I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
i
neer
ing
a
nd
Co
m
p
u
t
er
Science
Vo
l.
22
,
No
.
2
,
Ma
y
2
0
2
1
,
p
p
.
1078
~
1
0
8
6
I
SS
N:
2
5
02
-
4
7
5
2
,
DOI
: 1
0
.
1
1
5
9
1
/i
j
ee
cs.v
2
2
.i
2
.
pp
1
0
7
8
-
1
0
8
6
1078
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs.ia
esco
r
e.
co
m
T
ra
nsfer learning
w
ith
G
o
o
g
LeNet
for
det
ection o
f
lu
ng
cancer
M
ua
y
ed
S AL
-
H
us
einy
,
Ah
m
e
d S Sa
j
it
Co
ll
e
g
e
o
f
En
g
in
e
e
rin
g
,
Un
iv
e
rsit
y
o
f
Was
it
,
Ira
q
(Uo
W
),
Ira
q
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
a
n
6
,
2
0
2
1
R
ev
i
s
ed
Mar
7
,
2
0
2
1
A
cc
ep
ted
Ma
r
2
5
,
2
0
2
1
T
h
e
u
s
e
o
f
c
o
m
p
u
t
e
r
a
lg
o
r
i
t
h
m
s
h
a
s
g
a
i
n
e
d
m
o
m
e
n
t
u
m
i
n
f
i
l
l
i
n
g
/a
ss
i
s
t
i
n
g
r
o
l
e
s
o
f
s
p
e
c
i
a
l
is
t
s
e
s
p
e
c
i
a
l
ly
i
n
e
a
r
ly
d
i
a
g
n
o
s
i
s
sc
e
n
a
r
i
o
s
.
T
h
i
s
p
a
p
e
r
p
r
o
p
o
s
e
s
t
h
e
e
m
p
l
o
y
m
e
n
t
o
f
d
e
e
p
n
e
u
r
a
l
n
e
tw
o
r
k
s
(
D
NN
)
t
o
d
e
t
e
c
t
im
a
g
e
s
w
i
th
m
a
l
ig
n
a
n
t
n
o
d
u
l
e
s
o
f
l
u
n
g
c
o
m
p
u
t
e
d
t
o
m
o
g
ra
p
h
y
(CT
)
.
T
h
e
m
e
t
h
o
d
i
n
c
l
u
d
e
s
s
u
b
j
e
c
t
i
n
g
i
n
p
u
t
i
m
a
g
e
s
t
o
a
s
im
p
l
e
a
n
d
f
a
s
t
p
r
e
-
p
r
o
c
e
s
s
i
n
g
w
h
i
c
h
i
s
o
la
t
e
s
r
e
g
i
o
n
s
o
f
i
n
t
e
r
e
s
t
(
R
OI
)
,
t
h
a
t
’
s
t
h
e
l
u
n
g
s
d
o
m
i
n
a
te
d
a
r
e
a
,
r
i
d
d
i
n
g
t
h
e
i
m
a
g
e
s
o
f
o
t
h
e
r
s
u
r
r
o
u
n
d
i
n
g
t
i
s
s
u
e
s
a
n
d
a
r
t
if
a
c
t
s
.
C
e
n
t
e
r
e
d
a
n
d
s
i
z
e
n
o
rm
a
l
iz
e
d
im
a
g
e
s a
re
t
h
e
n
f
e
d
t
o
a
d
e
e
p
n
e
u
r
a
l
n
e
tw
o
r
k
f
o
r
tr
a
i
n
i
n
g
a
n
d
v
a
l
i
d
a
t
i
o
n
.
I
n
t
h
i
s
w
o
r
k
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
i
s
u
s
e
d
t
o
r
e
a
d
j
u
s
t
G
o
o
g
L
e
N
e
t
DN
N
t
o
l
e
a
r
n
t
h
i
s
m
e
d
i
c
a
l
d
a
t
a
.
T
h
i
s
i
n
c
l
u
d
e
s
a
l
l
o
w
i
n
g
f
i
n
a
l
l
a
y
e
r
s
o
f
t
h
e
DN
N
t
o
e
v
o
l
v
e
w
h
i
le
re
s
t
r
ic
t
i
n
g
d
e
e
p
l
a
y
e
r
s
.
I
n
t
h
i
s
s
e
t
t
i
n
g
,
a
r
o
u
g
h
,
u
n
p
r
o
c
e
s
se
d
d
a
t
a
se
t
,
t
h
e
IQ
-
OT
H
/NC
C
D
l
u
n
g
c
a
n
c
e
r
d
a
t
a
se
t
w
a
s
u
s
e
d
t
o
t
r
a
i
n
/
v
a
l
i
d
a
t
e
t
h
e
p
r
o
p
o
s
e
d
a
l
g
o
r
i
t
h
m
.
E
x
p
e
r
im
e
n
t
a
l
r
e
s
u
l
t
s
s
h
o
w
t
h
a
t
t
h
i
s
a
l
g
o
r
i
t
h
m
sc
o
r
e
s
9
4
.
3
8
%
a
c
c
u
r
a
c
y
,
w
h
i
c
h
o
u
t
p
e
r
f
o
rm
s
b
e
n
c
h
m
a
rk
m
e
t
h
o
d
p
r
e
v
i
o
u
s
ly
u
s
e
d
w
i
t
h
t
h
i
s
d
a
t
a
se
t
.
K
ey
w
o
r
d
s
:
A
r
ti
f
icial
i
n
tel
lig
e
n
ce
C
o
m
p
u
ter
v
is
io
n
C
T
s
ca
n
s
Dee
p
n
eu
r
al
n
e
t
w
o
r
k
L
u
n
g
ca
n
ce
r
T
r
an
s
f
er
lear
n
in
g
T
h
is
is
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
:
Mu
a
y
ed
S
AL
-
H
u
s
ei
n
y
Dep
ar
t
m
en
t o
f
E
lectr
ical
E
n
g
i
n
ee
r
in
g
C
o
lleg
e
o
f
E
n
g
in
ee
r
i
n
g
,
Un
i
v
e
r
s
it
y
o
f
W
asi
t
Un
i
v
er
s
it
y
C
i
t
y
,
A
l
-
R
ab
ie
Dis
t
.
,
A
l
-
Ku
t,
W
asit
,
I
r
aq
E
m
ail:
m
al
h
u
s
ei
n
y
@
u
o
w
asit
.
e
d
u
.
iq
1.
I
NT
RO
D
UCT
I
O
N
L
u
n
g
ca
n
ce
r
is
co
n
s
id
er
ed
am
o
n
g
th
e
f
ier
ce
s
t
f
o
r
m
s
o
f
d
is
ea
s
es
ac
r
o
s
s
t
h
e
w
o
r
ld
.
W
o
r
ld
h
ea
lth
o
r
g
an
izatio
n
(
W
HO)
esti
m
ate
s
th
at
m
o
r
e
th
a
n
1
.
7
6
m
ill
io
n
p
er
is
h
an
n
u
a
ll
y
d
u
e
to
th
i
s
t
y
p
e
o
f
ca
n
ce
r
,
[1
]
,
[
2]
.
On
e
o
f
t
h
e
m
o
s
t
i
m
p
o
r
tan
t
t
o
o
ls
f
o
r
th
e
d
iag
n
o
s
is
o
f
l
u
n
g
ca
n
ce
r
is
co
m
p
u
ted
to
m
o
g
r
ap
h
y
(
C
T
)
.
T
h
is
s
cr
ee
n
i
n
g
m
et
h
o
d
,
h
o
w
ev
er
,
r
eq
u
ir
es t
h
e
s
k
ill
s
o
f
a
tr
ai
n
ed
s
p
ec
ialis
t,
s
o
m
e
th
i
n
g
n
o
t a
l
w
a
y
s
av
ailab
le
[
3
]
.
T
h
e
u
s
e
o
f
th
e
d
ig
ital
p
r
o
ce
s
s
in
g
in
m
ed
icin
e
w
h
ic
h
s
tar
ted
in
ea
r
l
y
s
i
x
ti
es
o
f
las
t
ce
n
t
u
r
y
o
f
f
er
s
r
ed
u
cin
g
p
r
o
ce
s
s
in
g
t
i
m
e
a
n
d
i
m
p
r
o
v
e
m
en
t
in
s
p
ec
i
f
icit
y
a
n
d
s
en
s
iti
v
it
y
[
4
]
.
Als
o
,
th
e
s
tead
y
in
cr
ea
s
e
in
t
h
e
s
ca
le
an
d
n
u
m
b
er
o
f
m
ed
ical
d
ata
m
o
d
al
ities
is
co
n
s
id
er
ed
as
a
ch
allen
g
e
f
o
r
h
ea
lt
h
co
m
m
u
n
i
ties
.
C
o
m
p
u
ter
al
g
o
r
ith
m
s
ar
e
th
e
p
r
im
e
ca
n
d
id
ates
i
n
p
r
o
v
id
in
g
th
e
ca
p
ac
it
y
to
in
g
e
s
t
an
d
an
al
y
s
e
th
ese
d
ata.
T
h
e
m
ed
ical
d
ep
lo
y
m
e
n
t
o
f
th
e
s
e
alg
o
r
it
h
m
s
h
av
e
s
ee
n
r
elen
t
less
p
r
o
li
f
er
atio
n
i
n
r
ec
en
t
y
ea
r
s
[
5
]
.
T
h
ese
m
e
th
o
d
o
lo
g
ies
ar
e
f
u
r
th
er
s
p
ec
ialis
ed
ac
co
r
d
in
g
to
t
h
eir
f
i
n
al
p
r
o
d
u
ce
in
to
co
m
p
u
ter
aid
ed
d
etec
tio
n
(
C
A
De)
a
n
d
d
iag
n
o
s
is
(
C
A
D
x
)
s
y
s
te
m
s
[
6
]
.
T
h
e
ad
o
p
tio
n
o
f
C
A
D
s
y
s
te
m
s
en
m
ass
e
r
e
lies
o
n
ac
h
ie
v
i
n
g
h
ig
h
s
e
n
s
it
iv
it
y
,
cu
tti
n
g
f
al
s
e
p
o
s
itiv
es d
o
w
n
a
n
d
in
cr
ea
s
i
n
g
p
r
o
ce
s
s
i
n
g
s
p
ee
d
w
h
ile
m
a
in
t
ain
i
n
g
a
f
f
o
r
d
ab
le
co
s
ts
[
7
]
.
C
o
m
p
u
ter
is
ed
m
ed
ical
d
ata
an
al
y
s
i
s
u
s
u
a
ll
y
in
c
lu
d
es
p
r
e
-
p
r
o
ce
s
s
in
g
,
s
e
g
m
en
ta
ti
o
n
,
f
ea
t
u
r
e
ex
tr
ac
tio
n
an
d
clas
s
i
f
icatio
n
[8
]
-
[
11]
.
T
h
e
n
atu
r
e
o
f
tis
s
u
e
s
an
d
o
r
g
an
s
o
f
h
u
m
a
n
b
o
d
y
s
u
ch
as
th
o
s
e
o
f
l
u
n
g
s
m
ak
e
s
is
o
lat
io
n
o
f
ca
n
ce
r
n
o
d
u
le
a
d
e
m
a
n
d
in
g
ch
o
r
e
f
o
r
r
ad
io
lo
g
is
ts
w
i
th
f
ail
u
r
e
r
ates
o
f
ab
o
u
t
3
5
%
[
1
2
]
,
[
13]
.
S
p
ec
if
icall
y
,
t
h
e
lac
k
i
n
h
o
m
o
g
en
e
it
y
o
f
lu
n
g
ti
s
s
u
e
an
d
ab
u
n
d
an
ce
o
f
b
lo
o
d
v
es
s
els
a
n
d
b
r
o
n
ch
io
les
co
m
p
lica
tes
th
e
tas
k
o
f
s
e
g
m
e
n
tatio
n
[
1
4
]
,
[
15]
w
h
ic
h
i
s
t
h
e
in
it
ial
s
tep
i
n
d
etec
tio
n
an
d
d
iag
n
o
s
is
[
1
5
]
.
T
h
e
au
th
o
r
s
in
s
o
m
e
s
t
u
d
ies
[
1
6
]
d
ev
elo
p
ed
a
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
S
VM
)
b
ased
alg
o
r
i
th
m
w
h
er
e
g
e
n
etic
alg
o
r
ith
m
s
w
er
e
u
s
ed
to
s
ele
ct
th
e
o
p
ti
m
al
s
e
t
o
f
f
ea
t
u
r
e
s
,
th
e
ac
h
ie
v
ed
ac
cu
r
ac
y
w
a
s
9
3
.
6
6
%,
an
o
th
er
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4752
Tr
a
n
s
fer lea
r
n
in
g
w
ith
Go
o
g
LeN
et
fo
r
d
etec
tio
n
o
f lu
n
g
c
a
n
ce
r
(
Mu
a
ye
d
S
A
L
-
Hu
s
ein
y
)
1079
in
s
ta
n
ce
o
f
th
i
s
al
g
o
r
ith
m
w
h
i
ch
u
s
e
s
n
eu
r
al
n
e
t
w
o
r
k
s
is
clai
m
ed
to
h
a
v
e
ac
h
iev
ed
ac
cu
r
ac
y
o
f
9
5
.
8
7
%.
So
m
e
r
esear
ch
er
s
[
1
7
]
en
s
e
m
b
led
n
eu
r
al
n
et
w
o
r
k
s
g
u
id
ed
b
y
f
u
zz
y
lo
g
ic
clu
s
ter
in
g
to
s
eg
m
e
n
t
lu
n
g
ca
n
ce
r
i
m
ag
e
s
o
n
a
m
o
d
er
ate
d
ata
s
et,
t
h
is
w
a
s
u
s
ed
alo
n
g
w
it
h
s
o
m
e
p
h
y
s
ical
s
i
g
n
s
a
n
d
s
y
m
p
to
m
s
in
t
h
e
cla
s
s
i
f
icat
io
n
p
r
o
ce
s
s
.
Oth
er
r
esear
ch
er
s
[
1
8
]
also
u
s
ed
n
eu
r
al
n
et
w
o
r
k
s
w
it
h
s
el
f
-
o
r
g
a
n
izi
n
g
m
ap
s
to
d
ev
elo
p
th
eir
C
A
D
s
y
s
te
m
g
ai
n
i
n
g
ac
c
u
r
ac
y
o
f
9
0
.
6
3
%.
Dif
f
er
en
t
ap
p
r
o
ac
h
w
as
tak
en
i
n
[
1
9
]
,
th
e
f
o
cu
s
o
f
th
ei
r
s
tu
d
y
w
a
s
o
n
th
e
f
ate
o
f
p
atie
n
ts
o
f
l
u
n
g
ca
n
ce
r
b
y
tr
y
i
n
g
to
p
r
ed
ict
m
o
r
talit
y
r
ates,
th
eir
s
ta
ted
ac
cu
r
ac
y
w
a
s
9
6
%.
W
ith
th
e
in
tr
o
d
u
ctio
n
o
f
n
u
m
er
icall
y
f
ea
s
ib
le
d
ee
p
n
eu
r
al
n
et
w
o
r
k
s
a
n
e
w
w
av
e
o
f
C
AD
b
ased
lu
n
g
ca
n
ce
r
d
etec
tio
n
s
y
s
te
m
s
w
e
r
e
p
r
o
p
o
s
ed
w
it
h
s
u
c
ce
s
s
r
at
es
co
m
p
ar
ab
le
to
t
h
o
s
e
o
f
h
u
m
a
n
e
x
p
er
ts
.
T
h
e
au
th
o
r
s
o
f
[
2
0
]
u
s
ed
t
h
r
ee
t
y
p
es
o
f
clas
s
i
f
ier
s
,
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
et
w
o
r
k
(
C
NN)
,
d
ee
p
b
elief
n
et
w
o
r
k
(
DB
N)
,
an
d
s
tac
k
ed
d
en
o
is
i
n
g
au
to
e
n
co
d
er
(
SDA
E
)
o
n
t
h
e
L
I
D
C
/I
DR
I
d
atase
t,
w
h
er
e,
th
e
n
o
d
u
le
s
s
ca
n
s
ar
e
in
itial
l
y
s
e
g
m
e
n
ted
m
a
n
u
al
l
y
,
th
e
y
ar
e
t
h
en
d
o
w
n
s
a
m
p
le
d
to
5
2
X5
2
b
ef
o
r
e
f
ee
d
in
g
th
e
f
ea
t
u
r
es
to
t
h
e
class
i
f
ier
s
.
T
h
e
w
o
r
k
in
[
2
1
]
p
r
o
p
o
s
e
an
in
ter
esti
n
g
3
D
b
ase
d
f
ea
tu
r
es,
t
h
e
m
ed
ia
n
in
ten
s
it
y
p
r
o
j
ec
tio
n
(
MI
P
)
d
er
iv
ed
m
u
lti
-
v
ie
w
d
ata,
t
h
e
f
ea
tu
r
es a
r
e
th
e
n
u
s
ed
to
tr
ain
a
C
NN
to
e
x
tr
ac
t l
u
n
g
n
o
d
u
le
s
w
h
o
s
e
o
u
tp
u
t
is
f
ed
to
a
Gau
s
s
ia
n
p
r
o
ce
s
s
r
eg
r
es
s
io
n
(
GP
R
)
to
s
co
r
e
th
e
d
eg
r
ee
o
f
m
ali
g
n
an
c
y
.
Mic
r
o
s
co
p
ic
i
m
a
g
es
ar
e
u
s
ed
in
[
2
2
]
to
tr
ain
a
C
NN
f
o
r
au
to
m
atic
d
etec
tio
n
o
f
lu
n
g
ca
n
ce
r
.
T
h
e
alg
o
r
ith
m
p
r
o
p
o
s
ed
in
[
2
3
]
u
s
es
MI
P
3
D
f
ea
t
u
r
es
p
r
o
p
o
s
ed
in
[
2
1
]
w
it
h
t
h
e
s
a
m
e
d
ata
s
et
to
tr
ai
n
a
p
o
w
er
f
u
l
C
NN,
t
h
e
Go
o
g
L
e
Net
[
2
4
]
in
tr
an
s
f
e
r
lear
n
in
g
f
r
a
m
e
w
o
r
k
.
Di
f
f
er
en
t
ap
p
r
o
ac
h
is
f
o
llo
w
ed
in
[
2
5
]
w
h
er
e
a
s
h
o
r
t
p
ip
elin
e
is
u
t
ilized
b
y
r
ep
lacin
g
s
eg
m
e
n
tatio
n
b
y
t
h
r
es
h
o
ld
in
g
,
U
-
Net
i
s
th
e
n
u
s
ed
to
s
elec
t
i
m
a
g
es
w
it
h
n
o
d
u
le
ca
n
d
id
ates,
th
ese
ar
e
class
i
f
ied
b
y
u
s
i
n
g
3
D
C
NNs
,
Van
illa
an
d
Go
o
g
L
eNe
t,
D
ata
Scien
ce
B
o
w
l
[
2
6
]
an
d
lu
n
g
n
o
d
u
le
an
al
y
s
i
s
2016
[
2
7
]
d
atasets
w
er
e
u
s
e
d
to
tr
ain
an
d
test
t
h
e
p
r
o
p
o
s
ed
f
r
a
m
e
w
o
r
k
.
T
h
e
alg
o
r
ith
m
i
n
[
2
8
]
s
u
g
g
est
s
d
ep
lo
y
i
n
g
th
e
Alex
Ne
t
C
N
N
d
ir
ec
tl
y
to
cl
as
s
if
y
m
ali
g
n
an
c
y
i
n
lu
n
g
C
T
s
ca
n
s
o
f
th
e
I
Q
-
OT
H/NC
C
D
[
3
]
d
ataset,
w
h
ile
t
h
e
m
e
th
o
d
p
r
o
d
u
ce
s
g
o
o
d
r
esu
lt
s
,
th
er
e
ar
e
s
till
is
s
u
es
r
e
g
ar
d
in
g
t
h
e
ab
ilit
y
o
f
s
u
ch
ap
p
r
o
ac
h
to
p
r
o
d
u
ce
s
i
m
ilar
r
es
u
lt
s
w
it
h
d
i
f
f
er
e
n
t
d
ata
s
ets
o
r
w
it
h
a
u
g
m
e
n
tat
io
n
o
f
t
h
e
s
a
m
e
d
atase
t
u
s
u
al
l
y
e
m
p
lo
y
ed
to
i
m
p
r
o
v
e
th
e
r
eliab
ilit
y
o
f
al
g
o
r
ith
m
s
a
n
d
to
r
e
m
ed
y
t
h
e
is
s
u
e
s
o
f
o
v
er
f
itti
n
g
.
As
s
u
c
h
ex
p
er
i
m
en
tatio
n
a
n
d
d
ev
elo
p
m
en
t
ar
e
s
til
l
v
iab
le
f
o
r
C
A
D
s
y
s
te
m
s
to
g
ai
n
m
at
u
r
i
t
y
as
th
e
s
e
ar
e
s
till
v
u
ln
er
ab
le
to
h
i
g
h
f
alse
-
p
o
s
it
iv
e
r
ates
co
m
p
ar
ed
to
s
p
ec
ialis
t
[
1
2
]
.
I
n
th
is
s
t
u
d
y
a
f
r
a
m
e
w
o
r
k
i
s
p
r
esen
ted
to
s
er
v
e
i
n
th
e
d
ir
ec
tio
n
o
f
d
iag
n
o
s
i
n
g
l
u
n
g
ca
n
ce
r
i
n
C
T
s
ca
n
i
m
a
g
es.
T
h
e
r
est
o
f
t
h
is
p
ap
er
in
cl
u
d
es:
a
d
etailed
d
escr
ip
tio
n
o
f
th
e
m
et
h
o
d
s
u
s
ed
;
t
h
e
d
esig
n
o
f
t
h
e
ex
p
er
i
m
e
n
t
s
;
d
is
c
u
s
s
io
n
o
f
th
e
r
es
u
lts
;
an
d
f
i
n
all
y
,
th
e
co
n
c
lu
d
i
n
g
r
e
m
ar
k
s
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
T
h
e
p
r
o
p
o
s
ed
f
r
a
m
e
w
o
r
k
co
n
s
i
s
ts
o
f
a
s
et
o
f
p
r
e
-
p
r
o
ce
s
s
in
g
s
tep
s
,
f
o
llo
w
ed
b
y
th
e
s
etu
p
o
f
a
m
ac
h
in
e
lear
n
i
n
g
m
o
d
el.
T
h
e
m
o
d
el
i
s
tr
ai
n
ed
an
d
v
al
id
ated
b
y
u
s
in
g
i
m
a
g
e
d
ataset.
T
h
e
v
ar
io
u
s
s
ta
g
es o
f
t
h
e
f
r
a
m
e
w
o
r
k
ar
e
p
r
esen
ted
s
u
b
s
eq
u
en
tl
y
:
2
.
1
.
P
re
-
pro
ce
s
s
ing
T
h
e
m
ai
n
p
u
r
p
o
s
e
o
f
th
i
s
s
ta
g
e
is
to
h
elp
th
e
lear
n
i
n
g
m
o
d
el
f
o
cu
s
it
s
atte
n
tio
n
o
n
t
h
e
m
o
s
t
r
elev
a
n
t
f
ield
r
ath
er
t
h
a
n
ca
p
tu
r
i
n
g
u
n
r
elate
d
tis
s
u
es,
m
ar
k
er
s
an
d
o
t
h
er
clu
t
ter
.
T
h
e
o
u
tco
m
e
o
f
th
is
s
ta
g
e
i
s
a
r
eg
io
n
o
f
in
ter
es
t (
R
OI
)
co
n
tai
n
in
g
ce
n
tr
ed
an
d
s
ize
n
o
r
m
alis
ed
l
u
n
g
s
r
eg
io
n
,
t
h
is
i
s
s
u
m
m
ar
ized
:
a)
T
ex
tu
r
e
an
al
y
s
i
s
:
to
ac
h
iev
e
t
h
is
,
Gab
o
r
f
i
lter
[
2
9
]
is
ap
p
lied
.
T
h
ese
f
ilter
s
ar
e
b
an
d
p
as
s
f
ilter
s
n
o
r
m
all
y
u
tili
ze
d
i
n
i
m
ag
e
p
r
o
ce
s
s
i
n
g
t
o
em
p
h
asize
r
e
g
io
n
s
w
it
h
s
i
m
ilar
tex
t
u
r
e
[
3
0
]
as
s
h
o
w
n
i
n
2
nd
co
lu
m
n
o
f
Fig
u
r
e
1
,
T
h
is
is
th
e
ca
s
e
i
n
h
u
m
a
n
o
r
g
a
n
s
.
T
h
u
s
,
t
h
e
r
esp
o
n
s
e
o
f
t
h
e
f
ilter
to
a
n
i
n
p
u
t
i
m
ag
e
is
co
m
p
u
ted
in
(
1
)
:
(
́
́
)
(
́
)
(
1
)
w
h
er
e
́
an
d
,
́
Giv
en
t
h
at
is
th
e
s
in
u
s
o
id
al
w
a
v
ele
n
g
th
,
is
th
e
an
g
le
o
f
th
e
n
o
r
m
al
to
t
h
e
w
a
v
e
o
f
a
Gab
o
r
f
u
n
ctio
n
,
is
t
h
e
p
h
a
s
e
s
h
i
f
t,
is
th
e
G
au
s
s
ian
s
ta
n
d
ar
d
d
ev
iatio
n
an
d
is
th
e
asp
ec
t
r
atio
[
3
1
]
.
T
h
ese
p
ar
am
eter
s
w
er
e
s
elec
ted
em
p
ir
icall
y
to
m
a
x
i
m
ize
th
e
r
esp
o
n
s
e
o
f
th
e
f
ilter
i
n
th
e
l
u
n
g
s
’
t
is
s
u
e
r
eg
io
n
s
.
b)
Mo
r
p
h
o
lo
g
ical
o
p
er
atio
n
:
t
h
e
o
u
tp
u
t
o
f
t
h
e
p
r
ev
io
u
s
s
tep
is
f
ir
s
t,
th
r
e
s
h
o
ld
-
ed
to
co
n
v
er
t
t
h
e
i
m
a
g
e
i
n
to
b
lack
an
d
w
h
ite
m
as
k
.
I
m
ag
e
d
ilatio
n
f
o
llo
w
ed
b
y
er
o
s
io
n
ar
e
th
en
u
s
ed
to
f
ill
h
o
les
g
e
n
er
ated
d
u
e
to
f
ilter
i
n
g
a
n
d
to
p
r
o
d
u
ce
h
o
m
o
g
en
o
u
s
r
eg
io
n
s
.
T
h
e
lar
g
e
s
t
r
eg
io
n
w
h
ic
h
r
ep
r
esen
ts
th
e
s
u
r
r
o
u
n
d
i
n
g
o
f
p
atien
t b
o
d
y
i
s
th
e
n
r
e
m
o
v
ed
to
p
r
o
d
u
ce
th
e
lu
n
g
s
m
as
k
d
ep
i
cted
in
3
rd
co
lu
m
n
o
f
F
ig
u
r
e
1
.
c)
R
OI
e
x
tr
ac
tio
n
:
l
u
n
g
s
m
as
k
i
s
m
u
ltip
lied
b
y
t
h
e
o
r
ig
i
n
al
i
m
ag
e.
B
o
u
n
d
in
g
b
o
x
is
th
e
n
ca
l
cu
lated
w
h
ic
h
f
r
a
m
e
s
l
u
n
g
s
’
ex
tr
e
m
itie
s
.
Fi
n
all
y
,
i
m
a
g
es
ar
e
s
ize
n
o
r
m
ali
s
ed
.
E
x
a
m
p
le
R
OI
i
m
a
g
es
ar
e
s
h
o
w
n
i
n
las
t
co
lu
m
n
o
f
F
ig
u
r
e
1
.
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4752
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
22
,
No
.
2
,
Ma
y
2
0
2
1
:
1
0
7
8
-
1
0
8
6
1080
2
.
2
.
T
ra
ns
f
er
lea
rning
I
t
i
s
a
n
a
c
c
e
p
ta
b
l
e
a
p
p
r
o
a
c
h
in
a
r
t
i
f
ic
i
a
l
in
te
l
li
g
e
n
ce
(
A
I
)
(
p
ar
t
i
cu
l
a
r
ly
,
in
m
ac
h
in
e
l
e
a
r
n
in
g
(
M
L
)
)
a
s
w
ith
o
th
e
r
f
i
e
l
d
s
o
f
s
ci
en
c
e
th
a
t
k
n
o
w
l
e
d
g
e
an
d
e
x
p
e
r
i
en
ce
g
a
in
e
d
in
s
o
lv
i
n
g
c
e
r
ta
in
p
r
o
b
l
em
is
u
t
i
li
z
e
d
in
a
n
a
ly
zin
g
a
n
d
t
a
ck
l
in
g
o
th
e
r
r
el
a
t
e
d
(
an
d
ev
en
p
o
s
s
i
b
ly
u
n
r
e
l
at
e
d
)
p
r
o
b
l
em
s
an
d
ch
a
l
le
n
g
es
.
T
h
i
s
in
th
e
r
e
s
e
a
r
c
h
o
f
A
I
a
n
d
ML
is
c
a
l
l
e
d
t
r
an
s
f
e
r
l
e
a
r
n
in
g
[
3
2
]
.
T
h
e
m
o
t
iv
e
f
o
r
s
u
ch
t
r
en
d
i
s
ef
f
i
c
ie
n
cy
,
w
h
e
r
e
o
f
t
en
,
l
im
it
ed
c
o
m
p
u
t
in
g
r
e
s
o
u
r
c
es
d
e
t
e
r
r
e
s
e
a
r
c
h
.
Ut
i
l
iz
in
g
p
r
e
v
i
o
u
s
ly
t
r
ai
n
e
d
m
o
d
e
ls
h
e
l
p
b
r
i
d
g
e
s
o
m
e
g
a
p
s
,
y
e
t
,
c
a
r
ef
u
l
c
h
o
i
c
e
o
f
p
r
e
t
r
a
in
e
d
m
o
d
el
i
s
n
e
c
ess
a
r
y
t
o
en
s
u
r
e
f
in
e
tu
n
in
g
i
t
s
p
e
r
f
o
r
m
an
c
e
t
o
w
a
r
d
s
th
e
n
ew
d
o
m
a
in
p
r
o
d
u
c
e
f
e
as
i
b
l
e
r
e
s
u
lt
s
c
o
m
p
a
r
e
d
t
o
i
n
it
i
a
l
iz
in
g
th
e
en
ti
r
e
p
r
o
c
e
s
s
o
f
m
o
d
e
l
t
r
a
in
in
g
.
Fi
g
u
r
e
2
d
e
t
a
i
ls
th
is
p
r
o
c
e
s
s
.
I
n
iti
al
la
y
er
s
w
h
ic
h
ar
e
d
es
ig
n
ed
to
f
ir
e
o
n
lo
w
lev
el
f
ea
t
u
r
es
ar
e
lef
t
i
n
tact
as
it
i
s
a
s
s
u
m
ed
t
h
at
k
n
o
w
led
g
e
g
ai
n
ed
in
th
i
s
ar
ea
is
u
s
ef
u
l
i
n
th
e
tr
an
s
f
er
r
ed
d
o
m
ai
n
as
w
ell.
L
ater
la
y
er
s
w
h
ic
h
lear
n
d
o
m
a
in
s
p
ec
if
ic
(
h
ig
h
le
v
el)
f
ea
tu
r
es
a
r
e
r
ep
lace
d
an
d
tu
n
e
d
v
ia
t
h
e
s
u
b
s
eq
u
en
t
tr
ai
n
i
n
g
p
r
o
ce
s
s
.
T
h
e
o
v
er
all
tr
ai
n
i
n
g
ti
m
e
an
d
p
ar
a
m
eter
ad
j
u
s
t
m
e
n
t
is
s
i
g
n
if
ican
tl
y
r
ed
u
ce
d
as
m
o
s
t
o
f
th
e
m
o
d
el
is
i
n
h
er
i
te
d
f
r
o
m
t
h
e
o
r
ig
in
al
d
o
m
ai
n
tr
ain
i
n
g
[
3
4
]
.
Si
n
ce
cl
ass
i
f
icatio
n
ta
s
k
s
o
f
e
v
er
y
d
a
y
o
r
d
in
ar
y
o
b
j
ec
ts
is
in
es
s
en
ce
s
y
n
o
n
y
m
o
u
s
to
t
h
at
u
s
ed
in
m
ed
ical
ap
p
licatio
n
s
u
ch
as
t
h
e
clas
s
i
f
icatio
n
o
f
l
u
n
g
n
o
d
u
les
i
n
ter
m
s
o
f
ex
tr
ac
tin
g
an
d
lear
n
i
n
g
s
o
m
e
lo
w
-
lev
el
attr
ib
u
tes,
li
k
e
ed
g
es,
tex
t
u
r
e,
an
d
co
lo
r
i
n
f
o
r
m
atio
n
.
He
n
ce
,
th
e
u
tili
z
atio
n
o
f
p
r
e
-
tr
ain
ed
p
ar
am
eter
s
f
ac
ili
tates
lear
n
in
g
o
r
h
ig
h
-
lev
el
f
ea
t
u
r
e
in
o
r
d
er
to
cu
t
tr
ain
in
g
ti
m
e
an
d
im
p
r
o
v
e
m
o
d
e
l
p
er
f
o
r
m
a
n
ce
[
2
3
]
.
T
h
is
n
o
tio
n
is
p
ar
tic
u
lar
l
y
ex
er
ci
s
ed
h
er
e
b
y
r
ep
o
s
itio
n
i
n
g
t
h
e
p
ar
a
m
ete
r
s
o
f
a
w
ell
-
k
n
o
w
n
C
NN
m
o
d
eled
ar
o
u
n
d
I
m
a
g
e
Net
o
b
j
ec
ts
[
3
3
]
to
th
e
d
o
m
ai
n
o
f
lu
n
g
ca
n
ce
r
d
ata.
Fig
u
r
e
2
.
B
lo
ck
d
iag
r
a
m
s
h
o
win
g
t
h
e
p
r
o
ce
s
s
o
f
tr
a
n
s
f
er
lear
n
in
g
o
f
a
DN
N
tr
ain
ed
o
n
i
m
a
g
es
f
r
o
m
I
m
ag
eNe
t d
ataset
[
3
3
]
,
th
e
n
et
w
o
r
k
i
s
m
o
d
i
f
ied
to
lear
n
/clas
s
if
y
p
atter
n
s
f
r
o
m
d
i
f
f
er
en
t d
o
m
ai
n
(
ad
ap
ted
f
r
o
m
[
3
4
]
)
(
i)
(
ii)
(
iii)
(
iv
)
Fig
u
r
e
1
.
T
h
r
ee
s
tep
s
o
f
p
r
e
-
p
r
o
ce
s
s
in
g
ar
e
s
h
o
w
n
f
o
r
t
w
o
r
an
d
o
m
l
y
s
elec
ted
i
n
p
u
t i
m
ag
e
s
,
ea
ch
in
p
u
t i
m
a
g
e
an
d
th
e
s
u
b
s
eq
u
e
n
t p
r
ep
r
o
ce
s
s
in
g
ar
e
d
ep
icted
o
n
a
r
o
w
.
C
o
l
u
m
n
-
w
i
s
e,
in
p
u
t i
m
a
g
e
s
ar
e
in
(
i)
; te
x
tu
r
e
an
al
y
s
is
i
n
(
ii);
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
i
n
(
iii);
R
OI
ex
tr
ac
tio
n
in
(
i
v
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4752
Tr
a
n
s
fer lea
r
n
in
g
w
ith
Go
o
g
LeN
et
fo
r
d
etec
tio
n
o
f lu
n
g
c
a
n
ce
r
(
Mu
a
ye
d
S
A
L
-
Hu
s
ein
y
)
1081
2
.
3
.
G
o
o
g
L
eNe
t
Go
o
g
L
eNe
t
i
s
a
co
n
v
o
l
u
tio
n
a
l
n
eu
r
al
n
e
t
w
o
r
k
;
it
s
ar
ch
itect
u
r
e
is
s
h
o
w
n
i
n
F
ig
u
r
e
3
.
T
h
e
m
o
d
el
o
f
th
is
n
et
w
o
r
k
w
a
s
r
ep
o
r
ted
ly
d
ev
elo
p
ed
to
ac
co
u
n
t
f
o
r
h
ig
h
f
ea
tu
r
e
r
ep
r
esen
tatio
n
s
b
y
u
s
i
n
g
m
illi
o
n
ev
er
y
d
a
y
o
b
j
ec
t
i
m
a
g
es
in
c
lu
d
ed
i
n
a
h
u
g
e
d
ataset,
th
e
I
m
a
g
eNe
t
d
ataset
[
3
3
]
.
I
t
h
a
s
t
h
e
ca
p
ac
it
y
t
o
class
i
f
y
p
atter
n
s
o
f
ar
o
u
n
d
1
0
0
0
im
a
g
es
.
I
t
u
t
ilize
s
1
2
ti
m
es
le
s
s
p
ar
a
m
eter
s
th
a
n
A
le
x
n
et
[
3
5
]
.
Si
m
i
lar
to
o
th
er
n
eu
r
al
n
e
t
w
o
r
k
s
e
m
p
lo
y
ed
i
n
co
m
p
u
ter
v
i
s
io
n
ap
p
licatio
n
s
,
th
is
m
o
d
el
ac
ce
p
ts
i
m
ag
e
s
as
i
n
p
u
t
an
d
p
r
o
d
u
ce
s
lab
els
o
f
o
n
e
o
f
its
lear
n
ed
clas
s
es
to
g
et
h
er
with
th
e
le
v
el
o
f
co
n
f
id
e
n
ce
as
o
u
tp
u
t
[
3
4
]
.
T
h
e
ar
ch
itect
u
r
e
o
f
Go
o
g
L
e
Net
i
s
b
u
ilt
o
f
2
2
la
y
er
s
in
cl
u
d
i
n
g
9
in
ce
p
tio
n
m
o
d
u
le
s
.
T
h
e
m
o
d
if
ied
i
n
ce
p
tio
n
m
o
d
u
le
s
h
o
wn
in
F
i
g
u
r
e
4
u
s
es
lear
n
ab
le
f
ilter
s
[
2
4
]
w
i
th
s
iz
es
r
an
g
i
n
g
f
r
o
m
(
1
x
1
)
to
(
5
x
5
)
to
p
er
f
o
r
m
co
n
v
o
l
u
tio
n
in
p
ar
allel
w
h
ic
h
h
elp
s
ca
p
tu
r
e
f
ea
t
u
r
e
o
f
d
if
f
er
en
t le
v
els o
f
d
etails
[
3
5
]
.
Fig
u
r
e
3
.
T
h
e
ar
ch
itectu
r
e
o
f
Go
o
g
L
eNe
t
[
2
4
]
w
it
h
m
o
d
i
f
ie
d
in
ce
p
tio
n
m
o
d
u
le
(
d
ia
g
r
a
m
p
ar
titi
o
n
ed
to
f
it)
Fig
u
r
e
4
.
T
h
e
co
m
p
o
n
en
t
s
o
f
t
h
e
m
o
d
i
f
ied
in
ce
p
tio
n
m
o
d
u
le
u
s
ed
in
Go
o
g
L
e
Net
ar
ch
itect
u
r
e
[
2
4
]
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4752
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
22
,
No
.
2
,
Ma
y
2
0
2
1
:
1
0
7
8
-
1
0
8
6
1082
2
.
4
.
E
x
peri
m
ent
des
ig
n
I
n
th
is
s
tu
d
y
w
e
co
m
p
iled
t
h
e
p
u
b
licl
y
av
a
ilab
le
(
I
Q
-
OT
H/NC
C
D)
l
u
n
g
ca
n
ce
r
d
ataset
o
f
C
T
im
a
g
e
s
[
3
]
w
h
ic
h
w
as
u
s
ed
to
tr
ain
/
v
er
if
y
t
h
e
p
r
o
p
o
s
ed
lear
n
in
g
m
o
d
el.
T
h
e
d
ataset
w
as
r
ec
e
n
tl
y
g
at
h
er
ed
in
t
w
o
s
p
ec
ialis
t
h
o
s
p
itals
i
n
I
r
aq
d
u
r
in
g
t
h
e
las
t
q
u
ar
ter
o
f
2
0
1
9
.
T
h
e
d
ataset
is
f
o
cu
s
ed
m
ai
n
l
y
o
n
ca
te
g
o
r
izin
g
c
ases
i
n
o
n
e
o
f
th
r
ee
m
o
d
alit
ies
o
f
lu
n
g
ca
n
ce
r
:
n
o
r
m
al
(
NR
M)
,
b
en
i
g
n
(
B
GN)
an
d
m
alig
n
an
t
(
M
L
G)
,
all
p
er
f
o
r
m
ed
b
y
o
n
co
lo
g
i
s
ts
a
n
d
r
ad
io
l
o
g
is
ts
.
T
h
is
r
ep
o
s
ito
r
y
i
n
cl
u
d
es
ab
o
u
t
1
1
9
0
C
T
s
ca
n
i
m
a
g
es
w
i
th
d
i
m
en
s
io
n
s
5
1
2
*
5
1
2
.
T
h
e
i
m
a
g
es
co
m
e
f
r
o
m
1
1
0
ca
s
es
d
ec
i
m
a
ted
i
n
to
:
4
0
ca
s
es
cla
s
s
i
f
ied
as
M
L
G;
1
5
ca
s
e
s
class
i
f
ied
as
B
GN;
t
h
e
r
e
m
a
i
n
in
g
5
5
ca
s
es
w
er
e
cla
s
s
i
f
ie
d
as
NR
M.
T
h
e
tech
n
ical
d
etails
r
eg
ar
d
in
g
t
h
e
i
m
a
g
in
g
p
r
o
ce
s
s
ar
e
li
s
ted
in
[
3
]
.
Qu
ick
r
e
v
ie
w
o
f
t
h
i
s
d
ata
s
et
s
h
o
w
s
th
a
t
t
h
is
m
o
d
er
ate
s
ized
co
llectio
n
h
as
s
o
m
e
ir
r
eg
u
lar
ities
,
s
u
c
h
as
u
n
u
s
u
a
l
p
o
s
e
an
d
p
r
esen
ce
o
f
n
o
is
e.
T
h
is
m
a
k
es
t
h
e
i
m
a
g
e
s
in
ter
esti
n
g
to
u
s
e
to
m
ea
s
u
r
e
th
e
r
eliab
il
it
y
o
f
lea
r
n
in
g
m
o
d
el
s
.
T
h
e
u
s
er
i
n
ter
f
ac
e
(
UI
)
o
f
M
A
T
L
A
B
©
d
ee
p
lear
n
in
g
to
o
lb
o
x
s
h
o
w
i
n
g
t
h
e
tr
ai
n
in
g
p
ar
a
m
ete
r
s
o
f
Go
o
g
L
e
Net
b
y
u
s
i
n
g
t
h
e
I
Q
-
OT
H/NC
C
D
d
ataset
as s
h
o
w
n
i
n
F
i
g
u
r
e
5
.
Fig
u
r
e
5
.
T
h
e
u
s
er
in
ter
f
ac
e
(
UI
)
o
f
MA
T
L
A
B
©
d
ee
p
lear
n
in
g
to
o
lb
o
x
s
h
o
w
i
n
g
t
h
e
tr
ain
i
n
g
p
ar
a
m
e
ter
s
o
f
Go
o
g
L
eNe
t b
y
u
s
in
g
t
h
e
I
Q
-
O
T
H/NC
C
D
d
ataset
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
I
n
th
is
r
es
e
a
r
ch
t
r
an
s
f
e
r
l
ea
r
n
in
g
o
f
a
p
r
e
t
r
ain
e
d
d
e
e
p
n
eu
r
a
l
n
e
tw
o
r
k
,
th
e
G
o
o
g
L
e
Ne
t
w
as
em
p
l
o
y
e
d
t
o
d
e
v
e
l
o
p
a
l
e
a
r
n
in
g
m
o
d
el
o
f
t
h
e
I
Q
-
O
T
H
/
NC
C
D
lu
n
g
c
an
c
e
r
d
a
t
a
s
et
.
E
x
p
e
r
im
en
t
a
l
r
e
s
u
l
ts
s
h
o
w
th
a
t
th
e
t
r
ain
e
d
m
o
d
e
l
h
as
g
ain
e
d
a
n
o
v
e
r
al
l
ac
c
u
r
a
cy
o
f
9
4
.
3
8
%
o
n
th
e
v
al
id
a
t
i
o
n
d
a
ta
.
F
ig
u
r
e
6
a
s
s
h
o
w
s
t
h
e
p
r
o
g
r
e
s
s
o
f
t
h
e
t
r
a
in
in
g
a
n
d
v
a
l
i
d
at
i
o
n
o
v
e
r
1
7
8
8
i
t
e
r
a
t
i
o
n
d
i
v
i
d
e
d
i
n
t
o
1
2
e
p
o
c
h
s
.
I
n
th
e
t
o
p
p
l
o
t
w
h
i
ch
p
r
es
en
ts
t
h
e
a
c
cu
r
a
cy
,
i
t
i
s
n
o
ti
c
e
d
th
a
t
th
e
a
lg
o
r
i
th
m
s
ta
r
t
s
s
et
t
li
n
g
w
ith
in
t
h
e
v
i
cin
i
ty
o
f
i
t
s
f
in
a
l
v
a
lu
e
h
a
lf
w
ay
a
t
a
r
o
u
n
d
900
i
t
e
r
a
t
i
o
n
s
.
I
t
i
s
a
ls
o
n
o
t
i
ce
d
th
a
t
th
e
al
g
o
r
i
t
h
m
k
e
e
p
s
s
u
f
f
e
r
in
g
f
r
o
m
f
lu
ct
u
at
i
o
n
s
in
th
e
t
r
a
in
in
g
p
r
o
c
es
s
.
T
h
e
s
e
f
l
u
c
tu
a
ti
o
n
s
a
r
e
a
t
t
r
i
b
u
t
e
d
t
o
t
h
e
n
a
tu
r
e
o
f
th
e
d
a
t
a
w
h
o
s
e
im
ag
in
g
c
o
n
d
it
io
n
s
ca
n
b
e
d
es
c
r
i
b
e
d
as
n
o
n
-
u
n
i
f
o
r
m
.
A
d
d
it
i
o
n
a
l
ly
,
t
h
e
au
g
m
en
t
at
i
o
n
u
s
e
d
w
ith
th
e
d
a
t
a
i
s
c
o
m
p
li
c
a
t
in
g
th
is
is
s
u
e
.
N
e
v
e
r
th
e
l
e
s
s
,
i
t
is
a
s
s
u
m
e
d
th
a
t
s
u
ch
c
o
n
d
it
i
o
n
s
a
r
e
t
h
e
n
o
r
m
w
h
en
a
m
ac
h
in
e
l
e
a
r
n
in
g
m
o
d
el
i
s
d
e
p
l
o
y
e
d
t
o
r
e
a
l
p
r
o
b
l
em
s
w
h
e
r
e
d
at
a
c
o
n
d
i
t
i
o
n
s
a
r
e
f
a
r
f
r
o
m
p
e
r
f
e
c
t
.
B
o
t
t
o
m
r
o
w
o
f
F
i
g
u
r
e
6
s
h
o
w
t
h
e
l
o
s
s
a
s
s
o
c
i
at
e
d
w
ith
t
h
e
t
r
a
in
in
g
/v
al
i
d
a
t
i
o
n
p
r
o
c
e
d
u
r
e
.
F
i
g
u
r
e
7
p
r
e
s
en
t
s
e
x
am
p
l
e
im
ag
es
c
la
s
s
if
i
e
d
b
y
th
e
p
r
o
p
o
s
e
d
al
g
o
r
ith
m
.
T
h
e
r
esu
lts
also
s
h
o
w
th
at
t
h
e
tr
ain
ed
m
o
d
el
ac
h
iev
ed
9
5
.
0
8
%
s
en
s
it
iv
it
y
a
n
d
9
3
.
7
%
s
p
ec
if
icit
y
.
Sin
ce
t
h
e
p
r
i
m
ar
y
ai
m
o
f
cla
s
s
if
ica
tio
n
m
o
d
els
is
to
b
o
o
s
t
th
e
s
e
n
s
it
iv
i
t
y
w
h
ic
h
m
an
if
e
s
t
s
th
e
s
u
cc
e
s
s
o
r
h
i
t
r
ate,
w
h
ile
p
r
eser
v
in
g
th
e
s
p
e
cif
icit
y
w
h
ic
h
s
ig
n
i
f
ie
s
tr
u
e
n
eg
ati
v
e
r
ate
[
3
6
]
.
T
h
e
ac
h
iev
ed
v
alu
e
s
o
f
t
h
ese
m
ea
s
u
r
es
p
r
ese
n
ted
h
er
e
r
e
f
le
ct
th
e
r
eliab
il
it
y
o
f
t
h
e
p
r
o
p
o
s
ed
f
r
a
m
e
w
o
r
k
(
R
OI
e
x
tr
ac
tio
n
f
o
llo
w
ed
b
y
d
ee
p
lear
n
in
g
)
i
n
cla
s
s
i
f
y
in
g
s
i
m
ila
r
m
o
d
alit
y
o
f
d
ata
c
o
m
p
ar
ed
t
o
th
e
s
a
m
e
al
g
o
r
it
h
m
u
s
ed
w
it
h
o
u
t
p
r
e
-
p
r
o
ce
s
s
in
g
also
i
m
p
le
m
e
n
ted
b
y
t
h
e
a
u
th
o
r
s
,
w
h
ich
s
co
r
es
ar
o
u
n
d
7
0
%
ac
cu
r
ac
y
.
T
h
e
co
n
f
u
s
io
n
m
a
t
r
ix
o
f
t
h
e
d
is
c
u
s
s
ed
r
esu
lt
s
is
p
r
ese
n
ted
i
n
T
ab
le
1
.
I
t’
s
w
o
r
th
r
e
f
er
r
in
g
t
h
at
f
o
r
th
e
s
a
k
e
o
f
s
i
m
p
licit
y
o
f
p
r
esen
tatio
n
o
f
t
h
e
co
n
f
u
s
io
n
m
atr
ix
b
o
t
h
clas
s
es
o
f
B
GN
an
d
N
R
M
ar
e
co
n
s
i
d
er
ed
s
am
e
ca
teg
o
r
y
h
er
e.
T
h
is
co
n
f
o
r
m
s
to
th
e
in
itial
ai
m
o
f
t
h
is
p
ap
er
t
h
at
i
s
to
d
esi
g
n
a
m
o
d
el
ca
p
ab
le
o
f
d
etec
tin
g
m
al
ig
n
a
n
c
y
in
lu
n
g
C
T
s
ca
n
s
w
h
er
e
o
th
er
n
o
n
-
m
ali
g
n
a
n
t i
s
s
u
es a
r
e
n
o
t f
u
r
th
er
d
etailed
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4752
Tr
a
n
s
fer lea
r
n
in
g
w
ith
Go
o
g
LeN
et
fo
r
d
etec
tio
n
o
f lu
n
g
c
a
n
ce
r
(
Mu
a
ye
d
S
A
L
-
Hu
s
ein
y
)
1083
(
i)
(
ii)
Fig
u
r
e
6
.
P
r
o
g
r
ess
o
f
tr
ain
in
g
/
v
alid
atio
n
o
f
th
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
g
en
er
ated
b
y
t
h
e
M
A
T
L
A
B
® d
ee
p
lear
n
in
g
to
o
lb
o
x
o
v
er
1
2
e
p
o
ch
s
an
d
1
7
8
8
iter
at
io
n
s
: (
i)
ac
cu
r
ac
y
; (
ii
)
lo
s
s
Fig
u
r
e
7
.
Fo
u
r
r
an
d
o
m
l
y
s
elec
ted
ex
a
m
p
le
C
T
s
ca
n
i
m
a
g
es s
h
o
w
n
ab
o
v
e
ea
ch
i
s
t
h
e
class
if
icatio
n
lab
el
w
it
h
its
p
r
o
b
ab
ilit
y
a
s
p
r
o
d
u
ce
d
b
y
th
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
.
NR
M
r
ef
er
s
to
No
r
m
a
l,
B
GN
r
ef
er
s
to
B
en
ig
n
,
a
n
d
ML
G
r
e
f
er
s
to
Ma
lig
n
a
n
t
T
ab
le
1
.
C
o
n
f
u
s
io
n
m
atr
i
x
C
o
n
f
u
s
i
o
n
ma
t
r
i
x
P
r
e
d
i
c
t
e
d
c
l
a
ss
N
o
n
-
mal
i
g
n
a
n
t
(
p
o
si
t
i
v
e
)
M
a
l
i
g
n
a
n
t
(
n
e
g
a
t
i
v
e
)
A
c
t
u
a
l
c
l
a
ss
N
o
n
-
mal
i
g
n
a
n
t
T
P
=
1
1
6
F
N
=
6
M
a
l
i
g
n
a
n
t
F
P
=
8
T
N
=
1
1
9
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4752
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
22
,
No
.
2
,
Ma
y
2
0
2
1
:
1
0
7
8
-
1
0
8
6
1084
I
t’
s
w
o
r
th
n
o
ti
n
g
t
h
at
t
h
e
ti
m
e
co
n
s
u
m
ed
b
y
t
h
e
tr
ain
i
n
g
an
d
v
alid
atio
n
s
ta
g
es
i
s
5
7
m
in
u
tes
an
d
4
1
s
ec
o
n
d
s
,
w
h
ile
t
h
e
ti
m
e
f
o
r
s
t
a
m
p
f
o
r
p
r
e
-
p
r
o
ce
s
s
in
g
s
tep
w
a
s
les
s
th
a
n
2
s
ec
o
n
d
s
r
u
n
o
n
a
s
i
n
g
le
co
r
e
o
f
an
I
n
tel
C
o
r
e
-
i3
p
r
o
ce
s
s
o
r
p
o
w
er
ed
Ma
cB
o
o
k
Pro
m
ac
h
in
e
c
lo
ck
ed
at
2
.
5
GHz
w
i
th
1
6
GB
r
a
m
,
as
m
e
n
tio
n
ed
i
n
T
ab
le
1
.
T
h
ese
o
u
tco
m
es
w
e
r
e
also
co
m
p
ar
ed
to
th
o
s
e
r
e
co
r
d
ed
b
y
t
h
e
b
en
c
h
m
ar
k
al
g
o
r
ith
m
o
r
ig
i
n
all
y
p
r
o
p
o
s
ed
f
o
r
th
is
d
atase
t.
I
n
[
3
]
th
e
au
t
h
o
r
s
e
m
p
lo
y
ed
SV
M
to
p
er
f
o
r
m
c
lass
if
icatio
n
d
u
ties
.
T
h
e
p
r
o
p
o
s
ed
w
o
r
k
also
p
er
f
o
r
m
s
b
etter
th
a
n
t
h
e
al
g
o
r
ith
m
in
[
2
8
]
w
h
ic
h
e
m
p
lo
y
s
A
le
x
Net
to
cla
s
s
i
f
y
th
e
i
m
a
g
e
s
o
f
I
Q
-
OT
H/NC
C
D
d
ataset.
In
T
ab
le
2
,
it
is
ev
id
en
t
th
at
th
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
c
o
m
p
ar
ati
v
el
y
o
u
tp
er
f
o
r
m
s
in
all
th
r
ee
p
er
f
o
r
m
a
n
ce
m
ea
s
u
r
es.
I
t
i
s
clea
r
th
a
t
t
h
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
al
s
o
o
u
tp
er
f
o
r
m
s
o
th
er
ap
p
r
o
ac
h
es
i
m
p
le
m
en
ted
b
y
u
s
i
n
g
d
ataset
s
d
if
f
er
en
t
f
r
o
m
t
h
e
cu
r
r
en
t
o
n
e.
A
lt
h
o
u
g
h
,
ca
r
e
s
h
o
u
ld
b
e
tak
en
n
o
t
to
d
r
a
w
s
o
lid
co
n
clu
s
io
n
s
r
eg
ar
d
i
n
g
th
e
p
er
f
o
r
m
an
ce
o
f
t
h
is
alg
o
r
it
h
m
co
m
p
ar
ed
to
t
h
e
o
th
er
al
g
o
r
ith
m
s
u
n
les
s
te
s
ted
co
m
p
ar
ati
v
el
y
o
n
eq
u
al
g
r
o
u
n
d
s
.
T
ab
le
1
.
C
o
m
p
ar
is
o
n
o
f
p
er
f
o
r
m
an
ce
m
etr
ic
s
R
e
se
a
r
c
h
M
e
t
h
o
d
D
a
t
a
se
t
Ep
o
c
h
s
S
e
n
si
t
i
v
i
t
y
S
p
e
c
i
f
i
c
i
t
y
A
c
c
u
r
a
c
y
S
V
M
[
3
]
IQ
-
O
TH
/
N
C
C
D
/
8
9
.
0
3
3
%
9
3
.
6
6
2
%
8
9
.
8
8
%
M
I
P
-
C
N
N
[
2
3
]
LI
D
C
/
I
D
R
I
1
0
0
t
r
a
i
n
.
/
3
0
0
v
a
l
.
8
4
%
7
8
%
81
%
G
P
R
[
2
5
]
V
a
n
i
l
l
a
G
o
o
g
L
e
N
e
t
D
a
t
a
S
c
i
e
n
c
e
B
o
w
l
[
2
6
]
/
L
U
N
A
1
6
[
2
7
]
/
5
9
.
3
%
7
7
%
7
6
.
1
%
7
4
.
1
%
7
0
.
5
%
7
5
.
1
%
M
u
l
t
i
-
c
l
a
s
si
f
i
[
2
0
]
NN
BN
D
A
E
LI
D
C
/
I
D
R
I
1
0
0
/
/
7
9
.
7
6
%
8
1
.
1
9
%
7
9
.
2
9
%
T
u
mo
r
N
e
t
[
2
1
]
LI
D
C
/
I
D
R
I
1
0
,
0
0
0
i
t
e
r
.
/
/
8
2
.
4
7
%
M
i
c
r
o
sco
p
.
[
2
2
]
P
r
o
p
r
i
e
t
a
r
y
6
0
0
0
0
/
/
7
1
%
A
l
e
x
N
e
t
[
2
8
]
IQ
-
O
TH
/
N
C
C
D
1
0
0
9
5
.
7
1
%
9
5
%
9
3
.
5
4
%
P
r
o
p
o
se
d
IQ
-
O
TH
/
N
C
C
D
12
9
5
.
0
8
%
9
3
.
7
%
9
4
.
3
8
%
4.
CO
NCLU
SI
O
N
I
n
th
is
p
a
p
e
r
a
l
ea
r
n
i
n
g
alg
o
r
i
th
m
is
p
r
e
s
en
t
e
d
t
o
m
o
d
e
l
p
u
b
l
ic
ly
p
u
b
l
is
h
e
d
m
e
d
i
c
al
d
a
t
a
.
T
h
e
d
a
t
a
,
I
Q
-
O
T
H
/
N
C
C
D
lu
n
g
c
a
n
c
e
r
d
a
t
as
e
t
,
in
c
lu
d
es
a
c
o
l
l
e
c
ti
o
n
o
f
C
T
s
c
a
n
s
o
f
lu
n
g
r
e
g
i
o
n
f
o
r
s
u
b
je
c
t
s
w
i
th
p
r
o
b
a
b
l
e
l
u
n
g
ca
n
c
e
r
.
T
h
e
a
lg
o
r
i
th
m
p
r
o
p
o
s
e
s
t
o
u
s
e
c
o
m
p
u
t
a
ti
o
n
a
l
ly
in
ex
p
en
s
i
v
e
p
r
e
-
p
r
o
c
e
s
s
in
g
p
r
o
c
e
s
s
t
o
e
x
t
r
a
c
t lu
n
g
s
r
e
g
i
o
n
an
d
r
em
o
v
e
r
em
a
in
in
g
c
l
u
t
t
e
r
an
d
i
r
r
el
e
v
a
n
t
s
u
r
r
o
u
n
d
i
n
g
s
.
T
h
e
p
r
o
p
o
s
e
d
f
r
am
ew
o
r
k
a
ls
o
ex
p
l
o
r
e
s
th
e
b
e
n
ef
it
s
o
f
u
s
in
g
p
r
e
t
r
ai
n
e
d
d
e
e
p
n
e
u
r
al
n
etw
o
r
k
s
in
a
t
r
a
n
s
f
er
l
e
a
r
n
in
g
a
p
p
r
o
a
c
h
t
o
r
e
d
u
c
e
tim
e
an
d
r
e
s
o
u
r
c
es
.
T
h
i
s
i
s
a
ch
i
ev
e
d
b
y
r
et
r
a
in
in
g
G
o
o
g
L
e
N
e
t
t
o
a
c
c
o
m
m
o
d
a
t
e
f
o
r
t
h
e
f
e
a
tu
r
es
o
f
th
is
m
e
d
ic
a
l
d
a
t
a
.
E
m
p
i
r
i
c
a
l
o
u
t
c
o
m
es
s
h
o
w
th
at
t
h
e
p
r
o
p
o
s
e
d
a
lg
o
r
it
h
m
y
i
e
l
d
s
h
ig
h
e
r
a
c
c
u
r
ac
y
a
t
9
4
.
3
8
%
c
o
m
p
a
r
ed
t
o
th
e
a
lg
o
r
i
th
m
s
u
g
g
es
t
e
d
o
r
i
g
in
a
l
ly
w
i
th
th
is
d
a
t
a
s
e
t
w
h
i
ch
s
c
o
r
e
s
8
9
.
8
8
%
.
T
h
e
p
r
o
p
o
s
e
d
a
lg
o
r
it
h
m
a
ls
o
o
u
t
p
e
r
f
o
r
m
s
o
th
e
r
a
l
g
o
r
ith
m
s
im
p
l
em
en
t
e
d
o
n
d
if
f
e
r
en
t
d
a
ta
s
et
s
.
Fu
r
th
e
r
m
o
r
e
,
o
t
h
e
r
p
e
r
f
o
r
m
an
c
e
m
e
t
r
i
c
s
s
u
ch
as
s
en
s
i
ti
v
i
ty
an
d
s
p
e
c
if
i
ci
ty
w
e
r
e
c
o
m
p
u
t
e
d
.
T
h
es
e
m
e
t
r
i
cs
a
ls
o
s
u
p
p
o
r
t
th
e
a
b
o
v
e
v
e
r
d
i
c
t
w
h
ic
h
en
t
a
i
ls
th
a
t
th
i
s
a
p
p
r
o
a
ch
i
s
s
u
i
t
a
b
le
f
o
r
r
el
i
a
b
ly
cl
a
s
s
if
y
in
g
m
e
d
i
c
al
im
ag
e
r
y
.
T
h
e
m
e
r
g
e
r
o
f
s
u
i
t
a
b
l
e
p
r
e
-
p
r
o
c
es
s
in
g
w
ith
c
o
n
v
o
lu
t
i
o
n
al
n
eu
r
a
l
n
e
tw
o
r
k
s
in
th
is
f
ash
i
o
n
h
as
b
ee
n
v
a
li
d
a
t
e
d
b
y
u
s
in
g
th
e
s
am
e
d
e
e
p
le
a
r
n
in
g
m
o
d
e
l
tr
a
i
n
e
d
h
e
r
e
w
i
th
o
u
t
u
t
il
i
zi
n
g
R
O
I
ex
t
r
a
ct
i
o
n
(
p
r
e
-
p
r
o
c
e
s
s
in
g
)
,
th
e
l
at
e
r
v
a
r
i
an
t
r
e
s
u
l
t
e
d
in
m
u
ch
l
o
w
e
r
a
c
cu
r
a
cy
a
t
7
0
%
.
T
h
e
p
r
e
s
en
t
e
d
a
p
p
r
o
a
ch
,
a
ls
o
h
in
ts
a
t
th
e
p
r
o
s
p
ec
t
s
o
f
ex
p
an
d
i
n
g
th
e
u
ti
l
i
za
t
i
o
n
o
f
p
r
e
t
r
ain
e
d
m
o
d
e
l
s
as
a
m
ea
n
s
o
f
c
o
m
p
en
s
a
t
in
g
f
o
r
l
ac
k
o
f
c
o
m
p
u
t
a
ti
o
n
a
l
r
es
o
u
r
c
e
s
in
m
e
d
i
c
al
f
a
c
i
li
t
ie
s
an
d
ac
a
d
em
i
c
in
s
t
i
tu
t
i
o
n
s
w
h
il
e
in
th
e
s
am
e
tim
e
p
r
o
v
e
c
o
n
c
e
p
t
s
o
f
s
o
f
t
w
a
r
e
en
g
in
e
e
r
in
g
s
u
c
h
a
s
s
o
f
tw
a
r
e
r
e
u
s
a
b
il
i
ty
.
RE
F
E
R
E
NC
E
S
[1
]
A
.
A
su
n
th
a
a
n
d
A
.
S
ri
n
iv
a
sa
n
,
"
De
e
p
lea
rn
in
g
f
o
r
lu
n
g
Ca
n
c
e
r
d
e
tec
ti
o
n
a
n
d
c
las
sif
ica
ti
o
n
,
"
M
u
lt
i
me
d
ia
T
o
o
ls
a
n
d
Ap
p
li
c
a
ti
o
n
s,
v
o
l
.
7
9
,
n
o
.
1
1
,
p
p
.
7
7
3
1
-
7
7
6
2
,
2
0
2
0
.
[2
]
L
.
Nie
,
L
.
Zh
a
n
g
,
Y.
Ya
n
g
,
M
.
W
a
n
g
,
R.
Ho
n
g
,
a
n
d
T
.
-
S
.
Ch
u
a
,
"
Be
y
o
n
d
Do
c
to
rs:
F
u
tu
re
He
a
lt
h
P
re
d
icti
o
n
f
ro
m
M
u
lt
im
e
d
ia
a
n
d
M
u
lt
im
o
d
a
l
Ob
se
rv
a
ti
o
n
s,"
p
re
se
n
ted
a
t
th
e
Pro
c
e
e
d
in
g
s
o
f
t
h
e
2
3
rd
ACM
in
ter
n
a
ti
o
n
a
l
c
o
n
fer
e
n
c
e
o
n
M
u
lt
ime
d
i
a
,
Bris
b
a
n
e
,
A
u
stra
li
a
,
2
0
1
5
.
[
On
li
n
e
]
.
A
v
a
il
a
b
le:
d
o
i
:
1
0
.
1
1
4
5
/
2
7
3
3
3
7
3
.
2
8
0
6
2
1
7
.
[3
]
H.
F
.
A
l
-
Ya
sri
y
,
M
.
S
.
A
l
-
Hu
sie
n
y
,
F
.
Y.
M
o
h
se
n
,
E.
A
.
Kh
a
li
l,
a
n
d
Z.
S
.
Ha
ss
a
n
,
"
Ev
a
lu
a
ti
o
n
o
f
S
VM
P
e
rf
o
rm
a
n
c
e
in
th
e
De
te
c
ti
o
n
o
f
L
u
n
g
Ca
n
c
e
r
in
M
a
rk
e
d
CT
S
c
a
n
Da
tas
e
t,
"
In
d
o
n
e
sia
n
J
o
u
r
n
a
l
o
f
El
e
c
trica
l
En
g
i
n
e
e
rin
g
a
n
d
C
o
mp
u
ter
S
c
ien
c
e
,
v
o
l.
2
1
,
n
o
.
3
,
2
0
2
1
.
[4
]
E.
Em
irza
d
e
,
"
A
Co
m
p
u
ter
A
i
d
e
d
Dia
g
n
o
sis
S
y
ste
m
f
o
r
L
u
n
g
Ca
n
c
e
r
De
tec
ti
o
n
Us
in
g
S
V
M
,
"
M
a
ste
r
o
f
S
c
ien
c
e
,
T
h
e
Gr
a
d
u
a
te S
c
h
o
o
l
o
f
Ap
p
li
e
d
S
c
ien
c
e
s
, N
e
a
r
Eas
t
Un
iv
e
rsit
y
,
2
0
1
6
.
[5
]
M
.
S
.
A
L
-
Hu
se
in
y
,
N.
K.
A
b
b
a
s,
a
n
d
A
.
S
.
S
a
ji
t,
"
Dia
g
n
o
sis
o
f
a
rrh
y
th
m
ia
b
a
se
d
o
n
ECG
a
n
a
ly
sis
u
sin
g
CNN
,
"
Bu
ll
e
ti
n
o
f
El
e
c
trica
l
E
n
g
in
e
e
rin
g
a
n
d
I
n
fo
rm
a
t
ics
,
v
o
l.
9
,
n
o
.
3
,
p
p
.
9
8
8
-
9
9
5
,
2
0
2
0
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4752
Tr
a
n
s
fer lea
r
n
in
g
w
ith
Go
o
g
LeN
et
fo
r
d
etec
tio
n
o
f lu
n
g
c
a
n
ce
r
(
Mu
a
ye
d
S
A
L
-
Hu
s
ein
y
)
1085
[6
]
J.
Ch
o
,
K.
L
e
e
,
E.
S
h
in
,
G
.
Ch
o
y
,
a
n
d
S
.
Do
.
"
Ho
w
m
u
c
h
d
a
ta
is
n
e
e
d
e
d
t
o
train
a
m
e
d
ica
l
i
m
a
g
e
d
e
e
p
lea
rn
in
g
s
y
ste
m
to
a
c
h
iev
e
n
e
c
e
ss
a
r
y
h
ig
h
a
c
c
u
ra
c
y
,
"
2
0
1
6
.
[7
]
M
.
F
irm
in
o
,
A
.
H.
M
o
ra
is,
R.
M
.
M
e
n
d
o
ç
a
,
M
.
R.
Da
n
tas
,
H.
R.
He
k
is,
a
n
d
R.
V
a
len
ti
m
,
"
Co
m
p
u
ter
-
a
id
e
d
d
e
tec
ti
o
n
sy
ste
m
f
o
r
lu
n
g
c
a
n
c
e
r
in
c
o
m
p
u
ted
to
m
o
g
ra
p
h
y
s
c
a
n
s:
re
v
ie
w
a
n
d
f
u
tu
re
p
ro
sp
e
c
ts,
"
(in
e
n
g
),
Bi
o
me
d
En
g
O
n
li
n
e
,
v
o
l.
1
3
,
p
.
4
1
,
A
p
r
8
2
0
1
4
.
[8
]
E.
S
.
L
u
z
,
T
.
M
.
Nu
n
e
s,
V
.
H.
Alb
u
q
u
e
rq
u
e
,
J.
P
.
P
a
p
a
,
a
n
d
D.
M
e
n
o
tt
i,
"
ECG
a
rrh
y
th
m
ia
c
las
si
f
ic
a
ti
o
n
b
a
se
d
o
n
o
p
ti
m
u
m
-
p
a
th
f
o
re
st,"
Exp
e
rt S
y
ste
ms
wit
h
Ap
p
li
c
a
ti
o
n
s,
v
o
l.
4
0
,
p
p
.
3
5
6
1
-
3
5
7
3
,
2
0
1
3
.
[9
]
M
.
S
.
A
l
-
Hu
se
in
y
a
n
d
S
.
M
a
h
m
o
o
d
i,
"
S
h
a
p
e
re
g
istratio
n
u
sin
g
c
h
a
ra
c
teristic
f
u
n
c
ti
o
n
s,"
IET
Im
a
g
e
Pro
c
e
ss
in
g
,
v
o
l.
9
,
n
o
.
3
,
p
p
.
2
4
9
-
2
6
0
,
2
0
1
4
.
[1
0
]
S
.
M
a
h
m
o
o
d
i,
"
Ed
g
e
De
tec
ti
o
n
F
il
ter
b
a
se
d
o
n
M
u
m
f
o
rd
-
S
h
a
h
G
re
e
n
F
u
n
c
ti
o
n
,
"
S
IAM
J
o
u
rn
a
l
o
f
I
mg
in
g
S
c
ien
c
e
s,
v
o
l.
5
,
n
o
.
1
,
p
p
.
3
4
3
-
3
6
5
,
2
0
1
2
.
[1
1
]
S
.
M
a
h
m
o
o
d
i,
"
M
a
th
e
m
a
ti
c
a
l
A
n
a
l
y
sis
f
o
r
S
h
a
p
e
Re
g
istratio
n
b
a
se
d
o
n
S
ig
n
e
d
Dista
n
c
e
F
u
n
c
ti
o
n
s,
"
Un
iv
e
r
sit
y
o
f
S
o
u
t
h
a
m
p
to
n
I
n
stit
u
ti
o
n
a
l
Re
p
o
si
to
ry
,
2
0
1
1
,
v
o
l
.
so
t
o
n
:
2
7
2
9
6
5
.
[
O
n
li
n
e
]
.
[1
2
]
P
.
Y.
M
.
A
n
sh
a
d
a
n
d
S
.
S
.
Ku
m
a
r,
"
Re
c
e
n
t
m
e
th
o
d
s
f
o
r
th
e
d
e
tec
t
io
n
o
f
tu
m
o
r
u
sin
g
c
o
m
p
u
ter
a
id
e
d
d
iag
n
o
sis
—
A
re
v
ie
w
,
"
in
2
0
1
4
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Co
n
tro
l,
I
n
st
ru
me
n
ta
ti
o
n
,
Co
mm
u
n
ica
ti
o
n
&
Co
mp
u
ta
t
io
n
a
l
T
e
c
h
n
o
l
o
g
ies
,
2
0
1
4
,
p
p
.
1
0
1
4
-
1
0
1
9
.
[1
3
]
S
.
M
a
h
m
o
o
d
i,
M
.
S
.
A
L
-
Hu
se
in
y
,
a
n
d
M
.
S
.
Nix
o
n
,
"
S
im
il
a
rit
y
R
e
g
istratio
n
f
o
r
S
h
a
p
e
s
Ba
se
d
o
n
S
ig
n
e
d
Dista
n
c
e
F
u
n
c
ti
o
n
s,"
p
re
se
n
ted
a
t
th
e
I
n
ter
n
a
ti
o
n
a
l
S
y
mp
o
si
u
m o
n
Vi
s
u
a
l
Co
mp
u
ti
n
g
,
Cre
te,
G
re
e
c
e
,
2
0
1
2
.
[1
4
]
A
.
El
-
Ba
z
e
t
a
l.
,
"
Co
m
p
u
ter
-
Aid
e
d
Dia
g
n
o
sis
S
y
ste
m
s
f
o
r
Lu
n
g
Ca
n
c
e
r:
Ch
a
ll
e
n
g
e
s
a
n
d
M
e
th
o
d
o
l
o
g
ies
,
"
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
B
io
me
d
i
c
a
l
Ima
g
in
g
,
v
o
l
.
2
0
1
3
,
p
.
9
4
2
3
5
3
,
2
0
1
3
.
[1
5
]
S
.
M
a
h
m
o
o
d
i,
"
Em
p
h
y
se
m
a
d
iag
n
o
se
s
in
lu
n
g
s
u
sin
g
m
e
d
ica
l
i
m
a
g
in
g
m
e
th
o
d
s,"
p
re
se
n
ted
a
t
t
h
e
2
2
n
d
Ira
n
ian
Co
n
f
e
re
n
c
e
o
n
Bio
m
e
d
ica
l
En
g
in
e
e
ri
n
g
,
Ira
n
,
Isla
m
ic
Re
p
u
b
li
c
o
f
,
2
0
1
5
-
11
-
25
-
2
0
1
5
-
11
-
2
7
,
2
0
1
5
,
Ke
y
n
o
te
Sp
e
e
c
h
.
[
On
l
in
e
]
.
A
v
a
il
a
b
le:
h
tt
p
s
:/
/ep
rin
ts.
so
t
o
n
.
a
c
.
u
k
/
3
8
2
7
0
1
/
.
[1
6
]
J
.
M
.
D
i
a
z
,
R
.
C
.
P
i
n
o
n
,
a
n
d
G
.
S
o
l
a
n
o
,
"
L
u
n
g
c
a
n
c
e
r
c
l
a
s
s
i
f
i
c
a
t
i
o
n
u
s
i
n
g
g
e
n
e
t
i
c
a
l
g
o
r
i
t
h
m
t
o
o
p
t
i
m
i
z
e
p
r
e
d
i
c
t
i
o
n
m
o
d
e
l
s
,
"
i
n
T
h
e
5
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
I
n
f
o
r
m
a
t
i
o
n
,
I
n
t
e
l
l
i
g
e
n
c
e
,
S
y
s
t
e
m
s
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
2
0
1
4
,
p
p
.
1
-
6
.
[1
7
]
F
.
T
a
h
e
r
a
n
d
R.
S
a
m
m
o
u
d
a
,
"
L
u
n
g
c
a
n
c
e
r
d
e
tec
ti
o
n
b
y
u
sin
g
a
rti
f
icia
l
n
e
u
ra
l
n
e
t
w
o
rk
a
n
d
fu
z
z
y
c
lu
ste
rin
g
m
e
th
o
d
s,"
in
2
0
1
1
IEE
E
GCC Co
n
fer
e
n
c
e
a
n
d
Exh
i
b
it
i
o
n
,
2
0
1
1
,
p
p
.
2
9
5
-
2
9
8
.
[1
8
]
E.
Da
n
d
ıl
,
M
.
Ça
k
iro
ğ
lu
,
Z.
E
k
şi,
M
.
Öz
k
a
n
,
Ö.
K.
Ku
rt,
a
n
d
A
.
Ca
n
a
n
,
"
Artif
icia
l
n
e
u
ra
l
n
e
tw
o
rk
-
b
a
se
d
c
las
si
f
ica
ti
o
n
sy
ste
m
f
o
r
lu
n
g
n
o
d
u
les
o
n
c
o
m
p
u
ted
t
o
m
o
g
ra
p
h
y
sc
a
n
s,"
in
2
0
1
4
6
th
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
f
S
o
ft
Co
mp
u
ti
n
g
a
n
d
P
a
tt
e
rn
Rec
o
g
n
it
i
o
n
,
2
0
1
4
,
p
p
.
3
8
2
-
3
8
6
.
[1
9
]
A
d
a
¹
a
n
d
R.
Ka
u
r,
"
E
a
rl
y
d
e
tec
t
io
n
a
n
d
p
re
d
ictio
n
o
f
lu
n
g
c
a
n
c
e
r
su
rv
iv
a
l
u
sin
g
n
e
u
ra
l
n
e
t
w
o
rk
c
las
si
f
ier,"
e
d
:
IJA
IEM
,
2
0
1
3
.
[2
0
]
W
.
S
u
n
,
B.
Z
h
e
n
g
,
a
n
d
W
.
Qia
n
,
Co
mp
u
ter
a
i
d
e
d
l
u
n
g
c
a
n
c
e
r
d
ia
g
n
o
sis
wit
h
d
e
e
p
lea
rn
i
n
g
a
l
g
o
rit
h
ms
(S
P
I
E
M
e
d
ic
a
l
Im
a
g
in
g
).
S
P
IE,
2
0
1
6
.
[2
1
]
S
.
Hu
ss
e
in
,
R.
G
il
li
e
s,
K.
C
a
o
,
Q.
S
o
n
g
,
a
n
d
U.
Ba
g
c
i,
"
T
u
m
o
rNe
t
:
L
u
n
g
n
o
d
u
le
c
h
a
ra
c
teriz
a
ti
o
n
u
sin
g
m
u
lt
i
-
v
iew
Co
n
v
o
l
u
ti
o
n
a
l
Ne
u
ra
l
Ne
tw
o
rk
w
it
h
G
a
u
ss
ian
P
ro
c
e
ss
,
"
in
2
0
1
7
IE
EE
1
4
t
h
In
ter
n
a
ti
o
n
a
l
S
y
mp
o
si
u
m
o
n
Bi
o
me
d
ica
l
Ima
g
in
g
,
2
0
1
7
,
p
p
.
1
0
0
7
-
1
0
1
0
.
[2
2
]
A
.
T
e
ra
m
o
to
,
T
.
T
su
k
a
m
o
to
,
Y.
Kiriy
a
m
a
,
a
n
d
H.
F
u
ji
ta,
"
A
u
to
m
a
ted
Clas
sif
i
c
a
ti
o
n
o
f
L
u
n
g
Ca
n
c
e
r
Ty
p
e
s
f
ro
m
C
y
to
lo
g
ica
l
Im
a
g
e
s
Us
in
g
De
e
p
Co
n
v
o
l
u
ti
o
n
a
l
Ne
u
ra
l
Ne
tw
o
rk
s,
"
Bi
o
M
e
d
Res
e
a
rc
h
In
ter
n
a
t
io
n
a
l,
v
o
l.
2
0
1
7
,
p
.
4
0
6
7
8
3
2
,
2
0
1
7
.
[2
3
]
T
.
F
a
n
g
,
"
A
No
v
e
l
Co
m
p
u
ter
-
A
i
d
e
d
L
u
n
g
Ca
n
c
e
r
De
tec
ti
o
n
M
e
th
o
d
Ba
se
d
o
n
T
ra
n
sf
e
r
L
e
a
rn
in
g
f
r
o
m
G
o
o
g
LeN
e
t
a
n
d
M
e
d
ian
In
ten
sity
P
ro
jec
ti
o
n
s,"
2
0
1
8
IEE
E
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Co
mp
u
ter
a
n
d
Co
mm
u
n
ica
ti
o
n
En
g
i
n
e
e
rin
g
T
e
c
h
n
o
lo
g
y
(
CCET
),
p
p
.
2
8
6
-
2
9
0
,
2
0
1
8
.
[2
4
]
C.
S
z
e
g
e
d
y
e
t
a
l.
,
"
G
o
in
g
d
e
e
p
e
r
w
it
h
c
o
n
v
o
lu
ti
o
n
s,"
in
2
0
1
5
I
E
EE
Co
n
fer
e
n
c
e
o
n
Co
mp
u
ter
Vi
s
io
n
a
n
d
Pa
tt
e
rn
Rec
o
g
n
it
io
n
(
CVP
R)
,
2
0
1
5
,
p
p
.
1
-
9
.
[2
5
]
A
.
Ch
o
n
a
n
d
N.
Ba
lac
h
a
n
d
a
r,
"
De
e
p
Co
n
v
o
l
u
ti
o
n
a
l
Ne
u
ra
l
Ne
tw
o
rk
s f
o
r
L
u
n
g
Ca
n
c
e
r
De
te
c
ti
o
n
,
"
2
0
1
7
.
[2
6
]
Ka
g
g
l
e
.
Da
ta
S
c
ien
c
e
Bo
wl
2
0
1
7
,
Ka
g
g
le.
[
On
li
n
e
]
.
A
v
a
il
a
b
le:
h
tt
p
s://
ww
w
.
k
a
g
g
le.co
m
/c/d
a
ta
-
sc
ien
c
e
-
b
o
w
l
-
2017
[2
7
]
L
UN
A
.
L
u
n
g
n
o
d
u
le
a
n
a
lys
is
2
0
1
6
.
[
On
li
n
e
]
.
A
v
a
il
a
b
le:
h
tt
p
s://
l
u
n
a
1
6
.
g
ra
n
d
-
c
h
a
ll
e
n
g
e
.
o
rg
/
[2
8
]
H.
F
.
A
l
-
Ya
sri
y
,
M
.
S
.
A
l
-
Hu
sie
n
y
,
F
.
Y.
M
o
h
se
n
,
E
.
A
.
Kh
a
li
l,
a
n
d
Z.
S
.
Ha
ss
a
n
,
"
Dia
g
n
o
sis
o
f
L
u
n
g
Ca
n
c
e
r
Ba
se
d
o
n
CT
S
c
a
n
s Us
in
g
CNN
,
"
IOP
Co
n
fer
e
n
c
e
S
e
rie
s: M
a
ter
ia
l
s S
c
ien
c
e
a
n
d
E
n
g
i
n
e
e
rin
g
,
v
o
l.
9
2
8
,
2
0
2
0
.
[2
9
]
J.
G
.
Da
u
g
m
a
n
,
"
Co
m
p
lete
d
isc
re
te 2
-
D G
a
b
o
r
tran
sf
o
r
m
s b
y
n
e
u
ra
l
n
e
tw
o
rk
s
f
o
r
i
m
a
g
e
a
n
a
l
y
sis a
n
d
c
o
m
p
re
ss
io
n
,
"
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
Aco
u
st
ics
,
S
p
e
e
c
h
,
a
n
d
S
ig
n
a
l
Pro
c
e
ss
in
g
,
v
o
l.
3
6
,
n
o
.
7
,
p
p
.
1
1
6
9
-
1
1
7
9
,
1
9
8
8
.
[3
0
]
T
.
A
a
c
h
,
A
.
Ka
u
p
,
a
n
d
R.
M
e
ste
r,
"
On
tex
tu
re
a
n
a
l
y
sis:
L
o
c
a
l
e
n
e
rg
y
tran
s
f
o
r
m
s
v
e
rsu
s
q
u
a
d
ra
tu
re
f
il
ters
,
"
S
ig
n
a
l
P
ro
c
e
ss
in
g
,
v
o
l
.
4
5
,
n
o
.
2
,
p
p
.
1
7
3
-
1
8
1
,
1
9
9
5
.
[3
1
]
W
i
k
i
p
e
d
i
a
C
o
n
t
r
i
b
u
t
o
r
s
,
"
G
a
b
o
r
f
i
l
t
e
r
-
W
i
k
i
p
e
d
i
a
,
T
h
e
F
r
e
e
E
n
c
y
c
l
o
p
e
d
i
a
,
"
i
n
W
i
k
i
p
e
d
i
a
,
T
h
e
F
r
e
e
E
n
c
y
c
l
o
p
e
d
i
a
,
e
d
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
e
n
.
w
i
k
i
p
e
d
i
a
.
o
r
g
/
w
/
i
n
d
e
x
.
p
h
p
?
t
i
t
l
e
=
G
a
b
o
r
_
f
i
l
t
e
r
&
o
l
d
i
d
=
9
9
3
1
5
7
6
32
,
[
A
c
c
e
s
s
e
d
26
-
D
e
c
e
m
b
e
r
-
2020
]
.
[3
2
]
S
.
T
h
ru
n
a
n
d
L
.
P
ra
tt
,
L
e
a
rn
i
n
g
to
L
e
a
rn
.
S
p
r
in
g
e
r
US,
2
0
1
2
.
[3
3
]
A
.
Kriz
h
e
v
s
k
y
,
I.
S
u
tsk
e
v
e
r,
a
n
d
G
.
E.
Hin
to
n
,
"
Im
a
g
e
Ne
t
Clas
sif
ic
a
ti
o
n
w
it
h
De
e
p
Co
n
v
o
lu
ti
o
n
a
l
Ne
u
ra
l
Ne
tw
o
rk
s,"
in
NIPS
2
0
1
2
,
v
o
l.
A
d
v
a
n
c
e
s in
n
e
u
ra
l
in
f
o
rm
a
ti
o
n
p
r
o
c
e
ss
in
g
s
y
st
e
m
s 2
5
.
[3
4
]
m
a
t
h
w
o
r
k
s
.
c
o
m
.
"
T
r
a
n
s
f
e
r
L
e
a
r
n
i
n
g
U
s
i
n
g
P
r
e
t
r
a
i
n
e
d
N
e
t
w
o
r
k
.
"
[
O
n
l
i
n
e
]
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
u
k
.
m
a
t
h
w
o
r
k
s
.
c
o
m
/
h
e
l
p
/
d
e
e
p
l
e
a
r
n
i
n
g
/
u
g
/
t
r
a
n
s
f
e
r
-
l
e
a
r
n
i
n
g
-
u
s
i
n
g
-
p
r
e
t
r
a
i
n
e
d
-
n
e
t
w
o
r
k
.
h
t
m
l
.
(
A
c
c
e
s
s
e
d
2
5
2020.
[3
5
]
L
.
S
a
n
to
s.
"
A
rti
f
icia
l
In
e
ll
ig
e
n
c
e
.
"
G
it
Bo
o
k
.
[
On
li
n
e
]
.
A
v
a
il
a
b
le:
h
tt
p
s:/
/l
e
o
n
a
r
d
o
a
ra
u
j
o
sa
n
to
s
.
g
it
b
o
o
k
.
io
/artif
icia
l
-
in
telig
e
n
c
e
/ma
c
h
in
e
_
lea
rn
in
g
/d
e
e
p
_
lea
rn
i
n
g
/g
o
o
g
len
e
t
.
(
A
c
c
e
ss
e
d
2
0
2
0
).
[3
6
]
A
.
T
h
a
r
w
a
t,
"
Clas
si
f
ic
a
ti
o
n
a
ss
e
ss
m
e
n
t
m
e
th
o
d
s,"
Ap
p
li
e
d
Co
mp
u
ti
n
g
a
n
d
I
n
fo
rm
a
t
ics
,
2
0
2
0
.
d
o
i
:
1
0
.
1
0
1
6
/
j.
a
c
i.
2
0
1
8
.
0
8
.
0
0
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4752
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
22
,
No
.
2
,
Ma
y
2
0
2
1
:
1
0
7
8
-
1
0
8
6
1086
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
M
u
a
y
e
d
S
AL
-
H
u
se
i
n
y
re
c
e
iv
e
d
h
is
B
S
c
a
n
d
M
S
c
d
e
g
re
e
s
in
C
o
m
p
u
ter
a
n
d
S
o
f
tw
a
re
En
g
in
e
e
rin
g
f
o
r
m
AL
-
M
u
sta
n
siriy
a
Un
iv
e
rsit
y
,
Ira
q
,
in
2
0
0
2
a
n
d
2
0
0
5
re
sp
e
c
ti
v
e
l
y
.
He
re
c
e
iv
e
d
P
h
D
in
El
e
c
tro
n
ics
a
n
d
Co
m
p
u
ter
S
c
ien
c
e
f
ro
m
th
e
Un
iv
e
rsit
y
o
f
S
o
u
t
h
a
m
p
to
n
,
UK
,
in
2
0
1
2
.
He
re
c
e
iv
e
d
s
e
v
e
ra
l
re
s
e
a
rc
h
g
r
a
n
ts
a
s
w
e
ll
a
s
stu
d
y
sc
h
o
lars
h
ip
.
He
is
a
n
a
lu
m
n
u
s
o
f
th
e
F
u
lb
rig
h
t
V
isit
i
n
g
S
c
h
o
lars
p
r
o
g
ra
m
in
th
e
Un
iv
e
rsit
y
o
f
Ce
n
tral
O
k
lah
o
m
a
,
USA
,
in
2
0
1
4
.
Dr.
A
L
-
Hu
se
in
y
is
c
u
rre
n
tl
y
a
f
a
c
u
lt
y
m
e
m
b
e
r
a
n
d
De
a
n
o
f
Co
ll
e
g
e
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
I
n
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
in
W
a
sit
Un
iv
e
rsit
y
,
Ira
q
.
His
re
se
a
rc
h
in
tere
sts
in
c
l
u
d
e
c
o
m
p
u
ter
v
isio
n
,
m
a
c
h
in
e
lea
rn
in
g
,
b
i
o
m
e
d
ica
l
e
n
g
in
e
e
rin
g
,
Io
T
,
a
n
d
c
ry
p
to
g
ra
p
h
y
.
He
p
u
b
li
sh
e
d
se
v
e
ra
l
re
s
e
a
rc
h
p
a
p
e
rs
in
th
e
se
f
ield
s
a
n
d
a
u
th
o
re
d
a
b
o
o
k
i
n
c
ry
p
to
g
ra
p
h
y
.
His
c
u
rre
n
t
f
o
c
u
s
is
o
n
m
e
d
ica
l
in
f
o
rm
a
ti
c
s
a
n
d
d
e
v
e
lo
p
in
g
w
e
a
ra
b
les
f
o
r
m
e
d
ica
l
d
iag
n
o
sis
b
y
u
sin
g
m
a
c
h
in
e
lea
rn
in
g
.
Ahm
e
d
S
S
a
jit
is
a
f
a
c
u
lt
y
m
e
m
b
e
r
a
t
W
a
sit
u
n
iv
e
rsity
,
Ira
q
;
h
e
r
e
c
e
iv
e
d
h
is
B
S
c
in
Co
n
tr
o
l
a
n
d
Co
m
p
u
ter
En
g
in
e
e
rin
g
f
ro
m
th
e
Un
iv
e
rsit
y
o
f
T
e
c
h
n
o
lo
g
y
,
Ira
q
,
in
2
0
1
0
.
He
f
in
ish
e
d
h
is
M
S
c
in
Co
m
p
u
ter
En
g
in
e
e
rin
g
a
t
Ca
l
if
o
r
n
ia
S
tate
Un
iv
e
rsity
in
F
u
l
lerto
n
,
USA
,
in
2
0
1
7
.
He
w
a
s
g
ra
n
ted
a
n
a
ti
o
n
a
ll
y
p
re
stig
io
u
s
sc
h
o
lars
h
ip
f
ro
m
th
e
Ir
a
q
i
p
rim
e
m
in
ister’s
o
ff
ic
e
in
2
0
1
0
.
A
h
m
e
d
is
c
u
rre
n
tl
y
p
u
rs
u
in
g
h
is
P
h
D
in
p
h
o
r
o
n
ics
a
n
d
L
a
se
r
a
p
p
li
c
a
ti
o
n
s.
His
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
:
Co
m
p
u
ter
A
rc
h
it
e
c
tu
re
,
Ho
lo
g
r
a
m
s,
Ne
u
ra
l
Ne
tw
o
rk
s,
A
I,
A
l
g
o
rit
h
m
s,
a
n
d
P
h
a
se
Ch
a
n
g
e
M
a
teria
l
M
e
m
o
r
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.