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.
7
,
No
.
6
,
Dec
em
b
er
201
7
,
p
p
.
3
6
4
3
~
3
6
5
4
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v7
i
6
.
pp
3
6
4
3
-
3654
3643
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
jo
u
r
n
a
l.c
o
m/o
n
lin
e/in
d
ex
.
p
h
p
/I
JE
C
E
Neuro
endo
sco
py
Ada
pter
M
o
dule
Dev
elo
p
m
e
nt
for
Better
Bra
in
Tu
m
o
r I
m
a
g
e V
isua
liz
a
tion
Su
nil
L
.
B
a
ng
a
re
1
,
G
.
P
ra
deepini
2
,
Sh
r
is
ha
ila
p
pa
T
.
P
a
t
il
3
1
,2
De
p
a
rtm
e
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
&
En
g
in
e
e
rin
g
,
K.L
.
Un
iv
e
rsity
,
G
u
n
tu
r,
A
.
P
.
In
d
ia
3
De
p
a
rt
m
e
n
t
o
f
Co
m
p
u
ter E
n
g
in
e
e
rin
g
,
V
.
I
.
T
.
,
P
u
n
e
,
I
n
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
an
13
,
2
0
1
7
R
ev
i
s
ed
A
u
g
1
1
,
2
0
1
7
A
cc
ep
ted
A
u
g
30
,
2
0
1
7
T
h
e
issu
e
o
f
b
ra
in
m
a
g
n
e
ti
c
re
so
n
a
n
c
e
im
a
g
e
e
x
p
lo
ra
ti
o
n
t
o
g
e
th
e
r
w
it
h
c
las
si
f
ica
ti
o
n
re
c
e
iv
e
s
a
sig
n
if
ic
a
n
t
a
w
a
re
n
e
ss
in
re
c
e
n
t
y
e
a
rs.
In
d
e
e
d
,
v
a
rio
u
s
c
o
m
p
u
ter
-
a
id
e
d
-
d
iag
n
o
sis
so
l
u
ti
o
n
s
we
re
su
g
g
e
st
e
d
to
su
p
p
o
rt
ra
d
io
lo
g
ist
in
d
e
c
isio
n
-
m
a
k
in
g
.
In
th
is
c
ircu
m
sta
n
c
e
,
a
d
e
q
u
a
te
i
m
a
g
e
c
las
sif
ica
ti
o
n
is
e
x
tre
m
e
l
y
re
q
u
ired
a
s
it
is
th
e
m
o
st
c
o
m
m
o
n
c
rit
ica
l
b
ra
in
tu
m
o
rs
w
h
ich
o
f
ten
d
e
v
e
lo
p
f
ro
m
su
b
d
u
ra
l
h
e
m
a
to
m
a
c
e
ll
s,
w
h
ich
m
ig
h
t
b
e
c
o
m
m
o
n
t
y
p
e
in
a
d
u
lt
s.
In
h
e
a
lt
h
c
a
re
m
il
ieu
,
b
r
a
in
M
RIs
a
re
i
n
ten
d
e
d
f
o
r
id
e
n
ti
f
ica
ti
o
n
o
f
tu
m
o
r.
In
th
is
re
g
a
rd
,
v
a
rio
u
s
c
o
m
p
u
teriz
e
d
d
iag
n
o
sis
s
y
st
e
m
s
w
e
r
e
su
g
g
e
ste
d
to
h
e
lp
m
e
d
ica
l
p
ro
f
e
s
sio
n
a
ls
in
c
li
n
ica
l
d
e
c
isio
n
-
m
a
k
i
n
g
.
As
p
e
r
re
c
e
n
t
p
ro
b
lem
s,
Ne
u
ro
e
n
d
o
sc
o
p
y
is
th
e
g
o
ld
sta
n
d
a
rd
in
t
e
n
d
e
d
f
o
r
d
isc
o
v
e
rin
g
b
ra
in
tu
m
o
rs;
n
e
v
e
rth
e
les
s,
t
y
p
ica
l
Ne
u
ro
e
n
d
o
s
c
o
p
y
c
a
n
c
e
rtain
ly
o
v
e
rlo
o
k
rip
p
e
d
g
ro
w
th
s.
Ne
u
ro
e
n
d
o
sc
o
p
y
is
a
m
in
i
m
a
ll
y
-
in
v
a
siv
e
su
rg
ica
l
p
ro
c
e
d
u
re
i
n
w
h
ich
t
h
e
n
e
u
ro
s
u
rg
e
o
n
re
m
o
v
e
s
th
e
tu
m
o
r
th
r
o
u
g
h
s
m
a
ll
h
o
les
in
th
e
sk
u
ll
o
r
th
r
o
u
g
h
th
e
m
o
u
th
o
r
n
o
se
.
Ne
u
ro
e
n
d
o
sc
o
p
y
e
n
a
b
les
n
e
u
ro
s
u
rg
e
o
n
s
to
a
c
c
e
ss
a
re
a
s
o
f
th
e
b
ra
in
th
a
t
c
a
n
n
o
t
b
e
re
a
c
h
e
d
w
it
h
trad
it
io
n
a
l
su
rg
e
ry
to
re
m
o
v
e
th
e
tu
m
o
r
w
it
h
o
u
t
c
u
tt
i
n
g
o
r
h
a
r
m
in
g
o
th
e
r
p
a
rts
o
f
th
e
sk
u
ll
.
W
e
f
o
c
u
se
d
o
n
f
in
d
in
g
o
u
t
w
h
e
th
e
r
o
r
n
o
t
v
isu
a
l
im
a
g
e
s
o
f
tu
m
o
r
rip
p
e
d
les
io
n
s
e
n
d
e
d
u
p
b
e
in
g
m
u
c
h
b
e
tt
e
r
b
y
a
u
to
f
lu
o
re
sc
e
n
c
e
im
a
g
e
r
e
so
lu
ti
o
n
a
s
w
e
ll
a
s
n
a
rro
w
-
b
a
n
d
im
a
g
e
re
so
lu
ti
o
n
g
ra
p
h
ic
e
v
a
lu
a
ti
o
n
jo
i
n
tl
y
w
it
h
th
e
late
st
n
e
u
ro
e
n
d
o
sc
o
p
y
te
c
h
n
iq
u
e
.
A
lso
,
w
it
h
in
t
h
e
las
t
se
v
e
ra
l
y
e
a
r
s,
p
a
th
o
lo
g
y
lab
s
b
e
g
a
n
to
p
ro
c
e
e
d
in
th
e
d
irec
ti
o
n
o
f
a
n
e
n
ti
re
l
y
d
ig
it
a
l
w
o
rk
f
lo
w
,
u
sin
g
th
e
e
lec
tro
n
ic
slid
e
s
c
u
rre
n
tl
y
b
e
in
g
th
e
k
e
y
e
l
e
m
e
n
t
o
f
th
is
tec
h
n
iq
u
e
.
Be
sid
e
s
lo
ts
o
f
b
e
n
e
fit
s
re
g
a
rd
in
g
sto
ra
g
e
a
s
we
ll
a
s
e
x
p
lo
rin
g
c
a
p
a
b
il
it
ies
w
it
h
th
e
i
m
a
g
e
in
f
o
rm
a
ti
o
n
,
a
m
o
n
g
th
e
b
e
n
e
f
it
s
o
f
e
lec
tro
n
ic
slid
e
s
is
th
a
t
th
e
y
c
a
n
h
e
lp
th
e
a
p
p
li
c
a
ti
o
n
o
f
ima
g
e
a
n
a
l
y
sis
a
p
p
ro
a
c
h
e
s
w
h
ich
se
e
k
to
d
e
v
e
lo
p
q
u
a
n
ti
tat
i
v
e
a
tt
rib
u
tes
to
a
ss
ist
p
a
th
o
lo
g
is
ts
in
t
h
e
ir
w
o
rk
.
Ho
w
e
v
e
r,
s
y
ste
m
s
a
lso
h
a
v
e
so
m
e
d
iff
i
c
u
lt
ies
in
e
x
e
c
u
ti
o
n
a
n
d
h
a
n
d
l
in
g
.
He
n
c
e
,
su
c
h
c
o
n
v
e
n
ti
o
n
a
l
m
e
th
o
d
n
e
e
d
s au
to
m
a
ti
o
n
.
W
e
d
e
v
e
lo
p
e
d
a
n
d
e
m
p
lo
y
e
d
to
lo
o
k
f
o
r
th
e
tar
g
e
ted
im
p
o
rtan
c
e
a
lo
n
g
w
it
h
u
n
c
o
v
e
rin
g
th
e
b
e
st
-
f
o
c
u
se
d
g
ra
p
h
ic
p
o
siti
o
n
b
y
w
a
y
o
f
a
li
a
sin
g
se
a
r
c
h
m
e
th
o
d
in
c
o
rp
o
ra
ted
w
it
h
n
e
w
Ne
u
ro
e
n
d
o
sc
o
p
y
A
d
a
p
ter M
o
d
u
le (NA
M
)
tec
h
n
iq
u
e
.
K
ey
w
o
r
d
:
Neu
r
o
en
d
o
s
co
p
y
a
d
ap
ter
m
o
d
u
le
(
N
A
M)
Su
p
p
o
r
t
v
eto
r
m
ac
h
i
n
e
(
SVM)
P
ar
ticle
s
w
ar
m
o
p
ti
m
izati
o
n
(
P
SO)
Su
b
d
u
r
al
h
ae
m
ato
m
a
Me
d
ical
i
m
a
g
in
g
C
las
s
i
f
icatio
n
Vis
u
a
lizatio
n
Co
p
y
rig
h
t
©
2
0
1
7
I
n
stit
u
te o
f
Ad
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
:
Su
n
il
L
.
B
an
g
ar
e,
R
esear
ch
Sch
o
lar
,
Dep
ar
t
m
e
n
t
o
f
C
o
m
p
u
ter
Scien
ce
&
E
n
g
i
n
ee
r
in
g
,
K.
L
.
Un
i
v
er
s
i
t
y
,
G
u
n
tu
r
,
An
d
h
r
a
P
r
ad
esh
,
5
2
2
5
0
2
,
I
n
d
ia
E
m
ail:
s
u
n
il.b
an
g
ar
e@
g
m
ail.
c
o
m
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
6
,
Dec
em
b
er
2
0
1
7
:
3
6
4
3
–
3
6
5
4
3644
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
aim
o
f
th
i
s
an
al
y
ze
i
s
to
p
r
o
v
id
e
an
d
co
m
p
ar
e
th
r
ee
p
er
f
ec
tl
y
co
m
p
u
ter
ized
s
y
s
te
m
s
f
o
r
r
ap
id
class
i
f
icatio
n
o
f
b
r
ain
MR
i
m
ag
es
i
n
to
h
ea
lt
h
y
,
a
n
d
b
r
ain
w
it
h
t
h
e
t
u
m
o
r
i
n
f
ec
tio
n
.
I
t
f
u
n
ct
io
n
s
in
o
f
f
o
u
r
ac
tio
n
s
:
(
i)
Seg
m
en
tatio
n
o
f
th
e
i
m
a
g
e
to
g
et
i
m
ag
e
g
eo
m
etr
ic
b
o
u
n
d
ar
ies o
f
tu
m
o
r
ce
lls
;
(
ii)
C
o
m
p
u
tatio
n
o
f
t
h
e
A
F
D
s
i
g
n
atu
r
e
o
f
th
e
i
m
a
g
e;
(
iii)
E
x
tr
ac
tio
n
o
f
t
h
e
f
ea
tu
r
e
s
o
f
t
h
e
p
r
o
v
id
ed
A
FD; a
n
d
(
iv
)
C
las
s
i
f
icatio
n
b
y
Ne
u
r
o
en
d
o
s
co
p
y
A
d
ap
ter
Mo
d
u
le
(
N
A
M)
alo
n
g
w
it
h
p
er
f
o
r
m
a
n
ce
co
m
p
ar
is
o
n
ag
ai
n
s
t
s
u
p
p
o
r
t v
ec
to
r
m
ac
h
in
e
(
SVM
)
.
T
h
e
P
SO
alg
o
r
ith
m
[
1
]
is
e
m
p
lo
y
ed
f
o
r
p
r
e
-
p
r
o
ce
s
s
in
g
th
e
MR
i
m
ag
e
to
s
p
lit
u
p
th
e
v
it
al
ar
ea
s
o
f
th
e
b
r
ain
a
n
d
ad
d
itio
n
all
y
b
o
o
s
t
th
e
d
etec
tio
n
o
f
th
o
s
e
w
h
ich
i
n
d
icate
s
h
i
g
h
-
i
n
te
n
s
i
t
y
c
o
n
tr
ast
as
r
eg
ar
d
in
g
tu
m
o
r
.
T
h
e
p
u
r
p
o
s
e
is
t
o
id
en
ti
f
y
tu
m
o
r
b
y
d
is
co
v
er
in
g
t
h
e
p
ix
el
lev
e
l
d
is
s
i
m
i
lar
ities
ar
o
u
n
d
n
o
r
m
a
l
an
d
ab
n
o
r
m
al
tis
s
u
e
a
s
w
ell
as
t
h
e
p
r
esen
ce
o
f
ac
tiv
e
b
o
u
n
d
ar
y
ed
g
es.
W
ith
t
h
is
cir
c
u
m
s
ta
n
ce
,
th
e
MRI
p
ix
el
s
e
t
is
an
is
s
u
e,
an
d
to
id
en
ti
f
y
a
p
i
x
el
p
ar
titi
o
n
i
n
g
t
h
at
en
h
a
n
ce
s
th
e
b
et
w
ee
n
-
clas
s
d
ev
iatio
n
o
f
th
e
d
is
tr
ib
u
tio
n
o
f
in
te
n
s
it
y
d
e
g
r
ee
s
.
As p
er
tr
ad
it
io
n
al
s
o
lu
tio
n
,
P
SO
h
i
n
g
e
s
o
n
an
etio
lo
g
ical
m
etap
h
o
r
to
r
es
o
lv
e
t
h
ese
k
i
n
d
s
o
f
is
s
u
es
w
i
th
s
to
ch
a
s
tic
s
ea
r
ch
g
lo
b
al
o
p
ti
m
izatio
n
.
I
t
i
s
o
n
e
o
f
t
h
e
s
o
-
ca
lled
ev
o
l
u
tio
n
ar
y
co
m
p
u
tatio
n
p
ar
ad
ig
m
s
i
n
co
m
p
u
tat
io
n
al
in
tell
ig
e
n
ce
.
P
r
i
m
ar
il
y
P
SO
w
a
s
ac
c
u
s
to
m
ed
to
ac
h
ie
v
i
n
g
th
e
b
r
ai
n
MRI
s
eg
m
e
n
tatio
n
tas
k
d
u
e
to
t
h
e
p
er
f
o
r
m
a
n
ce
to
p
r
o
d
u
ce
ap
p
r
o
x
i
m
ate
w
a
y
s
to
o
co
m
p
licate
d
o
p
tim
iza
tio
n
i
s
s
u
e
w
it
h
th
e
co
m
p
eti
tiv
e
co
m
p
u
ta
tio
n
al
co
s
t
[
2
]
.
Fu
r
t
h
er
m
o
r
e,
P
SO
is
m
o
r
e
r
ap
id
l
y
an
d
co
n
s
is
te
n
t
th
a
n
g
e
n
etic
alg
o
r
ith
m
s
an
d
a
n
t
co
lo
n
y
o
p
t
i
m
izatio
n
[
3
]
b
u
t,
i
n
r
ea
l
li
f
e
a
p
p
licatio
n
,
s
i
n
ce
t
h
e
co
m
p
le
x
i
t
y
a
n
d
n
o
n
li
n
ea
r
it
y
o
f
m
a
n
y
p
r
o
b
le
m
s
,
t
h
e
tar
g
e
t
f
u
n
ctio
n
s
o
f
t
h
ese
p
r
o
b
le
m
s
ar
e
o
f
ten
d
is
cr
ete
an
d
o
f
m
u
lti
-
p
o
in
t
v
a
lu
e,
f
u
r
t
h
er
m
o
r
e,
t
h
e
m
o
d
eli
n
g
t
h
e
p
r
o
b
lem
its
e
lf
i
s
also
v
er
y
d
i
f
f
ic
u
lt
[
4
]
.
I
n
th
i
s
w
o
r
k
,
w
e
co
m
p
ar
ed
P
SO,
SVM,
an
d
n
e
w
l
y
d
ev
elo
p
ed
N
A
M
p
er
f
o
r
m
an
ce
s
.
N
A
M
e
n
ab
les
d
i
s
co
v
er
in
g
t
h
e
ex
is
te
n
ce
o
f
h
i
g
h
-
in
te
n
s
it
y
b
u
r
s
t
s
in
th
e
ar
r
a
y
o
f
t
h
e
i
m
ag
e
a
n
d
co
n
s
eq
u
e
n
tl
y
u
s
i
n
g
th
e
m
as
i
m
a
g
e
s
i
g
n
a
tu
r
es
to
d
is
tin
g
u
is
h
t
h
e
co
m
m
o
n
f
o
r
m
o
f
tu
m
o
r
a
f
f
ec
ted
MRIs.
Ma
n
y
o
f
th
e
p
r
ev
io
u
s
w
o
r
k
o
u
ts
r
ep
o
r
t
h
ig
h
clas
s
i
f
icatio
n
ac
cu
r
ac
ies,
b
u
t
ea
c
h
k
i
n
d
h
a
s
p
o
s
s
ib
le
n
eg
at
iv
e
a
s
p
ec
ts
.
T
h
e
d
e
m
a
n
d
f
o
r
m
a
n
u
a
ll
y
-
d
eter
m
in
ed
o
r
est
ab
lis
h
ed
r
eg
io
n
s
o
f
i
n
ter
es
t
s
in
th
e
M
R
I
[
5
]
is
w
id
el
y
s
tu
d
ied
.
T
h
e
m
o
s
t
p
r
o
b
le
m
atic
s
tu
d
ie
s
ar
e
o
v
er
co
m
in
g
th
e
lack
o
f
b
o
u
n
d
ar
ies,
p
o
o
r
co
n
tr
ast,
n
o
is
e,
s
u
p
er
v
i
s
ed
m
e
th
o
d
,
les
s
r
o
b
u
s
t,
less
e
f
f
icie
n
t
a
n
d
les
s
r
eliab
l
e
to
m
a
k
e
a
f
u
ll
u
s
e
o
f
o
b
tain
i
n
g
d
ata.
T
h
u
s
,
it
is
ex
p
ec
ted
th
at
i
m
a
g
e
p
r
o
ce
s
s
i
n
g
m
eth
o
d
s
w
i
ll
h
e
lp
to
s
o
lv
e
t
h
o
s
e
p
r
o
b
le
m
s
[
6
]
.
A
l
s
o
,
th
e
c
h
an
ce
o
f
co
r
r
u
p
tin
g
th
e
f
ea
t
u
r
e
s
et
u
s
i
n
g
n
o
i
s
e
alo
n
g
w
it
h
ar
ti
f
ac
t
s
d
u
e
to
ap
p
l
y
i
n
g
ex
ten
s
iv
e
ec
h
o
MRS
[
7
]
an
d
also
co
n
d
u
cti
n
g
ca
teg
o
r
izatio
n
i
n
s
tead
o
f
s
u
p
er
v
is
ed
clas
s
i
f
icatio
n
[
8
]
,
ca
teg
o
r
izatio
n
b
ein
g
les
s
s
u
itab
le
f
o
r
clin
ical
d
ia
g
n
o
s
i
s
.
E
v
en
t
u
all
y
,
e
v
en
th
o
u
g
h
s
o
lu
tio
n
i
n
[
9
]
b
r
o
u
g
h
t
ab
o
u
t
a
n
o
v
er
all
g
a
in
in
co
m
p
u
tatio
n
ti
m
e
w
it
h
o
u
t
a
n
y
d
ec
lin
e
in
ac
cu
r
ac
y
,
it
is
n
o
t
w
o
r
th
y
o
f
m
a
s
s
i
v
e
m
u
lt
i
-
d
atas
ets
[
1
0
]
.
Fu
r
th
er
m
o
r
e,
th
e
o
f
f
e
r
ed
m
o
d
el
in
[
1
1
]
is
d
ep
en
d
en
t
o
n
q
u
ite
a
f
e
w
co
n
s
tr
ain
t
s
to
b
e
tu
n
ed
an
d
a
m
u
l
titu
d
e
o
f
f
ea
t
u
r
es,
as
w
e
ll
a
s
b
ein
g
s
u
s
ce
p
tib
le
to
co
n
f
u
s
i
n
g
t
u
m
o
r
b
o
u
n
d
ar
ies.
L
ast
l
y
,
th
e
e
f
f
o
r
t
in
[
1
2
]
n
ec
es
s
itates
t
h
e
p
r
o
p
er
d
etec
tio
n
o
f
a
r
eg
i
o
n
o
f
i
n
ter
est
co
n
d
u
cted
m
a
n
u
a
ll
y
.
Hen
ce
,
as
o
f
th
e
f
ac
t
au
to
-
f
o
cu
s
alg
o
r
it
h
m
h
av
e
to
b
e
r
ea
l
-
ti
m
e,
th
e
co
n
v
en
tio
n
al
au
to
-
f
o
cu
s
alg
o
r
ith
m
co
u
ld
h
a
v
e
n
u
m
er
o
u
s
d
i
f
f
icu
lties
a
s
a
co
n
s
eq
u
e
n
ce
o
f
i
n
cr
ea
s
in
g
co
m
p
u
tat
io
n
s
as
t
h
e
v
o
lu
m
e
o
f
w
i
n
d
o
w
p
a
n
e
ar
ea
lo
ca
tio
n
ca
n
v
ar
y
co
n
s
id
er
ab
l
y
u
s
u
al
l
y
.
A
p
ar
tic
u
lar
p
r
o
b
le
m
s
tan
d
s
o
u
t
as
th
e
r
ep
ea
ted
in
ter
p
o
latio
n
o
f
au
to
-
f
o
cu
s
r
ec
eiv
es
r
etar
d
ed
as
a
co
n
s
eq
u
e
n
ce
o
f
in
cr
ea
s
in
g
ca
lc
u
latio
n
s
.
A
d
d
itio
n
al
l
y
,
t
h
e
lik
eli
h
o
o
d
lin
k
ed
to
d
e
-
f
o
cu
s
i
n
g
o
f
tar
g
eted
i
m
ag
e
co
u
ld
s
u
r
g
e
b
ec
au
s
e
o
f
t
h
e
m
is
m
atc
h
w
it
h
t
h
e
alg
o
r
it
h
m
p
er
f
o
r
m
a
n
ce
an
d
also
a
r
esu
lt
o
f
th
e
tar
g
et.
T
h
e
lo
ca
l
-
alia
s
i
n
g
-
s
a
m
p
l
in
g
m
eth
o
d
is
ev
id
en
tl
y
ap
p
lied
to
less
e
n
th
e
co
m
p
u
tatio
n
,
esp
ec
ial
l
y
in
m
ed
ical
-
m
icr
o
s
co
p
ic
-
d
ig
-
ca
m
.
Ho
w
e
v
er
,
i
n
s
id
e
t
h
e
L
o
ca
l
-
alias
in
g
tech
n
iq
u
e
,
m
an
y
ac
c
u
r
ate
s
p
ec
i
f
ics
m
a
y
lo
s
e
in
co
n
j
u
n
ctio
n
w
i
th
th
e
i
n
cr
ea
s
e
i
n
n
o
is
e
co
m
p
o
n
en
t
[
1
3
]
.
A
s
a
r
esu
lt,
t
h
e
len
s
d
ev
iate
s
th
r
o
u
g
h
o
u
t
t
h
e
ex
ac
t
p
o
s
itio
n
alo
n
g
w
ith
p
ic
tu
r
e
b
lu
r
m
a
y
p
er
h
ap
s
ta
k
e
p
lace
.
T
o
d
ef
ea
t
th
is
co
n
ce
r
n
,
t
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
“
N
A
M
in
te
r
p
o
latio
n
”
e
v
alu
ates
ev
er
y
s
i
n
g
le
m
is
p
lace
d
p
ix
el
th
r
o
u
g
h
a
n
ac
cu
m
u
lat
io
n
o
f
te
x
t
u
r
e
-
r
ele
v
an
t
s
e
m
i
-
lo
ca
l
p
ix
e
ls
w
h
e
n
u
s
in
g
t
h
e
r
e
g
u
lar
it
y
s
i
m
ilar
i
t
y
r
ec
u
r
r
en
tl
y
co
m
p
u
ted
th
r
o
u
g
h
s
o
m
e
s
ec
tio
n
s
o
f
s
e
v
er
al
p
r
o
p
o
r
tio
n
s
.
No
tab
ly
,
i
m
p
le
m
e
n
ti
n
g
N
A
M
-
iter
atio
n
,
cr
u
cial
clo
g
g
i
n
g
-
tex
t
u
r
e
p
ix
els
ar
e
in
v
o
l
v
ed
in
p
r
o
d
u
cin
g
s
y
m
p
to
m
s
as
w
e
ll
as
p
o
s
ter
io
r
an
al
y
s
i
s
,
an
d
a
n
o
ticea
b
l
y
b
ilater
al
d
etailed
tr
an
s
f
o
r
m
is
es
s
e
n
tia
l
f
o
r
r
eg
u
lar
izat
io
n
o
f
u
n
n
ec
ess
ar
y
clo
g
g
i
n
g
/p
las
m
a
la
y
e
r
r
em
o
v
al
f
r
o
m
th
e
i
m
a
g
e.
T
r
ial
an
d
er
r
o
r
ef
f
e
ct
s
co
n
s
id
er
ed
n
ex
t
to
ex
is
t
in
g
i
n
ter
p
o
latio
n
tech
n
iq
u
e
s
co
n
f
ir
m
o
u
r
ap
p
r
o
ac
h
ar
e
u
n
ab
le
j
u
s
t
co
n
s
id
er
ab
l
y
to
s
lo
w
u
p
th
e
a
lias
i
n
g
d
if
f
ic
u
lt
y
b
u
t
al
s
o
ac
q
u
ir
e
s
ig
n
i
f
ica
n
tl
y
b
etter
o
u
tco
m
es
th
r
o
u
g
h
o
u
t
a
r
an
g
e
o
f
v
ie
w
s
j
u
s
t
as
co
n
ce
r
n
in
g
q
u
an
titati
v
e
ev
al
u
at
io
n
to
g
eth
er
w
it
h
v
er
y
s
u
b
j
ec
tiv
e
i
m
a
g
e
q
u
alit
y
.
T
h
e
in
tr
o
d
u
ce
d
alg
o
r
i
th
m
m
a
y
b
e
ad
d
itio
n
all
y
u
t
ili
ze
d
as
a
p
o
s
s
ib
le
A
P
I
f
o
r
d
ig
ica
m
i
m
a
g
e
q
u
ali
t
y
i
m
p
r
o
v
e
m
en
t.
T
o
s
u
m
m
ar
ize,
e
m
p
lo
y
in
g
a
n
e
n
tire
l
y
n
e
w
n
eu
r
o
-
en
d
o
s
co
p
y
tech
n
iq
u
e
j
o
in
tl
y
w
i
th
au
to
f
o
cu
s
i
m
a
g
in
g
i
n
cr
ea
s
ed
f
ield
o
f
v
i
s
io
n
o
f
tu
m
o
r
r
ip
p
ed
g
r
o
w
t
h
s
,
co
r
r
esp
o
n
d
in
g
to
th
e
p
r
esen
ce
attain
ed
u
s
i
n
g
ch
r
o
m
o
n
e
u
r
o
-
e
n
d
o
s
co
p
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
N
A
M De
ve
lo
p
men
t fo
r
B
etter
B
r
a
in
Tu
mo
r
I
ma
g
e
V
is
u
a
liz
a
t
io
n
(
S
u
n
il L.
B
a
n
g
a
r
e
)
3645
Seh
g
al,
A
a
s
t
h
a,
et
al.
(
2
0
1
6
)
p
r
o
p
o
s
ed
m
eth
o
d
f
o
r
th
e
i
m
ag
e
d
iag
n
o
s
is
p
r
o
ce
s
s
f
o
r
b
r
a
in
ca
n
ce
r
,
clin
icia
n
s
o
f
te
n
n
ee
d
to
ac
h
ie
v
e
a
b
r
ai
n
b
io
p
s
y
to
g
et
h
i
s
to
lo
g
ical
d
ata.
Ho
w
e
v
er
,
t
h
er
e
ar
e
r
is
k
s
as
s
o
ciate
d
w
it
h
t
h
i
s
p
r
o
ce
d
u
r
e:
1
)
co
llected
s
a
m
p
les
ar
e
n
o
t
al
w
a
y
s
r
ep
r
esen
tati
v
e
o
f
th
e
tu
m
o
r
an
d
2
)
th
er
e
is
a
r
i
s
k
o
f
b
lo
o
d
v
ess
el
r
u
p
tu
r
e
w
h
e
n
th
e
s
a
m
p
le
is
ta
k
en
.
T
h
is
t
y
p
e
o
f
b
leed
in
g
o
cc
u
r
s
b
et
w
ee
n
0
.
3
an
d
5
9
.
8
%
o
f
th
e
ca
s
es,
an
d
t
h
e
m
o
r
tal
it
y
r
ate
c
an
b
e
as
h
ig
h
as
3
.
9
%.
Her
e
a
u
th
o
r
p
r
esen
ted
a
d
if
f
u
s
e
r
ef
le
ctan
ce
s
p
ec
tr
o
s
co
p
y
i
m
a
g
in
g
s
y
s
te
m
d
ir
ec
tl
y
i
n
te
g
r
ated
o
n
a
b
r
ain
b
io
p
s
y
n
e
ed
le
to
g
u
id
e
th
e
s
u
r
g
eo
n
d
u
r
in
g
n
ee
d
le
b
io
p
s
y
p
r
o
ce
d
u
r
es [
1
4
]
.
B
u
t th
is
tech
n
iq
u
e
lac
k
s
i
n
p
r
e
o
r
p
o
s
t
-
s
u
r
g
ical
an
al
y
s
i
s
.
A
l
s
o
,
as
p
er
Mo
h
y
eld
i
n
,
Ah
m
ed
,
et
al.
(
2
0
1
6
)
,
th
e
o
b
j
ec
t
o
f
th
eir
s
t
u
d
y
w
as
to
as
s
es
s
t
h
e
f
ea
s
ib
ili
t
y
,
ac
cu
r
ac
y
,
a
n
d
s
a
f
et
y
o
f
r
ea
l
-
ti
m
e
M
R
I
-
co
m
p
a
tib
le
f
r
a
m
ele
s
s
s
ter
eo
tactic
b
r
ai
n
b
io
p
s
y
.
T
h
e
ab
ilit
y
to
p
lace
a
b
io
p
s
y
ca
n
n
u
la
u
n
d
er
r
ea
l
-
ti
m
e
i
m
ag
i
n
g
g
u
id
an
ce
p
er
m
it
s
o
n
-
th
e
-
f
l
y
al
ter
atio
n
s
i
n
th
e
ca
n
n
u
la
tr
aj
ec
to
r
y
a
n
d
tip
p
lace
m
e
n
t.
R
ea
l
-
ti
m
e
i
m
a
g
in
g
d
u
r
i
n
g
M
R
I
-
g
u
id
ed
b
r
ai
n
b
io
p
s
y
p
r
o
v
id
es
p
r
ec
is
e,
s
af
e
tar
g
etin
g
o
f
b
r
ain
lesi
o
n
s
.
B
u
t,
ag
a
i
n
th
is
tec
h
n
iq
u
e
lack
s
in
i
m
a
g
e
en
h
a
n
ce
m
e
n
t a
n
d
u
s
ab
ili
t
y
[
1
5
]
.
Ma
g
n
e
tic
r
eso
n
a
n
ce
i
m
a
g
i
n
g
u
n
d
er
esti
m
ate
s
h
i
s
to
lo
g
ica
ll
y
id
en
ti
f
ied
tu
m
o
r
b
o
u
n
d
ar
ies,
p
ar
ticu
lar
l
y
f
o
r
w
o
u
n
d
s
t
h
at
h
a
v
e
a
h
i
g
h
i
m
a
g
in
g
s
u
s
p
icio
n
s
s
co
r
e
a
n
d
a
h
i
g
h
Glea
s
o
n
s
co
r
e
[
1
6
]
.
A
9
m
m
tr
ea
t
m
e
n
t
p
er
im
e
ter
all
ar
o
u
n
d
a
lesi
o
n
v
is
ib
le
o
n
m
a
g
n
et
ic
r
eso
n
a
n
ce
i
m
a
g
i
n
g
w
o
u
ld
ce
r
tai
n
l
y
co
n
s
is
ten
tl
y
m
a
k
e
p
o
s
itiv
e
tr
ea
t
m
en
t
m
et
h
o
d
o
f
th
e
en
tire
h
i
s
to
lo
g
ical
t
u
m
o
r
v
o
lu
m
e
d
u
r
i
n
g
f
o
ca
l
ab
lati
v
e
th
er
ap
y
.
Ho
w
ev
er
,
th
e
clo
g
g
i
n
g
la
y
er
is
p
r
ese
n
t
w
it
h
tu
m
o
r
ca
n
ce
r
g
r
o
w
th
h
e
n
ce
it
is
n
ec
e
s
s
ar
y
to
r
e
m
o
v
e
al
iasi
n
g
e
f
f
ec
t
ar
is
in
g
d
u
e
to
clo
g
g
i
n
g
/p
las
m
a
i
n
f
u
s
i
o
n
.
W
e
ar
e
f
o
cu
s
i
n
g
o
n
b
r
ain
ca
n
ce
r
ca
s
e,
b
u
t
t
h
e
s
o
l
u
tio
n
b
y
N
A
M
m
a
y
b
e
f
o
r
all
t
y
p
es o
f
r
e
g
io
n
t
u
m
o
r
d
etec
tio
n
.
T
h
e
ea
r
lier
in
v
esti
g
atio
n
h
a
s
m
en
tio
n
e
d
t
h
at
Neu
r
o
en
d
o
s
co
p
y
r
esec
tio
n
f
o
r
tu
m
o
r
g
r
o
w
t
h
s
e
n
d
s
u
p
w
it
h
a
7
6
–
9
0
%
d
ec
r
ea
s
e
[
1
7
]
in
th
e
c
h
a
n
ce
o
f
ac
q
u
ir
i
n
g
tu
m
o
r
m
e
lan
o
m
a.
E
v
e
n
s
o
,
ele
m
e
n
ts
lik
e
en
d
o
s
co
p
is
t’
s
i
n
s
er
tio
n
alo
n
g
w
ith
r
ev
u
ls
io
n
ab
ilit
y
,
ca
n
ce
r
o
u
s
g
r
o
w
t
h
p
o
s
itio
n
,
an
d
q
u
alit
y
o
f
co
lo
n
p
r
ep
,
an
d
ca
n
ce
r
o
u
s
g
r
o
w
t
h
d
ep
ictio
n
ca
n
less
e
n
th
e
p
o
ten
c
y
o
f
Ne
u
r
o
en
d
o
s
co
p
y
test
s
.
Fo
r
ex
a
m
p
le,
3
to
4
R
ip
p
e
d
tu
m
o
r
s
ar
e
t
y
p
icall
y
o
v
er
lo
o
k
ed
th
r
o
u
g
h
Neu
r
o
en
d
o
s
co
p
y
,
as
th
e
y
ar
e
co
m
p
lica
ted
to
d
i
s
co
v
er
w
h
e
n
u
s
i
n
g
th
e
t
y
p
ical
w
h
ite
-
li
g
h
t i
m
ag
e
r
eso
lu
tio
n
t
ec
h
n
iq
u
e.
Desp
ite
t
h
e
f
ac
t
t
h
at,
n
eu
r
o
-
e
n
d
o
s
co
p
y
g
iv
e
s
b
en
e
f
it
s
m
o
r
e
th
a
n
tr
ad
itio
n
al
Ne
u
r
o
en
d
o
s
co
p
y
,
th
e
m
et
h
o
d
is
m
o
r
e
co
m
p
lica
ted
as
w
ell
as
ti
m
e
-
co
n
s
u
m
in
g
.
No
w
ad
a
y
s
,
i
m
ag
e
-
e
n
h
a
n
ce
d
n
eu
r
o
-
en
d
o
s
co
p
y
co
n
tin
u
es
to
b
e
r
ec
o
m
m
e
n
d
e
d
to
f
u
r
th
er
i
m
p
r
o
v
e
m
icr
o
v
ascu
lar
co
m
p
ar
is
o
n
alo
n
g
w
i
th
th
e
a
s
s
i
s
t
in
t
h
e
in
s
ta
n
t
i
m
ag
e
r
eso
lu
tio
n
o
f
s
i
m
p
le
p
atter
n
s
a
n
d
also
s
h
ad
e
v
ar
ian
ce
s
.
A
d
i
g
ital
g
r
ap
h
ic
i
m
p
r
o
v
e
m
e
n
t
en
ta
il
s
i
m
a
g
e
p
r
o
ce
s
s
in
g
al
g
o
r
ith
m
s
.
Dif
f
er
en
t a
l
g
o
r
it
h
m
s
ar
e
alr
ea
d
y
r
ec
o
m
m
e
n
d
ed
ac
co
r
d
in
g
to
th
e
tar
g
et
s
a
n
d
als
o
th
e
m
o
d
alit
y
o
f
i
m
a
g
e
r
eso
lu
t
io
n
,
w
h
ic
h
in
c
lu
d
es
t
h
e
o
n
e
w
h
ic
h
i
m
p
r
o
v
e
s
th
e
s
h
ap
e
o
r
lig
h
t
p
atter
n
s
w
i
th
in
th
e
p
r
o
d
u
ce
d
g
r
ap
h
ics,
o
n
e
p
ar
ticu
lar
f
o
cu
s
i
n
g
i
m
a
g
e
co
m
p
ar
is
o
n
,
o
n
e
in
te
n
d
ed
f
o
r
co
lo
r
tr
an
s
f
o
r
m
atio
n
an
d
s
o
f
o
r
th
.
T
h
e
ap
p
licatio
n
o
f
e
x
is
t
in
g
s
tr
ate
g
ie
s
lik
e
a
u
to
r
es
o
lu
tio
n
a
n
d
also
s
m
all
i
m
a
g
e
r
eso
lu
tio
n
ca
n
r
es
u
lt
in
e
n
h
an
ce
m
e
n
ts
i
n
t
u
m
o
r
g
r
o
w
t
h
d
iag
n
o
s
is
r
ates,
d
esp
ite
t
h
e
f
ac
t
t
h
at
t
h
i
s
id
ea
is
d
eb
atab
le
[
1
8
]
.
A
ls
o
,
J
o
h
n
s
Ho
p
k
in
s
n
e
u
r
o
s
u
r
g
eo
n
m
a
y
b
e
u
s
i
n
g
f
id
u
cial
m
ar
k
er
s
to
tar
g
et
ce
r
t
ain
ar
ea
s
f
o
r
s
u
r
g
ical
r
esectio
n
o
r
tr
ea
t
m
e
n
t.
B
ef
o
r
e
t
h
e
s
u
r
g
er
y
,
a
C
T
o
r
MRI
s
ca
n
w
i
th
t
h
e
f
i
d
u
cials
in
p
lace
cr
ea
tes
a
3
-
D
m
ap
o
f
y
o
u
r
h
ea
d
.
Du
r
in
g
o
p
er
atio
n
,
t
h
e
f
id
u
c
ial
l
y
p
r
o
v
id
e
a
s
er
ie
s
o
f
r
ef
er
e
n
c
e
p
o
in
ts
f
o
r
th
e
s
u
r
g
eo
n
an
d
i
n
teg
r
ate
n
a
v
i
g
atio
n
d
u
r
in
g
s
u
r
g
er
y
w
it
h
t
h
e
3
-
D
m
ap
(
h
ttp
:/
/
www
.
h
o
p
k
in
s
m
ed
icin
e.
o
r
g
/)
.
T
h
is
i
m
a
g
i
n
g
-
as
s
i
s
ted
tech
n
iq
u
e
h
elp
s
d
o
cto
r
s
af
el
y
f
in
d
t
h
e
w
a
y
ar
o
u
n
d
t
h
e
b
r
ain
an
d
p
r
ec
is
el
y
lo
ca
te
tu
m
o
r
s
,
tar
g
ets
f
o
r
d
ee
p
b
r
ain
s
ti
m
u
latio
n
a
n
d
o
th
er
lesi
o
n
s
in
th
r
ee
d
i
m
en
s
io
n
s
as
h
e
o
r
s
h
e
w
o
r
k
s
,
al
m
o
s
t
lik
e
a
GP
S
s
y
s
te
m
.
T
h
is
k
in
d
o
f
co
m
b
in
ed
o
u
tco
m
es
h
as
b
ee
n
ac
q
u
ir
ed
w
h
e
n
u
s
i
n
g
th
e
n
eu
r
o
-
en
d
o
s
c
o
p
y
tec
h
n
iq
u
e
o
n
e
g
e
n
er
atio
n
b
ef
o
r
e
th
e
p
r
esen
t
o
n
e.
W
e
co
llected
ex
is
tin
g
r
es
u
lt
s
o
f
th
e
r
etr
o
s
p
ec
tiv
e
a
n
al
y
s
i
s
f
o
r
ev
alu
atio
n
o
f
v
is
ib
ili
t
y
o
f
t
u
m
o
r
tes
ted
w
it
h
P
SO
a
s
w
ell
as
S
VM
g
r
ap
h
ics,
b
u
t
w
e
d
id
n
o
t
te
s
t
3
D
i
m
a
g
i
n
g
tech
n
iq
u
e.
E
x
cl
u
s
i
v
el
y
,
w
e
f
o
cu
s
ed
o
n
ascer
tain
i
n
g
w
h
et
h
er
o
r
n
o
t
th
e
a
n
al
y
s
i
s
ab
ilit
ies
o
f
SV
M
an
d
P
SO
w
er
e
b
etter
t
h
a
n
t
h
at
o
f
o
u
r
n
e
w
l
y
d
ev
elo
p
ed
NA
M
(
Neu
r
o
en
d
o
s
co
p
y
A
d
a
p
ter
Mo
d
u
le)
f
o
r
th
e
v
i
s
u
al
izatio
n
o
f
t
u
m
o
r
g
r
o
w
t
h
s
w
h
ich
o
f
te
n
s
u
r
r
o
u
n
d
ed
b
y
u
n
e
v
en
clo
g
g
i
n
g
la
y
er
.
Ma
n
y
t
i
m
e
s
,
s
u
ch
clo
g
g
i
n
g
la
y
er
s
cr
ea
te
v
i
s
ib
il
it
y
o
b
s
tr
u
ctio
n
f
o
r
d
o
cto
r
s
.
Hen
ce
,
m
a
n
y
ti
m
es to
p
r
ed
ict
th
e
ca
n
ce
r
o
u
s
g
r
o
w
t
h
at
in
itial i
n
v
esti
g
atio
n
b
ec
o
m
e
a
m
b
ig
u
o
u
s
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
Neu
r
o
en
d
o
s
co
p
y
g
r
ap
h
ics
o
b
tain
ed
f
r
o
m
M
u
ltiS
p
ec
ial
it
y
Ho
s
p
ital
p
atie
n
t
(
id
en
t
i
t
y
/p
r
iv
ac
y
p
r
o
tecte
d
)
at
th
e
P
u
n
e,
I
n
d
ia
h
ad
b
ee
n
u
s
ed
f
o
r
th
i
s
c
u
r
r
en
t
r
esear
ch
.
E
x
is
ti
n
g
r
es
u
lts
wer
e
o
b
tain
ed
f
r
o
m
w
eb
/p
h
y
s
ician
’
s
d
ata
th
at
c
ar
r
ied
o
u
t
NA
M
a
n
d
UQI
b
y
tr
a
n
s
i
tio
n
i
n
g
in
itial
to
w
a
r
d
s
SVM
m
et
h
o
d
ac
co
m
p
a
n
ied
b
y
t
h
e
P
SO
m
a
n
n
er
.
Do
cto
r
s
u
s
ed
t
h
e
f
o
llo
w
i
n
g
eq
u
ip
m
en
t
i
n
th
is
s
tu
d
y
:
C
T
,
MRI
i
m
a
g
i
n
g
,
lig
h
t so
u
r
ce
s
,
an
d
v
id
eo
p
r
o
ce
s
s
o
r
s
.
Ulti
m
a
tel
y
,
th
e
SVM
cla
s
s
i
f
i
er
[
1
9
]
is
em
p
lo
y
ed
to
d
is
tin
g
u
i
s
h
b
et
w
ee
n
n
o
r
m
a
l
an
d
t
u
m
o
r
b
r
ai
n
i
m
a
g
es
in
f
l
u
e
n
ce
d
b
y
f
r
ac
tal
f
ea
tu
r
es.
T
h
e
SVM
is
d
ep
en
d
en
t
o
n
t
h
e
s
tatis
tical
lear
n
i
n
g
p
r
in
cip
le
w
h
ich
lo
ca
tes
th
e
s
u
itab
le
n
o
n
li
n
e
ar
h
y
p
er
p
lan
e
th
at
d
ec
r
ea
s
es
th
e
an
ticip
ated
class
if
icatio
n
er
r
o
r
.
I
t
w
a
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
6
,
Dec
em
b
er
2
0
1
7
:
3
6
4
3
–
3
6
5
4
3646
s
u
b
s
eq
u
en
t
l
y
e
f
f
ic
ien
t
l
y
i
m
p
l
e
m
en
ted
in
p
r
ev
io
u
s
w
o
r
k
s
[
2
0
-
2
2
]
.
SVM
p
r
o
v
ed
to
b
e
m
o
s
t
ef
f
icie
n
t
i
n
ir
is
i
m
a
g
e
p
r
o
ce
s
s
i
n
g
,
f
ea
t
u
r
e
e
x
tr
ac
tio
n
b
y
u
s
in
g
in
d
ep
en
d
e
n
t
c
o
m
p
o
n
en
t
an
al
y
s
is
.
T
h
e
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
is
u
s
ed
f
o
r
ir
is
class
i
f
icatio
n
a
n
d
r
ec
o
g
n
itio
n
[
2
3
]
.
T
h
e
SVM
m
et
h
o
d
is
p
ar
ticu
lar
l
y
i
m
p
o
r
tan
t
in
t
h
e
f
ield
o
f
m
ac
h
in
e
lear
n
i
n
g
.
T
h
er
e
ar
e
m
an
y
ad
v
a
n
ta
g
es
o
f
t
h
e
SVM
,
s
u
c
h
as
d
ir
ec
t
g
eo
m
etr
ic
i
n
t
er
p
r
etatio
n
,
eleg
an
t
m
at
h
e
m
a
tical
tr
ac
tab
ilit
y
,
a
n
d
h
ig
h
ac
cu
r
ac
y
[
2
4
]
.
SVM
b
ec
o
m
e
s
f
a
m
o
u
s
w
h
e
n
,
u
s
in
g
p
ix
el
m
ap
s
as
in
p
u
t;
it
g
iv
e
s
ac
c
u
r
ac
y
co
m
p
ar
ab
le
to
s
o
p
h
is
t
icate
d
n
e
u
r
al
n
et
w
o
r
k
s
w
it
h
elab
o
r
ated
f
ea
t
u
r
es
in
r
ec
o
g
n
itio
n
o
f
tas
k
[
2
5
]
.
Hen
ce
,
w
e
ad
o
p
ted
SVM
ad
v
an
ta
g
es to
b
u
ild
n
e
w
a
lg
o
r
ith
m
.
W
e
p
r
o
v
id
e
a
n
e
w
s
o
lu
tio
n
to
d
etec
t
th
e
t
u
m
o
r
i
n
b
r
ai
n
m
a
g
n
etic
r
eso
n
an
ce
i
m
ag
e
s
.
First,
th
e
r
ec
o
m
m
e
n
d
ed
au
to
m
a
ted
s
y
s
t
e
m
is
ea
s
y
to
e
m
p
lo
y
an
d
to
u
n
d
er
s
tan
d
.
I
t
is
m
o
s
tl
y
b
a
s
ed
o
n
s
eg
m
e
n
tatio
n
o
f
b
io
lo
g
ical
tis
s
u
e,
e
v
al
u
atio
n
o
f
t
h
e
d
ir
ec
tio
n
al
s
p
ec
tr
al
d
is
tr
i
b
u
tio
n
o
f
t
h
e
s
e
g
m
en
ted
i
m
a
g
e,
an
d
ev
al
u
atio
n
o
f
m
u
lti
-
s
ca
le
f
r
ac
tal
s
o
f
t
h
e
c
alcu
lated
d
ir
ec
tio
n
al
s
p
ec
tr
al
d
is
tr
ib
u
tio
n
.
C
er
tai
n
l
y
,
th
e
ex
is
te
n
ce
o
f
t
u
m
o
r
in
f
lu
e
n
ce
s
t
h
e
h
o
m
o
g
e
n
eit
y
o
f
th
e
b
io
lo
g
ical
tis
s
u
e
a
n
d
also
its
g
eo
m
etr
ical
b
eh
a
v
i
o
r
s
.
I
n
th
is
asp
ec
t,
d
ir
ec
tio
n
al
s
p
ec
tr
al
d
is
tr
ib
u
ti
o
n
en
ab
le
s
c
h
ar
ac
ter
izin
g
g
e
o
m
e
t
r
ical
p
atter
n
s
m
a
n
i
f
este
d
b
y
s
e
g
m
en
ta
tio
n
o
u
tco
m
es
an
d
a
s
w
ell
te
x
tu
r
al
f
ea
t
u
r
es
o
f
t
h
e
b
io
lo
g
ical
tis
s
u
e.
T
h
ese
p
atter
n
s
ca
n
b
e
co
n
s
id
er
ed
at
v
ar
io
u
s
s
ca
les
b
y
e
m
p
lo
y
in
g
f
r
ac
tal
esti
m
ates.
Seco
n
d
,
w
e
ass
e
s
s
th
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
u
n
d
er
NA
M
alg
o
r
it
h
m
i
s
ac
cu
s
to
m
ed
to
f
o
cu
s
in
g
t
h
e
t
u
m
o
r
a
f
f
ec
ted
r
eg
io
n
.
N
A
M
alg
o
r
it
h
m
i
s
i
n
ter
esti
n
g
as
it
h
elp
s
to
f
o
cu
s
o
n
s
e
g
m
e
n
tat
io
n
ap
p
r
o
ac
h
w
it
h
th
e
f
o
cu
s
o
n
tu
m
o
r
ed
g
e
co
r
r
ec
tn
ess
v
is
ib
ilit
y
.
T
h
ir
d
,
u
n
li
k
e
p
r
ev
io
u
s
w
o
r
k
s
[
1
6
–
1
8
]
,
o
u
r
p
r
o
p
o
s
ed
s
y
s
te
m
d
o
esn
'
t
n
ee
d
m
an
u
all
y
-
d
eter
m
i
n
ed
o
r
estab
lis
h
ed
r
eg
io
n
s
o
f
in
ter
est
s
.
A
l
s
o
,
o
u
r
au
to
m
ated
d
etec
tio
n
s
y
s
te
m
is
f
as
t
an
d
p
r
o
v
id
es
h
i
g
h
ac
c
u
r
ac
y
;
u
n
li
k
e
th
e
M
R
I
tech
n
iq
u
e
w
h
ic
h
is
u
s
ed
i
n
t
h
e
i
m
ag
e
c
l
ass
i
f
icatio
n
p
r
o
ce
s
s
th
at
g
e
n
er
ates
r
ea
l
-
t
i
m
e
p
s
e
u
d
o
g
r
ap
h
ic
s
.
I
n
f
u
r
th
er
s
ec
tio
n
2
.
1
,
w
e
p
r
esen
ted
co
m
p
u
tatio
n
al
m
et
h
o
d
s
u
s
ed
d
u
r
i
n
g
c
u
r
r
en
t r
esear
ch
an
al
y
s
i
s
.
2
.
1
.
Co
m
p
uta
t
io
na
l
M
et
ho
ds
T
h
e
o
v
er
all
t
u
m
o
r
/s
u
b
d
u
r
al
h
e
m
ato
m
a
d
etec
tio
n
an
d
an
al
y
s
is
s
y
s
te
m
i
s
d
escr
ib
ed
i
n
t
h
e
Fig
u
r
e1
,
w
h
er
e
b
r
ain
MRI
i
s
s
eg
m
e
n
ted
b
y
P
SO/S
V
M/
N
A
M.
T
h
e
o
b
tain
ed
s
eg
m
e
n
ted
i
m
a
g
e
i
s
f
u
r
t
h
er
p
r
o
ce
s
s
ed
to
o
b
tain
its
Au
to
-
f
o
cu
s
Dis
tr
ib
u
tio
n
(
AFD)
w
h
ic
h
a
u
to
m
at
icall
y
cu
ts
d
o
w
n
p
er
ip
h
er
al
f
o
cu
s
a
n
d
i
m
p
r
o
v
es
t
h
e
p
ix
e
l
b
o
u
n
d
ar
y
v
i
s
ib
ilit
y
o
f
t
u
m
o
r
.
I
n
d
ee
d
,
A
FD
i
s
s
u
itab
le
to
ch
a
r
ac
ter
ize
tex
tu
r
e
o
f
b
r
ai
n
b
io
l
o
g
ical
tis
s
u
e.
T
h
en
,
UQI
is
e
m
p
lo
y
ed
to
co
m
p
u
te
t
h
e
p
i
x
el
q
u
alit
y
o
f
t
h
e
AFD
o
f
in
p
u
t
b
r
ain
i
m
a
g
es.
I
n
t
h
i
s
r
e
g
ar
d
,
th
o
s
e
f
r
ac
ta
ls
ar
e
u
s
ed
to
f
o
r
m
t
h
e
f
ea
t
u
r
es
v
ec
to
r
.
Fi
n
all
y
,
t
h
e
N
A
M
is
tr
ain
ed
w
it
h
th
e
f
ea
t
u
r
es
v
ec
to
r
to
clas
s
i
f
y
a
g
iv
e
n
b
r
ain
MR
I
/n
e
u
r
o
-
e
n
d
o
s
co
p
y
i
m
ag
e
as
a
p
r
e
-
s
u
r
g
ical
o
r
p
o
s
t
-
s
u
r
g
ica
l
i
m
a
g
e
to
id
en
tify
w
h
eth
er
o
r
n
o
t
tu
m
o
r
af
f
ec
ted
p
o
r
tio
n
is
clea
r
.
A
ll
t
ec
h
n
iq
u
es;
i
n
clu
d
i
n
g
P
SO
-
b
as
ed
s
eg
m
e
n
tatio
n
al
g
o
r
ith
m
s
,
S
VM
an
d
N
A
M,
ar
e
d
escr
ib
ed
in
n
ex
t
s
u
b
s
ec
t
io
n
s
.
Fig
u
r
e
1
.
P
r
o
p
o
s
ed
s
y
s
te
m
f
o
r
tu
m
o
r
d
etec
tio
n
a
n
d
s
u
r
g
ical
a
n
al
y
s
i
s
2
.
1
.
1
.
P
a
rt
icle
S
w
a
rm
O
pti
miza
t
io
n T
ec
hn
iq
ues
P
ar
ticle
s
w
ar
m
o
p
ti
m
izatio
n
(
P
SO)
is
a
p
o
p
u
latio
n
-
b
a
s
ed
s
e
ar
ch
p
r
o
to
co
l p
r
esen
ted
b
y
Sa
l
i
m
L
a
h
m
ir
[
2
6
]
.
I
n
th
e
i
m
ag
e
s
eg
m
e
n
ta
tio
n
f
r
a
m
e
w
o
r
k
,
t
h
e
p
ix
el
s
e
t
m
atc
h
es
to
t
h
e
lo
o
k
u
p
s
p
ac
e,
alo
n
g
w
it
h
t
h
e
o
p
tim
u
m
s
o
l
u
tio
n
m
atc
h
es
to
ex
p
lo
itin
g
t
h
e
b
et
w
ee
n
-
clas
s
d
ev
iatio
n
o
f
th
e
d
is
tr
ib
u
tio
n
o
f
in
te
n
s
it
y
d
eg
r
ee
s
i
n
th
e
i
m
a
g
e.
Fo
llo
w
in
g
t
h
e
n
o
t
atio
n
i
n
[
2
7
]
n
o
tatio
n
,
at
ti
m
e
t
ea
ch
p
ar
ticle
m
o
v
e
s
i
n
a
s
ea
r
ch
s
p
ac
e
w
it
h
p
o
s
itio
n
an
d
v
elo
cit
y
w
h
ic
h
ar
e
d
ep
en
d
en
t o
n
lo
ca
l b
est p
o
s
itio
n
n
ei
g
h
b
o
r
h
o
o
d
b
est
an
d
g
lo
b
al
b
est
in
f
o
r
m
atio
n
as f
o
llo
w
s
:
+
1
=
+
1
1
(
̆
−
)
+
2
2
(
̆
−
)
+
3
3
(
̆
−
)
(
1
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
N
A
M De
ve
lo
p
men
t fo
r
B
etter
B
r
a
in
Tu
mo
r
I
ma
g
e
V
is
u
a
liz
a
t
io
n
(
S
u
n
il L.
B
a
n
g
a
r
e
)
3647
+
1
=
+
+
1
(
2
)
W
h
er
e
th
e
p
ar
a
m
eter
s
,
1
,
2
,
an
d
3
ar
e
r
esp
ec
tiv
el
y
t
h
e
w
ei
g
h
t
s
o
f
th
e
i
n
er
tial
i
n
f
l
u
e
n
ce
,
g
lo
b
a
l
b
est,
lo
ca
l
b
est,
an
d
n
eig
h
b
o
r
h
o
o
d
b
est
w
h
en
d
eter
m
i
n
i
n
g
th
e
n
e
w
v
elo
cit
y
.
T
h
e
p
ar
am
et
er
s
1
,
2
,
an
d
3
ar
e
v
ec
to
r
s
w
it
h
u
n
i
f
o
r
m
r
an
d
o
m
n
u
m
b
er
s
b
et
w
ee
n
0
an
d
1
ass
o
ciate
d
to
ea
ch
co
m
p
o
n
en
t
o
f
E
q
.
(
2
)
.
T
h
e
f
itn
es
s
f
u
n
ctio
n
u
s
ed
to
ev
al
u
ate
th
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
p
ar
ticles i
s
g
i
v
en
b
y
[
2
7
]
[
2
8
]
:
(
)
Ф
=
1
<
1
<
⋯
<
−
1
<
(
3
)
w
h
er
e,
is
th
e
b
etw
ee
n
-
clas
s
v
ar
ian
ce
o
f
th
e
i
m
a
g
e
in
te
n
s
it
y
d
is
tr
ib
u
tio
n
s
,
is
a
g
iv
en
t
h
r
esh
o
ld
,
is
th
e
i
n
te
n
s
it
y
le
v
el,
an
d
is
th
e
n
u
m
b
er
o
f
cla
s
s
es
i
n
to
w
h
ic
h
th
e
p
i
x
els
o
f
th
e
i
m
a
g
e
ar
e
d
iv
id
ed
(
b
y
g
r
o
u
p
in
g
p
ix
el
s
in
ter
m
s
o
f
th
eir
g
r
a
y
lev
e
ls
b
ee
n
w
i
th
i
n
a
s
p
ec
if
ied
r
an
g
e)
.
T
h
e
is
g
iv
e
n
b
y
[
2
9
]
to
s
o
lv
e
th
e
p
r
o
b
lem
w
i
th
P
SO
al
g
o
r
i
t
h
m
th
at
it
m
a
y
g
et
s
t
u
ck
i
n
a
s
u
b
-
o
p
tim
a
l
s
o
lu
tio
n
r
eg
io
n
[
3
0
]
[
3
1
]
.
I
n
o
r
d
er
t
o
s
o
lv
e
th
is
p
r
o
b
lem
,
w
e
in
tr
o
d
u
ce
d
th
e
NA
M
w
h
er
e
p
ix
el
s
elec
tio
n
is
e
m
p
lo
y
ed
to
escap
e
f
r
o
m
lo
ca
l
o
p
ti
m
a.
I
n
th
is
co
n
tex
t,
w
h
en
a
m
a
x
i
m
u
m
t
h
r
esh
o
ld
n
u
m
b
er
o
f
s
tep
s
is
r
ea
ch
ed
w
it
h
o
u
t
i
m
p
r
o
v
in
g
t
h
e
f
it
n
es
s
f
u
n
ctio
n
,
th
e
w
o
r
s
t
p
ar
ticle
is
d
elete
d
.
T
h
e
s
tep
s
co
u
n
ter
is
g
i
v
en
a
s
:
(
)
=
(
1
−
1
+
1
)
(
4
)
w
h
er
e
is
th
e
n
u
m
b
er
o
f
p
ar
ticles
d
elete
d
f
r
o
m
th
e
s
w
ar
m
w
h
e
n
th
er
e
is
n
o
i
m
p
r
o
v
e
m
en
t
in
f
it
n
es
s
d
u
r
in
g
an
iter
atio
n
.
A
n
e
w
s
w
ar
m
is
cr
ea
ted
w
it
h
a
p
r
o
b
ab
ilit
y
o
f
=
/
,
w
it
h
is
a
r
an
d
o
m
n
u
m
b
er
in
[
0
,
1
]
an
d
is
t
h
e
n
u
m
b
er
o
f
s
w
ar
m
s
.
A
r
ti
f
ici
al
I
n
telli
g
e
n
ce
tech
n
iq
u
e
s
r
el
ated
to
b
r
ain
ar
e
d
is
cu
s
s
ed
b
y
Su
tik
n
o
T
o
le
et
a
l [
3
2
]
.
Fra
ctio
n
al
ca
lc
u
l
u
s
b
a
s
ed
o
n
t
h
e
co
n
ce
p
t
o
f
f
r
ac
tio
n
al
d
if
f
er
en
tial
is
u
s
ed
to
co
n
tr
o
l
t
h
e
co
n
v
er
g
en
ce
r
ate.
T
h
e
f
r
ac
tio
n
al
d
i
f
f
er
e
n
tia
l
o
f
a
g
i
v
en
s
ig
n
al
(
)
is
g
i
v
en
b
y
:
[
(
)
]
=
1
∑
(
−
1
)
(
+
1
)
(
−
)
(
+
1
)
(
−
+
1
)
=
0
(
5
)
W
h
er
e
,
,
an
d
ar
e
r
esp
ec
tiv
ely
th
e
g
a
m
m
a
f
u
n
ctio
n
,
t
h
e
s
a
m
p
lin
g
p
er
io
d
,
an
d
th
e
tr
u
n
c
atio
n
o
r
d
er
.
2
.
1
.
2
.
Su
pp
o
rt
Vec
t
o
r
M
a
chi
ne
Cla
s
s
if
ier
Giv
e
n
a
tr
ai
n
in
g
s
et
{(
x
i,
y
i)
,
i
=
1
,
2
,
.
.
.
,
m
},
w
h
er
e
th
e
in
p
u
t
∈
,
an
d
clas
s
lab
els
∈
{
+
1
,
−
1
}
,
th
e
s
ep
ar
atio
n
h
y
p
er
-
p
la
n
e
f
o
r
a
n
o
n
lin
ea
r
S
VM
class
if
ier
[
2
2
]
h
as th
e
f
o
llo
w
i
n
g
f
o
r
m
:
(
)
=
(
∑
(
)
+
=
)
(
6
)
w
h
er
e
is
a
k
er
n
el
f
u
n
ctio
n
(
f
o
r
in
s
tan
ce
,
a
p
o
ly
n
o
m
ial
k
er
n
e
l)
,
is
th
e
L
ag
r
a
n
g
e
m
u
l
tip
lier
,
an
d
is
a
p
ar
am
eter
to
b
e
esti
m
a
ted
.
I
n
th
i
s
w
o
r
k
,
th
e
p
o
l
y
n
o
m
ial
d
eg
r
ee
w
a
s
v
ar
ied
f
r
o
m
2
to
4
.
Hig
h
er
o
r
d
er
s
w
er
e
ig
n
o
r
ed
b
ec
au
s
e
o
f
a
h
i
g
h
er
co
m
p
u
tatio
n
al
b
u
r
d
en
w
it
h
n
o
s
u
b
s
ta
n
tial
g
ai
n
i
n
cla
s
s
i
f
ic
atio
n
ac
cu
r
ac
y
f
r
o
m
o
u
r
ex
p
er
ien
ce
s
.
T
h
e
leav
e
-
o
n
e
-
o
u
t
cr
o
s
s
-
v
alid
atio
n
m
eth
o
d
is
ad
o
p
ted
f
o
r
v
alid
atio
n
o
f
e
x
p
er
i
m
e
n
ts
,
a
n
d
av
er
ag
e
an
d
s
ta
n
d
ar
d
d
ev
iatio
n
o
f
ac
cu
r
ac
y
r
ate,
s
en
s
iti
v
i
t
y
an
d
s
p
ec
i
f
icit
y
ar
e
ca
lcu
la
ted
to
ev
alu
ate
th
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
SVM.
I
n
o
u
r
w
o
r
k
,
p
o
s
iti
v
e
s
a
m
p
les
a
n
d
n
e
g
ativ
e
s
a
m
p
le
s
ar
e
r
esp
ec
tiv
el
y
p
r
e
-
s
u
r
g
ica
l
tu
m
o
r
a
n
d
p
o
s
t
s
u
r
g
ical
i
m
a
g
es.
T
h
e
N
AM
i
s
ad
o
p
ted
as
m
ai
n
cla
s
s
i
f
ier
a
lth
o
u
g
h
d
i
f
f
er
en
t
o
n
e
s
m
a
y
b
e
co
n
s
id
er
ed
a
n
d
co
m
p
ar
ed
u
s
i
n
g
UQI
.
F
u
r
t
h
er
s
ec
tio
n
2
.
1
.
3
p
r
o
v
id
es
th
e
N
A
M
tech
n
iq
u
e.
Fe
w
d
a
ta
m
in
i
n
g
alg
o
r
ith
m
s
p
r
o
p
o
s
ed
b
y
A
b
d
a
r
,
Mo
lo
u
d
,
et
al.
w
er
e
also
s
tu
d
ied
[
3
3
]
.
S.
L
.
B
an
g
ar
e
et
al.
h
av
e
p
r
o
p
o
s
ed
th
e
m
et
h
o
d
s
o
f
B
r
ain
tu
m
o
r
id
en
t
if
icatio
n
,
clas
s
i
f
icatio
n
an
d
S
V
M,
w
h
ic
h
w
er
e
al
s
o
s
t
u
d
ied
f
o
r
th
i
s
e
x
p
er
i
m
e
n
t
[
3
4
]
[
3
5
]
[
3
6
]
.
2
.
1
.
3
.
Neuro
-
E
nd
o
s
co
py
Ada
pte
r
M
o
du
le
T
h
e
NA
M
tech
n
iq
u
e
en
ab
les
in
s
ta
n
t
m
o
d
alit
y
tr
an
s
itio
n
i
n
g
th
r
o
u
g
h
P
SO
an
d
SVM
e
m
p
lo
y
i
n
g
a
Ma
tlab
co
n
tr
o
l
in
ter
f
ac
e.
W
e
ca
n
ea
s
il
y
u
s
e
r
ea
l
ti
m
e
i
m
a
g
e
s
ca
p
tu
r
ed
b
y
MRI
an
d
n
e
u
r
o
-
en
d
o
s
co
p
y
w
i
th
o
u
t
h
av
i
n
g
m
ag
n
i
f
icatio
n
h
ad
b
ee
n
ta
k
en
a
n
d
d
ig
itall
y
ag
ed
i
n
e
lectr
o
n
ic
d
ig
ita
l
h
ea
lth
ca
r
e
r
ec
o
r
d
s
.
T
h
e
g
r
ap
h
ic
s
w
er
e
b
ei
n
g
d
eter
m
i
n
ed
b
y
a
h
ig
h
l
y
s
k
illed
p
h
y
s
ician
b
li
n
d
ed
to
th
e
p
r
esen
t
s
t
u
d
y
’
s
ap
p
r
o
ac
h
es.
T
h
e
N
A
M
,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
6
,
Dec
em
b
er
2
0
1
7
:
3
6
4
3
–
3
6
5
4
3648
SVM
an
d
P
SO
g
r
ap
h
ic
s
f
o
r
ev
er
y
s
in
g
le
le
s
io
n
h
ad
b
ee
n
d
eliv
er
ed
elec
tr
o
n
icall
y
,
ar
b
itra
r
il
y
o
r
g
an
ized
,
an
d
a
Mic
r
o
s
o
f
t
P
o
w
er
P
o
in
t
d
e
m
o
n
s
tr
atio
n
w
as
d
e
v
elo
p
ed
.
T
h
ese
k
in
d
s
o
f
i
m
a
g
e
s
p
r
o
d
u
ce
d
b
y
N
AM
co
n
s
i
s
t
o
f
f
i
n
e
d
etail
s
w
h
ic
h
co
u
ld
d
is
ti
n
g
u
i
s
h
t
h
e
i
n
d
iv
id
u
al
o
r
t
h
e
l
esio
n
at
tr
ib
u
tes
f
o
r
n
o
r
m
a
l
b
r
ain
,
p
r
e
-
s
u
r
g
ical
a
n
d
p
o
s
t
s
u
r
g
ical
b
r
ain
i
m
a
g
es.
S
k
illed
h
ea
l
th
p
r
o
f
ess
io
n
als
to
o
k
p
ar
t
in
t
h
e
i
m
a
g
e
a
s
s
e
s
s
m
en
t
b
y
v
is
u
al
a
n
al
y
s
i
s
an
d
co
m
p
ar
ed
tech
n
icall
y
u
s
i
n
g
UQI
.
T
h
e
MRI
an
d
Neu
r
o
en
d
o
s
co
p
y
g
r
ap
h
ics
h
av
e
b
ee
n
i
n
tr
o
d
u
ce
d
to
ea
ch
o
n
e
o
f
t
h
e
d
o
cto
r
s
in
ar
b
itra
r
y
o
r
d
er
f
o
r
co
m
p
ar
ab
ilit
y
u
s
i
n
g
th
e
N
A
M.
Do
cto
r
s
o
b
tain
ed
ea
ch
o
n
e
o
f
t
h
e
C
T
,
MRI
an
d
n
eu
r
o
-
e
n
d
o
s
co
p
y
g
r
ap
h
ics
f
o
r
v
i
s
ib
ilit
y
o
f
r
ip
p
ed
tu
m
o
r
s
b
ased
o
n
t
h
e
p
u
r
s
u
in
g
r
a
n
g
e:
0
(
g
o
o
d
)
,
+2
(
b
etter
)
,
+3
(
ex
ce
lle
n
t
in
ca
s
e
o
f
NA
M)
,
–
2
(
p
o
o
r
)
,
an
d
also
–
1
(
v
er
y
p
o
o
r
)
.
T
h
e
f
ea
s
ib
le
h
ig
h
es
t
r
an
k
i
n
g
f
o
r
al
m
o
s
t
an
y
g
r
ap
h
ic
h
ad
b
ee
n
+7
,
alo
n
g
w
it
h
t
h
e
p
r
o
b
ab
le
lo
w
es
t
r
an
k
i
n
g
h
ad
b
ee
n
–
7
.
I
f
a
p
er
ce
p
tio
n
ac
q
u
ir
ed
a
c
o
m
p
lete
r
an
k
i
n
g
o
f
+7
o
r
ev
en
m
o
r
e,
th
e
g
r
ap
h
ic
w
as
t
h
o
u
g
h
t
to
b
e
en
h
an
ce
d
,
a
r
an
k
in
g
in
b
et
w
ee
n
+8
an
d
–
8
s
u
g
g
ested
n
o
tr
an
s
f
o
r
m
,
alo
n
g
w
ith
a
r
an
k
i
n
g
o
f
–
1
0
o
r
a
s
m
a
ller
a
m
o
u
n
t
p
o
in
ted
o
u
t
d
i
m
i
n
is
h
ed
p
r
esen
ce
.
T
h
is
w
o
r
k
m
o
r
eo
v
er
an
al
y
ze
d
t
h
e
ex
i
s
ti
n
g
Neu
r
o
en
d
o
s
co
p
ic
i
m
a
g
e
(
r
ef
er
Fig
u
r
e
2
b
ello
w
)
f
o
r
f
u
r
t
h
e
r
co
m
p
ar
is
o
n
w
it
h
NA
M
o
u
tp
u
t
i
m
ag
e
s
(
r
ef
er
F
ig
u
r
e
3
in
Sectio
n
3
)
.
Dis
s
i
m
ilar
ities
b
et
w
ee
n
i
m
a
g
es
h
a
v
e
b
ee
n
an
al
y
ze
d
b
y
v
is
u
al,
UQI
a
n
d
SP
SS
test
.
He
n
ce
,
u
s
in
g
Ya
m
e
n
'
s
f
o
r
m
u
la
we
g
o
t
P
-
v
al
u
e
<0
.
0
5
w
h
ich
m
ea
n
s
N
AM
m
et
h
o
d
is
s
u
p
er
io
r
th
an
SVM
a
n
d
P
SO.
Fig
u
r
e
2
.
Or
ig
in
al
Ne
u
r
o
en
d
o
s
co
p
y
(
S
u
b
d
u
r
al
He
m
ato
m
a)
i
m
ag
e
(
So
u
r
ce
: P
atien
t T
est R
e
p
o
r
t,
2
0
1
6
)
3.
NAM
I
T
E
RAT
I
V
E
I
N
T
E
RP
O
L
AT
I
O
N:
CL
I
NICA
L
CA
SE
S
T
UDY
Desp
ite
th
e
ad
v
a
n
ta
g
es,
th
er
e
ar
e
s
o
m
e
co
m
p
licat
in
g
is
s
u
e
s
ar
o
u
n
d
n
eu
r
o
en
d
o
s
co
p
y
t
h
at
m
i
g
h
t
n
o
t
al
w
a
y
s
b
e
ea
s
i
l
y
tack
led
o
r
av
o
id
ed
.
P
er
f
o
r
m
i
n
g
a
n
e
u
r
o
en
d
o
s
co
p
ic
p
r
o
ce
d
u
r
e
f
o
r
Su
b
d
u
r
al
He
m
ato
m
a
r
eq
u
ir
es
a
h
ig
h
l
y
s
p
ec
ialized
an
d
tr
ain
ed
n
eu
r
o
s
u
r
g
eo
n
,
m
ea
n
in
g
t
h
e
av
ailab
ili
t
y
o
f
t
h
i
s
p
r
o
ce
d
u
r
e
is
n
o
r
m
all
y
r
estricte
d
to
a
s
m
all
n
u
m
b
er
o
f
s
p
ec
ialized
m
ed
ical
f
ac
i
liti
es
[
2
8
]
.
T
h
e
o
p
er
ativ
e
en
v
ir
o
n
m
e
n
t
f
o
r
Su
b
d
u
r
al
He
m
ato
m
a
is
u
n
d
er
s
tan
d
ab
l
y
b
lo
o
d
y
,
w
h
ich
m
i
g
h
t
co
m
p
r
o
m
is
e
t
h
e
s
u
r
g
eo
n
’
s
f
i
eld
o
f
v
i
s
io
n
,
t
h
u
s
d
e
m
a
n
d
i
n
g
a
r
eg
u
lar
clea
n
s
i
n
g
o
f
t
h
e
en
d
o
s
co
p
e.
B
lo
o
d
clo
g
g
i
n
g
is
also
a
p
o
s
s
ib
le
p
r
o
b
lem
t
h
at
r
eq
u
ir
es
t
h
e
en
d
o
s
co
p
e
to
b
e
r
em
o
v
ed
to
b
e
p
r
o
p
er
ly
clea
n
ed
,
an
d
th
en
p
lace
d
b
ac
k
ag
a
in
i
n
s
id
e
th
e
p
atien
t
’
s
b
r
ain
.
Fin
al
l
y
,
t
h
e
o
p
er
atin
g
S
u
b
d
u
r
al
He
m
at
o
m
a
s
h
ea
th
’
s
s
m
a
ll
d
ia
m
eter
li
m
it
s
t
h
e
w
o
r
k
i
n
g
c
h
an
n
el
s
a
n
d
t
h
is
,
in
t
u
r
n
,
h
a
m
p
er
s
th
e
in
s
tr
u
m
e
n
ts
’
m
o
v
e
m
en
ts
.
A
ls
o
,
v
o
r
o
n
o
i
b
r
ain
m
ap
p
in
g
g
i
v
e
s
s
i
m
ilar
d
i
s
ad
v
an
tag
e
f
o
r
r
ea
l
ti
m
e
b
r
ain
tu
m
o
r
d
etec
tio
n
a
n
d
i
m
a
g
e
v
is
u
aliza
tio
n
.
Hen
ce
,
f
o
cu
s
in
g
o
v
er
to
o
v
er
co
m
e
s
u
c
h
d
is
ad
v
an
tag
e
s
w
e
d
ev
el
o
p
ed
th
e
N
A
M
tech
n
iq
u
e
wh
ich
ca
n
alia
s
t
h
e
clo
g
g
i
n
g
/b
lo
o
d
ef
f
ec
t a
n
d
n
eu
r
o
s
u
r
g
eo
n
s
ca
n
ab
le
to
v
ie
w
cl
ea
r
s
tate
o
f
b
r
ain
tu
m
o
r
.
I
n
li
n
e
w
it
h
t
h
e
o
p
in
io
n
t
h
r
o
u
g
h
w
h
ic
h
s
tan
d
ar
d
i
m
ag
e
c
h
ar
ac
ter
is
tic
s
r
ep
ea
ted
s
i
m
ila
r
lo
ca
tio
n
s
co
m
b
i
n
ed
w
i
t
h
t
h
e
a
s
s
es
s
m
en
t
o
f
a
n
ti
-
alia
s
i
n
g
in
t
h
e
ea
r
lier
s
e
g
m
en
t,
w
e
p
r
o
v
id
e
a
co
n
ti
n
u
in
g
Neu
r
o
en
d
o
s
co
p
y
au
to
-
f
o
c
u
s
i
n
ter
p
o
latio
n
s
tr
ate
g
y
o
f
a
n
ti
-
al
iasi
n
g
alo
n
g
w
it
h
lo
ca
tin
g
t
h
e
h
ig
h
-
q
u
al
it
y
i
m
a
g
e.
I
n
th
i
s
ar
ticle,
w
e
w
i
ll
tak
e
i
n
th
e
N
A
M
i
n
ter
p
o
latio
n
o
f
p
u
r
el
y
an
i
n
d
iv
id
u
al
iter
atio
n
i
n
a
d
d
itio
n
to
p
ar
ticu
lar
lo
ca
tio
n
m
ea
s
u
r
e
m
en
t.
I
n
s
id
e
th
e
f
o
llo
w
i
n
g
,
w
e
w
i
ll
ce
r
tain
l
y
d
escr
ib
e
ea
s
y
m
eth
o
d
s
to
ass
ess
r
e
g
io
n
s
i
m
ilar
it
y
,
a
s
s
o
ciate
d
w
it
h
th
e
ar
ea
an
al
y
s
is
f
o
r
d
r
o
p
p
ed
p
ix
el
i
m
p
o
r
tan
ce
ca
lcu
latio
n
.
R
eg
u
lar
d
ev
iati
o
n
r
ef
i
n
es
t
h
e
r
esu
lt
i
n
v
o
l
v
i
n
g
w
i
n
d
o
w
in
ad
d
itio
n
to
s
u
b
-
w
i
n
d
o
w
f
o
c
u
s
le
n
g
t
h
w
it
h
u
lti
m
ate
co
v
er
in
g
r
e
g
io
n
as
s
h
o
w
n
i
n
Fi
g
u
r
e
-
3
.
Fo
r
a
n
y
s
o
lit
ar
y
m
is
s
in
g
p
i
x
el,
f
u
n
ct
io
n
a
s
th
e
to
ler
a
n
ce
v
a
lu
e
to
tell
ap
ar
t
th
e
tex
tu
r
e
-
r
elev
an
t
p
i
x
els
b
y
m
ea
n
s
o
f
all
o
f
t
h
e
clo
s
e
p
i
x
el
s
.
Gen
er
all
y
,
b
ef
o
r
e
y
o
u
s
tar
t,
in
ter
p
o
lated
b
elief
s
r
eg
ar
d
in
g
d
r
o
p
p
ed
p
ix
els
ar
e
u
s
u
all
y
f
l
a
w
ed
all
o
f
w
h
ich
d
eli
v
er
th
e
d
y
s
f
u
n
ctio
n
to
th
e
co
m
p
u
tatio
n
o
f
t
h
e
w
i
n
d
o
w
d
i
m
en
s
io
n
s
.
Fro
m
t
h
e
c
u
r
io
s
it
y
r
eg
ar
d
in
g
e
x
a
m
in
at
io
n
,
w
e
p
r
esu
m
e
t
h
e
p
ar
tic
u
lar
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
N
A
M De
ve
lo
p
men
t fo
r
B
etter
B
r
a
in
Tu
mo
r
I
ma
g
e
V
is
u
a
liz
a
t
io
n
(
S
u
n
il L.
B
a
n
g
a
r
e
)
3649
v
ar
iatio
n
r
elatin
g
to
b
eg
i
n
with
in
ter
p
o
lated
v
a
lu
e
co
n
n
e
cted
w
ith
s
u
b
-
w
i
n
d
o
w
i
n
ad
d
itio
n
to
t
h
e
m
ai
n
w
i
n
d
o
w
-
v
al
u
e
to
w
atc
h
a
n
y
p
er
f
ec
tio
n
co
n
n
ec
ted
w
i
th
i
m
ag
e
ca
p
tu
r
e
f
o
c
u
s
w
it
h
ze
r
o
as
w
e
ll
as
s
ta
n
d
ar
d
d
ev
iatio
n
.
(
a)
Or
ig
in
al
I
m
a
g
e
(
b
)
NA
M
ap
p
lied
A
u
to
-
Fo
c
u
s
I
m
ag
e
Fig
u
r
e
3
.
Ou
tp
u
t I
m
ag
e
s
r
ep
r
esen
tat
io
n
o
f
S
u
b
d
u
r
al
He
m
ato
m
a
by
Ne
u
r
o
en
d
o
s
co
p
y
A
d
ap
ter
Mo
d
u
le
(
NA
M)
A
t
t
h
is
p
o
in
t,
w
e
ca
n
ea
s
il
y
n
o
tice
th
at
w
i
n
d
o
w
r
e
g
io
n
⨃
a
(
)
i
s
u
s
ed
s
i
n
ce
co
m
m
o
n
r
eg
io
n
,
b
ey
o
n
d
j
u
s
t
th
e
ea
r
l
y
in
ter
p
o
latio
n
u
n
til
n
(
)
.
A
ll
i
n
ter
p
o
lated
s
e
m
i
-
lo
ca
l
W
in
d
o
w
ar
ea
s
⨃
a
(
)
an
d
s
u
b
-
w
i
n
d
o
w
ar
ea
⨃
n
(
)
ar
e
u
s
u
a
ll
y
w
e
ll
b
alan
ce
d
u
s
in
g
t
h
e
r
eg
u
lar
r
eg
io
n
.
A
s
e
x
p
lain
ed
a
f
ter
u
s
in
g
t
h
is
m
u
lti
-
s
ca
l
e
ap
p
r
o
ac
h
,
to
f
ea
t
u
r
e
th
e
b
e
n
ef
its
o
f
ea
ch
a
n
d
ev
er
y
s
ig
n
i
f
ica
n
t
as
w
ell
as
s
m
al
l
-
s
ca
le
p
ix
e
l
lo
ca
tio
n
,
w
e
g
o
t
p
o
s
iti
v
e
r
es
u
lts
ca
u
s
ed
b
y
m
an
y
p
r
ec
ed
in
g
co
m
p
u
tatio
n
s
(
as
d
ep
icted
i
n
N
A
M
al
g
o
r
ith
m
)
j
o
in
tl
y
as
s
tr
ateg
ie
s
f
o
r
th
e
s
ca
led
-
d
o
w
n
m
ea
s
u
r
e
m
e
n
t
v
ar
i
atio
n
.
T
h
e
ac
tu
al
i
n
ter
p
o
late
d
v
alu
e
s
as
s
o
ciate
d
w
it
h
d
r
o
p
p
ed
p
ix
els
o
n
e
o
f
m
an
y
i
n
p
u
ts
is
b
r
o
ad
as
w
ell
as
d
iv
er
s
e.
W
i
n
d
o
w
r
eg
io
n
i
n
cl
u
d
es
m
er
el
y
p
ix
e
ls
an
d
it
is
co
n
ti
n
u
o
u
s
;
a
s
a
r
es
u
l
t,
th
e
s
u
b
-
w
in
d
o
w
r
eg
io
n
ca
n
b
e
u
s
ed
as
t
h
e
r
eg
u
lar
ar
ea
to
eq
u
al
j
u
s
t
ab
o
u
t
al
l
clo
s
e
b
y
o
r
ig
in
al
w
i
n
d
o
w
p
lace
s
it d
o
esn
'
t
m
atter
w
h
a
t in
p
u
t
th
e
y
ar
e
t
h
r
o
u
g
h
.
4.
NE
URO
E
NDO
SCO
P
Y
AD
AP
T
E
R
M
O
DULE
(
NAM
)
W
ith
t
h
e
an
ti
-
a
lias
i
n
g
p
er
s
p
ec
tiv
e,
it
's
ap
p
r
o
p
r
iate
to
s
elec
t
lar
g
er
r
eg
io
n
m
ea
s
u
r
e
m
e
n
ts
s
o
th
at
t
h
e
y
ca
n
u
n
co
v
er
a
lo
t
m
o
r
e
t
r
u
s
t
w
o
r
th
y
p
i
x
els.
T
h
e
m
o
r
e
i
m
p
o
r
tan
t
th
e
r
eg
io
n
s
iz
in
g
,
m
o
r
e
clo
s
e
b
y
p
i
x
els
ar
e
u
tili
ze
d
s
in
ce
t
h
e
m
et
h
o
d
to
lo
ca
te
clo
g
g
i
n
g
-
la
y
er
-
r
elev
a
n
t
p
ix
els
a
s
w
ell
a
s
r
ed
u
ce
t
h
e
p
r
ec
is
e
co
n
s
eq
u
e
n
c
e
r
elate
d
to
aliasi
n
g
ar
tif
ac
ts
.
N
ev
er
th
ele
s
s
,
th
e
g
r
ea
t
te
x
t
u
r
es
in
th
e
co
n
v
e
n
tio
n
al
i
m
a
g
e
r
e
q
u
ir
e
s
ca
led
-
d
o
w
n
r
eg
io
n
m
ea
s
u
r
e
m
en
t
s
to
o
b
tai
n
m
u
c
h
b
etter
q
u
ick
co
m
p
r
es
s
ed
f
ac
t
s
.
T
h
e
o
b
j
ec
ts
m
a
y
ap
p
ea
r
d
u
e
to
r
eg
io
n
m
is
m
atc
h
i
n
g
.
No
n
et
h
eles
s
,
it
m
a
y
b
e
d
i
f
f
icu
l
t
to
co
n
s
id
er
t
h
e
p
r
o
p
er
ar
ea
is
s
izin
g
o
b
s
er
v
ed
i
m
a
g
e
g
etti
n
g
aliasi
n
g
.
W
it
h
t
h
is
p
ar
ticu
lar
d
o
cu
m
en
t,
co
n
s
id
er
i
n
g
b
o
th
o
f
t
h
o
s
e
d
if
f
er
en
t
ite
m
s
,
w
e
s
u
g
g
e
s
t
r
ea
l
r
ec
u
r
r
i
n
g
m
u
lti
-
s
ca
le
i
n
ter
p
o
latio
n
s
o
l
u
t
io
n
s
w
h
ich
r
ep
licate
t
h
e
ac
tu
a
l
s
e
m
i
-
lo
ca
l
i
n
ter
p
o
latio
n
th
r
o
u
g
h
b
ig
to
s
ca
led
-
d
o
w
n
r
eg
io
n
s
iz
in
g
r
ec
u
r
r
en
t
l
y
.
Alg
o
rit
h
m
1
:
Neuro
endo
s
co
py
Ada
pte
r
M
o
du
le
I
np
ut:
M
R
I
/
Neu
r
o
en
d
o
s
co
p
y
m
ed
ical
i
m
ag
e
Step
-
1
.
I
n
itializ
in
g
b
o
u
n
d
ar
ies
o
f
w
i
n
d
o
w
:
(
a)
I
n
itial a
r
ea
o
f
i
m
a
g
e
p
o
in
t
=
0
,
I
n
itial c
o
o
r
d
in
ates o
f
w
i
n
d
o
w
=
(
0
,
0
)
i.
e.
s
tar
t p
o
in
t o
f
d
iag
o
n
al
le
n
g
t
h
o
f
i
m
a
g
e
p
o
in
t
w
i
n
d
o
w
(
b
)
Set step
s
ize
as p
er
s
tan
d
ar
d
d
ev
iatio
n
⊛
an
d
ar
ea
∩
(
)
(
c)
Set
as
th
e
r
ef
er
en
ce
d
s
ta
n
d
ar
d
d
ev
iatio
n
o
f
i
m
a
g
e
p
o
in
t
f
o
r
∩
(
)
(
d
)
R
ec
o
r
d
im
ag
e
b
o
u
n
d
ar
ies b
y
m
atc
h
i
n
g
p
i
x
el
co
o
r
d
in
ates
o
f
f
i
n
al
w
i
n
d
o
w
i
m
a
g
e
p
o
in
ts
(
e)
Sto
r
e
v
alu
es o
f
f
ir
s
t
w
in
d
o
w
i
m
a
g
e
p
o
in
t a
s
a
r
ef
er
e
n
ce
w
i
n
d
o
w
i
m
a
g
e
p
o
in
t
’
s
ar
ea
Step
-
2
.
I
ter
ativ
e
ap
p
r
o
x
i
m
atio
n
o
f
d
iag
o
n
al
m
a
tr
ix
:
(
a)
R
ec
o
r
d
cu
r
v
e
co
v
er
in
g
co
o
r
d
in
ates o
f
p
ar
allel
li
n
es o
f
d
ia
g
o
n
al
le
n
g
th
(
b
)
Ded
u
ct
w
i
n
d
o
w
i
m
a
g
e
p
o
in
t a
r
ea
f
r
o
m
c
u
r
v
e
co
v
er
i
n
g
ex
tr
a
b
o
u
n
d
ar
y
p
ix
e
ls
(
c)
R
esize
w
in
d
o
w
i
m
ag
e
p
o
in
t to
r
em
ai
n
i
n
g
cu
r
v
e
I
f
⊛
≦
c
urve
a
r
e
a
g
o
to
s
tep
-
1
else
C
o
n
s
id
er
n
ex
t c
u
r
v
e
f
o
r
p
ix
el
co
o
r
d
in
atio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
6
,
Dec
em
b
er
2
0
1
7
:
3
6
4
3
–
3
6
5
4
3650
Step
-
3
.
C
o
m
p
u
te
d
iag
o
n
a
l c
o
o
r
d
in
ates o
f
i
m
a
g
e
p
o
in
t
w
i
n
d
o
w
a
n
d
lo
ca
te
b
o
u
n
d
ar
y
p
ix
el
s.
Step
-
4
.
R
e
m
o
v
e
u
n
f
itted
c
u
r
v
e
p
ix
els.
Step
-
5
.
Ma
p
n
ex
t c
u
r
v
e
w
i
th
i
n
d
iag
o
n
al
p
ix
el
p
o
in
t
s
w
it
h
i
n
w
i
n
d
o
w
∪
an
d
also
∪
.
I
t
s
h
o
u
ld
b
e
p
o
in
ted
o
u
t
th
at
th
is
o
u
tco
m
e
o
r
ig
i
n
ati
n
g
f
r
o
m
j
u
s
t
ab
o
u
t
all
p
r
ev
io
u
s
iter
atio
n
s
ar
e
u
s
u
all
y
g
at
h
er
ed
co
llectiv
el
y
a
s
in
p
u
t
s
to
w
ar
d
s
th
e
f
o
llo
w
i
n
g
litt
le
d
i
m
en
s
io
n
v
er
s
io
n
,
w
h
ic
h
m
a
y
b
e
d
if
f
er
e
n
t
th
r
o
u
g
h
t
h
e
tr
ad
itio
n
al
r
ep
ea
ti
n
g
p
r
o
ce
d
u
r
e.
I
n
m
o
s
t
s
o
litar
y
v
er
s
io
n
,
p
r
ese
n
ted
th
e
r
eg
io
n
s
izin
g
to
g
et
h
er
w
it
h
in
p
u
t
s
w
h
ich
in
co
r
p
o
r
ated
th
e
o
u
tco
m
es
o
r
i
g
in
a
tin
g
f
r
o
m
j
u
s
t
ab
o
u
t
al
l
p
r
ev
io
u
s
iter
atio
n
s
,
ea
c
h
an
d
ev
er
y
in
ter
p
o
lated
s
e
m
i
-
lo
ca
l
w
i
n
d
o
w
-
p
o
r
t
r
eg
io
n
co
n
ce
n
tr
ated
th
r
o
u
g
h
p
ix
e
l
…
…
,
it
d
o
esn
'
t
m
atter
w
h
at
en
te
r
th
e
y
ar
e
o
r
ig
i
n
ati
n
g
f
r
o
m
,
i
s
a
ctu
all
y
co
n
s
id
er
ed
to
w
ar
d
s
d
r
o
p
p
ed
p
ix
el
…
.
.
f
o
cu
s
ed
ar
ea
.
R
ig
h
t
a
f
ter
d
eter
m
in
in
g
m
o
s
t
Neu
r
o
en
d
o
s
co
p
y
alo
n
g
w
it
h
r
elate
d
p
ath
o
lo
g
y
s
t
u
d
ies,
w
e
r
u
led
o
u
t
r
ev
ie
w
s
w
h
er
e:
(
1
)
T
h
er
e
s
ee
m
ed
to
b
e
m
u
lt
ip
le
Neu
r
o
en
d
o
s
co
p
y
co
u
r
s
e
o
f
ac
t
io
n
an
d
p
ath
o
lo
g
y
s
tate
m
e
n
t
ab
o
u
t
th
e
s
a
m
e
ti
m
e;
(
2
)
A
s
i
m
ilar
MRI
r
ec
o
r
d
id
en
tif
ie
d
an
d
m
er
g
ed
w
it
h
Neu
r
o
en
d
o
s
co
p
y
p
r
o
ce
s
s
;
(
3
)
T
h
e
in
d
iv
id
u
al
id
e
n
ti
f
ier
s
ee
m
ed
to
b
e
l
o
s
t a
n
d
(
4
)
T
h
e
p
ar
ticu
lar
d
o
cu
m
e
n
t
w
a
s
p
r
o
d
u
ce
d
u
s
in
g
co
m
p
u
ter
s
o
f
t
w
ar
e.
T
h
e
ap
p
licatio
n
en
ab
les
a
cl
in
ician
to
b
u
ild
a
Neu
r
o
e
n
d
o
s
co
p
y
d
o
cu
m
e
n
t
alo
n
g
w
i
th
p
r
e
-
p
o
p
u
lated
d
r
o
p
-
d
o
w
n
ch
o
ices.
C
o
n
s
id
er
i
n
g
t
h
at
t
h
e
s
t
u
d
ies
ar
e
p
r
o
d
u
c
ed
in
a
v
er
y
o
r
g
an
ized
s
tr
u
ct
u
r
e,
w
e
u
s
ed
i
m
ag
e
s
ca
p
tu
r
ed
d
u
r
in
g
MRI
a
n
d
Neu
r
o
en
d
o
s
co
p
y
.
A
l
s
o
,
w
e
i
n
co
r
p
o
r
ated
r
ep
o
r
ts
p
r
o
v
id
ed
b
y
d
o
cto
r
s
.
T
h
e
r
est
o
f
th
e
Neu
r
o
e
n
d
o
s
co
p
y
r
e
v
ie
w
s
w
it
h
i
n
o
u
r
SP
SS
a
n
al
y
s
is
te
s
t
h
ad
b
ee
n
d
eter
m
i
n
ed
an
d
also
tr
an
s
cr
ib
ed
.
Fo
llo
w
i
n
g
th
e
i
m
p
le
m
e
n
tin
g
o
f
th
e
Ma
tlab
ap
p
licatio
n
f
o
r
th
e
r
ep
o
r
ts
,
w
e
j
o
in
ed
th
e
r
ev
i
e
w
s
o
f
t
h
e
in
d
iv
id
u
al,
t
h
e
p
h
y
s
icia
n
w
h
o
ca
r
r
ied
o
u
t
th
e
p
r
o
ce
s
s
alo
n
g
w
it
h
th
e
m
ed
ical
ce
n
ter
e
x
ac
tl
y
w
h
er
e
it
h
ad
b
ee
n
co
n
d
u
cted
.
Fo
r
m
o
s
t
d
o
cto
r
s
,
th
er
e
w
a
s
t
h
e
e
n
tr
y
to
m
e
r
el
y
t
h
e
p
ar
tic
u
lar
s
u
b
s
et
o
f
th
e
n
e
u
r
o
-
s
u
r
g
er
y
d
ep
ar
tm
e
n
t
t
h
e
y
co
n
d
u
cted
at
Mu
lti
Sp
ec
ialit
y
Ho
s
p
ital.
Mo
s
t
o
f
th
e
s
e
d
o
cto
r
s
h
a
v
e
b
ee
n
i
n
s
p
ec
ialt
y
p
r
ac
tice
an
d
ca
r
r
ied
o
u
t
p
r
o
ce
s
s
es
a
t
Mu
ltiS
p
ec
ial
it
y
Ho
s
p
ita
ls
,
a
n
d
p
r
iv
ate
n
e
u
r
o
-
e
n
d
o
s
co
p
y
f
ac
ilit
ies
f
o
r
in
itial
p
atien
t a
s
s
e
s
s
m
e
n
ts
.
As
w
e
d
is
c
u
s
s
ed
th
e
p
atien
t
cl
in
ical
ca
s
e
d
etails
i
n
p
r
ev
io
u
s
Sectio
n
-
3
an
d
NA
M
al
g
o
r
ith
m
ic
d
etails
in
Sectio
n
-
4
,
n
e
x
t Sec
tio
n
-
5
p
r
o
v
id
es d
etails ab
o
u
t r
es
u
lt a
n
d
an
al
y
s
is
o
f
alg
o
r
it
h
m
ic
r
esear
ch
.
5.
RE
SU
L
T
S AN
D
AN
AL
Y
SI
S
MR
an
d
Neu
r
o
en
d
o
s
co
p
y
b
r
ain
i
m
a
g
es
w
er
e
o
b
tain
ed
f
r
o
m
th
e
M
u
ltiS
p
ec
ial
it
y
Ho
s
p
it
al
an
d
w
eb
f
o
r
v
alid
atio
n
o
f
t
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
.
T
h
e
s
et
in
cl
u
d
ed
1
5
im
a
g
es
o
f
n
o
r
m
a
l
(
h
ea
lt
h
y
)
b
r
ain
s
an
d
1
5
o
f
a
b
r
ain
t
u
m
o
r
.
T
h
e
Ma
t
lab
©
w
it
h
i
m
ag
e
p
r
o
ce
s
s
i
n
g
to
o
lb
o
x
s
o
f
t
w
ar
e
w
a
s
u
s
ed
f
o
r
i
m
ag
e
p
r
o
ce
s
s
i
n
g
a
n
d
class
i
f
icatio
n
.
Fi
g
u
r
e
-
4
s
h
o
w
s
an
ex
a
m
p
le
o
f
n
o
r
m
al
b
r
ain
M
R
I
an
d
its
s
e
g
m
en
tatio
n
r
es
u
lt
s
N
A
M.
T
o
q
u
an
titat
iv
el
y
e
v
al
u
ate
th
e
s
e
g
m
e
n
tatio
n
r
esu
lts
,
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
ea
ch
N
A
M
-
b
ased
s
eg
m
e
n
tatio
n
al
g
o
r
ith
m
i
s
ass
e
s
s
ed
b
y
co
m
p
u
ti
n
g
t
h
e
u
n
i
v
er
s
al
q
u
a
lit
y
in
d
e
x
(
UQI
)
(
h
ttp
://l
i
v
e.
ec
e.
u
te
x
as.e
d
u
/)
wh
ich
is
in
d
ep
en
d
en
t
o
f
th
e
i
m
ag
es
b
ei
n
g
test
ed
a
n
d
th
e
v
ie
w
i
n
g
co
n
d
itio
n
s
o
r
th
e
in
d
i
v
id
u
al
o
b
s
er
v
er
s
.
T
h
e
h
ig
h
e
r
th
e
i
n
d
ex
,
th
e
b
etter
is
th
e
s
eg
m
e
n
tatio
n
r
esu
lt.
Fo
r
in
s
ta
n
ce
,
th
e
v
al
u
es
o
f
UQI
o
b
tain
ed
f
o
r
h
ea
lt
h
y
b
r
ain
an
d
P
SO p
r
o
ce
s
s
ed
h
ea
lt
h
y
b
r
ain
ar
e
0
.
3
6
1
9
an
d
0
.
3
6
2
4
.
A
l
s
o
as
s
h
o
w
n
i
n
Fi
g
u
r
e
4
b
el
lo
w
,
t
h
e
v
a
lu
e
s
o
f
U
QI
o
b
tain
ed
f
o
r
SVM
an
d
N
AM
f
o
r
s
e
g
m
en
ta
tio
n
o
f
a
t
u
m
o
r
a
f
f
ec
ted
b
r
ain
i
m
ag
e
ar
e
r
esp
ec
ti
v
el
y
0
.
4
3
3
7
a
n
d
0
.
4
3
4
2
.
A
lth
o
u
g
h
t
h
ese
v
alu
es
ar
e
clo
s
e,
th
e
NA
M
al
g
o
r
it
h
m
y
ield
s
to
t
h
e
b
est r
esu
lt a
m
o
n
g
t
h
e
m
ac
h
i
n
e
lear
n
in
g
al
g
o
r
ith
m
s
.
Or
ig
i
n
al
i
m
ag
e
h
ea
lt
h
y
b
r
ain
[
1
2
]
P
SO i
m
ag
e
h
ea
lt
h
y
b
r
ai
n
[
1
2
]
SVM
tu
m
o
r
d
etec
tio
n
NA
M
t
u
m
o
r
d
etec
tio
n
Fig
u
r
e
4
.
Seg
m
en
ta
tio
n
r
es
u
lts
o
f
a
h
ea
lt
h
y
b
r
ain
(
o
r
ig
i
n
al
b
r
ain
i
m
ag
e
a
n
d
P
SO p
r
o
ce
s
s
ed
i
m
a
g
e)
an
d
tu
m
o
r
af
f
ec
ted
b
r
ain
u
s
i
n
g
SVM
a
n
d
NA
M.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
N
A
M De
ve
lo
p
men
t fo
r
B
etter
B
r
a
in
Tu
mo
r
I
ma
g
e
V
is
u
a
liz
a
t
io
n
(
S
u
n
il L.
B
a
n
g
a
r
e
)
3651
Ou
tco
m
e
s
r
e
g
ar
d
in
g
N
A
M,
SVM,
an
d
P
SO
ar
e
r
e
v
ea
led
in
T
ab
le
-
1
an
d
al
s
o
N
A
M,
en
h
an
ce
d
v
is
ib
il
it
y
s
ee
m
ed
to
b
e
id
en
tif
ied
7
1
.
4
%,
clo
s
e
b
y
v
i
s
ib
ilit
y
ap
p
ea
r
ed
to
b
e
o
b
tain
ed
in
SVM
6
3
.
1
%
s
u
f
f
icie
n
t
r
ea
s
o
n
f
o
r
P
SO,
m
u
ch
le
s
s
e
n
h
an
ce
d
v
i
s
ib
ilit
y
h
a
d
b
ee
n
s
ee
n
a
s
co
m
p
ar
ed
to
th
e
o
r
ig
in
a
l
i
m
a
g
e
a
s
2
7
.
9
%
f
o
r
th
e
an
al
y
ze
d
ca
s
e.
As
s
h
o
w
n
in
T
ab
le
1
b
ello
w
,
t
h
e
li
k
eli
h
o
o
d
o
f
en
h
an
ce
d
f
iel
d
o
f
v
i
s
io
n
o
f
N
A
M
w
a
s
ap
p
r
ec
iab
l
y
g
r
ea
ter
t
h
an
t
h
at
o
f
S
VM
a
n
d
P
SO
g
r
ap
h
ics
.
T
h
e
N
A
M
s
co
r
es
h
elp
ed
f
o
r
b
ein
g
in
cr
ea
s
ed
,
in
co
n
tr
as
t
to
SVM
an
d
P
SO
s
c
o
r
es
m
a
in
ta
in
ed
to
b
e
r
ed
u
ce
d
.
T
h
er
e
s
ee
m
ed
to
b
e
ab
s
o
lu
tel
y
n
o
s
i
g
n
i
f
ica
n
t
d
if
f
er
e
n
ce
alto
g
et
h
er
f
ield
o
f
v
is
io
n
s
co
r
es
b
et
w
ee
n
h
is
to
p
ath
o
lo
g
ical
co
n
cl
u
s
io
n
s
.
T
h
er
e
h
av
e
b
ee
n
n
o
s
ig
n
i
f
ica
n
t d
is
s
i
m
i
lar
ities
i
n
i
n
ter
-
o
b
s
er
v
er
u
n
d
er
s
tan
d
i
n
g
a
m
o
n
g
Ne
u
r
o
en
d
o
s
co
p
y
g
r
ap
h
ic
s
.
T
ab
le
1
.
C
o
m
p
ar
ativ
e
I
m
ag
e
Vis
ib
ilit
y
An
al
y
s
is
M
e
d
i
c
a
l
Im
a
g
e
T
e
c
h
n
i
q
u
e
V
i
si
b
i
l
i
t
y
(
%
)
a
s
p
e
r
U
QI
D
o
c
t
o
r
R
e
m
a
r
k
Pe
r
f
o
r
m
a
n
c
e
R
a
t
i
n
g
N
A
M
7
1
.
4
%
C
l
e
a
r
t
u
mo
r
r
e
g
i
o
n
v
i
s
i
b
i
l
i
t
y
w
i
t
h
c
l
o
g
g
i
n
g
a
l
i
a
si
n
g
e
f
f
e
c
t
Ex
c
e
l
l
e
n
t
f
o
r
p
r
e
-
su
r
g
i
c
a
l
i
n
v
e
st
i
g
a
t
i
o
n
S
V
M
6
3
.
1
%
T
r
a
d
i
t
i
o
n
a
l
p
e
r
f
o
r
man
c
e
w
i
t
h
p
l
a
sm
a
/
c
l
o
g
g
i
n
g
v
i
si
b
i
l
i
t
y
a
r
o
u
n
d
t
u
mo
r
G
o
o
d
f
o
r
su
r
g
i
c
a
l
e
v
e
n
t
s
PSO
2
7
.
9
%
C
o
mp
a
r
a
t
i
v
e
l
y
l
e
ss t
u
mo
r
r
e
g
i
o
n
v
i
si
b
i
l
i
t
y
G
o
o
d
f
o
r
p
r
i
mary
t
u
mo
r
p
o
si
t
i
o
n
i
n
v
e
st
i
g
a
t
i
o
n
s
P
r
ec
ed
in
g
w
o
r
k
h
as
ac
k
n
o
w
led
g
ed
th
is
v
ar
ian
ce
i
n
q
u
alit
y
a
m
o
n
g
th
e
d
is
ad
v
a
n
tag
e
s
o
f
Neu
r
o
en
d
o
s
co
p
y
a
s
b
ei
n
g
a
s
cr
ee
n
i
n
g
p
r
o
ce
s
s
ap
p
licati
o
n
f
o
r
tu
m
o
r
ca
n
ce
r
m
ali
g
n
an
c
y
tes
tin
g
.
Nich
e
o
r
g
an
izatio
n
s
,
a
s
w
e
ll
a
s
p
r
o
f
ess
io
n
als,
h
a
v
e
g
o
t
r
eq
u
es
ted
v
e
n
d
o
r
s
to
s
u
r
v
e
y
o
n
th
e
Neu
r
o
en
d
o
s
co
p
y
ef
f
ec
tiv
e
n
e
s
s
,
alt
h
o
u
g
h
co
u
p
l
es
o
f
d
o
cto
r
s
ar
e
co
n
f
ir
m
i
n
g
th
eir
q
u
alit
y
.
Ma
tlab
co
u
ld
b
e
o
n
e
m
et
h
o
d
o
f
g
en
er
ati
n
g
s
c
h
ed
u
le
ex
p
o
s
u
r
e
m
o
r
e
p
r
ev
ale
n
t.
O
u
r
f
in
al
r
esu
lt
s
e
m
p
h
a
s
ize
n
u
m
er
o
u
s
p
r
o
s
an
d
co
n
s
o
f
e
m
p
lo
y
i
n
g
M
R
I
clas
s
i
f
icatio
n
u
s
in
g
P
SO,
SVM
f
o
r
t
h
e
q
u
alit
y
o
f
m
ea
s
u
r
i
n
g
,
ap
p
l
y
i
n
g
in
f
o
r
m
atio
n
w
it
h
i
n
i
m
a
g
e
d
atab
ase.
T
h
e
s
i
g
n
if
ic
an
t
b
en
e
f
it
o
f
Ma
tlab
is
it
i
s
m
o
n
e
tar
il
y
p
o
s
s
ib
le.
An
o
th
e
r
ad
v
an
ta
g
e
i
s
t
h
at
Ma
tlab
en
ab
les
v
en
d
o
r
s
to
ca
r
r
y
o
n
m
ak
i
n
g
u
s
e
o
f
s
tan
d
ar
d
p
lo
t
w
h
e
n
e
v
er
ex
p
lai
n
i
n
g
a
f
f
e
cted
in
d
iv
id
u
al
ca
r
e
b
ec
o
m
e
s
ea
s
y
.
T
h
e
Ma
tlab
ap
p
licatio
n
also
o
f
f
er
s
a
n
ad
v
a
n
tag
e
t
h
at
it
i
s
u
s
u
al
l
y
e
f
f
o
r
tles
s
l
y
u
s
ed
to
i
n
clu
d
e
m
o
d
i
f
icat
io
n
s
to
w
ar
d
s
t
h
e
q
u
alit
y
i
m
a
g
e
p
r
o
ce
s
s
in
g
an
d
to
b
u
ild
d
ep
lo
y
ab
le
A
P
I
in
ter
f
a
ce
f
o
r
an
y
m
ed
ical
d
ev
ice
lik
e
n
eu
r
o
-
en
d
o
s
co
p
y
t
o
o
ls
.
Her
e,
w
e,
i
n
t
h
e
b
e
g
in
n
i
n
g
,
ca
r
r
y
o
u
t
tr
ial
s
o
f
o
u
r
m
et
h
o
d
co
m
b
in
ed
w
i
th
v
ar
io
u
s
k
i
n
d
s
r
eg
u
lar
ize
s
o
as
to
v
i
n
d
icate
t
h
e
n
ec
es
s
it
y
o
f
e
m
p
lo
y
in
g
t
h
e
N
AM
ap
p
r
o
ac
h
f
o
r
i
m
ag
e
en
h
an
ce
m
e
n
t.
C
o
n
s
eq
u
en
tl
y
,
w
e
an
al
y
ze
t
h
e
r
ec
o
m
m
e
n
d
ed
ap
p
r
o
ac
h
w
it
h
p
r
esen
t
N
A
M
au
to
-
f
o
cu
s
i
n
ter
p
o
latio
n
m
e
th
o
d
s
o
n
r
ea
l
-
ti
m
e
m
ed
ical
i
m
a
g
es
w
h
ic
h
ar
e
d
o
w
n
-
s
a
m
p
led
w
it
h
s
u
g
g
ested
tech
n
iq
u
e
in
m
i
n
i
m
izi
n
g
ali
asin
g
i
m
p
lica
tio
n
s
.
Ulti
m
a
tel
y
,
w
e
co
m
p
ar
ed
th
e
r
ec
o
m
m
e
n
d
ed
(
NA
M)
m
et
h
o
d
u
s
in
g
ex
i
s
ti
n
g
SVM
an
d
P
SO
tech
n
iq
u
es
to
id
en
ti
f
y
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
N
A
M.
I
n
th
e
f
u
n
d
a
m
en
tal
r
esear
ch
,
w
e
e
x
ec
u
ted
o
u
r
N
A
M
al
g
o
r
ith
m
ic
ap
p
r
o
ac
h
f
o
r
th
e
Ne
u
r
o
en
d
o
s
co
p
y
i
m
a
g
es,
r
esp
ec
ti
v
el
y
.
T
h
e
ex
p
er
i
m
e
n
tal
r
es
u
lt
s
ar
e
co
n
f
ir
m
ed
i
n
Fig
u
r
e
-
4
w
h
ic
h
i
n
t
u
r
n
d
is
p
la
y
s
a
n
eig
h
b
o
r
h
o
o
d
lo
ca
tio
n
o
f
th
e
tu
m
o
r
o
u
s
g
r
o
w
t
h
w
it
h
o
u
t
b
l
o
o
d
clo
g
g
in
g
o
r
p
las
m
a
in
cl
u
s
io
n
,
w
h
ic
h
o
f
te
n
co
n
s
is
tin
g
o
f
e
x
ce
s
s
i
v
e
aliasi
n
g
ar
ti
f
ac
t
s
.
W
e
ca
n
q
u
ic
k
l
y
n
o
tice
th
at
th
e
co
n
to
u
r
o
f
t
h
e
i
m
ag
e
p
atter
n
is
r
esto
r
ed
ad
eq
u
ately
;
e
v
e
n
s
o
,
th
e
p
er
i
m
eter
s
ar
e
clea
r
,
a
lo
n
g
w
i
th
t
h
e
p
i
x
els
b
e
o
b
s
er
v
ed
th
r
o
u
g
h
t
h
e
in
ter
p
o
lated
p
ix
els.
Ho
w
e
v
er
th
e
N
A
M
lead
s
to
th
e
s
p
ati
al
s
m
o
o
th
n
es
s
,
th
i
s
also
ad
d
itio
n
all
y
clea
r
s
th
e
v
is
ib
il
it
y
f
o
r
d
o
cto
r
s
.
R
ef
er
T
ab
le
2
b
ello
w
f
o
r
ev
al
u
atio
n
o
f
N
A
M
Alg
o
r
it
h
m
p
er
f
o
r
m
an
ce
to
g
et
h
er
w
it
h
ex
is
t
in
g
SVM
/P
SO te
ch
n
iq
u
e.
T
ab
le
2
.
C
o
m
p
ar
is
o
n
o
f
N
A
M
A
l
g
o
r
ith
m
E
f
f
ic
ien
c
y
w
it
h
S
VM
/P
SO M
eth
o
d
.
Im
a
g
e
S
V
M
/
PSO
M
e
t
h
o
d
N
A
M
A
l
g
o
r
i
t
h
m
M
i
n
D
i
f
f
M
i
n
D
i
f
f
M
R
I
a
n
d
N
e
u
r
o
e
n
d
o
sco
p
y
T
e
st
I
n
p
u
t
3
1
.
4
3
0
.
7
1
1
3
6
.
3
2
0
.
8
4
2
E
v
en
t
u
all
y
,
t
h
e
p
ix
el
s
ar
e
r
ec
o
g
n
izab
le
d
esp
ite
t
h
e
f
u
zz
y
in
ter
p
o
lated
f
o
r
m
s
.
Fig
u
r
e
-
4
r
ev
ea
ls
a
co
n
s
eq
u
e
n
ce
o
f
o
u
r
ap
p
r
o
ac
h
is
g
ett
in
g
N
A
M
r
e
g
u
lar
izatio
n
.
T
h
e
n
o
ticea
b
le
s
tr
u
ct
u
r
es
i
n
co
n
j
u
n
ctio
n
w
i
t
h
s
h
ar
p
en
ed
f
ac
ets
p
o
i
n
t
o
u
t
t
h
e
ad
v
a
n
ta
g
e
o
f
o
u
r
tech
n
i
q
u
e
o
f
N
A
M.
I
n
th
i
s
r
esear
c
h
,
w
e
m
ea
s
u
r
e
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
clo
g
g
in
g
f
ilter
i
m
a
g
in
g
p
r
o
to
co
l
(
au
to
-
f
o
cu
s
ap
p
r
o
ac
h
)
,
an
d
th
at
is
r
ec
o
m
m
e
n
d
ed
in
th
i
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
6
,
Dec
em
b
er
2
0
1
7
:
3
6
4
3
–
3
6
5
4
3652
p
ar
ticu
lar
p
ap
er
.
W
e
an
al
y
ze
d
th
e
i
m
ag
e
tak
e
n
b
y
Ne
u
r
o
en
d
o
s
co
p
y
u
n
it
a
n
d
p
atie
n
t
w
a
s
tr
ea
ted
at
M
u
lti
Sp
ec
ialit
y
Ho
s
p
ita
l,
P
u
n
e,
I
n
d
i
a.
6.
DIS
CU
SS
I
O
N
Desp
ite
th
e
f
ac
t
t
h
at
T
u
m
o
r
g
r
o
w
t
h
s
ar
e
m
u
c
h
le
s
s
w
id
es
p
r
ea
d
as
co
m
p
ar
ed
to
tu
m
o
r
p
o
ly
p
lo
id
y
w
o
u
n
d
s
,
r
ip
p
ed
s
k
i
n
les
io
n
s
p
o
s
s
ess
h
i
g
h
er
ca
n
ce
r
o
u
s
p
r
o
s
p
ec
tiv
e.
R
ip
p
ed
w
o
u
n
d
s
ar
e
lik
el
y
,
ev
e
n
s
o
,
to
b
e
m
o
r
e
ch
al
len
g
i
n
g
to
d
iag
n
o
s
e
e
m
p
lo
y
in
g
a
co
n
v
en
t
io
n
al
te
ch
n
iq
u
e
i
n
co
m
p
ar
is
o
n
w
it
h
p
o
ly
p
lo
id
y
g
r
o
w
t
h
s
,
p
ar
ticu
lar
l
y
f
o
r
less
-
e
x
p
er
ien
c
ed
n
eu
r
o
en
d
o
s
co
p
is
ts
.
O
u
r
o
u
tco
m
e
s
d
is
co
v
er
ed
th
a
t
N
AM
o
f
f
er
ed
ex
ce
lle
n
t
v
is
ib
il
it
y
as
co
m
p
ar
ed
to
t
h
at
o
f
SV
M/P
SO
m
eth
o
d
.
A
I
p
r
o
v
id
ed
h
ig
h
-
q
u
ali
t
y
v
i
s
u
al
i
m
ag
e
s
o
f
s
id
e
w
a
y
s
s
ca
tter
in
g
g
r
o
w
t
h
s
n
o
n
-
g
r
a
n
u
l
ar
ty
p
e.
So
m
e
o
th
er
s
cie
n
ti
f
ic
s
tu
d
ies
h
a
v
e
clai
m
ed
th
at
S
VM
is
r
elativ
el
y
a
g
o
o
d
ch
o
ice
f
o
r
d
i
f
f
er
e
n
tial
m
ed
ical
d
iag
n
o
s
i
s
.
E
v
e
n
s
o
,
r
eg
ar
d
less
o
f
w
h
eth
er
P
SO
i
s
h
elp
f
u
l
f
o
r
tu
m
o
r
ca
n
ce
r
o
u
s
g
r
o
w
t
h
p
r
o
g
n
o
s
is
c
o
n
tin
u
es to
b
e
q
u
esti
o
n
ab
le,
w
ith
ex
p
er
ts
co
n
f
ir
m
i
n
g
co
m
b
in
ed
r
esu
lts
[
9
]
.
A
l
s
o
,
SVM
Neu
r
o
e
n
d
o
s
co
p
y
h
av
i
n
g
a
tr
an
s
p
ar
en
t
co
v
er
d
is
co
v
er
ed
m
u
c
h
m
o
r
e
tu
m
o
r
n
e
o
p
lasi
as
as
co
m
p
ar
ed
to
p
er
f
o
r
m
ed
P
SO
w
it
h
n
o
tr
an
s
p
ar
en
t
co
v
er
,
ev
en
th
o
u
g
h
ca
r
r
ied
o
u
t
u
s
i
n
g
le
s
s
-
ex
p
er
ie
n
ce
d
d
o
cto
r
s
[
2
0
]
.
T
h
e
m
o
d
er
n
er
a
o
f
P
SO
h
as
b
ee
n
co
n
s
eq
u
e
n
tl
y
f
o
r
m
u
lated
alo
n
g
w
it
h
th
e
o
b
j
ec
tiv
e
o
f
e
n
h
a
n
ce
d
p
er
f
ec
tio
n
,
d
esp
ite
th
e
f
ac
t
t
h
at
p
o
s
s
ib
le
te
s
ts
a
n
al
y
z
in
g
th
i
s
d
ec
lar
e
h
av
e
n
'
t
b
ee
n
r
ec
e
n
tl
y
r
elea
s
ed
.
I
n
o
u
r
an
al
y
s
is
,
t
h
er
e
s
ee
m
ed
to
b
e
a
s
ig
n
i
f
ica
n
t
d
if
f
er
e
n
ce
in
t
h
e
s
t
an
d
ar
d
o
f
v
is
u
al
i
m
a
g
es
o
f
r
ip
p
ed
ca
n
ce
r
s
a
m
o
n
g
NA
M
an
d
S
VM
/P
SO.
O
u
r
r
e
s
u
lt
s
ap
p
l
y
i
n
g
N
A
M
w
h
ic
h
h
as
a
n
e
w
tec
h
n
iq
u
e
f
o
u
n
d
en
s
u
r
in
g
;
n
o
n
et
h
ele
s
s
,
th
er
e
w
as
clea
r
l
y
n
o
d
i
m
i
n
is
h
ed
f
ield
o
f
v
i
s
io
n
u
til
izin
g
NA
M
i
n
co
m
p
ar
is
o
n
w
it
h
P
SO.
T
h
is
p
ar
ticu
lar
o
u
tco
m
e
p
o
in
ted
o
u
t
s
o
m
e
li
k
eli
h
o
o
d
o
f
NA
M
p
r
e
-
Neu
r
o
en
d
o
s
co
p
y
g
r
o
u
n
d
s
ea
r
ch
.
Ad
d
itio
n
al
r
esear
ch
is
n
ec
es
s
ar
y
to
u
n
co
v
er
th
e
p
r
ac
tical
u
s
e
o
f
t
h
e
f
ield
o
f
v
is
io
n
alo
n
g
w
it
h
d
elec
tab
ilit
y
o
f
r
i
p
p
ed
w
o
u
n
d
s
w
h
ile
u
s
i
n
g
th
e
n
e
w
er
a
o
f
N
A
M
ap
p
r
o
ac
h
.
Ou
r
o
u
tco
m
e
s
d
e
m
o
n
s
tr
ated
th
at
N
A
M
tech
n
iq
u
e
co
u
ld
b
e
b
etter
th
a
n
P
SO N
eu
r
o
en
d
o
s
co
p
y
i
m
a
g
es
f
o
r
th
e
r
ec
o
g
n
itio
n
o
f
r
ip
p
ed
tu
m
o
r
s
r
e
g
io
n
v
is
ib
ili
t
y
.
7.
CO
NCLU
SI
O
N
I
n
s
u
m
m
ar
y
,
o
u
r
d
ata
r
ev
ea
le
d
th
at
N
A
M
is
a
g
o
o
d
ch
o
ice
f
o
r
en
h
an
ci
n
g
i
m
a
g
e
q
u
ali
t
y
t
h
r
o
u
g
h
o
u
t
ca
s
es
o
f
tu
m
o
r
r
ip
p
ed
ca
n
ce
r
s
.
I
n
th
is
p
ar
ticu
lar
p
ap
er
,
w
e
'
v
e
ad
v
is
ed
an
N
A
M
s
e
m
i
-
lo
ca
l
clo
g
g
i
n
g
-
a
la
y
er
-
aliasi
n
g
m
et
h
o
d
u
s
i
n
g
N
A
M
alg
o
r
ith
m
to
o
b
tain
g
o
o
d
q
u
al
it
y
i
m
ag
e
s
w
h
ich
ca
n
b
e
u
ti
li
ze
d
f
o
r
p
r
e
-
s
u
r
g
ica
l
an
d
p
o
s
t
-
s
u
r
g
ical
in
i
tial
i
n
v
es
tig
atio
n
s
.
T
h
is
tech
n
iq
u
e
ca
n
n
o
t
o
n
l
y
ac
h
ie
v
e
t
h
e
e
x
ce
lle
n
t
q
u
alit
y
o
f
i
m
a
g
es
b
u
t
also
le
s
s
e
n
alia
s
i
n
g
ar
ti
f
ac
ts
o
f
i
m
a
g
es
s
i
g
n
i
f
ican
t
l
y
s
o
t
h
at
d
o
cto
r
s
ca
n
v
i
s
u
al
ize
t
u
m
o
r
r
eg
io
n
v
er
y
w
ell
.
As
p
er
o
u
r
i
m
p
le
m
en
ta
tio
n
r
e
s
u
lt
s
a
n
d
p
h
y
s
icia
n
's
o
p
in
io
n
,
NA
M
ca
n
b
e
v
er
y
u
s
e
f
u
l
as
a
n
o
f
f
li
n
e
to
o
l.
A
l
s
o
co
n
ce
r
n
i
n
g
u
p
co
m
i
n
g
w
o
r
k
,
it
is
ap
p
r
o
p
r
iate
u
n
d
er
tak
es
a
tech
n
iq
u
e
o
f
ef
f
ec
ti
v
el
y
f
o
r
ti
f
y
in
g
th
e
co
m
p
u
tatio
n
al
p
er
f
o
r
m
a
n
ce
w
it
h
i
n
o
u
r
al
g
o
r
ith
m
s
o
as
to
en
tail
co
n
s
id
er
ab
l
y
lar
g
er
r
esear
ch
r
eg
io
n
ass
o
ciate
d
w
it
h
a
n
ti
-
alia
s
i
n
g
.
A
n
d
p
r
o
p
er
s
o
f
t
w
ar
e
p
ac
k
a
g
e
A
P
I
m
a
y
b
e
d
es
ig
n
ed
to
ap
p
ly
f
o
r
r
ea
l
ti
m
e
Neu
r
o
en
d
o
s
co
p
y
s
y
s
te
m
d
ev
el
o
p
m
e
n
t.
ACK
NO
WL
E
D
G
E
M
E
NT
S
W
e
ar
e
th
an
k
f
u
l
to
th
e
Dep
a
r
t
m
en
t
o
f
C
o
m
p
u
ter
Scie
n
ce
&
E
n
g
i
n
ee
r
in
g
,
K.
L
.
Un
i
v
er
s
it
y
,
A
.
P
.
I
n
d
ia
f
o
r
p
r
o
v
id
in
g
th
e
r
e
s
ea
r
ch
f
a
s
cilit
ies.,
Vi
s
h
w
a
k
ar
m
a
I
n
s
tit
u
te
o
f
T
ec
h
n
o
lo
g
y
,
P
u
n
e
a
n
d
Si
n
h
g
ad
A
ca
d
e
m
y
o
f
E
n
g
i
n
ee
r
in
g
,
P
u
n
e
f
o
r
p
r
o
v
id
i
n
g
t
h
e
s
u
p
p
o
r
t
an
d
g
u
id
a
n
ce
f
o
r
th
i
s
R
es
ea
r
ch
w
o
r
k
.
W
e
ar
e
th
an
k
f
u
l
to
Dr
.
M.
R
.
Nar
asi
n
g
a
R
ao
,
Dr
.
D.
R
aj
es
w
ar
a
R
a
o
,
an
d
Dr
.
N.
Srin
iv
a
s
f
r
o
m
K.
L
.
U
n
iv
er
s
it
y
,
A
.
P
.
,
I
n
d
ia
f
o
r
th
eir
m
o
ti
v
atio
n
,
g
u
id
a
n
ce
an
d
s
u
p
p
o
r
t
f
o
r
th
i
s
w
o
r
k
.
W
e
ar
e
also
th
an
k
f
u
l
to
Dr
.
K.
P
.
P
atil,
P
r
in
cip
al
an
d
P
r
o
f
.
A
.
N.
Ad
ap
an
a
w
ar
,
HO
D
-
I
T
f
r
o
m
Si
n
h
g
ad
A
ca
d
e
m
y
o
f
E
n
g
in
ee
r
i
n
g
,
P
u
n
e.
W
e
ar
e
th
an
k
f
u
l to
all
th
e
Me
m
b
er
s
o
f
th
e
Ho
s
p
ital
.
RE
F
E
R
E
NC
E
S
[1
]
R
y
a
lat,
M
o
h
a
m
m
a
d
Ha
sh
e
m
,
e
t
a
l
.
,
"
Ev
a
lu
a
ti
o
n
o
f
P
a
rti
c
le
S
w
a
r
m
Op
ti
m
isa
ti
o
n
f
o
r
M
e
d
ica
l
Im
a
g
e
S
e
g
m
e
n
tatio
n
"
,
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
S
y
s
tem
s
S
c
ien
c
e
2
0
1
6
.
S
p
rin
g
e
r
In
t
e
rn
a
ti
o
n
a
l
P
u
b
li
sh
i
n
g
,
v
o
l.
5
3
9
,
p
p
.
6
1
-
7
2
,
No
v
2
0
1
6
.
[2
]
Ch
e
n
,
Ru
e
y
-
M
a
w
,
a
n
d
Ch
u
in
-
M
u
W
a
n
g
,
"
M
RI
Br
a
in
T
issu
e
Clas
si
f
ica
ti
o
n
Us
in
g
Un
su
p
e
rv
ise
d
Op
ti
m
ize
d
Ex
ten
ics
-
Ba
se
d
M
e
th
o
d
s"
,
2
0
1
6
In
ter
n
a
ti
o
n
a
l
S
y
mp
o
siu
m
o
n
Co
mp
u
ter
,
Co
n
s
u
me
r
a
n
d
Co
n
tro
l
(
IS
3
C)
,
IEE
E,
Ju
ly
2
0
1
6
.
[3
]
M
e
h
d
i
H
o
ss
e
in
i,
e
t
a
l
.
,
"
Esti
m
a
ti
o
n
o
f
so
il
m
e
c
h
a
n
ica
l
re
sista
n
c
e
p
a
ra
m
e
ter
b
y
u
sin
g
p
a
rti
c
le
sw
a
r
m
o
p
ti
m
iz
a
ti
o
n
,
g
e
n
e
ti
c
a
lg
o
rit
h
m
a
n
d
m
u
lt
ip
le reg
re
ss
io
n
m
e
th
o
d
s"
,
S
o
il
a
n
d
T
il
l
a
g
e
Res
e
a
rc
h
,
v
o
l.
1
5
7
,
p
p
.
3
2
–
4
2
,
2
0
1
6
.
[4
]
Zh
a
o
,
C
u
ij
ie,
a
n
d
G
u
o
z
h
e
n
W
a
n
g
.
"
A
p
p
li
c
a
ti
o
n
o
f
Ch
a
o
ti
c
P
a
rti
c
le
S
w
a
r
m
Op
ti
m
iza
ti
o
n
in
W
a
v
e
let
Ne
u
ra
l
Ne
tw
o
rk
.
"
T
EL
KOM
NIKA
(
T
e
lec
o
mm
u
n
ic
a
ti
o
n
C
o
mp
u
ti
n
g
El
e
c
tro
n
ics
a
n
d
C
o
n
tr
o
l)
,
1
2
.
4
,
(2
0
1
4
):
9
9
7
-
1
0
0
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.