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.
15
,
No
.
4
,
A
u
g
u
s
t 2
0
2
6
,
p
p
.
3
5
0
5
~
3
5
1
7
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
15
.i
4
.
p
p
3
5
0
5
-
3
5
1
7
3505
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Co
mpa
riso
n of m
a
chine learning
al
g
o
rithms to iden
ti
fy
mus
cula
r atrop
hy
and mus
cula
r
dy
stro
phy
G
a
ng
a
B
ha
v
a
ni B
illa
,
Venk
a
t
eswa
ra
Ra
o
P
edda
da
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
a
n
d
En
g
i
n
e
e
r
i
n
g
,
K
o
n
e
r
u
La
k
s
h
ma
i
a
h
E
d
u
c
a
t
i
o
n
F
o
u
n
d
a
t
i
o
n
,
V
a
d
d
e
sw
a
r
a
m,
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
J
an
2
3
,
2
0
2
5
R
ev
is
ed
J
u
n
1
1
,
2
0
2
6
Acc
ep
ted
J
u
l 9
,
2
0
2
6
M
u
sc
u
lar
d
y
stro
p
h
y
is
a
g
e
n
e
t
ic
d
iso
r
d
e
r
c
h
a
ra
c
teriz
e
d
b
y
p
r
o
g
re
ss
iv
e
m
u
sc
le
we
a
k
n
e
ss
a
n
d
d
e
g
e
n
e
ra
t
io
n
,
wh
i
le
m
u
sc
u
lar
a
tr
o
p
h
y
in
v
o
lv
e
s
t
h
e
lo
ss
o
f
m
u
sc
le
ti
ss
u
e
d
u
e
to
p
a
t
h
o
lo
g
ica
l
c
o
n
d
it
i
o
n
s.
Ac
c
u
ra
te
d
iffe
re
n
ti
a
ti
o
n
b
e
twe
e
n
th
e
se
n
e
u
ro
m
u
sc
u
lar
d
iso
rd
e
rs
is
c
rit
ica
l
fo
r
e
ffe
c
ti
v
e
c
li
n
ica
l
d
iag
n
o
sis
a
n
d
trea
tme
n
t
p
lan
n
i
n
g
.
T
h
is
st
u
d
y
p
re
se
n
ts
a
m
a
c
h
in
e
lea
rn
in
g
–
b
a
se
d
fra
m
e
wo
rk
fo
r
c
las
sify
in
g
m
u
sc
u
lar
d
y
str
o
p
h
y
a
n
d
m
u
sc
u
la
r
a
tro
p
h
y
u
sin
g
s
u
rfa
c
e
e
lec
tro
m
y
o
g
ra
p
h
y
(
sEM
G
)
sig
n
a
ls.
sEM
G
d
a
ta
w
e
re
a
c
q
u
ired
a
n
d
p
ro
c
e
ss
e
d
t
o
e
x
trac
t
i
n
fo
r
m
a
ti
v
e
fe
a
tu
re
s
wit
h
l
o
w
c
o
m
p
u
tati
o
n
a
l
c
o
m
p
lex
it
y
.
S
e
v
e
ra
l
m
a
c
h
in
e
le
a
rn
in
g
c
las
sifiers
se
v
e
ra
l
tec
h
n
i
q
u
e
s
we
re
u
se
d
a
n
d
a
ss
e
ss
e
d
,
su
c
h
a
s
ra
n
d
o
m
fo
re
st
(RF
),
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
(S
VM),
g
ra
d
ien
t
b
o
o
sti
n
g
(G
B),
a
n
d
e
x
trem
e
g
ra
d
ien
t
b
o
o
stin
g
(
XG
B
o
o
st
).
To
f
u
rth
e
r
e
n
h
a
n
c
e
c
las
sifica
ti
o
n
p
e
rfo
rm
a
n
c
e
,
a
n
o
v
e
l
e
n
se
m
b
l
e
lea
rn
in
g
a
p
p
ro
a
c
h
c
o
m
b
in
i
n
g
m
u
lt
i
p
le
m
a
c
h
in
e
lea
rn
in
g
m
o
d
e
ls
wa
s
p
ro
p
o
se
d
.
Ex
p
e
rime
n
tal
re
su
lt
s
e
x
h
ib
it
th
a
t
th
e
su
g
g
e
ste
d
e
n
se
m
b
le
m
o
d
e
l
c
o
m
p
lete
s
sig
n
ifi
c
a
n
t
ly
h
i
g
h
e
r
g
ro
u
p
i
n
g
a
c
c
u
ra
ten
e
ss
a
n
d
ro
b
u
st
n
e
ss
c
o
m
p
a
re
d
to
in
d
i
v
id
u
a
l
c
las
sifiers
a
n
d
e
x
isti
n
g
m
e
th
o
d
s
re
p
o
rte
d
in
th
e
li
tera
tu
re
.
Th
e
re
su
lt
s
o
f
th
is
i
n
v
e
sti
g
a
ti
o
n
in
d
ic
a
te
th
a
t
e
n
se
m
b
le
-
b
a
se
d
m
a
c
h
in
e
lea
rn
in
g
tec
h
n
iq
u
e
s
i
n
teg
ra
ted
with
sE
M
G
a
n
a
ly
sis
p
r
o
v
i
d
e
a
re
l
iab
le
a
n
d
n
o
n
-
in
v
a
siv
e
d
iag
n
o
stic
a
id
,
with
stro
n
g
p
o
te
n
ti
a
l
fo
r
e
a
rly
d
e
te
c
ti
o
n
a
n
d
imp
ro
v
e
d
c
li
n
ica
l
m
a
n
a
g
e
m
e
n
t
o
f
n
e
u
ro
m
u
sc
u
lar d
is
o
rd
e
rs.
K
ey
w
o
r
d
s
:
E
lectr
o
m
y
o
g
r
ap
h
y
E
x
tr
em
e
g
r
a
d
ien
t b
o
o
s
tin
g
Gr
ad
ien
t b
o
o
s
tin
g
Mu
s
cu
lar
atr
o
p
h
y
an
d
d
y
s
tr
o
p
h
y
R
an
d
o
m
f
o
r
est
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
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
:
Gan
g
a
B
h
av
an
i Bi
lla
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
in
ee
r
in
g
,
K
o
n
er
u
L
ak
s
h
m
aiah
E
d
u
ca
tio
n
Fo
u
n
d
atio
n
Gr
ee
n
Field
s
,
Vad
d
eswar
am
,
An
d
h
r
a
Pra
d
esh
5
2
2
3
0
2
,
I
n
d
ia
E
m
ail:
b
h
av
an
icse1
0
@
g
m
ail.
c
o
m
1.
I
NT
RO
D
UCT
I
O
N
Mu
s
cle
wea
k
n
ess
an
d
g
r
ad
u
al
d
eter
io
r
atio
n
a
r
e
h
allm
ar
k
s
o
f
a
class
o
f
h
er
ed
itar
y
d
is
o
r
d
er
s
k
n
o
wn
as
m
u
s
cu
lar
d
y
s
tr
o
p
h
ies.
T
h
ey
a
r
e
b
r
o
u
g
h
t
o
n
b
y
m
u
tatio
n
s
in
th
e
g
en
es
th
at
co
d
e
f
o
r
m
u
s
cle
p
r
o
tein
s
an
d
ar
e
ch
ar
ac
ter
ized
b
y
p
r
o
g
r
ess
iv
e
d
eg
en
er
atio
n
o
f
s
k
eleta
l
m
u
s
c
les
[
1
]
–
[
3
]
.
M
u
s
cu
lar
d
y
s
tr
o
p
h
ies
ar
e
p
r
o
g
r
ess
iv
e,
s
y
s
tem
ic
m
u
s
cle
d
is
ea
s
es
th
at
ty
p
ically
ar
is
e
d
u
e
to
m
alf
u
n
ctio
n
in
g
o
r
ab
s
en
t
g
ly
c
o
p
r
o
tein
s
,
s
u
ch
as
d
y
s
tr
o
p
h
in
,
in
th
e
m
u
s
cle
m
em
b
r
an
e
[
1
]
,
[
2
]
,
[
4
]
.
T
h
e
r
ed
u
ctio
n
o
f
s
k
eleta
l
m
u
s
cle
m
ass
is
k
n
o
wn
as
m
u
s
cu
lar
atr
o
p
h
y
,
as
s
h
o
wn
i
n
Fig
u
r
e
1
.
Facto
r
s
s
u
ch
as
im
m
o
b
ilit
y
,
a
g
ein
g
,
m
aln
u
tr
iti
o
n
,
d
r
u
g
u
s
ag
e,
a
n
d
v
ar
io
u
s
n
eu
r
o
l
o
g
ical
o
r
m
u
s
cu
lo
s
k
eleta
l
d
is
o
r
d
er
s
c
an
c
o
n
t
r
ib
u
te
to
m
u
s
cu
lar
atr
o
p
h
y
.
M
u
s
cle
atr
o
p
h
y
lead
s
to
wea
k
n
ess
an
d
f
u
n
ctio
n
al
im
p
air
m
en
t
[
5
]
–
[
7
]
.
A
d
ef
in
in
g
f
ea
tu
r
e
o
f
all
m
u
s
cu
lar
d
y
s
tr
o
p
h
ies
is
th
e
co
n
tin
u
o
u
s
cy
cle
o
f
m
u
s
cle
f
ib
er
d
eg
e
n
er
atio
n
a
n
d
r
e
g
en
er
atio
n
,
as
illu
s
tr
ated
in
Fig
u
r
e
2
[
1
]
,
[
8
]
.
Me
d
ical
p
r
o
f
ess
io
n
als
em
p
lo
y
a
v
ar
iet
y
o
f
d
iag
n
o
s
tic
tech
n
iq
u
es
to
id
en
tify
m
y
o
p
ath
ies
an
d
th
eir
u
n
d
er
ly
in
g
ca
u
s
es.
T
h
ese
in
clu
d
e
elec
tr
o
m
y
o
g
r
ap
h
y
(
E
MG
)
a
n
d
b
l
o
o
d
test
s
to
d
eter
m
in
e
m
u
s
cle
en
z
y
m
e
lev
els
to
ass
es
s
m
u
s
cle
f
u
n
ctio
n
,
a
n
d
m
u
s
cle
b
io
p
s
y
f
o
r
ce
llu
lar
-
lev
el
ex
a
m
in
atio
n
o
f
m
u
s
cle
tis
s
u
e
.
I
n
ce
r
tain
ca
s
es
,
p
ar
ticu
lar
ly
wh
en
h
er
e
d
itar
y
m
y
o
p
ath
ies a
r
e
s
u
s
p
ec
ted
,
g
en
etic
test
in
g
[
2
]
,
[
7
]
,
[
9
]
is
also
r
ec
o
m
m
en
d
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.
15
,
No
.
4
,
Au
g
u
s
t 2
0
2
6
:
3
5
0
5
-
3
5
1
7
3506
Mu
s
cu
lar
atr
o
p
h
y
an
d
m
u
s
cu
l
ar
d
y
s
tr
o
p
h
y
ca
n
b
e
d
etec
ted
u
s
in
g
E
MG
,
as
s
h
o
wn
in
Fig
u
r
e
3
.
E
MG
d
etec
ts
an
o
m
alies
ass
o
ciate
d
with
m
u
s
cu
lar
wea
k
n
ess
,
s
tiff
n
ess
,
an
d
d
eg
en
e
r
atio
n
b
y
m
e
asu
r
in
g
th
e
elec
tr
ical
ac
tiv
ity
o
f
m
u
s
cles
in
r
esp
o
n
s
e
to
n
er
v
e
s
tim
u
latio
n
.
Du
e
to
its
n
o
n
-
in
v
asiv
e
n
atu
r
e
an
d
ab
ilit
y
to
ca
p
tu
r
e
r
ea
l
-
tim
e
m
u
s
cle
ac
tiv
ity
,
E
M
G
h
as
b
ec
o
m
e
a
v
alu
a
b
le
to
o
l
f
o
r
n
e
u
r
o
m
u
s
cu
lar
d
is
o
r
d
e
r
an
aly
s
is
[
1
0
]
–
[
1
3
]
.
Desp
ite
ex
ten
s
iv
e
r
esear
ch
o
n
E
MG
-
b
ased
d
iag
n
o
s
is
,
m
o
s
t
ex
is
tin
g
s
tu
d
ies
f
o
cu
s
o
n
b
i
n
a
r
y
class
if
icatio
n
o
r
r
ely
o
n
s
tatic
d
atasets
,
lim
itin
g
th
eir
a
p
p
licab
ilit
y
in
r
ea
l
-
t
im
e
clin
ical
s
ce
n
ar
io
s
[
1
0
]
–
[
1
5
]
.
T
h
er
ef
o
r
e,
t
h
e
p
r
im
ar
y
r
esear
ch
q
u
esti
o
n
ad
d
r
ess
ed
in
th
is
s
tu
d
y
is
:
ca
n
r
ea
l
-
tim
e
E
MG
s
ig
n
als,
w
h
en
an
aly
ze
d
u
s
in
g
m
ac
h
in
e
lear
n
i
n
g
a
n
d
d
ee
p
l
ea
r
n
in
g
tec
h
n
iq
u
es,
e
f
f
ec
tiv
el
y
d
is
tin
g
u
is
h
b
etwe
en
n
o
r
m
al
m
u
s
cle
co
n
d
itio
n
,
m
u
s
cu
lar
atr
o
p
h
y
,
an
d
m
u
s
cu
l
ar
d
y
s
tr
o
p
h
y
.
T
h
i
s
s
t
u
d
y
h
y
p
o
t
h
e
s
i
z
es
t
h
a
t
a
d
v
a
n
c
e
d
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
m
o
d
e
l
s
a
p
p
li
e
d
t
o
r
ea
l
-
t
i
m
e
E
M
G
d
a
t
a
c
a
n
a
c
h
i
e
v
e
h
i
g
h
e
r
c
l
a
s
s
i
f
ic
a
t
i
o
n
a
cc
u
r
a
cy
a
n
d
r
o
b
u
s
t
n
e
s
s
i
n
d
i
f
f
e
r
e
n
t
i
a
t
i
n
g
t
h
es
e
t
h
r
e
e
m
u
s
c
l
e
c
o
n
d
i
t
i
o
n
s
c
o
m
p
a
r
e
d
t
o
t
r
ad
i
t
i
o
n
a
l
s
t
a
ti
c
d
a
t
as
e
t
-
b
as
e
d
a
p
p
r
o
a
c
h
e
s
.
Fig
u
r
e
1
.
Mu
s
cu
lar
atr
o
p
h
y
Fig
u
r
e
2
.
Mu
s
cu
lar
d
y
s
tr
o
p
h
y
Fig
u
r
e
3
.
E
lectr
o
m
y
o
g
r
ap
h
y
te
s
t
T
h
e
s
ig
n
if
ican
ce
o
f
th
is
d
is
co
v
er
y
r
esid
es
in
its
ca
p
ac
ity
to
f
ac
ilit
ate
ea
r
ly
,
n
o
n
-
in
v
asiv
e,
an
d
r
ea
l
-
tim
e
d
iag
n
o
s
is
o
f
n
eu
r
o
m
u
s
cu
lar
p
r
o
b
lem
s
th
r
o
u
g
h
th
e
u
s
e
o
f
E
MG
s
ig
n
als,
wh
ich
ca
n
h
elp
d
o
cto
r
s
p
lan
tr
ea
tm
en
ts
an
d
in
ter
v
e
n
e
p
r
o
m
p
tly
[
2
]
,
[
3
]
,
[
1
6
]
.
T
h
e
s
u
g
g
ested
m
eth
o
d
im
p
r
o
v
es
d
iag
n
o
s
tic
r
eliab
ilit
y
b
y
b
etter
r
ef
lectin
g
p
h
y
s
io
lo
g
ical
m
u
s
cle
f
u
n
ctio
n
u
n
d
e
r
d
y
n
a
m
ic
co
n
d
itio
n
s
b
y
u
tili
zin
g
r
e
al
-
tim
e
E
MG
d
ata.
Alth
o
u
g
h
E
MG
-
b
ased
m
ac
h
i
n
e
lear
n
in
g
alg
o
r
ith
m
s
h
av
e
b
ee
n
r
esear
ch
ed
f
o
r
n
eu
r
o
m
u
s
cu
lar
p
r
o
b
lem
d
iag
n
o
s
is
in
th
e
p
ast,
m
o
s
t c
u
r
r
en
t r
esear
ch
co
n
ce
n
tr
ates o
n
b
in
ar
y
class
if
icatio
n
,
s
in
g
le
d
i
s
ea
s
e
id
en
tific
atio
n
,
o
r
tr
ad
itio
n
al
s
tan
d
alo
n
e
class
if
ier
s
[
1
0
]
–
[
1
5
]
.
T
h
e
p
r
o
p
o
s
ed
wo
r
k
ex
ten
d
s
th
e
s
tate
-
of
-
th
e
-
ar
t
b
y
o
f
f
er
in
g
an
en
s
em
b
le
lear
n
i
n
g
f
r
am
ewo
r
k
th
at
in
teg
r
ates
r
an
d
o
m
f
o
r
est
(
R
F),
g
r
ad
ie
n
t
b
o
o
s
tin
g
(
GB
)
,
an
d
ex
tr
em
e
g
r
ad
ien
t b
o
o
s
tin
g
(
XGBo
o
s
t)
m
o
d
els
f
o
r
th
e
class
if
icatio
n
o
f
m
u
s
cu
lar
atr
o
p
h
y
an
d
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
u
s
in
g
s
u
r
f
ac
e
elec
tr
o
m
y
o
g
r
ap
h
y
(
s
E
MG
)
d
ata.
T
h
e
en
s
em
b
le
ap
p
r
o
ac
h
b
le
n
d
s
th
e
a
d
v
an
ta
g
es
o
f
s
ev
er
al
class
if
ier
s
.
T
o
in
cr
ea
s
e
r
o
b
u
s
tn
ess
,
lo
wer
p
r
ed
ictio
n
v
a
r
ian
ce
(
VAR
)
,
an
d
im
p
r
o
v
e
g
en
e
r
aliza
tio
n
ab
i
lity
ac
r
o
s
s
v
ar
io
u
s
f
ea
tu
r
e
r
ep
r
esen
tatio
n
s
,
th
e
en
s
em
b
le
tech
n
iq
u
e
lev
er
ag
es th
e
ad
v
an
tag
es o
f
s
ev
er
al
class
if
ier
s
.
Ad
d
itio
n
ally
,
t
h
e
p
r
o
p
o
s
ed
f
r
am
ewo
r
k
o
f
f
e
r
s
a
u
n
if
ied
cla
s
s
if
icatio
n
s
tr
ateg
y
th
at
ca
n
d
is
cr
im
in
ate
m
u
ltip
le
n
eu
r
o
m
u
s
cu
lar
d
is
o
r
d
er
s
u
tili
zin
g
a
s
im
ilar
E
MG
an
aly
tic
p
ip
elin
e,
in
c
o
n
tr
ast
to
ea
r
lier
r
esear
ch
th
at
m
ain
ly
f
o
c
u
s
es
o
n
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
o
r
m
u
s
cu
lar
at
r
o
p
h
y
in
d
e
p
en
d
en
tly
.
I
n
o
r
d
er
to
p
r
o
v
id
e
lig
h
t
o
n
f
ea
tu
r
e
s
en
s
itiv
ity
an
d
m
o
d
el
s
tab
ilit
y
,
th
e
s
tu
d
y
ad
d
itio
n
ally
ex
am
in
es
class
if
ier
b
eh
a
v
io
r
u
n
d
e
r
v
ar
io
u
s
f
ea
tu
r
e
d
im
e
n
s
io
n
s
(
5
,
1
6
0
,
an
d
1
8
1
f
ea
tu
r
es).
T
h
ese
c
o
n
tr
ib
u
tio
n
s
en
ab
le
t
h
e
p
r
o
s
p
ec
tiv
e
u
s
e
o
f
E
MG
-
b
ased
d
iag
n
o
s
tic
s
y
s
tem
s
in
in
tel
lig
en
t
clin
ical
d
ec
is
io
n
-
s
u
p
p
o
r
t
en
v
ir
o
n
m
en
ts
an
d
im
p
r
o
v
e
th
eir
d
ep
en
d
a
b
ilit
y
.
E
ar
lier
r
esear
ch
co
n
ce
n
tr
ated
o
n
m
ac
h
in
e
lear
n
in
g
o
r
d
ee
p
l
ea
r
n
in
g
m
o
d
els
u
s
in
g
s
tatic
d
a
tasets
th
at
in
clu
d
ed
o
n
ly
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
o
r
m
u
s
cu
lar
atr
o
p
h
y
ca
s
es.
I
n
c
o
n
tr
ast,
th
e
p
r
esen
t
s
tu
d
y
ex
ten
d
s
E
MG
-
b
ased
an
aly
s
is
to
th
r
ee
d
is
tin
ct
m
u
s
cle
co
n
d
itio
n
s
b
y
in
tr
o
d
u
ci
n
g
m
ac
h
in
e
lear
n
in
g
an
d
d
ee
p
l
ea
r
n
in
g
alg
o
r
ith
m
s
tailo
r
ed
f
o
r
r
ea
l
-
tim
e
E
MG
d
at
a
d
u
r
in
g
m
u
s
cle
m
o
v
em
en
ts
.
T
h
e
r
em
ain
d
er
o
f
th
is
p
ap
er
i
s
o
r
g
an
ized
as
f
o
llo
ws.
Sectio
n
2
p
r
esen
ts
th
e
r
elate
d
wo
r
k
.
Sectio
n
3
d
escr
ib
es
th
e
p
r
o
p
o
s
ed
m
eth
o
d
o
lo
g
y
.
Sectio
n
4
p
r
esen
ts
th
e
r
esu
lts
an
d
d
is
cu
s
s
io
n
.
Fin
ally
,
s
ec
tio
n
5
co
n
clu
d
es th
e
s
tu
d
y
.
2.
RE
L
AT
E
D
WO
RK
Nu
m
er
o
u
s
s
tu
d
ies
h
av
e
e
x
am
i
n
ed
th
e
u
s
e
o
f
E
MG
,
m
ed
ical
im
ag
in
g
,
g
en
etic
an
aly
s
is
,
an
d
m
ac
h
in
e
lear
n
in
g
ap
p
r
o
ac
h
es
in
th
e
d
iag
n
o
s
is
an
d
ca
teg
o
r
is
atio
n
o
f
n
e
u
r
o
m
u
s
cu
lar
illn
ess
es,
n
am
ely
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
a
n
d
m
u
s
cu
lar
atr
o
p
h
y
.
T
h
r
o
u
g
h
m
ac
h
i
n
e
le
ar
n
in
g
an
d
d
ee
p
lear
n
in
g
tech
n
iq
u
es,
r
ec
en
t
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
C
o
mp
a
r
is
o
n
o
f m
a
ch
in
e
lea
r
n
i
n
g
a
lg
o
r
ith
ms to
id
en
tify m
u
s
cu
la
r
a
tr
o
p
h
y
a
n
d
…
(
Ga
n
g
a
B
h
a
va
n
i B
illa
)
3507
d
ev
elo
p
m
e
n
ts
in
a
r
tific
ial
in
te
llig
en
ce
h
av
e
s
ig
n
if
ican
tly
in
c
r
ea
s
ed
th
e
p
r
ec
is
io
n
a
n
d
ef
f
ec
tiv
en
ess
o
f
illn
ess
d
iag
n
o
s
is
.
I
n
o
r
d
er
to
h
ig
h
li
g
h
t
cu
r
r
en
t
lim
its
an
d
p
r
o
v
id
e
m
o
tiv
atio
n
f
o
r
th
e
p
lan
n
ed
s
tu
d
y
,
th
is
s
ec
tio
n
ex
am
in
es
th
e
m
o
s
t
p
er
tin
e
n
t
p
ap
er
s
co
n
ce
r
n
in
g
E
MG
-
b
ased
d
iag
n
o
s
tics
,
m
ac
h
in
e
lear
n
in
g
an
d
d
ee
p
lear
n
in
g
tech
n
iq
u
es,
an
d
clin
ical
in
v
est
ig
atio
n
s
o
f
m
u
s
cu
lar
illn
ess
es.
Ask
ar
in
ejad
et
a
l.
[
5
]
an
aly
z
ed
2
5
s
EMG
r
ea
d
in
g
s
ca
p
tu
r
ed
f
r
o
m
t
h
e
b
icep
s
m
u
s
cle,
c
o
m
p
r
is
in
g
1
7
h
ea
lth
y
a
n
d
8
atr
o
p
h
ic
s
a
m
p
les,
an
d
em
p
lo
y
ed
a
q
u
a
d
r
atic
d
is
cr
im
in
an
t
a
n
aly
s
is
(
QDA)
class
if
ier
to
ac
h
iev
e
im
p
r
o
v
ed
class
if
icati
o
n
ac
cu
r
a
cy
.
Vo
g
el
an
d
Hag
e
n
g
r
u
b
er
[
6
]
d
e
v
elo
p
e
d
an
s
E
MG
-
b
ased
in
ter
f
ac
e
th
at
en
ab
les in
d
iv
id
u
als with
atr
o
p
h
ic
m
u
s
cles to
co
n
tr
o
l a
r
o
b
o
tic
m
an
ip
u
lato
r
with
th
r
ee
d
eg
r
ee
s
o
f
f
r
e
ed
o
m
.
B
ao
et
a
l.
[
1
7
]
d
esig
n
e
d
a
f
o
u
r
-
class
m
o
to
r
im
a
g
er
y
-
b
ased
b
r
ain
–
co
m
p
u
ter
in
te
r
f
ac
e
s
y
s
te
m
f
o
r
a
s
u
b
ject
with
s
ev
er
e
s
p
in
al
m
u
s
cu
lar
atr
o
p
h
y
.
Sriv
astav
a
et
a
l.
[
1
8
]
d
em
o
n
s
tr
ated
th
at
co
m
b
in
in
g
b
o
t
h
q
u
an
titativ
e
m
u
s
cle
u
ltra
s
o
n
o
g
r
a
p
h
y
(
QM
U)
an
d
elec
tr
ical
im
p
ed
an
ce
m
y
o
g
r
ap
h
y
(
E
I
M)
d
ata
allo
w
s
h
ig
h
ly
ac
cu
r
ate
class
if
icatio
n
o
f
s
p
in
al
m
u
s
cu
lar
atr
o
p
h
y
-
a
f
f
ec
ted
m
u
s
cles.
B
ittma
n
n
et
a
l.
[
1
9
]
r
e
p
o
r
te
d
clin
ical
test
r
esu
lts
f
r
o
m
a
ca
s
e
s
tu
d
y
in
v
o
lv
in
g
a
f
iv
e
-
y
e
ar
-
o
l
d
g
ir
l
d
iag
n
o
s
ed
with
s
p
in
al
m
u
s
cu
lar
atr
o
p
h
y
.
Hass
an
et
a
l.
[
2
0
]
ass
es
s
ed
an
d
co
n
tr
asted
h
o
w
well
d
if
f
er
en
t
m
ac
h
in
e
lear
n
i
n
g
alg
o
r
ith
m
s
class
if
ied
d
ata.
L
eu
n
g
et
a
l.
[
2
1
]
p
r
o
v
id
e
d
a
th
o
r
o
u
g
h
an
aly
s
is
o
f
m
ac
h
in
e
lea
r
n
in
g
ap
p
licatio
n
s
in
g
en
etic
m
ed
icin
e,
em
p
h
asizin
g
ac
ce
s
s
ib
le
d
atasets
an
d
co
m
p
u
tatio
n
al
d
if
f
icu
lties
.
Sath
y
av
ik
asin
i
a
n
d
Vijay
a
[
1
]
p
r
o
p
o
s
ed
a
n
o
v
el
ap
p
r
o
ac
h
f
o
r
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
class
if
icatio
n
u
s
in
g
m
u
tated
g
e
n
e
s
eq
u
e
n
ce
s
an
d
p
o
s
itio
n
al
clo
n
in
g
tech
n
iq
u
es.
Mo
r
en
o
et
a
l.
[
4
]
u
s
in
g
in
t
eg
r
ativ
e
an
aly
s
is
,
a
g
en
e
e
x
p
r
ess
io
n
co
r
e
s
ig
n
at
u
r
e
f
o
r
Du
ch
en
n
e
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
was
f
o
u
n
d
,
an
d
p
r
o
m
is
in
g
th
e
r
ap
eu
t
ic
d
r
u
g
s
wer
e
id
en
tifie
d
.
Hetn
ar
s
k
a
et
a
l.
[
7
]
in
tr
o
d
u
ce
d
a
d
iag
n
o
s
tic
m
et
h
o
d
b
ased
o
n
c
r
ea
tin
e
p
h
o
s
p
h
o
k
in
ase
(
C
PK)
ac
tiv
ity
f
o
r
id
en
tify
in
g
s
p
in
al
m
u
s
cu
lar
atr
o
p
h
y
.
No
f
al
et
a
l.
[
2
2
]
u
tili
ze
d
n
etwo
r
k
an
al
y
s
is
an
d
m
ac
h
in
e
lear
n
i
n
g
tec
h
n
iq
u
es
to
i
d
en
tify
g
en
es
im
p
licated
in
s
p
in
al
m
u
s
cu
lar
atr
o
p
h
y
with
h
ig
h
ac
cu
r
ac
y
.
Dan
o
wsk
i
et
a
l.
[
9
]
in
v
esti
g
ated
s
er
u
m
cr
ea
tin
e
k
in
ase
(
C
K)
lev
els in
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
a
n
d
m
y
o
t
o
n
ia
d
y
s
tr
o
p
h
ica
to
s
u
p
p
o
r
t
d
i
s
ea
s
e
id
en
tific
atio
n
.
Keh
r
i
an
d
Awa
le
[
1
0
]
p
r
o
p
o
s
ed
an
E
MG
s
ig
n
al
an
aly
s
is
m
eth
o
d
th
at
u
s
es
ar
tific
ial
n
eu
r
al
n
etwo
r
k
s
(
ANN)
,
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es
(
SV
M)
,
an
d
wav
elet
tr
an
s
f
o
r
m
to
d
iag
n
o
s
e
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
.
Nar
d
o
et
a
l.
[
1
1
]
d
em
o
n
s
tr
ated
th
e
f
ea
s
ib
ilit
y
o
f
a
m
ac
h
in
e
-
lear
n
in
g
-
b
ased
m
e
th
o
d
f
o
r
p
r
ed
ictin
g
th
e
o
n
s
et
an
d
o
f
f
s
et
tim
in
g
o
f
m
u
s
cu
lar
r
ec
r
u
itm
en
t,
ac
h
ie
v
in
g
h
ig
h
p
e
r
f
o
r
m
an
ce
d
es
p
ite
s
ig
n
if
ican
t
v
ar
iab
ilit
y
in
s
E
MG
s
ig
n
als.
B
itar
et
a
l.
[
1
4
]
p
r
o
p
o
s
ed
a
n
E
MG
-
b
ased
m
ac
h
in
e
lear
n
in
g
f
r
am
ewo
r
k
f
o
r
th
e
d
etec
tio
n
o
f
Du
ch
en
n
e
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
.
Su
b
asi
et
a
l.
[
1
2
]
c
r
ea
ted
an
au
to
m
ated
ap
p
r
o
ac
h
f
o
r
class
if
y
in
g
E
M
G
s
ig
n
als
th
at
u
s
es
b
ag
g
in
g
an
d
d
is
cr
ete
wav
elet
tr
an
s
f
o
r
m
to
d
iag
n
o
s
e
n
eu
r
o
m
u
s
cu
lar
d
is
ea
s
es.
Kef
alas
et
a
l.
[
1
5
]
ass
ess
ed
h
o
w
well
a
u
to
m
ated
m
ac
h
in
e
lear
n
in
g
tech
n
iq
u
es
p
e
r
f
o
r
m
ed
in
id
en
tify
in
g
n
o
r
m
al
a
n
d
ab
n
o
r
m
al
EMG
d
ata.
I
n
am
et
a
l.
[
1
3
]
p
r
o
v
i
d
ed
a
c
o
n
cise
r
e
v
iew
o
f
s
tr
ateg
ies
u
s
ed
f
o
r
E
MG
s
ig
n
al
class
if
icatio
n
.
Sz
ajew
s
k
a
an
d
Ko
p
eć
[
8
]
an
al
y
ze
d
elec
tr
o
m
y
o
g
r
a
p
h
ic
p
atter
n
s
in
Du
ch
en
n
e
an
d
B
ec
k
er
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
ac
r
o
s
s
d
if
f
er
en
t stag
es o
f
m
u
s
cle
d
eg
en
er
atio
n
.
T
ag
h
izad
e
h
et
a
l
.
[
2
3
]
s
h
o
wn
h
o
w
d
ee
p
lear
n
in
g
tec
h
n
iq
u
es
ca
n
b
e
u
s
ed
to
q
u
ick
ly
an
d
ac
cu
r
ately
q
u
an
tify
r
o
tato
r
cu
f
f
m
u
s
cle
d
eg
en
er
atio
n
u
s
in
g
s
h
o
u
ld
er
co
m
p
u
ted
to
m
o
g
r
ap
h
y
(
C
T
)
d
atasets
.
Ma
s
tr
o
p
ietr
o
et
a
l.
[
2
4
]
s
u
g
g
ested
a
d
ee
p
lear
n
in
g
m
o
d
el
b
ased
o
n
s
win
tr
an
s
f
o
r
m
er
s
t
h
at
in
cr
ea
s
es
th
e
ac
cu
r
ac
y
o
f
class
if
y
in
g
v
ar
io
u
s
f
o
r
m
s
o
f
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
u
s
in
g
m
ag
n
etic
r
eso
n
a
n
ce
im
a
g
in
g
(
MRI
)
p
ictu
r
es.
Yan
g
et
a
l.
[
2
5
]
c
r
ea
ted
a
d
ee
p
lea
r
n
in
g
s
y
s
tem
with
g
r
ea
t
d
iag
n
o
s
tic
ac
cu
r
ac
y
f
o
r
id
e
n
tify
in
g
d
y
s
tr
o
p
h
in
o
p
ath
ies
f
r
o
m
th
ig
h
m
u
s
cle
MRI
im
ag
es.
Gar
ib
ald
i
et
a
l.
[
2
6
]
an
al
y
ze
d
m
u
s
cl
e
MR
I
p
at
te
r
n
s
i
n
m
y
o
t
o
n
ic
d
y
s
t
r
o
p
h
y
ty
p
e
1
(
DM
1
)
t
o
r
e
f
i
n
e
m
u
s
cl
e
i
n
v
o
l
v
e
m
e
n
t
ch
ar
ac
t
er
iz
ati
o
n
a
n
d
s
u
p
p
o
r
t
c
li
n
ic
al
t
r
ia
l
d
esi
g
n
.
T
o
b
o
r
e
e
t
a
l.
[
1
6
]
r
e
v
iew
e
d
d
e
ep
l
ea
r
n
i
n
g
–
b
as
ed
in
t
er
v
en
ti
o
n
s
f
o
r
a
d
d
r
ess
i
n
g
h
e
alt
h
c
ar
e
ch
all
e
n
g
es,
h
i
g
h
li
g
h
ti
n
g
k
e
y
c
o
n
s
id
e
r
at
io
n
s
i
n
b
i
o
m
ed
i
ca
l
ap
p
l
ica
ti
o
n
s
.
T
h
o
m
as
et
a
l.
[
2
]
r
e
p
o
r
ted
th
a
t
th
e
tim
e
to
d
iag
n
o
s
is
o
f
D
u
c
h
en
n
e
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
h
as
r
em
ain
e
d
lar
g
ely
u
n
ch
a
n
g
ed
o
v
er
a
1
5
-
y
ea
r
s
u
r
v
eillan
ce
p
er
io
d
.
Qu
i
n
liv
an
[
3
]
em
p
h
asized
th
e
cr
i
tical
im
p
o
r
tan
ce
o
f
ea
r
ly
d
iag
n
o
s
is
in
Du
ch
en
n
e
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
in
o
r
d
er
to
im
p
r
o
v
e
lo
n
g
-
ter
m
p
atien
t
o
u
tco
m
es.
Yan
et
a
l.
[
2
7
]
en
h
an
ce
d
ad
a
p
tab
ilit
y
ac
r
o
s
s
im
ag
in
g
task
s
b
y
th
e
in
tr
o
d
u
ctio
n
o
f
a
f
o
u
n
d
atio
n
m
o
d
el
f
o
r
u
n
iv
er
s
al
m
o
v
in
g
-
o
b
ject
s
eg
m
en
tatio
n
in
m
e
d
ical
im
ag
es.
B
u
s
h
b
y
et
a
l.
[
2
8
]
h
ig
h
lig
h
ted
clin
ical
d
iag
n
o
s
tic
d
elay
s
b
y
r
ec
o
g
n
izin
g
p
er
s
is
ten
t
s
h
o
r
tco
m
in
g
s
in
th
e
p
r
o
m
p
t
d
etec
tio
n
o
f
s
y
m
p
to
m
atic
Du
ch
en
n
e
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
.
Av
an
zo
et
a
l.
[
2
9
]
o
f
f
er
ed
a
co
m
p
r
eh
e
n
s
iv
e
ex
a
m
in
atio
n
o
f
m
ac
h
i
n
e
lear
n
in
g
an
d
d
ee
p
lear
n
in
g
tech
n
iq
u
es
u
s
ed
to
r
ad
io
m
ics
f
o
r
m
ed
ical
im
ag
i
n
g
.
Alth
o
u
g
h
Hu
y
s
m
an
s
et
a
l.
[
3
0
]
d
ev
el
o
p
ed
an
a
u
to
m
ate
d
MRI
-
b
ased
m
eth
o
d
f
o
r
q
u
an
t
if
y
in
g
m
u
s
cle
f
at
f
r
ac
tio
n
s
i
n
p
atien
ts
with
m
u
s
cu
lar
d
y
s
tr
o
p
h
ies,
th
e
s
tu
d
y
d
id
n
o
t
i
n
v
esti
g
ate
r
ea
l
-
tim
e
s
E
MG
-
b
ased
m
u
lti
-
class
class
if
icatio
n
u
s
in
g
m
ac
h
in
e
lea
r
n
in
g
o
r
en
s
em
b
le
lear
n
in
g
tech
n
iq
u
es.
Ov
er
all,
p
r
io
r
r
esear
ch
h
as
d
e
m
o
n
s
tr
ated
th
e
ef
f
icac
y
o
f
E
MG
,
m
ed
ical
im
ag
in
g
,
g
en
eti
c
an
aly
s
is
,
an
d
m
ac
h
in
e
lear
n
i
n
g
tech
n
iq
u
es
in
th
e
d
iag
n
o
s
is
o
f
n
eu
r
o
m
u
s
cu
lar
illn
ess
es.
U
s
in
g
E
M
G
s
ig
n
als,
m
ed
ical
im
ag
in
g
,
an
d
m
ac
h
in
e
lear
n
i
n
g
tech
n
iq
u
es,
n
u
m
er
o
u
s
s
tu
d
ies
h
av
e
ex
am
in
ed
m
u
s
cu
la
r
atr
o
p
h
y
,
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
,
o
r
s
p
in
al
m
u
s
cu
lar
atr
o
p
h
y
[
1
]
–
[
2
3
]
.
Desp
ite
th
ei
r
p
o
s
itiv
e
o
u
tco
m
es,
th
e
m
ajo
r
ity
o
f
th
ese
s
tu
d
ies
co
n
ce
n
tr
ated
o
n
o
f
f
lin
e
d
atasets
,
b
in
ar
y
class
if
icatio
n
,
o
r
a
s
in
g
le
d
is
ea
s
e.
T
h
eir
u
s
e
in
r
ea
l
-
tim
e
d
iag
n
o
s
is
is
th
er
ef
o
r
e
r
estricte
d
.
Fu
r
t
h
er
m
o
r
e,
v
er
y
f
ew
r
esear
ch
h
av
e
t
ak
en
in
to
ac
co
u
n
t
th
e
ca
teg
o
r
is
atio
n
o
f
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
,
m
u
s
cu
lar
atr
o
p
h
y
,
an
d
n
o
r
m
al
m
u
s
cle
h
ea
lth
with
in
a
u
n
if
ied
f
r
am
ewo
r
k
.
T
h
e
s
u
g
g
ested
s
tu
d
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.
15
,
No
.
4
,
Au
g
u
s
t 2
0
2
6
:
3
5
0
5
-
3
5
1
7
3508
o
f
f
er
s
a
r
ea
l
-
tim
e
s
E
MG
-
b
ased
en
s
em
b
le
lear
n
in
g
m
eth
o
d
f
o
r
m
u
lti
-
class
class
if
icatio
n
th
at
m
ak
es
u
s
e
o
f
R
F,
GB
,
an
d
XGBo
o
s
t
in
o
r
d
er
to
g
et
b
ey
o
n
d
th
ese
r
estrictio
n
s
.
To
in
c
r
ea
s
e
class
if
ic
atio
n
ac
cu
r
ac
y
an
d
d
ep
en
d
a
b
ilit
y
,
th
e
s
u
g
g
ested
a
p
p
r
o
ac
h
also
ass
ess
es c
la
s
s
if
ie
r
p
er
f
o
r
m
an
ce
u
s
in
g
v
a
r
io
u
s
f
e
atu
r
e
d
im
en
s
io
n
s
.
3.
M
E
T
H
O
D
T
h
e
s
u
g
g
ested
a
p
p
r
o
ac
h
is
d
is
p
lay
ed
in
Fig
u
r
e
4.
T
h
e
p
r
o
ce
d
u
r
e
o
f
m
o
n
ito
r
in
g
th
e
elec
tr
i
ca
l
ac
tiv
ity
g
en
er
ated
d
u
r
in
g
m
u
s
cle
ac
tiv
atio
n
is
k
n
o
wn
as
E
MG
,
o
r
m
y
o
elec
tr
ic
s
ig
n
al.
An
y
m
u
s
cle'
s
p
er
f
o
r
m
an
ce
ca
n
b
e
ex
am
in
e
d
b
y
r
ec
o
r
d
in
g
th
i
s
elec
tr
ical
ac
tiv
ity
.
Ma
ch
in
e
lear
n
in
g
tech
n
iq
u
es
ca
n
b
e
u
s
ed
to
p
r
ed
ict
wh
e
n
m
u
s
cle
an
o
m
alies will o
cc
u
r
.
Fig
u
r
e
4
.
B
lo
ck
d
iag
r
am
o
f
p
r
o
p
o
s
ed
m
eth
o
d
3
.
1
.
E
lect
ro
m
y
o
g
ra
ph
y
da
t
a
a
cquis
it
io
n
T
h
e
UC
I
Ma
ch
in
e
L
ea
r
n
in
g
R
ep
o
s
ito
r
y
'
s
r
esu
lts
p
r
o
v
id
ed
th
e
d
ataset
th
at
was
u
tili
ze
d
t
o
tr
ain
th
e
p
r
ed
ictiv
e
m
o
d
els.
T
h
e
Natio
n
al
Scien
ce
Fo
u
n
d
atio
n
,
a
d
i
v
is
io
n
o
f
t
h
e
C
en
ter
s
f
o
r
Dis
ea
s
e
C
o
n
tr
o
l
an
d
Pre
v
en
tio
n
(
C
DC
)
,
o
v
er
s
ee
s
UC
I
.
T
h
e
p
r
o
g
r
a
m
u
s
es
a
co
m
b
in
atio
n
o
f
p
h
y
s
ical
test
s
an
d
in
ter
v
iews
to
ass
es
s
th
e
h
ea
lth
a
n
d
n
u
tr
itio
n
al
s
tatu
s
o
f
Am
e
r
ican
s
.
E
lev
e
n
o
f
th
e
twen
ty
-
two
m
ale
in
d
iv
i
d
u
als
h
ad
a
p
r
o
f
ess
io
n
al
d
iag
n
o
s
is
o
f
a
s
ep
ar
ate
k
n
ee
ab
n
o
r
m
ality
.
T
h
ey
p
er
f
o
r
m
ed
th
r
ee
ex
er
cises
to
ass
ess
th
eir
k
n
ee
m
u
s
cle
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
C
o
mp
a
r
is
o
n
o
f m
a
ch
in
e
lea
r
n
i
n
g
a
lg
o
r
ith
ms to
id
en
tify m
u
s
cu
la
r
a
tr
o
p
h
y
a
n
d
…
(
Ga
n
g
a
B
h
a
va
n
i B
illa
)
3509
b
eh
av
io
r
,
walk
in
g
,
leg
f
lex
io
n
u
p
,
an
d
leg
e
x
ten
s
io
n
f
r
o
m
a
s
ea
ted
p
o
s
itio
n
.
T
h
e
ac
q
u
is
itio
n
p
r
o
ce
s
s
in
v
o
l
v
ed
th
e
u
s
e
o
f
a
g
o
n
io
m
eter
in
th
e
k
n
ee
an
d
f
o
u
r
elec
tr
o
d
es:
th
e
r
ec
tu
s
f
em
o
r
is
,
b
icep
s
f
em
o
r
i
s
,
s
em
iten
d
in
o
s
u
s
,
an
d
v
astu
s
m
ed
ialis
.
So
m
e
s
a
m
p
le
d
atasets
ar
e
s
h
o
wn
in
T
a
b
le
1
,
wh
ich
is
r
elate
d
to
ab
n
o
r
m
al
an
d
in
T
ab
le
2
wh
ich
is
r
elate
d
to
th
e
n
o
r
m
al
p
o
s
itio
n
o
f
t
h
e
k
n
ee
m
u
s
cle.
T
ab
le
1
.
Sam
p
le
E
MG
d
ataset
r
elate
d
to
ab
n
o
r
m
al
c
o
n
d
itio
n
o
f
p
atien
t in
g
ait
R
e
c
t
o
f
e
m
o
r
a
l
B
i
c
e
p
s
f
e
m
o
r
a
l
V
a
st
o
m
e
d
i
a
l
EM
G
s
e
m
i
t
e
n
d
i
n
o
s
u
s
F
l
e
x
o
-
e
x
t
e
n
s
i
o
n
-
0
.
0
5
4
8
-
0
.
0
0
1
5
0
.
0
4
2
0
.
0
2
8
5
-
9
0
.
0
7
9
5
-
0
.
0
2
5
6
0
.
0
7
8
0
.
0
3
9
7
-
9
0
.
0
8
4
-
0
.
0
2
9
3
0
.
0
6
9
0
.
0
3
4
5
-
9
0
.
0
8
0
2
-
0
.
0
2
0
3
0
.
0
5
2
5
0
.
0
3
4
5
-
9
0
.
0
8
4
7
-
0
.
0
0
0
8
0
.
0
4
1
2
0
.
0
3
3
7
-
9
0
.
0
9
3
7
0
.
0
0
9
0
.
0
3
8
2
0
.
0
0
8
2
-
9
0
.
0
8
9
2
0
.
0
0
6
0
.
0
3
1
5
-
0
.
0
3
1
5
-
8
.
9
0
.
0
8
7
7
0
.
0
1
0
5
0
.
0
1
6
5
-
0
.
0
4
2
-
8
.
8
0
.
0
9
3
0
.
0
1
8
-
0
.
0
0
9
1
-
0
.
0
2
6
3
-
8
.
7
0
.
0
9
3
7
0
.
0
2
7
7
-
0
.
0
2
8
5
-
0
.
0
1
8
1
-
8
.
7
T
ab
le
2
.
Sam
p
le
E
MG
d
ataset
r
elate
d
to
n
o
r
m
al
co
n
d
itio
n
o
f
p
atien
t in
g
ait
R
e
c
t
o
f
e
m
o
r
a
l
B
i
c
e
p
s fem
o
r
a
l
V
a
st
o
me
d
i
a
l
EM
G
s
e
mi
t
e
n
d
i
n
o
s
u
s
F
l
e
x
o
-
e
x
t
e
n
si
o
n
0
.
0
0
0
7
-
0
.
0
0
8
3
0
.
0
0
4
5
-
0
.
0
0
9
1
5
7
.
6
-
0
.
0
0
0
8
-
0
.
0
0
3
8
0
.
0
0
0
7
-
0
.
0
0
4
6
5
7
.
5
-
0
.
0
0
0
8
-
0
.
0
0
6
8
0
.
0
0
1
5
-
0
.
0
0
2
3
5
7
.
3
-
0
.
0
0
0
8
-
0
.
0
0
5
3
0
.
0
0
4
5
-
0
.
0
0
3
8
5
7
.
1
0
.
0
0
0
7
0
.
0
0
1
5
0
.
0
0
8
2
-
0
.
0
0
7
5
5
6
.
9
0
.
0
0
3
0
.
0
0
5
2
0
.
0
0
9
-
0
.
0
0
8
3
5
6
.
7
0
.
0
0
1
5
0
.
0
0
3
0
.
0
0
6
7
-
0
.
0
0
3
8
5
6
.
5
0
.
0
0
1
5
0
.
0
0
2
2
0
.
0
0
2
2
-
0
.
0
0
2
3
5
6
.
3
0
.
0
0
1
5
0
.
0
0
3
-
0
.
0
0
0
8
0
.
0
0
1
5
5
6
.
1
0
.
0
0
2
2
0
.
0
0
4
5
-
0
.
0
0
0
8
0
.
0
0
8
2
5
5
.
9
3
.
2
.
E
lect
ro
m
y
o
g
ra
ph
y
da
t
a
ins
t
rum
ent
a
t
io
n
B
io
m
etr
ics
em
p
lo
y
ed
MWX8
d
atalo
g
e
q
u
ip
m
e
n
t
with
e
ig
h
t
d
ig
ital
ch
an
n
els
an
d
f
o
u
r
an
alo
g
ch
an
n
els
—
f
o
u
r
f
o
r
s
am
p
lin
g
u
s
in
g
s
E
MG
an
d
o
n
e
f
o
r
g
o
n
io
m
etr
y
.
A
m
icr
o
SD
ca
r
d
wa
s
u
s
ed
to
s
to
r
e
th
e
d
ata
s
tr
aig
h
t
o
n
to
th
e
c
o
m
p
u
ter
'
s
in
ter
n
al
s
to
r
ag
e,
a
n
d
a
B
lu
e
to
o
th
a
d
ap
ter
was
u
s
ed
t
o
tr
an
s
m
it
it
to
r
ea
l
-
tim
e
d
atalo
g
s
o
f
twa
r
e.
T
h
e
p
r
o
g
r
a
m
'
s
s
am
p
lin
g
f
r
eq
u
en
cy
was
1
0
0
0
Hz
,
an
d
its
r
eso
lu
tio
n
was
1
4
b
its
.
T
h
er
e
ar
e
f
o
u
r
elec
tr
o
d
es
in
to
tal,
o
n
e
f
o
r
ea
ch
ch
an
n
el
(
1
–
4
)
an
d
o
n
e
f
o
r
th
e
tim
e
s
er
ies.
Fo
r
ev
er
y
s
u
b
ject
,
th
er
e
ar
e
ar
o
u
n
d
f
iv
e
s
h
ar
es o
r
m
o
ti
o
n
r
ep
etitio
n
s
in
ea
ch
s
er
ies as sh
o
wn
in
Fig
u
r
e
5
.
Fig
u
r
e
5
.
Mu
lti
-
a
r
r
ay
E
MG
te
s
t
3
.
3
.
E
lect
ro
m
y
o
g
ra
ph
y
s
ig
na
l pre
pro
ce
s
s
ing
a
nd
f
ea
t
ure
ex
t
ra
ct
i
o
n
T
h
e
ac
q
u
ir
e
d
s
E
MG
s
ig
n
als
wer
e
s
u
b
jecte
d
to
a
p
r
ep
r
o
ce
s
s
in
g
s
tag
e
to
im
p
r
o
v
e
s
ig
n
al
q
u
ality
an
d
r
ed
u
ce
th
e
i
n
f
lu
en
ce
o
f
n
o
is
e
an
d
in
te
r
-
s
u
b
ject
v
ar
iab
ilit
y
.
I
n
itially
,
th
e
r
aw
E
MG
s
ig
n
als
wer
e
i
n
s
p
ec
ted
f
o
r
m
is
s
in
g
v
alu
es
an
d
o
u
tlier
s
.
Sig
n
al
am
p
litu
d
es
wer
e
th
en
n
o
r
m
alize
d
u
s
in
g
m
in
-
m
ax
n
o
r
m
aliza
tio
n
,
wh
ich
s
ca
les th
e
s
ig
n
al
v
alu
es in
to
th
e
r
an
g
e
o
f
0
to
1
ac
c
o
r
d
in
g
to
:
n
o
r
m
=
−
−
No
r
m
aliza
tio
n
en
s
u
r
es
th
at
all
f
ea
tu
r
es
co
n
tr
ib
u
te
eq
u
ally
d
u
r
in
g
class
if
ier
tr
ain
in
g
an
d
p
r
ev
en
ts
attr
ib
u
tes
with
lar
g
er
m
ag
n
itu
d
es
f
r
o
m
d
o
m
in
atin
g
th
e
lear
n
in
g
p
r
o
ce
s
s
.
E
ac
h
E
MG
ch
an
n
el
h
ad
its
in
f
o
r
m
ativ
e
p
r
o
p
er
ties
ex
t
r
ac
ted
af
ter
n
o
r
m
aliza
tio
n
.
Am
o
n
g
th
e
s
tatis
t
ical
an
d
tem
p
o
r
al
ch
ar
ac
ter
is
tics
co
n
s
id
er
ed
wer
e
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.
15
,
No
.
4
,
Au
g
u
s
t 2
0
2
6
:
3
5
0
5
-
3
5
1
7
3510
m
ea
n
ab
s
o
lu
te
v
alu
e
(
MA
V)
,
r
o
o
t
m
ea
n
s
q
u
ar
e
(
R
MS)
,
V
AR
,
s
tan
d
ar
d
d
ev
iatio
n
(
STD
)
,
wav
ef
o
r
m
len
g
th
(
W
L
)
,
s
ig
n
al
en
er
g
y
,
m
ax
im
u
m
v
alu
e,
m
in
im
u
m
v
alu
e,
a
n
d
ze
r
o
cr
o
s
s
in
g
r
ate
(
Z
C
R
)
.
T
h
ese
ch
ar
ac
ter
is
tics
en
s
u
r
e
a
lo
w
co
m
p
u
tin
g
co
s
t
ap
p
r
o
p
r
iate
f
o
r
r
ea
l
-
tim
e
ex
e
cu
tio
n
wh
ile
ac
cu
r
ately
r
ep
r
e
s
en
tin
g
p
atter
n
s
o
f
m
u
s
cle
ac
tiv
atio
n
.
3
.
4
.
F
e
a
t
ure
s
elec
t
io
n
Usi
n
g
RF
f
ea
tu
r
e
im
p
o
r
tan
ce
an
aly
s
is
,
f
ea
tu
r
es
wer
e
r
an
k
ed
b
ased
o
n
th
eir
c
o
n
tr
ib
u
tio
n
to
class
if
icatio
n
p
er
f
o
r
m
an
ce
b
a
s
ed
o
n
im
p
u
r
ity
r
ed
u
ctio
n
s
c
o
r
es.
T
h
e
h
ig
h
est
-
r
an
k
ed
f
ea
t
u
r
es
wer
e
ch
o
s
en
to
cr
ea
te
r
ed
u
ce
d
f
ea
tu
r
e
s
ets
with
5
,
1
6
0
,
an
d
1
8
1
f
ea
tu
r
es.
T
h
is
m
eth
o
d
r
ed
u
ce
s
m
o
d
el
co
m
p
lex
ity
wh
ile
m
ain
tain
in
g
th
e
d
is
cr
im
in
ati
v
e
in
f
o
r
m
atio
n
r
eq
u
i
r
ed
f
o
r
p
r
ec
is
e
class
if
icatio
n
o
f
m
u
s
cu
lar
atr
o
p
h
y
a
n
d
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
.
3
.
5
.
Repro
du
cibili
t
y
g
uid
eli
nes
E
ac
h
m
ac
h
in
e
lear
n
in
g
m
o
d
el
was
im
p
lem
en
ted
u
s
in
g
Py
th
o
n
an
d
t
h
e
s
cik
it
-
l
ea
r
n
m
o
d
u
l
e.
I
n
o
r
d
er
to
en
s
u
r
e
o
b
jectiv
e
p
er
f
o
r
m
a
n
ce
esti
m
atio
n
,
ten
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
was
em
p
lo
y
ed
i
n
th
e
e
x
p
er
im
e
n
ts
.
W
h
ile
p
r
eser
v
in
g
class
b
alan
ce
,
th
e
d
ataset
was
r
an
d
o
m
l
y
s
p
lit
in
to
s
u
b
s
ets
f
o
r
test
in
g
an
d
tr
ai
n
in
g
.
T
o
p
r
o
v
id
e
f
air
co
m
p
a
r
is
o
n
s
b
et
wee
n
class
if
ier
s
,
h
y
p
er
p
ar
am
e
ter
s
ettin
g
s
wer
e
m
ain
tain
ed
t
h
r
o
u
g
h
o
u
t
th
e
test
s
.
T
h
e
p
u
b
licly
ac
ce
s
s
ib
le
UC
I
d
ataset
ca
n
b
e
u
s
ed
to
r
ep
r
o
d
u
c
e
th
e
en
tire
wo
r
k
f
lo
w,
th
is
co
v
er
s
m
o
d
el
tr
ain
in
g
,
ev
alu
atio
n
,
f
ea
t
u
r
e
ex
t
r
ac
tio
n
,
f
ea
tu
r
e
s
elec
tio
n
,
an
d
p
r
e
p
r
o
ce
s
s
in
g
.
3
.
6
.
P
re
dict
io
n
m
o
dels
Fin
d
in
g
p
atter
n
s
in
d
if
f
e
r
en
t
f
ea
tu
r
es
is
o
n
e
o
f
th
e
m
ain
g
o
a
ls
o
f
u
s
in
g
m
ac
h
in
e
lear
n
i
n
g
a
lg
o
r
ith
m
s
o
n
lar
g
e
d
atasets
.
B
o
th
m
u
s
cu
lar
atr
o
p
h
y
a
n
d
m
u
s
cu
lar
d
y
s
t
r
o
p
h
y
cr
iter
ia
wer
e
u
s
ed
to
la
b
el
th
e
d
ata
in
th
is
in
v
esti
g
atio
n
.
SVM,
GB
,
X
GB
o
o
s
t
,
an
d
RF
ar
e
th
e
s
u
p
er
v
is
ed
lear
n
in
g
m
eth
o
d
s
u
s
ed
in
th
is
wo
r
k
.
Py
th
o
n
s
cik
it
-
lear
n
was
u
s
ed
to
im
p
lem
en
t
ea
ch
al
g
o
r
ith
m
.
T
h
is
p
a
p
er
co
m
p
ar
e
s
th
e
al
g
o
r
ith
m
s
wh
ich
ar
e
p
r
o
v
i
d
e
d
b
etter
ac
cu
r
ac
y
i
n
th
e
an
al
y
s
is
o
f
s
E
MG
s
ig
n
als
to
id
en
tify
m
u
s
cu
lar
atr
o
p
h
y
an
d
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
d
is
ea
s
es.
B
y
co
m
p
ar
in
g
t
h
e
o
u
tco
m
es
o
f
th
ese
f
o
u
r
class
if
ier
s
u
s
in
g
ass
es
s
m
en
t
p
ar
am
eter
s
,
th
e
p
r
ed
ictio
n
ac
cu
r
ac
y
r
o
s
e.
T
h
e
b
est cla
s
s
if
ier
is
d
escr
ib
ed
b
y
th
e
ex
p
er
i
m
en
tal
r
esu
lt in
b
etwe
en
th
e
m
.
3
.
6
.
1
.
Su
pp
o
rt
v
ec
t
o
r
m
a
chi
ne
I
t
is
a
m
eth
o
d
f
o
r
r
eg
r
ess
io
n
a
n
d
class
if
icatio
n
.
E
ac
h
d
ata
p
o
in
t
in
SVM
is
p
lo
tted
in
n
-
d
i
m
en
s
io
n
al
s
p
ac
e,
wh
er
e
th
e
n
u
m
b
er
o
f
d
im
en
s
io
n
s
co
r
r
esp
o
n
d
s
to
th
e
a
ttrib
u
te
o
r
f
ea
t
u
r
e
co
u
n
t
as
s
h
o
w
in
Fig
u
r
e
6.
T
h
e
n
u
m
b
er
o
f
attr
ib
u
tes
is
d
en
o
ted
b
y
n
.
E
v
er
y
attr
ib
u
te'
s
v
al
u
e
co
r
r
esp
o
n
d
s
to
a
s
p
ec
if
ic
co
o
r
d
in
ate
,
a
n
d
th
e
d
ec
is
io
n
f
u
n
ctio
n
is
r
ep
r
esen
te
d
as sh
o
wn
in
(
1
)
.
=
+
(
1
)
T
h
e
h
y
p
er
p
lan
e
eq
u
atio
n
is
g
iv
en
b
y
(
2
)
.
+
=
0
(
2
)
Fig
u
r
e
6
.
Su
p
p
o
r
t
v
ec
to
r
s
m
ac
h
in
e
3
.
6
.
2
.
G
ra
dient
bo
o
s
t
ing
On
e
o
f
th
e
m
o
s
t
p
o
p
u
lar
ty
p
es
o
f
m
ac
h
in
e
lear
n
in
g
m
o
d
e
ls
is
th
e
p
r
ed
ictio
n
m
o
d
el.
I
n
m
ac
h
in
e
lear
n
in
g
,
th
e
GB
alg
o
r
ith
m
s
tan
d
s
o
u
t
f
o
r
its
ac
cu
r
ac
y
an
d
s
p
ee
d
o
f
p
r
ed
ictio
n
,
esp
ec
ially
wh
en
wo
r
k
in
g
wit
h
b
ig
an
d
co
m
p
licated
d
atasets
.
T
h
e
m
ath
em
atica
l
r
ep
r
esen
tatio
n
o
f
GB
is
s
h
o
wn
in
(
3
)
.
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
C
o
mp
a
r
is
o
n
o
f m
a
ch
in
e
lea
r
n
i
n
g
a
lg
o
r
ith
ms to
id
en
tify m
u
s
cu
la
r
a
tr
o
p
h
y
a
n
d
…
(
Ga
n
g
a
B
h
a
va
n
i B
illa
)
3511
(
)
=
−
1
(
)
+
ℎ
(
)
(
3
)
T
h
e
class
if
icatio
n
p
r
o
b
lem
'
s
lo
s
s
f
u
n
ctio
n
is
d
ef
in
ed
in
(
4
)
.
=
∑
(
−
(
)
)
(
4
)
3
.
6
.
3
.
E
x
t
re
m
e
g
ra
dient
bo
o
s
t
ing
A
s
tate
-
of
-
th
e
-
ar
t
m
ac
h
in
e
l
ea
r
n
in
g
tec
h
n
iq
u
e
n
o
te
d
f
o
r
its
ex
ce
p
tio
n
al
p
r
ed
ictio
n
ab
ilit
ies
is
XGBo
o
s
t.
T
h
e
f
o
r
m
u
la
f
o
r
l
o
s
s
f
u
n
ctio
n
o
r
m
ea
n
s
q
u
ar
ed
er
r
o
r
(
MSE
)
is
as sh
o
wn
in
(
5
)
.
(
)
=
min
∑
(
,
)
=
1
=
min
∑
(
,
)
2
=
1
(
5
)
3
.6
.4
.
Ra
nd
o
m
f
o
re
s
t
T
o
g
en
er
ate
p
r
ed
ictio
n
s
,
ea
c
h
tr
ee
is
v
o
ted
ac
r
o
s
s
u
s
in
g
th
e
RF
alg
o
r
ith
m
,
a
p
o
wer
f
u
l
tr
e
e
lear
n
in
g
tech
n
iq
u
e
in
m
ac
h
in
e
lear
n
in
g
.
B
o
th
class
if
icatio
n
an
d
r
e
g
r
ess
io
n
task
s
m
ak
e
ex
ten
s
iv
e
u
s
e
o
f
th
e
m
.
T
h
e
f
o
r
m
u
la
f
o
r
ca
lc
u
latin
g
th
e
f
ea
tu
r
e
im
p
o
r
tan
ce
is
d
ef
in
e
d
in
(
6
)
.
=
∑
(
)
(
6
)
3
.
7
.
E
v
a
lua
t
i
o
n pa
ra
m
e
t
er
s
Acc
u
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F
-
m
ea
s
u
r
e
ar
e
a
f
ew
d
ata
m
in
in
g
ev
alu
atio
n
m
ea
s
u
r
es
.
W
h
er
e
FP
,
FN,
T
P,
an
d
T
N
s
tan
d
f
o
r
f
a
ls
e
p
o
s
itiv
e,
f
alse
n
eg
ativ
e,
tr
u
e
p
o
s
iti
v
e,
an
d
tr
u
e
n
eg
ativ
e,
r
esp
ec
tiv
ely
.
As
s
tated
in
(
7
)
,
a
cc
u
r
ac
y
is
ca
lc
u
lated
b
y
d
i
v
id
in
g
th
e
n
u
m
b
e
r
o
f
co
r
r
ec
tly
id
e
n
tifie
d
i
n
s
tan
ce
s
b
y
th
e
d
ataset'
s
o
v
er
all
n
u
m
b
er
o
f
in
s
tan
ce
s
.
=
+
+
+
+
(
7
)
Acc
o
r
d
in
g
t
o
(
8
)
,
p
r
e
cisi
o
n
is
d
ef
in
ed
as th
e
m
ea
n
lik
elih
o
o
d
o
f
p
er
ti
n
en
t
r
etr
iev
al.
=
TP
+
FP
(
8
)
Acc
o
r
d
in
g
t
o
(
9
)
,
th
e
a
v
er
ag
e
l
ik
elih
o
o
d
o
f
f
u
ll r
etr
iev
al
is
th
e
r
ec
all.
=
TP
+
FN
(
9
)
As in
d
icate
d
in
(
10
)
,
th
e
F
-
m
e
asu
r
e
is
co
m
p
u
ted
u
tili
zin
g
b
o
th
p
r
ec
is
io
n
an
d
r
ec
all.
F
−
me
a
s
ure
=
2
×
(
×
)
Pr
ecis
i
o
n
+
Recal
l
(
10
)
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
4
.
1
.
Da
t
a
s
et
s
T
h
e
E
MG
d
ataset
m
en
tio
n
ed
i
n
Sectio
n
3
.
1
was
u
s
ed
f
o
r
th
e
ex
p
er
im
en
tal
ev
al
u
atio
n
.
T
h
r
ee
f
ea
tu
r
e
s
ets
(
5
,
1
6
0
,
an
d
1
8
1
f
ea
tu
r
e
s
)
wer
e
p
r
o
d
u
ce
d
f
o
r
class
if
icatio
n
af
ter
p
r
e
p
r
o
ce
s
s
in
g
an
d
f
ea
tu
r
e
ex
tr
ac
ti
o
n
.
B
o
th
th
e
s
u
g
g
ested
en
s
em
b
le
m
o
d
el
an
d
th
e
in
d
iv
i
d
u
al
m
ac
h
in
e
lear
n
in
g
m
o
d
els
wer
e
tr
ain
ed
an
d
test
ed
u
s
in
g
th
ese
f
ea
tu
r
e
s
ets.
Stan
d
ar
d
cr
iter
ia,
s
u
ch
as
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e,
wer
e
u
s
ed
t
o
ass
es
s
ea
ch
class
if
ier
'
s
p
er
f
o
r
m
an
ce
an
d
c
o
m
p
ar
e
h
o
w
well
it
id
en
tifie
d
n
o
r
m
al
m
u
s
cle
c
o
n
d
itio
n
,
m
u
s
cu
lar
atr
o
p
h
y
,
an
d
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
.
T
h
e
ac
cu
r
ac
y
,
co
m
p
leten
ess
,
an
d
v
o
lu
m
e
o
f
d
ata
i
n
th
e
UC
I
d
atab
ase
ar
e
well
-
k
n
o
wn
.
T
h
e
s
u
p
er
io
r
p
er
f
o
r
m
an
ce
o
f
th
e
p
r
o
p
o
s
e
d
en
s
em
b
le
-
b
ased
f
r
am
ewo
r
k
d
em
o
n
s
tr
ates
th
e
ef
f
ec
tiv
e
n
ess
o
f
c
o
m
b
in
in
g
co
m
p
lem
en
tar
y
lear
n
in
g
s
tr
ateg
ies
f
o
r
E
MG
s
ig
n
al
class
if
icatio
n
.
T
h
e
en
s
em
b
le
tech
n
i
q
u
e
b
en
e
f
ited
f
r
o
m
lo
wer
VAR
an
d
b
etter
g
en
er
al
izatio
n
ac
r
o
s
s
v
a
r
io
u
s
f
ea
tu
r
e
s
u
b
s
ets,
wh
er
ea
s
in
d
iv
id
u
al
cl
ass
if
ier
s
lik
e
SVM
an
d
GB
p
r
o
d
u
ce
d
co
m
p
etitiv
e
r
esu
lts
.
I
n
co
m
p
ar
is
o
n
to
in
d
ep
en
d
e
n
t
m
ac
h
in
e
lear
n
in
g
m
o
d
els,
th
e
f
in
d
in
g
s
u
g
g
ests
th
at
en
s
em
b
le
lear
n
i
n
g
ca
n
m
o
r
e
ac
c
u
r
ately
ca
p
t
u
r
e
th
e
in
tr
icate
an
d
n
o
n
-
lin
ea
r
p
atter
n
s
lin
k
ed
t
o
m
u
s
cle
atr
o
p
h
y
an
d
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
,
en
h
an
cin
g
d
iag
n
o
s
tic
r
eliab
ilit
y
.
4
.
2
.
M
a
chine le
a
rning
cla
s
s
i
f
iers
Fo
u
r
class
if
icatio
n
tech
n
iq
u
es
ar
e
u
s
ed
i
n
th
is
s
tu
d
y
an
d
ar
e
im
p
lem
en
te
d
in
Py
th
o
n
.
Pre
d
ictio
n
is
en
h
an
ce
d
by
th
e
u
s
ag
e
of
th
e
s
e
m
o
d
els.
To
d
ete
r
m
in
e
w
h
ich
class
if
ier
b
est
p
r
ed
icts
a
p
atien
t's
lik
elih
o
o
d
o
f
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.
15
,
No
.
4
,
Au
g
u
s
t 2
0
2
6
:
3
5
0
5
-
3
5
1
7
3512
d
ev
elo
p
in
g
m
u
s
cu
lar
atr
o
p
h
y
an
d
d
y
s
tr
o
p
h
y
illn
ess
es,
s
ev
er
al
class
if
ier
s
ar
e
co
m
p
a
r
ed
.
A
b
r
ief
e
x
p
lan
atio
n
o
f
th
ese
ca
teg
o
r
izatio
n
m
eth
o
d
s
an
d
class
if
ier
s
is
p
r
o
v
id
ed
as f
o
llo
ws:
i)
SVM
:
t
h
e
SVM
class
if
ier
p
er
f
o
r
m
a
n
ce
was
ev
alu
ate
d
u
s
i
n
g
ten
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
.
T
h
e
o
b
tain
ed
class
if
icatio
n
r
esu
lts
b
ased
o
n
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
a
n
d
F
-
m
ea
s
u
r
e
ar
e
p
r
esen
ted
in
Fig
u
r
e
7
.
ii)
GB
:
t
h
e
cla
s
s
if
icat
io
n
p
er
f
o
r
m
an
ce
o
f
GB
was
ev
alu
ated
u
s
in
g
th
e
s
am
e
ev
alu
atio
n
p
a
r
am
eter
s
.
T
h
e
o
b
tain
ed
ac
c
u
r
ac
y
p
er
f
o
r
m
an
c
e
is
p
r
esen
ted
in
Fig
u
r
e
8.
iii)
XGBo
o
s
t
:
th
e
p
er
f
o
r
m
an
ce
o
f
XGBo
o
s
t
in
cla
s
s
if
y
in
g
m
u
s
cu
lar
atr
o
p
h
y
an
d
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
u
s
in
g
s
E
MG
s
ig
n
als is
p
r
esen
ted
in
Fig
u
r
e
9
.
iv
)
RF
:
th
e
RF
cla
s
s
if
ier
p
er
f
o
r
m
an
ce
was
ev
alu
ated
u
s
in
g
th
e
s
am
e
ex
p
er
im
en
tal
s
etu
p
.
T
h
e
o
b
tain
e
d
ac
cu
r
ac
y
r
esu
lts
ar
e
p
r
esen
ted
in
Fig
u
r
e
1
0
.
Fig
u
r
e
7
.
Acc
u
r
ac
y
p
er
f
o
r
m
an
ce
o
f
SVM
Fig
u
r
e
8
.
Acc
u
r
ac
y
p
er
f
o
r
m
an
ce
o
f
GB
Fig
u
r
e
9
.
Acc
u
r
ac
y
p
er
f
o
r
m
an
ce
o
f
XGBo
o
s
t
Fig
u
r
e
1
0
.
Acc
u
r
ac
y
p
er
f
o
r
m
a
n
ce
o
f
RF
4
.
3
.
Co
m
pa
riso
n o
f
m
a
chine le
a
rning
a
lg
o
rit
hm
s
T
h
e
o
v
er
all
ac
cu
r
a
cy
p
er
f
o
r
m
an
ce
o
f
th
e
m
eth
o
d
s
is
s
h
o
wn
in
Fig
u
r
e
1
1
,
s
u
ch
as
SVM
as
in
Fig
u
r
e
1
1
(
a)
,
GB
as
in
Fig
u
r
e
1
1
(
b
)
,
XGBo
o
s
t
as
in
Fig
u
r
e
1
1
(
c)
,
a
n
d
RF
as
in
Fig
u
r
e
1
1
(
d
)
.
Fig
u
r
es
1
2
to
1
4
s
h
o
w
th
e
p
er
f
o
r
m
an
ce
o
f
v
ar
i
o
u
s
ev
alu
atio
n
p
ar
am
eter
s
in
th
e
p
r
ed
ictio
n
o
f
m
u
s
cu
lar
atr
o
p
h
y
a
n
d
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
d
is
ea
s
es.
T
h
e
ex
p
er
im
en
tal
r
esu
lts
s
h
o
w
th
e
co
m
p
ar
is
o
n
o
f
GB
,
XGBo
o
s
t
,
SVM
,
an
d
R
F
class
if
ier
s
an
d
ev
alu
ate
th
e
p
er
f
o
r
m
an
ce
o
n
t
h
e
b
ases
o
f
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all
,
an
d
F
-
m
ea
s
u
r
e.
I
n
ter
m
s
o
f
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F
-
m
ea
s
u
r
e
f
o
r
th
e
class
if
icatio
n
o
f
m
u
s
cu
lar
atr
o
p
h
y
an
d
m
u
s
cu
lar
d
y
s
tr
o
p
h
y
u
s
in
g
s
E
MG
s
ig
n
als,
th
e
ex
p
er
im
en
tal
f
in
d
in
g
s
s
h
o
w
th
at
th
e
RF
class
if
ier
p
er
f
o
r
m
s
b
etter
th
an
SVM,
GB
,
an
d
X
GB
o
o
s
t
.
T
h
e
ca
p
ac
ity
o
f
RF
to
m
an
a
g
e
h
ig
h
-
d
im
en
s
io
n
al
d
ata
an
d
s
u
cc
ess
f
u
lly
ca
p
tu
r
e
n
o
n
-
lin
ea
r
c
o
r
r
elatio
n
s
is
r
esp
o
n
s
ib
le
f
o
r
th
is
h
ig
h
er
p
er
f
o
r
m
an
ce
.
T
h
ese
r
esu
lts
alig
n
with
ea
r
lier
r
esear
ch
th
at
h
as
b
ee
n
d
o
cu
m
en
ted
in
liter
atu
r
e.
Fo
r
e
x
am
p
le,
Su
b
asi
et
a
l.
[
1
2
]
s
h
o
we
d
th
at
en
s
em
b
le
-
b
ase
d
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
C
o
mp
a
r
is
o
n
o
f m
a
ch
in
e
lea
r
n
i
n
g
a
lg
o
r
ith
ms to
id
en
tify m
u
s
cu
la
r
a
tr
o
p
h
y
a
n
d
…
(
Ga
n
g
a
B
h
a
va
n
i B
illa
)
3513
m
eth
o
d
s
g
r
ea
tly
in
cr
ea
s
e
class
if
icatio
n
ac
cu
r
ac
y
in
n
eu
r
o
m
u
s
cu
lar
d
is
o
r
d
er
id
en
tific
ati
o
n
b
y
p
r
o
p
o
s
in
g
an
EMG
-
b
ased
class
if
icatio
n
s
y
s
tem
em
p
lo
y
in
g
d
is
cr
ete
wa
v
elet
tr
an
s
f
o
r
m
a
n
d
b
ag
g
in
g
tech
n
iq
u
es.
T
h
e
ef
f
icac
y
o
f
en
s
em
b
le
an
d
tr
ee
-
b
ased
m
o
d
els
was
also
d
em
o
n
s
tr
ated
b
y
Kef
alas
et
a
l.
[
1
5
]
,
wh
o
d
em
o
n
s
tr
ated
th
at
au
to
m
ated
m
ac
h
i
n
e
lear
n
in
g
tech
n
i
q
u
es
p
r
o
d
u
ce
d
e
p
en
d
ab
le
p
er
f
o
r
m
a
n
ce
in
d
if
f
e
r
en
tiatin
g
b
etwe
en
n
o
r
m
al
an
d
p
at
h
o
lo
g
ical
E
M
G
d
ata.
Ad
d
itio
n
ally
,
th
e
au
th
o
r
s
[
1
0
]
,
[
1
4
]
h
a
v
e
r
ep
o
r
ted
g
o
o
d
class
if
icatio
n
ac
cu
r
ac
y
wh
e
n
u
s
in
g
m
ac
h
in
e
lear
n
in
g
ap
p
r
o
ac
h
es lik
e
SVM
an
d
ANN
f
o
r
E
MG
s
ig
n
al
an
a
ly
s
is
.
(
a)
(
b
)
(
c)
(
d
)
Fig
u
r
e
1
1
.
C
o
n
f
u
s
io
n
m
atr
ix
o
f
class
if
icatio
n
r
esu
lts
: (
a)
SV
M,
(
b
)
GB
,
(
c)
XGBo
o
s
t
,
an
d
(
d
)
RF
Fig
u
r
e
1
2
.
Mo
d
el
co
m
p
ar
is
o
n
f
o
r
1
6
0
f
ea
t
u
r
es
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.
15
,
No
.
4
,
Au
g
u
s
t 2
0
2
6
:
3
5
0
5
-
3
5
1
7
3514
Fig
u
r
e
1
3
.
Mo
d
el
co
m
p
ar
is
o
n
f
o
r
1
8
1
f
ea
t
u
r
es
Fig
u
r
e
1
4
.
Mo
d
el
co
m
p
ar
is
o
n
f
o
r
5
f
ea
tu
r
es
Ho
wev
er
,
th
e
cu
r
r
en
t
wo
r
k
e
x
p
an
d
s
th
e
an
aly
s
is
to
s
ev
er
al
cir
cu
m
s
tan
ce
s
,
b
o
o
s
tin
g
co
m
p
lex
ity
an
d
clin
ical
im
p
o
r
tan
ce
,
wh
er
ea
s
p
r
io
r
r
esear
ch
m
o
s
tly
co
n
c
en
tr
ated
o
n
b
in
a
r
y
class
if
icatio
n
p
r
o
b
lem
s
.
T
h
e
r
esu
lts
o
f
Has
s
an
et
a
l.
[
2
0
]
,
wh
o
em
p
h
asized
th
e
r
o
b
u
s
tn
ess
o
f
b
o
o
s
tin
g
alg
o
r
ith
m
s
in
class
if
icatio
n
ta
s
k
s
in
v
o
lv
in
g
co
m
p
licated
d
atas
ets,
ar
e
co
n
s
is
ten
t
with
th
e
p
er
f
o
r
m
an
ce
en
h
a
n
ce
m
en
t
s
ee
n
with
GB
an
d
XGBo
o
s
t
in
th
is
s
tu
d
y
.
RF
,
o
n
th
e
o
th
er
h
an
d
,
s
h
o
wed
s
u
p
er
io
r
g
e
n
er
aliza
tio
n
a
n
d
s
tab
ilit
y
o
v
e
r
v
a
r
io
u
s
f
ea
tu
r
e
s
ets (
1
6
0
,
1
8
1
,
an
d
5
f
ea
tu
r
es),
m
ak
in
g
it m
o
r
e
ap
p
r
o
p
r
iate
f
o
r
th
is
u
s
e
ca
s
e.
T
h
e
s
tu
d
y
was
ca
r
r
ied
o
u
t
u
s
in
g
a
p
u
b
licly
av
ailab
le
UC
I
d
ataset
with
a
s
m
all
n
u
m
b
er
o
f
p
ar
ticip
an
ts
,
d
esp
ite
th
e
s
u
g
g
e
s
ted
f
r
am
ewo
r
k
ac
h
iev
in
g
en
c
o
u
r
ag
in
g
class
if
icatio
n
r
esu
lts
.
T
h
e
d
ataset'
s
s
ize
an
d
c
o
n
tr
o
lled
ac
q
u
is
itio
n
c
o
n
d
itio
n
s
m
ay
n
o
t
ac
cu
r
atel
y
r
ef
lect
th
e
v
ar
iab
ilit
y
f
o
u
n
d
in
ac
tu
al
clin
ical
s
ettin
g
s
,
ev
en
th
o
u
g
h
it
o
f
f
e
r
s
a
h
elp
f
u
l
b
en
c
h
m
ar
k
f
o
r
co
m
p
ar
in
g
m
ac
h
i
n
e
lear
n
in
g
tech
n
iq
u
es.
E
MG
f
ea
tu
r
es
ca
n
b
e
g
r
ea
tly
in
f
l
u
e
n
ce
d
b
y
a
n
u
m
b
er
o
f
f
ac
t
o
r
s
,
in
clu
d
in
g
p
atien
t
ag
e,
d
is
ea
s
e
s
ev
er
ity
,
m
u
s
cle
g
r
o
u
p
v
ar
ian
ce
s
,
elec
tr
o
d
e
p
la
ce
m
en
t
ch
an
g
es,
a
n
d
s
ig
n
al
n
o
is
e.
As
a
r
esu
lt,
th
e
p
u
b
lis
h
ed
f
in
d
in
g
s
o
u
g
h
t
to
b
e
r
eg
ar
d
e
d
as
a
p
r
elim
in
ar
y
v
alid
atio
n
o
f
th
e
s
u
g
g
ested
f
r
am
ewo
r
k
.
T
o
f
u
r
th
er
ex
am
in
e
th
e
m
o
d
el'
s
r
o
b
u
s
tn
ess
,
g
en
er
aliza
b
ilit
y
,
an
d
clin
ical
r
eliab
ilit
y
,
f
u
tu
r
e
r
esear
ch
s
h
o
u
ld
u
s
e
lar
g
er
,
m
o
r
e
v
ar
ied
,
an
d
clin
ically
v
alid
ated
E
MG
d
ata
s
ets g
ath
er
ed
f
r
o
m
s
ev
er
al
h
ea
lth
ca
r
e
f
ac
ilit
ies.
Ad
d
itio
n
ally
,
th
is
s
tu
d
y
h
as
an
ed
g
e
o
v
er
p
r
ev
io
u
s
r
esear
ch
th
at
u
s
ed
s
tatic
d
atasets
b
ec
au
s
e
it
u
s
es
r
ea
l
-
tim
e
E
MG
d
ata.
Acc
o
r
d
i
n
g
to
Nar
d
o
et
a
l.
[
1
1
]
,
m
ac
h
in
e
lear
n
i
n
g
m
o
d
els
n
ee
d
to
b
e
r
esil
ien
t
to
th
e
co
n
s
id
er
ab
le
v
ar
iab
ilit
y
i
n
E
MG
s
ig
n
als.
T
h
is
v
ar
ia
b
ilit
y
is
h
an
d
led
well
b
y
th
e
s
u
g
g
e
s
ted
m
eth
o
d
,
wh
ich
im
p
r
o
v
es
d
iag
n
o
s
tic
p
er
f
o
r
m
an
ce
.
Ov
er
all,
th
e
r
esu
lts
s
u
p
p
o
r
t
cu
r
r
e
n
t
r
esear
ch
tr
en
d
s
an
d
s
h
o
w
th
at
en
s
em
b
le
-
b
ased
m
ac
h
in
e
lea
r
n
in
g
m
o
d
els
—
RF
in
p
ar
ticu
lar
—
ar
e
h
ig
h
ly
ef
f
ec
tiv
e
in
class
if
y
in
g
E
MG
Evaluation Warning : The document was created with Spire.PDF for Python.