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
.
6
,
Decem
b
er
20
25
,
p
p
.
5
9
5
4
~
5
9
6
7
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
15
i
6
.
pp
5
9
5
4
-
5
9
6
7
5954
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
M
a
chine learning
-
ba
sed cla
ss
ificati
o
n of loca
l muscle
f
a
tigue
using
elect
ro
my
o
g
ra
phy
sig
na
ls for enha
nced
rehab
ilitatio
n
o
utcom
es
Z
ha
nel
B
a
ig
a
ra
y
ev
a
1,
2
,
Ass
iy
a
B
o
lt
a
bo
y
ev
a
1,
2
,
B
a
g
l
a
n Im
a
nb
ek
1
,
K
a
s
s
y
m
bek O
zhik
en
o
v
2
,
Nurg
ul K
a
ry
m
s
s
a
k
o
v
a
1,
2
,
Ro
za
B
eisem
bek
o
v
a
2
1
D
e
p
a
r
t
me
n
t
o
f
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
B
i
g
D
a
t
a
,
F
a
c
u
l
t
y
o
f
I
n
f
o
r
ma
t
i
o
n
Te
c
h
n
o
l
o
g
y
,
A
l
-
F
a
r
a
b
i
K
a
z
a
k
h
N
a
t
i
o
n
a
l
U
n
i
v
e
r
si
t
y
,
A
l
mat
y
,
K
a
z
a
k
h
s
t
a
n
2
D
e
p
a
r
t
me
n
t
o
f
R
o
b
o
t
i
c
s a
n
d
T
e
c
h
n
i
c
a
l
M
e
a
n
s
o
f
A
u
t
o
m
a
t
i
o
n
,
I
n
st
i
t
u
t
e
o
f
A
u
t
o
ma
t
i
o
n
a
n
d
I
n
f
o
r
ma
t
i
o
n
Te
c
h
n
o
l
o
g
i
e
s
,
S
a
t
b
a
y
e
v
U
n
i
v
e
r
si
t
y
,
A
l
mat
y
,
K
a
z
a
k
h
st
a
n
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
y
2
6
,
2
0
2
5
R
ev
is
ed
Au
g
1
1
,
2
0
2
5
Acc
ep
ted
Sep
1
5
,
2
0
2
5
M
u
sc
le
fa
ti
g
u
e
is
a
k
e
y
fa
c
to
r
a
f
fe
c
ti
n
g
re
h
a
b
il
it
a
ti
o
n
p
ro
g
re
ss
,
s
a
fe
ty
,
a
n
d
p
a
ti
e
n
t
e
n
g
a
g
e
m
e
n
t.
Ac
c
u
ra
te
d
e
tec
ti
o
n
o
f
fa
ti
g
u
e
d
u
ri
n
g
p
h
y
sic
a
l
a
c
ti
v
it
y
re
m
a
in
s
a
c
h
a
ll
e
n
g
e
,
p
a
rti
c
u
larl
y
in
c
li
n
ica
l
a
n
d
re
m
o
te
se
tt
in
g
s.
Th
is
st
u
d
y
p
re
se
n
ts
th
e
d
e
v
e
l
o
p
m
e
n
t
o
f
a
n
i
n
tern
e
t
of
t
h
in
g
s
-
b
a
se
d
s
y
ste
m
fo
r
c
las
sify
in
g
lo
c
a
l
m
u
sc
le
fa
ti
g
u
e
u
si
n
g
su
rfa
c
e
e
lec
tro
m
y
o
g
ra
p
h
y
(E
M
G
)
sig
n
a
ls
a
n
d
m
a
c
h
in
e
lea
rn
i
n
g
.
A
we
a
ra
b
le
d
e
v
ice
wa
s
u
se
d
to
c
o
ll
e
c
t
re
a
l
-
ti
m
e
EM
G
d
a
ta
a
n
d
su
b
jec
ti
v
e
f
a
ti
g
u
e
ra
ti
n
g
s
fro
m
1
0
h
e
a
lt
h
y
p
a
rti
c
ip
a
n
ts
d
u
ri
n
g
su
sta
i
n
e
d
iso
m
e
tri
c
g
ri
p
e
x
e
rc
ise
s.
F
e
a
tu
re
e
x
trac
ti
o
n
wa
s
p
e
rfo
rm
e
d
on
-
d
e
v
ice
,
a
n
d
th
e
d
a
ta
we
re
tran
sm
it
ted
wire
les
sly
fo
r
a
n
a
ly
sis.
M
a
c
h
in
e
lea
rn
in
g
m
o
d
e
ls
i
n
c
lu
d
in
g
l
o
g
ist
ic
re
g
re
ss
io
n
,
d
e
c
isio
n
tree
(DT
),
ra
n
d
o
m
fo
re
st,
a
n
d
e
x
trem
e
g
ra
d
ien
t
b
o
o
stin
g
(XG
Bo
o
st)
we
re
train
e
d
t
o
c
las
sify
fa
ti
g
u
e
sta
tes
.
T
h
e
DT
m
o
d
e
l
a
c
h
i
e
v
e
d
t
h
e
h
i
g
h
e
st
a
c
c
u
ra
c
y
o
f
9
0
.
7
%
,
wit
h
a
p
re
c
isio
n
o
f
9
0
.
7
%
a
n
d
a
re
c
a
ll
o
f
9
0
.
9
%
.
S
HA
P
a
n
a
l
y
sis
re
v
e
a
led
ti
m
e
u
n
d
e
r
l
o
a
d
,
sm
o
k
in
g
,
a
n
d
a
lco
h
o
l
u
se
a
s
th
e
m
o
st
in
flu
e
n
ti
a
l
f
a
c
to
rs
in
fa
ti
g
u
e
c
las
sifica
ti
o
n
.
T
h
e
se
re
su
lt
s
sh
o
w
t
h
a
t
we
a
ra
b
le
EM
G
d
e
v
ice
s
c
o
m
b
in
e
d
wi
th
sm
a
rt
a
lg
o
rit
h
m
s
a
re
e
ffe
c
ti
v
e
fo
r
re
a
l
-
ti
m
e
fa
ti
g
u
e
m
o
n
it
o
r
in
g
d
u
ri
n
g
re
h
a
b
il
it
a
ti
o
n
.
K
ey
w
o
r
d
s
:
E
lectr
o
m
y
o
g
r
ap
h
y
s
ig
n
als
I
n
ter
n
et
o
f
th
in
g
s
Ma
ch
in
e
lear
n
in
g
Mu
s
cle
f
atig
u
e
R
eh
ab
ilit
atio
n
m
o
n
ito
r
in
g
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Kass
y
m
b
ek
Ozh
ik
en
o
v
Dep
ar
tm
en
t
o
f
R
o
b
o
tics
an
d
T
ec
h
n
ical
Me
an
s
o
f
A
u
to
m
at
io
n
,
I
n
s
titu
te
o
f
Au
to
m
atio
n
an
d
I
n
f
o
r
m
atio
n
T
ec
h
n
o
lo
g
ies,
Satb
ay
ev
Un
iv
e
r
s
ity
2
2
Satp
ae
v
Stre
et,
0
5
0
0
1
3
,
Al
m
aty
,
Kaz
ak
h
s
tan
E
m
ail: k
.
o
zh
ik
e
n
o
v
@
s
atb
ay
ev
.
u
n
iv
er
s
ity
1.
I
NT
RO
D
UCT
I
O
N
Mu
s
cle
f
atig
u
e
is
a
cr
itical
ar
ea
o
f
r
esear
ch
in
s
p
o
r
ts
m
ed
ic
in
e
an
d
o
cc
u
p
atio
n
al
h
ea
lth
[
1
]
.
I
t
is
an
ex
er
cise
-
in
d
u
ce
d
d
ec
r
ea
s
e
in
t
h
e
ab
ilit
y
to
p
r
o
d
u
ce
f
o
r
ce
[
2
]
.
Me
asu
r
in
g
f
atig
u
e
an
d
f
atig
a
b
ilit
y
is
cr
itical
to
u
n
d
er
s
tan
d
i
n
g
t
h
e
im
p
ac
t
o
f
th
e
d
is
ea
s
e,
d
eter
m
i
n
in
g
o
p
tim
al
lev
els
o
f
p
h
y
s
ical
ac
tiv
ity
,
tailo
r
in
g
r
eh
ab
ilit
atio
n
p
r
o
g
r
am
s
,
an
d
a
s
s
es
s
in
g
th
e
ef
f
ec
tiv
en
ess
o
f
th
er
ap
eu
tic
in
ter
v
e
n
tio
n
s
,
wh
ic
h
is
a
ce
n
tr
al
asp
ec
t
o
f
r
eh
a
b
ilit
atio
n
[
3
]
.
E
f
f
ec
tiv
e
f
atig
u
e
d
iag
n
o
s
is
ca
n
p
r
ev
en
t
m
u
s
cle
d
am
ag
e,
th
e
r
eb
y
in
cr
ea
s
in
g
th
e
s
af
ety
o
f
r
eh
ab
ilit
atio
n
ex
er
cises
.
I
t
h
as
im
p
o
r
tan
t
p
r
ac
tical
s
ig
n
if
ican
ce
f
o
r
th
e
ass
ess
m
en
t
o
f
m
u
s
cle
f
u
n
ctio
n
in
th
e
f
ield
o
f
r
e
h
ab
ilit
atio
n
m
e
d
icin
e
an
d
th
e
d
eter
m
in
atio
n
o
f
f
atig
u
e
in
r
e
h
ab
ilit
atio
n
s
cien
ce
[
4
]
.
Fatig
u
e
is
a
m
u
ltid
im
en
s
io
n
al
p
h
en
o
m
en
o
n
t
h
at
d
e
g
r
ad
e
s
p
h
y
s
ical
an
d
c
o
g
n
itiv
e
p
er
f
o
r
m
an
ce
,
u
ltima
tely
r
ed
u
cin
g
r
eh
a
b
ilit
atio
n
ef
f
icien
c
y
an
d
t
h
r
ea
ten
in
g
p
atien
t
s
af
ety
.
I
n
clin
ical
s
ettin
g
s
,
it
p
r
esen
ts
n
o
t
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
Ma
ch
in
e
lea
r
n
in
g
-
b
a
s
ed
cla
s
s
ifica
tio
n
o
f lo
ca
l m
u
s
cle
fa
tig
u
e
…
(
Zh
a
n
el
B
a
ig
a
r
a
ye
v
a
)
5955
o
n
ly
as
a
s
u
b
jectiv
e
lo
s
s
o
f
e
n
er
g
y
b
u
t
also
as
an
o
b
jectiv
e
f
all
in
n
eu
r
o
m
u
s
cu
lar
c
o
o
r
d
in
atio
n
a
n
d
m
o
to
r
f
u
n
ctio
n
[
5
]
.
C
h
r
o
n
ic
f
atig
u
e
i
n
p
o
s
t
-
s
tr
o
k
e
s
y
n
d
r
o
m
e
o
r
Par
k
in
s
o
n
’
s
d
is
ea
s
e
in
ten
s
if
ies
m
o
to
r
wea
k
n
ess
an
d
m
ar
k
ed
ly
in
cr
ea
s
es
f
all
r
is
k
,
jeo
p
ar
d
izin
g
s
af
ety
d
u
r
i
n
g
d
aily
ac
tiv
ities
an
d
th
er
ap
y
s
ess
io
n
s
[
6
]
.
B
io
m
ec
h
an
ically
,
f
atig
u
e
r
e
d
u
ce
s
n
eu
r
o
m
u
s
cu
lar
s
y
n
er
g
y
an
d
p
r
o
p
r
i
o
ce
p
tiv
e
ac
u
ity
,
cr
ea
tin
g
u
n
s
tab
le
m
o
v
em
en
t
p
atter
n
s
th
at
p
r
ed
is
p
o
s
e
in
d
iv
id
u
als
to
in
ju
r
ies
s
u
ch
as
an
ter
io
r
cr
u
ciate
lig
am
e
n
t
r
u
p
tu
r
e
[
5
]
,
[
7
]
.
C
o
g
n
itiv
ely
,
it
s
lo
ws
in
f
o
r
m
atio
n
p
r
o
ce
s
s
in
g
,
len
g
th
e
n
s
r
ea
ctio
n
tim
es,
an
d
u
n
d
er
m
in
es
d
ec
is
io
n
-
m
ak
in
g
d
u
r
in
g
u
n
ex
p
ec
ted
ev
e
n
ts
[
8
]
,
[
9
]
.
Per
s
is
ten
t
f
atig
u
e
ca
n
ex
te
n
d
r
e
h
ab
ilit
atio
n
tim
elin
es
[
1
0
]
an
d
er
o
d
e
p
atien
t
m
o
tiv
atio
n
[
1
1
]
.
R
o
u
tin
e
m
o
n
ito
r
in
g
is
th
er
e
f
o
r
e
ess
en
tial
f
o
r
tim
ely
a
d
ju
s
tm
en
t
o
f
in
t
er
v
en
tio
n
s
an
d
f
o
r
s
af
eg
u
ar
d
in
g
f
u
n
ctio
n
al
o
u
tco
m
es
[
9
]
.
Per
f
o
r
m
an
ce
f
atig
u
e
is
a
m
aj
o
r
ca
u
s
e
o
f
wo
r
k
-
r
elate
d
m
u
s
cu
lo
s
k
eleta
l
d
is
o
r
d
er
s
a
n
d
u
n
d
er
s
tan
d
in
g
its
im
p
ac
t d
u
r
in
g
m
an
u
al
h
an
d
lin
g
(
MH
T
)
is
cr
itical
to
p
r
ev
e
n
tin
g
s
u
ch
p
r
o
b
lem
s
[
1
2
]
.
Pra
ctica
l m
ea
s
u
r
em
en
t
r
em
ain
s
ch
allen
g
in
g
d
u
e
to
lim
itatio
n
s
in
eq
u
ip
m
en
t
s
ize
an
d
co
s
t.
I
n
v
asiv
e
an
d
ca
r
d
io
p
u
lm
o
n
ar
y
m
eth
o
d
s
ar
e
o
f
ten
in
c
o
n
v
e
n
ien
t
an
d
n
o
t
s
u
itab
le
f
o
r
th
e
d
aily
tr
ain
in
g
o
f
p
atien
ts
o
r
ath
letes
[
1
3
]
.
Mu
s
cle
f
atig
u
e
m
o
n
ito
r
in
g
is
in
d
em
a
n
d
f
o
r
d
aily
ac
tiv
ities
an
d
ex
er
cise,
in
clu
d
in
g
in
j
u
r
y
p
r
ev
e
n
tio
n
in
a
th
letes.
A
wea
r
ab
le
s
y
s
tem
is
n
ee
d
ed
t
h
at
n
o
t
o
n
l
y
n
o
tifie
s
th
e
u
s
er
o
f
th
e
o
n
s
et
o
f
f
atig
u
e
b
u
t
also
p
r
o
v
id
e
s
in
f
o
r
m
atio
n
o
n
a
r
elax
ed
s
tate
to
m
in
im
ize
th
e
r
is
k
o
f
in
ju
r
y
.
T
h
e
r
a
p
id
p
r
o
life
r
atio
n
o
f
w
ea
r
ab
le
s
en
s
o
r
s
an
d
I
n
ter
n
et
of
th
in
g
s
(
I
o
T
)
d
ev
ices
h
as
s
h
if
ted
th
e
p
ar
ad
ig
m
to
war
d
r
ea
l
-
tim
e
p
h
y
s
io
lo
g
ical
m
o
n
ito
r
in
g
,
m
a
k
in
g
f
atig
u
e
d
etec
tio
n
a
cr
it
ical
ap
p
licatio
n
in
tr
an
s
p
o
r
t,
in
d
u
s
tr
ial
s
af
ety
,
s
p
o
r
t
an
d
h
ea
lth
ca
r
e
[
1
4
]
.
Ad
v
an
ce
s
in
m
in
iatu
r
ized
b
i
o
s
en
s
o
r
s
,
lo
w
-
laten
cy
wir
eless
p
r
o
to
co
ls
an
d
ed
g
e
–
clo
u
d
in
f
r
astru
ctu
r
es
n
o
w
e
n
ab
le
co
n
tin
u
o
u
s
d
ata
ca
p
tu
r
e
an
d
p
r
o
ce
s
s
in
g
,
allo
win
g
s
y
s
tem
s
to
d
etec
t
s
u
b
tle
p
h
y
s
io
lo
g
ical
ch
a
n
g
es
[
1
5
]
.
Sev
er
al
r
ec
en
t
r
ev
iews
s
u
m
m
ar
is
e
I
o
T
-
b
ased
f
atig
u
e
m
o
n
ito
r
in
g
–
esp
ec
ially
f
o
r
d
r
iv
er
d
r
o
wsi
n
ess
,
wo
r
k
er
s
af
ety
an
d
h
ea
lth
m
an
a
g
e
m
en
t
–
y
et
h
i
g
h
lig
h
t
p
er
s
is
ten
t
ch
allen
g
es
in
r
ea
l
-
tim
e
p
r
o
ce
s
s
in
g
,
p
r
iv
ac
y
an
d
alg
o
r
ith
m
ad
a
p
tab
ilit
y
ac
r
o
s
s
h
eter
o
g
en
eo
u
s
s
ettin
g
s
[
1
4
]
.
E
m
er
g
in
g
tech
n
o
l
o
g
ies
s
u
ch
as
th
e
I
o
T
an
d
s
en
s
o
r
s
m
ak
e
it
p
o
s
s
ib
le
to
r
em
o
tely
m
o
n
ito
r
an
d
v
is
u
alize
p
h
y
s
ical
d
ata
[
1
6
]
.
Dete
ctio
n
o
f
m
u
s
cle
f
atig
u
e
u
s
in
g
E
MG
is
v
alu
ab
le
in
a
v
a
r
iety
o
f
ap
p
licatio
n
s
s
u
ch
as
s
p
o
r
ts
p
er
f
o
r
m
a
n
ce
,
r
eh
ab
ilit
atio
n
,
o
cc
u
p
atio
n
al
h
e
alth
an
d
d
r
iv
e
r
s
af
ety
.
I
t
h
elp
s
to
p
r
ev
en
t
f
atig
u
e
-
r
elate
d
in
ju
r
ies an
d
o
p
tim
ize
p
er
f
o
r
m
a
n
ce
.
Su
r
f
ac
e
elec
tr
o
m
y
o
g
r
a
p
h
y
(
E
MG
)
ca
n
b
e
u
s
ed
in
th
e
r
eh
ab
ilit
atio
n
o
f
m
u
s
cu
lo
s
k
eleta
l
d
is
o
r
d
er
s
t
o
ass
ess
th
e
co
o
r
d
in
atio
n
o
f
m
u
s
cles
th
at
s
tab
il
ize
a
jo
in
t
o
r
as
an
o
b
jectiv
e
to
o
l
to
ass
ess
m
u
s
cle
f
atig
u
e
[
1
7
]
.
Ma
ch
in
e
lear
n
in
g
m
o
d
els
u
s
i
n
g
E
MG
s
ig
n
al
p
r
o
p
e
r
ties
in
t
im
e
an
d
f
r
e
q
u
en
c
y
d
o
m
ain
s
ar
e
u
s
ed
to
d
if
f
er
en
ti
ate
m
u
s
cle
f
atig
u
e
lev
els
[
1
8
]
.
T
h
e
p
er
f
o
r
m
an
ce
o
f
m
o
d
els
m
ay
d
ep
en
d
o
n
th
e
th
r
esh
o
ld
s
u
s
ed
to
d
ef
in
e
f
atig
u
e
in
t
h
e
r
e
f
er
en
ce
m
ea
s
u
r
em
e
n
ts
,
an
d
th
e
c
h
o
ice
o
f
th
ese
th
r
esh
o
ld
s
in
f
lu
en
ce
s
wh
ich
p
r
e
d
icto
r
s
ar
e
c
o
n
s
id
er
ed
r
elev
an
t
[
1
5
]
.
Mo
s
t
p
r
ev
io
u
s
r
esear
ch
h
as
f
o
cu
s
ed
o
n
m
u
s
cle
f
atig
u
e
d
u
r
in
g
s
h
o
r
t
b
u
r
s
ts
o
f
m
ax
im
al
ef
f
o
r
t
o
r
alter
n
atin
g
co
n
tr
ac
tio
n
-
r
elax
atio
n
task
s
,
wh
ich
d
o
n
’
t
f
u
lly
ca
p
t
u
r
e
th
e
s
u
s
tain
ed
m
u
s
cle
ac
tiv
ity
co
m
m
o
n
in
ev
er
y
d
ay
r
eh
ab
ex
e
r
cises
.
Als
o
,
m
an
y
s
tu
d
ies
r
ely
o
n
ly
o
n
o
b
jectiv
e
E
MG
d
ata
an
d
d
o
n
’
t c
o
n
s
id
er
h
o
w
p
eo
p
le
ac
tu
ally
f
ee
l
d
u
r
in
g
th
e
task
.
T
h
is
s
tu
d
y
ad
d
r
ess
es
th
e
cr
itical
ch
allen
g
e
o
f
ac
cu
r
ately
d
etec
tin
g
m
u
s
cle
f
atig
u
e
—
a
co
m
p
lex
,
m
u
ltifa
cto
r
ial
p
h
en
o
m
en
o
n
th
at
im
p
ac
ts
r
eh
ab
ilit
atio
n
s
af
ety
an
d
ef
f
ec
tiv
e
n
ess
—
b
y
in
teg
r
atin
g
s
u
r
f
ac
e
E
MG
with
m
ac
h
in
e
lear
n
in
g
i
n
a
n
o
v
el
I
o
T
-
b
ased
wea
r
ab
le
s
y
s
tem
.
W
h
ile
p
r
io
r
wo
r
k
s
h
av
e
p
r
im
ar
ily
r
elied
o
n
b
r
ief
,
h
ig
h
-
in
ten
s
ity
co
n
tr
ac
ti
o
n
s
an
d
o
b
jectiv
e
E
MG
f
ea
tu
r
es
alo
n
e,
th
is
r
esear
ch
u
n
iq
u
ely
co
m
b
in
es
r
ea
l
-
tim
e
E
MG
s
ig
n
al
p
r
o
ce
s
s
in
g
with
s
u
b
jectiv
e
f
atig
u
e
r
a
tin
g
s
d
u
r
in
g
s
u
s
tain
ed
s
u
b
m
ax
im
al
g
r
ip
task
s
,
en
h
an
cin
g
ec
o
lo
g
ical
v
alid
ity
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
tem
,
Qim
y
l,
in
tr
o
d
u
ce
s
m
eth
o
d
o
lo
g
ica
l
ad
v
an
ce
m
en
t
b
y
en
ab
lin
g
o
n
-
d
e
v
ice
ex
tr
ac
tio
n
o
f
s
u
r
f
ac
e
E
MG
f
ea
tu
r
es
an
d
lo
w
-
laten
cy
wir
eless
d
ata
tr
an
s
m
is
s
io
n
,
th
er
eb
y
ad
d
r
ess
in
g
k
ey
lim
itatio
n
s
o
f
ex
is
tin
g
f
atig
u
e
m
o
n
ito
r
in
g
s
y
s
tem
s
th
at
d
ep
en
d
o
n
m
u
ltip
le
ex
ter
n
al
s
en
s
o
r
s
an
d
p
o
s
t
-
h
o
c
d
ata
an
aly
s
is
.
I
ts
ar
c
h
itectu
r
e
ac
c
o
m
m
o
d
ates
th
e
co
n
s
tr
ain
ts
o
f
m
o
b
ile
an
d
em
b
e
d
d
ed
en
v
ir
o
n
m
en
ts
,
with
a
co
m
p
ar
ativ
e
ev
alu
atio
n
o
f
class
if
ier
s
th
at
s
u
p
p
o
r
ts
in
tellig
en
t
d
ep
l
o
y
m
en
t
i
n
ad
a
p
tiv
e
r
eh
ab
ilit
atio
n
co
n
tex
ts
.
B
y
ali
g
n
in
g
p
h
y
s
io
lo
g
ical
s
ig
n
als
w
ith
u
s
er
-
r
ep
o
r
ted
f
atig
u
e
in
r
e
al
-
tim
e,
th
is
wo
r
k
co
n
tr
ib
u
tes
a
s
ca
lab
le,
p
r
ac
tical
s
o
lu
tio
n
to
lo
n
g
s
tan
d
in
g
ch
allen
g
es
in
f
atig
u
e
class
if
i
ca
tio
n
,
an
d
s
ets
a
f
o
u
n
d
atio
n
f
o
r
f
u
t
u
r
e
teler
e
h
a
b
ilit
atio
n
an
d
p
er
s
o
n
alize
d
h
e
alth
m
o
n
ito
r
in
g
s
y
s
tem
s
.
T
h
e
r
em
ain
d
er
o
f
t
h
e
ar
ticle
is
o
r
g
an
ize
d
as
f
o
llo
ws.
Sectio
n
1
p
r
esen
ts
th
e
in
tr
o
d
u
ctio
n
an
d
p
r
o
b
lem
s
tatem
en
t.
Sectio
n
2
p
r
o
v
i
d
es
a
cr
itical
liter
atu
r
e
r
ev
iew.
Sectio
n
3
d
etails th
e
ex
p
e
r
im
en
tal
m
eth
o
d
o
l
o
g
y
a
n
d
th
e
Qim
y
l d
ev
ice
ar
ch
itectu
r
e.
Sectio
n
4
r
ep
o
r
ts
an
d
d
is
cu
s
s
es
th
e
r
esu
lts
.
Sectio
n
5
co
n
clu
d
es
th
e
p
ap
er
a
n
d
o
u
tlin
es
d
ir
ec
tio
n
s
f
o
r
f
u
tu
r
e
r
esear
ch
.
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
Mu
s
cle
f
atig
u
e
is
t
h
e
g
r
ad
u
al
d
ec
lin
e
in
a
m
u
s
cle’
s
ab
ilit
y
to
g
en
er
ate
f
o
r
c
e,
wh
ic
h
is
co
m
m
o
n
l
y
o
b
s
er
v
ed
in
v
ar
io
u
s
co
n
d
itio
n
s
s
u
ch
as
m
u
ltip
le
s
cler
o
s
is
,
s
tr
o
k
e,
ch
r
o
n
ic
in
s
o
m
n
ia,
a
n
d
m
u
s
cu
lo
s
k
eleta
l
in
ju
r
ies
[
1
9
]
,
[
2
0
]
.
Du
e
to
th
e
s
u
b
jectiv
e
n
atu
r
e
o
f
f
atig
u
e
,
its
p
r
ec
is
e
m
ea
s
u
r
em
en
t
an
d
p
r
ed
ictio
n
r
em
ain
ch
allen
g
in
g
task
s
[
2
1
]
.
I
n
a
d
eq
u
ate
f
atig
u
e
m
an
a
g
em
en
t
ca
n
lead
to
in
ju
r
ies
an
d
p
o
o
r
er
r
e
h
ab
ilit
atio
n
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
.
6
,
Decem
b
e
r
20
25
:
5
9
5
4
-
5
9
6
7
5956
o
u
tco
m
es
[
2
2
]
.
Fatig
u
e
is
ty
p
ically
ass
es
s
ed
u
s
in
g
E
MG
s
i
g
n
als,
s
o
m
etim
es
co
m
b
in
ed
with
ac
ce
ler
o
m
eter
d
ata
an
d
o
th
er
p
h
y
s
io
lo
g
ical
i
n
d
icato
r
s
,
s
u
ch
as E
C
G
an
d
h
ea
r
t r
ate.
So
m
e
s
tu
d
ies u
tili
ze
s
u
r
f
ac
e
E
MG
d
u
r
in
g
r
o
b
o
tic
m
o
v
e
m
en
ts
to
ev
alu
at
e
th
e
im
p
ac
t
o
f
f
atig
u
e
o
n
jo
i
n
t
p
o
s
itio
n
s
en
s
e
[
2
3
]
.
I
n
r
eh
ab
ilit
atio
n
s
y
s
tem
s
,
f
atig
u
e
d
etec
tio
n
ac
ts
as a
s
ig
n
al
to
m
o
d
if
y
o
r
s
to
p
ex
er
cises
in
o
r
d
er
to
p
r
o
tect
th
e
p
atien
t
[
2
4
]
.
Ad
a
p
tiv
e
lo
ad
co
n
tr
o
l
m
eth
o
d
s
b
ased
o
n
o
n
lin
e
f
atig
u
e
d
etec
tio
n
ar
e
b
ei
n
g
d
ev
elo
p
e
d
to
o
p
tim
ize
tr
a
in
in
g
d
u
r
atio
n
an
d
in
ten
s
ity
.
A
f
atig
u
e
ass
es
s
m
en
t
m
o
d
el
f
o
r
v
ir
tu
al
r
ea
lity
h
as
also
b
ee
n
d
ev
elo
p
e
d
,
allo
win
g
f
o
r
in
d
iv
id
u
alize
d
f
atig
u
e
m
ea
s
u
r
em
e
n
t
with
o
u
t
ad
d
itio
n
al
s
en
s
o
r
s
[
2
5
]
.
I
n
ad
d
itio
n
,
d
ig
ital
twin
s
co
m
b
i
n
ed
with
m
ac
h
i
n
e
lear
n
in
g
m
eth
o
d
s
ar
e
u
s
ed
f
o
r
r
ea
l
-
tim
e
v
is
u
aliza
tio
n
a
n
d
f
o
r
ec
asti
n
g
,
wh
ich
ca
n
b
e
ap
p
lied
in
th
e
m
an
ag
em
en
t
o
f
r
e
h
ab
ilit
atio
n
p
r
o
ce
s
s
es
[
2
6
]
.
E
MG
p
lay
s
a
k
ey
r
o
le
in
ass
ess
in
g
m
u
s
cle
ac
tiv
ity
a
n
d
p
ati
en
t
co
n
d
itio
n
d
u
r
in
g
r
eh
a
b
ilit
atio
n
[
2
7
]
.
Su
r
f
ac
e
E
MG
allo
ws
th
e
d
etec
tio
n
o
f
m
u
s
cle
ac
tiv
ity
d
u
r
in
g
m
o
v
em
e
n
ts
an
d
ex
er
tio
n
,
w
h
ich
is
im
p
o
r
tan
t
f
o
r
d
iag
n
o
s
is
an
d
ev
alu
atio
n
o
f
f
u
n
ctio
n
al
ch
a
n
g
es
[
2
8
]
.
E
MG
is
wid
ely
u
s
ed
in
o
r
th
o
p
ed
ics,
n
eu
r
o
r
eh
ab
ilit
atio
n
,
s
p
o
r
ts
m
ed
icin
e,
an
d
th
e
s
tu
d
y
o
f
ag
in
g
p
r
o
ce
s
s
es
[
2
9
]
.
E
M
G
is
ap
p
lied
in
g
ait
r
eh
ab
ilit
at
io
n
,
co
n
tr
o
l
o
f
k
n
ee
ex
o
s
k
eleto
n
s
an
d
p
r
o
s
th
eses
,
an
d
in
th
e
ass
ess
m
en
t
o
f
m
o
t
io
n
p
ar
am
eter
s
(
s
u
c
h
as
s
p
ee
d
an
d
s
lo
p
e
)
d
u
r
in
g
r
o
b
o
tic
ass
is
tan
ce
.
E
MG
-
b
ased
b
io
f
ee
d
b
ac
k
co
n
tr
i
b
u
tes
to
i
m
p
r
o
v
i
n
g
s
tr
en
g
th
an
d
b
alan
c
e
in
s
tr
o
k
e
p
atien
ts
an
d
th
e
eld
er
l
y
[
3
0
]
.
C
o
m
b
in
i
n
g
E
MG
with
elec
tr
ical
s
tim
u
latio
n
tr
ig
g
er
ed
b
y
E
MG
s
ig
n
als
en
h
an
ce
s
h
an
d
f
u
n
ctio
n
in
ch
r
o
n
ic
s
tr
o
k
e
p
at
ien
ts
.
E
MG
an
aly
s
is
h
elp
s
d
et
ec
t
m
o
to
r
im
p
air
m
en
ts
,
ass
ess
co
o
r
d
in
atio
n
,
an
d
id
en
tify
s
p
asti
city
[
3
1
]
.
T
h
e
t
h
eo
r
y
o
f
m
u
s
cle
s
y
n
er
g
ies,
b
ased
o
n
E
MG
s
ig
n
al
d
ec
o
m
p
o
s
itio
n
,
is
u
s
ed
f
o
r
q
u
an
titativ
e
ass
ess
m
en
t
o
f
m
o
to
r
co
n
t
r
o
l
d
is
o
r
d
e
r
s
[
3
2
]
.
T
o
im
p
r
o
v
e
d
iag
n
o
s
tic
ac
cu
r
ac
y
an
d
co
n
tr
o
l
o
f
ass
is
tiv
e
d
ev
ices,
m
u
ltis
en
s
o
r
y
in
teg
r
atio
n
m
eth
o
d
s
a
r
e
b
ein
g
d
ev
elo
p
e
d
,
c
o
m
b
in
i
n
g
E
MG
with
o
th
e
r
b
io
m
ed
ical
s
ig
n
als
[
2
7
]
.
E
MG
p
r
o
ce
s
s
in
g
in
clu
d
es
n
o
is
e
s
u
p
p
r
ess
io
n
,
am
p
lific
atio
n
,
f
ilter
in
g
,
d
ec
o
m
p
o
s
itio
n
,
a
n
d
f
ea
tu
r
e
ex
tr
ac
tio
n
,
wh
ich
allo
ws cla
s
s
if
icatio
n
an
d
an
aly
s
is
o
f
s
ig
n
als ac
r
o
s
s
d
if
f
er
en
t f
ield
s
[
3
3
]
.
Ma
ch
in
e
lear
n
in
g
is
u
s
ed
f
o
r
t
h
e
au
to
m
atic
class
if
icatio
n
o
f
m
u
s
cle
f
atig
u
e
s
tates
b
ased
o
n
s
u
r
f
ac
e
E
MG
[
3
4
]
.
T
h
e
class
ical
an
d
m
o
s
t
wid
ely
u
s
ed
alg
o
r
ith
m
is
th
e
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
,
wh
ich
ac
h
iev
es
an
a
cc
u
r
ac
y
o
f
ar
o
u
n
d
9
0
%
in
d
is
tin
g
u
is
h
in
g
f
atig
u
e
d
an
d
n
o
n
-
f
atig
u
e
d
s
tates
[
2
2
]
.
T
h
is
in
v
o
lv
es
ex
tr
ac
tin
g
f
ea
tu
r
es
f
r
o
m
th
e
tim
e
an
d
f
r
eq
u
e
n
cy
d
o
m
ai
n
s
[
3
4
]
.
Fatig
u
e
class
if
icatio
n
d
u
r
in
g
d
y
n
am
ic
m
o
v
e
m
en
ts
,
s
u
ch
as
th
e
s
it
-
to
-
s
tan
d
test
,
h
as
b
ee
n
im
p
lem
en
ted
u
s
in
g
E
MG
an
d
ac
ce
ler
o
m
eter
d
a
ta
[
3
5
]
.
I
n
ad
d
itio
n
t
o
SVM,
r
ec
u
r
r
e
n
t
n
eu
r
al
n
etwo
r
k
s
s
u
ch
as
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
(
L
STM
)
a
r
e
u
s
e
d
to
a
cc
o
u
n
t
f
o
r
tem
p
o
r
al
d
ep
en
d
en
cies
in
E
MG
s
ig
n
als,
im
p
r
o
v
in
g
class
if
icatio
n
p
er
f
o
r
m
an
ce
[
3
4
]
.
Ma
ch
in
e
lear
n
in
g
is
also
u
s
ed
to
an
aly
ze
co
m
p
lex
n
o
n
lin
ea
r
ch
a
r
ac
ter
is
tics
o
f
E
MG
[
3
6
]
,
[
3
7
]
.
I
n
p
r
ac
tica
l
ap
p
licatio
n
s
,
f
atig
u
e
class
if
icatio
n
en
ab
les
th
e
cr
ea
tio
n
o
f
ad
ap
tiv
e
r
eh
ab
ilit
a
tio
n
s
y
s
tem
s
th
at
au
to
m
atica
ll
y
ad
ju
s
t
r
o
b
o
t
an
d
e
x
o
s
k
eleto
n
lo
ad
s
b
ased
o
n
th
e
p
atien
t’
s
co
n
d
itio
n
,
en
s
u
r
i
n
g
a
n
o
p
tim
al
tr
ai
n
in
g
p
r
o
ce
s
s
[
2
2
]
.
A
s
im
ilar
ap
p
r
o
ac
h
is
s
u
cc
e
s
s
f
u
lly
u
s
ed
in
th
e
o
il
an
d
g
as
in
d
u
s
tr
y
,
wh
er
e
L
STM
an
d
B
iLST
M
m
o
d
els
p
r
o
v
id
e
h
ig
h
ly
ac
cu
r
ate
p
r
ed
icti
o
n
s
o
f
o
p
er
atio
n
al
p
ar
am
eter
s
,
h
ig
h
lig
h
tin
g
t
h
e
p
o
ten
tial to
tr
an
s
f
er
th
ese
s
o
lu
ti
o
n
s
to
r
eh
a
b
ilit
atio
n
s
ettin
g
s
[
3
8
]
.
T
h
e
I
o
T
is
b
ein
g
ac
tiv
ely
in
teg
r
ated
in
to
r
eh
ab
ilit
atio
n
te
ch
n
o
lo
g
ies
f
o
r
r
em
o
te
m
o
n
it
o
r
in
g
an
d
m
an
ag
em
en
t
o
f
th
e
r
ec
o
v
er
y
p
r
o
ce
s
s
[
3
9
]
.
I
o
T
s
o
lu
tio
n
s
ar
e
ap
p
lied
in
wr
is
t
an
d
f
o
r
ea
r
m
r
eh
ab
ilit
atio
n
r
o
b
o
ts
,
as
well
as
in
k
n
ee
r
eh
ab
ilit
atio
n
s
y
s
tem
s
[
2
2
]
.
I
n
teg
r
ated
s
y
s
tem
s
h
av
e
b
ee
n
d
e
v
e
lo
p
ed
th
at
co
m
b
in
e
wea
r
ab
le
I
o
T
s
en
s
o
r
s
with
d
ee
p
lear
n
in
g
m
eth
o
d
s
f
o
r
p
e
r
s
o
n
alize
d
r
ec
o
g
n
itio
n
o
f
p
atien
t
a
ctiv
ity
at
h
o
m
e
an
d
in
“sm
ar
t
h
o
m
es”
[
3
9
]
.
T
h
ese
s
y
s
tem
s
en
ab
le
r
ea
l
-
tim
e
m
o
n
ito
r
in
g
o
f
p
h
y
s
io
lo
g
ic
al
p
ar
am
eter
s
an
d
m
o
v
em
en
ts
,
e
n
s
u
r
in
g
co
n
n
ec
ti
v
ity
b
etwe
en
th
e
p
atien
t
an
d
m
ed
ical
s
taf
f
v
ia
th
e
i
n
ter
n
et.
W
ea
r
ab
le
an
d
e
d
g
e
co
m
p
u
tin
g
d
ev
ices
with
b
u
ilt
-
in
s
en
s
o
r
s
an
d
B
lu
eto
o
th
s
u
p
p
o
r
t
p
lay
a
k
ey
r
o
le
in
m
ak
in
g
r
eh
ab
ilit
atio
n
m
o
r
e
m
o
b
ile
an
d
co
n
v
en
ien
t.
T
h
e
E
d
g
e
I
m
p
u
ls
e
p
latf
o
r
m
an
d
Ar
d
u
in
o
Nan
o
3
3
B
L
E
Sen
s
e
d
ev
ices
ar
e
u
s
ed
to
im
p
lem
en
t
m
ac
h
i
n
e
lear
n
i
n
g
o
n
ed
g
e
d
e
v
ices
in
o
r
th
o
p
e
d
ic
r
eh
ab
ilit
atio
n
[
2
5
]
.
I
o
T
also
en
ab
les
m
o
n
ito
r
in
g
o
f
v
ital
s
ig
n
s
(
E
C
G,
b
lo
o
d
p
r
e
s
s
u
r
e,
o
x
y
g
en
s
atu
r
atio
n
)
,
wh
i
ch
is
im
p
o
r
tan
t
in
tr
ea
tin
g
ch
r
o
n
ic
d
is
ea
s
es
[
3
9
]
.
Fu
r
th
er
m
o
r
e
,
I
o
T
s
y
s
tem
s
h
av
e
alr
ea
d
y
d
em
o
n
s
tr
ated
h
ig
h
a
cc
u
r
ac
y
a
n
d
ef
f
icien
c
y
in
m
an
ag
in
g
m
icr
o
clim
ate
an
d
m
o
n
ito
r
i
n
g
r
o
o
m
co
n
d
itio
n
s
,
wh
ich
m
ay
b
e
b
en
ef
icial
f
o
r
cr
ea
tin
g
o
p
tim
al
en
v
ir
o
n
m
e
n
ts
in
r
eh
ab
ilit
atio
n
zo
n
es
[
4
0
]
.
Desp
ite
th
ese
p
r
o
s
p
ec
ts
,
th
e
wid
esp
r
ea
d
co
m
m
er
cial
ad
o
p
tio
n
o
f
I
o
T
in
r
eh
ab
ilit
atio
n
r
em
ain
s
lim
ited
an
d
r
eq
u
ir
es f
u
r
t
h
er
r
e
s
ea
r
ch
an
d
d
ev
elo
p
m
en
t
[
4
1
]
.
T
h
e
d
etec
tio
n
an
d
p
r
e
d
ictio
n
o
f
m
u
s
cle
f
atig
u
e
r
em
ain
ch
all
en
g
in
g
task
s
d
u
e
to
th
e
s
u
b
jectiv
e
n
atu
r
e
o
f
f
atig
u
e
its
elf
an
d
th
e
lack
o
f
u
n
iv
er
s
al
ev
alu
atio
n
m
et
h
o
d
s
[
2
2
]
.
Mo
d
e
r
n
s
o
lu
tio
n
s
o
f
t
en
r
ely
o
n
ex
ter
n
al
p
h
y
s
io
lo
g
ical
d
ata
s
u
c
h
as
s
u
r
f
ac
e
E
MG
an
d
h
ea
r
t
r
ate,
wh
i
ch
r
eq
u
ir
e
ad
d
itio
n
al
s
en
s
o
r
s
,
in
cr
ea
s
e
co
s
ts
,
an
d
co
m
p
licate
s
y
s
tem
u
s
e
[
2
2
]
,
[
2
4
]
.
M
o
r
eo
v
er
,
s
u
ch
m
eth
o
d
s
m
ay
b
e
u
n
s
u
itab
le
f
o
r
p
a
tien
ts
with
lim
ited
p
h
y
s
ical
ab
ilit
ies,
s
u
ch
as
th
o
s
e
with
s
p
in
al
co
r
d
in
ju
r
ies.
I
n
s
u
f
f
icien
t
atten
tio
n
is
g
iv
e
n
t
o
an
aly
zin
g
f
atig
u
e
b
ased
s
o
lely
o
n
in
ter
n
al
s
y
s
tem
d
ata
(
e.
g
.
,
f
r
o
m
VR
o
r
r
o
b
o
tic
d
ev
ices)
,
wh
ic
h
co
u
ld
r
e
d
u
ce
d
e
p
en
d
e
n
ce
o
n
ex
ter
n
al
s
en
s
o
r
s
.
Alth
o
u
g
h
a
f
ew
s
tu
d
ies
u
s
e
s
im
p
le
m
o
d
els
s
u
ch
as
SVMs
with
a
lim
ited
n
u
m
b
er
o
f
f
ea
tu
r
es,
th
ey
h
i
g
h
lig
h
t
th
e
n
ee
d
f
o
r
m
o
r
e
ad
v
a
n
ce
d
an
d
ac
cu
r
ate
ap
p
r
o
ac
h
es
[
2
4
]
.
T
h
e
d
ev
elo
p
m
en
t
o
f
m
o
n
ito
r
in
g
an
d
r
eh
ab
ilit
atio
n
s
y
s
tem
s
is
also
co
n
s
tr
ain
ed
b
y
tech
n
ical
an
d
s
en
s
o
r
-
r
elate
d
lim
itatio
n
s
[
2
2
]
.
T
h
e
u
s
e
o
f
s
tep
p
er
m
o
to
r
s
with
lo
w
to
r
q
u
e,
I
M
U
v
ib
r
atio
n
s
,
a
n
d
th
e
n
ee
d
f
o
r
p
r
ec
is
e
s
en
s
o
r
p
o
s
itio
n
in
g
cr
ea
te
s
ig
n
if
ican
t
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
Ma
ch
in
e
lea
r
n
in
g
-
b
a
s
ed
cla
s
s
ifica
tio
n
o
f lo
ca
l m
u
s
cle
fa
tig
u
e
…
(
Zh
a
n
el
B
a
ig
a
r
a
ye
v
a
)
5957
ch
allen
g
es.
W
ir
ed
co
n
n
ec
tio
n
s
b
etwe
en
s
en
s
o
r
s
an
d
em
b
ed
d
ed
s
y
s
tem
s
r
ed
u
ce
m
o
b
ilit
y
,
wh
ile
s
u
r
f
ac
e
elec
tr
o
m
y
o
g
r
ap
h
y
(
s
E
MG
)
d
ata
q
u
ality
is
af
f
ec
ted
b
y
ar
tifa
cts
r
elate
d
to
s
k
in
co
n
tact
an
d
elec
tr
o
d
e
p
lace
m
en
t.
T
h
e
lim
ited
av
ail
ab
ilit
y
o
f
lar
g
e,
d
i
v
er
s
e,
an
d
o
p
e
n
s
en
s
o
r
d
atasets
,
alo
n
g
with
th
e
d
is
cr
ete
in
ter
p
r
etatio
n
o
f
co
n
tin
u
o
u
s
m
o
v
em
en
ts
,
h
in
d
er
s
th
e
ac
cu
r
ac
y
an
d
ad
a
p
tab
ilit
y
o
f
m
o
d
er
n
alg
o
r
ith
m
s
[
2
5
]
.
I
n
th
is
r
eg
ar
d
,
in
tellig
en
t
s
en
s
o
r
s
y
s
tem
s
u
s
in
g
I
o
T
an
d
ML
f
o
r
en
v
i
r
o
n
m
e
n
tal
m
o
n
ito
r
in
g
d
em
o
n
s
tr
ate
h
ig
h
s
en
s
itiv
ity
an
d
r
ea
l
-
tim
e
an
aly
s
is
ca
p
ab
ilit
ies,
wh
ich
co
u
ld
b
e
ad
ap
ted
f
o
r
m
o
n
ito
r
in
g
p
ati
en
ts
’
p
h
y
s
io
lo
g
ical
p
ar
am
eter
s
[
4
2
]
.
All
o
f
th
is
u
n
d
er
s
co
r
es
th
e
n
ee
d
f
o
r
a
c
o
m
p
r
eh
e
n
s
iv
e
r
ea
s
s
ess
m
en
t
o
f
b
o
th
s
en
s
in
g
an
d
an
aly
tical
ap
p
r
o
ac
h
es in
th
is
f
i
eld
.
I
n
th
is
s
tu
d
y
,
an
e
x
p
er
im
e
n
tal
in
v
esti
g
atio
n
an
d
d
ata
-
d
r
iv
e
n
an
aly
s
is
wer
e
co
n
d
u
cted
to
cla
s
s
if
y
lo
ca
l
m
u
s
cle
f
atig
u
e
d
u
r
in
g
is
o
m
etr
ic
h
an
d
g
r
i
p
ex
er
cises
u
s
in
g
elec
tr
o
m
y
o
g
r
ap
h
ic
(
E
MG
)
s
i
g
n
als.
T
h
e
p
r
im
ar
y
g
o
al
was
to
en
h
an
ce
r
eh
a
b
ilit
atio
n
o
u
tco
m
es
b
y
in
teg
r
atin
g
r
ea
l
-
tim
e
p
h
y
s
io
lo
g
ical
m
o
n
ito
r
in
g
with
m
ac
h
i
n
e
lear
n
in
g
-
b
ased
class
if
icatio
n
an
d
s
u
b
jectiv
e
f
atig
u
e
ass
ess
m
e
n
t.
R
ec
o
g
n
izin
g
th
e
ch
allen
g
es
o
f
ac
cu
r
ately
d
etec
tin
g
m
u
s
cle
f
atig
u
e
—
d
u
e
to
its
s
u
b
jectiv
e
an
d
m
u
ltifa
cto
r
ial
n
atu
r
e
—
th
e
s
tu
d
y
em
p
lo
y
e
d
a
d
u
al
ap
p
r
o
ac
h
.
A
cu
s
to
m
-
b
u
ilt
I
o
T
-
en
a
b
led
wea
r
ab
le
d
ev
ice
,
th
e
Qim
y
l
s
y
s
tem
,
was
u
s
ed
to
r
ec
o
r
d
a
n
d
p
r
o
ce
s
s
E
MG
s
ig
n
als.
Featu
r
e
ex
tr
ac
tio
n
m
e
th
o
d
s
wer
e
ap
p
lied
d
ir
ec
tly
o
n
th
e
d
ev
ice
to
g
en
er
ate
co
m
p
ac
t
d
escr
ip
to
r
s
s
u
ch
as
R
M
S,
MA
V,
P2
P,
an
d
MD
F,
wh
ich
wer
e
tr
an
s
m
itted
to
a
clo
u
d
s
er
v
e
r
f
o
r
s
to
r
ag
e
an
d
an
aly
s
is
.
Du
r
in
g
th
e
ex
p
er
im
en
t,
p
ar
ti
cip
an
ts
p
er
f
o
r
m
ed
s
u
s
tain
ed
s
u
b
m
ax
im
al
m
u
s
cl
e
co
n
tr
ac
tio
n
s
wh
ile
p
er
i
o
d
i
ca
lly
r
ep
o
r
tin
g
th
eir
p
e
r
ce
iv
ed
f
atig
u
e
lev
els,
en
ab
lin
g
alig
n
m
en
t
b
etwe
en
o
b
jectiv
e
E
MG
f
ea
tu
r
es
an
d
s
u
b
jectiv
e
s
en
s
atio
n
s
.
T
o
class
if
y
m
u
s
cle
s
tates
as
f
atig
u
ed
o
r
n
o
n
-
f
atig
u
e
d
,
s
ev
er
al
m
ac
h
in
e
lear
n
in
g
m
o
d
el
s
—
lo
g
is
tic
r
eg
r
ess
io
n
(
L
R
)
,
d
ec
is
io
n
tr
ee
(
DT
)
,
r
an
d
o
m
f
o
r
est
(
R
F)
,
an
d
XGBo
o
s
t
—
wer
e
tr
ain
ed
an
d
ev
alu
ated
.
XGBo
o
s
t
y
ield
ed
th
e
b
est
o
v
er
all
p
er
f
o
r
m
an
ce
,
wh
ile
RF
was
id
en
tifie
d
as
s
u
itab
le
f
o
r
r
ea
l
-
tim
e
em
b
ed
d
ed
a
p
p
licatio
n
s
.
T
h
e
f
in
d
i
n
g
s
d
em
o
n
s
tr
ate
th
e
f
ea
s
ib
ilit
y
an
d
ef
f
ec
tiv
e
n
ess
o
f
co
m
b
i
n
in
g
wea
r
ab
le
E
MG
s
en
s
in
g
with
in
tellig
en
t
class
if
icatio
n
alg
o
r
ith
m
s
to
s
u
p
p
o
r
t
ad
ap
tiv
e
r
e
h
ab
ilit
atio
n
,
p
er
s
o
n
alize
d
t
r
ain
in
g
,
an
d
ea
r
l
y
f
atig
u
e
d
etec
tio
n
.
T
h
is
in
teg
r
ated
ap
p
r
o
ac
h
o
f
f
er
s
a
s
ca
lab
le
s
o
lu
tio
n
f
o
r
m
o
d
er
n
teler
e
h
ab
ilit
atio
n
a
n
d
r
em
o
te
p
atien
t
m
o
n
ito
r
in
g
s
y
s
tem
s
.
Un
lik
e
m
an
y
p
r
ev
io
u
s
s
tu
d
ies,
wh
ich
o
f
ten
r
ely
s
o
lely
o
n
o
b
jectiv
e
E
MG
d
ata
with
o
u
t
in
c
o
r
p
o
r
ati
n
g
s
u
b
jectiv
e
f
atig
u
e
p
er
ce
p
tio
n
s
,
o
r
f
o
c
u
s
o
n
s
h
o
r
t,
m
ax
im
al
ef
f
o
r
t
b
u
r
s
ts
r
ath
er
t
h
an
s
u
s
t
ain
ed
s
u
b
m
a
x
im
al
co
n
tr
ac
tio
n
s
ty
p
ical
in
r
e
h
ab
ili
tatio
n
,
o
u
r
wo
r
k
ad
d
r
ess
es
th
ese
g
ap
s
b
y
s
y
n
ch
r
o
n
izin
g
r
ea
l
tim
e
p
h
y
s
io
lo
g
ical
s
ig
n
als
with
u
s
er
-
r
ep
o
r
te
d
f
atig
u
e
lev
els
d
u
r
i
n
g
r
ea
lis
tic
task
co
n
d
itio
n
s
.
M
o
r
eo
v
er
,
ex
is
tin
g
s
o
lu
tio
n
s
f
r
eq
u
e
n
tly
d
ep
en
d
o
n
m
u
ltip
le
ex
ter
n
al
s
en
s
o
r
s
,
lead
in
g
to
in
cr
ea
s
ed
co
m
p
lex
ity
a
n
d
r
ed
u
ce
d
p
r
ac
ticality
f
o
r
ev
er
y
d
a
y
u
s
e,
wh
e
r
ea
s
o
u
r
I
o
T
-
en
ab
le
d
wea
r
ab
le
p
latf
o
r
m
em
p
h
asizes
o
n
-
d
ev
ice
f
ea
tu
r
e
ex
tr
ac
tio
n
an
d
ef
f
icien
t
wir
eless
d
ata
tr
an
s
m
i
s
s
io
n
,
p
r
o
m
o
tin
g
lo
w
-
p
o
wer
a
n
d
m
o
b
ile
ap
p
licab
ilit
y
.
Ad
d
it
io
n
ally
,
wh
ile
p
r
io
r
wo
r
k
s
h
av
e
u
tili
ze
d
r
elativ
ely
s
im
p
le
m
ac
h
in
e
lear
n
i
n
g
m
o
d
els
o
r
f
ac
ed
lim
itatio
n
s
in
r
ea
l
tim
e
em
b
ed
d
e
d
im
p
lem
en
tatio
n
,
o
u
r
co
m
p
ar
a
tiv
e
an
aly
s
is
h
ig
h
lig
h
ts
ad
v
a
n
ce
d
alg
o
r
ith
m
s
lik
e
XGBo
o
s
t
f
o
r
ac
c
u
r
ac
y
an
d
R
an
d
o
m
Fo
r
est
f
o
r
e
m
b
ed
d
e
d
d
ep
l
o
y
m
en
t,
b
alan
ci
n
g
p
er
f
o
r
m
an
ce
an
d
o
p
er
atio
n
al
f
e
asib
ilit
y
.
T
h
u
s
,
o
u
r
s
tu
d
y
n
o
t
o
n
l
y
en
h
an
ce
s
f
ati
g
u
e
d
etec
tio
n
ac
cu
r
ac
y
b
u
t
also
ad
v
an
ce
s
p
r
ac
tical
in
teg
r
atio
n
in
to
ad
ap
tiv
e
r
eh
ab
ilit
atio
n
s
y
s
tem
s
,
o
v
er
co
m
in
g
k
e
y
tech
n
ical
an
d
m
eth
o
d
o
lo
g
ical
lim
itatio
n
s
id
en
tifie
d
in
liter
atu
r
e
.
3.
M
E
T
H
O
DO
L
O
G
Y
3
.
1
.
E
x
perim
ent
d
escript
io
n
T
en
h
ea
lth
y
v
o
lu
n
teer
s
(
8
m
al
e,
2
f
e
m
ale)
,
a
g
ed
1
9
–
2
2
,
p
ar
t
icip
ated
in
th
is
s
tu
d
y
.
T
o
en
s
u
r
e
in
ter
n
al
v
alid
ity
an
d
s
af
ety
,
p
ar
ticip
an
ts
wer
e
r
ec
r
u
ited
b
ased
o
n
s
tr
ict
in
clu
s
io
n
cr
iter
ia.
All
wer
e
f
r
ee
o
f
an
y
k
n
o
wn
m
u
s
cu
lo
s
k
eleta
l,
ca
r
d
io
v
ascu
lar
,
o
r
n
eu
r
o
lo
g
ical
co
n
d
itio
n
s
,
h
ad
n
o
u
p
p
er
lim
b
in
ju
r
ies
with
in
th
e
p
r
e
v
io
u
s
s
ix
m
o
n
th
s
,
an
d
wer
e
n
o
t
ta
k
in
g
m
ed
icatio
n
th
at
co
u
ld
af
f
ec
t
n
eu
r
o
m
u
s
cu
lar
f
u
n
ctio
n
.
E
ac
h
in
d
i
v
id
u
al
p
r
o
v
id
e
d
wr
itten
in
f
o
r
m
ed
co
n
s
en
t
af
ter
d
em
o
n
s
tr
atin
g
a
c
lear
u
n
d
er
s
tan
d
in
g
o
f
t
h
e
e
x
p
er
im
en
tal
p
r
o
to
co
l.
T
h
e
p
r
im
ar
y
o
b
jectiv
e
o
f
th
is
s
tu
d
y
was
to
in
v
esti
g
ate
th
e
d
ev
elo
p
m
en
t
o
f
lo
ca
lized
m
u
s
c
le
f
atig
u
e
d
u
r
in
g
a
b
r
ief
,
co
n
tr
o
lled
is
o
m
etr
ic
h
a
n
d
g
r
ip
task
.
A
k
ey
g
o
al
was
t
o
ex
p
lo
r
e
th
e
co
r
r
elatio
n
b
et
wee
n
s
E
MG
ac
tiv
ity
an
d
p
ar
ticip
an
ts
'
s
elf
-
r
ep
o
r
ted
s
en
s
atio
n
s
o
f
f
atig
u
e.
T
h
e
en
tire
p
r
o
ce
d
u
r
e
was
co
n
d
u
cted
in
a
s
in
g
le
s
e
s
s
io
n
last
in
g
ap
p
r
o
x
im
ately
th
r
ee
m
in
u
tes
p
er
p
ar
ticip
a
n
t,
u
s
in
g
a
ca
li
b
r
ated
h
an
d
g
r
ip
d
y
n
a
m
o
m
eter
t
o
q
u
an
tify
f
o
r
ce
o
u
tp
u
t.
T
h
e
ex
p
er
im
en
ta
l
s
ess
io
n
in
v
o
lv
e
d
two
co
n
s
ec
u
tiv
e
p
h
ases
.
First,
to
estab
lis
h
ea
ch
in
d
i
v
id
u
al’
s
m
ax
im
al
v
o
lu
n
tar
y
co
n
tr
ac
tio
n
(
MV
C
)
,
p
ar
ticip
an
ts
p
er
f
o
r
m
ed
th
r
ee
m
ax
im
al
is
o
m
etr
ic
co
n
tr
ac
tio
n
s
with
th
eir
d
o
m
in
an
t
h
an
d
.
E
ac
h
co
n
tr
ac
tio
n
last
ed
th
r
ee
s
ec
o
n
d
s
,
an
d
p
ar
ticip
a
n
ts
wer
e
m
o
tiv
ated
with
r
ea
l
-
tim
e
v
is
u
al
f
ee
d
b
ac
k
an
d
s
tan
d
a
r
d
i
ze
d
v
er
b
al
e
n
co
u
r
ag
em
e
n
t.
D
u
r
in
g
th
is
p
h
ase,
s
E
MG
s
ig
n
als
wer
e
co
n
tin
u
o
u
s
ly
r
ec
o
r
d
e
d
f
r
o
m
th
e
tar
g
et
f
o
r
ea
r
m
f
lex
o
r
m
u
s
cles.
T
h
e
r
o
o
t
m
ea
n
s
q
u
ar
e
(
R
MS)
v
alu
e
o
f
t
h
e
E
MG
s
ig
n
al
was
ca
lcu
lated
f
o
r
ea
c
h
tr
ial,
an
d
t
h
e
av
er
ag
e
o
f
t
h
e
th
r
ee
tr
ials
was
u
s
ed
to
d
ef
in
e
th
e
b
aselin
e
MV
C
r
ef
er
en
ce
,
en
s
u
r
in
g
in
te
r
-
p
ar
ticip
a
n
t
co
m
p
ar
ab
ilit
y
.
I
m
m
ed
iately
f
o
llo
win
g
th
e
MV
C
ass
ess
m
en
t,
p
ar
ticip
an
ts
b
eg
an
th
e
s
ec
o
n
d
p
h
ase
with
o
u
t
r
est
t
o
en
s
u
r
e
p
h
y
s
io
lo
g
ical
co
n
ti
n
u
ity
.
I
n
th
is
p
h
ase,
th
ey
m
ain
tain
ed
a
s
tead
y
is
o
m
etr
ic
co
n
tr
ac
tio
n
at
5
0
%
o
f
th
eir
in
d
i
v
id
u
al
MV
C
f
o
r
1
2
0
s
ec
o
n
d
s
.
T
h
is
p
r
o
to
co
l,
c
o
m
b
in
in
g
m
a
x
im
al
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
.
6
,
Decem
b
e
r
20
25
:
5
9
5
4
-
5
9
6
7
5958
ef
f
o
r
t
with
s
u
s
tain
ed
s
u
b
m
ax
im
al
ex
er
tio
n
,
is
wid
ely
ad
o
p
ted
in
f
atig
u
e
r
esear
c
h
to
s
im
u
late
r
ea
l
-
wo
r
l
d
m
u
s
cu
lar
d
em
an
d
s
an
d
en
s
u
r
e
a
co
n
tr
o
lled
i
n
d
u
ctio
n
o
f
f
atig
u
e
[
4
3
]
,
[
4
4
]
.
T
h
r
o
u
g
h
o
u
t
th
e
1
2
0
-
s
ec
o
n
d
s
u
s
tain
ed
co
n
tr
ac
tio
n
,
E
MG
ac
tiv
ity
was
co
n
tin
u
o
u
s
ly
r
ec
o
r
d
ed
to
tr
ac
k
ch
a
n
g
es
in
m
u
s
cle
ac
tiv
atio
n
.
Simu
ltan
eo
u
s
ly
,
p
ar
ticip
an
ts
p
r
o
v
id
ed
v
e
r
b
al
r
atin
g
s
o
f
th
eir
p
er
ce
iv
ed
f
atig
u
e
at
3
0
-
s
ec
o
n
d
in
ter
v
als.
T
h
ese
r
atin
g
s
wer
e
ca
p
tu
r
ed
u
s
in
g
a
0
-
1
0
n
u
m
er
ical
s
ca
le,
wh
er
e
0
in
d
icate
d
"n
o
f
atig
u
e"
a
n
d
1
0
r
ep
r
esen
ted
"m
ax
im
al
f
atig
u
e
.
"
Fo
r
a
n
aly
s
is
,
th
ese
s
elf
-
r
ep
o
r
ts
wer
e
th
en
ca
teg
o
r
ized
in
to
l
o
w
(
0
–
3
)
,
m
o
d
er
ate
(
4
–
6
)
,
an
d
h
ig
h
(
7
–
1
0
)
lev
els o
f
f
atig
u
e.
T
h
is
d
u
al
-
m
o
d
ality
ap
p
r
o
ac
h
,
wh
ich
in
teg
r
ates
o
b
jectiv
e
el
ec
tr
o
p
h
y
s
io
lo
g
ical
d
ata
with
s
u
b
jectiv
e
p
er
ce
p
tu
al
f
ee
d
b
ac
k
,
was
ch
o
s
en
to
p
r
o
v
id
e
a
co
m
p
r
eh
e
n
s
iv
e
an
d
ec
o
lo
g
ically
v
alid
ass
ess
m
en
t
o
f
f
atig
u
e
.
T
h
e
p
r
o
to
co
l'
s
d
esig
n
i
n
clu
d
e
d
s
ev
er
al
in
ten
tio
n
al
ch
o
ices
to
en
s
u
r
e
d
ata
in
teg
r
ity
.
T
h
e
ab
s
en
ce
o
f
a
r
est
p
er
io
d
b
etwe
en
t
h
e
two
p
h
as
es
was
cr
itical
to
p
r
ev
en
t
p
ar
tial
m
u
s
cle
r
ec
o
v
er
y
,
wh
ich
co
u
ld
co
n
f
o
u
n
d
th
e
f
atig
u
e
p
r
o
f
ile.
L
ik
ewise,
f
o
r
g
o
in
g
a
war
m
-
u
p
o
r
co
o
ld
o
wn
aim
ed
to
is
o
late
th
e
ac
u
te
m
u
s
cu
lar
r
esp
o
n
s
e
to
a
s
u
s
tain
ed
lo
ad
,
m
i
r
r
o
r
i
n
g
th
e
d
em
an
d
s
o
f
m
an
y
r
ea
l
-
wo
r
ld
task
s
.
I
n
s
u
m
m
ar
y
,
th
is
f
r
a
m
ewo
r
k
p
r
o
v
id
es
a
v
alid
ated
an
d
p
r
ac
tical
m
et
h
o
d
f
o
r
ass
ess
in
g
lo
ca
lized
m
u
s
cle
f
atig
u
e.
I
ts
b
r
ief
d
u
r
atio
n
,
m
in
im
al
eq
u
ip
m
en
t
r
eq
u
ir
em
e
n
ts
,
an
d
h
i
g
h
r
e
p
r
o
d
u
cib
ilit
y
m
ak
e
it
i
d
ea
lly
s
u
i
ted
f
o
r
i
n
teg
r
atio
n
in
to
m
o
b
il
e
h
ea
lth
p
latf
o
r
m
s
,
wea
r
ab
le
tech
n
o
lo
g
ies,
an
d
r
e
m
o
te
m
u
s
cu
lo
s
k
eleta
l e
v
alu
ati
o
n
s
y
s
tem
s
.
3
.
2
.
Q
i
m
y
l Io
T
d
ev
ice
T
o
en
s
u
r
e
ac
cu
r
ate
a
n
d
r
ep
r
o
d
u
cib
le
ac
q
u
is
itio
n
,
p
r
o
ce
s
s
in
g
,
an
d
an
al
y
s
is
o
f
s
u
r
f
ac
e
elec
tr
o
m
y
o
g
r
ap
h
ic
(
s
E
MG
)
s
i
g
n
als,
we
d
ev
elo
p
e
d
a
th
r
ee
-
tier
ed
s
y
s
tem
ar
ch
itectu
r
e
in
t
eg
r
atin
g
h
ar
d
war
e,
f
ir
m
war
e,
an
d
alg
o
r
ith
m
ic
m
o
d
u
les.
T
h
is
ar
ch
itectu
r
e
e
n
co
m
p
ass
es:
i)
an
ed
g
e
-
lev
el
I
o
T
d
ev
ice
(
Qim
y
l)
f
o
r
s
ig
n
al
ac
q
u
is
itio
n
an
d
p
r
ep
r
o
c
ess
in
g
,
ii)
a
ce
n
tr
al
s
er
v
er
f
o
r
s
tr
u
ctu
r
ed
d
ata
s
to
r
a
g
e
an
d
s
y
n
ch
r
o
n
izatio
n
,
an
d
iii)
a
m
ac
h
in
e
lear
n
i
n
g
p
i
p
elin
e
f
o
r
h
i
g
h
-
lev
el
class
if
icatio
n
o
f
m
u
s
cle
f
atig
u
e
s
tates,
as ill
u
s
tr
ated
in
Fig
u
r
e
1
.
Fig
u
r
e
1
.
Ar
c
h
itectu
r
e
o
f
th
e
m
u
lti
-
lev
el
s
y
s
tem
f
o
r
ac
q
u
is
itio
n
,
p
r
o
ce
s
s
in
g
,
s
to
r
ag
e,
a
n
d
a
n
aly
s
is
o
f
E
MG
s
ig
n
als
T
h
e
Qim
y
l
d
ev
ice
co
n
s
titu
tes
th
e
c
o
r
e
o
f
t
h
e
s
en
s
in
g
s
u
b
s
y
s
tem
an
d
is
co
m
p
o
s
ed
o
f
a
SEN0
2
4
0
s
E
MG
s
en
s
o
r
an
d
an
E
SP
3
2
m
icr
o
co
n
tr
o
ller
.
T
h
e
SEN0
2
4
0
m
o
d
u
le
ca
p
tu
r
es
d
i
f
f
er
en
tial
b
io
p
o
te
n
tials
f
r
o
m
th
e
s
k
in
s
u
r
f
ac
e
o
v
er
th
e
tar
g
e
t
m
u
s
cle
g
r
o
u
p
.
T
h
e
an
alo
g
s
ig
n
al
is
r
o
u
ted
to
th
e
h
ig
h
-
im
p
ed
an
ce
in
p
u
t
o
f
th
e
E
SP
3
2
,
wh
er
e
it
u
n
d
er
g
o
es
an
alo
g
-
to
-
d
ig
ital
co
n
v
er
s
io
n
.
Sam
p
lin
g
r
ates
b
etwe
en
1
0
0
Hz
an
d
1
0
0
0
Hz
wer
e
em
p
ir
ically
test
ed
,
with
5
0
0
Hz
s
elec
ted
as
th
e
d
ef
a
u
lt,
p
r
o
v
id
in
g
a
b
alan
ce
b
etwe
en
s
p
ec
tr
al
f
id
elity
(
p
r
eser
v
in
g
th
e
in
f
o
r
m
ativ
e
2
0
–
4
5
0
Hz
E
MG
b
an
d
)
an
d
p
o
wer
ef
f
icien
cy
.
A
d
ig
ital
4
t
h
-
o
r
d
e
r
B
u
tter
wo
r
th
b
an
d
-
p
ass
f
ilter
(
c
u
t
-
o
f
f
:
2
0
–
4
5
0
Hz)
is
a
p
p
lied
in
r
ea
l
-
tim
e
to
s
u
p
p
r
ess
m
o
tio
n
ar
tifa
cts
an
d
p
o
wer
lin
e
in
ter
f
er
en
ce
(
5
0
/6
0
Hz)
.
T
h
i
s
f
ilter
in
g
s
tag
e
r
ep
r
esen
ts
a
s
tan
d
ar
d
b
u
t
ess
en
tial
s
te
p
in
en
s
u
r
in
g
th
e
p
h
y
s
io
lo
g
ical
r
ele
v
an
ce
o
f
th
e
ac
q
u
ir
ed
s
ig
n
al.
Fo
llo
win
g
f
iltra
tio
n
,
th
e
s
ig
n
al
s
tr
ea
m
is
s
eg
m
en
ted
in
to
o
v
er
lap
p
i
n
g
2
0
0
m
s
win
d
o
ws
with
5
0
%
o
v
er
lap
to
r
etain
tem
p
o
r
al
r
eso
lu
tio
n
.
W
ith
in
ea
ch
win
d
o
w,
f
iv
e
v
alid
ated
tim
e
-
d
o
m
ain
an
d
f
r
e
q
u
en
c
y
-
d
o
m
ain
f
ea
tu
r
es
ar
e
ex
tr
ac
t
ed
:
r
o
o
t
m
ea
n
s
q
u
a
r
e
(
R
MS)
,
m
ea
n
ab
s
o
lu
te
v
alu
e
(
MA
V)
,
p
ea
k
-
to
-
p
ea
k
am
p
litu
d
e
(
P2
P),
m
e
d
ian
f
r
eq
u
en
cy
(
MD
F),
a
n
d
a
b
i
n
ar
y
a
ctiv
ity
f
lag
.
R
MS
an
d
MA
V,
co
m
m
o
n
l
y
u
s
ed
in
E
MG
an
aly
s
is
,
r
ef
lect
co
n
tr
ac
tio
n
in
ten
s
ity
,
wh
ile
P2
P
an
d
MD
F
p
r
o
v
id
e
in
s
ig
h
ts
in
t
o
s
ig
n
al
am
p
litu
d
e
d
y
n
am
ics
an
d
s
p
ec
tr
al
d
is
tr
ib
u
tio
n
.
T
h
e
b
in
ar
y
ac
tiv
ity
f
la
g
i
s
ca
lcu
lated
u
s
in
g
a
h
y
s
ter
esis
-
b
ased
r
u
le
ap
p
lied
to
R
MS
an
d
MA
V,
in
d
icatin
g
th
e
p
r
esen
ce
o
f
v
o
lu
n
tar
y
m
u
s
cu
lar
ac
tiv
atio
n
.
R
ath
er
th
an
tr
an
s
m
itti
n
g
r
aw
E
MG
s
ig
n
als,
wh
ich
ar
e
b
an
d
wid
th
-
in
ten
s
iv
e
an
d
s
u
s
ce
p
tib
l
e
to
lo
s
s
,
th
e
s
y
s
tem
tr
an
s
m
its
o
n
ly
co
m
p
ac
t
f
iv
e
-
elem
en
t
f
ea
tu
r
e
v
ec
to
r
s
with
tim
estam
p
s
v
ia
W
i
-
Fi.
T
h
is
a
p
p
r
o
ac
h
,
as
illu
s
tr
ated
in
Fig
u
r
e
1
,
s
ig
n
if
ican
tly
r
ed
u
ce
s
d
ata
lo
a
d
an
d
im
p
r
o
v
e
s
en
er
g
y
ef
f
icien
cy
,
m
ak
in
g
it su
itab
le
f
o
r
wea
r
ab
le,
b
atter
y
-
o
p
er
ated
s
y
s
tem
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
Ma
ch
in
e
lea
r
n
in
g
-
b
a
s
ed
cla
s
s
ifica
tio
n
o
f lo
ca
l m
u
s
cle
fa
tig
u
e
…
(
Zh
a
n
el
B
a
ig
a
r
a
ye
v
a
)
5959
T
h
e
s
er
v
er
b
ac
k
e
n
d
r
ec
eiv
es
t
h
ese
f
ea
tu
r
e
v
ec
to
r
s
in
r
ea
l
-
ti
m
e
an
d
s
to
r
es
th
em
in
a
s
tr
u
ctu
r
ed
tim
e
-
s
er
ies
f
o
r
m
at
to
g
eth
er
with
m
etad
ata
s
u
ch
as
tim
estam
p
s
,
s
ess
io
n
id
en
tifie
r
s
,
a
n
d
d
ev
ice
id
en
tifie
r
s
.
T
h
is
d
esig
n
s
u
p
p
o
r
ts
lo
n
g
itu
d
in
al
a
n
aly
s
is
,
tim
e
-
b
ased
ag
g
r
eg
ati
o
n
,
a
n
d
s
tab
le
s
to
r
a
g
e.
A
R
E
STf
u
l
API
in
ter
f
ac
e
f
ac
ilit
ates
r
ea
l
-
tim
e
ac
ce
s
s
f
o
r
m
o
n
it
o
r
in
g
an
d
v
is
u
aliza
tio
n
,
as
well
as
h
is
to
r
ical
d
ata
ex
p
o
r
t
f
o
r
o
f
f
lin
e
tr
ain
in
g
,
an
n
o
tatio
n
,
o
r
r
ev
iew
.
T
h
e
m
ac
h
in
e
lear
n
in
g
m
o
d
u
le
ad
d
r
ess
es
th
e
b
in
ar
y
class
i
f
icatio
n
task
o
f
d
etec
tin
g
th
e
o
n
s
et
o
f
m
u
s
cle
f
atig
u
e.
T
h
is
co
n
s
titu
tes
a
n
o
v
el
ap
p
licatio
n
o
f
s
u
p
e
r
v
is
ed
lear
n
in
g
in
ed
g
e
-
d
e
p
lo
y
ed
E
MG
s
y
s
tem
s
,
with
lab
els
d
er
iv
ed
f
r
o
m
r
ea
l
-
tim
e
u
s
er
in
p
u
t
an
d
f
o
r
ce
d
e
ca
y
cr
iter
ia.
Fo
u
r
class
if
icatio
n
alg
o
r
ith
m
s
wer
e
ev
alu
ated
:
lo
g
is
tic
r
eg
r
ess
io
n
,
d
ec
is
io
n
tr
ee
,
r
an
d
o
m
f
o
r
est,
an
d
XGBo
o
s
t.
Stra
tifie
d
5
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
was
u
s
ed
to
im
p
r
o
v
e
g
en
er
al
izatio
n
an
d
a
d
d
r
ess
class
im
b
alan
ce
.
Mo
d
el
p
er
f
o
r
m
a
n
ce
was
ass
e
s
s
ed
u
s
in
g
R
OC
-
AU
C
,
F1
-
s
co
r
e,
an
d
in
f
er
en
ce
laten
cy
.
XGBo
o
s
t
ac
h
iev
ed
th
e
h
ig
h
est
ac
cu
r
ac
y
u
n
d
er
lim
ited
tr
ain
in
g
d
ata,
wh
ile
th
e
r
an
d
o
m
f
o
r
est
m
o
d
el
s
h
o
wed
s
u
p
er
io
r
ex
ec
u
tio
n
s
p
ee
d
an
d
was
th
u
s
s
el
ec
ted
f
o
r
o
n
-
d
ev
ic
e
d
ep
lo
y
m
e
n
t.
T
h
is
en
ab
led
lo
c
al
in
f
er
en
ce
d
ir
ec
tly
o
n
th
e
E
SP
3
2
,
en
s
u
r
in
g
r
ea
l
-
tim
e
cl
ass
if
icatio
n
with
o
u
t
r
eq
u
ir
in
g
a
co
n
s
tan
t ser
v
er
co
n
n
ec
tio
n
.
T
h
e
p
h
y
s
ical
r
ea
lizatio
n
o
f
th
e
Qim
y
l
d
ev
ice
is
d
ep
icted
in
Fig
u
r
e
2
.
T
h
e
E
MG
s
en
s
o
r
is
attac
h
ed
o
v
er
th
e
f
lex
o
r
ca
r
p
i
r
ad
ialis
m
u
s
cle
u
s
in
g
m
ed
ical
-
g
r
a
d
e
a
d
h
esiv
e,
wh
ile
th
e
E
SP
3
2
m
ic
r
o
co
n
tr
o
ller
is
f
ix
ed
to
th
e
f
o
r
ea
r
m
with
a
lig
h
twei
g
h
t
s
tr
ap
.
T
h
is
ar
r
a
n
g
em
e
n
t
r
e
d
u
ce
s
wir
e
ten
s
io
n
an
d
m
o
v
e
m
en
t
-
in
d
u
ce
d
n
o
is
e,
wh
ile
p
r
eser
v
in
g
co
m
f
o
r
t
an
d
m
o
b
ilit
y
d
u
r
in
g
u
s
e.
I
f
n
etwo
r
k
co
n
n
ec
tiv
ity
is
tem
p
o
r
ar
ily
l
o
s
t,
f
ea
tu
r
e
v
ec
to
r
s
ar
e
b
u
f
f
e
r
ed
lo
ca
lly
an
d
au
to
m
atica
lly
tr
an
s
m
itted
o
n
ce
th
e
co
n
n
ec
tio
n
is
r
esto
r
ed
,
e
n
s
u
r
i
n
g
th
e
co
n
tin
u
ity
o
f
th
e
d
ataset.
Ov
er
all,
th
e
p
r
o
p
o
s
ed
s
y
s
tem
in
teg
r
ates
v
alid
a
ted
s
ig
n
al
p
r
o
ce
s
s
in
g
tech
n
iq
u
es
with
em
b
e
d
d
ed
m
ac
h
in
e
lear
n
in
g
to
d
eliv
er
a
n
en
er
g
y
-
e
f
f
icien
t,
l
o
w
-
laten
c
y
,
an
d
in
tellig
en
t
E
MG
m
o
n
i
to
r
in
g
s
o
l
u
tio
n
.
I
ts
d
esig
n
s
u
p
p
o
r
ts
p
er
s
o
n
alize
d
f
atig
u
e
tr
ac
k
in
g
an
d
a
d
ap
tiv
e
r
eh
ab
ilit
atio
n
in
r
ea
l
-
tim
e,
as
v
is
u
ally
s
u
m
m
ar
ize
d
in
Fig
u
r
e
1
.
Fig
u
r
e
2
.
I
o
T
d
e
v
ice
Qim
y
l
3
.
3
.
Da
t
a
p
re
pro
ce
s
s
ing
T
h
e
d
ataset
em
p
lo
y
ed
in
th
i
s
s
tu
d
y
was
o
b
tain
ed
t
h
r
o
u
g
h
s
u
r
f
ac
e
elec
tr
o
m
y
o
g
r
a
p
h
i
c
(
s
E
MG
)
r
ec
o
r
d
in
g
s
u
s
in
g
a
s
tan
d
ar
d
iz
ed
E
MG
s
en
s
o
r
o
v
er
a
1
2
0
-
s
ec
o
n
d
i
n
ter
v
al
f
o
r
ea
c
h
p
ar
ticip
an
t.
T
o
ca
p
tu
r
e
s
u
b
jectiv
e
p
er
ce
p
tio
n
s
o
f
f
atig
u
e,
i
n
d
iv
id
u
als
wer
e
p
r
o
m
p
t
ed
ev
e
r
y
3
0
s
ec
o
n
d
s
to
s
elf
-
ass
es
s
th
eir
f
atig
u
e
lev
el
u
s
in
g
a
1
0
-
p
o
in
t
L
ik
e
r
t
-
ty
p
e
s
ca
le.
Fo
r
a
n
aly
tical
p
u
r
p
o
s
es,
th
ese
s
elf
-
ass
ess
m
en
t
s
co
r
es
wer
e
s
u
b
s
eq
u
en
tly
d
is
cr
etize
d
in
t
o
th
r
ee
o
r
d
in
al
ca
teg
o
r
ies:
0
–
3
(
n
o
f
atig
u
e,
lab
ele
d
as
class
0
)
,
4
–
6
(
m
o
d
e
r
ate
f
atig
u
e,
class
1
)
,
an
d
7
–
1
0
(
h
i
g
h
f
atig
u
e,
class
2
)
.
T
h
is
p
r
o
c
ed
u
r
e
alig
n
s
with
r
ec
en
t
ef
f
o
r
t
s
to
p
air
s
u
b
jectiv
e
f
atig
u
e
r
atin
g
s
with
s
E
MG
d
ata
to
en
h
an
ce
m
o
d
el
tr
ain
in
g
an
d
class
if
icatio
n
[
4
5
]
,
an
d
th
e
ad
o
p
te
d
ca
teg
o
r
izatio
n
s
ch
em
e
was
s
p
ec
if
ically
ch
o
s
en
to
e
n
s
u
r
e
r
o
b
u
s
t
class
s
ep
ar
atio
n
in
s
u
p
er
v
is
ed
lear
n
in
g
task
s
,
co
n
s
is
ten
t w
ith
estab
lis
h
ed
p
r
ac
tices in
m
u
s
cle
f
atig
u
e
r
esear
ch
.
Un
lik
e
ce
r
tain
p
r
e
p
r
o
ce
s
s
in
g
p
ip
elin
es
th
at
in
tr
o
d
u
ce
s
y
n
th
etic
au
g
m
en
tatio
n
,
t
h
is
s
tu
d
y
ex
clu
s
iv
ely
u
tili
ze
d
o
r
ig
in
al
ex
p
e
r
im
en
tal
d
ata
to
p
r
eser
v
e
ec
o
l
o
g
ical
v
a
lid
ity
.
T
h
e
r
aw
d
ataset
was
in
itially
s
cr
ee
n
ed
f
o
r
d
u
p
licates
an
d
m
is
s
in
g
en
tr
ies
;
n
o
n
e
wer
e
f
o
u
n
d
,
c
o
n
f
ir
m
i
n
g
d
ata
in
teg
r
ity
.
C
o
n
tr
ar
y
to
t
y
p
ical
n
o
r
m
aliza
tio
n
p
r
o
ce
d
u
r
es,
n
o
s
ca
lin
g
tr
an
s
f
o
r
m
atio
n
s
wer
e
a
p
p
lied
at
th
is
s
tag
e,
s
in
ce
th
e
s
elec
ted
m
ac
h
in
e
lear
n
in
g
m
o
d
els
em
p
lo
y
ed
later
in
th
e
p
ip
elin
e
d
em
o
n
s
tr
ated
r
o
b
u
s
tn
ess
to
s
ca
le
h
eter
o
g
en
eit
y
.
Ad
d
itio
n
al
co
n
tex
tu
al
v
ar
iab
l
es
wer
e
in
co
r
p
o
r
ated
to
ex
p
l
o
r
e
th
eir
p
o
te
n
tial
in
f
lu
e
n
ce
o
n
f
atig
u
e
p
er
ce
p
tio
n
.
T
h
ese
in
clu
d
ed
d
em
o
g
r
ap
h
ic
an
d
life
s
ty
le
f
ac
t
o
r
s
s
u
ch
as
ag
e,
g
en
d
er
,
r
eg
u
lar
en
g
ag
e
m
en
t
in
s
p
o
r
ts
,
s
m
o
k
in
g
s
tatu
s
,
an
d
al
co
h
o
l
co
n
s
u
m
p
tio
n
.
C
ateg
o
r
ic
al
f
ea
tu
r
es
wer
e
en
co
d
e
d
n
u
m
er
ically
u
s
in
g
th
e
L
ab
elE
n
co
d
e
r
f
u
n
ctio
n
f
r
o
m
s
cik
it
-
lear
n
,
en
ab
lin
g
s
ea
m
less
in
teg
r
atio
n
in
to
d
o
wn
s
tr
ea
m
s
tatis
tical
an
d
m
ac
h
in
e
lear
n
in
g
m
o
d
els.
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
.
6
,
Decem
b
e
r
20
25
:
5
9
5
4
-
5
9
6
7
5960
A
co
r
r
elatio
n
a
n
aly
s
is
was
c
o
n
d
u
cte
d
to
ev
al
u
ate
in
ter
-
v
a
r
iab
le
r
elatio
n
s
h
ip
s
an
d
d
etec
t
p
o
ten
tial
m
u
ltico
llin
ea
r
ity
.
T
h
e
r
esu
ltin
g
Pear
s
o
n
co
r
r
elatio
n
m
atr
ix
,
s
h
o
wn
in
Fig
u
r
e
3
,
s
er
v
ed
as
a
d
iag
n
o
s
tic
to
o
l
f
o
r
f
ea
tu
r
e
s
elec
tio
n
an
d
d
im
e
n
s
io
n
ality
r
ed
u
ctio
n
.
Ap
p
ly
i
n
g
a
s
im
p
le
co
r
r
elatio
n
f
ilter
s
ig
n
if
ican
tly
m
itig
ates
m
u
ltico
llin
ea
r
ity
an
d
,
wh
en
co
m
b
in
ed
with
co
m
p
lem
en
ta
r
y
f
ea
tu
r
e
-
s
elec
tio
n
m
eth
o
d
s
,
r
elia
b
ly
en
h
an
ce
s
b
o
th
th
e
p
r
ed
ictiv
e
ac
cu
r
ac
y
an
d
r
o
b
u
s
tn
ess
o
f
th
e
m
o
d
els
[
4
6
]
.
As
in
d
icate
d
in
th
e
m
atr
ix
,
t
h
e
“f
atig
u
e”
s
co
r
e
d
em
o
n
s
tr
ated
a
n
e
x
tr
em
ely
h
ig
h
p
o
s
itiv
e
co
r
r
elatio
n
(
r
=
0
.
9
5
)
with
th
e
d
er
iv
ed
c
ateg
o
r
ical
v
a
r
iab
le
“f
atig
u
e_
class
,
”
co
n
f
ir
m
in
g
r
ed
u
n
d
an
cy
.
Acc
o
r
d
in
g
l
y
,
th
e
“f
atig
u
e”
co
lu
m
n
was
ex
clu
d
ed
to
p
r
ev
e
n
t
co
llin
ea
r
ity
-
in
d
u
ce
d
m
o
d
el
i
n
s
tab
ilit
y
.
Ad
d
itio
n
ally
,
t
h
e
“n
am
e”
co
lu
m
n
,
s
er
v
i
n
g
o
n
l
y
as
a
p
ar
ticip
a
n
t
id
en
tifie
r
,
was
d
is
ca
r
d
ed
d
u
e
t
o
its
lack
o
f
an
aly
tical
u
tili
ty
an
d
p
o
ten
tial
to
in
tr
o
d
u
ce
a
r
tifa
cts.
T
h
is
r
ig
o
r
o
u
s
p
r
ep
r
o
ce
s
s
in
g
p
ip
elin
e
en
s
u
r
es
th
at
o
n
ly
in
f
o
r
m
ativ
e
an
d
n
o
n
-
r
ed
u
n
d
an
t
f
ea
t
u
r
es
ar
e
r
etain
ed
f
o
r
f
u
r
t
h
er
an
aly
s
is
,
th
er
eb
y
en
h
a
n
cin
g
b
o
th
th
e
v
alid
ity
a
n
d
r
ep
r
o
d
u
ci
b
ilit
y
o
f
th
e
r
esear
c
h
.
T
h
e
u
s
e
o
f
d
o
m
ai
n
-
ju
s
tifie
d
d
is
cr
etiza
tio
n
,
ex
clu
s
io
n
cr
it
er
ia
b
ased
o
n
s
tatis
tical
d
iag
n
o
s
tics
,
an
d
s
tan
d
ar
d
ize
d
e
n
co
d
in
g
tech
n
iq
u
es
co
llectiv
ely
s
u
p
p
o
r
t th
e
m
eth
o
d
o
lo
g
ical
tr
an
s
p
a
r
en
cy
r
eq
u
ir
e
d
f
o
r
r
ep
licatio
n
.
Fig
u
r
e
3
.
C
o
r
r
elatio
n
m
atr
ix
3
.
4
.
M
a
chine
lea
rning
cla
s
s
i
f
ica
t
io
n
Su
p
er
v
is
ed
m
ac
h
in
e
-
lear
n
in
g
tech
n
iq
u
es
wer
e
em
p
lo
y
ed
to
co
n
s
tr
u
ct
a
m
o
d
el
th
at
q
u
an
tifie
s
p
ar
ticip
an
ts
’
f
atig
u
e
lev
els
f
r
o
m
elec
tr
o
m
y
o
g
r
a
p
h
y
(
E
MG
)
r
ec
o
r
d
in
g
s
to
g
eth
er
with
a
d
d
itio
n
al
co
n
te
x
tu
al
f
ac
to
r
s
.
Fo
u
r
co
m
p
lem
en
tar
y
class
if
ier
s
—
LR
,
RF
,
XGBo
o
s
t
an
d
DT
—
wer
e
s
elec
ted
b
ec
au
s
e
th
ey
m
an
ag
e
co
m
p
lex
,
s
tr
u
ct
u
r
ed
d
ata
ef
f
ec
tiv
ely
an
d
ca
p
tu
r
e
p
o
ten
ti
al
n
o
n
-
lin
ea
r
r
elatio
n
s
h
ip
s
with
in
it.
L
o
g
is
tic
r
eg
r
ess
io
n
s
er
v
ed
as
an
in
ter
p
r
etab
le
lin
ea
r
b
aselin
e
f
o
r
d
et
ec
tin
g
s
im
p
le
ass
o
ciatio
n
s
[
4
7
]
,
wh
er
ea
s
R
F
an
d
DT
,
as
tr
ee
-
b
ased
en
s
em
b
le
lear
n
er
s
,
m
o
d
elled
h
ier
ar
c
h
ical
p
atter
n
s
an
d
h
ig
h
-
o
r
d
er
f
ea
t
u
r
e
in
ter
ac
tio
n
s
[
4
8
]
,
[
4
9
]
.
E
x
tr
em
e
g
r
ad
ien
t
b
o
o
s
t
in
g
(
XGBo
o
s
t)
was
in
clu
d
ed
f
o
r
its
p
r
o
v
en
ca
p
ac
ity
to
r
ep
r
esen
t
in
tr
icate
v
ar
iab
le
d
ep
en
d
en
cies
wh
ile
m
ain
tain
in
g
s
tr
o
n
g
p
r
ed
ictiv
e
p
er
f
o
r
m
a
n
ce
[
5
0
]
.
Hy
p
e
r
-
p
ar
am
eter
o
p
tim
is
atio
n
f
o
r
ev
er
y
alg
o
r
ith
m
was
p
er
f
o
r
m
ed
with
an
ex
h
au
s
tiv
e
Gr
id
Sear
ch
,
en
s
u
r
in
g
ea
ch
m
o
d
el
was
tu
n
ed
to
th
e
s
p
ec
if
ic
ch
ar
ac
ter
is
tics
o
f
th
e
d
ataset.
I
m
p
lem
en
tatio
n
r
elied
o
n
th
e
Scik
it
-
lear
n
an
d
XGBo
o
s
t lib
r
ar
ies.
Mo
d
el
p
er
f
o
r
m
a
n
ce
was
ass
ess
ed
with
f
o
u
r
s
tan
d
ar
d
m
e
tr
ics.
Acc
u
r
ac
y
—
th
e
r
atio
o
f
co
r
r
ec
tly
class
if
ied
in
s
tan
ce
s
to
th
e
to
tal
n
u
m
b
er
o
f
p
r
ed
ictio
n
s
—
o
f
f
e
r
s
an
o
v
er
all
g
a
u
g
e
o
f
co
r
r
ec
t
n
ess
b
u
t
ca
n
m
ask
d
ef
icien
cies
wh
en
class
es
ar
e
im
b
alan
ce
d
.
Pre
cisi
o
n
,
d
ef
in
ed
as
th
e
p
r
o
p
o
r
ti
o
n
o
f
tr
u
e
p
o
s
itiv
es
am
o
n
g
all
p
o
s
itiv
e
p
r
ed
ictio
n
s
,
h
ig
h
lig
h
t
s
th
e
m
o
d
el’
s
r
esis
tan
ce
to
f
alse
p
o
s
itiv
es.
R
ec
all
m
ea
s
u
r
e
s
th
e
s
h
ar
e
o
f
tr
u
e
p
o
s
itiv
es
r
etr
iev
ed
f
r
o
m
all
ac
tu
al
p
o
s
itiv
es,
in
d
icatin
g
th
e
m
o
d
el’
s
s
en
s
itiv
ity
.
T
h
e
h
ar
m
o
n
ic
m
ea
n
o
f
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
Ma
ch
in
e
lea
r
n
in
g
-
b
a
s
ed
cla
s
s
ifica
tio
n
o
f lo
ca
l m
u
s
cle
fa
tig
u
e
…
(
Zh
a
n
el
B
a
ig
a
r
a
ye
v
a
)
5961
Pre
cisi
o
n
an
d
R
ec
all,
th
e
F1
-
s
co
r
e,
p
r
o
v
id
es
a
b
alan
ce
d
s
in
g
le
-
v
alu
e
s
u
m
m
ar
y
w
h
en
a
tr
ad
e
-
o
f
f
e
x
is
ts
b
etwe
en
th
e
two
.
T
h
e
m
etr
ics
wer
e
co
m
p
u
te
d
ac
co
r
d
in
g
t
o
t
h
e
ex
p
r
ess
io
n
s
:
=
+
+
+
+
(
1
)
=
+
(
2
)
=
+
(
3
)
1
=
2
×
×
+
(
4
)
Her
e,
T
P
r
ep
r
esen
ts
th
e
n
u
m
b
er
o
f
tr
u
e
-
p
o
s
itiv
e
p
r
ed
ictio
n
s
,
T
N
th
e
tr
u
e
-
n
eg
ativ
e
p
r
ed
ict
io
n
s
,
FP
th
e
f
al
s
e
-
p
o
s
itiv
e
p
r
ed
ictio
n
s
,
a
n
d
FN t
h
e
f
alse
-
n
eg
ativ
e
p
r
ed
ictio
n
s
.
T
ak
en
to
g
eth
er
,
th
e
ca
r
ef
u
ll
y
tu
n
e
d
class
if
ier
s
et,
th
e
c
lass
-
im
b
alan
ce
s
tr
ateg
y
,
an
d
th
e
f
o
u
r
co
m
p
lem
en
tar
y
m
etr
ics
p
r
o
v
i
d
e
a
co
m
p
r
e
h
en
s
iv
e
an
d
r
e
p
r
o
d
u
cib
le
ev
al
u
atio
n
f
r
am
ewo
r
k
[
5
1
]
.
T
h
is
s
tr
u
ctu
r
e
n
o
t
o
n
ly
h
ig
h
lig
h
ts
o
v
er
all
ac
cu
r
ac
y
b
u
t
also
clar
if
ies
ea
ch
m
o
d
el’
s
ten
d
en
c
y
to
war
d
f
alse
alar
m
s
o
r
m
is
s
ed
d
etec
tio
n
s
—
cr
itical
f
ac
to
r
s
wh
en
ass
ess
in
g
f
atig
u
e
f
r
o
m
E
MG
d
ata
—
th
er
eb
y
en
s
u
r
in
g
th
at
s
u
b
s
eq
u
e
n
t
co
m
p
ar
is
o
n
s
an
d
in
ter
p
r
etatio
n
s
r
est o
n
a
s
o
lid
,
tr
a
n
s
p
ar
en
t f
o
u
n
d
atio
n
.
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
an
aly
s
is
o
f
m
ac
h
i
n
e
lear
n
in
g
m
o
d
els
f
o
r
class
if
y
in
g
p
ar
ticip
an
ts
’
f
atig
u
e
lev
els
s
h
o
ws
v
ar
y
in
g
ac
cu
r
ac
y
an
d
ef
f
icien
cy
am
o
n
g
th
e
test
ed
alg
o
r
ith
m
s
.
T
h
e
DT
m
o
d
el
ac
h
iev
ed
th
e
b
est
r
esu
lts
with
an
ac
cu
r
ac
y
o
f
9
0
.
7
%,
alo
n
g
w
ith
h
ig
h
p
r
ec
is
io
n
(
9
0
.
7
%)
an
d
r
ec
all
(
9
0
.
9
%).
T
h
is
in
d
i
ca
tes
it
s
ab
ilit
y
to
ef
f
ec
tiv
ely
class
if
y
f
atig
u
e
lev
els with
a
g
o
o
d
b
alan
ce
b
etwe
en
co
m
p
leten
ess
an
d
p
r
ec
is
io
n
.
T
h
e
RF
m
o
d
el
d
em
o
n
s
tr
ated
an
ac
cu
r
ac
y
o
f
8
9
.
4
%,
p
r
ec
is
io
n
o
f
8
9
.
3
%,
an
d
r
ec
all
o
f
8
9
.
9
%.
T
h
e
class
if
icatio
n
r
esu
lts
ac
r
o
s
s
cl
ass
es
s
h
o
wed
h
ig
h
s
co
r
es
f
o
r
all
th
r
ee
f
atig
u
e
ca
teg
o
r
ies,
wi
th
p
ar
ticu
lar
ly
h
i
g
h
R
ec
all
f
o
r
clas
s
0
(
n
o
f
atig
u
e)
at
9
6
%.
Du
e
to
its
en
s
em
b
le
s
tr
u
ctu
r
e,
RF
ef
f
ec
tiv
ely
ca
p
tu
r
es
co
m
p
lex
d
ep
en
d
e
n
cies b
etwe
en
f
ea
tu
r
e
s
,
en
s
u
r
in
g
m
o
d
el
r
o
b
u
s
tn
ess
.
T
h
e
XGBo
o
s
t
m
o
d
el
s
h
o
wed
co
m
p
a
r
ab
le
r
esu
lts
with
an
ac
cu
r
ac
y
o
f
8
9
.
4
%,
an
d
p
r
ec
is
io
n
an
d
r
ec
all
b
o
th
at
8
9
.
4
%.
C
lass
-
wi
s
e
class
if
icatio
n
was
also
s
tab
le,
with
p
r
ec
is
io
n
an
d
r
ec
all
r
a
n
g
in
g
f
r
o
m
8
6
%
to
9
5
%.
T
h
is
m
eth
o
d
co
n
f
ir
m
ed
its
ef
f
ec
tiv
en
ess
in
th
e
f
atig
u
e
class
if
icatio
n
task
,
alth
o
u
g
h
it
s
lig
h
tly
lag
g
ed
b
eh
in
d
t
h
e
DT
in
o
v
er
all
ac
cu
r
ac
y
.
L
o
g
is
tic
r
eg
r
ess
io
n
ac
h
i
ev
ed
th
e
lo
west
ac
cu
r
ac
y
at
7
4
.
2
%.
I
ts
p
r
ec
is
io
n
an
d
r
ec
all
wer
e
o
n
ly
7
3
.
5
%
an
d
7
4
.
3
%,
r
esp
ec
tiv
ely
.
T
h
e
lin
ea
r
n
atu
r
e
o
f
t
h
is
m
o
d
el
lim
ited
its
ab
ilit
y
to
ca
p
tu
r
e
n
o
n
lin
ea
r
p
atter
n
s
in
t
h
e
d
ata,
wh
ich
led
to
m
o
r
e
er
r
o
r
s
in
class
if
y
in
g
f
atig
u
e
lev
el
s
.
T
h
u
s
,
am
o
n
g
th
e
test
ed
alg
o
r
i
th
m
s
,
th
e
Dec
is
io
n
T
r
ee
d
em
o
n
s
tr
ated
th
e
b
est
b
alan
ce
d
p
er
f
o
r
m
an
ce
,
as
s
h
o
wn
in
T
ab
le
1
,
f
o
llo
we
d
b
y
RF
an
d
XGBo
o
s
t
with
v
er
y
clo
s
e
r
esu
lts
.
L
o
g
is
tic
r
eg
r
ess
io
n
p
r
o
v
e
d
to
b
e
th
e
least
ef
f
ec
tiv
e
f
o
r
th
e
class
if
icatio
n
task
.
All
m
o
d
el
r
esu
lts
ar
e
p
r
esen
ted
in
class
if
icatio
n
r
ep
o
r
ts
an
d
m
etr
ics,
en
ab
lin
g
co
n
clu
s
io
n
s
ab
o
u
t
th
e
p
r
ac
tical
ap
p
licab
ilit
y
o
f
ea
ch
m
o
d
el
f
o
r
f
atig
u
e
an
aly
s
is
b
ased
o
n
E
MG
d
ata.
T
ab
le
1
.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
m
ac
h
in
e
lear
n
in
g
m
o
d
els wa
s
ev
alu
ated
f
o
r
class
if
y
in
g
m
u
s
c
le
f
atig
u
e
u
s
in
g
E
MG
s
ig
n
als
M
o
d
e
l
A
c
c
u
r
a
c
y
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
R
a
n
d
o
m
f
o
r
e
s
t
8
9
%
8
7
%
8
9
%
8
9
%
D
e
c
i
s
i
o
n
t
r
e
e
9
0
%
9
0
%
9
0
%
9
0
%
X
G
B
o
o
st
8
9
%
8
9
%
8
9
%
8
9
%
Lo
g
i
s
t
i
c
r
e
g
r
e
ssi
o
n
7
4
%
7
3
%
7
4
%
7
3
%
Fig
u
r
e
4
s
h
o
ws
th
e
co
n
f
u
s
io
n
m
atr
ix
f
o
r
th
e
f
atig
u
e
lev
el
class
if
icatio
n
m
o
d
el
with
th
r
ee
class
e
s
:
lo
w
(
0
)
,
m
ed
i
u
m
(
1
)
,
an
d
h
ig
h
(
2
)
.
T
h
e
m
atr
ix
in
d
icate
s
th
at
th
e
m
o
d
el
co
r
r
ec
tly
class
if
ied
7
6
in
s
tan
ce
s
o
f
lo
w
f
atig
u
e
lev
el,
with
1
in
s
t
an
ce
m
is
class
if
ied
as
m
ed
iu
m
an
d
2
as
h
i
g
h
.
Fo
r
t
h
e
m
e
d
iu
m
f
atig
u
e
le
v
el,
7
6
in
s
tan
ce
s
wer
e
co
r
r
ec
tly
id
en
tifie
d
,
wh
ile
9
wer
e
m
is
class
if
ied
as
lo
w
an
d
7
as
h
ig
h
.
R
eg
ar
d
in
g
th
e
h
i
g
h
f
atig
u
e
lev
el,
5
9
in
s
tan
ce
s
wer
e
co
r
r
ec
tly
class
if
ied
,
with
2
a
n
d
4
i
n
s
tan
ce
s
m
is
class
if
ied
a
s
lo
w
an
d
m
ed
i
u
m
,
r
esp
ec
tiv
ely
.
T
h
e
co
n
f
u
s
io
n
m
atr
ix
d
em
o
n
s
tr
ates
th
at
th
e
m
o
d
el
p
e
r
f
o
r
m
s
well,
with
m
o
s
t
m
is
class
if
ica
tio
n
s
o
cc
u
r
r
in
g
b
etwe
en
a
d
jace
n
t
f
atig
u
e
lev
els,
h
ig
h
lig
h
ti
n
g
t
h
e
ch
allen
g
e
o
f
p
r
ec
is
ely
d
is
tin
g
u
is
h
in
g
clo
s
ely
r
elate
d
class
es.
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
.
6
,
Decem
b
e
r
20
25
:
5
9
5
4
-
5
9
6
7
5962
Fig
u
r
e
5
s
h
o
ws
th
e
f
ea
tu
r
e
im
p
o
r
tan
ce
o
f
th
e
DT
m
o
d
el.
T
h
e
ch
ar
t
d
is
p
lay
s
th
e
r
elativ
e
in
f
lu
en
ce
o
f
ea
ch
f
ea
tu
r
e
o
n
th
e
m
o
d
el'
s
p
r
ed
ictio
n
s
.
T
h
e
m
o
s
t
im
p
o
r
tan
t
f
ea
tu
r
e
is
tim
e,
f
o
llo
wed
b
y
S
m
o
k
e
an
d
Alco
h
o
l
,
in
d
icatin
g
th
eir
s
ig
n
if
ican
t
i
m
p
ac
t
o
n
t
h
e
p
r
e
d
ictio
n
o
u
tc
o
m
es.
Oth
er
im
p
o
r
tan
t
f
ea
t
u
r
es
in
clu
d
e
Ag
e
an
d
s
p
o
r
t,
wh
ich
also
h
a
v
e
a
n
o
ticea
b
le
ef
f
ec
t o
n
th
e
m
o
d
el.
Fig
u
r
e
4
.
C
o
n
f
u
s
io
n
m
atr
i
x
Fig
u
r
e
5
.
Featu
r
e
i
m
p
o
r
ta
n
ce
s
I
t
is
ev
id
en
t
th
at
tim
e
h
as
a
s
tr
o
n
g
in
f
lu
en
ce
,
as
th
e
lo
n
g
er
th
e
lo
ad
is
ap
p
lied
,
th
e
g
r
ea
ter
t
h
e
f
atig
u
e
ex
p
er
ien
ce
d
—
an
o
b
s
er
v
atio
n
th
at
alig
n
s
with
th
e
k
n
o
w
n
p
h
y
s
io
lo
g
ical
ac
cu
m
u
latio
n
o
f
n
eu
r
o
m
u
s
cu
lar
f
atig
u
e
d
u
r
in
g
s
u
s
tain
ed
ex
er
tio
n
[
5
2
]
.
B
eh
av
io
r
al
f
ac
to
r
s
,
in
clu
d
in
g
s
m
o
k
i
n
g
an
d
alco
h
o
l
co
n
s
u
m
p
tio
n
,
also
em
er
g
ed
as
s
ig
n
if
ican
t
p
r
ed
icto
r
s
,
in
d
icatin
g
th
at
life
s
ty
le
h
ab
its
m
ay
co
n
tr
ib
u
te
m
ea
n
in
g
f
u
lly
to
th
e
d
ev
elo
p
m
e
n
t
an
d
p
er
ce
p
tio
n
o
f
f
atig
u
e.
L
ess
im
p
o
r
tan
t
f
ea
tu
r
es
in
clu
d
e
R
MS,
Gen
d
er
,
an
d
MD
F
(
Hz)
,
wh
ich
h
av
e
lo
wer
r
elativ
e
im
p
o
r
tan
ce
v
alu
es.
T
h
ese
r
esu
lts
co
n
f
ir
m
th
at
tim
e,
alo
n
g
with
ce
r
tain
b
io
lo
g
ical
an
d
b
eh
av
io
r
al
p
ar
a
m
eter
s
,
p
lay
s
a
k
ey
r
o
le
i
n
th
e
Dec
is
io
n
T
r
ee
m
o
d
el’
s
p
r
e
d
ictio
n
s
.
T
h
e
in
clu
s
io
n
o
f
b
o
t
h
p
h
y
s
io
lo
g
ical
an
d
b
eh
a
v
io
r
al
v
ar
iab
les
en
ab
les
th
e
m
o
d
el
to
r
ep
r
esen
t
f
atig
u
e
m
o
r
e
co
m
p
r
eh
e
n
s
iv
ely
,
o
f
f
er
in
g
m
ea
n
in
g
f
u
l
g
u
id
a
n
ce
f
o
r
th
e
d
esig
n
o
f
in
d
iv
i
d
u
alize
d
r
eh
ab
ilit
atio
n
p
r
o
to
c
o
ls
.
T
h
e
s
tu
d
y
aim
ed
to
d
ev
el
o
p
a
r
ea
l
-
tim
e,
I
o
T
-
e
n
ab
led
s
y
s
tem
f
o
r
class
if
y
in
g
lo
ca
l
m
u
s
cle
f
atig
u
e
u
s
in
g
s
u
r
f
ac
e
E
MG
s
ig
n
als,
b
eh
av
io
r
al
co
n
tex
t,
an
d
s
u
b
jectiv
e
f
atig
u
e
p
er
ce
p
tio
n
.
T
h
e
g
o
al
was
to
im
p
r
o
v
e
r
eh
ab
ilit
atio
n
o
u
tco
m
es
th
r
o
u
g
h
m
o
r
e
ac
cu
r
ate
an
d
p
e
r
s
o
n
alize
d
f
atig
u
e
d
etec
tio
n
.
A
k
ey
m
eth
o
d
o
lo
g
ical
in
n
o
v
atio
n
was
th
e
u
s
e
o
f
s
u
b
jectiv
e
f
atig
u
e
r
atin
g
s
as
th
e
tar
g
et
v
ar
iab
le,
p
r
ed
icted
f
r
o
m
a
co
m
b
in
atio
n
o
f
o
b
jectiv
e
E
MG
s
ig
n
al
f
ea
tu
r
es
an
d
b
eh
av
io
r
al
f
ac
to
r
s
s
u
ch
as
s
m
o
k
in
g
an
d
alco
h
o
l
co
n
s
u
m
p
tio
n
.
T
h
ese
f
ea
tu
r
es
wer
e
ex
tr
ac
ted
d
ir
ec
tl
y
o
n
a
wea
r
ab
le
d
ev
ice
(
Qim
y
l)
an
d
tr
an
s
m
itted
wir
eless
ly
f
o
r
a
n
aly
s
is
.
Sev
er
al
s
u
p
er
v
is
ed
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
—
LR
,
DT
,
RF
,
a
n
d
XGBo
o
s
t
—
wer
e
ap
p
lied
to
class
if
y
f
atig
u
e
s
tates,
with
th
e
DT
m
o
d
el
ac
h
iev
in
g
th
e
b
est
p
e
r
f
o
r
m
an
ce
a
t
9
0
.
7
%
ac
cu
r
ac
y
.
T
h
e
co
n
ce
p
tu
al
n
o
v
elty
lies
in
s
y
n
ch
r
o
n
izi
n
g
p
h
y
s
io
lo
g
ical
a
n
d
b
eh
a
v
io
r
al
in
p
u
ts
to
m
o
d
el
s
u
b
jectiv
e
f
atig
u
e
in
r
ea
l
tim
e
,
en
ab
lin
g
ef
f
icien
t
on
-
d
e
v
ice
p
r
o
ce
s
s
in
g
an
d
p
r
ac
tical
d
ep
lo
y
m
en
t
f
o
r
a
d
ap
tiv
e
t
eler
eh
ab
ilit
atio
n
.
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
Ma
ch
in
e
lea
r
n
in
g
-
b
a
s
ed
cla
s
s
ifica
tio
n
o
f lo
ca
l m
u
s
cle
fa
tig
u
e
…
(
Zh
a
n
el
B
a
ig
a
r
a
ye
v
a
)
5963
E
x
is
tin
g
s
tu
d
ies
o
n
m
u
s
cle
f
at
ig
u
e
ass
ess
m
en
t
h
av
e
p
r
ed
o
m
i
n
an
tly
ad
o
p
ted
o
b
jectiv
e
m
eth
o
d
o
lo
g
ies
b
ased
o
n
s
u
r
f
ac
e
E
MG
,
wit
h
lim
ited
co
n
s
id
er
atio
n
o
f
s
u
b
jectiv
e
p
er
ce
p
tio
n
s
.
On
e
lin
e
o
f
r
esear
ch
h
as
em
p
lo
y
ed
c
o
n
tin
u
o
u
s
wav
elet
tr
an
s
f
o
r
m
s
[
5
3
]
to
a
n
aly
ze
elec
tr
o
p
h
y
s
io
lo
g
ical
r
esp
o
n
s
es
ac
r
o
s
s
v
ar
y
in
g
ex
er
cise
in
ten
s
ities
,
f
o
cu
s
in
g
o
n
f
ea
tu
r
es
s
u
ch
as
f
r
eq
u
en
cy
d
r
o
p
-
o
f
f
s
,
b
u
r
s
t
p
o
wer
,
an
d
ar
ea
ch
an
g
es
to
in
f
e
r
lo
ca
lized
f
atig
u
e,
with
o
u
t
in
te
g
r
atin
g
p
a
r
ticip
an
t
-
r
e
p
o
r
ted
e
x
er
tio
n
.
Oth
er
in
v
esti
g
atio
n
s
h
av
e
u
s
ed
tim
e
a
n
d
f
r
eq
u
e
n
cy
d
o
m
ain
f
ea
tu
r
es
—
s
u
ch
as
m
e
d
ian
a
n
d
m
ea
n
f
r
eq
u
en
cy
s
h
if
ts
,
r
o
o
t
m
ea
n
s
q
u
ar
e
(
R
MS)
,
an
d
m
ea
n
ab
s
o
lu
te
v
alu
e
(
MA
V)
—
to
cla
s
s
if
y
f
atig
u
e
s
tates,
in
clu
d
in
g
p
r
o
lo
n
g
ed
f
atig
u
e
i
n
d
u
ce
d
b
y
tr
ea
d
m
ill
p
r
o
t
o
co
ls
[
5
4
]
.
W
h
ile
th
ese
ap
p
r
o
ac
h
es
d
em
o
n
s
tr
ated
h
ig
h
class
if
icatio
n
ac
cu
r
ac
y
u
s
in
g
m
ac
h
in
e
le
ar
n
in
g
alg
o
r
ith
m
s
,
th
ey
ass
ess
ed
f
atig
u
e
s
o
lely
th
r
o
u
g
h
ch
a
n
g
es
in
s
ig
n
al
p
r
o
p
e
r
ties
.
R
o
b
o
tic
p
r
o
t
o
co
ls
h
av
e
also
b
ee
n
p
r
o
p
o
s
ed
[
5
5
]
t
o
o
b
jectiv
ely
d
eter
m
in
e
t
h
e
o
n
s
et
o
f
f
atig
u
e
u
s
in
g
s
u
r
f
a
ce
E
MG
f
r
eq
u
en
cy
m
etr
ics
an
d
esti
m
ated
en
er
g
y
ex
p
en
d
itu
r
e
d
u
r
in
g
r
esis
tiv
e
m
o
to
r
task
s
.
Similar
ly
,
s
tu
d
ie
s
an
aly
zin
g
is
o
to
n
ic
co
n
tr
ac
ti
o
n
s
h
a
v
e
q
u
an
tifie
d
f
atig
u
e
th
r
o
u
g
h
s
p
ec
tr
al
ch
a
n
g
es
in
E
MG
s
ig
n
als
u
n
d
e
r
v
ar
y
i
n
g
lo
a
d
co
n
d
itio
n
s
[
5
6
]
.
Desp
ite
th
eir
m
eth
o
d
o
l
o
g
ical
r
ig
o
r
,
th
ese
w
o
r
k
s
co
n
s
is
ten
tly
o
m
it r
ea
l
-
tim
e
o
r
s
tr
u
ctu
r
e
d
s
u
b
jectiv
e
f
atig
u
e
r
atin
g
s
.
W
h
ile
p
r
ev
io
u
s
wo
r
k
s
h
av
e
r
e
lied
s
o
lely
o
n
o
b
jectiv
e
E
MG
f
ea
tu
r
es
to
ev
alu
ate
m
u
s
cle
f
a
tig
u
e,
th
is
s
tu
d
y
m
o
d
els
s
u
b
jectiv
e
f
atig
u
e
p
er
ce
p
tio
n
as
th
e
tar
g
et
v
ar
iab
le,
u
s
in
g
a
s
tr
u
ctu
r
ed
n
u
m
er
ical
r
atin
g
s
ca
le
co
llected
d
u
r
i
n
g
th
e
task
,
wh
ile
em
p
lo
y
in
g
o
b
jectiv
e
E
MG
s
ig
n
al
f
ea
tu
r
es
an
d
c
o
n
tex
tu
al
b
eh
av
io
r
al
f
ac
to
r
s
—
s
u
ch
as
s
m
o
k
in
g
an
d
alco
h
o
l
co
n
s
u
m
p
tio
n
—
as
p
r
ed
ictiv
e
in
p
u
ts
.
T
h
is
p
ar
ad
i
g
m
s
h
if
t
en
ab
les
th
e
m
o
d
el
to
lea
r
n
h
o
w
p
er
ce
i
v
ed
f
atig
u
e
co
r
r
elate
s
with
p
h
y
s
io
l
o
g
ical
an
d
b
eh
a
v
io
r
al
in
d
icato
r
s
,
o
f
f
er
i
n
g
a
m
o
r
e
p
er
s
o
n
alize
d
an
d
p
s
y
ch
o
l
o
g
ic
ally
in
f
o
r
m
ed
a
p
p
r
o
ac
h
to
f
at
ig
u
e
m
o
n
ito
r
in
g
.
B
y
s
y
n
ch
r
o
n
izin
g
E
MG
f
ea
tu
r
e
ex
tr
ac
tio
n
with
p
ar
ticip
a
n
ts
’
r
ea
l
-
tim
e
s
elf
-
r
ep
o
r
ts
d
u
r
in
g
s
u
s
tain
ed
is
o
m
etr
ic
co
n
tr
a
ctio
n
s
,
th
e
s
y
s
te
m
en
h
an
ce
s
th
e
ec
o
lo
g
ical
v
alid
i
ty
o
f
f
atig
u
e
d
etec
tio
n
.
T
h
e
a
b
ilit
y
to
m
o
d
el
s
u
b
jectiv
e
s
tat
es
d
ir
ec
tly
—
r
ath
er
th
an
in
f
er
t
h
em
s
o
lely
f
r
o
m
p
h
y
s
io
lo
g
ical
th
r
esh
o
ld
s
—
s
u
p
p
o
r
ts
m
o
r
e
r
esp
o
n
s
iv
e
an
d
a
d
a
p
tiv
e
r
e
h
ab
ilit
atio
n
s
tr
ateg
ies.
T
h
e
p
latf
o
r
m
’
s
r
em
o
te
m
o
n
ito
r
in
g
ca
p
a
b
ilit
ies
also
s
u
p
p
o
r
t
teler
eh
a
b
ilit
atio
n
,
allo
win
g
h
ea
lth
ca
r
e
p
r
o
v
id
e
r
s
to
d
eliv
er
cu
s
to
m
ize
d
ca
r
e.
B
ey
o
n
d
its
im
m
e
d
iate
u
tili
ty
,
th
e
p
r
o
p
o
s
ed
f
r
am
ewo
r
k
o
f
f
er
s
s
u
b
s
tan
tial
lo
n
g
-
ter
m
p
o
ten
tial f
o
r
th
e
a
d
v
an
ce
m
en
t o
f
ad
ap
tiv
e
r
eh
ab
ilit
atio
n
an
d
p
er
s
o
n
alize
d
h
ea
lth
m
o
n
ito
r
in
g
.
A
k
ey
in
n
o
v
a
ti
o
n
o
f
th
i
s
s
tu
d
y
li
es
in
i
t
s
d
u
a
l
-
in
p
u
t
m
o
d
e
li
n
g
ap
p
r
o
a
ch
,
wh
i
ch
co
m
b
in
e
s
o
b
jec
t
iv
e
s
u
r
f
ac
e
E
M
G
s
ig
n
al
f
e
atu
r
es
wi
th
co
n
tex
tu
a
l
b
eh
av
io
r
al
v
a
r
i
ab
l
es
s
u
ch
a
s
s
m
o
k
in
g
an
d
a
lco
h
o
l
co
n
s
u
m
p
t
io
n
.
B
y
t
r
ai
n
in
g
m
ac
h
i
n
e
lea
r
n
in
g
m
o
d
e
l
s
to
p
r
ed
i
ct
s
u
b
je
ct
iv
e
f
a
tig
u
e
r
at
i
n
g
s
d
er
iv
ed
f
r
o
m
p
ar
t
ic
ip
an
t
s
e
lf
-
as
s
e
s
s
m
en
t
s
,
th
e
s
y
s
t
em
b
r
id
g
es
th
e
g
ap
b
et
wee
n
m
e
a
s
u
r
ab
le
p
h
y
s
io
l
o
g
ic
al
ac
tiv
ity
an
d
in
d
iv
id
u
al
p
er
ce
p
tu
a
l
ex
p
er
ien
c
e.
T
h
is
m
o
d
el
in
g
s
tr
a
t
eg
y
a
ck
n
o
wl
ed
g
e
s
f
a
tig
u
e
n
o
t
m
er
e
ly
a
s
a
b
io
m
e
ch
an
ic
al
p
h
en
o
m
e
n
o
n
b
u
t
a
s
a
m
u
l
t
id
im
en
s
io
n
al
co
n
s
tr
u
c
t
in
f
lu
en
ce
d
b
y
b
o
th
in
ter
n
al
an
d
ex
t
er
n
a
l
f
ac
to
r
s
.
A
s
s
u
ch
,
it
f
a
ci
l
it
at
e
s
th
e
d
ev
elo
p
m
en
t
o
f
r
e
al
-
t
i
m
e
c
lo
s
ed
-
lo
o
p
r
e
h
ab
il
it
at
io
n
s
y
s
tem
s
th
at
c
an
d
y
n
am
i
ca
ll
y
ta
ilo
r
ex
er
c
i
s
e
p
a
r
am
e
ter
s
—
in
c
lu
d
in
g
in
t
en
s
i
ty
,
d
u
r
a
t
io
n
,
an
d
r
ec
o
v
er
y
in
te
r
v
al
s
—
b
a
s
e
d
o
n
th
e
ev
o
lv
in
g
in
ter
p
lay
b
e
tw
ee
n
p
h
y
s
io
lo
g
ic
al
lo
ad
an
d
p
er
s
o
n
al
to
ler
an
c
e.
T
h
e
ab
i
l
ity
to
m
o
d
el
h
o
w
f
at
ig
u
e
u
n
f
o
ld
s
o
v
er
t
im
e
,
a
s
p
er
ce
iv
ed
b
y
th
e
in
d
iv
id
u
al,
en
ab
l
e
s
cl
in
i
ci
an
s
an
d
in
t
el
lig
en
t
s
y
s
t
em
s
to
d
et
ec
t
e
ar
l
y
war
n
in
g
s
ig
n
s
o
f
o
v
er
ex
er
tio
n
an
d
in
ter
v
en
e
p
r
ee
m
p
t
iv
e
ly
to
p
r
ev
en
t
in
ju
r
y
o
r
d
i
s
en
g
a
g
em
en
t.
Mo
r
eo
v
er
,
th
e
d
em
o
n
s
tr
a
ted
f
e
a
s
ib
il
i
ty
o
f
o
n
-
d
ev
ic
e
f
ea
tu
r
e
ex
tr
a
ct
io
n
an
d
cl
a
s
s
if
ica
t
io
n
u
s
in
g
em
b
ed
d
ed
ML
m
o
d
e
l
s
o
n
E
S
P3
2
a
r
ch
it
ec
tu
r
e
p
av
e
s
th
e
way
f
o
r
ed
g
e
-
AI
r
eh
ab
il
i
t
at
io
n
s
y
s
te
m
s
th
a
t
a
r
e
p
o
r
tab
l
e,
en
e
r
g
y
-
e
f
f
i
ci
en
t
,
an
d
d
ep
lo
y
ab
le
at
s
ca
le
in
h
o
m
e
s
e
tt
in
g
s
.
T
h
i
s
ed
g
e
-
f
ir
s
t
p
ar
ad
ig
m
a
l
lev
ia
te
s
r
e
l
ian
c
e
o
n
clo
u
d
co
n
n
e
ct
iv
ity
,
r
ed
u
c
es
la
ten
cy
,
an
d
p
r
o
t
ec
t
s
p
at
ien
t
p
r
iv
a
cy
—
k
ey
r
eq
u
i
r
em
en
t
s
f
o
r
w
id
e
s
p
r
ea
d
ad
o
p
t
io
n
in
te
ler
eh
ab
il
it
at
io
n
an
d
r
em
o
te
eld
e
r
ly
c
ar
e
ap
p
li
ca
tio
n
s
.
W
it
h
s
u
ch
a
s
y
s
t
em
,
r
e
al
-
t
im
e
f
e
e
d
b
ac
k
lo
o
p
s
ca
n
b
e
en
ab
led
ev
en
in
lo
w
-
r
e
s
o
u
r
ce
o
r
o
f
f
l
in
e
en
v
ir
o
n
m
en
t
s
,
wh
er
e
co
n
v
en
t
io
n
a
l
clo
u
d
-
b
a
s
ed
s
y
s
tem
s
m
ay
f
ai
l.
I
t
o
p
en
s
av
en
u
e
s
f
o
r
lo
n
g
itu
d
in
al
an
aly
s
i
s
,
wh
er
e
tr
en
d
s
in
f
at
ig
u
e
p
r
o
g
r
e
s
s
io
n
ca
n
b
e
an
aly
z
ed
o
v
e
r
d
ay
s
o
r
wee
k
s
,
y
ie
ld
in
g
n
e
w
b
io
m
a
r
k
er
s
f
o
r
th
er
ap
y
ef
f
ec
tiv
en
e
s
s
,
o
v
er
tr
a
in
i
n
g
,
o
r
r
i
s
k
o
f
r
el
a
p
s
e
.
T
h
e
p
r
e
s
en
ted
s
t
u
d
y
o
f
f
er
s
an
i
n
te
r
d
i
s
cip
li
n
ar
y
co
n
tr
ib
u
t
io
n
at
t
h
e
in
t
er
s
ec
tio
n
o
f
e
le
ct
r
i
ca
l
en
g
in
ee
r
in
g
,
em
b
ed
d
ed
s
y
s
t
em
s
,
I
o
T
te
ch
n
o
l
o
g
y
,
an
d
m
ac
h
in
e
l
ea
r
n
in
g
,
t
ar
g
e
ti
n
g
cr
i
ti
ca
l
n
e
ed
s
in
m
o
d
er
n
h
ea
lt
h
ca
r
e
ap
p
li
ca
ti
o
n
s
.
T
h
e
p
r
o
p
o
s
ed
lo
w
-
p
o
we
r
wea
r
ab
le
d
ev
ic
e
e
n
ab
le
s
o
n
-
d
ev
i
ce
E
MG
s
ig
n
al
p
r
o
c
es
s
in
g
an
d
f
a
tig
u
e
cla
s
s
if
i
ca
tio
n
,
in
teg
r
at
in
g
p
h
y
s
i
o
lo
g
ic
al
an
d
lif
e
s
ty
le
d
a
ta
f
o
r
p
e
r
s
o
n
a
li
ze
d
m
o
n
ito
r
in
g
.
Al
ig
n
ed
wi
th
cu
r
r
en
t
tr
en
d
s
in
s
m
ar
t
h
ea
lth
s
y
s
t
em
s
,
th
i
s
wo
r
k
co
n
t
r
ib
u
te
s
to
ad
v
an
c
in
g
ad
ap
tiv
e
r
eh
ab
il
i
ta
tio
n
te
ch
n
o
lo
g
i
e
s
an
d
o
f
f
er
s
a
p
r
a
ct
ic
al
f
r
a
m
e
wo
r
k
f
o
r
f
u
tu
r
e
r
e
s
ea
r
ch
an
d
cl
in
i
ca
l
ap
p
l
ic
at
io
n
s
.
Ho
wev
er
,
s
ev
er
al
lim
itatio
n
s
wer
e
en
co
u
n
ter
e
d
d
u
r
in
g
th
e
ex
p
e
r
im
en
t
th
at
s
u
g
g
est
f
u
r
th
er
im
p
r
o
v
em
e
n
t.
T
h
e
s
am
p
le
s
i
ze
was
lim
ited
to
ten
h
ea
lth
y
in
d
iv
id
u
als
ag
ed
1
9
to
2
2
.
wh
ich
r
estricts
th
e
g
en
er
aliza
b
ilit
y
o
f
th
e
r
esu
l
ts
to
d
iv
er
s
e
clin
ical
p
o
p
u
l
atio
n
s
,
s
u
ch
as
o
ld
er
ad
u
lts
o
r
p
atien
ts
with
n
eu
r
o
m
u
s
cu
lar
co
n
d
itio
n
s
.
T
h
e
ex
er
cise
p
r
o
to
co
l
last
ed
ap
p
r
o
x
im
ately
th
r
ee
m
in
u
tes
to
r
ef
lect
ty
p
ical
r
eh
ab
ilit
atio
n
s
ess
io
n
s
,
y
et
th
is
s
h
o
r
t
d
u
r
atio
n
r
estricte
d
th
e
ab
ilit
y
to
o
b
s
er
v
e
lo
n
g
er
-
ter
m
f
atig
u
e
p
atter
n
s
o
r
r
ec
o
v
er
y
b
e
h
av
io
r
.
T
h
e
s
y
s
tem
also
r
elied
o
n
s
u
b
jectiv
e
s
e
lf
-
ass
ess
m
en
ts
o
f
f
atig
u
e,
wh
i
ch
m
ay
v
ar
y
d
u
e
to
d
if
f
er
en
ce
s
i
n
in
d
iv
id
u
al
p
ai
n
th
r
esh
o
ld
,
p
er
ce
p
tio
n
,
o
r
co
m
m
u
n
icatio
n
s
ty
le,
p
o
ten
ti
ally
af
f
ec
tin
g
th
e
co
n
s
is
ten
cy
o
f
th
e
d
ata.
Evaluation Warning : The document was created with Spire.PDF for Python.