T
E
L
K
O
M
N
I
K
A
T
elec
o
m
m
un
ica
t
io
n,
Co
m
pu
t
ing
,
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
19
,
No
.
4
,
A
u
g
u
s
t
2
0
2
1
,
p
p
.
1
2
6
0
~
1
2
7
2
I
SS
N:
1
6
9
3
-
6
9
3
0
,
ac
cr
ed
ited
First Gr
ad
e
b
y
Kem
en
r
is
tek
d
i
k
ti,
Dec
r
ee
No
: 2
1
/E/KPT
/2
0
1
8
DOI
: 1
0
.
1
2
9
2
8
/TE
L
KOM
NI
K
A.
v
1
9
i4
.
1
9
3
7
1
1260
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
An auto
ma
tic
sc
r
eening
appro
a
ch f
o
r obs
truct
iv
e sl
e
ep a
pnea
from p
ho
tople
thys
mo
g
ra
ph using
ma
chine learning
t
echniqu
es
Sm
ily
J
ey
a
J
o
t
hi E
.
1
,
Anitha
J
.
2
,
J
ud
e
H
ema
nth
3
1
De
p
a
rtme
n
t
o
f
Bi
o
m
e
d
ica
l
En
g
in
e
e
rin
g
,
Av
in
a
sh
il
in
g
a
m
In
stit
u
te f
o
r
Ho
m
e
S
c
ien
c
e
a
n
d
Hig
h
e
r
Ed
u
c
a
ti
o
n
f
o
r
Wo
m
e
n
,
Co
imb
a
to
re
,
I
n
d
ia
2
,3
De
p
a
rt
m
e
n
t
o
f
El
e
c
tro
n
ics
a
n
d
Co
m
m
u
n
ica
ti
o
n
E
n
g
i
n
e
e
rin
g
,
Ka
ru
n
y
a
I
n
stit
u
te o
f
Tec
h
n
o
lo
g
y
a
n
d
S
c
ien
c
e
s,
Co
imb
a
to
re
,
I
n
d
ia
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
an
5
,
2
0
2
1
R
ev
is
ed
Ma
r
1
7
,
2
0
2
1
Acc
ep
ted
Ma
r
3
0
,
2
0
2
1
Ob
stru
c
ti
v
e
sle
e
p
a
p
n
e
a
(OSA),
a
v
e
ry
c
o
m
m
o
n
sle
e
p
d
iso
rd
e
r
re
m
a
in
s
a
s
a
n
u
n
d
e
r
d
iag
n
o
se
d
r
o
o
t
c
a
u
se
fo
r
s
e
v
e
ra
l
c
a
rd
io
v
a
sc
u
lar
a
n
d
c
e
re
b
r
o
v
a
sc
u
lar
d
ise
a
se
s.
In
th
is
p
a
p
e
r,
we
p
ro
p
o
se
a
n
e
fficie
n
t
a
n
d
a
c
c
u
ra
te
sy
ste
m
th
a
t
u
ti
li
z
e
s
a
si
n
g
le
se
n
s
o
r
f
o
r
e
ff
e
c
ti
v
e
sc
re
e
n
in
g
o
f
OSA
u
sin
g
m
a
c
h
in
e
l
e
a
rn
in
g
a
lg
o
r
it
h
m
s.
T
h
e
a
u
to
m
a
ti
c
sc
re
e
n
in
g
s
y
ste
m
in
v
o
l
v
e
s
a
p
h
o
to
p
leth
y
sm
o
g
ra
m
(P
P
G
)
sig
n
a
l,
a
n
o
v
e
l
a
lg
o
rit
h
m
t
o
d
e
tec
t
a
n
d
re
m
o
v
e
th
e
c
o
rr
u
p
ted
p
a
rt
o
f
t
h
e
sig
n
a
l,
a
fe
a
tu
re
e
x
trac
ti
o
n
m
o
d
u
le
t
o
e
x
tra
c
t
se
v
e
ra
l
fe
a
tu
re
s
fro
m
t
h
e
P
P
G
wa
v
e
fo
rm
a
n
d
a
c
las
sifier
m
o
d
u
le
w
h
ic
h
h
e
lp
s
in
sc
re
e
n
in
g
fo
r
OSA
.
Th
e
e
lem
e
n
tal
id
e
a
b
e
h
in
d
t
h
is
wo
rk
is
t
h
a
t
th
e
re
is
a
c
h
a
ra
c
teristic
re
latio
n
sh
ip
b
e
twe
e
n
th
e
sh
a
p
e
o
f
th
e
P
P
G
wa
v
e
fo
rm
a
n
d
th
e
o
x
y
g
e
n
d
e
sa
tu
ra
ti
o
n
in
th
e
a
p
n
e
a
p
a
ti
e
n
ts.
Th
e
m
e
th
o
d
a
s
d
e
sc
rib
e
d
wa
s
tes
ted
o
n
2
8
5
su
b
jec
ts,
i
n
c
lu
siv
e
o
f
b
o
t
h
n
o
rm
a
l
a
n
d
a
p
n
e
a
p
a
ti
e
n
t
s,
a
n
d
th
e
re
su
lt
s
we
re
o
b
tain
e
d
a
fter
10
-
f
o
l
d
-
c
ro
ss
v
a
l
id
a
ti
o
n
o
f
th
e
d
if
fe
re
n
t
m
a
c
h
in
e
lea
rn
in
g
tec
h
n
iq
u
e
s
v
iz.
,
u
n
iv
a
riate
re
g
re
ss
io
n
,
m
u
l
ti
v
a
riate
r
e
g
re
ss
io
n
,
su
p
p
o
rt
v
e
c
t
o
r
m
a
c
h
in
e
a
n
d
ra
n
d
o
m
fo
re
st.
Th
e
b
e
st r
e
su
lt
s in
sc
re
e
n
in
g
OSA
we
re
o
b
tain
e
d
fr
o
m
ra
n
d
o
m
f
o
re
st
a
lg
o
rit
h
m
with
t
h
e
h
i
g
h
e
st
p
e
rfo
rm
a
n
c
e
(Ac
c
:
9
8
.
0
%
,
S
e
n
:
9
8
.
6
%
,
S
p
e
c:
9
9
.
3
%
)
fo
r
a
ll
t
h
e
c
o
m
b
in
e
d
fe
a
tu
re
s.
Th
e
p
ro
p
o
se
d
w
o
rk
is
a
n
e
ffe
c
ti
v
e
sy
s
tem
fo
r
a
u
to
m
a
ti
c
sc
re
e
n
in
g
o
f
O
S
A
fro
m
a
sin
g
le
P
P
G
se
n
so
r,
th
e
re
b
y
re
d
u
c
in
g
t
h
e
n
e
e
d
f
o
r
a
v
e
ry
e
x
p
e
n
siv
e
a
n
d
o
v
e
rn
i
g
h
t
p
o
ly
s
o
m
n
o
g
ra
p
h
y
sle
e
p
stu
d
y
.
K
ey
w
o
r
d
s
:
Mu
ltiv
ar
iate
r
eg
r
ess
io
n
Ob
s
tr
u
ctiv
e
s
leep
ap
n
ea
Ph
o
to
p
leth
y
s
m
o
g
r
am
R
an
d
o
m
f
o
r
est
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
Un
iv
ar
iate
r
eg
r
ess
io
n
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r
th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
J
u
d
e
Hem
an
th
Dep
ar
tm
en
t o
f
E
lectr
o
n
ics an
d
C
o
m
m
u
n
icatio
n
E
n
g
in
ee
r
i
n
g
Kar
u
n
y
a
I
n
s
titu
te
o
f
T
ec
h
n
o
lo
g
y
an
d
Scein
ce
s
Kar
u
n
y
a
Nag
a
r
,
C
o
im
b
ato
r
e,
6
4
1
1
1
4
,
T
a
m
iln
ad
u
,
I
n
d
ia
E
m
ail:
ju
d
eh
em
an
t
h
@
k
ar
u
n
y
a
.
ed
u
1.
I
NT
RO
D
UCT
I
O
N
Sleep
d
is
o
r
d
er
s
h
as
n
o
w
b
ec
o
m
e
a
v
e
r
y
co
m
m
o
n
h
ea
lth
c
o
n
d
itio
n
a
f
f
ec
tin
g
ab
o
u
t
2
to
4
%
o
f
th
e
ad
u
lt
p
o
p
u
latio
n
an
d
h
a
v
e
e
f
f
ec
t
o
n
s
ev
er
al
asp
ec
ts
o
f
life
.
Am
o
n
g
th
e
s
ix
ty
d
if
f
e
r
e
n
t
s
leep
d
is
o
r
d
er
s
id
en
tifie
d
b
y
th
e
I
n
ter
n
atio
n
al
C
lass
if
icatio
n
o
f
Sleep
Dis
o
r
d
er
s
[
1
]
,
o
b
s
tr
u
ctiv
e
s
leep
ap
n
ea
(
OSA)
is
o
n
e
o
f
th
e
m
o
s
t
co
m
m
o
n
o
n
e,
ch
ar
ac
ter
ized
b
y
ep
is
o
d
es
o
f
co
m
p
lete,
in
ter
m
itten
t
o
r
p
ar
ti
al
o
b
s
tr
u
ctio
n
an
d
r
ep
etitiv
e
co
llap
s
e
o
f
th
e
u
p
p
er
air
way
d
u
r
in
g
s
leep
.
R
esear
ch
r
ep
o
r
ts
o
n
th
e
p
r
ev
alen
c
e
o
f
th
e
s
y
n
d
r
o
m
e
am
o
n
g
th
e
a
d
u
lt p
o
p
u
latio
n
es
tim
ates th
at
it i
s
m
u
ch
h
ig
h
er
i
.
e.
ab
o
u
t
5
0
% in
p
atien
ts
with
ca
r
d
io
v
ascu
lar
an
d
ce
r
eb
r
o
v
ascu
lar
d
is
ea
s
es
[
2
]
.
Ob
s
tr
u
ctiv
e
s
leep
ap
n
ea
is
cu
r
r
en
tly
d
iag
n
o
s
ed
u
s
in
g
p
o
ly
s
o
m
n
o
g
r
ap
h
y
(
PS
G)
an
d
is
co
n
s
id
er
ed
as
th
e
g
o
ld
s
tan
d
ar
d
.
PS
G
is
u
s
u
ally
p
er
f
o
r
m
ed
in
s
leep
la
b
s
as
an
o
v
e
r
n
ig
h
t
s
leep
s
tu
d
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
A
n
a
u
to
m
a
tic
s
creen
in
g
a
p
p
r
o
a
ch
fo
r
o
b
s
tr
u
ctive
s
leep
a
p
n
e
a
fr
o
m
…
(
S
mily
Je
ya
Jo
t
h
i.
E
)
1261
Du
e
to
th
e
lim
itatio
n
o
f
s
leep
lab
s
in
h
o
s
p
itals
an
d
clin
ics
an
d
d
u
e
to
c
o
m
p
le
x
ity
o
f
t
h
e
ab
o
v
e
d
iag
n
o
s
tic
tech
n
iq
u
e,
t
h
e
p
r
o
ce
d
u
r
e
is
q
u
ite
ex
p
e
n
s
iv
e
an
d
c
u
m
b
er
s
o
m
e
to
u
s
e
[
3
]
.
T
h
e
r
ef
o
r
e
,
th
e
n
ee
d
f
o
r
a
n
alter
n
ativ
e,
s
im
p
le
an
d
r
o
b
u
s
t scr
ee
n
in
g
o
f
OSA
is
ju
s
tifie
d
b
y
th
e
s
ize
o
f
th
e
p
o
p
u
latio
n
a
f
f
ec
ted
.
Sev
er
al
ap
p
r
o
ac
h
es
alter
n
ativ
e
to
PS
G
h
av
e
b
ee
n
p
r
o
p
o
s
ed
wh
ich
aid
s
in
th
e
d
ay
tim
e
m
o
n
ito
r
in
g
o
f
OSA
u
s
in
g
in
d
iv
id
u
al
s
ig
n
als
v
iz.
,
r
esp
i
r
ato
r
y
,
elec
tr
o
ca
r
d
i
o
g
r
ap
h
y
(
E
C
G)
,
p
h
o
t
o
p
leth
y
s
m
o
g
r
ap
h
y
(
PP
G)
,
E
E
G,
s
n
o
r
in
g
lo
u
d
n
ess
,
Sp
O
2
o
r
co
m
b
i
n
atio
n
o
f
two
o
r
m
o
r
e
s
ig
n
als.
As
a
r
ep
lace
m
en
t
to
th
ese
m
eth
o
d
s
an
d
to
th
e
o
v
e
r
n
ig
h
t
PS
G
s
tu
d
ie
s
,
th
is
wo
r
k
p
r
o
p
o
s
es
th
e
u
s
e
o
f
a
n
o
n
-
in
v
asiv
e,
l
o
w
-
c
o
s
t
d
ev
ice
s
u
ch
as
p
h
o
to
p
let
h
y
s
m
o
g
r
ap
h
y
(
PP
G)
th
at
ca
n
b
e
ef
f
o
r
tles
s
ly
ap
p
lie
d
ev
en
to
o
u
tp
atien
ts
.
Ucar
et
a
l.
[
4
]
d
escr
ib
es
an
a
p
p
r
o
a
c
h
t
o
w
a
r
d
s
d
e
t
e
r
m
i
n
i
n
g
t
h
e
c
o
n
n
e
c
t
i
o
n
b
e
t
w
e
e
n
t
h
e
r
e
s
p
i
r
a
t
o
r
y
a
r
r
e
s
t
s
a
n
d
P
PG
s
i
g
n
a
l
o
f
O
S
A
p
a
t
i
e
n
ts
a
n
d
h
a
s
d
e
p
i
ct
e
d
s
i
g
n
i
f
i
ca
n
t
as
s
o
ci
a
t
i
o
n
i
n
u
s
i
n
g
PP
G
s
i
g
n
a
l
f
o
r
d
i
a
g
n
o
s
is
o
f
O
S
A
[
4
]
.
T
h
i
s
s
i
g
n
a
l
c
a
n
ea
s
il
y
a
s
s
e
s
s
t
h
e
t
ac
t
i
le
a
r
te
r
i
a
l
p
a
l
p
a
ti
o
n
o
n
t
h
e
f
i
n
g
e
r
’
s
ca
p
i
l
la
r
i
es
b
y
v
i
r
t
u
e
o
f
t
h
e
d
i
f
f
e
r
e
n
c
es
i
n
l
i
g
h
t
a
b
s
o
r
p
t
i
o
n
[
5
]
.
Pre
cisely
,
a
s
im
p
le
an
d
r
o
b
u
s
t n
o
n
-
in
v
asiv
e
d
ev
ice
th
at
ca
n
p
r
o
v
id
e
r
ea
l tim
e
s
cr
ee
n
in
g
f
o
r
OSA
is
th
e
s
u
b
ject
o
f
in
ter
est d
u
e
t
o
its
ea
s
y
av
ailab
ilit
y
an
d
less
d
is
co
m
f
o
r
t to
th
e
p
atien
ts
[
6
]
.
An
ap
p
r
o
ac
h
f
o
r
au
t
o
m
ated
r
ec
o
g
n
itio
n
o
f
OSA
f
r
o
m
E
C
G
r
ec
o
r
d
in
g
s
s
h
o
ws
f
alse
n
eg
ativ
e
r
esu
lts
wh
en
p
atien
ts
wi
th
a
h
is
to
r
y
o
f
ca
r
d
io
v
ascu
lar
d
is
ea
s
es
wer
e
in
clu
d
ed
[
7
]
.
Mo
s
t
r
ec
en
t
r
es
ea
r
ch
ap
p
r
o
ac
h
o
n
au
to
m
ated
OSA
d
etec
tio
n
u
s
in
g
ca
r
d
io
p
u
lm
o
n
ar
y
s
ig
n
al
th
at
u
tili
ze
s
b
o
th
h
ea
r
t
r
ate
s
ig
n
al
an
d
r
esp
ir
atio
n
r
ate
s
ig
n
al
d
e
m
o
n
s
tr
ates
a
r
ed
u
ce
d
s
en
s
itiv
ity
an
d
s
p
ec
if
i
cit
y
f
o
r
s
u
b
ject
-
s
p
ec
if
ic
cr
o
s
s
v
a
lid
atio
n
in
s
p
ite
o
f
its
co
m
p
lex
in
s
tr
u
m
en
ta
tio
n
a
n
d
ca
lib
r
atio
n
p
r
o
ce
d
u
r
es [
8
]
.
An
o
th
er
s
tu
d
y
th
at
m
ak
es u
s
e
o
f
ca
r
d
io
r
esp
ir
ato
r
y
m
o
d
el
-
b
ased
d
ata
-
d
r
iv
en
ap
p
r
o
ac
h
f
o
r
OSA
d
etec
tio
n
co
m
b
in
es
m
ea
s
u
r
em
en
t
s
ig
n
als
f
r
o
m
v
ar
io
u
s
s
en
s
o
r
m
o
d
alities
with
th
e
m
at
h
em
atica
l
m
o
d
el
o
f
ca
r
d
io
r
esp
ir
ato
r
y
s
y
s
tem
in
th
e
co
n
te
x
t
o
f
im
p
r
o
v
in
g
th
e
d
etec
tio
n
p
er
f
o
r
m
an
ce
o
f
OSA.
A
lim
itatio
n
o
f
th
is
m
eth
o
d
is
th
at
th
e
tim
e
d
elay
an
d
th
r
esh
o
ld
v
a
lu
es
o
f
Sp
O
2
v
ar
y
f
r
o
m
o
n
e
in
d
iv
id
u
al
to
an
o
th
e
r
.
T
h
e
d
if
f
icu
lty
lies
in
th
e
f
ac
t
th
at
a
g
en
er
alize
d
an
d
s
u
b
ject
-
in
d
ep
en
d
en
t v
alu
e
co
u
ld
n
o
t
b
e
o
b
tain
ed
[9
]
.
A
r
ec
en
t
r
esear
ch
p
er
s
p
ec
tiv
e
to
ac
cu
r
ately
d
etec
t
s
leep
ap
n
ea
h
as
p
u
t
f
o
r
war
d
th
e
s
ig
n
if
ican
ce
o
f
l
o
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
n
etwo
r
k
b
ased
o
n
R
R
in
ter
v
a
l
s
ig
n
als.
T
h
e
m
a
in
lim
itatio
n
o
f
t
h
e
wo
r
k
h
ap
p
e
n
s
to
b
e
f
r
o
m
t
h
e
lo
w
h
o
ld
v
alid
atio
n
ac
c
u
r
a
cy
o
f
8
1
.
3
0
%
an
d
th
e
in
s
u
f
f
icien
t
win
d
o
w
s
ize
ch
o
s
en
to
ex
t
r
ac
t th
e
R
R
in
ter
v
als [
1
0
]
.
2.
B
ACK
G
RO
UND
I
t
is
im
p
o
r
tan
t
to
u
n
d
er
s
tan
d
t
h
e
p
h
y
s
io
lo
g
ical
f
ac
to
r
s
th
at
s
u
p
p
o
r
t
th
e
th
eo
r
etica
l
esti
m
ati
o
n
o
f
OSA
ch
ar
ac
ter
is
tics
f
r
o
m
th
e
PP
G
s
ig
n
al.
T
h
e
f
o
llo
win
g
s
ess
io
n
d
escr
ib
es
th
e
r
elatio
n
b
etwe
en
t
h
ese
p
h
y
s
io
lo
g
ical
f
ac
to
r
s
an
d
th
e
s
ig
n
al
p
r
o
ce
s
s
in
g
tech
n
i
q
u
es
u
s
ed
t
o
ex
tr
ac
t
th
e
f
ea
tu
r
es
r
elev
an
t
to
ch
a
n
g
es
in
PP
G
d
u
e
to
OSA.
T
h
e
h
em
o
d
y
n
am
ics
an
d
th
e
ca
r
d
io
v
ascu
lar
ac
tiv
ity
o
f
an
OSA
p
atien
t
s
way
b
etw
ee
n
v
en
tilato
r
y
an
d
ap
n
eic
p
er
i
o
d
s
,
d
u
e
to
r
ec
u
r
r
e
n
t
ap
n
ea
s
.
T
h
e
co
n
s
eq
u
en
ce
s
o
f
r
e
p
etitiv
e
s
u
r
g
es
cu
r
b
th
e
n
o
r
m
al
h
ea
r
t
r
ate
a
n
d
b
lo
o
d
p
r
ess
u
r
e
an
d
th
u
s
co
n
tr
i
b
u
te
to
th
e
ad
v
e
r
s
e
ef
f
ec
ts
o
n
th
e
ca
r
d
io
v
ascu
lar
ac
tiv
ity
.
H
y
p
o
x
ia
-
a
co
n
d
itio
n
wh
er
e
o
x
y
g
en
s
u
p
p
ly
to
th
e
ti
s
s
u
es
ar
e
r
ed
u
ce
d
,
m
ag
n
if
ied
n
eg
ativ
e
in
tr
a
-
th
o
r
ac
ic
p
r
ess
u
r
e
p
r
o
m
p
ted
d
u
e
t
o
o
b
s
tr
u
cted
p
h
ar
y
n
x
a
n
d
s
leep
ar
o
u
s
als
ar
e
th
e
m
ain
p
ath
o
p
h
y
s
io
lo
g
ical
f
ea
tu
r
es
o
f
OSA.
T
h
e
v
a
r
iab
ilit
y
in
h
ea
r
t
r
ate
m
ay
d
if
f
er
am
o
n
g
i
n
d
iv
id
u
als
an
d
it
d
ep
en
d
s
o
n
th
e
h
y
p
o
x
ia
s
ev
er
ity
an
d
in
tr
in
s
ic
h
y
p
o
x
ic
ch
em
o
s
en
s
itiv
ity
.
2
.
1
.
P
a
t
ho
ph
y
s
io
lo
g
y
o
f
O
S
A
s
y
nd
ro
m
e
Ob
esit
y
,
th
ick
en
ed
p
h
ar
y
n
g
ea
l
walls,
r
ed
u
ce
d
m
u
s
cle
to
n
e
o
f
n
aso
-
p
h
a
r
y
n
x
d
u
r
in
g
s
leep
;
h
y
p
er
ten
s
io
n
an
d
o
t
h
er
p
ath
o
l
o
g
ies
co
n
tr
ib
u
te
to
OSA
[
1
1
]
.
I
t
also
af
f
ec
ts
th
e
g
en
er
al
h
em
o
d
y
n
a
m
ics
an
d
th
e
s
tate
o
f
th
e
au
to
n
o
m
o
u
s
s
y
s
tem
an
d
it
i
s
r
elate
d
to
t
h
e
in
d
iv
id
u
al’
s
d
em
o
g
r
ap
h
ic
an
d
m
o
r
p
h
o
lo
g
ic
p
ar
am
eter
s
[
1
2
]
.
T
h
u
s
,
m
ea
s
u
r
em
en
t
o
f
p
ar
am
ete
r
s
lik
e
ag
e,
s
ex
,
weig
h
t,
a
n
d
h
eig
h
t
an
d
b
o
d
y
m
ass
in
d
e
x
r
em
ain
s
a
p
r
e
-
r
eq
u
is
ite.
2
.
2
.
O
SA a
nd
blo
o
d v
is
co
s
it
y
B
lo
o
d
v
is
co
s
ity
is
th
e
in
ter
n
al
r
esis
tan
ce
o
f
f
er
e
d
b
y
th
e
b
l
o
o
d
ag
ain
s
t
s
h
ea
r
f
o
r
ce
s
a
n
d
is
d
eter
m
in
e
d
b
y
th
e
v
is
co
s
ity
o
f
p
lasma
,
h
e
m
ato
cr
it
an
d
th
e
b
io
m
ec
h
a
n
ical
p
r
o
p
e
r
ties
o
f
r
e
d
b
l
o
o
d
ce
lls
[
1
3
]
.
C
h
an
g
es
i
n
th
e
r
h
e
o
lo
g
ical
p
r
o
p
er
ties
o
f
b
lo
o
d
a
n
d
p
lasma
lea
d
s
to
an
in
cr
ea
s
e
in
b
lo
o
d
cl
o
ttin
g
a
n
d
p
er
h
ap
s
r
em
ain
s
to
b
e
a
v
ital
f
ac
to
r
in
tr
ig
g
er
in
g
c
ar
d
io
v
ascu
lar
co
m
p
licatio
n
s
d
u
e
to
OSA
[
1
4
]
.
T
h
er
e
is
s
tr
o
n
g
ev
id
e
n
ce
t
h
at
th
e
b
lo
o
d
v
is
co
s
ity
an
d
th
e
p
lasm
a
v
is
co
s
ity
ar
e
ab
n
o
r
m
ally
h
i
g
h
in
OSA
p
atien
ts
[
1
5
]
.
T
h
e
f
lu
ctu
atio
n
s
in
th
e
b
lo
o
d
p
r
ess
u
r
e
ad
v
e
r
s
ely
af
f
ec
t
t
h
e
r
esp
o
n
s
e
o
f
b
lo
o
d
v
ess
el
wall
an
d
th
er
eb
y
m
o
d
if
y
th
e
v
ess
el
co
m
p
lian
ce
[
1
6
]
.
T
h
e
ef
f
ec
t o
f
b
lo
o
d
v
is
co
s
ity
an
d
b
lo
o
d
v
ess
el
co
m
p
lian
ce
ar
e
r
ef
lecte
d
o
n
th
e
s
h
ap
e
o
f
PP
G
wav
ef
o
r
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
4
,
Au
g
u
s
t 2
0
2
1
:
1
2
6
0
-
1
2
7
2
1262
2
.
3
.
O
SA a
nd
hea
rt
r
a
t
e
v
a
r
ia
bil
it
y
T
h
e
ass
o
ciatio
n
b
etwe
en
s
ev
e
r
ity
o
f
OSA
an
d
h
ea
r
t
r
ate
v
a
r
iab
ilit
y
h
as
b
ee
n
s
tu
d
ied
ex
te
n
s
iv
ely
b
y
d
o
in
g
tim
e
d
o
m
ain
an
d
f
r
eq
u
en
cy
d
o
m
ai
n
an
aly
s
is
o
f
h
ea
r
t
r
ate
v
ar
iab
ilit
y
(
HR
V)
,
s
lee
p
ar
o
u
s
als,
o
x
y
g
e
n
d
esatu
r
atio
n
an
d
o
t
h
er
s
leep
p
ar
am
eter
s
.
T
h
e
f
r
e
q
u
en
c
y
d
o
m
ain
in
d
ices
ten
d
to
y
ield
a
b
etter
r
esu
lt
wh
e
n
co
m
p
ar
ed
to
tim
e
d
o
m
ain
in
d
i
ce
s
[
1
7
]
.
Hen
ce
f
o
r
th
,
t
h
e
v
ar
ia
b
ilit
y
in
th
e
h
ea
r
t
r
ate
d
u
e
to
OSA
is
r
ef
lecte
d
in
th
e
p
o
wer
s
p
ec
tr
u
m
o
f
th
e
h
ea
r
t r
ate.
2
.
4
.
O
SA a
nd
ba
ro
re
f
lex
es
T
h
is
r
elate
s
th
e
h
ea
r
t
r
at
e
an
d
th
e
b
lo
o
d
p
r
ess
u
r
e.
T
h
e
f
ac
t
u
n
d
er
ly
i
n
g
is
th
at,
an
i
n
cr
ea
s
e
in
b
lo
o
d
p
r
ess
u
r
e
ac
tiv
ates
th
e
ca
r
o
tid
s
in
u
s
an
d
a
o
r
tic
ar
c
h
b
a
r
o
r
ec
e
p
to
r
s
wh
ich
in
h
ib
its
th
e
s
y
m
p
ath
etic
o
u
tf
lo
w
an
d
r
ed
u
ce
s
th
e
h
ea
r
t
r
ate.
T
h
e
o
p
p
o
s
ite
ef
f
ec
t
is
ex
p
er
ien
ce
d
d
u
e
t
o
a
d
e
c
r
ea
s
e
in
b
lo
o
d
p
r
ess
u
r
e
[
1
8
]
.
So
lar
o
et
a
l.
[
1
9
]
d
escr
ib
e
th
e
s
p
ec
tr
al
p
r
o
p
er
ties
o
f
h
ea
r
t
r
ate
v
ar
iab
ilit
y
an
d
its
r
elatio
n
to
b
lo
o
d
p
r
ess
u
r
e
lev
els.
Faza
n
J
r
et
a
l
.
[
2
0
]
h
av
e
in
v
esti
g
ated
t
h
e
r
elat
io
n
b
etwe
en
h
ea
r
t
r
ate
v
ar
i
ab
ilit
y
an
d
h
u
m
an
b
ar
o
r
ef
le
x
es.
2
.
5
.
O
SA a
nd
brea
t
hin
g
ra
t
e
I
t
h
as
b
ee
n
m
ad
e
clea
r
th
r
o
u
g
h
s
ev
er
al
s
tu
d
ies
th
at
th
e
r
e
s
p
ir
ato
r
y
f
r
eq
u
e
n
cy
h
as
e
f
f
ec
t
s
o
n
b
lo
o
d
p
r
ess
u
r
e
lev
els
[
2
1
]
.
Sig
n
al
p
r
o
ce
s
s
in
g
tech
n
iq
u
es
to
ex
t
r
a
ct
th
e
lo
w
f
r
eq
u
e
n
cy
c
o
m
p
o
n
en
ts
o
f
PP
G
s
ig
n
al
h
elp
in
m
e
asu
r
in
g
th
e
r
esp
ir
at
io
n
r
ate.
Du
e
to
th
e
well
-
k
n
o
wn
f
ac
t
o
f
h
ig
h
n
o
n
-
lin
ea
r
ity
an
d
n
o
n
-
s
tatio
n
ar
y
p
r
o
p
er
ties
o
f
b
i
o
s
ig
n
als
,
em
p
ir
ical
m
o
d
e
d
ec
o
m
p
o
s
itio
n
(
E
MD
)
alg
o
r
ith
m
is
b
est
s
u
ited
an
d
is
ca
p
ab
le
o
f
ac
cu
r
ately
esti
m
atin
g
th
e
r
es
p
ir
ato
r
y
r
ate
f
r
o
m
PP
G
s
i
g
n
al.
T
h
is
p
r
o
v
id
es
an
alter
n
ativ
e
to
u
s
in
g
a
s
ep
ar
ate
s
en
s
o
r
m
o
d
u
le
f
o
r
m
o
n
ito
r
i
n
g
r
esp
ir
atio
n
r
ate.
3.
M
E
T
H
O
DO
L
O
G
Y
A
d
e
t
a
il
e
d
a
c
c
o
u
n
t
o
f
t
h
e
m
a
te
r
i
a
l
s
a
n
d
m
e
t
h
o
d
o
l
o
g
i
e
s
u
s
e
d
i
n
t
h
e
v
a
r
i
o
u
s
m
o
d
u
l
es
o
f
t
h
e
p
r
o
p
o
s
e
d
s
y
s
te
m
a
r
e
d
e
s
c
r
i
b
e
d
i
n
t
h
is
s
es
s
i
o
n
.
I
n
c
o
n
t
r
a
s
t
t
o
t
h
e
a
b
o
v
e
d
i
s
c
u
s
s
e
d
m
e
t
h
o
d
s
t
h
e
a
p
p
r
o
a
c
h
p
r
o
p
o
s
e
d
i
n
t
h
is
p
a
p
e
r
m
e
a
s
u
r
es
t
h
e
d
i
f
f
e
r
e
n
c
e
i
n
t
h
e
t
r
a
n
s
i
t
t
i
m
e
b
e
t
w
ee
n
t
h
e
s
u
c
c
ess
i
v
e
P
PG
w
a
v
e
f
o
r
m
a
n
d
a
l
s
o
m
e
as
u
r
e
s
t
h
e
r
e
p
e
r
c
u
s
s
i
o
n
s
o
f
p
h
y
s
i
o
l
o
g
i
c
al
a
l
t
e
r
a
ti
o
n
s
i
n
f
li
c
t
e
d
o
n
t
h
e
s
h
a
p
e
o
f
t
h
e
P
PG
w
a
v
e
f
o
r
m
a
n
d
o
n
t
h
e
h
e
a
r
t
r
a
t
e
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
tem
is
o
r
g
a
n
ized
in
to
t
h
r
ee
m
o
d
u
les
as
d
ep
icted
in
t
h
e
f
lo
w
d
iag
r
am
s
h
o
wn
in
Fig
u
r
e
1
.
T
h
e
PP
G
wav
ef
o
r
m
o
b
tain
ed
f
r
o
m
a
s
im
p
le
f
i
n
g
er
tip
PP
G
s
en
s
o
r
m
o
v
es
th
r
o
u
g
h
a
clea
n
s
ig
n
al
d
etec
to
r
m
o
d
u
le
th
at
elim
in
a
tes
co
r
r
u
p
ted
u
n
clea
n
s
ig
n
als
wh
ich
s
et
f
o
o
t
in
d
u
e
to
a
r
tifa
cts
an
d
f
in
g
e
r
m
o
v
em
en
ts
.
T
h
is
m
o
d
u
le
e
n
s
u
r
es
th
at
th
e
s
y
s
tem
d
o
es
n
o
t
r
e
q
u
ir
e
an
y
r
e
-
ca
lib
r
atio
n
with
ch
a
n
g
e
in
in
d
iv
id
u
als
o
r
ch
a
n
g
e
o
v
e
r
ti
m
e.
Un
co
r
r
u
p
te
d
an
d
c
o
n
s
ec
u
tiv
ely
r
ec
o
r
d
ed
4
0
0
0
s
am
p
les
ar
e
p
r
o
ce
s
s
ed
u
s
in
g
m
o
v
in
g
av
er
a
g
e
alg
o
r
ith
m
to
f
u
r
th
er
s
m
o
o
th
e
n
th
e
s
ig
n
al
.
T
h
e
s
ig
n
al
p
r
o
ce
s
s
in
g
an
d
f
ea
tu
r
e
e
x
tr
ac
tio
n
m
o
d
u
le
p
r
o
ce
s
s
es
th
e
s
ig
n
al
to
ex
tr
ac
t
s
ev
er
al
r
elate
d
f
ea
tu
r
es
f
r
o
m
th
e
clea
n
PP
G
wa
v
ef
o
r
m
.
Fin
ally
,
a
class
if
i
er
m
o
d
u
le
m
ak
es
a
d
ec
is
io
n
b
etwe
en
th
e
n
o
r
m
al
an
d
ab
n
o
r
m
al
v
alu
es
o
f
th
e
v
ar
iab
l
es
with
th
e
h
elp
o
f
th
e
ex
tr
ac
ted
f
ea
t
u
r
es f
ed
t
o
it.
Fig
u
r
e
1
.
B
lo
ck
d
iag
r
am
o
f
th
e
au
to
m
atic
OSA
s
cr
ee
n
in
g
s
y
s
tem
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
A
n
a
u
to
m
a
tic
s
creen
in
g
a
p
p
r
o
a
ch
fo
r
o
b
s
tr
u
ctive
s
leep
a
p
n
e
a
fr
o
m
…
(
S
mily
Je
ya
Jo
t
h
i.
E
)
1263
T
h
e
tr
ain
in
g
a
n
d
test
in
g
o
f
th
e
s
y
s
tem
was
ca
r
r
ied
o
u
t
with
2
8
5
s
u
b
jects
en
c
o
m
p
ass
in
g
i
n
d
iv
id
u
als
f
r
o
m
v
ar
ied
ag
e
g
r
o
u
p
s
,
h
ea
lth
y
,
in
d
iv
id
u
als
with
h
y
p
er
ten
s
i
o
n
a
n
d
ca
r
d
io
v
ascu
lar
d
is
ea
s
e,
an
d
OSA
p
atien
ts
.
T
h
e
in
p
u
t
PP
G
s
ig
n
al
r
ec
o
r
d
e
d
with
t
h
e
h
elp
o
f
a
f
in
g
er
tip
p
u
ls
e
o
x
im
eter
d
ev
ice
(
Ger
m
an
s
en
s
o
r
)
an
d
th
e
r
ef
er
en
ce
ap
n
ea
s
ig
n
als
o
b
tain
ed
f
r
o
m
th
e
ap
n
ea
d
atab
ase
f
r
o
m
Ph
y
s
io
n
et
wer
e
u
s
ed
f
o
r
t
r
ain
in
g
th
e
s
y
s
tem
.
T
h
e
d
atab
ase
in
clu
d
es
1
6
9
m
a
le
an
d
1
1
6
f
em
ale
s
u
b
jects,
ag
ed
b
etwe
en
2
0
to
8
5
y
ea
r
s
.
T
h
e
d
atab
ase
wid
ely
in
co
r
p
o
r
ates
y
o
u
n
g
h
ea
lth
y
i
n
d
iv
id
u
als,
d
i
ab
etic
p
atien
ts
,
b
lo
o
d
p
r
ess
u
r
e
p
atien
ts
an
d
s
leep
ap
n
ea
p
atien
ts
with
a
f
o
cu
s
to
c
o
r
r
elate
th
e
s
y
m
p
to
m
s
co
m
m
o
n
t
o
s
leep
a
p
n
ea
,
b
l
o
o
d
s
u
g
ar
an
d
b
lo
o
d
p
r
ess
u
r
e.
R
ea
l
tim
e
d
ig
italized
PP
G
s
ig
n
als
r
ec
o
r
d
ed
u
s
in
g
a
u
n
iv
er
s
al
s
er
ial
b
u
s
(
USB
)
f
in
g
er
-
tip
p
u
ls
e
o
x
im
eter
s
en
s
o
r
,
ar
e
obt
ain
ed
as
a
s
eq
u
en
ce
,
h
av
in
g
a
s
am
p
lin
g
r
ate
o
f
1
0
0
0
s
am
p
les
p
er
s
ec
o
n
d
.
T
h
e
PP
G
s
ig
n
al
is
g
iv
en
as
in
p
u
t
to
th
e
clea
n
s
ig
n
al
d
etec
tio
n
m
o
d
u
le
w
h
ich
id
e
n
tifie
s
a
clea
n
an
d
u
n
c
o
r
r
u
p
ted
s
ig
n
al
X
(T
)
in
co
r
p
o
r
ated
with
s
ev
er
al
s
eg
m
en
ts
/f
r
am
es
h
av
i
n
g
a
f
r
a
m
e
len
g
th
o
f
4
0
0
0
co
n
s
ec
u
tiv
ely
r
ec
o
r
d
ed
s
am
p
les.
T
im
e
d
u
r
atio
n
o
f
two
m
in
u
te
was
s
elec
ted
s
o
as
to
ac
co
m
m
o
d
ate
s
u
f
f
icien
t
s
a
m
p
les
f
o
r
esti
m
atio
n
o
f
h
ea
r
t
r
ate
an
d
r
esp
ir
ato
r
y
r
ate
f
r
o
m
th
e
PP
G
s
ig
n
al.
T
h
e
o
u
tp
u
t
o
f
th
e
clea
n
s
ig
n
al
d
etec
to
r
is
f
ed
to
th
e
s
ig
n
al
p
r
o
ce
s
s
in
g
m
o
d
u
le,
wh
o
s
e
o
u
tp
u
t
v
ec
to
r
X
F
co
n
tai
n
in
g
a
s
et
o
f
f
ea
tu
r
es a
r
e
f
ed
t
o
th
e
m
ac
h
i
n
e
lear
n
in
g
m
o
d
u
l
e.
3
.
1
.
Clea
n si
g
na
l det
ec
t
io
n
m
o
du
le
T
h
e
u
ltima
te
aim
o
f
th
e
clea
n
s
ig
n
al
d
etec
to
r
is
to
elim
in
ate
th
e
co
r
r
u
p
ted
s
ig
n
al
len
g
th
as
s
h
o
wn
in
Fig
u
r
e
2
an
d
to
s
elec
t
th
e
f
r
am
e
th
at
h
as
clea
n
s
ig
n
al
a
s
d
e
p
icted
in
Fig
u
r
e
3
.
T
h
e
PP
G
s
ig
n
al
r
ec
o
r
d
e
d
at
th
e
f
in
g
er
tip
is
u
s
u
ally
v
u
ln
e
r
ab
le
to
s
p
u
r
io
u
s
p
ea
k
s
,
n
o
is
e
g
en
e
r
ated
d
u
e
to
m
o
v
em
e
n
ts
,
s
ig
n
al
d
is
to
r
tio
n
d
u
e
to
in
itial
tr
an
s
ien
t
ir
r
eg
u
lar
ities
,
an
d
s
ig
n
al
s
atu
r
atio
n
.
T
h
e
o
u
tp
u
t
o
f
th
e
clea
n
s
ig
n
al
d
etec
to
r
is
a
v
ec
to
r
co
n
s
is
tin
g
o
f
‘
n
’
n
u
m
b
e
r
o
f
f
r
am
es
with
ea
ch
f
r
am
e
h
o
l
d
in
g
4
0
0
0
clea
n
d
ata
s
am
p
le
s
o
b
ta
in
ed
f
r
o
m
th
e
PP
G
s
ig
n
al.
T
im
e
r
eq
u
ir
ed
t
o
ac
q
u
ir
e
o
n
e
f
r
am
e
o
f
4
0
0
0
co
n
s
ec
u
tiv
e
s
am
p
les
is
4
s
,
wh
ich
is
ca
p
ab
le
o
f
tr
ap
p
i
n
g
s
ev
er
al
h
ea
r
t
b
ea
ts
.
R
ed
u
cin
g
th
e
f
r
am
e
len
g
th
b
elo
w
4
s
m
ig
h
t
g
r
asp
v
e
r
y
f
ew
h
ea
r
t
b
ea
ts
r
esu
ltin
g
in
an
u
n
r
eliab
le
d
ata.
I
n
o
r
d
er
to
d
i
s
cr
im
in
ate
b
etwe
en
clea
n
a
n
d
co
r
r
u
p
ted
s
ig
n
al,
a
s
et
o
f
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es
ar
e
ex
tr
ac
ted
f
r
o
m
t
h
e
tim
e
-
d
o
m
ain
PP
G
wav
ef
o
r
m
,
its
f
ir
s
t
an
d
s
ec
o
n
d
d
er
iv
ativ
e
.
Fig
u
r
e
2
.
C
o
r
r
u
p
ted
PP
G
s
ig
n
al
with
in
itial ir
r
eg
u
lar
ities
Fig
u
r
e
3
.
C
lean
PP
G
s
ig
n
al
af
ter
elim
in
atin
g
th
e
c
o
r
r
u
p
ted
s
i
g
n
al
len
g
th
3.
1
.
1
.
S
et
o
f
m
o
rpho
lo
g
ica
l f
ea
t
ures
T
h
e
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es
o
f
in
ter
est
ex
tr
ac
ted
f
o
r
f
ea
tu
r
e
v
ec
to
r
wer
e
p
u
ls
e
h
eig
h
t/s
y
s
to
lic
p
ea
k
,
d
iast
o
lic
p
ea
k
,
au
g
m
e
n
tatio
n
i
n
d
ex
,
p
u
ls
e
in
ter
v
al,
p
u
ls
e
wid
th
,
p
u
ls
e
ar
ea
,
p
ea
k
to
p
ea
k
i
n
ter
v
al,
n
o
tc
h
tim
e,
s
y
s
to
lic
p
ea
k
tim
e
an
d
d
iast
o
li
c
p
ea
k
tim
e
as
s
h
o
wn
i
n
Fig
u
r
e
4
.
Pu
ls
e
wid
th
i.e
.
,
t
h
e
tim
e
b
etwe
en
o
n
s
et
an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
4
,
Au
g
u
s
t 2
0
2
1
:
1
2
6
0
-
1
2
7
2
1264
en
d
o
f
p
u
ls
e
wav
e
clea
r
ly
co
r
r
elate
s
with
th
e
v
ascu
lar
r
esis
tan
ce
wh
ich
is
d
u
e
to
p
r
o
lo
n
g
e
d
in
cr
ea
s
e
in
b
lo
o
d
p
r
ess
u
r
e.
T
ab
le
1
d
ef
in
es
th
e
s
et
o
f
m
o
r
p
h
o
lo
g
ical
p
ar
am
et
er
s
ex
tr
ac
ted
.
Pu
ls
e
ar
ea
is
m
ea
s
u
r
ed
as
th
e
to
tal
ar
ea
u
n
d
er
th
e
PP
G
cu
r
v
e,
wh
ich
is
also
u
s
ed
a
s
an
in
d
icato
r
o
f
to
tal
p
er
ip
h
er
al
r
esis
tan
ce
.
T
h
e
au
g
m
en
tatio
n
in
d
ex
q
u
an
tifie
s
th
e
c
o
n
tr
ib
u
tio
n
m
ad
e
b
y
t
h
e
wav
e
r
ef
le
ctio
n
o
n
th
e
s
y
s
to
lic
ar
ter
ial
p
r
ess
u
r
e
[
2
2
]
.
I
n
ad
d
itio
n
to
th
e
a
b
o
v
e
-
m
en
tio
n
ed
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es
wh
ich
ev
alu
ate
t
h
e
ca
r
d
io
v
as
cu
lar
f
u
n
ctio
n
s
,
th
e
m
ax
im
u
m
am
p
litu
d
e
o
f
th
e
f
i
r
s
t
d
er
iv
ativ
e
an
d
s
ec
o
n
d
d
er
i
v
ativ
e
o
f
th
e
PP
G
wav
ef
o
r
m
is
also
in
co
r
p
o
r
ated
in
o
u
r
s
tu
d
y
.
T
h
e
v
elo
city
o
f
b
lo
o
d
f
lo
w
in
th
e
f
in
g
e
r
is
r
ep
r
esen
ted
b
y
th
e
f
ir
s
t
d
er
i
v
a
tiv
e
an
d
its
s
ec
o
n
d
d
er
iv
ativ
e
s
p
ea
k
s
m
u
ch
ab
o
u
t
th
e
ar
ter
ial
s
tiff
n
ess
wh
ich
is
an
e
v
id
en
t
s
y
m
p
to
m
o
f
OSA.
T
h
e
p
u
ls
e
tr
a
n
s
it
tim
e
m
ea
s
u
r
em
en
t (
PTT
)
in
O
SA p
atien
ts
d
ep
icts
th
at
th
er
e
i
s
an
in
cr
ea
s
e
in
PTT
d
u
e
to
ar
t
er
ial
s
tiff
n
e
ss.
T
h
e
en
er
g
y
p
r
o
f
ile,
T
K
x
(
t
)
an
d
s
p
ec
tr
al
p
o
wer
d
is
tr
ib
u
tio
n
o
f
PP
G
s
ig
n
al,
S
pen
o
f
ea
ch
f
r
a
m
e
wer
e
esti
m
ated
u
s
in
g
T
ea
g
er
-
Kaiser
en
er
g
y
o
p
er
ato
r
a
n
d
s
p
ec
tr
al
en
tr
o
p
y
o
p
er
ato
r
r
esp
ec
tiv
el
y
[
2
3
]
,
[
2
4
]
.
Fig
u
r
e
4
.
C
h
ar
ac
ter
is
tic
f
ea
tu
r
es d
er
iv
ed
f
r
o
m
PP
G
wav
ef
o
r
m
T
ab
le
1
.
E
x
tr
ac
ted
f
ea
tu
r
es f
r
o
m
th
e
PP
G
wav
ef
o
r
m
F
e
a
t
u
r
e
D
e
f
i
n
i
t
i
o
n
x
(
s
y
st
o
l
i
c
p
e
a
k
)
M
a
x
i
m
u
m
a
m
p
l
i
t
u
d
e
o
f
t
h
e
si
g
n
a
l
y
(
d
i
a
s
t
o
l
i
c
p
e
a
k
)
A
mp
l
i
t
u
d
e
o
f
t
h
e
d
i
a
s
t
o
l
i
c
p
e
a
k
y
/
x
(
a
u
g
me
n
t
a
t
i
o
n
i
n
d
e
x
)
A
mp
l
i
t
u
d
e
o
f
d
i
a
st
o
l
i
c
p
e
a
k
/
a
mp
l
i
t
u
d
e
o
f
t
h
e
s
i
g
n
a
l
t
pi
(
p
u
l
se
i
n
t
e
r
v
a
l
)
Ti
me
t
a
k
e
n
f
o
r
o
n
e
c
o
m
p
l
e
t
e
P
P
G
p
e
r
i
o
d
t
pp
(
p
e
a
k
t
o
p
e
a
k
i
n
t
e
r
v
a
l
)
Ti
me
e
l
a
p
se
d
b
e
t
w
e
e
n
t
w
o
su
c
c
e
ssi
v
e
p
e
a
k
s
t
1
(
sy
st
o
l
i
c
p
e
a
k
t
i
me)
Ti
me
a
t
w
h
i
c
h
s
y
st
o
l
i
c
p
e
a
k
o
c
c
u
r
t
3
(
d
i
a
s
t
o
l
i
c
p
e
a
k
t
i
me)
Ti
me
a
t
w
h
i
c
h
d
i
a
st
o
l
i
c
p
e
a
k
o
c
c
u
r
N
t
(
N
o
t
c
h
t
i
me)
Ti
me
a
t
w
h
i
c
h
n
o
t
c
h
o
c
c
u
r
S
D
o
f
a
m
p
l
i
t
u
d
e
S
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
o
f
t
h
e
a
mp
l
i
t
u
d
e
s
1
st
d
e
r
i
v
a
t
i
v
e
_
m
a
x
a
m
p
M
a
x
i
m
u
m
a
m
p
l
i
t
u
d
e
o
f
t
h
e
f
i
r
s
t
d
e
r
i
v
a
t
i
v
e
2
nd
d
e
r
i
v
a
t
i
v
e
_
ma
x
a
m
p
M
a
x
i
m
u
m
a
m
p
l
i
t
u
d
e
o
f
t
h
e
se
c
o
n
d
d
e
r
i
v
a
t
i
v
e
3.
1
.
2
.
Cla
s
s
if
ier
a
lg
o
rit
hm
T
h
e
f
ea
tu
r
e
v
ec
to
r
X
n
(f)
is
th
u
s
co
n
s
tr
u
cted
b
y
cl
u
s
ter
in
g
all
th
e
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es
a
n
d
en
e
r
g
y
lev
els o
f
PP
G
s
ig
n
al
co
n
tain
ed
in
ea
ch
f
r
am
e
o
f
4
0
0
0
s
am
p
le
s
.
X
n
(f)
=
[
x
,
y
,
y
/x
,
t
pi
,
t
pi
, t
1
, t
3
,
Nt,
SD,
1
st
d
er
iv
e
m
a
x
_
am
p
,
2
nd
d
er
iv
e
m
ax
_
am
p
,
T
K
x
(
t)
, S
pen
]
T
T
h
e
class
if
ier
tak
es
X
n
(f)
a
s
it
s
in
p
u
t
an
d
d
is
tin
g
u
is
h
es
b
etwe
en
g
o
o
d
s
ig
n
al
an
d
b
ad
s
ig
n
al
b
ased
o
n
n
o
is
e
an
d
s
ig
n
al
lo
s
s
.
A
th
r
esh
o
l
d
-
b
as
ed
class
if
ier
m
ak
es
a
class
if
icatio
n
d
ec
is
io
n
b
ased
o
n
a
v
alu
e
o
f
lin
ea
r
co
m
b
in
atio
n
o
f
all
th
e
ab
o
v
e
s
tated
f
ea
tu
r
es.
T
h
e
th
r
esh
o
ld
s
in
itially
wer
e
r
eso
lv
ed
b
y
in
s
p
ec
tio
n
f
r
o
m
a
s
et
o
f
ex
am
p
les an
d
tr
ain
e
d
u
s
in
g
a
s
et
o
f
h
an
d
-
lab
eled
d
atab
ase.
T
h
e
alg
o
r
ith
m
is
d
esig
n
ed
in
s
u
ch
a
way
to
elim
in
ate
th
e
in
c
o
m
in
g
s
ig
n
a
ls
f
o
r
th
e
f
ir
s
t
1
5
s
ec
o
n
d
s
s
o
as
to
elim
in
ate
th
e
s
p
u
r
io
u
s
click
s
an
d
m
o
tio
n
ar
tifa
cts
in
th
e
s
ig
n
al
d
u
e
to
f
i
n
g
er
m
o
v
em
e
n
ts
.
T
h
e
s
u
b
jects
wer
e
ad
v
is
ed
to
b
e
i
n
a
r
elax
ed
s
tate
at
th
e
tim
e
o
f
r
ec
o
r
d
in
g
th
e
r
ea
l
tim
e
PP
G
s
ig
n
als
u
s
in
g
USB
b
ased
f
in
g
er
-
tip
p
u
ls
e
o
x
im
ete
r
,
with
a
s
am
p
lin
g
r
ate
o
f
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
A
n
a
u
to
m
a
tic
s
creen
in
g
a
p
p
r
o
a
ch
fo
r
o
b
s
tr
u
ctive
s
leep
a
p
n
e
a
fr
o
m
…
(
S
mily
Je
ya
Jo
t
h
i.
E
)
1265
1
0
0
0
s
am
p
les/
s
ec
.
i
.
e.
,
ea
ch
d
ata
ac
q
u
is
itio
n
p
r
o
v
id
es
1
0
0
0
s
am
p
les.
Data
s
am
p
le
s
o
b
tain
ed
f
r
o
m
4
co
n
s
ec
u
tiv
e
ac
q
u
is
itio
n
s
r
ec
o
r
d
ed
f
o
r
4
s
,
ar
e
s
to
r
ed
in
a
n
ar
r
ay
wh
ich
f
o
r
m
s
o
n
e
f
r
am
e/seg
m
en
t
h
av
in
g
4
0
0
0
s
am
p
les
alto
g
eth
er
.
T
h
e
class
if
ier
wo
r
k
s
o
n
th
e
b
asis
o
f
ce
r
t
ain
o
r
d
ai
n
ed
r
u
les
an
d
f
u
n
ctio
n
s
wr
itten
b
y
h
a
n
d
af
ter
an
aly
zi
n
g
v
ar
io
u
s
s
ig
n
al
s
an
d
m
ak
es
a
d
ec
is
io
n
b
etw
ee
n
g
o
o
d
an
d
b
ad
f
r
am
e/seg
m
en
t.
On
ce
a
clea
n
s
ig
n
al
o
f
a
m
i
n
u
te
o
r
m
o
r
e
co
n
s
ec
u
tiv
e
s
am
p
les
wer
e
o
b
tai
n
ed
,
th
e
ac
q
u
is
itio
n
s
to
p
s
an
d
th
e
s
ig
n
al
p
r
o
ce
ed
s
to
th
e
s
ig
n
al
p
r
o
ce
s
s
in
g
m
o
d
u
le
f
o
r
f
ea
tu
r
e
ex
t
r
ac
tio
n
.
T
h
e
clin
ical
p
ar
am
eter
s
o
f
th
e
s
tu
d
y
g
r
o
u
p
ar
e
s
h
o
wn
in
T
ab
le
2
.
T
ab
le
2
.
C
lin
ical
p
ar
am
eter
s
o
f
th
e
s
tu
d
y
g
r
o
u
p
S
u
b
j
e
c
t
s
A
g
e
G
e
n
d
e
r
M/F
H
e
i
g
h
t
cm
W
e
i
g
h
t
Kg
R
e
s
p
.
R
a
t
e
/
mi
n
H
e
a
r
t
R
a
t
e
/
mi
n
S
p
O
2
B
G
L
mg
/
d
l
BP
S
y
s/
d
i
a
B
M
I
K
g
/
m
2
Lu
n
g
d
i
s
o
r
d
e
r
S
1
74
M
1
5
9
92
21
62
97
2
2
2
1
5
4
/
5
0
3
6
.
4
y
e
s
S
2
44
F
1
5
0
85
16
84
98
1
0
7
1
2
2
/
8
2
3
7
.
8
no
S
3
40
M
1
3
8
77
17
1
1
1
99
3
5
8
1
3
8
/
7
3
4
0
.
4
no
S
4
48
M
1
6
1
69
16
1
0
1
99
4
7
8
1
2
8
/
8
5
2
6
.
6
no
S
5
52
F
1
4
6
71
17
82
99
1
2
0
1
0
6
/
4
9
3
3
.
3
no
S
6
63
F
1
5
3
89
12
90
99
1
2
8
1
4
9
/
8
7
38
no
S
7
42
F
1
5
9
73
14
88
99
1
4
3
1
1
6
/
8
5
2
8
.
9
no
S
8
70
M
1
5
6
65
18
71
98
96
1
2
1
/
8
2
2
6
.
4
no
S
9
52
M
1
4
5
76
19
75
97
1
2
5
1
1
8
/
7
6
3
4
.
2
y
e
s
S
10
49
M
1
5
2
89
16
86
97
1
1
5
1
0
0
/
6
5
3
8
.
5
y
e
s
3.
2
.
Sig
na
l pro
ce
s
s
ing
m
o
du
le
T
h
is
s
ec
tio
n
s
u
m
m
ar
izes
th
e
d
if
f
er
en
t
s
ig
n
al
p
r
o
ce
s
s
in
g
alg
o
r
ith
m
s
u
s
ed
f
o
r
co
m
p
u
ti
n
g
f
ea
tu
r
es
r
elate
d
to
OSA
f
r
o
m
PP
G
s
ig
n
al.
Sig
n
al
d
u
r
atio
n
o
f
m
o
r
e
t
h
an
a
m
i
n
u
te
co
m
p
r
is
in
g
o
f
at
least
1
5
f
r
am
es
o
f
co
n
s
ec
u
tiv
ely
r
ec
o
r
d
e
d
g
o
o
d
s
ig
n
al
is
g
iv
en
as
in
p
u
t
to
th
e
s
ig
n
al
p
r
o
ce
s
s
in
g
m
o
d
u
le
.
T
h
e
o
u
t
p
u
t
o
f
th
is
m
o
d
u
le
is
a
v
ec
to
r
co
m
p
r
is
in
g
o
f
an
ag
g
r
e
g
ate
o
f
all
th
e
f
ea
tu
r
es
an
d
f
ed
as
in
p
u
t
to
th
e
m
ac
h
in
e
lear
n
in
g
m
o
d
u
le.
T
h
e
len
g
th
o
f
th
e
s
ig
n
al
an
d
t
h
e
n
u
m
b
er
o
f
s
am
p
le
s
s
elec
ted
ar
e
s
u
f
f
icien
t
en
o
u
g
h
to
co
m
p
u
te
h
ea
r
t
r
ate,
r
esp
ir
atio
n
r
ate
an
d
t
h
eir
v
ar
iab
ilit
y
r
esp
ec
tiv
ely
.
3.
2
.
1
.
E
s
t
im
a
t
io
n o
f
re
s
pira
t
o
ry
ra
te
E
s
tim
atio
n
o
f
r
esp
ir
ato
r
y
r
ate
f
r
o
m
PP
G
s
ig
n
al
is
an
alter
n
ativ
e
m
eth
o
d
to
u
s
in
g
a
s
ep
ar
ate
s
en
s
o
r
-
am
p
lifie
r
u
n
it.
Sev
er
al
s
ig
n
al
p
r
o
ce
s
s
in
g
tech
n
i
q
u
es
wh
ich
wo
r
k
b
y
e
x
tr
ac
tin
g
th
e
r
esp
ir
atio
n
tr
en
d
em
b
ed
d
e
d
in
th
e
PP
G
wav
ef
o
r
m
h
av
e
b
ee
n
p
u
t f
o
r
th
[
2
5
]
,
[
2
6
].
I
t is a
wel
l
-
kn
o
wn
f
ac
t th
at
th
e
b
io
s
ig
n
als ar
e
h
ig
h
ly
n
o
n
-
li
n
ea
r
an
d
n
o
n
-
s
tatio
n
ar
y
,
n
ev
er
t
h
eless
,
test
r
esu
lts
r
ev
ea
l
th
at
th
e
em
p
ir
ical
m
o
d
e
d
ec
o
m
p
o
s
itio
n
(
E
MD
)
alg
o
r
ith
m
is
b
est
s
u
ited
f
o
r
n
o
n
-
lin
ea
r
s
ig
n
als
an
d
is
ca
p
ab
le
o
f
ac
cu
r
ately
esti
m
atin
g
th
e
r
es
p
ir
ato
r
y
ra
te
f
r
o
m
PP
G
s
ig
n
al.
As
o
u
r
aim
is
to
d
ev
elo
p
a
s
y
s
tem
w
h
ich
au
to
m
atica
lly
s
cr
ee
n
s
f
o
r
OSA
p
atien
ts
,
th
e
m
ea
s
u
r
em
en
ts
ar
e
ass
o
ciate
d
t
o
th
e
in
d
iv
id
u
al’
s
h
em
o
d
y
n
am
ics
an
d
ca
r
d
io
v
ascu
lar
ac
tiv
ity
.
Hen
ce
,
ef
f
o
r
ts
to
ac
cu
r
ately
m
ea
s
u
r
e
h
ea
r
t
r
ate
an
d
r
esp
ir
ato
r
y
r
ate
f
r
o
m
PP
G
s
ig
n
al
h
av
e
b
ee
n
ca
r
r
ied
o
u
t
as
th
ey
ar
e
th
e
p
ar
am
et
er
s
clo
s
ely
ass
o
ciate
d
with
OSA
an
d
h
as b
ee
n
p
u
b
lis
h
ed
in
o
u
r
p
r
ev
io
u
s
wo
r
k
[
27
].
3.
2
.
2
.
H
ea
rt
ra
t
e
v
a
ria
bil
it
y
a
nd
o
x
y
g
en
s
a
t
ura
t
io
n
Sev
er
al
s
tu
d
ies
p
r
o
v
ed
th
at
th
er
e
is
a
s
tr
o
n
g
ev
id
en
t
ial
r
elatio
n
b
etwe
en
OSA
an
d
h
ea
r
t
r
ate
v
ar
iab
ilit
y
wh
ich
is
r
ec
o
r
d
ed
as
th
e
v
ar
iatio
n
in
th
e
tim
e
in
ter
v
al
b
etwe
en
h
ea
r
tb
ea
ts
.
I
n
th
is
wo
r
k
,
th
e
v
ar
iab
ilit
y
in
p
ea
k
-
to
-
p
ea
k
in
t
er
v
al
o
f
PP
G
s
ig
n
al
h
as
b
ee
n
co
m
p
u
ted
u
s
in
g
s
lo
p
e
d
etec
t
io
n
alg
o
r
ith
m
.
T
h
e
p
ea
k
s
wer
e
d
etec
ted
f
r
o
m
th
e
ze
r
o
cr
o
s
s
in
g
s
o
f
th
e
s
ig
n
al,
r
ath
er
th
an
m
ea
s
u
r
in
g
th
e
s
ig
n
al
p
ea
k
s
.
I
n
itially
a
b
an
d
p
ass
f
ilter
was
u
s
ed
to
a
tten
u
ate
th
e
d
ich
r
o
tic
n
o
tch
e
f
f
ec
t
an
d
to
r
em
o
v
e
th
e
m
ea
n
.
T
h
is
m
eth
o
d
was
f
o
u
n
d
to
b
e
ac
cu
r
ate
in
m
ea
s
u
r
in
g
th
e
tim
e
in
ter
v
al
b
etwe
en
th
e
alter
n
ate
ze
r
o
cr
o
s
s
in
g
s
an
d
th
e
d
if
f
er
e
n
ce
i
n
th
e
tim
e
in
ter
v
al
g
iv
es
t
h
e
m
ea
s
u
r
e
o
f
h
ea
r
t
r
ate
v
ar
iab
ilit
y
.
T
h
e
m
ea
n
v
al
u
e
o
f
th
e
p
er
ip
h
er
al
o
x
y
g
e
n
s
atu
r
atio
n
r
an
g
e
was
d
ir
ec
tly
g
iv
en
b
y
th
e
USB
b
ased
f
in
g
er
-
tip
p
u
ls
e
o
x
im
ete
r
o
v
er
t
h
e
en
tire
r
ec
o
r
d
in
g
d
u
r
atio
n
.
T
h
e
f
ea
tu
r
e
v
ec
to
r
w
as
f
in
ally
co
n
s
tr
u
cted
b
y
cl
u
s
ter
in
g
all
t
h
e
ab
o
v
e
m
en
tio
n
e
d
f
ea
tu
r
es,
in
clu
d
in
g
p
h
y
s
io
lo
g
ical
p
a
r
am
eter
s
,
m
o
r
p
h
o
lo
g
ical
p
ar
am
eter
o
f
th
e
PP
G
wav
ef
o
r
m
,
T
ea
g
er
-
Kaiser
en
er
g
y
lev
el,
an
d
en
tr
o
p
y
o
f
th
e
s
ig
n
al,
h
ea
r
t
r
a
te,
r
esp
ir
atio
n
r
ate,
h
ea
r
t
r
ate
v
ar
iab
ilit
y
an
d
o
x
y
g
en
s
atu
r
at
io
n
.
T
h
e
s
tatis
tical
f
ea
tu
r
es
s
u
ch
as
m
ea
n
,
M
an
d
s
tan
d
ar
d
d
ev
iatio
n
,
SD
wer
e
a
ls
o
ca
lcu
lated
in
o
r
d
er
to
ca
p
t
u
r
e
th
e
r
elatio
n
s
h
i
p
b
etwe
en
h
ea
r
t
r
ate
v
ar
iab
ilit
y
an
d
OSA.
T
h
e
o
u
tp
u
t
o
f
th
e
s
ig
n
al
p
r
o
ce
s
s
in
g
m
o
d
u
le
is
f
ed
as
in
p
u
t
to
th
e
class
if
ier
m
o
d
u
le,
wh
ich
is
tr
a
in
ed
with
all
th
e
f
ea
tu
r
es o
f
th
e
o
u
tp
u
t
v
ec
to
r
as g
i
v
en
b
el
o
w:
X
F
=
[
A,
W
,
B
MI
,
B
P,
B
GL
,
PTT
,
x
,
y
,
y
/x
,
t
pi
,
t
pi,
t1
,
t3
,
T
K
x
(
t)
,
Sp
en
,
HR
,
R
R
,
HR
V,
Sp
O
2,
1
st
d
e
r
iv
e
m
ax
_
am
p
,
2
nd
d
e
r
iv
e
m
ax
_
am
p
,
M,
SD]
T
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
4
,
Au
g
u
s
t 2
0
2
1
:
1
2
6
0
-
1
2
7
2
1266
3.
3
.
Cla
s
s
if
ier
m
o
du
le
T
h
e
m
ac
h
in
e
lear
n
in
g
m
o
d
u
le
is
s
u
p
p
o
s
ed
to
in
f
er
a
p
h
y
s
io
lo
g
ical
f
u
n
ctio
n
th
at
r
elate
s
th
e
f
ea
tu
r
es
ex
tr
ac
ted
f
r
o
m
th
e
PP
G
s
ig
n
al
an
d
th
e
d
esire
d
tar
g
et
[
2
8
].
L
iter
atu
r
e
clea
r
ly
p
o
r
tr
a
y
s
th
a
t
m
o
s
t
o
f
th
e
wo
r
k
in
v
o
lv
ed
in
th
e
d
etec
tio
n
o
f
OSA
h
as
u
s
ed
n
e
u
r
al
n
etwo
r
k
-
b
ased
class
if
ier
[
2
9
]
.
I
t
is
a
ls
o
u
n
d
e
r
s
to
o
d
th
at
d
ee
p
lear
n
in
g
n
eu
r
al
n
etwo
r
k
s
h
av
e
alwa
y
s
o
u
tp
er
f
o
r
m
e
d
th
e
s
h
allo
w
n
eu
r
al
n
etwo
r
k
s
an
d
co
n
v
o
lu
tio
n
a
l
n
eu
r
al
n
etwo
r
k
is
th
e
wid
ely
u
s
ed
class
if
ier
in
th
e
r
ec
en
t
y
ea
r
s
[
3
0
]
,
[
3
1
]
.
I
n
o
n
e
o
f
th
e
ap
p
r
o
ac
h
es
to
d
etec
t
r
esp
ir
ato
r
y
ar
r
ests
in
OSA
p
atien
ts
u
s
in
g
PP
G
s
ig
n
al,
class
if
ier
s
lik
e
k
-
n
ea
r
est
n
eig
h
b
o
u
r
s
class
if
icatio
n
,
r
ad
ial
b
asis
f
u
n
ctio
n
n
eu
r
al
n
e
two
r
k
,
p
r
o
b
ab
ilis
tic
n
eu
r
al
n
etwo
r
k
,
m
u
ltil
ay
er
f
ee
d
f
o
r
war
d
n
eu
r
al
n
etwo
r
k
an
d
en
s
em
b
le
class
if
icatio
n
h
a
v
e
b
ee
n
c
o
m
p
ar
e
d
b
y
th
e
a
u
th
o
r
s
an
d
th
eir
r
esu
lts
d
e
p
ict
a
test
i
n
g
ac
c
u
r
ac
y
r
ate
o
f
9
7
.
0
7
%
with
m
u
ltil
ay
er
f
ee
d
f
o
r
war
d
n
e
u
r
al
n
etwo
r
k
[
4
]
.
I
n
o
u
r
wo
r
k
th
r
ee
d
if
f
er
en
t
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
wer
e
test
ed
o
n
t
h
e
b
asis
o
f
f
lex
ib
ilit
y
,
ac
cu
r
ac
y
;
s
m
ar
t
en
o
u
g
h
to
d
e
al
with
n
o
is
y
s
ig
n
als,
non
-
r
e
q
u
ir
em
e
n
t
o
f
r
ep
ea
te
d
ca
lib
r
atio
n
an
d
s
tab
ilit
y
in
ter
m
s
o
f
its
p
er
f
o
r
m
an
ce
.
Mo
r
e
o
v
er
,
th
e
d
ata
ty
p
e
in
v
o
lv
ed
in
o
u
r
w
o
r
k
is
in
th
e
tab
u
lar
f
o
r
m
a
n
d
it
h
as
f
ac
ilit
ated
ea
s
y
im
p
lem
en
tatio
n
o
f
th
e
wo
r
k
u
s
in
g
th
ese
alg
o
r
ith
m
s
.
T
h
e
s
tr
u
ct
u
r
e
o
f
e
ac
h
class
if
ier
an
d
m
eth
o
d
o
f
tr
ain
in
g
in
v
o
lv
ed
a
r
e
d
escr
ib
ed
in
th
is
s
ec
tio
n
.
3.
3
.
1
.
Univ
a
ria
t
e
re
g
re
s
s
io
n (
UR)
Un
iv
ar
aite
r
eg
r
ess
io
n
also
k
n
o
wn
as
lin
ea
r
r
e
g
r
ess
io
n
is
a
s
u
p
er
v
is
ed
m
ac
h
i
n
e
lear
n
in
g
alg
o
r
ith
m
u
s
ed
f
o
r
d
ata
an
aly
s
is
.
T
h
e
r
elatio
n
s
h
ip
b
etwe
en
a
s
ca
lar
d
ep
en
d
en
t
v
a
r
iab
le
an
d
o
n
e
o
r
m
o
r
e
th
an
o
n
e
ex
p
lan
ato
r
y
v
ar
iab
le
ca
n
b
e
m
o
d
eled
b
y
s
im
p
le
lin
ea
r
r
e
g
r
ess
io
n
an
d
m
u
ltip
le
lin
ea
r
r
e
g
r
ess
io
n
ap
p
r
o
ac
h
es
r
esp
ec
tiv
ely
.
L
in
ea
r
p
r
ed
icto
r
f
u
n
ctio
n
s
,
also
ca
lled
as
lin
ea
r
m
o
d
els
ar
e
u
s
ed
in
m
o
d
eli
n
g
th
e
d
ata
an
d
to
esti
m
ate
th
e
u
n
k
n
o
wn
m
o
d
el
p
ar
am
eter
s
f
r
o
m
a
s
et
o
f
d
ata
p
o
in
ts
.
I
t
is
a
m
eth
o
d
u
s
ed
to
f
it
th
e
b
es
t
s
tr
aig
h
t
lin
e
b
etwe
en
a
s
et
o
f
d
ata
p
o
in
ts
,
an
d
th
e
o
b
tain
ed
s
tr
aig
h
t
lin
e
is
u
s
ed
as
a
m
o
d
el
to
p
r
ed
ict
th
e
v
alu
e
o
f
v
ar
iab
le
f
r
o
m
an
in
p
u
t
v
ar
ia
b
le
.
T
h
e
s
tr
aig
h
t
lin
e
r
ep
r
esen
ts
th
e
b
est
esti
m
ate
o
f
th
e
y
v
alu
e
f
o
r
e
v
er
y
in
p
u
t o
f
.
Gen
er
al
f
o
r
m
o
f
lin
e
ar
r
eg
r
ess
io
n
class
if
ier
is
g
iv
e
n
in
(
1
)
.
=
+
(1
)
W
h
er
e
is
th
e
in
p
u
t
v
alu
e,
is
th
e
s
lo
p
e
o
f
th
e
b
est
f
it
lin
e
an
d
is
th
e
p
o
in
t
wh
er
e
=
0
an
d
in
ter
s
ec
ts
th
e
-
ax
is
.
T
h
e
s
lo
p
e
o
f
th
e
lin
e
ca
n
b
e
ca
lcu
lated
u
s
in
g
(
2
)
.
=
∑
(
−
̅
)
(
−
̅
)
∑
(
−
̅
)
2
(2
)
W
h
er
e
̅
an
d
̅
ar
e
th
e
m
ea
n
o
f
in
d
ep
en
d
en
t
v
a
r
iab
le
an
d
d
e
p
en
d
en
t
v
ar
iab
le
r
esp
ec
tiv
el
y
a
n
d
an
d
ar
e
th
e
v
alu
es
o
f
in
d
ep
en
d
en
t
v
ar
i
ab
le
an
d
d
ep
en
d
en
t
v
ar
iab
le
r
esp
ec
tiv
ely
.
T
h
is
attem
p
t
was
m
ad
e
to
m
o
d
el
th
e
co
r
r
elatio
n
b
etwe
en
th
e
s
et
o
f
24
f
ea
tu
r
es,
b
o
th
d
em
o
g
r
a
p
h
ic
an
d
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es,
ex
tr
ac
ted
in
th
e
p
r
ev
io
u
s
m
o
d
u
les,
wh
ich
ar
e
tak
en
to
b
e
th
e
in
d
e
p
en
d
e
n
t
v
ar
iab
les
an
d
th
e
ap
n
ea
-
h
y
p
o
p
n
ea
in
d
ex
(
AHI
)
as
th
e
d
ep
en
d
e
n
t
v
ar
iab
le.
Fro
m
s
ev
er
al
r
esea
r
ch
es
it
i
s
m
ad
e
clea
r
th
at
th
e
OSA
s
ev
er
ity
ca
n
b
e
d
ef
in
ed
b
ased
o
n
AHI
i
n
d
ex
an
d
is
r
ec
o
m
m
en
d
ed
b
y
th
e
Am
e
r
ican
A
ca
d
em
y
o
f
s
leep
m
ed
icin
e.
I
t
v
ar
ies
f
r
o
m
m
ild
(5
≤
AHI
≤
1
5
e
v
en
ts
/h
o
u
r
)
,
m
o
d
er
ate
(
1
5
≤
AHI
≤
3
0
e/h
)
to
s
ev
er
e
(
AHI
≥
3
0
e/h
)
.
I
n
d
ep
e
n
d
e
n
t
o
f
th
e
ass
o
ciate
d
s
y
m
p
to
m
s
,
OSA
ca
n
b
e
d
iag
n
o
s
ed
in
p
atien
t
s
with
a
f
r
eq
u
en
cy
o
f
o
b
s
tr
u
ctiv
e
r
esp
ir
ato
r
y
d
is
tu
r
b
an
ce
s
g
r
ea
ter
t
h
an
1
5
e/
h
.
L
in
ea
r
r
eg
r
ess
io
n
s
er
v
e
s
to
i
n
ter
p
r
et
th
e
f
u
n
ctio
n
al
r
elati
o
n
s
h
ip
b
etwe
en
AHI
an
d
ea
ch
o
f
th
e
f
ea
tu
r
es
in
d
iv
id
u
ally
,
an
d
th
e
n
p
r
ed
ict
t
h
e
f
u
t
u
r
e
v
al
u
e
o
f
th
e
tar
g
et
v
a
r
iab
le,
i.e
.
,
AHI
in
o
u
r
ca
s
e,
b
ased
o
n
th
e
r
elatio
n
s
h
ip
i
n
ter
p
r
eted
.
T
h
e
alg
o
r
ith
m
f
o
r
lin
ea
r
r
eg
r
ess
io
n
was
im
p
lem
en
ted
i
n
m
at
lab
an
d
th
e
b
est
f
it
s
tr
aig
h
t
-
lin
e
eq
u
atio
n
s
r
elatin
g
v
a
r
io
u
s
f
ea
tu
r
es
with
AHI
wer
e
o
b
tain
e
d
f
r
o
m
th
e
lin
ea
r
r
eg
r
ess
io
n
p
lo
t.
T
h
e
r
elatio
n
b
etwe
en
ea
ch
o
f
th
e
f
ea
tu
r
es
as
m
en
tio
n
e
d
i
n
T
ab
le
1
,
with
AHI
was
in
ter
p
r
eted
u
s
in
g
lin
ea
r
r
eg
r
ess
io
n
m
o
d
el.
Per
h
ap
s
it
was
lab
o
u
r
-
i
n
ten
s
iv
e
to
p
r
ed
ic
t
th
e
y
v
alu
e
f
o
r
ea
c
h
in
d
iv
id
u
al
f
ea
tu
r
e
co
n
tain
ed
in
th
e
f
ea
tu
r
e
v
ec
to
r
X
F
,
f
r
o
m
th
e
s
ig
n
al
p
r
o
ce
s
s
in
g
m
o
d
u
le
.
T
h
e
an
aly
s
is
o
f
th
e
o
b
s
er
v
at
io
n
s
r
ev
ea
led
th
at
f
ew
o
f
th
e
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es
v
iz.
,
s
y
s
to
lic
p
ea
k
,
d
iast
o
lic
p
ea
k
,
au
g
m
e
n
tatio
n
in
d
e
x
,
t
pi
,
t
pp
ex
tr
ac
ted
f
r
o
m
th
e
PP
G
s
ig
n
al
h
ad
a
cl
o
s
e
r
elatio
n
s
h
ip
with
th
e
tar
g
e
t
v
ar
iab
le
an
d
h
ad
g
o
o
d
p
r
ed
i
ctio
n
ac
cu
r
ac
y
.
T
h
e
eq
u
atio
n
s
r
elatin
g
s
y
s
to
lic
p
ea
k
(
x
)
with
AHI
an
d
au
g
m
e
n
tatio
n
in
d
ex
(
y/x
)
with
AHI
o
b
tai
n
ed
f
r
o
m
t
h
e
lin
ea
r
r
eg
r
ess
io
n
p
lo
t is g
iv
e
n
in
(
3
)
an
d
(
4
)
re
s
p
ec
tiv
ely
.
=
(
1
.
1471
)
∗
+
(
20
.
7108
)
(
3
)
=
(
(
68
.
0199
)
∗
+
(
27
.
7509
)
)
(
4
)
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
A
n
a
u
to
m
a
tic
s
creen
in
g
a
p
p
r
o
a
ch
fo
r
o
b
s
tr
u
ctive
s
leep
a
p
n
e
a
fr
o
m
…
(
S
mily
Je
ya
Jo
t
h
i.
E
)
1267
T
h
e
lin
ea
r
r
e
g
r
ess
io
n
class
if
ie
r
was
s
elec
ted
b
ec
au
s
e
o
f
its
s
im
p
licity
in
p
r
o
g
r
am
m
in
g
an
d
ea
s
e
o
f
p
r
ed
ictin
g
n
u
m
er
ical
v
al
u
es.
I
t
was
ab
le
to
p
r
ed
ict
th
e
s
ev
e
r
ity
o
f
OSA
o
v
er
a
wid
e
r
an
g
e
an
d
th
e
p
r
e
d
icted
v
alu
es
wer
e
clo
s
e
to
th
e
r
ef
er
en
ce
v
alu
es
with
a
h
ig
h
est
er
r
o
r
p
er
ce
n
tag
e
o
f
±
7
as
s
h
o
wn
in
T
ab
le
3
.
T
h
e
m
ajo
r
d
is
ad
v
an
tag
e
with
s
u
c
h
a
lin
ea
r
p
r
ed
icto
r
is
th
at
th
e
in
ab
ilit
y
to
co
n
f
ir
m
th
e
d
ia
g
n
o
s
is
o
f
OSA
in
a
p
atien
t
o
n
th
e
b
asis
o
f
a
s
in
g
l
e
f
ea
tu
r
e
p
ar
a
m
eter
.
I
f
r
eq
u
i
r
e
d
to
in
clu
d
e
all
th
e
f
ea
tu
r
es,
it
wo
u
ld
b
e
a
lab
o
u
r
-
in
ten
s
iv
e
an
d
tim
e
-
c
o
n
s
u
m
in
g
p
r
o
ce
s
s
.
T
ab
le
3
.
Pre
d
icted
ap
n
ea
-
h
y
p
o
p
n
ea
in
d
e
x
(
AHI
)
f
r
o
m
th
r
ee
f
ea
tu
r
es u
s
in
g
lin
ea
r
r
e
g
r
ess
io
n
S
u
b
j
e
c
t
R
e
f
.
A
H
I
e
/
h
S
y
st
o
l
i
c
p
e
a
k
(
x
)
P
u
l
s
e
i
n
t
e
r
v
a
l
(
t
pi
)
P
e
a
k
t
o
p
e
a
k
i
n
t
e
r
v
a
l
(
t
pp
)
P
e
a
k
v
o
l
t
.
(
mv
)
P
r
e
d
.
A
H
I
e
/
h
Er
r
.
%
t
pi
(
sec
)
P
r
e
d
.
A
H
I
e
/
h
Er
r
.
%
t
pp
(
sec
)
P
r
e
d
.
A
H
I
e
/
h
Er
r
.
%
S1
4
1
.
0
2
3
.
8
9
2
.
7
5
0
.
7
4
3
.
9
8
0
.
5
1
.
2
5
3
.
7
7
5
.
7
5
S2
7
0
.
9
8
7
.
0
1
-
0
.
1
4
0
.
7
6
6
.
9
8
7
0
.
1
8
1
.
1
9
6
.
9
7
0
.
4
2
S3
OSA
17
0
.
7
6
1
6
.
2
7
8
4
.
2
4
0
.
9
9
1
6
.
8
6
0
.
8
2
1
.
4
3
16
5
.
8
8
S4
OSA
23
0
.
6
1
2
2
.
7
6
1
.
0
4
1
.
0
7
2
1
.
9
8
4
.
4
3
1
.
3
7
2
1
.
4
7
6
.
6
5
S5
OSA
27
0
.
6
9
2
7
.
0
3
4
-
0
.
1
2
1
.
3
8
26
3
.
7
1
.
4
0
2
6
.
8
4
0
.
6
3.
3
.
2
.
M
ultiv
a
ria
t
e
re
g
re
s
s
io
n (
M
R)
I
n
o
r
d
er
t
o
o
v
er
co
m
e
th
is
s
n
ag
o
f
p
r
ed
ictin
g
AHI
in
d
ex
f
o
r
ea
ch
in
d
iv
i
d
u
al
f
ea
tu
r
e,
we
att
em
p
ted
t
o
im
p
lem
en
t
th
e
s
tatis
tical
tec
h
n
iq
u
e:
m
u
ltip
le
lin
ea
r
r
e
g
r
e
s
s
io
n
/m
u
ltiv
ar
iate
r
eg
r
ess
io
n
,
wh
ich
u
s
e
s
ev
er
al
in
d
ep
en
d
en
t/p
r
ed
icto
r
v
ar
iab
l
es.
I
t
cr
ea
tes
a
lin
ea
r
r
elatio
n
s
h
ip
in
th
e
f
o
r
m
o
f
a
s
tr
aig
h
t
lin
e
t
h
at
b
est
ap
p
r
o
x
im
ates
all
th
e
in
d
iv
id
u
al
d
ata
p
o
in
t
s
an
d
also
h
elp
s
to
d
eter
m
in
e
th
e
p
o
ten
tial
v
a
r
iab
les
th
at
ca
n
b
e
im
p
o
r
tan
t
p
r
ed
icto
r
s
f
o
r
a
g
i
v
en
d
ep
e
n
d
en
t
v
ar
iab
le
,
AHI
in
o
u
r
ca
s
e.
T
h
e
g
en
er
al
m
o
d
el
o
f
m
u
ltiv
ar
iate
r
eg
r
ess
io
n
f
o
r
n
v
a
r
iab
les is
g
iv
en
b
y
(
5
)
.
=
1
1
+
2
2
+
⋯
+
(5
)
wh
er
e
b
1
to
b
n
ar
e
th
e
r
e
g
r
ess
io
n
co
ef
f
icien
ts
r
ep
r
esen
tin
g
t
h
e
v
alu
e
at
wh
ich
th
e
d
ep
en
d
en
t
v
ar
iab
le
ch
an
g
es
wh
en
th
e
in
d
ep
en
d
en
t
v
a
r
iab
l
e
ch
an
g
es.
T
h
e
eq
u
atio
n
r
elat
in
g
s
y
s
to
lic
p
ea
k
(
x
)
,
p
ea
k
to
p
ea
k
in
ter
v
al
(
t
pp
)
an
d
au
g
m
en
tatio
n
i
n
d
ex
(
y
/x
)
with
AHI
o
b
tain
ed
f
r
o
m
t
h
e
m
u
ltip
le
lin
ea
r
r
eg
r
ess
io
n
p
lo
t is g
iv
en
in
6
.
=
(
1
.
1471
)
∗
+
(
20
.
7108
)
∗
+
(
13
.
1487
)
∗
+
1
.
0
9
8
(6
)
C
o
m
b
in
atio
n
s
o
f
m
o
r
p
h
o
lo
g
i
ca
l,
s
p
ec
tr
al
an
d
s
tatis
tical
f
e
atu
r
es
wer
e
ch
o
s
en
o
n
a
tr
ial
-
an
d
-
er
r
o
r
b
asis
an
d
all
p
o
s
s
ib
le
co
m
b
in
atio
n
s
wer
e
p
r
ac
tically
im
p
le
m
en
ted
.
As
d
is
p
lay
ed
i
n
T
ab
l
e
4
,
r
esp
ir
atio
n
r
ate
an
d
h
ea
r
t
r
ate
co
m
b
in
ed
with
f
ew
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es
o
f
PP
G
d
ep
icted
th
e
least
er
r
o
r
p
er
ce
n
tag
e.
Yet,
th
e
p
r
ed
icted
ap
n
ea
-
h
y
p
o
p
n
ea
in
d
ex
f
o
r
OSA
p
atien
ts
s
h
o
we
d
less
er
ac
cu
r
ac
y
o
v
er
a
wid
e
r
an
g
e.
Ap
ar
t
f
r
o
m
s
elec
tin
g
th
e
r
ig
h
t
co
m
b
i
n
atio
n
o
f
th
e
f
ea
t
u
r
es
wh
ich
was
a
lab
o
r
-
i
n
ten
s
iv
e
p
r
o
ce
s
s
,
th
e
c
o
m
p
u
tatio
n
al
tim
e
r
eq
u
ir
ed
is
less
co
m
p
ar
ed
to
o
th
er
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
.
T
ab
le
4
.
Pre
d
icted
ap
n
ea
-
h
y
p
o
p
n
ea
in
d
e
x
(
AHI
)
c
o
m
b
in
i
n
g
f
iv
e
f
ea
tu
r
es u
s
in
g
MR tec
h
n
i
q
u
e
S
u
b
j
e
c
t
s
R
e
f
.
A
H
I
e
/
h
S
y
st
o
l
i
c
p
e
a
k
(
x
)
mv
P
e
a
k
t
o
p
e
a
k
i
n
t
e
r
v
a
l
(
t
pp
)
Re
s
p
.
R
a
t
e
/
m
i
n
P
u
l
s
e
I
n
t
e
r
v
a
l
(
t
pi
)
H
e
a
r
t
R
a
t
e
b
e
a
t
s/
mi
n
P
r
e
d
i
c
t
e
d
A
H
I
e
/
h
%
Er
r
o
r
S
1
4
1
.
3
8
0
.
7
7
1
5
.
8
0
.
7
5
74
3
.
9
3
1
.
7
5
S
2
5
1
.
0
9
0
.
8
3
15
0
.
7
3
76
4
.
8
9
2
.
2
S
3
7
1
.
0
3
0
.
9
8
1
5
.
6
0
.
8
1
81
6
.
7
4
0
.
8
6
S
4
9
0
.
9
2
0
.
8
9
1
3
.
3
0
.
7
9
78
8
.
5
7
4
.
7
7
S
5(O
S
A
)
11
0
.
7
9
1
.
1
5
1
0
.
8
0
.
9
3
83
9
.
6
8
12
S
6(O
S
A
)
17
0
.
7
7
1
.
1
9
1
0
.
2
0
.
9
8
71
1
5
.
0
4
1
1
.
5
2
S
7(O
S
A
)
20
0
.
6
5
1
.
2
9
.
8
1
.
0
4
68
1
6
.
9
9
1
5
.
0
5
S
8(O
S
A
)
24
0
.
7
1
1
.
1
8
6
.
5
1
.
2
5
64
2
0
.
3
8
1
5
.
0
8
3.
3
.
3
.
Su
pp
o
rt
v
ec
t
o
r
m
a
chi
nes
(
SV
M
)
A
d
ee
p
lear
n
in
g
alg
o
r
ith
m
w
h
ich
p
er
f
o
r
m
s
s
u
p
er
v
is
ed
lear
n
in
g
o
n
a
s
et
o
f
lab
eled
tr
ain
i
n
g
d
ata
is
u
s
ed
[
7
]
,
[
2
6
]
.
B
ased
o
n
th
e
i
n
f
er
en
ce
o
b
tai
n
ed
f
r
o
m
th
e
tr
ain
in
g
s
et,
th
e
al
g
o
r
ith
m
co
n
s
tr
u
cts
a
s
et
o
f
h
y
p
e
r
p
lan
es
in
a
h
ig
h
d
im
en
s
io
n
a
l
s
p
ac
e
wh
er
e
th
e
n
o
n
-
lin
ea
r
n
-
d
im
en
s
io
n
al
in
p
u
t
v
ec
to
r
is
m
ap
p
ed
in
to
a
K
-
d
im
en
s
io
n
al
f
ea
tu
r
e
s
p
ac
e
v
ia
a
k
er
n
el
tr
ick
.
SVM
u
s
e
s
a
k
er
n
el
f
u
n
ctio
n
to
tr
an
s
f
o
r
m
t
h
e
in
p
u
t
d
ata
in
to
a
h
ig
h
er
d
im
en
s
io
n
s
p
ac
e
an
d
th
en
class
if
y
th
e
d
ata
in
to
two
cl
ass
es
b
y
co
n
s
tr
u
ctin
g
a
s
ep
ar
at
in
g
h
y
p
er
p
la
n
e
in
th
e
tr
an
s
f
o
r
m
ed
s
p
ac
e.
T
h
e
d
ata
v
ec
to
r
s
n
ea
r
est
to
th
e
h
y
p
er
p
lan
e
in
th
e
tr
an
s
f
o
r
m
ed
s
p
ac
e
ar
e
th
e
s
u
p
p
o
r
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
4
,
Au
g
u
s
t 2
0
2
1
:
1
2
6
0
-
1
2
7
2
1268
v
ec
to
r
s
.
E
ac
h
in
p
u
t
d
ata
v
ec
to
r
is
m
ap
p
ed
in
to
a
h
ig
h
er
d
im
en
s
io
n
f
ea
tu
r
e
s
p
ac
e
K
v
ia
a
n
o
n
-
lin
ea
r
m
a
p
p
in
g
Φ(x
)
.
I
n
(
7
)
d
e
f
in
e
th
e
e
q
u
atio
n
o
f
th
e
h
y
p
er
p
la
n
e
s
ep
ar
atin
g
th
e
d
ata
in
t
o
two
d
if
f
er
en
t c
l
ass
es.
(
)
=
(
)
(7
)
(
)
=
∑
=
1
(
)
+
0
=
0
(8
)
wh
er
e
W
=
[
w
0
,
w
1
,
w
2
,
….
.
,
w
K
]
is
th
e
weig
h
t
v
ec
to
r
.
L
in
ea
r
k
er
n
el
an
d
s
ec
o
n
d
-
o
r
d
er
p
o
ly
n
o
m
ial
k
er
n
el
f
u
n
ctio
n
s
wer
e
e
x
p
er
im
e
n
ted
with
d
if
f
er
en
t
v
alu
es
o
f
th
e
r
eg
u
lar
izatio
n
p
ar
am
eter
C
(
C
=
0
.
1
,
1
,
5
,
8
)
as
s
h
o
wn
in
T
ab
le
5
.
T
h
e
v
alu
e
o
f
p
ar
am
ete
r
C
h
as
b
ee
n
id
e
n
t
if
ied
o
n
th
e
b
asis
o
f
lo
west
m
is
clas
s
if
icatio
n
r
ate
o
n
th
e
test
in
g
d
ataset
an
d
th
e
b
est
s
u
ited
co
n
f
ig
u
r
atio
n
f
o
r
t
h
e
p
o
ly
n
o
m
ial
k
er
n
el
f
u
n
cti
o
n
s
wer
e
estab
lis
h
ed
b
ased
o
n
in
d
ep
en
d
en
t
tr
ials
u
s
in
g
th
e
test
in
g
d
ata.
T
h
e
alg
o
r
ith
m
was
im
p
lem
en
ted
u
s
in
g
SVM
to
o
lb
o
x
f
o
r
Ma
tlab
.
T
ab
le
5
.
Pre
d
icted
ap
n
ea
-
h
y
p
o
p
n
ea
in
d
e
x
(
AHI
)
u
s
in
g
SVM
(
s
ec
o
n
d
-
o
r
d
er
p
o
ly
n
o
m
ial
k
e
r
n
el
with
C
=1
0
)
S
u
b
j
e
c
t
R
e
f
.
A
H
I
e
/
h
R
e
s
p
i
r
a
t
o
r
y
R
a
t
e
/
mi
n
S
p
O
2
M
o
r
p
h
o
l
o
g
i
c
a
l
F
e
a
t
u
r
e
R
e
s
p
.
r
a
t
e
/
mi
n
P
r
e
d
.
A
H
I
e
/
h
Er
r
.
%
O
2
sat
u
.
P
r
e
d
.
A
H
I
e
/
h
Er
r
.
%
P
r
e
d
.
A
H
I
e
/
h
Er
r
.
%
S1
3
1
6
.
5
6
3
.
2
5
-
8
.
3
99
3
.
0
3
-
1
.
0
3
.
5
7
-
1
9
.
0
S2
7
1
5
.
8
6
.
6
5
.
7
1
99
6
.
9
8
0
.
2
6
.
2
7
1
0
.
4
2
S3
OSA
14
1
0
.
2
1
2
.
8
8
.
5
7
97
1
3
.
5
6
3
.
0
12
1
4
.
2
8
S4
OSA
21
8
.
3
2
0
.
6
1
.
9
0
96
20
4
.
0
2
2
.
7
6
-
8
.
3
8
S5
OSA
33
7
.
7
30
9
.
0
9
94
3
0
.
7
6
.
0
2
8
.
8
1
2
.
7
2
As
r
esp
ir
atio
n
r
ate
an
d
o
x
y
g
e
n
s
atu
r
atio
n
f
ea
tu
r
es
ar
e
clo
s
ely
ass
o
ciate
d
with
O
SA,
cla
s
s
if
icatio
n
was
d
o
n
e
f
o
r
th
ese
two
f
ea
tu
r
es
s
ep
ar
ately
an
d
a
clas
s
if
icati
o
n
u
s
in
g
all
th
e
m
o
r
p
h
o
lo
g
ica
l
f
ea
tu
r
es
co
m
b
in
ed
was
also
ex
ec
u
ted
.
T
ab
le
5
s
h
o
ws
th
e
class
if
ier
r
esu
lt
s
o
f
r
esp
ir
ato
r
y
r
ate,
Sp
O
2
an
d
t
h
e
c
o
m
b
in
ed
f
ea
tu
r
e
s
et.
T
h
e
p
o
ly
n
o
m
ial
k
er
n
el
d
ep
ict
ed
h
ig
h
er
ac
c
u
r
ac
y
,
s
en
s
itiv
ity
an
d
s
p
ec
if
icity
co
m
p
ar
ed
t
o
th
e
lin
ea
r
k
er
n
el
class
if
icatio
n
.
Sp
O
2
h
ad
th
e
h
i
g
h
est
s
en
s
itiv
ity
an
d
ac
c
u
r
ac
y
wh
ile
r
esp
ir
ato
r
y
r
ate
h
ad
th
e
h
ig
h
est
s
p
ec
if
icity
.
T
h
e
p
e
r
f
o
r
m
an
ce
o
f
th
e
class
i
f
ier
f
o
r
s
ep
a
r
ate
f
ea
t
u
r
e
h
ad
s
h
o
wn
b
etter
r
esu
lts
th
an
th
e
c
o
m
b
in
ed
f
ea
tu
r
e
s
et
.
I
n
g
en
e
r
al,
th
e
ac
cu
r
ac
y
in
c
r
ea
s
ed
with
in
cr
ea
s
e
in
C
v
alu
e.
T
h
e
h
i
g
h
est
p
er
f
o
r
m
a
n
ce
(
Acc
:
9
6
.
5
%,
Sen
:
9
8
.
0
%,
Sp
ec
:
9
9
.
3
%)
was
ac
h
iev
ed
with
th
e
s
ec
o
n
d
-
o
r
d
er
p
o
ly
n
o
m
ial
with
C
=1
0
f
o
r
Sp
O
2
.
T
h
e
class
if
icatio
n
o
f
ap
n
eic
an
d
n
o
r
m
al
s
ig
n
als
b
ased
o
n
o
n
e
s
in
g
le
f
ea
tu
r
e
ca
n
n
o
t
b
e
i
n
co
r
p
o
r
ated
in
to
au
to
m
atic
s
cr
ee
n
in
g
alg
o
r
ith
m
s
.
3.
3
.
4
.
Ra
nd
o
m
f
o
re
s
t
(
RF
)
R
F
is
a
s
u
p
er
v
is
ed
m
ac
h
in
e
le
ar
n
in
g
tec
h
n
iq
u
e
u
s
ed
f
o
r
clas
s
if
icatio
n
an
d
r
e
g
r
ess
io
n
.
I
t
o
p
er
ates
b
y
cr
ea
tin
g
d
ec
is
io
n
tr
ee
s
b
ased
o
n
th
e
f
ea
tu
r
e
p
a
r
am
eter
s
d
u
r
i
n
g
th
e
tr
ain
in
g
p
r
o
ce
s
s
an
d
ac
q
u
ir
e
th
e
p
r
e
d
ictio
n
f
r
o
m
ea
ch
o
f
th
e
p
ar
am
e
ter
[
2
8
]
.
T
h
e
p
r
ec
is
io
n
o
f
th
e
r
esu
lt
in
cr
ea
s
es with
in
cr
ea
s
e
in
th
e
n
u
m
b
er
o
f
tr
ee
s
an
d
also
av
o
id
s
o
v
er
f
itti
n
g
o
f
th
e
m
o
d
el.
T
h
e
al
g
o
r
ith
m
w
o
r
k
s
i
n
two
s
tag
es,
o
n
e
is
th
e
co
n
s
tr
u
ctio
n
o
f
t
h
e
f
o
r
est
wh
ich
is
co
m
p
letely
a
r
an
d
o
m
p
r
o
ce
s
s
an
d
th
e
o
th
er
is
to
m
ak
e
a
p
r
ed
ictio
n
f
r
o
m
th
e
clas
s
if
ier
f
o
r
m
ed
in
th
e
f
ir
s
t
s
tag
e.
I
n
itially
th
e
alg
o
r
ith
m
r
an
d
o
m
ly
s
elec
ts
“
m
”
f
ea
t
u
r
es
f
r
o
m
a
to
tal
o
f
“
n
”
f
ea
tu
r
es,
wh
er
e
m<<n
.
Usi
n
g
th
e
b
est
s
p
lit
p
o
in
t,
n
o
d
es
an
d
d
au
g
h
ter
n
o
d
es
ar
e
f
o
r
m
ed
am
o
n
g
t
h
e
r
a
n
d
o
m
ly
s
elec
ted
m
f
ea
tu
r
es.
T
ab
le
6
d
e
p
ic
ts
th
e
p
er
f
o
r
m
a
n
ce
o
f
r
a
n
d
o
m
f
o
r
est
alg
o
r
it
h
m
f
o
r
m
o
r
p
h
o
l
o
g
ical,
s
tatis
t
ical
an
d
co
m
b
in
ed
f
ea
tu
r
es r
esp
ec
tiv
ely
u
s
in
g
2
0
d
ec
is
io
n
tr
ee
s
.
T
ab
le
6
.
Pre
d
icted
ap
n
ea
-
h
y
p
o
p
n
ea
in
d
e
x
(
AHI
)
u
s
in
g
r
a
n
d
o
m
f
o
r
est with
2
0
d
ec
is
io
n
tr
ee
s
S
u
b
j
e
c
t
R
e
f
.
A
H
I
e
/
h
M
o
r
p
h
o
l
o
g
i
c
a
l
f
e
a
t
u
r
e
s
S
t
a
t
i
st
i
c
a
l
f
e
a
t
u
r
e
s
C
o
m
b
i
n
e
d
f
e
a
t
u
r
e
s
P
r
e
d
.
A
H
I
e
/
h
Er
r
.
%
P
r
e
d
.
A
H
I
e
/
h
Er
r
.
%
P
r
e
d
.
A
H
I
e
/
h
Er
r
.
%
S1
4
3
.
7
8
5
.
5
3
.
4
8
1
3
.
0
3
.
9
7
0
.
7
S2
7
6
.
4
1
8
.
0
5
.
7
1
8
.
0
6
.
7
7
3
.
2
S3
OSA
17
1
5
.
5
8
.
0
1
7
.
8
-
4
.
0
16
5
.
0
S4
OSA
23
2
1
.
3
7
.
0
2
0
.
5
1
0
.
0
2
2
.
8
0
.
8
S5
OSA
27
2
5
.
7
4
.
0
2
6
.
4
2
.
0
2
6
.
6
1
.
4
8
T
h
e
r
an
d
o
m
f
o
r
est
is
co
n
s
tr
u
cted
b
y
r
ep
ea
tin
g
th
e
ab
o
v
e
s
tep
s
u
n
til
th
e
r
eq
u
ir
ed
n
u
m
b
e
r
o
f
n
o
d
es
an
d
tr
ee
s
a
r
e
f
o
r
m
ed
.
T
h
e
r
a
n
d
o
m
f
o
r
est
alg
o
r
ith
m
p
r
ed
icts
th
e
o
u
tp
u
t
b
y
an
aly
zi
n
g
th
e
t
est
f
ea
tu
r
es
an
d
th
e
r
u
les
o
f
ea
ch
o
f
th
e
d
ec
is
io
n
tr
ee
s
an
d
s
to
r
es
th
e
p
r
ed
icted
o
u
tp
u
t.
Af
ter
co
m
p
u
tin
g
th
e
v
o
tes
f
o
r
ea
ch
o
f
th
e
p
r
ed
ic
ted
tar
g
et,
th
e
alg
o
r
ith
m
id
en
tifie
s
th
e
tar
g
et
with
th
e
h
ig
h
est
v
o
tin
g
as
th
e
f
in
al
p
r
e
d
icted
o
u
tp
u
t.
I
n
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
A
n
a
u
to
m
a
tic
s
creen
in
g
a
p
p
r
o
a
ch
fo
r
o
b
s
tr
u
ctive
s
leep
a
p
n
e
a
fr
o
m
…
(
S
mily
Je
ya
Jo
t
h
i.
E
)
1269
p
r
o
p
o
s
ed
wo
r
k
,
th
e
in
p
u
t
f
ea
t
u
r
e
s
et
in
clu
d
es
a
to
tal
o
f
2
4
f
ea
tu
r
es
an
d
4
0
0
0
d
ata
s
am
p
les
o
f
PP
G
s
ig
n
al.
T
h
e
n
u
m
b
er
o
f
d
ec
is
io
n
tr
ee
s
ch
o
s
en
in
itially
was
1
0
a
n
d
la
ter
2
0
.
T
h
e
p
er
f
o
r
m
an
ce
o
f
th
e
alg
o
r
ith
m
was
ex
am
in
ed
f
o
r
m
o
r
p
h
o
lo
g
ical
f
ea
tu
r
es
s
ep
ar
ately
,
s
tatis
tical
an
d
s
p
ec
tr
al
f
ea
tu
r
es
s
ep
a
r
ately
an
d
f
in
ally
f
o
r
all
th
e
co
m
b
i
n
ed
f
ea
tu
r
es.
T
h
e
h
i
g
h
est
p
er
f
o
r
m
an
ce
(
Acc
:
9
8
.
0
%,
Sen
:
9
8
.
6
%,
Sp
ec
:
9
9
.
3
%)
was
p
o
r
tr
ay
ed
f
o
r
a
r
an
d
o
m
f
o
r
est
o
f
2
0
d
ec
is
io
n
tr
ess
,
tr
ain
ed
with
th
e
en
tire
f
ea
tu
r
e
v
ec
to
r
.
T
h
e
o
u
tp
u
t
o
f
t
h
e
class
if
ier
is
an
ag
g
r
eg
atio
n
o
f
th
e
o
u
tp
u
ts
o
f
all
th
e
d
ec
is
io
n
tr
ee
s
o
f
th
e
f
o
r
est,
wh
ich
r
ed
u
ce
s
t
h
e
v
a
r
ian
ce
an
d
b
ias.
T
h
e
ex
ec
u
tio
n
tim
e
o
f
th
e
alg
o
r
ith
m
was
v
er
y
h
ig
h
co
m
p
ar
ed
to
o
th
er
r
eg
r
ess
io
n
al
g
o
r
ith
m
s
,
a
s
th
e
class
if
ier
f
ir
s
t
p
er
f
o
r
m
s
s
elec
tio
n
o
f
r
an
d
o
m
s
am
p
les
f
r
o
m
t
h
e
g
iv
e
n
f
ea
tu
r
e
v
ec
to
r
X
F
,
f
o
llo
wed
b
y
co
n
s
tr
u
ctio
n
o
f
a
d
ec
is
io
n
tr
ee
f
o
r
ev
er
y
s
am
p
l
e.
Pre
d
icted
r
esu
lts
ar
e
o
b
tain
ed
f
r
o
m
ea
c
h
d
ec
is
io
n
tr
ee
a
n
d
th
e
c
o
m
p
u
tatio
n
r
eq
u
ir
ed
h
er
e
ar
e
m
e
r
ely
co
m
p
ar
is
o
n
s
o
f
o
n
e
f
ea
tu
r
e
a
t
ea
ch
n
o
d
e
o
f
th
e
t
r
ee
.
B
y
in
tr
o
d
u
cin
g
co
m
p
lete
r
an
d
o
m
n
ess
in
th
e
s
elec
tio
n
o
f
th
e
s
am
p
le
f
ea
tu
r
e
at
e
v
er
y
n
o
d
e
o
f
th
e
d
ec
is
io
n
tr
ee
,
an
d
as
th
e
class
if
ier
tech
n
iq
u
e
m
a
k
es
co
m
p
a
r
is
o
n
b
etwe
en
o
n
ly
o
n
e
f
ea
tu
r
e
at
e
ac
h
n
o
d
e,
th
e
s
y
s
tem
is
m
o
r
e
p
o
wer
f
u
l
i
n
f
i
n
d
in
g
th
e
co
r
r
elatio
n
s
o
f
th
e
in
p
u
ts
with
o
u
t
th
e
n
ee
d
f
o
r
in
p
u
t
s
ca
lin
g
.
T
h
e
p
er
f
o
r
m
an
ce
o
f
th
e
alg
o
r
ith
m
was
ex
ce
llen
t
with
a
lo
w
er
r
o
r
r
at
e.
T
h
e
alg
o
r
ith
m
was
co
n
f
ig
u
r
ed
b
ased
o
n
t
h
e
n
u
m
b
e
r
o
f
f
ea
tu
r
es
co
m
p
ar
ed
at
ea
ch
n
o
d
e
o
f
t
h
e
tr
ee
an
d
th
e
n
u
m
b
er
o
f
d
ec
is
io
n
t
r
ee
s
an
d
i
m
p
lem
en
ted
th
r
o
u
g
h
Ma
tlab
.
4.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
T
h
e
ex
p
er
im
en
tal
wo
r
k
in
v
o
lv
ed
f
ir
s
t
o
f
id
en
tify
in
g
th
e
b
est
class
if
ier
alg
o
r
ith
m
f
o
r
th
e
g
o
o
d
s
ig
n
al
d
etec
tio
n
m
o
d
u
le.
Seco
n
d
ly
,
t
h
e
s
elec
tio
n
o
f
f
ea
tu
r
es
th
at
co
u
ld
m
ax
im
ize
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
was
an
im
p
o
r
tan
t
cr
iter
i
o
n
.
T
h
o
u
g
h
a
g
e,
weig
h
t,
b
o
d
y
m
ass
in
d
ex
wer
e
in
clu
d
ed
in
t
h
e
f
ea
tu
r
e
v
ec
t
o
r
,
th
ey
wer
e
n
o
n
-
p
r
ed
ictiv
e
p
ar
am
eter
s
an
d
wer
e
u
s
ed
o
n
ly
to
im
p
r
o
v
e
t
h
e
ac
cu
r
ac
y
o
f
th
e
s
y
s
tem
.
T
h
e
o
v
er
all
ef
f
icien
cy
o
f
th
e
class
if
ier
alg
o
r
ith
m
s
is
d
ef
in
ed
b
ased
o
n
its
ab
ilit
y
to
d
is
tin
g
u
is
h
th
e
n
o
r
m
al
an
d
th
e
ab
n
o
r
m
al
s
ig
n
als
co
r
r
ec
tly
.
T
h
e
s
en
s
itiv
ity
o
f
th
e
class
if
ier
i
s
d
ef
in
ed
a
s
th
e
p
er
ce
n
tag
e
o
f
ap
n
eic
s
ig
n
als
c
o
r
r
ec
tly
class
if
ied
an
d
s
p
ec
if
icity
o
f
t
h
e
class
if
ier
is
th
e
p
er
ce
n
tag
e
o
f
n
o
r
m
al
s
ig
n
als
co
r
r
ec
tly
class
if
ied
.
Un
iv
ar
iate
an
d
m
u
ltiv
a
r
iate
r
eg
r
ess
io
n
s
er
v
e
to
in
ter
p
r
et
t
h
e
f
u
n
ctio
n
al
r
elatio
n
s
h
ip
b
et
wee
n
AHI
an
d
ea
ch
o
f
th
e
f
ea
tu
r
es in
d
i
v
id
u
ally
an
d
i
n
g
r
o
u
p
s
r
esp
ec
tiv
ely
,
an
d
th
en
p
r
ed
ict
th
e
f
u
tu
r
e
v
alu
e
o
f
th
e
tar
g
et
v
ar
iab
le.
T
h
e
f
o
r
m
e
r
al
g
o
r
ith
m
was
s
elec
ted
f
o
r
its
s
im
p
licity
an
d
its
p
r
e
d
ictin
g
ac
cu
r
a
cy
,
with
th
e
h
ig
h
est
er
r
o
r
p
er
ce
n
tag
e
o
f
±
7
.
T
h
e
p
r
ed
icted
o
u
t
p
u
ts
f
o
r
R
R
,
t
pp
,
t
pi
,
HR
V
wer
e
v
er
y
clo
s
e
to
th
e
tar
g
et.
T
h
e
alg
o
r
ith
m
d
e
p
icted
co
n
s
is
ten
cy
o
f
th
e
p
r
ed
icted
o
u
tp
u
ts
o
v
e
r
a
wid
e
r
an
g
e
o
f
s
ig
n
als.
T
h
e
o
n
ly
d
is
ad
v
a
n
tag
e
was
th
at,
to
in
clu
d
e
all
th
e
m
o
r
p
h
o
lo
g
ical,
s
tatis
tical
an
d
s
p
ec
tr
al
f
ea
tu
r
es
was
a
lab
o
r
-
in
te
n
s
iv
e
p
r
o
ce
s
s
.
T
h
is
was
o
v
er
co
m
e
b
y
u
s
in
g
t
h
e
m
u
ltiv
ar
iate
r
eg
r
ess
io
n
as
i
t
s
er
v
ed
to
i
n
ter
p
r
et
th
e
f
u
n
ctio
n
al
r
elatio
n
s
h
ip
b
etwe
en
AHI
an
d
s
e
v
er
al
o
f
th
e
f
ea
tu
r
es
ad
d
ed
to
g
et
h
er
,
an
d
th
en
p
r
e
d
ict
th
e
f
u
tu
r
e
v
alu
e
o
f
th
e
tar
g
et
v
ar
iab
le.
T
h
e
alg
o
r
ith
m
s
h
o
wed
m
u
ch
n
o
n
-
lin
ea
r
ity
b
et
wee
n
n
o
r
m
al
a
n
d
a
p
n
eic
s
ig
n
als
an
d
p
o
r
tr
ay
ed
o
v
er
-
f
itti
n
g
f
o
r
s
ev
er
al
n
o
r
m
al
PP
G
s
ig
n
als.
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
wo
r
k
ed
r
elativ
ely
well
c
o
m
p
a
r
ed
to
t
h
e
a
b
o
v
e
tech
n
iq
u
e
as
i
n
T
ab
le
7
.
Ox
y
g
en
s
atu
r
atio
n
f
ea
t
u
r
es
p
r
o
v
id
ed
a
clea
r
m
a
r
g
in
o
f
s
ep
a
r
atio
n
b
etwe
en
th
e
n
o
r
m
al
a
n
d
th
e
a
p
n
eic
s
ig
n
als.
T
h
e
h
ig
h
est
p
er
f
o
r
m
an
ce
(
A
cc
:
9
6
.
5
%,
Sen
:
9
8
.
0
%,
Sp
ec
:
.
3
%)
was
ac
h
iev
ed
with
th
e
s
ec
o
n
d
-
o
r
d
er
p
o
ly
n
o
m
ial
with
C
=1
0
f
o
r
Sp
O
2
.
T
h
e
s
ec
o
n
d
o
r
d
er
p
o
ly
n
o
m
ial
k
er
n
el
d
e
p
icted
h
ig
h
er
a
cc
u
r
ac
y
,
s
en
s
itiv
ity
an
d
s
p
ec
if
icity
co
m
p
a
r
ed
to
t
h
e
lin
ea
r
k
er
n
el
class
if
ier
.
Sp
O
2
h
ad
t
h
e
h
i
g
h
est
s
en
s
itiv
ity
an
d
ac
cu
r
ac
y
wh
ile
r
esp
ir
ato
r
y
r
ate
h
a
d
th
e
h
ig
h
e
s
t
s
p
ec
if
icity
.
T
h
e
p
er
f
o
r
m
an
c
e
o
f
t
h
e
class
if
ier
f
o
r
s
ep
ar
ate
f
ea
tu
r
e
h
ad
s
h
o
wn
b
etter
r
esu
lts
th
an
th
e
co
m
b
in
e
d
f
ea
tu
r
e
s
et.
I
n
g
e
n
er
al,
th
e
ac
cu
r
ac
y
in
c
r
ea
s
ed
with
in
cr
ea
s
e
in
C
v
alu
e.
T
h
e
r
eliab
ilit
y
an
d
s
tab
ilit
y
o
f
th
e
m
ac
h
in
e
lear
n
in
g
alg
o
r
it
h
m
s
wer
e
ex
am
in
ed
b
y
cr
o
s
s
-
v
alid
atin
g
th
eir
p
er
f
o
r
m
a
n
ce
o
n
th
e
s
am
e
d
atab
ase
f
o
r
m
u
ltip
le
tim
es.
T
h
e
en
tire
d
ataset
was
s
h
u
f
f
led
r
an
d
o
m
l
y
an
d
was
s
p
lit
in
to
1
0
g
r
o
u
p
s
an
d
a
1
0
cr
o
s
s
-
f
o
ld
v
alid
atio
n
was
p
e
r
f
o
r
m
e
d
.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
r
an
d
o
m
f
o
r
est
alg
o
r
ith
m
ex
am
in
ed
all
th
e
co
m
b
in
ed
f
ea
t
u
r
es
d
ep
icted
ex
ce
llen
t
ac
cu
r
ac
y
as
s
ee
n
in
T
ab
le
8
.
T
h
e
h
ig
h
est
p
er
f
o
r
m
an
ce
(
Acc
:
9
8
.
0
%,
Se
n
:
9
8
.
6
%,
Sp
ec
:
9
9
.
3
%)
was
p
o
r
tr
ay
ed
f
o
r
a
r
an
d
o
m
f
o
r
est
o
f
2
0
d
ec
is
io
n
tr
ess
,
tr
ain
ed
with
th
e
en
tire
f
ea
tu
r
e
v
ec
to
r
.
I
n
cr
ea
s
e
in
th
e
n
u
m
b
er
o
f
d
ec
is
io
n
tr
ee
s
b
ey
o
n
d
2
0
s
h
o
wed
n
o
en
h
an
ce
m
e
n
t
in
its
p
er
f
o
r
m
an
ce
.
T
h
e
p
er
f
o
r
m
a
n
ce
was
b
ett
er
(
Acc
:
9
6
.
8
%,
Sen
:
9
5
.
9
%,
Sp
ec
:
9
5
.
9
%)
with
1
0
d
ec
is
io
n
tr
ee
s
f
o
r
th
e
m
o
r
p
h
o
lo
g
ic
al
f
ea
tu
r
es
o
n
ly
an
d
th
e
co
m
p
u
tatio
n
tim
e
tak
e
n
was
also
v
e
r
y
less
,
b
u
t
ag
ain
a
d
ec
is
io
n
ca
n
n
o
t
b
e
m
a
d
e
b
ased
o
n
o
n
l
y
f
ew
p
a
r
am
et
er
s
.
T
h
e
co
m
p
u
tatio
n
tim
e
r
eq
u
ir
e
d
was
m
o
r
e
f
o
r
th
e
co
m
b
in
ed
f
ea
tu
r
e
v
ec
to
r
m
o
d
el,
b
u
t
th
e
p
r
ed
ictio
n
ac
cu
r
ac
y
was
also
th
e
h
i
g
h
est.
Ou
t
o
f
s
am
p
le
test
in
g
a
n
d
c
r
o
s
s
v
alid
atio
n
m
eth
o
d
o
lo
g
y
w
as
u
s
ed
f
o
r
tu
n
in
g
th
e
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
to
im
p
r
o
v
e
th
eir
p
er
f
o
r
m
an
ce
.
As
s
h
o
wn
in
T
ab
le
s
7
an
d
8
,
th
e
r
an
d
o
m
f
o
r
e
s
t
a
l
g
o
r
it
h
m
g
a
v
e
b
e
t
t
e
r
r
es
u
l
ts
c
o
m
p
a
r
e
d
t
o
o
t
h
e
r
t
e
c
h
n
iq
u
e
s
,
w
it
h
t
h
e
p
e
r
f
e
c
t
s
et
o
f
f
ea
t
u
r
e
s
u
s
e
d
a
n
d
w
it
h
b
e
s
t
s
t
r
u
c
t
u
r
e
d
e
v
i
s
e
d
.
R
a
n
d
o
m
f
o
r
e
s
t
d
e
p
i
c
t
e
d
v
e
r
y
l
e
s
s
v
a
r
i
ab
i
l
i
t
y
i
n
d
i
c
at
i
n
g
b
e
t
t
e
r
s
t
a
b
i
li
t
y
o
f
t
h
e
m
e
t
h
o
d
.
T
h
e
o
t
h
e
r
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
it
h
m
s
s
h
o
w
e
d
m
u
c
h
v
a
r
i
a
b
i
l
it
y
wh
e
n
u
s
e
d
o
n
t
h
e
t
e
s
t
i
n
g
d
a
t
a
b
ase
a
g
a
i
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.