Content uploaded by Zainab Alansari
Author content
All content in this area was uploaded by Zainab Alansari on Jul 06, 2018
Content may be subject to copyright.
Content uploaded by Zainab Alansari
Author content
All content in this area was uploaded by Zainab Alansari on Jul 05, 2018
Content may be subject to copyright.
The Internet of Things Adoption in Healthcare
Applications
Zainab Alansari
1,2
, Nor Badrul Anuar
1
and Amirrudin
Kamsin
1
Safeeullah Soomro
2
and Mohammad Riyaz Belgaum
2
1
Department of Computer Science & Information Technology
2
Department of Computer Studies
University of Malaya AMA International University
Kuala Lumpur, Malaysia Salmabad , Kingdom of Bahrain
z.alansari@siswa.um.edu.my, { badrul & amir}@ um.edu.my { s.soomro & bmdriyaz}@ amaiu.edu.bh
Abstract - The Internet of Things (IoT) integrated with
various healthcare applications and medical fields such as remote
care system for patients, warning systems for emergencies, fitness
programs, chronic diseases and also elderly care such as heart
rate checking system, blood pressure measurement system, health
check systems, artificial heart rate provider and hearing aids.
Furthermore, IoT adopted applications monitor the treatment or
drugs quantity process. Additionally, many applications are
produced based on IoT adoption in healthcare systems that are
used by doctors to monitor their patients after discharging from
hospital. The aim of this study is to give priority to the adequate
healthcare field. The study distinguishes different users of IoT in
healthcare systems as well as its functions and preferences. The
Fuzzy Analytic Hierarchy Process (FAHP) along with the
development analysis of Chang, Da-Yong has been used to
prioritize the IoT adoption in healthcare applications.
Index Terms - Internet of Things, Healthcare applications,
FAHP.
I. INTRODUCTION
Internet of Things is an expression which is used to
describe the objects that have soft communication through the
internet. In everyday life, we walk among smart cities that
calculate our movement in the moment, reports our energy
consumption to power plants, and even more, the smart selling
devices send their product selling information [1]. But could
also healthcare systems enter this gaming of big data and move
along with technology? How to adopt these significant
innovations in healthcare application?
What the companies are trying to reach is to fill the gap
between the patient and the doctor. At the moment, technology
is advanced enough to measure our heart rate, motility, glucose
levels and much more by ourselves. Nowadays, the hands of
people who walk in the streets decorated with smart devices
that regularly observe their health condition. But the question
is that how we can share our information with specialists to use
them usefully? The main problem is efficiency and the way of
using these great innovations is a challenge as well;
furthermore, many doctors doubt the security and the
information confidentiality. In many cases, patients limit to
telephone counseling instead of medical appointments with
doctors [2]. In such cases, if a wireless blood pressure
monitoring device is available, most meetings change to the
virtual meeting which counts as a considerable saving.
Internet of things can somehow reconstruct the healthcare
systems so all people from anywhere in the world can access
the desired medical facilities [3]. Moreover, frequent
examinations and capturing the moment-to-moment health
indices have an incredible impact on disease prevention and
timely treatment. What is important is to implement demanded
infrastructure for new technologies in healthcare systems.
The internet of things can remodel and modernize the
healthcare systems indeed, and it influences the way people,
machines, and software interact [4]. The modest helps of the
IoT in healthcare systems are remote care system for patients,
warning systems for emergencies, fitness programs, chronic
diseases, equipment management, monitoring temperature, air
quality, energy consumption and elderly care such as heart rate
measuring system, blood pressure monitoring machine, health
check systems, artificial heart rate provider and hearing aids.
In more advanced cases, the IoT adopted applications monitor
the treatment or drugs quantity process. Additionally, many
applications are designed based on IoT adoption in healthcare
systems which are used by doctors to monitor their patients
after discharging from hospital. To date, a limited number of
studies conducted in the context of IoT adoption in healthcare
applications and it seems to be imperative to prioritize some
domains of IoT that have the essential dormant in healthcare
application adoption.
The aim of this study is to give priority to the operative
fields of the healthcare applications to advance IoT. Section
two distinguishes different users of IoT in healthcare systems
as well as its capacities and preferences. Section three
endeavors the revised new conceptual framework which
proposed in the previous paper. In section four we use FAHP
along with the development analysis of Chang, Da-Yong [5] to
prioritize the IoT adoption in healthcare applications.
Additionally, section five presents the final weight of each
three primary criteria and the nine sub-criteria and renders
their ranking. Finally, section six delivers a conclusion and
reasons the future studies.
AUTHOR'S COPYAUTHOR'S COPY
** AUTHOR'S COPY **
This is a pre-print of an article published in IEEE Xplore on 26 March 2018 for a Conference in Bangkok, Thailand
held on 7-8 Aug. 2017 2017 IEEE 3rd International Conference on Engineering Technologies and Social Sciences
(ICETSS) and is available online with DOI: 10.1109/ICETSS.2017.8324138
II. IOT IN HEALTHCARE SYSTEMS
Internet of Things is a modern technology that involves in
every field. Day by day, more devices increasingly connect to
the internet and exchange information.
A. Different Functions of IoT in Healthcare Systems
Commonly, IoT adoptions in healthcare applications
utilize the following goals:
1) Continuous collection of life signs: life signs covering
blood pressure, heart beat, body temperature, and respiration
use creative tools equipped with connectivity systems that are
continuously gathering and storing information 24/7.
2) Periodical collection of life signs: life signs comprising
blood pressure, heart beat, body temperature, and respiration
are adopting smart devices which by connecting to the system
collect and store the reports periodically at adjusted times.
3) Continuous collection of specific parameters
associated with common chronic diseases: specific parameters
associated with common chronic diseases including blood
sugar, blood fat, body water percentage, stress levels and risk
of seizure use productive tools including connectivity systems
that are continuously gathering and storing information 24/7.
4) Periodical collection of specific parameters associated
with common chronic diseases: specific parameters related to
common chronic diseases including blood sugar, blood fat,
body water percentage, stress levels and risk of seizure are
adopting smart devices which by connecting to the system
collect and store the reports periodically at adjusted times.
5) Tracking and monitoring: every object (people, health
equipment, etc.) And their abilities to interact with the wireless
sensor network devices are monitored, measured and
controlled 24/7 by connecting bases that are located
everywhere with great connections capacity.
6) Remote assistance: healthcare and life services can be
provided remotely through internet-optimized devices such as
emergency, first aid, healthy houses, diet and medication
management, telemedicine and remote diagnostics, social
health networks, etc.
7) Data management: by connecting global IoT, every
healthcare data (support, diagnosis, treatment, recovery,
medicine, management, finance and even daily activities) can
be used in the entire series of data collection and management.
Data can be classified and regained from the systems as a
detailed and compiled report on the disease, city, state, time,
patient, doctors, and physician.
8) Sending smart contents to user: through the given life
signs and precise parameters associated with common chronic
diseases that is collected per person and according to their
defined threshold levels, in case of obvious symptoms, the
system has the capability to change a person's profile page
accordingly and automatically send the articles or educational
films described that disease to the user's screen.
9) Enterprises integration: with the help of IoT in
healthcare systems, integrated interagency information can be
obtained. This feature enables authorized persons such as
doctors, nurses, physiotherapists, and radiologist to access all
medical information of the patient at any place [6].
B. Advantages of IoT in Healthcare Systems
The main advantages of this technology in healthcare
systems are as follows:
1) Costs reduction: with the ability to meet and examine
patients remotely, in-person visit’s cost can be reduced.
Besides, with the advent of home care equipment, many
patients can be hospitalized and monitored at their homes.
2) Treatment outcomes: since the monitoring is consistent,
continuous and automated, all data are stored in the cloud and
sent to doctor regularly; the treatment processes moved
accurately. The use of this method can ensure the timely
medical care to assess the recovery process.
3) Disease management: when a person’s health signs are
steadily recorded and reported, the disease can be identified
and treated before its progression.
4) Errors reduction: detailed and precise data that are
collected automatically and freed of human errors can
dramatically lessen the rate of medical errors and its associated
financial and critical costs.
5) Patient’s satisfaction: the emphasis on patient's
requirements, data accuracy, timely treatment, costs reduction,
reduction of repeated visit, recording of recovery process, and
the most significant is the patient actively involvement in
treatment process which all possess a positive impact on him.
6) Medication management: IoT helps the patients in the
exact usage of drugs as well as the pharmacies and healthcare
centers to prevent the drug waste [7].
III. CONCEPTUAL FRAMEWORK
This research is a revised version of the author’s previous
article [8]; consequently the same conceptual framework is
used to identify the criteria’s influencing IoT healthcare
adoption. The proposed conceptual model has developed
based on the incorporation of findings from previous related
studies and meetings with specialists. According to the
conducted researches, the model presented considers
innovative as it confers a new framework for classifying the
criteria influencing IoT adoption in healthcare systems. Figure
1 exhibits the conceptual model which prioritizes IoT in health
sector according to sustainable progress and based on the
Fuzzy Analytical Hierarchy Process (FAHP) method.
Fig. 1 The Conceptual Model of the study based on FAHP.
Selecting IoT Applications in Health Sector
Economic
Prosperity
Quality of Life Environmental
Protection
Medical Frid
g
es
Fall Detection
Patient Surveillance
Chronic Disease
H
yg
ienic Hand Control
Slee
p
Control
Dental Health
S
p
ortsmen Care
Ultraviolet Radiation
AUTHOR'S COPYAUTHOR'S COPY
** AUTHOR'S COPY **
To design the conceptual model, the previous relevant
studies were reviewed, and after identifying the key criteria
and interviewing with experts, the criteria that are affecting the
IoT healthcare system adoption have been divided into three
main criteria of economic prosperity, environmental protection
and quality of life. According to the available published
papers, each of these criteria is divided into nine sub-criteria.
The following are some discussion on the criteria and sub-
criteria of the proposed new conceptual model.
IoT European Research Cluster (IERC) has been
presented a comprehensive classification of relevant areas of
IoT in smart health. Some usages are service type, and some
are a product. Related areas of IoT in the health sector (smart
health) are:
1) Medical Fridges (internal temperature protective
control): some organic elements must be kept in containers
with certain conditions (temperature). IoT can well assume
this task and cause objects interaction.
2) Fall Detection: this usage focused on helping the
physically challenged and elderly in their lives so that they can
live independently.
3) Sportsmen Care: the application used to measure the
weight, sleep, exercise, blood pressure and other relevant
parameters for professional athletes.
4) Patient Surveillance: used for remote in-hospital
monitoring, (especially the elderly) or used for patient’s home
care.
5) Chronic Disease Management: taking care of patients
with chronic diseases while there is no need of physical
attendance. This technology reduces the presence of people in
hospitals and results in lower costs, reduces hospital stay and
reduces traffic (even reduces fuel consumption).
6) Ultraviolet Radiation: UV rays measurement and
notifying the people to stop to enter certain areas or refrain of
exposure to UV rays at certain hours.
7) Hygienic Hand Control: by linking devices such as
designed RFID for emissions measurement, environmental
pollution could be identified.
8) Sleep Control: devices that by linking to individuals,
identifies some signs such as heart rate, blood pressure during
sleep and the data may be collected and analyzed after.
9) Dental Health: Bluetooth-enabled toothbrush with the
help of Smartphone apps records someone’s brushing
information to study the person’s brushing habits and share
the statistics with the dentist [9].
IV. METHODOLOGY
A. Fuzzy Analytic Hierarchy Process (FAHP)
FAHP is a method which by developing AHP based on
fuzzy logic, allows the researchers to use the non-exact data in
the analysis. In this study, the researchers used the triangular
fuzzy numbers proposed by Hu et al. [10]. It can be seen in
table I.
TABLE I
Q
UALITATIVE TERM AND THEIR CORRESPONDING TRIANGULAR FUZZY
NUMBERS
Linguistic
variables
Positive
triangular fuzzy
number
Positive reciprocal triangular
fuzzy number
Extremely strong (9, 9, 9) (0.11, 0.11, 0.11)
Intermediate (7, 8, 9) (0.11, 0.13, 0.14)
Very strong (6, 7, 8) (0.13, 0.14, 0.17)
Intermediate (5, 6, 7) (0.14, 0.17, 0.20)
Strong (4, 5, 6) (0.17, 0.20, 0.25)
Intermediate (3, 4, 5) (0.20, 0.25, 0.33)
Moderately Strong (2, 3, 4) (0.25, 0.33, 0.5)
Intermediate (1, 2, 3) (0.33, 0.5, 1)
Equally Strong (1, 1, 1) (1, 1, 1)
B. Modified Fuzzy Analytic Hierarchy Process (FAHP) with
Development Analysis of Chang, Da-Yong
For the implementation of AHP with fuzzy logic, in 1996,
Chinese researchers called Chang presented a development
analysis. The numbers that are used in this method are
triangular fuzzy numbers. Each triangular fuzzy number is
shown with “(1),” and “(2),” is subjected to the “(3),” joint
functions.
()
umlA ,,
~
=
(1)
uml ≤≤
(2)
() ()( )
()( )
°
°
¯
°
°
®
−−
−−
=
0
/
/
0
~
muxu
lmlx
xu
ux
uxm
mxl
lx
>
≤≤
≤≤
<
(3)
In the paired comparisons method, for each of the matrix
rows, the Sk value which is also a triangular fuzzy number is
calculated using “(4).”
[]
1
111
k
S
−
===
¦¦¦
×=
m
i
n
jij
n
jkj
MM
(4)
The
k
represents the number of rows, the
i
and
j
represents options and criteria respectively. In this method,
after calculating Sk, the large degree towards each other has to
be calculated. In general, if
1
M
and
2
M
are two triangular
fuzzy numbers, the large degree of
1
M
towards
2
M
defined
as “(5).”
()
()()
¯
®
=≥
=≥
2121
21
1
MMhgtMMV
MMV
¿
¾
½
≤
≥
21
21
mifm
mifm
(5)
So that can have “(6).”
()
()
()()
2211
12
21 lmum
ul
MMhgt −−−
−
=
(6)
Furthermore, to calculate the weights of the criteria in the
matrix of pairwise comparisons the “(7),” is used:
(){}
iknkSSVXW
kii
≠=≥=
;,...,2,1min)(
(7)
AUTHOR'S COPYAUTHOR'S COPY
** AUTHOR'S COPY **
Therefore, measuring the vector weight is calculated
using “(8),” which is the non-weighted coefficients vector of
FAHP.
[]
T
n
xWxWxWW )(),...,(),(
21
′′′
=
′
(8)
These matrices weighted with the help of “(9).”
¦
′
′
=
i
i
i
W
W
W
(9)
To achieve a holistic perspective in decision-making
which is an outcome of the entire expert’s opinion, the paired
comparison matrices must be combined. To achieve this
purpose, the geometric mean method used in group decision-
making. The “(10),” shows the relationship between the
compositions of the experts. In this definition,
L
is the
number of decision makers.
k
k
L
ijLij
XX
1
1
~¸
¸
¹
·
¨
¨
©
§
=
∏
=
(10)
In most of the available resources such as books and
articles, the FAHP along with Chang’s development analysis
is used. This method is used sometimes in calculating the
negative weights of criteria and sub-criteria which are
considered as its primary defect. For preventing the
calculation of negative weights and solving these problems,
Chang proposed that firstly the decision matrix converts to
normal cellular and then run the FAHP along with Chang’s
development analysis. Most of the researchers didn’t pay
attention to this solution. This study uses the modified
Chang’s method which is normalization of cellular matrix
before using the technique, to prevent the calculation of
negative weights.
Reliability assess in multi-criteria decision-making
techniques is different from reliability assess in statistics. For
each extracted decision matrix from the expert’s view, the
percentage of incompatibility must be calculated. Thus, it
shows if there is any reasonable consistency between paired
comparisons decision makers. To determine the reliability of
fuzzy decision making, the incompatibility percentage of each
of the final matrix is calculated. If the definite matrix of paired
comparisons is consistent, paired comparisons is a compatible
matrix phase 16. Therefore, the fuzzy decision matrix changes
to matrices containing final numbers. For this purpose, the
Center of Area (CA) method is used which means converting
the fuzzy numbers to non-fuzzy numbers. Calculation of this
approach for fuzzy numbers is shown in “(11).”
()( )
l
lmlu
CA +
−+−
=3
(11)
V. ANALYSIS AND RESULTS
This research is benefited from a group decision-making
process, and decision tables which are included in this section
are all the result of the geometric mean of all of the fifteen
completed questionnaire by experts, doctors, managers and
IoT specialist in Bahrain (for the brevity sake, presenting of
each table and FAHP was prevented).
In the following, integrated matrix derived from expert
opinion is given. After calculating the percentage of
inconsistency tables, all were obtained a result less than 1.0,
which represents its consistency and reliability. Finally, it
proceeded to criteria prioritizing using Chang’s finding
developed methods. To prevent the calculation of negative
weights, first converted the regular decisions matrices to
cellular matrices and then applied FAHP with Chang’s
development analysis approach. It is notable that for
shortening the article, the details of each calculating were
filtered, and only cellular normalized decision tables are
shown.
A. Weighting and Prioritization Criteria and Sub-Criteria
As stated, in Chang’s development analysis approach, the
integrated regular decision matrix is cellular, and Chang’s
algorithm is implemented. Then by the implementation of
AHP and Chang’s approach, the weight of each criterion is
calculated. As can be seen in table II, the economic prosperity
criteria with the weight of 0.413 achieved the first rank, the
quality of life with the weight of 0.386 set next and the criteria
of environmental protection with the weight of 0.201 achieved
the last rank.
TABLE II
T
HE FINAL WEIGHT AND RANK OF THE THREE TOP CRITERIA
Criteria’s sign Criteria’s name Criteria’s
Weight
Priority
C1 Economic prosperity 0.413 1
C2 Environmental protection 0.201 3
C3 Quality of Life 0.386 2
After calculating the weight of main criteria’s, the weight
of sub- criteria's has to be estimated and prioritized. The
results of these calculations presented in Table III.
TABLE III
T
HE FINAL WEIGHT AND RANK OF ALL NINE SUB
-
CRITERIA
Sub-criteria Final sub-criteria’s weight Priority
Sleep Control 0.158 1
Sportsmen Care 0.132 2
Dental Health 0.126 3
Medical Fridges 0.112 4
Patient Surveillance 0.109 5
Fall Detection 0.105 6
Ultraviolet Radiation 0.101 7
Hygienic Hand Control 0.093 8
Chronic Disease Management 0.064 9
Among the nine sub-criteria influencing the IoT
healthcare systems adoption, the sleep control (with a final
weight of 0.158 is placed in the first priority and chronic
disease management (with a final weight of 0.064) is
positioned in the last priority. Other sub-criteria priorities are
shown in table III.
AUTHOR'S COPYAUTHOR'S COPY
** AUTHOR'S COPY **
VI. CONCLUSION AND FUTURE STUDIES
This study is a revised and expanded version of an article
entitled “The Rise of Internet of Things (IoT) in Big
Healthcare Data: Review and Open Research Issues” which
used the proposed new conceptual model by identifying the
criteria influencing IoT healthcare systems adoption. Previous
studies rarely explore the issues raised in this research or just
have been referred shortly. The study, by the use of
interviews, expert opinions and review of previously related
researches provided a new conceptual model which includes
three main criteria intention in the use of IoT as a new
technology. Each of these three criteria also covers nine other
sub-criteria. This research prioritizes aspects of the conceptual
model using modified fuzzy analytic hierarchy process.
Undoubtedly, this research can be the basis for making
other research application model to improve the healthcare
systems acceptance of IoT and to compare the healthcare
situations in future studies. This study helps the hospitals and
the healthcare organizations to have a right decision for
implementing the IoT in the context of suitable construction
framework for the use of internet in every househeld health
monitoring machines. On the other hand, it helps the
manufactories to focus on the company's capital factors which
have the greatest influence on the acceptance of this new
technology.
Although the internet of things can transform the
treatment area in healthcare systems, the collection of this big
volume of sensitive data itself is a big challenge. The
incorrectly share of big data can have adverse effects on
treatment areas and also the reliability of the collected data.
Besides, retrieve, record and digitize this volume of health
data undoubtedly pressurise the databases. With common
health trackers and wearable monitoring physical parameters,
the specialists who receive and study these data need special
equipments and a data center. Furthermore, health services
need a standard language through which they can share the
information with each other. Security is considered as another
major challenge for the smart healthcare systems. With the
increasing volume of collected data, the need of protecting
these data against cyber attacks also increase.
Internet of things architecture consists of a powerful
analytics engine, a cloud platform or a virtualized
infrastructure. These tools should be used and managed
somehow to immediately report and resolve any anomalies in
network traffic, user access or system errors. When it comes
to healthcare systems, the slightest interruption in the system
operation may be at the expense of human life.
REFERENCES
[1] Barnaghi, Payam, Wei Wang, Cory Henson, and Kerry Taylor.
"Semantics for the Internet of Things: early progress and back to the
future." International Journal on Semantic Web and Information Systems
(IJSWIS) 8, no. 1 (2012): 1-21.
[2] Weber, Rolf H. "Internet of Things–New security and privacy
challenges." Computer law & security review 26, no. 1 (2010): 23-30.
[3] Atzori, Luigi, Antonio Iera, and Giacomo Morabito. "The internet of
things: A survey." Computer networks 54, no. 15 (2010): 2787-2805.
[4] Gubbi, Jayavardhana, Rajkumar Buyya, Slaven Marusic, and Marimuthu
Palaniswami. "Internet of Things (IoT): A vision, architectural elements,
and future directions." Future generation computer systems 29, no. 7
(2013): 1645-1660.
[5] Chang, Da-Yong. "Applications of the extent analysis method on fuzzy
AHP." European journal of operational research 95, no. 3 (1996): 649-
655.
[6] Bui, Nicola, and Michele Zorzi. "Health care applications: a solution
based on the internet of things." In Proceedings of the 4th International
Symposium on Applied Sciences in Biomedical and Communication
Technologies, p. 131. ACM, 2011.
[7] Li, Shancang, Li Da Xu, and Shanshan Zhao. "The internet of things: a
survey." Information Systems Frontiers 17, no. 2 (2015): 243-259.
[8] Zainab Alansari, Safeeulah Soomro, Mohammad Riyaz Belgaum, and
Shahaboddin Shamshirband. " The Rise of Internet of Things (IoT) in Big
Healthcare Data: Review and Open Research Issues." International
Conference on Advanced Computing and Intelligent Engineering, India
(2016).
[9] Vermesan, Ovidiu, and Peter Friess, eds. Internet of things-from research
and innovation to market deployment. Aalborg: River Publishers, 2014.
[10] Hu, Allen H., Chia-Wei Hsu, Tsai-Chi Kuo, and Wei-Cheng Wu. "Risk
evaluation of green components to hazardous substance using FMEA and
FAHP." Expert Systems with Applications 36, no. 3 (2009): 7142-7147.
AUTHOR'S COPYAUTHOR'S COPY
View publication statsView publication stats
** AUTHOR'S COPY **