I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
1
5
,
No
.
3
,
J
u
n
e
2
0
2
6
,
p
p
.
2337
~
2
3
4
6
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
15
.i
3
.
p
p
2
3
3
7
-
2
3
4
6
2337
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Bra
in t
umo
r d
ete
ction using
V
G
G
-
1
6
mo
del
Aicha
O
us
s
o
u
s
,
Abderra
hm
a
ne
E
z
-
Z
a
ho
ut,
So
um
ia
Z
it
i
D
e
p
a
r
t
me
n
t
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
,
F
a
c
u
l
t
y
o
f
S
c
i
e
n
c
e
s,
M
o
h
a
mm
e
d
V
U
n
i
v
e
r
si
t
y
i
n
R
a
b
a
t
,
R
a
b
a
t
,
M
o
r
o
c
c
o
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
2
9
,
2
0
2
4
R
ev
is
ed
Ma
r
2
,
2
0
2
6
Acc
ep
ted
Ap
r
2
0
,
2
0
2
6
Re
se
a
rc
h
in
m
e
d
ica
l
ima
g
e
a
n
a
ly
sis,
sp
e
c
ifi
c
a
ll
y
th
ro
u
g
h
d
e
e
p
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
two
rk
s,
a
d
d
re
ss
e
s
th
e
c
h
a
ll
e
n
g
e
s
o
f
m
a
n
u
a
ll
y
a
n
a
ly
z
i
n
g
larg
e
m
a
g
n
e
ti
c
re
so
n
a
n
c
e
ima
g
in
g
(
M
RI
)
ima
g
e
v
o
l
u
m
e
s
fo
r
b
ra
i
n
t
u
m
o
r
d
e
te
c
ti
o
n
.
Th
e
m
a
n
u
a
l
a
n
a
l
y
sis
is
ti
m
e
-
c
o
n
su
m
i
n
g
,
ted
i
o
u
s
,
a
n
d
p
r
o
n
e
to
i
n
a
c
c
u
ra
c
ies
d
u
e
to
su
b
tl
e
v
isu
a
l
sim
il
a
rit
ies
b
e
t
we
e
n
n
o
rm
a
l
ti
ss
u
e
a
n
d
tu
m
o
r
c
e
ll
s.
Th
is
re
se
a
rc
h
a
ims
to
a
u
t
o
m
a
te
tu
m
o
r
d
e
tec
ti
o
n
,
in
c
re
a
sin
g
a
c
c
u
ra
c
y
a
n
d
e
fficie
n
c
y
in
m
e
d
ica
l
trea
tme
n
ts.
Th
is
stu
d
y
a
ime
d
to
d
e
v
e
l
o
p
a
m
o
d
e
l
c
a
p
a
b
le
o
f
c
las
sify
in
g
b
ra
in
tu
m
o
rs
2
D
M
RI
ima
g
e
s,
a
n
d
th
e
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
two
r
k
(
CNN
)
-
b
a
se
d
m
o
d
e
l
su
c
c
e
ss
fu
ll
y
a
c
h
iev
e
d
a
n
a
c
c
u
ra
c
y
o
f
9
9
.
2
1
%
b
u
t
su
ffe
re
d
fr
o
m
n
o
ti
c
e
a
b
le
Ov
e
rfit
ti
n
g
.
Im
p
lem
e
n
ti
n
g
t
h
e
in
d
e
p
e
n
d
e
n
t
tes
ts
se
t
a
n
d
e
a
rly
sto
p
p
in
g
m
it
ig
a
ted
th
is
issu
e
,
m
a
k
in
g
th
e
m
o
d
e
l
m
o
re
re
li
a
b
le
f
o
r
p
r
o
d
u
c
ti
o
n
d
e
p
l
o
y
m
e
n
t
a
n
d
d
e
m
o
n
st
ra
ti
n
g
i
ts
p
o
ten
ti
a
l
i
n
su
p
p
o
r
ti
n
g
p
h
y
sic
i
a
n
s
in
d
e
tec
ti
n
g
b
ra
i
n
tu
m
o
rs
,
t
h
e
re
b
y
e
n
h
a
n
c
in
g
trea
tme
n
t
e
fficie
n
c
y
.
Th
e
u
se
o
f
P
y
th
o
n
,
Ten
s
o
rF
l
o
w,
a
n
d
Ke
ra
s
fa
c
il
it
a
ted
th
e
d
e
v
e
l
o
p
m
e
n
t
o
f
t
h
e
p
ro
p
o
se
d
so
l
u
ti
o
n
,
fo
c
u
sin
g
o
n
a
d
i
v
e
rse
se
t
o
f
M
RI
ima
g
e
s
with
v
a
ry
in
g
tu
m
o
r
siz
e
s,
lo
c
a
ti
o
n
s,
sh
a
p
e
s,
a
n
d
in
ten
siti
e
s.
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
m
o
d
el
Dee
p
lear
n
in
g
Ma
g
n
etic
r
eso
n
an
ce
im
a
g
in
g
Seg
m
en
tatio
n
VGG
-
16
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
:
Aich
a
Ou
s
s
o
u
s
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
,
Facu
lty
o
f
Scien
ce
s
,
Mo
h
am
m
ed
V
Un
iv
er
s
ity
in
R
ab
at
R
ab
at,
Mo
r
o
cc
o
E
m
ail
: a
ich
a_
o
u
s
s
o
u
s
@
u
m
5
.
a
c.
m
a
1.
I
NT
RO
D
UCT
I
O
N
Dee
p
lear
n
in
g
s
ig
n
if
ican
tly
i
m
p
ac
ts
th
e
m
ed
ical
f
ield
[
1
]
,
in
clu
d
in
g
th
e
d
e
v
elo
p
m
en
t
o
f
n
ew
d
r
u
g
s
,
im
p
r
o
v
e
d
clin
ical
d
ec
is
io
n
-
m
ak
in
g
,
an
d
in
n
o
v
ativ
e
m
e
d
icin
e
[
2
]
.
Me
d
ical
im
ag
in
g
,
wh
ich
in
v
o
lv
es
non
-
in
v
asiv
e
v
is
u
aliza
tio
n
o
f
th
e
b
o
d
y
'
s
in
ter
io
r
,
is
cr
itical
f
o
r
d
iag
n
o
s
is
an
d
tr
ea
tm
en
t
,
r
ely
in
g
o
n
im
ag
e
s
eg
m
en
tatio
n
to
en
h
a
n
ce
ef
f
ec
tiv
en
ess
[
3
]
.
I
n
co
n
tex
t
o
f
b
r
ai
n
tu
m
o
r
class
if
icatio
n
,
m
ed
ica
l
im
ag
e
p
r
o
ce
s
s
in
g
ev
alu
ates
3
D
d
atasets
f
r
o
m
c
o
m
p
u
ted
to
m
o
g
r
ap
h
y
(
CT
)
o
r
m
ag
n
etic
r
eso
n
an
ce
im
ag
in
g
(
MRI
)
s
ca
n
n
er
s
to
d
iag
n
o
s
e
d
is
ea
s
es,
p
lan
s
u
r
g
er
ies,
an
d
co
n
d
u
ct
r
esear
ch
.
R
ad
io
lo
g
is
ts
,
p
h
y
s
ician
s
,
an
d
en
g
in
ee
r
s
an
aly
ze
h
u
m
an
an
ato
m
y
t
h
r
o
u
g
h
m
ea
s
u
r
es,
s
tatis
tical
an
aly
s
is
,
an
d
s
im
u
latio
n
m
o
d
els
in
c
o
r
p
o
r
at
in
g
r
ea
l
an
ato
m
ical
g
eo
m
etr
ies,
en
h
a
n
cin
g
i
n
s
ig
h
ts
.
T
u
m
o
r
s
r
ef
e
r
to
ab
n
o
r
m
al
c
ell
g
r
o
wth
,
wh
ile
ca
n
ce
r
is
a
ty
p
e
o
f
tu
m
o
r
[
4
]
.
Dee
p
lear
n
in
g
h
as
d
em
o
n
s
tr
ated
p
r
o
m
is
in
g
r
esu
lts
in
au
to
m
ated
b
r
ain
tu
m
o
r
clas
s
if
icatio
n
,
en
h
an
cin
g
d
iag
n
o
s
tic
ac
cu
r
ac
y
an
d
f
ac
ilit
atin
g
tim
ely
i
n
ter
v
en
tio
n
s
[
5
]
,
[
6
]
.
Var
io
u
s
ar
c
h
itectu
r
es,
s
u
ch
as
co
n
v
o
l
u
tio
n
al
n
eu
r
al
n
etwo
r
k
s
(
C
NNs
)
, U
-
Net,
an
d
V
-
Net,
h
av
e
b
ee
n
ev
alu
ate
d
o
n
p
o
p
u
lar
d
atasets
lik
e
b
r
ain
tu
m
o
r
s
eg
m
e
n
tatio
n
c
h
allen
g
e
(
B
r
aT
S
)
,
in
ter
n
et
b
r
ain
s
eg
m
en
tatio
n
r
e
p
o
s
ito
r
y
(
I
B
SR
)
,
an
d
m
ed
ical
im
ag
e
co
m
p
u
tin
g
an
d
c
o
m
p
u
ter
ass
is
ted
in
ter
v
e
n
tio
n
(
MI
C
C
AI
)
[
7
]
,
[
8
]
.
Ho
wev
er
,
ch
allen
g
es
r
e
m
ain
,
in
clu
d
in
g
th
e
n
ee
d
f
o
r
lar
g
e
,
an
n
o
tated
d
at
asets
to
ad
d
r
ess
is
s
u
es
lik
e
c
lass
im
b
alan
ce
,
d
ata
s
ca
r
city
,
an
d
in
ter
-
s
ca
n
n
er
v
ar
iab
ilit
y
.
Fu
r
th
er
r
esear
ch
is
ess
en
tial
to
im
p
r
o
v
e
d
iag
n
o
s
tic
ac
cu
r
ac
y
a
n
d
clin
ical
d
ec
is
io
n
-
m
ak
in
g
in
n
eu
r
o
-
o
n
co
l
o
g
y
[
9
]
.
Key
c
h
a
llen
g
es
an
d
f
u
tu
r
e
d
ir
ec
tio
n
s
in
clu
d
e
cr
ea
tin
g
lar
g
e
-
s
ca
le,
m
u
lti
-
m
o
d
al,
an
d
an
n
o
tated
d
atasets
,
s
tan
d
ar
d
iz
in
g
ev
al
u
atio
n
m
etr
ics,
an
d
i
n
teg
r
atin
g
ex
p
lain
a
b
le
ar
tific
i
al
in
tellig
en
ce
in
to
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
3
,
J
u
n
e
2
0
2
6
:
2
3
3
7
-
2
3
4
6
2338
clin
ical
en
v
ir
o
n
m
en
ts
[
1
0
]
.
D
esp
ite
th
e
ch
allen
g
es
in
m
ed
ic
al
im
ag
e
an
aly
s
is
,
d
ee
p
lear
n
in
g
tec
h
n
iq
u
es
h
av
e
im
p
r
o
v
e
d
b
r
ain
tu
m
o
r
class
i
f
icatio
n
,
b
en
e
f
itin
g
p
atien
t
o
u
tco
m
es
b
y
in
c
r
ea
s
in
g
ac
cu
r
ac
y
an
d
r
e
d
u
cin
g
m
an
u
al
lab
o
r
.
T
h
ese
ad
v
a
n
c
em
en
ts
co
n
tr
ib
u
te
to
a
b
ette
r
u
n
d
er
s
tan
d
in
g
o
f
b
r
ain
tu
m
o
r
s
an
d
o
p
tim
ize
tr
ea
tm
en
t stra
teg
ies f
o
r
r
esear
ch
er
s
,
clin
ician
s
,
an
d
m
e
d
ical
im
ag
in
g
p
r
o
f
ess
io
n
als.
Gli
o
m
a
s
e
g
m
e
n
ta
ti
o
n
,
a
c
r
u
cia
l
as
p
ec
t
o
f
m
e
d
i
ca
l
im
a
g
e
p
r
o
ce
s
s
i
n
g
,
in
v
o
l
v
es
d
e
te
cti
n
g
t
u
m
o
r
s
i
n
t
h
e
b
r
ai
n
a
n
d
s
p
i
n
al
c
o
r
d
.
MRI
d
a
ta
d
e
r
i
v
e
d
f
r
o
m
cl
in
ica
l
s
c
an
s
o
r
a
r
ti
f
ic
ial
d
ata
b
ases
[
1
1
]
c
a
n
b
e
c
o
m
p
l
ex
d
u
e
to
v
a
r
i
ati
o
n
s
i
n
s
ca
n
n
i
n
g
e
q
u
i
p
m
e
n
t
a
n
d
t
ec
h
n
i
q
u
es,
r
es
u
l
ti
n
g
i
n
i
n
te
n
s
it
y
d
i
f
f
e
r
e
n
c
es
a
cr
o
s
s
i
m
a
g
es
.
T
h
e
ch
all
en
g
e
is
f
u
r
t
h
e
r
c
o
m
p
l
ic
ate
d
b
y
th
e
n
ec
ess
it
y
o
f
u
s
i
n
g
m
u
lti
p
le
m
o
d
a
liti
es
to
d
is
ti
n
g
u
is
h
tu
m
o
r
s
u
b
-
r
e
g
i
o
n
s
.
2.
RE
L
AT
E
D
WO
RK
Dee
k
s
h
a
et
a
l.
[
1
2
]
in
tr
o
d
u
c
e
a
C
NN
-
b
ased
m
o
d
el
f
o
r
c
lass
if
y
in
g
p
r
ev
alen
t
b
r
ain
tu
m
o
r
ty
p
es,
in
clu
d
in
g
m
en
in
g
i
o
m
as,
g
lio
m
as,
an
d
p
itu
itar
y
ad
e
n
o
m
as
,
u
s
in
g
MRI
im
ag
es.
T
h
e
m
o
d
el
is
co
n
s
tr
u
cted
u
s
in
g
Ker
as
an
d
T
e
n
s
o
r
Flo
w,
in
c
o
r
p
o
r
ati
n
g
lay
e
r
s
o
f
co
n
v
o
l
u
tio
n
,
p
o
o
lin
g
,
an
d
c
o
m
p
let
e
co
n
n
ec
tiv
ity
.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
was
tr
ain
ed
an
d
test
ed
o
n
a
d
ataset
o
f
2
5
3
MRI
s
ca
n
s
,
ac
h
iev
in
g
an
a
cc
u
r
ac
y
o
f
9
2
.
6
2
%.
T
h
e
C
NN
m
o
d
el'
s
p
er
f
o
r
m
an
ce
was
b
en
ch
m
ar
k
ed
a
g
ain
s
t
tr
ad
itio
n
al
class
if
ier
s
s
u
ch
as
k
-
n
ea
r
est
n
eig
h
b
o
r
s
(
KNN
)
,
d
ec
is
io
n
tr
ee
(
DT
)
,
an
d
r
an
d
o
m
f
o
r
ests
(
R
F)
,
with
t
h
e
C
NN
m
o
d
el
d
em
o
n
s
tr
atin
g
s
u
p
er
io
r
ac
cu
r
ac
y
.
Ad
d
itio
n
ally
,
th
e
a
u
th
o
r
s
d
ev
e
lo
p
ed
a
web
in
ter
f
ac
e
f
o
r
c
o
n
v
en
ien
t a
cc
ess
an
d
u
tili
za
tio
n
o
f
th
e
m
o
d
el.
Feb
r
ian
to
et
a
l.
[
1
3
]
c
o
n
d
u
ct
ed
a
s
tu
d
y
o
n
d
etec
tin
g
b
r
ain
tu
m
o
r
s
in
MRI
im
ag
es
u
s
in
g
C
NNs.
A
d
ataset
o
f
2
,
0
6
5
im
ag
es
was
d
iv
id
ed
in
to
7
0
%
tr
ain
in
g
,
1
5
%
test
in
g
,
an
d
1
5
%
v
alid
atio
n
.
T
wo
C
NN
m
o
d
els
wer
e
p
r
o
p
o
s
ed
a
n
d
c
o
m
p
ar
e
d
:
m
o
d
el
1
with
o
n
e
c
o
n
v
o
lu
tio
n
lay
er
an
d
m
o
d
el
2
with
two
co
n
v
o
l
u
tio
n
lay
e
r
s
.
Mo
d
el
2
o
u
tp
er
f
o
r
m
e
d
m
o
d
el
1
,
with
9
6
%
av
g
.
ac
cu
r
ac
y
o
n
tr
ain
in
g
d
ata,
9
3
%
o
n
test
d
ata,
an
d
an
F
1
-
s
co
r
e
o
f
9
2
%.
Ho
wev
er
,
m
o
d
el
2
to
o
k
lo
n
g
e
r
f
o
r
tr
ain
in
g
.
T
h
e
s
tu
d
y
co
n
f
ir
m
ed
C
NN's
ef
f
ec
tiv
en
ess
in
d
iag
n
o
s
in
g
b
r
ain
tu
m
o
r
s
(
9
3
%
ac
cu
r
ac
y
,
0
.
2
3
2
6
4
lo
s
s
)
an
d
em
p
h
asized
th
e
im
p
ac
t
o
f
co
n
v
o
lu
tio
n
lay
er
s
o
n
class
if
icatio
n
q
u
ality
.
I
m
a
g
e
a
u
g
m
en
tatio
n
i
n
cr
ea
s
ed
d
iv
e
r
s
ity
an
d
im
p
r
o
v
ed
class
if
icatio
n
r
esu
lts
.
T
h
e
s
tu
d
y
s
u
g
g
ested
th
at
u
s
in
g
m
o
r
e
i
m
ag
es a
n
d
f
o
cu
s
i
n
g
o
n
s
p
ec
if
ic
tu
m
o
r
ty
p
es c
o
u
ld
f
u
r
t
h
er
en
h
an
ce
class
if
icatio
n
p
er
f
o
r
m
a
n
ce
.
Gay
ath
r
i
et
a
l.
[
1
4
]
p
r
o
v
i
d
e
an
o
v
er
v
iew
o
f
v
ar
io
u
s
s
tu
d
i
es
th
at
h
av
e
u
tili
ze
d
C
NN
s
,
m
u
lti
-
lay
er
p
er
ce
p
tr
o
n
s
(
ML
Ps
)
,
an
d
m
u
l
tiv
ar
iab
le
r
eg
r
ess
io
n
an
d
n
eu
r
al
n
etwo
r
k
m
o
d
els
f
o
r
d
etec
ti
n
g
an
d
s
eg
m
e
n
tin
g
d
if
f
er
en
t
ty
p
es
o
f
ca
n
ce
r
s
,
in
cl
u
d
in
g
b
r
ain
tu
m
o
r
s
,
b
lad
d
e
r
ca
n
ce
r
,
an
d
lu
n
g
ca
n
ce
r
,
th
r
o
u
g
h
m
ed
ical
im
ag
in
g
.
T
h
e
r
ev
iew
ad
d
r
ess
es
th
e
ch
allen
g
es
an
d
lim
itatio
n
s
o
f
th
ese
tech
n
iq
u
es,
s
u
ch
as
t
h
e
r
eq
u
ir
e
m
en
t
f
o
r
d
o
m
ain
-
s
p
ec
if
ic
e
x
p
er
t
in
ter
p
r
etatio
n
,
s
ig
n
if
ican
t
a
n
ato
m
ical
v
ar
iatio
n
s
,
an
d
c
o
n
ce
r
n
s
r
elate
d
to
im
ag
e
q
u
ality
.
T
h
e
au
th
o
r
s
cu
s
to
m
ized
an
d
tr
ain
ed
v
is
u
al
g
eo
m
etr
y
g
r
o
u
p
1
6
-
lay
e
r
(
VGG
-
16
)
m
o
d
el
o
n
a
d
ataset
co
n
s
is
tin
g
o
f
1
,
6
5
5
b
r
ain
MR
I
im
ag
es
with
tu
m
o
r
s
an
d
1
,
5
9
8
tu
m
o
r
-
f
r
ee
im
ag
es,
ac
h
iev
in
g
9
4
%
ac
cu
r
ac
y
af
ter
h
y
p
e
r
p
ar
am
eter
o
p
tim
iz
atio
n
.
T
h
ey
also
ev
alu
ate
h
o
w
well
th
e
VGG
-
1
6
m
o
d
el
p
er
f
o
r
m
s
co
m
p
ar
ed
to
o
th
er
tech
n
i
q
u
es f
o
r
d
etec
tin
g
b
r
ain
tu
m
o
r
s
,
co
n
f
ir
m
in
g
its
ab
ilit
y
to
id
en
tify
t
h
ese
tu
m
o
r
s
ac
cu
r
ately
.
Ad
d
itio
n
ally
,
th
e
au
th
o
r
s
h
ig
h
lig
h
t
th
e
n
ec
ess
ity
o
f
ca
r
ef
u
l
e
v
alu
atio
n
an
d
u
s
e
o
f
d
ee
p
lear
n
in
g
to
o
ls
in
m
ed
ical
en
v
ir
o
n
m
en
ts
.
T
h
ey
em
p
h
asize
th
e
n
ee
d
to
co
n
s
id
er
eth
ical
co
n
ce
r
n
s
an
d
th
e
p
o
s
s
ib
le
r
ep
er
cu
s
s
io
n
s
o
f
f
alse
p
o
s
itiv
e
s
an
d
n
e
g
ativ
es
.
T
h
ey
also
s
tr
ess
th
e
im
p
o
r
tan
ce
o
f
c
r
ea
tin
g
clea
r
p
r
o
to
co
ls
f
o
r
th
e
in
co
r
p
o
r
atio
n
o
f
th
ese
to
o
l
s
in
to
h
ea
lth
ca
r
e
p
r
ac
tice.
Ma
h
m
u
d
et
a
l.
[
1
5
]
in
tr
o
d
u
c
e
a
C
NN
f
r
am
ewo
r
k
s
p
ec
if
i
ca
lly
tailo
r
ed
f
o
r
e
f
f
ec
tiv
e
b
r
ain
tu
m
o
r
id
en
tific
atio
n
u
s
in
g
MRI
.
T
h
i
s
f
r
am
ewo
r
k
is
co
m
p
ar
ed
to
a
n
u
m
b
er
o
f
o
th
er
m
o
d
els,
s
u
ch
as
R
esNet
-
5
0
,
VGG
-
1
6
,
an
d
I
n
ce
p
tio
n
-
V
3
.
E
v
alu
atio
n
is
b
ased
o
n
cr
iter
i
a
in
clu
d
in
g
ac
cu
r
ac
y
,
r
ec
all,
l
o
s
s
,
an
d
ar
ea
u
n
d
er
th
e
cu
r
v
e
(
AUC).
T
h
e
f
in
d
in
g
s
in
d
icate
th
at
th
e
s
u
g
g
ested
C
NN
f
r
am
ewo
r
k
o
u
ts
h
in
es
th
e
o
th
er
m
o
d
els
in
id
en
tify
in
g
b
r
ain
t
u
m
o
r
s
f
r
o
m
a
co
llectio
n
o
f
3
,
2
6
4
MR im
ag
es.
Gay
ath
r
i
an
d
Ku
m
ar
[
1
6
]
in
v
e
s
tig
ate
a
tr
an
s
f
er
lear
n
in
g
-
b
as
ed
VGG
-
1
6
m
o
d
el
f
o
r
t
h
e
s
eg
m
en
tatio
n
an
d
class
if
icatio
n
o
f
b
r
ai
n
tu
m
o
r
s
u
s
in
g
MRI
s
ca
n
s
.
T
h
is
an
aly
s
is
em
p
lo
y
s
th
e
B
r
aT
S
2
0
1
8
d
ataset.
T
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
ac
h
iev
es
a
n
ac
cu
r
ac
y
o
f
9
9
.
6
%,
9
5
.
3
5
%,
an
d
9
4
%
f
o
r
v
ar
i
o
u
s
tu
m
o
r
lo
ca
tio
n
s
,
ex
ce
ed
in
g
r
esu
lts
f
r
o
m
ea
r
lier
m
eth
o
d
s
.
I
t a
ls
o
attain
s
a
clas
s
if
icatio
n
a
cc
u
r
ac
y
o
f
9
9
.
6
%,
b
etter
th
an
co
m
p
etin
g
m
eth
o
d
s
lik
e
s
o
f
tm
ax
en
s
em
b
le
(
9
9
.
1
%)
an
d
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM
)
-
r
ad
ial
b
asis
f
u
n
ctio
n
(
R
B
F
)
(
9
7
.
7
0
%).
R
ag
h
u
v
an
s
h
i
an
d
Dh
a
r
iwal
[
1
7
]
an
aly
ze
th
e
e
f
f
ec
tiv
en
ess
o
f
C
NNs,
VGG
-
1
6
,
an
d
a
d
d
iti
o
n
al
d
ee
p
lear
n
in
g
m
o
d
els
in
d
etec
tin
g
b
r
ain
tu
m
o
r
s
th
r
o
u
g
h
MRI
im
ag
in
g
.
T
h
ey
also
d
is
cu
s
s
th
e
c
lin
ical
im
p
licatio
n
s
o
f
th
ese
tech
n
o
lo
g
ies,
em
p
h
as
izin
g
th
eir
p
o
ten
tial
to
aid
r
ad
i
o
lo
g
is
ts
in
b
r
ain
d
iag
n
o
s
tics
an
d
im
p
r
o
v
e
p
atien
t
o
u
tco
m
es.
T
h
e
a
u
th
o
r
s
s
tr
ess
t
h
e
cr
itical
r
o
le
o
f
d
ee
p
lear
n
in
g
in
th
e
m
e
d
ical
f
ield
an
d
its
s
ig
n
if
ican
t in
f
lu
e
n
ce
o
n
b
r
ai
n
tu
m
o
r
r
esear
ch
.
I
s
m
ail
et
a
l.
[
1
8
]
r
ep
o
r
t
o
n
t
h
e
u
s
e
o
f
th
e
VGG
-
1
6
s
tr
u
ctu
r
e
o
f
C
NNs
f
o
r
ca
te
g
o
r
izin
g
m
ed
ica
l
im
ag
es,
s
p
ec
if
ically
tar
g
etin
g
d
atasets
r
elate
d
to
b
r
ain
t
u
m
o
r
s
an
d
Alzh
eim
e
r
'
s
d
is
ea
s
e.
T
h
ey
h
ig
h
lig
h
t
th
e
im
p
o
r
tan
ce
o
f
class
if
y
in
g
m
ed
ical
im
ag
es
an
d
h
o
w
C
NNs,
p
ar
ticu
lar
ly
VGG
-
1
6
,
ca
n
im
p
r
o
v
e
p
r
ec
is
io
n
an
d
d
e
p
en
d
a
b
ilit
y
in
th
is
ar
ea
.
T
wo
d
atasets
wer
e
cr
ea
t
ed
,
with
d
ata
a
u
g
m
e
n
tatio
n
tech
n
iq
u
es
s
u
ch
as
r
o
tatio
n
,
s
ca
lin
g
,
a
n
d
f
lip
p
in
g
em
p
lo
y
ed
to
in
cr
ea
s
e
th
e
d
ata
s
et
s
ize
an
d
c
o
u
n
ter
class
im
b
alan
ce
s
.
T
h
e
r
esu
lts
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
B
r
a
in
tu
mo
r
d
etec
tio
n
u
s
in
g
V
GG
-
1
6
mo
d
el
(
A
ich
a
Ou
s
s
o
u
s
)
2339
s
h
o
w
th
at
th
e
m
o
d
el
is
v
e
r
y
e
f
f
ec
tiv
e
at
d
etec
tin
g
ir
r
eg
u
lar
iti
es
in
m
ed
ical
im
ag
es,
esp
ec
ial
ly
with
in
th
e
b
r
ain
tu
m
o
r
d
ataset.
3.
M
E
T
H
O
D
T
h
e
m
eth
o
d
o
lo
g
y
co
n
s
is
ts
o
f
s
ev
er
al
k
ey
s
tag
es.
I
n
itially
,
we
ac
q
u
ir
ed
a
p
u
b
licly
av
ailab
le
d
ataset
o
f
b
in
ar
y
-
class
b
r
ai
n
MRI
im
ag
es
f
r
o
m
t
h
e
Kag
g
le
p
latf
o
r
m
,
wh
ich
in
cl
u
d
es
a
t
o
tal
o
f
2
5
3
im
ag
es:
1
5
5
wit
h
tu
m
o
r
s
an
d
9
8
with
o
u
t.
Alth
o
u
g
h
th
e
o
r
ig
in
al
im
a
g
es
v
ar
ied
in
s
ize,
we
s
tan
d
ar
d
ized
th
em
to
a
u
n
if
o
r
m
d
im
en
s
io
n
b
e
f
o
r
e
f
ee
d
in
g
th
e
m
in
to
th
e
m
o
d
el
f
o
r
tr
ai
n
in
g
.
All
im
ag
es
ar
e
o
f
h
ig
h
q
u
ality
,
an
d
Fig
u
r
e
1
illu
s
tr
ates
s
ix
ex
am
p
les
f
r
o
m
th
e
d
ataset.
Fo
llo
win
g
th
is
,
w
e
p
r
e
-
p
r
o
ce
s
s
ed
th
e
d
atasets
t
o
p
r
ep
ar
e
th
e
m
f
o
r
an
aly
s
is
.
W
e
em
p
lo
y
ed
th
e
VGG
-
1
6
m
o
d
el
as
o
u
r
d
ee
p
C
NN
ar
ch
itectu
r
e
to
tr
ain
th
e
im
a
g
es.
T
o
v
alid
ate
o
u
r
r
esu
lts
,
we
ass
e
s
s
ed
p
er
f
o
r
m
a
n
ce
m
etr
ics s
u
ch
as a
cc
u
r
ac
y
a
n
d
lo
s
s
.
Fig
u
r
e
1
.
B
r
ain
tu
m
o
r
MRI
im
ag
es
3
.
1
.
Da
t
a
s
et
co
llect
i
o
n a
nd
pre
-
pro
ce
s
s
ing
Fo
r
th
is
wo
r
k
,
th
e
d
ataset
o
f
b
r
ain
MRI
im
ag
es
is
u
tili
ze
d
f
o
r
d
etec
tin
g
b
r
ain
tu
m
o
r
s
.
I
n
to
tal,
we
g
ath
er
ed
2
5
3
im
ag
es,
wh
ich
in
clu
d
ed
1
5
5
with
tu
m
o
r
s
an
d
9
8
with
o
u
t,
f
ea
tu
r
in
g
a
r
an
g
e
o
f
tu
m
o
r
s
izes,
s
h
ap
es,
lo
ca
tio
n
s
,
an
d
in
ten
s
ities
.
T
h
is
d
ata
s
et
aim
s
to
d
is
tin
g
u
is
h
b
etwe
en
im
ag
es
co
n
tain
in
g
tu
m
o
r
s
an
d
th
o
s
e
with
h
ea
lth
y
tis
s
u
e.
T
o
m
itig
ate
o
v
er
f
itti
n
g
an
d
en
h
an
ce
t
h
e
m
o
d
el'
s
ef
f
ec
tiv
en
ess
,
we
em
p
lo
y
e
d
d
ata
au
g
m
en
tatio
n
s
tr
ateg
ies.
Data
au
g
m
en
tatio
n
in
v
o
lv
es
a
p
p
ly
in
g
r
an
d
o
m
m
o
d
if
icatio
n
s
to
t
h
e
tr
ain
in
g
im
ag
es,
lead
in
g
t
o
th
e
cr
ea
tio
n
o
f
n
ew
d
ata
v
ar
iatio
n
s
.
B
y
u
s
in
g
th
ese
tech
n
iq
u
es,
we
en
h
an
ce
d
b
o
t
h
th
e
d
iv
er
s
ity
an
d
s
ize
o
f
th
e
d
ataset,
th
er
eb
y
lo
wer
in
g
th
e
ch
an
ce
o
f
o
v
er
f
itti
n
g
[
1
9
]
.
I
n
th
is
m
eth
o
d
,
th
e
I
m
a
g
eDa
taGe
n
er
ato
r
class
f
r
o
m
Ker
as
was
u
tili
ze
d
f
o
r
d
ata
a
u
g
m
en
tatio
n
.
V
ar
io
u
s
p
ar
a
m
e
ter
s
wer
e
s
et
to
s
p
ec
if
y
th
e
ty
p
es
an
d
lev
els
o
f
tr
an
s
f
o
r
m
atio
n
s
to
b
e
ap
p
lied
,
s
u
ch
as
r
o
tatio
n
,
s
h
if
tin
g
in
wid
th
an
d
h
eig
h
t,
r
escalin
g
,
s
h
ea
r
in
g
,
ad
ju
s
tin
g
b
r
ig
h
tn
ess
,
an
d
f
lip
p
in
g
im
ag
es
b
o
th
h
o
r
izo
n
tally
an
d
v
er
tically
.
B
y
r
a
n
d
o
m
l
y
i
m
p
lem
en
tin
g
th
ese
m
o
d
if
icatio
n
s
o
n
th
e
tr
ain
in
g
im
ag
es,
t
h
e
m
o
d
el
b
ec
o
m
es
m
o
r
e
r
o
b
u
s
t
a
n
d
b
etter
e
q
u
ip
p
e
d
to
m
an
ag
e
v
ar
iatio
n
s
f
ac
ed
in
r
ea
l
-
life
s
itu
atio
n
s
.
3
.
2
.
Dee
p
lea
rning
mo
dels
T
h
e
C
NN
m
o
d
el
is
o
n
e
o
f
th
e
to
p
tech
n
iq
u
es
in
m
ed
ical
im
ag
e
an
aly
s
is
is
th
e
C
NN
.
T
h
is
s
p
ec
if
ic
k
in
d
o
f
a
r
tific
ial
n
eu
r
al
n
et
wo
r
k
is
aim
ed
at
r
ec
o
g
n
izin
g
an
d
p
r
o
ce
s
s
in
g
im
ag
es,
m
ak
in
g
it
p
ar
ticu
lar
ly
ef
f
ec
tiv
e
f
o
r
task
s
th
at
in
v
o
l
v
e
u
n
d
er
s
tan
d
in
g
co
m
p
lex
v
is
u
al
elem
en
ts
.
C
NN
s
ca
n
au
to
m
atica
lly
ca
teg
o
r
ize
b
r
ain
t
u
m
o
r
s
th
r
o
u
g
h
a
s
y
s
tem
atic
m
eth
o
d
th
at
in
v
o
lv
es
b
o
th
tr
ain
i
n
g
an
d
v
alid
atio
n
s
ta
g
es.
I
n
th
is
r
esear
c
h
,
th
e
b
r
ai
n
im
ag
i
n
g
d
ataset
was so
u
r
ce
d
f
r
o
m
th
e
p
u
b
licly
a
v
ailab
le
co
llectio
n
o
n
Kag
g
le.
T
h
e
d
ec
is
io
n
t
o
u
s
e
a
C
NN
f
o
r
o
u
r
m
o
d
el
co
m
es
f
r
o
m
its
in
ter
co
n
n
ec
ted
p
a
r
a
m
eter
s
an
d
s
h
ar
ed
f
ea
tu
r
es,
wh
ich
im
p
r
o
v
e
its
ef
f
ec
tiv
en
ess
in
im
ag
e
class
if
icatio
n
task
s
.
Dr
awin
g
in
s
p
ir
ati
o
n
f
r
o
m
th
e
m
en
tal
f
u
n
ctio
n
s
o
f
cr
ea
tiv
e
an
im
als
an
d
s
o
p
h
is
ticated
co
g
n
itiv
e
ab
ilit
ies,
C
NNs
ex
em
p
lify
an
in
n
o
v
ativ
e
ca
teg
o
r
y
o
f
d
e
ep
n
eu
r
al
n
etwo
r
k
s
(
DNNs)
.
T
h
e
d
esig
n
o
f
a
C
NN
is
m
ad
e
u
p
o
f
a
lay
er
e
d
s
tr
u
ctu
r
e
th
at
in
clu
d
es
f
u
lly
co
n
n
ec
te
d
lay
er
s
,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
3
,
J
u
n
e
2
0
2
6
:
2
3
3
7
-
2
3
4
6
2340
co
n
v
o
l
u
tio
n
al
lay
er
s
,
p
o
o
lin
g
lay
er
s
,
f
latten
in
g
lay
e
r
s
,
an
d
o
u
tp
u
t
lay
e
r
s
.
E
ac
h
C
NN
h
as
a
d
is
tin
ctiv
e
s
et
o
f
im
ag
es
th
at
v
ar
y
in
n
u
m
b
er
,
s
ize,
an
d
ty
p
e,
alo
n
g
with
d
if
f
er
en
t
ac
tiv
atio
n
f
u
n
ctio
n
s
.
T
h
e
s
ettin
g
s
o
f
th
e
C
NN
ar
e
estab
lis
h
ed
th
r
o
u
g
h
test
in
g
an
d
e
x
p
er
im
e
n
tal
ad
ju
s
tm
e
n
ts
.
Fu
n
d
am
en
tally
,
a
C
NN
co
n
tain
s
n
u
m
e
r
o
u
s
lay
er
s
,
with
th
e
m
ain
o
n
es
b
ei
n
g
th
e
co
n
v
o
lu
tio
n
al
lay
er
an
d
th
e
s
u
b
s
am
p
lin
g
lay
er
.
T
h
e
co
n
v
o
l
u
tio
n
al
lay
er
d
etec
ts
f
ea
tu
r
es
in
th
e
im
ag
e,
en
ab
lin
g
th
e
n
etwo
r
k
to
ca
p
t
u
r
e
cr
itical
attr
ib
u
tes
th
at
r
em
ain
s
im
ilar
ac
r
o
s
s
d
if
f
er
en
t
p
ar
ts
o
f
th
e
im
ag
e.
T
h
is
lay
er
b
o
o
s
ts
th
e
m
o
d
el'
s
ca
p
ab
ilit
y
to
id
en
tify
p
atter
n
s
an
d
ch
a
r
ac
ter
is
tics
,
wh
ich
ar
e
th
en
u
tili
ze
d
f
o
r
cla
s
s
if
icatio
n
.
C
o
n
v
er
s
ely
,
th
e
s
u
b
s
am
p
lin
g
lay
e
r
d
e
cr
ea
s
es
th
e
d
im
en
s
io
n
s
o
f
t
h
e
in
p
u
t
im
ag
e,
h
elp
in
g
t
o
f
o
r
m
a
v
ec
to
r
r
e
p
r
esen
tatio
n
as
th
e
in
f
o
r
m
atio
n
m
o
v
es
th
r
o
u
g
h
th
e
C
NN.
T
h
is
p
r
o
ce
d
u
r
e,
k
n
o
wn
as
p
o
o
lin
g
,
co
n
d
en
s
es
th
e
d
ata
wh
ile
p
r
eser
v
in
g
ess
en
tial
f
ea
tu
r
es
n
ee
d
ed
f
o
r
p
r
ec
is
e
class
if
icatio
n
[
2
0
]
.
T
h
e
VGG
-
1
6
m
o
d
el
is
an
ad
v
an
ce
d
d
ee
p
C
NN
th
at
co
n
s
is
ts
o
f
m
u
ltip
le
lay
er
s
,
s
u
ch
as c
o
n
v
o
lu
tio
n
al
an
d
f
u
lly
co
n
n
ec
ted
lay
er
s
,
ar
r
an
g
e
d
in
a
s
eq
u
e
n
tial
m
a
n
n
er
t
o
p
r
o
ce
s
s
an
d
class
if
y
MRI
im
ag
es.
T
h
is
ar
ch
itectu
r
e
en
a
b
les
th
e
m
o
d
el
to
id
en
tify
in
tr
icate
f
ea
tu
r
e
s
f
r
o
m
in
p
u
t
im
ag
es,
wh
ich
a
id
s
in
th
e
ac
cu
r
ate
id
en
tific
atio
n
an
d
class
if
icatio
n
o
f
b
r
ain
tu
m
o
r
s
[
2
1
]
.
VGG
-
1
6
co
n
tain
s
a
to
tal
o
f
1
6
la
y
er
s
,
wh
ich
in
clu
d
e
1
3
co
n
v
o
lu
tio
n
al
lay
e
r
s
an
d
3
f
u
lly
co
n
n
ec
ted
lay
er
s
.
T
h
e
c
o
n
v
o
lu
tio
n
al
lay
er
s
u
tili
ze
s
m
all
f
ilter
s
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
th
r
o
u
g
h
co
n
v
o
l
u
tio
n
,
ef
f
ec
tiv
ely
ca
p
tu
r
i
n
g
s
p
atial
ar
r
an
g
e
m
en
ts
an
d
l
o
ca
l
ch
ar
a
cter
is
tics
with
in
th
e
im
ag
es.
T
h
is
allo
ws
th
e
m
o
d
el
to
u
n
d
er
s
tan
d
r
ep
r
esen
tatio
n
s
at
d
if
f
e
r
en
t
lev
els
o
f
d
etail.
At
th
e
co
n
cl
u
s
io
n
o
f
th
e
n
etwo
r
k
,
th
e
f
u
lly
co
n
n
ec
t
ed
lay
er
s
,
also
k
n
o
wn
as
d
en
s
e
lay
er
s
,
p
er
f
o
r
m
class
if
icatio
n
task
s
b
ased
o
n
th
e
f
ea
tu
r
es
o
b
tain
ed
f
r
o
m
th
e
co
n
v
o
lu
tio
n
al
lay
er
s
.
T
h
e
y
c
o
m
b
in
e
th
e
ex
tr
ac
te
d
f
ea
t
u
r
es
to
m
ak
e
p
r
ed
ictio
n
s
o
n
wh
eth
er
b
r
ai
n
tu
m
o
r
s
ar
e
p
r
es
en
t
in
th
e
MRI
im
ag
es.
B
y
la
y
er
in
g
s
ev
er
al
c
o
n
v
o
lu
tio
n
al
a
n
d
f
u
lly
co
n
n
ec
ted
lay
er
s
,
th
e
VGG
-
1
6
m
o
d
el
s
k
illfu
lly
ca
p
tu
r
es
s
u
b
tle
d
etails
an
d
co
m
p
le
x
co
n
n
ec
tio
n
s
in
b
r
ain
tu
m
o
r
im
ag
e
r
y
,
lead
in
g
to
im
p
r
ess
iv
e
ac
cu
r
ac
y
in
tu
m
o
r
d
etec
tio
n
.
T
h
e
s
tr
u
ctu
r
e
o
f
th
e
p
r
o
p
o
s
ed
n
etwo
r
k
m
o
d
el
is
illu
s
tr
ated
in
Fig
u
r
e
2
.
Fig
u
r
e
2
.
T
h
e
p
r
o
p
o
s
ed
n
etwo
r
k
m
o
d
el'
s
s
tr
u
ctu
r
e
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
A
tr
an
s
f
er
lear
n
in
g
ap
p
r
o
ac
h
u
s
in
g
a
VGG
-
1
6
m
o
d
el
p
r
e
-
tr
ain
ed
is
em
p
lo
y
e
d
o
n
th
e
I
m
ag
eNe
t
d
atab
ase
to
class
if
y
MRI
im
ag
es
f
o
r
b
r
ain
t
u
m
o
r
d
ete
ctio
n
.
T
h
e
m
eth
o
d
o
lo
g
y
in
v
o
lv
ed
r
etr
ie
v
in
g
t
h
e
p
r
e
-
tr
ain
ed
VGG
-
1
6
m
o
d
el
to
ex
tr
ac
t
f
ea
tu
r
es
f
r
o
m
MRI
im
ag
es,
th
en
r
ep
lacin
g
t
h
e
f
in
al
class
if
icatio
n
lay
er
with
a
b
in
ar
y
o
u
tp
u
t
lay
er
f
o
r
"tu
m
o
r
p
r
esen
t/ab
s
en
t"
cl
ass
if
icatio
n
wh
ile
r
etain
in
g
th
e
o
th
er
la
y
er
s
to
p
r
eser
v
e
ac
q
u
ir
ed
k
n
o
wled
g
e
.
T
h
e
cu
s
to
m
ar
ch
itectu
r
e
ad
d
ed
a
f
latten
lay
er
to
tr
a
n
s
f
o
r
m
th
e
d
ata
in
to
a
v
ec
to
r
,
f
o
llo
wed
b
y
a
d
en
s
e
la
y
er
with
a
s
o
f
tm
ax
ac
tiv
atio
n
f
u
n
ctio
n
f
o
r
b
i
n
ar
y
class
if
icati
o
n
.
B
ef
o
r
e
tr
ain
i
n
g
,
MRI
im
ag
es
wer
e
p
r
ep
r
o
ce
s
s
ed
b
y
r
esizin
g
to
2
2
4
×
2
2
4
p
ix
els
an
d
e
n
h
an
ce
d
u
s
in
g
d
ata
au
g
m
en
tatio
n
tech
n
iq
u
es,
in
clu
d
in
g
r
o
tatio
n
,
s
h
if
tin
g
,
f
lip
p
in
g
,
an
d
b
r
ig
h
tn
ess
ad
ju
s
tm
en
t,
to
im
p
r
o
v
e
class
if
icatio
n
p
er
f
o
r
m
an
ce
a
n
d
p
r
ev
en
t
o
v
er
f
itti
n
g
.
T
h
e
m
o
d
el
was
co
n
f
ig
u
r
e
d
with
t
h
e
Ad
am
o
p
tim
izer
,
ca
teg
o
r
ical
cr
o
s
s
-
en
tr
o
p
y
as
th
e
lo
s
s
f
u
n
c
tio
n
,
an
d
ac
cu
r
ac
y
as
th
e
p
r
i
m
ar
y
ev
alu
atio
n
m
etr
ic.
T
r
ain
in
g
was
co
n
d
u
cted
o
v
er
1
0
ep
o
c
h
s
with
2
6
tr
ai
n
in
g
s
tep
s
,
ac
h
iev
in
g
a
n
im
p
r
ess
iv
e
ac
cu
r
ac
y
o
f
9
9
.
2
1
%.
T
h
is
r
e
p
r
esen
ts
a
s
u
b
s
tan
tial a
d
v
an
ce
m
en
t in
g
r
ad
ien
t d
escen
t o
p
tim
izatio
n
te
ch
n
iq
u
es,
with
th
e
r
esu
lts
illu
s
tr
ated
in
Fig
u
r
e
3
.
T
h
e
m
o
d
el
was
tr
ain
ed
f
o
r
1
0
ep
o
ch
s
,
ea
ch
co
m
p
r
is
in
g
2
6
s
tep
s
an
d
tak
in
g
7
1
to
9
3
s
ec
o
n
d
s
to
co
m
p
lete.
T
r
ain
in
g
lo
s
s
d
ec
r
ea
s
ed
s
ig
n
if
ican
tly
f
r
o
m
0
.
9
7
8
7
in
th
e
f
ir
s
t
ep
o
ch
to
0
.
1
2
7
6
b
y
th
e
ten
th
ep
o
ch
,
wh
ile
tr
ain
in
g
ac
cu
r
ac
y
im
p
r
o
v
ed
f
r
o
m
6
9
.
1
7
%
to
9
4
.
0
7
%
.
Valid
atio
n
m
etr
ics
also
s
h
o
wed
p
o
s
itiv
e
tr
en
d
s
,
with
v
alid
atio
n
lo
s
s
d
r
o
p
p
i
n
g
f
r
o
m
0
.
7
6
8
3
to
0
.
0
4
9
6
an
d
v
a
lid
atio
n
ac
cu
r
ac
y
r
is
in
g
f
r
o
m
6
6
.
8
0
%
to
9
9
.
2
%.
No
tab
ly
,
v
alid
atio
n
ac
cu
r
ac
y
co
n
s
is
ten
tly
s
u
r
p
ass
ed
tr
ain
in
g
ac
cu
r
ac
y
,
s
u
g
g
esti
n
g
ef
f
ec
tiv
e
d
ata
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
B
r
a
in
tu
mo
r
d
etec
tio
n
u
s
in
g
V
GG
-
1
6
mo
d
el
(
A
ich
a
Ou
s
s
o
u
s
)
2341
au
g
m
en
tatio
n
,
p
o
te
n
tial
o
v
er
f
itti
n
g
,
o
r
is
s
u
es
with
d
ataset
s
p
litt
in
g
.
Desp
ite
a
m
in
o
r
f
lu
ct
u
atio
n
in
e
p
o
ch
9
,
th
e
m
o
d
el
s
tab
ilized
b
y
ep
o
c
h
1
0
,
ac
h
ie
v
in
g
h
ig
h
p
er
f
o
r
m
a
n
ce
m
etr
ics
o
v
er
all.
Ho
wev
er
,
th
e
u
n
u
s
u
ally
h
i
g
h
v
alid
atio
n
ac
cu
r
ac
y
r
aises
co
n
ce
r
n
s
ab
o
u
t
p
o
s
s
ib
le
o
v
er
f
itti
n
g
o
r
d
ata
leak
ag
e.
Fig
u
r
e
3
.
Ou
tc
o
m
e
g
e
n
er
ated
b
y
o
u
r
s
u
g
g
ested
a
p
p
r
o
ac
h
I
n
Fig
u
r
es 4
an
d
5
,
th
e
an
aly
s
is
o
f
th
e
tr
ain
in
g
an
d
v
alid
atio
n
lo
s
s
es
r
ev
ea
ls
a
p
o
s
it
iv
e
lear
n
in
g
tr
en
d
,
with
b
o
th
m
etr
ics
d
ec
r
ea
s
in
g
o
v
er
tim
e,
p
ar
ticu
lar
ly
s
teep
ly
in
t
h
e
in
itial
ep
o
c
h
s
b
ef
o
r
e
lev
elin
g
o
f
f
.
I
n
itially
,
tr
ain
in
g
lo
s
s
was
h
ig
h
er
th
an
v
alid
atio
n
lo
s
s
,
b
u
t
th
ey
c
o
n
v
er
g
ed
ar
o
u
n
d
e
p
o
ch
6
,
with
v
alid
atio
n
lo
s
s
b
ec
o
m
in
g
s
lig
h
tly
lo
wer
th
er
e
af
ter
.
Ho
wev
er
,
a
n
o
tab
le
s
p
i
k
e
in
tr
ain
i
n
g
l
o
s
s
ar
o
u
n
d
ep
o
ch
8
s
u
g
g
ests
s
o
m
e
in
s
tab
ilit
y
,
p
o
s
s
ib
ly
d
u
e
to
a
ch
allen
g
in
g
b
atc
h
o
f
tr
ain
i
n
g
d
ata.
I
n
ter
m
s
o
f
ac
c
u
r
ac
y
,
b
o
th
tr
ain
in
g
,
an
d
v
alid
atio
n
m
etr
ics
in
cr
ea
s
ed
as
ex
p
ec
ted
,
b
u
t
v
alid
atio
n
a
cc
u
r
ac
y
co
n
s
is
ten
tly
ex
ce
ed
e
d
tr
ain
in
g
ac
c
u
r
ac
y
,
wh
ich
is
u
n
u
s
u
al
an
d
m
a
y
in
d
icate
p
ec
u
liar
ities
in
th
e
d
atase
t
o
r
t
h
e
d
ata
au
g
m
en
tatio
n
p
r
o
ce
s
s
.
B
y
th
e
en
d
o
f
th
e
tr
ain
in
g
,
b
o
th
ac
cu
r
ac
ies
wer
e
im
p
r
ess
iv
ely
h
ig
h
,
with
v
alid
atio
n
ac
cu
r
ac
y
a
p
p
r
o
a
ch
in
g
1
0
0
%.
T
h
is
in
d
icate
s
clea
r
lear
n
i
n
g
p
r
o
g
r
e
s
s
an
d
g
o
o
d
g
en
e
r
aliza
tio
n
,
as e
v
id
en
ce
d
b
y
th
e
clo
s
e
alig
n
m
en
t
o
f
tr
ain
in
g
an
d
v
alid
atio
n
m
etr
ics.
Ho
wev
er
,
th
e
co
n
s
is
ten
tly
h
ig
h
er
v
alid
atio
n
ac
cu
r
ac
y
r
aises
co
n
ce
r
n
s
ab
o
u
t
p
o
ten
tial
o
v
er
f
itti
n
g
o
r
d
ata
leak
ag
e,
w
ar
r
an
tin
g
f
u
r
th
er
in
v
esti
g
atio
n
.
T
h
e
s
p
ik
e
i
n
tr
ain
i
n
g
lo
s
s
also
h
ig
h
lig
h
ts
a
n
ee
d
to
ass
ess
m
o
d
el
s
tab
ilit
y
.
T
o
m
itig
ate
th
e
r
is
k
s
o
f
o
v
er
f
itti
n
g
,
wh
ich
ar
is
e
f
r
o
m
t
h
e
m
o
d
el
m
e
m
o
r
izin
g
f
ea
tu
r
es
o
f
th
e
tr
ain
i
n
g
s
et
r
ath
er
th
an
lear
n
i
n
g
th
em
,
an
d
f
r
o
m
a
n
im
b
alan
ce
o
r
in
s
u
f
f
icien
t
s
ize
in
th
e
v
alid
atio
n
s
et,
we
f
ir
s
t p
er
f
o
r
m
ed
a
s
et
o
f
in
d
ep
en
d
en
t te
s
ts
an
d
th
en
ap
p
lied
ea
r
ly
s
to
p
p
i
n
g
.
Fig
u
r
e
4
.
T
r
ain
in
g
a
n
d
v
alid
atio
n
lo
s
s
es
Fig
u
r
e
5
.
T
r
ain
in
g
a
n
d
v
alid
atio
n
ac
cu
r
ac
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
3
,
J
u
n
e
2
0
2
6
:
2
3
3
7
-
2
3
4
6
2342
T
h
e
d
a
t
a
s
e
t
i
s
d
i
v
id
ed
in
t
o
a
t
r
a
i
n
in
g
s
e
t
,
a
t
e
s
t
s
e
t
,
an
d
a
v
a
l
i
d
a
t
io
n
s
e
t
.
T
h
e
n
p
e
r
f
o
r
m
e
d
a
s
e
t
o
f
i
n
d
ep
e
n
d
en
t
t
e
s
t
s
.
T
h
e
n
ew
r
e
s
u
l
t
s
,
a
s
i
l
l
u
s
t
r
a
t
ed
i
n
F
i
g
u
r
e
s
6
an
d
7
,
d
em
o
n
s
t
r
a
t
e
a
s
i
g
n
if
i
c
a
n
t
i
m
p
r
o
v
em
e
n
t
i
n
m
o
d
e
l
p
e
r
f
o
r
m
a
n
c
e
f
o
l
lo
w
i
n
g
t
h
e
s
e
m
o
d
if
i
c
a
t
i
o
n
s
,
wh
e
r
e
F
i
g
u
r
e
7
(
a
)
s
h
o
w
s
t
r
a
in
i
n
g
an
d
v
a
l
i
d
a
t
io
n
a
c
cu
r
ac
y
a
n
d
F
i
g
u
r
e
7
(
b
)
s
h
o
w
s
t
r
a
in
i
n
g
an
d
v
a
l
i
d
a
t
io
n
l
o
s
s
e
s
.
W
i
t
h
a
f
i
n
a
l
t
e
s
t
a
c
cu
r
a
c
y
o
f
8
6
.
2
7
%
a
n
d
a
te
s
t
l
o
s
s
r
ed
u
ce
d
to
0
.
2
6
1
0
,
th
e
m
o
d
e
l
n
o
w
e
x
h
i
b
i
t
s
b
e
t
te
r
g
en
e
r
a
l
i
z
a
t
i
o
n
c
ap
ab
i
l
i
t
i
e
s
.
T
h
e
t
r
i
p
ar
t
i
t
e
d
i
v
i
s
i
o
n
o
f
t
h
e
d
a
t
as
e
t
(
t
r
a
in
i
n
g
/
v
a
l
i
d
a
t
io
n
/
te
s
t
)
s
i
g
n
i
f
i
ca
n
t
ly
m
i
t
ig
a
t
ed
th
e
p
r
e
v
i
o
u
s
l
y
o
b
s
er
v
ed
o
v
er
f
i
t
t
in
g
p
r
o
b
le
m
.
H
o
w
ev
e
r
,
a
s
l
i
g
h
t
te
n
d
e
n
cy
to
w
a
r
d
s
o
v
e
r
f
i
t
t
i
n
g
i
s
s
t
i
l
l
o
b
s
e
r
v
e
d
a
f
t
er
e
p
o
c
h
6
,
s
u
g
g
e
s
t
in
g
t
h
a
t
a
n
e
ar
l
y
s
t
o
p
p
i
n
g
s
t
r
a
t
e
g
y
c
o
u
l
d
f
u
r
th
er
o
p
t
i
m
i
z
e
th
e
r
e
s
u
l
t
s
.
I
m
p
l
e
m
en
t
i
n
g
in
d
e
p
en
d
en
c
e
t
e
s
t
s
a
n
d
a
s
e
p
ar
a
t
e
s
e
t
o
f
t
e
s
t
s
e
n
s
u
r
e
s
a
m
o
r
e
r
i
g
o
r
o
u
s
a
n
d
r
e
l
ia
b
l
e
e
v
a
lu
a
t
io
n
o
f
t
h
e
m
o
d
e
l
's
t
r
u
e
p
r
ed
i
c
t
i
v
e
p
o
w
e
r
.
Fig
u
r
e
6
.
Ou
tc
o
m
e
g
e
n
er
ated
b
y
o
u
r
s
u
g
g
ested
a
p
p
r
o
ac
h
af
t
er
im
p
lem
en
tin
g
in
d
e
p
en
d
e
n
t te
s
ts
s
et
(
a)
(
b
)
Fig
u
r
e
7
.
T
r
ain
in
g
a
n
d
v
alid
atio
n
r
esu
lts
af
ter
im
p
lem
e
n
tin
g
an
in
d
ep
e
n
d
en
t te
s
tin
g
s
et
o
n
(
a)
m
o
d
el
ac
c
u
r
ac
y
an
d
(
b
)
m
o
d
el
lo
s
s
T
h
e
r
esu
lts
o
b
tain
ed
a
f
ter
i
m
p
lem
en
tin
g
ea
r
ly
s
to
p
p
in
g
,
illu
s
tr
ated
in
Fig
u
r
es
8
an
d
9
,
r
ev
ea
l
a
r
em
ar
k
ab
le
im
p
r
o
v
em
en
t
in
th
e
m
o
d
el'
s
p
er
f
o
r
m
a
n
ce
,
w
h
er
e
Fig
u
r
e
9
(
a)
s
h
o
ws
tr
ain
in
g
an
d
v
alid
atio
n
ac
cu
r
ac
y
an
d
Fig
u
r
e
9
(
b
)
s
h
o
ws
tr
ain
in
g
a
n
d
v
alid
atio
n
lo
s
s
es
.
T
est
ac
cu
r
ac
y
n
o
w
r
ea
ch
es
8
4
.
3
1
%
with
a
lo
s
s
o
f
0
.
2
8
4
6
,
d
em
o
n
s
tr
atin
g
ex
ce
llen
t
g
en
er
aliza
tio
n
ca
p
ab
i
liti
es.
T
r
ain
in
g
s
to
p
p
e
d
au
to
m
atica
lly
af
ter
1
0
ep
o
ch
s
,
allo
win
g
th
e
m
o
d
el
to
co
n
v
er
g
e
to
an
o
p
tim
al
p
o
in
t
with
o
u
t
ex
ce
s
s
iv
e
o
v
er
f
itti
n
g
.
C
o
m
p
ar
ed
to
th
e
p
r
ev
io
u
s
r
esu
lts
(
8
6
.
2
7
%),
th
e
s
lig
h
t
d
ec
r
ea
s
e
in
test
ac
cu
r
ac
y
to
8
4
.
3
1
%
is
ac
tu
ally
a
p
o
s
itiv
e
in
d
icato
r
:
it
s
u
g
g
ests
th
at
th
e
m
o
d
el
p
r
io
r
it
izes
r
o
b
u
s
tn
ess
an
d
g
en
e
r
aliza
tio
n
o
v
er
ex
ce
s
s
iv
e
m
em
o
r
iza
tio
n
.
T
h
e
ac
c
u
r
ac
y
g
r
ap
h
s
h
o
ws
m
o
r
e
s
tab
le
an
d
p
r
ed
ictab
le
cu
r
v
es,
with
less
v
o
latilit
y
th
an
b
ef
o
r
e
.
E
ar
ly
s
to
p
p
in
g
p
r
e
v
en
ted
p
r
o
lo
n
g
ed
tr
ain
in
g
b
ey
o
n
d
t
h
e
o
p
tim
al
p
o
in
t,
t
h
u
s
en
s
u
r
in
g
a
b
etter
-
b
alan
ce
d
m
o
d
el,
m
o
r
e
r
eliab
le
f
o
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
B
r
a
in
tu
mo
r
d
etec
tio
n
u
s
in
g
V
GG
-
1
6
mo
d
el
(
A
ich
a
Ou
s
s
o
u
s
)
2343
p
r
o
d
u
ctio
n
,
an
d
ca
p
ab
le
o
f
m
ain
tain
in
g
co
n
s
is
ten
t
p
er
f
o
r
m
an
ce
o
n
n
ew
d
ata.
T
ab
le
1
s
h
o
ws
a
d
etailed
co
m
p
ar
is
o
n
o
f
th
e
r
esu
lts
o
b
ta
in
ed
at
th
e
th
r
ee
s
tag
es o
f
m
o
d
el
im
p
r
o
v
em
e
n
t.
Fig
u
r
e
8
.
Ou
tc
o
m
e
g
e
n
er
ated
b
y
o
u
r
s
u
g
g
ested
a
p
p
r
o
ac
h
af
t
er
im
p
lem
en
tin
g
an
ea
r
l
y
s
to
p
p
in
g
s
tr
ateg
y
(
a)
(
b
)
Fig
u
r
e
9
.
T
r
ain
in
g
a
n
d
v
alid
atio
n
r
esu
lts
af
ter
im
p
lem
e
n
tin
g
an
im
p
lem
en
tin
g
an
ea
r
ly
s
to
p
p
in
g
s
tr
ateg
y
o
n
(
a)
m
o
d
el
ac
cu
r
ac
y
an
d
(
b
)
m
o
d
el
lo
s
s
T
ab
le
1
.
C
o
m
p
a
r
ativ
e
tab
le
o
f
m
o
d
el
p
er
f
o
r
m
a
n
ce
s
C
r
i
t
e
r
i
a
I
n
i
t
i
a
l
M
o
d
e
l
I
n
d
e
p
e
n
d
e
n
t
t
e
st
s
se
t
E
a
r
l
y
st
o
p
p
i
n
g
st
r
a
t
e
g
y
Te
st
a
c
c
u
r
a
c
y
86
.
2
7
%
84
.
3
1
%
Te
st
l
o
ss
0
.
2
6
1
0
0
.
2
8
4
6
Tr
a
i
n
i
n
g
a
c
c
u
r
a
c
y
~
9
3
%
(
é
p
o
q
u
e
8
)
~
9
6
%
(
é
p
o
q
u
e
8
)
~
9
9
.
3
0
%
(
é
p
o
q
u
e
1
0
)
V
a
l
i
d
a
t
i
o
n
a
c
c
u
r
a
cy
99
.
2
%
(
é
p
o
q
u
e
1
0
)
~
8
2
%
(
é
p
o
q
u
e
1
0
)
~
8
2
.
0
5
%
(
é
p
o
q
u
e
1
0
)
Tr
a
i
n
i
n
g
l
o
ss
~0
.
05
~0
.
12
~0
.
05
V
a
l
i
d
a
t
i
o
n
l
o
ss
0
.
9
9
2
~0
.
53
~0
.
54
N
u
mb
e
r
o
f
e
p
o
c
h
s
10
10
1
0
(
w
i
t
h
e
a
r
l
y
s
t
o
p
p
i
n
g
)
O
v
e
r
f
i
t
t
i
n
g
S
e
v
e
r
e
(
f
r
o
m
p
e
r
i
o
d
6
-
7)
M
o
d
e
r
a
t
e
(
sl
i
g
h
t
a
f
t
e
r
e
p
o
c
h
5
-
6)
M
i
n
i
m
a
l
(
c
o
n
t
r
o
l
e
f
f
e
c
t
i
v
e
l
y
)
C
u
r
v
e
s
t
a
b
i
l
i
t
y
Lo
w
(
st
r
o
n
g
d
i
v
e
r
g
e
n
c
e
)
A
v
e
r
a
g
e
(
r
e
d
u
c
e
d
d
i
v
e
r
g
e
n
c
e
)
Ex
c
e
l
l
e
n
t
(
h
a
r
m
o
n
i
o
u
s c
o
n
v
e
r
g
e
n
c
e
)
D
a
t
a
d
i
v
i
si
o
n
Tr
a
i
n
/
v
a
l
i
d
a
t
i
o
n
o
n
l
y
Tr
a
i
n
/
v
a
l
i
d
a
t
i
o
n
/
t
e
st
Tr
a
i
n
/
v
a
l
i
d
a
t
i
o
n
/
t
e
st
G
e
n
e
r
a
l
i
z
a
t
i
o
n
c
a
p
a
c
i
t
y
V
e
r
y
l
o
w
G
o
o
d
Ex
c
e
l
l
e
n
t
4
.
1
.
Co
m
pa
ra
t
iv
e
a
na
ly
s
is
wit
h e
x
is
t
ing
t
ec
hn
iqu
e
s
Sev
er
al
r
ec
en
t
s
tu
d
ies
h
av
e
in
v
esti
g
ated
C
NN
-
b
ased
m
eth
o
d
s
f
o
r
b
r
ai
n
tu
m
o
r
d
etec
tio
n
in
MRI
im
ag
es,
with
v
a
r
y
in
g
s
u
cc
ess
lev
els
as
s
h
o
wn
in
T
ab
le
2
.
R
esear
ch
b
y
I
r
s
h
ei
d
at
an
d
D
u
wair
i
[
2
2
]
u
s
ed
a
two
-
s
tag
e
C
NN
ap
p
r
o
ac
h
t
h
a
t
f
ir
s
t
em
p
lo
y
ed
b
o
u
n
d
i
n
g
b
o
x
es
f
o
r
tu
m
o
r
l
o
ca
lizatio
n
,
a
n
d
th
e
n
c
o
n
d
u
cted
class
if
icatio
n
.
T
h
e
s
tu
d
y
in
[
2
3
]
s
h
o
wed
th
e
u
s
e
o
f
Mo
b
ile
Net
v
2
,
I
n
ce
p
tio
n
V
3
,
an
d
R
esNet5
0
ar
ch
itectu
r
es,
r
ep
o
r
tin
g
9
8
.
3
%,
9
8
.
6
%,
an
d
9
8
.
6
%
at
5
,
10
,
an
d
2
0
-
f
o
ld
c
r
o
s
s
-
v
alid
atio
n
.
Me
an
wh
ile,
So
wr
ir
ajan
et
a
l.
[
2
4
]
ex
am
in
ed
VGG
-
1
6
,
ac
h
iev
in
g
9
7
.
9
%
ac
cu
r
ac
y
in
t
h
eir
t
ests
.
An
o
th
er
im
p
o
r
tan
t
wo
r
k
in
[
2
5
]
co
m
b
i
n
ed
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
3
,
J
u
n
e
2
0
2
6
:
2
3
3
7
-
2
3
4
6
2344
VGG
-
1
6
with
g
r
ay
-
le
v
el
co
-
o
cc
u
r
r
e
n
ce
m
atr
ix
(
GL
C
M)
f
ea
tu
r
es
f
o
r
b
r
ain
MRI
an
al
y
s
is
,
r
ea
ch
in
g
9
6
%
ac
cu
r
ac
y
.
T
h
e
p
r
o
p
o
s
ed
VG
G
-
1
6
tr
an
s
f
er
lear
n
in
g
f
r
am
e
wo
r
k
g
r
ea
tly
s
u
r
p
ass
es
th
ese
m
eth
o
d
s
,
d
eliv
er
in
g
h
ig
h
er
class
if
icatio
n
ac
cu
r
ac
y
at
9
9
.
2
1
%.
T
h
e
n
o
ta
b
le
in
cr
ea
s
e
in
ac
cu
r
ac
y
h
ig
h
lig
h
ts
th
e
ef
f
ec
tiv
en
ess
an
d
r
eliab
ilit
y
o
f
o
u
r
ap
p
r
o
ac
h
f
o
r
au
to
m
ated
b
r
ain
tu
m
o
r
d
etec
ti
o
n
an
d
class
if
icatio
n
.
T
ab
le
2
.
C
o
m
p
a
r
is
o
n
with
ex
i
s
tin
g
tech
n
iq
u
es
R
e
f
e
r
e
n
c
e
A
l
g
o
r
i
t
h
m
A
c
c
u
r
a
c
y
(
%)
[
2
2
]
C
N
N
9
6
.
7
[
2
3
]
M
o
b
i
l
e
N
e
t
V
2
,
I
n
c
e
p
t
i
o
n
V
3
,
a
n
d
R
e
sN
e
t
5
0
9
8
.
3
,
9
8
.
6
,
a
n
d
9
8
.
6
a
t
f
i
v
e
,
t
e
n
,
a
n
d
2
0
-
f
o
l
d
c
r
o
ss
-
v
a
l
i
d
a
t
i
o
n
[
2
4
]
VGG
-
16
9
7
.
9
[
2
5
]
VGG
-
1
6
w
i
t
h
G
LC
M
96
P
r
o
p
o
se
d
m
e
t
h
o
d
VGG
-
16
9
9
.
2
1
5.
CO
NCLU
SI
O
N
T
u
m
o
r
s
p
o
s
e
a
m
ajo
r
d
an
g
e
r
to
p
eo
p
le
b
ec
au
s
e
m
alig
n
an
t
c
ells
ca
n
in
v
ad
e
n
ea
r
b
y
tis
s
u
es
an
d
s
p
r
ea
d
to
d
if
f
er
e
n
t
p
a
r
ts
o
f
th
e
b
o
d
y
.
T
im
ely
i
d
en
tific
atio
n
o
f
b
r
ain
tu
m
o
r
s
is
ess
en
tial
f
o
r
d
eliv
er
in
g
t
h
e
b
est
tr
ea
tm
en
t.
T
h
is
r
esear
ch
u
tili
ze
s
C
NN
s
to
cr
ea
te
a
m
o
d
el
th
at
class
if
ie
s
M
R
I
s
ca
n
s
o
f
b
r
ain
tu
m
o
r
s
.
A
n
o
tab
le
b
en
ef
it
o
f
th
is
m
eth
o
d
is
its
ca
p
ac
ity
to
a
u
to
m
atica
lly
d
etec
t
in
tr
icate
f
ea
tu
r
es
in
m
u
lti
-
m
o
d
al
MRI
s
ca
n
s
,
allo
win
g
it
to
d
if
f
er
en
tiate
b
et
wee
n
b
r
ain
r
eg
io
n
s
with
tu
m
o
r
s
an
d
th
o
s
e
th
at
ar
e
h
ea
lth
y
.
T
h
e
m
o
d
el
s
h
o
ws
r
em
ar
k
ab
le
r
esu
lts
,
r
ea
ch
in
g
a
n
ac
cu
r
ac
y
o
f
9
9
%
wh
en
ev
al
u
ated
o
n
a
s
m
all
s
elec
tio
n
o
f
b
r
ain
tu
m
o
r
d
atasets
th
at
h
av
e
u
n
e
v
en
class
if
icatio
n
s
.
Ho
wev
er
,
d
esp
ite
th
ese
en
co
u
r
ag
in
g
f
in
d
in
g
s
,
th
e
m
o
d
el
en
co
u
n
ter
s
s
ev
er
al
is
s
u
es
th
at
co
u
ld
af
f
ec
t
its
s
u
cc
ess
in
p
r
ac
tical
m
ed
ical
s
ettin
g
s
.
T
o
tack
le
th
ese
c
h
allen
g
e
s
,
f
u
tu
r
e
in
itiativ
es
s
h
o
u
ld
aim
to
in
cr
ea
s
e
a
n
d
d
i
v
er
s
if
y
th
e
d
ataset,
u
s
e
3
D
C
NN
f
o
r
th
o
r
o
u
g
h
s
p
atial
a
n
aly
s
is
,
an
d
a
d
d
m
o
r
e
d
ata
ty
p
es
lik
e
p
atien
t
m
ed
ical
h
is
to
r
y
an
d
g
e
n
etic
d
et
ails
.
I
m
p
r
o
v
i
n
g
th
e
m
o
d
el'
s
tr
an
s
p
ar
en
c
y
an
d
d
ep
en
d
a
b
ilit
y
ca
n
b
e
ac
co
m
p
lis
h
ed
b
y
u
s
in
g
e
x
p
lain
a
b
le
AI
s
tr
ateg
ies,
q
u
an
tify
i
n
g
u
n
ce
r
tai
n
ty
,
an
d
co
n
d
u
ctin
g
co
m
p
r
e
h
en
s
iv
e
test
in
g
.
Mo
r
eo
v
e
r
,
in
v
esti
g
atin
g
f
ed
er
ated
lear
n
in
g
,
co
n
tin
u
o
u
s
lear
n
in
g
s
y
s
tem
s
,
an
d
en
s
em
b
le
tech
n
iq
u
es
ca
n
en
h
an
ce
th
e
m
o
d
el'
s
ef
f
ec
tiv
en
ess
an
d
ad
a
p
tab
ilit
y
.
L
astl
y
,
th
o
r
o
u
g
h
clin
ical
v
alid
atio
n
an
d
ad
h
er
en
ce
to
et
h
ical
an
d
r
eg
u
lato
r
y
g
u
id
elin
e
s
ar
e
cr
u
cial
f
o
r
t
h
e
ef
f
ec
tiv
e
i
n
co
r
p
o
r
atio
n
o
f
th
e
m
o
d
el
in
to
h
ea
lth
ca
r
e
p
r
ac
ti
ce
s
,
m
ak
in
g
s
u
r
e
th
at
it
n
o
t
o
n
ly
o
p
er
ates
well
b
u
t
also
co
m
p
lies
with
all
n
ec
ess
ar
y
leg
al
an
d
eth
ical
s
ta
n
d
ar
d
s
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
s
tate
n
o
f
u
n
d
in
g
in
v
o
lv
ed
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Aich
a
Ou
s
s
o
u
s
✓
✓
✓
✓
✓
✓
✓
✓
✓
A
b
d
e
r
r
a
h
m
a
n
e
E
z
-
Z
a
h
o
u
t
✓
✓
✓
✓
✓
So
u
m
ia
Z
iti
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
T
h
e
au
t
h
o
r
d
ec
lar
es
t
h
at
th
e
r
e
ar
e
n
o
k
n
o
wn
co
n
f
licts
o
f
in
ter
est
ass
o
ciate
d
with
th
is
p
u
b
licatio
n
.
T
h
er
e
ar
e
n
o
f
in
a
n
cial
o
r
p
e
r
s
o
n
al
r
elatio
n
s
h
ip
s
th
at
co
u
ld
in
ap
p
r
o
p
r
iately
in
f
l
u
en
ce
o
r
b
ias
th
e
co
n
ten
t
o
f
th
is
wo
r
k
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
B
r
a
in
tu
mo
r
d
etec
tio
n
u
s
in
g
V
GG
-
1
6
mo
d
el
(
A
ich
a
Ou
s
s
o
u
s
)
2345
I
NF
O
RM
E
D
CO
NS
E
N
T
W
e
h
av
en
'
t
u
s
ed
a
p
r
ed
ef
in
ed
p
er
s
o
n
;
we
h
av
e
ac
q
u
ir
e
d
a
p
u
b
licly
av
ailab
le
d
ataset
o
f
b
in
ar
y
-
class
b
r
ain
MRI
im
ag
es f
r
o
m
th
e
K
ag
g
le
p
latf
o
r
m
.
E
T
H
I
CAL AP
P
RO
V
AL
T
h
is
s
ec
tio
n
n
o
t
a
p
p
licab
le.
T
h
is
r
esear
ch
d
id
n
o
t
in
v
o
lv
e
h
u
m
a
n
s
u
b
jects,
h
u
m
an
b
io
lo
g
ical
m
ater
ials
,
o
r
ex
p
er
im
en
tal
p
r
o
ce
d
u
r
es
o
n
an
im
als.
T
h
e
wo
r
k
was
co
n
d
u
cted
s
o
lely
o
n
co
m
p
u
tatio
n
al
m
o
d
els,
p
u
b
licly
av
ailab
le
d
atasets
,
o
r
n
o
n
‑
s
en
s
itiv
e
d
ata
th
at
d
i
d
n
o
t
r
eq
u
i
r
e
in
ter
v
en
tio
n
with
liv
in
g
o
r
g
a
n
is
m
s
.
T
h
er
ef
o
r
e,
eth
ical
ap
p
r
o
v
al
f
r
o
m
an
in
s
titu
tio
n
al
r
ev
iew
b
o
ar
d
o
r
a
n
im
al
eth
ics
c
o
m
m
itte
e
was
n
o
t
n
ec
ess
ar
y
f
o
r
th
is
s
tu
d
y
.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
au
th
o
r
s
c
o
n
f
i
r
m
t
h
at
th
e
d
ata
s
u
p
p
o
r
ti
n
g
t
h
e
f
i
n
d
in
g
s
o
f
t
h
is
s
t
u
d
y
a
r
e
a
v
a
ila
b
l
e
wit
h
i
n
t
h
e
a
r
ti
cle
.
RE
F
E
R
E
NC
E
S
[
1
]
A
.
O
u
ss
o
u
s,
A
.
Ez
-
Za
h
o
u
t
,
S
.
Zi
t
i
,
a
n
d
A
.
O
u
sso
u
s,
“
A
n
o
v
e
r
v
i
e
w
o
f
t
h
e
mo
s
t
e
f
f
i
c
i
e
n
t
me
t
h
o
d
s
f
o
r
p
r
e
d
i
c
t
i
n
g
h
e
a
l
t
h
c
a
r
e
d
i
s
o
r
d
e
r
s,”
i
n
AI
P
C
o
n
f
e
re
n
c
e
Pr
o
c
e
e
d
i
n
g
s
,
2
0
2
3
,
v
o
l
.
2
8
1
4
,
n
o
.
1
,
d
o
i
:
1
0
.
1
0
6
3
/
5
.
0
1
5
0
0
2
5
.
[
2
]
A
.
O
u
sso
u
s
,
A
.
Ez
-
Z
a
h
o
u
t
,
a
n
d
S
.
Zi
t
i
,
“
R
e
v
i
e
w
i
n
g
c
h
r
o
n
i
c
a
i
l
m
e
n
t
s:
p
r
e
d
i
c
t
i
n
g
d
i
se
a
ses
w
i
t
h
a
m
u
l
t
i
-
s
y
mp
t
o
m
a
p
p
r
o
a
c
h
,
”
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
En
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
.
3
5
,
n
o
.
1
,
p
p
.
4
1
8
–
4
2
7
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s
.
v
3
5
.
i
1
.
p
p
4
1
8
-
4
2
7
.
[
3
]
D
.
S
.
P
r
a
b
h
a
a
n
d
J
.
S
.
K
u
m
a
r
,
“
P
e
r
f
o
r
man
c
e
e
v
a
l
u
a
t
i
o
n
o
f
i
ma
g
e
s
e
g
m
e
n
t
a
t
i
o
n
u
s
i
n
g
o
b
j
e
c
t
i
v
e
met
h
o
d
s
,
”
I
n
d
i
a
n
J
o
u
r
n
a
l
o
f
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
9
,
n
o
.
8
,
2
0
1
6
,
d
o
i
:
1
0
.
1
7
4
8
5
/
i
j
st
/
2
0
1
6
/
v
9
i
8
/
8
7
9
0
7
.
[
4
]
A
.
M
u
st
a
q
e
e
m,
A
.
Ja
v
e
d
,
a
n
d
T.
F
a
t
i
ma,
“
A
n
e
f
f
i
c
i
e
n
t
b
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
a
l
g
o
r
i
t
h
m
u
si
n
g
w
a
t
e
r
sh
e
d
a
n
d
t
h
r
e
sh
o
l
d
i
n
g
b
a
se
d
seg
m
e
n
t
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
I
m
a
g
e
,
G
ra
p
h
i
c
s
a
n
d
S
i
g
n
a
l
Pr
o
c
e
ssi
n
g
,
v
o
l
.
4
,
n
o
.
1
0
,
p
p
.
3
4
–
3
9
,
2
0
1
2
,
d
o
i
:
1
0
.
5
8
1
5
/
i
j
i
g
sp
.
2
0
1
2
.
1
0
.
0
5
.
[
5
]
M
.
I
.
S
h
a
r
i
f
,
J
.
P
.
Li
,
J.
A
mi
n
,
a
n
d
A
.
S
h
a
r
i
f
,
“
A
n
i
m
p
r
o
v
e
d
f
r
a
m
e
w
o
r
k
f
o
r
b
r
a
i
n
t
u
mo
r
a
n
a
l
y
s
i
s
u
si
n
g
M
R
I
b
a
s
e
d
o
n
Y
O
LO
v
2
a
n
d
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
C
o
m
p
l
e
x
&
I
n
t
e
l
l
i
g
e
n
t
S
y
st
e
m
s
,
v
o
l
.
7
,
p
p
.
2
0
2
3
–
2
0
3
6
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s4
0
7
4
7
-
021
-
0
0
3
1
0
-
3.
[
6
]
A
.
R
a
z
a
e
t
a
l
.
,
“
A
h
y
b
r
i
d
d
e
e
p
l
e
a
r
n
i
n
g
-
b
a
s
e
d
a
p
p
r
o
a
c
h
f
o
r
b
r
a
i
n
t
u
m
o
r
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
E
l
e
c
t
r
o
n
i
c
s
,
v
o
l
.
1
1
,
n
o
.
7
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
e
l
e
c
t
r
o
n
i
c
s1
1
0
7
1
1
4
6
.
[
7
]
P
.
A
g
r
a
w
a
l
,
N
.
K
a
t
a
l
,
a
n
d
N
.
H
o
o
d
a
,
“
S
e
g
m
e
n
t
a
t
i
o
n
a
n
d
c
l
a
ssi
f
i
c
a
t
i
o
n
o
f
b
r
a
i
n
t
u
m
o
r
u
si
n
g
3
D
-
U
N
e
t
d
e
e
p
n
e
t
w
o
r
k
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
g
n
i
t
i
v
e
C
o
m
p
u
t
i
n
g
i
n
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
3
,
p
p
.
1
9
9
–
2
1
0
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
j
c
c
e
.
2
0
2
2
.
1
1
.
0
0
1
.
[
8
]
M
.
F
.
A
h
a
me
d
e
t
a
l
.
,
“
A
r
e
v
i
e
w
o
n
b
r
a
i
n
t
u
m
o
r
se
g
me
n
t
a
t
i
o
n
b
a
s
e
d
o
n
d
e
e
p
l
e
a
r
n
i
n
g
me
t
h
o
d
s
w
i
t
h
f
e
d
e
r
a
t
e
d
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,
”
C
o
m
p
u
t
e
r
i
ze
d
Me
d
i
c
a
l
I
m
a
g
i
n
g
a
n
d
G
ra
p
h
i
c
s
,
v
o
l
.
1
1
0
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
mp
me
d
i
m
a
g
.
2
0
2
3
.
1
0
2
3
1
3
.
[
9
]
T.
M
a
g
a
d
z
a
a
n
d
S
.
V
i
r
i
r
i
,
“
D
e
e
p
l
e
a
r
n
i
n
g
f
o
r
b
r
a
i
n
t
u
mo
r
s
e
g
me
n
t
a
t
i
o
n
:
a
su
r
v
e
y
o
f
s
t
a
t
e
-
of
-
t
h
e
-
a
r
t
,
”
J
o
u
rn
a
l
o
f
I
m
a
g
i
n
g
,
v
o
l
.
7
,
n
o
.
2
,
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
j
i
m
a
g
i
n
g
7
0
2
0
0
1
9
.
[
1
0
]
E.
S
.
B
i
r
a
t
u
,
F
.
S
c
h
w
e
n
k
e
r
,
Y
.
M
.
A
y
a
n
o
,
a
n
d
T.
G
.
D
e
b
e
l
e
e
,
“
A
s
u
r
v
e
y
o
f
b
r
a
i
n
t
u
mo
r
s
e
g
me
n
t
a
t
i
o
n
a
n
d
c
l
a
ssi
f
i
c
a
t
i
o
n
a
l
g
o
r
i
t
h
ms,
”
J
o
u
rn
a
l
o
f
I
m
a
g
i
n
g
,
v
o
l
.
7
,
n
o
.
9
,
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
j
i
m
a
g
i
n
g
7
0
9
0
1
7
9
.
[
1
1
]
M
.
P
r
a
s
t
a
w
a
,
E.
B
u
l
l
i
t
t
,
a
n
d
G
.
G
e
r
i
g
,
“
S
i
m
u
l
a
t
i
o
n
o
f
b
r
a
i
n
t
u
mo
r
s
i
n
M
R
i
ma
g
e
s
f
o
r
e
v
a
l
u
a
t
i
o
n
o
f
se
g
m
e
n
t
a
t
i
o
n
e
f
f
i
c
a
c
y
,
”
Me
d
i
c
a
l
I
m
a
g
e
A
n
a
l
y
s
i
s
,
v
o
l
.
1
3
,
n
o
.
2
,
p
p
.
2
9
7
–
3
1
1
,
2
0
0
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
med
i
a
.
2
0
0
8
.
1
1
.
0
0
2
.
[
1
2
]
K
.
N
.
D
e
e
k
s
h
a
,
D
.
M
,
A
.
V
G
i
r
i
s
h
,
A
.
S
.
B
h
a
t
,
a
n
d
L.
H
,
“
C
l
a
ssi
f
i
c
a
t
i
o
n
o
f
b
r
a
i
n
t
u
mo
r
a
n
d
i
t
s
t
y
p
e
s
u
si
n
g
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
i
n
2
0
2
0
I
EEE
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
f
o
r
I
n
n
o
v
a
t
i
o
n
i
n
T
e
c
h
n
o
l
o
g
y
(
I
N
O
C
O
N
)
,
2
0
2
0
,
p
p
.
1
–
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
N
O
C
O
N
5
0
5
3
9
.
2
0
2
0
.
9
2
9
8
3
0
6
.
[
1
3
]
D
.
C
.
F
e
b
r
i
a
n
t
o
,
I
.
S
o
e
s
a
n
t
i
,
a
n
d
H
.
A
.
N
u
g
r
o
h
o
,
“
C
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
f
o
r
b
r
a
i
n
t
u
mo
r
d
e
t
e
c
t
i
o
n
,
”
i
n
I
O
P
C
o
n
f
e
re
n
c
e
S
e
r
i
e
s:
Ma
t
e
r
i
a
l
s
S
c
i
e
n
c
e
a
n
d
En
g
i
n
e
e
ri
n
g
,
2
0
2
0
,
v
o
l
.
7
7
1
,
d
o
i
:
1
0
.
1
0
8
8
/
1
7
5
7
-
8
9
9
X
/
7
7
1
/
1
/
0
1
2
0
3
1
.
[
1
4
]
P
.
G
a
y
a
t
h
r
i
,
A
.
D
h
a
v
i
l
e
sw
a
r
a
p
u
,
S
.
I
b
r
a
h
i
m
,
R
.
P
a
u
l
,
a
n
d
R
.
G
u
p
t
a
,
“
Ex
p
l
o
r
i
n
g
t
h
e
p
o
t
e
n
t
i
a
l
o
f
V
G
G
-
1
6
a
r
c
h
i
t
e
c
t
u
r
e
f
o
r
a
c
c
u
r
a
t
e
b
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
”
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
e
rs,
Me
c
h
a
n
i
c
a
l
a
n
d
Ma
n
a
g
e
m
e
n
t
,
v
o
l
.
2
,
n
o
.
2
,
2
0
2
3
,
d
o
i
:
1
0
.
5
7
1
5
9
/
g
a
d
l
.
j
c
mm
.
2
.
2
.
2
3
0
5
6
.
[
1
5
]
M
.
I
.
M
a
h
m
u
d
,
M
.
M
a
mu
n
,
a
n
d
A
.
A
b
d
e
l
g
a
w
a
d
,
“
A
d
e
e
p
a
n
a
l
y
si
s
o
f
b
r
a
i
n
t
u
mo
r
d
e
t
e
c
t
i
o
n
f
r
o
m
M
R
i
ma
g
e
s
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
n
e
t
w
o
r
k
s,”
Al
g
o
ri
t
h
m
s
,
v
o
l
.
1
6
,
n
o
.
4
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
a
1
6
0
4
0
1
7
6
.
[
1
6
]
T
.
G
a
y
a
t
h
r
i
a
n
d
S
.
K
.
K
u
mar
,
“
B
r
a
i
n
t
u
mo
r
se
g
me
n
t
a
t
i
o
n
a
n
d
c
l
a
ssi
f
i
c
a
t
i
o
n
u
si
n
g
C
N
N
p
r
e
-
t
r
a
i
n
e
d
V
G
G
-
1
6
m
o
d
e
l
i
n
M
R
I
i
ma
g
e
s,
”
I
I
U
M
En
g
i
n
e
e
r
i
n
g
J
o
u
r
n
a
l
,
v
o
l
.
2
5
,
n
o
.
2
,
p
p
.
1
9
6
–
2
1
1
,
2
0
2
4
,
d
o
i
:
1
0
.
3
1
4
3
6
/
i
i
u
mej
.
v
2
5
i
2
.
2
9
6
3
.
[
1
7
]
S
.
R
a
g
h
u
v
a
n
s
h
i
a
n
d
S
.
D
h
a
r
i
w
a
l
,
“
Th
e
V
G
G
1
6
me
t
h
o
d
i
s
a
p
o
w
e
r
f
u
l
t
o
o
l
f
o
r
d
e
t
e
c
t
i
n
g
b
r
a
i
n
t
u
m
o
r
s
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,
”
i
n
RAi
S
E
-
2
0
2
3
,
MDPI
E
n
g
i
n
e
e
r
i
n
g
Pro
c
e
e
d
i
n
g
s
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
e
n
g
p
r
o
c
2
0
2
3
0
5
9
0
4
6
.
[
1
8
]
A
.
R
.
I
smai
l
,
S
.
Q
.
N
i
sa
,
S
.
A
.
S
h
a
h
a
r
u
d
d
i
n
,
S
.
I
.
M
a
s
n
i
,
a
n
d
S
.
A
.
S
.
A
mi
n
,
“
U
t
i
l
i
si
n
g
V
G
G
-
1
6
o
f
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
f
o
r
m
e
d
i
c
a
l
i
m
a
g
e
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
P
e
rv
a
si
v
e
C
o
m
p
u
t
i
n
g
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
v
o
l
.
1
0
,
n
o
.
1
,
p
p
.
1
1
3
–
1
1
8
,
2
0
2
4
,
d
o
i
:
1
0
.
3
1
4
3
6
/
i
j
p
c
c
.
v
1
0
i
1
.
4
6
0
.
[
1
9
]
D
.
S
a
n
t
o
s,
“
B
r
a
i
n
t
u
mo
r
d
e
t
e
c
t
i
o
n
u
s
i
n
g
t
h
e
V
G
G
-
1
6
mo
d
e
l
:
a
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
Pre
p
r
i
n
t
s
,
2
0
2
3
,
d
o
i
:
1
0
.
2
0
9
4
4
/
p
r
e
p
r
i
n
t
s
2
0
2
3
0
8
.
0
9
8
3
.
v
1
.
[
2
0
]
Z.
R
a
sh
e
e
d
e
t
a
l
.
,
“
B
r
a
i
n
t
u
mo
r
c
l
a
ssi
f
i
c
a
t
i
o
n
f
r
o
m M
R
I
u
si
n
g
i
ma
g
e
e
n
h
a
n
c
e
me
n
t
a
n
d
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
t
e
c
h
n
i
q
u
e
s,
”
Bra
i
n
S
c
i
e
n
c
e
s
,
v
o
l
.
1
3
,
n
o
.
9
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
b
r
a
i
n
sc
i
1
3
0
9
1
3
2
0
.
[
2
1
]
M
.
H
u
ma
y
u
n
,
R
.
S
u
j
a
t
h
a
,
S
.
N
.
A
l
m
u
a
y
q
i
l
,
a
n
d
N
.
Z
.
J
h
a
n
j
h
i
,
“
A
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
w
i
t
h
a
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
f
o
r
t
h
e
c
l
a
ssi
f
i
c
a
t
i
o
n
o
f
l
u
n
g
c
a
r
c
i
n
o
m
a
,
”
H
e
a
l
t
h
c
a
r
e
,
v
o
l
.
1
0
,
n
o
.
6
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
h
e
a
l
t
h
c
a
r
e
1
0
0
6
1
0
5
8
.
[
2
2
]
S
.
I
r
sh
e
i
d
a
t
a
n
d
R
.
D
u
w
a
i
r
i
,
“
B
r
a
i
n
t
u
mo
r
d
e
t
e
c
t
i
o
n
u
s
i
n
g
a
r
t
i
f
i
c
i
a
l
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
1
1
t
h
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
f
o
rm
a
t
i
o
n
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
C
S
)
,
A
p
r
.
2
0
2
0
,
p
p
.
1
9
7
–
2
0
3
.
[
2
3
]
S
.
Jai
n
a
n
d
V
.
Ja
i
n
“
N
o
v
e
l
a
p
p
r
o
a
c
h
t
o
c
l
a
s
si
f
y
b
r
a
i
n
t
u
mo
r
b
a
s
e
d
o
n
t
r
a
n
s
f
e
r
l
e
a
r
n
i
n
g
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
15
,
n
o
.
4
,
2
0
3
1
–
2
0
3
8
,
A
p
r
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s4
1
8
7
0
-
023
-
0
1
2
5
9
-
x.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
3
,
J
u
n
e
2
0
2
6
:
2
3
3
7
-
2
3
4
6
2346
[
2
4
]
S
.
R
.
S
o
w
r
i
r
a
j
a
n
,
S
.
B
a
l
a
su
b
r
a
m
a
n
i
a
n
,
a
n
d
R
.
S
.
P
.
R
a
j
,
“
M
R
I
b
r
a
i
n
t
u
m
o
r
c
l
a
ssi
f
i
c
a
t
i
o
n
u
si
n
g
a
h
y
b
r
i
d
V
G
G
1
6
-
N
A
D
E
mo
d
e
l
,
”
Bra
z
i
l
i
a
n
Ar
c
h
i
v
e
s
o
f
Bi
o
l
o
g
y
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
66
,
2
0
2
3
,
d
o
i
:
1
0
.
1
5
9
0
/
1
6
7
8
-
4
3
2
4
-
2
0
2
3
2
2
0
0
7
1
.
[
2
5
]
H
.
K
i
b
r
i
y
a
,
R
.
A
mi
n
,
J.
K
i
m,
M
.
N
a
w
a
z
,
a
n
d
R
.
G
a
n
t
a
ss
i
,
“
A
n
o
v
e
l
a
p
p
r
o
a
c
h
f
o
r
b
r
a
i
n
t
u
m
o
r
c
l
a
ssi
f
i
c
a
t
i
o
n
u
s
i
n
g
a
n
e
n
s
e
m
b
l
e
o
f
d
e
e
p
a
n
d
h
a
n
d
-
c
r
a
f
t
e
d
f
e
a
t
u
r
e
s,
”
S
e
n
s
o
rs
,
v
o
l
.
2
3
,
n
o
.
1
0
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
3
1
0
4
6
9
3
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Aic
h
a
O
u
ss
o
u
s
Ph
.
D
is
stu
d
e
n
t
i
n
F
a
c
u
lt
y
o
f
S
c
ien
c
e
s
,
M
o
h
a
m
m
e
d
V
Un
i
v
e
rsity
in
Ra
b
a
t.
S
h
e
is
c
o
m
p
u
ter
sc
ien
c
e
e
n
g
in
e
e
r
,
a
d
e
g
re
e
fr
o
m
t
h
e
Na
ti
o
n
a
l
S
c
h
o
o
l
o
f
A
p
p
l
ied
S
c
ien
c
e
s,
Ag
a
d
ir,
M
o
ro
c
c
o
,
in
2
0
1
3
.
Li
c
e
n
se
d
i
p
lo
m
a
in
M
a
th
e
m
a
ti
c
s
a
n
d
Co
m
p
u
ter
S
c
ien
c
e
s
fro
m
th
e
F
a
c
u
lt
y
o
f
S
c
ien
c
e
s,
Ib
n
Zo
h
r
Un
iv
e
rsity
,
Ag
a
d
ir,
M
o
r
o
c
c
o
,
i
n
2
0
1
0
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
a
ich
a
_
o
u
ss
o
u
s
@u
m
5
.
a
c
.
m
a
.
Abd
e
r
r
a
h
m
a
n
e
Ez
-
Za
h
o
u
t
is
c
u
rre
n
tl
y
a
n
a
ss
o
c
iate
p
ro
fe
ss
o
r
a
t
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
s
,
F
a
c
u
lt
y
o
f
S
c
ien
c
e
s
,
M
o
h
a
m
m
e
d
V
Un
i
v
e
r
sity
.
He
g
ra
d
u
a
ted
P
h
.
D.
i
n
Co
m
p
u
ter
S
c
ien
c
e
s
fro
m
ENS
I
AS
Co
ll
e
g
e
o
f
IT
.
H
e
is
a
n
a
c
ti
v
e
m
e
m
b
e
r
o
f
IP
S
S
Tea
m
(in
telli
g
e
n
t
p
r
o
c
e
ss
in
g
sy
ste
m
s
a
n
d
se
c
u
rit
y
).
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
a
b
d
e
rra
h
m
a
n
e
_
e
z
z
a
h
o
u
t@
u
m
5
.
a
c
.
m
a
.
S
o
u
m
ia
Ziti
is
a
fu
ll
p
ro
fe
ss
o
r
a
n
d
re
se
a
rc
h
e
r
a
t
th
e
F
a
c
u
lt
y
o
f
S
c
ien
c
e
s
o
f
M
o
h
a
m
m
e
d
V
U
n
iv
e
rsit
y
in
Ra
b
a
t
sin
c
e
2
0
0
7
.
S
h
e
o
b
tain
e
d
h
e
r
P
h
.
D.
in
Co
m
p
u
ter
S
c
ien
c
e
sp
e
c
ializin
g
in
g
ra
p
h
t
h
e
o
r
y
fro
m
th
e
Un
iv
e
rsity
o
f
Orle
a
n
s
in
F
ra
n
c
e
,
a
lo
n
g
with
a
d
ip
l
o
m
a
in
a
d
v
a
n
c
e
d
stu
d
ies
in
f
u
n
d
a
m
e
n
t
a
l
c
o
m
p
u
ter
sc
ien
c
e
.
S
h
e
a
lso
h
o
l
d
s
a
Ba
c
c
a
lau
re
a
te
in
M
a
th
e
m
a
ti
c
a
l
S
c
ien
c
e
s
a
n
d
c
o
m
p
lete
d
h
e
r
u
n
d
e
rg
ra
d
u
a
te
stu
d
ies
i
n
M
a
th
e
m
a
ti
c
s
a
n
d
P
h
y
sic
s
,
sp
e
c
ializin
g
i
n
m
a
th
e
m
a
ti
c
s,
a
t
Ha
ss
a
n
II
Un
iv
e
rsity
i
n
M
o
ro
c
c
o
.
F
u
rt
h
e
rm
o
re
,
sh
e
e
a
rn
e
d
a
m
a
ste
r’s d
e
g
re
e
in
S
c
ien
c
e
a
n
d
Tec
h
n
o
l
o
g
y
in
c
o
m
p
u
ter sc
ien
c
e
fro
m
th
e
sa
m
e
in
stit
u
ti
o
n
.
He
r
re
se
a
rc
h
in
tere
sts
e
n
c
o
m
p
a
ss
a
wid
e
ra
n
g
e
o
f
t
o
p
ics
i
n
c
lu
d
in
g
g
ra
p
h
th
e
o
ry
,
in
f
o
rm
a
ti
o
n
sy
ste
m
s,
a
rti
ficia
l
in
telli
g
e
n
c
e
,
d
a
ta
sc
ien
c
e
,
so
ftwa
re
d
e
v
e
lo
p
m
e
n
t,
d
a
tab
a
se
m
o
d
e
li
n
g
,
b
i
g
d
a
ta,
c
ry
p
to
g
ra
p
h
y
,
a
n
d
n
u
m
e
rica
l
m
e
th
o
d
s
a
n
d
sim
u
latio
n
s.
S
h
e
h
a
s
c
o
n
tri
b
u
ted
e
x
ten
siv
e
l
y
to
th
e
se
field
s
wi
th
o
v
e
r
t
h
a
n
e
ig
h
t
y
p
u
b
li
c
a
ti
o
n
s
i
n
e
ste
e
m
e
d
in
tern
a
ti
o
n
a
l
j
o
u
r
n
a
ls
a
n
d
c
o
n
fe
re
n
c
e
s.
Ad
d
it
io
n
a
ll
y
,
sh
e
p
l
a
y
s
a
p
i
v
o
tal
ro
le
in
c
o
o
r
d
i
n
a
ti
n
g
,
p
a
rti
c
i
p
a
ti
n
g
o
r
a
ss
e
ss
in
g
in
v
a
ri
o
u
s
e
d
u
c
a
ti
o
n
a
l
a
n
d
so
c
i
o
-
e
c
o
n
o
m
ic
o
r
re
a
se
rc
h
p
ro
jec
ts.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
so
u
m
ia.ziti
@fsr.u
m
5
.
a
c
.
m
a
.
Evaluation Warning : The document was created with Spire.PDF for Python.