Towards a Comprehensive Data Analytics Framework for Smart Healthcare Services
Introduction
In the last decade, we have been witnessing a continuous increase in lifestyle-related illnesses as a result of various factors such lack of exercise, poor diet, pollution and the addictive habit of smoking. For example, a recent study has estimated that 9% of deaths across the world in 2008 were caused by physical inactivity and merely improving the physical activity levels can affect the life expectancy of the world's population by an increase of 0.68 years [23]. In addition, current guidelines for a healthy life are recommending that adults should spend about 150 minutes of their week on physical activity. However, the study reported that 1/3 of adults are not performing sufficient physical activity which leads to increasing risk of the adults developing certain diseases such as heart-related diseases and diabetes. Our diets have also profoundly differed in the last decades. For example, the consumption of fast food and processed food has been rising continuously across the world resulting in the growing intake of salt, fat, simple sugars and sweeteners [41]. In addition, there has been major increases in the consumption of meat and a corresponding reduction in the consumption of vegetables, whole-grain foods and non-citrus fruits. All of such updates together lead to major increases in the number of calories being consumed, increasing the obesity levels and resulting in severe health threats. Therefore, diseases such as cardiovascular diseases, cancer and diabetes have become the major cause of death across the world.
In general, the medical advancements, improved handling of communicable diseases, and better diet have increased peoples's life expectancy. For example, some reports estimated that life expectancy has increase by 13 years during the course of the 21st century.1 The UN expects that, life expectancy is going to increase from 68 years in 2005–2010 to 81 years in 2095–2100.2 Therefore, global aging and its associated effect on the performance of health services are acknowledged as an increasing phenomenon over the last decades. Many countries have been affected by the challenge of an aging population where older individuals represent a bigger share of the size of their population. In particular, the number of people with an age of 65 years old or older is expected to increase from an estimation of 524 million in 2010 to about 1.5 billion in 2050 [13]. Such demographic changes are consequently leading to rising demands for healthcare services and higher government expenditures because it is commonly expected that elder individual are naturally more vulnerable to health problems and chronic diseases. Such increasing expenses on healthcare services represent a crucial challenge for almost every government.
In principle, the economics of healthcare systems is getting a lot of attention because of the current dynamics of global demographics. In particular, the spending on healthcare services is commonly a high priority of the internal political agenda and discussions in almost all countries. It is expected that the healthcare costs will account for 20–30% of GDP in some countries by 2050 [17], [21], a percentage that is financially unsustainable. In principle, costs continue to increase leading to a consequent need to change the focus of the approaches of healthcare services from that of a reactive model to a model that utilizes predictive healthcare mechanisms. Governments consider smarter healthcare as an effective way of improving quality while minimizing service cost. Establishing such models requires monitoring and diagnosing various data sources for the sake of being able to achieve accurate and effective predictions. Additionally, applications for providing the support of delivering in-home exercise programs for improving the balance and strength of older adults can serve as a preventative action for health problems such as falls. Other applications include activity monitoring, fitness measurements via important signs for monitoring and calorie-intake tracking where various product offerings overlap with main public health advices on activity, exercise, managing one's diet, and observing for early indications of health-related problems. Such indications will be significantly amplified as governments suffer from increasing healthcare expenses in the long run, increasing the opportunities for incorporating sensing-based healthcare platforms and solution.
In principle, the above challenges confirm that the current reactive models of healthcare systems have become unsustainable. Therefore, there have been increasing calls for various major changes in the mechanisms of providing the healthcare services. For example, the healthcare services need to be predictive and proactive to limit the occurrence of expensive acute health episodes [37]. In addition, the healthcare services need to be individualized, rather than population-based in order to guarantee the delivery of the right treatment. Furthermore, the delivery process of care services need to be decentralized from hospitals to the community and the home. In practice, Information and Communication Technologies (ICT) can play a main role in achieving all of these goals and can represent an effective solution to deliver manageable models of patient services in home and community locations. For example, sensing technology [41] can play a main role on monitoring the main health status indicators of an individual directly or indirectly via ambient monitoring of day-to-day patterns where in-house healthcare has become a main component of the Internet-of-Things (IOT) [30]. In addition, wearable sensing technology [49] is designed to monitor an individual's vital signs at all times, i.e 24/7, where alerts can be communicated to a medical staff/caregiver in the case that a certain limit is reached or in the case an abnormal event such as the collapse of a patient gets observed. Big data analytics services [57] can monitor and detect vital signs and other various measurements which can be provided to physicians or healthcare providers to support them in the diagnosis process. In addition, this diagnosis can be even automated so that it can minimize or remove the need to visit the physician for simple illnesses such as flu and other more common illnesses.
This article focuses on analyzing how the recent advancements of ICT can be effectively exploited and integrated for tackling the above mentioned challenges and contribute towards the state-of-the-art of healthcare services. In particular, we focus on exploiting the advancement in the areas of sensor technologies, cloud computing, Internet-Of-Things and big data analytics systems as emerging technologies, which are made possible by the remarkable progress in various complementary advancements including network communication speed, computational capabilities and data storage capacities that provide various advantages and characteristics that can contribute towards improving the efficiency and effectiveness of healthcare services.
The main contribution of this paper is two-fold; Firstly, analyzing the state-of-the-art in the aforementioned key enabling areas and technologies, and identifying the challenges and gaps for the realization of an integrated and comprehensive smart healthcare data analytics solution. Secondly, drawing on the findings of this analytical study, we propose an integrated and comprehensive framework for big data analytics services in smart healthcare networks, SmartHealth, which addresses the revealed challenges and fills in the identified gaps. The Framework acts as a roadmap for the research in the area of big data analytics in smart healthcare applications.
The rest of this paper is organized as follows: the analytical study starts in Section 2 by presenting and analyzing related work and open challenges in the area of smart healthcare systems, which reveals a set of enabling technologies towards the realization of an integrated and comprehensive data analytics smart healthcare solution. We then dedicated separate Sections to discuss and analyze each of these enabling technologies. In particular, sensing technologies is explicated in Section 3, cloud computing and its application to the domain of healthcare is discussed in Section 4. This is followed by a discussion about Big Data storage and processing systems in Section 5. Based on this comprehensive analytical study, an integrated and comprehensive framework for big data analytics in healthcare is derived in Section 6, which represents the second main contribution of this paper. Sample application scenarios and use cases of the proposed framework is presented in Section 7. The paper is then concluded in Section 8, highlighting future work directions.
Section snippets
Related work and open challenges
The term Cyber Physical System represents the umbrella term that integrates and exploits recent ICT advancement in sensing computing, cloud computing, Internet-Of-Things, and big data storage and analytics, with numerous applications in various domains, such as healthcare, manufacturing, traffic, logistics, etc. A cyber-physical system (CPS) is a system of collaborating computational elements controlling physical entities [58]. It is a new revolution in sensing, computing and communication that
Sensing technologies
Sensing is a pervasively used technology in nearly every aspect of hospital-based service starting from the simplest digital thermometer to complex laser-guided surgical tools [41]. For example, imaging sensing technologies (e.g., magnetic, X-ray), positron emission tomography (PET), computed tomography (CT) and ultrasound are commonly used technologies for providing the medical staff with several insights into the health status of every patient. These sensors have played a crucial role in
Cloud computing
Cloud computing has been acknowledged on the top of Gartner's list of the ten most disruptive technologies of the next years [1]. It represents a paradigm shift in the field of ICT which has been already emerging to change the ways of how businesses deal with their storage and computing resources [18]. In principle, cloud computing represents an emerging paradigm for the process of provisioning computing resources and infrastructure. This paradigm shifts the location of the computing resources
Big data storage and processing systems
In general, using the mining metaphor, data represent the new gold where analytics systems represent the machinery that mines, molds and mints it. In practice, healthcare systems across the world are facing the challenge of information overload in caring for patients. Healthcare analytics is defined as a set of computer-based methods, processes and workflows for transforming raw health data into meaningful insights, new discovery and knowledge that helps in making more effective healthcare
SmartHealth framework
In general, due to recent advancements in sensor devices and other related technologies, the cost of data acquisition has reduced dramatically. In principle, while the initial setup costs are relatively high, the continuous data acquisition remains very cheap. In addition, these initial costs are continuously going down with the continuous advancements in sensor technologies. In practice, sensing-based patient monitoring generates much more data than healthcare professionals are able to
Use cases and application scenarios
In general, the volume of healthcare data is expected to continue growing dramatically in the years ahead. In practice, utilizing the recent advancements on ICT to effectively analyze and utilize such big data can bring about significant benefits for healthcare organizations ranging from single-physician offices and multi-provider groups to large hospital networks in several use cases and application scenarios. In particular, healthcare analytics can be leveraged in several applications with
Conclusion and future work
In this article, we analyzed how the recent advancements of ICT can be effectively exploited and integrated for tackling the above mentioned challenges and contribute towards the state-of-the-art of healthcare services. In particular, we focused on exploiting the advancements in the areas of sensor technologies, cloud computing, Internet-Of-Things and Big data analytics systems as emerging technologies that can significantly contribute towards improving the efficiency and effectiveness of
References (76)
- et al.
The internet of things: a survey
Comput. Netw.
(2010) - et al.
Pathways: a school-based, randomized controlled trial for the prevention of obesity in American Indian schoolchildren
Am. J. Clin. Nutr.
(2003) - et al.
Effect of physical inactivity on major non-communicable diseases worldwide: an analysis of burden of disease and life expectancy
Lancet
(July 2012) - et al.
From wireless sensor networks towards cyber physical systems
Pervasive Mob. Comput.
(2011) - et al.
Design and evaluation of a portable optical-based biosensor for testing whole blood prothrombin time
Talanta
(2013) Gartner top ten disruptive technologies for 2008 to 2012: emerging trends and technologies roadshow
(2008)- et al.
An event-driven sensor virtualization approach for internet of things cloud
- et al.
A survey on sensor-cloud: architecture, applications, and approaches
Int. J. Distrib. Sens. Netw.
(2013) - et al.
Accurate measurement of concurrent flexion–extension and internal–external rotations in smart knee prostheses
IEEE Trans. Biomed. Eng.
(2013) - et al.
Above the clouds: a Berkeley view of cloud computing
(2009)
Smart checklists for human-intensive medical systems
The meaningful use of big data: four perspectives–four challenges
ACM SIGMOD Rec.
Digital disease detection, harnessing the web for public health surveillance
N. Engl. J. Med.
Health Analytics: Gaining the Insights to Transform Health Care
Ageing and gait variability, a population-based study of older people
Age Ageing
Google trends: a web-based tool for real-time surveillance of disease outbreaks
Clin. Infect. Dis.
Mapreduce: simplified data processing on large clusters
Commun. ACM
Medical cyber physical systems and bigdata platforms
Health care costs: a market-based view
McKinsey Q.
Cloud Computing: Concepts, Technology, & Architecture
Multi-sensor context-awareness in mobile devices and smart artifacts
Mob. Netw. Appl.
Citizens as sensors: the world of volunteered geography
GeoJournal
Review of cyber-physical system in healthcare
Int. J. Distrib. Sens. Netw.
Challenges and research directions in medical cyber physical systems
Proc. IEEE
Rehospitalizations among patients in the medicare fee-for-service program
N. Engl. J. Med.
Mobile ECG measurement and analysis system using mobile phone as the base station
Adaptive real-time data management for cyber-physical systems
The Big-Data Revolution in us Health Care: Accelerating Value and Innovation
Continuous monitoring of vital constants for mobile users: the MobiHealth approach
Internet of things
MLbase: a distributed machine-learning system
Managing deadline miss ratio and sensor data freshness in real-time databases
IEEE Trans. Knowl. Data Eng.
Cyber physical systems: design challenges
Challenges and research directions in medical cyber-physical systems
Proc. IEEE
Review of cyber-physical system architecture
Pregel: a system for large-scale graph processing
Cited by (131)
Machine intelligence and medical cyber-physical system architectures for smart healthcare: Taxonomy, challenges, opportunities, and possible solutions
2023, Artificial Intelligence in MedicineOf digital transformation in the healthcare (systematic review of the current state of the literature)
2024, Health and Technology