READ THE ARTICLE THAT IS LISTED BELOW THEN:

Pretend you are in favor of this article and discuss these questions:

-what skills are needed to integrate and use analytic systems

-procedures needed to manage health information systems

-what to seek for a positive return on investment (ROI)

-what you learned from the article. Use competing facts, figures and assumptions.

ARTICLE:

Abstract

Healthcare informatics and analytics (HCI&A), also known as healthcare information technology (HIT), healthcare IS (HIS), and so on, has rapidly evolved with the emerge of advanced data analytics technologies applied to the medical domain. Currently, HCI&A has emerged as an important area of study for both practitioners and academic researchers. Accordingly, this emerging field has prompted for an inquiry of the opportunities and challenges related to management of healthcare data, and the application of advanced data analytics to the contemporary healthcare industry. In order to contribute to the literature of healthcare informatics and analytics, this study proposes an HCI&A framework under the context of big data, which covers four important segments such as the underlying technologies, system applications, system evaluations, and emerging research areas. Based on the key features and capabilities of underpinning technologies, the evolution of HCI&A are conceptualized by three stages, namely HCI&A 1.0, HCI&A 2.0, and HCI&A 3.0. By analyzing the technological growth and current research trends, this study outlines the trend map of HCI&A for education and knowledge transfer. We also contributed to conduct a bibliographic study on healthcare informatics and healthcare information systems. To the best of our knowledge, our study is among the very few comprehensive bibliographic studies about HCI&A. We hope that our study can contribute to supplement contemporary thoughts on HCI&A research, and facilitate the related knowledge transfer to the healthcare industry.

1.Introduction

Healthcare informatics and analytics (HCI&A) represents the applications of advanced technologies and data analytics in healthcare. Recently, HCI&A and the almost parallel discipline of big data analytics have become more and more popular for both practitioners and researchers. Over the last 20 years, new inventions in information and communication technologies, artificial intelligence, and advanced data analytics have already shifted the healthcare system toward a more efficient and effective mode than ever before. Therefore, HCI&A has emerged as a dominant field of study presenting the measurement and value of data affiliated problems to be answered in current healthcare industries. Regarding these matters, recently, the President's Council of Advisors on Science and Technology (PCAST) published a report on the application of information technology in US healthcare.1This statistic branded a data-centric method to realize the prospects of health technologies. Moreover, according to the National Academy of Engineering, cutting-edge medical care informatics is highlighted as one of the 14 grand challenges of engineering (Yang H. et al., 2014). In business settings, advanced healthcare technologies occupy a superior position to encounter different long-term requirements(Benko and Wilson, 2003). Currently, several research initiatives have also featured the significant development of healthcare technologies. For example, the National Institute of Health (NIH) and US National Science Foundation (NSF) are working together and have issued some joint program solicitations to invite research in data science and medical informatics (i.e., Smart and Connected Health (SCH)2and Computational and Data-Enabled Science and Engineering (CDS&E)3). Moreover, the collaboration with the NIH Institute of General Medical Sciences, division of mathematical science of the NSF has also released joint initiatives to support bioinformatics research(NSF 2013).

From the research outcomes, it is verified that information can be an extraordinary resource for healthcare industries and the benefits related to medical data and analysis in different healthcare organizations have helped show appreciable attention to HCI&A. HCI&A attracts researchers from multiple disciplines including computer science, social science, business, physics, biology, and medicine. The multidisciplinary approach to HCI&A showed some taxing challenges and promotes numerous inventive solutions. However, there is no clear and organized study about the advancement of HCI&A to date. To get a clear view of the state-of-the-art technologies, it is imperative to conduct a thorough study of the HCI&A and such study should range from evolution of the HCI concepts to the future direction of research. To fill-up such gap, in this study, we present an overall synopsis of healthcare informatics and analytics, underscoring its numerous challenges and opportunities. The key contribution of the paper is noted below:

We summarized key concepts of informatics which enables HCI&A to exists. In this regard, the evolution of HCI&A from its beginning to the present days is presented. The roles of the computer networks, social media, ecommerce and web technologies are investigated and summarized.

From an extensive study of the literature, we figure out the key technologies which enable the state-of-the-art HCI&A. The importance and uses of the key technologies in healthcare are figured out.

The state-of-the-art research outcomes are summarized. We divide the research efforts into two broad categories: HCI&A research from informatics and analytics perspective, and SCI& research from information systems perspective. Research efforts and results in each category are analyzed and summarized.

We also carry out bibliographic study of research paper published in different journals and conference to date on HCI&A. From the study, we figure out the trend of research in the field in academia and industries.

The final contribution is to present the state of the current education systems in academia to address the needs of HCI&A.

The overall structure of the paper is shown inFig.1. InSection2, the evolution of HCI&A from its genesis will be discussed in detail, which will be followed by a thorough description of the state-of-the-art HCI&A technologies inSection3. Then, the applications of HCI&A will be discussed from the Big data point of view inSection4. Different domains of current research on HCI&A in academia and industries are described inSection5. We then report a bibliographic study on health informatics and information system literature inSection6. Before concluding the paper inSection8, we present some challenges and prospects of HCI&A education and development inSection7.

2.HCI&A evaluation: key characteristics and capabilities

Since the birth of modern computing, around the 1950s, the terminformaticshas been used by researchers(Ledley etal., 1959). Healthcare informatics started to be a well-liked factor for practitioners since the late 1990s, and from then all healthcare entities such as patients, physicians, and administrators have regularly been enchanted by the capabilities that advanced tools and techniques might have in medicine and medical care services(Acampora etal., 2013). More recently, big data and big data analytic methods have been widely used in healthcare applications. Big data in healthcare also introduces some complicated issues, such as non-uniform data distribution and parallel processing with a large number of variables, which are inefficiently handled by existing analytical methods. Big healthcare data is devastating not only because of its volume but also the heterogeneous nature of data and speed at which it must be managed(Frost, 2015). In order to extract knowledge from big data, a healthcare system requires unconventional and mature data storage, management, analysis, and visualization tools and techniques. However, in this study, we use HCI&A as an integrated label and manage big data analytics as a new paradigm that enables new guidelines for HCI&A research.

2.1.HCI&A 1.0

In the context of data management and analytics, HCI&A began from the database techniques which are widely used in various healthcare settings from the 1970s(Brook etal., 1976). During that time, IBM designed a Patient Order Management and Communication System (POMCS) which was used by the Coral Gables Variety Children's Hospital in order to achieve three objectives: increase revenue, develop personnel productivity and cost savings, and improve patient care quality(Wiederhold, 1981b). These data management systems rely heavily on health data collection, extraction, and analysis technologies(Wiederhold, 1981a). As a data-centric perspective, HCI&A can be treated as HCI&A 1.0 where data is fully structured, homogeneous and often stored in relational database management systems (RDBMS).

Moreover, the artificial intelligence and data analytics of the medical domain originated from three other important dimensions - health-medicine, web, and data (seeFig.2). The concept of HCI&A 1.0 is that it is an umbrella term which mostly includes Web 1.0 technologies, Health 1.0 and Medicine 1.0 applications, services and tools. In Web 1.0, hospitals or healthcare organizations produce information without any interaction with patients; it is mainly one way of publishing content. In the healthcare context, the goal of Web 1.0 was to establish an online presence of service providers and make their information available to all customers (mainly patients) at any time. As HCI&A 1.0 includes Web 1.0 tools and techniques, its technologies include all core web protocols (i.e. HTML, HTTP), emerging web protocols (i.e. XML), and hypertext. HCI&A 1.0 technologies have some functional limitations: they are sometimes not able to forward participation between customers and service providers. Medicine 1.0 and Health 1.0 involve broad concepts which include mainly a provider centric approach. However, they define the combination of healthcare data management systems through the use of database technologies such as warehousing. HCI&A 1.0 also includes different statistical analytic tools, datamining tools and techniques for classification, segmentation, clustering, and analysis of health data. Different HCI&A tools have already been consolidated into the top commercial healthcare informatics platforms developed by IBM, Oracle, and Microsoft(Mead, 2006). In the context of software applications, HCI&A 1.0 includes some software packages for extraction, transformation and loading (ETL), online analytical processing (OLAP), database querying and reporting, data mining, and visualization. However, from data collection to knowledge discovery, there are several intelligence and analytical capabilities that must be contained in HCI&A 1.0.

2.2.HCI&A 2.0

In 2004, the term Web 2.0 was introduced to enable users to add information or content to the web, thereby allowing users to collaborate and interact using social media dialog. From that time, the internet has become increasingly popular and now forms a predominant part of our daily life(Oh etal., 2005). Web 1.0 was mostly unidirectional and so it allowed only direct interaction between providers and customers, whereas Web 2.0 allows the managing and assembling of large global crowds with common interests in social interaction(Lau etal., 2012). Moreover, Web 2.0 can deal with unstructured data resources -social media feeds, microblogs, weblogs, wikis, and other things that sometimes aren't stored in a database. Thus, in Web 2.0, patients can share their feelings and emotions on chat logs, online forums, and microblogs (Twitter), and these applications also support the design of cost saving e-healthcare solutions. When Web2.0 technologies are functionalized to the medical domain, the term health 2.0 is used to synthesize health and Web 2.0(Hughes etal., 2008), and Medicine 2.0 represents the fusion of medicine and Web 2.0(Van De Belt etal., 2010). However, according to the study in of Hughes B. etal. (2008), there are no potential differences between Health 2.0 and Medicine 2.0, and both of them are new concepts and are still developing(Van De Belt etal., 2010). Web 2.0 technologies, Health 2.0/Medicine 2.0, and advanced datamining technologies have ushered in a novel and thrilling era of HCI&A 2.0 research in the inaugurating period of social media, focused on data and network analytics for unstructured web content. With HCI&A 2.0 patients and professionals are enabled to collaborate with each other for developing collaborative healthcare. They transform their respective role/view using social networks where network content is heterogeneous. Moreover, in the healthcare context, RDBMS-based service information, comprehensive and IP-based exploration and reciprocal records that are accumulated smoothly through cookies and server records have become an innovative gold mine for understanding patients' demands and extracting new prospects of the health industry. HCI&A 2.0 applications can systematically assemble large volumes of timely reactions and feelings from both patients and professionals for different types of healthcare analytics.

According to the Gartner Hype cycle for healthcare (July, 2012), there are four major windows in the priority matrix for healthcare provider applications and systems. Several HCI&A 2.0 systems contain high value in the five to ten-year window(Shaffer, 2012). Moreover, Gartner healthcare platforms emphasize the application of intelligent agents, machine learning, neural networking, and text mining in order to improve quality, control operating costs, better engage patients, improve operational efficiency and meet the growing and challenging health services demand. New computer science and information system courses in artificial intelligence have emerged to address the required methodological exercise and training.

2.3.HCI&A 3.0

Web 3.0, also called the third generation of the worldwide web, promises to be a more mature and emerging web where improved 'trails' for information recovery will be designed, and an advanced scope for cognitive and intelligent processing of information will also be constructed(Lassila and &Hendler, 2007). Web 3.0 is likely to have a big effect on medicine and health(Giustini, 2007). Over the last five years, the new version of the Web Health 3.0/Medicine 3.0 has revolutionized healthcare systems through improving patient-professional communication, presenting context aware services, and offering intelligent and automated healthcare(Nash, 2008). The term HCI&A 3.0 represents the broader concept and is an umbrella term which includes all emerging concepts of Web 3.0, Medicine 3.0/Health 3.0, big data, mobile systems (e.g. m-health), and also smart systems. HCI&A 3.0 contains all mobile HCI&A technologies which are emerging and some are foreseen for the near future. Academic research on mobile HCI&A is still in its nascent phase. However, HCI&A 3.0 technologies can support a wide range of advanced applications and opportunities including mobile telemedicine, sensor-based monitoring, ubiquitous medical services, universal access to healthcare data, intensive monitoring, and lifestyle incentive management. The future HCI&A 3.0 systems will require the integration of mature and scalable techniques in big data analytics, social network analysis (SNA), and spatial-temporal analysis with existing datamining-based HCI&A 2.0 systems.

We have investigated the studies of the Gartner Hype Cycle for healthcare during 2005-2013, and summarized the key characteristics, challenges, and opportunities of HCI&A 1.0, 2.0, and 3.0. inTable1. The healthcare business community has started to take some important steps to adopt HCI&A for its needs. IS researchers are enjoying unique opportunities for the unprecedented capabilities of HCI&A, though they are also confronting some challenges due to different uncertainties associated with emerging healthcare informatics and analytics. However, as HCI&A is a sociotechnical matter and the IS discipline is also a multidisciplinary area, IS study needs to carefully evaluate research, future plans and guidelines, and curricula from HCI&A 1.0 to HCI&A 3.0.

3.HCI&A technologies and scopes

Recently, many emerging healthcare technologies have reshaped existing HCI&A research capabilities. In the healthcare context, researchers have shown that different intelligent and analytical tools have led to more effective process redesigns of medical services (i.e., diagnosis and medicine), in decision support system (e.g. scheduling), in management (e. g. supply chain), and in communication systems (e.g. global networking systems)(Yeow and Goh, 2015). Moreover, recently, the data deluge from multiple sources has caused a new era, "Big Data," to emerge that has soundlessly descended on every section of healthcare industries, from diagnosis and surgery departments to administration and management sections. In the healthcare context, an overwhelming amount of web-based, mobile, and sensor-generated data is arriving on a terabyte and even exabyte scale(Raghupathi and &Raghupathi, 2014). New drugs, diagnoses, and prognoses discovery, and insight can be attained from the highly detailed, contextualized, and rich content of relevant data (Groves etal., 2013).

However, over the last three decades, the development of healthcare technologies at both academic and industry level has reflected the inception, growth, and maturation of several tools and techniques in different technological streams (seeFig.2). After investigating the existing literature, we identified three major technical streams of healthcare informatics and analytics. They are software agent, machine learning, and cloud computing. Different tools and techniques under these streams are being developed over time to incorporate different definitions of healthcare problems that relate to information systems and technologies. The remaining part of this section describes each major technology with appropriate examples in the medical domain.

3.1.Software agent

In the healthcare domain, software agents are one of the most exciting research paradigms for developing software applications and they must be able to perceive the physical and virtual world around it using different sensing devices(Wooldridge, 2009). Moreover, agent-based computing systems in healthcare have been addressed as 'the next significant revolution in healthcare service development through software'(Isern etal., 2007, September)and 'the emerging software innovation to shift health paradigm to smart healthcare system' (Yuan etal., 2005). From the 1990s there has been a growing interest in the research and practice on agent-based application in healthcare (Cortes etal., July 2002)(Shankararaman, June 2000), and recently, some significant research works are also beginning to appear which reap big data benefits in the healthcare industry through multi-agent systems (MASs)4(Felisberto etal., 2015)(Twardowski and Ryzko, 2014, August).

Moreover, intelligent agents have already been deployed in different tasks such as health information retrieval from large volume data sources(Abasolo and Gomez, 2000, September);(Baujard etal., 1998), decision support systems for diagnosis and care(Baujard etal., 1998), patient, doctor, and nurse scheduling(Rossetti etal., 1999), co-operation among different medical components (e.g., diagnosis and medicine) to manage pervasive healthcare(Su etal., 2011, September);),(Chan etal., 2008)the development of education, training, and services(Zaharakis etal., 1998), and medical information sharing(Mohan etal., 2009). Over the last decade, a number of software agent tools have been included in healthcare research and practices. For instance, the National Electronic Library for Health (NeLH) (Nealon and Moreno, 2003), context-aware hospital information system (CHIS) (Tentori etal., 2006), and Virtual electronic patient record (VEPR) (Cruz-Correia etal., 2005) are the multi-agent tools for healthcare management; CARREL (Cabanillas etal., 2003), CARREL+ (Tolchinsky etal., 2006), Medical information agents (MIA) (Vermeulen etal., 2009;(Braun etal., 2007), and healthcare services (HeCaSe2) (Isern etal., 2007, September) are the multi-agent tools that support the making of appropriate medical decisions; K4Care and Aingeru are the agent tools for pervasive healthcare; and SHARE-IT (Corts etal., 2007;Walliser etal., 2008), Geriatric Ambient Intelligence (GerAmI) (Corchado etal., 2008) are examples of integrated system agent tools.

3.2.Machine learning

In recent years, a good number of research works have been appearing in the biomedical engineering and artificial intelligent literature, which describe the application of machine learning (ML) techniques to design classifiers for anomalies (diseases and viruses) detection or medical diagnosis. Moreover, there has been a dramatic increase in the application of machine learning methods, tools, and techniques that can help solve diagnostic and prognostic complications in advanced healthcare systems. Machine learning is being used to increase the performance of clinical parameters and their combinations of the variety of medical prognoses such as prediction of disease progression, decision support for therapy or surgery, knowledge extraction from emerging research and practice, and overall healthcare system management (Magoulas and Prentza, 2001). The machine learning approach is also being used to analyze heterogeneous health data (e.g., X-ray reports, ECG reports, tomography reports, temperature, pulse, and blood pressure reports). In existing literature, we see the intellectual growth of machine learning in medical applications via several different research studies, and through investigating literature we also identify four major ML application streams in healthcare: ML in diagnosis (Kononenko, 2001), ML in prognosis and treatment (Frunza etal., 2011;Kukar etal., 1996), ML in drug discovery (Burbidge etal., 2001), and ML in surgery (Lanfranco etal., 2004).

Medical diagnostic reasoning is the most remarkable application area of intelligent systems. Therefore, currently different ML tools and techniques are well suited to analyzing medical diagnosis in specialized diagnostic problems (Kononenko, 2001). Learning-Classification-Propagation (LCP) (Zhou etal., 2008, September), expert system for thyroid disease diagnosis (ESTDD) (Kele & Kele, 2008) are examples of ML-based diagnosis tools. Machine learning techniques allow computers to learn from the past examples, to extract hard-to-discern patterns from high volume heterogeneous noisy datasets, and to predict intended anomalies or diseases using different statistical, probabilistic, and optimization techniques. PredictSNP (Bendl etal., 2014), CanPredict (Kaminker etal., 2007) are two well-known tools for prediction disease-related mutation. Deep learning is also used in estimating prognosis and guide therapy in patients with pulmonary hypertension and adult congenital heart diseases (Diller etal., 2019). The estimation accuracy is more than 90%. In case of drug discovery, IBM has introduced its own health ML applications in drug discovery since its early days.5Recently, Google has also entered into the drug discovery challenge and has started to work as a host company that raises and makes economic value by working on medicine inventions with the help of ML tools and techniques.6GPCRpred is another same type ML-based drug discovery tool that can predict families and subfamilies of GPCRs(G-protein coupled receptors) from the dipeptide composition of proteins (Bhasin and Raghava, 2004).

3.3.Cloud computing

In the big data era, healthcare systems have almost become a data-centric and collaborative endeavor (Almashaqbeh etal., 2014). The application of health data with different text mining or data mining algorithms entails a growing demand for dynamic, scalable resources (Koh and Tan, 2011;Raja etal., 2008). In order to present the proper communication between different resources, cloud computing is a new, fashionable and fast rising area of development in the healthcare context (Mell and &Grance, 2011). Moreover, in the healthcare framework, permanent infrastructure investments are hard to evaluate, and adjustable on-demand services are required to meet dynamic demands (Griebel etal., 2015). Therefore, these resources are employed temporarily which can produce some viable solutions to fulfill such type of requirements. By adopting the cloud model, all technical processes of a remote organization will be migrated to the vendor's cloud-computing infrastructure where all the manipulations will be made and preserved (Armbrust etal., 2009). Recently some vendor organizations like IBM and Microsoft enable other organizations to access their (vendor) expensive infrastructure (e.g. hardware, software, and experts) through using a new "pay-as-you-go" model (Armbrust etal., 2010). Therefore, adopting cloud health organization can minimize infrastructure setup costs because the cloud computing providers will shoulder it.

In the cloud computing environment, preserving confidential sensitive healthcare data and compliance with key regulations such as the HIPAA is a matter of great concern, which may also cause a regulatory backlash and hinder further organizational innovation (Kuo, 2011). By using different privacy preserving methods and security tools, these security and privacy risks can easily be controlled and then healthcare organizations can certainly take advantage of cloud computing solutions and gain numerous benefits such as helping to improve the quality of care and minimizing overall healthcare expenditures (Muir, 2011;Wang and Tan, 2010, October).

In order to meet the healthcare demands, several third-party vendors have started to design cloud computing tools and different healthcare organizations are embedded in these vendors in customizing these tools with suitable security components to enhance their business. Cloudkick,7LogicMonitor (Logic Monitor 2012), and Pandora FMS (Pandora 2011) are examples of cloud computing tools, which are designed for healthcare monitoring. Moreover, by using the cloud, healthcare organizations are collaborating with their partners to show better healthcare services (Kabachinski, 2011). For instance, Microsoft HealthVault8is cloud-based platform where Microsoft collaborated with Kaiser Permanente, and Google Health uses a cloud named "gcloud"9to obtain health information from Cleveland Clinic's MyChart program (Kabachinski, 2011).

4.HCI&A applications: big data context

Healthcare Informatics and Analytics applications are growing in different dimensions in the health arena; they are now being actively diffused via systems such as electronic medical records (EMR), personal health records (PMR), clinical decision support systems (CDSS), and computerized provider order entry (CPOE) (Romanow etal., 2012). HCI&A is elaborately used and also provides leverage for different prospects generated by the large volume of medical data and multidimensional analytics in the medical domain required in various potential and high-influence application areas (Avison and Young, 2007;Goldschmidt, 2005). Four potential and high-impact applications of HCI&A are - (1) Health insurance and cost (2) Health administration and policy (3) Smart Healthcare and services, and (4) Security and privacy. In the healthcare context, different subdomains use different datasets and analytical approaches, where researchers and practitioners have to adopt or develop appropriate systems to generate the targeted results.

4.1.Health insurance and cost

Currently big data analytics and different health ITs are broadly used by insurers, personal or other healthcare funders to identify cost overruns that might constitute anomalies (Srinivasan and Arunasalam, 2013). Insurance and payment data is collected from customer claims and hospital discharges data including client opinions and behavioral data using the web. Big data enabled technologies (i.e., machine learning) allow us to consider interconnected claims and payments, whereas previously, systems tended to focus on each claim individually, and could not formally make judgments based on clues, which could possibly abuse, waste, fraud and errors (Groves etal., 2013). Therefore, advanced HealthITs present not just effective analytics but also relational explanations for actions that help facilitate transparent solutions. Though designing effective analytics is a challenging matter because of data complexities and enormous human factors, big data and its technologies could carry significant financial and healthcare benefits over large populations and timeframes. Big data tools and techniques are enabling private health insurance funds to recover hidden cost overruns, and are ushering in a new era of high-quality patient care at lower cost (Bates etal., 2014). For instance, in order to make effective decisions, predictive models of healthcare insurance in Australia use three levels of analytics- Admission-Level Analytics, Aggregated-level analytics, and Contract-level analytics. In the big data environment, these analytics extract anomalous admissions, and compare the performance of providers with respect to cost effectiveness and quality of care anomalies (Srinivasan and Arunasalam, 2013).

4.2.Health administration and policy

The proliferation of "smart" devices and the growth of electronic communication have generated much excitement for designing new healthcare setups and policies. As the current healthcare standard (Health/Medicine- 2.0,3.0) allows participatory, online, and rich multimedia, there is a great opportunity for adopting HCI&A applications in healthcare management and policy design. Healthcare administration and policy are related to four important factors- healthcare management, planning, resource allocation and decision-making (James etal., 2011). They will be imperative for health administration and policymakers to figure out how to overcome healthcare inconsistencies. In the big data epoch, an enormous amount of health data is being generated to facilitate organizations and persons to innovate and grow. Through converting these vast resources (data) into information the art of healthcare management system is advancing to achieve the objectives. For instance, the National Electronic Library for Health (NeLH) is a data resource that offers a portal to extract evidence-based health data on the internet (Nealon and Moreno, 2003). In order to provide effective and efficient healthcare administration (e.g., automatic resource planning, efficient scheduling among all entities, and accurate real-time medical decisions), the health industry should deploy some big data-enabled intelligent tools and techniques. Moreover, these technologies reduce overall healthcare expenditure, and according to the Mckinsey reports, such technologies can support the unlocking of more than $300 billion a year in additional costs throughout the US healthcare sector (James etal., 2011).

4.3.Smart care and services

Smart health integrates ideas from ubiquitous computing and ambient intelligence applied to predictive, personalized, preventive and participatory healthcare systems (Rcker etal., 2014). Smart health is strongly connected to the concepts of wellness and wellbeing (Suryadevara and Mukhopadhyay, 2014) and includes a large volume of data, collected by large amounts of biomedical sensors, (e.g., temperature, heart rate, blood pressure, and breathing rate), genomic-driven big data (genotyping, gene expression, and sequencing data), payer-provider big data (electronic health records, insurance records, and pharmacy prescriptions), and social media data (patients' status, feedback, and responses) actuators, to observe and predict patients' physical and mental conditions. Smart health is a nascent but promising field of study at the intersection of medical informatics, public health, and also business, alluding to intelligent healthcare services or enhanced cognitive capabilities through the IoT (internet of things). Big data applications in healthcare organizations can provide significant benefits which include detecting diseases at an early stage when they can be prescribed more easily and effectively. The major initiatives of the National Science Foundation (NSF) related to big health data analytics is the NSF Smart Health and Wellbeing (SHB)10program. The main goal of the SHB program is to address ICT issues in the big data context that support a much-needed revolution in healthcare from being reactive and hospital-centered to proactive and patient-centered, and accentuate wellbeing rather than disease control (Chen etal., 2012b).

In smart systems, healthcare workers are enabled to review and update a patient's medical data from every positional setting using handheld devices (Arnrich etal., 2010 a;Arnrich etal., 2010 b). Besides, HCI&A under integrated systems is used in the synchronization of actions that have to be performed to provide smart (autonomous, interactive, and intelligent) healthcare to citizens. These intelligent systems integrate different AI techniques with a specific purpose among different purposes (i.e. diagnoses, treatments, therapy, and surgery) under the umbrella of e-health.

4.4.Healthcare security and safety

Prognosticating threatening events and measuring healthcare security risk in real-time are highly needed and also very challenging issues in the burgeoning healthcare industry. Due to the inclusion of different advanced technological approaches, the healthcare industry is being revolutionized through presenting smart healthcare services (Agrawal etal., 2007), but otherwise industry is encountering a deluge of avant-garde attacks ranging from Distributed Denial of Service (DDoS) to secret disrupting software (Patil and &Seshadri, 2014, June;Pramanik etal., 2017a). Moreover, according to the Institute of Medicine report, security and safety issues have received a lot of attention in the US since 2000 (Institute of Medicine 2000). That report presented a real panorama of health security and safety problems in details; it also provides some guidelines on how an organization should be reformed to improve the security system in regard to healthcare. Since that report, US healthcare industries have been uninterruptedly struggling to develop coherent programs to refine security and safety systems, and also these programs have been examined substantially. In recent years, using heterogeneous data (i.e. financial and spatial data) from the advanced healthcare monitoring systems, HCI&A will steer new directions in real-time security intelligence, and using the analytical outcomes, healthcare providers can take preemptive measures before affecting the healthcare systems (Bates etal., 2014;Meingast etal., 2006, August).Table2summarizes the rising HCI&A applications, the nature of data, analytical approaches, and contributions in different sections of healthcare implementations.