D. Chen, R. Zhang, and X. Zhu, “Leveraging Shannon entropy to validate the transition between ICD-10 and ICD-11,” Entropy, vol. 20, no. 10, p. 769, 2018. [52] O. Gureje, R. Lewis-Fernandez, B. J. Hall, and G. M. Reed, “Systematic inclusion of culture-related information in ICD- 11,” World Psychiatry, vol. 18, no. 3, pp. 357-358, 2019. [53] World Health Organization, “ICD-11 implementation or transition guide,” 2019, https://icd.who.int/docs/ICD-11% 20Implementation%20or%20Transition%20Guide_v105.pdf. International Journal of Advances in Rotating Machinery Multimedia Journal of The Scientific Journal of Engineering World Journal Sensors Hindawi Hindawi Publishing Corporation Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Journal of Control Science and Engineering Advances in Civil Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com Journal of Journal of Electrical and Computer Robotics Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 VLSI Design Advances in OptoElectronics International Journal of Modelling & Aerospace International Journal of Simulation Navigation and in Engineering Engineering Observation Hindawi Hindawi Hindawi Hindawi Volume 2018 Volume 2018 Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com www.hindawi.com www.hindawi.com Volume 2018 International Journal of Active and Passive International Journal of Antennas and Advances in Chemical Engineering Propagation Electronic Components Shock and Vibration Acoustics and Vibration Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Healthcare Engineering Hindawi Publishing Corporation

A Bibliometric Analysis of the Development of ICD-11 in Medical Informatics

Download PDF
13 pages

Loading next page...
 
/lp/hindawi-publishing-corporation/a-bibliometric-analysis-of-the-development-of-icd-11-in-medical-yZoVvR0hyL

References (52)

Publisher
Hindawi Publishing Corporation
Copyright
Copyright © 2019 Donghua Chen et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ISSN
2040-2295
eISSN
2040-2309
DOI
10.1155/2019/1649363
Publisher site
See Article on Publisher Site

Abstract

Hindawi Journal of Healthcare Engineering Volume 2019, Article ID 1649363, 12 pages https://doi.org/10.1155/2019/1649363 Research Article A Bibliometric Analysis of the Development of ICD-11 in Medical Informatics 1 1 2 1 Donghua Chen , Runtong Zhang , Hongmei Zhao, and Jiayi Feng Department of Information Management, School of Economics and Management, Beijing Jiaotong University, Beijing 100044, China Peking University People’s Hospital, Beijing 100044, China Correspondence should be addressed to Runtong Zhang; rtzhang@bjtu.edu.cn Received 3 September 2019; Revised 7 November 2019; Accepted 22 November 2019; Published 26 December 2019 Academic Editor: Cesare F. Valenti Copyright © 2019 Donghua Chen et al. +is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +e International Classification of Diseases (ICD), which is used to group and report health conditions and factors, provides a basis for healthcare statistics. +e 11th revision of the ICD (ICD-11) released by the World Health Organization provides stakeholders with novel perspectives on solving the complexity of critical problems in medical informatics. +is study conducts a bibliometric analysis of research published over the period of 1989–2018 to examine the development of ICD-related research and its trends. First, over 4000 ICD-related papers spanning the 30-year period are retrieved from the Web of Science database. +en, based on the meta data of the selected papers, time trend analysis is performed to examine the development of different ICD revisions. Finally, the keywords and topics of these papers are analyzed and visualized using VOSViewer and CiteSpace. Our findings indicate that ICD-11-related research has grown rapidly in recent years compared with studies on ICD-9 and ICD-10. Moreover, the most popular research directions of ICD-11 include the topics psychiatry, psychology, information science, library science, and behavioral science. In terms of perspectives, information system-related research is more common than big data- and knowledge discovery-related work. However, the popularity of big data- and knowledge discovery-related developments has grown in recent years. +e use of ICD-11 facilitates the development of medical informatics from the perspectives of information systems, big data, and knowledge discovery. to monitor epidemics and threats toward public health, 1. Introduction monitors the expenditure burden shouldered by patients, +e International Classification of Diseases (ICD) [1], which evaluates the progress in achieving public health objectives, was developed by the World Health Organization (WHO), determines the obligation of member states of the WHO to plays a crucial role in governments’ reporting, grouping, and provide free or subsidized medical services to their pop- statistical analyses of diseases and other health-related ulations, and develops appropriate healthcare services [3]. factors. +e wide use of ICD makes it a global standard for +erefore, the ICD is key to the sustainable development of medical big data research [4]. diagnostic health information and enables sustainable and systematic recording, analysis, interpretation, and com- ICD standards have been used in medicine and parison of mortality and morbidity rates of different healthcare for over 100 years. +e first ICD standards ini- countries at different time points. +e ICD also covers tially focused on the statistics of the causes of death. In 1946, various signs, symptoms, abnormal findings, complaints, the Interim Commission of the WHO was entrusted to take and social factors suitable for studies on financial aspects, over the revision of the ICD and introduced a method for such as billing or resource allocation, and provides a basis for disease classification [5]. At present, the most widely used big data in personalized healthcare [2]. Moreover, the ICD version of the ICD is its 10th revision (ICD-10), which was provides an information framework that allows stakeholders endorsed by the Forty-third World Health Assembly in 2 Journal of Healthcare Engineering Novel concepts in ICD-11, such as stem codes and post- 1990. After over a decade of revisions by numerous countries based on Internet-based maintenance platforms, the WHO coordination, are proposed to overcome the challenges encountered by ICD-10 in recent years because the latter is released the final version of the 11th revision of ICD (ICD- 11) in June 18, 2018, to provide a new de facto standard of now outdated in the clinical and classification points of view. disease coding for the twenty-first century [6]. +e ICD-11 For example, stem codes containing all pertinent in- was submitted to the 144th Executive Board Meeting in formation in a precombined fashion in ICD-11 are referred January 2019 and then to the 72nd World Health Assembly to as “pre-coordination;” when additional detail pertaining in May 2019 [7]. Following endorsement, the member states to a single condition is described by combining multiple of the WHO are expected to begin reporting on the basis of codes, the code combination is referred to as “post-co- ordination.” ICD-11 also allows stakeholders to operate in ICD-11 on January 1, 2022. +e development of new ICD standards is expected to an electronic environment and capture more information, especially for morbidity use cases. In summary, the newly revolutionize global medical informatics within the next few decades. Over the past 20 years, the ICD-10 has been widely proposed ICD-11 is more suitable for disease coding than past revisions of ICD in the new era of medical informatics. utilized to classify healthcare information. For example, ICD-10-coded hospital big data offer new opportunities for +is study aims to conduct a bibliometric analysis to monitoring flu epidemics [8]. Numerous ICD-10 national examine the development of ICD-11-based studies in modifications have been developed to adapt actual use in healthcare. +e data sources and search strings are first different countries. +e ICD-11 at present is ready for testing determined. Time trend analysis is then performed based on and implementation in accordance with the specific time- the selected papers. Keywords and topics are analyzed and lines and requirements of different countries [9]. +e visualized to summarize the main findings of our study from the perspectives of information systems, big data, and structure and design of the newly proposed ICD-11 are based on clinical practices over the past few decades and knowledge discovery. Finally, we discuss and conclude our work. differ considerably from those of its previous revisions [10]. ICD-11-coded medical records provide the basis of massive health statistics with the latest development of big data- 2. Materials and Methods driven intelligent healthcare using big data analytical plat- forms such as Apache Hadoop and Spark [11, 12]. However, Figure 1 illustrates the flowchart of this bibliometric analysis the increasing use of ICD-11 in medical and health big data of ICD-related research. First, we determine relevant key- reduces the applicability and relevance of past analytical words and conduct search strategies to retrieve ICD-related methods because the ICD-11 features new code schemes and research. Second, the Web of Science database is used to concepts that differ from previous ICDs, such as stem codes retrieve relevant publications. +ird, time trend analysis of representing entities or groupings of high relevance or ICD-related papers is performed. Finally, the analysis results clinical conditions that should always be described as a single are visualized from three perspectives. category. Appropriate utilization of ICD-11 for the analysis of mortality, morbidity, epidemiology, case mixing, quality and safety, primary care, and detailed information from 2.1. Data Sources. +e data sources of ICD-related work medical and health big data are essential to provide the basis published over a certain time period are selected to facilitate for big data research in health informatics [13]. this bibliometric analysis. Many researchers have applied ICD Introduction of the novel concepts of ICD-11 can to their research since 1990, when the ICD-10 was first en- overcome the problems of previous ICD revisions. +e dorsed. Given that the ICD-9 and ICD-10 have played a foundation component and content model are key concepts crucial role in promoting the development of medical in- in ICD-11. +e foundation component is a multidimen- formatics in the last 30 years [17], the trends of relevant sional collection of all ICD entities. +e content model studies in this period should be examined. +e Web of Science describes several specific diseases or disorders and is defined database is used to obtain high-quality papers. However, we by 13 attributes, namely, ICD entity title, classification acknowledge that the database may not contain several properties, textual definitions, terms, body system/body valuable papers in this field. We will synthesize and discuss part, temporal properties, subtype property severity, man- relevant literature. Relevant articles published over the period ifestation properties, causal properties, functioning prop- of 1989–2018 are retrieved from the Web of Science database erties, specific condition properties, treatment properties, by searching the keywords “ICD-9,” “ICD-10,” and “ICD-11” and diagnostic criteria. +e content model also illustrates in the article title field in the core set of the Web of Science background knowledge that provides the basis for the sys- database. +is process can search studies relevant to the ICD tematic definition of each ICD entity to enable computer- national modification because the name of ICD national ization. New disorders, such as gaming disorder, which modifications includes the keyword “ICD.” Table 1 shows the remains controversial, are introduced in ICD-11 [14]. In statistics of the publications selected from the database. contrast to ICD-10, ICD-11 is established on the basis of +e distribution of ICD-related subjects in the selected ontology models [15]. Several value sets in ICD-11 are publications over 30 years is illustrated in Figure 2. Psychiatry derived from external ontologies, such as the Systematized and psychology are the most popular subjects in ICD-related Nomenclature of Medicine–Clinical Terms (SNOMED CT) clinical research. Information science and library science are [16], which has played an important role in healthcare. other popular research fields that may focus on the Journal of Healthcare Engineering 3 big data. Finally, an overlay visualization of the existing 54 Begin publications related to ICD and knowledge discovery is presented to investigate the state of knowledge discovery Determine searching using ICD codes. +e distribution of keywords from these keywords and three perspectives is examined and discussed. strategies Existing literature 2.4. Topic Analysis. +e topics of ICD-related research to Determine data promote healthcare are as follows. sources of publications Web of science database 2.4.1. Information System Perspective. ICD codes provide Examine the trends of the basis of structured medical big data in healthcare. Most ICD-related research From three work used natural language processing and machine- perspectives learning techniques for textual analysis. Without pro- fessional clinical inspection, such as evaluation of the proper coding of the clinical statuses of patients, the collected data Information Big data-driven Knowledge may be imprecise. ICD codes also enable automated clas- systems analysis discovery sification of diagnostic terms, such as application of com- puter-assisted coding in Spanish [18]. +e ICD is useful for solving such problems and produces structured data that Overlay visualization of analyzing improve the reliability of results from big data analysis. results of the selected publications 2.4.2. Big Data Perspective. ICD codes can be related to Discussion different perspectives of big data in healthcare. Analysis of massive individual data from the perspectives of different End sources, dimensions, and time points often reveals trends that traditional medical research approaches cannot Figure 1: Flowchart of the bibliometric analysis in this study. show [19]. However, the contents of existing medical big data are occasionally incorrect, incomplete, and even un- available; few datasets are complete and valuable for research development of ICD standards. Behavioral science and purposes. +e precision and reliability of analyzing ICD- neuroscience neurology are related to clinical research. Health coded results in big data-driven algorithms rely on the science services and science technology are key fields focusing coding quality of ICD when ICD coders encode their on improvements in the practical use of ICD in medical and medical records [20]. health informatics. Research on health science services and science technology is key to intelligent healthcare. 2.4.3. Knowledge Discovery Perspective. +e performance of ICD-related analysis generally relies on changes in the main 2.2. Time Trend Analysis. Time trend analysis aims to ex- diagnosis in the discharge summary of patients or the ac- amine the development of different revisions of ICD over the curacy of techniques for extracting information from patient past 30 years. First, the trends of publications related to ICD- records by medical institutions [21]. Professional and 9, ICD-10, and ICD-11 over the selected period are ex- technical requirements for practitioners, especially for fresh amined. A timeline view of ICD-related research is then used coders, are stringent because ICD coders must establish a to analyze ICD-related topics extracted from the keywords clear disease classification framework in their mind. Disease- of publications retrieved for the period 2009–2018. Finally, related concepts and relations could be retrieved and as- three perspectives, namely, medical information systems, big sociated with other knowledge sources in medical domains data, and knowledge discovery, illustrate the trends of the with the use of ICD to facilitate clinical knowledge discovery number of publications for the past 30 years. from ICD-coded data [22]. 2.3. Keyword Analysis. +ree perspectives of ICD-related research are examined through overlay visualization of the 2.5. Tools for Visualization. VOSViewer [23] and CiteSpace publications. First, an overlay visualization of 234 publi- [24] are used to visualize the search results and examine the cations collected by using the keywords “ICD” and “in- key information and trends of publications on the use of formation systems” from the core set of the Web of Science ICD. database is presented to examine studies related to the During network visualization using VOSViewer, items are represented by their label by default by a circle. +e size implementation of ICD in medical information systems. Second, an overlay visualization of 51 publications related to of the label and circle of an item are determined by the weight of the item. +e higher the weight of an item, the the use of ICD in big data analytics is presented to examine the trends of ICD-related research from the perspective of larger the label and circle of this item. +e color of an item is 4 Journal of Healthcare Engineering Table 1: Main types of publications retrieved for bibliometric analysis. Number of ICD-9-related Publication type Number of ICD-10-related publications Number of ICD-11-related publications publications Article 390 1359 350 Other 178 367 101 Abstract 143 249 112 Meeting 67 89 21 Letter 31 40 31 Editorial 16 83 93 Review 8 116 66 Case report 3 8 1 Clinical trial 3 26 4 News 3 30 9 Reference material 3 8 1 Total 845 2375 789 Number of publications 0 100 200 300 400 500 600 700 Psychiatry Psychology Information Science Library Science Behavioral Sciences Neurosciences Neurology Health Care Sciences Services Science Technology Other Topics Pediatrics Public Environmental Occupational Health Mathematics Geriatrics Gerontology Pathology Demography General Internal Medicine Pharmacology Pharmacy Substance Abuse Toxicology Genetics Heredity Allergy Computer Science Sociology Immunology Mathematical Computational Biology Infectious Diseases Social Sciences Other Topics Figure 2: Research directions related to ICD over the period 1989–2018. determined by the cluster to which the item belongs. Lines titles of publications containing the keywords “ICD-9,” between items represent links. In the overlay visualization “ICD-10,” and “ICD-11” and searching for publication using VOSViewer, a color bar is shown at the bottom right topics containing the strings “ICD-9,” “ICD-10,” and corner of the graphic. +e color bar is shown only if colors “ICD-11.” are determined by item scores, which indicates how scores Figure 3(a) indicates that the number of ICD-11-related are mapped to colors. papers in 2017 exceeds that of ICD-10-related papers in an During timeline visualization using CiteSpace, time is analysis by searching titles of publications. +e numbers of mapped to the horizontal position, and clusters are arranged publications related to ICD-10 gradually increased over the along these horizontal lines. Users can adjust a complex set period of 1989 and 2018, peaked at approximately 160 of parameters to control the analysis process as well as publications in 2014, and then rapidly decreased to 87 in interact and manipulate the visualization of a knowledge 2019. However, the number of ICD-9-related research each domain. year remained stable between 20 and 30. +e figure shows that ICD-11 has become the focus of ICD-related studies. Figure 3(b) indicates that the numbers of publications 3. Results related to ICD-9, ICD-10, and ICD-11 approximately in- 3.1. Trends of ICD-Related Research. Figure 3 shows the creased over each publication year in an analysis by changes in number of publications by publication year searching topics of publications. +e number of publications using different searching strategies, namely, searching related to ICD-11 each year is much smaller than those of Journal of Healthcare Engineering 5 1990 1995 2000 2005 2010 2015 2020 Publication year Title ICD-11 ICD-10 ICD-9 (a) 1990 1995 2000 2005 2010 2015 2020 Publication year Topic ICD-11 ICD-10 ICD-9 (b) Figure 3: Changes in numbers of publications related to ICD-9, ICD-10, and ICD-11 with publication year according to different search strategies. (a) Publication titles containing ICD-9, ICD-10, and ICD-11. (b) Publication topics containing ICD-9, ICD-10, and ICD-11. ICD-9- and ICD-10-related research. +e number of ICD-9- data and knowledge discovery. +e number of publica- related research publications exceeded that of ICD-10-re- tions related to information systems, big data, and lated research between 2012 and 2018. +e number of knowledge discovery increased roughly each year but publications related to ICD-11 continually increased to over studies on information systems were published much 200 in 2019. ICD-11 related research may include keywords earlier and with greater frequency than studies on big data of ICD-9 and ICD-10. Overall, research topics on ICD-11 research and knowledge discovery. +e trends of ICD- and, in turn, the number of relevant publications, began to related research may be expected to play a crucial role in big data analysis based on ICD-coded data and knowl- show an upward trend in 2006. Figure 4 depicts a timeline view of ICD-related research edge-based systems. trends over the period of 2009–2018 by using CiteSpace. +e visualization results in the figure demonstrate the ICD-re- 3.2. Distribution of Keywords. +e distribution of keywords lated topics extracted from the keywords of the retrieved in this bibliometric analysis is summarized in Table 2. +e publications from 2009 to 2018. A larger circle in the figure Total Link Strength (TLS) attribute in the table indicates the indicates a higher popularity of the corresponding topics in number of links of an item with other items and the total the year; conversely, a small circle indicates that the key- strength of the links of an item with other items, respectively. word-related research is less popular. +e topics in the figure +e tabular results show the top 10 keywords involved in the are clustered into seven groups, namely, “ICD-11 defini- three perspectives according to our analysis using VOS- tion,” “relevant specifier,” “obsessive–compulsive disorder,” Viewer. Network visualizations of the keywords in different “healthcare-related harm,” “false positive problem,” “ab- perspectives also follow. normal anxiety,” and “gender incongruence.” Figure 5 illustrates the trends of the number of pub- lications related to the three perspectives discussed earlier. 3.3. Perspective of Information Systems. +e overlay visual- Results indicate that, over the last 30 years, ICD has been ization with respect to the perspective of information sys- more extensively used in information systems than in big tems is shown in Figure 6; Table 2 (Part A) shows the Number of papers Number of papers 6 Journal of Healthcare Engineering Figure 4: Timeline view of ICD-related research over the period of 2009–2018. 30 +e visualization presented in Figure 6 implies that the adaptation process for ICD-11 will be more efficient than that for ICD-10 despite the 10 years required to develop the clinical modification of ICD-10. +e politicking and issues encountered over the past 10 years need not be repeated. 3.4. Perspective of Big Data. Table 2 (Part B) shows the statistics of popular keywords in big data research. +e number of publications related to big data-related re- 0 search is smaller than that related to the use of in- formation systems, as shown in Figure 7. First, big data- 1980 1990 2000 2010 2020 related research (8/51) with ICD is a promising field Publication year because the concept of big data in research only emerged Perspective after 2016. Second, big data research is associated with Information systems data mining, machine-learning algorithms, and man- Big data agement, whereas past research aspects focused on Knowledge clinical research, such as mortality (5/51), breast cancer (3/54), adolescents, and risk. +ird, connections (5/51) Figure 5: Trends of the numbers of publications related to in- formation systems, big data, and knowledge discovery. exist between big data and ICD; these connections in- clude statistics of mortality (5/51) and healthcare (3/51). statistics of popular keywords in medical information sys- Other additional valuable information can be found in tems. First, the network indicates that studies related to Figure 7. information systems are often associated with research on +e visualization in Figure 7 provides the past and administrative data (18/234), classification (20/234), and present trends of research and their connections to the mortality-related data (18/234). Second, ICD-11-related development of different versions of ICDs from different research is becoming a popular trend and often related to perspectives. +e results encourage the use of ICD, especially patient safety, population, mental disorders, and clinical ICD-11, in big data-driven algorithms. Big data-driven al- utilities, although ICD-10 also provides a broad research gorithms can adopt machine-learning-based methods that foundation that covers various needs in medical information enhance the statistics of ICD-11-coded big data for future systems (26/234). +ird, the network visualization illustrates mortality and morbidity research. major research directions that future testing and imple- mentation of ICD-11 should follow. Finally, the visualized results show the trends of the transition of use from ICD-9 3.5. Perspective of Knowledge Discovery. Table 2 (Part C) and ICD-10 to ICD-11. Figure 6 shows that the publication shows the statistics of popular keywords related to years of ICD-9- and ICD-10-related studies (green lines) are knowledge discovery. First, the results in Figure 8 show that between 2008 and 2012 and that the publication years of studies on knowledge discovery are usually related to the ICD-11-related studies (yellow lines) are between 2013 and analysis of ontology-based models for prediction, hospital 2018. management (4/54), and quality of life (4/54). Second, the Number of publications Journal of Healthcare Engineering 7 Table 2: Distribution of top 10 keywords in our bibliometric analysis from the perspectives of information systems, big data, and knowledge discovery. Perspective Keywords Occurrence Total link strength Administrative data 18 82 ICD-10 26 77 Mortality 18 72 Validation 12 67 Care 15 66 (A) Information systems Classification 20 56 Complications 10 52 Quality 9 50 Accuracy 9 47 ICD-9-CM 10 47 Big data 8 14 Care 6 13 ICD 5 13 Big cities 4 10 Mortality 5 10 (B) Big data Racial disparities 4 10 Survival 3 10 Follow-up 3 9 Health 3 9 Breast cancer 3 8 Implantable cardioverter-defibrillator 4 12 Quality of life 4 12 Care 4 9 Classification 8 9 ICD 6 9 (C) Knowledge discovery Knowledge 4 8 Management 4 8 Anxiety 6 7 Sudden cardiac death 4 7 Diagnosis 4 6 Figure 6: Overlay visualization of 320 publications related to the use of ICD in information systems published over the period of 2006–2018. 8 Journal of Healthcare Engineering Figure 7: Overlay visualization of 51 publications related to the use of ICD in big data over the period of 2008–2018. Figure 8: Overlay visualization of 54 publications related to the use of ICD in knowledge discovery. use of ICD-10 in knowledge discovery is currently a hot improving predictions and risk management from the spot; thus, ICD-11 may be expected to become increasingly clinical perspective. popular in the future. +ird, external knowledge sources (4/ +e visualization analysis in Figure 8 indicates that the 54), such as guidelines and databases in the medical field, number of publications related to knowledge discovery is are necessary for knowledge discovery from medical rec- smaller than that related to the implementation of in- formation systems and big data-driven analysis. +e results ords. +ird, most research related to knowledge discovery is disease based. +erefore, ICD-based knowledge dis- provide new perspectives on the use of the newly designed covery in healthcare is insufficient but promising for ICD-11 to enhance knowledge discovery in medical big data. Journal of Healthcare Engineering 9 coding nomenclature should be developed for coders, 4. Discussion medical staff, nurses, and allied health providers (e.g., re- +is study conducts a bibliometric analysis of research spiratory, physical, and occupational therapists). +e WHO published over the period of 1989–2018 to examine the has developed ICD-11 Application Programming Interfaces development of ICD-related research and its trends. +e (ICD-API) and its container version of the ICD-API to time trend analysis indicates that ICD-11 related research support ICD-11 implementation in hospitals [33]. Fourth, has grown rapidly in recent years compared with ICD-9 and physician practices may face financial and operational ICD-10 studies and that the popular research directions of burdens from ICD-10 implementation and other techno- ICD-11 include the topics psychiatry, psychology, in- logical requirements. Finally, the move from the diagnosis formation science, library science, and behavioral science. In and procedural codes of ICD-9 and ICD-10 to those of ICD- terms of perspective, information system-related research is 11 may raise concerns about protected health information more common than big data- and knowledge discovery- security and privacy risks. related work. Information system research is associated with Automated ICD coding for medical records based on keywords including “administrative data,” “ICD-10,” diagnostic information is the most popular research di- “mortality,” and “validation.” However, trends also show rection taken by ICD coding experts to improve the effi- that big data- and knowledge discovery-related research has ciency and accuracy of ICD coding in hospitals. Automated become more popular in recent years. Big data-related re- coding is a complicated computer-aided process involving search is associated with keywords such as “healthcare” and numerous task-oriented algorithms, such as natural lan- “mortality,” while knowledge discovery-based research is guage processing techniques [34] and semantic web tech- related to keywords such as “quality-of-life,” “management,” nology [35], that allow utilization of ICD coding rules to and “anxiety.” +e use of ICD-11 has facilitated the de- support coding of medical records. +e WHO provides us velopment of medical informatics from the perspectives of with a simple ICD-11 coding tool online to demonstrate the information systems, big data, and knowledge discovery. use of ICD-11 coding [36]. +e accuracy of ICD coding tasks +e release of ICD-11 affects the future implementation mainly relies on the abilities of coders from the medical of other standards in the medical field [25, 26]. For example, record department. +ese coders can code the summary of ICD is the most important reference for categorizing dis- diagnosis in electronic health records after patient discharge. ease-related groups in disease-based payment in a hospital. If the coders have questions related to a patient’s records, In this case, the necessity and accuracy of ICD coding is they will ask doctors to clarify the information to maintain highly important. +e use of artificial intelligent techniques medical record quality. However, inexperienced ICD coders also frees clinical coders from the burden of coding records produce medical records with poor coding quality that, in [27]. +erefore, developing big data-driven intelligent al- turn, provide poor-quality big data for analysis [37]. Oc- gorithms that automatically learn massive information from casionally, several semantic web-based approaches for medical record pathological sections and image data to harvesting multilingual textual definitions in ICD-11 must provide guidance for diagnosis and disease treatment and be used among different countries [38]. +e use of the establish different disease models have become more crucial MapReduce model and proper expert knowledge in auto- in the era of intelligent healthcare than in the past [28–30]. mated ICD coding provides high-accuracy and efficient By extracting and structuring ICD-11-coded data and uti- statistics for electronic health records [39]. +us, novel lizing expert knowledge, such as ICD-11 and SNOMED CT, coding algorithms for ICD-11 are necessary to revise and the algorithm with the use of ICD-11 could hold potential adapt actual scenes in big data-driven healthcare. value for solving critical healthcare problems that cannot be Given that ontology modeling can represent knowledge solved by traditional ICD-10. and support knowledge reasoning in specific fields, an +e ICD was initially established to provide mortality ontology-based ICD is naturally suitable for providing a and morbidity statistics [31]. +e future use of ICD-11 in knowledge base for decision-making in healthcare [40, 41]. healthcare expands the utility of ICD-11-coded big data in Researchers can also develop a customized version of healthcare. Facilitating statistical analysis by using ICD-11 WebProteg ´ e´ [42] to support the collaborative development for decision-making based on big data is a key concern of of ICD-11 content. Other ontology-based algorithms may governments and the WHO because these institutions may also be useful to enhance ICD-11-based decision-making require an overview of healthcare data to improve national [43]. +erefore, an important use of ICD-11 is the imple- healthcare policies and provide early warnings for diseases mentation of knowledge discovery in healthcare by trans- and risks [26, 32]. +e information systems of hospitals will forming big data into healthcare knowledge [44]. Big data- likely entail multiple upgrades to support the transition from driven algorithms for knowledge discovery using ICD-11 are ICD-9 and ICD-10 to ICD-11. Numerous reasons for these related to numerous application scenes. For example, re- upgrades may be cited. First, ICD-11 adoption may require searchers may search for a specific diagnostic item in the considerable technological modifications, such as ontology context of disease classification or establish a conceptual modeling for IT vendors, trading partners, external knowledge network from the narratives of electronic health reporting entities, and third-party payors. Second, pro- records by using the entities and their relationships found in ductivity loss is anticipated in functional areas that routinely ICD-11. +rough involvement with other big data sources use ICD-9 and ICD-10 codes. +ird, training programs for related to healthcare, such as massive medical records, ICD- new/revised clinical documentation requirements and 11-based big data analysis algorithms for knowledge 10 Journal of Healthcare Engineering discovery can provide considerable insights into the po- ICD-11-based research is revolutionizing medical in- tential value of big data in healthcare [45]. In these cases, the formatics. +e potential value of the general features, con- algorithm usually requires external knowledge sources for cepts, and code structures of ICD-11 to naturally support big ICD-11-based applications. data-driven medical informatics is examined, and findings According to the current analysis, the ICD has been illustrate the potential uses of ICD-11 in statistical analysis, proven to be the most important component of healthcare automated ICD-11 coding, and knowledge discovery in big information systems for clinical research, medical moni- data in healthcare. +e results further suggest that stake- toring, and public health management on a global scale [46]. holders should be aware of the future use of ICD-11 to ICD-based statistics from the big data perspective [47] in- overcome the challenges encountered in earlier imple- cludes the causes of death, diseases, injuries, and symptoms, mentations of ICD-11 in medical informatics. Substantial as well as diagnostic and external disease factors. Hospitals time and money are required to hire staff in the fields of in different regions can use ICD codes to share and compare medical research, information technology, and adminis- equivalent medical data and promote medical and financial tration to complete the transition from ICD-10 to ICD-11. information management [48]. However, the transition from ICD-10 to ICD-11 complicates the further develop- Data Availability ment of support tools for medical information systems [49–51]. Although the ICD is valuable to research on +e data used to support the findings of this study are healthcare-related diseases, implementing this system in available from the corresponding author upon request. each member state of the WHO is difficult. An inevitable tension exists between the incorporation of locally relevant Conflicts of Interest material and the essential purpose of ICD-11, which is to reliably convey clinical information across diverse bound- +e authors declare that there are no conflicts of interest aries [52]. For example, over 20 years since ICD-10 was first regarding the publication of this paper. released, only about 100 countries have reached ICD-10 standards because the number of codes has increased. In Acknowledgments addition, doctors’ workloads have increased after adoption of the ICD-11 because patients’ diseases, diagnoses, and +is work was supported in part by a key project of National treatment must be recorded as accurately and as precisely as Natural Science Foundation of China (71532002), a major possible. Medical institutions have had to upgrade their project of the National Social Science Foundation of China healthcare information systems to adapt to the needs of (18ZDA086), a key project of Beijing Social Science Foun- ICD-11 coding. Substantial time and money are required to dation Research Base (18JDGLA017), and the Fundamental hire staff in the fields of medical research, information Research Funds for the Central Universities of China technology, and administration to complete the transition (2018YJS072). from ICD-10 to ICD-11 [53]. +e limitations of this study are as follows. First, the References publication data used in this study mainly come from Web of Science, one of the most reputed indexing databases for [1] World Health Organization, “ICD-11,” 2019, https://icd.who. publications. +e ICD is used worldwide for different int. [2] M. Viceconti, P. Hunter, and R. Hose, “Big data, big purposes, and the outcomes are not always documented as knowledge: big data for personalized healthcare,” IEEE scientific published data. Although several other relevant Journal of Biomedical and Health Informatics, vol. 19, no. 4, studies are discussed, other work outside the database may pp. 1209–1215, 2015. not be fully considered in this study. Second, the main [3] L. K. Jones, R. Pulk, M. R. Gionfriddo, M. A. Evans, and contents of the selected publications were not retrieved for D. Parry, “Utilizing big data to provide better health at lower analysis; instead, meta data of the publications were con- cost,” American Journal of Health-System Pharmacy, vol. 75, sidered in keyword and topic analyses. +ird, we focused on no. 7, pp. 427–435, 2018. studies related to the perspective of medical informatics, [4] G. B. Mitropoulos, “+e DSM-ICD diagnostic approach as an rather than clinical perspectives; a thorough review of such essential bridge between the patient and the “big data”,” medical research may be needed to further analyze findings Psychiatriki, vol. 29, no. 3, pp. 249–256, 2018. in the main contents. Finally, the three perspectives used as a [5] T. Lancet, “ICD-11: a brave attempt at classifying a new world,” 3e Lancet, vol. 391, no. 10139, p. 2476, 2018. basis for analysis were selected subjectively; ICD-11 can [6] T. Lancet, “ICD-11,” 3e Lancet, vol. 393, no. 10188, p. 2275, provide other implicit uses for medical informatics in the future. [7] B. Pocai, “+e ICD-11 has been adopted by the world health assembly,” World Psychiatry, vol. 18, no. 3, pp. 371-372, 2019. 5. Conclusions [8] G. Bouzille, ´ C. Poirier, B. Campillo-Gimenez et al., “Leveraging hospital big data to monitor flu epidemics,” +is study conducts a bibliometric analysis on the devel- Computer Methods and Programs in Biomedicine, vol. 154, opment of ICD-related research from three perspectives. +e pp. 153–160, 2018. analysis results indicate that ICD-11-related studies have [9] T. Lancet, “Revising the ICD: explaining the WHO approach,” rapidly developed in recent years. Further development of 3e Lancet, vol. 388, no. 10059, pp. 2476-2477, 2016. Journal of Healthcare Engineering 11 [10] World Health Organization, “ICD-11 reference guide,” 2019, Engineering Conference, pp. 156-157, Shreveport, LA, USA, https://icd.who.int/icd11refguide/en/index.html. 2016. [11] M. G. Weiner, “Point: is ICD-10 diagnosis coding important [26] V. Poznyak, G. M. Reed, and M. E. Medina-Mora, “Aligning in the era of big data? Yes,” Chest, vol. 153, no. 5, the ICD-11 classification of disorders due to substance use pp. 1093–1095, 2018. with global service needs,” Epidemiology and Psychiatric [12] F. Farhangmehr, “Statistical approaches for big data analytics Sciences, vol. 27, no. 3, pp. 212–218, 2017. and machine learning: data-driven network reconstruction [27] J. Simeone, T. D. Bhagnai, T. Rhodes, and M. Reynolds, and predictive modeling of time series biological systems,” “Comparison of ICD-9 to ICD-10 crosswalks derived by physician and clinical coder vs. automated methods,” Value in ProQuest Dissertations Publishing, University of California, San Diego, CA, USA, 2014. Health, vol. 20, no. 8, p. A741, 2017. [28] M. Wang, A. Cyhaniuk, D. L. Cooper, and N. Iyer, “Identi- [13] D. Gu, J. Li, X. Li, and C. Liang, “Visualizing the knowledge structure and evolution of big data research in healthcare fication of people with acquired hemophilia in a large elec- informatics,” International Journal of Medical Informatics, tronic health record database,” Journal of Blood Medicine, vol. 98, pp. 22–32, 2017. vol. 8, no. 8, pp. 89–97, 2017. [14] C. H. Ko, H. C. Lin, P. C. Lin, and J. Y. Yen, “Validity, [29] T. Lancet, “ICD-11: in praise of good data,” 3e Lancet In- functional impairment and complications related to Internet fectious Diseases, vol. 18, no. 8, p. 813, 2018. gaming disorder in the DSM-5 and gaming disorder in the [30] C. C. Yang and P. Veltri, “Intelligent healthcare informatics in big data era,” Artificial Intelligence in Medicine, vol. 65, no. 2, ICD-11,” Australian & New Zealand Journal of Psychiatry, Article ID 0004867419881499, 2019. pp. 75–77, 2015. [31] S. Brocco, P. Vercellion, Ca. A. Goldoni, N. Alba, M. Gatti, [15] J. M. Rodrigues, S. Schulz, A. Rector, K. Spackman, J. Millar, and J. J. Campbell, “ICD-11 and SNOMED CT common and D. Agostini, “‘Bridge coding’ ICD-9, ICD-10 and effects ontology: circulatory system,” Studies in Health Technology on mortality statistics,” Epidemiologia & Prevenzione, vol. 34, and Informatics, vol. 205, pp. 1043–1047, 2017. no. 3, pp. 109–119, 2010. [16] G. Jiang, H. R. Solbrig, R. J. G. Chalmers, K. Spackman, [32] F. Mesle, J. Vallin, and G. Rogers, “+e effect of ICD-10 on A. L. Rector, and C. G. Chute, “A case study of ICD-11 continuity in cause-of-death statistics. +e example of anatomy value set extraction from SNOMED CT,” in Pro- France,” Population, vol. 63, no. 2, pp. 347–359, 2008. [33] World Health Organization, “ICD-API,” 2019, https://icd. ceedings of the International Conference on Biomedical On- tology, pp. 122–138, Buffalo, NY, USA, July 2011. who.int/icdapi. [34] D. Chen, R. Zhang, and R. G. Qiu, “Leveraging semantics in [17] K. Bushra, J. Nathalie, M. Amy et al., “Systematic review of validated case definitions for diabetes in ICD-9-coded and WordNet to facilitate the computer-assisted coding of ICD- 11,” IEEE Journal of Biomedical and Health Informatics, p. 1, ICD-10-coded data in adult populations,” BMJ Open, vol. 6, no. 8, Article ID e009952, 2016. 2019. [18] A. P´erez, K. Gojenola, A. Casillas, M. Oronoz, and A. D´ıaz de [35] G. Jiang, H. R. Solbrig, and C. G. Chute, “Using semantic web Ilarraza, “Computer aided classification of diagnostic terms in technology to support ICD-11 textual definitions authoring,” Spanish,” Expert Systems with Applications, vol. 42, no. 6, Journal of Biomedical Semantics, vol. 4, no. 1, p. 11, 2013. [36] World Health Organization, “ICD-11 coding tool,” 2019, pp. 2949–2958, 2015. [19] C. Austin and F. Kusumoto, “+e application of big data in https://icd.who.int/ct11/icd11_mms/en/release. [37] A. Nishio, E. Kimura, R. Seto, Y. Sato, and H. Mizushima, medicine: current implications and future directions,” Journal of Interventional Cardiac Electrophysiology, vol. 47, no. 1, “+e survey for determining knowledge-related problems in the dissemination of ICD-11,” Studies in Health Technology pp. 51–59, 2016. [20] D. A. Southern, M. Hall, D. E. White et al., “Opportunities and and Informatics, vol. 264, pp. 1865-1866, 2019. [38] G. Jiang, H. R. Solbrig, and C. G. Chute, “A semantic web- challenges for quality and safety applications in ICD-11: an international survey of users of coded health data,” In- based approach for harvesting multilingual textual definitions ternational Journal for Quality in Health Care, vol. 28, no. 1, from Wikipedia to support ICD-11 revision,” in Proceedings of the ICBO 2013 Workshops: International Workshop on pp. 129–135, 2016. [21] M. Halper, H. Gu, Y. Perl, and C. Ochs, “Abstraction networks Vaccine and Drug Ontology Studies, VDOS 2013, International Workshop on Definitions in Ontologies, DO 2013, Part of the for terminologies: supporting management of ‘Big knowl- edge’,” Artificial Intelligence in Medicine, vol. 64, no. 4, Semantic Trilogy 2013, CEUR-WS, vol. 1061, p. 1061, Annecy, pp. 1–16, 2015. France, October 2013. [22] C. Sideris, S. Shaikh, H. Kalantarian, and M. A. Sarrafzadeh, [39] K. Grolinger, M. Hayes, W. A. Hlgashino, A. L’Heureux, “A big-data platform for medical knowledge extraction from D. S. Allison, and M. A. M. Capretz, “Challenges for Map- electronic health records: automatic assignment of ICD-9 Reduce in big data,” in Proceedings of the 2014 IEEE 10th codes,” in Proceedings of the ACM International Conference on World Congress on Services, pp. 182–189, Anchorage, AK, USA, June 2014. Pervasive Technologies Related to Assistive Environments, p. 77, Island, Greece, November 2016. [40] M. B. First, G. M. Reed, S. E. Hyman, and S. Saxena, “+e development of the ICD-11 clinical descriptions and di- [23] A. Perianes-Rodriguez, L. Waltman, and N. J. van Eck, “Constructing bibliometric networks: a comparison between agnostic guidelines for mental and behavioural disorders,” full and fractional counting,” Journal of Informetrics, vol. 10, World Psychiatry, vol. 14, no. 1, pp. 82–90, 2015. no. 4, pp. 1178–1195, 2016. [41] S. Nitsuwat and W. Paoin, “Development of ICD-10-TM [24] C. Chen, “Science mapping: a systematic review of the lit- ontology for a semi-automated morbidity coding system in erature,” Journal of Data and Information Science, vol. 2, no. 2, +ailand,” Methods of Information in Medicine, vol. 51, no. 6, pp. 1–40, 2017. pp. 519–528, 2012. [25] S. M. Krishnan, “Application of analytics to big data in [42] T. Tudorache, N. Csongor, N. F. Noy, and M. A. Musen, healthcare,” in Proceedings of the 32nd Southern Biomedical “WebProteg´ ´ e: a collaborative ontology editor and knowledge 12 Journal of Healthcare Engineering acquisition tool for the web,” Semantic Web, vol. 4, no. 1, pp. 89–99, 2013. [43] Y. Lou, S. W. Tu, C. Nyulas, T. Tudorache, R. J. G. Chalmers, and M. A. Musen, “Use of ontology structure and Bayesian models to aid the crowdsourcing of ICD-11 sanctioning rules,” Journal of Biomedical Informatics, vol. 68, pp. 20–34, [44] H. Yang, E. Kundakcioglu, J. Li et al., “Healthcare intelligence: turning data into knowledge,” IEEE Intelligent Systems, vol. 29, no. 3, pp. 54–68, 2014. [45] M. Rotmensch, Y. Halpern, A. Tlimat, S. Horng, and D. Sontag, “Learning a health knowledge graph from elec- tronic medical records,” Scientific Reports, vol. 7, no. 1, p. 5994, 2017. [46] P. Mirhaji, D. Allemang, R. Coyne, and S. W. Casscells, “Improving the public health information network through semantic modeling,” IEEE Intelligent Systems, vol. 22, no. 3, pp. 13–17, 2007. [47] G. Rey, K. Bounebache, and C. Rondet, “Causes of deaths data, linkages and big data perspectives,” Journal of Forensic and Legal Medicine, vol. 57, pp. 37–40, 2018. [48] V. Sundararajan, P. S. Romano, H. Quan et al., “Capturing diagnosis-timing in ICD-coded hospital data: recommenda- tions from the WHO ICD-11 topic advisory group on quality and safety,” International Journal for Quality in Health Care, vol. 27, no. 4, pp. 328–333, 2015. [49] J. Seare, J. Yang, S. Yu, and V. Zarotsky, “Building a bridge: ICD-9-CM to ICD-10-CM mapping challenges and solu- tions,” Value in Health, vol. 17, no. 3, p. A187, 2014. [50] K. W. Fung, R. Richesson, M. Smerek et al., “Preparing for the ICD-10-CM transition: automated methods for translating ICD codes in clinical phenotype definitions,” eGEMs, vol. 4, no. 1, p. 1211, 2016. [51] D. Chen, R. Zhang, and X. Zhu, “Leveraging Shannon entropy to validate the transition between ICD-10 and ICD-11,” Entropy, vol. 20, no. 10, p. 769, 2018. [52] O. Gureje, R. Lewis-Fernandez, B. J. Hall, and G. M. Reed, “Systematic inclusion of culture-related information in ICD- 11,” World Psychiatry, vol. 18, no. 3, pp. 357-358, 2019. [53] World Health Organization, “ICD-11 implementation or transition guide,” 2019, https://icd.who.int/docs/ICD-11% 20Implementation%20or%20Transition%20Guide_v105.pdf. International Journal of Advances in Rotating Machinery Multimedia Journal of The Scientific Journal of Engineering World Journal Sensors Hindawi Hindawi Publishing Corporation Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Journal of Control Science and Engineering Advances in Civil Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com Journal of Journal of Electrical and Computer Robotics Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 VLSI Design Advances in OptoElectronics International Journal of Modelling & Aerospace International Journal of Simulation Navigation and in Engineering Engineering Observation Hindawi Hindawi Hindawi Hindawi Volume 2018 Volume 2018 Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com www.hindawi.com www.hindawi.com Volume 2018 International Journal of Active and Passive International Journal of Antennas and Advances in Chemical Engineering Propagation Electronic Components Shock and Vibration Acoustics and Vibration Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018

Journal

Journal of Healthcare Engineering – Hindawi Publishing Corporation

Published: Dec 26, 2019

There are no references for this article.