T
E
L
K
O
MN
I
K
A
T
el
eco
m
m
u
n
i
ca
t
i
o
n
,
C
o
m
p
u
t
i
n
g
,
E
l
ect
ro
n
i
cs
a
n
d
C
o
n
t
ro
l
Vo
l
.
18
, N
o
.
5
,
O
ct
o
b
er
2020
,
pp
.
2432~
2438
I
S
S
N
:
1693
-
6930,
a
c
c
r
e
di
t
e
d F
i
r
s
t
G
r
a
de
by K
e
m
e
nr
i
s
t
e
kdi
kt
i
,
D
e
c
r
e
e
N
o:
21/
E
/
K
P
T
/
2018
D
O
I
:
10.
12928/
T
E
L
K
O
M
N
I
K
A
.
v18i
5.
15523
2432
Jou
r
n
al
h
om
e
page
:
ht
t
p:
/
/
j
our
nal
.
uad
.
ac
.
i
d/
i
nde
x
.
php/
T
E
L
K
O
M
N
I
K
A
C
a
nc
ero
us
l
un
g
no
d
ul
e
de
t
e
ct
i
o
n
i
n
co
m
p
ut
ed
t
o
m
o
g
ra
phy
i
m
age
s
Ay
m
a
n
Ab
u
Ba
k
e
r
1
,
Y
a
zeed
G
h
a
d
i
2
1
De
pa
r
tm
e
n
t
of
E
le
c
tr
ic
a
l
E
ng
in
e
e
r
i
ng,
A
pp
lie
d
S
c
ie
nc
e
P
r
iva
te
Un
ive
r
si
ty,
J
or
da
n
2
De
pa
r
tm
e
n
t of
E
le
c
tr
ic
a
l E
ng
in
e
e
r
i
ng,
A
l Ai
n U
ni
ve
r
s
it
y of
S
c
ie
nc
e
a
nd
Te
c
h
no
lo
gy,
U
ni
te
d Ar
a
b Em
ir
a
te
s
A
rt
i
cl
e I
n
f
o
AB
S
T
RACT
A
r
tic
le
h
is
to
r
y
:
R
ecei
v
ed
J
a
n 20,
2020
R
ev
i
s
ed
M
a
r
28,
2020
A
ccep
t
ed
J
un 12,
2020
Dia
gn
os
is the
c
om
p
ute
d
t
om
o
gr
a
p
hy im
a
ge
s
(
C
T
-
i
m
a
ge
s)
i
s o
ne
of
the
im
a
ge
s t
ha
t m
a
y ta
ke
a
lot
of
tim
e
i
n dia
gn
os
is b
y th
e
r
a
dio
lo
gi
st a
n
d m
a
y
m
iss
som
e
of
c
a
nc
e
r
o
us n
od
ule
s i
n t
he
se
im
a
ge
s.
T
he
r
e
f
or
e
,
in
th
is
pa
pe
r
a
ne
w
no
ve
l
e
n
ha
nc
e
m
e
nt a
n
d
de
te
c
ti
on c
a
nc
e
r
ou
s
nod
ul
e
a
lg
or
i
t
h
m
is
pr
o
po
se
d to d
ia
g
no
se
a
C
T
-
im
a
ge
s.
The
n
ove
l a
lg
or
i
th
m
is div
ide
d in
to t
hr
e
e
m
a
in
sta
ge
s.
I
n
f
ir
st
sta
ge
,
su
sp
ic
i
ou
s r
e
gi
on
s a
r
e
e
n
ha
nc
e
d
us
in
g m
o
dif
ie
d
L
oG
a
lgor
it
hm
.
The
n i
n s
ta
ge
tw
o,
a
p
ote
nt
ia
l c
a
nc
e
r
o
us
no
du
le
w
as
de
te
c
te
d
ba
se
d on v
is
ua
l a
p
pe
a
r
a
nc
e
i
n l
un
g.
F
ina
l
ly,
f
i
ve
te
x
t
ur
e
f
e
a
tur
e
s a
na
l
ys
is
a
lg
or
i
thm
is im
ple
m
e
nte
d to r
e
d
uc
e
num
be
r
of
de
te
c
t
e
d F
P
r
e
gion
s.
Th
is
a
lg
or
i
thm
is e
va
l
ua
te
d u
si
ng
6
0 c
a
se
s (
n
or
m
a
l
a
n
d c
a
n
c
e
r
ou
s c
a
se
s)
,
a
n
d i
t
sho
ws a
h
ig
h se
n
si
ti
vi
ty i
n de
te
c
ti
ng t
he
c
a
nc
e
r
o
us l
u
ng
n
od
ule
s wi
th TP
r
a
ti
on
9
7%
a
nd
w
it
h
F
P
r
a
ti
o
2
5
c
lu
ste
r
/im
a
ge
.
Ke
y
wo
r
d
s
:
C
an
cer
d
et
ect
i
o
n
C
om
put
e
d t
om
ogr
a
phy
L
u
n
g
can
cer
T
ex
t
u
r
e f
eat
u
r
es
L
ap
l
aci
an
f
ilte
r
T
his
is
a
n
o
pe
n
ac
c
e
s
s
ar
tic
le
u
nde
r
the
CC
B
Y
-
SA
lic
e
n
se
.
C
or
r
e
s
pon
di
n
g A
u
t
h
or
:
A
ym
a
n
Ab
u
B
ak
er
,
D
ep
ar
t
m
en
t
o
f
E
l
ect
r
i
cal
an
d
C
o
m
p
u
t
er
E
n
g
i
n
eer
i
n
g
,
A
p
p
l
i
ed
S
ci
en
ce P
r
i
v
at
e U
n
i
v
er
s
i
t
y
,
A
m
m
a
n,
J
or
da
n.
E
ma
il:
a
_a
buba
ke
r
@
a
s
u.
e
du.
j
o
1.
I
NT
RO
DUC
T
I
O
N
L
ung c
a
nc
e
r
i
s
one
o
f
t
he
m
os
t
r
e
l
e
va
nt
publ
i
c
h
e
a
l
t
h i
s
s
ue
s
i
n U
ni
t
e
d s
t
a
t
e
s
,
E
u
r
ope
a
nd
M
i
ddl
e
E
a
st
[
1
, 2
]
.
E
ar
l
y
D
et
ect
i
o
n
an
d
t
r
eat
m
en
t
o
f
t
h
i
s
t
y
p
es
o
f
can
cer
i
s
r
eq
u
i
r
e t
o
ef
f
ect
i
v
el
y
o
v
e
r
co
m
e t
h
i
s
bur
de
n.
C
h
e
st
X
-
r
ay
co
n
s
i
d
er
ed
a ch
eap
es
t
m
e
t
h
o
d
as
an
i
n
i
t
i
al
d
et
ect
i
o
n
o
f
l
u
n
g
can
cer
.
C
om
put
e
d
t
om
ogr
a
phy (
C
T
)
a
s
a
s
e
c
ond di
a
gnos
i
s
s
t
a
ge
i
s
t
he
be
s
t
i
m
a
gi
n
g
mo
d
a
lity
f
o
r
th
e
d
e
te
c
tio
n
o
f
s
ma
ll
pul
m
ona
r
y nodul
e
s
,
pa
r
t
i
c
ul
a
r
l
y s
i
nc
e
t
he
i
nt
r
odu
c
t
i
on of
t
he
he
l
i
c
a
l
t
e
c
hnol
ogy
[3
,
4
]
.
T
h
e
C
T
i
m
ag
es
ar
e
a
hi
gh r
e
s
ol
ut
i
on
i
m
a
ge
s
w
i
t
h h
i
gh a
m
ount
of
da
t
a
s
t
or
a
ge
.
T
he
r
e
f
or
e
,
r
e
s
e
a
r
c
he
r
s
t
r
i
e
s
t
o he
l
p
t
he
r
a
di
ol
ogi
s
t
t
o e
a
s
i
l
y pr
oc
e
s
s
t
he
s
e
huge
i
m
a
ge
a
nd
a
ut
om
a
t
i
c
a
l
l
y de
t
e
c
t
e
d t
he
pot
e
nt
i
a
l
nodul
e
l
ung
c
a
nc
e
r
s
us
i
ng
c
om
put
e
r
a
i
de
d di
a
gnos
i
s
s
ys
t
e
m
(
C
A
D
)
[
5
,
6
]
.
D
e
t
e
c
t
i
on nodul
e
l
ung
c
a
nc
e
r
i
s
one
of
t
he
m
os
t
di
f
f
i
c
ul
t
cas
es
f
o
r
t
h
e
r
ad
i
o
l
o
g
i
s
t
s
p
eci
al
l
y
i
n
C
T
i
m
ag
es
s
i
n
ce
t
h
e v
i
s
u
al
ap
p
ear
an
ce o
f
t
u
m
o
r
s
i
s
n
o
t
cl
ear
o
r
s
ur
r
oundi
ng w
i
t
h
pa
r
e
nc
hym
a
l
t
i
s
s
ue
i
n
C
T
-
I
m
ag
es
[7
, 8
]
.
T
h
er
ef
o
r
e,
t
h
e
v
i
s
u
al
ap
p
ear
an
ce f
o
r
can
cer
o
u
s
nodul
e
s
ha
ve
s
i
m
i
l
a
r
vi
s
ua
l
c
ha
r
a
c
t
e
r
i
s
t
i
c
s
of
nor
m
a
l
t
i
s
s
ue
s
[
9]
.
T
he
r
e
f
o
r
e
,
t
hi
s
pa
pe
r
p
r
opos
e
d
a
nove
l
d
et
ect
i
o
n
an
d
cl
as
s
i
f
i
cat
i
o
n
m
et
h
o
d
f
o
r
can
cer
o
u
s
cel
l
s
.
T
h
r
ee m
ai
n
s
t
ag
es
ar
e
u
s
ed
t
o
accu
r
at
el
y
d
et
ect
t
he
c
a
nc
e
r
ous
nodul
e
s
i
n t
he
C
T
l
ung
i
m
a
ge
s
[
1
0]
.
T
he
pa
pe
r
p
r
e
s
e
nt
a
br
i
e
f
de
s
c
r
i
pt
i
on
a
bout
t
he
l
ung C
T
ima
g
e
in
s
e
c
tio
n
2
.
T
h
e
n
lite
r
a
tu
r
e
r
e
vi
e
w
s
e
c
t
i
on w
hi
c
h pr
e
s
e
nt
e
d i
n s
e
c
t
i
on
3
.
P
r
opos
e
d de
t
e
c
t
i
on a
nd
c
la
s
s
if
ic
a
tio
n
a
lg
o
r
ith
m is
in
tr
o
d
u
c
e
d
in
S
e
c
tio
n
4
.
F
i
na
l
l
y
,
di
s
c
us
s
i
on a
nd c
onc
l
us
i
on a
r
e
pr
e
s
e
nt
e
d i
n
s
ect
i
o
n
5
a
nd s
e
c
t
i
on
6
r
es
p
ect
i
v
el
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
C
anc
e
r
ou
s
l
ung nodul
e
de
t
e
c
t
i
on i
n
c
om
put
e
d t
om
ogr
aphy
i
m
age
s
(
A
y
m
an A
bu
B
ak
e
r
)
2433
2.
DAT
AB
AS
E
T
h
e d
at
ab
as
e t
h
at
ar
e u
s
ed
i
n t
h
i
s
pa
pe
r
i
n
dow
nl
oa
de
d f
r
om
C
a
nc
e
r
I
m
a
gi
ng A
r
c
hi
ve
(
T
C
I
A
)
,
w
hi
c
h i
s
or
ga
ni
z
e
d i
nt
o
pur
pos
e
-
bui
l
t
c
ol
l
e
c
t
i
ons
o
f
s
ubj
e
c
t
s
[
11]
.
T
hi
s
da
t
a
ba
s
e
ha
s
di
f
f
e
r
e
nt
ki
nd
of
hi
gh
r
e
s
ol
ut
i
o
n i
m
a
ge
s
o
f
can
cer
t
y
p
e an
d
/
o
r
an
at
o
m
i
cal
s
i
t
e
(
l
u
n
g
,
b
r
ai
n
,
et
c.
)
i
n
co
m
m
on
.
M
or
e
ove
r
,
t
he
da
t
a
ba
s
e
i
nc
l
ude
a
hu
ge
da
t
a
s
e
t
o
f
M
RI
,
CT
,
a
n
d
X
-
R
ay
i
m
ag
es
t
h
at
ar
e s
t
o
r
ed
as
D
I
C
O
M
f
i
l
e f
o
r
m
at
[
12]
.
3.
LI
T
ER
A
TU
R
E R
EV
I
EW
M
a
ny a
ut
hor
s
pr
opos
e
d d
i
f
f
e
r
e
nt
C
A
D
t
e
c
hni
que
s
t
o he
l
p
t
he
r
a
di
ol
ogi
s
t
i
n d
i
a
gnos
e
s
l
ung c
a
nc
e
r
i
n
CT
i
m
a
ge
s
.
O
ne
o
f
t
he
s
e
t
e
c
hni
que
s
i
s
us
i
ng L
a
pl
a
c
i
a
n of
G
a
us
s
i
a
n f
i
l
t
e
r
a
s
i
n
[
13]
.
T
he
i
r
a
l
g
or
i
t
hm
i
s
d
i
vi
de
d
t
o t
w
o m
a
i
n s
t
a
ge
s
,
t
r
a
ns
m
i
s
s
i
on of
hi
gh i
nt
e
ns
i
t
y
an
d
L
O
G
F
i
l
t
er
.
T
h
es
e s
t
ag
es
ar
e ap
p
l
i
ed
cal
cu
l
at
e
t
h
e
co
n
t
r
as
t
di
f
f
e
r
e
nc
e
s
i
ns
i
de
a
nd
out
s
i
de
r
e
gi
on
of
i
nt
e
r
e
s
t
.
S
na
ke
a
l
gor
i
t
hm
i
s
a
not
he
r
e
nha
nc
e
m
e
nt
a
nd
s
e
gm
e
nt
a
t
i
on
t
ool
us
e
d
t
o
f
i
nd
t
he
i
nt
e
r
na
l
e
ne
r
gy
i
n C
T
i
m
a
ge
l
i
ke
V
i
ve
ka
na
nda
n D
.
et
al
[
1
4
]
.
M
es
s
ay
et
al
.
[
1
5
]
i
m
pl
e
m
e
nt
l
oc
a
l
c
ont
r
a
s
t
e
nha
nc
e
m
e
nt
f
i
l
t
e
r
t
ha
t
f
ol
l
ow
t
he
nodu
l
e
e
nha
nc
e
m
e
nt
m
e
t
hod
t
o
d
et
ect
t
h
e ch
es
t
r
ad
i
o
g
r
ap
h
s
.
D
o
t
s
,
l
i
n
es
,
an
d
p
l
an
es
en
h
an
cem
en
t
t
ech
n
i
q
u
es
ar
e u
s
ed
b
y
L
i
et
al
.
[
1
6
]
.
S
uc
h
t
h
e
se
en
h
an
cem
en
t
t
ech
n
i
q
u
es
c
a
n be
i
m
pl
e
m
e
nt
e
d on
s
p
eci
f
i
c
s
ha
pe
s
a
nd s
uppr
e
s
s
ot
he
r
obj
e
c
t
s
.
S
o
,
as
an
in
itia
l s
te
p
th
e
C
T
-
I
m
ag
e i
s
b
l
u
r
r
ed
t
h
en
a G
au
s
s
i
an
k
er
n
el
f
ilte
r
(
G
K
F
)
o
f
a
n
o
d
u
le
s
iz
e
is
imp
le
m
e
n
te
d
to
de
t
e
c
t
t
he
nodul
e
.
As
f
in
a
l s
ta
g
e
,
mu
lti
-
s
cal
es
G
K
F
a
r
e
us
e
d a
s
t
o f
i
nd a
m
a
t
c
h w
i
t
h t
he
nodul
e
s
i
z
e
.
Y
am
am
o
t
o
et
al
.
[
1
7
]
u
s
ed
a s
t
at
i
s
t
i
cal
en
h
an
cem
en
t
f
i
l
t
er
t
o
en
h
an
ce C
T
i
m
ag
es
.
T
h
ey
i
m
p
l
em
en
t
ed
th
e
Q
u
o
it
F
ilte
r
th
a
t
h
a
s
la
r
g
e
r
in
g
a
n
d
d
is
k
f
ilte
r
s
.
D
if
f
e
r
e
n
tia
l in
te
n
s
ity
v
a
lu
e
p
r
o
c
e
d
u
r
e
is
imp
le
me
n
te
d
in
t
hi
s
a
l
gor
i
t
hm
be
t
w
e
e
n t
he
i
nt
e
r
na
l
a
nd
e
xt
e
r
na
l
d
i
s
ks
i
n or
de
r
t
o
f
i
nd
a
pot
e
nt
i
a
l
r
e
gi
on
of
i
nt
e
r
e
s
t
.
A
not
he
r
s
e
gm
e
nt
a
t
i
on a
l
gor
i
t
hm
i
s
pr
e
s
e
nt
e
d by
M
aci
ej
D
aj
n
o
w
i
ec et
al
.
[
1
8
]
.
I
n
itia
lly
,
a
n
o
p
timu
m th
r
e
s
h
o
ld
v
a
lu
e
w
as
cal
cu
l
at
ed
b
a
s
ed
o
n
d
i
f
f
er
en
t
i
m
ag
e d
at
a s
et
s
.
T
hr
e
s
hol
d a
l
gor
i
t
hm
t
he
n i
s
i
m
pl
e
m
e
nt
e
d t
o s
e
gm
e
nt
l
unge
r
e
g
io
n
f
r
o
m C
T
ima
g
e
.
A
f
te
r
th
a
t,
c
onne
c
t
i
ng c
om
pone
nt
l
a
be
l
i
ng
(
C
C
L
)
is
a
p
p
lie
d
to
d
e
te
c
t r
e
ma
in
in
g
r
e
gi
ons
i
n t
he
i
m
a
ge
.
T
h
e
re
fo
re
,
l
ung
nodul
e
c
a
n s
uc
c
e
s
s
f
ul
l
y s
e
gm
e
nt
e
d but
ha
ve
l
obe
s
w
i
t
h nodul
e
s
r
e
gi
on.
O
n t
he
ot
he
r
ha
nd,
C
A
D
s
ys
t
e
m
i
s
a
pow
e
r
f
ul
t
ool
i
n de
t
e
c
t
i
ng
c
a
nc
e
r
ous
nodul
e
s
i
n t
he
C
T
i
m
a
ge
s
.
A
not
he
r
C
A
D
s
ys
t
e
m
t
o de
t
e
c
t
c
a
nc
e
r
ous
l
ung
r
e
g
i
on i
s
pr
opos
e
d by
L
ee et
al
.
[
1
9
]
.
A
u
th
o
r
s
imp
le
me
n
t
G
A
to
s
e
t a
ta
r
g
e
t
p
o
s
itio
n
a
n
d
a
ls
o
to
ma
tc
h
th
e
b
e
s
t
te
mp
la
te
ima
g
e
f
r
o
m
pr
e
vi
ous
da
t
a
s
e
t
s
.
F
our
ge
ne
r
a
t
i
on
s
e
ve
r
y t
i
m
e
a
r
e
e
s
t
a
bl
i
s
he
d ba
s
e
d on
gr
e
y
l
e
ve
l
of
G
a
us
s
i
a
n di
s
t
r
i
but
i
on.
A
lg
o
r
ith
m s
e
n
s
itiv
ity
w
a
s
85%
i
n
d
et
ect
i
n
g
c
a
nc
e
r
ous
nodul
e
s
i
n t
he
C
T
-
i
m
ag
es
.
O
t
h
er
au
t
h
o
r
l
i
k
e O
p
f
er
et
al
.
[
20]
us
e
d di
s
t
a
nc
e
tr
a
n
s
f
o
r
ma
tio
n
te
c
hni
que
t
o
i
m
p
r
ove
t
he
pe
r
f
or
m
a
nc
e
of
C
A
D
s
ys
t
e
m
.
T
he
di
s
t
a
nc
e
t
r
a
ns
f
or
m
a
t
i
ons
of
va
r
i
ous
t
hr
e
s
hol
ds
a
nd s
ubs
e
qu
e
nt
c
r
e
s
t
l
i
ne
e
xt
r
a
c
t
i
on a
r
e
us
e
d e
nha
nc
e
t
he
C
A
D
s
e
n
s
i
t
i
vi
t
y.
C
A
D
s
ys
t
e
m
pe
r
f
or
m
a
nc
e
i
s
s
l
i
ght
l
y i
m
pr
ove
d w
he
n us
i
ng t
hi
s
t
e
c
hni
que
s
i
nc
e
i
t
p
r
oduc
e
l
a
r
ge
num
be
r
of
F
P
r
e
gi
ons
.
A
not
he
r
C
A
D
s
ys
t
e
m
i
s
us
e
d by
M
or
e
ove
r
,
G
ol
o
s
i
o e
t
a
l
.
[
21]
t
o de
t
e
c
t
c
a
nc
e
r
ous
nodul
e
s
i
n
C
T
i
m
a
ge
s
.
T
he
y us
e
d m
ul
t
i
-
a
da
pt
i
ve
t
hr
e
s
hol
d s
ur
f
a
c
e
t
r
i
a
ngul
a
t
i
on a
ppr
oa
c
h i
n t
he
de
t
e
c
t
i
on a
l
gor
i
t
hm
.
T
he
pr
opos
e
d
a
lg
o
r
ith
m s
h
ow
good s
e
ns
i
t
i
vi
t
y i
n
de
t
e
c
t
i
on
c
a
nc
e
r
nodul
e
s
but
m
a
ny de
t
e
c
t
e
d
F
P
a
r
e
pr
e
s
e
nt
s
.
K.
D
e
va
ki
e
t
a
l
.
[
22]
de
ve
l
op
a
t
e
c
hni
que
t
o
a
c
c
ur
a
t
e
l
y s
e
gm
e
nt
l
ung
r
e
gi
on
f
r
om
C
T
-
i
m
a
ge
s
.
T
he
pr
opos
e
d
a
l
gor
i
t
hm
c
a
n e
f
f
i
c
i
e
nt
l
y s
e
gm
e
nt
t
he
l
ung
r
e
gi
ons
f
r
om
C
T
-
i
m
ag
es
.
4.
R
ES
EA
R
C
H
M
ETH
O
D
E
nha
nc
e
m
e
nt
a
nd de
t
e
c
t
i
on
of
t
he
c
a
nc
e
r
ous
nodul
e
s
t
e
c
hni
que
i
s
p
r
opos
e
d i
n
t
hi
s
pa
pe
r
.
T
he
pr
opos
e
d m
e
t
hod
i
s
m
a
i
nl
y di
vi
de
d
t
o t
h
r
e
e
s
t
a
ge
s
.
I
ni
t
i
a
l
l
y,
t
he
c
a
nc
e
r
ous
l
ung nodu
l
e
s
w
i
l
l
be
e
nha
nc
e
d
us
i
ng L
a
pl
a
c
i
a
n f
i
l
t
e
r
.
T
he
n i
n
t
he
s
e
c
ond s
t
a
ge
,
t
he
a
ve
r
a
ge
de
t
e
c
t
i
on
a
l
gor
i
t
hm
w
i
l
l
be
us
e
d t
o de
t
e
c
t
a p
o
t
en
t
i
al
can
cer
o
u
s
l
ung
nodul
e
(
P
C
L
N
)
r
e
gi
ons
i
n
C
T
-
i
m
ag
es
.
F
i
n
al
,
f
i
v
e
t
ex
t
u
r
e
f
eat
u
r
es
ar
e
u
s
ed
t
o
h
el
p
i
n r
e
duc
i
ng num
be
r
of
de
t
e
c
t
e
d F
P
r
e
gi
ons
.
F
i
gur
e
1,
s
how
t
he
p
r
opos
e
d a
l
gor
i
t
hm
s
t
a
ge
s
.
4
.1
.
C
a
nc
e
r
o
us
l
ung
no
dul
e
e
nha
nc
e
m
e
nt
T
he
f
unda
m
e
nt
a
l
ope
r
a
t
i
on ne
e
de
d t
o a
s
s
i
s
t
c
a
nc
e
r
ous
nod
ul
e
i
n C
T
-
i
m
ag
e i
s
co
n
t
r
as
t
en
h
an
c
em
en
t
.
I
n m
a
ny i
m
a
ge
pr
oc
e
s
s
i
ng a
ppl
i
c
a
t
i
ons
,
t
he
L
a
pl
a
c
i
a
n f
i
l
t
e
r
i
s
one
of
t
he
s
i
m
pl
e
s
t
a
nd e
f
f
e
c
t
i
ve
t
e
c
hni
que
s
f
or
i
nt
e
ns
i
t
y e
nha
nc
e
m
e
nt
t
ha
t
pr
e
s
e
nt
e
d i
n a
s
(
1
)
.
L
ap
l
aci
an
f
i
l
t
er
i
m
p
r
o
v
es
co
n
t
r
as
t
o
f
t
h
e can
cer
o
u
s
n
o
d
u
le
s
in
CT
-
i
m
a
ge
by a
ppl
yi
ng t
he
L
a
pl
a
c
i
a
n m
a
s
k of
s
i
z
e
9
×
9
.
(
,
)
=
(
,
)
+
[
2
(
,
)
]
(
1)
w
h
er
e S
(
x,
y
)
i
s
t
he
i
nt
e
ns
i
t
y va
l
ue
f
or
t
he
p
r
oc
e
s
s
e
d i
m
a
ge
,
f
(
x
,
y
)
t
he
i
nt
e
ns
i
t
y va
l
ue
f
o
r
t
he
i
nput
i
m
a
ge
a
nd
c
i
s
c
ons
i
de
r
a
s
one
i
n t
h
i
s
pa
pe
r
.
C
a
nc
e
r
ous
nodul
e
s
a
ppe
a
r
on
di
gi
t
i
z
e
d C
T
a
s
s
m
a
l
l
r
e
gi
ons
,
w
i
t
h
i
nt
e
ns
i
t
y va
l
ue
s
hi
ghe
r
t
ha
n
t
he
i
r
s
ur
r
oundi
ng ba
c
kgr
ound.
T
he
s
i
z
e
of
c
a
nc
e
r
ous
n
odul
e
s
i
s
us
ua
l
l
y l
e
s
s
t
ha
n 4 m
m
[
23
]
.
S
o
it
is
n
o
t e
a
s
y
to
e
nha
nc
e
t
he
c
a
nc
e
r
ous
nodul
e
s
r
e
gi
ons
s
i
nc
e
s
u
r
r
oundi
ng l
ung t
i
s
s
ue
m
a
ke
s
t
he
a
bnor
m
a
l
i
t
y a
r
e
a
s
a
l
m
os
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1693
-
6930
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
,
Vo
l
.
18
, N
o
.
5
,
O
c
t
obe
r
2020:
2432
-
2438
2434
i
nvi
s
i
bl
e
.
T
he
r
e
f
or
e
,
t
he
m
odi
f
i
e
d a
ve
r
a
ge
f
i
l
t
e
r
i
s
i
m
p
l
e
m
e
nt
e
d t
o
s
m
oot
h t
he
e
dge
s
of
t
he
p
r
oc
e
s
s
e
d
L
ap
l
aci
an
en
h
an
cem
en
t
i
m
ag
e.
T
h
i
s
i
n
cas
e w
i
l
l
s
l
i
g
h
t
l
y
en
h
an
ce t
h
e
can
cer
o
u
s
nodul
e
s
r
e
gi
ons
t
o
b
e eas
i
l
y
d
et
ect
ed
i
n
t
h
e n
ex
t
s
t
ag
e.
A
f
t
e
r
e
xt
e
ns
i
ve
a
na
l
ys
i
s
of
60
can
cer
o
u
s
C
T
i
m
ag
es
,
w
e co
n
cl
u
d
ed
t
h
at
al
l
can
cer
o
u
s
nodul
e
s
ha
ve
gr
e
y s
c
a
l
e
va
l
ue
s
i
n t
he
r
a
nge
f
r
om
80 t
o 230
[
24,
25
]
.
I
n
acco
r
d
an
ce w
i
t
h
t
h
es
e o
b
s
er
v
at
i
o
n
s
,
each
C
T
-
i
m
ag
e
i
s
pr
oc
e
s
s
e
d us
i
ng t
he
m
odi
f
i
e
d a
ve
r
a
ge
f
i
l
t
e
r
t
ha
t
pr
e
s
e
nt
e
d on
(
2
)
,
=
1
2
3
0
=
8
0
(
2
)
w
he
r
e
S
k i
s
t
he
i
nt
e
ns
i
t
y va
l
ue
f
or
t
he
pr
oc
e
s
s
e
d i
m
a
ge
,
r
j
t
he
i
nt
e
ns
i
t
y va
l
ue
f
or
t
he
i
npu
t
i
m
a
ge
a
nd m
a
nd n
ar
e t
h
e m
as
k
s
i
ze.
A
f
t
e
r
pr
oc
e
s
s
i
ng t
he
L
a
pl
a
c
i
a
n
a
nd m
odi
f
i
e
d a
ve
r
a
ge
f
i
l
t
e
r
,
t
he
c
a
nc
e
r
ous
nodul
e
r
e
gi
ons
be
c
om
e
s
l
i
ght
l
y br
i
ght
e
r
c
o
r
r
e
s
pondi
ng t
o
t
he
ne
i
ghbor
r
e
gi
ons
a
s
s
how
n i
n F
i
gur
e
2
.
T
hi
s
w
i
l
l
a
s
s
i
s
t
i
n
de
t
e
c
t
i
on t
he
r
e
gi
on o
f
i
nt
e
r
e
s
t
t
ha
t
w
i
l
l
be
di
s
c
us
s
e
d i
n t
he
ne
xt
s
e
c
t
i
on.
F
ig
ur
e
1
.
M
C
d
et
ect
i
o
n
an
d
cl
as
s
i
f
i
cat
i
o
n
F
i
gur
e
2.
C
a
nc
e
r
ous
nodul
e
e
nha
nc
e
m
e
nt
r
e
s
ul
t
4
.
2
.
P
o
t
en
t
i
a
l
C
a
n
cero
u
s
l
ung
no
dul
e
s
C
a
nc
e
r
ous
nodul
e
s
a
ppe
a
r
on di
gi
t
i
z
e
d
C
T
-
ima
g
e
s
a
s
s
ma
ll r
e
g
io
n
s
,
w
i
th
in
te
n
s
ity
v
a
lu
e
s
h
ig
h
e
r
t
ha
n t
he
i
r
s
ur
r
oundi
ng
ba
c
kgr
ound.
S
o
i
n or
de
r
t
o
de
t
e
c
t
t
he
s
e
r
e
gi
on,
t
w
o c
onc
e
nt
r
i
c
c
i
r
c
ul
a
r
m
a
s
ks
a
r
e
us
e
d
a
s
s
how
n i
n F
i
gur
e
3.
W
he
n c
e
nt
e
r
e
d
on
t
he
c
a
nc
e
r
ou
s
nodul
e
,
t
he
i
nne
r
m
a
s
ke
d r
e
gi
on
i
nc
l
ude
d
t
he
c
a
nc
e
r
ous
nodul
e
w
hi
l
e
t
he
ou
t
e
r
m
a
s
ke
d r
e
g
i
on i
nc
l
ude
d
t
he
s
ur
r
oundi
ng
r
e
gi
on
.
T
he
i
nne
r
a
nd out
e
r
co
n
cen
t
r
i
c ci
r
cu
l
ar
m
as
k
’
s
s
i
ze
ar
e
d
es
i
g
n
b
as
ed
o
n
C
T
-
I
m
a
ge
r
e
s
ol
ut
i
on.
T
h
es
e
m
as
k
s
w
er
e
t
es
t
ed
o
n
6
0
CT
-
i
m
ag
e an
d
t
h
ey
w
er
e e
f
f
e
c
tiv
e
in
d
e
te
c
tin
g
a
ll
th
e
s
u
s
p
ic
io
u
s
r
e
gi
ons
i
n
C
T
-
I
m
ag
es
.
D
et
ect
i
o
n
t
h
e
pot
e
nt
i
a
l
c
a
nc
e
r
ous
l
ung nodul
e
s
(
P
C
LN
)
c
l
us
t
e
r
i
s
de
s
i
gne
d ba
s
e
d on t
he
f
a
c
t
t
ha
t
t
he
c
a
nc
e
r
ous
nodul
e
s
a
r
e
br
i
ght
e
r
t
ha
n
t
he
ne
i
ghbor
pi
xe
l
s
.
T
h
er
ef
o
r
e,
i
n
or
de
r
t
o
s
e
l
e
c
t
P
C
L
N
t
w
o
c
ondi
t
i
ons
s
houl
d
be
s
a
t
i
s
f
i
e
d,
a
ve
r
a
ge
va
l
ue
f
or
t
he
i
nne
r
m
a
s
k s
houl
d be
g
r
e
a
t
e
r
t
ha
n out
e
r
m
a
s
k a
nd
t
he
i
n
t
e
ns
i
t
y pi
xe
l
va
l
ue
of
t
h
e
c
e
nt
e
r
of
t
he
i
nne
r
m
a
s
k s
houl
d be
t
he
hi
ghe
s
t
i
nt
e
ns
i
t
y
i
n t
he
m
a
s
k.
A
f
t
e
r
pr
oc
e
s
s
i
ng 60 C
T
-
im
a
ge
s
us
i
ng P
C
L
N
a
lg
o
r
ith
m,
it w
a
s
n
o
tic
e
d
th
a
t a
ll
c
a
nc
e
r
ous
nodul
e
s
i
n t
he
C
T
-
ima
g
e
s
ar
e d
et
ect
ed
but
m
a
ny de
t
e
c
t
e
d
f
a
l
se
pos
i
t
i
ve
(
F
P
)
r
e
gi
ons
a
s
s
how
n i
n
F
i
gur
e
4.
T
hi
s
i
n c
a
s
e
w
i
l
l
r
e
duc
e
t
he
s
e
n
s
i
t
i
vi
t
y of
t
he
pr
opos
e
d C
A
D
s
y
s
t
em
.
T
h
er
ef
o
r
e,
t
ex
t
u
r
e f
eat
u
r
e an
al
y
s
i
s
w
i
l
l
b
e ap
p
l
i
ed
t
o
r
ed
u
ce n
u
m
b
er
o
f
d
et
ect
ed
F
P
cl
u
s
t
er
s
an
d
in
c
r
e
a
s
e
th
e
s
e
n
s
itiv
ity
o
f
th
is
C
A
D
s
y
s
te
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
C
anc
e
r
ou
s
l
ung nodul
e
de
t
e
c
t
i
on i
n
c
om
put
e
d t
om
ogr
aphy
i
m
age
s
(
A
y
m
an A
bu
B
ak
e
r
)
2435
F
ig
ur
e
.
3
.
T
w
o
co
n
cen
t
r
i
c
ci
r
cu
l
ar
m
as
k
s
F
i
gur
e
4.
P
C
L
N
pr
oc
e
s
s
i
ng r
e
s
ul
t
4
.
3
.
T
ex
t
u
re
f
ea
t
u
re ex
t
ra
ct
i
o
n
C
a
nc
e
r
ous
nodul
e
s
ar
e
a
s
ma
ll s
iz
e
le
s
s
th
a
n
4
mm
w
it
h
a
br
i
gh
t
r
e
gi
on
co
m
p
a
r
in
g
w
ith
t
he
s
ur
r
oundi
ng r
e
gi
ons
i
n
t
he
l
ung.
S
o,
i
n or
de
r
t
o ha
ve
a
pl
a
t
f
o
r
m
t
ha
t
c
a
n he
l
p
r
a
di
ol
ogi
s
t
i
n d
i
a
gnos
i
s
t
he
s
e
t
ype
s
of
t
um
or
s
,
w
e
ne
e
d t
o
f
oc
us
on t
he
vi
e
w
c
ha
r
a
c
t
e
r
i
s
t
i
c
s
of
c
a
nc
e
r
ous
nodul
e
s
i
n
CT
-
I
m
ag
es
.
A
s
f
ir
s
t s
te
p
,
th
e
c
a
nc
e
r
ous
nodul
e
s
ar
e
en
h
an
ce
d
a
nd
a
pot
e
nt
i
a
l
c
a
nc
e
r
ous
nodul
e
s
a
lg
o
r
ith
m
is
imp
l
e
me
n
te
d
t
o
d
et
ect
can
cer
o
u
s
cl
u
s
t
er
s
i
n
CT
-
i
m
ag
es
.
B
o
th
a
lg
o
r
ith
ms
a
r
e
e
f
f
e
c
tiv
e
ly
d
et
ec
t
t
he
c
a
nc
e
r
ous
r
e
gi
ons
i
n
t
he
C
T
i
m
a
ge
but
w
i
t
h hi
gh
num
be
r
of
f
a
l
s
e
po
s
i
t
i
ve
(
F
P
)
r
e
gi
ons
.
I
n
t
hi
s
s
e
c
t
i
on
,
w
e
imp
le
me
n
t a
n
e
w
ap
p
r
o
ach
t
o
r
e
duc
e
num
be
r
of
F
P
r
e
gi
ons
us
i
ng
t
e
xt
ur
e
f
e
a
t
ur
e
s
.
H
a
v
i
ng m
i
ni
m
um
num
be
r
o
f
F
P
r
e
gi
on w
i
l
l
e
nha
nc
e
t
he
pe
r
f
or
m
a
nc
e
of
our
C
A
D
s
ys
t
e
m
.
F
ir
s
t o
r
d
e
r
t
ex
t
u
r
e
f
eat
u
r
es
b
as
e
i
s
us
e
d
i
n
t
hi
s
pa
pe
r
t
o
r
e
duc
e
de
t
e
c
t
e
d num
be
r
o
f
F
P
r
e
gi
ons
.
S
o,
t
w
o
d
at
as
et
s
ar
e g
en
er
at
ed
t
o
ach
i
ev
e
t
hi
s
goa
l
.
T
P
da
t
a
s
e
t
i
s
ge
ne
r
a
t
e
d us
i
ng
223 e
xpe
r
t
l
y
i
de
nt
i
f
i
e
d (
a
c
t
ua
l
)
T
P
cl
u
s
t
er
s
an
d
t
h
e s
eco
n
d
d
at
as
et
w
h
i
ch
i
s
F
P
d
at
as
et
u
s
ed
4
1
5
act
u
al
F
P
cl
u
s
t
er
s
.
T
h
en
f
i
v
e t
ex
t
u
r
e
f
eat
u
r
es
ar
e cal
cu
l
at
ed
b
as
ed
on t
he
s
e
da
t
a
s
e
t
s
(
T
P
a
nd F
P
)
.
T
he
se
f
eat
u
r
e
s
ar
e
s
u
m
m
ar
i
zed
i
n
(4
)
-
(
8)
.
(
)
=
ℎ
(
)
/
2
4
0
=
4
0
(
3)
W
he
r
e
h(
i
)
i
s
t
he
i
nt
e
ns
i
t
y h
i
s
t
ogr
a
m
a
nd M
,
N
a
r
e
t
he
i
m
a
ge
r
e
gi
on’
s
he
i
ght
a
nd w
i
dt
h
r
e
s
pe
c
t
i
ve
l
y.
−
T
h
e
m
o
d
i
f
i
ed
m
ean
f
eat
u
r
e
=
(
)
2
4
0
=
4
0
(
4)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1693
-
6930
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
,
Vo
l
.
18
, N
o
.
5
,
O
c
t
obe
r
2020:
2432
-
2438
2436
−
T
h
e
m
odi
f
i
e
d
e
nt
r
opy
f
e
a
t
ur
e
=
−
(
)
2
[
(
)
]
2
4
0
=
4
0
(
5)
−
T
h
e
m
odi
f
i
e
d
s
t
a
nda
r
d de
vi
a
t
i
on
f
e
a
t
ur
e
=
(
−
)
2
(
)
2
4
0
=
4
0
(
6)
−
T
h
e
m
odi
f
i
e
d
t
hi
r
d
or
de
r
of
m
om
e
nt
f
e
a
t
ur
e
3
=
(
−
)
3
(
)
2
4
0
=
4
0
(
7)
−
T
h
e
m
odi
f
i
e
d
kur
t
os
i
s
f
e
a
t
ur
e
=
−
4
(
−
)
4
(
)
−
3
2
4
0
=
4
0
(
8)
T
h
e v
al
u
es
o
f
t
h
es
e f
eat
u
r
es
ar
e p
r
es
en
t
ed
as
s
h
o
w
n
i
n
F
i
g
u
r
e 5
.
F
i
gur
e
5
.
T
h
e
fi
rs
t
o
rd
e
r
fe
a
t
u
re
5.
RE
S
UL
T
S
AND ANAL
YS
I
S
T
he
c
a
nc
e
r
ous
nodul
e
e
nha
nc
e
m
e
nt
a
nd d
e
t
e
c
t
i
on
a
l
gor
i
t
hm
i
s
a
ppl
i
e
d on 60 C
T
-
i
m
a
ge
s
.
T
he
n,
f
i
ve
t
e
xt
ur
e
f
e
a
t
ur
e
s
a
r
e
ge
ne
r
a
t
e
d (
E
nt
r
opy
,
m
e
a
n,
S
T
D
,
ku
r
t
os
i
s
a
nd s
ke
w
ne
s
s
)
w
hi
c
h w
e
r
e
pr
oc
e
s
s
e
d us
i
ng
s
t
at
i
s
t
i
cal
an
al
y
s
i
s
t
o
r
ed
u
ce t
h
e d
et
ect
ed
F
P
r
eg
i
o
n
s
.
T
h
e
a
l
gor
i
t
hm
i
s
s
ubj
e
c
t
i
ve
l
y e
va
l
ua
t
e
d us
i
ng t
hr
e
e
r
a
di
ol
ogi
s
t
,
w
he
r
e
num
be
r
o
f
de
t
e
c
t
e
d F
P
r
e
gi
ons
a
r
e
c
ount
e
d pe
r
i
m
a
ge
.
A
l
s
o T
P
pe
r
c
e
nt
a
ge
of
e
a
c
h i
m
a
ge
i
s
al
s
o
r
eco
d
ed
.
F
i
n
al
l
y
,
t
h
e
av
er
ag
e o
f
d
et
ect
ed
F
P
r
eg
i
o
n
an
d
T
P
p
er
cen
t
ag
e i
s
p
r
es
en
t
ed
i
n
t
h
e
T
ab
l
e 1
af
t
er
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
C
anc
e
r
ou
s
l
ung nodul
e
de
t
e
c
t
i
on i
n
c
om
put
e
d t
om
ogr
aphy
i
m
age
s
(
A
y
m
an A
bu
B
ak
e
r
)
2437
pr
oc
e
s
s
i
ng 60 C
T
-
i
m
ag
e.
F
r
o
m
T
ab
l
e
1
,
i
t
i
s
e
v
i
de
nt
t
ha
t
our
a
l
gor
i
t
hm
a
c
hi
e
ve
s
a
good pe
r
f
or
m
a
nc
e
i
n
de
t
e
c
t
i
ng c
a
nc
e
r
ous
nodul
e
s
but
s
t
i
l
l
num
be
r
o
f
F
P
i
s
s
l
i
ght
l
y
hi
gh.
F
i
gur
e
6
s
how
t
he
r
e
s
ul
t
s
of
C
T
p
r
o
ces
s
i
n
g
s
t
ag
es
w
h
er
e t
h
e i
m
ag
e p
r
o
ce
s
s
e
d u
s
i
ng e
nha
nc
e
m
e
nt
a
l
gor
i
t
hm
,
t
he
n de
t
e
c
t
i
on r
e
gi
on of
i
nt
e
r
e
s
t
.
F
i
n
al
l
y
t
h
e t
ex
t
u
r
e f
eat
u
r
e
an
al
y
s
i
s
i
s
i
m
p
l
em
en
t
ed
t
o
r
ed
u
ce o
f
t
h
e d
et
ect
ed
F
P
r
eg
i
o
n
s
.
(
a)
(
b)
(
c)
(
d)
F
ig
ur
e
6
.
A
ccu
r
at
e
d
et
ect
i
o
n
o
f
can
cer
o
u
s
;
(
a
)
o
r
ig
in
a
l ima
g
e
,
(
b)
i
m
ag
e en
h
an
cem
en
t
,
(c
) i
ma
g
e
w
ith
P
L
C
N
,
(d
)
t
ex
t
u
r
e f
eat
u
r
es
an
al
y
s
i
s
T
ab
l
e
1
.
I
ma
g
e
q
u
a
lity
e
v
a
lu
a
tio
n
A
v
er
ag
e o
f
d
et
ect
ed
F
P
r
eg
i
o
n
(
cl
u
s
t
er
)
A
v
e
ra
g
e
o
f T
P
(%)
R
a
d
io
lo
g
is
t 1
28
97%
R
a
d
io
lo
g
is
t 2
23
98%
R
a
d
io
lo
g
is
t 2
24
96%
6.
CO
NCL
US
I
O
N
T
hi
s
pa
pe
r
pr
e
s
e
nt
s
t
he
ongoi
ng e
f
f
or
t
i
n e
nha
nc
i
ng a
nd de
t
e
c
t
i
ng t
he
c
a
nc
e
r
ous
nodul
e
s
i
n t
he
C
T
i
m
a
ge
s
.
T
hi
s
pr
opos
e
s
a
l
gor
i
t
hm
i
s
di
vi
de
d
t
o
t
hr
e
e
pha
s
e
s
.
T
he
c
a
nc
e
r
ous
nodu
l
e
s
a
r
e
e
nha
nc
e
d us
i
ng
t
h
e L
ap
l
aci
an
f
i
l
t
er
.
T
h
en
,
t
h
e av
er
ag
e f
i
l
t
er
i
s
m
o
d
i
f
i
e
d
ba
s
e
d on t
he
l
ow
e
r
a
nd
uppe
r
gr
e
y
l
e
ve
l
s
of
t
he
c
a
nc
e
r
ous
nodul
e
s
i
n t
he
C
T
i
m
a
ge
s
.
T
hi
s
i
nc
a
s
e
,
s
l
i
ght
l
y e
nha
nc
e
s
t
he
c
a
n
c
e
r
ous
no
dul
e
s
i
n
t
he
m
a
m
m
ogr
a
m
i
m
a
ge
s
.
I
n
t
he
s
e
c
ond pha
s
e
,
t
he
pot
e
nt
i
a
l
c
a
nc
e
r
ous
nodul
e
s
r
e
gi
ons
a
r
e
de
t
e
c
t
e
d us
i
ng
mu
lti s
ta
tis
tic
a
l f
ilte
r
w
h
ic
h
r
e
s
u
lts
la
r
g
e
n
u
m
b
e
r
o
f
F
P
r
e
g
io
n
s
.
F
in
a
lly
,
te
x
tu
r
e
f
e
a
tu
r
e
a
n
a
ly
s
is
is
i
m
pl
e
m
e
nt
e
d t
o
r
e
duc
e
t
he
de
t
e
c
t
e
d F
P
r
e
gi
ons
.
A
s
a
r
e
s
ul
t
,
t
he
p
r
opos
e
d a
l
gor
i
t
hm
i
s
s
ubj
e
c
t
i
ve
l
y a
nd
obj
e
c
t
i
ve
l
y t
e
s
t
e
d on 60 C
T
i
m
a
ge
s
a
nd i
t
s
how
s
t
ha
t
i
s
a
l
gor
i
t
hm
c
a
n de
t
e
c
t
t
he
c
a
nc
e
r
ous
nodul
e
s
w
i
t
h
an
av
er
ag
e r
at
e 9
7
.
6
%
w
i
t
h
F
P
r
e
gi
ons
of
25 c
l
us
t
e
r
.
A
C
K
N
O
W
LED
G
EM
EN
TS
T
he
a
u
t
h
or
i
s
gr
a
t
e
f
u
l
t
o t
he
A
p
pl
i
e
d S
c
i
e
n
c
e
P
r
i
v
a
t
e
U
ni
ve
r
s
i
t
y,
A
m
m
a
n,
J
or
d
a
n,
f
or
t
h
e
f
ul
l
f
i
n
a
nc
i
a
l
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1693
-
6930
T
E
L
KOM
NI
KA
T
el
eco
m
m
u
n
C
om
put
E
l
C
ont
r
o
l
,
Vo
l
.
18
, N
o
.
5
,
O
c
t
obe
r
2020:
2432
-
2438
2438
R
EF
ER
EN
C
ES
[1
]
S
.
K
.
S
ha
h,
M
.
F
.
M
c
Nitt
-
Gr
a
y
,
S
.
R
.
R
oge
r
s,
J
.
G
.
G
oldi
n,
R
.
D
.
S
uh,
J
.
W
.
S
a
y
r
e
,
e
t
al
.
,
“
C
om
p
ute
r
a
i
de
d
c
ha
r
a
c
te
r
iz
a
t
io
n of
t
he
s
ol
ita
r
y p
ulm
ona
r
y no
du
le
u
si
ng
vol
um
e
tr
ic
a
n
d c
o
ntr
a
st e
n
ha
nc
e
m
e
n
t f
e
a
tur
e
s
,
”
Ac
ade
mi
c
Ra
di
ol
ogy
,
v
ol.
12
, n
o
.
10
, p
p
.
13
10
-
13
19
,
20
05
[2
]
S
.
B
.
Lo
,
M
.
T
.
F
r
e
e
d
m
a
n
,
L
.
B
.
G
illi
s,
C
.
S
.
W
hite
,
S
.
K
.
Mu
n
,
“
C
om
p
ute
r
-
Ai
de
d
De
te
c
ti
on of
L
u
ng N
od
ule
s
on
CT
w
i
th a
C
om
pu
te
r
iz
e
d P
ulm
ona
r
y Ve
sse
l S
up
pr
e
sse
d F
unc
ti
on,
”
A
me
r
ic
a
n J
ou
rn
al
of R
oe
n
tge
no
lo
gy
,
vol.
21
0
,
no
.
3
, p
p
.
48
0
-
48
8
,
20
18
.
[3
]
W
.
Li
,
“
R
e
c
e
nt pr
ogr
e
ss
in c
om
p
ute
r
-
a
ide
d d
ia
g
no
si
s o
f
lun
g n
od
ule
s o
n t
hi
n
-
se
c
ti
on C
T
,
”
C
om
pu
te
r
ize
d Me
d
ic
a
l
I
ma
gi
ng G
ra
ph
,
v
ol.
31
,
no.
4
-
5
, p
p
.
24
8
-
2
57
,
20
07
.
[4
]
F
ur
qa
n S
ha
uka
t
,
G
uli
sta
n R
a
ja
,
Ale
j
a
n
dr
o F
.
F
r
a
ngi
,
“
C
om
pute
r
-
a
ide
d de
te
c
ti
on of
l
un
g no
du
le
s
: a
r
e
vie
w,
”
J
ou
rn
al
o
f Me
dic
al I
m
ag
in
g
,
vo
l.
6
, n
o
.
2
,
pp
.
2
09
01
-
201
9
11
,
20
19
.
[5
]
Ani
ta
C
ha
u
dha
r
y,
S
o
ni
t
S
u
khr
a
j S
in
gh,
“
L
u
ng C
a
nc
e
r
De
te
c
ti
on
on C
T I
m
a
ge
s B
y
Us
in
g I
m
a
ge
P
r
oc
e
ss
in
g
,
”
2
0
12
I
nte
rn
at
io
na
l C
on
fe
re
nc
e
o
n C
om
pu
ti
ng
Sc
ie
nc
e
s
,
pp
.
14
2
-
14
6,
20
12.
[6
]
J
.
P
u
,
B
.
Z
he
ng,
J
.
K
.
L
e
a
d
e
r
,
e
t
a
l
.,
“
An a
utom
a
te
d C
T ba
se
d l
un
g nod
ule
de
te
c
t
io
n sc
he
m
e
us
in
g ge
om
e
t
r
ic
a
na
l
ys
is of
si
gne
d di
sta
nc
e
f
ie
ld
,”
M
ed
ic
al
Phy
s
i
cs
,
vo
l.
35
,
no.
8
,
p
p
.
34
53
-
34
61
,
20
08
.
[7
]
S
.
M
a
tsum
o
to,
Y
.
Ohn
o,
H
.
Y
a
m
a
g
a
t
a
e
t
a
l
.
,
“
C
om
pu
te
r
-
a
i
de
d de
t
e
c
t
io
n of
lun
g no
du
le
s on m
u
lt
ide
te
c
t
or
r
o
w
c
om
p
ute
d t
om
o
gr
a
p
hy
us
in
g t
hr
e
e
-
d
im
e
n
si
ona
l a
na
ly
si
s of
n
od
ule
c
a
n
di
da
te
s a
n
d t
he
ir
sur
r
ou
nd
in
gs
,
”
Ra
di
at
i
on
M
e
d
ic
i
ne
,
vo
l.
26
,
no
.
9
, p
p
.
9
-
5
62
,
20
08
.
[8
]
S
te
ve
n
S
c
h
a
l
e
k
a
m
p
,
Bram v
an
G
inne
k
e
n,
Em
m
e
li
ne
Koe
da
m
,
M
ir
a
n
da
M
.
S
noe
r
e
n,
e
t
a
l
.
,
“
C
om
pute
r
-
ai
d
ed
De
te
c
ti
on I
m
pr
ove
s De
te
c
t
io
n of
P
ulm
ona
r
y No
d
ule
s in C
he
s
t R
a
d
io
gr
a
p
hs be
yo
nd
the
S
up
por
t
by
B
one
-
su
ppr
e
sse
d I
m
a
ge
s,
”
Ra
di
ol
og
y
,
vo
l.
2
72
,
no.
1
,
p
p
.
120
1
-
12
20
,
20
14
.
[9
]
J
. R
.
S
ousa
,
A.
C
.
S
ilv
a
,
A
. C
.
d
e P
ai
v
a et
al
.
, “
M
e
thodol
ogy f
o
r
a
utom
a
tic
de
t
e
c
tion
of
lung nodul
e
s in
c
om
put
e
r
i
z
e
d
t
o
m
o
g
rap
h
y
i
mag
es
,”
C
omput
e
r M
e
thods
and Programs
i
n
B
io
m
e
di
c
ine
,
vol.
98
,
no.
1
,
pp.
1
-
14
,
2010
.
[
10]
F
a
r
id
od
di
n S
ha
r
ia
t
ya
,
M
o
jta
ba
M
ou
sa
v
ib
,
“
A
pp
li
c
a
t
io
n of
C
AD
s
ys
te
m
s f
or
the
a
u
tom
a
tic
de
te
c
ti
on
of
l
un
g
nod
ul
e
s,
”
I
n
fo
rm
at
ic
s i
n Me
dic
ine
U
nl
oc
k
e
d
,
v
ol.
15
,
n
o
.
1
0,
pp.
10
01
-
10
13
,
20
19
.
[
11]
N
BIA
,
“
L
I
DC
l
un
g
no
du
le
im
a
ge
da
ta
ba
se
,
”
Na
tio
na
l
Can
cer Ima
g
i
ng
A
r
c
h
ive
.
[
On
li
ne
]
.
A
va
i
la
ble
:
htt
ps
:/
/im
a
gi
ng.
nc
i.
ni
h.
go
v/
.
[
12]
Yin
gr
u Z
ha
o,
G
e
e
r
tr
u
ida
H.
d
e
B
oc
k,
R
oz
e
m
a
r
ij
n V
lie
ge
n
tha
r
t,
R
ob
J.
va
n
Kla
ve
r
e
n,
Yi
ng
Wa
ng,
Lu
c
a
B
ogo
n
i,
P
im
A.
de
Jon
g,
W
i
lle
m
P
.
M
a
li,
P
e
te
r
M
.
A.
va
n Ooi
j
e
n
a
n
d
M
a
tt
hi
js O
ud
ke
r
k,
“
Perfo
rman
ce o
f co
mp
u
t
er
-
a
id
e
d
de
te
c
ti
on o
f
pu
lm
o
na
r
y
no
du
le
s in lo
w
-
d
ose
C
T: c
om
p
a
r
is
on w
it
h d
ou
ble
r
e
a
di
ng
by
no
du
le
v
ol
um
e
,
”
Eu
ro
pe
an
Ra
di
ol
ogy
,
v
ol.
22
, p
p
.
20
76
-
20
84
,
20
12
.
[
13]
Z
he
ng
ha
o S
hi,
J
un B
a
i,
L
if
e
n
g He
,
Ts
uy
os
hi
Na
ka
m
ur
a
,
Qua
nz
hu
Ya
o,
Hi
de
n
or
iI
to
h “
A M
e
th
od
f
or
En
ha
nc
i
ng
L
ung N
od
ule
s i
n C
he
s
t R
a
d
io
gr
a
p
hs b
y U
se
of
L
oG
F
il
t
e
r
,
”
200
9 2n
d I
n
te
r
na
ti
on
al C
o
ng
re
s
s o
n I
ma
ge
a
nd S
ig
na
l
Pro
c
e
s
si
ng
,
Ti
a
nj
in
,
pp.
1
-
4,
Oc
t 1
7
-
19
,
20
09.
[
14]
D
.
Vive
ka
na
nda
n,
S
un
il R
e
tm
in
R
a
j,
“
A
F
e
a
tur
e
E
xtr
a
c
ti
on
M
o
de
l f
or
A
sse
ss
in
g
th
e
G
r
owt
h
of
L
ung C
a
nc
e
r
in
C
om
pu
te
r
Ai
de
d Dia
gn
os
is
,
”
IE
E
E
-
I
n
te
r
na
ti
on
al
C
o
nfe
re
nc
e
on
Re
c
e
nt T
re
n
ds
in I
nf
or
ma
ti
on
T
e
c
h
no
l
ogy
,
ICRT
IT
MIT
,
pp
.
9
53
-
9
58,
C
he
nna
i,
I
nd
ia
,
Ju
ne
5,
20
11.
[
15]
T
.
M
e
ssa
y,
R
.
H
a
r
d
i
e
,
S
.
Ro
g
ers
, “
A ne
w c
om
p
uta
ti
ona
l
ly e
f
f
ic
ie
nt C
A
D s
ys
te
m
f
or
pu
lm
o
na
r
y
no
du
le
de
te
c
t
io
n i
n
C
T im
a
ge
r
y
,”
M
e
d
i
c
al
I
ma
g
e
An
al
y
si
s
,
v
ol.
14
,
no.
3
,
pp.
390
-
4
06
,
20
10
.
[
16]
Q.
Li
,
F.
Li
,
K
.
Doi
,
“
C
o
m
pute
r
iz
e
d
d
e
te
c
tion
of
lun
g nodul
e
s
in
thin
-
se
c
tion
C
T
im
a
ge
s
by
use
o
f
s
e
le
c
t
ive
en
h
an
c
em
en
t
fi
l
t
ers
an
d
an
au
t
o
m
at
ed
ru
l
e
-
b
as
ed
cl
as
s
i
fi
e
r
,
”
A
ca
d
em
i
c
Radiol
ogy
,
vol.
15
,
no.
2
,
pp.
165
-
175
,
2008
.
[
17]
S
.
Ya
m
a
m
ot
o,
M
.
M
a
tsum
ot
o,
Y
.
Ta
te
no,
T
.
I
in
um
a
,
T
.
M
a
tsum
ot
o,
“
Qu
oi
t
f
il
te
r
-
a
ne
w f
i
lte
r
ba
se
d
o
n
m
a
the
m
a
tic
a
l m
or
ph
ol
og
y t
o e
xtr
a
c
t t
he
is
ola
te
d
sha
do
w,
a
nd
i
ts
a
pp
lic
a
ti
on
t
o a
u
tom
a
tic
de
te
c
t
io
n of
lu
ng
c
a
nc
e
r
in X
-
ra
y
CT
,”
P
roc
e
e
di
ng
s of
13
th I
n
te
r
na
ti
on
al C
on
fe
r
e
nc
e
on
Pa
tte
rn Re
c
og
ni
ti
on
,
pp
.
3
-
7
,
19
96
.
[
18]
M
a
c
ie
j Da
jn
ow
ie
c
,
Ja
va
dA
lir
e
z
a
ie
,
“
C
om
p
ute
r
S
im
ul
a
t
i
on
f
or
S
e
gm
e
nta
ti
on
of
L
ung
N
od
ule
s
o
n
CT
Imag
e
s
,
”
2
0
04
I
EE
E I
nte
rn
at
io
na
l C
o
nfe
re
nc
e
o
n Sy
ste
ms,
M
an a
nd C
y
be
r
ne
t
ic
s
,
v
ol.
5,
pp
.
44
91
-
44
96,
20
04
[
19]
Y
.
L
e
e
,
T
.
H
a
r
a
,
H
.
F
ujita
,
S
.
I
toh,
T
.
I
sh
iga
ki
,
“
A
ut
om
a
te
d de
te
c
t
io
n of
p
ulm
ona
r
y no
du
le
s i
n he
lic
a
l C
T im
a
ge
s
ba
se
d on a
n im
pr
ove
d
te
m
pla
te
-
m
a
tc
hi
ng te
c
h
ni
que
,
”
I
EEE T
ra
ns
ac
t
io
ns o
n Me
dic
al I
m
ag
in
g
,
vo
l.
20
,
no.
7
,
pp.
5
95
-
6
04
,
20
01
.
[
20]
R
.
O
p
fer,
R
.
W
i
e
m
k
e
r
,
“
P
e
r
f
or
m
a
nc
e
a
na
ly
si
s
f
or
c
om
p
ute
r
a
ide
d lu
ng no
du
le
de
te
c
ti
on on
L
I
DC
da
ta
,
”
Pro
c
e
e
d
in
gs o
f S
PI
E
-
T
he
I
n
te
r
na
ti
on
al
Soc
ie
ty
f
or
O
pti
c
al
E
ng
in
e
e
ri
ng
,
pp.
1
-
9
,
20
07
.
[
21]
B
.
G
olosi
o,
G
.
L
.
M
a
s
a
l
a
,
A
.
P
ic
c
ioli,
P
.
Ol
iva
,
M
.
C
a
r
pine
ll
i
,
“
A no
ve
l m
u
lt
i
thr
e
sh
ol
d m
e
th
od f
or
n
o
dul
e
de
te
c
ti
on i
n l
un
g C
T
,”
M
e
d
i
c
al
Ph
y
s
ic
s,
v
ol.
36
,
no.
8
,
p
p
.
360
7
-
36
18
,
20
09
.
[
22]
K.
D
e
v
a
k
i
,
V.
M
ur
a
li
B
ha
ska
r
a
n,
“
S
tu
dy of
C
om
p
ute
d Tom
o
gr
a
p
hy I
m
a
ge
s of
the
L
ung
s,
”
201
1 I
nte
r
na
ti
o
na
l
C
onf
e
re
nc
e
on
Re
c
e
n
t T
re
n
ds
in I
n
fo
rm
at
io
n T
e
c
h
no
lo
g
y
(
I
C
RT
I
T
)
,
pp.
8
37
-
8
42,
20
11
.
[
23]
B
.
S
a
hine
r
,
L
.
M
.
Ha
djii
sk
i,
H
.
P
.
C
ha
n,
J
.
S
hi,
P
.
N
.
C
a
s
c
a
d
e
,
E
.
A
.
K
a
z
e
r
o
o
n
i
,
e
t
a
l
., “
Ef
f
e
c
t of
C
AD
on
r
a
di
ol
og
is
ts’
de
te
c
t
io
n of
lu
ng n
od
ule
s on t
hor
a
c
ic
C
T sc
a
n
s: o
bse
r
ve
r
pe
r
f
or
m
a
nc
e
st
ud
y
,
”
Pr
oc
SPI
E Me
d
I
ma
gi
ng
,
pp
.
1
-
7
,
20
07
.
[
24]
A
.
A
.
A
b
u
B
ak
er
,
R
.
S
.
Q
a
hwa
ji
,
S
ta
n
I
pson
,
M
ohm
m
a
d
H
.
S
a
l
e
h,
“
On
e
S
c
a
n C
onne
c
te
d
C
o
m
pone
nt
L
a
b
e
ll
ing
Te
c
hnique
,
”
2007
I
EEE
I
nte
rnational
C
onfe
r
e
nc
e
on
Sign
al Proc
e
ssing and
C
ommunic
ations
,
pp.
1283
-
1286
,
2007
.
[
25]
R
.
C
.
Ha
r
die
,
S
.
K
.
R
oge
r
s,
T
.
W
ils
on,
A
.
R
oge
r
s
,
“
P
e
r
f
or
m
a
nc
e
a
na
ly
si
s
of
a
ne
w
c
om
pu
te
r
a
ide
d
de
te
c
ti
on
s
ys
te
m
f
or
ide
nt
if
y
in
g l
un
g no
d
u
le
s
on c
he
st r
a
di
ogr
a
ph
s
,
”
M
e
d
ic
a
l
I
m
age
An
al
ys
i
s
,
vo
l.
12
,
n
o
.
3
,
pp.
24
0
-
2
58
,
20
08
.
Evaluation Warning : The document was created with Spire.PDF for Python.