I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
15
,
No
.
2
,
A
p
r
il
20
26
,
p
p
.
1
3
2
7
~
1
3
3
8
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
15
.i
2
.
p
p
1
3
2
7
-
1
3
3
8
1327
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Att
rib
ute
o
ptimiz
a
tion
to impro
v
e
brea
st cancer
pre
diction
using
ma
chine lea
rning
t
ech
niques
Ra
g
ha
v
endra
Srini
v
a
s
a
ia
h
1
,
Sa
nto
s
h K
um
a
r
J
a
nk
a
t
t
i
2
,
N
ira
nja
na
Sh
ra
v
a
na
bela
g
o
l
a
J
ina
cha
nd
ra
3
,
M
a
njuna
t
h Ra
m
a
nn
a
L
a
m
a
ni
4
,
B
ella
m
Vij
a
y
a
L
a
k
s
hm
i
5
,
Ris
hita
B
helwa
5
1
D
e
p
a
r
t
me
n
t
o
f
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
D
a
t
a
S
c
i
e
n
c
e
E
n
g
i
n
e
e
r
i
n
g
,
S
c
h
o
o
l
o
f
E
n
g
i
n
e
e
r
i
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
C
H
R
I
S
T
U
n
i
v
e
r
si
t
y
,
B
a
n
g
a
l
o
r
e
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
D
a
y
a
n
a
n
d
a
S
a
g
a
r
U
n
i
v
e
r
si
t
y
,
B
a
n
g
a
l
o
r
e
,
I
n
d
i
a
3
D
e
p
a
r
t
me
n
t
o
f
M
e
c
h
a
n
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
,
C
H
R
I
S
T
U
n
i
v
e
r
s
i
t
y
,
B
a
n
g
a
l
o
r
e
,
I
n
d
i
a
4
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
M
o
o
d
l
a
k
a
t
t
e
I
n
st
i
t
u
t
e
o
f
Te
c
h
n
o
l
o
g
y
,
K
u
n
d
a
p
u
r
a
,
I
n
d
i
a
5
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
C
H
R
I
S
T
U
n
i
v
e
r
si
t
y
,
B
a
n
g
a
l
o
r
e
,
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
u
l
13
,
2
0
2
4
R
ev
is
ed
J
an
8
,
2
0
2
6
Acc
ep
ted
J
an
25
,
2
0
2
6
Bre
a
st
c
a
n
c
e
r
(BC)
a
rise
s
wh
e
n
c
e
ll
s
g
ro
w
o
u
t
o
f
c
o
n
tro
l.
It
a
ffe
c
ts
wo
m
e
n
m
o
re
th
a
n
m
e
n
.
S
e
e
k
i
n
g
c
a
n
c
e
r
trea
tme
n
t
c
a
n
b
e
b
o
th
c
o
stl
y
a
n
d
ti
m
e
-
c
o
n
su
m
in
g
,
wit
h
tes
t
re
su
lt
s
sp
a
n
n
i
n
g
fr
o
m
a
fe
w
h
o
u
rs
t
o
se
v
e
ra
l
we
e
k
s.
Th
e
d
u
ra
ti
o
n
o
f
th
e
se
tes
ts
d
e
p
e
n
d
s
o
n
t
h
e
n
u
m
b
e
r
o
f
a
tt
rib
u
tes
with
in
th
e
d
a
tas
e
t.
Th
is
re
se
a
rc
h
p
a
p
e
r
e
n
d
e
a
v
o
rs
to
o
p
ti
m
ize
th
e
d
a
tas
e
t
a
tt
rib
u
tes
a
n
d
fin
d
th
e
a
c
c
u
ra
c
y
o
f
t
h
e
o
p
ti
m
iz
e
d
d
a
tas
e
t.
Th
e
p
rima
ry
g
o
a
l
is
to
re
d
u
c
e
fe
a
tu
re
s
u
sin
g
re
c
u
rsiv
e
fe
a
tu
re
e
li
m
in
a
ti
o
n
to
m
in
imiz
e
t
h
e
ti
m
e
tak
e
n
fo
r
th
e
tes
t
re
s
u
lt
.
T
h
is
wo
rk
d
isc
u
ss
e
s
th
e
m
a
c
h
in
e
lea
rn
in
g
tec
h
n
iq
u
e
a
n
d
th
e
ra
n
d
o
m
fo
re
st
(R
F
)
a
lg
o
r
it
h
m
,
wh
ich
h
e
l
p
s
d
e
term
in
e
th
e
p
a
ra
m
e
ter
a
c
c
u
ra
c
y
o
n
th
e
Wi
sc
o
n
sin
BC
d
iag
n
o
stic
d
a
tas
e
t
.
Th
e
m
e
th
o
d
a
c
h
iev
e
s
a
n
a
c
c
u
ra
c
y
o
f
9
6
.
4
9
%
wit
h
o
n
l
y
e
ig
h
tee
n
a
tt
rib
u
tes
.
It
h
a
s
a
id
e
d
t
h
e
h
e
a
lt
h
c
a
re
in
d
u
stry
in
fi
n
d
in
g
BC
i
n
les
s ti
m
e
a
n
d
imp
r
o
v
i
n
g
th
e
trea
tme
n
t.
K
ey
w
o
r
d
s
:
Attr
ib
u
te
o
p
tim
izatio
n
B
r
ea
s
t c
an
ce
r
p
r
ed
ictio
n
Ma
ch
in
e
lear
n
in
g
R
an
d
o
m
f
o
r
est cla
s
s
if
ier
W
is
co
n
s
in
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
:
R
ag
h
av
en
d
r
a
Srin
iv
asaiah
Dep
ar
tm
e
n
t
o
f
Ar
tif
ici
al
I
n
t
ell
i
g
e
n
c
e
a
n
d
D
ata
Sc
ie
n
c
e
E
n
g
i
n
ee
r
i
n
g
,
Sc
h
o
o
l
o
f
E
n
g
i
n
e
e
r
i
n
g
an
d
T
e
c
h
n
o
l
o
g
y
C
HR
I
ST
Un
iv
er
s
ity
Kan
m
an
ik
e,
Ku
m
b
alg
u
d
u
,
M
y
s
o
r
e
R
o
ad
,
B
an
g
alo
r
e
-
74
,
I
n
d
i
a
E
m
ail: r
ag
h
av
.
tr
g
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
er
e
ar
e
m
a
n
y
ca
s
es in
th
e
w
o
r
ld
wh
er
e
m
o
s
t w
o
m
en
h
a
v
e
b
r
ea
s
t c
an
ce
r
(
B
C
)
.
I
t is a
ty
p
e
o
f
ca
n
ce
r
ca
u
s
ed
b
y
t
h
e
o
v
er
g
r
o
wth
o
f
s
p
ec
if
ic
ce
lls
in
th
e
h
u
m
a
n
b
o
d
y
.
As
th
e
ca
n
ce
r
is
d
etec
ted
in
a
later
s
tag
e,
th
e
tr
ea
tm
en
t
is
a
f
in
a
n
cial
is
s
u
e
as
its
p
r
ice
in
cr
ea
s
es,
esp
ec
ially
in
ad
v
an
ce
d
s
tag
es.
Pro
m
p
t
an
d
p
r
ec
is
e
B
C
d
iag
n
o
s
is
is
es
s
en
tial
f
o
r
ef
f
icien
t
tr
ea
tm
en
t
p
lan
n
in
g
a
n
d
i
m
p
r
o
v
i
n
g
p
atien
t
o
u
tc
o
m
es
[
1
]
.
Ma
ch
in
e
lear
n
in
g
(
ML
)
tech
n
iq
u
es
h
av
e
b
ec
o
m
e
in
cr
ea
s
in
g
ly
p
o
wer
f
u
l
in
m
e
d
ical
r
esear
ch
o
v
e
r
th
e
p
ast
f
ew
y
ea
r
s
,
p
ar
ticu
la
r
ly
in
d
is
ea
s
e
p
r
ed
ictio
n
an
d
d
ia
g
n
o
s
is
.
So
m
e
ML
tec
h
n
iq
u
es
lik
e
ar
tific
ial
n
e
u
r
al
n
etwo
r
k
s
(
ANN)
,
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
(
SVM)
,
k
-
n
ea
r
est n
eig
h
b
o
r
s
(
K
NN
)
,
an
d
d
ec
en
tr
alize
d
ML
ap
p
r
o
ac
h
es lik
e
f
ed
er
ated
lear
n
in
g
ar
e
also
u
s
ed
to
d
ia
g
n
o
s
e
B
C
[
2
]
−
[
8
]
.
Sev
er
al
s
id
e
ef
f
ec
ts
m
u
s
t
b
e
ad
d
r
ess
ed
b
ef
o
r
e
tr
ea
tin
g
a
p
atien
t.
An
o
n
co
lo
g
is
t
will
b
e
ab
le
to
id
en
tify
a
n
d
d
iag
n
o
s
e
B
C
with
th
e
h
elp
o
f
m
a
g
n
etic
r
eso
n
a
n
ce
im
ag
in
g
(
MRI
)
s
ca
n
,
an
d
tis
s
u
e
b
io
p
s
y
,
wh
ich
is
a
tim
e
-
co
n
s
u
m
in
g
p
r
o
ce
s
s
[
9
]
.
Scr
ee
n
in
g
is
a
ty
p
e
o
f
test
th
at
id
en
tifie
s
m
o
s
t
ca
n
ce
r
ce
l
ls
q
u
ick
ly
an
d
ea
r
ly
[
1
0
]
.
I
m
ag
e
p
r
o
ce
s
s
in
g
tech
n
iq
u
es,
s
u
ch
as
co
n
tr
ast
-
r
estricte
d
ad
ap
tiv
e
h
is
to
g
r
a
m
eq
u
a
lizatio
n
(
C
L
AHE
)
,
r
em
o
v
e
n
o
is
e
f
r
o
m
th
e
im
ag
e
an
d
e
n
h
an
ce
im
ag
e
q
u
ality
b
y
i
n
ten
s
if
y
in
g
p
ix
el
v
alu
es
an
d
ce
ll
a
r
ea
s
to
p
r
o
d
u
ce
a
clea
r
er
p
ictu
r
e
with
s
h
ar
p
er
d
etails,
f
ac
ilit
atin
g
th
e
id
en
tific
atio
n
o
f
o
b
jects
an
d
b
o
u
n
d
ar
ies.
Fu
zz
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
2
,
Ap
r
il
20
26
:
1
3
2
7
-
1
3
3
8
1328
SVM,
B
ay
esian
class
if
ier
,
a
n
d
r
a
n
d
o
m
f
o
r
est
(
R
F)
ap
p
r
o
ac
h
es
ar
e
u
s
ed
to
class
if
y
th
ese
p
r
e
-
p
r
o
ce
s
s
ed
p
ictu
r
es
[
1
1
]
.
T
h
er
e
ar
e
also
s
ev
er
al
ca
s
es
wh
er
e
B
C
is
id
en
tifie
d
th
r
o
u
g
h
b
l
o
o
d
r
ep
o
r
ts
.
B
C
ca
n
b
e
id
en
tifie
d
b
y
a
lu
m
p
o
r
m
ass
in
th
e
b
r
ea
s
t
tis
s
u
e,
an
d
n
o
t
all
lu
m
p
s
ar
e
ca
n
ce
r
o
u
s
,
b
u
t
m
am
m
o
g
r
ap
h
ic
im
ag
es
ca
n
id
en
tify
t
h
o
s
e
th
at
ar
e
[
1
2
]
.
B
C
is
a
co
m
p
le
x
co
n
d
itio
n
th
at
h
as
m
an
y
m
o
lec
u
lar
a
n
d
s
y
m
p
to
m
m
an
i
f
estatio
n
s
.
I
t is ess
en
tia
l to
id
en
tify
th
e
ch
ar
ac
ter
is
tics
with
h
ig
h
p
r
ed
ic
tab
ilit
y
am
o
n
g
d
if
f
er
en
t v
a
r
iab
les to
m
ak
e
r
eliab
le
an
d
u
n
d
er
s
tan
d
a
b
le
m
o
d
els [
1
3
]
.
On
e
p
o
p
u
lar
f
ea
t
u
r
e
s
elec
tio
n
(
FS
)
tech
n
iq
u
e
n
am
ed
r
ec
u
r
s
iv
e
f
ea
tu
r
e
ex
tr
ac
tio
n
(
R
FE)
s
er
v
es
as
a
p
o
wer
f
u
l
to
o
l
f
o
r
a
s
y
s
tem
atic
r
ed
u
ctio
n
o
f
u
n
n
ec
ess
ar
y
d
ata,
wh
ich
in
cr
ea
s
es
th
e
ef
f
icien
cy
an
d
in
ter
p
r
etab
ilit
y
o
f
m
o
d
els
f
o
r
p
r
ed
ictio
n
.
T
h
e
p
r
im
ar
y
m
o
tiv
atio
n
f
o
r
th
is
r
esear
ch
is
ex
p
lo
r
atio
n
o
f
th
e
u
s
ag
e
o
f
R
FE
in
th
e
co
n
tex
t
o
f
B
C
p
r
ed
ictio
n
.
R
FE
’
s
ab
ilit
y
to
ass
ess
an
d
r
an
k
v
a
r
iab
les
b
ased
o
n
th
eir
i
m
p
o
r
ta
n
ce
cr
ea
tes
s
co
p
e
f
o
r
a
n
ew
s
et
o
f
f
ea
tu
r
es
th
at
h
elp
p
r
ed
ict
B
C
m
o
s
t
ef
f
ec
tiv
ely
[
1
4
]
.
T
h
e
f
o
llo
win
g
s
ec
tio
n
s
wo
u
ld
ex
p
lo
r
e
in
d
etail
tech
n
o
lo
g
ies
u
s
ed
,
f
ea
t
u
r
es
o
f
th
e
d
ataset,
an
d
ex
p
e
r
im
en
t
o
u
tco
m
e
ca
r
r
ied
o
u
t
u
s
in
g
R
FE
f
o
r
p
r
e
d
ictin
g
B
C
.
T
h
e
a
im
is
to
h
ig
h
lig
h
t
t
h
e
b
e
n
ef
its
b
r
o
u
g
h
t
b
y
attr
ib
u
te
r
ed
u
ctio
n
in
p
r
ed
ictin
g
B
C
ac
cu
r
ately
an
d
m
o
d
el
i
n
ter
p
r
e
tab
ilit
y
,
wh
ich
co
u
l
d
p
av
e
th
e
way
f
o
r
f
u
r
th
e
r
ef
f
o
r
ts
f
o
r
ea
r
ly
d
iag
n
o
s
is
an
d
p
er
s
o
n
alize
d
m
e
d
icatio
n
.
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
T
o
u
n
d
er
s
tan
d
th
e
d
if
f
er
en
t
a
lg
o
r
ith
m
s
u
s
ed
f
o
r
p
r
ed
ictin
g
an
d
d
iag
n
o
s
in
g
B
C
,
d
if
f
er
e
n
t
r
esear
ch
p
u
b
licatio
n
s
ar
e
s
tu
d
ied
.
W
h
e
n
it
co
m
es
to
t
h
e
p
r
ed
ictio
n
o
f
B
C
,
s
o
m
e
o
f
th
ese
alg
o
r
ith
m
s
illu
s
tr
ate
d
if
f
er
e
n
t
lev
els
o
f
ac
cu
r
ac
y
.
T
h
e
n
ee
d
f
o
r
th
e
u
s
ag
e
o
f
a
d
v
an
ce
d
m
o
d
els
f
o
r
th
e
b
etter
id
en
tific
atio
n
o
f
B
C
is
d
escr
ib
ed
th
r
o
u
g
h
an
au
to
m
ate
d
d
etec
ti
o
n
m
ec
h
a
n
is
m
b
ased
o
n
a
n
en
s
em
b
le
o
f
class
if
ier
s
.
T
h
e
ch
a
llen
g
es
f
ac
ed
wh
e
n
id
en
tify
in
g
B
C
at
an
ea
r
ly
s
tag
e
d
u
e
to
th
e
s
m
all
s
ize
o
f
th
e
ca
n
ce
r
ce
lls
ar
e
d
escr
ib
e
d
ex
ten
s
iv
ely
.
T
h
e
tim
e
-
co
n
s
u
m
in
g
p
r
o
ce
s
s
o
f
test
in
g
is
al
s
o
em
p
h
asized
.
T
h
e
ANN
an
d
an
en
s
em
b
le
o
f
ML
alg
o
r
ith
m
s
ar
e
s
o
m
e
o
f
th
e
d
if
f
er
en
t
ML
tech
n
iq
u
es
u
s
ed
.
W
h
en
u
s
in
g
all
o
f
th
e
v
ar
iab
les
with
in
th
e
d
ataset,
th
e
p
r
o
p
o
s
ed
m
eth
o
d
illu
s
tr
ates
an
ac
cu
r
ac
y
r
ate
o
f
9
8
.
8
3
%
[
1
5
]
.
T
h
e
f
iv
e
d
if
f
er
en
t
ML
alg
o
r
ith
m
s
em
p
l
o
y
ed
in
th
e
d
ataset
ar
e
ANN,
R
F,
K
NN,
lo
g
is
tic
r
eg
r
es
s
io
n
(
L
R
)
,
an
d
SV
M.
A
cc
u
r
ac
y
,
s
p
ec
if
icity
,
s
e
n
s
itiv
ity
,
F1
-
s
co
r
e
,
p
r
ec
is
io
n
,
n
eg
ativ
e
p
r
ed
ictiv
e
v
alu
es,
f
alse
n
eg
ativ
es,
an
d
f
a
ls
e
p
o
s
itiv
es
ar
e
th
e
m
etr
ics
e
m
p
lo
y
ed
t
o
test
th
e
p
er
f
o
r
m
an
ce
o
f
d
if
f
er
en
t
tec
h
n
iq
u
es.
As
ev
id
e
n
t
f
r
o
m
th
e
r
esu
lts
,
th
e
ANN
em
p
lo
y
s
all
th
e
v
ar
iab
les
to
ac
h
iev
e
h
ig
h
er
ac
c
u
r
ac
y
o
f
9
8
.
5
7
%
[
1
6
]
.
I
n
c
o
n
s
id
er
in
g
m
ax
im
izin
g
ac
c
u
r
ac
y
a
n
d
m
in
im
izin
g
er
r
o
r
s
,
th
e
p
r
ed
ictio
n
o
f
B
C
th
r
o
u
g
h
ML
was
ex
p
lo
r
ed
.
I
n
co
r
r
ec
tin
g
th
e
er
r
o
r
s
in
ex
is
tin
g
m
eth
o
d
s
,
th
er
e
was
an
o
p
p
o
r
tu
n
ity
to
im
p
r
o
v
e
p
r
ed
ic
tio
n
m
o
d
els.
Fo
u
r
d
if
f
er
en
t
m
o
d
els
o
f
ML
alg
o
r
ith
m
s
⎯
SVM,
ANN,
L
R
,
an
d
RF
⎯
wer
e
u
s
ed
in
th
e
d
ataset
th
r
o
u
g
h
th
e
J
u
p
y
ter
en
v
ir
o
n
m
en
t.
Fro
m
th
e
ex
p
er
im
en
tal
r
esu
lts
,
L
R
u
ti
lized
all
th
e
v
ar
iab
les an
d
p
er
f
o
r
m
ed
b
etter
th
an
o
th
e
r
m
o
d
els in
ac
c
u
r
ac
y
m
o
d
els [
1
7
]
.
A
th
o
r
o
u
g
h
d
is
cu
s
s
io
n
is
g
iv
en
o
n
s
eg
m
en
tatio
n
-
b
ased
ML
an
d
ef
f
ec
tiv
e
im
a
g
e
p
r
o
ce
s
s
in
g
m
eth
o
d
s
f
o
r
B
C
d
iag
n
o
s
is
.
T
h
e
in
p
u
t
d
ata
f
o
r
th
is
wo
r
k
is
m
am
m
o
g
r
ap
h
y
p
ictu
r
es.
T
o
im
p
r
o
v
e
im
ag
e
q
u
ality
,
th
e
C
L
AHE
m
eth
o
d
is
em
p
lo
y
e
d
,
wh
ich
h
elp
s
r
e
d
u
ce
n
o
is
e
in
th
e
im
a
g
es
an
d
en
h
a
n
ce
s
im
ag
e
q
u
ality
.
T
ec
h
n
iq
u
es
lik
e
f
u
zz
y
SVM,
R
F,
an
d
B
ay
esian
class
if
ier
g
r
o
u
p
th
e
p
r
e
p
r
o
ce
s
s
ed
im
ag
es.
Fro
m
th
e
r
esu
lt
o
b
tain
ed
,
f
u
zz
y
SVM
p
er
f
o
r
m
s
b
etter
th
a
n
t
h
e
o
th
er
m
eth
o
d
s
with
a
n
ac
c
u
r
ac
y
o
f
9
4
%
[
1
8
]
.
A
n
ew
n
ested
en
s
em
b
le
tech
n
i
q
u
e
f
o
r
au
t
o
m
ated
B
C
d
iag
n
o
s
is
was
in
tr
o
d
u
ce
d
,
d
em
o
n
s
tr
atin
g
a
r
esear
c
h
g
a
p
r
elate
d
to
th
e
lim
ited
ex
p
lo
r
ati
o
n
o
f
h
y
p
e
r
p
ar
am
eter
tu
n
in
g
an
d
FS
.
I
n
th
is
wo
r
k
,
Me
ta
class
es
ar
e
u
tili
ze
d
in
c
o
n
ju
n
ctio
n
with
cr
o
s
s
-
v
alid
atio
n
tec
h
n
iq
u
es
f
o
r
m
o
d
el
ev
alu
atio
n
to
d
is
tin
g
u
is
h
b
etwe
en
b
e
n
ig
n
b
r
ea
s
t
tu
m
o
r
s
an
d
m
alig
n
an
t
ca
n
ce
r
s
.
Fro
m
th
e
r
esu
lt
o
b
tain
ed
it
was
f
o
u
n
d
th
at
SV
-
B
ay
es
Net
-
3
-
Me
taC
las
s
i
f
ier
an
d
SV
-
n
aïv
e
B
ay
es
-
3
-
Me
taC
las
s
if
ier
ac
h
iev
ed
ac
cu
r
ac
y
o
f
9
8
.
0
7
% with
co
m
p
lete
attr
ib
u
tes [
1
9
]
.
Dif
f
er
en
t
ML
tech
n
iq
u
es
an
d
d
ee
p
lear
n
in
g
(
DL
)
alg
o
r
ith
m
s
d
etec
t
b
en
ig
n
an
d
m
alig
n
an
t
tu
m
o
r
s
.
Mo
d
els
s
u
ch
as
SVM,
L
R
,
m
u
ltil
ay
er
p
er
ce
p
tr
o
n
(
ML
P)
,
A
NN,
an
d
K
NN
wer
e
a
p
p
lied
to
th
e
d
ataset,
an
d
th
e
r
esu
lts
wer
e
co
m
p
ar
ed
.
T
h
e
co
m
p
ar
is
o
n
o
f
th
e
r
esu
lts
f
o
u
n
d
th
at
th
e
ANN
ac
h
iev
ed
an
ac
cu
r
ac
y
o
f
9
9
.
3
%
u
s
in
g
th
e
c
o
m
p
lete
s
et
o
f
attr
ib
u
tes
[
2
0
]
.
A
co
m
p
ar
ativ
e
a
n
aly
s
is
o
f
ML
alg
o
r
ith
m
s
f
o
r
B
C
p
r
ed
ictio
n
was
co
n
d
u
cte
d
,
id
en
tif
y
in
g
c
h
allen
g
es
r
elate
d
to
d
ata
s
ize
an
d
th
e
lim
itatio
n
s
o
f
d
ec
is
io
n
tr
e
es
(
DT
)
in
s
p
ec
if
ic
s
ce
n
ar
io
s
.
A
d
ataset
is
s
u
b
jecte
d
to
s
ev
er
al
tec
h
n
iq
u
es,
i
n
clu
d
in
g
SVM,
K
NN,
DT
,
K
-
m
e
an
s
,
an
d
ANN,
f
o
r
th
e
ea
r
ly
d
ia
g
n
o
s
is
o
f
b
en
ig
n
an
d
m
alig
n
a
n
t
ca
n
ce
r
.
SVM
was
d
eter
m
in
ed
to
h
av
e
ac
c
u
r
ac
y
o
f
9
7
.
1
4
%
wh
en
all
ch
ar
ac
ter
is
tics
wer
e
u
s
ed
[
2
1
]
.
T
h
e
n
ec
ess
ity
to
f
in
d
th
e
b
est cla
s
s
if
icatio
n
f
ea
tu
r
es a
n
d
co
m
b
in
e
r
ad
io
m
ics
an
d
g
en
o
m
es
d
ata
was
th
e
m
ain
f
o
cu
s
o
f
t
h
e
e
x
p
lo
r
atio
n
o
f
ML
alg
o
r
ith
m
s
f
o
r
B
C
ty
p
e
ca
teg
o
r
izatio
n
.
T
r
ip
le
-
n
eg
ativ
e
an
d
n
o
n
-
tr
ip
le
-
n
eg
ativ
e
B
C
wer
e
clas
s
if
ied
u
s
in
g
g
en
e
ex
p
r
ess
io
n
d
ata
an
d
th
e
ML
tech
n
iq
u
e.
SVM,
K
-
m
ea
n
s
,
n
aïv
e
B
ay
es,
an
d
DT
a
r
e
th
e
f
o
u
r
class
if
icatio
n
m
o
d
els
th
at
ar
e
co
m
p
ar
ed
.
T
h
e
o
u
tco
m
e
u
n
eq
u
i
v
o
ca
lly
s
h
o
wn
th
at
ML
alg
o
r
ith
m
s
o
u
tp
er
f
o
r
m
o
t
h
er
tech
n
iq
u
es
[
2
2
]
.
T
o
class
if
y
p
atien
ts
in
to
g
r
o
u
p
s
n
o
ca
n
ce
r
,
ca
n
ce
r
,
a
n
d
n
o
n
-
ca
n
ce
r
o
u
s
,
th
e
r
esear
c
h
er
s
u
s
ed
ML
an
d
DL
tech
n
iq
u
es
f
o
r
th
e
id
en
tific
atio
n
o
f
B
C
f
r
o
m
th
e
th
er
m
o
g
r
ap
h
ic
i
m
ag
e.
T
h
r
ee
class
if
icatio
n
s
y
s
tem
s
ar
e
u
s
ed
:
R
F,
SVM,
an
d
co
n
v
o
l
u
tio
n
n
e
u
r
al
n
etwo
r
k
(
C
NN)
.
C
NN
h
as b
ee
n
f
o
u
n
d
m
o
r
e
ef
f
icien
t th
an
o
t
h
er
s
y
s
tem
s
as p
r
o
v
ed
i
n
[
2
3
]
.
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
A
ttr
ib
u
te
o
p
timis
a
tio
n
to
imp
r
o
ve
b
r
ea
s
t c
a
n
ce
r
p
r
ed
ictio
n
u
s
in
g
ma
ch
in
e
…
(
R
a
g
h
a
ve
n
d
r
a
S
r
in
iva
s
a
ia
h
)
1329
A
d
etailed
d
is
cu
s
s
io
n
o
n
p
ictu
r
e
f
o
r
m
atio
n
a
n
d
p
r
ep
r
o
ce
s
s
in
g
m
eth
o
d
s
r
elate
d
t
o
th
e
d
etec
tio
n
o
f
B
C
was
in
tr
o
d
u
ce
d
.
I
n
a
n
ef
f
o
r
t
to
im
p
r
o
v
e
th
e
p
r
ec
is
io
n
o
f
t
h
is
m
o
d
el,
th
e
im
p
o
r
tan
ce
o
f
th
e
in
teg
r
atio
n
o
f
ar
tific
ial
in
tellig
en
ce
(
AI
)
tec
h
n
iq
u
es
an
d
n
o
v
el
m
eth
o
d
o
lo
g
ies
is
em
p
h
asized
in
th
is
p
ap
er
.
T
h
e
s
ig
n
if
ica
n
ce
o
f
v
ar
y
in
g
th
e
s
ize
o
f
th
e
u
s
e
d
d
ata
in
d
er
iv
in
g
b
r
o
ad
er
asp
ec
ts
was
em
p
h
asized
in
th
e
th
o
r
o
u
g
h
d
is
cu
s
s
io
n
,
s
u
g
g
esti
n
g
a
p
r
o
b
a
b
le
ar
ea
o
f
r
esear
ch
f
o
r
th
e
f
u
tu
r
e
[
2
4
]
.
T
h
e
m
o
s
t
f
atal
an
d
life
-
th
r
ea
ten
in
g
ty
p
e
o
f
ca
n
ce
r
is
B
C
,
wh
er
ein
it
is
f
ir
s
t
d
is
c
o
v
er
ed
wh
en
b
r
ea
s
t
en
lar
g
em
en
t
h
ap
p
en
s
.
E
ar
ly
d
iag
n
o
s
is
is
th
er
ef
o
r
e
cr
u
cial.
Ma
m
m
o
g
r
a
p
h
y
a
n
d
u
ltra
s
o
u
n
d
m
eth
o
d
s
ar
e
ty
p
ically
em
p
l
o
y
ed
f
o
r
th
e
d
etec
tio
n
.
ML
te
ch
n
iq
u
es
lik
e
C
NN
ca
n
b
e
u
s
ed
to
d
etec
t
m
am
m
o
g
r
am
s
.
E
ac
h
lay
e
r
o
f
th
e
C
NN
id
en
tifie
s
th
e
f
ea
tu
r
es
an
d
p
atter
n
s
th
at
h
elp
ef
f
icien
tly
f
in
d
an
o
m
alies.
A
n
ap
p
r
o
ac
h
b
ased
o
n
B
r
ea
s
eNe
t
-
SVM
i
s
em
p
lo
y
ed
o
n
th
e
d
ig
ital
d
atab
ase
f
o
r
s
cr
ee
n
in
g
m
am
m
o
g
r
a
p
h
y
(
DDSM)
d
atasets
to
au
to
m
atica
lly
d
etec
t
an
d
class
if
y
B
C
.
Acc
o
r
d
in
g
to
t
h
e
r
esu
lts
,
th
e
m
o
d
el
ac
h
iev
ed
ac
cu
r
ac
y
o
f
9
9
.
1
6
%
[
2
5
]
.
B
C
is
o
n
e
o
f
th
e
m
o
s
t
cr
itical
wo
r
ld
wid
e
h
ea
lth
is
s
u
es,
an
d
m
o
s
t
o
f
ten
it
af
f
ec
ts
wo
m
en
.
T
h
e
g
r
a
d
ien
t
b
o
o
s
tin
g
(
GB
)
m
eth
o
d
is
ap
p
lied
to
th
e
d
ataset
to
id
en
tify
t
h
e
v
ital
cr
itical
f
ac
to
r
s
.
T
h
e
GB
m
eth
o
d
was
em
p
lo
y
ed
f
o
r
d
i
s
ea
s
e
cla
s
s
if
icatio
n
.
T
o
ev
alu
ate
th
e
m
o
d
el
’
s
p
er
f
o
r
m
an
ce
,
v
ar
i
o
u
s
cr
iter
ia,
in
clu
d
in
g
s
en
s
itiv
ity
,
ac
cu
r
ac
y
,
s
p
ec
if
icity
,
F1
-
s
co
r
e,
an
d
p
o
s
itiv
e
an
d
n
eg
ativ
e
p
r
ed
ictiv
e
v
alu
es,
wer
e
u
s
ed
.
All m
eth
o
d
s
ac
h
iev
ed
1
0
0
% a
cc
u
r
ac
y
[
2
6
]
.
FS
is
th
e
p
r
o
ce
s
s
o
f
attem
p
tin
g
to
p
r
ev
e
n
t
th
e
m
u
ltip
licity
o
f
f
ea
tu
r
es,
wh
ich
is
th
e
m
o
s
t
s
ig
n
if
ican
t
p
r
o
b
lem
in
d
is
ea
s
e
d
iag
n
o
s
is
.
FS
m
eth
o
d
s
h
elp
in
d
etec
tin
g
th
e
ess
en
tial f
ea
tu
r
es th
at
co
n
tr
ib
u
te
ef
f
ec
tiv
ely
to
th
e
m
o
d
els'
p
er
f
o
r
m
an
ce
im
p
r
o
v
em
en
t.
FS
m
eth
o
d
s
h
elp
eli
m
in
ate
an
d
r
em
o
v
e
u
n
n
ec
ess
a
r
y
d
ata
[
2
7
]
.
T
h
r
ee
class
if
ier
s
b
ased
o
n
ANN,
g
en
etic
alg
o
r
ith
m
s
(
GA)
,
an
d
p
a
r
t
icle
s
war
m
o
p
tim
izatio
n
(
PSO
)
with
FS
m
eth
o
d
s
ar
e
ap
p
lied
o
n
th
e
W
is
co
n
s
in
d
ataset.
I
t
was
f
o
u
n
d
th
at
th
e
PS
O
cla
s
s
if
ier
s
ac
h
iev
ed
an
ac
cu
r
ac
y
o
f
9
7
.
2
%,
s
p
ec
if
icity
o
f
9
5
.
6
%
an
d
s
en
s
itiv
ity
o
f
9
8
%
;
GA
class
if
ie
r
s
attain
ed
an
ac
cu
r
ac
y
o
f
9
6
.
6
%,
s
p
ec
if
icity
o
f
9
3
.
7
%
an
d
s
en
s
itiv
ity
o
f
9
7
.
5
%
;
an
d
ANN
class
if
ier
s
ac
h
ie
v
ed
an
ac
c
u
r
ac
y
o
f
9
7
.
3
%,
s
p
ec
if
icity
o
f
9
5
.
1
%
an
d
s
en
s
itiv
ity
o
f
9
8
.
4
%
[
2
8
]
.
MRI
is
id
en
tifie
d
as
a
p
o
ten
t
ial
ca
n
d
id
ate
f
o
r
d
ir
ec
tin
g
n
ea
r
-
in
f
r
ar
e
d
s
p
ec
tr
al
to
m
o
g
r
a
p
h
y
,
wh
ich
e
n
h
an
ce
s
th
e
s
p
ec
if
icity
an
d
s
en
s
itiv
ity
o
f
B
C
d
iag
n
o
s
is
.
Ho
wev
er
,
th
e
d
if
f
icu
lty
in
lig
h
t
p
r
o
p
a
g
atio
n
in
th
e
MRI
im
ag
e
s
af
f
ec
ts
th
e
p
er
f
o
r
m
an
ce
o
f
s
p
ec
tr
al
to
m
o
g
r
ap
h
y
.
T
o
o
v
er
c
o
m
e
th
e
p
r
o
b
lem
s
,
a
3
D
s
p
ec
tr
al
im
ag
e
was
d
ev
e
lo
p
ed
g
u
id
ed
b
y
MR,
wh
ich
ac
h
iev
ed
an
ac
cu
r
ac
y
o
f
8
9
.
5
%,
s
p
ec
if
icity
o
f
9
2
.
9
%,
s
en
s
itiv
ity
o
f
8
7
.
5
%
,
an
d
a
r
ec
eiv
er
o
p
er
atin
g
c
h
ar
ac
ter
is
tic
(
R
O
C
)
cu
r
v
e
o
f
0
.
9
8
[
2
9
]
.
R
FE
is
an
FS
tech
n
iq
u
e
th
at
c
o
n
tin
u
ally
attem
p
ts
to
s
elec
t
th
e
m
o
s
t
cr
itical
f
ea
tu
r
es,
p
r
im
ar
ily
f
o
c
u
s
in
g
o
n
class
if
icatio
n
ac
cu
r
ac
y
an
d
th
e
lear
n
ed
m
o
d
el.
R
FE
wo
r
k
s
b
y
s
eq
u
en
tially
r
em
o
v
i
n
g
th
e
wo
r
s
t
f
ea
tu
r
es
,
wh
ich
r
ed
u
ce
s
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
tech
n
iq
u
e.
R
FE
wo
r
k
s
b
y
u
s
in
g
th
e
b
ac
k
war
d
elim
in
atio
n
tec
h
n
iq
u
e,
wh
ich
in
v
o
lv
es
elim
in
atin
g
attr
ib
u
tes
to
r
ed
u
ce
ef
f
icien
cy
r
ec
u
r
s
iv
ely
[
3
0
]
,
[
3
1
]
.
R
FE
s
u
f
f
er
s
f
r
o
m
p
r
o
b
lem
s
s
u
ch
as
in
co
n
s
is
ten
cies
with
th
e
f
ea
tu
r
e
r
an
k
in
g
cr
iter
io
n
a
n
d
th
e
m
ax
im
u
m
m
ar
g
in
co
n
ce
p
t,
as
th
e
co
m
p
u
tatio
n
o
f
th
e
cr
iter
io
n
is
d
o
n
e
lo
ca
lly
.
A
d
d
itio
n
ally
,
th
e
r
e
is
a
lack
o
f
g
lo
b
al
m
ea
s
u
r
em
en
t o
f
f
ea
tu
r
e
i
m
p
o
r
tan
ce
,
wh
ich
is
n
o
t g
u
ar
a
n
teed
to
b
e
o
p
tim
a
l,
an
d
a
h
ig
h
r
is
k
o
f
o
v
er
f
itti
n
g
[
3
2
]
,
[
3
3
]
.
E
ar
ly
d
iag
n
o
s
is
o
f
B
C
is
th
e
b
est
way
to
cu
r
e
th
e
d
is
ea
s
e.
T
o
s
o
lv
e
th
e
p
r
o
b
lem
s
ass
o
ciate
d
with
er
r
o
r
s
in
d
iag
n
o
s
in
g
th
e
d
is
ea
s
e,
a
h
y
b
r
i
d
m
o
d
el
c
o
m
b
in
in
g
p
r
in
cip
al
co
m
p
o
n
e
n
t
an
aly
s
is
(
PC
A)
an
d
SVM
i
s
p
r
o
p
o
s
ed
.
PC
A
was
u
s
ed
to
s
elec
t
f
ea
tu
r
es
in
th
e
f
i
r
s
t
cy
cle
an
d
r
ed
u
ce
th
e
n
u
m
b
er
o
f
f
ea
tu
r
es
in
th
e
s
ec
o
n
d
cy
cle.
T
h
e
r
ed
u
ce
d
f
ea
tu
r
es
ar
e
f
ed
in
to
SVM
f
o
r
r
is
k
ass
es
s
m
en
t
an
d
d
iag
n
o
s
is
;
th
e
p
r
o
p
o
s
ed
m
o
d
el
ac
h
iev
ed
a
n
ac
c
u
r
ac
y
o
f
9
7
.
6
2
%,
s
p
ec
if
icity
o
f
1
0
0
%,
a
n
d
s
en
s
itiv
ity
o
f
9
5
.
2
4
%
[
3
4
]
.
I
n
r
ec
en
t
y
ea
r
s
,
m
a
n
y
co
m
p
u
ter
ized
d
iag
n
o
s
tic
s
y
s
te
m
s
h
av
e
b
ee
n
d
e
v
elo
p
e
d
to
r
e
d
u
ce
h
u
m
an
er
r
o
r
s
an
d
h
elp
p
h
y
s
ician
s
d
iag
n
o
s
e
d
is
ea
s
es
ef
f
ec
tiv
ely
.
An
atte
m
p
t
was
m
ad
e
to
cr
ea
te
a
c
o
m
p
u
ter
-
aid
ed
d
iag
n
o
s
is
s
y
s
t
em
u
tili
zin
g
p
atter
n
r
ec
o
g
n
itio
n
s
o
f
twar
e.
A
h
y
b
r
i
d
tech
n
i
q
u
e
was
p
r
o
p
o
s
ed
b
y
co
m
b
in
in
g
L
R
an
d
PC
A,
in
wh
ich
PC
A
was
u
s
ed
f
o
r
FS
an
d
L
R
f
o
r
class
if
icatio
n
o
f
B
C
tu
m
o
r
s
.
T
h
e
h
y
b
r
id
m
eth
o
d
ac
h
iev
e
d
an
ac
c
u
r
ac
y
o
f
1
0
0
%
b
y
o
u
tp
er
f
o
r
m
in
g
m
a
n
y
e
x
is
tin
g
m
eth
o
d
s
,
wh
ic
h
in
clu
d
ed
r
ed
u
cin
g
t
h
e
q
u
ality
o
f
th
e
attr
ib
u
tes,
th
e
n
u
m
b
e
r
o
f
attr
ib
u
tes,
an
d
r
esp
o
n
s
e
tim
e
[
3
5
]
.
I
r
r
elev
an
t
a
n
d
d
u
p
lica
ted
f
ea
tu
r
es
lead
t
o
a
r
e
d
u
c
tio
n
in
p
r
ed
ictio
n
ac
cu
r
ac
y
an
d
also
m
ak
e
th
e
s
y
s
tem
am
b
ig
u
o
u
s
.
T
h
e
liter
atu
r
e
co
llectio
n
an
d
s
eg
r
eg
atio
n
p
r
o
v
id
e
a
b
etter
o
v
er
v
iew
o
f
c
u
r
r
en
t
s
tu
d
ies,
a
llo
w
f
o
r
a
d
ee
p
er
u
n
d
er
s
tan
d
i
n
g
o
f
th
e
r
esear
ch
la
n
d
s
ca
p
e,
id
en
tify
g
ap
s
,
a
n
d
p
r
o
v
id
e
a
f
o
u
n
d
atio
n
f
o
r
s
u
b
s
eq
u
en
t
r
esear
ch
wo
r
k
o
n
attr
ib
u
te
o
p
tim
izatio
n
f
o
r
B
C
p
r
ed
ictio
n
.
T
h
r
o
u
g
h
t
h
is
ap
p
r
o
ac
h
,
th
e
m
e
d
ical
co
s
ts
an
d
test
tim
e
will b
e
r
ed
u
ce
d
.
3.
M
E
T
H
O
D
Her
e
ar
e
th
e
s
tep
s
in
v
o
lv
ed
in
th
e
p
r
o
ce
s
s
o
f
f
in
d
i
n
g
th
e
ac
cu
r
ac
y
o
f
th
e
o
p
tim
ized
d
ataset
an
d
th
e
o
r
ig
in
al
d
ataset
u
s
in
g
ML
tec
h
n
iq
u
es:
i)
C
o
llectin
g
d
ataset:
th
e
f
ir
s
t
p
r
o
ce
s
s
in
th
is
u
n
d
er
tak
i
n
g
i
s
th
e
co
llectio
n
o
f
th
e
d
atas
et
th
at
will
b
e
em
p
lo
y
ed
in
test
in
g
an
d
tr
ain
in
g
th
e
ML
m
o
d
els
as
well
a
s
ev
alu
atin
g
th
e
ac
c
u
r
ac
y
o
f
B
C
d
etec
tio
n
.
T
h
is
d
ataset
ca
n
b
e
o
b
tain
e
d
f
r
o
m
d
if
f
er
e
n
t
s
o
u
r
ce
s
in
th
e
f
o
r
m
o
f
d
atab
ases
o
r
s
p
r
ea
d
s
h
ee
ts
.
I
n
th
is
p
ap
er
,
th
e
d
ataset
b
ein
g
u
tili
ze
d
h
as b
ee
n
o
b
tain
ed
f
r
o
m
Kag
g
le.
ii)
C
h
ec
k
in
g
f
o
r
m
is
s
in
g
v
alu
es:
m
is
s
in
g
v
alu
es
n
ee
d
to
b
e
v
er
i
f
ied
af
ter
co
llectio
n
.
T
h
e
r
e
m
ay
b
e
m
is
s
in
g
v
alu
es
in
th
e
d
ata
f
o
r
s
ev
er
a
l
r
ea
s
o
n
s
,
in
clu
d
in
g
m
is
tak
es
in
th
e
p
r
o
ce
s
s
o
f
d
ata
en
tr
y
.
T
h
e
ab
o
v
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
2
,
Ap
r
il
20
26
:
1
3
2
7
-
1
3
3
8
1330
p
r
o
ce
d
u
r
e
is
ca
lled
d
ata
p
r
ep
r
o
ce
s
s
in
g
.
T
h
e
r
o
w
i
n
f
o
r
m
atio
n
s
h
o
u
l
d
b
e
d
elete
d
in
ca
s
e
m
is
s
in
g
v
alu
es
ex
is
t
in
th
e
d
ata.
Oth
er
wis
e,
th
e
u
s
er
ca
n
p
r
o
ce
e
d
to
th
e
n
ex
t
p
h
ase
to
s
p
lit
th
e
d
ataset
v
ia
th
e
ML
t
ec
h
n
iq
u
e
n
am
ed
r
a
n
d
o
m
f
o
r
e
s
t c
lass
if
ier
(
R
FC
)
.
iii)
Dele
te
attr
ib
u
tes:
th
e
u
s
er
n
ee
d
s
to
d
elete
th
e
r
o
w
i
n
f
o
r
m
a
tio
n
f
o
r
t
h
e
m
is
s
in
g
v
alu
e,
i
f
th
er
e
ar
e
an
y
,
f
r
o
m
th
is
s
tep
b
ec
a
u
s
e
it will h
elp
o
b
tain
th
e
co
r
r
ec
t
p
o
s
itiv
e
r
esu
lt.
iv
)
Sp
litt
in
g
d
ataset
u
s
in
g
R
FC
:
i
f
th
er
e
ar
e
n
o
m
is
s
in
g
v
alu
es
in
th
e
d
ataset,
th
e
n
th
e
d
atase
t
ca
n
b
e
s
p
li
t
in
to
th
e
tr
ain
in
g
d
ata
s
et
an
d
th
e
test
in
g
d
ata
s
et
b
y
u
s
in
g
th
e
R
FC
tech
n
iq
u
e.
v)
I
n
itial
ac
cu
r
ac
y
:
th
e
ML
al
g
o
r
ith
m
is
tr
ain
ed
o
n
a
t
r
ain
in
g
d
ataset
af
ter
th
e
d
ataset
h
as
b
ee
n
s
p
lit.
Af
ter
tr
ain
in
g
th
e
alg
o
r
ith
m
,
it
is
id
ea
l
to
ch
ec
k
h
o
w
it
p
er
f
o
r
m
ed
u
s
in
g
a
test
in
g
d
ataset.
T
h
e
i
n
itial
ac
cu
r
ac
y
o
f
th
e
alg
o
r
ith
m
is
o
th
e
r
wis
e
ca
lled
th
e
ac
cu
r
ac
y
o
f
t
h
e
o
r
i
g
in
al
d
ataset.
v
i)
R
FE
to
r
ed
u
ce
th
e
d
ataset:
R
FE
is
a
tech
n
iq
u
e
u
s
ed
to
r
e
d
u
ce
th
e
n
u
m
b
er
o
f
f
ea
tu
r
es
in
th
e
d
ataset.
Op
tim
izin
g
th
e
n
u
m
b
er
o
f
f
e
atu
r
es
ca
n
en
h
a
n
ce
p
er
f
o
r
m
a
n
ce
an
d
r
ed
u
ce
th
e
tim
e
r
e
q
u
ir
ed
to
p
r
e
d
ict
B
C
.
R
FE
r
em
o
v
es
th
e
f
ea
tu
r
es
th
at
h
av
e
th
e
m
o
s
t
n
eg
lig
ib
le
im
p
ac
t
o
n
th
e
m
o
d
el
’
s
ac
cu
r
ac
y
.
T
h
i
s
p
r
o
ce
s
s
is
r
ep
ea
ted
u
n
til
a
d
es
ir
ed
n
u
m
b
er
o
f
f
ea
tu
r
es
is
r
ea
ch
ed
.
T
h
is
p
ap
er
r
ed
u
ce
s
th
e
d
ataset
b
y
6
0
%
o
f
th
e
o
r
ig
in
al
d
ataset.
T
h
e
m
i
n
im
ized
d
ataset
is
r
ed
u
ce
d
to
1
8
attr
ib
u
tes af
ter
u
s
in
g
th
e
R
FE
tech
n
iq
u
e.
As
s
h
o
wn
in
Fig
u
r
e
1
,
th
is
p
ap
er
u
tili
ze
s
th
e
R
F
alg
o
r
ith
m
an
d
R
FE
to
ac
h
iev
e
th
e
d
e
s
ir
ed
r
esu
lt.
T
h
e
d
etailed
im
p
lem
en
tatio
n
s
tep
s
ar
e
d
escr
ib
ed
in
Alg
o
r
ith
m
1
.
T
h
e
R
FC
d
eter
m
in
es
th
e
r
esu
lt
b
y
co
n
s
id
er
in
g
t
h
e
o
u
t
p
u
ts
o
f
b
in
ar
y
tr
ee
s
.
T
h
e
R
FC
tech
n
iq
u
e
u
s
ed
in
th
is
p
a
p
er
ex
p
lain
s
h
o
w
to
f
in
d
th
e
ac
cu
r
ac
y
o
f
th
e
o
r
ig
in
al
d
ataset
an
d
class
if
y
it
in
to
two
s
ets:
a
tr
ain
in
g
s
et
an
d
a
test
in
g
s
et.
T
h
e
test
in
g
s
et
is
u
s
ed
f
o
r
p
er
f
o
r
m
in
g
test
s
,
wh
i
le
th
e
tr
ain
in
g
s
et
is
u
s
ed
to
tr
ain
th
e
s
y
s
tem
.
T
h
e
R
F
C
i
s
u
s
ed
o
n
ce
ag
ain
af
ter
R
FE
is
ap
p
lied
to
th
e
d
ataset.
R
F
C
p
er
f
o
r
m
s
th
e
s
am
e
o
p
er
at
io
n
o
n
t
h
e
r
ed
u
ce
d
f
ea
tu
r
e
d
at
aset.
Fig
u
r
e
1
.
I
m
p
lem
en
te
d
p
r
o
ce
s
s
Alg
o
r
ith
m
1
.
Pro
p
o
s
ed
R
FE
–
R
F
C
p
r
o
ce
d
u
r
e
f
o
r
f
ea
tu
r
e
s
elec
tio
n
an
d
ac
c
u
r
ac
y
ev
alu
atio
n
Step
1
: Beg
in
Step
2
:
Up
lo
ad
th
e
d
ataset
to
c
o
lab
an
d
lo
ad
th
e
d
ata.
Step
3
: D
iv
id
e
th
e
d
ataset
in
to
X
(
f
ea
tu
r
e)
an
d
Y
la
b
els (
tar
g
et)
.
Step
4
:
C
h
o
o
s
e
a
ML
alg
o
r
ith
m
(
R
FE)
f
o
r
FS
.
Step
5
:
T
h
e
en
tire
d
ata
is
th
en
r
an
k
ed
b
ased
o
n
th
e
im
p
o
r
tan
ce
o
f
th
e
attr
ib
u
tes
u
s
in
g
th
e
R
FE
alg
o
r
ith
m
.
T
h
e
f
ea
tu
r
es
ar
e
th
e
n
s
elec
ted
b
as
ed
o
n
a
f
ix
e
d
p
e
r
ce
n
tag
e
(
6
0
%),
an
d
th
e
least
im
p
o
r
tan
t
attr
ib
u
tes
ar
e
r
em
o
v
ed
b
ased
o
n
th
e
r
an
k
in
g
.
Step
6
: Rep
ea
t step
5
u
n
til th
e
d
esire
d
n
u
m
b
er
o
f
f
ea
tu
r
es is
s
elec
ted
.
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
A
ttr
ib
u
te
o
p
timis
a
tio
n
to
imp
r
o
ve
b
r
ea
s
t c
a
n
ce
r
p
r
ed
ictio
n
u
s
in
g
ma
ch
in
e
…
(
R
a
g
h
a
ve
n
d
r
a
S
r
in
iva
s
a
ia
h
)
1331
Step
7
:
Use th
e
d
esire
d
p
r
o
p
er
ties
to
ev
alu
ate
th
e
ac
cu
r
ac
y
o
f
th
e
R
FC
.
Step
8
: D
is
p
lay
th
e
ac
cu
r
ac
y
Step
9
: E
n
d
Acc
u
r
ac
y
c
alcu
latio
n
f
o
r
m
u
la
e
u
s
in
g
R
FC
:
test
s
et:
=
t
r
u
e
lab
els
_
=
p
r
ed
icted
lab
el
(
u
s
in
g
th
e
R
F)
Acc
u
r
ac
y
is
ca
lcu
lated
as sh
o
wn
in
(
1
)
an
d
(
2
)
.
=
×
100
(
1
)
=
∑
(
[
i
]
=
=
=
1
_
[
]
)
×
100
(
2
)
N=
to
tal
n
u
m
b
er
o
f
s
am
p
les.
R
FE
is
an
FS
tech
n
iq
u
e
th
at
s
elec
ts
a
s
u
b
s
et
o
f
r
elev
an
t f
ea
tu
r
es f
r
o
m
a
lar
g
e
r
d
ataset.
I
t w
ill id
en
tify
an
d
r
em
o
v
e
less
cr
itical
f
ea
tu
r
es
b
y
wo
r
k
in
g
with
an
L
R
m
o
d
el.
T
h
is
ap
p
r
o
ac
h
is
b
ased
o
n
a
g
r
ee
d
y
s
ea
r
ch
tech
n
iq
u
e.
Min
im
izatio
n
o
f
d
a
taset
u
s
in
g
R
FE:
Featu
r
e
s
elec
tio
n
:
−
L
et
X
b
e
th
e
o
r
ig
i
n
al
f
ea
tu
r
e
m
atr
ix
with
d
im
e
n
s
io
n
s
m
×
n
,
wh
e
r
e,
m
=
n
u
m
b
er
o
f
s
am
p
les
an
d
n
=
n
u
m
b
er
o
f
f
ea
tu
r
es.
−
C
alcu
latin
g
th
e
n
u
m
b
er
o
f
f
ea
t
u
r
es to
s
elec
t: n
u
m
_
f
ea
t
u
r
es_
t
o
_
s
elec
t
=
⌊
0
.
6
0
×
n
⌋
.
R
ed
u
ce
d
f
ea
tu
r
e
m
atr
ix
:
−
T
r
an
s
f
o
r
m
t
h
e
o
r
i
g
in
al
f
ea
tu
r
e
m
atr
ix
X
u
s
in
g
R
FE
:
=
r
f
e.
f
i
t_
tr
an
s
f
o
r
m
(
X,
Y)
.
=
th
e
r
ed
u
ce
d
f
ea
tu
r
e
m
atr
ix
with
d
im
en
s
io
n
s
m
×
n
u
m
_
f
ea
tu
r
es_
to
_
s
elec
t.
T
a
b
le
1
d
is
p
la
y
s
a
cc
u
r
ac
y
,
a
tt
r
ib
u
tes
,
an
d
th
e
t
im
e
ta
k
en
t
o
c
alc
u
l
ate
ac
cu
r
ac
y
i
n
s
ec
o
n
d
s
.
T
h
e
r
es
u
lts
s
h
o
wn
a
r
e
o
b
t
ai
n
e
d
f
r
o
m
th
e
cu
r
r
e
n
t
s
tu
d
y
u
s
i
n
g
R
FC
an
d
R
FE
te
c
h
n
iq
u
es.
T
h
e
o
r
i
g
i
n
a
l
d
at
ase
t
c
o
n
t
ai
n
s
3
2
att
r
i
b
u
tes
,
a
n
d
a
f
te
r
a
p
p
l
y
i
n
g
t
h
e
FS
m
e
th
o
d
,
th
e
n
u
m
b
e
r
o
f
a
ttri
b
u
t
es
was
r
e
d
u
c
ed
to
1
8
,
w
h
il
e
m
ai
n
ta
in
in
g
t
h
e
s
am
e
ac
c
u
r
ac
y
as
t
h
e
o
r
i
g
i
n
al
d
a
tase
t.
F
r
o
m
t
h
e
r
es
u
l
t
o
b
t
ai
n
e
d
t
h
e
m
o
s
t
ess
e
n
ti
al
att
r
i
b
u
t
es
i
d
e
n
ti
f
i
ed
a
r
e
as
f
o
ll
o
ws:
m
ea
n
r
a
d
i
u
s
,
m
e
an
te
x
t
u
r
e,
m
e
an
p
e
r
i
m
e
te
r
,
m
e
an
s
m
o
o
t
h
n
ess
,
m
ea
n
c
o
n
ca
v
i
ty
,
m
ea
n
c
o
n
ca
v
e
p
o
i
n
ts
,
m
ea
n
s
y
m
m
e
tr
y
,
te
x
t
u
r
e
e
r
r
o
r
,
a
r
e
a
e
r
r
o
r
,
c
o
n
c
av
it
y
e
r
r
o
r
,
w
o
r
s
t
r
ad
iu
s
,
w
o
r
s
t
t
e
x
t
u
r
e,
w
o
r
s
t
p
e
r
i
m
et
er
,
w
o
r
s
t
s
m
o
o
t
h
n
ess
,
wo
r
s
t
c
o
n
c
av
it
y
,
wo
r
s
t
c
o
n
ca
v
e
p
o
i
n
ts
,
w
o
r
s
t s
y
m
m
et
r
y
,
a
n
d
w
o
r
s
t
f
r
a
cta
l
d
i
m
en
s
i
o
n
.
T
ab
le
1
.
Dis
p
lay
s
ac
cu
r
ac
y
,
at
tr
ib
u
tes,
an
d
th
e
tim
e
tak
en
to
ca
lcu
late
ac
cu
r
ac
y
in
s
ec
o
n
d
s
D
a
t
a
s
e
t
Ti
me
t
a
k
e
n
(
i
n
se
c
o
n
d
s)
A
c
c
u
r
a
c
y
(
%)
O
r
i
g
i
n
a
l
d
a
t
a
s
e
t
0
.
2
6
9
6
.
4
9
R
e
d
u
c
e
d
d
a
t
a
se
t
(
6
0
%)
0
.
2
2
9
6
.
4
9
T
h
is
s
tu
d
y
u
tili
ze
s
th
e
p
u
b
licl
y
av
ailab
le
BC
W
is
co
n
s
in
(
d
i
ag
n
o
s
tic
)
d
ataset
f
r
o
m
UC
I
/Kag
g
le
[
3
6
]
,
wh
ich
co
n
tain
s
5
6
9
in
s
tan
ce
s
with
3
0
n
u
m
e
r
ic
p
r
e
d
ictiv
e
f
e
atu
r
es
d
er
iv
ed
f
r
o
m
d
ig
itized
f
in
e
-
n
ee
d
le
asp
ir
ate
(
FNA)
im
ag
es.
T
h
e
tar
g
et
v
a
r
iab
le
is
b
in
ar
y
:
m
alig
n
an
t
(
2
1
2
ca
s
es)
an
d
b
e
n
ig
n
(
3
5
7
ca
s
es).
T
h
er
e
ar
e
n
o
m
is
s
in
g
v
alu
es,
an
d
th
e
‘
id
’
co
lu
m
n
was
d
is
ca
r
d
ed
.
T
h
e
d
ataset
was
s
tr
atif
ied
an
d
s
p
lit
in
to
7
5
%
tr
ain
in
g
(
4
2
7
s
am
p
les)
an
d
2
5
%
test
in
g
(
1
4
2
s
am
p
les)
s
ets
u
s
in
g
test
_
s
ize
=0
.
2
5
,
s
tr
atif
y
=y
,
an
d
r
an
d
o
m
_
s
tate
=4
2
.
All
ex
p
er
im
en
ts
em
p
lo
y
ed
s
cik
it
-
lear
n
’
s
R
FC
with
th
e
f
o
llo
win
g
f
ix
ed
h
y
p
e
r
p
ar
am
eter
s
:
n
_
esti
m
ato
r
s
=1
0
0
,
m
a
x
_
d
e
p
t
h
=
n
o
n
e
,
m
i
n
_
s
a
m
p
l
e
s
_
s
p
li
t
=
2
,
m
i
n
_
s
a
m
p
le
s
_
le
a
f
=
1
,
c
l
a
s
s
_
w
e
i
g
h
t
=
‘
b
a
l
a
n
ce
d
’
,
r
a
n
d
o
m
_
s
t
a
t
e
=
4
2
.
R
FE
was
ap
p
lied
u
s
in
g
t
h
e
s
am
e
R
F
as
th
e
b
ase
esti
m
ato
r
,
n
_
f
ea
tu
r
es_
to
_
s
elec
t
=1
8
,
s
tep
=
1
,
a
n
d
s
co
r
in
g
=‘
ac
cu
r
ac
y
’
,
a
n
d
r
ed
u
ce
s
th
e
f
ea
tu
r
e
s
et
f
r
o
m
3
0
t
o
1
8
(
4
0
%
r
ed
u
ctio
n
)
.
All
r
ep
o
r
ted
r
esu
lts
ar
e
o
b
tain
ed
v
ia
5
-
f
o
ld
s
tr
atif
ied
cr
o
s
s
-
v
alid
atio
n
o
n
th
e
t
r
ain
in
g
s
et
(
Stra
tifie
d
KFo
ld
,
s
h
u
f
f
le
=
t
r
u
e
,
r
an
d
o
m
_
s
tate
=4
2
)
.
Fin
al
p
er
f
o
r
m
an
ce
o
n
th
e
h
eld
-
o
u
t
test
s
et
i
s
al
s
o
r
ep
o
r
ted
.
Me
tr
ic
s
in
clu
d
e
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
al
l,
F1
-
s
co
r
e,
s
p
e
cif
icity
,
an
d
ar
ea
u
n
d
e
r
th
e
cu
r
v
e
(
AUC
)
-
R
OC
(
m
ea
n
±
s
tan
d
ar
d
d
e
v
iatio
n
wh
er
e
ap
p
licab
le)
.
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
ab
le
2
s
h
o
ws
th
e
c
o
m
p
a
r
is
o
n
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
wit
h
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
o
th
e
r
ex
is
tin
g
r
esear
ch
u
s
in
g
th
e
d
i
f
f
er
en
t
FS
m
eth
o
d
s
o
n
th
e
d
ataset
in
ter
m
s
o
f
th
e
n
u
m
b
er
o
f
attr
i
b
u
tes
u
s
ed
an
d
th
e
ac
cu
r
ac
y
ac
h
iev
e
d
.
I
t
ca
n
b
e
o
b
s
er
v
ed
th
at
th
e
ex
is
tin
g
FS
m
eth
o
d
s
ac
h
iev
e
b
etter
ac
cu
r
ac
y
with
f
ewe
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
2
,
Ap
r
il
20
26
:
1
3
2
7
-
1
3
3
8
1332
attr
ib
u
tes
co
m
p
ar
e
d
to
th
e
p
r
o
p
o
s
ed
m
eth
o
d
;
h
o
wev
er
,
t
h
e
e
x
is
tin
g
tech
n
iq
u
es
d
o
n
o
t
f
o
c
u
s
o
n
th
e
tim
e
tak
e
n
f
o
r
p
r
ed
ictio
n
.
T
ab
le
3
d
ep
icts
th
e
h
y
p
er
p
ar
a
m
eter
s
an
d
e
x
p
e
r
im
en
tal
s
ettin
g
s
.
T
ab
le
2
.
Per
f
o
r
m
an
ce
co
m
p
a
r
is
o
n
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
w
ith
ex
is
tin
g
m
eth
o
d
s
b
ased
o
n
s
elec
ted
attr
ib
u
tes
FS
met
h
o
d
N
u
mb
e
r
o
f
f
e
a
t
u
r
e
s se
l
e
c
t
e
d
A
c
c
u
r
a
c
y
P
S
O
[
3
6
]
12
9
9
.
8
2
M
o
d
i
f
i
e
d
b
a
t
a
l
g
o
r
i
t
h
m
[
3
7
]
10
9
8
.
7
0
G
A
[
2
8]
9
-
11
9
7
.
1
3
R
EF [
3
8
]
8
9
7
.
5
T
ab
le
3
.
Hy
p
er
p
ar
a
m
eter
s
an
d
ex
p
er
im
en
tal
s
ettin
g
s
C
o
m
p
o
n
e
n
t
S
e
t
t
i
n
g
D
a
t
a
s
e
t
BC
W
i
sc
o
n
si
n
(
d
i
a
g
n
o
st
i
c
)
,
5
6
9
×
3
0
Tr
a
i
n
/
t
e
s
t
s
p
l
i
t
7
5
%/
2
5
%,
st
r
a
t
i
f
i
e
d
,
r
a
n
d
o
m_
s
t
a
t
e
=
4
2
C
r
o
ss
-
v
a
l
i
d
a
t
i
o
n
5
-
f
o
l
d
s
t
r
a
t
i
f
i
e
d
RF
n
_
e
st
i
ma
t
o
r
s
=
1
0
0
,
c
l
a
ss
_
w
e
i
g
h
t
=
'b
a
l
a
n
c
e
d
',
r
a
n
d
o
m_
s
t
a
t
e
=
4
2
R
F
E
e
st
i
mat
o
r
=
R
a
n
d
o
mF
o
r
e
st
,
n
_
f
e
a
t
u
r
e
s
_
t
o
_
sel
e
c
t
=
1
8
,
s
t
e
p
=
1
,
s
c
o
r
i
n
g
=
a
c
c
u
r
a
c
y
S
o
f
t
w
a
r
e
P
y
t
h
o
n
3
.
1
2
,
s
c
i
k
i
t
-
l
e
a
r
n
1
.
5
.
0
,
p
a
n
d
a
s 2
.
2
,
N
u
mP
y
1
.
2
6
Fig
u
r
e
2
f
ac
ilit
ates
a
co
m
p
ar
is
o
n
o
f
ac
cu
r
ac
y
,
th
e
n
u
m
b
e
r
o
f
f
ea
tu
r
es,
an
d
th
e
tim
e
r
eq
u
ir
ed
to
co
m
p
u
te
ac
cu
r
ac
y
f
o
r
b
o
th
th
e
o
r
ig
in
al
an
d
o
p
tim
ized
d
ata
s
ets.
I
n
Fig
u
r
e
2
,
th
e
b
lu
e
b
a
r
d
en
o
tes
th
e
tim
e
tak
en
b
y
t
h
e
o
r
i
g
in
al
d
ataset
an
d
th
e
o
p
tim
ized
d
ataset,
wh
il
e
th
e
m
ar
o
o
n
b
a
r
s
h
o
ws
th
e
n
u
m
b
er
o
f
attr
ib
u
tes
co
n
s
id
er
ed
b
y
th
e
o
r
i
g
in
al
a
n
d
th
e
o
p
tim
ize
d
d
ataset,
an
d
t
h
e
g
r
ee
n
b
ar
s
h
o
ws
th
e
ac
cu
r
ac
y
ac
h
ie
v
ed
b
y
th
e
o
r
ig
in
al
a
n
d
t
h
e
o
p
tim
ized
d
at
aset.
R
ed
u
cin
g
th
e
n
u
m
b
er
o
f
attr
ib
u
tes
in
th
e
o
p
tim
ized
d
at
aset
m
ain
tain
s
th
e
s
am
e
ac
cu
r
ac
y
as
th
e
o
r
ig
in
al
d
ataset
b
u
t
d
im
in
is
h
es
th
e
tim
e
tak
en
to
ca
lcu
late
ac
cu
r
ac
y
.
Sp
ec
if
ically
,
with
a
4
0
%
r
ed
u
ctio
n
i
n
th
e
d
ataset
(
attr
ib
u
tes
r
ed
u
ce
d
f
r
o
m
3
2
to
1
8
)
,
th
e
tim
e
ta
k
en
t
o
f
in
d
th
e
ac
cu
r
ac
y
d
ec
r
ea
s
es
to
3
u
n
its
co
m
p
ar
ed
to
th
e
o
r
ig
in
al
d
ataset.
Fig
u
r
e
2
.
C
o
m
p
a
r
in
g
o
r
ig
in
al
an
d
o
p
tim
ize
d
d
ataset
ac
cu
r
ac
y
,
attr
ib
u
tes
,
an
d
tim
e
T
h
u
s
,
we
m
ay
d
ed
u
ce
th
at
th
e
tim
e
n
ee
d
ed
to
f
o
r
ec
ast
B
C
d
ec
r
ea
s
es
in
p
r
o
p
o
r
tio
n
to
th
e
n
u
m
b
er
o
f
q
u
alities
.
I
n
ad
d
itio
n
to
h
elp
i
n
g
p
atien
ts
b
y
cu
ttin
g
d
o
w
n
o
n
test
len
g
th
,
tr
ea
tm
en
t tim
e,
ex
p
en
s
es,
an
d
waitin
g
tim
es
f
o
r
test
r
esu
lts
,
th
i
s
ti
m
e
r
ed
u
ctio
n
also
h
el
p
s
lo
wer
B
C
’
s
o
v
er
all
d
ea
th
r
ate.
I
n
co
n
clu
s
io
n
,
f
ewe
r
ch
ar
ac
ter
is
tics
r
esu
lt
in
a
m
o
r
e
s
u
cc
ess
f
u
l
p
r
ed
ictio
n
p
r
o
ce
d
u
r
e,
g
u
ar
an
teein
g
p
r
o
m
p
t
an
d
ef
f
icien
t
m
ed
ical
ca
r
e
f
o
r
p
atien
ts
an
d
e
v
en
tu
all
y
h
elp
in
g
to
m
a
n
ag
e
an
d
lo
we
r
th
e
m
o
r
tality
r
ate
r
elate
d
to
B
C
.
T
h
e
ac
cu
r
ac
y
o
f
th
e
s
u
g
g
ested
ap
p
r
o
ac
h
an
d
th
e
n
u
m
b
er
o
f
attr
ib
u
tes
em
p
lo
y
ed
in
p
r
ed
ictio
n
ar
e
co
n
tr
asted
with
th
o
s
e
o
f
o
th
e
r
ap
p
r
o
ac
h
es
in
Fig
u
r
e
3
.
Fig
u
r
e
3
illu
s
tr
ates
h
o
w
s
o
m
e
cu
r
r
en
t
ap
p
r
o
ac
h
es,
wh
ich
m
ak
e
u
s
e
o
f
th
e
en
tire
c
o
llectio
n
o
f
f
ea
tu
r
es,
p
er
f
o
r
m
b
etter
th
an
th
e
s
u
g
g
ested
ap
p
r
o
ac
h
.
T
h
e
ac
cu
r
a
cy
ac
h
iev
ed
b
y
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
is
n
ea
r
ly
th
at
o
f
th
e
e
x
is
tin
g
m
et
h
o
d
s
,
an
d
it
u
s
es
o
n
ly
1
8
attr
ib
u
tes,
wh
ich
is
ap
p
r
o
x
im
ately
6
0
%
o
f
th
e
to
tal
attr
ib
u
tes.
I
n
th
is
p
ap
er
,
th
e
p
r
ed
ictio
n
tim
e
is
r
ed
u
ce
d
,
wh
ich
will
also
d
ec
r
ea
s
e
th
e
tim
e
r
eq
u
ir
ed
f
o
r
test
in
g
(
a
n
aly
zin
g
ca
n
ce
r
ce
lls
)
,
as
well
as
th
e
tim
e
s
p
en
t
o
n
test
r
esu
lts
an
d
m
ed
ical
co
s
ts
.
26
22
32
18
9
6
.
4
9
9
6
.
4
9
0
20
40
60
80
1
0
0
1
2
0
Or
i
g
i
n
a
l
Dat
a
se
t
O
p
t
i
m
i
sed
D
a
t
a
set
V
a
l
u
e
D
a
t
a
se
t
t
y
p
e
Ti
me
N
o
.
o
f
A
t
t
r
i
b
u
t
e
s
A
c
c
u
r
a
c
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
A
ttr
ib
u
te
o
p
timis
a
tio
n
to
imp
r
o
ve
b
r
ea
s
t c
a
n
ce
r
p
r
ed
ictio
n
u
s
in
g
ma
ch
in
e
…
(
R
a
g
h
a
ve
n
d
r
a
S
r
in
iva
s
a
ia
h
)
1333
Fig
u
r
e
4
p
r
esen
ts
th
e
o
v
er
all
ev
alu
atio
n
o
f
th
e
R
FE
-
o
p
tim
ized
R
F
m
o
d
el,
s
h
o
win
g
h
ig
h
s
co
r
es
f
o
r
ac
cu
r
ac
y
(
9
6
.
4
9
%),
p
r
ec
is
io
n
(
9
5
.
8
9
%),
r
ec
all
(
9
8
.
5
9
%)
,
a
n
d
F1
-
s
co
r
e
(
9
7
.
2
2
%).
T
h
ese
r
esu
lts
co
n
f
ir
m
th
at
th
e
1
8
-
f
ea
tu
r
e
m
o
d
el
p
r
o
v
id
e
s
r
eliab
le
an
d
clin
ically
m
ea
n
in
g
f
u
l
d
ia
g
n
o
s
tic
p
er
f
o
r
m
an
ce
.
T
h
e
co
n
f
u
s
io
n
m
atr
ix
an
d
p
er
f
o
r
m
a
n
ce
m
etr
i
cs
b
ar
ch
ar
t
in
Fig
u
r
e
5
s
h
o
w
co
n
s
is
ten
tly
h
ig
h
v
alu
es
f
o
r
a
cc
u
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e
.
L
o
w
m
is
class
if
icatio
n
r
ates
ar
e
s
h
o
wn
in
th
e
f
ig
u
r
e'
s
co
n
f
u
s
io
n
m
at
r
ix
,
wh
ich
d
is
p
lay
s
o
n
ly
o
n
e
f
alse
n
e
g
ativ
e
an
d
th
r
ee
f
alse
p
o
s
itiv
es.
Ov
er
all,
Fi
g
u
r
e
5
s
h
o
ws
th
at
f
o
r
b
o
th
b
e
n
ig
n
an
d
m
alig
n
an
t
in
s
tan
ce
s
,
th
e
m
o
d
el
o
f
f
er
s
co
n
s
is
ten
t a
n
d
r
eliab
le
ca
teg
o
r
iz
atio
n
.
Fig
u
r
e
3
.
C
o
m
p
a
r
in
g
p
r
o
p
o
s
ed
m
eth
o
d
with
e
x
is
tin
g
m
eth
o
d
s
Fig
u
r
e
4
.
P
r
o
p
o
s
ed
m
o
d
el
-
p
e
r
f
o
r
m
an
ce
m
et
r
ics
Fig
u
r
e
5
.
Pro
p
o
s
ed
m
o
d
el
-
co
n
f
u
s
io
n
m
atr
ix
9
8
.
8
3
9
8
.
5
7
94
9
8
.
0
7
9
7
.
2
8
9
.
5
9
6
.
4
9
32
32
32
32
32
32
18
0
20
40
60
80
1
0
0
1
2
0
En
sem
b
l
e
me
t
h
o
d
[
1
5
]
A
N
N
[
1
6
]
F
u
z
z
y
S
V
M
[
1
8
]
En
sem
b
l
e
C
l
a
ssi
f
i
e
r
s
[
1
9
]
P
S
O
[
2
8
]
M
R
[
2
9
]
P
r
o
p
o
sed
M
e
t
h
o
d
A
c
c
u
r
a
c
y
M
e
t
h
o
d
A
c
c
u
r
a
c
y
No
.
o
f
Att
r
i
b
u
t
e
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
2
,
Ap
r
il
20
26
:
1
3
2
7
-
1
3
3
8
1334
E
x
ce
llen
t
s
en
s
itiv
ity
with
f
e
w
f
alse
p
o
s
itiv
es
is
in
d
icate
d
b
y
Fig
u
r
e
6
R
OC
cu
r
v
e,
wh
ich
r
is
es
s
tr
o
n
g
ly
in
th
e
u
p
p
er
-
lef
t
ar
ea
.
T
h
e
f
ig
u
r
e
’
s
AUC
s
co
r
e
o
f
0
.
9
9
6
in
d
icate
s
th
at
th
er
e
is
alm
o
s
t
p
er
f
ec
t
d
is
cr
im
in
atio
n
b
etwe
en
th
e
tw
o
g
r
o
u
p
s
.
T
h
u
s
,
Fig
u
r
e
6
v
alid
ates
th
at
f
ea
tu
r
e
r
ed
u
cti
o
n
th
r
o
u
g
h
R
FE
d
o
es
n
o
t
co
m
p
r
o
m
is
e
m
o
d
el
p
er
f
o
r
m
an
ce
.
Fig
u
r
e
7
p
r
esen
ts
th
e
p
r
ec
is
io
n
–
re
ca
ll
cu
r
v
e
,
wh
ich
m
ain
tain
s
h
ig
h
p
r
ec
is
io
n
ac
r
o
s
s
n
ea
r
ly
th
e
e
n
tire
r
ec
all
r
an
g
e.
T
h
e
c
u
r
v
e
clea
r
ly
s
h
o
ws
an
av
er
ag
e
p
r
ec
is
io
n
o
f
0
.
9
9
7
,
s
ig
n
if
y
in
g
m
in
im
al
f
alse
-
p
o
s
itiv
e
p
r
e
d
ictio
n
s
.
As
d
e
p
icted
i
n
Fig
u
r
e
7
,
th
e
m
o
d
el
p
er
f
o
r
m
s
r
eliab
ly
e
v
en
i
n
th
e
p
r
esen
ce
o
f
class
im
b
alan
ce
,
m
ak
in
g
it su
itab
le
f
o
r
m
ed
ic
al
d
iag
n
o
s
tic
task
s
.
Fig
u
r
e
6
.
Pro
p
o
s
ed
m
o
d
el
-
R
OC
cu
r
v
e
Fig
u
r
e
7
.
Pro
p
o
s
ed
m
o
d
el
-
PR
cu
r
v
e
T
h
e
1
8
f
ea
tu
r
es
r
etain
ed
b
y
R
FE+
R
F
ar
e:
m
ea
n
r
ad
iu
s
,
m
ea
n
tex
tu
r
e,
m
ea
n
p
e
r
im
eter
,
m
ea
n
ar
ea
,
m
ea
n
s
m
o
o
th
n
ess
,
m
ea
n
co
n
c
av
ity
,
m
ea
n
co
n
ca
v
e
p
o
in
ts
,
m
ea
n
s
y
m
m
etr
y
,
m
ea
n
f
r
ac
tal
d
im
en
s
io
n
,
r
ad
iu
s
er
r
o
r
,
te
x
tu
r
e
er
r
o
r
,
ar
ea
er
r
o
r
,
wo
r
s
t
r
ad
iu
s
,
wo
r
s
t
tex
tu
r
e,
wo
r
s
t
p
er
im
eter
,
wo
r
s
t
ar
ea
,
wo
r
s
t
s
m
o
o
th
n
ess
,
an
d
wo
r
s
t
co
n
ca
v
ity
.
T
h
ese
alig
n
clo
s
ely
with
clin
ically
in
ter
p
r
etab
le
m
o
r
p
h
o
lo
g
ica
l
d
escr
ip
to
r
s
.
T
h
e
co
m
p
ar
is
o
n
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el
with
ex
is
tin
g
m
o
d
els
is
s
h
o
wn
in
Fig
u
r
e
8
.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
ac
h
iev
es
ac
cu
r
ac
y
alm
o
s
t
as
h
ig
h
as
s
o
m
e
o
f
t
h
e
ex
is
tin
g
m
eth
o
d
s
,
b
u
t
with
o
n
ly
1
8
f
ea
tu
r
es.
T
h
e
tr
ain
in
g
ac
cu
r
ac
y
was
9
9
.
9
%
–
1
0
0
%,
wh
ile
th
e
5
-
f
o
ld
C
V
ac
cu
r
ac
y
was
9
6
.
4
3
%
(
g
ap
<0
.
6
%),
co
n
f
ir
m
in
g
m
in
im
al
o
v
e
r
f
itti
n
g
.
T
h
e
lear
n
in
g
c
u
r
v
es
ar
e
s
h
o
w
n
in
Fig
u
r
e
9
.
T
h
e
im
p
o
r
ta
n
ce
o
f
ea
ch
f
ea
tu
r
e
a
n
d
its
co
n
tr
i
b
u
tio
n
to
th
e
o
v
er
all
ac
cu
r
ac
y
is
s
h
o
wn
in
Fig
u
r
e
1
0
.
Fig
u
r
e
8
.
C
o
m
p
a
r
is
o
n
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el
a
n
d
ex
is
tin
g
m
o
d
els
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
A
ttr
ib
u
te
o
p
timis
a
tio
n
to
imp
r
o
ve
b
r
ea
s
t c
a
n
ce
r
p
r
ed
ictio
n
u
s
in
g
ma
ch
in
e
…
(
R
a
g
h
a
ve
n
d
r
a
S
r
in
iva
s
a
ia
h
)
1335
Fig
u
r
e
9
.
L
ea
r
n
in
g
cu
r
v
es
Fig
u
r
e
1
0
.
1
8
f
ea
tu
r
es selecte
d
b
y
R
FE+
R
F
5.
CO
NCLU
SI
O
N
C
o
n
d
u
c
ti
n
g
m
o
r
e
ac
c
u
r
a
te
p
r
e
d
ic
ti
o
n
s
f
o
r
B
C
,
th
is
r
ese
a
r
c
h
u
n
d
er
li
n
es
th
e
i
m
p
o
r
t
an
ce
o
f
m
i
n
im
izi
n
g
th
e
ch
ar
ac
t
e
r
is
ti
cs
o
f
t
h
e
d
atas
et.
T
h
is
s
t
u
d
y
h
e
lp
s
i
n
d
esi
g
n
i
n
g
an
a
cc
u
r
at
e,
ef
f
i
cie
n
t
,
a
n
d
f
r
ie
n
d
l
y
p
r
e
d
ic
ti
o
n
m
o
d
el
f
o
r
p
at
ie
n
ts
s
u
f
f
e
r
i
n
g
f
r
o
m
B
C
b
y
ap
p
l
y
i
n
g
R
F
E
t
o
i
d
en
ti
f
y
t
h
e
m
o
s
t
i
m
p
o
r
t
an
t
c
h
ar
ac
t
er
is
ti
cs.
Min
im
i
zi
n
g
c
h
a
r
a
cte
r
is
tics
to
6
0
%
h
el
p
s
i
n
i
n
c
r
e
asi
n
g
ac
c
u
r
ac
y
,
ac
ce
l
er
ati
n
g
tes
t
r
es
u
lts
,
a
n
d
p
r
o
v
id
in
g
b
e
tte
r
tr
e
at
m
e
n
t c
h
o
ices
f
o
r
p
a
tie
n
ts
wit
h
B
C
.
T
h
is
ap
p
r
o
ac
h
h
el
p
s
p
at
ie
n
ts
g
ai
n
a
cc
ess
t
o
m
o
r
e
a
p
p
r
o
p
r
iat
e
t
r
e
at
m
e
n
t
ch
o
i
ce
s
a
n
d
co
u
l
d
r
ed
u
ce
t
h
e
d
ea
t
h
to
ll
d
u
e
t
o
B
C
,
p
ar
tic
u
l
ar
l
y
in
t
h
e
p
r
eli
m
i
n
a
r
y
s
ta
g
es
.
ACK
NO
WL
E
DG
M
E
N
T
S
W
e
wo
u
ld
lik
e
to
th
an
k
C
HR
I
ST
Un
iv
er
s
ity
f
o
r
s
u
p
p
o
r
tin
g
u
s
in
co
m
p
letin
g
th
is
wo
r
k
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
s
tate
n
o
f
u
n
d
in
g
in
v
o
lv
ed
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
2
,
Ap
r
il
20
26
:
1
3
2
7
-
1
3
3
8
1336
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
R
a
g
h
a
v
e
n
d
r
a
S
r
i
n
i
v
a
s
a
i
a
h
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
S
a
n
t
o
s
h
K
u
m
a
r
J
a
n
k
a
t
t
i
✓
✓
✓
✓
✓
✓
✓
✓
Nir
an
jan
a
Sh
r
av
an
ab
elag
o
la
J
in
ac
h
an
d
r
a
✓
✓
✓
✓
✓
✓
✓
Ma
n
ju
n
ath
R
am
an
n
a
L
am
an
i
✓
✓
✓
✓
✓
B
ellam
Vijay
a
L
ak
s
h
m
i
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
R
is
h
ita
B
h
elwa
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
t
h
at
s
u
p
p
o
r
t
th
e
f
in
d
in
g
s
o
f
th
i
s
s
tu
d
y
will
b
e
a
v
a
ilab
le
in
h
ttp
s
://www.
k
ag
g
le.
co
m
/d
atasets
/u
cim
l/b
r
ea
s
t
-
ca
n
ce
r
-
wis
co
n
s
in
-
d
ata
.
RE
F
E
R
E
NC
E
S
[
1
]
S
.
A
s
i
f
e
t
a
l
.
,
“
A
d
v
a
n
c
e
m
e
n
t
s
a
n
d
p
r
o
s
p
e
c
t
s
o
f
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
i
n
m
e
d
i
c
a
l
d
i
a
g
n
o
s
t
i
c
s
:
u
n
v
e
i
l
i
n
g
t
h
e
f
u
t
u
r
e
o
f
d
i
a
g
n
o
s
t
i
c
p
r
e
c
i
s
i
o
n
,
”
A
r
c
h
i
v
e
s
o
f
C
o
m
p
u
t
a
t
i
o
n
a
l
M
e
t
h
o
d
s
i
n
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
3
2
,
n
o
.
2
,
p
p
.
8
5
3
–
8
8
3
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
8
3
1
-
0
2
4
-
1
0
1
4
8
-
w.
[
2
]
I
.
S
e
t
h
e
t
a
l
.
,
“
U
se
o
f
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
i
n
b
r
e
a
s
t
s
u
r
g
e
r
y
:
a
n
a
r
r
a
t
i
v
e
r
e
v
i
e
w
,
”
G
l
a
n
d
S
u
rg
e
ry
,
v
o
l
.
1
3
,
n
o
.
3
,
p
p
.
3
9
5
–
4
1
1
,
2
0
2
4
,
d
o
i
:
1
0
.
2
1
0
3
7
/
g
s
-
23
-
4
1
4
.
[
3
]
V
.
La
h
o
u
r
a
e
t
a
l
.
,
“
C
l
o
u
d
c
o
mp
u
t
i
n
g
-
b
a
s
e
d
f
r
a
m
e
w
o
r
k
f
o
r
b
r
e
a
st
c
a
n
c
e
r
d
i
a
g
n
o
s
i
s
u
si
n
g
e
x
t
r
e
m
e
l
e
a
r
n
i
n
g
ma
c
h
i
n
e
,
”
D
i
a
g
n
o
st
i
c
s
,
v
o
l
.
1
1
,
n
o
.
2
,
F
e
b
.
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
d
i
a
g
n
o
s
t
i
c
s
1
1
0
2
0
2
4
1
.
[
4
]
T.
K
h
a
t
e
r
e
t
a
l
.
,
“
A
n
e
x
p
l
a
i
n
a
b
l
e
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
mo
d
e
l
f
o
r
t
h
e
c
l
a
ssi
f
i
c
a
t
i
o
n
o
f
b
r
e
a
st
c
a
n
c
e
r
,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
1
3
,
p
p
.
5
6
1
8
–
5
6
3
3
,
2
0
2
5
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
3
.
3
3
0
8
4
4
6
.
[
5
]
Y
.
S
u
p
r
i
y
a
a
n
d
R
.
C
h
e
n
g
o
d
e
n
,
“
B
r
e
a
s
t
c
a
n
c
e
r
p
r
e
d
i
c
t
i
o
n
u
si
n
g
sh
a
p
e
l
y
a
n
d
g
a
m
e
t
h
e
o
r
y
i
n
f
e
d
e
r
a
t
e
d
l
e
a
r
n
i
n
g
e
n
v
i
r
o
n
me
n
t
,
”
I
E
EE
Ac
c
e
ss
,
v
o
l
.
1
2
,
p
p
.
1
2
3
0
1
8
–
1
2
3
0
3
7
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
4
.
3
4
2
4
9
3
4
.
[
6
]
A
.
S
h
a
r
ma
a
n
d
P
.
K
.
M
i
sh
r
a
,
“
P
e
r
f
o
r
man
c
e
a
n
a
l
y
si
s
o
f
mac
h
i
n
e
l
e
a
r
n
i
n
g
b
a
sed
o
p
t
i
mi
z
e
d
f
e
a
t
u
r
e
sel
e
c
t
i
o
n
a
p
p
r
o
a
c
h
e
s
f
o
r
b
r
e
a
st
c
a
n
c
e
r
d
i
a
g
n
o
si
s,”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
4
,
n
o
.
4
,
p
p
.
1
9
4
9
–
1
9
6
0
,
Ju
n
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s4
1
8
7
0
-
021
-
0
0
6
7
1
-
5.
[
7
]
S
.
Li
u
,
H
.
L
i
u
,
S
.
F
a
n
,
L
.
S
o
n
g
,
a
n
d
Z.
W
a
n
g
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
e
n
a
b
l
e
s
l
e
g
a
l
r
i
s
k
a
ssess
me
n
t
i
n
i
n
t
e
r
n
e
t
h
e
a
l
t
h
c
a
r
e
u
s
i
n
g
H
I
P
A
A
d
a
t
a
,
”
S
c
i
e
n
t
i
f
i
c
Re
p
o
r
t
s
,
v
o
l
.
1
5
,
n
o
.
1
,
A
u
g
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
8
-
0
2
5
-
1
3
7
2
0
-
x.
[
8
]
T.
J.
P
a
d
a
mse
e
e
t
a
l
.
,
“
R
i
sk
-
m
a
n
a
g
e
men
t
d
e
c
i
si
o
n
-
ma
k
i
n
g
d
a
t
a
f
r
o
m
a
c
o
mm
u
n
i
t
y
-
b
a
se
d
s
a
mp
l
e
o
f
r
a
c
i
a
l
l
y
d
i
v
e
r
se
w
o
m
e
n
a
t
h
i
g
h
r
i
s
k
o
f
b
r
e
a
st
c
a
n
c
e
r
:
r
a
t
i
o
n
a
l
e
,
met
h
o
d
s,
a
n
d
s
a
mp
l
e
c
h
a
r
a
c
t
e
r
i
s
t
i
c
s
o
f
t
h
e
d
a
u
g
h
t
e
r
s
i
st
e
r
m
o
t
h
e
r
p
r
o
j
e
c
t
su
r
v
e
y
,
”
Br
e
a
s
t
C
a
n
c
e
r R
e
se
a
rc
h
,
v
o
l
.
2
6
,
n
o
.
1
,
J
a
n
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
8
6
/
s1
3
0
5
8
-
0
2
3
-
0
1
7
5
3
-
x.
[
9
]
H
.
A
l
j
u
a
i
d
,
N
.
A
l
t
u
r
k
i
,
N
.
A
l
s
u
b
a
i
e
,
L.
C
a
v
a
l
l
a
r
o
,
a
n
d
A
.
L
i
o
t
t
a
,
“
C
o
m
p
u
t
e
r
-
a
i
d
e
d
d
i
a
g
n
o
s
i
s
f
o
r
b
r
e
a
s
t
c
a
n
c
e
r
c
l
a
ssi
f
i
c
a
t
i
o
n
u
s
i
n
g
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
s
a
n
d
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
,
”
C
o
m
p
u
t
e
r
M
e
t
h
o
d
s
a
n
d
Pro
g
r
a
m
s
i
n
Bi
o
m
e
d
i
c
i
n
e
,
v
o
l
.
2
2
3
,
A
u
g
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
mp
b
.
2
0
2
2
.
1
0
6
9
5
1
.
[
1
0
]
M
.
K
.
D
.
M
e
n
o
n
a
n
d
J
.
R
o
d
r
i
g
u
e
s,
“
Ef
f
i
c
i
e
n
t
u
l
t
r
a
w
i
d
e
b
a
n
d
r
a
d
a
r
b
a
se
d
n
o
n
i
n
v
a
si
v
e
e
a
r
l
y
b
r
e
a
st
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
1
1
,
p
p
.
8
4
2
1
4
–
8
4
2
2
7
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
3
.
3
3
0
3
3
3
3
.
[
1
1
]
S
.
C
h
a
u
d
h
u
r
y
e
t
a
l
.
,
“
Ef
f
e
c
t
i
v
e
i
ma
g
e
p
r
o
c
e
ss
i
n
g
a
n
d
s
e
g
m
e
n
t
a
t
i
o
n
-
b
a
s
e
d
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
f
o
r
d
i
a
g
n
o
s
i
s
o
f
b
r
e
a
st
c
a
n
c
e
r
,
”
C
o
m
p
u
t
a
t
i
o
n
a
l
a
n
d
M
a
t
h
e
m
a
t
i
c
a
l
M
e
t
h
o
d
s
i
n
M
e
d
i
c
i
n
e
,
v
o
l
.
2
0
2
2
,
p
p
.
1
–
6
,
A
p
r
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
2
/
6
8
4
1
3
3
4
.
[
1
2
]
M
.
L.
M
a
r
i
n
o
v
i
c
h
e
t
a
l
.
,
“
A
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
(
AI
)
f
o
r
b
r
e
a
s
t
c
a
n
c
e
r
s
c
r
e
e
n
i
n
g
:
b
r
e
a
s
t
scr
e
e
n
p
o
p
u
l
a
t
i
o
n
-
b
a
se
d
c
o
h
o
r
t
st
u
d
y
o
f
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
,
”
e
Bi
o
M
e
d
i
c
i
n
e
,
v
o
l
.
9
0
,
A
p
r
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
b
i
o
m.
2
0
2
3
.
1
0
4
4
9
8
.
[
1
3
]
U
.
N
a
see
m
e
t
a
l
.
,
“
A
n
a
u
t
o
m
a
t
i
c
d
e
t
e
c
t
i
o
n
o
f
b
r
e
a
s
t
c
a
n
c
e
r
d
i
a
g
n
o
si
s
a
n
d
p
r
o
g
n
o
si
s b
a
s
e
d
o
n
mac
h
i
n
e
l
e
a
r
n
i
n
g
u
si
n
g
e
n
sem
b
l
e
o
f
c
l
a
ss
i
f
i
e
r
s,”
I
EEE
Ac
c
e
ss
,
v
o
l
.
1
0
,
p
p
.
7
8
2
4
2
–
7
8
2
5
2
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
2
.
3
1
7
4
5
9
9
.
[
1
4
]
L
.
S
a
n
g
,
Z
.
L
i
u
,
C
.
H
u
a
n
g
,
J
.
X
u
,
a
n
d
H
.
W
a
n
g
,
“
M
u
l
t
i
p
a
r
a
m
e
t
r
i
c
M
R
I
-
b
a
s
e
d
r
a
d
i
o
m
i
c
s
n
o
m
o
g
r
a
m
f
o
r
p
r
e
d
i
c
t
i
n
g
t
h
e
h
o
r
m
o
n
e
r
e
c
e
p
t
o
r
s
t
a
t
u
s
o
f
h
e
r
2
-
p
o
s
i
t
i
v
e
b
r
e
a
s
t
c
a
n
c
e
r
,
”
C
l
i
n
i
c
a
l
R
a
d
i
o
l
o
g
y
,
v
o
l
.
7
9
,
n
o
.
1
,
p
p
.
6
0
–
6
6
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
r
a
d
.
2
0
2
3
.
0
9
.
0
1
3
.
[
1
5
]
B
.
O
.
M
a
c
a
u
l
a
y
,
B
.
S
.
A
r
i
b
i
s
a
l
a
,
S
.
A
.
A
k
a
n
d
e
,
B
.
A
.
A
k
i
n
n
u
w
e
s
i
,
a
n
d
O
.
A
.
O
l
a
b
a
n
j
o
,
“
B
r
e
a
s
t
c
a
n
c
e
r
r
i
s
k
p
r
e
d
i
c
t
i
o
n
i
n
A
f
r
i
c
a
n
w
o
m
e
n
u
s
i
n
g
r
a
n
d
o
m
f
o
r
e
s
t
c
l
a
s
s
i
f
i
e
r
,
”
C
a
n
c
e
r
T
r
e
a
t
m
e
n
t
a
n
d
R
e
s
e
a
r
c
h
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
v
o
l
.
2
8
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
t
a
r
c
.
2
0
2
1
.
1
0
0
3
9
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.