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.
37
,
No
.
3
,
Ma
r
ch
20
25
,
p
p
.
2
0
0
9
~
20
20
I
SS
N:
2
502
-
4
7
52
,
DOI
: 1
0
.
1
1
5
9
1
/ijee
cs
.v
37.
i
3
.
p
p
200
9
-
20
20
2009
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs
.
ia
esco
r
e.
co
m
O
ptimi
zed d
ense
co
nv
o
lutiona
l net
wo
rk wit
h condi
ti
o
na
l
a
utoreg
ress
iv
e va
lue
-
at
-
r
isk for
chr
o
nic kidney
disea
se dete
ction
throug
h gro
up
-
ba
sed sea
rch
Chet
a
n Nim
ba
Aher
1
,
Arc
ha
na
Ra
j
esh
Da
t
e
2
,
Sh
ridev
i S.
Va
s
ek
a
r
3
,
P
riy
a
nk
a
T
up
e
-
Wa
g
hm
a
re
4
,
Am
ra
pa
li Shi
v
a
j
ira
o
Cha
v
a
n
1
1
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
En
g
i
n
e
e
r
i
n
g
,
A
I
S
S
M
S
I
n
st
i
t
u
t
e
o
f
I
n
f
o
r
m
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
P
u
n
e
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
E&T
C
E
n
g
i
n
e
e
r
i
n
g
,
H
S
B
P
V
T’
s GO
I
,
F
a
c
u
l
t
y
o
f
E
n
g
i
n
e
e
r
i
n
g
,
A
h
me
d
n
a
g
a
r
,
I
n
d
i
a
3
D
e
p
a
r
t
me
n
t
o
f
E&T
C
f
r
o
m SC
TR
’
s
P
u
n
e
I
n
st
i
t
u
t
e
o
f
C
o
m
p
u
t
e
r
Te
c
h
n
o
l
o
g
y
,
P
u
n
e
,
I
n
d
i
a
4
S
y
m
b
i
o
si
s
I
n
st
i
t
u
t
e
o
f
Te
c
h
n
o
l
o
g
y
,
S
y
mb
i
o
si
s
I
n
t
e
r
n
a
t
i
o
n
a
l
(
D
e
e
me
d
U
n
i
v
e
r
si
t
y
)
,
P
u
n
e
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ap
r
4
,
2
0
2
4
R
ev
is
ed
Sep
29
,
2
0
2
4
Acc
ep
ted
Oct
7
,
2
0
2
4
Ch
ro
n
ic
k
i
d
n
e
y
d
ise
a
se
(CKD
)
is
th
e
g
ra
d
u
a
l
d
e
c
re
a
se
in
re
n
a
l
fu
n
c
ti
o
n
a
li
t
y
th
a
t
lea
d
s
to
k
id
n
e
y
fa
il
u
re
o
r
d
a
m
a
g
e
.
Th
is
d
ise
a
se
is
th
e
m
o
st
se
v
e
re
wo
rld
wi
d
e
h
e
a
lt
h
c
o
n
d
it
i
o
n
th
a
t
k
i
ll
s
n
u
m
e
ro
u
s
p
e
o
p
le
e
v
e
ry
y
e
a
r
a
s
a
n
o
u
tco
m
e
o
f
h
e
re
d
i
tary
fa
c
to
rs
a
n
d
wo
rse
li
fe
sty
les
.
As
CKD
p
ro
g
re
ss
e
s,
it
b
e
c
o
m
e
s
d
iffi
c
u
l
t
t
o
d
iag
n
o
se
.
Ut
il
izin
g
re
g
u
lar
d
o
c
t
o
r
c
o
n
su
lt
a
ti
o
n
d
a
ta
fo
r
e
v
a
lu
a
ti
n
g
d
i
v
e
rse
p
h
a
se
s
o
f
CKD
c
a
n
a
ss
ist
in
e
a
rli
e
r
d
e
tec
ti
o
n
a
n
d
ti
m
e
l
y
in
fe
re
n
c
e
.
F
u
rth
e
rm
o
re
,
e
ffe
c
tu
a
l
d
e
tec
ti
o
n
m
e
th
o
d
s
a
re
v
it
a
l
o
w
i
n
g
to
a
n
in
c
re
a
se
d
c
o
u
n
t
o
f
p
a
ti
e
n
ts
wit
h
CKD
.
He
re
,
g
ro
u
p
se
a
rc
h
c
o
n
d
it
io
n
a
l
a
u
to
re
g
re
ss
iv
e
v
a
l
u
e
-
at
-
risk
b
a
se
d
d
e
n
se
c
o
n
v
o
lu
ti
o
n
a
l
n
e
two
rk
(G
S
CAV
iaR
-
De
n
se
Ne
t)
is
in
tro
d
u
c
e
d
fo
r
CKD
d
e
tec
ti
o
n
.
F
irstl
y
,
c
h
r
o
n
i
c
d
a
ta
is
a
c
q
u
ired
fr
o
m
t
h
e
d
a
tas
e
t
a
n
d
M
i
n
-
M
a
x
n
o
rm
a
li
z
a
ti
o
n
is
u
ti
li
z
e
d
t
o
p
re
-
p
ro
c
e
ss
c
o
n
sid
e
re
d
c
h
r
o
n
ic
k
id
n
e
y
d
a
ta.
Th
e
re
a
fter,
fe
a
tu
re
se
lec
ti
o
n
(F
S
)
is
p
e
rf
o
rm
e
d
b
a
se
d
o
n
T
o
p
so
e
sim
il
a
rit
y
.
Las
tl
y
,
CKD
d
e
tec
ti
o
n
is
e
x
e
c
u
ted
b
y
d
e
n
se
c
o
n
v
o
lu
ti
o
n
a
l
n
e
two
rk
(De
n
se
Ne
t)
a
n
d
g
ro
u
p
se
a
rc
h
c
o
n
d
i
ti
o
n
a
l
a
u
to
re
g
re
ss
iv
e
v
a
lu
e
-
at
-
risk
(G
S
CAV
iaR)
is
e
m
p
lo
y
e
d
to
trai
n
D
e
n
s
e
N
e
t
.
H
o
we
v
e
r
,
G
S
C
AV
i
a
R
i
s
d
e
s
i
g
n
e
d
b
y
i
n
c
o
r
p
o
r
a
t
i
n
g
a
g
r
o
u
p
s
e
a
r
c
h
o
p
t
i
m
i
z
e
r
(G
S
O
)
w
i
t
h
a
c
o
n
d
i
t
i
o
n
a
l
a
u
t
o
r
e
g
r
e
s
s
i
v
e
v
a
l
u
e
-
at
-
r
is
k
(C
A
V
i
a
R
)
m
o
d
e
l
.
A
d
d
i
t
i
o
n
a
l
l
y
,
G
S
C
A
V
ia
R
-
D
e
n
s
e
N
e
t
a
c
q
u
i
r
e
d
a
m
a
x
i
m
a
l
a
c
c
u
r
a
c
y
o
f
a
b
o
u
t
9
1
.
5
%
,
s
e
n
s
i
t
i
v
i
t
y
o
f
a
b
o
u
t
9
2
.
8
%
a
n
d
s
p
e
c
i
f
i
c
i
t
y
o
f
a
b
o
u
t
9
0
.
7
%
.
K
ey
w
o
r
d
s
:
C
h
r
o
n
ic
k
id
n
ey
d
is
ea
s
e
Den
s
e
co
n
v
o
lu
tio
n
al
n
etwo
r
k
Gr
o
u
p
s
ea
r
ch
o
p
tim
izer
Min
-
m
ax
n
o
r
m
aliza
tio
n
T
o
p
s
o
e
s
im
ilar
ity
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
:
C
h
etan
Nim
b
a
Ah
er
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
E
n
g
i
n
ee
r
in
g
,
AI
SS
MS
I
n
s
ti
tu
te
o
f
I
n
f
o
r
m
atio
n
T
ec
h
n
o
lo
g
y
Pu
n
e,
Ma
h
ar
ash
tr
a,
4
1
1
0
0
1
,
I
n
d
ia
E
m
ail: c
h
etan
.
ah
er
0
7
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
p
r
o
b
lem
o
f
ch
r
o
n
ic
k
id
n
ey
d
is
ea
s
e
(
C
KD)
d
etec
tio
n
i
s
a
s
ig
n
if
ican
t
ch
allen
g
e
in
th
e
m
ed
ical
f
ield
.
C
KD
i
s
a
ter
m
th
at
d
ef
i
n
es
to
th
e
s
tate,
wh
er
eu
p
o
n
k
id
n
ey
s
ca
n
n
o
lo
n
g
er
f
ilter
b
lo
o
d
m
o
r
e
ef
f
icien
tl
y
[
1
]
.
Pre
v
io
u
s
r
esear
ch
h
as
h
i
g
h
lig
h
ted
v
ar
io
u
s
m
et
h
o
d
o
lo
g
ie
s
f
o
r
ea
r
l
y
d
etec
tio
n
,
b
u
t
lim
itatio
n
s
r
em
ain
.
T
h
is
r
esear
ch
ad
d
r
ess
es
th
ese
lim
i
tatio
n
s
b
y
in
tr
o
d
u
cin
g
a
n
o
v
el
ap
p
r
o
ac
h
th
at
lev
e
r
ag
es
th
e
o
p
tim
ized
d
en
s
e
co
n
v
o
l
u
tio
n
al
n
etwo
r
k
(
Den
s
eNe
t)
with
co
n
d
itio
n
al
au
t
o
r
eg
r
ess
iv
e
v
alu
e
-
at
-
r
is
k
(
C
AViaR)
f
o
r
C
KD
d
etec
tio
n
th
r
o
u
g
h
g
r
o
u
p
-
b
ased
s
ea
r
ch
.
T
h
is
ap
p
r
o
ac
h
n
o
t
o
n
l
y
im
p
r
o
v
es
d
etec
tio
n
ac
cu
r
ac
y
b
u
t
also
p
r
o
v
i
d
es
a
m
o
r
e
d
etailed
an
aly
s
is
o
f
r
is
k
f
ac
to
r
s
.
Mo
r
eo
v
er
,
C
KD
is
ca
teg
o
r
ized
b
y
a
g
r
ad
u
al
d
ec
r
ea
s
e
in
k
id
n
ey
f
u
n
ctio
n
i
n
g
th
at
d
a
m
ag
es
r
en
al
o
r
g
an
f
u
n
ctio
n
s
.
Owin
g
to
t
h
e
lack
o
f
o
b
v
io
u
s
s
y
m
p
to
m
s
in
ea
r
lier
s
tag
es,
th
e
b
eg
in
n
i
n
g
o
f
r
en
al
f
ailu
r
es
m
a
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
37
,
No
.
3
,
Ma
r
ch
20
25
:
2
0
0
9
-
20
20
2010
f
ir
s
tly
n
o
t
h
av
e
b
ee
n
id
en
tifi
ed
[
2
]
.
I
t
is
clea
r
th
at
C
KD
af
f
ec
ts
an
y
p
er
s
o
n
an
d
s
o
m
e
o
f
p
eo
p
le
a
r
e
m
o
r
e
v
u
ln
er
ab
le
to
th
is
d
is
ea
s
e
th
a
n
o
th
er
s
p
ar
ticu
lar
ly
,
p
atien
ts
h
av
in
g
h
ea
r
t
p
r
o
b
lem
s
,
d
iab
etes,
an
d
ab
n
o
r
m
al
p
o
tass
iu
m
o
r
ca
lciu
m
le
v
els.
As
C
KD
in
cr
ea
s
es,
b
o
d
y
m
ay
co
llect
to
o
m
u
ch
q
u
an
tity
o
f
f
l
u
id
,
waste
p
r
o
d
u
cts
an
d
elec
tr
o
l
y
tes
[
1
]
.
C
KD
is
an
in
cr
ea
s
in
g
ly
s
er
io
u
s
co
n
d
it
io
n
in
t
h
e
cu
r
r
en
t
a
g
ein
g
co
m
m
u
n
ity
.
T
h
e
a
g
ed
p
o
p
u
latio
n
an
d
r
elate
d
h
ig
h
h
y
p
er
ten
s
io
n
en
h
an
ce
th
e
o
c
cu
r
r
en
ce
o
f
h
y
p
er
g
ly
c
em
ia
a
n
d
h
y
p
er
lip
id
em
ia,
th
er
eb
y
in
c
r
ea
s
in
g
C
KD
in
cid
en
ce
[
3
]
.
I
n
ad
d
itio
n
,
C
KD
is
a
co
m
m
o
n
ca
teg
o
r
y
o
f
k
id
n
e
y
d
is
ea
s
e
th
at
ca
n
b
e
o
n
ly
cu
r
e
d
ef
f
o
r
tles
s
ly
wh
en
it
is
d
etec
ted
at
ea
r
lier
s
tag
es
[
4
]
.
Ho
wev
er
,
th
is
d
is
ea
s
e
h
as
n
o
s
y
m
p
to
m
s
in
its
ea
r
lier
s
tag
e;
test
in
g
is
an
o
n
ly
m
o
d
e
to
id
en
tify
w
h
eth
er
p
atien
t is af
f
ec
ted
with
k
id
n
ey
d
is
ea
s
e
o
r
n
o
t [
5
]
.
E
ar
lier
d
etec
tio
n
o
f
C
KD
in
its
b
eg
in
n
in
g
p
h
ases
ca
n
ass
is
t
th
e
p
atien
ts
in
g
ettin
g
ef
f
ec
tu
al
tr
ea
tm
en
ts
an
d
th
en
,
p
r
e
v
en
t
th
e
d
ev
elo
p
m
en
t
o
f
en
d
-
s
ta
g
e
r
en
al
d
is
ea
s
e
(
E
SR
D)
th
at
n
ee
d
s
a
k
id
n
ey
tr
an
s
p
lan
t
o
r
d
ialy
s
is
to
en
h
a
n
ce
th
e
p
atien
t’
s
life
[
5
]
.
Hen
ce
,
ce
r
tain
b
l
o
o
d
an
d
u
r
in
e
test
s
ar
e
tak
en
f
o
r
d
etec
tin
g
C
KD.
Ho
wev
er
,
d
et
ec
tin
g
C
KD
at
th
e
s
tar
tin
g
s
tag
es
is
n
o
t
s
im
p
le
with
o
u
t
ac
c
u
r
ate
ex
am
in
atio
n
s
[
6
]
.
A
d
d
itio
n
ally
,
C
KD
h
as
h
ig
h
er
m
o
r
tality
a
n
d
m
o
r
b
id
i
ty
,
with
a
c
o
m
p
r
eh
en
s
iv
e
im
p
ac
t
o
n
th
e
h
u
m
an
b
o
d
y
.
C
KD
d
iag
n
o
s
is
is
cr
u
cial
an
d
m
ay
b
e
ca
p
ab
le
o
f
o
b
ta
in
in
g
tim
ely
tr
ea
tm
e
n
ts
as
it
is
an
ir
r
ev
e
r
s
ib
le
an
d
p
r
o
g
r
ess
iv
e
p
ath
o
lo
g
ic
s
y
n
d
r
o
m
e
[
7
]
.
Acc
u
r
ate
m
a
n
ag
e
m
en
t
o
f
C
KD
is
p
iv
o
tal
f
o
r
p
r
o
tectin
g
th
e
f
u
n
ctio
n
ality
o
f
k
id
n
ey
s
,
d
ec
r
ea
s
in
g
d
is
ea
s
e
d
ev
elo
p
m
en
t
an
d
en
h
an
cin
g
p
atien
t
r
esu
lts
[
8
]
.
T
h
e
e
x
is
tin
g
r
esear
ch
er
s
h
av
e
r
e
v
ea
led
th
at
m
ac
h
in
e
lear
n
in
g
(
ML
)
a
n
d
d
ee
p
lear
n
i
n
g
(
DL
)
m
et
h
o
d
s
ca
n
b
e
em
p
lo
y
ed
f
o
r
th
e
ac
cu
r
ate
d
iag
n
o
s
is
o
f
C
KD
[
9
]
.
E
m
p
lo
y
in
g
DL
’
s
s
k
ill
d
is
co
v
er
y
ab
ilit
ies
lik
e
cla
s
s
if
icatio
n
an
d
d
ata
m
in
in
g
ap
p
r
o
ac
h
es
[
10
]
,
it
is
p
r
esen
tly
p
r
o
b
ab
le
to
m
a
n
ag
e
v
alu
ab
le
an
d
h
u
g
e
d
ata
f
o
r
en
h
an
cin
g
clin
ical
p
r
o
g
n
o
s
is
an
d
d
iag
n
o
s
is
in
d
ec
is
io
n
-
m
ak
in
g
[
1
1
]
,
[
1
2
]
.
W
h
en
h
ea
lth
ca
r
e
p
r
o
v
id
er
s
i
n
teg
r
ate
th
is
d
ata
with
o
th
er
in
f
o
r
m
atio
n
s
o
u
r
ce
s
,
th
ey
ca
n
d
ev
elo
p
n
ewe
r
s
o
lu
tio
n
s
with
an
ass
is
tan
ce
o
f
p
r
ed
ictiv
e
an
aly
s
is
f
o
r
ea
r
lier
C
KD
d
etec
tio
n
,
r
elate
d
h
ea
lth
th
r
ea
ts
an
d
ev
en
p
r
escr
ip
tiv
e
an
aly
s
is
f
o
r
th
e
p
r
ec
is
io
n
m
ed
icin
es [
1
3
]
.
T
h
e
v
ital
aim
is
to
in
tr
o
d
u
c
e
GSC
AVia
R
-
Den
s
eNe
t
f
o
r
C
KD
d
etec
tio
n
.
C
KD
is
p
r
o
g
r
ess
iv
ely
ac
k
n
o
wled
g
e
d
as
a
wo
r
ld
wid
e
h
ea
lth
is
s
u
e
an
d
an
im
p
o
r
t
an
t
d
eter
m
in
an
t
o
f
wo
r
s
e
h
e
alth
r
esu
lts
.
I
n
th
is
r
esear
ch
,
ch
r
o
n
ic
k
id
n
e
y
d
ata
is
tak
en
f
r
o
m
a
s
p
ec
if
ic
d
ata
s
et.
T
h
en
,
p
r
e
-
p
r
o
ce
s
s
in
g
is
c
o
n
d
u
cte
d
u
tili
zin
g
Min
-
Ma
x
n
o
r
m
aliza
tio
n
.
Af
t
er
p
r
e
-
p
r
o
c
ess
in
g
o
f
d
ata,
FS
is
ac
co
m
p
lis
h
ed
b
ased
o
n
T
o
p
s
o
e
s
im
ilar
ity
.
Fin
ally
,
C
KD
is
d
etec
ted
u
ti
lizin
g
Den
s
eNe
t
an
d
GSC
A
ViaR
d
o
es
its
tr
ain
in
g
.
Ho
wev
er
,
GSC
AViaR
i
s
d
esig
n
ed
b
y
j
o
in
in
g
GSO
with
C
AViaR.
Pro
p
o
s
ed
GSC
AViaR
-
Den
s
eNe
t
f
o
r
C
KD
d
etec
tio
n
:
No
wad
ay
s
,
ea
r
lier
d
etec
tio
n
o
f
C
KD
an
d
i
ts
co
m
p
lex
ities
s
ee
m
to
b
e
v
er
y
cr
u
cial
f
o
r
en
h
an
cin
g
a
p
atie
n
t’
s
life
.
Her
e,
th
e
d
etec
tio
n
o
f
C
KD
is
co
n
d
u
cte
d
b
y
De
n
s
eNe
t.
Ho
wev
er
,
De
n
s
eNe
t
is
tr
ain
ed
u
tili
zin
g
GSC
AVia
R
wh
ich
is
m
o
d
elled
b
y
co
m
b
in
in
g
GSO
with
C
AVia
R
.
C
KD
is
al
s
o
ter
m
ed
as
ch
r
o
n
ic
r
en
al
d
is
ea
s
e,
wh
er
ein
k
id
n
ey
s
f
ail
to
f
u
n
ctio
n
g
r
a
d
u
ally
.
Fo
r
r
ed
u
cin
g
th
e
ch
an
c
es
o
f
C
KD
th
at
lead
to
k
id
n
ey
tr
an
s
p
lan
ta
tio
n
o
r
d
ialy
s
is
,
ea
r
lier
C
KD
d
etec
tio
n
is
cr
u
cial.
T
h
is
m
o
tiv
ated
,
T
o
d
esig
n
a
m
eth
o
d
to
d
etec
t
C
KD
b
y
r
ev
iewin
g
cu
r
r
e
n
t
ap
p
r
o
ac
h
es
d
ev
elo
p
ed
f
o
r
C
KD
d
etec
tio
n
.
T
h
e
r
ev
iewe
d
tech
n
iq
u
es a
lo
n
g
with
th
ei
r
ad
v
a
n
ta
g
es a
n
d
ch
allen
g
es a
r
e
in
ter
p
r
eted
in
th
is
p
ar
t.
Saif
et
a
l.
[
1
4
]
d
esig
n
ed
a
d
ee
p
en
s
em
b
le
m
o
d
el
f
o
r
C
KD
p
r
ed
ictio
n
.
I
t
en
h
a
n
ce
d
co
m
p
licated
f
ea
tu
r
e
d
ep
ictio
n
s
an
d
p
er
f
o
r
m
ed
b
etter
in
class
if
icatio
n
tas
k
s
.
Nev
er
th
eless
,
th
is
m
eth
o
d
f
ailed
to
in
v
esti
g
ate
th
e
r
o
b
u
s
tn
ess
o
f
th
is
m
o
d
el.
R
ao
et
a
l.
[
1
5
]
p
r
esen
ted
a
f
u
s
io
n
DL
m
o
d
el
f
o
r
th
e
p
r
e
d
ic
tio
n
o
f
C
KD.
T
h
is
ap
p
r
o
ac
h
was
id
ea
l
f
o
r
m
ed
ical
ap
p
licatio
n
s
,
wh
ich
em
p
lo
y
d
ata
in
d
if
f
er
e
n
t
f
o
r
m
ats.
Ho
wev
er
,
s
ize
o
f
d
ataset
was
n
o
t
ex
p
an
d
ed
,
an
d
it
d
id
n
o
t
ass
ess
it
s
g
en
er
aliza
b
ilit
y
to
d
iv
er
s
e
p
o
p
u
latio
n
s
.
I
n
tr
o
d
u
ce
d
th
e
n
o
v
el
weig
h
t
co
n
v
o
lu
tio
n
n
e
u
r
al
n
etwo
r
k
(
NW
C
NN)
f
o
r
d
iag
n
o
s
in
g
C
KD
[
1
6
]
.
T
h
is
tech
n
iq
u
e
ef
f
icien
tl
y
h
an
d
led
m
is
s
in
g
d
ata
im
p
u
t
atio
n
s
,
ev
en
th
o
u
g
h
it
f
ailed
to
id
en
tify
th
e
s
ev
er
ity
lev
el
o
f
C
KD
wh
ile
im
p
r
o
v
in
g
g
e
n
er
aliza
tio
n
p
e
r
f
o
r
m
an
ce
.
I
s
m
ail
[
1
]
d
ev
el
o
p
ed
a
s
n
a
k
e
-
o
p
tim
ized
f
r
am
ewo
r
k
ter
m
e
d
C
KD
-
SO
f
o
r
ea
r
lier
id
en
tif
icatio
n
o
f
C
KD.
T
h
is
ap
p
r
o
ac
h
p
r
o
v
i
d
ed
ea
r
ly
in
ter
f
e
r
en
ce
s
th
at
d
ec
r
e
ased
h
ig
h
tr
o
u
b
le
o
f
C
KD
-
ass
o
ciate
d
d
is
ea
s
es
as
well
as m
o
r
tality
,
b
u
t sti
ll,
it f
ac
ed
s
to
r
ag
e
an
d
co
m
p
u
tatio
n
al
ch
allen
g
es:
A
f
ew
d
em
er
its
o
f
ex
is
tin
g
C
KD
d
etec
tio
n
m
eth
o
d
s
co
llected
f
o
r
r
ev
iew
ar
e
e
x
p
lain
ed
b
e
lo
w.
−
T
h
e
tech
n
iq
u
e
d
ev
el
o
p
ed
in
[
1
]
was
o
n
ly
s
u
itab
le
f
o
r
p
o
p
u
latio
n
s
tu
d
y
an
d
it
d
id
n
o
t
ass
is
t
clin
ical
ex
p
er
ts
with
ea
ch
p
atien
t.
Mo
r
eo
v
er
,
it r
e
q
u
ir
ed
m
u
ch
m
em
o
r
y
s
to
r
ag
e
a
n
d
len
g
th
y
tr
ai
n
in
g
tim
e.
−
Fu
s
io
n
DL
m
o
d
el
[
1
5
]
h
ad
b
etter
f
lex
ib
ilit
y
an
d
d
u
r
ab
ilit
y
,
ev
e
n
th
o
u
g
h
it
f
ailed
t
o
e
n
h
an
ce
illn
ess
p
r
ed
ictio
n
s
,
tr
ea
tm
en
t
an
d
p
r
e
v
en
tio
n
,
t
h
er
eb
y
im
p
r
o
v
in
g
p
a
tien
t c
ar
e
an
d
r
esu
lts
.
−
C
KD
d
etec
tio
n
in
its
ea
r
ly
p
h
a
s
es
ca
n
p
r
e
v
en
t
s
er
io
u
s
h
ea
lth
p
r
o
b
lem
s
.
Ho
wev
er
,
ac
cu
r
ac
y
o
f
tr
a
d
itio
n
al
ap
p
r
o
ac
h
es
f
o
r
d
etec
tin
g
C
K
D
is
d
ec
r
ea
s
ed
f
r
eq
u
en
tly
o
win
g
to
t
h
eir
d
e
p
en
d
e
n
ce
o
n
a
li
m
ited
n
u
m
b
er
o
f
b
io
lo
g
ical
f
ea
tu
r
es.
T
h
i
s
s
t
u
d
y
i
n
t
r
o
d
u
c
e
s
a
h
y
b
r
i
d
m
o
d
e
l
t
h
a
t
c
o
m
b
i
n
e
s
t
h
e
G
SC
A
Vi
aR
m
o
d
e
l
w
it
h
D
e
n
s
e
N
et
f
o
r
d
e
t
e
c
t
i
n
g
C
K
D
.
B
y
m
e
r
g
i
n
g
s
t
a
t
is
t
ic
a
l
m
o
d
e
l
l
i
n
g
w
it
h
d
e
e
p
l
e
a
r
n
i
n
g
,
t
h
i
s
n
o
v
e
l
a
p
p
r
o
a
c
h
s
i
g
n
i
f
i
c
a
n
t
l
y
b
o
o
s
ts
p
r
e
d
i
c
t
i
o
n
a
c
c
u
r
a
c
y
a
n
d
r
o
b
u
s
t
n
e
s
s
,
e
v
e
n
w
i
t
h
d
i
v
e
r
s
e
p
a
ti
en
t
d
a
t
a
.
T
h
e
r
e
s
e
a
r
c
h
n
o
t
o
n
l
y
a
c
h
i
e
v
e
s
s
u
p
e
r
i
o
r
p
e
r
f
o
r
m
a
n
c
e
c
o
m
p
a
r
e
d
t
o
e
x
i
s
t
i
n
g
m
e
t
h
o
d
s
b
u
t
a
l
s
o
p
r
o
v
i
d
e
s
a
t
h
o
r
o
u
g
h
m
e
t
h
o
d
o
l
o
g
i
c
a
l
f
r
a
m
e
w
o
r
k
t
h
a
t
o
t
h
e
r
r
e
s
e
a
r
c
h
e
r
s
c
a
n
u
s
e
a
n
d
e
x
p
a
n
d
u
p
o
n
,
a
d
v
a
n
c
i
n
g
t
h
e
f
i
e
l
d
o
f
C
K
D
d
e
t
e
ct
i
o
n
.
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
Op
timiz
ed
d
en
s
e
co
n
vo
lu
tio
n
a
l n
etw
o
r
k
w
ith
co
n
d
itio
n
a
l a
u
to
r
eg
r
ess
ive
… (
C
h
eta
n
N
imb
a
A
h
er
)
2011
T
h
e
s
tr
u
ctu
r
e
o
f
t
h
e
p
ap
er
i
s
as
f
o
llo
ws:
s
ec
tio
n
2
d
etai
ls
th
e
p
r
o
p
o
s
ed
GSC
AViaR
-
Den
s
eNe
t
m
eth
o
d
o
l
o
g
y
,
an
d
s
ec
tio
n
3
p
r
esen
ts
th
e
r
esu
lts
alo
n
g
s
id
e
th
e
e
x
p
er
im
e
n
tal
s
etu
p
,
d
ataset
d
escr
ip
tio
n
,
ev
alu
atio
n
m
etr
ics,
an
d
a
c
o
m
p
ar
ativ
e
an
aly
s
is
with
ex
is
tin
g
ap
p
r
o
ac
h
es.
Fin
ally
,
s
ec
tio
n
4
co
n
clu
d
es
with
a
d
is
cu
s
s
io
n
o
f
th
e
f
in
d
in
g
s
an
d
th
eir
im
p
licatio
n
s
,
o
f
f
er
in
g
a
t
h
o
r
o
u
g
h
o
v
er
v
iew
o
f
th
e
r
esear
ch
o
u
tc
o
m
es.
2.
M
E
T
H
O
D
T
o
d
etec
t
C
KD,
s
p
ec
if
ic
b
l
o
o
d
an
d
u
r
in
e
test
s
m
u
s
t
b
e
tak
en
an
d
th
er
ef
o
r
e,
C
KD
d
etec
t
io
n
at
its
ea
r
lier
p
h
ase
is
n
o
t
s
im
p
le
with
o
u
t
a
p
p
r
o
p
r
iate
test
s
.
Her
e,
GSC
AV
iaR
-
Den
s
eNe
t
is
p
r
esen
ted
f
o
r
d
etec
tin
g
C
KD.
I
n
itially
,
ch
r
o
n
ic
k
id
n
e
y
d
ata
is
o
b
tain
ed
f
r
o
m
a
p
ar
t
icu
lar
d
ataset.
T
h
e
d
ata
is
p
r
e
-
p
r
o
ce
s
s
ed
b
y
Min
-
Ma
x
n
o
r
m
aliza
tio
n
.
T
h
en
,
f
e
atu
r
es
ar
e
s
elec
ted
b
ased
o
n
T
o
p
s
o
e
s
im
ilar
ity
.
L
astl
y
,
C
KD
is
d
etec
ted
b
y
em
p
lo
y
in
g
Den
s
eNe
t
an
d
it
is
tr
ain
ed
b
y
GSC
AViaR.
Mo
r
e
o
v
er
,
GSC
AViaR
is
d
ev
is
ed
b
y
in
teg
r
atin
g
GSO
with
C
AVia
R
.
Fig
u
r
e
1
ex
h
i
b
its
a
p
icto
r
ial
illu
s
tr
atio
n
o
f
GSC
AVia
R
-
Den
s
eNe
t f
o
r
C
KD
d
etec
tio
n
.
Fig
u
r
e
1
.
A
p
icto
r
ial
illu
s
tr
atio
n
o
f
GSC
AViaR
-
Den
s
eNe
t f
o
r
C
KD
d
etec
tio
n
2
.
1
.
Acquis
it
io
n o
f
chro
nic kidn
ey
da
t
a
T
h
e
c
h
r
o
n
i
c
k
i
d
n
e
y
d
a
t
a
i
s
a
c
q
u
i
r
e
d
f
r
o
m
t
h
e
d
a
t
a
s
e
t
[
1
6
]
t
o
c
a
r
r
y
o
u
t
C
K
D
d
e
t
e
c
t
i
o
n
a
n
d
i
t
i
s
g
i
v
e
n
b
y
,
=
{
1
,
2
,
.
.
.
,
ℎ
,
.
.
.
,
}
(
1
)
h
er
e,
ℎ
r
ep
r
esen
ts
ℎ
ℎ
in
p
u
t c
h
r
o
n
ic
k
id
n
ey
d
ata
w
h
er
ea
s
to
tal
d
ata
i
n
th
e
d
ataset
R
is
im
p
lied
as
.
2
.
2
.
P
re
-
pro
ce
s
s
ing
utilizing
M
in
-
M
a
x
no
rm
a
liza
t
io
n
Data
p
r
e
-
p
r
o
ce
s
s
in
g
is
ca
r
r
ied
o
u
t to
im
p
u
te
m
is
s
in
g
d
ata
an
d
r
ec
o
g
n
ize
th
e
v
ar
ia
b
les,
wh
ich
m
u
s
t b
e
co
n
s
id
er
ed
in
p
r
ed
ictio
n
s
y
s
tem
s
.
Her
e,
Min
-
Ma
x
n
o
r
m
al
izatio
n
is
em
p
lo
y
ed
f
o
r
d
ata
p
r
e
-
p
r
o
ce
s
s
in
g
b
y
co
n
s
id
er
in
g
ℎ
with
d
im
en
s
io
n
×
as
in
p
u
t.
Min
-
m
a
x
n
o
r
m
aliza
ti
o
n
[
1
7
]
is
th
e
ea
s
iest
m
eth
o
d
,
wh
er
ein
th
is
tech
n
iq
u
e
is
ca
p
ab
le
o
f
f
itti
n
g
d
ata
in
a
p
r
e
-
d
ef
in
e
d
b
o
u
n
d
ar
y
with
th
e
p
r
e
-
d
ef
in
e
d
b
o
u
n
d
a
r
y
.
I
t
ca
n
b
e
f
o
r
m
u
lated
as f
o
llo
ws,
ℎ
=
−
∗
(
−
)
+
(
2
)
wh
er
e,
[
,
]
m
en
tio
n
s
p
r
e
-
d
ef
in
e
d
b
o
u
n
d
a
r
y
,
d
en
o
tes
a
r
a
n
g
e
o
f
ac
tu
al
d
ata
an
d
ℎ
s
p
ec
if
ies
p
r
e
-
p
r
o
ce
s
s
ed
d
ata
with
d
im
e
n
s
io
n
×
.
2
.
3
.
F
S ba
s
ed
o
n
t
o
ps
o
e
s
im
i
la
rit
y
An
in
ten
tio
n
o
f
FS
is
to
d
ete
ct
m
o
s
t
in
f
o
r
m
ativ
e
an
d
s
ig
n
if
ican
t
s
u
b
s
et
o
f
th
e
f
ea
tu
r
es
in
ce
r
tain
d
atab
ases
.
Mo
r
eo
v
er
,
it
d
is
ca
r
d
s
th
e
f
ea
tu
r
es
th
at
ar
e
r
ed
u
n
d
an
t
o
r
n
o
t
ap
p
r
o
p
r
iate.
Her
e,
f
ea
tu
r
es
ar
e
s
elec
ted
b
ased
o
n
T
o
p
s
o
e
s
im
ilar
ity
b
y
tak
in
g
ℎ
with
d
im
en
s
io
n
×
as
in
p
u
t.
T
o
p
s
o
e
s
im
ilar
ity
[
1
8
]
co
m
p
u
tes
th
e
d
is
tan
ce
b
etwe
en
two
p
r
o
b
ab
il
ity
d
is
tr
ib
u
tio
n
s
an
d
it c
a
n
b
e
ca
lcu
lated
b
y
,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
37
,
No
.
3
,
Ma
r
ch
20
25
:
2
0
0
9
-
20
20
2012
=
∑
(
(
2
+
)
+
(
2
+
)
)
=
1
(
3
)
Her
e,
in
d
icate
s
ca
n
d
id
ate
f
ea
tu
r
es
an
d
n
o
tes
tar
g
et.
Af
te
r
co
m
p
u
tin
g
T
o
p
s
o
e
s
im
ilar
ity
f
o
r
in
d
iv
id
u
al
f
ea
tu
r
es,
to
p
f
ea
tu
r
es
with
h
ig
h
er
v
alu
es
ar
e
s
ele
cted
.
T
h
e
o
u
tco
m
e
af
ter
FS
is
s
y
m
b
o
lized
as
ℎ
with
d
im
en
s
io
n
×
,
wh
er
e
>
.
2
.
4
.
CK
D
det
ec
t
i
o
n ut
ilizin
g
DenseNet
E
ar
ly
d
iag
n
o
s
is
an
d
d
etec
tio
n
o
f
C
KD
is
m
o
r
e
cr
itical
f
o
r
s
to
p
p
in
g
th
e
d
ev
elo
p
m
en
t
to
k
id
n
e
y
f
ailu
r
es.
Her
e,
De
n
s
en
et
is
e
m
p
lo
y
ed
to
d
etec
t
C
KD
b
y
o
b
tain
in
g
ℎ
with
d
im
en
s
io
n
×
as
in
p
u
t
an
d
Den
s
en
et
is
tr
ain
ed
b
y
GSC
AViaR.
Fu
r
th
er
m
o
r
e,
GSC
AViaR is
d
esig
n
ed
b
y
m
er
g
in
g
GS
O
with
C
AVia
R
.
2
.
4
.
1
Arc
hite
ct
ure
o
f
DenseNet
Den
s
eNe
t
[
1
9
]
lin
k
s
in
d
iv
id
u
a
l
lay
er
s
to
all
o
th
er
lay
er
s
in
a
f
ee
d
-
f
o
r
war
d
(
FF
)
m
an
n
e
r
.
C
o
n
s
id
er
a
n
im
ag
e
ℎ
,
wh
ich
is
g
iv
en
to
a
c
o
n
v
o
lu
ti
o
n
al
(
c
o
n
v
)
n
etwo
r
k
.
I
t
co
n
tain
s
lay
er
s
,
ea
ch
o
n
e
i
m
p
lem
en
ts
th
e
non
-
lin
ea
r
ity
tr
an
s
f
o
r
m
atio
n
(
.
)
,
wh
er
ein
in
d
ex
es
a
lay
er
.
(
.
)
r
ef
er
s
to
co
m
p
o
s
ite
f
u
n
ctio
n
i
n
g
o
f
o
p
er
atio
n
s
lik
e
b
atc
h
n
o
r
m
al
izatio
n
(
B
N)
,
co
n
v
,
r
ec
tifie
d
lin
ea
r
u
n
it
(
R
eL
U)
o
r
p
o
o
l
in
g
.
An
o
u
tp
u
t
o
f
ℎ
lay
er
is
d
en
o
ted
as
.
(
a)
R
esNet
s
R
esNet
s
in
clu
d
e
s
k
ip
-
co
n
n
ec
tio
n
s
th
at
b
y
p
ass
a
n
o
n
-
li
n
e
ar
ity
tr
an
s
f
o
r
m
atio
n
with
th
e
id
en
tity
o
p
er
atio
n
a
n
d
it is
m
o
d
eled
b
y
,
=
(
−
1
)
+
−
1
(
4
)
A
b
en
ef
it
o
f
R
esNets
is
th
at
a
g
r
a
d
ien
t
ca
n
d
ir
ec
tly
f
l
o
w
b
y
m
ea
n
s
o
f
i
d
en
tity
o
p
er
atio
n
f
r
o
m
t
h
e
later
lay
er
s
to
th
e
ea
r
lies
t
lay
er
s
.
Nev
e
r
th
e
less
,
id
en
tity
o
p
er
atio
n
an
d
th
e
o
u
tco
m
e
o
f
r
ein
teg
r
ated
b
y
a
s
u
m
m
atio
n
th
at
m
ay
d
elay
i
n
f
o
r
m
atio
n
f
lo
w
i
n
a
n
etwo
r
k
.
(
b
)
Den
s
e
co
n
n
ec
tiv
ity
Fo
r
en
h
an
cin
g
in
f
o
r
m
atio
n
f
lo
w
am
o
n
g
s
t
lay
er
s
,
d
iv
e
r
s
e
co
n
n
ec
tiv
ity
p
atter
n
is
d
ev
elo
p
ed
an
d
d
ir
ec
t
ass
o
ciatio
n
s
f
r
o
m
an
y
lay
e
r
to
ev
er
y
s
u
cc
ee
d
in
g
lay
er
.
Acc
o
r
d
in
g
ly
,
ℎ
lay
er
ac
ce
p
ts
f
ea
tu
r
e
m
ap
s
o
f
e
v
er
y
p
r
ev
io
u
s
lay
e
r
,
ℎ
,
.
.
.
,
−
1
as a
n
in
p
u
t.
=
(
[
ℎ
,
1
,
.
.
.
,
−
1
]
)
(
5
)
h
er
e,
[
ℎ
,
1
,
.
.
.
,
−
1
]
in
d
icate
s
co
n
ca
te
n
atio
n
o
f
f
ea
t
u
r
e
m
ap
s
g
en
er
ated
in
t
h
e
lay
er
s
0
,
.
.
.
,
−
1
.
Du
e
to
its
d
en
s
e
co
n
n
ec
tiv
ity
,
th
is
n
et
w
o
r
k
is
s
p
ec
if
ied
as De
n
s
eNe
t.
(
c)
C
o
m
p
o
s
ite
o
p
er
atio
n
(
.
)
is
d
ef
in
ed
as
a
co
m
p
o
s
ite
o
p
er
atio
n
o
f
th
e
th
r
ee
f
o
llo
win
g
f
u
n
ctio
n
s
s
u
ch
as
B
N,
f
o
llo
we
d
b
y
R
eL
U
an
d
3
×
3
co
n
v
.
(
d
)
Po
o
lin
g
lay
er
s
A
n
i
m
p
o
r
t
a
n
t
s
e
g
m
e
n
t
o
f
c
o
n
v
n
e
t
w
o
r
k
s
i
s
t
h
e
d
o
w
n
-
s
a
m
p
l
i
n
g
l
a
y
e
r
s
,
w
h
i
c
h
v
a
r
y
i
n
f
e
a
t
u
r
e
m
a
p
d
i
m
e
n
s
i
o
n
s
.
F
o
r
f
a
c
i
li
t
at
i
n
g
d
o
w
n
-
s
a
m
p
l
i
n
g
i
n
t
h
is
s
t
r
u
c
tu
r
e
,
n
e
t
w
o
r
k
i
s
d
i
v
i
d
e
d
i
n
t
o
n
u
m
e
r
o
u
s
d
e
n
s
e
l
y
a
s
s
o
ci
a
t
e
d
d
e
n
s
e
b
l
o
c
k
s
.
T
h
e
l
ay
e
r
s
a
m
i
d
b
l
o
c
k
s
a
r
e
r
e
f
e
r
r
e
d
as
t
r
a
n
s
i
t
i
o
n
la
y
e
r
s
t
h
at
p
e
r
f
o
r
m
p
o
o
l
i
n
g
a
n
d
c
o
n
v
.
(
e)
Gr
o
wth
r
ate
I
f
an
in
d
iv
id
u
al
o
p
er
atio
n
g
en
er
ates
f
ea
tu
r
e
m
ap
s
,
it p
u
r
s
u
es th
at
ℎ
lay
er
h
as
0
+
×
(
−
1
)
in
p
u
t
f
ea
tu
r
e
m
ap
s
,
wh
e
r
ein
0
r
ep
r
esen
ts
th
e
co
u
n
t
o
f
ch
an
n
e
ls
in
an
i
n
p
u
t
lay
er
.
A
h
y
p
er
p
ar
am
eter
t
is
s
p
ec
if
ied
as th
e
g
r
o
wth
r
ate
o
f
th
e
n
etw
o
r
k
.
(f)
B
o
ttlen
ec
k
lay
er
s
E
v
en
th
o
u
g
h
in
d
iv
id
u
al
lay
e
r
o
n
ly
g
en
e
r
ates
t
o
u
tp
u
t
f
ea
t
u
r
e
m
ap
s
,
it
g
en
er
ally
h
as
s
ev
er
al
in
p
u
ts
.
T
h
e
1
×
1
co
n
v
is
p
r
esen
te
d
as
th
e
b
o
ttlen
ec
k
lay
e
r
,
b
ef
o
r
e
an
i
n
d
iv
id
u
al
3
×
3
co
n
v
r
ed
u
ce
s
th
e
co
u
n
t
o
f
a
n
in
p
u
t f
ea
tu
r
e
m
ap
a
n
d
th
e
r
ef
o
r
e,
to
d
ec
r
ea
s
e
co
m
p
u
tatio
n
al
e
f
f
ec
tiv
en
ess
.
(
g
)
C
o
m
p
r
ess
io
n
Fo
r
en
h
an
cin
g
s
y
s
tem
co
m
p
a
ctn
ess
,
th
e
co
u
n
t
o
f
f
ea
tu
r
e
m
ap
s
at
th
e
tr
an
s
itio
n
lay
er
s
is
r
ed
u
ce
d
.
I
f
th
e
d
en
s
e
b
lo
c
k
co
m
p
r
is
es
f
ea
tu
r
e
m
ap
s
,
p
u
r
s
u
in
g
tr
an
s
itio
n
lay
er
is
p
e
r
m
itted
to
g
en
er
ate
[
]
o
u
tp
u
t
f
ea
tu
r
e
m
a
p
s
,
wh
er
ein
0
<
≤
1
is
m
e
n
tio
n
ed
as
a
co
m
p
r
ess
io
n
f
a
cto
r
.
T
h
e
C
KD
-
d
etec
ted
o
u
t
p
u
t
f
r
o
m
Den
s
eNe
t is im
p
lied
as
ℎ
an
d
D
en
s
eNe
t m
o
d
el
is
s
h
o
wn
in
Fi
g
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
Op
timiz
ed
d
en
s
e
co
n
vo
lu
tio
n
a
l n
etw
o
r
k
w
ith
co
n
d
itio
n
a
l a
u
to
r
eg
r
ess
ive
… (
C
h
eta
n
N
imb
a
A
h
er
)
2013
Fig
u
r
e
2
.
Den
s
eNe
t m
o
d
el
2
.
4
.
2
.
T
ra
ini
ng
o
f
DenseNet
utilizing
G
SCAVia
R
GSO
[
2
0
]
is
th
e
n
atu
r
e
-
e
n
th
u
s
ed
o
p
tim
izatio
n
a
p
p
r
o
ac
h
th
at
ca
n
r
eso
lv
e
v
ar
i
o
u
s
d
iv
er
s
e
o
p
tim
izatio
n
tr
o
u
b
les.
GSO
is
in
s
p
ir
ed
b
y
th
e
s
ea
r
ch
in
g
attr
ib
u
tes
o
f
an
i
m
als
in
u
s
u
al
life
.
T
h
is
a
lg
o
r
ith
m
is
em
p
lo
y
ed
f
o
r
d
is
co
v
er
in
g
ex
ce
llen
t
o
u
tco
m
e
o
v
er
th
e
g
r
o
u
p
o
f
ca
n
d
id
ate
s
o
lu
tio
n
s
to
r
eso
lv
e
an
y
o
p
ti
m
izatio
n
is
s
u
es
b
y
id
en
tify
in
g
m
i
n
im
al
o
r
m
ax
i
m
al
o
b
jectiv
e
f
u
n
ctio
n
s
f
o
r
p
ar
ticu
lar
p
r
o
b
lem
s
.
C
AVia
R
[
2
1
]
s
p
ec
if
ies
an
ev
o
lu
tio
n
o
f
q
u
an
tile
o
v
er
tim
e
u
tili
zin
g
a
r
em
ar
k
ab
le
k
i
n
d
o
f
au
to
r
e
g
r
ess
iv
e
p
r
o
ce
d
u
r
e
.
C
AViaR
m
o
d
el
is
ca
p
ab
le
to
ad
ap
t
n
ewe
r
th
r
ea
t
en
v
ir
o
n
m
en
ts
.
Her
e,
GSO
is
c
o
m
b
in
ed
with
C
AViaR to
d
esi
g
n
GSC
AVia
R
th
at
is
m
o
r
e
s
u
itab
le
f
o
r
t
r
ain
in
g
D
en
s
eNe
t f
o
r
d
etec
tin
g
C
KD.
−
Gr
o
u
p
s
ea
r
ch
p
o
s
itio
n
en
c
o
d
in
g
T
h
e
lear
n
in
g
p
ar
am
eter
o
f
De
n
s
eNe
t
is
co
n
tin
u
o
u
s
ly
tu
n
e
d
in
s
ea
r
ch
s
p
ac
e
f
o
r
ac
q
u
ir
in
g
s
u
p
er
io
r
o
u
tco
m
es,
in
s
u
c
h
a
m
an
n
er
th
at
=
[
1
×
]
.
−
Fit
n
ess
f
u
n
ctio
n
T
h
e
f
itn
ess
f
u
n
ctio
n
is
ev
al
u
a
ted
b
y
id
e
n
tify
in
g
v
a
r
iatio
n
a
m
o
n
g
s
t
tar
g
et
an
d
De
n
s
eNe
t
o
u
tco
m
es
th
at
ca
n
b
e
s
p
ec
if
ied
as,
=
1
∑
[
ℎ
−
ℎ
]
ℎ
=
1
2
(
6
)
h
er
e,
ℎ
in
d
icate
s
tar
g
eted
o
u
t
p
u
t
,
ℎ
m
en
tio
n
s
Den
s
eNe
t o
u
tp
u
t w
h
er
ea
s
s
p
ec
if
ies to
tal
d
ata.
GSC
AV
iaR
p
er
f
o
r
m
s
th
e
f
o
llo
win
g
s
tep
s
to
attain
th
e
b
est o
u
tco
m
e.
Ste
p 1
:
I
nitia
lizing
o
f
s
o
lutio
n
Firstl
y
,
a
g
r
o
u
p
o
f
ca
n
d
i
d
ate
ag
en
ts
th
at
is
ter
m
ed
as
g
r
o
u
p
an
d
in
d
iv
id
u
al
ag
e
n
ts
s
p
ec
if
ied
as
m
em
b
er
s
ar
e
r
a
n
d
o
m
ly
in
itialized
.
I
t c
an
b
e
f
o
r
m
u
lated
b
y
,
=
{
1
,
2
,
.
.
.
,
,
.
.
.
,
}
(
7
)
wh
er
e,
im
p
lies
ℎ
ca
n
d
id
ate
s
o
lu
tio
n
,
d
en
o
tes to
tal
v
ar
iab
les in
a
p
o
p
u
latio
n
.
Ste
p 2
:
Co
m
pu
t
ing
o
bje
ct
iv
e
f
un
ct
io
n
I
t
is
d
eter
m
in
ed
b
y
tak
i
n
g
th
e
d
if
f
er
en
ce
a
m
o
n
g
s
t
Den
s
eNe
t
an
d
tar
g
eted
o
u
tp
u
ts
,
wh
ich
is
ca
lcu
lated
u
tili
zin
g
(
6
)
.
Ste
p 3
:
P
ro
du
cing
s
t
a
g
e
An
ap
ex
is
th
e
e
x
is
tin
g
lo
ca
tio
n
o
f
th
e
p
r
o
d
u
ce
r
.
I
n
GSO,
a
p
r
o
d
u
ce
r
p
er
f
o
r
m
s
at
ℎ
iter
ati
o
n
as
m
en
tio
n
ed
b
elo
w.
A
p
r
o
d
u
ce
r
in
v
esti
g
ates
at
ze
r
o
an
d
th
er
e
af
ter
ex
am
in
e
b
esid
es
u
s
in
g
s
to
ch
asti
c
test
in
g
o
f
th
r
ee
p
o
s
itio
n
s
in
th
e
v
alid
atio
n
p
lac
e.
T
h
e
f
ir
s
t c
r
iter
io
n
at
a
ze
r
o
r
ate
ca
n
b
e
illu
s
tr
ated
b
y
,
=
+
ℜ
1
(
)
(
8
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
37
,
No
.
3
,
Ma
r
ch
20
25
:
2
0
0
9
-
20
20
2014
A
p
o
in
t in
th
e
r
ig
h
t
-
h
an
d
s
id
e
h
y
p
er
c
u
b
e
ca
n
b
e
g
i
v
en
b
y
,
ℜ
=
+
ℜ
1
(
+
ℜ
2
/
2
)
(
9
)
A
p
o
in
t in
th
e
lef
t
-
h
an
d
s
id
e
h
y
p
er
cu
b
e
is
m
o
d
elled
as,
=
+
ℜ
1
(
−
ℜ
2
/
2
)
(
1
0
)
Her
e,
ℜ
1
∈
̸
1
th
at
s
p
ec
if
ies
to
n
o
r
m
ally
d
is
tr
ib
u
ted
s
to
ch
asti
c
v
alu
e
h
av
i
n
g
m
ea
n
=0
a
n
d
t
h
e
s
tan
d
ar
d
d
ev
iatio
n
(
SD)
as 1
.
ℜ
2
∈
̸
−
1
im
p
lies
to
s
to
ch
asti
c
v
alu
es
th
at
ar
e
d
is
tr
ib
u
ted
u
n
if
o
r
m
l
y
in
a
r
an
g
e
0
an
d
1
.
I
f
a
b
etter
r
e
g
io
n
h
as
s
u
p
e
r
io
r
f
itn
ess
v
alu
e
th
an
its
ex
is
tin
g
lo
ca
tio
n
,
th
en
it
m
o
v
es
to
th
is
r
eg
io
n
.
Or
else,
it st
ay
s
in
its
p
r
esen
t lo
ca
tio
n
an
d
c
h
an
g
es its
h
ea
d
t
o
a
n
ewe
r
an
g
le
as
:
+
1
=
+
ℜ
2
(
1
1
)
Her
e,
∈
̸
1
en
o
tes m
ax
im
al
a
d
ju
s
tin
g
lo
ca
tio
n
.
I
f
a
p
r
o
d
u
ce
r
is
n
o
t
ca
p
ab
le
o
f
ac
q
u
ir
in
g
s
u
p
er
io
r
s
ea
r
ch
s
p
ac
e
af
ter
o
u
t
o
f
iter
atio
n
s
,
it
e
m
p
lo
y
s
th
e
lead
er
b
ac
k
to
0
∘
.
+
=
(
1
2
)
Her
e,
∈
̸
1
in
d
icate
s
co
n
s
tan
t v
alu
e.
Ste
p 4
:
Scro
un
g
ing
s
t
a
g
e
At
an
in
d
iv
id
u
al
iter
atio
n
,
v
a
r
io
u
s
g
r
o
u
p
in
g
ag
en
ts
ar
e
s
elec
ted
as
s
cr
o
u
n
g
er
s
.
At
ℎ
r
ed
u
n
d
an
cy
,
s
p
ac
e
co
p
y
in
g
attr
ib
u
te
o
f
ℎ
s
cr
o
u
n
g
e
r
is
im
p
lied
as st
o
ch
asti
c
walk
in
g
n
ea
r
e
r
a
p
r
o
d
u
ce
r
.
+
1
=
+
ℜ
3
∘
(
−
)
(
1
3
)
+
1
=
(
1
−
ℜ
3
)
+
ℜ
3
∘
(
1
4
)
Fro
m
C
AViaR,
th
e
ex
p
r
ess
io
n
ca
n
b
e
g
iv
en
as,
=
0
+
∑
=
1
−
+
∑
=
1
(
−
)
(
1
5
)
C
o
n
s
id
er
,
=
=
2
,
th
er
ef
o
r
e
a
b
o
v
e
e
q
u
atio
n
b
ec
o
m
es,
=
0
+
1
−
1
+
0
−
2
+
1
−
1
+
2
(
−
2
)
(
1
6
)
Su
b
s
titu
te
(
1
6
)
in
(
1
4
)
an
d
th
u
s
,
th
e
eq
u
atio
n
b
ec
o
m
es,
+
1
=
(
0
+
1
−
1
+
0
−
2
+
1
−
1
+
2
(
−
2
)
)
(
1
−
ℜ
3
)
+
ℜ
3
∘
(
1
7
)
T
h
e
ab
o
v
e
e
x
p
r
ess
io
n
is
an
u
p
d
ated
eq
u
ati
o
n
o
f
GSC
AViaR,
wh
er
ein
ℜ
3
n
o
tes
u
n
if
o
r
m
s
to
ch
asti
c
s
eq
u
en
ce
v
alu
es
r
an
g
i
n
g
b
etwe
en
0
a
n
d
1
,
r
ef
er
s
to
a
p
r
o
d
u
ce
r
at
ℎ
iter
atio
n
wh
er
ea
s
∘
in
d
icate
s
p
r
o
d
u
ct
th
at
co
m
p
u
tes a
p
r
o
d
u
ct
o
f
tw
o
v
ec
to
r
s
.
Ste
p 5
:
Dis
persio
n
s
t
a
g
e
I
n
GSO,
it
m
ak
es
clas
s
if
icat
io
n
if
ℎ
T
h
e
o
f
f
e
r
s
ag
en
t
is
d
is
p
er
s
ed
.
At
ℎ
s
ea
r
ch
,
it
d
ev
el
o
p
s
s
ch
o
last
ic
f
r
o
n
t lo
ca
tio
n
a
n
d
t
h
en
,
it o
b
tain
s
r
an
d
o
m
d
is
tan
c
e
th
at
ca
n
b
e
m
en
tio
n
e
d
b
y
,
=
.
ℜ
1
(
1
8
)
T
h
en
,
n
ewe
r
lo
ca
tio
n
s
ca
n
b
e
f
o
r
m
u
lated
as,
+
1
=
+
(
+
1
)
(
1
9
)
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
Op
timiz
ed
d
en
s
e
co
n
vo
lu
tio
n
a
l n
etw
o
r
k
w
ith
co
n
d
itio
n
a
l a
u
to
r
eg
r
ess
ive
… (
C
h
eta
n
N
imb
a
A
h
er
)
2015
Ste
p 6
:
T
er
m
ina
t
io
n
GSC
AV
iaR
i
s
ter
m
in
ated
af
ter
o
b
tain
in
g
th
e
b
est s
o
lu
tio
n
b
y
co
n
tin
u
o
u
s
ex
ec
u
tio
n
o
f
th
e
ab
o
v
e
s
tep
s
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
o
u
tco
m
es
ac
h
iev
ed
b
y
GSC
AVia
R
-
Den
s
eNe
t
th
at
i
s
d
esig
n
ed
f
o
r
C
KD
d
etec
tio
n
ar
e
elu
cid
ated
in
th
is
p
ar
t.
GSC
AVia
R
-
De
n
s
eNe
t
o
u
tp
er
f
o
r
m
ed
ex
is
tin
g
m
eth
o
d
s
,
ac
h
iev
in
g
9
1
.
5
%
ac
cu
r
ac
y
,
9
2
.
8
%
s
en
s
itiv
ity
,
an
d
9
0
.
7
%
s
p
ec
if
icity
with
9
0
%
tr
ain
in
g
d
ata.
I
n
co
m
p
ar
is
o
n
,
th
e
Dee
p
en
s
e
m
b
le
m
o
d
el,
Fu
s
io
n
DL
m
o
d
el,
NW
C
NN,
an
d
C
KD
-
SO h
ad
lo
wer
m
etr
ics ac
r
o
s
s
th
e
b
o
ar
d
.
3
.
1
.
E
x
perim
ent
s
et
up
T
h
e
GSC
AV
iaR
-
Den
s
eNe
t
m
o
d
el
f
o
r
C
KD
d
etec
tio
n
was
im
p
lem
en
ted
u
s
in
g
th
e
PYTH
ON
to
o
l.
T
h
e
im
p
lem
en
tatio
n
in
v
o
lv
ed
th
e
u
s
e
o
f
v
ar
io
u
s
lib
r
ar
ies,
in
clu
d
in
g
T
en
s
o
r
Flo
w,
Ker
as,
an
d
s
cik
it
-
lear
n
,
t
o
b
u
ild
a
n
d
t
r
ain
th
e
Den
s
eNe
t
ar
ch
itectu
r
e
in
teg
r
ated
wi
th
th
e
C
AViaR
m
o
d
el.
T
h
e
ex
p
er
im
e
n
ts
wer
e
co
n
d
u
cte
d
o
n
a
h
ig
h
-
p
er
f
o
r
m
a
n
ce
co
m
p
u
tin
g
en
v
ir
o
n
m
e
n
t
t
o
en
s
u
r
e
th
e
e
f
f
icien
t
tr
ain
i
n
g
o
f
th
e
m
o
d
el
o
n
th
e
C
KD
d
ataset.
3
.
2
.
Da
t
a
s
et
des
cr
iptio
n
T
h
e
s
tu
d
y
u
tili
ze
s
th
e
C
KD
d
ataset,
wh
ich
c
o
m
p
r
is
es
4
0
0
p
atien
t
r
ec
o
r
d
s
c
o
llected
.
T
h
e
d
ataset
in
clu
d
es
k
ey
f
ea
tu
r
es
s
u
ch
as
ag
e,
g
en
d
er
,
b
lo
o
d
p
r
ess
u
r
e,
s
er
u
m
cr
ea
tin
in
e,
an
d
g
l
o
m
e
r
u
lar
f
iltra
tio
n
r
ate
(
GFR
)
[
1
6
]
.
3
.
3
.
E
v
a
lua
t
i
o
n m
et
rics
Acc
u
r
ac
y
,
s
p
ec
if
icity
,
an
d
s
en
s
itiv
ity
ar
e
co
n
s
id
er
ed
f
o
r
ev
alu
atin
g
th
e
GSC
AVia
R
-
Den
s
eNe
t
m
o
d
el.
Acc
u
r
ac
y
m
ea
s
u
r
es
o
v
er
all
co
r
r
ec
tn
ess
,
s
p
ec
if
icity
ass
es
s
es
th
e
id
en
tific
atio
n
o
f
n
eg
ativ
e
ca
s
es,
an
d
s
en
s
itiv
ity
ev
alu
ates
th
e
d
etec
tio
n
o
f
p
o
s
itiv
e
ca
s
es.
T
o
g
eth
e
r
,
th
ese
m
etr
ics
p
r
o
v
i
d
e
a
co
m
p
r
eh
en
s
iv
e
v
iew
o
f
th
e
m
o
d
el'
s
p
er
f
o
r
m
a
n
ce
[
2
2
]
,
[
2
3
]
.
3
.
3
.
1
.
Acc
ura
cy
Acc
u
r
ac
y
im
p
lies
a
p
er
ce
n
ta
g
e
o
f
ex
ac
tly
d
etec
ted
ca
s
es
as
p
o
s
itiv
e
a
n
d
n
eg
ativ
e
C
KD
b
y
th
e
s
y
s
tem
o
u
t o
f
o
v
er
all
ca
s
es e
s
tim
ated
th
at
ca
n
b
e
co
m
p
u
ted
b
y
,
=
+
+
+
+
(
2
0
)
Her
e,
in
d
icate
s
tr
u
e
p
o
s
itiv
e
(
T
P),
r
ep
r
esen
ts
tr
u
e
n
e
g
ativ
e
(
T
N)
,
s
p
ec
if
ies
f
alse
p
o
s
i
tiv
e
(
FP
)
an
d
n
o
tes f
alse n
eg
ativ
e
(
FN)
.
3
.
3
.
2
Sp
ec
if
icit
y
Sp
ec
if
icity
co
m
p
u
tes
a
p
r
o
p
o
r
tio
n
o
f
T
N
in
s
tan
ce
s
th
at
ar
e
ac
cu
r
ately
d
etec
ted
b
y
a
m
o
d
el
an
d
it
is
ev
alu
ated
as,
=
+
(
2
1
)
3
.
3
.
3
Sens
it
iv
it
y
Sen
s
itiv
ity
ev
alu
ates
a
p
r
o
p
o
r
tio
n
o
f
T
P
in
s
tan
ce
s
th
at
ar
e
p
er
f
ec
tly
d
etec
ted
b
y
a
s
y
s
tem
,
wh
ich
is
g
iv
en
b
y
,
=
+
(
2
2
)
3
.
4
.
Co
m
pa
ra
t
iv
e
t
ec
hn
iqu
es
T
h
e
Dee
p
E
n
s
em
b
le
m
o
d
el
[
8
]
,
Fu
s
io
n
DL
m
o
d
el
[
1
5
]
,
No
v
el
W
eig
h
t
C
o
n
v
o
lu
tio
n
al
Neu
r
al
Netwo
r
k
(
NW
C
NN)
[
9
]
,
a
n
d
Sn
ak
e
-
e
f
f
icien
t
Featu
r
e
Selectio
n
-
b
ased
Fra
m
ewo
r
k
(
C
KD
-
SO)
[
1
]
ar
e
co
n
s
id
er
ed
c
o
m
p
ar
ativ
e
m
eth
o
d
s
to
d
em
o
n
s
tr
ate
th
e
ef
f
ec
tiv
e
n
ess
o
f
GSC
AVia
R
-
Den
s
eNe
t
[
2
4
]
,
[
2
5
]
.
3
.
5
.
Co
m
pa
ra
t
iv
e
a
s
s
ess
m
ent
T
h
e
esti
m
atio
n
o
f
GSC
AVia
R
-
Den
s
eNe
t
i
s
p
er
f
o
r
m
ed
b
y
ass
es
s
in
g
k
ey
m
etr
ics
wh
ile
v
ar
y
in
g
th
e
tr
ain
in
g
d
ata
a
n
d
u
tili
zin
g
K
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
.
T
h
is
ap
p
r
o
ac
h
e
n
s
u
r
es
th
at
th
e
m
o
d
e
l's
p
er
f
o
r
m
an
ce
is
r
o
b
u
s
t a
n
d
co
n
s
is
ten
t a
cr
o
s
s
d
i
f
f
er
en
t su
b
s
ets o
f
th
e
d
ata,
h
el
p
in
g
to
m
in
im
ize
b
ias an
d
v
ar
i
an
ce
[
2
6
]
,
[
2
7
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
37
,
No
.
3
,
Ma
r
ch
20
25
:
2
0
0
9
-
20
20
2016
3
.
5
.
1
.
Ana
ly
s
is
o
f
t
ra
ini
ng
da
t
a
Fig
u
r
e
3
r
ep
r
esen
ts
th
e
a
n
al
y
s
is
o
f
GSC
AViaR
-
Den
s
eNe
t
co
n
ce
r
n
in
g
ev
alu
atio
n
m
ea
s
u
r
es
b
y
ch
an
g
in
g
tr
ain
in
g
d
ata.
I
n
th
i
s
s
ec
tio
n
,
v
alu
es
attain
ed
b
y
GSC
AV
iaR
-
Den
s
eNe
t
an
d
co
n
v
en
tio
n
al
m
eth
o
d
s
wh
ile
tr
ain
in
g
d
ata=
9
0
%
ar
e
ex
p
lain
ed
.
Fig
u
r
e
3
(
a)
in
ter
p
r
ets
th
e
ass
es
s
m
en
t
o
f
GSC
A
ViaR
-
Den
s
eNe
t
wi
th
r
eg
ar
d
t
o
ac
cu
r
ac
y
.
GSC
AVia
R
-
Den
s
eNe
t
attain
ed
an
ac
cu
r
ac
y
o
f
0
.
9
1
5
wh
er
ea
s
th
e
Dee
p
en
s
em
b
le
m
o
d
el,
Fu
s
io
n
DL
m
o
d
el,
NW
C
NN
an
d
C
KD
-
SO
ac
q
u
ir
ed
0
.
7
4
9
,
0
.
7
8
4
,
0
.
8
1
9
a
n
d
0
.
8
5
4
im
p
lies
en
h
an
ce
m
e
n
t
in
p
er
f
o
r
m
an
ce
ab
o
u
t
1
8
.
1
5
7
%,
1
4
.
3
0
8
%,
1
0
.
4
8
7
%
an
d
6
.
6
5
5
%.
E
v
alu
atio
n
o
f
GSC
AVia
R
-
Den
s
eNe
t
in
ter
m
s
o
f
s
en
s
itiv
ity
is
s
h
o
wn
in
Fig
u
r
e
3
(
b
)
.
T
h
e
s
en
s
itiv
ity
o
b
tain
ed
b
y
GSC
AViaR
-
Den
s
eNe
t
is
0
.
9
2
8
wh
ile
th
e
v
alu
e
ac
h
iev
e
d
b
y
th
e
Dee
p
e
n
s
em
b
le
m
o
d
el
is
0
.
7
3
9
,
th
e
F
u
s
io
n
DL
m
o
d
el
is
0
.
7
8
4
,
N
W
C
N
N
is
0
.
8
0
5
an
d
C
KD
-
SO
is
0
.
8
5
4
.
I
t
ex
p
licates
im
p
r
o
v
em
en
t
in
p
er
f
o
r
m
an
ce
ab
o
u
t
2
0
.
4
5
2
%,
1
5
.
5
8
5
%,
1
3
.
3
1
6
%
an
d
8
.
0
4
5
%.
Fig
u
r
e
3
(
c)
m
e
n
tio
n
s
th
e
esti
m
atio
n
o
f
GSC
AVia
R
-
Den
s
eNe
t
r
eg
ar
d
in
g
s
p
ec
if
ici
ty
.
Dee
p
en
s
em
b
le
m
o
d
el,
Fu
s
io
n
DL
m
o
d
el,
NW
C
N
N
an
d
C
KD
-
SO
o
b
tain
ed
s
p
ec
if
icity
o
f
0
.
7
3
4
,
0
.
7
8
5
,
0
.
8
0
6
an
d
0
.
8
5
4
wh
er
ea
s
GSC
AVia
R
-
Den
s
eN
et
ac
q
u
ir
ed
0
.
9
0
7
.
T
h
is
d
escr
ib
es
en
h
an
cin
g
in
p
e
r
f
o
r
m
an
ce
ab
o
u
t
1
9
.
1
3
2
%,
1
3
.
4
5
3
%,
1
1
.
1
9
4
% a
n
d
5
.
9
0
0
%.
(
a)
(
b
)
(
c)
Fig
u
r
e
3
.
C
o
m
p
a
r
ativ
e
an
aly
s
i
s
b
ased
o
n
tr
ain
in
g
d
ata
:
(
a
)
ac
cu
r
ac
y
,
(
b
)
s
en
s
itiv
ity
,
an
d
(
c)
s
p
ec
if
icity
3
.
5
.
2
.
Ana
ly
s
is
re
g
a
rding
K
-
f
o
ld
Ass
es
s
m
en
t
o
f
GSC
AVia
R
-
Den
s
eNe
t
r
eg
ar
d
in
g
e
v
alu
at
io
n
m
ea
s
u
r
es
b
y
v
ar
y
i
n
g
K
-
f
o
ld
is
d
em
o
n
s
tr
ated
in
Fig
u
r
e
4
.
T
h
e
v
alu
es o
b
tain
ed
b
y
co
n
s
id
er
e
d
tech
n
i
q
u
es
wh
ile
K
-
f
o
ld
=
9
a
r
e
illu
s
tr
ated
in
th
is
p
ar
t.
E
v
al
u
atio
n
o
f
GSC
AVia
R
-
Den
s
eNe
t
in
r
esp
ec
tiv
e
to
a
c
cu
r
ac
y
is
s
p
ec
if
ied
in
Fig
u
r
e
4
(
a)
.
T
h
e
ac
c
u
r
ac
y
ac
q
u
ir
ed
b
y
GSC
AViaR
-
Den
s
eNe
t is 0
.
9
2
8
wh
er
ea
s
th
e
v
alu
e
attain
ed
b
y
th
e
Dee
p
E
n
s
em
b
le
m
o
d
el
is
0
.
7
3
9
,
th
e
Fu
s
io
n
DL
m
o
d
el
is
0
.
7
8
6
,
NW
C
NN
i
s
0
.
8
0
6
an
d
C
KD
-
SO
i
s
0
.
8
5
3
.
I
t
elu
cid
a
tes
en
h
an
ce
m
en
t
in
p
er
f
o
r
m
an
ce
ab
o
u
t
1
7
.
8
3
0
%
,
1
1
.
3
1
3
%,
9
.
0
6
8
%
an
d
3
.
7
5
5
%.
Fig
u
r
e
4
(
b
)
p
r
esen
ts
an
esti
m
atio
n
o
f
GSC
AV
iaR
-
Den
s
eNe
t
with
r
elatio
n
to
s
en
s
itiv
ity
.
Dee
p
e
n
s
em
b
le
m
o
d
el,
Fu
s
io
n
DL
m
o
d
el,
NW
C
NN
an
d
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
Op
timiz
ed
d
en
s
e
co
n
vo
lu
tio
n
a
l n
etw
o
r
k
w
ith
co
n
d
itio
n
a
l a
u
to
r
eg
r
ess
ive
… (
C
h
eta
n
N
imb
a
A
h
er
)
2017
C
KD
-
SO
at
tain
ed
s
en
s
itiv
ity
o
f
0
.
7
3
2
,
0
.
7
8
5
,
0
.
8
0
5
a
n
d
0
.
8
5
4
w
h
er
ea
s
GSC
AVia
R
-
Den
s
eNe
t
ac
h
iev
ed
0
.
8
9
8
.
T
h
is
in
d
icate
s
im
p
r
o
v
e
m
en
t
in
p
er
f
o
r
m
an
ce
ab
o
u
t
1
8
.
4
7
5
%,
1
2
.
6
4
0
%,
1
0
.
4
0
7
%
an
d
4
.
8
9
6
%.
An
aly
s
is
o
f
GSC
AViaR
-
Den
s
eNe
t
co
n
s
id
er
in
g
s
p
ec
if
icity
is
d
elin
ea
te
d
in
Fig
u
r
e
4
(
c)
.
GSC
AViaR
-
Den
s
eNe
t
o
b
tain
ed
a
s
p
ec
if
icity
o
f
0
.
9
1
0
wh
er
ea
s
th
e
Dee
p
e
n
s
em
b
le
m
o
d
el,
Fu
s
io
n
DL
m
o
d
el,
NW
C
NN
an
d
C
KD
-
SO
attain
ed
0
.
7
3
7
,
0
.
7
8
6
,
0
.
8
0
4
an
d
0
.
8
5
3
s
ig
n
if
ies
p
er
f
o
r
m
an
ce
en
h
a
n
c
em
en
t
o
f
ab
o
u
t
1
8
.
9
5
3
%,
1
3
.
5
8
4
%,
1
1
.
5
8
1
%
an
d
6
.
2
2
2
%.
(
a)
(
b
)
(
c)
Fig
u
r
e
4
.
C
o
m
p
a
r
ativ
e
an
aly
s
i
s
b
ased
o
n
K
-
f
o
ld
:
(
a)
ac
c
u
r
ac
y
,
(
b
)
s
en
s
itiv
ity
,
an
d
(
c
)
s
p
ec
i
f
icity
3
.
6
.
Co
m
pa
ra
t
iv
e
dis
cus
s
io
n
GSC
AV
iaR
-
Den
s
eNe
t
ac
q
u
ir
ed
s
u
p
er
io
r
r
esu
lts
wh
ile
co
m
p
ar
in
g
with
ex
is
tin
g
s
ch
em
es
lik
e
th
e
Dee
p
en
s
em
b
le
m
o
d
el,
Fu
s
io
n
DL
m
o
d
el,
NW
C
N
N
an
d
C
KD
-
SO.
T
h
e
d
is
cu
s
s
io
n
ta
b
le
o
f
ass
es
s
m
en
ts
p
er
f
o
r
m
ed
is
illu
s
tr
ated
in
T
ab
le
1
.
W
h
en
tr
ain
in
g
d
ata=
9
0
%,
GSC
AVia
R
-
Den
s
eNe
t
ac
h
iev
ed
9
1
.
5
%
o
f
ac
cu
r
ac
y
w
h
er
ea
s
th
e
Dee
p
e
n
s
em
b
le
m
o
d
el,
Fu
s
io
n
DL
m
o
d
el,
NW
C
NN
an
d
C
KD
-
SO
o
b
tain
ed
7
4
.
9
%,
7
8
.
4
%,
8
1
.
9
%
an
d
8
5
.
4
%.
T
h
is
d
escr
ib
es
th
at
GSC
AViaR
-
Den
s
eNe
t
is
ca
p
ab
le
o
f
d
etec
tin
g
p
o
s
s
ib
le
s
y
m
p
to
m
s
o
f
C
KD.
Sen
s
itiv
ity
ac
q
u
ir
e
d
b
y
th
e
Dee
p
en
s
e
m
b
le
m
o
d
el
is
7
3
.
9
%,
Fu
s
io
n
DL
m
o
d
el
is
7
8
.
4
%,
NW
C
NN
is
8
0
.
5
%
an
d
C
KD
-
SO
is
8
5
.
4
%
wh
ile
s
en
s
itiv
ity
attain
ed
b
y
GSC
AViaR
-
Den
s
eNe
t
is
9
2
.
8
%
wh
ile
tr
ain
in
g
d
ata
is
9
0
%.
I
t
elu
cid
ates
th
at
G
SC
A
ViaR
-
Den
s
eNe
t
d
etec
ted
ea
ch
p
er
s
o
n
at
r
is
k
f
o
r
C
KD.
Dee
p
en
s
em
b
le
m
o
d
el,
Fu
s
io
n
DL
m
o
d
el,
NW
C
NN
an
d
C
KD
-
SO
ac
h
iev
ed
a
s
p
ec
if
icity
o
f
7
3
.
4
%,
7
8
.
5
%,
8
0
.
6
%
an
d
8
5
.
4
%
wh
er
ea
s
GSC
AVia
R
-
Den
s
eNe
t
ac
q
u
ir
ed
a
s
p
ec
if
icity
o
f
ab
o
u
t
9
0
.
7
%.
T
h
is
in
d
icate
s
th
at
GSC
AV
iaR
-
Den
s
eNe
t
p
er
f
ec
tly
id
en
tifie
d
in
d
i
v
id
u
als
wh
o
h
av
e
C
KD.
Fro
m
th
e
ass
ess
m
en
ts
co
n
d
u
cte
d
,
it
ca
n
b
e
co
n
clu
d
e
d
th
at
GSC
AViaR
-
Den
s
eNe
t
i
s
th
e
b
etter
ap
p
r
o
ac
h
f
o
r
C
KD
d
etec
tio
n
as
it
ac
h
iev
ed
9
1
.
5
%
ac
cu
r
ac
y
,
9
2
.
8
% sen
s
itiv
ity
an
d
9
0
.
7
% sp
ec
if
icity
f
o
r
9
0
% o
f
tr
ain
in
g
d
ata.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
37
,
No
.
3
,
Ma
r
ch
20
25
:
2
0
0
9
-
20
20
2018
T
ab
le
1
.
C
o
m
p
a
r
ativ
e
d
is
cu
s
s
io
n
o
f
GSC
AVia
R
-
Den
s
eNe
t
S
e
t
u
p
s
M
e
t
r
i
c
s
/
M
e
t
h
o
d
s
D
e
e
p
e
n
s
e
mb
l
e
mo
d
e
l
F
u
si
o
n
D
L
mo
d
e
l
N
W
C
N
N
C
K
D
-
SO
P
r
o
p
o
se
d
G
S
C
A
V
i
a
R
-
D
e
n
seN
e
t
Tr
a
i
n
i
n
g
d
a
t
a
=
9
0
%
A
c
c
u
r
a
c
y
7
4
.
9
%
7
8
.
4
%
8
1
.
9
%
8
5
.
4
%
9
1
.
5
%
S
e
n
s
i
t
i
v
i
t
y
7
3
.
9
%
7
8
.
4
%
8
0
.
5
%
8
5
.
4
%
9
2
.
8
%
S
p
e
c
i
f
i
c
i
t
y
7
3
.
4
%
7
8
.
5
%
8
0
.
6
%
8
5
.
4
%
9
0
.
7
%
K
-
f
o
l
d
=
9
A
c
c
u
r
a
c
y
7
2
.
9
%
7
8
.
6
%
8
0
.
6
%
8
5
.
3
%
8
8
.
7
%
S
e
n
s
i
t
i
v
i
t
y
7
3
.
2
%
7
8
.
5
%
8
0
.
5
%
8
5
.
4
%
8
9
.
8
%
S
p
e
c
i
f
i
c
i
t
y
7
3
.
7
%
7
8
.
6
%
8
0
.
4
%
8
5
.
3
%
9
1
%
4.
CO
NCLU
SI
O
N
C
KD
s
p
ec
if
ies
an
im
p
air
m
en
t o
f
th
e
k
id
n
ey
s
th
at
g
ets
wo
r
s
e
o
v
er
tim
e.
I
t
is
a
d
eter
io
r
atin
g
is
s
u
e
th
at
ca
u
s
es
wo
r
ld
wid
e
tr
o
u
b
le
as
t
h
e
ex
is
tin
g
r
em
e
d
ial
ch
o
ices
ar
e
n
o
t
e
f
f
ec
tiv
e.
E
f
f
ec
tu
al
tr
ea
tm
en
t
an
d
ea
r
lier
d
iag
n
o
s
in
g
a
r
e
s
ig
n
if
ican
t
to
av
o
id
C
KD
p
r
o
g
r
ess
io
n
.
Fu
r
th
er
m
o
r
e
,
ea
r
lier
d
etec
tio
n
o
f
C
KD
i
s
v
ital
to
s
av
e
n
u
m
er
o
u
s
p
eo
p
le.
As
an
o
u
tc
o
m
e,
v
a
r
io
u
s
r
esear
ch
e
r
s
ar
e
p
r
esen
tly
co
n
ce
n
tr
ated
o
n
d
e
v
elo
p
in
g
p
r
o
f
icien
t
tech
n
iq
u
es
to
d
etec
t
C
KD.
Ho
wev
er
,
m
o
s
t
o
f
th
e
ap
p
r
o
ac
h
es
ar
e
tim
e
-
c
o
n
s
u
m
in
g
to
id
en
tify
C
KD.
I
n
th
is
r
esear
ch
,
GSC
AVia
R
-
Den
s
eN
et
is
n
ewly
d
esig
n
ed
f
o
r
C
KD
d
etec
tio
n
.
At
f
ir
s
t,
ch
r
o
n
ic
k
i
d
n
ey
d
ata
is
tak
en
f
r
o
m
a
s
p
ec
if
ic
d
ataset.
T
h
en
,
p
r
e
-
p
r
o
ce
s
s
in
g
o
f
co
n
s
id
er
ed
d
ata
is
ac
co
m
p
lis
h
ed
b
y
Min
-
Ma
x
n
o
r
m
aliza
tio
n
.
Af
ter
th
at,
FS
is
ca
r
r
ied
o
u
t
f
o
r
s
elec
tin
g
ap
p
r
o
p
r
iate
f
ea
t
u
r
e
s
f
o
r
d
etec
tio
n
p
r
o
ce
s
s
.
T
h
e
f
e
atu
r
es
ar
e
s
elec
ted
b
ased
o
n
T
o
p
s
o
e
s
im
ilar
ity
.
A
t
last
,
C
KD
i
s
d
etec
ted
u
tili
zin
g
Den
s
eNe
t
an
d
th
e
tr
ai
n
in
g
o
f
Den
s
eNe
t
is
d
o
n
e
b
y
GSC
AVia
R
.
Mo
r
eo
v
er
,
G
SC
AVia
R
i
s
p
r
esen
ted
b
y
jo
in
in
g
GSO
with
C
AViaR.
I
n
ad
d
itio
n
,
GSC
AVia
R
-
Den
s
eNe
t
attain
ed
m
ax
im
u
m
ac
cu
r
ac
y
,
s
en
s
itiv
ity
an
d
s
p
ec
if
icity
o
f
ab
o
u
t
9
1
.
5
%,
9
2
.
8
%
an
d
9
0
.
7
%
wh
ile
co
n
s
id
er
ed
t
r
ain
in
g
d
ata
is
9
0
%.
GSC
AViaR
-
Den
s
eNe
t
d
em
o
n
s
tr
ated
s
u
p
er
io
r
p
er
f
o
r
m
an
ce
in
C
KD
d
etec
tio
n
,
h
ig
h
lig
h
tin
g
its
p
o
t
en
tial
f
o
r
ea
r
ly
d
iag
n
o
s
is
.
Fu
tu
r
e
wo
r
k
m
a
y
ex
p
lo
r
e
o
p
tim
i
zin
g
th
e
m
o
d
el
f
o
r
b
r
o
ad
e
r
d
atasets
to
en
h
an
ce
g
e
n
er
aliza
b
ilit
y
.
RE
F
E
R
E
NC
E
S
[
1
]
W
.
N
.
I
smai
l
,
“
S
n
a
k
e
-
e
f
f
i
c
i
e
n
t
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
-
b
a
s
e
d
f
r
a
mew
o
r
k
f
o
r
p
r
e
c
i
se
e
a
r
l
y
d
e
t
e
c
t
i
o
n
o
f
c
h
r
o
n
i
c
k
i
d
n
e
y
d
i
s
e
a
se
,
”
D
i
a
g
n
o
s
t
i
c
s
,
v
o
l
.
1
3
,
n
o
.
1
5
,
p
.
2
5
0
1
,
Ju
l
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
d
i
a
g
n
o
s
t
i
c
s1
3
1
5
2
5
0
1
.
[
2
]
M
.
M
a
j
i
d
e
t
a
l
.
,
“
U
s
i
n
g
e
n
sem
b
l
e
l
e
a
r
n
i
n
g
a
n
d
a
d
v
a
n
c
e
d
d
a
t
a
m
i
n
i
n
g
t
e
c
h
n
i
q
u
e
s
t
o
i
m
p
r
o
v
e
t
h
e
d
i
a
g
n
o
si
s
o
f
c
h
r
o
n
i
c
k
i
d
n
e
y
d
i
s
e
a
se
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
A
d
v
a
n
c
e
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
4
,
n
o
.
1
0
,
p
p
.
4
7
0
–
4
8
0
,
2
0
2
3
,
d
o
i
:
1
0
.
1
4
5
6
9
/
I
JA
C
S
A
.
2
0
2
3
.
0
1
4
1
0
5
0
.
[
3
]
M
.
C
.
Tsa
i
e
t
a
l
.
,
“
R
i
s
k
p
r
e
d
i
c
t
i
o
n
m
o
d
e
l
f
o
r
c
h
r
o
n
i
c
k
i
d
n
e
y
d
i
s
e
a
s
e
i
n
t
h
a
i
l
a
n
d
u
si
n
g
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
S
H
A
P
,
”
D
i
a
g
n
o
s
t
i
c
s
,
v
o
l
.
1
3
,
n
o
.
2
3
,
p
.
3
5
4
8
,
N
o
v
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
d
i
a
g
n
o
s
t
i
c
s1
3
2
3
3
5
4
8
.
[
4
]
M
.
D
.
B
a
sar
a
n
d
A
.
A
k
a
n
,
“
D
e
t
e
c
t
i
o
n
o
f
c
h
r
o
n
i
c
k
i
d
n
e
y
d
i
sea
s
e
b
y
u
si
n
g
e
n
sem
b
l
e
c
l
a
s
si
f
i
e
r
s,”
i
n
Pr
o
c
.
2
0
1
7
1
0
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
s E
l
e
c
t
r
i
c
a
l
a
n
d
E
l
e
c
t
r
o
n
i
c
s E
n
g
i
n
e
e
r
i
n
g
(
ELECO)
,
2
0
1
7
,
p
p
.
5
4
4
–
5
4
7
.
[
5
]
M
.
A
l
mas
o
u
d
a
n
d
T
.
E.
W
a
r
d
,
“
D
e
t
e
c
t
i
o
n
o
f
c
h
r
o
n
i
c
k
i
d
n
e
y
d
i
se
a
se
u
si
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
w
i
t
h
l
e
a
s
t
n
u
m
b
e
r
o
f
p
r
e
d
i
c
t
o
r
s,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
d
v
a
n
c
e
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
0
,
n
o
.
8
,
p
p
.
8
9
–
9
6
,
2
0
1
9
,
d
o
i
:
1
0
.
1
4
5
6
9
/
i
j
a
c
s
a
.
2
0
1
9
.
0
1
0
0
8
1
3
.
[
6
]
A
.
N
i
s
h
a
n
t
h
a
n
d
T.
Th
i
r
u
v
a
r
a
n
,
“
I
d
e
n
t
i
f
y
i
n
g
i
m
p
o
r
t
a
n
t
a
t
t
r
i
b
u
t
e
s
f
o
r
e
a
r
l
y
d
e
t
e
c
t
i
o
n
o
f
c
h
r
o
n
i
c
k
i
d
n
e
y
d
i
s
e
a
s
e
,
”
I
EEE
R
e
v
i
e
w
s
i
n
Bi
o
m
e
d
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
1
1
,
p
p
.
2
0
8
–
2
1
6
,
2
0
1
7
,
d
o
i
:
1
0
.
1
1
0
9
/
R
B
M
E.
2
0
1
7
.
2
7
8
7
4
8
0
.
[
7
]
J.
Q
i
n
,
L
.
C
h
e
n
,
Y
.
Li
u
,
C
.
Li
u
,
C
.
F
e
n
g
,
a
n
d
B
.
C
h
e
n
,
“
A
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
me
t
h
o
d
o
l
o
g
y
f
o
r
d
i
a
g
n
o
si
n
g
c
h
r
o
n
i
c
k
i
d
n
e
y
d
i
s
e
a
s
e
,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
8
,
p
p
.
2
0
9
9
1
–
2
1
0
0
2
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
1
9
.
2
9
6
3
0
5
3
.
[
8
]
M
.
S
.
A
r
i
f
,
A
.
M
u
k
h
e
i
mer,
a
n
d
D
.
A
si
f
,
“
E
n
h
a
n
c
i
n
g
t
h
e
e
a
r
l
y
d
e
t
e
c
t
i
o
n
o
f
c
h
r
o
n
i
c
k
i
d
n
e
y
d
i
s
e
a
s
e
:
a
r
o
b
u
st
mac
h
i
n
e
l
e
a
r
n
i
n
g
mo
d
e
l
,
”
Bi
g
D
a
t
a
a
n
d
C
o
g
n
i
t
i
v
e
C
o
m
p
u
t
i
n
g
,
v
o
l
.
7
,
n
o
.
3
,
p
.
1
4
4
,
A
u
g
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
b
d
c
c
7
0
3
0
1
4
4
.
[
9
]
T.
S
a
r
o
j
a
a
n
d
Y
.
K
a
l
p
a
n
a
,
“
H
y
b
r
i
d
mi
ssi
n
g
d
a
t
a
i
m
p
u
t
a
t
i
o
n
a
n
d
n
o
v
e
l
w
e
i
g
h
t
c
o
n
v
o
l
u
t
i
o
n
n
e
u
r
a
l
n
e
t
w
o
r
k
c
l
a
ssi
f
i
e
r
f
o
r
c
h
r
o
n
i
c
k
i
d
n
e
y
d
i
se
a
se
d
i
a
g
n
o
s
i
s,
”
M
e
a
s
u
r
e
m
e
n
t
:
S
e
n
s
o
rs
,
v
o
l
.
2
7
,
p
.
1
0
0
7
1
5
,
J
u
n
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
mea
se
n
.
2
0
2
3
.
1
0
0
7
1
5
.
[
1
0
]
G
.
H
u
a
n
g
,
Z
.
Li
u
,
L.
V
a
n
D
e
r
M
a
a
t
e
n
,
a
n
d
K
.
Q
.
W
e
i
n
b
e
r
g
e
r
,
“
D
e
n
s
e
l
y
c
o
n
n
e
c
t
e
d
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
t
w
o
r
k
s,”
P
ro
c
e
e
d
i
n
g
s
-
3
0
t
h
I
EEE
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
si
o
n
a
n
d
P
a
t
t
e
r
n
Re
c
o
g
n
i
t
i
o
n
,
C
V
PR
2
0
1
7
,
v
o
l
.
2
0
1
7
-
J
a
n
u
a
r
y
,
p
p
.
2
2
6
1
–
2
2
6
9
,
2
0
1
7
,
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
.
2
0
1
7
.
2
4
3
.
[
1
1
]
R
.
F
.
E
n
g
l
e
a
n
d
S
.
M
a
n
g
a
n
e
l
l
i
,
“
C
A
V
i
a
R
:
C
o
n
d
i
t
i
o
n
a
l
a
u
t
o
r
e
g
r
e
ss
i
v
e
v
a
l
u
e
a
t
r
i
s
k
b
y
r
e
g
r
e
ss
i
o
n
q
u
a
n
t
i
l
e
s,”
J
o
u
r
n
a
l
o
f
B
u
s
i
n
e
ss
a
n
d
Ec
o
n
o
m
i
c
S
t
a
t
i
st
i
c
s
,
v
o
l
.
2
2
,
n
o
.
4
,
p
p
.
3
6
7
–
3
8
1
,
O
c
t
.
2
0
0
4
,
d
o
i
:
1
0
.
1
1
9
8
/
0
7
3
5
0
0
1
0
4
0
0
0
0
0
0
3
7
0
.
[
1
2
]
K
.
A
d
a
m,
I
.
I
.
M
o
h
d
,
a
n
d
Y
.
I
b
r
a
h
i
m,
“
A
n
a
l
y
z
i
n
g
t
h
e
i
n
st
r
u
c
t
i
o
n
s
v
u
l
n
e
r
a
b
i
l
i
t
y
o
f
d
e
n
s
e
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
t
w
o
r
k
o
n
G
P
U
S
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
t
ri
c
a
l
a
n
d
C
o
m
p
u
t
e
r
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
1
1
,
n
o
.
5
,
p
p
.
4
4
8
1
–
4
4
8
8
,
O
c
t
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
1
i
5
.
p
p
4
4
8
1
-
4
4
8
8
.
[
1
3
]
S
.
A
k
t
e
r
e
t
a
l
.
,
“
C
o
mp
r
e
h
e
n
s
i
v
e
p
e
r
f
o
r
ma
n
c
e
a
ssess
me
n
t
o
f
d
e
e
p
l
e
a
r
n
i
n
g
mo
d
e
l
s
i
n
e
a
r
l
y
p
r
e
d
i
c
t
i
o
n
a
n
d
r
i
sk
i
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
c
h
r
o
n
i
c
k
i
d
n
e
y
d
i
s
e
a
s
e
,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
9
,
p
p
.
1
6
5
1
8
4
–
1
6
5
2
0
6
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
1
.
3
1
2
9
4
9
1
.
[
1
4
]
D
.
S
a
i
f
,
A
.
M
.
S
a
r
h
a
n
,
a
n
d
N
.
M
.
El
sh
e
n
n
a
w
y
,
“
D
e
e
p
-
k
i
d
n
e
y
:
a
n
e
f
f
e
c
t
i
v
e
d
e
e
p
l
e
a
r
n
i
n
g
f
r
a
m
e
w
o
r
k
f
o
r
c
h
r
o
n
i
c
k
i
d
n
e
y
d
i
sea
s
e
p
r
e
d
i
c
t
i
o
n
,
”
H
e
a
l
t
h
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
a
n
d
S
y
st
e
m
s
,
v
o
l
.
1
2
,
n
o
.
1
,
D
e
c
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
3
7
5
5
-
0
2
3
-
0
0
2
6
1
-
8.
[
1
5
]
P
.
K
.
R
a
o
,
S
.
C
h
a
t
t
e
r
j
e
e
,
K
.
N
a
g
a
r
a
j
u
,
S
.
B
.
K
h
a
n
,
A
.
A
l
m
u
s
h
a
r
r
a
f
,
a
n
d
A
.
I
.
A
l
h
a
r
b
i
,
“
F
u
s
i
o
n
o
f
g
r
a
p
h
a
n
d
t
a
b
u
l
a
r
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
s
f
o
r
p
r
e
d
i
c
t
i
n
g
c
h
r
o
n
i
c
k
i
d
n
e
y
d
i
s
e
a
s
e
,
”
D
i
a
g
n
o
s
t
i
c
s
,
v
o
l
.
1
3
,
n
o
.
1
2
,
p
.
1
9
8
1
,
J
u
n
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
d
i
a
g
n
o
s
t
i
c
s
1
3
1
2
1
9
8
1
.
[
1
6
]
“
C
h
r
o
n
i
c
k
i
d
n
e
y
d
i
s
e
a
se
d
a
t
a
s
e
t
,
”
U
C
I
Ma
c
h
i
n
e
L
e
a
rn
i
n
g
Re
p
o
si
t
o
ry
.
h
t
t
p
s
:
/
/
a
r
c
h
i
v
e
.
i
c
s.
u
c
i
.
e
d
u
/
m
l
/
d
a
t
a
s
e
t
s
/
c
h
r
o
n
i
c
_
k
i
d
n
e
y
_
d
i
s
e
a
se
#
.
[
1
7
]
S
.
G
.
K
.
P
a
t
r
o
a
n
d
K
.
K
.
S
a
h
u
,
“
N
o
r
ma
l
i
z
a
t
i
o
n
:
a
p
r
e
p
r
o
c
e
ssi
n
g
st
a
g
e
,
”
I
a
r
j
se
t
,
p
p
.
2
0
–
2
2
,
M
a
r
.
2
0
1
5
,
d
o
i
:
1
0
.
1
7
1
4
8
/
i
a
r
j
se
t
.
2
0
1
5
.
2
3
0
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.