I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
41
,
No
.
2
,
Feb
r
u
ar
y
20
2
6
,
p
p
.
773
~
781
I
SS
N:
2
502
-
4
7
52
,
DOI
: 1
0
.
1
1
5
9
1
/ijee
cs
.v
41
.
i
2
.
pp
773
-
7
8
1
773
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs
.
ia
esco
r
e.
co
m
H
y
brid S
VM
–
AN
N
sy
stem f
o
r
auto
ma
ted
M
RI
dia
g
n
o
sis
of
a
nterior cru
cia
te
lig
a
ment inj
u
ries
Sa
zwa
n Sy
a
f
iq M
a
zla
n
1
,
Aziz
i Mi
s
k
o
n
1
,
Sh
a
riza
l A
hm
a
d So
bri
2
1
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
,
D
e
p
a
r
t
me
n
t
o
f
E
l
e
c
t
r
i
c
a
l
El
e
c
t
r
o
n
i
c
E
n
g
i
n
e
e
r
i
n
g
,
N
a
t
i
o
n
a
l
D
e
f
e
n
c
e
U
n
i
v
e
r
si
t
y
M
a
l
a
y
si
a
,
K
u
a
l
a
L
u
m
p
u
r
,
M
a
l
a
y
si
a
2
S
c
h
o
o
l
o
f
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
N
o
t
t
i
n
g
h
a
m
Tr
e
n
t
U
n
i
v
e
r
si
t
y
,
N
o
t
t
i
n
g
h
a
m
,
U
n
i
t
e
d
K
i
n
g
d
o
m
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
1
,
2
0
2
5
R
ev
is
ed
Dec
30
,
2
0
2
5
Acc
ep
ted
J
an
11
,
2
0
2
6
An
terio
r
c
r
u
c
iate
li
g
a
m
e
n
t
(A
CL)
tea
rs
a
re
a
fre
q
u
e
n
t
c
a
u
s
e
o
f
k
n
e
e
in
sta
b
il
it
y
,
y
e
t
m
a
g
n
e
ti
c
re
so
n
a
n
c
e
ima
g
in
g
(
M
RI
)
i
n
terp
re
tati
o
n
re
m
a
in
s
ti
m
e
-
c
o
n
su
m
in
g
a
n
d
o
b
se
rv
e
r
-
d
e
p
e
n
d
e
n
t.
Th
is
p
a
p
e
r
p
re
se
n
ts
a
n
a
u
to
m
a
ted
M
RI
fra
m
e
wo
rk
fo
r
ACL
in
j
u
r
y
sc
re
e
n
in
g
a
n
d
se
v
e
rit
y
g
ra
d
in
g
u
sin
g
a
h
y
b
rid
s
u
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
–
a
rti
ficia
l
n
e
u
ra
l
n
e
two
r
k
(S
V
M
–
AN
N)
m
o
d
e
l.
A
b
a
lan
c
e
d
d
a
tas
e
t
o
f
6
0
0
sa
g
it
tal
k
n
e
e
M
RI
ima
g
e
s
fro
m
Ho
sp
it
a
l
Taip
in
g
(
n
o
rm
a
l,
p
a
rti
a
l
tea
r,
c
o
m
p
lete
tea
r)
wa
s
sta
n
d
a
rd
ize
d
v
i
a
re
siz
in
g
,
re
g
io
n
-
of
-
i
n
tere
st
c
ro
p
p
in
g
,
c
o
n
tras
t
e
n
h
a
n
c
e
m
e
n
t,
n
o
ise
fil
t
e
rin
g
,
a
n
d
se
g
m
e
n
tatio
n
.
M
o
r
p
h
o
lo
g
ica
l
a
n
d
te
x
tu
re
fe
a
tu
re
s
we
re
e
x
tr
a
c
ted
a
n
d
re
d
u
c
e
d
u
sin
g
p
ri
n
c
ip
a
l
c
o
m
p
o
n
e
n
t
a
n
a
ly
sis
(
P
CA).
T
h
e
S
VM
p
e
r
fo
rm
s
th
e
in
it
ial
sc
re
e
n
in
g
(in
j
u
re
d
v
s.
n
o
n
-
in
ju
re
d
)
a
n
d
sa
m
p
les
p
re
d
icte
d
a
s
in
ju
re
d
a
re
p
a
ss
e
d
to
th
e
a
rti
ficia
l
n
e
u
ra
l
n
e
two
rk
(
ANN
)
to
c
las
sify
se
v
e
ri
ty
.
Us
in
g
c
o
n
fu
si
o
n
-
m
a
tri
x
a
n
d
re
c
e
iv
e
r
o
p
e
ra
ti
n
g
c
h
a
ra
c
teristic
(
ROC
)
e
v
a
lu
a
ti
o
n
,
th
e
p
ro
p
o
se
d
s
y
ste
m
a
c
h
iev
e
d
8
6
.
2
%
o
v
e
ra
ll
a
c
c
u
ra
c
y
a
n
d
8
1
.
7
%
se
n
siti
v
it
y
,
with
th
e
AN
N
re
a
c
h
in
g
a
p
p
r
o
x
ima
tely
9
5
%
a
c
c
u
ra
c
y
o
n
i
n
j
u
re
d
c
a
se
s
fo
rwa
rd
e
d
f
o
r
g
ra
d
in
g
.
A
c
li
n
icia
n
u
sa
b
il
it
y
su
r
v
e
y
i
n
d
ica
ted
h
i
g
h
a
c
c
e
p
tan
c
e
(~
9
5
%
),
su
p
p
o
r
ti
n
g
t
h
e
fe
a
sib
il
it
y
o
f
d
e
p
lo
y
m
e
n
t
a
s
a
li
g
h
twe
ig
h
t
d
e
c
isio
n
-
su
p
p
o
rt
t
o
o
l
.
Li
m
it
a
ti
o
n
s
in
c
lu
d
e
re
li
a
n
c
e
o
n
sin
g
le
sa
g
i
tt
a
l
slice
s
a
n
d
si
n
g
le
-
se
q
u
e
n
c
e
d
a
ta;
fu
tu
re
wo
r
k
will
i
n
c
o
r
p
o
ra
te
m
u
lt
i
-
sli
c
e
/3
D
a
n
d
m
u
lt
i
-
se
q
u
e
n
c
e
M
RI
to
imp
ro
v
e
s
e
n
siti
v
it
y
a
n
d
g
e
n
e
ra
li
z
a
b
il
it
y
.
K
ey
w
o
r
d
s
:
An
ter
io
r
cr
u
ciate
lig
a
m
en
t
Ar
tific
ial
n
eu
r
al
n
etwo
r
k
Dig
ital im
ag
e
p
r
o
ce
s
s
in
g
MRI
k
n
ee
im
ag
in
g
Or
th
o
p
ed
ic
in
j
u
r
y
d
etec
tio
n
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
:
Sh
ar
izal
Ah
m
ad
So
b
r
i
Sch
o
o
l o
f
Scien
ce
a
n
d
T
ec
h
n
o
lo
g
y
,
No
ttin
g
h
am
T
r
en
t U
n
iv
e
r
s
ity
No
ttin
g
h
am
NG1
1
8
NS,
Un
ited
Kin
g
d
o
m
E
m
ail: sh
ar
izal.
ah
m
ad
s
o
b
r
i@
n
tu
.
ac
.
u
k
1.
I
NT
RO
D
UCT
I
O
N
An
ter
io
r
cr
u
ciate
lig
am
en
t
(
A
C
L
)
in
ju
r
ies
ar
e
am
o
n
g
th
e
m
o
s
t
f
r
eq
u
en
t
s
er
io
u
s
lig
am
en
t
in
ju
r
ies
o
f
th
e
k
n
ee
in
s
p
o
r
ts
m
ed
icin
e
a
n
d
ar
e
a
m
ajo
r
ca
u
s
e
o
f
f
u
n
ctio
n
al
in
s
tab
ilit
y
wh
en
u
n
tr
ea
ted
.
Acc
u
r
ate
d
etec
tio
n
an
d
g
r
ad
in
g
a
r
e
clin
ically
im
p
o
r
tan
t
to
g
u
id
e
tim
ely
m
a
n
ag
em
en
t
an
d
to
r
ed
u
ce
s
ec
o
n
d
ar
y
m
e
n
is
ca
l
o
r
ch
o
n
d
r
al
d
a
m
ag
e.
Ma
g
n
etic
r
eso
n
an
ce
im
a
g
in
g
(
MRI
)
i
n
Fig
u
r
e
1
is
wid
ely
u
s
ed
to
e
v
alu
ate
AC
L
f
ib
e
r
co
n
tin
u
ity
,
s
ig
n
al
ch
a
n
g
es,
a
n
d
ass
o
ciate
d
in
tr
a
-
ar
ticu
lar
f
in
d
in
g
s
[
1
]
,
[
2
]
.
Fig
u
r
e
1
(
a)
i
s
k
n
ee
an
ato
m
y
o
f
AC
L
in
ju
r
y
an
d
Fig
u
r
e
1
(
b
)
i
s
ac
tu
al
MRI
k
n
ee
.
Alth
o
u
g
h
MRI
p
r
o
v
id
es
ex
ce
llen
t
s
o
f
t‑
tis
s
u
e
co
n
tr
ast
f
o
r
ev
alu
atin
g
AC
L
in
teg
r
ity
,
i
n
ter
p
r
etatio
n
s
till
d
ep
en
d
s
o
n
r
ea
d
er
ex
p
er
tis
e
an
d
ca
n
s
h
o
w
in
ter
‑
o
b
s
er
v
er
v
ar
iab
ilit
y
,
p
ar
ti
cu
lar
ly
f
o
r
p
ar
tial
tear
s
an
d
s
u
b
tle
s
ig
n
al
ab
n
o
r
m
alities
.
T
h
ese
lim
itatio
n
s
m
o
tiv
ate
co
m
p
u
ter
‑
aid
ed
d
iag
n
o
s
is
s
y
s
tem
s
th
at
ca
n
o
f
f
e
r
o
b
jectiv
e
a
n
d
r
ep
r
o
d
u
cib
le
d
ec
is
io
n
s
u
p
p
o
r
t [
1
]
,
[
3
]
,
[
4
]
.
I
n
m
an
y
clin
ical
s
ettin
g
s
,
MRI
is
in
ter
p
r
eted
b
y
r
ad
i
o
lo
g
y
s
er
v
ices
an
d
th
en
in
te
g
r
ated
with
o
r
th
o
p
e
d
ic
ass
ess
m
en
t.
T
h
is
m
u
lti‑step
wo
r
k
f
l
o
w
ca
n
co
n
t
r
ib
u
te
to
d
elay
s
a
n
d
v
ar
iab
ilit
y
in
g
r
ad
in
g
wh
e
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
41
,
No
.
2
,
Feb
r
u
a
r
y
20
2
6
:
7
7
3
-
7
8
1
774
ca
s
e
v
o
lu
m
es
a
r
e
h
i
g
h
o
r
wh
en
ex
p
er
tis
e
is
lim
ited
.
Au
t
o
m
atin
g
p
a
r
ts
o
f
th
e
im
ag
e‑
b
a
s
ed
ass
ess
m
en
t
m
ay
h
elp
s
tan
d
ar
d
ize
g
r
ad
in
g
an
d
s
h
o
r
ten
th
e
tim
e
to
a
p
r
eli
m
in
ar
y
r
e
p
o
r
t,
esp
ec
ially
i
n
r
eso
u
r
ce
‑
co
n
s
tr
ain
ed
en
v
ir
o
n
m
en
ts
[
5
]
.
R
ec
en
t
r
e
s
ea
r
ch
h
as
d
em
o
n
s
tr
ated
s
tr
o
n
g
p
er
f
o
r
m
a
n
ce
f
o
r
AC
L
tear
d
etec
tio
n
u
s
in
g
m
u
lti‑seq
u
en
ce
r
ad
io
m
ics
a
n
d
m
o
d
e
r
n
m
ac
h
in
e
lear
n
in
g
.
Fo
r
ex
am
p
le,
C
h
en
g
et
a
l.
[
3
]
u
s
ed
m
u
lti‑seq
u
e
n
ce
(
T
1
‑
weig
h
ted
an
d
PD‑
weig
h
t
ed
)
MRI
r
a
d
io
m
ics
with
a
n
s
u
p
p
o
r
t
v
ec
t
o
r
m
ac
h
in
e
(
SVM
)
class
if
ier
an
d
r
ep
o
r
ted
h
ig
h
ar
ea
u
n
d
e
r
th
e
cu
r
v
e
(
AUC
)
,
s
en
s
itiv
ity
,
an
d
s
p
ec
if
icity
.
Dee
p
lear
n
in
g
a
p
p
r
o
ac
h
es
h
a
v
e
also
b
ee
n
a
p
p
lied
to
AC
L
tear
d
e
tectio
n
an
d
lo
ca
lizatio
n
o
n
k
n
ee
MRI
[
3
]
.
Ho
wev
er
,
m
an
y
h
ig
h
‑
p
e
r
f
o
r
m
in
g
m
eth
o
d
s
r
e
q
u
ir
e
m
u
lti‑seq
u
e
n
ce
o
r
m
u
lti‑sli
ce
/3
D
in
p
u
ts
,
in
ten
s
iv
e
f
ea
t
u
r
e
e
x
tr
ac
tio
n
,
an
d
s
u
b
s
tan
tial
co
m
p
u
tin
g
r
eso
u
r
ce
s
,
wh
ich
c
an
h
in
d
er
r
ap
id
d
ep
l
o
y
m
en
t
o
n
lig
h
tweig
h
t
p
l
atf
o
r
m
s
an
d
li
m
it
in
ter
p
r
etab
ilit
y
[
6
]
-
[
8
]
.
T
h
is
s
tu
d
y
tar
g
ets
a
c
o
m
p
lem
en
tar
y
n
ich
e
b
y
p
r
o
p
o
s
in
g
a
f
u
lly
au
to
m
ated
,
co
m
p
u
tatio
n
ally
ef
f
icien
t
p
ip
elin
e
th
at
u
s
es
s
tan
d
ar
d
iz
ed
p
r
ep
r
o
ce
s
s
in
g
,
in
ter
p
r
etab
le
m
o
r
p
h
o
m
etr
ic
f
ea
t
u
r
es,
p
r
in
cip
al
co
m
p
o
n
en
t
an
aly
s
is
(
PC
A
)
f
o
r
d
im
en
s
io
n
ality
r
ed
u
ctio
n
,
a
n
d
a
h
y
b
r
i
d
S
VM
–
ar
tific
ial
n
eu
r
al
n
etwo
r
k
(
ANN
)
ar
ch
itectu
r
e
to
i)
s
cr
ee
n
f
o
r
in
ju
r
y
an
d
ii)
g
r
ad
e
s
ev
er
ity
.
Acc
o
r
d
in
g
ly
,
t
h
e
s
y
s
tem
is
ev
alu
ated
u
s
in
g
co
n
f
u
s
io
n
m
atr
ices,
r
ec
eiv
er
o
p
er
atin
g
ch
a
r
ac
ter
is
tic
(
R
OC
)
an
aly
s
i
s
,
an
d
clin
i
ca
lly
r
elev
an
t
m
etr
ics
(
ac
c
u
r
a
cy
,
s
en
s
itiv
ity
,
an
d
s
p
ec
if
icity
)
.
T
h
e
r
esu
lts
s
ec
tio
n
r
ep
o
r
ts
b
o
th
th
e
s
cr
ee
n
in
g
p
er
f
o
r
m
a
n
ce
(
in
ju
r
y
v
s
n
o
n
‑
in
j
u
r
y
)
a
n
d
t
h
e
g
r
a
d
in
g
p
er
f
o
r
m
an
ce
(
n
o
r
m
al/p
a
r
tial/co
m
p
lete
tear
)
[
9
]
.
C
lin
ically
,
AC
L
in
ju
r
y
ass
ess
m
en
t
ty
p
ically
co
m
b
in
es
p
atien
t
h
is
to
r
y
,
p
h
y
s
ical
ex
am
in
atio
n
(
e.
g
.
,
L
ac
h
m
an
,
p
i
v
o
t‑
s
h
if
t,
a
n
d
an
ter
io
r
d
r
awe
r
test
s
)
an
d
MRI
co
n
f
ir
m
atio
n
.
No
r
m
al
AC
L
s
o
f
ten
ap
p
ea
r
as
a
co
n
tin
u
o
u
s
lo
w‑
s
ig
n
al
b
an
d
,
wh
er
ea
s
p
ar
tial
tear
s
m
ay
s
h
o
w
wav
in
ess
,
in
cr
ea
s
ed
s
ig
n
al,
o
r
f
o
ca
l f
i
b
er
d
is
r
u
p
tio
n
; c
o
m
p
lete
tear
s
s
h
o
w
d
is
co
n
tin
u
ity
o
r
n
o
n
‑
v
is
u
ali
za
tio
n
[
1
]
.
(
a)
(
b
)
Fig
u
r
e
1
.
MRI
is
wid
ely
u
s
ed
t
o
(
a)
k
n
ee
an
at
o
m
y
o
f
AC
L
in
ju
r
y
an
d
(
b
)
ac
tu
al
MRI
k
n
ee
Dig
ital
im
ag
e
p
r
o
ce
s
s
in
g
h
as
em
er
g
ed
as
a
cr
itical
e
n
ab
le
r
o
f
au
to
m
ated
m
ed
ical
im
ag
e
an
aly
s
is
b
ec
au
s
e
it
allo
ws
s
y
s
tem
atic
e
n
h
an
ce
m
e
n
t
an
d
s
eg
m
en
tatio
n
o
f
an
ato
m
ical
s
tr
u
ctu
r
es.
B
asic
o
p
er
atio
n
s
s
u
ch
as
r
esizin
g
,
cr
o
p
p
in
g
,
c
o
n
tr
ast
ad
ju
s
tm
en
t,
f
ilter
in
g
,
th
r
esh
o
l
d
in
g
a
n
d
b
in
ar
y
in
v
er
s
io
n
ar
e
u
s
ed
to
s
tan
d
ar
d
ize
in
p
u
t
im
ag
es
an
d
p
r
ep
a
r
e
f
o
r
s
u
b
s
eq
u
en
t
s
eg
m
en
tatio
n
an
d
f
ea
tu
r
e
ex
tr
ac
tio
n
[
9
]
,
[
1
0
]
.
I
n
AC
L
M
R
I
s
tu
d
ies,
th
ese
m
eth
o
d
s
ar
e
ap
p
lied
to
i
s
o
late
th
e
lig
am
en
t f
r
o
m
s
u
r
r
o
u
n
d
in
g
b
o
n
e
,
ca
r
tilag
e
an
d
s
o
f
t tis
s
u
e
s
,
u
s
u
ally
b
y
f
o
cu
s
in
g
o
n
a
r
eg
io
n
o
f
in
t
er
est
an
d
th
en
p
er
f
o
r
m
in
g
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
to
s
h
ar
p
en
s
tr
u
ctu
r
a
l
b
o
u
n
d
ar
ies
[1
1
]
,
[
1
2
]
.
T
h
e
wo
r
k
o
f
Ma
zla
n
an
d
co
lleag
u
es
d
escr
ib
es
a
co
m
p
r
e
h
en
s
iv
e
p
i
p
elin
e
th
at
in
clu
d
es
r
esizin
g
to
a
s
tan
d
ar
d
f
o
r
m
at,
cr
o
p
p
in
g
to
a
r
eg
io
n
d
e
ter
m
in
ed
co
llectiv
ely
b
y
m
e
d
ical
ex
p
er
ts
an
d
co
m
p
u
tatio
n
al
an
aly
s
is
,
co
n
tr
ast
en
h
an
ce
m
e
n
t
to
co
r
r
ec
t
d
ar
k
s
ca
n
s
a
n
d
n
o
is
e
s
u
p
p
r
ess
io
n
u
s
in
g
m
e
d
ian
o
r
av
er
ag
e
f
ilter
s
,
f
o
llo
wed
b
y
s
eg
m
en
tatio
n
u
s
in
g
d
ilatio
n
,
e
r
o
s
io
n
,
b
o
u
n
d
ar
y
t
r
ac
in
g
an
d
r
eg
io
n
-
p
r
o
p
ag
atio
n
tech
n
iq
u
es.
T
h
is
p
r
o
g
r
ess
io
n
r
ef
lects
a
g
en
er
al
co
n
s
en
s
u
s
in
th
e
liter
atu
r
e
th
at
a
r
o
b
u
s
t
p
r
e
-
p
r
o
ce
s
s
in
g
s
tag
e
is
ess
en
tial f
o
r
an
y
s
u
cc
ess
f
u
l a
u
to
m
ated
AC
L
d
iag
n
o
s
is
s
y
s
tem
.
B
ey
o
n
d
p
r
ep
r
o
ce
s
s
in
g
,
s
eg
m
e
n
tatio
n
an
d
f
ea
tu
r
e
e
x
tr
ac
tio
n
ar
e
ce
n
tr
al
to
d
ig
ital
in
ter
p
r
etatio
n
o
f
AC
L
im
ag
es.
An
ato
m
ical
s
tu
d
ies
h
av
e
r
ep
o
r
ted
co
n
s
id
er
ab
le
v
ar
iatio
n
in
AC
L
s
h
ap
e
an
d
tex
tu
r
e,
p
r
o
m
p
tin
g
ca
lls
f
o
r
q
u
an
titativ
e
an
aly
s
is
o
f
lig
am
e
n
t
m
o
r
p
h
o
lo
g
y
r
ath
e
r
th
an
s
o
lely
v
is
u
al
in
s
p
ec
tio
n
[
1
3
]
.
R
esear
ch
h
as
s
h
o
wn
th
at
p
ar
am
ete
r
s
s
u
ch
a
s
lig
am
en
t
s
ize,
ar
ea
,
o
r
ien
tati
o
n
an
g
le
a
n
d
elo
n
g
atio
n
co
r
r
e
late
with
d
if
f
er
en
t
in
ju
r
y
lev
els,
wh
er
e
s
ize
o
f
ten
s
ep
ar
ates
in
tact
f
r
o
m
s
ev
er
ely
d
am
ag
ed
lig
am
en
ts
,
ar
ea
d
is
tin
g
u
is
h
es
p
ar
tial
f
r
o
m
n
o
r
m
al
in
j
u
r
ies,
an
g
le
r
e
f
lects
co
m
p
lete
tear
s
an
d
elo
n
g
atio
n
in
d
icate
s
th
e
d
e
g
r
ee
o
f
s
tr
etch
in
g
o
r
f
ib
er
d
is
r
u
p
tio
n
[
3
]
,
[
1
4
]
,
[
1
5
]
.
T
h
ese
in
s
ig
h
ts
s
u
p
p
o
r
t
th
e
ex
tr
ac
tio
n
o
f
s
h
ap
e
d
escr
ip
to
r
s
as
co
r
e
f
ea
tu
r
es
f
o
r
m
ac
h
in
e
-
lear
n
i
n
g
-
b
ased
AC
L
class
if
ier
s
[1
6
]
-
[
18
]
as sh
o
ws in
F
ig
u
r
e
2.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
Hyb
r
id
S
V
M
–
A
N
N
s
ys
tem
fo
r
a
u
to
ma
ted
mri
d
ia
g
n
o
s
is
o
f a
n
teri
o
r
…
(
S
a
z
w
a
n
S
ya
fiq
Ma
z
la
n
)
775
Fig
u
r
e
2
.
Var
iatio
n
s
h
ap
e
o
f
th
e
k
n
ee
[
1
3
]
Ma
ch
in
e
lear
n
in
g
,
p
ar
ticu
lar
ly
s
u
p
er
v
is
ed
lear
n
in
g
,
is
n
o
w
f
i
r
m
ly
estab
lis
h
ed
in
th
e
f
ield
o
f
m
ed
ical
im
ag
e
an
aly
s
is
[
4
]
.
T
ec
h
n
iq
u
es
s
u
ch
as
SVM
,
ANN
,
lo
g
is
tic
r
eg
r
ess
io
n
an
d
d
ec
is
io
n
tr
ee
s
h
av
e
b
ee
n
ex
ten
s
iv
ely
u
s
ed
f
o
r
class
if
icatio
n
task
s
in
v
o
lv
in
g
tu
m
o
r
s
,
o
r
g
an
s
eg
m
en
tatio
n
an
d
lesi
o
n
d
etec
tio
n
[
19
]
-
[2
1
]
.
I
n
th
e
c
o
n
tex
t
o
f
AC
L
,
Ma
z
lan
’
s
ea
r
ly
wo
r
k
in
tr
o
d
u
ce
d
f
u
zz
y
-
in
f
er
en
ce
s
y
s
tem
s
u
s
in
g
r
u
le
-
b
ased
lo
g
ic
d
er
iv
ed
f
r
o
m
ex
p
er
t
k
n
o
wled
g
e
f
o
r
MRI
-
b
ased
class
if
icatio
n
,
illu
s
tr
atin
g
th
e
f
ea
s
ib
ilit
y
o
f
au
to
m
atin
g
AC
L
d
iag
n
o
s
is
[
19
]
,
[
2
0
]
.
L
ater
s
tu
d
ies
ex
ten
d
ed
th
is
ap
p
r
o
ac
h
to
em
b
r
ac
e
f
u
lly
d
ata
-
d
r
iv
e
n
m
eth
o
d
s
,
in
clu
d
in
g
SVM
-
b
ased
class
if
icat
io
n
an
d
co
m
p
ar
ativ
e
an
aly
s
es
o
f
em
b
ed
d
ed
s
y
s
tem
s
f
o
r
AC
L
MR
I
d
iag
n
o
s
is
,
wh
ich
ex
am
in
ed
t
h
e
f
ea
s
ib
ilit
y
o
f
im
p
lem
en
tin
g
AC
L
d
ia
g
n
o
s
tic
to
o
ls
o
n
d
i
f
f
er
en
t
h
ar
d
wa
r
e
p
latf
o
r
m
s
[2
1
]
.
Fu
r
th
er
wo
r
k
p
r
o
p
o
s
ed
a
co
m
p
lete
AC
L
d
iag
n
o
s
is
s
y
s
tem
u
s
in
g
im
ag
e
p
r
o
ce
s
s
in
g
an
d
SVM,
wh
er
e
th
e
au
t
h
o
r
s
r
ep
o
r
ted
im
p
r
o
v
ed
p
er
f
o
r
m
a
n
ce
an
d
p
r
o
v
id
ed
a
s
tep
p
i
n
g
s
to
n
e
to
war
d
s
h
y
b
r
id
ar
c
h
itectu
r
es
th
at
in
teg
r
ate
m
o
r
e
th
an
o
n
e
class
if
ier
[1
7
]
,
[
2
1
]
,
[
2
2
]
as in
(
1
)
an
d
(
2
).
{
(
1
,
1
)
,
(
2
,
2
)
,
(
3
,
3
)
…
,
(
,
)
}
(
1
)
(
,
)
+
=
0
(
2
)
T
h
e
p
ar
am
eter
is
s
ca
lar
,
wh
ile
w
is
d
im
en
s
io
n
al.
T
h
e
v
e
cto
r
w
co
o
r
d
in
ate
is
p
er
p
en
d
icu
lar
to
s
ep
ar
atin
g
p
lan
e
an
d
th
e
o
f
f
s
et
p
ar
am
eter
b
is
ad
d
ed
t
o
in
cr
ea
s
e
th
e
m
ar
g
in
g
ap
.
R
em
o
v
i
n
g
p
ar
am
ete
r
b
,
will
ca
u
s
e
th
e
p
lan
e
g
o
in
g
th
r
o
u
g
h
th
e
o
r
ig
in
(
0
,
0
)
p
o
i
n
t
an
d
r
estrict
an
y
s
o
lu
tio
n
.
T
h
e
p
ar
a
llel
p
lan
es
ca
n
b
e
d
escr
ib
e
d
in
(
3
)
a
n
d
(
4
)
[
23
].
.
+
=
1
(
3
)
.
+
=
−
1
(
4
)
T
h
e
co
n
d
itio
n
o
f
tr
ain
in
g
d
at
a
m
u
s
t
b
e
s
ep
ar
a
b
le,
in
o
r
d
er
f
o
r
t
h
o
s
e
p
lan
es
to
h
av
e
n
o
i
n
ter
s
ec
t
p
o
in
ts
an
d
m
ax
im
ize
th
eir
d
is
tan
ce
.
I
n
g
eo
m
etr
y
,
d
eter
m
in
e
th
e
d
is
tan
ce
b
etwe
en
th
o
s
e
p
lan
es
r
ep
r
esen
t
as
2
|
|
in
m
in
im
izin
g
|
w|
.
2.
M
AT
E
R
I
AL
A
ND
M
E
T
H
O
D
T
h
e
m
eth
o
d
o
lo
g
y
o
f
th
is
s
tu
d
y
is
d
esig
n
ed
to
co
n
s
tr
u
ct
a
f
u
lly
au
to
m
ated
d
iag
n
o
s
tic
f
r
am
ewo
r
k
f
o
r
AC
L
in
ju
r
ies
u
s
in
g
MRI
an
d
to
r
ef
lect
t
h
e
s
tep
s
b
y
w
h
ich
clin
ician
s
ass
ess
s
u
ch
i
n
ju
r
ies
in
p
r
ac
tice.
T
h
e
o
v
er
all
wo
r
k
f
lo
w
ex
ten
d
s
p
r
ev
io
u
s
wo
r
k
b
y
Ma
zlan
,
w
h
o
d
ev
elo
p
e
d
an
AC
L
d
iag
n
o
s
is
s
y
s
tem
th
at
in
teg
r
ates d
ig
ital im
ag
e
p
r
o
ce
s
s
in
g
,
f
ea
tu
r
e
r
ed
u
cti
o
n
an
d
an
en
s
em
b
le
o
f
SVM
an
d
ANN
c
lass
if
ier
s
f
o
r
MRI
-
b
ased
in
ju
r
y
d
iag
n
o
s
is
,
an
d
d
e
m
o
n
s
tr
ated
its
f
ea
s
ib
ilit
y
in
a
h
o
s
p
ital settin
g
as in
Fig
u
r
e
3
[
1
8
]
,
[
2
2
]
,
[
2
4]
.
Fig
u
r
e
3
.
Stru
ctu
r
e
o
f
a
s
u
p
er
v
is
ed
lear
n
in
g
s
y
s
tem
S
upe
rvise
d Le
a
rning A
lgorit
hm
I
nput
F
e
a
tur
e
Ne
ura
l
Ne
twork
F
e
a
tur
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
41
,
No
.
2
,
Feb
r
u
a
r
y
20
2
6
:
7
7
3
-
7
8
1
776
T
h
e
f
ir
s
t
s
tag
e
co
n
ce
r
n
s
MR
I
d
ata
ac
q
u
is
itio
n
an
d
s
tan
d
ar
d
izatio
n
.
All
MRI
im
ag
es
u
s
ed
in
th
is
r
esear
ch
wer
e
o
b
tain
e
d
f
r
o
m
t
h
e
r
a
d
io
lo
g
y
d
e
p
ar
tm
en
t
o
f
H
o
s
p
ital
T
aip
in
g
,
Ma
lay
s
ia.
MRI
ac
q
u
is
itio
n
d
etails
(
to
ad
d
r
ess
s
ca
n
n
er
-
r
elate
d
v
a
r
iab
ilit
y
)
:
all
k
n
ee
MR
I
ex
am
i
n
atio
n
s
wer
e
ac
q
u
ir
e
d
at
Ho
s
p
ital
T
aip
in
g
u
s
in
g
a
co
n
s
is
ten
t
clin
ical
k
n
ee
p
r
o
to
c
o
l
to
m
in
im
ize
i
n
tr
a
-
d
ataset
v
ar
iab
ilit
y
.
T
h
e
s
ca
n
n
er
m
o
d
el,
f
ield
s
tr
en
g
th
:
[
1
.
5
T
/3
T
]
,
an
d
k
n
ee
c
o
il:
[
to
b
e
s
p
ec
if
ied
]
.
T
h
e
s
ag
ittal
im
ag
es
u
s
ed
in
th
is
s
tu
d
y
wer
e
ex
p
o
r
t
ed
f
r
o
m
th
e
r
o
u
tin
e
p
r
o
to
co
l.
All
im
ag
es
wer
e
a
n
o
n
y
m
ize
d
at
e
x
p
o
r
t,
co
n
v
er
ted
to
g
r
ay
s
ca
le,
an
d
r
esize
d
f
o
r
s
tan
d
ar
d
ized
p
r
o
ce
s
s
in
g
.
Sli
ce
s
elec
tio
n
an
d
g
r
o
u
n
d
tr
u
th
lab
elin
g
:
ea
ch
s
am
p
le
in
th
e
d
ataset
co
r
r
esp
o
n
d
s
to
a
2
D
s
ag
ittal
s
lice
im
ag
e.
R
ef
er
en
ce
lab
els
(
n
o
r
m
al,
p
a
r
tial
tear
,
c
o
m
p
le
te
tear
)
wer
e
ass
ig
n
ed
b
ased
o
n
th
e
f
in
al
clin
ical
in
ter
p
r
etatio
n
o
f
th
e
f
u
ll
MRI
ex
am
in
atio
n
(
m
u
lti‑sli
ce
s
er
ies)
an
d
co
r
r
esp
o
n
d
in
g
r
a
d
io
lo
g
y
/
o
r
th
o
p
ed
ic
ass
es
s
m
en
t,
an
d
th
e
n
m
a
p
p
ed
to
th
e
ex
p
o
r
ted
s
lice
u
s
ed
as
i
n
p
u
t
to
th
e
alg
o
r
ith
m
.
B
ec
au
s
e
th
e
AC
L
co
u
r
s
e
is
o
b
liq
u
e
r
elativ
e
t
o
th
e
s
ag
ittal
p
lan
e,
u
s
in
g
a
s
in
g
le
s
lice
ca
n
m
is
s
f
o
ca
l
f
ib
er
d
is
r
u
p
tio
n
;
th
er
ef
o
r
e,
f
u
tu
r
e
wo
r
k
will
ex
ten
d
th
e
in
p
u
t
f
r
o
m
a
s
in
g
le
s
lice
to
m
u
lti‑sli
ce
o
r
3
D
v
o
lu
m
e
a
n
aly
s
is
an
d
/o
r
s
ag
itt
al
r
ec
o
n
s
tr
u
ctio
n
s
alig
n
e
d
with
t
h
e
AC
L
tr
ajec
to
r
y
[
3
]
,
[
1
4
]
.
T
h
e
d
ataset
co
m
p
r
is
es
6
0
0
s
ag
ittal
-
p
lan
e
k
n
ee
MRI
im
ag
es o
f
g
r
a
y
s
ca
le
f
o
r
m
at,
ea
ch
with
d
im
en
s
io
n
s
o
f
5
0
0
×5
0
0
p
ix
els
.
Am
o
n
g
th
ese,
2
0
0
r
e
p
r
esen
t
n
o
r
m
al
AC
L
s
tr
u
ctu
r
es,
2
0
0
d
e
p
ict
p
ar
tial
tear
s
a
n
d
2
0
0
co
r
r
esp
o
n
d
to
co
m
p
lete
o
r
co
m
p
lete
tear
s
,
r
esu
ltin
g
in
a
b
ala
n
ce
d
th
r
ee
-
class
d
ataset.
T
h
e
im
ag
es
wer
e
ca
p
tu
r
e
d
u
n
d
er
s
tan
d
ar
d
ized
s
ettin
g
s
d
esig
n
ed
to
m
in
im
ize
m
o
v
e
m
en
t
ar
tef
ac
ts
an
d
o
t
h
er
s
o
u
r
ce
s
o
f
n
o
is
e,
s
u
ch
as
p
h
y
s
io
lo
g
ic
m
o
tio
n
a
n
d
m
ag
n
etic
s
u
s
ce
p
tib
ilit
y
ef
f
ec
ts
,
s
o
as
to
en
s
u
r
e
a
co
n
s
is
ten
t
v
is
u
aliza
tio
n
o
f
th
e
k
n
ee
an
ato
m
y
.
MRI
was
s
elec
ted
b
ec
au
s
e
it
o
f
f
er
s
ex
ce
llen
t
s
o
f
t
-
tis
s
u
e
co
n
tr
ast
wh
ile
u
s
i
n
g
r
ad
io
f
r
eq
u
en
cy
r
ad
iatio
n
co
n
s
id
er
ed
s
af
e
co
m
p
ar
ed
with
i
n
v
asiv
e
im
ag
in
g
m
o
d
alities
lik
e
ar
t
h
r
o
s
co
p
y
[
2
2
]
.
On
ce
co
llected
,
th
e
MRI
im
ag
es
u
n
d
er
g
o
a
n
i
m
ag
e
p
r
ep
r
o
ce
s
s
in
g
p
ip
elin
e
t
h
at
s
tan
d
ar
d
izes
r
eso
lu
tio
n
,
en
h
an
ce
s
co
n
tr
ast
an
d
p
r
ep
ar
es
th
e
im
ag
es
f
o
r
s
eg
m
e
n
tatio
n
an
d
f
ea
tu
r
e
ex
tr
ac
tio
n
.
T
h
e
p
r
ep
r
o
ce
s
s
in
g
is
im
p
lem
en
ted
in
MA
T
L
AB
an
d
b
e
g
in
s
b
y
r
esizin
g
ea
ch
MRI
to
a
u
n
if
o
r
m
s
ize,
wh
ich
f
ac
ilit
ates
later
p
r
o
ce
s
s
in
g
s
t
ep
s
an
d
e
n
s
u
r
es
th
at
r
eg
io
n
-
of
-
in
ter
est
o
p
er
atio
n
s
r
ef
er
en
ce
co
n
s
is
ten
t
co
o
r
d
in
ate
s
.
T
h
e
s
y
s
tem
th
en
ap
p
lies
cr
o
p
p
in
g
to
is
o
late
th
e
AC
L
r
eg
io
n
.
T
h
r
o
u
g
h
co
llab
o
r
ativ
e
an
aly
s
is
in
v
o
lv
in
g
m
ed
i
ca
l
ex
p
er
ts
an
d
in
s
p
ec
tio
n
o
f
6
0
0
MRI
s
am
p
les,
a
cr
o
p
p
in
g
win
d
o
w
lo
ca
ted
at
[
3
0
7
3
1
5
0
1
9
0
]
f
o
r
r
esized
im
ag
es
o
f
[
3
2
0
2
2
0
]
was
id
en
tifie
d
as
o
p
tim
al
f
o
r
ca
p
tu
r
in
g
th
e
lig
am
e
n
t
ar
ea
b
e
twee
n
th
e
f
em
u
r
a
n
d
tib
ia.
T
h
i
s
ch
o
ice
is
co
n
s
is
ten
t
with
p
r
i
o
r
im
ag
in
g
s
tu
d
ies
th
at
r
ec
o
m
m
en
d
f
o
c
u
s
in
g
o
n
an
ato
m
ically
r
elev
an
t
r
eg
io
n
s
to
im
p
r
o
v
e
e
f
f
icien
cy
a
n
d
r
ed
u
ce
ir
r
elev
an
t
v
ar
iatio
n
in
s
u
b
s
eq
u
en
t p
r
o
ce
s
s
in
g
as in
(
5
)
[
2
5
]
,
[
2
6
]
.
(
,
)
=
(
×
)
,
(
×
)
(
5
)
W
h
er
e
b
y
is
th
e
r
esize
g
a
in
.
Fo
llo
win
g
cr
o
p
p
in
g
,
c
o
n
tr
ast
en
h
an
ce
m
e
n
t
is
ap
p
lied
to
r
e
d
is
tr
ib
u
te
p
ix
el
in
ten
s
ities
an
d
im
p
r
o
v
e
th
e
v
is
ib
ilit
y
o
f
lig
am
en
t
s
tr
u
ctu
r
es,
esp
ec
ially
in
s
ca
n
s
th
at
in
itially
ap
p
ea
r
d
ar
k
o
r
ex
h
ib
it
lo
w
co
n
tr
ast.
T
h
is
en
h
an
ce
m
en
t
in
c
r
ea
s
es
th
e
ef
f
ec
tiv
e
g
r
ay
s
ca
le
d
y
n
am
i
c
r
an
g
e
av
ailab
le
f
o
r
AC
L
an
d
ad
jace
n
t
tis
s
u
es,
m
ak
in
g
it e
asier
to
d
is
tin
g
u
is
h
lig
am
en
t f
ib
er
s
f
r
o
m
s
u
r
r
o
u
n
d
in
g
s
tr
u
ctu
r
es a
s
illu
s
tr
ate
in
(
6
)
.
(
1
,
1
,
2
,
2
)
=
(
1
2
)
,
(
1
2
)
,
(
2
2
)
,
(
1
2
)
(
6
)
No
is
e
is
r
ed
u
ce
d
u
s
in
g
m
ed
ia
n
an
d
m
ea
n
f
ilt
er
s
to
s
m
o
o
th
in
ten
s
ity
v
a
r
iatio
n
s
an
d
s
u
p
p
r
ess
s
m
all
ar
tef
ac
ts
wh
ile
p
r
eser
v
in
g
e
d
g
es,
h
elp
in
g
s
tan
d
ar
d
ize
im
ag
e
ap
p
ea
r
an
ce
an
d
im
p
r
o
v
e
s
eg
m
en
tatio
n
r
eliab
ilit
y
.
Seg
m
en
tatio
n
th
en
is
o
lates
th
e
AC
L
b
y
th
r
esh
o
ld
in
g
g
r
a
y
s
ca
le
im
ag
es
to
b
in
ar
y
,
f
o
llo
we
d
b
y
m
o
r
p
h
o
l
o
g
ical
r
ef
in
em
en
t:
d
ilatio
n
b
r
id
g
es
g
ap
s
b
etwe
en
lig
am
en
t
p
ix
els
an
d
er
o
s
io
n
r
em
o
v
es
is
o
lated
n
o
is
e
an
d
s
h
ar
p
en
s
b
o
u
n
d
ar
ies.
B
o
u
n
d
ar
y
t
r
ac
in
g
ex
tr
ac
ts
a
clo
s
ed
lig
am
e
n
t
co
n
to
u
r
,
a
n
d
r
eg
io
n
-
b
ase
d
lab
elin
g
(
r
eg
i
o
n
p
r
o
p
a
g
atio
n
)
r
etain
s
th
e
co
n
n
ec
ted
AC
L
o
b
ject
wh
ile
r
em
o
v
in
g
u
n
r
elate
d
s
tr
u
ct
u
r
es
[1
0
]
,
[
2
4
]
.
T
h
is
f
u
lly
au
to
m
ated
p
r
o
ce
s
s
y
ield
s
an
AC
L
m
ask
th
at
p
r
eser
v
es
lig
am
en
t
g
eo
m
etr
y
an
d
lo
ca
tio
n
f
o
r
m
o
r
p
h
o
m
etr
ic
an
aly
s
is
,
f
o
llo
win
g
k
ey
s
tag
es
d
escr
ib
ed
b
y
Ma
zlan
[
1
8
]
,
[
2
2
]
.
Featu
r
e
ex
tr
ac
tio
n
t
h
en
r
ep
r
esen
ts
th
e
s
eg
m
en
ted
AC
L
u
s
in
g
q
u
a
n
titativ
e
d
escr
ip
to
r
s
f
o
r
m
ac
h
in
e
lear
n
in
g
.
Gu
id
e
d
b
y
o
r
th
o
p
e
d
ic
k
n
o
wled
g
e
a
n
d
p
r
io
r
s
tu
d
ies
[
3
]
,
[
1
3
]
-
[
1
5
]
,
f
e
atu
r
es
in
clu
d
e
s
ize,
ar
ea
,
o
r
ie
n
tatio
n
an
g
le,
elo
n
g
atio
n
,
asp
ec
t
r
atio
,
cir
cu
lar
i
ty
,
ce
n
tr
o
id
,
a
n
d
r
elate
d
r
eg
io
n
p
r
o
p
er
ties
(
T
ab
le
1
)
.
Size
an
d
ar
ea
ca
p
tu
r
e
li
g
am
e
n
t
ex
ten
t
an
d
p
o
ten
tial
f
ib
e
r
lo
s
s
,
an
g
le
r
ef
lects
alig
n
m
en
t
d
is
r
u
p
tio
n
ty
p
ical
o
f
co
m
p
lete
tear
s
,
a
n
d
elo
n
g
atio
n
/asp
ec
t
r
atio
/cir
c
u
lar
ity
d
escr
ib
e
s
h
ap
e
ch
an
g
es
ass
o
ciate
d
with
s
tr
etch
in
g
o
r
s
tr
u
ctu
r
al
d
am
ag
e.
T
h
e
f
u
ll
s
et
o
f
f
ea
tu
r
es
ca
n
b
e
h
ig
h
-
d
im
en
s
io
n
al
an
d
m
a
y
co
n
tain
r
ed
u
n
d
an
t
o
r
co
r
r
elate
d
v
ar
iab
les,
p
r
in
cip
al
co
m
p
o
n
en
t
an
aly
s
is
is
em
p
lo
y
ed
f
o
r
d
im
e
n
s
io
n
ali
ty
r
ed
u
ctio
n
.
PC
A
tr
an
s
f
o
r
m
s
th
e
o
r
ig
i
n
al
v
a
r
iab
l
es
in
to
a
s
m
aller
s
et
o
f
u
n
c
o
r
r
elate
d
p
r
in
cip
al
co
m
p
o
n
e
n
ts
th
at
ca
p
tu
r
e
m
o
s
t
o
f
th
e
v
ar
ian
ce
in
th
e
f
ea
tu
r
e
s
p
ac
e.
I
n
ea
r
lier
AC
L
class
if
icatio
n
wo
r
k
,
Ma
zlan
s
h
o
wed
th
at
f
o
u
r
p
r
in
cip
al
co
m
p
o
n
en
ts
ar
e
s
u
f
f
icien
t
to
p
r
eser
v
e
th
e
d
is
cr
im
in
ativ
e
i
n
f
o
r
m
atio
n
n
ec
ess
ar
y
to
d
is
tin
g
u
is
h
n
o
r
m
al,
p
ar
tial
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
Hyb
r
id
S
V
M
–
A
N
N
s
ys
tem
fo
r
a
u
to
ma
ted
mri
d
ia
g
n
o
s
is
o
f a
n
teri
o
r
…
(
S
a
z
w
a
n
S
ya
fiq
Ma
z
la
n
)
777
an
d
cr
u
cial
in
j
u
r
ies,
wh
ile
s
im
p
lify
in
g
class
if
ier
d
esig
n
a
n
d
m
itig
atin
g
o
v
er
f
itti
n
g
.
Acc
o
r
d
in
g
l
y
,
th
is
s
tu
d
y
u
s
es
th
e
lead
in
g
p
r
in
cip
al
c
o
m
p
o
n
en
ts
as
in
p
u
ts
to
th
e
m
ac
h
in
e
-
lear
n
i
n
g
m
o
d
els.
T
h
e
m
ac
h
in
e
-
lear
n
in
g
ar
ch
itectu
r
e
c
o
n
s
i
s
ts
o
f
two
l
ev
els:
an
SVM
-
b
ased
s
cr
ee
n
i
n
g
s
tag
e
a
n
d
an
ANN
-
b
ased
class
if
icatio
n
s
tag
e.
T
h
e
s
cr
ee
n
in
g
s
tag
e
ad
d
r
ess
es
th
e
b
in
a
r
y
q
u
esti
o
n
o
f
wh
eth
er
a
g
iv
e
n
MRI
s
ca
n
s
h
o
ws
a
n
AC
L
in
ju
r
y
.
Fo
r
th
is
p
u
r
p
o
s
e,
an
SVM
is
tr
ai
n
ed
o
n
th
e
p
r
in
cip
al
co
m
p
o
n
en
ts
to
s
ep
ar
ate
in
ju
r
ed
f
r
o
m
n
o
n
-
in
ju
r
e
d
ca
s
es,
lev
er
ag
in
g
th
e
al
g
o
r
ith
m
’
s
a
b
ilit
y
to
co
n
s
tr
u
ct
o
p
tim
al
s
ep
ar
atin
g
h
y
p
e
r
p
lan
es
with
m
ax
im
al
m
ar
g
in
s
.
Hy
p
er
p
ar
a
m
eter
s
s
u
ch
as
th
e
k
er
n
el
ty
p
e
an
d
r
eg
u
lar
izatio
n
f
ac
to
r
s
ar
e
tu
n
e
d
em
p
i
r
ically
u
s
in
g
tr
ain
in
g
an
d
v
alid
atio
n
s
ets,
with
p
er
f
o
r
m
a
n
ce
ass
ess
ed
th
r
o
u
g
h
c
o
n
f
u
s
io
n
m
atr
ices
an
d
R
OC
cu
r
v
es.
Sam
p
les
class
if
ied
as
n
o
n
-
in
ju
r
e
d
ar
e
lab
eled
as
n
o
r
m
al
an
d
n
o
t
p
ass
ed
to
f
u
r
th
er
p
r
o
ce
s
s
in
g
,
wh
er
ea
s
s
am
p
les
class
if
ied
a
s
in
ju
r
ed
ar
e
f
o
r
war
d
ed
to
th
e
A
NN.
T
ab
le
1
.
Par
am
eter
e
x
tr
ac
tio
n
F
e
a
t
u
r
e
D
e
scri
p
t
i
o
n
X
-
a
x
i
s
El
o
n
g
a
t
i
o
n
p
i
x
e
l
i
n
x
-
a
x
i
s
.
Y
-
a
x
i
s
El
o
n
g
a
t
i
o
n
p
i
x
e
l
i
n
y
-
a
x
i
s
.
P
e
r
i
me
t
e
r
To
t
a
l
p
i
x
e
l
a
l
o
n
g
A
C
L
b
o
u
n
d
a
r
y
.
A
r
e
a
Tr
i
a
n
g
u
l
a
r
a
r
e
a
i
n
A
C
L
.
A
v
e
r
a
g
e
p
i
x
e
l
A
n
a
v
e
r
a
g
e
p
i
x
e
l
o
f
A
C
L
o
b
j
e
c
t
i
n
i
m
a
g
e
.
C
i
r
c
u
l
a
r
i
t
y
C
i
r
c
u
l
a
r
a
r
e
a
i
n
A
C
L
.
A
sp
e
c
t
r
a
t
i
o
R
a
t
i
o
v
e
r
t
i
c
a
l
p
i
x
e
l
l
i
n
e
w
i
t
h
h
o
r
i
z
o
n
t
a
l
p
i
x
e
l
l
i
n
e
.
A
n
g
l
e
El
o
n
g
a
t
i
o
n
d
e
g
r
e
e
A
C
L
.
N
u
mb
e
r
si
d
e
To
t
a
l
p
i
x
e
l
A
C
L
o
b
j
e
c
t
i
n
i
m
a
g
e
.
T
h
e
ANN,
im
p
lem
en
ted
as
a
f
ee
d
-
f
o
r
war
d
n
etwo
r
k
tr
ain
ed
with
th
e
L
ev
en
b
er
g
–
Ma
r
q
u
ar
d
t
alg
o
r
ith
m
,
p
er
f
o
r
m
s
m
u
lti
-
class
class
if
icatio
n
in
to
n
o
r
m
al,
p
ar
tial
tear
an
d
co
m
p
lete
tear
c
ateg
o
r
ies.
I
t
ac
ce
p
ts
th
e
p
r
in
cip
al
c
o
m
p
o
n
en
ts
as
in
p
u
ts
an
d
p
r
o
d
u
ce
s
a
th
r
ee
-
cl
ass
o
u
tp
u
t.
Du
r
in
g
tr
ain
in
g
,
t
h
e
n
etwo
r
k
weig
h
ts
ar
e
iter
ativ
ely
u
p
d
ated
u
n
til
c
o
n
v
er
g
en
ce
is
ac
h
iev
ed
b
ased
o
n
p
er
f
o
r
m
an
ce
m
etr
ics
s
u
ch
as
m
ea
n
-
s
q
u
ar
ed
er
r
o
r
a
n
d
ac
cu
r
ac
y
o
n
tr
ain
i
n
g
,
v
alid
atio
n
an
d
test
s
ets.
Ma
zlan
’
s
th
esis
r
ep
o
r
ts
th
at,
with
ap
p
r
o
p
r
iat
e
co
n
f
ig
u
r
atio
n
,
th
e
ANN
ac
h
iev
es
n
ea
r
-
p
er
f
ec
t
r
eg
r
ess
io
n
v
a
lu
es
an
d
h
ig
h
class
-
s
p
ec
if
ic
p
er
f
o
r
m
a
n
ce
in
d
ices.
T
h
ese
f
in
d
in
g
s
in
f
o
r
m
th
e
n
et
wo
r
k
d
esig
n
a
n
d
tr
ai
n
in
g
s
tr
at
eg
y
u
s
ed
in
th
e
p
r
esen
t stu
d
y
a
s
in
F
ig
u
r
e
4
.
Fig
u
r
e
4
.
ANN
ap
p
lied
f
o
r
th
e
AC
L
d
iag
n
o
s
is
s
tr
u
ctu
r
e
Per
f
o
r
m
an
ce
e
v
alu
atio
n
in
v
o
lv
es
b
o
th
q
u
an
titativ
e
an
d
q
u
alitativ
e
m
ea
s
u
r
es.
Qu
an
titativ
ely
,
th
e
s
y
s
tem
is
ev
alu
ated
u
s
in
g
ac
c
u
r
ac
y
,
s
en
s
itiv
ity
,
s
p
ec
if
icity
,
co
n
f
u
s
io
n
m
atr
ices
an
d
R
OC
cu
r
v
es
f
o
r
th
e
SVM,
ANN
an
d
th
e
co
m
b
in
ed
h
y
b
r
id
s
y
s
tem
.
Qu
alitativ
ely
,
clin
i
ca
l
v
alid
atio
n
is
o
b
tai
n
ed
b
y
co
m
p
ar
in
g
s
y
s
tem
o
u
tp
u
ts
ag
ain
s
t
th
e
d
iag
n
o
s
es
o
f
o
r
t
h
o
p
e
d
ic
an
d
r
ad
i
o
lo
g
y
s
p
ec
ialis
ts
o
n
s
elec
ted
ca
s
es,
an
d
b
y
ad
m
i
n
is
ter
in
g
an
o
n
lin
e
q
u
esti
o
n
n
air
e
to
m
e
d
ical
p
r
ac
titi
o
n
e
r
s
to
g
au
g
e
u
s
ab
ilit
y
,
u
s
e
f
u
ln
ess
a
n
d
af
f
o
r
d
a
b
ilit
y
.
T
h
e
s
y
s
tem
is
u
ltima
tely
im
p
lem
en
ted
o
n
two
p
latf
o
r
m
s
:
MA
T
L
AB
,
s
er
v
in
g
as
th
e
d
ev
elo
p
m
e
n
t
an
d
p
r
o
ce
s
s
in
g
en
v
ir
o
n
m
en
t,
an
d
an
A
n
d
r
o
id
ap
p
licatio
n
th
at
d
eliv
e
r
s
a
lig
h
tweig
h
t d
iag
n
o
s
tic
to
o
l
f
o
r
m
o
b
ile
u
s
e.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
r
esu
lts
o
f
th
is
s
tu
d
y
d
em
o
n
s
tr
ate
th
at
th
e
p
r
o
p
o
s
ed
SVM
–
ANN
en
s
em
b
le
s
y
s
tem
ca
n
ac
cu
r
ately
d
iag
n
o
s
e
AC
L
co
n
d
itio
n
s
u
s
in
g
MRI
an
d
th
at
it
alig
n
s
clo
s
ely
with
clin
ical
ex
p
ec
tatio
n
s
.
Qu
an
titativ
e
p
er
f
o
r
m
an
ce
is
s
u
m
m
ar
ized
u
s
in
g
co
n
f
u
s
io
n
m
atr
ices
an
d
a
g
g
r
eg
ated
m
etr
ics
f
o
r
th
e
SV
M
s
cr
ee
n
in
g
m
o
d
u
le,
th
e
ANN
class
if
icatio
n
m
o
d
u
l
e
an
d
th
e
i
n
teg
r
ated
h
y
b
r
id
s
y
s
tem
.
T
h
e
p
r
o
ce
s
s
in
p
u
t
o
b
tai
n
ed
d
ata
f
r
o
m
f
o
u
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
41
,
No
.
2
,
Feb
r
u
a
r
y
20
2
6
:
7
7
3
-
7
8
1
778
f
ea
tu
r
es
ex
p
lain
ed
in
p
r
ev
io
u
s
ch
ap
ter
.
Fig
u
r
e
5
(
a)
d
o
es
n
o
t
ap
p
ly
th
e
k
-
m
ea
n
tech
n
iq
u
e,
wh
ile
F
ig
u
r
e
5
(
b
)
ap
p
lied
with
2
−
2
o
f
(
K)
a
n
d
F
ig
u
r
e
5
(
c)
with
d
if
f
er
e
n
t
(
K)
p
ar
a
m
eter
o
f
2
−
4
.
R
ed
d
o
ts
m
ea
n
AC
L
in
ju
r
y
a
n
d
g
r
ee
n
d
o
ts
m
ea
n
n
o
n
-
AC
L
in
j
u
r
y
.
(
a)
(
b
)
(
c)
Fig
u
r
e
5
.
Data
d
is
tr
ib
u
tio
n
:
(
a)
with
o
u
t th
e
k
-
m
ea
n
clu
s
ter
,
(
b
)
2
−
2
k
-
m
ea
n
clu
s
ter
,
an
d
(
c)
2
−
4
k
-
m
ea
n
clu
s
ter
All
th
e
clu
s
ter
ed
p
ar
am
eter
s
b
ased
o
n
f
o
u
r
f
ea
tu
r
es
(
cir
cu
lar
ity
,
asp
ec
t
r
atio
,
an
g
le
an
d
n
u
m
b
er
s
id
e)
ca
n
b
e
ea
s
ily
id
e
n
tifie
d
f
r
o
m
ea
ch
g
r
o
u
p
u
s
in
g
k
-
m
ea
n
clu
s
ter
wh
ich
u
s
ed
co
n
s
tan
t
p
ar
a
m
eter
o
f
C
=2
3
with
γ
=2
−4
.
T
h
is
f
ig
u
r
e
is
th
e
b
est test
in
g
p
ar
am
eter
ap
p
lied
o
n
b
o
th
clu
s
ter
; h
ig
h
er
p
ar
am
eter
m
ig
h
t p
r
o
d
u
ce
lar
g
e
m
ar
g
in
er
r
o
r
,
wh
ile
a
lo
w
p
ar
am
eter
p
r
o
d
u
ce
s
less
b
o
u
n
d
ar
y
clu
s
ter
.
T
h
e
ap
p
r
o
x
im
ate
co
ef
f
icien
t e
x
p
lain
s
th
e
r
elatio
n
b
etwe
en
in
ju
r
y
d
ata
a
n
d
n
o
n
-
in
j
u
r
y
d
ata
as
T
a
b
le
2
s
h
o
ws
th
e
r
esu
lt
o
f
s
cr
ee
n
in
g
u
s
in
g
SVM
wit
h
d
if
f
er
en
t
p
ar
am
eter
s
o
f
g
am
m
a.
Fro
m
h
e
r
e,
th
e
m
o
s
t
s
u
itab
le
v
alu
e
f
its
f
o
r
t
h
is
s
y
s
te
m
is
at
γ
=
-
4
with
an
av
er
ag
e
h
i
g
h
est ap
p
r
o
x
im
ate
c
o
ef
f
icien
t o
f
7
9
%.
T
ab
le
2
.
SVM
s
cr
ee
n
in
g
with
th
e
v
ar
iab
le
g
a
m
m
a
p
a
r
am
eter
P
h
a
se
D
a
t
a
S
c
h
e
me
γ
=
0
γ=
-
2
γ=
-
4
γ=
-
8
Tr
a
i
n
i
n
g
7
0
%
4
0
.
4
0
%
5
2
.
1
0
%
7
7
.
4
0
%
6
1
.
4
0
%
Te
st
i
n
g
2
0
%
4
7
.
1
0
%
5
4
.
7
0
%
8
1
.
2
0
%
7
2
.
5
0
%
V
a
l
i
d
a
t
i
o
n
1
0
%
3
8
.
9
0
%
5
9
.
3
0
%
8
2
.
5
0
%
6
8
.
6
0
%
T
h
e
ANN
ap
p
lied
L
ev
en
b
er
g
-
Ma
r
q
u
ar
d
t
(
L
M)
tech
n
iq
u
e.
T
h
e
L
M
is
a
v
er
y
s
im
p
le
m
eth
o
d
f
o
r
ap
p
r
o
x
im
atin
g
a
f
u
n
ctio
n
in
cl
ass
if
y
in
g
AC
L
in
ju
r
ies
L
M
ab
le
to
r
ev
ea
l
p
o
ten
tially
co
m
p
le
x
r
elatio
n
s
h
ip
s
f
o
r
all
f
ea
tu
r
es
in
th
is
AC
L
d
iag
n
o
s
is
s
y
s
tem
.
I
t
is
u
n
ab
le
to
e
s
tim
ate
th
e
o
u
tp
u
t,
wh
ich
r
esu
lt
ty
p
e
o
f
AC
L
in
ju
r
ies.
Fig
u
r
e
6
i
n
d
icate
s
s
y
s
tem
p
er
f
o
r
m
an
ce
a
n
d
F
ig
u
r
e
7
s
h
o
w
th
e
p
er
f
o
r
m
a
n
ce
with
o
u
t tr
ain
in
g
.
Fig
u
r
e
6
.
Sy
s
tem
p
e
r
f
o
r
m
an
ce
T
h
e
tr
ain
in
g
e
p
o
ch
o
f
7
6
iter
atio
n
s
o
r
ep
o
c
h
h
as
tu
n
e
wei
g
h
ted
f
r
o
m
o
r
ig
in
v
al
u
e
0
.
0
0
1
to
0
.
0
1
,
wh
er
eb
y
lo
wer
weig
h
t
p
r
o
d
u
c
es
s
m
aller
d
ata
an
d
th
e
im
p
ac
t
o
n
g
r
ap
h
s
h
o
ws
tr
ain
in
g
d
at
a
p
atter
n
o
f
10
−
3
.
Fig
u
r
e
6
p
lo
ts
g
r
ap
h
f
o
r
tr
ain
i
n
g
s
tate,
th
e
g
r
ad
ien
t
a
g
r
ap
h
s
h
o
ws
g
r
ad
ien
t
v
alu
e
d
ec
r
ea
s
i
n
g
f
r
o
m
0
.
0
0
0
7
to
9
.
6
8
1
x
10
−
6
with
7
6
ep
o
c
h
s
.
T
h
e
d
ec
r
ea
s
in
g
is
tu
n
ed
to
m
ax
im
i
ze
m
atch
in
g
b
etwe
en
tr
ain
in
g
d
ata
an
d
test
d
ata,
wh
ile
Fig
u
r
e
7
s
h
o
ws
th
r
ee
g
r
ap
h
s
wh
ich
is
tr
ain
in
g
,
te
s
t
an
d
o
v
er
all
(
all)
.
T
r
ain
in
g
r
eg
r
ess
i
o
n
p
lo
tted
at
0
.
9
8
0
8
,
wh
ile
th
e
v
alid
atio
n
r
eg
r
ess
io
n
s
h
o
ws
at
0
.
9
8
6
3
a
n
d
test
r
eg
r
ess
io
n
at
0
.
9
6
5
7
.
Fro
m
h
er
e,
th
e
all
g
r
ap
h
s
h
o
ws
m
o
s
t
o
f
th
e
d
ata
lay
o
n
th
e
f
it
lin
e.
T
h
e
o
v
e
r
all
p
er
f
o
r
m
an
ce
o
f
tr
ain
i
n
g
,
v
alid
atio
n
a
n
d
test
s
h
o
ws
r
eg
r
ess
io
n
o
f
0
.
9
7
8
5
.
F
r
o
m
h
er
e,
wh
en
all
d
ata
p
lo
tted
in
a
s
in
g
le
g
r
ap
h
,
o
b
v
io
u
s
ly
s
ee
n
m
o
s
t
o
f
th
e
d
ata
ar
e
n
o
t la
y
o
n
th
e
f
it lin
e.
T
h
er
ef
o
r
e,
a
f
u
r
th
e
r
tr
ain
in
g
is
r
eq
u
ir
ed
to
en
s
u
r
e
m
o
s
t o
f
th
e
d
ata
lay
o
n
th
e
f
it
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
Hyb
r
id
S
V
M
–
A
N
N
s
ys
tem
fo
r
a
u
to
ma
ted
mri
d
ia
g
n
o
s
is
o
f a
n
teri
o
r
…
(
S
a
z
w
a
n
S
ya
fiq
Ma
z
la
n
)
779
lin
e
an
d
p
r
o
d
u
ce
d
a
m
o
r
e
ac
c
u
r
ate
s
y
s
tem
.
A
co
n
f
u
s
io
n
m
atr
ix
m
eth
o
d
in
th
is
r
esear
ch
co
n
s
is
ts
o
f
th
e
h
y
b
r
id
co
m
b
in
atio
n
o
f
SVM
an
d
ANN,
it
i
s
u
s
ed
f
o
r
a
d
ec
is
io
n
s
u
p
p
o
r
t
to
o
l
th
at
u
s
es
a
m
o
d
el
o
f
d
ec
is
io
n
s
an
d
p
o
s
s
ib
le
co
n
s
eq
u
e
n
ce
s
in
clu
d
in
g
c
h
an
ce
ev
en
t
o
u
tc
o
m
es,
r
e
s
o
u
r
ce
co
s
ts
an
d
u
tili
ty
.
Fo
r
t
h
e
id
en
tific
atio
n
o
f
AC
L
in
ju
r
y
,
th
e
r
esu
lts
ar
e
t
y
p
ically
p
r
esen
ted
in
ter
m
s
o
f
th
e
co
n
f
u
s
io
n
m
atr
ix
.
T
h
i
s
m
atr
ix
s
h
o
ws
th
e
d
is
p
o
s
itio
n
s
o
f
th
e
s
et
o
f
in
s
t
an
ce
s
in
a
m
atr
ix
f
o
r
m
.
Su
p
p
o
s
e
an
id
en
tific
atio
n
s
y
s
tem
in
v
o
lv
es
o
n
ly
tw
o
class
es,
wh
er
e
ea
ch
h
as
its
ass
o
ciate
d
clas
s
lab
els.
T
h
er
e
will
b
e
f
o
u
r
p
o
s
s
ib
le
o
u
tco
m
es
in
th
is
ca
s
e
b
ec
au
s
e
it is
a
b
in
ar
y
ca
s
e.
T
ab
le
3
s
h
o
ws d
etails r
esu
lt o
n
h
y
b
r
id
s
y
s
tem
.
Fig
u
r
e
7
.
Sy
s
tem
p
e
r
f
o
r
m
an
ce
with
o
u
t c
o
m
p
lete
tr
ain
in
g
T
ab
le
3
.
Deta
ils
r
esu
lt f
o
r
h
y
b
r
id
d
iag
n
o
s
is
s
y
s
tem
o
f
AC
L
S
e
v
e
r
i
t
y
M
i
l
d
M
o
d
e
r
a
t
e
S
e
v
e
r
e
A
c
c
u
r
a
c
y
(
%)
8
5
.
2
7
4
%
8
7
.
7
3
1
%
8
7
.
3
6
9
%
Er
r
o
r
r
a
t
e
(
%)
<
1
5
%
<
1
4
%
<
1
4
%
S
e
n
s
i
t
i
v
i
t
y
(
%)
7
9
.
3
5
1
%
8
0
.
9
2
4
%
8
2
.
4
6
3
%
T
ab
le
3
s
u
m
m
a
r
izes
th
e
h
y
b
r
id
s
y
s
tem
p
er
f
o
r
m
an
ce
f
o
r
th
r
ee
s
ev
er
ity
lev
els
(
m
ild
,
m
o
d
er
ate,
an
d
s
ev
er
e)
,
wh
ich
co
r
r
esp
o
n
d
to
i
n
cr
ea
s
in
g
d
eg
r
ee
s
o
f
AC
L
f
ib
er
d
is
r
u
p
tio
n
in
t
h
is
d
ataset.
I
m
p
o
r
tan
tly
,
clin
ical
tr
ea
tm
en
t
d
ec
is
io
n
s
(
e.
g
.
,
s
u
r
g
er
y
v
s
.
r
eh
ab
ilit
atio
n
)
d
e
p
en
d
o
n
a
d
d
itio
n
al
f
ac
to
r
s
s
u
ch
as
k
n
ee
in
s
tab
ilit
y
,
p
atien
t
ac
tiv
ity
d
em
an
d
s
,
an
d
co
n
co
m
itan
t
m
en
is
ca
l
o
r
ch
o
n
d
r
al
in
ju
r
ies;
th
er
ef
o
r
e,
th
e
s
y
s
tem
o
u
tp
u
t
s
h
o
u
ld
b
e
in
ter
p
r
ete
d
as im
ag
in
g
‑
b
ased
s
ev
er
ity
g
r
a
d
in
g
r
ath
er
th
a
n
a
d
ir
ec
t tr
ea
tm
en
t r
ec
o
m
m
en
d
atio
n
[
2
]
,
[
1
4]
.
Per
f
o
r
m
an
ce
c
o
m
p
a
r
is
o
n
an
d
r
ef
in
em
en
t:
th
e
h
y
b
r
id
SVM
–
ANN
s
y
s
tem
attain
s
s
en
s
itiv
it
y
v
alu
es
o
f
ap
p
r
o
x
im
ately
7
9
–
8
2
%
ac
r
o
s
s
s
ev
er
ity
s
tr
ata
(
T
ab
le
3
)
,
w
h
ich
is
lo
wer
th
an
s
ev
er
al
r
e
ce
n
t
m
u
lti‑seq
u
en
ce
r
ad
io
m
ics
an
d
d
ee
p
lear
n
in
g
ap
p
r
o
ac
h
es.
Fo
r
in
s
tan
ce
,
C
h
en
g
et
a
l.
[
3
]
r
e
p
o
r
te
d
v
alid
a
tio
n
s
en
s
itiv
ity
an
d
s
p
ec
if
icity
o
f
0
.
8
5
7
an
d
0
.
8
2
9
(
AUC
0
.
9
2
7
)
u
s
in
g
m
u
lti‑s
eq
u
en
ce
MRI
r
ad
io
m
ics
with
an
SVM
class
if
ier
.
T
h
e
p
er
f
o
r
m
an
ce
g
a
p
is
lik
ely
in
f
lu
en
ce
d
b
y
(
i)
th
e
u
s
e
o
f
a
s
in
g
le
2
D
s
ag
ittal
s
lice
r
ath
er
t
h
an
m
u
lti‑sli
ce
/3
D
in
p
u
ts
,
(
ii)
s
ca
n
n
er
/p
r
o
t
o
co
l
v
ar
iab
ilit
y
n
o
t
f
u
lly
m
o
d
eled
,
an
d
(
iii)
th
e
u
s
e
o
f
h
an
d
‑
cr
af
ted
m
o
r
p
h
o
m
etr
i
c
f
ea
tu
r
es
in
s
tead
o
f
r
ich
er
tex
t
u
r
e/r
ad
io
m
ics
f
ea
tu
r
es
[1
2
]
,
[
2
5
]
.
Fu
tu
r
e
r
ef
in
em
e
n
ts
will
p
r
io
r
itize
m
u
lti‑sli
ce
ag
g
r
eg
atio
n
(
e.
g
.
,
s
lice‑
wis
e
v
o
tin
g
o
r
3
D
m
o
d
elin
g
)
,
p
r
o
to
co
l
h
ar
m
o
n
izatio
n
,
f
ea
tu
r
e
ex
p
an
s
io
n
to
war
d
r
ad
io
m
ics,
an
d
c
o
m
p
ar
ativ
e
b
en
c
h
m
ar
k
i
n
g
a
g
ain
s
t
lig
h
tweig
h
t
co
n
v
o
lu
tio
n
al
n
e
u
r
a
l
n
etwo
r
k
(
C
NN
)
b
ac
k
b
o
n
es su
itab
le
f
o
r
m
o
b
ile
d
ep
lo
y
m
e
n
t
[
5
]
,
[
1
4
]
,
[
2
6
]
.
4.
CO
NCLU
SI
O
N
T
h
e
r
esu
lts
s
h
o
w
th
at
a
s
tr
u
ctu
r
ed
p
ip
elin
e
s
tan
d
ar
d
ized
p
r
ep
r
o
ce
s
s
in
g
,
in
ter
p
r
eta
b
le
m
o
r
p
h
o
m
et
r
ic
f
ea
tu
r
e
ex
tr
ac
tio
n
,
PC
A
-
b
ased
r
ed
u
ctio
n
,
an
d
h
y
b
r
id
cla
s
s
if
icatio
n
ca
n
en
ab
le
au
to
m
ated
AC
L
in
ju
r
y
s
cr
ee
n
in
g
a
n
d
MRI
-
b
ased
s
ev
er
ity
g
r
ad
in
g
as
a
d
ec
is
io
n
-
s
u
p
p
o
r
t
to
o
l
(
n
o
t
a
r
ep
lace
m
e
n
t
f
o
r
e
x
p
er
t
r
ea
d
i
n
g
)
.
A
k
ey
s
tr
en
g
th
is
th
e
two
-
s
tag
e
d
esig
n
:
th
e
SVM
f
ir
s
t
f
ilte
r
s
clea
r
ly
n
o
r
m
al
ca
s
es,
r
ed
u
c
in
g
ANN
wo
r
k
lo
ad
an
d
allo
win
g
th
e
ANN
to
f
o
cu
s
o
n
g
r
ad
in
g
in
ju
r
ed
s
ca
n
s
.
T
h
is
r
ef
lects
clin
ical
r
ea
s
o
n
in
g
an
d
,
b
ased
o
n
co
n
f
u
s
io
n
m
atr
ices
an
d
p
er
f
o
r
m
an
ce
co
m
p
ar
is
o
n
s
,
r
e
d
u
ce
s
n
o
r
m
al
-
ca
s
e
m
is
class
if
icatio
n
wh
ile
im
p
r
o
v
in
g
d
is
cr
im
in
atio
n
b
etwe
en
p
ar
tia
l
an
d
co
m
p
lete
tear
s
,
co
n
s
is
t
en
t
with
Ma
zlan
’
s
f
in
d
in
g
s
o
n
h
y
b
r
id
s
y
s
tem
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
41
,
No
.
2
,
Feb
r
u
a
r
y
20
2
6
:
7
7
3
-
7
8
1
780
C
lin
ically
m
o
tiv
ated
ch
o
ices
(
r
eg
io
n
o
f
in
ter
es
t
(
R
OI
)
d
ef
in
i
tio
n
an
d
m
o
r
p
h
o
lo
g
y
-
b
ased
d
escr
ip
to
r
s
lin
k
ed
to
lig
am
en
t
co
n
tin
u
ity
/ap
p
ea
r
an
ce
)
also
im
p
r
o
v
e
tr
an
s
p
ar
en
c
y
v
er
s
u
s
b
lack
-
b
o
x
m
o
d
els
an
d
s
u
p
p
o
r
t
f
u
tu
r
e
ex
ten
s
io
n
to
r
ad
io
m
ics
o
r
d
ee
p
f
ea
tu
r
es.
A
p
r
elim
in
ar
y
clin
ician
ev
alu
atio
n
f
o
u
n
d
t
h
e
in
t
er
f
a
ce
an
d
o
u
tp
u
ts
u
n
d
er
s
tan
d
a
b
le
f
o
r
s
cr
ee
n
in
g
an
d
g
r
ad
in
g
;
h
o
we
v
er
,
b
r
o
a
d
e
r
m
u
lti
-
d
is
cip
lin
ar
y
an
d
p
r
o
s
p
ec
tiv
e
ass
ess
m
en
t
s
esp
ec
ially
in
clu
d
in
g
m
o
r
e
m
u
s
cu
lo
s
k
eleta
l
r
ad
io
lo
g
is
ts
ar
e
n
ee
d
ed
to
q
u
an
tify
clin
ical
im
p
ac
t.
L
im
itatio
n
s
in
clu
d
e
r
elian
ce
o
n
a
s
in
g
le
s
ag
ittal
s
lice
(
r
is
k
in
g
m
is
s
ed
f
o
ca
l
d
is
r
u
p
tio
n
d
u
e
to
th
e
AC
L
’
s
o
b
liq
u
e
co
u
r
s
e)
,
p
o
ten
tial
s
ca
n
n
er
/p
r
o
to
co
l v
ar
i
ab
ilit
y
,
an
d
a
m
o
d
est
d
ataset
with
h
an
d
-
cr
a
f
ted
f
ea
tu
r
es
th
a
t
m
ay
u
n
d
er
p
er
f
o
r
m
m
u
lti
-
s
eq
u
en
ce
r
ad
io
m
ics.
R
ea
l
-
wo
r
ld
d
ep
lo
y
m
en
t
will
r
eq
u
ir
e
m
u
lti
-
ce
n
ter
p
r
o
s
p
ec
tiv
e
v
alid
atio
n
,
p
r
o
t
o
co
l
h
ar
m
o
n
izatio
n
,
a
n
d
th
r
esh
o
ld
ca
lib
r
atio
n
;
f
u
tu
r
e
wo
r
k
will
e
x
p
an
d
d
ata
ac
r
o
s
s
s
ca
n
n
er
s
/s
eq
u
en
ce
s
,
a
d
d
m
u
lti
-
s
lice
ag
g
r
eg
atio
n
,
a
n
d
b
en
ch
m
ar
k
lig
h
tweig
h
t
d
ee
p
-
lear
n
in
g
b
ac
k
b
o
n
es su
itab
le
f
o
r
e
d
g
e/m
o
b
ile
u
s
e
.
Ov
er
all,
th
is
s
tu
d
y
p
r
esen
ts
a
f
u
lly
au
to
m
ated
f
r
am
ewo
r
k
in
teg
r
atin
g
p
r
ep
r
o
ce
s
s
in
g
,
PC
A
-
b
ase
d
f
ea
tu
r
e
r
ed
u
ctio
n
,
SVM
s
cr
ee
n
in
g
,
an
d
ANN
g
r
ad
in
g
in
a
u
n
if
ied
ar
c
h
itectu
r
e.
On
a
b
ala
n
ce
d
d
ataset
o
f
6
0
0
k
n
ee
MRI
im
ag
es
f
r
o
m
Ho
s
p
ital
T
aip
in
g
,
th
e
s
y
s
tem
ac
h
ie
v
ed
~8
6
%
ac
c
u
r
ac
y
a
n
d
>8
1
%
s
en
s
itiv
ity
,
with
R
OC
an
aly
s
i
s
in
d
icatin
g
s
tr
o
n
g
d
is
cr
im
in
ativ
e
p
er
f
o
r
m
a
n
ce
.
E
x
p
e
r
t
v
er
if
icatio
n
an
d
p
r
ac
titi
o
n
er
s
u
r
v
ey
f
ee
d
b
ac
k
s
u
g
g
est
th
e
o
u
tp
u
ts
alig
n
with
r
o
u
tin
e
p
r
ac
tice
an
d
t
h
at
th
e
to
o
l
is
u
s
ab
le
a
n
d
p
o
ten
tially
co
s
t
-
ef
f
ec
tiv
e
f
o
r
d
ep
l
o
y
m
en
t.
T
h
is
wo
r
k
co
n
tr
ib
u
tes
b
y
em
b
ed
d
in
g
ex
p
er
t
k
n
o
wled
g
e
in
to
an
en
d
-
to
-
en
d
p
latf
o
r
m
th
at
co
m
b
in
es
im
ag
e
p
r
o
ce
s
s
in
g
,
m
ac
h
in
e
l
ea
r
n
in
g
,
a
n
d
c
o
n
f
u
s
io
n
-
m
atr
i
x
-
b
ased
ev
al
u
atio
n
to
r
ed
u
ce
m
an
u
al
wo
r
k
lo
ad
,
s
tan
d
ar
d
ize
d
ec
is
io
n
s
,
an
d
s
u
p
p
o
r
t
co
n
s
is
ten
t
AC
L
d
iag
n
o
s
is
.
Fu
tu
r
e
wo
r
k
will
f
o
cu
s
o
n
m
u
lti
-
ce
n
ter
ex
p
an
s
io
n
,
lear
n
i
n
g
f
ea
tu
r
es
d
ir
ec
tly
f
r
o
m
r
aw
im
ag
es
u
s
in
g
d
ee
p
lear
n
in
g
,
an
d
tig
h
t
er
in
teg
r
atio
n
with
clin
ical
in
f
o
r
m
atio
n
tec
h
n
o
lo
g
y
(
IT
)
s
y
s
tem
s
to
im
p
r
o
v
e
r
o
b
u
s
tn
ess
,
ac
cu
r
ac
y
,
an
d
a
d
o
p
tio
n
in
r
o
u
tin
e
o
r
th
o
p
e
d
ic
an
d
s
p
o
r
ts
m
ed
icin
e
p
r
ac
tice.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
s
tate
n
o
f
u
n
d
in
g
in
v
o
lv
ed
.
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
Data
av
ailab
ilit
y
is
n
o
t
ap
p
li
ca
b
le
to
th
is
p
ap
er
as
n
o
n
e
w
d
ata
wer
e
cr
ea
ted
o
r
an
aly
ze
d
in
th
is
s
tu
d
y
.
RE
F
E
R
E
NC
E
S
[
1
]
L.
Y
a
o
,
A
.
G
e
n
t
i
l
i
,
L.
L
.
S
e
e
g
e
r
,
a
n
d
H
.
M
.
D
o
,
“
A
n
t
e
r
i
o
r
c
r
u
c
i
a
t
e
l
i
g
a
me
n
t
i
n
j
u
r
y
:
M
R
i
m
a
g
i
n
g
d
i
a
g
n
o
si
s
a
n
d
p
a
t
t
e
r
n
s
o
f
i
n
j
u
r
y
,
”
Ra
d
i
o
l
o
g
i
c
C
l
i
n
i
c
s
o
f
N
o
rt
h
Am
e
ri
c
a
,
v
o
l
.
3
3
,
n
o
.
4
,
p
p
.
7
7
1
–
7
8
6
,
1
9
9
5
.
[
2
]
R
.
K
o
t
s
i
f
a
k
i
e
t
a
l
.
,
“
A
s
p
e
t
a
r
c
l
i
n
i
c
a
l
p
r
a
c
t
i
c
e
g
u
i
d
e
l
i
n
e
o
n
r
e
h
a
b
i
l
i
t
a
t
i
o
n
a
f
t
e
r
a
n
t
e
r
i
o
r
c
r
u
c
i
a
t
e
l
i
g
a
me
n
t
r
e
c
o
n
s
t
r
u
c
t
i
o
n
,
”
Bri
t
i
sh
J
o
u
rn
a
l
o
f
S
p
o
r
t
s
Me
d
i
c
i
n
e
,
v
o
l
.
5
7
,
n
o
.
9
,
p
p
.
5
0
0
–
5
1
4
,
M
a
y
2
0
2
3
,
d
o
i
:
1
0
.
1
1
3
6
/
b
j
s
p
o
r
t
s
-
2
0
2
2
-
1
0
6
1
5
8
.
[
3
]
Q
.
C
h
e
n
g
,
H
.
Li
n
,
J
.
Z
h
a
o
,
X
.
L
u
,
a
n
d
Q
.
W
a
n
g
,
“
A
p
p
l
i
c
a
t
i
o
n
o
f
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
-
b
a
s
e
d
mu
l
t
i
-
se
q
u
e
n
c
e
M
R
I
r
a
d
i
o
m
i
c
s
i
n
d
i
a
g
n
o
si
n
g
a
n
t
e
r
i
o
r
c
r
u
c
i
a
t
e
l
i
g
a
me
n
t
t
e
a
r
s,”
J
o
u
r
n
a
l
o
f
O
rt
h
o
p
a
e
d
i
c
S
u
r
g
e
r
y
a
n
d
Re
se
a
r
c
h
,
v
o
l
.
1
9
,
n
o
.
1
,
p
.
9
9
,
Jan
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
8
6
/
s
1
3
0
1
8
-
024
-
0
4
6
0
2
-
5.
[
4
]
C
.
M
.
B
i
sh
o
p
,
Pa
t
t
e
rn
Re
c
o
g
n
i
t
i
o
n
a
n
d
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
.
N
e
w
Y
o
r
k
:
S
p
r
i
n
g
e
r
,
2
0
0
6
.
[
5
]
M
.
M
e
r
c
u
r
i
o
e
t
a
l
.
,
“
D
e
e
p
l
e
a
r
n
i
n
g
mo
d
e
l
s
t
o
d
e
t
e
c
t
a
n
t
e
r
i
o
r
c
r
u
c
i
a
t
e
l
i
g
a
m
e
n
t
i
n
j
u
r
y
o
n
M
R
I
:
a
c
o
mp
r
e
h
e
n
si
v
e
r
e
v
i
e
w
,
”
D
i
a
g
n
o
s
t
i
c
s
,
v
o
l
.
1
5
,
n
o
.
6
,
p
.
7
7
6
,
M
a
r
.
2
0
2
5
,
d
o
i
:
1
0
.
3
3
9
0
/
d
i
a
g
n
o
s
t
i
c
s1
5
0
6
0
7
7
6
.
[
6
]
C
.
Q
u
e
t
a
l
.
,
“
A
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
f
o
r
a
n
t
e
r
i
o
r
c
r
u
c
i
a
t
e
l
i
g
a
m
e
n
t
r
u
p
t
u
r
e
l
o
c
a
l
i
z
a
t
i
o
n
o
n
k
n
e
e
M
R
i
m
a
g
e
s,
”
Fr
o
n
t
i
e
rs
i
n
Bi
o
e
n
g
i
n
e
e
r
i
n
g
a
n
d
Bi
o
t
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
0
,
S
e
p
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
8
9
/
f
b
i
o
e
.
2
0
2
2
.
1
0
2
4
5
2
7
.
[
7
]
Y
.
M
i
n
a
m
o
t
o
e
t
a
l
.
,
“
A
u
t
o
ma
t
e
d
d
e
t
e
c
t
i
o
n
o
f
a
n
t
e
r
i
o
r
c
r
u
c
i
a
t
e
l
i
g
a
me
n
t
t
e
a
r
s
u
s
i
n
g
a
d
e
e
p
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
B
MC
Mu
s
c
u
l
o
s
k
e
l
e
t
a
l
D
i
s
o
r
d
e
rs
,
v
o
l
.
2
3
,
n
o
.
1
,
p
.
5
7
7
,
D
e
c
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
8
6
/
s
1
2
8
9
1
-
0
2
2
-
0
5
5
2
4
-
1.
[
8
]
J.
D
o
n
g
e
t
a
l.
,
“
A
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
e
n
h
a
n
c
e
s c
l
i
n
i
c
i
a
n
s’
d
i
a
g
n
o
st
i
c
a
c
c
u
r
a
c
y
t
o
o
v
e
r
9
0
%
f
o
r
a
n
t
e
r
i
o
r
c
r
u
c
i
a
t
e
l
i
g
a
m
e
n
t
r
u
p
t
u
r
e
d
e
t
e
c
t
i
o
n
o
n
M
R
I
,
”
Ar
t
h
r
o
sc
o
p
y
,
2
0
2
4
.
[
9
]
R
.
C
.
G
o
n
z
a
l
e
z
a
n
d
R
.
E.
W
o
o
d
s
,
D
i
g
i
t
a
l
I
m
a
g
e
Pr
o
c
e
ss
i
n
g
,
4
t
h
e
d
.
2
0
1
8
.
[
10
]
J.
C
A
N
N
Y
,
“
A
c
o
m
p
u
t
a
t
i
o
n
a
l
a
p
p
r
o
a
c
h
t
o
e
d
g
e
d
e
t
e
c
t
i
o
n
,
”
i
n
Re
a
d
i
n
g
s
i
n
C
o
m
p
u
t
e
r
Vi
s
i
o
n
,
E
l
se
v
i
e
r
,
1
9
8
7
,
p
p
.
1
8
4
–
2
0
3
.
[
11
]
S
.
H
a
y
k
i
n
,
“
N
e
u
r
a
l
n
e
t
w
o
r
k
s
a
n
d
l
e
a
r
n
i
n
g
ma
c
h
i
n
e
s,
”
i
n
Pe
a
rs
o
n
Pre
n
t
i
c
e
H
a
l
l
N
e
w
J
e
r
sey
U
S
A
9
3
6
p
L
i
n
k
s
,
v
o
l
.
3
,
2
0
0
8
,
p
.
9
0
6
.
[
12
]
I
.
I
r
mak
c
i
,
S
.
M
.
A
n
w
a
r
,
D
.
A
.
To
r
i
g
i
a
n
,
a
n
d
U
.
B
a
g
c
i
,
“
D
e
e
p
l
e
a
r
n
i
n
g
f
o
r
mu
sc
u
l
o
s
k
e
l
e
t
a
l
i
ma
g
e
a
n
a
l
y
s
i
s,”
C
o
n
f
e
re
n
c
e
Re
c
o
rd
-
Asi
l
o
m
a
r
C
o
n
f
e
re
n
c
e
o
n
S
i
g
n
a
l
s
,
S
y
st
e
m
s
a
n
d
C
o
m
p
u
t
e
rs
,
v
o
l
.
2
0
1
9
-
N
o
v
e
mb
e
r
,
p
p
.
1
4
8
1
–
1
4
8
5
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
I
EEEC
O
N
F
4
4
6
6
4
.
2
0
1
9
.
9
0
4
8
6
7
1
.
[
13
]
C
.
P
.
S
.
K
u
l
se
n
g
,
V
.
N
a
i
n
a
ma
l
a
i
,
E.
G
r
ø
v
i
k
,
J.
-
T
.
G
e
i
t
u
n
g
,
A
.
Å
r
ø
e
n
,
a
n
d
K
.
-
I
.
G
j
e
s
d
a
l
,
“
A
u
t
o
ma
t
i
c
s
e
g
me
n
t
a
t
i
o
n
o
f
h
u
m
a
n
k
n
e
e
a
n
a
t
o
my
b
y
a
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
a
p
p
l
y
i
n
g
a
3
D
M
R
I
p
r
o
t
o
c
o
l
,
”
BM
C
M
u
sc
u
l
o
sk
e
l
e
t
a
l
D
i
so
r
d
e
r
s
,
v
o
l
.
2
4
,
n
o
.
1
,
p
.
4
1
,
Jan
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
8
6
/
s1
2
8
9
1
-
023
-
0
6
1
5
3
-
y.
[
1
4]
C
.
Y
u
,
C
.
Q
i
u
,
a
n
d
Z
.
Z
h
a
n
g
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
f
o
r
a
n
t
e
r
i
o
r
c
r
u
c
i
a
t
e
l
i
g
a
m
e
n
t
a
n
d
m
e
n
i
sc
u
s
a
n
a
l
y
si
s
i
n
k
n
e
e
M
R
I
:
a
c
o
m
p
r
e
h
e
n
si
v
e
r
e
v
i
e
w
,
”
D
i
sp
l
a
y
s
,
v
o
l
.
8
8
,
p
.
1
0
3
0
3
2
,
Ju
l
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
d
i
s
p
l
a
.
2
0
2
5
.
1
0
3
0
3
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
Hyb
r
id
S
V
M
–
A
N
N
s
ys
tem
fo
r
a
u
to
ma
ted
mri
d
ia
g
n
o
s
is
o
f a
n
teri
o
r
…
(
S
a
z
w
a
n
S
ya
fiq
Ma
z
la
n
)
781
[1
5
]
A
.
C
.
K
a
r
a
a
n
d
F
.
H
a
r
d
a
l
a
ç
,
“
D
e
t
e
c
t
i
o
n
a
n
d
c
l
a
ssi
f
i
c
a
t
i
o
n
o
f
K
n
e
e
i
n
j
u
r
i
e
s
f
r
o
m
M
R
I
mag
e
s
U
si
n
g
t
h
e
M
R
N
e
t
d
a
t
a
s
e
t
w
i
t
h
p
r
o
g
r
e
ss
i
v
e
l
y
o
p
e
r
a
t
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
met
h
o
d
s,
”
M
a
c
h
i
n
e
L
e
a
rn
i
n
g
a
n
d
K
n
o
w
l
e
d
g
e
E
x
t
r
a
c
t
i
o
n
,
v
o
l
.
3
,
n
o
.
4
,
p
p
.
1
0
0
9
–
1
0
2
9
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
ma
k
e
3
0
4
0
0
5
0
.
[1
6
]
S
.
S
.
M
a
z
l
a
n
,
“
A
u
t
o
ma
t
e
d
d
i
a
g
n
o
s
i
s
o
f
a
n
t
e
r
i
o
r
c
r
u
c
i
a
t
e
l
i
g
a
me
n
t
(
A
C
L)
k
n
e
e
i
n
j
u
r
i
e
s
u
si
n
g
M
R
I
b
a
se
d
o
n
i
m
a
g
e
p
r
o
c
e
ss
i
n
g
a
n
d
mac
h
i
n
e
l
e
a
r
n
i
n
g
,
”
U
n
i
v
.
K
u
a
l
a
L
u
m
p
u
r
,
2
0
1
9
.
[1
7
]
S
.
S
.
M
a
z
l
a
n
,
M
.
Z.
A
y
o
b
,
a
n
d
Z.
A
.
K
a
d
i
r
,
“
A
n
t
e
r
i
o
r
c
r
u
c
i
a
t
e
l
i
g
a
m
e
n
t
(
A
C
L)
d
i
a
g
n
o
si
s s
y
s
t
e
m
.
”
[
18
]
S
.
S
.
M
a
z
l
a
n
,
M
.
Z.
A
y
o
b
,
Z.
A
.
K
.
B
a
k
t
i
,
S
.
R
.
M
u
k
h
t
a
r
,
a
n
d
N
.
A
.
W
a
h
a
b
,
“
D
e
v
e
l
o
p
me
n
t
o
f
a
n
t
e
r
i
o
r
c
r
u
c
i
a
t
e
l
i
g
a
m
e
n
t
(
A
C
L
)
d
i
a
g
n
o
si
s
sy
s
t
e
m
u
si
n
g
i
m
a
g
e
p
r
o
c
e
ssi
n
g
,
”
J
o
u
r
n
a
l
o
f
F
u
n
d
a
m
e
n
t
a
l
a
n
d
Ap
p
l
i
e
d
S
c
i
e
n
c
e
s
,
v
o
l
.
9
,
n
o
.
5
S
,
p
.
4
3
9
,
J
a
n
.
2
0
1
8
,
d
o
i
:
1
0
.
4
3
1
4
/
j
f
a
s.
v
9
i
5
s.
3
1
.
[
19
]
S
.
S
.
M
a
z
l
a
n
,
M
.
Z
.
A
y
o
b
,
Z
.
A
.
K
a
d
i
r
,
a
n
d
S
.
R
.
M
u
k
h
t
a
r
,
“
D
e
v
e
l
o
p
m
e
n
t
o
f
a
n
t
e
r
i
o
r
c
r
u
c
i
a
t
e
l
i
g
a
m
e
n
t
(
A
C
L
)
K
n
e
e
i
n
j
u
r
y
c
l
a
s
s
i
f
i
c
a
t
i
o
n
s
y
s
t
e
m
f
r
o
m
m
a
g
n
e
t
i
c
r
e
s
o
n
a
n
c
e
i
m
a
g
e
(
M
R
I
)
u
s
i
n
g
f
u
z
z
y
i
n
f
e
r
e
n
c
e
s
y
s
t
e
m
.
”
P
r
o
c
.
S
c
i
.
E
n
g
.
T
e
c
h
n
o
l
.
N
a
t
.
C
o
n
f
.
,
2
0
1
5
.
[2
0
]
S
.
S
.
M
a
z
l
a
n
,
M
.
I
.
Y
u
s
o
f
,
a
n
d
S
.
S
.
H
a
f
sh
a
r
,
“
A
d
e
v
e
l
o
p
m
e
n
t
o
f
a
n
t
e
r
i
o
r
c
r
u
c
i
a
t
e
l
i
g
a
m
e
n
t
(
A
c
l
)
d
i
a
g
n
o
s
i
s
s
y
s
t
e
m
u
si
n
g
f
u
z
z
y
i
n
f
e
r
e
n
c
e
s
y
s
t
e
m,
”
Ma
l
a
y
si
a
n
J
o
u
r
n
a
l
o
f
I
n
d
u
st
r
i
a
l
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
,
n
o
.
1
,
2
0
1
6
.
[2
1
]
S
.
S
.
M
a
z
l
a
n
,
M
.
Z.
A
y
o
b
,
a
n
d
Z.
A
.
K
.
B
a
k
t
i
,
“
A
n
t
e
r
i
o
r
c
r
u
c
i
a
t
e
l
i
g
a
men
t
(
A
C
L)
i
n
j
u
r
y
c
l
a
ss
i
f
i
c
a
t
i
o
n
sy
s
t
e
m
u
s
i
n
g
s
u
p
p
o
r
t
v
e
c
t
o
r
ma
c
h
i
n
e
(
S
V
M
)
,
”
i
n
2
0
1
7
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
E
n
g
i
n
e
e
r
i
n
g
T
e
c
h
n
o
l
o
g
y
a
n
d
T
e
c
h
n
o
p
r
e
n
e
u
rsh
i
p
(
I
C
E
2
T
)
,
S
e
p
.
2
0
1
7
,
v
o
l
.
2
0
1
7
-
J
a
n
u
a
,
p
p
.
1
–
5
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
E2
T.
2
0
1
7
.
8
2
1
5
9
6
0
.
[
2
2]
U
.
C
o
r
t
e
s
,
C
o
r
i
n
n
a
(
A
T&
TB
e
l
l
La
b
s.,
H
o
h
n
d
e
l
,
N
J0
7
7
3
3
a
n
d
U
.
V
l
a
d
i
m
i
r
,
V
a
p
n
i
k
(
A
T&T
B
e
l
l
La
b
s.,
H
o
h
n
d
e
l
,
N
J
0
7
7
3
3
,
“
S
u
p
p
o
r
t
-
v
e
c
t
o
r
n
e
t
w
o
r
k
s
,
”
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
,
v
o
l
.
2
9
7
,
n
o
.
2
0
,
p
p
.
2
7
3
–
2
9
7
,
1
9
9
5
.
[
2
3]
R
.
A
n
d
r
a
d
e
,
R
.
P
e
r
e
i
r
a
,
R
.
V
a
n
C
i
n
g
e
l
,
J
.
B
.
S
t
a
a
l
,
a
n
d
J.
Esp
r
e
g
u
e
i
r
a
-
M
e
n
d
e
s,
“
H
o
w
s
h
o
u
l
d
c
l
i
n
i
c
i
a
n
s
r
e
h
a
b
i
l
i
t
a
t
e
p
a
t
i
e
n
t
s
a
f
t
e
r
A
C
L
r
e
c
o
n
s
t
r
u
c
t
i
o
n
?
A
sy
st
e
mat
i
c
r
e
v
i
e
w
o
f
c
l
i
n
i
c
a
l
p
r
a
c
t
i
c
e
g
u
i
d
e
l
i
n
e
s
(
C
P
G
s)
w
i
t
h
a
f
o
c
u
s
o
n
q
u
a
l
i
t
y
a
p
p
r
a
i
sal
(
A
G
R
EE
I
I
)
,
”
Bri
t
i
sh
J
o
u
rn
a
l
o
f
S
p
o
r
t
s
Me
d
i
c
i
n
e
,
v
o
l
.
5
4
,
n
o
.
9
,
p
p
.
5
1
2
–
5
1
9
,
M
a
y
2
0
2
0
,
d
o
i
:
1
0
.
1
1
3
6
/
b
j
s
p
o
r
t
s
-
2
0
1
8
-
1
0
0
3
1
0
.
[
2
4
]
N
.
O
t
s
u
,
“
A
t
h
r
e
s
h
o
l
d
s
e
l
e
c
t
i
o
n
me
t
h
o
d
f
r
o
m
g
r
a
y
-
l
e
v
e
l
h
i
s
t
o
g
r
a
ms,”
I
E
EE
T
r
a
n
s
a
c
t
i
o
n
s
o
n
S
y
st
e
m
s,
Ma
n
,
a
n
d
C
y
b
e
r
n
e
t
i
cs
,
v
o
l
.
9
,
n
o
.
1
,
p
p
.
6
2
–
6
6
,
1
9
7
9
.
[
2
5
]
C
.
W
.
A
.
P
f
i
r
r
man
n
,
M
.
Z
a
n
e
t
t
i
,
J.
R
o
m
e
r
o
,
a
n
d
J.
H
o
d
l
e
r
,
“
F
e
m
o
r
a
l
t
r
o
c
h
l
e
a
r
d
y
sp
l
a
s
i
a
:
M
R
f
i
n
d
i
n
g
s,
”
R
a
d
i
o
l
o
g
y
,
v
o
l
.
2
1
6
,
n
o
.
3
,
p
p
.
8
5
8
–
8
6
4
,
S
e
p
.
2
0
0
0
,
d
o
i
:
1
0
.
1
1
4
8
/
r
a
d
i
o
l
o
g
y
.
2
1
6
.
3
.
r
0
0
se
3
8
8
5
8
.
[
2
6
]
D
.
A
.
R
u
b
i
n
,
“
M
R
i
ma
g
i
n
g
o
f
t
h
e
k
n
e
e
m
e
n
i
sc
i
,
”
Ra
d
i
o
l
o
g
i
c
C
l
i
n
i
c
s
o
f
N
o
rt
h
Am
e
r
i
c
a
,
v
o
l
.
3
5
,
n
o
.
1
,
p
p
.
2
1
–
4
4
,
1
9
9
7
,
d
o
i
:
1
0
.
1
0
1
6
/
s
0
0
3
3
-
8
3
8
9
(
2
2
)
0
0
5
7
7
-
2.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
S
a
z
wa
n
S
y
a
f
iq
M
a
z
la
n
is
a
re
se
a
rc
h
e
r
a
t
th
e
Na
ti
o
n
a
l
De
fe
n
c
e
Un
iv
e
rsity
o
f
M
a
lay
sia
(UPNM
)
wh
o
se
wo
r
k
sp
a
n
s
e
n
v
ir
o
n
m
e
n
tal
m
o
d
e
li
n
g
,
a
rt
ifi
c
ial
in
telli
g
e
n
c
e
,
ro
b
o
ti
c
s
,
e
m
b
e
d
d
e
d
sy
ste
m
s,
a
n
d
m
e
d
ica
l
ima
g
in
g
.
His
i
n
terd
isc
ip
li
n
a
ry
re
se
a
rc
h
fo
c
u
se
s
o
n
d
e
v
e
lo
p
i
n
g
c
o
m
p
u
tati
o
n
a
l
fra
m
e
wo
rk
s
t
h
a
t
in
teg
ra
te
a
rti
ficia
l
i
n
telli
g
e
n
c
e
f
o
r
d
iag
n
o
stic
ima
g
in
g
a
n
d
e
n
v
iro
n
m
e
n
tal
f
o
re
c
a
stin
g
.
Am
o
n
g
h
is
n
o
tab
le
c
o
n
tri
b
u
ti
o
n
s
is
th
e
2
0
2
2
a
rti
c
le
“
In
c
o
rp
o
ra
ti
n
g
AN
N
with
P
CR
f
o
r
p
ro
g
re
ss
iv
e
d
e
v
e
l
o
p
i
n
g
o
f
a
ir
p
o
ll
u
ti
o
n
in
d
e
x
f
o
re
c
a
st
”
p
u
b
li
sh
e
d
i
n
P
la
n
n
i
n
g
M
a
lay
sia
,
w
h
e
re
h
e
p
ro
p
o
se
d
a
h
y
b
rid
m
o
d
e
l
c
o
m
b
i
n
in
g
p
rin
c
i
p
a
l
c
o
m
p
o
n
e
n
t
re
g
re
ss
io
n
(
P
CR)
a
n
d
AN
N
to
p
r
e
d
ict
t
h
e
a
ir
p
o
l
lu
tan
t
i
n
d
e
x
(API
)
in
S
e
lan
g
o
r
.
I
n
t
h
e
fiel
d
o
f
m
e
d
ica
l
ima
g
in
g
,
h
e
a
u
th
o
re
d
th
e
b
o
o
k
“
Dia
g
n
o
si
n
g
a
n
teri
o
r
c
r
u
c
iate
li
g
a
m
e
n
t
(ACL)
k
n
e
e
in
ju
ries
M
RI:
a
n
e
n
se
m
b
le
sc
re
e
n
in
g
a
n
d
lea
rn
i
n
g
sy
ste
m
”
a
n
d
c
o
-
d
e
v
e
l
o
p
e
d
a
n
a
u
t
o
m
a
ted
ACL
d
iag
n
o
sis
sy
ste
m
u
sin
g
a
d
v
a
n
c
e
d
ima
g
e
-
p
r
o
c
e
ss
in
g
a
n
d
m
a
c
h
in
e
-
lea
rn
in
g
m
e
th
o
d
s.
His
re
se
a
r
c
h
c
o
n
ti
n
u
e
s
to
e
m
p
h
a
siz
e
p
ra
c
ti
c
a
l
i
m
p
lem
e
n
tatio
n
s
o
f
AI
fo
r
c
li
n
ica
l
a
n
d
e
n
v
iro
n
m
e
n
tal
d
e
c
isio
n
s
u
p
p
o
rt
.
He
c
a
n
b
e
re
a
c
h
e
d
a
t
e
m
a
il
:
sa
z
w
a
n
@u
p
n
m
.
e
d
u
.
m
y
.
Az
izi
Mi
sko
n
re
se
a
rc
h
c
o
n
tri
b
u
ti
o
n
s
sp
a
n
b
i
o
m
e
d
ica
l
e
n
g
i
n
e
e
rin
g
,
f
o
c
u
si
n
g
o
n
ste
m
c
e
ll
d
iffere
n
ti
a
ti
o
n
a
n
d
t
h
e
i
n
flu
e
n
c
e
o
f
m
a
g
n
e
ti
c
field
s
o
n
c
e
ll
u
lar
b
e
h
a
v
io
r
.
In
2
0
2
4
,
h
e
co
-
a
u
th
o
re
d
a
sy
ste
m
a
ti
c
re
v
iew
in
P
e
e
rJ
d
e
taili
n
g
m
o
rp
h
o
l
o
g
ica
l
a
n
d
sta
in
i
n
g
tec
h
n
iq
u
e
s
c
rit
ica
l
fo
r
a
n
a
ly
z
i
n
g
o
ste
o
b
las
t
a
n
d
o
ste
o
c
las
t
fo
rm
a
ti
o
n
.
His
i
n
t
e
rd
isc
ip
l
in
a
r
y
b
o
d
y
o
f
wo
rk
a
lso
c
o
v
e
rs
b
i
o
se
n
so
rs,
d
r
u
g
d
e
li
v
e
ry
m
e
th
o
d
o
l
o
g
ies
,
EE
G
-
b
a
se
d
p
ro
sth
e
t
ic
c
o
n
tr
o
l,
d
e
fe
n
se
-
re
late
d
ti
ss
u
e
e
n
g
in
e
e
rin
g
,
fra
c
tu
re
m
e
c
h
a
n
ics
si
m
u
latio
n
s,
a
n
d
e
v
e
n
c
y
b
e
rse
c
u
rit
y
stra
teg
y
d
e
v
e
lo
p
m
e
n
t.
He
c
a
n
b
e
c
o
n
tac
te
d
a
t
e
m
a
il
:
a
z
izim
isk
o
n
@
u
p
n
m
.
e
d
u
.
m
y
.
S
h
a
r
iza
l
Ah
m
a
d
S
o
b
r
i
is
a
r
e
se
a
rc
h
e
r
a
t
No
tt
in
g
h
a
m
Tren
t
Un
iv
e
rsit
y
(NTU),
UK
,
wh
o
se
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
las
e
r
m
a
teria
ls
p
ro
c
e
ss
in
g
,
a
d
v
a
n
c
e
d
m
a
c
h
in
i
n
g
,
su
sta
in
a
b
le
a
n
d
a
d
d
it
iv
e
m
a
n
u
f
a
c
tu
rin
g
,
a
n
d
e
n
g
i
n
e
e
rin
g
m
a
n
a
g
e
m
e
n
t.
He
h
a
s
a
u
t
h
o
re
d
n
u
m
e
ro
u
s
jo
u
rn
a
l
a
rti
c
les
a
n
d
b
o
o
k
c
h
a
p
ters
i
n
t
h
e
se
field
s,
c
o
n
tri
b
u
ti
n
g
to
a
d
v
a
n
c
e
s
i
n
p
re
c
isio
n
m
a
n
u
fa
c
tu
r
in
g
,
c
o
m
p
o
site
m
a
teria
l
p
ro
c
e
ss
in
g
,
a
n
d
su
sta
in
a
b
le
m
a
teria
l
tec
h
n
o
l
o
g
ies
.
His
wo
r
k
p
a
rti
c
u
la
rly
f
o
c
u
se
s
o
n
a
d
v
a
n
c
e
d
m
a
c
h
in
in
g
o
f
c
o
m
p
o
site
m
a
teria
ls
e
sp
e
c
ially
c
a
rb
o
n
fi
b
e
r
–
re
in
f
o
rc
e
d
p
o
l
y
m
e
rs
(CF
RP
)
u
sin
g
las
e
r
a
n
d
h
y
b
ri
d
p
r
o
c
e
ss
e
s
a
s
we
ll
a
s
d
e
v
e
l
o
p
m
e
n
t
o
f
s
u
sta
in
a
b
le
w
o
o
d
-
b
a
se
d
p
r
o
d
u
c
ts,
g
r
e
e
n
m
a
teria
ls,
su
rfa
c
e
a
n
d
c
h
e
m
ica
l
m
o
d
ifi
c
a
ti
o
n
s
fo
r
e
n
v
iro
n
m
e
n
tal
a
p
p
li
c
a
ti
o
n
s.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
sh
a
riza
l.
a
h
m
a
d
so
b
ri@
n
tu
.
a
c
.
u
k
.
Evaluation Warning : The document was created with Spire.PDF for Python.