I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
pu
t
er
E
ng
ineering
(
I
J
E
CE
)
Vo
l.
15
,
No
.
6
,
Decem
b
er
20
25
,
p
p
.
5
6
5
5
~
5
6
6
6
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
15
i
6
.
pp
5
6
5
5
-
5
6
6
6
5655
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
M
a
chine
lea
rning
mo
del f
o
r acc
ura
te
pre
diction o
f
c
o
ro
na
ry
a
rter
y
disea
se by
inco
rpo
ra
ting err
o
r re
duct
io
n met
ho
do
lo
g
ies
Sa
ntho
s
h G
up
t
a
Do
g
ipa
rt
hi
1
,
J
a
y
a
nthi K
.
1
,
Aj
it
h Ana
ntha
k
rish
na
P
illa
i
2
, K
.
Na
k
k
ee
r
a
n
3
1
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
o
n
i
c
s a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
En
g
i
n
e
e
r
i
n
g
,
P
u
d
u
c
h
e
r
r
y
Te
c
h
n
o
l
o
g
i
c
a
l
U
n
i
v
e
r
si
t
y
,
P
u
d
u
c
h
e
r
r
y
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
C
a
r
d
i
o
l
o
g
y
,
K
a
u
v
e
r
y
H
o
sp
i
t
a
l
-
R
a
d
i
a
l
R
o
a
d
,
C
h
e
n
n
a
i
,
Ta
m
i
l
n
a
d
u
,
I
n
d
i
a
3
S
c
h
o
o
l
o
f
E
n
g
i
n
e
e
r
i
n
g
,
F
r
a
s
e
r
N
o
b
l
e
B
u
i
l
d
i
n
g
,
U
n
i
v
e
r
s
i
t
y
o
f
A
b
e
r
d
e
e
n
,
A
b
e
r
d
e
e
n
, U
n
i
t
e
d
K
i
n
g
d
o
m
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ap
r
9
,
2
0
2
5
R
ev
is
ed
J
u
l 1
6
,
2
0
2
5
Acc
ep
ted
Sep
1
4
,
2
0
2
5
Co
ro
n
a
r
y
a
rtery
d
ise
a
se
(CAD
)
re
m
a
in
s
a
lea
d
in
g
c
a
u
se
o
f
m
o
rtalit
y
wo
rld
wi
d
e
,
wit
h
a
n
e
sp
e
c
ially
h
i
g
h
b
u
rd
e
n
i
n
d
e
v
e
lo
p
in
g
c
o
u
n
tri
e
s
su
c
h
a
s
In
d
ia.
I
n
li
g
h
t
o
f
i
n
c
re
a
sin
g
p
a
t
ien
t
lo
a
d
s
a
n
d
li
m
it
e
d
m
e
d
ica
l
re
so
u
rc
e
s,
th
e
re
is
a
n
u
r
g
e
n
t
n
e
e
d
fo
r
a
c
c
u
r
a
te
a
n
d
re
li
a
b
le
d
ia
g
n
o
stic
su
p
p
o
rt
sy
ste
m
s.
Th
is
stu
d
y
i
n
tro
d
u
c
e
s
a
m
a
c
h
in
e
lea
rn
in
g
(M
L)
fra
m
e
wo
rk
th
a
t
a
ims
to
e
n
h
a
n
c
e
CAD
p
re
d
ictio
n
a
c
c
u
ra
c
y
b
y
s
p
e
c
ifi
c
a
ll
y
a
d
d
re
ss
in
g
th
e
re
d
u
c
ti
o
n
o
f
fa
lse
n
e
g
a
ti
v
e
s
(F
N),
wh
ic
h
a
r
e
c
rit
ica
l
in
m
e
d
ica
l
d
iag
n
o
st
ics
.
Util
izin
g
a
sta
c
k
e
d
e
n
se
m
b
le
m
o
d
e
l
c
o
m
p
risin
g
f
iv
e
b
a
se
c
las
sifiers
a
n
d
a
m
e
ta
-
c
las
sifier,
th
e
fra
m
e
wo
rk
in
teg
r
a
tes
c
o
st
-
se
n
siti
v
e
lea
rn
in
g
,
c
las
sifica
ti
o
n
th
re
sh
o
l
d
tu
n
i
n
g
,
e
n
g
in
e
e
re
d
fe
a
t
u
re
s,
a
n
d
m
a
n
u
a
l
we
ig
h
ti
n
g
stra
teg
ies
.
Th
e
m
o
d
e
l
wa
s
d
e
v
e
lo
p
e
d
u
sin
g
a
c
li
n
ica
ll
y
a
c
q
u
ired
d
a
tas
e
t
fro
m
th
e
Ja
wa
h
a
rlal
In
stit
u
te
o
f
p
o
stg
ra
d
u
a
te
m
e
d
i
c
a
l
e
d
u
c
a
ti
o
n
a
n
d
re
se
a
rc
h
(
JIPM
ER),
c
o
n
sistin
g
o
f
4
2
8
p
a
ti
e
n
t
re
c
o
r
d
s
with
3
6
o
ri
g
in
a
l
fe
a
tu
re
s.
Ev
a
lu
a
ti
o
n
m
e
tri
c
s
sh
o
w
t
h
a
t
t
h
e
p
ro
p
o
se
d
m
o
d
e
l
a
c
h
iev
e
d
a
n
a
c
c
u
ra
c
y
o
f
9
2
.
1
9
%
,
se
n
s
it
iv
it
y
o
f
9
8
%
,
a
n
d
a
n
F
1
-
s
c
o
re
o
f
9
5
.
1
5
%
.
T
h
e
se
imp
ro
v
e
m
e
n
ts
a
re
sig
n
ifi
c
a
n
t
in
a
c
li
n
ica
l
c
o
n
tex
t
,
p
o
ten
ti
a
ll
y
re
d
u
c
in
g
m
isse
d
d
ia
g
n
o
se
s
a
n
d
imp
ro
v
i
n
g
p
a
ti
e
n
t
o
u
tco
m
e
s.
T
h
e
m
o
d
e
l
is
i
n
ten
d
e
d
f
o
r
d
e
p
l
o
y
m
e
n
t
in
c
a
rd
io
lo
g
y
o
u
t
p
a
ti
e
n
t
se
tt
in
g
s
a
n
d
d
e
m
o
n
stra
tes
a
sc
a
lab
le,
a
d
a
p
tab
le
a
p
p
ro
a
c
h
to
m
e
d
ica
l
d
iag
n
o
stics
.
K
ey
w
o
r
d
s
:
Fals
e
n
eg
ativ
es
Featu
r
e
en
g
in
ee
r
i
n
g
Misclas
s
if
icatio
n
co
s
t
T
h
r
esh
o
ld
a
d
ju
s
tm
en
t
W
eig
h
tin
g
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
:
San
th
o
s
h
Gu
p
ta
Do
g
ip
ar
th
i
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
,
Pu
d
u
ch
er
r
y
T
ec
h
n
o
lo
g
ical
Un
iv
er
s
ity
Pu
d
u
ch
er
r
y
6
0
5
0
1
4
,
I
n
d
ia
E
m
ail: san
th
o
s
h
g
u
p
ta@
p
tu
n
iv
.
ed
u
.
in
1.
I
NT
RO
D
UCT
I
O
N
C
ar
d
io
v
ascu
lar
d
is
ea
s
es,
esp
ec
ially
co
r
o
n
ar
y
ar
ter
y
d
is
ea
s
e
(
C
AD)
,
ar
e
a
p
r
ed
o
m
in
an
t
ca
u
s
e
o
f
m
o
r
b
id
ity
an
d
m
o
r
tality
g
lo
b
a
lly
,
co
n
tr
ib
u
tin
g
t
o
ap
p
r
o
x
im
a
tely
2
8
.
1
%
o
f
all
d
ea
th
s
in
I
n
d
ia
alo
n
e
[
1
]
.
Giv
e
n
I
n
d
ia'
s
p
o
p
u
latio
n
s
u
r
p
ass
in
g
1
.
4
4
b
illi
o
n
an
d
a
n
ex
is
tin
g
d
o
cto
r
-
to
-
p
atien
t
r
atio
s
ig
n
if
ican
tly
b
elo
w
th
e
W
o
r
ld
Hea
lth
Or
g
an
izatio
n
’
s
r
ec
o
m
m
en
d
e
d
1
:1
0
0
0
[
2
]
,
t
h
e
ex
is
tin
g
h
ea
lth
ca
r
e
s
y
s
tem
f
ac
es
a
m
ass
iv
e
b
u
r
d
e
n
in
tim
ely
an
d
ac
c
u
r
at
e
d
iag
n
o
s
is
.
T
h
is
s
ce
n
ar
io
h
ig
h
lig
h
ts
th
e
n
ee
d
f
o
r
in
tellig
en
t
d
ec
is
io
n
-
s
u
p
p
o
r
t
s
y
s
tem
s
ca
p
ab
le
o
f
allev
iatin
g
th
e
d
iag
n
o
s
tic
wo
r
k
lo
ad
,
p
ar
ti
cu
lar
ly
in
r
eso
u
r
ce
-
co
n
s
tr
ain
e
d
en
v
ir
o
n
m
en
ts
.
T
h
e
ce
n
tr
al
p
r
o
b
lem
in
cu
r
r
e
n
t
m
ac
h
in
e
lear
n
in
g
(
ML
)
b
a
s
ed
d
iag
n
o
s
tic
m
o
d
els
f
o
r
C
AD
lies
in
th
eir
m
is
class
if
icatio
n
ten
d
en
cies,
esp
ec
ially
f
alse
n
eg
ativ
es
(
FN)
ca
s
e
s
wh
er
e
d
is
e
ased
p
atien
ts
ar
e
m
is
class
if
ied
as
h
ea
lth
y
.
T
h
is
p
o
s
es
a
s
ev
er
e
r
is
k
in
cli
n
ical
s
ettin
g
s
,
wh
er
e
ea
r
ly
d
etec
tio
n
is
cr
u
cial.
T
r
ad
itio
n
al
m
o
d
els
o
f
ten
o
p
ti
m
ize
f
o
r
ac
c
u
r
ac
y
o
r
AUC,
in
ad
v
er
ten
tly
n
eg
lectin
g
s
en
s
itiv
ity
an
d
e
r
r
o
r
co
s
ts
.
A
f
alse
n
eg
ativ
e
in
C
AD
d
iag
n
o
s
tics
ca
n
d
elay
life
-
s
av
i
n
g
in
ter
v
en
tio
n
s
an
d
m
is
lead
clin
ical
d
ec
is
io
n
s
.
T
h
er
ef
o
r
e,
r
ed
u
ci
n
g
FN
m
u
s
t
b
e
a
co
r
e
o
b
jectiv
e
f
o
r
an
y
ML
m
o
d
el
aim
ed
at
C
AD
p
r
ed
i
ctio
n
.
Sev
er
al
ML
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
6
,
Decem
b
e
r
20
25
:
5
6
5
5
-
5
6
6
6
5656
m
eth
o
d
o
l
o
g
ies ar
e
av
ailab
le
to
r
ed
u
ce
th
e
FN d
ep
en
d
in
g
o
n
t
h
e
s
p
ec
if
ics o
f
d
if
f
er
en
t f
ea
t
u
r
e
ch
ar
ac
ter
is
tics
o
f
th
e
in
p
u
t
d
ataset
an
d
th
e
m
o
d
el
alg
o
r
it
h
m
im
p
lem
en
ted
.
I
n
th
e
b
r
o
ad
er
lan
d
s
ca
p
e
o
f
p
r
ed
ictiv
e
m
o
d
elin
g
,
s
ev
er
al
m
eth
o
d
o
l
o
g
ies
s
u
ch
as
ad
ju
s
tin
g
th
e
class
if
ica
tio
n
th
r
esh
o
ld
[
3
]
,
c
o
s
t
-
s
en
s
itiv
e
lear
n
in
g
[
4
]
,
r
esam
p
lin
g
tech
n
iq
u
es
[
3
]
,
f
ea
tu
r
e
en
g
i
n
ee
r
in
g
[
5
]
,
alg
o
r
ith
m
s
elec
tio
n
[
6
]
,
d
ata
au
g
m
en
tatio
n
[
7
]
,
e
r
r
o
r
an
aly
s
is
[
3
]
,
en
s
em
b
le
m
eth
o
d
s
[
8
]
,
m
o
d
el
ca
lib
r
atio
n
[
9
]
,
ac
tiv
e
lear
n
in
g
[
1
0
]
,
m
an
u
al
an
d
p
er
f
o
r
m
an
ce
-
b
ased
weig
h
t
ad
ju
s
tm
en
t
[
3
]
,
[
1
0
]
,
h
y
p
er
p
ar
am
eter
tu
n
i
n
g
[
1
1
]
,
an
d
cu
s
to
m
s
tack
in
g
[
1
2
]
ar
e
co
m
m
o
n
ly
ad
o
p
ted
t
o
im
p
r
o
v
e
class
if
icatio
n
o
u
tco
m
es,
p
ar
ticu
lar
l
y
b
y
r
ed
u
cin
g
f
alse
n
eg
ati
v
es.
Ou
t
o
f
th
ese
m
eth
o
d
o
l
o
g
ies,
d
ep
e
n
d
in
g
u
p
o
n
th
e
p
r
e
d
ictio
n
m
o
d
el
d
ev
elo
p
e
d
,
d
if
f
er
en
t
m
et
h
o
d
o
lo
g
ies
ar
e
to
b
e
in
co
r
p
o
r
ated
to
r
e
d
u
ce
th
e
FN.
Selecte
d
s
et
o
f
m
eth
o
d
o
lo
g
ies
f
o
r
h
ea
r
t
d
is
ea
s
e
p
r
ed
ictio
n
s
y
s
tem
s
with
b
etter
p
r
ed
ictio
n
ac
cu
r
ac
y
an
d
r
ec
al
l
ar
e
v
ital
in
I
n
d
ia
to
co
m
b
at
th
e
g
r
o
win
g
h
ea
lth
is
s
u
es
d
u
e
to
ca
r
d
io
v
ascu
la
r
d
is
ea
s
es.
T
h
ey
f
ac
ilit
ate
ea
r
ly
d
etec
tio
n
,
o
p
tim
ize
r
eso
u
r
ce
u
tili
za
tio
n
,
r
ed
u
ce
h
ea
lth
ca
r
e
co
s
ts
,
an
d
s
u
p
p
o
r
t
b
etter
h
ea
lth
ca
r
e
o
u
tco
m
es,
c
o
n
tr
ib
u
tin
g
t
o
o
v
e
r
all
s
o
cieta
l w
ell
-
b
ein
g
.
Sev
er
al
wo
r
k
s
in
liter
atu
r
e
h
a
v
e
ex
p
l
o
r
ed
ML
-
b
ased
a
p
p
r
o
ac
h
es
f
o
r
C
AD
d
etec
tio
n
.
Fo
r
in
s
tan
ce
,
Kh
an
n
a
et
a
l.
[
1
3
]
c
o
m
p
ar
e
d
class
if
icatio
n
alg
o
r
ith
m
s
li
k
e
SVM
an
d
n
eu
r
al
n
etwo
r
k
s
f
o
r
h
ea
r
t
d
is
ea
s
e
p
r
ed
ictio
n
,
wh
ile
Ma
in
i
et
a
l.
[
1
4
]
em
p
h
asized
th
e
u
tili
ty
o
f
tailo
r
ed
p
r
ed
ictio
n
m
o
d
els f
o
r
I
n
d
ian
p
o
p
u
latio
n
s
.
Ho
wev
er
,
th
ese
s
tu
d
ies
p
r
im
a
r
ily
f
o
cu
s
ed
o
n
ac
cu
r
ac
y
en
h
a
n
ce
m
en
t
with
o
u
t
a
d
etailed
s
tr
ateg
y
f
o
r
h
an
d
lin
g
FN
er
r
o
r
s
.
Similar
ly
,
Z
r
iq
at
et
a
l.
[
1
5
]
a
n
d
B
ab
u
et
a
l.
[
1
6
]
d
is
cu
s
s
ed
im
p
r
o
v
ed
ML
ar
ch
i
tectu
r
es
b
u
t
lack
ed
tar
g
eted
er
r
o
r
-
c
o
s
t
an
aly
s
is
o
r
ex
p
licit
th
r
esh
o
ld
o
p
tim
izat
io
n
m
eth
o
d
s
.
T
h
u
s
,
d
esp
ite
t
h
e
p
r
o
g
r
ess
in
ML
ap
p
licatio
n
s
f
o
r
C
AD,
a
s
ig
n
if
ican
t g
ap
r
em
ai
n
s
in
h
an
d
lin
g
er
r
o
r
s
en
s
itiv
ity
,
p
a
r
ticu
lar
ly
r
ed
u
cin
g
FNs
.
T
h
is
s
tu
d
y
p
r
o
p
o
s
es
an
en
s
em
b
le
ML
m
o
d
el
tailo
r
ed
f
o
r
FN
r
ed
u
ctio
n
in
C
AD
p
r
ed
ictio
n
.
T
h
e
m
o
d
el
co
m
b
in
es
f
iv
e
d
iv
e
r
s
e
b
ase
lear
n
er
s
(
lin
ea
r
an
d
n
o
n
-
lin
ea
r
SVM,
k
-
n
ea
r
est
n
eig
h
b
o
r
s
,
r
an
d
o
m
f
o
r
est,
an
d
Ad
aBo
o
s
t)
with
a
s
tack
in
g
m
eta
-
class
if
ier
.
I
t
in
co
r
p
o
r
at
es
s
ev
er
al
n
o
v
el
s
tr
ateg
ies:
co
s
t
-
s
en
s
itiv
e
lear
n
in
g
[
4
]
t
o
p
e
n
alize
FN,
th
r
esh
o
l
d
ad
ju
s
tm
en
t
b
ased
o
n
p
r
ec
i
s
io
n
-
r
ec
all
tr
ad
e
-
o
f
f
s
[
3
]
,
d
o
m
ain
-
d
r
iv
e
n
m
a
n
u
al
weig
h
t
ass
ig
n
m
en
t
[
3
]
,
an
d
d
er
iv
atio
n
o
f
m
ea
n
in
g
f
u
l
co
m
p
o
s
ite
f
ea
tu
r
es
s
u
ch
as
p
u
ls
e
p
r
ess
u
r
e,
MA
P,
an
d
E
C
G
ab
n
o
r
m
ality
s
co
r
es.
T
h
ese
m
o
d
if
icatio
n
s
ar
e
aim
ed
at
f
in
e
-
tu
n
i
n
g
th
e
m
o
d
el
to
r
ed
u
ce
m
is
d
iag
n
o
s
es
wh
ile
m
ain
tain
in
g
g
e
n
er
aliza
tio
n
ab
ilit
y
.
Ou
r
i
n
n
o
v
ati
o
n
l
ies
in
t
h
e
i
n
te
g
r
ati
o
n
o
f
t
h
es
e
F
N
-
r
e
d
u
c
ti
o
n
m
et
h
o
d
o
lo
g
i
es
i
n
t
o
a
cli
n
i
ca
l
l
y
v
a
li
d
at
ed
en
s
em
b
l
e
m
o
d
el
.
Un
li
k
e
e
ar
li
er
s
t
u
d
ies
,
t
h
is
a
p
p
r
o
a
c
h
em
p
h
asiz
es
m
e
d
i
ca
l
s
af
et
y
b
y
l
o
we
r
i
n
g
FN
ca
s
es
f
r
o
m
s
ix
t
o
t
wo
o
n
a
4
2
8
-
p
ati
e
n
t
d
at
ase
t,
i
m
p
r
o
v
in
g
r
ec
all
f
r
o
m
9
4
%
t
o
9
8
%
a
s
u
b
s
t
a
n
ti
al
le
ap
i
n
d
i
a
g
n
o
s
t
ic
r
el
ia
b
il
it
y
.
T
h
is
w
o
r
k
n
o
t
o
n
l
y
a
d
d
r
ess
es
a
c
r
it
ic
al
c
li
n
ic
al
ch
all
en
g
e
b
u
t
als
o
s
e
ts
a
r
e
p
l
i
ca
b
le
p
r
e
ce
d
e
n
t
f
o
r
d
e
p
l
o
y
i
n
g
ML
s
y
s
t
em
s
i
n
r
e
al
-
w
o
r
ld
h
o
s
p
ita
l
w
o
r
k
f
l
o
ws.
T
h
e
p
r
o
p
o
s
e
d
m
o
d
el
is
c
u
r
r
e
n
tl
y
b
ei
n
g
p
r
ep
a
r
e
d
f
o
r
d
e
p
l
o
y
m
en
t
at
J
I
P
ME
R
a
n
d
m
ay
b
e
a
d
a
p
t
ed
f
o
r
o
th
er
d
is
e
ase
d
o
m
ai
n
s
wit
h
s
i
m
il
ar
f
ea
tu
r
e
s
ets.
T
h
e
r
em
ain
d
er
o
f
th
is
p
a
p
er
i
s
o
r
g
an
ized
as
f
o
llo
ws:
s
ec
tio
n
2
p
r
esen
ts
th
e
p
r
o
p
o
s
ed
m
o
d
el
an
d
its
th
eo
r
etica
l f
o
u
n
d
atio
n
.
Sectio
n
3
d
etails th
e
m
eth
o
d
o
lo
g
y
u
s
ed
in
b
u
ild
in
g
a
n
d
tu
n
in
g
th
e
m
l m
o
d
el.
Sectio
n
4
r
ep
o
r
ts
th
e
r
esu
lts
an
d
p
r
o
v
i
d
es
a
co
m
p
r
eh
en
s
iv
e
d
is
cu
s
s
io
n
co
m
p
ar
i
n
g
th
e
m
o
d
el’
s
p
er
f
o
r
m
an
ce
with
ex
is
tin
g
m
eth
o
d
s
.
Fin
ally
,
s
ec
t
io
n
5
c
o
n
clu
d
es th
e
p
ap
er
an
d
o
u
tlin
es d
ir
ec
tio
n
s
f
o
r
f
u
t
u
r
e
wo
r
k
.
2.
P
RO
P
O
SE
D
M
O
D
E
L
F
O
R
CAD
P
RE
D
I
CT
I
O
N
T
h
is
s
ec
tio
n
p
r
esen
ts
th
e
d
e
s
ig
n
an
d
th
e
o
r
etica
l
f
o
u
n
d
atio
n
o
f
th
e
p
r
o
p
o
s
ed
m
ac
h
in
e
lear
n
in
g
f
r
am
ewo
r
k
f
o
r
p
r
ed
ictin
g
c
o
r
o
n
ar
y
ar
ter
y
d
is
ea
s
e.
I
t
f
u
r
th
e
r
elab
o
r
ates
o
n
th
e
in
co
r
p
o
r
ati
o
n
o
f
m
u
ltip
le
f
alse
n
eg
ativ
e
(
FN)
r
ed
u
ctio
n
s
tr
at
eg
ies
to
th
e
e
n
s
em
b
le
lear
n
i
n
g
s
tr
u
ctu
r
e
to
im
p
r
o
v
e
d
iag
n
o
s
tic
r
eliab
ilit
y
b
y
m
in
im
izin
g
FN e
r
r
o
r
s
,
wh
ich
ar
e
cr
itical
in
m
ed
ical
d
ec
is
io
n
-
m
ak
in
g
.
2
.
1
.
St
a
ck
ed
ens
em
ble a
rc
hite
ct
ure
T
h
e
p
r
o
p
o
s
ed
m
ac
h
in
e
lea
r
n
i
n
g
m
o
d
el
is
d
esig
n
ed
as
a
s
tack
ed
en
s
em
b
le
f
r
a
m
ewo
r
k
th
a
t
in
teg
r
ates
m
u
ltip
le
b
ase
class
if
ier
s
to
lev
er
ag
e
th
e
in
d
iv
id
u
al
s
tr
en
g
th
s
o
f
ea
ch
.
Sp
ec
if
ically
,
it
co
m
b
i
n
es
f
iv
e
class
if
ier
s
v
iz.
,
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
(
SVM)
with
b
o
t
h
lin
ea
r
an
d
n
o
n
lin
ea
r
k
er
n
els
[
1
7
]
,
[
1
8
]
,
k
-
n
ea
r
est
n
ei
g
h
b
o
r
s
(
KNN)
[
1
9
]
,
r
an
d
o
m
f
o
r
est
(
R
F)
[
2
0
]
,
an
d
Ad
aBo
o
s
t
[
2
1
]
f
o
llo
wed
b
y
a
m
eta
-
class
if
ier
th
at
s
y
n
th
esizes
th
eir
o
u
tp
u
ts
f
o
r
f
in
al
d
ec
is
io
n
-
m
a
k
in
g
.
E
n
s
em
b
le
lear
n
in
g
m
et
h
o
d
s
,
p
ar
ticu
lar
ly
s
tack
in
g
,
ar
e
k
n
o
wn
to
e
n
h
an
ce
m
o
d
el
g
en
er
aliza
tio
n
an
d
p
er
f
o
r
m
an
ce
ac
r
o
s
s
d
iv
e
r
s
e
d
atasets
.
I
n
th
is
c
o
n
tex
t,
th
e
c
h
o
ic
e
o
f
class
if
ier
s
was
in
f
o
r
m
e
d
b
y
th
eir
co
m
p
lem
en
tar
y
n
atu
r
e:
SVM
f
o
r
m
a
r
g
in
-
b
ased
s
ep
ar
atio
n
,
KNN
f
o
r
lo
ca
l
d
ec
is
io
n
b
o
u
n
d
ar
ies,
R
F
f
o
r
f
ea
tu
r
e
im
p
o
r
tan
ce
an
d
b
a
g
g
in
g
,
an
d
Ad
aBo
o
s
t
f
o
r
h
an
d
lin
g
d
if
f
icu
lt
-
to
-
class
if
y
in
s
tan
ce
s
.
T
h
e
m
eta
-
class
if
ier
in
th
e
f
in
al
lay
er
ca
p
tu
r
es
a
n
d
b
alan
ce
s
th
ese
b
eh
av
io
r
s
to
m
in
im
ize
o
v
er
all
class
if
icatio
n
er
r
o
r
,
p
ar
tic
u
lar
ly
f
alse n
eg
ativ
es.
2
.
2
.
I
nte
g
ra
t
io
n o
f
f
a
ls
e
neg
a
t
iv
e
re
du
ct
io
n m
et
ho
do
lo
g
i
es
A
d
is
t
in
g
u
is
h
i
n
g
i
n
n
o
v
a
ti
o
n
in
o
u
r
m
o
d
el
is
t
h
e
i
n
c
o
r
p
o
r
at
io
n
o
f
m
u
lt
ip
le
s
t
r
a
te
g
ies
e
x
p
li
cit
ly
tar
g
eti
n
g
t
h
e
r
e
d
u
cti
o
n
o
f
f
a
ls
e
n
e
g
a
ti
v
es
(
FN
)
,
a
c
r
iti
ca
l
c
o
n
ce
r
n
i
n
m
e
d
i
ca
l
d
i
ag
n
o
s
ti
cs.
F
i
r
s
t,
a
c
o
s
t
-
s
e
n
s
i
ti
v
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Ma
ch
in
e
lea
r
n
in
g
mo
d
el
fo
r
a
cc
u
r
a
te
p
r
ed
ictio
n
o
f c
o
r
o
n
a
r
y
…
(
S
a
n
t
h
o
s
h
Gu
p
ta
Do
g
i
p
a
r
th
i
)
5657
lea
r
n
i
n
g
a
p
p
r
o
a
ch
is
e
m
p
lo
y
e
d
w
h
e
r
e
th
e
m
is
cl
ass
i
f
ic
ati
o
n
co
s
t
o
f
FN
is
s
et
s
ig
n
i
f
i
ca
n
tl
y
h
i
g
h
e
r
t
h
a
n
t
h
a
t
o
f
f
als
e
p
o
s
it
iv
es,
t
h
e
r
e
b
y
in
f
l
u
e
n
ci
n
g
th
e
m
o
d
el
t
o
b
e
m
o
r
e
co
n
s
e
r
v
at
iv
e
wh
e
n
p
r
e
d
ic
ti
n
g
n
e
g
a
ti
v
e
o
u
t
co
m
es
.
Sec
o
n
d
,
m
a
n
u
al
wei
g
h
ti
n
g
is
ap
p
l
ie
d
t
o
b
ase
class
if
ie
r
s
b
ase
d
o
n
t
h
ei
r
h
is
to
r
i
ca
l
p
er
f
o
r
m
an
ce
a
n
d
cl
in
ic
al
in
s
i
g
h
t
t
h
is
em
p
h
asi
ze
s
m
o
r
e
r
el
ia
b
le
m
o
d
els
i
n
t
h
e
f
i
n
al
d
e
cisi
o
n
.
T
h
i
r
d
,
th
e
d
ec
is
i
o
n
th
r
esh
o
l
d
is
o
p
ti
m
iz
e
d
u
s
i
n
g
a
p
r
ec
is
i
o
n
-
r
e
ca
ll
t
r
a
d
e
-
o
f
f
,
s
ele
cti
n
g
a
v
al
u
e
t
h
at
m
a
x
i
m
iz
es
th
e
F
1
-
s
c
o
r
e
,
w
h
i
c
h
i
s
p
ar
tic
u
l
ar
ly
s
u
it
ed
f
o
r
i
m
b
al
an
ce
d
d
atas
ets
.
F
in
a
lly
,
d
o
m
ai
n
-
s
p
ec
if
ic
e
n
g
i
n
ee
r
ed
f
ea
tu
r
es
s
u
c
h
as
p
u
ls
e
p
r
e
s
s
u
r
e,
m
e
an
a
r
t
er
ial
p
r
ess
u
r
e,
E
C
G
a
b
n
o
r
m
a
lit
y
s
c
o
r
e,
a
n
d
c
o
m
o
r
b
id
it
y
c
o
u
n
ts
a
r
e
i
n
tr
o
d
u
c
ed
to
e
n
h
a
n
ce
m
o
d
el
in
t
er
p
r
e
ta
b
il
it
y
a
n
d
p
r
e
d
ic
ti
o
n
s
t
r
e
n
g
t
h
.
T
h
is
m
u
l
t
i
-
p
r
o
n
g
e
d
s
tr
ate
g
y
c
o
ll
ec
ti
v
el
y
r
e
d
u
c
es
F
N
i
n
s
t
a
n
ce
s
,
m
ak
i
n
g
t
h
e
m
o
d
el
m
o
r
e
tr
u
s
tw
o
r
t
h
y
a
n
d
cl
in
ica
ll
y
v
ia
b
le
.
S
im
i
la
r
FN
-
r
e
d
u
cti
o
n
s
t
r
ate
g
i
es
h
a
v
e
b
e
e
n
s
u
cc
ess
f
u
l
l
y
a
p
p
li
e
d
i
n
r
ec
e
n
t
wo
r
k
s
u
s
i
n
g
c
o
s
t
-
s
en
s
iti
v
e
e
n
s
em
b
le
m
et
h
o
d
s
an
d
t
h
r
esh
o
l
d
-
m
o
v
i
n
g
te
ch
n
i
q
u
es
[
2
2
]
–
[
2
4
]
.
3.
M
E
T
H
O
DS
T
h
is
s
ec
tio
n
o
u
tlin
es
th
e
en
d
-
to
-
en
d
m
eth
o
d
o
lo
g
y
ad
o
p
ted
f
o
r
b
u
ild
in
g
th
e
p
r
o
p
o
s
ed
ML
m
o
d
el
f
o
r
co
r
o
n
a
r
y
a
r
ter
y
d
is
ea
s
e
p
r
e
d
ic
tio
n
.
I
t
co
v
er
s
th
e
d
ataset
ch
a
r
ac
ter
is
tics
,
f
ea
tu
r
e
en
g
in
ee
r
in
g
s
tr
ateg
ies,
m
o
d
el
tr
ain
in
g
p
r
o
ce
s
s
,
an
d
ev
al
u
atio
n
m
etr
ics.
T
h
e
m
eth
o
d
o
l
o
g
ical
p
ip
elin
e
e
n
s
u
r
es
h
ig
h
r
ep
r
o
d
u
cib
ilit
y
an
d
tr
an
s
p
ar
en
cy
i
n
d
ata
p
r
o
ce
s
s
in
g
,
m
o
d
el
d
ev
elo
p
m
en
t,
a
n
d
p
e
r
f
o
r
m
a
n
ce
an
aly
s
is
.
3
.
1
.
Da
t
a
s
et
des
cr
iptio
n
A
d
ataset
was
p
r
ep
ar
ed
b
y
co
llectin
g
th
e
d
em
o
g
r
a
p
h
ical,
cl
in
ical
ass
ess
m
en
t,
E
C
G,
lab
an
d
E
C
HO
f
ea
tu
r
es
o
f
4
2
8
p
atien
ts
f
r
o
m
Dep
ar
tm
en
t
o
f
C
ar
d
io
lo
g
y
,
J
I
PME
R
,
Pu
d
u
ch
er
r
y
.
T
h
e
d
ata
s
et
h
as
3
6
d
if
f
er
en
t
f
ea
tu
r
es
with
th
e
last
f
ea
tu
r
e
in
d
icatin
g
in
b
in
ar
y
v
alu
es
ab
o
u
t
th
e
p
r
esen
ce
‘
1
’
o
r
ab
s
en
ce
‘
0
’
o
f
c
o
r
o
n
a
r
y
ar
ter
y
d
is
ea
s
e
(
C
AD)
.
Fu
ll d
etails ab
o
u
t th
e
r
em
ain
i
n
g
3
5
f
e
atu
r
es a
r
e
p
r
o
v
id
ed
in
T
ab
le
1
.
T
ab
le
1
.
Deta
ils
o
f
3
5
f
ea
tu
r
es
f
r
o
m
t
h
e
J
I
PME
R
C
AD
d
atas
et
F
e
a
t
u
r
e
N
a
me
D
a
t
a
R
a
n
g
e
Ty
p
e
F
e
a
t
u
r
e
C
a
t
e
g
o
r
y
A
g
e
(
Y
e
a
r
s)
20
–
83
N
u
meri
c
a
l
D
e
mo
g
r
a
p
h
i
c
a
l
W
e
i
g
h
t
(
k
g
)
50
–
90
N
u
meri
c
a
l
D
e
mo
g
r
a
p
h
i
c
a
l
Le
n
g
t
h
(
c
m)
1
3
0
–
180
N
u
meri
c
a
l
D
e
mo
g
r
a
p
h
i
c
a
l
G
e
n
d
e
r
1
–
M
,
0
–
F
B
i
n
a
r
y
D
e
mo
g
r
a
p
h
i
c
a
l
B
M
I
(
kg
m
2
⁄
)
1
9
.
3
7
–
3
5
.
5
N
u
meri
c
a
l
D
e
mo
g
r
a
p
h
i
c
a
l
D
i
a
b
e
t
e
s
M
e
l
l
i
t
u
s
1
–
Y
,
0
–
N
B
i
n
a
r
y
D
e
mo
g
r
a
p
h
i
c
a
l
H
y
p
e
r
t
e
n
si
o
n
1
–
Y
,
0
–
N
B
i
n
a
r
y
D
e
mo
g
r
a
p
h
i
c
a
l
C
u
r
r
e
n
t
S
mo
k
e
r
1
–
Y
,
0
–
N
B
i
n
a
r
y
D
e
mo
g
r
a
p
h
i
c
a
l
Ex
-
sm
o
k
e
r
1
–
Y
,
0
–
N
B
i
n
a
r
y
D
e
mo
g
r
a
p
h
i
c
a
l
D
y
sl
i
p
i
d
e
mi
a
1
–
Y
,
0
–
N
B
i
n
a
r
y
D
e
mo
g
r
a
p
h
i
c
a
l
S
y
st
o
l
i
c
B
P
(
mm
H
g
)
90
–
1
8
9
N
u
meri
c
a
l
C
l
i
n
i
c
a
l
A
ss
e
ssm
e
n
t
D
i
a
st
o
l
i
c
B
P
(
mm
H
g
)
52
–
1
1
2
N
u
meri
c
a
l
C
l
i
n
i
c
a
l
A
ss
e
ssm
e
n
t
P
u
l
s
e
r
a
t
e
(
/
mi
n
)
44
–
1
4
0
N
u
meri
c
a
l
C
l
i
n
i
c
a
l
A
ss
e
ssm
e
n
t
A
n
g
i
n
a
1
–
Y
,
0
–
N
B
i
n
a
r
y
C
l
i
n
i
c
a
l
A
ss
e
ssm
e
n
t
D
y
sp
n
e
a
1
–
Y
,
0
–
N
B
i
n
a
r
y
C
l
i
n
i
c
a
l
A
ss
e
ssm
e
n
t
R
h
y
t
h
m
0
–
S
i
n
u
s,
1
–
F
i
b
r
i
l
l
a
t
i
o
n
B
i
n
a
r
y
EC
G
Q
-
W
a
v
e
1
–
Y
,
0
–
N
B
i
n
a
r
y
EC
G
Q
S
W
a
v
e
1
–
Y
,
0
–
N
B
i
n
a
r
y
EC
G
Q
R
S
C
o
m
p
l
e
x
1
–
Y
,
0
–
N
B
i
n
a
r
y
EC
G
A
x
i
s D
e
v
i
a
t
i
o
n
1
–
Y
,
0
–
N
B
i
n
a
r
y
EC
G
ST
-
T
c
h
a
n
g
e
s
1
–
Y
,
0
–
N
B
i
n
a
r
y
EC
G
T
-
W
a
v
e
1
–
Y
,
0
–
N
B
i
n
a
r
y
EC
G
S
T
e
l
e
v
a
t
i
o
n
1
–
Y
,
0
–
N
B
i
n
a
r
y
EC
G
S
T
d
e
p
r
e
ssi
o
n
1
–
Y
,
0
–
N
B
i
n
a
r
y
EC
G
T
i
n
v
e
r
si
o
n
1
–
Y
,
0
–
N
B
i
n
a
r
y
EC
G
Le
f
t
v
e
n
t
r
i
c
u
l
a
r
h
y
p
e
r
t
r
o
p
h
y
1
–
Y
,
0
–
N
B
i
n
a
r
y
EC
G
P
o
o
r
R
W
a
v
e
P
r
o
g
r
e
ss
i
o
n
1
–
Y
,
0
–
N
B
i
n
a
r
y
EC
G
B
u
n
d
l
e
B
r
a
n
c
h
B
l
o
c
k
0
–
A
b
se
n
c
e
,
1
–
L
B
B
B
/
R
B
B
B
N
o
mi
n
a
l
EC
G
R
B
S
(
m
g
/
d
l
)
60
–
7
3
3
N
u
meri
c
a
l
La
b
a
n
d
E
C
H
O
C
r
e
a
t
i
n
i
n
e
(
mg
/
d
l
)
0
.
3
–
2
.
7
N
u
meri
c
a
l
La
b
a
n
d
E
C
H
O
B
l
o
o
d
U
r
e
a
(
m
g
/
d
l
)
9
.
3
–
70
N
u
meri
c
a
l
La
b
a
n
d
E
C
H
O
H
a
e
mo
g
l
o
b
i
n
(
g
m/
d
l
)
7
.
1
–
22
N
u
meri
c
a
l
La
b
a
n
d
E
C
H
O
P
l
a
t
e
l
e
t
C
o
u
n
t
(
1
0
0
0
/
ml
)
90
–
5
8
6
N
u
meri
c
a
l
La
b
a
n
d
E
C
H
O
Ej
e
c
t
i
o
n
F
r
a
c
t
i
o
n
(
%)
15
–
66
N
u
meri
c
a
l
La
b
a
n
d
E
C
H
O
R
e
g
i
o
n
a
l
w
a
l
l
m
o
t
i
o
n
a
b
n
o
r
ma
l
i
t
y
0
–
N
o
r
ma
l
,
1
–
A
b
n
o
r
mal
B
i
n
a
r
y
La
b
a
n
d
E
C
H
O
3
.
2
.
Da
t
a
prepro
ce
s
s
ing
a
nd
f
ea
t
ure
eng
ineering
Prio
r
to
m
o
d
el
t
r
ain
in
g
,
th
e
r
aw
clin
ical
d
ataset
o
b
tain
e
d
f
r
o
m
J
I
PME
R
u
n
d
e
r
wen
t
s
y
s
tem
atic
p
r
ep
r
o
ce
s
s
in
g
.
T
h
is
in
clu
d
ed
h
an
d
lin
g
m
is
s
in
g
v
alu
es,
o
u
tlier
tr
ea
tm
en
t,
an
d
n
o
r
m
aliz
atio
n
o
f
co
n
tin
u
o
u
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
6
,
Decem
b
e
r
20
25
:
5
6
5
5
-
5
6
6
6
5658
v
ar
iab
les
s
u
ch
as
ag
e,
s
y
s
to
lic/d
iast
o
lic
b
lo
o
d
p
r
ess
u
r
e,
an
d
ejec
tio
n
f
r
ac
tio
n
.
C
ateg
o
r
ic
al
v
ar
iab
les
s
u
ch
as
g
en
d
er
,
s
m
o
k
in
g
s
tatu
s
,
an
d
E
C
G
in
d
icato
r
s
wer
e
en
co
d
e
d
in
to
n
u
m
er
ical
r
ep
r
esen
tati
o
n
s
u
s
in
g
o
n
e
-
h
o
t
o
r
o
r
d
in
al
e
n
co
d
in
g
s
ch
em
es,
a
s
ap
p
r
o
p
r
iate.
Featu
r
e
e
n
g
in
ee
r
in
g
was
co
n
d
u
cted
t
o
en
h
an
ce
th
e
m
o
d
el'
s
d
is
cr
im
in
ativ
e
p
o
wer
.
T
en
d
er
iv
ed
f
ea
tu
r
es
wer
e
ad
d
ed
,
in
cl
u
d
in
g
clin
ically
s
ig
n
if
ican
t
co
n
s
tr
u
cts
lik
e
p
u
ls
e
p
r
ess
u
r
e
(
s
y
s
to
lic
m
in
u
s
d
iast
o
lic
B
P),
m
ea
n
ar
te
r
ial
p
r
ess
u
r
e
(
MA
P),
co
m
o
r
b
i
d
ity
co
u
n
t,
E
C
G
ab
n
o
r
m
ality
s
co
r
e,
an
d
r
atio
s
s
u
ch
as
R
B
S
to
B
MI
.
Ad
d
itio
n
ally
,
ca
teg
o
r
ical
r
ec
o
d
in
g
f
o
r
v
ar
iab
les
s
u
ch
as
h
ea
r
t
r
ate
an
d
ejec
tio
n
f
r
ac
tio
n
was
in
tr
o
d
u
c
ed
to
m
a
p
th
ese
in
to
clin
ically
in
ter
p
r
etab
le
ca
teg
o
r
ies.
T
h
e
co
m
b
in
e
d
u
s
e
o
f
r
aw
an
d
en
g
in
ee
r
e
d
f
ea
tu
r
es
y
ield
ed
a
to
tal
o
f
4
6
p
r
ed
icto
r
s
,
p
r
o
v
id
in
g
a
r
ich
er
an
d
m
o
r
e
r
o
b
u
s
t
in
p
u
t
s
p
ac
e
f
o
r
tr
ain
in
g
th
e
e
n
s
em
b
le
m
o
d
el.
3
.
3
.
F
N
re
du
ct
io
n
m
et
ho
do
l
o
g
ies
I
n
d
ev
elo
p
in
g
a
m
ed
ical
d
ia
g
n
o
s
tic
to
o
l
to
p
er
f
o
r
m
b
in
ar
y
class
if
icatio
n
as,
p
atien
t
w
i
th
d
is
ea
s
e
(
p
o
s
itiv
e
class
)
o
r
with
o
u
t
d
is
ea
s
e
(
n
eg
ativ
e
class
)
,
th
er
e
ar
e
two
p
o
s
s
ib
le
m
is
clas
s
if
icati
o
n
er
r
o
r
s
n
am
ely
FP
an
d
FN a
r
e
av
ailab
le.
I
n
th
is
s
ce
n
ar
io
:
−
A
FN
is
wh
en
th
e
ML
to
o
l
in
co
r
r
ec
tly
class
if
ies
a
d
is
ea
s
ed
p
atien
t
as
h
ea
lth
y
,
wh
i
ch
ca
n
b
e
v
er
y
d
etr
im
en
tal
in
ter
m
s
o
f
p
atien
t
h
ea
lth
.
−
A
FP
is
wh
en
th
e
ML
to
o
l
in
co
r
r
ec
tly
class
if
ies
a
h
ea
lth
y
p
atien
t
as
d
is
ea
s
ed
,
wh
i
ch
ca
n
lead
to
u
n
n
ec
ess
ar
y
s
tr
ess
an
d
f
u
r
th
er
ex
p
en
s
iv
e
test
s
.
I
n
th
e
f
o
llo
win
g
s
u
b
s
ec
tio
n
s
,
d
etailed
in
f
o
r
m
atio
n
ab
o
u
t
m
o
d
if
icatio
n
s
to
r
e
d
u
ce
th
e
FN
an
d
p
ip
elin
e
ar
ch
itectu
r
e
o
f
t
h
e
tu
n
ed
ML
m
o
d
el
ar
e
ex
p
lain
ed
.
3
.
3
.
1
.
Co
s
t
-
s
ens
it
iv
e
lea
rning
Mo
s
t
o
f
th
e
ML
m
eth
o
d
o
lo
g
ies
ass
u
m
es
th
e
m
is
class
if
i
ca
tio
n
er
r
o
r
s
ar
e
j
u
s
tifie
d
as
th
ey
a
r
e
in
h
er
en
tly
p
r
o
v
id
ed
b
y
th
e
m
o
d
el
its
elf
.
R
ath
er
th
an
s
u
ch
a
j
u
s
tific
atio
n
,
th
e
in
co
r
r
ec
t
p
r
e
d
ictio
n
s
o
f
FP
o
r
FN
s
h
o
u
ld
b
e
s
ee
n
as
a
q
u
esti
o
n
o
n
th
e
r
eliab
ilit
y
o
f
th
e
ML
m
o
d
el
p
r
e
d
ictio
n
.
I
n
co
r
r
ec
t
p
r
ed
ictio
n
s
o
f
FN
an
d
FP
co
m
p
ar
ed
to
th
e
T
P
an
d
T
N
lead
s
to
r
ed
u
ctio
n
in
th
e
r
e
ca
ll
an
d
h
en
ce
lo
wer
ac
cu
r
ac
y
lev
el
f
o
r
th
e
f
in
al
p
r
ed
icted
v
alu
es
o
f
th
e
d
if
f
e
r
en
t
f
ea
tu
r
es
b
y
th
e
ML
m
o
d
el.
T
o
av
o
id
s
u
ch
e
r
r
o
r
s
in
p
r
ed
ictio
n
b
y
th
e
en
s
em
b
le
m
o
d
el,
a
m
is
class
if
i
ca
tio
n
co
s
t
m
at
r
ix
is
i
n
tr
o
d
u
ce
d
to
tr
ain
all
b
ase
class
if
ier
s
t
o
b
e
f
o
llo
we
d
b
y
a
m
eta
class
if
ier
to
r
e
d
u
ce
th
e
wr
o
n
g
p
r
e
d
ictio
n
s
.
T
o
r
e
f
lect
t
h
e
s
ev
er
ity
o
f
er
r
o
r
s
,
a
co
s
t
m
atr
ix
is
d
ef
i
n
ed
s
u
ch
th
at
co
s
t
o
f
a
FN
is
h
ig
h
e
r
th
a
n
a
FP
.
T
h
u
s
,
ca
lcu
lated
elem
en
ts
f
o
r
th
e
c
o
s
t
m
atr
ix
ar
e
n
o
t
to
b
e
s
im
p
ly
u
s
ed
as
a
m
u
ltip
licatio
n
f
ac
to
r
,
r
a
th
er
to
b
e
u
s
ed
as
an
i
n
f
lu
e
n
cin
g
f
ac
t
o
r
o
n
h
o
w
th
e
cla
s
s
if
ier
ev
alu
ates
its
d
ec
is
io
n
.
I
n
th
e
f
o
llo
win
g
we
p
r
o
v
id
e
t
h
e
d
etails
o
n
h
o
w
th
e
co
s
t
m
atr
ix
ca
n
b
e
u
s
ed
to
a
d
ju
s
t
th
e
p
r
ed
icte
d
v
alu
es b
y
th
e
ML
m
o
d
el.
Ver
y
co
m
m
o
n
m
eth
o
d
o
l
o
g
y
i
s
to
m
o
d
if
y
t
h
e
d
ec
is
io
n
th
r
es
h
o
ld
o
f
th
e
class
if
ier
u
s
in
g
th
e
elem
en
ts
ca
lcu
lated
f
o
r
th
e
co
n
s
tr
u
ctio
n
o
f
th
e
co
s
t
m
atr
ix
.
L
et
u
s
d
en
o
te
th
e
p
r
e
d
icted
p
r
o
b
ab
il
ities
o
f
th
e
p
o
s
itiv
e
class
as
p
(
o
u
tp
u
t
o
f
t
h
e
class
if
ier
)
,
an
d
ass
u
m
e
th
e
d
ef
a
u
lt
th
r
esh
o
ld
v
alu
e
f
o
r
class
if
y
in
g
a
p
o
s
itiv
e
in
s
tan
ce
as
0
.
5
.
I
f
t
h
e
co
s
t
o
f
a
FN
(
C
FN
)
is
h
ig
h
er
th
a
n
th
e
co
s
t
o
f
a
FP
(
C
FP
)
,
th
er
e
is
a
r
eq
u
ir
em
e
n
t
to
d
ec
r
ea
s
e
th
e
th
r
esh
o
ld
v
alu
e
to
r
ed
u
ce
th
e
FN.
On
th
e
o
th
er
h
an
d
,
if
th
e
C
FP
is
g
r
ea
ter
th
an
th
e
C
FN
,
th
er
e
is
a
r
eq
u
ir
em
e
n
t
to
in
cr
ea
s
e
th
e
th
r
esh
o
ld
v
alu
e
to
r
ed
u
ce
th
e
FP
.
T
h
e
a
d
ju
s
ted
th
r
esh
o
ld
(
T
ad
ju
s
t
ed
)
ca
n
b
e
ca
lcu
late
d
as:
=
+
,
an
d
th
e
c
o
r
r
esp
o
n
d
in
g
p
r
e
d
ictio
n
is
ad
ju
s
ted
as f
o
llo
ws:
If
p
≥
T
a
djus
ted
,
class
if
y
as p
o
s
itiv
e,
If
p
<
T
a
djus
ted
,
class
if
y
as n
eg
ativ
e.
T
h
ese
ad
ju
s
tm
en
ts
b
alan
ce
th
e
co
s
ts
a
s
s
o
ciate
d
with
th
e
FP
an
d
FN
as
p
er
th
e
s
p
ec
if
ic
c
o
s
t
m
atr
ix
.
Als
o
,
it
en
s
u
r
es
th
at
th
e
m
o
d
e
l’
s
p
r
ed
ictio
n
s
alig
n
with
th
e
s
p
ec
if
ic
co
s
t
co
n
s
id
er
atio
n
s
,
p
o
ten
tially
lead
in
g
to
b
etter
o
u
tco
m
es
in
r
ea
l
-
wo
r
ld
s
ce
n
ar
io
s
wh
er
e
d
if
f
er
en
t
ty
p
e
s
o
f
er
r
o
r
s
,
n
am
ely
,
ty
p
e
-
I
an
d
ty
p
e
-
I
I
er
r
o
r
s
[
1
4
]
h
av
e
d
if
f
e
r
en
t
co
n
s
eq
u
en
ce
s
.
I
n
th
is
r
esear
ch
s
tu
d
y
,
th
e
m
is
c
lass
if
icatio
n
co
s
t
m
atr
ix
u
s
ed
i
s
[
0
,
17
,
20
,
0
]
f
o
r
[
TP
,
FP
,
FN
,
TN
]
wh
er
e
all
th
e
b
ase
class
if
ier
s
an
d
m
eta
class
if
ier
ar
e
tr
ain
ed
an
d
test
ed
with
th
eir
r
esp
ec
tiv
e
d
atasets
an
d
co
s
t m
atr
ix
.
3
.
3
.
2
.
M
a
nu
a
l
weig
ht
a
djustm
ent
Ma
n
u
al
weig
h
t
ad
ju
s
tm
en
t
in
v
o
lv
es
th
e
m
o
d
if
icatio
n
o
f
th
e
p
r
ed
ictio
n
weig
h
tin
g
s
in
th
e
in
d
iv
id
u
al
b
ase
m
o
d
el’
s
en
s
em
b
le
to
im
p
r
o
v
e
th
e
o
v
e
r
all
p
er
f
o
r
m
a
n
ce
.
T
h
is
m
eth
o
d
o
lo
g
y
lev
er
ag
es
d
o
m
ain
ex
p
er
tis
e
to
p
r
o
v
id
e
p
r
o
p
er
s
ig
n
if
ica
n
ce
to
m
o
d
els th
at
ar
e
ex
p
ec
ted
to
p
e
r
f
o
r
m
b
etter
in
ce
r
tai
n
asp
ec
ts
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Ma
ch
in
e
lea
r
n
in
g
mo
d
el
fo
r
a
cc
u
r
a
te
p
r
ed
ictio
n
o
f c
o
r
o
n
a
r
y
…
(
S
a
n
t
h
o
s
h
Gu
p
ta
Do
g
i
p
a
r
th
i
)
5659
C
o
n
s
id
er
an
en
s
em
b
le
o
f
N
=
5
b
ase
m
o
d
els.
L
et
m
atr
ix
X
(
4
2
8
×
3
5
)
b
e
th
e
in
p
u
t
f
ea
tu
r
e
m
atr
ix
,
an
d
y
(
4
2
8
×
1
)
b
e
th
e
ta
r
g
et
v
ec
to
r
.
T
h
e
p
r
ed
ictio
n
f
r
o
m
th
e
i
th
m
o
d
e
l
f
o
r
a
g
iv
e
n
in
p
u
t
x
(
elem
en
ts
o
f
m
atr
ix
X
)
is
d
en
o
ted
as
̂
(
)
.
T
h
e
en
s
em
b
le
p
r
e
d
ictio
n
is
t
y
p
ically
a
weig
h
ted
co
m
b
in
atio
n
o
f
t
h
ese
in
d
iv
id
u
al
p
r
ed
ictio
n
s
:
̂
e
ns
e
mble
(
)
=
∑
=
1
̂
(
)
,
wh
er
e
w
i
ar
e
th
e
weig
h
tin
g
s
ass
ig
n
ed
to
t
h
e
p
r
ed
ictio
n
s
f
r
o
m
e
ac
h
b
ase
m
o
d
el.
I
n
m
a
n
u
al
we
ig
h
t
ad
ju
s
tm
en
t,
th
ese
weig
h
tin
g
s
w
i
ar
e
s
et
b
ased
o
n
th
e
f
ea
tu
r
e
’
s
d
o
m
ain
ex
p
er
tis
e.
Fo
r
o
u
r
J
I
PME
R
d
ataset,
th
e
m
an
u
al
weig
h
tin
g
s
u
s
ed
ar
e
[
0
.
5
,
0
.
8
,
0
.
7
,
0
.
9
,
0
.
8
5
]
an
d
th
e
f
in
al
m
o
d
el
s
co
r
es
ar
e
o
b
tain
ed
b
y
m
u
ltip
ly
in
g
with
weig
h
tin
g
s
an
d
p
r
ed
ictio
n
s
co
r
es
o
f
r
esp
ec
tiv
e
tr
ain
an
d
test
d
atasets
.
3
.
3
.
3
.
T
hresh
o
ld
a
dju
s
t
m
ent
Ad
ju
s
tin
g
th
e
d
ec
is
io
n
th
r
esh
o
ld
o
f
a
class
if
ier
ca
n
b
r
in
g
a
tr
ad
e
-
o
f
f
b
etwe
en
th
e
p
r
e
cisi
o
n
an
d
r
ec
all.
T
h
e
o
p
tim
al
th
r
esh
o
ld
ca
n
b
e
d
eter
m
i
n
ed
b
y
m
ax
im
i
zin
g
th
e
F1
-
Sco
r
e
a
n
d
th
e
n
e
ce
s
s
ar
y
s
tep
s
to
b
e
f
o
llo
wed
ar
e:
−
Pre
d
ictio
n
s
co
r
es
:
f
o
r
ea
c
h
test
in
s
tan
ce
,
o
b
tain
th
e
p
r
e
d
icted
p
r
o
b
a
b
ilit
y
s
co
r
es
p
̂
(
x
)
f
r
o
m
th
e
m
o
d
el.
−
T
h
r
esh
o
ld
s
:
d
ef
i
n
e
a
r
an
g
e
f
o
r
p
o
s
s
ib
le
th
r
esh
o
ld
s
θ
to
ev
al
u
ate.
Fo
r
ea
c
h
th
r
esh
o
ld
θ
,
cla
s
s
if
y
th
e
d
ata
p
o
in
ts
as:
̂
(
)
=
{
1
if
̂
(
)
≥
0
if
̂
(
)
<
−
C
o
m
p
u
te
m
etr
ics
:
f
o
r
ea
c
h
th
r
esh
o
ld
θ
,
co
m
p
u
te
p
r
ec
is
io
n
an
d
r
ec
all.
−
F1
-
Sco
r
e
ca
lcu
latio
n
:
ca
lcu
lat
e
th
e
F1
-
Sco
r
e
f
o
r
ea
c
h
th
r
esh
o
ld
.
−
Op
tim
al
th
r
esh
o
ld
:
s
elec
t
th
e
t
h
r
esh
o
ld
θ
∗
th
at
m
ax
im
izes th
e
F1
-
Sco
r
e:
∗
=
a
r
g
F1
-
S
co
r
e
(
)
.
3
.
3
.
4
.
E
ng
ineere
d/
deriv
ed
f
e
a
t
ures
Der
iv
ed
f
ea
tu
r
es,
also
k
n
o
wn
as
en
g
in
ee
r
ed
f
e
atu
r
es,
ar
e
n
ew
v
ar
iab
les
cr
ea
ted
b
y
t
r
an
s
f
o
r
m
in
g
o
r
co
m
b
in
in
g
ex
is
tin
g
f
ea
tu
r
es.
T
h
ese
f
ea
tu
r
es c
an
ca
p
tu
r
e
m
o
r
e
en
tan
g
led
r
elatio
n
s
h
ip
s
in
ter
co
n
n
ec
tin
g
th
e
d
ata
th
at
r
aw
f
ea
tu
r
es
m
ig
h
t
n
o
t
r
e
v
ea
l.
I
n
th
e
c
o
n
tex
t
o
f
m
ed
ica
l
d
ata,
d
e
r
iv
ed
f
ea
tu
r
es
ca
n
b
e
p
ar
ticu
lar
ly
u
s
ef
u
l
in
en
h
an
ci
n
g
th
e
p
r
ed
ictiv
e
c
ap
ac
ity
an
d
ca
p
ab
ilit
y
o
f
th
e
m
o
d
el
b
y
in
co
r
p
o
r
atin
g
d
o
m
a
in
k
n
o
wled
g
e
an
d
s
p
ec
if
ic
m
ed
ical
in
s
ig
h
ts
.
I
ts
m
ain
p
u
r
p
o
s
e
is
:
−
B
y
ca
p
tu
r
in
g
a
d
d
itio
n
al
in
f
o
r
m
atio
n
th
at
r
aw
f
ea
t
u
r
es
alo
n
e
m
ig
h
t
n
o
t
p
r
o
v
id
e,
d
er
iv
ed
f
ea
tu
r
es
ca
n
h
elp
to
im
p
r
o
v
e
th
e
ac
cu
r
ac
y
,
p
r
ec
i
s
io
n
,
r
ec
all,
an
d
o
v
er
all
p
er
f
o
r
m
an
ce
o
f
t
h
e
ML
m
o
d
els.
−
Der
iv
ed
f
ea
tu
r
es
ca
n
o
f
ten
m
ak
e
th
e
m
o
d
els
m
o
r
e
i
n
ter
p
r
etab
le
b
y
b
r
in
g
i
n
g
o
u
t
th
e
im
p
o
r
tan
t
r
elatio
n
s
h
ip
s
an
d
p
atter
n
s
in
t
h
e
d
ata
th
at
ar
e
m
ea
n
in
g
f
u
l
to
th
e
n
atu
r
e
o
f
th
e
d
ataset
(
m
ed
ical
in
o
u
r
ca
s
e
s
tu
d
y
)
.
−
Pro
p
er
ly
e
n
g
in
ee
r
ed
f
ea
tu
r
es
ca
n
h
elp
to
r
ed
u
ce
th
e
in
f
lu
en
ce
o
f
th
e
n
o
is
y
d
ata
in
t
h
e
p
r
e
d
ictio
n
p
r
o
ce
s
s
b
y
f
o
c
u
s
in
g
m
o
r
e
o
n
th
e
r
ele
v
an
t a
s
p
ec
ts
o
f
th
e
d
ata.
Der
iv
ed
f
ea
tu
r
es
ca
n
b
e
o
f
d
if
f
er
en
t
k
in
d
s
s
u
ch
as,
s
tatis
tical,
tem
p
o
r
al,
tr
an
s
f
o
r
m
atio
n
,
etc.
Dep
en
d
in
g
o
n
t
h
e
p
r
e
d
ictio
n
r
eq
u
ir
em
en
ts
an
y
p
ar
ticu
lar
f
ea
tu
r
e
k
in
d
ca
n
b
e
ad
o
p
ted
.
Nev
er
th
eless
,
in
clu
s
io
n
o
f
m
o
r
e
f
ea
tu
r
e(
s
)
to
th
e
e
x
is
tin
g
d
ataset
f
ea
tu
r
es will in
tr
o
d
u
ce
d
if
f
er
e
n
t c
o
n
s
tr
ain
ts
to
th
e
ML
m
o
d
el
lik
e
th
e
o
v
er
f
itti
n
g
,
r
elev
an
ce
an
d
c
o
m
p
lex
ity
.
A
to
tal
o
f
1
0
d
er
i
v
ed
f
ea
t
u
r
e
s
ar
e
o
b
tain
ed
f
o
r
th
is
wo
r
k
,
wh
er
e
th
eir
n
a
m
es,
r
elatio
n
s
with
o
th
er
f
ea
tu
r
es a
n
d
th
ei
r
s
ig
n
if
ican
ce
ar
e
lis
ted
b
elo
w
.
−
Pu
ls
e
p
r
ess
u
r
e
: I
t is d
ef
in
ed
as
:
P
u
ls
e
P
r
ess
u
r
e
= S
ysto
lic
B
P
–
Dia
s
tol
ic
BP.
T
h
is
d
er
iv
ed
f
ea
tu
r
e
p
r
o
v
i
d
es in
s
ig
h
ts
in
to
ca
r
d
io
v
ascu
lar
h
e
alth
an
d
ar
ter
ial
s
tiff
n
ess
.
−
Me
an
ar
ter
ial
p
r
ess
u
r
e
(
MA
P
):
I
t
r
ep
r
esen
ts
th
e
av
e
r
ag
e
p
r
ess
u
r
e
in
a
p
atien
t’
s
ar
ter
ies
d
u
r
in
g
a
ca
r
d
iac
cy
cle.
I
t is a
u
s
ef
u
l in
d
icato
r
o
f
p
er
f
u
s
io
n
p
r
ess
u
r
e
o
f
th
e
o
r
g
an
s
.
I
t is d
ef
in
ed
as:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
6
,
Decem
b
e
r
20
25
:
5
6
5
5
-
5
6
6
6
5660
MAP
= (
S
ys
tol
ic
BP
+
2
×
Dia
s
to
lic
B
P
)
/3
.
−
C
o
m
o
r
b
id
ity
co
u
n
t
:
T
h
is
d
er
iv
ed
f
ea
t
u
r
e
c
o
u
n
ts
t
h
e
p
atien
t’
s
n
u
m
b
er
o
f
co
m
o
r
b
id
c
o
n
d
itio
n
s
.
T
h
e
p
r
esen
ce
o
f
m
u
ltip
le
co
m
o
r
b
id
ities
[
s
u
ch
as
d
iab
etes
m
ellitu
s
(
DM
)
,
h
y
p
er
ten
s
io
n
(
HT
N)
,
an
d
d
y
s
lip
id
em
ia
(
DL
P)]
ca
n
s
ig
n
if
ican
tly
im
p
ac
t
th
e
p
atien
t’
s
o
v
er
all
h
ea
lth
a
n
d
r
is
k
p
r
o
f
ile
.
I
t
is
u
s
ef
u
l
t
o
ass
es
s
th
e
b
u
r
d
en
o
f
c
h
r
o
n
ic
d
is
ea
s
es
o
n
a
p
atien
t
an
d
h
el
p
s
in
s
tr
atif
y
in
g
r
is
k
a
n
d
tail
o
r
in
g
tr
ea
tm
en
t
p
lan
s
.
I
t is d
ef
in
ed
as:
C
o
mo
r
b
id
ity
C
o
u
n
t =
DM +
HTN
+ D
LP
.
−
Hea
r
t
r
ate
ca
te
g
o
r
y
:
I
t
ca
teg
o
r
izes
th
e
p
atien
t’
s
h
ea
r
t
r
ate
in
t
o
th
r
ee
lev
els:
L
o
w,
No
r
m
al,
a
n
d
Hig
h
.
Hea
r
t
r
ate
is
a
c
r
itical
v
ital
s
ig
n
th
a
t
ca
n
i
n
d
icate
u
n
d
er
l
y
in
g
co
n
d
itio
n
s
s
u
ch
as
b
r
a
d
y
ca
r
d
ia
(
l
o
w
h
ea
r
t
r
ate)
,
tach
y
ca
r
d
ia
(
h
ig
h
h
ea
r
t
r
ate)
,
an
d
n
o
r
m
al
h
ea
r
t
f
u
n
ctio
n
in
g
.
T
h
is
d
e
r
iv
ed
f
ea
tu
r
e
h
elp
s
to
s
wif
tly
id
en
tify
an
y
ab
n
o
r
m
alities
in
th
e
h
ea
r
t
r
ate
th
at
r
eq
u
ir
e
im
m
ed
iate
m
e
d
ical
atten
tio
n
.
I
t is co
m
p
u
ted
as:
Hea
r
t_
R
a
te_
C
a
teg
o
r
y
=
d
is
cret
iz
e
(
d
a
ta
.
P
R
,
[
0
,
6
0
,
1
0
0
,
I
n
f]
,
‘
ca
teg
o
r
ica
l’
,
‘
Lo
w
’
,
‘
N
o
r
ma
l’
,
‘
Hig
h
’
)
.
−
E
C
G
ab
n
o
r
m
alities
co
u
n
t
:
T
h
is
d
er
iv
ed
f
ea
tu
r
e
ad
d
s
u
p
v
a
r
io
u
s
ab
n
o
r
m
alities
f
o
u
n
d
in
a
n
E
C
G
r
ea
d
in
g
.
E
ac
h
o
f
th
ese
c
o
m
p
o
n
en
ts
(
Q
W
,
QS,
QR
S
C
)
[
2
2
]
,
[
2
3
]
r
ep
r
esen
ts
d
if
f
er
en
t
ty
p
es
o
f
E
C
G
ab
n
o
r
m
alities
th
at
ca
n
in
d
icate
v
ar
io
u
s
h
ea
r
t
co
n
d
itio
n
s
.
Su
m
o
f
th
ese
c
o
m
p
o
n
e
n
ts
b
ec
o
m
es
y
et
o
t
h
er
d
er
iv
ed
f
ea
tu
r
e
th
at
p
r
o
v
id
es
a
co
n
s
o
lid
ated
m
ea
s
u
r
e
o
f
th
e
o
v
er
all
E
C
G
ab
n
o
r
m
ality
b
u
r
d
en
,
wh
ich
c
an
b
e
u
s
ef
u
l
to
p
r
ed
ict
ca
r
d
iac
e
v
en
ts
an
d
th
e
s
ev
er
ity
o
f
th
e
h
ea
r
t d
is
ea
s
e.
I
t
is
ca
lcu
lated
as:
E
C
G_
A
b
n
o
r
ma
liti
es_
C
o
u
n
t =
QW+QS
+QR
S
C
+S
TT
+
TW
+
S
TE+S
TD+T
I
+LVH+P
R
W
+B
B
B
.
−
R
an
d
o
m
b
lo
o
d
s
u
g
a
r
to
B
MI
r
atio
:
I
t
m
ea
s
u
r
es
th
e
r
elatio
n
s
h
ip
b
etwe
en
r
an
d
o
m
b
lo
o
d
s
u
g
ar
(
R
B
S)
lev
els
an
d
b
o
d
y
m
ass
in
d
e
x
(
B
MI
)
.
I
t
h
el
p
s
in
u
n
d
er
s
tan
d
i
n
g
h
o
w
b
lo
o
d
g
lu
c
o
s
e
lev
els
ar
e
af
f
ec
ted
b
y
b
o
d
y
weig
h
t.
T
h
is
r
atio
ca
n
b
e
p
ar
ticu
lar
ly
u
s
ef
u
l
in
m
an
ag
i
n
g
d
iab
etes
an
d
o
b
esit
y
-
r
ela
ted
co
n
d
itio
n
s
,
p
r
o
v
id
i
n
g
in
s
ig
h
ts
in
to
t
h
e
m
et
ab
o
lic
s
tatu
s
o
f
th
e
p
atien
t.
T
h
is
r
atio
is
ca
lcu
lated
as:
R
B
S
_
B
MI_
R
a
tio
=
RBS
B
MI
⁄
.
−
L
ef
t
v
en
tr
icu
la
r
ejec
tio
n
f
r
ac
ti
o
n
(
L
VE
F)
C
ateg
o
r
y
:
T
h
is
d
e
r
iv
ed
f
ea
t
u
r
e
q
u
an
tifie
s
th
e
a
m
o
u
n
t
o
f
b
lo
o
d
p
u
m
p
e
d
b
y
t
h
e
lef
t
v
en
t
r
icle
in
ea
ch
h
ea
r
t
co
n
tr
ac
tio
n
.
C
ateg
o
r
izin
g
L
VE
F
in
t
o
L
o
w,
No
r
m
al,
an
d
Hig
h
h
elp
s
in
ass
ess
in
g
th
e
f
u
n
ctio
n
al
s
tatu
s
o
f
th
e
h
ea
r
t.
A
lo
w
L
VE
F
in
d
icate
s
h
ea
r
t
f
ailu
r
e
o
r
ca
r
d
io
m
y
o
p
ath
y
,
wh
ile
n
o
r
m
al
an
d
h
ig
h
ca
teg
o
r
ies
ar
e
in
d
icatio
n
s
o
f
g
o
o
d
h
ea
r
t
f
u
n
ctio
n
in
g
.
T
h
is
ca
teg
o
r
izatio
n
is
cr
u
cial
f
o
r
d
iag
n
o
s
in
g
a
n
d
m
o
n
ito
r
in
g
h
ea
r
t
co
n
d
itio
n
s
.
L
VE
F c
ateg
o
r
y
is
co
m
p
u
ted
as:
LVE
F
_
C
a
teg
o
r
y
=
d
is
cr
etiz
e(
d
a
ta
.
LVE
F
,
[
0
,
4
0
,
5
5
,
I
n
f]
,
‘
ca
teg
o
r
ica
l’
,
‘
Lo
w
’
,
‘
N
o
r
ma
l’
,
‘
Hig
h
’
)
.
−
An
g
in
a
r
elativ
e
r
i
s
k
:
R
elativ
e
r
is
k
v
alu
es
ar
e
b
ased
o
n
th
e
g
en
d
er
an
d
ag
e
f
o
r
p
atien
ts
with
ty
p
ical
an
g
in
a
(
ANG)
[
2
5
]
.
T
h
ese
v
alu
es
ar
e
m
an
u
ally
ass
ig
n
ed
b
ased
o
n
p
r
ed
ef
in
e
d
r
is
k
ca
teg
o
r
ies
as
s
p
ec
if
ied
b
y
th
e
ca
r
d
io
lo
g
is
ts
.
T
h
e
p
r
o
b
a
b
ilit
y
v
alu
es a
s
s
ig
n
ed
f
o
r
C
AD
as p
er
th
e
ca
teg
o
r
y
d
e
p
icted
in
T
a
b
le
2
.
T
ab
le
2
.
Pro
b
ab
ilit
y
o
f
C
AD
b
ased
o
n
ag
e
a
n
d
g
en
d
e
r
A
g
e
Ty
p
i
c
a
l
A
n
g
i
n
a
M
e
n
W
o
me
n
30
-
39
0
.
7
6
0
.
2
6
40
-
49
0
.
8
7
0
.
5
5
50
-
59
0
.
9
3
0
.
7
3
60
-
69
0
.
9
4
0
.
8
6
−
Diab
etes
m
ellitu
s
r
elativ
e
r
is
k
:
r
elativ
e
r
is
k
f
o
r
p
atien
ts
with
d
iab
etes
m
ellitu
s
(
DM
)
h
av
e
t
wo
-
to
f
o
u
r
-
f
o
ld
p
o
s
s
ib
ilit
y
o
f
d
ev
elo
p
in
g
co
r
o
n
ar
y
d
is
ea
s
e
[
2
6
]
.
So
,
p
r
esen
c
e
o
f
DM
is
m
an
u
ally
s
et
to
0
.
7
an
d
ab
s
en
ce
as
0
.
3
.
−
Sm
o
k
in
g
s
tatu
s
:
s
m
o
k
in
g
s
tatu
s
is
an
o
th
er
cr
itical
d
eter
m
i
n
an
t
o
f
v
ar
io
u
s
h
ea
lth
r
is
k
s
.
C
u
r
r
en
t
s
m
o
k
er
s
an
d
ex
-
s
m
o
k
er
s
h
av
e
d
if
f
e
r
en
t
r
is
k
p
r
o
f
iles
co
m
p
ar
ed
t
o
in
d
i
v
id
u
als
wh
o
h
a
v
e
n
e
v
er
s
m
o
k
ed
.
T
h
is
f
ea
tu
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Ma
ch
in
e
lea
r
n
in
g
mo
d
el
fo
r
a
cc
u
r
a
te
p
r
ed
ictio
n
o
f c
o
r
o
n
a
r
y
…
(
S
a
n
t
h
o
s
h
Gu
p
ta
Do
g
i
p
a
r
th
i
)
5661
co
m
b
in
es
th
e
in
f
o
r
m
atio
n
f
r
o
m
C
u
r
r
en
t_
S
mo
ke
r
an
d
E
x
_
S
mo
ke
r
f
ea
tu
r
es
in
to
a
s
in
g
le
ca
teg
o
r
ical
v
ar
iab
le.
I
t
u
s
es
th
e
ca
teg
o
r
ical
v
alu
es
o
f
th
e
two
s
m
o
k
i
n
g
f
ea
tu
r
e
to
m
ap
in
to
a
s
in
g
le
v
alu
e
th
at
co
r
r
esp
o
n
d
s
to
o
n
e
o
f
th
e
f
o
llo
win
g
th
r
ee
ca
teg
o
r
ies
: i)
0
f
o
r
‘
N
ev
er S
mo
ke
d
’,
ii)
0
.
5
f
o
r
‘
E
x_
S
mo
ke
r
’,
an
d
iii)
1
f
o
r
‘
C
u
r
r
en
t_
S
mo
ke
r
’.
B
y
co
m
b
in
in
g
in
to
a
s
in
g
le
ca
teg
o
r
ical
v
ar
ia
b
le,
th
e
co
m
p
lex
ity
o
f
h
a
n
d
lin
g
m
u
ltip
le
r
elate
d
f
ea
tu
r
es
is
r
ed
u
ce
d
.
T
h
is
h
elp
s
in
b
etter
in
ter
p
r
etatio
n
a
n
d
a
n
aly
s
is
.
3
.
4
.
M
o
del
t
ra
ini
ng
a
nd
ev
a
l
ua
t
io
n
T
h
e
p
ip
elin
e
a
r
ch
itectu
r
e
o
f
th
e
tu
n
ed
ML
m
o
d
el
is
d
e
p
icted
in
Fig
u
r
e
1
.
T
h
is
wo
r
k
f
lo
w
c
ap
tu
r
es
all
s
tag
es
o
f
th
e
m
o
d
elin
g
p
r
o
ce
s
s
,
in
clu
d
in
g
d
ata
p
r
e
-
p
r
o
ce
s
s
in
g
,
d
er
i
v
ed
f
ea
tu
r
e
cr
ea
tio
n
,
m
o
d
el
tr
ain
in
g
with
co
s
t
-
s
en
s
itiv
e
lear
n
in
g
,
m
a
n
u
al
weig
h
tin
g
s
,
an
d
class
if
icatio
n
th
r
esh
o
l
d
ad
ju
s
tm
en
t
to
o
p
tim
ize
p
r
e
d
ictiv
e
p
er
f
o
r
m
an
ce
.
T
h
e
p
r
o
p
o
s
ed
p
ip
elin
e
o
f
f
er
s
a
s
tr
u
ctu
r
e
d
a
n
d
e
f
f
ec
tiv
e
m
eth
o
d
o
lo
g
y
f
o
r
h
an
d
lin
g
co
m
p
le
x
m
ed
ical
d
atasets
.
I
n
itially
,
th
e
d
ataset
was
p
ar
titi
o
n
ed
in
to
7
0
%
f
o
r
tr
ain
in
g
an
d
3
0
%
f
o
r
test
in
g
.
Af
te
r
tr
ain
in
g
,
th
e
m
o
d
el
p
r
ed
ictio
n
s
wer
e
v
alid
ated
ac
r
o
s
s
th
e
f
u
ll
d
ataset
u
s
in
g
cr
o
s
s
-
v
alid
ated
p
r
ed
ictio
n
s
to
en
s
u
r
e
r
o
b
u
s
tn
ess
.
T
h
e
m
o
d
elin
g
p
r
o
ce
s
s
s
tar
ts
b
y
lo
ad
in
g
th
e
tr
ai
n
in
g
an
d
test
s
ets,
f
o
llo
wed
b
y
p
r
ep
r
o
ce
s
s
in
g
an
d
f
ea
tu
r
e
s
elec
tio
n
.
T
h
e
f
i
v
e
b
ase
class
if
ier
s
ar
e
tr
ain
ed
,
th
eir
o
u
tp
u
ts
f
ed
i
n
to
a
m
eta
-
class
if
ier
,
an
d
th
e
en
s
em
b
le
m
o
d
el
is
ev
alu
ate
d
u
s
in
g
a
test
d
ataset.
Fo
r
th
e
b
ase
co
n
f
ig
u
r
atio
n
(
with
o
u
t
an
y
FN
-
r
ed
u
ctio
n
m
eth
o
d
s
)
,
th
e
tr
ain
in
g
p
r
o
c
ed
u
r
e
in
c
lu
d
es
th
e
f
o
llo
win
g
s
tep
s
:
(
i)
L
o
ad
th
e
d
ataset
an
d
p
er
f
o
r
m
a
7
0
:3
0
tr
ain
-
test
s
p
lit;
(
ii)
Ap
p
ly
d
ata
c
lean
in
g
an
d
f
ea
tu
r
e
s
elec
tio
n
;
(
iii)
T
r
ain
th
e
en
s
e
m
b
le
m
o
d
el
u
s
in
g
f
iv
e
b
ase
class
if
ier
s
an
d
a
m
eta
-
lear
n
e
r
;
(
iv
)
T
est
th
e
f
u
ll
d
ataset
o
n
th
e
tr
ain
ed
m
o
d
el;
(
v
)
C
o
m
p
a
r
e
p
r
e
d
icted
v
alu
es
with
ac
tu
al
lab
els
to
g
en
e
r
ate
th
e
co
n
f
u
s
io
n
m
atr
ix
;
(
v
i)
C
o
m
p
u
te
ev
alu
a
tio
n
m
etr
ics
s
u
ch
as
ac
c
u
r
a
cy
(
AC
C
)
,
p
r
ec
is
io
n
(
P),
r
e
ca
ll/s
en
s
itiv
ity
(
S),
s
p
ec
if
icity
(
SP
)
,
F1
-
s
co
r
e,
Ma
tth
ew’
s
C
o
r
r
elatio
n
C
o
ef
f
icien
t (
MCC
)
,
an
d
ar
ea
u
n
d
er
th
e
c
u
r
v
e
(
AUC).
T
h
e
p
ip
elin
e
en
s
u
r
es in
ter
p
r
et
ab
ilit
y
an
d
r
ep
r
o
d
u
cib
ilit
y
wh
i
le
en
ab
lin
g
th
e
u
s
e
o
f
en
s
em
b
l
e
m
eth
o
d
s
with
in
teg
r
ated
FN
r
ed
u
ctio
n
.
A
d
etailed
co
m
p
ar
is
o
n
o
f
th
e
p
er
f
o
r
m
an
ce
m
etr
ics
f
o
r
t
h
e
b
aselin
e
m
o
d
el
(
with
o
u
t
FN
r
e
d
u
ctio
n
)
is
p
r
e
s
en
ted
later
in
s
ec
tio
n
4
(
R
esu
lts
an
d
Dis
cu
s
s
io
n
)
,
wh
e
r
e
T
ab
le
3
r
ep
o
r
ts
th
e
co
n
f
u
s
io
n
m
atr
ix
an
d
ass
o
ciate
d
ev
alu
atio
n
o
u
tco
m
es.
Fig
u
r
e
1
.
Pip
elin
e
a
r
ch
itectu
r
e
o
f
th
e
p
r
o
p
o
s
ed
ML
m
o
d
el
,
in
clu
d
in
g
p
r
ep
r
o
ce
s
s
in
g
,
f
ea
tu
r
e
en
g
in
ee
r
in
g
,
FN
-
r
ed
u
ctio
n
,
tr
ai
n
in
g
,
a
n
d
e
v
alu
atio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
6
,
Decem
b
e
r
20
25
:
5
6
5
5
-
5
6
6
6
5662
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
is
s
ec
tio
n
p
r
esen
ts
an
d
a
n
al
y
ze
s
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
p
r
o
p
o
s
ed
ML
m
o
d
el
f
o
r
C
AD
p
r
ed
ictio
n
,
f
o
cu
s
in
g
f
ir
s
t
o
n
th
e
b
aselin
e
co
n
f
ig
u
r
atio
n
with
o
u
t
an
y
er
r
o
r
-
r
ed
u
ctio
n
m
eth
o
d
o
lo
g
ies,
f
o
llo
wed
b
y
im
p
r
o
v
em
e
n
ts
ac
h
iev
ed
t
h
r
o
u
g
h
th
e
in
te
g
r
atio
n
o
f
in
d
i
v
id
u
al
an
d
co
m
b
in
e
d
f
alse
n
eg
at
iv
e
(
FN)
r
ed
u
ctio
n
s
tr
ateg
ies.
4
.
1
.
B
a
s
eline
m
o
del per
f
o
r
ma
nce
T
ab
le
3
s
u
m
m
ar
izes
th
e
c
o
n
f
u
s
io
n
m
atr
ix
a
n
d
p
e
r
f
o
r
m
an
ce
m
etr
ics
f
o
r
th
e
b
aselin
e
ML
m
o
d
el
tr
ain
ed
an
d
ev
al
u
ated
with
o
u
t
in
co
r
p
o
r
atin
g
an
y
FN
-
r
e
d
u
c
tio
n
m
eth
o
d
o
lo
g
ies.
T
h
is
co
n
f
ig
u
r
atio
n
u
s
es
th
e
f
u
ll
en
s
em
b
le
p
i
p
elin
e
with
o
r
ig
in
al
an
d
d
e
r
iv
ed
f
ea
tu
r
es
b
u
t
ap
p
lies
n
o
c
o
s
t
-
s
en
s
itiv
e
lear
n
in
g
,
m
an
u
al
weig
h
tin
g
,
t
h
r
esh
o
ld
tu
n
i
n
g
,
o
r
e
n
g
in
ee
r
e
d
s
tr
ateg
ies.
As
s
h
o
wn
,
wh
ile
th
e
m
o
d
el
d
em
o
n
s
tr
ates
r
esp
ec
tab
le
o
v
er
all
ac
cu
r
ac
y
a
n
d
AUC,
i
t
s
till
s
u
f
f
er
s
f
r
o
m
a
r
elativ
el
y
h
ig
h
er
f
alse
n
eg
ativ
e
(
FN)
co
u
n
t,
wh
ich
is
a
cr
itical
co
n
ce
r
n
in
clin
ical
ap
p
licatio
n
s
wh
er
e
u
n
d
ia
g
n
o
s
ed
C
AD
ca
s
e
s
ca
n
h
av
e
s
ev
er
e
c
o
n
s
eq
u
en
ce
s
.
T
h
e
r
esu
lts
f
r
o
m
T
ab
le
3
h
ig
h
lig
h
t
th
e
n
ee
d
to
ad
d
r
ess
th
e
F
N
is
s
u
e
d
ir
ec
tly
.
T
o
ev
alu
ate
t
h
e
im
p
ac
t
o
f
FN
-
r
ed
u
ctio
n
s
tr
ateg
ies,
ad
d
itio
n
al
ex
p
e
r
im
en
ts
wer
e
c
o
n
d
u
cted
,
a
p
p
ly
in
g
ea
ch
m
eth
o
d
in
d
iv
id
u
ally
an
d
in
co
m
b
in
atio
n
.
T
a
b
le
3
.
Pe
r
f
o
r
m
a
n
c
e
m
e
tr
ics
o
f
t
h
e
b
as
eli
n
e
m
o
d
e
l w
it
h
o
u
t
FN
r
e
d
u
ci
n
g
m
e
th
o
d
o
l
o
g
ies
S
.
N
o
.
D
a
t
a
s
e
t
C
o
n
f
u
s
i
o
n
M
a
t
r
i
x
P
e
r
f
o
r
ma
n
c
e
M
e
t
r
i
c
s
TN
FP
FN
TP
A
C
C
P
S
SP
F1
M
C
C
AUC
1.
JI
P
M
ER
20
8
6
94
8
9
.
0
6
%
9
2
.
1
6
%
9
4
.
0
0
%
7
1
.
4
3
%
9
3
.
0
7
%
6
7
.
2
3
%
9
2
.
0
4
%
4
.
2
.
I
nd
iv
idu
a
l F
N
-
re
du
ct
io
n
m
et
ho
do
lo
g
ies
Her
e
we
r
ep
o
r
t
th
e
ef
f
ec
tiv
e
n
ess
o
f
th
e
ML
m
o
d
el
in
ac
c
u
r
ate
p
r
ed
ictio
n
d
u
e
to
th
e
in
c
o
r
p
o
r
atio
n
o
f
FN
r
ed
u
ctio
n
m
eth
o
d
o
lo
g
ies
d
etailed
ea
r
lier
.
First,
we
d
etail
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
ML
m
o
d
el
with
o
n
ly
o
n
e
o
f
th
e
FN
r
e
d
u
ctio
n
m
eth
o
d
o
lo
g
ies
is
in
co
r
p
o
r
ated
.
I
n
T
ab
le
4
,
all
a
v
ailab
le
f
ea
t
u
r
es
f
r
o
m
th
e
J
I
PME
R
d
ataset
ar
e
ap
p
lied
a
n
d
p
r
o
v
i
d
ed
all
th
e
p
er
f
o
r
m
an
ce
m
etr
ics
o
f
th
e
m
o
d
el
with
in
d
iv
i
d
u
al
m
eth
o
d
o
lo
g
ies
u
tili
ze
d
in
th
e
s
tu
d
y
i.e
.
,
en
g
in
ee
r
ed
f
ea
tu
r
es,
co
s
t
-
s
en
s
itiv
e
lear
n
in
g
,
m
a
n
u
al
weig
h
tin
g
s
p
r
o
v
is
io
n
a
n
d
ca
lcu
latio
n
o
f
n
ew
th
r
esh
o
ld
s
u
s
in
g
p
r
ec
is
io
n
-
r
ec
all
tr
ad
e
-
o
f
f
.
First
m
o
d
el
i.e
.
,
b
aselin
e
m
o
d
el
(
S.No
.
1
in
T
ab
le
4
)
th
at
u
s
ed
o
n
ly
th
e
o
r
ig
in
al
f
ea
t
u
r
es
r
esu
lted
in
m
ax
im
u
m
p
er
ce
n
ta
g
es
f
o
r
v
ar
io
u
s
m
etr
ics
as:
ac
cu
r
ac
y
8
9
.
0
6
%,
p
r
ec
is
io
n
9
2
.
1
6
%,
r
ec
all
9
4
%,
F1
-
s
co
r
e
9
3
.
0
7
%
a
n
d
AUC
9
2
.
0
4
%.
T
h
ese
p
er
f
o
r
m
an
ce
m
etr
ics
ar
e
n
o
t
th
e
b
est
we
wis
h
ed
f
o
r
as
th
e
o
r
ig
in
al
d
ataset
f
ea
tu
r
es
h
as
o
n
ly
ca
p
tu
r
ed
th
e
ess
en
tial
ch
ar
ac
ter
is
tics
r
eq
u
ir
ed
f
o
r
ac
cu
r
ate
class
if
icatio
n
.
Seco
n
d
m
o
d
el
(
S.No
.
2
in
T
a
b
le
4
)
is
ad
d
itio
n
o
f
en
g
in
ee
r
ed
f
ea
tu
r
es
alo
n
g
wit
h
th
e
o
r
i
g
in
al
f
ea
tu
r
es
s
h
o
wed
s
lig
h
t
im
p
r
o
v
em
en
t
i
n
th
e
p
r
ec
is
io
n
9
3
%
a
n
d
r
ec
all
9
3
%,
b
u
t
with
a
m
ar
g
i
n
al
d
ec
r
ea
s
e
in
AUC
9
1
%.
M
ar
g
in
al
d
ec
r
ea
s
e
in
th
e
AUC
m
etr
ics
o
f
th
is
ca
s
e
ca
n
b
e
attr
ib
u
te
d
to
th
e
n
o
is
e
an
d
/o
r
r
ed
u
n
d
an
c
y
in
tr
o
d
u
ce
d
b
y
th
e
i
n
clu
s
io
n
o
f
th
e
d
e
r
iv
ed
f
ea
tu
r
es
f
o
r
th
e
tr
ain
in
g
an
d
p
r
ed
ictio
n
.
T
h
ir
d
m
o
d
el
(
S.No
.
3
in
T
ab
le
4
)
is
th
e
in
co
r
p
o
r
atio
n
o
f
th
e
c
o
s
t
m
atr
ix
th
at
wo
r
k
s
o
n
ly
o
n
th
e
o
r
ig
i
n
al
f
ea
tu
r
es
to
r
ed
u
ce
th
e
FN,
p
r
o
v
id
ed
s
i
g
n
if
ican
t
p
e
r
f
o
r
m
an
ce
im
p
r
o
v
em
en
t
in
ac
c
u
r
ac
y
9
1
.
4
1
%,
r
ec
all
9
7
%
an
d
F1
-
s
co
r
e
9
4
.
6
3
%,
alo
n
g
with
a
s
lig
h
t
co
m
p
r
o
m
is
e
in
A
UC
9
0
.
3
9
%.
T
h
is
im
p
r
o
v
em
e
n
t
o
f
th
e
ac
cu
r
ac
y
is
a
d
ir
ec
t
r
ef
lectio
n
o
f
th
e
h
ig
h
r
ed
u
ctio
n
o
f
FN
m
etr
ic
in
th
is
m
o
d
el.
E
ith
er
f
o
u
r
th
(
S.No
.
4
in
T
ab
le
4
)
m
o
d
el
th
at
in
co
r
p
o
r
ate
d
m
an
u
al
weig
h
tin
g
s
o
r
t
h
e
f
if
th
(
S.No
.
5
in
T
ab
le
4
)
m
o
d
el
th
at
in
co
r
p
o
r
ated
th
r
esh
o
ld
ad
j
u
s
tm
en
ts
to
th
e
o
r
ig
in
al
f
ea
t
u
r
es
d
ataset
d
id
n
o
t
s
h
o
w
an
y
s
ig
n
if
ican
t
ch
a
n
g
e
in
th
e
p
er
f
o
r
m
an
ce
co
m
p
ar
ed
t
o
th
e
f
ir
s
t
m
o
d
el
(
S.No
.
1
in
T
ab
le
4
)
.
T
h
is
in
d
icate
s
th
at
th
er
e
ar
e
n
o
ad
d
ed
ad
v
an
tag
es
in
u
s
in
g
b
o
th
m
an
u
al
weig
h
tin
g
s
an
d
th
r
esh
o
ld
ad
ju
s
tm
en
ts
m
eth
o
d
o
lo
g
ies
to
g
eth
er
f
o
r
ac
c
u
r
ate
p
r
ed
ictio
n
o
f
J
I
PME
R
C
AD
m
ed
ical
k
in
d
o
f
d
atasets
.
T
h
e
m
an
u
al
weig
h
tin
g
s
d
id
n
o
t
alter
th
e
m
o
d
el’
s
d
ec
is
io
n
b
o
u
n
d
ar
y
en
o
u
g
h
to
ca
u
s
e
an
y
ap
p
r
ec
iab
le
ch
an
g
e
in
p
er
f
o
r
m
an
ce
.
As
th
e
o
r
i
g
in
al
m
o
d
el
alr
ea
d
y
in
clu
d
ed
th
e
o
p
tim
al
th
r
esh
o
ld
in
th
e
p
r
ed
ictio
n
p
r
o
ce
s
s
,
th
e
f
if
th
m
o
d
el
th
at
in
co
r
p
o
r
ated
th
e
th
r
esh
o
ld
ad
ju
s
tm
en
ts
h
as
n
o
e
f
f
ec
t
o
n
r
ed
u
cin
g
th
e
FN.
T
h
e
m
o
d
e
l's
p
er
f
o
r
m
an
ce
m
etr
ics
s
u
g
g
est
th
at
th
e
d
ef
au
lt
th
r
esh
o
ld
was a
p
p
r
o
p
r
iate,
a
n
d
an
y
f
u
r
th
er
a
d
ju
s
tm
en
ts
d
id
n
o
t e
n
h
an
ce
o
r
d
eg
r
a
d
e
th
e
m
o
d
el’
s
p
er
f
o
r
m
an
ce
.
4
.
3
.
Co
m
bin
a
t
io
n o
f
F
N
-
re
du
ct
io
n m
et
ho
do
lo
g
ies
I
n
T
ab
le
5
,
we
r
ep
o
r
t
th
e
ef
f
ec
tiv
en
ess
o
f
th
e
ML
m
o
d
el
in
p
r
ed
ictin
g
th
e
J
I
PME
R
d
ataset
wh
en
co
m
b
in
atio
n
o
f
m
o
r
e
th
a
n
o
n
e
FN r
ed
u
cin
g
m
et
h
o
d
o
lo
g
ies a
r
e
in
co
r
p
o
r
ated
.
First co
m
b
in
a
tio
n
m
o
d
el
(
S.No
.
1
in
T
ab
le
5
)
th
at
u
s
ed
o
r
ig
i
n
a
l
an
d
d
er
iv
e
d
f
ea
tu
r
es
alo
n
g
with
m
an
u
al
weig
h
tin
g
s
p
r
o
v
id
ed
th
e
s
tr
o
n
g
est
p
er
f
o
r
m
an
ce
with
ac
cu
r
ac
y
9
0
.
6
3
%,
r
ec
all
9
6
.
0
0
%
an
d
F1
-
s
co
r
e
9
4
.
1
2
%,
wh
ile
AU
C
g
o
t
d
ec
r
ea
s
ed
to
9
1
.
0
0
%.
Seco
n
d
co
m
b
in
atio
n
m
o
d
el
(
S.No
.
2
in
T
ab
le
5
)
u
s
ed
o
r
ig
in
al
an
d
d
e
r
iv
ed
f
ea
tu
r
es
alo
n
g
with
th
r
esh
o
ld
ad
ju
s
tm
en
ts
,
m
ir
r
o
r
ed
th
e
f
ir
s
t
co
m
b
in
atio
n
m
o
d
e
l’
s
p
er
f
o
r
m
an
ce
m
etr
ics
in
d
ica
tin
g
th
at
th
er
e
is
n
o
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Ma
ch
in
e
lea
r
n
in
g
mo
d
el
fo
r
a
cc
u
r
a
te
p
r
ed
ictio
n
o
f c
o
r
o
n
a
r
y
…
(
S
a
n
t
h
o
s
h
Gu
p
ta
Do
g
i
p
a
r
th
i
)
5663
ad
d
itio
n
al
b
en
ef
it
c
o
m
p
a
r
ed
t
o
f
ir
s
t
co
m
b
in
atio
n
m
o
d
el
th
at
u
s
ed
o
r
ig
in
al
a
n
d
d
er
iv
e
d
f
ea
tu
r
es
alo
n
g
wit
h
m
an
u
al
weig
h
tin
g
s
.
T
h
e
th
ir
d
co
m
b
in
atio
n
m
o
d
el
(
S.No
.
3
in
T
ab
le
5
)
u
s
es
o
r
ig
in
al
a
n
d
d
e
r
iv
ed
f
ea
tu
r
es
with
a
co
s
t
m
atr
ix
,
d
em
o
n
s
tr
ates
b
alan
ce
d
p
e
r
f
o
r
m
an
ce
m
et
r
ics
with
an
ac
cu
r
ac
y
8
9
.
0
6
%,
p
r
e
cisi
o
n
9
3
.
0
0
%,
an
d
an
AUC
9
0
.
6
3
%
b
u
t
n
o
t
p
r
o
v
id
e
a
n
y
s
ig
n
i
f
ican
t
im
p
r
o
v
em
en
t
co
m
p
ar
ed
to
f
ir
s
t
(
o
r
th
ir
d
)
c
o
m
b
in
atio
n
m
o
d
el.
B
y
o
b
s
er
v
in
g
th
e
f
ir
s
t
th
r
ee
co
m
b
in
atio
n
m
o
d
el
c
ases
,
th
e
ad
d
itio
n
o
f
d
er
iv
e
d
f
ea
tu
r
es
with
an
y
m
eth
o
d
o
l
o
g
y
d
id
n
o
t
s
h
o
w
s
u
b
s
tan
tial
im
p
r
o
v
em
e
n
t
d
u
e
to
p
o
ten
tial
n
o
is
e
o
r
r
e
d
u
n
d
an
cy
.
T
h
e
f
o
u
r
t
h
co
m
b
in
atio
n
m
o
d
el
(
S.No
.
4
i
n
T
ab
le
5
)
u
s
ed
o
r
i
g
in
al
f
ea
t
u
r
es
alo
n
g
with
m
an
u
al
wei
g
h
tin
g
s
an
d
a
co
s
t
m
atr
ix
,
r
esu
lted
in
p
er
ce
n
tag
es
o
f
p
er
f
o
r
m
an
c
e
m
etr
ics
as:
ac
cu
r
ac
y
9
1
.
4
1
%,
p
r
ec
is
io
n
9
2
.
3
8
%,
F1
s
co
r
e
9
4
.
6
3
% a
n
d
r
ec
all
9
7
.
0
0
%.
T
h
e
f
if
th
co
m
b
in
atio
n
m
o
d
el
(
S.No
.
5
in
T
ab
le
5
)
u
s
es o
r
ig
in
al
f
ea
tu
r
es a
lo
n
g
with
m
an
u
al
weig
h
tin
g
s
an
d
th
r
e
s
h
o
ld
ad
ju
s
tm
en
t
m
ain
tain
s
with
th
e
o
r
ig
in
al
m
o
d
el’
s
(
S.No
.
1
in
T
ab
le
4
)
p
er
f
o
r
m
an
ce
.
T
h
e
s
ix
th
co
m
b
in
atio
n
m
o
d
el
(
S.No
.
6
in
T
ab
le
5
)
u
tili
zin
g
a
co
s
t
m
atr
ix
an
d
th
r
esh
o
l
d
ad
ju
s
tm
en
t
with
o
r
ig
in
al
f
ea
tu
r
es,
em
er
g
es
as
th
e
b
est
ML
p
r
ed
ictin
g
alg
o
r
ith
m
m
o
d
el,
r
es
u
lted
in
m
ax
im
u
m
p
er
ce
n
tag
es
f
o
r
v
ar
io
u
s
m
etr
i
cs
as:
ac
cu
r
ac
y
9
2
.
1
9
%,
r
ec
all
9
8
.
0
0
%,
F1
s
co
r
e
9
5
.
1
5
%,
an
d
MCC
7
6
.
0
8
%,
th
o
u
g
h
its
AUC
i
s
s
lig
h
tly
l
o
wer
at
9
0
.
3
9
%.
B
y
co
m
p
ar
i
n
g
th
e
last
th
r
ee
co
m
b
in
atio
n
m
o
d
el
ca
s
es,
th
e
ef
f
ec
tiv
en
ess
o
f
th
e
co
s
t
m
atr
ix
in
m
o
d
els
h
ig
h
lig
h
ts
its
im
p
o
r
tan
ce
in
im
p
r
o
v
in
g
r
e
ca
ll
b
y
r
ed
u
cin
g
th
e
FN
b
y
v
er
y
g
o
o
d
a
m
o
u
n
t.
T
ab
le
4
.
Per
f
o
r
m
an
ce
m
etr
ics o
f
th
e
b
aselin
e
m
o
d
el
with
in
d
iv
id
u
al
FN m
eth
o
d
o
lo
g
ies
S.
N
o
.
C
a
t
e
g
o
r
y
N
o
.
o
f
F
e
a
t
u
r
e
s
C
o
n
f
u
s
i
o
n
M
a
t
r
i
x
P
e
r
f
o
r
ma
n
c
e
M
e
t
r
i
c
s
TN
FP
FN
TP
A
C
C
P
S
SP
F1
M
C
C
AUC
1
B
a
se
l
i
n
e
m
o
d
e
l
36
20
8
6
94
8
9
.
0
6
%
9
2
.
1
6
%
9
4
.
0
0
%
7
1
.
4
3
%
9
3
.
0
7
%
6
7
.
2
3
%
9
2
.
0
4
%
2
En
g
i
n
e
e
r
e
d
f
e
a
t
u
r
e
s
46
21
7
7
93
8
9
.
0
6
%
9
3
.
0
0
%
9
3
.
0
0
%
7
5
.
0
0
%
9
3
.
0
0
%
6
8
.
0
0
%
9
1
.
0
0
%
3
C
o
s
t
-
se
n
si
t
i
v
e
l
e
a
r
n
i
n
g
36
20
8
3
97
9
1
.
4
1
%
9
2
.
3
8
%
9
7
.
0
0
%
7
1
.
4
3
%
9
4
.
6
3
%
7
3
.
6
8
%
9
0
.
3
9
%
4
M
a
n
u
a
l
w
e
i
g
h
t
a
d
j
u
st
m
e
n
t
36
20
8
6
94
8
9
.
0
6
%
9
2
.
1
6
%
9
4
.
0
0
%
7
1
.
4
3
%
9
3
.
0
7
%
6
7
.
2
3
%
9
2
.
0
4
%
5
Th
r
e
s
h
o
l
d
a
d
j
u
s
t
me
n
t
36
20
8
6
94
8
9
.
0
6
%
9
2
.
1
6
%
9
4
.
0
0
%
7
1
.
4
3
%
9
3
.
0
7
%
6
7
.
2
3
%
9
2
.
0
4
%
T
ab
le
5
.
Per
f
o
r
m
an
ce
m
etr
ics o
f
th
e
b
aselin
e
m
o
d
el
with
co
m
b
in
atio
n
o
f
d
if
f
er
en
t FN
m
et
h
o
d
o
lo
g
ies
S.
N
o
.
C
a
t
e
g
o
r
y
N
o
.
o
f
F
e
a
t
u
r
e
s
C
o
n
f
u
s
i
o
n
M
a
t
r
i
x
P
e
r
f
o
r
ma
n
c
e
M
e
t
r
i
c
s
TN
FP
FN
TP
A
C
C
P
S
SP
F1
M
C
C
AUC
1
En
g
i
n
e
e
r
e
d
f
e
a
t
u
r
e
s
+
M
a
n
u
a
l
w
e
i
g
h
t
a
d
j
u
s
t
me
n
t
46
20
8
4
96
9
0
.
6
3
%
9
2
.
3
1
%
9
6
.
0
0
%
7
1
.
4
3
%
9
4
.
1
2
%
7
1
.
4
2
%
9
1
.
0
0
%
2
En
g
i
n
e
e
r
e
d
f
e
a
t
u
r
e
s
+
Th
r
e
s
h
o
l
d
a
d
j
u
s
t
me
n
t
46
20
8
4
96
9
0
.
6
3
%
9
2
.
3
1
%
9
6
.
0
0
%
7
1
.
4
3
%
9
4
.
1
2
%
7
1
.
4
2
%
9
1
.
0
0
%
3
C
o
s
t
-
se
n
si
t
i
v
e
+
En
g
i
n
e
e
r
e
d
f
e
a
t
u
r
e
s
46
21
7
7
93
8
9
.
0
6
%
9
3
.
0
0
%
9
3
.
0
0
%
7
5
.
0
0
%
9
3
.
0
0
%
6
8
.
0
0
%
9
0
.
6
3
%
4
M
a
n
u
a
l
w
e
i
g
h
t
+
C
o
st
-
se
n
si
t
i
v
e
36
20
8
3
97
9
1
.
4
1
%
9
2
.
3
8
%
9
7
.
0
0
%
7
1
.
4
3
%
9
4
.
6
3
%
7
3
.
6
8
%
9
0
.
3
9
%
5
Th
r
e
s
h
o
l
d
+
M
a
n
u
a
l
w
e
i
g
h
t
36
20
8
6
94
8
9
.
0
6
%
9
2
.
1
6
%
9
4
.
0
0
%
7
1
.
4
3
%
9
3
.
0
7
%
6
7
.
2
3
%
9
2
.
0
4
%
6
C
o
s
t
-
se
n
si
t
i
v
e
+
Th
r
e
s
h
o
l
d
a
d
j
u
s
t
me
n
t
36
20
8
2
98
9
2
.
1
9
%
9
2
.
4
5
%
9
8
.
0
0
%
7
1
.
4
3
%
9
5
.
1
5
%
7
6
.
0
8
%
9
0
.
3
9
%
I
n
th
e
tech
n
iq
u
es
r
ep
o
r
ted
i
n
th
e
liter
atu
r
e
[
1
4
]
,
[
1
6
]
,
[
2
7
]
g
o
o
d
p
r
ed
ictio
n
ac
cu
r
ac
y
f
o
r
th
e
r
esp
ec
tiv
e
d
ataset
u
tili
ze
d
wer
e
ac
h
iev
ed
eith
er
b
y
b
u
ild
in
g
a
n
ew
s
o
p
h
is
ticated
ML
alg
o
r
ith
m
o
r
d
e
v
elo
p
in
g
a
d
ee
p
er
n
etwo
r
k
.
I
n
p
ar
ticu
la
r
,
v
e
r
y
f
ew
o
f
th
em
ar
e
co
n
ce
n
tr
atin
g
o
n
r
ed
u
ctio
n
o
f
t
y
p
e
–
I
an
d
ty
p
e
–
I
I
er
r
o
r
s
[
1
4
]
.
I
n
th
is
wo
r
k
,
we
in
c
o
r
p
o
r
ated
v
ar
i
o
u
s
m
eth
o
d
o
lo
g
ies
to
f
o
c
u
s
m
ain
ly
o
n
r
ed
u
ctio
n
o
f
FN
as
th
e
m
o
d
el
alm
o
s
t
p
ick
ed
o
u
t
th
o
s
e
p
atien
ts
wh
o
ar
e
h
av
in
g
h
ea
r
t
d
is
ea
s
e.
Fro
m
T
ab
le
4
an
d
5
,
it
is
ev
id
en
t
th
at
th
e
m
o
d
el
is
well
s
u
ited
f
o
r
s
ele
ctin
g
m
eth
o
d
o
lo
g
ies
lik
e
co
s
t
-
s
en
s
itiv
e
an
d
th
r
esh
o
ld
ad
j
u
s
tm
en
t
as
FN
ar
e
r
ed
u
ce
d
to
a
co
u
n
t
o
f
o
n
l
y
2
p
atien
ts
wh
ich
r
esu
lted
in
an
ac
cu
r
ac
y
im
p
r
o
v
em
e
n
t
f
r
o
m
8
9
.
0
6
%
to
9
2
.
1
9
%.
T
h
is
p
r
o
v
es
th
at
th
e
p
r
o
p
o
s
ed
id
ea
o
f
d
if
f
er
en
t
m
eth
o
d
o
lo
g
ies
ca
n
wo
r
k
as
a
b
etter
ML
f
r
a
m
ewo
r
k
to
p
r
o
v
id
e
ef
f
icien
t
p
r
e
d
ictio
n
m
o
d
el
wit
h
r
ed
u
ctio
n
o
f
ty
p
e
-
I
a
n
d
t
y
p
e
-
I
I
er
r
o
r
s
.
T
h
is
r
esu
lt
d
em
o
n
s
tr
ates
th
e
m
ax
im
u
m
clin
ical
v
alu
e
b
y
s
u
cc
ess
f
u
lly
m
in
im
izin
g
life
-
th
r
ea
ten
in
g
m
is
clas
s
if
icatio
n
s
wh
ile
p
r
eser
v
in
g
h
ig
h
d
iag
n
o
s
tic
p
r
ec
is
io
n
,
h
ig
h
lig
h
tin
g
t
h
e
o
p
t
im
al
tr
ad
e
-
o
f
f
ac
h
iev
ed
b
y
in
t
eg
r
atin
g
m
u
ltip
le
e
r
r
o
r
r
e
d
u
cti
o
n
s
tr
ateg
ies.
I
t
is
im
p
o
r
tan
t
to
n
o
te
th
at
th
e
r
e
p
o
r
ted
f
alse
n
e
g
ativ
e
(
FN)
c
o
u
n
t
o
f
2
was
o
b
tain
ed
f
r
o
m
th
e
3
0
%
test
s
et,
co
r
r
esp
o
n
d
in
g
to
ap
p
r
o
x
im
at
ely
1
2
8
p
atien
ts
.
T
h
is
co
u
n
t
r
ef
lects
th
e
m
o
d
el’
s
ev
alu
atio
n
p
er
f
o
r
m
an
ce
o
n
h
eld
-
o
u
t
d
ata
an
d
n
o
t th
e
en
tir
e
d
ataset.
5.
CO
NCLU
SI
O
N
I
n
th
is
r
esear
c
h
s
tu
d
y
,
a
s
tack
ed
e
n
s
em
b
le
-
b
ased
ML
class
if
ier
m
o
d
el
is
d
ev
elo
p
ed
f
o
r
co
r
o
n
ar
y
ar
ter
y
d
is
ea
s
e
p
r
ed
ictio
n
o
f
J
I
PME
R
o
u
t
-
p
atien
t
d
ataset
wit
h
h
ig
h
ac
c
u
r
ac
y
.
T
h
e
r
esear
c
h
wo
r
k
r
e
p
o
r
ted
i
n
th
is
p
ap
er
is
f
o
c
u
s
ed
o
n
r
ed
u
cin
g
th
e
FN
to
im
p
r
o
v
e
th
e
o
b
tain
ed
p
r
ed
ictio
n
ac
c
u
r
ac
y
.
T
h
is
is
ac
h
iev
ed
b
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
6
,
Decem
b
e
r
20
25
:
5
6
5
5
-
5
6
6
6
5664
tu
n
in
g
o
r
m
o
d
if
y
i
n
g
th
e
s
tack
ed
en
s
em
b
le
m
o
d
el
with
FN
r
ed
u
ctio
n
m
eth
o
d
o
lo
g
ies.
E
x
p
er
im
en
tal
r
esu
lts
s
h
o
wed
a
clea
r
r
e
d
u
ctio
n
in
F
N
co
u
n
t
an
d
im
p
r
o
v
ed
p
er
f
o
r
m
an
ce
ac
r
o
s
s
p
r
ec
is
io
n
,
r
ec
all,
F1
-
s
co
r
e,
an
d
AUC
m
etr
ics.
No
tab
ly
,
co
m
b
in
in
g
c
o
s
t
-
s
en
s
itiv
e
lear
n
in
g
an
d
t
h
r
e
s
h
o
ld
ad
ju
s
tm
en
t
r
e
d
u
ce
d
FN to
ju
s
t
two
ca
s
es
in
th
e
3
0
%
test
s
et,
with
a
co
r
r
esp
o
n
d
in
g
F1
-
s
co
r
e
o
f
9
5
.
1
5
%
an
d
r
ec
all
o
f
9
8
%
h
ig
h
lig
h
tin
g
th
e
clin
ical
im
p
ac
t
o
f
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
o
lo
g
y
.
An
y
d
ataset
h
av
in
g
s
im
ilar
f
ea
tu
r
es
lik
e
th
e
J
I
PME
R
m
ed
ical
r
ec
o
r
d
d
ataset
co
n
s
id
er
ed
in
t
h
is
r
esear
ch
ca
n
u
tili
ze
o
u
r
p
r
o
p
o
s
ed
ML
m
o
d
el
with
in
co
r
p
o
r
atio
n
o
f
s
u
itab
le
FN
r
ed
u
cin
g
m
eth
o
d
o
l
o
g
ies f
o
r
ac
c
u
r
ate
p
r
ed
ictio
n
.
T
h
is
wo
r
k
h
as
s
u
b
s
tan
tial
im
p
licatio
n
s
f
o
r
r
ea
l
-
tim
e
d
ec
is
io
n
s
u
p
p
o
r
t
in
ca
r
d
io
lo
g
y
,
esp
ec
ially
in
r
eso
u
r
ce
-
lim
ited
s
ettin
g
s
.
T
h
e
d
ev
el
o
p
ed
m
o
d
el
is
n
o
w
b
ei
n
g
p
r
ep
ar
e
d
f
o
r
d
e
p
lo
y
m
e
n
t
with
in
th
e
clin
ical
wo
r
k
f
lo
w
at
J
I
PME
R
,
with
p
lan
s
f
o
r
r
ea
l
-
wo
r
ld
v
alid
atio
n
an
d
ex
p
an
s
io
n
to
o
t
h
er
ca
r
d
io
v
ascu
lar
co
n
d
itio
n
s
.
Fu
tu
r
e
r
esear
ch
will
f
o
cu
s
o
n
in
teg
r
atin
g
th
is
p
r
ed
ictiv
e
f
r
am
ewo
r
k
with
d
ee
p
lear
n
in
g
-
b
ased
s
eg
m
en
tatio
n
m
o
d
els
u
s
in
g
an
g
i
o
g
r
a
p
h
y
im
ag
es,
th
er
eb
y
a
d
v
an
ci
n
g
to
wa
r
d
a
u
n
i
f
ied
an
d
au
to
m
ated
d
iag
n
o
s
tic
p
ip
elin
e.
Ad
d
itio
n
ally
,
th
e
p
r
o
p
o
s
ed
m
o
d
el
ca
n
b
e
f
u
r
th
er
e
n
h
an
ce
d
b
y
in
c
o
r
p
o
r
atin
g
r
ec
en
t
d
e
v
e
lo
p
m
en
ts
in
h
y
b
r
id
an
d
co
s
t
-
b
ased
e
n
s
em
b
le
lear
n
in
g
tech
n
iq
u
es to
im
p
r
o
v
e
p
r
e
d
ictiv
e
ac
cu
r
ac
y
an
d
clin
ical
a
d
ap
tab
ilit
y
.
ACK
NO
WL
E
DG
M
E
N
T
S
T
h
e
au
th
o
r
s
wo
u
ld
lik
e
to
ac
k
n
o
wled
g
e
Dr
.
Ak
in
ch
a
n
B
h
ar
d
waj,
Fo
r
m
er
Sen
io
r
R
esid
en
ts
,
Dep
ar
tm
en
t
o
f
C
ar
d
io
lo
g
y
,
J
I
PME
R
,
Pu
d
u
ch
er
r
y
f
o
r
ex
ten
d
in
g
h
is
s
u
p
p
o
r
t
to
wa
r
d
s
cr
ea
tio
n
o
f
t
h
e
d
atab
ase
u
s
ed
in
th
is
s
tu
d
y
.
W
e
also
lik
e
to
th
a
n
k
th
e
J
I
MPE
R
I
n
s
titu
te
E
th
ics
C
o
m
m
ittee
f
o
r
p
er
m
itti
n
g
u
s
to
u
s
e
th
ei
r
p
atien
ts
’
d
ata
f
o
r
th
is
r
esear
ch
s
tu
d
y
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
No
f
u
n
d
s
,
g
r
an
ts
,
o
r
o
th
e
r
s
u
p
p
o
r
t w
as r
ec
eiv
ed
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
San
th
o
s
h
Gu
p
ta
Do
g
ip
ar
th
i
✓
✓
✓
✓
✓
✓
J
ay
an
th
i K
✓
✓
✓
✓
✓
✓
Ajith
An
an
th
ak
r
is
h
n
a
Pil
lai
✓
✓
✓
✓
K
Nak
k
ee
r
an
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
T
h
e
au
th
o
r
s
r
ev
ea
ls
th
at
th
ey
h
av
e
n
o
co
n
f
licts
o
f
in
ter
est to
d
is
clo
s
e.
I
NF
O
RM
E
D
CO
NS
E
N
T
B
ein
g
a
r
et
r
o
s
p
ec
tiv
e
s
tu
d
y
,
th
e
i
n
f
o
r
m
ed
co
n
s
en
t
f
o
r
th
is
r
esear
ch
s
tu
d
y
is
waiv
ed
o
f
f
,
as
an
o
n
y
m
ized
p
atien
t
d
ata
h
as b
ee
n
p
r
o
v
id
ed
.
Hen
ce
f
o
r
th
eth
i
cs a
p
p
r
o
v
al
is
s
u
f
f
icien
t
f
o
r
t
h
is
r
esear
ch
s
tu
d
y
.
E
T
H
I
CAL AP
P
RO
V
AL
E
th
ical
ap
p
r
o
v
al
f
o
r
th
is
s
tu
d
y
was
o
b
tain
ed
f
r
o
m
t
h
e
I
n
s
titu
te
E
th
ics
C
o
m
m
ittee
o
f
J
I
PME
R
(
I
E
C
Ap
p
r
o
v
al
No
.
1
0
8
7
/
2
0
1
9
/OB
S,
d
ated
1
2
th
Au
g
u
s
t
2
0
2
2
)
.
T
h
e
co
m
m
ittee
g
r
an
ted
p
e
r
m
i
s
s
io
n
f
o
r
th
e
u
s
e
o
f
p
atien
t d
ata
in
th
is
r
esear
ch
.
Evaluation Warning : The document was created with Spire.PDF for Python.