I
AE
S
I
n
t
e
r
n
at
ion
al
Jou
r
n
al
of
Ar
t
if
icial
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
Vol.
14
,
No.
4
,
Augus
t
2025
,
pp.
2797
~
2804
I
S
S
N:
2252
-
8938
,
DO
I
:
10
.
11591/i
jai
.
v
14
.i
4
.
pp
27
97
-
2804
2797
Jou
r
n
al
h
omepage
:
ht
tp:
//
ij
ai
.
iaes
c
or
e
.
c
om
In
t
e
g
r
at
in
g
r
an
d
om
f
or
e
st
an
d
ge
n
e
t
i
c
al
gor
it
h
m
s f
o
r
i
m
p
r
ove
d
k
id
n
e
y d
is
e
as
e
p
r
e
d
ic
t
io
n
B
om
m
an
ah
all
i
Ve
n
k
at
agiriyap
p
a
Raghavend
r
a
1
,
Anan
d
k
u
m
ar
Ra
m
ap
p
a
An
n
iger
i
1
,
Jogi
p
alya
S
h
ivan
an
j
ap
p
a
S
r
ik
a
n
t
am
u
r
t
h
y
1
,
Gu
r
u
r
aj
Raghavend
r
ar
ao
S
at
t
iger
i
2
1
D
e
pa
r
tm
e
nt
of
M
e
c
h
a
ni
c
a
l
E
ngi
ne
e
r
in
g, J
S
S
A
c
a
de
my of
T
e
c
h
ni
c
a
l
E
duc
a
ti
on,
V
is
ve
s
va
r
a
y
a
T
e
c
hnol
ogi
c
a
l
U
ni
ve
r
s
it
y,
B
e
nga
lu
r
u, I
ndi
a
2
D
e
pa
r
tm
e
nt
of
M
e
c
h
a
ni
c
a
l
E
ngi
ne
e
r
in
g, V
is
hw
a
na
th
r
a
o D
e
s
hp
a
nde
I
ns
ti
tu
te
of
T
e
c
hnol
ogy, Ha
li
ya
l,
I
ndi
a
Ar
t
icle
I
n
f
o
AB
S
T
RA
CT
A
r
ti
c
le
h
is
tor
y
:
R
e
c
e
ived
Nov
21
,
2024
R
e
vis
e
d
J
un
16
,
2025
Ac
c
e
pted
J
ul
10
,
2025
T
h
i
s
w
o
rk
o
ffer
s
a
n
o
v
e
l
met
h
o
d
fo
r
p
re
d
i
c
t
i
n
g
ch
r
o
n
i
c
k
i
d
n
e
y
d
i
s
eas
e
(CK
D
)
b
y
c
o
mb
i
n
i
n
g
ran
d
o
m
fo
re
s
t
(RF)
c
l
as
s
i
f
i
cat
i
o
n
w
i
t
h
g
en
e
t
i
c
al
g
o
ri
t
h
m
(G
A
)
t
o
o
p
t
i
mi
ze
i
mp
o
rt
a
n
t
p
arame
t
ers
.
T
h
e
d
at
a
s
et
co
m
p
ri
s
es
1
,
6
5
9
p
at
i
en
t
s
w
i
t
h
5
1
cl
i
n
i
cal
p
arame
t
ers
.
T
h
e
s
u
g
g
es
t
ed
met
h
o
d
emp
h
a
s
i
ze
s
t
h
e
o
p
t
i
m
i
zat
i
o
n
o
f
ran
d
o
m
s
t
at
e
v
al
u
es
,
t
e
s
t
s
i
ze,
an
d
e
s
s
e
n
t
i
al
h
y
p
erp
arame
t
ers
,
s
u
c
h
as
t
h
e
n
u
mb
er
o
f
t
rees
i
n
t
h
e
f
o
r
es
t
,
t
h
e
l
eas
t
n
u
m
b
er
o
f
s
am
p
l
e
s
n
ee
d
ed
at
a
l
eaf
n
o
d
e,
an
d
t
h
e
s
mal
l
es
t
n
u
m
b
er
o
f
s
am
p
l
e
s
n
eces
s
ary
t
o
s
p
l
i
t
a
n
i
n
t
er
n
al
n
o
d
e.
T
h
e
o
p
t
i
m
i
zat
i
o
n
p
r
o
ces
s
i
s
co
n
d
u
ct
e
d
i
n
t
w
o
s
t
ag
e
s
:
t
h
e
fi
r
s
t
s
t
a
g
e
o
p
t
i
m
i
zes
t
h
e
ran
d
o
m
s
t
a
t
e
a
n
d
t
e
s
t
s
i
ze,
w
h
i
l
e
t
h
e
s
eco
n
d
s
t
ag
e
f
o
cu
s
es
o
n
h
y
p
er
p
aramet
er
s
.
T
h
r
o
u
g
h
e
x
t
e
n
s
i
v
e
s
i
mu
l
at
i
o
n
s
o
v
er
5
0
r
u
n
s
,
t
h
e
s
t
u
d
y
d
em
o
n
s
t
ra
t
es
t
h
at
t
h
e
o
p
t
i
m
i
ze
d
mo
d
el
ac
h
i
e
v
e
s
an
accu
racy
ra
n
g
i
n
g
fro
m
0
.
9
4
5
1
t
o
0
.
9
7
3
8
.
T
h
e
re
s
u
l
t
s
i
n
d
i
cat
e
a
ma
x
i
m
u
m
i
n
crea
s
e
i
n
accu
rac
y
o
f
2
.
0
9
%
,
s
h
o
w
ca
s
i
n
g
t
h
e
effect
i
v
en
es
s
o
f
t
h
e
G
A
-
RF
i
n
t
eg
ra
t
ed
ap
p
ro
ac
h
i
n
en
h
a
n
ci
n
g
m
o
d
e
l
p
erfo
rma
n
ce.
T
h
i
s
w
o
r
k
p
r
o
v
i
d
e
s
v
al
u
ab
l
e
i
n
s
i
g
h
t
s
i
n
t
o
t
h
e
i
mp
ac
t
o
f
p
aramet
er
o
p
t
i
mi
zat
i
o
n
o
n
mac
h
i
n
e
l
earn
i
n
g
(ML
)
mo
d
e
l
s
,
p
art
i
cu
l
arl
y
i
n
med
i
cal
d
i
ag
n
o
s
t
i
cs
,
a
n
d
o
ffer
s
a
ro
b
u
s
t
frame
w
o
r
k
fo
r
d
e
v
el
o
p
i
n
g
h
i
g
h
l
y
accu
ra
t
e
p
red
i
ct
i
v
e
mo
d
el
s
.
K
e
y
w
o
r
d
s
:
C
hr
onic
kidney
dis
e
a
s
e
Ge
ne
ti
c
a
lgor
it
hm
M
a
c
hine
lea
r
ning
Optim
iza
ti
on
P
r
e
dictive
model
R
a
ndom
f
or
e
s
t
Th
i
s
i
s
a
n
o
p
en
a
c
ces
s
a
r
t
i
c
l
e
u
n
d
e
r
t
h
e
CC
B
Y
-
SA
l
i
ce
n
s
e.
C
or
r
e
s
pon
din
g
A
u
th
or
:
B
omm
a
na
ha
ll
i
Ve
nka
tagir
iyappa
R
a
gha
ve
ndr
a
De
pa
r
tm
e
nt
of
M
e
c
ha
nica
l
E
nginee
r
ing
,
J
S
S
Ac
a
de
my
of
T
e
c
hnica
l
E
duc
a
ti
on
Vis
ve
s
va
r
a
ya
T
e
c
hnologi
c
a
l
Unive
r
s
it
y
B
e
nga
lur
u,
I
ndia
E
mail:
bv
r
js
s
@gmail.
c
om
1.
I
NT
RODU
C
T
I
ON
T
he
tr
a
ini
ng
a
nd
tes
ti
ng
method
f
o
r
c
las
s
if
ying
bi
ologi
c
a
l
inf
or
mation
in
mac
hine
lea
r
ning
(
M
L
)
is
be
c
omi
ng
incr
e
a
s
ingl
y
s
igni
f
ica
nt.
T
he
r
e
s
e
a
r
c
he
r
f
oc
us
e
s
c
a
r
e
f
ull
y
on
c
hoos
ing
the
r
ight
methods
of
c
las
s
if
ica
ti
on
a
nd
de
ve
lopi
ng
the
pr
e
dictive
mode
l.
F
e
w
s
tudi
e
s
ha
ve
e
xa
mi
ne
d
c
las
s
if
ica
ti
on
a
nd
pr
e
dictive
model
de
ve
lopm
e
nt.
L
it
e
r
a
tur
e
s
tudi
e
s
a
r
e
p
r
e
domi
na
ntl
y
theor
e
ti
c
a
l;
ther
e
e
xis
ts
no
pr
a
c
ti
c
a
l
m
ode
l
f
or
s
a
mpl
e
s
e
lec
ti
on
in
the
tr
a
ini
ng
a
nd
tes
ti
ng
pr
oc
e
s
s
.
T
he
r
e
s
e
a
r
c
h
f
oc
us
e
s
on
type
of
c
las
s
if
ica
ti
on,
da
ta
tr
a
ini
ng
a
nd
tes
ti
ng
is
f
ound
the
gr
e
y
a
r
e
a
.
I
n
or
de
r
to
buil
d
a
nd
f
it
a
good
model
in
ML
,
t
his
a
r
ti
c
le
pr
ovides
a
n
in
-
de
pth
a
na
lys
is
of
the
s
ize
s
of
both
the
tr
a
ini
ng
a
nd
tes
ti
ng
da
tas
e
ts
.
T
his
wor
k
e
xa
mi
ne
s
the
e
f
f
e
c
t
of
r
a
ndom
s
tate
(
s
huf
f
li
ng
the
da
ta
s
e
t)
,
tes
t
s
ize
(
da
ta
s
e
t
f
or
tr
a
ini
ng
a
nd
tes
ti
ng)
,
a
nd
hype
r
-
pa
r
a
mete
r
s
on
pe
r
f
or
manc
e
in
r
a
ndom
f
or
e
s
t
(
R
F
)
ML
.
Ac
c
ur
a
c
y
is
a
metr
ic
that
c
ompar
e
s
the
pe
r
f
o
r
manc
e
metr
i
c
s
of
the
tes
t
a
nd
the
pr
e
dicte
d
da
ta
f
r
om
the
model
.
T
he
ge
ne
ti
c
a
lgor
it
hm
(
GA
)
is
int
e
gr
a
ted
int
o
R
F
c
las
s
if
ica
ti
on
in
two
s
tage
s
.
T
he
pu
r
pos
e
of
the
GA
is
to
opti
mi
z
e
th
e
pa
r
a
mete
r
s
du
r
ing
the
p
r
o
c
e
s
s
of
model
buil
ding.
T
he
s
im
ulation
r
e
s
ult
s
(
a
c
c
ur
a
c
y)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
Ar
ti
f
I
ntell
,
Vol.
14
,
No.
4
,
Augus
t
20
25
:
279
7
-
2804
2798
s
how
that
the
r
a
ndom
s
tate
,
tes
t
r
a
ti
o,
a
nd
hype
r
-
pa
r
a
mete
r
s
a
r
e
the
c
r
it
e
r
ia
that
ha
ve
a
dir
e
c
t
im
pa
c
t
on
the
model's
c
or
r
e
c
tnes
s
.
T
o
c
r
e
a
te
a
r
e
li
a
ble
s
ys
tem,
a
GA
s
houl
d
be
im
pleme
nted
int
o
the
R
F
c
las
s
if
ica
ti
on
to
f
ine
-
tune
the
c
or
r
e
c
tnes
s
of
the
r
e
s
ult
s
.
T
he
R
F
a
l
gor
it
hm
i
s
a
po
we
r
f
u
l
a
n
d
ve
r
s
a
ti
le
ML
m
e
th
od,
s
u
it
a
b
le
f
or
c
l
a
s
s
if
i
c
a
t
io
n
a
n
d
r
e
gr
e
s
s
i
on.
T
h
e
f
e
a
t
ur
e
s
of
t
he
R
F
a
l
gor
it
hm,
s
u
c
h
a
s
a
c
c
u
r
a
c
y,
r
o
bu
s
tn
e
s
s
in
h
a
n
dli
ng
mi
s
s
ing
v
a
l
u
e
s
,
s
c
a
la
bi
li
t
y,
n
on
pa
r
a
me
tr
i
c
s
u
pp
or
t
t
o
h
a
n
dli
ng
c
o
mp
le
x
i
nt
e
r
a
c
ti
o
n
s
,
r
o
bu
s
tn
e
s
s
to
noi
s
y
d
a
t
a
a
nd
f
it
t
he
e
f
f
e
c
ti
ve
mo
d
e
l
f
or
pr
e
d
ic
ti
on
in
di
a
g
no
s
in
g
kidn
e
y
d
i
s
e
a
s
e
.
M
a
n
y
r
a
n
do
mi
z
e
d
a
l
gor
it
hm
s
in
s
kle
a
r
n
u
s
e
t
h
e
r
a
nd
om
s
t
a
t
e
to
s
e
l
e
c
t
th
e
r
a
nd
om
s
e
e
d
to
f
e
e
d
to
t
h
e
p
s
e
ud
o
-
r
a
n
do
m
nu
mb
e
r
g
e
ne
r
a
t
or
.
T
h
e
m
o
s
t
p
o
pul
a
r
i
nt
e
g
e
r
s
a
r
e
0
a
n
d
42.
W
h
e
n
u
s
in
g
a
n
i
nte
g
e
r
f
or
t
h
e
r
a
nd
om
s
t
a
t
e
(
c
ha
ng
in
g
th
e
or
de
r
o
f
t
r
a
in
in
g
s
a
m
pl
e
s
yi
e
ld
s
v
a
r
ie
d
r
e
s
u
lt
s
)
,
t
h
e
f
u
nc
ti
on
wil
l
p
r
o
du
c
e
th
e
s
a
m
e
r
e
s
ul
t
s
a
c
r
o
s
s
d
if
f
e
r
e
nt
e
x
e
c
u
ti
o
n
s
.
T
h
e
r
e
s
u
lt
s
o
nl
y
c
h
a
n
ge
if
t
he
int
e
ge
r
v
a
lu
e
i
s
a
l
t
e
r
e
d
,
a
nd
i
n
c
r
e
a
s
i
ng
th
e
n
um
be
r
of
r
un
s
wil
l
l
ik
e
l
y
d
e
c
r
e
a
s
e
t
h
e
v
a
r
ia
n
c
e
.
T
h
e
r
e
a
s
o
n
f
or
v
a
r
i
a
n
c
e
i
n
pe
r
f
o
r
m
a
n
c
e
i
s
a
s
a
mp
le
t
ha
t
i
s
t
oo
s
m
a
ll
(
a
n
d/
or
m
a
n
y
f
e
a
tu
r
e
s
/c
la
s
s
e
s
w
hi
c
h
i
s
t
oo
hi
gh
)
,
wh
ic
h
c
a
u
s
e
s
th
e
mo
de
l
s
to
o
ve
r
-
f
i
t.
T
o
de
c
r
e
a
s
e
t
he
v
a
r
ia
n
c
e
by
i
n
c
r
e
a
s
i
ng
th
e
num
b
e
r
of
r
u
n
s
.
C
hr
onic
kidney
dis
e
a
s
e
(
C
KD
)
is
a
long
-
ter
m
he
a
lt
h
c
ondit
ion
that
typ
ica
ll
y
las
ts
a
l
if
e
ti
me
a
nd
a
r
is
e
s
due
to
k
idney
c
a
nc
e
r
or
dim
ini
s
he
d
ki
dne
y
f
unc
ti
on
.
T
he
pr
ogr
e
s
s
ion
of
thi
s
c
hr
onic
il
lnes
s
c
a
n
be
s
topped
or
s
lowe
d
down
to
the
point
whe
r
e
the
pa
t
ient’
s
li
f
e
c
a
n
be
s
us
taine
d
only
thr
ough
dialys
is
or
s
ur
ge
r
y
[
1]
.
E
a
r
l
ier
de
tec
ti
on
o
f
C
KD
is
di
f
f
icult
a
mong
pa
ti
e
nts
,
due
to
no
s
ympt
oms
a
nd
va
r
ying
r
a
tes
o
f
kidney
dis
e
a
s
e
pr
ogr
e
s
s
ion.
T
im
e
ly
a
nd
p
r
e
c
is
e
pr
e
dict
ion
of
kidney
dis
e
a
s
e
is
e
s
s
e
nti
a
l
f
or
e
f
f
e
c
ti
ve
dis
e
a
s
e
mana
ge
ment
[
2]
.
T
he
th
ir
d
objec
ti
ve
of
the
UN
'
s
s
us
taina
bl
e
de
ve
lopm
e
nt
goa
ls
(
S
DG
)
f
oc
us
e
s
on
good
he
a
lt
h
a
nd
we
ll
-
be
ing,
highl
ight
ing
the
g
r
owing
c
h
a
ll
e
nge
s
pos
e
d
by
non
-
c
omm
unica
bl
e
dis
e
a
s
e
s
.
One
of
the
S
DG
tar
ge
ts
f
or
2030
is
to
r
e
duc
e
pr
e
matur
e
d
e
a
ths
f
r
om
non
-
c
omm
unica
ble
dis
e
a
s
e
s
by
one
-
t
hir
d
[
3]
.
Kidne
y
inj
ur
y
is
ir
r
e
ve
r
s
ibl
e
a
nd
c
a
n
a
dva
nc
e
to
e
nd
-
s
tage
r
e
na
l
dis
e
a
s
e
(
E
S
R
D)
,
e
ve
ntually
r
e
quir
i
ng
r
e
na
l
r
e
plac
e
ment
ther
a
py
(
R
R
T
)
due
to
the
los
s
of
r
e
maining
kidney
f
unc
ti
on
[
4]
–
[
6]
.
Ac
c
or
ding
to
Z
ha
o
e
t
al
.
[
7]
,
t
r
e
a
ti
ng
C
KD
a
nd
r
e
na
l
f
a
il
ur
e
is
e
xpe
ns
ive
a
nd
of
ten
inef
f
e
c
ti
ve
.
E
a
r
ly
a
nd
a
c
c
ur
a
te
diagnos
is
,
a
long
with
ti
mely
t
r
e
a
tm
e
nt,
is
e
s
s
e
nti
a
l
f
or
e
f
f
e
c
ti
ve
ma
na
ge
ment
of
C
KD
.
T
his
s
tudy
a
im
s
to
de
s
ign
a
nd
va
li
da
te
a
pr
e
dictive
model
f
o
r
identif
y
i
ng
C
KD
.
I
n
pr
e
vious
r
e
s
e
a
r
c
h,
P
a
l
[
4]
e
mpl
oye
d
thr
e
e
ML
a
lgo
r
it
hms
—
logi
s
ti
c
r
e
gr
e
s
s
ion
(
L
R
)
,
de
c
is
ion
tr
e
e
(
DT
)
,
a
nd
s
uppor
t
ve
c
tor
mac
hine
(
S
VM
)
—
to
c
ons
tr
uc
t
a
p
r
e
dictive
model.
S
im
il
a
r
ly,
Kha
li
d
e
t
al
.
[
8]
int
r
oduc
e
d
a
h
ybr
id
a
ppr
oa
c
h
that
int
e
gr
a
ted
Ga
us
s
ian
n
a
ïve
B
a
ye
s
(
f
or
gr
a
dient
boos
ti
ng)
a
nd
a
DT
a
s
the
ba
s
e
lea
r
ne
r
,
with
a
R
F
model
s
e
r
ving
a
s
the
meta
-
c
las
s
if
ier
.
D
e
ba
l
a
nd
S
it
ote
[
3]
inves
ti
ga
ted
C
KD
us
ing
pr
e
dictive
mod
e
ls
s
uc
h
a
s
R
F
,
S
VM
,
a
nd
DT
.
I
n
a
nother
s
tudy
,
S
a
if
e
t
al
.
[
9]
pr
opos
e
d
th
r
e
e
di
f
f
e
r
e
nt
mo
de
ls
a
im
e
d
a
t
p
r
e
dicting
C
KD
6
to
12
mont
hs
be
f
or
e
c
li
nica
l
s
y
mpt
oms
a
ppe
a
r
,
e
mpl
oying
s
ophis
ti
c
a
ted
a
ppr
oa
c
he
s
li
ke
c
onvolut
ional
ne
ur
a
l
ne
twor
ks
(
C
NN
s
)
,
long
s
hor
t
-
ter
m
memor
y
(
L
S
T
M
)
models
,
a
nd
de
e
p
e
ns
e
mbl
e
lea
r
ning
tec
hniques
.
R
a
hman
e
t
al
.
[
10]
f
oc
us
e
d
on
e
nha
nc
ing
c
las
s
if
ica
ti
on
pe
r
f
or
manc
e
by
a
pplyi
ng
f
e
a
tur
e
s
e
l
e
c
ti
on
methods
s
uc
h
a
s
r
e
c
ur
s
ive
f
e
a
tur
e
e
li
m
inati
on
(
R
F
E
)
a
nd
the
B
or
uta
a
lgor
it
hm,
a
long
with
mul
ti
ple
pe
r
f
or
manc
e
metr
ics
,
to
identi
f
y
opti
mal
c
las
s
if
ie
rs
,
s
tr
iki
ng
a
ba
lanc
e
be
twe
e
n
high
a
c
c
ur
a
c
y
a
nd
low
c
omput
a
ti
ona
l
c
os
t.
Additi
ona
ll
y,
L
e
i
e
t
al
.
[
2]
c
onduc
ted
a
c
ompr
e
he
ns
ive
meta
-
a
na
lys
is
to
e
va
luate
how
a
c
c
ur
a
tely
ML
tec
hnique
s
c
a
n
diagnos
e
the
pr
ogr
e
s
s
ion
of
kidney
dis
e
a
s
e
.
Dr
it
s
a
s
a
nd
T
r
igka
[
11]
p
r
opos
e
d
a
ML
-
ba
s
e
d
s
tr
a
tegy
f
or
a
s
s
e
s
s
ing
C
KD
r
is
k,
leve
r
a
ging
a
r
a
nge
of
models
including
p
r
oba
bil
is
ti
c
,
tr
e
e
-
ba
s
e
d,
a
nd
e
ns
e
mbl
e
a
ppr
oa
c
he
s
s
u
c
h
a
s
S
VM
,
LR
,
s
tocha
s
ti
c
gr
a
dient
de
s
c
e
nt
(
S
GD
)
,
a
r
ti
f
icia
l
ne
ur
a
l
ne
twor
ks
(
AN
N)
,
a
nd
k
-
ne
a
r
e
s
t
ne
ighbor
s
(k
-
NN
)
.
L
e
i
e
t
al
.
[
2
]
pe
r
f
or
med
a
s
ys
tema
ti
c
meta
-
a
na
lys
is
to
a
s
s
e
s
s
the
di
a
gnos
ti
c
a
c
c
ur
a
c
y
of
M
L
a
lgor
it
hms
f
or
kidney
dis
e
a
s
e
pr
ogr
e
s
s
ion.
Dr
it
s
a
s
a
nd
T
r
igka
[
11]
de
ve
loped
a
n
M
L
methodology
to
pr
e
dict
C
KD
r
is
k,
uti
li
z
ing
p
r
oba
bil
is
ti
c
,
t
r
e
e
-
ba
s
e
d,
a
nd
e
ns
e
mbl
e
lea
r
ning
models
,
including
S
VM
,
L
R
,
S
GD
,
AN
N,
a
nd
k
-
NN.
L
im
e
t
al
.
[
12]
r
e
view
e
d
C
KD
a
nd
noted
that
C
ox
r
e
g
r
e
s
s
ion
modeling
wa
s
the
mos
t
c
omm
only
us
e
d
method
a
mong
the
f
e
w
s
tudi
e
s
e
xa
mi
ne
d.
Aoki
e
t
al
.
[
13]
e
xplor
e
d
the
a
ppli
c
a
ti
on
of
ML
tec
hniques
,
including
R
F
s
ur
vival
models
,
to
s
tudy
C
KD
in
the
U
.
S
.
,
f
oc
us
ing
on
labor
a
tor
y
-
de
r
ived
r
is
k
f
a
c
tor
s
a
s
pr
e
dictor
s
of
e
s
ti
mate
d
glom
e
r
ular
f
il
tr
a
ti
on
r
a
te
(
e
GFR
)
.
B
ins
a
wa
d
[
14]
a
na
lyze
d
the
c
or
r
e
lation
be
twe
e
n
kidney
f
unc
ti
on
a
nd
e
lec
tr
oc
a
r
diogr
a
m
(
E
C
G)
r
e
a
dings
us
ing
a
n
opti
mi
z
e
d
R
F
model,
de
mons
tr
a
ti
ng
s
upe
r
ior
pe
r
f
o
r
manc
e
in
ter
ms
of
c
las
s
if
ica
ti
on
a
c
c
ur
a
c
y
(
C
A)
,
f
a
l
s
e
pos
it
ive
r
a
te
(
F
P
R
)
,
a
nd
t
r
ue
pos
it
ive
r
a
te
(
T
P
R
)
whe
n
c
omp
a
r
e
d
to
other
methods
.
He
ma
e
t
al
.
[
15
]
,
ut
il
izi
ng
both
s
tanda
r
d
a
nd
r
e
a
l
-
ti
me
da
tas
e
ts
,
a
s
s
e
s
s
e
d
the
e
f
f
e
c
ti
ve
ne
s
s
of
va
r
ious
M
L
a
lgor
it
hms
—
s
uc
h
a
s
k
-
NN
,
R
F
,
DT
,
g
r
a
dient
boos
ti
ng
,
a
nd
e
xt
r
e
me
gr
a
dient
boos
t
ing
(
XG
B
oos
t
)
—
in
f
or
e
c
a
s
ti
ng
C
KD
.
T
a
kka
va
taka
r
n
e
t
al
.
[
16]
f
oc
us
e
d
on
s
tage
-
4
C
KD
,
e
mpl
oying
f
our
dif
f
e
r
e
nt
models
—
L
ASS
O
r
e
g
r
e
s
s
ion,
R
F
,
XG
B
oos
t,
a
nd
ANN
—
to
pr
e
dict
the
pr
ogr
e
s
s
ion
to
e
nd
-
s
tage
k
idney
dis
e
a
s
e
(
E
S
KD
)
.
S
a
nmar
c
hi
e
t
al
.
[
17]
p
r
ovided
a
c
ompr
e
he
ns
ive
r
e
view
of
M
L
methodologi
e
s
us
e
d
in
C
KD
r
e
s
e
a
r
c
h,
outl
ini
ng
both
the
potential
a
d
va
ntage
s
a
nd
li
mi
tations
of
thes
e
tec
hniques
in
d
iagnos
is
,
pr
ognos
is
,
a
nd
dis
e
a
s
e
mana
ge
ment.
Z
hu
e
t
al
.
[
18
]
de
ve
loped
a
pipeline
to
pr
oc
e
s
s
lon
git
udinal
e
lec
t
r
onic
he
a
lt
h
r
e
c
or
ds
(
E
HR
s
)
a
nd
a
ppli
e
d
r
e
c
ur
r
e
nt
ne
ur
a
l
ne
twor
ks
(
R
NN
s
)
to
f
o
r
e
c
a
s
t
the
pr
ogr
e
s
s
ion
of
C
KD
f
r
om
s
tage
s
I
I
/
I
I
I
to
I
V/V.
Additi
ona
ll
y,
G
hos
h
a
nd
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
Ar
ti
f
I
ntell
I
S
S
N:
2252
-
8938
I
ntegr
ati
ng
r
andom
for
e
s
t
and
ge
ne
ti
c
algor
it
hms
f
or
…
(
B
omm
anahall
i
V
e
nk
atagi
r
iyappa
R
aghav
e
n
dr
a
)
2799
Kha
ndoke
r
[
19]
e
va
luate
d
M
L
-
ba
s
e
d
pr
e
diction
models
f
or
C
KD
a
nd
highl
ight
e
d
the
int
e
r
p
r
e
tabili
ty
of
S
ha
pley
a
ddit
ive
e
xplana
ti
ons
(
S
HA
P
)
a
nd
l
oc
a
l
int
e
r
pr
e
table
model
-
a
gnos
ti
c
e
xplana
ti
ons
(
L
I
M
E
)
f
r
a
mew
or
ks
,
whic
h
o
f
f
e
r
va
luable
c
li
nica
l
ins
ight
s
f
or
he
a
lt
hc
a
r
e
pr
o
f
e
s
s
ionals
.
Altur
ki
e
t
a
l
.
[
20]
s
tudi
e
d
C
KD
pr
e
diction
us
ing
R
F
with
a
c
c
ur
a
c
y
o
f
9
2.
85%
with
s
ynthetic
mi
no
r
it
y
ove
r
s
a
mpl
ing
t
e
c
hnique
(
S
M
OT
E
)
.
R
a
hman
e
t
al
.
[
10
]
us
e
d
ML
a
lgor
it
hm
f
or
C
KD
pr
e
diction
with
99
.
75%
a
c
c
ur
a
c
y.
R
a
jea
s
hwa
r
i
a
nd
Ar
une
s
h
[
21]
us
e
d
de
e
p
c
onvolut
ional
ne
ur
a
l
ne
twor
k
(
DC
NN
)
a
nd
modi
f
ied
e
xtr
e
me
r
a
ndom
f
or
e
s
t
(
M
E
R
F
)
a
ppr
oa
c
he
s
we
r
e
us
e
d
to
p
r
e
dict
C
KD
wit
h
98.
5%
a
c
c
ur
a
c
y.
F
r
om
the
li
ter
a
tu
r
e
r
e
view
,
i
t
is
lea
r
ne
d
that
s
e
ve
r
a
l
s
tudi
e
s
a
r
e
a
va
il
a
ble
on
C
KD
p
r
e
diction
th
r
ough
mac
hine
-
lea
r
ning
a
ppr
oa
c
he
s
.
I
n
the
s
tudy
of
C
KD
,
va
r
ious
pa
r
a
mete
r
s
li
ke
da
tas
e
t
s
ize
,
qua
li
ty
of
da
tas
e
t,
a
nd
the
ti
mi
ng
of
da
ta
c
oll
e
c
ti
on
play
a
c
r
uc
ial
r
o
le.
I
n
the
pr
e
s
e
nt
s
tudy,
a
f
oc
us
on
C
KD
pr
e
dicti
on
us
ing
M
L
models
f
or
the
big
da
tas
e
t
is
c
ons
ider
e
d
by
a
two
-
s
tage
hybr
id
model
of
R
F
a
nd
GA
.
B
y
thi
s
a
ppr
oa
c
h,
the
a
c
c
ur
a
c
y
of
th
e
hybr
id
model
im
p
r
ove
s
.
2.
M
E
T
HO
D
Dia
gnos
ing
kidney
dis
e
a
s
e
is
a
c
r
uc
ial
a
s
pe
c
t
o
f
he
a
lt
hc
a
r
e
,
whic
h
f
oc
us
e
s
on
i
mpr
oving
he
a
lt
h
thr
ough
the
p
r
e
ve
nti
on,
diagnos
is
,
tr
e
a
tm
e
nt,
a
n
d
mana
ge
ment
of
d
is
e
a
s
e
s
a
nd
im
pa
ir
ments
.
T
he
e
a
r
ly
diagnos
is
he
lps
s
oc
iety
to
unde
r
go
e
a
r
ly
tr
e
a
tm
e
nt
a
nd
he
nc
e
a
void
f
ur
ther
da
mage
to
the
he
a
lt
h.
Da
ta
a
bout
kidney
dis
e
a
s
e
is
c
oll
e
c
ted
f
r
om
ope
n
s
our
c
e
at
htt
ps
:/
/www
.
ka
ggle.
c
om/
da
tas
e
ts
/r
a
biee
lkhar
oua
/chr
onic
-
kidney
-
dis
e
a
s
e
-
da
ta
s
e
t
-
a
n
a
lys
is
.
T
he
da
ta
of
1
,
659
pe
r
s
ons
te
s
ted
f
or
51
va
r
ious
pa
r
a
mete
r
s
be
f
or
e
diagnos
ing
the
C
KD
.
A
pr
e
dictive
model
is
ge
ne
r
a
ted
us
ing
1
,
659
pe
r
s
o
ns
with
51
tes
ted
pa
r
a
mete
r
s
.
T
he
RF
c
las
s
if
ier
is
uti
li
z
e
d
f
or
c
a
tegor
izing
the
da
ta
to
f
it
the
model
a
nd
to
p
r
e
dict
.
T
he
e
f
f
e
c
ti
ve
ne
s
s
of
the
model
is
de
t
e
r
mi
ne
d
thr
ough
the
a
c
c
ur
a
c
y,
whic
h
c
ompa
r
e
s
or
de
viati
on
of
the
da
ta
c
ons
ider
e
d
a
nd
da
ta
ge
ne
r
a
ted
thr
ough
the
model.
Highe
r
a
c
c
ur
a
c
y
im
pli
e
s
the
model
is
e
f
f
e
c
ti
ve
in
p
r
e
diction.
T
he
e
f
f
e
c
ti
ve
ne
s
s
of
the
model
in
ter
ms
of
a
c
c
ur
a
c
y
in
the
R
F
c
las
s
i
f
ica
ti
on
de
pe
nds
u
pon
va
r
ious
pa
r
a
mete
r
s
v
iz,
r
a
ndom
s
tate
,
tes
t
s
ize
,
a
nd
hype
r
pa
r
a
mete
r
s
.
Ke
y
hype
r
pa
r
a
mete
r
s
in
R
F
c
las
s
if
ica
ti
on
include
the
number
o
f
t
r
e
e
s
(
n_e
s
ti
mator
s
)
,
whic
h
indi
c
a
tes
how
many
DT
s
the
model
will
ge
ne
r
a
t
e
dur
ing
tr
a
ini
ng
,
s
e
lec
ti
ng
the
be
s
t
one
thr
ough
major
it
y
voti
ng.
Othe
r
im
po
r
tant
pa
r
a
mete
r
s
a
r
e
the
mi
nim
um
number
of
s
a
mpl
e
s
r
e
quir
e
d
a
t
a
lea
f
node
(
mi
n_s
a
mpl
e
s
_lea
f
)
,
whic
h
e
ns
ur
e
s
that
a
s
pli
t
p
oint
a
t
a
ny
de
pth
lea
ve
s
a
t
lea
s
t
the
s
pe
c
if
ied
m
ini
mum
number
of
t
r
a
ini
ng
s
a
mpl
e
s
i
n
both
br
a
nc
he
s
,
a
nd
the
mi
nim
um
number
of
s
a
mpl
e
s
ne
e
de
d
to
s
pli
t
a
n
int
e
r
na
l
node
.
I
n
thi
s
r
e
s
e
a
r
c
h,
a
GA
is
e
mpl
oye
d
to
e
nha
nc
e
th
e
model's
a
c
c
ur
a
c
y
in
two
s
tage
s
.
S
tage
1
,
du
r
ing
the
c
las
s
if
ica
ti
on
of
the
da
ta,
a
nd
s
tage
2,
in
tuni
ng
the
hype
r
pa
r
a
mete
r
s
.
I
n
s
tage
1,
the
pr
e
c
is
io
n
of
the
model
in
ter
ms
of
a
c
c
ur
a
c
y
is
a
na
lyze
d
by
s
huf
f
li
ng
the
da
ta
s
e
ts
a
long
with
the
tes
t
da
tas
e
t
s
ize
.
T
he
opti
mi
z
a
ti
on
of
thes
e
two
pa
r
a
mete
r
s
f
oc
us
e
s
o
n
a
c
c
ur
a
c
y
thr
ough
the
us
e
of
a
GA
.
F
igur
e
1
s
hows
a
r
c
hit
e
c
tur
e
of
the
pr
oc
e
s
s
.
S
tage
1:
T
he
lowe
r
a
nd
uppe
r
bound
of
the
r
a
ndom
s
tate
a
nd
tes
t
s
ize
is
s
e
t
a
s
[
13]
R
a
ndom_s
tate
=
[
0,
100]
T
e
s
t_s
ize
=
[
0.
1,
0.
5
]
S
tage
1
&2
:
GA
pa
r
a
mete
r
s
s
e
t
to
M
a
xim
um
number
of
it
e
r
a
ti
ons
:30
P
opulation
s
ize
:30
M
utation
p
r
oba
bil
it
y
=
0.
1
E
li
t
r
a
ti
o
=
0
.
01
C
r
os
s
ove
r
p
r
oba
bil
it
y
=
0
.
85
P
a
r
e
nts
p
or
ti
on
=
0.
3
C
r
os
s
ove
r
type
=
Unif
or
m
GA
is
us
e
d
to
opti
mi
z
e
hype
r
pa
r
a
mete
r
s
f
o
r
R
F
c
l
a
s
s
if
ica
ti
ons
e
mpl
oying
e
volut
ionar
y
s
tr
a
tegie
s
to
s
e
a
r
c
h
f
or
the
be
s
t
hype
r
pa
r
a
mete
r
s
e
t.
GA
ar
e
ba
s
e
d
on
na
tur
a
l
s
e
lec
ti
on
pr
inciples
a
nd
e
mpl
oy
tec
hniques
li
ke
s
e
lec
ti
on,
c
r
os
s
ove
r
,
a
nd
mut
a
ti
on
to
e
volve
s
olut
ions
to
opti
mi
z
a
ti
on
pr
ob
lems
.
T
he
opt
im
ize
d
r
a
ndom
s
tate
a
nd
tes
t
s
ize
,
de
ter
mi
ne
d
f
or
model
a
c
c
ur
a
c
y,
a
r
e
s
ubs
e
que
ntl
y
us
e
d
a
s
input
pa
r
a
mete
r
s
in
the
s
e
c
ond
leve
l
of
the
GA
to
r
e
f
ine
the
hype
r
pa
r
a
mete
r
s
of
the
R
F
c
las
s
if
ier
.
T
his
a
ddit
ional
s
tep
e
nha
nc
e
s
the
hype
r
pa
r
a
mete
r
s
to
a
c
hieve
be
tt
e
r
a
c
c
ur
a
c
y
f
or
mo
de
l
f
it
ti
ng
a
nd
pr
e
diction
[
22]
–
[
24]
.
T
his
s
tudy
f
oc
us
e
s
on
opti
mi
z
ing
the
r
a
ndom
s
tate
,
tes
t
s
ize
,
a
nd
thr
e
e
ke
y
hype
r
pa
r
a
mete
r
s
:
the
n_e
s
ti
mator
s
,
the
mi
n_s
a
mpl
e
s
_lea
f
,
a
nd
the
mi
ni
mum
number
o
f
s
a
mpl
e
s
r
e
quir
e
d
f
or
a
s
pli
t
.
T
he
s
im
ulation
is
te
r
mi
na
ted
if
the
c
ondit
ions
(
low
e
r
/and
uppe
r
bound
)
a
r
e
not
s
a
ti
s
f
ied
Hype
r
p
a
r
a
mete
r
s
lowe
r
a
nd
uppe
r
bound
s
e
t
to
:
[
25]
Numbe
r
of
t
r
e
e
s
(
n_e
s
ti
mator
s
)
=
[
1,
100]
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
Ar
ti
f
I
ntell
,
Vol.
14
,
No.
4
,
Augus
t
20
25
:
279
7
-
2804
2800
M
ini
mum
s
a
mpl
e
s
lea
f
(
mi
n_s
a
mpl
e
s
_lea
f
)
=
[
1,
1
0]
M
ini
mum
s
a
mpl
e
s
to
s
pli
t
(
)
=
[
2
,
10
]
F
igur
e
1.
Ar
c
hit
e
c
tur
e
of
GA
int
e
gr
a
ted
with
R
F
3.
RE
S
UL
T
S
AN
D
DI
S
CU
S
S
I
ON
T
he
pr
e
dictive
model
is
ge
ne
r
a
ted
a
nd
s
im
ulations
we
r
e
pe
r
f
or
med
with
R
F
c
las
s
if
ica
ti
on
a
s
f
oll
ows
the
a
c
c
ur
a
c
y
obtaine
d
in
thes
e
s
im
ulations
is
tabul
a
ted:
‒
T
e
s
t
s
ize
10
to
50
a
nd
R
a
ndom
S
tate
0
to
100
‒
I
ntegr
a
ti
ng
GA
wi
th
R
F
in
two
leve
ls
,
S
tage
1:
T
o
opti
mi
z
e
tes
t
s
ize
a
nd
r
a
ndom
s
tate
,
S
tage
2:
I
nput
of
opti
mi
z
e
d
tes
t
s
ize
a
nd
r
a
ndom
s
tate
to
opt
im
ize
hype
r
pa
r
a
mete
r
s
s
uc
h
a
s
N
e
s
ti
mator
s
,
mi
nim
um
lea
f
,
a
nd
mi
nim
um
s
pli
t.
T
he
objec
t
ive
of
the
GA
is
to
i
mpr
ove
the
a
c
c
ur
a
c
y
of
the
c
las
s
if
ica
ti
on
by
s
e
lec
ti
ng
a
ppr
opr
iate
e
f
f
e
c
ti
ng
pa
r
a
mete
r
s
.
T
he
s
im
ulations
we
r
e
c
onduc
ted
f
or
tes
t
s
ize
10
to
50
a
nd
r
a
ndom
s
tate
0
to
100.
T
he
a
c
c
ur
a
c
y
obtaine
d
is
tabula
ted
in
T
a
ble
1
.
T
he
s
im
ulation
r
e
s
ult
in
T
a
ble
1
r
e
ve
a
ls
that
a
maximum
a
c
c
ur
a
c
y
o
f
0.
9518
is
obtaine
d
f
o
r
tes
t
s
ize
10
f
or
r
a
ndom
s
tage
s
10
&
100.
I
ntegr
a
ti
ng
GA
with
R
F
in
two
leve
ls
,
s
ta
ge
1:
t
o
opti
mi
z
e
tes
t
s
ize
a
nd
r
a
ndom
s
tate
,
s
tage
2:
i
n
put
of
op
ti
mi
z
e
d
tes
t
s
i
z
e
a
nd
r
a
ndom
s
tate
to
opti
mi
z
e
hype
r
pa
r
a
mete
r
s
s
uc
h
a
s
N
E
s
ti
mator
s
,
mi
nim
um
l
e
a
f
,
a
nd
mi
nim
um
s
pit
.
A
tot
a
l
o
f
50
s
im
ulation
r
u
ns
we
r
e
c
onduc
ted,
a
nd
the
f
indi
ngs
a
r
e
s
umm
a
r
ize
d
in
T
a
ble
2.
T
a
ble
1
.
S
im
ulation
r
e
s
ult
of
tes
t
s
ize
a
nd
r
a
ndom
s
tate
on
a
c
c
ur
a
c
y
T
e
s
t
s
iz
e
R
a
ndom
s
ta
te
0
10
20
30
40
50
60
70
80
90
100
A
c
c
ur
a
c
y
10
0.8916
0.9518
0.9337
0.9036
0.9337
0.9157
0.9337
0.9036
0.9036
0.9398
0.9518
15
0.8956
0.9116
0.9438
0.9157
0.9277
0.9197
0.9277
0.8916
0.9116
0.9197
0.9157
20
0.9006
0.9127
0.9367
0.9157
0.9187
0.9096
0.9277
0.9066
0.9187
0.9066
0.9157
25
0.9012
0.8988
0.9181
0.9205
0.9205
0.9181
0.9205
0.9060
0.9157
0.9060
0.9084
30
0.9076
0.8956
0.9197
0.9237
0.9197
0.9137
0.9217
0.9157
0.9177
0.9116
0.9096
35
0.9105
0.9002
0.9157
0.9243
0.9243
0.9191
0.9208
0.9157
0.9243
0.9105
0.9053
40
0.9142
0.9066
0.9142
0.9217
0.9292
0.9187
0.9187
0.9172
0.9187
0.9157
0.9066
45
0.9170
0.9090
0.9116
0.9183
0.9210
0.9183
0.9157
0.9264
0.9237
0.9157
0.9116
50
0.9157
0.9096
0.9072
0.9229
0.9193
0.9205
0.9157
0.9277
0.9205
0.9120
0.9157
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
Ar
ti
f
I
ntell
I
S
S
N:
2252
-
8938
I
ntegr
ati
ng
r
andom
for
e
s
t
and
ge
ne
ti
c
algor
it
hms
f
or
…
(
B
omm
anahall
i
V
e
nk
atagi
r
iyappa
R
aghav
e
n
dr
a
)
2801
T
a
ble
2.
S
im
ulation
r
e
s
ult
of
GA
int
e
g
r
a
ted
R
F
in
t
wo
leve
ls
N
o. of
R
un
S
ta
ge
1:
O
ut
put
of
G
A
t
o opt
im
iz
e
r
a
ndom s
ta
te
a
nd t
e
s
t
s
i
z
e
S
ta
ge
2:
O
ut
put
of
G
A
t
o opt
im
iz
e
hype
r
-
pa
r
a
me
te
r
s
R
a
ndom s
ta
te
T
e
s
t
s
iz
e
R
a
ndom s
ta
te
T
e
s
t
s
iz
e
R
a
ndom s
ta
te
1
97
0.416
0.9507
26
2
8
0.9522
2
34
0.103
0.9593
8
2
5
0.9650
3
44
0.207
0.9623
13
1
9
0.9652
4
86
0.100
0.9581
31
1
3
0.9640
5
14
0.110
0.9529
85
2
3
0.9528
6
44
0.180
0.9658
7
8
7
0.9623
7
77
0.110
0.9581
14
1
3
0.9685
8
14
0.220
0.9539
22
1
6
0.9593
9
3
0.200
0.9451
4
7
9
0.9542
10
20
0.160
0.9513
9
2
5
0.9550
11
77
0.128
0.9624
26
2
10
0.9671
12
37
0.130
0.9509
11
2
5
0.9554
13
97
0.350
0.9504
30
2
2
0.9504
14
9
0.130
0.9577
17
1
5
0.9624
15
12
0.380
0.9504
17
6
5
0.9504
16
14
0.220
0.9526
26
1
2
0.9582
17
44
0.140
0.9612
27
7
4
0.9655
18
44
0.150
0.9592
47
1
2
0.9632
19
14
0.220
0.9539
4
5
5
0.9566
20
44
0.100
0.9651
8
4
10
0.9651
21
21
0.120
0.9598
12
5
6
0.9648
22
34
0.110
0.9558
12
1
7
0.9613
23
89
0.110
0.9572
10
5
10
0.9626
24
32
0.100
0.9538
11
5
9
0.9595
25
21
0.144
0.9585
21
2
8
0.9627
26
77
0.130
0.9630
13
4
7
0.9722
27
97
0.360
0.9502
31
4
7
0.9518
28
14
0.220
0.9570
56
2
6
0.9569
29
86
0.100
0.9540
3
7
6
0.9655
30
9
0.110
0.9529
6
2
2
0.9738
31
14
0.190
0.9527
17
1
5
0.9558
32
44
0.200
0.9585
33
1
10
0.9614
33
9
0.120
0.9543
28
4
10
0.9593
34
44
0.600
0.9590
5
10
4
0.9664
35
34
0.110
0.9545
18
1
4
0.9600
36
14
0.220
0.9530
26
1
3
0.9558
37
44
0.130
0.9628
13
1
3
0.9720
38
14
0.220
0.9562
7
7
6
0.9616
39
21
0.210
0.9590
3
2
6
0.9692
40
14
0.200
0.9527
61
1
4
0.9556
41
3
0.210
0.9484
14
4
4
0.9512
42
44
0.180
0.9623
25
1
4
0.9617
43
44
0.110
0.9568
33
1
10
0.9675
44
9
0.130
0.9526
5
6
6
0.9668
45
21
0.130
0.9628
8
4
7
0.9674
46
44
0.140
0.9657
21
1
4
0.9699
47
77
0.130
0.9593
2
10
4
0.9683
48
86
0.130
0.9471
5
4
10
0.9519
49
21
0.130
0.9631
87
4
5
0.9631
50
9
0.15
0.9547
4
10
8
0.967
T
he
r
e
s
ult
of
GA
in
s
tage
1
a
nd
s
tage
2
indi
c
a
tes
t
ha
t
the
a
c
c
ur
a
c
y
va
r
ies
f
r
om
0
.
9451
to
0.
9738
f
o
r
opti
mi
z
e
d
r
a
ndom
s
tate
a
nd
tes
t
s
ize
.
I
n
s
tage
2,
the
a
c
c
ur
a
c
y
is
de
c
r
e
a
s
e
d
in
4
r
uns
,
with
no
c
ha
nge
with
1
r
un
out
of
50
,
a
nd
in
the
r
e
maining
s
im
ulation
r
u
n
the
a
c
c
ur
a
c
y
incr
e
a
s
e
d
to
a
maxi
mum
of
0
.
0209
(
2.
09%
)
.
45
r
uns
out
of
50
indi
c
a
ted
that
int
e
gr
a
ti
ng
GA
wit
h
R
F
in
s
tage
s
1
a
nd
2
im
pr
ove
d
the
a
c
c
ur
a
c
y.
T
hi
s
f
ur
ther
lea
ds
to
f
it
ti
ng
the
pr
e
dictive
model
mo
r
e
e
f
f
e
c
ti
ve
ly.
T
he
GA
-
ge
ne
r
a
ted
s
im
ulation
of
s
tage
1
(
50
th
r
un
in
T
a
ble
2
)
s
hows
in
F
igur
e
2
that
the
a
c
c
ur
a
c
y
obtaine
d
is
0.
9547
with
r
a
ndom
s
tage
9
a
nd
tes
t
s
ize
0.
15.
F
igur
e
2
indi
c
a
tes
that
no
im
pr
ove
me
nt
in
the
r
e
s
ult
wa
s
f
ound
a
f
ter
23
it
e
r
a
ti
ons
.
S
im
il
a
r
ly,
F
igur
e
3
r
e
ve
a
ls
s
tage
2
-
GA
s
im
ulation
to
opti
mi
z
e
hype
r
pa
r
a
mete
r
s
.
T
he
a
c
c
ur
a
c
y
obtaine
d
in
thi
s
r
u
n
is
0.
967.
T
hr
e
e
pa
r
a
mete
r
s
s
uc
h
a
s
the
number
of
tr
e
e
(
4)
,
mi
nim
um
lea
ve
s
(
10)
,
a
nd
mi
ni
mum
lea
f
to
s
pit
(
8)
a
r
e
opti
mi
z
e
d
f
or
a
n
a
c
c
ur
a
c
y
is
0.
967
in
the
5
0
th
r
un
a
s
s
hown
in
T
a
ble
2
.
T
his
s
hows
that
GA
im
p
r
ove
s
the
a
c
c
ur
a
c
y
f
r
o
m
s
tage
1
to
s
tage
2.
T
he
de
tail
e
d
DT
is
de
picte
d
in
F
igur
e
4
f
or
p
ictor
ial
r
e
p
r
e
s
e
ntation.
T
he
s
im
ulation
is
c
a
r
r
ied
out
us
ing
P
ython
v
3
.
3
with
8
GB
R
AM
,
with
GA
f
unc
ti
on
a
nd
ML
a
lgor
it
hm
in
W
in
dows
10
ope
r
a
ti
ng
s
ys
tem
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
Ar
ti
f
I
ntell
,
Vol.
14
,
No.
4
,
Augus
t
20
25
:
279
7
-
2804
2802
F
igur
e
2.
S
tate
1
-
GA
s
im
ulation
to
opti
mi
z
e
r
a
ndom
s
tate
a
nd
tes
t
s
ize
,
a
c
c
ur
a
c
y
is
the
objec
ti
ve
f
unc
ti
o
n
F
igur
e
3.
S
tage
2
-
GA
s
im
ulation
to
opti
mi
z
e
hype
r
pa
r
a
mete
r
s
F
igur
e
4.
P
icto
r
ial
r
e
p
r
e
s
e
ntation
of
the
DT
4.
CONC
L
USI
ON
T
his
s
tudy
pr
e
s
e
nts
a
nove
l
a
ppr
oa
c
h
to
opti
mi
z
ing
the
pe
r
f
or
manc
e
of
an
R
F
c
las
s
if
ier
b
y
int
e
gr
a
ti
ng
a
GA
f
or
the
pr
e
diction
o
f
C
KD
us
ing
a
da
tas
e
t
of
1659
pa
ti
e
nts
with
51
pa
r
a
met
e
r
s
.
B
y
s
ys
tema
ti
c
a
ll
y
opti
mi
z
ing
the
r
a
ndom
s
tate
,
tes
t
s
ize
,
a
nd
hype
r
pa
r
a
mete
r
s
in
a
two
-
s
tage
pr
oc
e
s
s
,
th
e
method
e
f
f
e
c
ti
ve
ly
e
nha
nc
e
s
the
a
c
c
ur
a
c
y
of
the
R
F
mode
l.
T
he
opti
mi
z
a
ti
on
pr
oc
e
s
s
,
c
onduc
ted
ove
r
50
s
im
ulation
r
uns
,
de
mons
tr
a
tes
a
s
igni
f
ica
nt
im
pr
ove
ment
in
model
a
c
c
ur
a
c
y,
r
a
nging
f
r
om
0
.
9451
to
0.
973
8,
with
a
maxim
um
incr
e
a
s
e
of
2.
09
%
.
T
he
s
e
f
indi
ngs
high
li
ght
the
c
r
it
ica
l
r
ole
of
opti
mi
z
ing
both
model
pa
r
a
mete
r
s
a
nd
hype
r
pa
r
a
mete
r
s
to
e
nha
nc
e
the
pr
e
dictive
c
a
pa
bil
it
ies
of
ML
models
,
e
s
pe
c
ially
withi
n
the
r
e
a
lm
of
medic
a
l
diagnos
ti
c
s
.
T
he
c
ombi
na
ti
on
of
GA
with
R
F
not
only
e
nha
nc
e
s
model
pe
r
f
or
manc
e
but
a
ls
o
e
s
tablis
he
s
a
s
tr
ong
a
nd
r
e
li
a
ble
f
r
a
mew
or
k
f
or
i
mpr
oving
p
r
e
diction
a
c
c
ur
a
c
y
in
c
li
n
ica
l
s
e
tt
ings
.
F
u
tur
e
s
tudi
e
s
may
c
on
s
ider
a
pplyi
ng
thi
s
opti
mi
z
a
ti
on
s
tr
a
tegy
to
dif
f
e
r
e
nt
dis
e
a
s
e
s
a
nd
da
ta
s
e
ts
to
f
ur
ther
a
s
s
e
s
s
it
s
e
f
f
e
c
ti
ve
ne
s
s
a
nd
ve
r
s
a
ti
li
ty.
AC
KNOWL
E
DGE
M
E
NT
S
Author
s
thank
the
o
r
ga
niza
ti
on
f
o
r
e
nc
our
a
ging
us
to
c
onduc
t
r
e
s
e
a
r
c
h
in
mul
ti
-
dis
c
ipl
inar
y
a
r
e
a
s
.
F
UN
DI
NG
I
NF
ORM
AT
I
ON
T
his
r
e
s
e
a
r
c
h
wa
s
c
a
r
r
ied
out
indepe
nde
ntl
y,
wi
th
no
f
unding
invol
ve
d
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
Ar
ti
f
I
ntell
I
S
S
N:
2252
-
8938
I
ntegr
ati
ng
r
andom
for
e
s
t
and
ge
ne
ti
c
algor
it
hms
f
or
…
(
B
omm
anahall
i
V
e
nk
atagi
r
iyappa
R
aghav
e
n
dr
a
)
2803
AU
T
HO
R
CONT
RI
B
U
T
I
ONS
S
T
AT
E
M
E
N
T
T
his
jour
na
l
us
e
s
the
C
ontr
ibut
o
r
R
oles
T
a
xo
nomy
(
C
R
e
diT
)
to
r
e
c
ognize
indi
vidual
a
uthor
c
ontr
ibut
ions
,
r
e
duc
e
a
utho
r
s
hip
dis
putes
,
a
nd
f
a
c
il
it
a
te
c
oll
a
bor
a
ti
on.
Nam
e
of
Au
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
B
omm
a
na
ha
ll
i
Ve
nka
tagir
iyappa
R
a
gha
ve
ndr
a
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Ana
ndkumar
R
a
mappa
Annige
r
i
✓
✓
✓
✓
✓
✓
✓
J
ogipalya
S
hivana
njappa
S
r
ikanta
mur
thy
✓
✓
✓
✓
✓
✓
✓
Gur
ur
a
j
R
a
gha
ve
ndr
a
r
a
o
S
a
tt
iger
i
✓
✓
✓
✓
✓
✓
C
:
C
onc
e
pt
ua
li
z
a
ti
on
M
:
M
e
th
odol
ogy
So
:
So
f
twa
r
e
Va
:
Va
li
da
ti
on
Fo
:
Fo
r
ma
l
a
na
ly
s
is
I
:
I
nve
s
ti
ga
ti
on
R
:
R
e
s
our
c
e
s
D
:
D
a
ta
C
ur
a
ti
on
O
:
W
r
it
in
g
-
O
r
ig
in
a
l
D
r
a
f
t
E
:
W
r
it
in
g
-
R
e
vi
e
w
&
E
di
ti
ng
Vi
:
Vi
s
ua
li
z
a
ti
on
Su
:
Su
pe
r
vi
s
io
n
P
:
P
r
oj
e
c
t
a
dmi
ni
s
tr
a
ti
on
Fu
:
Fu
ndi
ng a
c
qui
s
it
io
n
CONF
L
I
CT
OF
I
NT
E
RE
S
T
S
T
AT
E
M
E
N
T
T
he
a
uthor
s
de
c
lar
e
that
ther
e
a
r
e
no
c
onf
li
c
ts
of
in
ter
e
s
t
r
e
late
d
to
th
is
s
tudy
DA
T
A
AV
AI
L
A
B
I
L
I
T
Y
T
he
da
ta
that
s
uppor
t
thi
s
s
tudy
a
r
e
op
e
nly
a
va
il
a
ble
in
Ka
ggle
da
ta
s
e
t
li
nk
a
t
htt
ps
:/
/www
.
ka
ggle.
c
om/
da
tas
e
ts
/r
a
biee
lkhar
oua
/c
hr
onic
-
kidney
-
dis
e
a
s
e
-
da
ta
s
e
t
-
a
na
lys
i
s
.
RE
F
E
RE
NC
E
S
[
1]
M
. A
. I
s
la
m, M
. Z
. H
. M
a
ju
mde
r
, a
nd M
. A
. H
us
s
e
in
, “
C
hr
oni
c
ki
dne
y di
s
e
a
s
e
pr
e
di
c
ti
on ba
s
e
d on ma
c
hi
ne
l
e
a
r
ni
ng a
lg
or
it
hms
,”
J
our
nal
of
P
at
hol
ogy
I
nf
or
m
at
ic
s
, vol
. 14, 2023, doi:
10.1016/j
.j
pi
.2023.100189.
[
2]
N
.
L
e
i
e
t
al
.
,
“
M
a
c
hi
ne
le
a
r
ni
ng
a
lg
or
it
hms
’
a
c
c
ur
a
c
y
in
pr
e
d
ic
ti
ng
ki
dne
y
di
s
e
a
s
e
pr
ogr
e
s
s
io
n:
a
s
ys
te
ma
ti
c
r
e
vi
e
w
a
nd
m
e
ta
-
a
na
ly
s
is
,”
B
M
C
M
e
di
c
al
I
nf
or
m
at
ic
s
and
D
e
c
is
io
n M
ak
in
g
, vol
.
22, no. 1, De
c
. 2022, doi:
10.1186/s
12911
-
022
-
01951
-
1.
[
3]
D
.
A
.
D
e
ba
l
a
nd
T
.
M
.
S
it
ot
e
,
“
C
hr
oni
c
ki
dne
y
di
s
e
a
s
e
pr
e
d
ic
ti
on
us
in
g
ma
c
hi
ne
l
e
a
r
ni
ng
te
c
hni
que
s
,”
J
ou
r
nal
of
B
ig
D
at
a
,
vol
. 9, no. 1, Nov. 2022
, doi
:
10.1186/s
40537
-
022
-
00657
-
5.
[
4]
S
.
P
a
l,
“
C
hr
oni
c
k
id
ne
y
di
s
e
a
s
e
pr
e
di
c
ti
on
us
in
g
ma
c
hi
ne
le
a
r
ni
ng
te
c
hni
que
s
,”
B
io
m
e
di
c
al
M
at
e
r
ia
ls
&
D
e
v
ic
e
s
,
vol
.
1,
no
.
1,
p
p. 534
–
540, M
a
r
. 2023, doi:
10.1007/s
44174
-
022
-
00027
-
y.
[
5]
S
. J
. G
il
be
r
t
a
nd D
. E
. W
e
in
e
r
,
N
at
io
nal
k
id
ne
y
f
oundati
on’
s
pr
ime
r
on k
id
ne
y
di
s
e
as
e
s
,
C
hi
n
a
:
E
ls
e
vi
e
r
, 2017.
[
6]
L
.
J
.
L
o
e
t
al
.
,
“
D
ia
ly
s
is
-
r
e
qui
r
in
g
a
c
ut
e
r
e
na
l
f
a
il
ur
e
in
c
r
e
a
s
e
s
th
e
r
is
k
of
p
r
ogr
e
s
s
iv
e
c
hr
oni
c
ki
dne
y
di
s
e
a
s
e
,”
K
id
ne
y
I
nt
e
r
nat
io
nal
, vol
. 76, no. 8, pp. 893
–
899, Oc
t.
2009, doi:
10.1038/ki
.2009.289.
[
7]
J
.
Z
ha
o
e
t
al
.
,
“
A
n
e
a
r
ly
pr
e
di
c
ti
on
mode
l
f
or
c
hr
oni
c
ki
dne
y
di
s
e
a
s
e
,”
Sc
i
e
nt
if
ic
R
e
por
ts
,
vol
.
12,
no.
1,
F
e
b.
20
22,
doi
:
10.1
038/
s
41598
-
022
-
06665
-
y.
[
8]
H
.
K
ha
li
d,
A
.
K
h
a
n,
M
.
Z
.
K
ha
n,
G
.
M
e
hmood, a
nd
M
.
S
.
Q
ur
e
s
hi
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
hybr
id
mode
l
f
or
th
e
pr
e
di
c
ti
on
of
c
hr
oni
c
ki
dne
y di
s
e
a
s
e
,”
C
om
put
at
io
nal
I
nt
e
ll
ig
e
nc
e
and N
e
ur
os
c
ie
nc
e
,
vol
. 2023, no. 1, J
a
n. 2023, doi:
10.1155/
2023/9266889.
[
9]
D
.
S
a
if
,
A
.
M
.
S
a
r
ha
n,
a
nd
N
.
M
.
E
ls
he
nna
w
y,
“
D
e
e
p
-
ki
dne
y:
a
n
e
f
f
e
c
ti
ve
de
e
p
le
a
r
ni
ng
f
r
a
me
w
or
k
f
or
c
hr
oni
c
ki
dne
y
di
s
e
a
s
e
pr
e
di
c
ti
on,”
H
e
al
th
I
nf
or
m
at
io
n Sc
ie
nc
e
and Sy
s
te
m
s
, vol
. 12, n
o. 1, De
c
. 2023, doi:
10.1007/s
13755
-
023
-
00261
-
8.
[1
0]
M
.
M
.
R
a
hma
n,
M
.
A
l
-
A
mi
n,
a
nd
J
.
H
o
s
s
a
in
,
“
M
a
c
hi
n
e
le
a
r
n
in
g
mode
ls
f
or
c
hr
oni
c
ki
dne
y
di
s
e
a
s
e
di
a
gnos
i
s
a
nd
pr
e
di
c
ti
on,”
B
io
m
e
di
c
al
Si
gnal
P
r
oc
e
s
s
in
g and C
ont
r
ol
, vol
. 87,
J
a
n. 2024,
doi
:
10.1016/j
.bs
pc
.2023.105368.
[
11]
E
.
D
r
it
s
a
s
a
nd
M
.
T
r
ig
ka
,
“
M
a
c
hi
ne
le
a
r
ni
ng
t
e
c
hni
que
s
f
or
c
hr
oni
c
ki
dne
y
di
s
e
a
s
e
r
is
k
pr
e
di
c
ti
on,”
B
ig
D
at
a
and
C
ogni
t
iv
e
C
om
put
in
g
, vol
. 6, no. 3, S
e
p. 2022, doi:
10.3390/bdcc
6030098.
[
12]
D
. K
. E
. L
im
e
t
al
.
, “
P
r
e
di
c
ti
on mode
ls
us
e
d i
n t
he
pr
ogr
e
s
s
io
n
o
f
c
hr
oni
c
ki
dne
y di
s
e
a
s
e
:
a
s
c
opi
ng r
e
vi
e
w
,”
P
L
O
S O
N
E
, vol
. 17,
no. 7, J
ul
. 2022, doi:
10.1371/j
our
na
l.
pone
.0271619.
[
13]
J
.
A
oki
e
t
al
.
,
“
C
K
D
pr
ogr
e
s
s
io
n
pr
e
di
c
ti
on
in
a
di
ve
r
s
e
U
S
popula
ti
on:
a
ma
c
hi
ne
-
le
a
r
ni
ng
mode
l,
”
K
id
ne
y
M
e
di
c
in
e
,
vol
.
5,
no. 9, S
e
p. 2023, doi:
10.1016/j
.xkme
.2023.100692.
[
14]
M
.
B
in
s
a
w
a
d,
“
E
nha
nc
in
g
ki
dne
y
di
s
e
a
s
e
pr
e
di
c
ti
on
w
it
h
op
ti
mi
z
e
d
f
or
e
s
t
a
nd
E
C
G
s
ig
na
l
s
da
ta
,
”
H
e
li
y
on
,
vol
.
10,
no.
10,
M
a
y 2024, doi:
10.1016/j
.he
li
yon.2024.e
30792.
[
15]
K
.
H
e
ma
,
K
.
M
e
e
na
,
a
nd
R
.
P
a
ndi
a
n,
“
A
na
ly
z
e
th
e
im
pa
c
t
of
f
e
a
tu
r
e
s
e
le
c
ti
on
te
c
hni
que
s
in
th
e
e
a
r
ly
pr
e
di
c
ti
on
of
C
K
D
,”
I
nt
e
r
nat
io
nal
J
our
nal
of
C
ogni
ti
v
e
C
om
put
in
g i
n E
ngi
ne
e
r
in
g
,
vol
. 5, pp. 66
–
77, 2024, doi:
10.1016/j
.i
jc
c
e
.2023.12.002.
[
16]
K
. T
a
kka
va
t
a
k
a
r
n, W
. O
h, E
. C
he
ng,
G
. N
. N
a
dka
r
ni
, a
nd
L
. C
ha
n, “
M
a
c
hi
ne
l
e
a
r
ni
ng mode
ls
t
o pr
e
di
c
t
e
nd
-
s
ta
ge
ki
dne
y di
s
e
a
s
e
in
c
hr
oni
c
ki
dne
y di
s
e
a
s
e
s
ta
ge
4,
”
B
M
C
N
e
phr
ol
ogy
, vol
. 24, n
o. 1, De
c
. 2023, doi:
10.1186/s
12882
-
023
-
03424
-
7.
[
17]
F
.
S
a
nma
r
c
hi
,
C
.
F
a
nc
oni
,
D
.
G
ol
in
e
ll
i,
D
.
G
or
i,
T
.
H
e
r
na
nde
z
-
B
ous
s
a
r
d,
a
nd
A
.
C
a
podi
c
i,
“
P
r
e
di
c
t,
di
a
gnos
e
,
a
nd
tr
e
a
t
c
hr
oni
c
ki
dne
y
di
s
e
a
s
e
w
it
h
ma
c
hi
ne
le
a
r
ni
ng:
a
s
ys
te
ma
ti
c
li
te
r
a
tu
r
e
r
e
vi
e
w
,”
J
our
nal
of
N
e
phr
ol
ogy
,
vol
.
36,
no.
4,
pp.
1101
–
1117,
F
e
b. 2023, doi:
10.1007/s
406
20
-
023
-
01573
-
4.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
Ar
ti
f
I
ntell
,
Vol.
14
,
No.
4
,
Augus
t
20
25
:
279
7
-
2804
2804
[
18]
Y
.
Z
hu,
D
.
B
i,
M
.
S
a
unde
r
s
,
a
nd
Y
.
J
i,
“
P
r
e
di
c
ti
on
of
c
hr
oni
c
ki
dne
y
di
s
e
a
s
e
pr
ogr
e
s
s
io
n
us
in
g
r
e
c
ur
r
e
nt
ne
ur
a
l
ne
twor
k
a
nd
e
le
c
tr
oni
c
he
a
lt
h r
e
c
or
ds
,”
Sc
ie
nt
if
ic
R
e
po
r
ts
, vol
. 13, no. 1, De
c
. 2023, doi:
10.1038/s
41598
-
023
-
49271
-
2.
[
19
]
S
.
K
.
G
hos
h
a
nd
A
.
H
.
K
ha
ndoke
r
,
“
I
nve
s
ti
ga
ti
on
on
e
xpl
a
in
a
bl
e
ma
c
hi
ne
le
a
r
ni
ng
mode
ls
to
pr
e
di
c
t
c
hr
oni
c
ki
dne
y
di
s
e
a
s
e
s
,”
Sc
ie
nt
if
ic
R
e
por
ts
, vol
. 14, no. 1, F
e
b. 2024, doi:
10.1038/s
4159
8
-
024
-
54375
-
4.
[
20]
N
.
A
lt
ur
ki
e
t
al
.
,
“
I
mp
r
ovi
ng
pr
e
di
c
ti
on
of
c
hr
on
ic
ki
dne
y
di
s
e
a
s
e
us
in
g
K
N
N
im
put
e
d
S
M
O
T
E
f
e
a
tu
r
e
s
a
nd
T
r
io
N
e
t
mode
l,
”
C
om
put
e
r
M
ode
li
ng i
n E
ngi
ne
e
r
in
g & Sc
ie
nc
e
s
, vol
. 139, no. 3,
pp. 3513
–
3534, 2024, doi:
10.32604/cme
s
.2023.045868.
[
21]
S
.
R
a
je
a
s
hw
a
r
i
a
nd
K
.
A
r
une
s
h,
“
C
hr
oni
c
di
s
e
a
s
e
pr
e
di
c
ti
o
n
w
i
th
de
e
p
c
onvolut
io
n
ba
s
e
d
modi
f
ie
d
e
xt
r
e
me
-
r
a
ndom
f
or
e
s
t
c
la
s
s
if
ie
r
,”
B
io
m
e
di
c
al
Si
gnal
P
r
oc
e
s
s
in
g and C
ont
r
ol
, vol
. 87,
J
a
n. 2024, doi:
10.1016/j
.bs
pc
.2023.105425.
[
22]
L
. B
r
e
im
a
n, “
R
a
ndom f
or
e
s
ts
,”
M
ac
hi
ne
L
e
a
r
ni
ng
, vol
. 45, no. 1, pp. 5
–
32, Oc
t.
2001, d
oi
:
10.1023/A:1010933
404324.
[
23]
J
.
H
.
H
ol
la
nd,
A
dapt
at
io
n
in
na
tu
r
al
and
ar
ti
fi
c
ia
l
s
y
s
te
m
s
.
U
ni
te
d
S
ta
te
s
:
M
I
T
P
r
e
s
s
,
C
a
mbr
id
ge
,
M
A
,
1992
,
doi
:
10.7551/m
it
pr
e
s
s
/1
090.001.0001.
[
24]
C
. C
he
n, A
. L
ia
w
, a
nd L
. B
r
e
im
a
n,
“
U
s
in
g r
a
ndom f
or
e
s
t
to
l
e
a
r
n i
mba
la
n
c
e
d da
ta
,
”
U
C
B
e
r
k
e
le
y
St
at
is
ti
c
s
D
e
par
tm
e
nt
,
pp. 1
-
12,
2004.
[
25]
E
.
A
kkur
a
nd
A
.
C
.
Ö
z
tü
r
k,
“
C
hr
oni
c
ki
dne
y
di
s
e
a
s
e
pr
e
di
c
ti
on
w
it
h
s
ta
c
ke
d
e
n
s
e
mbl
e
-
ba
s
e
d
mode
l,
”
A
lp
ha
J
our
nal
of
E
ngi
ne
e
r
in
g and A
ppl
ie
d Sc
ie
nc
e
s
, vol
. 1, no. 1, pp. 50
–
61, 202
3
.
B
I
OG
RA
P
HI
E
S
OF
AU
T
HO
RS
D
r.
Bo
m
m
a
n
a
ha
l
l
i
V
enk
a
ta
g
i
r
i
y
a
ppa
R
a
g
ha
v
endra
h
o
l
d
s
a
b
ac
h
el
o
r's
d
eg
ree
in
mec
h
an
i
cal
e
n
g
i
n
eer
i
n
g
an
d
a
Ph
.
D
.
i
n
mec
h
an
i
cal
e
n
g
i
n
eer
i
n
g
s
c
i
en
ce
s
fro
m
V
i
s
v
e
s
v
ara
y
a
T
ech
n
o
l
o
g
i
ca
l
U
n
i
v
ers
i
t
y
,
Bel
a
g
av
i
,
In
d
i
a.
Cu
rre
n
t
l
y
,
h
e
s
erv
e
s
as
a
n
as
s
o
c
i
at
e
p
ro
f
es
s
o
r
i
n
t
h
e
D
ep
ar
t
men
t
o
f
Mech
an
i
cal
E
n
g
i
n
eer
i
n
g
at
J
SS
A
cad
em
y
o
f
T
ech
n
i
cal
E
d
u
ca
t
i
o
n
,
Ben
g
al
u
ru
.
H
i
s
res
earc
h
i
n
t
eres
t
s
i
n
cl
u
d
e
s
ch
e
d
u
l
i
n
g
,
t
h
e
s
t
u
d
y
o
f
mach
i
n
i
n
g
p
arame
t
ers
,
o
p
t
i
m
i
zat
i
o
n
,
an
d
met
a
h
eu
r
i
s
t
i
c
a
l
g
o
ri
t
h
m
s
.
H
e
h
as
p
u
b
l
i
s
h
ed
mo
re
t
h
an
2
2
p
ap
er
s
i
n
p
eer
-
rev
i
ew
e
d
j
o
u
r
n
al
s
an
d
i
s
a
memb
er
o
f
t
h
e
In
s
t
i
t
u
t
i
o
n
o
f
E
n
g
i
n
ee
r
s
.
H
e
can
b
e
co
n
t
act
e
d
at
e
mai
l
:
b
v
r
j
s
s
@
g
mai
l
.
co
m
.
D
r.
A
na
ndkum
a
r
R
a
m
a
ppa
A
nn
i
g
er
i
h
o
l
d
s
a
Ph
.
D
.
i
n
mech
a
n
i
ca
l
en
g
i
n
eeri
n
g
fro
m
IIT
Ma
d
ras
,
In
d
i
a,
a
n
d
h
as
a
s
t
r
o
n
g
aca
d
emi
c
f
o
u
n
d
at
i
o
n
i
n
t
h
e
fi
e
l
d
.
H
e
i
s
cu
rre
n
t
l
y
s
erv
i
n
g
as
a
Pro
fes
s
o
r
i
n
t
h
e
D
e
p
art
me
n
t
o
f
Mech
a
n
i
cal
E
n
g
i
n
eer
i
n
g
at
J
SS
A
cad
em
y
o
f
T
ech
n
i
ca
l
E
d
u
cat
i
o
n
,
Ben
g
al
u
ru
.
H
i
s
res
earc
h
fo
cu
s
es
o
n
o
p
t
i
mi
za
t
i
o
n
an
d
s
mart
s
t
ru
c
t
u
re
,
an
d
h
e
h
as
co
n
t
ri
b
u
t
ed
s
i
g
n
i
fi
c
an
t
l
y
t
o
t
h
e
fi
el
d
t
h
r
o
u
g
h
t
h
e
p
u
b
l
i
cat
i
o
n
o
f
o
v
er
2
5
p
ap
er
s
i
n
p
eer
-
rev
i
e
w
ed
j
o
u
r
n
al
s
.
H
e
can
b
e
c
o
n
t
act
e
d
at
e
mai
l
:
a
n
an
d
a
n
n
i
g
er
i
@
g
mai
l
.
co
m
.
D
r.
J
o
g
i
p
a
l
y
a
Sh
i
v
a
na
nj
a
ppa
Sr
i
ka
nt
a
m
urth
y
h
o
l
d
s
a
Bach
e
l
o
r'
s
d
e
g
ree
i
n
mech
an
i
cal
e
n
g
i
n
eer
i
n
g
a
n
d
a
Ph
.
D
.
i
n
t
h
e
s
ame
f
i
el
d
.
Cu
rren
t
l
y
,
h
e
s
erv
e
s
as
an
a
s
s
i
s
t
an
t
p
ro
fe
s
s
o
r
i
n
t
h
e
D
ep
ar
t
men
t
o
f
Mech
a
n
i
ca
l
E
n
g
i
n
ee
ri
n
g
at
J
SS
A
cad
em
y
o
f
T
ech
n
i
c
al
E
d
u
cat
i
o
n
,
Ben
g
a
l
u
r
u
.
H
i
s
res
earc
h
i
n
t
eres
t
s
i
n
c
l
u
d
e
o
p
t
i
mi
zat
i
o
n
an
d
f
i
n
i
t
e
e
l
emen
t
an
al
y
s
i
s
.
H
e
i
s
al
s
o
a
memb
er
o
f
t
h
e
In
s
t
i
t
u
t
i
o
n
o
f
E
n
g
i
n
e
ers
.
H
e
can
b
e
co
n
t
act
e
d
at
e
mai
l
:
j
s
s
r
i
k
an
t
amu
r
t
h
y
@
j
s
s
at
e
b
.
ac.
i
n
.
Gurura
j
R
a
g
ha
v
endra
r
a
o
Sa
tt
i
g
er
i
i
s
g
rad
u
a
t
e
d
i
n
mech
an
i
cal
e
n
g
i
n
eer
i
n
g
,
p
res
e
n
t
l
y
w
o
rk
i
n
g
as
a
s
s
o
ci
a
t
e
p
ro
fe
s
s
o
r
i
n
t
h
e
D
ep
art
men
t
o
f
Mec
h
an
i
cal
E
n
g
i
n
eeri
n
g
,
K
L
S
V
i
s
h
w
an
a
t
h
ra
o
D
es
h
p
an
d
e
In
s
t
i
t
u
t
e
o
f
T
ech
n
o
l
o
g
y
,
H
al
i
y
al
,
K
arn
a
t
ak
a.
H
i
s
res
earch
i
n
t
eres
t
i
s
j
et
i
m
p
i
n
g
em
en
t
.
H
e
i
s
a
memb
er
i
n
I
n
s
t
i
t
u
t
e
o
f
E
n
g
i
n
eers
.
H
e
ca
n
b
e
c
o
n
t
act
e
d
at
e
ma
i
l
:
g
rs
@
k
l
s
v
d
i
t
.
ed
u
.
i
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.