Int
ern
at
i
onal
Journ
al of Ele
ctrical
an
d
Co
mput
er
En
gin
eeri
ng
(IJ
E
C
E)
Vo
l.
10
,
No.
5
,
Octo
be
r
2020
,
pp.
4918
~
4927
IS
S
N: 20
88
-
8708
,
DOI:
10
.11
591/
ijece
.
v10
i
5
.
pp
4918
-
49
27
4918
Journ
al h
om
e
page
:
http:
//
ij
ece.i
aesc
or
e.c
om/i
nd
ex
.ph
p/IJ
ECE
An i
nn
ov
ativ
e IoT
s
er
vice for
m
ed
ical
d
iagnosis
Sa
fi
a Abb
as
Depa
rt
m
ent
o
f
C
om
pute
r
Scie
n
ce
,
Facu
lty
of
C
om
pute
r and
I
nform
at
ion
Scie
n
ce
s,
Prince
ss
Nourah
bint Abdulra
hm
an
Univer
si
t
y
,
K
ingdom of
S
audi
A
rab
ia
,
Ain
Sham
s
Univer
sit
y
,
Eg
y
pt
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
J
ul
16
, 2
019
Re
vised
Ma
r
16
,
2020
Accepte
d
Ma
r
2
6
, 202
0
Due
to
t
he
m
isdia
gnose
of
dis
ea
ses
that
in
creased
re
ce
nt
l
y
i
n
a
sca
r
i
l
y
m
anne
r,
m
an
y
re
sea
rch
ers
d
evot
e
d
the
ir
eff
orts
an
d
depl
o
y
ed
tech
nologi
es
t
o
improve
the
m
edi
c
al
di
agnosis
proc
ess
and
r
educ
ing
the
r
esult
ed
r
isk
.
Acc
ordingly
,
this
pape
r
proposed
arc
hitect
u
re
of
a
c
y
b
er
-
m
edi
c
ine
servic
e
for
m
edi
ca
l
dia
gnos
is,
base
d
intern
et
of
thi
ngs
(
Io
T)
and
cl
oud
in
fra
struct
ur
e
(Ia
aS).
Thi
s
ser
vic
e
off
ers
a
s
har
ed
env
ironment
for
m
edica
l
dat
a
,
an
d
ext
ra
ct
ed
knowl
edge
and
f
indings
bet
wee
n
p
at
i
ent
s
and
doc
tors
in
an
int
er
ac
t
ive,
sec
ure
d,
e
la
sti
c
a
nd
rel
i
abl
e
wa
y
.
It
pre
d
ic
ts
the
m
edi
c
a
l
dia
gnosis
and
p
r
ovide
s
an
appr
o
pria
t
e
tr
ea
tment
for
the
g
ive
n
s
ym
ptoms
and
m
edi
ca
l
conditi
ons
base
d
on
m
ult
ipl
e
class
ifiers
to
assure
hi
gh
accurac
y
.
Moreove
r,
i
t
entail
s
different
fun
ct
ion
al
ities
such
as
on
-
demand
sea
rch
ing
fo
r
scie
ntific
pap
ers
and
disea
ses
desc
ription
for
unre
cognize
d
co
m
bina
ti
on
of
s
y
m
ptoms
using
web
cra
wl
er
to
enr
ic
h
the
r
esult
s.
W
her
e
such
sea
rch
i
n
g
result
s
from
cr
awle
r,
are
proc
essed,
an
aly
z
ed
and
add
ed
to
the
resid
ent
knowledge
base
(KB)
to
ac
hi
e
ve
ada
p
ta
bi
li
t
y
and
subs
idi
ze
t
he
service
pre
dictive
ab
il
i
t
y.
Ke
yw
or
d
s
:
Cy
ber
-
m
edici
ne
e
-
h
eal
th
IoT
Copyright
©
202
0
Instit
ut
e
o
f Ad
vanc
ed
Engi
n
ee
r
ing
and
S
cienc
e
.
Al
l
rights re
serv
ed
.
Corres
pond
in
g
Aut
h
or
:
Safia
Abbas,
Dep
a
rtm
ent o
f C
om
pu
te
r
Scie
nce
,
Faculty
of
C
om
pu
te
r
an
d
I
nf
or
m
at
ion
S
ci
en
ces
,
Pr
inces
s
N
oura
h
bin
t
A
bdulra
hm
an
Un
i
ver
si
ty
,
Kingdom
o
f
Saudi
Ar
a
bia
,
Ain Sham
s Un
iversity
, Cair
o, Egy
pt.
Em
a
il
:
sa
m
ah
m
ou
d@pnu.e
du.sa
,
sa
fia_a
bbas@cis.as
u.
e
du.e
g
1.
INTROD
U
CTION
In
s
om
e
instan
ces,
m
any
peo
ple
su
f
fer
se
ve
rely
fr
om
a
con
diti
on
bro
ught
abo
ut
by
m
is
diag
nosis.
Mi
sd
ia
gnos
is r
esults
in
s
om
e
conditi
ons
w
hi
ch
co
ul
d
be
a
voide
d
by doin
g
pro
per
d
ia
gn
osi
s,
the o
utc
ome
m
a
y
resu
lt
in
death
and
lo
ng
li
fe
disabili
ti
es.
A
ddit
ion
al
ly
,
sev
eral
causes
le
a
d
to
m
isdiagno
sis,
su
c
h
as
di
seases
that
exit
regu
la
rly
,
inadequ
at
e
inform
ation
ab
ou
t
the
pa
rtic
ular
il
lnes
s,
an
d
m
iss
or
de
rin
g
of
re
quire
d
la
borator
y
te
st.
Re
searche
rs
ha
ve
m
ade
us
e
of
data
sci
ence
,
inter
net
a
nd
inf
or
m
at
ion
te
chnolo
gy
to
dis
cov
e
r
pro
per
rem
edies
fo
r
s
uc
h
m
isdiagnosis
em
erg
encies
that
incr
eases
day
by
da
y.
Ba
sed
on
t
he
resea
rch
done
by
sp
eci
al
iz
ed
aut
hors
who
ha
ve
earn
e
d
the
tr
ust
of
tra
pp
i
ng
data
sci
ence,
m
isdiagnosis
c
auses
se
ver
al
pro
blem
s
fo
r
patie
nts.
A
dd
it
io
nally
,
it
has
bee
n
note
d
that
sever
al
sci
entifi
c
te
r
m
s
t
end
to
s
how
ap
pro
pr
ia
te
strat
egies
in
the
healt
h
do
m
ai
n
that
app
l
ie
s
su
ch
te
c
hn
ologies.
T
he
r
esearche
rs
i
ntr
oduce
d
the
im
plem
entat
ion
of
te
rm
s,
su
c
h
as
e
Healt
h
[
1
-
3]
an
d
m
Healt
h
[
4,
5].
Howe
ver,
cy
be
r
m
edici
ne
ha
s
bec
om
e
the
m
os
t
widely
app
li
ed
te
chn
iq
ue
i
n
t
he
present
days
[6
-
8].
Applic
at
ion
of
cy
ber
m
edicin
e
in
the
ea
rl
y
nin
et
ie
s,
w
hi
ch
re
pr
e
sents
the
e
Healt
h,
intr
oduce
d
the
a
pp
li
cat
ion
of
i
ntern
et
co
m
m
un
ic
at
ion
t
o
deliver
a
he
al
th
care
syst
em
est
ablished
base
d
on
a
dis
si
m
il
ar
com
pu
te
r
data
base
a
nd
for
be
arin
g
m
anag
e
m
ent
ap
plica
tio
n
to
gethe
r
by
the
inte
rn
et
[9
-
13]
.
A
fter
wards,
Io
T
te
ch
no
l
ogy
was
intr
oduc
ed,
w
hich
ac
com
m
od
at
ed
di
ff
ere
nt
ob
j
ect
s
fixe
d
with
s
of
t
war
e,
sen
sor
a
nd
netw
o
r
k
c
onne
ct
ivit
y
m
a
inly
to
co
op
e
rate
f
or
data
c
ollec
ti
on
a
nd
exc
ha
nge
[
14
-
16
]
.
M
any
ap
plica
ti
on
s
that
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
An
i
nnov
ative
IoT
se
rvi
ce for
med
ic
al d
i
agnosis
(
Safi
a
A
bb
as
)
4919
te
nd
to
a
pply
su
c
h
te
ch
niques
in
t
he
he
al
th
do
m
ai
n
ha
ve
bee
n
in
ve
ste
d
in
healt
h
do
m
ai
n
su
cc
essfu
ll
y.
Io
T
te
ch
nolo
gy
al
lowed
deli
ver
y
of
ser
vice
s
to
patie
n
ts,
thr
ough
the
int
ern
et
.
Also
,
t
he
te
chnolo
gy
al
lowed
patie
nts
to
sto
re
inf
or
m
at
ion
relat
ing
to
th
ei
r
healt
h
co
ndit
ion
s
for
use
by
sp
eci
fic
consulta
nts.
M
edical
serv
ic
es
wer
e
therefo
re
co
nducted
th
rou
gh
the
internet,
w
it
ho
ut
nec
essar
il
y
seei
ng
the
do
ct
or
face
t
o
face.
Diff
e
re
nt
s
hortcom
ing
s
res
ulted
th
r
ough
the
ap
plica
ti
on
of
I
oT
te
c
hn
i
qu
e
,
s
uch
as
le
a
kin
g
of
t
he
patie
nt
’
s
pr
i
vate
in
form
at
ion
t
hroug
h
the
transm
issio
n
m
edia.
Io
T
con
ce
pt
inclu
ded
f
unct
ion
al
resour
ces
s
uc
h
as
sens
or
,
s
of
t
wa
re,
an
d
netw
ork
co
n
necti
vity
,
wh
ic
h
c
omm
un
ic
at
es
to
each
oth
e
r
f
or
data
colle
ct
ion
an
d
exch
a
nge
[
17,
18
]
.
Applic
at
io
n
of
I
oT
co
nce
pt
res
ults
into
so
m
e
add
ed
a
dvanta
ge,
s
uch
as
rem
ote
con
trol
of
patie
nt’s
healt
h
a
nd
fitness
inf
or
m
at
ion
,
a
n
al
ert
c
onditi
on
duri
ng
an
e
m
erg
ency,
an
d
rem
ote
co
ntro
l
of
sign
ific
a
nt
m
e
dical
treatm
ent
par
am
et
ers.
I
oT
be
com
es
ben
efici
al
in
the
healt
h
care
do
m
ai
n
since
it
app
li
es
t
h
e
p
o
w
e
r
o
f
t
h
e
i
n
t
e
l
l
i
g
e
n
c
e
a
n
d
s
m
a
r
t
d
e
v
i
c
e
s
l
i
k
e
d
t
o
g
e
t
h
e
r
,
f
o
r
t
h
e
p
r
o
v
i
s
i
o
n
o
f
e
f
f
e
c
t
i
v
e
i
n
f
o
r
m
a
t
i
o
n
r
e
p
o
s
i
t
o
r
i
e
s
.
T
h
e
s
i
g
n
i
f
i
c
a
n
t
i
n
f
o
r
m
a
t
i
o
n
i
s
t
h
e
n
e
v
a
l
u
a
t
e
d
a
n
d
a
n
a
l
y
z
e
d
,
f
o
r
e
f
f
e
c
t
i
v
e
u
s
e
i
n
t
h
e
h
e
a
l
t
h
c
a
r
e
d
o
m
a
i
n
.
Currentl
y,
the
idea
of
cl
oud
com
pu
ti
ng
i
n
the
eHealt
h
dom
ai
n
is
la
rg
e
ly
sp
rea
d.
It
c
om
pr
ise
s
of
sever
al
la
ye
rs
inv
ol
ving
cl
oud
c
om
pu
ti
ng
infr
ast
r
uct
ur
e
(I
aaS
)
an
d
s
of
t
war
e
as
a
serv
ic
e
(
SaaS
)
la
ye
rs.
The
m
ai
n
act
ivit
y
involves
the
pro
vision
of
upgr
a
da
ble,
require
d,
a
da
pt
able
an
d
prote
ct
ed
at
m
os
ph
e
re
f
or
healt
h
care
do
m
ai
n
that
gu
ar
antees
reduce
d
exp
e
ns
es
an
d
protect
ed
ser
vi
ces
[16,
19
-
21
]
.
Su
ch
act
ivit
ie
s
are
consi
der
e
d
as
t
he
m
ai
n
con
cer
n
an
d
sig
nifica
nt
chall
en
ges
i
n
cy
ber
m
edici
ne
sect
or.
Cl
ou
d
com
pu
ti
n
g
m
ai
nly
cov
e
rs
a
wide
range
of
act
ivi
ti
es
as
com
par
ed
to
cy
ber
m
edici
ne
w
hic
h
is
lim
i
te
d
to
som
e
extent.
Acc
ordi
ng
to
the
resear
ch
ers,
t
he
im
ple
m
entat
ion
of
I
oT
base
d
cl
ou
d
dicta
te
s
a
w
el
l
-
groun
ded
de
ci
sion
-
m
akin
g
m
od
el
that
acce
ss
ce
rtai
n
be
ne
fits,
su
c
h
as
data
colle
ct
ion
a
nd
ada
ptati
on,
avail
abili
ty
of
data,
in
w
hich
bot
h
the
patie
nts
a
nd
co
nsult
ants
can
ha
ve
ac
cess
to,
ef
fecti
ve
data
sh
a
ring
an
d
t
ran
s
m
issi
on
,
a
nd
secure
transm
issi
on
of
data
be
twee
n
diff
e
re
nt
par
ti
es.
Re
searc
hers
sugg
est
the
i
m
ple
m
entat
ion
of
cl
ou
d
com
pu
ti
ng
con
ce
pts,
m
ai
n
ly
f
or ef
fecti
ve
operati
on i
n
th
e h
eal
th
dom
ain
.
The
rest
of
thi
s
pap
e
r
is
orga
nized
as
fo
ll
ows:
Sect
ion
2
pro
vid
es
a
li
te
ratur
e
s
urvey
in
the
IoT
a
nd
e
-
healt
h.
Sect
ion
3,
e
xpla
ins
the
diag
nosti
c
Ser
vice
Bl
ock
Diag
ram
and
it
s
functi
onal
it
ie
s.
Sect
io
n
4
presents
the
interact
io
n
of
the
us
e
rs
an
d
the
ser
vice
with
scr
eens
ho
ts.
Sect
ion
5,
gi
ves
the
co
nclu
sio
n
a
nd
the futu
re
wor
ks
.
2.
RELATE
D
W
ORK
Re
centl
y,
num
ero
us
stu
die
s
and
ap
plica
ti
on
s
ha
ve
been
c
onduct
ed
on
and
pro
vid
e
d
as
Ho
m
e
-
diag
nos
is
In
rece
nt
ye
ars,
div
e
rse
resea
rch
es
a
nd
ap
plica
ti
ons
of
I
oT
an
d
inter
net
-
ba
se
d
cl
ou
d
infr
a
struct
ur
e
us
in
g
big
data
con
ce
pts
as
a
Ho
m
e
-
diag
nos
is
serv
ic
e
ha
ve
bee
n
c
onduct
ed
a
nd
ap
plied.
Seve
ral
of
the
stu
dies
c
on
ce
r
ned
are
ba
sed
on how
t
o
util
iz
e
the IoT
,
w
hic
h
m
on
it
ors
a
nd
a
naly
zes
the p
at
ie
nts’ h
eal
th
conditi
ons
a
nd
pro
vid
e
al
erts
f
or
a
ny
c
riti
cal
cases.
Wh
il
e
sever
al
othe
r
a
pp
li
cat
io
ns
a
nd
st
ud
ie
s
m
entione
d
are
how
t
he
cl
oud
i
nfrastr
uctur
e
(
IaaS)
an
d
big
data
a
re
dep
l
oyed
i
n
t
he
ir
a
ppli
cat
ions
in
orde
r
to
a
chieve
a p
e
rv
asi
ve,
on
-
dem
and
se
r
vice with
secu
re
da
ta
tran
sm
issi
on
c
hannels.
The
aut
hors
i
n
[
2
2
]
re
gard
the
m
ai
n
aspect
of
IoT
as
connecti
ng
heter
og
e
ne
ou
s
entit
ie
s
and
assem
bling
la
rg
e
am
ou
nts
of
data,
th
us
sai
d,
i
n
co
ntext
t
o
the
e
-
healt
h
en
vironm
ent,
IoT
is
re
gard
ed
a
s
the
proce
ss
of connecti
ng d
at
a
ab
ou
t
the
pat
ie
nt
to
facil
it
at
e
treatm
ent
effe
ct
ively
and
e
f
fici
ently
,
as
we
ll
as
to
receive
m
or
e
com
pr
ehe
ns
ive
know
le
dg
e
.
T
he
auth
ors
il
lum
inate
that
in
evita
bly
healt
hc
are
perso
nnel
will
hav
e
m
utu
al
knowle
dge
a
nd
acce
s
sibil
it
y
to/f
or
a
patie
nt'
s
data.
T
he
aut
hors
pro
po
s
e
a
sm
art
rem
ote
diag
nosis
deci
sion
s
upport
m
od
el
,
wh
ic
h
de
plo
ys
the
pr
ovided
visi
ons
of
Io
T
to
the
e
-
he
al
th
env
ir
onm
ents
.
The
pr
opos
e
d
m
od
el
seem
s
to
be
ad
apti
ve;
the
aut
hors
offe
red
no
sp
eci
fic
guid
el
ine,
m
e
tho
dolo
gy,
or
te
c
hniq
ue
f
or
t
he
decisi
on
m
aking
proc
e
ss
of
the
sm
art
m
od
el
.
This
pa
per
offe
red
a
n
a
bs
tract
ob
se
rv
at
io
n
of
c
omm
on
issues
in
the
cl
oud
en
vir
on
m
ents,
i.e.,
sec
ur
it
y
in
the
transm
i
ssion
of
data
and
t
he
avail
abi
li
ty
of
data, as
w
el
l as
quickly
cit
ed pre
-
existi
ng s
ol
ution
s
.
The
a
uthors
i
n
[2
3
]
ta
rg
et
e
d
people
in
S
uboptim
al
Healt
h
Stat
us
(SHS
),
known
as
“t
he
third
sta
te
”,
wh
ic
h
m
eans
to
be
betwee
n
the
sta
te
of
bein
g
healt
hy
and
fall
in
g
il
l.
To
ai
d
peopl
e
in
getti
ng
di
sease
pr
eca
utio
n
knowle
dge
easi
ly
,
they
de
velo
pe
d
ho
m
e
sel
f
-
ca
re
se
rv
ic
es
bas
e
d
on
their
des
ign
of
a
distrib
ute
d
Luce
ne
-
base
d
search
cl
us
te
r
wh
ic
h
dep
l
oys
the
cl
oud
Ia
aS.
U
sin
g
this
desig
n,
t
hey
are
able
t
o
ac
hieve
scal
abili
ty
dat
a
retrievin
g,
a
naly
sis,
an
d
pr
ivacy
protect
io
n.
F
or
data
an
al
ysi
s,
the
appl
ic
at
ion
us
es
f
or
m
al
con
ce
pt
c
om
pu
ta
ti
on
a
nd
bl
oo
m
filt
er
signa
ture.
T
heir
a
ppli
cat
ion
not
only
prov
i
des
a
n
on
-
dem
and
s
tora
ge
m
od
el
bu
t
al
s
o
pr
ov
i
des
a
n
el
ast
ic
scal
able
m
od
el
that
m
anag
es
the
r
us
h
ho
ur
acce
ss.
O
n
t
he
ot
he
r
ha
nd,
howe
ver,
an
offli
ne
L
uce
ne
data
file
is
ne
eded
t
o
be
pr
o
cesse
d
aut
oma
ti
cal
ly
or
m
a
nu
al
ly
.
Alt
hough
a
n
autom
at
ic
fo
r
m
at
ion
f
or
s
uc
h
a
file
ap
pears
to
be
diff
ic
ult,
it
is
sti
ll
has
the
abili
ty
to
e
njo
y
the
el
ast
ic
it
y
bu
t
loses it
s self
-
a
dap
ta
ti
on
wh
il
e in the
onli
ne
m
od
e.
The
a
uthor
s
pr
esent
a
s
urvey
in
[
2
4
]
t
ha
t
dis
cusses
t
he
po
te
ntial
chall
en
ge
s
that
co
uld
be
faced
us
i
ng
rem
ote
m
on
it
or
in
g
te
ch
nolo
gi
es
that
prese
ntly
exist
an
d
im
plem
enting
the
m
in
I
oT
a
nd
c
loud
e
nv
i
ronm
ents.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
5
,
Oct
ob
e
r 2
020
:
49
18
-
49
27
4920
Po
te
ntial
chall
eng
e
s
m
entioned
thr
ough
ou
t
the
i
m
ple
m
en
ta
ti
on
co
uld
be
;
the
la
te
ncy
of
da
ta
tra
nsfer
on
tim
e
-
crit
ic
al
task
s
of
ag
gr
e
gat
ed
data,
heter
ogene
ous
data
acqu
isi
ti
on
fro
m
an
arr
ay
of
r
eso
ur
ces
an
d
s
ens
or
s,
nu
m
erous
am
o
un
ts
of
data
a
nd
a
naly
ses
from
Io
T
based
sensors
in
rel
at
ion
to
both
m
achine
le
arn
i
ng
a
nd
visu
al
iz
at
ion
phases
in
ord
e
r
to
receive
the
m
os
t
accurate
diagnosis.
More
ov
e
r,
the
authors
disc
usse
d
the
be
ne
fits
an
d
po
sit
ive
heal
th
im
pact
of
he
al
th
m
on
it
or
ing
an
d
t
he
m
a
nag
em
ent
of
usi
ng
I
oT
se
ns
i
ng
with
cl
oud
pro
cessi
ng,
f
ur
t
her
m
ore,
offer
i
ng
a
pr
oacti
ve
sc
hem
e
for
the
progn
os
is
of
disease
s
at
it
s
incipie
nt
sta
te
,
al
ong
with
pre
ven
ti
on,
c
ur
e,
and
c
om
pr
ehe
ns
ive
m
anag
em
ent
of
he
al
th
over
disease
.
This
s
urvey
e
xh
i
bits
per
s
onal
iz
ed
m
anag
em
ent
and
t
reatm
ent
a
pp
li
cable
t
o
a
patie
nt'
s
sp
ec
ific
ci
rcu
m
stan
ces,
w
hich
al
so
ai
ds
healt
h
ca
r
e
org
anizat
ion
s
in
t
he
r
e
du
ct
io
n of
cost a
nd subsi
di
zat
ion
.
The
a
uthors
i
n
[2
5
]
pro
posed
a
m
ob
il
e
healt
hcar
e
a
pp
li
cat
ion
that
m
on
it
or
s
a
nd
dia
gnose
s
diabete
s
and
the
seve
rity
of
the
diseas
e.
T
he
a
uthors
pro
pose
us
in
g
a
ne
w
hy
br
id
te
chn
iq
ue
cal
le
d
F
uzzy
R
ul
e
-
base
d
Neural
Cl
assif
ie
r
as
a
decisi
on
s
upport
sys
tem
fo
r
dia
gnos
in
g.
T
he
m
od
ule
f
or
the
de
ci
sion
-
m
akin
g
dat
a
red
ist
ri
bu
te
s
a
reposit
ory
of
da
ta
set
s/reco
rds
of
diabetes
s
ym
pto
m
s
retrieved
f
ro
m
experim
ental
data
from
the
UC
I
reposi
tory,
hosp
it
al
s,
an
d
se
nsors
f
r
om
wear
able
de
vices
at
ta
che
d
to
the
hum
a
n
body.
This
da
ta
is
store
d
in
a
cl
oud
e
nvir
on
m
ent
in
a
Ha
doop
file
syst
em
t
o
ve
rify
scal
a
bili
ty
.
To
ens
ure
the
pr
otect
ion
a
nd
confide
ntial
it
y
of
th
e
patie
nts'
m
edical
reco
r
ds
m
ulti
ple
encr
y
ption
/
dec
ryptio
n
m
et
ho
ds
a
re
im
ple
m
ente
d.
Althou
gh
this
pro
po
se
d
m
ob
il
e
app
li
cat
ion
is
a
decisi
on
s
uppo
rt
syst
e
m
t
hat
easi
ly
enj
oy
s
con
fi
den
ti
al
it
y
and
scal
abili
ty
,
howev
e
r,
it
la
cks
a
necessa
ry
in
te
ll
igent
infl
ue
nce
i
n
pr
e
dicti
on,
as
yo
u
m
us
t
go
th
r
ough
m
ulti
ple
cl
oud
repo
sit
or
ie
s
in
sea
rc
h
of
m
at
ches.
A
n
intel
li
gen
t
fact
or
is
need
e
d
i
n
pr
e
dicti
ng
ne
w
cases
that
do
not
so
le
ly
resu
lt
in
the
Ha
doop
DB
file
syste
m
.
Ba
gg
in
g
Boo
tst
ra
ppin
g
m
ay
increase
bette
r
accura
cy
in
the
decisi
on
m
aking
proce
ss,
in
w
hich
m
or
e
than
on
e
cl
assi
fie
r
is
i
m
ple
m
ented
as
co
ns
ul
ta
nts
in
the d
ia
gnos
i
ng
process
.
The
a
uthors
in
[26]
pr
opos
e
an
in
dustria
l
I
oT
dr
i
ven
heal
thcare
ec
os
yst
e
m
ref
er
red
to
as
Heath
I
oT,
wh
e
re
tw
o
dis
ti
nct
sh
are
hold
ers
are
li
nk
e
d
to
f
or
m
a
co
m
plex
Healt
hIoT
ec
os
yst
em
,
sh
a
reholde
r
s
from
var
yi
ng
par
ti
e
s
co
ns
ist
of,
ph
a
rm
aceuti
cal
and
healt
h
i
ndus
t
ry
orga
niz
at
ion
s,
t
o
pati
ents
an
d
s
peci
al
ist
s
.
Con
ce
ntrati
ng
on
EC
G
analy
sis,
the
auth
ors
us
ed
waterm
ark
i
ng
process
es
and
wav
el
e
t
transfor
m
at
ion
to
secur
e
data
bot
h
di
gital
analy
zed
an
d
a
na
lo
g.
Exam
ples
of
su
c
h
ECG
m
anipu
la
ti
ons
ha
ve
al
so
bee
n
pr
opose
d
by
m
ulti
ple
authors
in
[2
3
,
2
5
,
26
]
.
H
ow
e
ve
r,
a
uthors
i
n
t
his
resea
rc
h
re
m
ark
ably
ha
ve
an
a
dv
a
ntage
du
e
t
o
waterm
ark
ing
for
data
tra
nsm
issi
on
of
EC
G
sig
nal
thr
ou
gh
t
he
cl
ou
d.
Au
t
hors
e
xp
ec
t
us
in
g
su
c
h
H
eal
thIo
T
ecosyst
em
wil
l
no
t
only
al
low
sec
ure
an
d
fast
data
tr
ansm
issi
on
s
be
tween
par
ti
es
bu
t
offer
rea
l
-
tim
e
m
on
it
or
ing an
d av
oid
unnece
s
sary hos
pital
an
d h
um
an
erro
rs.
The
pro
pose
d
Io
T
ser
vice
is
set
up
throu
gh
var
io
us
ty
pes
of
se
ns
ors,
M
2M
o
r
rem
otely
,
for
the
m
on
it
or
in
g
of
th
e
pat
ie
nts’
healt
h,
i.e.,
ECG
for
he
artbeat
dis
ord
er
detect
ion.
T
his
ser
vice
will
no
t
on
ly
ai
d
patie
nts
bu
t
al
so
s
pe
ci
al
ist
s
and
healt
hcar
e
pro
vid
ers
,
the
ser
vice
pr
ovi
des
th
e
abili
ty
to
m
a
neuve
r
betwee
n
ass
or
t
ed
ty
pes
of
dat
a,
scans
(im
agi
ng),
dig
it
al
an
d
anal
og,
th
us
offer
i
ng
a
n
ar
r
ay
of
f
unct
iona
li
ti
es,
su
c
h
as
but
not
lim
it
ed
to
diag
nosis
thr
ough
sym
pto
m
s
analy
sis,
disease
predict
io
n
thr
ough
patie
nt
dat
a
m
on
it
or
ing,
a
nd
prob
i
ng
f
or
i
nfor
m
at
ion
on
sp
eci
fic
sym
pto
m
s
an
d
diseas
es.
F
ur
t
her
m
ore,
this
propose
d
I
oT
serv
ic
e
diag
nosti
c
pr
oce
ss
is
est
ablished
t
hroug
h
the
us
e
of
the
baggin
g
bootstra
p
c
on
ce
pt,
al
lowi
ng
m
ulti
ple
cl
assifi
ers
to
e
xpos
e
the
i
nter
op
e
ra
bili
ty
a
m
ong
a
co
ntrast
of
sym
pto
m
s,
thu
s
,
directi
ng
t
o
the
m
os
t
id
eal
and
favor
a
ble
diag
no
sis
f
ro
m
vari
ou
s
re
po
sit
ori
es
of
the
patie
nts'
data,
pr
ov
i
ding
high
acc
uracy
in
the
de
ci
sion
m
aking
proces
s.
3.
DIAGN
OSTI
C
SE
RVICE
BL
OCK
DI
A
GRAM
The
pro
posed
m
edical
diagnostic
serv
ic
e
em
plo
ys
bo
th
the
Io
T
an
d
the
cl
oud
inf
rastr
uc
tur
e,
ai
m
ing
to
(i)
pro
vi
de
a
cy
ber
-
m
edici
ne
ser
vice
in
the
cl
oud
la
ye
rs
as
softwa
r
e
as
a
serv
ic
e
(S
aaS)
t
o
sh
a
re
al
l
extracte
d
knowle
dge
an
d
fin
dings
betwee
n
diff
e
re
nt
par
ti
es,
(ii)
harness
hybri
d
m
achine
le
arn
in
g
te
chn
iq
ues
to
analy
ze
var
i
ou
s
in
pu
t
data
ty
p
es
as
sym
pt
om
s
and
pr
e
dic
t
the
su
it
able
di
agnosis,
(iii
)
im
pr
ov
e
t
he
acc
ur
acy
of
the
decisi
on
m
aking
pro
cess
by
ap
plyi
ng
dif
fer
e
nt
cl
assifi
ers,
(iv)
offer
var
i
ou
s
f
un
ct
io
nalit
ie
s
to
ai
d
do
ct
or
s
a
nd
pa
ti
ents
searching
f
or
ne
w
diseases
and
unrec
ognized
c
om
bi
nations
of
sym
pto
m
s.
The
ser
vice
has
t
wo
dif
fer
e
nt
m
od
es,
ei
th
er
to
sin
g
in
as
a
doct
or
or
as
a
patie
nt
.
Eac
h
m
od
e
has
it
s
own
f
unct
ion
a
li
ti
es,
wh
ic
h
goin
g
to
be
discusse
d
la
te
r.
As
s
how
n
in
Fig
ure
1,
t
he
ser
vi
ce
con
ta
ins
t
hr
ee
m
ai
n
m
od
ules
.
The follo
wing
su
bse
ct
ions
will
d
isc
uss sel
ect
ed parts
of the
serv
ic
e,
for m
or
e
detai
ls see [
27
]
.
3.1
.
Data pr
ovider
(D
P
)
m
odul
e
This
m
od
ule
c
on
si
ders
va
rio
us
in
put/
ou
t
pu
t
data
interact
io
ns
th
r
ough
the
whole
se
rv
ic
e,
su
c
h
data
i
s
gen
e
rated
fro
m
m
ult
iple
IoT
sources
a
n
d
enjoys
diff
e
re
nt
struct
ur
es
a
nd
ty
pe
s,
see
[27]
f
or
m
or
e
detai
ls.
Wh
e
re,
f
or
ECG
m
easur
ing
,
i
t
is
an
op
ti
on
al
ph
ase
do
ne
by
a
har
dware
,
ar
du
i
no
kit,
wh
ic
h
was
de
sig
ne
d
an
d
dev
el
op
e
d,
see
Fig
ur
e
2(
a
),
to
fit
the
serv
ic
e
as
a
po
rta
ble
de
vice
that
can
be
easi
ly
plu
gged
in.
Us
ually
,
hear
t
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
An
i
nnov
ative
IoT
se
rvi
ce for
med
ic
al d
i
agnosis
(
Safi
a
A
bb
as
)
4921
beats
an
d
bl
ood
press
ur
e
c
onsidere
d
as
an
i
m
po
rtant
pr
e
-
a
naly
sis
te
st
fo
r
patie
nt
seve
rity
conditi
on
s.
It
ai
ds
in
pr
e
dicti
ng
an
d
disc
ov
e
rin
g
t
he
crit
ic
al
cases
su
c
h
as
m
yocard
ia
l
in
far
ct
i
on,
i
f
the
patie
nt
su
f
fers
f
r
om
chest
pain.
T
he
Ard
uino
kit
Fig
ure
2
co
nverts
the
A
nalo
g
data
i
nto
di
gital
one
an
d
draws
the
co
rr
es
pondin
g
BPM
diag
ram
si
m
il
a
r
to
the
ECG
pl
oting
,
as
see
n
in
Fig
ur
e
2(b
).
It
m
easur
es
the
hear
t
beats
pe
r
m
inu
te
(BPM)
and
plo
ts
the
el
ect
r
ocardio
gr
a
ph
(
ECG)
f
or
pri
nt
ing
,
as
do
ct
or
s
can
rea
d,
a
nd
t
hen
t
he
res
ulted
grap
h
is
trac
ed
an
d
analy
zed, by t
he
m
ast
er classif
ie
r
a
gen
t,
to
s
po
t
on the
up
-
norm
al
r
eadin
gs base
d o
n
Ta
ble
1.
Fig
ure
1
.
Dia
gnos
ti
c s
oft
w
are
as a
s
e
r
vice
b
l
ock
d
ia
gr
am
(
a)
(
b)
Fig
ure
2
.
(a
)
T
he devel
op
e
d s
ens
or
,
(
b)
T
he plot
diag
ram
f
or BPM
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
5
,
Oct
ob
e
r 2
020
:
49
18
-
49
27
4922
Table
1
.
S
ta
tus
for
BPM
m
eas
ur
em
ents
BPM
Statu
s
<6
0
pb
m
Slo
w
6
0
>= and
=<10
0
No
r
m
al
>1
0
0
Fast
3.2
.
Master
c
lassifier
ag
e
nt mod
ule
This
m
od
ule
is
respo
ns
ible
of
c
ontrolli
ng
diff
e
re
nt
cl
assifi
ers
based
on
the
ty
pe
of
da
ta
prov
i
de
d
from
the D
P m
odule
as
foll
ows:
-
An
al
ogue
a
nd
scans
data:
bo
t
h
ty
pes
of
dat
a,
ei
ther
anal
ogue
gathe
re
d
by
sensors
or
s
cann
e
d,
se
nt
to
the
cl
assifi
er
CA1
th
r
ough
the
m
ast
er
cl
as
sifie
r
age
nt.
C
A1
a
naly
ze
an
d
im
ple
m
ents
i
m
age
processi
ng
te
chn
iq
ues
i
n
order
t
o
re
veal
the
dis
order
inf
or
m
at
ion
.
A
s
seen
i
n
Fig
ure
3,
CA
1
c
he
cks
if
t
he
data
analo
gu
e
or
Sc
ans.
F
or
t
he
f
orm
er
ty
pe,
it
c
onve
rts
the
ana
logue
into
dig
i
t
al
and
store
s
it
in
the
form
o
f
plo
t
dia
gr
am
i
m
ages.
The
n,
ei
ther
plo
t
di
agr
am
s
or
scans'
i
m
ages
a
re
proces
sed
,
analy
zed,
a
nd
interp
reted
, s
ea
rch
i
ng for ab
norm
al
p
at
te
rn
s.
Fig
ure
3. CA
1 pipeli
ne c
ha
rt
-
Digital
data:
a
ny
ot
her
data
t
ypes,
rather
th
an
a
nalo
gu
e
a
nd
scan
s,
c
ons
idere
d
di
gital
and
m
anipu
la
te
d
us
in
g
dif
fer
e
nt
cl
assifi
cat
ion
al
gorithm
s.
The
m
ast
er
cl
assif
ie
r
age
nt
gath
ers
dif
fer
e
nt
di
gital
data
fr
om
the
DP
m
od
ul
e,
and
distrib
ut
es
the
req
ui
re
d
ta
sk
s
bet
we
en
cl
assifi
ers
from
CA2
to
CA4
,
for
m
or
e
detai
ls see [
27]
.
3.3
.
In
ferenc
e m
odul
e
The
i
nf
e
ren
ce
m
od
ule
is
co
nsi
der
e
d
as
t
he
data
la
ke
f
or
s
ym
pto
m
s,
diag
no
sis
,
diseases
,
kn
ow
le
dg
e
and
pe
rsonal
inf
or
m
at
ion
of
diff
e
re
nt
par
ti
e
s,
associat
e
d
w
it
h
in
-
database
analy
sis.
Wh
e
re,
the
data
is
store
d
in
dif
fer
e
nt
for
m
s
su
ch
as
D
BM
S
struct
ur
e
d
ta
bles,
K
no
wled
ge
base
(
KB),
an
d
m
et
a
-
data
file
s
as
a
bs
tract
descr
i
ption f
or
the pre
viously
stor
e
d kno
wled
ge.
T
he
a
naly
sis o
f
the
data is
done
t
hro
ugh
t
he data
b
as
e en
gine
that
interact
s
with
th
e
cl
assif
ie
r
m
od
ule
to
r
eply
the
querie
s.
I
f
t
he
e
xisted
data
is
no
t
s
uffici
ent
to
ext
ract
th
e
require
d
know
le
dg
e
f
or
the
qu
e
ries,
the
e
ng
i
ne
will
r
un
a
Web
cra
wl
er
to
sea
rch
f
or
the
m
os
t
re
le
van
t
knowle
dge
require
d
to
r
e
ply t
he qu
e
ry.
For
m
or
e d
et
ai
ls ab
ou
t
dif
fer
e
nt
phases
a
nd inte
r
op
e
ra
bili
ty
see
[27].
4.
SERVICE
-
U
S
ER I
NTER
A
CTIO
N
E
X
A
MPLE
Her
e
is
a
n
init
ia
l
ru
n
f
or
the
pro
po
se
d
se
rv
i
ce,
in
w
hich
,
a
real
interface
screen
s
hoots
i
s
entai
le
d
to
sh
ow
the
i
nter
act
ion
of
the
s
erv
ic
e
with
the
us
ers
,
ei
ther
pa
ti
ents
or
do
ct
or
s
.
O
ur
il
lustr
at
ion
s
are
wr
it
te
n
in
it
al
ic
f
on
t.
User
>>
ru
n
the
progr
am
.
As show
n
in
Fi
g
ure
4,
t
he
sta
r
t
u
p me
nu ap
pe
ar
e
d
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
An
i
nnov
ative
IoT
se
rvi
ce for
med
ic
al d
i
agnosis
(
Safi
a
A
bb
as
)
4923
Fig
ure
4. The
s
ta
rt
-
up
window
Now us
ers hav
e fou
r opti
ons, f
or
simp
li
ci
ty
, we wi
l
l sho
w t
he
m
od
e
th
at is su
it
ab
le
for
bo
th u
sers (p
atient and
do
ct
or
s)
.
User
>>
press
m
easur
em
ents b
utto
n
Ser
vice>>
r
ea
ds t
he
se
nsor
sig
nals a
nd con
ve
rt them
into
B
PM
diag
ram
as seen
in Fi
g
ure
5.
Fig
ure
5. BPM
c
ha
rt
Th
e cl
as
sif
ie
r
ag
e
nt wil
l a
na
l
yze the re
adin
gs a
nd
s
pecif
y if
there is
a dis
order.
User
>>
cho
os
e
Diag
nose
Ser
vice>>
Fig
ur
e
6
is s
how
n
.
Fig
ure
6
.
G
e
nd
er
s
el
ect
io
n
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
5
,
Oct
ob
e
r 2
020
:
49
18
-
49
27
4924
User
>>
sup
pos
e that
fem
al
e
is sel
ect
ed
Ser
vice>>
interact
with
the
us
er
t
o
sel
ect
sp
eci
fic
pa
rt
of
the
bo
dy
to
be
dia
gnos
e
d,
and
t
hen
t
he
r
el
at
ed
com
m
on
sy
m
pto
m
s w
il
l be show
e
d.
User
>>
selec
t
so
m
e sy
m
pto
m
s as s
how
n
in
Fig
ure
7.
Fig
ure
7
.
Rel
at
ed
sym
pto
m
s p
resen
at
io
n
If
the
sym
pt
oms
m
atc
h
any
pr
e
-
exi
sti
ng
disease
i
n
the
K
B
wi
th
high
acc
uracy,
t
he
servi
ce
w
il
l
sh
ow
the d
i
agnosti
c
results
.
Ser
vice>>
sho
w
the
relat
ed
diseases
an
d
order
e
d
them
fr
om
hig
he
st
prob
a
bili
ty
t
o
lowe
st,
as
s
how
n
in
Fig
ure
8.
Fig
ure
8. Dia
gnos
is
res
ults
order
e
d
If no
m
atch
f
ound, th
e
servic
e
init
iate t
he
cr
aw
le
r to
se
ar
ch
throu
gh the we
b abo
ut most
r
el
evan
t
diag
nosis.
Ser
vice>>
pres
ent the
res
ults
base
d on cra
wling
,
as i
n
Fi
g
ur
e
9(a) a
nd Fi
g
ure
9(b)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
An
i
nnov
ative
IoT
se
rvi
ce for
med
ic
al d
i
agnosis
(
Safi
a
A
bb
as
)
4925
(a)
(b)
Fig
ure
9
.
(a
)
R
esults d
oes
not
m
at
ch
with s
pe
ci
fic p
re
determ
ined
acc
uracy
,
(
b
)
Re
s
ults ha
ve
b
ee
n fou
nd
Suppose
th
at t
he user is
a
pre
ntice
doctor
, w
ho wa
nts to
k
no
w
sp
eci
fi
c in
formati
on
about s
pecif
ic
sympto
ms
or
disease
Ser
vice>>
en
a
ble user t
o
sea
r
ch fo
r disea
se
by n
am
e o
r
sea
rch f
or sp
eci
fic sym
pto
m
s
User
>>
cho
os
e
sym
pto
m
s n
am
e and
w
rite
“it
ching
t
he
s
ki
n”,
a
s s
how
n
i
n
Fi
g
ure
10(a
)
OR
User
>>
cho
os
e
d
ise
ase a
nd
wri
te
“cancer”, as
shown i
n
Fi
g
ure
10(b)
User >>
pr
ess
Searc
h bu
tt
on
(a)
(b)
Fig
ure
10
.
(a)
Searc
h by sym
pto
m
,
(b)
Sea
r
ch by disease
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
5
,
Oct
ob
e
r 2
020
:
49
18
-
49
27
4926
Ser
vice>>
cra
wl for t
he res
ults an
d
s
how
t
he
p
e
rtinent i
nfo
rm
ation
afte
r
proces
sin
g,
as
s
how
i
n
Fi
g
ure
11(a
)
and Fig
ure
10
(
b)
.
(a)
(b)
Fig
ure
11
.
(a)
Crawl
er
res
ult
s
after
proces
sin
g for the
ide
ntifie
d
sym
pto
m
,
(b)
C
ra
wl
er
res
ult
s
f
or the i
de
ntifie
d disease
5.
CONCL
US
I
O
N
The
pe
r
vasive
ness
of
m
isdia
gnos
es
a
nd
it
s
undesire
d
co
nse
quences
of
wrong
treat
m
e
nt
that
coul
d
le
ad
to
death
or
li
felong
disa
bili
ti
es
pr
opel
researc
hers
in
the
cy
ber
-
m
edici
ne
do
m
ai
n
to
harness
te
ch
nolo
gy
and
reduce
t
he
risk
of
m
is
diag
noses.
T
hi
s
pap
e
r
prop
ose
d
a
ne
w
cy
ber
-
m
edici
ne
serv
ic
e
for
m
edical
diag
nosis
as
SaaS
m
edical
layer
on
cl
oud
I
aaS.
The
se
r
vice
m
ai
nly
con
sist
s
of
thre
e
m
ai
n
m
od
ules,
dat
a
pro
vid
er
m
od
ul
e,
in
w
hich
he
te
rogen
e
ous
t
ypes
of
data
a
re
ente
red
by
diff
e
re
nt
ty
pes
of
us
e
rs
us
in
g
M2M
sens
or
s
data,
scans
,
or
di
gital
data.
M
ast
er
cl
assifi
e
r
age
nt
m
od
ul
e,
w
her
e
m
ulti
ple
cl
assifi
ers
are
i
m
ple
m
ented
to
im
pr
ov
e
dec
isi
on
m
aking
process
a
nd
m
ini
m
iz
e
the
m
i
sd
ia
gn
os
e
ris
ks.
I
nf
e
ren
ce
m
odule,
that
con
ta
in
s
th
e
data
en
gin
e
a
nd
d
at
a
sou
rce
s
for
al
l
us
er
s
a
nd
n
ee
ded
d
ise
ases
inf
or
m
at
i
on,
it
us
es
cra
wler
to
search
f
or
unre
cognized
c
om
bin
at
io
n
of
sy
m
pto
m
s
and
adap
t
it
sel
f
wit
h
new
fin
dings.
Desp
it
e
the
propose
d
serv
ic
e
is
in
the
early
i
m
p
l
e
m
entat
ion
ph
ase,
it
aims
t
o
ve
rify
the
avail
abili
ty
,
on
dem
and
,
secu
re
data
transm
issi
on
,
and
m
or
e
acc
ur
at
e
diag
no
si
s,
not
only
for
patie
nts
bu
t
al
so
f
or
s
pec
ia
li
st
and
heal
th
care
pro
vid
er
s.
Mo
r
eov
e
r,
the
se
rvi
ce
of
fe
rs
va
riou
s
functi
ons
to
ad
d
m
or
e
fle
xib
il
it
y
su
ch
as,
but
no
t
lim
ited
to
,
diag
nosis
thr
ough
sym
pto
m
s
analy
sis,
dise
ase
pr
e
dicti
on
thr
ough
patie
nt
data
m
on
it
ori
ng
,
a
nd
pro
bi
ng
for
inf
or
m
at
ion
on
sp
eci
fic
sym
pto
m
s
and
disea
ses.
I
n
the
fu
t
ur
e
,
detai
ls
of
pr
e
processi
ng,
i
m
ple
m
entat
io
n
an
d
decisi
on m
aking
processes
are
goin
g
to
b
e
d
i
scusse
d
a
nd a
na
ly
zed in
detai
ls.
ACKN
OWLE
DGME
NT
T
h
i
s
r
e
s
e
a
r
c
h
w
a
s
fu
nde
d
b
y
the
D
e
a
n
s
h
i
p
o
f
S
c
i
e
n
t
i
f
i
c
R
e
s
e
a
r
c
h
a
t
P
r
i
n
c
e
s
s
N
o
u
r
a
h
b
i
n
t
A
b
d
u
l
r
a
h
m
a
n
U
n
i
v
e
r
s
i
t
y
t
h
r
o
u
g
h
t
h
e
F
a
s
t
-
t
r
a
c
k
R
e
s
e
a
r
c
h
F
u
n
d
i
n
g
P
r
o
g
r
a
m
.
REFERE
NCE
S
[1]
O.
Verm
esa
n
and
P.
Friess
,
“
Digit
izing
The
Industr
y
Int
ern
e
t
of
Thi
ngs
Co
nnec
t
ing
the
P
hy
sic
al,
Digit
a
l
a
nd
Virtua
l
W
orlds
,
”
Rive
r
Publisher
s,
2016.
[2]
S.
Kum
ar,
e
t al
.
,
“
Mobile
health:
Revol
uti
on
iz
ing
hea
l
thc
ar
e thr
ou
gh
tra
nsdisc
ipl
in
ar
y
rese
arc
h
,”
C
omputer
,
vol
.
46
,
no.
1
,
pp
.
28
-
35
,
2013.
[3]
P.
Chat
te
r
jee
a
nd
R.
L.
Arm
ent
ano
,
“
Inte
rn
et
of
thi
ngs
for
a
sm
art
and
ubiqui
tou
s
eheal
th
sy
st
em
,”
i
n
7
th
Inte
rnational
Co
nfe
renc
e
on
Co
mputati
onal
In
telli
gen
ce
and
Co
mm
unic
ati
on
Ne
tworks
(
CICN
)
,
Jabal
pur,
Ind
ia
,
pp.
178
-
183
,
20
15.
[4]
P.
Verm
a
and
S.
K.
Sood
,
“
Clou
d
-
ce
ntr
ic
Io
T
b
a
sed
disea
se
dia
g
nosis
hea
lt
h
ca
r
e
fra
m
ework
,”
Jo
urnal
of
Ambi
en
t
Inte
lligen
ce and H
umanized
Computing
,
vol
.
9
,
n
o.
5
,
pp
.
1293
-
1
309,
2017
.
[5]
D.
Hail
e
y
,
e
t
al.
,
“
A
rev
ie
w
of
guide
li
n
es
for
ref
err
al
of
p
at
i
ent
s
to
slee
p
l
abor
a
t
o
rie
s
[technol
og
y
rep
ort
no
55]
,”
Techni
ca
l
report
,
Ottawa:
Canad
ian
Coordinati
n
g
O
ffi
c
e
for
Hea
lt
h
Te
chnol
ogy
Assess
ment
,
200
5.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
An
i
nnov
ative
IoT
se
rvi
ce for
med
ic
al d
i
agnosis
(
Safi
a
A
bb
as
)
4927
[6]
G.
L.
Kreps
and
L.
Neuh
ause
r
,
“
New
dire
ctions
i
n
eHe
alth
comm
unic
a
ti
o
n:
Oppo
rtuni
ties
and
challe
ng
es
,”
Pat
i
ent
Educ
ati
on
and
C
ounseli
ng
,
vol
.
7
8,
no
.
3
,
pp
.
329
-
336,
2010
.
[7]
S.
M.
R.
Islam,
et
al.,
“
The
in
t
ern
et
of
th
ings
for
hea
lt
h
ca
r
e:
A
comprehe
nsive
surve
y
,”
I
EE
E
Ac
ce
ss
,
vol
.
3,
pp.
678
-
708
,
20
15.
[8]
D.
La
ke
,
e
t
al.,
“
Inte
rne
t
of
th
i
ngs:
Archi
tectur
al
fra
m
ework
f
or
ehe
a
lt
h
sec
ur
ity
,”
Journal
of
ICT
,
vol.
3&4
,
pp.
301
-
328
,
20
14.
[9]
M.
Sham
im
H.
and
G.
Muham
m
ad
,
“
Cloud
assisted
industri
a
l
int
ern
et
of
thi
n
gs
(IIoT
)
ena
bl
e
d
fra
m
ework
for
hea
l
th
m
onit
orin
g
,”
Comput
er
Ne
tworks
Journal
,
vol.
101
,
pp
.
192
-
202,
2016
.
[10]
Z.
Kirta
v
a,
e
t
al
.,
“m
hea
lt
h
for
ca
rdiac
patie
nts
te
le
m
onit
ori
ng
and
int
egr
a
t
ed
ca
re
,”
i
n
15
th
Inte
rnationa
l
Confe
renc
e
on
e
-
Healt
h
N
et
wo
rking,
App
li
ca
tions
and
Serv
ic
e
s
(Heal
thc
om
2
013)
,
Lis
bon,
Portugal
,
pp
.
21
-
25,
2013.
[11]
J.
Moham
m
ed,
et
al
.
,
“
Inte
rne
t
of
thi
ngs:
Remote
pat
i
ent
m
onit
ori
ng
using
w
eb
se
rvic
es
and
cl
oud
computing
,”
i
n
IEE
E
Int
ernational
Confe
renc
e
on
Inte
rnet
of
Things
(
iTh
ings
2014)
an
d
IEE
E
Gr
ee
n
Computing
and
Comm
unic
ati
ons (
Gr
ee
nCom
)
and
IEEE
C
ybe
r,
P
hysic
al
and
So
cial
Computing
(
CPSCom)
,
pp.
25
6
-
263
,
20
14
.
[12]
S.
Uribe
and
R.
J.
Marino
,
“
Inte
r
net
and
informat
ion
technolog
y
u
se
b
y
dental
stud
ent
s in
Ch
il
e
,”
E
urpoean
Journal
of
Denta
l
Edu
ca
ti
on
,
vol. 10
,
no
.
3,
pp.
162
-
168,
2006.
[13]
M.
Ri
y
adh
,
et
a
l.
,
“
Archi
t
ec
t
ing
the
Inte
rne
t
of
Thi
ngs:
Stat
e
o
f
the
Art,
”
Boo
k
cha
pte
r
in
Ro
bots
and
Sensor
Clouds,
Springer
Inte
rn
at
ion
al Publishing
,
vol. 36, pp. 55
-
75,
2015
.
[14]
W
.
Li
n,
e
t
a
l.,
“
A
cl
oud
-
base
d
fra
m
ework
for
hom
e
-
dia
gnosi
s
service
over
bi
g
m
edi
cal
da
ta
,”
The
Journal
of
Syste
ms
and
Sof
t
ware
,
vol
.
102
,
p
p
192
-
206,
2014
.
[15]
I.
K.
Indra
jit
an
d
S.
Nangpa
l.
“C
y
ber
m
ed
ic
in
e
and
c
y
ber
he
al
t
h
ca
re
,
r
eview
of
m
edi
ci
ne
on
the
int
ern
e
t
,”
M
ed
i
cal
J
ournal,
Armed Forces
India
,
vo
l.
57
,
no
.
3
,
pp
.
2
15
-
220,
2001
.
[16]
I.
Ishaq,
e
t
al.,
“
Inte
rne
t
of
th
ings
virt
ual
ne
t
works
:
Bringi
ng
net
work
virt
ua
li
z
at
ion
to
r
esourc
e
-
constr
ai
n
ed
devi
c
es
,”
i
n
IE
E
E
Inte
rnationa
l
Confe
renc
e
on
Gr
ee
n
Computing
and
Comm
u
n
ic
ati
ons
(
Gr
ee
nCom)
,
Be
sançon,
Paris
,
pp.
2
9
3
-
3
00,
2012
.
[17]
M.
Hass
ana
liera
gh
,
e
t
a
l.,
“
He
alt
h
m
onit
oring
an
d
m
ana
gement
u
sing
int
ern
et
-
of
-
thi
ngs
(IoT
)
s
en
sing
with
c
loud
-
base
d
proc
essin
g:
Opportunities
and
cha
l
le
nges
,”
i
n
IE
EE
In
ter
nati
onal
Conf
e
renc
e
on
Serv
i
c
es
Computing
,
pp.
285
-
292
,
20
15.
[18]
P.
S.
J.
Kum
ar
a
nd
A.
S.
Chai
tra
,
“
A
Survey
on
Cloud
Com
puti
ng
base
d
Hea
lt
h
Care
for
D
ia
be
t
es:
Anal
y
s
is
and
Diagnosis
,”
i
n
I
OSR
Journal
o
f Com
pute
r E
ngin
ee
ring
(
IOSR
-
JC
E)
,
v
ol. 17,
no.
4
,
pp
.
109
-
117
,
2
015.
[19]
G.
E
y
senb
ac
h
,
et
al.,
“
Shopping
aro
und
th
e
int
ern
et
tod
a
y
and
tomorro
w:
towar
ds
the
m
il
le
nnium
of
c
y
b
ermedi
c
ine,
”
Brit
ish
Me
di
cal
Journal,
vo
l. 31
9,
no
.
7220
,
p
.
1
294,
1999
.
[20]
T.
N.
Arvani
t
is,
et
al.,
“
Towa
rds
the
Mill
enni
um
of
Cy
b
ermedi
c
i
ne
,
MEDN
ET
99
Abs
tra
ct
Book,
”
The
4th
W
orld
Congre
ss
on
the
Inte
rne
t
in
Medi
ci
ne
,
1999
.
[21]
P.
M.
Ku
m
ar,
et
al
.
,
“
Cloud
and
IoT
base
d
disea
se
pre
di
ct
ion
and
dia
gnosis
sy
stem
for
hea
l
th
ca
re
using
fuz
z
y
neur
al c
l
assifier
,”
Fu
ture
Gen
erati
on
Comput
er
S
yste
ms
,
vol
.
86
,
pp.
527
-
534
,
20
18.
[22]
Y.
Li
,
e
t
al.,
“
En
abl
ing
he
al
th
m
o
nit
oring
as
a
ser
vic
e
in
th
e
cl
oud
,”
i
n
IE
EE
/
ACM
7th
Int
ernati
on
al
Confe
renc
e
o
n
Util
ity
and
Clou
d
Computing
,
London,
UK
,
pp.
1
27
-
137
,
2014
.
[23]
W
.
Li
n,
W
.
Dou
,
Z.
Zhou
and
C.
Li
uc.
A
cl
oud
-
b
ase
d
fra
m
ework
for
hom
edi
agnosis
servic
e
over
big
m
edi
ca
l
data.
The
Journa
l
of
Sy
stems
and
Soft
ware
,
El
sev
ie
r
,
v
ol.
102
,
pp
192
–
206,
2014
.
[24]
H.
Oh,
et
al
.
,
“
W
hat
is
eHe
al
th
(3):
A
sy
stema
ti
c
rev
ie
w
of
publi
shed
definit
io
ns
,”
Journal
of
Me
dic
a
l
Inte
rne
t
Re
search
,
vol
.
7
,
no.
1,
pp.
1
-
11,
2005.
[25]
Y.
Hu
and
G.
Bai
,
“
A
s
y
stem
at
i
c
li
t
erature
r
evi
e
w
of
c
loud
computing
in
eHe
alth
,”
Health
Informatic
s
-
A
n
Inte
rnational
Jo
urnal
(
HIIJ)
,
vol.
3,
no
.
4
,
pp
.
11
-
20,
2014
.
[26]
L.
Hu
,
et
al.,
“
Soft
ware
defined
hea
l
thc
ar
e
net
w
orks
,”
i
n
IEE
E
Wirel
ess
Communi
cat
ions
,
vol.
22,
no.
6,
pp.
67
-
75,
2015.
[27]
S.
Abbas,
“
C
y
be
r
-
m
edi
ci
ne
servi
ce
for
m
edi
ca
l
di
agnosis
base
d
on
IoT
and
cl
oud
infra
struct
ur
e,
”
The
Inte
rnational
Confe
renc
e
on
A
rtif
ic
a
l
In
te
l
li
gen
t
and
C
omput
er
V
ision
(
AICV
’20
)
,
pp.
617
-
627
,
2
4
Mar.
2020.
BIOGR
AP
H
Y
O
F
AU
TH
OR
Safi
a
Abbas
wo
rks
as
associa
t
e
profe
ss
or
in
the
Depa
rtment
of
Com
pute
r
Scie
n
ce
,
Fa
cult
y
of
Com
pute
r
and
In
form
at
ion
Scie
n
ce
s,
Princ
ess Nourah
bint
abdul
r
ahman
Univer
sit
y
,
KS
A,
during
2019
-
2020,
and
Univer
sit
y
of
Ain
Sham
s,
Cai
ro,
Eg
y
pt
dur
ing
2016
-
2018.
During
2006
-
2011,
s
he
rec
e
ive
d
the
Ph.D.
from
the
Gradua
te
Schoo
l
of
Scie
nc
e
and Te
chno
log
y
,
Nii
gat
a
Univ
ersity
,
Japa
n
.
A
strong
the
m
e
of
her
w
ork
is
in
the
sw
arm
opti
m
iz
ers,
and
sec
uri
t
y
in
cl
oud,
Med
ical
Diagnosis
using
m
ac
hine
learni
n
g
and
D
at
a
m
ining.
Evaluation Warning : The document was created with Spire.PDF for Python.