I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
A
I
)
V
ol
. 13, No. 4, D
e
c
e
m
be
r
2024, pp.
4301
~
4311
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v13.i4.pp
4301
-
4311
4301
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
Pr
oac
t
i
ve
c
e
r
vi
c
al
c
an
c
e
r
r
i
sk
ass
e
ss
m
e
n
t
u
si
n
g
d
at
a
-
d
r
i
ve
n
an
al
yt
i
c
s
S
r
e
e
la
t
h
a
1,
2
, V
r
in
d
a S
h
iv
as
h
e
t
t
y
3
1
D
e
pa
r
t
m
e
nt
of
I
nf
or
m
a
t
i
on
S
c
i
e
nc
e
a
nd
E
ngi
ne
e
r
i
ng, S
a
i
V
i
dya
I
ns
t
i
t
ut
e
of
T
e
c
hnol
ogy, V
i
s
ve
s
va
r
a
ya
T
e
c
hnol
ogi
c
a
l
U
ni
ve
r
s
i
t
y,
B
e
l
a
ga
vi
, I
ndi
a
2
D
e
pa
r
t
m
e
nt
of
C
om
put
e
r
S
c
i
e
nc
e
a
nd
E
ngi
ne
e
r
i
ng, P
r
e
s
i
de
nc
y U
ni
ve
r
s
i
t
y, B
a
nga
l
or
e
, I
ndi
a
3
D
e
pa
r
t
m
e
nt
of
I
nf
or
m
a
t
i
on
S
c
i
e
nc
e
a
nd E
ngi
ne
e
r
i
ng, S
a
i
V
i
dya
I
ns
t
i
t
ut
e
of
T
e
c
hnol
ogy, B
a
nga
l
or
e
, I
ndi
a
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
O
c
t
22, 2023
R
e
vi
s
e
d
F
e
b 17, 2024
A
c
c
e
pt
e
d
M
a
r
21, 2024
This
study
introduces
a
sophisticated
predictive
model
integrating
clinical
and
lifestyle
data
addressing
the
critical
public
health
challenge
of
c
ervical
cancer,
particular
ly
in
regions
lacking
routine
screenings
.
Leveraging
data
-
driven
analytics,
the
proposed
model
undergoes
compreh
ensive
preprocessing,
including
exploratory
data
analysis,
missing
value
imputation,
and
feature
extractio
n.
Feature
selection
is
carrie
d
out
usi
ng
the
XGBoost
classifier
to
ensure
model
efficacy.
Data
normalization
an
d
class
balance
via
oversampling
techniques
are
applied,
with
model
vali
dation
conducted
through
stratifi
ed
cross
-
validation.
The
optimized
feature
vector
is
then
employed
to
train
a
LightGBM
model.
Utilizing
a
retros
pective
dataset
of
858
patients
from
the
Hospital
Universitario
de
C
aracas,
Venezuela,
comprising
demographic,
lifestyle,
and
medical
history
data,
the
LightGBM
model
achieves
an
impressive
accuracy
of
98%,
outperf
orming
similar
existing
approache
s.
The
study
outcome
demonstrate
s
the
effectivenes
s
of
t
he
proposed
data
modelling
framework
and
feature
selection,
along
with
the
choice
of
LightGBM
as
a
suitable
classifie
r.
The
proposed
predictive
framework
can
efficiently
aid
healthcare
professionals
in prioritizing
high
-
risk pati
ents for furt
her evaluation
and interven
tion.
K
e
y
w
o
r
ds
:
C
e
r
vi
c
a
l
c
a
nc
e
r
L
if
e
s
ty
le
da
ta
M
a
c
hi
ne
l
e
a
r
ni
ng
M
ul
ti
f
a
c
to
r
ia
l
c
li
ni
c
a
l
da
ta
R
is
k pr
e
di
c
ti
on
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
S
r
e
e
la
th
a
D
e
pa
r
tm
e
nt
of
C
om
put
e
r
S
c
ie
nc
e
a
nd E
ngi
ne
e
r
in
g, P
r
e
s
id
e
nc
y
U
ni
ve
r
s
it
y
B
a
nga
lo
r
e
, I
ndi
a
E
m
a
il
:
s
r
e
e
la
th
a
.pk@
pr
e
s
id
e
nc
yuni
ve
r
s
it
y.i
n
1.
I
N
T
R
O
D
U
C
T
I
O
N
C
e
r
vi
c
a
l
c
a
nc
e
r
is
a
s
e
ve
r
e
di
s
e
a
s
e
th
a
t
a
f
f
e
c
t
s
m
il
li
ons
of
w
o
m
e
n
w
or
ld
w
id
e
.
I
t
is
c
ha
r
a
c
te
r
iz
e
d
by
th
e
a
bnor
m
a
l
gr
ow
th
of
c
e
ll
s
in
th
e
c
e
r
vi
x,
th
e
op
e
ni
ng
to
th
e
u
te
r
us
[
1]
.
T
he
l
e
a
di
ng
c
a
u
s
e
of
c
e
r
vi
c
a
l
c
a
nc
e
r
is
pe
r
s
is
te
nt
in
f
e
c
ti
on
w
it
h
c
e
r
ta
in
ty
pe
s
of
hum
a
n
pa
pi
ll
om
a
vi
r
us
(
H
P
V
)
,
a
pr
e
va
le
nt
vi
r
us
th
a
t
is
s
pr
e
a
d
th
r
ough
s
e
xua
l
c
ont
a
c
t
[
2]
.
M
os
t
pe
opl
e
w
ho
a
r
e
in
f
e
c
te
d
w
it
h
H
P
V
c
le
a
r
th
e
vi
r
us
on
th
e
ir
ow
n,
but
s
om
e
pe
opl
e
de
ve
lo
p
a
pe
r
s
is
te
nt
in
f
e
c
ti
on,
w
hi
c
h
c
a
n
le
a
d
to
th
e
de
ve
lo
pm
e
nt
of
pr
e
c
a
nc
e
r
ous
c
e
ll
s
,
w
hi
c
h
c
a
n
e
ve
nt
ua
ll
y
tu
r
n
in
to
c
a
nc
e
r
[
3]
.
M
e
di
c
a
l
a
dva
nc
e
s
ha
v
e
le
d
to
s
e
ve
r
a
l
w
a
ys
to
pr
e
ve
nt
a
nd
tr
e
a
t
e
a
r
ly
-
s
ta
ge
c
e
r
vi
c
a
l
c
a
nc
e
r
. U
nf
or
tu
na
te
ly
, t
he
r
e
a
li
ty
i
s
ha
r
s
h;
m
a
ny w
om
e
n a
r
e
di
a
gnos
e
d w
it
h c
e
r
vi
c
a
l
c
a
nc
e
r
w
he
n t
he
di
s
e
a
s
e
ha
s
r
e
a
c
h
e
d
a
n
a
dva
nc
e
d
s
ta
g
e
,
m
a
ki
ng
tr
e
a
tm
e
nt
m
or
e
c
om
pl
e
x,
e
xpe
n
s
iv
e
,
a
nd
le
s
s
li
ke
ly
to
be
s
uc
c
e
s
s
f
ul
[
4]
.
D
e
s
pi
te
a
va
il
a
bl
e
pr
e
ve
nt
iv
e
m
e
c
h
a
ni
s
m
s
,
e
a
r
ly
de
te
c
ti
on
a
nd
m
a
na
ge
m
e
nt
of
c
e
r
vi
c
a
l
c
a
nc
e
r
r
e
m
a
in
s
a
s
ig
ni
f
ic
a
nt
he
a
lt
h
pr
obl
e
m
.
L
a
te
di
a
gnos
is
of
c
e
r
vi
c
a
l
c
a
nc
e
r
is
ve
r
y
c
om
m
on
a
s
m
a
ny
w
om
e
n
a
r
e
di
a
gnos
e
d
w
it
h
a
dva
nc
e
d
c
e
r
vi
c
a
l
c
a
n
c
e
r
,
m
a
in
ly
due
to
a
la
c
k
of
r
e
gul
a
r
s
c
r
e
e
ni
ng,
e
c
onomi
c
ba
r
r
ie
r
s
t
o
he
a
lt
hc
a
r
e
a
nd
th
e
is
s
ue
of
s
oc
io
-
c
ul
tu
r
a
l
[
5]
,
[
6]
.
T
he
W
or
ld
H
e
a
lt
h
O
r
ga
ni
z
a
ti
on
(
W
H
O
)
r
e
c
om
m
e
nds
th
a
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
. 13, No. 4, D
e
c
e
m
be
r
2024:
4301
-
4311
4302
w
om
e
n
a
ge
d
30
-
49
unde
r
go
c
e
r
vi
c
a
l
c
a
nc
e
r
s
c
r
e
e
ni
ng
a
t
le
a
s
t
onc
e
e
ve
r
y
f
iv
e
ye
a
r
s
,
ye
t
a
dh
e
r
e
nc
e
to
th
e
s
e
gui
de
li
ne
s
is
in
s
uf
f
ic
ie
nt
,
pa
r
ti
c
ul
a
r
ly
in
lo
w
-
r
e
s
our
c
e
s
e
tt
in
gs
a
nd
r
e
gi
ons
w
it
h
li
m
it
e
d
a
w
a
r
e
ne
s
s
a
bout
c
e
r
vi
c
a
l
c
a
nc
e
r
a
nd
it
s
im
pl
ic
a
ti
ons
[
7]
.
T
he
r
e
a
r
e
m
a
ny
ba
r
r
ie
r
s
to
c
e
r
vi
c
a
l
c
a
nc
e
r
s
c
r
e
e
ni
ng.
O
ne
ba
r
r
ie
r
is
th
a
t
w
om
e
n m
a
y not know a
bout
t
he
r
is
k
f
a
c
to
r
s
a
nd s
ym
pt
om
s
of
c
e
r
vi
c
a
l
c
a
nc
e
r
, s
o t
he
y do not
r
e
a
li
z
e
how
im
por
ta
nt
it
is
to
ge
t
s
c
r
e
e
ne
d
r
e
gul
a
r
ly
[
8]
.
A
not
he
r
ba
r
r
ie
r
i
s
th
a
t
w
om
e
n
m
a
y
not
be
c
om
f
or
ta
bl
e
ta
lk
in
g
a
bout
th
e
ir
r
e
pr
oduc
ti
ve
he
a
lt
h,
e
s
p
e
c
ia
ll
y
w
it
h
m
a
le
doc
to
r
s
.
I
n
c
on
s
e
r
va
ti
ve
s
oc
ie
ti
e
s
,
w
om
e
n
m
a
y
be
he
s
it
a
nt
to
ta
lk
a
bout
th
e
ir
r
e
pr
oduc
ti
ve
he
a
lt
h
be
c
a
us
e
of
c
ul
tu
r
a
l
nor
m
s
or
pe
r
s
ona
l
in
hi
bi
ti
ons
[
9]
.
T
hi
s
c
a
n
le
a
d
to
de
la
ye
d
or
m
is
s
e
d
di
a
gnos
e
s
.
A
not
he
r
ba
r
r
ie
r
is
th
e
a
va
il
a
bi
li
ty
a
nd
a
f
f
or
da
bi
l
it
y
of
s
c
r
e
e
ni
ng
f
a
c
il
it
ie
s
.
I
n
m
a
ny
pl
a
c
e
s
,
pr
im
a
r
il
y
r
ur
a
l
a
r
e
a
s
,
th
e
r
e
is
no
in
f
r
a
s
tr
uc
tu
r
e
f
or
r
e
gul
a
r
,
a
f
f
or
da
bl
e
c
e
r
vi
c
a
l
c
a
nc
e
r
s
c
r
e
e
ni
ng
[
10]
.
T
hi
s
m
a
k
e
s
it
di
f
f
ic
ul
t
f
or
w
om
e
n
to
a
c
c
e
s
s
th
e
s
e
e
s
s
e
nt
ia
l
h
e
a
lt
hc
a
r
e
s
e
r
vi
c
e
s
.
T
he
r
e
f
or
e
,
th
e
r
e
is
a
n
ur
ge
nt
ne
e
d
f
or
a
n
e
f
f
e
c
ti
ve
s
ol
ut
io
n
th
a
t
c
a
n
f
a
c
il
it
a
te
e
a
r
ly
de
te
c
ti
on
of
c
e
r
vi
c
a
l
c
a
nc
e
r
.
T
he
e
m
e
r
ge
n
c
e
of
a
r
ti
f
ic
ia
l
in
te
ll
ig
e
nc
e
(
A
I
)
te
c
hnol
ogy
in
he
a
lt
hc
a
r
e
is
r
e
vol
ut
io
ni
z
in
g
th
e
di
a
gnos
ti
c
pr
oc
e
s
s
[
11]
.
U
s
in
g
c
li
ni
c
a
l
da
ta
a
nd
pa
ti
e
nt
hi
s
to
r
ie
s
,
A
I
-
ba
s
e
d
m
e
th
ods
c
a
n
id
e
nt
if
y
pa
tt
e
r
ns
a
nd
c
or
r
e
la
ti
ons
th
a
t
m
a
y
not
be
m
uc
h
e
xpl
or
e
d
us
in
g
tr
a
di
ti
ona
l
a
na
ly
ti
c
a
l
m
e
th
ods
[
12]
.
H
ow
e
ve
r
,
w
he
n
th
e
da
ta
c
om
e
s
f
r
om
c
li
ni
c
a
l
s
e
tt
in
gs
,
it
is
di
f
f
ic
ul
t
to
bui
ld
a
c
c
ur
a
te
pr
e
di
c
ti
ve
m
ode
ls
th
a
t
c
a
n
be
c
li
ni
c
a
ll
y
a
c
c
e
pt
a
bl
e
be
c
a
us
e
of
t
he
i
nhe
r
e
nt
he
te
r
oge
ne
it
y a
nd c
om
pl
e
xi
t
y of
t
hi
s
f
or
m
of
da
ta
[
13]
,
[
14]
.
I
n
r
e
c
e
nt
s
ta
te
-
of
-
th
e
-
a
r
t
w
or
ks
,
m
a
ny
r
e
s
e
a
r
c
he
r
s
h
a
ve
of
f
e
r
e
d
r
is
k
pr
e
di
c
ti
on
m
ode
ls
to
va
r
io
us
c
li
ni
c
a
l
c
ont
e
xt
s
.
I
n
c
e
r
vi
c
a
l
c
a
n
c
e
r
,
th
e
e
xi
s
ti
ng
s
c
he
m
e
s
a
dop
t
va
r
io
us
a
tt
r
ib
ut
e
s
,
in
vol
ve
c
ol
le
c
ti
ng
c
li
ni
c
a
l
da
ta
a
nd
de
m
ogr
a
phi
c
in
f
or
m
a
ti
on
f
r
om
w
om
e
n
a
nd
a
ppl
yi
ng
di
f
f
e
r
e
nt
m
a
c
hi
ne
le
a
r
ni
ng
a
ppr
oa
c
he
s
to
m
ode
l
r
is
k
f
a
c
to
r
s
.
I
ja
z
e
t
al
.
[
15]
de
ve
lo
p
e
d
a
pr
e
di
c
ti
on
m
od
e
l
to
pr
e
di
c
t
c
e
r
vi
c
a
l
c
a
nc
e
r
by
e
v
a
lu
a
ti
ng
r
is
k
f
a
c
to
r
s
.
T
hi
s
m
ode
l
us
e
s
noi
s
e
-
r
e
s
is
ta
nt
de
n
s
it
y
c
lu
s
te
r
in
g
a
nd
a
nom
a
ly
is
ol
a
ti
on
tr
e
e
s
to
ha
ndl
e
out
li
e
r
s
in
th
e
da
ta
a
nd
s
ynt
he
ti
c
m
in
or
it
y
ove
r
-
s
a
m
pl
in
g
te
c
hni
que
(
S
M
O
T
E
)
te
c
hni
que
to
ba
la
nc
e
th
e
da
ta
s
e
t.
T
he
m
ode
l
is
ba
s
e
d
on
a
r
a
ndom
f
or
e
s
t
c
la
s
s
if
ie
r
f
or
c
la
s
s
if
ic
a
t
io
n
ta
s
ks
.
L
il
hor
e
e
t
al
.
[
16]
a
ddr
e
s
s
e
d
th
e
li
m
it
a
ti
ons
of
t
he
s
ta
nda
r
d P
a
p s
m
e
a
r
e
xa
m
in
a
ti
on f
or
c
e
r
vi
c
a
l
c
a
nc
e
r
, w
hi
c
h c
a
n pr
oduc
e
m
a
ny f
a
ls
e
-
pos
it
iv
e
r
e
s
ul
ts
due
to
hum
a
n
e
r
r
or
.
T
he
a
ut
hor
s
in
tr
oduc
e
d
a
n
e
f
f
ic
ie
nt
f
e
a
tu
r
e
s
e
le
c
ti
on
a
nd
pr
e
di
c
ti
on
m
ode
l
us
in
g
B
or
ut
a
a
na
ly
s
is
a
nd
th
e
s
uppor
t
ve
c
to
r
m
a
c
hi
ne
(
S
V
M
)
m
e
t
hod.
H
ow
e
ve
r
,
th
is
s
tu
dy
di
d
no
t
di
s
c
us
s
th
e
m
ode
l'
s
s
c
a
la
bi
li
ty
t
o a
da
pt
t
o c
ha
ngi
ng c
a
nc
e
r
s
c
r
e
e
ni
ng guide
li
ne
s
a
nd pr
a
c
ti
c
e
s
. P
ut
r
i
e
t
al
.
[
17]
c
ons
id
e
r
e
d
a
c
a
s
e
s
tu
dy of
c
e
r
vi
c
a
l
c
a
nc
e
r
i
n I
ndone
s
ia
, w
he
r
e
i
t
r
a
nks
a
s
t
he
s
e
c
ond de
a
dl
ie
s
t
f
or
m
of
c
a
nc
e
r
a
f
te
r
b
r
e
a
s
t
c
a
nc
e
r
.
T
he
r
e
s
e
a
r
c
h
pr
e
s
e
nt
e
d
a
m
e
th
odol
ogy
c
om
bi
ni
ng
r
e
gi
on
-
a
w
a
r
e
s
e
gm
e
nt
a
ti
on,
te
xt
ur
e
f
e
a
tu
r
e
a
nd
a
r
ti
f
ic
ia
l
ne
ur
a
l
ne
twor
k
(
A
N
N
)
to
lo
c
a
li
z
e
c
e
r
vi
c
a
l
c
a
nc
e
r
a
r
e
a
s
a
nd
c
a
te
gor
ie
s
C
T
im
a
ge
s
in
to
nor
m
a
l
a
nd
a
bnor
m
a
l.
C
ur
ia
[
18]
in
tr
oduc
e
s
a
n
e
ns
e
m
bl
e
m
e
th
od
f
or
c
e
r
vi
c
a
l
c
a
nc
e
r
f
or
e
c
a
s
ti
ng
f
oc
us
in
g
on
m
in
im
iz
in
g
e
r
r
or
s
a
nd
f
a
ls
e
pos
it
iv
e
s
.
I
t
a
l
s
o
in
te
gr
a
te
s
e
xpl
a
in
a
bi
li
ty
a
nd
in
te
r
pr
e
ta
bi
li
ty
f
e
a
tu
r
e
s
to
m
a
ke
th
e
m
ode
l'
s
r
e
s
ul
ts
a
nd
de
c
is
io
n
-
m
a
ki
ng
pr
oc
e
s
s
e
s
m
or
e
pr
e
c
is
e
a
nd
m
or
e
unde
r
s
ta
nda
bl
e
in
th
e
c
ont
e
xt
of
c
li
ni
c
a
l
de
c
is
io
ns
.
Y
oune
s
z
a
de
e
t
al
.
[
19]
r
e
vi
e
w
e
d
r
e
c
e
nt
A
I
-
ba
s
e
d
a
ppr
oa
c
he
s
to
im
pr
ove
c
om
put
e
r
-
a
id
e
d
di
a
gnos
ti
c
(
C
A
D
)
s
ys
te
m
s
f
or
c
e
r
vi
c
a
l
c
a
nc
e
r
s
c
r
e
e
ni
ng
im
a
ge
s
.
T
he
y
id
e
nt
if
ie
d
a
r
e
a
s
w
he
r
e
f
ur
th
e
r
r
e
s
e
a
r
c
h
is
ne
e
de
d
a
nd
out
li
ne
d
f
ut
ur
e
di
r
e
c
ti
ons
.
D
hi
vya
e
t
al
.
[
20]
e
va
lu
a
te
d
th
e
pe
r
f
or
m
a
nc
e
of
di
f
f
e
r
e
nt
s
upe
r
vi
s
e
d
c
la
s
s
if
ie
r
s
f
or
a
ut
om
a
ti
c
c
e
r
vi
c
a
l
tu
m
our
c
la
s
s
if
ic
a
ti
on.
T
he
y
de
ve
lo
pe
d
a
n
opt
im
iz
e
d
m
e
ta
-
le
a
r
ni
ng
a
lg
or
it
hm
to
tu
ne
th
e
hype
r
pa
r
a
m
e
te
r
s
of
th
e
c
la
s
s
if
ie
r
s
a
nd
s
e
le
c
t
th
e
be
s
t
m
ode
l
f
or
th
e
gi
ve
n
da
ta
s
e
t.
Z
hou
e
t
al
.
[
21]
de
ve
lo
pe
d
a
f
r
a
m
e
w
or
k
in
c
or
po
r
a
ti
ng
va
r
io
u
s
s
ta
ti
s
ti
c
a
l
in
di
c
a
to
r
s
to
a
na
ly
z
e
r
is
k
f
a
c
to
r
s
m
or
e
e
f
f
ic
ie
nt
ly
ba
s
e
d
on
th
or
ough
c
or
r
e
la
ti
on
a
na
ly
s
i
s
a
nd
c
lo
s
e
r
e
la
ti
ons
hi
p
s
a
m
ong
in
di
c
a
to
r
s
d
e
r
iv
e
d
f
r
om
e
xi
s
ti
ng
li
te
r
a
tu
r
e
.
P
r
iy
a
a
nd
K
a
r
th
ik
e
ya
n
[
22]
a
ddr
e
s
s
th
e
li
m
it
a
ti
ons
of
e
xi
s
ti
ng
m
ode
ls
,
s
uc
h
a
s
lo
w
a
c
c
ur
a
c
y
a
nd
hi
gh
pr
oc
e
s
s
in
g
ti
m
e
in
c
e
r
vi
c
a
l
c
a
nc
e
r
de
te
c
ti
on.
T
he
ir
m
e
th
od
us
e
s
S
M
O
T
E
to
a
ddr
e
s
s
c
la
s
s
im
ba
la
nc
e
is
s
ue
s
,
th
e
a
r
ti
f
ic
ia
l
be
e
c
ol
ony
(
A
B
C
)
opt
im
iz
a
ti
on a
lg
or
it
hm
to
s
e
le
c
t
th
e
m
os
t
e
s
s
e
nt
ia
l
f
e
a
tu
r
e
s
,
a
nd
th
e
lo
ng
s
hor
t
-
te
r
m
m
e
m
or
y
(
L
S
T
M
)
le
a
r
ni
ng
m
ode
l
to
c
la
s
s
if
y
c
e
r
vi
c
a
l
c
a
nc
e
r
ba
s
e
d
on
th
e
s
e
le
c
te
d
f
e
a
tu
r
e
s
.
Ş
e
nt
ür
k
a
nd
U
z
un
[
23]
a
ppl
ie
d
a
m
e
di
a
n
f
il
te
r
to
r
e
m
ove
noi
s
e
f
r
om
P
a
p
s
m
e
a
r
im
a
ge
s
a
nd
us
e
d
tr
a
ns
f
e
r
le
a
r
ni
ng
f
o
r
e
a
r
ly
c
e
r
vi
c
a
l
c
a
nc
e
r
di
a
gnos
is
.
M
uda
w
i
a
nd
A
la
z
e
b
[
24]
c
om
pa
r
e
d
th
e
pe
r
f
or
m
a
nc
e
of
di
f
f
e
r
e
nt
c
la
s
s
ic
a
l
s
upe
r
vi
s
e
d
c
la
s
s
if
ie
r
s
f
or
e
a
r
ly
c
e
r
vi
c
a
l
c
a
nc
e
r
pr
e
di
c
ti
on.
T
he
y
f
ound
th
a
t
r
a
ndom
f
or
e
s
t
a
nd
de
c
is
io
n
tr
e
e
c
la
s
s
if
ie
r
s
a
c
hi
e
v
e
d
th
e
hi
ghe
s
t
c
la
s
s
if
ic
a
ti
on s
c
or
e
s
.
T
a
ni
m
u
e
t
al
.
[
25]
in
te
gr
a
te
d
r
e
c
ur
s
iv
e
f
e
a
tu
r
e
e
li
m
in
a
ti
on
a
nd
L
1
r
e
gul
a
r
iz
a
ti
on
m
e
th
ods
to
id
e
nt
if
y
th
e
m
os
t
pe
r
ti
ne
nt
a
tt
r
ib
ut
e
s
f
o
r
th
e
de
c
is
io
n
tr
e
e
c
la
s
s
if
ie
r
f
or
pr
e
c
is
e
c
e
r
vi
c
a
l
c
a
nc
e
r
pr
e
di
c
ti
ons
.
H
e
nc
e
, i
t
c
a
n be
r
e
a
li
z
e
d t
ha
t
th
e
e
xi
s
ti
ng l
it
e
r
a
tu
r
e
c
ons
is
ts
of
a
w
id
e
r
a
nge
of
s
ol
ut
io
ns
t
o de
ve
lo
p a
pr
e
di
c
ti
ve
m
ode
l
f
or
c
e
r
vi
c
a
l
c
a
nc
e
r
.
S
in
c
e
c
li
ni
c
a
l
da
ta
of
te
n
s
uf
f
e
r
s
f
r
om
da
ta
in
c
ons
is
te
nc
y
a
nd
in
c
om
pl
e
te
ne
s
s
.
I
t
ha
s
b
e
e
n
id
e
nt
if
ie
d
th
a
t
m
o
s
t
e
xi
s
ti
ng w
or
ks
l
a
c
k
s
ophi
s
ti
c
a
te
d
f
e
a
tu
r
e
e
ngi
ne
e
r
in
g
a
nd
a
r
e
m
or
e
s
pe
c
if
ic
to
a
ddr
e
s
s
in
g
e
it
he
r
c
la
s
s
im
ba
la
nc
e
pr
obl
e
m
s
or
f
e
a
tu
r
e
s
e
le
c
ti
on.
S
om
e
s
tu
di
e
s
c
ons
id
e
r
e
d
a
ddr
e
s
s
in
g
bot
h
but
w
e
r
e
s
ubj
e
c
te
d
to
bi
a
s
e
d
pr
e
di
c
ti
ve
le
a
r
ni
ng,
hi
nde
r
in
g
th
e
ir
a
ppl
ic
a
bi
li
ty
in
r
e
a
l
-
w
or
ld
a
ppl
ic
a
ti
ons
.
T
hi
s
pa
pe
r
in
tr
oduc
e
s
a
nove
l
c
e
r
vi
c
a
l
c
a
n
c
e
r
r
is
k
a
s
s
e
s
s
m
e
nt
f
r
a
m
e
w
or
k,
e
m
pl
oyi
ng
th
e
pow
e
r
of
da
ta
a
na
ly
ti
c
s
to
s
c
r
ut
in
iz
e
va
r
io
us
r
is
k
f
a
c
to
r
s
.
T
h
e
pr
opo
s
e
d
f
r
a
m
e
w
or
k
th
e
n
ut
il
iz
e
s
A
I
a
lg
or
it
hm
s
to
in
f
e
r
f
r
om
c
li
ni
c
a
l
da
ta
,
pr
e
di
c
ti
ng
th
e
pr
oba
bi
li
ty
of
in
d
iv
id
ua
ls
w
it
h
a
hi
gh
r
is
k
of
de
ve
lo
pi
ng
c
e
r
vi
c
a
l
c
a
nc
e
r
. T
he
n
e
xt
s
e
c
ti
on pr
e
s
e
nt
s
t
he
f
r
a
m
e
w
or
k de
s
ig
n a
nd t
he
i
m
pl
e
m
e
nt
a
ti
on pr
oc
e
dur
e
a
dopt
e
d.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
P
r
oac
ti
v
e
c
e
r
v
ic
al
c
anc
e
r
r
i
s
k
as
s
e
s
s
m
e
nt
u
s
in
g data
-
dr
iv
e
n an
al
y
ti
c
s
(
Sr
e
e
la
th
a)
4303
2.
M
E
T
H
O
D
T
hi
s
s
e
c
ti
on
pr
e
s
e
nt
s
th
e
pr
opos
e
d
s
ys
te
m
a
na
ly
ti
c
a
l
d
e
s
ig
n
a
nd
a
de
ta
il
e
d
di
s
c
u
s
s
io
n
of
th
e
c
om
put
in
g
pr
oc
e
dur
e
a
dopt
e
d
in
th
e
s
y
s
te
m
im
pl
e
m
e
nt
a
ti
on
f
o
r
c
e
r
vi
c
a
l
c
a
nc
e
r
r
is
k
pr
e
di
c
ti
on
ba
s
e
d
on
th
e
in
put
pa
ti
e
nt
'
s
de
m
ogr
a
phi
c
a
nd
ha
bi
ts
-
r
e
la
te
d
in
f
o
r
m
a
ti
on.
B
a
s
ic
a
ll
y,
th
e
pr
opos
e
d
r
e
s
e
a
r
c
h
f
oc
us
e
s
on
de
ve
lo
pi
ng
a
n
e
a
r
ly
de
te
c
ti
on
m
ode
l
f
or
c
e
r
vi
c
a
l
c
a
nc
e
r
,
w
hi
c
h
is
of
ut
m
os
t
im
por
ta
nc
e
,
gi
ve
n
th
e
r
e
a
l
-
w
or
l
d
c
ha
ll
e
nge
s
s
ur
r
ounding
c
a
nc
e
r
di
a
gnos
is
a
nd
it
s
a
s
s
oc
ia
t
e
d
h
e
a
lt
h
im
pl
ic
a
ti
ons
.
T
he
m
a
in
obj
e
c
ti
ve
of
th
e
pr
opos
e
d
m
e
th
odol
ogy
is
to
c
r
e
a
te
a
n
a
c
c
ur
a
t
e
a
nd
r
e
li
a
bl
e
pr
e
di
c
ti
ve
m
ode
l
th
a
t
c
a
n
be
ut
il
iz
e
d
to
id
e
nt
if
y
pa
ti
e
nt
s
w
ho
a
r
e
a
t
r
is
k
of
c
e
r
vi
c
a
l
c
a
nc
e
r
.
T
hi
s
m
ode
l
w
il
l
a
s
s
is
t
he
a
lt
hc
a
r
e
pr
of
e
s
s
io
na
l
s
in
m
a
ki
ng
in
f
or
m
e
d
de
c
is
io
ns
a
bout
w
hi
c
h
pa
ti
e
nt
s
s
houl
d
unde
r
go
m
or
e
c
om
pr
e
he
ns
iv
e
e
xa
m
in
a
ti
ons
.
T
he
s
c
he
m
a
ti
c
a
r
c
hi
te
c
tu
r
e
of
t
he
pr
opos
e
d c
e
r
vi
c
a
l
c
a
n
c
e
r
r
is
k pr
e
di
c
ti
on f
r
a
m
e
w
or
k i
s
i
ll
us
tr
a
te
d i
n F
ig
ur
e
1.
F
ig
ur
e
1. I
ll
us
tr
a
te
s
t
he
s
c
he
m
a
ti
c
a
r
c
hi
te
c
tu
r
e
of
t
he
pr
opos
e
d f
r
a
m
e
w
or
k, f
ol
lo
w
in
g a
bl
oc
k
-
ba
s
e
d w
or
kf
lo
w
T
he
m
e
th
odol
ogy
a
dopt
e
d
in
th
e
pr
opos
e
d
s
y
s
te
m
a
dopt
s
a
s
ys
te
m
a
ti
c
im
pl
e
m
e
nt
a
ti
on
pr
oc
e
dur
e
.
T
he
f
ir
s
t
s
te
p c
a
r
r
ie
d out
i
n t
he
s
ys
te
m
de
ve
lo
pm
e
nt
i
s
e
xpl
or
a
t
or
y da
ta
a
na
ly
s
is
(
E
D
A
)
t
o ga
in
in
s
ig
ht
s
a
bout
th
e
da
ta
s
e
t
a
tt
r
ib
ut
e
s
a
nd
unde
r
s
ta
nd
th
e
c
om
pl
e
xi
ty
of
th
e
da
ta
s
e
t.
T
hi
s
pha
s
e
in
vol
ve
s
vi
s
ua
l
a
na
ly
s
is
,
c
he
c
ki
ng
m
is
s
in
g
va
lu
e
s
a
nd
pe
r
f
or
m
in
g
de
s
c
r
ip
ti
ve
a
n
a
ly
s
is
. N
e
xt
,
a
pr
e
pr
oc
e
s
s
in
g
m
e
c
h
a
ni
s
m
is
a
dopt
e
d
to
pr
ovi
de
s
ui
ta
bl
e
tr
e
a
tm
e
nt
to
th
e
da
ta
s
e
t
to
a
ddr
e
s
s
a
ny
out
li
e
r
s
a
nd
in
c
ons
is
te
nc
ie
s
.
F
ur
th
e
r
,
s
ta
ti
s
ti
c
a
l
a
na
ly
s
is
is
don
e
to
und
e
r
s
ta
nd
a
nd
id
e
nt
if
y
pot
e
nt
ia
l
r
e
l
a
ti
ons
h
ip
s
be
twe
e
n
r
is
k
a
tt
r
ib
ut
e
s
a
nd
out
c
om
e
c
la
s
s
la
be
ls
.
T
hi
s
pr
oc
e
s
s
he
lp
s
to
a
n
a
ly
z
e
a
nd
s
e
le
c
t
th
e
pr
e
di
c
ti
v
e
m
ode
l'
s
r
e
le
va
nt
a
tt
r
ib
ut
e
s
.
T
h
e
s
tu
dy
th
e
n
pe
r
f
or
m
s
a
r
a
nki
ng of
t
he
f
e
a
tu
r
e
s
unde
r
c
ons
id
e
r
a
ti
on us
in
g a
n
e
xt
r
a
-
tr
e
e
c
la
s
s
if
ie
r
t
e
s
t.
A
f
te
r
w
a
r
ds
, t
he
m
os
t
dom
in
a
ti
ng
f
e
a
tu
r
e
s
a
r
e
s
e
le
c
t
e
d
ba
s
e
d
on
th
e
r
a
nki
ng s
c
or
e
. T
h
e
to
p
opt
im
a
l
f
e
a
tu
r
e
s
a
r
e
th
e
n
ve
c
to
r
iz
e
d a
nd
s
ubj
e
c
te
d
to
th
e
le
a
r
ni
ng
a
na
ly
ti
c
s
-
ba
s
e
d
pr
e
di
c
ti
ve
ta
s
k.
T
he
s
tu
dy
in
th
i
s
pha
s
e
e
m
pl
oye
d
va
r
io
us
c
om
put
a
ti
ona
l
in
te
ll
ig
e
nc
e
,
pa
r
ti
c
ul
a
r
ly
a
s
ha
ll
ow
m
a
c
hi
ne
le
a
r
ni
ng
c
la
s
s
if
ie
r
,
w
hi
c
h
is
tr
a
in
e
d
us
in
g
s
e
le
c
te
d
f
e
a
tu
r
e
s
a
nd
d
e
pl
oye
d
to
p
e
r
f
or
m
pr
e
di
c
ti
on
of
th
e
pr
e
s
e
n
c
e
of
hi
gh
-
r
is
k
H
P
V
s
tr
a
in
s
or
c
e
r
vi
c
a
l
c
a
nc
e
r
r
is
k.
T
he
be
nc
hm
a
r
ki
ng
of
th
e
pr
opos
e
d
s
y
s
te
m
is
done
c
onc
e
r
ni
ng
c
onf
us
io
n
m
a
tr
ix
,
a
c
c
ur
a
c
y,
f
1
-
s
c
or
e
,
a
nd
c
om
pa
r
a
ti
ve
a
na
ly
s
is
w
it
h di
f
f
e
r
e
nt
m
a
c
hi
ne
l
e
a
r
ni
ng
m
ode
ls
a
nd s
im
il
a
r
e
xi
s
ti
ng s
tu
di
e
s
.
T
he
pr
opos
e
d
s
y
s
te
m
r
e
ta
in
s
hi
gh
ut
il
it
y
in
th
e
r
e
a
l
-
w
or
ld
he
a
lt
hc
a
r
e
a
ppl
ic
a
ti
on.
I
t
c
a
n
be
de
pl
oye
d
in
hos
pi
ta
ls
,
c
li
ni
c
s
,
a
nd
s
c
r
e
e
ni
ng
c
e
nt
r
e
s
,
w
he
r
e
it
c
a
n
s
e
r
ve
a
s
a
s
uppor
t
s
ys
te
m
f
or
th
e
in
it
ia
l
s
c
r
e
e
ni
ng
of
c
e
r
vi
c
a
l
c
a
n
c
e
r
a
nd
h
e
lp
in
a
be
tt
e
r
a
nd
m
or
e
r
a
pi
d
de
c
is
io
n
-
m
a
ki
ng
pr
oc
e
s
s
.
F
or
e
xa
m
pl
e
,
if
th
e
pa
ti
e
nt
is
id
e
nt
if
ie
d
or
pr
e
di
c
te
d
a
s
a
t
r
is
k
of
c
e
r
vi
c
a
l
c
a
nc
e
r
,
th
e
e
xpe
r
ts
th
e
n
pr
io
r
it
is
e
pa
ti
e
nt
s
f
or
f
ur
th
e
r
e
xa
m
in
a
ti
ons
a
nd
in
te
r
ve
nt
io
n
us
in
g a
n
a
dva
n
c
e
d
s
c
r
e
e
ni
ng
pr
oc
e
s
s
,
ul
ti
m
a
te
ly
l
e
a
di
ng
to
e
a
r
li
e
r
de
te
c
ti
on
of
c
a
nc
e
r
a
nd t
r
e
a
tm
e
nt
, t
he
r
e
by be
tt
e
r
pa
ti
e
nt
out
c
om
e
s
.
2.1. Dat
as
e
t
d
e
s
c
r
ip
t
io
n
T
he
c
e
r
vi
c
a
l
c
a
nc
e
r
da
ta
s
e
t
us
e
d
in
th
is
s
tu
dy
w
a
s
obt
a
in
e
d
f
r
o
m
th
e
m
a
c
hi
ne
le
a
r
ni
ng
r
e
pos
it
or
y
o
f
th
e
U
ni
ve
r
s
it
y
of
C
a
li
f
or
ni
a
,
I
r
vi
ne
(
U
C
I
)
[
26
]
.
T
hi
s
da
ta
s
e
t
is
e
s
s
e
nt
ia
l
f
or
pr
e
di
c
ti
ng
r
is
k
f
a
c
to
r
s
a
nd
di
a
gnos
in
g
c
e
r
vi
c
a
l
c
a
nc
e
r
.
I
t
w
a
s
c
a
r
e
f
ul
ly
pr
e
pa
r
e
d
us
in
g
da
ta
c
ol
le
c
te
d
f
r
om
858
pa
ti
e
nt
s
a
t
th
e
H
os
pi
ta
l
U
ni
ve
r
s
it
a
r
io
de
C
a
r
a
c
a
s
i
n V
e
ne
z
u
e
la
. T
h
e
s
pe
c
if
ic
de
t
a
il
s
of
t
he
da
ta
s
e
t
a
r
e
out
li
ne
d i
n
T
a
bl
e
1.
B
a
s
e
d
on
th
e
s
ta
ti
s
ti
c
s
pr
ovi
de
d
in
T
a
bl
e
1,
it
c
a
n
be
a
na
ly
z
e
d
t
ha
t
th
e
da
ta
s
e
t
in
c
lu
d
e
s
de
m
ogr
a
phi
c
in
f
or
m
a
ti
on,
li
f
e
s
ty
le
ha
bi
ts
,
a
nd
m
e
di
c
a
l
hi
s
to
r
y.
T
hi
s
da
ta
s
e
t
c
ont
a
in
s
f
our
c
r
uc
ia
l
r
e
s
pons
e
va
r
ia
bl
e
s
f
or
c
e
r
vi
c
a
l
c
a
nc
e
r
pr
e
di
c
ti
ve
m
ode
ll
in
g:
H
in
s
e
lm
a
nn,
S
c
hi
ll
e
r
,
C
it
ol
ogy,
a
nd
B
io
ps
y.
T
he
s
e
f
e
a
tu
r
e
s
a
r
e
e
s
s
e
nt
ia
l
f
or
di
a
gnos
in
g
c
e
r
vi
c
a
l
c
a
nc
e
r
in
pa
ti
e
nt
s
.
T
he
ir
a
c
c
um
ul
a
ti
on
pr
ovi
de
s
a
c
om
pr
e
he
ns
iv
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
. 13, No. 4, D
e
c
e
m
be
r
2024:
4301
-
4311
4304
r
e
pr
e
s
e
nt
a
ti
on
of
th
e
va
r
ia
bl
e
s
th
a
t
in
f
lu
e
nc
e
c
e
r
vi
c
a
l
c
a
nc
e
r
.
T
he
da
ta
s
e
t
c
ont
a
in
s
m
is
s
in
g
va
lu
e
s
,
a
s
s
om
e
pa
ti
e
nt
s
de
c
li
ne
d t
o a
n
s
w
e
r
c
e
r
ta
in
que
s
ti
ons
due
t
o pr
iv
a
c
y c
on
c
e
r
ns
.
T
a
bl
e
1. S
um
m
a
r
y of
t
he
c
e
r
vi
c
a
l
c
a
nc
e
r
r
is
k da
ta
s
e
t
S
I
. N
o
A
t
t
r
i
but
e
s
D
a
t
a
T
ype
D
e
s
c
r
i
pt
i
on
0
A
ge
i
nt
32
A
ge
of
a
w
om
a
n
1
N
um
be
r
of
s
e
xua
l
pa
r
t
ne
r
s
i
nt
32
T
ot
a
l
num
be
r
of
s
e
xua
l
pa
r
t
ne
r
s
2
F
i
r
s
t
s
e
xua
l
i
nt
e
r
c
our
s
e
i
nt
32
A
ge
of
a
w
om
a
n w
he
n s
he
h
a
d he
r
f
i
r
s
t
s
e
xua
l
i
nt
e
r
c
our
s
e
.
3
N
um
of
pr
e
gna
nc
i
e
s
i
nt
32
T
ot
a
l
num
be
r
of
t
i
m
e
s
t
he
w
om
a
n got
pr
e
gna
nt
.
4
S
m
oke
s
bool
W
he
t
he
r
t
he
w
om
e
n s
m
oke
s
or
not
.
5
S
m
oke
s
(
ye
a
r
s
)
i
nt
32
N
um
be
r
of
ye
a
r
s
f
or
w
hi
c
h t
he
w
om
a
n i
s
s
m
oki
ng.
6
S
m
oke
s
(
pa
c
ks
/
ye
a
r
)
i
nt
32
T
ot
a
l
num
be
r
of
pa
c
ke
t
s
of
c
i
ga
r
e
t
t
e
s
pe
r
ye
a
r
t
he
w
om
a
n s
m
oke
s
.
7
H
or
m
ona
l
C
ont
r
a
c
e
pt
i
ve
s
bool
W
he
t
he
r
t
he
w
om
e
n us
e
hor
m
ona
l
c
ont
r
a
c
e
pt
i
ve
s
or
not
.
8
H
or
m
ona
l
C
ont
r
a
c
e
pt
i
ve
s
(
ye
a
r
s
)
i
nt
32
T
ot
a
l
ye
a
r
s
f
or
w
hi
c
h c
ont
r
a
c
e
pt
i
ve
m
e
t
hod w
a
s
us
e
d by w
om
e
n.
9
I
U
D
bool
T
he
i
nt
r
a
ut
e
r
i
ne
c
ont
r
a
c
e
pt
i
ve
de
vi
c
e
w
a
s
us
e
d or
not
.
10
I
U
D
(
ye
a
r
s
)
i
nt
32
F
or
how
m
a
ny ye
a
r
s
t
he
I
U
D
w
a
s
us
e
d.
11
S
T
D
s
bool
T
he
pr
e
s
e
nc
e
of
s
e
xua
l
l
y t
r
a
ns
m
i
t
t
e
d di
s
e
a
s
e
s
(
S
T
D
)
.
12
S
T
D
s
(
num
be
r
)
i
nt
32
T
ot
a
l
num
be
r
of
S
T
D
pr
e
s
e
nt
w
i
t
h t
he
pa
t
i
e
nt
.
13
S
T
D
s
:
c
ondyl
om
a
t
os
i
s
bool
T
he
pr
e
s
e
nc
e
of
c
ondyl
om
a
t
os
i
s
w
i
t
h t
he
pa
t
i
e
nt
.
14
S
T
D
s
:
c
e
r
vi
c
a
l
c
ondyl
om
a
t
os
i
s
bool
T
he
pr
e
s
e
nc
e
of
c
e
r
vi
c
a
l
c
ondyl
om
a
t
os
i
s
.
15
S
T
D
s
:
va
gi
na
l
c
ondyl
om
a
t
os
i
s
bool
T
he
pr
e
s
e
nc
e
of
va
gi
na
l
c
ondyl
om
a
t
os
i
s
.
16
S
T
D
s
:
vul
vo
-
pe
r
i
ne
a
l
c
ondyl
om
a
t
os
i
s
bool
T
he
pr
e
s
e
nc
e
of
vul
vo
-
pe
r
i
ne
a
l
c
ondyl
om
a
t
os
i
s
.
17
S
T
D
s
:
s
yphi
l
i
s
bool
T
he
pr
e
s
e
nc
e
of
s
yphi
l
i
s
.
18
S
T
D
s
:
pe
l
vi
c
i
nf
l
a
m
m
a
t
or
y di
s
e
a
s
e
bool
T
he
pr
e
s
e
nc
e
of
pe
l
vi
c
i
nf
l
a
m
m
a
t
or
y di
s
e
a
s
e
.
19
S
T
D
s
:
ge
ni
t
a
l
he
r
pe
s
bool
T
he
pr
e
s
e
nc
e
of
ge
ni
t
a
l
he
r
pe
s
.
20
S
T
D
s
:
m
ol
l
us
c
um
c
ont
a
gi
os
um
bool
T
he
pr
e
s
e
nc
e
of
m
ol
l
us
c
um
c
ont
a
gi
os
um
.
21
S
T
D
s
:
A
I
D
S
bool
T
he
pr
e
s
e
nc
e
of
A
I
D
S
i
n t
he
pa
t
i
e
nt
.
22
S
T
D
s
:
H
I
V
bool
T
he
pr
e
s
e
nc
e
of
H
I
V
i
n t
he
pa
t
i
e
nt
.
23
S
T
D
s
:
H
e
pa
t
i
t
i
s
B
bool
T
he
pr
e
s
e
nc
e
of
he
pa
t
i
t
i
s
B
i
n t
he
pa
t
i
e
nt
s
.
24
S
T
D
s
:
H
P
V
bool
T
he
pr
e
s
e
nc
e
of
H
P
V
i
n t
he
pa
t
i
e
nt
s
.
25
S
T
D
s
:
N
um
be
r
of
di
a
gnos
e
s
i
nt
32
T
he
t
ot
a
l
num
be
r
of
t
i
m
e
s
t
he
S
T
D
s
ha
ve
be
e
n di
a
gnos
e
d.
26
S
T
D
s
:
T
i
m
e
s
i
nc
e
f
i
r
s
t
di
a
gnos
i
s
i
nt
32
T
he
t
ot
a
l
num
be
r
of
ye
a
r
s
s
i
nc
e
t
he
f
i
r
s
t
di
a
gnos
e
.
27
S
T
D
s
:
T
i
m
e
s
i
nc
e
l
a
s
t
di
a
gnos
i
s
i
nt
32
T
he
t
ot
a
l
num
be
r
of
ye
a
r
s
e
l
a
ps
e
d s
i
nc
e
t
he
l
a
s
t
di
a
gnos
e
.
28
D
x:
C
a
nc
e
r
bool
T
he
pr
e
s
e
nc
e
of
c
a
nc
e
r
a
f
t
e
r
t
he
di
a
gnos
e
.
29
D
x:
C
I
N
bool
T
he
pr
e
s
e
nc
e
of
c
e
r
vi
c
a
l
i
nt
r
a
e
pi
t
he
l
i
a
l
ne
opl
a
s
i
a
.
30
D
x:
H
P
V
bool
T
he
pr
e
s
e
nc
e
of
hum
a
n
pa
pi
l
l
om
a
vi
r
us
e
s
(
H
P
V
)
.
31
Dx
bool
T
he
pr
e
s
e
nc
e
of
C
I
N
or
H
P
V
.
32
H
i
ns
e
l
m
a
nn
bool
A
c
ol
pos
c
opy t
e
s
t
t
o e
x
a
m
i
ne
a
m
a
gni
f
i
e
d vi
e
w
of
t
he
c
e
r
vi
x
33
S
c
hi
l
l
e
r
bool
I
odi
ne
t
e
s
t
t
o
di
a
gnos
e
c
e
r
vi
c
a
l
c
a
n
c
e
r
.
34
C
i
t
ol
ogy
bool
A
P
a
P
s
m
e
a
r
s
t
e
s
t
, he
l
p
s
de
t
e
c
t
a
bnor
m
a
l
c
e
l
l
s
i
n t
he
c
e
r
vi
x
35
B
i
ops
y
bool
A
s
ur
gi
c
a
l
pr
oc
e
dur
e
w
he
r
e
a
s
m
a
l
l
t
i
s
s
ue
i
s
r
e
m
ove
d f
r
om
t
he
c
e
r
vi
x
2.2. P
r
e
p
r
oc
e
s
s
in
g
P
r
e
pr
oc
e
s
s
in
g
is
a
n
im
por
ta
nt
ope
r
a
ti
on
in
da
ta
-
dr
iv
e
n
a
na
l
yt
ic
s
a
s
it
a
lo
ne
c
ont
r
ib
ut
e
s
70%
to
pe
r
f
or
m
a
nc
e
im
pr
ove
m
e
nt
.
T
he
in
it
ia
l
pr
e
p
r
oc
e
s
s
in
g
s
te
p
in
vol
ve
s
im
por
ti
ng
th
e
da
ta
s
e
t
a
nd
c
onduc
ti
ng
E
D
A
to
unde
r
s
ta
nd
it
s
c
ha
r
a
c
te
r
is
ti
c
s
a
nd
in
he
r
e
nt
c
om
pl
e
xi
ti
e
s
of
th
e
da
ta
s
e
t.
D
ur
in
g
E
D
A
,
m
is
s
in
g
da
ta
in
s
ta
nc
e
s
r
e
pr
e
s
e
nt
e
d
a
s
'
?
'
w
e
r
e
id
e
nt
if
ie
d.
T
o
s
t
a
nda
r
di
z
e
th
e
ha
ndl
in
g
of
th
e
s
e
m
is
s
in
g
va
lu
e
s
,
'
?
'
in
s
t
a
nc
e
s
w
e
r
e
r
e
pl
a
c
e
d
w
it
h
'
not
a
num
be
r
'
(
NAN
)
.
I
m
put
in
g
m
is
s
in
g
va
lu
e
s
is
c
r
it
ic
a
l
f
or
e
ns
ur
in
g
da
ta
in
te
gr
it
y
a
nd
a
c
c
ur
a
c
y,
th
e
r
e
by
im
pr
ovi
ng
m
ode
l
pr
e
c
is
io
n
a
nd
m
it
ig
a
ti
ng
p
ot
e
nt
ia
l
bi
a
s
e
s
.
T
he
da
ta
s
e
t
c
ont
a
in
s
num
e
r
ic
a
l
a
nd
c
a
te
gor
ic
a
l
a
tt
r
ib
ut
e
s
,
s
o
th
e
im
put
a
ti
on
a
ppr
oa
c
h
va
r
ie
d
a
c
c
or
di
ngl
y.
F
or
num
e
r
ic
a
l
a
tt
r
ib
ut
e
s
,
m
is
s
in
g
va
lu
e
s
w
e
r
e
s
ub
s
ti
tu
te
d w
it
h t
he
m
e
di
a
n va
lu
e
of
non
-
nul
l
e
nt
r
i
e
s
i
n t
he
r
e
s
pe
c
ti
ve
f
e
a
tu
r
e
, a
nd f
or
c
a
te
gor
ic
a
l
a
tt
r
ib
ut
e
s
,
m
is
s
in
g
va
lu
e
s
w
e
r
e
r
e
pl
a
c
e
d
w
it
h
th
e
m
ode
,
i.
e
.,
th
e
m
os
t
f
r
e
que
nt
ly
oc
c
ur
r
in
g
non
-
nul
l
va
lu
e
w
it
hi
n t
he
f
e
a
tu
r
e
.
A
ne
w
c
ol
um
n
ti
tl
e
d
"
c
a
nc
e
r
r
is
k"
w
a
s
a
ls
o
in
tr
oduc
e
d
to
a
ug
m
e
nt
th
e
da
ta
s
e
t'
s
pr
e
di
c
ti
ve
c
a
p
a
c
it
y.
T
hi
s
c
ol
um
n c
om
bi
ne
s
th
e
e
nt
r
ie
s
f
r
om
th
e
"
D
x:
C
a
nc
e
r
"
a
nd
"
D
x:
C
I
N
"
c
ol
um
ns
.
I
t
is
im
por
ta
nt
to
not
e
th
a
t
"
D
x:
C
I
N
"
is
in
di
c
a
ti
ve
of
a
pr
e
c
a
nc
e
r
ous
c
ondi
ti
on,
in
di
c
a
ti
ng
th
a
t
pa
ti
e
nt
s
di
a
gnos
e
d
w
it
h
C
I
N
ha
ve
a
n
in
c
r
e
a
s
e
d
r
is
k
of
de
ve
lo
pi
ng
in
va
s
iv
e
c
a
n
c
e
r
.
T
he
m
ot
iv
a
ti
on
f
or
a
ddi
ng
th
is
c
ol
um
n
w
a
s
th
e
r
e
la
ti
ve
ly
lo
w
in
c
id
e
nc
e
r
a
te
s
in
th
e
in
di
vi
dua
l
c
ol
um
ns
:
"
D
x:
C
a
nc
e
r
"
ha
d
1
8
pos
it
iv
e
in
s
ta
n
c
e
s
,
a
nd
"
D
x:
C
I
N
"
ha
d
9.
B
y
m
e
r
gi
ng
th
e
da
ta
f
r
om
th
e
s
e
two
c
ol
um
ns
,
a
to
ta
l
of
27
in
s
ta
nc
e
s
in
di
c
a
ti
ng
c
a
nc
e
r
di
a
gno
s
is
w
e
r
e
obt
a
in
e
d,
w
hi
c
h i
s
m
or
e
r
e
pr
e
s
e
nt
a
ti
ve
of
t
he
unde
r
ly
in
g r
is
k f
a
c
to
r
s
.
2.3. F
e
at
u
r
e
e
xt
r
ac
t
io
n
an
d
s
e
le
c
t
io
n
D
im
e
ns
io
na
li
ty
r
e
duc
ti
on i
s
c
r
uc
ia
l
in
da
ta
pr
e
pr
oc
e
s
s
in
g a
s
i
t
e
li
m
in
a
te
s
i
r
r
e
le
va
nt
f
e
a
tu
r
e
s
a
nd only
r
e
ta
in
s
im
por
ta
nt
a
nd
hi
ghl
y
c
or
r
e
la
te
d
f
e
a
tu
r
e
s
. T
hi
s
pr
oc
e
s
s
s
i
m
pl
if
ie
s
th
e
da
ta
s
e
t
by
r
e
m
ovi
ng
unn
e
c
e
s
s
a
r
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
P
r
oac
ti
v
e
c
e
r
v
ic
al
c
anc
e
r
r
i
s
k
as
s
e
s
s
m
e
nt
u
s
in
g data
-
dr
iv
e
n an
al
y
ti
c
s
(
Sr
e
e
la
th
a)
4305
f
e
a
tu
r
e
s
a
nd
e
na
bl
e
s
th
e
m
ode
l
to
id
e
nt
if
y
a
nd
u
ti
li
z
e
ke
y
pa
tt
e
r
ns
in
th
e
da
ta
m
or
e
e
f
f
e
c
ti
ve
ly
.
B
y
r
e
duc
in
g
th
e
num
be
r
of
f
e
a
tu
r
e
s
,
th
e
m
ode
l
be
c
om
e
s
m
or
e
e
f
f
ic
ie
nt
a
n
d
le
s
s
pr
one
to
ove
r
f
it
ti
ng,
w
he
r
e
it
le
a
r
ns
to
m
e
m
or
ie
s
th
e
tr
a
in
in
g
da
ta
r
a
th
e
r
th
a
n
g
e
ne
r
a
li
z
e
to
ne
w
,
un
s
e
e
n
d
a
ta
.
I
t
a
l
s
o
r
e
duc
e
s
th
e
c
om
put
a
ti
ona
l
c
om
pl
e
xi
ty
in
da
ta
pr
oc
e
s
s
in
g
a
nd
dur
in
g
m
ode
l
tr
a
in
in
g.
T
he
i
ni
ti
a
l
s
te
p
in
th
e
pr
opos
e
d
f
e
a
tu
r
e
e
ngi
ne
e
r
in
g
pr
oc
e
s
s
in
vol
ve
s
s
pl
it
ti
ng
th
e
da
ta
s
e
t
in
to
tr
a
in
in
g
a
nd
te
s
ti
ng
s
ubs
e
ts
u
s
in
g
a
70
-
30%
s
pl
it
r
a
ti
o.
F
ol
lo
w
in
g
th
is
pa
r
ti
ti
oni
ng,
nor
m
a
li
z
a
ti
on
is
pe
r
f
or
m
e
d
to
s
c
a
le
th
e
num
e
r
ic
a
l
a
tt
r
ib
ut
e
s
of
th
e
d
a
ta
s
e
t
to
a
s
ta
nda
r
di
z
e
d
r
a
nge
,
ty
pi
c
a
ll
y
[
0,1]
.
T
hi
s
nor
m
a
li
z
a
ti
on
e
ns
ur
e
s
th
a
t
no
s
in
gl
e
f
e
a
tu
r
e
di
s
pr
opor
ti
ona
te
ly
in
f
lu
e
nc
e
s
th
e
m
ode
l
due
t
o i
ts
s
c
a
le
.
n
o
r
m
a
l
iz
e
d
(
x
)
=
x
−
min
(
x
)
ma
x
(
x
)
−
min
(
x
)
(
1)
T
he
s
tu
dy
th
e
n
p
e
r
f
or
m
s
f
e
a
tu
r
e
s
e
le
c
ti
on
us
in
g
th
e
E
xt
r
a
T
r
e
e
c
la
s
s
if
ie
r
,
a
tr
e
e
-
ba
s
e
d
e
n
s
e
m
bl
e
le
a
r
ni
ng
te
c
hni
que
.
U
nl
ik
e
r
e
gul
a
r
de
c
is
io
n
tr
e
e
s
th
a
t
c
hoos
e
t
he
be
s
t
s
pl
it
a
m
ong
a
s
e
t
of
a
va
il
a
bl
e
f
e
a
tu
r
e
s
,
E
xt
r
a
T
r
e
e
s
e
le
c
t
s
f
e
a
tu
r
e
s
a
nd
s
pl
it
poi
nt
s
a
t
r
a
ndom,
w
hi
c
h
he
lp
s
r
e
duc
e
th
e
va
r
ia
nc
e
.
T
he
de
c
is
io
n
-
m
a
ki
ng
pr
oc
e
s
s
of
th
e
tr
e
e
is
r
e
gul
a
te
d
by
e
nt
r
opy
va
lu
e
,
w
hi
c
h
he
lp
s
i
n
de
te
r
m
in
in
g
th
e
pur
it
y
of
a
s
pl
it
,
num
e
r
ic
a
ll
y
gi
ve
n
a
s
(
2)
.
(
)
=
+
+
2
(
+
)
−
−
2
(
−
)
(
2)
W
he
r
e
in
(
2)
S
is
th
e
s
e
t
of
s
a
m
pl
e
s
,
p
+
is
th
e
pr
opor
ti
on
o
f
pos
it
iv
e
s
a
m
pl
e
s
a
nd
p
−
is
th
e
pr
opor
ti
on
of
ne
ga
ti
ve
s
a
m
pl
e
s
.
T
h
e
E
xt
r
a
T
r
e
e
c
la
s
s
if
ie
r
a
ggr
e
ga
te
s
m
ul
ti
pl
e
s
uc
h
tr
e
e
s
a
nd
c
om
put
e
s
a
n
a
ve
r
a
ge
to
de
te
r
m
in
e
th
e
im
por
ta
nc
e
of
e
a
c
h
f
e
a
tu
r
e
.
T
he
s
ig
ni
f
ic
a
nc
e
of
a
f
e
a
tu
r
e
is
di
r
e
c
tl
y
r
e
la
te
d
to
th
e
f
r
e
que
nc
y
w
it
h
w
hi
c
h
it
a
ppe
a
r
s
in
th
e
tr
e
e
s
a
nd
th
e
de
pt
h
a
t
w
hi
c
h
it
a
ppe
a
r
s
.
O
nc
e
th
e
f
e
a
tu
r
e
im
por
ta
nc
e
is
de
r
iv
e
d,
th
e
s
e
va
lu
e
s
a
r
e
s
ubj
e
c
te
d
to
s
ta
nda
r
di
z
a
ti
on
ope
r
a
ti
on
to
e
n
s
ur
e
th
a
t
th
e
c
ol
le
c
ti
ve
im
por
ta
nc
e
a
c
r
os
s
a
ll
f
e
a
tu
r
e
s
s
um
s
up
to
on
e
.
M
a
th
e
m
a
ti
c
a
ll
y,
gi
ve
n
a
s
e
t
of
f
e
a
tu
r
e
im
por
ta
nc
e
va
lu
e
s
F
,
th
e
nor
m
a
li
z
e
d
im
por
ta
nc
e
f
or
f
e
a
tu
r
e
i
a
s
i
n
(
3)
.
=
∑
=
1
(
3)
W
he
r
e
,
n
is
th
e
to
ta
l
num
be
r
of
f
e
a
tu
r
e
s
.
F
ol
lo
w
in
g
th
is
pr
oc
e
dur
e
,
th
e
s
tu
dy
s
e
le
c
ts
th
e
m
o
s
t
s
ig
ni
f
ic
a
nt
f
e
a
tu
r
e
s
a
nd
e
x
c
lu
de
s
th
e
le
a
s
t
s
ig
ni
f
ic
a
nt
a
tt
r
ib
ut
e
s
.
T
hi
s
pr
oc
e
s
s
not
onl
y
r
e
f
in
e
s
th
e
m
ode
l
but
a
ls
o
e
nha
nc
e
s
it
s
pe
r
f
or
m
a
nc
e
a
nd
di
m
in
is
he
s
c
om
put
a
ti
ona
l
o
ve
r
he
a
d,
of
f
e
r
in
g
a
m
or
e
e
f
f
ic
ie
nt
le
a
r
ni
ng
e
nvi
r
onm
e
nt
. F
ig
ur
e
2 i
ll
us
tr
a
te
s
t
he
t
op s
e
le
c
te
d f
e
a
tu
r
e
s
, f
ol
lo
w
in
g c
r
os
s
-
c
or
r
e
la
ti
on a
na
ly
s
is
i
n F
ig
ur
e
3.
F
ig
ur
e
2. I
ll
us
tr
a
te
s
t
he
s
c
he
m
a
ti
c
a
r
c
hi
te
c
tu
r
e
of
t
he
pr
opos
e
d f
r
a
m
e
w
or
k, f
ol
lo
w
in
g a
bl
oc
k
-
ba
s
e
d w
or
kf
lo
w
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
. 13, No. 4, D
e
c
e
m
be
r
2024:
4301
-
4311
4306
F
ig
ur
e
3. I
ll
us
tr
a
te
s
t
he
P
e
a
r
s
on
c
or
r
e
la
ti
on plot
f
or
t
he
t
op 15 s
e
le
c
te
d f
e
a
tu
r
e
s
F
ig
ur
e
3
pr
e
s
e
nt
s
a
P
e
a
r
s
on
c
or
r
e
la
ti
on
pl
ot
f
or
th
e
to
p
15
s
e
le
c
te
d
f
e
a
tu
r
e
s
.
A
P
e
a
r
s
on c
or
r
e
la
ti
on
is
a
s
ta
ti
s
ti
c
a
l
m
e
a
s
ur
e
t
ha
t
qua
nt
if
ie
s
t
he
s
tr
e
ngt
h a
nd dir
e
c
ti
on of
t
he
l
in
e
a
r
r
e
la
ti
ons
hi
p be
twe
e
n t
w
o va
r
ia
bl
e
s
.
T
he
vi
s
ua
l
a
n
a
ly
s
is
r
e
ve
a
ls
th
e
r
e
la
ti
ons
hi
p
s
be
tw
e
e
n
va
r
ia
bl
e
s
,
w
it
h
c
o
e
f
f
ic
ie
nt
s
r
a
ngi
ng
f
r
om
-
1
(
ne
ga
ti
ve
r
e
la
ti
ons
hi
p)
to
1
(
pos
it
iv
e
r
e
la
ti
ons
hi
p)
.
I
t
c
a
n
be
obs
e
r
ve
d
th
a
t
th
e
f
e
a
tu
r
e
D
x:
H
P
V
is
s
tr
ongl
y
pos
it
iv
e
l
y
c
or
r
e
la
te
d
w
it
h
C
a
nc
e
r
_r
is
k
(
0.718944)
,
I
U
D
dur
a
ti
on
a
nd
p
r
e
s
e
nc
e
a
ls
o
c
or
r
e
la
te
po
s
it
iv
e
ly
(
0.749288)
,
S
m
oki
ng
ye
a
r
s
a
nd
pa
c
ks
/y
e
a
r
e
xhi
bi
t
a
s
im
il
a
r
tr
e
nd
(
0.7243
20)
,
r
e
c
e
nt
S
T
D
di
a
gnos
e
s
c
or
r
e
la
te
w
it
h
th
e
ti
m
e
s
in
c
e
t
he
f
ir
s
t
di
a
gnos
is
(
0.935614)
, a
nd a
ge
h
a
s
a
m
ode
r
a
t
e
r
e
la
ti
ons
hi
p w
it
h pr
e
gna
nc
ie
s
(
0.525892)
.
2.4. M
od
e
l
t
r
ai
n
in
g
I
n t
he
t
r
a
in
in
g pha
s
e
, t
he
da
ta
s
e
t
is
s
pl
it
i
nt
o t
e
s
ti
ng a
nd t
r
a
in
in
g s
ubs
e
ts
, w
he
r
e
80%
of
t
he
da
ta
s
e
t
is
c
ons
id
e
r
e
d
f
or
tr
a
in
in
g
a
nd
va
li
da
ti
on
a
nd
20%
is
c
ons
id
e
r
e
d
f
or
te
s
ti
ng
pur
pos
e
s
.
T
he
s
tu
dy
im
pl
e
m
e
nt
s
di
f
f
e
r
e
nt
s
upe
r
vi
s
e
d
c
la
s
s
if
ie
r
s
to
id
e
nt
if
y
th
e
on
e
th
a
t
e
xhi
bi
te
d
th
e
hi
ghe
s
t
pr
e
c
is
io
n
a
nd
r
e
li
a
bi
li
ty
in
pr
e
di
c
ti
ons
a
nd
r
is
k
e
va
lu
a
ti
on.
A
s
ig
ni
f
ic
a
nt
c
ha
ll
e
nge
a
ddr
e
s
s
e
d
dur
in
g
th
is
m
ode
l
s
e
le
c
ti
on
pr
oc
e
s
s
is
th
e
c
la
s
s
im
ba
la
n
c
e
is
s
ue
,
w
he
r
e
one
c
la
s
s
dom
in
a
te
s
th
e
ot
he
r
,
p
ot
e
nt
ia
ll
y
bi
a
s
in
g
th
e
m
ode
l'
s
pr
e
di
c
ti
ons
.
T
o
m
it
ig
a
te
th
is
,
ove
r
s
a
m
pl
in
g
vi
a
th
e
S
M
O
T
E
is
e
m
pl
oye
d.
A
dd
it
io
na
ll
y,
di
f
f
e
r
e
nt
s
upe
r
vi
s
e
d
c
la
s
s
if
ie
r
s
,
s
uc
h
a
s
S
V
M
,
r
a
ndom
f
or
e
s
ts
,
n
a
iv
e
B
a
ye
s
,
K
-
ne
a
r
e
s
t
ne
ig
hbor
s
(
K
N
N
s
)
,
L
ig
ht
G
B
M
,
a
nd
A
da
B
oos
t,
a
r
e
ut
il
iz
e
d,
a
s
e
a
c
h
of
f
e
r
in
g
di
s
ti
nc
t
le
a
r
ni
ng
pa
tt
e
r
ns
w
i
th
th
e
ir
ow
n
a
dva
nt
a
ge
s
.
H
ype
r
pa
r
a
m
e
te
r
opt
im
iz
a
ti
on
us
e
d
th
e
gr
id
s
e
a
r
c
h
te
c
hni
que
to
r
e
f
in
e
a
nd
c
a
li
br
a
te
th
e
m
ode
ls
.
T
hi
s
pr
oc
e
s
s
in
vol
ve
d
s
y
s
te
m
a
ti
c
a
ll
y
e
xpl
or
in
g
a
r
a
nge
of
hype
r
pa
r
a
m
e
te
r
va
lu
e
s
t
o
f
in
d t
he
c
om
bi
na
ti
on
t
ha
t
r
e
s
ul
te
d i
n t
he
be
s
t
m
ode
l
pe
r
f
or
m
a
nc
e
. F
or
e
a
c
h
m
ode
l,
hype
r
pa
r
a
m
e
te
r
s
,
s
uc
h
a
s
C
(
r
e
gul
a
r
iz
a
ti
on
pa
r
a
m
e
te
r
)
a
nd
ke
r
ne
l
ty
pe
f
or
S
V
M
,
num
be
r
o
f
tr
e
e
s
a
nd
m
a
xi
m
um
de
pt
h
f
or
r
a
ndom
f
or
e
s
ts
,
a
nd
l
e
a
r
ni
ng
r
a
te
a
nd
b
oos
ti
ng
r
ounds
f
or
L
ig
ht
G
B
M
,
w
e
r
e
tu
ne
d
to
a
c
hi
e
ve
opt
im
a
l
pe
r
f
or
m
a
nc
e
.
H
ype
r
pa
r
a
m
e
te
r
s
s
uc
h
a
s
a
lp
ha
f
or
n
a
iv
e
B
a
ye
s
a
nd
K
f
or
K
N
N
s
w
e
r
e
a
ls
o
opt
im
iz
e
d.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
P
r
oac
ti
v
e
c
e
r
v
ic
al
c
anc
e
r
r
i
s
k
as
s
e
s
s
m
e
nt
u
s
in
g data
-
dr
iv
e
n an
al
y
ti
c
s
(
Sr
e
e
la
th
a)
4307
3.
R
E
S
U
L
T
S
A
N
D
D
I
S
C
U
S
S
I
O
N
T
he
pr
opos
e
d
s
y
s
te
m
w
a
s
de
s
ig
ne
d
a
nd
d
e
ve
lo
pe
d
u
s
in
g
P
yt
hon
s
c
r
ip
ti
ng
in
th
e
A
na
c
ond
a
e
nvi
r
onm
e
nt
.
T
hi
s
s
e
c
ti
on
pr
e
s
e
nt
s
th
e
pe
r
f
or
m
a
nc
e
of
th
e
pr
opos
e
d
s
ys
te
m
w
it
h
di
f
f
e
r
e
nt
c
la
s
s
if
ic
a
ti
on
m
ode
ls
.
T
he
e
f
f
e
c
ti
ve
ne
s
s
of
th
e
pr
opos
e
d
c
om
put
a
ti
ona
l
f
r
a
m
e
w
or
k
is
a
ls
o
s
how
n
th
r
ough
a
c
om
pa
r
a
ti
ve
a
na
ly
s
is
w
it
h
s
im
il
a
r
e
xi
s
ti
ng
r
e
s
e
a
r
c
h
w
or
k.
T
h
e
pe
r
f
or
m
a
nc
e
of
th
e
pr
opos
e
d
s
ys
te
m
is
m
e
a
s
ur
e
d
in
te
r
m
s
of
a
c
c
ur
a
c
y, pr
e
c
is
io
n, r
e
c
a
ll
, F
1
-
s
c
or
e
, a
nd t
h
e
r
e
c
e
iv
e
r
ope
r
a
ti
ng c
ha
r
a
c
te
r
is
ti
c
(
R
O
C
)
c
ur
ve
.
Ac
c
ur
a
c
y
=
(
T
r
ue
p
o
s
it
iv
e
s
+
T
r
ue
n
e
g
a
t
iv
e
s
)
/
(
T
o
t
a
l
n
um
b
e
r
of
p
r
e
d
ic
t
io
n
s
)
(
4)
A
c
c
ur
a
c
y
is
th
e
pr
opor
ti
on
of
a
ll
c
or
r
e
c
t
pr
e
di
c
ti
ons
,
i.
e
.
th
e
ove
r
a
ll
c
or
r
e
c
tn
e
s
s
of
th
e
m
ode
l.
H
ow
e
ve
r
,
it
s
de
pe
nda
n
c
e
on
c
or
r
e
c
t
pr
e
di
c
ti
ons
f
or
a
ll
c
la
s
s
e
s
c
a
n
be
m
is
l
e
a
di
ng
in
c
a
s
e
w
he
n
da
ta
s
e
t
is
s
e
ve
r
e
ly
im
ba
la
nc
e
d
(
i.
e
.,
th
e
r
e
a
r
e
m
or
e
ne
ga
ti
ve
c
a
s
e
s
th
a
n
pos
it
iv
e
c
a
s
e
s
)
.
I
n
th
is
c
a
s
e
,
th
e
m
ode
l
onl
y
ne
e
ds
t
o pr
e
di
c
t
a
ll
c
a
s
e
s
a
s
ne
g
a
ti
ve
t
o a
c
hi
e
ve
hi
gh a
c
c
ur
a
c
y.
=
/
+
(
5)
P
r
e
c
is
io
n
is
ba
s
i
c
a
ll
y
a
po
s
it
iv
e
pr
e
di
c
ti
ve
va
lu
e
th
a
t
m
e
a
s
ur
e
s
t
he
pr
opor
ti
on
of
pos
it
iv
e
pr
e
di
c
ti
ons
th
a
t
a
r
e
a
c
tu
a
ll
y
c
or
r
e
c
t.
P
r
e
c
is
io
n
is
a
good
m
e
a
s
ur
e
of
a
m
o
de
l'
s
a
bi
li
ty
to
id
e
nt
i
f
y
pos
it
iv
e
c
a
s
e
s
.
A
hi
gh
a
c
c
ur
a
c
y s
c
or
e
m
e
a
ns
t
he
m
od
e
l
doe
s
not
pr
oduc
e
m
a
ny f
a
l
s
e
p
os
it
iv
e
s
.
=
/
(
+
)
(
6)
R
e
c
a
ll
is
th
e
tr
ue
po
s
it
iv
e
r
a
te
,
w
hi
c
h
m
e
a
s
ur
e
s
th
e
pr
opor
ti
on
of
a
c
tu
a
l
pos
it
iv
e
s
th
a
t
a
r
e
c
or
r
e
c
tl
y
pr
e
di
c
te
d.
P
e
r
f
or
m
a
nc
e
m
e
tr
ic
s
a
r
e
a
good
m
e
a
s
ur
e
of
a
m
ode
l'
s
a
bi
li
ty
to
c
a
pt
ur
e
a
ll
pos
it
iv
e
c
a
s
e
s
.
A
hi
gh
r
e
c
a
ll
in
di
c
a
te
s
th
a
t
th
e
m
ode
l
is
not
m
is
s
in
g
m
a
ny
tr
ue
pos
it
iv
e
r
e
s
ul
ts
.
T
he
F
1
s
c
or
e
r
e
f
e
r
s
to
th
e
ha
r
m
oni
c
m
e
a
n
of
pr
e
c
is
io
n
a
nd
r
e
c
a
ll
m
e
tr
ic
s
,
pr
ovi
di
ng
a
ba
la
nc
e
b
e
twe
e
n
th
e
two
w
he
n
th
e
c
la
s
s
di
s
tr
ib
ut
io
n
is
une
ve
n.
1
=
2
×
(
×
)
/
(
+
)
(
7)
T
he
e
va
lu
a
ti
on
of
th
e
pr
opos
e
d
pr
e
di
c
ti
on
m
ode
l
a
ls
o
c
ons
id
e
r
s
th
e
R
O
C
c
ur
ve
,
w
hi
c
h
s
how
s
th
e
pe
r
f
or
m
a
nc
e
of
th
e
c
la
s
s
if
ic
a
ti
on
m
ode
l
a
t
a
ll
pos
s
ib
le
th
r
e
s
ho
ld
s
.
I
t
is
c
r
e
a
te
d
by
pl
ot
ti
ng
th
e
tr
ue
pos
it
iv
e
r
a
te
(
T
P
R
)
ve
r
s
us
th
e
f
a
ls
e
pos
it
iv
e
r
a
te
(
F
P
R
)
a
t
di
f
f
e
r
e
nt
th
r
e
s
hol
ds
.
T
P
R
is
de
f
in
e
d
a
s
th
e
pr
opor
ti
on
of
a
c
tu
a
l
pos
it
iv
e
e
xa
m
pl
e
s
th
a
t
a
r
e
c
or
r
e
c
tl
y
pr
e
di
c
te
d,
a
nd
F
P
R
is
th
e
pr
opor
ti
on
of
a
c
tu
a
l
ne
ga
ti
ve
e
xa
m
pl
e
s
th
a
t
a
r
e
i
nc
or
r
e
c
tl
y pr
e
di
c
te
d. T
he
hi
ghe
r
t
he
a
r
e
a
und
e
r
t
he
c
ur
ve
(
AUC
)
, t
he
be
tt
e
r
t
he
m
ode
l
pe
r
f
or
m
a
nc
e
.
3.1. Nu
m
e
r
ic
al
ou
t
c
om
e
T
he
out
c
om
e
s
ta
ti
s
ti
c
s
pr
e
s
e
nt
e
d
in
T
a
bl
e
2
s
how
th
e
pe
r
f
or
m
a
nc
e
of
th
e
d
if
f
e
r
e
nt
popu
la
r
c
la
s
s
if
ic
a
ti
on
m
ode
ls
a
c
r
os
s
s
e
ve
r
a
l
c
r
it
ic
a
l
m
e
tr
ic
s
,
na
m
e
ly
a
c
c
ur
a
c
y,
pr
e
c
is
io
n,
r
e
c
a
ll
,
f
1
-
s
c
or
e
,
a
nd
R
O
C
s
c
or
e
.
B
a
s
e
d
on
th
e
c
r
it
ic
a
l
a
na
ly
s
is
of
th
e
a
c
c
ur
a
c
y
s
c
or
e
,
i
t
ha
s
be
e
n
id
e
nt
if
ie
d
th
a
t
th
e
r
a
ndom
f
or
e
s
t
c
la
s
s
if
ie
r
a
c
hi
e
ve
d
th
e
hi
ghe
s
t
a
c
c
ur
a
c
y
s
c
or
e
a
t
98
.
06
%
.
F
ol
lo
w
e
d
c
lo
s
e
ly
by
bot
h
th
e
S
V
M
a
nd
L
ig
ht
G
B
M
c
la
s
s
if
ic
a
ti
on
m
ode
ls
,
e
a
c
h
a
c
hi
e
vi
ng
a
n a
c
c
ur
a
c
y of
97
.
67
%
.
T
a
bl
e
2. A
na
ly
s
is
of
t
he
nume
r
ic
a
l
out
c
om
e
s
ta
ti
s
ti
c
s
f
or
c
la
s
s
if
ic
a
ti
on mode
l
A
c
c
ur
a
c
y
P
r
e
c
i
s
i
on
R
e
c
a
l
l
F1
-
S
c
or
e
R
O
C
S
c
or
e
S
V
M
0.976744
0.973173
0.976744
0.972961
0.96
R
a
ndom
F
or
e
s
t
0.980620
0.979559
0.980620
0.979994
0.95
N
a
i
ve
B
a
ye
s
0.968992
0.972774
0.968992
0.970653
0.85
KNN
0.957364
0.973420
0.957364
0.963550
0.96
L
i
ght
G
B
M
0.976744
0.976744
0.976744
0.976744
0.97
A
da
B
oos
t
0.965116
0.967207
0.965116
0.966107
0.80
It
is
t
o be
not
e
d t
ha
t
th
e
hi
gh
er
a
c
c
ur
a
c
y i
ndi
c
a
te
s
t
he
ov
e
r
a
ll
c
o
r
r
e
c
tn
e
s
s
of
t
he
c
la
s
s
if
ic
a
ti
on
m
ode
l
s
a
nd
it
s
s
ui
ta
bi
li
ty
f
or
a
gi
ve
n
da
t
a
s
e
t
.
H
ow
e
ve
r
,
w
he
n
th
e
d
a
ta
s
e
t
s
uf
f
e
r
s
f
r
om
c
la
s
s
im
ba
la
n
c
e
pr
obl
e
m
,
pe
r
f
or
m
a
nc
e
a
na
ly
s
is
c
ons
id
e
r
in
g
ot
he
r
m
e
tr
ic
s
s
uc
h
a
s
pr
e
c
i
s
io
n
a
nd
r
e
c
a
ll
be
c
om
e
s
a
c
r
it
ic
a
l
is
s
ue
.
T
he
p
r
e
c
is
io
n
c
a
pt
ur
e
s
th
e
m
ode
l'
s
c
a
pa
bi
li
ty
to
c
or
r
e
c
tl
y
c
la
s
s
if
y
pos
it
iv
e
in
s
ta
nc
e
s
out
of
th
os
e
pr
e
di
c
te
d
a
s
pos
it
iv
e
.
B
a
s
e
d
on
th
e
num
e
r
ic
a
l
out
c
om
e
,
it
c
a
n
be
s
e
e
n
th
a
t
t
he
r
a
ndom
f
o
r
e
s
t
m
ode
l
a
ga
in
out
pe
r
f
o
r
m
e
d
w
it
h
a
pr
e
c
is
io
n
of
97
.
95
%
,
th
ough
th
e
di
f
f
e
r
e
nc
e
w
it
h
ot
he
r
to
p
-
pe
r
f
or
m
in
g
m
ode
ls
li
ke
K
N
N
(
97
.
34
%
)
a
nd
S
V
M
(
97
.
31
%
)
is
r
e
la
ti
ve
ly
m
in
im
a
l.
H
e
r
e
h
ig
h
pr
e
c
is
io
n
s
c
or
e
im
pl
ie
s
a
lo
w
e
r
f
a
ls
e
-
pos
it
iv
e
r
a
te
,
e
ns
ur
in
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
. 13, No. 4, D
e
c
e
m
be
r
2024:
4301
-
4311
4308
th
a
t
th
e
pr
e
di
c
ti
ons
m
a
de
a
r
e
r
e
li
a
bl
e
.
I
n
te
r
m
s
of
r
e
c
a
ll
r
a
t
e
,
w
hi
c
h
r
e
f
le
c
ts
th
e
a
bi
li
ty
of
th
e
c
la
s
s
if
ic
a
ti
on
m
ode
ls
in
id
e
nt
if
yi
ng
a
ll
pos
s
ib
le
pos
it
iv
e
in
s
t
a
nc
e
s
,
r
a
ndom
f
or
e
s
t
c
la
s
s
if
ie
r
m
a
in
ta
in
s
it
s
s
upe
r
io
r
it
y
w
it
h
a
s
c
or
e
of
98
.
06
%
.
H
ow
e
v
e
r
,
bot
h
S
V
M
a
nd
L
ig
ht
G
B
M
a
c
hi
e
v
e
d
s
im
il
a
r
pe
r
f
or
m
a
nc
e
a
s
w
e
ll
, e
a
c
h
pr
e
s
e
nt
in
g
a
r
e
c
a
ll
of
97
.
67
%
.
T
he
F
1
-
S
c
or
e
, r
e
pr
e
s
e
nt
in
g t
he
ha
r
m
oni
c
m
e
a
n of
pr
e
c
is
io
n a
nd r
e
c
a
ll
, i
s
s
ig
ni
f
ic
a
nt
w
he
n
de
a
li
ng
w
it
h
da
ta
s
e
ts
w
it
h
im
ba
la
nc
e
d
c
la
s
s
di
s
tr
ib
ut
io
ns
.
H
e
r
e
,
r
a
ndom
f
or
e
s
t
m
a
in
ta
in
s
it
s
dom
in
a
nc
e
w
it
h
a
n
F
1
s
c
or
e
of
97.99%
.
F
ur
th
e
r
m
or
e
,
L
ig
ht
G
B
M
e
xhi
bi
ts
un
iq
ue
c
ha
r
a
c
te
r
is
ti
c
s
w
it
h
th
e
s
a
m
e
F
1
s
c
or
e
,
pr
e
c
is
io
n
a
nd
r
e
c
a
ll
of
97.67%
,
in
di
c
a
ti
ng
ba
l
a
nc
e
d
p
e
r
f
or
m
a
nc
e
in
te
r
m
s
of
f
a
ls
e
po
s
it
iv
e
s
a
nd
f
a
ls
e
ne
ga
ti
ve
s
.
F
in
a
ll
y,
th
e
R
O
C
s
c
or
e
r
e
pr
e
s
e
nt
s
th
e
m
ode
l'
s
a
bi
li
ty
to
di
s
ti
ngui
s
h
c
la
s
s
e
s
f
or
a
ll
th
r
e
s
hol
ds
;
th
e
c
la
s
s
if
ie
r
L
ig
ht
G
B
M
a
c
hi
e
ve
s
a
n
R
O
C
s
c
or
e
of
97%
, out
pe
r
f
or
m
in
g a
ll
ot
he
r
c
la
s
s
if
ie
r
s
.
3.2. Vis
u
al
ou
t
c
om
e
F
ig
ur
e
4
pr
ovi
de
s
a
de
ta
il
e
d
c
om
pa
r
is
on
of
th
e
pe
r
f
or
m
a
nc
e
m
e
tr
ic
s
f
or
th
r
e
e
of
th
e
to
p
-
pe
r
f
or
m
in
g
c
la
s
s
if
ic
a
ti
on
m
ode
l
s
:
S
V
M
,
r
a
ndom
f
or
e
s
t
,
a
nd
L
ig
ht
G
B
M
.
F
or
e
a
c
h
m
ode
l,
th
e
pe
r
f
or
m
a
nc
e
is
vi
s
ua
li
s
e
d
th
r
ough
it
s
c
onf
us
io
n
m
a
tr
ix
,
w
hi
c
h
il
lu
s
tr
a
te
s
th
e
a
c
c
ur
a
c
y
of
pr
e
di
c
ti
ons
in
te
r
m
s
of
tr
ue
pos
it
iv
e
s
a
nd
tr
ue
ne
ga
ti
ve
s
,
a
nd
th
e
R
O
C
c
ur
ve
,
w
hi
c
h
e
va
lu
a
te
s
th
e
tr
a
de
-
of
f
be
twe
e
n
s
e
n
s
it
iv
it
y
a
nd
s
pe
c
if
ic
it
y.
A
na
ly
s
is
of
F
ig
ur
e
s
4(
a
)
a
nd
4(
b)
r
e
ve
a
ls
a
la
r
ge
num
be
r
of
tr
ue
ne
ga
ti
ve
s
(
248)
a
nd
a
ve
r
y
s
m
a
ll
num
be
r
of
f
a
ls
e
pos
it
iv
e
s
(
2)
.
I
n
te
r
m
s
of
pos
it
iv
e
c
a
te
gor
ie
s
,
it
c
or
r
e
c
tl
y
id
e
nt
if
ie
d
4
in
s
ta
n
c
e
s
,
but
a
l
s
o
ha
d
4
f
a
ls
e
ne
ga
ti
ve
s
.
T
hi
s
i
ndi
c
a
te
s
a
s
li
ght
de
c
r
e
a
s
e
i
n t
he
s
e
ns
it
iv
it
y of
t
he
S
V
M
m
ode
l.
T
he
S
V
M
m
ode
l
a
c
hi
e
ve
d a
n R
O
C
s
c
or
e
of
96%
a
nd
s
uc
c
e
s
s
f
ul
ly
di
s
ti
ngui
s
he
d
be
twe
e
n
pos
it
iv
e
a
nd
ne
ga
ti
ve
c
la
s
s
e
s
a
t
di
f
f
e
r
e
nt
th
r
e
s
hol
ds
.
F
ig
ur
e
s
4(
c
)
a
nd
4(
d)
a
na
ly
s
is
s
how
s
th
a
t
th
e
r
a
ndom
f
or
e
s
t
m
ode
l
id
e
nt
if
ie
d
248
tr
ue
ne
ga
ti
ve
s
a
nd
onl
y
2
f
a
ls
e
pos
it
iv
e
s
.
F
or
th
e
pos
it
iv
e
c
la
s
s
,
it
a
c
c
ur
a
t
e
ly
pr
e
di
c
te
d
5
in
s
ta
nc
e
s
a
nd
onl
y
3
f
a
ls
e
ne
ga
ti
ve
s
.
T
hi
s
c
onf
ig
ur
a
ti
on
s
li
ght
ly
im
pr
ove
s
s
e
ns
it
iv
it
y
c
om
pa
r
e
d
to
S
V
M
.
F
ur
th
e
r
m
or
e
,
th
e
R
O
C
s
c
or
e
of
th
e
r
a
ndom
f
or
e
s
t
m
ode
l
is
95%
,
w
hi
c
h
is
a
lm
o
s
t
c
om
pa
r
a
bl
e
to
S
V
M
.
M
e
a
nw
hi
le
,
a
c
a
r
e
f
ul
a
na
ly
s
is
of
F
ig
ur
e
s
4(
e
)
a
nd
4(
f
)
s
how
s
th
a
t
L
ig
ht
G
B
M
a
ls
o
h
a
s
248
tr
ue
ne
ga
ti
ve
s
a
nd
2
f
a
ls
e
pos
it
iv
e
s
,
s
im
il
a
r
to
th
e
r
a
ndom
f
or
e
s
t
c
la
s
s
if
ie
r
.
I
t
ha
s
th
e
s
a
m
e
a
c
c
ur
a
te
pr
e
di
c
ti
on
s
,
i.
e
.
5
pos
it
iv
e
c
la
s
s
pr
e
di
c
ti
ons
a
nd
3
f
a
l
s
e
ne
ga
ti
ve
c
la
s
s
pr
e
di
c
ti
on
s
.
L
ig
ht
G
B
M
s
ta
nds
out
w
it
h
a
R
O
C
s
c
or
e
of
97%
,
s
li
ght
ly
hi
ghe
r
th
a
n
S
V
M
a
nd
r
a
ndom f
or
e
s
t
. T
hi
s
e
xc
e
ll
e
nt
R
O
C
s
c
or
e
i
ndi
c
a
t
e
s
t
ha
t
L
ig
ht
G
B
M
ha
s
a
be
tt
e
r
a
bi
li
ty
t
o di
s
ti
ngui
s
h c
la
s
s
e
s
a
t
di
f
f
e
r
e
nt
th
r
e
s
hol
ds
.
A
hi
gh
R
O
C
s
c
or
e
in
di
c
a
te
s
a
m
ode
l'
s
s
tr
e
ngt
h
in
m
a
in
ta
in
in
g
a
hi
gh
tr
ue
pos
it
iv
e
r
a
te
.
A
c
c
or
di
ng
to
th
e
r
e
s
ul
t
s
ta
ti
s
ti
c
s
,
L
ig
ht
G
B
M
m
ode
ls
out
pe
r
f
o
r
m
ot
he
r
c
la
s
s
if
ic
a
ti
on
m
ode
ls
on
a
ll
m
e
tr
ic
s
.
I
t
a
c
hi
e
ve
s
t
he
hi
gh
e
s
t
a
c
c
ur
a
c
y, pr
e
c
is
io
n, r
e
c
a
ll
, F
1 s
c
or
e
, a
nd R
O
C
s
c
or
e
.
(
a
)
(
b)
(
c
)
(
d)
(
e
)
(f)
F
ig
ur
e
4. I
ll
us
tr
a
ti
on of
vi
s
ua
l
out
c
om
e
of
t
op c
la
s
s
if
ie
r
s
:
(
a
)
S
V
M
c
onf
us
io
n m
a
tr
ix
,
(
b)
S
V
M
R
O
C
,
(
c
)
r
a
ndom f
or
e
s
t
c
onf
us
io
n m
a
tr
ix
,
(
d)
r
a
ndom f
or
e
s
t
R
O
C
,
(
e
)
L
ig
ht
G
B
M
c
onf
us
io
n m
a
tr
ix
, a
nd
(
f
)
L
ig
ht
G
B
M
R
O
C
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
P
r
oac
ti
v
e
c
e
r
v
ic
al
c
anc
e
r
r
i
s
k
as
s
e
s
s
m
e
nt
u
s
in
g data
-
dr
iv
e
n an
al
y
ti
c
s
(
Sr
e
e
la
th
a)
4309
3.3. Us
e
c
as
e
s
c
e
n
ar
io
B
a
s
e
d
on
th
e
c
om
pr
e
he
n
s
iv
e
a
na
ly
s
is
of
th
e
r
e
s
ul
t
s
,
th
e
L
ig
ht
G
B
M
m
ode
l
e
m
e
r
ge
d
a
s
th
e
be
s
t
m
ode
l
due
t
o i
ts
c
ons
is
te
n
c
y, a
c
c
ur
a
c
y, pr
e
c
is
io
n, r
e
c
a
ll
a
nd hig
h R
O
C
s
c
or
e
.
T
he
r
e
f
or
e
, f
or
pa
ti
e
nt
s
a
t
r
is
k of
c
e
r
vi
c
a
l
c
a
nc
e
r
,
th
e
pr
oba
bi
li
ty
s
c
or
e
f
r
o
m
th
is
m
ode
l
c
a
n
be
u
s
e
d
to
c
a
lc
ul
a
te
th
e
r
is
k
pe
r
c
e
nt
a
ge
,
e
xpr
e
s
s
e
d
a
s
in
(
8)
.
(
=
1
)
=
1
1
+
e
−
∑
r
a
w
s
c
o
r
e
(
8)
W
he
r
e
(
=
1
)
is
t
he
pr
oba
bi
li
ty
t
ha
t
th
e
i
ns
ta
nc
e
be
lo
ngs
t
o c
la
s
s
1 (
e
.g., "
a
t
r
is
k"
)
,
a
nd t
he
∑
r
a
w
s
c
or
e
i
s
th
e
s
um
of
th
e
out
pu
ts
of
a
ll
th
e
tr
e
e
s
f
or
th
a
t
in
s
ta
nc
e
.
T
he
(
8
)
is
th
e
lo
gi
s
ti
c
f
unc
ti
on,
w
hi
c
h
m
a
ps
th
e
r
a
w
s
c
or
e
be
twe
e
n 0 a
nd 1, yie
ld
in
g a
pr
oba
bi
li
ty
va
lu
e
.
F
or
e
xa
m
pl
e
, i
f
(
=
1
)
is
0.8, it
m
e
a
ns
th
a
t
th
e
r
e
i
s
a
n
80%
pr
oba
bi
li
ty
,
a
c
c
or
di
ng
to
th
e
m
ode
l,
th
a
t
th
e
p
a
ti
e
nt
i
s
a
t
r
is
k.
I
n
a
pr
a
c
ti
c
a
l
s
e
tt
in
g,
c
ons
id
e
r
a
26
-
ye
a
r
-
ol
d
w
om
a
n
w
ho
ha
s
ha
d
8
s
e
xua
l
pa
r
tn
e
r
s
,
be
e
n
pr
e
g
na
nt
twi
c
e
,
s
m
oke
s
a
n
a
ve
r
a
g
e
of
25
pa
c
ks
of
c
ig
a
r
e
tt
e
s
pe
r
ye
a
r
,
a
nd
ha
s
ha
d
3
S
T
D
di
a
gnos
e
s
.
W
it
hout
a
n
H
P
V
di
a
gnos
is
,
th
e
m
ode
l
put
s
he
r
r
is
k
of
c
e
r
vi
c
a
l
c
a
nc
e
r
a
t
18%
. C
onve
r
s
e
ly
,
a
n H
P
V
di
a
gnos
is
in
c
r
e
a
s
e
s
he
r
r
is
k
to
a
n
e
s
ti
m
a
te
d
90
pe
r
c
e
nt
.
A
not
he
r
e
xa
m
pl
e
is
a
40
-
ye
a
r
-
ol
d
w
om
a
n
w
it
h
3
s
e
xu
a
l
pa
r
tn
e
r
s
w
ho
s
ta
r
te
d
ha
vi
ng
s
e
x
w
he
n
s
h
e
w
a
s
26
ye
a
r
s
ol
d.
B
a
s
e
d
on
th
e
s
e
a
tt
r
ib
ut
e
s
a
lo
ne
,
he
r
r
is
k
is
onl
y
0.09%
;
how
e
ve
r
,
a
n
H
P
V
di
a
gnos
is
in
c
r
e
a
s
e
s
th
is
r
is
k
to
99%
.
T
he
s
e
e
x
a
m
pl
e
s
hi
ghl
ig
ht
th
e
s
ig
ni
f
ic
a
nt
im
pa
c
t
of
H
P
V
s
ta
tu
s
on
c
e
r
vi
c
a
l
c
a
nc
e
r
r
is
k.
F
a
c
to
r
s
th
a
t
in
c
r
e
a
s
e
th
e
r
is
k
of
H
P
V
in
c
lu
de
ha
vi
ng
m
ul
ti
pl
e
s
e
xua
l
pa
r
tn
e
r
s
,
ha
vi
ng
s
e
x
to
o
e
a
r
ly
,
c
onc
ur
r
e
nt
s
e
xu
a
ll
y
tr
a
ns
m
it
te
d
in
f
e
c
ti
ons
(
s
uc
h
a
s
c
hl
a
m
ydi
a
,
gonor
r
he
a
,
s
yphi
li
s
,
H
I
V
/AI
D
S
)
,
a
nd
a
c
om
pr
om
is
e
d
im
m
une
s
ys
te
m
.
A
ddi
ti
ona
ll
y,
a
not
h
e
r
r
e
a
s
on
i
s
th
a
t
pr
e
-
e
xi
s
ti
ng
m
e
di
c
a
l
c
ondi
ti
ons
,
ha
r
m
f
ul
li
f
e
s
ty
le
h
a
bi
ts
,
or
s
m
oki
ng ma
y l
e
a
d t
o a
w
e
a
k
e
ne
d i
m
m
une
s
y
s
te
m
, w
hi
c
h c
a
n
a
l
s
o l
e
a
d t
o s
qua
m
ou
s
c
e
ll
c
e
r
vi
c
a
l
c
a
nc
e
r
.
3.4. Com
p
ar
at
iv
e
an
al
ys
is
F
ig
ur
e
5
pr
ovi
de
s
a
c
om
pa
r
a
ti
ve
a
na
ly
s
is
to
de
m
on
s
tr
a
te
th
e
e
f
f
e
c
ti
ve
ne
s
s
of
th
e
pr
opos
e
d
c
e
r
vi
c
a
l
c
a
nc
e
r
r
is
k
pr
e
di
c
ti
on
m
od
e
l
w
it
h
e
xi
s
ti
ng
s
im
il
a
r
r
e
s
e
a
r
c
h
w
o
r
ks
.
T
he
r
is
k
pr
e
di
c
ti
on
m
ode
l
G
a
us
s
ia
n
n
a
iv
e
B
a
ye
s
u
s
e
d
in
th
e
s
tu
dy
[
27]
ha
s
a
n
a
c
c
ur
a
c
y
of
81%
,
pr
e
c
is
io
n
is
86%
a
nd
a
c
hi
e
ve
d
th
e
r
e
c
a
ll
r
a
te
w
it
h
100%
.
T
he
e
ns
e
m
bl
e
m
ode
l
ut
il
iz
e
d
in
[
18]
e
xhi
bi
te
d
s
up
e
r
io
r
a
c
c
ur
a
c
y
a
t
95%
,
boa
s
ti
ng
a
pe
r
f
e
c
t
pr
e
c
i
s
io
n
s
c
or
e
.
H
ow
e
v
e
r
,
it
s
r
e
c
a
ll
is
c
om
pa
r
a
ti
ve
ly
lo
w
e
r
a
t
67%
.
S
um
a
n
a
nd
H
ood
a
[
28]
a
dopt
e
d
d
e
c
is
io
n
tr
e
e
,
yi
e
ld
in
g a
c
c
ur
a
c
y of
93%
,
w
hi
le
t
he
p
r
e
c
is
io
n a
nd r
e
c
a
ll
w
e
r
e
8
9%
a
nd 96%
.
L
u
e
t
al
.
[
29]
e
m
pl
oye
d
lo
gi
s
ti
c
r
e
gr
e
s
s
io
n
a
nd a
tt
a
in
e
d a
n a
c
c
ur
a
c
y of
82%
, but
i
ts
pr
e
c
is
io
n a
n
d r
e
c
a
ll
s
c
or
e
s
, a
t
45%
a
nd 21%
, r
e
s
pe
c
ti
ve
ly
,
a
r
e
qui
te
lo
w
e
r
t
ha
n ot
he
r
m
ode
ls
.
F
ig
ur
e
5. S
how
s
c
om
pa
r
a
ti
ve
a
na
ly
s
is
i
n t
e
r
m
s
of
a
c
c
ur
a
c
y, pr
e
c
is
io
n, a
nd r
e
c
a
ll
r
a
te
A
kt
e
r
e
t
al
.
[
30]
us
in
g
de
c
is
io
n
tr
e
e
,
s
how
e
d
th
e
s
a
m
e
a
c
c
ur
a
c
y
of
93%
,
but
va
r
ie
d
pa
r
ti
c
ul
a
r
ly
in
r
e
c
a
ll
,
w
it
h
a
r
is
e
to
100%
.
T
he
pr
opos
e
d
r
is
k
pr
e
di
c
ti
on
f
r
a
m
e
w
or
k
(
P
r
op)
ha
r
ne
s
s
e
s
th
e
pot
e
nt
ia
l
of
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
, V
ol
. 13, No. 4, D
e
c
e
m
be
r
2024:
4301
-
4311
4310
L
ig
ht
G
B
M
th
a
t
out
pe
r
f
or
m
s
th
e
m
ode
ls
m
e
nt
io
ne
d
a
bove
w
it
h
a
c
c
ur
a
c
y,
pr
e
c
is
io
n,
a
nd
r
e
c
a
ll
,
a
ll
r
e
a
c
hi
ng
98%
.
T
hi
s
hi
ghl
ig
ht
s
th
e
e
f
f
ic
a
c
y
of
th
e
L
ig
ht
G
B
M
m
ode
l
a
n
d
s
how
s
it
s
b
a
la
nc
e
d
a
ppr
oa
c
h
a
nd
c
ons
is
te
nt
pe
r
f
or
m
a
nc
e
a
c
r
os
s
a
ll
m
e
tr
ic
s
,
s
ugge
s
ti
ng
it
s
s
ui
ta
bi
li
ty
a
s
a
r
e
li
a
bl
e
pr
e
di
c
it
ve
m
od
e
l
in
th
e
pr
op
s
oe
d
f
r
a
m
e
w
or
k f
or
c
e
r
vi
c
a
l
c
a
nc
e
r
r
is
k pr
e
di
c
ti
on.
4.
C
O
N
C
L
U
S
I
O
N
T
hi
s
s
tu
dy
ha
s
pr
e
s
e
nt
ed
a
r
obus
t
da
ta
-
dr
iv
e
n
a
ppr
oa
c
h
de
s
i
gne
d
to
pr
oa
c
ti
ve
ly
id
e
nt
if
y
c
e
r
vi
c
a
l
c
a
nc
e
r
r
is
k,
e
na
bl
in
g
he
a
lt
hc
a
r
e
pr
of
e
s
s
io
na
ls
to
m
a
ke
in
f
or
m
e
d
de
c
is
io
ns
a
bout
p
a
ti
e
nt
c
a
r
e
.
B
y
a
ppl
yi
ng
c
om
put
a
ti
ona
l
in
te
ll
ig
e
nc
e
a
lg
or
it
hm
s
to
c
li
ni
c
a
l
a
nd
pa
ti
e
nt
l
if
e
s
ty
le
da
ta
,
th
is
r
e
s
e
a
r
c
h
w
or
k
a
tt
e
m
pt
e
d
t
o
im
pr
ove
e
a
r
ly
de
te
c
ti
on
o
f
c
e
r
vi
c
a
l
c
a
nc
e
r
a
nd
in
c
r
e
a
s
e
di
a
gnos
ti
c
a
c
c
ur
a
c
y
a
nd
e
f
f
ic
ie
nc
y.
T
hi
s
s
tu
dy
id
e
nt
if
ie
d
ke
y
f
e
a
tu
r
e
s
a
nd
c
or
r
e
la
ti
ons
f
or
bui
ld
in
g
a
n
e
f
f
e
c
ti
ve
c
a
nc
e
r
pr
e
di
c
ti
on
m
ode
l
us
in
g
th
e
L
ig
ht
G
B
M
c
la
s
s
if
ie
r
th
r
ough
e
f
f
ic
ie
nt
a
na
ly
s
is
a
nd
pr
e
pr
o
c
e
s
s
in
g
of
a
c
e
r
vi
c
a
l
c
a
nc
e
r
da
ta
s
e
t.
T
he
e
f
f
e
c
ti
ve
ne
s
s
of
t
he
pr
opos
e
d pr
e
di
c
ti
on mode
l
is
f
ul
ly
ve
r
if
ie
d
w
it
h a
n a
c
c
ur
a
c
y of
up t
o 98%
. B
y f
a
c
il
it
a
ti
ng
e
a
r
ly
a
nd
a
c
c
ur
a
te
de
te
c
ti
on
of
c
e
r
vi
c
a
l
c
a
nc
e
r
,
th
e
pr
opos
e
d
r
is
k
pr
e
di
c
ti
on
m
ode
l
e
na
bl
e
s
he
a
lt
hc
a
r
e
pr
of
e
s
s
io
na
ls
to
pr
io
r
it
iz
e
a
nd
f
a
s
t
-
tr
a
c
k
e
xa
m
in
a
ti
on
a
nd
t
r
e
a
tm
e
nt
of
hi
gh
-
r
is
k
pa
ti
e
nt
s
.
T
hi
s
pr
oa
c
ti
ve
a
ppr
oa
c
h
to
r
is
k
a
na
ly
s
is
a
nd
p
a
ti
e
nt
c
a
r
e
h
a
s
th
e
pot
e
nt
ia
l
to
s
ig
ni
f
ic
a
nt
ly
im
pr
ove
c
li
ni
c
a
l
out
c
om
e
s
a
nd
in
c
r
e
a
s
e
t
he
l
ik
e
li
hood of
s
uc
c
e
s
s
f
ul
t
r
e
a
tm
e
nt
a
nd r
e
c
ov
e
r
y. F
ur
th
e
r
m
or
e
, i
nt
e
gr
a
ti
ng t
he
pr
opos
e
d pr
e
di
c
ti
ve
m
ode
l
in
to
da
il
y
he
a
lt
hc
a
r
e
pr
a
c
ti
c
e
i
s
e
xp
e
c
te
d
to
im
pr
ove
p
a
ti
e
nt
c
a
r
e
s
ta
nda
r
ds
a
nd
opt
im
iz
e
th
e
a
ll
oc
a
ti
on
of
m
e
di
c
a
l
r
e
s
our
c
e
s
.
I
n
f
ut
ur
e
,
th
e
s
tu
dy
e
xt
e
nd
s
th
e
s
c
ope
of
th
e
pr
opos
e
d
pr
e
di
c
ti
ve
m
ode
l
w
it
h
s
ophi
s
ti
c
a
te
d de
e
p l
e
a
r
ni
ng mode
l
to
di
a
gnos
e
c
e
r
vi
c
a
l
c
a
nc
e
r
us
in
g
pe
p s
m
e
a
r
i
m
a
ge
s
.
R
E
F
E
R
E
N
C
E
S
[
1]
M
.
A
r
byn
e
t
al
.
,
“
E
s
t
i
m
a
t
e
s
of
i
nc
i
de
nc
e
a
nd
m
or
t
a
l
i
t
y
of
c
e
r
vi
c
a
l
c
a
nc
e
r
i
n
2018:
a
w
or
l
dw
i
de
a
na
l
ys
i
s
,
”
T
he
L
anc
e
t
G
l
obal
H
e
al
t
h
, vol
. 8, no. 2, pp. e
191
–
e
203, 2020, doi
:
10.1016/
S
2214
-
109X
(
19)
3048
2
-
6.
[
2]
K
. S
. O
kuna
de
, “
H
um
a
n pa
pi
l
l
om
a
vi
r
us
a
nd c
e
r
vi
c
a
l
c
a
nc
e
r
,”
J
ou
r
nal
of
O
bs
t
e
t
r
i
c
s
and G
y
na
e
c
ol
ogy
, vol
. 40, no. 5, pp. 602
–
608
,
2020, doi
:
10.1080/
01443615.2019.1634030.
[
3]
J
.
H
ube
r
,
A
.
M
ue
l
l
e
r
,
M
.
S
a
i
l
e
r
,
a
nd
P
.
A
.
R
e
gi
dor
,
“
H
um
a
n
pa
pi
l
l
om
a
vi
r
us
pe
r
s
i
s
t
e
nc
e
or
c
l
e
a
r
a
nc
e
a
f
t
e
r
i
nf
e
c
t
i
on
i
n
r
e
pr
oduc
t
i
ve
a
ge
.
W
ha
t
i
s
t
he
s
t
a
t
us
?
R
e
vi
e
w
of
t
he
l
i
t
e
r
a
t
u
r
e
a
nd
ne
w
da
t
a
of
a
va
gi
na
l
ge
l
c
ont
a
i
ni
ng
s
i
l
i
c
a
t
e
di
oxi
de
,
c
i
t
r
i
c
a
c
i
d,
a
nd s
e
l
e
ni
t
e
,”
W
om
e
n’
s
H
e
al
t
h
, vol
. 17, 2021, doi
:
10.1177/
17455065211020702.
[
4]
C
.
A
.
B
ur
m
e
i
s
t
e
r
e
t
al
.
,
“
C
e
r
vi
c
a
l
c
a
nc
e
r
t
he
r
a
pi
e
s
:
C
ur
r
e
nt
c
ha
l
l
e
nge
s
a
nd
f
ut
ur
e
pe
r
s
pe
c
t
i
ve
s
,”
T
um
ou
r
V
i
r
us
R
e
s
e
a
r
c
h
,
vol
.
13,
2022, doi
:
10.1016/
j
.t
vr
.2022.200238.
[
5]
R
.
M
e
hr
ot
r
a
a
nd
K
.
Y
a
da
v,
“
C
e
r
vi
c
a
l
C
a
nc
e
r
:
f
or
m
ul
a
t
i
on
a
nd
i
m
pl
e
m
e
nt
a
t
i
on
of
G
ovt
of
I
ndi
a
gui
de
l
i
ne
s
f
or
s
c
r
e
e
ni
ng
a
nd
m
a
na
ge
m
e
nt
,”
I
ndi
an J
our
nal
of
G
y
ne
c
ol
ogi
c
O
n
c
ol
ogy
, vol
. 20, no. 1, 2022, d
oi
:
10.1007/
s
40944
-
021
-
00602
-
z.
[
6]
L
. A
l
l
a
hqol
i
e
t
al
.
, “
D
e
l
a
ye
d
c
e
r
vi
c
a
l
c
a
nc
e
r
di
a
gno
s
i
s
:
a
s
y
s
t
e
m
a
t
i
c
r
e
vi
e
w
,”
E
ur
ope
an R
e
v
i
e
w
f
or
M
e
di
c
al
and P
har
m
ac
ol
ogi
c
al
Sc
i
e
nc
e
s
, vol
. 26, no. 22, pp. 8467
–
8480, 2022, doi
:
10.26355/
e
ur
r
e
v_202211_30382.
[
7]
S
. S
. S
ha
s
t
r
i
e
t
al
.
, “
S
e
c
onda
r
y pr
e
ve
nt
i
on of
c
e
r
vi
c
a
l
c
a
nc
e
r
:
A
S
C
O
r
e
s
our
c
e
–
s
t
r
a
t
i
f
i
e
d gui
de
l
i
ne
upda
t
e
,”
J
C
O
G
l
obal
O
nc
ol
ogy
,
no. 8, 2022, doi
:
10.1200/
go.22.00217.
[
8]
M
.
A
ki
nl
ot
a
n,
J
.
N
.
B
ol
i
n,
J
.
H
e
l
du
s
e
r
,
C
.
O
j
i
nna
ka
,
A
.
L
i
c
hor
a
d,
a
nd
D
.
M
c
C
l
e
l
l
a
n,
“
C
e
r
vi
c
a
l
c
a
nc
e
r
s
c
r
e
e
ni
ng
ba
r
r
i
e
r
s
a
nd
r
i
s
k
f
a
c
t
or
know
l
e
dge
a
m
ong
uni
n
s
ur
e
d
w
om
e
n,”
J
our
nal
of
C
om
m
uni
t
y
H
e
al
t
h
,
vol
.
42,
no.
4,
pp.
770
–
778,
2017,
doi
:
10.1007/
s
10900
-
017
-
0316
-
9.
[
9]
S
.
S
i
ngh
a
nd
S
.
B
a
d
a
ya
,
“
F
a
c
t
or
s
i
nf
l
ue
nc
i
ng
upt
a
ke
of
c
e
r
vi
c
a
l
c
a
n
c
e
r
s
c
r
e
e
n
i
ng
a
m
ong
w
om
e
n
i
n
i
ndi
a
:
a
ho
s
pi
t
a
l
-
ba
s
e
d
pi
l
o
t
s
t
udy,”
J
our
nal
of
C
om
m
uni
t
y
M
e
di
c
i
ne
&
H
e
al
t
h E
duc
at
i
on
, 2012, doi
:
10.4172/
2161
-
0711.1000157.
[
10]
A
.
B
.
B
a
ns
a
l
,
A
.
P
.
P
a
kha
r
e
,
N
.
K
a
poor
,
R
.
M
e
hr
ot
r
a
,
a
nd
A
.
M
.
K
oka
ne
,
“
K
n
ow
l
e
dge
,
a
t
t
i
t
ude
,
a
nd
pr
a
c
t
i
c
e
s
r
e
l
a
t
e
d
t
o
c
e
r
vi
c
a
l
c
a
nc
e
r
a
m
ong
a
dul
t
w
om
e
n:
A
ho
s
pi
t
a
l
-
ba
s
e
d
c
r
os
s
-
s
e
c
t
i
ona
l
s
t
udy,”
J
ou
r
nal
of
N
at
ur
al
S
c
i
e
nc
e
,
B
i
ol
ogy
and
M
e
di
c
i
ne
,
vol
.
6
,
no. 2, pp. 324
–
328, 2015, doi
:
10.4103/
0976
-
9668.159993.
[
11]
T
.
U
.
Z
a
m
a
n
e
t
al
.
,
“
A
r
t
i
f
i
c
i
a
l
i
nt
e
l
l
i
ge
nc
e
:
t
he
m
a
j
or
r
ol
e
i
t
pl
a
ye
d
i
n
t
he
m
a
na
ge
m
e
nt
of
he
a
l
t
hc
a
r
e
dur
i
ng
C
O
V
I
D
-
19
pa
nde
m
i
c
,”
I
A
E
S
I
nt
e
r
nat
i
onal
J
our
nal
of
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
nc
e
,
vol
.
12,
no.
2,
pp.
505
–
513,
2023,
doi
:
10.11591/
i
j
a
i
.v12.i
2.pp505
-
513.
[
12]
J
.
J
e
ys
hr
i
a
nd
M
.
K
ow
s
i
ga
n,
“
A
c
om
pr
e
he
ns
i
ve
a
s
s
e
s
s
m
e
nt
of
r
e
c
e
nt
a
dva
nc
e
s
i
n
c
e
r
vi
c
a
l
c
a
nc
e
r
de
t
e
c
t
i
on
f
or
a
ut
om
a
t
e
d
s
c
r
e
e
ni
ng,”
I
m
age
P
r
oc
e
s
s
i
ng and I
nt
e
l
l
i
ge
nt
C
om
put
i
ng Sy
s
t
e
m
s
, pp. 171
–
184,
2022, doi
:
10.1201/
9781003267782
-
11.
[
13]
W
.
S
un,
Z
.
C
a
i
,
Y
.
L
i
,
F
.
L
i
u,
S
.
F
a
ng,
a
nd
G
.
W
a
ng,
“
D
a
t
a
pr
oc
e
s
s
i
ng
a
n
d
t
e
xt
m
i
ni
ng
t
e
c
hnol
ogi
e
s
on
e
l
e
c
t
r
oni
c
m
e
di
c
a
l
r
e
c
or
ds
:
A
r
e
vi
e
w
,”
J
our
nal
of
H
e
al
t
hc
a
r
e
E
ngi
ne
e
r
i
ng
, vol
. 2018, 2018, doi
:
1
0.1155/
2018/
4302425.
[
14]
A
.
K
a
l
a
nt
a
r
i
,
A
.
K
a
m
s
i
n,
S
.
S
ha
m
s
hi
r
ba
nd,
A
.
G
a
ni
,
H
.
A
.
-
R
okny,
a
nd
A
.
T
.
C
hr
onopoul
os
,
“
C
om
put
a
t
i
ona
l
i
nt
e
l
l
i
ge
nc
e
a
ppr
oa
c
he
s
f
or
c
l
a
s
s
i
f
i
c
a
t
i
on
of
m
e
di
c
a
l
da
t
a
:
S
t
a
t
e
-
of
-
t
he
-
a
r
t
,
f
ut
u
r
e
c
ha
l
l
e
nge
s
a
nd
r
e
s
e
a
r
c
h
di
r
e
c
t
i
ons
,”
N
e
ur
oc
om
put
i
ng
,
vol
.
276, pp. 2
–
22, 2018, doi
:
10.1016/
j
.ne
uc
om
.2017.01.126.
[
15]
M
.
F
.
I
j
a
z
,
M
.
A
t
t
i
que
,
a
nd
Y
.
S
on,
“
D
a
t
a
-
dr
i
ve
n
c
e
r
vi
c
a
l
c
a
nc
e
r
pr
e
di
c
t
i
on
m
ode
l
w
i
t
h
out
l
i
e
r
de
t
e
c
t
i
on
a
nd
ove
r
-
s
a
m
pl
i
ng
m
e
t
hods
,”
Se
ns
o
r
s
, vol
. 20, no. 10, 2020, doi
:
10.3390/
s
20102809.
[
16]
U
.
K
.
L
i
l
hor
e
e
t
al
.
,
“
H
ybr
i
d
m
ode
l
f
or
de
t
e
c
t
i
on
of
c
e
r
vi
c
a
l
c
a
nc
e
r
us
i
ng
c
a
us
a
l
a
na
l
ys
i
s
a
nd
m
a
c
hi
ne
l
e
a
r
ni
ng
t
e
c
hni
que
s
,
”
C
om
put
at
i
onal
and M
at
he
m
at
i
c
al
M
e
t
hods
i
n M
e
di
c
i
ne
, vol
. 2022, 2022, doi
:
10.1155/
2022/
4688327.
[
17]
E
. R
. P
ut
r
i
, A
. Z
a
r
ka
s
i
, P
. P
r
a
j
i
t
no, a
nd D
. S
. S
oe
j
oko, “
A
r
t
i
f
i
c
i
a
l
ne
ur
a
l
ne
t
w
or
k f
or
c
e
r
vi
c
a
l
a
bnor
m
a
l
i
t
i
e
s
de
t
e
c
t
i
on on c
om
put
e
d
t
om
ogr
a
phy
i
m
a
ge
s
,”
I
A
E
S
I
nt
e
r
nat
i
onal
J
ou
r
nal
of
A
r
t
i
f
i
c
i
al
I
nt
e
l
l
i
ge
n
c
e
,
vol
.
12,
no.
1,
pp.
171
–
179,
2023,
doi
:
10.11591/
i
j
a
i
.v12.i
1.pp171
-
179.
[
18]
F
.
C
ur
i
a
,
“
C
e
r
vi
c
a
l
c
a
nc
e
r
r
i
s
k
pr
e
di
c
t
i
on
w
i
t
h
r
obus
t
e
ns
e
m
bl
e
a
nd
e
xpl
a
i
na
bl
e
bl
a
c
k
boxe
s
m
e
t
hod,
”
H
e
al
t
h
and
T
e
c
hnol
ogy
,
vol
. 11, no. 4, pp. 875
–
885, 2021, doi
:
10.1007/
s
12553
-
021
-
00554
-
6.
[
19]
N
.
Y
oune
s
z
a
de
,
M
.
M
a
r
j
a
ni
,
a
nd
C
.
P
.
P
e
i
,
“
D
e
e
p
l
e
a
r
ni
ng
i
n
c
e
r
vi
c
a
l
c
a
n
c
e
r
di
a
gnos
i
s
:
a
r
c
hi
t
e
c
t
ur
e
,
oppor
t
uni
t
i
e
s
,
a
nd
ope
n
r
e
s
e
a
r
c
h c
ha
l
l
e
nge
s
,”
I
E
E
E
A
c
c
e
s
s
, vol
. 11, pp. 6133
–
6149, 2023, doi
:
10.1109/
A
C
C
E
S
S
.2023.3235833.
Evaluation Warning : The document was created with Spire.PDF for Python.