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.
16
,
No
.
5
,
Octo
b
er
20
26
,
p
p
.
2
8
3
6
~
2
8
4
6
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
16
i
5
.
pp
2
8
3
6
-
2
8
4
6
2836
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
H
ea
rt
disea
se pre
diction t
hro
ug
h f
e
a
ture
o
ptimi
za
tion,
ensem
bles, and
t
h
resho
ld calibra
tio
n
M
ehd
i Mo
s
t
o
f
i,
T
o
ura
j
B
a
ni
Ro
s
t
a
m
M
a
s
t
e
r
o
f
D
a
t
a
A
n
a
l
y
si
s D
e
p
a
r
t
m
e
n
t
,
U
n
i
v
e
r
si
t
y
o
f
N
i
a
g
a
r
a
F
a
l
l
s
,
N
i
a
g
a
r
a
F
a
l
l
s
,
O
n
t
a
r
i
o
,
C
a
n
a
d
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
y
1
5
,
2
0
2
6
R
ev
is
ed
Au
g
1
3
,
2
0
2
6
Acc
ep
ted
Au
g
1
9
,
2
0
2
6
Early
d
e
tec
ti
o
n
o
f
c
a
rd
i
o
v
a
sc
u
lar
d
ise
a
se
is
e
ss
e
n
ti
a
l
f
o
r
re
d
u
c
i
n
g
m
o
rtalit
y
a
n
d
e
n
a
b
li
n
g
ti
m
e
ly
c
li
n
ica
l
i
n
ter
v
e
n
ti
o
n
.
Alth
o
u
g
h
m
a
c
h
in
e
lea
rn
i
n
g
m
o
d
e
ls
a
re
wid
e
ly
u
se
d
f
o
r
h
e
a
rt
d
ise
a
se
p
re
d
ictio
n
,
m
a
n
y
stu
d
ies
fo
c
u
s
p
r
i
m
a
ril
y
o
n
o
v
e
ra
ll
a
c
c
u
ra
c
y
ra
t
h
e
r
t
h
a
n
s
e
n
siti
v
it
y
,
wh
ich
is
c
rit
ica
l
in
m
e
d
ica
l
sc
re
e
n
in
g
tas
k
s.
T
h
is
stu
d
y
p
r
o
p
o
se
d
a
stru
c
tu
re
d
m
u
lt
i
-
sta
g
e
o
p
t
imiz
a
ti
o
n
fra
m
e
wo
rk
th
a
t
in
teg
ra
ted
fe
a
t
u
re
se
lec
ti
o
n
,
e
n
se
m
b
le
m
o
d
e
l
t
u
n
in
g
,
a
n
d
se
n
siti
v
it
y
-
fo
c
u
se
d
d
e
c
isi
o
n
-
t
h
re
sh
o
l
d
a
d
ju
stm
e
n
t
with
in
a
u
n
ifi
e
d
p
re
d
ictiv
e
p
ip
e
li
n
e
.
F
e
a
tu
re
se
lec
ti
o
n
wa
s c
o
n
d
u
c
te
d
u
si
n
g
re
c
u
rsi
v
e
fe
a
tu
re
e
li
m
in
a
ti
o
n
with
c
ro
ss
-
v
a
li
d
a
ti
o
n
(RF
ECV)
a
n
d
imp
o
rtan
c
e
-
b
a
se
d
tec
h
n
iq
u
e
s,
re
su
lt
in
g
in
a
n
o
p
ti
m
a
l
su
b
se
t
o
f
1
4
p
re
d
ict
o
rs
.
E
n
se
m
b
le
c
las
sifiers
in
c
lu
d
in
g
ra
n
d
o
m
fo
re
st,
e
x
tra
tree
s
,
a
n
d
XG
Bo
o
st
we
re
o
p
ti
m
ize
d
th
r
o
u
g
h
h
y
p
e
r
p
a
ra
m
e
ter
tu
n
i
n
g
u
si
n
g
Ra
n
d
o
m
ize
d
S
e
a
rc
h
CV
.
To
p
ri
o
rit
ize
se
n
siti
v
it
y
,
t
h
e
c
las
sifica
ti
o
n
th
re
sh
o
l
d
wa
s
c
a
li
b
ra
ted
fro
m
0
.
5
to
0
.
3
.
In
th
e
e
v
a
lu
a
ti
o
n
sp
li
t,
re
c
a
ll
i
n
c
re
a
se
d
to
0
.
9
6
1
wh
il
e
m
a
in
tai
n
in
g
str
o
n
g
d
isc
rimin
a
ti
v
e
p
e
rfo
rm
a
n
c
e
(AU
C
≈
0
.
9
3
)
.
R
o
b
u
stn
e
ss
wa
s
f
u
rth
e
r
a
ss
e
ss
e
d
u
sin
g
stra
ti
fie
d
c
ro
ss
-
v
a
li
d
a
ti
o
n
a
n
d
b
o
o
tstrap
re
s
a
m
p
li
n
g
,
p
ro
d
u
c
i
n
g
a
m
e
a
n
re
c
a
ll
o
f
0
.
9
0
2
with
a
9
5
%
c
o
n
fi
d
e
n
c
e
in
terv
a
l
o
f
0
.
8
4
2
–
0
.
9
5
9
.
Th
e
re
su
lt
s
d
e
m
o
n
stra
te
d
th
a
t
in
te
g
ra
ti
n
g
fe
a
tu
re
re
fin
e
m
e
n
t,
e
n
se
m
b
le
lea
rn
in
g
,
a
n
d
th
re
sh
o
l
d
c
a
li
b
ra
ti
o
n
imp
r
o
v
e
s
th
e
se
n
siti
v
it
y
a
n
d
re
li
a
b
il
it
y
o
f
h
e
a
rt
d
ise
a
se
p
re
d
ictio
n
m
o
d
e
ls.
K
ey
w
o
r
d
s
:
E
n
s
em
b
le
lear
n
in
g
Featu
r
e
o
p
tim
izatio
n
Hea
r
t d
is
ea
s
e
p
r
ed
ictio
n
Hy
p
er
p
ar
a
m
eter
tu
n
in
g
Ma
ch
in
e
lear
n
in
g
T
h
r
esh
o
ld
ca
lib
r
atio
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
:
Me
h
d
i M
o
s
to
f
i
Ma
s
ter
o
f
Data
An
aly
s
is
Dep
ar
tm
en
t,
Un
iv
er
s
ity
o
f
Niag
ar
a
Falls
4
3
4
2
Q
u
ee
n
Stre
et,
Niag
ar
a
Falls
,
ON,
L
2
E
7
J
7
,
C
an
ad
a
E
m
ail: m
eh
d
i.m
o
s
to
f
i@
u
n
f
c
.
c
a
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
s
u
b
s
tan
tial
g
lo
b
al
b
u
r
d
e
n
o
f
ca
r
d
io
v
ascu
lar
d
is
ea
s
e
m
ak
es
tim
ely
id
en
tific
atio
n
o
f
in
d
i
v
id
u
als
at
elev
ated
r
is
k
an
im
p
o
r
tan
t
o
b
jectiv
e
in
clin
ical
d
ec
is
io
n
-
m
ak
in
g
.
R
eliab
le
p
r
ed
ictiv
e
m
o
d
els
ca
n
h
elp
clin
ician
s
r
ec
o
g
n
ize
in
d
iv
i
d
u
als
at
elev
ated
ca
r
d
io
v
ascu
la
r
r
is
k
a
n
d
s
u
p
p
o
r
t
ea
r
lier
in
t
er
v
en
tio
n
.
Ma
ch
in
e
lear
n
in
g
h
as
in
cr
ea
s
in
g
ly
b
e
en
ap
p
lied
to
clin
ical
d
atas
ets
to
s
u
p
p
o
r
t
r
is
k
esti
m
atio
n
an
d
d
ata
-
d
r
iv
en
ass
es
s
m
en
t o
f
ca
r
d
io
v
ascu
lar
co
n
d
itio
n
s
[
1
]
,
[
2
]
.
Hea
r
t
d
is
ea
s
e
p
r
ed
ictio
n
h
as
b
ee
n
in
v
esti
g
ated
u
s
in
g
a
b
r
o
ad
r
an
g
e
o
f
alg
o
r
ith
m
s
,
f
r
o
m
lo
g
is
tic
r
eg
r
ess
io
n
an
d
s
u
p
p
o
r
t v
ec
to
r
m
ac
h
in
es
to
d
ec
is
io
n
tr
ee
s
an
d
en
s
em
b
le
ap
p
r
o
ac
h
es
s
u
ch
as
r
an
d
o
m
f
o
r
es
t
an
d
XGBo
o
s
t
[
3
]
.
T
h
ese
m
eth
o
d
s
ca
n
ca
p
tu
r
e
n
o
n
lin
ea
r
in
ter
ac
tio
n
s
am
o
n
g
clin
ical
p
r
ed
icto
r
s
an
d
h
av
e
d
em
o
n
s
tr
ated
s
tr
o
n
g
p
er
f
o
r
m
an
ce
in
m
ed
ical
class
if
icatio
n
task
s
.
R
ec
en
t
s
tu
d
ies
f
u
r
th
er
d
em
o
n
s
tr
ate
th
at
m
ac
h
in
e
lear
n
in
g
tech
n
i
q
u
es
ca
n
ef
f
ec
tiv
ely
id
e
n
tify
c
o
m
p
lex
p
atter
n
s
in
clin
ical
d
atasets
an
d
im
p
r
o
v
e
ca
r
d
io
v
ascu
lar
r
is
k
p
r
ed
ictio
n
[
4
]
,
[
5
]
.
Ho
wev
er
,
m
an
y
ex
is
tin
g
s
tu
d
ies
p
r
im
ar
ily
f
o
cu
s
o
n
m
ax
im
izin
g
o
v
er
all
ac
cu
r
ac
y
,
wh
ile
less
atten
tio
n
h
as
b
ee
n
g
iv
en
to
o
p
tim
izin
g
r
ec
all
in
clin
ical
s
cr
ee
n
in
g
s
ce
n
ar
io
s
wh
er
e
f
ailin
g
to
d
etec
t p
o
s
itiv
e
ca
s
es c
an
h
av
e
s
er
io
u
s
co
n
s
eq
u
en
ce
s
[
6
]
.
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
Hea
r
t d
is
ea
s
e
p
r
ed
ictio
n
th
r
o
u
g
h
fea
t
u
r
e
o
p
timiz
a
tio
n
,
en
s
e
mb
les,
a
n
d
th
r
esh
o
ld
c
a
lib
r
a
tio
n
(
Meh
d
i Mo
s
to
fi
)
2837
An
o
th
er
lim
itatio
n
in
ex
is
tin
g
wo
r
k
is
th
e
lack
o
f
in
te
g
r
ate
d
o
p
tim
izatio
n
s
tr
ateg
ies
th
at
in
teg
r
ate
f
ea
tu
r
e
s
elec
tio
n
,
m
o
d
el
tu
n
in
g
,
an
d
d
ec
is
io
n
-
th
r
esh
o
ld
ca
li
b
r
atio
n
with
in
a
u
n
i
f
ied
p
r
ed
i
ctiv
e
p
ip
elin
e
[
7
]
.
I
n
m
ed
ical
p
r
e
d
ictio
n
task
s
,
th
r
esh
o
ld
s
elec
tio
n
p
lay
s
a
n
im
p
o
r
tan
t
r
o
le
in
b
alan
cin
g
r
ec
all
an
d
p
r
ec
is
io
n
,
p
ar
ticu
lar
ly
w
h
en
t
h
e
g
o
al
is
to
m
i
n
im
ize
f
alse
n
eg
ativ
es.
Mo
r
eo
v
er
,
m
an
y
s
tu
d
ies
r
el
y
s
o
lely
o
n
s
in
g
le
ev
alu
atio
n
s
p
lits
with
o
u
t p
r
o
v
i
d
in
g
s
tatis
tical
v
alid
atio
n
to
ass
ess
th
e
r
o
b
u
s
tn
ess
o
f
th
e
r
esu
lts
[
8
]
.
T
o
ad
d
r
ess
th
ese
lim
itatio
n
s
,
t
h
is
s
tu
d
y
p
r
o
p
o
s
es
a
s
tr
u
ctu
r
e
d
m
ac
h
i
n
e
lear
n
i
n
g
f
r
am
ewo
r
k
f
o
r
h
ea
r
t
d
is
ea
s
e
p
r
ed
ictio
n
th
at
in
teg
r
ates
f
ea
tu
r
e
s
elec
tio
n
,
e
n
s
em
b
le
m
o
d
elin
g
,
a
n
d
s
en
s
itiv
ity
-
f
o
cu
s
ed
t
h
r
esh
o
ld
o
p
tim
izatio
n
[
9
]
.
T
h
e
f
r
am
ew
o
r
k
is
ev
alu
ated
u
s
in
g
s
ev
er
a
l
m
ac
h
in
e
lear
n
in
g
m
o
d
els
a
n
d
v
alid
ated
u
s
in
g
s
tatis
t
ical
tech
n
iq
u
es
to
en
s
u
r
e
r
o
b
u
s
tn
ess
an
d
r
eliab
ilit
y
o
f
p
r
ed
ictiv
e
p
er
f
o
r
m
an
ce
.
T
h
e
m
ain
co
n
tr
i
b
u
tio
n
s
o
f
th
is
s
tu
d
y
ar
e
as f
o
llo
ws:
a.
A
s
tr
u
ctu
r
ed
p
r
ed
ictio
n
f
r
am
e
wo
r
k
th
at
in
teg
r
ates
f
ea
tu
r
e
s
elec
tio
n
,
en
s
em
b
le
m
o
d
elin
g
,
an
d
s
en
s
itiv
ity
-
f
o
cu
s
ed
th
r
esh
o
ld
o
p
tim
izatio
n
f
o
r
h
ea
r
t d
is
ea
s
e
r
is
k
class
if
icatio
n
.
b.
A
s
en
s
itiv
ity
-
f
o
cu
s
ed
th
r
esh
o
l
d
o
p
tim
izatio
n
d
esig
n
ed
f
o
r
m
ed
ical
s
cr
ee
n
in
g
s
ce
n
ar
io
s
wh
er
e
m
in
im
izin
g
f
alse n
eg
ativ
es is
cr
itical.
c.
A
s
tatis
tically
v
alid
ated
ev
alu
atio
n
,
in
clu
d
in
g
c
r
o
s
s
-
v
alid
atio
n
s
tab
ilit
y
an
aly
s
is
,
b
o
o
ts
tr
ap
co
n
f
i
d
en
ce
in
ter
v
als,
an
d
Mc
Nem
ar
s
ig
n
if
ican
ce
test
in
g
f
o
r
m
o
d
el
c
o
m
p
ar
is
o
n
.
Ma
ch
in
e
lear
n
i
n
g
tech
n
iq
u
es
h
av
e
b
ee
n
wid
ely
a
p
p
lied
to
s
u
p
p
o
r
t
th
e
p
r
ed
ictio
n
an
d
d
i
ag
n
o
s
is
o
f
ca
r
d
io
v
ascu
lar
d
is
ea
s
es.
W
it
h
th
e
g
r
o
win
g
av
ailab
ilit
y
o
f
clin
ical
d
atasets
an
d
elec
tr
o
n
ic
h
ea
lth
r
e
co
r
d
s
,
p
r
ed
ictiv
e
m
o
d
els
h
av
e
b
ee
n
d
ev
elo
p
ed
to
id
en
tif
y
p
atie
n
ts
at
r
is
k
o
f
h
ea
r
t
d
is
ea
s
e
u
s
in
g
d
em
o
g
r
ap
h
ic,
p
h
y
s
io
lo
g
ical,
an
d
d
iag
n
o
s
tic
v
ar
iab
les.
A
v
a
r
iety
o
f
m
ac
h
i
n
e
lear
n
in
g
alg
o
r
ith
m
s
h
av
e
b
ee
n
ex
p
l
o
r
ed
in
t
h
is
co
n
tex
t to
im
p
r
o
v
e
class
if
icati
o
n
p
er
f
o
r
m
a
n
ce
an
d
s
u
p
p
o
r
t c
l
in
ical
d
ec
is
io
n
-
m
ak
in
g
[
9
]
–
[
1
1
]
.
T
r
ad
itio
n
al
s
tatis
tical
an
d
m
ac
h
in
e
lear
n
in
g
m
o
d
els
s
u
ch
as
lo
g
is
tic
r
eg
r
ess
io
n
h
av
e
o
f
ten
b
ee
n
u
s
ed
as
b
aselin
e
ap
p
r
o
a
ch
es
d
u
e
t
o
th
eir
in
te
r
p
r
etab
ilit
y
a
n
d
e
f
f
ec
tiv
en
ess
in
b
in
ar
y
class
if
icatio
n
task
s
[
1
2
]
.
I
n
ad
d
itio
n
,
alg
o
r
ith
m
s
ca
p
ab
le
o
f
m
o
d
elin
g
n
o
n
lin
ea
r
r
elatio
n
s
h
ip
s
,
in
clu
d
in
g
s
u
p
p
o
r
t v
ec
to
r
m
ac
h
in
es
an
d
tr
ee
-
b
ased
m
o
d
els,
h
av
e
b
ee
n
wi
d
ely
ap
p
lied
in
h
ea
lth
ca
r
e
an
aly
tics
[
1
3
]
.
T
h
ese
m
eth
o
d
s
ar
e
ab
le
to
ca
p
tu
r
e
co
m
p
lex
in
ter
ac
tio
n
s
am
o
n
g
clin
ical
p
r
ed
icto
r
s
a
n
d
h
av
e
s
h
o
wn
p
r
o
m
is
in
g
p
er
f
o
r
m
a
n
ce
in
d
is
ea
s
e
r
is
k
p
r
ed
ictio
n
p
r
o
b
lem
s
[
1
4
]
,
[
1
5
]
.
E
n
s
em
b
le
lear
n
in
g
tech
n
iq
u
es
h
av
e
g
ai
n
ed
p
ar
ticu
lar
atten
ti
o
n
in
m
e
d
ical
p
r
e
d
ictio
n
task
s
b
ec
au
s
e
th
ey
in
teg
r
ate
m
u
ltip
le
b
ase
le
ar
n
er
s
to
im
p
r
o
v
e
p
r
e
d
ictiv
e
a
cc
u
r
ac
y
a
n
d
r
o
b
u
s
tn
ess
.
Me
th
o
d
s
s
u
ch
as
r
an
d
o
m
f
o
r
est
an
d
ex
tr
a
tr
ee
s
ag
g
r
eg
a
te
m
u
ltip
le
d
ec
is
io
n
tr
ee
s
to
r
ed
u
ce
v
ar
ian
ce
a
n
d
im
p
r
o
v
e
g
en
er
aliza
tio
n
[
1
6
]
.
B
o
o
s
tin
g
-
b
ased
alg
o
r
ith
m
s
,
in
clu
d
in
g
ex
tr
e
m
e
g
r
ad
ien
t
b
o
o
s
tin
g
(
XGBo
o
s
t)
,
f
u
r
th
er
en
h
an
ce
p
r
e
d
ictiv
e
p
er
f
o
r
m
an
ce
b
y
s
eq
u
en
tially
b
u
ild
in
g
m
o
d
els
th
at
f
o
cu
s
o
n
co
r
r
ec
tin
g
p
r
e
v
io
u
s
er
r
o
r
s
[
1
7
]
.
R
ec
en
t
r
esear
ch
also
h
ig
h
lig
h
ts
th
e
ef
f
ec
tiv
en
e
s
s
o
f
en
s
em
b
le
an
d
h
y
b
r
id
lea
r
n
in
g
f
r
am
ewo
r
k
s
f
o
r
im
p
r
o
v
in
g
ca
r
d
i
o
v
ascu
lar
d
is
ea
s
e
p
r
ed
ictio
n
u
s
in
g
cli
n
ical
d
atasets
[
1
8
]
,
[
1
9
]
.
T
h
ese
en
s
em
b
le
m
eth
o
d
s
ar
e
th
e
r
e
f
o
r
e
well
s
u
ited
to
h
ea
lth
ca
r
e
d
atasets
th
at
o
f
ten
c
o
n
tain
n
o
n
lin
ea
r
p
atter
n
s
an
d
h
eter
o
g
en
e
o
u
s
v
ar
ia
b
les.
Featu
r
e
s
elec
tio
n
is
an
o
th
er
im
p
o
r
tan
t
co
m
p
o
n
e
n
t
o
f
p
r
ed
i
ctiv
e
m
o
d
elin
g
b
ec
au
s
e
it
h
el
p
s
id
en
tify
th
e
m
o
s
t
i
n
f
o
r
m
ativ
e
v
ar
iab
l
es
wh
ile
r
ed
u
cin
g
m
o
d
el
co
m
p
lex
ity
.
T
ec
h
n
iq
u
es
s
u
ch
a
s
r
ec
u
r
s
iv
e
f
ea
tu
r
e
elim
in
atio
n
with
cr
o
s
s
-
v
alid
at
io
n
(
R
FECV)
ar
e
co
m
m
o
n
ly
u
s
ed
to
d
eter
m
in
e
o
p
tim
al
s
u
b
s
ets
o
f
p
r
ed
icto
r
s
wh
ile
m
ain
tain
in
g
p
r
ed
ictiv
e
p
er
f
o
r
m
an
ce
[
2
0
]
.
I
n
ad
d
itio
n
,
th
e
ch
o
ice
o
f
class
if
icatio
n
th
r
esh
o
ld
p
lay
s
an
im
p
o
r
tan
t
r
o
le
in
m
e
d
ical
s
cr
ee
n
in
g
task
s
.
B
ec
au
s
e
f
alse
n
eg
ativ
es
m
ay
h
av
e
s
er
io
u
s
clin
ical
co
n
s
eq
u
en
ce
s
,
ad
ju
s
tin
g
th
e
d
ec
is
io
n
th
r
esh
o
ld
ca
n
im
p
r
o
v
e
r
ec
all
an
d
i
n
cr
ea
s
e
th
e
d
etec
tio
n
o
f
h
i
g
h
-
r
is
k
p
atien
ts
[
2
1
]
.
R
ec
en
t
s
tu
d
ies
al
s
o
em
p
h
asize
th
e
g
r
o
win
g
im
p
o
r
tan
ce
o
f
ex
p
lain
ab
le
an
d
in
ter
p
r
eta
b
le
m
ac
h
in
e
lear
n
in
g
tech
n
iq
u
es in
h
ea
lth
c
ar
e
p
r
ed
i
ctio
n
m
o
d
els to
im
p
r
o
v
e
tr
an
s
p
ar
en
cy
a
n
d
clin
ical
tr
u
s
t
[
2
2
]
.
Desp
ite
ex
ten
s
iv
e
r
esear
ch
i
n
m
ac
h
i
n
e
lear
n
in
g
-
b
ased
h
ea
r
t
d
is
ea
s
e
p
r
ed
ictio
n
,
m
a
n
y
s
tu
d
ies
p
r
im
ar
ily
f
o
cu
s
o
n
im
p
r
o
v
in
g
o
v
e
r
all
class
if
icatio
n
ac
cu
r
ac
y
.
Featu
r
e
s
elec
tio
n
,
m
o
d
e
l
o
p
tim
izatio
n
,
a
n
d
th
r
esh
o
ld
ca
lib
r
atio
n
a
r
e
o
f
ten
in
v
esti
g
ated
s
ep
ar
ately
r
ath
er
th
an
in
teg
r
ated
with
in
a
u
n
if
i
ed
f
r
am
ewo
r
k
[
2
3
]
.
R
ec
en
t
r
ev
iews
co
n
f
ir
m
t
h
at
m
o
r
e
c
o
m
p
r
e
h
en
s
iv
e
m
ac
h
in
e
lear
n
in
g
f
r
am
ew
o
r
k
s
th
at
in
teg
r
ate
m
u
ltip
l
e
o
p
tim
izatio
n
s
tr
ateg
ies
ar
e
s
till
n
ee
d
ed
t
o
im
p
r
o
v
e
c
ar
d
io
v
a
s
cu
lar
d
is
ea
s
e
d
etec
tio
n
an
d
e
n
s
u
r
e
r
o
b
u
s
t
m
o
d
el
v
alid
atio
n
[
2
4
]
,
[
2
5
]
.
C
o
n
s
eq
u
en
tly
,
t
h
er
e
r
em
ain
s
a
n
ee
d
f
o
r
s
tr
u
ctu
r
e
d
a
p
p
r
o
ac
h
es
t
h
at
in
teg
r
ate
th
ese
o
p
tim
izatio
n
s
tr
ateg
ies
to
im
p
r
o
v
e
d
etec
tio
n
p
er
f
o
r
m
an
ce
wh
ile
alig
n
in
g
p
r
ed
ictiv
e
m
o
d
els
with
clin
ical
p
r
io
r
ities
.
2.
M
E
T
H
O
D
T
h
is
s
tu
d
y
p
r
o
p
o
s
es
a
m
u
lti
-
s
tag
e
o
p
tim
izatio
n
f
r
am
ewo
r
k
f
o
r
h
ea
r
t
d
is
ea
s
e
p
r
ed
ictio
n
th
a
t
im
p
r
o
v
es
m
o
d
el
p
e
r
f
o
r
m
an
ce
t
h
r
o
u
g
h
th
r
ee
co
o
r
d
in
ate
d
o
p
tim
izatio
n
lay
er
s
:
f
ea
tu
r
e
o
p
tim
izatio
n
,
m
o
d
el
o
p
tim
izatio
n
,
an
d
d
ec
is
io
n
o
p
tim
izatio
n
.
T
h
e
f
r
am
ewo
r
k
b
eg
in
s
with
b
aselin
e
m
o
d
el
co
n
s
tr
u
ctio
n
an
d
p
r
o
g
r
ess
iv
ely
en
h
an
ce
s
p
r
e
d
ictiv
e
p
er
f
o
r
m
a
n
ce
th
r
o
u
g
h
f
ea
t
u
r
e
s
elec
tio
n
,
h
y
p
er
p
ar
am
eter
tu
n
in
g
,
a
n
d
s
en
s
itiv
ity
-
f
o
cu
s
ed
th
r
esh
o
ld
o
p
tim
izatio
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.
16
,
No
.
5
,
Octo
b
e
r
20
26
:
2
8
3
6
-
2
8
4
6
2838
T
h
e
p
r
o
p
o
s
ed
f
r
am
ewo
r
k
f
o
llo
ws
a
s
tr
u
ctu
r
ed
m
ac
h
in
e
lear
n
in
g
p
ip
elin
e
t
h
at
in
c
lu
d
es
d
ata
p
r
ep
r
o
ce
s
s
in
g
,
f
ea
tu
r
e
o
p
tim
izatio
n
,
en
s
em
b
le
m
o
d
el
tr
ain
in
g
,
th
r
esh
o
ld
ca
lib
r
atio
n
,
an
d
s
tatis
tical
v
alid
atio
n
.
T
h
e
o
v
e
r
all
r
esear
ch
wo
r
k
f
lo
w
o
f
th
e
p
r
o
p
o
s
ed
f
r
am
ew
o
r
k
is
illu
s
tr
ated
in
Fig
u
r
e
1
.
Fig
u
r
e
1
.
E
x
p
er
im
e
n
tal
m
eth
o
d
o
lo
g
y
an
d
v
alid
atio
n
w
o
r
k
f
l
o
w
o
f
th
e
p
r
o
p
o
s
ed
h
ea
r
t
d
is
ea
s
e
p
r
ed
ictio
n
f
r
am
ewo
r
k
2
.1
.
Da
t
a
s
et
des
cr
iptio
n
T
h
e
ex
p
e
r
im
en
ts
wer
e
c
o
n
d
u
cted
u
s
in
g
th
e
p
u
b
licly
av
ailab
le
Kag
g
le
h
ea
r
t
f
ailu
r
e
p
r
ed
ictio
n
d
ataset.
T
h
e
d
ataset
co
n
tain
s
d
em
o
g
r
a
p
h
ic
an
d
clin
ical
at
tr
ib
u
tes
in
clu
d
in
g
ag
e,
c
h
o
lest
er
o
l
lev
el,
r
esti
n
g
b
lo
o
d
p
r
ess
u
r
e,
elec
tr
o
ca
r
d
io
g
r
am
(
E
C
G)
m
ea
s
u
r
em
en
ts
,
an
d
ex
er
cise
-
in
d
u
ce
d
an
g
i
n
a.
T
h
e
tar
g
et
v
ar
iab
le
r
ep
r
esen
ts
th
e
p
r
esen
ce
o
r
ab
s
en
ce
o
f
h
ea
r
t d
is
ea
s
e,
f
o
r
m
in
g
a
b
in
ar
y
class
if
icatio
n
task
.
T
h
e
d
ataset
co
n
s
is
ts
o
f
s
tr
u
ctu
r
ed
clin
ical
attr
ib
u
tes
co
m
m
o
n
ly
ass
o
ciate
d
with
ca
r
d
io
v
a
s
cu
lar
r
is
k
ass
es
s
m
en
t.
W
h
ile
it
is
p
u
b
li
cly
av
ailab
le
an
d
wid
ely
u
s
ed
f
o
r
b
en
ch
m
a
r
k
in
g
m
ac
h
in
e
lear
n
in
g
m
o
d
els,
it
r
ep
r
esen
ts
ag
g
r
e
g
ated
p
atien
t
r
ec
o
r
d
s
r
ath
e
r
th
an
d
ata
co
llec
ted
f
r
o
m
a
s
in
g
le
clin
ical
in
s
ti
tu
tio
n
.
As
a
r
esu
lt,
d
etailed
clin
ical
p
r
o
v
en
an
ce
s
u
ch
as
h
o
s
p
ital
s
o
u
r
ce
o
r
co
m
p
r
eh
en
s
iv
e
p
atien
t
d
em
o
g
r
a
p
h
ic
d
is
tr
ib
u
tio
n
s
is
lim
ited
.
Nev
er
th
eless
,
th
e
in
clu
d
ed
v
ar
iab
les
co
r
r
esp
o
n
d
to
s
tan
d
ar
d
clin
ical
r
is
k
f
ac
to
r
s
u
s
ed
in
ca
r
d
io
v
ascu
lar
d
is
ea
s
e
ass
ess
m
en
t,
s
u
p
p
o
r
tin
g
th
e
r
ele
v
an
ce
o
f
th
e
d
ataset
f
o
r
p
r
ed
ict
iv
e
m
o
d
elin
g
.
T
h
e
d
ataset
co
n
tain
s
9
1
8
o
b
s
er
v
atio
n
s
an
d
1
2
v
ar
iab
les,
in
c
lu
d
in
g
th
e
b
in
ar
y
tar
g
et
v
ar
iab
le.
Of
th
e
9
1
8
o
b
s
er
v
atio
n
s
,
5
0
8
c
o
r
r
esp
o
n
d
t
o
p
atien
ts
with
h
ea
r
t d
is
ea
s
e
a
n
d
4
1
0
to
p
atien
ts
with
o
u
t h
ea
r
t d
is
ea
s
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
Hea
r
t d
is
ea
s
e
p
r
ed
ictio
n
th
r
o
u
g
h
fea
t
u
r
e
o
p
timiz
a
tio
n
,
en
s
e
mb
les,
a
n
d
th
r
esh
o
ld
c
a
lib
r
a
tio
n
(
Meh
d
i Mo
s
to
fi
)
2839
2
.2
.
Da
t
a
prepro
ce
s
s
ing
Prio
r
to
m
o
d
el
tr
ain
in
g
,
c
ateg
o
r
ical
v
ar
iab
les
wer
e
tr
a
n
s
f
o
r
m
ed
u
s
in
g
o
n
e
-
h
o
t
e
n
co
d
in
g
to
cr
ea
t
e
n
u
m
er
ical
r
ep
r
esen
tatio
n
s
s
u
itab
le
f
o
r
m
o
d
el
tr
ain
in
g
.
T
h
e
o
r
ig
in
al
d
ataset
co
n
tain
ed
n
o
m
is
s
in
g
v
alu
es.
Ho
wev
er
,
p
h
y
s
io
lo
g
ically
i
m
p
lau
s
ib
le
v
alu
es,
in
clu
d
in
g
ze
r
o
ch
o
lest
er
o
l
an
d
r
esti
n
g
b
lo
o
d
p
r
ess
u
r
e
m
ea
s
u
r
em
en
ts
an
d
n
e
g
ativ
e
Old
p
ea
k
v
al
u
es,
wer
e
i
d
en
tifi
ed
d
u
r
in
g
d
ata
s
cr
ee
n
in
g
an
d
r
ec
o
d
e
d
as
m
is
s
in
g
v
alu
es.
T
h
e
r
esu
ltin
g
d
ata
wer
e
th
en
p
r
ep
a
r
ed
f
o
r
s
u
b
s
eq
u
e
n
t
f
ea
tu
r
e
s
elec
tio
n
an
d
m
o
d
e
l
d
ev
elo
p
m
en
t
.
No
im
p
u
tatio
n
o
r
r
ec
o
r
d
d
eletio
n
was
p
er
f
o
r
m
ed
af
ter
th
is
r
ec
o
d
in
g
;
th
e
r
esu
ltin
g
m
is
s
in
g
v
a
lu
es
wer
e
r
etain
ed
an
d
h
a
n
d
led
th
r
o
u
g
h
th
e
n
ativ
e
m
is
s
in
g
-
v
alu
e
m
ec
h
a
n
is
m
s
o
f
th
e
tr
ee
-
b
ased
esti
m
ato
r
s
.
2
.
3
.
E
x
perim
ent
a
l
s
et
up
T
h
e
ex
p
e
r
im
en
tal
co
n
f
ig
u
r
ati
o
n
co
r
r
esp
o
n
d
in
g
to
th
e
wo
r
k
f
lo
w
in
Fig
u
r
e
1
is
d
escr
ib
e
d
b
elo
w
t
o
s
u
p
p
o
r
t
r
ep
r
o
d
u
cib
ilit
y
.
T
o
e
n
s
u
r
e
r
ep
r
o
d
u
cib
ilit
y
,
all
ex
p
er
im
en
ts
wer
e
co
n
d
u
cted
u
s
i
n
g
a
co
n
tr
o
lled
an
d
co
n
s
is
ten
t p
ip
elin
e.
a.
Data
s
p
li
tti
n
g
:
T
h
e
d
a
tas
et
was
d
i
v
i
d
e
d
i
n
t
o
tr
ai
n
i
n
g
(
8
0
%
)
a
n
d
test
in
g
(
2
0
%
)
s
ets
u
s
i
n
g
s
t
r
ati
f
ie
d
s
a
m
p
li
n
g
to
p
r
es
er
v
e
class
d
is
t
r
i
b
u
ti
o
n
.
All
p
r
ep
r
o
ce
s
s
in
g
,
f
ea
t
u
r
e
s
elec
tio
n
,
an
d
m
o
d
el
tr
ain
in
g
wer
e
p
er
f
o
r
m
ed
ex
clu
s
iv
ely
o
n
t
h
e
tr
ain
in
g
s
et,
wh
ile
th
e
test
s
et
was r
eser
v
ed
f
o
r
f
in
al
ev
alu
atio
n
.
b.
C
r
o
s
s
-
v
alid
atio
n
an
d
f
ea
tu
r
e
s
elec
tio
n
:
Featu
r
e
s
elec
tio
n
wa
s
p
er
f
o
r
m
e
d
u
s
in
g
R
FECV
wi
th
an
XGBo
o
s
t
esti
m
ato
r
an
d
5
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
,
o
p
tim
izin
g
f
o
r
r
ec
all.
T
h
is
p
r
o
ce
d
u
r
e
id
en
tifie
d
an
o
p
tim
al
s
u
b
s
et
o
f
1
4
p
r
e
d
icto
r
s
.
c.
Mo
d
el
d
ev
el
o
p
m
en
t
an
d
h
y
p
e
r
p
ar
am
eter
o
p
tim
izatio
n
:
T
h
r
ee
m
o
d
els
wer
e
e
v
alu
ated
:
r
a
n
d
o
m
f
o
r
est,
ex
tr
a
tr
e
es,
an
d
XGB
o
o
s
t.
H
y
p
e
r
p
a
r
am
e
te
r
t
u
n
in
g
was
co
n
d
u
ct
ed
u
s
i
n
g
R
a
n
d
o
m
ize
d
S
ea
r
c
h
C
V
wit
h
5
0
ite
r
ati
o
n
s
an
d
5
-
f
o
l
d
c
r
o
s
s
-
v
al
id
ati
o
n
,
o
p
ti
m
i
zi
n
g
r
ec
all
.
T
h
e
s
ea
r
ch
s
p
ac
e
in
clu
d
ed
k
e
y
p
ar
am
ete
r
s
s
u
ch
as
th
e
n
u
m
b
er
o
f
esti
m
ato
r
s
,
m
a
x
im
u
m
d
ep
th
,
lear
n
in
g
r
ate,
s
u
b
s
am
p
lin
g
r
atio
,
a
n
d
f
ea
tu
r
e
s
am
p
li
n
g
r
atio
.
d.
T
h
r
esh
o
ld
ca
lib
r
atio
n
:
A
class
if
icatio
n
th
r
esh
o
l
d
o
f
0
.
3
was
ap
p
lied
to
p
r
e
d
icted
p
r
o
b
a
b
ilit
ies
to
p
r
io
r
itize
r
ec
all.
T
h
e
d
ec
is
io
n
th
r
esh
o
ld
was
s
elec
ted
u
s
in
g
o
u
t
-
of
-
f
o
ld
p
r
ed
ictio
n
s
g
en
er
ate
d
f
r
o
m
t
h
e
tr
ain
in
g
d
ata
d
u
r
in
g
cr
o
s
s
-
v
alid
atio
n
.
T
h
r
es
h
o
ld
p
er
f
o
r
m
an
ce
was
th
e
r
ef
o
r
e
e
v
alu
ated
with
o
u
t
u
s
in
g
t
h
e
h
el
d
-
o
u
t
test
s
et,
wh
ich
was r
eser
v
ed
ex
clu
s
iv
ely
f
o
r
f
i
n
al
m
o
d
el
ass
ess
m
en
t
.
e.
Mo
d
el
ev
alu
atio
n
:
Per
f
o
r
m
a
n
c
e
was
ev
alu
ated
o
n
th
e
u
n
s
ee
n
test
s
et
u
s
in
g
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
ar
ea
u
n
d
e
r
th
e
R
OC
cu
r
v
e
(
AUC)
.
I
n
ad
d
itio
n
,
5
-
f
o
ld
c
r
o
s
s
-
v
alid
atio
n
was
u
s
ed
to
ass
ess
m
o
d
el
s
tab
ilit
y
,
with
th
e
E
x
tr
a
T
r
ee
s
m
o
d
el
ac
h
iev
in
g
a
r
ec
all
o
f
0
.
8
8
4
±
0
.
0
3
0
.
f.
I
m
p
lem
en
tatio
n
d
etails:
All
ex
p
er
im
en
ts
wer
e
im
p
lem
en
ted
in
Py
th
o
n
u
s
in
g
th
e
s
cik
it
-
lear
n
an
d
XGBo
o
s
t
lib
r
ar
ies,
with
a
f
ix
ed
r
an
d
o
m
s
ee
d
(
_
=
4
2
)
to
en
s
u
r
e
r
ep
r
o
d
u
cib
ilit
y
.
T
h
e
e
x
p
er
im
e
n
tal
p
ip
elin
e
was e
x
ec
u
ted
in
th
e
Go
o
g
le
C
o
lab
en
v
ir
o
n
m
en
t to
m
ain
tain
co
n
s
is
ten
t c
o
m
p
u
tatio
n
al
s
ettin
g
s
.
2
.
4
.
B
a
s
eline
m
o
del c
o
ns
t
ruct
io
n
T
o
estab
lis
h
a
r
ef
er
en
ce
p
e
r
f
o
r
m
an
ce
lev
el,
th
r
ee
e
n
s
em
b
le
lear
n
in
g
alg
o
r
ith
m
s
wer
e
in
itia
lly
tr
ain
ed
u
s
in
g
d
ef
au
lt
h
y
p
er
p
ar
am
eter
s
:
r
an
d
o
m
f
o
r
est,
ex
tr
a
tr
ee
s
,
X
GB
o
o
s
t.
T
h
ese
alg
o
r
ith
m
s
wer
e
s
elec
ted
b
ec
au
s
e
en
s
em
b
le
tr
ee
-
b
ased
m
o
d
els
h
av
e
d
em
o
n
s
tr
ated
s
tr
o
n
g
p
er
f
o
r
m
an
ce
in
s
tr
u
ctu
r
e
d
h
ea
lth
ca
r
e
d
atasets
d
u
e
to
th
eir
ab
ilit
y
to
ca
p
t
u
r
e
n
o
n
lin
e
ar
r
elatio
n
s
h
ip
s
an
d
in
ter
ac
tio
n
s
am
o
n
g
clin
ical
v
ar
iab
les.
2
.
5
.
F
e
a
t
ure
o
ptim
iz
a
t
io
n
(
l
a
y
er
1)
Featu
r
e
s
elec
tio
n
was
p
er
f
o
r
m
ed
u
s
in
g
th
r
ee
co
m
p
le
m
en
tar
y
tech
n
iq
u
es:
SelectFr
o
m
Mo
d
el,
p
er
m
u
tatio
n
im
p
o
r
tan
ce
,
an
d
R
FE
C
V.
T
h
e
R
FE
C
V
p
r
o
ce
d
u
r
e
was
u
s
ed
to
id
en
tify
a
n
o
p
tim
al
s
u
b
s
et
o
f
p
r
ed
ictiv
e
v
a
r
iab
les,
r
esu
ltin
g
in
a
r
ed
u
ce
d
f
ea
tu
r
e
s
et
co
n
tai
n
in
g
1
4
p
r
e
d
icto
r
s
.
2
.
6
.
M
o
del
o
ptim
iza
t
io
n (
la
y
er
2)
Af
ter
s
elec
tin
g
th
e
o
p
tim
al
f
e
atu
r
e
s
u
b
s
et,
th
e
e
n
s
em
b
le
m
o
d
els
wer
e
r
etr
ain
e
d
u
s
in
g
th
e
r
ed
u
ce
d
f
ea
tu
r
e
s
et.
T
o
f
u
r
th
er
im
p
r
o
v
e
p
r
ed
ictiv
e
p
er
f
o
r
m
an
ce
,
h
y
p
er
p
ar
a
m
eter
tu
n
in
g
was
co
n
d
u
cted
u
s
in
g
R
an
d
o
m
ized
Sear
ch
C
V.
T
h
e
tu
n
in
g
p
r
o
ce
s
s
ex
p
lo
r
ed
d
if
f
er
en
t
co
n
f
ig
u
r
atio
n
s
o
f
k
ey
m
o
d
el
p
ar
am
ete
r
s
,
in
clu
d
in
g
th
e
n
u
m
b
er
o
f
esti
m
ato
r
s
,
m
ax
im
u
m
tr
ee
d
ep
t
h
,
lear
n
in
g
r
ate,
an
d
s
u
b
s
am
p
lin
g
p
ar
am
ete
r
s
.
T
h
r
o
u
g
h
th
is
p
r
o
ce
s
s
,
th
e
m
o
d
els we
r
e
r
ef
in
ed
to
im
p
r
o
v
e
p
r
ed
ictiv
e
p
er
f
o
r
m
an
ce
an
d
g
en
er
aliza
tio
n
.
2
.
7
.
Dec
is
io
n
o
ptim
iza
t
io
n
(
l
a
y
er
3)
T
h
e
th
ir
d
o
p
tim
izatio
n
lay
er
f
o
cu
s
es
o
n
im
p
r
o
v
in
g
th
e
d
ec
is
io
n
-
m
ak
in
g
b
e
h
av
io
r
o
f
th
e
cl
ass
if
ier
.
I
n
m
ed
ical
s
cr
ee
n
in
g
a
p
p
licatio
n
s
,
m
in
im
izin
g
f
alse
n
e
g
ativ
es
is
cr
itical
b
ec
au
s
e
f
ailin
g
t
o
d
etec
t
a
p
atien
t
with
h
ea
r
t d
is
ea
s
e
m
ay
d
elay
t
r
ea
tm
en
t a
n
d
i
n
cr
ea
s
e
h
ea
lth
r
is
k
s
.
Fo
r
th
is
r
ea
s
o
n
,
th
e
p
r
o
p
o
s
ed
f
r
am
ewo
r
k
ad
o
p
ts
a
s
en
s
itiv
ity
-
f
o
cu
s
ed
th
r
esh
o
l
d
o
p
tim
izatio
n
to
alig
n
m
o
d
el
b
e
h
av
io
r
with
clin
ical
p
r
io
r
ities
.
I
n
s
tead
o
f
f
o
cu
s
i
n
g
s
o
lely
o
n
o
v
er
all
class
if
icatio
n
ac
cu
r
ac
y
,
th
e
m
o
d
el
ev
alu
atio
n
p
r
io
r
itizes
r
ec
all
(
s
en
s
itiv
ity
)
to
im
p
r
o
v
e
th
e
d
etec
tio
n
o
f
p
atien
ts
wh
o
ar
e
lik
ely
to
h
av
e
h
ea
r
t d
is
ea
s
e.
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.
16
,
No
.
5
,
Octo
b
e
r
20
26
:
2
8
3
6
-
2
8
4
6
2840
2
.
8
.
T
hresh
o
ld
ca
lib
ra
t
io
n
B
in
ar
y
class
if
ier
s
co
m
m
o
n
l
y
ass
ig
n
class
lab
els
u
s
in
g
a
p
r
o
b
a
b
ilit
y
cu
to
f
f
o
f
0
.
5
;
h
o
wev
er
,
th
is
co
n
v
en
tio
n
al
cu
to
f
f
m
ay
n
o
t
b
e
ap
p
r
o
p
r
iate
wh
e
n
s
en
s
itiv
ity
is
th
e
p
r
im
ar
y
clin
ical
o
b
j
ec
tiv
e
.
Ho
wev
er
,
in
m
ed
ical
s
cr
ee
n
in
g
task
s
th
is
th
r
esh
o
ld
m
ay
lead
to
m
is
s
ed
p
o
s
itiv
e
ca
s
e
s
,
wh
ich
is
u
n
d
esira
b
le
b
ec
au
s
e
f
ailin
g
to
d
etec
t h
ea
r
t d
is
ea
s
e
ca
n
d
elay
clin
ical
in
ter
v
en
tio
n
.
T
o
id
en
tify
an
ap
p
r
o
p
r
iate
d
ec
is
io
n
b
o
u
n
d
ar
y
,
a
th
r
esh
o
ld
s
wee
p
an
aly
s
is
wa
s
p
er
f
o
r
m
ed
b
y
ev
alu
atin
g
m
o
d
el
p
er
f
o
r
m
an
c
e
ac
r
o
s
s
th
r
esh
o
ld
s
r
an
g
in
g
f
r
o
m
0
.
1
t
o
0
.
8
.
Fo
r
ea
ch
th
r
esh
o
ld
,
r
ec
all
an
d
p
r
ec
is
io
n
wer
e
ca
lcu
lated
t
o
a
n
aly
ze
th
e
tr
ad
e
-
o
f
f
b
etwe
en
s
en
s
itiv
ity
an
d
f
alse p
o
s
itiv
e
p
r
ed
ictio
n
s
.
As
th
e
d
ec
is
io
n
b
o
u
n
d
a
r
y
was
lo
wer
ed
,
s
en
s
itiv
ity
in
c
r
ea
s
ed
,
alth
o
u
g
h
th
is
im
p
r
o
v
em
en
t
was
ac
co
m
p
an
ied
b
y
a
p
r
o
g
r
ess
iv
e
d
ec
lin
e
in
p
r
ec
is
io
n
.
A
t
h
r
esh
o
ld
o
f
0
.
3
p
r
o
v
id
ed
a
b
alan
ce
d
tr
ad
e
-
o
f
f
,
ac
h
iev
in
g
h
ig
h
r
ec
all
wh
ile
m
ain
tain
in
g
ac
ce
p
tab
le
p
r
ec
is
io
n
.
B
ased
o
n
th
is
r
esu
lt,
th
e
cla
s
s
if
icatio
n
th
r
esh
o
ld
was
ca
li
b
r
ated
to
0
.
3
in
o
r
d
e
r
to
p
r
i
o
r
itize
th
e
d
etec
tio
n
o
f
h
ig
h
-
r
is
k
p
atien
ts
.
T
h
is
ad
ju
s
tm
en
t
in
cr
ea
s
es
s
en
s
itiv
ity
an
d
r
e
d
u
ce
s
th
e
lik
elih
o
o
d
o
f
f
alse
n
eg
ativ
es,
wh
ich
is
p
ar
ticu
lar
ly
im
p
o
r
tan
t
in
m
ed
ical
s
cr
ee
n
in
g
ap
p
licatio
n
s
.
T
ab
le
1
p
r
esen
ts
th
e
ef
f
ec
t
o
f
d
if
f
er
en
t
th
r
esh
o
ld
v
alu
es
o
n
m
o
d
el
p
er
f
o
r
m
a
n
ce
,
illu
s
tr
atin
g
th
e
tr
a
d
e
-
o
f
f
b
etwe
en
r
ec
a
ll
an
d
p
r
ec
is
io
n
.
As
th
e
th
r
esh
o
l
d
d
ec
r
ea
s
es,
r
ec
all
im
p
r
o
v
es,
in
d
icatin
g
b
etter
d
etec
tio
n
o
f
p
o
s
itiv
e
ca
s
es,
wh
ile
p
r
ec
is
io
n
g
r
ad
u
ally
d
ec
lin
es
d
u
e
to
an
in
cr
ea
s
e
in
f
alse
p
o
s
itiv
es.
A
th
r
esh
o
ld
o
f
0
.
3
p
r
o
v
id
es
a
b
alan
ce
d
tr
a
d
e
-
o
f
f
,
ac
h
iev
in
g
h
ig
h
r
ec
all
wh
ile
m
ain
tain
in
g
ac
ce
p
tab
le
p
r
ec
is
io
n
.
T
h
is
r
esu
lt
s
u
p
p
o
r
ts
th
e
u
s
e
o
f
th
r
esh
o
ld
ca
lib
r
atio
n
as
an
ef
f
ec
tiv
e
s
t
r
ateg
y
f
o
r
p
r
io
r
itizin
g
s
en
s
itiv
ity
in
m
ed
ical
s
cr
ee
n
in
g
ap
p
licatio
n
s
.
T
ab
le
1
r
ep
o
r
ts
v
alid
atio
n
-
s
et
p
er
f
o
r
m
an
ce
u
s
ed
f
o
r
th
r
esh
o
l
d
s
elec
tio
n
,
wh
er
ea
s
th
e
f
in
al
h
eld
-
o
u
t
test
-
s
et
p
er
f
o
r
m
an
ce
o
b
tain
ed
a
f
ter
ap
p
ly
in
g
th
e
s
elec
ted
th
r
esh
o
ld
i
s
r
ep
o
r
ted
s
ep
ar
ately
in
T
ab
le
2
.
T
ab
le
1
.
E
f
f
ec
t o
f
t
h
r
esh
o
ld
v
a
lu
es o
n
m
o
d
el
p
er
f
o
r
m
a
n
ce
M
e
t
r
i
c
T=
0
.
1
T=
0
.
2
T=
0
.
3
T=
0
.
4
T=
0
.
5
T=
0
.
6
T=
0
.
7
T=
0
.
8
Th
r
e
s
h
o
l
d
0
.
1
0
.
2
0
.
3
0
.
4
0
.
5
0
.
6
0
.
7
0
.
8
R
e
c
a
l
l
0
.
9
4
0
.
9
1
0
.
9
0
0
.
8
9
0
.
8
4
0
.
8
4
0
.
8
2
0
.
8
2
P
r
e
c
i
s
i
o
n
0
.
8
2
0
.
8
5
0
.
8
5
0
.
8
8
0
.
8
9
0
.
8
9
0
.
8
9
0
.
9
1
T
ab
le
2
.
Fin
al
m
o
d
el
p
er
f
o
r
m
a
n
ce
af
ter
th
r
esh
o
ld
ca
lib
r
atio
n
M
o
d
e
l
A
c
c
u
r
a
c
y
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
AUC
R
a
n
d
o
m F
o
r
e
s
t
0
.
8
2
1
0
.
7
7
2
0
.
9
6
1
0
.
9
3
2
Ex
t
r
a
Tr
e
e
s
0
.
8
3
2
0
.
7
8
4
0
.
9
6
1
0
.
9
3
4
X
G
B
o
o
st
0
.
8
7
0
0
.
8
4
8
0
.
9
3
1
0
.
9
2
9
2
.
9
.
E
v
a
lua
t
i
o
n
m
et
rics
Mo
d
el
p
er
f
o
r
m
a
n
ce
was
ass
ess
ed
f
r
o
m
co
m
p
lem
en
tar
y
p
er
s
p
ec
tiv
es.
Acc
u
r
ac
y
r
ep
r
e
s
en
ted
th
e
p
r
o
p
o
r
tio
n
o
f
all
ca
s
es
class
if
i
ed
co
r
r
ec
tly
,
wh
er
ea
s
p
r
ec
is
io
n
q
u
a
n
tifie
d
th
e
r
eliab
ilit
y
o
f
p
o
s
itiv
e
p
r
ed
ictio
n
s
.
R
ec
all
m
ea
s
u
r
ed
th
e
p
r
o
p
o
r
ti
o
n
o
f
ac
tu
al
h
ea
r
t
-
d
is
ea
s
e
ca
s
es
s
u
cc
e
s
s
f
u
lly
d
etec
ted
an
d
was
tr
ea
ted
as
th
e
p
r
in
cip
al
s
cr
ee
n
in
g
-
o
r
ien
te
d
m
etr
ic.
AUC
s
u
m
m
ar
ized
th
e
ab
ilit
y
o
f
ea
ch
class
if
ier
to
d
is
tin
g
u
is
h
p
o
s
itiv
e
f
r
o
m
n
eg
ativ
e
ca
s
es
ac
r
o
s
s
p
o
s
s
ib
le
d
ec
is
io
n
th
r
esh
o
ld
s
.
R
OC
cu
r
v
es
wer
e
a
d
d
itio
n
ally
ex
am
in
e
d
to
ch
ar
ac
ter
ize
d
is
cr
im
in
atio
n
o
v
er
a
r
an
g
e
o
f
d
ec
is
io
n
b
o
u
n
d
ar
ies.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
is
s
ec
tio
n
ev
alu
ates
th
e
ef
f
ec
tiv
en
ess
o
f
th
e
p
r
o
p
o
s
ed
m
u
lti
-
s
tag
e
o
p
tim
izatio
n
f
r
am
e
wo
r
k
.
T
h
e
ex
p
er
im
en
ts
an
aly
ze
how
ea
ch
s
tag
e
o
f
th
e
p
ip
elin
e
—
in
clu
d
i
n
g
f
ea
tu
r
e
o
p
tim
izatio
n
,
m
o
d
e
l o
p
tim
izatio
n
,
an
d
th
r
esh
o
ld
ca
lib
r
atio
n
—
af
f
ec
ts
p
r
ed
ictiv
e
p
er
f
o
r
m
a
n
ce
.
T
h
r
ee
en
s
em
b
le
class
if
ier
s
,
r
an
d
o
m
f
o
r
est,
ex
tr
a
tr
ee
s
,
XGBo
o
s
t
,
wer
e
ev
alu
ated
u
s
in
g
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
a
n
d
AUC.
T
h
e
r
esu
lts
ar
e
p
r
ese
n
ted
p
r
o
g
r
ess
iv
ely
to
illu
s
tr
ate
p
er
f
o
r
m
a
n
ce
im
p
r
o
v
em
en
ts
ac
r
o
s
s
th
e
o
p
tim
izatio
n
s
tag
es.
3
.1
.
B
a
s
eline
m
o
del per
f
o
r
ma
nce
T
h
e
f
ir
s
t
e
x
p
er
im
e
n
t
e
v
alu
ates
th
e
b
aselin
e
p
e
r
f
o
r
m
an
ce
o
f
t
h
e
th
r
ee
en
s
em
b
le
m
o
d
els
tr
ai
n
ed
u
s
in
g
th
e
f
u
ll
f
ea
tu
r
e
s
et
an
d
d
ef
au
lt
h
y
p
er
p
a
r
am
eter
s
.
T
h
is
estab
lis
h
es
a
r
ef
er
en
ce
p
o
in
t
p
r
io
r
to
ap
p
ly
i
n
g
o
p
tim
izatio
n
tech
n
iq
u
es.
Fig
u
r
e
2
s
h
o
ws
th
at
r
a
n
d
o
m
f
o
r
est
an
d
ex
tr
a
tr
ee
s
ac
h
iev
e
s
im
ilar
ac
cu
r
ac
y
lev
els,
wh
ile
ex
tr
a
tr
ee
s
d
em
o
n
s
tr
ate
s
lig
h
tly
h
ig
h
e
r
r
ec
all.
XGBo
o
s
t
s
h
o
ws
co
m
p
etitiv
e
p
r
ec
is
i
o
n
b
u
t
lo
wer
r
ec
all
co
m
p
ar
ed
to
th
e
o
th
er
e
n
s
em
b
le
m
o
d
els.
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
Hea
r
t d
is
ea
s
e
p
r
ed
ictio
n
th
r
o
u
g
h
fea
t
u
r
e
o
p
timiz
a
tio
n
,
en
s
e
mb
les,
a
n
d
th
r
esh
o
ld
c
a
lib
r
a
tio
n
(
Meh
d
i Mo
s
to
fi
)
2841
Fig
u
r
e
2
.
B
aselin
e
m
o
d
el
p
er
f
o
r
m
an
ce
3
.
2
.
P
er
f
o
r
m
a
nce
a
f
t
er
f
ea
t
ure
o
ptim
iza
t
io
n
Featu
r
e
s
elec
tio
n
was
ap
p
lied
u
s
in
g
SelectFr
o
m
Mo
d
el,
p
er
m
u
tatio
n
im
p
o
r
tan
ce
,
an
d
R
FECV.
R
FEC
V
id
en
tifie
d
an
o
p
tim
al
s
u
b
s
et
o
f
1
4
f
ea
tu
r
es,
wh
ich
was
u
s
ed
to
r
etr
ain
th
e
m
o
d
el
s
.
T
ab
le
3
p
r
esen
ts
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
e
n
s
em
b
le
m
o
d
els
af
ter
f
ea
tu
r
e
o
p
tim
izatio
n
u
s
in
g
R
FECV,
s
h
o
win
g
th
e
im
p
ac
t
o
f
th
e
r
ed
u
ce
d
f
ea
t
u
r
e
s
et
o
n
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
AUC.
T
h
e
r
esu
lts
in
d
icate
th
a
t
f
ea
tu
r
e
s
elec
tio
n
s
lig
h
tly
im
p
r
o
v
es
o
v
er
all
p
e
r
f
o
r
m
a
n
ce
wh
ile
r
ed
u
cin
g
m
o
d
el
co
m
p
lex
ity
,
with
th
e
ex
tr
a
tr
ee
s
m
o
d
e
l
ac
h
iev
in
g
th
e
h
ig
h
est ac
cu
r
ac
y
an
d
r
ec
all
am
o
n
g
th
e
ev
alu
at
ed
m
o
d
els.
Featu
r
e
o
p
tim
izatio
n
s
lig
h
tly
i
m
p
r
o
v
e
d
o
v
er
all
p
er
f
o
r
m
a
n
ce
wh
ile
r
ed
u
cin
g
m
o
d
el
co
m
p
l
ex
ity
.
T
h
e
ex
tr
a
tr
ee
s
m
o
d
el
s
h
o
wed
th
e
lar
g
est
im
p
r
o
v
em
e
n
t,
ac
h
iev
in
g
th
e
h
ig
h
est
ac
cu
r
ac
y
(
0
.
8
8
6
)
an
d
r
ec
all
(
0
.
9
0
2
)
.
R
an
d
o
m
f
o
r
est
an
d
XGBo
o
s
t
ex
h
ib
ited
o
n
ly
m
in
o
r
ch
a
n
g
es,
in
d
icatin
g
th
at
th
e
r
ed
u
ce
d
f
ea
tu
r
e
s
u
b
s
et
p
r
eser
v
ed
m
o
s
t
o
f
th
e
p
r
e
d
ictiv
e
in
f
o
r
m
atio
n
.
Ov
er
all,
th
e
r
esu
lts
s
u
g
g
est
th
at
elim
in
atin
g
r
ed
u
n
d
a
n
t
f
ea
tu
r
es
ca
n
s
im
p
lify
th
e
m
o
d
el
wh
ile
m
ain
tain
in
g
s
tr
o
n
g
p
r
e
d
ictiv
e
p
er
f
o
r
m
an
ce
.
T
ab
le
3
.
Per
f
o
r
m
an
ce
af
ter
f
ea
tu
r
e
o
p
tim
izatio
n
(
R
FECV)
M
o
d
e
l
A
c
c
u
r
a
c
y
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
AUC
R
a
n
d
o
m F
o
r
e
s
t
0
.
8
7
5
0
.
8
8
3
0
.
8
9
2
0
.
9
3
2
Ex
t
r
a
Tr
e
e
s
0
.
8
8
6
0
.
8
9
3
0
.
9
0
2
0
.
9
3
7
X
G
B
o
o
st
0
.
8
5
9
0
.
8
8
0
0
.
8
6
3
0
.
9
1
9
3
.3
.
P
er
f
o
r
m
a
nce
a
f
t
er
mo
del o
ptim
iza
t
io
n
Hy
p
er
p
ar
a
m
eter
tu
n
in
g
was
p
er
f
o
r
m
e
d
u
s
in
g
R
an
d
o
m
ized
S
ea
r
ch
C
V
to
im
p
r
o
v
e
m
o
d
el
p
e
r
f
o
r
m
a
n
ce
b
y
ex
p
lo
r
in
g
d
if
f
er
en
t
c
o
m
b
i
n
atio
n
s
o
f
k
ey
p
ar
am
eter
s
s
u
ch
as
th
e
n
u
m
b
er
o
f
esti
m
ato
r
s
,
tr
ee
d
e
p
th
,
a
n
d
lear
n
in
g
r
ate.
Mo
d
el
o
p
tim
iza
tio
n
led
to
n
o
ticea
b
le
im
p
r
o
v
e
m
en
ts
in
p
r
ed
ictiv
e
p
er
f
o
r
m
a
n
ce
,
p
ar
ticu
lar
ly
f
o
r
th
e
en
s
em
b
le
m
o
d
els.
Af
ter
tu
n
in
g
,
b
o
th
ex
tr
a
tr
ee
s
an
d
XGBo
o
s
t
s
h
o
wed
im
p
r
o
v
ed
ac
cu
r
ac
y
an
d
r
ec
all
co
m
p
ar
ed
t
o
th
eir
b
aselin
e
co
n
f
ig
u
r
atio
n
s
.
T
h
e
ex
tr
a
tr
ee
s
c
lass
if
ier
ac
h
iev
ed
th
e
b
est
p
er
f
o
r
m
an
ce
,
r
ea
ch
in
g
a
r
ec
all
o
f
0
.
9
4
1
wh
ile
m
ai
n
tain
in
g
h
ig
h
o
v
er
all
ac
cu
r
a
cy
.
T
h
ese
r
esu
lts
in
d
icate
th
at
h
y
p
er
p
ar
a
m
eter
o
p
tim
izatio
n
en
a
b
les
en
s
em
b
l
e
m
o
d
els
to
b
etter
ca
p
tu
r
e
p
at
ter
n
s
in
th
e
clin
ical
d
ata,
im
p
r
o
v
in
g
t
h
e
d
etec
tio
n
o
f
p
atien
ts
with
h
ea
r
t
d
is
ea
s
e.
Fig
u
r
e
3
illu
s
tr
ates
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
e
n
s
em
b
le
m
o
d
els
af
ter
h
y
p
er
p
ar
am
eter
o
p
tim
izatio
n
,
h
ig
h
lig
h
tin
g
th
e
im
p
r
o
v
em
en
ts
in
ac
cu
r
ac
y
an
d
r
ec
all
ac
h
iev
ed
th
r
o
u
g
h
p
ar
am
eter
t
u
n
in
g
.
T
h
e
f
i
g
u
r
e
s
h
o
ws
th
at
th
e
ex
tr
a
tr
ee
s
a
n
d
XGBo
o
s
t
m
o
d
els
ex
h
i
b
it
n
o
ticea
b
le
g
ai
n
s
in
p
r
ed
ictiv
e
p
e
r
f
o
r
m
an
ce
co
m
p
a
r
ed
to
th
eir
b
aselin
e
co
n
f
ig
u
r
at
io
n
s
.
3
.
4
.
P
er
f
o
r
m
a
nce
a
f
t
er
t
hresh
o
ld ca
lib
ra
t
io
n
T
o
p
r
io
r
itize
s
en
s
itiv
ity
in
m
ed
ical
s
cr
ee
n
in
g
,
th
e
class
if
icatio
n
th
r
esh
o
ld
was
r
ed
u
ce
d
f
r
o
m
th
e
d
ef
au
lt
v
alu
e
o
f
0
.
5
to
0
.
3
.
T
h
is
ad
ju
s
tm
en
t
in
cr
ea
s
es
th
e
d
etec
tio
n
o
f
p
o
s
itiv
e
ca
s
es.
I
n
th
e
s
in
g
le
ev
alu
atio
n
s
p
lit,
th
e
ex
tr
a
tr
e
es
m
o
d
el
ac
h
iev
ed
a
r
ec
all
o
f
0
.
9
6
1
af
ter
th
r
esh
o
ld
ca
lib
r
atio
n
.
Ho
wev
er
,
b
o
o
ts
tr
ap
r
esam
p
lin
g
r
ev
ea
led
a
m
ea
n
r
ec
all
o
f
0
.
9
0
2
(
9
5
%
C
I
:
0
.
8
4
2
–
0
.
9
5
9
)
,
i
n
d
icatin
g
th
at
t
h
e
s
in
g
le
-
s
p
lit
r
esu
lt
r
ep
r
esen
ts
an
o
p
tim
is
tic
u
p
p
er
-
b
o
u
n
d
esti
m
ate.
T
ab
le
2
p
r
esen
ts
th
e
f
in
al
p
er
f
o
r
m
an
ce
o
f
th
e
en
s
em
b
le
m
o
d
els
af
ter
th
r
esh
o
l
d
ca
lib
r
atio
n
,
h
i
g
h
lig
h
tin
g
th
e
im
p
ac
t
o
f
th
e
ad
ju
s
ted
d
ec
is
io
n
t
h
r
e
s
h
o
ld
o
n
ac
c
u
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
AUC.
T
h
e
r
esu
lts
s
h
o
w
th
at
l
o
wer
in
g
th
e
th
r
esh
o
l
d
s
ig
n
if
ica
n
tly
i
n
c
r
ea
s
es
r
ec
all
ac
r
o
s
s
all
m
o
d
els,
with
ex
tr
a
tr
ee
s
an
d
r
an
d
o
m
f
o
r
est
ac
h
iev
in
g
th
e
h
ig
h
est
r
ec
all
o
f
0
.
9
6
1
,
wh
ile
m
ain
tain
in
g
ac
ce
p
tab
le
lev
els o
f
p
r
ec
is
io
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.
16
,
No
.
5
,
Octo
b
e
r
20
26
:
2
8
3
6
-
2
8
4
6
2842
Fig
u
r
e
3
.
Per
f
o
r
m
an
c
e
o
f
e
n
s
em
b
le
m
o
d
els af
ter
h
y
p
e
r
p
ar
am
eter
o
p
tim
izatio
n
,
illu
s
tr
atin
g
i
m
p
r
o
v
e
m
en
ts
in
ac
cu
r
ac
y
an
d
r
ec
all
co
m
p
ar
e
d
to
b
aselin
e
co
n
f
i
g
u
r
atio
n
s
,
wit
h
n
o
tab
le
g
ain
s
o
b
s
er
v
e
d
f
o
r
e
x
tr
a
t
r
ee
s
an
d
XGBo
o
s
t m
o
d
els
3
.5
.
RO
C
curv
e
a
na
ly
s
is
R
OC
an
aly
s
is
wa
s
u
s
ed
to
ex
am
in
e
h
o
w
well
th
e
th
r
ee
class
if
ier
s
s
ep
ar
ated
p
atien
ts
with
an
d
with
o
u
t
h
ea
r
t
d
is
ea
s
e
o
v
er
th
e
f
u
ll
r
an
g
e
o
f
class
if
icatio
n
cu
to
f
f
s
.
R
ath
er
th
an
ev
alu
ati
n
g
p
r
e
d
ictio
n
s
at
a
s
in
g
le
th
r
esh
o
ld
,
th
e
cu
r
v
es
s
u
m
m
ar
ize
th
e
r
elatio
n
s
h
ip
b
etwe
en
co
r
r
e
ctly
d
etec
tin
g
p
o
s
itiv
e
ca
s
es
an
d
in
co
r
r
ec
tly
lab
elin
g
n
e
g
ativ
e
c
ases
as p
o
s
itiv
e.
Fig
u
r
e
4
s
h
o
ws
th
at
all
th
r
ee
m
o
d
els
d
em
o
n
s
tr
ate
s
tr
o
n
g
d
is
cr
im
in
ativ
e
ca
p
ab
ilit
y
,
with
A
UC
v
alu
es
ex
ce
ed
in
g
0
.
9
2
.
R
an
d
o
m
f
o
r
e
s
t
an
d
ex
tr
a
tr
ee
s
ac
h
iev
e
n
e
ar
ly
id
en
tical
R
OC
cu
r
v
es
wi
th
AUC
v
alu
es
o
f
ap
p
r
o
x
im
ately
0
.
9
3
2
an
d
0
.
9
3
4
,
r
esp
ec
tiv
ely
,
wh
ile
XGBo
o
s
t
s
h
o
ws
a
s
l
ig
h
tly
lo
wer
AUC
o
f
0
.
9
2
9
.
XGBo
o
s
t
s
h
o
ws
a
s
lig
h
tly
lo
wer
AUC
o
f
0
.
9
2
2
,
alth
o
u
g
h
i
ts
p
er
f
o
r
m
an
ce
r
em
ain
s
s
tr
o
n
g
.
Ov
er
all,
th
e
R
O
C
an
aly
s
is
co
n
f
ir
m
s
th
at
en
s
em
b
le
lear
n
in
g
m
eth
o
d
s
p
r
o
v
id
e
ef
f
ec
tiv
e
class
if
icatio
n
p
er
f
o
r
m
an
ce
f
o
r
h
ea
r
t
d
is
ea
s
e
p
r
ed
ictio
n
.
T
h
e
o
b
s
er
v
ed
p
r
ec
is
io
n
–
r
ec
all
tr
ad
e
-
o
f
f
f
u
r
th
er
s
u
p
p
o
r
ts
th
e
ef
f
ec
tiv
en
ess
o
f
th
r
esh
o
l
d
ca
lib
r
atio
n
in
im
p
r
o
v
i
n
g
s
en
s
itiv
ity
wh
ile
m
ain
tain
in
g
ac
ce
p
t
ab
le
p
r
ec
is
io
n
.
Fig
u
r
e
4
.
R
OC
cu
r
v
e
co
m
p
ar
i
s
o
n
f
o
r
e
n
s
em
b
le
m
o
d
els
T
h
e
s
im
ilar
ity
o
f
th
e
th
r
ee
R
OC
p
r
o
f
iles
in
d
icate
s
th
at
th
eir
r
an
k
i
n
g
ab
ilit
y
r
em
ain
e
d
b
r
o
a
d
ly
co
m
p
ar
ab
le
ac
r
o
s
s
d
ec
is
io
n
b
o
u
n
d
ar
ies.
T
h
eir
h
i
g
h
AUC
v
alu
es
also
s
h
o
w
th
at
th
e
m
o
d
els
r
etain
e
d
s
u
b
s
tan
tial
d
is
cr
im
in
ato
r
y
ca
p
ac
ity
in
d
e
p
en
d
e
n
tly
o
f
t
h
e
s
elec
ted
o
p
e
r
atin
g
th
r
esh
o
l
d
.
T
h
e
s
u
b
s
eq
u
en
t
in
cr
ea
s
e
in
r
ec
all
af
te
r
th
r
esh
o
ld
ad
ju
s
tm
en
t
th
er
ef
o
r
e
r
esu
lted
p
r
im
ar
ily
f
r
o
m
ch
a
n
g
in
g
th
e
d
ec
is
io
n
r
u
le
r
ath
er
th
an
f
r
o
m
alter
in
g
t
h
e
u
n
d
er
ly
in
g
r
an
k
in
g
a
b
ilit
y
o
f
th
e
class
if
ier
s
.
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
Hea
r
t d
is
ea
s
e
p
r
ed
ictio
n
th
r
o
u
g
h
fea
t
u
r
e
o
p
timiz
a
tio
n
,
en
s
e
mb
les,
a
n
d
th
r
esh
o
ld
c
a
lib
r
a
tio
n
(
Meh
d
i Mo
s
to
fi
)
2843
3
.6
.
St
a
t
is
t
ica
l
v
a
lid
a
t
io
n
v
ia
bo
o
t
s
t
ra
p r
esa
m
pli
ng
T
h
e
s
tab
ilit
y
o
f
th
e
o
p
tim
ized
m
o
d
el
af
ter
th
r
esh
o
ld
a
d
ju
s
tm
en
t
was
f
u
r
th
er
ex
am
in
ed
th
r
o
u
g
h
b
o
o
ts
tr
ap
r
esam
p
lin
g
o
f
th
e
h
e
ld
-
o
u
t te
s
t d
ata
.
A
to
tal
o
f
2
,
0
0
0
b
o
o
ts
tr
ap
s
am
p
les we
r
e
g
e
n
er
ated
b
y
s
am
p
lin
g
with
r
ep
lace
m
en
t
f
r
o
m
th
e
t
est
d
ata.
Per
f
o
r
m
an
ce
m
etr
ic
s
wer
e
co
m
p
u
ted
f
o
r
ea
ch
r
e
s
am
p
led
d
ataset
to
esti
m
ate
v
ar
iab
ilit
y
an
d
co
n
s
t
r
u
ct
9
5
%
c
o
n
f
id
e
n
ce
in
ter
v
als.
Fig
u
r
e
5
p
r
esen
ts
b
o
o
ts
tr
ap
esti
m
ates
o
f
m
o
d
el
p
er
f
o
r
m
an
ce
with
9
5
% c
o
n
f
i
d
en
ce
in
ter
v
als.
T
h
e
b
o
o
ts
tr
ap
r
esu
lts
in
d
ica
te
th
at
th
e
m
o
d
el
p
er
f
o
r
m
a
n
ce
r
em
ain
s
s
tab
le
u
n
d
er
r
esam
p
lin
g
v
ar
iab
ilit
y
.
R
ec
all
r
em
ain
s
co
n
s
is
ten
tly
h
ig
h
,
s
u
p
p
o
r
tin
g
th
e
ef
f
ec
tiv
en
ess
o
f
th
e
s
en
s
itiv
ity
-
f
o
cu
s
ed
th
r
esh
o
ld
o
p
tim
izatio
n
.
R
ec
all
ac
h
iev
ed
a
b
o
o
ts
tr
ap
m
ea
n
o
f
0
.
9
0
2
(
9
5
% C
I
: 0
.
8
4
2
–
0
.
9
5
9
)
.
Fig
u
r
e
5
.
B
o
o
ts
tr
ap
p
er
f
o
r
m
an
ce
m
etr
ics with
9
5
% c
o
n
f
i
d
en
ce
in
ter
v
als
3
.7
.
Cro
s
s
-
v
a
lid
a
t
io
n sta
bil
it
y
a
na
ly
s
is
Mo
d
el
r
o
b
u
s
tn
ess
ac
r
o
s
s
alte
r
n
ativ
e
d
ata
p
ar
titi
o
n
s
was
f
u
r
th
er
ex
a
m
in
ed
u
s
in
g
s
tr
atif
ied
5
-
f
o
l
d
cr
o
s
s
-
v
alid
atio
n
.
Fo
r
ea
c
h
f
o
l
d
,
th
e
r
ec
all
s
co
r
e
was
co
m
p
u
ted
,
an
d
th
e
r
esu
lts
wer
e
s
u
m
m
ar
ized
as
m
ea
n
±
s
tan
d
ar
d
d
ev
iatio
n
ac
r
o
s
s
f
o
ld
s
.
T
h
e
cr
o
s
s
-
v
alid
atio
n
r
ec
all
s
co
r
es
wer
e
as
f
o
llo
ws:
R
an
d
o
m
Fo
r
est
ac
h
iev
ed
0
.
8
7
2
±
0
.
0
3
3
,
ex
tr
a
t
r
ee
s
ac
h
i
ev
ed
0
.
8
8
4
±
0
.
0
3
0
,
an
d
XGBo
o
s
t a
ch
iev
ed
0
.
8
6
5
±
0
.
0
2
8
.
T
h
e
r
elativ
ely
s
m
all
s
tan
d
ar
d
d
ev
iatio
n
s
in
d
icate
th
at
m
o
d
el
p
er
f
o
r
m
a
n
ce
r
em
ain
s
s
tab
le
ac
r
o
s
s
d
if
f
er
en
t
t
r
ain
in
g
–
v
alid
atio
n
p
ar
titi
o
n
s
.
Am
o
n
g
th
e
e
v
alu
ate
d
m
o
d
els,
e
x
tr
a
tr
ee
s
o
b
tain
ed
th
e
h
ig
h
est
av
e
r
ag
e
r
ec
all,
alth
o
u
g
h
th
e
d
if
f
e
r
en
ce
s
am
o
n
g
t
h
e
m
o
d
els
wer
e
r
elat
iv
ely
s
m
all.
T
h
is
s
u
g
g
ests
th
at
all
th
r
ee
e
n
s
em
b
le
m
eth
o
d
s
d
e
m
o
n
s
tr
ate
c
o
m
p
ar
ab
le
s
en
s
itiv
ity
wh
en
ev
a
lu
ated
u
n
d
er
cr
o
s
s
-
v
alid
atio
n
,
co
n
f
ir
m
in
g
th
e
r
o
b
u
s
tn
ess
o
f
th
e
p
r
ed
ictiv
e
f
r
am
ewo
r
k
.
3
.8
.
St
a
t
is
t
ica
l
mo
del c
o
m
pa
riso
n us
i
ng
M
cNe
m
a
r’
s
t
est
Pair
wis
e
Mc
Nem
ar
test
s
wer
e
u
s
ed
to
d
eter
m
in
e
wh
eth
er
th
e
class
if
ier
s
m
ad
e
m
ater
ially
d
if
f
e
r
en
t
er
r
o
r
s
o
n
t
h
e
s
am
e
h
eld
-
o
u
t
o
b
s
er
v
atio
n
s
.
B
ec
au
s
e
th
e
test
ev
alu
ates
d
is
ag
r
ee
m
en
ts
b
etwe
en
p
air
ed
p
r
ed
ictio
n
s
r
ath
er
th
a
n
co
m
p
a
r
in
g
ag
g
r
eg
ate
p
er
f
o
r
m
a
n
ce
m
ea
s
u
r
es,
it
p
r
o
v
id
es
a
d
ir
ec
t
s
tatis
tical
a
s
s
es
s
m
en
t
o
f
wh
eth
er
o
n
e
class
if
ier
p
r
o
d
u
ce
d
a
d
if
f
e
r
en
t
p
atter
n
o
f
c
o
r
r
ec
t
an
d
in
c
o
r
r
ec
t
class
if
icatio
n
s
f
r
o
m
an
o
th
e
r
.
T
ab
le
4
s
u
m
m
ar
izes th
e
r
esu
ltin
g
p
air
wis
e
co
m
p
ar
is
o
n
s
.
All p
air
wis
e
p
-
v
alu
es we
r
e
g
r
ea
ter
th
an
0
.
0
5
.
C
o
n
s
eq
u
en
tly
,
th
e
o
b
s
er
v
ed
d
if
f
er
en
ce
s
am
o
n
g
r
an
d
o
m
f
o
r
est,
ex
tr
a
t
r
ee
s
,
an
d
XGBo
o
s
t
d
id
n
o
t
p
r
o
v
id
e
s
u
f
f
icien
t
ev
id
en
ce
o
f
d
i
f
f
er
en
t
er
r
o
r
b
e
h
av
io
r
o
n
t
h
e
h
eld
-
o
u
t
s
am
p
le.
Alth
o
u
g
h
e
x
tr
a
t
r
ee
s
p
r
o
d
u
ce
d
s
lig
h
tly
s
tr
o
n
g
er
v
alu
es
f
o
r
s
o
m
e
p
er
f
o
r
m
a
n
ce
m
ea
s
u
r
es,
th
e
s
tatis
t
ical
co
m
p
ar
is
o
n
in
d
icate
s
th
at
th
is
ap
p
ar
en
t a
d
v
an
tag
e
co
u
ld
r
ea
s
o
n
ab
ly
ar
is
e
f
r
o
m
s
a
m
p
lin
g
v
a
r
iatio
n
.
T
ab
le
4
.
Mc
Nem
ar
test
r
esu
lts
f
o
r
p
ai
r
wis
e
co
m
p
ar
is
o
n
o
f
en
s
em
b
le
class
if
ier
s
RF
ET
XGB
RF
–
0
.
7
7
2
8
0
.
3
4
2
8
ET
0
.
7
7
2
8
–
0
.
4
7
9
5
XGB
0
.
3
4
2
8
0
.
4
7
9
5
–
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.
16
,
No
.
5
,
Octo
b
e
r
20
26
:
2
8
3
6
-
2
8
4
6
2844
C
o
llectiv
ely
,
th
e
f
in
d
in
g
s
in
d
i
ca
te
th
at
co
o
r
d
in
atin
g
f
ea
t
u
r
e
s
elec
tio
n
,
h
y
p
er
p
a
r
am
eter
o
p
ti
m
izatio
n
,
an
d
d
ec
is
io
n
-
th
r
esh
o
ld
ad
ju
s
tm
en
t
ca
n
en
h
an
ce
m
o
d
el
s
en
s
itiv
ity
wh
ile
p
r
eser
v
in
g
r
eliab
le
p
r
ed
ictiv
e
p
er
f
o
r
m
an
ce
.
Featu
r
e
o
p
tim
iz
atio
n
u
s
in
g
R
FECV
r
ed
u
ce
d
th
e
d
ataset
to
1
4
p
r
ed
icto
r
s
wh
ile
m
ain
tain
in
g
s
tr
o
n
g
p
r
e
d
ictiv
e
p
er
f
o
r
m
a
n
c
e.
T
h
e
lim
ited
ch
a
n
g
e
o
b
s
er
v
ed
in
R
an
d
o
m
Fo
r
est
an
d
X
GB
o
o
s
t
af
ter
f
ea
tu
r
e
r
ed
u
ctio
n
s
u
g
g
ests
th
at
th
ese
en
s
em
b
le
m
eth
o
d
s
ar
e
r
elativ
ely
r
o
b
u
s
t
to
r
ed
u
n
d
an
t v
ar
iab
le
s
.
I
n
co
n
tr
ast,
ex
tr
a
t
r
ee
s
s
h
o
wed
clea
r
er
im
p
r
o
v
e
m
en
ts
,
in
d
icatin
g
th
at
m
o
d
els
with
s
tr
o
n
g
e
r
r
a
n
d
o
m
izatio
n
m
ay
b
en
ef
it
m
o
r
e
f
r
o
m
ca
r
ef
u
lly
s
elec
ted
f
ea
tu
r
e
s
u
b
s
ets
.
Hy
p
e
r
p
ar
am
ete
r
tu
n
in
g
f
u
r
th
e
r
en
h
an
ce
d
m
o
d
el
s
en
s
itiv
ity
,
p
ar
ticu
lar
ly
f
o
r
th
e
e
n
s
em
b
le
m
eth
o
d
s
.
T
h
e
ex
tr
a
tr
ee
s
class
if
ier
ac
h
iev
ed
th
e
lar
g
est
r
e
ca
ll
im
p
r
o
v
e
m
en
t
,
h
ig
h
lig
h
tin
g
th
e
im
p
o
r
tan
ce
o
f
p
ar
am
eter
c
o
n
f
ig
u
r
atio
n
i
n
ca
p
tu
r
in
g
c
o
m
p
lex
n
o
n
lin
ea
r
p
atter
n
s
in
clin
ical
d
ata.
T
h
e
m
o
s
t
s
u
b
s
tan
tial
p
er
f
o
r
m
an
ce
ch
an
g
e
o
cc
u
r
r
ed
af
te
r
th
r
esh
o
ld
ca
lib
r
atio
n
.
W
h
ile
r
ec
all
r
ea
ch
ed
0
.
9
6
1
in
th
e
ca
lib
r
ated
ev
al
u
atio
n
s
p
lit,
b
o
o
ts
tr
ap
r
esam
p
l
in
g
an
d
cr
o
s
s
-
v
alid
atio
n
i
n
d
i
ca
ted
a
m
o
r
e
s
tab
le
r
ec
all
r
an
g
e
o
f
a
p
p
r
o
x
im
ately
0
.
8
8
–
0
.
9
0
.
T
h
is
d
if
f
e
r
en
ce
h
ig
h
lig
h
ts
h
o
w
a
s
in
g
le
f
a
v
o
r
ab
le
s
p
lit
m
ay
o
v
er
esti
m
ate
m
o
d
el
p
er
f
o
r
m
an
ce
,
wh
er
ea
s
r
ep
ea
ted
s
am
p
lin
g
p
r
o
v
id
es
a
m
o
r
e
r
elia
b
le
esti
m
ate.
T
h
e
co
n
s
is
ten
cy
b
etwe
en
b
o
o
ts
tr
a
p
an
d
cr
o
s
s
-
v
alid
atio
n
r
esu
lts
s
u
g
g
ests
th
at
th
e
s
en
s
itiv
ity
-
f
o
cu
s
ed
t
h
r
esh
o
ld
o
p
tim
izatio
n
p
r
o
d
u
ce
s
a
s
tab
le
im
p
r
o
v
e
m
en
t r
ath
e
r
th
an
a
s
p
lit
-
s
p
ec
if
ic
ef
f
ec
t.
Fin
ally
,
Mc
Nem
ar
’
s
test
in
d
icate
d
th
at
th
e
d
if
f
er
e
n
ce
s
am
o
n
g
th
e
ev
alu
ate
d
en
s
em
b
le
m
o
d
els
wer
e
n
o
t
s
tatis
tical
ly
s
ig
n
if
ican
t.
Alth
o
u
g
h
e
x
tr
a
tr
ee
s
ac
h
iev
ed
s
lig
h
tly
h
ig
h
er
r
ec
all
an
d
AU
C
v
alu
es,
th
e
th
r
ee
m
o
d
els d
em
o
n
s
tr
ated
s
tatis
tica
lly
co
m
p
ar
a
b
le
p
r
e
d
ictiv
e
ca
p
ab
ilit
y
wh
en
p
r
o
p
er
l
y
o
p
tim
iz
ed
.
Un
lik
e
p
r
io
r
s
tu
d
ies
th
at
ev
alu
ate
f
ea
tu
r
e
s
elec
tio
n
,
m
o
d
el
o
p
tim
izatio
n
,
an
d
th
r
esh
o
ld
ca
lib
r
atio
n
in
d
ep
en
d
en
tly
,
th
is
wo
r
k
d
e
m
o
n
s
tr
ates
th
at
th
eir
in
teg
r
at
io
n
with
in
a
u
n
if
ied
f
r
am
ew
o
r
k
lea
d
s
to
m
o
r
e
co
n
s
is
ten
t
an
d
clin
ically
m
ea
n
in
g
f
u
l
im
p
r
o
v
em
en
ts
in
s
en
s
itiv
ity
.
C
o
m
p
ar
ed
to
e
x
is
tin
g
ap
p
r
o
ac
h
es
th
at
p
r
im
ar
ily
f
o
c
u
s
o
n
ac
cu
r
ac
y
,
t
h
e
p
r
o
p
o
s
ed
f
r
am
ewo
r
k
em
p
h
asizes
s
en
s
itiv
ity
-
f
o
cu
s
ed
th
r
esh
o
ld
o
p
tim
izatio
n
,
wh
ich
is
m
o
r
e
s
u
itab
le
f
o
r
m
e
d
ical
s
cr
ee
n
in
g
ap
p
licatio
n
s
wh
er
e
m
in
im
izin
g
f
alse n
eg
ativ
es is
cr
itical.
T
h
ese
f
in
d
in
g
s
h
a
v
e
im
p
o
r
tan
t
im
p
licatio
n
s
f
o
r
clin
ical
d
ec
i
s
io
n
s
u
p
p
o
r
t
s
y
s
tem
s
,
wh
er
e
p
r
io
r
itizin
g
s
en
s
itiv
ity
ca
n
im
p
r
o
v
e
ea
r
l
y
d
etec
tio
n
o
f
h
ig
h
-
r
is
k
p
atien
ts
.
T
h
e
p
r
o
p
o
s
ed
f
r
am
ewo
r
k
ca
n
b
e
e
x
ten
d
e
d
to
o
th
er
m
ed
ical
p
r
ed
ictio
n
task
s
wh
er
e
ea
r
ly
d
etec
tio
n
is
ess
en
tial.
4.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
p
r
o
p
o
s
ed
a
m
u
l
ti
-
s
tag
e
o
p
tim
izatio
n
f
r
am
ew
o
r
k
f
o
r
h
ea
r
t
d
is
ea
s
e
p
r
ed
i
ctio
n
th
at
in
teg
r
ates
R
FECV
-
b
ased
f
ea
tu
r
e
s
elec
tio
n
,
en
s
em
b
le
m
o
d
e
l
h
y
p
e
r
p
ar
am
ete
r
tu
n
in
g
,
an
d
s
en
s
itiv
ity
-
f
o
cu
s
ed
th
r
esh
o
ld
o
p
tim
izatio
n
with
i
n
a
u
n
if
ied
m
ac
h
in
e
lear
n
in
g
p
ip
elin
e.
Fo
llo
win
g
th
r
esh
o
ld
ad
ju
s
tm
en
t,
th
e
f
r
am
ewo
r
k
ac
h
ie
v
ed
a
r
ec
all
o
f
u
p
to
0
.
9
6
1
o
n
th
e
ev
alu
atio
n
s
p
lit
wh
ile
r
etain
in
g
s
tr
o
n
g
class
-
d
is
cr
im
in
atio
n
ca
p
ab
ilit
y
.
Ad
d
itio
n
al
v
alid
at
io
n
u
s
in
g
b
o
o
ts
tr
ap
r
esam
p
li
n
g
an
d
cr
o
s
s
-
v
alid
atio
n
co
n
f
i
r
m
ed
s
tab
le
r
ec
all
p
er
f
o
r
m
an
ce
in
th
e
r
an
g
e
o
f
ap
p
r
o
x
im
ately
0
.
8
8
–
0
.
9
0
,
in
d
icatin
g
th
at
th
e
r
ec
all
-
f
o
cu
s
ed
o
p
tim
izatio
n
p
r
o
v
id
es
co
n
s
is
ten
t
im
p
r
o
v
e
m
en
ts
r
ath
er
th
an
is
o
lated
g
ain
s
.
Ov
er
all,
th
e
f
in
d
in
g
s
s
u
p
p
o
r
t
th
e
v
alu
e
o
f
co
m
b
in
in
g
f
ea
tu
r
e
r
ef
in
em
e
n
t,
en
s
em
b
le
-
m
o
d
el
o
p
tim
izati
o
n
,
an
d
th
r
esh
o
l
d
ad
ju
s
tm
en
t
to
p
r
o
d
u
ce
m
o
r
e
d
ep
en
d
a
b
le
ca
r
d
io
v
ascu
lar
r
is
k
p
r
ed
ictio
n
s
.
T
h
e
f
in
d
in
g
s
s
h
o
u
ld
b
e
in
ter
p
r
eted
in
lig
h
t
o
f
th
e
s
tu
d
y
’
s
r
elian
ce
o
n
o
n
e
s
tr
u
ctu
r
ed
d
ata
s
et,
wh
ich
lim
its
d
ir
ec
t
ex
ten
s
io
n
o
f
t
h
e
r
ep
o
r
te
d
p
er
f
o
r
m
a
n
ce
to
o
th
er
p
atien
t
p
o
p
u
latio
n
s
o
r
clin
ical
en
v
ir
o
n
m
en
ts
.
Fu
r
th
er
e
v
alu
atio
n
s
h
o
u
ld
th
er
ef
o
r
e
u
s
e
in
d
e
p
en
d
e
n
t
an
d
m
u
lti
-
in
s
titu
tio
n
al
d
ata.
Fu
tu
r
e
ex
ten
s
io
n
s
co
u
ld
also
in
co
r
p
o
r
ate
SHAP,
L
I
ME
,
o
r
r
elate
d
ex
p
lan
atio
n
m
et
h
o
d
s
to
ex
am
in
e
h
o
w
in
d
iv
id
u
al
p
r
ed
icto
r
s
in
f
lu
e
n
ce
m
o
d
el
o
u
tp
u
ts
an
d
t
o
im
p
r
o
v
e
th
e
tr
an
s
p
ar
en
c
y
o
f
th
e
f
r
a
m
e
wo
r
k
f
o
r
p
o
ten
tial c
lin
ical
u
s
e.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
e
au
th
o
r
s
d
ec
lar
e
th
at
n
o
e
x
ter
n
al
f
u
n
d
in
g
,
r
esear
ch
g
r
an
t,
o
r
r
esear
ch
c
o
n
tr
ac
t
was
r
e
ce
iv
ed
f
o
r
th
is
s
tu
d
y
.
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
Me
h
d
i M
o
s
to
f
i
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
T
o
u
r
aj
B
an
i
R
o
s
tam
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
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
Hea
r
t d
is
ea
s
e
p
r
ed
ictio
n
th
r
o
u
g
h
fea
t
u
r
e
o
p
timiz
a
tio
n
,
en
s
e
mb
les,
a
n
d
th
r
esh
o
ld
c
a
lib
r
a
tio
n
(
Meh
d
i Mo
s
to
fi
)
2845
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
d
ec
lar
e
n
o
co
n
f
lict o
f
in
ter
est.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
u
s
ed
in
t
h
is
s
tu
d
y
ar
e
p
u
b
licly
av
ailab
le
.
T
h
e
Hea
r
t
Failu
r
e
Pre
d
ictio
n
Data
s
et
s
u
p
p
o
r
tin
g
th
e
f
in
d
in
g
s
o
f
t
h
is
r
esear
ch
is
av
ailab
le
o
n
Kag
g
le
at:
h
ttp
s
:
//w
w
w
.
ka
g
g
le.
co
m/d
a
ta
s
ets/
fed
eso
r
ia
n
o
/h
ea
r
t
-
fa
ilu
r
e
-
p
r
ed
ictio
n
.
T
h
e
p
u
b
licl
y
ac
ce
s
s
ib
le
d
ataset
was u
s
ed
f
o
r
all
an
aly
s
es c
o
n
d
u
cted
i
n
t
h
is
s
tu
d
y
.
RE
F
E
R
E
NC
E
S
[
1
]
M
.
A
.
B
o
u
q
e
n
t
a
r
e
t
a
l
.
,
“
Ea
r
l
y
h
e
a
r
t
d
i
se
a
se
p
r
e
d
i
c
t
i
o
n
u
s
i
n
g
f
e
a
t
u
r
e
e
n
g
i
n
e
e
r
i
n
g
a
n
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
,
”
H
e
l
i
y
o
n
,
v
o
l
.
1
0
,
n
o
.
1
9
,
p
.
e
3
8
7
3
1
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
h
e
l
i
y
o
n
.
2
0
2
4
.
e
3
8
7
3
1
.
[
2
]
H
.
El
-
S
o
f
a
n
y
,
B
.
B
o
u
a
l
l
e
g
u
e
,
a
n
d
Y
.
M
.
A
.
El
-
La
t
i
f
,
“
A
p
r
o
p
o
s
e
d
t
e
c
h
n
i
q
u
e
f
o
r
p
r
e
d
i
c
t
i
n
g
h
e
a
r
t
d
i
se
a
se
u
s
i
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
a
n
d
a
n
e
x
p
l
a
i
n
a
b
l
e
A
I
met
h
o
d
,
”
S
c
i
e
n
t
i
f
i
c
Re
p
o
r
t
s
,
v
o
l
.
1
4
,
n
o
.
1
,
p
.
2
3
2
7
7
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
3
8
/
s4
1
5
9
8
-
0
2
4
-
7
4
6
5
6
-
2.
[
3
]
P
.
I
a
c
o
b
e
sc
u
,
V
.
M
a
r
i
n
a
,
C
.
A
n
g
h
e
l
,
a
n
d
A
.
D
.
A
n
g
h
e
l
e
,
“
Ev
a
l
u
a
t
i
n
g
b
i
n
a
r
y
c
l
a
ss
i
f
i
e
r
s
f
o
r
c
a
r
d
i
o
v
a
sc
u
l
a
r
d
i
sea
se
p
r
e
d
i
c
t
i
o
n
:
En
h
a
n
c
i
n
g
e
a
r
l
y
d
i
a
g
n
o
s
t
i
c
c
a
p
a
b
i
l
i
t
i
e
s,”
J
o
u
rn
a
l
o
f
C
a
r
d
i
o
v
a
sc
u
l
a
r
D
e
v
e
l
o
p
m
e
n
t
a
n
d
D
i
s
e
a
s
e
,
v
o
l
.
1
1
,
n
o
.
1
2
,
p
.
3
9
6
,
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
j
c
d
d
1
1
1
2
0
3
9
6
.
[
4
]
H
.
A
.
A
l
-
A
l
s
h
a
i
k
h
e
t
a
l
.
,
“
C
o
m
p
r
e
h
e
n
si
v
e
e
v
a
l
u
a
t
i
o
n
a
n
d
p
e
r
f
o
r
ma
n
c
e
a
n
a
l
y
si
s
o
f
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
i
n
h
e
a
r
t
d
i
sea
se
p
r
e
d
i
c
t
i
o
n
,
”
S
c
i
e
n
t
i
f
i
c
Re
p
o
r
t
s
,
v
o
l
.
1
4
,
n
o
.
1
,
p
.
7
8
1
9
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
8
-
024
-
5
8
4
8
9
-
7.
[
5
]
S
.
M
.
G
a
n
i
e
,
P
.
K
.
D
.
P
r
a
ma
n
i
k
,
a
n
d
Z.
Z
h
a
o
,
“
En
s
e
m
b
l
e
l
e
a
r
n
i
n
g
w
i
t
h
e
x
p
l
a
i
n
a
b
l
e
A
I
f
o
r
i
m
p
r
o
v
e
d
h
e
a
r
t
d
i
s
e
a
se
p
r
e
d
i
c
t
i
o
n
b
a
s
e
d
o
n
m
u
l
t
i
p
l
e
d
a
t
a
s
e
t
s
,
”
S
c
i
e
n
t
i
f
i
c
Re
p
o
r
t
s
,
v
o
l
.
1
5
,
n
o
.
1
,
p
.
1
3
9
1
2
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
3
8
/
s4
1
5
9
8
-
0
2
5
-
9
7
5
4
7
-
6.
[
6
]
C
.
M
.
B
h
a
t
t
,
P
.
P
a
t
e
l
,
T.
G
h
e
t
i
a
,
a
n
d
P
.
L.
M
a
z
z
e
o
,
“
Ef
f
e
c
t
i
v
e
h
e
a
r
t
d
i
sea
se
p
r
e
d
i
c
t
i
o
n
u
si
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
,
”
Al
g
o
r
i
t
h
m
s
,
v
o
l
.
1
6
,
n
o
.
2
,
F
e
b
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
a
1
6
0
2
0
0
8
8
.
[
7
]
T.
B
a
n
e
r
j
e
e
a
n
d
İ
.
P
a
ç
a
l
,
“
A
s
y
s
t
e
m
a
t
i
c
r
e
v
i
e
w
o
f
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
i
n
h
e
a
r
t
d
i
se
a
se
p
r
e
d
i
c
t
i
o
n
,
”
T
u
rk
i
s
h
J
o
u
rn
a
l
o
f
Bi
o
l
o
g
y
,
v
o
l
.
4
9
,
n
o
.
S
I
-
5
,
p
p
.
6
0
0
–
6
3
4
,
2
0
2
5
,
d
o
i
:
1
0
.
5
5
7
3
0
/
1
3
0
0
-
0
1
5
2
.
2
7
6
6
.
[
8
]
H
.
S
.
S
i
p
a
i
l
,
N
.
A
h
m
a
d
,
a
n
d
N
.
M
.
N
o
o
r
,
“
H
e
a
r
t
d
i
se
a
se
p
r
e
d
i
c
t
i
o
n
u
si
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
,
”
i
n
1
s
t
N
a
t
i
o
n
a
l
Bi
o
m
e
d
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
C
o
n
f
e
r
e
n
c
e
,
N
BE
C
2
0
2
1
,
2
0
2
1
,
p
p
.
4
8
–
5
2
.
d
o
i
:
1
0
.
1
1
0
9
/
N
B
E
C
5
3
2
8
2
.
2
0
2
1
.
9
6
1
8
7
5
3
.
[
9
]
A
.
T
i
w
a
r
i
,
A
.
C
h
u
g
h
,
a
n
d
A
.
S
h
a
r
ma
,
“
En
s
e
m
b
l
e
f
r
a
m
e
w
o
r
k
f
o
r
c
a
r
d
i
o
v
a
scu
l
a
r
d
i
se
a
se
p
r
e
d
i
c
t
i
o
n
,
”
C
o
m
p
u
t
e
r
s
i
n
Bi
o
l
o
g
y
a
n
d
Me
d
i
c
i
n
e
,
v
o
l
.
1
4
6
,
p
.
1
0
5
6
2
4
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
m
p
b
i
o
me
d
.
2
0
2
2
.
1
0
5
6
2
4
.
[
1
0
]
A
.
B
.
T
e
sh
a
l
e
e
t
a
l
.
,
“
A
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
i
m
p
r
o
v
e
s
r
i
sk
p
r
e
d
i
c
t
i
o
n
i
n
c
a
r
d
i
o
v
a
sc
u
l
a
r
d
i
s
e
a
s
e
,
”
G
e
ro
S
c
i
e
n
c
e
,
v
o
l
.
4
7
,
n
o
.
4
,
p
p
.
5
4
9
7
–
5
5
0
2
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
3
5
7
-
0
2
4
-
0
1
4
3
8
-
z.
[
1
1
]
T.
L
i
u
,
A
.
K
r
e
n
t
z
,
L.
L
u
,
a
n
d
V
.
C
u
r
c
i
n
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
b
a
s
e
d
p
r
e
d
i
c
t
i
o
n
m
o
d
e
l
s
f
o
r
c
a
r
d
i
o
v
a
s
c
u
l
a
r
d
i
s
e
a
se
r
i
s
k
u
s
i
n
g
e
l
e
c
t
r
o
n
i
c
h
e
a
l
t
h
r
e
c
o
r
d
s
d
a
t
a
:
s
y
st
e
mat
i
c
r
e
v
i
e
w
a
n
d
me
t
a
-
a
n
a
l
y
s
i
s,
”
E
u
ro
p
e
a
n
H
e
a
r
t
J
o
u
rn
a
l
-
D
i
g
i
t
a
l
H
e
a
l
t
h
,
v
o
l
.
6
,
n
o
.
1
,
p
p
.
7
–
2
2
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
9
3
/
e
h
j
d
h
/
z
t
a
e
0
8
0
.
[
1
2
]
C
.
H
.
C
h
e
n
g
,
B
.
J.
Le
e
,
O
.
N
.
N
f
o
r
,
C
.
H
.
H
si
a
o
,
Y
.
C
.
H
u
a
n
g
,
a
n
d
Y
.
P
.
Li
a
w
,
“
U
s
i
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
-
b
a
se
d
a
l
g
o
r
i
t
h
ms
t
o
c
o
n
st
r
u
c
t
c
a
r
d
i
o
v
a
sc
u
l
a
r
r
i
s
k
p
r
e
d
i
c
t
i
o
n
m
o
d
e
l
s
f
o
r
Ta
i
w
a
n
e
s
e
a
d
u
l
t
s
b
a
se
d
o
n
t
r
a
d
i
t
i
o
n
a
l
a
n
d
n
o
v
e
l
r
i
sk
f
a
c
t
o
r
s,”
BM
C
Me
d
i
c
a
l
I
n
f
o
rm
a
t
i
c
s
a
n
d
D
e
c
i
s
i
o
n
M
a
k
i
n
g
,
v
o
l
.
2
4
,
n
o
.
1
,
p
.
1
9
9
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
8
6
/
s1
2
9
1
1
-
0
2
4
-
0
2
6
0
3
-
2.
[
1
3
]
S
.
H
o
ssa
i
n
,
M
.
K
.
H
a
s
a
n
,
M
.
O
.
F
a
r
u
k
,
N
.
A
k
t
a
r
,
R
.
H
o
ssa
i
n
,
a
n
d
K
.
H
o
ssa
i
n
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
f
o
r
p
r
e
d
i
c
t
i
n
g
c
a
r
d
i
o
v
a
sc
u
l
a
r
d
i
se
a
se
i
n
B
a
n
g
l
a
d
e
s
h
:
e
v
i
d
e
n
c
e
f
r
o
m
a
c
r
o
s
s
-
sec
t
i
o
n
a
l
st
u
d
y
i
n
2
0
2
3
,
”
BM
C
C
a
rd
i
o
v
a
s
c
u
l
a
r
D
i
so
r
d
e
rs
,
v
o
l
.
2
4
,
n
o
.
1
,
p
.
2
1
4
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
8
6
/
s
1
2
8
7
2
-
0
2
4
-
0
3
8
8
3
-
2.
[
1
4
]
F
.
A
sad
i
,
R
.
H
o
ma
y
o
u
n
f
a
r
,
Y
.
M
e
h
r
a
l
i
,
C
.
M
a
sc
i
,
S
.
Ta
l
e
b
i
,
a
n
d
F
.
Za
y
e
r
i
,
“
D
e
t
e
c
t
i
o
n
o
f
c
a
r
d
i
o
v
a
s
c
u
l
a
r
d
i
s
e
a
se
c
a
s
e
s
u
s
i
n
g
a
d
v
a
n
c
e
d
t
r
e
e
-
b
a
s
e
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
,
”
S
c
i
e
n
t
i
f
i
c
Re
p
o
rt
s
,
v
o
l
.
1
4
,
n
o
.
1
,
p
.
2
2
2
3
0
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
3
8
/
s4
1
5
9
8
-
024
-
7
2
8
1
9
-
9.
[
1
5
]
R
.
K
u
m
a
r
e
t
a
l
.
,
“
A
c
o
m
p
r
e
h
e
n
s
i
v
e
r
e
v
i
e
w
o
f
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
f
o
r
h
e
a
r
t
d
i
s
e
a
se
p
r
e
d
i
c
t
i
o
n
:
c
h
a
l
l
e
n
g
e
s
,
t
r
e
n
d
s,
e
t
h
i
c
a
l
c
o
n
s
i
d
e
r
a
t
i
o
n
s,
a
n
d
f
u
t
u
r
e
d
i
r
e
c
t
i
o
n
s
,
”
Fr
o
n
t
i
e
r
s
i
n
A
rt
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
8
,
p
.
1
5
8
3
4
5
9
,
2
0
2
5
,
d
o
i
:
1
0
.
3
3
8
9
/
f
r
a
i
.
2
0
2
5
.
1
5
8
3
4
5
9
.
[
1
6
]
E.
S
a
k
y
i
-
Y
e
b
o
a
h
e
t
a
l
.
,
“
H
e
a
r
t
d
i
s
e
a
se
p
r
e
d
i
c
t
i
o
n
u
s
i
n
g
e
n
s
e
mb
l
e
t
r
e
e
a
l
g
o
r
i
t
h
ms:
A
s
u
p
e
r
v
i
se
d
l
e
a
r
n
i
n
g
p
e
r
s
p
e
c
t
i
v
e
,
”
A
p
p
l
i
e
d
C
o
m
p
u
t
a
t
i
o
n
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
S
o
f
t
C
o
m
p
u
t
i
n
g
,
v
o
l
.
2
0
2
5
,
n
o
.
1
,
p
.
1
9
8
9
8
1
3
,
2
0
2
5
,
d
o
i
:
1
0
.
1
1
5
5
/
a
c
i
s
/
1
9
8
9
8
1
3
.
[
1
7
]
T.
C
h
e
n
a
n
d
C
.
G
u
e
s
t
r
i
n
,
“
X
G
B
o
o
s
t
:
A
sca
l
a
b
l
e
t
r
e
e
b
o
o
st
i
n
g
sy
s
t
e
m
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
AC
M
S
I
G
K
D
D
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
K
n
o
w
l
e
d
g
e
D
i
sc
o
v
e
ry
a
n
d
D
a
t
a
Mi
n
i
n
g
,
A
u
g
.
2
0
1
6
,
v
o
l
.
1
3
-
17
-
A
u
g
u
st
-
2
0
1
6
,
p
p
.
7
8
5
–
7
9
4
.
d
o
i
:
1
0
.
1
1
4
5
/
2
9
3
9
6
7
2
.
2
9
3
9
7
8
5
.
[
1
8
]
T.
V
u
e
t
a
l
.
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
m
o
d
e
l
f
o
r
p
r
e
d
i
c
t
i
n
g
c
o
r
o
n
a
r
y
h
e
a
r
t
d
i
s
e
a
s
e
r
i
s
k
:
D
e
v
e
l
o
p
me
n
t
a
n
d
v
a
l
i
d
a
t
i
o
n
u
s
i
n
g
i
n
s
i
g
h
t
s
f
r
o
m
a
Ja
p
a
n
e
s
e
p
o
p
u
l
a
t
i
o
n
–
b
a
se
d
s
t
u
d
y
,
”
J
MIR
C
a
rd
i
o
,
v
o
l
.
9
,
p
.
e
6
8
0
6
6
,
2
0
2
5
,
d
o
i
:
1
0
.
2
1
9
6
/
6
8
0
6
6
.
[
1
9
]
G
.
G
u
l
e
t
a
l
.
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
e
n
s
e
mb
l
e
me
t
h
o
d
s
f
o
r
c
a
r
d
i
o
v
a
s
c
u
l
a
r
d
i
s
e
a
se
p
r
e
d
i
c
t
i
o
n
:
A
sy
s
t
e
m
a
t
i
c
r
e
v
i
e
w
o
f
a
p
p
r
o
a
c
h
e
s,
p
e
r
f
o
r
m
a
n
c
e
t
r
e
n
d
s,
a
n
d
r
e
sea
r
c
h
c
h
a
l
l
e
n
g
e
s,”
C
o
m
p
u
t
e
rs
,
v
o
l
.
1
5
,
n
o
.
1
,
p
.
2
5
,
2
0
2
6
,
d
o
i
:
1
0
.
3
3
9
0
/
c
o
m
p
u
t
e
r
s
1
5
0
1
0
0
2
5
.
[
2
0
]
M
.
A
.
N
a
s
e
r
,
A
.
A
.
M
a
j
e
e
d
,
M
.
A
l
sa
b
a
h
,
T.
R
.
A
l
-
S
h
a
i
k
h
l
i
,
a
n
d
K
.
M
.
K
a
k
y
,
“
A
r
e
v
i
e
w
o
f
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
’
s
r
o
l
e
i
n
c
a
r
d
i
o
v
a
sc
u
l
a
r
d
i
se
a
se
p
r
e
d
i
c
t
i
o
n
:
R
e
c
e
n
t
a
d
v
a
n
c
e
s
a
n
d
f
u
t
u
r
e
c
h
a
l
l
e
n
g
e
s,”
Al
g
o
r
i
t
h
m
s
,
v
o
l
.
1
7
,
n
o
.
2
,
p
.
7
8
,
F
e
b
.
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
a
1
7
0
2
0
0
7
8
.
[
2
1
]
S
.
H
.
P
a
r
k
,
J
.
C
h
o
i
,
a
n
d
J.
S
.
B
y
e
o
n
,
“
K
e
y
p
r
i
n
c
i
p
l
e
s
o
f
c
l
i
n
i
c
a
l
v
a
l
i
d
a
t
i
o
n
,
d
e
v
i
c
e
a
p
p
r
o
v
a
l
,
a
n
d
i
n
su
r
a
n
c
e
c
o
v
e
r
a
g
e
d
e
c
i
si
o
n
s
o
f
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
,
”
K
o
re
a
n
J
o
u
rn
a
l
o
f
Ra
d
i
o
l
o
g
y
,
v
o
l
.
2
2
,
n
o
.
3
,
p
p
.
4
4
2
–
4
5
3
,
2
0
2
1
,
d
o
i
:
1
0
.
3
3
4
8
/
k
j
r
.
2
0
2
1
.
0
0
4
8
.
[
2
2
]
M
.
B
a
i
r
y
,
K
.
C
h
a
d
a
g
a
,
N
.
S
a
mp
a
t
h
i
l
a
,
R
.
V
.
A
r
j
u
n
a
n
,
a
n
d
G
.
M
.
B
a
i
r
y
,
“
A
n
e
x
p
l
a
i
n
a
b
l
e
a
n
a
l
y
t
i
c
a
l
a
p
p
r
o
a
c
h
t
o
h
e
a
r
t
a
t
t
a
c
k
d
e
t
e
c
t
i
o
n
u
s
i
n
g
b
i
o
m
a
r
k
e
r
s
a
n
d
n
a
t
u
r
e
-
i
n
s
p
i
r
e
d
a
l
g
o
r
i
t
h
ms
,
”
H
e
a
l
t
h
c
a
re
An
a
l
y
t
i
c
s
,
v
o
l
.
8
,
p
.
1
0
0
4
0
7
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
h
e
a
l
t
h
.
2
0
2
5
.
1
0
0
4
0
7
.
Evaluation Warning : The document was created with Spire.PDF for Python.