I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
,
p
p
.
2410
~
2
4
2
2
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
15
.i
3
.
p
p
2
4
1
0
-
2
4
2
2
2410
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Ens
uring
ef
fect
iv
e ce
rv
ica
l cance
r
disea
ses
diag
no
sis
sy
stem
using
ensem
ble
m
a
chine learning
m
o
del
O
luwa
t
o
bi Ak
inl
a
de
1
,
J
a
f
a
r
Abdo
lla
hi
2
,
M
is
a
n P
a
ul E
t
ch
ie
3
,
Su
nd
a
y
Adeo
la
Aj
a
g
be
4
,
5
,
O
luwa
s
ey
i
O
m
o
t
a
y
o
Ala
bi
6
,
B
a
m
bo
Ay
o
Adey
a
nju
7
1
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
B
i
r
mi
n
g
h
a
m
C
i
t
y
U
n
i
v
e
r
si
t
y
,
B
i
r
mi
n
g
h
a
m
,
U
n
i
t
e
d
K
i
n
g
d
o
m
2
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
En
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
,
C
e
n
t
r
a
l
Te
h
r
a
n
B
r
a
n
c
h
,
I
sl
a
mi
c
A
z
a
d
U
n
i
v
e
r
si
t
y
,
Te
h
r
a
n
,
I
r
a
n
3
S
c
h
o
o
l
o
f
I
n
f
o
r
ma
t
i
c
s,
C
o
mp
u
t
i
n
g
a
n
d
C
y
b
e
r
S
y
s
t
e
ms,
N
o
r
t
h
e
r
n
A
r
i
z
o
n
a
U
n
i
v
e
r
s
i
t
y
,
F
l
a
g
st
a
f
f
,
U
n
i
t
e
d
S
t
a
t
e
s
4
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
U
n
i
v
e
r
s
i
t
y
o
f
Z
u
l
u
l
a
n
d
,
R
i
c
h
a
r
d
s
B
a
y
,
S
o
u
t
h
A
f
r
i
c
a
5
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
En
g
i
n
e
e
r
i
n
g
,
A
b
i
o
l
a
A
j
i
m
o
b
i
Te
c
h
n
i
c
a
l
U
n
i
v
e
r
si
t
y
,
I
b
a
d
a
n
,
N
i
g
e
r
i
a
6
D
e
p
a
r
t
me
n
t
o
f
M
e
c
h
a
n
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
,
L
e
a
d
C
i
t
y
U
n
i
v
e
r
si
t
y
,
I
b
a
d
a
n
,
N
i
g
e
r
i
a
7
I
n
t
e
r
n
a
t
i
o
n
a
l
B
u
s
i
n
e
ss Sc
h
o
o
l
,
U
n
i
v
e
r
si
t
y
o
f
Li
n
c
o
l
n
,
L
i
n
c
o
l
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
Au
g
22
,
2
0
2
4
R
ev
is
ed
Mar
3
,
2
0
2
6
Acc
ep
ted
Ap
r
22
,
2
0
2
6
Wo
rld
wi
d
e
,
c
e
rv
ica
l
c
a
n
c
e
r
is
a
m
a
jo
r
p
u
b
li
c
h
e
a
lt
h
c
o
n
c
e
rn
,
a
n
d
e
a
rly
d
e
tec
ti
o
n
is
e
ss
e
n
ti
a
l
fo
r
su
c
c
e
ss
fu
l
m
a
n
a
g
e
m
e
n
t
a
n
d
trea
tme
n
t.
En
se
m
b
le
m
a
c
h
in
e
lea
rn
in
g
(M
L)
h
a
s
e
m
e
r
g
e
d
a
s
a
p
ro
m
isin
g
a
p
p
ro
a
c
h
f
o
r
imp
ro
v
i
n
g
th
e
a
c
c
u
ra
c
y
a
n
d
re
li
a
b
il
it
y
o
f
c
e
rv
ica
l
c
a
n
c
e
r
d
iag
n
o
sis
sy
ste
m
s.
T
h
is
stu
d
y
e
v
a
lu
a
ted
t
h
e
p
e
rf
o
rm
a
n
c
e
o
f
v
a
rio
u
s
e
n
se
m
b
le
M
L
m
o
d
e
ls
fo
r
c
e
rv
ica
l
c
a
n
c
e
r
d
iag
n
o
sis
u
sin
g
a
larg
e
d
a
tas
e
t
o
f
c
e
rv
ica
l
c
e
ll
ima
g
e
s.
T
h
e
p
e
rfo
rm
a
n
c
e
o
f
d
iffere
n
t
e
n
se
m
b
l
e
m
o
d
e
ls
wa
s
c
o
m
p
a
re
d
,
i
n
c
lu
d
in
g
ra
n
d
o
m
fo
re
st
(R
F)
,
g
ra
d
ie
n
t
b
o
o
sti
n
g
(
G
B)
,
a
n
d
sta
c
k
in
g
,
with
c
o
n
v
e
n
ti
o
n
a
l
M
L
m
o
d
e
ls,
su
c
h
a
s
lo
g
isti
c
re
g
re
ss
io
n
a
n
d
s
u
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
(S
VM).
Th
e
b
a
g
g
i
n
g
e
n
se
m
b
le m
o
d
e
l
d
e
v
e
lo
p
e
d
re
p
o
rte
d
train
i
n
g
o
f
0
.
9
9
1
6
6
7
,
0
.
8
2
7
5
8
6
,
1
.
0
,
a
n
d
0
.
9
0
5
6
6
fo
r
a
c
c
u
ra
c
y
,
p
re
c
isio
n
,
re
c
a
ll
,
a
n
d
F
-
m
e
a
su
re
(F
1
-
sc
o
re
)
,
re
sp
e
c
ti
v
e
ly
.
F
u
rt
h
e
rm
o
re
,
t
h
e
in
terp
re
tab
il
it
y
o
f
e
n
se
m
b
le
m
o
d
e
ls
wa
s
in
v
e
stig
a
te
d
u
si
n
g
fe
a
tu
re
imp
o
rtan
c
e
a
n
d
p
a
rti
a
l
d
e
p
e
n
d
e
n
c
e
p
lo
ts.
Th
e
in
terp
re
tab
il
it
y
a
n
a
ly
sis
re
v
e
a
led
th
a
t
e
n
se
m
b
le
m
o
d
e
ls
c
a
n
p
r
o
v
i
d
e
v
a
lu
a
b
le
in
sig
h
ts
in
t
o
th
e
k
e
y
fe
a
tu
re
s
a
n
d
fa
c
to
rs
th
a
t
c
o
n
t
rib
u
te
to
c
e
rv
i
c
a
l
c
a
n
c
e
r
d
iag
n
o
sis.
I
n
c
o
n
c
l
u
sio
n
,
th
e
f
in
d
i
n
g
s
su
g
g
e
st
t
h
a
t
e
n
se
m
b
le
M
L
is
a
p
ro
m
isin
g
a
p
p
ro
a
c
h
fo
r
d
e
v
e
l
o
p
in
g
a
c
c
u
ra
te
a
n
d
re
li
a
b
le
c
e
rv
ica
l
c
a
n
c
e
r
d
iag
n
o
sis sy
ste
m
s a
n
d
imp
ro
v
i
n
g
th
e
in
ter
p
re
tab
il
it
y
o
f
t
h
e
se
m
o
d
e
l
s.
K
ey
w
o
r
d
s
:
C
er
v
ical
ca
n
ce
r
Dee
p
lear
n
in
g
Diag
n
o
s
is
E
n
s
em
b
le
m
o
d
els
Ma
ch
in
e
lear
n
in
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
:
Olu
wasey
i
Om
o
tay
o
Alab
i
Dep
ar
tm
en
t o
f
Me
ch
an
ical
E
n
g
in
ee
r
in
g
,
L
ea
d
C
ity
Un
iv
er
s
it
y
T
o
ll
-
Gate
Ar
ea
,
Of
f
Ob
a
Otu
d
ek
o
Av
en
u
e,
I
b
ad
an
,
Nig
er
ia
E
m
ail: a
lab
i.o
lu
wasey
i@
lcu
.
ed
u
.
n
g
1.
I
NT
RO
D
UCT
I
O
N
Ma
n
y
s
tu
d
ies
h
av
e
em
p
lo
y
ed
ar
tific
ial
in
tellig
en
ce
(
AI
)
te
ch
n
iq
u
es
as
to
o
ls
to
s
o
lv
e
h
ea
lth
ca
r
e
-
r
elate
d
is
s
u
es,
p
ar
ticu
lar
ly
,
a
n
d
th
e
p
er
f
o
r
m
an
c
e
h
as
b
ee
n
s
o
en
co
u
r
ag
in
g
.
Am
o
n
g
th
ese
s
tu
d
ies
was
h
ea
r
t
d
is
ea
s
e
p
r
ed
ictio
n
u
s
in
g
in
ter
n
et
o
f
th
in
g
s
(
I
o
T
)
-
en
a
b
led
m
ac
h
in
e
lear
n
in
g
(
ML
)
[
1
]
.
T
h
e
a
u
t
o
m
a
t
e
d
d
i
a
g
n
o
s
is
o
f
i
n
f
e
c
t
i
o
u
s
d
i
s
ea
s
es
u
s
i
n
g
t
h
e
I
o
T
-
e
n
a
b
l
e
d
A
I
m
o
d
e
l
h
a
s
o
f
f
e
r
e
d
s
u
s
t
a
i
n
a
b
l
e
p
r
o
g
n
o
s
i
s
a
n
d
r
e
d
u
c
e
d
d
i
a
g
n
o
s
is
e
r
r
o
r
s
s
i
g
n
i
f
i
c
a
n
t
l
y
w
i
t
h
i
m
p
r
o
v
e
d
a
c
c
u
r
a
c
y
[
2
]
.
T
h
e
in
teg
r
atio
n
o
f
I
o
T
-
e
n
ab
led
AI
ap
p
licatio
n
s
im
p
r
o
v
es
h
ea
lth
ca
r
e
d
eliv
er
y
in
v
ar
i
o
u
s
way
s
.
Alth
o
u
g
h
th
e
class
ical
d
ee
p
lear
n
in
g
(
DL
)
tech
n
iq
u
e
s
o
f
AI
in
ex
is
ten
ce
f
o
cu
s
o
n
a
u
to
m
atic
d
is
ea
s
e
d
iag
n
o
s
is
s
y
s
tem
s
wh
er
e
p
ar
am
eter
s
ettin
g
,
s
u
ch
as
lear
n
in
g
r
ate
s
elec
tio
n
an
d
weig
h
t
m
an
ag
em
e
n
t
,
ar
e
m
aj
o
r
co
n
ce
r
n
s
[
3
]
–
[
5
]
.
Me
an
w
h
ile,
lim
ited
atten
tio
n
is
p
aid
to
th
e
asp
ec
t
o
f
m
an
ag
em
en
t
u
s
in
g
au
t
o
m
ate
d
d
ev
ices
th
at
in
teg
r
ated
en
s
em
b
led
DL
-
b
ased
s
y
s
tem
s
an
d
I
o
T
co
m
p
u
tin
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
E
n
s
u
r
in
g
effec
tive
ce
r
vica
l c
a
n
ce
r
d
is
ea
s
es d
ia
g
n
o
s
is
s
yste
m
u
s
in
g
…
(
Olu
w
a
to
b
i A
kin
l
a
d
e
)
2411
m
o
d
els
f
o
r
r
ea
l
-
tim
e
d
is
ea
s
e
d
iag
n
o
s
is
.
T
h
e
tech
n
o
lo
g
ies
’
in
h
er
en
t
ad
v
a
n
tag
es
in
th
e
in
teg
r
atio
n
o
f
an
en
s
em
b
led
DL
-
b
ased
s
y
s
tem
an
d
I
o
T
co
m
p
u
tin
g
h
av
e
m
u
c
h
p
o
wer
t
o
d
iag
n
o
s
e
m
an
y
d
is
ea
s
es
at
a
tim
e
an
d
aid
in
th
e
m
a
n
ag
em
en
t
o
f
d
is
ea
s
e
d
iag
n
o
s
is
o
f
d
is
ea
s
es.
T
h
er
ef
o
r
e
,
th
is
r
esear
ch
aim
s
t
o
ev
alu
ate
en
s
em
b
le
ML
-
b
ased
s
y
s
tem
s
f
o
r
ce
r
v
ica
l c
an
ce
r
d
is
ea
s
e
d
iag
n
o
s
is
.
Sp
ec
if
ic
o
b
jectiv
es a
r
e
to
:
−
T
o
s
im
u
late
th
e
r
is
k
f
ac
to
r
s
o
f
ce
r
v
ical
ca
n
ce
r
d
is
ea
s
es u
s
in
g
en
s
em
b
le
ML
-
b
ased
m
o
d
els.
−
T
o
ev
alu
ate
th
e
en
s
em
b
le
ML
-
b
ased
m
o
d
els
u
s
in
g
v
a
r
io
u
s
s
tate
-
of
-
th
e
-
ar
t
m
etr
ics
lik
e
ac
cu
r
ac
y
,
tr
ain
in
g
tim
e,
s
en
s
itiv
ity
,
s
p
ec
if
icity
,
r
ec
all,
F
-
m
ea
s
u
r
e
(
F1
-
s
co
r
e
)
,
a
n
d
p
r
ec
is
io
n
.
T
h
is
p
ap
er
is
s
tr
u
ctu
r
ed
as
f
o
llo
ws.
Sectio
n
2
d
elin
ea
tes
th
e
liter
atu
r
e
r
ev
iew
p
er
tin
en
t
to
th
is
s
tu
d
y
.
Sectio
n
3
d
elin
ea
tes
th
e
m
eth
o
d
o
lo
g
y
,
s
ch
em
atic
f
r
am
ew
o
r
k
,
an
d
d
ata
p
r
ep
ar
atio
n
o
f
th
e
s
am
p
le
d
ataset,
an
d
elu
cid
ates
th
e
d
iag
n
o
s
is
m
o
d
el,
p
ar
am
eter
s
,
a
n
d
m
etr
ics
.
S
ec
tio
n
4
p
r
esen
ted
an
d
a
n
aly
ze
d
th
e
r
esu
lt
ex
p
er
im
en
tatio
n
.
Sectio
n
5
d
e
lin
ea
tes
th
e
co
n
clu
s
io
n
an
d
p
r
o
s
p
ec
tiv
e
en
d
ea
v
o
r
s
in
co
m
b
a
tin
g
ce
r
v
ical
ca
n
ce
r
b
y
en
s
em
b
le
ML
m
eth
o
d
o
lo
g
i
es.
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
T
h
e
f
o
u
r
th
m
o
s
t
p
r
ev
ale
n
t
d
is
ea
s
e
am
o
n
g
wo
m
e
n
wo
r
ld
wid
e,
ce
r
v
ical
ca
n
ce
r
is
a
s
ig
n
if
ic
an
t
ca
u
s
e
o
f
illn
ess
an
d
m
o
r
tality
.
Sy
s
t
em
s
f
o
r
d
iag
n
o
s
in
g
ce
r
v
ical
ca
n
ce
r
h
a
v
e
b
ee
n
d
e
v
elo
p
e
d
u
s
in
g
a
lo
t
o
f
ML
ap
p
r
o
ac
h
es.
T
h
e
ac
cu
r
ac
y
o
f
c
er
v
ical
ca
n
ce
r
d
iag
n
o
s
is
h
as im
p
r
o
v
e
d
b
ec
a
u
s
e
o
f
th
e
u
s
e
o
f
en
s
em
b
le
ML
.
R
ec
en
t
s
tu
d
ies
o
n
en
s
em
b
le
ML
f
o
r
ce
r
v
ical
ca
n
ce
r
d
i
ag
n
o
s
is
s
y
s
tem
s
wer
e
r
ev
iewe
d
in
th
e
liter
atu
r
e.
Nu
m
er
o
u
s
s
tu
d
ies
h
av
e
d
em
o
n
s
tr
ated
th
at
ML
ca
n
in
cr
ea
s
e
th
e
p
r
ec
is
io
n
o
f
s
y
s
tem
s
u
s
ed
to
d
iag
n
o
s
e
ce
r
v
ical
ca
n
ce
r
.
Fo
r
in
s
tan
ce
,
Ad
e
k
u
n
le
et
a
l.
[
6
]
cr
ea
ted
an
en
s
em
b
le
m
o
d
el
b
ased
o
n
r
an
d
o
m
f
o
r
est
(
R
F)
an
d
g
r
a
d
ien
t
b
o
o
s
tin
g
(
GB
)
tech
n
iq
u
es,
wh
ic
h
h
ad
a
9
4
.
6
%
ac
cu
r
ac
y
r
ate.
T
h
e
ac
cu
r
ac
y
o
f
a
n
en
s
em
b
le
m
o
d
el
s
tu
d
ied
,
wh
ich
co
m
b
in
ed
d
ec
is
io
n
tr
ee
(
DT
)
,
RF
,
an
d
s
u
p
p
o
r
t
v
ec
t
o
r
m
ac
h
in
e
(
SVM)
tech
n
iq
u
es,
was
9
5
.
5
%.
An
e
n
s
em
b
le
m
o
d
el
b
ased
o
n
th
e
ad
ap
tiv
e
b
o
o
s
tin
g
(
Ad
aBo
o
s
t
)
,
DT
,
an
d
K
-
n
ea
r
est
n
eig
h
b
o
r
s
’
alg
o
r
ith
m
s
was
cr
e
ated
[
7
]
an
d
ac
h
iev
ed
a
9
2
.
1
%
ac
cu
r
ac
y
r
ate.
Ad
d
itio
n
ally
,
s
o
m
e
s
tu
d
ies
h
av
e
lo
o
k
ed
in
to
th
e
ap
p
licatio
n
o
f
en
s
em
b
le
m
o
d
els
b
ased
o
n
DL
f
o
r
ce
r
v
ical
ca
n
ce
r
d
iag
n
o
s
is
s
y
s
tem
s
.
Fo
r
in
s
tan
ce
,
B
ah
ath
iq
et
a
l.
[
8
]
cr
ea
ted
an
en
s
em
b
le
m
o
d
el
b
as
ed
o
n
a
co
m
b
in
atio
n
o
f
r
ec
u
r
r
en
t
n
eu
r
al
n
etwo
r
k
(
R
NN)
an
d
co
n
v
o
lu
tio
n
al
n
e
u
r
al
n
etwo
r
k
(
C
NN)
alg
o
r
ith
m
s
,
an
d
it
ac
h
iev
ed
a
9
8
.
4
%
ac
cu
r
ac
y
r
ate.
T
h
e
au
th
o
r
also
e
x
p
lo
r
e
r
ec
en
t d
e
v
elo
p
m
en
ts
in
ML
-
b
ased
ce
r
v
ic
al
ca
n
ce
r
d
iag
n
o
s
is
s
y
s
tem
s
in
a
r
ev
iew
s
tu
d
y
.
T
h
e
r
ev
iew
em
p
h
asizes
en
s
em
b
le
ML
'
s
p
o
ten
tial
to
in
c
r
ea
s
e
th
e
p
r
ec
is
io
n
o
f
ce
r
v
i
ca
l
ca
n
ce
r
d
iag
n
o
s
is
.
T
h
e
ar
ticle
also
g
i
v
es
an
o
v
er
v
iew
o
f
th
e
p
er
f
o
r
m
an
ce
o
f
s
ev
er
al
en
s
em
b
le
m
o
d
els
cr
ea
ted
f
o
r
ce
r
v
ical
ca
n
ce
r
d
iag
n
o
s
is
.
B
a
h
ath
iq
et
a
l.
[
8
]
h
a
v
e
also
wr
itten
a
r
ev
iew
ar
ticle
th
at
f
o
cu
s
es
o
n
th
e
u
s
e
o
f
en
s
em
b
le
ML
f
o
r
th
e
class
if
icatio
n
o
f
ce
r
v
ical
ca
n
ce
r
ce
ll
s
.
T
h
e
p
ap
er
o
f
f
er
s
a
th
o
r
o
u
g
h
e
x
am
in
atio
n
o
f
m
u
ltip
le
en
s
em
b
le
m
o
d
els
c
r
ea
ted
f
o
r
th
e
d
etec
tio
n
o
f
ce
r
v
ical
ca
n
ce
r
,
alo
n
g
with
p
er
f
o
r
m
an
ce
in
d
icato
r
s
.
T
h
e
p
a
p
er
also
d
is
cu
s
s
es
th
e
d
if
f
ic
u
lties
in
cr
ea
tin
g
e
n
s
em
b
le
m
o
d
els
a
n
d
o
f
f
er
s
p
o
s
s
ib
le
s
o
lu
tio
n
s
.
C
asalla
s
et
a
l
[
9
]
c
o
n
d
u
cted
a
th
o
r
o
u
g
h
s
tu
d
y
a
n
d
m
eta
-
a
n
aly
s
is
to
ass
ess
th
e
ef
f
ec
tiv
e
n
ess
o
f
s
ev
er
al
ML
m
o
d
els,
in
clu
d
in
g
en
s
em
b
le
m
o
d
els,
f
o
r
ce
r
v
ical
ca
n
ce
r
d
iag
n
o
s
is
.
T
h
e
a
n
al
y
s
is
,
w
h
i
ch
lo
o
k
e
d
a
t
2
6
p
ap
e
r
s
on
en
s
em
b
l
e
m
o
d
els
p
er
f
o
r
m
e
d
b
ette
r
i
n
t
er
m
s
o
f
ac
cu
r
ac
y
an
d
s
en
s
i
ti
v
it
y
th
a
n
o
t
h
e
r
ML
m
o
d
els.
T
h
e
w
o
r
k
a
ls
o
h
i
g
h
li
g
h
ts
th
e
n
ee
d
f
o
r
a
d
d
iti
o
n
al
s
t
u
d
y
to
c
r
ea
t
e
tr
u
s
tw
o
r
t
h
y
e
n
s
e
m
b
le
m
o
d
e
ls
f
o
r
t
h
er
ap
e
u
tic
u
s
a
g
e
.
T
h
e
s
ta
te
o
f
M
L
-
b
ase
d
ce
r
v
ic
al
c
an
ce
r
d
iag
n
o
s
is
s
y
s
te
m
s
is
co
v
er
ed
i
n
a
r
ev
iew
,
wit
h
a
n
e
m
p
h
as
is
o
n
th
e
a
p
p
lic
ati
o
n
o
f
en
s
em
b
l
e
m
o
d
els
.
T
h
e
s
t
u
d
y
r
e
p
o
r
te
d
v
a
r
i
o
u
s
en
s
em
b
l
e
m
o
d
els
cr
ea
t
e
d
f
o
r
d
ia
g
n
o
s
i
n
g
ce
r
v
ica
l
ca
n
ce
r
a
n
d
th
e
ir
e
f
f
e
cti
v
e
n
ess
m
ea
s
u
r
es
.
T
h
e
r
e
v
ie
w
als
o
h
i
g
h
li
g
h
ts
th
e
p
o
te
n
ti
al
o
f
DL
-
b
as
ed
e
n
s
em
b
le
m
o
d
els
f
o
r
im
p
r
o
v
i
n
g
t
h
e
ac
c
u
r
ac
y
o
f
c
er
v
ic
al
ca
n
c
e
r
d
ia
g
n
o
s
is
.
T
h
e
u
s
e
o
f
e
n
s
e
m
b
le
ML
f
o
r
t
h
e
d
ete
cti
o
n
a
n
d
d
ia
g
n
o
s
i
s
o
f
c
e
r
v
ica
l
ca
n
c
er
i
n
l
o
w
-
r
eso
u
r
c
e
s
etti
n
g
s
is
t
h
e
s
u
b
je
ct
o
f
a
n
o
t
h
e
r
r
ev
iew
s
t
u
d
y
[
1
0
]
.
T
h
e
ar
ticle
g
av
e
a
n
in
v
esti
g
a
tiv
e
r
ep
o
r
t
o
f
s
ev
er
al
en
s
em
b
le
m
o
d
els
cr
ea
ted
f
o
r
ce
r
v
ical
ca
n
ce
r
d
iag
n
o
s
is
an
d
h
o
w
well
th
ey
wo
r
k
in
lo
w
-
r
eso
u
r
ce
en
v
ir
o
n
m
en
ts
.
T
h
e
p
ap
er
also
co
v
e
r
s
th
e
d
if
f
icu
lties
o
f
d
ev
elo
p
in
g
an
d
im
p
lem
en
tin
g
en
s
em
b
le
m
o
d
els
in
lo
w
-
r
eso
u
r
ce
e
n
v
ir
o
n
m
en
ts
.
S
y
s
tem
atic
r
ev
iew
an
d
m
eta
-
an
aly
s
is
ass
es
s
es
th
e
ef
f
icac
y
o
f
s
ev
e
r
al
ML
m
o
d
els,
in
clu
d
in
g
en
s
em
b
le
m
o
d
els,
f
o
r
ce
r
v
ical
ca
n
ce
r
d
etec
tio
n
[
1
1
]
.
T
h
e
r
esear
c
h
,
wh
ich
in
clu
d
ed
2
1
ex
p
e
r
im
e
n
ts
,
co
n
clu
d
ed
th
at
e
n
s
em
b
le
m
o
d
els
p
er
f
o
r
m
ed
b
etter
in
ter
m
s
o
f
ac
cu
r
ac
y
a
n
d
s
en
s
itiv
ity
th
an
o
th
e
r
ML
m
o
d
els.
T
h
e
ar
ticle
also
i
d
e
n
tifie
s
th
e
n
ee
d
f
o
r
f
u
r
th
er
r
esear
ch
to
d
ev
elo
p
r
o
b
u
s
t
an
d
r
eliab
le
en
s
em
b
le
m
o
d
els
f
o
r
clin
ical
u
s
e.
T
h
e
a
p
p
li
ca
ti
o
n
o
f
e
n
s
e
m
b
le
ML
f
o
r
t
h
e
ea
r
l
y
d
e
te
cti
o
n
o
f
ce
r
v
i
ca
l
c
an
ce
r
is
th
e
s
u
b
j
ec
t
o
f
a
r
ev
iew
p
a
p
e
r
[
1
2
]
.
T
h
e
p
ap
er
r
e
p
o
r
ted
th
e
n
u
m
er
o
u
s
en
s
em
b
le
m
o
d
els
th
at
h
av
e
b
ee
n
cr
ea
ted
f
o
r
d
iag
n
o
s
in
g
ce
r
v
ical
ca
n
ce
r
as
well
as
th
eir
p
er
f
o
r
m
an
ce
m
ea
s
u
r
es.
T
h
e
p
o
s
s
ib
ilit
y
o
f
en
s
em
b
le
m
o
d
els f
o
r
i
d
en
tify
in
g
p
r
ec
an
ce
r
o
u
s
l
esio
n
s
an
d
l
o
wer
in
g
ce
r
v
ical
ca
n
ce
r
in
cid
en
ce
is
a
ls
o
co
v
er
ed
in
t
h
e
ar
ticle.
T
h
e
im
p
o
r
ta
n
ce
o
f
en
s
em
b
le
ML
in
en
h
an
cin
g
th
e
in
ter
p
r
etab
ilit
y
a
n
d
e
x
p
lain
ab
i
lity
o
f
ce
r
v
ical
ca
n
ce
r
d
iag
n
o
s
is
s
y
s
tem
s
wa
s
co
v
er
ed
b
y
Kau
s
h
ik
et
a
l.
[
1
3
]
.
T
h
e
ar
ticle
g
av
e
a
g
e
n
er
al
o
v
er
v
iew
o
f
th
e
in
ter
p
r
etab
ilit
y
o
f
s
ev
er
al
e
n
s
em
b
le
m
o
d
e
ls
cr
ea
ted
f
o
r
th
e
d
etec
tio
n
o
f
ce
r
v
ical
ca
n
ce
r
.
T
h
e
r
e
v
iew
also
an
al
y
ze
s
th
e
d
if
f
icu
lties
with
en
s
em
b
le
m
o
d
els'
in
ter
p
r
etab
ilit
y
an
d
o
f
f
er
s
alter
n
ativ
e
r
e
m
ed
ie
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
4
1
0
-
2
4
2
2
2412
T
h
e
VGG1
9
(
tr
an
s
f
er
lear
n
in
g
o
r
T
L
)
m
o
d
el
an
d
co
lp
o
s
co
p
y
en
s
em
b
le
n
etwo
r
k
(
C
YE
NE
T
)
wer
e
th
e
two
DL
C
NN
ar
ch
itectu
r
es
s
u
g
g
ested
in
th
e
r
esear
ch
to
u
s
e
co
lp
o
s
co
p
y
p
ictu
r
es
to
d
iag
n
o
s
e
ce
r
v
ical
ca
n
ce
r
[
1
4
]
.
VGG1
9
was
u
s
ed
as
a
TL
f
o
r
th
e
s
tu
d
ies
in
th
e
C
NN
ar
ch
itectu
r
e.
A
u
n
iq
u
e
m
o
d
el
ca
lled
C
YE
NE
T
was
d
ev
elo
p
ed
to
a
u
to
m
atica
lly
id
en
tif
y
ce
r
v
ical
ca
n
ce
r
s
u
s
in
g
co
lp
o
s
co
p
y
im
a
g
es.
T
h
e
ac
c
u
r
ac
y
,
s
p
ec
if
icity
,
an
d
s
en
s
itiv
ity
o
f
th
e
d
ev
elo
p
ed
m
o
d
el
wer
e
a
s
s
es
s
ed
.
T
h
e
class
if
icatio
n
ac
cu
r
ac
y
f
o
r
VGG1
9
was
7
3
.
3
%.
T
h
e
o
u
tco
m
es
f
o
r
VGG1
9
(
T
L
)
wer
e
p
r
e
d
o
m
in
an
tly
f
av
o
r
a
b
le.
T
h
e
k
a
p
p
a
s
c
o
r
e
o
f
th
e
VGG1
9
m
o
d
el
in
d
icate
s
th
at
it
is
class
if
ied
with
in
th
e
m
o
d
er
ate
r
a
n
g
e.
T
h
e
ex
p
er
im
e
n
tal
r
esu
lts
in
d
icate
d
th
at
t
h
e
p
r
o
p
o
s
ed
C
YE
NE
T
h
ad
h
i
g
h
s
en
s
itiv
ity
,
s
p
ec
if
icity
,
an
d
k
ap
p
a
v
alu
es
o
f
9
2
.
4
%,
9
6
.
2
%,
an
d
8
8
%,
r
esp
ec
tiv
ely
.
T
h
e
class
if
icatio
n
ac
c
u
r
ac
y
o
f
C
YE
NE
T
m
o
d
el
h
as
r
is
en
to
9
2
.
3
%,
r
ep
r
esen
tin
g
a
1
9
%
im
p
r
o
v
em
e
n
t o
v
e
r
VGG1
9
(
T
L
)
m
o
d
el.
T
h
e
s
tu
d
y
p
r
esen
ted
a
n
in
g
e
n
io
u
s
ap
p
r
o
ac
h
f
o
r
em
p
lo
y
in
g
ML
alg
o
r
ith
m
s
to
f
o
r
ec
ast
ce
r
v
ical
ca
n
ce
r
.
T
h
e
p
r
o
p
o
s
ed
r
esear
ch
m
eth
o
d
o
l
o
g
y
co
m
p
r
is
es
f
o
u
r
s
tag
es:
d
ataset
ex
am
in
atio
n
,
d
ata
p
r
e
-
p
r
o
ce
s
s
in
g
,
p
r
ed
ictiv
e
m
o
d
el
s
elec
tio
n
(
PMS)
,
an
d
p
s
eu
d
o
-
c
o
d
e
d
ev
el
o
p
m
en
t.
T
h
e
PMS sectio
n
r
ec
o
r
d
s
ex
p
er
im
en
ts
with
v
ar
io
u
s
estab
lis
h
ed
ML
ap
p
r
o
ac
h
es,
in
clu
d
in
g
DT
s
,
lo
g
is
tic
r
eg
r
ess
io
n
,
SVM
s
,
K
-
n
ea
r
est
n
eig
h
b
o
r
s
alg
o
r
ith
m
s
,
RF
,
an
d
XGBo
o
s
t
.
I
n
th
e
co
n
te
x
t
o
f
ce
r
v
ical
ca
n
ce
r
p
r
ed
ictio
n
,
th
e
R
F,
DT
,
Ad
aBo
o
s
t
,
an
d
GB
alg
o
r
ith
m
s
g
et
th
e
h
i
g
h
est
class
if
icatio
n
s
co
r
e
o
f
1
0
0
%.
T
h
e
SVM
h
as
b
ee
n
f
o
u
n
d
to
ac
h
iev
e
9
9
%
ac
cu
r
ac
y
.
T
h
e
co
m
p
u
tatio
n
al
co
m
p
lex
ity
o
f
co
n
v
en
tio
n
al
ML
m
eth
o
d
s
is
ass
e
s
s
ed
to
m
ea
s
u
r
e
th
e
m
o
d
els'
ef
f
icac
y
[
1
5
]
.
3.
M
E
T
H
O
D
T
h
is
s
tu
d
y
u
s
ed
a
f
r
am
ewo
r
k
p
r
esen
ted
in
Fig
u
r
e
1
.
T
h
e
f
r
am
ewo
r
k
d
escr
ib
es
th
e
c
a
n
c
er
d
ataset
co
llectio
n
an
d
s
o
u
r
ce
,
th
e
d
ata
p
r
ep
r
o
ce
s
s
in
g
,
th
e
en
s
em
b
le
ML
m
o
d
els
b
u
ild
in
g
,
an
d
th
e
p
er
f
o
r
m
an
c
e
ev
alu
atio
n
f
o
r
th
e
d
e
v
elo
p
e
d
m
o
d
el
.
T
h
is
s
tu
d
y
estab
lis
h
es
an
ev
o
lu
tio
n
ar
y
e
n
s
em
b
le
lear
n
in
g
s
y
s
tem
u
tili
zin
g
s
tack
in
g
to
p
r
ed
ict
v
ar
io
u
s
ty
p
es
o
f
illn
ess
es.
B
ef
o
r
e
an
aly
s
is
,
th
e
d
ata
wer
e
s
an
itized
,
an
d
m
is
s
in
g
v
alu
es a
n
d
o
u
tlier
s
wer
e
id
en
ti
f
ied
an
d
r
ep
lace
d
with
m
ed
ian
v
alu
es.
Fig
u
r
e.
1
.
Sch
e
m
atic
o
f
th
e
m
o
d
el
p
r
o
p
o
s
ed
f
r
am
ewo
r
k
3
.
1
.
Da
t
a
s
et
3
.
1
.
1
.
Ris
k
f
a
c
t
o
rs c
er
v
ica
l c
a
ncer
da
t
a
s
et
info
r
m
a
t
io
n
T
h
e
d
ataset
o
f
ty
p
e
2
ce
r
v
ic
al
ca
n
ce
r
with
7
6
8
r
ec
o
r
d
s
a
n
d
n
in
e
u
s
ef
u
l
f
ac
to
r
s
was
u
s
ed
in
th
is
in
v
esti
g
atio
n
f
r
o
m
h
ttp
s
://ar
ch
iv
e.
ics.u
ci.
ed
u
/m
l/d
atasets
/C
er
v
ical+
ca
n
ce
r
+%2
8
R
is
k
+F
ac
to
r
s
%2
9
.
T
ab
le
1
co
m
p
r
is
es
o
f
th
e
d
escr
ip
tio
n
o
f
th
e
r
is
k
f
ac
to
r
s
o
f
ce
r
v
ical
c
an
ce
r
d
ataset
in
f
o
r
m
atio
n
an
d
T
ab
le
2
in
clu
d
es
a
lis
t
o
f
th
ese
f
ea
tu
r
es
an
d
th
e
a
ttrib
u
tes
o
f
th
e
d
ataset.
2
0
%
o
f
th
e
d
ata
is
u
s
ed
f
o
r
test
in
g
,
wh
ile
8
0
%
is
u
s
ed
f
o
r
tr
ain
in
g
.
T
h
e
d
ata
was
g
ath
er
ed
in
C
ar
ac
as,
Ven
ez
u
ela
at
th
e
Ho
s
p
ital
Un
iv
er
s
itar
io
d
e
C
ar
ac
as
.
T
h
e
co
llectio
n
co
n
tain
s
d
em
o
g
r
ap
h
ic
in
f
o
r
m
atio
n
,
r
o
u
tin
es,
an
d
p
r
io
r
m
ed
ical
r
ec
o
r
d
s
f
o
r
8
5
8
p
eo
p
le.
Sev
e
r
al
p
atien
ts
d
ec
lin
ed
to
r
e
p
ly
to
s
o
m
e
o
f
th
e
q
u
esti
o
n
s
o
u
t
o
f
r
es
p
ec
t f
o
r
t
h
eir
p
r
i
v
ac
y
[
1
6
]
.
T
ab
l
e
1
.
Descr
ip
tio
n
o
f
th
e
r
is
k
f
ac
to
r
s
ce
r
v
ical
ca
n
ce
r
d
ataset
in
f
o
r
m
atio
n
R
i
s
k
f
a
c
t
o
r
s c
e
r
v
i
c
a
l
c
a
n
c
e
r
d
a
t
a
s
e
t
D
a
t
a
s
e
t
S
a
mp
l
e
s
i
z
e
F
e
a
t
u
r
e
s
i
z
e
i
n
c
l
u
d
i
n
g
c
l
a
ss
l
a
b
e
l
C
l
a
s
ses
P
r
e
sen
c
e
o
f
a
mi
ssi
n
g
a
t
t
r
i
b
u
t
e
P
r
e
sen
c
e
o
f
mi
ssi
n
g
n
o
i
sy
R
i
s
k
f
a
c
t
o
r
o
f
c
e
r
v
i
c
a
l
c
a
n
c
e
r
d
a
t
a
set
8
5
8
36
2
Y
e
s
Y
e
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
E
n
s
u
r
in
g
effec
tive
ce
r
vica
l c
a
n
ce
r
d
is
ea
s
es d
ia
g
n
o
s
is
s
yste
m
u
s
in
g
…
(
Olu
w
a
to
b
i A
kin
l
a
d
e
)
2413
T
ab
le
2
.
T
h
e
attr
ib
u
te
o
f
th
e
c
er
v
ical
ca
n
ce
r
d
ataset
R
o
w
F
e
a
t
u
r
e
Ty
p
e
A
t
t
r
i
b
u
t
e
c
h
a
r
a
c
t
e
r
i
st
i
c
s
1
A
g
e
i
n
t
a
g
e
2
N
u
mb
e
r
o
f
se
x
u
a
l
p
a
r
t
n
e
r
s
i
n
t
3
F
i
r
st
s
e
x
u
a
l
i
n
t
e
r
c
o
u
r
se
i
n
t
4
N
u
m
o
f
p
r
e
g
n
a
n
c
i
e
s
i
n
t
5
S
mo
k
e
s
b
o
o
l
6
S
mo
k
e
s (
y
e
a
r
s)
b
o
o
l
y
e
a
r
s
7
S
mo
k
e
s (
p
a
c
k
s
/
y
e
a
r
)
b
o
o
l
p
a
c
k
s
/
y
e
a
r
8
H
o
r
mo
n
a
l
co
n
t
r
a
c
e
p
t
i
v
e
s
b
o
o
l
9
H
o
r
mo
n
a
l
c
o
n
t
r
a
c
e
p
t
i
v
e
s (
y
e
a
r
s)
i
n
t
y
e
a
r
s
10
I
U
D
b
o
o
l
11
I
U
D
(
y
e
a
r
s)
i
n
t
y
e
a
r
s
12
S
TD
s
b
o
o
l
13
S
TD
s (n
u
mb
e
r
)
i
n
t
n
u
m
b
e
r
14
S
TD
s:
c
o
n
d
y
l
o
m
a
t
o
si
s
b
o
o
l
15
S
TD
s:
c
e
r
v
i
c
a
l
c
o
n
d
y
l
o
ma
t
o
s
i
s
b
o
o
l
16
S
TD
s:
v
a
g
i
n
a
l
c
o
n
d
y
l
o
m
a
t
o
si
s
b
o
o
l
17
S
TD
s:
v
u
l
v
o
-
p
e
r
i
n
e
a
l
c
o
n
d
y
l
o
ma
t
o
s
i
s
b
o
o
l
18
S
TD
s:
s
y
p
h
i
l
i
s
b
o
o
l
19
S
TD
s:
p
e
l
v
i
c
i
n
f
l
a
mm
a
t
o
r
y
d
i
se
a
se
b
o
o
l
20
S
TD
s:
g
e
n
i
t
a
l
h
e
r
p
e
s
b
o
o
l
21
S
TD
s:
m
o
l
l
u
s
c
u
m c
o
n
t
a
g
i
o
s
u
m
b
o
o
l
22
S
TD
s:
A
I
D
S
b
o
o
l
23
S
TD
s:
H
I
V
b
o
o
l
24
S
TD
s:
H
e
p
a
t
i
t
i
s
B
b
o
o
l
25
S
TD
s:
H
P
V
b
o
o
l
26
S
TD
s:
nu
m
b
e
r
o
f
d
i
a
g
n
o
ses
i
n
t
27
S
TD
s:
ti
me
si
n
c
e
f
i
r
s
t
d
i
a
g
n
o
si
s
i
n
t
28
S
TD
s:
t
i
me
si
n
c
e
l
a
st
d
i
a
g
n
o
s
i
s
i
n
t
29
D
x
:
c
a
n
c
e
r
b
o
o
l
30
D
x
:
C
I
N
b
o
o
l
31
D
x
:
H
P
V
b
o
o
l
32
Dx
b
o
o
l
33
H
i
n
se
l
ma
n
n
:
t
a
r
g
e
t
v
a
r
i
a
b
l
e
b
o
o
l
34
S
c
h
i
l
l
e
r
:
t
a
r
g
e
t
v
a
r
i
a
b
l
e
b
o
o
l
35
C
y
t
o
l
o
g
y
:
t
a
r
g
e
t
v
a
r
i
a
b
l
e
b
o
o
l
36
B
i
o
p
s
y
:
t
a
r
g
e
t
v
a
r
i
a
b
l
e
b
o
o
l
3
.
1
.
2
.
At
t
ribute
info
rma
t
io
n
T
h
e
ce
r
v
ical
ca
n
ce
r
d
ataset
p
r
esen
ts
a
s
tr
u
ctu
r
ed
c
o
llectio
n
o
f
p
atien
t
-
le
v
el
clin
ical,
b
e
h
a
v
io
r
al,
an
d
d
iag
n
o
s
tic
attr
ib
u
tes
d
esig
n
ed
to
s
u
p
p
o
r
t
p
r
e
d
ictiv
e
m
o
d
elin
g
.
I
t
in
clu
d
es
a
m
ix
o
f
in
teg
er
an
d
b
o
o
lea
n
v
ar
iab
les
ca
p
tu
r
in
g
d
em
o
g
r
a
p
h
ics
(
e.
g
.
,
ag
e)
,
r
e
p
r
o
d
u
cti
v
e
h
is
to
r
y
(
e.
g
.
,
p
r
eg
n
a
n
cies)
,
life
s
ty
le
f
ac
to
r
s
(
e.
g
.
,
s
m
o
k
in
g
h
ab
its
)
,
co
n
tr
ac
ep
tiv
e
u
s
e,
an
d
d
etailed
s
ex
u
ally
tr
an
s
m
itted
d
is
ea
s
e
(
STD)
in
d
icato
r
s
.
Ad
d
itio
n
ally
,
tem
p
o
r
al
f
ea
tu
r
es
s
u
ch
as
y
ea
r
s
o
f
ex
p
o
s
u
r
e
an
d
tim
e
s
in
ce
d
iag
n
o
s
is
en
r
i
ch
th
e
d
ataset.
T
h
e
o
u
tco
m
e
v
a
r
iab
les
ar
e
r
ep
r
esen
ted
b
y
f
o
u
r
b
in
ar
y
tar
g
ets
—
Hin
s
elm
an
n
,
Sch
iller
,
C
y
to
lo
g
y
,
an
d
B
io
p
s
y
—
p
r
o
v
id
i
n
g
m
u
ltip
le
p
er
s
p
ec
tiv
e
s
f
o
r
ce
r
v
ical
ca
n
ce
r
d
etec
tio
n
an
d
class
if
icatio
n
task
s
.
3
.
2
.
Da
t
a
prepro
ce
s
s
ing
T
h
e
in
v
esti
g
atio
n
o
f
s
u
b
s
tan
tial
h
ea
lth
ca
r
e
d
atab
ases
m
a
y
b
e
in
ter
esti
n
g
u
s
in
g
th
e
k
n
o
wled
g
e
d
is
co
v
er
y
i
n
d
ata
b
ase
s
(
KDD)
ap
p
r
o
ac
h
.
B
ec
au
s
e
it
in
f
lu
en
c
es
th
e
q
u
ality
o
f
t
h
e
o
u
tco
m
es
p
r
o
d
u
ce
d
b
y
d
ata
m
in
in
g
tec
h
n
iq
u
es,
d
ata
p
r
e
-
p
r
o
ce
s
s
in
g
,
wh
ich
tak
es
u
p
ar
o
u
n
d
8
0
%
o
f
t
h
e
to
tal
p
r
o
ject
tim
e,
is
a
s
ig
n
if
ica
n
t
s
tep
in
th
e
KDD
p
r
o
ce
s
s
.
T
o
p
r
ep
ar
e
t
h
e
d
ata
f
o
r
an
ML
m
o
d
el
to
ac
ce
p
t,
d
ata
p
r
ep
r
o
ce
s
s
in
g
is
ess
en
tial.
I
t
is
en
s
u
r
ed
th
at
th
e
m
o
d
el
o
n
ly
l
ea
r
n
s
f
r
o
m
tr
ain
in
g
d
ata
an
d
th
at
test
in
g
d
ata
i
s
u
s
ed
to
ev
alu
ate
th
e
m
o
d
el'
s
p
er
f
o
r
m
an
ce
.
As
a
r
esu
lt,
tr
ai
n
in
g
an
d
test
d
ata
wer
e
s
ep
ar
ated
f
r
o
m
th
e
d
ataset
th
at
was
u
s
ed
.
7
0
%
o
f
th
e
d
ataset
is
m
ad
e
u
p
o
f
tr
ain
in
g
d
ata,
wh
ile
th
e
r
em
ain
in
g
3
0
%
is
s
p
lit
b
etwe
en
test
an
d
v
alid
atio
n
d
ata.
T
h
e
d
ata
was
in
itially
f
u
lly
s
cr
am
b
led
[
1
7
]
.
Pre
p
r
o
ce
s
s
in
g
th
e
d
ata
is
th
e
in
itial
s
tag
e
in
cr
ea
tin
g
an
ML
m
o
d
el.
R
ea
l
-
wo
r
ld
d
ata
is
f
r
eq
u
en
tly
in
ex
ac
t
in
ter
m
s
o
f
attr
ib
u
te
v
alu
es
an
d
tr
en
d
s
,
in
co
m
p
lete
,
in
co
n
s
is
ten
t,
an
d
in
ac
cu
r
ate
(
in
clu
d
in
g
er
r
o
r
s
o
r
o
u
tlier
s
)
.
T
h
is
is
wh
er
e
d
ata
p
r
ep
ar
atio
n
co
m
es
in
to
p
lay
:
it
o
r
g
an
izes,
clea
n
s
,
an
d
f
o
r
m
ats
th
e
r
aw
d
ata
s
o
th
at
ML
m
o
d
els
ca
n
u
s
e
it.
T
h
e
d
ata
m
u
s
t
b
e
p
r
o
p
er
l
y
p
r
ep
a
r
e
d
to
g
en
er
ate
b
etter
r
esu
lts
f
r
o
m
th
e
m
o
d
el
u
s
ed
i
n
ML
ap
p
licatio
n
s
.
So
m
e
ML
m
o
d
els
n
ee
d
th
e
d
ata
in
a
ce
r
tain
f
o
r
m
at,
s
o
n
u
ll
v
alu
es
f
r
o
m
th
e
in
itial
r
aw
d
at
a
s
et
m
u
s
t
b
e
p
r
o
ce
s
s
ed
b
ef
o
r
e
th
e
alg
o
r
ith
m
ca
n
b
e
ap
p
lied
[
1
8
]
.
T
h
e
way
th
e
d
ata
s
et
is
o
r
g
an
ized
s
h
o
u
ld
also
b
e
tak
en
in
to
ac
co
u
n
t
in
o
r
d
er
t
o
r
u
n
v
ar
io
u
s
ML
a
n
d
DL
alg
o
r
ith
m
s
s
im
u
ltan
eo
u
s
ly
an
d
s
elec
t th
e
b
est o
f
th
em
.
T
h
e
p
r
ep
r
o
ce
s
s
in
g
m
eth
o
d
s
u
s
ed
in
th
is
ar
ticle
ar
e
as f
o
llo
ws:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
4
1
0
-
2
4
2
2
2414
−
Han
d
lin
g
n
u
ll
v
alu
es:
th
er
e
ar
e
a
f
ew
n
u
ll
v
alu
es
in
e
v
er
y
r
e
al
-
wo
r
ld
d
ataset.
No
m
o
d
el
ca
n
m
an
ag
e
n
u
ll
o
r
n
o
t a
n
u
m
b
er
(
Na
N)
v
ar
ia
b
les o
n
its
o
wn
,
wh
eth
er
th
e
is
s
u
e
is
o
n
e
o
f
class
if
icatio
n
,
r
eg
r
ess
io
n
,
o
r
an
y
o
th
er
k
in
d
,
s
o
we
m
u
s
t in
ter
v
e
n
e.
−
Stan
d
ar
d
izatio
n
:
in
th
e
p
r
ep
r
o
ce
s
s
in
g
p
r
o
ce
s
s
,
th
is
is
a
cr
itical
s
tep
.
I
n
ML
,
d
ata
ar
e
s
t
an
d
ar
d
ized
b
y
ch
an
g
in
g
th
em
s
o
th
at
t
h
e
m
ea
n
is
0
an
d
t
h
e
s
tan
d
ar
d
d
ev
iati
o
n
is
1
.
3
.
3
.
B
uil
din
g
a
nd
t
ra
ini
ng
t
he
ens
em
b
le
m
a
chine le
a
rni
ng
m
o
del
T
o
f
o
r
ec
ast
d
if
f
er
en
t
ty
p
es
o
f
s
ick
n
ess
,
th
is
r
esear
ch
d
ev
el
o
p
s
an
ev
o
lu
tio
n
a
r
y
en
s
em
b
le
lear
n
in
g
s
y
s
tem
b
ased
o
n
s
tack
in
g
.
Prio
r
to
lear
n
in
g
,
th
e
d
ata
was
clea
n
ed
,
an
d
th
e
m
is
s
in
g
v
alu
es
an
d
o
u
tlier
s
wer
e
lo
ca
ted
an
d
im
p
u
ted
with
th
e
m
ed
ian
v
alu
es.
I
n
o
r
d
e
r
t
o
m
a
x
i
m
iz
e
cl
ass
if
ic
ati
o
n
ac
cu
r
ac
y
wh
ile
r
e
d
u
ci
n
g
en
s
em
b
l
e
co
m
p
le
x
i
ty
,
b
ase
le
ar
n
e
r
s
el
ec
ti
o
n
em
p
l
o
y
s
a
v
ar
i
ety
o
f
m
ac
h
i
n
e
-
l
ea
r
n
i
n
g
o
p
ti
m
iz
ati
o
n
t
ec
h
n
i
q
u
es
.
W
h
e
n
it
co
m
es
to
m
o
d
el
c
o
m
b
i
n
at
io
n
,
t
h
e
p
r
e
m
is
e
lea
r
n
e
r
s
'
p
r
e
d
ic
ti
o
n
s
ar
e
c
o
m
b
i
n
ed
u
s
i
n
g
a
m
et
a
-
class
if
ie
r
k
n
o
w
n
as
b
a
g
g
in
g
,
b
o
o
s
ti
n
g
,
a
n
d
A
d
aB
o
o
s
t
.
T
a
b
l
e
3
p
r
es
e
n
ts
t
h
e
h
y
p
e
r
p
a
r
a
m
et
e
r
t
u
n
i
n
g
M
L
al
g
o
r
it
h
m
s
u
s
ed
.
T
ab
le
3
.
Hy
p
er
p
ar
a
m
eter
tu
n
i
n
g
ML
alg
o
r
ith
m
s
Te
c
h
n
i
q
u
e
H
y
p
e
r
p
a
r
a
me
t
e
r
B
a
g
g
i
n
g
p
a
r
a
m
e
t
e
r
s =
{ 'ma
x
_
sam
p
l
e
s
':
[
0
.
7
,
0
.
8
,
0
.
9
,
1
]
,
'max
_
f
e
a
t
u
r
e
s':
[
0
.
7
,
0
.
8
,
0
.
9
,
1
]
,
'n
_
e
st
i
ma
t
o
r
s'
:
[
1
0
,
2
0
,
3
0
,
4
0
,
5
0
]
,
}
B
o
o
st
i
n
g
c
r
i
t
e
r
i
o
n
=
'f
r
i
e
d
ma
n
_
mse',
i
n
i
t
=
N
o
n
e
,
l
e
a
r
n
i
n
g
_
r
a
t
e
=
0
.
1
,
l
o
ss='
d
e
v
i
a
n
c
e
',
m
a
x
_
d
e
p
t
h
=
3
,
max
_
f
e
a
t
u
r
e
s=N
o
n
e
,
ma
x
_
l
e
a
f
_
n
o
d
e
s
=
N
o
n
e
,
mi
n
_
i
mp
u
r
i
t
y
_
s
p
l
i
t
=
1
e
-
0
7
,
mi
n
_
sam
p
l
e
s_
l
e
a
f
=
1
,
mi
n
_
sa
mp
l
e
s
_
s
p
l
i
t
=
2
,
m
i
n
_
w
e
i
g
h
t
_
f
r
a
c
t
i
o
n
_
l
e
a
f
=
0
.
0
,
n
_
e
st
i
ma
t
o
r
s
=
1
0
0
,
p
r
e
so
r
t
=
'a
u
t
o
',
r
a
n
d
o
m
_
s
t
a
t
e
=
N
o
n
e
,
S
u
b
s
a
m
p
l
e
=
1
.
0
,
v
e
r
b
o
se
=
0
,
w
a
r
m
_
s
t
a
r
t
=
F
a
l
se
,
b
a
s
e
_
e
st
i
ma
t
o
r
=
d
c
l
f
A
d
a
B
o
o
st
n
_
e
st
i
ma
t
o
r
s
=
5
0
,
l
e
a
r
n
i
n
g
_
r
a
t
e
=
0
.
5
,
a
l
g
o
r
i
t
h
m='S
A
M
M
E.
R
'
,
r
a
n
d
o
m
_
s
t
a
t
e
=1
T
h
e
s
ev
er
al
class
if
ier
s
th
at
m
ak
e
u
p
an
en
s
em
b
le
lear
n
in
g
s
y
s
tem
ar
e
wh
at
d
eter
m
in
e
its
s
u
cc
ess
.
I
t
is
im
p
o
s
s
ib
le
to
f
ix
a
p
o
ten
tial
er
r
o
r
if
all
class
if
ier
s
p
r
o
d
u
ce
th
e
s
am
e
r
esu
lts
.
T
h
ey
ar
e
th
er
ef
o
r
e
m
o
r
e
p
r
o
n
e
to
m
ak
e
v
ar
i
o
u
s
m
is
tak
es
o
n
v
ar
io
u
s
s
am
p
les.
I
f
ea
ch
class
if
ier
ex
h
i
b
its
a
d
i
f
f
er
en
t
er
r
o
r
,
y
o
u
ca
n
co
m
b
in
e
th
em
s
tr
ateg
ically
to
lo
wer
t
h
e
o
v
er
all
e
r
r
o
r
.
A
b
r
o
ad
s
e
t
o
f
class
if
ier
s
is
th
er
ef
o
r
e
r
eq
u
ir
ed
.
T
h
er
e
a
r
e
n
u
m
er
o
u
s
way
s
to
attain
th
is
d
iv
er
s
ity
,
as
s
h
o
wn
in
Fig
u
r
e
2
[
1
9
]
:
i)
t
r
ain
class
if
ier
s
u
s
in
g
a
v
ar
iety
o
f
tr
ain
in
g
d
atasets
,
ii)
u
s
e
s
ev
er
al
class
if
ier
s
'
v
ar
io
u
s
tr
ain
in
g
s
ettin
g
s
,
an
d
iii)
m
ak
e
u
s
e
o
f
s
ev
er
al
class
if
ier
s
.
Fig
u
r
e
2
.
Pro
p
er
m
et
h
o
d
s
o
f
e
n
s
em
b
le
lear
n
in
g
Ma
n
ag
er
s
to
d
ay
d
is
r
eg
ar
d
th
em
o
win
g
to
a
lack
o
f
in
f
o
r
m
a
tio
n
ab
o
u
t
h
o
w
to
u
s
e
th
e
v
ar
io
u
s
d
ata
av
ailab
le
to
u
s
.
O
n
th
e
o
t
h
er
h
an
d
,
i
f
th
ese
o
s
ten
s
ib
ly
u
n
im
p
o
r
tan
t
f
ac
ts
ar
e
d
elib
er
ate
ly
s
av
ed
an
d
later
m
in
ed
,
it
will
p
r
o
d
u
ce
a
wea
lth
o
f
k
n
o
wled
g
e
an
d
aid
in
m
a
n
ag
er
ial
d
ec
is
io
n
s
.
T
h
is
s
tu
d
y
em
p
lo
y
s
s
tatis
tical
an
d
p
r
o
b
ab
ilis
tic
m
eth
o
d
s
o
n
th
e
d
ataset,
alo
n
g
with
p
r
ep
r
o
ce
s
s
in
g
to
elim
in
ate
r
ed
u
n
d
an
t
an
d
m
is
s
in
g
d
ata,
to
ex
tr
ac
t
f
ea
tu
r
es
ex
h
i
b
itin
g
g
r
ea
ter
v
ar
ian
ce
i
n
d
iab
etes
c
o
m
p
licatio
n
s
.
A
g
en
etic
alg
o
r
ith
m
,
in
co
n
j
u
n
ctio
n
with
lo
g
is
tic
r
eg
r
ess
io
n
a
n
d
RF
,
is
u
tili
ze
d
to
i
d
en
tify
th
e
o
p
tim
al
f
ea
tu
r
es
f
o
r
ac
c
u
r
ate
d
iag
n
o
s
is
o
f
th
e
tar
g
et
class
[
1
9
]
.
Fig
u
r
e
3
s
h
o
ws th
e
p
r
o
p
o
s
ed
f
lo
wch
ar
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
E
n
s
u
r
in
g
effec
tive
ce
r
vica
l c
a
n
ce
r
d
is
ea
s
es d
ia
g
n
o
s
is
s
yste
m
u
s
in
g
…
(
Olu
w
a
to
b
i A
kin
l
a
d
e
)
2415
Fig
u
r
e
3
.
Su
g
g
ested
f
lo
wch
a
r
t
3
.
4
.
P
er
f
o
r
m
a
nce
m
e
a
s
urin
g
a
t
t
ribute
s
Py
th
o
n
was
th
e
p
r
o
g
r
am
m
in
g
lan
g
u
ag
e
u
tili
ze
d
f
o
r
th
e
s
tu
d
y
'
s
co
d
in
g
,
wh
ile
J
u
p
y
ter
No
teb
o
o
k
was
u
s
ed
f
o
r
im
p
lem
en
tatio
n
.
T
h
e
m
o
d
el
with
t
h
e
h
ig
h
est
d
e
g
r
ee
o
f
p
r
ed
ictab
ilit
y
was
ch
o
s
en
o
u
t
o
f
all
th
e
m
o
d
els
[
1
7
]
.
T
ab
le
4
p
r
esen
ts
th
e
co
n
f
u
s
io
n
m
atr
ix
f
o
r
t
h
e
m
o
d
el
ev
alu
atio
n
.
T
h
e
ter
m
s
f
alse
n
eg
ativ
e
(
FN)
an
d
f
alse
p
o
s
itiv
e
(
FP
)
d
en
o
te
th
e
q
u
an
tity
o
f
f
alse
n
eg
ativ
e
an
d
f
alse
p
o
s
i
tiv
e
s
am
p
les,
r
esp
ec
tiv
ely
.
T
h
e
n
u
m
b
e
r
o
f
g
e
n
u
in
e
n
e
g
ativ
e
o
r
tr
u
e
p
o
s
itiv
e
s
am
p
les
is
also
r
ep
r
esen
ted
b
y
th
e
t
r
u
e
n
eg
ativ
e
(
T
N)
an
d
t
r
u
e
p
o
s
itiv
e
(
T
P)
v
a
r
iab
les.
Sen
s
itiv
ity
g
au
g
es
h
o
w
m
a
n
y
co
r
r
ec
tly
s
tated
p
o
s
itiv
es
th
er
e
ar
e.
Ad
d
itio
n
ally
,
Sp
ec
if
icity
g
au
g
es
h
o
w
ac
cu
r
ately
n
eg
ativ
es
ar
e
d
is
cr
im
in
ated
ag
ain
s
t.
A
r
ea
u
n
d
er
th
e
cu
r
v
e
(
AUC)
is
a
m
etr
ic
u
s
ed
to
m
ea
s
u
r
e
th
e
ef
f
ec
tiv
en
ess
o
f
a
class
if
ier
.
T
h
e
F
-
m
ea
s
u
r
e
(
F1
-
s
co
r
e
)
also
d
em
o
n
s
tr
ated
th
e
b
in
ar
y
m
o
d
el'
s
ac
cu
r
ac
y
.
Ad
d
itio
n
ally
,
th
e
F1
-
s
co
r
e
was u
s
ed
to
est
im
ate
th
e
ef
f
icien
c
y
to
test
f
o
r
co
n
s
is
ten
cy
an
d
v
ar
iety
[
1
7
]
.
Du
e
to
th
e
n
u
m
er
o
u
s
e
v
alu
atin
g
th
e
co
r
r
ec
tn
ess
o
f
m
u
lti
-
class
alg
o
r
ith
m
s
n
ee
d
s
s
o
p
h
is
ticated
m
eth
o
d
o
lo
g
ies
d
u
e
t
o
th
e
s
ize
an
d
v
ar
iety
o
f
th
e
d
atasets
test
ed
,
th
e
r
an
g
e
o
f
tech
n
iq
u
es
u
s
ed
,
an
d
th
e
f
ea
t
u
r
es
o
f
th
e
d
ata
s
et,
wh
ich
in
clu
d
es
b
alan
ce
d
an
d
u
n
b
alan
c
ed
d
ata.
Acc
u
r
ac
y
,
s
en
s
itiv
ity
,
an
d
p
r
ess
u
r
e
ar
e
ju
s
t
a
f
ew
o
f
th
e
c
h
ar
ac
ter
is
tics
u
s
ed
to
g
au
g
e
h
o
w
well
th
ese
alg
o
r
ith
m
s
f
u
n
ctio
n
[
1
8
]
.
User
s
ca
n
ass
es
s
h
o
w
s
u
cc
ess
f
u
lly
a
d
ev
elo
p
ed
class
if
icatio
n
m
o
d
el
an
aly
ze
s
tex
tu
al
m
ater
ial
b
y
u
n
d
er
s
tan
d
in
g
t
h
ese
in
d
icato
r
s
.
T
r
ad
itio
n
ally
,
th
e
s
o
le
f
u
n
d
am
en
tal
cr
iter
io
n
o
f
ev
alu
at
io
n
an
d
g
e
n
er
alit
y
r
ep
o
r
ted
in
th
e
f
ield
o
f
m
u
lti
-
class
p
r
o
b
lem
s
is
th
e
ac
cu
r
a
cy
d
er
iv
e
d
f
r
o
m
th
e
class
if
icatio
n
[
1
8
]
with
th
e
f
o
llo
win
g
d
e
f
in
itio
n
:
a
d
d
itio
n
a
lly
,
in
v
er
s
e
m
ea
s
u
r
em
en
t is ex
p
lain
ed
.
I
n
th
is
s
ec
tio
n
,
we'
ll g
o
th
r
o
u
g
h
s
o
m
e
o
f
th
e
m
o
s
t u
s
ed
m
etr
ics f
o
r
ev
al
u
atin
g
m
o
d
els.
T
ab
le
4
.
C
o
n
f
u
s
io
n
m
atr
ix
C
o
n
f
u
s
i
o
n
ma
t
r
i
x
a
c
t
u
a
l
c
l
a
ss
C
l
a
s
si
f
i
e
d
a
s
:
n
e
g
a
t
i
ve
C
l
a
s
si
f
i
e
d
a
s
:
p
o
s
i
t
i
v
e
N
e
g
a
t
i
v
e
TN
FP
P
o
si
t
i
v
e
FN
TP
3
.
4
.
1
.
M
et
rics f
o
r
cla
s
s
if
ica
t
io
n
T
h
ese
f
o
u
r
o
u
tco
m
es
ar
e
c
o
m
m
o
n
l
y
p
lo
tted
o
n
a
co
n
f
u
s
io
n
m
atr
ix
.
T
h
e
s
ce
n
ar
i
o
o
f
b
in
ar
y
ca
teg
o
r
izatio
n
is
s
h
o
wn
in
t
h
e
co
n
f
u
s
io
n
m
atr
ix
b
elo
w.
T
h
is
m
atr
ix
wo
u
ld
b
e
co
n
s
tr
u
cted
af
ter
m
ak
in
g
p
r
ed
ictio
n
s
b
ased
o
n
y
o
u
r
te
s
t
d
ata
an
d
d
esig
n
atin
g
ea
c
h
p
r
ed
ictio
n
as
o
n
e
o
f
th
e
f
o
u
r
lik
ely
o
u
tco
m
es
m
en
tio
n
ed
ab
o
v
e
[
1
7
]
.
T
h
e
th
r
ee
p
r
im
ar
y
m
etr
ics
u
s
ed
to
ass
ess
a
class
if
icatio
n
m
o
d
el
ar
e
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
an
d
r
ec
all.
−
Acc
u
r
ac
y
:
ac
cu
r
ac
y
i
s
m
ea
s
u
r
ed
as
a
p
r
o
p
o
r
tio
n
o
f
co
r
r
e
ctly
p
r
ed
icted
v
alu
es
f
o
r
th
e
test
d
ata.
I
t'
s
s
im
p
le
to
ca
lcu
late
[
3
]
b
y
d
iv
i
d
in
g
th
e
n
u
m
b
er
o
f
ac
cu
r
ate
p
r
ed
ictio
n
s
b
y
t
h
e
to
tal
f
o
r
ec
asts
.
=
(
1
)
−
Pre
cisi
o
n
is
d
ef
in
ed
as
t
h
e
p
er
ce
n
tag
e
o
f
r
elev
a
n
t
ex
am
p
les
am
o
n
g
all
th
e
ex
a
m
p
les
ex
p
ec
ted
to
b
el
o
n
g
to
a
s
p
ec
if
ic
class
[
2
0
]
.
=
+
(
2
)
−
R
ec
all
is
d
ef
in
ed
as
th
e
p
er
c
en
tag
e
o
f
e
x
am
p
les
p
r
ed
icted
to
b
elo
n
g
to
a
class
d
iv
id
ed
b
y
th
e
to
ta
l
n
u
m
b
er
o
f
e
x
am
p
les th
at
ac
tu
a
lly
b
elo
n
g
to
th
e
class
[
2
0
]
.
=
+
(
3
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
4
1
0
-
2
4
2
2
2416
−
Sp
ec
if
icity
is
n
eg
ativ
e
in
cid
en
ce
s
ar
e
ex
p
r
ess
ed
as
a
s
h
ar
e
o
f
all
ac
tu
ally
n
eg
ativ
e
in
cid
en
ts
.
T
h
er
ef
o
r
e
,
th
e
to
tal
n
u
m
b
e
r
o
f
n
eg
ativ
e
ex
am
p
les
i
n
th
e
co
llectio
n
i
s
r
ep
r
esen
ted
b
y
th
e
n
u
m
er
a
to
r
(
T
N+
FP
)
.
Similar
to
r
ec
o
llectio
n
,
b
u
t
with
an
em
p
h
asis
o
n
n
eg
ativ
e
o
cc
u
r
r
en
ce
s
.
Ho
w
m
a
n
y
h
ea
lth
y
p
atien
ts
,
f
o
r
in
s
tan
ce
,
wer
e
in
f
o
r
m
ed
th
e
y
d
id
n
'
t
h
av
e
ca
n
ce
r
wh
en
t
h
ey
ac
tu
ally
d
id
?
I
t
is
an
e
x
am
in
atio
n
t
o
d
eter
m
in
e
h
o
w
d
is
tin
ct
th
e
cla
s
s
if
icatio
n
s
ar
e
[
2
1
]
,
[
2
2
]
.
=
TN
T
N+
F
P
(
4
)
−
F1
-
s
co
r
e
is
h
ar
m
o
n
io
u
s
,
r
ec
all
an
d
p
r
ec
is
io
n
a
r
e
eq
u
al.
B
ec
a
u
s
e
th
is
co
m
p
o
n
en
t
weig
h
s
b
o
th
,
th
e
g
r
ea
ter
th
e
F1
-
s
co
r
e,
th
e
b
etter
.
As
y
o
u
ca
n
s
ee
,
a
lo
w
n
u
m
er
ato
r
d
r
am
atica
lly
lo
wer
s
t
h
e
u
ltima
te
F1
-
s
co
r
e.
A
m
o
d
el
th
er
ef
o
r
e
p
er
f
o
r
m
s
well
in
th
e
F1
-
s
co
r
e
if
th
e
an
ticip
a
ted
p
o
s
itiv
es
ar
e
in
d
ee
d
p
o
s
itiv
e
(
p
r
ec
is
io
n
)
an
d
if
it
d
o
es
n
o
t
o
v
e
r
lo
o
k
p
o
s
itiv
es
an
d
f
o
r
ec
ast
th
em
n
eg
ativ
ely
(
r
ec
all)
[
2
3
]
–
[
2
5
]
.
T
h
e
J
u
p
y
ter
No
teb
o
o
k
p
latf
o
r
m
'
s
An
ac
o
n
d
a
en
v
ir
o
n
m
e
n
t
an
d
th
e
Py
th
o
n
p
r
o
g
r
a
m
m
in
g
lan
g
u
ag
e,
v
er
s
io
n
3
.
1
0
,
wer
e
u
s
ed
to
c
r
ea
te
th
is
ar
ticle.
T
h
e
im
p
lem
e
n
tatio
n
d
etails
ar
e
p
r
esen
ted
in
th
e
s
o
f
twar
e
r
e
q
u
ir
em
en
ts
as
s
h
o
wn
in
T
ab
le
5
.
T
ab
le
5
.
So
f
twar
e
r
eq
u
ir
e
m
en
t
s
D
i
st
r
i
b
u
t
i
o
n
A
n
a
c
o
n
d
a
n
a
v
i
g
a
t
o
r
a
n
d
G
o
o
g
l
e
C
o
l
a
b
P
a
c
k
a
g
e
s
M
a
t
p
l
o
t
l
i
b
,
N
u
mP
y
,
P
a
n
d
a
s
,
S
c
i
-
I
T
l
e
a
r
n
La
n
g
u
a
g
e
P
y
t
h
o
n
3
.
1
0
I
D
E
Ju
p
y
t
e
r
N
o
t
e
b
o
o
k
(
G
o
o
g
l
e
C
o
l
a
b
)
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
4
.
1
.
Ris
k
f
a
ct
o
rs c
er
v
ica
l c
a
ncer
da
t
a
s
et
info
rma
t
io
n
I
n
o
r
d
e
r
t
o
e
x
p
l
o
r
e
t
h
e
r
i
s
k
f
ac
t
o
r
o
f
t
h
e
ce
r
v
i
c
a
l
c
a
n
c
e
r
d
ata
s
e
t
a
c
q
u
i
r
e
d
,
i
n
i
t
i
al
d
a
t
a
e
x
p
lo
r
a
t
i
o
n
w
as
done
,
a
n
d
t
h
e
d
e
t
a
i
ls
a
r
e
n
u
m
b
e
r
o
f
t
r
a
i
n
i
n
g
d
a
t
a:
6
0
0
,
n
u
m
b
e
r
o
f
t
e
s
t
d
a
t
a
:
2
5
8
,
t
r
a
i
n
:
(
6
0
0
,
3
5
)
,
t
e
s
t
:
(
2
5
8
,
3
5
)
,
(
6
0
0
,
3
5
)
,
(
2
5
8
,
3
5
)
.
Fig
u
r
e
4
s
h
o
ws
th
e
im
p
o
r
tan
t
f
ea
tu
r
e
b
ased
o
n
th
e
b
a
g
g
in
g
alg
o
r
ith
m
f
o
r
r
is
k
f
ac
to
r
s
s
elec
tio
n
o
f
th
e
ce
r
v
ical
ca
n
ce
r
d
ataset
,
a
n
d
Fig
u
r
e
5
s
h
o
ws
th
e
o
v
er
lap
p
ed
a
n
d
5
-
f
o
ld
co
n
f
u
s
io
n
m
atr
ix
r
esu
lts
o
f
th
e
two
-
class
class
i
f
icatio
n
task
b
ag
g
i
n
g
alg
o
r
ith
m
s
.
T
ab
le
6
is
th
e
tr
ain
in
g
an
d
test
p
er
f
o
r
m
a
n
ce
r
esu
lts
o
f
th
e
b
ag
g
in
g
e
n
s
em
b
l
e
m
o
d
el.
T
h
e
co
n
f
u
s
io
n
m
atr
ices
h
av
e
b
ee
n
u
s
ed
to
f
o
r
ec
ast
th
e
r
e
s
u
lts
o
f
th
e
m
o
d
els
th
at
co
r
r
e
late
to
th
e
ac
tu
al
v
alu
es
to
e
v
alu
ate
th
e
ef
f
ec
tiv
en
ess
o
f
d
if
f
er
en
t
ML
m
o
d
els.
T
h
e
p
r
o
p
o
r
tio
n
o
f
d
ataset
in
s
tan
ce
s
th
at
m
o
d
els
ac
cu
r
ately
p
r
e
d
icted
wh
ile
attem
p
tin
g
to
class
if
y
th
em
is
s
h
o
wn
i
n
th
e
co
n
f
u
s
io
n
m
atr
ices.
Fo
u
r
alter
n
ativ
e
o
u
tco
m
es
f
r
o
m
th
e
p
r
ed
ictio
n
m
eth
o
d
ar
e
d
esi
g
n
ated
as
T
N,
FN,
FP
,
an
d
T
P.
T
h
e
co
n
f
u
s
io
n
m
atr
ices
ar
e
also
u
s
ed
to
co
n
s
tr
u
ct
o
th
er
k
ey
m
ea
s
u
r
em
e
n
ts
o
r
in
d
icato
r
s
,
s
u
ch
as
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e.
T
ab
le
7
is
th
e
tr
ain
i
n
g
an
d
test
p
er
f
o
r
m
an
ce
r
esu
lt
s
o
f
th
e
b
ag
g
in
g
alg
o
r
ith
m
.
Fig
u
r
e
4
.
I
m
p
o
r
tan
t f
ea
tu
r
es
b
ased
o
n
th
e
b
ag
g
in
g
alg
o
r
ith
m
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
E
n
s
u
r
in
g
effec
tive
ce
r
vica
l c
a
n
ce
r
d
is
ea
s
es d
ia
g
n
o
s
is
s
yste
m
u
s
in
g
…
(
Olu
w
a
to
b
i A
kin
l
a
d
e
)
2417
F
i
g
u
r
e
5
.
O
v
e
r
l
a
p
p
e
d
a
n
d
5
-
f
o
l
d
c
o
n
f
u
s
i
o
n
m
a
t
r
i
x
r
e
s
u
lt
s
o
f
t
h
e
t
w
o
-
c
l
as
s
c
l
as
s
i
f
i
c
a
ti
o
n
t
a
s
k
b
a
g
g
i
n
g
a
l
g
o
r
i
t
h
m
s
T
ab
le
6
.
T
r
ai
n
in
g
a
n
d
test
p
er
f
o
r
m
an
ce
r
esu
lts
o
f
b
ag
g
in
g
e
n
s
em
b
le
m
o
d
el
O
p
e
r
a
t
i
o
n
A
c
c
u
r
a
c
y
R
e
c
a
l
l
P
r
e
c
i
s
i
o
n
F1
-
sc
o
r
e
Tr
a
i
n
i
n
g
p
e
r
f
o
r
ma
n
c
e
0
.
9
9
1
6
6
7
0
.
8
2
7
5
8
6
1
.
0
0
.
9
0
5
6
6
Te
st
i
n
g
p
e
r
f
o
r
ma
n
c
e
0
.
9
3
0
2
3
3
0
.
1
3
3
3
3
3
0
.
2
8
5
7
1
4
0
.
1
8
1
8
1
8
C
las
s
if
icatio
n
r
ep
o
r
t
b
ased
o
n
class
if
ied
ca
s
es
is
s
p
ec
if
ied
in
th
e
b
o
x
b
elo
w,
alo
n
g
with
th
e
co
r
r
esp
o
n
d
in
g
class
if
ier
s
'
p
er
f
o
r
m
an
ce
o
v
er
ac
cu
r
ac
y
,
p
r
e
cisi
o
n
,
F1
-
s
co
r
e
,
r
ec
all,
an
d
r
ec
eiv
er
o
p
er
atin
g
ch
ar
ac
ter
is
tic
(
R
OC
)
v
alu
es.
T
h
e
r
esu
lts
in
d
icate
th
at
th
e
b
ag
g
in
g
m
o
d
el
is
th
e
m
o
s
t
a
cc
u
r
ate.
T
h
er
e
f
o
r
e,
co
m
p
ar
ed
to
o
th
e
r
class
if
ier
s
,
th
e
b
ag
g
i
n
g
ML
class
if
ier
c
an
p
r
ed
ict
th
e
lik
elih
o
o
d
o
f
d
is
ea
s
e
with
g
r
ea
ter
ac
cu
r
ac
y
.
T
h
e
R
OC
ar
ea
o
f
class
if
icatio
n
m
eth
o
d
s
is
s
h
o
wn
in
Fig
u
r
e
6
.
T
h
e
illn
ess
d
atase
t w
ith
th
e
attr
ib
u
te
s
is
u
s
ed
in
th
is
r
esear
c
h
to
ev
alu
ate
th
e
b
a
g
g
in
g
alg
o
r
ith
m
.
I
t
ca
n
b
e
o
b
s
er
v
e
d
th
at,
with
9
8
.
6
6
%
ac
cu
r
ac
y
,
9
4
%
p
r
ec
is
io
n
,
9
9
%
r
ec
all,
a
n
d
9
6
%
F1
-
s
co
r
e,
th
e
b
ag
g
in
g
class
if
ier
p
er
f
o
r
m
s
th
e
b
est.
9
6
%
F1
-
s
co
r
e
af
ter
th
e
b
ag
g
in
g
m
eth
o
d
,
9
9
%
ac
c
u
r
ac
y
,
9
4
% p
r
ec
is
io
n
,
9
9
% r
e
ca
ll.
T
h
e
ca
p
ac
ity
o
f
t
h
e
v
a
r
io
u
s
ML
alg
o
r
ith
m
s
to
d
iag
n
o
s
e
tr
u
e
p
o
s
itiv
e
r
ate
(
T
PR
)
v
s
.
f
alse
p
o
s
itiv
e
r
ate
(
F
P
R
)
is
r
ep
r
esen
ted
g
r
ap
h
ically
b
y
th
e
R
OC
cu
r
v
e.
T
h
e
AUC,
wh
ich
r
ep
r
esen
ts
t
h
e
r
ec
eiv
er
o
p
e
r
atin
g
cu
r
v
e'
s
p
r
o
b
a
b
ilit
y
s
h
ap
e,
is
u
s
ed
to
ass
es
s
h
o
w
well
th
e
class
es
ca
n
b
e
d
is
tin
g
u
is
h
ed
f
r
o
m
o
n
e
a
n
o
th
e
r
.
T
h
e
r
esu
lts
o
f
s
ev
er
al
class
if
icatio
n
m
o
d
e
ls
f
o
r
d
eter
m
in
in
g
wh
eth
er
a
p
atien
t
h
as
d
iab
ete
s
o
r
n
o
t
ar
e
s
h
o
wn
in
t
h
e
R
OC
cu
r
v
e.
I
n
ter
m
s
o
f
R
OC
an
d
AUC
f
o
r
ca
n
ce
r
illn
ess
p
r
ed
ictio
n
,
class
if
ier
s
lik
e
GB
,
n
aiv
e
B
ay
es
(
NB
)
,
SVM,
an
d
DT
o
u
tp
er
f
o
r
m
o
t
h
er
class
if
ier
s
.
T
h
e
test
in
g
d
ata
s
et
v
er
if
ied
th
e
tr
a
in
ed
m
o
d
els'
p
r
ed
ictio
n
p
er
f
o
r
m
an
ce
.
T
ab
le
7
.
T
r
ai
n
in
g
a
n
d
test
p
er
f
o
r
m
an
ce
r
esu
lts
o
f
b
ag
g
in
g
al
g
o
r
ith
m
Tr
a
i
n
i
n
g
a
c
c
u
r
a
c
y
Te
st
i
n
g
a
c
c
u
r
a
c
y
TP
FP
FN
TN
R
O
C
A
U
C
sco
r
e
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
sc
o
r
e
s
u
p
p
o
r
t
0
1
0
1
0
1
9
9
.
1
6
9
3
.
0
2
2
4
1
2
13
2
4
9
.
5
8
9
4
.
0
0
.
0
0
9
9
.
0
0
.
0
0
9
7
.
0
0
.
0
0
Fig
u
r
e
6
.
T
h
e
R
OC
cu
r
v
e
f
o
r
t
h
e
b
ag
g
in
g
m
o
d
el
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
26
:
2
4
1
0
-
2
4
2
2
2418
4
.
2
.
B
o
o
s
t
ing
C
lass
if
ic
ati
o
n
r
e
p
o
r
t
b
ase
d
o
n
class
if
i
e
d
c
ases
is
s
p
e
ci
f
i
ed
i
n
t
h
e
b
o
x
b
el
o
w,
a
lo
n
g
wit
h
th
e
co
r
r
es
p
o
n
d
i
n
g
cl
ass
i
f
ie
r
s
'
p
e
r
f
o
r
m
a
n
ce
o
v
e
r
ac
cu
r
ac
y
,
p
r
ec
i
s
io
n
,
F
1
-
s
c
o
r
e
,
r
ec
a
ll,
a
n
d
R
OC
v
al
u
es.
R
es
u
l
ts
in
d
ic
at
e
t
h
at
t
h
e
b
o
o
s
ti
n
g
m
o
d
el
h
as
th
e
g
r
ea
t
est
ac
cu
r
ac
y
.
I
n
co
m
p
a
r
is
o
n
t
o
o
t
h
e
r
class
i
f
i
er
s
,
t
h
e
b
o
o
s
ti
n
g
ML
class
i
f
i
er
c
an
p
r
e
d
i
ct
th
e
l
i
k
el
ih
o
o
d
o
f
d
is
e
ase
w
it
h
g
r
e
ate
r
ac
c
u
r
a
cy
.
T
h
e
ill
n
ess
d
a
tas
et
w
it
h
t
h
e
af
o
r
e
m
e
n
ti
o
n
e
d
a
ttr
ib
u
t
es
is
u
s
e
d
i
n
t
h
is
r
ese
a
r
c
h
t
o
e
v
al
u
at
e
t
h
e
b
a
g
g
i
n
g
te
ch
n
i
q
u
e
.
Fig
u
r
e
7
d
ep
icts
th
e
o
v
e
r
la
p
p
e
d
a
n
d
5
-
f
o
ld
c
o
n
f
u
s
io
n
m
at
r
i
x
r
esu
lts
o
f
t
h
e
tw
o
-
class
c
lass
i
f
i
ca
t
io
n
t
ask
,
b
o
o
s
t
in
g
al
g
o
r
it
h
m
s
th
e
co
n
f
u
s
io
n
m
a
tr
i
x
f
o
r
t
h
e
b
o
o
s
tin
g
m
et
h
o
d
.
T
h
e
R
OC
r
e
g
i
o
n
o
f
cl
ass
i
f
ic
ati
o
n
m
et
h
o
d
s
R
O
C
f
o
r
t
h
e
b
o
o
s
t
in
g
m
o
d
el
is
s
h
o
w
n
i
n
Fi
g
u
r
e
8
,
an
d
T
a
b
l
e
8
p
r
es
e
n
ts
tr
ai
n
i
n
g
an
d
tes
t
p
e
r
f
o
r
m
a
n
c
e
r
es
u
l
ts
o
f
t
h
e
b
o
o
s
ti
n
g
m
o
d
el
.
Fig
u
r
e
7
.
Ov
e
r
lap
p
ed
an
d
5
-
f
o
ld
co
n
f
u
s
io
n
m
atr
ix
r
esu
lts
o
f
th
e
two
-
class
class
if
icatio
n
task
b
o
o
s
tin
g
alg
o
r
ith
m
s
Fig
u
r
e
8
.
T
h
e
R
OC
cu
r
v
e
f
o
r
t
h
e
b
o
o
s
tin
g
m
o
d
el
T
ab
le
8
.
T
r
ai
n
in
g
a
n
d
test
p
er
f
o
r
m
an
ce
r
esu
lts
o
f
b
o
o
s
tin
g
m
o
d
el
Tr
a
i
n
i
n
g
a
c
c
u
r
a
c
y
Te
st
i
n
g
a
c
c
u
r
a
c
y
TP
FP
FN
TN
R
O
C
A
U
C
sc
o
r
e
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
sc
o
r
e
s
u
p
p
o
r
t
9
8
.
2
2
9
6
.
4
5
5
7
1
0
10
19
9
8
.
3
3
9
8
.
2
7
1
0
0
9
9
.
1
3
4
.
3
.
Ada
B
o
o
s
t
I
n
th
is
s
tu
d
y
,
th
e
b
o
o
s
tin
g
class
if
ier
wo
r
k
s
b
est,
with
9
8
%
a
cc
u
r
ac
y
,
9
4
%
p
r
ec
is
io
n
,
9
9
%
r
ec
all,
an
d
9
6
%
F1
-
s
co
r
e.
W
ith
9
8
%
ac
cu
r
ac
y
,
9
4
%
p
r
ec
is
io
n
,
9
9
%
r
ec
all,
an
d
9
6
%
F1
-
s
co
r
e
af
ter
th
e
b
o
o
s
tin
g
alg
o
r
ith
m
.
C
lass
if
icatio
n
r
ep
o
r
t
b
ased
o
n
class
if
ied
ca
s
es
is
s
p
ec
if
ied
in
th
e
b
o
x
b
elo
w,
alo
n
g
with
th
e
co
r
r
esp
o
n
d
in
g
class
if
ier
s
'
p
er
f
o
r
m
an
ce
o
v
e
r
ac
cu
r
ac
y
,
p
r
ec
i
s
io
n
,
F1
-
s
co
r
e
,
r
ec
all,
a
n
d
R
OC
v
alu
es.
R
esu
lts
in
d
icate
th
at
th
e
Ad
aBo
o
s
t
m
o
d
el
h
as
th
e
h
ig
h
est
d
eg
r
ee
o
f
ac
c
u
r
ac
y
.
T
h
er
e
f
o
r
e,
co
m
p
ar
ed
to
o
th
er
class
if
ier
s
,
th
e
Ad
aBo
o
s
t
ML
class
if
ier
ca
n
p
r
ed
ict
th
e
l
ik
elih
o
o
d
o
f
d
is
ea
s
e
with
g
r
ea
ter
ac
cu
r
ac
y
.
T
h
e
illn
ess
d
ataset
with
th
e
attr
ib
u
tes
is
u
s
ed
in
th
is
r
esear
ch
to
ev
alu
ate
th
e
b
ag
g
in
g
tech
n
iq
u
e.
Fig
u
r
e
9
d
ep
icts
th
e
Ad
aBo
o
s
t
alg
o
r
ith
m
'
s
co
n
f
u
s
i
o
n
m
atr
ix
.
T
h
e
R
OC
r
eg
io
n
o
f
class
if
icatio
n
m
eth
o
d
s
is
s
h
o
wn
in
Fig
u
r
e
1
0
,
an
d
T
ab
le
9
p
r
esen
ts
th
e
tr
ain
i
n
g
an
d
test
p
er
f
o
r
m
an
ce
r
esu
lt
s
o
f
th
e
Ad
aBo
o
s
t a
lg
o
r
ith
m
.
Fig
u
r
e
9
.
C
o
n
f
u
s
io
n
m
atr
i
x
r
e
s
u
lts
o
f
th
e
two
-
class
class
if
icatio
n
task
f
o
r
Ad
aBo
o
s
t a
lg
o
r
ith
m
s
Fig
u
r
e
1
0
.
T
h
e
R
OC
cu
r
v
e
f
o
r
th
e
Ad
aBo
o
s
t m
o
d
el
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
E
n
s
u
r
in
g
effec
tive
ce
r
vica
l c
a
n
ce
r
d
is
ea
s
es d
ia
g
n
o
s
is
s
yste
m
u
s
in
g
…
(
Olu
w
a
to
b
i A
kin
l
a
d
e
)
2419
T
ab
le
9
.
T
r
ai
n
in
g
a
n
d
test
p
er
f
o
r
m
an
ce
r
esu
lts
o
f
Ad
aBo
o
s
t
alg
o
r
ith
m
Tr
a
i
n
i
n
g
a
c
c
u
r
a
c
y
Te
st
i
n
g
a
c
c
u
r
a
c
y
TP
FP
FN
TN
R
O
C
A
U
C
sco
r
e
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
sc
o
r
e
s
u
p
p
o
r
t
0
1
0
1
0
1
9
4
%
9
2
%
2
13
4
2
3
9
0
.
9
5
0
.
3
3
0
.
9
8
0
.
1
3
0
.
9
7
0
.
1
9
T
h
e
b
ag
g
i
n
g
cl
ass
i
f
ie
r
w
o
r
k
s
b
est,
wi
th
9
5
% a
c
cu
r
a
c
y
,
9
5
%
p
r
e
cisi
o
n
,
9
8
%
r
e
ca
ll
,
a
n
d
9
7
%
F1
-
s
c
o
r
e.
W
ith
9
8
%
ac
cu
r
ac
y
,
9
4
%
p
r
ec
is
io
n
,
9
9
%
r
ec
all,
an
d
9
6
%
F1
-
s
co
r
e
af
te
r
th
e
b
a
g
g
in
g
alg
o
r
ith
m
.
Af
ter
co
m
p
ar
in
g
s
ev
er
al
class
if
icati
o
n
alg
o
r
ith
m
s
,
th
e
en
s
em
b
le
c
lass
if
ier
u
s
in
g
th
e
af
o
r
em
en
tio
n
ed
ch
ar
ac
ter
is
tics
o
u
tp
er
f
o
r
m
s
th
e
o
th
er
s
with
an
ac
cu
r
ac
y
o
f
8
7
%.
Fo
llo
win
g
th
e
b
o
o
s
tin
g
class
if
ier
,
th
e
b
a
g
g
in
g
an
d
Ad
aBo
o
s
t
class
if
ier
al
s
o
p
er
f
o
r
m
s
b
etter
,
with
an
ac
cu
r
ac
y
o
f
8
8
%.
T
o
s
o
lv
e
th
e
alg
o
r
ith
m
m
o
r
e
ef
f
ec
tiv
ely
,
m
an
y
ex
tr
a
Py
th
o
n
lib
r
ar
ies
a
r
e
lo
ad
ed
.
T
h
e
ess
en
tial
lib
r
a
r
ies,
in
clu
d
in
g
p
an
d
as,
N
u
m
Py
,
s
cik
it
-
lear
n
,
a
n
d
m
atp
lo
tlib
,
h
av
e
b
ee
n
im
p
o
r
ted
.
B
ag
g
in
g
:
0
.
8
7
7
1
0
4
(
0
.
0
1
3
3
7
2
)
,
b
o
o
s
tin
g
:
0
.
8
3
9
3
9
4
(
0
.
0
0
0
4
1
0
)
,
an
d
Ad
aBo
o
s
t
:
0
.
8
7
7
1
0
4
(
0
.
0
1
3
3
7
2
)
.
T
h
e
c
o
m
p
ar
is
o
n
o
f
th
e
th
r
ee
m
o
d
els
is
p
r
esen
ted
g
r
ap
h
i
ca
lly
in
Fig
u
r
e
1
1
.
T
h
e
p
r
ed
ictio
n
an
d
ea
r
l
y
d
iag
n
o
s
is
o
f
d
is
ea
s
es
m
ig
h
t
b
e
co
m
p
letely
alter
ed
th
an
k
s
to
ML
,
a
b
r
an
c
h
o
f
AI
.
T
h
is
wo
r
k
'
s
p
r
im
ar
y
o
b
jectiv
e
was
to
d
ev
elo
p
an
d
im
p
le
m
en
t
illn
ess
p
r
ed
ictio
n
u
s
in
g
a
v
ar
iety
o
f
ML
ap
p
r
o
ac
h
es,
as
well
as
to
d
o
o
u
tp
u
t
an
aly
s
is
o
n
th
o
s
e
tech
n
iq
u
es
to
id
e
n
tify
th
e
b
est
class
if
ier
with
th
e
g
r
ea
test
lev
el
o
f
ac
cu
r
ac
y
,
wh
ich
we
wer
e
ab
le
to
d
o
.
I
n
th
is
r
esear
ch
,
we
e
v
alu
ate
d
a
v
ar
iety
o
f
ML
class
if
icatio
n
m
eth
o
d
s
to
atta
in
h
ig
h
-
p
er
f
o
r
m
an
ce
ac
cu
r
ac
y
.
C
o
m
p
ar
in
g
b
o
o
s
tin
g
class
if
icatio
n
m
eth
o
d
s
to
b
ag
g
in
g
an
d
Ad
aBo
o
s
t
alg
o
r
it
h
m
s
r
ev
ea
ls
th
at
th
e
latter
ar
e
m
o
r
e
ef
f
icien
t
an
d
p
r
o
v
id
e
s
u
p
er
io
r
r
esu
lts
.
T
h
e
b
ag
g
in
g
an
d
Ad
aBo
o
s
t
alg
o
r
it
h
m
'
s
clas
s
if
icatio
n
ac
cu
r
ac
y
was
8
8
%.
T
h
e
FPR
is
s
h
o
wn
o
n
th
e
x
-
a
x
is
,
wh
ile
th
e
g
en
u
in
e
p
o
s
itiv
e
r
ate
is
s
h
o
wn
o
n
th
e
y
-
ax
is
.
R
OC
cu
r
v
e
r
ep
r
esen
ts
th
e
b
est
p
r
ed
ictio
n
s
m
ad
e
u
s
in
g
th
e
b
ag
g
in
g
m
o
d
el
(
Fi
g
u
r
e
6
)
,
bo
o
s
tin
g
m
o
d
el
(
Fig
u
r
e
8
)
,
an
d
Ad
aBo
o
s
t
m
o
d
el
(
Fig
u
r
e
1
0
)
,
r
esp
ec
tiv
ely
,
with
an
d
with
o
u
t
h
y
p
e
r
p
ar
am
eter
a
d
ju
s
tm
en
t.
Fig
u
r
e
1
1
.
C
o
m
p
ar
is
o
n
o
f
th
e
m
o
d
els
5.
CO
NCLU
SI
O
N
T
h
e
co
n
clu
s
io
n
is
d
esig
n
ed
to
h
elp
en
s
em
b
le
ML
h
as
e
m
er
g
ed
as
a
p
r
o
m
is
in
g
ap
p
r
o
ac
h
f
o
r
im
p
r
o
v
in
g
th
e
ac
cu
r
ac
y
an
d
r
eliab
ilit
y
o
f
ce
r
v
ical
ca
n
ce
r
d
iag
n
o
s
is
s
y
s
tem
s
.
T
h
is
s
t
u
d
y
ev
alu
ate
d
th
e
p
er
f
o
r
m
an
ce
o
f
v
a
r
io
u
s
en
s
em
b
le
ML
m
o
d
els
an
d
c
o
m
p
ar
e
d
th
em
with
co
n
v
en
tio
n
al
ML
m
o
d
els.
T
h
e
r
esu
lts
s
h
o
wed
th
at
e
n
s
em
b
le
ML
m
o
d
els
o
u
tp
e
r
f
o
r
m
ed
co
n
v
en
tio
n
al
m
o
d
els
in
ter
m
s
o
f
ac
cu
r
a
cy
,
s
en
s
itiv
ity
,
an
d
s
p
ec
if
icity
.
Mo
r
eo
v
e
r
,
in
ter
p
r
e
tab
ilit
y
an
aly
s
is
u
s
in
g
f
ea
tu
r
e
im
p
o
r
tan
ce
a
n
d
p
ar
tial
d
ep
e
n
d
en
ce
p
lo
ts
s
h
o
wed
Evaluation Warning : The document was created with Spire.PDF for Python.