I
AE
S
I
nte
rna
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
9
,
No
.
1
,
Ma
r
ch
2
0
2
0
,
p
p
.
91
~
99
I
SS
N:
2252
-
8938
,
DOI
: 1
0
.
1
1
5
9
1
/i
j
ai.
v
9
.i
1
.
p
p
91
-
99
91
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Autis
m
spe
ctru
m
diso
rder c
la
ss
ific
a
tion o
n
elect
ro
enc
epha
lo
g
ra
m
sig
na
l using
deep learning
alg
o
rith
m
N.
A
Ali
1
,
A.
R
Sy
a
f
ee
za
2
,
A.
S J
a
a
f
a
r
3
,
M
.
K
M
o
hd
F
it
ri
Alif
4
1,
2,
3
F
a
c
u
l
ty
o
f
El
e
c
tro
n
ics
a
n
d
Co
m
p
u
ter E
n
g
in
e
e
rin
g
,
Un
iv
e
rsiti
T
e
k
n
ik
a
l
M
a
la
y
sia
M
e
la
k
a
,
Ha
n
g
Tu
a
h
Ja
y
a
,
7
6
1
0
0
Du
rian
T
u
n
g
g
a
l,
M
e
lak
a
,
M
a
la
y
si
a
4
Co
n
tr
o
l
a
n
d
M
e
c
h
a
tr
o
n
ics
E
n
g
in
e
e
rin
g
d
e
p
a
rtm
e
n
t,
F
a
c
u
lt
y
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
,
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
a
lay
sia
,
8
1
3
0
0
,
Jo
h
o
r
Ba
h
ru
,
Jo
h
o
r,
M
a
lay
sia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
No
v
1
0
,
2
0
19
R
ev
i
s
ed
Feb
6
,
2
0
2
0
A
cc
ep
ted
Feb
1
7
,
2
0
2
0
A
u
ti
s
m
S
p
e
c
tru
m
Diso
rd
e
r
(A
S
D)
is
a
n
e
u
r
o
d
e
v
e
lo
p
m
e
n
tal
th
a
t
im
p
a
c
t
th
e
so
c
ial
in
tera
c
ti
o
n
a
n
d
c
o
m
m
u
n
ica
ti
o
n
sk
il
ls.
Dia
g
n
o
sis
o
f
ASD
is
o
n
e
o
f
th
e
d
if
f
icu
lt
p
ro
b
lem
s
f
a
c
in
g
re
se
a
rc
h
e
rs.
T
h
is
re
se
a
rc
h
wo
rk
a
im
e
d
to
re
v
e
a
l
th
e
d
if
f
e
re
n
t
p
a
tt
e
rn
b
e
tw
e
e
n
a
u
ti
stic
a
n
d
n
o
rm
a
l
c
h
il
d
re
n
v
ia
e
lec
tro
e
n
c
e
p
h
a
lo
g
ra
m
(EE
G
)
b
y
u
sin
g
t
h
e
d
e
e
p
lea
rn
in
g
a
lg
o
rit
h
m
.
T
h
e
b
ra
i
n
sig
n
a
l
d
a
tab
a
se
u
se
d
p
a
tt
e
rn
re
c
o
g
n
it
io
n
w
h
e
re
th
e
e
x
trac
ted
f
e
a
tu
re
s
w
il
l
u
n
d
e
rg
o
th
e
m
u
lt
il
a
y
e
r
p
e
rc
e
p
tro
n
n
e
tw
o
rk
f
o
r
th
e
c
las
sif
ica
ti
o
n
p
r
o
c
e
ss
.
T
h
e
p
ro
m
isin
g
m
e
th
o
d
to
p
e
rf
o
r
m
th
e
c
las
si
f
ica
ti
o
n
is
th
ro
u
g
h
a
d
e
e
p
lea
rn
in
g
a
lg
o
rit
h
m
,
w
h
ich
is
c
u
rre
n
tl
y
a
w
e
ll
-
k
n
o
w
n
a
n
d
su
p
e
ri
o
r
m
e
th
o
d
i
n
t
h
e
p
a
tt
e
rn
re
c
o
g
n
it
io
n
f
ield
.
T
h
e
p
e
rf
o
r
m
a
n
c
e
m
e
a
su
re
f
o
r
th
e
c
la
ss
if
ic
a
ti
o
n
w
o
u
ld
b
e
th
e
a
c
c
u
ra
c
y
.
T
h
e
h
ig
h
e
r
p
e
rc
e
n
tag
e
m
e
a
n
s
th
e
m
o
re
e
ffe
c
ti
v
e
n
e
ss
f
o
r
th
e
A
S
D
d
iag
n
o
sis.
T
h
is
c
a
n
b
e
se
e
n
a
s
th
e
g
ro
u
n
d
w
o
rk
f
o
r
a
p
p
ly
in
g
a
n
e
w
a
lg
o
rit
h
m
f
o
r
f
u
rth
e
r
d
e
v
e
lo
p
m
e
n
t
d
iag
n
o
sis
o
f
a
u
ti
sm
to
se
e
h
o
w
th
e
trea
t
m
e
n
t
is w
o
rk
in
g
a
s
w
e
ll
in
f
u
tu
re
.
K
ey
w
o
r
d
s
:
Au
ti
s
m
s
p
ec
tr
u
m
d
is
o
r
d
er
Dee
p
l
ea
r
n
in
g
E
lectr
o
en
ce
p
h
alo
g
r
a
m
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
:
Nu
r
A
lis
a
Ali,
Dep
ar
t
m
en
t o
f
C
o
m
p
u
ter
E
n
g
i
n
ee
r
in
g
,
Facu
lt
y
o
f
E
lectr
o
n
ics an
d
C
o
m
p
u
ter
E
n
g
i
n
ee
r
i
n
g
,
UT
eM
,
7
6
1
0
0
Han
g
T
u
ah
J
a
y
a,
Ma
lac
ca
,
Ma
la
y
s
ia.
E
m
ail:
ali
s
a@
u
te
m
.
ed
u
.
m
y
1.
I
NT
RO
D
UCT
I
O
N
Au
ti
s
m
Sp
ec
tr
u
m
D
is
o
r
d
er
(
ASD)
is
a
s
y
n
d
r
o
m
e
th
at
ad
v
er
s
el
y
a
f
f
ec
t
a
ch
i
ld
w
h
er
e
t
h
e
b
eh
av
io
r
al
s
y
m
p
to
m
s
s
tar
t
to
ap
p
ea
r
d
u
r
i
n
g
th
e
f
ir
s
t
y
ea
r
o
f
li
f
e
[
1
]
.
T
h
is
ea
r
l
y
ch
ild
h
o
o
d
o
n
s
et
in
c
lu
d
es
s
y
m
p
to
m
s
s
u
c
h
as
lac
k
i
n
s
o
cial
i
n
ter
ac
tio
n
an
d
v
er
y
s
lo
w
la
n
g
u
a
g
e
s
k
ill
s
d
ev
elo
p
m
en
t
as
s
tated
b
y
r
esear
ch
er
s
[2
-
3]
.
A
co
n
tin
u
o
u
s
ch
ar
ac
ter
s
an
d
b
eh
av
io
r
al
ass
e
s
s
m
e
n
t
is
co
n
d
u
cted
b
y
s
p
ec
iali
s
t
in
o
r
d
er
to
d
etec
t
au
tis
tic
p
r
esen
ce
in
a
ch
ild
.
A
d
o
cu
m
en
ted
an
al
y
s
is
d
o
n
e
b
y
p
ed
iatr
ics
s
tated
th
at,
an
au
ti
s
tic
ch
i
ld
at
ap
p
r
o
x
im
a
tel
y
2
4
m
o
n
t
h
s
,
ar
e
s
till
u
n
ab
le
t
o
p
r
o
d
u
ce
t
w
o
m
ea
n
i
n
g
f
u
l
wo
r
d
th
at
d
o
n
o
t
in
v
o
lv
e
i
m
ita
tin
g
an
d
r
ep
ea
tin
g
.
Desp
ite
s
o
m
u
ch
r
e
s
ea
r
ch
b
ei
n
g
co
n
d
u
cted
,
t
h
e
e
x
ac
t
f
ac
to
r
s
to
w
h
y
th
i
s
d
is
o
r
d
er
o
cc
u
r
s
r
e
m
ai
n
u
n
a
n
s
w
er
ed
.
As
o
f
w
h
y
t
h
i
s
at
y
p
ical
b
eh
a
v
io
r
is
v
er
y
d
if
f
ic
u
lt
to
d
etec
t
i
s
m
a
y
b
e
d
u
e
to
th
e
b
ar
el
y
n
o
ticea
b
le
ch
an
g
es
o
f
th
e
p
r
i
m
ar
y
n
e
u
r
al
i
m
p
air
m
en
t
its
elf
.
T
h
e
r
elatio
n
o
f
A
SD
w
i
th
E
E
G
s
ig
n
al
is
t
h
er
e
is
a
s
ig
n
i
f
ica
n
t
d
ec
lin
e
o
f
E
E
G
co
m
p
le
x
it
y
p
er
ce
iv
ed
in
a
u
ti
s
tic
c
h
ild
.
T
h
e
n
o
te
w
o
r
t
h
y
d
if
f
er
e
n
ce
s
w
er
e
o
b
s
er
v
ed
b
et
w
ee
n
b
r
ain
r
eg
io
n
i
n
th
e
r
i
g
h
t
h
e
m
is
p
h
er
e
an
d
th
e
ce
n
tr
al
co
r
tex
r
ep
r
esen
ted
b
y
[4
]
.
E
E
G
s
i
g
n
al
s
ar
e
elec
t
r
ical
v
o
ltag
e
tr
i
g
g
er
ed
o
n
th
e
elec
tr
o
d
es
b
y
b
r
ain
elec
tr
o
m
ag
n
etic
s
i
g
n
al
s
(
B
E
MS)
[
5
]
.
A
t
t
h
e
h
o
s
p
ital,
a
s
p
ec
ialized
tech
n
ician
’
s
m
ea
s
u
r
es,
m
ar
k
s
an
d
p
u
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
I
n
t J
A
r
ti
f
I
n
tell
,
Vo
l.
9
,
No
.
1
,
Ma
r
ch
20
20
:
91
–
99
92
ab
o
u
t
1
6
-
2
5
elec
tr
o
d
es
o
n
th
e
p
atie
n
ts
.
Ma
n
y
r
esear
ch
s
h
o
w
s
th
at
t
h
e
E
E
G
s
i
g
n
a
ls
,
w
h
ic
h
co
n
s
i
s
ts
o
f
A
lp
h
a,
B
eta,
Gam
m
a,
Delta
a
n
d
T
h
eta,
o
f
A
S
D
ch
ild
i
s
n
o
t t
h
e
s
a
m
e
as c
o
m
p
ar
ed
to
n
o
n
-
A
S
D
ch
ild
[
6
]
.
Dee
p
L
ea
r
n
i
n
g
is
a
s
u
b
s
et
o
f
m
ac
h
in
e
lear
n
in
g
.
A
s
m
e
n
tio
n
ed
b
y
r
esear
ch
er
s
[7
-
8]
,
d
ee
p
lear
n
in
g
in
cl
u
d
es
t
h
e
u
s
e
o
f
a
lo
t
o
f
h
i
d
d
en
n
eu
r
o
n
s
an
d
la
y
er
s
,
u
s
u
a
ll
y
m
o
r
e
t
h
a
n
t
w
o
,
a
s
a
n
ar
ch
i
tectu
r
al
b
en
e
f
it,
i
n
ad
d
itio
n
w
it
h
n
e
w
tr
ai
n
i
n
g
p
l
atf
o
r
m
;
w
h
e
n
d
ata
av
ailab
ilit
y
in
cr
ea
s
e
s
,
d
ee
p
lear
n
in
g
s
y
s
te
m
s
ca
n
d
ev
elo
p
g
r
ad
u
all
y
a
n
d
f
il
l
in
t
h
e
g
ap
s
w
h
er
e
h
u
m
a
n
i
n
ter
p
r
etatio
n
is
n
o
t
p
o
s
s
ib
le.
R
eg
ar
d
les
s
it
b
ein
g
m
ac
h
in
e
o
r
d
ee
p
lear
n
in
g
,
b
o
th
co
n
s
is
t
s
o
f
s
u
p
er
v
is
ed
lear
n
i
n
g
an
d
u
n
s
u
p
er
v
is
ed
lear
n
in
g
.
S
u
p
er
v
is
ed
lear
n
in
g
i
s
d
ev
elo
p
in
g
p
r
ed
ictiv
e
m
et
h
o
d
b
ased
o
n
i
n
p
u
t
an
d
o
u
tp
u
t
d
ata.
W
h
ile
u
n
s
u
p
er
v
is
ed
lear
n
i
n
g
is
g
r
o
u
p
i
n
g
a
n
d
in
ter
p
r
etin
g
d
ata
b
ased
o
n
in
p
u
t
d
ata
o
n
l
y
.
Ho
w
ev
er
,
th
e
s
t
u
d
y
co
n
d
u
c
te
d
b
y
[
9
]
m
en
t
io
n
ed
th
at
d
ee
p
lear
n
in
g
ca
n
b
e
an
ex
ce
lle
n
t
to
o
l
to
b
e
ap
p
lied
in
clin
ical
b
r
ain
i
m
a
g
i
n
g
.
T
h
i
s
i
s
f
u
r
t
h
er
s
u
p
p
o
r
ted
b
y
s
tu
d
ie
s
co
n
d
u
cted
b
y
[
1
0
]
s
tati
n
g
th
a
t
d
ee
p
lear
n
i
n
g
i
s
a
v
er
y
r
eliab
le
m
e
th
o
d
to
q
u
ick
l
y
ad
d
r
ess
p
atie
n
t
h
ea
lt
h
co
n
d
i
tio
n
b
y
clas
s
i
f
y
i
n
g
o
f
d
is
ea
s
e
s
,
d
el
ete
r
ed
u
n
d
an
c
y
b
u
t
a
t
t
h
e
s
a
m
e
ti
m
e
m
ai
n
tai
n
s
co
r
r
ec
tn
e
s
s
a
n
d
p
r
ec
is
io
n
o
f
d
is
ea
s
e
id
en
ti
f
icatio
n
.
E
v
en
t
u
all
y
,
t
h
e
p
r
o
ce
s
s
o
f
d
etec
tin
g
b
ec
o
m
es
d
is
ea
s
es
le
s
s
ti
m
e
co
n
s
u
m
in
g
.
Hea
lt
h
ca
r
e
p
er
s
o
n
n
el
ca
n
b
e
ce
r
tai
n
o
n
th
eir
d
iag
n
o
s
i
s
in
t
h
e
d
ec
is
io
n
-
m
ak
in
g
s
ta
g
e.
T
h
is
p
ap
er
is
o
r
g
an
ized
as
f
o
llo
w
s
:
T
h
e
n
e
x
t
s
ec
t
io
n
d
es
cr
ib
es
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
an
d
t
h
e
d
atab
ase
u
s
ed
.
T
h
e
r
esu
lts
a
n
d
an
al
y
s
is
ar
e
s
h
o
w
n
i
n
Sectio
n
3
.
Fin
all
y
,
Sect
io
n
4
co
n
clu
d
e
s
th
i
s
p
ap
er
.
T
h
e
n
u
m
b
er
o
f
c
h
ild
r
en
d
etec
ted
w
i
th
Au
t
is
m
Sp
ec
tr
u
m
D
i
s
o
r
d
er
(
A
SD)
h
as
b
ee
n
i
n
cr
ea
s
in
g
i
n
t
h
e
p
ast
f
e
w
y
ea
r
s
as
s
tated
b
y
t
h
e
Min
is
tr
y
o
f
Hea
lth
,
Ma
la
y
s
ia.
A
tte
n
tio
n
h
as
b
ee
n
b
r
o
u
g
h
t
to
th
i
s
n
eu
r
o
d
ev
elo
p
m
e
n
tal
i
m
p
air
m
en
t
d
is
ea
s
e
d
u
e
to
t
h
e
u
n
k
n
o
w
n
ca
u
s
e
o
f
t
h
is
d
i
s
o
r
d
er
.
Diag
n
o
s
in
g
ASD
b
ef
o
r
e
th
e
ag
e
o
f
th
r
e
e
i
s
v
er
y
c
h
alle
n
g
in
g
.
T
h
is
i
s
b
ec
au
s
e;
ASD
is
h
ig
h
l
y
as
s
o
ciate
d
w
ith
ei
th
er
t
h
e
o
v
er
-
ab
u
n
d
a
n
ce
o
r
v
er
y
lo
w
n
e
u
r
o
n
co
n
n
ec
tio
n
o
f
th
e
b
r
ain
w
ir
es.
Ho
w
e
v
e
r
,
th
e
f
o
r
m
atio
n
o
f
eith
er
t
h
e
o
v
er
-
ab
u
n
d
an
ce
o
r
v
er
y
lo
w
n
e
u
r
o
n
co
n
n
ec
tio
n
d
u
r
in
g
c
h
ild
g
r
o
w
t
h
i
s
a
v
er
y
s
lo
w
p
r
o
g
r
ess
t
h
at
it
is
h
ar
d
l
y
n
o
ticea
b
le
[
1
1
]
.
T
h
e
s
i
tu
atio
n
i
s
m
u
c
h
m
o
r
e
al
ar
m
in
g
s
i
n
ce
t
h
er
e
is
n
o
t
y
et
a
q
u
an
ti
f
iab
le
tech
n
o
lo
g
y
u
s
e
d
to
ex
ac
tl
y
ad
d
r
ess
th
is
d
i
s
o
r
d
er
.
ASD
is
n
o
t
ex
ac
t
l
y
g
e
n
etica
ll
y
r
elate
d
,
th
u
s
s
p
ec
ialis
t
ca
n
n
o
t
ex
ac
tl
y
p
r
ed
ict
if
t
h
e
ch
ild
i
s
au
tis
tic
o
r
n
o
t
s
in
ce
b
ir
th
.
Ma
n
y
r
ea
l
-
li
f
e
ca
s
es
s
h
o
w
t
h
at
if
a
c
h
ild
h
as
a
u
tis
t
ic
s
ib
lin
g
s
t
h
at
w
i
ll
n
o
t
n
ec
es
s
ar
il
y
m
ea
n
th
e
ch
ild
w
ill
b
e
au
t
is
tic
a
s
well.
Fu
r
t
h
er
m
o
r
e,
ac
co
r
d
in
g
to
M
y
t
h
ili
[
1
2
]
,
ea
ch
ch
ild
w
it
h
au
tis
m
s
h
o
w
s
v
er
y
d
is
tin
g
u
is
h
ab
le
b
eh
a
v
io
r
al
s
y
m
p
to
m
s
.
Fo
r
ex
a
m
p
le,
s
o
m
e
ch
ild
is
v
er
y
w
ell
co
n
v
er
s
ed
,
w
h
ile
s
o
m
e
b
ar
el
y
u
tter
s
w
o
r
d
s
.
So
m
e
ch
ild
w
it
h
au
ti
s
m
i
s
v
er
y
m
u
c
h
attac
h
ed
an
d
s
o
m
e
j
u
s
t
d
id
n
o
t
k
n
o
w
h
o
w
to
e
x
p
r
ess
e
m
o
tio
n
s
.
So
m
e
a
u
tis
t
ic
ch
il
d
ev
en
h
a
s
th
e
s
a
m
e
b
eh
av
i
o
r
as
n
o
r
m
a
l
ch
ild
w
it
h
v
er
y
m
i
n
i
m
al
b
eh
a
v
io
r
d
if
f
er
e
n
ce
.
T
h
er
e
is
a
v
er
y
th
i
n
b
ar
r
ier
th
at
s
ep
ar
ates
t
h
e
n
o
r
m
al
w
it
h
au
tis
t
ic.
T
h
u
s
,
t
h
es
e
s
m
al
l
b
ar
r
ier
s
o
f
b
eh
av
io
r
d
if
f
er
en
ce
s
ar
e
th
e
m
ain
ch
a
l
len
g
es t
h
at
k
ee
p
r
esear
ch
er
s
’
atten
tio
n
to
th
is
s
y
n
d
r
o
m
e.
E
lectr
o
en
ce
p
h
alo
g
r
a
m
(
E
E
G)
is
a
m
et
h
o
d
u
s
ed
i
n
p
u
r
p
o
s
e
t
o
f
in
d
o
u
t
t
h
e
b
r
ain
f
u
n
ctio
n
r
eg
ar
d
less
th
e
m
e
n
tal
s
tate
s
o
f
t
h
e
i
n
d
i
v
id
u
als
[
1
3
]
.
E
E
G
p
r
o
v
id
es
r
o
b
u
s
t
p
ar
a
m
eter
s
w
h
er
e
it
ca
n
ex
a
m
i
n
e
t
h
e
b
r
ai
n
ac
tiv
it
y
at
r
est
s
tate
an
d
ac
ti
v
e
s
tate.
E
E
G
ca
n
also
s
h
o
w
w
h
i
ch
p
ar
t
o
f
th
e
b
r
ain
i
s
ac
ti
v
e
wh
en
d
o
in
g
s
p
ec
i
f
ic
task
s
an
d
th
e
e
f
f
ec
t
o
f
t
h
o
s
e
s
p
ec
if
ic
tas
k
s
.
T
h
er
e
ar
e
f
iv
e
t
y
p
es
o
f
f
r
eq
u
en
c
y
b
an
d
s
as
s
o
ciate
d
w
it
h
E
E
G
s
ig
n
al
n
a
m
el
y
t
h
e
alp
h
a,
b
eta,
g
a
m
m
a,
th
e
ta
an
d
d
elta.
As
af
o
r
e
m
e
n
tio
n
ed
ea
ch
o
f
th
i
s
b
r
ain
w
av
e
h
as
t
h
eir
o
w
n
f
r
eq
u
e
n
c
y
r
a
n
g
e
an
d
ea
c
h
is
as
s
o
ciate
d
w
it
h
its
o
w
n
p
h
y
s
io
lo
g
ical
c
h
ar
ac
ter
is
tic
s
wh
ich
g
o
v
er
n
s
it
s
o
w
n
co
g
n
iti
v
e
p
r
o
p
er
ties
.
D
u
r
in
g
E
E
G
r
ec
o
r
d
in
g
,
p
atien
t
w
ill
b
e
s
h
o
w
n
s
er
ies
o
f
s
i
m
u
lat
io
n
w
h
ile
s
itt
in
g
d
o
w
n
.
A
ll
t
h
ese
s
i
m
u
latio
n
s
h
a
v
e
a
c
er
tain
w
a
y
o
f
a
f
f
ec
ti
n
g
all
t
h
e
f
i
v
e
f
r
eq
u
e
n
c
y
b
an
d
s
to
s
t
i
m
u
late.
I
n
c
h
ild
w
it
h
ASD,
all
t
h
ese
f
i
v
e
t
y
p
es
o
f
f
r
eq
u
en
c
y
b
an
d
ar
e
i
n
ter
r
u
p
ted
an
d
d
o
es
n
o
t
s
h
o
w
r
es
u
lt
i
n
a
cc
o
r
d
an
ce
w
it
h
t
h
e
g
iv
e
n
s
i
m
u
lat
io
n
.
T
h
e
A
lp
h
a
b
an
d
i
s
k
n
o
w
n
to
b
e
av
ailab
le
w
h
e
n
a
p
er
s
o
n
i
s
i
n
r
elax
ed
m
o
o
d
an
d
ar
e
a
w
a
k
e
.
T
h
ey
ar
e
also
ass
o
ciate
d
w
it
h
ti
m
in
g
a
n
d
co
g
n
it
iv
e
i
n
h
ib
itio
n
.
Ne
x
t,
th
e
b
eta
b
an
d
s
ar
e
en
g
a
g
ed
w
it
h
aler
t
n
es
s
,
ac
tiv
e
task
i
n
v
o
lv
e
m
e
n
t
a
n
d
m
o
to
r
b
eh
av
io
r
.
T
h
e
g
a
m
m
a
b
a
n
d
i
s
an
ea
r
l
y
s
e
n
s
o
r
y
r
e
s
p
o
n
d
,
w
h
er
e
it
h
elp
s
w
it
h
f
ea
t
u
r
e
b
in
d
i
n
g
in
s
e
n
s
o
r
y
p
r
o
ce
s
s
in
g
.
F
in
al
l
y
,
th
e
d
elta
b
an
d
is
ac
tiv
e
d
u
r
i
n
g
d
ee
p
s
le
ep
an
d
ev
e
n
t
t
h
at
is
r
elate
d
to
s
lo
w
w
a
v
es
s
u
ch
a
s
d
etec
tio
n
o
f
atten
tio
n
an
d
w
a
k
e
f
u
l
n
e
s
s
.
T
h
e
tab
le
co
m
p
u
tes
t
h
e
E
E
G
s
ig
n
al
w
a
v
e
f
o
r
m
o
f
h
u
m
a
n
b
ei
n
g
is
s
h
o
w
n
i
n
T
ab
le
1
.
W
h
ile
Fi
g
u
r
e
1
ill
u
s
tr
ates
s
a
m
p
le
o
f
E
E
G
s
i
g
n
al
f
r
o
m
a
n
o
r
m
al
p
er
s
o
n
.
T
ab
le
1
.
E
E
G
B
r
ain
Sig
n
al
o
f
No
r
m
a
l H
u
m
a
n
B
ein
g
[
1
4
]
L
e
v
e
l
F
r
e
q
u
e
n
c
y
r
a
n
g
e
A
p
p
r
o
x
i
mat
e
EEG
l
a
b
e
l
1
64
-
1
2
8
H
z
H
i
g
h
G
a
mm
a
2
32
-
6
4
H
z
G
a
mm
a
3
16
-
3
2
H
z
B
e
t
a
4
8
-
1
6
H
z
A
l
p
h
a
5
4
-
8
H
z
T
h
e
t
a
6
0
-
4
H
z
D
e
l
t
a
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
A
u
tis
m
s
p
ec
tr
u
m
d
is
o
r
d
er c
la
s
s
ifica
tio
n
o
n
elec
tr
o
en
ce
p
h
a
lo
g
r
a
m
s
ig
n
a
l u
s
in
g
…
(
N
.
A
A
li
)
93
A
cr
o
s
s
m
a
n
y
s
t
u
d
ies,
it
is
s
h
o
w
n
th
a
t
a
c
h
ild
w
it
h
ASD
is
co
n
s
is
ten
t
in
p
r
o
d
u
cin
g
at
y
p
i
ca
l
n
e
u
r
al
ac
tiv
it
y
d
u
r
in
g
t
h
e
f
r
eq
u
e
n
c
y
b
an
d
s
i
m
u
latio
n
.
Hen
ce
,
E
E
G
s
ig
n
al
r
e
v
ea
ls
h
ig
h
h
o
p
e
in
s
ee
k
in
g
m
o
r
e
f
i
n
d
in
g
s
to
lo
ca
lize
th
e
ex
ac
t
ar
ea
in
s
id
e
th
e
b
r
ain
w
h
er
e
th
e
alter
atio
n
b
eg
in
s
.
T
h
is
is
f
u
r
th
er
s
u
p
p
o
r
ted
b
y
[
1
4
]
,
s
tati
n
g
th
a
t
b
y
ca
lc
u
lati
n
g
h
o
w
q
u
ic
k
t
h
e
b
r
ai
n
r
ea
ct
to
v
i
s
u
al
an
d
a
u
d
io
s
ti
m
u
l
i
w
il
l
p
o
s
itiv
el
y
i
m
p
ac
t
i
n
class
i
f
y
in
g
a
u
tis
m
a
n
d
d
iag
n
o
s
e
th
e
d
is
o
r
d
er
ea
r
lier
.
Dee
p
lear
n
in
g
co
m
b
i
n
es
f
ea
t
u
r
e
ex
tr
ac
tio
n
an
d
class
i
f
icatio
n
p
r
o
ce
s
s
[
1
5
]
.
Featu
r
e
ex
tr
ac
tio
n
is
th
e
p
r
o
ce
s
s
o
f
p
r
o
d
u
ci
n
g
a
d
en
s
e
b
u
t
h
ig
h
l
y
m
ea
n
i
n
g
f
u
l
d
i
g
ital
p
r
esen
tatio
n
o
f
s
o
m
eo
n
e
f
u
n
d
a
m
en
tal
b
io
m
etr
ic
attr
ib
u
te.
I
n
t
y
p
ical
m
ac
h
i
n
e
l
ea
r
n
in
g
,
t
h
e
i
m
p
o
r
tan
t
f
ea
t
u
r
e
n
ee
d
s
to
b
e
m
a
n
u
al
l
y
e
x
tr
ac
te
d
an
d
class
i
f
y
u
s
in
g
o
n
e
o
r
m
o
r
e
class
i
f
ier
alg
o
r
it
h
m
.
I
n
o
r
d
er
to
ex
tr
ac
t
f
ea
tu
r
es,
th
er
e
ar
e
t
w
o
m
o
s
t
p
o
p
u
lar
m
eth
o
d
s
u
s
ed
w
h
ich
ar
e
th
e
f
ilter
m
et
h
o
d
s
an
d
th
e
w
r
ap
p
er
m
e
th
o
d
s
.
A
n
e
x
a
m
p
le
o
f
f
ilter
m
et
h
o
d
s
an
d
w
r
ap
p
er
m
et
h
o
d
s
ar
e
ch
i
s
q
u
ar
ed
test
an
d
r
ec
u
r
s
i
v
e
f
ea
tu
r
e
eli
m
in
at
io
n
al
g
o
r
ith
m
r
esp
ec
tiv
el
y
.
As
f
o
r
clas
s
if
ier
,
m
o
s
t
co
m
m
o
n
class
i
f
ier
s
ar
e
Fi
s
h
er
L
i
n
ea
r
Dis
cr
i
m
i
n
a
n
t
An
al
y
s
is
(
F
L
D
A
)
,
R
an
d
o
m
Fo
r
e
s
t
a
n
d
Su
p
p
o
r
t
Vec
to
r
Ma
ch
in
e.
I
n
m
an
y
r
esear
c
h
,
in
o
r
d
er
to
o
b
t
ain
t
h
e
b
est cla
s
s
if
icatio
n
ac
c
u
r
ac
y
,
t
w
o
o
r
m
o
r
e
class
if
icatio
n
alg
o
r
it
h
m
w
ill b
e
u
s
ed
[
1
6
]
.
T
h
u
s
,
it
is
a
g
r
ee
d
th
at
s
i
n
ce
d
ee
p
lear
n
i
n
g
c
an
s
u
cc
es
s
f
u
ll
y
co
m
b
in
e
f
ea
t
u
r
e
ex
tr
ac
tio
n
an
d
class
i
f
ier
to
g
et
h
er
,
it c
an
d
ec
r
ea
s
e
th
e
e
x
p
er
i
m
e
n
ti
n
g
a
n
d
s
ele
ctin
g
f
ea
t
u
r
e
s
p
r
o
ce
s
s
.
On
e
o
f
th
e
m
o
s
t
p
o
p
u
lar
d
ee
p
lear
n
in
g
m
et
h
o
d
s
is
t
h
e
C
o
n
v
o
lu
tio
n
al
Ne
u
r
al
Net
w
o
r
k
,
w
h
ich
is
also
ca
lled
as
C
NN
[
1
7
]
.
A
s
m
en
ti
o
n
ed
b
ef
o
r
e,
lik
e
an
y
o
t
h
er
d
ee
p
lear
n
in
g
m
et
h
o
d
,
th
e
f
ea
tu
r
e
th
at
is
n
ee
d
ed
to
b
e
ex
tr
ac
ted
d
o
es
n
o
t
h
a
v
e
to
b
e
o
u
tlin
ed
.
T
h
e
C
NN
al
g
o
r
ith
m
w
il
l
au
to
m
atica
ll
y
e
x
tr
icate
t
h
e
m
o
s
t
d
is
tin
g
u
is
h
i
n
g
ch
ar
ac
ter
i
s
tics
w
it
h
o
u
t i
n
cl
u
d
in
g
t
h
e
o
p
ti
m
iz
atio
n
a
n
d
p
r
o
p
ag
atio
n
s
tep
s
.
T
h
is
is
m
ad
e
p
o
s
s
ib
le
ev
en
w
it
h
o
u
t
u
s
i
n
g
f
ea
tu
r
e
e
x
tr
ac
to
r
d
u
e
to
th
e
tr
ai
n
i
n
g
o
f
d
ata
s
tep
p
r
esen
t
in
th
e
C
NN
alg
o
r
it
h
m
.
T
h
e
ex
a
m
p
le
s
o
f
ap
p
licatio
n
s
u
til
iz
ed
C
NN
th
r
o
u
g
h
s
t
u
d
ies b
y
[
1
8
–
2
1
]
.
Dee
p
lear
n
i
n
g
p
ar
ticu
lar
l
y
C
NN
is
d
ep
lo
y
ed
d
u
e
to
t
h
e
ca
p
ab
ilit
y
o
f
t
h
e
al
g
o
r
ith
m
t
o
ac
tu
all
y
r
ec
o
g
n
ize
u
n
iq
u
e
f
ea
t
u
r
es
o
f
e
lectr
ical
b
r
ain
w
a
v
e
p
atter
n
f
r
o
m
E
E
G
s
i
g
n
al.
Ma
n
y
s
t
u
d
ies
h
av
e
s
h
o
w
n
s
u
cc
e
s
s
in
i
n
te
g
r
atin
g
E
E
G
s
i
g
n
al
i
n
to
th
e
C
NN
al
g
o
r
it
h
m
.
E
E
G
is
n
o
n
-
li
n
ea
r
an
d
h
i
g
h
l
y
in
tr
icate
s
ig
n
al
th
at
r
ec
o
r
d
s
i
m
p
o
r
tan
t
d
ata.
T
h
is
d
ata
p
o
r
tr
a
y
s
t
h
e
d
if
f
er
en
ce
s
o
f
o
n
e
h
u
m
an
b
e
i
n
g
to
a
n
o
th
er
.
Si
n
ce
E
E
G
is
v
er
y
co
m
p
le
x
an
d
n
o
n
-
l
in
ea
r
i
n
n
at
u
r
e,
m
a
n
y
li
n
ea
r
clas
s
i
f
ier
m
eth
o
d
s
co
u
ld
n
o
t
ac
c
u
r
atel
y
d
etec
t
t
h
is
s
ig
n
a
l
as
E
E
G.
T
h
u
s
,
C
NN
h
as
b
ec
o
m
e
in
cr
ea
s
in
g
l
y
p
o
p
u
lar
an
d
i
s
b
ec
o
m
i
n
g
a
h
ig
h
l
y
p
r
ef
er
r
ed
n
e
w
tech
n
iq
u
e
in
s
ig
n
al
d
etec
tio
n
an
d
clas
s
i
f
icatio
n
.
T
h
e
t
h
eo
r
y
o
f
C
NN
is
m
ai
n
l
y
in
s
p
ir
ed
b
y
t
h
e
b
r
ain
o
f
h
u
m
a
n
b
ei
n
g
.
C
NN
is
a
n
eu
r
al
n
et
w
o
r
k
t
h
at
co
n
s
i
s
ts
o
f
m
u
l
tila
y
er
p
er
ce
p
tr
o
n
(
ML
P
)
.
E
ac
h
m
u
l
tila
y
er
p
er
ce
p
tr
o
n
s
er
v
es
its
o
w
n
s
p
ec
ial
p
u
r
p
o
s
es
w
ith
its
o
w
n
ar
r
an
g
e
m
en
t
o
f
e
x
ec
u
tio
n
a
s
s
h
o
w
n
in
Fi
g
u
r
e
2
.
I
n
all
M
L
P
,
th
er
e
m
u
s
t
b
e
an
in
p
u
t
la
y
er
w
h
ich
co
m
es
f
r
o
m
a
n
in
p
u
t
d
ata,
a
m
i
n
i
m
u
m
o
f
o
n
e
h
id
d
en
la
y
er
an
d
f
in
a
ll
y
a
n
o
u
t
p
u
t
la
y
er
t
h
at
w
il
l
p
r
ed
icts
th
e
o
u
tp
u
t
f
r
o
m
t
h
e
in
p
u
t la
y
er
[
2
2
]
.
Fig
u
r
e
1
.
Sa
m
p
le
o
f
E
E
G
s
i
g
n
al
f
r
o
m
n
o
r
m
al
p
er
s
o
n
Fig
u
r
e
2
.
Mu
ltil
a
y
er
P
er
ce
p
tr
o
n
Net
w
o
r
k
(
M
L
P
)
T
h
er
e
ar
e
s
ev
er
al
la
y
er
s
th
a
t
ar
e
co
m
m
o
n
l
y
k
n
o
w
n
i
n
C
NN
s
u
c
h
as
co
n
v
o
l
u
tio
n
al
la
y
er
-
wh
ich
s
et
o
f
lear
n
ab
le
f
i
lter
s
k
n
o
w
n
as
k
er
n
el
an
d
p
r
o
d
u
ce
o
u
tp
u
t
f
e
atu
r
e
m
ap
s
t
h
at
w
il
l
b
e
o
u
t
p
u
t
f
o
r
n
ex
t
la
y
er
.
A
cti
v
atio
n
la
y
er
s
i
m
p
li
f
ies
b
a
ck
-
p
r
o
p
ag
atio
n
an
d
r
ed
u
ce
r
e
d
u
n
d
an
cie
s
.
P
o
o
lin
g
la
y
er
s
i
s
to
d
ec
r
ea
s
e
s
ize
o
f
f
ea
t
u
r
e
m
ap
s
f
r
o
m
co
n
v
o
lu
ti
o
n
al
la
y
er
.
F
u
ll
y
co
n
n
ec
ted
l
a
y
er
o
r
d
en
s
e
la
y
er
is
r
e
s
p
o
n
s
ib
le
f
o
r
th
e
o
u
tp
u
t
p
r
ed
ictio
n
s
an
d
s
o
f
t
m
a
x
la
y
er
w
h
ich
i
s
an
ac
ti
v
atio
n
f
u
n
cti
o
n
th
at
p
u
r
p
o
s
e
is
to
f
i
n
d
m
ea
n
s
q
u
ar
e
er
r
o
r
.
T
h
is
C
NN
la
y
er
s
is
e
x
ec
u
ted
in
ca
s
ca
d
in
g
m
an
n
er
.
Fi
g
u
r
e
3
s
h
o
w
s
an
ex
a
m
p
le
o
f
C
NN
la
y
er
s
f
r
o
m
w
o
r
k
s
b
y
[
2
3
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
I
n
t J
A
r
ti
f
I
n
tell
,
Vo
l.
9
,
No
.
1
,
Ma
r
ch
20
20
:
91
–
99
94
T
h
er
e
ar
e
also
f
ea
tu
r
e
m
ap
s
,
w
ei
g
h
ts
,
s
a
m
p
le
s
ize,
A
d
a
m
o
p
tim
izatio
n
a
n
d
ep
o
ch
t
h
at
n
ee
d
to
b
e
d
eter
m
in
ed
b
ef
o
r
e
tr
ain
i
n
g
p
r
o
ce
s
s
w
h
ic
h
is
co
m
m
o
n
i
n
d
ee
p
lear
n
in
g
.
Mo
s
t
ti
m
e,
w
h
e
n
m
o
r
e
la
y
er
s
ar
e
ad
d
ed
,
th
e
b
etter
th
e
lear
n
in
g
p
r
o
ce
s
s
w
h
ic
h
lead
s
to
b
etter
class
i
f
icatio
n
ac
cu
r
ac
y
.
Fo
r
th
e
h
id
d
en
la
y
er
s
an
d
its
o
u
tp
u
t
la
y
er
s
,
an
ac
ti
v
atio
n
f
u
n
ctio
n
i
s
n
ee
d
ed
.
So
m
e
w
o
r
d
s
s
u
c
h
as
R
e
L
u
,
d
r
o
p
o
u
t
an
d
f
latte
n
is
r
elati
v
el
y
n
e
w
f
u
n
ctio
n
t
h
at
e
x
is
t
in
C
N
N
ap
p
licatio
n
[
2
4
-
26]
.
R
eL
u
is
th
e
ac
ti
v
atio
n
f
u
n
ctio
n
u
s
ed
i
n
ev
er
y
co
n
v
o
lu
tio
n
la
y
er
,
d
r
o
p
o
u
t
is
to
p
r
ev
en
t
o
v
er
f
itti
n
g
d
ata
in
f
u
ll
y
co
n
n
ec
te
d
la
y
er
an
d
f
latte
n
la
y
er
th
at
is
u
s
e
to
f
latte
n
d
ata
to
1
D
ar
r
ay
o
n
l
y
.
Fig
u
r
e
3
.
T
h
e
s
tan
d
ar
d
C
NN
L
a
y
er
[
2
3
]
2.
RE
S
E
ARCH
M
E
T
H
O
D
T
h
e
m
et
h
o
d
o
lo
g
y
f
o
r
th
i
s
r
e
s
ea
r
ch
i
s
d
i
v
id
ed
in
to
t
h
r
ee
s
ec
tio
n
s
.
T
h
e
f
i
r
s
t
s
ec
tio
n
d
i
s
cu
s
s
es
t
h
e
ex
is
t
in
g
d
atab
ase,
th
e
s
ec
o
n
d
s
ec
tio
n
d
i
s
cu
s
s
e
s
ap
p
lied
p
r
e
-
p
r
o
ce
s
s
in
g
m
eth
o
d
a
n
d
t
h
e
f
i
n
al
s
ec
tio
n
d
is
c
u
s
s
es
th
e
d
ee
p
lear
n
in
g
d
esi
g
n
.
T
h
is
s
y
s
te
m
r
an
o
n
2
.
3
GHz
I
n
tel
C
o
r
e
i5
2
.
5
GHz
w
i
th
a
n
NVI
DI
A
Ge
Fo
r
ce
p
r
o
ce
s
s
o
r
th
at
h
a
s
6
1
0
M
b
m
e
m
o
r
y
.
T
h
e
P
y
th
o
n
3
.
5
.
2
lan
g
u
a
g
e,
Op
en
So
u
r
ce
C
o
m
p
u
t
er
Vis
io
n
L
ib
r
ar
y
(
Op
en
C
V)
an
d
T
en
s
o
r
f
lo
w
w
e
r
e
u
s
ed
f
o
r
p
r
e
-
p
r
o
ce
s
s
in
g
an
d
to
d
esig
n
t
h
e
C
N
N
m
o
d
el.
2.
1
.
Resea
rc
h f
lo
w
cha
rt
T
h
is
r
esear
ch
ca
r
r
ied
o
u
t
th
e
d
esig
n
o
f
an
alg
o
r
it
h
m
to
d
etec
t
au
tis
m
.
I
t
is
in
te
g
r
ated
b
y
u
s
in
g
E
E
G
s
ig
n
al.
C
NN
tec
h
n
iq
u
e
w
ill
b
e
u
s
ed
in
th
i
s
p
r
o
j
ec
t
as
a
b
in
ar
y
clas
s
i
f
icatio
n
.
I
n
o
r
d
er
to
en
s
u
r
e
th
e
ex
p
ec
ted
r
esu
lt
w
i
ll
b
e
o
b
tain
ed
,
s
e
v
e
r
al
m
aj
o
r
s
tep
s
n
ee
d
to
b
e
c
o
n
d
u
ct
s
u
c
h
as
d
ata
co
llec
ti
o
n
s
,
i
m
p
le
m
e
n
ti
n
g
,
test
i
n
g
,
a
n
d
tr
o
u
b
les
h
o
o
tin
g
.
T
h
ese
s
tep
s
ar
e
u
s
ed
to
an
al
y
z
e
th
e
d
ata
an
d
o
u
tp
u
t.
Fig
u
r
e
4
s
h
o
w
s
th
e
s
tep
s
i
n
p
er
f
o
r
m
in
g
t
h
is
r
esear
c
h
.
Fig
u
r
e
4
.
Ma
in
d
ev
elo
p
m
e
n
t i
n
v
o
l
v
ed
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
A
u
tis
m
s
p
ec
tr
u
m
d
is
o
r
d
er c
la
s
s
ifica
tio
n
o
n
elec
tr
o
en
ce
p
h
a
lo
g
r
a
m
s
ig
n
a
l u
s
in
g
…
(
N
.
A
A
li
)
95
2.
2
.
E
x
is
t
ing
da
t
a
s
et
T
h
e
d
ataset
is
o
b
tain
ed
f
r
o
m
ex
is
t
in
g
d
ataset
f
r
o
m
o
t
h
er
r
esear
ch
er
s
o
f
r
elate
d
p
ap
er
w
o
r
k
.
T
h
e
d
ataset
is
r
eq
u
es
t
f
o
r
m
all
y
b
y
e
m
a
il
to
r
esear
ch
e
s
th
a
t
h
a
v
e
d
o
n
e
r
esear
ch
b
y
u
s
in
g
E
E
G
s
ig
n
al
w
h
ic
h
i
s
o
b
tain
ed
f
r
o
m
Un
iv
er
s
it
y
Ki
n
g
A
b
d
u
l
A
ziz,
J
ed
d
ah
,
Sa
u
d
i
A
r
ab
ia.
T
h
e
d
ataset
o
b
tai
n
ed
co
n
s
is
ts
o
f
t
w
e
n
t
y
f
iles
,
1
2
n
o
r
m
al
an
d
8
f
o
r
d
is
o
r
d
er
s
.
T
h
e
d
ataset
w
a
s
r
ec
o
r
d
ed
in
r
elax
i
n
g
s
tate
i
n
o
r
d
er
to
o
b
tain
as
m
an
y
ar
tif
ac
t
-
f
r
ee
E
E
G
d
ata
as
p
o
s
s
ib
le.
T
h
e
d
ataset
d
iv
id
ed
in
to
t
w
o
g
r
o
u
p
s
:
n
o
r
m
al
g
r
o
u
p
(
t
w
elv
e
s
h
ea
lth
y
v
o
lu
n
teer
s
u
b
j
ec
ts
ag
ed
9
-
1
6
y
ea
r
s
o
ld
)
an
d
au
ti
s
tic
g
r
o
u
p
(
eig
h
t
w
it
h
a
u
tis
m
tr
ai
ts
a
g
ed
1
0
-
1
6
y
ea
r
s
o
ld
)
.
T
h
e
E
E
G
s
ig
n
a
ls
w
as
ac
q
u
ir
ed
at
a
s
a
m
p
lin
g
r
ate
o
f
2
5
6
Hz
u
s
in
g
ac
ti
v
e
elec
tr
o
d
es
an
d
th
e
ac
tiv
e
d
ig
ital
E
E
G
a
m
p
li
f
ier
an
d
r
ec
o
r
d
in
g
s
y
s
t
e
m
f
r
o
m
B
C
I
2
0
0
0
s
o
f
t
w
ar
e.
T
h
e
d
ata
ac
q
u
is
itio
n
s
y
s
te
m
h
a
s
1
6
ch
an
n
el
s
alto
g
eth
er
,
w
h
ic
h
ar
e
lab
eled
b
ased
o
n
th
e
1
0
–
2
0
in
ter
n
atio
n
al
s
ta
n
d
ar
d
ac
q
u
is
itio
n
s
y
s
te
m
.
2.
3
.
P
re
pro
ce
s
s
ing
o
f
E
E
G
s
ig
na
l
On
ce
t
h
e
d
ataset
i
s
o
b
tain
ed
,
th
e
r
a
w
d
ataset
is
p
r
e
-
p
r
o
ce
s
s
ed
.
T
h
e
r
a
w
d
ataset
o
b
tain
e
d
is
in
t
h
e
f
o
r
m
o
f
m
atr
ix
w
h
er
e
it
i
s
ar
r
an
g
ed
a
s
s
u
c
h
C
h
a
n
n
el
(
elec
tr
o
d
es
n
u
m
b
er
)
×
Fre
q
u
en
c
y
(
Hz
)
×
T
im
e
(
s
)
.
E
ac
h
o
f
t
h
e
d
ataset
h
a
s
a
n
eq
u
iv
a
le
n
t
c
h
a
n
n
el
an
d
s
a
m
p
lin
g
f
r
eq
u
en
c
y
w
h
ic
h
is
1
6
elec
tr
o
d
es
ch
an
n
el
an
d
2
5
6
Hz
f
r
eq
u
en
c
y
r
esp
ec
ti
v
el
y
.
Ho
w
e
v
er
,
th
e
ti
m
e
ta
k
en
,
t o
f
th
e
s
a
m
p
lin
g
f
r
eq
u
e
n
c
y
i
s
d
if
f
er
en
t
f
o
r
ea
ch
d
ataset.
Fo
r
ex
a
m
p
le,
d
ataset
p
er
s
o
n
No
r
m
al#
1
is
ar
r
an
g
ed
in
1
6
×2
5
6
×1
1
2
,
w
h
ic
h
m
ea
n
s
t
h
e
s
a
m
p
le
is
tak
e
n
f
o
r
1
1
2
s
.
Fo
r
d
ataset
p
er
s
o
n
No
r
m
al#
2
i
s
ar
r
an
g
ed
in
1
6
×2
5
6
×
9
5
w
h
ic
h
in
d
icate
s
th
a
t th
e
s
a
m
p
le
is
t
ak
en
f
o
r
9
5
s
o
n
l
y
.
A
ll
t
h
e
d
atase
t
i
s
p
r
e
-
p
r
o
ce
s
s
e
d
ac
co
r
d
in
g
to
t
h
e
ti
m
e
ta
k
e
n
to
r
ec
o
r
d
th
e
d
ata.
T
h
is
s
tep
i
s
d
o
n
e
b
y
u
s
i
n
g
MA
T
L
A
B
s
o
f
t
w
ar
e.
T
h
u
s
,
f
o
r
p
er
s
o
n
No
r
m
al#
1
,
it
is
p
r
e
-
p
r
o
ce
s
s
ed
m
an
u
all
y
f
o
r
1
1
2
ti
m
es
an
d
is
s
to
r
ed
in
.
csv
f
i
le
f
o
r
m
at.
T
h
e
f
ile
is
n
a
m
e
r
an
d
o
m
l
y
.
Fo
r
ex
a
m
p
le,
at
ti
m
e,
t
=
1
s
f
o
r
p
er
s
o
n
No
r
m
al#
1
,
it
is
lab
el
as
‘
a1
.
cs
v
’
,
w
h
er
e
‘
a
’
i
s
to
in
d
icate
all
f
ile
f
o
r
p
er
s
o
n
No
r
m
a
l#
1
w
h
ile
‘
1
’
i
n
d
icate
t
h
e
ti
m
e,
t
i
n
1
s
.
At
ti
m
e,
t
=2
s
,
t
h
e
f
ile
is
s
a
v
e
as
‘
a2
.
cs
v
’
.
T
h
is
i
s
d
o
n
e
f
o
r
all
d
ataset.
T
ab
le
2
s
h
o
w
s
th
e
r
an
d
o
m
lab
eli
n
g
f
o
r
ea
ch
d
ataset
u
n
til i
t c
o
m
p
leted
f
o
r
th
e
w
h
o
le
d
ataset
co
n
s
i
s
te
d
o
f
2
0
s
u
b
j
ec
ts
.
T
ab
le
2
.
R
an
d
o
m
lab
elli
n
g
o
f
d
ataset
D
a
t
a
se
t
L
a
b
e
l
N
o
r
mal
#
1
a
N
o
r
mal
#
2
b
A
u
t
i
sm#
1
A
A
u
t
i
sm#
2
B
Nex
t,
th
e
al
l
d
ataset
is
co
m
b
i
n
ed
in
a
f
ile
a
n
d
is
n
a
m
ed
‘
all.
c
s
v
’
f
ile.
T
h
is
m
a
k
es
u
p
a
to
tal
o
f
1
7
,
1
3
6
d
atasets
i
n
a
f
ile.
A
ll
d
ataset
is
n
o
w
at
1
6
×2
5
6
m
a
tr
ix
f
o
r
m
.
T
h
e
n
e
x
t
s
tep
o
f
p
r
e
-
p
r
o
ce
s
s
i
n
g
is
to
lab
el
t
h
e
d
ataset
to
n
o
r
m
al
p
er
s
o
n
o
r
au
tis
tic
p
atie
n
t.
Fo
r
n
o
r
m
al
p
er
s
o
n
,
th
e
lab
el
i
s
s
e
t
to
1
an
d
f
o
r
au
tis
t
ic
p
atien
t
;
it
is
s
et
to
2
.
A
t
t
h
e
f
i
n
al
co
lu
m
n
,
th
e
d
ata
is
s
et
to
eith
er
1
o
r
2
ac
co
r
d
in
g
l
y
b
ased
o
n
its
cla
s
s
if
icatio
n
o
n
eit
h
er
n
o
r
m
al
o
r
au
tis
m
.
Fin
a
ll
y
,
t
h
e
d
ata
is
i
m
p
le
m
e
n
ted
in
t
o
a
p
r
e
-
p
r
o
ce
s
s
in
g
al
g
o
r
ith
m
th
at
is
u
s
ed
f
o
r
au
g
m
e
n
tatio
n
an
d
r
e
m
o
v
al
o
f
n
o
is
e
u
s
i
n
g
r
a
n
d
o
m
s
h
u
f
f
li
n
g
an
d
w
h
ite
Ga
u
s
s
ia
n
n
o
is
e.
2.
4
.
Desig
n o
f
deep
lea
rning
a
lg
o
rit
h
m
T
h
e
d
ee
p
lear
n
in
g
m
o
d
el
is
d
e
s
ig
n
ed
to
f
i
t
t
h
e
E
E
G
d
ata
in
2
D
m
atr
ix
f
o
r
m
.
T
h
e
m
o
d
el
is
s
h
o
w
n
i
n
Fig
u
r
e
5
p
r
o
d
u
ce
d
in
P
y
th
o
n
.
T
h
e
p
r
o
p
o
s
ed
d
ee
p
lear
n
in
g
m
o
d
el
w
h
ic
h
u
s
e
C
NN
ar
ch
itect
u
r
e
h
as
a
to
tal
o
f
6
la
y
er
s
w
h
ic
h
co
n
s
is
t
s
o
f
t
h
r
ee
co
n
v
o
l
u
tio
n
al
la
y
er
s
,
o
n
e
f
lat
ten
la
y
er
an
d
t
w
o
d
en
s
e
(
f
u
ll
y
co
n
n
ec
ted
)
la
y
er
s
.
Af
ter
ea
ch
co
n
v
o
lu
tio
n
al
la
y
er
,
b
atch
n
o
r
m
aliza
tio
n
is
ap
p
lied
.
T
h
is
m
o
d
el
also
u
s
ed
Gau
s
s
ia
n
d
r
o
p
o
u
t
w
ei
g
h
ts
.
T
h
i
s
is
b
ec
au
s
e
th
e
p
r
o
to
co
ls
o
f
tr
an
s
f
er
lear
n
in
g
s
tated
th
a
t
w
h
en
C
N
N
i
s
lear
n
ed
f
r
o
m
s
cr
atc
h
,
it
n
ee
d
s
to
b
e
s
tar
ted
w
it
h
r
a
n
d
o
m
Ga
u
s
s
ian
d
is
tr
ib
u
tio
n
s
.
N
ex
t,
A
d
a
m
o
p
ti
m
izatio
n
is
s
et
to
0
.
0
0
0
1
w
i
th
2
0
0
ep
o
ch
s
.
So
m
e
ac
tiv
at
io
n
f
u
n
c
tio
n
is
also
u
s
ed
i
n
th
i
s
C
NN
m
o
d
el.
T
h
e
R
eL
u
ac
tiv
at
io
n
f
u
n
ctio
n
is
u
s
ed
in
ev
er
y
co
n
v
o
l
u
tio
n
a
l
la
y
er
w
h
i
le
th
e
f
i
n
al
d
e
n
s
e
la
y
er
u
s
ed
S
o
f
t
m
a
x
ac
t
iv
at
io
n
.
T
h
e
f
o
llo
win
g
T
ab
le
3
s
h
o
w
s
h
o
w
th
e
to
tal
n
u
m
b
er
o
f
d
atas
et
is
d
is
tr
ib
u
ted
f
o
r
m
o
d
el
tr
ai
n
in
g
,
v
alid
atio
n
a
n
d
in
f
er
en
ce
.
T
ab
le
3
.
Dis
tr
ib
u
tio
n
o
f
d
atase
t
M
o
d
e
l
D
i
st
r
i
b
u
t
i
o
n
(
%)
T
r
a
i
n
i
n
g
S
e
t
8
0
%
V
a
l
i
d
a
t
i
o
n
S
e
t
1
0
%
T
e
st
S
e
t
1
0
%
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
I
n
t J
A
r
ti
f
I
n
tell
,
Vo
l.
9
,
No
.
1
,
Ma
r
ch
20
20
:
91
–
99
96
Fig
u
r
e
5
.
Desig
n
o
f
Dee
p
L
ea
r
n
in
g
Mo
d
el
P
er
f
o
r
m
a
nce
a
s
s
e
s
s
m
e
nt
m
a
t
rix
Gen
er
all
y
,
t
h
e
p
r
ed
ictiv
e
m
o
d
el
d
er
iv
ed
is
r
etr
iev
ed
w
it
h
a
n
u
m
b
er
o
f
ev
al
u
atio
n
m
ea
s
u
r
es.
Fo
r
a
class
i
f
icatio
n
o
f
p
r
ed
icted
class
es
b
et
w
ee
n
A
S
D
an
d
No
n
-
ASD,
T
ab
le
4
s
h
o
w
s
t
h
e
p
o
s
s
ib
le
an
s
w
er
f
o
r
a
te
s
t
ca
s
e
p
r
ed
ictio
n
.
T
ab
le
4
.
C
o
n
f
u
s
io
n
m
atr
i
x
f
o
r
d
iag
n
o
s
i
s
o
n
A
S
D
P
r
e
d
i
c
t
e
d
C
l
a
ss
A
c
t
u
a
l
C
l
a
ss
A
S
D
N
o
n
-
A
S
D
A
S
D
T
r
u
e
P
o
si
t
i
v
e
(
T
P
)
F
a
l
s
e
N
e
g
a
t
i
v
e
(
F
N
)
N
o
n
-
A
S
D
F
a
l
s
e
P
o
si
t
i
v
e
(
F
P
)
T
r
u
e
N
e
g
a
t
i
v
e
(
T
N
)
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
clas
s
i
f
icatio
n
r
esu
lt
is
a
n
al
y
ze
d
b
y
t
h
e
co
n
f
u
s
io
n
m
atr
i
x
o
f
t
h
e
te
s
t
f
ile.
T
h
e
co
n
f
u
s
io
n
m
a
tr
ix
i
s
a
p
ar
ticu
l
ar
tab
le
la
y
o
u
t
t
h
at
allo
w
s
v
i
s
u
aliza
tio
n
o
f
th
e
p
er
f
o
r
m
an
c
e
o
f
an
al
g
o
r
ith
m
.
C
las
s
i
f
icatio
n
ac
c
u
r
ac
y
(
1
)
is
o
n
e
o
f
t
h
e
co
m
m
o
n
e
v
alu
a
ti
o
n
m
ea
s
u
r
e
s
.
T
h
e
n
u
m
b
er
o
f
test
ca
s
e
s
t
h
at
h
av
e
b
ee
n
co
r
r
ec
tly
cla
s
s
i
f
ied
ca
n
b
e
id
en
tifie
d
b
y
u
s
i
n
g
th
is
m
ea
s
u
r
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
A
u
tis
m
s
p
ec
tr
u
m
d
is
o
r
d
er c
la
s
s
ifica
tio
n
o
n
elec
tr
o
en
ce
p
h
a
lo
g
r
a
m
s
ig
n
a
l u
s
in
g
…
(
N
.
A
A
li
)
97
(
%
)
=
|
+
|
|
+
+
+
|
(
1
)
3.
RE
SU
L
T
& AN
A
L
YS
I
S
3
.
1
.
T
ra
ini
ng
re
s
ult
T
h
e
tr
ain
i
n
g
p
r
o
ce
s
s
i
s
r
ep
ea
t
ed
f
o
r
f
iv
e
ti
m
es.
E
ac
h
ti
m
e
t
h
e
tr
ai
n
i
n
g
p
r
o
ce
s
s
e
n
d
ed
,
th
e
r
esu
l
t
o
f
th
e
tr
ai
n
in
g
i
s
o
v
er
w
r
iti
n
g
an
d
au
to
m
atica
l
l
y
s
to
r
ed
in
a
web
f
ile.
T
h
is
w
eb
f
ile
w
ill
b
e
u
p
d
ated
ac
co
r
d
in
g
l
y
w
h
e
n
th
e
co
d
e
tr
ain
i
n
g
is
e
x
e
cu
ted
.
T
h
u
s
,
w
h
en
f
u
tu
r
e
tr
ai
n
in
g
i
s
d
o
n
e,
th
e
r
es
u
lt
f
r
o
m
p
r
ev
io
u
s
tr
ai
n
i
n
g
i
s
u
s
ed
as
a
b
asis
f
o
r
th
at
m
o
d
el
tr
ain
i
n
g
.
T
h
e
f
ir
s
t
m
o
d
el
t
r
ain
in
g
ac
h
iev
ed
ac
cu
r
ac
y
o
f
7
0
%,
th
e
s
ec
o
n
d
tr
ain
i
n
g
ac
h
ie
v
ed
ac
cu
r
ac
y
o
f
7
9
% a
n
d
th
e
n
e
x
t t
h
r
ee
m
o
d
el
tr
ain
i
n
g
ac
h
ie
v
ed
a
co
n
s
is
te
n
t
ac
cu
r
ac
y
o
f
±
8
0
%.
W
h
en
a
tr
ai
n
i
n
g
ac
c
u
r
ac
y
d
id
n
o
t
in
cr
ea
s
e
a
f
ter
s
ev
er
al
tr
ai
n
in
g
p
r
o
ce
s
s
,
t
h
is
i
n
d
ica
te
t
h
e
s
to
p
p
in
g
cr
iter
io
n
.
T
h
e
s
to
p
p
in
g
cr
iter
io
n
s
h
o
w
s
th
at
eit
h
er
m
o
r
e
d
ataset
n
ee
d
to
b
e
ad
d
e
d
o
r
th
e
C
NN
m
o
d
el
n
ee
d
s
to
b
e
i
m
p
r
o
v
ed
in
o
r
d
er
to
ac
h
ie
v
e
h
i
g
h
er
ac
c
u
r
ac
y
.
Fi
g
u
r
e
6
(
a)
an
d
6
(
b
)
s
h
o
w
s
tr
ain
in
g
p
r
o
g
r
ess
o
f
t
h
e
C
N
N
u
s
i
n
g
t
h
e
o
b
tain
ed
s
a
m
p
les,
an
d
t
h
e
r
esp
ec
ti
v
e
f
lu
ct
u
ati
o
n
s
i
n
ac
c
u
r
ac
y
a
n
d
lo
s
s
m
etr
ics
ac
r
o
s
s
ea
c
h
tr
ain
i
n
g
ep
o
ch
s
.
As s
h
o
w
n
i
n
t
h
e
p
r
ev
io
u
s
f
i
g
u
r
es,
th
e
o
b
tai
n
ed
tr
ain
i
n
g
ac
c
u
r
ac
y
is
o
n
l
y
8
0
%.
No
te
t
h
at
t
h
e
ac
cu
r
ac
y
o
b
tain
ed
is
b
ased
o
n
o
n
ly
t
w
e
n
t
y
d
at
ase
ts
an
d
s
i
x
la
y
er
s
o
f
C
NN
m
o
d
el.
A
lt
h
o
u
g
h
co
n
v
o
l
u
tio
n
to
2
D
m
atr
i
x
f
o
r
m
i
s
p
o
s
s
ib
le,
th
e
r
es
u
lt
o
b
tain
ed
w
o
u
ld
n
o
t
b
e
as
w
el
l
a
s
th
e
r
es
u
lt
a
s
d
ata
tr
ain
ed
u
s
i
n
g
i
n
a
2
D
i
m
ag
e.
T
h
is
is
d
u
e
to
t
h
e
2
D
co
n
v
o
l
u
tio
n
p
r
o
ce
s
s
is
n
o
t
ab
le
to
s
eize
co
m
p
letel
y
th
e
s
p
atial
r
e
latio
n
s
h
ip
b
et
w
ee
n
elec
tr
o
d
es.
T
h
is
m
ea
n
s
th
at
E
E
G
d
ata,
o
r
b
asically
s
ig
n
al
i
n
g
en
er
al,
h
av
e
s
m
al
l
a
m
o
u
n
t
o
f
s
p
atial
r
e
g
u
lar
it
ies
f
o
r
th
e
2
D
C
NN
to
d
er
iv
e
f
r
o
m
.
(
a)
(
b
)
Fig
u
r
e
6
.
T
r
ain
in
g
p
r
o
g
r
ess
,
(
a)
A
cc
u
r
ac
y
,
(
b
)
L
o
s
s
3
.
2
.
T
est
re
s
ult
An
al
g
o
r
ith
m
ca
lled
th
e
i
n
f
er
e
n
ce
al
g
o
r
ith
m
i
s
d
ev
e
lo
p
ed
to
test
w
h
et
h
er
th
e
d
ev
elo
p
ed
C
NN
m
o
d
el
ca
n
co
r
r
ec
tly
cla
s
s
i
f
y
a
g
i
v
en
E
E
G
s
ig
n
al
f
r
o
m
a
r
an
d
o
m
d
ataset.
T
h
e
in
f
er
en
ce
al
g
o
r
ith
m
lo
ad
ed
th
e
C
N
N
m
o
d
el
th
at
is
s
to
r
ed
in
.
h
5
f
ile
.
T
h
en
an
E
E
G
d
ata
s
et
w
h
ich
is
eit
h
er
n
o
r
m
a
l
o
r
au
ti
s
tic
p
at
ien
t
E
E
G
i
s
lo
ad
ed
.
T
h
e
p
r
e
d
icted
o
u
tp
u
t
is
p
r
in
t
ed
.
T
h
e
d
esig
n
o
f
d
ee
p
lear
n
in
g
m
o
d
el
in
t
h
i
s
p
r
o
j
ec
t
d
id
n
o
t
ac
h
ie
v
e
h
i
g
h
class
i
f
icatio
n
o
f
ac
cu
r
ac
y
,
t
h
u
s
p
r
ed
icted
o
u
tp
u
t
p
r
o
d
u
ce
d
is
r
ath
er
in
co
n
s
is
te
n
t.
T
h
e
p
r
o
d
u
ce
d
r
esu
lt
s
o
m
eti
m
es
g
iv
e
s
a
p
o
s
iti
v
e
r
e
s
u
lt
an
d
v
ice
v
er
s
a.
T
h
e
co
n
s
i
s
ten
c
y
o
f
t
h
e
p
r
in
ted
r
es
u
lt
is
clea
r
l
y
d
u
e
to
th
e
p
er
ce
n
tag
e
o
f
ac
c
u
r
ac
y
o
b
tain
ed
d
u
r
in
g
tr
ai
n
in
g
s
e
t.
4.
CO
NCLU
SI
O
NS
I
n
co
n
cl
u
s
io
n
,
th
e
o
b
j
ec
tiv
es
o
f
th
i
s
r
esear
ch
ar
e
ac
h
ie
v
ed
w
i
th
s
o
m
e
i
m
p
r
o
v
e
m
en
t
th
a
t
co
u
ld
b
e
m
ad
e
f
o
r
f
u
t
u
r
e
s
t
u
d
ies.
Sin
c
e
th
er
e
is
l
i
m
ited
d
ataset
a
v
a
ilab
le
p
u
b
licl
y
o
f
E
E
G
b
r
ain
s
ig
n
al
o
n
au
t
is
m
p
atien
t,
th
e
d
ataset
is
f
o
r
m
all
y
r
eq
u
ested
f
r
o
m
s
ev
er
al
u
n
i
v
er
s
ities
i
n
s
id
e
an
d
o
u
ts
id
e
th
e
co
u
n
tr
y
.
T
h
e
d
ataset
test
ed
i
n
t
h
is
r
esear
ch
o
b
tain
e
d
f
r
o
m
th
e
Ki
n
g
A
b
d
u
laziz
U
n
iv
er
s
it
y
,
J
ed
d
ah
,
Sau
d
i
A
r
ab
ia
w
ith
to
tal
o
f
o
n
l
y
2
0
p
e
r
s
o
n
s
.
T
h
e
d
ee
p
lear
n
in
g
m
o
d
el
u
s
in
g
C
N
N
is
d
ev
elo
p
ed
w
ith
a
to
tal
o
f
s
i
x
la
y
er
s
.
As
m
e
n
tio
n
ed
,
ea
ch
la
y
er
i
n
C
NN
p
la
y
s
m
aj
o
r
r
o
le
in
en
s
u
r
i
n
g
t
h
e
tr
ai
n
i
n
g
an
d
lear
n
in
g
p
r
o
ce
s
s
s
u
cc
es
s
f
u
l
.
T
h
e
to
tal
o
f
d
ata
tr
ain
i
n
g
is
f
i
v
e
t
i
m
e
s
w
i
th
a
co
n
s
is
ten
ce
ac
c
u
r
ac
y
r
ate.
T
h
e
s
to
p
p
in
g
cr
iter
io
n
s
h
o
w
s
t
h
at
f
o
r
t
h
is
s
p
ec
i
f
ic
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
I
n
t J
A
r
ti
f
I
n
tell
,
Vo
l.
9
,
No
.
1
,
Ma
r
ch
20
20
:
91
–
99
98
C
NN
m
o
d
el,
t
h
e
m
ax
i
m
u
m
a
ch
iev
ab
le
ac
c
u
r
ac
y
i
s
ar
o
u
n
d
eig
h
t
y
p
er
ce
n
t
o
n
l
y
.
T
h
e
ev
al
u
atio
n
o
f
t
h
e
d
ee
p
lear
n
in
g
m
o
d
el
s
h
o
w
s
i
n
co
n
s
is
ten
c
y
o
f
th
e
p
r
in
ted
r
esu
lt
w
h
ic
h
is
th
e
d
ir
ec
t
ca
u
s
e
o
f
th
e
lo
w
ac
c
u
r
ac
y
.
A
lt
h
o
u
g
h
th
e
ac
h
ie
v
ed
ac
cu
r
ac
y
is
o
n
l
y
ab
o
u
t
8
0
%,
th
is
s
h
o
w
s
t
h
at
it
h
as
h
i
g
h
er
p
o
ten
tial
i
n
d
ev
elo
p
in
g
b
etter
alg
o
r
ith
m
b
y
u
s
i
n
g
m
o
r
e
co
m
p
le
x
d
ee
p
lear
n
in
g
m
o
d
e
l a
n
d
h
av
in
g
m
o
r
e
d
ataset.
T
h
e
m
aj
o
r
co
n
tr
ib
u
tio
n
o
f
t
h
i
s
r
esear
ch
to
w
ar
d
s
t
h
e
s
o
ciet
y
i
s
in
p
r
o
d
u
cin
g
an
al
ter
n
ati
v
e
m
et
h
o
d
to
m
ak
e
d
etec
tio
n
o
n
t
h
e
p
r
esen
c
e
o
f
au
tis
m
i
n
a
ch
i
ld
.
T
h
e
cu
r
r
en
t
d
iag
n
o
s
e
m
et
h
o
d
o
f
A
SD
is
v
er
y
m
u
ch
ti
m
e
co
n
s
u
m
i
n
g
,
t
h
u
s
i
t
i
s
n
o
t
r
elia
b
le
s
in
ce
s
t
u
d
ies
h
a
v
e
s
h
o
w
n
th
at
ASD
c
h
ild
s
u
f
f
er
s
f
r
o
m
m
an
y
s
id
e
e
f
f
ec
t
o
f
ASD
s
i
n
ce
y
o
u
n
g
ag
e
s
u
ch
a
s
v
i
s
u
a
l
i
m
p
a
ir
m
e
n
t.
F
u
r
t
h
er
m
o
r
e,
a
f
u
ll
y
d
ev
elo
p
an
d
h
i
g
h
ac
c
u
r
ac
y
s
y
s
te
m
co
u
ld
b
e
e
m
p
lo
y
ed
b
y
th
e
Mi
n
is
tr
y
o
f
Hea
lth
as
a
n
e
w
d
ia
g
n
o
s
i
n
g
m
eth
o
d
o
f
ASD
to
h
i
g
h
r
is
k
ch
i
ld
r
en
b
ased
o
n
E
E
G.
I
n
ter
m
s
o
f
ec
o
n
o
m
i
c,
it
ca
n
g
r
ea
tl
y
r
ed
u
ce
th
e
ti
m
e
ta
k
e
n
to
d
etec
t
A
S
D
t
h
u
s
ea
r
l
y
tr
ea
t
m
e
n
t
co
u
ld
b
e
p
lan
n
ed
b
y
p
ed
iatr
ics to
p
r
o
v
id
e
b
etter
h
ea
lth
s
u
p
p
o
r
t to
th
e
au
ti
s
tic
p
atie
n
t.
ACK
NO
WL
E
D
G
E
M
E
NT
S
Au
t
h
o
r
s
ar
e
g
r
ate
f
u
l
to
U
n
i
v
er
s
iti
T
ek
n
i
k
al
Ma
la
y
s
ia
Me
l
ak
a
f
o
r
th
e
f
i
n
a
n
cial
s
u
p
p
o
r
t
th
r
o
u
g
h
f
u
n
d
a
m
en
ta
l
r
esear
ch
g
r
an
t
F
R
GS/2
0
1
8
/FKEKK
-
C
E
R
I
A
/F
0
0
3
6
3
.
A
u
th
o
r
s
a
ls
o
w
a
n
t
to
th
an
k
r
esear
ch
er
at
Kin
g
A
b
d
u
laziz
U
n
i
v
er
s
it
y
,
J
e
d
d
ah
,
Sau
d
i
A
r
ab
ia
in
p
r
o
v
id
in
g
t
h
e
d
ataset.
RE
F
E
R
E
NC
E
S
[1
]
E.
B.
Jo
h
n
s
o
n
,
Au
t
ism
S
p
e
c
tru
m
Diso
rd
e
rs
:
T
h
e
W
o
rld
wi
d
e
Ch
a
r
m
An
d
C
h
a
ll
e
n
g
e
Of
Au
t
ism
S
p
e
c
tru
m
Diso
rd
e
rs
,
S
p
e
c
ial
Ed
.
Bra
d
f
o
rd
,
Uk
:
Em
e
ra
ld
G
ro
u
p
P
u
b
li
sh
i
n
g
,
2
0
1
4
.
[2
]
M
.
S
h
a
r
d
a
Et
Al
.
,
“
L
a
n
g
u
a
g
e
A
b
il
it
y
P
re
d
icts
Co
rti
c
a
l
S
tr
u
c
tu
re
A
n
d
Co
v
a
rian
c
e
In
Bo
y
s
W
it
h
Au
ti
sm
S
p
e
c
tru
m
Diso
rd
e
r,
”
Ce
re
b
.
Co
rte
x
,
Vo
l.
2
7
,
No
.
3
,
P
p
.
1
8
4
9
–
1
8
6
2
,
2
0
1
7
.
[3
]
J.
Je
n
n
in
g
s
Du
n
la
p
,
“
A
u
ti
sm
S
p
e
c
tru
m
Diso
rd
e
r
S
c
re
e
n
in
g
A
n
d
E
a
rl
y
A
c
ti
o
n
,
”
J
.
Nu
rs
e
Pra
c
t.
,
Vo
l.
1
5
,
No
.
7
,
P
p
.
496
–
5
0
1
,
2
0
1
9
.
[4
]
T
.
L
iu
,
Y.
Ch
e
n
,
D.
Ch
e
n
,
C
.
L
i,
Y.
Qiu
,
A
n
d
J.
W
a
n
g
,
“
A
lt
e
re
d
El
e
c
tro
e
n
c
e
p
h
a
lo
g
ra
m
Co
m
p
le
x
it
y
In
A
u
ti
stic
Ch
il
d
re
n
S
h
o
w
n
By
T
h
e
M
u
lt
isc
a
le E
n
tro
p
y
A
p
p
ro
a
c
h
,
”
Ne
u
ro
re
p
o
rt
,
V
o
l
.
2
8
,
N
o
.
3
,
P
p
.
1
6
9
–
1
7
3
,
2
0
1
7
.
[5
]
P
.
R.
P
.
Ho
o
le
Et
Al
.
,
“
A
u
ti
s
m
,
Eeg
A
n
d
Bra
in
El
e
c
tro
m
a
g
n
e
ti
c
s
Re
se
a
r
c
h
,
”
2
0
1
2
Ie
e
e
-
Emb
s
Co
n
f.
Bi
o
me
d
.
E
n
g
.
S
c
i.
Ie
c
b
e
s 2
0
1
2
,
No
.
De
c
e
m
b
e
r,
P
p
.
5
4
1
–
5
4
3
,
2
0
1
2
.
[6
]
N.
F
a
u
z
a
n
A
n
d
N.
H.
A
m
r
a
n
,
“
Bra
in
W
a
v
e
s
A
n
d
Co
n
n
e
c
ti
v
it
y
Of
A
u
ti
s
m
S
p
e
c
tru
m
Diso
rd
e
rs,”
Pro
c
e
d
ia
-
S
o
c
.
Beh
a
v
.
S
c
i
.
,
Vo
l.
1
7
1
,
P
p
.
8
8
2
–
8
9
0
,
2
0
1
5
.
[7
]
L
.
Bo
te
-
Cu
riel,
S
.
M
u
ñ
o
z
-
Ro
m
e
ro
,
A
.
G
e
rre
ro
-
Cu
ries
e
s,
A
n
d
J.
L
.
Ro
j
o
-
Á
lv
a
re
z
,
“
De
e
p
L
e
a
rn
in
g
A
n
d
Big
Da
ta
In
He
a
lt
h
c
a
re
:
A
Do
u
b
le Rev
iew
F
o
r
Crit
ica
l
Be
g
in
n
e
rs,”
Ap
p
l.
S
c
i.
,
V
o
l
.
9
,
No
.
1
1
,
2
0
1
9
.
[8
]
F
.
C.
M
o
ra
b
it
o
Et
Al.
,
“
De
e
p
Co
n
v
o
lu
ti
o
n
a
l
Ne
u
ra
l
Ne
tw
o
rk
s
F
o
r
Clas
si
f
ica
ti
o
n
Of
M
il
d
Co
g
n
it
iv
e
I
m
p
a
ired
A
n
d
A
lzh
e
i
m
e
r’s
Dise
a
s
e
P
a
ti
e
n
ts
F
r
o
m
S
c
a
lp
Eeg
Re
c
o
rd
in
g
s,”
2
0
1
6
Ie
e
e
2
n
d
In
t
.
F
o
ru
m
Res
.
T
e
c
h
n
o
l.
S
o
c
.
I
n
d
.
L
e
v
e
ra
g
in
g
A
Better
T
o
mo
rr
o
w,
Rt
si 2
0
1
6
,
2
0
1
6
.
[9
]
W
.
H.
L
.
P
i
n
a
y
a
,
A
.
M
e
c
h
e
ll
i,
An
d
J.
R.
S
a
t
o
,
“
Us
in
g
De
e
p
A
u
to
e
n
c
o
d
e
rs
T
o
Id
e
n
ti
fy
A
b
n
o
rm
a
l
Bra
in
S
tr
u
c
tu
ra
l
P
a
tt
e
r
n
s
In
Ne
u
ro
p
sy
c
h
iatric
Dis
o
rd
e
rs:
A
Larg
e
-
S
c
a
le
M
u
lt
i
-
S
a
m
p
le
S
tu
d
y
,
”
Hu
m.
Bra
in
M
a
p
p
.
,
V
o
l.
4
0
,
No
.
3
,
P
p
.
9
4
4
–
9
5
4
,
2
0
1
9
.
[1
0
]
G
.
Ch
o
,
J.
Yim
,
Y.
Ch
o
i,
J.
Ko
,
A
n
d
S
.
H.
L
e
e
,
“
Re
v
ie
w
O
f
M
a
c
h
in
e
L
e
a
rn
in
g
A
lg
o
rit
h
m
s
F
o
r
Dia
g
n
o
sin
g
M
e
n
ta
l
Ill
n
e
ss
,
”
Psy
c
h
ia
try
In
v
e
stig
.
,
Vo
l
.
1
6
,
No
.
4
,
P
p
.
2
6
2
–
2
6
9
,
2
0
1
9
.
[1
1
]
W
.
Bo
sl,
A
.
T
iern
e
y
,
H.
T
a
g
e
r
-
F
lu
sb
e
rg
,
A
n
d
C.
Ne
lso
n
,
“
Eeg
Co
m
p
lex
it
y
A
s
A
Bio
m
a
rk
e
r
F
o
r
Au
ti
sm
S
p
e
c
tru
m
Diso
rd
e
r
Risk
,
”
Bmc
M
e
d
.
,
V
o
l
.
9
,
2
0
1
1
.
[1
2
]
M. S.
My
thil
i
A
nd
A
. R
.
M. S
ha
na
v
a
s
,
“
A
St
ud
y
O
n
A
utis
m
Spe
c
tr
um
D
i
s
orde
r
s
U
s
ing
C
la
s
s
if
i
c
a
ti
on
T
e
c
hn
ique
s
,”
Ijcs
it
,
Vo
l.
5
,
No
.
6
,
P
p
.
7
2
8
8
–
7
2
9
1
,
2
0
1
4
.
[1
3
]
J.
B.
E
w
e
n
Et
Al
.
,
“
De
c
r
e
a
s
e
d
M
o
d
u
latio
n
Of
Ee
g
Os
c
il
latio
n
s
In
Hig
h
-
F
u
n
c
ti
o
n
i
n
g
A
u
ti
s
m
D
u
rin
g
A
M
o
to
r
Co
n
tr
o
l
T
a
sk
,
”
V
o
l.
1
0
,
No
.
M
a
y
,
P
p
.
1
–
1
1
,
2
0
1
6
.
[1
4
]
T
.
He
u
n
is
Et
Al
.
,
“
Re
c
u
rre
n
c
e
Qu
a
n
ti
f
ica
ti
o
n
A
n
a
ly
sis
Of
Re
stin
g
S
tate
Eeg
S
ig
n
a
ls
I
n
A
u
ti
sm
S
p
e
c
tru
m
Diso
rd
e
r
-
A
S
y
ste
m
a
ti
c
M
e
th
o
d
o
l
o
g
ica
l
Ex
p
lo
ra
ti
o
n
Of
T
e
c
h
n
ica
l
A
n
d
De
m
o
g
ra
p
h
ic
Co
n
f
o
u
n
d
e
rs
In
T
h
e
S
e
a
r
c
h
F
o
r
Bio
m
a
rk
e
rs,”
Bmc
M
e
d
.
,
V
o
l.
1
6
,
No
.
1
,
P
p
.
1
–
1
7
,
2
0
1
8
.
[1
5
]
I.
V
a
silev
,
D.
S
late
r,
G
.
S
p
a
c
a
g
n
a
,
P
.
Ro
e
lan
ts,
A
n
d
V
.
Z
o
c
c
a
,
Pyth
o
n
De
e
p
L
e
a
rn
i
n
g
:
Exp
l
o
rin
g
De
e
p
L
e
a
rn
i
n
g
T
e
c
h
n
iq
u
e
s
An
d
Ne
u
r
a
l
Ne
two
rk
Arc
h
it
e
c
tu
re
s
W
it
h
Pyto
rc
h
,
Ke
ra
s
An
d
T
e
n
so
rfl
o
w
,
2
n
d
E
d
it
i
o
.
Bir
m
in
g
h
a
m
,
U
k
:
P
a
c
k
t
P
u
b
li
s
h
in
g
,
2
0
1
8
.
[1
6
]
M
.
J.
A
lh
a
d
d
a
d
Et
Al
.
,
“
Dia
g
n
o
sis
A
u
ti
s
m
B
y
F
ish
e
r
L
in
e
a
r
Disc
rim
in
a
n
t
A
n
a
l
y
sis
F
ld
a
V
ia
Eeg
,
”
V
o
l
.
4
,
No
.
2
,
P
p
.
4
5
–
5
4
,
2
0
1
2
.
[1
7
]
G
a
v
in
Ha
c
k
e
li
n
g
,
M
a
ste
rin
g
M
a
c
h
in
e
L
e
a
rn
i
n
g
W
it
h
S
c
ikit
-
L
e
a
rn
S
e
c
o
n
d
,
2
n
d
Ed
i
ti
o
.
P
a
c
k
t
P
u
b
li
sh
in
g
,
2
0
1
7
.
[1
8
]
P
.
M
a
rz
u
k
i,
A
.
R.
S
y
a
f
e
e
z
a
,
A
.
N.
A
li
sa
,
A
n
d
M
.
K.
M
.
F
.
A
li
f
,
“
A
n
Im
p
ro
v
e
d
O
f
M
a
la
y
sia
n
L
ice
n
se
P
late
s
De
tec
ti
o
n
Us
in
g
De
e
p
Co
n
v
o
lu
ti
o
n
a
l
Ne
u
ra
l
Ne
tw
o
rk
s,”
S
y
mp
.
E
lec
tr.
M
e
c
h
a
tro
n
ics
Ap
p
l.
S
c
i.
2
0
1
8
,
V
o
l
.
2
0
1
8
,
No
.
No
v
e
m
b
e
r,
P
p
.
7
5
–
7
6
,
2
0
1
8
.
[1
9
]
N.
A
.
A
li
,
A
.
R.
S
y
a
f
e
e
z
a
,
L
.
J
.
G
e
o
k
,
Y.
C.
W
o
n
g
,
N.
A
.
Ha
m
id
,
A
n
d
A
.
S
.
Ja
a
f
a
r,
“
D
e
sig
n
Of
A
u
to
m
a
t
e
d
Co
m
p
u
ter
-
A
id
e
d
Clas
sif
i
c
a
ti
o
n
O
f
Bra
in
T
u
m
o
r
Us
in
g
De
e
p
L
e
a
r
n
in
g
,
”
In
In
tell
ig
e
n
t
An
d
I
n
ter
a
c
t
ive
Co
mp
u
ti
n
g
,
2
0
1
9
,
P
p
.
2
8
5
–
2
9
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
A
u
tis
m
s
p
ec
tr
u
m
d
is
o
r
d
er c
la
s
s
ifica
tio
n
o
n
elec
tr
o
en
ce
p
h
a
lo
g
r
a
m
s
ig
n
a
l u
s
in
g
…
(
N
.
A
A
li
)
99
[2
0
]
A
.
R.
S
y
a
f
e
e
z
a
,
S
.
S
.
L
ie
w
,
A
n
d
R.
Ba
k
h
teri,
“
Co
n
v
o
l
u
ti
o
n
a
l
Ne
u
r
a
l
Ne
t
w
o
rk
F
o
r
F
a
c
e
Re
c
o
g
n
it
io
n
W
it
h
P
o
se
A
n
d
Ill
u
m
in
a
ti
o
n
V
a
riati
o
n
,
”
I
n
t.
J
.
En
g
.
T
e
c
h
n
o
l
.
Co
n
v
o
l
u
ti
o
n
a
l
,
V
o
l.
6
,
No
.
1
,
P
p
.
4
4
–
5
7
,
2
0
1
4
.
[2
1
]
M
.
K.
M
.
F
.
A
li
f
,
P
.
M
.
,
A
.
R.
S
y
a
fe
e
z
a
,
A
n
d
N.
A
.
A
li
,
“
F
u
se
d
Co
n
v
o
lu
ti
o
n
a
l
Ne
u
ra
l
Ne
t
w
o
rk
F
o
r
F
a
c
ial
Ex
p
re
ss
io
n
Re
c
o
g
n
it
i
o
n
,
”
Pro
c
.
S
y
mp
.
El
e
c
tr.
M
e
c
h
a
tro
n
ics
Ap
p
l.
S
c
i.
2
0
1
8
,
P
p
.
7
3
–
7
4
,
2
0
1
8
.
[2
2
]
R.
Dje
m
a
l,
K.
A
lsh
a
ra
b
i,
S
.
Ib
ra
h
im
,
A
n
d
A
.
A
lsu
wa
il
e
m
,
“
Ee
g
-
Ba
se
d
Co
m
p
u
ter
A
id
e
d
Dia
g
n
o
sis
Of
A
u
ti
s
m
S
p
e
c
tru
m
Diso
rd
e
r
Us
in
g
W
a
v
e
le
t,
En
t
ro
p
y
,
A
n
d
A
n
n
,
”
Bi
o
me
d
Re
s.
In
t.
,
V
o
l.
2
0
1
7
,
2
0
1
7
.
[2
3
]
B.
T
a
n
g
Et
Al
.
,
“
De
e
p
c
h
a
rt:
C
o
m
b
in
in
g
De
e
p
Co
n
v
o
lu
ti
o
n
a
l
Ne
tw
o
rk
s
A
n
d
De
e
p
Be
li
e
f
Ne
t
w
o
r
k
s
In
Ch
a
rt
Clas
sif
ic
a
ti
o
n
,
”
S
i
g
n
a
l
Pr
o
c
e
ss
in
g
,
V
o
l
.
1
2
4
,
P
p
.
1
5
6
–
1
6
1
,
2
0
1
6
.
[2
4
]
S
.
S
.
B,
D.
M
a
h
a
p
a
tra,
Z.
G
e
,
A
n
d
R.
Ch
a
k
ra
v
o
rt
y
,
L
e
a
rn
in
g
Fo
r
W
e
a
k
ly
S
u
p
e
rv
ise
d
L
o
c
a
li
za
ti
o
n
Of
Ch
e
s
t
Pa
th
o
lo
g
ies
In
X
-
R
a
y
Ima
g
e
s
,
Vo
l.
1
.
S
p
r
in
g
e
r
In
ter
n
a
ti
o
n
a
l
P
u
b
li
s
h
in
g
,
2
0
1
8
.
[2
5
]
M
.
A
.
Bu
jan
g
A
n
d
T
.
H.
A
d
n
a
n
,
“
Re
q
u
irem
e
n
ts
F
o
r
M
in
im
u
m
S
a
m
p
le
S
ize
F
o
r
S
e
n
siti
v
it
y
An
d
S
p
e
c
if
icit
y
A
n
a
l
y
si
s,”
J
.
Cli
n
.
Di
a
g
n
o
stic R
e
s.
,
Vo
l.
1
0
,
No
.
1
0
,
P
p
.
Ye
0
1
–
Ye
0
6
,
2
0
1
6
.
[2
6
]
H.
Ra
jag
u
ru
,
“
A
n
a
l
y
si
s
O
f
P
a
c
L
e
a
rn
in
g
Ba
se
d
Ba
y
e
sia
n
O
p
ti
m
iza
ti
o
n
W
it
h
A
u
to
e
n
c
o
d
e
rs
F
o
r
Ep
il
e
p
sy
Clas
sif
ic
a
ti
o
n
F
r
o
m
Eeg
S
ig
n
a
ls,”
v
o
l.
8
,
n
o
.
1
2
,
p
p
.
2
0
6
–
2
1
2
,
2
0
1
7
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
Nu
r
A
li
sa
A
li
c
u
rre
n
tl
y
is
a
P
h
D
stu
d
e
n
t
in
U
n
iv
e
rsiti
T
e
k
n
ik
a
l
M
a
lay
si
a
M
e
lak
a
(U
T
e
M
),
M
a
la
y
sia
.
S
h
e
re
c
e
iv
e
d
h
e
r
Hig
h
e
r
Na
ti
o
n
a
l
Di
p
lo
m
a
(HN
D)
in
El
e
c
tro
n
ics
E
n
g
in
e
e
rin
g
f
ro
m
Brit
ish
M
a
lay
sia
n
In
stit
u
te,
U
n
i
KL
-
BM
I
(2
0
0
3
)
a
n
d
BEn
g
i
n
E
lec
tro
n
ics
f
ro
m
Un
iv
e
rsit
y
o
f
S
u
rre
y
(UN
iS
),
Un
it
e
d
Ki
n
g
d
o
m
in
2
0
0
6
.
S
h
e
re
c
e
iv
e
d
h
e
r
M
a
ste
r
De
g
re
e
in
Co
m
p
u
ter
S
y
ste
m
s
f
ro
m
Un
iv
e
rsit
y
o
f
S
o
u
t
h
A
u
stra
li
a
(Un
iS
A
),
A
u
stra
li
a
in
2
0
0
9
.
Cu
rre
n
tl
y
,
h
e
r
P
h
D
re
se
a
rc
h
in
v
o
lv
e
s e
m
b
e
d
d
e
d
sy
ste
m
s,
m
a
c
h
in
e
a
n
d
d
e
e
p
lea
rn
in
g
;
im
a
g
e
a
n
d
sig
n
a
l
p
r
o
c
e
ss
in
g
.
Dr.
S
y
a
f
e
e
z
a
A
h
m
a
d
Ra
d
z
i
re
c
e
i
v
e
d
h
e
r
B.
En
g
d
e
g
re
e
in
El
e
c
tri
c
a
l
-
El
e
c
tro
n
ic
En
g
in
e
e
rin
g
i
n
2
0
0
3
a
n
d
h
e
r
M
.
En
g
d
e
g
re
e
in
El
e
c
tri
c
a
l
–
El
e
c
tro
n
ic
&
T
e
lec
o
m
m
u
n
ica
ti
o
n
E
n
g
in
e
e
rin
g
i
n
2
0
0
5
f
ro
m
Un
iv
e
rsiti
Tek
n
o
lo
g
i
M
a
lay
sia
.
S
h
e
a
lso
re
c
e
iv
e
d
h
e
r
P
h
.
D
d
e
g
re
e
in
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
f
ro
m
th
e
sa
m
e
u
n
iv
e
rsity
in
2
0
1
4
.
S
h
e
is
c
u
rre
n
tl
y
a
S
e
n
io
r
L
e
c
tu
re
r
a
t
t
h
e
F
a
c
u
lt
y
o
f
El
e
c
tro
n
ic
En
g
in
e
e
rin
g
a
n
d
Co
m
p
u
ter
En
g
in
e
e
rin
g
,
Un
iv
e
rsiti
T
e
k
n
ik
a
l
M
a
la
y
sia
M
e
la
k
a
(UT
e
M
).
S
h
e
h
a
s
b
e
e
n
a
n
a
c
a
d
e
m
ici
a
n
in
UT
e
M
sin
c
e
2
0
0
6
.
S
h
e
d
e
d
ica
tes
h
e
rse
lf
to
u
n
iv
e
rsity
tea
c
h
in
g
a
n
d
c
o
n
d
u
c
ti
n
g
re
se
a
r
c
h
.
He
r
re
se
a
rc
h
f
ield
s
in
c
lu
d
e
e
m
b
e
d
d
e
d
s
y
ste
m
,
p
a
tt
e
rn
re
c
o
g
n
it
io
n
,
m
a
c
h
in
e
lea
rn
in
g
,
d
e
e
p
lea
rn
in
g
,
im
a
g
e
p
ro
c
e
ss
in
g
,
b
i
o
m
e
tri
c
,
e
tc
.
Dr.
A
b
d
S
h
u
k
u
r
Ja
’a
f
a
r
re
c
e
i
v
e
d
b
o
t
h
f
irst
a
n
d
m
a
ste
r
d
e
g
re
e
f
ro
m
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
a
la
y
sia
(U
T
M
)
in
Ba
c
h
e
lo
r
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
(
2
0
0
2
)
a
n
d
M
a
ste
r
o
f
En
g
in
e
e
rin
g
i
n
El
e
c
tro
n
ic
a
n
d
T
e
lec
o
m
m
u
n
ica
ti
o
n
(2
0
0
5
).
He
jo
in
e
d
Un
iv
e
rsiti
T
e
k
n
ik
a
l
M
a
la
y
sia
M
e
l
a
k
a
(UT
e
M
)
a
s
a
lec
tu
re
r
in
2
0
0
5
a
n
d
re
c
e
iv
e
d
P
h
D
in
C
o
m
m
u
n
ica
ti
o
n
S
y
ste
m
f
ro
m
L
a
n
c
a
ste
r
Un
iv
e
rsit
y
,
UK
.
Cu
rre
n
tl
y
h
is
re
se
a
rc
h
in
tere
st
o
n
RF
,
M
icro
w
a
v
e
c
ircu
it
s
a
n
d
a
lg
o
rit
h
m
d
e
v
e
lo
p
m
e
n
t
f
o
r
in
d
o
o
r
p
o
siti
o
n
i
n
g
a
n
d
n
a
v
ig
a
ti
o
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.