T
E
L
KO
M
NI
K
A
,
V
ol
.
14,
N
o.
3,
S
ept
em
ber
20
16,
pp.
10
90
~
109
8
I
S
S
N
:
1
693
-
6
930
,
ac
c
r
edi
t
ed
A
b
y
D
IK
T
I,
D
e
c
r
e
e
N
o
:
58/
D
I
K
T
I
/
K
ep/
2013
D
O
I
:
10.
12928/
T
E
LK
O
M
N
I
K
A
.
v
1
4
i
3
.
3440
10
90
R
ec
ei
v
ed
J
anu
ar
y
2
7
,
201
6
;
R
ev
i
s
ed
Ma
y
22
,
20
1
6
;
A
c
c
ept
ed
J
u
ne
8
,
2
01
6
B
r
i
gh
tne
s
s
a
n
d
C
on
tr
a
s
t M
odi
fi
c
a
ti
o
n i
n
Ult
r
aso
n
o
g
r
ap
h
y
I
m
ag
es
Usi
n
g
E
d
g
e
D
et
ec
t
io
n
Resu
l
t
s
R
e
tn
o
S
u
p
r
i
y
a
n
ti
*
1
,
S
u
w
i
tn
o
2
,
Y
o
g
i
R
a
m
a
d
h
a
n
i
3
,
H
a
r
i
s
B
W
i
d
o
d
o
4
,
T
u
ti
k
I
d
a
R
o
s
a
n
ti
5
1
,2
,3
E
l
e
c
t
r
i
c
a
l
E
ng
i
neer
i
ng D
ep
ar
t
m
e
nt
,
J
end
er
al
S
oed
i
r
m
a
n U
ni
v
er
s
i
t
y
4
,5
M
edi
c
a
l F
a
c
u
lt
y
,
J
ender
al
S
oedi
r
m
an U
ni
v
er
s
i
t
y
,
J
l
.
H
R
.
B
oenj
a
m
i
n 70
8 P
ur
w
ok
er
t
o
,
P
hone:
+
62
-
281
-
6
352
92
*
C
or
r
es
po
ndi
ng a
ut
hor
,
e
-
ma
i
l
:
r
et
no_
s
u
pr
i
y
ant
i
@
uns
oed.
ac
.
i
d
A
b
st
r
act
C
ur
r
ent
l
y
,
u
l
t
r
as
ono
gr
aph
y
d
ev
i
c
e be
c
om
e
i
m
po
r
t
ant
equ
i
pm
ent
f
or
s
up
por
t
i
ng d
i
ag
n
os
i
s
in
di
es
as
e
s
.
U
nf
or
t
unet
al
y
,
a
l
o
t
of
u
l
t
r
a
s
ono
gr
ap
hy
i
m
age
s
d
o
not
p
r
ov
i
de
en
ough
i
n
f
or
m
at
i
o
n f
or
s
u
ppor
t
i
n
g
di
agn
os
i
s
es
pec
i
al
l
y
i
m
age
s
pr
oduc
ed by
l
ow
-
r
e
s
ol
ut
i
o
n
u
l
t
r
a
s
ono
gr
ap
hy
.
I
t
i
s
c
a
us
e
d
b
y
i
m
age
qual
i
t
y
t
h
a
t
has
been
pr
od
uc
ed
i
s
i
nad
e
quat
e
be
c
au
s
e
o
f
no
i
s
e
.
T
h
i
s
r
es
e
ar
c
h
ai
m
s
t
o
i
m
pr
ov
e
i
m
age
qua
l
i
t
y
b
y
m
odi
f
y
i
ng br
i
ght
n
es
s
and c
o
nt
r
as
t
t
o t
he ed
ge det
ec
t
i
on
al
gor
i
t
hm
s
.
B
y
m
odi
f
y
i
ng t
h
e
br
i
ght
n
es
s
and
c
ont
r
as
t
w
i
l
l
c
a
us
e t
he v
a
l
ue o
f
s
t
and
ar
d de
v
i
at
i
on o
f
t
he ul
t
r
a
s
ono
gr
ap
hy
i
m
age
i
s
l
ow
er
ed.
R
ai
s
i
n
g
s
et
t
i
ng
v
al
u
es
w
i
l
l
c
au
s
e
d
ev
i
at
i
o
n
s
t
andar
d
v
a
l
ue
b
ec
om
e
s
m
al
l
er
,
and
a
l
s
o
t
he
r
es
u
l
t
of
s
t
a
nd
ar
d
de
v
i
at
i
on
i
s
i
nv
er
s
e
l
y
pr
opor
t
i
on
al
t
o t
he
v
al
ue
of
R
M
S
E
.
T
he r
es
u
l
t
s
s
h
ow
t
hat
t
hi
s
m
odi
f
i
c
a
t
i
on
c
a
n
i
m
pr
ov
e
i
m
ag
e
qual
i
t
y
by
r
edu
c
i
n
g no
i
s
e
s
i
gni
f
i
c
an
t
l
y
.
Ke
y
w
o
rd
s
:
ul
t
r
as
ono
gr
aph
y
i
m
age,
l
ow
r
es
o
l
ut
i
on,
br
i
g
ht
ne
s
s
,
c
on
t
r
as
t
,
s
t
and
ar
d d
ev
i
at
i
o
n,
r
edu
c
i
ng n
oi
s
e
C
o
p
y
r
i
g
h
t
©
20
16 U
n
i
ver
si
t
a
s A
h
mad
D
ah
l
an
.
A
l
l
r
i
g
h
t
s r
eser
ved
.
1
.
I
n
tr
o
d
u
c
ti
o
n
T
he dev
e
l
opm
ent
of
i
nf
or
m
at
i
o
n t
ec
hn
ol
og
y
(I
T
)
has
an
i
m
pac
t
on
t
he d
ev
el
o
pm
ent
of
v
ar
i
ous
as
pec
t
s
of
hum
an l
i
f
e.
O
ne
as
pec
t
t
h
at
af
f
ec
t
ed t
he
de
v
e
l
opm
ent
of
t
he I
T
w
or
l
d
i
s
hea
l
t
h.
O
ne
par
t
of
I
T
i
s
di
gi
t
a
l
i
m
age pr
oc
es
s
i
n
g.
O
n
t
he
ot
her
ha
nd,
i
n
t
he
f
i
el
d
of
heal
t
h t
her
e
ar
e
t
ec
hn
ol
ogi
es
t
o
m
ax
i
m
i
z
e
t
he
r
ol
e
of
di
gi
t
al
i
m
age
pr
oc
es
s
i
ng
i
n
s
up
por
t
i
ng
h
eal
t
h
s
er
v
i
c
es
es
pec
i
a
l
l
y
f
or
di
ag
nos
i
ng
w
hi
c
h i
s
t
h
e us
e of
ul
t
r
as
onogr
aph
y
eq
ui
pm
ent
.
U
l
t
r
as
onogr
aph
y
i
s
us
ed
as
di
agn
os
t
i
c
ai
d
b
y
v
i
s
ua
l
i
z
ed
hum
an
bod
y
or
gans
s
uc
h
ex
am
pl
e
i
s
ut
er
us
ar
ea.
E
x
am
i
nat
i
o
n
us
i
ng
u
l
t
r
as
o
nogr
a
ph
y
f
or
ut
er
us
ar
ea
i
s
v
er
y
us
ef
ul
es
pec
i
al
l
y
f
or
hi
g
h
-
r
i
sk
pr
egn
anc
i
es
.
U
l
t
r
as
on
ogr
aph
y
ex
am
i
nat
i
on i
s
on
e m
et
hod of
s
c
r
eeni
n
g
f
or
c
hec
k
i
ng
pr
egn
anc
y
i
s
c
ons
i
d
er
ed
s
af
e,
non
-
i
n
v
as
i
v
e
,
ac
c
ur
at
e
and
ef
f
ec
t
i
v
e.
C
ur
r
ent
l
y
,
u
l
t
r
as
onogr
a
ph
y
i
m
age has
e
v
o
l
v
ed,
w
hi
c
h
f
or
m
er
l
y
onl
y
us
e
t
w
o
-
di
m
ens
i
on
al
ul
t
r
as
o
nogr
aph
y
i
m
age i
s
no
w
a
t
hr
ee
-
d
i
m
ens
i
onal
ul
t
r
as
o
und
i
m
age
and al
s
o
f
o
ur
-
di
m
ens
i
ona
l
ul
t
r
as
ono
g
r
aph
y
i
m
age.
U
nf
or
t
unat
e
l
y
,
es
p
ec
i
a
l
l
y
i
n I
nd
ones
i
a,
a
h
i
gh
-
r
es
o
l
ut
i
o
n u
l
t
r
as
on
ogr
a
ph
y
w
i
t
h
m
os
t
l
y
on
l
y
a
va
i
l
a
b
l
e
at
l
ar
ge
hos
p
i
t
a
l
s
i
n
bi
g c
i
t
i
es
,
bec
aus
e
of
t
he h
i
g
h pr
i
c
e
of
t
he
eq
ui
p
m
ent
.
O
n t
he
ot
her
ha
nd,
pat
i
e
nt
s
s
pr
ea
d i
n al
l
ar
ea i
nc
l
ud
i
n
g i
n s
m
al
l
c
i
t
y
w
h
er
e hos
p
i
t
a
l
s
or
heal
t
h c
l
i
ni
c
s
hav
e t
w
o
-
d
i
m
ent
i
on
al
u
l
t
r
a
s
onogr
a
ph
y
on
l
y
.
T
he i
m
age obt
ai
ned b
y
u
l
t
r
as
on
og
r
aph
y
de
v
i
c
e,
s
o
m
et
i
m
es
hav
e
a l
os
s
of
qua
l
i
t
y
t
hat
c
a
n b
e r
a
nges
c
ont
r
as
t
,
g
eom
et
r
i
c
di
s
t
or
t
i
on,
f
u
z
z
i
nes
s
or
noi
s
e
[
1]
.
F
ig
ur
e
1 s
h
o
w
s
a
n ex
am
pl
e of
i
m
age pr
od
uc
ed b
y
t
w
o d
i
m
ent
i
on
al
ul
t
r
a
s
onogr
a
ph
y
.
F
ig
ur
e
1.
A
n
E
x
am
pl
e I
m
age of
2
-
D
U
l
t
r
as
on
ogr
ap
h
y
F
i
gur
e
2.
E
x
am
pl
es
of
i
nput
i
m
ages
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
K
A
I
S
S
N
:
1
693
-
6
930
B
r
i
g
ht
nes
s
a
nd C
ont
r
as
t
M
odi
f
i
c
at
i
on
i
n
U
l
t
r
as
o
nogr
ap
hy
I
ma
ge
s U
si
n
g
…
(
R
e
t
no S
upr
i
y
ant
i
)
1091
A
c
c
or
di
n
g t
o F
i
g
ur
e
1,
t
he
i
m
age i
s
not
y
et
f
ul
l
y
has
good i
m
age
qu
al
i
t
y
t
hat
h
as
not
pr
ov
i
de
d c
l
e
ar
i
nf
or
m
at
i
on
abou
t
w
h
at
i
s
c
ont
a
i
n
ed t
h
er
ei
n,
f
or
ex
am
pl
e,
i
s
i
nf
or
m
at
i
on on
t
he
s
hape a
nd l
a
y
o
ut
of
t
he ut
e
r
us
t
hat
ha
v
e
di
f
f
er
ent
l
a
y
o
ut
i
n
eac
h u
l
at
r
as
o
nogr
a
ph
y
i
m
age.
W
hi
l
e
f
or
an
ac
c
ur
at
e
di
agn
os
i
s
w
e
ne
ed
t
o
t
he
ac
c
ur
ac
y
o
f
t
he
s
hape
a
nd
l
a
y
out
of
t
he
ut
er
us
ar
ea,
t
her
ef
or
e
t
he
l
o
w
-
r
e
s
o
lu
t
io
n
im
a
ge
w
i
l
l
be
a
pr
o
bl
em
.
O
ne
of
i
m
age
pr
oc
es
s
i
ng
t
ec
hni
ques
t
h
at
c
an
be
us
ed
t
o
obt
ai
n
t
he
s
hape
an
d
l
a
y
o
ut
of
t
he
ut
er
us
ar
ea
i
s
us
i
ng
ed
ge
de
t
ec
t
i
o
n
m
et
hod.
E
dg
e
det
ec
t
i
o
n
m
et
hod
i
s
us
ed
t
o
ob
t
ai
n
t
he
e
dg
es
of
obj
ec
t
s
.
T
her
e
ar
e
s
ev
er
a
l
oper
a
t
or
s
on
edge
de
t
ec
t
i
on
i
nc
l
ude
C
ann
y
o
per
at
or
,
L
ap
l
ac
i
a
n
of
G
aus
s
i
an
oper
a
t
or
,
P
r
e
w
i
t
t
op
er
at
or
,
R
ober
t
oper
a
t
or
an
d
S
ob
el
o
per
at
or
.
H
o
w
ev
er
,
ac
c
or
di
ng
t
o
t
he
e
dge
det
e
c
t
i
on
pr
oc
es
s
,
ul
t
r
as
on
ogr
a
ph
y
i
m
ages
s
h
oul
d ha
v
e
go
od
qua
l
i
t
y
.
Lo
w
r
es
o
l
ut
i
o
n
q
ual
i
t
y
of
t
he
i
m
age
gr
e
a
t
l
y
i
nf
l
ue
nc
es
t
he
r
es
ul
t
s
of
edge
det
ec
t
i
o
n.
T
o s
ol
v
e t
h
i
s
pr
ob
l
em
,
one
of
w
hi
c
h
i
s
c
apa
bl
e
of
af
f
ec
t
i
ng t
he qu
al
i
t
y
of
t
he ul
t
r
as
on
ogr
a
ph
y
i
m
age i
s
t
he br
i
ght
n
es
s
,
c
ont
r
as
t
an
d noi
s
e
of
t
he
ul
t
r
as
on
ogr
a
ph
y
i
m
age.
T
he
f
i
nal
go
al
of
t
h
i
s
r
es
ea
r
c
h
i
s
t
o
i
m
pr
ov
e
t
h
e
qu
al
i
t
y
of
t
he
i
m
age
pr
oduc
e
d
b
y
t
w
o
-
di
m
ent
i
on
al
u
l
t
r
as
on
ogr
a
ph
y
.
T
her
ef
or
e i
t
i
s
ex
pec
t
ed t
h
at
us
i
n
g t
w
o
-
di
m
ens
i
on
al
ul
t
r
as
on
ogr
a
ph
y
i
n r
ur
a
l
ar
eas
c
an be op
t
i
m
i
z
ed
i
n or
der
t
o i
m
pr
ov
e pe
op
l
e he
al
t
h qua
l
i
t
y
i
n t
hi
s
ar
eas
.
N
auer
[
2]
i
n hi
s
r
es
ear
c
h
i
n
v
es
t
i
gat
e
d t
he ef
f
ec
t
of
t
ube t
e
ns
i
on r
e
duc
t
i
on on
i
m
ag
e
c
ont
r
as
t
a
nd
i
m
age
qu
al
i
t
y
i
n
pe
di
at
r
i
c
t
em
por
al
b
one
c
om
put
ed
t
om
ogr
aph
y
(
C
T
)
.
Ma
sse
y
[
3]
i
de
nt
i
f
i
e
d
6
c
om
m
on
i
m
age
c
apt
ur
e
and
an
al
y
s
i
s
pr
o
bl
em
ar
eas
i
n
s
ub
l
i
n
gua
l
s
i
d
e
-
s
t
r
eam
dar
k
-
f
i
el
d
v
i
d
eos
:
i
l
l
um
i
nat
i
on,
d
ur
at
i
o
n,
f
oc
us
,
c
ont
e
nt
,
s
t
a
bi
l
i
t
y
,
a
nd pr
es
s
ur
e.
T
he c
r
i
t
er
i
a
i
nt
r
o
d
uc
ed
ar
e an obj
ec
t
i
v
e w
a
y
t
o as
s
es
s
t
he qua
l
i
t
y
of
i
m
age ac
qui
s
i
t
i
o
n,
w
i
t
h t
he g
oa
l
of
s
el
ec
t
i
ng
v
i
d
eos
of
adeq
uat
e
q
ual
i
t
y
f
or
an
al
y
s
i
s
.
C
as
e
[4
]
s
um
m
a
r
i
z
e
d
t
he
pr
i
nc
i
pl
es
of
nuc
l
ear
c
a
r
di
ol
og
y
s
i
ng
l
e
phot
on
em
i
s
s
i
on c
om
put
ed t
om
ogr
aph
y
(
S
P
E
C
T
)
a
nd
pos
i
t
r
o
n
em
i
s
s
i
on t
om
ogr
ap
h
y
(
P
E
T
)
i
m
agi
ng an
d t
ec
hn
i
q
ues
f
or
m
ai
nt
ai
ni
n
g qu
al
i
t
y
:
f
r
om
t
he c
al
i
br
at
i
on of
i
m
agi
ng
equi
pm
ent
t
o
pos
t
pr
oc
es
s
i
ng
t
ec
h
ni
que
s
.
P
l
ant
on
[
5]
r
e
v
i
e
w
e
d
t
he
ul
t
r
as
on
ogr
ap
h
y
(
U
S
)
di
a
g
nos
t
i
c
c
r
i
t
er
i
a
,
t
he U
S
per
f
or
m
anc
e i
n
t
he
di
a
gnos
i
s
and
gr
ad
i
n
g of
h
epat
i
c
s
t
eat
os
i
s
,
t
h
e U
S
s
t
e
at
os
i
s
m
odel
s
,
but
al
s
o
i
t
s
l
i
m
i
t
at
i
o
ns
i
n
t
h
e
di
a
gnos
i
s
of
s
t
eat
os
i
s
.
I
n
addi
t
i
o
n,
t
he
y
al
s
o
d
i
s
c
us
s
ed 2
m
oder
n
m
et
hods
of
as
s
es
s
i
ng hepa
t
i
c
s
t
eat
os
i
s
us
i
ng ul
t
r
as
ou
nds
,
nam
el
y
t
he c
om
put
er
i
z
ed pr
oc
es
s
i
ng
of
dat
a f
or
m
i
ng t
he U
S
i
m
age and
t
he c
o
nt
r
o
l
l
e
d
at
t
e
nuat
i
o
n par
am
et
er
m
eas
ur
ed w
i
t
h
uni
di
m
ens
i
ona
l
t
r
ans
i
e
nt
el
as
t
ogr
a
ph
y
.
S
er
bes
[
6]
d
i
d r
es
ear
c
h a
bou
t
t
he
den
oi
s
i
ng
per
f
or
m
anc
e
qua
dr
at
ur
e
s
i
gn
al
s
.
T
he
y
e
v
a
l
uat
ed
a
nd
c
om
par
ed
w
i
t
h
t
he
ot
her
s
b
y
us
i
ng
s
i
m
ul
at
ed
an
d
r
ea
l
q
uadr
at
ur
e
s
i
gn
al
s
.
T
he
qu
ant
i
t
a
t
i
v
e
r
es
ul
t
s
d
em
ons
t
r
at
ed
t
ha
t
t
he
m
odi
f
i
ed
dua
l
-
t
r
ee
-
c
om
pl
ex
-
w
av
el
e
t
-
t
r
ans
f
or
m
-
bas
ed
den
oi
s
i
ng
out
per
f
or
m
s
t
he
c
on
v
ent
i
on
al
di
s
c
r
et
e
w
av
el
e
t
t
r
ans
f
or
m
w
i
t
h t
he
s
a
m
e l
ev
e
l
of
c
om
put
at
i
o
nal
c
om
pl
ex
i
t
y
an
d ex
h
i
bi
t
s
al
m
os
t
equal
per
f
or
m
anc
e t
o t
he
dua
l
-
t
r
ee c
om
pl
e
x
w
av
el
e
t
t
r
ans
f
or
m
w
i
t
h a
l
m
os
t
hal
f
c
o
m
p
ut
at
i
on
al
c
os
t
.
C
ie
c
h
o
le
w
s
k
i
[
7]
des
c
r
i
bed
t
w
o ac
t
i
v
e c
ont
our
m
odel
s
:
t
he edge
-
b
as
ed m
odel
and t
he r
eg
i
o
n
-
bas
ed
m
odel
m
a
k
i
ng
us
e
of
a
m
o
r
phol
o
gi
c
a
l
ap
pr
oac
h,
bot
h
des
i
gne
d
f
or
ex
t
r
ac
t
i
ng
t
he
gal
l
b
l
ad
der
s
hap
e f
r
o
m
ul
t
r
as
onogr
a
ph
y
i
m
ages
.
T
he ac
t
i
v
e c
ont
o
ur
m
odel
s
w
er
e ap
p
l
i
ed t
o
ul
t
r
as
on
ogr
a
ph
y
i
m
ages
w
i
t
hout
l
es
i
o
ns
and t
o t
hos
e
s
how
i
ng s
p
ec
i
f
i
c
di
s
eas
e
uni
t
s
,
n
am
el
y
,
anat
om
i
c
al
c
hanges
l
i
k
e f
o
l
ds
and t
ur
ns
of
t
he gal
l
bl
adder
as
w
el
l
as
pol
y
ps
a
nd gal
l
s
t
on
es
.
T
he
y
al
s
o pr
es
e
nt
s
m
odi
f
i
c
at
i
o
ns
of
t
he ed
ge
-
bas
e
d
m
odel
,
s
uc
h
as
t
he m
et
hod
f
or
r
e
m
ov
i
ng
s
e
lf
-
c
r
os
s
i
ngs
and
l
oo
ps
or
t
he
m
et
hod
of
dam
peni
ng
t
he
i
nf
l
at
i
on
f
or
c
e
w
h
i
c
h
m
ov
es
nodes
i
f
t
he
y
ap
pr
oac
h t
he e
dge
b
ei
n
g det
er
m
i
ned.
S
anc
he
z
[
8]
pr
op
os
ed
a us
ef
ul
t
ool
f
or
i
dent
i
f
yi
n
g
pat
i
ent
s
at
hi
g
h
r
i
s
k
of
s
t
r
ok
e
and
s
el
ec
t
i
n
g
t
h
os
e
w
h
o
c
an
be
nef
i
t
m
os
t
f
r
o
m
r
e
v
as
c
ul
ar
i
z
at
i
on
t
her
ap
i
es
s
uc
h
as
c
ar
ot
i
d
e
ndar
t
er
ec
t
om
y
a
nd
s
t
ent
i
ng
.
C
h
i
f
or
[
9]
dem
ons
t
r
at
ed
t
h
at
p
er
i
o
dont
al
ul
t
r
as
on
ogr
a
ph
y
i
s
a
r
e
l
i
a
bl
e
m
et
hod
w
i
t
h
w
hi
c
h
t
o
i
de
nt
i
f
y
an
d
e
v
a
l
uat
e
t
h
e
at
t
ac
hm
ent
l
ev
e
l
of
t
he
gi
n
gi
v
a
l
j
unc
t
i
on
al
ep
i
t
he
l
i
um
.
V
at
ans
e
v
er
[
10]
pr
opos
e
d
f
et
al
ne
ur
oi
m
agi
ng
s
t
u
d
y
t
ha
t
pr
ov
i
de
nor
m
al
pos
t
er
i
or
f
os
s
a gr
o
w
t
h
t
r
aj
ec
t
or
i
es
d
ur
i
ng
t
he
s
ec
on
d a
nd t
hi
r
d t
r
i
m
es
t
er
s
of
pr
egn
anc
y
v
i
a s
em
i
-
aut
om
at
i
c
s
egm
ent
at
i
on of
r
ec
ons
t
r
uc
t
ed f
et
al
br
ai
n
MR
i
m
ages
and t
o
as
s
es
s
c
o
m
m
on
c
er
ebel
l
ar
m
al
f
or
m
at
i
ons
i
n
c
om
par
i
s
on
w
i
t
h
t
h
e
r
ef
er
enc
e
d
at
a.
A
n
obe
t
t
i
[
1
1]
s
ho
w
ed
a
hi
gh
c
or
r
el
at
i
o
n
bet
w
ee
n t
w
o m
odal
i
t
i
es
t
o i
de
nt
i
f
y
pos
s
i
b
l
e
m
al
po
s
i
t
i
o
ni
n
g
of
a
c
at
het
er
r
es
u
l
t
i
ng
f
r
om
c
annu
l
at
i
on
of
c
ent
r
al
v
ei
ns
,
and
i
t
s
c
om
pl
i
c
at
i
o
ns
.
T
he
l
es
s
t
i
m
e
r
equi
r
e
d
t
o
p
er
f
or
m
ul
t
r
as
onogr
a
ph
y
al
l
o
w
s
ea
r
l
i
er
us
e
of
t
he
c
at
het
er
f
or
t
he
ad
m
i
ni
s
t
r
at
i
on
of
ac
ut
e t
her
a
pi
es
t
hat
c
an b
e l
i
f
e
-
s
av
i
ng f
or
t
he c
r
i
t
i
c
al
l
y
i
l
l
pa
t
i
ent
s
.
T
anak
a
[
12]
ex
am
i
ned t
he
c
l
i
n
i
c
al
ut
i
l
i
t
y
of
t
he m
al
i
g
nanc
y
gr
a
di
ng s
y
s
t
em
f
or
hepa
t
oc
el
l
u
l
ar
c
ar
c
i
nom
a (
H
C
C
)
us
i
ng a
c
o
m
bi
nat
i
on
of
2 di
f
f
er
ent
c
ont
r
as
t
-
enh
anc
ed u
l
t
r
as
o
nogr
a
ph
y
i
m
ages
.
C
hi
em
[
13]
c
om
par
ed
em
er
genc
y
ph
y
s
i
c
i
a
n
-
per
f
or
m
ed
pel
v
i
c
ul
t
r
as
o
nogr
a
ph
y
(
E
P
P
U
)
w
i
t
h
r
ad
i
o
l
og
y
de
par
t
m
ent
-
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1
6
9
3
-
6
930
T
E
L
KO
M
NI
K
A
V
o
l.
14
,
N
o
.
3,
S
ept
em
ber
2016
:
10
90
–
1
098
1092
per
f
or
m
ed
pel
v
i
c
ul
t
r
as
o
n
ogr
ap
h
y
(
R
P
P
U
)
i
n
em
er
genc
y
d
epar
t
m
ent
(
E
D
)
f
em
al
e
pat
i
e
nt
s
r
equi
r
i
ng
pel
v
i
c
ul
t
r
as
o
no
gr
aph
y
an
d
t
h
ei
r
o
ut
c
om
e
s
i
n
r
e
l
a
t
i
o
n
t
o
E
D
l
eng
t
h
of
s
t
a
y
,
E
D
r
eadm
i
s
s
i
on,
and a
l
t
er
na
t
i
v
e di
ag
nos
i
s
,
w
i
t
hi
n a 14
-
da
y
f
ol
l
o
w
-
u
p
per
i
od.
W
ang
[
14]
i
n
v
es
t
i
gat
ed
and c
om
par
ed c
ont
r
as
t
-
en
hanc
ed
ul
t
r
as
o
und (
C
E
U
S
)
i
n t
he c
har
ac
t
er
i
s
at
i
on o
f
hi
s
t
ol
og
i
c
al
l
y
pr
ov
e
n f
oc
al
no
du
l
ar
h
y
per
pl
as
i
a (
F
N
H
)
w
i
t
h c
ont
r
as
t
-
enha
nc
ed c
om
put
ed t
om
ogr
aph
y
(
C
E
C
T
)
.
H
i
z
uk
ur
i
[
15]
de
v
el
op
ed a c
o
m
put
er
i
z
ed d
et
er
m
i
nat
i
o
n s
c
he
m
e
f
or
hi
s
t
ol
o
gi
c
a
l
c
l
as
s
i
f
i
c
at
i
o
n of
br
eas
t
m
as
s
b
y
us
i
ng
obj
e
c
t
i
v
e
f
eat
ur
es
c
or
r
es
po
nd
i
n
g
t
o
c
l
i
ni
c
i
ans
'
s
ubj
ec
t
i
v
e
i
m
pr
es
s
i
ons
f
or
i
m
age
f
eat
ur
es
on
ul
t
r
as
on
ogr
ap
hi
c
i
m
ages
.
A
c
c
or
di
n
g
t
o
t
hes
e
r
es
ear
c
hes
,
i
t
s
e
em
s
t
hat
m
os
t
of
r
es
ear
c
her
us
i
ng
c
om
pl
ex
m
et
hod
t
o
opt
i
m
i
z
e
u
l
t
r
as
onogr
aph
y
f
unc
t
i
on.
I
n
ot
he
r
hand,
s
om
e
r
es
ear
c
her
al
s
o em
phas
i
z
e t
o i
m
pr
ov
e i
m
age
qual
i
t
y
i
n ot
her
c
as
e s
uc
h as
W
ang
[
1
6]
in
h
is
r
es
ear
c
h pur
p
os
e t
o
i
m
age de
noi
s
i
ng
i
s
t
o
r
es
t
or
e t
he or
i
gi
na
l
i
m
age
w
i
t
ho
ut
noi
s
e f
r
om
t
he
noi
s
e
i
m
age,
a
nd
a
t
t
h
e
s
a
m
e
t
i
m
e
m
ai
nt
ai
n
t
h
e
d
et
a
i
l
ed
i
nf
or
m
at
i
on
of
t
he
i
m
age
as
m
uc
h
as
pos
s
i
bl
e.
Z
an
g
[
1
7]
pr
op
os
ed a n
e
w
h
y
br
i
d al
gor
i
t
hm
f
or
t
he i
m
age ed
ge ex
t
r
ac
t
i
o
n and r
ef
i
n
i
n
g,
w
hi
c
h
c
om
bi
ned
t
he
ge
net
i
c
al
gor
i
t
hm
and
an
t
c
ol
o
n
y
al
g
or
i
t
hm
.
W
u
J
i
e
[
18]
pr
o
pos
ed
m
edi
an
f
i
l
t
er
i
n
g
al
gor
i
t
hm
t
o
enhan
c
e
t
ar
get
s
;
an
d
t
he
t
ar
g
et
s
ar
e
s
har
pene
d
b
y
us
i
n
g
l
a
t
er
al
i
nh
i
bi
t
i
o
n
al
g
or
i
t
hm
,
t
he
ed
ge
of
t
a
r
get
s
i
s
out
l
i
ned.
I
n
or
d
er
t
o
get
r
el
i
ab
l
e
t
ar
g
et
r
e
gi
o
n,
a
da
pt
i
v
e
t
hr
es
hol
d s
egm
ent
at
i
on a
l
g
or
i
t
hm
i
s
us
ed t
o ex
t
r
ac
t
need t
ar
g
et
r
eg
i
on,
and c
ha
r
ac
t
er
i
s
t
i
c
s
of
t
ar
get
i
s
us
ed
t
o d
i
s
t
i
ngu
i
s
h
m
ul
t
i
pl
e
t
ar
ge
t
s
.
A
c
c
or
di
n
g
t
o
t
he
m
ai
n
goa
l
of
our
r
es
ear
c
h
f
or
s
uppor
t
i
ng
h
ea
l
t
h
s
er
v
i
c
e
t
ec
hn
ol
o
g
y
f
or
r
ur
al
ar
ea,
i
n t
h
i
s
pap
er
w
e w
i
l
l
em
phas
i
z
e f
or
opt
i
m
i
z
i
n
g ul
t
r
as
ono
gr
ap
h
y
i
m
age q
u
a
l
i
t
y b
y
app
l
y
i
ng
a s
i
m
pl
e a
nd r
o
b
us
t
m
et
hod.
I
n
our
pr
e
v
i
ou
s
r
es
ear
c
h
[
19
-
26
]
,
w
e
de
v
e
l
op
ed s
om
e
s
i
m
pl
e and eas
y
t
o
us
e t
ec
hno
l
og
y
t
o s
u
ppor
t
hea
l
t
h s
er
v
i
c
e
i
n r
ur
a
l
ar
e
a b
as
ed on
i
m
age
pr
oc
es
s
i
ng
an
d
ex
per
t
s
y
s
t
em
.
I
t
i
s
i
m
pl
em
ent
ed i
n
s
om
e
ar
eas
o
f
di
s
eas
es
s
u
c
h
as
c
at
ar
ac
t
,
hi
g
h
r
i
s
k
pr
egnanc
y
,
c
er
v
i
c
al
c
anc
er
a
nd
e
t
c
.
W
e
al
s
o
opt
i
m
i
z
e
d
s
om
e
equi
pm
en
t
f
or
ac
qui
r
i
n
g
dat
a
s
uc
h
as
di
g
i
t
a
l
c
am
er
a,
s
m
ar
t
phone,
l
o
w
-
c
os
t
pa
nor
am
i
c
,
por
t
abl
e
U
S
G
.
I
n
t
hi
s
pap
er
,
w
e
w
il
l o
p
t
im
i
z
e
lo
w
-
c
os
t
u
l
t
r
as
onogr
a
ph
y
w
h
er
e t
h
i
s
m
ac
hi
ne a
v
a
i
l
ab
i
l
i
y
i
s
v
er
y
l
i
m
i
t
ed i
n
dev
el
opi
ng c
o
unt
r
i
es
s
uc
h i
n I
nd
ones
i
a.
2.
R
e
sea
r
ch
M
et
h
o
d
2.
1.
D
at
a
A
cq
u
i
si
t
i
o
n
A
l
l
dat
a
us
ed
i
n
t
h
i
s
r
es
ear
c
h
w
er
e
ob
t
ai
ne
d
f
r
om
G
ener
al
H
os
i
t
al
of
B
a
n
y
um
as
R
ege
nc
y
.
D
at
a
i
s
u
l
t
r
as
on
o
gr
aph
y
i
m
age
w
i
t
h
.
jp
g
ex
t
e
ns
i
on
as
s
ho
w
n
i
n
F
i
g
ur
e 2
.
2.
2.
I
n
cr
e
ased
I
m
ag
e B
r
i
g
h
t
n
ess
T
he bas
i
c
oper
a
t
i
o
n i
s
us
ual
l
y
do
ne i
n t
h
e i
m
age i
s
br
i
ght
nes
s
enh
anc
em
ent
.
T
hi
s
oper
at
i
o
n i
s
per
f
or
m
ed t
o i
nc
r
eas
e t
he
br
i
g
ht
nes
s
of
an i
m
age.
I
f
an i
m
age s
t
i
l
l
h
as
a l
o
w
br
i
ght
nes
s
qua
l
i
t
y
i
t
c
an b
e don
e br
i
ght
n
es
s
enha
nc
em
ent
pr
oc
es
s
.
Mat
hem
at
i
c
al
l
y
[
27
]
, th
e
i
nc
r
eas
ed
br
i
g
ht
nes
s
i
s
do
ne
b
y
ad
di
n
g
a
c
ons
t
a
nt
t
o
t
he
v
al
u
e
of
t
he
e
nt
i
r
e
pi
x
el
.
T
he
add
i
t
i
on
of
br
i
ght
n
es
s
c
an b
e
w
r
i
t
t
en
as
des
c
r
i
be
d i
n
E
qua
t
i
on 1
.
(
,
)
=
(
,
)
+
(
1
)
2
.
3
.
C
o
n
tr
a
s
t S
tr
e
tc
h
i
n
g
T
he c
ont
r
as
t
i
n
an
i
m
age
s
t
at
es
t
he
d
i
s
t
r
i
bu
t
i
o
n of
l
i
ght
a
nd
dar
k
s
hades
of
c
ol
or
.
A
gr
a
y
-
s
c
al
e
i
m
age
i
s
s
ai
d t
o
hav
e
a l
o
w
c
o
nt
r
as
t
w
he
n t
he d
i
s
t
r
i
b
ut
i
on
of
c
ol
or
t
e
nd
t
o
nar
r
o
w
t
he
r
ange
of
gr
a
y
l
e
v
e
l
s
.
C
o
nv
er
s
e
l
y
,
i
f
t
he
i
m
age
ha
s
a
hi
gh
c
on
t
r
as
t
r
ang
e
of
gr
a
y
l
e
v
e
l
s
di
s
t
r
i
b
ut
e
d
o
v
er
w
i
de
[
27]
.
C
ont
r
as
t
s
t
r
et
c
h
i
ng
pr
oc
es
s
c
oul
d
b
e
do
ne
b
y
m
ul
t
i
p
l
y
i
n
g
a
c
o
ns
t
ant
i
m
age and
i
t
c
o
ul
d
be
w
r
i
t
t
e
n i
n
E
q
uat
i
o
n 2.
(
,
)
=
(
,
)
(
2)
2
.
4
.
B
r
i
g
h
tn
e
s
s
a
n
d
C
o
n
tr
a
s
t C
o
m
b
i
n
a
ti
o
n
O
per
at
i
ons
of
br
i
g
ht
n
es
s
i
nc
r
eas
i
n
g an
d c
ont
r
as
t
s
t
r
et
c
hi
n
g c
ou
l
d b
e
don
e
s
i
m
ul
t
aneous
l
y
w
i
t
h t
he a
i
m
t
o i
m
pr
ov
e i
m
age qual
i
t
y
.
I
n g
ener
al
,
a c
om
bi
nat
i
on of
t
he t
w
o
oper
at
i
o
ns
c
an b
e
w
r
i
t
t
en
a
s
des
c
i
bed
i
n
E
quat
i
o
n 3
[
2
7]
.
(
,
)
=
(
,
)
+
(
3)
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
K
A
I
S
S
N
:
1
693
-
6
930
B
r
i
g
ht
nes
s
a
nd C
ont
r
as
t
M
odi
f
i
c
at
i
on
i
n
U
l
t
r
as
o
nogr
ap
hy
I
ma
ge
s U
si
n
g
…
(
R
e
t
no S
upr
i
y
ant
i
)
1093
2
.
5
.
D
e
v
i
a
ti
o
n
S
ta
n
d
a
r
d
S
t
an
dar
d de
v
i
at
i
on i
s
a v
ar
i
at
i
o
n of
dat
a di
s
t
r
i
b
ut
i
on of
al
l
t
h
e dat
a
.
T
he s
m
al
l
er
v
al
ue of
s
pr
eadi
ng
m
eans
f
e
w
er
v
ar
i
at
i
ons
i
n
dat
a
v
a
l
ues
.
I
f
s
p
r
eadi
ng
i
s
0
,
t
hen
t
he
v
al
u
e
of
a
l
l
dat
a
i
s
t
he
s
am
e.
T
he
gr
eat
er
v
al
u
e
of
t
h
e
d
at
a
s
pr
ead
i
ng
m
eans
i
nc
r
eas
i
ng
l
y
v
a
r
i
ed.
S
t
and
ar
d
dev
i
at
i
o
n c
an
be c
a
l
c
ul
at
ed
us
i
ng
t
he
f
ol
l
o
w
i
n
g f
or
m
ul
a
[
1]
as
des
c
r
i
be
d i
n
E
qua
t
i
on 4.
(
4)
W
he
r
e
in
:
σ
=
d
ev
i
at
i
o
n s
t
an
dar
d,
=
D
at
a
t
o I
,
x
’
=
dat
a a
v
er
a
ge,
n=
num
ber
of
dat
a
2
.
6
.
R
o
o
t M
e
a
n
S
q
u
a
r
e
E
r
r
o
r
(R
M
S
E
)
O
bs
er
v
at
i
on of
c
hanges
i
n i
m
age af
t
er
i
m
age pr
oc
es
s
i
ng c
an be do
ne b
y
l
ook
i
ng
di
r
ec
t
l
y
o
n t
he
i
m
age.
H
o
w
ev
er
,
t
o
m
eas
ur
e
quant
i
t
a
ti
v
e
l
y
,
it
c
an b
e do
ne b
y
c
al
c
ul
a
t
i
n
g t
he
v
a
l
ue
of
R
M
S
E
(
R
oot
Me
an
S
q
uar
e
E
r
r
or
)
.
R
M
S
E
i
s
t
h
e r
oot
of
t
he
M
S
E
(
Mea
n
S
quar
e
d E
r
r
or
)
.
RM
S
E
c
an
be c
a
l
c
ul
at
e
d u
s
i
ng t
he
E
q
u
at
i
on 5
[
1]
.
=
√
(
5)
MS
E
v
a
l
u
e c
al
c
u
l
at
i
on
i
s
t
he a
v
er
ag
e s
quar
e
d er
r
or
bet
w
e
en t
h
e or
i
gi
n
al
i
m
ages
w
i
t
h
t
he
i
m
age pr
oc
es
s
i
ng r
es
u
l
t
s
.
MS
E
c
an
be c
a
l
c
ul
at
ed
u
s
i
ng t
he
E
qu
at
i
on 6.
=
1
[
(
,
)
−
′
(
,
)
]
2
(
6)
W
h
er
ei
n
:
X
=
i
m
age w
i
dt
h
(
pi
x
el
)
,
Y=
c
i
t
r
a
he
i
g
ht
(
pi
x
el
),
I
=
i
m
age
p
ix
e
l
v
al
ue
bef
or
e noi
s
e
r
educ
t
i
o
n,
I
’=
i
m
age
p
i
x
el
v
al
u
e af
t
er
no
i
s
e r
ed
uc
t
i
on.
3.
R
e
su
l
t
s
a
n
d
A
n
a
l
y
s
i
s
3.
1.
C
o
n
v
e
r
ti
n
g
I
m
ag
e t
o
G
r
a
y
s
cal
e
T
he
f
i
r
s
t
s
t
ep i
n t
hi
s
s
y
s
t
e
m
i
s
t
o c
onv
er
t
i
np
ut
i
m
ages
w
hi
c
h
i
s
i
n R
G
B
(
R
ed
-
G
r
een
-
B
l
u
e)
t
o gr
a
y
s
c
al
e t
h
at
h
as
onl
y
o
ne
v
al
ue t
hat
i
s
gr
a
y
.
T
hi
s
pr
oc
es
s
ai
m
s
t
o s
i
m
pl
i
f
y
t
he
ul
t
r
as
on
ogr
a
ph
y
i
m
age
t
hat
i
s
s
t
i
l
l
i
n
t
h
e
f
or
m
o
f
R
G
B
w
hi
c
h
has
t
hr
e
e
c
ons
t
i
t
u
ent
c
o
m
ponent
s
,
nam
el
y
R
,
G
,
a
nd
B
,
c
on
v
er
t
ed
i
nt
o
gr
a
y
s
c
al
e
f
or
m
w
hi
c
h
has
onl
y
one
c
om
ponent
of
w
hi
c
h
i
s
gr
a
y
.
I
n t
h
e gr
a
y
s
c
a
l
e i
m
age hand
l
e on
l
y
s
h
ades
of
bl
a
c
k
and w
h
i
t
e pr
od
uc
es
gr
a
y
ef
f
ec
t
.
F
i
gur
e
3 s
ho
w
s
a
n ex
am
pl
e of
c
on
v
er
s
i
o
n f
r
o
m
R
G
B
t
o gr
a
y
s
c
a
le
im
a
ge i
n
our
ex
p
er
i
m
ent
.
G
r
a
y
s
c
a
l
e
i
m
age
i
s
ob
t
ai
ne
d
b
y
c
al
c
ul
a
t
i
n
g
t
he
av
er
ag
e
v
a
l
ue
of
t
he
c
ol
or
c
om
ponent
s
R
,
G
,
B
.
T
he c
al
c
ul
at
i
on
pr
oc
es
s
c
ar
r
i
ed ou
t
on
t
he
en
t
i
r
e p
i
x
el
i
m
age
.
F
i
gur
e
3.
R
G
B
t
o G
r
a
y
s
c
al
e C
on
v
er
s
i
on
F
i
gur
e 4.
A
n
E
x
am
pl
e o
f
C
r
opp
i
ng P
r
oc
es
s
U
s
i
ng T
em
pl
at
e M
at
c
hi
ng
Met
h
od
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1
6
9
3
-
6
930
T
E
L
KO
M
NI
K
A
V
o
l.
14
,
N
o
.
3,
S
ept
em
ber
2016
:
10
90
–
1
098
1094
3.
2.
D
et
er
m
i
n
e
U
te
r
u
s
A
r
e
a
T
o det
er
m
i
ne ut
er
us
ar
e
a,
w
e ap
pl
i
ed
t
em
pl
at
e m
at
c
hi
ng m
et
hod.
T
em
pl
at
e m
at
c
hi
n
g i
s
us
ed t
o f
i
n
d t
h
e ex
i
s
t
e
nc
e
of
a des
i
r
ed
obj
ec
t
i
n an
i
m
age b
y
us
i
n
g t
h
e t
em
pl
at
e i
m
age as
a
r
ef
er
e
nc
e i
m
age
[
27]
.
S
el
e
c
t
i
on of
i
m
age t
e
m
pl
at
es
t
o
det
er
m
i
ne des
i
r
ed
ar
e
a
is
v
er
y
i
m
por
t
ant
i
n
t
he
m
et
hod
of
t
em
pl
at
e
m
at
c
hi
ng.
Mor
e
pr
ec
i
s
e
s
el
ec
t
i
on
of
t
em
pl
at
e
i
m
age
is
h
i
gh
er
pr
es
ent
at
i
o
n s
uc
c
es
s
t
em
p
l
at
e m
at
c
hi
n
g m
et
hod.
I
f
t
he s
el
ec
t
i
o
n i
s
n
ot
a
ppr
op
r
i
at
e t
em
pl
at
e
t
hen
t
he
det
er
m
i
nat
i
o
n of
t
h
e ut
er
us
ar
ea
w
i
l
l
f
ai
l
.
S
el
e
c
t
i
on
of
t
em
pl
at
e c
ou
l
d
be
done
bas
e
d o
n
s
hape
and
i
m
age
d
egr
ad
at
i
on.
B
ot
h
v
ar
i
ab
l
es
c
ou
l
d
b
e e
v
al
ut
e
d b
y
i
m
age hi
s
t
o
gr
am
[
28]
.
An
ex
am
pl
e of
det
er
m
i
ni
ng
ut
e
r
us
ar
ea i
n o
ur
ex
per
i
m
ent
i
s
s
how
n i
n F
i
gur
e
4.
3.
3.
D
ev
i
at
i
o
n
S
ta
n
d
a
r
d
T
he s
t
andar
d de
v
i
at
i
on i
s
u
s
ed t
o c
al
c
ul
at
e
v
ar
i
at
i
ons
i
n t
he d
i
s
t
r
i
but
i
o
n of
c
ol
or
i
nt
ens
i
t
y
on an ul
t
r
as
ono
gr
aph
y
i
m
age.
T
abl
e 1 des
c
r
i
bes
s
om
e ex
am
pl
es
of
dev
i
a
t
i
o
n
s
t
andar
d
of
our
gr
a
y
s
c
al
e i
m
age a
nd t
em
pl
at
e
i
m
age c
andi
dat
e.
T
abl
e 1.
V
a
l
u
es
of
I
m
age S
t
an
dar
d
D
e
v
i
a
t
i
o
n
N
o
I
m
age i
den
t
i
t
y
S
t
andar
d
D
ev
i
at
i
on V
al
ue
G
r
ay
s
c
al
e i
m
age
t
e
m
pl
at
e
i
m
age
c
andi
da
t
e
1
13
67,
36
56,
87
2
14
64,
69
54,
41
3
15
69,
25
56,
71
4
16
69,
13
56
5
23
67,
28
57,
65
6
26
67,
28
46,
55
7
34
64,
36
56,
47
8
35
65,
56
56,
16
9
52
67,
27
62,
09
10
53
70,
27
61,
48
11
57
59,
36
63,
94
12
58
62,
35
66,
73
A
v
er
age
66,
18
57,
92
A
c
c
or
di
n
g t
o T
abl
e 1,
s
e
l
ec
t
i
on
of
t
em
pl
at
e i
m
age i
s
bas
ed o
n a
v
er
ag
e o
f
s
t
andar
d
dev
i
at
i
o
n v
a
l
ue
bet
w
een g
r
a
y
s
c
a
l
e i
m
age and c
an
d
i
dat
e t
em
pl
at
e i
m
age.
A
c
c
or
di
n
g t
o t
he
s
t
andar
d
de
v
i
at
i
on
v
a
l
ues
,
t
he m
os
t
qual
i
f
i
ed
i
m
age i
s
i
m
age w
i
t
h
i
de
nt
i
t
y
23,
26,
5
2
an
d
35
.
H
o
w
e
v
er
,
s
uc
c
es
s
per
c
ent
age
as
a
t
em
pl
at
e
i
s
t
h
e
bes
t
w
a
y
t
o
de
t
er
m
i
ne t
h
e b
es
t
t
em
pl
at
e.
I
f
t
he
t
em
pl
at
e
i
m
age
c
andi
dat
es
h
av
e
hi
gher
per
c
e
nt
ag
e
t
h
an
ot
h
er
,
i
t
w
i
l
l
be
c
hoos
en
as
a t
em
pl
at
e i
m
age.
T
abl
e
2 d
es
c
r
i
bed
per
c
ent
a
ge of
s
uc
c
es
s
t
em
pl
at
e
.
A
c
c
or
di
n
g t
o T
abl
e
2,
i
m
age23
has
hi
gher
p
er
c
ent
a
ge t
ha
n ot
her
,
i
t
i
s
a
bo
ut
83.
3%
.
T
her
ef
or
e f
or
t
he nex
t
pr
oc
es
s
,
w
e us
ed
i
m
age23
as
a
n i
m
age t
em
pl
at
e
i
n
our
ex
per
i
m
ent
.
3
.
4
.
B
r
i
g
h
tn
e
s
s
a
n
d
C
o
n
tr
a
s
t M
o
d
i
fi
c
a
ti
o
n
T
hi
s
pr
oc
es
s
ai
m
s
t
o
i
m
pr
ov
e
our
i
m
ages
qua
l
i
t
y
b
y
m
odi
f
y
i
ng
br
i
ght
nes
s
an
d
c
o
nt
r
as
t
.
Mat
h
em
at
i
c
al
l
y
,
m
odi
f
i
y
i
ng
pr
oc
es
s
is
b
y
ap
pl
y
i
ng e
quat
i
on
3.
B
r
i
ght
nes
s
c
on
s
t
ant
v
al
u
e
w
i
l
l
i
nc
r
eas
e c
o
l
or
i
nt
e
ns
i
t
y
v
a
l
ues
f
or
eac
h p
i
x
el
i
n
ac
c
or
danc
e t
o t
h
e g
i
v
en c
o
ns
t
a
nt
v
al
ue.
T
hi
s
al
s
o
a
pp
l
i
es
t
o
t
he
i
m
age
m
ul
t
i
pl
i
c
at
i
on
aga
i
ns
t
t
o
c
ont
r
as
t
c
ons
t
a
nt
v
a
l
ue.
T
her
ef
or
e,
b
y
m
odi
f
y
i
n
g c
ont
r
as
t
and
br
i
g
ht
nes
s
w
i
l
l
l
e
ad t
o i
nc
r
eas
e
c
ol
or
i
nt
e
ns
i
t
y
v
al
ues
f
or
al
l
pi
x
e
l
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
K
A
I
S
S
N
:
1
693
-
6
930
B
r
i
g
ht
nes
s
a
nd C
ont
r
as
t
M
odi
f
i
c
at
i
on
i
n
U
l
t
r
as
o
nogr
ap
hy
I
ma
ge
s U
si
n
g
…
(
R
e
t
no S
upr
i
y
ant
i
)
1095
T
abl
e 2.
P
er
c
ent
ag
e of
U
t
e
r
us
D
et
ec
t
i
on
NO
T
e
m
pl
at
e
N
um
ber
o
f
i
m
age
det
ec
t
ed
U
t
er
us
det
e
c
t
ed
(
i
m
age
i
dent
i
t
y
)
N
um
ber
of
f
ai
l
det
ec
t
ed
U
t
er
us
not
det
ec
t
ed
(
i
m
age
i
dent
i
t
y
)
S
u
c
c
e
ss
per
c
ent
age
(%
)
1
i
m
age
13
9
13,
14,
15,
16
,
23,
52,
53,
57,
58
3
26,
34,
35,
75
2
i
m
age
14
8
13,
14,
15,
16
,
23,
52,
53,
58
4
26,
34,
35,
57
66,
6
3
i
m
age
15
7
13,
14,
15,
16
,
52,
53,
57
5
23,
26,
34,
35,
58
58,
3
4
i
m
age
16
8
13,
14,
15,
16
,
52,
53,
57,
58
4
23,
26,
34,
35
66,
6
5
i
m
age
23
10
13,
15,
23,
26
,
34,
35,
52,
53,
57
,
58
2
14,
16
83,
3
6
i
m
age
26
6
23,
26,
34,
52
,
57,
58
6
13,
14,
15,
16,
35,
53
50
7
i
m
age
34
7
13,
34,
35,
52
,
5
3
,
57,
58
5
14,
15,
16 ,
23,
26,
58,
3
8
i
m
age
35
7
13,
15,
35,
52
,
53,
57,
58
5
14,
16,
23,
26,
34,
58,
3
9
i
m
age
52
2
52,
53
10
13,
14,
15,
16,
23,
26,
34,
35,
57,
58
16,
6
10
i
m
age
54
2
52,
53
10
13,
14,
15,
16,
23,
26,
34,
35,
57,
58
16,
6
11
i
m
age
57
5
13,
14,
23,
57
,
58
7
15,
16,
26,
34,
35,
52,
54
41,
6
12
i
m
age
58
5
13,
23,
34,
53
,
58
7
14,
15,
16,
26,
35,
52,
57
41,
6
T
abl
e
3.
S
t
and
ar
d D
e
v
i
at
i
o
n an
d
R
M
SE
V
al
u
es
U
s
i
n
g
C
ann
y
O
p
er
at
or
P
r
oc
es
s
D
ev
i
at
i
on
s
t
andar
d
R
M
SE
P
r
oc
es
s
D
ev
i
at
i
on
S
t
andar
d
R
M
SE
O
r
i
gi
nal
i
m
age
58,
8889
O
r
i
gi
nal
I
m
age
58,
8889
G
r
ay
s
c
al
e c
onv
er
s
i
on
59,
3695
G
r
ay
s
c
al
e c
onv
er
s
i
on
59,
3695
N
oi
s
e r
e
m
ov
i
ng
60,
207
N
oi
s
e r
e
m
ov
i
ng
60,
207
W
i
t
hout
m
odi
f
i
c
a
t
i
on
60,
207
0
M
odi
f
i
c
at
i
on
9,
6671
59,
9278
A
ppl
i
ed m
ed
i
an
f
i
l
t
er
60,
2872
4,
6661
A
ppl
i
ed
m
edi
an f
i
l
t
er
9,
6573
60,
0772
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0
,
01 0,
02
]
0,
1455
232,
7248
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0
,
01 0,
02
]
0,
0865
232,
7362
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0
,
02 0,
03
]
0
,
1394
232,
7259
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0
,
01 0,
02
]
]
0,
0836
232,
7364
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0
,
03 0,
04
]
0,
1322
232,
7274
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0
,
01 0,
02
]
0,
0814
232,
7365
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0
,
04 0,
05
]
0,
1246
23
2,
7287
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0
,
01 0,
02
]
0,
0788
232,
7367
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0
,
05 0,
06
]
0,
1178
232,
7298
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0
,
01 0,
02
]
0,
075
232,
7369
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1
6
9
3
-
6
930
T
E
L
KO
M
NI
K
A
V
o
l.
14
,
N
o
.
3,
S
ept
em
ber
2016
:
10
90
–
1
098
1096
T
abl
e
4.
S
t
and
ar
d D
e
v
i
at
i
o
n an
d
R
M
SE
V
al
u
es
U
s
i
n
g
S
ob
el
O
p
e
r
at
or
P
r
oc
es
s
D
ev
i
at
i
on
s
t
andar
d
R
M
SE
P
r
oc
es
s
D
ev
i
at
i
on
S
t
andar
d
R
M
SE
O
r
i
gi
nal
i
m
age
58,
8889
O
r
i
gi
nal
I
m
age
58,
8889
G
r
ay
s
c
al
e c
onv
er
s
i
on
59,
3695
G
r
ay
s
c
al
e c
onv
er
s
i
on
59,
3695
N
oi
s
e r
e
m
ov
i
ng
60,
207
N
oi
s
e r
e
m
ov
i
ng
60,
207
W
i
t
hout
m
odi
f
i
c
a
t
i
on
60,
207
0
M
odi
f
i
c
at
i
on
9,
6671
59,
9278
A
ppl
i
ed m
ed
i
an
f
i
l
t
er
60,
2872
4,
6661
A
ppl
i
ed
m
edi
an f
i
l
t
er
9,
6573
60,
0772
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[0
,0
1
]
0,
1554
232,
723
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0
,
01]
0,
0898
232,
736
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0,
0
2
]
0,
1234
232,
729
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0
,
02
]
0,
0764
232,
7368
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[0
,0
3
]
0,
1051
232,
731
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0,
03]
0,
0675
232,
7373
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[0
,0
4
]
0,
0945
232,
733
E
dge de
t
ec
t
i
o
n
u
s
i
n
g
t
hr
es
hol
d
[0
,0
4]
0,
0526
232,
7379
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[0
,0
5
]
0,
0881
232,
734
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0
,
05
]
0,
0317
232,
7386
T
abl
e
5
.
S
t
and
ar
d
D
e
v
ia
t
io
n
an
d
R
M
SE
V
al
u
es
U
s
i
n
g
P
r
e
w
i
t
t
O
per
a
t
or
P
r
oc
es
s
D
ev
i
at
i
on
s
t
a
ndar
d
R
M
SE
P
r
oc
es
s
D
ev
i
at
i
on
S
t
andar
d
O
r
i
gi
nal
i
m
age
58,
8889
O
r
i
gi
nal
I
m
age
58,
8889
R
MS
E
G
r
ay
s
c
al
e c
onv
er
s
i
on
59,
3695
G
r
ay
s
c
al
e c
onv
er
s
i
on
59,
3695
N
oi
s
e r
e
m
ov
i
ng
60,
207
N
oi
s
e r
e
m
ov
i
ng
60,
207
W
i
t
hout
m
odi
f
i
c
a
t
i
on
60,
207
0
M
odi
f
i
c
at
i
on
9,
6671
59
,
9278
A
ppl
i
ed m
ed
i
an
f
i
l
t
er
60,
2872
4,
6661
A
ppl
i
ed
m
edi
an f
i
l
t
er
9,
6573
60,
0772
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[0
,0
1
]
0,
1543
232,
7
2
3
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0
,
01]
0,
0896
232,
736
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[0
,0
2
]
0,
1219
232,
7
29
E
dge
det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0
,
02
]
0,
0757
232,
7369
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[0
,0
3
]
0,
1038
232,
7
23
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0,
03]
0,
0673
232,
7373
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[0
,0
4
]
0,
0935
232,
73
3
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0,
0
4]
0,
0488
232,
7381
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[0
,0
5
]
0,
0874
232,
73
4
E
dge det
e
c
t
i
on
us
i
ng
t
hr
es
hol
d
[
0
,
05
]
0,
0309
232,
7386
3
.
5
.
E
d
g
e
D
e
te
c
ti
o
n
a
n
d
C
a
l
c
u
l
a
ti
n
g
O
f S
t
a
n
d
a
r
d
D
e
v
i
a
ti
o
n
a
n
d
R
M
S
E
T
hi
s
pr
oc
es
s
ai
m
s
t
o
get
t
h
e
edg
e
s
ha
pe
of
ut
er
us
.
W
e
di
d
t
h
i
s
ex
per
i
m
ent
b
y
a
p
pl
y
i
n
g
s
ev
er
a
l
ed
ge
det
ec
t
i
o
n
op
e
r
at
or
s
s
uc
h as
c
ann
y
,
s
o
bel
,
and
P
r
e
w
i
t
t
.
T
ab
l
e 3
t
o T
abl
e
5 s
ho
w
an
ex
am
pl
e
r
es
ul
t
of
eac
h
m
et
hod
i
nc
l
ud
i
ng
de
v
i
at
i
on
s
t
a
ndar
d
and
R
M
S
E
v
al
ues
,
i
n
her
e
w
e
us
e
i
m
age57 as
a
c
as
e.
A
c
c
or
di
n
g T
abl
e
3
t
o
T
abl
e
5 f
or
a
l
l
ed
ge
det
ec
t
i
o
n
op
er
at
or
s
,
s
t
an
dar
d
de
v
i
at
i
on
v
a
l
ues
f
or
bot
h
c
on
di
t
i
o
ns
w
i
t
hou
t
m
odi
f
y
i
n
g
a
nd
w
i
t
h
m
odi
f
y
i
ng
ar
e
a
l
m
os
t
t
he
s
am
e,
t
he
di
f
f
er
enc
e
i
s
t
oo s
m
al
l
.
H
o
w
e
v
er
,
de
v
i
a
t
i
on s
t
a
ndar
d
v
a
l
ues
f
or
edge d
et
ec
t
i
on r
es
u
l
t
w
h
e
n t
her
e i
s
no
m
odi
f
i
c
at
i
on
i
s
l
ar
ger
t
h
an
i
n
m
odi
f
i
c
at
i
o
n
c
ondi
t
i
o
n.
I
t
i
s
c
aus
ed
t
hi
s
i
m
age
has
been
i
m
pai
r
ed
s
t
andar
d
de
v
i
at
i
on
v
a
l
ue
du
e t
o t
he
br
i
gh
t
nes
s
a
nd c
on
t
r
as
t
m
odi
f
i
c
at
i
on.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
K
A
I
S
S
N
:
1
693
-
6
930
B
r
i
g
ht
nes
s
a
nd C
ont
r
as
t
M
odi
f
i
c
at
i
on
i
n
U
l
t
r
as
o
nogr
ap
hy
I
ma
ge
s U
si
n
g
…
(
R
e
t
no S
upr
i
y
ant
i
)
1097
4
.
C
o
n
c
l
u
s
i
o
n
A
c
c
or
di
n
g t
o t
h
e r
es
ul
t
s
a
bov
e,
w
e c
onc
l
ud
e:
(
i
)
ed
ge det
ec
t
i
o
n i
n u
l
t
r
as
o
nog
r
aph
y
i
m
ages
w
h
i
c
h ha
v
e
bee
n m
odi
f
i
e
d br
i
ght
n
es
s
and c
on
t
r
as
t
pr
ov
i
di
ng g
ood s
hap
e
ut
er
us
an
d j
us
t
hav
i
n
g a b
i
t
no
i
s
e.
T
hat
m
eans
br
i
ght
nes
s
an
d c
ont
r
as
t
m
odi
f
i
c
at
i
on c
ou
l
d
i
m
pr
ov
e i
m
age
qua
l
i
t
y
.
(
i
i
)
B
r
i
gh
t
nes
s
and
c
ont
r
as
t
m
odi
f
i
c
at
i
on
c
aus
e
s
t
a
ndar
de
v
i
a
t
i
on
v
al
u
e
de
c
r
eas
ed.
T
he
hi
g
her
v
a
l
ue
s
et
t
i
ng,
t
he
s
m
al
l
er
d
ev
i
at
i
o
n
v
al
ue
g
ener
at
ed.
(
i
i
i
)
D
ur
i
ng
i
m
age pr
oc
es
s
i
ng,
s
t
andar
d
de
v
i
a
t
i
o
n v
al
ue i
s
i
nv
er
s
e
l
y
pr
opor
t
i
o
nal
t
o
R
MS
E
v
al
u
e.
(
i
v
)
G
ener
al
l
y
,
a
l
l
edg
e
det
ec
t
i
on o
per
at
or
abl
e t
o
pr
oduc
e go
od e
dge d
et
e
c
t
i
on i
f
us
i
n
g ac
c
ur
at
e t
hr
es
hol
d.
I
n o
ur
ex
per
i
m
ent
,
c
ann
y
op
er
at
o
r
pr
ov
i
de t
h
e bes
t
r
es
u
l
t
w
hen us
i
ng t
hr
es
ho
l
d [
0.
02
0.
03]
w
hi
l
e f
or
s
obel
and
pr
e
w
i
t
t
op
er
at
or
,
t
he b
es
t
r
es
ul
t
i
s
obt
a
i
n
ed
w
h
en us
i
ng t
hr
es
ho
l
d [
0.
00
2]
.
H
i
gh
er
t
hr
es
hol
d
pr
o
v
i
de
d
e
dge de
t
ec
t
i
o
n
r
es
ul
t
obt
ai
ned
w
i
l
l
get
t
i
ng w
or
s
t
.
A
c
k
n
o
w
l
e
d
g
e
m
en
t
s
T
hi
s
w
or
k
i
s
s
uppor
t
ed b
y
D
i
r
ec
t
or
a
t
e G
en
er
al
of
H
i
gh
er
E
d
uc
at
i
on t
hr
o
ugh
H
i
bah
St
r
a
t
e
g
i
s
N
a
s
i
o
n
a
l
(
ST
R
AN
AS)
.
R
ef
er
en
ces
[1
]
R
Mu
n
i
r,
P
engol
a
han C
i
t
r
a D
i
gi
t
a
l
deng
an P
end
ek
a
t
an
A
l
gor
i
t
m
i
k
.
1s
t
e
d.
B
and
ung,
I
ndone
s
i
a
:
P
ener
bi
t
I
nf
or
m
at
i
k
a.
2
004.
[2
]
C
B
N
auer
,
C
Z
ub
l
er
,
C
W
e
i
s
s
t
an
ner
,
C
S
t
i
e
ger
,
P
S
enn
,
A
A
r
nol
d.
R
ad
i
at
i
on
do
s
e
o
pt
i
m
i
z
at
i
o
n
i
n
pedi
a
t
r
i
c
t
em
por
a
l
bon
e
c
o
m
put
ed
t
om
o
gr
ap
h
y
:
I
nf
l
u
enc
e
of
t
ube
t
ens
i
on
on
i
m
age c
o
nt
r
as
t
an
d
i
m
ag
e qu
al
i
t
y
.
N
eur
or
a
di
o
l
og
y
.
2012;
53
(
3
):
2
47
-
25
4.
[3
]
M
J
M
as
s
ey
,
E
LaR
oc
hel
l
e,
G
N
aj
ar
r
o,
A
K
ar
m
a
c
har
l
a,
N
I
A
r
nol
d,
R
T
r
z
ec
i
ak
,
S
S
hapi
r
o
.
T
he
m
i
c
r
oc
i
r
c
u
l
at
i
on i
m
a
ge qu
al
i
t
y
s
c
or
e:
D
ev
el
op
m
en
t
and
pr
el
i
m
i
n
ar
y
ev
al
uat
i
on
of
a
pr
opo
s
e
d
appr
oa
c
h
t
o
gr
adi
n
g
qua
l
i
t
y
of
i
m
age
a
c
qu
i
s
i
t
i
o
n
f
or
bed
s
i
d
e
v
i
deom
i
c
r
o
s
c
opy
.
J
.
C
r
i
t.
C
a
r
e
.
2
013;
28
(
6
)
:
913
-
9
17.
[4
]
P
Ca
s
e
J
A,
M
D
B
at
em
an
T
M.
T
ak
i
ng
t
he
per
f
ec
t
n
uc
l
ea
r
i
m
ag
e:
Q
ual
i
t
y
c
ont
r
o
l
,
ac
q
ui
s
i
t
i
o
n,
an
d
pr
oc
e
s
s
i
ng
t
ec
h
ni
q
ues
f
or
c
ar
di
ac
S
P
E
C
T
,
P
E
T
,
and hy
br
i
d i
m
a
gi
n
g.
J
.
N
u
c
l
.
C
ar
di
ol
.
2
013;
20
(
5
):
891
-
90
7.
[5
]
H
Lups
o
r
-
P
l
at
on,
M
S
t
ef
ane
s
c
u,
D
M
ur
es
an,
M
F
l
or
ea,
M
E
S
zá
s
z
,
A
M
ani
u,
R
B
ad
ea.
N
oni
nv
as
i
v
e
as
s
es
s
m
e
nt
of
l
i
v
er
s
t
ea
t
os
i
s
u
s
i
ng
ul
t
r
as
o
und m
et
hod
s
.
M
e
d
. U
l
tr
as
on.
2014
;
16
(
3
):
23
6
-
24
5.
[6
]
G
S
er
bes
,
N
A
y
di
n
.
D
e
noi
s
i
n
g per
f
o
r
m
a
nc
e
of
m
od
i
f
i
e
d d
ual
-
t
r
e
e c
om
p
l
ex
w
a
v
el
et
t
r
a
ns
f
or
m
f
o
r
pr
oc
e
s
s
i
ng qua
d
r
at
ur
e e
m
bo
l
i
c
do
ppl
er
s
i
gna
l
s
.
M
ed.
B
i
o
l
.
E
ng.
C
om
put
.
201
4;
52
(
1
):
29
-
43
.
[7
]
M
C
i
ec
hol
ew
s
k
i
,
J
C
ho
c
ho
l
ow
i
c
z
.
Q
uad
r
at
ur
e s
i
gn
al
s
.
C
om
p
ut
.
B
i
o
l
.
M
ed.
201
3;
43
(
12
):
22
38
-
235
5
.
[8
]
P
M
ar
t
i
nez
-
S
anc
h
ez
,
A
V
A
l
ex
andr
ov
.
U
l
t
r
a
s
on
ogr
ap
hy
of
c
a
r
ot
i
d
pl
aqu
e
f
or
t
he
pr
ev
en
t
i
o
n
of
s
t
r
ok
e
.
E
x
per
t
R
ev
.
C
ar
di
o
v
a
s
c
.
T
her
.
2013
;
11
(
10
)
:
1
425
–
144
0
.
[9
]
R
C
hi
f
or
,
M
E
B
adea,
D
M
i
t
r
ea,
I
C
B
a
dea
,
M
C
r
i
s
an,
I
C
hi
f
or
,
R
Av
ra
m
.
C
om
p
ut
e
r
-
a
s
s
i
s
t
ed
i
dent
i
f
i
c
at
i
on
of
t
he
gi
n
gi
v
al
s
ul
c
u
s
and
per
i
odo
nt
al
ep
i
t
he
l
i
al
j
unc
t
i
on
on
hi
gh
-
f
r
eque
nc
y
ul
t
r
a
s
ou
n
d
i
m
ag
es
.
M
ed.
U
l
t
r
as
o
n.
2
015
;
17
(
3
):
273
–
279.
[
10]
D
V
at
an
s
ev
er
,
V
K
y
r
i
ak
o
po
ul
ou,
J
M
A
lls
o
p
,
M
F
o
x,
A
C
hew
,
J
V
H
aj
nal
,
M
A
R
ut
her
f
or
d
.
M
ul
t
i
di
m
en
s
i
o
nal
a
nal
y
s
i
s
of
f
e
t
al
pos
t
er
i
o
r
f
os
s
a i
n hea
l
t
h an
d di
s
e
as
e.
T
he C
er
ebe
l
l
um
.
2
013
;
12
(
5
):
632
–
64
4
.
[
11]
M
A
nobet
t
i
,
A
C
oppa
,
F
B
ul
l
et
t
i
,
S
P
i
az
z
a,
P
N
az
er
i
an
,
A
C
on
t
i
,
R
P
i
ni
.
V
er
i
f
i
c
at
i
on
of
c
or
r
ec
t
c
ent
r
a
l
v
enous
c
a
t
het
er
pl
a
c
em
ent
i
n
t
he
em
er
gen
c
y
depar
t
m
ent
:
C
om
par
i
s
on
bet
w
een
ul
t
r
a
s
o
no
gr
aphy
an
d
c
he
s
t
r
ad
i
ogr
aphy
.
I
nt
er
n.
E
m
er
g.
M
ed.
2013
;
8
(
2
):
1
7
3
–
1
80
.
[
12]
H
T
anak
a,
H
I
i
j
i
m
a,
A
H
i
ga
s
hi
ur
a,
K
Y
oh,
A
I
s
hi
i
,
T
T
ak
as
h
i
m
a,
S
N
i
s
h
i
g
uc
hi
.
N
ew
m
al
i
gn
ant
gr
adi
n
g
s
y
s
t
em
f
or
hepa
t
oc
el
l
ul
ar
c
a
r
c
i
no
m
a u
s
i
n
g t
he s
onaz
oi
d
c
ont
r
a
s
t
age
nt
f
or
ul
t
r
as
on
ogr
aphy
.
J
.
G
as
t
r
oen
t
er
o
l
.
20
14
;
49
(
4
):
7
55
–
763
.
[
13]
A
T
C
hi
em
,
C
H
C
han,
D
Y
I
br
ahi
m
,
C
L A
nder
s
o
n,
D
S
W
u,
C
J
G
ila
n
i,
JC
F
o
x
.
P
el
v
i
c
ul
t
r
as
onogr
a
phy
and l
engt
h of
s
t
ay
i
n t
h
e
E
D
:
A
n ob
s
er
v
at
i
o
nal
s
t
u
dy
.
A
m
.
J
.
E
m
er
g.
M
ed.
2014
;
32
(
12
):
14
64
–
14
69.
[
14]
W
W
a
n
g,
L
C
hen
,
M
Lu,
G
L
i
u,
S
S
hen,
Z
Xu
,
L
Z
hou
.
C
ont
r
a
s
t
-
en
han
c
ed
ul
t
r
as
oun
d
f
eat
ur
e
s
o
f
hi
s
t
ol
og
i
c
al
l
y
pr
ov
en f
oc
a
l
no
dul
ar
hy
per
pl
a
s
i
a
:
D
i
agn
os
t
i
c
per
f
or
m
anc
e c
o
m
par
ed w
i
t
h
c
ont
r
a
s
t
-
enhan
c
ed
C
T
.
E
ur
.
R
adi
ol
.
20
13
;
23
(
9
):
2
546
–
255
4.
[
15]
A
H
i
z
uk
ur
i
,
R
N
a
k
ay
am
a
,
Y
K
as
hi
k
ur
a
,
H
T
ak
as
e
,
H
K
aw
anak
a,
T
O
gaw
a,
S
T
s
ur
uok
a
.
C
om
put
er
i
z
ed
det
er
m
i
nat
i
on
s
c
h
em
e
f
or
hi
s
t
o
l
og
i
c
a
l
c
l
a
s
s
i
f
i
c
at
i
on
of
br
e
as
t
m
as
s
us
i
ng
obj
ec
t
i
v
e
f
eat
ur
e
s
c
or
r
es
p
ond
i
ng
t
o
c
l
i
ni
c
i
an
s
’
s
ub
j
e
c
t
i
v
e
i
m
pr
es
s
i
o
n
s
on
ul
t
r
a
s
on
ogr
ap
hi
c
i
m
a
ge
s
.
J
.
D
i
g
i
t.
I
m
agi
ng
.
201
3
;
26
(
5
):
95
8
–
97
0.
[1
6]
J
W
ang
, D
Z
han
g
.
I
m
age
D
enoi
s
i
ng B
as
e
d on A
r
t
i
f
i
c
i
a
l
B
ee C
ol
ony
and B
P
N
eur
al
N
et
w
or
k
.
T
EL
KO
M
N
I
KA
T
el
ec
om
m
uni
c
at
i
on
C
om
put
.
E
l
ec
t
r
on.
C
o
n
t
ro
l
.
2
015
;
13
(
2
):
614
.
[
17]
X
Z
hang
,
S L
i
u
.
I
m
ag
e E
d
ge F
eat
ur
e E
x
t
r
ac
t
i
on
and
R
ef
i
ni
ng B
as
ed o
n G
enet
i
c
-
A
nt
C
ol
on
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1
6
9
3
-
6
930
T
E
L
KO
M
NI
K
A
V
o
l.
14
,
N
o
.
3,
S
ept
em
ber
2016
:
10
90
–
1
098
1098
A
l
gor
i
t
hm
.
T
EL
KO
M
N
I
KA
T
el
ec
om
m
uni
c
at
i
o
n C
om
put
.
E
l
e
c
t
r
on.
C
ont
r
ol
.
201
5
;
13
(
1
)
:
11
8
.
[
18]
W
J
i
e,
F
Z
ur
e
n,
W
Lei
,
Z
F
eng
,
L
W
an
g
.
H
i
gh R
e
c
og
n
i
t
i
on R
at
i
o I
m
age P
r
o
c
e
s
s
i
ng A
l
gor
i
t
h
m
of
M
i
c
r
o
E
l
ec
t
r
i
c
a
l
C
o
m
po
nent
s
i
n O
pt
i
c
al
M
i
c
r
o
s
c
o
pe
.
T
EL
KO
M
N
I
K
A
T
el
e
c
om
m
uni
c
at
i
on C
om
p
ut
.
E
l
e
c
t
r
on
.
C
ont
r
ol
.
201
4
;
12
(
4
):
911.
[
19]
R
S
upr
i
y
ant
i
,
H
H
abe,
M
K
i
dode,
S
N
agat
a
.
A
s
i
m
pl
e
and
r
obus
t
m
et
hod
t
o
s
c
r
ee
n
c
at
ar
a
c
t
s
us
ing
s
pe
c
ul
ar
r
ef
l
ec
t
i
on app
ear
an
c
e
.
P
r
oc
.
S
P
I
E
6
915 M
edi
c
a
l
I
m
agi
ng
.
20
08.
[
20]
R
S
upr
i
y
ant
i
,
H
H
abe,
M
K
i
dode,
S
N
agat
a
.
C
om
pac
t
C
at
ar
a
c
t
S
c
r
ee
ni
n
g S
y
s
t
em
:
D
es
i
g
n an
d
P
r
ac
t
i
c
al
D
at
a
A
c
qui
s
i
t
i
on
.
I
nt
er
nat
i
ona
l
C
onf
er
en
c
e
on
I
ns
t
r
um
en
t
at
i
on,
C
o
m
m
uni
c
at
i
on,
I
nf
or
m
at
i
o
n
T
ec
hnol
o
gy
and B
i
o
m
edi
c
a
l
E
ngi
ne
er
i
n
g (
I
C
I
C
I
-
BM
E)
.
2009.
[
21]
R
S
upr
i
y
ant
i
,
H
H
abe,
M
K
i
dode,
S
N
aga
t
a
.
E
x
t
r
a
c
t
i
ng
ap
pear
an
c
e
i
nf
or
m
at
i
o
n
i
ns
i
de
t
he
p
upi
l
f
o
r
c
at
ar
ac
t
s
c
r
e
eni
n
g
.
P
r
oc
eed
i
n
gs
of
t
h
e 11t
h I
A
P
R
C
onf
er
en
c
e on M
ac
hi
ne V
i
s
i
on
A
pp
l
i
c
at
i
ons
.
200
9
:
342
–
3
45.
[
22]
R
S
upr
i
y
ant
i
,
E
P
r
anat
a,
Y
R
am
ad
han
i
,
TI
R
os
ant
i
.
S
ep
ar
abi
l
i
t
y
f
i
l
t
er
f
or
l
o
c
a
l
i
z
i
ng
ab
no
r
m
al
pu
pi
l
:
I
dent
i
f
i
c
at
i
on of
i
n
put
i
m
a
ge
.
T
el
k
om
n
i
k
a
.
20
13
;
11
(
4
):
783
–
7
90.
[
23]
R
S
upr
i
y
ant
i
,
D
P
ut
r
i
,
E
M
ur
dy
ant
or
o,
HB
W
id
o
do
.
Co
m
par
i
ng edg
e det
e
c
t
i
on m
et
hod
s
t
o l
oc
a
l
i
z
e
ut
er
u
s
ar
ea o
n ul
t
r
a
s
o
und i
m
a
ge
.
P
r
oc
.
of
201
3 3r
d I
nt
.
C
on
f
.
on I
ns
t
r
um
ent
a
t
i
on
,
C
om
m
u
ni
c
a
t
i
on
s
,
I
nf
or
m
at
i
on T
ec
hno
l
.
,
an
d B
i
o
m
edi
c
a
l
E
ngi
neer
i
ng:
S
c
i
en
c
e
and T
ec
hno
l
.
f
or
I
m
pr
ov
e
m
ent
of
H
eal
t
h
,
S
af
et
y
,
an
d E
nv
i
r
o
n
., IC
IC
I
-
BM
E
.
2013
:
152
–
155.
[
24]
R
S
upr
i
y
ant
i
,
A
F
ar
i
z
,
T
S
ept
i
a
na,
E
M
ur
dy
ant
or
o,
Y
R
am
ad
h
ani
,
HB
W
id
o
do
.
A
S
i
m
pl
e
S
c
r
eeni
n
g
f
o
r
Hi
g
h
-
R
i
s
k
P
r
e
gna
nc
i
es
i
n R
ur
al
A
r
ea
s
B
a
s
ed E
x
per
t
S
y
s
t
e
m
.
T
el
k
om
ni
k
a
.
2
015
;
13
(
2
)
:
6
61
–
669
.
[
25]
H
B
W
i
dodo
,
A
S
oel
a
i
m
an,
Y
R
am
adh
ani
,
R
S
upr
i
y
ant
i
.
C
al
c
ul
a
t
i
ng
C
o
nt
r
a
s
t
S
t
r
et
c
h
i
ng
V
ar
i
ab
l
es
i
n
O
r
der
t
o I
m
pr
ov
e D
e
nt
al
R
a
di
ol
ogy
I
m
ag
e Q
ual
i
t
y
.
I
n
t
er
n
at
i
on
al
C
onf
er
en
c
e on E
n
gi
n
eer
i
ng
a
n
d
T
ec
hnol
o
gy
f
or
S
u
s
t
a
i
nab
l
e D
ev
el
opm
ent
(
I
C
E
T
4S
D
)
.
2015.
[
26]
R
S
upr
i
y
ant
i
,
U
E
r
f
ay
ant
o,
Y
R
a
m
adhan
i
,
E
M
ur
dy
ant
or
o,
H
B
W
ido
d
o.
B
l
o
od P
r
es
s
ur
e M
obi
l
e
M
oni
t
or
i
ng
f
or
P
r
e
gnan
t
W
om
an
B
as
e
d
A
ndr
oi
d
S
y
s
t
em
.
I
nt
er
nat
i
on
al
C
onf
er
en
c
e
o
n
E
ngi
ne
er
i
n
g
and T
ec
h
nol
o
gy
f
or
S
u
s
t
a
i
nab
l
e D
ev
el
op
m
en
t
(
I
C
E
T
4S
D
)
.
20
15.
[
27]
A
K
adi
r
,
A
S
oes
a
nt
o.
T
eor
i
d
an A
pl
i
k
as
i
P
e
ngo
l
ah
an C
i
t
r
a
. Y
ogy
ak
ar
t
a I
ndon
es
i
a:
P
en
er
bi
t
A
nd
i
.
2013.
[
28]
R
S
upr
i
y
ant
i
,
I
A
P
r
a
dan
a,
E
M
ur
dy
ant
or
o
,
HB
W
id
o
do
.
Lo
c
al
i
z
i
ng U
t
er
us
R
e
gi
on f
r
om
L
ow
-
R
e
s
o
lu
t
i
on
U
l
t
r
as
o
nogr
aphy
D
ev
i
c
e
U
s
i
ng
T
em
pl
at
e
M
at
c
hi
ng
M
et
hod
.
I
nt
.
J
.
B
i
o
s
c
i
.
B
i
oc
hem
.
B
i
oi
nf
or
m
a.
201
4
;
4
(
5
):
30
4
–
3
11.
Evaluation Warning : The document was created with Spire.PDF for Python.