I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
pu
t
er
E
ng
ineering
(
I
J
E
CE
)
Vo
l.
15
,
No
.
5
,
Octo
b
er
20
25
,
p
p
.
4
9
1
6
~
4
9
3
2
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
15
i
5
.
pp
4
9
1
6
-
4
9
3
2
4916
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Fuzzy clus
tering
o
ptimiza
tion ba
se
d artificial
bee
co
lo
ny
a
lg
o
rithm f
o
r bra
in ma
g
netic
r
eso
na
nce
ima
g
ing
ima
g
e seg
menta
tion
Cha
k
ir
M
o
k
hta
ri,
M
o
ha
mm
ed
Deba
k
la
,
B
o
ud
j
ela
l Mef
t
a
h
LI
S
Y
S
La
b
o
r
a
t
o
r
y
,
D
e
p
a
r
t
m
e
n
t
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
,
F
a
c
u
l
t
y
o
f
E
x
a
c
t
S
c
i
e
n
c
e
s,
U
n
i
v
e
r
s
i
t
y
M
u
s
t
a
p
h
a
S
t
a
m
b
o
u
l
i
o
f
M
a
sc
a
r
a
,
M
a
s
c
a
r
a
,
A
l
g
e
r
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Feb
1
1
,
2
0
2
5
R
ev
is
ed
J
u
n
5
,
2
0
2
5
Acc
ep
ted
J
u
n
3
0
,
2
0
2
5
In
b
ra
in
m
a
g
n
e
ti
c
re
so
n
a
n
c
e
ima
g
in
g
(
M
RI)
a
n
a
ly
sis,
ima
g
e
c
lu
ste
rin
g
i
s
re
g
a
rd
e
d
a
s
o
n
e
o
f
th
e
m
o
st
c
ru
c
ial
tas
k
s.
It
is
fre
q
u
e
n
tl
y
e
m
p
lo
y
e
d
t
o
e
stim
a
te
a
n
d
v
isu
a
li
z
e
b
ra
in
a
n
a
to
m
ica
l
stru
c
tu
re
s,
id
e
n
t
ify
p
a
th
o
l
o
g
ica
l
re
g
io
n
s,
a
n
d
a
ss
ist
in
g
u
id
i
n
g
s
u
r
g
ica
l
p
ro
c
e
d
u
re
s.
F
u
z
z
y
c
-
m
e
a
n
s
a
lg
o
rit
h
m
(F
CM
)
is
wid
e
l
y
u
se
d
in
t
h
e
M
R
I
ima
g
e
se
g
m
e
n
tatio
n
p
ro
c
e
ss
.
H
o
we
v
e
r
,
it
h
a
s
b
e
e
n
se
v
e
ra
l
we
a
k
n
e
ss
e
s
su
c
h
a
s
n
o
ise
se
n
si
ti
v
i
ty
,
stu
c
k
in
lo
c
a
l
o
p
ti
m
u
m
a
n
d
iss
u
e
s
with
p
a
ra
m
e
ters
in
it
ializa
ti
o
n
.
T
o
a
d
d
re
ss
th
e
se
F
CM
p
ro
b
lem
s
,
th
is
p
a
p
e
r
p
re
se
n
ts
a
n
o
v
e
l
fu
z
z
y
o
p
t
imiz
a
ti
o
n
m
e
th
o
d
th
a
t
e
n
h
a
n
c
e
s
b
ra
in
M
RI
ima
g
e
se
g
m
e
n
tatio
n
b
y
in
te
g
ra
ti
n
g
th
e
a
rt
ifi
c
ial
b
e
e
c
o
lo
n
y
(ABC)
a
lg
o
rit
h
m
with
F
CM
c
lu
ste
rin
g
tec
h
n
i
q
u
e
s.
Th
e
p
ro
p
o
se
d
m
e
th
o
d
se
e
k
s
t
o
o
p
ti
m
ize
m
u
lt
ip
le
F
CM
p
a
ra
m
e
ters
sim
u
lt
a
n
e
o
u
sl
y
,
in
c
lu
d
in
g
t
h
e
o
b
jec
ti
v
e
fu
n
c
ti
o
n
,
n
u
m
b
e
r
o
f
c
l
u
ste
rs,
a
n
d
c
lu
st
e
r
c
e
n
ter
v
a
lu
e
s.
T
h
e
m
e
th
o
d
wa
s
e
v
a
lu
a
te
d
o
n
b
o
t
h
sim
u
late
d
a
n
d
c
li
n
ica
l
b
ra
in
M
R
ima
g
e
s,
with
a
n
e
m
p
h
a
sis
o
n
se
g
m
e
n
ti
n
g
wh
it
e
m
a
tt
e
r,
g
re
y
m
a
tt
e
r,
a
n
d
c
e
re
b
ro
sp
in
a
l
flu
i
d
re
g
i
o
n
s.
Ex
p
e
rime
n
tal
re
su
lt
s
d
e
m
o
n
stra
te
sig
n
ifi
c
a
n
t
imp
ro
v
e
m
e
n
ts
in
se
g
m
e
n
tati
o
n
a
c
c
u
ra
c
y
,
a
c
h
iev
in
g
a
Ja
c
c
a
rd
sim
il
a
rit
y
(JS)
o
f
n
e
a
rly
1
,
a
p
a
rti
ti
o
n
c
o
e
fficie
n
t
in
d
e
x
(
P
CI)
o
f
0
.
9
2
,
a
n
d
a
Da
v
ie
s
-
Bo
u
ld
in
in
d
e
x
(DBI) o
f
0
.
4
1
,
o
u
tp
e
rf
o
rm
in
g
o
th
e
r
sta
ts
o
f
t
h
e
a
rts
m
e
th
o
d
s.
K
ey
w
o
r
d
s
:
B
ee
co
lo
n
y
alg
o
r
ith
m
B
r
ain
m
ag
n
etic
r
eso
n
a
n
ce
im
ag
in
g
s
eg
m
en
tatio
n
Fu
zz
y
clu
s
ter
in
g
Fu
zz
y
c
-
m
ea
n
s
Op
tim
izatio
n
m
eth
o
d
s
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
C
h
ak
ir
Mo
k
h
tar
i
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
S
cie
n
ce
,
Facu
lty
o
f
E
x
ac
t Scie
n
ce
s
,
Un
iv
er
s
ity
o
f
Ma
s
ca
r
a
B
P 3
0
5
,
R
o
u
te
d
e
Ma
m
o
u
n
ia,
Ma
s
ca
r
a,
2
9
0
0
0
,
Alg
er
ia
E
m
ail:
m
o
k
h
tar
i.c
@
u
n
iv
-
m
ascar
a.
d
z
AB
B
RE
V
I
A
T
I
O
NS
Ab
b
r
ev
.
Descr
ip
tio
n
Ab
b
r
ev
.
Descr
ip
tio
n
AB
C
Ar
tific
ial
b
ee
co
lo
n
y
GM
M
Gau
s
s
ian
m
ix
tu
r
e
m
o
d
els
AFC
M
Ad
ap
tiv
e
f
u
zz
y
c
-
m
ea
n
s
GW
A
Gr
ay
wo
lf
alg
o
r
ith
m
C
NN
C
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
s
L
DC
F
C
M
L
o
ca
l d
en
s
ity
clu
s
ter
in
g
f
u
zz
y
c
-
m
ea
n
s
C
SF
C
er
eb
r
o
s
p
in
al
f
lu
id
ML
Ma
ch
in
e
lear
n
in
g
DB
I
Dav
ies
-
B
o
u
ld
in
in
d
ex
MRI
Ma
g
n
etic
r
eso
n
an
ce
im
a
g
in
g
DL
Dee
p
lear
n
in
g
PC
I
Par
titi
o
n
co
ef
f
icien
t in
d
ex
DPSO
Dy
n
am
ic
p
ar
ticle
s
war
m
o
p
tim
izatio
n
PSO
Par
ticle
s
war
m
o
p
tim
izatio
n
FC
M
Fu
zz
y
c
-
m
ea
n
s
alg
o
r
ith
m
R
DO
R
ain
d
r
o
p
o
p
tim
izer
FP
SO
FC
M
F
u
z
z
y
PS
O
f
o
r
FC
M
SVM
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es
GA
Gen
etic
alg
o
r
ith
m
WM
W
h
ite
m
atter
GM
Gr
ay
m
atter
W
OA
W
h
ale
o
p
tim
izatio
n
alg
o
r
ith
m
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:
2088
-
8
7
0
8
F
u
z
z
y
clu
s
ter
in
g
o
p
timiz
a
tio
n
b
a
s
ed
a
r
tifi
cia
l b
ee
co
lo
n
y
a
l
g
o
r
ith
m
fo
r
…
(
C
h
a
kir Mo
kh
ta
r
i
)
4917
1.
I
NT
RO
D
UCT
I
O
N
B
r
ain
MRI
im
ag
e
s
eg
m
en
tati
o
n
is
a
cr
itical
task
in
m
e
d
ic
al
im
ag
in
g
,
e
n
ab
lin
g
th
e
d
elin
ea
tio
n
an
d
an
aly
s
is
o
f
an
ato
m
ical
s
tr
u
ctu
r
es,
p
ath
o
lo
g
ical
r
eg
i
o
n
s
,
an
d
f
u
n
ctio
n
al
ar
ea
s
with
in
th
e
b
r
ain
.
I
t
p
lay
s
a
p
iv
o
tal
r
o
le
in
d
ia
g
n
o
s
in
g
n
e
u
r
o
lo
g
ical
d
is
o
r
d
er
s
,
p
lan
n
in
g
tr
ea
tm
en
ts
,
an
d
m
o
n
ito
r
in
g
d
is
ea
s
e
p
r
o
g
r
ess
io
n
.
Ov
er
th
e
p
ast
f
ew
d
ec
a
d
es,
s
ig
n
if
ican
t
ad
v
an
ce
m
en
ts
h
a
v
e
b
ee
n
m
ad
e
i
n
s
eg
m
en
tatio
n
tech
n
iq
u
es,
d
r
iv
en
b
y
th
e
in
cr
ea
s
in
g
av
ailab
ilit
y
o
f
h
ig
h
-
r
eso
lu
tio
n
MRI
d
ata
an
d
th
e
d
ev
elo
p
m
en
t
o
f
s
o
p
h
is
ticated
co
m
p
u
tatio
n
al
m
eth
o
d
s
[
1
]
.
H
o
wev
er
,
d
esp
it
e
th
ese
ad
v
a
n
ce
m
en
ts
,
b
r
ain
MRI
s
eg
m
en
tatio
n
r
e
m
ain
s
a
ch
allen
g
in
g
p
r
o
b
lem
d
u
e
to
th
e
in
h
er
e
n
t
co
m
p
lex
it
y
o
f
b
r
ain
s
tr
u
ctu
r
es,
v
a
r
iab
ilit
y
ac
r
o
s
s
in
d
i
v
id
u
als,
a
n
d
lim
itatio
n
s
in
im
ag
in
g
tech
n
o
lo
g
y
[
2
]
,
[
3
]
.
T
h
e
cu
r
r
en
t
s
tate
o
f
b
r
ain
MRI
s
eg
m
en
tatio
n
is
ch
ar
ac
ter
ized
b
y
a
d
iv
er
s
e
ar
r
ay
o
f
m
eth
o
d
s
,
r
a
n
g
in
g
f
r
o
m
tr
ad
itio
n
al
ap
p
r
o
ac
h
es to
m
o
d
er
n
d
ee
p
lear
n
in
g
-
b
ased
tech
n
iq
u
es
[
4
]
.
T
r
ad
itio
n
al
m
eth
o
d
s
f
o
r
b
r
ain
MRI
s
eg
m
en
tatio
n
r
ely
o
n
i
n
ten
s
ity
v
alu
es,
s
p
atial
in
f
o
r
m
atio
n
,
an
d
an
ato
m
ical
p
r
io
r
s
to
d
if
f
e
r
e
n
tiate
s
tr
u
ctu
r
es.
T
ec
h
n
iq
u
e
s
in
clu
d
e
th
r
esh
o
ld
in
g
,
r
eg
i
o
n
-
b
ased
m
eth
o
d
s
(
e.
g
.
,
r
e
g
io
n
g
r
o
win
g
an
d
wate
r
s
h
ed
alg
o
r
ith
m
)
,
ed
g
e
d
ete
ctio
n
(
e.
g
.
,
So
b
el
a
n
d
C
an
n
y
)
,
ac
tiv
e
co
n
to
u
r
s
,
atlas
-
b
ased
m
eth
o
d
s
,
an
d
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
.
W
h
ile
th
ese
m
eth
o
d
s
h
av
e
co
n
tr
ib
u
t
ed
to
s
eg
m
en
tatio
n
,
th
ey
f
ac
e
ch
allen
g
es
s
u
ch
as
in
ten
s
ity
in
h
o
m
o
g
en
eities,
an
d
an
ato
m
ical
v
ar
ia
b
ilit
y
,
as
well
as
r
elian
ce
o
n
m
an
u
al
in
ter
v
en
tio
n
an
d
lo
ca
l in
f
o
r
m
atio
n
,
wh
ich
lim
its
th
eir
ac
cu
r
ac
y
an
d
g
e
n
er
aliza
b
ilit
y
.
T
h
ese
lim
itatio
n
s
h
av
e
s
p
u
r
r
ed
th
e
d
ev
elo
p
m
e
n
t
o
f
ad
v
a
n
ce
d
tech
n
iq
u
es
lik
e
m
ac
h
i
n
e
lear
n
in
g
an
d
d
e
ep
lear
n
in
g
,
w
h
ich
au
to
m
ate
co
m
p
le
x
p
atter
n
r
ec
o
g
n
itio
n
.
Ho
wev
er
,
tr
a
d
i
tio
n
al
m
eth
o
d
s
r
em
ain
r
elev
an
t
f
o
r
s
p
ec
if
ic
ap
p
licatio
n
s
an
d
as p
r
ep
r
o
ce
s
s
in
g
s
tep
s
in
m
o
d
e
r
n
s
eg
m
e
n
tatio
n
p
ip
elin
es
[
5
]
,
[
6
]
.
Ma
ch
in
e
lear
n
in
g
,
in
clu
d
in
g
s
u
p
er
v
is
ed
an
d
u
n
s
u
p
er
v
is
ed
tech
n
iq
u
es,
h
as
s
ig
n
if
ican
tly
ad
v
an
ce
d
b
r
ain
MRI
s
eg
m
e
n
tatio
n
b
y
o
f
f
er
in
g
m
o
r
e
r
o
b
u
s
t,
d
ata
-
d
r
iv
en
ap
p
r
o
ac
h
es
c
o
m
p
ar
e
d
to
tr
ad
itio
n
al
m
eth
o
d
s
.
Su
p
er
v
is
ed
m
eth
o
d
s
lik
e
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
es
an
d
r
an
d
o
m
f
o
r
ests
u
s
e
lab
eled
d
atasets
to
lear
n
co
m
p
lex
p
atter
n
s
,
im
p
r
o
v
in
g
s
eg
m
en
tat
io
n
ac
cu
r
ac
y
.
H
o
wev
er
,
th
eir
s
u
cc
ess
d
ep
en
d
s
o
n
h
ig
h
-
q
u
ality
an
n
o
tated
d
ata
,
wh
ich
is
co
s
tly
an
d
tim
e
-
c
o
n
s
u
m
in
g
t
o
p
r
o
d
u
ce
,
a
n
d
th
ey
o
f
te
n
s
tr
u
g
g
le
to
g
en
e
r
alize
ac
r
o
s
s
d
if
f
e
r
en
t
im
ag
in
g
p
r
o
to
co
ls
o
r
p
o
p
u
lat
io
n
s
.
Un
s
u
p
er
v
is
ed
m
eth
o
d
s
,
s
u
ch
as
k
-
m
ea
n
s
clu
s
ter
in
g
,
Gau
s
s
ian
m
ix
tu
r
e
m
o
d
els,
an
d
f
u
zz
y
c
-
m
ea
n
s
,
g
r
o
u
p
p
ix
els
b
ased
o
n
s
im
ilar
ity
with
o
u
t
lab
eled
d
ata,
m
ak
in
g
th
em
u
s
ef
u
l
f
o
r
ex
p
lo
r
ato
r
y
an
aly
s
is
.
Ho
wev
er
,
th
ey
lack
p
r
ec
is
io
n
f
o
r
cl
in
ical
ap
p
licatio
n
s
d
u
e
to
r
el
ian
ce
o
n
lo
w
-
lev
el
f
ea
tu
r
es a
n
d
s
en
s
itiv
ity
to
n
o
is
e
an
d
ar
tifa
cts.
B
o
th
ap
p
r
o
ac
h
es f
ac
e
ch
allen
g
es lik
e
in
ten
s
ity
in
h
o
m
o
g
en
eities,
n
o
is
e,
class
im
b
alan
ce
,
an
d
h
ig
h
co
m
p
u
tatio
n
al
c
o
s
ts
,
wh
ich
ca
n
d
e
g
r
ad
e
p
er
f
o
r
m
an
ce
a
n
d
lim
it
s
ca
lab
ilit
y
.
W
h
ile
m
ac
h
in
e
lear
n
in
g
r
e
m
ain
s
r
elev
an
t
in
s
p
ec
if
ic
ap
p
licatio
n
s
an
d
h
y
b
r
id
p
ip
el
in
es,
its
ch
allen
g
es
h
ig
h
lig
h
t th
e
n
ee
d
f
o
r
c
o
n
tin
u
ed
in
n
o
v
atio
n
in
b
r
ain
MRI
s
e
g
m
en
tatio
n
[
7
]
.
T
h
e
r
is
e
o
f
d
ee
p
lear
n
in
g
,
p
ar
ticu
lar
ly
co
n
v
o
lu
tio
n
al
n
e
u
r
al
n
etwo
r
k
s
,
r
ev
o
lu
tio
n
ized
b
r
ain
MRI
s
eg
m
en
tatio
n
b
y
en
ab
lin
g
au
to
m
atic
lear
n
in
g
o
f
h
ier
a
r
ch
i
ca
l
f
ea
tu
r
es
f
r
o
m
r
aw
d
ata.
Ar
ch
itectu
r
es
lik
e
U
-
Net,
with
its
co
n
tr
ac
tin
g
an
d
ex
p
a
n
s
iv
e
p
ath
s
co
n
n
ec
te
d
b
y
s
k
ip
co
n
n
ec
tio
n
s
,
ex
ce
ll
ed
in
ca
p
tu
r
in
g
f
in
e
d
etails
an
d
ac
h
iev
in
g
s
tate
-
of
-
th
e
-
ar
t
r
esu
lts
.
Fu
lly
co
n
v
o
lu
ti
o
n
al
n
etwo
r
k
s
(
FC
Ns)
f
u
r
th
er
ad
v
an
ce
d
th
e
f
ield
b
y
en
ab
lin
g
en
d
-
to
-
e
n
d
,
p
i
x
el
-
wis
e
s
eg
m
en
tatio
n
with
o
u
t
h
an
d
cr
af
te
d
f
ea
tu
r
es.
Ho
w
ev
er
,
d
ee
p
lear
n
in
g
m
eth
o
d
s
f
a
ce
ch
allen
g
es,
in
cl
u
d
in
g
th
e
n
ee
d
f
o
r
la
r
g
e,
h
ig
h
-
q
u
ality
a
n
n
o
tated
d
atasets
,
w
h
ich
ar
e
co
s
tly
an
d
tim
e
-
co
n
s
u
m
in
g
t
o
p
r
o
d
u
ce
.
L
im
ited
d
ataset
d
iv
er
s
ity
ca
n
h
i
n
d
er
m
o
d
el
p
er
f
o
r
m
a
n
ce
an
d
g
en
er
aliza
tio
n
,
ev
e
n
with
d
ata
au
g
m
en
tatio
n
.
Ad
d
i
tio
n
ally
,
th
e
h
ig
h
co
m
p
u
tatio
n
al
co
s
t
o
f
tr
ain
in
g
,
esp
ec
ially
f
o
r
3
D
v
o
lu
m
et
r
ic
d
ata,
p
o
s
es
s
ca
lab
ilit
y
an
d
ac
ce
s
s
ib
ilit
y
is
s
u
es,
p
ar
ticu
lar
ly
in
r
eso
u
r
ce
-
c
o
n
s
tr
ain
ed
s
ettin
g
s
.
Desp
ite
th
ese
lim
itatio
n
s
,
d
ee
p
lear
n
in
g
r
em
ain
s
a
tr
an
s
f
o
r
m
ativ
e
a
p
p
r
o
ac
h
in
b
r
ai
n
MRI
s
eg
m
en
tatio
n
[
2
]
,
[
8
]
,
[
9
]
.
I
n
th
is
a
r
ticle,
we
ad
v
o
ca
te
f
o
r
th
e
h
y
b
r
id
izatio
n
o
f
th
e
f
u
zz
y
c
-
m
ea
n
s
m
eth
o
d
[
1
0
]
ap
p
lied
to
b
r
ain
MRI
im
ag
e
s
eg
m
e
n
tatio
n
,
p
o
s
itio
n
in
g
it
as
a
co
m
p
ellin
g
a
lter
n
ativ
e
to
p
u
r
ely
m
ac
h
in
e
l
ea
r
n
in
g
a
n
d
d
ee
p
lear
n
in
g
ap
p
r
o
ac
h
es.
W
h
ile
M
L
an
d
DL
m
eth
o
d
s
h
av
e
r
ev
o
l
u
tio
n
ized
m
ed
ical
im
ag
e
s
eg
m
en
tatio
n
with
th
eir
ab
ilit
y
to
lear
n
co
m
p
lex
p
atter
n
s
an
d
ac
h
ie
v
e
s
tate
-
of
-
th
e
-
a
r
t
r
esu
lts
,
th
ey
co
m
e
with
s
ig
n
if
ican
t
ch
allen
g
es,
in
clu
d
in
g
th
e
n
ee
d
f
o
r
lar
g
e
an
n
o
tated
d
atasets
,
h
ig
h
co
m
p
u
tatio
n
al
co
s
ts
,
an
d
lim
ited
in
ter
p
r
etab
ilit
y
.
I
n
co
n
tr
ast,
f
u
zz
y
c
-
m
ea
n
s
,
a
well
-
estab
lis
h
ed
u
n
s
u
p
er
v
is
ed
clu
s
ter
in
g
tech
n
iq
u
e,
o
f
f
er
s
s
ev
er
al
u
n
iq
u
e
ad
v
an
tag
es
th
at
ca
n
b
e
e
n
h
an
ce
d
th
r
o
u
g
h
h
y
b
r
id
izatio
n
,
m
ak
in
g
it
a
v
iab
le
an
d
ef
f
icien
t
s
o
lu
tio
n
f
o
r
b
r
ai
n
MRI
s
eg
m
en
tatio
n
.
FC
M
is
p
ar
ticu
lar
ly
well
-
s
u
ited
f
o
r
m
e
d
ical
im
ag
in
g
d
u
e
to
its
ab
ilit
y
to
h
an
d
le
th
e
in
h
er
e
n
t
am
b
ig
u
ity
an
d
u
n
ce
r
tain
ty
in
tis
s
u
e
b
o
u
n
d
ar
ies.
Un
lik
e
tr
ad
itio
n
al
h
ar
d
clu
s
ter
in
g
m
et
h
o
d
s
,
FC
M
allo
w
s
p
ix
els
to
b
elo
n
g
to
m
u
ltip
le
cl
u
s
ter
s
with
v
ar
y
in
g
d
eg
r
ee
s
o
f
m
em
b
er
s
h
ip
,
r
ef
lectin
g
th
e
p
ar
tial
v
o
lu
m
e
e
f
f
ec
t
o
f
ten
o
b
s
er
v
e
d
in
MRI
d
ata.
T
h
is
f
lex
ib
ilit
y
m
ak
es F
C
M
h
ig
h
ly
ef
f
ec
tiv
e
f
o
r
s
eg
m
en
tin
g
b
r
ain
tis
s
u
es
s
u
ch
as
GM
,
W
M,
an
d
C
SF
,
wh
er
e
in
ten
s
ity
d
is
tr
ib
u
tio
n
s
o
f
ten
o
v
e
r
lap
.
Ho
wev
er
,
tr
ad
itio
n
al
FC
M
p
r
esen
ts
s
er
io
u
s
lim
itatio
n
s
wh
ich
ca
n
d
eg
r
ad
e
its
p
er
f
o
r
m
an
ce
in
co
m
p
lex
M
R
I
d
atasets
.
−
Firstl
y
,
it n
ee
d
s
th
e
r
ig
h
t n
u
m
b
er
o
f
clu
s
ter
s
wh
ich
is
n
o
t a
v
ailab
le
in
all
ca
s
es.
−
Seco
n
d
ly
,
it
is
v
er
y
s
en
s
itiv
e
to
in
itializatio
n
,
d
ef
er
en
t
clu
s
ter
ce
n
ter
s
in
itializatio
n
ca
n
lead
to
d
ef
er
e
n
t
clu
s
ter
in
g
r
esu
lts
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
5
,
Octo
b
e
r
20
25
:
4
9
1
6
-
4
9
3
2
4918
−
T
h
ir
d
ly
,
d
u
e
to
th
e
p
r
i
n
cip
le
o
f
th
e
iter
ativ
e
o
p
tim
izatio
n
o
f
a
co
s
t
f
u
n
ctio
n
,
it
is
s
tr
o
n
g
ly
s
en
s
itiv
e
to
th
e
p
r
o
b
lem
s
o
f
lo
ca
l
m
in
im
a.
T
h
ese
ch
allen
g
es
ca
n
lead
to
s
u
b
o
p
tim
al
s
eg
m
en
tatio
n
r
esu
lts
,
p
ar
ticu
lar
ly
in
co
m
p
lex
MRI
d
atasets
with
in
ten
s
ity
in
h
o
m
o
g
en
eities o
r
o
v
er
lap
p
in
g
tis
s
u
e
d
is
tr
ib
u
tio
n
s
.
T
o
ad
d
r
ess
th
ese
lim
itatio
n
s
,
we
p
r
o
p
o
s
e
a
h
y
b
r
id
ap
p
r
o
ac
h
th
at
in
teg
r
ates
FC
M
with
AB
C
o
p
tim
izatio
n
[
1
1
]
.
T
h
is
h
y
b
r
id
ap
p
r
o
ac
h
,
r
e
f
er
r
ed
to
as
FC
M
-
AB
C
o
p
tim
izer
,
lev
er
a
g
es
th
e
s
tr
en
g
th
s
o
f
b
o
th
m
eth
o
d
s
to
ad
d
r
ess
th
e
lim
i
tatio
n
s
o
f
tr
ad
itio
n
al
FC
M
wh
ile
en
h
an
cin
g
its
ac
cu
r
ac
y
,
r
o
b
u
s
tn
ess
,
an
d
ef
f
icien
cy
.
T
h
e
in
te
g
r
atio
n
o
f
AB
C
with
FC
M
is
p
ar
ticu
lar
ly
ju
s
tifie
d
in
th
e
c
o
n
t
ex
t
o
f
b
r
ain
MRI
s
eg
m
en
tatio
n
d
u
e
t
o
th
e
u
n
iq
u
e
ch
allen
g
es
p
o
s
ed
b
y
m
e
d
ic
al
im
ag
in
g
d
ata.
B
r
ain
MRI
i
m
ag
es
o
f
ten
ex
h
ib
it
h
ig
h
v
a
r
iab
ilit
y
in
in
ten
s
ity
,
a
n
d
an
ato
m
ical
s
tr
u
ctu
r
es,
m
ak
in
g
it
d
if
f
icu
lt
f
o
r
tr
a
d
itio
n
al
m
eth
o
d
s
to
ac
h
ie
v
e
co
n
s
is
ten
t
an
d
ac
cu
r
ate
r
esu
lts
.
AB
C
-
F
C
M
o
p
tim
izer
ad
d
r
es
s
es
th
ese
ch
allen
g
es
b
y
co
m
b
in
in
g
th
e
f
lex
ib
ilit
y
o
f
FC
M
in
h
an
d
lin
g
u
n
ce
r
tain
t
y
with
th
e
g
lo
b
al
o
p
tim
izatio
n
ca
p
ab
ilit
ies o
f
AB
C
.
Mo
r
eo
v
er
,
th
e
h
y
b
r
id
AB
C
-
FC
M
ap
p
r
o
ac
h
ali
g
n
s
with
t
h
e
n
ee
d
f
o
r
in
te
r
p
r
etab
le
an
d
clin
ically
r
elev
an
t
s
eg
m
e
n
tatio
n
m
eth
o
d
s
.
Un
lik
e
d
ee
p
lea
r
n
in
g
m
o
d
e
ls
,
wh
ich
o
f
ten
o
p
er
ate
as
“
b
lack
b
o
x
es,
”
AB
C
-
FC
M
p
r
o
v
id
es
tr
an
s
p
ar
en
t
an
d
in
tu
itiv
e
r
esu
lts
,
m
ak
in
g
it
ea
s
ier
f
o
r
clin
ician
s
to
u
n
d
er
s
tan
d
an
d
tr
u
s
t
th
e
s
eg
m
en
tatio
n
o
u
tc
o
m
es.
T
h
is
is
p
ar
ticu
lar
ly
im
p
o
r
tan
t
in
m
ed
ical
ap
p
licatio
n
s
,
wh
er
e
in
ter
p
r
etab
ilit
y
an
d
ex
p
lain
ab
ilit
y
ar
e
c
r
itical
f
o
r
c
lin
ical
d
ec
is
io
n
-
m
ak
in
g
.
T
h
e
in
teg
r
atio
n
o
f
AB
C
with
FC
M
ad
d
r
ess
es
s
ev
er
al
k
ey
ch
allen
g
es in
b
r
ain
MRI
s
eg
m
e
n
tatio
n
:
−
I
m
p
r
o
v
ed
in
itializatio
n
:
AB
C
's
g
lo
b
al
s
ea
r
ch
ca
p
ab
ilit
ies
o
p
tim
ize
in
itial
clu
s
ter
ce
n
ter
s
,
r
ed
u
cin
g
t
h
e
r
is
k
o
f
p
o
o
r
in
itializatio
n
an
d
en
h
a
n
cin
g
s
eg
m
en
tatio
n
ac
cu
r
ac
y
.
−
E
s
ca
p
e
f
r
o
m
lo
ca
l
o
p
tim
a:
AB
C
h
elp
s
F
C
M
av
o
id
lo
ca
l
o
p
tim
a
b
y
ex
p
lo
r
in
g
n
ew
r
e
g
io
n
s
o
f
th
e
s
o
lu
tio
n
s
p
ac
e,
en
s
u
r
in
g
t
h
at
clu
s
ter
ce
n
ter
s
co
n
v
er
g
e
clo
s
er
to
t
h
e
g
l
o
b
al
o
p
tim
u
m
.
−
C
o
m
p
u
tatio
n
al
ef
f
icien
cy
:
Alt
h
o
u
g
h
AB
C
ad
d
s
co
m
p
lex
ity
,
its
ef
f
icien
t
s
ea
r
ch
m
ec
h
an
is
m
o
f
ten
lead
s
to
f
aster
co
n
v
er
g
en
ce
,
b
alan
cin
g
ac
cu
r
ac
y
an
d
co
m
p
u
tatio
n
al
c
o
s
t.
−
Ad
ap
tab
ilit
y
to
co
m
p
lex
b
r
ain
s
tr
u
ctu
r
es:
AB
C
's
ad
ap
tiv
e
r
e
f
in
em
en
t
o
f
clu
s
ter
ce
n
ter
s
m
a
k
es
it
ef
f
ec
tiv
e
f
o
r
s
eg
m
en
tin
g
co
m
p
lex
b
r
a
in
s
tr
u
ctu
r
es
(
e.
g
.
,
g
r
ay
m
at
ter
,
wh
ite
m
atter
,
ce
r
eb
r
o
s
p
in
al
f
lu
id
)
a
n
d
p
ath
o
lo
g
ical
r
e
g
io
n
s
(
e.
g
.
,
tu
m
o
r
s
)
,
h
an
d
lin
g
th
e
v
ar
iab
ilit
y
a
n
d
in
tr
icac
y
o
f
b
r
ain
MRI
d
ata
.
Ou
r
g
o
al
is
to
e
n
h
an
ce
s
eg
m
en
tatio
n
ac
c
u
r
ac
y
b
y
o
p
ti
m
izin
g
th
e
FC
M
alg
o
r
ith
m
th
r
o
u
g
h
th
e
s
im
u
ltan
eo
u
s
o
p
tim
izatio
n
o
f
i
ts
k
ey
p
ar
am
ete
r
s
,
in
clu
d
in
g
t
h
e
o
b
jectiv
e
f
u
n
ctio
n
,
th
e
n
u
m
b
er
o
f
clu
s
ter
s
,
an
d
th
e
clu
s
ter
ce
n
ter
v
al
u
es,
u
s
in
g
th
e
AB
C
alg
o
r
ith
m
.
On
ce
t
h
e
o
p
tim
al
n
u
m
b
er
o
f
cl
u
s
ter
s
an
d
clu
s
ter
ce
n
te
r
v
alu
es
ar
e
d
eter
m
i
n
ed
,
t
h
e
cla
s
s
if
icatio
n
o
f
all
p
ix
els
is
p
er
f
o
r
m
ed
u
s
in
g
th
e
m
em
b
er
s
h
ip
d
eg
r
ee
m
atr
i
x
.
Ou
r
ap
p
r
o
ac
h
lev
er
a
g
es
th
e
AB
C
alg
o
r
ith
m
'
s
s
tr
o
n
g
o
p
tim
izatio
n
ca
p
ab
ilit
ies,
wh
ich
en
s
u
r
e
t
h
e
d
is
co
v
er
y
o
f
t
h
e
g
lo
b
al
o
p
tim
u
m
an
d
allo
w
f
o
r
in
d
iv
id
u
als
o
f
v
ar
y
in
g
s
i
ze
s
in
th
e
i
n
itial
p
o
p
u
latio
n
.
T
h
ese
p
r
o
p
e
r
ties
s
ig
n
if
ican
tly
im
p
r
o
v
e
th
e
FC
M
alg
o
r
ith
m
,
lead
in
g
to
m
o
r
e
ef
f
ec
tiv
e
clu
s
ter
in
g
.
B
y
in
teg
r
atin
g
AB
C
with
FC
M,
o
u
r
p
r
o
p
o
s
ed
m
eth
o
d
ad
d
r
ess
es
cr
itical
ch
allen
g
es
in
f
u
zz
y
clu
s
ter
in
g
,
s
u
ch
as
d
eter
m
in
in
g
th
e
ap
p
r
o
p
r
iate
n
u
m
b
er
o
f
clu
s
ter
s
,
id
en
tify
in
g
o
p
tim
al
clu
s
ter
ce
n
ter
s
,
an
d
ac
h
iev
in
g
t
h
e
o
p
tim
al
v
alu
e
o
f
t
h
e
o
b
jectiv
e
f
u
n
ctio
n
,
all
in
a
u
n
i
f
ied
an
d
s
im
u
ltan
e
o
u
s
m
an
n
er
.
T
h
e
r
em
ain
d
er
o
f
th
e
p
ap
e
r
is
o
r
g
an
ized
as
f
o
llo
ws:
s
ec
t
io
n
1
in
tr
o
d
u
ce
s
th
e
p
ap
er
.
Sectio
n
2
r
ev
iews
r
elev
an
t
s
tu
d
ies
o
n
o
p
tim
izin
g
b
r
ain
MRI
im
ag
e
s
eg
m
en
tatio
n
u
s
in
g
f
u
zz
y
tech
n
iq
u
es.
Sectio
n
3
p
r
esen
ts
th
e
p
r
o
p
o
s
ed
clu
s
te
r
in
g
m
eth
o
d
b
ased
o
n
th
e
FC
M
-
A
B
C
o
p
tim
izer
.
Sectio
n
4
d
is
cu
s
s
es
th
e
ex
p
er
im
en
tatio
n
an
d
r
esu
lts
,
f
o
llo
wed
b
y
t
h
e
co
n
clu
s
io
n
in
s
ec
tio
n
5
.
2.
RE
L
AT
E
D
WO
RK
FC
M
m
eth
o
d
as
u
n
s
u
p
er
v
is
ed
ap
p
r
o
ac
h
is
wid
ely
s
tu
d
ied
an
d
u
s
ed
as
a
p
o
wer
f
u
l
to
o
l
in
a
wid
e
r
an
g
e
o
f
a
p
p
licatio
n
s
an
d
s
u
cc
ess
f
u
lly
ap
p
lied
in
m
ed
ical
im
ag
e
s
eg
m
en
tatio
n
.
I
n
th
e
f
ield
o
f
b
r
ain
MRI
im
ag
e
s
eg
m
en
tatio
n
,
FC
M
alg
o
r
ith
m
is
ex
ten
s
iv
ely
u
tili
ze
d
d
u
e
t
o
its
ab
ilit
y
to
h
an
d
le
th
e
u
n
ce
r
tain
t
y
an
d
co
m
p
lex
ity
o
f
m
e
d
ical
im
ag
e
s
[
1
2
]
–
[
1
5
]
.
On
e
o
f
its
m
ain
ad
v
an
tag
es
is
its
ab
ilit
y
to
p
r
o
d
u
ce
s
m
o
o
th
a
n
d
ac
cu
r
ate
s
eg
m
en
tatio
n
s
,
m
ak
i
n
g
it
a
v
alu
ab
le
to
o
l
f
o
r
m
ed
i
ca
l
d
iag
n
o
s
is
an
d
tr
ea
tm
en
t
p
lan
n
in
g
.
Ho
wev
e
r
,
th
e
alg
o
r
ith
m
h
as
s
o
m
e
lim
it
atio
n
s
,
s
u
ch
as
it
s
s
en
s
itiv
ity
to
n
o
is
e
an
d
in
ten
s
ity
in
h
o
m
o
g
en
eity
,
wh
ich
ca
n
lead
to
m
is
class
if
icatio
n
.
Mo
r
eo
v
e
r
,
FC
M
r
eq
u
ir
es
p
r
io
r
k
n
o
wled
g
e
o
f
th
e
n
u
m
b
er
o
f
clu
s
ter
s
,
an
d
its
co
m
p
u
tatio
n
al
c
o
s
t c
an
b
e
h
ig
h
,
esp
ec
ially
f
o
r
lar
g
e
m
e
d
ical
d
atasets
.
I
n
r
e
c
e
n
t
y
e
a
r
s
,
r
e
s
e
a
r
c
h
e
r
s
o
f
t
e
n
i
n
t
e
g
r
a
te
FC
M
w
it
h
p
r
e
p
r
o
c
e
s
s
i
n
g
t
e
c
h
n
i
q
u
es
,
h
y
b
r
i
d
m
o
d
e
l
s
a
n
d
o
p
t
i
m
i
z
a
ti
o
n
m
e
t
h
o
d
s
,
s
u
c
h
as
p
a
r
t
i
c
l
e
s
w
a
r
m
o
p
t
i
m
i
z
a
ti
o
n
,
g
e
n
e
t
i
c
a
l
g
o
r
i
t
h
m
s
,
a
r
t
i
f
i
c
i
al
b
e
e
c
o
l
o
n
y
,
a
n
d
g
r
a
y
w
o
l
f
a
l
g
o
r
i
t
h
m
,
t
o
i
m
p
r
o
v
e
c
lu
s
t
e
r
i
n
g
a
c
c
u
r
ac
y
,
e
n
h
a
n
c
e
r
o
b
u
s
t
n
e
s
s
,
a
n
d
r
e
d
u
c
e
c
o
m
p
u
t
ati
o
n
a
l
c
o
m
p
l
e
x
i
t
y
i
n
b
r
a
i
n
M
R
I
s
e
g
m
e
n
ta
t
i
o
n
.
S
o
n
g
e
t
a
l
.
[
1
6
]
p
r
o
p
o
s
e
d
a
f
u
z
z
y
c
-
m
e
a
n
s
cl
u
s
t
e
r
i
n
g
m
o
d
e
l
wi
t
h
s
p
a
t
i
a
l
c
o
n
s
t
r
ai
n
t
f
o
r
u
n
s
u
p
e
r
v
i
s
e
d
s
e
g
m
e
n
t
a
ti
o
n
o
f
b
r
a
i
n
m
a
g
n
e
t
i
c
r
es
o
n
a
n
c
e
i
m
ag
e
s
.
T
h
e
y
i
n
c
o
r
p
o
r
a
t
e
t
h
e
s
p
a
ti
a
l
d
is
t
a
n
c
e
a
n
d
t
h
e
g
r
a
y
l
e
v
e
l
i
n
f
o
r
m
a
t
i
o
n
b
e
t
w
e
en
t
h
e
l
o
c
a
l
n
e
i
g
h
b
o
r
h
o
o
d
p
i
x
e
l
s
,
c
o
m
b
i
n
e
d
w
i
t
h
t
h
e
n
o
n
-
l
i
n
e
ar
w
e
i
g
h
t
i
n
g
f
o
r
m
i
n
t
h
e
s
i
m
il
a
r
i
t
y
m
e
as
u
r
e
o
f
t
h
e
f
u
z
z
y
c
l
u
s
t
e
r
i
n
g
.
I
n
t
h
e
p
r
o
p
o
s
e
d
m
e
t
h
o
d
[
1
7
]
f
u
z
z
y
k
e
r
n
e
l
s
e
e
d
s
el
e
c
ti
o
n
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:
2088
-
8
7
0
8
F
u
z
z
y
clu
s
ter
in
g
o
p
timiz
a
tio
n
b
a
s
ed
a
r
tifi
cia
l b
ee
co
lo
n
y
a
l
g
o
r
ith
m
fo
r
…
(
C
h
a
kir Mo
kh
ta
r
i
)
4919
t
e
c
h
n
i
q
u
e
i
s
u
s
e
d
t
o
d
e
f
i
n
e
t
h
e
c
o
m
p
l
e
t
e
b
r
a
i
n
M
R
I
i
m
a
g
e
i
n
t
o
d
i
f
f
e
r
e
n
t
g
r
o
u
p
s
o
f
s
i
m
i
l
a
r
i
n
t
e
n
s
i
t
y
.
A
m
o
n
g
t
h
e
s
e
g
r
o
u
p
s
t
h
e
m
o
s
t
a
c
c
u
r
a
t
e
k
e
r
n
e
l
s
a
r
e
s
e
l
e
c
t
e
d
e
m
p
i
r
i
ca
ll
y
t
h
a
t
s
h
o
w
h
i
g
h
e
s
t
r
e
s
e
m
b
l
an
c
e
w
i
t
h
t
h
e
t
u
m
o
r
.
T
h
e
c
o
n
c
e
p
t
o
f
f
u
z
z
i
n
e
s
s
h
e
l
p
s
m
a
k
i
n
g
t
h
e
s
e
le
c
t
i
o
n
e
v
e
n
a
t
t
h
e
b
o
u
n
d
a
r
y
r
e
g
i
o
n
s
.
Gen
etic
alg
o
r
ith
m
s
h
av
e
b
ee
n
ap
p
lied
to
f
u
zz
y
cl
u
s
ter
in
g
f
o
r
MRI
s
eg
m
en
tatio
n
.
I
n
s
tu
d
y
[
1
8
]
an
in
n
o
v
ativ
e
a
p
p
r
o
ac
h
t
o
b
r
ai
n
MRI
im
ag
e
s
eg
m
e
n
tatio
n
was
p
r
esen
ted
.
T
h
e
r
esear
ch
er
s
en
h
an
ce
d
th
e
tr
ad
itio
n
al
FC
M
alg
o
r
ith
m
b
y
u
s
in
g
GA
f
o
r
p
a
r
am
ete
r
o
p
tim
izatio
n
,
wh
ich
s
ig
n
i
f
ican
tly
im
p
r
o
v
ed
s
eg
m
en
tatio
n
ac
cu
r
ac
y
.
T
h
ey
in
teg
r
ate
f
u
zz
y
s
et
th
eo
r
y
,
f
u
zz
y
m
etr
ics,
an
d
Su
g
en
o
n
eg
atio
n
p
r
in
ci
p
les.
W
h
en
test
ed
o
n
th
e
B
r
aT
S2
0
1
8
d
ataset,
th
eir
m
o
d
if
ie
d
ap
p
r
o
ac
h
o
u
tp
er
f
o
r
m
in
g
th
e
co
n
v
en
tio
n
al
FC
M
m
eth
o
d
.
T
h
is
ad
v
an
ce
m
e
n
t
is
p
ar
ticu
lar
ly
s
ig
n
if
ican
t
f
o
r
m
ed
ical
im
ag
in
g
an
aly
s
is
,
as
it
b
etter
h
an
d
les
th
e
ch
allen
g
es o
f
u
n
ce
r
tain
ty
,
n
o
is
e,
an
d
am
b
ig
u
ity
i
n
MRI
im
ag
es.
Par
ticle
s
war
m
o
p
tim
izatio
n
h
as
b
ee
n
ex
ten
s
iv
ely
u
tili
ze
d
to
en
h
an
ce
FC
M
b
y
o
p
tim
izin
g
its
clu
s
ter
ce
n
ter
s
.
Fo
r
in
s
tan
ce
,
PS
O
-
FC
M
alg
o
r
ith
m
s
aim
to
r
ed
u
ce
th
e
im
p
ac
t
o
f
l
o
ca
l
m
in
im
a
an
d
im
p
r
o
v
e
s
eg
m
en
tatio
n
r
o
b
u
s
tn
ess
b
y
g
lo
b
ally
s
ea
r
ch
in
g
f
o
r
b
etter
cl
u
s
ter
co
n
f
ig
u
r
atio
n
s
.
T
h
ese
m
eth
o
d
s
h
av
e
s
h
o
wn
p
r
o
m
is
e
in
im
p
r
o
v
in
g
s
eg
m
e
n
tatio
n
ac
cu
r
ac
y
an
d
co
m
p
u
t
atio
n
al
ef
f
icien
cy
.
Dh
an
ac
h
an
d
r
a
an
d
C
h
an
u
[
1
9
]
co
m
b
in
e
d
y
n
am
ic
p
ar
ticle
s
war
m
o
p
tim
izatio
n
(
DPSO)
with
th
e
F
C
M
alg
o
r
ith
m
,
ad
d
r
ess
in
g
FC
M
's
lim
itatio
n
s
s
u
ch
as
s
en
s
it
iv
ity
to
in
itial
v
alu
es
an
d
n
o
is
e.
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
d
y
n
am
ic
ally
ad
ju
s
ts
in
er
tia
weig
h
t
an
d
lear
n
in
g
p
ar
am
ete
r
s
,
en
h
an
cin
g
g
l
o
b
al
s
ea
r
ch
c
ap
ab
ilit
ies
wh
ile
in
co
r
p
o
r
atin
g
a
n
o
is
e
r
ed
u
ctio
n
m
ec
h
an
is
m
b
ased
o
n
s
u
r
r
o
u
n
d
in
g
p
ix
el
attr
ib
u
tes.
T
h
e
m
et
h
o
d
s
h
o
ws
im
p
r
o
v
e
d
r
o
b
u
s
tn
ess
an
d
ac
cu
r
ac
y
in
s
eg
m
en
tatio
n
,
m
a
k
in
g
it
a
s
ig
n
if
ican
t
ad
v
an
ce
m
en
t
in
im
a
g
e
p
r
o
ce
s
s
in
g
.
Ma
h
esa
an
d
W
ib
o
wo
[
2
0
]
p
r
esen
t
a
n
o
p
tim
izatio
n
m
eth
o
d
f
o
r
b
r
ai
n
tu
m
o
r
im
ag
e
s
eg
m
e
n
tatio
n
u
s
in
g
f
u
zz
y
c
-
m
ea
n
s
en
h
an
ce
d
b
y
PSO
.
T
h
e
r
esear
ch
ad
d
r
ess
es
th
e
in
ef
f
icien
cies
o
f
m
an
u
al
tu
m
o
r
s
eg
m
en
tatio
n
,
wh
ich
d
ela
y
s
p
atien
t
tr
ea
tm
en
t.
B
y
o
p
tim
izin
g
th
e
o
b
jectiv
e
f
u
n
ctio
n
o
f
FC
M,
th
e
p
r
o
p
o
s
ed
FC
M
-
PS
O
m
eth
o
d
ac
h
iev
e
d
lo
wer
o
b
jectiv
e
f
u
n
ctio
n
v
al
u
es
ac
r
o
s
s
s
ix
M
R
I
T
2
im
ag
es,
d
em
o
n
s
tr
atin
g
im
p
r
o
v
e
d
s
eg
m
en
tatio
n
ac
cu
r
ac
y
co
m
p
ar
e
d
to
th
e
o
r
ig
in
al
FC
M.
T
h
e
f
in
d
in
g
s
s
u
g
g
est
th
at
in
te
g
r
atin
g
PS
O
w
ith
FC
M
ca
n
s
ig
n
if
ican
tly
en
h
an
ce
th
e
r
eliab
ilit
y
o
f
au
to
m
ated
b
r
ain
t
u
m
o
r
s
eg
m
en
tatio
n
,
f
ac
ilit
atin
g
tim
ely
m
ed
ical
d
ec
is
io
n
s
.
Kav
ith
a
an
d
Pra
b
ak
a
r
an
[
2
1
]
p
r
esen
t
an
ap
p
r
o
ac
h
f
o
r
b
r
ai
n
tu
m
o
r
d
etec
tio
n
u
s
in
g
a
h
y
b
r
i
d
m
eth
o
d
c
o
m
b
in
in
g
ass
u
r
ed
c
o
n
v
er
g
en
ce
p
ar
ticle
s
war
m
o
p
tim
izatio
n
(
AC
PS
O)
an
d
FC
M
clu
s
ter
in
g
.
I
t
em
p
h
asizes
th
e
im
p
o
r
tan
ce
o
f
p
r
e
-
p
r
o
ce
s
s
in
g
tech
n
iq
u
es,
p
ar
ticu
lar
l
y
th
e
a
d
ap
tiv
e
b
ilater
al
f
ilter
.
T
h
e
s
tu
d
y
co
m
p
ar
es
v
ar
io
u
s
s
eg
m
en
tatio
n
tech
n
iq
u
es,
co
n
clu
d
in
g
th
at
th
e
p
r
o
p
o
s
ed
m
eth
o
d
s
ig
n
if
ica
n
tly
en
h
an
c
es
tu
m
o
r
d
etec
tio
n
ac
cu
r
ac
y
.
Sem
ch
ed
in
e
an
d
Mo
u
s
s
ao
u
i
[
2
2
]
p
r
o
p
o
s
ed
a
n
o
v
el
in
itializatio
n
ap
p
r
o
ac
h
f
o
r
th
e
f
u
zz
y
c
-
m
ea
n
s
alg
o
r
ith
m
b
ased
o
n
f
u
zz
y
p
ar
ticle
s
war
m
o
p
tim
izatio
n
(
FP
SO)
ap
p
lied
to
b
r
ai
n
MR
im
ag
e
s
eg
m
en
tatio
n
.
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
u
s
es
th
e
FP
SO
alg
o
r
ith
m
to
g
et
th
e
in
it
ial
clu
s
ter
ce
n
ter
s
o
f
FC
M
ac
co
r
d
in
g
to
a
n
ew
f
itn
ess
f
u
n
cti
o
n
wh
ich
co
m
b
in
es
f
u
zz
y
clu
s
ter
v
alid
ity
in
d
ices.
Gr
ay
wo
lf
o
p
tim
izatio
n
(
GW
O)
h
as
b
ee
n
ef
f
ec
tiv
ely
u
s
ed
t
o
o
p
tim
ize
FC
M
b
y
s
ea
r
ch
in
g
f
o
r
th
e
b
est
clu
s
ter
ce
n
tr
o
id
s
,
lead
in
g
to
i
m
p
r
o
v
e
d
clu
s
ter
in
g
ac
cu
r
ac
y
an
d
r
o
b
u
s
tn
ess
.
B
y
in
teg
r
atin
g
GW
O
with
F
C
M,
th
e
o
p
tim
izatio
n
p
r
o
ce
s
s
av
o
i
d
s
lo
ca
l
m
in
im
a
an
d
en
h
an
ce
s
clu
s
ter
in
g
p
er
f
o
r
m
an
ce
,
m
a
k
in
g
it
p
ar
ticu
lar
ly
u
s
ef
u
l
in
co
m
p
lex
im
ag
e
s
eg
m
en
tatio
n
an
d
d
ata
clu
s
ter
in
g
task
s
[
2
3
]
.
Sin
g
h
et
a
l.
[
2
4
]
in
tr
o
d
u
ce
a
n
o
v
el
im
ag
e
s
eg
m
en
tatio
n
m
eth
o
d
co
m
b
in
in
g
s
p
atial
f
u
zz
y
c
-
m
ea
n
s
(
SF
C
M)
clu
s
ter
in
g
with
th
e
GW
O,
ter
m
ed
SF
C
M
GW
O,
to
en
h
an
ce
th
e
a
cc
u
r
ac
y
o
f
MRI
im
ag
e
s
eg
m
e
n
tatio
n
.
T
h
e
s
tu
d
y
d
e
m
o
n
s
tr
at
es
th
at
SF
C
MG
W
O
o
u
tp
er
f
o
r
m
s
tr
ad
itio
n
al
SF
C
M
an
d
GA
-
b
ased
SF
C
M
(
G
ASFC
M)
in
s
eg
m
en
tatio
n
tas
k
s
,
as
ev
id
en
ce
d
b
y
im
p
r
o
v
e
d
clu
s
ter
in
g
v
alid
ity
f
u
n
ctio
n
s
.
T
h
e
e
f
f
ec
tiv
en
es
s
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
is
v
alid
ated
th
r
o
u
g
h
co
m
p
ar
ativ
e
an
al
y
s
is
o
n
two
b
r
ain
MRI
im
ag
es,
wh
er
e
it
ac
h
iev
es
s
u
p
er
io
r
p
er
f
o
r
m
a
n
ce
.
Nay
ak
et
a
l.
[
2
5
]
a
n
o
v
el
o
b
jectiv
e
f
u
n
ctio
n
ca
lle
d
f
u
zz
y
en
tr
o
p
y
cl
u
s
ter
in
g
wi
th
lo
ca
l
s
p
atial
in
f
o
r
m
atio
n
a
n
d
b
ias
co
r
r
ec
tio
n
(
FECS
B
)
was
p
r
o
p
o
s
ed
to
en
h
an
ce
th
e
ac
cu
r
ac
y
o
f
MRI
b
r
ain
im
ag
e
s
eg
m
e
n
tatio
n
.
T
h
e
p
r
o
p
o
s
ed
h
y
b
r
i
d
ap
p
r
o
ac
h
m
a
x
im
izes
th
e
e
f
f
ic
ien
cy
o
f
FECS
B
in
MRI
b
r
ai
n
im
ag
e
s
eg
m
en
tatio
n
b
y
c
o
m
b
in
in
g
PS
O
with
GW
O.
T
h
e
PS
O
-
GW
O
c
lu
s
t
er
in
g
m
eth
o
d
o
u
tp
e
r
f
o
r
m
s
th
e
co
n
v
en
tio
n
al
FC
M
m
eth
o
d
,
as
s
h
o
wn
b
y
th
e
ex
p
er
im
en
tal
f
i
n
d
in
g
s
.
AB
C
h
as
b
ee
n
em
p
lo
y
e
d
to
o
p
tim
ize
FC
M
f
o
r
MRI
b
r
ain
s
eg
m
en
tatio
n
.
T
h
ese
m
eth
o
d
s
f
o
cu
s
o
n
im
p
r
o
v
in
g
co
n
v
er
g
e
n
ce
s
p
ee
d
an
d
s
eg
m
en
tatio
n
ac
cu
r
ac
y
in
co
m
p
lex
MRI
d
atasets
.
Fo
r
in
s
tan
ce
,
th
e
s
tu
d
y
in
[
2
6
]
i
n
tr
o
d
u
ce
s
a
n
ew
m
et
h
o
d
f
o
r
MRI
b
r
ain
tu
m
o
r
s
eg
m
en
tatio
n
th
at
co
m
b
in
es
th
e
A
B
C
alg
o
r
ith
m
with
FC
M
clu
s
ter
in
g
.
I
t a
d
d
r
ess
es th
e
ch
allen
g
es o
f
s
eg
m
en
tin
g
s
im
ilar
tex
tu
r
e
f
ield
s
in
MRI
im
ag
es b
y
em
p
l
o
y
in
g
a
f
itn
ess
f
u
n
ctio
n
b
ased
o
n
t
wo
-
d
im
en
s
io
n
al
g
r
ey
en
tr
o
p
y
,
d
er
iv
e
d
f
r
o
m
d
is
cr
ete
wav
e
let
tr
an
s
f
o
r
m
s
.
T
h
e
AB
C
alg
o
r
ith
m
o
p
tim
izes
th
r
esh
o
ld
esti
m
atio
n
,
r
esu
ltin
g
i
n
ef
f
icien
t
s
eg
m
en
tatio
n
with
m
in
im
ized
n
o
is
e.
E
x
p
er
im
en
tal
r
esu
lts
d
em
o
n
s
tr
ate
clea
r
tu
m
o
r
d
elin
ea
tio
n
in
s
eg
m
en
ted
im
a
g
es,
en
h
a
n
cin
g
tu
m
o
r
in
ten
s
it
y
v
is
ib
ilit
y
.
Alo
m
o
u
s
h
et
a
l.
[
2
7
]
a
s
p
atial
in
f
o
r
m
atio
n
o
f
f
u
zz
y
clu
s
ter
in
g
-
b
ased
m
ea
n
b
est ar
tific
ial
b
ee
co
lo
n
y
alg
o
r
ith
m
(
SF
C
M
-
Me
an
AB
C
)
is
p
r
esen
ted
.
T
h
is
alg
o
r
ith
m
aim
s
to
en
h
an
ce
m
ed
ical
im
ag
e
s
eg
m
en
tatio
n
,
p
ar
ticu
lar
ly
f
o
r
Ph
a
n
to
m
MR
I
b
r
ain
im
ag
es.
SF
C
M
-
Me
an
AB
C
in
teg
r
ates
s
p
atial
in
f
o
r
m
atio
n
to
m
itig
ate
n
o
is
e
ef
f
ec
ts
an
d
e
m
p
lo
y
s
t
h
e
Me
an
AB
C
alg
o
r
ith
m
to
b
alan
ce
ex
p
l
o
r
atio
n
an
d
e
x
p
l
o
itatio
n
,
im
p
r
o
v
in
g
clu
s
ter
ce
n
ter
o
p
tim
izatio
n
.
T
h
e
m
eth
o
d
p
r
o
v
e
d
p
a
r
ticu
lar
ly
ef
f
ec
tiv
e
at
r
ed
u
cin
g
n
o
i
s
e
s
en
s
itiv
ity
wh
ile
m
ain
tain
in
g
ac
cu
r
ate
s
eg
m
en
t
atio
n
r
esu
lts
co
m
p
ar
e
d
to
g
r
o
u
n
d
tr
u
th
im
a
g
e.
Au
th
o
r
s
in
[
2
8
]
co
m
b
i
n
e
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
5
,
Octo
b
e
r
20
25
:
4
9
1
6
-
4
9
3
2
4920
co
n
ce
p
t
o
f
th
e
FC
M
an
d
f
o
u
r
-
ch
ain
q
u
a
n
tu
m
b
ee
co
lo
n
y
o
p
tim
izatio
n
(
F
QABC
)
.
T
h
e
FQAB
C
alg
o
r
ith
m
o
v
er
co
m
es
th
e
d
r
awb
ac
k
s
o
f
FC
M
wh
ich
is
s
en
s
itiv
e
to
in
itial
clu
s
ter
in
g
ce
n
ter
s
.
Per
f
o
r
m
an
ce
ev
alu
atio
n
ex
p
er
im
en
ts
with
FC
M,
FAB
C
an
d
FQAB
C
h
av
e
b
ee
n
d
o
n
e
o
n
r
ea
l
a
n
d
m
a
g
n
etic
r
eso
n
an
ce
im
ag
es.
T
h
e
ex
p
er
im
en
tal
r
esu
lts
s
h
o
w
th
at
th
e
FQAB
C
alg
o
r
ith
m
is
m
o
r
e
ef
f
ec
tiv
e.
Oth
er
s
tu
d
ies
h
av
e
em
p
lo
y
e
d
th
e
wh
ale
o
p
tim
izatio
n
alg
o
r
ith
m
(
W
OA)
to
en
h
an
ce
FC
M
f
o
r
MRI
b
r
ain
im
ag
e
s
eg
m
en
tatio
n
.
B
y
r
ef
in
in
g
clu
s
ter
ce
n
tr
o
id
s
an
d
en
h
a
n
cin
g
th
e
g
lo
b
al
s
ea
r
ch
ca
p
ab
ilit
y
,
W
OA
-
F
C
M
im
p
r
o
v
es
s
eg
m
en
tatio
n
ac
cu
r
ac
y
,
s
p
ee
d
s
u
p
co
n
v
er
g
e
n
ce
,
an
d
en
h
an
ce
s
r
o
b
u
s
tn
ess
ag
ain
s
t
n
o
is
e
an
d
in
ten
s
ity
v
ar
iatio
n
s
,
m
ak
i
n
g
it a
p
r
o
m
is
in
g
ap
p
r
o
ac
h
f
o
r
MRI
b
r
ain
an
aly
s
is
.
A
n
o
v
el
im
ag
e
s
eg
m
en
tatio
n
m
eth
o
d
co
m
b
in
in
g
FC
M
with
th
e
W
OA
to
en
h
an
ce
s
eg
m
en
t
atio
n
ac
cu
r
ac
y
an
d
n
o
is
e
r
ed
u
ctio
n
was p
r
esen
ted
in
[
2
9
]
.
T
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
ad
d
r
ess
es FC
M
's lim
i
tatio
n
s
,
s
u
ch
as sen
s
it
iv
ity
to
in
itial v
alu
es a
n
d
n
o
is
e,
b
y
u
tili
zin
g
W
OA’
s
g
lo
b
al
o
p
ti
m
izatio
n
ca
p
a
b
ilit
ies.
E
x
p
er
i
m
en
tal
ev
alu
atio
n
s
o
n
s
y
n
th
e
tic
an
d
MRI
im
a
g
es
with
v
ar
io
u
s
n
o
is
e
ty
p
es
s
h
o
w
th
at
th
e
ap
p
r
o
ac
h
s
u
r
p
ass
es
e
x
is
tin
g
tech
n
iq
u
es,
s
u
ch
as
FC
M
an
d
s
tan
d
alo
n
e
W
OA,
b
y
ac
h
iev
in
g
l
o
wer
m
ea
n
s
q
u
ar
e
e
r
r
o
r
(
MSE
)
a
n
d
h
ig
h
er
p
ea
k
s
ig
n
al
-
to
-
n
o
is
e
r
atio
(
PS
NR
)
.
A
s
tu
d
y
in
[
2
9
]
in
tr
o
d
u
ce
s
a
n
ew
ap
p
r
o
ac
h
f
o
r
im
ag
e
s
eg
m
e
n
tatio
n
b
ased
o
n
th
e
W
OA
an
d
FC
M
a
lg
o
r
ith
m
.
Sin
ce
ex
p
lo
r
atio
n
an
d
ex
p
lo
itatio
n
p
h
ases
ar
e
p
er
f
o
r
m
ed
in
n
ea
r
ly
e
q
u
al
n
u
m
b
er
s
o
f
iter
at
io
n
s
s
ep
ar
ately
,
th
e
W
OA
s
im
u
ltan
eo
u
s
ly
s
h
o
ws
b
etter
av
o
id
an
ce
f
r
o
m
lo
ca
l o
p
t
im
a
an
d
s
u
p
er
io
r
co
n
v
er
g
e
n
ce
s
p
ee
d
.
T
o
v
alid
ate
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
p
r
o
p
o
s
ed
s
y
s
tem
,
ex
p
er
im
en
ts
ar
e
co
n
d
u
cted
o
n
s
y
n
th
etic
an
d
MRI
I
m
ag
es
b
y
tak
in
g
v
ar
io
u
s
ty
p
es o
f
n
o
is
e
an
d
th
e
f
in
d
in
g
s
in
d
icate
t
h
at
th
e
p
r
o
p
o
s
ed
m
eth
o
d
is
m
o
r
e
ef
f
icien
t
an
d
ef
f
ec
tiv
e
.
R
ec
en
t
s
tu
d
ies
h
av
e
also
ex
p
lo
r
ed
r
ain
d
r
o
p
o
p
tim
izer
f
o
r
FC
M
in
MRI
b
r
ain
s
eg
m
en
tatio
n
.
I
n
s
tu
d
y
[
3
0
]
an
im
p
r
o
v
e
d
FC
M
clu
s
te
r
in
g
m
eth
o
d
o
p
tim
ized
with
t
h
e
r
ain
d
r
o
p
alg
o
r
ith
m
(
FC
M
-
R
O)
f
o
r
b
r
ain
MRI
s
eg
m
en
tatio
n
was
in
tr
o
d
u
ce
d
.
T
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
in
co
r
p
o
r
ates
a
h
y
b
r
id
f
ilter
f
o
r
n
o
is
e
r
ed
u
cti
o
n
,
ac
h
iev
in
g
a
well
p
a
r
titi
o
n
c
o
ef
f
icien
t
(
PC
)
an
d
p
ar
titi
o
n
en
tr
o
p
y
(
PE)
v
al
u
es
ac
r
o
s
s
f
iv
e
MRI
im
ag
es,
s
ig
n
if
ican
tly
o
u
tp
er
f
o
r
m
in
g
t
r
ad
itio
n
al
FC
M.
T
h
e
s
tu
d
y
d
em
o
n
s
tr
ates
th
at
FC
M
-
R
O
ef
f
ec
tiv
ely
e
x
tr
ac
ts
lesi
o
n
s
,
th
er
eb
y
im
p
r
o
v
i
n
g
d
ia
g
n
o
s
tic
ac
cu
r
ac
y
in
m
ed
ical
i
m
ag
in
g
.
3.
P
RO
P
O
SE
D
M
E
T
H
O
D
I
n
th
is
s
ec
tio
n
,
p
r
io
r
to
d
elv
i
n
g
in
t
o
t
h
e
d
etails
o
f
th
e
p
r
o
p
o
s
ed
FC
M
-
AB
C
o
p
tim
izer
m
eth
o
d
,
we
will
f
ir
s
t
r
ev
iew
th
e
FC
M
an
d
AB
C
alg
o
r
ith
m
s
.
T
h
is
f
o
u
n
d
atio
n
al
o
v
e
r
v
iew
is
ess
en
tial
f
o
r
u
n
d
er
s
tan
d
in
g
h
o
w
th
ese
two
m
eth
o
d
o
lo
g
ie
s
ar
e
in
teg
r
ated
to
ad
d
r
ess
th
e
lim
itatio
n
s
o
f
tr
ad
itio
n
al
F
C
M,
p
ar
ticu
lar
ly
in
ter
m
s
o
f
p
a
r
am
eter
in
itializatio
n
,
clu
s
ter
ce
n
ter
o
p
tim
izatio
n
,
an
d
t
h
e
ch
allen
g
e
o
f
lo
ca
l
m
i
n
im
a.
B
y
r
ev
is
itin
g
th
e
co
r
e
p
r
in
ci
p
les
an
d
m
ec
h
a
n
is
m
s
o
f
b
o
th
alg
o
r
ith
m
s
,
we
aim
to
p
r
o
v
id
e
a
c
o
m
p
r
e
h
en
s
iv
e
co
n
tex
t
f
o
r
th
e
d
ev
elo
p
m
e
n
t
o
f
o
u
r
h
y
b
r
id
a
p
p
r
o
ac
h
,
h
ig
h
lig
h
tin
g
th
e
s
y
n
e
r
g
is
tic
b
en
ef
its
th
at
ar
is
e
f
r
o
m
th
eir
co
m
b
in
atio
n
.
Ad
d
itio
n
ally
,
t
h
is
b
ac
k
g
r
o
u
n
d
will
f
ac
ilit
ate
a
clea
r
e
r
u
n
d
e
r
s
tan
d
in
g
o
f
h
o
w
t
h
e
p
r
o
p
o
s
ed
o
p
tim
izer
e
n
h
an
ce
s
th
e
r
o
b
u
s
tn
ess
an
d
ac
cu
r
ac
y
o
f
b
r
ain
im
a
g
e
s
eg
m
e
n
tatio
n
task
s
,
s
ettin
g
th
e
s
tag
e
f
o
r
its
ap
p
licatio
n
i
n
co
m
p
lex
r
ea
l
-
wo
r
ld
s
ce
n
ar
i
o
s
.
3
.
1
.
F
uzzy
c
-
m
e
a
ns
a
lg
o
rit
h
m
T
h
e
FC
M
alg
o
r
ith
m
b
el
o
n
g
s
to
th
e
f
am
ily
o
f
clu
s
ter
in
g
alg
o
r
ith
m
s
b
ased
o
n
f
u
zz
y
f
u
n
ctio
n
o
p
tim
izatio
n
.
T
h
e
s
tan
d
ar
d
v
er
s
io
n
is
f
ir
s
tly
in
tr
o
d
u
ce
d
b
y
Du
n
n
a
n
d
g
e
n
er
alize
d
b
y
B
ez
d
ek
[
1
0
]
.
I
t
h
as
u
n
d
er
g
o
n
e
m
an
y
in
ter
v
en
tio
n
s
lead
in
g
t
o
a
lo
t
o
f
al
g
o
r
ith
m
s
.
All
th
ese
alg
o
r
ith
m
s
ar
e
co
n
s
id
er
e
d
as
s
o
f
t
clu
s
ter
in
g
in
t
h
e
way
th
at
ea
c
h
elem
en
t
o
f
t
h
e
d
ata
to
b
e
cl
u
s
ter
ed
m
ay
b
elo
n
g
to
m
o
r
e
t
h
an
o
n
e
clu
s
ter
with
d
ef
er
en
t
d
eg
r
ee
s
o
f
m
em
b
er
s
h
ip
.
T
h
e
o
b
jectiv
e
f
u
n
ctio
n
is
o
p
tim
ized
in
an
iter
ativ
e
way
an
d
at
th
e
en
d
o
f
th
e
p
r
o
ce
s
s
; e
ac
h
elem
en
t is ass
ig
n
ed
to
th
e
clu
s
ter
in
wh
ich
it
h
as th
e
h
ig
h
est m
em
b
er
s
h
i
p
.
L
et
=
(
1
,
2
,
…
,
)
an
im
ag
e
o
f
N
p
ix
els
to
b
e
clu
s
ter
ed
in
t
o
K
(
2
<
≤
)
clu
s
ter
s
,
wh
er
e
r
ep
r
esen
ts
d
ata
f
ea
tu
r
es.
T
h
e
s
tan
d
ar
d
FC
M
o
b
jectiv
e
f
u
n
cti
o
n
[
1
0
]
is
f
o
r
m
u
lated
as
(
1
)
:
(
,
)
=
∑
∑
,
2
(
,
)
=
1
=
1
(
1
)
an
d
=
(
1
,
2
,
…
,
)
ar
e
th
e
m
em
b
er
s
h
ip
s
d
e
g
r
ee
s
m
atr
ix
an
d
a
v
ec
to
r
o
f
clu
s
ter
s
ce
n
ter
s
r
esp
ec
tiv
ely
.
∈
[
1
,
∞
[
is
to
c
o
n
tr
o
l
f
u
zz
in
ess
,
2
(
,
)
is
th
e
g
r
a
y
s
ca
le
E
u
clid
ea
n
d
is
tan
c
e
an
d
,
is
th
e
m
em
b
er
s
h
ip
d
eg
r
ee
o
f
th
e
p
ix
el
in
th
e
ℎ
clu
s
ter
wh
ich
m
u
s
t c
h
ec
k
t
h
e
f
o
ll
o
win
g
co
n
s
tr
ain
ts
:
∀
∈
[
1
,
]
,
∈
[
1
,
]
:
∑
,
=
1
=
1
,
∈
[
0
,
1
]
,
0
≤
∑
,
≤
=
1
(
2
)
An
alter
n
ate
o
p
tim
izatio
n
is
ap
p
lied
o
n
th
e
m
em
b
er
s
h
ip
f
u
n
ctio
n
,
an
d
clu
s
ter
s
ce
n
ter
s
u
s
in
g
(
3
)
a
n
d
(
4
)
:
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:
2088
-
8
7
0
8
F
u
z
z
y
clu
s
ter
in
g
o
p
timiz
a
tio
n
b
a
s
ed
a
r
tifi
cia
l b
ee
co
lo
n
y
a
l
g
o
r
ith
m
fo
r
…
(
C
h
a
kir Mo
kh
ta
r
i
)
4921
,
=
(
2
(
,
)
)
1
1
−
∑
(
2
(
,
)
)
1
1
−
=
1
(
3
)
an
d
=
∑
,
=
1
∑
,
=
1
(
4
)
T
h
e
FC
M
alg
o
r
ith
m
b
eg
in
s
with
a
r
an
d
o
m
in
itializatio
n
o
f
clu
s
ter
ce
n
ter
s
an
d
iter
ativ
ely
u
p
d
ates
th
em
u
s
in
g
f
o
r
m
u
las
(
3
)
an
d
(
4
)
u
n
til
n
o
f
u
r
th
er
im
p
r
o
v
em
e
n
t
in
th
eir
p
o
s
itio
n
s
is
o
b
s
er
v
e
d
.
On
ce
t
h
e
clu
s
ter
ce
n
ter
s
ar
e
s
tab
ilized
,
ea
ch
p
i
x
el
j
in
th
e
im
ag
e
is
ass
ig
n
ed
to
th
e
clu
s
ter
f
o
r
wh
ich
it
h
as
t
h
e
m
ax
im
u
m
f
u
zz
y
m
em
b
er
s
h
ip
d
e
g
r
ee
.
T
h
is
p
r
o
ce
s
s
en
s
u
r
es
th
at
ev
er
y
p
ix
el
is
ass
o
ciate
d
w
ith
th
e
m
o
s
t
r
el
ev
an
t
clu
s
ter
b
ased
o
n
its
d
eg
r
ee
o
f
b
el
o
n
g
in
g
n
ess
,
as d
eter
m
in
ed
b
y
th
e
alg
o
r
ith
m
'
s
iter
ativ
e
o
p
tim
izatio
n
.
As
d
is
cu
s
s
ed
in
s
ec
tio
n
2
,
f
o
r
m
u
latin
g
a
g
lo
b
al
s
o
lu
tio
n
th
a
t
ef
f
ec
tiv
ely
ac
co
u
n
ts
f
o
r
all
p
ar
am
eter
s
o
f
FC
M
alg
o
r
ith
m
p
r
esen
ts
s
ig
n
if
ican
t
ch
allen
g
es.
I
n
f
ac
t,
t
o
ad
d
r
ess
th
ese
ch
allen
g
es
a
n
d
s
o
lv
e
th
e
c
o
m
p
lex
o
p
tim
izatio
n
p
r
o
b
lem
p
o
s
ed
b
y
th
e
FC
M
alg
o
r
ith
m
,
we
p
r
o
p
o
s
e
in
th
is
wo
r
k
a
n
ev
o
lu
tio
n
ar
y
alg
o
r
ith
m
(
E
A)
b
ased
o
n
th
e
AB
C
a
lg
o
r
ith
m
.
B
y
in
teg
r
atin
g
th
e
AB
C
alg
o
r
ith
m
in
to
th
e
FC
M
f
r
am
ew
o
r
k
,
o
u
r
p
r
o
p
o
s
ed
m
eth
o
d
s
ee
k
s
to
s
im
u
ltan
e
o
u
s
ly
o
p
tim
ize
m
u
ltip
le
p
ar
a
m
eter
s
,
in
clu
d
in
g
th
e
n
u
m
b
e
r
o
f
clu
s
ter
s
,
th
eir
in
itializatio
n
,
an
d
th
e
o
v
e
r
all
o
b
jectiv
e
f
u
n
ctio
n
.
T
h
e
s
tr
en
g
th
s
o
f
th
e
AB
C
alg
o
r
ith
m
s
u
c
h
as
its
s
tr
o
n
g
g
lo
b
al
s
ea
r
ch
ca
p
ab
ilit
y
,
s
im
p
licity
,
an
d
ea
s
e
o
f
im
p
lem
en
tatio
n
a
r
e
lev
er
a
g
ed
t
o
e
n
s
u
r
e
th
at
th
e
FC
M
alg
o
r
ith
m
o
p
er
ates a
t its
f
u
ll p
o
ten
tial,
d
eliv
er
in
g
m
o
r
e
ac
cu
r
ate
an
d
r
e
liab
le
r
esu
lts
.
3
.
2
.
Art
if
ici
a
l
bee
co
lo
ny
a
l
g
o
rit
hm
AB
C
alg
o
r
ith
m
is
an
ev
o
lu
tio
n
ar
y
alg
o
r
ith
m
b
io
-
in
s
p
ir
e
d
[
1
1
]
.
I
t
im
itates
th
e
h
o
n
ey
b
e
e
s
war
m
s
in
f
o
o
d
f
o
r
ag
in
g
an
d
s
u
cc
ess
f
u
lly
ap
p
lied
in
v
ar
i
o
u
s
o
p
tim
izatio
n
p
r
o
b
lem
s
.
I
t
o
p
e
r
ates
th
r
o
u
g
h
t
h
e
co
llab
o
r
atio
n
o
f
t
h
r
ee
ty
p
es
o
f
b
ee
s
:
em
p
lo
y
e
d
b
ee
s
,
o
n
lo
o
k
er
b
ee
s
,
an
d
s
co
u
t b
ee
s
,
ea
ch
with
d
is
tin
ct
r
o
les
in
th
e
s
ea
r
ch
f
o
r
n
ec
tar
(
o
r
o
p
tim
al
s
o
lu
tio
n
s
)
.
T
h
e
em
p
l
o
y
ed
b
ee
s
ar
e
r
esp
o
n
s
ib
le
f
o
r
e
x
p
lo
itin
g
k
n
o
wn
f
o
o
d
s
o
u
r
ce
s
.
E
ac
h
em
p
lo
y
ed
b
ee
r
ep
r
esen
ts
a
p
o
ten
tial
s
o
lu
tio
n
an
d
ass
ess
e
s
its
q
u
ality
b
ased
o
n
a
f
itn
ess
f
u
n
ctio
n
.
T
h
ey
s
e
ar
ch
in
th
e
v
icin
ity
o
f
th
eir
ass
ig
n
ed
f
o
o
d
s
o
u
r
ce
an
d
ca
n
ad
ju
s
t
th
eir
p
o
s
itio
n
to
im
p
r
o
v
e
th
e
s
o
lu
tio
n
.
I
f
a
b
ee
f
in
d
s
a
b
etter
s
o
lu
tio
n
,
it
s
h
ar
es
th
is
in
f
o
r
m
atio
n
with
th
e
o
n
lo
o
k
e
r
b
e
es.
T
h
e
later
m
o
n
ito
r
th
e
q
u
a
lity
o
f
f
o
o
d
s
o
u
r
ce
s
s
h
ar
ed
b
y
em
p
lo
y
ed
b
ee
s
.
T
h
e
y
u
tili
ze
a
p
r
o
b
a
b
ilit
y
-
b
ased
s
elec
tio
n
m
ec
h
an
is
m
to
ch
o
o
s
e
wh
ich
f
o
o
d
s
o
u
r
ce
to
ex
p
lo
r
e
b
ased
o
n
its
f
itn
ess
.
B
y
co
n
ce
n
tr
atin
g
o
n
th
e
m
o
s
t
p
r
o
m
is
in
g
s
o
u
r
ce
s
,
o
n
lo
o
k
er
b
ee
s
co
n
tr
ib
u
te
to
th
e
ex
p
l
o
itatio
n
p
h
ase
o
f
th
e
alg
o
r
ith
m
ca
lled
also
lo
ca
l
s
ea
r
ch
,
f
u
r
th
er
r
ef
in
i
n
g
th
e
s
ea
r
ch
f
o
r
o
p
tim
al
s
o
lu
tio
n
s
.
T
h
e
s
co
u
t
b
ee
s
p
r
e
s
en
t
th
e
ex
p
lo
r
ativ
e
p
h
ase
an
d
th
ey
ar
e
r
esp
o
n
s
ib
le
f
o
r
e
x
p
lo
r
in
g
n
ew
a
r
ea
s
o
f
th
e
s
ea
r
ch
s
p
ac
e
to
d
is
co
v
er
n
ew
f
o
o
d
s
o
u
r
ce
s
.
T
h
eir
r
an
d
o
m
s
ea
r
ch
h
el
p
s
m
ain
tain
d
iv
er
s
ity
in
th
e
p
o
p
u
latio
n
a
n
d
p
r
ev
e
n
ts
th
e
al
g
o
r
ith
m
f
r
o
m
g
ettin
g
tr
ap
p
e
d
in
lo
ca
l
o
p
tim
a.
T
h
r
o
u
g
h
th
e
co
o
r
d
in
ate
d
e
f
f
o
r
ts
o
f
th
ese
th
r
ee
ty
p
es o
f
b
ee
s
,
th
e
AB
C
a
lg
o
r
ith
m
ef
f
icien
tly
e
x
p
lo
r
es a
n
d
e
x
p
lo
its
th
e
s
o
lu
tio
n
s
p
ac
e.
T
h
e
AB
C
alg
o
r
ith
m
b
eg
in
s
f
o
o
d
f
o
r
ag
in
g
(
s
o
lu
tio
n
s
ea
r
ch
)
b
y
p
r
o
d
u
ci
n
g
r
a
n
d
o
m
ly
an
in
itial
p
o
p
u
la
tio
n
o
f
NS
b
ee
s
in
s
ea
r
ch
s
p
ac
e
ac
co
r
d
i
n
g
to
(
5
)
:
=
+
(
0
,
1
)
∗
(
−
)
=
1
,
…
,
(
5
)
wh
er
e
is
a
b
ee
,
an
d
ar
e
th
e
u
p
p
er
an
d
th
e
lo
we
r
v
alu
es o
f
t
h
e
s
ea
r
ch
s
p
ac
e
r
esp
ec
tiv
ely
.
Af
ter
th
e
in
itializatio
n
p
h
ase,
th
e
AB
C
alg
o
r
ith
m
ev
alu
ates
th
e
in
itial
p
o
p
u
latio
n
an
d
p
e
r
f
o
r
m
s
th
e
th
r
ee
f
o
llo
win
g
s
tep
s
u
n
til
c
o
n
v
er
g
en
ce
to
th
e
o
p
tim
al
g
l
o
b
al
s
o
lu
tio
n
(
s
atis
f
ac
to
r
y
f
it
n
ess
)
o
r
m
ax
im
u
m
iter
atio
n
s
.
Step
1
: E
m
p
lo
y
e
d
b
ee
p
h
ase
−
E
ac
h
em
p
lo
y
ed
b
ee
g
en
e
r
ates a
n
ew
s
o
lu
tio
n
in
t
h
e
n
eig
h
b
o
r
h
o
o
d
u
s
in
g
ex
p
r
ess
io
n
(
6
)
:
+
1
=
+
(
−
)
=
1
,
…
,
(
6
)
wh
er
e
is
a
r
an
d
o
m
n
u
m
b
er
in
th
e
r
an
g
e
[
−
1
,
1
]
,
an
d
ar
e
th
e
i
th
s
o
l
u
tio
n
an
d
th
e
b
est
s
o
lu
tio
n
o
f
k
th
iter
atio
n
r
esp
ec
tiv
ely
an
d
+
1
r
ep
r
esen
ts
th
e
u
p
d
ate
d
s
o
lu
tio
n
.
−
E
v
alu
ate
th
e
n
ew
s
o
lu
tio
n
’
s
f
it
n
ess
.
−
I
f
th
e
n
ew
s
o
lu
tio
n
is
b
etter
,
u
p
d
ate
th
e
cu
r
r
en
t so
lu
tio
n
an
d
m
em
o
r
ize
th
e
n
ew
o
n
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
5
,
Octo
b
e
r
20
25
:
4
9
1
6
-
4
9
3
2
4922
Step
2
: O
n
lo
o
k
e
r
b
ee
p
h
ase
−
E
ac
h
o
n
lo
o
k
er
b
ee
s
elec
ts
a
s
o
u
r
ce
f
o
o
d
with
a
p
r
o
b
a
b
ilit
y
p
r
o
p
o
r
tio
n
ally
to
t
h
e
q
u
ality
o
f
th
e
n
ec
ta
r
(
th
e
s
o
lu
tio
n
)
.
T
h
e
p
r
o
b
a
b
ilit
y
o
f
s
elec
tin
g
th
e
s
o
u
r
ce
f
o
o
d
is
ca
lcu
lated
ac
co
r
d
in
g
to
(
7
)
:
=
(
)
∑
(
)
=
1
=
1
,
…
,
(
7
)
wh
er
e
(
)
is
th
e
f
itn
ess
o
f
th
e
s
o
lu
tio
n
.
−
Gen
er
ate
n
ew
s
o
lu
tio
n
f
o
r
t
h
e
s
elec
ted
f
o
o
d
s
o
u
r
ce
u
s
in
g
(
6
)
−
Up
d
ate
s
o
lu
tio
n
s
if
im
p
r
o
v
em
en
ts
ar
e
f
o
u
n
d
.
Step
3
: Sco
u
t b
ee
p
h
ase
−
If
an
y
f
o
o
d
s
o
u
r
ce
p
r
esen
ts
n
o
im
p
r
o
v
e
m
en
ts
f
o
r
a
n
u
m
b
er
o
f
cy
cles,
it is
ab
an
d
o
n
ed
.
−
I
f
s
o
,
r
ep
lace
it with
a
n
ew
r
an
d
o
m
f
o
o
d
s
o
u
r
ce
u
s
in
g
(
5
)
.
−
R
etu
r
n
to
th
e
em
p
lo
y
ed
b
ee
p
h
ase
.
Step
4
: T
er
m
in
atio
n
−
If
th
e
s
to
p
p
in
g
cr
iter
io
n
is
m
et
o
r
th
e
m
ax
im
al
iter
atio
n
n
u
m
b
er
is
r
ea
ch
ed
,
r
etu
r
n
th
e
b
e
s
t
b
ee
(
o
p
tim
al
s
o
lu
tio
n
)
.
T
o
r
ea
ch
th
e
g
lo
b
al
o
p
tim
u
m
,
th
e
AB
C
Alg
o
r
ith
m
b
alan
ce
b
etwe
en
ex
p
l
o
itativ
e
s
ea
r
ch
an
d
e
x
p
lo
r
at
o
r
y
s
ea
r
ch
an
d
th
e
b
o
th
in
r
an
d
o
m
m
an
n
er
.
3
.
3
.
P
r
o
po
s
ed
F
CM
-
AB
C
o
p
t
im
izer
m
e
t
ho
d
I
n
th
is
wo
r
k
,
a
n
ew
en
h
an
ce
m
en
t
o
f
FC
M
ca
lled
FC
M
-
A
B
C
o
p
tim
izer
i
s
in
tr
o
d
u
ce
d
;
it
is
b
ased
o
n
th
e
AB
C
a
lg
o
r
ith
m
.
Alth
o
u
g
h
th
e
FC
M
h
as a
d
v
an
tag
es lik
e
ef
f
icac
y
,
s
im
p
licity
an
d
co
m
p
u
tatio
n
al
ef
f
icien
cy
,
it
n
o
n
eth
eless
h
as
m
ajo
r
d
r
aw
b
ac
k
s
s
u
ch
as
n
u
m
b
er
o
f
clu
s
t
er
s
,
clu
s
ter
ce
n
ter
s
v
alu
es
an
d
is
ea
s
ily
tr
ap
p
ed
in
lo
ca
l
o
p
tim
a.
So
,
th
e
m
ai
n
o
b
jectiv
e
is
to
o
v
er
c
o
m
e
th
ese
m
ajo
r
d
r
aw
b
ac
k
s
th
at
will
af
f
ec
t
th
e
clu
s
ter
in
g
in
ter
m
o
f
p
r
ec
is
io
n
.
Fo
r
th
is
p
u
r
p
o
s
e,
we
im
p
r
o
v
e
th
e
FC
M
clu
s
ter
in
g
b
y
e
x
p
lo
itin
g
AB
C
al
g
o
r
ith
m
in
o
r
d
e
r
to
f
in
d
s
im
u
ltan
eo
u
s
ly
th
e
r
ig
h
t
n
u
m
b
er
o
f
clu
s
ter
s
an
d
th
e
o
p
tim
al
clu
s
ter
s
ce
n
ter
s
f
o
r
a
g
iv
en
im
ag
e
I
o
f
N
p
ix
els.
AB
C
alg
o
r
ith
m
co
m
b
i
n
es
b
etwe
en
ex
p
lo
itatio
n
an
d
ex
p
lo
r
atio
n
to
f
in
d
th
e
o
p
ti
m
al
v
alu
es
o
f
FC
M
p
ar
am
eter
s
.
I
t e
n
s
u
r
es th
e
s
ea
r
ch
in
g
in
all
d
ir
ec
tio
n
s
in
th
e
s
o
lu
tio
n
s
p
ac
e.
T
o
ac
h
iev
e
t
h
is
o
b
jectiv
e,
f
ir
s
t,
ea
ch
b
ee
b
i
co
n
s
is
ts
o
f
a
v
e
cto
r
co
m
p
r
is
in
g
two
p
a
r
ts
.
T
h
e
f
ir
s
t
p
ar
t
m
ain
tain
s
th
e
n
u
m
b
er
o
f
clu
s
ter
s
wh
ile
th
e
s
ec
o
n
d
m
ain
tai
n
s
th
e
v
alu
es
o
f
th
e
ce
n
te
r
s
o
f
th
ese
clu
s
ter
s
in
Fig
u
r
e
1
.
N
b
c
i
1
2
………….
Fig
u
r
e
1
.
Stru
ctu
r
e
o
f
a
b
ee
wh
er
e
is
th
e
n
u
m
b
er
o
f
clu
s
te
r
s
o
f
th
e
im
ag
e
to
b
e
s
eg
m
en
t
ed
.
T
h
is
n
u
m
b
er
is
b
etwe
en
2
an
d
m
ax
im
u
m
n
u
m
b
er
o
f
clu
s
ter
s
(
)
.
is
th
e
v
a
lu
e
o
f
th
e
ce
n
ter
o
f
th
e
b
ee
b
i
wh
ich
is
t
h
e
g
r
ey
lev
els
o
f
th
e
in
p
u
t im
ag
e
I
.
Seco
n
d
,
we
d
ev
el
o
p
a
n
ew
o
b
jectiv
e
f
u
n
ctio
n
F
in
o
r
d
er
t
o
ev
alu
ate
s
o
lu
tio
n
s
f
itn
ess
.
T
h
i
s
f
u
n
ctio
n
en
s
u
r
es
th
e
o
p
tim
al
v
alu
es
o
f
th
e
clu
s
ter
’
s
ce
n
ter
s
an
d
th
e
r
ig
h
t
n
u
m
b
er
o
f
clu
s
ter
s
.
I
t
ex
p
lo
its
th
e
o
b
jectiv
e
f
u
n
ctio
n
o
f
th
e
FC
M
alg
o
r
ith
m
an
d
a
v
alid
ity
in
d
e
x
.
I
t is d
e
f
in
ed
as:
(
)
=
1
1
(
)
+
2
2
(
)
=
1
,
…
,
(
8
)
w
h
e
r
e
1
(
)
c
o
r
r
e
s
p
o
n
d
s
t
o
t
h
e
s
t
an
d
a
r
d
F
C
M
o
b
j
e
c
ti
v
e
f
u
n
c
t
i
o
n
,
w
h
i
c
h
m
i
n
i
m
i
z
es
t
h
e
w
e
i
g
h
t
e
d
s
u
m
o
f
s
q
u
a
r
e
d
d
i
s
t
a
n
c
es
b
e
t
we
e
n
d
a
ta
p
o
i
n
t
s
a
n
d
c
l
u
s
t
e
r
c
e
n
te
r
s
.
T
h
e
s
e
c
o
n
d
t
e
r
m
,
2
(
)
,
r
e
p
r
e
s
e
n
t
s
a
c
l
u
s
t
e
r
i
n
g
v
a
l
i
d
i
t
y
i
n
d
e
x
t
h
a
t
e
v
al
u
a
t
es
t
h
e
q
u
a
l
i
t
y
o
f
t
h
e
r
e
s
u
lt
i
n
g
p
a
r
t
i
ti
o
n
s
i
n
t
e
r
m
s
o
f
c
o
m
p
a
c
t
n
e
s
s
a
n
d
s
e
p
a
r
a
t
i
o
n
.
T
h
e
w
e
i
g
h
t
s
W
1
a
n
d
W
2
c
o
n
t
r
o
l
t
h
e
r
e
l
a
t
i
v
e
i
m
p
o
r
t
a
n
c
e
o
f
e
a
c
h
c
o
m
p
o
n
e
n
t
i
n
t
h
e
o
v
e
r
a
l
l
o
p
t
i
m
iza
t
i
o
n
p
r
o
c
e
s
s
.
T
h
e
m
o
tiv
atio
n
b
e
h
in
d
th
is
h
y
b
r
id
f
o
r
m
u
latio
n
lies
in
ad
d
r
es
s
in
g
th
e
lim
itatio
n
s
o
f
u
s
in
g
FC
M
alo
n
e.
W
h
ile
F
C
M
ef
f
ec
tiv
ely
m
in
im
izes
in
tr
a
-
clu
s
ter
v
ar
ian
ce
,
it
d
o
es
n
o
t
in
h
er
en
tly
e
n
s
u
r
e
well
-
s
ep
ar
ated
o
r
m
ea
n
in
g
f
u
l
clu
s
ter
s
,
esp
ec
ial
ly
wh
en
th
e
o
p
tim
al
n
u
m
b
e
r
o
f
clu
s
ter
s
is
u
n
k
n
o
wn
o
r
th
e
d
ata
co
n
tain
s
o
v
er
lap
p
i
n
g
s
tr
u
ctu
r
es.
I
n
co
r
p
o
r
atin
g
a
v
alid
ity
i
n
d
ex
as
an
ad
d
itio
n
al
cr
iter
io
n
en
h
a
n
ce
s
th
e
ab
ilit
y
o
f
th
e
alg
o
r
ith
m
to
i
d
en
tify
m
o
r
e
c
o
m
p
ac
t a
n
d
d
is
tin
ct
clu
s
ter
s
,
th
er
eb
y
im
p
r
o
v
i
n
g
o
v
er
all
s
eg
m
en
tatio
n
q
u
ality
.
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:
2088
-
8
7
0
8
F
u
z
z
y
clu
s
ter
in
g
o
p
timiz
a
tio
n
b
a
s
ed
a
r
tifi
cia
l b
ee
co
lo
n
y
a
l
g
o
r
ith
m
fo
r
…
(
C
h
a
kir Mo
kh
ta
r
i
)
4923
B
y
co
m
b
in
in
g
b
o
t
h
o
b
jectiv
es,
th
e
p
r
o
p
o
s
ed
f
u
n
ctio
n
en
ab
les
a
b
alan
ce
d
tr
ad
e
-
o
f
f
b
etwe
en
m
in
im
izin
g
with
in
-
clu
s
ter
d
is
to
r
tio
n
(
v
ia
FC
M)
a
n
d
m
ax
i
m
izin
g
clu
s
ter
v
alid
ity
(
v
ia
th
e
in
d
ex
)
.
T
h
is
d
u
al
-
o
b
jectiv
e
ap
p
r
o
ac
h
p
r
o
v
es
p
a
r
ticu
lar
ly
b
en
e
f
icial
in
co
m
p
lex
ap
p
licatio
n
s
s
u
ch
as
b
r
ain
MRI
s
eg
m
en
tatio
n
,
wh
er
e
ac
cu
r
ate
a
n
d
in
ter
p
r
eta
b
le
clu
s
ter
in
g
is
ess
en
tial f
o
r
d
iag
n
o
s
tic
r
eliab
ilit
y
.
B
o
t
h
w
e
i
g
h
ts
W
1
a
n
d
W
2
c
a
n
b
e
a
d
j
u
s
t
e
d
d
e
p
e
n
d
i
n
g
o
n
t
h
e
s
p
e
c
i
f
i
c
r
e
q
u
i
r
e
m
e
n
ts
o
f
t
h
e
a
p
p
li
c
a
t
i
o
n
o
r
b
a
s
e
d
o
n
p
r
i
o
r
k
n
o
w
l
e
d
g
e
a
b
o
u
t
t
h
e
d
a
t
a
s
t
r
u
c
t
u
r
e
.
A
c
c
o
r
d
i
n
g
t
o
t
h
e
s
t
r
u
ct
u
r
e
o
f
b
e
e
b
i
,
1
i
s
d
e
f
i
n
e
d
a
s
(
9
)
:
1
(
)
=
∑
∑
,
2
(
,
)
=
1
=
1
(
9
)
ar
e
th
e
im
ag
e
p
ix
els an
d
ar
e
t
h
e
E
u
clid
ea
n
d
is
tan
ce
.
2
is
a
clu
s
ter
v
alid
ity
in
d
ex
,
k
n
o
wn
as
th
e
I
Mb
alan
ce
d
in
d
ex
(
I
MI
I
n
d
e
x
)
,
p
r
o
p
o
s
ed
b
y
L
iu
et
a
l.
[
3
1
]
to
id
en
tify
th
e
o
p
tim
al
n
u
m
b
er
o
f
clu
s
ter
s
.
I
t is f
o
r
m
ally
d
ef
in
ed
i
n
(
1
0
)
.
2
(
)
=
∑
∑
,
2
(
,
)
=
1
∑
,
2
=
1
=
1
≠
,
2
(
,
)
+
≠
,
2
(
,
)
(
1
0
)
wh
er
e
,
=
∑
,
=
1
∑
,
=
1
.
3
.
3
.
1
.
G
ener
a
l st
eps
o
f
t
he
F
CM
-
AB
C
o
ptim
izer
T
h
e
g
en
er
al
s
tep
s
o
f
t
h
e
FC
M
-
AB
C
o
p
tim
izer
m
eth
o
d
a
r
e
o
u
tlin
ed
as
f
o
llo
ws,
in
teg
r
atin
g
th
e
s
tr
en
g
th
s
o
f
th
e
FC
M
alg
o
r
i
th
m
an
d
th
e
AB
C
o
p
tim
izatio
n
tech
n
i
q
u
e
to
ac
h
iev
e
r
o
b
u
s
t
an
d
ac
c
u
r
ate
s
eg
m
en
tatio
n
r
esu
lts
:
Step
1
:
I
n
itializatio
n
:
we
s
et
t
h
e
m
ax
im
u
m
n
u
m
b
er
o
f
clu
s
ter
s
,
an
d
th
e
n
u
m
b
er
o
f
cy
cle
,
th
en
an
in
itial
p
o
p
u
latio
n
o
f
b
ee
s
is
g
en
er
ated
in
wh
ich
ea
ch
b
ee
,
in
its
f
ir
s
t
p
ar
t
o
u
g
h
t
to
b
e
ass
ig
n
ed
a
r
an
d
o
m
v
alu
e
in
th
e
r
an
g
e
[
2
,
]
,
wh
ile
ea
ch
v
alu
e
in
s
ec
o
n
d
p
ar
t
is
in
itialized
r
an
d
o
m
l
y
u
s
in
g
(
5
)
ac
co
r
d
in
g
to
th
e
g
r
ey
le
v
els
o
f
th
e
im
a
g
e
I
.
Fo
r
ea
ch
b
ee
b
i
,
we
s
et
th
e
co
u
n
te
r
“
no
-
imp
r
o
ve
men
t
-
cy
cle
”
to
0
.
Step
2
:
Fit
n
es
s
ev
alu
atio
n
:
af
ter
ca
lcu
latin
g
th
e
m
em
b
er
s
h
ip
v
alu
e
,
f
o
r
ea
ch
clu
s
ter
ce
n
ter
s
o
f
th
e
b
ee
(
=
1
,
…
,
)
u
s
in
g
(
3
)
,
we
e
v
alu
ate
th
e
f
i
tn
ess
o
f
all
th
e
b
ee
s
in
th
e
p
o
p
u
latio
n
,
(
)
ac
co
r
d
in
g
to
th
e
(
8
)
.
T
h
e
b
ee
with
th
e
b
e
s
t c
o
n
f
ig
u
r
atio
n
is
s
to
r
ed
.
Step
3
:
E
m
p
l
o
y
ed
b
ee
p
h
ase
:
in
th
is
s
tep
,
ea
ch
em
p
lo
y
ed
b
ee
g
en
er
ates
a
n
ew
s
o
lu
tio
n
i
n
th
e
n
eig
h
b
o
r
h
o
o
d
ac
co
r
d
in
g
t
o
(
6
)
.
I
t c
o
n
s
is
ts
o
f
m
o
d
if
y
in
g
ea
ch
ce
n
te
r
o
f
ea
c
h
b
ee
s
lig
h
tly
to
f
in
d
a
b
etter
p
o
s
itio
n
th
r
o
u
g
h
lo
ca
l
ex
p
lo
r
atio
n
wit
h
o
u
t
af
f
ec
tin
g
th
e
n
u
m
b
er
o
f
clu
s
ter
s
.
T
h
en
,
th
e
n
ew
s
o
l
u
tio
n
’
s
f
itn
ess
is
ev
alu
ated
.
I
f
th
e
n
e
w
s
o
lu
tio
n
is
b
etter
,
u
p
d
ate
th
e
cu
r
r
e
n
t
s
o
lu
tio
n
.
Oth
er
wis
e
in
cr
ea
s
e
th
e
co
u
n
ter
“
no
-
imp
r
o
ve
men
t
-
cy
cl
e
”
.
Step
4
:
On
lo
o
k
er
b
ee
p
h
ase
:
b
ased
o
n
th
e
f
itn
ess
v
alu
es,
w
e
ass
ig
n
p
r
o
b
ab
ilit
y
to
ea
ch
s
o
lu
tio
n
u
s
in
g
(
7
)
.
Acc
o
r
d
in
g
t
o
th
ese
p
r
o
b
ab
ilit
ies,
ea
ch
o
n
lo
o
k
er
b
ee
ch
o
o
s
es a
s
o
lu
tio
n
an
d
ap
p
lies
m
o
d
i
f
icatio
n
s
u
s
in
g
(
6
)
to
f
u
r
th
e
r
r
ef
in
e
th
e
clu
s
ter
s
ce
n
ter
s
.
Step
5
:
Sco
u
t
b
ee
p
h
ase
:
to
en
h
an
ce
th
e
ca
p
ab
ilit
y
to
e
x
p
lo
it
th
e
g
lo
b
al
s
ea
r
ch
,
we
s
o
r
t
th
e
b
ee
s
ac
co
r
d
in
g
to
(
1
0
)
an
d
we
ab
an
d
o
n
all
b
ee
s
th
at
th
e
“
no
-
imp
r
o
ve
men
t
-
cy
cle
”
ex
ce
ed
s
.
I
f
an
y
ab
an
d
o
n
ed
b
ee
b
elo
n
g
s
to
th
e
h
ig
h
est
b
ee
s
,
we
r
ep
lace
th
e
ab
an
d
o
n
ed
b
ee
s
with
n
ew
co
n
f
ig
u
r
atio
n
s
,
a
r
an
d
o
m
n
u
m
b
er
o
f
cl
u
s
ter
s
an
d
n
ew
cl
u
s
ter
ce
n
ter
s
u
s
in
g
(
5
)
,
else
w
e
k
ee
p
th
e
n
u
m
b
er
o
f
clu
s
ter
s
an
d
we
r
eset
r
an
d
o
m
l
y
o
n
l
y
th
e
clu
s
ter
ce
n
t
er
s
.
Step
6
:
L
o
o
p
:
s
tep
s
f
r
o
m
2
to
5
ar
e
r
e
p
ea
ted
u
n
til
th
e
o
b
je
ctiv
e
f
u
n
ctio
n
b
ec
am
e
less
th
an
a
th
r
esh
o
ld
o
r
m
ax
im
u
m
n
u
m
b
e
r
o
f
iter
atio
n
s
is
r
ea
ch
ed
.
S
t
e
p
7
:
T
e
r
m
i
n
at
i
o
n
:
f
i
n
al
l
y
,
we
u
s
e
t
h
e
b
es
t
c
o
n
f
i
g
u
r
a
ti
o
n
s
to
r
e
d
o
f
t
h
e
n
u
m
b
e
r
o
f
c
l
u
s
t
e
r
s
a
n
d
t
h
e
i
r
c
e
n
t
e
r
s
to
p
e
r
f
o
r
m
a
l
as
t
c
a
l
c
u
l
a
t
i
o
n
o
f
p
i
x
e
l
m
e
m
b
e
r
s
h
i
p
s
,
a
c
c
o
r
d
i
n
g
t
o
F
C
M
.
W
e
a
s
s
i
g
n
e
a
c
h
p
i
x
e
l
o
f
t
h
e
i
m
a
g
e
t
o
c
e
n
t
e
r
f
o
r
w
h
i
c
h
t
h
e
m
e
m
b
e
r
s
h
i
p
s
,
i
s
h
i
g
h
e
r
f
o
r
t
h
e
p
u
r
p
o
s
e
t
o
g
e
n
e
r
a
t
e
t
h
e
s
e
g
m
e
n
t
e
d
i
m
a
g
e
.
3
.
3
.
2
.
F
CM
-
AB
C
o
ptim
izer
a
lg
o
rit
hm
Ou
r
p
r
o
p
o
s
ed
m
eth
o
d
is
s
u
m
m
ar
ized
in
th
e
p
s
eu
d
o
c
o
d
e
p
r
esen
ted
in
Fig
u
r
e
2
.
T
h
e
p
s
eu
d
o
co
d
e
o
u
tlin
es
th
e
k
ey
s
tep
s
an
d
lo
g
ic
o
f
th
e
FC
M
-
AB
C
o
p
ti
m
izer
,
h
ig
h
lig
h
tin
g
h
o
w
th
e
AB
C
alg
o
r
ith
m
is
in
teg
r
ated
with
th
e
FC
M
f
r
am
ewo
r
k
to
ac
h
iev
e
r
o
b
u
s
t
an
d
ac
cu
r
ate
s
eg
m
e
n
tatio
n
r
esu
l
ts
.
E
ac
h
s
tep
in
th
e
p
s
eu
d
o
co
d
e
co
r
r
esp
o
n
d
s
to
a
s
p
ec
if
ic
p
h
ase
o
f
th
e
o
p
tim
iza
tio
n
p
r
o
ce
s
s
,
en
s
u
r
in
g
clar
ity
an
d
r
ep
r
o
d
u
cib
ilit
y
o
f
th
e
m
eth
o
d
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
5
,
Octo
b
e
r
20
25
:
4
9
1
6
-
4
9
3
2
4924
FC
M
-
A
B
C
o
p
tim
izer
alg
o
r
ith
m
Input:
original image
I
1.
fix the parameters
MaxNbc, NS, є, NBcycle, L lowest bees, MaxIteration.
2.
generate randomly an initial population of bees
(
=
1
,
2
,
…
,
)
.
3.
it=0
4.
for each bee
, fix
“
no
-
improvement
-
cycle
”
to
0
.
5.
repeat
6.
it=it+1
7.
for each bee
calculate the membership value
,
using (3)
calculate the fitness function
(
)
according to (8).
8.
endfor
9.
select the lowest fitness
Fl,
memorize the best solution
Bbest
.
1
0
.
for each bee
i
generate a new solution
bnew
according to (6).
evaluate the
bnew
’s fitness.
If
bnew
is better,
b
i
=
bnew
.
else
“
no
-
improvement
-
cycle
”
++.
calculate the solution probability using (7).
1
1
.
endfor
1
2
.
applied
greedy
algorithm
to
update
solutions
tha
t
have
the
highest
proba
bilities
using (6).
1
3
.
evaluate their fitness according to (10).
1
4
.
ElitBee
=
L lowest bees
1
5
.
for each bee
if
“
no
-
improvement
-
cycle
”
>
NBcycle
if
∈
,
re
pl
ac
e
with
ne
w
cl
us
te
rs
ce
nt
er
s
wi
th
ou
t
a
ff
ec
ti
ng
th
e
nu
mb
er
of
clusters
Nbc
i
else generate a new solution for
according to (5).
1
6
.
endfor
1
7
.
until (
Fl
<
є or it
>=
MaxIteration)
1
8
.
Calculate the membership value
,
according to
Bbest
.
Fig
u
r
e
2
.
Ps
eu
d
o
co
d
e
o
f
FC
M
-
AB
C
o
p
tim
izer
4.
E
XP
E
R
I
M
E
N
T
A
L
RE
SUL
T
S
T
h
e
p
e
r
f
o
r
m
a
n
c
e
o
f
t
h
e
FC
M
-
A
B
C
o
p
ti
m
i
z
e
r
a
l
g
o
r
i
t
h
m
d
ep
e
n
d
s
o
n
s
e
v
e
r
a
l
k
e
y
p
a
r
a
m
e
te
r
s
.
T
h
ese
p
a
r
a
m
e
t
e
r
s
a
r
e
s
e
l
e
ct
e
d
t
o
b
a
l
an
c
e
e
x
p
l
o
r
a
t
i
o
n
,
e
x
p
l
o
it
a
ti
o
n
,
a
n
d
c
o
m
p
u
t
a
t
i
o
n
a
l
e
f
f
i
c
i
e
n
c
y
.
T
h
e
p
o
p
u
l
a
t
i
o
n
s
i
z
e
r
e
f
e
r
s
t
o
t
h
e
t
o
t
a
l
n
u
m
b
e
r
o
f
b
e
e
s
,
i
n
c
l
u
d
i
n
g
e
m
p
l
o
y
e
d
,
o
n
l
o
o
k
e
r
,
a
n
d
s
c
o
u
t
b
e
es
,
t
y
p
i
c
a
ll
y
s
e
t
b
e
t
w
ee
n
5
0
a
n
d
1
0
0
.
T
h
i
s
r
a
n
g
e
b
a
l
a
n
c
e
s
e
x
p
l
o
r
a
t
i
o
n
a
n
d
c
o
m
p
u
t
a
t
i
o
n
a
l
e
f
f
i
c
i
e
n
c
y
:
a
l
a
r
g
e
r
p
o
p
u
l
a
ti
o
n
e
n
h
a
n
c
e
s
s
o
l
u
t
i
o
n
d
i
v
e
r
s
i
t
y
a
n
d
s
e
a
r
c
h
s
p
a
ce
e
x
p
l
o
r
a
t
i
o
n
,
h
e
l
p
i
n
g
a
v
o
i
d
l
o
c
a
l
o
p
t
i
m
a
,
w
h
il
e
a
s
m
al
l
e
r
s
i
ze
r
e
d
u
c
e
s
c
o
m
p
u
t
a
ti
o
n
a
l
o
v
e
r
h
e
a
d
.
F
o
r
b
r
a
i
n
M
R
I
s
e
g
m
e
n
t
a
t
i
o
n
,
a
p
o
p
u
l
a
t
i
o
n
s
iz
e
o
f
5
0
i
s
c
h
o
s
e
n
a
s
i
t
e
f
f
ec
t
i
v
ely
e
x
p
l
o
r
e
s
t
h
e
h
i
g
h
-
d
i
m
e
n
s
i
o
n
a
l
s
e
a
r
c
h
s
p
a
c
e
o
f
c
lu
s
t
e
r
c
e
n
t
e
r
s
wi
t
h
o
u
t
i
n
c
u
r
r
i
n
g
e
x
c
e
s
s
i
v
e
c
o
m
p
u
t
a
ti
o
n
a
l
c
o
s
t
s
.
I
n
o
u
r
i
m
p
l
e
m
e
n
t
a
ti
o
n
,
t
h
e
m
ax
i
m
u
m
n
u
m
b
e
r
o
f
i
t
e
r
a
ti
o
n
s
i
s
s
e
t
t
o
3
0
0
.
T
y
p
i
c
a
ll
y
,
v
a
l
u
es
b
et
w
e
e
n
1
0
0
a
n
d
5
0
0
i
t
e
r
a
ti
o
n
s
a
r
e
r
e
c
o
m
m
e
n
d
e
d
i
n
o
p
t
i
m
i
z
a
t
i
o
n
t
as
k
s
,
i
n
c
l
u
d
i
n
g
m
e
d
i
c
al
i
m
a
g
e
s
e
g
m
e
n
t
a
t
i
o
n
.
T
h
e
n
u
m
b
e
r
o
f
i
t
e
r
a
t
i
o
n
s
p
l
a
y
s
a
c
r
u
c
i
a
l
r
o
le
i
n
b
a
l
a
n
c
i
n
g
e
x
p
l
o
r
a
t
i
o
n
a
n
d
c
o
m
p
u
t
a
t
i
o
n
a
l
e
f
f
i
c
i
e
n
c
y
,
t
h
e
h
i
g
h
e
r
t
h
e
n
u
m
b
e
r
,
t
h
e
m
o
r
e
t
h
o
r
o
u
g
h
l
y
t
h
e
a
l
g
o
r
i
t
h
m
c
a
n
e
x
p
l
o
r
e
t
h
e
s
e
a
r
c
h
s
p
a
c
e
a
n
d
r
e
f
i
n
e
p
o
t
e
n
t
i
a
l
s
o
l
u
ti
o
n
s
.
H
o
w
e
v
e
r
,
t
h
is
a
l
s
o
r
e
s
u
lt
s
i
n
i
n
c
r
ea
s
e
d
c
o
m
p
u
t
a
t
i
o
n
t
i
m
e
.
I
n
t
h
e
c
o
n
t
e
x
t
o
f
b
r
a
i
n
M
R
I
s
e
g
m
e
n
t
at
i
o
n
,
w
h
e
r
e
c
o
n
v
e
r
g
e
n
c
e
i
s
o
f
t
e
n
a
c
h
i
e
v
e
d
w
i
t
h
i
n
t
h
is
r
a
n
g
e
,
3
0
0
i
t
e
r
a
t
i
o
n
s
p
r
o
v
i
d
e
a
r
e
a
s
o
n
a
b
l
e
t
r
a
d
e
-
o
f
f
b
e
t
w
e
e
n
a
c
c
u
r
a
c
y
a
n
d
p
e
r
f
o
r
m
a
n
c
e
,
a
l
l
o
w
i
n
g
t
h
e
al
g
o
r
i
t
h
m
t
o
c
o
n
v
e
r
g
e
e
f
f
e
c
t
i
v
el
y
w
i
t
h
o
u
t
u
n
n
e
c
e
s
s
a
r
y
r
es
o
u
r
c
e
co
n
s
u
m
p
t
i
o
n
.
T
o
a
v
o
i
d
s
t
a
g
n
a
t
i
o
n
i
n
a
l
o
c
a
l
m
i
n
i
m
u
m
,
w
e
s
e
t
m
a
x
i
m
u
m
n
u
m
b
e
r
o
f
c
y
c
l
e
s
(
)
t
o
1
0
,
w
h
i
c
h
l
i
m
i
ts
t
h
e
n
u
m
b
e
r
o
f
c
o
n
s
e
c
u
t
iv
e
c
y
c
l
e
s
w
i
t
h
o
u
t
i
m
p
r
o
v
e
m
e
n
t
a
n
d
h
e
l
p
s
m
a
i
n
t
a
i
n
a
b
a
l
a
n
c
e
b
e
t
w
e
e
n
e
x
p
l
o
r
a
t
i
o
n
a
n
d
e
x
p
l
o
i
t
a
ti
o
n
d
u
r
i
n
g
t
h
e
o
p
ti
m
i
z
a
ti
o
n
p
r
o
c
e
s
s
.
I
n
t
h
e
o
b
je
ct
i
v
e
f
u
n
c
ti
o
n
,
t
h
e
w
e
i
g
h
ts
1
a
n
d
2
a
r
e
b
o
t
h
s
e
t
t
o
0
.
5
,
e
n
s
u
r
i
n
g
a
b
a
l
a
n
c
e
d
c
o
n
t
r
i
b
u
t
i
o
n
o
f
t
h
e
i
n
d
i
v
i
d
u
a
l
c
o
m
p
o
n
e
n
t
s
i
n
t
h
e
o
p
t
i
m
i
z
a
t
i
o
n
p
r
o
c
e
s
s
.
4
.
1
.
M
et
rics us
ed
f
o
r
s
eg
m
e
nta
t
io
n e
v
a
lua
t
io
n
T
h
e
ev
alu
atio
n
o
f
b
r
ain
M
R
I
s
eg
m
en
tatio
n
p
er
f
o
r
m
an
c
e
r
elies
o
n
s
ev
er
al
m
etr
ics
to
q
u
a
n
tify
ac
cu
r
ac
y
,
r
o
b
u
s
tn
ess
,
an
d
c
o
n
s
is
ten
cy
[
3
2
]
.
I
n
ca
s
es
wh
er
e
th
e
g
r
o
u
n
d
tr
u
th
is
av
ailab
le,
we
u
s
e
J
ac
ca
r
d
Similar
ity
Me
tr
ic
.
I
n
ca
s
es
wh
er
e
th
e
g
r
o
u
n
d
tr
u
th
is
u
n
a
v
ailab
le,
it
b
ec
o
m
es
n
ec
ess
ar
y
to
r
ely
o
n
in
ter
n
al
v
alid
atio
n
in
d
ices
to
ev
alu
ate
th
e
q
u
ality
o
f
th
e
clu
s
ter
in
g
r
esu
lts
.
B
y
u
tili
zin
g
th
ese
in
d
i
ce
s
in
co
m
b
in
atio
n
,
we
ca
n
o
b
tain
a
c
o
m
p
r
e
h
en
s
i
v
e
ev
alu
atio
n
o
f
th
e
cl
u
s
ter
in
g
o
u
tc
o
m
es,
en
s
u
r
in
g
th
at
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
ac
h
iev
es
o
p
tim
al
p
er
f
o
r
m
an
c
e
ev
en
in
th
e
ab
s
en
ce
o
f
g
r
o
u
n
d
tr
u
th
in
f
o
r
m
atio
n
.
T
h
is
ap
p
r
o
ac
h
n
o
t
o
n
ly
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:
2088
-
8
7
0
8
F
u
z
z
y
clu
s
ter
in
g
o
p
timiz
a
tio
n
b
a
s
ed
a
r
tifi
cia
l b
ee
co
lo
n
y
a
l
g
o
r
ith
m
fo
r
…
(
C
h
a
kir Mo
kh
ta
r
i
)
4925
en
h
an
ce
s
th
e
r
eliab
ilit
y
o
f
th
e
s
eg
m
en
tatio
n
p
r
o
ce
s
s
b
u
t
also
en
ab
les
m
ea
n
in
g
f
u
l
co
m
p
ar
is
o
n
s
with
o
th
er
clu
s
ter
in
g
tech
n
iq
u
es u
n
d
e
r
s
im
ilar
co
n
d
itio
n
s
.
4
.
1
.
1
.
J
a
cc
a
rd
s
im
ila
rit
y
m
et
ric
T
h
e
J
ac
ca
r
d
s
im
ilar
ity
(
o
r
J
ac
ca
r
d
I
n
d
e
x
)
m
ea
s
u
r
es
th
e
o
v
e
r
lap
b
etwe
en
two
s
ets,
in
th
is
ca
s
e,
th
e
s
eg
m
en
ted
r
eg
i
o
n
an
d
th
e
g
r
o
u
n
d
tr
u
th
.
I
t is d
e
f
in
ed
as
(
1
1
)
:
=
|
∩
|
|
∪
|
(
1
1
)
wh
er
e
an
d
ar
e
th
e
to
tal
n
u
m
b
er
o
f
p
ix
els
lab
eled
in
to
th
e
cl
u
s
ter
k
id
en
tifie
d
b
y
th
e
clu
s
te
r
in
g
alg
o
r
ith
m
an
d
th
e
g
r
o
u
n
d
tr
u
th
r
esp
ec
tiv
ely
.
T
h
e
clu
s
ter
k
is
well
d
etec
ted
wh
en
th
e
v
alu
e
o
f
is
n
ea
r
1
.
4
.
1
.
2
.
P
a
rt
it
i
o
n
co
ef
f
icient
in
dex
T
h
e
p
a
r
t
i
t
i
o
n
c
o
e
f
f
i
c
i
e
n
t
i
n
d
e
x
(
P
C
I
)
m
e
a
s
u
r
es
t
h
e
f
u
z
z
i
n
e
s
s
o
f
t
h
e
c
l
u
s
t
e
r
i
n
g
r
e
s
u
l
t
.
I
t
i
s
d
e
f
i
n
e
d
a
s
(
1
2
)
:
=
1
∑
∑
2
=
1
=
1
(
1
2
)
4
.
1
.
3
.
P
a
rt
it
i
o
n
ent
ro
py
ind
ex
T
h
e
p
ar
titi
o
n
en
tr
o
p
y
in
d
ex
(
P
E
I
)
m
ea
s
u
r
es
th
e
u
n
ce
r
tain
t
y
o
r
r
an
d
o
m
n
ess
in
th
e
clu
s
ter
in
g
r
esu
lt.
I
t
is
d
ef
in
ed
as
(
1
3
)
:
=
1
∑
∑
=
1
=
1
(
)
(
1
3
)
4
.
1
.
4
.
Da
v
ies
-
B
o
uld
in
i
n
dex
T
h
e
Dav
ies
-
B
o
u
ld
in
i
n
d
ex
(
D
B
I
)
m
ea
s
u
r
es
th
e
c
o
m
p
ac
t
n
ess
an
d
s
ep
ar
atio
n
o
f
clu
s
ter
s
.
I
t
is
d
ef
in
ed
as
(
14)
:
=
1
∑
≠
(
+
,
)
=
1
(
1
4
)
wh
er
e
S
i
is
th
e
m
ea
n
d
is
tan
ce
b
etwe
en
th
e
ce
n
ter
o
f
t
h
e
clu
s
ter
I
an
d
all
th
e
p
o
in
ts
b
elo
n
g
in
g
to
t
h
is
clu
s
ter
an
d
,
d
en
o
tes th
e
d
is
tan
ce
b
etw
ee
n
th
e
ce
n
tr
o
id
s
o
f
th
e
clu
s
ter
s
I
an
d
J
.
4
.
2
.
E
x
perim
ent
a
l
re
s
ults
o
n si
m
ula
t
ed
bra
in
M
R
i
m
a
g
es
T
h
e
f
o
llo
win
g
ex
p
e
r
im
en
ts
w
er
e
co
n
d
u
cted
u
s
in
g
s
im
u
lated
b
r
ain
d
atab
ase
(
SB
D)
[
3
3
]
.
T
h
e
SB
D
p
r
o
v
id
es
s
y
n
th
etic
MRI
b
r
ain
im
ag
es
with
k
n
o
wn
g
r
o
u
n
d
tr
u
th
s
eg
m
en
tatio
n
s
,
m
ak
in
g
it
id
ea
l
f
o
r
v
alid
atin
g
s
eg
m
en
tatio
n
alg
o
r
ith
m
s
.
T
h
e
im
ag
es
s
im
u
late
d
if
f
er
en
t
i
n
ten
s
ity
in
h
o
m
o
g
e
n
eities,
an
d
s
lice
th
ick
n
ess
e
s
,
m
im
ick
in
g
r
ea
l
-
wo
r
l
d
MRI
c
h
allen
g
es.
T
h
is
d
atab
ase
in
cl
u
d
es
g
r
o
u
n
d
tr
u
th
in
f
o
r
m
atio
n
f
o
r
tis
s
u
e
o
f
W
M
,
GM
,
an
d
C
SF
.
I
t
o
f
f
er
s
a
c
o
n
tr
o
lled
s
ettin
g
to
ass
ess
th
e
alg
o
r
ith
m
’
s
ac
c
u
r
ac
y
a
n
d
i
ts
ab
ilit
y
to
h
an
d
l
e
in
ten
s
ity
in
h
o
m
o
g
en
eity
e
f
f
ec
t
iv
ely
.
T
h
e
p
r
o
p
o
s
ed
FC
M
-
AB
C
o
p
tim
izer
m
eth
o
d
was
in
itially
test
ed
o
n
a
T
1
-
weig
h
ted
b
r
ain
MR
im
ag
e
with
d
im
en
s
io
n
s
o
f
217
×
181
p
ix
els,
w
h
ich
in
clu
d
es
2
0
%
g
r
ay
s
ca
le
n
o
n
-
u
n
if
o
r
m
ity
to
s
im
u
late
r
e
al
-
wo
r
ld
im
ag
in
g
ch
allen
g
es.
T
h
e
p
r
im
ar
y
o
b
jectiv
e
o
f
th
is
ap
p
licati
o
n
was
to
ac
cu
r
ately
s
eg
m
en
t
an
d
id
en
tif
y
cr
itical
b
r
ain
r
eg
io
n
s
,
n
am
ely
W
M,
GM
,
an
d
C
SF
.
T
h
ese
tis
s
u
e
ty
p
es
ar
e
f
u
n
d
am
en
tal
f
o
r
r
ad
io
lo
g
is
ts
in
th
eir
an
aly
s
is
an
d
d
iag
n
o
s
is
o
f
v
ar
i
o
u
s
n
eu
r
o
lo
g
ical
d
is
o
r
d
er
s
an
d
d
is
ea
s
es.
F
i
g
u
r
e
3
p
r
o
v
id
e
s
a
v
i
s
u
a
l
r
e
p
r
e
s
e
n
t
a
t
io
n
o
f
th
e
s
e
g
m
en
t
a
t
i
o
n
r
e
s
u
l
t
s
,
a
l
lo
w
i
n
g
f
o
r
a
d
i
r
e
c
t
c
o
m
p
a
r
i
s
o
n
o
f
th
e
p
e
r
f
o
r
m
a
n
c
e
o
f
f
o
u
r
d
if
f
er
e
n
t
a
l
g
o
r
i
th
m
s
:
F
C
M
,
G
A
-
F
C
M
,
F
C
M
A
-
E
S
,
a
n
d
t
h
e
p
r
o
p
o
s
e
d
F
C
M
-
A
B
C
o
p
t
i
m
i
z
er
m
e
th
o
d
.
T
o
p
r
o
v
id
e
c
o
n
t
ex
t
,
t
h
e
o
r
i
g
i
n
a
l
b
r
a
in
i
m
ag
e
i
s
s
h
o
w
n
i
n
F
i
g
u
r
e
3
(
a)
,
w
h
i
l
e
i
t
s
c
o
r
r
e
s
p
o
n
d
in
g
g
r
o
u
n
d
t
r
u
t
h
s
f
o
r
W
M
,
G
M
,
a
n
d
C
S
F
a
r
e
d
i
s
p
l
a
y
ed
i
n
F
ig
u
r
e
3
(
b
)
.
T
h
e
s
eg
m
en
t
e
d
i
m
a
g
e
s
p
r
o
d
u
c
ed
b
y
th
e
F
C
M
,
G
A
-
F
C
M
,
F
C
M
A
-
E
S
,
an
d
F
C
M
-
A
B
C
o
p
t
im
i
z
e
r
m
e
th
o
d
s
ar
e
p
r
e
s
en
t
ed
i
n
F
i
g
u
r
e
s
3
(
c
)
,
3
(
d
)
,
3
(
e)
,
an
d
3
(
f
)
,
r
e
s
p
e
c
t
iv
e
l
y
.
Fro
m
Fig
u
r
e
3
,
it
is
clea
r
th
at
th
e
p
r
o
p
o
s
ed
FC
M
-
AB
C
o
p
tim
izer
m
eth
o
d
o
u
tp
er
f
o
r
m
s
th
e
o
th
e
r
m
eth
o
d
s
in
ter
m
s
o
f
ac
c
u
r
ate
ly
ex
tr
ac
tin
g
b
r
ain
tis
s
u
es.
A
clo
s
er
ex
am
in
atio
n
r
e
v
ea
ls
th
at
th
e
FC
M
-
AB
C
o
p
tim
izer
m
et
h
o
d
ef
f
ec
tiv
el
y
m
ain
tain
s
r
e
g
io
n
al
h
o
m
o
g
e
n
eity
,
e
n
s
u
r
in
g
th
at
t
h
e
s
eg
m
en
ted
r
eg
io
n
s
ar
e
co
n
s
is
ten
t
an
d
u
n
if
o
r
m
.
Ad
d
i
tio
n
ally
,
th
e
alg
o
r
ith
m
p
r
eser
v
es
m
o
r
e
d
etailed
in
f
o
r
m
atio
n
f
r
o
m
th
e
o
r
ig
in
al
MR
im
ag
e,
wh
ich
is
cr
u
cial
f
o
r
m
ain
tain
in
g
th
e
i
n
teg
r
ity
o
f
th
e
an
ato
m
ical
s
tr
u
ctu
r
es
b
e
in
g
an
aly
ze
d
.
T
h
is
ab
ilit
y
to
r
etain
f
i
n
e
d
etails
is
p
ar
ticu
lar
ly
a
d
v
an
ta
g
eo
u
s
in
m
ed
ical
im
a
g
in
g
ap
p
licatio
n
s
,
wh
er
e
s
u
b
tle
Evaluation Warning : The document was created with Spire.PDF for Python.