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.
1
4
,
No
.
6
,
Dec
em
b
er
2
0
2
5
,
p
p
.
4
8
0
2
~
4
8
1
3
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
14
.i
6
.
p
p
4
8
0
2
-
4
8
1
3
4802
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
M
ulti
-
pha
se
featu
re selec
tion for
de
tect
io
n of
epi
theli
a
l ov
a
ria
n
ca
ncer using
e
nse
mble ma
chine
lea
rning
t
ech
niques
Su
m
a
P
a
la
ni Su
bra
m
a
ny
a
,
S
um
a
K
un
cha
Venk
a
t
a
pa
t
hia
h
D
e
p
a
r
t
me
n
t
of
E
l
e
c
t
r
o
n
i
c
s
a
n
d
C
o
mm
u
n
i
c
a
t
i
o
n
E
n
g
i
n
e
e
r
i
n
g
,
R
a
m
a
i
a
h
I
n
s
t
i
t
u
t
e
o
f
T
e
c
h
n
o
l
o
g
y
,
A
f
f
i
l
i
a
t
e
d
t
o
V
i
s
v
e
sv
a
r
a
y
a
Te
c
h
n
o
l
o
g
i
c
a
l
U
n
i
v
e
r
si
t
y
,
B
e
l
a
g
a
v
i
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
an
3
0
,
2
0
2
5
R
ev
is
ed
Sep
1
2
,
2
0
2
5
Acc
ep
ted
Oct
1
8
,
2
0
2
5
Ep
it
h
e
l
ial
o
v
a
rian
c
a
rc
in
o
m
a
is
o
n
e
of
th
e
m
o
st
p
re
v
a
len
t
c
a
u
se
s
o
f
d
e
a
th
.
Ti
m
e
ly
o
v
a
rian
c
a
n
c
e
r
d
iag
n
o
sis
i
s
sig
n
i
fica
n
t
f
o
r
b
e
tt
e
rin
g
p
a
ti
e
n
t
o
u
tco
m
e
s
a
n
d
ra
tes
o
f
su
rv
i
v
a
l.
F
o
r
p
ro
g
n
o
stic
a
n
d
d
iag
n
o
stic
e
v
a
l
u
a
ti
o
n
o
f
m
a
li
g
n
a
n
c
ies
,
AI
-
b
a
se
d
m
a
c
h
in
e
lea
rn
in
g
a
l
g
o
r
it
h
m
s
a
re
u
se
d
.
T
h
is
n
o
v
e
l
tec
h
n
iq
u
e
is
u
n
d
o
u
b
ted
l
y
a
n
e
ffe
c
ti
v
e
t
o
o
l
t
h
a
t
m
a
y
a
id
in
se
lec
ti
n
g
t
h
e
b
e
st
c
o
u
rse
o
f
a
c
ti
o
n
.
T
h
e
c
o
ll
e
c
ti
o
n
o
f
d
a
ta co
m
p
risin
g
1
5
0
p
a
ti
e
n
ts co
n
tain
e
d
a
n
e
x
ten
siv
e
se
lec
ti
o
n
o
f
c
li
n
ica
l
c
h
a
ra
c
teristics
a
n
d
m
a
rk
e
rs
o
f
t
u
m
o
rs.
Th
e
re
c
u
rsiv
e
fe
a
tu
re
e
li
m
in
a
ti
o
n
(
RF
E)
a
n
d
c
o
r
re
latio
n
c
o
e
fficie
n
t
fe
a
tu
re
se
lec
ti
o
n
tec
h
n
i
q
u
e
s
we
re
a
ss
imilate
d
to
p
ick
th
e
fe
a
t
u
re
s
fo
r
t
h
e
m
a
c
h
in
e
lea
rn
in
g
m
o
d
e
l
,
su
c
h
a
s
a
g
e
,
C
A
-
1
2
5
,
t
u
m
o
r
late
ra
li
ty
,
siz
e
,
t
u
m
o
r
ty
p
e
,
g
ra
d
e
o
f
tu
m
o
r
,
a
n
d
In
ter
n
a
ti
o
n
a
l
F
e
d
e
ra
ti
o
n
o
f
G
y
n
e
c
o
lo
g
y
a
n
d
Ob
ste
tri
c
s
(
F
IG
O
)
sta
g
e
.
Th
e
stu
d
y
’s
fin
d
i
n
g
s
in
d
ica
te
th
a
t
th
e
b
a
se
m
o
d
e
l
a
c
c
u
ra
c
y
wa
s
a
ro
u
n
d
9
6
%
,
se
n
siti
v
it
y
9
3
%
,
a
n
d
sp
e
c
ifi
c
it
y
1
0
0
%
.
Us
in
g
e
n
se
m
b
le
c
las
sifica
ti
o
n
,
a
c
c
u
ra
c
y
wa
s
a
ro
u
n
d
9
6
%
,
se
n
siti
v
it
y
9
8
%
,
a
n
d
sp
e
c
ifi
c
it
y
9
4
%
f
o
r
t
h
e
RF
E
tec
h
n
iq
u
e
.
B
y
o
b
tain
i
n
g
a
d
e
e
p
e
r
u
n
d
e
rsta
n
d
in
g
o
f
t
h
e
ir
d
e
c
isio
n
-
m
a
k
i
n
g
p
r
o
c
e
ss
,
e
x
p
lain
a
b
le
a
rti
ficia
l
i
n
telli
g
e
n
c
e
m
a
k
e
s
so
p
h
ist
ica
ted
m
a
c
h
in
e
lea
rn
in
g
m
e
th
o
d
s
e
a
sie
r
to
e
x
p
lain
.
Be
fo
re
b
e
g
in
n
i
n
g
trea
tme
n
t,
th
is
re
se
a
rc
h
o
ffe
rs
c
ru
c
ial
d
a
ta
fo
r
th
e
d
iag
n
o
sis
a
n
d
p
ro
g
n
o
sis
a
ss
e
ss
m
e
n
t
o
f
in
d
i
v
id
u
a
ls wi
th
e
p
it
h
e
li
a
l
o
v
a
rian
c
a
n
c
e
r
(
EOC
)
.
K
ey
w
o
r
d
s
:
C
o
r
r
elatio
n
co
ef
f
icie
n
t
E
n
s
em
b
le
class
if
ier
Ma
ch
in
e
lear
n
in
g
Ov
ar
ian
ca
n
ce
r
R
ec
u
r
s
iv
e
f
ea
tu
r
e
elim
in
atio
n
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Su
m
a
Palan
i Su
b
r
am
an
y
a
Dep
ar
tm
en
t o
f
E
lectr
o
n
ics an
d
C
o
m
m
u
n
icatio
n
E
n
g
in
ee
r
i
n
g
,
R
am
aiah
I
n
s
titu
te
o
f
T
ec
h
n
o
l
o
g
y
Af
f
iliated
to
Vis
v
esv
ar
ay
a
T
e
ch
n
o
lo
g
ical
Un
iv
er
s
ity
B
elag
av
i
-
5
9
0
0
1
8
,
I
n
d
ia
E
m
ail: su
m
ap
1
9
9
4
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Ov
ar
ian
ca
n
ce
r
r
em
ain
s
a
m
a
jo
r
h
ea
lth
co
n
ce
r
n
b
ec
a
u
s
e
o
f
th
e
d
if
f
icu
lties
in
tr
ea
tin
g
it,
wh
ich
ar
e
m
o
s
tly
r
elate
d
to
d
elay
ed
id
en
tific
atio
n
an
d
th
e
in
tr
icac
y
o
f
its
d
is
ea
s
e
d
etec
tio
n
.
I
n
d
ia
is
ex
p
ec
ted
to
s
ee
4
9
,
6
4
4
n
ew
ca
s
es
o
f
o
v
ar
ian
ca
n
ce
r
in
2
0
2
5
[
1
]
.
T
h
is
is
a
m
in
o
r
r
is
e
o
v
e
r
th
e
4
3
,
8
8
6
i
n
s
tan
ce
s
th
at
wer
e
p
r
ed
icted
f
o
r
2
0
2
0
.
Ad
d
itio
n
a
lly
,
th
e
f
i
v
e
-
y
ea
r
s
u
r
v
i
v
al
r
at
e
f
r
o
m
2
0
1
5
to
2
0
2
1
was
5
1
.
6
%
[
2
]
,
[
3
]
.
T
h
ese
o
u
tco
m
es
h
i
g
h
lig
h
t
th
e
im
p
licatio
n
o
f
ea
r
ly
d
etec
tio
n
an
d
th
er
a
p
eu
tic
s
tr
ateg
ies,
an
ar
ea
o
f
co
n
tin
u
o
u
s
r
esear
ch
n
ee
d
t
o
b
e
d
ev
elo
p
e
d
[
4
]
,
[
5
]
.
T
h
e
d
iag
n
o
s
tic
m
eth
o
d
s
th
at
ar
e
em
p
lo
y
e
d
to
id
en
tif
y
o
v
ar
ian
ca
n
ce
r
in
clu
d
e
l
o
wer
ab
d
o
m
en
test
s
,
m
ed
ical
im
ag
in
g
p
r
o
ce
d
u
r
es
s
u
ch
as
tr
an
s
v
ag
in
al
u
ltra
s
o
u
n
d
(
T
VUS)
,
m
a
g
n
etic
r
eso
n
a
n
ce
im
ag
in
g
(
MRI)
,
an
d
b
lo
o
d
test
s
.
T
h
e
m
o
s
t
p
r
ev
alen
t
b
io
m
a
r
k
er
s
a
r
e
ca
n
ce
r
an
tig
e
n
1
2
5
(
C
A
-
1
2
5
)
,
C
A1
9
–
9,
ca
r
cin
o
em
b
r
y
o
n
ic
an
tig
e
n
(
C
E
A)
,
an
d
h
u
m
an
ep
i
d
id
y
m
is
p
r
o
tein
4
(
HE
4
)
.
T
h
e
m
o
s
t
r
e
s
p
o
n
s
ib
le
tech
n
iq
u
e
f
o
r
d
eter
m
in
in
g
o
v
ar
ian
ca
n
ce
r
is
h
is
to
p
ath
o
lo
g
y
b
ec
au
s
e
it
ex
am
in
es
tis
s
u
e
s
am
p
les
at
th
e
ce
llu
lar
lev
el
[
6
]
.
T
h
e
r
eliab
ilit
y
o
f
T
VUS
an
d
p
elv
ic
ex
am
in
atio
n
i
n
s
eg
r
eg
ati
n
g
b
etwe
en
b
en
ig
n
an
d
m
alig
n
an
t
tu
m
o
r
s
is
v
er
y
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
Mu
lti
-
p
h
a
s
e
fea
tu
r
e
s
elec
tio
n
fo
r
d
etec
tio
n
o
f e
p
ith
eli
a
l o
va
r
i
a
n
ca
n
ce
r
u
s
in
g
…
(
S
u
ma
P
a
la
n
i S
u
b
r
a
ma
n
ya
)
4803
lo
w.
Similar
ly
,
it
m
ay
b
e
d
if
f
icu
lt
f
o
r
MRI
to
r
eliab
ly
id
en
tify
wh
eth
er
a
tu
m
o
r
is
m
alig
n
an
t,
b
u
t
it
ca
n
p
r
o
d
u
ce
p
r
ec
is
e
im
ag
es
o
f
th
e
o
v
ar
ies
an
d
th
e
s
u
r
r
o
u
n
d
in
g
tis
s
u
es.
HE
4
,
ca
r
b
o
h
y
d
r
ate
an
tig
en
7
2
-
4
(
C
A7
2
-
4)
,
a
n
d
ca
r
b
o
h
y
d
r
ate
a
n
tig
en
1
2
5
(
C
A
-
1
2
5
)
ar
e
im
p
o
r
tan
t
ca
n
ce
r
b
io
m
ar
k
er
s
th
at
ca
n
id
en
tify
f
em
ale
p
elv
ic
tu
m
o
r
s
[
7
]
–
[
9
]
.
A
ty
p
ical
co
u
r
s
e
o
f
m
ed
icatio
n
f
o
r
o
v
ar
ia
n
ca
n
ce
r
in
v
o
l
v
es
s
u
r
g
er
y
f
o
llo
wed
b
y
ch
e
m
o
th
er
ap
y
[
1
0
]
.
B
u
t
m
o
s
t
o
f
th
e
p
atien
ts
ev
e
n
tu
ally
ex
p
e
r
ien
ce
a
r
ec
u
r
r
en
ce
of
th
e
d
is
ea
s
e
th
at
is
u
s
u
ally
i
n
cu
r
ab
le,
p
r
im
ar
il
y
as
a
co
n
s
eq
u
en
ce
o
f
th
e
f
o
r
m
atio
n
o
f
d
r
u
g
r
esis
tan
ce
.
B
ec
au
s
e
o
f
th
e
s
ev
er
e
s
id
e
ef
f
ec
ts
an
d
ex
p
en
s
e
o
f
ch
em
o
th
er
a
p
y
,
p
r
ec
is
io
n
m
ed
i
cin
e
aim
s
to
class
if
y
p
atien
ts
wh
o
s
e
m
alig
n
an
cies a
r
e
r
esis
tan
t to
th
e
tr
ea
tm
e
n
t.
Ar
tific
ial
in
tellig
en
ce
(
AI
)
is
ess
en
tial
in
th
e
t
h
er
ap
eu
tic
in
d
u
s
tr
y
f
o
r
ca
n
ce
r
d
ia
g
n
o
s
is
an
d
d
etec
tio
n
[
1
1
]
.
I
n
a
r
is
in
g
n
u
m
b
er
o
f
m
e
d
ical
ap
p
licatio
n
s
,
AI
is
em
er
g
in
g
as
a
f
ea
s
ib
le
alter
n
ativ
e
f
o
r
d
ec
is
io
n
-
m
a
k
in
g
alg
o
r
ith
m
s
.
I
n
t
h
e
m
e
d
ical
f
i
eld
,
m
ac
h
i
n
e
lear
n
in
g
is
f
r
eq
u
en
tly
u
s
ed
to
ev
alu
ate
p
atie
n
t
d
ata
a
n
d
p
r
o
v
id
e
ea
r
ly
d
is
ea
s
e
d
iag
n
o
s
es.
T
h
e
m
ac
h
in
e
r
ec
ei
v
es
th
e
p
atien
t'
s
k
ey
ch
ar
ac
ter
is
tics
as
in
p
u
t
a
n
d
o
u
tp
u
ts
a
p
r
ec
is
e
d
iag
n
o
s
is
[
1
2
]
.
I
n
n
o
v
ativ
e
ap
p
r
o
ac
h
es
u
s
in
g
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
h
o
ld
s
ig
n
i
f
ican
t
p
r
o
m
is
e
f
o
r
d
iag
n
o
s
in
g
ca
n
ce
r
an
d
f
o
r
ec
as
tin
g
th
e
co
u
r
s
e
o
f
d
is
ea
s
e.
Ma
n
y
r
esear
ch
er
s
h
av
e
w
o
r
k
ed
o
n
m
ac
h
in
e
lear
n
i
n
g
alg
o
r
ith
m
s
in
o
r
d
er
to
ac
c
u
r
ately
d
etec
t
o
v
a
r
ian
ca
n
ce
r
u
s
in
g
th
ese
b
io
m
ar
k
er
s
.
Ar
ez
z
o
et
a
l
.
[
1
2
]
u
s
ed
m
ac
h
in
e
lear
n
i
n
g
al
g
o
r
ith
m
s
o
n
u
ltra
s
o
u
n
d
i
m
ag
es
to
ca
lc
u
late
th
e
1
2
-
m
o
n
th
s
u
r
v
iv
al
p
er
io
d
f
o
r
o
v
a
r
ian
ca
n
ce
r
p
atien
ts
.
Fu
r
th
e
r
f
iv
e
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
was
u
s
ed
to
tr
ain
a
n
d
v
alid
ate
th
r
e
e
d
is
tin
ct
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
,
lo
g
is
tic
r
eg
r
ess
io
n
(
L
R
)
,
r
an
d
o
m
f
o
r
est
(
R
F)
,
an
d
k
-
n
ea
r
est
n
eig
h
b
o
r
s
(
KNN)
to
f
o
r
ec
ast
a
12
-
m
o
n
th
s
u
r
v
iv
al
p
e
r
io
d
.
T
h
e
h
ig
h
est
p
er
f
o
r
m
an
ce
ac
c
u
r
ac
y
was
9
3
.
7
%.
Z
iy
am
b
e
et
al
.
[
1
3
]
ap
p
lied
a
co
n
v
o
l
u
tio
n
n
e
u
r
al
n
etwo
r
k
to
h
is
to
p
ath
o
lo
g
ical
im
ag
es
to
p
r
ed
ict
an
d
d
ia
g
n
o
s
e
o
v
ar
ian
c
an
ce
r
an
d
ac
h
iev
ed
an
ac
cu
r
ac
y
o
f
9
4
%.
T
h
e
p
r
in
cip
al
in
te
n
t
o
f
th
is
r
esear
ch
is
to
em
p
lo
y
en
s
em
b
l
e
-
b
ased
m
ac
h
in
e
-
lea
r
n
in
g
al
g
o
r
ith
m
s
to
ass
es
s
th
e
p
r
e
-
o
p
e
r
ativ
e
s
tatu
s
o
f
th
o
s
e
d
iag
n
o
s
ed
with
o
v
a
r
ian
ca
n
ce
r
.
T
h
e
m
o
s
t
im
p
o
r
ta
n
t
f
ea
tu
r
es
,
s
u
ch
as
ag
e,
tu
m
o
r
later
ality
,
s
ize,
tu
m
o
r
ty
p
e,
tu
m
o
r
g
r
ad
e,
I
n
ter
n
atio
n
al
Fed
er
atio
n
o
f
Gy
n
ec
o
l
o
g
y
a
n
d
Ob
s
tetr
ics
(
FIG
O
)
s
tag
e
,
an
d
C
A
-
1
2
5
,
a
r
e
s
elec
ted
u
s
in
g
two
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
es.
T
h
e
r
e
is
a
clo
s
e
ass
o
ciatio
n
b
etwe
en
th
e
c
lin
ical
f
ac
to
r
s
a
n
d
th
e
ef
f
ec
tiv
e
tu
m
o
r
m
ar
k
er
C
A
-
1
2
5
.
T
h
is
will
allo
w
th
e
d
o
cto
r
s
to
tr
ea
t
th
e
p
atien
ts
ap
p
r
o
p
r
iately
,
wh
ich
will in
cr
ea
s
e
p
atien
ts
'
lo
n
g
ev
it
y
.
T
h
e
p
a
p
er
is
d
esig
n
e
d
as
f
o
llo
ws:
t
h
e
p
r
io
r
in
v
esti
g
atio
n
s
c
o
n
d
u
cted
f
o
r
th
e
d
iag
n
o
s
is
an
d
d
etec
tio
n
o
f
o
v
ar
ian
ca
n
ce
r
ar
e
co
v
er
ed
in
s
ec
tio
n
2
.
A
th
o
r
o
u
g
h
ex
p
lan
atio
n
o
f
ea
ch
elem
en
t
o
f
th
e
s
u
g
g
ested
f
r
am
ewo
r
k
is
g
iv
en
i
n
s
ec
tio
n
3
.
Sectio
n
4
p
r
esen
ts
th
e
f
i
n
d
in
g
s
an
d
a
n
an
aly
s
is
o
f
th
e
r
esear
ch
.
Sectio
n
5
p
r
esen
ts
th
e
s
tu
d
y
'
s
co
n
clu
s
io
n
s
.
2.
L
I
T
E
R
AT
U
RE
SU
RVE
Y
T
h
e
r
e
c
e
n
t
s
t
u
d
i
es
e
m
p
l
o
y
e
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
m
o
d
e
l
s
o
n
s
ig
n
i
f
i
c
a
n
t
b
i
o
m
a
r
k
e
r
s
f
o
r
t
h
e
d
e
t
e
c
t
i
o
n
o
f
o
v
a
r
i
a
n
c
a
n
c
e
r
.
Di
f
f
e
r
e
n
t
m
a
ch
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
it
h
m
s
p
r
o
p
o
s
e
d
b
y
L
a
v
a
n
y
a
a
n
d
Pa
s
u
p
a
t
h
i
[
1
4
]
i
n
c
l
u
d
e
K
NN
,
s
u
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
(
S
V
M
)
,
d
e
c
i
s
i
o
n
t
r
e
es
(
D
T
)
,
f
o
l
l
o
w
ed
b
y
m
a
x
v
o
t
i
n
g
,
b
o
o
s
t
i
n
g
,
b
a
g
g
i
n
g
,
a
n
d
s
t
a
c
k
i
n
g
.
D
a
t
a
w
as
c
o
ll
e
c
t
e
d
f
r
o
m
K
a
g
g
l
e
.
T
o
s
e
l
e
ct
t
h
e
f
e
a
t
u
r
es
m
in
i
m
u
m
r
e
d
u
n
d
a
n
c
y
m
a
x
i
m
u
m
r
e
l
e
v
a
n
c
e
(
M
R
M
R
)
a
l
g
o
r
i
t
h
m
w
a
s
u
s
e
d
.
SV
M
h
a
s
8
5
%
a
c
c
u
r
a
c
y
,
a
n
d
s
t
a
c
k
i
n
g
8
9
%
.
W
i
b
o
w
o
et
al
.
[
1
5
]
d
i
s
c
u
s
s
e
d
t
h
e
c
l
a
s
s
i
f
i
c
a
ti
o
n
o
f
o
v
a
r
i
a
n
c
a
n
c
e
r
u
s
i
n
g
K
NN
an
d
S
V
M
a
n
d
a
c
h
i
e
v
e
d
a
cl
a
s
s
i
f
i
c
a
t
i
o
n
a
c
c
u
r
a
c
y
o
f
9
0
.
4
7
%
f
o
r
K
N
N
.
Ah
am
ad
et
a
l
.
[
1
6
]
f
o
cu
s
ed
o
n
en
s
em
b
le
m
o
d
els
in
ad
d
it
io
n
to
m
ac
h
in
e
lea
r
n
in
g
tec
h
n
iq
u
es
to
ca
teg
o
r
ize
b
etwe
en
h
ea
lth
y
a
n
d
ca
n
ce
r
o
u
s
p
atien
ts
.
Var
io
u
s
s
ig
n
if
ican
t
B
io
m
ar
k
er
s
u
s
ed
in
th
e
s
tu
d
y
ar
e
CA
-
1
2
5
,
HE
-
4
,
C
E
A,
an
d
C
A1
9
-
9
.
Ov
er
all,
th
is
wo
r
k
attain
ed
an
ac
cu
r
ac
y
o
f
9
1
%.
A
m
ac
h
in
e
lear
n
in
g
m
o
d
el,
p
r
o
p
o
s
ed
b
y
T
ale
b
et
a
l
.
[
1
7
]
,
u
s
es
m
ac
h
in
e
lear
n
i
n
g
alg
o
r
ith
m
s
to
p
r
o
g
r
ess
in
th
e
p
r
ec
is
io
n
o
f
o
v
ar
ian
ca
n
ce
r
d
ia
g
n
o
s
is
.
T
h
e
m
o
d
el
i
s
s
im
u
lated
u
s
in
g
MA
T
L
AB
2
0
2
1
a.
Per
f
o
r
m
an
ce
is
ass
ess
e
d
u
s
in
g
a
v
ar
iatio
n
o
f
s
tatis
tical
m
etr
ics u
s
in
g
th
e
p
r
o
p
o
s
ed
m
o
d
el
,
with
an
ac
c
u
r
ac
y
o
f
9
7
.
1
6
%.
Ah
am
ad
et
al
.
[
1
6
]
id
en
tif
y
m
ajo
r
b
lo
o
d
b
io
m
ar
k
e
r
s
lik
e
C
A
-
1
2
5
,
C
A
1
9
-
9
,
C
E
A
,
an
d
HE
-
4
,
alo
n
g
with
o
th
er
cr
itical
p
ar
am
eter
s
.
T
h
e
s
tu
d
y
d
is
cu
s
s
es
th
e
ap
p
licatio
n
o
f
s
ev
er
al
m
ac
h
in
e
lear
n
in
g
m
eth
o
d
s
,
em
p
h
asizin
g
th
e
n
e
ed
f
o
r
ea
r
ly
id
e
n
tific
atio
n
to
im
p
r
o
v
e
p
atien
t
o
u
tco
m
es,
i
n
clu
d
in
g
DT
,
R
F,
SVM,
g
r
ad
ien
t
b
o
o
s
tin
g
m
a
ch
in
e
(
GB
M
)
,
L
R
,
lig
h
t
g
r
a
d
ien
t
b
o
o
s
tin
g
m
ac
h
in
e
(
L
G
B
M
)
,
an
d
ex
tr
em
e
g
r
ad
ien
t
b
o
o
s
tin
g
(
XGB
)
,
in
c
ateg
o
r
izin
g
o
v
a
r
ian
ca
n
ce
r
p
at
ien
ts
b
ased
o
n
clin
ical
d
ata
w
ith
an
ac
c
u
r
ac
y
o
f
9
1
%.
W
an
g
et
al
.
[
1
8
]
p
er
f
o
r
m
ed
in
itial
s
cr
ee
n
in
g
o
f
o
v
ar
i
an
ca
n
ce
r
b
ased
o
n
th
e
r
is
k
o
f
o
v
ar
ia
n
m
alig
n
an
cy
alg
o
r
ith
m
(
R
OM
A)
.
T
h
e
b
io
m
ar
k
er
s
HE
-
4
,
C
A
-
1
2
5
wer
e
u
s
ed
an
d
o
b
tain
ed
an
AUC
o
f
0
.
9
1
f
o
r
R
OM
A.
B
ast
et
a
l
.
[
1
9
]
talk
a
b
o
u
t
u
s
in
g
b
io
m
ar
k
er
s
to
d
iag
n
o
s
e
o
v
ar
ian
ca
n
ce
r
e
ar
ly
,
s
u
c
h
as
C
A
-
1
2
5
,
m
icr
o
R
NAs,
ctDNA
,
an
d
m
eth
y
lated
DNA.
A
p
er
f
o
r
m
an
ce
o
f
9
8
% is
ac
h
iev
ed
f
o
r
all
th
e
co
n
tr
o
l su
b
ject
s
.
W
ib
o
wo
et
al
.
[
1
5
]
m
ain
o
b
jectiv
e
was
to
clas
s
if
y
o
v
ar
ian
ca
n
ce
r
u
s
in
g
m
ac
h
in
e
lear
n
in
g
p
r
o
ce
d
u
r
es,
n
a
m
ely
KNN
a
n
d
SVM.
T
h
is
p
a
p
er
e
x
p
lain
s
h
o
w
well
m
ac
h
in
e
lear
n
in
g
p
r
o
c
ess
es
s
u
ch
as
KNN
an
d
SVM
ca
teg
o
r
ize
o
v
a
r
ian
ca
n
ce
r
ca
s
es;
in
th
is
p
ar
ticu
lar
r
esear
ch
,
KNN
p
er
f
o
r
m
ed
b
e
tter
th
an
SVM
with
an
ac
cu
r
ac
y
o
f
9
0
.
4
7
%.
T
o
in
cr
ea
s
e
th
e
cu
r
r
e
n
t
b
io
m
ar
k
e
r
co
m
b
in
atio
n
m
o
d
el'
s
ab
ilit
y
to
class
if
y
o
v
ar
ian
ca
n
ce
r
,
So
n
g
et
al
.
[
2
0
]
aim
t
o
in
clu
d
e
m
en
o
p
au
s
al
d
ata.
T
h
e
ar
ea
u
n
d
er
th
e
R
OC
cu
r
v
e
o
f
0
.
9
8
5
s
p
ec
if
ies
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.
1
4
,
No
.
6
,
Dec
em
b
er
2
0
2
5
:
4
8
0
2
-
4
8
1
3
4804
th
at
th
e
ty
p
ical
with
m
en
o
p
a
u
s
al
d
ata
ac
h
iev
es
b
etter
th
a
n
th
e
m
o
d
el
with
o
u
t
clin
ical
d
ata
in
th
e
s
tu
d
y
'
s
ev
alu
atio
n
f
o
r
o
v
ar
ian
ca
n
ce
r
s
cr
ee
n
in
g
.
T
o
in
cr
ea
s
e
th
e
d
etec
tio
n
ac
c
u
r
ac
y
Ab
u
zi
n
ad
ah
et
al
.
[
2
1
]
u
s
ed
en
s
em
b
le
m
o
d
els
s
u
ch
a
s
b
ag
g
in
g
an
d
b
o
o
s
tin
g
wh
ich
in
clu
d
e
5
0
f
ea
tu
r
es.
C
o
m
b
in
in
g
en
s
em
b
le
m
o
d
els
lo
wer
s
th
e
v
ar
ian
c
e
f
o
r
p
r
ec
is
e
o
v
a
r
ian
ca
n
ce
r
d
etec
tio
n
,
an
d
th
e
r
esu
ltin
g
ac
cu
r
ac
y
o
f
9
6
.
8
7
%
was
ac
h
iev
ed
.
I
n
o
r
d
er
to
g
en
er
at
e
a
n
ew
p
r
ed
ictiv
e
d
iag
n
o
s
is
,
Paik
et
a
l
.
[
2
2
]
ass
ess
ed
th
e
ac
cu
r
ac
y
o
f
g
r
a
d
i
en
t
b
o
o
s
tin
g
(
GB
)
u
s
in
g
c
o
n
v
en
tio
n
al
s
tatis
tical
m
eth
o
d
s
.
T
h
e
r
esear
ch
p
r
e
s
en
ted
h
o
w
m
ac
h
in
e
lear
n
i
n
g
tech
n
iq
u
es,
p
ar
ticu
la
r
ly
GB
,
ca
n
s
u
r
p
ass
co
n
v
en
tio
n
al
s
tatis
tical
m
eth
o
d
s
in
p
r
ec
is
ely
f
o
r
ec
asti
n
g
th
e
in
d
iv
id
u
al
o
u
tco
m
es
o
f
ep
ith
elial
o
v
ar
ian
ca
n
ce
r
(
E
OC
)
p
atien
ts
with
an
ar
ea
u
n
d
er
th
e
c
u
r
v
e
(
AUC
)
v
alu
e
o
f
0
.
8
3
0
.
Nay
ak
et
al
.
[
2
3
]
d
escr
ib
ed
RF
as
a
m
ac
h
in
e
lear
n
in
g
class
if
ier
wh
ich
r
esu
lted
in
an
ac
cu
r
ac
y
o
f
9
1
%.
T
ab
le
1
d
etails
ab
o
u
t
th
e
wo
r
k
ca
r
r
ied
o
u
t
in
th
e
r
ec
en
t liter
atu
r
e.
T
h
e
m
o
s
t
r
e
c
e
n
t
s
t
u
d
y
f
o
c
u
s
ed
o
n
d
e
t
e
c
ti
n
g
o
v
a
r
i
a
n
c
a
n
c
e
r
s
o
l
el
y
t
h
r
o
u
g
h
b
i
o
m
a
r
k
e
r
s
.
Ho
w
e
v
e
r
,
i
n
o
r
d
e
r
t
o
o
f
f
e
r
a
m
e
a
n
i
n
g
f
u
l
d
i
a
g
n
o
s
t
i
c
t
o
o
l
f
o
r
e
v
a
l
u
a
ti
n
g
t
h
e
p
r
e
-
o
p
e
r
a
t
i
v
e
c
o
n
d
i
t
i
o
n
o
f
o
v
a
r
i
a
n
c
a
n
c
e
r
s
,
cl
i
n
i
c
al
c
h
a
r
a
c
t
e
r
is
t
i
cs
a
n
d
b
i
o
m
a
r
k
e
r
s
m
u
s
t
b
e
i
n
t
e
g
r
a
t
e
d
.
T
h
e
r
e
f
o
r
e
,
t
h
e
k
e
y
c
o
n
t
r
i
b
u
t
i
o
n
o
f
t
h
i
s
w
o
r
k
a
s
f
o
l
l
o
w
s
:
i
)
t
o
c
h
o
o
s
e
t
h
e
b
e
s
t
c
l
i
n
i
ca
l
c
h
a
r
a
ct
e
r
i
s
t
i
cs
t
h
a
t
a
r
e
ess
e
n
t
i
al
f
o
r
th
e
p
r
o
m
p
t
d
e
t
e
c
t
i
o
n
o
f
o
v
a
r
i
a
n
c
a
n
c
e
r
,
t
w
o
f
e
a
t
u
r
e
s
e
l
e
ct
i
o
n
s
t
r
a
t
e
g
ie
s
a
r
e
u
s
e
d
,
i
i
)
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
e
n
s
em
b
l
e
m
o
d
e
l
s
a
r
e
e
m
p
l
o
y
e
d
to
a
c
h
i
e
v
e
t
h
e
ea
r
l
y
d
i
a
g
n
o
s
i
s
a
n
d
p
r
o
g
n
o
s
i
s
o
f
o
v
a
r
i
a
n
c
a
n
c
e
r
,
a
n
d
i
i
i
)
m
ac
h
in
e
l
e
a
r
n
i
n
g
a
l
g
o
r
it
h
m
s
a
r
e
f
u
r
t
h
e
r
e
n
h
a
n
c
e
d
b
y
e
x
p
l
a
i
n
a
b
l
e
a
r
t
i
f
i
c
i
al
i
n
t
e
ll
i
g
e
n
c
e
(
X
A
I
)
,
w
h
i
c
h
m
a
k
e
s
t
h
e
d
e
c
i
s
i
o
n
-
m
a
k
i
n
g
p
r
o
c
e
d
u
r
e
s
e
x
p
li
c
i
t
,
c
o
m
p
r
e
h
e
n
s
i
b
l
e
a
n
d
e
f
f
e
c
t
i
v
e
.
T
ab
le
1
.
Sy
n
o
p
s
is
o
f
p
e
r
tin
en
t
r
esear
ch
R
e
f
.
M
e
t
h
o
d
u
se
d
S
a
mp
l
e
s
i
z
e
P
e
r
f
o
r
ma
n
c
e
p
a
r
a
m
e
t
e
r
s
Li
mi
t
a
t
i
o
n
s
[
2
4
]
LR
,
S
V
M
,
R
F
,
DT
,
K
N
N
3
5
0
p
a
t
i
e
n
t
s
i
n
t
h
e
K
a
g
g
l
e
d
a
t
a
s
e
t
,
a
n
d
a
m
o
n
g
t
h
e
5
0
p
a
r
a
m
e
t
e
r
s
a
r
e
a
g
e
,
m
e
n
o
p
a
u
s
e
,
a
n
d
t
y
p
e
.
A
c
c
u
r
a
c
y
f
r
o
m
RF
w
a
s
9
0
.
4
%
w
i
t
h
t
h
e
t
o
p
f
e
a
t
u
r
e
s
s
e
l
e
c
t
e
d
w
e
r
e
a
g
e
,
CA
-
1
2
5
,
m
e
n
o
p
a
u
s
e
,
H
E
-
4,
n
e
u
t
r
o
p
h
i
l
s
(
N
E
U
)
,
a
n
d
w
i
t
h
r
a
n
d
o
m
f
e
a
t
u
r
e
s
,
a
c
c
u
r
a
c
y
w
a
s
a
r
o
u
n
d
8
1
.
9
%
.
A
c
c
u
r
a
t
e
i
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
o
v
a
r
i
a
n
c
a
n
c
e
r
n
e
c
e
ssi
t
a
t
e
s
a
b
e
t
t
e
r
c
o
mp
r
e
h
e
n
s
i
o
n
o
f
t
h
e
d
i
s
e
a
se'
s
p
a
t
h
o
p
h
y
si
o
l
o
g
y
.
[
2
5
]
GB
m
a
c
h
i
n
e
,
S
V
M
,
R
F
,
c
o
n
d
i
t
i
o
n
a
l
R
F
(
C
R
F
)
,
n
a
ï
v
e
B
a
y
e
s
(
N
B
)
,
n
e
u
r
a
l
n
e
t
w
o
r
k
,
a
n
d
E
l
a
s
t
i
c
N
e
t
.
1
0
1
i
n
d
i
v
i
d
u
a
l
s
w
i
t
h
n
o
r
m
a
l
o
v
a
r
i
a
n
t
u
m
o
r
s
a
n
d
3
3
4
p
a
t
i
e
n
t
s wi
t
h
EO
C
.
RF
g
a
v
e
t
h
e
b
e
s
t
a
c
c
u
r
a
c
y
a
n
d
A
U
C
,
o
f
6
9
.
0
%
a
n
d
0
.
7
6
0
,
r
e
s
p
e
c
t
i
v
e
l
y
.
R
o
b
u
st
v
a
l
i
d
a
t
i
o
n
a
n
d
c
l
i
n
i
c
a
l
i
mp
l
e
m
e
n
t
a
t
i
o
n
i
n
i
t
i
a
t
i
v
e
s,
e
x
t
e
r
n
a
l
v
a
l
i
d
a
t
i
o
n
i
n
p
o
t
e
n
t
i
a
l
g
r
o
u
p
s i
s
r
e
q
u
i
r
e
d
.
[
2
6
]
R
F
a
l
g
o
r
i
t
h
m w
i
t
h
a
1
0
-
f
o
l
d
c
r
o
ss
-
v
a
l
i
d
a
t
i
o
n
t
e
c
h
n
i
q
u
e
.
1
5
7
ser
u
m
sam
p
l
e
s
f
r
o
m
h
e
a
l
t
h
y
n
o
n
-
c
a
n
c
e
r
c
o
n
t
r
o
l
s
a
n
d
1
4
3
f
r
o
m
o
v
a
r
i
a
n
c
a
n
c
e
r
p
a
t
i
e
n
t
s
F
i
v
e
b
i
o
m
a
r
k
e
r
s
,
A
p
o
l
i
p
o
p
r
o
t
e
i
n
[
A
p
o
]
A
1
,
A
p
o
A
2
,
H
E
-
4
,
C
A
-
1
2
5
,
a
n
d
c
a
n
c
e
r
a
n
t
i
g
e
n
1
5
-
3
,
sh
o
w
e
d
9
3
.
7
1
%
s
e
n
s
i
t
i
v
i
t
y
a
n
d
9
3
.
6
3
%
sp
e
c
i
f
i
c
i
t
y
F
o
c
u
se
d
so
l
e
l
y
o
n
d
i
a
g
n
o
st
i
c
a
c
c
u
r
a
c
y
r
a
t
h
e
r
t
h
a
n
f
o
r
e
c
a
st
i
n
g
r
e
s
p
o
n
s
e
t
o
t
h
e
r
a
p
y
o
r
s
u
r
v
i
v
a
l
[
2
7
]
A
n
a
d
d
i
t
i
o
n
a
l
b
e
n
e
f
i
t
i
s
mu
l
t
i
v
a
r
i
a
t
e
a
n
a
l
y
si
s
w
i
t
h
t
u
m
o
r
b
i
o
mar
k
e
r
s
i
n
a
d
d
i
t
i
o
n
t
o
u
l
t
r
a
s
o
n
o
g
r
a
p
h
y
.
W
o
me
n
a
b
o
v
e
5
0
y
e
a
r
s
a
r
e
d
e
f
i
n
e
d
S
e
n
s
i
t
i
v
i
t
y
a
n
d
s
p
e
c
i
f
i
c
i
t
y
o
f
C
A
-
1
2
5
w
i
t
h
C
A
1
9
–
9
,
EG
F
R
,
G
-
C
S
F
,
Eo
t
a
x
i
n
,
I
L
-
2
R
,
c
V
C
A
M
,
a
n
d
M
I
F
w
e
r
e
9
8
.
2
%
a
n
d
9
8
.
7
%,
r
e
sp
e
c
t
i
v
e
l
y
.
La
c
k
o
f
a
d
e
t
a
i
l
e
d
d
a
t
a
s
e
t
sp
e
c
i
f
i
c
a
t
i
o
n
,
mi
ss
i
n
g
i
n
f
o
r
mat
i
o
n
a
b
o
u
t
t
h
e
d
i
se
a
se
st
a
g
e
s,
u
n
k
n
o
w
n
i
m
a
g
i
n
g
i
n
p
u
t
s
,
o
r
b
i
o
m
a
r
k
e
r
s
[
2
8
]
RF
,
DT
,
G
a
u
ssi
a
n
N
B
,
A
d
a
B
o
o
st
,
a
n
d
LR
.
3
4
9
p
a
t
i
e
n
t
s
RF
c
l
a
ss
i
f
i
e
r
h
a
d
t
h
e
g
r
e
a
t
e
st
v
a
l
i
d
a
t
i
o
n
a
c
c
u
r
a
c
y
a
t
9
9
%
.
L
a
c
k
o
f
c
l
a
r
i
f
i
c
a
t
i
o
n
r
e
g
a
r
d
i
n
g
t
h
e
d
a
t
a
s
o
u
r
c
e
s
,
s
u
c
h
a
s
i
m
a
g
i
n
g
,
b
i
o
m
a
r
k
e
r
s
,
o
r
c
l
i
n
i
c
a
l
r
e
c
o
r
d
s
.
N
o
m
e
c
h
a
n
i
s
m
f
o
r
i
n
t
e
g
r
a
t
i
n
g
c
l
i
n
i
c
a
l
p
r
o
c
e
d
u
r
e
s
[
2
9
]
P
a
r
a
m
e
t
e
r
o
p
t
i
m
i
z
a
t
i
o
n
a
n
d
f
e
a
t
u
r
e
w
e
i
g
h
t
i
n
g
f
o
r
o
v
a
r
i
a
n
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
.
W
e
i
g
h
t
s
m
a
x
i
m
i
z
a
t
i
o
n
-
l
e
a
s
t
a
b
s
o
l
u
t
e
s
h
r
i
n
k
a
g
e
a
n
d
s
e
l
e
c
t
i
o
n
o
p
e
r
a
t
o
r
(
L
A
S
S
O
)
r
e
g
u
l
a
r
i
z
a
t
i
o
n
a
n
d
a
d
v
e
r
s
e
d
r
u
g
e
v
e
n
t
(
A
D
E
)
w
i
t
h
c
r
o
s
s
-
v
a
l
i
d
a
t
i
o
n
e
r
r
o
r
.
2
3
5
p
a
t
i
e
n
t
s
W
i
t
h
K
N
N
,
9
7
.
2
4
%
a
c
c
u
r
a
c
y
a
n
d
9
6
.
4
8
%
(
me
a
n
)
a
c
c
u
r
a
c
y
w
i
t
h
S
V
M
Th
e
i
n
a
b
i
l
i
t
y
t
o
c
o
mp
r
e
h
e
n
d
t
h
e
o
p
t
i
m
i
z
a
t
i
o
n
p
r
o
c
e
sses
mak
e
s
c
l
i
n
i
c
a
l
i
mp
l
e
m
e
n
t
a
t
i
o
n
d
i
f
f
i
c
u
l
t
.
La
c
k
o
f
t
e
st
i
n
g
t
o
v
e
r
i
f
y
r
o
b
u
st
n
e
s
s
o
n
a
se
p
a
r
a
t
e
c
o
h
o
r
t
.
[
3
0
]
M
u
l
t
i
c
l
a
ss
S
V
M
,
ANN
,
a
n
d
N
a
i
v
e
4
9
3
d
a
t
a
f
r
o
m
t
h
e
c
a
n
c
e
r
g
e
n
o
m
e
a
t
l
a
s
p
r
o
g
r
a
m
(
TC
G
A
)
p
o
r
t
a
l
M
u
l
t
i
c
l
a
ss
S
V
M
=
9
8
%
P
r
o
v
i
d
e
s
a
s
u
mm
a
r
y
o
f
p
r
e
v
i
o
u
s
r
e
se
a
r
c
h
.
D
o
e
s
n
o
t
a
ssess
w
h
e
t
h
e
r
t
h
e
a
s
sesse
d
mac
h
i
n
e
l
e
a
r
n
i
n
g
mo
d
e
l
s
a
r
e
c
l
i
n
i
c
a
l
l
y
a
p
p
l
i
c
a
b
l
e
.
3.
M
E
T
H
O
DO
L
O
G
Y
T
h
e
f
o
l
l
o
w
i
n
g
a
r
e
t
h
e
k
e
y
p
h
a
s
e
s
t
h
at
a
r
e
n
e
c
ess
a
r
y
i
n
t
h
e
d
i
ag
n
o
s
i
s
o
f
o
v
a
r
i
a
n
ca
n
c
e
r
:
p
r
e
-
p
r
o
c
e
s
s
i
n
g
,
f
e
a
t
u
r
e
s
e
l
e
ct
i
o
n
te
c
h
n
i
q
u
e
s
,
m
a
c
h
i
n
e
l
ea
r
n
i
n
g
m
o
d
e
ls
,
a
n
d
m
o
d
e
l
e
x
p
l
a
n
at
i
o
n
p
r
o
c
es
s
,
a
s
i
n
d
i
c
at
e
d
i
n
t
h
e
A
l
g
o
r
i
t
h
m
1
.
As
s
h
o
wn
in
Fig
u
r
e
1
,
p
r
e
-
p
r
o
ce
s
s
in
g
th
e
d
ata
is
th
e
in
itia
l
s
tep
in
th
e
p
r
o
ce
s
s
.
I
t
in
v
o
lv
es
th
e
d
r
o
p
n
a
(
)
m
et
h
o
d
t
o
r
ep
lace
t
h
e
m
is
s
in
g
v
alu
es o
f
th
e
r
e
q
u
ir
e
d
f
ea
tu
r
e
c
o
lu
m
n
f
r
o
m
th
e
d
at
aset.
Fo
llo
win
g
th
e
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
Mu
lti
-
p
h
a
s
e
fea
tu
r
e
s
elec
tio
n
fo
r
d
etec
tio
n
o
f e
p
ith
eli
a
l o
va
r
i
a
n
ca
n
ce
r
u
s
in
g
…
(
S
u
ma
P
a
la
n
i S
u
b
r
a
ma
n
ya
)
4805
p
r
e
-
p
r
o
ce
s
s
in
g
,
th
e
d
ata
is
d
iv
id
ed
in
to
two
s
ec
tio
n
s
as
tes
tin
g
an
d
tr
ain
in
g
.
Usi
n
g
an
ass
o
r
tm
en
t
o
f
f
ea
tu
r
e
s
elec
tio
n
an
d
m
ac
h
in
e
lear
n
i
n
g
tech
n
iq
u
es,
t
h
e
s
u
g
g
ested
m
o
d
el
is
ev
alu
ated
u
s
in
g
s
tan
d
ar
d
m
etr
ics
to
d
eter
m
in
e
th
e
p
r
e
-
o
p
er
ativ
e
s
tate
o
f
o
v
ar
ian
ca
n
ce
r
.
T
o
i
m
p
r
o
v
e
in
ter
p
r
etab
ilit
y
,
t
h
e
c
o
n
s
eq
u
en
ce
o
f
th
e
ch
o
s
en
f
ea
tu
r
es is
d
em
o
n
s
tr
ated
u
s
in
g
an
X
AI
tech
n
iq
u
e.
Alg
o
r
ith
m
1
.
Stru
ctu
r
e
d
m
ac
h
in
e
lear
n
in
g
w
o
r
k
f
lo
w
f
o
r
o
v
ar
ian
ca
n
ce
r
b
io
m
ar
k
er
class
if
icatio
n
with
d
u
al
f
ea
tu
r
e
s
elec
tio
n
s
tr
ateg
ies
1.
Preprocess and load the data using
load_data ('DATASET_1.xlsx')
2.
Set up features and target
X=data.drop ('CA125'); y=(data ['CA125'] >35)
3.
Data split (50% testing, 50% training)
split
(X, y)=X_train, X_test, y_train, y_test
4.
Select features:
i)
Based on correlation (threshold=0.1)
ii)
Based on
the recursive feature elimination (RFE)
5.
Develop and assess models:
i)
Using features chosen by correlation
ii)
On features chosen by RFE
6.
Use SHAP to interpret the optimal model.
7.
Create graphs of related performance metrics
Fig
u
r
e
1
.
Flo
w
d
ia
g
r
am
o
f
th
e
p
r
o
p
o
s
ed
wo
r
k
3
.
1
.
Da
t
a
s
et
Data
b
ases
f
r
o
m
R
am
aiah
Me
d
ical
C
o
lleg
e,
B
an
g
alo
r
e
,
we
r
e
u
s
ed
in
t
h
is
s
tu
d
y
.
I
t
h
as
a
to
tal
o
f
2
8
ch
a
r
ac
ter
is
tics
,
in
clu
d
in
g
t
h
e
tu
m
o
r
m
ar
k
er
C
A
-
1
2
5
.
Se
v
en
m
o
s
t
s
ig
n
if
ica
n
t
f
ea
tu
r
es
ar
e
s
elec
ted
,
s
u
ch
as
ag
e,
tu
m
o
r
later
ality
,
s
ize,
tu
m
o
r
ty
p
e
,
tu
m
o
r
g
r
ad
e
,
an
d
F
I
GO
s
tag
e.
T
h
er
e
ar
e
8
4
ca
s
es
o
f
o
v
ar
ian
ca
n
ce
r
an
d
6
6
ca
s
es o
f
n
o
n
-
o
v
ar
ian
c
an
ce
r
am
o
n
g
th
e
1
5
0
in
d
iv
id
u
als in
th
e
d
ataset.
3.
2
.
P
re
-
pro
ce
s
s
ing
T
h
r
o
u
g
h
o
u
t
th
e
d
ata
ass
ess
m
en
t
p
r
o
ce
s
s
,
we
p
r
eser
v
ed
as
m
u
ch
o
f
th
e
o
r
ig
i
n
al
d
ata
as
p
o
s
s
ib
le
to
g
u
ar
an
tee
th
at
it
co
u
ld
b
e
u
s
ed
co
m
p
letely
.
T
h
e
s
ize
co
lu
m
n
was
m
is
s
in
g
1
0
%
o
f
th
e
v
al
u
e.
T
o
d
ea
l
with
th
e
m
is
s
in
g
v
alu
es,
th
e
m
ed
ian
is
s
p
ec
if
ically
ca
lcu
lated
.
3.
3
.
Sepa
ra
t
ing
t
he
d
a
t
a
s
et
T
h
e
d
ataset
was
co
llected
f
r
o
m
R
am
aiah
Me
d
ical
C
o
lleg
e,
B
an
g
alo
r
e.
I
t
was
d
iv
id
ed
i
n
to
tr
ain
in
g
an
d
test
in
g
p
h
ases
.
Sp
ec
if
ically
,
6
0
% o
f
th
e
d
ata
was u
s
ed
f
o
r
tr
ain
in
g
,
wh
ile
th
e
r
em
ain
in
g
4
0
% wa
s
u
s
ed
f
o
r
test
in
g
.
3.
4
.
F
e
a
t
ure
s
elec
t
io
n t
ec
hn
iqu
es
T
h
e
f
ea
tu
r
e
s
elec
tio
n
m
et
h
o
d
s
ar
e
a
m
o
r
e
u
n
d
er
s
tan
d
a
b
le
an
d
p
r
ac
tical
way
f
o
r
m
o
d
el
cr
ea
tio
n
b
y
tak
in
g
p
er
tin
e
n
t
in
f
o
r
m
atio
n
o
u
t
o
f
th
e
r
aw
d
ata.
T
h
e
o
p
tim
al
f
ea
tu
r
e
tech
n
o
lo
g
y
ca
n
d
if
f
er
b
ec
au
s
e
o
f
t
h
e
s
p
ec
if
ic
d
ataset,
th
e
is
s
u
e
b
ein
g
s
o
lv
ed
,
an
d
th
e
alg
o
r
ith
m
s
e
m
p
lo
y
ed
.
I
t
o
f
te
n
r
e
q
u
ir
es
a
d
ee
p
co
m
p
r
eh
e
n
s
io
n
o
f
th
e
f
ac
ts
an
d
ex
te
n
s
iv
e
d
o
m
ain
ex
p
e
r
tis
e.
T
o
d
eter
m
i
n
e
th
e
ef
f
icac
y
o
f
th
e
f
ea
t
u
r
es
th
at
ar
e
cr
ea
ted
,
iter
ativ
e
f
ea
tu
r
e
en
g
in
ee
r
in
g
n
ec
ess
itate
s
test
in
g
an
d
v
alid
atio
n
.
3.
4
.
1.
Rec
urs
iv
e
f
ea
t
ure
elimin
a
t
io
n t
ec
hn
iqu
e
T
h
e
least
im
p
o
r
tan
t
ch
ar
ac
ter
is
tics
ar
e
p
r
o
g
r
ess
iv
ely
r
em
o
v
ed
u
s
in
g
R
FE
,
a
b
ac
k
war
d
elim
in
atio
n
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
[
3
1
]
.
I
t
p
er
f
o
r
m
s
g
r
ad
in
g
o
f
e
ac
h
f
ea
tu
r
e
ac
co
r
d
i
n
g
to
th
e
wa
y
th
e
m
o
d
el
p
e
r
f
o
r
m
s
.
B
y
p
r
o
g
r
ess
iv
ely
r
em
o
v
in
g
f
e
atu
r
es,
R
FE
r
ed
u
ce
s
p
r
ed
icto
r
d
ep
en
d
e
n
ce
.
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.
1
4
,
No
.
6
,
Dec
em
b
er
2
0
2
5
:
4
8
0
2
-
4
8
1
3
4806
3
.
4
.
2
.
Co
rr
ela
t
io
n
c
o
ef
f
icient
T
h
e
lin
ea
r
r
elatio
n
s
h
ip
a
m
o
n
g
f
ea
tu
r
es
an
d
t
h
e
d
esire
d
v
ar
iab
le
is
m
ea
s
u
r
ed
b
y
th
e
c
o
r
r
elatio
n
co
ef
f
icien
t
[
3
1
]
.
Hig
h
c
o
r
r
elat
io
n
f
ea
tu
r
es
ar
o
u
n
d
±
1
is
d
ee
m
ed
s
ig
n
if
ican
t.
W
h
en
ch
ar
ac
ter
is
tics
h
av
e
lin
ea
r
p
r
ed
ictiv
e
p
o
wer
,
it wo
r
k
s
b
est.
3
.
5
.
M
a
chine
lea
rning
a
nd
ens
em
ble m
o
dels
T
h
e
f
ea
tu
r
es
ch
o
s
en
b
y
f
ea
tu
r
e
s
elec
tio
n
ap
p
r
o
ac
h
es
ar
e
u
s
ed
b
y
m
ac
h
in
e
lear
n
i
n
g
m
o
d
e
ls
,
wh
ich
ar
e
cr
u
cial
f
o
r
d
ec
is
io
n
-
m
ak
in
g
.
Fu
r
th
er
to
im
p
r
o
v
e
th
e
p
e
r
f
o
r
m
an
ce
f
r
o
m
b
ase
m
o
d
els
,
e
n
s
em
b
le
ap
p
r
o
ac
h
es
ar
e
u
s
ed
.
T
h
e
r
e
ar
e
two
s
tag
es
in
th
is
s
ec
tio
n
.
3
.
5
.
1
.
Sta
g
e
1
:
m
a
chine
lea
rning
m
o
dels
T
h
is
m
eth
o
d
allo
ws
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
to
lear
n
an
d
id
en
tify
p
atter
n
s
in
th
e
d
ata
co
llectiv
ely
b
y
p
r
o
v
id
in
g
th
em
with
ac
ce
s
s
to
th
e
tr
ai
n
in
g
d
ata
b
ase
[
3
2
]
.
Var
io
u
s
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
ar
e
em
p
lo
y
ed
in
th
i
s
wo
r
k
.
T
h
ese
in
clu
d
e
SVM
,
KNN
m
o
d
el,
DT
,
an
d
LR
.
3
.
5
.
2
.
Sta
g
e
2
:
ens
em
ble mo
dels
E
n
s
em
b
le
m
o
d
els
s
u
ch
as
v
o
tin
g
class
if
ier
,
s
tak
in
g
,
b
ag
g
in
g
an
d
b
o
o
s
tin
g
ar
e
u
s
ed
to
p
r
o
g
r
ess
th
e
p
er
f
o
r
m
an
ce
o
f
b
ase
m
o
d
els
[
3
2
]
.
W
h
en
a
co
m
p
le
x
d
ataset
is
u
n
av
ailab
le,
m
ac
h
i
n
e
lear
n
in
g
m
o
d
els
s
h
o
u
ld
b
e
u
s
ed
i
n
s
tead
o
f
d
ee
p
lear
n
in
g
m
o
d
els.
Als
o
,
E
n
s
em
b
le
m
o
d
els
ar
e
f
o
u
n
d
to
b
e
b
ett
er
p
er
f
o
r
m
in
g
th
a
n
in
d
iv
id
u
al
b
ase
m
ac
h
in
e
lea
r
n
in
g
m
o
d
els
f
o
r
th
e
a
b
o
v
e
m
en
tio
n
ed
r
ea
s
o
n
.
Fig
u
r
e
2
in
d
icate
s
d
if
f
er
e
n
t
en
s
em
b
le
m
o
d
els
o
b
tain
ed
b
y
co
m
b
i
n
in
g
in
d
ep
en
d
en
t
b
a
s
e
m
ac
h
in
e
lear
n
in
g
m
o
d
els'
p
r
ed
ictio
n
s
.
T
h
e
en
s
em
b
le
m
o
d
els ar
e
elab
o
r
at
ed
in
th
is
s
ec
tio
n
:
‒
Vo
tin
g
class
if
ier
:
v
o
tin
g
class
if
ier
s
[
2
2
]
a
r
e
m
ac
h
i
n
e
lear
n
i
n
g
m
o
d
els
th
at
p
r
e
d
ict
an
o
u
t
p
u
t
class
b
ased
o
n
wh
ich
m
o
d
el
h
as th
e
b
est c
h
an
ce
o
f
p
r
o
d
u
cin
g
th
e
tar
g
et
class
.
‒
B
ag
g
in
g
:
a
m
eta
-
alg
o
r
ith
m
ca
lled
b
o
o
ts
tr
ap
ag
g
r
eg
atin
g
,
s
o
m
etim
es
r
ef
er
r
ed
to
as
b
ag
g
i
n
g
[
2
2
]
,
aim
s
to
in
cr
ea
s
e
th
e
ac
c
u
r
ac
y
an
d
co
n
s
is
ten
cy
o
f
m
ac
h
in
e
lear
n
in
g
tech
n
iq
u
es
u
s
ed
in
an
aly
tical
r
eg
r
ess
io
n
an
d
class
if
icatio
n
.
I
t
r
ed
u
ce
s
v
ar
i
an
ce
an
d
h
elp
s
av
o
i
d
o
v
e
r
f
it
tin
g
.
I
n
ac
co
r
d
a
n
ce
with
(
1
)
an
d
(
2
)
,
DT
tech
n
iq
u
es
ar
e
ty
p
ically
ap
p
li
ed
with
it.
On
e
s
p
ec
ial
ap
p
li
ca
tio
n
o
f
th
e
m
o
d
el
av
e
r
ag
in
g
ap
p
r
o
ac
h
is
b
ag
g
in
g
.
ℎ
=
∑
|
−
|
=
1
(
1
)
=
(
∑
(
−
)
=
1
)
1
(
2
)
‒
Stack
in
g
:
i
n
o
r
d
er
to
i
n
cr
ea
s
e
p
r
ed
ictio
n
p
er
f
o
r
m
an
ce
,
s
tack
i
n
g
is
an
en
s
em
b
le
lear
n
in
g
s
tr
ateg
y
[
2
2
]
in
m
ac
h
in
e
lear
n
in
g
th
at
in
te
g
r
at
es
b
ase
m
o
d
els,
also
k
n
o
wn
a
s
b
ase
lear
n
er
s
.
T
h
e
g
o
al
is
to
cr
ea
te
a
b
etter
o
v
er
all
m
o
d
el
b
y
s
tr
ateg
icall
y
co
m
b
in
i
n
g
th
e
o
u
tp
u
ts
o
f
m
an
y
m
o
d
els
an
d
u
tili
zin
g
t
h
eir
r
esp
ec
tiv
e
s
tr
en
g
th
s
.
‒
B
o
o
s
tin
g
:
XGB
is
an
o
th
er
te
ch
n
iq
u
e
f
o
r
en
h
an
cin
g
m
ac
h
i
n
e
lear
n
in
g
[
2
2
]
.
An
ex
tr
em
e
v
ar
ian
t
o
f
th
e
GB
tech
n
iq
u
e
is
th
e
ex
tr
em
e
GB
alg
o
r
ith
m
,
s
o
m
etim
es k
n
o
wn
as XG
B
.
G
B
an
d
XGB d
if
f
er
p
r
im
ar
ily
in
th
at
th
e
f
o
r
m
er
em
p
l
o
y
s
a
r
eg
u
lar
izatio
n
tech
n
iq
u
e
.
I
t
is
a
r
eg
u
lar
ized
f
o
r
m
o
f
th
e
cu
r
r
en
tly
em
p
lo
y
ed
g
r
ad
ien
t
-
b
o
o
s
tin
g
tech
n
i
q
u
e.
T
h
is
ex
p
lain
s
wh
y
XG
B
o
u
tp
er
f
o
r
m
s
a
tr
ad
itio
n
al
GB
m
eth
o
d
.
Ad
d
itio
n
ally
,
it p
er
f
o
r
m
s
b
etter
in
d
atasets
th
at
co
n
tain
b
o
th
n
u
m
er
ical
an
d
ca
teg
o
r
ical
v
ar
i
ab
les.
Fig
u
r
e
2
.
B
lo
ck
d
iag
r
am
o
f
e
n
s
em
b
le
m
o
d
el
3
.
6
.
P
er
f
o
r
m
a
nce
i
nd
ica
t
o
rs
T
h
e
ef
f
ec
tiv
e
n
ess
o
f
t
h
e
m
ac
h
in
e
lear
n
i
n
g
m
eth
o
d
s
t
h
at
h
a
v
e
b
ee
n
im
p
lem
e
n
ted
is
ev
al
u
ated
u
s
in
g
f
o
u
r
d
if
f
er
e
n
t c
r
iter
ia.
T
h
e
y
ar
e:
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
Mu
lti
-
p
h
a
s
e
fea
tu
r
e
s
elec
tio
n
fo
r
d
etec
tio
n
o
f e
p
ith
eli
a
l o
va
r
i
a
n
ca
n
ce
r
u
s
in
g
…
(
S
u
ma
P
a
la
n
i S
u
b
r
a
ma
n
ya
)
4807
i)
T
r
u
e
p
o
s
itiv
e
(
T
p
)
:
it
is
u
s
ed
t
o
f
o
r
ec
ast th
e
e
v
en
t v
alu
e
m
o
r
e
p
r
ec
is
ely
.
ii)
F
a
l
s
e
p
o
s
i
t
i
v
e
(
F
p
)
:
in
e
s
s
e
n
c
e
,
t
h
i
s
t
e
c
h
n
i
q
u
e
i
s
e
m
p
l
o
y
e
d
t
o
a
s
c
e
r
t
a
i
n
t
h
e
e
r
r
o
n
e
o
u
s
v
a
l
u
e
o
f
a
n
o
c
c
u
r
r
e
n
c
e
.
iii)
T
r
u
e
n
e
g
ativ
e
(
T
n
)
: t
h
e
p
u
r
p
o
s
e
o
f
th
is
m
etr
ic
is
to
p
r
e
d
ict
wh
en
an
e
v
en
t v
al
u
e
will n
o
t o
c
cu
r
.
iv
)
Fals
e
n
eg
ativ
e
(
Fn
)
:
t
h
is
is
u
s
ed
wh
en
th
e
n
o
e
v
en
t v
al
u
e
is
wr
o
n
g
ly
p
r
ed
icted
.
P
r
ec
is
io
n
,
r
ec
all,
F1
-
s
co
r
e
,
a
n
d
ac
cu
r
ac
y
ar
e
th
e
f
o
u
r
p
er
f
o
r
m
an
ce
in
d
icato
r
s
.
‒
Pre
cisi
o
n
:
t
h
e
m
o
d
el'
s
q
u
ality
is
r
ef
er
r
e
d
to
as
p
r
ec
is
io
n
.
I
n
s
im
p
le
ter
m
s
,
th
e
m
o
s
t
g
e
n
u
i
n
ely
p
o
s
itiv
e
o
u
t o
f
all
f
av
o
r
a
b
le
p
r
e
d
ictio
n
s
as p
er
(
3
)
[
2
3
]
.
=
+
(
3
)
‒
R
ec
all:
t
h
e
r
atio
ca
n
b
e
ca
lcu
l
ated
as sh
o
wn
in
(
4
)
[
2
3
]
.
=
+
(
4
)
‒
F1
-
s
c
o
r
e:
e
r
r
o
n
e
o
u
s
p
o
s
i
ti
v
e
an
d
e
r
r
o
n
e
o
u
s
n
eg
ati
v
e
r
es
u
l
t
s
a
r
e
a
ls
o
t
ak
e
n
in
to
co
n
s
i
d
e
r
at
io
n
i
n
t
h
is
p
e
r
f
o
r
m
a
n
c
e.
T
h
e
r
e
f
o
r
e
,
i
t w
o
r
k
s
ef
f
e
cti
v
el
y
w
it
h
b
o
t
h
b
ala
n
c
ed
a
n
d
im
b
a
la
n
c
ed
d
at
a
s
ets
as
p
er
(
5
)
[
2
3
]
.
1
−
s
c
or
e
=
2
(
∗
)
+
(
5
)
‒
Acc
u
r
ac
y
:
t
h
is
is
co
m
p
u
ted
b
y
d
iv
id
in
g
th
e
to
tal
n
u
m
b
er
o
f
s
am
p
les
b
y
th
e
n
u
m
b
er
o
f
ex
am
p
les
th
at
wer
e
co
r
r
ec
tly
id
e
n
tifie
d
.
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
p
r
esen
ce
o
f
o
v
ar
ian
ca
n
c
er
was
d
eter
m
in
ed
b
y
an
al
y
zi
n
g
a
n
u
m
b
er
o
f
m
ed
ical
ch
ar
ac
ter
is
tic
s
th
at
w
er
e
in
clu
d
ed
as
p
ar
t
o
f
t
h
e
d
ataset.
T
h
e
m
o
s
t
im
p
o
r
tan
t
f
ea
tu
r
es
,
s
u
ch
as
a
g
e,
C
A
-
1
2
5
,
tu
m
o
r
later
ality
,
s
ize,
tu
m
o
r
ty
p
e,
g
r
ad
e
o
f
tu
m
o
r
,
an
d
FIG
O
s
tag
e
,
f
o
r
th
e
an
aly
s
is
wer
e
s
elec
ted
.
T
a
b
le
2
g
iv
es
th
e
d
etailed
in
f
o
r
m
atio
n
o
n
th
e
p
atien
ts
in
th
e
d
ataset.
T
ab
le
2
.
Key
i
n
f
o
r
m
atio
n
o
f
th
e
p
atien
ts
in
th
e
d
ataset
P
a
r
a
m
e
t
e
r
s
C
a
t
e
g
o
r
i
e
s
To
t
a
l
n
o
.
of
p
a
t
i
e
n
t
s
(
N
=
1
5
0
)
P
e
r
c
e
n
t
a
g
e
A
g
e
>
5
0
63
42
>
5
0
63
42
C
A
-
1
2
5
>
3
5
U
/
m
L
84
56
<
3
5
U
/
m
L
66
50
S
i
z
e
>
4
.
8
cm
1
3
4
89
<
4
.
8
cm
16
10
Tu
m
o
u
r
l
a
t
e
r
a
l
i
t
y
U
n
i
l
a
t
e
r
a
l
125
83
B
i
l
a
t
e
r
a
l
25
17
Tu
m
o
u
r
t
ype
S
e
r
o
u
s
98
65
E
n
d
o
m
e
t
r
o
i
d
27
18
C
l
e
a
r
c
e
l
l
6
40
G
e
r
m c
e
l
l
0
0
S
e
x
c
o
r
d
B
r
e
n
n
e
r
4
3
M
u
c
i
n
o
u
s
15
10
G
r
a
d
e
o
f
t
u
m
o
u
r
G
r
a
d
e
I
8
5
G
r
a
d
e
I
I
1
1
4
76
G
r
a
d
e
I
I
I
5
3
G
r
a
d
e
I
V
23
15
F
i
g
o
st
a
g
e
g
r
o
u
p
S
t
a
g
e
I
60
40
S
t
a
g
e
I
I
6
4
S
t
a
g
e
I
I
I
82
55
S
t
a
g
e
I
V
2
1
I
n
th
is
s
tu
d
y
,
Py
th
o
n
(
3
.
8
)
is
th
e
co
r
e
p
r
o
g
r
am
m
in
g
lan
g
u
ag
e.
T
h
e
f
o
llo
win
g
Py
th
o
n
lib
r
ar
ies
ar
e
u
s
ed
:
Pan
d
as
f
o
r
lo
ad
in
g
an
d
p
r
o
ce
s
s
in
g
,
Nu
m
Py
to
p
er
f
o
r
m
b
asic
o
p
er
atio
n
s
,
s
cik
it
-
lear
n
f
o
r
m
ac
h
i
n
e
lear
n
in
g
alg
o
r
ith
m
s
,
a
n
d
p
lo
t
s
wer
e
cr
ea
ted
u
s
in
g
Ma
tp
lo
t
lib
.
SciPy
was
u
s
ed
to
ca
lcu
l
ate
p
-
v
alu
es
,
a
n
d
a
Sh
ap
ley
ad
d
itiv
e
e
x
p
lan
atio
n
s
(
SHAP
)
p
lo
t
was
u
tili
ze
d
f
o
r
th
e
m
o
d
el
ex
p
lan
at
io
n
p
r
o
ce
s
s
.
T
h
e
h
y
p
er
p
ar
am
eter
s
u
s
ed
f
o
r
m
ac
h
in
e
lear
n
in
g
m
o
d
els ar
e
d
is
cu
s
s
ed
in
T
ab
le
3
.
T
h
e
s
tatis
tica
l
im
p
o
r
tan
ce
o
f
a
f
ea
tu
r
e'
s
a
s
s
o
ciatio
n
with
th
e
tar
g
et
v
ar
iab
le
is
d
eter
m
in
ed
in
f
ea
tu
r
e
s
elec
tio
n
u
s
in
g
a
p
-
v
alu
e.
T
h
e
clin
ical
f
ea
tu
r
es
ar
e
class
if
ied
as
h
ig
h
ly
s
ig
n
if
ican
t,
m
ar
g
in
a
lly
s
ig
n
if
ican
t,
an
d
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.
1
4
,
No
.
6
,
Dec
em
b
er
2
0
2
5
:
4
8
0
2
-
4
8
1
3
4808
non
-
s
ig
n
if
ica
n
t
b
ased
o
n
th
e
p
-
v
alu
es.
T
h
e
f
o
llo
win
g
f
ea
tu
r
es
ar
e
h
ig
h
ly
s
ig
n
if
ican
t
(
p
<
0
.
0
0
1
)
:
tu
m
o
r
ty
p
e
(
p
=3
.
2
8
e
-
0
7
)
,
tu
m
o
r
later
ality
(
p
=6
.
6
9
e
-
0
8
)
,
g
r
a
d
e
o
f
t
u
m
o
r
(
p
=
1
.
0
5
e
-
0
5
)
,
FIG
OSTA
GE
GR
OU
P_
I
I
I
C
(
p
=1
.
2
1
e
-
2
6
)
.
T
h
e
f
o
llo
win
g
f
ea
tu
r
es
ar
e
m
ar
g
i
n
ally
s
ig
n
if
ican
t
(
0
.
0
1
<p
<0
.
0
5
)
:
s
ize
(
p
=0
.
0
4
0
)
,
FIG
OSTA
GE
G
R
OU
P_
I
I
B
(
p
=0
.
0
1
3
)
.
T
h
e
f
o
llo
win
g
f
ea
tu
r
es
ar
e
least
s
ig
n
if
ican
t
(
p
≥
0
.
0
5
)
:
a
g
e
(
p
=0
.
2
4
2
)
,
FIG
OSTA
GE
G
R
OU
P_
I
C
3
(
p
=0
.
0
8
3
)
.
T
ab
le
3
.
Hy
p
er
p
ar
a
m
eter
s
f
o
r
m
ac
h
in
e
lear
n
in
g
m
o
d
el
M
o
d
e
l
H
y
p
e
r
p
a
r
a
me
t
e
r
s
LR
max
_
i
t
e
r
,
p
e
n
a
l
t
y
,
C
,
s
o
l
v
e
r
S
V
M
k
e
r
n
e
l
,
C
,
g
a
m
ma,
p
r
o
b
a
b
i
l
i
t
y
K
NN
n
_
n
e
i
g
h
b
o
r
s,
w
e
i
g
h
t
s,
m
e
t
r
i
c
DT
max
_
d
e
p
t
h
,
mi
n
_
s
a
m
p
l
e
s_
s
p
l
i
t
,
c
r
i
t
e
r
i
o
n
V
o
t
i
n
g
c
l
a
s
si
f
i
e
r
e
st
i
mat
o
r
s,
v
o
t
i
n
g
B
a
g
g
i
n
g
c
l
a
ssi
f
i
e
r
e
st
i
mat
o
r
,
n
_
e
s
t
i
m
a
t
o
r
s,
ma
x
_
s
a
m
p
l
e
s
,
b
o
o
t
s
t
r
a
p
B
o
o
st
i
n
g
n
_
e
st
i
ma
t
o
r
s
,
l
e
a
r
n
i
n
g
_
r
a
t
e
,
b
a
se
_
e
st
i
mat
o
r
S
t
a
c
k
i
n
g
c
l
a
ss
i
f
i
e
r
e
st
i
mat
o
r
s,
f
i
n
a
l
_
e
st
i
ma
t
o
r
4
.
1
.
Rela
t
io
ns
hip
bet
wee
n t
u
m
o
r
s
t
a
g
e,
s
ize,
la
t
er
a
lit
y
,
a
n
d CA
-
125
T
h
e
m
o
s
t
co
m
m
o
n
l
y
r
ep
o
r
te
d
tu
m
o
r
m
ar
k
er
is
C
A
-
1
2
5
,
wh
ich
h
as
b
ee
n
lin
k
e
d
to
a
n
u
m
b
er
o
f
h
is
to
lo
g
ical
ab
n
o
r
m
alities
th
a
t
m
ay
aid
in
ea
r
ly
d
etec
tio
n
.
Ov
ar
ian
ca
n
ce
r
is
ty
p
ically
d
etec
ted
in
th
e
later
s
tag
es
an
d
is
u
n
d
iag
n
o
s
ed
in
its
ea
r
ly
s
tag
es.
C
A
-
1
2
5
lev
els
g
r
ea
ter
th
an
3
5
U/m
l
ar
e
r
eg
ar
d
ed
to
b
e
elev
ated
.
Fig
u
r
e
3
in
f
er
s
th
at
th
e
p
atien
ts
in
th
e
ad
v
an
ce
d
s
tag
e
h
av
e
s
m
aller
tu
m
o
r
s
an
d
h
ig
h
lev
el
s
o
f
C
A
-
125.
T
h
is
f
ac
t w
o
u
ld
b
e
b
en
e
f
icial
to
th
e
d
o
cto
r
s
to
c
h
ec
k
th
e
m
etastas
i
s
co
n
d
itio
n
.
Als
o
,
t
h
e
n
e
x
t
p
ar
am
ete
r
,
b
ila
t
er
al
E
OC
,
h
a
d
a
wo
r
s
e
p
r
o
g
n
o
s
is
t
h
an
u
n
il
at
er
al
E
OC
.
A
d
d
it
io
n
al
ly
,
as
ill
u
s
tr
at
ed
i
n
Fi
g
u
r
e
4
,
w
e
ca
n
s
e
e
t
h
at
C
A
-
1
2
5
is
h
i
g
h
f
o
r
t
h
e
b
i
lat
er
al
E
OC
.
M
o
r
e
o
v
er
,
s
er
o
u
s
an
d
en
d
o
m
et
r
o
id
ca
r
ci
n
o
m
as
c
o
m
p
r
is
e
th
e
b
u
lk
o
f
t
h
e
p
o
p
u
l
ati
o
n
.
C
o
m
p
ar
ed
t
o
e
n
d
o
m
etr
io
i
d
c
ar
ci
n
o
m
a,
s
e
r
o
u
s
ca
r
ci
n
o
m
a
t
y
p
i
ca
l
ly
e
x
h
i
b
its
g
r
e
ate
r
ag
g
r
ess
io
n
a
n
d
a
p
o
o
r
e
r
p
r
o
g
n
o
s
is
.
S
u
r
v
iv
al
r
a
tes
a
r
e
g
r
ea
tl
y
i
m
p
ac
te
d
b
y
th
e
f
ac
t
th
at
it
is
ty
p
i
ca
l
ly
i
d
e
n
t
if
i
ed
a
t st
a
g
es
th
at
a
r
e
m
o
r
e
a
d
v
a
n
c
e
d
.
T
h
e
r
e
f
o
r
e,
a
g
r
ea
te
r
C
A
-
1
2
5
le
v
el
is
s
ee
n
f
o
r
s
e
r
o
u
s
ca
r
ci
n
o
m
a,
w
h
i
c
h
i
s
i
n
d
ic
ate
d
as
t
y
p
e
1
an
d
t
y
p
e
2
in
d
i
ca
tes
s
e
r
o
u
s
a
n
d
e
n
d
o
m
etr
o
i
d
c
ar
ci
n
o
m
a
as
p
e
r
F
ig
u
r
e
5
.
T
h
e
ess
en
tia
l
f
e
atu
r
es
a
r
e
v
it
al
f
o
r
p
r
ec
is
e
cl
a
s
s
if
i
ca
t
io
n
a
n
d
p
r
e
d
ic
ti
o
n
.
T
h
e
r
es
u
lts
o
f
v
a
r
i
o
u
s
class
i
f
i
er
s
,
s
u
c
h
as
L
R
,
D
T
,
S
VM
,
K
NN,
s
ta
ck
in
g
,
b
a
g
g
in
g
,
an
d
v
o
ti
n
g
cl
ass
if
ie
r
s
,
we
r
e
u
s
ed
f
o
r
th
e
p
r
ed
icti
o
n
o
f
o
v
a
r
ia
n
ca
n
ce
r
.
T
h
e
r
es
u
lts
o
f
v
a
r
i
o
u
s
class
if
i
e
r
s
ar
e
d
is
cu
s
s
ed
i
n
s
ec
t
io
n
s
4
.
2
a
n
d
4
.
3
.
Fig
u
r
e
3
.
I
n
f
lu
e
n
ce
o
f
h
ig
h
lev
el
o
f
C
A
-
1
2
5
o
n
tu
m
o
r
s
ize
Fig
u
r
e
4
.
I
n
v
esti
g
atio
n
o
f
later
ality
with
C
A
-
125
Fig
u
r
e
5
.
Ass
o
ciatio
n
o
f
tu
m
o
r
ty
p
e
with
C
A
-
1
2
5
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
Mu
lti
-
p
h
a
s
e
fea
tu
r
e
s
elec
tio
n
fo
r
d
etec
tio
n
o
f e
p
ith
eli
a
l o
va
r
i
a
n
ca
n
ce
r
u
s
in
g
…
(
S
u
ma
P
a
la
n
i S
u
b
r
a
ma
n
ya
)
4809
4.
2
.
B
a
s
e
mo
dels
Acc
u
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
F
1
-
s
co
r
e,
s
en
s
itiv
ity
,
an
d
s
p
ec
if
icity
wer
e
th
e
p
r
im
ar
y
m
ea
s
u
r
es
u
s
ed
in
th
e
p
er
f
o
r
m
an
ce
ev
al
u
atio
n
.
T
h
e
f
o
llo
win
g
ar
e
t
h
e
p
er
f
o
r
m
a
n
ce
r
esu
lts
th
at
wer
e
attain
ed
.
T
ab
le
4
lis
ts
all
th
e
ev
alu
atio
n
m
etr
ics,
it
is
o
b
s
er
v
ed
th
at
th
e
L
R
f
r
o
m
th
e
R
FE
tech
n
iq
u
e
o
u
tp
e
r
f
o
r
m
s
th
e
o
t
h
er
f
ea
tu
r
e
s
elec
tio
n
alg
o
r
ith
m
s
in
ter
m
s
o
f
ac
cu
r
a
cy
o
f
9
6
%
an
d
o
th
er
p
er
f
o
r
m
an
ce
p
a
r
am
eter
s
.
Acc
u
r
ac
y
,
w
h
ich
is
m
ea
s
u
r
ed
as
th
e
p
er
ce
n
tag
e
o
f
co
r
r
ec
tly
ca
t
eg
o
r
ized
ca
s
es
in
a
d
ataset,
was
th
e
p
r
im
ar
y
e
v
alu
atio
n
m
etr
ic
em
p
lo
y
ed
i
n
th
is
r
ev
iew
p
r
o
ce
d
u
r
e.
T
ab
le
4
.
E
v
alu
atio
n
m
et
r
ics o
f
b
ase
m
o
d
els
F
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
t
e
c
h
n
i
q
u
e
B
a
se
m
o
d
e
l
s
A
c
c
u
r
a
c
y
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
s
c
o
r
e
S
e
n
s
i
t
i
v
i
t
y
S
p
e
c
i
f
i
c
i
t
y
R
F
E
t
e
c
h
n
i
q
u
e
S
V
M
92
1
0
0
85
9
2
.
1
8
5
.
8
3
1
0
0
K
NN
7
8
.
6
9
4
.
2
6
7
.
5
7
8
.
1
7
6
7
.
5
9
3
.
8
1
DT
9
4
.
6
9
5
.
2
8
9
3
.
0
6
9
5
.
2
8
9
5
.
2
8
9
3
.
8
1
LR
96
1
0
0
9
3
.
0
6
9
6
.
3
1
9
3
.
0
6
1
0
0
C
o
r
r
e
l
a
t
i
o
n
c
o
e
f
f
i
c
i
e
n
t
S
V
M
88
95
83
8
8
.
7
8
3
.
6
9
3
.
8
1
K
NN
92
93
93
9
2
.
9
9
3
.
0
6
9
0
.
9
5
DT
9
3
.
3
3
9
5
.
5
6
9
3
.
0
6
9
4
.
1
2
9
3
.
0
6
9
4
.
2
9
LR
88
9
5
.
2
8
8
3
.
6
1
8
8
.
7
4
8
3
.
6
1
9
3
.
8
1
4.
3
.
E
ns
em
ble
m
o
dels
T
h
er
e
is
a
n
o
ticea
b
le
im
p
r
o
v
em
en
t
i
n
ac
cu
r
ac
y
,
p
r
ec
i
s
io
n
,
r
ec
all,
F1
-
s
co
r
e,
s
en
s
i
tiv
ity
,
an
d
s
p
ec
if
icity
wh
en
co
m
p
a
r
ed
to
th
e
b
ase
m
o
d
els.
No
tab
ly
,
th
e
b
o
o
s
tin
g
m
o
d
el
s
h
o
ws
th
e
h
ig
h
est
ac
cu
r
ac
y
o
f
9
6
%
f
r
o
m
th
e
R
FE
tech
n
iq
u
e
.
T
ab
le
5
s
h
o
ws
all
th
e
ev
alu
atio
n
m
etr
ics
s
co
r
es
o
f
th
e
en
s
em
b
le
m
o
d
el.
T
h
e
aim
o
f
t
h
is
in
v
esti
g
atio
n
is
to
d
eter
m
in
e
th
e
o
p
tim
al
an
d
m
o
s
t
ef
f
icien
t
ap
p
r
o
ac
h
f
o
r
th
e
d
etec
tio
n
o
f
o
v
a
r
ian
ca
n
ce
r
tu
m
o
r
s
,
as
well
as
th
e
elem
en
ts
th
at
co
n
tr
i
b
u
te
to
t
h
e
s
u
p
er
io
r
p
er
f
o
r
m
a
n
ce
o
f
a
s
p
ec
if
ic
en
s
em
b
le
lear
n
in
g
s
tr
ateg
y
o
v
er
o
th
er
s
.
AI
-
b
ase
d
m
a
ch
in
e
l
ea
r
n
i
n
g
al
g
o
r
i
th
m
s
ar
e
em
p
l
o
y
e
d
f
o
r
t
h
e
p
r
o
g
n
o
s
ti
c
a
n
d
d
i
ag
n
o
s
ti
c
ass
ess
m
e
n
t
o
f
ca
n
ce
r
s
.
Fi
g
u
r
e
6
p
r
ese
n
ts
th
e
co
m
p
a
r
at
iv
e
p
e
r
f
o
r
m
an
ce
o
f
d
i
f
f
er
e
n
t
m
o
d
e
ls
u
s
ed
in
th
is
s
tu
d
y
.
As
i
ll
u
s
t
r
at
ed
in
Fig
u
r
e
6
(
a
)
,
b
as
e
m
o
d
els
h
a
v
e
a
s
e
n
s
i
ti
v
it
y
o
f
9
3
.
0
6
%
a
n
d
s
p
ec
i
f
ic
it
y
o
f
1
0
0
%,
w
h
il
e
F
ig
u
r
e
6
(
b
)
d
em
o
n
s
t
r
at
es
th
a
t
en
s
em
b
l
e
m
o
d
els
h
a
v
e
a
s
en
s
i
ti
v
it
y
o
f
9
7
.
7
%
a
n
d
s
p
ec
i
f
i
cit
y
o
f
9
3
.
8
5
%
wit
h
a
n
ac
cu
r
a
cy
o
f
9
6
%
.
T
ab
le
5
.
Acc
u
r
ac
y
o
f
en
s
em
b
l
e
m
o
d
els
F
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
t
e
c
h
n
i
q
u
e
B
a
se
m
o
d
e
l
s
A
c
c
u
r
a
c
y
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
sc
o
r
e
S
e
n
s
i
t
i
v
i
t
y
S
p
e
c
i
f
i
c
i
t
y
R
F
E
t
e
c
h
n
i
q
u
e
V
o
t
i
n
g
c
l
a
s
si
f
i
e
r
93
1
0
0
88
9
3
.
4
6
88
1
0
0
B
a
g
g
i
n
g
9
4
.
6
9
7
.
7
9
3
.
0
6
9
5
.
2
7
9
3
.
0
6
9
6
.
6
7
B
o
o
st
i
n
g
96
9
5
.
5
6
9
7
.
7
9
6
.
6
98
94
S
t
a
c
k
i
n
g
9
3
.
3
9
5
.
2
9
3
.
0
6
9
4
.
1
9
3
.
0
6
9
3
.
8
1
C
o
r
r
e
l
a
t
i
o
n
c
o
e
f
f
i
c
i
e
n
t
V
o
t
i
n
g
c
l
a
s
si
f
i
e
r
88
9
5
.
2
8
3
.
6
8
8
.
7
8
3
.
6
9
3
.
8
B
a
g
g
i
n
g
8
9
.
3
9
5
.
2
8
8
.
6
9
1
.
7
8
8
.
6
9
0
.
9
B
o
o
st
i
n
g
9
3
.
3
9
5
.
5
9
3
.
0
6
9
4
.
1
2
9
3
.
0
6
9
4
.
2
9
S
t
a
c
k
i
n
g
92
9
3
.
3
9
3
.
0
6
9
3
.
0
7
9
3
.
0
6
9
0
.
9
(
a)
(
b
)
Fig
u
r
e
6
.
Per
f
o
r
m
an
c
e
e
v
alu
at
io
n
f
o
r
R
FE
tech
n
iq
u
es
of
(
a)
b
ase
m
o
d
els an
d
(
b
)
e
n
s
em
b
le
m
o
d
els
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.
1
4
,
No
.
6
,
Dec
em
b
er
2
0
2
5
:
4
8
0
2
-
4
8
1
3
4810
4.
4
.
M
o
del int
er
pret
a
t
io
ns
XAI
f
ac
ilit
ates
th
e
v
is
u
aliza
ti
o
n
o
f
wh
eth
er
a
ce
r
tain
f
ea
t
u
r
e
is
lin
k
ed
to
a
m
o
d
el'
s
p
r
ed
ictio
n
s
.
T
h
e
m
o
d
el
p
r
o
jectio
n
s
ar
e
d
is
p
lay
ed
o
n
t
h
e
y
-
a
x
is
,
wh
ile
th
e
v
a
lu
e
d
is
tr
ib
u
tio
n
o
f
th
e
f
ea
tu
r
e
is
d
is
p
lay
ed
o
n
t
h
e
x
-
ax
is
.
B
ased
o
n
th
e
SHAP
p
lo
t,
p
o
s
itiv
e
v
alu
es
r
ep
r
esen
t
h
ig
h
er
p
r
e
d
ictio
n
r
is
k
,
wh
er
ea
s
n
eg
ativ
e
v
al
u
e
in
d
icate
s
a
d
ec
r
ea
s
e
in
th
e
p
r
e
d
ictio
n
r
is
k
.
T
h
e
SHAP
p
lo
t,
as
s
h
o
wn
in
Fig
u
r
e
7
,
p
r
o
v
id
es a
n
o
v
er
v
iew
o
f
th
e
f
ea
tu
r
es
in
f
lu
e
n
cin
g
th
e
m
o
d
e
l’
s
p
r
ed
ictio
n
.
F
ig
u
r
e
s
7
(
a)
a
n
d
7
(
b
)
in
d
icate
th
at
t
h
e
tu
m
o
r
s
ize,
later
ality
,
an
d
FIG
O
th
ir
d
s
tag
e
(
I
I
I
C
)
h
av
e
th
e
g
r
ea
test
in
f
lu
en
ce
o
n
t
h
e
p
r
ed
ictio
n
a
n
d
ar
e
m
o
s
t
s
ig
n
if
ican
t
in
r
elatio
n
to
CA
-
1
2
5
.
T
h
e
m
o
d
el'
s
p
r
ed
icted
ac
cu
r
ac
y
will
b
e
d
eg
r
ad
ed
t
o
a
g
r
ea
ter
o
r
less
er
d
e
g
r
ee
b
a
s
ed
o
n
a
g
e
to
war
d
s
th
e
lef
t.
(
a)
(
b
)
Fig
u
r
e
7
.
SHAP p
lo
t f
o
r
(
a)
co
r
r
elatio
n
co
ef
f
icien
t a
n
d
(
b
)
R
FE
4.
5
.
Co
m
pa
ra
t
iv
e
a
na
ly
s
is
wit
h t
he
ex
is
t
ing
re
s
ults
T
ab
le
6
p
r
esen
ts
a
co
m
p
a
r
is
o
n
b
etwe
en
th
e
p
r
o
p
o
s
ed
s
tr
ateg
y
f
o
r
m
e
d
ical
d
ataset
d
iag
n
o
s
is
an
d
p
r
ev
io
u
s
in
v
esti
g
atio
n
s
.
T
h
e
r
esu
lts
s
h
o
w
th
at
th
e
p
r
o
p
o
s
ed
m
eth
o
d
o
u
tp
er
f
o
r
m
s
ea
r
lier
s
tu
d
ies.
T
h
e
ac
cu
r
ac
y
o
f
p
r
e
d
ictio
n
is
th
e
h
ig
h
est f
o
r
th
e
p
r
o
p
o
s
ed
wo
r
k
,
r
ea
c
h
in
g
9
6
%.
T
ab
le
6
.
C
o
m
p
a
r
is
o
n
with
ex
i
s
tin
g
r
esu
lts
Ref
N
o
.
of
s
e
l
e
c
t
e
d
f
e
a
t
u
r
e
s
M
e
t
h
o
d
s
A
c
c
u
r
a
c
y
[
2
5
]
8
f
e
a
t
u
r
e
s
(
r
e
t
r
o
s
p
e
c
t
i
v
e
s
t
u
d
y
)
G
B
M
,
S
V
M
,
R
F
,
C
R
F
,
N
B
,
n
e
u
r
a
l
n
e
t
w
o
r
k
,
a
n
d
El
a
s
t
i
c
N
et
H
i
g
h
e
s
t
a
c
c
u
r
a
c
y
w
a
s
o
b
t
a
i
n
e
d
f
r
o
m
RF
w
i
t
h
6
9
%
[
1
4
]
To
p
18
f
e
a
t
u
r
e
s
w
e
r
e
s
e
l
e
c
t
e
d
f
r
o
m
K
a
g
g
l
e
d
a
t
a
b
a
s
e
K
N
N
,
S
V
M
,
DT
,
a
n
d
e
n
s
e
m
b
l
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
su
c
h
as
ma
x
v
o
t
i
n
g
,
b
o
o
st
i
n
g
,
b
a
g
g
i
n
g
,
a
n
d
s
t
a
c
k
i
n
g
.
H
i
g
h
e
s
t
a
c
c
u
r
a
c
y
w
i
t
h
b
a
s
e
mo
d
e
l
-
8
5
%
a
n
d
e
n
sem
b
l
e
m
o
d
e
l
–
89%
P
r
o
p
o
s
e
d
W
o
r
k
To
p
7
f
e
a
t
u
r
e
s
w
e
r
e
s
e
l
e
c
t
e
d
(
R
a
m
a
i
a
h
M
e
d
i
c
a
l
C
o
l
l
e
g
e
)
K
N
N
,
S
V
M
,
a
n
d
DT
a
l
o
n
g
w
i
t
h
e
n
se
mb
l
e
m
o
d
e
l
s
su
c
h
as
ma
x
v
o
t
i
n
g
,
b
o
o
s
t
i
n
g
,
b
a
g
g
i
n
g
,
a
n
d
s
t
a
c
k
i
n
g
H
i
g
h
e
s
t
a
c
c
u
r
a
c
y
w
i
t
h
b
a
s
e
mo
d
e
l
-
9
6
%
a
n
d
e
n
sem
b
l
e
mo
d
e
l
–
9
6
%
4.
6
.
Dis
cus
s
io
n
On
e
o
f
th
e
m
o
s
t
d
if
f
ic
u
lt
p
at
h
o
lo
g
ical
t
y
p
es
is
E
OC
;
tr
ea
tm
en
t
ch
o
ices
d
iv
er
g
e
d
e
p
en
d
in
g
o
n
th
e
ty
p
e
o
f
tu
m
o
r
.
B
u
t
in
p
r
ac
tic
e,
f
ailin
g
to
f
o
r
ec
ast
th
e
ty
p
e
o
f
tu
m
o
r
o
f
ten
lea
d
s
to
eith
e
r
a
p
o
o
r
ly
p
lan
n
e
d
o
p
er
atio
n
o
r
a
m
is
d
iag
n
o
s
is
o
f
th
e
p
at
h
o
lo
g
ical
ty
p
e,
b
o
t
h
o
f
wh
ic
h
ca
n
h
av
e
a
s
u
b
s
t
an
tial
im
p
ac
t
o
n
th
e
p
atien
t.
Ov
ar
ian
m
ali
g
n
an
cies
m
u
s
t th
er
ef
o
r
e
b
e
d
ia
g
n
o
s
ed
a
s
s
o
o
n
as p
o
s
s
ib
le.
T
h
is
s
tu
d
y
in
ten
d
s
to
d
eter
m
in
e
th
e
b
est
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
e
f
o
r
p
r
e
d
ictin
g
th
e
p
r
eo
p
er
ativ
e
d
iag
n
o
s
is
o
f
o
v
ar
ian
ca
n
ce
r
,
b
y
m
e
r
g
in
g
it
with
m
ac
h
i
n
e
lear
n
in
g
alg
o
r
ith
m
s
.
Am
o
n
g
th
e
m
an
y
f
ac
ets
o
f
o
v
ar
ian
ca
n
ce
r
ar
e
s
ev
e
r
al
h
is
to
ty
p
es
with
v
ar
y
in
g
g
r
ad
es
a
n
d
clin
ical
s
tag
es.
T
h
er
ef
o
r
e,
a
cr
u
cial
tactic
f
o
r
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
Mu
lti
-
p
h
a
s
e
fea
tu
r
e
s
elec
tio
n
fo
r
d
etec
tio
n
o
f e
p
ith
eli
a
l o
va
r
i
a
n
ca
n
ce
r
u
s
in
g
…
(
S
u
ma
P
a
la
n
i S
u
b
r
a
ma
n
ya
)
4811
d
eliv
er
in
g
in
d
iv
id
u
alize
d
,
o
p
ti
m
al
h
ea
lth
ca
r
e
is
p
r
ed
ictin
g
t
h
e
clin
ical
ch
a
r
ac
ter
is
tics
o
f
o
v
ar
ian
ca
n
ce
r
u
s
in
g
p
r
eo
p
e
r
ativ
e
d
ata
an
d
ca
teg
o
r
i
zin
g
p
atien
ts
b
ased
o
n
p
r
o
g
n
o
s
is
.
Acc
o
r
d
in
g
to
T
ab
le
6
,
t
h
e
m
aj
o
r
ity
o
f
r
esear
ch
was
co
n
d
u
cted
u
tili
zin
g
m
ac
h
in
e
lear
n
in
g
a
p
p
r
o
ac
h
es
to
p
er
f
o
r
m
class
if
icatio
n
o
n
b
i
o
m
ar
k
er
s
alo
n
e,
s
u
ch
as
ag
e,
C
A
-
1
2
5
,
m
en
o
p
au
s
e,
HE
-
4
,
a
n
d
NE
U
[
1
4
]
,
[
2
5
]
.
I
n
ad
d
itio
n
to
class
if
icatio
n
,
d
eter
m
in
in
g
th
e
p
r
eo
p
er
ativ
e
s
tatu
s
is
th
e
p
r
im
ar
y
g
o
al,
wh
ic
h
is
co
n
tin
g
en
t u
p
o
n
th
e
s
ev
er
al
f
o
r
m
s
o
f
E
OC
,
in
clu
d
in
g
s
er
o
u
s
,
en
d
o
m
etr
o
i
d
,
clea
r
ce
ll,
an
d
m
u
cin
o
u
s
ca
r
ci
n
o
m
a.
T
h
ese
f
o
r
m
s
also
im
p
ac
t
in
p
r
o
g
n
o
s
is
as
well.
So
,
C
A
-
1
2
5
is
co
n
s
id
er
ed
th
e
g
o
ld
tu
m
o
r
m
a
r
k
er
,
wh
ic
h
h
as
d
ep
en
d
en
cies
o
n
v
a
r
io
u
s
o
th
e
r
clin
ical
f
ac
t
o
r
s
[
2
2
]
,
[
2
3
]
.
T
h
er
e
f
o
r
e,
in
t
h
is
wo
r
k
,
C
A
-
1
2
5
r
ea
d
in
g
s
a
r
e
co
n
s
id
er
e
d
alo
n
g
with
s
ev
er
al
o
th
er
f
ea
t
u
r
es,
s
u
ch
as tu
m
o
r
s
ize,
later
ality
,
FI
GO
s
tag
e,
an
d
tu
m
o
r
ty
p
e.
T
h
e
d
ataset
u
s
ed
in
th
is
s
tu
d
y
m
ak
es
it
clea
r
th
at
p
atien
ts
in
s
tag
es
o
f
ca
n
ce
r
I
I
,
I
I
I
,
a
n
d
I
V
h
ad
s
m
aller
tu
m
o
r
s
an
d
s
ig
n
if
ica
n
tly
h
ig
h
e
r
C
A
-
1
2
5
lev
els
th
an
th
o
s
e
in
s
tag
e
I
,
wh
ich
i
s
in
th
e
m
etastati
c
co
n
d
itio
n
.
T
h
is
in
f
o
r
m
atio
n
o
n
s
tag
in
g
ass
is
t
s
th
e
d
o
cto
r
s
in
d
ec
id
in
g
f
u
r
th
er
tr
ea
tm
e
n
t
,
s
u
ch
as
s
u
r
g
er
y
,
ch
em
o
th
er
a
p
y
,
a
n
d
o
th
er
tr
ea
tm
en
ts
,
wh
ich
also
h
el
p
s
to
an
aly
ze
th
e
a
g
g
r
ess
iv
en
ess
o
f
th
e
ca
n
ce
r
.
T
h
ese
f
in
d
in
g
s
f
u
r
th
er
em
p
h
asize
th
e
n
ee
d
to
co
n
s
id
er
b
o
th
t
h
e
b
ilater
al
an
d
u
n
ilater
al
elem
en
t
s
,
with
b
ilater
ality
b
ein
g
m
o
r
e
c
o
m
m
o
n
ly
s
ee
n
i
n
th
o
s
e
wh
o
ar
e
at
th
e
ev
o
lv
e
d
s
tag
e.
T
h
is
d
ata
ab
o
u
t
th
e
o
r
ig
in
o
f
th
e
tu
m
o
r
an
d
its
p
o
ten
tial
f
o
r
s
p
r
ea
d
ca
n
b
e
in
f
er
r
ed
f
r
o
m
its
s
ize
an
d
later
ality
,
o
r
wh
eth
er
it
is
o
n
eith
er
o
r
b
o
th
o
f
th
e
o
v
ar
ies.
Kn
o
win
g
th
e
ty
p
e
o
f
t
u
m
o
r
,
aid
s
p
h
y
s
ician
s
in
d
eter
m
in
in
g
th
e
b
est co
u
r
s
e
o
f
th
er
ap
y
an
d
f
o
r
ec
asti
n
g
th
e
p
atien
t'
s
p
r
o
g
n
o
s
is
.
T
h
e
d
ep
en
d
en
cies
o
f
th
e
ab
o
v
e
p
a
r
am
eter
s
with
C
A
-
1
2
5
r
ea
d
in
g
s
ar
e
ev
id
en
t
f
r
o
m
Fig
u
r
e
s
3
to
5.
T
h
er
ef
o
r
e,
u
s
in
g
e
f
f
ec
tiv
e
f
ea
t
u
r
e
s
elec
tio
n
p
r
o
ce
d
u
r
es,
th
is
wo
r
k
tak
es
in
to
ac
c
o
u
n
t
e
v
er
y
p
o
ten
tial
p
ar
am
eter
f
o
r
an
e
f
f
ec
tiv
e
d
iag
n
o
s
is
an
d
p
r
o
g
n
o
s
is
o
f
o
v
ar
ian
ca
n
ce
r
,
in
clu
d
in
g
a
g
e,
C
A
-
1
2
5
,
tu
m
o
r
later
ality
,
s
ize,
tu
m
o
r
ty
p
e,
g
r
ad
e
o
f
tu
m
o
r
,
an
d
FIG
O
s
tag
e.
T
h
e
b
est
ac
cu
r
ac
y
o
f
9
6
%
with
s
en
s
itiv
ity
9
3
%
an
d
s
p
ec
if
icity
1
0
0
%
f
o
r
LR
an
d
9
6
%
with
s
en
s
itiv
ity
9
8
%
an
d
s
p
ec
if
icity
9
4
%
f
o
r
b
o
o
s
tin
g
class
if
ier
s
u
n
d
er
th
e
R
FE
tech
n
iq
u
e
wer
e
p
r
o
jecte
d
b
y
e
n
s
em
b
le
class
if
ier
s
in
co
n
ju
n
ctio
n
with
b
as
e
m
ac
h
in
e
lear
n
i
n
g
class
if
ier
s
.
As
p
er
T
ab
le
6
,
th
is
s
tu
d
y
ac
h
iev
es
th
e
h
ig
h
est
ac
cu
r
ac
y
o
f
9
6
%
in
p
r
ed
ictin
g
th
e
p
r
e
-
o
p
er
ativ
e
an
aly
s
is
o
f
o
v
a
r
ian
ca
n
ce
r
co
m
p
ar
ed
t
o
o
th
e
r
p
r
ev
io
u
s
s
tu
d
ies.
T
h
ese
r
esu
lts
,
h
o
wev
er
,
i
m
p
ly
th
at
AI
co
u
l
d
o
f
f
er
u
s
ef
u
l
p
r
e
o
p
er
ativ
e
b
io
m
ar
k
er
b
ased
d
iag
n
o
s
tic
d
ata,
en
ab
lin
g
a
cu
s
to
m
ized
tr
ea
tm
en
t
p
lan
p
r
io
r
to
t
h
e
m
ain
clin
ical
ap
p
r
o
ac
h
in
E
OC
.
5.
CO
NCLU
SI
O
N
AND
F
U
T
U
RE
WO
RK
E
ar
ly
d
etec
tio
n
ca
n
im
p
r
o
v
e
th
e
p
r
o
g
n
o
s
is
o
f
o
v
ar
ian
ca
n
ce
r
,
an
ag
g
r
ess
iv
e
co
n
d
itio
n
.
T
h
er
ef
o
r
e,
in
o
r
d
er
to
d
eliv
e
r
p
e
r
s
o
n
alize
d
,
o
p
tim
al
h
ea
lth
c
ar
e,
it
is
ess
en
tial
to
u
s
e
p
r
eo
p
e
r
ativ
e
d
at
a
to
an
ticip
ate
t
h
e
clin
ical
ch
ar
ac
ter
is
tics
o
f
o
v
a
r
ian
ca
n
ce
r
a
n
d
t
o
class
if
y
p
ati
en
ts
ac
co
r
d
in
g
to
th
ei
r
p
r
o
g
n
o
s
is
.
T
h
is
r
esear
ch
u
s
es
r
ea
l
-
tim
e
d
ata
f
r
o
m
R
a
m
aiah
Me
d
ical
C
o
lleg
e
in
B
an
g
alo
r
e
to
i
n
v
esti
g
ate
th
e
a
p
p
licatio
n
o
f
p
r
e
-
o
p
er
ativ
e
an
aly
s
is
in
t
h
e
s
u
cc
ess
f
u
l
d
etec
tio
n
o
f
o
v
ar
ia
n
ca
n
ce
r
.
T
h
e
r
e
ar
e
1
5
0
p
ati
en
t
r
ec
o
r
d
s
in
th
e
d
atab
ases
,
an
d
ea
ch
o
n
e
h
as
2
8
f
ea
tu
r
es.
T
h
e
d
ataset
co
n
s
is
ts
o
f
6
6
r
ec
o
r
d
s
f
o
r
n
o
n
-
ca
n
ce
r
o
u
s
p
atien
ts
an
d
8
4
r
ec
o
r
d
s
f
o
r
ca
n
ce
r
o
u
s
p
atie
n
ts
.
Sev
en
m
o
s
t
s
ig
n
if
ican
t
ch
ar
ac
ter
is
tics
f
r
o
m
th
e
d
ataset
a
g
e,
C
A
-
1
2
5
,
t
u
m
o
r
later
ality
,
s
ize,
tu
m
o
r
ty
p
e,
g
r
a
d
e
o
f
tu
m
o
r
,
an
d
FIG
O
s
tag
e
wer
e
ch
o
s
en
u
s
in
g
two
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
s
:
R
FE
an
d
co
r
r
elatio
n
co
ef
f
ici
en
t
ap
p
r
o
ac
h
.
T
o
g
et
th
e
b
e
s
t
r
esu
lt
in
f
o
r
ec
asti
n
g
th
e
p
r
e
-
o
p
er
ativ
e
s
tate,
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
ar
e
u
s
ed
in
co
n
ju
n
ctio
n
with
en
s
em
b
le
m
o
d
els
,
in
clu
d
in
g
v
o
tin
g
class
if
ier
s
,
s
ta
c
k
in
g
,
b
a
g
g
in
g
,
an
d
b
o
o
s
tin
g
.
Ou
r
test
f
in
d
i
n
g
s
s
h
o
w
t
h
a
t,
f
o
r
LR
,
we
o
b
tain
ed
9
6
%
ac
cu
r
ac
y
f
o
r
th
e
b
ase
m
o
d
el
an
d
9
6
%
ac
cu
r
ac
y
f
o
r
t
h
e
b
o
o
s
tin
g
en
s
em
b
le
f
o
r
th
e
R
FE
f
ea
tu
r
e
s
elec
tio
n
tech
n
iq
u
e.
W
ith
th
e
aid
o
f
X
AI
,
th
is
an
aly
s
is
o
f
f
er
s
v
alu
ab
le
in
s
ig
h
ts
in
to
f
ea
tu
r
e
s
ele
ctio
n
,
m
o
d
el
co
m
p
lex
ity
,
an
d
ac
cu
r
ac
y
,
o
f
f
e
r
in
g
g
u
id
an
ce
f
o
r
e
n
h
an
cin
g
m
ac
h
in
e
lear
n
in
g
m
eth
o
d
s
u
tili
ze
d
in
ca
n
ce
r
o
u
tco
m
e
p
r
ed
ictio
n
d
ec
is
io
n
-
m
a
k
in
g
.
Sin
ce
h
is
to
lo
g
y
is
r
eg
ar
d
ed
as
th
e
g
o
ld
s
tan
d
a
r
d
in
t
h
e
d
iag
n
o
s
is
o
f
o
v
ar
ia
n
ca
n
ce
r
,
th
is
wo
r
k
ca
n
b
e
f
u
r
th
er
en
h
an
ce
d
b
y
in
teg
r
atin
g
h
is
to
p
ath
o
lo
g
y
im
ag
es
with
b
io
m
a
r
k
er
s
f
o
r
th
e
s
am
e
in
d
iv
id
u
al
s
.
Ad
d
itio
n
ally
,
to
im
p
r
o
v
e
th
e
class
if
icatio
n
ac
cu
r
ac
y
,
a
b
esp
o
k
e
m
o
d
el
ca
n
b
e
cr
ea
ted
u
s
in
g
an
FP
GA
an
d
a
GPU
-
b
ased
s
y
s
tem
.
As
a
r
e
s
u
lt,
E
OC
ca
n
b
e
d
etec
ted
an
d
class
if
ied
m
o
r
e
q
u
ick
l
y
,
p
o
te
n
tially
ex
ten
d
in
g
th
e
p
atien
ts
'
life
tim
e.
ACK
NO
WL
E
DG
E
M
E
NT
T
h
e
au
t
h
o
r
s
w
i
s
h
t
o
e
x
p
r
e
s
s
g
r
a
t
i
tu
d
e
to
D
r
.
Ma
n
g
a
l
a
G
o
u
r
i
S
.
R
.
,
P
r
o
f
.
an
d
H
ea
d
,
De
p
a
r
t
m
en
t
o
f
P
a
t
h
o
lo
g
y
,
R
am
a
i
a
h
M
ed
i
c
al
C
o
l
l
e
g
e
,
B
a
n
g
a
lo
r
e
,
f
o
r
p
r
o
v
id
i
n
g
v
a
lu
a
b
le
s
u
g
g
e
s
t
i
o
n
s
t
o
co
n
d
u
c
t
t
h
i
s
r
e
s
e
a
r
ch
wo
r
k
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
is
r
esear
ch
was n
o
t f
u
n
d
ed
b
y
an
y
ag
en
c
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.