I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
pu
t
er
E
ng
ineering
(
I
J
E
CE
)
Vo
l.
15
,
No
.
5
,
Octo
b
er
20
25
,
p
p
.
4
8
5
6
~
4
8
6
4
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
15
i
5
.
pp
4
8
5
6
-
4
8
6
4
4856
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Facia
l ima
g
e ana
ly
sis
f
o
r autism
sp
ectr
um
diso
rder
detec
tion in
toddlers
using
d
e
ep learning
and
t
ra
nsfer
lea
rning
Anup
a
m
Da
s,
P
ra
s
a
nt
K
um
a
r
P
a
t
t
na
i
k
,
Anj
a
n B
a
nd
y
o
pa
dh
y
a
y
S
c
h
o
o
l
o
f
C
o
m
p
u
t
e
r
E
n
g
i
n
e
e
r
i
n
g
,
K
a
l
i
n
g
a
I
n
st
i
t
u
t
e
o
f
I
n
d
u
st
r
i
a
l
T
e
c
h
n
o
l
o
g
y
(
K
I
I
T)
D
e
e
me
d
t
o
b
e
U
n
i
v
e
r
si
t
y
,
B
h
u
b
a
n
e
sw
a
r
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
2
6
,
2
0
2
4
R
ev
is
ed
J
u
n
1
,
2
0
2
5
Acc
ep
ted
J
u
l 3
,
2
0
2
5
Au
ti
sm
sp
e
c
tru
m
d
iso
rd
e
r
(AS
D)
is
a
n
e
u
ro
lo
g
ica
l
il
ln
e
ss
th
a
t
m
a
n
ifes
ts
it
se
lf
th
ro
u
g
h
re
stricte
d
a
n
d
re
p
e
a
ted
a
c
ti
v
it
y
p
a
tt
e
rn
s,
fri
v
o
l
o
u
s
o
r
re
c
id
iv
ist
in
tere
sts
o
r
h
o
b
b
ies
a
n
d
c
o
n
siste
n
t
h
a
n
d
ica
p
s
t
o
s
o
c
ial
i
n
tera
c
ti
o
n
s
a
n
d
e
x
c
h
a
n
g
e
s.
Be
tt
e
r
re
su
lt
s
a
n
d
e
a
rly
i
n
terv
e
n
t
io
n
a
re
d
e
p
e
n
d
e
n
t
u
p
o
n
t
h
e
e
a
rly
id
e
n
ti
fica
ti
o
n
o
f
p
e
o
p
le
wit
h
ASD
.
Do
c
to
rs em
p
l
o
y
a
v
a
riety
o
f
tec
h
n
iq
u
e
s to
a
n
ti
c
ip
a
te
a
u
ti
sm
,
in
c
l
u
d
i
n
g
g
e
n
e
ti
c
tes
ti
n
g
,
n
e
u
r
o
p
s
y
c
h
o
l
o
g
ica
l
tes
ti
n
g
,
h
e
a
rin
g
a
n
d
v
isio
n
sc
re
e
n
in
g
s,
a
n
d
d
ia
g
n
o
stic
in
ter
v
iew
s
.
In
a
d
d
it
i
o
n
to
re
q
u
iri
n
g
m
o
re
ti
m
e
a
n
d
m
o
n
e
y
,
th
e
trad
it
i
o
n
a
l
d
iag
n
o
sis
a
p
p
ro
a
c
h
m
a
k
e
s
th
e
p
a
re
n
ts
o
f
c
h
i
ld
re
n
wit
h
e
x
te
n
siv
e
d
e
v
e
l
o
p
m
e
n
tal
a
b
n
o
rm
a
li
ti
e
s
fe
e
l
to
o
in
a
d
e
q
u
a
te
to
d
isc
lo
se
t
h
e
ir
c
o
n
d
it
io
n
.
S
o
,
we
n
e
e
d
a
to
o
l
t
h
a
t
c
a
n
d
e
tec
t
a
u
ti
sm
e
a
rly
i
n
les
s
ti
m
e
a
n
d
m
o
n
e
y
.
M
a
c
h
in
e
lea
rn
in
g
m
e
th
o
d
s
c
a
n
b
e
u
se
d
to
f
u
lfi
ll
th
is
c
rit
e
rio
n
.
I
n
t
h
is
s
tu
d
y
,
d
e
e
p
lea
rn
i
n
g
with
tran
sfe
r
lea
rn
in
g
(VG
G
-
1
6
)
is
u
se
d
to
d
e
tec
t
a
u
ti
sm
th
ro
u
g
h
fa
c
ial
ima
g
e
s
o
f
c
h
i
ld
re
n
a
n
d
a
c
h
iev
e
d
a
lmo
st
9
7
%
a
c
c
u
ra
c
y
.
T
h
e
su
g
g
e
ste
d
m
o
d
e
l
sig
n
ifi
c
a
n
tl
y
imp
ro
v
e
s
a
c
c
u
ra
c
y
a
n
d
sa
v
e
s
ti
m
e
a
n
d
m
o
n
e
y
b
y
u
sin
g
fa
c
e
fe
a
tu
re
s
i
n
p
h
o
to
s
o
f
c
h
il
d
re
n
to
id
e
n
t
ify
e
a
rly
a
u
ti
sm
t
e
n
d
e
n
c
ies
in
c
h
il
d
re
n
.
K
ey
w
o
r
d
s
:
Au
tis
m
s
p
ec
tr
u
m
d
is
o
r
d
er
Dee
p
lear
n
in
g
Facial
im
ag
e
an
aly
s
is
Ma
ch
in
e
lear
n
in
g
T
r
an
s
f
er
lear
n
i
n
g
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
:
An
u
p
am
Das
Sch
o
o
l o
f
C
o
m
p
u
ter
E
n
g
in
ee
r
i
n
g
,
Kalin
g
a
I
n
s
titu
te
o
f
I
n
d
u
s
t
r
ial
T
ec
h
n
o
lo
g
y
(
K
I
I
T
)
Dee
m
ed
to
b
e
Un
iv
er
s
ity
B
h
u
b
an
eswar
,
Pin
-
7
5
1
0
2
4
,
O
d
is
h
a,
I
n
d
ia
E
m
ail:
an
u
k
iit2
3
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Au
tis
m
s
p
ec
tr
u
m
d
is
o
r
d
e
r
(
ASD)
[
1
]
ac
t
as
a
n
eu
r
o
lo
g
i
ca
l
d
ev
elo
p
m
en
tal
d
is
o
r
d
er
t
h
at
af
f
ec
ts
s
o
cializa
tio
n
as
well
a
s
co
m
m
u
n
icatio
n
.
E
ar
ly
d
iag
n
o
s
is
is
cr
u
cial
as
ASD
ca
n
im
p
ac
t
s
o
c
ial,
ac
ad
em
ic,
an
d
p
r
o
f
ess
io
n
al
asp
ec
ts
o
f
life
.
M
an
y
ch
il
d
r
en
s
h
o
w
s
ig
n
s
with
i
n
th
e
f
ir
s
t
y
ea
r
,
s
u
ch
as
r
ed
u
ce
d
ey
e
co
n
tact,
lack
o
f
in
ter
est
in
ca
r
eg
iv
e
r
s
,
o
r
d
elay
ed
r
esp
o
n
s
e
to
n
am
es
[
2
]
,
[
3
]
.
So
m
e
m
a
y
r
eg
r
ess
b
etwe
en
1
8
–
2
4
m
o
n
th
s
,
lo
s
in
g
ac
q
u
ir
ed
s
k
ills
.
Sy
m
p
t
o
m
s
v
ar
y
in
s
ev
er
ity
a
n
d
im
p
ac
t
o
n
f
u
n
ctio
n
in
g
,
m
ak
in
g
as
s
ess
m
en
t
co
m
p
lex
.
Ar
tific
ial
in
tellig
en
ce
(
AI
)
a
n
d
m
ac
h
in
e
lear
n
in
g
(
ML
)
[
4
]
,
[
5
]
a
r
e
r
e
v
o
lu
tio
n
izin
g
ASD
d
iag
n
o
s
is
an
d
tr
ea
tm
en
t,
o
f
f
er
in
g
f
aster
,
m
o
r
e
ac
cu
r
ate,
a
n
d
s
ca
lab
le
ap
p
r
o
ac
h
es
b
y
a
n
aly
zin
g
la
r
g
e
d
atasets
an
d
id
en
tify
i
n
g
s
u
b
tle
p
atter
n
s
[
6
]
,
[
7
]
.
Glo
b
ally
,
ASD
af
f
ec
ts
ab
o
u
t
1
in
1
0
0
c
h
ild
r
en
,
in
f
lu
en
c
ed
b
y
g
en
etic
a
n
d
en
v
ir
o
n
m
en
tal
f
ac
to
r
s
[
8
]
.
D
iag
n
o
s
is
r
elies
o
n
o
b
s
er
v
i
n
g
b
eh
av
io
r
an
d
d
ev
el
o
p
m
en
ta
l
m
iles
to
n
es,
with
s
p
ec
ialis
t
s
ab
le
to
p
r
o
v
id
e
r
e
liab
le
ass
e
s
s
m
en
ts
b
y
ag
e
tw
o
[
9
]
.
E
ar
l
y
in
ter
v
en
tio
n
s
ig
n
if
ican
tly
im
p
r
o
v
es
d
ev
elo
p
m
e
n
tal
o
u
tco
m
es
[
1
0
]
–
[
1
2
]
em
p
h
asizin
g
th
e
n
ee
d
f
o
r
p
r
o
m
p
t tr
ea
tm
en
t to
m
ax
im
iz
e
p
o
ten
tial.
ASD
wh
ich
was
f
ir
s
t
id
en
tifi
ed
in
2
0
1
3
,
is
a
d
ev
elo
p
m
en
tal
illn
ess
ch
ar
ac
ter
ized
b
y
lim
ited
an
d
r
ep
etitiv
e
b
eh
a
v
io
r
al
p
atter
n
s
,
in
ter
ests
,
o
r
h
o
b
b
ies
in
ad
d
itio
n
to
p
er
s
is
ten
t
ch
alle
n
g
es
with
s
o
cial
en
g
ag
em
e
n
t
an
d
c
o
m
m
u
n
icatio
n
[
1
3
]
.
K
e
y
s
y
m
p
to
m
s
o
f
A
u
tis
m
ca
n
b
e
s
ee
n
in
Fig
u
r
e
1
.
I
t
h
as
s
u
p
er
s
ed
ed
th
e
ea
r
lier
n
o
m
e
n
clatu
r
e
f
o
r
d
is
o
r
d
er
s
lik
e
Asp
er
g
er
'
s
s
y
n
d
r
o
m
e
an
d
a
u
tis
m
d
is
o
r
d
e
r
th
at
wer
e
co
n
s
id
er
e
d
to
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
F
a
cia
l ima
g
e
a
n
a
lysi
s
fo
r
a
u
tis
m
s
p
ec
tr
u
m
d
is
o
r
d
er d
etec
ti
o
n
in
to
d
d
lers
u
s
in
g
…
(
A
n
u
p
a
m
Da
s
)
4857
b
e
o
n
“
t
h
e
g
r
ea
t
c
o
n
tin
u
u
m
”
o
f
au
tis
m
[
1
4
]
,
[
1
5
]
.
E
v
en
th
o
u
g
h
au
tis
m
h
as
p
r
o
b
ab
l
y
b
ee
n
ar
o
u
n
d
f
o
r
a
wh
ile,
Dr
.
L
eo
Kan
n
e
r
p
r
o
v
id
e
d
th
e
f
ir
s
t
clin
ical
d
escr
ip
tio
n
o
f
th
e
co
n
d
itio
n
in
1
9
4
3
[
1
6
]
.
E
l
ev
en
ch
ild
r
e
n
,
eig
h
t
b
o
y
s
an
d
th
r
ee
g
ir
ls
,
wer
e
d
iag
n
o
s
ed
b
y
Dr
.
Kan
n
er
,
th
e
cr
ea
to
r
o
f
th
e
n
atio
n
'
s
f
ir
s
t
p
ed
iatr
ic
p
s
y
ch
iatr
i
c
p
r
o
g
r
a
m
,
with
wh
at
h
e
ca
lled
“
au
tis
tic
d
is
tu
r
b
an
ce
s
o
f
af
f
ec
t
iv
e
co
n
tact
”
[
1
7
]
.
Ov
er
th
e
Atlan
tic,
at
a
b
o
u
t
th
e
s
am
e
tim
e,
a
p
ed
iatr
ician
f
r
o
m
Au
s
tr
ia
n
am
ed
Han
s
Asp
er
g
er
was
tr
ea
tin
g
a
s
im
ilar
s
et
o
f
k
id
s
.
L
ater
,
a
m
ild
er
v
er
s
io
n
o
f
a
u
tis
m
was r
ef
er
r
ed
to
as
“
Asp
er
g
er
s
y
n
d
r
o
m
e
”
in
h
is
h
o
n
o
r
.
R
esear
ch
er
s
h
av
e
n
o
t
d
eter
m
i
n
ed
th
e
s
p
ec
if
ic
f
ac
to
r
s
ca
u
s
in
g
a
u
tis
m
s
in
ce
th
ey
b
eliev
e
th
at
m
u
ltip
le
g
en
etic
elem
en
ts
alo
n
g
s
id
e
e
n
v
ir
o
n
m
en
tal
f
ac
to
r
s
p
lay
a
co
m
b
in
ed
r
o
le.
T
h
e
o
d
d
s
o
f
d
ev
elo
p
in
g
a
u
tis
m
in
cr
ea
s
e
in
ca
s
es
with
eith
er
g
en
etic
ab
n
o
r
m
alities
o
r
a
f
am
i
ly
m
ed
ical
b
ac
k
g
r
o
u
n
d
.
Valid
atin
g
a
d
iag
n
o
s
is
o
f
au
tis
m
b
ec
o
m
es
p
o
s
s
ib
le
wh
en
ch
ild
r
en
r
ea
c
h
ea
r
ly
ch
ild
h
o
o
d
th
r
o
u
g
h
s
y
s
tem
atic
b
eh
av
io
r
al
ass
ess
m
en
t
s
s
u
p
p
o
r
ted
b
y
d
ev
elo
p
m
en
tal
h
is
to
r
y
r
ev
iew.
B
eh
av
io
r
al
d
iag
n
o
s
is
f
o
r
Au
tis
m
r
elies
o
n
m
u
ltip
le
ex
p
er
t
p
r
o
f
ess
io
n
als
in
clu
d
in
g
p
ed
i
atr
ician
s
alo
n
g
with
p
s
y
ch
o
l
o
g
is
ts
an
d
s
p
ee
ch
-
lan
g
u
a
g
e
p
ath
o
lo
g
is
ts
.
E
ar
ly
tr
ea
tm
en
t
an
d
th
er
ap
ies
s
u
ch
as
s
o
cial
s
k
ills
tr
ain
in
g
to
g
eth
er
with
s
p
ee
ch
th
er
ap
y
an
d
o
cc
u
p
atio
n
al
th
er
ap
y
an
d
b
eh
av
i
o
r
al
in
ter
v
e
n
tio
n
s
h
elp
en
h
an
ce
th
e
life
ex
p
er
ie
n
ce
o
f
th
e
af
f
ec
ted
p
e
r
s
o
n
s
with
au
tis
m
s
p
ec
tr
u
m
d
is
o
r
d
er
d
esp
ite
th
e
c
o
n
d
itio
n
b
ein
g
i
n
cu
r
a
b
le.
T
h
e
C
en
t
er
s
o
f
Dis
ea
s
e
C
o
n
tr
o
l
a
n
d
Pre
v
en
tio
n
(
C
DC
)
p
u
b
lis
h
ed
n
ew
d
ata
o
n
th
e
f
r
e
q
u
en
cy
o
f
au
tis
m
in
th
e
p
o
p
u
l
atio
n
o
f
ch
ild
r
en
: 1
in
3
6
in
t
h
e
Un
ited
States
,
1
in
3
6
in
Ar
izo
n
a
[
1
8
]
,
[
1
9
]
.
T
h
e
a
u
t
is
m
a
n
d
d
e
v
e
l
o
p
m
e
n
ta
l
d
i
s
a
b
i
l
it
i
es
m
o
n
i
t
o
r
i
n
g
n
etw
o
r
k
s
e
n
t
u
p
d
a
t
e
d
d
at
a
t
o
t
h
e
C
DC
a
n
d
p
r
e
v
e
n
t
i
o
n
o
n
M
a
r
c
h
2
3
.
A
c
c
o
r
d
i
n
g
t
o
t
h
e
l
a
t
e
s
t
d
a
t
a
,
1
i
n
3
6
A
m
e
r
i
c
a
n
c
h
i
l
d
r
e
n
a
g
e
d
8
r
e
c
e
i
v
e
d
a
d
i
a
g
n
o
s
i
s
o
f
A
S
D
i
n
2
0
2
0
[
2
0
]
.
C
o
m
p
a
r
e
d
t
o
t
h
e
p
r
e
v
i
o
u
s
l
y
s
t
at
e
d
p
r
e
v
a
le
n
c
e
o
f
1
i
n
4
4
i
n
2
0
1
8
,
t
h
i
s
n
u
m
b
e
r
i
n
d
i
c
a
t
e
s
a
n
i
n
c
r
e
a
s
e
as
s
h
o
w
n
i
n
Fi
g
u
r
e
2
.
Fig
u
r
e
1
.
Key
s
y
m
p
to
m
s
o
f
A
u
tis
m
Fig
u
r
e
2
.
Pre
v
ale
n
ce
r
ate
o
f
A
u
tis
m
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
5
,
Octo
b
e
r
20
25
:
4
8
5
6
-
4
8
6
4
4858
T
h
e
c
u
r
r
e
n
t
t
ec
h
n
i
q
u
e
s
f
o
r
a
u
t
is
m
d
i
a
g
n
o
s
i
s
r
e
q
u
i
r
e
e
x
te
n
s
i
v
e
c
o
s
ts
y
e
t
p
r
o
d
u
c
e
s
u
b
j
ec
t
i
v
e
r
e
s
u
l
ts
t
h
u
s
c
a
u
s
i
n
g
e
a
r
l
y
i
n
t
e
r
v
e
n
t
i
o
n
t
o
b
e
d
e
l
a
y
e
d
.
R
e
s
ea
r
c
h
e
r
s
r
e
q
u
i
r
e
a
n
a
u
t
o
m
a
t
e
d
a
n
d
c
o
s
t
-
e
f
f
e
c
t
iv
e
d
e
t
e
c
t
i
o
n
s
y
s
t
e
m
t
h
a
t
i
d
e
n
t
i
f
i
e
s
A
SD
d
u
r
i
n
g
i
ts
i
n
i
t
i
al
s
t
a
g
e
s
.
T
h
i
s
w
o
r
k
f
i
l
ls
a
c
r
u
c
i
a
l
k
n
o
w
le
d
g
e
g
a
p
t
h
r
o
u
g
h
V
G
G
-
1
6
d
e
e
p
l
e
a
r
n
i
n
g
a
n
a
l
y
s
i
s
o
f
f
a
c
i
al
i
m
ag
e
s
t
o
d
e
v
e
l
o
p
a
n
e
f
f
i
c
i
e
n
t
,
n
o
n
-
c
o
n
t
a
c
t
,
a
n
d
p
r
e
c
i
s
e
m
o
d
e
l
f
o
r
d
e
t
e
c
t
i
n
g
A
SD
,
w
h
i
c
h
r
e
m
o
v
e
s
c
o
n
v
e
n
t
i
o
n
a
l
d
ia
g
n
o
s
i
s
t
e
c
h
n
i
q
u
e
s
'
r
e
q
u
i
r
e
m
e
n
t
s
.
T
h
i
s
r
e
s
e
a
r
c
h
p
r
e
s
e
n
ts
a
n
A
I
a
l
g
o
r
i
t
h
m
b
a
s
e
d
o
n
d
e
e
p
l
e
a
r
n
i
n
g
(
C
N
Ns
)
a
l
o
n
g
w
i
t
h
t
r
a
n
s
f
e
r
l
e
a
r
n
i
n
g
(
V
G
G
-
1
6
)
f
o
r
a
n
a
l
y
zi
n
g
p
i
c
t
u
r
e
s
o
f
t
o
d
d
l
e
r
f
a
c
es
as
a
n
A
S
D
d
e
t
e
c
t
i
o
n
t
o
o
l
.
Us
i
n
g
a
p
r
e
-
t
r
a
in
e
d
V
G
G
-
1
6
m
o
d
e
l
a
l
l
o
ws
t
h
e
a
p
p
r
o
a
c
h
t
o
e
f
f
e
c
tiv
e
l
y
e
x
t
r
a
ct
AS
D
-
r
el
a
t
e
d
f
a
c
ia
l
f
e
at
u
r
e
s
w
h
i
c
h
l
e
a
d
s
t
o
9
9
.
5
0
%
c
l
ass
i
f
i
c
at
i
o
n
a
c
c
u
r
a
c
y
.
U
s
i
n
g
t
h
i
s
m
e
t
h
o
d
h
e
a
l
t
h
c
a
r
e
p
r
o
v
i
d
e
r
s
g
ai
n
a
f
a
s
te
r
a
n
d
m
o
r
e
a
f
f
o
r
d
a
b
l
e
d
i
a
g
n
o
s
t
i
c
t
e
c
h
n
i
q
u
e
t
h
a
t
n
e
e
d
s
n
o
s
u
r
g
e
r
y
.
A
n
o
n
l
i
n
e
A
S
D
s
c
r
e
e
n
i
n
g
t
o
o
l
d
e
r
i
v
e
d
f
r
o
m
t
h
i
s
s
o
l
u
t
i
o
n
w
il
l
h
e
l
p
d
et
e
c
t
AS
D
ea
r
l
y
f
o
r
i
n
t
e
r
v
e
n
t
i
o
n
p
u
r
p
o
s
e
s
w
i
t
h
o
u
t
n
e
e
d
i
n
g
e
x
t
e
n
s
i
v
e
cl
i
n
i
c
al
e
v
a
lu
a
t
i
o
n
s
.
2.
M
E
T
H
O
D
An
aly
zin
g
ASD
is
c
r
u
cial,
b
u
t
d
iag
n
o
s
in
g
it
ca
n
b
e
c
h
allen
g
in
g
s
in
ce
it
lack
s
a
clin
ical
b
en
ch
m
ar
k
,
e.
g
.
b
lo
o
d
test
,
to
d
eter
m
in
e
th
e
is
s
u
e.
T
o
co
m
e
u
p
with
a
co
n
clu
s
io
n
,
p
r
o
f
ess
io
n
als
tak
e
in
to
co
n
s
id
er
atio
n
o
f
th
e
d
ev
el
o
p
m
en
tal
h
is
to
r
y
o
f
t
h
e
teen
ag
e
r
.
A
n
aly
zin
g
ASD
i
s
cr
u
cial
b
ec
a
u
s
e,
in
th
e
ab
s
en
ce
o
f
a
d
iag
n
o
s
is
,
it
ca
n
ca
u
s
e
a
g
r
ea
t
d
ea
l
o
f
d
is
tr
ess
an
d
co
n
f
u
s
io
n
f
o
r
th
e
u
n
id
en
tifie
d
p
er
s
o
n
ab
o
u
t
a
v
ar
iety
o
f
d
ail
y
d
if
f
icu
lties
.
T
h
is
m
ay
r
esu
lt in
d
is
r
u
p
tiv
e
b
e
h
av
io
r
s
a
n
d
s
o
ci
al
d
is
en
g
ag
em
en
t.
I
n
r
esp
o
n
s
e
to
th
e
n
ee
d
,
th
e
ac
ce
s
s
ib
le
d
ata
as
s
o
ciate
d
with
au
tis
m
an
d
its
an
aly
s
i
s
h
av
e
b
ee
n
d
ec
o
n
s
tr
u
cted
.
A
m
o
d
el
h
as
b
ee
n
d
e
v
elo
p
e
d
to
m
a
n
ag
e
q
u
esti
o
n
n
air
es
in
a
q
u
ick
an
d
ea
s
y
way
f
o
r
d
iag
n
o
s
in
g
.
I
n
o
r
d
e
r
to
an
aly
ze
au
tis
m
th
r
o
u
g
h
a
ch
ild
'
s
f
ac
ial
p
ictu
r
e,
a
d
ataset
o
f
f
ac
ial
im
ag
es
h
as
b
ee
n
u
s
ed
th
at
will
h
elp
t
h
e
m
o
d
el
to
p
r
ep
ar
e
,
test
,
an
d
a
p
p
r
o
v
e
.
Nex
t,
u
s
in
g
C
NNs
an
d
ad
d
e
d
tr
an
s
f
er
lear
n
i
n
g
tech
n
iq
u
es,
a
m
o
d
el
h
as b
ee
n
d
ev
elo
p
e
d
.
2
.
1
.
Da
t
a
s
et
I
n
th
is
r
esear
ch
,
th
e
d
ata
was
tak
en
o
f
f
th
e
p
u
b
licly
av
aila
b
le
Kag
g
le
[
2
1
]
.
T
wo
d
if
f
e
r
e
n
t
ty
p
es
o
f
d
atasets
h
av
e
b
ee
n
u
s
ed
in
th
is
an
aly
s
is
.
C
o
n
s
o
lid
ated
is
th
e
n
am
e
o
f
th
e
wo
r
k
o
u
t
s
et.
Au
tis
tic
an
d
n
o
n
-
au
tis
tic
ar
e
it
s
two
s
u
b
-
in
d
ic
es.
Deta
il
s
o
f
th
e
tr
ain
in
g
,
v
alid
atio
n
,
an
d
test
d
ata
ar
e
co
n
s
id
er
ed
as
p
er
T
ab
les
1
,
2
,
an
d
3
ac
c
o
r
d
in
g
l
y
.
Fig
u
r
e
3
illu
s
tr
ates
th
e
d
at
a
p
r
e
-
p
r
o
ce
s
s
in
g
p
ip
elin
e
u
s
e
d
in
th
is
s
tu
d
y
.
DL
alo
n
g
with
th
e
m
eth
o
d
o
f
t
r
an
s
f
er
lear
n
in
g
is
also
ap
p
lied
t
o
r
ec
o
g
n
izin
g
au
tis
m
in
th
e
g
i
v
en
s
tu
d
y
b
y
u
s
in
g
ch
ild
f
ac
ial
p
h
o
to
s
[
2
2
]
–
[
2
5
]
.
T
ab
le
1
.
Deta
ils
o
f
th
e
tr
ai
n
in
g
d
ata
u
s
ed
f
o
r
class
if
icatio
n
C
l
a
s
ses
N
o
.
o
f
i
ma
g
e
s
Ty
p
e
o
f
i
ma
g
e
s
A
u
t
i
st
i
c
1
4
7
0
j
p
e
g
N
o
n
-
A
u
t
i
st
i
c
1
4
7
0
j
p
e
g
T
ab
le
2
.
Deta
ils
o
f
th
e
v
alid
atio
n
d
ata
u
s
ed
f
o
r
class
if
icatio
n
C
l
a
s
ses
N
o
.
o
f
i
ma
g
e
s
Ty
p
e
o
f
i
ma
g
e
s
A
u
t
i
st
i
c
1
0
0
j
p
e
g
N
o
n
-
A
u
t
i
st
i
c
1
0
0
j
p
e
g
T
ab
le
3
.
Deta
ils
o
f
th
e
test
d
at
a
u
s
ed
f
o
r
class
if
icatio
n
C
l
a
s
ses
N
o
.
o
f
i
ma
g
e
s
Ty
p
e
o
f
i
ma
g
e
s
A
u
t
i
st
i
c
1
0
0
j
p
e
g
N
o
n
-
A
u
t
i
st
i
c
1
0
0
j
p
e
g
Fig
u
r
e
3
.
Data
pr
e
-
p
r
o
ce
s
s
in
g
im
ag
es
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
F
a
cia
l ima
g
e
a
n
a
lysi
s
fo
r
a
u
tis
m
s
p
ec
tr
u
m
d
is
o
r
d
er d
etec
ti
o
n
in
to
d
d
lers
u
s
in
g
…
(
A
n
u
p
a
m
Da
s
)
4859
2
.
2
.
Co
nv
o
lutio
na
l
neura
l net
wo
rk
(
CNN)
T
h
e
f
o
u
n
d
atio
n
o
f
DL
,
a
cr
u
c
ial
b
r
an
ch
o
f
ML
,
is
n
e
u
r
al
n
e
two
r
k
s
.
An
in
p
u
t
lay
er
,
a
n
o
u
tp
u
t
lay
er
,
an
d
o
n
e
o
r
m
o
r
e
h
i
d
d
en
lay
e
r
s
ar
e
all
p
r
esen
t
i
n
th
ese
n
e
two
r
k
s
.
No
d
es
ar
e
c
o
n
n
ec
te
d
b
y
th
r
esh
o
ld
s
an
d
weig
h
ts
in
ea
ch
lay
er
.
Data
is
p
as
s
ed
to
th
e
n
ex
t
lay
er
wh
en
a
n
o
d
e'
s
o
u
tp
u
t
s
u
r
p
ass
es
its
th
r
esh
o
ld
;
o
th
er
wis
e,
it
s
tay
s
d
o
r
m
an
t.
T
h
r
ee
p
r
im
a
r
y
lay
er
s
ar
e
p
o
o
lin
g
,
co
n
v
o
lu
tio
n
al
a
n
d
f
u
lly
co
n
n
ec
ted
.
T
h
e
ce
n
tr
al
co
n
s
titu
en
t
o
f
a
C
NN
is
wh
er
e
th
e
m
ain
p
r
o
ce
s
s
in
g
is
d
o
n
e
.
I
n
o
r
d
er
t
o
id
en
tify
p
ar
ticu
la
r
f
ea
tu
r
es
with
in
a
n
in
p
u
t
im
ag
e,
th
is
lay
er
ap
p
lie
s
a
f
ilter
,
o
cc
asio
n
ally
ca
lled
a
k
er
n
el,
wh
ich
is
a
tin
y
m
atr
ix
o
f
weig
h
ts
th
at
m
o
v
es
o
v
e
r
th
e
r
ec
ep
tiv
e
f
ie
ld
.
T
h
e
p
o
o
lin
g
lay
e
r
co
m
es
af
ter
th
e
co
n
v
o
lu
tio
n
al
lay
e
r
in
a
C
NN,
is
an
ess
en
tial
co
m
p
o
n
en
t.
Similar
to
th
e
co
n
v
o
l
u
tio
n
al
lay
er
,
t
h
e
p
o
o
lin
g
lay
er
also
ca
r
r
ies
o
u
t
o
p
er
atio
n
s
t
h
at
in
v
o
lv
e
s
wee
p
in
g
ac
r
o
s
s
th
e
in
p
u
t im
ag
e,
b
u
t f
o
r
a
d
if
f
er
en
t
r
ea
s
o
n
.
Fu
lly
co
n
n
ec
ted
lay
er
w
h
ich
c
lass
if
ies
im
ag
es
u
s
in
g
th
e
p
r
o
p
er
ties
co
llected
f
r
o
m
th
e
p
r
i
o
r
lay
er
s
,
i
s
cr
u
cial
to
th
e
f
in
al
s
tag
es o
f
a
C
NN.
Fig
u
r
e
4
d
ep
icts
th
e
ar
ch
itectu
r
e
o
f
th
e
C
NN
u
s
ed
in
t
h
is
s
tu
d
y
.
A
n
eu
r
o
n
in
a
lay
er
ab
o
v
e
in
d
icate
s
th
at
all
o
th
er
n
eu
r
o
n
s
in
th
e
lay
er
b
elo
w
it,
ar
e
co
n
n
ec
ted
wh
en
it
is
co
n
s
id
er
ed
to
b
e
f
u
lly
in
te
r
co
n
n
ec
ted
.
Fig
u
r
e
4
.
C
NN
ar
ch
itectu
r
e
2
.
3
.
T
ra
ns
f
er
l
ea
rning
B
y
u
s
in
g
in
f
o
r
m
atio
n
f
r
o
m
o
n
e
task
o
r
d
ataset
to
en
h
an
ce
a
m
o
d
el'
s
p
er
f
o
r
m
an
ce
o
n
a
s
ep
ar
ate
b
u
t
r
elate
d
task
,
tr
an
s
f
e
r
lear
n
i
n
g
is
a
ML
tech
n
iq
u
e.
T
o
p
u
t
it
a
n
o
th
er
way
,
tr
a
n
s
f
er
lear
n
in
g
m
ak
es
b
etter
u
s
e
o
f
k
n
o
wled
g
e
ac
q
u
ir
ed
i
n
o
n
e
co
n
tex
t
to
en
h
an
ce
g
en
er
aliza
tio
n
in
an
o
th
e
r
.
As
s
h
o
wn
in
Fig
u
r
e
5
,
ap
p
licatio
n
s
f
o
r
tr
an
s
f
er
lear
n
in
g
ar
e
n
u
m
er
o
u
s
an
d
r
a
n
g
e
f
r
o
m
d
ee
p
le
ar
n
in
g
m
o
d
el
tr
ain
in
g
to
d
ata
s
cien
ce
r
eg
r
ess
io
n
p
r
o
b
lem
-
s
o
lv
in
g
.
T
h
at
is
esp
e
cially
attr
ac
tiv
e
f
o
r
th
e
latter
,
co
n
s
id
er
in
g
th
e
v
o
lu
m
e
o
f
d
ata
r
eq
u
ir
e
d
to
b
u
ild
d
ee
p
n
e
u
r
al
n
etwo
r
k
s
.
Fig
u
r
e
5
.
T
r
an
s
f
er
lear
n
in
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
5
,
Octo
b
e
r
20
25
:
4
8
5
6
-
4
8
6
4
4860
2
.
3
.
1
.
Adv
a
nta
g
es
Sh
o
r
ten
ed
tr
ain
i
n
g
less
tr
ain
in
g
tim
e
tr
an
s
f
er
lear
n
in
g
ca
n
s
av
e
a
v
ast
am
o
u
n
t
o
f
tr
ain
i
n
g
tim
e
b
y
r
ely
in
g
o
n
th
e
k
n
o
wled
g
e
ac
q
u
ir
ed
in
th
e
p
ast.
I
n
s
tead
o
f
s
tar
tin
g
f
r
o
m
s
cr
atch
,
m
o
d
el
s
f
in
e
-
tu
n
e
ex
is
tin
g
weig
h
ts
,
s
av
in
g
co
m
p
u
tatio
n
al
r
eso
u
r
ce
s
an
d
s
p
ee
d
in
g
u
p
d
ev
elo
p
m
en
t.
I
m
p
r
o
v
ed
p
er
f
o
r
m
an
ce
with
l
ess
d
ata.
On
e
o
f
th
e
b
ig
g
est
ad
v
a
n
tag
es
is
im
p
r
o
v
ed
p
e
r
f
o
r
m
an
ce
,
esp
ec
ially
wh
en
d
ea
lin
g
with
s
m
all
d
ataset
s
.
Pre
-
tr
ain
ed
m
o
d
els
ca
n
g
en
er
alize
b
etter
an
d
av
o
id
o
v
er
f
itti
n
g
s
in
ce
th
ey
h
av
e
al
r
ea
d
y
b
ee
n
tr
ain
ed
o
n
lar
g
e
d
atasets
.
L
o
wer
co
m
p
u
tatio
n
a
l
co
s
ts
s
in
ce
th
e
m
o
d
el
d
o
es
n
ot
n
ee
d
to
lear
n
f
r
o
m
th
e
g
r
o
u
n
d
u
p
,
tr
a
n
s
f
er
lear
n
in
g
r
ed
u
ce
s
th
e
n
ee
d
f
o
r
p
o
wer
f
u
l
h
a
r
d
war
e
an
d
lo
n
g
tr
ain
in
g
p
er
io
d
s
,
m
ak
in
g
it
m
o
r
e
co
s
t
-
ef
f
ec
tiv
e
an
d
ac
ce
s
s
ib
le.
E
f
f
ec
tiv
e
with
lim
ited
lab
elled
d
ata
tr
an
s
f
er
lea
r
n
in
g
is
p
ar
ticu
lar
ly
h
elp
f
u
l
i
n
s
ce
n
ar
io
s
wh
er
e
lab
eled
d
ata
is
h
ar
d
to
co
m
e
b
y
.
T
h
e
k
n
o
wled
g
e
f
r
o
m
a
r
elate
d
d
o
m
ain
h
elp
s
th
e
m
o
d
el
lear
n
u
s
ef
u
l
r
ep
r
esen
tatio
n
s
ev
en
wh
en
th
e
tar
g
et
d
o
m
ai
n
lack
s
s
u
f
f
icien
t
lab
eled
ex
am
p
les.
2
.
4
.
VG
G
1
6
VGG1
6
f
u
n
ctio
n
s
as
a
wi
d
ely
ad
o
p
ted
d
ee
p
lear
n
in
g
m
o
d
el
wh
ich
s
u
cc
ess
f
u
lly
id
en
tifie
s
9
2
7
im
ag
es
with
in
its
1
,
0
0
0
ca
teg
o
r
y
d
ataset.
Stak
eh
o
ld
e
r
s
ch
o
o
s
e
VGG1
6
f
o
r
m
an
y
d
ee
p
lear
n
in
g
task
s
b
ec
au
s
e
its
d
ef
in
ed
d
esig
n
to
g
eth
er
with
its
s
im
p
le
p
r
o
g
r
am
m
in
g
an
d
ab
ilit
y
to
p
e
r
f
o
r
m
tr
an
s
f
er
lear
n
in
g
.
T
h
e
n
etwo
r
k
u
n
f
o
ld
s
in
to
1
6
co
m
p
o
s
itio
n
b
lo
ck
s
th
at
in
clu
d
e
1
3
co
n
v
o
lu
tio
n
al
o
p
er
atio
n
s
alo
n
g
with
5
m
ax
-
p
o
o
lin
g
o
p
e
r
atio
n
s
an
d
3
f
u
lly
co
n
n
ec
te
d
o
p
er
atio
n
s
.
T
h
e
s
y
s
tem
an
aly
ze
s
2
2
4
×
2
2
4
R
GB
p
h
o
to
s
wh
ile
u
s
in
g
3
×3
co
n
v
o
lu
tio
n
al
k
er
n
el
s
tr
u
ctu
r
es
with
s
tr
id
e
s
et
at
1
to
o
b
tain
p
r
ec
is
e
f
ea
tu
r
es
f
r
o
m
im
ag
es.
T
wo
ty
p
es
o
f
lay
er
s
with
p
ar
allel
o
p
er
atio
n
s
ar
e
ap
p
lied
to
r
ed
u
ce
d
im
en
s
io
n
s
th
r
o
u
g
h
2
×2
m
a
x
-
p
o
o
lin
g
with
a
s
tr
id
e
o
f
2
.
T
h
e
f
ilter
co
u
n
t
in
cr
ea
s
es
s
y
s
tem
atica
lly
f
r
o
m
C
o
n
v
-
1
(
6
4
f
i
lter
s
)
to
C
o
n
v
-
2
(
1
2
8
f
ilter
s
)
,
C
o
n
v
-
3
(
2
5
6
f
ilter
s
)
an
d
en
d
s
at
C
o
n
v
-
4
an
d
C
o
n
v
-
5
(
5
1
2
f
ilter
s
ea
ch
)
.
T
h
e
m
o
d
el
ap
p
lies
4
0
9
6
n
e
u
r
o
n
s
to
ea
ch
o
f
its
f
ir
s
t
two
f
u
lly
co
n
n
ec
ted
lay
er
s
f
o
llo
wed
b
y
th
e
f
in
al
1
0
0
0
-
class
So
f
tMa
x
ac
tiv
ated
lay
er
.
T
h
e
alter
n
atio
n
p
atter
n
b
etwe
en
p
o
o
lin
g
an
d
c
o
n
v
o
lu
tio
n
lay
er
s
lead
s
to
en
h
an
ce
d
f
ea
tu
r
e
ex
tr
ac
tio
n
a
b
ilit
ies
w
h
ile
in
cr
ea
s
in
g
th
e
class
if
icatio
n
p
r
ec
is
io
n
.
T
r
an
s
f
er
lear
n
i
n
g
allo
ws
VGG1
6
to
o
b
tain
d
o
m
ain
-
s
p
ec
if
ic
ac
c
u
r
ac
y
t
h
r
o
u
g
h
b
r
ie
f
ad
d
itio
n
al
tr
ain
in
g
p
r
o
ce
s
s
es.
T
h
e
co
m
b
in
atio
n
o
f
VGG1
6
m
o
d
el'
s
ex
ce
llen
t
ac
cu
r
ac
y
p
e
r
f
o
r
m
a
n
ce
to
g
et
h
er
with
its
ad
ap
tab
le
p
r
e
-
tr
ain
ed
weig
h
ts
m
ak
es
it
id
ea
l
f
o
r
u
s
e
i
n
m
e
d
ical
im
a
g
in
g
b
esid
es
ac
tin
g
as
a
to
o
l
f
o
r
o
b
ject
d
etec
tio
n
an
d
f
ac
ial
r
ec
o
g
n
itio
n
an
d
au
to
n
o
m
o
u
s
s
y
s
tem
s
ap
p
licatio
n
s
.
T
h
e
b
en
ch
m
ar
k
s
tatu
s
o
f
th
e
d
ee
p
lear
n
in
g
f
r
am
ewo
r
k
k
ee
p
s
its
p
lace
as
a
s
tan
d
ar
d
m
o
d
el
th
at
p
r
o
v
id
es
a
s
o
lid
co
m
b
in
atio
n
b
etwe
en
d
ep
th
an
d
o
p
er
atio
n
al
e
f
f
icien
cy
with
h
i
g
h
ac
cu
r
ac
y
ac
r
o
s
s
d
if
f
er
en
t c
o
m
p
u
ter
v
is
io
n
ap
p
licatio
n
s
.
2.
5.
E
v
a
lua
t
i
o
n o
f
m
o
del
T
h
e
p
er
f
o
r
m
an
ce
ass
ess
m
en
t
o
f
class
if
icatio
n
m
o
d
els
u
s
e
s
ac
cu
r
ac
y
to
g
eth
er
with
p
r
e
cisi
o
n
an
d
r
ec
all
ex
p
r
ess
ed
th
r
o
u
g
h
(
1
),
(
2
)
an
d
(
3
)
.
Mo
d
el
ac
cu
r
ac
y
m
ea
s
u
r
es
th
e
b
alan
ce
b
etwe
en
co
r
r
ec
tly
f
o
r
ec
asted
r
esu
lts
ag
ain
s
t
all
p
r
ed
ictio
n
s
m
ad
e
o
n
th
e
test
in
g
d
ata.
A
cc
u
r
ac
y
d
eter
m
in
es
th
e
p
er
f
o
r
m
an
ce
m
etr
ics
b
y
co
u
n
tin
g
th
e
to
tal
n
u
m
b
er
o
f
f
o
r
ec
asted
r
esu
lts
wh
ile
ac
co
u
n
tin
g
f
o
r
co
r
r
ec
tly
p
r
ed
icte
d
ca
s
es.
=
+
(
1
)
Pre
cisi
o
n
s
tan
d
s
f
o
r
th
e
q
u
o
t
ien
t
b
etwe
en
ac
tu
al
p
o
s
itiv
e
m
atch
es
f
o
r
all
o
f
th
e
p
o
s
itiv
e
f
o
r
ec
asts
.
Mo
d
el
s
h
o
ws its
ca
p
ab
ilit
y
o
f
r
ec
o
g
n
i
zin
g
v
alid
e
x
am
p
les f
r
o
m
a
p
a
r
ticu
lar
f
ield
.
=
+
(
2
)
R
ec
all
d
em
o
n
s
tr
ates
th
e
r
elati
o
n
s
h
ip
b
etwe
en
ac
tu
al
p
o
s
itiv
e
in
s
tan
ce
s
an
d
tr
u
e
p
o
s
itiv
e
p
r
ed
ictio
n
s
to
to
tal
class
in
s
tan
ce
s
.
T
h
e
m
ea
s
u
r
e
in
d
icate
s
wh
eth
er
th
e
m
o
d
el
p
r
o
p
er
l
y
id
en
tifie
s
all
r
elev
an
t
ex
am
p
les
f
r
o
m
an
ass
ig
n
ed
class
.
=
+
(
3
)
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
C
o
n
v
o
lu
tio
n
al
n
e
u
r
al
n
etwo
r
k
s
th
at
also
b
en
ef
it
f
r
o
m
T
L
m
eth
o
d
:
th
e
m
o
d
el
is
tr
ain
e
d
o
n
1
,
4
7
0
f
ac
ial
p
h
o
to
s
o
f
ch
ild
r
en
wh
o
ar
e
au
tis
tic
an
d
1
,
4
7
0
wh
o
ar
e
n
o
t;
s
elec
ted
th
e
f
ea
tu
r
es
f
o
r
d
eter
m
in
in
g
ex
p
licitn
ess
,
af
f
ec
tab
ilit
y
,
an
d
ac
cu
r
ac
y
o
f
t
h
e
p
r
ed
icted
m
o
d
el.
C
NN
wo
r
k
s
with
p
r
e
-
tr
ai
n
ed
VGG1
6
v
er
s
io
n
o
f
I
m
a
g
eNe
t,
th
e
ac
tiv
ato
r
o
f
s
ig
m
o
id
,
Ad
am
o
p
tim
izer
,
an
d
a
8
-
e
p
s
ilo
n
b
in
a
r
y
l
o
s
s
f
u
n
ctio
n
,
wh
ich
a
r
e
s
h
o
wn
in
Fig
u
r
e
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
F
a
cia
l ima
g
e
a
n
a
lysi
s
fo
r
a
u
tis
m
s
p
ec
tr
u
m
d
is
o
r
d
er d
etec
ti
o
n
in
to
d
d
lers
u
s
in
g
…
(
A
n
u
p
a
m
Da
s
)
4861
Fig
u
r
e
6
.
E
p
o
ch
s
3
.
1
.
Rec
t
if
ied
lin
ea
r
un
it
f
un
ct
io
n
(
ReLU )
T
h
is
is
ex
p
r
ess
ed
as
(
4
)
:
(
)
=
x
+
|
x
|
2
(
4
)
w
h
er
e:
(
)
is
th
e
o
u
tp
u
t
o
f
th
e
R
eL
U
f
u
n
ctio
n
;
x
is
th
e
in
p
u
t
v
ar
iab
le
;
|
x
|
r
ep
r
esen
ts
th
e
ab
s
o
lu
te
v
alu
e
o
f
x
,
wh
ich
e
n
s
u
r
es
th
at
th
e
o
u
tp
u
t
is
eith
er
x
(
if
x
is
p
o
s
itiv
e)
o
r
0
(
if
x
is
n
eg
ativ
e)
.
B
e
ca
u
s
e
o
f
its
ea
s
e
o
f
u
s
e
an
d
ca
p
ac
ity
to
s
u
cc
ess
f
u
lly
ad
d
n
o
n
-
lin
ea
r
ity
to
th
e
m
o
d
el,
R
eL
U
is
f
r
eq
u
e
n
tly
u
s
ed
as
an
ac
tiv
atio
n
f
u
n
ctio
n
i
n
n
eu
r
al
n
etwo
r
k
s
.
3
.
2
.
Sig
m
o
id
f
un
ct
io
n
B
ec
au
s
e
it
co
m
p
r
ess
es
in
p
u
t
v
alu
es
in
to
th
is
r
an
g
e,
th
is
f
u
n
ctio
n
is
h
elp
f
u
l
f
o
r
b
i
n
ar
y
cl
ass
if
icatio
n
task
s
,
p
r
o
d
u
cin
g
v
alu
es b
etwe
en
0
an
d
1
i
n
(
5
)
:
=
1
1
+
−
(
5
)
T
h
e
v
al
u
e
r
an
g
e
o
f
A
is
f
r
o
m
0
to
1
.
wh
er
e:
is
th
e
o
u
t
p
u
t
o
f
t
h
e
f
u
n
ctio
n
;
is
th
e
b
ase
o
f
t
h
e
n
atu
r
al
lo
g
ar
ith
m
;
is
th
e
in
p
u
t v
ar
ia
b
l
e.
T
h
e
v
al
u
e
r
a
n
g
e
o
f
A
is
f
r
o
m
0
to
1
,
r
ef
e
r
r
in
g
to
th
at
A
g
o
es
to
0
as
x
g
o
es
t
o
n
eg
ativ
e
in
f
i
n
ity
an
d
A
g
o
es
to
1
as
x
g
o
es
t
o
p
o
s
itiv
e
in
f
in
ity
.
T
h
e
f
o
llo
win
g
lis
ts
o
f
th
e
p
r
e
d
ictio
n
m
o
d
el’
s
o
u
tco
m
es,
alo
n
g
with
its
p
er
f
o
r
m
an
ce
m
etr
ics
an
d
ex
p
ec
tatio
n
s
f
o
r
lear
n
in
g
a
n
d
ad
ap
tin
g
a
r
e
d
is
p
lay
e
d
.
Acc
u
r
ac
y
,
lo
s
s
es,
an
d
co
n
f
u
s
io
n
m
atr
ix
ar
e
s
h
o
wn
i
n
Fig
u
r
es
7
,
8
an
d
9
r
esp
ec
tiv
ely
:
ac
cu
r
ac
y
:
0
.
9
6
6
7
,
lo
s
s
:
0
.
1
0
1
1
,
v
al
-
ac
c
u
r
ac
y
:
0
.
9
9
5
0
,
v
al
-
lo
s
s
: 0
.
0
3
9
1
.
T
ab
l
e
4
d
is
p
lay
s
th
e
class
if
icatio
n
p
er
f
o
r
m
an
ce
m
etr
ics.
Fig
u
r
e
7
.
Acc
u
r
ac
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
5
,
Octo
b
e
r
20
25
:
4
8
5
6
-
4
8
6
4
4862
Fig
u
r
e
8
.
L
o
s
s
es
Fig
u
r
e
9
.
C
o
n
f
u
s
io
n
m
atr
i
x
T
ab
le
4
.
C
lass
if
icatio
n
p
er
f
o
r
m
an
ce
m
etr
ics
C
l
a
s
ses
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
S
c
o
r
e
S
u
p
p
o
r
t
0
0
.
9
8
0
.
9
9
0
.
9
9
1
5
0
1
0
.
9
9
0
.
9
8
0
.
9
8
1
5
0
A
c
c
u
r
a
c
y
0
.
9
8
5
3
0
0
M
a
c
r
o
A
v
g
0
.
9
8
5
0
.
9
8
5
0
.
9
8
5
3
0
0
W
e
i
g
h
t
e
d
A
v
g
0
.
9
8
5
0
.
9
8
5
0
.
9
8
5
3
0
0
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
f
o
r
d
etec
tin
g
ASD
u
s
in
g
f
ac
ial
im
ag
es
with
d
ee
p
lear
n
in
g
an
d
tr
a
n
s
f
e
r
lear
n
in
g
tech
n
iq
u
es
(
VGG
-
1
6
)
d
em
o
n
s
tr
ated
h
ig
h
e
f
f
icien
cy
an
d
a
cc
u
r
ac
y
.
T
h
e
m
o
d
el
ac
c
u
r
ac
y
in
th
e
tr
ain
in
g
an
d
v
alid
atio
n
was
9
6
.
6
7
%
an
d
9
9
.
5
0
,
r
esp
ec
tiv
ely
with
co
r
r
esp
o
n
d
in
g
lo
s
s
v
alu
es
o
f
0
.
1
0
1
1
an
d
0
.
0
3
9
1
,
r
esp
ec
tiv
ely
.
T
h
ese
r
esu
lts
h
ig
h
lig
h
t
th
e
m
o
d
el
’
s
s
tr
o
n
g
p
r
ed
ictiv
e
ca
p
ab
ilit
y
an
d
g
e
n
er
aliza
tio
n
ac
r
o
s
s
d
atasets
.
T
h
e
lo
w
lo
s
s
v
alu
es
s
u
g
g
est
m
in
im
al
er
r
o
r
in
p
r
ed
ictio
n
,
an
d
th
e
h
ig
h
v
alid
atio
n
ac
cu
r
ac
y
im
p
lies
r
esil
ien
ce
to
o
v
er
f
itti
n
g
.
T
h
e
co
n
f
u
s
io
n
m
atr
ix
as
s
h
o
wn
in
Fig
u
r
e
9
,
s
h
o
ws
m
in
im
al
m
is
clas
s
if
icatio
n
s
,
in
d
icatin
g
th
e
m
o
d
el’
s
ab
ilit
y
t
o
d
is
ce
r
n
s
u
b
tle
f
ac
ial
f
ea
t
u
r
e
v
ar
iatio
n
s
in
d
icativ
e
o
f
au
tis
m
.
L
ev
er
ag
in
g
tr
an
s
f
er
lear
n
in
g
th
r
o
u
g
h
th
e
p
r
e
-
tr
ain
ed
VGG
-
1
6
m
o
d
el
r
ed
u
ce
d
co
m
p
u
tatio
n
al
co
s
t a
n
d
im
p
r
o
v
e
d
ac
cu
r
ac
y
.
T
h
e
m
o
d
el
ef
f
ec
tiv
ely
h
an
d
led
b
in
a
r
y
class
if
icatio
n
task
s
u
s
in
g
th
e
b
in
ar
y
cr
o
s
s
-
en
tr
o
p
y
lo
s
s
f
u
n
ctio
n
an
d
s
ig
m
o
id
ac
tiv
atio
n
,
y
ield
in
g
h
ig
h
s
en
s
itiv
ity
an
d
s
p
ec
if
icity
.
C
o
m
p
ar
ed
t
o
p
r
i
o
r
r
esear
ch
as
s
h
o
wn
in
T
a
b
le
5
,
th
e
m
o
d
el
o
u
tp
er
f
o
r
m
ed
o
th
e
r
ap
p
r
o
ac
h
es
r
eg
ar
d
in
g
ac
cu
r
ac
y
,
s
en
s
itiv
ity
,
in
a
d
d
itio
n
t
o
ef
f
icien
cy
o
f
co
m
p
u
tatio
n
,
d
e
m
o
n
s
tr
atin
g
its
ap
p
licab
ilit
y
f
o
r
r
ea
l
-
wo
r
l
d
s
ce
n
ar
io
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
F
a
cia
l ima
g
e
a
n
a
lysi
s
fo
r
a
u
tis
m
s
p
ec
tr
u
m
d
is
o
r
d
er d
etec
ti
o
n
in
to
d
d
lers
u
s
in
g
…
(
A
n
u
p
a
m
Da
s
)
4863
T
h
is
ap
p
r
o
ac
h
r
ed
u
ce
s
th
e
ti
m
e
an
d
c
o
s
t
o
f
tr
a
d
itio
n
al
d
ia
g
n
o
s
tic
m
eth
o
d
s
,
o
f
f
e
r
in
g
a
n
o
n
-
in
v
asiv
e,
d
ep
lo
y
a
b
le
to
o
l
f
o
r
ea
r
ly
d
et
ec
tio
n
o
f
ASD.
Ho
wev
er
,
c
h
allen
g
es
r
elate
d
to
d
ataset
d
iv
er
s
ity
r
em
ain
,
an
d
ex
p
an
d
i
n
g
th
e
d
ataset
an
d
in
c
o
r
p
o
r
ati
n
g
m
u
ltimo
d
al
d
ata
co
u
ld
en
h
an
c
e
th
e
m
o
d
el
’
s
g
en
e
r
aliza
b
ilit
y
.
Fu
r
th
er
r
esear
ch
in
to
m
o
d
el
in
ter
p
r
eta
b
ilit
y
co
u
ld
in
c
r
ea
s
e
tr
u
s
t a
m
o
n
g
clin
ician
s
an
d
f
am
ilies
,
s
o
lid
if
y
in
g
its
r
o
le
as a
v
alu
ab
le
to
o
l
f
o
r
ea
r
ly
d
ia
g
n
o
s
is
.
T
ab
le
5
.
C
o
m
p
a
r
is
o
n
with
ex
i
s
tin
g
wo
r
k
:
m
o
d
el
tr
ain
in
g
an
d
v
alid
atio
n
ac
c
u
r
ac
y
M
o
d
e
l
Tr
a
i
n
i
n
g
a
c
c
u
r
a
c
y
V
a
l
i
d
a
t
i
o
n
a
c
c
u
r
a
c
y
Jah
a
n
a
r
a
a
n
d
P
a
d
m
a
n
a
b
h
a
n
[
2
3
]
0
.
9
6
1
0
0
.
8
4
6
7
P
r
o
p
o
se
d
m
o
d
e
l
w
i
t
h
V
G
G
-
16
0
.
9
6
6
7
0
.
9
9
5
0
4.
CO
NCLU
SI
O
N
E
ar
ly
d
etec
tio
n
o
f
a
u
tis
m
is
cr
u
cial
f
o
r
s
ev
er
al
r
ea
s
o
n
s
,
as
it
ca
n
lead
to
s
ig
n
if
ican
t
p
o
s
itiv
e
o
u
tco
m
es
f
o
r
in
d
i
v
id
u
als
o
n
t
h
e
au
tis
m
s
p
ec
tr
u
m
.
I
f
a
m
ac
h
in
e
lear
n
in
g
m
o
d
el
ca
n
d
o
th
at
in
less
tim
e
in
an
ef
f
icien
t
m
an
n
er
,
th
en
th
is
will
b
e
a
g
r
ea
t
ac
h
iev
em
e
n
t
f
o
r
th
e
m
ed
ical
f
ield
.
I
n
th
is
s
tu
d
y
,
a
ML
class
if
icatio
n
m
o
d
el
with
a
f
ac
ial
im
ag
e
d
at
aset
th
r
o
u
g
h
C
NN
-
s
ty
le
d
ee
p
l
ea
r
n
in
g
an
d
tr
an
s
f
er
lea
r
n
in
g
(
VGG
-
1
6
)
to
d
etec
t
au
tis
m
h
as
b
ee
n
p
r
o
p
o
s
ed
.
I
n
th
e
f
u
tu
r
e
,
u
s
in
g
th
is
m
o
d
el,
an
o
n
lin
e
a
p
p
licatio
n
ca
n
b
e
d
ev
elo
p
e
d
wh
er
e
p
ar
en
ts
ca
n
u
p
lo
ad
im
a
g
es o
f
t
h
eir
ch
ild
r
e
n
an
d
f
in
d
o
u
t th
e
p
o
s
s
ib
ilit
y
o
f
au
tis
m
at
an
ea
r
l
y
s
tag
e.
ACK
NO
WL
E
DG
M
E
N
T
S
Ou
r
s
in
ce
r
e
g
r
atitu
d
e
g
o
es
o
u
t
to
th
e
Me
itY,
Go
v
e
r
n
m
en
t
o
f
I
n
d
ia,
f
o
r
th
eir
s
tead
f
ast
ass
is
tan
ce
th
r
o
u
g
h
th
e
s
ch
o
la
r
s
h
ip
o
f
f
er
e
d
b
y
th
e
Vis
v
esv
ar
ay
a
Ph
.
D.
S
ch
em
e.
RE
F
E
R
E
NC
E
S
[
1
]
D
.
S
i
n
g
h
,
M
.
R
a
k
h
r
a
,
a
n
d
S
.
A
g
g
a
r
w
a
l
,
“
A
u
t
i
sm
s
p
e
c
t
r
u
m
d
i
s
o
r
d
e
r
d
e
t
e
c
t
i
o
n
u
s
i
n
g
t
h
e
D
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s,
”
i
n
2
0
2
2
2
nd
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
T
e
c
h
n
o
l
o
g
i
c
a
l
A
d
v
a
n
c
e
m
e
n
t
s
i
n
C
o
m
p
u
t
a
t
i
o
n
a
l
S
c
i
e
n
c
e
s
(
I
C
T
A
C
S
)
,
O
c
t
.
2
0
2
2
,
p
p
.
7
6
1
–
7
6
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
TA
C
S
5
6
2
7
0
.
2
0
2
2
.
9
9
8
8
4
4
2
.
[
2
]
F
.
Z
h
u
,
S
.
W
a
n
g
,
W
.
L
i
u
,
H
.
Zh
u
,
M
.
Li
,
a
n
d
X
.
Zo
u
,
“
A
m
u
l
t
i
m
o
d
a
l
mac
h
i
n
e
l
e
a
r
n
i
n
g
s
y
st
e
m
i
n
e
a
r
l
y
s
c
r
e
e
n
i
n
g
f
o
r
t
o
d
d
l
e
r
s
w
i
t
h
a
u
t
i
sm
s
p
e
c
t
r
u
m
d
i
s
o
r
d
e
r
s
b
a
se
d
o
n
t
h
e
r
e
s
p
o
n
s
e
t
o
n
a
m
e
,
”
Fr
o
n
t
i
e
rs
i
n
Ps
y
c
h
i
a
t
r
y
,
v
o
l
.
2
6
,
n
o
.
1
4
,
p
.
1
0
3
9
2
9
3
,
Ja
n
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
8
9
/
f
p
s
y
t
.
2
0
2
3
.
1
0
3
9
2
9
3
.
[
3
]
S
.
M
.
H
a
s
a
n
,
M
.
P
.
U
d
d
i
n
,
M
.
A
l
M
a
mu
n
,
M
.
I
.
S
h
a
r
i
f
,
A
.
U
l
h
a
q
,
a
n
d
G
.
K
r
i
s
h
n
a
mo
o
r
t
h
y
,
“
A
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
f
r
a
m
e
w
o
r
k
f
o
r
e
a
r
l
y
-
st
a
g
e
d
e
t
e
c
t
i
o
n
o
f
a
u
t
i
sm
s
p
e
c
t
r
u
m
d
i
so
r
d
e
r
s,
”
I
n
d
u
s
t
ri
a
l
En
g
i
n
e
e
ri
n
g
J
o
u
r
n
a
l
,
v
o
l
.
5
4
,
n
o
.
0
2
,
p
p
.
1
4
2
–
1
5
6
,
2
0
2
5
,
d
o
i
:
1
0
.
3
6
8
9
3
/
i
e
j
.
2
0
2
5
.
v
5
2
i
2
.
0
1
5
.
[
4
]
F
.
Th
a
b
t
a
h
a
n
d
D
.
P
e
e
b
l
e
s
,
“
A
n
e
w
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
m
o
d
e
l
b
a
s
e
d
o
n
i
n
d
u
c
t
i
o
n
o
f
r
u
l
e
s
f
o
r
a
u
t
i
sm d
e
t
e
c
t
i
o
n
,
”
H
e
a
l
t
h
I
n
f
o
rm
a
t
i
c
s
J
o
u
rn
a
l
,
v
o
l
.
2
6
,
n
o
.
1
,
p
p
.
2
6
4
–
2
8
6
,
M
a
r
.
2
0
2
0
,
d
o
i
:
1
0
.
1
1
7
7
/
1
4
6
0
4
5
8
2
1
8
8
2
4
7
1
1
.
[
5
]
O
.
P
.
J
e
n
a
,
B
.
B
h
u
s
h
a
n
,
a
n
d
U
.
K
o
se
,
“
M
a
c
h
i
n
e
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
me
d
i
c
a
l
d
a
t
a
a
n
a
l
y
t
i
c
s
a
n
d
h
e
a
l
t
h
c
a
r
e
a
p
p
l
i
c
a
t
i
o
n
s
,
”
Ma
c
h
i
n
e
L
e
a
rn
i
n
g
a
n
d
D
e
e
p
L
e
a
rn
i
n
g
i
n
M
e
d
i
c
a
l
D
a
t
a
A
n
a
l
y
t
i
c
s
a
n
d
H
e
a
l
t
h
c
a
re
A
p
p
l
i
c
a
t
i
o
n
s
,
p
p
.
1
–
2
7
9
,
2
0
2
2
,
d
o
i
:
1
0
.
1
2
0
1
/
9
7
8
1
0
0
3
2
2
6
1
4
7
.
[
6
]
K
.
K
.
H
y
d
e
e
t
a
l
.
,
“
A
p
p
l
i
c
a
t
i
o
n
s
o
f
s
u
p
e
r
v
i
se
d
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
i
n
a
u
t
i
sm
s
p
e
c
t
r
u
m d
i
s
o
r
d
e
r
r
e
sea
r
c
h
:
A
r
e
v
i
e
w
,
”
Re
v
i
e
w
J
o
u
rn
a
l
o
f
Au
t
i
sm
a
n
d
D
e
v
e
l
o
p
m
e
n
t
a
l
D
i
s
o
rd
e
rs
,
v
o
l
.
6
,
n
o
.
2
,
p
p
.
1
2
8
–
1
4
6
,
J
u
n
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/
s
4
0
4
8
9
-
019
-
0
0
1
5
8
-
x.
[
7
]
B
.
Tw
a
l
a
a
n
d
E.
M
o
l
l
o
y
,
“
O
n
e
f
f
e
c
t
i
v
e
l
y
p
r
e
d
i
c
t
i
n
g
a
u
t
i
sm
s
p
e
c
t
r
u
m
d
i
s
o
r
d
e
r
t
h
e
r
a
p
y
u
si
n
g
a
n
e
n
sem
b
l
e
o
f
c
l
a
ss
i
f
i
e
r
s,
”
S
c
i
e
n
t
i
f
i
c
Re
p
o
r
t
s
,
v
o
l
.
1
3
,
n
o
.
1
,
p
.
1
9
9
5
7
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
3
8
/
s4
1
5
9
8
-
0
2
3
-
4
6
3
7
9
-
3.
[
8
]
J.
Ze
i
d
a
n
e
t
a
l
.
,
“
G
l
o
b
a
l
p
r
e
v
a
l
e
n
c
e
o
f
a
u
t
i
sm
:
A
sy
s
t
e
m
a
t
i
c
r
e
v
i
e
w
u
p
d
a
t
e
,
”
Au
t
i
sm
R
e
se
a
rc
h
,
v
o
l
.
1
5
,
n
o
.
5
,
p
p
.
7
7
8
–
7
9
0
,
M
a
y
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
2
/
a
u
r
.
2
6
9
6
.
[
9
]
C
.
L
o
r
d
,
S
.
R
i
si
,
P
.
S
.
D
i
L
a
v
o
r
e
,
C
.
S
h
u
l
ma
n
,
A
.
Th
u
r
m,
a
n
d
A
.
P
i
c
k
l
e
s,
“
A
u
t
i
sm
f
r
o
m
2
t
o
9
y
e
a
r
s
o
f
a
g
e
,
”
Ar
c
h
i
v
e
s
o
f
G
e
n
e
ra
l
Psy
c
h
i
a
t
r
y
,
v
o
l
.
6
3
,
n
o
.
6
,
p
.
6
9
4
,
J
u
n
.
2
0
0
6
,
d
o
i
:
1
0
.
1
0
0
1
/
a
r
c
h
p
s
y
c
.
6
3
.
6
.
6
9
4
.
[
1
0
]
J.
S
.
H
a
n
d
l
e
ma
n
a
n
d
S
.
H
a
r
r
i
s
,
Pr
e
sc
h
o
o
l
e
d
u
c
a
t
i
o
n
p
r
o
g
ra
m
s
f
o
r c
h
i
l
d
re
n
w
i
t
h
a
u
t
i
sm
,
TX
:
P
r
o
-
Ed
.
2
0
0
1
.
[
1
1
]
N
a
t
i
o
n
a
l
R
e
se
a
r
c
h
C
o
u
n
c
i
l
;
D
i
v
i
s
i
o
n
o
f
B
e
h
a
v
i
o
r
a
l
a
n
d
S
o
c
i
a
l
S
c
i
e
n
c
e
s
a
n
d
E
d
u
c
a
t
i
o
n
;
C
o
mm
i
t
t
e
e
o
n
Ed
u
c
a
t
i
o
n
a
l
I
n
t
e
r
v
e
n
t
i
o
n
s
f
o
r
C
h
i
l
d
r
e
n
w
i
t
h
A
u
t
i
sm,
E
d
u
c
a
t
i
n
g
c
h
i
l
d
re
n
w
i
t
h
a
u
t
i
sm
.
W
a
s
h
i
n
g
t
o
n
,
D
.
C
.
:
N
a
t
i
o
n
a
l
A
c
a
d
e
m
i
e
s
P
r
e
ss,
2
0
0
1
.
[
1
2
]
A
.
D
a
s,
P
.
P
a
t
t
a
n
a
i
k
,
S
.
M
u
k
h
e
r
j
e
e
,
S
.
Tu
r
j
y
a
,
a
n
d
A
.
B
a
n
d
o
p
a
d
h
y
a
y
,
“
E
a
r
l
y
a
u
t
i
sm
s
p
e
c
t
r
u
m
d
i
s
o
r
d
e
r
scre
e
n
i
n
g
i
n
t
o
d
d
l
e
r
s
:
A
c
o
m
p
r
e
h
e
n
si
v
e
st
a
c
k
e
d
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
i
n
g
a
n
d
D
i
g
i
t
a
l
S
y
st
e
m
s
,
v
o
l
.
1
7
,
n
o
.
1
,
p
p
.
1
–
1
2
,
J
a
n
.
2
0
2
5
,
d
o
i
:
1
0
.
1
2
7
8
5
/
i
j
c
d
s/
1
5
7
1
0
1
0
9
5
0
.
[
1
3
]
A
meric
a
n
P
sy
c
h
i
a
t
r
i
c
A
sso
c
i
a
t
i
o
n
,
D
i
a
g
n
o
s
t
i
c
a
n
d
s
t
a
t
i
s
t
i
c
a
l
m
a
n
u
a
l
o
f
m
e
n
t
a
l
d
i
s
o
rd
e
rs
,
5
e
d
.
A
r
l
i
n
g
t
o
n
,
V
A
.
:
A
m
e
r
i
c
a
n
P
sy
c
h
i
a
t
r
i
c
P
u
b
l
i
s
h
i
n
g
,
2
0
1
3
.
[
1
4
]
T.
G
r
a
n
d
i
n
,
T
h
i
n
k
i
n
g
i
n
p
i
c
t
u
res
:
M
y
l
i
f
e
w
i
t
h
a
u
t
i
sm
.
2
0
0
6
.
[
1
5
]
B
.
K
h
a
n
,
S
.
M
.
B
h
a
t
t
i
,
a
n
d
A
.
A
k
r
a
m
,
“
A
u
t
i
sm
s
p
e
c
t
r
u
m
d
i
s
o
r
d
e
r
d
e
t
e
c
t
i
o
n
i
n
c
h
i
l
d
r
e
n
v
i
a
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
s
b
a
s
e
d
o
n
f
a
c
i
a
l
i
ma
g
e
s,
”
B
u
l
l
e
t
i
n
o
f
B
u
si
n
e
ss
a
n
d
Ec
o
n
o
m
i
c
s (
BBE)
,
v
o
l
.
1
3
,
n
o
.
1
,
p
p
.
5
5
7
–
5
7
2
,
2
0
2
4
,
d
o
i
:
1
0
.
6
1
5
0
6
/
0
1
.
0
0
2
4
1
.
[
1
6
]
F
.
R
.
V
o
l
k
mar
a
n
d
A
.
K
l
i
n
,
I
ssu
e
s
i
n
t
h
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
a
u
t
i
sm
a
n
d
rel
a
t
e
d
c
o
n
d
i
t
i
o
n
s
.
2
0
0
5
.
[
1
7
]
K
a
n
n
e
r
L
,
“
A
u
t
i
st
i
c
d
i
st
u
r
b
a
n
c
e
s
o
f
a
f
f
e
c
t
i
v
e
c
o
n
t
a
c
t
,
”
Ac
t
a
P
a
e
d
o
p
s
y
c
h
i
a
t
r
i
c
a
,
v
o
l
.
3
5
,
n
o
.
4
,
p
p
.
1
0
0
–
1
3
6
,
1
9
6
8
.
[
1
8
]
S
.
Z
a
u
d
e
r
e
r
,
“
6
9
+
a
u
t
i
s
m
s
t
a
t
i
s
t
i
c
s:
h
o
w
m
a
n
y
p
e
o
p
l
e
h
a
v
e
a
u
t
i
sm
?
”
A
c
c
e
sse
d
:
M
a
y
0
2
,
2
0
2
2
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
w
w
w
.
c
r
o
ssr
i
v
e
r
t
h
e
r
a
p
y
.
c
o
m
/
a
u
t
i
sm
-
st
a
t
i
st
i
c
s
.
[
1
9
]
Y
.
Lo
f
t
u
s
,
“
A
u
t
i
sm
st
a
t
i
s
t
i
c
s
y
o
u
n
e
e
d
t
o
k
n
o
w
i
n
2
0
2
2
,
”
A
u
t
i
sm
Pa
r
e
n
t
i
n
g
Ma
g
a
zi
n
e
.
p
.
1
,
2
0
2
4
,
A
c
c
e
sse
d
:
D
e
c
.
2
9
,
2
0
2
3
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
w
w
w
.
a
u
t
i
smp
a
r
e
n
t
i
n
g
ma
g
a
z
i
n
e
.
c
o
m/
a
u
t
i
sm
-
s
t
a
t
i
st
i
c
s
/
.
[
2
0
]
S
A
R
R
C
,
“
C
D
C
r
e
l
e
a
ses
n
e
w
a
u
t
i
sm
p
r
e
v
a
l
e
n
c
e
:
1
i
n
3
1
c
h
i
l
d
r
e
n
i
n
t
h
e
U
.
S
.
I
d
e
n
t
i
f
i
e
d
w
i
t
h
a
u
t
i
sm,
”
2
0
2
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
5
,
Octo
b
e
r
20
25
:
4
8
5
6
-
4
8
6
4
4864
h
t
t
p
s
:
/
/
a
u
t
i
smc
e
n
t
e
r
.
o
r
g
/
a
u
t
i
sm
p
r
e
v
a
l
e
n
c
e
/
(
a
c
c
e
ss
e
d
M
a
y
2
0
,
1
B
C
)
.
[
2
1
]
C
.
S
e
n
o
l
,
“
A
u
t
i
sm
_
i
m
a
g
e
_
d
a
t
a
.
”
K
a
g
g
l
e
h
t
t
p
s
:
/
/
w
w
w
.
k
a
g
g
l
e
.
c
o
m
/
d
a
t
a
se
t
s/
c
i
h
a
n
0
6
3
/
a
u
t
i
sm
-
i
ma
g
e
-
d
a
t
a
.
[
2
2
]
F
.
W
.
A
l
s
a
a
d
e
a
n
d
M
.
S
.
A
l
z
a
h
r
a
n
i
,
“
C
l
a
s
si
f
i
c
a
t
i
o
n
a
n
d
d
e
t
e
c
t
i
o
n
o
f
a
u
t
i
sm
sp
e
c
t
r
u
m
d
i
s
o
r
d
e
r
b
a
se
d
o
n
d
e
e
p
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
,
”
C
o
m
p
u
t
a
t
i
o
n
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
N
e
u
r
o
sc
i
e
n
c
e
,
v
o
l
.
2
0
2
2
,
p
p
.
1
–
1
0
,
F
e
b
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
2
/
8
7
0
9
1
4
5
.
[
2
3
]
S
.
Jah
a
n
a
r
a
a
n
d
S
.
P
a
d
ma
n
a
b
h
a
n
,
“
D
e
t
e
c
t
i
n
g
a
u
t
i
sm
f
r
o
m
f
a
c
i
a
l
i
ma
g
e
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
d
v
a
n
c
e
R
e
s
e
a
r
c
h
,
I
d
e
a
s
a
n
d
I
n
n
o
v
a
t
i
o
n
s
i
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
7
,
n
o
.
2
,
p
p
.
2
1
9
–
2
2
5
,
2
0
2
1
,
d
o
i
:
1
0
.
1
3
1
4
0
/
R
G
.
2
.
2
.
3
5
2
6
8
.
3
5
2
0
2
.
[
2
4
]
A
.
S
u
k
u
mar
a
n
,
“
A
st
u
d
y
o
n
a
u
t
i
sm
d
e
t
e
c
t
i
o
n
f
r
o
m
f
a
c
i
a
l
i
m
a
g
e
s,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
f
o
r
Re
s
e
a
r
c
h
i
n
A
p
p
l
i
e
d
S
c
i
e
n
c
e
a
n
d
En
g
i
n
e
e
ri
n
g
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
1
,
n
o
.
1
1
,
p
p
.
2
3
4
4
–
2
3
4
8
,
N
o
v
.
2
0
2
3
,
d
o
i
:
1
0
.
2
2
2
1
4
/
i
j
r
a
se
t
.
2
0
2
3
.
5
6
9
0
8
.
[
2
5
]
A
.
D
a
s,
P
.
K
.
P
a
t
t
n
a
i
k
,
a
n
d
A
.
B
a
n
d
y
o
p
a
d
h
y
a
y
,
“
Ea
r
l
y
i
n
t
e
r
v
e
n
t
i
o
n
i
n
a
u
t
i
s
m sp
e
c
t
r
u
m d
i
s
o
r
d
e
r
:
H
y
b
r
i
d
i
z
i
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
s
w
i
t
h
e
n
s
e
m
b
l
e
m
e
t
h
o
d
s
f
o
r
e
n
h
a
n
c
e
d
d
e
t
e
c
t
i
o
n
u
si
n
g
f
a
c
i
a
l
i
ma
g
e
r
y
,
”
i
n
2
0
2
5
3
rd
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
t
e
l
l
i
g
e
n
t
S
y
st
e
m
s,
A
d
v
a
n
c
e
d
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
(
I
S
AC
C
)
,
F
e
b
.
2
0
2
5
,
p
p
.
7
8
1
–
7
8
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
S
A
C
C
6
5
2
1
1
.
2
0
2
5
.
1
0
9
6
9
3
0
8
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Anu
p
a
m
Da
s
is
e
n
g
a
g
e
d
a
s
a
fu
ll
-
ti
m
e
re
se
a
r
c
h
sc
h
o
lar
a
s
p
a
rt
o
f
th
e
Visv
e
sv
a
ra
y
a
P
h
.
D.
sc
h
e
m
e
,
s
u
p
p
o
rted
b
y
M
e
it
Y,
G
o
v
e
rn
m
e
n
t
o
f
In
d
ia.
H
o
ld
in
g
a
n
M
.
Tec
h
.
d
e
g
re
e
fro
m
Wes
t
Be
n
g
a
l
Un
iv
e
r
sity
o
f
Tec
h
n
o
l
o
g
y
,
h
e
b
o
a
sts
a
n
e
x
ten
siv
e
p
u
b
li
c
a
ti
o
n
re
c
o
rd
in
b
o
t
h
n
a
ti
o
n
a
l
a
n
d
in
tern
a
ti
o
n
a
l
jo
u
rn
a
ls
a
n
d
c
o
n
fe
re
n
c
e
s.
Wi
t
h
o
v
e
r
a
d
e
c
a
d
e
o
f
e
x
p
e
rien
c
e
in
tea
c
h
in
g
,
h
is
e
x
p
e
rti
se
sp
a
n
s
v
a
rio
u
s
d
o
m
a
in
s,
in
c
l
u
d
i
n
g
AI
,
m
a
c
h
in
e
lea
rn
in
g
,
so
c
ia
l
n
e
two
rk
a
n
a
ly
sis,
g
ra
p
h
d
a
tab
a
se
s,
a
n
d
o
p
ti
m
iza
ti
o
n
tec
h
n
i
q
u
e
s.
He
c
a
n
b
e
re
a
c
h
e
d
a
t
a
n
u
k
i
it
2
3
@g
m
a
il
.
c
o
m
.
Pra
sa
n
t
K
u
m
a
r
Pa
ttn
a
ik
is
c
u
rre
n
tl
y
se
rv
i
n
g
a
s
a
p
ro
fe
ss
o
r
a
t
KIIT
Un
iv
e
rsity
,
Bh
u
b
a
n
e
sw
a
r,
In
d
ia.
He
h
o
ld
s
a
P
h
.
D.
i
n
c
o
m
p
u
ter
sc
ien
c
e
a
n
d
is
a
S
e
n
io
r
M
e
m
b
e
r
o
f
th
e
IEE
E.
He
h
a
s
a
ls
o
c
o
n
tri
b
u
te
d
t
o
t
h
e
fiel
d
s
o
f
c
l
o
u
d
c
o
m
p
u
ti
n
g
,
m
o
b
il
e
c
o
m
p
u
ti
n
g
,
b
ra
in
-
c
o
m
p
u
ter
i
n
terfa
c
e
,
a
n
d
p
ri
v
a
c
y
p
ro
tec
ti
o
n
o
v
e
r
t
h
e
p
a
st
fif
tee
n
y
e
a
rs.
In
a
d
d
it
io
n
t
o
b
e
c
o
m
i
n
g
a
fe
ll
o
w
o
f
th
e
I
n
stit
u
ti
o
n
o
f
El
e
c
tro
n
ics
a
n
d
Tele
c
o
m
m
u
n
ica
ti
o
n
En
g
in
e
e
rs
(IE
TE
),
Dr
.
P
a
tt
n
a
ik
h
a
s
a
n
u
m
b
e
r
o
f
re
se
a
rc
h
p
u
b
li
c
a
ti
o
n
s
in
p
re
stig
io
u
s
in
tern
a
ti
o
n
a
l
j
o
u
rn
a
ls
a
n
d
c
o
n
fe
re
n
c
e
s.
He
c
a
n
b
e
re
a
c
h
e
d
a
t
p
a
tn
a
ik
p
ra
sa
n
t@g
m
a
il
.
c
o
m
.
Anja
n
Ba
n
d
y
o
p
a
d
h
y
a
y
a
ss
istan
t
p
r
o
fe
s
so
r
a
t
Ka
li
n
g
a
In
sti
tu
te
o
f
I
n
d
u
strial
Tec
h
n
o
l
o
g
y
,
c
o
m
p
lete
d
h
is
P
h
.
D.
a
t
NIT,
Du
r
g
a
p
u
r
,
u
n
d
e
r
t
h
e
M
HRD
’s
Visv
e
sv
a
ra
y
a
P
h
.
D
.
F
e
ll
o
ws
h
i
p
.
His
re
se
a
rc
h
fo
c
u
se
s
o
n
a
lg
o
r
it
h
m
ic
g
a
m
e
th
e
o
ry
,
p
a
r
ti
c
u
larly
m
e
c
h
a
n
ism
d
e
sig
n
,
with
in
tere
sts
sp
a
n
n
in
g
c
lo
u
d
,
f
o
g
c
o
m
p
u
ti
n
g
,
I
o
T,
h
e
a
lt
h
c
a
re
a
n
d
m
e
tav
e
rse
.
He
h
o
ld
s
a
n
M
.
Tec
h
.
in
I
n
fo
rm
a
ti
o
n
S
e
c
u
rit
y
fro
m
NIT,
Du
rg
a
p
u
r
.
Wi
th
4
2
r
e
se
a
rc
h
p
a
p
e
rs
p
u
b
li
sh
e
d
i
n
v
a
rio
u
s
d
o
m
a
in
s
.
He
h
a
s
g
a
r
n
e
re
d
se
v
e
ra
l
b
e
st
p
a
p
e
r
a
wa
rd
s,
in
c
lu
d
in
g
a
t
3
P
G
CIC.
Dr
.
Ba
n
d
y
o
p
a
d
h
y
a
y
c
u
rre
n
tl
y
se
r
v
e
s
a
s
th
e
n
o
d
a
l
o
ffice
r
o
f
th
e
Vis
v
e
sv
a
ra
y
a
Ph
.
D
.
S
c
h
e
m
e
u
n
d
e
r
th
e
M
e
it
Y,
G
o
v
e
rn
m
e
n
t
o
f
I
n
d
ia.
He
c
a
n
b
e
re
a
c
h
e
d
a
t
a
n
jan
.
b
a
n
d
y
o
p
a
d
h
y
a
y
fc
s@
k
ii
t.
a
c
.
in
.
Evaluation Warning : The document was created with Spire.PDF for Python.