I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
3
7
,
No
.
3
,
Ma
r
ch
20
2
5
,
p
p
.
1
8
0
4
~
1
814
I
SS
N:
2
502
-
4
7
52
,
DOI
: 1
0
.
1
1
5
9
1
/ijee
cs
.v
3
7
.
i
3
.
pp
1
8
0
4
-
1
8
1
4
1804
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs
.
ia
esco
r
e.
co
m
Lev
erag
ing
ima
g
e
f
usio
n and
trans
f
er learning
for en
ha
nced
tumo
r diag
no
sis
Nirm
a
la
j
y
o
t
hi Na
riset
t
y
1
,
K
un
da
Su
re
s
h B
a
b
u
2
,
Sn
e
ha
B
a
na
la
3
,
Aka
s
h Re
dd
y
K
a
nd
i
3
,
Sree
j
a
Nuk
a
ra
pu
3
,
J
a
g
rut
hi M
ek
a
la
3
1
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
K
o
n
e
r
u
L
a
k
sh
m
a
i
a
h
E
d
u
c
a
t
i
o
n
F
o
u
n
d
a
t
i
o
n
,
H
y
d
e
r
a
b
a
d
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
N
a
r
a
s
a
r
a
o
p
e
t
a
E
n
g
i
n
e
e
r
i
n
g
C
o
l
l
e
g
e
(
A
u
t
o
n
o
m
o
u
s)
,
N
a
r
a
sar
a
o
p
e
t
a
,
I
n
d
i
a
3
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
-
A
I
M
L
a
n
d
I
o
T,
V
a
l
l
u
r
u
p
a
l
l
i
N
a
g
e
sw
a
r
a
R
a
o
V
i
g
n
a
n
a
J
y
o
t
h
i
I
n
st
i
t
u
t
e
o
f
En
g
i
n
e
e
r
i
n
g
a
n
d
Te
c
h
n
o
l
o
g
y
,
H
y
d
e
r
a
b
a
d
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
May
16
,
2
0
2
4
R
ev
is
ed
Sep
26
,
2
0
2
4
Acc
ep
ted
Oct
7
,
2
0
2
4
In
In
d
ia,
t
h
e
n
u
m
b
e
r
o
f
b
ra
in
t
u
m
o
r
c
a
se
s
a
re
in
c
re
a
sin
g
ra
p
i
d
ly
.
Co
m
p
a
re
d
to
th
e
c
u
rre
n
t
trea
tme
n
t
a
p
p
ro
a
c
h
e
s th
e
re
is a n
e
e
d
fo
r
e
fficie
n
t
a
n
d
a
d
v
a
n
c
e
d
d
iag
n
o
stic
a
p
p
ro
a
c
h
e
s.
Do
c
t
o
r
p
rima
ril
y
u
se
m
a
g
n
e
ti
c
re
so
n
a
n
c
e
ima
g
in
g
(M
RI)
a
n
d
c
o
m
p
u
te
d
to
m
o
g
ra
p
h
y
(CT)
sc
a
n
s
fo
r
d
iag
n
o
sis,
e
a
c
h
with
it
s
o
wn
se
t
o
f
a
d
v
a
n
tag
e
s
a
n
d
li
m
it
a
ti
o
n
s.
Ac
c
u
ra
te
d
iag
n
o
sis
o
f
b
ra
in
tu
m
o
r
s
re
q
u
ires
d
e
tailed
in
f
o
rm
a
ti
o
n
a
b
o
u
t
th
e
t
u
m
o
r
’
s
t
y
p
e
,
siz
e
,
lo
c
a
ti
o
n
,
a
n
d
p
ro
li
fe
ra
ti
o
n
ra
te.
B
u
t
a
ll
t
h
is
in
f
o
rm
a
ti
o
n
m
u
st
b
e
e
stim
a
ted
a
c
c
u
ra
tely
a
n
d
p
re
c
ise
ly
.
Ev
e
n
t
h
o
u
g
h
M
RI
a
n
d
CT
p
ro
v
i
d
e
a
n
a
to
m
ica
l
i
n
fo
rm
a
ti
o
n
re
g
a
rd
in
g
tu
m
o
rs,
th
e
y
m
a
y
n
o
t
a
lwa
y
s
c
o
rre
c
tl
y
c
las
sify
th
e
tu
m
o
r
h
e
n
c
e
,
o
ften
re
q
u
iri
n
g
th
e
a
d
d
i
ti
o
n
a
l
b
io
p
sy
f
o
r
m
o
re
d
e
tailed
a
n
a
ly
sis.
T
h
is
p
a
p
e
r
a
ims
to
d
e
m
o
n
stra
te
th
e
fu
si
o
n
o
f
M
RI
a
n
d
CT
sc
a
n
s
b
y
g
e
n
e
ra
ti
n
g
a
fu
se
d
ima
g
e
u
sin
g
tra
n
sfe
r
lea
rn
in
g
wi
t
h
c
o
n
v
o
lu
ti
o
n
n
e
u
ra
l
n
e
two
rk
s
(
CNN
)
th
a
t
p
o
ss
e
ss
e
s
a
ll
th
e
im
p
o
rta
n
t
i
n
fo
r
m
a
ti
o
n
fr
o
m
b
o
t
h
m
o
d
a
li
ti
e
s
y
ield
s
b
e
tt
e
r
re
su
lt
s
th
a
n
u
si
n
g
M
RI
a
n
d
CT
sc
a
n
s
a
lo
n
e
.
Th
is
f
u
se
d
sc
a
n
h
a
s
th
e
p
o
te
n
ti
a
l
to
h
ig
h
li
g
h
t
su
b
tl
e
d
e
tails
t
h
a
t
m
a
y
b
e
o
v
e
rlo
o
k
e
d
o
n
in
d
iv
i
d
u
a
l
s
c
a
n
s.
Th
is
p
ro
p
o
se
d
sy
ste
m
re
tain
s
a
ll
t
h
e
b
e
st
p
o
ss
i
b
le
f
u
n
c
ti
o
n
a
li
t
ies
fo
r
m
e
d
ica
l
a
p
p
ro
a
c
h
e
s
to
b
ra
i
n
tu
m
o
r
d
e
tec
ti
o
n
.
It
p
a
v
e
s
th
e
wa
y
to
o
ffe
r
a
m
u
c
h
m
o
re
imp
a
c
tfu
l
h
e
a
lt
h
c
a
re
a
p
p
ro
a
c
h
to
p
a
ti
e
n
ts
b
y
d
e
li
v
e
rin
g
a
c
c
u
ra
te
a
n
d
re
li
a
b
le
sta
ti
stics
,
e
n
h
a
n
c
i
n
g
th
e
d
iag
n
o
stic
a
c
c
u
ra
c
y
(9
6
.
4
3
%
)
p
re
c
e
d
e
n
t
t
o
a
n
y
sin
g
le m
o
d
a
li
ty
.
K
ey
w
o
r
d
s
:
B
r
ain
tu
m
o
r
d
etec
tio
n
C
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
Fu
s
io
n
I
m
ag
e
class
if
icatio
n
T
r
an
s
f
er
lear
n
i
n
g
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
:
Nir
m
alajy
o
th
i N
ar
is
etty
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
in
ee
r
in
g
,
K
o
n
er
u
L
ak
s
h
m
aiah
E
d
u
ca
tio
n
Fo
u
n
d
atio
n
Hy
d
er
ab
ad
,
T
elan
g
a
n
a,
I
n
d
ia
E
m
ail: n
ir
m
ala.
n
ar
is
etty
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
B
r
ain
tu
m
o
r
s
ar
e
ca
u
s
ed
b
y
th
e
p
r
o
life
r
atio
n
o
f
ce
lls
ab
n
o
r
m
ally
with
in
th
e
b
r
ain
wh
ich
d
is
tu
r
b
s
th
e
b
r
ain
’
s
wo
r
k
in
g
f
u
n
ctio
n
alities
an
d
ca
u
s
es
s
ev
er
e
m
e
d
ical
i
lln
ess
b
y
p
r
ess
u
r
izin
g
in
s
id
e
th
e
s
k
u
ll
[
1
]
.
T
h
is
p
h
en
o
m
en
o
n
if
h
ap
p
en
in
g
s
h
o
u
ld
b
e
d
etec
ted
at
ea
r
ly
s
tag
es to
co
n
tr
o
l
f
u
r
th
e
r
co
m
p
licatio
n
s
an
d
n
eu
r
o
lo
g
ica
l
s
y
m
p
to
m
s
.
Fo
r
th
is
p
u
r
p
o
s
e,
th
er
e
ar
e
two
m
e
d
ical
m
o
d
alities
n
am
ely
m
ag
n
etic
r
eso
n
an
ce
im
ag
in
g
(
MRI
)
an
d
co
m
p
u
ted
t
o
m
o
g
r
ap
h
y
(
CT
)
s
ca
n
s
.
T
h
e
m
en
tio
n
ed
s
ca
n
s
ar
e
v
er
y
r
eliab
le
an
d
well
-
es
tab
lis
h
ed
p
o
wer
f
u
l
d
iag
n
o
s
tic
to
o
ls
.
As
th
e
r
eliab
i
lity
is
r
elativ
e,
th
is
p
ap
er
p
r
o
p
o
s
es
a
m
o
r
e
p
r
o
m
is
in
g
s
o
lu
tio
n
wh
ich
is
a
f
u
s
io
n
o
f
MRI
an
d
C
T
im
ag
es.
T
h
e
m
en
tio
n
ed
s
ca
n
s
ar
e
v
er
y
r
eliab
le
an
d
well
-
estab
li
s
h
ed
p
o
wer
f
u
l
d
iag
n
o
s
tic
to
o
ls
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
Leve
r
a
g
in
g
ima
g
e
fu
s
io
n
a
n
d
tr
a
n
s
fer lea
r
n
in
g
fo
r
en
h
a
n
ce
d
tu
mo
r
d
ia
g
n
o
s
is
(
N
ir
ma
la
jyo
t
h
i Na
r
is
etty
)
1805
T
h
e
p
r
im
a
r
y
a
d
v
an
ta
g
es
o
f
a
C
T
-
s
ca
n
n
er
ar
e
to
ac
q
u
i
r
e
p
h
y
s
ical
in
f
o
r
m
atio
n
,
s
u
ch
as
s
ize
,
co
n
d
itio
n
an
d
h
o
m
o
g
e
n
eities
[
2
]
,
[
3
]
.
D
esp
ite
co
n
tin
u
o
u
s
ad
v
an
ce
m
e
n
ts
in
h
ea
lth
ca
r
e
tech
n
o
l
o
g
y
,
m
ac
h
in
e
lear
n
in
g
h
as
d
em
o
n
s
tr
ated
cu
ttin
g
-
ed
g
e
p
o
ten
tial
in
r
esh
ap
in
g
p
atien
t
ca
r
e
p
r
o
ce
d
u
r
es
an
d
f
av
o
r
ab
le
o
u
tco
m
es.
Ma
ch
in
e
lear
n
in
g
o
f
f
er
s
p
r
o
m
is
in
g
ap
p
r
o
ac
h
es,
as
a
r
esu
lt,
h
ea
lth
ca
r
e
p
r
o
f
ess
io
n
als
m
ay
u
s
e
th
is
k
n
o
wled
g
e
to
aid
a
n
d
en
h
an
ce
th
eir
wo
r
k
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
tem
is
im
p
lem
e
n
ted
u
s
in
g
m
ac
h
in
e
lear
n
i
n
g
m
o
d
els
an
d
h
e
n
ce
tr
u
s
ted
with
th
e
m
o
s
t
ac
cu
r
ate
r
esu
lts
.
MRI
s
u
r
p
ass
es
C
T
in
v
is
u
alizin
g
s
o
f
t
tis
s
u
es
an
d
h
en
ce
b
o
o
s
tin
g
ac
cu
r
ac
y
[
4
]
.
T
h
e
b
r
a
i
n
s
c
a
n
m
u
s
t
b
e
e
x
a
m
in
e
d
v
e
r
y
c
a
r
e
f
u
l
l
y
s
i
n
c
e
it
is
co
m
p
o
s
e
d
o
f
e
x
t
r
e
m
e
l
y
s
o
f
t
a
n
d
s
m
o
o
t
h
t
is
s
u
e
s
[
5
]
.
So
m
etim
es
it
i
s
d
if
f
icu
lt
to
d
r
ag
o
n
ize
th
e
tu
m
o
r
with
MRI
wh
en
th
e
s
ize
o
f
th
e
tu
m
o
r
is
v
er
y
s
m
all.
T
h
at
is
d
u
r
in
g
t
h
e
in
itial
s
tag
es
’
d
etec
tio
n
o
f
th
e
lo
ca
tio
n
o
f
th
e
tu
m
o
r
b
ec
o
m
es
a
r
is
k
y
task
b
ec
au
s
e
d
if
f
er
en
tiatin
g
f
r
o
m
s
u
r
r
o
u
n
d
in
g
tis
s
u
es
is
d
if
f
icu
lt.
W
h
er
ea
s
C
T
s
ca
n
s
h
i
g
h
lig
h
t
clea
r
cu
t
b
o
n
e
d
etails
wh
ich
p
u
lls
th
eir
ab
ilit
y
to
d
is
tin
g
u
is
h
s
o
f
t tis
s
u
es d
o
wn
.
Alth
o
u
g
h
MRI
an
d
C
T
h
av
e
th
eir
s
p
ec
ial
im
p
o
r
tan
ce
,
m
ask
in
g
s
o
m
e
s
u
b
tle
ab
n
o
r
m
alities
in
ca
s
es
en
co
u
n
ter
e
d
m
ay
lead
to
lim
itatio
n
s
in
d
etec
tin
g
tu
m
o
r
s
wh
ich
in
tu
r
n
m
ak
es
it
n
ec
ess
ar
y
to
d
ev
elo
p
a
m
u
c
h
m
o
r
e
p
r
o
m
is
in
g
s
o
lu
tio
n
.
Fo
r
th
e
f
o
r
e
g
o
in
g
ex
p
lan
atio
n
,
ev
a
lu
atin
g
th
em
u
s
in
g
a
s
in
g
le
im
ag
in
g
m
et
h
o
d
m
a
y
b
e
in
ad
eq
u
ate.
Diag
n
o
s
in
g
b
r
a
in
tu
m
o
r
s
ca
n
b
e
co
m
p
lex
f
o
r
r
ad
io
lo
g
is
ts
ev
en
af
ter
r
ev
iewin
g
im
ag
es.
T
h
is
is
b
ec
au
s
e
d
if
f
e
r
en
t
im
a
g
in
g
t
ec
h
n
iq
u
es,
lik
e
C
T
s
ca
n
s
a
n
d
MRIs
s
h
o
wn
in
Fig
u
r
e
s
1
a
n
d
2
,
p
r
o
v
id
e
co
m
p
lem
en
tar
y
in
f
o
r
m
atio
n
.
E
ac
h
m
o
d
ality
ex
ce
ls
at
r
ev
ea
lin
g
s
p
ec
if
ic
d
etails
ab
o
u
t
th
e
tu
m
o
r
a
n
d
s
u
r
r
o
u
n
d
in
g
ce
lls
[
6
]
.
T
o
o
v
er
co
m
e
th
e
a
b
o
v
e
r
esear
ch
g
ap
s
,
th
is
s
tu
d
y
in
v
esti
g
ates
th
e
MRI
an
d
C
T
f
u
s
io
n
m
er
g
in
g
f
o
r
th
e
s
tr
en
g
th
s
o
f
b
o
th
m
o
d
alities
an
d
g
en
er
at
in
g
a
m
o
r
e
co
m
p
r
eh
en
s
iv
e
i
m
ag
e
o
f
b
r
ain
s
ca
n
s
u
r
p
ass
in
g
b
o
th
th
e
in
d
iv
id
u
al
s
ca
n
s
.
Fig
u
r
e
1
.
C
T
s
ca
n
o
f
b
r
ain
tu
m
o
r
Fig
u
r
e
2
.
MRI
s
ca
n
b
r
ain
tu
m
o
r
As
an
alter
n
ativ
e,
f
u
s
io
n
tec
h
n
iq
u
es
s
u
ch
as
d
is
cr
ete
wav
elet
tr
an
s
f
o
r
m
s
(
DW
T
s
)
an
d
em
p
ir
ical
m
o
d
e
d
ec
o
m
p
o
s
itio
n
(
E
MD
s
)
ar
e
o
f
ten
u
s
ed
as
th
ey
en
s
u
r
e
th
e
lo
s
s
d
u
r
in
g
th
e
f
u
s
io
n
p
r
o
ce
s
s
is
m
in
im
al
to
ex
tr
ac
t
s
alien
t
f
ea
tu
r
es
ac
cu
r
ately
[
7
]
.
B
r
ain
tu
m
o
r
s
ar
e
p
r
im
ar
ily
class
if
ied
as
b
en
i
g
n
an
d
m
alig
n
a
n
t
in
d
icatin
g
h
ea
lth
y
a
n
d
u
n
h
ea
lt
h
y
r
esp
ec
tiv
ely
[
5
]
.
C
o
m
p
a
r
ativ
ely
t
o
tr
ad
itio
n
al
m
et
h
o
d
s
,
co
n
v
o
lu
tio
n
al
n
eu
r
a
l
n
etwo
r
k
s
(
C
NNs
)
h
av
e
d
e
m
o
n
s
tr
ated
s
u
p
er
io
r
s
eg
m
e
n
tatio
n
,
class
if
icatio
n
,
an
d
f
ea
tu
r
e
ex
tr
ac
tio
n
ca
p
ab
ilit
ies.
T
h
e
p
r
o
p
o
s
ed
wo
r
k
ad
o
p
ted
a
C
NN
b
ased
ar
ch
itectu
r
e
to
ca
teg
o
r
ize
b
r
ain
tu
m
o
r
s
as
C
NN
ca
n
ex
tr
ac
t
s
tr
u
c
tu
r
e
d
a
n
d
im
p
o
r
tan
t
f
ea
tu
r
es
f
r
o
m
m
ed
ical
im
a
g
es.
T
h
is
class
if
ier
ca
teg
o
r
ize
s
b
r
ain
tu
m
o
r
s
in
t
o
f
o
u
r
ty
p
es
n
am
ely
g
li
o
m
a,
m
en
in
g
io
m
a
,
p
itu
itar
y
,
an
d
n
o
tu
m
o
r
.
Fro
m
th
e
p
r
ec
ed
in
g
in
f
o
r
m
atio
n
,
f
u
s
ed
im
ag
es p
r
o
b
a
b
ly
d
eliv
e
r
m
o
r
e
ef
f
ec
tiv
e
r
esu
lts
wh
en
co
m
p
ar
e
d
to
MRI
alo
n
e.
An
im
p
r
o
v
is
ed
m
eth
o
d
f
o
r
p
r
ed
ictin
g
th
e
b
r
ain
tu
m
o
r
o
v
er
a
r
o
b
u
s
t
u
s
er
in
ter
f
ac
e
(
UI
)
is
th
e
k
ey
o
b
jectiv
e
o
f
t
h
is
r
esear
ch
wo
r
k
.
T
h
e
UI
is
d
esig
n
e
d
in
s
u
ch
a
way
th
at
an
y
b
o
d
y
ca
n
e
f
f
o
r
tles
s
ly
o
p
er
ate
it.
T
h
e
co
m
p
lete
wo
r
k
o
f
th
e
a
r
ticle
i
s
o
r
g
an
ize
d
in
t
o
s
ev
er
al
s
ec
tio
n
s
.
T
h
e
v
ar
io
u
s
s
tu
d
ies
r
elate
d
to
th
e
b
r
ai
n
tu
m
o
r
ar
e
co
v
er
e
d
in
s
ec
tio
n
2
.
T
h
e
p
r
o
p
o
s
ed
m
ater
ial
a
n
d
m
eth
o
d
ar
e
ex
p
lain
ed
in
s
ec
tio
n
3
.
Sectio
n
4
p
r
esen
ts
th
e
r
esu
lts
in
g
r
ea
t
d
ep
th
,
wh
ile
s
ec
tio
n
5
d
elv
es
i
n
to
m
o
r
e
d
etail.
Fin
ally
,
s
ec
ti
o
n
6
co
n
clu
d
es
b
y
d
r
awin
g
co
n
clu
s
io
n
s
.
2.
L
I
T
E
R
AT
U
R
RE
VI
E
W
Sev
er
al
s
tu
d
ies
h
av
e
co
n
ce
n
t
r
ated
o
n
d
ev
elo
p
i
n
g
an
d
en
h
an
cin
g
th
e
p
e
r
f
o
r
m
an
ce
o
f
b
r
ain
tu
m
o
r
d
etec
tio
n
an
d
class
if
icatio
n
.
C
h
an
d
r
ash
ek
ar
an
d
Sre
ed
e
v
i
[
6
]
p
r
o
p
o
s
ed
a
s
tu
d
y
u
s
in
g
n
o
n
s
u
b
s
am
p
le
d
co
n
to
u
r
let
tr
a
n
s
f
o
r
m
(
NSC
T
)
an
d
NSC
T
-
SR
tech
n
iq
u
es,
s
p
ar
s
e
r
ep
r
esen
tatio
n
,
a
n
d
n
eu
r
al
n
etwo
r
k
s
f
o
r
en
h
an
ce
d
im
ag
e
f
u
s
io
n
.
R
an
i
an
d
L
alith
ak
u
m
ar
i
[
8
]
p
r
esen
ted
a
h
y
b
r
id
f
u
s
io
n
m
o
d
el
th
at
in
teg
r
ated
MRI
an
d
C
T
im
ag
es
f
o
r
d
etec
tin
g
b
r
ai
n
tu
m
o
r
s
in
clu
d
es
EMD
,
DW
T
,
n
eu
r
al
n
etwo
r
k
s
,
f
o
r
ef
f
ec
t
iv
e
m
ed
ical
im
ag
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
7
,
No
.
3
,
Ma
r
ch
20
2
5
:
1
804
-
1
8
1
4
1806
f
u
s
io
n
f
r
ee
f
r
o
m
d
is
to
r
tio
n
.
So
lan
k
i
et
a
l.
[
3
]
in
v
esti
g
at
ed
th
e
ef
f
ec
tiv
e
n
ess
o
f
d
if
f
e
r
en
t
s
eg
m
en
tatio
n
alg
o
r
ith
m
s
f
o
r
tu
m
o
r
id
en
tifi
ca
tio
n
.
Am
in
et
a
l.
[
9
]
h
a
v
e
d
ev
elo
p
ed
a
n
ew
a
p
p
r
o
ac
h
in
class
if
y
in
g
b
r
ain
tu
m
o
r
s
in
MRI
s
ca
n
s
.
T
h
is
h
y
b
r
id
m
o
d
el,
ca
lled
C
NN
-
k
-
n
e
ar
est
n
eig
h
b
o
r
s
(
K
NN
)
,
co
m
b
in
es
th
e
s
tr
en
g
th
s
o
f
C
NNs
an
d
th
e
K
NN
a
lg
o
r
ith
m
.
A
n
o
v
el
m
eth
o
d
o
f
d
et
ec
tin
g
b
r
ain
tu
m
o
r
s
au
to
m
ati
ca
lly
u
s
in
g
MRI
is
p
r
esen
ted
b
y
Sap
r
a
et
a
l.
[
1
0
]
.
W
ater
s
h
ed
s
eg
m
en
tatio
n
b
r
a
in
tu
m
o
r
d
etec
tio
n
,
in
t
h
e
d
o
m
ain
o
f
b
i
o
m
ed
ical
im
ag
e
p
r
o
ce
s
s
in
g
d
ed
icate
d
t
o
b
r
ai
n
tu
m
o
r
d
etec
tio
n
,
t
h
e
r
esear
ch
h
as
b
ee
n
co
n
d
u
cte
d
b
y
Dh
ag
e
et
a
l.
[
1
1
]
.
T
h
e
s
tu
d
y
b
y
Nan
m
a
r
an
et
a
l
.
[
7
]
,
titl
ed
“
B
r
ain
t
u
m
o
r
clas
s
if
icatio
n
an
d
d
etec
tio
n
u
s
in
g
h
y
b
r
i
d
d
ee
p
tu
m
o
r
n
etwo
r
k
,
”
ad
d
r
ess
es
th
e
cr
u
cial
is
s
u
e
o
f
ac
cu
r
ately
class
if
y
in
g
an
d
id
en
tif
y
in
g
b
r
ain
tu
m
o
r
s
.
A
d
is
tin
g
u
is
h
in
g
ap
p
r
o
ac
h
o
f
b
r
ain
t
u
m
o
r
d
ete
ctio
n
also
class
if
icatio
n
u
s
in
g
MRI
J
av
er
ia
Am
in
[
1
2
]
p
r
o
p
o
s
ed
an
au
to
m
ated
m
eth
o
d
u
s
in
g
MRI
th
at
e
f
f
ec
tiv
ely
d
is
tin
g
u
is
h
es
b
e
twe
en
ca
n
ce
r
o
u
s
an
d
n
o
n
-
ca
n
c
er
o
u
s
b
r
ai
n
s
ca
n
s
.
Nan
d
ee
s
h
[
1
3
]
in
tr
o
d
u
ce
d
a
p
r
o
m
is
in
g
s
o
lu
tio
n
:
a
m
u
ltimo
d
al
im
ag
e
f
u
s
io
n
(
MM
I
F)
alg
o
r
ith
m
th
at
c
o
m
b
in
es
th
e
s
tr
en
g
th
s
o
f
d
if
f
e
r
en
t
im
a
g
in
g
m
o
d
alities
to
cr
ea
te
a
r
ich
er
tap
estry
o
f
in
f
o
r
m
atio
n
,
u
ltima
tely
en
h
a
n
cin
g
b
r
ain
t
u
m
o
r
d
etec
tio
n
ac
cu
r
ac
y
.
Z
aid
i
et
a
l
.
[
1
4
]
ex
p
lo
r
e
th
e
clin
ical
v
alu
e
o
f
c
o
m
b
in
in
g
PET
,
C
T
,
an
d
MRI
s
ca
n
s
u
s
in
g
f
u
s
io
n
im
ag
in
g
.
T
h
ey
h
ig
h
lig
h
t
h
o
w
th
is
ap
p
r
o
a
ch
p
lay
s
a
cr
itical
r
o
le
in
d
iag
n
o
s
in
g
,
s
tag
in
g
,
an
d
m
o
n
ito
r
in
g
tr
ea
tm
en
t
r
esp
o
n
s
es
f
o
r
p
atien
ts
with
m
alig
n
an
t
d
is
ea
s
es.
Ab
d
allatif
et
a
l.
[
1
5
]
u
tili
ze
wav
elet
tr
an
s
f
o
r
m
f
o
r
its
ab
ilit
y
t
o
d
e
co
m
p
o
s
e
im
ag
es
i
n
to
d
if
f
er
en
t
f
r
eq
u
e
n
cy
b
an
d
s
,
allo
win
g
t
h
em
to
s
elec
tiv
el
y
m
er
g
e
i
n
f
o
r
m
atio
n
f
r
o
m
ea
c
h
m
o
d
ality
w
h
ile
p
r
eser
v
in
g
ess
en
tial
f
ea
tu
r
es.
K
u
m
ar
[
1
6
]
liter
atu
r
e
s
u
r
v
e
y
ex
p
lo
r
es
a
co
m
p
r
e
h
en
s
iv
e
an
aly
s
is
o
f
s
eg
m
en
tatio
n
tech
n
iq
u
es
f
o
r
b
r
ain
tu
m
o
r
MRI
an
aly
s
is
,
o
v
er
lap
p
in
g
b
o
th
th
e
tr
a
d
itio
n
al
m
eth
o
d
o
lo
g
ies
lik
e
Ots
u
’
s
th
r
esh
o
l
d
in
g
,
wate
r
s
h
ed
t
r
an
s
f
o
r
m
ati
o
n
,
an
d
K
-
m
ea
n
s
c
lu
s
ter
in
g
,
as
well
as
a
d
v
a
n
c
ed
m
eth
o
d
s
s
u
ch
as
HAAR
DW
T
an
d
C
NNs.
L
o
g
eswar
i
an
d
Kar
n
a
n
[
1
7
]
p
r
esen
ted
an
e
n
h
an
ce
d
ap
p
r
o
a
ch
o
f
b
r
ain
tu
m
o
r
d
etec
tio
n
u
s
in
g
s
eg
m
en
tatio
n
u
s
in
g
s
o
f
t c
o
m
p
u
tin
g
.
T
h
e
s
tu
d
y
[
1
8
]
in
tr
o
d
u
ce
s
a
wo
r
k
in
g
m
et
h
o
d
o
lo
g
y
co
m
b
i
n
in
g
C
NNs
with
th
e
“
Ad
ap
tiv
e
d
y
n
a
m
ic
s
in
e
-
co
s
in
e
f
itn
ess
g
r
ey
wo
lf
o
p
tim
iza
tio
n
(
ADSC
FGW
O)
”
alg
o
r
ith
m
f
o
r
ac
cu
r
ately
d
et
ec
tin
g
b
r
ai
n
tu
m
o
r
an
d
class
if
icatio
n
.
I
n
‘
C
T
s
ca
n
b
ased
b
r
ain
tu
m
o
r
r
ec
o
g
n
itio
n
an
d
e
x
tr
ac
tio
n
u
s
in
g
p
r
ewitt
an
d
m
o
r
p
h
o
lo
g
ical
d
ilat
io
n
’
p
r
o
p
o
s
ed
b
y
So
n
i
an
d
R
ai
[
1
9
]
,
d
ev
e
l
o
p
ed
a
s
y
s
tem
m
ain
ly
co
n
s
is
ts
o
f
6
s
tep
s
n
am
ely
,
d
ata
wr
an
g
lin
g
u
s
in
g
a
d
ap
tiv
e
h
is
to
g
r
am
eq
u
aliza
tio
n
,
th
r
esh
o
l
d
in
g
,
p
r
ewitt
ed
g
e
d
is
co
v
er
y
f
i
lter
,
m
o
r
p
h
o
lo
g
ical
d
ilatio
n
p
r
o
ce
s
s
,
f
lo
o
d
f
illi
n
g
,
R
OI
ex
tr
ac
tio
n
.
An
o
th
er
s
tu
d
y
o
n
B
r
ain
T
u
m
o
r
p
r
o
p
o
s
ed
b
y
Kam
il
in
[
2
0
]
,
in
tr
o
d
u
ce
s
a
m
ac
h
in
e
-
lear
n
in
g
ap
p
r
o
ac
h
f
o
r
c
alcu
latin
g
t
u
m
o
r
s
ize
with
h
i
g
h
ac
c
u
r
a
cy
,
o
u
t
p
er
f
o
r
m
in
g
ex
is
tin
g
m
eth
o
d
s
u
s
ed
in
C
T
s
ca
n
s
.
R
ah
m
an
et
a
l.
[
2
1
]
p
r
o
p
o
s
ed
a
n
o
v
el
I
o
T
-
b
ased
s
y
s
tem
in
b
r
ain
tu
m
o
r
d
etec
tio
n
u
s
in
g
p
h
y
s
ical
ac
tiv
ities
o
n
MRI
s
ca
n
s
.
T
h
is
m
eth
o
d
is
co
s
t
-
ef
f
ec
tiv
e
an
d
a
v
o
id
s
th
e
c
h
allen
g
es
ass
o
ciate
d
with
im
ag
e
s
eg
m
e
n
tatio
n
in
MRI
-
b
ased
tec
h
n
iq
u
es.
Ab
b
o
o
d
et
a
l.
[
2
2
]
p
r
o
p
o
s
ed
an
au
to
m
ate
d
b
r
ain
tu
m
o
r
class
if
icatio
n
s
y
s
tem
u
s
in
g
s
h
allo
w
d
ee
p
lear
n
in
g
m
o
d
els.
Fo
u
r
m
o
d
els,
Alex
Net,
VGG1
6
,
Go
o
g
leNe
t,
an
d
R
esNet5
0
,
a
r
e
ev
alu
ated
.
E
l
Ham
d
ao
u
i
et
a
l.
p
r
o
p
o
s
es
a
n
ew
i
n
tellig
en
t
s
y
s
tem
f
o
r
b
r
ai
n
tu
m
o
r
d
etec
tio
n
an
d
class
if
icatio
n
u
s
in
g
R
MI
im
ag
es
in
[
2
3
]
.
I
t
tack
les
th
e
c
h
allen
g
e
o
f
lim
ited
tr
ain
in
g
d
ata
b
y
em
p
l
o
y
in
g
d
ee
p
tr
a
n
s
f
er
le
ar
n
in
g
f
r
o
m
p
r
e
-
tr
ai
n
ed
C
NNs
o
n
a
lar
g
e
im
a
g
e
d
ataset.
A
n
im
p
r
o
v
ed
m
et
h
o
d
th
an
ar
tific
ial
n
eu
r
al
n
etwo
r
k
(
ANN)
an
d
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es
(
SVM
)
tech
n
iq
u
es
is
an
ticip
ated
in
[
2
4
]
wh
ich
in
clu
d
es
f
o
u
r
d
is
tin
ctiv
e
m
o
d
els,
k
-
m
ea
n
s
clu
s
ter
in
g
s
eg
m
en
tatio
n
,
h
ig
h
c
o
n
ce
n
tr
at
io
n
s
lu
r
r
y
d
is
p
o
s
al
(
HC
SD)
m
eth
o
d
,
ex
t
r
ac
tio
n
o
f
f
ea
tu
r
es,
a
n
d
KNN
class
if
ier
.
A
h
y
b
r
id
f
ea
tu
r
e
ex
tr
ac
tio
n
m
eth
o
d
was
u
s
ed
b
ased
o
n
DW
T
an
d
p
r
in
cip
al
c
o
m
p
o
n
en
t
an
aly
s
is
to
id
en
tif
y
th
e
b
r
ain
tu
m
o
r
[
2
5
]
.
B
r
ain
tu
m
o
r
s
p
o
s
e
a
s
ig
n
if
ican
t
ch
allen
g
e
in
h
ea
lth
ca
r
e
.
Acc
u
r
ate
an
d
tim
ely
d
etec
tio
n
is
cr
itical
f
o
r
ef
f
ec
tiv
e
tr
ea
tm
en
t,
y
et
b
o
t
h
MRI
as
well
as
C
T
s
ca
n
s
,
th
e
cu
r
r
en
t
d
iag
n
o
s
tic
h
av
e
lim
itatio
n
s
.
T
h
e
ex
is
tin
g
r
esear
ch
o
n
b
r
ain
t
u
m
o
r
d
ete
ctio
n
with
im
a
g
e
f
u
s
io
n
a
n
d
n
eu
r
al
n
etwo
r
k
s
,
th
er
e
ar
e
s
till
s
o
m
e
g
ap
s
a
n
d
o
p
p
o
r
tu
n
ities
f
o
r
f
u
r
th
er
ex
p
lo
r
atio
n
o
f
s
tr
u
ct
u
r
al
s
im
ilar
ity
i
n
d
ex
(
SS
I
M
)
,
p
ea
k
s
ig
n
al
-
to
-
n
o
is
e
r
atio
(
PS
NR
),
an
d
m
ea
n
s
q
u
ar
e
e
r
r
o
r
(
MSE
)
v
alu
es
alo
n
g
with
ac
cu
r
ac
y
.
Ma
n
y
s
tu
d
ies
m
en
tio
n
co
n
ce
r
n
s
ab
o
u
t
h
ig
h
C
PU
tim
e
co
n
s
u
m
p
tio
n
,
esp
ec
ially
in
r
ea
l
-
wo
r
ld
clin
ical
s
ettin
g
s
wh
er
e
q
u
ick
d
iag
n
o
s
is
is
cr
u
cial.
T
h
er
e
’
s
a
n
ee
d
f
o
r
r
esear
ch
f
o
cu
s
ed
o
n
o
p
tim
izin
g
alg
o
r
ith
m
s
to
r
ed
u
ce
ex
ec
u
tio
n
tim
e
a
n
d
r
eso
u
r
ce
r
eq
u
ir
em
en
ts
with
o
u
t
co
m
p
r
o
m
is
in
g
ac
cu
r
ac
y
.
I
n
s
p
ite
o
f
th
e
p
r
o
m
is
in
g
r
esu
lts
o
f
ex
is
tin
g
r
esear
ch
,
th
er
e
is
s
till
a
lack
o
f
clin
ical
v
alid
atio
n
o
f
th
ese
m
o
d
els.
MRI
-
C
T
f
u
s
io
n
o
f
f
er
s
a
p
r
o
m
is
in
g
leap
f
o
r
war
d
i
n
b
r
ai
n
tu
m
o
r
d
etec
tio
n
.
T
h
e
f
u
s
io
n
o
f
MRI
an
d
C
T
en
h
an
ce
s
h
ea
lth
ca
r
e
p
r
o
f
ess
io
n
als
’
ca
p
ac
ity
to
m
ak
e
b
etter
d
ec
is
io
n
s
an
d
s
o
lv
e
p
r
o
b
lem
s
b
y
lev
er
a
g
in
g
b
o
th
o
f
th
eir
s
tr
en
g
th
s
an
d
o
v
er
co
m
i
n
g
th
eir
d
r
awb
ac
k
s
.
3.
M
E
T
H
O
D
T
h
e
p
r
ev
alen
ce
o
f
b
r
ain
tu
m
o
r
s
is
o
n
th
e
r
is
e.
I
t
is
n
ec
e
s
s
ar
y
to
d
ev
elo
p
a
co
m
p
r
eh
e
n
s
iv
e
r
is
k
ass
es
s
m
en
t
s
y
s
tem
in
o
r
d
er
to
d
ea
l
with
th
is
is
s
u
e.
T
o
ac
h
iev
e
o
p
tim
al
p
r
ed
ictio
n
ac
cu
r
ac
y
,
s
tr
in
g
en
t
m
ac
h
in
e
lear
n
in
g
m
eth
o
d
o
l
o
g
ies
m
u
s
t
b
e
im
p
lem
en
ted
.
T
h
is
s
ec
tio
n
d
is
cu
s
s
es
th
e
s
ev
er
al
s
te
p
s
in
v
o
lv
ed
i
n
th
e
m
eth
o
d
o
l
o
g
y
o
f
th
e
p
r
o
p
o
s
ed
s
y
s
tem
.
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
o
l
o
g
y
c
o
n
tain
s
o
f
th
e
f
o
llo
win
g
s
tag
es.
i)
c
o
llectio
n
o
f
b
r
ain
t
u
m
o
r
MRI
an
d
C
T
s
ca
n
,
ii)
g
en
e
r
atin
g
th
e
f
u
s
ed
im
a
g
es
,
an
d
i
ii)
clas
s
if
icatio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
Leve
r
a
g
in
g
ima
g
e
fu
s
io
n
a
n
d
tr
a
n
s
fer lea
r
n
in
g
fo
r
en
h
a
n
ce
d
tu
mo
r
d
ia
g
n
o
s
is
(
N
ir
ma
la
jyo
t
h
i Na
r
is
etty
)
1807
3
.1
.
Co
llect
i
o
n o
f
bra
in t
um
o
r
M
RI a
nd
C
T
s
ca
n
Fo
r
an
y
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
,
th
e
d
ataset
p
lay
s
a
k
ey
r
o
le.
T
h
e
d
atasets
ar
e
d
o
wn
lo
ad
ed
f
r
o
m
Kag
g
le
ea
ch
f
o
r
MRI
an
d
f
u
s
ed
im
ag
es
g
en
e
r
ated
f
r
o
m
M
R
I
an
d
C
T
im
ag
es
ex
clu
s
iv
el
y
.
T
h
e
d
atasets
ar
e
in
ter
n
ally
f
u
r
th
er
d
iv
id
ed
i
n
to
ty
p
es
o
f
tu
m
o
r
s
n
am
el
y
g
lio
m
a,
m
en
in
g
io
m
a,
p
itu
itar
y
,
a
n
d
n
o
tu
m
o
r
.
E
ac
h
d
ataset
co
n
s
is
ts
o
f
2
,
9
6
1
tr
ain
i
n
g
im
a
g
es
wh
ich
i
n
clu
d
e
8
2
6
g
lio
m
a,
2
4
7
m
en
in
g
io
m
a,
8
2
7
p
itu
itar
y
,
an
d
3
2
7
n
o
tu
m
o
r
im
ag
es.
Ap
ar
t
f
r
o
m
tr
ain
in
g
th
e
d
ataset
co
n
s
is
ts
o
f
3
9
3
test
in
g
im
ag
es
wh
ich
in
clu
d
e
1
0
0
g
lio
m
a
,
1
1
5
m
e
n
in
g
io
m
a,
1
0
4
n
o
t
u
m
o
r
,
an
d
7
4
p
itu
itar
y
im
ag
e
s
.
T
h
e
d
ataset
is
s
p
lit
in
to
tr
ain
in
g
an
d
test
in
g
r
esp
ec
tiv
ely
.
T
h
e
v
alid
atio
n
d
ata
is
s
p
lit
in
to
a
r
atio
o
f
9
0
:1
0
.
A
ten
-
f
o
ld
cr
o
s
s
v
alid
atio
n
is
ap
p
lied
to
av
o
i
d
th
e
o
v
er
f
itti
n
g
,
wh
e
r
e
o
n
e
s
et
ev
er
y
tim
e
is
u
s
ed
to
v
alid
ate,
e
x
ce
p
t
th
at
all
o
th
er
s
ets
ar
e
u
s
ed
f
o
r
tr
ain
in
g
.
T
h
is
p
r
o
ce
s
s
is
p
er
f
o
r
m
ed
iter
ativ
ely
f
o
r
all
th
e
1
0
f
o
ld
s
g
iv
in
g
1
0
d
if
f
er
en
t
r
esu
lts
f
o
r
ea
ch
iter
atio
n
.
T
h
e
av
er
ag
e
o
f
all
th
ese
r
esu
lts
is
co
n
s
id
er
ed
as
th
e
r
esu
lt
(
b
ase
d
o
n
wh
ich
th
e
tr
ain
in
g
a
n
d
te
s
tin
g
ac
cu
r
ac
ies
ar
e
co
m
p
ar
ed
)
.
3
.2
.
G
ener
a
t
ing
t
he
f
us
ed
im
a
g
es
T
h
e
s
ec
o
n
d
p
a
r
t
o
f
th
e
m
et
h
o
d
o
lo
g
y
is
a
m
u
ltimo
d
al
f
u
s
io
n
o
f
MRI
an
d
C
T
s
ca
n
s
to
g
e
n
er
ate
th
e
f
u
s
ed
im
ag
e
to
p
r
e
p
ar
e
th
e
d
at
aset.
Mu
ltimo
d
al
f
u
s
io
n
o
f
M
R
I
an
d
C
T
im
ag
es is
a
tech
n
iq
u
e
to
f
u
s
e
two
b
r
ain
s
ca
n
s
o
f
th
e
s
am
e
p
atien
t
as
s
h
o
wn
in
Fig
u
r
e
3
.
T
h
e
f
u
s
io
n
m
eth
o
d
o
l
o
g
y
is
ca
r
r
ied
o
u
t
in
two
m
ain
p
h
ases
.
T
h
e
in
itial
s
tep
is
r
eg
is
tr
atio
n
,
to
b
r
in
g
MRI
a
n
d
C
T
im
ag
e
s
in
to
a
co
m
m
o
n
co
o
r
d
in
ate
s
y
s
tem
,
L
an
d
m
ar
k
-
b
ased
im
ag
e
r
eg
is
tr
atio
n
is
p
e
r
f
o
r
m
e
d
b
y
s
elec
tin
g
a
m
in
im
u
m
o
f
5
a
n
d
a
m
a
x
im
u
m
o
f
1
0
co
o
r
d
in
ate
p
o
in
ts
(
C
T
p
o
in
ts
an
d
MRI
p
o
in
ts
)
t
o
m
ap
th
e
im
ag
es
g
e
n
er
atin
g
r
eg
is
ter
ed
im
ag
es
f
o
r
b
o
t
h
M
R
I
an
d
C
T
s
ca
n
s
b
y
p
r
ep
r
o
ce
s
s
in
g
th
e
im
ag
es
u
s
in
g
tech
n
i
q
u
es
lik
e
r
e
f
lectio
n
,
s
ca
lin
g
,
tr
a
n
s
latio
n
to
g
e
n
er
ate
an
o
p
tim
u
m
tr
an
s
f
o
r
m
atio
n
m
at
r
ix
f
o
r
th
e
g
iv
en
C
T
im
ag
e
k
ee
p
in
g
it
f
ix
ed
an
d
m
o
d
if
y
i
n
g
MRI
ac
co
r
d
in
g
to
C
T
.
No
w,
th
e
r
esu
ltin
g
r
eg
is
ter
ed
im
a
g
es
ca
n
b
e
d
ir
ec
tly
f
u
s
ed
to
g
en
er
ate
a
f
u
s
ed
im
ag
e
th
at
co
n
tain
s
all
th
e
in
f
o
r
m
atio
n
f
r
o
m
b
o
th
MRI
an
d
C
T
s
ca
n
s
.
Fig
u
r
e
3
.
Me
th
o
d
o
lo
g
y
o
f
MRI
an
d
C
T
s
ca
n
s
f
u
s
io
n
u
s
in
g
V
GG
-
19
T
h
e
s
ec
o
n
d
s
tep
is
th
e
m
ain
f
u
s
io
n
p
er
f
o
r
m
e
d
u
s
in
g
tr
a
n
s
f
er
lear
n
in
g
an
d
DW
T
.
T
r
a
n
s
f
er
l
ea
r
n
in
g
is
a
tech
n
iq
u
e
to
u
s
e
th
e
i
n
tellig
en
ce
f
r
o
m
a
p
r
e
-
t
r
ain
ed
f
r
a
m
ewo
r
k
to
g
ain
f
aster
p
r
o
g
r
e
s
s
in
f
u
r
t
h
er
task
s
.
v
is
u
al
g
eo
m
etr
y
g
r
o
u
p
(
VGG
-
1
9
)
C
NN
ar
ch
itectu
r
e
is
u
s
ed
in
th
e
p
r
o
p
o
s
ed
f
u
s
io
n
p
r
o
ce
s
s
.
DW
T
u
tili
ze
s
th
e
Py
W
av
elets
lib
r
ar
y
(
‘
p
y
wt
’
)
,
an
d
th
e
f
u
n
ctio
n
ap
p
lies
th
e
DW
T
to
b
o
th
th
e
M
R
I
an
d
C
T
im
ag
es
u
s
in
g
th
e
‘
h
aa
r
’
wav
elet.
T
h
is
d
ec
o
m
p
o
s
es
th
e
im
ag
es
in
to
ap
p
r
o
x
im
atio
n
(
L
L
)
an
d
d
etails
(
L
H,
HL
,
HH)
co
ef
f
icien
ts
ac
r
o
s
s
d
if
f
er
en
t
s
ca
les.
E
ac
h
s
et
o
f
co
ef
f
icien
ts
(
L
L
,
L
H,
HL
,
HH)
f
r
o
m
b
o
th
MRI
an
d
C
T
im
ag
es
ar
e
s
av
e
d
as sep
ar
ate
g
r
ay
s
ca
le
im
ag
es.
Fo
r
ea
ch
co
e
f
f
icien
t set,
a
f
u
s
i
o
n
p
r
o
ce
s
s
is
co
n
d
u
cted
.
T
h
is
f
u
s
io
n
p
r
o
ce
s
s
b
eg
in
s
b
y
co
n
v
er
tin
g
in
p
u
t
im
ag
es
to
t
h
e
YC
b
C
r
co
lo
r
s
p
ac
e,
a
co
m
m
o
n
co
lo
r
s
p
a
ce
u
s
ed
in
im
ag
e
p
r
o
ce
s
s
in
g
.
I
f
th
e
in
p
u
t
im
ag
es
a
r
e
n
o
t
g
r
a
y
s
ca
le,
th
ey
ar
e
c
o
n
v
er
ted
to
YC
b
C
r
an
d
n
o
r
m
alize
d
.
Oth
e
r
wis
e,
g
r
ay
s
ca
le
im
ag
es
ar
e
d
ir
ec
tly
n
o
r
m
alize
d
.
No
r
m
aliza
tio
n
in
v
o
l
v
es
co
n
v
er
tin
g
th
e
p
ix
el
v
alu
es
to
th
e
r
an
g
e
[
0
,
1
]
.
T
h
e
n
o
r
m
alize
d
im
ag
es
ar
e
th
en
co
n
v
e
r
ted
in
to
Py
T
o
r
ch
t
en
s
o
r
s
,
f
ac
ilit
atin
g
f
u
r
th
er
p
r
o
ce
s
s
in
g
with
in
th
e
d
ee
p
lear
n
in
g
f
r
am
ewo
r
k
.
N
o
w,
ea
ch
in
p
u
t
im
a
g
e
is
p
ass
ed
th
r
o
u
g
h
th
e
m
o
d
el,
wh
ich
ex
tr
ac
ts
f
ea
tu
r
e
m
ap
s
f
r
o
m
t
h
e
im
ag
es
an
d
co
m
b
in
es
th
em
to
cr
ea
te
“
s
u
m
m
a
p
s
”
r
ep
r
esen
ti
n
g
t
h
e
im
p
o
r
tan
ce
o
f
d
if
f
er
en
t
f
ea
tu
r
es a
cr
o
s
s
im
ag
es.
T
h
e
f
u
s
io
n
s
tr
ateg
y
in
v
o
lv
e
s
ca
lcu
latin
g
weig
h
ts
f
o
r
e
ac
h
in
p
u
t
im
ag
e
b
ased
o
n
its
f
ea
tu
r
e
im
p
o
r
tan
ce
wh
ich
i
n
o
t
h
er
te
r
m
s
is
ca
lled
weig
h
ted
f
u
s
io
n
.
T
h
e
weig
h
ts
ar
e
o
b
tain
ed
b
y
a
p
p
ly
in
g
a
‘
s
o
f
tm
ax
’
fu
n
ctio
n
to
th
e
f
ea
tu
r
e
m
a
p
s
an
d
in
ter
p
o
latin
g
th
em
to
m
atc
h
th
e
d
im
en
s
io
n
s
o
f
th
e
in
p
u
t
im
ag
es.
T
h
e
in
p
u
t
im
ag
es
ar
e
th
e
n
m
u
ltip
lied
b
y
th
eir
r
esp
ec
tiv
e
weig
h
ts
an
d
s
u
m
m
ed
t
o
p
r
o
d
u
ce
th
e
f
u
s
ed
im
a
g
e.
A
m
ax
f
u
s
io
n
s
tr
ateg
y
is
ap
p
l
ied
to
en
s
u
r
e
th
e
p
r
eser
v
at
io
n
o
f
s
alien
t
f
ea
tu
r
es
ac
r
o
s
s
im
a
g
es.
T
h
is
in
v
o
l
v
es
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
7
,
No
.
3
,
Ma
r
ch
20
2
5
:
1
804
-
1
8
1
4
1808
iter
ativ
ely
co
m
p
ar
in
g
an
d
u
p
d
atin
g
th
e
f
u
s
ed
im
ag
e
to
p
r
eser
v
e
th
e
m
ax
im
u
m
v
alu
ed
p
ix
e
ls
f
r
o
m
ea
ch
in
p
u
t
im
ag
e.
Fin
ally
en
s
u
r
in
g
co
m
p
atib
ilit
y
with
s
tan
d
ar
d
im
ag
e
f
o
r
m
ats,
t
h
e
f
u
s
ed
im
ag
e
is
r
ec
o
n
s
tr
u
cte
d
a
n
d
r
e
t
u
r
n
e
d
a
s
a
n
a
r
r
a
y
o
f
p
i
x
e
l
v
a
l
u
e
s
o
n
p
e
r
f
o
r
m
i
n
g
i
n
v
e
r
s
e
w
a
v
e
l
e
t
t
r
a
n
s
f
o
r
m
(
I
D
W
T
)
r
o
o
t
i
n
g
t
o
i
n
c
r
e
a
s
e
i
n
b
r
a
i
n
t
u
m
o
r
c
l
a
s
s
i
f
ic
a
t
i
o
n
a
c
c
u
r
a
c
y
,
r
o
b
u
s
t
n
e
s
s
,
a
n
d
e
f
f
e
c
t
i
v
en
e
s
s
d
et
e
c
t
i
o
n
o
f
t
u
m
o
r
e
v
e
n
d
u
r
i
n
g
i
n
i
t
i
a
l
s
t
a
g
es
.
3
.3
.
B
ra
in t
um
o
r
det
ec
t
io
n
T
h
e
th
ir
d
p
ar
t
is
im
p
lem
en
tin
g
a
b
r
ain
tu
m
o
r
class
if
icatio
n
m
o
d
el
o
n
ex
clu
s
iv
e
MRI,
ex
c
lu
s
iv
e
C
T
,
an
d
f
u
s
ed
im
ag
es
to
co
m
p
ar
e
th
e
ac
cu
r
ac
y
attain
ed
b
y
e
ac
h
o
f
th
e
m
o
d
el
p
r
o
v
in
g
f
u
s
ed
im
ag
es
ten
d
to
class
if
y
b
r
ain
tu
m
o
r
m
u
ch
b
e
tter
th
an
MRI
an
d
C
T
alo
n
e.
T
h
e
d
etailed
p
r
o
ce
d
u
r
e
is
d
e
p
icted
in
Fig
u
r
e
4
.
T
h
e
m
u
lti
-
class
clas
s
if
icatio
n
o
f
b
r
ain
tu
m
o
r
is
p
er
f
o
r
m
e
d
u
s
in
g
C
NN
ar
ch
itectu
r
e.
T
h
e
in
p
u
t
d
ataset,
co
n
s
is
tin
g
o
f
im
ag
es,
is
p
r
e
-
p
r
o
ce
s
s
ed
an
d
th
en
s
p
lit
in
to
t
r
ain
in
g
an
d
test
in
g
s
ets.
T
h
e
d
ataset
co
n
s
is
t
s
o
f
m
u
ltip
le
class
es,
s
o
lab
el
en
co
d
in
g
is
p
er
f
o
r
m
ed
wh
e
r
e
d
if
f
er
en
t
class
es
ar
e
en
co
d
ed
as
n
u
m
er
ical
v
alu
es
to
f
ac
ilit
ate
m
o
d
el
tr
ain
in
g
u
s
in
g
T
en
s
o
r
Flo
w/Ker
as u
tili
ties
.
T
h
e
C
NN
ar
ch
itectu
r
e
co
m
p
r
i
s
es
m
u
ltip
le
co
n
v
o
lu
tio
n
al,
m
ax
-
p
o
o
lin
g
,
d
en
s
e,
an
d
d
r
o
p
o
u
t
lay
er
s
t
o
ex
tr
ac
t
h
ier
ar
ch
ical
f
ea
tu
r
es
an
d
p
r
e
v
en
t
o
v
e
r
f
itti
n
g
.
T
h
e
m
o
d
el
s
tar
ts
with
a
co
n
v
o
lu
ti
o
n
al
lay
er
u
s
in
g
3
2
f
ilter
s
,
f
o
llo
wed
b
y
a
r
ec
tifie
d
lin
ea
r
u
n
it
(
R
eL
U)
ac
tiv
atio
n
f
u
n
ctio
n
f
o
r
n
o
n
-
lin
ea
r
ity
.
T
h
e
n
etwo
r
k
th
en
u
s
es
s
ev
er
al
co
n
v
o
lu
tio
n
al
l
ay
er
s
with
in
cr
ea
s
in
g
f
ilter
s
izes
(
6
4
,
1
2
8
,
2
5
6
)
to
p
r
o
g
r
ess
iv
ely
ex
t
r
ac
t
m
o
r
e
co
m
p
lex
f
ea
tu
r
es
f
r
o
m
th
e
im
ag
es.
I
n
b
etwe
en
th
ese
lay
er
s
,
m
ax
-
p
o
o
lin
g
with
a
2
x
2
win
d
o
w
r
ed
u
ce
s
th
e
s
ize
o
f
th
e
d
ata
an
d
lo
wer
s
co
m
p
u
tatio
n
a
l
co
s
t.
Dr
o
p
o
u
t
lay
e
r
s
(
s
et
at
3
0
%
d
r
o
p
o
u
t
r
ate)
ar
e
in
s
er
ted
to
av
o
id
th
e
m
o
d
e
l
f
r
o
m
o
v
er
f
itti
n
g
o
n
th
e
tr
ain
i
n
g
d
ata.
Dr
o
p
o
u
t
lay
er
s
b
y
a
d
r
o
p
o
u
t
r
ate
o
f
0
.
3
ar
e
ad
d
e
d
to
r
e
g
u
lar
ize
th
e
m
o
d
el
a
n
d
o
v
e
r
co
n
m
e
o
v
er
f
it
tin
g
.
Flatten
ed
f
ea
tu
r
e
m
a
p
s
a
r
e
f
ed
in
to
d
e
n
s
ely
co
n
n
ec
ted
lay
er
s
to
p
e
r
f
o
r
m
class
if
icatio
n
.
T
wo
d
en
s
e
lay
er
s
with
5
1
2
u
n
its
an
d
R
eL
U
ac
tiv
atio
n
f
u
n
ctio
n
s
ar
e
ad
d
e
d
to
ca
p
tu
r
e
h
ig
h
-
lev
e
l
ab
s
tr
ac
tio
n
s
f
r
o
m
th
e
ex
tr
ac
te
d
f
ea
tu
r
es.
T
h
e
f
in
al
o
u
tp
u
t
lay
er
co
n
s
is
ts
o
f
f
o
u
r
u
n
its
with
s
o
f
tm
ax
ac
tiv
atio
n
,
p
r
o
d
u
cin
g
p
r
o
b
ab
ilit
ies
f
o
r
ea
ch
class
in
th
e
class
if
icatio
n
task
.
T
o
t
r
ain
th
e
m
o
d
el
ef
f
ec
tiv
ely
,
it
is
s
et
u
p
t
o
m
in
im
ize
er
r
o
r
s
in
its
p
r
ed
ictio
n
s
.
T
h
is
is
ac
h
iev
ed
u
s
in
g
a
s
p
ec
if
ic
m
ath
em
atica
l
f
u
n
ctio
n
(
ca
teg
o
r
ical
cr
o
s
s
-
en
tr
o
p
y
)
a
n
d
an
o
p
tim
izatio
n
alg
o
r
ith
m
(
Ad
am
)
th
at
h
elp
s
t
h
e
m
o
d
el
lear
n
f
r
o
m
its
m
is
tak
es
an
d
im
p
r
o
v
e
its
ac
cu
r
ac
y
o
v
er
tim
e.
T
h
e
m
o
d
el
is
tr
ain
ed
d
esig
n
ed
f
o
r
1
7
e
p
o
ch
s
with
a
v
alid
atio
n
s
p
lit
o
f
1
0
%
a
n
d
an
ea
r
ly
s
to
p
p
in
g
cr
iter
io
n
to
p
r
e
v
en
t
o
v
er
f
itti
n
g
.
All
th
e
p
ar
am
eter
s
ettin
g
s
o
f
th
e
m
o
d
el
ar
e
lis
ted
in
T
ab
le
1
.
Mo
d
el
p
er
f
o
r
m
an
ce
is
m
o
n
ito
r
ed
u
s
in
g
ac
cu
r
ac
y
as
th
e
ass
es
s
m
en
t
m
etr
ic.
T
h
e
tr
ain
ed
m
o
d
el
’
s
p
er
f
o
r
m
an
ce
is
ev
alu
ated
o
n
th
e
test
s
et
to
ass
ess
its
g
en
er
aliza
tio
n
ab
ilit
y
an
d
ef
f
ec
tiv
e
n
ess
in
class
if
y
in
g
u
n
s
ee
n
d
ata.
Fig
u
r
e
4
.
Me
th
o
d
o
lo
g
y
o
f
b
r
ain
tu
m
o
r
c
lass
if
icatio
n
u
s
in
g
C
NN
T
ab
le
1
.
Par
am
eter
s
u
s
ed
i
n
C
NN
ar
ch
itectu
r
e
M
o
d
e
l
S
e
q
u
e
n
t
i
a
l
C
o
n
v
o
2
D
9
M
a
x
p
o
o
l
i
n
g
4
D
r
o
p
o
u
t
6
D
e
n
se
3
F
l
a
t
t
e
n
1
A
c
t
i
v
a
t
i
o
n
F
u
n
c
t
i
o
n
R
e
LU
,
S
o
f
t
M
a
x
Lo
ss f
u
n
c
t
i
o
n
C
a
t
e
g
o
r
i
c
a
l
c
r
o
ss
e
n
t
r
o
p
y
O
p
t
i
mi
z
e
r
A
d
a
m
M
e
t
r
i
c
s
A
c
c
u
r
a
c
y
I
mag
e
i
n
p
u
t
s
h
a
p
e
(
1
5
0
,
1
5
0
,
3
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
Leve
r
a
g
in
g
ima
g
e
fu
s
io
n
a
n
d
tr
a
n
s
fer lea
r
n
in
g
fo
r
en
h
a
n
ce
d
tu
mo
r
d
ia
g
n
o
s
is
(
N
ir
ma
la
jyo
t
h
i Na
r
is
etty
)
1809
T
h
is
b
r
ain
tu
m
o
r
class
if
icatio
n
m
o
d
el
is
ca
r
r
ied
o
u
t
o
n
a
n
ex
clu
s
iv
e
MRI
d
ataset
an
d
f
u
s
ed
im
ag
es
d
ataset.
T
h
e
ac
cu
r
ac
y
o
f
all
t
h
r
ee
m
o
d
els
is
co
m
p
a
r
ed
to
an
aly
ze
wh
ich
d
ataset
is
m
o
r
e
ef
f
icien
t
th
a
n
th
e
o
th
er
two
d
atasets
in
class
if
y
i
n
g
th
e
tu
m
o
r
p
r
ec
is
ely
.
T
h
e
u
s
er
m
u
s
t
u
p
lo
ad
MRI
an
d
C
T
p
r
e
-
p
r
o
ce
s
s
ed
s
ca
n
s
in
th
e
web
s
ite
s
h
o
wn
in
Fig
u
r
e
5
an
d
en
ter
th
e
n
u
m
b
er
o
f
p
o
in
ts
f
o
r
r
eg
is
tr
atio
n
.
T
h
is
r
eg
is
tr
atio
n
is
a
lan
d
m
ar
k
-
b
ased
r
eg
is
tr
atio
n
w
ith
a
m
i
n
im
u
m
o
f
5
an
d
m
a
x
i
m
u
m
o
f
1
0
r
eg
is
tr
atio
n
p
o
in
ts
ca
n
b
e
s
elec
ted
b
y
th
e
u
s
er
m
an
u
ally
d
u
r
in
g
r
eg
is
tr
atio
n
.
I
n
th
e
p
r
e
v
io
u
s
s
tep
th
e
s
elec
tio
n
o
f
co
m
m
o
n
co
o
r
d
in
ate
p
o
in
ts
is
p
er
f
o
r
m
e
d
as
s
h
o
w
n
in
Fig
u
r
e
6
,
an
d
th
e
co
r
r
esp
o
n
d
in
g
r
esu
lt
o
f
th
e
s
elec
tio
n
is
d
is
p
lay
ed
in
web
p
ag
e
s
h
o
wn
in
Fig
u
r
e
7
.
I
t
s
h
o
ws
th
e
r
eg
is
ter
ed
MRI
an
d
r
e
g
is
ter
ed
C
T
s
ca
n
s
d
is
p
lay
ed
.
T
h
e
b
u
tt
o
n
to
p
er
f
o
r
m
f
u
s
io
n
o
f
th
e
r
e
g
is
ter
ed
im
ag
es
is
g
iv
en
b
el
o
w
th
e
d
is
p
lay
ed
im
a
g
es.
Fig
u
r
e
5
.
W
eb
lan
d
in
g
to
g
iv
e
MRI
an
d
C
T
s
ca
n
s
o
f
s
am
e
p
a
tien
t a
s
in
p
u
ts
alo
n
g
with
n
u
m
b
er
o
f
r
eg
is
tr
atio
n
p
o
in
ts
Fig
u
r
e
6
.
Dis
p
lay
o
f
u
p
lo
ad
ed
MRI
an
d
C
T
s
ca
n
s
to
MRI
an
d
C
T
f
o
r
co
m
m
o
n
co
o
r
d
in
ate
p
o
in
ts
o
n
b
o
th
f
o
r
r
eg
is
tr
atio
n
Fig
u
r
e
7
.
Dis
p
lay
o
f
r
eg
is
ter
ed
s
elec
t scan
s
r
ea
d
y
f
u
s
io
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
7
,
No
.
3
,
Ma
r
ch
20
2
5
:
1
804
-
1
8
1
4
1810
I
f
u
s
er
is
s
u
r
e
ab
o
u
t
th
e
co
m
m
o
n
co
o
r
d
in
ate
s
p
ac
e
th
e
f
u
s
io
n
o
f
t
h
e
r
e
g
is
ter
ed
s
ca
n
s
ca
n
b
e
in
itiated
.
T
h
e
f
u
s
io
n
r
esu
lt
o
f
MRI
an
d
C
T
s
ca
n
s
is
d
is
p
lay
ed
in
Fig
u
r
e
8
web
p
a
g
e.
T
h
is
f
u
s
ed
im
ag
e
co
n
tain
s
th
e
co
m
b
in
ed
i
n
f
o
r
m
atio
n
o
f
MR
I
an
d
C
T
s
ca
n
s
.
Hen
ce
o
v
er
c
o
m
in
g
in
d
iv
id
u
al
d
is
ad
v
a
n
tag
es
o
f
b
o
th
s
ca
n
s
to
s
o
m
e
ex
ten
t.
Fig
u
r
e
8
.
Dis
p
lay
o
f
f
in
ally
f
u
s
ed
im
ag
e
o
f
r
eg
is
ter
ed
MRI
a
n
d
C
T
s
ca
n
4.
RE
SU
L
T
S
AND
D
I
S
C
US
SI
O
N
A
f
u
s
ed
im
ag
e
’
s
q
u
ality
is
ass
es
s
ed
b
y
th
r
ee
cr
iter
io
n
m
ea
s
u
r
es:
MSE
,
PS
N
R
,
an
d
SS
I
M.
B
y
co
m
p
ar
in
g
th
r
ee
ch
a
r
ac
ter
is
tics
o
f
two
im
ag
es,
SS
I
M
ca
lcu
lates
h
o
w
s
im
ilar
th
ey
ar
e,
in
c
lu
d
in
g
lu
m
i
n
an
ce
,
co
n
tr
ast,
an
d
s
tr
u
ctu
r
e.
A
h
i
g
h
SS
I
M
v
alu
e
i
n
d
icate
s
a
g
o
o
d
p
r
eser
v
atio
n
o
f
s
tr
u
ctu
r
al
in
f
o
r
m
atio
n
.
PS
NR
v
alu
es
g
r
ea
ter
t
h
an
1
in
d
icate
b
etter
r
ec
o
n
s
tr
u
ctio
n
q
u
ality
b
etwe
en
f
u
s
ed
an
d
r
ef
er
e
n
ce
p
ictu
r
es.
A
PS
NR
v
alu
e
o
f
2
1
.
0
9
is
m
o
d
er
ate,
s
u
g
g
esti
n
g
t
h
at
th
e
f
u
s
ed
im
ag
e
h
as
a
d
ec
en
t
lev
el
o
f
o
v
er
al
l
f
id
elity
.
I
t
is
also
less
co
m
p
ar
ed
to
th
e
v
alu
e
o
f
PS
NR
in
[
1
1
]
.
T
h
e
MSE
ca
l
cu
lates
d
if
f
er
en
ce
s
at
th
e
p
ix
e
l
lev
el,
b
u
t
ig
n
o
r
es
s
tr
u
ctu
r
al
o
r
p
er
ce
p
t
u
al
s
im
ilar
ity
,
s
o
SS
I
M
is
m
o
r
e
ac
c
u
r
at
e.
SS
I
M
m
ea
s
u
r
es
h
o
w
well
t
h
e
f
u
s
io
n
alg
o
r
ith
m
p
r
eser
v
es
s
tr
u
ctu
r
al
d
etails,
wh
ich
is
a
m
ajo
r
g
o
al
o
f
im
ag
e
f
u
s
io
n
.
SS
I
M
s
h
o
u
ld
b
e
weig
h
ed
alo
n
g
s
id
e
PS
NR
an
d
MSE
ev
en
if
th
ey
o
f
f
er
m
o
r
e
co
n
tex
t
r
eq
u
ir
in
g
p
r
ese
r
v
atio
n
o
f
s
h
ap
es,
ed
g
es,
an
d
tex
tu
r
es,
s
u
ch
as
m
ed
ical
im
ag
in
g
,
r
em
o
te
s
en
s
in
g
,
a
n
d
p
h
o
t
o
g
r
ap
h
y
.
Acc
o
r
d
in
g
to
T
ab
le
2
,
t
h
e
v
alu
e
o
f
S
SIM
0
.
9
1
in
d
icate
s
th
at
th
e
f
u
s
ed
im
ag
e
r
etain
s
th
e
s
h
ap
e,
ed
g
e,
a
n
d
tex
tu
r
e
o
f
t
h
e
r
ef
er
en
ce
i
m
ag
e.
Hen
ce
,
t
h
e
m
o
d
el
is
r
eliab
le
f
o
r
th
e
f
u
s
io
n
o
f
MRI
an
d
C
T
s
ca
n
s
.
T
h
e
two
m
o
d
els
im
p
lem
en
te
d
u
s
in
g
ex
clu
s
iv
e
MRI
an
d
f
u
s
ed
im
ag
es
r
esp
ec
tiv
el
y
ar
e
co
m
p
a
r
ed
b
ased
o
n
ac
cu
r
ac
y
.
T
h
e
f
in
d
in
g
s
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
o
lo
g
y
s
h
o
ws
th
at
f
u
s
ed
im
ag
es
ca
n
p
er
f
o
r
m
ef
f
ec
tiv
ely
in
d
etec
tin
g
tu
m
o
r
s
.
T
h
e
tr
ain
in
g
ac
cu
r
ac
y
an
d
v
alid
atio
n
lo
s
s
o
f
m
o
d
el
m
u
s
t
b
e
ev
alu
ated
to
k
n
o
w
h
o
w
th
e
m
o
d
el
u
n
d
er
f
its
o
r
o
v
e
r
f
its
in
p
r
o
v
id
ed
d
ata.
As
esti
m
ated
,
th
e
m
o
d
el
test
ed
u
s
in
g
f
u
s
ed
im
ag
es
r
esu
lted
in
9
4
.
6
3
%
tr
ain
in
g
ac
cu
r
ac
y
wh
ich
is
a
b
etter
s
co
r
e
co
m
p
a
r
ed
to
th
e
m
o
d
el
tr
ain
e
d
u
s
in
g
MRI
with
a
9
2
.
9
8
%
ac
cu
r
ac
y
.
Ad
d
itio
n
ally
,
test
in
g
ac
cu
r
ac
y
u
s
in
g
f
u
s
ed
im
ag
es
(
9
3
.
9
8
%)
is
f
ar
b
etter
th
an
MRI
alo
n
e
with
a
lo
wer
6
7
.
9
7
%.
T
h
ese
r
esu
lts
s
u
f
f
icien
tly
p
r
o
v
e
th
at
f
u
s
ed
im
ag
es
ca
n
m
o
r
e
p
r
ec
is
ely
an
d
ac
cu
r
ately
class
if
y
b
r
ain
tu
m
o
r
s
th
an
MRI
o
r
C
T
tak
en
in
d
iv
id
u
ally
.
T
h
e
r
ea
s
o
n
f
o
r
th
e
h
i
g
h
er
ac
cu
r
ac
y
o
f
t
h
e
f
u
s
ed
im
ag
es
is
th
at
th
e
f
u
s
e
d
im
ag
es
co
n
tain
all
t
h
e
im
p
o
r
tan
t
in
f
o
r
m
atio
n
f
r
o
m
th
e
C
T
as
well
as
MRI
im
ag
es r
esu
ltin
g
in
p
ea
k
in
g
m
o
d
el
co
r
r
ec
tn
ess
.
T
ab
le
2.
Fu
s
io
n
m
o
d
el
m
etr
ics
an
d
r
esu
lts
M
e
t
r
i
c
s
V
a
l
u
e
s
SSIM
0
.
9
1
P
S
N
R
2
1
.
0
9
M
S
E
5
0
6
.
3
2
T
h
is
r
esear
ch
ex
a
m
in
es
h
o
w
well
b
r
ain
tu
m
o
r
s
m
ay
b
e
d
e
tecte
d
b
y
co
m
b
i
n
in
g
C
T
a
n
d
MRI
d
ata
u
s
in
g
th
e
DW
T
.
Go
o
d
s
tr
u
ctu
r
al
s
im
ilar
ity
an
d
ac
ce
p
tab
le
n
o
is
e
lev
els
in
th
e
f
u
s
ed
p
ictu
r
e
ar
e
s
h
o
wn
b
y
t
h
e
ass
es
s
m
en
t
m
etr
ics
(
SS
I
M,
P
SNR
,
an
d
MSE
)
,
wh
ich
v
ali
d
ate
s
u
cc
ess
f
u
l
im
ag
e
f
u
s
io
n
.
T
h
e
ac
cu
r
ac
y
o
f
tr
ain
in
g
a
n
d
ass
ess
in
g
th
e
f
u
s
ed
m
o
d
el
was
s
ig
n
if
ican
tly
h
i
g
h
er
t
h
an
t
h
at
o
f
m
o
d
els
b
ase
d
o
n
in
d
iv
id
u
al
MRI
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
Leve
r
a
g
in
g
ima
g
e
fu
s
io
n
a
n
d
tr
a
n
s
fer lea
r
n
in
g
fo
r
en
h
a
n
ce
d
tu
mo
r
d
ia
g
n
o
s
is
(
N
ir
ma
la
jyo
t
h
i Na
r
is
etty
)
1811
o
r
C
T
d
ata.
T
h
is
d
em
o
n
s
tr
ates
th
e
b
e
n
ef
it
o
f
im
a
g
e
f
u
s
io
n
:
th
e
f
u
s
ed
im
a
g
e
o
f
f
er
s
a
m
o
r
e
co
m
p
lete
v
iew
o
f
th
e
b
r
ain
,
en
a
b
lin
g
th
e
m
o
d
el
to
class
if
y
tu
m
o
r
s
b
etter
.
T
h
ese
r
esu
lts
s
u
g
g
est
in
co
r
p
o
r
atin
g
im
ag
e
f
u
s
io
n
in
t
o
th
e
d
iag
n
o
s
tic
wo
r
k
f
lo
w
co
u
ld
en
h
an
ce
b
r
ain
tu
m
o
r
d
etec
tio
n
ac
cu
r
ac
y
.
T
h
e
ca
r
ef
u
l
o
b
s
er
v
atio
n
s
f
r
o
m
th
e
Fig
u
r
e
9
is
th
at
th
e
m
o
d
el
tr
ain
ed
o
n
f
u
s
ed
im
ag
es
ac
h
iev
ed
a
r
em
ar
k
ab
le
tr
ain
in
g
ac
c
u
r
ac
y
o
f
9
4
.
6
3
%,
s
h
o
wca
s
in
g
a
n
i
m
p
r
o
v
e
m
en
t
c
o
m
p
ar
e
d
t
o
th
e
9
2
.
9
8
%
ac
c
u
r
ac
y
o
b
tain
ed
with
th
e
m
o
d
el
tr
ain
ed
o
n
MRI
s
ca
n
s
alo
n
e.
T
h
is
ad
v
an
tag
e
tr
an
s
lated
to
test
in
g
ac
cu
r
ac
y
as
well.
T
h
e
f
u
s
ed
m
o
d
el
ac
h
iev
ed
a
test
i
n
g
ac
cu
r
ac
y
o
f
9
3
.
9
8
%,
wh
er
ea
s
th
e
m
o
d
el
tr
ain
ed
o
n
MRI
s
ca
n
s
alo
n
e
y
ield
ed
a
m
u
c
h
lo
wer
ac
cu
r
ac
y
o
f
6
7
.
9
7
%.
T
h
is
s
u
b
s
tan
tial
d
if
f
er
en
ce
h
ig
h
lig
h
ts
th
e
p
o
w
er
o
f
im
ag
e
f
u
s
io
n
.
Fu
s
ed
im
ag
es
o
f
f
er
a
m
o
r
e
co
m
p
lete
v
iew
o
f
th
e
b
r
ain
b
y
c
o
m
b
in
in
g
th
e
p
r
ec
is
e
b
o
n
e
s
tr
u
ctu
r
e
d
ata
f
r
o
m
C
T
s
ca
n
s
with
th
e
d
etailed
s
o
f
t
ti
s
s
u
e
in
f
o
r
m
atio
n
f
r
o
m
MRI
s
ca
n
s
.
Usi
n
g
th
is
ex
ten
s
iv
e
i
n
f
o
r
m
atio
n
,
th
e
m
o
d
el
ca
n
class
if
y
tu
m
o
r
s
m
o
r
e
ac
cu
r
ately
,
wh
ich
m
a
y
r
esu
lt in
ea
r
lier
d
iag
n
o
s
is
an
d
b
etter
p
atien
t o
u
tco
m
es.
Fig
u
r
e
9
.
Acc
u
r
ac
y
v
ar
iatio
n
f
o
r
b
o
t
h
m
o
d
els
T
h
e
tr
ain
in
g
ac
cu
r
ac
y
an
d
v
al
id
atio
n
ac
cu
r
ac
y
o
f
m
o
d
el
m
u
s
t
b
e
ev
alu
ated
to
k
n
o
w
h
o
w
th
e
m
o
d
el
f
its
in
p
r
o
v
id
ed
d
ata.
Fig
u
r
e
s
1
0
an
d
1
1
s
h
o
ws
th
e
tr
ain
in
g
an
d
v
alid
atio
n
o
f
lo
s
s
an
d
ac
cu
r
ac
y
v
ar
iatio
n
th
r
o
u
g
h
all
1
7
ep
o
c
h
s
.
T
h
e
tr
ain
in
g
ac
cu
r
ac
y
co
n
s
tan
tly
p
e
ak
ed
an
d
v
alid
atio
n
o
v
er
all
r
esu
lts
ar
e
ef
f
icien
t.
T
h
e
tr
ain
in
g
ac
c
u
r
ac
y
co
n
s
ta
n
tly
d
ec
r
ea
s
ed
an
d
v
alid
atio
n
o
v
er
all
r
esu
l
ts
ar
e
ef
f
icien
t
a
s
th
ey
d
ec
r
ea
s
ed
in
th
e
en
d
.
T
h
e
tr
ain
in
g
ac
cu
r
ac
y
an
d
v
al
id
atio
n
ac
cu
r
ac
y
o
f
m
o
d
el
m
u
s
t
b
e
ev
alu
ated
to
k
n
o
w
h
o
w
th
e
m
o
d
el
f
its
in
p
r
o
v
id
ed
d
ata.
Fig
u
r
e
1
2
s
h
o
ws
th
e
tr
ain
in
g
an
d
v
al
id
atio
n
ac
cu
r
ac
y
v
ar
iatio
n
th
r
o
u
g
h
all
1
7
ep
o
ch
s
.
T
h
e
tr
ain
in
g
ac
cu
r
ac
y
co
n
s
tan
tly
p
ea
k
ed
an
d
v
alid
atio
n
o
v
e
r
all
r
esu
lts
ar
e
ef
f
icien
t.
E
v
en
th
o
u
g
h
v
alid
atio
n
ac
cu
r
ac
y
h
as
m
a
n
y
u
p
s
an
d
d
o
wn
s
th
e
o
v
er
all
r
esu
lts
f
o
r
f
u
s
ed
m
o
d
el
ar
e
e
f
f
icien
t
th
an
MRI
m
o
d
el.
T
h
e
tr
ain
in
g
ac
cu
r
ac
y
an
d
v
alid
ati
o
n
lo
s
s
o
f
m
o
d
el
m
u
s
t b
e
ev
alu
ated
to
k
n
o
w
h
o
w
th
e
m
o
d
el
u
n
d
er
f
its
o
r
o
v
e
r
f
its
in
p
r
o
v
id
ed
d
ata.
Fig
u
r
e
1
3
s
h
o
ws
th
e
tr
ain
in
g
an
d
v
alid
atio
n
lo
s
s
v
ar
iatio
n
th
r
o
u
g
h
all
1
7
e
p
o
ch
s
.
T
h
e
tr
ain
in
g
ac
c
u
r
ac
y
co
n
s
tan
tly
d
ec
r
ea
s
ed
,
a
n
d
v
alid
atio
n
o
v
e
r
all
r
esu
lts
ar
e
ef
f
icien
t
as
th
ey
d
ec
r
ea
s
ed
in
th
e
en
d
.
B
y
u
s
in
g
f
u
s
ed
im
a
g
es,
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
o
lo
g
y
s
ig
n
if
ican
tly
r
ed
u
ce
d
th
e
v
alu
e
o
f
PS
NR
f
r
o
m
[
1
1
]
an
d
ac
h
iev
es a
n
im
p
r
o
v
ed
ac
c
u
r
ac
y
(
9
6
.
4
3
%)
wh
en
c
o
m
p
a
r
ed
to
th
e
s
tate
-
of
-
th
e
-
ar
t m
et
h
o
d
s
[
1
2
]
.
Fig
u
r
e
1
0
.
T
r
ain
in
g
an
d
v
alid
a
ti
o
n
lo
s
s
f
o
r
MRI
m
o
d
el
Fig
u
r
e
1
1
.
T
r
ain
in
g
an
d
v
alid
a
tio
n
ac
cu
r
ac
y
f
o
r
MRI
m
o
d
el
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
7
,
No
.
3
,
Ma
r
ch
20
2
5
:
1
804
-
1
8
1
4
1812
Fig
u
r
e
1
2
.
T
r
ain
in
g
an
d
v
alid
a
tio
n
ac
cu
r
ac
y
o
f
f
u
s
ed
m
o
d
el
Fig
u
r
e
1
3
.
T
r
ain
in
g
an
d
v
alid
a
tio
n
lo
s
s
f
o
r
Fu
s
ed
m
o
d
el
5.
CO
NCLU
SI
O
N
C
o
m
p
ar
in
g
th
e
ex
is
tin
g
m
eth
o
d
with
a
n
ew
c
u
ttin
g
-
ed
g
e
m
et
h
o
d
,
d
ee
p
lear
n
in
g
m
eth
o
d
s
wer
e
f
o
u
n
d
to
b
e
m
o
r
e
ef
f
ec
tiv
e
f
o
r
b
r
ain
tu
m
o
r
s
eg
m
e
n
tatio
n
b
ased
u
p
o
n
MRI
im
ag
es.
T
h
e
cu
r
r
en
t
s
tu
d
y
p
r
o
p
o
s
es
th
e
h
y
b
r
id
f
u
s
io
n
b
ased
b
r
ain
tu
m
o
r
class
if
icatio
n
u
n
p
r
ec
e
d
e
n
ted
ap
p
licatio
n
u
s
in
g
d
ee
p
lear
n
in
g
a
n
d
d
ata
an
aly
s
is
in
d
etec
tin
g
b
r
ai
n
tu
m
o
r
s
.
T
h
e
VGG
-
1
9
is
u
s
ed
f
o
r
f
u
s
in
g
th
e
MRI
an
d
C
T
im
ag
es.
L
ater
th
ese
im
ag
es
ar
e
in
p
u
t
to
th
e
C
NN
m
o
d
el.
B
y
u
s
in
g
f
u
s
ed
im
a
g
es,
th
e
p
r
o
p
o
s
ed
m
eth
o
d
o
l
o
g
y
s
ig
n
if
ican
tly
r
ed
u
ce
d
th
e
v
alu
e
o
f
PS
NR
an
d
ac
h
i
ev
es
an
im
p
r
o
v
ed
ac
cu
r
ac
y
(
9
6
.
4
3
%)
wh
e
n
co
m
p
ar
ed
t
o
th
e
s
tate
-
of
-
th
e
-
ar
t
m
eth
o
d
s
.
I
n
c
o
n
tr
ast
to
th
e
p
r
e
-
tr
ain
ed
m
o
d
els
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
ac
h
iev
es
b
est
lev
el
a
cc
u
r
ac
y
.
Ho
wev
er
,
th
is
s
tu
d
y
o
n
ly
e
x
am
in
ed
o
n
e
m
o
d
el
a
n
d
d
i
d
n
o
t
c
o
m
p
ar
e
it
with
o
th
er
m
o
d
els
an
d
tim
e
co
m
p
le
x
ity
.
Fu
r
th
er
m
o
r
e
,
th
e
d
ataset
u
s
ed
in
th
is
s
tu
d
y
is
f
o
r
m
Kag
g
le,
wh
ich
m
ay
n
o
t
r
ep
r
esen
t
th
e
wh
o
le
p
o
p
u
latio
n
.
Fo
r
ad
d
r
ess
in
g
all
o
f
th
ese
lim
itatio
n
s
,
f
u
tu
r
e
s
tu
d
ies
will
i
n
v
esti
g
ate
s
u
p
er
v
is
ed
m
ac
h
in
e
lear
n
in
g
m
eth
o
d
s
,
en
s
em
b
le
m
o
d
els,
an
d
tim
e
c
o
m
p
lex
ity
.
I
n
ad
d
itio
n
,
a
we
b
ap
p
licatio
n
ca
n
b
e
u
p
lo
ad
e
d
in
a
clo
u
d
-
b
ase
d
en
v
ir
o
n
m
en
t
o
n
ce
it
is
r
ea
d
y
i
n
f
u
ll
-
f
le
d
g
ed
s
o
th
at
it
ca
n
b
e
ac
ce
s
s
ed
f
r
o
m
an
y
wh
e
r
e
in
th
e
wo
r
ld
.
T
h
is
s
tu
d
y
m
ay
ass
is
t in
p
r
o
v
id
in
g
d
o
ct
o
r
s
a
r
o
b
u
s
t
an
d
en
h
a
n
ce
d
way
t
o
tr
ea
t p
atien
ts
ef
f
ec
tiv
ely
.
RE
F
E
R
E
NC
E
S
[
1
]
J.
R
.
C
a
r
o
n
,
B
.
L
i
t
t
ma
n
n
,
D
.
C
h
i
t
r
a
d
e
v
i
,
a
n
d
M
.
K
r
i
s
h
n
a
m
u
r
t
h
y
,
“
A
s
u
r
v
e
y
o
n
M
R
I
b
a
s
e
d
a
u
t
o
ma
t
e
d
b
r
a
i
n
t
u
m
o
r
se
g
me
n
t
a
t
i
o
n
t
e
c
h
n
i
q
u
e
s,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Ad
v
a
n
c
e
s i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
3
,
n
o
.
1
1
,
2
0
1
4
.
[
2
]
S
.
B
.
K
a
n
g
,
Y
.
L
i
,
X
.
T
o
n
g
,
a
n
d
H
.
Y
.
S
h
u
m
,
“
I
mag
e
-
b
a
se
d
r
e
n
d
e
r
i
n
g
,
”
Fo
u
n
d
a
t
i
o
n
s
a
n
d
T
r
e
n
d
s
i
n
C
o
m
p
u
t
e
r
G
ra
p
h
i
c
s
a
n
d
Vi
si
o
n
,
v
o
l
.
2
,
n
o
.
3
,
p
p
.
1
7
3
–
2
5
8
,
2
0
0
6
,
d
o
i
:
1
0
.
1
5
6
1
/
0
6
0
0
0
0
0
0
1
2
.
[
3
]
S
.
S
o
l
a
n
k
i
,
U
.
P
.
S
i
n
g
h
,
S
.
S
.
C
h
o
u
h
a
n
,
a
n
d
S
.
Ja
i
n
,
“
B
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
a
n
d
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
s
i
n
g
i
n
t
e
l
l
i
g
e
n
c
e
t
e
c
h
n
i
q
u
e
s
:
a
n
o
v
e
r
v
i
e
w
,
”
I
EE
E
A
c
c
e
ss
,
v
o
l
.
1
1
,
p
p
.
1
2
8
7
0
–
1
2
8
8
6
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
3
.
3
2
4
2
6
6
6
.
[
4
]
D
.
H
a
r
i
t
h
a
,
“
C
o
m
p
a
r
a
t
i
v
e
s
t
u
d
y
o
n
b
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
t
e
c
h
n
i
q
u
e
s
,
”
i
n
2
0
1
6
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
S
i
g
n
a
l
Pr
o
c
e
ssi
n
g
,
C
o
m
m
u
n
i
c
a
t
i
o
n
,
P
o
w
e
r
a
n
d
Em
b
e
d
d
e
d
S
y
s
t
e
m
(
S
C
O
P
ES
)
,
O
c
t
.
2
0
1
6
,
p
p
.
1
3
8
7
–
1
3
9
2
,
d
o
i
:
1
0
.
1
1
0
9
/
S
C
O
P
ES.
2
0
1
6
.
7
9
5
5
6
6
8
.
[
5
]
G
.
A
.
A
mr
a
n
e
t
a
l
.
,
“
B
r
a
i
n
t
u
m
o
r
c
l
a
ssi
f
i
c
a
t
i
o
n
a
n
d
d
e
t
e
c
t
i
o
n
u
si
n
g
h
y
b
r
i
d
d
e
e
p
t
u
m
o
r
n
e
t
w
o
r
k
,
”
El
e
c
t
r
o
n
i
c
s
(
S
w
i
t
z
e
rl
a
n
d
)
,
v
o
l
.
1
1
,
n
o
.
2
1
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
e
l
e
c
t
r
o
n
i
c
s
1
1
2
1
3
4
5
7
.
[
6
]
L.
C
h
a
n
d
r
a
sh
e
k
a
r
a
n
d
S
r
e
e
d
e
v
i
A
.
,
“
A
h
y
b
r
i
d
m
u
l
t
i
mo
d
a
l
me
d
i
c
a
l
i
ma
g
e
f
u
si
o
n
t
e
c
h
n
i
q
u
e
f
o
r
C
T
a
n
d
M
R
I
b
r
a
i
n
i
m
a
g
e
s,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
e
r
Vi
s
i
o
n
a
n
d
I
m
a
g
e
Pr
o
c
e
ssi
n
g
,
v
o
l
.
8
,
n
o
.
3
,
p
p
.
1
–
1
5
,
2
0
1
8
,
d
o
i
:
1
0
.
4
0
1
8
/
i
j
c
v
i
p
.
2
0
1
8
0
7
0
1
0
1
.
[
7
]
R
.
N
a
n
mara
n
e
t
a
l
.
,
“
I
n
v
e
st
i
g
a
t
i
n
g
t
h
e
r
o
l
e
o
f
i
ma
g
e
f
u
s
i
o
n
i
n
b
r
a
i
n
t
u
mo
r
c
l
a
ssi
f
i
c
a
t
i
o
n
m
o
d
e
l
s
b
a
se
d
o
n
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
m
f
o
r
p
e
r
s
o
n
a
l
i
z
e
d
me
d
i
c
i
n
e
,
”
C
o
m
p
u
t
a
t
i
o
n
a
l
a
n
d
Ma
t
h
e
m
a
t
i
c
a
l
Me
t
h
o
d
s
i
n
M
e
d
i
c
i
n
e
,
v
o
l
.
2
0
2
2
,
p
p
.
1
–
1
3
,
F
e
b
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
2
/
7
1
3
7
5
2
4
.
[
8
]
V
.
A
.
R
a
n
i
a
n
d
S
.
La
l
i
t
h
a
k
u
mar
i
,
“
A
h
y
b
r
i
d
f
u
s
i
o
n
mo
d
e
l
f
o
r
b
r
a
i
n
t
u
m
o
r
i
mag
e
s o
f
M
R
I
a
n
d
C
T,
”
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
2
0
2
0
I
EE
E
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
m
u
n
i
c
a
t
i
o
n
a
n
d
S
i
g
n
a
l
Pro
c
e
ssi
n
g
,
I
C
C
S
P
2
0
2
0
,
p
p
.
1
3
1
2
–
1
3
1
6
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
S
P
4
8
5
6
8
.
2
0
2
0
.
9
1
8
2
3
7
1
.
[
9
]
J.
A
mi
n
,
M
.
S
h
a
r
i
f
,
M
.
Y
a
sm
i
n
,
a
n
d
S
.
L.
F
e
r
n
a
n
d
e
s,
“
A
d
i
s
t
i
n
c
t
i
v
e
a
p
p
r
o
a
c
h
i
n
b
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
a
n
d
c
l
a
ssi
f
i
c
a
t
i
o
n
u
s
i
n
g
M
R
I
,
”
P
a
t
t
e
r
n
Re
c
o
g
n
i
t
i
o
n
L
e
t
t
e
rs
,
v
o
l
.
1
3
9
,
p
p
.
1
1
8
–
1
2
7
,
N
o
v
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
a
t
r
e
c
.
2
0
1
7
.
1
0
.
0
3
6
.
[
1
0
]
P
.
S
a
p
r
a
,
R
.
S
i
n
g
h
,
a
n
d
S
.
K
h
u
r
a
n
a
,
“
B
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
u
si
n
g
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
S
c
i
e
n
c
e
a
n
d
Mo
d
e
r
n
En
g
i
n
e
e
ri
n
g
(
I
J
I
S
ME
)
,
2
0
1
3
.
[
1
1
]
P
.
D
h
a
g
e
,
M
.
R
.
P
h
e
g
a
d
e
,
a
n
d
S
.
K
.
S
h
a
h
,
“
W
a
t
e
r
s
h
e
d
se
g
me
n
t
a
t
i
o
n
b
r
a
i
n
t
u
mo
r
d
e
t
e
c
t
i
o
n
,
”
2
0
1
5
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Pe
rv
a
si
v
e
C
o
m
p
u
t
i
n
g
:
A
d
v
a
n
c
e
C
o
m
m
u
n
i
c
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
a
n
d
A
p
p
l
i
c
a
t
i
o
n
f
o
r
S
o
c
i
e
t
y
,
I
C
PC
2
0
1
5
,
2
0
1
5
,
d
o
i
:
1
0
.
1
1
0
9
/
P
ER
V
A
S
I
V
E.
2
0
1
5
.
7
0
8
6
9
6
7
.
[
1
2
]
B
.
S
r
i
n
i
v
a
s
a
n
d
G
.
S
.
R
a
o
,
“
A
h
y
b
r
i
d
C
N
N
-
K
N
N
m
o
d
e
l
f
o
r
M
R
I
b
r
a
i
n
t
u
mo
r
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Re
c
e
n
t
T
e
c
h
n
o
l
o
g
y
a
n
d
E
n
g
i
n
e
e
r
i
n
g
(
I
J
RTE)
,
v
o
l
.
8
,
n
o
.
2
,
p
p
.
5
2
3
0
–
5
2
3
5
,
J
u
l
.
2
0
1
9
,
d
o
i
:
1
0
.
3
5
9
4
0
/
i
j
r
t
e
.
B
1
0
5
1
.
0
7
8
2
1
9
.
[
1
3
]
M
.
D
.
N
a
n
d
e
e
sh
,
“
M
u
l
t
i
mo
d
a
l
i
ma
g
e
f
u
s
i
o
n
u
s
i
n
g
h
y
b
r
i
d
a
l
g
o
r
i
t
h
ms
f
o
r
b
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
,
”
2
0
2
1
I
E
EE
My
so
re
S
u
b
S
e
c
t
i
o
n
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
,
M
y
su
r
u
C
o
n
2
0
2
1
,
p
p
.
5
6
6
–
5
7
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
M
y
s
u
r
u
C
o
n
5
2
6
3
9
.
2
0
2
1
.
9
6
4
1
5
9
2
.
[
1
4
]
H
.
Za
i
d
i
,
M
.
L.
M
o
n
t
a
n
d
o
n
,
a
n
d
A
.
A
l
a
v
i
,
“
T
h
e
c
l
i
n
i
c
a
l
r
o
l
e
o
f
f
u
si
o
n
i
m
a
g
i
n
g
u
si
n
g
P
ET
,
C
T,
a
n
d
M
R
i
ma
g
i
n
g
,
”
PET
C
l
i
n
i
c
s
,
v
o
l
.
3
,
n
o
.
3
,
p
p
.
2
7
5
–
2
9
1
,
2
0
0
8
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
p
e
t
.
2
0
0
9
.
0
3
.
0
0
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
Leve
r
a
g
in
g
ima
g
e
fu
s
io
n
a
n
d
tr
a
n
s
fer lea
r
n
in
g
fo
r
en
h
a
n
ce
d
tu
mo
r
d
ia
g
n
o
s
is
(
N
ir
ma
la
jyo
t
h
i Na
r
is
etty
)
1813
[
1
5
]
M
.
H
.
A
b
d
a
l
l
a
t
i
f
,
H
.
E
i
ssa
,
a
n
d
L.
G
.
B
e
n
e
d
r
e
ss
,
“
W
a
v
e
l
e
t
t
r
a
n
sf
o
r
m
b
a
s
e
d
i
ma
g
e
f
u
s
i
o
n
f
o
r
b
r
a
i
n
t
u
mo
r
d
e
t
e
c
t
i
o
n
,
”
2
0
2
2
I
EE
E
2
n
d
I
n
t
e
r
n
a
t
i
o
n
a
l
M
a
g
h
re
b
M
e
e
t
i
n
g
o
f
t
h
e
C
o
n
f
e
re
n
c
e
o
n
S
c
i
e
n
c
e
s
a
n
d
T
e
c
h
n
i
q
u
e
s
o
f
Au
t
o
m
a
t
i
c
C
o
n
t
r
o
l
a
n
d
C
o
m
p
u
t
e
r
En
g
i
n
e
e
ri
n
g
,
MI
-
S
T
A
2
0
2
2
-
Pro
c
e
e
d
i
n
g
,
p
p
.
2
8
3
–
2
8
7
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
M
I
-
S
TA
5
4
8
6
1
.
2
0
2
2
.
9
8
3
7
6
5
5
.
[
1
6
]
A
.
K
u
m
a
r
,
“
S
t
u
d
y
a
n
d
a
n
a
l
y
s
i
s
o
f
d
i
f
f
e
r
e
n
t
s
e
g
me
n
t
a
t
i
o
n
m
e
t
h
o
d
s
f
o
r
b
r
a
i
n
t
u
m
o
r
M
R
I
a
p
p
l
i
c
a
t
i
o
n
,
”
M
u
l
t
i
m
e
d
i
a
T
o
o
l
s
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
8
2
,
n
o
.
5
,
p
p
.
7
1
1
7
–
7
1
3
9
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
0
4
2
-
0
2
2
-
1
3
6
3
6
-
y.
[
1
7
]
T.
L
o
g
e
sw
a
r
i
a
n
d
M
.
K
a
r
n
a
n
,
“
A
n
i
mp
r
o
v
e
d
i
mp
l
e
m
e
n
t
a
t
i
o
n
o
f
b
r
a
i
n
t
u
mo
r
d
e
t
e
c
t
i
o
n
u
si
n
g
s
o
f
t
c
o
m
p
u
t
i
n
g
,
”
2
n
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
m
m
u
n
i
c
a
t
i
o
n
S
o
f
t
w
a
re
a
n
d
N
e
t
w
o
r
k
s,
I
C
C
S
N
2
0
1
0
,
p
p
.
1
4
7
–
1
5
1
,
2
0
1
0
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
S
N
.
2
0
1
0
.
1
0
.
[
1
8
]
R
.
R
.
M
.
A
m
b
i
l
y
P
.
K
.
,
S
h
i
n
e
P
.
Jam
e
s,
“
B
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
u
si
n
g
i
ma
g
e
f
u
s
i
o
n
a
n
d
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
En
g
i
n
e
e
ri
n
g
Re
s
e
a
r
c
h
a
n
d
G
e
n
e
r
a
l
S
c
i
e
n
c
e
,
v
o
l
.
2
,
n
o
.
2
,
M
a
r
c
h
-
A
p
r
i
l
,
2
0
1
5
,
p
p
.
1
3
8
3
–
1
3
8
8
,
2
0
1
5
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
:
/
/
w
w
w
.
i
j
i
sm
e
.
o
r
g
/
w
p
-
c
o
n
t
e
n
t
/
u
p
l
o
a
d
s/
p
a
p
e
r
s
/
v
1
i
9
/
I
0
4
2
5
0
8
1
9
1
3
.
p
d
f
.
[
1
9
]
A
.
S
o
n
i
a
n
d
A
.
R
a
i
,
“
C
T
sca
n
b
a
se
d
b
r
a
i
n
t
u
m
o
r
r
e
c
o
g
n
i
t
i
o
n
a
n
d
e
x
t
r
a
c
t
i
o
n
u
si
n
g
p
r
e
w
i
t
t
a
n
d
m
o
r
p
h
o
l
o
g
i
c
a
l
d
i
l
a
t
i
o
n
,
”
2
0
2
1
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
e
r
C
o
m
m
u
n
i
c
a
t
i
o
n
a
n
d
I
n
f
o
rm
a
t
i
c
s,
I
C
C
C
I
2
0
2
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
C
I
5
0
8
2
6
.
2
0
2
1
.
9
4
0
2
6
7
7
.
[
2
0
]
M
.
Y
.
K
a
mi
l
,
“
B
r
a
i
n
t
u
m
o
r
a
r
e
a
c
a
l
c
u
l
a
t
i
o
n
u
si
n
g
m
o
r
p
h
o
l
o
g
i
c
a
l
o
p
e
r
a
t
i
o
n
s,”
I
O
S
R
J
o
u
rn
a
l
o
f
C
o
m
p
u
t
e
r
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
1
7
,
n
o
.
A
p
r
i
l
,
p
p
.
1
2
8
–
1
3
1
,
2
0
1
5
.
[
2
1
]
M
.
L.
R
a
h
m
a
n
,
A
.
W
.
R
e
z
a
,
a
n
d
S
.
I
.
S
h
a
b
u
j
,
“
A
n
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
-
b
a
se
d
a
u
t
o
ma
t
i
c
b
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
s
y
st
e
m,”
I
n
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
El
e
c
t
r
i
c
a
l
E
n
g
i
n
e
e
ri
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
5
,
n
o
.
1
,
p
p
.
2
1
4
–
2
2
2
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s
.
v
2
5
.
i
1
.
p
p
2
1
4
-
2
2
2
.
[
2
2
]
A
.
A
.
A
b
b
o
o
d
,
Q
.
M
.
S
h
a
l
l
a
l
,
a
n
d
M
.
A
.
F
a
d
h
e
l
,
“
A
u
t
o
ma
t
e
d
b
r
a
i
n
t
u
m
o
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
s
i
n
g
v
a
r
i
o
u
s
d
e
e
p
l
e
a
r
n
i
n
g
mo
d
e
l
s:
a
c
o
m
p
a
r
a
t
i
v
e
st
u
d
y
,
”
I
n
d
o
n
e
s
i
a
n
J
o
u
r
n
a
l
o
f
El
e
c
t
ri
c
a
l
E
n
g
i
n
e
e
ri
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
2
,
n
o
.
1
,
p
p
.
2
5
2
–
2
5
9
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s
.
v
2
2
.
i
1
.
p
p
2
5
2
-
2
5
9
.
[
2
3
]
H
.
E
l
H
a
m
d
a
o
u
i
e
t
a
l
.
,
"
H
i
g
h
p
r
e
c
i
si
o
n
b
r
a
i
n
t
u
mo
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
mo
d
e
l
b
a
se
d
o
n
d
e
e
p
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
a
n
d
s
t
a
c
k
i
n
g
c
o
n
c
e
p
t
s,"
I
n
d
o
n
e
si
a
n
J
o
u
r
n
a
l
o
f
E
l
e
c
t
r
i
c
a
l
En
g
i
n
e
e
r
i
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
4
,
n
o
.
1
,
p
p
.
1
6
7
-
1
7
7
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s.
v
2
4
.
i
1
.
p
p
1
6
7
-
1
7
7
.
[
2
4
]
C.
P
.
S
a
mj
i
t
h
R
a
j
a
n
d
R
.
S
h
r
e
e
j
a
,
“
A
u
t
o
m
a
t
i
c
b
r
a
i
n
t
u
mo
r
t
i
ss
u
e
d
e
t
e
c
t
i
o
n
i
n
T
-
1
w
e
i
g
h
t
e
d
M
R
I
,
”
Pro
c
e
e
d
i
n
g
s
o
f
2
0
1
7
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
I
n
n
o
v
a
t
i
o
n
s
i
n
I
n
f
o
rm
a
t
i
o
n
,
Em
b
e
d
d
e
d
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
S
y
st
e
m
s,
I
C
I
I
EC
S
2
0
1
7
,
v
o
l
.
2
0
1
8
-
Jan
u
a
r
y
,
p
p
.
1
–
4
,
2
0
1
7
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
I
I
EC
S
.
2
0
1
7
.
8
2
7
6
0
9
4
.
[
2
5
]
D
.
M
.
T
o
u
f
i
q
,
A
.
M
.
S
a
g
h
e
e
r
,
a
n
d
H
.
V
e
i
si
,
“
B
r
a
i
n
t
u
m
o
r
i
d
e
n
t
i
f
i
c
a
t
i
o
n
w
i
t
h
a
h
y
b
r
i
d
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
me
t
h
o
d
b
a
s
e
d
o
n
d
i
s
c
r
e
t
e
w
a
v
e
l
e
t
t
r
a
n
sf
o
r
m
a
n
d
p
r
i
n
c
i
p
l
e
c
o
mp
o
n
e
n
t
a
n
a
l
y
s
i
s,”
B
u
l
l
e
t
i
n
o
f
El
e
c
t
ri
c
a
l
En
g
i
n
e
e
ri
n
g
a
n
d
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
1
0
,
n
o
.
5
,
p
p
.
2
5
8
8
–
2
5
9
7
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
9
1
/
e
e
i
.
v
1
0
i
5
.
3
0
1
3
.
B
I
O
G
RAP
H
I
E
S
OF
A
UT
H
O
R
S
Dr
.
Nir
m
a
la
j
y
o
t
h
i
N
a
r
ise
tty
e
a
rn
e
d
h
e
r
D
o
c
to
ra
te
i
n
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
fro
m
Ac
h
a
ry
a
Na
g
a
rju
n
a
Un
iv
e
rsity
,
G
u
n
tu
r,
An
d
h
r
a
P
ra
d
e
sh
.
S
h
e
is
c
u
rre
n
tl
y
wo
rk
i
n
g
a
s
As
so
c
iate
P
ro
fe
ss
o
r
in
t
h
e
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
Ko
n
e
ru
Lak
s
h
m
a
iah
Ed
u
c
a
ti
o
n
F
o
u
n
d
a
ti
o
n
,
B
o
wra
m
p
e
t,
Hy
d
e
ra
b
a
d
-
5
0
0
0
4
3
,
Tela
n
g
a
n
a
,
In
d
ia
.
S
h
e
h
a
s
m
o
re
th
a
n
1
3
y
e
a
rs
o
f
tea
c
h
in
g
e
x
p
e
rien
c
e
a
n
d
3
.
5
y
e
a
rs
o
f
re
se
a
rc
h
e
x
p
e
rien
c
e
.
P
u
b
l
ish
e
d
1
8
re
se
a
rc
h
p
u
b
li
c
a
ti
o
n
s
i
n
re
p
u
ted
i
n
tern
a
ti
o
n
a
l
j
o
u
r
n
a
ls.
G
u
id
e
d
1
7
UG
p
ro
jec
ts
.
S
h
e
p
u
b
li
s
h
e
d
1
I
n
d
ian
P
a
ten
t.
H
e
r
re
se
a
r
c
h
in
tere
sts
in
c
lu
d
e
d
a
ta
m
in
in
g
,
m
a
c
h
in
e
lea
rn
in
g
,
d
e
e
p
lea
rn
in
g
,
a
n
d
c
lo
u
d
c
o
m
p
u
ti
n
g
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
n
irma
la.n
a
rise
tt
y
@g
m
a
il
.
c
o
m
.
Dr
.
K
unda
S
u
r
e
sh
B
a
b
u
As
so
c
iate
P
ro
fe
ss
o
r
i
n
t
h
e
De
p
a
rtm
e
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
,
Na
ra
sa
ra
o
p
e
t
En
g
in
e
e
rin
g
Co
ll
e
g
e
(A
u
to
n
o
m
o
u
s),
Na
ra
sa
ra
o
p
e
ta
-
AP,
h
e
h
a
s
m
o
re
th
a
n
1
0
y
e
a
rs
o
f
e
x
p
e
rien
c
e
.
He
h
a
s
p
u
b
l
ish
e
d
se
v
e
ra
l
re
se
a
rc
h
p
a
p
e
rs
i
n
re
p
u
ted
j
o
u
r
n
a
ls
a
n
d
c
o
n
fe
re
n
c
e
s.
He
is
a
m
e
m
b
e
r
o
f
IS
TE
li
fe
ti
m
e
.
He
h
a
s
c
o
m
p
lete
d
h
is
P
h
.
D
.
in
Co
m
p
u
ter S
c
ien
c
e
a
n
d
E
n
g
i
n
e
e
rin
g
,
VIT
-
AP
Un
i
v
e
rsity
,
Am
a
ra
v
a
ti
,
An
d
h
ra
P
ra
d
e
sh
,
In
d
ia
i
n
t
h
e
y
e
a
r
2
0
2
3
.
He
h
a
s
o
b
tain
e
d
h
is
M
.
Tec
h
fro
m
S
a
c
e
t,
V
e
tap
a
lem
,
Ch
irala
in
CS
E
in
th
e
y
e
a
r
2
0
1
0
.
C
o
m
p
lete
d
B.
Tec
h
fro
m
V.R.
S
&
Y.R.
N
Co
ll
e
g
e
,
Ch
irala
Affili
a
ted
to
I
J
NTUH
Un
iv
e
rsity
i
n
t
h
e
y
e
a
r
2
0
0
8
.
His
re
se
a
rc
h
in
tere
sts
a
r
e
m
o
b
il
e
c
o
m
p
u
ti
n
g
,
c
o
m
p
u
ter
n
e
two
rk
,
in
fo
rm
a
ti
o
n
se
c
u
ri
ty
,
m
a
c
h
in
e
lea
rn
in
g
,
d
e
e
p
lea
rn
in
g
,
a
n
d
c
y
b
e
r
se
c
u
rit
y
.
He
c
a
n
b
e
c
o
n
tac
te
d
a
t
e
m
a
il
:
su
re
sh
k
u
n
d
a
5
4
6
@
g
m
a
il
.
c
o
m
.
S
n
e
h
a
Ba
n
a
l
a
B
-
Tec
h
I
V
y
e
a
r,
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
-
Art
ifi
c
ial
In
telli
g
e
n
c
e
a
n
d
M
a
c
h
in
e
Lea
rn
in
g
,
Va
ll
u
ru
p
a
ll
i
Na
g
e
sw
a
ra
Ra
o
Vig
n
a
n
a
Jy
o
th
i
I
n
stit
u
te
o
f
En
g
in
e
e
rin
g
a
n
d
Tec
h
n
o
lo
g
y
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
sn
e
h
a
b
a
n
a
la2
8
1
1
@
g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.