See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/324008399

Concurrence of Big Data Analytics and Healthcare: A Systematic Review

Article  in  International Journal of Medical Informatics · June 2018

DOI: 10.1016/j.ijmedinf.2018.03.013

CITATIONS READS

9 2,662

2 authors:

Nishita Mehta Anil Pandit

Symbiosis International University Chellaram Diabetes Institute
2 PUBLICATIONS   9 CITATIONS    34 PUBLICATIONS   14 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Hospital & Health care Management View project

All content following this page was uploaded by Nishita Mehta on 29 March 2018.

The user has requested enhancement of the downloaded file.

 International Journal of Medical Informatics 114 (2018) 57–65

Contents lists available at ScienceDirect

International Journal of Medical Informatics

journal homepage: www.elsevier.com/locate/ijmedinf

Review article

Concurrence of big data analytics and healthcare: A systematic review T

a,⁎ b
Nishita Mehta , Anil Pandit
a
Symbiosis International University, Pune, India
b
Symbiosis Institute of Health Sciences, Pune, India

A R T I C LE I N FO A B S T R A C T

Keywords: Background: The application of Big Data analytics in healthcare has immense potential for improving the quality
Big data of care, reducing waste and error, and reducing the cost of care.
Analytics Purpose: This systematic review of literature aims to determine the scope of Big Data analytics in healthcare
Healthcare including its applications and challenges in its adoption in healthcare. It also intends to identify the strategies to
Predictive analytics
overcome the challenges.
Evidence-based medicine
Data sources: A systematic search of the articles was carried out on five major scientific databases: ScienceDirect,
PubMed, Emerald, IEEE Xplore and Taylor & Francis. The articles on Big Data analytics in healthcare published
in English language literature from January 2013 to January 2018 were considered.
Study selection: Descriptive articles and usability studies of Big Data analytics in healthcare and medicine were
selected.
Data extraction: Two reviewers independently extracted information on definitions of Big Data analytics; sources
and applications of Big Data analytics in healthcare; challenges and strategies to overcome the challenges in
healthcare.
Results: A total of 58 articles were selected as per the inclusion criteria and analyzed. The analyses of these
articles found that: (1) researchers lack consensus about the operational definition of Big Data in healthcare; (2)
Big Data in healthcare comes from the internal sources within the hospitals or clinics as well external sources
including government, laboratories, pharma companies, data aggregators, medical journals etc.; (3) natural
language processing (NLP) is most widely used Big Data analytical technique for healthcare and most of the
processing tools used for analytics are based on Hadoop; (4) Big Data analytics finds its application for clinical
decision support; optimization of clinical operations and reduction of cost of care (5) major challenge in
adoption of Big Data analytics is non-availability of evidence of its practical benefits in healthcare.
Conclusion: This review study unveils that there is a paucity of information on evidence of real-world use of Big
Data analytics in healthcare. This is because, the usability studies have considered only qualitative approach
which describes potential benefits but does not take into account the quantitative study. Also, majority of the
studies were from developed countries which brings out the need for promotion of research on Healthcare Big
Data analytics in developing countries.

1. Introduction delivery, rather than a central asset source for competitive advantages”
[1]. As the electronic health data remains largely underutilized and
Over the last decade, there has been a rapid digitalization across the hence wasted [2], there is a need for converting the raw data into
industries. Healthcare has also undergone this digital transformation meaningful and actionable information [3,4].
with an increase in use of Electronic Medical Records (EMRs); Much of the highly valuable healthcare data is in unstructured or
Healthcare Information Systems (HIS); and handheld, wearable and semi-structured form. Added to it, the complex, dynamic and hetero-
smart devices. As a result, a massive amount and variety of health-re- geneous characteristics of the data [5–7] renders it difficult to extract
lated data today, is in digital form, which includes −omics data, socio- useful information using traditional data analytical tools & techniques
demographics data and insurance claims data apart from clinical data. [8]. In fact, there is a finite human ability to process this data without
This high-quality healthcare data offers potential value for optimizing effective decision support [9]. This creates the need for integration of
care delivery, but it is still “perceived as a by-product of healthcare Big Data analytics into healthcare. Big Data analytics has the ability to

⁎
Corresponding author.
E-mail address: [email protected] (N. Mehta).

https://doi.org/10.1016/j.ijmedinf.2018.03.013
Received 22 January 2018; Accepted 23 March 2018
1386-5056/ © 2018 Elsevier B.V. All rights reserved.
N. Mehta, A. Pandit International Journal of Medical Informatics 114 (2018) 57–65

analyze a wide variety of complex data and generate valuable insights

which otherwise would not have been possible. When applied to the
healthcare data, it has the potential to identify patterns and lead to
improved healthcare quality & reduced costs and enable timely deci-
sion-making [6,8,10,11]. As per the report by McKinsey Global Institute
[12], by utilizing Big Data effectively, US Healthcare can create a value
of more than $300 billion every year, of which two-third would be in
the form of reducing healthcare expenditure by about 8%. Using Big
Data technology, hidden knowledge can be uncovered using automated
analysis of outcomes [13].
Advancement in cloud computing and increased deployment of
EMRs enable easy access to longitudinal patient data [14]. The in-
tegration of longitudinal patient data with data from disparate, struc-
tured and unstructured Big Data sources offer the potential of com-
prehensive understanding of diseases at a considerably higher pace
[15,16]. The ability of Big Data analytics to identify disease hetero-
geneity allows quick and accurate diagnosis and assessment of therapies
[3,17–19]. By linking data from different sources and discerning pat-
terns, the predictive power of Big Data analytics can also be used for
transforming continuous real-time data into valuable information. This
is of utmost importance in emergency medical situations as it can mean
the difference between life and death [10].
The promising value of Big Data technology in healthcare has cre-
ated an increasing interest of academic & industry investigators.
Nevertheless, there have been only a few literature reviews and the
literature remains largely fragmented. The purpose of this research
therefore is to gain a comprehensive understanding of current outlook
on this technology. It aims at answering the research question on: How
“Big Data analytics” fits in healthcare environment to enhance its value?
Accordingly, this review explores the conceptual aspect of applying Big
Data analytics to healthcare and its significance in enhancing care de-
livery and business worth. It also describes the challenges posed in
leveraging Big Data analytics in healthcare and the need for approaches
to overcome them.
Fig. 1. Research Process.

2. Review method
IC 2: Articles published between 2013 and 2018
A systematic review was conducted for capturing relevant literature IC 3: Only articles published in English
from different sources, focusing on the following objectives: To capture the literature relevant to the research interest, the arti-
cles with their primary emphasis on traditional analytics in healthcare

• To determine different perspectives to definition and concepts of Big were excluded.

Data in healthcare
• To explore the sources of Big Health Data
• To identify Big Data analytical techniques and technologies in 2.3. Study selection
healthcare
• To illustrate the potential benefits and applications of Big Data The procedure for search and selection of research material was
carried out in the following four phases (Fig. 1):
within healthcare
• To present strategies for tackling the challenges of Big Data appli- 1. The search for publications on electronic databases containing
cation within healthcare
keywords “big data” or “big data analytics”, and “healthcare” or
By investigating these objectives in detail, this review will make a “medicine” or “biomedicine”
significant contribution in understanding the overall context and the 2. Scrutiny of the title, abstract and keywords of identified articles and
future application of Big Data techniques in the healthcare domain. selection of the significant articles on the basis of selection criteria
3. Perusal of articles that were not eliminated in the previous phase for
2.1. Information sources the review
4. Scanning of cross-reference articles for detailed study
A search for articles was made on following databases:
ScienceDirect, PubMed, Emerald, IEEE Xplore and Taylor & Francis.
The references included in these articles were also scanned for a thor- 2.4. Quality assessment
ough review.
During the review, activities ensuring the quality of the search
2.2. Selection criteria process were undertaken. The web searches were made in incognito
mode to avoid any influence of historical searches. From initial sear-
To select the literature for inclusion in the literature review, fol- ches, the authors manually extracted relevant papers and articles. The
lowing inclusion criteria were used: analysis and evaluation of abstracts was carried out and the authors
IC 1: Articles that deal with Big Data analytics in healthcare verified which articles were to be included or excluded from the study.

58
N. Mehta, A. Pandit International Journal of Medical Informatics 114 (2018) 57–65

3. Results genotype, phenotype, genomics data, ICD codes [3,7,26]; but the un-
structured data includes memos, clinical notes, prescriptions, medical
The study selection process was followed for each of the databases. imaging, EHRs, lifestyle, environmental, and health economics data
In ScienceDirect, Emerald and Taylor & Francis, the initial searches [3,7,26,27]. The challenge for Big Data analytics is to deal with this
yielded a larger number of hits (3324, 3167 and 5816). While the heterogeneous data in order to generate insights for improved health-
search resulted in 23 articles on PubMed and 60 articles on IEEE Xplore. care outcomes.
After the examination of title, a total of 122 articles were selected as per For defining Big Data in healthcare, Auffray et al. [26] focused on
the selection criteria. The abstracts and keywords of these articles were the types of healthcare data, while authors like Raghupathi & Raghu-
scrutinized and 87 relevant articles were chosen for further study. The pathi [10], Karen et al. [28], Tan et al. [29] emphasized on the re-
full text of these 87 articles was read. Articles discussing technical as- quirement of analytical and management tools. According to Liyanage
pects of Big Data computing and statistical models for Big Data analy- et al. [30], a quantitative definition of Big Data is difficult because the
tics in healthcare were beyond the scope of this review and were thus volume aspect of Big Data is relative to the time of definition and would
excluded. Of the 87 articles screened, a total of 58 articles were in- change with the advancement in technologies. Hansen et al. [25] and
cluded in this review. Roski et al. [31] concentrate on its analytical ability. On the other hand,
The literature included in this review comprises mainly of de- Bates et al. [18], Dinov [25] and Bian et al. [32] focuses on the char-
scriptive articles and usability studies. On the basis of the main research acteristics of Big Data. These characteristics are defined as 5Vs: volume,
objectives, the content from these articles was extracted and the articles velocity, variety, veracity and value.
were organized into different groups: Big Data analytics definition and The volume refers to the quantity of Big Data in healthcare, which is
concepts, sources of Big Data in healthcare, Big Data techniques for estimated to increase dramatically to 35 zettabytes by 2020 [33]. The
healthcare analytics, application and potential benefits of Big Data in variety refers to the different types of healthcare Big Data collected
healthcare and challenges in Big Data analytics in healthcare. The fol- including their heterogeneous characteristics [8] and structured and
lowing section summarizes the findings in each of these categories. unstructured nature of medical data [7]. The velocity is the speed of
data generation (i.e. real-time patient data [7]) as well as data collec-
3.1. Concept of big data and its definitional perspectives tion. The veracity refers to sources that influence accuracy such as in-
consistencies, missing data, ambiguities, deception, fraud, duplication,
The concept of Big Data was framed in late 1990s when Michael Cox spam and latency [34]. Veracity and data quality issues are of acute
and David Ellsworth [20] considered visualization as a problem of Big concern in healthcare because life or death decisions depend on having
Data. But one of the early definitions of Big Data was given by Francis the accurate information [10]. Lastly, the value represents cost-benefit
X. Diebold [21] in 2000 when he referred to Big Data as “explosion in to the decision maker through the ability to take meaningful action
the quantity (and sometimes, quality) of available and potentially re- based on insights derived from data [34]. These characteristics of
levant data”. Later the key dimensions of Big Data – volume, velocity and medical data contribute to their complexity: diversity of health-related
variety: the 3Vs – were derived from the study of Doug Laney in 2001 ailments and their co-morbidities; the heterogeneity of treatments and
[22]. Manyika et al. [12], in their report, regarded value as another outcomes [21]; differences in clinical workflows, practice standards,
important element of Big Data. With regards to Big Data in healthcare, patient populations, available technologies, and referral resources [35].
Feldman et al. [23] introduced veracity as yet another critical feature. Specific to healthcare, Dinov [25] presented two important char-
Several definitions of Big Data currently exist based on the character- acteristics of Big Data: their energy and life-span. The energy en-
istics of Big Data (Table 1). However, Scruggs et al. [24] state that the capsulates the holistic information content included in the data. Hence,
definition of Big Data is beyond the scope of these characteristics and the energy of the aggregated dataset is much higher than individual
should extend to its potential “to be useful and reused, accumulate database, rendering it more beneficial for exploring associations. The
value over time, and innovate a multi-dimensional, systems-level un- life-span of Big Data is in terms of its value past time of acquisition
derstanding”. which decays at an exponential rate. As per the definition of Big Data by
In healthcare, “Big Data includes heterogeneous, multi-spectral, Ghani et al. [36], Big Data is not limited by the purpose of answering a
incomplete and imprecise observations (e.g., diagnosis, demographics, specific question. On the other hand, Baro et al. [37] defines healthcare
treatment, prevention of disease, illness, injury, and physical and big data in terms of the number of statistical individuals (n) and the
mental impairments) derived from different sources using incongruent number of variables (p). According to him, Big Data are datasets with
sampling” [25]. Some of these data are structured and they focus on Log (n * p) ≥ 7, and they have the properties of great variety and high

Table 1
Big Data Definitions.
Authors Definition

Raghupathi & Raghupathi [10] By definition, big data in healthcare refers to electronic health data sets so large and complex that they are difficult (or impossible) to manage
with traditional software and/or hardware; nor can they be easily managed with traditional or common data management tools and methods
Auffray et al. [26] “Big data in health” encompasses high volume, high diversity biological, clinical, environmental, and lifestyle information collected from single
individuals to large cohorts, in relation to their health and wellness status, at one or several time points.
Bates et al. [18] By big data, we refer to the high volume, variety, and potential for the rapid accumulation of data
Dinov [25] Big healthcare data refers to complex datasets that have some unique characteristics, beyond their large size, that both facilitate and convolute
the process of extraction of actionable knowledge about an observable phenomenon
Karen et al. [29] Big Data is a term used to describe data sets with such large volume or complexity that conventional data processing methods are not good
enough to deal with them.
Tan et al. [30] Big data has been referred to as data that are too complex and large that cannot be processed and managed by traditional data processing tools
Hansen et al. [25] In addition to just having more data, Big Data also generally refers to the application of machine learning for analyzing the data sets.
Roski et al. [31] Big data–that is, the sophisticated and rapid analysis of massive amounts of diverse Information
Bian et al. [32] Big data is commonly defined through the 4 Vs: volume (scale or quantity of data), velocity (speed and analysis of real-time or near-real-time
data), variety (different forms of data, often from disparate data sources), and veracity (quality assurance of the data).
Ghani et al. [36] The term big data refers to ultra-large bodies of data that have not been prospectively limited in size or scope by the intent to address specific
research questions or disease conditions, and that grow continuously and rapidly.
Baro et al. [37] Big data can be defined as datasets with Log (n∗p) ≥7. Properties of big data are its great variety and high velocity.

59
N. Mehta, A. Pandit International Journal of Medical Informatics 114 (2018) 57–65

velocity. As described by Asante-Korang & Jacobs [3] and Groves et al. [47],
By analyzing the literature, it is evident that although the sig- by incorporating descriptive and comparative analytics, healthcare or-
nificance of big data in strengthening healthcare is recognized and ganizations have seen improved quality of care. However, they state
understood, there is still a lack of consensus on the operational defi- that the long-term tangible benefits can be accrued with utilization of
nition of big data in healthcare. Thus, the examination of definitions predictive analytics. According to the literature, predictive analytics
from previous studies allows discernment of the common elements. can be used for prediction of high-cost patients, readmissions, triage,
decompensation (when a patient’s condition worsens), adverse events,
3.2. Sources of healthcare big data and treatment optimization for diseases affecting multiple organ system
[18,27,36,48]. Some of the Big Data Analytical Techniques used in
Data in healthcare are disorganized and distributed, coming from healthcare are shown in Table 3.
various sources and having different structures and forms [38]. Furthermore, Mohammed et al. [39] highlighted some of the ap-
Healthcare Big Data includes data on physiological, behavioral, mole- plications of Big Data Technologies like MapReduce and Hadoop for
cular, clinical, environmental exposure, medical imaging, disease healthcare analytics which was supported by other researchers:
management, medication prescription history, nutrition, or exercise
parameters [26]. Some of the primary sources of Big Data in healthcare • MapReduce has the ability to improve the performance of common
are administrative databases (insurance claims and pharmaceuticals), signal detection algorithms for pharmacovigilance at approximately
clinical databases, electronic health record data [11], and laboratory linear speedup rates [59].
information system data [39]. The other sources of data [11] are bio- • Algorithms based on the Hadoop distributed platform can refine
metric data (wearable or sensor generated [35]), patient-reported data protein structure alignments more accurately than existing algo-
(standardized health surveys), data from social media [35], medical rithms [60].
imaging data, and biomarker data, including all the spectrum of ‘omics’ • MapReduce based algorithms can improve the performance of
data (that is, genomic, proteomic, and metabolomic data). Miller [40] neural signal processing [61].
identifies two main sources of health Big Data to be genomics-driven • Image reconstruction algorithms accelerate the reconstruction pro-
Big Data (genotyping, gene expression, sequencing data) and payer–- cess [62].
provider Big Data (electronic health records, insurance records, phar-
macy prescription, patient feedback and responses). On the other hand, MapReduce framework has been also been used by Markonis et al.
Swan [41] categorized Big Data streams into (a) Traditional medical [63] for finding optimal parameters for lung texture classification and
data obtained from EMRs, medication history and lab reports which to increase the speed of medical image processing.
assist in a better understanding of disease outcomes and optimizing Peek et al. [64], in their study discussed about some of the Hadoop-
healthcare delivery; (b) “Omics” data including genomics, micro- based Big Data processing tools like Oozie and Pig which can be used
biomics, proteomics, and metabolomics, which helps in understanding for batch processing; and non-Hadoop processing tools like Storm,
the mechanisms of diseases and accelerate the individualization of Spark, Hive and GraphLab which can be used for streaming data ana-
medical treatments (c) Data from social media, wearables & sensors lysis. Regardless of these potential applications, there is a need for
which provides the information about behavior and lifestyle of in- analytical tools to offer parallelization, in order to enable the timely
dividuals. Thus, the healthcare data comes from internal sources such processing of data [5].
as EMRs, CPOE, imaging data and biomedical data, as well as external
data sources, such as government, insurance claims/billing, R&D la- 3.4. Application of big data analytics in healthcare
boratories, and social media [10,42].
According to Belle et al. [2], healthcare data is spread among dif- Big Data analytics has the potential to transform business and
ferent healthcare systems, health insurers, researchers, government clinical models for smart and efficient delivery of care [34]. It enables
entities. Huang et al. [43] recognizes that Big Data in precision medi- integration of de-identified health information to allow secondary uses
cine comes from four different stakeholders: (a) Government and large of data [65]. Also, by recognizing patterns and deciphering associations
companies, (b) Smaller stakeholders such as academic groups and it can facilitate autonomous-decision making [66]. In clinical practice,
technology, biotech, and device startups, (c) Health care providers and Big Data analytics can help early detection of disease, accurate pre-
payers, and (d) Not-for-profit foundations and patient advocacy groups. diction of disease trajectory, and identification of deviation from
Clinical data such as vital signs, past medical history, medications, healthy state, changed disease trajectories and detection of fraud. By
immunizations and medical imaging can be derived from electronic providing this information, it helps the healthcare organizations in
health records, CPOE, clinical decision support systems medication personalization of predictions, targeted-treatment and cost-effective-
administration records, laboratory and pharmaceutical records [10,44], ness of care, and reduction in waste of resources; and by giving ac-
cohort studies, government surveys & clinical trials [20]. Adminis- tionable recommendations to individuals it encourages them maintain
trative data, on the other hand, contains patient demographic data and themselves in good health [10,18,26]. Big Data presents an opportunity
visit information, admit date, discharge date, ICD diagnosis & proce- to detect relatively low-frequency events that nonetheless can have
dure codes, admit source, discharge disposition and claims data such as significant clinical impact. Apart from that, clinical data integration and
charges for the visit, payer and reimbursement [44]. Table 2 sum- its effective usage support a vast range of applications, such as disease
marizes the sources of different types of healthcare data. surveillance, clinical decision support systems, and individual health-
care management; improvement of health-process efficiency; enhance-
3.3. Big data analytical techniques and technologies in healthcare ment of healthcare quality and reduction of healthcare cost [8,67].
Sukumar et al. [34] reveals that integrating Big Data analytics into
The multi-dimensional healthcare data – medical images (X-ray, healthcare can provide answer to eight important questions in health-
MRI images), biomedical signals (EEG, ECG, EMG etc.), audio tran- care − 1. How are costs for various aspects of health care likely to rise
scripts, handwritten prescriptions and structured data from EMRs [45] in the future? 2. How are certain policy changes impacting cost and
− and its dynamicity and complexity makes it difficult to analyze them. behavior? 3. How do health care costs vary geographically? 4. Can
There is paucity of analytical strategies that can handle such hetero- fraudulent claims be detected? 5. What treatment options seem most
geneous data and facilitate decision-making [46]. The literature men- effective for various diseases? 6. Why do some providers seem to have
tions some of the analytical approaches which can apply to healthcare better health outcomes? 7. Why do patients choose one provider over
and medicine. another? 8. Are there early signs of an epidemic?

60
N. Mehta, A. Pandit International Journal of Medical Informatics 114 (2018) 57–65

Table 2
Sources of Healthcare Data (Adapted from [15]).
Type Description Source

Clinical Electronic Medical Records (EMRs) Detailed patient-related information (physician prescriptions, medications, medical Hospitals & Clinics
history)
Diagnostic Diagnostic Results (imaging results, laboratory reports) Laboratories
Radiology Departments
Biomarkers Molecular data (genomic, proteomic, transcriptomic, metabolomic) Diagnostic Companies
Ancillary Administrative data (admission, discharge, transfer) & financial data (claims) Hospitals & Clinics
Data Aggregators
Claims Medical Claims Medical reimbursement data (procedures, hospital stay, insurance policy details) Payers
Data Aggregators
Prescription Claims Prescription reimbursement data (drugs, dose, duration) Payers
Data Aggregators
Clinical Research Clinical Trails Design parameters (compound, size, end points) Pharma Companies
Medical Journals
Patient-generated Data Social Media Community discussions Web Health Portals
Social Media Websites
Wearable & Sensors Wellness & lifestyle data (smartphones, fitness monitors) Device Data Systems

Table 3
Big Data Analytical Techniques in Healthcare.
Big Data Analytical Technique Healthcare Application Studies By

Cluster Analysis Determination of obesity clusters for identifying high-risk groups; Clark et. al. [49]
Determination of population clusters with specific health determinants for treatment of chronic diseases Swain [50], Schatz [51]
Data Mining Bio-signal monitoring for health-related abnormalities; Forkan et. al. [52]
Determination of epidemics; Ghani et. al. [36]
Inductive reasoning and exploratory data analysis in healthcare Roski et. al. [31]
Graph Analytics Analysis of hospital performance across various quality measures Downing et. al. [53]
Machine Learning Prediction of disease risk; Chen et. al. [54]
Assessment of the hospital performance; Downing et. al. [53]
Determination of epidemics Ghani et. al. [36]
Natural Language Processing (NLP) Improvement of efficiency of care and controlling costs; Wang et. al. [27]
Providing training, consultation and treatments; Khalifa and Meystre [55]
Identification of high-risk factors; Martin-Sanchez et. al. [5]
Extraction of information from clinical notes; Roski et. al. [31]
Reducing likelihood of morbidity & mortality
Neural Networks Diagnosis of chronic diseases; Al-Jumeily et. al. [56]
Prediction of patients’ future disease Martin-Sanchez et. al. [5]
Pattern Recognition Improvement of public health surveillance Luxton [57]
Spatial Analysis Extracting meaningful population-level insights by using visual, spatial and advanced analytics Amirian et al. [58]

Table 4 treatment heterogeneity; 5) precision medicine and clinical decision

Big Data Applications in Healthcare. support; 6) quality of care and performance measurement; 7) public
Application Area Studies By: health; and 8) research applications. Raghupathi & Raghupathi [10]
state that, ‘the areas in which advanced analytical techniques yield the
Genomics Maia et al. [70] greatest results include: pinpointing patients who are the greatest
Drug Discovery & Clinical Szlezák et al. [15]; Taglang & Jackson [46];
consumers of health resources or at the greatest risk for adverse out-
Research Wong et al. [72]
Personalized Healthcare Viceconti et al. [73]
comes; providing individuals with the information they need to make
Precision Medicine Leff & Yang [9]; Weng & Kahn [35]; Huang informed decisions and more effectively manage their own health as
et al. [43] well as more easily adopt and track healthier behaviors; identifying
Elderly Care Jiang et al. [74] treatments, programs and processes that do not deliver demonstrable
Mental Health Geerts et al. [75]
benefits or cost too much; reducing readmissions by identifying en-
Cardiovascular Disease Asante-Korang & Jacobs [3]; Rumsfeld et al.
[11]; Mandawat et al. [76]; Kim [77] vironmental or lifestyle factors that increase risk or trigger adverse
Diabetes Bellazi et al. [68]; Kumar Sarvana [48] events and adjusting treatment plans accordingly; improving outcomes
Gynecology Erekson & Iglesia [78] by examining vitals from at-home health monitors; managing popula-
Nephrology Nandkarni et al. [79] tion health by detecting vulnerabilities within patient populations
Oncology Mandawat et al. [76]; Maia et al. [70]; Naqa
[80]
during disease outbreaks or disasters; and bringing clinical, financial
Ophthalmology Clark et al. [49] and operational data together to analyze resource utilization produc-
Urology Ghani et al. [36] tively and in real time’. Electronic phenotyping is another area which
can successfully exploit Big Data technologies for ascertaining a clinical
condition or characteristic (phenotype) [68,69]. These studies show
Belle et al. [2] identified three major areas for the application of Big that there is a vast potential of Big Data analytics in Healthcare. It is
Data analytics in Healthcare: Image Processing, Signal Processing and beyond the scope of this review to encompass each of these potential
Genomics. On the other hand, the eight areas of application of Big Data applications. Table 4 summarizes some of the usability studies ac-
analytics to improve healthcare as per Rumsfeld et al. [11] include: 1) cording to their application areas.
predictive modelling for risk and resource use; 2) population manage- Apart from these clinical benefits and applications mentioned
ment; 3) drug and medical device safety surveillance; 4) disease and above, the literature also presents operational and financial benefits of

61
N. Mehta, A. Pandit International Journal of Medical Informatics 114 (2018) 57–65

Big Data analytics. From among all the articles examined, three articles • Training key personnel to use Big Data analytics: In order to extract
[6,27,70] unveil the business value in healthcare. Findings from the meaningful insights and valuable information from Big Data,
study of Wang & Hajli [6], exhibits that benefits of Big Data analytics healthcare professionals should be trained with Big Data analytics
are improved IT effectiveness and efficiency, and optimization of clin- competencies. This is critical for healthcare, because incorrect in-
ical operations. terpretation of the reports generated could lead to unanticipated
consequences [27].
3.5. Challenges in big data analytics in healthcare and strategies to • Incorporating cloud computing into the organization's Big Data analytics:
overcome them The challenge of storage of voluminous data can be tackled by
making use of cloud computing. This would enable small and
Even with huge potential benefits, the healthcare industry is in its medium sized hospitals and care organizations to eliminate cost and
nascent stage for adoption of Big Data analytics. With the huge amount data storage issues [27].
of data available, there is a lack of knowledge about which data to use
and for what purpose [15]. Another major challenge that healthcare According to Wang et al. [27], a shift of focus from technology tools
faces is the lack of appropriate IT infrastructure [8,27,81] and transi- to the managerial, economic, and strategic impacts of Big Data analytics
tion from use of paper-based records to use of distributed data pro- and exploration of effective path for acquiring healthcare business
cessing [64,82]. The resistance for redesigning processes and approving value would enable exploit the benefits of Big Data analytics.
technology that influences the health care system [39,70] and need for
huge initial investment [15,83], makes it more difficult to utilize Big 4. Discussion
Data technology. Studies show that because of the lack of knowledge
about the best algorithm and tool for analysis [71] and unavailability of 4.1. Main findings
trained clinical scientists and Big Data managers for interpretation of
Big Data outcomes [3,26,81,83], healthcare remains far from realizing This systematic review assessed the emerging landscape of Big Data
the potential of Big Data analytics. A major concern with the use of Big analytics for healthcare. Specifically, it identified the best available
Data analytics in Healthcare is the processing of information without literature about the concept of Big Data analytics, sources of Big Health
human supervision which might lead to erroneous conclusions [84,85]. Data, Big Data analytical techniques for clinical data, implementation
According to Raghupathi and Raghupathi [10], there is a need for a of Big Data analytics in healthcare, causes for underutilization of Big
simple, convenient and transparent Big Data analytics system which can Data analytics in healthcare and strategies to mitigate them.
be applied for real-time cases. The concept of Big Data covers a wide range of definitions extending
From the technical point of view, challenges include integration of from the data that is difficult to manage using traditional analytical
structured, semi-structured and unstructured data from a variety of tools, to the characterization of big data as having large volume, high
resources [26]. Studies show that the main technical issues in Big Data velocity, huge variety and varied veracity. While most of the studies
analytics include siloed/fragmented data [10,15,45], limitations of define Big Data in terms of the aforementioned characteristics, one of
observational data [11,43], validation [11], data structure issues, data the studies state that Big Data in healthcare is also characterized by
standardization issues [10,32,43,75,86], data inaccuracy and incon- having energy and life-span [25], but the literature lacks detailed de-
sistency (veracity) [15,20,32,34], data reliability [87], semantic inter- scription of these characteristics especially with regards to healthcare.
operability [25,64,87], network bandwidth, scalability, and cost [2]. Studies have demonstrated that Big Data analytics differ from tradi-
The problems such as missing data and the risk of false-positive asso- tional analytical approach in terms that instead of tracking care quality
ciations [11,88] also add to it. Security issues such as Big Data breaches and outcomes in retrospective view by using deductive reasoning, it
can be significant threat in healthcare [3,35]. uses inductive reasoning for prospective analysis of data [91]. These
Patient privacy and confidentiality are of utmost importance in techniques of data analysis are hypotheses-generating rather than hy-
healthcare. But data sharing between various stakeholders for deriving potheses-testing since they focus on finding association and correlation
insights, can deepen the concern for privacy [8,39,43,70,83,85]. Ac- in the observational data and not on the casual relationship between
cording to Mittelstadt et al. [83] informed consent and privacy are the variables. But it needs for the testing of hypotheses before applying
key areas of concern. Lack of data protocols and standards are some of results into a clinical practice [92]. Few of the studies state that there is
the governance issues faced by Big Data analytics in healthcare [2]. One a need for the application of human judgement and supervision on the
of the studies by Lee and Yoon [20] states that one of the prominent insights obtained from application of Big Data analytics in healthcare
reasons for the lack of clinical integration of Big Data technology is the [84,93], as it would prevent the occurrence of adverse events which
dearth of evidence of practical benefits of Big Data analytics in result from relying solely on Big Data analytics. Big Data analytics can,
healthcare. thus, lead to a shift from experience-based to evidence-based decision-
In order to overcome these challenges, various strategies were found making in healthcare [10,94].
in the literature. The strategies for curbing the aforementioned issues Most studies have shown that there are a large number of sources of
include: healthcare Big Data. Clinical and administrative data in healthcare
comes from various sources including healthcare providers, labora-
• Implementing (big) data governance: Due to poor governance, tories, diagnostic companies, insurance companies, pharmaceutical
healthcare organizations incur huge financial costs in IT investment firms, not-for-profit organizations, government and web-health portals
[23]. With appropriate data governance, the enterprise-wide data [38–44]. The literature on Big Data techniques and technologies that
resources can be leveraged effectively to create business value are applied to clinical data is largely fragmented. Most of this literature
[27,83]. highlights the use of natural language processing for clinical as well as
• Developing an information sharing culture: Information sharing and operational applications. As identified from the review, other Big Data
aggregation of data can address the issue of interoperability and techniques which find application in healthcare are cluster analysis,
enable effective utilization of the Big Data analytical and predictive data mining, graph analytics, machine learning, neural networks, pat-
capabilities [27,45]. tern recognition and spatial analysis. Studies showed that in most of the
• Employing security measures: Strong encryption of data, validation of cases, Hadoop and tools which run on top of Hadoop are used for
source of data, access control and authentication [89] and de- processing of patient-data [59–64], but since they are batch-processing
identification [90] are some of the measures for securing the data tools, newer tools like Storm, Spark and GraphLab have started finding
and maintaining confidentiality. their application for streaming and real-time data [64].

62
N. Mehta, A. Pandit International Journal of Medical Informatics 114 (2018) 57–65

Majority of the studies reviewed, concerned the application of Big 5. Conclusion

Data analytics in different areas of healthcare. According to them, Big
Data analytics finds application in clinical decision support; persona- Big Data analytics has emerged as a new frontier for enhancing
lized medicine; and optimization of clinical operations and cost-effec- healthcare delivery. With the opportunities created by digital and in-
tiveness of care. Thus, the integration of Big Data technology into formation revolution, healthcare industry can exploit the potential
healthcare can not only improve the quality of care, but enable early benefits of leveraging Big Data technology. Big Data analytics increas-
identification of high-risk patients by making use of real-time analytics ingly provides value to healthcare by improving healthcare quality and
and hence can benefit by saving lives. Studies on the use of Big Data outcomes and providing cost-effective care. The predictive nature and
analytics for cardiovascular diseases, diabetes, oncology, elderly care, pattern-recognition aspect of Big Data analytics enable the shift from
gynecology, and clinical research have shown that it can enable de- experience-based medicine to evidence-based medicine. Through its
livery of timely care and cost-saving by eliminating inefficiencies systematic review, the study presents a useful starting point for the
[70–80]. application of Big Data analytics in future healthcare research. In ad-
Despite the tremendous value-addition with use of Big Data analy- dition, the study reflects that once the scope of Big Data analytics is
tical tools, healthcare industry still lags in adoption of this technology defined; its characteristics and features are understood; and challenges
due to many challenges. The non-availability of appropriate IT infra- are properly tackled, its application will maximize the healthcare value
structure, huge investment costs associated with implementing analy- through promoting the extensive usage of insights.
tical tools, data privacy & security issues, fragmented data ownership
and technical challenges such as data quality & multi-dimensionality of Author contributions
data are some of the issues. One of the studies identified lack of evi-
dence of practical benefits as a major cause behind the reluctance for Study conception and design: Nishita Mehta
using Big Data analytics in healthcare [20]. Few studies highlighted Acquisition of data: Nishita Mehta, Dr. Anil Pandit
that there is a shortage of skilled Big Data analysts with knowledge of Analysis and interpretation of data: Nishita Mehta, Dr. Anil
healthcare, who have the ability to identify right data and right tools for Pandit
analysis of health-related data and interpret insights obtained after Drafting of manuscript: Nishita Mehta, Dr. Anil Pandit
analysis, which makes the use of technology difficult. The complexity of Critical revision: Nishita Mehta, Dr. Anil Pandit
Big Data analytical systems, is also one of the factors for limited use of
technology for healthcare applications [10]. Strategies for mitigating Conflict of interest
these challenges are hence required in order to realize its full potential.
Some of these strategies include change in organizational culture; None.
health-information exchange; training of key healthcare personnel;
development of simple-transparent Big Data systems; use of cloud sto- Summary Points
rage and distributed data processing; and strengthening of IT security.

What was already known on the topic:

4.2. Gaps and implications for future research

• Most of the studies reviewed, have a relatively narrow scope with • With digital evolution in healthcare, there is a huge amount of
health data available
• which remains unexploited and hence is wasted.
limited practical application. The published work discussed about

• Big Data analytics (BDA) has the ability to process health data
the potential application of Big Data analytics but there was no
evidence of real-world cases about the application in healthcare.
•
and generate valuable insights which can help improve the
None of the usability studies on Big Data analytics included in this
clinical, operational and financial outcomes.
•
review discussed about quantitative results by the usage of tech-
Despite the potential applications of BDA, healthcare industry
nology. Most of these studies used qualitative approach to explain
is slow in leveraging Big Data initiatives.
the benefits and challenges of using Big Data technology for
healthcare applications, while the use of quantitative approach will
What this study has added:
provide evidence for the practical benefits and will enable wide-

• Technology readiness for BDA is very low in healthcare ma-

scale adoption of technology.
• Majority of the studies included in review, were from the developed
jorly because of the non availability of skilled and trained
countries. It is essential to promote the research on Big Data ana-
clinical Big Data analysts.
•
lytics in healthcare in the developing countries, since that will en-
There is minimal evidence on how BDA can improve the
able delivery of better quality care.
quality of care, nor are there any economic assessment stu-
dies regarding its cost-effectiveness.
•
4.3. Limitations
There is a need for a user-friendly and transparent Big Data
system for wider adoption of BDA in healthcare.
•
While the literature covers information about Big Data analytics and
Shift in organizational culture and training of key healthcare
its role in healthcare and medicine, current research has few limita-
personnel, will enable effective utilization of BDA for better
tions. First, the contents of this study consists of a systematic review of
healthcare outcomes
the current status of Big Data technology in healthcare, but it does not
take into consideration the technical details regarding the im-
plementation and results obtained in each of the study reviewed.
Second, there is a heterogeneity in documentation since the literature References
contains disparate sources of information on definition of Big Data,
techniques of Big Data analytics and their application and challenges in [1] T.B. Murdoch, A.S. Detsky, The Inevitable Application of Big Data to Health Care
healthcare. Finally, despite the use of a systematic approach for review, 309 (2016), pp. 5–6.
[2] A. Belle, R. Thiagarajan, S.M.R. Soroushmehr, F. Navidi, D.A. Beard, K. Najarian,
the inclusion of studies on ‘big data analytics’ in ‘healthcare’ for this Big data analytics in healthcare, Hindawi Publ Corp. 2015 (2015) 1–16, http://dx.
review was based on subjective judgement, hence the cross-reference doi.org/10.1155/2015/370194.
articles were also considered for this review. [3] A. Asante-Korang, J.P. Jacobs, Big Data and paediatric cardiovascular disease in the

63
N. Mehta, A. Pandit International Journal of Medical Informatics 114 (2018) 57–65

era of transparency in healthcare, Cardiol. Young 26 (2016) 1597–1602, http://dx. [32] Jiang Bian, David Maslov, Mayer Angel Mayer, Siven Seevanayanagam, L. Toldo,
doi.org/10.1017/S1047951116001736. Challenges and opportunities of big data in health care: a systematic review, J. Med.
[4] Z. Goli-Malekabadi, M. Sargolzaei-Javan, M.K. Akbari, An effective model for store Internet Res. 4 (2016), http://dx.doi.org/10.2196/medinform.5359.
and retrieve big health data in cloud computing, Comput. Methods Programs [33] Hao Gui, Rong Zheng, C. Ma, An architecture for healthcare big data management
Biomed. 132 (2016) 75–82, http://dx.doi.org/10.1016/j.cmpb.2016.04.016. and analysis, Nternational Conf. Heal. Inf Sci. (2016) 154–160.
[5] F.J. Martin-Sanchez, V. Aguiar-Pulido, G.H. Lopez-Campos, N. Peek, L. Sacchi, [34] R. Sreenivas Sukumar, Ramachandran Natarajan, R. Ferrell, Quality of Big Data in
Secondary use and analysis of big data collected for patient care, IMIA Yearbook 26 health care, Int. J. Health Care Qual. Assur. 28 (2015) 621–634.
(2017) 1–10, http://dx.doi.org/10.15265/IY-2017-008. [35] C. Weng, M.G. Kahn, Clinical research informatics for big data and precision
[6] Y. Wang, N. Hajli, Exploring the path to big data analytics success in healthcare, J. medicine, Yearbook Med. Inf. (2016) 211–218, http://dx.doi.org/10.15265/IY-
Bus. Res. 70 (2017) 287–299, http://dx.doi.org/10.1016/j.jbusres.2016.08.002. 2016-019.
[7] B. Cyganek, M. Graña, B. Krawczyk, A. Kasprzak, P. Porwik, K. Walkowiak, [36] K.R. Ghani, K. Zheng, J.T. Wei, C.P. Friedman, Harnessing big data for health care
M. Woźniak, A survey of big data issues in electronic health record analysis, Appl. and research: are urologists ready? Eur. Urol. 66 (2014) 975–977, http://dx.doi.
Artif. Intell. 30 (2016) 497–520, http://dx.doi.org/10.1080/08839514.2016. org/10.1016/j.eururo.2014.07.032.
1193714. [37] Emilie Baro, Samuel Degoul, Regis Beuscart, E. Chazard, Toward a literature-driven
[8] F.F. Costa, Big data in biomedicine, Drug Discov. Today 19 (2014) 433–440, http:// definition of big data in healthcare, BioMed. Res. Int. (2015).
dx.doi.org/10.1016/j.drudis.2013.10.012. [38] N. Rouse, B. William, Serban Understanding and Managing the Complexity of
[9] D.R. Leff, G.-Z. Yang, Big data for precision medicine, Engineering 1 (2015) Healthcare, (2014).
277–279, http://dx.doi.org/10.15302/J-ENG-2015075. [39] E.A. Mohammed, B.H. Far, C. Naugler, Applications of the MapReduce program-
[10] W. Raghupathi, V. Raghupathi, Big data analytics in healthcare: promise and po- ming framework to clinical big data analysis: current landscape and future trends,
tential, Heal. Inf. Sci. Syst. 2 (2014) 3, http://dx.doi.org/10.1186/2047-2501-2-3. BioData Min. 7 (2014) 22, http://dx.doi.org/10.1186/1756-0381-7-22.
[11] J.S. Rumsfeld, K.E. Joynt, T.M. Maddox, Big data analytics to improve cardiovas- [40] K. Miller, Big data analytics in biomedical research, Biomed. Comput. Rev. (2012)
cular care: promise and challenges, Nat. Rev. Cardiol. 13 (2016) 350–359, http:// 15–21.
dx.doi.org/10.1038/nrcardio.2016.42. [41] M. Swan, The quantified self: fundamental disruption in big data science and bio-
[12] J. Manyika, M. Chui, B. Brown, et al., Big Data:The Next Frontier for Innovation, logical discovery, Big Data 1 (2013).
Competition, and Productivity, McKinsey Global Institute, 2011, 2018. [42] Michael J. Ward, Keith A. Marsolo, C.M. Froehle, Applications of business analytics
[13] L. Haar, Big Data Expected to Have Big Impact on Diagnostic Imaging, (2014) in healthcare, Bus. Horiz. 57 (2014) 571–582.
http://www.diagnosticimaging.com/siim-2014/big-data-expected-have-big- [43] B.E. Huang, W. Mulyasasmita, G. Rajagopal, The path from big data to precision
impact-diagnostic-imaging. medicine, Expert Rev. Precis. Med. Drug Dev. 1 (2016) 129–143, http://dx.doi.org/
[14] M. Grindle, J. Kavathekar, D. Wan, A new era for the healthcare industry–cloud 10.1080/23808993.2016.1157686.
computing changes the game, Accenture (2013). [44] P.S. Bradley, Implications of big data analytics on population health management,
[15] N. Szlezák, M. Evers, J. Wang, L. Pérez, The role of big data and advanced analytics Big Data 1 (2013) 152–159.
in drug discovery, development, and commercialization, Clin. Pharmacol. Ther. 95 [45] D.V. Dimitrov, Medical internet of things and big data in healthcare, Healthcare Inf.
(2014) 492–495, http://dx.doi.org/10.1038/clpt.2014.29. Res. 22 (2016) 156–163, http://dx.doi.org/10.4258/hir.2016.22.3.156.
[16] W.D. Kirsh, Why Longitudinal Data Is Crucial to Making Better Care Decisions, [46] G. Taglang, D.B. Jackson, Use of big data in drug discovery and clinical trials,
(2017) https://www.healthdatamanagement.com/opinion/why-longitudinal-data- Gynecol. Oncol. 141 (2016) 17–23, http://dx.doi.org/10.1016/j.ygyno.2016.02.
is-crucial-to-making-better-care-decisions. 022.
[17] E. Capobianco, Systems and precision medicine approaches to diabetes hetero- [47] S. Groves, Basel Van Kayyali, David Knott, P. Kuiken, The Big Data Revolution in
geneity: a Big Data perspective, Clin. Transl. Med. 6 (2017) 23, http://dx.doi.org/ Healthcare: Accelerating Value and Innovation, (2013).
10.1186/s40169-017-0155-4. [48] N.M. Saravana Kumar, T. Eswari, P. Sampath, S. Lavanya, Predictive methodology
[18] D.W. Bates, S. Saria, L. Ohno-Machado, A. Shah, G. Escobar, Big data in health care: for diabetic data analysis in big data, Procedia Comput. Sci. 50 (2015) 203–208,
using analytics to identify and manage high-risk and high-cost patients, Health Aff. http://dx.doi.org/10.1016/j.procs.2015.04.069.
33 (2014) 1123–1131, http://dx.doi.org/10.1377/hlthaff.2014.0041. [49] A. Clark, J.Q. Ng, N. Morlet, J.B. Semmens, Big data and ophthalmic research, Surv.
[19] H. Krumholz, Big data and new knowledge In medicine: the thinking training, and Ophthalmol. 61 (2016) 443–465, http://dx.doi.org/10.1016/j.survophthal.2016.
tools needed for a learning health system, Health Aff. 33 (2014) 1163–1170. 01.003.
[20] M. Cox, D. Ellsworth, Application-controlled demand paging for out-of-core vi- [50] A. Swain, Mining big data to support decision making in healthcare, J. Inf. Technol.
sualization, Proceedings. Vis. ’97 (Cat. No. 97CB36155) (1997) 235–244, http://dx. Case Appl. Res. 18 (2016) 141–154, http://dx.doi.org/10.1080/15228053.2016.
doi.org/10.1109/VISUAL.1997.663888. 1245522.
[21] F.X. Diebold, Big data dynamic factor models for macroeconomic measuring and [51] B. Schatz, National surveys of population health: big data analytics for mobile
forecasting, Adv. Econ. Econom. Eighth World Congr. Econom. Soc. (2003) health monitors, Big Data 3 (2015) 219–229, http://dx.doi.org/10.1089/big.2015.
115–122. 0021.
[22] D. Laney, META delta, Appl. Deliv. Strateg. 949 (2001) 4, http://dx.doi.org/10. [52] A. Forkan, I. Khalil, M. Atiquzzaman, ViSiBiD, A learning model for early discovery
1016/j.infsof.2008.09.005. and real-time prediction of severe clinical events using vital signs as big data,
[23] B. Feldman, E.M. Martin, T. Skotnes, Big data in healthcare–hype and hope, Dr. Comput. Networks 113 (2017) 244–257, http://dx.doi.org/10.1016/j.comnet.
Bonnie 360 degree, Bus. Dev. Digit. Heal. 2012 (2013) 122–125 http://www.riss. 2016.12.019.
kr/link?id=A99883549. [53] N. Downing, A. Cloninger, A. Venkatesh, A. Hsieh, E. Drye, R. Coifman, et al.,
[24] S.B. Scruggs, Karol Watson, Andrew I. Su, Henning Hermjakob, J.R. Yates 3rd, Describing the performance of U. S. hospitals by applying big data analytics, PLoS
M.L. Lindsey, P. Ping, Harnessing the heart of big data, Circ. Res. 116 (2015) One 12 (2017) e0179603, http://dx.doi.org/10.1371/journal.pone.0179603.
1115–1119. [54] M. Chen, Y. Hao, K. Hwang, L. Wang, L. Wang, Disease prediction by machine
[25] I.D. Dinov, Volume and value of big healthcare data, J. Med. Stat. Inf. 4 (2016), learning over big data from healthcare communities, IEEE Access. 5 (2017)
http://dx.doi.org/10.7243/2053-7662-4-3. 8869–8879, http://dx.doi.org/10.1109/access.2017.2694446.
[26] C. Auffray, R. Balling, I. Barroso, L. Bencze, M. Benson, J. Bergeron, E. Bernal- [55] A. Khalifa, S. Meystre, Adapting existing natural language processing resources for
Delgado, N. Blomberg, C. Bock, A. Conesa, S. Del Signore, C. Delogne, P. Devilee, cardiovascular risk factors identification in clinical notes, J. Biomed. Inf. 58 (2015)
A. Di Meglio, M. Eijkemans, P. Flicek, N. Graf, V. Grimm, H.-J. Guchelaar, Y.- S128–S132, http://dx.doi.org/10.1016/j.jbi.2015.08.002.
K. Guo, I.G. Gut, A. Hanbury, S. Hanif, R.-D. Hilgers, Honrado Á, D.R. Hose, [56] D. Al-jumeily, A. Hussain, C. Mallucci, C. Oliver, Applied Computing in Medicine
J. Houwing-Duistermaat, T. Hubbard, S.H. Janacek, H. Karanikas, T. Kievits, and Health, (2015).
M. Kohler, A. Kremer, J. Lanfear, T. Lengauer, E. Maes, T. Meert, W. Müller, [57] D. Luxton, Artificial Intelligence in Behavioral and Mental Health Care, (2015).
D. Nickel, P. Oledzki, B. Pedersen, M. Petkovic, K. Pliakos, M. Rattray, J.R. i Màs, [58] P. Amirian, F. van Loggerenberg, T. Lang, A. Thomas, R. Peeling, A. Basiri,
R. Schneider, T. Sengstag, X. Serra-Picamal, W. Spek, L.A.I. Vaas, O. van Batenburg, S.N. Goodman, Using big data analytics to extract disease surveillance information
M. Vandelaer, P. Varnai, P. Villoslada, J.A. Vizcaíno, J.P.M. Wubbe, G. Zanetti, from point of care diagnostic machines, Pervasive Mob. Comput. (2017), http://dx.
Making sense of big data in health research: towards an EU action plan, Genome doi.org/10.1016/j.pmcj.2017.06.013.
Med. 8 (2016) 71, http://dx.doi.org/10.1186/s13073-016-0323-y. [59] W. Wang, K. Haerian, H. Salmasian, R. Harpaz, H. Chase, C. Friedman, A drug-
[27] Y. Wang, L.A. Kung, T.A. Byrd, Big data analytics: understanding its capabilities and adverse event extraction algorithm to support pharmacovigilance knowledge
potential benefits for healthcare organizations, Technol. Forecast. Soc. Change mining from PubMed citations, AMIA Annu. Symp. Proc. 1464-70 (2011)
(2015), http://dx.doi.org/10.1016/j.techfore.2015.12.019. (22195210).
[28] K.Y. He, D. Ge, M.M. He, Big data analytics for genomic medicine, Int. J. Mol. Sci. [60] C.-L. Hung, Y.-L. Lin, Implementation of a parallel protein structure alignment
18 (2017) 1–18, http://dx.doi.org/10.3390/ijms18020412. service on cloud, Int. J. Genomics. 2013 (2013) 1–8, http://dx.doi.org/10.1155/
[29] S.S.-L. Tan, G. Gao, S. Koch, Big data and analytics in healthcare, Methods Inf. Med. 2013/439681.
54 (2015) 546–547, http://dx.doi.org/10.3414/ME15-06-1001. [61] L. Wang, D. Chen, R. Ranjan, S.U. Khan, J. Kołodziej, J. Wang, Parallel processing of
[30] H. Liyanage, S. De Lusignan, S. Liaw, C. Kuziemsky, F. Mold, P. Krause, D. Fleming, massive EEG data with MapReduce, Proc. Int. Conf. Parallel Distrib. Syst.–ICPADS.
Big data usage patterns in the health care domain: a use case driven approach (2012) 164–171, http://dx.doi.org/10.1109/ICPADS.2012.32.
applied to the assessment of vaccination benefits and risks contribution of the IMIA [62] B. Meng, G. Pratx, L. Xing, Ultrafast and scalable cone-beam CT reconstruction
primary healthcare working group big data for assessing vaccination benefits and using MapReduce in a cloud computing environment, Med. Phys. 38 (2011)
risks: a, IMIA, Yearbook Med. Inf. (2014) 27–35. 6603–6609, http://dx.doi.org/10.1118/1.3660200.
[31] J. Roski, G.W. Bo-Linn, T.A. Andrews, Creating value in health care through big [63] D. Markonis, R. Schaer, I. Eggel, H. Muller, A. Depeursinge, Using MapReduce for
data: opportunities and policy implications, Health Aff. 33 (2014) 1115–1122, large-Scale medical image analysis, 2012 IEEE Second Int. Conf. Healthcare
http://dx.doi.org/10.1377/hlthaff.2014.0147. Informatics, Imaging Syst. Biol. (2012) 1, http://dx.doi.org/10.1109/HISB.2012.8.

64
N. Mehta, A. Pandit International Journal of Medical Informatics 114 (2018) 57–65

[64] N. Peek, J. Holmes, J. Sun, Technical challenges for big data in biomedicine and 1124–1129, http://dx.doi.org/10.1016/j.jmig.2015.07.003.
health: data sources, infrastructure, and analytics, IMIA Yearbook 9 (2014) 42–47, [79] G.N. Nadkarni, S.G. Coca, C.M. Wyatt, Big data in nephrology: promises and pitfalls,
http://dx.doi.org/10.15265/iy-2014-0018. Kidney Int. 90 (2016) 240–241, http://dx.doi.org/10.1016/j.kint.2016.06.003.
[65] I. Cano, A. Tenyi, E. Vela, F. Miralles, J. Roca, Perspectives on Big Data applications [80] I. El Naqa, Perspectives on making big data analytics work for oncology, Methods
of health information, Curr. Opin. Syst. Biol. 3 (2017) 36–42, http://dx.doi.org/10. 111 (2016) 32–44, http://dx.doi.org/10.1016/j.ymeth.2016.08.010.
1016/j.coisb.2017.04.012. [81] S. Fodeh, Q. Zeng, Mining Big Data in biomedicine and health care, J. Biomed. Inf.
[66] A. Asante-Korang, J.P. Jacobs, Big Data and paediatric cardiovascular disease in the 63 (2016) 400–403, http://dx.doi.org/10.1016/j.jbi.2016.09.014.
era of transparency in healthcare, Cardiol. Young 26 (2016) 1597–1602, http://dx. [82] F. Zhang, J. Cao, S. Khan, K. Li, K. Hwang, A task-level adaptive MapReduce fra-
doi.org/10.1017/S1047951116001736. mework for real-time streaming data in healthcare applications, Future Gener.
[67] T. Schultz, Turning healthcare challenges into big data opportunities: a use-case Comput. Syst. 43-44 (2015) 149–160, http://dx.doi.org/10.1016/j.future.2014.06.
review across the pharmaceutical development lifecycle, Bull. Assoc. Inf. Sci. 009.
Technol. (2013). [83] J. Wu, H. Li, S. Cheng, Z. Lin, The promising future of healthcare services: when big
[68] Riccardo Bellazzi, Arianna Dagliati, Lucia Sacchi, D. Segangi, Big data technologies: data analytics meets wearable technology, Inf. Manag. 53 (2016) 1020–1033,
new opportunities for diabetes management, J. Diabetes Sci. Technol. 9 (2015) http://dx.doi.org/10.1016/j.im.2016.07.003.
1119–1125. [84] G. Asokan, V. Asokan, Leveraging big data to enhance the effectiveness of one
[69] L.J. Frey, L. Lenert, G. Lopez-Campos, EHR big data deep phenotyping. health in an era of health informatics, J. Epidemiol. Global Health 5 (2015)
Contribution of the IMIA Genomic Medicine Working Group, Yearbook Med. Inf. 9 311–314.
(2014) 206–211, http://dx.doi.org/10.15265/IY-2014-0006. [85] M. Grossglauser, H. Saner, Data-driven healthcare: from patterns to actions, Eur. J.
[70] Y. Wang, L. Kung, W.Y.C. Wang, C.G. Cegielski, An integrated big data analytics- Prev. Cardiol. 21 (2014) 14–17, http://dx.doi.org/10.1177/2047487314552755.
enabled transformation model: application to health care, Inf Manag. (2017) 1–16, [86] R. Budhiraja, R. Thomas, M. Kim, S. Redline, The role of big data in the manage-
http://dx.doi.org/10.1016/j.im.2017.04.001. ment of sleep-disordered Breathing, Sleep Med. Clin. 11 (2016) 241–255, http://dx.
[71] A.T. Maia, S.-J. Sammut, A. Jacinta-Fernandes, S.-F. Chin, Big data in cancer doi.org/10.1016/j.jsmc.2016.01.009.
genomics, Curr. Opin. Syst. Biol. 4 (2017) 78–84, http://dx.doi.org/10.1016/j. [87] S. Salas-Vega, A. Haimann, E. Mossialos, Big data and health care: challenges and
coisb.2017.07.007. opportunities for coordinated policy development in the EU, Health Syst. Reform 1
[72] H.T. Wong, Q. Yin, Y.Q. Guo, K. Murray, D.H. Zhou, D. Slade, Big data as a new (2015) 285–300, http://dx.doi.org/10.1080/23288604.2015.1091538.
approach in emergency medicine research, J. Acute Dis. 4 (2015) 178–179, http:// [88] T. McNutt, K. Moore, H. Quon, Needs and challenges for big data in radiation on-
dx.doi.org/10.1016/j.joad.2015.04.003. cology, Int. J. Radiation Oncol., Biol., Phys. 95 (2016) 909–915, http://dx.doi.org/
[73] M. Viceconti, P. Hunter, D. Hose, Big Data, big knowledge: big data for personalised 10.1016/j.ijrobp.2015.11.032.
healthcare, IEEE J. Biomed. Heal. Inf. 2194 (2015) 1, http://dx.doi.org/10.1109/ [89] R. Naydenov, D. Liveri, L. Dupre, E. Chalvatzi, C. Skouloudi, Big Data Security:
JBHI.2015.2406883. Good Practices and Recommendations on the Security of Big Data Systems,
[74] P. Jiang, J. Winkley, C. Zhao, R. Munnoch, G. Min, L.T. Yang, An intelligent in- European Union Agency for Network and Information Security (ENISA), Greece,
formation forwarder for healthcare big data systems with distributed wearable 2015https://www.enisa.europa.eu/publications/big-data-security/at_download/
sensors, IEEE Syst. J. (2014) 1–9, http://dx.doi.org/10.1109/JSYST.2014.2308324. fullReport.
[75] H. Geerts, P.A. Dacks, V. Devanarayan, M. Haas, Z.S. Khachaturian, M.F. Gordon, [90] J. Adler-Milstein, A. Jha, Healthcare's big data challenge, Am. J. Manage. Care 19
S. Maudsley, K. Romero, D. Stephenson, Big data to smart data in Alzheimer’s (2013) 537–538.
disease: the brain health modeling initiative to foster actionable knowledge, [91] C.H. Lee, H.-J. Yoon, Medical big data: promise and challenges, Kidney Res Clin.
Alzheimer’s Dement 12 (2016) 1014–1021, http://dx.doi.org/10.1016/j.jalz.2016. Pract. 36 (2017) 3–11, http://dx.doi.org/10.23876/j.krcp.2017.36.1.3.
04.008. [92] H. Care, B. Outcomes, Leveraging Big Data and Analytics in Healthcare and Life
[76] A. Mandawat, A.E. Williams, S.A. Francis, Cardio-oncology: the role of big data, Sciences: Enabling Personalized Medicine for High-Quality Care, Better Outcomes
Heart Fail. Clin. 13 (2017) 403–408, http://dx.doi.org/10.1016/j.hfc.2016.12.010. Developing Technologies to Meet Big Data Challenges, (2012).
[77] J. Kim, Big Data, Health informatics, and the future of cardiovascular medicine, J. [93] Ernst, Young, Is the next evolution of big data big judgment? (2015) http://www.
Am. Coll. Cardiol. 69 (2017) 899–902, http://dx.doi.org/10.1016/j.jacc.2017.01. ey.com/gl/en/issues/business-environment/ey-analytics (accessed August 14,
006. 2017).
[78] E.A. Erekson, C.B. Iglesia, Improving patient outcomes in gynecology: the role of [94] H.J. Watson, Tutorial: big data analytics: concepts, technologies, and applications,
large data registries and big data analytics, J. Min. Invasive Gynecol. 22 (2015) Commun. Assoc. Inf. Syst. 34 (1) (2014) 1247–1268.

65

View publication stats