T
E
L
K
O
M
N
I
K
A
T
elec
o
m
m
un
ica
t
io
n,
Co
m
pu
t
ing
,
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
19
,
No
.
4
,
A
u
g
u
s
t
2
0
2
1
,
p
p
.
1
2
8
4
~
1
2
9
0
I
SS
N:
1
6
9
3
-
6
9
3
0
,
ac
cr
ed
ited
First Gr
ad
e
b
y
Kem
en
r
is
tek
d
i
k
ti,
Dec
r
ee
No
: 2
1
/E/KPT
/2
0
1
8
DOI
: 1
0
.
1
2
9
2
8
/TE
L
KOM
NI
K
A.
v
1
9
i4
.
1
8
8
7
4
1284
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
Lung
cancer clas
s
ificatio
n bas
ed on
CT
sca
n ima
g
e b
y
apply
ing
FCM seg
men
tatio
n and
neu
ra
l net
wo
rk t
ec
hnique
Ahm
a
d Z
o
eba
d F
o
ea
dy
,
Siti
Ria
Riqm
a
wa
t
in
,
Dia
n Ca
nd
ra
Rini No
v
it
a
s
a
ri
De
p
a
rtme
n
t
o
f
M
a
t
h
e
m
a
ti
c
s,
UIN
S
u
n
a
n
Am
p
e
l
S
u
ra
b
a
y
a
,
In
d
o
n
e
sia
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Sep
7
,
2
0
2
0
R
ev
is
ed
J
an
8
,
2
0
2
1
Acc
ep
ted
J
an
2
1
,
2
0
2
1
Th
e
n
u
m
b
e
r
o
f
p
e
o
p
le
wi
th
lu
n
g
c
a
n
c
e
r
h
a
s
re
a
c
h
e
d
a
p
p
ro
x
im
a
tely
2
.
0
9
m
il
li
o
n
p
e
o
p
le
w
o
rld
wi
d
e
.
O
u
t
o
f
9
.
0
6
m
il
li
o
n
c
a
se
s
o
f
d
e
a
th
,
1
.
7
6
m
il
li
o
n
p
e
o
p
le
d
ie
d
u
e
t
o
l
u
n
g
c
a
n
c
e
r.
L
u
n
g
c
a
n
c
e
r
c
a
n
b
e
a
u
to
m
a
ti
c
a
ll
y
id
e
n
ti
fied
u
sin
g
a
c
o
m
p
u
ter
-
a
id
e
d
d
ia
g
n
o
sis
sy
st
em
(CAD
)
su
c
h
a
s
ima
g
e
p
ro
c
e
ss
in
g
.
Th
e
ste
p
s
tak
e
n
fo
r
e
a
rly
d
e
tec
ti
o
n
a
re
p
re
-
p
ro
c
e
ss
in
g
fe
a
tu
re
e
x
trac
ti
o
n
,
a
n
d
c
las
sifica
ti
o
n
.
P
re
-
p
r
o
c
e
ss
in
g
is
c
a
rried
o
u
t
i
n
se
v
e
ra
l
sta
g
e
s,
n
a
m
e
ly
g
ra
y
sc
a
le
ima
g
e
s,
n
o
ise
re
m
o
v
a
l,
a
n
d
c
o
n
tr
a
st
li
m
it
e
d
a
d
a
p
ti
v
e
h
ist
o
g
r
a
m
e
q
u
a
li
z
a
ti
o
n
.
Th
is
ima
g
e
fe
a
tu
re
e
x
trac
ted
u
si
n
g
g
ra
y
lev
e
l
c
o
-
o
c
c
u
rre
n
c
e
m
a
tri
x
(
G
LCM
)
a
n
d
c
las
sified
u
si
n
g
2
m
e
th
o
d
o
f
n
e
u
ra
l
n
e
two
rk
wh
ich
is
fe
e
d
fo
r
wa
rd
n
e
u
ra
l
n
e
two
rk
(F
F
NN
)
d
a
n
fe
e
d
b
a
c
k
w
a
rd
n
e
u
ra
l
n
e
two
r
k
(
F
BNN
).
T
h
i
s
re
se
a
rc
h
a
ims
to
o
b
tai
n
th
e
b
e
st
n
e
u
ra
l
n
e
t
wo
rk
m
o
d
e
l
to
c
las
sify
l
u
n
g
c
a
n
c
e
r
a
.
Ba
se
d
o
n
train
in
g
ti
m
e
a
n
d
a
c
c
u
ra
c
y
,
th
e
b
e
st
m
e
th
o
d
o
f
F
F
NN
is
k
e
rn
e
l
e
x
trem
e
lea
rn
in
g
m
a
c
h
in
e
(KE
LM
),
with
a
train
i
n
g
ti
m
e
o
f
1
2
se
c
o
n
d
s
a
n
d
a
n
a
c
c
u
ra
c
y
o
f
9
3
.
4
5
%
,
wh
il
e
t
h
e
b
e
st
m
e
th
o
d
o
f
F
BNN
is
Ba
c
k
p
r
o
p
a
g
a
ti
o
n
with
a
train
in
g
ti
m
e
o
f
1
8
m
in
u
tes
0
4
se
c
o
n
d
s a
n
d
a
n
a
c
c
u
ra
c
y
o
f
9
7
.
5
%
.
K
ey
w
o
r
d
s
:
ANFI
S
ELM
FC
M
GL
C
M
KE
L
M
L
u
n
g
ca
n
ce
r
R
NN
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
:
Dian
C
an
d
r
a
R
in
i N
o
v
itas
ar
i
Dep
ar
tm
en
t o
f
Ma
th
em
atics
UI
N
Su
n
an
Am
p
el
Su
r
a
b
ay
a
Ah
m
ad
Yan
i N
o
.
1
1
7
W
o
n
o
co
lo
Su
r
ab
ay
a
6
0
2
3
7
,
I
n
d
o
n
esia
E
m
ail:
d
ian
cr
in
i@
u
in
s
b
y
.
ac
.
id
1.
I
NT
RO
D
UCT
I
O
N
L
u
n
g
c
an
ce
r
is
a
m
alig
n
a
n
t
tu
m
o
r
th
at
ir
r
eg
u
lar
ly
g
r
o
ws
in
o
n
e
o
r
b
o
th
lu
n
g
s
with
a
n
ab
n
o
r
m
al
an
d
ir
r
eg
u
lar
d
ir
ec
tio
n
o
f
s
p
r
ea
d
.
C
an
ce
r
ce
lls
s
p
r
ea
d
f
r
o
m
th
e
l
u
n
g
s
th
r
o
u
g
h
t
h
e
b
lo
o
d
s
tr
ea
m
o
r
ly
m
p
h
f
lu
id
th
at
s
u
r
r
o
u
n
d
s
th
e
lu
n
g
tis
s
u
e.
C
a
n
ce
r
ce
lls
ca
n
also
s
p
r
ea
d
to
an
o
th
er
o
r
g
a
n
,
wh
ich
is
ca
lled
th
e
m
etastatic
p
r
o
ce
s
s
[
1
]
.
T
h
er
e
ar
e
f
o
u
r
ty
p
es
o
f
lu
n
g
ca
n
ce
r
th
at
is
s
q
u
am
o
u
s
ce
ll
ca
r
cin
o
m
a,
s
m
all
ce
ll
ca
r
cin
o
m
a,
lar
g
e
ce
ll
ca
r
cin
o
m
a,
an
d
a
d
en
o
ca
r
c
in
o
m
a
[
2
]
.
Acc
o
r
d
in
g
to
th
e
W
o
r
ld
Hea
lth
Or
g
an
izatio
n
(
W
HO)
in
its
r
ep
o
r
t o
n
ca
n
ce
r
wh
ich
s
tated
th
at
in
2
0
1
8
p
eo
p
le
with
lu
n
g
ca
n
ce
r
ac
co
u
n
ted
f
o
r
a
r
o
u
n
d
2
.
0
9
m
illi
o
n
p
eo
p
le
w
o
r
ld
wid
e,
an
d
1
.
7
6
m
illi
o
n
p
e
o
p
le
d
ied
o
f
lu
n
g
ca
n
ce
r
o
u
t
o
f
9
.
0
6
m
illi
o
n
ca
s
es
o
f
d
ea
th
[
3
]
.
T
h
e
Am
er
ican
C
an
ce
r
So
ciety
esti
m
ates
th
at
b
y
2
0
2
0
Am
er
ic
a
h
as
2
2
8
,
8
2
0
n
ew
ca
s
es
o
f
lu
n
g
ca
n
ce
r
an
d
1
3
5
,
7
2
0
d
ea
t
h
s
.
T
h
is
r
ep
o
r
t
is
b
ased
o
n
esti
m
ates
f
o
r
o
n
e
y
ea
r
i
n
2
0
2
0
[
4
]
.
T
h
is
d
is
ea
s
e
is
ch
a
r
ac
te
r
ized
b
y
p
ain
ac
c
o
m
p
an
ie
d
b
y
s
h
o
r
tn
ess
o
f
b
r
ea
th
b
ec
au
s
e
th
e
ca
n
ce
r
ce
lls
f
ill
t
h
e
s
p
ac
e
in
th
e
lu
n
g
s
,
an
d
th
e
ca
p
ac
ity
o
f
th
e
lu
n
g
s
f
o
r
ai
r
s
to
r
ag
e
is
g
ettin
g
n
ar
r
o
wer
[
5
]
.
T
h
is
d
is
ea
s
e
is
a
ls
o
a
f
r
eq
u
e
n
t
ca
u
s
e
o
f
o
th
e
r
c
an
ce
r
s
b
ec
a
u
s
e
it
s
p
r
ea
d
s
r
ap
i
d
ly
in
th
e
b
o
d
y
an
d
in
th
e
lu
n
g
s
[
6
]
.
L
u
n
g
ca
n
ce
r
is
o
f
ten
ca
lled
p
r
im
ar
y
ca
n
ce
r
b
e
ca
u
s
e
it
is
th
e
b
eg
in
n
i
n
g
o
f
th
e
f
o
r
m
atio
n
o
f
o
th
e
r
ca
n
ce
r
ce
lls
in
th
e
b
o
d
y
[
7
]
.
T
o
id
en
tif
y
th
is
d
is
ea
s
e
ca
n
b
e
d
o
n
e
b
y
d
o
in
g
a
c
o
m
p
u
ter
ize
d
to
m
o
g
r
ap
h
y
(
CT
)
s
ca
n
o
r
m
ag
n
etic
r
eso
n
a
n
ce
im
ag
in
g
(
MRI
)
.
Ho
we
v
er
,
m
o
s
t
p
atien
ts
p
r
ef
er
to
u
s
e
C
T
s
ca
n
s
b
ec
au
s
e
o
f
th
e
l
o
w
co
s
t a
n
d
r
elativ
ely
ac
cu
r
ate
r
e
s
u
lts
.
C
T
s
ca
n
is
a
co
m
p
ilatio
n
o
f
x
-
r
a
y
im
ag
es tak
en
f
r
o
m
d
if
f
er
en
t a
n
g
les with
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Lu
n
g
ca
n
ce
r
cla
s
s
ifica
tio
n
b
a
s
ed
o
n
C
T
s
ca
n
ima
g
e
b
y
… (
A
h
ma
d
Zo
eb
a
d
F
o
ea
d
y
)
1285
c
o
m
p
i
l
a
t
i
o
n
p
r
o
c
e
s
s
i
n
g
t
o
t
a
k
e
p
i
c
t
u
r
e
s
o
f
p
a
r
t
i
c
u
l
a
r
b
o
d
y
p
a
r
t
s
.
I
n
g
e
n
e
r
a
l
,
a
C
T
s
c
a
n
im
a
g
e
i
s
b
l
a
c
k
a
n
d
w
h
i
t
e
[
8
]
.
T
h
e
r
e
a
r
e
4
s
t
a
g
e
s
i
n
l
u
n
g
c
a
n
c
e
r
t
h
a
t
i
s
s
t
a
g
e
I
,
I
I
,
I
I
I
,
a
n
d
I
V
.
S
t
a
g
e
I
,
I
I
,
a
n
d
I
I
I
c
a
n
b
e
p
r
e
d
i
c
t
e
d
b
y
a
n
a
x
i
a
l
C
T
s
c
a
n
o
f
t
h
e
l
u
n
g
s
.
W
h
i
le
St
a
g
e
I
V
p
a
ti
e
n
ts
c
a
n
f
e
e
l
t
h
e
s
y
m
p
t
o
m
s
b
e
ca
u
s
e
t
h
e
c
a
n
c
e
r
h
a
s
s
p
r
e
a
d
t
h
r
o
u
g
h
o
u
t
t
h
e
b
o
d
y
a
n
d
c
a
n
n
o
t
b
e
i
d
e
n
t
i
f
i
e
d
f
r
o
m
t
h
e
C
T
s
c
a
n
[
9
]
.
S
t
a
g
e
I
l
u
n
g
c
a
n
c
e
r
i
s
s
a
i
d
t
o
b
e
b
e
n
i
g
n
b
e
c
a
u
s
e
i
t
d
o
e
s
n
o
t
k
i
ll
t
h
e
p
at
i
e
n
t
.
H
o
we
v
e
r
,
w
h
e
n
t
h
e
ca
n
c
e
r
r
e
a
c
h
es
s
ta
g
e
I
I
o
n
w
a
r
d
s
,
i
t
is
cl
as
s
i
f
i
e
d
as
m
a
li
g
n
a
n
t
b
e
c
a
u
s
e
i
t
h
a
s
a
m
o
r
t
a
li
t
y
r
a
t
e
o
f
6
0
-
7
0
%
,
a
n
d
i
s
d
i
f
f
i
c
u
lt
t
o
c
u
r
e
.
T
h
e
r
e
f
o
r
e
,
e
a
r
l
y
d
e
t
e
c
t
i
o
n
o
f
l
u
n
g
c
a
n
c
e
r
i
s
r
e
q
u
i
r
e
d
t
o
d
e
t
e
r
m
i
n
e
w
h
e
t
h
e
r
a
c
a
n
c
e
r
is
m
a
l
i
g
n
a
n
t
o
r
n
o
t
s
o
t
h
at
t
r
e
a
t
m
e
n
t
c
a
n
b
e
d
o
n
e
a
s
s
o
o
n
as
p
o
s
s
i
b
l
e
[
1
0
]
.
T
h
is
lu
n
g
ca
n
ce
r
ea
r
ly
d
etec
tio
n
s
y
s
tem
u
s
es
a
co
m
p
u
ter
-
aid
e
d
d
iag
n
o
s
is
s
y
s
tem
(
C
A
D)
s
u
ch
as
im
ag
e
p
r
o
ce
s
s
in
g
.
T
h
e
s
tep
s
tak
en
f
o
r
ea
r
ly
d
etec
tio
n
ar
e
p
r
e
-
p
r
o
ce
s
s
in
g
f
ea
tu
r
e
ex
tr
ac
tio
n
,
an
d
class
if
icatio
n
[
1
1
]
.
Pre
-
p
r
o
ce
s
s
in
g
is
ca
r
r
ie
d
o
u
t
in
s
ev
er
al
s
tag
es,
n
a
m
ely
,
g
r
ay
s
ca
le
im
a
g
es,
n
o
is
e
r
e
m
o
v
al,
h
is
to
g
r
am
eq
u
aliza
tio
n
,
an
d
Fu
zz
y
C
-
M
ea
n
s
(
FC
M)
s
eg
m
en
tatio
n
.
FC
M
is
th
e
b
est
clu
s
ter
m
eth
o
d
,
an
d
all
p
a
r
am
eter
s
m
u
s
t
b
e
p
r
e
-
d
eter
m
in
ed
[
1
2
]
.
B
alaf
ar
et
al.
s
tated
th
at
m
e
d
i
ca
l
im
ag
es
h
a
d
a
lo
t
o
f
n
o
is
e
an
d
in
h
o
m
o
g
en
eity
.
B
ased
o
n
th
e
p
a
p
er
h
e
wr
o
te,
it
p
r
o
v
ed
th
at
FC
M
was
th
e
b
est
s
eg
m
en
tatio
n
m
et
h
o
d
f
o
r
m
ed
ical
im
ag
e
d
ata
s
u
ch
as
C
T
s
ca
n
s
an
d
MRI
b
ec
au
s
e
it
co
u
ld
r
ed
u
ce
n
o
is
e
an
d
m
ax
im
ize
th
e
f
ea
t
u
r
e
an
d
b
a
ck
g
r
o
u
n
d
s
elec
ti
o
n
p
r
o
ce
s
s
[
1
3
]
.
Hu
an
g
et
a
l
.
o
n
ce
ap
p
lied
FC
M
s
eg
m
en
tatio
n
to
C
T
b
r
ain
im
ag
es
to
d
etec
t
b
r
ain
ca
n
ce
r
.
I
n
th
is
r
esear
ch
,
it
was
ex
p
lain
ed
th
at
FC
M
co
u
ld
e
x
tr
ac
t
th
e
m
ass
f
ea
tu
r
e
o
n
C
T
im
a
g
es
well
s
o
th
at
it
was
s
u
itab
le
f
o
r
u
s
e
in
ca
n
ce
r
d
etec
tio
n
[
1
4
]
.
Ap
a
r
t
f
r
o
m
C
T
s
ca
n
im
ag
es
o
f
th
e
b
r
ai
n
,
FC
M
s
eg
m
en
t
atio
n
h
as
also
b
ee
n
ap
p
lied
to
id
e
n
tify
l
u
n
g
ca
n
ce
r
,
as
was
d
o
n
e
b
y
Dh
awa
r
e
et
a
l
.
I
n
th
is
r
esear
ch
,
th
e
FC
M
s
eg
m
en
tatio
n
was
co
m
b
in
ed
wit
h
th
e
g
r
ay
lev
el
c
o
-
o
cc
u
r
r
en
ce
m
atr
ix
(
GL
C
M
)
m
eth
o
d
,
wh
ich
is
s
u
itab
le
f
o
r
e
x
tr
ac
tin
g
th
e
tex
tu
r
e
o
f
an
im
a
g
e.
T
h
e
ap
p
licatio
n
o
f
th
is
m
eth
o
d
h
as su
cc
ess
f
u
lly
id
en
tifie
d
th
e
m
ass
p
o
s
itio
n
[
1
5
]
.
I
m
ag
e
d
ata
will
b
e
ea
s
ier
to
p
r
o
ce
s
s
wh
en
it
is
n
u
m
er
ic
d
ata.
T
h
e
ap
p
licatio
n
o
f
f
ea
tu
r
e
ex
t
r
ac
tio
n
is
r
eq
u
ir
ed
to
c
r
ea
te
n
u
m
er
ical
d
ata
f
r
o
m
th
e
im
ag
e.
I
n
t
h
is
ca
s
e,
th
e
m
eth
o
d
u
s
ed
in
f
ea
tu
r
e
ex
tr
ac
tio
n
is
th
e
GL
C
M
[
1
6
]
.
T
h
e
n
ex
t
s
tag
e
is
th
e
class
if
icatio
n
.
T
h
is
s
tag
e
is
u
s
ed
to
d
eter
m
in
e
t
h
e
s
tag
e
o
f
lu
n
g
ca
n
ce
r
.
T
h
e
r
esu
lt
o
f
f
ea
tu
r
e
e
x
tr
ac
tio
n
f
r
o
m
th
e
GL
C
M
will
b
e
th
e
in
itial
in
p
u
t
o
f
th
is
p
r
o
ce
s
s
.
I
n
th
e
cl
ass
if
icatio
n
p
r
o
ce
s
s
it
s
elf
,
th
e
d
ata
will
b
e
d
iv
id
e
d
in
to
two
p
ar
ts
,
n
am
ely
tr
ai
n
in
g
d
ata
a
n
d
test
in
g
d
ata.
I
n
t
h
is
r
esear
ch
,
s
ev
er
al
n
eu
r
al
n
etwo
r
k
m
eth
o
d
s
ar
e
u
s
ed
to
f
o
r
m
t
h
e
b
est
m
o
d
el
o
f
a
s
y
s
tem
.
Neu
r
al
n
etwo
r
k
h
as
b
ee
n
ap
p
lied
to
C
T
s
ca
n
d
ata
b
y
T
h
an
am
m
al
an
d
Su
d
h
a
[
1
7
]
.
T
h
is
r
esear
c
h
u
s
ed
o
n
e
o
f
th
e
n
eu
r
al
n
etw
o
r
k
m
et
h
o
d
s
th
at
is
B
ac
k
p
r
o
p
ag
atio
n
.
T
h
is
r
esear
c
h
s
h
o
wed
th
at
th
e
ap
p
licatio
n
o
f
s
eg
m
en
tatio
n
co
u
l
d
ap
p
ly
ac
c
u
r
ac
y
to
th
e
n
e
u
r
al
n
etwo
r
k
m
eth
o
d
.
T
h
e
ac
c
u
r
ac
y
o
b
tain
ed
in
t
h
is
r
esear
ch
was
9
5
%
[
1
7
]
.
T
h
e
a
p
p
licatio
n
o
f
n
eu
r
al
n
etwo
r
k
f
o
r
lu
n
g
ca
n
ce
r
C
T
s
ca
n
d
ata
h
as
b
ee
n
co
n
d
u
cte
d
b
y
Ar
u
lm
u
r
u
g
an
et
a
l.
[
1
8
]
an
d
Sh
au
k
at
et
a
l
.
[
1
9
]
.
T
h
e
r
esear
ch
ex
p
lain
ed
th
at
th
e
n
eu
r
al
n
et
wo
r
k
co
u
ld
class
if
y
a
C
T
s
ca
n
im
ag
e
well
with
an
av
er
ag
e
ac
cu
r
ac
y
o
f
9
4
.
5
%.
Ho
wev
er
,
th
is
r
esear
ch
d
em
o
n
s
tr
ated
th
at
B
ac
k
p
r
o
p
ag
atio
n
h
ad
a
s
lo
w
tr
ain
in
g
tim
e
[
1
8
]
,
[
1
9
]
.
B
ased
o
n
th
is
ca
s
e,
lu
n
g
ca
n
ce
r
will
b
e
id
en
t
if
ied
b
ased
o
n
C
T
s
ca
n
d
ata
to
d
eter
m
in
e
th
e
b
est
n
eu
r
al
n
etw
o
r
k
m
o
d
el
th
at
ca
n
b
e
u
s
ed
.
T
h
is
r
esear
ch
will
co
n
d
u
ct
tr
ials
o
n
two
n
eu
r
al
n
et
w
o
r
k
m
o
d
els,
n
am
ely
f
ee
d
-
f
o
r
war
d
(
FF
NN)
an
d
f
ee
d
b
ac
k
war
d
(
FB
NN)
.
FF
NN
is
a
n
eu
r
al
n
etwo
r
k
th
at
s
en
d
s
d
ata
o
r
i
n
p
u
t
in
o
n
e
d
ir
ec
tio
n
,
th
at
is
th
r
o
u
g
h
th
e
in
p
u
t
n
o
d
e
an
d
o
u
t
at
th
e
o
u
tp
u
t
n
o
d
e
[
2
0
]
.
FB
NN
is
a
n
eu
r
al
n
etwo
r
k
th
at
s
en
d
s
d
ata
o
r
in
p
u
t
in
two
di
r
ec
tio
n
s
th
at
is
th
r
o
u
g
h
th
e
in
p
u
t
n
o
d
e
to
th
e
o
u
tp
u
t
n
o
d
e
a
n
d
b
ac
k
ag
ain
to
th
e
n
o
d
e
in
p
u
t
[
2
1
]
.
B
ased
o
n
s
ev
er
al
r
esear
ch
r
ev
iews
ab
o
v
e,
au
to
m
atic
d
etec
tio
n
o
f
lu
n
g
ca
n
ce
r
w
ill
b
e
ca
r
r
ied
o
u
t
u
s
in
g
b
o
th
m
eth
o
d
s
to
m
ax
im
ize
th
e
r
esu
lts
o
b
tain
ed
.
T
h
is
r
ese
ar
ch
aim
s
to
ac
h
iev
e
th
e
b
est
n
eu
r
al
n
etwo
r
k
m
o
d
el
to
class
if
y
lu
n
g
ca
n
ce
r
.
2.
P
RE
L
I
M
I
NAR
I
E
S
2
.
1
.
T
he
f
uzzy
C
-
m
ea
ns
s
eg
m
ent
a
t
io
n a
l
g
o
rit
hm
(
F
CM
)
Fu
zz
y
C
-
m
ea
n
s
(
FC
M)
is
a
d
ata
clu
s
ter
in
g
tech
n
i
q
u
e
in
wh
ich
th
e
ex
is
ten
ce
o
f
ea
c
h
d
ata
p
o
in
t
in
a
clu
s
ter
is
d
eter
m
in
ed
b
y
t
h
e
d
eg
r
ee
o
f
m
em
b
er
s
h
ip
.
FC
M
s
eg
m
en
tatio
n
is
th
e
s
ep
ar
atio
n
o
f
th
e
b
ac
k
g
r
o
u
n
d
with
f
ea
tu
r
es
b
y
clu
s
ter
in
g
th
e
im
ag
e
m
atr
ix
[
2
2
]
.
T
h
e
in
iti
al
s
tep
r
eq
u
ir
ed
is
th
e
in
itializatio
n
o
f
th
e
in
itial
FC
M
in
p
u
ts
s
u
ch
as
iter
atio
n
s
,
m
u
ltip
le
clu
s
ter
s
,
er
r
o
r
s
,
an
d
weig
h
ts
.
T
h
en
,
t
h
e
m
em
b
e
r
s
h
ip
m
atr
ix
(
)
is
in
itialized
r
an
d
o
m
ly
,
wh
er
e
is
th
e
ce
n
ter
o
f
th
e
clu
s
ter
is
t
h
e
in
p
u
t
m
atr
ix
,
is
th
e
in
itial
weig
h
t
with
th
e
d
ef
au
lt v
al
u
e
o
f
2
,
a
n
d
is
a
clu
s
ter
[2
3
]
.
T
h
e
ce
n
ter
o
f
th
e
clu
s
ter
is
ca
lcu
lated
u
s
in
g
(
1
)
.
=
∑
(
(
)
∗
)
=
1
∑
(
)
=
1
(
1
)
Af
ter
o
b
tain
i
n
g
th
e
clu
s
ter
ce
n
ter
,
th
e
n
th
e
o
b
jectiv
e
f
u
n
cti
o
n
is
ca
lcu
lated
u
s
in
g
th
e
f
o
r
m
u
la
s
h
o
wn
in
(
2
)
a
n
d
t
h
e
c
h
an
g
e
in
th
e
m
em
b
er
s
h
ip
v
alu
e
m
atr
ix
(
)
is
c
alcu
lated
u
s
in
g
(
3
)
.
T
h
e
iter
ati
o
n
is
s
aid
to
s
to
p
wh
en
th
e
m
in
im
u
m
er
r
o
r
o
r
m
ax
im
u
m
iter
atio
n
is
r
ea
ch
ed
[2
4
]
.
=
∑
∑
(
[
∑
(
−
)
2
=
1
]
(
)
)
=
1
=
1
(
2
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
4
,
Au
g
u
s
t 2
0
2
1
:
1
2
8
2
-
1
2
9
0
1286
=
[
∑
(
−
)
2
=
1
]
−
1
−
1
∑
[
∑
(
−
)
2
=
1
]
=
1
−
1
−
1
(
3
)
2
.
2
.
Neura
l
net
wo
r
k
Neu
r
al
n
etwo
r
k
s
ar
e
a
s
et
o
f
alg
o
r
ith
m
s
wh
ich
is
h
u
m
an
b
r
ain
m
o
d
eled
lik
e
an
d
d
esig
n
ed
t
o
r
ec
o
g
n
ize
p
atter
n
s
[2
5
]
.
L
i
k
e
th
e
b
r
ain
,
an
ar
tifi
cial
n
eu
r
al
n
etwo
r
k
is
a
co
llectio
n
o
f
co
n
n
ec
ted
u
n
i
ts
th
at
ca
n
also
b
e
ca
lled
n
eu
r
o
n
s
.
C
o
n
n
ec
tio
n
s
b
etwe
en
n
eu
r
o
n
s
in
an
ar
tific
ial
n
eu
r
al
n
etwo
r
k
ca
n
ca
r
r
y
s
ig
n
als
in
th
e
f
o
r
m
o
f
r
ea
l
v
alu
es
th
at
d
eter
m
i
n
e
th
e
weig
h
t
o
r
s
tr
en
g
t
h
o
f
th
e
s
ig
n
al
[2
6
]
.
I
n
p
u
t
d
ata
o
r
in
f
o
r
m
at
io
n
s
en
t
b
y
n
eu
r
o
n
s
ca
n
b
e
s
in
g
le
o
r
m
u
ltip
le.
W
h
er
e
is
th
e
n
eu
r
o
n
o
u
t
p
u
t,
is
th
e
s
ig
n
al
weig
h
t,
is
th
e
n
e
u
r
o
n
in
p
u
t,
an
d
is
th
e
b
ias
[2
7
]
.
T
h
er
e
ar
e
s
ev
er
a
l
n
eu
r
al
n
etwo
r
k
s
im
p
lem
en
te
d
b
ased
o
n
m
ath
em
atica
l
o
p
er
atio
n
s
,
an
d
a
s
et
o
f
p
ar
am
eter
s
is
r
eq
u
ir
e
d
to
d
eter
m
in
e
th
e
o
u
tp
u
t
[2
5
]
,
[
2
7
]
.
−
Feed
f
o
r
war
d
n
eu
r
al
n
etwo
r
k
(
FF
NN)
FF
NN
is
a
n
eu
r
al
n
etwo
r
k
th
at
s
en
d
s
d
ata
o
r
in
p
u
t in
o
n
e
d
ir
e
ctio
n
,
th
at
is
th
r
o
u
g
h
th
e
in
p
u
t
n
o
d
e
an
d
o
u
t
at
th
e
o
u
tp
u
t
n
o
d
e
.
T
h
er
e
a
r
e
s
ev
er
al
f
ee
d
-
f
o
r
war
d
n
eu
r
al
n
etwo
r
k
m
eth
o
d
s
,
th
at
is
r
ad
ial
b
asis
f
u
n
ctio
n
NN
(
R
B
FN
N)
,
ex
tr
em
e
lear
n
in
g
m
ac
h
in
e
(
E
L
M)
,
k
er
n
el
e
x
tr
em
e
lear
n
in
g
m
ac
h
in
e
(
KE
L
M)
,
a
n
d
p
er
ce
p
tr
o
n
[2
0
]
.
Fig
u
r
e
1
s
h
o
ws
th
e
d
esig
n
o
f
t
h
e
FF
NN
alg
o
r
ith
m
with
cir
cles
in
a
n
etwo
r
k
th
at
f
o
r
m
s
n
eu
r
o
n
s
in
an
ar
tific
ial
n
eu
r
al
n
etwo
r
k
[2
8
]
.
−
Feed
b
ac
k
war
d
n
eu
r
al
n
etwo
r
k
(
FB
NN)
FB
NN
is
a
n
eu
r
al
n
etwo
r
k
th
a
t
s
en
d
s
d
ata
o
r
in
p
u
t
in
two
d
ir
ec
tio
n
s
th
at
is
th
r
o
u
g
h
th
e
in
p
u
t
n
o
d
e
to
th
e
o
u
tp
u
t
n
o
d
e
an
d
b
ac
k
ag
ai
n
to
t
h
e
n
o
d
e
i
n
p
u
t.
T
h
er
e
ar
e
s
ev
er
al
b
ac
k
war
d
n
e
u
r
al
n
etwo
r
k
m
eth
o
d
s
t
h
at
is
B
ac
k
p
r
o
p
ag
atio
n
,
r
ec
u
r
r
en
t
n
eu
r
al
n
etwo
r
k
(
R
NN)
,
ad
a
p
tiv
e
n
e
u
r
o
-
f
u
zz
y
in
f
e
r
en
ce
s
y
s
tem
(
ANFI
S),
an
d
s
elf
-
o
r
g
an
izin
g
m
ap
(
SOM)
.
Fig
u
r
e
2
s
h
o
ws
th
e
d
esig
n
o
f
an
FB
NN
in
a
n
etwo
r
k
,
in
wh
ich
th
er
e
ar
e
n
o
d
es
(
cir
cles)
co
n
n
ec
ted
b
y
e
d
g
es
th
at
f
o
r
m
n
eu
r
o
n
s
in
a
d
u
m
m
y
n
etwo
r
k
.
T
h
e
m
o
v
em
e
n
t
in
FB
NN
o
cc
u
r
s
in
f
ee
d
b
ac
k
b
ec
au
s
e
th
e
in
p
u
t d
at
a
is
s
en
t in
two
d
ir
ec
tio
n
s
[
2
1
]
.
Fig
u
r
e
1
.
Feed
-
f
o
r
war
d
n
e
u
r
al
n
etwo
r
k
alg
o
r
ith
m
[2
5
]
Fig
u
r
e
2
.
Feed
b
ac
k
war
d
n
eu
r
al
n
etwo
r
k
alg
o
r
ith
m
[2
5
]
3.
RE
S
E
ARCH
M
E
T
H
O
D
3
.1
.
Resea
rc
h
t
y
pe
T
h
is
r
esear
ch
is
d
escr
ip
tiv
e
q
u
an
titativ
e
r
esear
ch
b
ec
a
u
s
e
it
in
v
o
lv
es
a
lo
t
o
f
ca
lcu
latio
n
s
to
f
in
d
o
u
t
th
e
r
esu
lts
,
an
d
th
e
d
ata
p
r
o
ce
s
s
in
g
m
u
s
t
b
e
an
al
y
ze
d
in
ea
c
h
s
tag
e.
C
alcu
latio
n
s
ar
e
ca
r
r
ied
o
u
t
in
all
p
r
o
ce
s
s
es,
wh
ile
d
ata
an
aly
s
is
is
ca
r
r
ied
o
u
t w
h
en
t
h
e
d
ata
h
as b
ee
n
p
r
o
ce
s
s
ed
an
d
o
b
tain
s
th
e
r
esu
lts
.
T
h
is
r
esear
ch
h
av
e
3
m
ain
p
r
o
ce
s
s
wh
ich
is
p
r
e
-
p
r
o
c
ess
in
g
,
f
ea
tu
r
e
e
x
tr
ac
tio
n
,
a
n
d
class
if
icatio
n
.
3
.2
.
Da
t
a
c
o
llect
io
n a
nd
a
na
ly
s
is
T
h
e
d
ata
was
o
b
tain
ed
f
r
o
m
a
ca
n
ce
r
im
ag
i
n
g
a
r
ch
iv
e
o
f
3
5
1
im
ag
es
co
n
s
is
tin
g
o
f
s
ev
e
r
al
s
tag
es
th
at
is
7
2
d
ata
o
f
s
tag
e
I
,
7
7
d
ata
o
f
s
tag
e
I
I
,
a
n
d
2
0
2
d
ata
o
f
s
tag
e
I
I
I
.
Fu
r
th
er
m
o
r
e,
th
e
d
at
a
is
p
r
o
ce
s
s
ed
in
to
p
r
e
-
p
r
o
ce
s
s
in
g
,
f
ea
t
u
r
e
e
x
tr
ac
tio
n
,
an
d
th
e
n
class
if
icatio
n
i
s
ca
r
r
ied
o
u
t
to
d
eter
m
in
e
wh
eth
er
th
e
ca
n
ce
r
is
m
alig
n
an
t
o
r
b
e
n
ig
n
.
T
esti
n
g
th
e
ev
al
u
atio
n
d
ata
b
e
g
in
s
with
th
e
f
ea
tu
r
e
ex
tr
ac
tio
n
p
r
o
ce
s
s
co
n
s
is
tin
g
o
f
g
r
ay
s
ca
le
im
ag
es,
n
o
is
e
r
em
o
v
al,
h
is
to
g
r
a
m
e
q
u
aliza
tio
n
,
a
n
d
FC
M
s
eg
m
en
tatio
n
.
T
h
en
t
h
e
f
ea
tu
r
es
ar
e
tak
en
u
s
in
g
th
e
GL
C
M
m
eth
o
d
,
an
d
f
in
ally
,
th
e
d
ata
f
r
o
m
th
e
f
ea
tu
r
e
ex
tr
ac
tio
n
is
u
s
ed
as
in
p
u
t
to
th
e
n
eu
r
al
n
etwo
r
k
class
if
icatio
n
.
T
h
e
en
tire
s
er
ie
s
o
f
r
esear
ch
s
tag
es
ca
n
b
e
s
e
en
in
th
e
f
lo
w
ch
ar
t
in
Fig
u
r
e
3
.
Af
ter
th
e
f
ea
tu
r
e
ex
tr
ac
t
io
n
r
esu
lts
ar
e
o
b
tain
ed
,
th
en
th
e
t
r
ain
in
g
d
ata
an
d
test
in
g
d
ata
ar
e
d
is
tr
ib
u
ted
to
b
e
in
clu
d
ed
i
n
th
e
n
e
u
r
al
n
etwo
r
k
m
eth
o
d
.
T
h
e
n
e
u
r
al
n
etwo
r
k
m
eth
o
d
s
u
s
ed
n
ex
t
a
r
e
d
iv
id
ed
in
to
two
th
at
is
FF
NN
an
d
FB
NN.
I
n
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Lu
n
g
ca
n
ce
r
cla
s
s
ifica
tio
n
b
a
s
ed
o
n
C
T
s
ca
n
ima
g
e
b
y
… (
A
h
ma
d
Zo
eb
a
d
F
o
ea
d
y
)
1287
FF
NN,
th
e
m
eth
o
d
s
u
s
ed
ar
e
E
L
M,
KE
L
M,
p
er
ce
p
t
r
o
n
,
an
d
R
B
FNN,
wh
ile
in
FB
NN,
th
e
m
eth
o
d
s
u
s
ed
ar
e
B
ac
k
p
r
o
p
ag
atio
n
,
R
NN,
ANF
I
S,
an
d
SOM.
T
h
e
tr
ain
in
g
d
at
a
is
u
s
ed
to
f
o
r
m
th
e
n
eu
r
al
n
e
two
r
k
m
eth
o
d
,
an
d
th
e
test
in
g
d
ata
is
u
s
ed
t
o
test
th
e
s
y
s
tem
ac
cu
r
ac
y
lev
el.
Fin
ally
,
th
e
d
ata
o
n
lu
n
g
ca
n
ce
r
i
s
d
iv
id
ed
in
to
t
h
r
ee
class
es th
at
i
s
th
e
s
tag
e
I
,
s
tag
e
I
I
,
an
d
s
tag
e
I
I
I
.
Fig
u
r
e
3
.
Gr
a
p
h
ical
ab
s
tr
ac
t o
f
lu
n
g
ca
n
ce
r
class
if
icatio
n
b
ased
o
n
C
T
s
ca
n
im
ag
e
by
a
p
p
ly
in
g
FC
M
s
eg
m
en
tatio
n
an
d
n
eu
r
al
n
etw
o
r
k
tech
n
iq
u
e
4.
RE
SU
L
T
AND
DI
SCUS
SI
O
N
Sev
er
al
p
r
o
ce
s
s
es
ar
e
r
eq
u
ir
e
d
to
d
etec
t
th
e
lev
el
o
f
m
ali
g
n
an
cy
o
f
lu
n
g
ca
n
ce
r
.
T
h
e
r
esu
lts
o
f
p
r
e
-
p
r
o
ce
s
s
in
g
ca
n
b
e
s
ee
n
in
Fig
u
r
e
4
.
Fig
u
r
e
4
(
a
)
is
a
g
r
ay
s
ca
le
im
ag
e
an
d
th
e
3
3
m
ed
ian
f
ilter
r
esu
lt
s
h
o
w
n
in
Fig
u
r
e
4
(
b
)
.
T
h
e
n
ex
t
s
tep
is
th
e
h
is
to
g
r
am
eq
u
aliza
tio
n
s
h
o
wn
in
Fig
u
r
e
4
(
c
)
.
T
h
e
last
s
tag
e
o
f
p
r
e
-
p
r
o
ce
s
s
in
g
is
th
e
FC
M
s
eg
m
en
tatio
n
wh
o
s
e
r
esu
lt
s
h
o
wn
in
Fig
u
r
e
4
(
d
)
,
wh
er
e
th
e
ex
i
s
tin
g
im
ag
e
h
as n
o
b
ac
k
g
r
o
u
n
d
,
an
d
o
n
ly
f
ea
tu
r
es
o
f
th
e
im
a
g
e
ar
e
tak
en
.
Dete
r
m
in
atio
n
o
f
clu
s
ter
s
in
th
e
FC
M
s
eg
m
en
tatio
n
i
n
th
is
r
esear
ch
u
s
in
g
a
tr
ial
s
y
s
tem
.
T
h
e
tr
ial
s
y
s
tem
aim
s
to
d
eter
m
in
e
th
e
b
est
f
ea
tu
r
e
an
d
b
ac
k
g
r
o
u
n
d
s
ep
ar
atio
n
.
T
h
is
r
esear
ch
u
s
es
a
tr
ial
o
f
th
e
n
u
m
b
e
r
o
f
cl
u
s
ter
s
2
,
3
,
an
d
5
.
T
h
e
b
est
r
esu
l
ts
ac
h
iev
ed
in
th
is
s
tu
d
y
ar
e
th
e
n
u
m
b
er
o
f
c
lu
s
ter
s
3
wh
ich
ca
n
b
e
s
ee
n
in
Fi
g
u
r
e
4
(
d
)
b
ec
au
s
e
it
h
as
h
ig
h
co
n
tr
ast
o
n
f
ea
tu
r
es
an
d
d
is
g
u
is
es
th
e
b
ac
k
g
r
o
u
n
d
im
ag
e.
Nex
t,
f
ea
tu
r
e
ex
tr
ac
tio
n
u
s
in
g
f
o
r
p
ar
a
m
eter
s
th
at
is
en
er
g
y
,
co
r
r
elatio
n
,
h
o
m
o
g
en
eity
,
a
n
d
co
n
tr
ast.
Sam
p
le
d
ata
f
r
o
m
f
ea
tu
r
e
ex
tr
ac
t
io
n
ca
n
b
e
s
ee
n
in
T
a
b
le
1
.
(
a)
(
b
)
(
c)
(
d
)
Fig
u
r
e
4
.
(
a
)
Gr
ay
s
ca
le,
(
b
)
M
ed
ian
Fil
ter
,
(
c)
His
to
g
r
am
E
q
u
aliza
tio
n
,
(
d
)
FC
M
Seg
m
en
tatio
n
T
h
e
n
ex
t
p
r
o
ce
s
s
is
class
if
icat
io
n
d
iv
id
e
d
in
to
two
s
y
s
tem
s
,
n
am
ely
FF
NN
an
d
FB
NN.
FF
NN
u
s
e
s
E
L
M,
KE
L
M,
R
B
FNN,
an
d
p
e
r
ce
p
tr
o
n
m
eth
o
d
s
,
wh
ile
FB
NN
u
s
es
B
ac
k
p
r
o
p
ag
atio
n
,
SO
M,
ANFI
S,
an
d
R
NN
m
eth
o
d
s
.
T
h
e
d
ata
f
r
o
m
th
e
f
ea
tu
r
e
ex
tr
ac
tio
n
is
d
iv
id
e
d
i
n
to
2
th
at
is
tr
ain
in
g
d
ata
an
d
test
in
g
d
ata.
Data
s
h
ar
in
g
u
s
es th
e
K
-
f
o
l
d
cr
o
s
s
v
alid
atio
n
m
eth
o
d
with
=
5
.
As
s
ee
n
f
r
o
m
T
ab
le
2
th
at
ea
c
h
m
eth
o
d
h
as
d
if
f
er
e
n
t
ac
cu
r
a
cy
v
alu
es.
T
h
e
p
ar
am
eter
s
u
s
e
d
f
o
r
FF
NN
an
d
FB
NN
m
eth
o
d
s
ar
e
th
e
s
am
e
th
at
is
h
id
d
en
n
o
d
es
o
f
1
0
0
.
T
h
e
r
esu
lts
o
b
tain
ed
ar
e
KE
L
M
is
th
e
b
est
m
eth
o
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
4
,
Au
g
u
s
t 2
0
2
1
:
1
2
8
2
-
1
2
9
0
1288
to
class
if
y
lu
n
g
ca
n
ce
r
f
r
o
m
FF
NN.
T
h
is
i
s
b
ec
au
s
e
KE
L
M
h
as
im
p
lem
en
ted
a
k
er
n
el
alg
o
r
ith
m
s
o
th
at
it
ca
n
m
ap
d
ata
to
a
h
ig
h
er
d
im
en
s
io
n
wh
ich
m
ak
es
it
ea
s
ier
f
o
r
th
e
s
y
s
tem
to
r
ec
o
g
n
ize
d
ata
p
atte
r
n
s
.
W
h
ile
th
e
b
est
FB
NN
m
et
h
o
d
is
B
ac
k
p
r
o
p
a
g
atio
n
.
T
h
is
is
b
ec
au
s
e
B
ac
k
p
r
o
p
a
g
atio
n
is
k
n
o
wn
as
wei
g
h
t
u
p
d
atin
g
,
wh
ich
allo
ws
th
e
s
y
s
tem
to
r
ec
o
g
n
ize
d
ata
p
atter
n
s
b
etter
i
n
ea
ch
ite
r
atio
n
.
T
h
e
wo
r
s
t
ac
cu
r
ac
y
am
o
n
g
th
e
m
eth
o
d
s
is
R
NN.
T
h
is
is
b
ec
au
s
e
R
NN
i
s
a
m
eth
o
d
b
ased
o
n
d
ata
s
eq
u
e
n
cin
g
an
d
it
is
m
o
r
e
s
u
itab
le
f
o
r
u
s
e
in
p
r
ed
ictio
n
s
th
at
h
av
e
a
co
r
r
elatio
n
b
etwe
en
d
ata.
T
h
e
r
esu
lts
o
f
th
e
co
m
p
ar
is
o
n
o
f
th
e
FF
NN
an
d
FB
NN
m
eth
o
d
s
ca
n
b
e
s
ee
n
in
Fig
u
r
e
s
5
an
d
6
,
r
esp
e
ctiv
ely
.
T
ab
le
1
.
Sam
p
le
o
f
GL
C
M
r
esu
lts
C
o
n
t
r
a
st
C
o
r
r
e
l
a
t
i
o
n
En
e
r
g
y
H
o
mo
g
e
n
e
i
t
y
La
b
e
l
0
.
5
5
4
4
0
.
7
6
8
1
0
.
9
0
1
4
0
.
9
7
6
2
S
t
a
g
e
I
0
.
4
8
8
1
0
.
7
6
8
3
0
.
9
2
9
2
0
.
9
8
4
4
1
.
1
0
0
3
0
.
7
1
2
0
0
.
8
3
7
6
0
.
9
5
4
4
0
.
5
4
2
5
0
.
8
0
9
3
0
.
8
9
4
3
0
.
9
7
7
5
0
.
4
7
5
0
0
.
6
0
1
8
0
.
9
3
9
6
0
.
9
8
3
2
S
t
a
g
e
I
I
0
.
4
9
7
1
0
.
7
3
5
8
0
.
9
2
6
4
0
.
9
8
1
8
0
.
4
3
0
1
0
.
7
4
8
3
0
.
9
3
1
2
0
.
9
8
3
9
0
.
6
0
8
3
0
.
6
9
1
1
0
.
9
1
7
5
0
.
9
7
7
6
0
.
5
1
2
4
0
.
7
4
9
0
0
.
9
2
3
6
0
.
9
8
1
5
S
t
a
g
e
I
I
I
0
.
5
1
5
6
0
.
6
7
1
5
0
.
9
3
4
3
0
.
9
8
2
2
0
.
4
5
1
3
0
.
8
9
4
2
0
.
9
3
4
2
0
.
9
8
4
6
0
.
6
1
4
2
0
.
6
0
2
4
0
.
8
5
3
5
0
.
9
7
0
3
T
ab
le
2
.
C
lass
if
icatio
n
r
esu
lt
o
f
FF
NN
an
d
FB
NN
with
1
0
0
h
id
d
en
n
o
d
es
F
F
N
N
F
B
N
N
M
e
t
h
o
d
F
o
l
d
A
c
c
u
r
a
c
y
M
e
t
h
o
d
F
o
l
d
A
c
c
u
r
a
c
y
ELM
F
o
l
d
1
8
8
.
6
5
%
S
O
M
F
o
l
d
1
9
1
.
5
0
%
F
o
l
d
2
8
8
.
0
0
%
F
o
l
d
2
9
2
.
0
0
%
F
o
l
d
3
9
2
.
0
0
%
F
o
l
d
3
7
5
.
4
5
%
F
o
l
d
4
9
0
.
5
0
%
F
o
l
d
4
7
7
.
4
5
%
F
o
l
d
5
8
2
.
4
5
%
F
o
l
d
5
8
0
.
0
0
%
M
a
x
9
2
.
0
0
%
M
a
x
9
2
.
0
0
%
K
EL
M
F
o
l
d
1
9
0
.
5
0
%
R
N
N
F
o
l
d
1
7
7
.
5
0
%
F
o
l
d
2
9
0
.
0
0
%
F
o
l
d
2
7
9
.
5
0
%
F
o
l
d
3
9
2
.
5
0
%
F
o
l
d
3
8
2
.
7
5
%
F
o
l
d
4
9
2
.
2
5
%
F
o
l
d
4
8
1
.
0
0
%
F
o
l
d
5
8
8
.
4
5
%
F
o
l
d
5
7
7
.
2
5
%
M
a
x
9
2
.
5
0
%
M
a
x
8
2
.
7
5
%
R
B
F
N
N
F
o
l
d
1
8
5
.
4
5
%
A
N
F
I
S
F
o
l
d
1
8
8
.
6
5
%
F
o
l
d
2
7
4
.
4
5
%
F
o
l
d
2
8
5
.
8
5
%
F
o
l
d
3
8
2
.
1
5
%
F
o
l
d
3
9
3
.
4
5
%
F
o
l
d
4
8
4
.
0
0
%
F
o
l
d
4
8
9
.
4
5
%
F
o
l
d
5
7
7
.
6
5
%
F
o
l
d
5
8
8
.
6
5
%
M
a
x
8
5
.
4
5
%
M
a
x
9
3
.
4
5
%
P
e
r
c
e
p
t
r
o
n
F
o
l
d
1
8
9
.
5
0
%
B
a
c
k
p
r
o
p
a
g
a
t
i
o
n
F
o
l
d
1
9
4
.
5
0
%
F
o
l
d
2
8
8
.
6
5
%
F
o
l
d
2
9
5
.
7
5
%
F
o
l
d
3
9
0
.
0
0
%
F
o
l
d
3
9
2
.
5
0
%
F
o
l
d
4
7
5
.
6
5
%
F
o
l
d
4
9
5
.
4
5
%
F
o
l
d
5
8
2
.
5
0
%
F
o
l
d
5
9
6
.
0
0
%
M
a
x
9
0
.
0
0
%
M
a
x
9
6
.
0
0
%
A
s
s
h
o
w
n
i
n
t
h
e
g
r
a
p
h
i
n
F
i
g
u
r
e
6
,
o
v
e
r
a
l
l
,
t
h
e
a
c
c
u
r
a
c
y
l
e
v
e
l
s
o
f
t
h
e
f
o
u
r
m
e
t
h
o
d
s
r
i
s
e
s
l
i
g
h
t
l
y
a
n
d
s
t
a
b
i
l
y
w
h
i
c
h
p
r
o
v
e
s
t
h
a
t
F
F
N
N
w
i
t
h
t
h
e
s
e
f
o
u
r
m
e
t
h
o
d
s
h
a
s
a
h
i
g
h
m
a
t
c
h
t
o
r
e
c
o
g
n
i
z
e
t
h
e
p
a
t
t
e
r
n
s
o
f
l
u
n
g
c
a
n
c
e
r
d
a
t
a
.
T
h
e
n
,
t
h
e
e
n
t
i
r
e
F
F
N
N
m
e
t
h
o
d
s
r
e
a
c
h
t
h
e
i
r
h
i
g
h
e
s
t
p
o
i
n
t
o
f
a
c
c
u
r
a
c
y
a
t
t
h
e
h
i
d
d
e
n
n
o
d
e
s
o
f
1
0
0
-
2
5
0
.
T
h
e
r
e
f
o
r
e
,
t
h
e
t
r
a
i
n
i
n
g
e
x
p
e
r
i
m
e
n
t
w
a
s
c
a
r
r
i
e
d
o
u
t
u
s
i
n
g
1
0
0
h
i
d
d
e
n
n
o
d
e
s
b
e
c
a
u
s
e
f
r
o
m
s
e
v
e
r
a
l
N
N
m
e
t
h
o
d
s
,
t
h
e
a
v
e
r
a
g
e
h
a
s
t
h
e
b
e
s
t
a
c
c
u
r
a
c
y
a
t
1
0
0
h
o
d
e
e
n
n
o
d
e
s
.
A
f
t
e
r
t
h
a
t
t
h
e
a
c
c
u
r
a
c
y
l
e
v
e
l
s
o
f
t
h
e
f
o
u
r
m
e
t
h
o
d
s
s
l
o
w
l
y
d
e
c
l
i
n
e
.
T
h
i
s
i
s
b
e
c
a
u
s
e
t
h
e
i
n
c
r
e
a
s
i
n
g
n
u
m
b
e
r
o
f
h
i
d
d
e
n
n
o
d
e
s
r
e
s
u
l
t
s
i
n
o
v
e
r
l
a
p
p
i
n
g
c
a
s
e
s
s
o
t
h
a
t
t
h
e
r
e
i
s
d
a
t
a
t
h
a
t
i
s
n
o
t
i
n
c
l
u
d
e
d
i
n
a
n
y
c
l
a
s
s
.
O
u
t
o
f
t
h
e
f
o
u
r
F
F
N
N
m
e
t
h
o
d
s
,
K
E
L
M
h
a
s
t
h
e
h
i
g
h
e
s
t
a
c
c
u
r
a
c
y
o
f
9
3
.
4
5
%
a
t
h
i
d
d
e
n
n
o
d
e
s
o
f
2
5
0
.
A
s
s
e
e
n
i
n
t
h
e
g
r
a
p
h
i
n
F
i
g
u
r
e
6
o
n
t
h
e
r
e
s
u
l
t
s
o
f
F
B
N
N
t
h
a
t
R
N
N
i
s
l
e
s
s
c
a
p
a
b
l
e
o
f
c
l
a
s
s
i
f
y
i
n
g
l
u
n
g
c
a
n
c
e
r
.
T
h
e
r
e
s
u
l
t
s
o
f
R
N
N
h
a
v
e
a
f
a
i
r
l
y
f
a
r
r
a
n
g
e
f
r
o
m
t
h
e
o
t
h
e
r
F
B
N
N
m
e
t
h
o
d
s
.
R
N
N
h
a
s
t
h
e
h
i
g
h
e
s
t
a
c
c
u
r
a
c
y
o
f
8
2
.
7
5
%
a
t
t
h
e
h
i
d
d
e
n
n
o
d
e
s
o
f
1
0
0
.
T
h
e
A
N
F
I
S
a
n
d
S
O
M
m
e
t
h
o
d
s
h
a
v
e
a
h
i
g
h
a
c
c
u
r
a
c
y
a
t
h
i
d
d
e
n
n
o
d
e
1
0
0
w
i
t
h
a
l
m
o
s
t
t
h
e
s
a
m
e
a
c
c
u
r
a
c
y
o
f
9
3
.
4
5
%
f
o
r
A
N
F
I
S
a
n
d
9
2
%
f
o
r
S
O
M
.
I
n
t
h
i
s
r
e
s
e
a
r
c
h
,
t
h
e
h
i
g
h
e
s
t
a
c
c
u
r
a
c
y
o
f
F
B
N
N
w
a
s
a
c
h
i
e
v
e
d
u
s
i
n
g
t
h
e
b
a
c
k
p
r
o
p
a
g
a
t
i
o
n
m
e
t
h
o
d
w
i
t
h
t
h
e
h
i
g
h
e
s
t
a
c
c
u
r
a
c
y
o
f
9
7
.
5
%
.
I
n
a
d
d
i
t
i
o
n
t
o
a
c
c
u
r
a
c
y
,
i
t
i
s
n
e
c
e
s
s
a
r
y
t
o
p
a
y
a
t
t
e
n
t
i
o
n
t
o
t
h
e
p
e
r
f
o
r
m
a
n
c
e
o
f
a
m
o
d
e
l
b
a
s
e
d
o
n
t
h
e
r
e
q
u
i
r
e
d
t
r
a
i
n
i
n
g
t
i
m
e
.
T
h
e
r
e
s
u
l
t
s
o
n
t
r
a
i
n
i
n
g
t
i
m
e
c
a
n
b
e
s
e
e
n
i
n
T
a
b
l
e
3
.
T
h
e
s
e
r
e
s
u
l
t
s
a
r
e
a
l
s
o
c
o
m
p
a
r
e
d
w
i
t
h
s
y
s
t
e
m
s
t
h
a
t
d
o
n
o
t
i
m
p
l
e
m
e
n
t
F
C
M
s
e
g
m
e
n
t
a
t
i
o
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Lu
n
g
ca
n
ce
r
cla
s
s
ifica
tio
n
b
a
s
ed
o
n
C
T
s
ca
n
ima
g
e
b
y
… (
A
h
ma
d
Zo
eb
a
d
F
o
ea
d
y
)
1289
T
ab
le
3
s
h
o
ws
th
at
th
e
ap
p
licatio
n
o
f
FC
M
s
eg
m
en
tatio
n
ca
n
af
f
ec
t
th
e
r
esu
lts
o
f
th
e
ac
c
u
r
ac
y
o
f
th
e
s
y
s
tem
b
ein
g
cr
ea
ted
.
T
h
is
is
s
h
o
wn
b
y
c
o
m
p
ar
i
n
g
s
y
s
tem
s
th
at
u
s
e
FC
M
s
eg
m
en
tatio
n
an
d
s
y
s
tem
s
th
at
u
s
e
th
r
esh
o
ld
in
g
.
T
h
e
o
v
er
all
NN
m
eth
o
d
h
as
b
etter
ac
cu
r
ac
y
if
th
er
e
is
FC
M
s
eg
m
en
tatio
n
.
T
ab
le
3
also
r
e
v
ea
ls
th
at
th
e
FF
NN
m
eth
o
d
d
o
es
n
o
t
r
eq
u
ir
e
lo
n
g
tr
ai
n
in
g
tim
e,
w
h
ile
th
e
FB
NN
m
eth
o
d
r
e
q
u
ir
e
s
lo
n
g
tr
ain
in
g
tim
e.
T
h
is
is
b
ec
au
s
e
th
e
FB
NN
m
eth
o
d
r
eq
u
i
r
es iter
atio
n
in
th
e
tr
ain
in
g
p
r
o
ce
s
s
,
wh
ile
t
h
e
FF
NN
m
eth
o
d
d
o
es n
o
t.
B
ased
o
n
tr
ain
in
g
tim
e
a
n
d
ac
cu
r
ac
y
,
th
e
b
est
FF
NN
m
eth
o
d
is
KE
L
M
with
a
tr
ain
in
g
tim
e
o
f
1
2
s
ec
o
n
d
s
an
d
an
ac
cu
r
ac
y
o
f
9
3
.
4
5
%,
wh
ile
th
e
b
est
F
B
NN
m
eth
o
d
is
B
a
ck
p
r
o
p
ag
atio
n
with
a
tr
ain
in
g
tim
e
o
f
1
8
m
in
u
tes
0
4
s
ec
o
n
d
s
an
d
an
ac
cu
r
ac
y
o
f
9
7
.
5
%.
T
h
e
r
esu
lts
o
f
a
s
y
s
t
em
ca
n
also
b
e
s
ee
n
f
r
o
m
t
h
e
er
r
o
r
.
E
r
r
o
r
ca
n
b
e
d
ef
in
ed
as
th
e
h
ig
h
est
ac
cu
r
a
cy
wh
ich
is
(
)
=
100%
−
an
d
it
ca
n
b
e
co
n
clu
d
ed
t
h
at
th
e
b
est
FF
NN
m
eth
o
d
,
KE
L
M,
h
as
an
er
r
o
r
o
f
6
.
5
5
%,
wh
ile
th
e
b
est
FB
N
N
m
eth
o
d
,
n
am
ely
B
ac
k
p
r
o
p
a
g
atio
n
,
h
as
an
er
r
o
r
o
f
2
.
5
%.
I
n
th
is
r
esear
ch
,
r
esear
ch
es
s
till
lack
ed
in
p
r
e
-
p
r
o
ce
s
s
in
g
s
tag
e
b
ec
au
s
e
th
e
d
ata
u
s
ed
wer
e
n
o
t
b
ala
n
ce
d
.
T
h
er
ef
o
r
e,
d
ata
a
u
g
m
en
tatio
n
is
n
ee
d
ed
t
o
o
v
e
r
co
m
e
th
e
im
b
alan
ce
o
f
th
e
d
ata.
On
e
o
f
th
e
m
et
h
o
d
s
th
at
ca
n
b
e
u
s
ed
f
o
r
d
ata
au
g
m
en
tatio
n
is
co
n
d
itio
n
al
g
en
e
r
ativ
e
ad
v
er
s
ial
n
etwo
r
k
s
(
G
ANs)
co
n
d
u
cted
b
y
Sh
an
q
in
g
Gu
.
Ma
n
is
h
a
Ped
n
e
k
ar
,
an
d
R
o
b
er
t Slater
[2
9
].
Fig
u
r
e
5
.
Gr
a
p
h
o
f
FF
NN
r
esu
lt
Fig
u
r
e
6
.
Gr
a
p
h
o
f
FB
NN
r
esu
lt
T
ab
le
3
.
T
r
ai
n
in
g
tim
e
a
n
d
b
est ac
cu
r
ac
y
co
m
p
ar
is
o
n
M
o
d
e
l
M
e
t
h
o
d
N
u
mb
e
r
o
f
H
i
d
d
e
n
N
o
d
e
F
C
M
S
e
g
me
n
t
a
t
i
o
n
Th
r
e
s
h
o
l
d
i
n
g
A
c
c
u
r
a
c
y
Tr
a
i
n
i
n
g
T
i
me
A
c
c
u
r
a
c
y
Tr
a
i
n
i
n
g
T
i
me
F
F
N
N
ELM
1
0
0
9
2
.
0
0
%
5
se
c
o
n
d
8
9
.
6
0
%
5
se
c
o
n
d
K
EL
M
2
5
0
9
3
.
4
5
%
1
2
se
c
o
n
d
9
0
.
5
0
%
1
0
se
c
o
n
d
R
B
F
N
N
2
0
0
8
6
.
5
0
%
8
se
c
o
n
d
8
2
.
0
0
%
8
se
c
o
n
d
P
e
r
c
e
p
t
r
o
n
1
5
0
9
2
.
1
5
%
1
1
se
c
o
n
d
8
9
.
7
5
%
1
0
se
c
o
n
d
F
B
N
N
S
O
M
2
0
0
9
1
.
4
5
%
1
4
mi
n
u
t
e
s
3
2
sec
o
n
d
8
5
.
1
5
%
1
3
mi
n
u
t
e
s
1
2
sec
o
n
d
R
N
N
5
0
0
8
2
.
0
0
%
2
8
mi
n
u
t
e
s
1
2
sec
o
n
d
7
7
.
5
0
%
2
7
mi
n
u
t
e
s
4
8
sec
o
n
d
A
N
F
I
S
1
0
0
9
3
.
4
5
%
1
2
mi
n
u
t
e
s
4
1
sec
o
n
d
9
1
.
0
0
%
1
0
mi
n
u
t
e
s
5
3
sec
o
n
d
B
a
c
k
p
r
o
p
a
g
a
t
i
o
n
1
5
0
9
7
.
5
0
%
1
8
mi
n
u
t
e
s
0
4
sec
o
n
d
9
6
.
5
0
%
1
7
mi
n
u
t
e
s
1
5
sec
o
n
d
5.
CO
NCLU
SI
O
N
T
h
is
r
esear
ch
aim
s
to
o
b
tain
th
e
b
est
n
eu
r
al
n
etwo
r
k
m
o
d
el
to
class
if
y
lu
n
g
ca
n
ce
r
.
T
h
e
FC
M
s
eg
m
en
tatio
n
h
as
s
u
cc
ee
d
e
d
i
n
in
cr
ea
s
in
g
t
h
e
ac
cu
r
ac
y
o
f
t
h
e
lu
n
g
ca
n
ce
r
class
if
icatio
n
s
y
s
tem
.
T
h
is
ex
p
lain
s
th
at
th
e
FC
M
s
eg
m
en
tatio
n
is
ab
le
to
id
en
tify
t
h
e
m
ass
in
lu
n
g
ca
n
ce
r
s
o
t
h
at
th
e
class
if
ier
ca
n
r
ec
o
g
n
ize
th
e
p
atter
n
well.
B
ased
o
n
tr
ain
in
g
tim
e
an
d
ac
cu
r
ac
y
,
th
e
b
est FF
NN
m
eth
o
d
is
KE
L
M
with
a
tr
ain
in
g
tim
e
o
f
1
2
s
ec
o
n
d
s
an
d
an
ac
cu
r
a
cy
o
f
9
3
.
4
5
%,
wh
ile
th
e
b
est
FB
NN
m
eth
o
d
is
B
ac
k
p
r
o
p
ag
atio
n
with
a
tr
ain
in
g
tim
e
o
f
1
8
m
in
u
tes 0
4
s
ec
o
n
d
s
a
n
d
an
ac
cu
r
ac
y
o
f
9
7
.
5
%.
RE
F
E
R
E
NC
E
S
[1
]
P
.
Ka
l
a
v
a
th
i
a
n
d
A.
D
h
a
v
a
p
a
n
d
iam
m
a
l,
“
S
e
g
m
e
n
tatio
n
o
f
Lu
n
g
Tu
m
o
r
i
n
C
T
S
c
a
n
Im
a
g
e
s
u
si
n
g
F
A
-
F
CM
Alg
o
rit
h
m
s,”
v
o
l.
1
8
,
n
o
.
5
,
p
p
.
7
4
-
7
9
,
2
0
1
6
,
d
o
i:
1
0
.
9
7
9
0
/
0
6
6
1
-
1
8
0
5
0
4
7
4
7
9
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
4
,
Au
g
u
s
t 2
0
2
1
:
1
2
8
2
-
1
2
9
0
1290
[2
]
K.
In
a
m
u
ra
,
“
Lu
n
g
c
a
n
c
e
r:
u
n
d
e
rsta
n
d
i
n
g
it
s
m
o
lec
u
lar
p
a
th
o
lo
g
y
a
n
d
t
h
e
2
0
1
5
WHO
c
las
s
ifi
c
a
ti
o
n
,
”
Fro
n
t.
O
n
c
o
l.
,
v
o
l.
7
,
p
.
1
9
3
,
2
0
1
7
,
d
o
i:
1
0
.
3
3
8
9
/
fo
n
c
.
2
0
1
7
.
0
0
1
9
3
.
[3
]
A.
A.
Ad
jei,
“
Lu
n
g
c
a
n
c
e
r
wo
rl
d
wid
e
,
”
J
.
T
h
o
ra
c
.
O
n
c
o
l.
,
v
o
l.
1
4
,
n
o
.
6
,
p
.
9
5
6
,
2
0
1
9
.
[4
]
T
h
e
A
m
e
r
ic
a
n
Ca
n
c
e
r
S
o
c
ie
t
y
m
e
d
i
c
a
l
a
n
d
e
d
i
t
o
r
i
a
l
c
o
n
t
e
n
t
t
e
a
m
,
“
K
e
y
S
t
a
t
i
s
t
ic
s
f
o
r
L
u
n
g
C
a
n
c
e
r
,
”
Am
e
r
ic
a
n
C
a
n
c
e
r
S
o
c
i
e
t
y
,
2
0
2
0
.
h
t
t
p
s
:
/
/
w
w
w
.
c
a
n
c
e
r
.
o
r
g
/
c
a
n
c
e
r
/
l
u
n
g
-
c
a
n
c
e
r
/
a
b
o
u
t
/
k
e
y
-
s
t
a
t
i
s
t
ic
s
.
h
t
m
l
(a
c
c
e
ss
e
d
A
u
g
.
0
9
,
2
0
2
0
)
.
[5
]
M
.
M
u
fti
,
S
.
C
h
i
n
g
,
S
.
F
a
rjam
i,
S
.
S
h
a
h
a
n
g
ian
,
a
n
d
S
.
S
o
b
n
o
sk
y
,
“
A
c
a
se
se
ries
o
f
two
p
a
ti
e
n
ts
p
re
se
n
ti
n
g
wi
th
p
e
rica
rd
ial
e
ffu
si
o
n
a
s
first
m
a
n
i
fe
sta
ti
o
n
o
f
n
o
n
-
sm
a
ll
c
e
ll
l
u
n
g
c
a
n
c
e
r
with
BRA
F
m
u
tatio
n
a
n
d
e
x
p
re
ss
io
n
o
f
PD
-
L1
,
”
W
o
rl
d
J
.
On
c
o
l.
,
v
o
l.
9
,
n
o
.
2
,
p
.
5
6
,
2
0
1
8
,
d
o
i:
1
0
.
1
4
7
4
0
/
wjo
n
1
0
9
2
w.
[6
]
V.
W.
Q.
Lo
u
e
t
a
l.
,
“
Re
sp
irato
r
y
sy
m
p
to
m
s,
sle
e
p
,
a
n
d
q
u
a
li
ty
o
f
li
fe
in
p
a
ti
e
n
ts
wit
h
a
d
v
a
n
c
e
d
l
u
n
g
c
a
n
c
e
r,
”
J
.
Pa
in
S
y
mp
to
m
M
a
n
a
g
e
.
,
v
o
l
.
5
3
,
n
o
.
2
,
p
p
.
2
5
0
–
2
5
6
,
2
0
1
7
,
d
o
i
:
1
0
.
1
0
1
6
/
j.
jp
a
in
s
y
m
m
a
n
.
2
0
1
6
.
0
9
.
0
0
6
.
[7
]
M
.
P
.
Ya
v
r
o
p
o
u
l
o
u
e
t
a
l.
,
“
Dista
n
t
lu
n
g
m
e
tas
tas
e
s
c
a
u
se
d
b
y
a
h
isto
lo
g
ica
ll
y
b
e
n
i
g
n
p
h
o
sp
h
a
t
u
ric
m
e
se
n
c
h
y
m
a
l
tu
m
o
r,
”
E
n
d
o
c
rin
o
l.
d
ia
b
e
tes
M
e
t
a
b
.
c
a
se
re
p
o
rts
,
v
o
l
.
2
0
1
8
,
n
o
.
1
,
2
0
1
8
,
d
o
i:
1
0
.
1
5
3
0
/E
DM
-
18
-
0
0
2
3
.
[8
]
S
.
M
a
k
a
ju
,
P
.
W.
C.
P
ra
sa
d
,
A.
Als
a
d
o
o
n
,
A.
K.
S
in
g
h
,
a
n
d
A.
E
lch
o
u
e
m
i,
“
Lu
n
g
c
a
n
c
e
r
d
e
tec
ti
o
n
u
sin
g
CT
sc
a
n
ima
g
e
s,”
Pro
c
e
d
ia
C
o
mp
u
t.
S
c
i.
,
v
o
l.
1
2
5
,
p
p
.
1
0
7
-
1
1
4
,
2
0
1
8
,
d
o
i:
1
0
.
1
0
1
6
/
j.
p
r
o
c
s.2
0
1
7
.
1
2
.
0
1
6
.
[9
]
S
.
P
a
r
k
e
t
a
l
.
,
“
Vo
lu
m
e
d
o
u
b
l
in
g
ti
m
e
s
o
f
l
u
n
g
a
d
e
n
o
c
a
rc
in
o
m
a
s:
c
o
rre
latio
n
wit
h
p
re
d
o
m
in
a
n
t
h
ist
o
lo
g
ic
su
b
ty
p
e
s
a
n
d
p
r
o
g
n
o
sis,”
Ra
d
io
l
o
g
y
,
v
o
l.
2
9
5
,
n
o
.
3
,
p
p
.
7
0
3
-
7
1
2
,
2
0
2
0
,
d
o
i:
1
0
.
1
1
4
8
/ra
d
io
l
.
2
0
2
0
1
9
1
8
3
5
.
[1
0
]
L.
Ni
e
t
a
l
.
,
“
He
a
rt
V5
p
re
d
icts
c
a
rd
iac
e
v
e
n
ts
in
u
n
re
se
c
tab
le
lu
n
g
c
a
n
c
e
r
p
a
ti
e
n
ts
u
n
d
e
r
g
o
i
n
g
c
h
e
m
o
ra
d
iatio
n
,
”
J
.
T
h
o
r
a
c
.
D
is.
,
v
o
l
.
1
1
,
n
o
.
6
,
p
.
2
2
2
9
,
2
0
1
9
,
d
o
i:
1
0
.
2
1
0
3
7
/
jt
d
.
2
0
1
9
.
0
6
.
2
9
.
[1
1
]
M
.
Th
o
h
i
r,
A.
Z.
F
o
e
a
d
y
,
D.
C
.
R
.
No
v
it
a
sa
ri,
A.
Z.
Arifi
n
,
B.
Y.
P
h
iad
e
lv
ira,
a
n
d
A.
H
.
As
y
h
a
r,
“
Cl
a
ss
ifi
c
a
ti
o
n
o
f
Co
lp
o
sc
o
p
y
Da
ta
Us
in
g
G
LCM
-
S
VM
o
n
Ce
r
v
ica
l
Ca
n
c
e
r,
”
in
2
0
2
0
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Arti
fi
c
ia
l
In
telli
g
e
n
c
e
in
I
n
fo
rm
a
ti
o
n
a
n
d
C
o
mm
u
n
ic
a
ti
o
n
(ICAI
IC)
,
2
0
2
0
,
p
p
.
3
7
3
-
3
7
8
.
[1
2
]
A.
Ka
p
o
o
r
a
n
d
A.
S
i
n
g
h
a
l,
“
A
c
o
m
p
a
ra
ti
v
e
stu
d
y
o
f
K
-
M
e
a
n
s,
K
-
M
e
a
n
s+
+
a
n
d
F
u
z
z
y
C
-
M
e
a
n
s
c
lu
ste
rin
g
a
lg
o
rit
h
m
s,”
in
2
0
1
7
3
r
d
i
n
ter
n
a
ti
o
n
a
l
c
o
n
fer
e
n
c
e
o
n
c
o
mp
u
ta
t
i
o
n
a
l
i
n
telli
g
e
n
c
e
&
c
o
mm
u
n
ic
a
t
io
n
tec
h
n
o
lo
g
y
(CICT
)
,
2
0
1
7
,
p
p
.
1
-
6
,
d
o
i
:
1
0
.
1
1
0
9
/CIACT.
2
0
1
7
.
7
9
7
7
2
7
2
.
[1
3
]
M
.
A.
Ba
lafa
r,
“
F
u
z
z
y
C
-
m
e
a
n
b
a
se
d
b
ra
in
M
RI
se
g
m
e
n
tatio
n
a
lg
o
r
it
h
m
s,”
Arti
f.
I
n
tell.
Rev
.
,
v
o
l.
4
1
,
n
o
.
3
,
p
p
.
4
4
1
-
4
4
9
,
2
0
1
4
,
d
o
i:
1
0
.
1
0
0
7
/s
1
0
4
6
2
-
0
1
2
-
9
3
1
8
-
2.
[1
4
]
H.
Hu
a
n
g
,
F
.
M
e
n
g
,
S
.
Zh
o
u
,
F
.
Jia
n
g
,
a
n
d
G
.
M
a
n
o
g
a
ra
n
,
“
Bra
i
n
ima
g
e
se
g
m
e
n
tatio
n
b
a
se
d
o
n
F
CM
c
lu
ste
rin
g
a
lg
o
rit
h
m
a
n
d
ro
u
g
h
se
t,
”
IEE
E
A
c
c
e
ss
,
v
o
l.
7
,
p
p
.
1
2
3
8
6
-
1
2
3
9
6
,
2
0
1
9
,
d
o
i:
1
0
.
1
1
0
9
/ACCES
S
.
2
0
1
9
.
2
8
9
3
0
6
3
.
[1
5
]
B.
U.
Dh
a
wa
re
a
n
d
A.
C.
P
ise
,
“
Lu
n
g
c
a
n
c
e
r
d
e
tec
ti
o
n
u
si
n
g
b
a
y
a
se
in
c
las
sifier
a
n
d
F
CM
se
g
m
e
n
tatio
n
,
”
i
n
2
0
1
6
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
A
u
to
ma
ti
c
Co
n
tro
l
a
n
d
Dy
n
a
mic
Op
ti
miza
ti
o
n
T
e
c
h
n
iq
u
e
s
(ICACDOT)
,
2
0
1
6
,
p
p
.
1
7
0
-
1
7
4
,
d
o
i:
1
0
.
1
1
0
9
/ICAC
DO
T.
2
0
1
6
.
7
8
7
7
5
7
2
.
[1
6
]
A.
Z.
F
o
e
a
d
y
,
D.
C.
R.
No
v
it
a
sa
ri,
A.
H.
As
y
h
a
r,
a
n
d
M
.
F
irma
n
sja
h
,
“
Au
t
o
m
a
ted
Dia
g
n
o
sis
S
y
st
e
m
o
f
Dia
b
e
ti
c
Re
ti
n
o
p
a
t
h
y
Us
in
g
G
LCM
M
e
t
h
o
d
a
n
d
S
V
M
Clas
sifier,”
2
0
1
8
5
t
h
In
t.
C
o
n
f.
E
lec
tr.
En
g
.
Co
mp
u
t.
S
c
i.
In
f
o
rm
a
ti
c
s
,
p
p
.
1
5
4
-
1
6
0
,
2
0
1
8
,
d
o
i:
1
0
.
1
1
0
9
/e
e
c
si.2
0
1
8
.
8
7
5
2
7
2
6
.
[1
7
]
K.
K.
T
h
a
n
a
m
m
a
l
a
n
d
J.
S
.
J.
S
u
d
h
a
,
“
En
h
a
n
c
e
m
e
n
t
o
f
f
issu
re
u
sin
g
b
a
c
k
p
r
o
p
a
g
a
ti
o
n
n
e
u
r
a
l
n
e
two
rk
a
n
d
se
g
m
e
n
tatio
n
o
f
l
o
b
e
s i
n
CT
sc
a
n
ima
g
e
,
”
In
t.
J
.
Bi
o
me
d
.
En
g
.
T
e
c
h
n
o
l.
,
v
o
l.
2
0
,
n
o
.
1
,
p
p
.
1
-
1
1
,
2
0
1
6
.
[1
8
]
R.
Aru
lmu
r
u
g
a
n
a
n
d
H.
An
a
n
d
a
k
u
m
a
r,
“
Early
d
e
tec
ti
o
n
o
f
l
u
n
g
c
a
n
c
e
r
u
sin
g
wa
v
e
let
fe
a
tu
re
d
e
sc
rip
to
r
a
n
d
fe
e
d
fo
rwa
rd
b
a
c
k
p
ro
p
a
g
a
ti
o
n
n
e
u
ra
l
n
e
two
rk
s
c
las
sifier,”
in
C
o
mp
u
ta
ti
o
n
a
l
V
isio
n
a
n
d
Bi
o
I
n
sp
ir
e
d
Co
mp
u
ti
n
g
,
S
p
rin
g
e
r,
2
0
1
8
,
p
p
.
1
0
3
-
1
1
0
.
[1
9
]
F
.
S
h
a
u
k
a
t,
G
.
Ra
ja,
R.
As
h
ra
f,
S
.
Kh
a
li
d
,
M
.
Ah
m
a
d
,
a
n
d
A.
Al
i,
“
Artifi
c
ial
n
e
u
ra
l
n
e
two
r
k
b
a
se
d
c
las
sifica
ti
o
n
o
f
lu
n
g
n
o
d
u
les
i
n
CT
ima
g
e
s
u
sin
g
i
n
ten
sity
,
sh
a
p
e
a
n
d
tex
t
u
re
fe
a
tu
re
s,”
J
.
Amb
ien
t
In
tell.
H
u
ma
n
iz.
Co
mp
u
t.
,
v
o
l.
1
0
,
n
o
.
1
0
,
p
p
.
4
1
3
5
-
4
1
4
9
,
2
0
1
9
,
d
o
i:
1
0
.
1
0
0
7
/s
1
2
6
5
2
-
0
1
9
-
0
1
1
7
3
-
w.
[2
0
]
A.
Lo
m
u
sc
io
a
n
d
L.
M
a
g
a
n
ti
,
“
A
n
a
p
p
ro
a
c
h
to
re
a
c
h
a
b
i
li
ty
a
n
a
ly
s
is
fo
r
fe
e
d
-
f
o
rwa
rd
re
l
u
n
e
u
ra
l
n
e
two
rk
s,”
a
rXiv
Pre
p
r.
a
rXiv1
7
0
6
.
0
7
3
5
1
,
2
0
1
7
.
[2
1
]
L.
N.
M
.
Taw
fiq
a
n
d
O
.
M
.
S
a
li
h
,
“
De
sig
n
S
u
i
tab
le
F
e
e
d
F
o
rwa
rd
N
e
u
ra
l
Ne
two
rk
t
o
S
o
lv
e
Tro
e
sc
h
’s
P
ro
b
lem
,
”
S
c
i.
In
t.
(Lah
o
re
),
v
o
l.
3
1
,
p
p
.
4
1
-
4
8
,
2
0
1
9
.
[2
2
]
D.
C.
R.
No
v
it
a
sa
ri
e
t
a
l
.
,
“
Id
e
n
ti
f
y
Ed
u
c
a
ti
o
n
Qu
a
li
t
y
Ba
se
d
o
n
Isla
m
ic
S
e
n
io
r
Hig
h
S
c
h
o
o
l
Da
ta
in
K
o
m
p
e
ti
si
S
a
in
s
M
a
d
ra
sa
h
Us
in
g
F
u
z
z
y
C
-
M
e
a
n
s
Clu
ste
rin
g
,
”
in
S
ma
rt
T
re
n
d
s
in
C
o
m
p
u
t
in
g
a
n
d
Co
mm
u
n
ica
ti
o
n
s:
Pro
c
e
e
d
in
g
s
o
f
S
ma
rtCo
m
2
0
2
0
,
S
p
rin
g
e
r,
2
0
2
0
,
p
p
.
4
8
7
-
4
9
5
.
[2
3
]
Y.
K.
Du
b
e
y
a
n
d
M
.
M
.
M
u
s
h
rif
,
“
F
CM
c
lu
ste
rin
g
a
lg
o
rit
h
m
s
fo
r
se
g
m
e
n
tatio
n
o
f
b
ra
i
n
M
R
ima
g
e
s,”
Ad
v
.
Fu
zz
y
S
y
st.
,
v
o
l.
2
0
1
6
,
2
0
1
6
,
d
o
i:
1
0
.
1
1
5
5
/
2
0
1
6
/
3
4
0
6
4
0
6
.
[2
4
]
B.
T
h
a
m
a
ra
ich
e
lv
i
a
n
d
G
.
Ya
m
u
n
a
,
“
G
a
u
ss
ian
k
e
r
n
e
l
-
b
a
se
d
F
CM
se
g
m
e
n
tati
o
n
o
f
b
ra
in
M
RI
with
BP
N
N
c
las
sifica
ti
o
n
,
”
In
t.
J
.
Bi
o
me
d
.
E
n
g
.
T
e
c
h
n
o
l
.
,
v
o
l.
2
0
,
n
o
.
2
,
p
p
.
1
1
6
-
1
3
1
,
2
0
1
6
,
d
o
i:
1
0
.
1
5
0
4
/IJBET
.
2
0
1
6
.
0
7
4
1
9
8
.
[2
5
]
O.
I.
Ab
i
o
d
u
n
e
t
a
l.
,
“
Co
m
p
re
h
e
n
siv
e
re
v
ie
w
o
f
a
rti
ficia
l
n
e
u
ra
l
n
e
two
rk
a
p
p
li
c
a
ti
o
n
s
t
o
p
a
tt
e
r
n
re
c
o
g
n
it
i
o
n
,
”
IEE
E
Acc
e
ss
,
v
o
l.
7
,
p
p
.
1
5
8
8
2
0
-
1
5
8
8
4
6
,
2
0
1
9
,
d
o
i:
1
0
.
1
1
0
9
/ACCE
S
S
.
2
0
1
9
.
2
9
4
5
5
4
5
.
[2
6
]
Y.
Xu
,
H.
Ya
o
,
a
n
d
K.
Li
n
,
“
An
o
v
e
rv
iew
o
f
n
e
u
ra
l
n
e
two
rk
s
f
o
r
d
ru
g
d
isc
o
v
e
ry
a
n
d
th
e
i
n
p
u
ts u
se
d
,
”
Exp
e
rt
O
p
in
.
Dr
u
g
Disc
o
v
.
,
v
o
l.
1
3
,
n
o
.
1
2
,
p
p
.
1
0
9
1
-
1
1
0
2
,
2
0
1
8
,
d
o
i
:
1
0
.
1
0
8
0
/
1
7
4
6
0
4
4
1
.
2
0
1
8
.
1
5
4
7
2
7
8
.
[2
7
]
M
.
Isla
m
,
G
.
Ch
e
n
,
a
n
d
S
.
Jin
,
“
An
Ov
e
rv
iew
o
f
Ne
u
ra
l
Ne
two
rk
,
”
Am.
J
.
Ne
u
ra
l
Ne
two
rk
s
Ap
p
l.
,
v
o
l.
5
,
n
o
.
1
,
p
p
.
7
-
1
1
,
2
0
1
9
,
d
o
i:
1
0
.
1
1
6
4
8
/
j.
a
j
n
n
a
.
2
0
1
9
0
5
0
1
.
1
2
.
[2
8
]
L.
D.
M
e
d
u
s,
T
.
Ia
k
y
m
c
h
u
k
,
J.
V.
F
ra
n
c
e
s
-
Vill
o
ra
,
M
.
Ba
taller
-
M
o
m
p
e
á
n
,
a
n
d
A.
Ro
sa
d
o
-
M
u
ñ
o
z
,
“
A
No
v
e
l
S
y
st
o
li
c
P
a
ra
ll
e
l
Ha
rd
wa
re
Arc
h
it
e
c
t
u
re
fo
r
th
e
F
P
G
A
Ac
c
e
lera
ti
o
n
o
f
F
e
e
d
fo
rwa
rd
Ne
u
ra
l
Ne
two
rk
s,”
IEE
E
Acc
e
ss
,
v
o
l.
7
,
p
p
.
7
6
0
8
4
-
7
6
1
0
3
,
2
0
1
9
,
d
o
i:
1
0
.
1
1
0
9
/ACCES
S
.
2
0
1
9
.
2
9
2
0
8
8
5
.
[2
9
]
S
.
G
u
,
M
.
P
e
d
n
e
k
a
r,
a
n
d
R.
S
late
r,
“
Im
p
ro
v
e
Im
a
g
e
Clas
sifica
ti
o
n
Us
in
g
Da
ta
Au
g
m
e
n
tatio
n
a
n
d
Ne
u
ra
l
Ne
two
rk
s,
”
S
M
U Da
ta
S
c
i.
Rev
.
,
v
o
l.
2
,
n
o
.
2
,
p
.
1
,
2
0
1
9
.
Evaluation Warning : The document was created with Spire.PDF for Python.