I
nte
rna
t
io
na
l J
o
urna
l
o
f
Adv
a
nces in Applie
d Science
s
(
I
J
AAS)
Vo
l.
6
,
No
.
1
,
M
ar
ch
2
0
1
7
,
p
p
.
70
~
76
I
SS
N:
2252
-
8814
70
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
jo
u
r
n
a
l.c
o
m/o
n
lin
e/in
d
ex
.
p
h
p
/I
J
AAS
Cla
ss
ificatio
n of
H
ea
rt
Ra
te
Da
ta
Using
BF
O
-
K
FC
M
Clusteri
ng
a
nd I
m
p
ro
v
ed Ex
tre
m
e Lea
r
ning
M
a
chine Clas
sifie
r
R.
k
a
v
it
ha
1
,
T
.
Chri
s
t
o
ph
er
2
1
De
p
a
rtme
n
t
o
f
c
o
m
p
u
ter sc
ien
c
e
,
P
S
G
R
Krish
n
a
m
m
a
l
Co
ll
e
g
e
f
o
r
w
o
m
e
n
,
Co
im
b
a
to
re
,
In
d
ia
2
De
p
a
rtme
n
t
o
f
Re
se
a
rc
h
&
P
G
c
o
m
p
u
ter sc
ien
c
e
,
G
o
v
e
rn
m
e
n
t
A
rt
s Co
ll
e
g
e
,
Co
im
b
a
to
re
,
In
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
an
1
5
,
2
0
1
6
R
ev
i
s
ed
Feb
1
5
,
2
0
1
7
A
cc
ep
ted
Feb
2
2
,
2
0
1
7
A
n
El
e
c
tro
c
a
rd
io
g
ra
m
o
r
ECG
is
a
n
e
lec
tri
c
a
l
re
c
o
rd
in
g
o
f
th
e
h
e
a
rt
a
n
d
is
u
se
d
in
t
h
e
in
v
e
stig
a
ti
o
n
o
f
h
e
a
rt
d
ise
a
se
.
T
h
e
h
e
a
rt
ra
te
v
a
ries
n
o
t
o
n
ly
in
re
latio
n
t
o
t
h
e
c
a
rd
iac
d
e
m
a
n
d
b
u
t
is
a
lso
a
f
f
e
c
ted
b
y
th
e
p
re
se
n
c
e
o
f
c
a
rd
iac
d
ise
a
se
a
n
d
d
iab
e
tes
.
F
u
rth
e
rm
o
re
,
it
h
a
s
b
e
e
n
sh
o
w
n
th
a
t
He
a
rt
Ra
t
e
V
a
riab
il
i
ty
(HRV
)
m
a
y
b
e
u
se
d
a
s
a
n
e
a
rl
y
in
d
ica
to
r
o
f
c
a
rd
iac
d
ise
a
se
su
sc
e
p
ti
b
il
it
y
a
n
d
t
h
e
p
re
se
n
c
e
o
f
d
iab
e
tes
.
T
h
e
re
f
o
re
,
th
e
h
e
a
rt
ra
te
v
a
riab
il
it
y
m
a
y
b
e
u
se
d
f
o
r
e
a
rly
c
li
n
ica
l
sc
re
e
n
in
g
o
f
th
e
se
d
is
e
a
se
s.
T
h
e
g
e
n
e
ra
li
z
a
ti
o
n
p
e
rf
o
r
m
a
n
c
e
o
f
t
h
e
S
V
M
c
las
sif
ier
is
n
o
t
su
ff
ici
e
n
t
f
o
r
th
e
c
o
rre
c
t
c
las
si
f
ica
ti
o
n
o
f
h
e
a
rt
ra
te
d
a
ta.
T
o
o
v
e
rc
o
m
e
th
is
p
ro
b
lem
th
e
Im
p
ro
v
e
d
Ex
tre
m
e
L
e
a
rn
in
g
M
a
c
h
in
e
(IE
L
M
)
c
las
sif
ier
is
u
se
d
w
h
ich
w
o
rk
s
b
y
se
a
r
c
h
in
g
f
o
r
th
e
b
e
st
v
a
lu
e
o
f
th
e
p
a
ra
m
e
ter
s,
a
n
d
u
p
stre
a
m
b
y
lo
o
k
in
g
f
o
r
th
e
b
e
st
su
b
se
t
o
f
fe
a
tu
re
s
u
sin
g
Ba
c
teria
l
F
o
ra
g
in
g
Op
ti
m
iza
t
io
n
(B
F
O)
th
a
t
f
e
e
d
th
e
c
las
sif
ier.
In
th
is
w
o
rk
,
n
i
n
e
li
n
e
a
r
a
n
d
n
o
n
li
n
e
a
r
f
e
a
tu
re
s
a
re
e
x
trac
ted
f
ro
m
th
e
HRV
sig
n
a
ls.
Af
ter t
h
e
p
re
p
ro
c
e
ss
in
g
,
fe
a
tu
re
e
x
trac
ti
o
n
is
d
o
n
e
a
lo
n
g
w
it
h
f
e
a
tu
re
se
lec
ti
o
n
u
sin
g
B
F
O
f
o
r
d
a
ta
re
d
u
c
ti
o
n
.
T
h
e
n
,
p
ro
p
o
se
d
a
sc
h
e
m
e
to
in
teg
ra
te
Ke
rn
e
l
F
u
z
z
y
C
-
M
e
a
n
s
(KFCM
)
c
lu
ste
rin
g
a
n
d
Clas
sif
ier t
o
im
p
ro
v
e
th
e
a
c
c
u
ra
c
y
re
su
lt
f
o
r
EC
G
b
e
a
t
c
las
sif
i
c
a
ti
o
n
.
T
h
e
re
su
lt
s
sh
o
w
th
a
t
th
e
p
ro
p
o
se
d
m
e
th
o
d
is
e
f
fe
c
ti
v
e
f
o
r
c
la
ss
i
f
ica
ti
o
n
o
f
h
e
a
rt
ra
te d
a
ta,
w
it
h
a
n
a
c
c
e
p
tab
le h
ig
h
a
c
c
u
ra
c
y
.
K
ey
w
o
r
d
:
E
C
G
s
i
g
n
a
ls
clas
s
i
f
icatio
n
I
m
p
r
o
v
ed
E
x
tr
e
m
e
L
ea
r
n
i
n
g
Ma
ch
i
n
e
(
E
L
M)
Ker
n
el
F
u
zz
y
C
-
Me
an
s
Su
p
p
o
r
t
Vec
to
r
Ma
ch
in
e
(
SVM)
Co
p
y
rig
h
t
©
201
7
In
s
t
it
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
R
.
k
a
v
it
h
a
,
Dep
ar
t
m
en
t o
f
co
m
p
u
ter
s
c
ien
ce
,
Kr
is
h
n
a
m
m
a
l Co
lle
g
e
f
o
r
w
o
m
en
,
C
o
i
m
b
ato
r
e
,
I
n
d
ia.
E
m
ail:
r
k
a
v
it
h
a2
5
0
9
@
g
m
ail.
c
o
m
1.
I
NT
RO
D
UCT
I
O
N
E
C
G
is
a
m
et
h
o
d
to
m
ea
s
u
r
e
an
d
r
ec
o
r
d
d
if
f
er
en
t
elec
tr
ical
p
o
ten
tials
o
f
th
e
h
ea
r
t
.
W
ille
m
E
in
t
h
o
v
e
n
d
ev
elo
p
ed
th
e
E
C
G
m
et
h
o
d
in
t
h
e
ea
r
l
y
1
9
0
0
s
.
T
h
e
o
r
ig
in
o
f
t
h
e
elec
tr
ical
a
ctiv
it
y
m
ea
s
u
r
ed
b
y
E
C
G
i
s
i
n
t
h
e
m
u
s
cle
f
ib
er
s
o
f
d
if
f
er
en
t
p
ar
ts
o
f
t
h
e
h
ea
r
t
.
T
h
e
elec
tr
ical
p
o
ten
tial
g
e
n
er
ated
b
y
e
lectr
ical
ac
tiv
it
y
in
ca
r
d
iac
tis
s
u
e
is
m
ea
s
u
r
ed
o
n
th
e
s
u
r
f
ac
e
o
f
t
h
e
h
u
m
a
n
b
o
d
y
.
C
u
r
r
en
t
f
lo
w
,
in
th
e
f
o
r
m
o
f
io
n
s
,
s
ig
n
al
s
co
n
tr
ac
tio
n
o
f
ca
r
d
iac
m
u
s
c
le
f
ib
er
s
lead
in
g
to
th
e
h
ea
r
t's
p
u
m
p
i
n
g
ac
tio
n
.
I
t
is
a
n
o
n
p
e
r
s
is
te
n
t
r
ec
o
r
d
in
g
p
r
o
d
u
ce
d
b
y
a
n
elec
tr
o
ca
r
d
io
g
r
ap
h
ic
d
ev
ice.
T
h
e
r
ec
o
g
n
itio
n
a
n
d
clas
s
i
f
icatio
n
o
f
t
h
e
E
C
G
b
ea
t
s
i
s
a
v
er
y
i
m
p
o
r
tan
t ta
s
k
in
t
h
e
co
r
o
n
ar
y
in
te
n
s
iv
e
u
n
it,
w
h
er
e
th
e
class
i
f
icatio
n
o
f
t
h
e
E
C
G
b
ea
ts
is
e
s
s
e
n
tial to
o
l
f
o
r
th
e
d
iag
n
o
s
is
.
E
C
G
o
f
f
e
r
s
ca
r
d
io
lo
g
is
ts
w
it
h
u
s
e
f
u
l
in
f
o
r
m
atio
n
ab
o
u
t
t
h
e
r
h
y
th
m
a
n
d
f
u
n
ctio
n
in
g
o
f
th
e
h
ea
r
t.
T
h
er
ef
o
r
e,
its
an
a
l
y
s
is
r
ep
r
esen
ts
an
ef
f
icie
n
t
w
a
y
to
d
ete
ct
an
d
tr
ea
t
d
if
f
er
en
t
k
in
d
s
o
f
ca
r
d
iac
d
is
ea
s
es
u
p
to
n
o
w
;
m
a
n
y
al
g
o
r
ith
m
s
h
a
v
e
b
ee
n
d
ev
e
lo
p
ed
f
o
r
th
e
r
ec
o
g
n
itio
n
an
d
c
la
s
s
i
f
icatio
n
o
f
E
C
G
s
ig
n
al.
So
m
e
o
f
t
h
e
m
u
s
e
ti
m
e
an
d
s
o
m
e
u
s
e
f
r
eq
u
en
c
y
d
o
m
ai
n
f
o
r
d
ep
ictio
n
.
B
ased
o
n
th
at
m
a
n
y
s
p
ec
if
ic
attr
ib
u
tes
ar
e
d
ef
in
ed
,
allo
w
in
g
th
e
r
ec
o
g
n
itio
n
b
et
w
ee
n
t
h
e
b
ea
ts
b
elo
n
g
in
g
to
d
if
f
er
e
n
t
p
ath
o
lo
g
ical
clas
s
es
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
AA
S
I
SS
N:
2252
-
8814
Tit
le
o
f m
a
n
u
s
crip
t is sh
o
r
t a
n
d
clea
r
,
imp
lies
r
esea
r
ch
r
esu
l
ts
(
F
i
r
s
t A
u
th
o
r
)
71
T
h
e
E
C
G
w
a
v
e
f
o
r
m
s
m
a
y
b
e
d
if
f
er
e
n
t
f
o
r
t
h
e
s
a
m
e
p
atie
n
t
to
s
u
c
h
e
x
te
n
t
th
a
t
t
h
e
y
ar
e
u
n
lik
e
ea
c
h
o
th
er
a
n
d
at
th
e
s
a
m
e
ti
m
e
ali
k
e
f
o
r
d
if
f
e
r
en
t t
y
p
es o
f
b
ea
ts
[
1
]
.
Hea
r
t
R
ate
Var
iab
ilit
y
i
s
u
s
e
d
to
m
ea
s
u
r
e
th
e
v
ar
iatio
n
s
in
h
ea
r
t
s
ig
n
als
a
n
d
m
o
r
e
s
p
ec
if
icall
y
v
ar
iatio
n
s
p
er
u
n
it
ti
m
e
o
f
t
h
e
n
u
m
b
er
o
f
h
ea
r
tb
ea
ts
.
T
h
e
E
lectr
o
ca
r
d
io
g
r
am
is
t
h
e
elec
tr
ic
al
ac
tiv
it
y
o
f
h
ea
r
t
an
d
g
e
n
er
ates
elec
tr
ical
s
ig
n
a
ls
w
h
ic
h
ar
e
ca
lled
a
s
P
QR
S
T
w
a
v
es.
T
h
e
m
o
s
t
i
m
p
o
r
tan
t
w
av
e
is
th
e
QR
S
co
m
p
le
x
.
Hea
r
t
b
ea
t
d
ep
en
d
s
o
n
th
e
t
i
m
e
in
ter
v
al
b
et
w
ee
n
t
w
o
Q
R
S
co
m
p
le
x
w
a
v
es
w
h
i
ch
is
ca
l
led
as
R
-
R
in
ter
v
a
l [
2
]
.
No
r
m
all
y
Hea
lt
h
y
p
er
s
o
n
s
p
r
ese
n
t la
r
g
e
v
alu
e
s
o
f
H
R
V.
P
r
ed
ictio
n
o
f
HR
V
a
n
al
y
s
i
s
i
s
o
n
e
o
f
t
h
e
m
aj
o
r
r
esear
ch
to
p
ics
f
r
o
m
la
s
t
t
w
o
d
ec
ad
es.
Se
v
er
al
m
et
h
o
d
s
f
o
r
h
ea
r
t
r
ate
v
ar
iab
ilit
y
w
e
r
e
p
r
o
p
o
s
ed
am
o
n
g
th
e
m
,
Sp
ec
tr
al
Me
t
h
o
d
s
b
ased
o
n
F
FT
,
n
o
n
li
n
ea
r
a
p
p
r
o
ac
h
,
in
clu
d
i
n
g
m
ar
k
o
v
ch
ai
n
m
o
d
el
ar
e
w
id
el
y
u
s
ed
[
3
,
4
]
.
Neu
r
al
Net
w
o
r
k
a
n
d
Ma
ch
i
n
e
L
ea
r
n
i
n
g
m
e
th
o
d
s
ar
e
o
n
e
o
f
th
e
p
o
w
er
f
u
l
m
et
h
o
d
s
to
class
i
f
y
a
n
d
p
r
ed
ict
th
e
Hea
r
t
r
ate
v
ar
iab
il
it
y
p
atter
n
s
.
Var
io
u
s
s
tu
d
ie
s
h
av
e
b
ee
n
d
o
n
e
u
s
i
n
g
d
if
f
er
en
t
cla
s
s
i
f
icatio
n
m
et
h
o
d
s
lik
e
S
u
p
p
o
r
t V
ec
to
r
Ma
ch
i
n
e,
Neu
r
al
Net
w
o
r
k
,
a
n
d
W
av
elet
T
r
an
s
f
o
r
m
P
C
A
etc
.
I
n
th
i
s
w
o
r
k
,
I
E
L
M
is
u
s
ed
f
o
r
C
lass
i
f
icatio
n
o
f
H
R
V
d
at
a.
B
y
u
s
i
n
g
I
E
L
M,
n
o
t
o
n
l
y
r
ed
u
ce
th
e
c
lass
i
f
icatio
n
ti
m
e,
b
u
t
also
i
m
p
r
o
v
e
th
e
cla
s
s
i
f
icatio
n
ac
cu
r
ac
y
s
i
g
n
i
f
ican
tl
y
.
I
n
t
h
is
w
o
r
k
a
f
ter
f
ea
t
u
r
e
ex
tr
ac
tio
n
th
e
s
elec
tio
n
o
f
b
es
t
f
ea
t
u
r
es
is
d
o
n
e
u
s
i
n
g
B
FO
w
h
ic
h
i
s
u
s
ed
to
r
ed
u
ce
t
h
e
d
i
m
en
s
io
n
o
f
f
ea
tu
r
e
s
in
Hea
r
tb
ea
t
cla
s
s
i
f
icatio
n
w
i
th
cl
u
s
t
er
i
n
g
u
s
i
n
g
KF
C
M.
T
h
e
r
e
m
a
in
i
n
g
p
ar
t
o
f
t
h
e
p
ap
er
is
o
r
g
an
ized
as
f
o
llo
w
s
.
Sectio
n
2
d
escr
ib
es
t
h
e
p
r
ep
r
o
ce
s
s
in
g
a
n
d
f
ea
t
u
r
e
ex
tr
ac
tio
n
alo
n
g
w
i
th
f
ea
tu
r
e
s
elec
tio
n
p
r
o
ce
s
s
.
Sectio
n
4
d
escr
ib
es
th
e
I
E
L
M
s
y
s
te
m
f
o
r
HR
V
clas
s
if
ica
tio
n
.
Sectio
n
5
d
escr
ib
es
t
h
e
ex
p
er
i
m
e
n
tal
d
esi
g
n
in
cl
u
d
in
g
th
e
e
v
al
u
atio
n
r
es
u
lt
s
o
f
th
is
s
y
s
te
m
.
Fi
n
all
y
,
Secti
o
n
7
co
n
clu
d
es t
h
e
p
ap
er
.
2.
M
AT
E
RIAL
S AN
D
M
E
T
H
O
DS
I
n
t
h
is
p
ap
er
,
th
e
HR
V
s
i
g
n
al
ex
p
lo
r
ed
as
t
h
e
b
asic
s
ig
n
al
t
o
class
i
f
y
ca
r
d
iac
ar
r
h
y
t
h
m
ias
in
to
f
i
v
e
class
es
:
No
r
m
al
Si
n
u
s
R
h
y
t
h
m
(
NS
R
)
,
P
r
e
m
atu
r
e
Ven
tr
ic
u
lar
C
o
n
tr
ac
tio
n
(
P
VC
)
,
A
tr
i
al
Fib
r
illatio
n
(
A
F),
Ven
tr
ic
u
lar
Fib
r
illatio
n
(
VF)
an
d
an
d
2
°
Hea
r
t
B
lo
ck
(
B
I
I
)
.
T
h
e
HR
V
ar
r
h
y
t
h
m
ia
d
ata,
o
b
tain
ed
u
s
in
g
th
e
E
C
G
d
ata
f
r
o
m
th
e
MI
T
-
B
I
H
A
r
r
h
y
t
h
m
ia
Data
b
ase.
T
h
e
an
al
y
s
is
is
ca
r
r
ie
d
o
u
t
in
th
r
ee
s
tag
es.
First
a
p
r
ep
r
o
ce
s
s
in
g
p
r
o
ce
d
u
r
e
is
u
s
ed
to
r
em
o
v
e
t
h
e
i
n
ter
f
er
i
n
g
s
ig
n
a
ls
f
r
o
m
t
h
e
E
C
G
s
.
I
n
t
h
e
s
ec
o
n
d
s
ta
g
e,
ti
m
e
an
d
f
r
eq
u
e
n
c
y
d
o
m
ai
n
an
d
n
o
n
li
n
ea
r
m
et
h
o
d
s
ar
e
ap
p
lied
to
ex
tr
ac
t
co
r
r
esp
o
n
d
in
g
f
ea
tu
r
e
s
.
I
n
th
e
t
h
ir
d
s
tag
e
th
e
ex
tr
a
cted
f
ea
t
u
r
es
ar
e
r
ed
u
ce
d
to
attain
th
e
o
p
tim
al
f
ea
tu
r
es
u
s
i
n
g
B
FO
an
d
b
ased
o
n
th
ese
th
e
cl
u
s
ter
i
n
g
p
r
o
ce
s
s
is
d
o
n
e
u
s
in
g
K
FC
M.
T
h
ese
r
esu
lts
ar
e
g
i
v
e
n
as
in
p
u
t
to
I
E
L
M
clas
s
i
f
ier
.
T
h
e
o
v
er
all
ar
ch
itectu
r
e
d
iag
r
a
m
i
s
ill
u
s
tr
ated
in
Fi
g
.
1
.
Nex
t
m
ater
ials
a
n
d
m
et
h
o
d
s
ar
e
d
escr
ib
ed
.
T
h
en
t
h
e
d
i
f
f
er
en
t
s
tep
s
o
f
t
h
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
ar
e
ex
p
lain
ed
.
Fin
a
ll
y
r
esu
lts
o
b
t
ain
ed
o
n
t
h
e
MI
T
-
B
I
H
ar
r
h
y
t
h
m
ia
d
atab
ase
ar
e
p
r
esen
ted
.
a.
P
re
pro
ce
s
s
ing
An
d F
ea
t
ure
E
x
t
ra
ct
io
n Alo
ng
Wit
h F
ea
t
ure
Select
io
n P
ro
ce
s
s
2
.
1
.
1
P
re
pro
ce
s
s
ing
t
he
Sig
na
l
T
h
e
HR
V
d
ata
u
s
ed
in
th
is
p
r
o
j
ec
t
is
g
en
er
ated
f
r
o
m
th
e
E
C
G
s
ig
n
al
s
p
r
o
v
id
ed
b
y
t
h
e
MI
T
-
B
I
H
d
atab
ase.
A
t
f
ir
s
t,
it
is
n
ec
e
s
s
ar
y
to
e
x
tr
ac
t
t
h
e
HR
V
s
i
g
n
al
s
f
r
o
m
t
h
e
E
C
G
s
i
g
n
al
s
.
Gen
er
all
y
,
m
a
n
y
in
ter
f
er
i
n
g
s
ig
n
al
s
s
u
c
h
as
t
h
e
m
ai
n
s
5
0
Hz,
th
e
e
lectr
o
m
y
o
g
r
a
m
(
E
MG
)
s
i
g
n
als
an
d
also
th
e
b
aseli
n
e
w
a
n
d
er
in
g
ca
n
af
f
ec
t t
h
e
e
x
tr
a
ctio
n
p
r
o
ce
s
s
.
Hen
ce
,
t
h
e
s
e
in
t
er
f
er
in
g
s
i
g
n
als ar
e
r
e
m
o
v
ed
f
r
o
m
t
h
e
i
n
p
u
t
E
C
G
s
ig
n
al
u
s
i
n
g
a
5
-
1
5
Hz
b
an
d
p
ass
f
i
lter
.
T
h
en
th
e
s
i
g
n
a
l
is
p
r
o
ce
s
s
ed
u
s
i
n
g
t
h
r
es
h
o
ld
in
g
an
d
th
e
r
es
u
lta
n
t
s
ig
n
al
is
id
e
n
ti
f
ied
as H
R
V
s
i
g
n
al.
2
.
1
.
2
F
ea
t
ure
E
x
t
ra
ct
io
n
T
h
e
n
ex
t
s
tep
i
n
t
h
e
b
lo
ck
d
iag
r
a
m
i
s
th
e
f
ea
t
u
r
e
ex
tr
ac
tio
n
.
Gen
er
all
y
,
t
h
e
ca
r
d
io
v
ascu
lar
s
y
s
te
m
d
em
o
n
s
tr
ate
s
b
o
th
li
n
ea
r
an
d
n
o
n
li
n
ea
r
b
eh
a
v
io
r
.
T
h
er
ef
o
r
e,
in
t
h
is
w
o
r
k
,
co
m
b
i
n
atio
n
s
o
f
li
n
ea
r
a
n
d
n
o
n
li
n
ea
r
f
ea
t
u
r
es a
r
e
co
n
s
id
er
ed
.
L
inea
r
a
na
ly
s
is
:
T
i
m
e
do
m
a
i
n f
ea
t
ures
Fo
u
r
co
m
m
o
n
l
y
u
s
ed
ti
m
e
d
o
m
ai
n
p
ar
am
e
ter
s
o
f
th
e
HR
V
s
ig
n
al
w
h
ich
ar
e
d
ir
ec
tl
y
ex
tr
ac
ted
f
r
o
m
th
e
R
R
in
ter
v
al
ti
m
e
s
er
ie
s
ar
e:
•
M
e
a
n
H
R:
T
h
e
m
ea
n
v
al
u
e
o
f
th
e
h
ea
r
t
r
ate
w
it
h
i
n
o
n
e
m
i
n
u
te
in
ea
c
h
s
eg
m
e
n
t.
I
n
s
tan
ta
n
eo
u
s
h
ea
r
t
r
ate
(
b
ea
t p
er
m
i
n
u
te)
is
eq
u
al
to
6
0
d
iv
id
ed
b
y
ea
ch
R
-
R
i
n
te
r
v
al
(
s
ec
o
n
d
)
.
•
ST
D
H
R:
T
h
e
s
tan
d
ar
d
d
ev
iatio
n
o
f
I
n
s
tan
tan
eo
u
s
h
ea
r
t r
at
e
in
ea
ch
s
e
g
m
e
n
t.
pNN5
0
:
T
h
e
n
u
m
b
er
o
f
s
u
cc
es
s
iv
e
d
if
f
er
en
ce
o
f
6
4
R
-
R
i
n
ter
v
al
s
t
h
at
d
i
f
f
er
s
m
o
r
e
t
h
an
5
0
m
s
,
r
esp
ec
tiv
el
y
,
d
i
v
id
ed
b
y
6
4
.
H
RV
t
ria
ng
ula
r
in
dex
:
T
h
is
r
ef
er
s
to
t
h
e
i
n
te
g
r
al
o
f
t
h
e
h
is
to
g
r
a
m
(
i.e
.
T
o
tal
n
u
m
b
er
o
f
R
R
in
ter
v
a
ls
)
d
iv
id
ed
b
y
th
e
h
ei
g
h
t o
f
th
e
h
is
to
g
r
a
m
.
A
b
in
w
id
t
h
o
f
1
/1
2
8
is
s
elec
ted
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8814
IJ
AA
S
Vo
l.
6
,
No
.
1
,
M
ar
ch
20
1
7
:
xx
–
xx
72
F
re
qu
ency
do
m
a
in f
ea
t
ures
A
lt
h
o
u
g
h
t
h
e
ti
m
e
d
o
m
ai
n
p
a
r
a
m
eter
s
ar
e
e
f
f
ec
ti
v
e,
t
h
e
y
d
o
n
o
t
h
av
e
th
e
ab
ilit
y
o
f
d
is
c
r
i
m
i
n
atio
n
b
et
w
ee
n
t
h
e
s
y
m
p
at
h
etic
an
d
p
ar
as
y
m
p
at
h
etic
co
n
te
n
ts
o
f
t
h
e
H
R
V
s
i
g
n
al.
Hi
g
h
-
f
r
eq
u
e
n
c
y
(
HF)
b
an
d
(
0
.
2
-
0
.
5
Hz)
o
f
HR
V
s
i
g
n
al
s
h
o
ws
th
e
ca
r
d
iac
v
a
g
al
ac
ti
v
it
ies
s
u
c
h
as
R
esp
ir
ato
r
y
Si
n
u
s
A
r
r
h
y
t
h
m
ia
(
R
S
A
)
.
I
n
f
ac
t,
HF
co
m
p
o
n
e
n
t
s
ar
e
co
n
s
id
er
ed
as
th
e
o
r
ig
in
o
f
p
ar
as
y
m
p
a
th
et
ic
ac
tiv
ities
o
f
th
e
ca
r
d
io
v
ascu
lar
s
y
s
te
m
.
On
t
h
e
o
th
er
h
a
n
d
,
th
e
lo
w
-
f
r
eq
u
en
c
y
(
L
F)
b
an
d
(
0
-
0
.
2
Hz)
is
r
elate
d
to
th
e
b
ar
o
r
ec
ep
to
r
co
n
tr
o
l
an
d
is
m
ed
iated
b
y
s
y
m
p
at
h
etic
s
y
s
t
e
m
s
.
I
n
th
i
s
p
r
o
j
ec
t,
th
e
p
o
w
e
r
s
p
ec
tr
al
d
en
s
it
y
(
P
SD)
f
o
r
th
e
HF
an
d
L
F
b
an
d
s
ar
e
ca
lcu
lated
an
d
th
e
r
atio
o
f
th
e
L
F
an
d
HF
b
an
d
s
p
o
w
er
(
L
F/H
F)
is
co
n
s
id
er
ed
as
th
e
Fre
q
u
en
c
y
d
o
m
a
i
n
f
ea
t
u
r
e
o
f
th
e
H
R
V
s
i
g
n
al.
No
nli
nea
r
a
na
ly
s
is
T
h
e
HR
V
s
i
g
n
al
an
a
l
y
s
is
b
y
u
s
e
o
f
m
et
h
o
d
s
o
n
n
o
n
li
n
ea
r
d
y
n
a
m
ic
s
lead
s
to
v
er
y
v
a
lu
ab
l
e
in
f
o
r
m
atio
n
f
o
r
p
h
y
s
io
lo
g
ical
in
ter
p
r
etatio
n
o
f
th
e
h
ea
r
t.
H
en
ce
,
f
o
u
r
d
if
f
er
en
t
n
o
n
lin
ea
r
p
ar
am
eter
s
o
f
t
h
e
HR
V
s
i
g
n
al
ar
e
u
s
ed
i
n
th
is
wo
r
k
.
SD1
/SD2
:
P
o
in
ca
r
e
p
lo
t
is
a
g
r
ap
h
ical
r
ep
r
esen
tatio
n
o
f
t
h
e
co
r
r
elatio
n
b
et
w
ee
n
s
u
cc
es
s
i
v
e
R
R
in
ter
v
als.
T
h
is
is
o
b
tain
ed
p
lo
ttin
g
ea
ch
R
R
i
n
ter
v
al
(
R
R
(
n
+1
)
)
as
a
f
u
n
ctio
n
o
f
t
h
e
p
r
ev
io
u
s
in
ter
v
al
(
R
R
(
n
)
)
in
R
R
i
n
ter
v
al
t
i
m
e
s
er
ies.
T
h
is
p
lo
t
is
q
u
an
tit
iv
el
y
a
n
a
l
y
ze
d
ca
lcu
latin
g
t
h
e
s
ta
n
d
ar
d
d
e
v
iatio
n
o
f
th
e
d
is
tan
ce
s
o
f
t
h
e
ti
m
e
s
er
ies
p
o
in
ts
f
r
o
m
t
h
e
li
n
es
y
=
x
a
n
d
y
=
x
+
2
,
in
w
h
ic
h
is
th
e
m
e
an
o
f
all
v
a
lu
e
s
o
f
R
R
in
ter
v
al
t
i
m
e
s
er
ies.
T
h
ese
v
a
lu
e
s
ar
e
n
a
m
ed
SD1
an
d
SD2
r
esp
ec
ti
v
el
y
.
I
n
f
ac
t,
S
D
1
r
ep
r
esen
ts
th
e
f
a
s
t
b
ea
t
-
to
-
b
ea
t
v
ar
iab
ilit
y
,
w
h
ile
SD2
d
escr
ib
es
th
e
r
elativ
el
y
lo
n
g
-
ter
m
v
ar
iab
ilit
y
in
t
h
e
HR
V
s
i
g
n
a
l.
I
n
th
is
w
o
r
k
,
SD
1
/SD2
is
u
s
ed
as
th
e
f
ir
s
t
n
o
n
lin
ea
r
f
ea
t
u
r
e
w
h
ic
h
is
ex
tr
ac
t
ed
f
r
o
m
H
R
V
s
eg
m
e
n
ts
.
L
L
E
:
T
h
e
L
ar
g
est
L
y
ap
u
n
o
v
E
x
p
o
n
e
n
t
p
r
o
v
id
es
u
s
e
f
u
l
in
f
o
r
m
atio
n
ab
o
u
t
t
h
e
d
ep
en
d
en
c
y
o
f
s
y
s
te
m
o
n
in
itial
co
n
d
itio
n
s
an
d
a
p
o
s
iti
v
e
L
y
ap
u
n
o
v
e
x
p
o
n
e
n
t
co
n
f
ir
m
s
t
h
e
e
x
is
te
n
ce
o
f
c
h
ao
s
i
n
t
h
e
s
y
s
te
m
.
Fo
r
ca
lcu
lati
n
g
L
L
E
,
a
p
o
in
t
is
s
elec
ted
in
t
h
e
r
ec
o
n
s
tr
u
cted
p
h
ase
s
p
ac
e
o
f
t
h
e
s
y
s
te
m
a
n
d
all
n
ei
g
h
b
o
r
p
o
in
ts
r
esid
i
n
g
w
it
h
in
a
p
r
ed
ef
i
n
ed
r
ad
iu
s
e
ar
e
d
eter
m
i
n
ed
.
As
t
h
e
s
y
s
te
m
e
v
o
lv
e
s
,
t
h
e
m
ea
n
d
is
ta
n
ce
s
b
et
w
ee
n
th
e
tr
aj
ec
to
r
y
o
f
th
e
i
n
itial
p
o
in
t
a
n
d
th
e
tr
aj
ec
to
r
ies
o
f
th
e
n
ei
g
h
b
o
r
p
o
in
ts
ar
e
ca
lcu
lated
.
T
h
en
th
e
lo
g
ar
it
h
m
o
f
t
h
ese
m
ea
n
v
al
u
es
p
lo
ts
a
g
ai
n
s
t
th
e
ti
m
e
a
n
d
th
e
s
lo
p
e
o
f
t
h
e
r
es
u
lti
n
g
li
n
e
ar
e
co
n
s
id
er
ed
as
L
L
E
.
T
h
e
e
m
b
ed
d
in
g
d
i
m
e
n
s
io
n
an
d
th
e
lag
ar
e
s
elec
ted
to
b
e
m
=
1
0
an
d
t
=
1
,
r
esp
ec
tiv
el
y
.
T
h
e
t
h
r
esh
o
ld
d
i
s
tan
ce
e
is
s
elec
ted
to
b
e
m
S
D,
w
h
er
e
SD
i
s
t
h
e
s
ta
n
d
ar
d
d
ev
iatio
n
o
f
t
h
e
R
R
ti
m
e
s
er
ie
s
.
Sp
E
n:
T
h
e
Sp
ec
tr
al
E
n
tr
o
p
y
s
h
o
w
s
t
h
e
co
m
p
le
x
it
y
o
f
t
h
e
i
n
p
u
t
ti
m
e
s
er
ies
(
H
R
V
s
eg
m
e
n
t)
i
n
t
h
e
f
r
eq
u
en
c
y
d
o
m
a
in
.
L
ar
g
e
v
a
l
u
es
o
f
Sp
E
n
s
h
o
w
h
ig
h
ir
r
eg
u
lar
it
y
an
d
s
m
a
ller
v
al
u
e
s
o
f
it
in
d
icate
m
o
r
e
r
eg
u
lar
ti
m
e
s
er
ie
s
.
T
h
e
Sh
a
n
n
o
n
’
s
c
h
a
n
n
e
l e
n
tr
o
p
y
i
s
u
s
ed
to
esti
m
ate
t
h
e
s
p
ec
tr
al
e
n
tr
o
p
y
o
f
th
e
p
r
o
ce
s
s
as:
=
−
∑
l
og
w
h
er
e
is
th
e
P
DF
(
p
r
o
b
ab
ilit
y
d
e
n
s
i
t
y
f
u
n
ctio
n
)
v
al
u
e
at
f
r
eq
u
en
c
y
f
.
Heu
r
i
s
ticall
y
,
th
e
en
tr
o
p
y
i
s
en
ter
p
r
eted
as
a
m
ea
s
u
r
e
o
f
u
n
ce
r
tain
t
y
ab
o
u
t
t
h
e
ev
e
n
t
at
f
.
T
h
u
s
e
n
tr
o
p
y
ca
n
b
e
u
s
ed
a
s
a
m
ea
s
u
r
e
o
f
s
y
s
te
m
co
m
p
lex
i
t
y
.
T
h
e
s
p
ec
tr
al
en
tr
o
p
y
H
d
escr
ib
es th
e
co
m
p
lex
i
t
y
o
f
th
e
H
R
V.
D2
:
T
h
e
C
o
r
r
elatio
n
Dim
e
n
s
io
n
is
a
m
ea
s
u
r
e
o
f
co
m
p
le
x
it
y
o
f
th
e
ti
m
e
s
er
ies
an
d
d
eter
m
i
n
es
th
e
m
i
n
i
m
u
m
n
u
m
b
er
o
f
d
y
n
a
m
ic
v
ar
iab
les
w
h
ich
ca
n
m
o
d
el
t
h
e
s
y
s
te
m
.
2
.
1
.
3
F
ea
t
ure
Select
io
n Usin
g
B
F
O
I
n
th
is
s
ec
tio
n
,
f
ea
t
u
r
es
o
f
h
e
ar
t
r
ate
ar
e
ex
tr
ac
ted
.
T
h
e
ex
tr
ac
ted
f
ea
tu
r
es
ar
e
r
ed
u
ce
d
f
u
r
th
er
b
y
u
s
i
n
g
B
ac
ter
ia
Fo
r
ag
in
g
Op
ti
m
izatio
n
to
r
e
m
o
v
e
r
ed
u
n
d
an
c
y
an
d
ir
r
elev
a
n
t
f
ea
t
u
r
es.
T
h
e
r
esu
lti
n
g
f
ea
t
u
r
e
s
u
b
s
et
(
o
b
tain
ed
b
y
B
FO)
is
t
h
e
m
o
s
t r
ep
r
esen
tati
v
e
s
u
b
s
et
a
n
d
is
u
s
ed
to
i
m
p
r
o
v
e
th
e
cla
s
s
if
icatio
n
r
esu
lt.
B
a
ct
er
ia
Repre
s
ent
a
t
io
n
E
ac
h
b
ac
ter
ia’
s
p
o
s
it
io
n
r
e
p
r
esen
t
o
n
e
p
o
s
s
ib
le
s
o
lu
ti
o
n
(
f
ea
t
u
r
e
s
u
b
s
et)
r
eq
u
ir
ed
f
o
r
f
ac
e
r
ec
o
g
n
itio
n
.
T
h
e
n
u
m
b
er
o
f
d
i
m
e
n
s
io
n
s
o
f
s
ea
r
c
h
s
p
ac
e
is
m
w
h
er
e
m
is
t
h
e
len
g
t
h
o
f
f
ea
tu
r
e
v
ec
to
r
(
FV)
ex
tr
ac
ted
f
r
o
m
s
ec
tio
n
2
.
1
.
2
.
I
n
ea
ch
d
i
m
e
n
s
io
n
o
f
s
ea
r
ch
s
p
ac
e,
b
ac
ter
ia
p
o
s
itio
n
i
s
1
o
r
0
,
w
h
er
e
1
o
r
0
in
d
icate
s
t
h
at
t
h
i
s
f
ea
t
u
r
e
is
s
elec
ted
o
r
n
o
t
s
elec
ted
,
r
esp
ec
tiv
el
y
,
as
r
eq
u
ir
ed
f
ea
tu
r
e
f
o
r
n
ex
t
g
e
n
er
atio
n
.
I
n
th
e
ea
ch
i
ter
atio
n
o
f
c
h
e
m
o
t
ax
is
s
tep
,
ea
ch
b
ac
ter
ia
tu
m
b
les
to
th
e
n
e
w
r
an
d
o
m
p
o
s
iti
o
n
.
P
o
s
itio
n
o
f
th
b
ac
ter
ia
in
th
c
h
e
m
o
ta
x
is
a
n
d
th
r
ep
r
o
d
u
ctio
n
s
tep
is
d
ef
i
n
ed
as:
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
AA
S
I
SS
N:
2252
-
8814
Tit
le
o
f m
a
n
u
s
crip
t is sh
o
r
t a
n
d
clea
r
,
imp
lies
r
esea
r
ch
r
esu
l
ts
(
F
i
r
s
t A
u
th
o
r
)
73
(
,
)
=
1
2
,
.
.
(
1
)
W
h
er
e,
m
is
t
h
e
len
g
t
h
o
f
f
e
atu
r
e
v
ec
to
r
ex
tr
ac
ted
.
E
ac
h
=
1
o
r
0
(
=
1
,
2
,
.
.
)
Dep
en
d
in
g
u
p
o
n
w
h
e
th
er
zt
h
f
ea
t
u
r
e
is
s
e
l
ec
ted
o
r
n
o
t f
o
r
th
e
n
ex
t i
ter
ati
o
n
.
Fit
n
e
s
s
F
u
n
ctio
n
:
I
n
ea
ch
g
e
n
e
r
atio
n
,
ea
ch
b
ac
ter
iu
m
i
s
ev
al
u
ated
,
an
d
a
v
alu
e
o
f
g
o
o
d
n
es
s
o
r
f
itn
es
s
is
r
etu
r
n
ed
b
y
a
f
it
n
es
s
f
u
n
ctio
n
.
T
h
is
ev
o
l
u
tio
n
i
s
d
r
iv
e
n
b
y
th
e
f
itn
e
s
s
f
u
n
c
tio
n
F
[
5
]
.
L
et
1
,
2
,
.
.
.
,
an
d
1
,
2
.
.
.
d
en
o
te
t
h
e
clas
s
es
an
d
n
u
m
b
er
o
f
s
i
g
n
als
w
it
h
i
n
ea
ch
clas
s
,
r
esp
ec
tiv
e
l
y
.
L
et
1
,
2
,
an
d
0
b
e
th
e
m
ea
n
s
o
f
co
r
r
esp
o
n
d
in
g
class
es a
n
d
th
e
g
r
an
d
m
ea
n
i
n
th
e
f
ea
t
u
r
e
s
p
ac
e,
ca
n
b
e
ca
lcu
lated
as:
=
1
∑
(
)
,
=
1
∀
=
1
,
2
,
…
,
(
2
)
W
h
er
e
(
i)
,
=
1
,
2
,
…
,
,
r
ep
r
esen
ts
t
h
e
s
a
m
p
le
h
ea
r
t
r
ate
s
ig
n
al
f
r
o
m
cla
s
s
an
d
g
r
an
d
m
ea
n
0
is
:
0
=
1
∑
=
1
(
3
)
W
h
er
e
is
th
e
to
tal
n
u
m
b
er
o
f
h
ea
r
t r
ate
d
ata
s
ig
n
al
o
f
all
th
e
class
e
s
.
T
h
u
s
t
h
e
b
et
w
ee
n
cla
s
s
s
ca
tte
r
f
it
n
es
s
f
u
n
ctio
n
is
co
m
p
u
ted
a
s
f
o
llo
w
s
:
=
√
∑
(
−
0
)
(
−
0
)
=
1
(
4
)
T
h
e
alg
o
r
ith
m
p
r
o
p
o
s
ed
f
o
r
f
ea
t
u
r
e
ex
tr
ac
tio
n
u
s
in
g
B
FO
.
T
h
er
e
ar
e
ce
r
tain
v
ar
iatio
n
s
i
n
B
F
O
alg
o
r
ith
m
u
s
ed
i
n
th
is
w
o
r
k
.
Firstl
y
,
s
tep
6
.
6
o
f
th
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
m
o
v
es
t
h
e
b
ac
ter
ia
b
ac
k
to
its
p
r
ev
io
u
s
p
o
s
itio
n
i
f
c
u
r
r
en
t
p
o
s
itio
n
i
s
les
s
s
u
itab
le
(
c
h
ec
k
ed
u
s
i
n
g
f
itn
e
s
s
f
u
n
c
tio
n
)
.
So
in
t
h
i
s
al
g
o
r
ith
m
,
b
ac
ter
ia
h
av
e
“m
e
m
o
r
y
”
as
t
h
e
y
r
e
m
e
m
b
er
t
h
eir
p
r
ev
io
u
s
p
o
s
itio
n
.
Seco
n
d
l
y
,
as
t
h
er
e
ar
e
ch
a
n
ce
s
t
h
at
b
ac
ter
ia
m
a
y
g
et
s
tr
u
ck
i
n
lo
ca
l
o
p
tim
a,
eli
m
in
a
tio
n
d
is
p
er
s
al
r
em
o
v
es
b
ac
ter
ia
f
r
o
m
its
cu
r
r
en
t
p
o
s
itio
n
an
d
m
o
v
e
s
it
to
“
r
an
d
o
m
”
n
e
w
p
o
s
itio
n
.
I
n
t
h
e
p
r
o
p
o
s
ed
alg
o
r
it
h
m
,
p
o
s
itio
n
o
f
b
ac
ter
ia
is
d
ec
id
ed
r
an
d
o
m
l
y
i
n
th
e
ea
ch
iter
atio
n
.
T
h
er
e
is
n
o
n
ee
d
o
f
u
s
in
g
E
li
m
i
n
atio
n
D
is
p
er
s
al.
2
.
2
.
K
er
nelized
F
uzzy
C
-
M
ea
ns
Alg
o
rit
h
m
T
h
e
w
el
l
-
k
n
o
w
n
C
o
v
er
’
s
th
eo
r
e
m
s
tates
t
h
at
if
a
d
ataset
is
n
o
t
lin
ea
r
l
y
s
ep
ar
ab
le;
tr
an
s
f
o
r
m
it
in
to
a
h
ig
h
er
d
i
m
en
s
io
n
al
s
p
ac
e
n
o
n
-
lin
ea
r
l
y
.
T
h
u
s
,
th
e
n
e
w
l
y
o
b
tain
ed
d
ataset
is
m
o
r
e
lik
el
y
to
b
e
lin
ea
r
l
y
s
ep
ar
ab
le.
Her
e,
th
e
n
o
n
-
l
in
e
ar
tr
an
s
f
o
r
m
atio
n
o
f
t
h
e
d
atas
et
in
to
a
h
i
g
h
er
d
i
m
en
s
io
n
al
s
p
ac
e
is
p
er
f
o
r
m
ed
w
it
h
s
o
m
e
n
o
n
-
li
n
ea
r
k
er
n
el
f
u
n
ctio
n
s
.
T
h
e
co
n
ce
p
t
o
f
“
k
er
n
el”
i
s
f
ir
s
t
attr
ac
ted
g
r
ea
t
atten
tio
n
w
it
h
t
h
e
in
tr
o
d
u
ctio
n
o
f
s
u
p
p
o
r
t v
ec
to
r
m
ac
h
in
e
s
(
SVM)
b
y
C
o
r
tes a
n
d
Vap
n
ik
[
6
]
.
T
h
is
id
ea
is
th
e
n
u
tili
ze
d
b
y
Z
h
a
n
g
an
d
C
h
e
n
[
7
]
in
f
u
zz
y
c
-
m
ea
n
s
c
lu
s
ter
in
g
a
lg
o
r
it
h
m
a
n
d
th
e
y
p
r
o
p
o
s
ed
th
e
k
er
n
eli
ze
d
f
u
zz
y
c
-
m
ea
n
s
alg
o
r
ith
m
(
KF
C
M)
.
T
h
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
i
s
r
ea
lized
b
y
r
estatin
g
th
e
d
is
ta
n
ce
f
u
n
ctio
n
in
FC
M
al
g
o
r
ith
m
w
it
h
a
k
er
n
el
-
b
ased
f
u
n
c
tio
n
.
T
h
e
o
b
j
ec
tiv
e
f
u
n
ctio
n
o
f
th
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
t
h
e
n
ca
n
b
e
g
iv
e
n
as
f
o
llo
w
s
:
=
∑
∑
‖
(
)
−
(
)
‖
2
=
1
=
1
=
2
∑
∑
(
1
−
(
,
)
)
=
1
=
1
(
5
)
W
h
er
e
is
th
e
m
ap
p
in
g
f
u
n
cti
o
n
.
I
n
[
7
]
u
s
ed
a
Gau
s
s
ia
n
k
er
n
el
f
u
n
ct
io
n
f
o
r
n
o
n
-
li
n
ea
r
m
a
p
p
in
g
o
f
th
e
d
ataset.
A
Ga
u
s
s
ian
k
er
n
el
f
u
n
ctio
n
ca
n
b
e
g
i
v
e
n
w
it
h
th
e
f
o
llo
w
in
g
f
o
r
m
u
la:
(
,
)
=
(
−
(
,
)
2
2
)
(
6
)
T
h
en
b
y
s
u
b
s
tit
u
ti
n
g
E
q
.
(
6
)
in
to
E
q
.
(
5
)
=
2
∑
∑
(
1
−
(
−
(
,
)
2
2
)
)
=
1
=
1
(
7
)
B
y
s
etti
n
g
t
h
e
d
er
iv
ati
v
e
o
f
t
h
e
o
b
j
ec
tiv
e
f
u
n
ctio
n
to
ze
r
o
,
an
d
ca
n
b
e
f
o
u
n
d
as f
o
llo
w
s
:
=
(
1
−
(
,
)
)
−
1
(
−
1
)
∑
(
1
−
(
,
)
)
−
1
(
−
1
)
=
1
=
∑
(
∙
)
=
1
∑
=
1
(
∙
)
(
8
)
T
h
u
s
,
b
y
ap
p
l
y
i
n
g
K
FC
M
cl
u
s
ter
in
g
o
n
HR
V,
o
b
tain
ed
cl
u
s
t
er
s
w
i
th
i
n
ac
ti
v
e
s
tate
a
n
d
ac
ti
v
e
s
tate
r
es
u
lt.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8814
IJ
AA
S
Vo
l.
6
,
No
.
1
,
M
ar
ch
20
1
7
:
xx
–
xx
74
2
.
3
.
H
ea
rt
Ra
t
e
Da
t
a
cl
a
s
s
if
ica
t
io
n Usin
g
I
E
L
M
E
x
tr
e
m
e
lear
n
i
n
g
m
ac
h
in
e
(
E
L
M)
i
s
a
n
e
f
f
icien
t
al
g
o
r
ith
m
f
o
r
s
i
n
g
le
-
h
id
d
en
la
y
er
f
ee
d
f
o
r
w
ar
d
n
eu
r
al
n
et
w
o
r
k
s
(
SLFNs),
wh
ich
ca
n
p
r
o
d
u
ce
g
o
o
d
g
e
n
e
r
aliza
tio
n
p
er
f
o
r
m
a
n
ce
in
m
o
s
t
ca
s
e
s
a
n
d
lear
n
th
o
u
s
an
d
s
o
f
ti
m
es
f
a
s
ter
t
h
an
co
n
v
e
n
tio
n
al
p
o
p
u
lar
al
g
o
r
ith
m
s
.
Ho
w
e
v
er
,
th
e
p
er
f
o
r
m
an
ce
o
f
E
L
M
is
s
en
s
iti
v
e
to
th
e
in
itialized
n
u
m
b
er
o
f
h
id
d
en
n
e
u
r
o
n
s
.
I
n
s
o
m
e
tr
ad
itio
n
al
m
et
h
o
d
s
,
th
e
n
u
m
b
er
o
f
h
id
d
en
n
o
d
es
is
g
r
ad
u
all
y
in
cr
ea
s
ed
b
y
a
f
i
x
ed
in
ter
v
al
to
s
elec
t
th
e
n
ea
r
l
y
o
p
ti
m
al
n
u
m
b
er
o
f
n
o
d
es
f
o
r
E
L
M,
w
h
er
ea
s
t
h
ese
m
et
h
o
d
s
ar
e
o
f
a
litt
le
b
it
o
f
co
m
p
le
x
it
y
a
n
d
q
u
ite
ti
m
e
-
co
n
s
u
m
i
n
g
.
T
h
is
w
o
r
k
p
r
o
p
o
s
es
an
i
m
p
r
o
v
ed
E
L
M
b
ased
o
n
KFC
M
clu
s
ter
in
g
,
w
h
ic
h
d
o
es
n
o
t
n
ee
d
to
d
ef
in
e
th
e
n
u
m
b
er
o
f
h
id
d
en
n
o
d
es
in
ad
v
an
ce
m
a
n
u
a
ll
y
an
d
r
a
n
d
o
m
l
y
.
T
h
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
au
to
m
atica
ll
y
d
eter
m
in
e
s
t
h
e
n
u
m
b
er
o
f
h
id
d
en
n
o
d
es
f
o
r
d
if
f
er
en
t
d
ata
s
ets.
E
m
p
ir
ical
s
t
u
d
y
o
f
KF
C
M
-
b
as
ed
E
L
M
o
n
s
e
v
er
al
co
m
m
o
n
l
y
u
s
ed
clas
s
i
f
icatio
n
b
en
ch
m
ar
k
p
r
o
b
le
m
s
s
h
o
w
s
t
h
at
it a
ch
ie
v
es b
etter
p
er
f
o
r
m
an
ce
co
m
p
ar
ed
w
ith
t
h
e
s
ta
n
d
ar
d
E
L
M.
T
h
e
m
ai
n
id
ea
o
f
t
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
lies
in
d
eter
m
i
n
i
n
g
t
h
e
n
u
m
b
er
o
f
h
id
d
en
n
eu
r
o
n
s
i
n
t
h
e
SLFNs.
I
n
E
L
M
[
8
]
alg
o
r
ith
m
,
t
h
e
n
u
m
b
er
o
f
h
id
d
en
n
e
u
r
o
n
s
i
s
r
an
d
o
m
l
y
c
h
o
s
e
n
i
n
th
e
b
eg
i
n
n
i
n
g
o
f
lear
n
in
g
,
r
esu
lti
n
g
i
n
u
n
ce
r
ta
in
n
et
w
o
r
k
ar
ch
itect
u
r
e.
Us
u
all
y
is
n
o
t
k
n
o
w
n
t
h
at
h
o
w
ap
p
r
o
p
r
iate
it
is
to
r
an
d
o
m
l
y
d
eter
m
i
n
e
t
h
e
n
u
m
b
er
o
f
h
id
d
en
n
o
d
es
f
o
r
d
i
f
f
er
e
n
t
d
ate
s
ets.
I
n
m
o
s
t
ca
s
es,
t
h
e
n
u
m
b
er
o
f
h
id
d
en
n
o
d
es
is
g
r
ad
u
all
y
i
n
cr
ea
s
ed
b
y
f
i
x
ed
i
n
ter
v
al
an
d
t
h
e
n
ea
r
l
y
o
p
ti
m
al
n
u
m
b
er
o
f
n
o
d
es
f
o
r
E
L
M
is
th
e
n
s
elec
ted
b
ased
o
n
cr
o
s
s
-
v
alid
atio
n
m
et
h
o
d
,
w
h
ic
h
,
h
o
w
e
v
er
,
is
q
u
ite
ti
m
e
-
co
n
s
u
m
in
g
in
t
h
e
e
x
p
er
i
m
e
n
t.
Hu
a
n
g
et
al.
[
9
]
p
r
o
p
o
s
ed
an
in
cr
e
m
e
n
tal
al
g
o
r
ith
m
r
ef
er
r
e
d
to
as
in
cr
e
m
en
ta
l
ex
tr
e
m
e
l
ea
r
n
in
g
m
ac
h
i
n
e
(
I
-
E
L
M)
b
y
i
n
cr
ea
s
i
n
g
h
id
d
en
n
o
d
es
o
n
e
b
y
o
n
e
u
n
ti
l
th
e
n
u
m
b
er
o
f
h
id
d
en
n
o
d
es
h
as
e
x
c
ee
d
ed
th
e
p
r
ed
ef
in
ed
m
ax
i
m
u
m
n
u
m
b
er
o
r
th
e
r
esid
u
al
er
r
o
r
b
ec
o
m
es
les
s
th
an
th
e
e
x
p
ec
ted
o
n
e.
Su
b
s
e
q
u
en
tl
y
,
a
co
n
v
e
x
in
cr
e
m
e
n
tal
e
x
tr
e
m
e
lear
n
in
g
m
ac
h
in
e
(
C
I
-
E
L
M)
w
a
s
p
r
o
p
o
s
ed
in
Hu
an
g
et
al.
[
1
0
]
.
Dif
f
er
en
t
f
r
o
m
I
-
E
L
M,
CI
-
E
L
M
r
ec
alcu
la
tes
t
h
e
o
u
t
p
u
t
w
e
ig
h
t
s
o
f
th
e
e
x
is
t
in
g
h
id
d
en
n
o
d
e
af
ter
a
n
e
w
h
id
d
e
n
n
o
d
e
is
ad
d
ed
.
I
t
o
b
tain
s
a
f
a
s
ter
co
n
v
er
g
e
n
ce
r
ate
w
h
ile
r
e
m
ain
in
g
th
e
I
-
E
L
M’
s
ef
f
icie
n
c
y
.
I
n
o
r
d
er
to
d
ec
r
ea
s
e
th
e
n
e
t
w
o
r
k
co
m
p
le
x
it
y
a
n
d
o
b
tain
m
o
r
e
c
o
m
p
ac
t
n
et
w
o
r
k
ar
ch
itec
tu
r
e,
Hu
a
n
g
e
t
al.
[
1
1
]
p
r
o
p
o
s
ed
an
en
h
an
ce
d
m
et
h
o
d
f
o
r
I
-
E
L
M
(
ca
lled
E
I
-
E
L
M)
.
A
t
ea
c
h
lear
n
i
n
g
s
tep
,
s
e
v
er
al
h
id
d
en
n
o
d
es
ar
e
r
an
d
o
m
l
y
g
e
n
er
ated
an
d
a
m
o
n
g
th
e
m
t
h
e
h
id
d
en
n
o
d
e
lead
in
g
to
th
e
lar
g
e
s
t
r
esid
u
al
er
r
o
r
d
ec
r
ea
s
in
g
w
ill
o
n
l
y
b
e
ad
d
ed
to
th
e
ex
is
t
in
g
n
e
t
w
o
r
k
.
Fen
g
et
al.
[
1
2
]
later
p
r
o
p
o
s
ed
an
ap
p
r
o
ac
h
r
ef
er
r
ed
to
as
er
r
o
r
m
i
n
i
m
ized
ex
tr
e
m
e
lear
n
i
n
g
m
ac
h
in
e
(EM
-
E
L
M)
,
w
h
ic
h
ad
d
s
r
an
d
o
m
h
id
d
en
n
o
d
es
to
S
L
FN
s
o
n
e
b
y
o
n
e
o
r
g
r
o
u
p
b
y
g
r
o
u
p
an
d
in
cr
e
m
en
ta
ll
y
u
p
d
ates
t
h
e
o
u
tp
u
t
w
eig
h
t
s
d
u
r
in
g
th
e
g
r
o
w
t
h
o
f
t
h
e
n
et
w
o
r
k
.
Ho
w
e
v
er
,
th
er
e
e
x
i
s
ts
m
o
r
e
o
r
les
s
co
m
p
lex
it
y
in
t
h
e
alg
o
r
it
h
m
s
ab
o
v
e,
all
o
f
w
h
ic
h
g
r
ad
u
a
ll
y
i
n
cr
ea
s
e
th
e
n
u
m
b
er
o
f
h
id
d
en
n
o
d
es
d
u
r
in
g
th
e
d
ata
p
r
o
ce
s
s
in
g
.
I
n
o
r
d
er
to
r
ed
u
ce
th
e
co
m
p
le
x
it
y
o
f
d
eter
m
i
n
in
g
t
h
e
n
u
m
b
er
o
f
h
id
d
en
n
e
u
r
o
n
s
,
a
K
FC
M
-
b
ased
E
L
M
al
g
o
r
ith
m
is
p
r
o
p
o
s
ed
in
th
is
s
ec
ti
o
n
i
n
v
ie
w
o
f
t
h
e
ad
v
an
ta
g
e
s
o
f
A
P
clu
s
ter
in
g
.
Fir
s
tl
y
,
K
FC
M
u
s
ed
to
clu
s
ter
t
h
e
g
i
v
e
n
s
a
m
p
les.
Sec
o
n
d
l
y
,
th
e
n
u
m
b
er
o
f
h
id
d
en
n
o
d
es in
th
e
S
L
FN is d
eter
m
in
e
d
b
y
t
h
e
n
u
m
b
er
o
f
clu
s
ter
s
af
ter
th
e
KF
C
M
clu
s
t
er
in
g
.
S
u
m
m
ar
izi
n
g
,
th
e
p
r
o
p
o
s
ed
KFC
M
-
b
ased
E
L
M
al
g
o
r
ith
m
ca
r
r
ies
o
u
t
t
h
e
f
o
llo
w
in
g
p
r
o
ce
s
s
i
n
g
:
KFC
M
-
b
ased
E
L
M
Alg
o
r
it
h
m
: G
iv
e
n
a
tr
ain
i
n
g
s
et
an
d
ac
tiv
atio
n
f
u
n
ct
io
n
(
)
=
1
(
1
+
(
−
)
)
Step
1
:
Sep
ar
ate
th
e
d
ata
s
et
s
in
to
t
w
o
n
o
n
-
o
v
er
lap
p
in
g
s
u
b
s
ets
i
n
cl
u
d
in
g
eq
u
al
n
u
m
b
er
o
f
s
a
m
p
les
f
o
r
tr
ain
i
n
g
a
n
d
tes
ti
n
g
.
Step
2
: A
ll t
h
e
in
p
u
ts
(
attr
ib
u
t
es)
in
th
e
tr
ai
n
i
n
g
s
u
b
s
et
ar
e
n
o
r
m
al
ized
in
to
th
e
r
a
n
g
e
[
0
,
1
]
.
Step
3
:
KFC
M
cl
u
s
ter
i
n
g
.
C
a
lcu
late
t
h
e
n
u
m
b
er
o
f
cl
u
s
ter
s
o
f
th
e
tr
ai
n
i
n
g
s
u
b
s
e
t
an
d
d
eter
m
i
n
e
th
e
h
id
d
e
n
n
o
d
e
s
ize
in
S
L
FN a
cc
o
r
d
in
g
t
o
it.
Step
4
: Ran
d
o
m
l
y
as
s
ig
n
h
id
d
en
n
o
d
e
p
ar
a
m
eter
s
: i
n
p
u
t
w
ei
g
h
t
v
ec
to
r
an
d
h
id
d
en
n
o
d
e
b
ias
.
Step
5
: Calcu
late
t
h
e
h
id
d
en
la
y
er
o
u
tp
u
t
m
atr
i
x
H
u
s
in
g
t
h
e
lear
n
in
g
s
u
b
s
et.
Step
6
: Calcu
late
t
h
e
o
u
tp
u
t
weig
h
t
m
atr
i
x
β.
3.
E
XP
E
R
I
M
E
NT
A
L
RE
SUL
T
S AN
D
D
I
S
C
USS
I
O
N
T
h
e
MI
T
-
B
I
H
ar
r
h
y
t
h
m
ia
d
at
ab
ase
is
u
s
ed
a
s
t
h
e
d
ata
s
o
u
r
ce
in
th
i
s
s
tu
d
y
w
h
ic
h
i
s
d
ev
elo
p
ed
b
y
Ma
r
k
&
Mo
o
d
y
(
1
9
9
7
)
[
A
v
ailab
le:
h
ttp
://ec
g
.
m
it.e
d
u
/d
b
i
n
f
o
.
h
t
m
l
]
.
T
h
e
d
atab
ase
co
n
tain
s
4
8
r
ec
o
r
d
in
g
s
.
E
ac
h
h
as
d
u
r
atio
n
o
f
3
0
m
i
n
u
tes
an
d
in
c
lu
d
es
t
w
o
lead
s
;
t
h
e
m
o
d
if
ied
li
m
b
lead
I
I
an
d
o
n
e
o
f
t
h
e
m
o
d
if
ied
lead
s
V1
,
V2
,
V4
o
r
V5
.
T
h
e
s
a
m
p
li
n
g
f
r
eq
u
en
c
y
is
3
6
0
Hz
,
th
e
d
ata
ar
e
b
an
d
p
ass
f
i
lter
ed
at
0
.
1
–
1
0
0
Hz
an
d
th
e
r
eso
l
u
tio
n
is
2
0
0
s
a
m
p
l
es
p
er
m
V.
T
w
e
n
t
y
-
th
r
ee
o
f
t
h
e
r
ec
o
r
d
in
g
s
ar
e
in
te
n
d
ed
to
s
er
v
e
as
a
r
ep
r
esen
tativ
e
s
a
m
p
le
o
f
r
o
u
ti
n
e
cli
n
ical
r
ec
o
r
d
in
g
s
an
d
2
5
r
ec
o
r
d
in
g
s
co
n
tai
n
co
m
p
lex
v
e
n
tr
ic
u
lar
,
j
u
n
ctio
n
al,
a
n
d
s
u
p
r
av
e
n
tr
icu
lar
ar
r
h
y
t
h
m
ias.
T
h
er
e
ar
e
o
v
er
1
0
9
,
0
0
0
lab
elled
v
en
tr
icu
lar
b
ea
ts
f
r
o
m
1
5
d
if
f
er
e
n
t
h
ea
r
tb
ea
t
t
y
p
es.
T
h
er
e
is
a
lar
g
e
d
if
f
er
en
ce
in
t
h
e
n
u
m
b
er
o
f
ex
a
m
p
les
in
ea
c
h
h
e
ar
t
b
ea
t
ty
p
e.
T
h
e
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
AA
S
I
SS
N:
2252
-
8814
Tit
le
o
f m
a
n
u
s
crip
t is sh
o
r
t a
n
d
clea
r
,
imp
lies
r
esea
r
ch
r
esu
l
ts
(
F
i
r
s
t A
u
th
o
r
)
75
lar
g
est
c
lass
is
―No
r
m
al
b
ea
t‖
w
it
h
ab
o
u
t
7
5
,
0
0
0
ex
am
p
les
an
d
th
e
s
m
alles
t
clas
s
i
s
―Su
p
r
av
e
n
tr
ic
u
lar
p
r
em
at
u
r
e
b
ea
t‖
(
SP
)
w
it
h
j
u
s
t
t
w
o
e
x
a
m
p
les.
T
h
e
d
atab
ase
is
i
n
d
ex
ed
b
o
th
i
n
ti
m
i
n
g
in
f
o
r
m
at
io
n
a
n
d
b
ea
t
class
i
f
icatio
n
.
Fo
r
m
o
r
e
d
etails ab
o
u
t M
I
T
B
I
H
A
r
r
h
y
th
m
ia
d
atab
ase
s
ee
[
1
4
]
.
A
to
tal
o
f
1
2
r
ec
o
r
d
s
ar
e
u
s
e
d
f
r
o
m
th
e
d
atab
ase
a
n
d
a
to
t
al
o
f
3
0
,
8
7
3
b
ea
ts
;
2
2
,
4
7
6
n
o
r
m
al
b
ea
ts
,
5
,
3
9
4
a
b
n
o
r
m
al
P
VC
ar
r
h
y
t
h
m
ia
b
ea
ts
,
a
n
d
3
,
0
0
3
o
th
er
ar
r
h
y
t
h
m
ic
b
ea
ts
ar
e
e
x
tr
ac
ted
.
T
h
e
d
atab
ase
in
d
ex
f
iles
u
s
ed
f
r
o
m
d
atab
ase
to
lo
c
ate
b
ea
ts
i
n
E
C
G
s
i
g
n
a
ls
.
Fo
r
ex
a
m
p
le
t
h
e
p
r
o
p
o
s
ed
w
o
r
k
cl
ass
i
f
ies
w
it
h
t
h
e
s
es
f
i
v
e
clas
s
es
s
u
c
h
a
s
n
o
r
m
al
s
in
u
s
r
h
y
t
h
m
(
NS
R
)
,
p
r
e
m
at
u
r
e
v
en
tr
icu
lar
co
n
tr
ac
tio
n
(
P
V
C
)
,
atr
ial
f
ib
r
illatio
n
(
A
F),
v
e
n
tr
ic
u
lar
f
ib
r
illatio
n
(
VF)
an
d
an
d
2
°
h
ea
r
t
b
lo
ck
(
B
I
I
)
.
I
f
f
o
r
ex
a
m
p
le
a
s
e
g
m
e
n
t
o
f
H
R
V
w
it
h
th
e
VF
ar
r
h
y
t
h
m
ia
is
cla
s
s
i
f
ied
as
th
e
VF,
th
e
n
it
i
s
s
aid
th
at
t
h
e
s
eg
m
en
t
is
clas
s
i
f
ied
T
P
.
O
n
th
e
o
t
h
er
h
an
d
i
f
a
n
o
n
-
VF
s
e
g
m
e
n
t
is
cla
s
s
i
f
ied
as
n
o
n
-
VF,
t
h
e
n
it
is
s
aid
t
h
at
t
h
e
s
eg
m
e
n
t
i
s
clas
s
i
f
ied
T
N.
A
n
y
n
o
n
-
VF
s
eg
m
e
n
t
w
h
ic
h
is
c
lass
if
ied
a
VF
s
e
g
m
e
n
t
b
y
m
is
ta
k
e
w
i
ll
p
r
o
d
u
ce
a
FP
,
w
h
ile
an
y
VF
s
e
g
m
e
n
t
w
h
ich
is
class
i
f
ied
a
n
o
n
-
VF
s
e
g
m
en
t
b
y
m
i
s
ta
k
e
w
i
ll
p
r
o
d
u
ce
a
FN
r
esu
lt.
Fo
r
th
e
ev
a
lu
at
io
n
o
f
p
r
o
p
o
s
ed
class
if
ier
,
a
to
tal
o
f
1
3
1
7
s
eg
m
e
n
t
s
,
w
h
ic
h
ar
e
o
b
tain
ed
o
n
th
e
MI
T
-
B
I
H
ar
r
h
y
th
m
ia
d
atab
ase,
w
er
e
u
s
ed
an
d
it
co
n
s
is
ted
o
f
8
3
5
NSR
s
eg
m
e
n
t
s
,
5
7
PVC
s
e
g
m
en
t
s
,
3
2
2
A
F
s
eg
m
e
n
ts
,
7
8
VF
s
e
g
m
en
ts
a
n
d
2
5
B
I
I
s
e
g
m
e
n
ts
.
T
h
e
i
m
p
le
m
en
ta
tio
n
w
a
s
ex
p
er
i
m
e
n
ted
o
n
a
v
ar
iet
y
o
f
d
ataset
s
a.
Acc
ura
cy
Co
m
pa
ri
s
o
n
T
h
e
p
r
o
p
o
s
ed
I
E
L
M
-
P
SO
alg
o
r
ith
m
p
r
o
d
u
ce
s
b
etter
ac
cu
r
a
c
y
r
ate
w
h
ic
h
i
s
m
u
c
h
g
r
ea
ter
ac
cu
r
ac
y
r
esu
lt
s
th
a
n
ex
i
s
ti
n
g
al
g
o
r
ith
m
s
s
u
c
h
as
SVM,
E
SVM
-
P
SO.
W
h
en
t
h
e
n
u
m
b
er
o
f
f
e
atu
r
es
i
n
cr
ea
s
es
t
h
e
ac
cu
r
ac
y
o
f
t
h
e
r
es
u
lt
i
s
i
n
cr
ea
s
es.
T
h
is
ap
p
r
o
ac
h
p
r
o
d
u
ce
s
h
i
g
h
ac
c
u
r
ac
y
r
ate
w
h
e
n
co
m
p
ar
ed
to
ex
is
t
in
g
s
y
s
te
m
.
b.
F1
-
m
ea
s
ure
Co
m
pa
ri
s
o
n
F1
-
m
ea
s
u
r
e
is
d
ef
i
n
ed
as t
h
e
h
ar
m
o
n
ic
m
ea
n
o
f
p
r
ec
is
io
n
a
n
d
r
ec
all.
A
g
o
o
d
class
i
f
ier
is
ass
u
m
ed
to
h
av
e
a
h
ig
h
F1
-
m
ea
s
u
r
e,
w
h
i
ch
i
n
d
icate
s
t
h
at
th
e
clas
s
if
ier
p
er
f
o
r
m
s
w
e
ll
w
i
th
r
e
s
p
ec
t
to
b
o
th
p
r
ec
is
io
n
(
P
)
an
d
r
ec
all
(
R
)
.
1
−
=
2
×
+
.
T
h
e
p
r
o
p
o
s
e
d
I
E
L
M
-
B
F
O
b
ased
F
S
al
g
o
r
ith
m
p
r
o
d
u
ce
s
h
i
g
h
F1
-
m
ea
s
u
r
e
is
h
i
g
h
er
th
a
n
th
e
e
x
is
t
in
g
alg
o
r
it
h
m
s
s
u
ch
a
s
S
VM
,
E
SVM
-
P
SO.
W
h
en
t
h
e
n
u
m
b
er
o
f
f
ea
t
u
r
es
in
cr
ea
s
es
th
e
F1
-
m
ea
s
u
r
e
o
f
t
h
e
r
es
u
lt
is
in
cr
ea
s
e
s
.
T
h
is
ap
p
r
o
ac
h
p
r
o
d
u
ce
s
ef
f
ec
ti
v
e
F1
-
m
ea
s
u
r
e
r
ate
w
h
e
n
co
m
p
ar
ed
to
ex
is
t
in
g
s
y
s
te
m
.
c.
E
x
ec
utio
n T
i
m
e
Co
m
pa
riso
n
T
h
e
b
est
ac
cu
r
ac
y
r
ate
is
ac
h
i
ev
ed
u
s
i
n
g
th
e
li
n
ea
r
an
d
n
o
n
-
lin
ea
r
f
ea
t
u
r
e
v
ec
to
r
a
n
d
t
h
e
B
FO
-
b
ased
f
ea
t
u
r
e
s
elec
t
io
n
a
lg
o
r
it
h
m
u
s
in
g
o
n
l
y
les
s
n
u
m
b
er
o
f
s
e
lec
ted
f
ea
t
u
r
es
w
h
ich
is
s
i
g
n
i
f
ic
an
tl
y
i
m
p
r
o
v
es
t
h
e
ac
cu
r
ac
y
a
n
d
r
ed
u
ce
s
th
e
ti
m
e
co
m
p
le
x
it
y
.
d.
M
e
a
n Abs
o
lute
E
rr
o
r
I
n
s
tati
s
tic
s
,
th
e
m
ea
n
ab
s
o
lu
t
e
er
r
o
r
(
MA
E
)
is
a
q
u
an
tit
y
u
s
ed
to
m
ea
s
u
r
e
h
o
w
clo
s
e
f
o
r
ec
asts
o
r
p
r
ed
ictio
n
s
ar
e
to
th
e
ev
en
t
u
al
o
u
tco
m
es.
T
h
e
m
ea
n
ab
s
o
lu
te
er
r
o
r
is
g
iv
en
b
y
;
=
1
∑
|
−
|
=
1
=
1
∑
|
|
=
1
As
th
e
n
a
m
e
s
u
g
g
est
s
,
th
e
m
e
an
ab
s
o
lu
te
er
r
o
r
is
an
av
er
a
g
e
o
f
th
e
ab
s
o
lu
te
er
r
o
r
s
|
|
=
|
−
|
,
w
h
er
e
is
t
h
e
p
r
ed
ictio
n
an
d
th
e
tr
u
e
v
alu
e.
Fig
.
5
s
h
o
w
s
t
h
e
No
.
o
f
s
elec
te
d
f
ea
tu
r
es
f
o
r
d
if
f
er
en
t
f
ea
t
u
r
e
v
ec
to
r
d
i
m
en
s
io
n
s
an
d
cla
s
s
i
f
icatio
n
o
f
h
ea
r
t
r
ate
d
ata
is
d
o
n
e
u
s
i
n
g
m
et
h
o
d
s
s
u
c
h
a
s
t
h
e
I
E
L
M
-
B
FO
b
ased
FS
,
E
S
VM
-
P
SO
b
ased
FS
an
d
SV
M
w
it
h
o
u
t
f
ea
t
u
r
e
s
elec
tio
n
al
g
o
r
ith
m
s
.
T
h
e
b
est
ac
cu
r
ac
y
r
a
te
is
ac
h
ie
v
ed
u
s
i
n
g
t
h
e
li
n
ea
r
an
d
n
o
n
-
li
n
ea
r
f
ea
t
u
r
e
v
ec
to
r
an
d
th
e
B
FO
-
b
a
s
ed
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
it
h
m
u
s
in
g
o
n
l
y
less
n
u
m
b
er
o
f
s
elec
ted
f
ea
tu
r
es
a
n
d
th
e
cla
s
s
i
f
icatio
n
is
d
o
n
e
u
s
i
n
g
KF
C
M
b
ased
I
E
L
M
is
s
ig
n
if
ican
t
l
y
r
ed
u
ce
s
t
h
e
M
AE
an
d
r
ed
u
ce
s
t
h
e
co
m
p
u
tatio
n
al
co
m
p
le
x
it
y
.
4.
CO
NCLU
SI
O
N
I
n
th
i
s
w
o
r
k
,
a
K
FC
M
-
b
ased
E
L
M
al
g
o
r
ith
m
i
s
p
r
o
p
o
s
ed
t
o
au
to
m
atica
l
l
y
d
eter
m
i
n
e
th
e
n
u
m
b
er
o
f
n
eu
r
o
n
s
i
n
t
h
e
h
id
d
en
la
y
er
o
f
S
L
FN.
KF
C
M
c
lu
s
ter
in
g
m
et
h
o
d
ap
p
lied
in
t
h
e
a
lg
o
r
ith
m
p
r
o
v
id
es
t
h
e
r
elativ
el
y
r
ea
s
o
n
ab
le
n
u
m
b
er
o
f
h
id
d
en
n
o
d
es
b
ef
o
r
e
tr
ain
i
n
g
t
h
e
g
i
v
e
n
h
ea
r
t
r
ate
d
ata.
T
h
e
B
FO
h
as
b
ee
n
u
s
ed
f
o
r
th
e
f
ea
tu
r
es
s
elec
t
io
n
.
A
p
er
f
o
r
m
an
ce
co
m
p
ar
is
o
n
o
f
t
h
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
with
t
h
e
E
SVM
-
P
SO
an
d
SVM
m
et
h
o
d
s
h
as b
ee
n
c
ar
r
ied
o
u
t o
n
h
ea
r
t r
ate
d
ata.
T
h
e
ex
p
er
i
m
e
n
tal
r
es
u
lt
s
i
n
d
icat
e
th
at
t
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
ac
h
iev
e
s
i
m
p
r
o
v
ed
tes
tin
g
ac
cu
r
ac
y
co
m
p
ar
ed
to
th
e
E
SVM
-
P
SO a
n
d
SVM.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8814
IJ
AA
S
Vo
l.
6
,
No
.
1
,
M
ar
ch
20
1
7
:
xx
–
xx
76
RE
F
E
R
E
NC
E
S
[1
]
Oso
w
s
k
i,
S.,
L
in
h
,
T
.
H.
(
2
0
0
1
)
.
E
C
G
b
ea
t
r
ec
o
g
n
itio
n
u
s
in
g
f
u
zz
y
h
y
b
r
id
n
eu
r
al
n
e
t
w
o
r
k
.
I
E
E
E
Tr
a
n
s
.
B
io
med
.
E
n
g
.
4
8
(
1
1
)
,
1
2
6
5
–
1
2
7
1
.
[2
]
A
r
g
y
r
o
Ka
m
p
o
u
r
ak
i,
Geo
r
g
e
Ma
n
is
,
a
n
d
C
h
r
is
to
p
h
o
r
o
s
Ni
k
o
u
,
Me
m
b
er
,
I
E
E
E
,
”
Hea
r
tb
ea
t
T
im
e
Ser
ies
C
las
s
i
f
icatio
n
W
ith
S
u
p
p
o
r
t
Vec
to
r
Ma
ch
i
n
es”,
I
E
E
E
T
r
a
n
s
a
ctio
n
s
On
I
n
fo
r
ma
tio
n
Tech
n
o
lo
g
y
I
n
B
io
med
icin
e
,
VOL
.
1
3
,
NO
4
J
u
le
2
0
0
9
[3
]
M.
Ka
m
a
th
an
d
E
.
Falle
n
,
“
P
o
w
er
s
p
ec
tr
al
a
n
al
y
s
is
o
f
HR
V:
A
n
o
n
i
n
v
asi
v
e
s
i
g
n
at
u
r
e
o
f
ca
r
d
iac
au
to
n
o
m
ic
f
u
n
ctio
n
s
,
” C
r
it.
R
ev
.
B
io
med
.
E
n
g
.
,
v
o
l.
2
1
,
p
p
.
2
4
5
–
3
1
1
,
1
9
9
3
.
[4
]
R
.
Sil
ip
o
,
G.
Dec
o
,
R
.
Ver
g
a
s
s
o
la,
an
d
C
.
Gr
e
m
i
g
n
i,
“A
c
h
ar
ac
ter
izatio
n
o
f
H
R
V
’
s
n
o
n
lin
ea
r
h
id
d
en
d
y
n
a
m
ics
b
y
m
ea
n
s
o
f
M
ar
k
o
v
m
o
d
els,”
I
E
E
E
Tr
a
n
s
.
B
io
m
ed
.
E
n
g
.
,
v
o
l.
4
6
,
n
o
.
8
,
p
p
.
9
7
8
–
9
8
6
,
A
u
g
.
1999.
[5
]
C
.
L
iu
an
d
H.
W
ec
h
s
ler
,
“
E
v
o
lu
tio
n
ar
y
P
u
r
s
u
it
a
n
d
I
ts
A
p
p
licatio
n
to
Face
R
ec
o
g
n
itio
n
,
”
I
E
E
E
Tr
a
n
s
.
P
a
tter
n
A
n
a
lysi
s
a
n
d
Ma
ch
i
n
e
I
n
tellig
en
ce
,
v
o
l.
2
2
,
n
o
.
6
,
p
p
.
5
7
0
-
5
8
2
,
2
0
0
0
[6
]
C
.
C
o
r
tes,
V.
Vap
n
i
k
,
Ma
ch
i
n
e
L
ea
r
n
i
n
g
2
0
(
1
9
9
5
)
2
7
3
–
2
9
3
.
[7
]
Dao
-
Qia
n
g
Z
h
a
n
g
,
So
n
g
-
C
an
C
h
e
n
,
A
n
o
v
el
k
er
n
e
lized
f
u
z
z
y
C
-
m
ea
n
s
al
g
o
r
ith
m
w
it
h
ap
p
licatio
n
i
n
m
ed
ical
i
m
ag
e
s
eg
m
e
n
tatio
n
,
A
r
tifi
cia
l I
n
tellig
en
ce
in
Med
i
cin
e
3
2
(
S
ep
temb
er (
1
)
)
(
2
0
0
4
)
3
7
–
50,
[8
]
Kar
p
a
g
ac
h
el
v
i,
S.,
A
r
t
h
an
ar
i,
M.
,
&
Si
v
a
k
u
m
ar
,
M.
(
2
0
1
1
)
.
C
las
s
i
f
icatio
n
o
f
E
C
G
s
i
g
n
als
u
s
i
n
g
ex
tr
e
m
e
L
ea
r
n
i
n
g
Ma
ch
in
e.
co
mp
u
ter a
n
d
in
f
o
r
ma
tio
n
s
cien
ce
,
4
(
1
)
,
p
4
2
.
[9
]
Gu
a
n
g
-
B
in
H
u
a
n
g
,
L
ei
C
h
e
n
,
C
h
ee
-
Kh
eo
n
g
S
ie
w
,
"
Un
iv
er
s
al
A
p
p
r
o
x
i
m
a
tio
n
U
s
in
g
I
n
cr
e
m
en
tal
C
o
n
s
tr
u
c
ti
v
e
Feed
f
o
r
w
ar
d
N
et
w
o
r
k
s
w
i
th
R
a
n
d
o
m
Hid
d
en
No
d
es"
,
I
E
E
E
Tr
a
n
s
a
ctio
n
s
o
n
N
eu
r
a
l
N
etw
o
r
k
s
,
v
o
l.
1
7
,
n
o
.
4
,
p
p
.
8
7
9
-
8
9
2
,
2
0
0
6
.
[1
0
]
Gu
a
n
g
-
B
in
H
u
an
g
,
L
ei
C
h
e
n
,
"
C
o
n
v
ex
I
n
cr
e
m
en
ta
l
E
x
tr
e
m
e
L
ea
r
n
i
n
g
Ma
ch
in
e"
,
N
eu
r
o
co
mp
u
tin
g
,
v
o
l.
7
0
,
n
o
.
1
6
-
1
8
,
p
p
.
3
0
5
6
-
3
0
6
2
,
2
0
0
7
.
[1
1
]
Gu
a
n
g
-
B
in
H
u
a
n
g
,
L
ei
C
h
e
n
,
"
E
n
h
an
ce
d
R
a
n
d
o
m
Sear
ch
B
ased
I
n
cr
e
m
en
ta
l
E
x
tr
e
m
e
L
ea
r
n
i
n
g
Ma
ch
i
n
e"
,
N
eu
r
o
co
mp
u
tin
g
,
v
o
l.
7
1
,
n
o
.
1
6
-
1
8
,
p
p
.
3
4
6
0
-
3
4
6
8
,
2
0
0
8
.
[1
2
]
Gu
o
-
R
u
i
Fe
n
g
,
G
u
a
n
g
-
B
i
n
H
u
an
g
,
Qi
n
g
-
P
in
g
L
i
n
,
R
o
b
er
t
Ga
y
.
"
E
r
r
o
r
Min
i
m
ized
E
x
tr
e
m
e
L
ea
r
n
i
n
g
Ma
ch
i
n
e
w
it
h
Gr
o
w
t
h
o
f
Hi
d
d
en
No
d
es
an
d
I
n
cr
e
m
en
tal
L
ea
r
n
in
g
"
,
I
E
E
E
Tr
a
n
s
a
ctio
n
s
o
n
N
eu
r
a
l
N
etw
o
r
k
s
,
v
o
l.
2
0
,
n
o
.
8
,
p
p
.
1
3
5
2
-
1
3
5
7
,
2
0
0
9
.
[1
3
]
G.
B
.
M
o
o
d
y
,
an
d
R
.
G.
Ma
r
k
,
T
h
e
i
m
p
ac
t
o
f
th
e
MI
T
/B
I
H
ar
r
h
y
t
h
m
ia
d
atab
ase
,
I
E
E
E
E
n
g
.
Med
.
B
i
o
l.
Ma
g
.
,
v
o
l.
2
0
,
n
o
.
3
,
p
p
.
45
–
5
0
,
Ma
y
-
J
u
n
.
2
0
0
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.