I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
9
,
No
.
2
,
A
p
r
il
201
9
,
p
p
.
9
8
2
~9
9
1
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v9
i
2
.
pp
982
-
9
9
1
982
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e
.
co
m/
jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
I
JE
C
E
Fra
m
ew
o
rk
f
o
r p
ro
g
ress
iv
e seg
m
e
ntatio
n of c
hest
r
a
dio
g
ra
ph
for ef
ficien
t
dia
g
n
o
sis
of inert reg
io
ns
Sa
v
it
ha
S.
K
.
1
,
N.
C.
N
a
v
ee
n
2
1
De
p
a
rtme
n
t
o
f
CS
E,
V
isv
e
sv
a
ra
y
a
Tec
h
n
o
lo
g
ica
l
Un
iv
e
rsity
,
Be
l
a
g
a
v
i,
Ka
rn
a
tak
a
,
In
d
ia
2
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
&
En
g
in
e
e
rin
g
,
JSS
A
T
E,
Be
n
g
a
lu
ru
,
In
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
n
30
,
2
0
1
8
R
ev
i
s
ed
Oct
27
,
2
0
1
8
A
cc
ep
ted
No
v
20
,
2
0
1
8
S
e
g
m
e
n
tatio
n
is
o
n
e
o
f
th
e
m
o
st
e
ss
e
n
ti
a
l
ste
p
s
re
q
u
ired
t
o
id
e
n
t
if
y
th
e
in
e
rt
o
b
jec
t
i
n
t
h
e
c
h
e
st
x
-
ra
y
.
A
re
v
ie
w
w
it
h
th
e
e
x
isti
n
g
se
g
m
e
n
tatio
n
tec
h
n
iq
u
e
s
to
w
a
rd
s
c
h
e
st
x
-
r
a
y
a
s
we
ll
a
s
o
th
e
r
v
it
a
l
o
rg
a
n
s
w
a
s
p
e
r
f
o
r
m
e
d
.
T
h
e
m
a
in
o
b
jec
ti
v
e
wa
s
to
f
in
d
w
h
e
th
e
r
e
x
isti
n
g
s
y
ste
m
o
ff
e
r
s
a
c
c
u
ra
c
y
a
t
t
h
e
c
o
st
o
f
re
c
u
rsiv
e
a
n
d
c
o
m
p
lex
o
p
e
ra
ti
o
n
s.
T
h
e
p
r
o
p
o
se
d
sy
ste
m
c
o
n
tri
b
u
tes
t
o
in
tro
d
u
c
e
a
f
ra
m
e
w
o
rk
th
a
t
c
a
n
o
ff
e
r
a
g
o
o
d
b
a
lan
c
e
b
e
tw
e
e
n
c
o
m
p
u
tati
o
n
a
l
p
e
rf
o
r
m
a
n
c
e
a
n
d
se
g
m
e
n
tatio
n
p
e
rf
o
r
m
a
n
c
e
.
G
i
v
e
n
a
n
in
p
u
t
o
f
c
h
e
st
x
-
ra
y
,
th
e
sy
st
e
m
o
ffe
rs
p
ro
g
re
ss
iv
e
s
e
a
rc
h
f
o
r
si
m
il
a
r
i
m
a
g
e
o
n
th
e
b
a
sis
o
f
sim
il
a
rit
y
sc
o
re
w
it
h
q
u
e
ried
im
a
g
e
.
Re
g
io
n
-
b
a
se
d
sh
a
p
e
d
e
sc
rip
to
r
is
a
p
p
li
e
d
f
o
r
e
x
tra
c
ti
n
g
th
e
fe
a
tu
re
e
x
c
lu
siv
e
l
y
f
o
r
id
e
n
ti
fy
in
g
th
e
lu
n
g
re
g
io
n
f
ro
m
th
e
th
o
ra
c
ic
re
g
io
n
f
o
ll
o
w
e
d
b
y
c
o
n
to
u
r
a
d
j
u
stm
e
n
t.
T
h
e
f
in
a
l
se
g
m
e
n
tatio
n
o
u
tco
m
e
sh
o
w
s
a
c
c
u
ra
te
id
e
n
ti
f
ica
ti
o
n
f
o
ll
o
w
e
d
b
y
se
g
m
e
n
tatio
n
o
f
a
p
ica
l
a
n
d
c
o
sto
p
h
re
n
ic
re
g
io
n
o
f
lu
n
g
.
Co
m
p
a
ra
ti
v
e
a
n
a
l
y
sis
p
ro
v
e
d
th
a
t
p
ro
p
o
se
d
s
y
ste
m
o
ff
e
rs b
e
tt
e
r
se
g
m
e
n
tatio
n
p
e
rf
o
rm
a
n
c
e
in
c
o
n
tras
t
t
o
e
x
isti
n
g
sy
ste
m
.
K
ey
w
o
r
d
s
:
C
h
e
s
t r
ad
io
g
r
ap
h
C
h
e
s
t x
-
r
a
y
Dis
ea
s
e
d
etec
tio
n
L
u
n
g
Seg
m
en
tatio
n
Co
p
y
rig
h
t
©
2
0
1
9
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Sav
it
h
a
S.
K.
,
Dep
ar
t
m
en
t
o
f
C
SE,
Vis
v
e
s
v
ar
a
y
a
T
ec
h
n
o
lo
g
ical
Un
i
v
er
s
it
y
,
B
elag
av
i,
Kar
n
a
tak
a,
I
n
d
ia
.
E
m
ail:
s
a
v
it
h
a
s
s
k
8
0
@
g
m
ail.
c
o
m
1.
I
NT
RO
D
UCT
I
O
N
I
r
r
esp
ec
tiv
e
o
f
p
r
esen
ce
o
f
v
a
r
io
u
s
f
o
r
m
s
o
f
r
ad
io
lo
g
ical
e
v
alu
atio
n
,
c
h
es
t
x
-
r
a
y
is
o
n
e
o
f
th
e
m
o
s
t
f
r
eq
u
en
tl
y
p
r
escr
ib
ed
b
y
t
h
e
d
o
cto
r
[
1
]
.
I
t
is
b
ec
au
s
e
ch
es
t
x
-
r
a
y
is
o
n
e
o
f
t
h
e
m
o
s
t
ec
o
n
o
m
ical
i
m
ag
in
g
p
r
ac
tices
th
at
r
ev
ea
l
m
u
lt
ip
le
s
et
o
f
cli
n
ical
i
n
f
o
r
m
atio
n
ab
o
u
t
ch
e
s
t
e.
g
.
b
lo
o
d
v
es
s
els,
p
r
o
b
lem
s
as
s
o
ciate
d
w
it
h
h
ea
r
t,
p
o
s
s
ib
le
f
r
ac
t
u
r
e
o
f
r
ib
ca
g
es,
ac
tu
al
co
n
d
itio
n
o
f
lu
n
g
s
,
o
u
tl
in
e
o
f
h
ea
r
t,
an
y
f
o
r
m
o
f
ca
lci
u
m
d
ep
o
s
it,
an
y
ab
n
o
r
m
al
c
h
a
n
g
e
s
af
ter
s
u
r
g
er
y
,
p
o
s
itio
n
o
f
i
m
p
lan
tab
le
d
ev
ices,
etc
[
2
],
[
3
]
.
Ho
w
e
v
er
,
a
ch
est
x
-
r
ay
al
s
o
co
n
s
i
s
ts
o
f
v
ar
io
u
s
p
r
o
b
lem
s
e.
g
.
ill
u
m
i
n
atio
n
,
r
o
tat
io
n
o
f
o
b
j
ec
t,
o
p
ac
ity
,
ab
n
o
r
m
alit
y
i
n
co
n
tr
ast a
n
d
b
r
ig
h
t
n
es
s
,
etc.
Su
c
h
p
r
o
b
le
m
s
ac
ts
as
a
n
i
m
p
ed
i
m
e
n
t
to
w
a
r
d
s
ac
cu
r
ate
d
iag
n
o
s
is
o
f
an
y
d
is
ea
s
e
r
elate
d
to
ch
est.
He
n
c
e,
f
o
r
an
e
f
f
ec
tiv
e
d
iag
n
o
s
i
s
,
it
is
n
ec
es
s
ar
y
t
h
at
th
e
f
o
c
u
s
o
f
t
h
e
r
ad
io
lo
g
is
t
/
p
h
y
s
icia
n
s
h
o
u
ld
b
e
o
n
l
y
co
n
ce
n
tr
ated
to
id
e
n
ti
f
y
t
h
e
co
r
r
ec
t lo
ca
tio
n
o
f
ch
e
s
t.
T
h
is
p
r
o
b
le
m
is
e
f
f
ec
ti
v
el
y
s
o
l
v
ed
b
y
s
eg
m
e
n
tatio
n
p
r
o
ce
s
s
[
4
]
.
B
asicall
y
,
w
h
en
a
ch
es
t
r
ad
i
o
g
r
ap
h
i
s
s
u
b
j
ec
ted
to
s
eg
m
e
n
tatio
n
,
it
ca
n
b
e
ex
p
ec
ted
t
h
at
t
h
e
o
u
tco
m
e
r
e
s
u
lt
s
h
o
u
ld
b
e
s
p
lit
tin
g
o
f
t
h
e
c
h
est
r
ad
io
g
r
ap
h
s
i
n
to
it
s
s
m
al
ler
co
m
p
o
n
e
n
t i
n
t
h
e
f
o
r
m
o
f
o
b
j
ec
t o
r
r
eg
io
n
[
5
]
.
I
t
b
asically
as
s
is
t
s
in
s
e
g
r
eg
ati
n
g
u
n
n
ec
e
s
s
ar
y
d
e
tailin
g
(
i.e
.
b
ac
k
g
r
o
u
n
d
)
f
r
o
m
th
e
ess
e
n
tial
o
b
j
ec
t
(
f
o
r
eg
r
o
u
n
d
)
,
s
o
th
a
t
t
h
e
d
ia
g
n
o
s
is
ca
n
b
e
ca
r
r
ied
o
u
t
w
it
h
b
etter
ac
cu
r
ac
y
.
An
o
t
h
er
b
en
e
f
it
o
f
ca
r
r
y
in
g
o
u
t
s
eg
m
e
n
tatio
n
p
r
o
ce
s
s
i
n
ch
e
s
t
x
-
r
a
y
i
s
t
h
at
it
as
s
is
t
s
i
n
q
u
a
n
tify
i
n
g
th
e
tis
s
u
e
s
,
id
en
t
if
ica
ti
o
n
o
f
a
n
y
o
d
d
itie
s
,
esti
m
ati
n
g
v
o
l
u
m
e
o
f
tar
g
et
o
b
j
ec
t
(
e.
g
.
m
al
ig
n
a
n
t
/
b
e
n
i
g
n
m
as
s
es),
etc
[
6
]
.
I
n
a
n
u
ts
h
ell,
th
e
co
r
e
o
b
j
ec
tiv
e
o
f
th
e
i
m
a
g
e
s
eg
m
e
n
tatio
n
m
eth
o
d
s
is
to
co
n
s
tr
u
ct
a
s
i
m
p
l
er
r
ep
r
esen
tatio
n
o
f
th
e
tar
g
et
ed
o
b
j
ec
t
o
r
r
eg
io
n
an
d
t
h
er
ef
o
r
e
t
h
e
o
u
tco
m
e
o
f
s
eg
m
e
n
tatio
n
p
r
o
ce
s
s
i
s
n
e
v
er
s
i
n
g
le
u
n
i
t.
I
t
i
s
a
s
et
o
f
co
m
p
o
n
en
t
t
h
at
co
u
ld
co
llectiv
el
y
f
o
r
m
ed
t
h
e
ac
t
u
a
l
i
m
a
g
e.
I
n
ex
i
s
ti
n
g
s
y
s
te
m
,
th
er
e
ar
e
v
ar
io
u
s
a
u
t
h
o
r
s
th
a
t
h
a
v
e
ca
r
r
ied
o
u
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
F
r
a
mewo
r
k
fo
r
p
r
o
g
r
es
s
ive
s
e
g
men
ta
tio
n
o
f c
h
est r
a
d
io
g
r
a
p
h
f
o
r
efficien
t d
ia
g
n
o
s
is
o
f in
ert…
(
S
a
vith
a
S
.
K
.
)
983
r
esear
ch
w
o
r
k
to
w
ar
d
s
s
e
g
m
en
tatio
n
co
n
s
i
d
er
i
n
g
m
ed
i
ca
l
i
m
a
g
es
e.
g
.
[
7
]
-
[
9
]
.
Ho
w
e
v
er
,
co
m
p
lex
ities
ass
o
ciate
d
w
it
h
th
e
ch
e
s
t
x
-
r
ay
ar
e
q
u
ite
d
i
f
f
er
e
n
t
co
m
p
ar
ed
to
o
th
er
m
ed
ical
i
m
a
g
es.
A
t
p
r
esen
t
ti
m
e
s
,
t
h
er
e
ar
e
o
n
ly
t
w
o
f
ix
ed
p
o
s
itio
n
s
f
o
r
ca
p
tu
r
in
g
c
h
est
x
-
r
a
y
i.e
.
p
o
s
ter
io
r
-
an
ter
io
r
p
o
s
itio
n
a
n
d
later
al
p
o
s
itio
n
.
A
lt
h
o
u
g
h
,
th
e
s
e
p
o
s
itio
n
s
o
f
f
er
s
9
0
%
v
is
u
al
izatio
n
to
m
a
x
i
m
u
m
cli
n
ical
p
r
o
b
lem
s
ass
o
ciate
d
w
ith
c
h
est
,
b
u
t
s
o
m
et
i
m
e
i
t
v
eil
s
v
is
u
aliz
atio
n
to
s
o
m
e
cr
itical
p
r
o
b
lem
s
e.
g
.
n
o
d
u
le
s
,
m
a
s
s
es,
l
u
m
p
s
,
lesi
o
n
s
o
w
i
n
g
to
p
r
esen
ce
o
f
b
o
n
es
(
i.e
.
r
ib
ca
s
es,
cla
v
icle
b
o
n
e,
etc)
.
T
h
er
ef
o
r
e,
th
e
b
ig
g
er
c
h
alle
n
g
e
i
n
p
er
f
o
r
m
i
n
g
s
eg
m
e
n
tatio
n
i
s
to
f
i
n
d
s
u
c
h
o
cc
lu
d
ed
o
b
j
ec
ts
an
d
ex
tr
ac
t th
e
tar
g
eted
o
b
j
ec
t
as f
o
r
eg
r
o
u
n
d
.
Du
e
to
co
m
p
le
x
it
y
a
s
s
o
ciate
d
w
it
h
p
er
f
o
r
m
in
g
s
eg
m
e
n
tat
io
n
o
f
s
u
c
h
i
n
er
t
r
e
g
io
n
o
f
th
e
c
h
est,
m
aj
o
r
ity
o
f
t
h
e
ex
i
s
ti
n
g
r
esear
ch
w
o
r
k
is
f
o
u
n
d
to
ad
o
p
t
m
a
ch
in
e
lear
n
i
n
g
tec
h
n
iq
u
es
o
r
iter
ativ
e
tech
n
iq
u
e
s
o
n
t
h
e
b
asi
s
o
f
ce
r
tain
p
r
ed
ef
in
ed
i
n
f
o
r
m
atio
n
.
A
lt
h
o
u
g
h
,
a
ll
t
h
ese
e
x
i
s
ti
n
g
tech
n
iq
u
es
ar
e
clai
m
ed
to
o
f
f
er
g
o
o
d
s
eg
m
e
n
tatio
n
p
er
f
o
r
m
a
n
ce
,
b
u
t
th
eir
co
m
p
u
tatio
n
al
p
er
f
o
r
m
a
n
ce
is
q
u
ite
q
u
e
s
t
io
n
ab
le
o
w
i
n
g
to
in
v
o
l
v
e
m
en
t
o
f
h
ig
h
er
d
eg
r
ee
o
f
r
ec
u
r
s
i
v
e
o
p
er
atio
n
.
T
h
er
e
f
o
r
e,
th
e
p
r
esen
t
m
a
n
u
s
cr
ip
t
in
tr
o
d
u
ce
s
a
s
i
m
p
le
an
d
co
s
t
ef
f
ec
ti
v
e
f
r
a
m
e
w
o
r
k
th
at
is
ca
p
ab
le
o
f
p
er
f
o
r
m
in
g
s
eg
m
e
n
tatio
n
u
s
in
g
p
r
o
g
r
ess
i
v
e
ap
p
r
o
ac
h
th
at
h
as
ac
ted
as
a
b
etter
alter
n
ativ
e
o
f
r
ec
u
r
s
iv
e
ap
p
r
o
ac
h
in
e
x
is
tin
g
s
y
s
te
m
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
also
i
n
tr
o
d
u
ce
s
a
n
an
al
y
tical
m
o
d
el
to
clai
m
t
h
e
s
e
g
m
e
n
tat
io
n
p
er
f
o
r
m
an
ce
.
Sectio
n
2
d
is
c
u
s
s
es
ab
o
u
t
t
h
e
e
x
is
tin
g
r
esear
ch
co
n
tr
ib
u
tio
n
f
o
llo
w
ed
b
y
b
r
ief
o
u
tli
n
i
n
g
o
f
p
r
o
b
le
m
s
as
s
o
ciate
d
w
it
h
ex
is
tin
g
s
y
s
te
m
i
n
Sectio
n
3
.
A
d
o
p
ted
r
esear
ch
m
e
th
o
d
o
lo
g
y
o
f
p
r
o
p
o
s
ed
s
y
s
te
m
i
s
b
r
ief
ed
i
n
Sectio
n
4
f
o
llo
w
ed
b
y
e
lab
o
r
ated
d
is
cu
s
s
io
n
o
f
alg
o
r
ith
m
d
esi
g
n
i
n
Sectio
n
5
.
Dis
cu
s
s
io
n
o
f
r
esu
l
ts
ac
co
m
p
lis
h
ed
in
th
e
p
r
o
p
o
s
ed
s
tu
d
y
is
ca
r
r
ied
o
u
t
in
Sectio
n
6
an
d
f
i
n
all
y
t
h
e
co
n
cl
u
d
in
g
r
e
m
ar
k
s
i
s
d
o
n
e
in
Secti
o
n
7
.
2.
RE
L
AT
E
D
WO
RK
T
h
is
s
ec
tio
n
b
r
ief
s
ab
o
u
t
t
h
e
ex
i
s
ti
n
g
r
esear
ch
tec
h
n
iq
u
es
ca
r
r
ied
o
u
t
to
w
ar
d
s
s
e
g
m
e
n
tat
io
n
tech
n
iq
u
es
i
n
m
ed
ical
i
m
a
g
e
s
.
R
esear
ch
er
s
h
av
e
r
ev
ie
w
e
d
ab
o
u
t
ex
is
tin
g
i
m
ag
i
n
g
te
ch
n
iq
u
es
o
f
c
h
est
r
ad
io
g
r
ap
h
s
[
1
0
]
an
d
th
er
eb
y
ex
te
n
d
t
h
e
d
is
c
u
s
s
io
n
in
th
e
lin
e
o
f
c
h
es
t
r
ad
io
g
r
ap
h
e
x
p
licitl
y
.
A
t
p
r
ese
n
t
,
th
er
e
ar
e
v
ar
io
u
s
r
an
g
es
o
f
te
ch
n
iq
u
es
i
m
p
le
m
e
n
ted
f
o
r
s
e
g
m
e
n
tat
io
n
.
T
h
e
w
o
r
k
ca
r
r
ied
o
u
t
b
y
W
an
g
a
n
d
Gu
o
[
1
1
]
h
av
e
p
r
es
en
ted
a
co
m
b
i
n
ed
i
m
p
le
m
e
n
tatio
n
o
f
id
en
tify
i
n
g
s
k
in
b
o
u
n
d
ar
y
,
s
eg
m
e
n
ti
n
g
co
n
to
u
r
r
eg
io
n
s
,
a
n
d
r
e
f
in
e
m
e
n
t
o
f
l
u
n
g
r
eg
io
n
.
Mo
s
t
r
ec
e
n
tl
y
,
a
u
n
iq
u
e
s
e
g
m
e
n
tatio
n
p
r
o
ce
s
s
o
f
s
e
g
r
e
g
ati
n
g
h
ea
r
t
f
r
o
m
t
h
e
lu
n
g
f
ield
w
a
s
i
n
tr
o
d
u
ce
d
b
y
Da
i
et
a
l.
[
1
2
]
.
T
h
e
au
th
o
r
s
h
a
v
e
u
s
ed
co
n
v
o
l
u
tio
n
-
b
ased
s
eg
m
e
n
tatio
n
tech
n
iq
u
e
i
n
o
r
d
er
to
co
n
s
tr
u
ct
a
n
et
w
o
r
k
t
h
at
ca
n
d
is
cr
etize
b
et
w
ee
n
t
h
e
g
r
o
u
n
d
tr
u
th
i
n
f
o
r
m
atio
n
a
n
d
s
y
n
t
h
esized
m
a
s
k
.
A
d
o
p
tio
n
o
f
th
r
es
h
o
ld
-
b
ased
s
eg
m
e
n
tati
o
n
s
ch
e
m
e
ca
n
b
e
o
b
s
er
v
ed
in
th
e
w
o
r
k
o
f
S
h
i
et
a
l.
[
1
3
]
.
T
h
e
au
th
o
r
s
h
av
e
u
s
ed
r
an
d
o
m
w
al
k
al
g
o
r
ith
m
in
o
r
d
er
to
s
eg
m
en
t
l
u
n
g
f
r
o
m
c
h
est
r
eg
io
n
f
u
r
t
h
er
cu
r
v
at
u
r
e
-
b
ased
tec
h
n
iq
u
e
w
a
s
u
til
ized
to
s
m
o
o
th
e
n
t
h
e
co
n
to
u
r
s
.
A
n
o
n
-
co
n
v
e
n
tio
n
al
t
ec
h
n
iq
u
e
o
f
v
ec
to
r
q
u
an
tizat
io
n
h
a
s
b
ee
n
f
o
u
n
d
to
ass
is
t
in
s
e
g
m
en
tatio
n
as
well
f
o
r
ch
est
r
ad
io
g
r
ap
h
s
as
s
ee
n
in
t
h
e
w
o
r
k
o
f
Han
et
a
l.
[
1
4
]
.
T
h
e
w
o
r
k
ca
r
r
ied
o
u
t
b
y
S
h
e
n
et
a
l.
[
1
5
]
h
as
u
s
ed
a
ch
ai
n
co
d
in
g
tec
h
n
iq
u
e
alo
n
g
w
it
h
s
u
p
er
v
i
s
ed
lear
n
i
n
g
al
g
o
r
ith
m
f
o
r
ca
r
r
y
i
n
g
o
u
t
l
u
n
g
s
e
g
m
en
tatio
n
f
o
c
u
s
i
n
g
o
n
ac
cu
r
ac
y
.
Si
m
i
lar
d
ir
ec
tio
n
o
f
e
m
p
h
a
s
is
to
w
ar
d
s
ac
cu
r
ac
y
i
n
s
eg
m
e
n
tatio
n
p
er
f
o
r
m
a
n
ce
was
also
ca
r
r
ied
o
u
t
b
y
C
h
ae
et
a
l.
[
1
6
]
.
T
h
e
au
th
o
r
co
n
tr
ib
u
ted
to
p
r
esen
t
a
tec
h
n
i
q
u
e
f
o
r
r
ec
o
n
s
tr
u
cti
n
g
r
eg
io
n
o
f
s
e
g
m
en
ta
tio
n
t
h
er
eb
y
e
n
h
a
n
cin
g
s
eg
m
e
n
tatio
n
p
er
f
o
r
m
a
n
ce
.
Gill
et
a
l.
[
1
7
]
h
av
e
p
r
esen
t
ed
an
atlas
-
b
ased
m
o
d
el
f
o
r
ca
r
r
y
i
n
g
o
u
t
s
e
g
m
e
n
tatio
n
u
s
in
g
a
f
f
in
e
tr
an
s
f
o
r
m
atio
n
s
c
h
e
m
e.
Ng
o
an
d
C
ar
n
eir
o
[
1
8
]
h
av
e
p
r
ese
n
ted
a
d
ee
p
lear
n
in
g
m
ec
h
a
n
i
s
m
u
s
i
n
g
le
v
el
s
et
me
t
h
o
d
f
o
r
lu
n
g
s
eg
m
e
n
tatio
n
.
T
h
e
tech
n
iq
u
e
p
r
ese
n
ts
g
o
o
d
o
p
tim
izatio
n
to
w
ar
d
s
s
h
a
p
e
f
ea
tu
r
es
d
u
r
i
n
g
s
eg
m
e
n
tatio
n
.
Fil
h
o
et
a
l.
[
1
9
]
h
a
v
e
i
m
p
le
m
en
ted
s
i
n
g
u
lar
it
y
-
b
ased
tec
h
n
iq
u
e
i
n
te
g
r
ated
with
r
e
g
io
n
-
g
r
o
w
i
n
g
m
et
h
o
d
an
d
th
r
es
h
o
ld
in
g
s
c
h
e
m
e
to
g
et
h
er
to
p
er
f
o
r
m
l
u
n
g
s
eg
m
e
n
tatio
n
.
R
u
iz
et
a
l.
[
2
0
]
h
av
e
p
r
esen
ted
a
h
ea
r
t
s
eg
m
e
n
tatio
n
tec
h
n
iq
u
e
u
s
in
g
t
h
r
esh
o
ld
i
n
g
b
ased
s
c
h
e
m
e,
f
ilter
i
n
g
,
an
d
m
o
r
p
h
o
l
o
g
ical
o
p
er
atio
n
s
.
Far
ag
et
a
l.
[
2
1
]
h
av
e
in
tr
o
d
u
ce
d
a
m
o
d
el
th
a
t
i
m
p
le
m
e
n
ts
s
h
ap
e
m
o
d
u
le
alo
n
g
w
it
h
s
tat
is
tical
i
n
f
o
r
m
atio
n
ab
o
u
t
th
e
in
ten
s
it
y
.
T
h
e
tec
h
n
iq
u
e
also
u
s
es
a
d
en
s
it
y
esti
m
atio
n
m
et
h
o
d
u
s
in
g
n
o
n
-
p
ar
am
etr
ic
ap
p
r
o
ac
h
f
o
r
ef
f
ec
tiv
e
l
u
n
g
s
e
g
m
e
n
tat
io
n
.
L
as
s
en
et
a
l.
[
2
2
]
h
av
e
ad
o
p
ted
a
w
ater
s
h
ed
al
g
o
r
ith
m
f
o
r
ac
h
ie
v
in
g
b
etter
class
i
f
icatio
n
o
f
lu
n
g
s
ec
tio
n
d
u
r
in
g
tr
an
s
f
o
r
m
atio
n
p
r
o
ce
s
s
.
Ho
w
e
v
er
,
th
e
p
r
o
ce
s
s
in
v
o
l
v
e
s
lack
o
f
e
f
f
icie
n
c
y
in
lear
n
i
n
g
p
r
o
ce
s
s
p
r
io
r
to
p
e
r
f
o
r
m
s
e
g
m
e
n
tat
io
n
.
S
u
ch
p
r
o
b
le
m
s
o
f
lear
n
in
g
w
er
e
ad
d
r
ess
ed
in
t
h
e
w
o
r
k
o
f
Feu
l
n
er
et
a
l.
[
2
3
]
,
[
2
4
]
b
y
u
s
i
n
g
d
is
cr
i
m
in
at
iv
e
ap
p
r
o
ac
h
f
o
r
s
e
g
m
e
n
ti
n
g
l
y
m
p
h
n
o
d
e.
T
h
e
au
th
o
r
s
h
a
v
e
u
tili
ze
d
g
r
ap
h
c
u
t
alg
o
r
it
h
m
f
o
r
ca
r
r
y
i
n
g
o
u
t
s
e
g
m
en
tati
o
n
.
J
ir
ap
atn
ak
u
l
et
a
l.
[
2
5
]
h
av
e
u
s
ed
s
u
r
f
ac
e
esti
m
atio
n
tech
n
iq
u
e
to
id
en
ti
f
y
an
d
s
e
g
m
e
n
t
p
u
l
m
o
n
ar
y
m
ass
es
f
r
o
m
c
h
est
p
o
r
tio
n
.
L
o
e
t
a
l.
[
2
6
]
h
av
e
u
s
ed
r
eg
io
n
g
r
o
w
in
g
m
ec
h
an
i
s
m
alo
n
g
w
it
h
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
in
o
r
d
er
to
p
er
f
o
r
m
s
eg
m
e
n
tatio
n
.
T
h
e
tech
n
iq
u
e
al
s
o
u
s
es
f
u
zz
y
lo
g
ic
as
w
ell
a
s
an
ato
m
ical
m
o
d
elin
g
u
s
i
n
g
s
e
m
an
t
ic
f
ea
t
u
r
e
s
f
o
r
en
h
an
c
in
g
t
h
e
s
eg
m
e
n
tatio
n
p
er
f
o
r
m
a
n
ce
.
U
s
ag
e
o
f
f
u
zz
y
th
eo
r
y
h
a
s
also
b
ee
n
r
ep
o
r
ted
in
th
e
w
o
r
k
o
f
Z
h
o
u
et
a
l.
[
2
7
]
f
o
r
ass
is
tin
g
i
n
s
eg
m
e
n
tatio
n
.
T
h
e
tech
n
iq
u
e
u
s
e
s
co
r
r
elatio
n
o
f
p
i
x
els
i
n
o
r
d
er
to
p
er
f
o
r
m
d
etec
tio
n
.
Fu
zz
y
cl
u
s
ter
in
g
tech
n
iq
u
e
w
a
s
also
r
ep
o
r
ted
to
b
e
u
s
ed
in
th
e
w
o
r
k
o
f
J
i
et
a
l.
[
2
8
]
f
o
r
ad
d
r
ess
in
g
s
eg
m
e
n
tatio
n
p
r
o
b
l
e
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
9
,
No
.
2
,
A
p
r
il 2
0
1
9
:
9
8
2
-
9
9
1
984
P
u
et
a
l.
[
2
9
]
,
[
3
0
]
h
av
e
p
r
e
s
en
ted
a
u
n
iq
u
e
tr
ee
-
b
ased
s
ch
e
m
e
co
n
s
id
er
in
g
cu
r
v
a
tu
r
e
o
f
h
u
m
an
air
w
a
y
f
o
r
ass
is
ti
n
g
i
n
t
h
r
es
h
o
ld
-
b
ased
lu
n
g
s
e
g
m
en
tatio
n
.
T
h
e
s
ch
e
m
e
al
s
o
p
r
o
v
ed
th
at
ad
o
p
tio
n
o
f
r
ad
ial
-
b
asis
f
u
n
ct
io
n
ca
n
s
i
g
n
if
ican
tl
y
a
s
s
i
s
t i
n
id
en
ti
f
ic
at
io
n
o
f
f
i
s
s
u
r
es
o
v
er
l
u
n
g
s
u
r
f
ac
es.
Usa
g
e
o
f
n
e
u
r
al
n
et
w
o
r
k
an
d
w
av
e
lets
is
r
ep
o
r
ted
to
en
h
a
n
ce
t
h
e
s
eg
m
e
n
tatio
n
m
et
h
o
d
s
as
d
i
s
cu
s
s
ed
i
n
w
o
r
k
o
f
C
e
y
lan
et
a
l.
[
3
1
]
.
T
h
e
au
th
o
r
s
h
a
v
e
also
u
s
ed
r
eg
io
n
-
g
r
o
w
i
n
g
m
et
h
o
d
f
o
r
p
er
f
o
r
m
i
n
g
s
e
g
m
e
n
tatio
n
.
S
h
i
k
a
ta
et
a
l.
[
3
2
]
h
av
e
d
is
cu
s
s
ed
a
tr
ee
-
b
ased
alg
o
r
ith
m
t
h
at
w
o
r
k
s
alo
n
g
w
it
h
E
ig
en
v
al
u
es
f
o
r
co
n
s
tr
u
cti
n
g
ter
m
i
n
al
p
o
r
tio
n
s
o
f
th
e
tis
s
u
es
o
n
l
u
n
g
s
.
C
h
e
n
et
a
l.
[
3
3
]
h
av
e
p
r
esen
ted
a
tech
n
iq
u
e
th
at
u
s
es
en
er
g
y
f
u
n
ct
io
n
al
o
n
g
w
it
h
k
er
n
el
f
o
r
ass
is
tin
g
i
n
s
eg
m
e
n
ta
tio
n
.
T
h
e
co
m
p
lete
m
o
d
eli
n
g
is
ca
r
r
ied
o
u
t
co
n
s
id
er
in
g
s
tati
s
tical
ap
p
r
o
ac
h
.
A
d
o
p
tio
n
o
f
r
eg
io
n
g
r
o
w
i
n
g
tec
h
n
iq
u
e
h
a
s
also
b
ee
n
r
ep
o
r
te
d
in
th
e
w
o
r
k
ca
r
r
ied
o
u
t
b
y
J
ian
g
et
a
l.
[
3
4
]
.
T
h
r
esh
o
ld
in
g
-
b
ased
s
eg
m
e
n
tatio
n
s
c
h
e
m
e
w
a
s
w
it
n
es
s
ed
to
o
f
f
er
b
etter
s
y
s
t
e
m
p
er
f
o
r
m
a
n
ce
d
u
r
i
n
g
s
eg
m
e
n
tat
io
n
p
r
o
ce
s
s
as seen
i
n
t
h
e
w
o
r
k
o
f
L
i
u
et
a
l.
[
3
5
]
.
T
h
er
e
ar
e
v
ar
io
u
s
liter
at
u
r
es
t
o
w
ar
d
s
s
e
g
m
en
tatio
n
p
r
o
b
le
m
s
th
a
t
h
a
s
also
b
ee
n
co
n
s
id
er
in
g
d
if
f
er
e
n
t
m
ed
ical
ca
s
e
s
t
u
d
y
ap
ar
t
f
r
o
m
lu
n
g
s
.
E
n
i
k
o
v
a
n
d
An
to
n
[
3
6
]
h
av
e
u
s
ed
m
ac
h
i
n
e
lear
n
in
g
tec
h
n
iq
u
e
f
o
r
s
eg
m
e
n
ti
n
g
i
m
a
g
es
w
it
h
ce
r
v
ical
s
p
in
es.
T
h
r
es
h
o
ld
in
g
-
b
as
ed
s
ch
e
m
e
w
a
s
ad
o
p
ted
b
y
W
an
g
et
a
l.
[
3
7
]
f
o
r
f
aster
p
r
o
ce
s
s
o
f
m
ed
ical
i
m
a
g
e
s
eg
m
e
n
tatio
n
.
J
ian
g
et
a
l.
[
3
8
]
h
av
e
u
s
ed
lev
el
s
et
m
et
h
o
d
alo
n
g
w
ith
r
e
g
io
n
g
r
o
w
i
n
g
f
o
r
p
er
f
o
r
m
i
n
g
s
eg
m
en
tatio
n
o
f
b
r
ai
n
i
m
a
g
es.
I
n
co
r
p
o
r
atio
n
o
f
g
a
m
e
th
e
o
r
y
f
o
r
i
m
p
r
o
v
in
g
s
eg
m
e
n
tatio
n
w
a
s
d
is
cu
s
s
ed
i
n
w
o
r
k
o
f
Z
h
o
n
g
an
d
W
u
[
3
9
]
.
I
n
teg
r
ated
im
p
le
m
e
n
tatio
n
o
f
ed
g
e
as
w
ell
a
s
r
eg
io
n
w
a
s
r
ep
o
r
ted
in
th
e
w
o
r
k
o
f
L
u
o
et
a
l.
[
4
0
]
w
h
er
e
t
h
e
s
eg
m
e
n
tatio
n
p
er
f
o
r
m
a
n
ce
h
as
b
ee
n
o
p
ti
m
ized
u
s
i
n
g
s
w
ar
m
in
te
lli
g
en
ce
.
B
en
ef
its
o
f
m
u
lt
i
-
t
h
r
es
h
o
ld
in
g
s
c
h
e
m
es
f
o
r
s
eg
m
e
n
ti
n
g
b
r
ain
i
m
ag
es
w
a
s
d
is
cu
s
s
ed
b
y
L
i
u
et
a
l.
[
4
1
]
.
Stu
d
y
to
w
a
r
d
s
au
to
m
ated
s
e
g
m
e
n
tatio
n
i
s
p
u
t
f
o
r
w
ar
d
b
y
Me
c
h
r
ez
et
a
l.
[
4
2
]
co
n
s
id
er
in
g
b
r
ain
i
m
a
g
e
s
.
T
h
e
au
t
h
o
r
s
h
a
v
e
u
s
ed
a
r
ec
u
r
s
i
v
e
p
atch
o
f
i
m
ag
e
s
i
n
o
r
d
er
to
p
er
f
o
r
m
s
e
g
m
e
n
tat
io
n
.
S
i
m
ilar
b
r
ain
i
m
a
g
es
w
er
e
also
in
v
esti
g
a
ted
f
o
r
s
eg
m
en
ta
tio
n
b
y
C
o
n
g
et
a
l.
[
4
3
]
w
h
o
u
s
ed
f
ield
esti
m
atio
n
tech
n
iq
u
e.
Ak
h
a
v
a
n
an
d
Faez
[
4
4
]
illu
s
tr
ated
th
e
s
e
g
m
e
n
tat
io
n
a
lg
o
r
it
h
m
f
o
r
R
et
in
al
b
lo
o
d
v
ess
el
i
m
a
g
e
b
y
u
s
i
n
g
F
u
zz
y
an
d
m
ed
ial
f
il
ter
ap
p
r
o
ac
h
an
d
f
o
u
n
d
e
f
f
ec
ti
v
e
in
d
etec
tio
n
o
f
r
etin
a
l
b
lo
o
d
v
es
s
els.
A
r
esear
ch
to
w
ar
d
s
au
to
m
atic
m
ed
ical
i
m
ag
e
s
eg
m
e
n
tatio
n
is
f
o
u
n
d
in
Sead
a
et
a
l.
[
4
5
]
an
d
n
a
m
ed
th
e
m
o
d
el
a
s
"
ascen
d
in
g
ao
r
ta"
.
T
h
e
m
o
d
el
o
u
tco
m
es
w
it
h
n
ea
l
y
9
5
%
o
f
a
cc
u
r
ac
y
i
n
s
e
g
m
e
n
tatio
n
.
A
D
o
u
b
l
y
tr
u
n
ca
ted
K
-
m
ea
n
cl
u
s
ter
i
n
g
an
d
L
ap
lace
m
i
x
tu
r
e
m
o
d
el
i
s
p
r
esen
ted
f
o
r
i
m
a
g
e
s
e
g
m
e
n
tatio
n
b
y
J
y
o
t
h
ir
m
a
y
i
e
t
a
l.
[
4
6
]
.
I
n
th
i
s
,
d
if
f
er
en
t p
i
x
elled
i
m
a
g
es
w
er
e
a
n
al
y
ze
d
t
h
e
f
o
r
s
u
p
er
io
r
it
y
o
f
m
o
d
el
i
n
s
e
g
m
e
n
tatio
n
.
Hen
ce
,
th
er
e
ar
e
m
u
ltip
le
tec
h
n
iq
u
es
f
o
r
ass
i
s
ti
n
g
in
s
eg
m
en
tatio
n
o
f
m
ed
ical
i
m
ag
e
i
n
liter
atu
r
es.
T
h
e
n
ex
t sectio
n
o
u
tli
n
e
s
th
e
p
r
o
b
lem
s
a
s
s
o
ciate
d
w
it
h
ex
i
s
ti
n
g
s
y
s
te
m
.
3.
P
RO
B
L
E
M
I
DE
NT
I
F
I
CA
T
I
O
N
Af
ter
r
e
v
ie
w
i
n
g
th
e
ex
i
s
ti
n
g
s
y
s
te
m
o
f
s
e
g
m
en
ta
tio
n
,
f
o
llo
w
i
n
g
p
r
o
b
le
m
s
h
as
b
ee
n
id
en
ti
f
ied
:
i)
u
s
ag
e
o
f
th
r
es
h
o
ld
in
g
,
r
u
le,
m
o
d
el
b
ased
,
ed
g
e
-
b
ased
,
p
i
x
el
-
w
is
e
cla
s
s
i
f
icatio
n
,
an
d
r
eg
io
n
-
b
ased
s
ch
e
m
e
s
ar
e
f
r
eq
u
en
tl
y
u
s
ed
f
o
r
p
er
f
o
r
m
i
n
g
s
e
g
m
e
n
tatio
n
.
T
h
e
s
ig
n
i
f
ica
n
t
p
r
o
b
lem
s
o
f
all
th
e
ad
o
p
ted
m
et
h
o
d
s
ar
e
its
ass
u
m
p
tio
n
s
b
ein
g
h
i
g
h
l
y
h
eu
r
i
s
tic.
Hen
ce
,
th
e
y
ar
e
n
o
t
m
u
ch
ap
p
licab
le
f
o
r
p
er
f
o
r
m
i
n
g
s
ca
lab
le
s
eg
m
e
n
tatio
n
p
er
f
o
r
m
an
ce
a
n
d
ca
n
b
e
u
s
ed
o
n
l
y
i
n
p
r
eli
m
i
n
ar
y
s
tag
e
s
.
T
h
is
p
r
o
b
lem
is
m
o
r
e
h
i
g
h
f
o
r
r
u
le
-
b
ased
tech
n
iq
u
es.
i
i)
u
s
ag
e
o
f
d
ef
o
r
m
ab
le
m
o
d
el
-
b
ased
te
ch
n
iq
u
es
s
u
f
f
er
s
f
r
o
m
s
i
g
n
if
i
ca
n
t
li
m
ita
tio
n
s
o
f
h
ig
h
er
co
n
tr
ast
a
n
d
p
r
o
m
i
n
en
t
o
cc
lu
s
io
n
ca
u
s
ed
b
y
ed
g
es
o
f
r
ib
ca
s
es,
h
ig
h
er
d
ep
en
d
en
c
y
o
f
r
ef
er
en
ce
m
o
d
el,
an
d
s
ca
tter
ed
s
o
l
u
tio
n
f
r
o
m
m
u
ltip
le
i
n
ter
n
al
a
ttrib
u
te
s
d
u
r
in
g
s
eg
m
e
n
tatio
n
,
iii)
u
s
a
g
e
o
f
m
ac
h
i
n
e
lear
n
in
g
ap
p
r
o
ac
h
also
r
esu
lts
in
h
i
g
h
er
d
ep
en
d
en
cies
o
n
tr
ain
in
g
o
p
er
atio
n
to
b
e
ca
r
r
ie
d
o
u
t
o
n
d
ataset,
iv
)
lo
w
er
co
n
v
er
g
e
n
ce
s
p
ee
d
p
er
f
o
r
m
an
ce
,
v
)
m
aj
o
r
it
y
o
f
t
h
e
s
ch
e
m
es
p
er
f
o
r
m
s
eg
m
e
n
tatio
n
b
ase
d
o
n
p
ix
el
r
elate
d
in
f
o
r
m
atio
n
t
h
at
i
n
cr
ea
s
es
ac
cu
r
ac
y
o
f
s
e
g
m
e
n
tatio
n
at
th
e
co
s
t
o
f
co
m
p
u
t
atio
n
al
co
m
p
le
x
it
y
.
E
x
is
ti
n
g
m
ac
h
in
e
lear
n
i
n
g
m
e
ch
an
i
s
m
p
er
f
o
r
m
s
h
ig
h
l
y
iter
a
tiv
e
o
p
er
atio
n
t
h
at
i
n
cr
ea
s
es
th
e
ex
ec
u
tio
n
ti
m
e
o
f
th
e
s
eg
m
e
n
tatio
n
p
r
o
ce
s
s
.
T
h
er
ef
o
r
e,
th
er
e
i
s
a
n
ee
d
o
f
p
r
o
ce
s
s
t
h
at
p
er
f
o
r
m
s
les
s
iter
a
tiv
e
o
p
er
atio
n
a
n
d
m
o
r
e
p
r
o
g
r
es
s
i
v
e
s
e
g
m
en
tati
o
n
s
tep
s
th
at
co
u
ld
o
f
f
er
a
b
etter
b
alan
ce
b
et
w
ee
n
ac
cu
r
a
c
y
a
n
d
co
m
p
u
ta
tio
n
ti
m
e.
W
e
p
u
t
f
o
r
w
ar
d
lo
g
ic
t
h
at
an
e
f
f
ec
ti
v
e
s
e
g
m
en
ta
tio
n
alg
o
r
ith
m
al
w
a
y
s
d
e
m
an
d
a
g
o
o
d
eq
u
ilib
r
iu
m
b
et
w
ee
n
co
m
p
u
tat
io
n
al
co
m
p
lex
it
y
a
n
d
ac
cu
r
ac
y
a
n
d
th
er
eb
y
m
ai
n
tai
n
an
o
p
ti
m
al
s
ca
la
b
ilit
y
p
er
f
o
r
m
a
n
ce
.
T
h
e
n
ex
t sectio
n
d
is
c
u
s
s
e
s
ab
o
u
t th
e
ad
o
p
ted
r
esear
ch
m
et
h
o
d
o
lo
g
y
in
o
r
d
er
to
ad
d
r
ess
s
u
c
h
p
r
o
b
lem
s
.
4.
P
RO
P
O
SE
D
M
E
T
H
O
DO
L
O
G
Y
T
h
e
p
r
o
p
o
s
ed
s
tu
d
y
is
a
n
e
x
te
n
s
io
n
o
f
p
r
e
v
io
u
s
w
o
r
k
[
4
7
]
w
h
er
e
t
h
e
e
m
p
h
as
is
w
a
s
o
n
i
n
co
r
p
o
r
atin
g
m
u
lti
-
le
v
el
o
f
i
m
ag
e
p
r
e
-
p
r
o
ce
s
s
i
n
g
o
f
c
h
es
t
r
ad
io
g
r
ap
h
.
T
h
is
p
ar
t
o
f
t
h
e
s
tu
d
y
i
n
tr
o
d
u
c
es
a
s
i
m
p
le
a
n
d
y
et
n
o
v
el
s
eg
m
e
n
ta
tio
n
p
o
lic
y
as
s
h
o
w
ca
s
ed
in
Fi
g
u
r
e
1
.
A
d
o
p
tin
g
a
n
al
y
tical
r
esear
ch
m
eth
o
d
o
lo
g
y
,
th
e
p
r
o
p
o
s
ed
s
y
s
te
m
ta
k
es
th
e
in
p
u
t
o
f
a
q
u
er
ied
i
m
a
g
e
a
n
d
is
ca
p
ab
le
o
f
ex
tr
ac
ti
n
g
h
o
r
izo
n
tal
a
n
d
v
er
tical
p
ix
el
in
f
o
r
m
atio
n
ass
o
ciate
d
w
it
h
eit
h
er
in
te
n
s
it
y
o
r
p
r
o
j
ec
tio
n
.
A
d
o
p
tio
n
o
f
th
i
s
s
tr
ate
g
y
e
n
s
u
r
es
m
iti
g
ati
n
g
a
n
y
f
o
r
m
o
f
ad
v
er
s
e
e
f
f
ec
t
o
f
r
o
tatio
n
s
in
th
e
c
h
est
r
ad
io
g
r
a
p
h
s
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
al
s
o
in
tr
o
d
u
ce
s
u
s
ag
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
F
r
a
mewo
r
k
fo
r
p
r
o
g
r
es
s
ive
s
e
g
men
ta
tio
n
o
f c
h
est r
a
d
io
g
r
a
p
h
f
o
r
efficien
t d
ia
g
n
o
s
is
o
f in
ert…
(
S
a
vith
a
S
.
K
.
)
985
o
f
co
n
te
n
t
-
b
ased
i
m
a
g
e
r
etr
iev
al
tech
n
iq
u
e
i
n
o
r
d
er
to
n
ar
r
o
w
d
o
w
n
t
h
e
m
atc
h
i
n
g
i
m
ag
e
s
.
T
h
e
n
ar
r
o
w
n
es
s
o
f
th
e
m
atc
h
ed
i
m
a
g
e
is
f
u
r
th
er
b
o
o
s
ted
w
it
h
a
n
aid
o
f
s
i
m
i
lar
it
y
s
co
r
e
b
et
w
ee
n
t
h
e
q
u
er
ied
i
m
a
g
e
an
d
d
ataset.
T
h
e
o
u
tco
m
e
o
f
t
h
is
p
r
o
ce
s
s
r
esu
lt
s
i
n
s
et
o
f
5
m
a
tch
i
n
g
i
m
a
g
es
ar
r
an
g
ed
o
n
th
e
b
asis
o
f
t
h
eir
r
an
k
s
.
T
h
e
i
m
ag
e
w
it
h
1
s
t
r
an
k
i
s
c
o
n
s
id
er
ed
f
o
r
f
u
r
th
er
p
r
o
ce
s
s
i
n
g
w
h
er
e
a
n
e
x
tr
ac
tio
n
o
f
f
e
at
u
r
e
is
ca
r
r
ied
o
u
t
u
s
i
n
g
v
is
u
al
d
e
s
cr
ip
to
r
.
Fin
a
ll
y
,
a
s
i
m
p
le
i
m
a
g
e
r
e
g
is
tr
at
io
n
p
r
o
ce
s
s
i
s
i
m
p
le
m
e
n
ted
th
at
co
u
ld
f
u
r
t
h
er
ascer
tain
t
h
e
m
ap
p
in
g
o
f
i
n
p
u
t
to
tar
g
et
i
m
a
g
e.
I
n
t
h
e
e
n
tire
p
r
o
ce
s
s
,
ac
cu
r
ac
y
o
f
t
h
e
m
atc
h
ed
i
m
a
g
e
i
s
h
i
g
h
l
y
en
s
u
r
ed
an
d
th
er
ef
o
r
e
a
co
n
to
u
r
-
b
ased
ad
j
u
s
t
m
e
n
t
is
f
u
r
th
er
ca
r
r
ied
o
u
t
to
elim
i
n
ate
an
y
s
elec
t
io
n
o
f
u
n
n
ec
es
s
ar
y
b
o
u
n
d
ar
y
o
r
r
eg
io
n
o
f
th
e
lu
n
g
.
A
p
p
l
y
in
g
cu
b
ic
s
p
lin
e
in
ter
p
o
latio
n
f
u
r
t
h
er
m
a
k
e
t
h
e
s
y
s
te
m
m
o
r
e
en
h
a
n
ce
d
en
o
u
g
h
to
p
er
f
o
r
m
s
eg
m
e
n
tatio
n
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
i
s
al
s
o
ca
p
ab
le
o
f
p
er
f
o
r
m
in
g
s
eg
m
e
n
tatio
n
as
f
o
llo
w
in
g
i
)
s
eg
m
e
n
ti
n
g
co
m
p
lete
lu
n
g
r
eg
io
n
,
ii)
s
eg
m
en
t
in
g
ap
ical
r
eg
io
n
,
an
d
iii)
s
eg
m
e
n
ti
n
g
co
s
to
p
h
r
en
ic
r
eg
i
o
n
i
n
o
r
d
er
to
en
s
u
r
e
f
u
r
t
h
er
a
p
p
licatio
n
o
f
th
e
o
u
tco
m
e
w
i
t
h
r
esp
ec
t
to
cli
n
ical
in
f
er
en
ce
o
f
c
h
est r
a
d
io
g
r
ap
h
s
.
Q
u
e
r
i
e
d
I
m
a
g
e
J
S
R
T
d
a
t
a
s
e
t
S
e
l
e
c
t
e
d
A
n
a
l
y
s
i
s
O
p
t
i
o
n
S
u
m
o
f
i
n
t
e
n
s
i
t
y
S
u
m
o
f
P
r
o
j
e
c
t
i
o
n
C
B
I
R
S
i
m
i
l
a
r
i
t
y
S
c
o
r
e
Q
u
e
r
y
I
m
a
g
e
a
n
d
s
e
t
o
f
m
a
t
c
h
e
d
i
m
a
g
e
s
w
i
t
h
r
a
n
k
s
F
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
V
i
s
u
a
l
D
e
s
c
r
i
p
t
o
r
I
m
a
g
e
r
e
g
i
s
t
r
a
t
i
o
n
C
o
n
t
o
u
r
A
d
j
u
s
t
m
e
n
t
I
n
t
e
r
p
o
l
a
t
i
o
n
S
e
g
m
e
n
t
a
t
i
o
n
O
p
t
i
o
n
-
1
O
p
t
i
o
n
-
2
Fig
u
r
e
1
.
Sch
e
m
atic
r
ep
r
esen
t
atio
n
o
f
p
r
o
ce
s
s
f
lo
w
o
f
p
r
o
p
o
s
ed
s
eg
m
e
n
tatio
n
5.
AL
G
O
RI
T
H
M
DE
SI
G
N
T
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
e
m
p
h
a
s
izes
o
n
th
e
in
co
r
p
o
r
atin
g
a
p
r
ec
is
e
s
eg
m
e
n
tatio
n
p
r
o
ce
s
s
th
at
ca
n
f
i
n
all
y
a
s
s
is
t
s
in
d
is
ea
s
e
d
iag
n
o
s
is
f
r
o
m
th
e
g
i
v
en
c
h
e
s
t
r
ad
io
g
r
ap
h
s
.
T
h
e
m
ec
h
an
i
s
m
ass
is
ts
in
f
o
r
m
u
lati
n
g
a
b
etter
s
ea
r
ch
co
n
d
itio
n
b
y
f
o
c
u
s
i
n
g
o
n
th
e
e
n
tire
p
r
o
ce
s
s
o
f
m
a
k
i
n
g
t
h
e
i
n
p
u
t
i
m
ag
e
r
ea
d
y
f
o
r
s
e
g
m
e
n
tat
io
n
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
co
n
s
is
t
s
o
f
5
s
eq
u
e
n
tia
l
al
g
o
r
ith
m
s
in
o
r
d
er
to
p
er
f
o
r
m
a
n
e
f
f
e
ctiv
e
s
eg
m
e
n
tatio
n
d
if
f
er
e
n
t
f
r
o
m
e
x
is
tin
g
s
y
s
te
m
i.e
.
i)
A
l
g
o
r
ith
m
f
o
r
p
r
o
ce
s
s
i
n
g
t
h
e
in
p
u
t
i
m
a
g
e,
ii)
A
l
g
o
r
ith
m
f
o
r
co
n
te
n
t
-
b
ased
i
m
a
g
e
r
etr
ie
v
al,
iii)
A
l
g
o
r
ith
m
f
o
r
lo
ca
l
f
ea
t
u
r
e
s
elec
ti
o
n
,
iv
)
A
l
g
o
r
ith
m
f
o
r
i
m
ag
e
r
eg
is
tr
atio
n
p
r
o
ce
s
s
,
an
d
v
)
A
lg
o
r
it
h
m
f
o
r
s
e
g
m
e
n
t
atio
n
.
T
h
is
s
ec
tio
n
d
is
c
u
s
s
es a
b
o
u
t a
ll th
e
al
g
o
r
ith
m
s
i
n
b
r
ief
.
1
.
A
lg
o
r
it
h
m
f
o
r
p
r
o
ce
s
s
in
g
t
h
e
in
p
u
t i
m
a
g
e
I
n
p
u
t:
I
Ou
tp
u
t: H
1
, H
2
Star
t
1.
in
it
I
2.
I
=r
(
I
)
3.
S
r
=∑I
4.
4
.
S
c
=∑(
I
,
2
)
5.
[H
1
b
1
]
=h
is
t(
S
r
m
ax
(
S
r
))
6.
[H
2
b
2
]
=h
is
t(
S
c
,
m
a
x
(
S
c
))
E
n
d
T
h
e
alg
o
r
ith
m
tak
e
s
t
h
e
in
p
u
t
I
(
I
n
p
u
t
I
m
a
g
e)
t
h
at
a
f
ter
p
r
o
ce
s
s
in
g
r
es
u
lt
s
i
n
H1
(
Ho
r
izo
n
tal
P
r
o
j
ec
tio
n
)
,
H2
(
v
er
tical
p
r
o
j
e
ctio
n
)
.
T
h
e
in
p
u
t
i
m
ag
e
I
is
d
i
g
itized
f
o
llo
w
ed
b
y
ap
p
l
y
i
n
g
r
esizin
g
o
p
er
atio
n
r
(
L
in
e
-
2
)
.
T
h
e
alg
o
r
ith
m
co
m
p
u
tes
t
h
e
s
u
m
m
atio
n
o
f
r
o
w
w
i
s
e
ele
m
en
t
s
(
L
i
n
e
-
3
)
as
well
as
co
lu
m
n
w
i
s
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
9
,
No
.
2
,
A
p
r
il 2
0
1
9
:
9
8
2
-
9
9
1
986
ele
m
e
n
ts
(
L
in
e
-
4
)
.
His
to
g
r
a
m
is
f
o
r
m
ed
co
n
s
id
er
in
g
th
e
r
o
w
ele
m
e
n
t
s
co
r
r
esp
o
n
d
in
g
t
o
Sr
(
L
in
e
-
5
)
an
d
co
lu
m
n
s
ec
tio
n
co
n
s
i
s
ts
o
f
al
l
th
e
m
a
x
i
m
u
m
v
al
u
es
o
f
Sc
(
L
i
n
e
-
6
)
.
T
h
is
o
p
er
atio
n
i
s
es
s
e
n
t
ial
in
th
e
p
r
o
p
o
s
ed
s
y
s
te
m
as
it
o
f
f
er
s
th
e
co
m
p
l
ete
s
e
g
m
e
n
tat
io
n
p
r
o
ce
s
s
to
b
e
u
n
d
er
tak
e
n
eit
h
er
o
n
s
u
m
m
atio
n
o
f
h
o
r
izo
n
tal
-
v
er
tical
ele
m
en
ts
o
r
o
n
p
r
o
jectio
n
o
f
h
o
r
izo
n
tal
-
v
er
tical
ele
m
e
n
ts
.
T
h
e
p
r
o
j
ec
tio
n
o
p
er
atio
n
ass
is
ts
i
n
m
iti
g
ati
n
g
an
y
f
o
r
m
o
f
r
o
tatio
n
s
in
t
h
e
in
p
u
t
i
m
a
g
e
u
s
i
n
g
v
er
tical
an
d
h
o
r
izo
n
tal
o
r
t
h
o
g
o
n
al
p
r
o
j
ec
tio
n
.
T
h
is
m
a
k
es
t
h
e
i
n
p
u
t
i
m
ag
e
f
r
ee
f
r
o
m
an
y
f
o
r
m
s
o
f
r
o
tatio
n
al
e
f
f
ec
t.
T
h
e
n
ex
t
s
tag
e
o
f
th
e
al
g
o
r
ith
m
is
to
ap
p
ly
co
n
te
n
t
-
b
ased
i
m
ag
e
r
et
r
iev
al
p
r
o
ce
s
s
as s
h
o
w
n
b
elo
w:
2
.
A
lg
o
r
it
h
m
f
o
r
co
n
te
n
t
-
b
a
s
e
d
im
a
g
e
r
etr
iev
a
l
I
n
p
u
t: ϕ
Ou
tp
u
t: S
ro
, S
co
, β
Star
t
1.
Fo
r
i=1
: ϕ
2.
[(H
10
b
10
)
(
H
20
b
20
)]
ϕ
(
m
i)
,
w
h
er
e
a=
1
,
2
,
3
,
4
3.
S
ro
=∑(
a
ϕ
)
,
a=
5
4.
S
co
=∑(
(
a
ϕ
)
,
2
)
5.
E
n
d
6.
Fo
r
n
=1
:p
7.
ix
1
=b
1
==
p
(
n
)
,
p
1
(
n
)
=H
10
(
ix
1
)
8.
ix
2
=b
10
==
p
(
n
)
,
p
2
(
n
)
=H
10
(
ix
2
)
9.
E
n
d
10.
Fo
r
n
=1
:q
11.
ix
1
=b
2
==
q
(
n
)
,
q
1
(
n
)
=H
1
(
ix
1
)
12.
ix
2
=b
20
==
q
(
n
)
,
q
2
(
n
)
=H
20
(
ix
2
)
13.
E
n
d
14.
g
et
r
w
=n
/(
n
+
m
)
15.
Fo
r
(
x
y
)
=1
:(
n
m
)
16.
p
t
=p
t
+√
{p
1
(
x
)
.
p
2
(
x
)
}
17.
q
t
=q
t
+√
{q
1
(
x
)
.
q
2
(
x
)
}
18.
E
n
d
19.
β
α
.
q
t
+(
1
-
α
)
.
q
t
20.
(
m
_
I
m
_
m
)
ϕ
(
a
ix
(
1
)
)
,
w
h
e
r
e
a=
5
an
d
6
21.
(S
ro
S
co
)
=∑(
m
_
I
)
22.
o
b
tain
(
S
co
H
10
b
10
H
20
b
20
)
ϕ
(
a,
ix
(
1
)
)
,
w
h
er
e
a=
1
-
4
E
n
d
T
h
is
alg
o
r
it
h
m
tak
e
s
t
h
e
i
n
p
u
t
o
f
ϕ
(
d
atab
ase)
th
at
a
f
ter
p
r
o
ce
s
s
i
n
g
y
ield
s
to
Sro
(
s
u
m
o
f
h
o
r
izo
n
tal
in
te
n
s
it
y
)
,
Sco
(
s
u
m
o
f
v
er
tic
al
in
te
n
s
it
y
)
,
β
(
s
i
m
ilar
it
y
m
e
asu
r
e)
.
A
s
p
ec
if
ic
i
m
ag
e
d
ata
s
et
ϕ
i
s
u
s
ed
w
h
er
e
th
e
f
ir
s
t
a
n
d
s
ec
o
n
d
r
o
w
ele
m
en
t
co
r
r
esp
o
n
d
s
to
h
o
r
izo
n
t
al
p
r
o
j
ec
ti
o
n
H1
0
an
d
b
1
0
w
h
ile
t
h
ir
d
an
d
f
o
u
r
t
h
r
o
w
ele
m
en
ts
co
r
r
esp
o
n
d
to
v
er
tical
p
r
o
j
ec
tio
n
H2
0
an
d
b
2
0
r
esp
ec
tiv
el
y
(
L
i
n
e
-
2
)
.
T
h
e
alg
o
r
ith
m
al
s
o
co
m
p
u
tes
t
h
e
s
u
m
m
atio
n
o
f
r
o
w
-
w
i
s
e
(
Sro
)
an
d
co
lu
m
n
-
w
i
s
e
(
Sco
)
e
lem
en
t
r
esp
ec
ti
v
el
y
(
L
i
n
e
-
3
an
d
L
in
e
-
4
)
.
Ho
w
e
v
er
,
th
e
al
g
o
r
ith
m
ch
o
o
s
es
to
co
n
s
id
er
s
u
m
o
f
h
o
r
izo
n
tal
an
d
v
er
tical
in
ten
s
it
y
(
i.e
.
Sro
an
d
Sco
)
o
n
ly
u
p
o
n
s
elec
tio
n
o
f
s
u
m
m
atio
n
o
f
h
o
r
izo
n
tal
an
d
v
er
tical
el
e
m
en
ts
.
On
t
h
e
o
th
er
h
a
n
d
,
th
e
alg
o
r
ith
m
s
elec
ts
h
o
r
izo
n
tal
a
n
d
v
er
tical
p
r
o
j
ec
ts
(
i.e
.
b
1
0
,
H1
0
an
d
b
2
0
,
H2
0
)
u
p
o
n
s
elec
tio
n
o
f
h
o
r
izo
n
tal
-
v
er
tical
p
r
o
j
ec
tio
n
.
T
h
e
n
ex
t
p
ar
t o
f
th
e
alg
o
r
it
h
m
co
n
s
id
er
s
s
i
m
ilar
it
y
co
e
f
f
icie
n
t p
t
h
at
co
m
p
u
te
s
t
h
e
co
m
m
o
n
ele
m
e
n
ts
b
et
w
ee
n
th
e
p
r
o
j
ec
tio
n
s
b
1
an
d
b
1
0
(
L
in
e
-
6
)
.
C
o
n
s
id
er
i
n
g
al
l t
h
e
v
a
lu
es o
f
p
,
t
h
e
alg
o
r
it
h
m
ch
ec
k
s
t
h
e
p
r
o
j
ec
tio
n
b
1
o
f
in
p
u
t
i
m
a
g
e
to
b
e
m
atc
h
i
n
g
w
i
th
u
n
it
v
al
u
e
o
f
p
r
o
j
ec
tio
n
in
t
h
e
d
atab
ase.
On
l
y
t
h
e
h
o
r
izo
n
t
al
p
r
o
j
ec
tio
n
s
H1
0
ar
e
co
n
s
id
er
ed
f
o
r
f
u
r
th
er
p
r
o
ce
s
s
in
g
i.e
.
p
1
(
L
i
n
e
-
7
an
d
L
i
n
e
-
8
)
.
T
h
e
s
i
m
ilar
o
p
er
atio
n
is
ca
r
r
ied
o
u
t
f
o
r
v
er
tical
p
r
o
j
ec
tio
n
w
i
th
s
i
m
i
l
ar
it
y
co
ef
f
icie
n
t
q
(
L
in
e
-
1
0
)
f
o
r
ca
lcu
latin
g
q
1
an
d
q
2
(
L
in
e
-
1
1
an
d
L
in
e
-
1
2
)
.
T
h
e
alg
o
r
ith
m
al
s
o
co
m
p
u
tes
th
e
r
elativ
e
w
ei
g
h
t
r
w
u
s
i
n
g
m
at
h
e
m
a
tical
ex
p
r
es
s
io
n
s
h
o
w
n
in
L
i
n
e
-
1
4
,
w
h
er
e
th
e
v
a
r
iab
le
n
a
n
d
m
co
r
r
esp
o
n
d
s
to
len
g
t
h
o
f
p
a
n
d
q
.
T
h
e
n
ex
t
p
ar
t
o
f
t
h
e
al
g
o
r
ith
m
ca
l
cu
lates
a
te
m
p
o
r
ar
y
v
ar
iab
les
p
t
(
L
in
e
-
1
6
)
an
d
q
t
(
L
i
n
e
-
1
7
)
i
n
o
r
d
er
to
co
m
p
u
te
a
f
in
al
s
i
m
ilar
it
y
co
e
f
f
i
cien
t
β
u
s
i
n
g
t
h
e
ex
p
r
ess
io
n
s
h
o
w
n
i
n
L
in
e
-
1
9
.
A
s
o
r
tin
g
p
r
o
ce
s
s
in
d
escen
d
in
g
o
r
d
er
is
ca
r
r
ied
o
u
t
f
o
r
th
e
o
b
tain
ed
v
alu
e
o
f
s
i
m
ilar
it
y
co
ef
f
icie
n
t.
A
m
atr
ix
m
_
I
an
d
m
_
m
r
ep
r
esen
t
s
m
atc
h
ed
i
m
ag
e
a
n
d
(
o
n
g
o
in
g
)
m
atc
h
i
n
g
i
m
a
g
e
r
esp
ec
tiv
el
y
(
L
i
n
e
-
2
0
)
.
Fi
n
all
y
,
s
u
m
m
atio
n
s
o
f
r
o
w
-
w
is
e
a
n
d
co
lu
m
n
-
w
i
s
e
ele
m
e
n
t
s
ar
e
ca
p
tu
r
ed
(
i.e
.
Sro
an
d
Sco
)
f
r
o
m
m
atc
h
ed
i
m
a
g
e
m
_
I
(
L
in
e
-
2
1
)
f
o
llo
w
ed
b
y
ac
q
u
i
r
in
g
o
f
s
u
m
o
f
h
o
r
izo
n
tal
i
n
te
n
s
it
y
(
Sro
)
,
s
u
m
o
f
v
er
tical
i
n
te
n
s
it
y
(
Sco
)
,
h
o
r
iz
o
n
tal
p
r
o
j
ec
tio
n
(
b
1
0
,
H1
0
)
,
an
d
v
er
tica
l
p
r
o
j
ec
tio
n
(
b
2
0
,
h
2
0
)
as
s
h
o
w
n
i
n
L
i
n
e
-
2
2
.
T
h
er
ef
o
r
e,
th
e
g
o
o
d
av
ail
ab
ilit
y
o
f
p
r
o
j
ec
tio
n
-
b
ased
in
f
o
r
m
atio
n
s
i
g
n
if
ican
tl
y
a
s
s
is
ts
to
m
i
tig
a
te
an
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
F
r
a
mewo
r
k
fo
r
p
r
o
g
r
es
s
ive
s
e
g
men
ta
tio
n
o
f c
h
est r
a
d
io
g
r
a
p
h
f
o
r
efficien
t d
ia
g
n
o
s
is
o
f in
ert…
(
S
a
vith
a
S
.
K
.
)
987
f
o
r
m
s
o
f
m
i
s
ali
g
n
m
e
n
t
i
n
t
h
e
i
m
ag
e.
A
t
t
h
e
s
a
m
e
ti
m
e,
th
e
a
lg
o
r
ith
m
o
f
co
n
te
n
t
-
b
ased
i
m
a
g
e
r
etr
iev
al
r
esu
lts
in
g
o
o
d
n
u
m
b
er
o
f
m
o
s
t r
elev
an
t i
m
ag
e
s
th
er
eb
y
r
ed
u
c
in
g
t
h
e
ef
f
o
r
t o
f
s
ea
r
ch
to
w
ar
d
s
b
est
f
it
i
m
a
g
e.
T
h
e
alg
o
r
it
h
m
also
m
a
k
es
u
s
e
o
f
s
i
m
i
la
r
it
y
co
ef
f
icie
n
t
i
n
o
r
d
er
to
co
n
f
ir
m
t
h
e
r
elev
an
c
y
s
co
r
e
o
f
t
h
e
q
u
er
ie
d
i
m
a
g
e
w
it
h
t
h
at
o
f
t
h
e
d
ata
s
et.
Up
o
n
s
e
lectio
n
o
f
t
h
e
b
e
s
t
m
atch
ed
i
m
a
g
e,
th
e
i
m
ag
e
i
s
n
o
w
r
ea
d
y
to
b
e
ex
tr
ac
ted
f
o
r
its
f
ea
tu
r
e
u
s
i
n
g
t
h
e
alg
o
r
it
h
m
b
elo
w
:
3
.
A
lg
o
r
it
h
m
f
o
r
lo
ca
l f
ea
t
u
r
e
s
elec
tio
n
I
n
p
u
t: I
Ou
tp
u
t: i
m
3
Star
t
1.
[
(
im
1
im
2
)
(
d
es
1
d
es
2
)
(
lo
c
1
lo
c
2
)
]
=σ
(
p
1
p
2
)
2.
Fo
r
i=1
: d
es
1
3.
[
v
.
in
]
s
o
r
t(
ic(
d
p
)
)
4.
I
f
(
v
(
1
)
<d
R
.
v
(
2
)
)
5.
m
(
i)
=in
(
1
)
6.
E
ls
e
7.
m
(
i)
=0
8.
E
n
d
9.
im
3
p
1
|
|
p
2
E
n
d
T
h
is
alg
o
r
ith
m
is
r
esp
o
n
s
ib
le
f
o
r
ex
tr
ac
tin
g
lo
ca
l
f
ea
t
u
r
es
f
r
o
m
t
h
e
m
atc
h
ed
i
m
a
g
e
o
b
tain
ed
f
r
o
m
co
n
ten
t
-
b
ased
i
m
ag
e
r
etr
ie
v
al
p
r
o
ce
s
s
.
T
h
e
alg
o
r
ith
m
d
esi
g
n
i
s
b
ased
o
n
v
is
u
al
d
escr
ip
to
r
o
f
r
eg
io
n
-
b
ased
s
h
ap
e
m
et
h
o
d
.
T
h
e
alg
o
r
ith
m
tak
e
s
t
h
e
i
n
p
u
t
o
f
I
(
p
r
o
ce
s
s
ed
in
p
u
t
i
m
a
g
e)
t
h
at
a
f
ter
p
r
o
ce
s
s
i
n
g
y
ie
ld
s
i
m
3
(
i
m
ag
e
w
it
h
lo
ca
l
f
ea
tu
r
e)
.
T
h
e
v
ar
iab
les
p
1
an
d
p
2
r
ep
r
esen
ts
I
*
2
5
5
an
d
m
_
I
*
2
5
5
r
esp
ec
ti
v
el
y
.
T
h
e
alg
o
r
ith
m
i
m
p
le
m
e
n
t
s
a
f
u
n
ct
io
n
σ
f
o
r
p
er
f
o
r
m
in
g
t
h
e
e
x
tr
ac
tio
n
o
f
lo
ca
l
f
ea
tu
r
es
in
t
h
e
f
o
r
m
o
f
s
ets
e.
g
.
(
i
m
1
i
m
2
)
(
d
es1
d
es2
)
(
lo
c1
lo
c2
)
as
s
h
o
w
n
in
L
i
n
e
-
1
.
T
h
e
alg
o
r
ith
m
co
m
p
u
te
s
th
e
v
ec
to
r
o
f
d
o
t
p
r
o
d
u
ct
d
p
o
f
f
ir
s
t
d
escr
ip
to
r
(
d
es1
)
an
d
tr
an
s
p
o
s
e
o
f
s
ec
o
n
d
d
escr
ip
to
r
(
L
i
n
e
-
3
)
.
T
h
is
o
p
er
atio
n
i
s
f
u
r
th
er
f
o
llo
w
ed
b
y
s
o
r
tin
g
o
p
er
atio
n
in
o
r
d
er
to
o
b
tain
v
a
n
d
i
n
m
ap
s
(
L
i
n
e
-
3
)
.
T
h
e
alg
o
r
ith
m
t
h
en
c
h
ec
k
s
o
f
th
e
v
(
1
)
is
f
o
u
n
d
m
o
r
e
th
a
n
d
is
tan
ce
r
at
io
d
R
m
u
ltip
lied
w
it
h
v
(
2
)
th
a
n
t
h
e
u
n
i
t in
d
e
x
is
ass
i
g
n
ed
to
m
atc
h
(
L
i
n
e
-
5
)
o
r
else
ze
r
o
is
as
s
i
g
n
ed
(
L
i
n
e
-
7
)
.
T
h
e
d
is
ta
n
ce
r
atio
d
R
o
n
l
y
k
ee
p
m
a
tch
es
i
n
w
h
ic
h
t
h
e
r
atio
o
f
v
ec
to
r
an
g
les
f
r
o
m
t
h
e
%
n
ea
r
est
to
s
ec
o
n
d
n
ea
r
e
s
t
n
e
ig
h
b
o
u
r
is
le
s
s
th
a
n
d
R
.
B
o
th
th
e
i
m
a
g
es
o
b
tai
n
ed
i.e
.
p
1
an
d
p
2
as
th
e
o
u
tco
m
e
.
Th
e
p
r
o
ce
s
s
ed
i
m
ag
e
i
s
f
u
r
t
h
e
r
s
u
b
j
ec
ted
to
r
eg
is
tr
atio
n
p
r
o
c
ess
as s
h
o
w
n
b
elo
w
:
4
.
A
lg
o
r
it
h
m
f
o
r
i
m
a
g
e
r
eg
i
s
tr
atio
n
p
r
o
ce
s
s
I
n
p
u
t: I
,
m
_
I
Ou
tp
u
t: I
reg
,
m
w
Star
t
1.
let
I
1
an
d
I
2
b
e
I
an
d
m
_
I
2.
[I
reg
m
2
]
μ
1
(I
2
I
1
,
ρ
)
3.
t
form
μ
2
(I
2
I
1
,
ρ
)
4.
mw
γ
(
m
_
m
,
t
fo
r
m
)
E
n
d
T
h
is
alg
o
r
ith
m
is
m
ai
n
l
y
r
es
p
o
n
s
ib
le
f
o
r
p
er
f
o
r
m
in
g
i
m
a
g
e
r
eg
is
tr
atio
n
p
r
o
ce
s
s
w
h
ic
h
tak
es
t
h
e
in
p
u
t
o
f
I
(
p
r
o
ce
s
s
ed
in
p
u
t
i
m
ag
e)
,
m
_
I
(
Ma
tc
h
ed
i
m
a
g
e)
i
n
o
r
d
er
to
g
e
n
er
ate
t
h
e
o
u
tp
u
t
o
f
I
r
eg
(
r
eg
i
s
ter
ed
i
m
a
g
e)
,
m
w
(
co
n
to
u
r
ad
j
u
s
te
d
i
m
ag
e)
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
i
m
p
le
m
en
t
s
a
f
u
n
ctio
n
as
a
co
n
f
ig
u
r
atio
n
s
u
itab
le
f
o
r
r
eg
is
ter
in
g
i
m
ag
e
s
u
s
i
n
g
m
u
lti
m
o
d
al
ap
p
r
o
ac
h
.
T
h
is
f
u
n
ctio
n
f
o
r
m
u
lates
a
n
o
p
tim
izer
as
w
el
l
as
m
etr
ic
co
n
f
i
g
u
r
atio
n
i
n
o
r
d
er
to
r
eg
is
ter
th
e
i
n
te
n
s
it
y
-
b
as
ed
i
m
ag
e.
T
h
e
v
ar
iab
le
ρ
d
e
p
icts
a
s
et
o
f
b
o
t
h
o
p
tim
izer
an
d
a
m
etr
ic
as
a
s
tr
u
ct
u
r
e
w
it
h
e
x
p
licit
i
n
f
o
r
m
atio
n
ab
o
u
t
d
if
f
er
e
n
t
o
p
ti
m
izer
p
r
o
p
e
r
ties
f
o
r
o
b
tain
in
g
co
n
v
er
g
en
ce
o
v
er
a
g
lo
b
al
m
a
x
i
m
u
m
.
T
h
e
a
lg
o
r
i
th
m
i
m
p
le
m
en
ts
a
f
u
n
ctio
n
μ
1
f
o
r
r
eg
i
s
ter
in
g
an
i
m
a
g
e
(
L
in
e
-
2
)
as
w
ell
a
s
f
u
n
ctio
n
μ
2
f
o
r
est
i
m
ati
n
g
t
h
e
g
eo
m
etr
ic
tr
an
s
f
o
r
m
a
tio
n
r
eq
u
ir
e
to
s
tab
ilize
t
h
e
m
o
v
i
n
g
i
m
a
g
e
(
L
in
e
-
3
)
.
T
h
e
n
ex
t
s
tep
o
f
th
e
al
g
o
r
ith
m
is
t
o
p
er
f
o
r
m
tr
an
s
f
o
r
m
atio
n
o
f
m
atc
h
ed
i
m
a
g
e
m
_
m
as
p
er
th
e
g
eo
m
etr
ic
tr
a
n
s
f
o
r
m
atio
n
ex
h
ib
ited
b
y
t
f
o
r
m
.
H
en
ce
,
t
h
e
o
u
tco
m
e
m
w
(
L
in
e
-
4
)
ca
n
b
e
co
n
s
id
er
ed
as
a
tr
a
n
s
f
o
r
m
ed
i
m
a
g
e
th
at
h
as
b
etter
ad
j
u
s
t
m
e
n
t
o
f
t
h
e
co
n
to
u
r
s
o
f
th
e
l
u
n
g
r
e
g
io
n
s
.
T
h
e
n
ex
t
p
ar
t
o
f
t
h
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
9
,
No
.
2
,
A
p
r
il 2
0
1
9
:
9
8
2
-
9
9
1
988
alg
o
r
ith
m
co
n
t
in
u
es
f
o
r
f
i
n
al
s
tep
o
f
s
eg
m
en
tatio
n
p
r
o
ce
s
s
as
s
h
o
w
n
b
elo
w
:
5
.
A
lg
o
r
it
h
m
f
o
r
id
en
ti
f
y
in
g
u
p
p
er
an
d
lo
w
er
lu
n
g
s
e
g
m
en
t
I
n
p
u
t: I
,
m
w
Ou
tp
u
t: u
,
l
Star
t
1.
y
τ
(
I
)
2.
b
w
=λ(
m
w
)
3.
[(x
s
y
s
)]
co
(
)
,
w
h
er
e
co
=
f
(
b
w
)
4.
[x
o1
y
o1
]
Ω
(
y
,
v
ar
)
5.
5
.
(
m
1
m
2
)
{a
r
g
m
in
(y
s1
y
s2
)
ar
g
m
ax
(y
s1
y
s2
)}
6.
u
I
(
n
r
)
7.
l
I
(
en
d
-
n
r
)
E
n
d
T
h
e
alg
o
r
ith
m
co
n
s
id
er
s
t
h
e
in
p
u
t
a
s
tr
a
n
s
f
o
r
m
ed
i
m
ag
e
m
w
o
b
tain
ed
f
r
o
m
th
e
p
r
io
r
alg
o
r
ith
m
(
L
in
e
-
2
)
an
d
u
s
e
s
a
f
u
n
ctio
n
λ
f
o
r
b
in
ar
izin
g
t
h
e
i
m
a
g
e
an
d
th
er
eb
y
r
esu
l
tan
t
i
m
a
g
e
o
f
b
w
is
o
b
tain
ed
(
L
in
e
-
2
)
.
T
h
e
n
ex
t
s
tep
is
to
o
b
tain
th
e
b
in
ar
ized
b
o
u
n
d
ar
ies
co
o
f
th
e
p
r
o
ce
s
s
ed
i
m
ag
e
b
w
(
L
i
n
e
-
3
)
.
T
h
e
s
y
s
te
m
th
an
e
x
tr
ac
ts
t
h
e
b
in
ar
ized
b
o
u
n
d
ar
ies
w
i
th
r
esp
ec
t
to
ex
p
lic
it
r
o
w
s
i
n
o
r
d
er
to
o
b
tain
(
x
s
1
y
s
1
)
an
d
(
x
s
2
y
s
2
)
.
A
f
ilter
i
n
g
f
u
n
ctio
n
τ
is
i
n
v
o
k
ed
o
n
i
m
a
g
e
I
(
L
in
e
-
1
)
f
o
ll
o
w
ed
b
y
ap
p
ly
in
g
r
ec
u
r
s
i
v
e
p
r
o
ce
s
s
Ω
(
L
in
e
-
4
)
co
n
s
id
er
in
g
d
i
f
f
er
en
t
v
ar
iab
le
s
v
ar
.
Fi
n
all
y
,
t
h
e
s
eg
m
e
n
ted
i
m
a
g
e
is
g
e
n
er
ated
b
y
co
n
j
o
in
in
g
(
x
s
y
s
)
a
n
d
(
x
o
y
o
)
p
o
in
t
s
as
f
o
llo
w
s
:
t
h
e
al
g
o
r
ith
m
e
x
tr
ac
t
s
t
h
e
m
i
n
i
m
u
m
ar
g
u
m
e
n
t
s
o
f
(
y
s
1
y
s
2
)
an
d
m
ax
i
m
u
m
ar
g
u
m
e
n
ts
o
f
(
y
s
1
y
s
2
)
in
o
r
d
er
to
f
o
r
m
m
1
an
d
m
2
r
esp
ec
t
iv
el
y
(
L
i
n
e
-
5
)
.
T
h
e
alg
o
r
ith
m
ex
tr
ac
ts
n
u
m
b
er
o
f
r
o
w
s
n
r
a
n
d
n
u
m
b
er
o
f
co
l
u
m
n
s
n
c
f
r
o
m
in
p
u
t
i
m
a
g
e
I
a
n
d
o
b
tain
s
th
e
u
p
p
er
p
o
r
tio
n
u
(
L
in
e
-
6
)
a
n
d
lo
w
er
p
o
r
tio
n
l
(
L
in
e
-
7
)
.
A
clo
s
er
l
o
o
k
in
to
th
e
al
g
o
r
ith
m
s
h
o
ws
th
at
p
r
o
p
o
s
ed
s
eg
m
e
n
tatio
n
tec
h
n
iq
u
e
is
q
u
i
te
d
if
f
er
e
n
t
f
r
o
m
all
co
n
v
e
n
tio
n
a
l
s
eg
m
e
n
tatio
n
tech
n
iq
u
e
in
f
o
llo
w
i
n
g
m
a
n
n
er
v
iz.
i)
it
i
s
p
r
o
g
r
ess
iv
e
a
n
d
n
o
t
iter
ativ
e
as
s
ee
n
in
ex
is
ti
n
g
tech
n
iq
u
es
i
n
liter
atu
r
e,
ii)
th
e
s
eg
m
e
n
tat
io
n
o
p
er
atio
n
is
m
o
r
e
in
to
o
p
tim
iz
in
g
th
e
d
etec
tio
n
p
er
f
o
r
m
an
ce
u
n
less
ex
is
ti
n
g
s
y
s
te
m
th
a
t
o
n
l
y
f
o
cu
s
es
o
n
d
etec
tio
n
,
ii
i)
th
e
i
m
ag
e
r
e
g
is
tr
atio
n
p
r
o
ce
s
s
f
o
llo
w
ed
b
y
co
n
to
u
r
ad
j
u
s
t
m
e
n
t
f
u
r
th
er
f
i
n
e
-
t
u
n
e
s
th
e
ed
g
es
o
f
lu
n
g
r
eg
io
n
w
it
h
o
u
t
e
v
en
ap
p
l
y
i
n
g
co
n
v
e
n
tio
n
al
ed
g
e
-
b
ased
s
e
g
m
e
n
tat
io
n
p
r
o
ce
s
s
.
T
h
is
c
au
s
e
s
th
e
alg
o
r
it
h
m
to
o
b
tain
m
o
r
e
ac
c
u
r
ate
in
f
o
r
m
atio
n
ab
o
u
t
l
u
n
g
r
eg
io
n
an
d
p
er
f
o
r
m
s
f
aster
id
en
ti
f
i
ca
tio
n
o
f
it.
T
h
e
n
ex
t
s
ec
tio
n
o
u
tlin
e
s
th
e
r
es
u
lt
o
b
tain
ed
b
y
p
r
o
p
o
s
ed
s
tu
d
y
.
6.
RE
SU
L
T
ANAL
YSI
S
T
h
e
ass
es
s
m
e
n
t
o
f
t
h
e
p
r
o
p
o
s
ed
s
t
u
d
y
h
as
b
ee
n
ca
r
r
ied
o
u
t
u
s
i
n
g
J
ap
an
ese
So
ciet
y
R
a
d
io
lo
g
ical
T
ec
h
n
o
lo
g
y
(
J
SR
T
)
d
ataset
o
f
ch
e
s
t
r
ad
io
g
r
ap
h
w
h
er
e
th
e
s
i
ze
o
f
t
h
e
i
m
ag
e
s
i
s
8
,
1
9
2
KB
.
A
ll
th
e
i
m
ag
e
s
ar
e
g
r
a
y
-
s
ca
le
i
m
ag
e
s
w
i
th
co
lo
r
d
ep
th
o
f
1
2
b
it.
T
h
is
s
ize
is
q
u
ite
b
i
g
g
er
co
m
p
ar
ed
to
co
n
v
e
n
tio
n
al
m
ed
ical
i
m
a
g
es.
T
h
e
s
tu
d
y
o
u
tco
m
e
o
f
t
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
h
a
s
b
ee
n
co
m
p
ar
ed
w
it
h
th
e
ex
i
s
ti
n
g
s
e
g
m
e
n
tat
io
n
tech
n
iq
u
es.
T
h
er
e
ar
e
d
iv
er
s
e
s
tu
d
ie
s
in
liter
atu
r
e
s
p
er
tain
i
n
g
to
s
e
g
m
en
ta
tio
n
a
n
d
h
e
n
ce
w
e
c
h
o
o
s
e
t
h
e
o
n
e
th
at
i
s
f
r
eq
u
en
tl
y
ad
o
p
ted
in
s
eg
m
e
n
tatio
n
i.e
.
ed
g
e
b
ase
d
,
th
r
esh
o
ld
-
b
ased
,
an
d
r
eg
i
o
n
-
b
ased
s
c
h
e
m
es
.
T
ab
le
1
h
ig
h
li
g
h
ts
th
e
o
u
tco
m
e
o
f
co
m
p
ar
ativ
e
a
n
al
y
s
is
o
f
p
r
o
p
o
s
ed
an
d
ex
is
tin
g
s
y
s
te
m
w
it
h
r
esp
ec
t
to
E
d
g
e
-
b
ased
,
th
r
es
h
o
ld
-
b
ased
,
an
d
r
eg
io
n
-
b
ased
.
T
ab
le
1
.
Ou
tco
m
e
o
f
C
o
m
p
ar
a
tiv
e
An
al
y
s
is
P
e
r
f
o
r
man
c
e
F
a
c
t
o
r
s
Ex
i
st
i
n
g
[
4
7
]
P
r
o
p
o
se
d
Ed
g
e
-
b
a
se
d
T
h
r
e
sh
o
l
d
-
b
a
se
d
R
e
g
i
o
n
-
b
a
se
d
Ex
e
c
u
t
i
o
n
T
i
me
(
s)
1
.
0
1
9
6
8
1
.
0
2
6
5
4
.
2
8
4
1
0
.
5
4
8
7
A
c
c
u
r
a
c
y
(
%)
9
7
%
9
8
%
9
4
%
9
9
%
T
h
e
r
eg
io
n
-
w
is
e
s
e
g
m
e
n
tat
io
n
p
r
o
ce
s
s
r
es
u
lt
s
i
n
g
en
er
at
io
n
o
f
to
o
m
u
c
h
r
ed
u
n
d
an
t
in
f
o
r
m
atio
n
o
n
th
e
b
asis
o
f
ce
r
tai
n
p
r
e
-
d
ef
i
n
ed
f
ac
to
r
.
T
h
er
ef
o
r
e,
ex
ec
u
tio
n
ti
m
e
i
s
m
o
r
e
co
m
p
ar
ed
to
o
th
er
s
.
E
d
g
e
-
b
ased
b
ein
g
t
h
e
m
o
s
t
f
r
eq
u
e
n
tl
y
ad
o
p
ted
tech
n
iq
u
e
th
at
es
s
en
tia
l
l
y
d
ep
en
d
s
o
n
eit
h
er
h
is
to
g
r
a
m
o
r
g
r
ad
ie
n
t
-
b
a
s
ed
in
f
o
r
m
atio
n
th
a
t in
cr
ea
s
es t
h
e
ac
cu
r
ac
y
co
m
p
ar
ed
to
r
eg
io
n
-
b
ased
.
Ho
w
ev
er
,
it
s
p
er
f
o
r
m
a
n
ce
is
n
ea
r
l
y
s
i
m
ila
r
to
th
r
esh
o
ld
-
b
ased
s
eg
m
e
n
tati
o
n
tech
n
iq
u
e.
T
h
e
p
r
i
m
e
r
ea
s
o
n
f
o
r
en
h
an
ce
d
ac
cu
r
ac
y
o
f
p
r
o
p
o
s
ed
s
y
s
te
m
ar
e
i)
n
ar
r
o
w
i
n
g
o
f
s
ea
r
ch
o
b
j
ec
ts
f
r
o
m
d
ataset
o
n
th
e
b
asis
o
f
s
i
m
ilar
it
y
s
co
r
e,
ii)
u
s
ag
e
o
f
r
eg
io
n
-
b
ased
s
h
ap
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
F
r
a
mewo
r
k
fo
r
p
r
o
g
r
es
s
ive
s
e
g
men
ta
tio
n
o
f c
h
est r
a
d
io
g
r
a
p
h
f
o
r
efficien
t d
ia
g
n
o
s
is
o
f in
ert…
(
S
a
vith
a
S
.
K
.
)
989
d
escr
ip
to
r
f
o
r
f
u
r
th
er
en
s
u
r
i
n
g
th
e
co
r
r
ec
t
s
elec
tio
n
o
f
l
u
n
g
r
eg
io
n
,
iii)
i
m
a
g
e
r
eg
is
tr
at
io
n
p
r
o
ce
s
s
in
teg
r
ated
w
it
h
co
n
to
u
r
ad
j
u
s
t
m
en
t
to
f
i
n
all
y
id
en
ti
f
y
t
h
e
l
u
n
g
r
e
g
io
n
f
o
llo
w
ed
b
y
s
p
i
n
e
i
n
ter
p
o
latio
n
,
an
d
iv
)
ca
p
ab
ilit
y
to
id
en
ti
f
y
t
h
e
ap
ical
a
n
d
co
s
t
o
p
h
r
en
ic
r
eg
io
n
s
.
Mo
r
eo
v
er
u
s
ag
e
o
f
f
ea
t
u
r
es
to
p
er
f
o
r
m
s
e
g
m
e
n
tat
io
n
f
u
r
th
er
r
ed
u
ce
s
th
e
s
ea
r
c
h
ti
m
e
f
o
r
s
i
m
ilar
o
b
j
ec
t c
au
s
in
g
r
ed
u
cti
o
n
in
ex
ec
u
tio
n
t
i
m
e
f
o
r
p
r
o
p
o
s
e
d
s
y
s
te
m
.
7.
CO
NCLU
SI
O
N
T
h
e
co
n
tr
ib
u
tio
n
o
f
ex
i
s
ti
n
g
liter
atu
r
es
to
w
ar
d
s
s
eg
m
e
n
tat
io
n
p
r
o
ce
s
s
in
c
h
e
s
t
x
-
r
a
y
h
as
n
o
t
y
et
ad
d
r
ess
ed
th
e
p
r
o
b
lem
a
s
s
o
cia
ted
w
it
h
t
h
e
p
o
s
itio
n
o
f
ca
p
tu
r
in
g
th
e
r
ad
io
g
r
ap
h
.
T
h
e
p
r
o
p
o
s
ed
s
tu
d
y
co
n
s
id
er
s
ce
r
tain
s
et
o
f
p
r
o
b
le
m
s
e.
g
.
i)
co
n
v
en
tio
n
al
p
o
s
itio
n
o
f
ca
p
tu
r
in
g
c
h
est
x
-
r
a
y
li
m
it
s
th
e
v
is
ib
il
it
y
o
f
ce
r
tai
n
in
er
t
o
b
j
ec
t
an
d
m
a
y
p
o
ten
t
ial
ac
ts
as
i
m
p
ed
i
m
e
n
t
i
n
s
eg
m
e
n
tat
io
n
p
r
o
ce
s
s
,
ii)
e
x
i
s
tin
g
s
e
g
m
e
n
tat
io
n
tech
n
iq
u
es
ar
e
m
o
r
e
in
cli
n
ed
in
u
s
i
n
g
iter
ativ
e
o
p
er
atio
n
th
at
o
f
f
er
s
ac
cu
r
ac
y
at
th
e
co
s
t
o
f
co
m
p
u
tatio
n
a
l
co
m
p
le
x
it
y
,
iii)
m
o
r
e
f
r
eq
u
e
n
t
l
y
e
x
is
ti
n
g
tec
h
n
iq
u
es e
.
g
.
r
u
le
b
ased
,
m
ac
h
i
n
e
-
lear
n
in
g
b
ase
d
,
th
r
esh
o
ld
-
b
ased
,
r
eg
io
n
-
b
ased
,
ed
g
e
-
b
ased
d
o
e
s
n
’
t
a
s
s
i
s
t
m
u
c
h
i
n
ad
d
r
ess
i
n
g
t
h
e
p
r
o
b
lem
o
f
p
r
o
g
r
ess
i
v
e
s
eg
m
e
n
tatio
n
.
A
s
i
m
p
le
a
n
d
n
o
v
el
p
r
o
g
r
ess
i
v
e
s
eg
m
e
n
tatio
n
p
r
o
ce
s
s
i
s
i
n
tr
o
d
u
ce
d
w
h
er
e
e
m
p
h
a
s
is
was
o
n
r
ed
u
ctio
n
o
f
iter
atio
n
s
to
w
ar
d
s
s
eg
m
e
n
tat
io
n
p
r
o
ce
s
s
.
T
h
e
s
tu
d
y
o
u
t
co
m
e
s
h
o
w
s
e
n
h
a
n
ce
d
ac
cu
r
ac
y
a
n
d
r
ed
u
ce
d
ex
ec
u
t
io
n
ti
m
e
as c
o
m
p
ar
ed
to
f
r
eq
u
en
t
l
y
u
s
ed
s
e
g
m
en
ta
tio
n
alg
o
r
ith
m
s
.
RE
F
E
R
E
NC
E
S
[1
]
Ca
m
p
il
h
o
,
M
.
Ka
m
e
l,
“
I
m
a
g
e
An
a
ly
sis
a
n
d
Re
c
o
g
n
it
io
n
:
7
t
h
In
t
e
rn
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
,
”
ICI
A
R
2
0
1
0
,
P
ó
v
o
a
d
e
V
a
rz
im
,
P
o
r
tu
g
a
l,
Ju
n
e
2
1
-
2
3
,
2
0
1
0
,
Pro
c
e
e
d
in
g
s”,
S
p
ri
n
g
e
r
S
c
ien
c
e
&
Bu
sin
e
ss
M
e
d
ia
,
p
p
.
4
6
0
,
2
0
1
0
.
[2
]
Da
v
id
A
tt
w
o
o
d
,
A
n
n
e
S
a
k
d
in
a
w
a
t,
“
X
-
Ra
y
s
a
n
d
Ex
tre
m
e
Ultr
a
v
io
let
Ra
d
iatio
n
:
P
ri
n
c
ip
les
a
n
d
A
p
p
li
c
a
ti
o
n
s
,
”
C
a
m
b
rid
g
e
Un
iv
e
rsit
y
P
re
ss
,
2
0
1
7
.
[3
]
S
.
V
it
u
lan
o
,
“
Im
a
g
e
:
E
-
lea
rn
in
g
,
Un
d
e
rsta
n
d
i
n
g
,
In
f
o
rm
a
ti
o
n
R
e
tri
e
v
a
l
a
n
d
M
e
d
ica
l
:
P
ro
c
e
e
d
in
g
s
o
f
th
e
F
irst
In
t
e
rn
a
ti
o
n
a
l
W
o
rk
sh
o
p
”
,
Ca
g
li
a
ri,
Italy
,
9
-
1
0
J
u
n
e
2
0
0
3
,
”
W
o
rl
d
S
c
ien
ti
f
ic,
2
0
1
0
.
[4
]
Ch
risto
p
h
e
r
Clark
e
,
A
n
th
o
n
y
Du
x
,
“
Ch
e
st X
-
ra
y
s f
o
r
M
e
d
ica
l
S
tu
d
e
n
ts
,
”
J
o
h
n
W
il
e
y
&
S
o
n
s
,
2
0
1
7
.
[5
]
T
.
Do
h
i,
I.
S
a
k
u
m
a
,
H.
L
iao
,
“
M
e
d
ica
l
Im
a
g
in
g
a
n
d
A
u
g
m
e
n
ted
Re
a
li
ty
:
4
th
In
tern
a
t
i
o
n
a
l
W
o
rk
sh
o
p
T
o
k
y
o
,
Ja
p
a
n
,
A
u
g
u
st 1
-
2
,
2
0
0
8
,
P
r
o
c
e
e
d
i
n
g
s,”
S
p
rin
g
e
r S
c
ien
c
e
&
Bu
sin
e
ss
M
e
d
ia
,
p
p
.
4
4
1
,
2
0
0
8
.
[6
]
A
ll
a
n
H.
G
o
ro
ll
,
A
lb
e
rt
G
.
M
u
ll
e
y
,
“
P
rim
a
r
y
Ca
r
e
M
e
d
icin
e
:
Off
ice
Ev
a
lu
a
ti
o
n
a
n
d
M
a
n
a
g
e
m
e
n
t
o
f
th
e
A
d
u
lt
P
a
ti
e
n
t
,
”
L
i
p
p
i
n
c
o
t
t
W
il
li
a
ms
&
W
il
k
in
s
,
2
0
0
9
.
[7
]
S
.
Ja
e
g
e
r,
A
.
Ka
r
a
rg
y
ris,
S
.
Ca
n
d
e
m
ir,
J.
S
ieg
e
l
m
a
n
,
L
.
F
o
li
o
,
S
.
A
n
tan
i,
a
n
d
G
.
T
h
o
m
a
,
“
A
u
to
m
a
t
ic
sc
re
e
n
in
g
f
o
r
tu
b
e
rc
u
l
o
sis
in
c
h
e
st
ra
d
io
g
ra
p
h
s:
a
su
rv
e
y
,
”
Qu
a
n
ti
t
a
ti
v
e
ima
g
i
n
g
i
n
me
d
icin
e
a
n
d
su
rg
e
ry
,
v
o
l.
3
,
n
o
.
2
,
pp.
8
9
,
2
0
1
3
.
[8
]
W
.
S
.
H.M
.
W
.
A
h
m
a
d
,
W
.
M
.
Di
y
a
n
a
W
.
Za
k
i,
a
n
d
M
.
F
.
A
.
F
a
u
z
i,
“
L
u
n
g
se
g
m
e
n
tatio
n
o
n
sta
n
d
a
rd
a
n
d
m
o
b
il
e
c
h
e
st
ra
d
io
g
ra
p
h
s
u
si
n
g
o
rien
te
d
G
a
u
ss
ian
d
e
riv
a
ti
v
e
s
f
il
ter
,
”
Bi
o
me
d
ica
l
e
n
g
in
e
e
rin
g
o
n
li
n
e
,
v
o
l.
1
4
,
n
o
.
1
,
pp.
2
0
,
2
0
1
5
.
[9
]
P
.
Ca
m
p
a
d
e
ll
i
a
n
d
E.
Ca
sira
g
h
i,
“
L
u
n
g
f
ield
se
g
m
e
n
tatio
n
in
d
ig
it
a
l
p
o
ste
ro
-
a
n
teri
o
r
c
h
e
st
ra
d
io
g
ra
p
h
s,”
Pa
tt
e
rn
Rec
o
g
n
it
io
n
a
n
d
Ima
g
e
An
a
lys
is
,
p
p
.
7
3
6
-
7
4
5
,
2
0
0
5
.
[1
0
]
S
.
K.
S
a
v
it
h
a
a
n
d
N.
C.
Na
v
e
e
n
,
“
S
tu
d
y
f
o
r
A
ss
e
ss
in
g
th
e
Ad
v
a
n
c
e
m
e
n
t
o
f
I
m
a
g
in
g
T
e
c
h
n
i
q
u
e
s
in
Ch
e
st
Ra
d
io
g
ra
p
h
ic Im
a
g
e
s,”
Co
mm
u
n
ica
ti
o
n
s
o
n
A
p
p
li
e
d
e
lec
tro
n
ics
,
v
o
l.
4
,
p
p
.
2
2
-
3
4
,
2
0
1
7
.
[1
1
]
J.
W
a
n
g
a
n
d
H.
G
u
o
,
“
A
u
to
m
a
ti
c
A
p
p
ro
a
c
h
f
o
r
L
u
n
g
S
e
g
m
e
n
tatio
n
w
it
h
Ju
x
ta
-
P
leu
ra
l
No
d
u
les
f
ro
m
T
h
o
ra
c
ic
CT
Ba
se
d
o
n
C
o
n
t
o
u
r
T
ra
c
in
g
a
n
d
C
o
rre
c
ti
o
n
,
”
Co
mp
u
ta
ti
o
n
a
l
a
n
d
m
a
th
e
ma
t
ica
l
me
th
o
d
s
in
me
d
ici
n
e
,
2
0
1
6
.
[1
2
]
W
.
Da
i,
J.
Do
y
le,
X
.
L
ian
g
,
H.
Z
h
a
n
g
,
N.
D
o
n
g
,
Y.
L
i,
a
n
d
E.
P
.
X
in
g
,
“
S
CA
N:
S
tru
c
tu
re
C
o
rre
c
ti
n
g
A
d
v
e
rs
a
rial
Ne
tw
o
rk
f
o
r
C
h
e
st X
-
Ra
y
s Or
g
a
n
S
e
g
m
e
n
tatio
n
,
”
AR
XIV
Pre
p
ri
n
t
AR
XIV
:1
7
0
3
.
0
8
7
7
0
,
2
0
1
7
.
[1
3
]
Z.
S
h
i,
J.
M
a
,
M
.
Zh
a
o
,
Y.
L
iu
,
Y.
F
e
n
g
,
M
.
Zh
a
n
g
,
L
.
He
,
a
n
d
K.
S
u
z
u
k
i,
“
M
a
n
y
is
B
e
tt
e
r
th
a
n
On
e
:
A
n
In
teg
ra
ti
o
n
o
f
M
u
lt
ip
le
S
im
p
le
S
trate
g
ies
f
o
r
A
c
c
u
ra
te
L
u
n
g
S
e
g
m
e
n
tatio
n
in
CT
Im
a
g
e
s
,
”
Bi
o
me
d
Res
e
a
rc
h
In
ter
n
a
t
io
n
a
l
,
2
0
1
6
.
[1
4
]
H.
Ha
n
,
L
.
L
i,
F
.
Ha
n
,
B.
S
o
n
g
,
W
.
M
o
o
re
a
n
d
Z.
L
ian
g
,
“
F
a
st
a
n
d
A
d
a
p
ti
v
e
De
tec
ti
o
n
o
f
P
u
l
m
o
n
a
ry
No
d
u
les
i
n
T
h
o
ra
c
ic
CT
I
m
a
g
e
s
Us
in
g
a
H
iera
rc
h
i
c
a
l
V
e
c
to
r
Qu
a
n
ti
z
a
ti
o
n
S
c
h
e
m
e
,
”
in
IEE
E
J
o
u
rn
a
l
o
f
B
io
me
d
ica
l
a
n
d
He
a
lt
h
I
n
fo
rm
a
ti
c
s
,
v
o
l.
1
9
,
n
o
.
2
,
p
p
.
6
4
8
-
6
5
9
,
M
a
rc
h
2
0
1
5
.
[1
5
]
S
.
S
h
e
n
,
A
.
A
.
T
.
Bu
i,
J.
C
o
n
g
,
a
n
d
W
.
Hs
u
,
“
A
n
a
u
to
m
a
ted
lu
n
g
se
g
m
e
n
tatio
n
a
p
p
r
o
a
c
h
u
si
n
g
b
i
d
irec
ti
o
n
a
l
c
h
a
i
n
c
o
d
e
s to
im
p
ro
v
e
n
o
d
u
le d
e
tec
ti
o
n
a
c
c
u
ra
c
y
,
”
Co
mp
u
ter
s in
b
i
o
lo
g
y
a
n
d
me
d
icin
e
, v
o
l.
5
7
,
p
p
.
1
3
9
-
1
4
9
,
2
0
1
5
.
[1
6
]
S
-
H.
Ch
a
e
,
D.
M
o
o
n
,
D.
G
.
L
e
e
,
a
n
d
S
.
B.
P
a
n
,
“
M
e
d
ica
l
im
a
g
e
s
e
g
m
e
n
tatio
n
f
o
r
m
o
b
il
e
e
lec
tro
n
i
c
p
a
ti
e
n
t
c
h
a
rts
us
in
g
n
u
m
e
rica
l
m
o
d
e
li
n
g
o
f
Io
T
,
”
J
o
u
rn
a
l
o
f
Ap
p
li
e
d
M
a
th
e
ma
ti
c
s
,
p
p
.
8
,
2
0
1
4
.
[1
7
]
G
.
G
i
ll
,
M
.
T
o
e
w
s
a
n
d
R.
R.
Be
i
c
h
e
l,
“
Ro
b
u
st
in
it
ial
iza
ti
o
n
o
f
a
c
ti
v
e
sh
a
p
e
m
o
d
e
ls
f
o
r
lu
n
g
se
g
m
e
n
tatio
n
in
CT
sc
a
n
s:
a
f
e
a
tu
re
-
b
a
se
d
a
tl
a
s ap
p
ro
a
c
h
,
”
J
o
u
rn
a
l
o
f
Bi
o
me
d
ica
l
Ima
g
in
g
,
p
p
.
7
,
2
0
1
4
.
[1
8
]
T
.
A
.
Ng
o
a
n
d
G
.
C
a
rn
e
iro
,
“
L
u
n
g
se
g
m
e
n
tatio
n
in
c
h
e
st
ra
d
io
g
ra
p
h
s
u
sin
g
d
istan
c
e
re
g
u
lariz
e
d
le
v
e
l
se
t
a
n
d
d
e
e
p
-
stru
c
tu
re
d
lea
rn
i
n
g
a
n
d
in
f
e
re
n
c
e
,
”
2
0
1
5
IEE
E
In
ter
n
a
ti
o
n
a
l
Co
n
fe
re
n
c
e
o
n
Ima
g
e
Pr
o
c
e
ss
in
g
(
ICIP)
,
Qu
e
b
e
c
Cit
y
,
QC,
2
0
1
5
,
p
p
.
2
1
4
0
-
2
1
4
3
.
[1
9
]
P
.
P
.
R
.
F
i
lh
o
,
P
.
C.
C
o
rtez
,
a
n
d
V
.
H.
C.
d
.
A
lb
u
q
u
e
r
q
u
e
,
“
3
D
se
g
m
e
n
tatio
n
a
n
d
v
isu
a
li
z
a
ti
o
n
o
f
lu
n
g
a
n
d
it
s
stru
c
tu
re
s
u
si
n
g
CT
i
m
a
g
e
s o
f
th
e
th
o
ra
x
,
”
J
o
u
r
n
a
l
o
f
B
io
me
d
ica
l
S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
v
o
l.
6
,
n
o
.
1
1
,
p
p
.
1
0
9
9
,
2
0
1
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
9
,
No
.
2
,
A
p
r
il 2
0
1
9
:
9
8
2
-
9
9
1
990
[2
0
]
J.
L
-
Ru
iz,
J.
M
-
S
á
n
c
h
e
z
,
M
.
C.
B
-
Ju
m
il
la,
M
.
M
-
Lara
,
R.
V
-
M
o
n
e
d
e
ro
,
a
n
d
J.
L
.
S
-
G
ó
m
e
z
,
“
A
u
to
m
a
ti
c
i
m
a
g
e
-
b
a
se
d
se
g
m
e
n
tat
io
n
o
f
th
e
h
e
a
rt
f
ro
m
C
T
sc
a
n
s,”
EURA
S
IP
J
o
u
rn
a
l
o
n
Im
a
g
e
a
n
d
Vi
d
e
o
Pro
c
e
ss
in
g
,
n
o
.
1
,
pp.
52
,
2
0
1
4
.
[2
1
]
A
.
F
a
r
a
g
,
H.
E.
A
.
E.
M
u
n
im
,
J.
H.
G
r
a
h
a
m
a
n
d
A
.
A
.
F
a
r
a
g
,
“
A
No
v
e
l
A
p
p
ro
a
c
h
f
o
r
L
u
n
g
No
d
u
les
S
e
g
m
e
n
tatio
n
in
Ch
e
st
CT
Us
in
g
L
e
v
e
l
S
e
ts,
”
in
I
EE
E
T
ra
n
s
a
c
ti
o
n
s
o
n
Ima
g
e
Pro
c
e
ss
in
g
,
v
o
l
.
2
2
,
n
o
.
1
2
,
p
p
.
5
2
0
2
-
5
2
1
3
,
De
c
.
2
0
1
3
.
[2
2
]
L
a
ss
e
n
,
E.
M
.
v
a
n
R
ik
x
o
o
rt,
M
.
S
c
h
m
id
t,
S
.
Ke
rk
stra
,
B.
v
a
n
G
in
n
e
k
e
n
a
n
d
J.
M
.
Ku
h
n
i
g
k
,
“
A
u
to
m
a
ti
c
S
e
g
m
e
n
tatio
n
o
f
th
e
P
u
lm
o
n
a
ry
L
o
b
e
s
F
ro
m
Ch
e
st
C
T
S
c
a
n
s
Ba
se
d
o
n
F
issu
re
s,
V
e
ss
e
ls,
a
n
d
Bro
n
c
h
i
,
”
in
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
M
e
d
ica
l
Im
a
g
i
n
g
,
v
o
l.
3
2
,
n
o
.
2
,
p
p
.
2
1
0
-
2
2
2
,
F
e
b
.
2
0
1
3
.
[2
3
]
J.
F
e
u
ln
e
r,
S
.
K.
Zh
o
u
,
M
.
Ha
m
m
o
n
,
J
.
Ho
r
n
e
g
g
e
r,
a
n
d
D.
Co
m
a
n
i
c
iu
,
“
Ly
m
p
h
n
o
d
e
d
e
tec
ti
o
n
a
n
d
s
e
g
m
e
n
tatio
n
in
c
h
e
st
CT
d
a
ta
u
sin
g
d
isc
rim
in
a
ti
v
e
lea
rn
in
g
a
n
d
a
sp
a
ti
a
l
p
r
io
r
,
”
M
e
d
ica
l
im
a
g
e
a
n
a
lys
is
,
v
o
l.
1
7
,
n
o
.
2
,
pp.
2
5
4
-
2
7
0
,
2
0
1
3
.
[2
4
]
J.
F
e
u
ln
e
r,
S
.
K.
Zh
o
u
,
M
.
Ha
m
m
o
n
,
J
.
Ho
rn
e
g
g
e
r,
a
n
d
D.
Co
m
a
n
iciu
,
“
S
e
g
m
e
n
tatio
n
b
a
se
d
f
e
a
t
u
re
s
f
o
r
ly
m
p
h
n
o
d
e
d
e
tec
ti
o
n
f
ro
m
3
-
d
c
h
e
st
c
t
,
”
In
I
n
ter
n
a
ti
o
n
a
l
W
o
rk
sh
o
p
o
n
M
a
c
h
in
e
L
e
a
rn
i
n
g
in
M
e
d
ica
l
Im
a
g
i
n
g
,
S
p
ri
n
g
e
r
,
Be
rli
n
,
He
id
e
l
b
e
rg
,
p
p
.
9
1
-
9
9
,
2
0
1
1
.
[2
5
]
A
.
C.
Jira
p
a
tn
a
k
u
l,
Y.D.
M
u
lma
n
,
A
.
P
.
Re
e
v
e
s,
D.
F
.
Ya
n
k
e
lev
it
z
,
a
n
d
C.
I.
He
n
sc
h
k
e
,
“
S
e
g
m
e
n
tatio
n
o
f
ju
x
tap
leu
ra
l
p
u
lm
o
n
a
ry
n
o
d
u
les
u
sin
g
a
ro
b
u
st su
rf
a
c
e
e
sti
m
a
te
,
”
J
o
u
rn
a
l
o
f
Bi
o
me
d
ica
l
Ima
g
in
g
,
p
p
.
1
5
,
2
0
1
1
[2
6
]
P
.
L
o
,
J.
G
o
ld
in
,
D
.
Oria
,
A
.
Ba
n
o
la,
a
n
d
M
.
Bro
w
n
,
“
Histo
ric
a
u
t
o
m
a
ted
lu
n
g
se
g
m
e
n
tatio
n
m
e
th
o
d
:
P
e
rf
o
rm
a
n
c
e
o
n
L
OLA
1
1
d
a
ta
se
t
,
”
I
n
Pr
o
c
.
4
th
In
ter
n
.
M
ICCAI
W
o
rk
sh
o
p
o
n
P
u
lmo
n
a
ry
Ima
g
e
A
n
a
lys
is
,
T
o
ro
n
to
,
Ca
n
a
d
a
,
p
p
.
2
5
7
-
2
6
0
.
2
0
1
1
.
[2
7
]
N.
Zh
o
u
,
T
.
Ya
n
g
,
a
n
d
S
.
Z
h
a
n
g
,
“
A
n
i
m
p
ro
v
e
d
F
CM
m
e
d
ica
l
ima
g
e
se
g
m
e
n
tatio
n
a
lg
o
rit
h
m
b
a
se
d
o
n
M
M
T
D
,
”
Co
mp
u
t
a
ti
o
n
a
l
a
n
d
ma
t
h
e
ma
ti
c
a
l
me
th
o
d
s
in
me
d
icin
e
,
2
0
1
4
.
[2
8
]
S
.
Ji,
B.
W
e
i,
Z.
Yu
,
G
.
Ya
n
g
,
a
n
d
Y.
Yi
n
,
“
A
n
e
w
m
u
lt
istag
e
m
e
d
ica
l
se
g
m
e
n
tatio
n
m
e
th
o
d
b
a
se
d
o
n
s
u
p
e
r
p
ix
e
l
a
n
d
f
u
z
z
y
c
lu
ste
rin
g
,
”
Co
mp
u
ta
ti
o
n
a
l
a
n
d
ma
t
h
e
ma
ti
c
a
l
me
th
o
d
s i
n
me
d
icin
e
,
2
0
1
4
[2
9
]
J.
P
u
,
C.
F
u
h
rm
a
n
,
W
.
F
.
G
o
o
d
,
F
.
C.
S
c
iu
rb
a
a
n
d
D.
G
u
r,
“
A
Diff
e
r
e
n
ti
a
l
Ge
o
m
e
tri
c
A
p
p
ro
a
c
h
to
A
u
to
m
a
ted
S
e
g
m
e
n
tatio
n
o
f
Hu
m
a
n
A
ir
w
a
y
T
re
e
,
”
in
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
M
e
d
ica
l
Ima
g
in
g
,
v
o
l.
3
0
,
n
o
.
2
,
p
p
.
2
6
6
-
2
7
8
,
F
e
b
.
2
0
1
1
.
[3
0
]
J.
P
u
e
t
a
l.
,
“
P
u
lm
o
n
a
ry
L
o
b
e
S
e
g
m
e
n
tatio
n
in
CT
Ex
a
m
in
a
ti
o
n
s
Us
i
n
g
Im
p
li
c
it
S
u
rf
a
c
e
F
it
ti
n
g
,
”
in
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
M
e
d
ica
l
Im
a
g
i
n
g
,
v
o
l.
2
8
,
n
o
.
1
2
,
p
p
.
1
9
8
6
-
1
9
9
6
,
De
c
.
2
0
0
9
.
[3
1
]
M
.
Ce
y
lan
,
Y.
Oz
b
a
y
,
O.N.
U
c
a
n
a
n
d
E
.
Yild
ir
im
,
“
A
n
o
v
e
l
m
e
t
h
o
d
f
o
r
l
u
n
g
se
g
m
e
n
tatio
n
o
n
c
h
e
st
CT
i
m
a
g
e
s:
c
o
m
p
lex
-
v
a
lu
e
d
a
rti
f
icia
l
n
e
u
ra
l
n
e
tw
o
rk
w
it
h
c
o
m
p
lex
w
a
v
e
l
e
t
tran
sf
o
r
m
,
”
T
u
rk
ish
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
i
n
e
e
rin
g
&
Co
mp
u
ter
S
c
ien
c
e
s
,
v
o
l.
1
8
,
n
o
.
4
,
p
p
.
6
1
3
-
6
2
4
,
2
0
1
0
.
[3
2
]
H.
S
h
ik
a
ta,
G
.
M
c
L
e
n
n
a
n
,
E.
A
.
Ho
ffm
a
n
,
a
n
d
M
.
S
o
n
k
a
,
“
S
e
g
m
e
n
tatio
n
o
f
p
u
lm
o
n
a
ry
v
a
s
c
u
lar
t
re
e
s
f
ro
m
th
o
ra
c
ic
3
D CT
im
a
g
e
s
,
”
J
o
u
rn
a
l
o
f
Bi
o
m
e
d
ica
l
Ima
g
in
g
,
p
p
.
2
4
,
2
0
0
9
.
[3
3
]
Ch
e
n
,
Q
-
H.
Zo
u
,
W
-
S
.
C
h
e
n
,
a
n
d
Y.
L
i,
“
A
f
a
st
re
g
io
n
-
b
a
se
d
se
g
m
e
n
tatio
n
m
o
d
e
l
w
it
h
G
a
u
s
sia
n
k
e
rn
e
l
o
f
f
ra
c
ti
o
n
a
l
o
r
d
e
r
,
”
A
d
v
a
n
c
e
s in
M
a
th
e
ma
ti
c
a
l
Ph
y
sic
s
,
2
0
1
3
.
[3
4
]
H.
Jia
n
g
,
B.
H
e
,
D.
F
a
n
g
,
Z.
M
a
,
B.
Ya
n
g
,
a
n
d
L
.
Zh
a
n
g
,
“
A
re
g
i
o
n
g
ro
w
in
g
v
e
ss
e
l
se
g
m
e
n
tatio
n
a
lg
o
rit
h
m
b
a
se
d
o
n
s
p
e
c
tru
m
in
f
o
r
m
a
ti
o
n
,
”
Co
mp
u
ta
ti
o
n
a
l
a
n
d
ma
th
e
ma
ti
c
a
l
me
th
o
d
s i
n
me
d
ici
n
e
,
2
0
1
3
.
[3
5
]
J.
L
iu
,
J.
Zh
e
n
g
,
Q.
T
a
n
g
,
a
n
d
W
.
Jin
,
“
M
in
im
u
m
e
rro
r
th
re
sh
o
l
d
in
g
se
g
m
e
n
tatio
n
a
lg
o
rit
h
m
b
a
se
d
o
n
3
d
g
ra
y
sc
a
l
e
h
isto
g
ra
m
,
”
M
a
th
e
ma
ti
c
a
l
Pro
b
le
ms
in
En
g
i
n
e
e
rin
g
,
2
0
1
4
.
[3
6
]
E.
T
.
En
ik
o
v
a
n
d
R.
A
n
to
n
,
“
Im
a
g
e
se
g
m
e
n
tatio
n
a
n
d
a
n
a
ly
sis
o
f
flex
io
n
-
e
x
ten
sio
n
ra
d
io
g
ra
p
h
s
o
f
c
e
rv
ica
l
sp
in
e
s
,
”
J
o
u
rn
a
l
o
f
me
d
ica
l
e
n
g
in
e
e
rin
g
,
2
0
1
4
.
[3
7
]
W
.
Wan
g
,
L
.
Du
a
n
,
a
n
d
Y.
W
a
n
g
,
“
F
a
st I
m
a
g
e
S
e
g
m
e
n
tatio
n
Us
in
g
Tw
o
-
Di
m
e
n
sio
n
a
l
Otsu
Ba
se
d
o
n
Esti
m
a
ti
o
n
o
f
Distrib
u
ti
o
n
A
lg
o
rit
h
m
,
”
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
,
2
0
1
7
.
[3
8
]
W
.
Jia
n
g
,
Z.
Zh
o
u
,
X
.
Di
n
g
,
X
.
De
n
g
,
L
.
Zo
u
,
a
n
d
B
.
L
i,
“
L
e
v
e
l
S
e
t
Ba
se
d
Hip
p
o
c
a
m
p
u
s
S
e
g
m
e
n
tatio
n
i
n
M
R
Im
a
g
e
s
w
it
h
I
m
p
ro
v
e
d
In
it
ializa
ti
o
n
Us
in
g
Re
g
io
n
G
ro
w
in
g
,
”
Co
mp
u
t
a
ti
o
n
a
l
a
n
d
ma
th
e
ma
t
i
c
a
l
me
th
o
d
s
in
me
d
icin
e
,
2
0
1
7
.
[3
9
]
J.
Zh
o
n
g
a
n
d
H.
W
u
,
“
Ev
o
lu
ti
o
n
a
ry
G
a
m
e
A
l
g
o
rit
h
m
f
o
r
I
m
a
g
e
S
e
g
m
e
n
tatio
n
,
”
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
Co
mp
u
ter
E
n
g
in
e
e
rin
g
,
2
0
1
7
.
[4
0
]
Y.
L
u
o
,
L
.
L
iu
,
Q.
Hu
a
n
g
,
a
n
d
X
.
L
i,
“
A
No
v
e
l
S
e
g
m
e
n
tatio
n
A
p
p
ro
a
c
h
Co
m
b
in
in
g
Re
g
io
n
-
a
n
d
Ed
g
e
-
Ba
se
d
In
f
o
rm
a
ti
o
n
f
o
r
Ultras
o
u
n
d
Im
a
g
e
s
,
”
Bi
o
M
e
d
re
se
a
rc
h
in
ter
n
a
ti
o
n
a
l
,
2
0
1
7
.
[4
1
]
S
.
L
iu
,
X
.
S
h
e
n
,
Y.
F
e
n
g
,
a
n
d
H.
Ch
e
n
,
“
A
No
v
e
l
Histo
g
ra
m
Re
g
io
n
M
e
rg
in
g
Ba
se
d
M
u
lt
it
h
re
sh
o
l
d
S
e
g
m
e
n
tatio
n
A
l
g
o
rit
h
m
f
o
r
M
R
Bra
in
Im
a
g
e
s
,
”
In
ter
n
a
ti
o
n
a
l
jo
u
rn
a
l
o
f
b
i
o
me
d
ica
l
ima
g
in
g
,
2
0
1
7
.
[4
2
]
R.
M
e
c
h
re
z
,
J.
G
o
ld
b
e
rg
e
r,
a
n
d
H.
G
re
e
n
sp
a
n
,
“
P
a
tch
-
b
a
se
d
se
g
m
e
n
tatio
n
w
it
h
sp
a
ti
a
l
c
o
n
siste
n
c
y
:
a
p
p
li
c
a
ti
o
n
to
M
S
les
io
n
s in
b
ra
i
n
M
RI
,
”
J
o
u
rn
a
l
o
f
Bi
o
me
d
ica
l
Im
a
g
i
n
g
,
v
o
l.
3
,
2
0
1
6
.
[4
3
]
W
.
Co
n
g
,
J.
S
o
n
g
,
K.
L
u
a
n
,
H.
L
ian
g
,
L
.
W
a
n
g
,
X
.
M
a
,
a
n
d
J.
L
i,
“
A
M
o
d
if
ied
Bra
in
M
R
I
m
a
g
e
S
e
g
m
e
n
tatio
n
a
n
d
Bias
F
ield
Esti
m
a
ti
o
n
M
o
d
e
l
Ba
se
d
o
n
L
o
c
a
l
a
n
d
G
lo
b
a
l
In
f
o
r
m
a
t
io
n
,
”
Co
m
p
u
t
a
ti
o
n
a
l
a
n
d
ma
th
e
m
a
ti
c
a
l
me
th
o
d
s
in
me
d
icin
e
,
2
0
1
6
.
[4
4
]
Ra
z
ieh
A
k
h
a
v
a
n
,
Ka
ri
m
F
a
e
z
,
“
A No
v
e
l
Re
ti
n
a
l
Blo
o
d
V
e
ss
e
l
S
e
g
m
e
n
tatio
n
A
lg
o
rit
h
m
u
sin
g
F
u
z
z
y
se
g
m
e
n
tatio
n
,
’
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
E
lec
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
o
l.
4
,
n
o
.
4
,
p
p
.
5
6
1
-
5
7
2
,
A
u
g
u
st 2
0
1
4
.
[4
5
]
No
h
a
A
.
S
e
a
d
a
,
S
a
fw
a
t
H
a
m
a
d
,
M
o
sta
f
a
G
.
M
.
M
o
sta
f
a
,
“
M
o
d
e
l
-
b
a
s
e
d
A
u
to
m
a
ti
c
S
e
g
m
e
n
tatio
n
o
f
A
sc
e
n
d
in
g
A
o
rta
f
ro
m
M
u
lt
im
o
d
a
li
ty
M
e
d
ic
a
l
Da
ta
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
Co
mp
u
ter
E
n
g
i
n
e
e
rin
g
(
IJ
ECE
)
,
v
o
l.
6
,
n
o
.
6
,
p
p
.
3
1
6
1
-
3
1
7
3
,
De
c
e
m
b
e
r
2
0
1
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
F
r
a
mewo
r
k
fo
r
p
r
o
g
r
es
s
ive
s
e
g
men
ta
tio
n
o
f c
h
est r
a
d
io
g
r
a
p
h
f
o
r
efficien
t d
ia
g
n
o
s
is
o
f in
ert…
(
S
a
vith
a
S
.
K
.
)
991
[4
6
]
T
.
J
y
o
th
irma
y
i,
K.
S
rin
iv
a
sa
Ra
o
,
P
.
S
r
i
n
iv
a
sa
Ra
o
,
Ch
.
S
a
ty
a
n
a
ra
y
a
n
a
,
“
I
m
a
g
e
S
e
g
m
e
n
tatio
n
Ba
se
d
o
n
Do
u
b
ly
T
ru
n
c
a
ted
G
e
n
e
ra
li
z
e
d
L
a
p
lac
e
M
ix
tu
re
M
o
d
e
l
a
n
d
K
M
e
a
n
s
Cl
u
ste
rin
g
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
Co
mp
u
ter
E
n
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
o
l.
6
,
n
o
.
5
,
p
p
.
2
1
8
8
-
2
1
9
6
,
o
c
to
b
e
r
2
0
1
6
.
[4
7
]
S
a
v
it
h
a
S
.
K
a
n
d
N.
C
.
Na
v
e
e
n
,
“
A
lg
o
rit
h
m
f
o
r
p
re
-
p
ro
c
e
ss
in
g
c
h
e
st
-
x
-
ra
y
u
sin
g
m
u
lt
i
-
lev
e
l
e
n
h
a
n
c
e
m
e
n
t
o
p
e
ra
ti
o
n
,
”
2
0
1
6
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
W
ire
les
s
Co
mm
u
n
ica
ti
o
n
s,
S
i
g
n
a
l
Pr
o
c
e
ss
in
g
a
n
d
Ne
two
rk
in
g
(
W
iS
PNE
T
)
,
Ch
e
n
n
a
i,
2
0
1
6
,
p
p
.
2
1
8
2
-
2
1
8
6
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
S
a
v
ith
a
S
.
K
.
,
w
o
rk
in
g
a
s
A
s
sis
tan
t
P
r
o
f
e
ss
o
r
in
Ba
n
g
a
lo
re
In
sti
tu
te
o
f
T
e
c
h
n
o
lo
g
y
,
B
e
n
g
a
lu
ru
,
Ka
rn
a
tak
a
in
Co
m
p
u
ter
S
c
ien
c
e
De
p
a
rtme
n
t.
S
h
e
h
a
s
f
in
ish
e
d
h
e
r
B.
E
f
ro
m
B
a
n
g
a
lo
re
Un
iv
e
rsit
y
,
In
d
ia.
S
h
e
h
a
s d
o
n
e
h
e
r
M
.
T
e
c
h
f
ro
m
Ku
v
e
m
p
u
Un
iv
e
rsity
,
In
d
ia.
S
h
e
is
p
u
rs
u
in
g
P
h
D
f
ro
m
V
T
U,
Be
lag
a
v
i,
In
d
ia.
He
r
A
re
a
o
f
In
tere
st
is
in
Dig
it
a
l
Im
a
g
e
P
ro
c
e
ss
in
g
,
Da
tab
a
se
M
a
n
a
g
e
m
e
n
t
S
y
st
e
m
a
n
d
Da
ta
M
in
in
g
.
S
h
e
h
a
s
p
u
b
slish
e
d
6
p
a
p
e
rs
i
n
n
a
ti
o
n
a
l
a
n
d
in
tern
a
ti
o
n
j
o
u
r
n
a
ls
a
n
d
Co
n
f
e
re
n
c
e
s.
S
h
e
h
a
s aro
u
n
d
1
2
y
e
a
rs o
f
a
c
a
d
e
m
ic ex
p
e
rien
c
e
.
Dr
.
Na
v
e
e
n
N
.
C
.
,
w
o
rk
in
g
a
s
P
ro
f
e
ss
o
r,
De
p
a
rtm
e
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
&
E
n
g
in
e
e
rin
g
,
JSS
AT
E,
Be
n
g
a
lu
ru
,
In
d
ia.
He
h
a
s
c
o
m
p
lete
d
B.
E
f
ro
m
Ba
n
g
a
lo
re
Un
iv
e
rsit
y
,
In
d
ia.
M
.
E
f
ro
m
Ba
n
g
a
lo
re
Un
iv
e
rsit
y
,
In
d
ia.
He
h
a
s
c
o
m
p
lete
d
h
is
P
h
D
f
ro
m
S
R
M
Un
iv
e
rsit
y
,
Ch
e
n
n
a
i,
In
d
ia.
He
h
a
s
a
ro
u
n
d
1
6
y
e
a
rs
o
f
a
c
a
n
d
e
m
ic
e
x
p
e
rien
c
e
a
n
d
3
y
e
a
rs
o
f
in
d
u
strial
e
x
p
e
rien
c
e
.
He
h
a
s
p
u
b
li
sh
e
d
1
2
p
a
p
e
rs i
n
n
a
ti
o
n
a
l
a
n
d
i
n
tern
a
ti
o
n
jo
u
rn
a
ls
a
n
d
c
o
n
f
e
re
n
c
e
s.
Evaluation Warning : The document was created with Spire.PDF for Python.