Indonesi
an
Journa
l
of El
ect
ri
cal Engineer
ing
an
d
Comp
ut
er
Scie
nce
Vo
l.
1
3
,
No.
3
,
Ma
rch
201
9
, p
p.
8
6
1
~
8
6
5
IS
S
N: 25
02
-
4752, DO
I: 10
.11
591/ijeecs
.v1
3
.i
3
.pp
8
6
1
-
8
6
5
861
Journ
al h
om
e
page
:
http:
//
ia
es
core.c
om/j
ourn
als/i
ndex.
ph
p/ij
eecs
Hunger
and stre
ss monit
or
in
g system usin
g galv
anic skin
response
Utta
r
a Go
gate
1
, Jagdish B
akal
2
1
chol
ar
Th
adomal
Shah
ani E
ngi
nee
ring
Coll
ege,
Mum
bai
Univer
sit
y
,
Ind
ia
2
Shivaj
ira
o
S
Jo
ndhal
e
Coll
ege o
f
Engi
n
ee
ring
,
Mum
bai
Univer
sit
y
,
India
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Sep
21
, 201
8
Re
vised
N
ov
2
7
, 2
018
Accepte
d
Dec
2
2
, 201
8
In
the
world,
roughl
y
m
ore
than
3
m
il
li
on
peo
ple
are
aff
ec
t
ed
b
y
m
uscle
power
loss
diseas
es
which
c
ause
la
ck
o
f
comm
unic
ation.
Furth
er,
t
his
la
ck
of
comm
unic
at
ion
i
s
a
seve
re
problem
with
hunger
and
stress,
which
som
et
imes
prove
s
to
be
fat
al.
It
is
essential
to
m
onit
or
ph
y
sic
al
par
amet
ers
such
as
hunger
and
m
ent
al
str
ess
of
t
he
pa
ti
en
t.
To
observe
hunger
and
str
ess
,
gal
van
ic
skin
re
spons
e
(GS
R)
s
ensor
is
a
m
ost
pre
fer
red
m
et
ho
d
because
i
t
m
ea
sures
the
co
nduct
an
ce
of
skin
due
to
sw
ea
ting.
W
e
have
de
signed
new
hunger
and
stres
s
m
onit
oring
s
ystem
(HS
-
MS
)
tha
t
h
el
ps
pa
tient
s
and
th
ei
r
ca
re
ta
k
ers
to
m
onit
or
stress
an
d
hunger
l
evel.
HS
-
MS
is
two
el
e
ct
rod
e
s
s
y
stem
tha
t
at
t
a
che
s
to
two
fin
ger
s
(inde
x
and
m
iddl
e)
of
the
pat
ie
n
t
and
m
onit
ors
starva
ti
on
and
te
nsion.
Acc
ording
to
the
response
of
the
sensor
,
ca
re
ta
k
ers
ca
n
i
dent
if
y
hunger
a
nd
eve
n
the
m
e
nta
l
stress
conditions
of
the
pat
i
ent
.
In
a
ll,
35
pat
ie
n
ts
were
exa
m
ine
d
using
HS
-
M
S
sy
stem
and
20
of
the
m
were
used
as
a
cont
ro
l.
O
ver
all
s
y
st
em
accura
c
y
w
as
foun
d
out
to
b
e
86.
6%
and
r
esponse
ti
m
e
of
H
S
-
MS
5
sec
onds
.
Sensiti
v
ity
wit
h
respe
ct
t
o
hunger
and
str
e
ss
was
about
99.
9%
and
73
.
3%
respe
ctive
l
y
.
T
hus
HS
-
MS
ca
n
be
ide
n
ti
fi
ed
as
a
hel
pful
tool
to
provide
comfort
to
the
patient
ea
sil
y
and
ca
n
b
e
used
in
h
ospita
ls or
hom
es
with
lon
ely
el
d
erly
p
eopl
e
.
Ke
yw
or
ds:
Galva
nic s
kin
r
esp
on
se
Hun
ger detect
ion
Stress m
easur
e
m
ent
Copyright
©
201
9
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
:
Utta
ra
Gogate
,
Re
search
schol
ar T
hadom
al
Sh
aha
ni E
ngine
erin
g
Coll
e
ge,
Mum
bai U
niv
e
rsity
, Ind
ia
.
+91
-
998702
6692
Em
a
il
:
uttara.go
gate
16@
gm
ail.co
m
1.
INTROD
U
CTION
In
the
hosp
it
a
l,
va
rio
us
ph
y
siolo
gical
pa
ra
m
et
ers
of
the
patie
nt
a
re
m
on
it
or
ed
c
on
ti
nu
ously
by
trai
ned
nu
rsing
sta
ff
an
d
doct
or
s
.
Vita
l
body
par
am
et
ers
s
uch
as
te
m
perat
ur
e,
hea
rt
rate,
a
nd
pulse
ra
te
an
d
blood
ox
y
gen
l
evel
of
the
patie
nt
are
to
be
c
on
ti
nu
o
usl
y
m
on
it
ore
d
so
tha
t
healt
h
conditi
on
of
the
patie
nt
can
be
kep
t
unde
r
co
ntro
l.
D
ue
to
certai
n
m
usc
le
power
loss
diseases
s
uc
h
as
Am
yotrophic
la
te
ral
scl
ero
sis
(A
L
S)
, S
to
ke,
Tay
-
Sach
s
dise
ase,
patie
nts b
e
com
e
hig
hly
de
pende
nt
on
th
ei
r
relat
ives
or care
-
t
ak
ers.
In
m
any
cases,
patie
nts
even
ca
nnot
ex
pr
ess
hunger
or
dem
and
f
ood.
In
s
uch
case
s
,
the
hunger
a
nd
stress
m
on
it
or
i
ng
sys
-
te
m
(H
S
-
MS)
syst
em
ca
n
help
ca
re
-
ta
ke
rs
or
hosp
it
al
at
te
nd
ants
in
find
i
ng
out
patie
nts
hu
ng
e
r.
Ev
en
HS
-
MS
syst
e
m
ca
n
hel
p
these
ca
re
-
ta
ke
rs
fin
din
g
a
ny
stress
on
patie
nt,
w
ho
cannot
ex
press
feeli
ng
s
li
ke
f
ear
or
irrita
ti
on
.
Me
ntal
conditi
on
of
the
patie
nt
is
al
so
i
m
po
rta
nt
f
or
fast
rec
ov
e
r
y
of
a
patie
nt.
Fo
r
e
xam
ple,
f
or
th
e
hear
t
patie
nt,
a
ny stress
can
lead to
a se
rio
us
conditi
on l
ike t
he heart at
ta
ck
.
Stress
a
nd
hunger
ca
n
be
m
e
asur
e
d
with
a
si
m
ple
syst
e
m
com
pr
isi
ng
of
a
Galva
nic
S
kin
Re
spo
nse
sens
or
. I
t
is
ob
serv
e
d
that
s
kin
secret
m
or
e
s
weat
in
stresse
d
co
ndit
ion
s
.
Galva
nic
sk
i
n
r
esp
on
se
se
nsor
can
be
us
e
d
to
ide
ntif
y
any
su
ch
i
n
ci
den
ce
s.
P
hysio
log
ist
s
belie
ve
in
the
fact
that,
in
increase
d
st
ress
co
ndit
ion
m
any
of
the
f
ollow
i
ng
sym
pto
m
s
are
obser
ved
i
n
patie
nts,
li
ke
ra
pid
breat
hing,
i
ncr
ea
sed
he
art
beat
rate,
excess
sweati
ng, dr
op
in bo
dy tem
per
at
ur
e a
nd tensi
on on m
us
cl
es.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
1
3
, N
o.
3
,
Ma
rc
h
201
9
:
8
6
1
–
8
6
5
862
Ma
ny
researc
he
rs
ha
ve
co
ntri
bu
te
d
in
de
vel
op
m
ent
of
we
arab
le
instr
ume
nts
m
easur
ing
the
hu
m
an
stress.
T
hey
ad
apted
diff
e
re
nt
te
chn
iq
ues
li
ke
com
pu
te
r
vis
ion
te
ch
niques
[1
]
to
e
xtract
the
vis
ual
evid
ence
s
li
ke
facial
e
xpressio
ns
,
wea
r
able
em
otion
m
on
it
or
ing
usi
n
g
s
kin
te
m
per
at
ur
e
,
s
kin
c
onduct
ance
,
a
nd
pulse
wav
e
[
2].
Als
o
qu
a
ntifyi
ng
st
ress
le
vels
bas
ed
on
m
easur
e
s
of
Hea
rt
-
Ra
te
Var
ia
bili
ty
(H
RV
)
us
i
ng
re
li
able,
high
-
pr
eci
si
on
instru
m
entat
ion
an
d
sync
hro
nized
m
easur
e
m
ents
[3
,
4],
usi
ng
wear
a
ble
instr
um
ents
li
ke
glove
con
ta
ini
ng
s
kin
te
m
per
at
ur
e
and
G
SR
se
nsor
we
re
repo
rted
[5,
6].
Re
posit
ion
a
ble
C
he
st
Stra
ps
Syst
e
m
fo
r
card
ia
c
a
nd
res
pirati
on
rec
ord
ing
s
a
re
re
ported
rece
ntly
[7
]
.
Ma
ny
oth
er
re
searche
rs
us
ed
m
et
ho
ds
to
m
easur
e
any
on
e
or
a
com
bin
at
ion
of
diff
e
re
nt
body
par
am
et
ers
li
ke
hear
t
rate,
re
sp
irat
io
n
rate,
blood
pr
ess
ure
or
sk
i
n
cond
uctance
et
c.
[8
-
13
]
.
Most
of
a
bove
m
ent
ion
e
d
li
te
ratur
e
work
is
f
oc
use
d
on
stress
m
easur
em
ent
an
d
ve
ry
few re
ports
on
hunger
m
easur
e
m
ent.
All
the
researc
h
wor
k
re
porte
d
in
li
te
r
at
ur
e
was
done
us
i
ng
m
or
e
tha
n
one
se
nsor
or
m
et
ho
ds
an
d
m
os
tly
in
fu
ll
y
con
tr
olled
env
i
ronm
ent.
In
the
li
te
ratur
e
m
any
kin
d
of
hunger
are
e
nl
ist
ed
li
ke
ph
y
sic
al
,
e
m
otion
al
et
c.
w
hich
ca
n
be
detect
ed
by
ps
yc
ho
l
og
ist
s
us
in
g
ps
yc
holo
gical
te
ch
ni
qu
e
s
a
nd
ther
apies.
In
physi
ology
an
at
te
m
pt
of
hunger
detect
ion
is
m
ai
nly
done
t
o
co
ntr
ol
ob
e
sit
y
or
dem
entia
in
li
ter
at
ur
e
.
They
use
d
m
et
hods
li
ke
obs
erv
i
ng
i
ntake
patte
rn
s
,
by
quest
io
ning
[
14
]
,
and
sug
gesting
hunge
r
scal
e
[15]
.
No
s
pecific
m
et
ho
d
is
sugg
e
ste
d
to
de
t
ect
hu
nger
f
or
patie
nts
or
el
der
ly
peop
le
who
are
total
ly
dep
e
ndent
on
ot
her
s.
In
t
his
pa
per
we
hav
e d
esc
rib
ed
de
sig
n
an
d
i
m
ple
m
entat
ion
of
H
S
-
MS
sys
tem
wh
ic
h
is
a
n
at
te
m
pt
to
detect
hu
m
an
stress
an
d
hung
er
us
in
g
s
kin
c
onduct
ance.
G
SR
sens
or
is
use
d
to
m
ea
su
re
sk
in
co
nducta
nce
by
at
ta
ching
se
nsor
to
fin
ger
ti
ps
of
a
pe
rs
on.
GS
R
sig
nals
are
colle
ct
ed
and
se
nd
to
c
on
t
ro
ll
er
us
in
g
inbuil
t
analo
g
to
dig
it
al
con
ve
rter
.
S
ign
al
is
then
s
ent
to
PC
us
in
g
wireless
c
om
m
un
ic
at
ion
as
sh
ow
n
in
Figure
1
HS
MS
a
rch
it
e
ct
ur
e.
This
m
et
hod
is
ba
sed
on
t
he
fact
t
ha
t
GS
R
of
a
person
va
ries
in
c
ase
of
any
stre
ss
an
d
hunger
tha
n
no
rm
al
read
in
gs
,
as
sh
ow
n
in
Figure
2
.
O
ur
pr
oto
ty
pe
H
S
-
M
S
syst
e
m
is
a
wear
a
ble
an
d
wireless
syst
e
m
that can
be use
d by el
der
ly
people a
nd p
at
ie
nts
on t
heir
own
.
Figure
1. H
S
-
MS sy
ste
m
arc
hitec
ture. R
ead
ing
s
are
ac
quir
ed usin
g GSR
sens
or
an
d
c
on
ver
te
d
in
to dig
it
al
sign
al
us
in
g A
r
du
i
no N
a
no
on boar
d AD
C
. E
SP 826
6 node
MC
U
is t
he
u
s
ed
to
tra
ns
m
it
t
he data
ove
r W
i
-
Fi
towa
rd
s
PC
Figure
2 Ard
uin
o
Na
no is
us
e
d
to
acq
uire
r
e
adin
gs
from
G
SR sens
or a
nd
sen
d
it
over
W
i
Fi us
i
ng ESP8
266.
The
si
gn
al
c
on
diti
on
in
g
ci
rcu
i
t i
s u
se
d for int
erf
aci
ng
GS
R
s
ens
or
with
A
t
o D c
onve
rter
of Ard
uino
Na
no.
GS
R se
nsor
is
a senso
r
f
or sta
nd
a
r
d
rea
dout
of the c
onduct
ance
value
from
the scal
e.
(B)
Im
age of
exp
e
rim
ental
setup
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
Hun
ger and
stress mo
nitorin
g syst
em usi
ng galva
nic skin
re
sp
onse
(
Uttar
a G
ogate
)
863
2.
RESEA
R
CH MET
HO
D
2.1.
S
ystem
D
escri
pt
i
on
HS
-
M
S
syst
em
con
sist
s
of
sens
or
s
a
nd
A
rduin
o
bo
a
r
d
to
colle
c
t
an
d
process
data
f
ro
m
sensor
.
The
data
is
th
en
sent
to
c
om
pu
te
r
syst
e
m
fo
r
st
or
a
ge
an
d
dis
play
u
sin
g
wireless
c
om
m
un
ic
at
ion
.
Figure
3
sh
ows
t
he
H
S
-
MS
syst
e
m
(A
)
an
d
the
e
xpe
rim
ental
set
up
(B).
Syst
em
c
on
sist
s
of
GSR
sensor
wh
ic
h
ta
ke
s
sk
in
co
nducta
nc
e
as
se
nsor
in
pu
t
a
nd
pro
duces
anal
og
vol
ta
ge
sig
nals
as
sens
or
ou
t
pu
t.
Sig
nal
c
onditi
on
i
ng
ci
rcu
it
co
ns
ist
s
o
f
an am
plifier
and
volt
age
di
vid
er
circuit t
hat sh
i
fts volt
a
ge
le
vel
of GS
R sens
or b
et
we
en 0 to
5V
DC
.
Si
gn
al
co
ndit
ion
er
outpu
t
is
the
n
c
onve
rted
to
di
gital
sign
al
s
us
i
ng
i
nbuilt
A
to
D
c
onve
rter
ci
r
cuit
of
Ardu
i
no
Na
no
boar
d.
O
utp
ut
f
r
om
Ar
duin
o
Na
no
boar
d
is
the
n
sent
to
com
pu
te
r
us
i
ng
wireless
com
m
un
ic
at
ion
m
od
ule
WiFi
ESP
82
66 no
de
MC
U.
Figure
3. Stan
da
rd v
al
u
es at t
he
outp
ut of
GSR
sen
s
or
Figure
3
(
A)
D
iffer
e
nt
val
ues
ob
ta
ine
d
with
the
stresse
d
conditi
on
a
nd
r
el
ax
co
nd
it
io
n.
In
stres
sed
c
onditi
on
the
a
ver
a
ge
value
goes
belo
w
300
and
sta
ndar
d
de
viati
on
is
ve
r
y
high.
In
rela
x
c
onditi
on
the
valu
e
is
above
400
with
ne
gligi
ble
sta
nd
a
rd
de
vi
at
ion
.
Fi
gure
3
(B)
I
ndic
at
es
Hun
ger
sta
tu
s
reado
uts
from
HS
-
M
S
syst
e
m
.
It
is
alw
ay
s
obser
ve
d
that
durin
g
h
unge
r
the
value
fall
s
belo
w
40
0
but
rem
ai
ns
above
300
with
sli
gh
t
sta
nd
a
rd
de
via
ti
on
value
.
Fi
gure
3
(C)
Stres
s
le
vels
a
ff
ect
e
d
by
m
edita
ti
o
n,
dee
p
br
eat
hin
g
an
d
stu
dyin
g.
As
per
t
he
syst
em
reado
ut
we
fou
nd
m
edita
ti
on
and
dee
p
br
eat
hing
tra
ns
la
te
s
stressed
t
o
rela
x
sta
te
.
O
n
th
e
oth
e
r
hand,
stu
dyin
g
cha
ng
e
s
rela
x
sta
te
to
stresse
d
sta
te
.
Fig
ur
e
3
(
D)
Ef
fect
of
tim
e
of
the
day
on
syst
em
read
out.
At
7
AM
a
nd
11
AM
m
os
t
of
the
sub
j
ec
ts
wer
e
f
ound
hung
ry.
At
9.00,
14.00
a
nd
22.
00
al
m
os
t
al
l
the
read
i
ng
s
we
re
fou
n
d
into
a
r
el
axed
cat
eg
ory
.
At
19.
00
P
M
m
os
t
of
th
e
te
st
cases
were
stresse
d
a
nd
few
of
them
wer
e
ly
ing
into
a
hung
r
y
sta
te
.
Ther
e
was
no
sig
nific
ant
diff
e
re
nce
as
per
the
plac
e
(r
esi
de
nce,
c
ollege,
office) c
oncer
ne
d.
2.2.
E
xp
eri
m
ent
al
Proce
du
re
The
e
xperim
e
ntati
on
f
or
str
ess
an
d
hunge
r
m
easur
em
ent
is
co
nducte
d
on
a
gro
up
of
35
healt
hy
people
(eac
h
f
or
hu
ng
e
r
an
d
for
stress)
in
re
gu
la
r
en
viron
m
ent
at
diff
ere
nt
locat
ion
s
li
ke
residen
ce
,
of
fice
and
colle
ge
cam
pus.
Test
wa
s
car
ried
on
pe
op
le
from
age
gro
up
of
18
t
o
45
and
al
m
os
t
equ
al
num
ber
of
m
al
es
and
fem
al
es.
Ou
t
of
35
sam
ples
20
we
re
us
e
d
an
d
co
ntr
ol
(A
l
read
y
known
as
hungr
y
or
stresse
d)
and
15
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
1
3
, N
o.
3
,
Ma
rc
h
201
9
:
8
6
1
–
8
6
5
864
read
i
ng
s
wer
e
us
e
d
for
blin
d
te
sti
ng
wh
ic
h
was
then
ver
ifi
ed
from
us
er
feed
bac
k
gi
ven
pr
io
r
to
the
te
st
ing.
The
read
i
ng
w
ere
ta
ke
n
i
n
diff
e
ren
t
c
onditi
on
s
s
uch
as
be
fore
a
nd
after
m
editation,
de
ep
br
eat
hing,
w
hile
stud
yi
ng
dif
fic
ult
su
bject
s
at
diff
e
re
nt
tim
e
l
ike
early
m
or
ni
ng
,
la
te
night,
befor
e
a
nd
aft
er
lun
c
h
tim
e
and
at
diff
e
re
nt p
la
ce
s also like
co
ll
e
ge
cam
pu
s a
nd at
ho
m
e.
2.3.
Algori
thm
f
or
H
un
ger
a
nd
St
re
ss
De
tection
a)
Atta
ch
se
nsor
input li
ds t
o fin
ger
s
ti
ps
of p
e
r
so
n.
b)
GS
R
values
ar
e colle
ct
ed, co
nv
e
rted
fro
m
an
al
og to
d
i
gital
b
y
Ardu
i
no
usi
ng
serial
co
m
m
un
ic
at
ion
.
c)
Pr
oc
esse
d Ard
uino
ou
t
pu
t i
s
s
ent to la
ptop/c
om
pu
te
r usin
g wirele
ss c
omm
un
ic
at
io
n ASP
8266
W
i
Fi.
d)
If
G
SR
value
i
s
above
40
0
th
en
set
Stat
e=
relax,
el
se
if
G
SR
value
li
es
in
betw
een
300
to
400
the
n
se
t
Stat
e= hun
gr
y
el
se set Sta
te
=
stressed
.
In
ab
ove
al
go
r
it
h
m
ste
p
4
va
lues
w
ere
deci
ded
after
c
on
tr
ol
ex
pe
rim
ent
with
20
rea
dings
eac
h
f
or
hunger
and
for st
ress.
3.
RESU
LT
S
AND A
N
ALYSIS
We
ha
ve
ta
ke
n
diff
e
ren
t
GS
R
read
in
gs
to
de
te
ct
stressed
an
d
relaxe
d
c
ondi
ti
on
s
sim
il
arly
to
ob
se
r
ve
hungry
a
nd
rel
axed
co
ndit
ion
as
sho
wn
i
n
Figure
3(
A
)
a
nd
(
B
)
.
It
is
obs
erv
e
d
that
t
he
GS
R
rea
di
ng
s
hows
high
value
s
in
relaxed
sta
te
.
In
hung
ry
co
nd
it
io
n
rea
dings
are
f
ound
out
to
be
belo
w
400
but
abov
e
300.
Wh
e
n
read
i
ngs
wer
e
belo
w
300
a
nd
with
high
va
riance
,
we
f
ound
ou
t
the
sta
tus
to
be
stresse
d.
W
e
can
con
cl
ud
e
t
hat
GS
R
rea
dings
wer
e
gen
e
rall
y
above
40
0
in
r
el
axed
sta
te
.
We
ha
ve
ob
se
r
ved
t
hat
the
sta
te
to
be
relaxe
d
with
no
hunger
a
nd
no
stress.
Fig
ure
3(
C)
sho
ws
th
e
read
outs
of
HS
MS
syst
em
wh
e
n
te
st
su
bj
ect
was
unde
r
m
edita
tio
n,
dee
p
br
eat
hing
an
d
st
ud
y
ing
.
We
fou
nd
the
eff
ect
of
th
ese
conditi
on
s
on
th
e
stress
le
vel
of
par
ti
ci
pa
nts.
F
r
om
the
values
recorde
d
a
fter
m
edita
ti
on
an
d
befo
re
m
edita
t
ion
a
nd
deep
breat
hing
e
xer
ci
se,
it
is fou
nd out th
at
stressed st
at
e of the
sub
j
ect
h
as
ch
a
nged
to rel
ax
sta
te
.
On
oth
e
r
ha
nd
read
i
ng
s
be
for
e
stud
yi
ng
an
d
after
stu
dying
sh
ows
tra
nsfo
rm
ation
f
ro
m
relaxed
sta
te
to
stressed
sta
te
.
Figure
3(
D
)
,
sho
ws
the
e
ff
ect
of
tim
e
of
the
day
on
syst
e
m
read
out.
Dif
fer
e
nt
sta
te
of
relaxati
on
at
di
ff
e
ren
t
ti
m
e
was
recorde
d.
M
os
t
of
the
s
ubje
ct
s
found
ou
t
to
be
hungry
at
7.00
a
nd
11.
00
tim
e
po
i
nts.
At 9
.
00
, 1
4.0
0
an
d
22.
00, al
m
os
t al
l
t
he
read
i
ngs w
e
re f
ou
nd
int
o
a relaxe
d
cat
egor
y. At 1
9.00
m
os
t of
the
te
st
cases
wer
e
stre
ssed
and
few
of
th
e
m
wer
e
ly
ing
into
a
hung
ry
sta
te
.
The
rea
di
ngs
we
re
ta
ke
n
at
diff
e
re
nt
place
s
but
the
re
was
no
sig
nifica
nt
diff
e
re
nce
as
pe
r
the
place
(r
e
sidence
,
c
olleg
e,
office
)
c
on
c
ern
e
d.
All
res
ults
co
ul
d
detect
hung
er.
Sam
e
par
ti
ci
pan
ts
par
ti
ci
pated
3
ti
m
es
in
detect
io
n
of
hunger
on
dif
fer
e
nt
days. T
he o
btain
ed
r
es
ults
were co
nf
irm
ed
f
or r
e
peatabil
it
y.
In
case
of
s
om
e
par
ti
ci
pan
t
s
GS
R
value
increase
d
e
ven
after
c
on
s
um
i
ng
te
a
or
c
off
ee.
HS
-
MS
syst
e
m
will
be
further
m
od
i
fied
as
wea
ra
ble
syst
e
m
and
m
easur
ing
di
ff
ere
nt
body
vital
par
am
et
e
rs
li
ke
tem
per
at
ur
e
,
he
art
rate,
pu
lse
rate
and
ECG
.
Using
al
l
these
par
am
et
ers
or
a
com
bin
at
ion
of
any
one
with
GS
R
values
m
or
e
accurate
te
sts
for
stre
ss
can
be
pe
rfor
m
ed.
T
his
m
od
el
can
be
us
e
d
as
prec
autiona
ry
m
easur
e
for
gi
ving
al
erts
to
careta
ke
rs
in
case
of
increa
se
d
stress
o
r
un
c
omm
on
patte
rn
s
in
any
vital
body
par
am
et
er.
Thi
s is u
se
f
ul in
c
ase o
f heart
or
stroke
patie
nts
and elde
rly
p
e
op
le
sta
yi
ng
at
ho
m
e alon
e.
4.
CONCL
US
I
O
N
In
hos
pital
s
w
hen
pa
ti
ents
ar
e
in
unco
ns
ci
ous
or
sem
i
-
consci
ou
s
sta
te
,
unde
r
t
reatm
ent
with
hea
vy
m
edici
ne
do
se
s
found
ha
r
d
to
detect
act
ual
hung
e
r.
Also
in
case
of
patie
nts
su
f
fer
i
ng
fro
m
hear
t
diseases,
st
rok
e
et
c.
it
is
req
ui
red
to
kee
p
th
e
patie
nts
stres
s
fr
ee
.
H
S
-
MS
syst
e
m
is
us
ed
to
detect
stre
ss
an
d
hunger
us
in
g
GS
R
sens
or.
It
detect
s
hu
nge
r
wit
h
86.
6%
accu
rac
y
and
99.
9%
sensiti
vity
.
We
ha
ve
su
ccess
fu
ll
y
de
te
ct
ed
stress
with
66.
66%
accuracy
an
d
73.33%
se
ns
it
ivit
y.
Av
e
rag
e
respo
ns
e
tim
e
of
th
e
syst
e
m
is
5
seconds.
It
is
diff
i
cult
to
ide
ntify
stress
a
nd
hunger
sim
ultaneo
us
ly
beca
us
e
we
f
ound
in
str
ess
e
d
conditi
on
e
ff
ec
t
of
hunger
is
neg
li
gi
ble.
W
e
al
so
fo
un
d
tha
t,
there
was
no
sign
ific
ant
dif
fer
e
nce
on
hunger
as
per
the
place
is
co
ncerne
d;
on
the
oth
e
r
ha
nd,
ti
m
e
of
the
da
y
has
sig
nific
ant
ef
fect
on
hunge
r
because
of
the
body cyc
le
.
ACKN
OWLE
DGE
MEN
TS
Au
t
hors
would
li
ke
to
tha
nk
s
tud
e
nts
an
d
sta
ff
of
S.S
.
Jon
dhal
e
Coll
ege
of
En
gin
ee
rin
g
for
s
upport
and
enc
oura
ge
m
ent
in
car
ryi
ng
out
this
w
or
k.
We
w
ou
l
d
a
lso
li
ke
t
o
tha
nk
Dr.
T
haku
r
a
nd
D
r.
Ma
rath
e
from
M. D. T
haku
r
Mem
or
ia
l m
ult
ispeci
al
ty
h
osp
it
al
an
d
i
ntensi
ve
ca
re
un
it
.
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
Hun
ger and
stress mo
nitorin
g syst
em usi
ng galva
nic skin
re
sp
onse
(
Uttar
a G
ogate
)
865
REFERE
NCE
S
[1]
Liao
,
W
.,
Zha
ng
,
W
.,
Zhu
,
Z
.
,
Ji
,
Q
.
A
real
-
ti
me
human
stress
mo
nit
oring
system
using
dynamic
B
aye
sian
ne
twork
.
In:
2005
IEEE
Com
pute
r
So
ci
e
t
y
Conf
ere
n
c
e
on
Com
puter
Vision
and
Patt
ern
Rec
ogn
it
io
n
(
CVP
R2005
)
-
W
orkshops
,
San
Diego,
CA
,
US
A,
pp
.
70
–
80
(
20
05
).
doi
:10.
1109
/
CVP
R
.
2005.
39
4
9.
[2]
Yoon,
Sungh
y
u
n,
Sim
,
Jai
K
y
o
ung,
Cho,
Young
-
Ho,
A
Fl
e
xi
bl
e
and
W
earable
Hum
an
Stress
Monit
oring
Pa
tch
Sci
en
ti
fic
R
e
ports
2016/03/
23/onl
in
e
6
-
23468,
htt
p
://dx.doi.org/
10
.
1
038/srep2346810.
1038/srep2346
8
,
htt
ps://
ww
w.nat
ure
.
com/a
rticle
s/
srep23468#supple
m
ent
ar
y
-
infor
m
at
ion
.
[3]
E.
Jovanov
,
A
.
O'
Donnell
Lords
,
D.
Raskovic,
P
.
G.
Cox,
R.
Ad
hami
and
F
.
An
dra
sik,
Str
ess
m
onit
oring
us
ing
a
distri
bute
d
wire
l
ess
int
el
li
g
ent
se
nsor
sy
stem,
in
I
EE
E
Engi
n
ee
rin
g
in
Me
dic
ine
an
d
Bi
ology
Maga
zine
,
vol.
22,
no.
3,
pp.
49
-
55,
Ma
y
-
June
2003.
d
oi:
10.
1109
/ME
MB.2003.
12136
26
,
htt
p://ieeexp
lore.
ie
e
e.
org/st
amp/stamp.j
sp
?
tp=&ar
num
ber
=1213626&
isnum
ber
=27
287
.
[4]
M.
Khan,
Z.
Ri
zvi
,
M.
Z.
Shaik
h,
W
.
Kaz
m
i,
A.
Shaikh,
Desig
n
and
implementa
ti
on
of
in
te
l
li
g
ent
hum
an
stress
m
onit
oring
s
y
st
e
m
,
Inte
rnationa
l Journal
of
Innov
ati
on
and
Scient
i
c Re
-
search
10(
1),
179
-
190
(201
4).
[5]
M.V.
Vill
ar
ej
o
,
B.
G.
Za
p
ira
in
,
A.M.
Zorr
illa
,
Sensors
12(5),
A
Stress
Sensor
B
ase
d
on
Galva
ni
c
Skin
Response
(GS
R)
Control
led b
y
Z
igBee,
Se
nsor
s
2012
;
12
(5)
:
6075
-
6101;
https
:/
/doi
.
or
g/10.
3
390/s120506075
.
[6]
T.
Huang,
J.
Xi
ong,
L.
Xuan
,
D
e
s
i
g
n
a
n
d
I
m
p
l
e
m
e
n
t
a
t
i
o
n
o
f
a
W
i
r
e
l
e
s
s
H
e
a
l
t
h
c
a
r
e
S
y
s
t
e
m
B
a
s
e
d
o
n
G
a
l
v
a
n
i
c
S
k
i
n
R
e
s
p
o
n
s
e
,
Inte
rna
ti
ona
l
Confer
ence
on
Applie
d
Inform
at
i
c
s
and
Co
m
m
uni
ca
t
ion
(
Springer
,
2011),
pp.
337
-
343.
[7]
J.
Ferre
ira,
L.
A
lva
re
z,
R.
Bu
en
d
a,
D.
A
y
ll
on,
C.
Ll
er
ena,
R.
GilPit
a,
F.
Seo
a
ne,
Bi
o
impedance
-
Based
We
arab
le
Me
asur
eme
nt
Instrum
ent
ati
on
fo
r
Study
ing
the
A
utomati
c
Nerv
e
Syste
m
Re
spons
e
to
Stressful
Working
Co
ndit
io
ns,
in
Journal
of
P
h
y
sics:
Confer
e
nce
Ser
ie
s,
(IO
P
Publishing,
2
013),
vol
.
434
,
p.
0
12015,
do
i:
10
.
1088/1742
-
6596/434/
1/012
015.
[8]
L.
Cl
ift
on,
D.A.
Cli
fton
,
M.A.
P
imente
l
,
P.J.
W
a
tki
nson,
L
.
T
aras
senko,
Predic
t
i
ve
Monitori
ng
o
f
Mobile
Pat
ie
nt
s
b
y
Com
bin
ing
Cli
nical
Obs
erv
at
ions
with
Dat
a
from
W
ea
rab
l
e
Sensors
,
IEEE
journal
of
b
iomedi
ca
l
and
h
eal
t
h
inf
orm
ati
cs
2014
;
18(3), 722:
216
8
-
2194.
[9]
K.
Malhi,
S.
C.
Mukhopadh
y
a
y
,
J.
Schneppe
r
,
M.
Hae
fke
,
and
H.
Ewa
ld
,
A
Zigbee
-
base
d
wea
r
abl
e
ph
y
sio
logi
c
al
par
amete
rs
m
oni
-
tori
ng
s
y
stem,
I
EE
E
Sensors
J
.
,
2012
;
12
(
3
):
423
-
430
.
[10]
J.
Choi,
B
.
Ahm
ed,
R.
Guti
err
ez
-
Os
una,
Dev
elopm
ent
and
ev
a
lua
ti
on
of
an
a
m
bula
tor
y
stress
m
onit
or
base
d
on
wea
rab
l
e
sensors
.
,
IEE
E
tran
sacti
ons
on
inf
orm
ati
on
te
ch
nology
in
bio
medic
in
e,
2012
;
16(2):
279
-
86.
doi:
10.
1109
/T
IT
B.
2011.
2169804
.
Epub
2011
Sep
29.
[11]
S.
Saha,
P.
N
ag
,
M.K.
Ra
y
,
in
Control
,
A
com
ple
t
e
Vi
rtua
l
Instrum
ent
for
Me
a
sur
ing
and
ANAL
Y
ZING
Hum
an
Stress
in
R
eal
Time,
Instrum
en
tat
ion
,
En
erg
y
and
Com
m
unic
a
ti
on
(CI
EC),
20
14
Inte
rn
at
ion
al
Confer
en
ce
on
(IE
EE,
2014)
,
p
p.
81
-
85
.
[12]
A.
Sano,
R.
W
.
Pica
rd
,
Stress
Re
c
ognit
ion
using W
earable
Senso
rs
and
Mobil
e
P
hones
,
in
Affecti
ve
Com
puti
ng
a
nd
Inte
lligen
t
Int
eract
ion
(ACII),
20
13
Hum
ai
ne
As
sociation
Conf
erence
on
(IE
EE,
2
013),
pp
.
671
-
67
6.
[13]
F.T
.
Sun,
C.
Kuo,
H.T.
Ch
eng,
S.
Buthpi
t
i
y
a
,
P.
Coll
ins,
M.
Gri
ss
,
Ac
t
iv
i
ty
-
awar
e
Me
n
tal
S
tress
Dete
c
ti
on
Us
in
g
Phy
siologi
cal
S
ensors
,
in
Inte
rn
at
ion
al
Confer
e
nce
on
Mobil
e
Com
puti
ng,
Applicati
ons,
and
Servic
es
(Spring
er
,
2010),
pp
.
282
-
3
01.
[14]
R.
E.
Kl
ei
nm
an,
J.
M.
Murph
y
,
K.M.
W
ie
neke,
M.
S.
Desm
ond,
A.
Schi,
J.A
.
Gapinski,
Us
e
of
a
Singl
e
-
Quest
io
n
Scre
eni
ng
Tool
t
o
Detect
Hunge
r
in
Famil
i
es
Attendi
ng
a
Nei
gh
borhood
Healt
h
Cent
er
,
Am
bulat
or
y
Pa
edi
a
trics,
2007;
7
(
4
):
278
-
284.
[15]
R.
H.
B
arn
es.
Mi
nd/bod
y
in
te
gra
t
ion:
Essenti
a
l
r
e
adi
ngs
in
b
iofe
ed
bac
k
(1980)
.
Evaluation Warning : The document was created with Spire.PDF for Python.