I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
V
ol
. 14, No. 5, O
c
to
be
r
2025
, pp.
3926
~
3933
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
14
.i
5
.pp
3926
-
3933
3926
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
H
yb
r
i
d
i
z
e
d
d
e
e
p
l
e
ar
n
i
n
g
m
od
e
l
w
i
t
h
n
ove
l
r
e
c
om
m
e
n
d
e
r
f
or
p
r
e
d
i
c
t
i
n
g c
r
i
t
i
c
al
i
t
y st
at
e
of
p
at
i
e
n
t
u
si
n
g
M
IM
IC
-
IV
d
at
ase
t
S
ar
ik
a K
h
op
e
1
, D
e
e
p
al
i
K
ot
am
b
k
ar
2
,
R
am
a V
as
an
t
h
a A
d
ir
aj
u
3
, S
m
it
a S
u
h
as
B
at
t
al
w
ar
4
1
D
e
pa
r
t
m
e
nt
of
E
l
e
c
t
r
oni
c
s
a
nd
T
e
l
e
c
om
m
uni
c
a
t
i
on, G
H
R
a
i
s
oni
C
ol
l
e
ge
of
E
ngi
ne
e
r
i
ng a
nd M
a
na
ge
m
e
nt
, P
une
, I
ndi
a
2
D
e
pa
r
t
m
e
nt
of
E
l
e
c
t
r
oni
c
s
E
ngi
ne
e
r
i
ng,
R
a
m
de
oba
ba
U
ni
ve
r
s
i
t
y, N
a
gpur
,
I
ndi
a
3
D
e
pa
r
t
m
e
nt
of
E
l
e
c
t
r
oni
c
s
a
nd C
om
m
uni
c
a
t
i
on E
ngi
ne
e
r
i
ng, A
di
t
ya
U
ni
ve
r
s
i
t
y,
K
a
ki
na
da
, I
ndi
a
4
D
e
pa
r
t
m
e
nt
of
A
r
t
i
f
i
c
i
a
l
I
nt
e
l
l
i
ge
nc
e
a
nd
M
a
c
hi
ne
L
e
a
r
ni
ng
, G
H
R
a
i
s
oni
C
ol
l
e
ge
of
E
ngi
ne
e
r
i
ng a
nd M
a
na
ge
m
e
nt
, P
une
, I
ndi
a
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
D
e
c
23, 2024
R
e
vi
s
e
d
J
ul
2, 2025
A
c
c
e
pt
e
d
A
ug 6, 2025
The
contribution
of
machine
learning
towards
prediction
of
critical
s
tate
of
patient
is
the
prime
focus
of
the
current
study.
The
review
of
current
approaches
of
machine
learning
has
been
witness
ed
with
various
shortcomings.
Hence,
the
proposed
study
adopts
medical
informatio
n
mart
for
intensive
care
(MIMIC
-
IV)
dataset
in
order
to
develop
a
novel
ana
lytical
model
that
can
predict
the
criticality
state
of
patient
in
their
next
vis
it.
The
model
has
been
designed
by
hybridizing
convolut
ion
neural
network
(CNN)
and
long
short
-
term
memory
(LSTM)
which
takes
the
discrete
in
put
of
hospital
and
individual
patient
information
in
each
visit.
The
concat
enated
feature
is
then
subjected
to
a
newly
introduced
recommender
module
which
offers
implicit
feedback
by
assigning
a
ranking
score.
The
final
pre
dictive
outcome
of
study
offers
criticality
rank.
The
study
model
is
bench
marked
with
existing
machine
learning
approaches
to
find
54%
of
inc
reased
accuracy and 7
0% of reduced
processi
ng tim
e.
K
e
y
w
o
r
d
s
:
C
onvolut
io
n ne
ur
a
l
ne
twor
k
C
r
it
ic
a
li
ty
s
ta
te
L
ong s
hor
t
-
te
r
m
m
e
m
or
y
M
a
c
hi
ne
l
e
a
r
ni
ng
M
I
M
I
C
-
IV
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
S
a
r
ik
a
K
hope
D
e
pa
r
tm
e
nt
of
E
le
c
tr
oni
c
s
a
nd T
e
le
c
om
m
uni
c
a
ti
on, G
H
R
a
is
oni
C
ol
le
ge
of
E
ngi
ne
e
r
in
g a
nd M
a
na
ge
m
e
nt
N
e
w
G
a
t
N
o 1200, Domkhe
l
R
oa
d, W
a
ghol
i,
P
une
-
412207,
M
a
ha
r
a
s
ht
r
a
S
ta
te
, I
ndi
a
E
m
a
il
:
s
a
r
ik
a
.khope
@
gm
a
il
.c
om
1.
I
N
T
R
O
D
U
C
T
I
O
N
T
he
r
e
i
s
a
s
i
gni
f
ic
a
nt
le
v
e
l
of
c
ont
r
ib
ut
io
n
m
a
de
by
m
a
c
hi
n
e
l
e
a
r
ni
ng
to
w
a
r
ds
th
e
he
a
lt
hc
a
r
e
s
y
s
t
e
m
th
a
t
h
a
s
s
u
c
c
e
s
s
f
ul
l
y
tr
a
n
s
f
or
m
e
d
h
e
a
lt
h
c
a
r
e
m
a
na
g
e
m
e
n
t,
m
e
di
c
a
l
r
e
s
e
a
r
c
h,
a
nd
pa
ti
e
nt
c
a
r
e
[
1]
.
T
h
e
c
or
e
te
c
hn
ol
ogi
c
a
l
a
dv
a
nc
e
m
e
nt
of
m
a
c
hi
n
e
le
a
r
ni
ng
c
a
n
be
r
e
a
li
z
e
d
by
it
s
hi
ghl
y
im
pr
ov
e
d
m
e
di
c
a
l
im
a
gi
ng
a
nd
di
a
gno
s
ti
c
s
th
a
t
c
ont
r
ib
u
te
s
to
w
a
r
d
s
de
t
e
c
ti
o
n
of
va
r
i
ous
di
s
e
a
s
e
s
in
it
s
e
a
r
ly
s
t
a
ge
s
.
F
ur
th
e
r
,
h
os
pi
ta
l
ope
r
a
ti
ons
a
r
e
s
ub
s
ta
nt
i
a
ll
y
opt
im
i
z
e
d
u
s
in
g
m
a
c
hi
ne
le
a
r
n
in
g
r
ig
ht
f
r
om
in
ve
nt
or
y
c
ont
r
ol
a
nd
s
t
a
f
f
s
c
h
e
dul
i
ng
to
b
e
d
m
a
na
ge
m
e
nt
.
A
t
th
e
s
a
m
e
t
im
e
,
va
r
io
u
s
ty
pe
s
of
p
a
ti
e
n
t
da
t
a
,
e
.g.
l
if
e
s
ty
le
f
a
c
t
or
s
,
ge
n
e
ti
c
in
f
or
m
a
ti
on
, c
li
ni
c
a
l
hi
s
to
r
y,
c
a
n
b
e
us
e
d by
m
a
c
hi
ne
le
r
a
ni
ng
t
ow
a
r
d
s
pr
e
di
c
ti
ng
pr
ob
a
bi
li
ty
of
va
r
io
u
s
r
a
ng
e
s
of
di
s
e
a
s
e
s
[
2]
.
T
he
pr
e
di
c
ti
on
of
pr
ogr
e
s
s
i
on
of
di
s
e
a
s
e
s
c
a
n
b
e
no
w
po
s
s
ib
le
by
in
ve
s
ti
g
a
ti
ng
th
e
p
a
tt
e
r
n
s
of
da
ta
o
ve
r
ti
m
e
us
i
ng
m
a
c
hi
n
e
le
a
r
ni
ng.
T
hi
s
a
ls
o
a
s
s
i
s
ts
i
n
of
f
e
r
in
g
pe
r
s
ona
l
iz
e
d
tr
e
a
t
m
e
nt
f
or
v
a
r
io
u
s
pa
ti
e
nt
s
uf
f
e
r
in
g
f
r
om
c
hr
oni
c
di
s
e
a
s
e
s
e
.
g.,
a
s
t
hm
a
,
hyp
e
r
te
n
s
io
n
,
a
nd
di
a
be
t
e
s
.
C
o
ns
i
de
r
in
g
r
e
a
l
-
ti
m
e
da
t
a
f
r
om
c
li
ni
c
a
l
s
e
tt
in
g
(
e
.
g.,
e
l
e
c
tr
o
ni
c
he
a
lt
h
r
e
c
or
d
s
)
,
m
a
c
h
in
e
le
a
r
n
in
g
c
a
n
a
s
s
e
s
s
th
e
r
i
s
k
pe
r
t
a
in
in
g
t
o
s
p
e
c
if
i
c
pa
ti
e
n
t,
of
f
e
r
dr
u
g
in
t
e
r
a
c
t
io
n,
a
nd
g
e
ne
r
a
te
a
le
r
t
s
y
s
te
m
f
or
pot
e
nt
ia
l
is
s
ue
s
.
H
ow
e
ve
r
,
m
or
ta
li
t
y
pr
e
di
c
ti
o
n
is
s
ti
ll
o
ne
of
t
he
c
ha
l
le
ngi
ng t
op
ic
w
it
hi
n m
a
c
h
in
e
l
e
a
r
n
in
g i
n
he
a
l
th
c
a
r
e
s
e
c
to
r
[
3]
.
T
he
pr
im
e
r
e
a
s
on b
e
hi
n
d t
hi
s
is
a
bs
e
nc
e
of
h
ig
h
-
qu
a
li
ty
l
ongi
tu
di
na
l
da
t
a
, n
oi
s
y
da
t
a
, i
m
ba
l
a
n
c
e
d d
a
ta
, a
nd mi
s
s
i
ng d
a
ta
.
A
pa
r
t
f
r
om
th
is
,
th
e
in
he
r
it
a
nc
e
of
bi
a
s
f
r
om
th
e
m
e
di
c
a
l
da
ta
is
qui
te
pos
s
ib
le
in
m
a
c
hi
ne
le
a
r
ni
ng
w
hi
le
th
e
y
unde
r
go
tr
a
in
in
g
ope
r
a
ti
on.
H
e
nc
e
,
s
u
c
h
tr
a
in
e
d
m
ode
l
of
f
e
r
s
m
or
e
a
m
pl
if
ic
a
ti
on
of
bi
a
s
e
s
th
a
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
H
y
br
id
iz
e
d de
e
p l
e
ar
ni
ng m
od
e
l
w
it
h nov
e
l
r
e
c
om
m
e
nd
e
r
f
or
p
r
e
di
c
ti
ng c
r
it
ic
al
it
y
s
ta
te
…
(
Sar
ik
a
K
hope
)
3927
f
in
a
ll
y
le
a
ds
to
out
li
e
r
s
.
A
t
th
e
s
a
m
e
ti
m
e
,
in
tr
oduc
ti
on
of
bi
a
s
ne
s
s
is
a
l
s
o
f
e
a
s
ib
le
dur
in
g
s
e
le
c
ti
on
of
f
e
a
tu
r
e
in
or
de
r
to
c
a
r
r
y
out
pr
e
di
c
ti
ve
a
na
ly
s
is
.
A
not
he
r
bi
gge
r
c
h
a
ll
e
nge
a
s
s
oc
ia
te
d
w
it
h
m
a
c
hi
ne
le
a
r
ni
ng
is
na
r
r
ow
e
d
s
c
ope
of
ge
ne
r
a
li
z
a
ti
on,
w
hi
c
h
m
e
a
ns
m
ode
l
de
s
ig
ne
d
us
in
g
one
s
c
e
na
r
io
of
pa
ti
e
nt
a
nd
di
s
e
a
s
e
m
a
y
not
be
a
ppl
ic
a
bl
e
w
he
n
s
om
e
of
th
e
pot
e
nt
ia
l
d
e
pe
nd
e
nc
ie
s
c
ha
nge
s
.
B
e
c
a
u
s
e
of
a
ll
th
e
s
e
is
s
ue
s
,
de
ve
lo
pi
ng
a
r
e
li
a
bl
e
m
a
c
hi
ne
le
a
r
ni
ng
m
ode
l
f
or
m
or
ta
li
ty
pr
e
di
c
ti
on,
e
s
pe
c
ia
ll
y
f
or
pa
ti
e
nt
a
dm
it
te
d
in
in
te
ns
iv
e
c
a
r
e
uni
t
(
I
C
U
)
is
s
ti
ll
a
pot
e
nt
ia
l
c
h
a
ll
e
nge
[
4]
–
[
6]
.
T
hi
s
r
e
duc
e
s
th
e
r
e
li
a
bi
li
ty
a
tt
r
ib
ut
e
ove
r
va
r
io
us
s
e
tt
in
gs
of
he
a
lt
hc
a
r
e
uni
t
s
.
I
n
or
de
r
to
a
ddr
e
s
s
a
ll
t
he
s
e
im
pe
ndi
ng
is
s
ue
s
,
th
e
r
e
i
s
s
ti
ll
a
be
tt
e
r
c
ha
nc
e
to
w
or
k
on
it
.
O
ne
w
a
y
to
do
s
o
is
by
a
dopt
in
g
a
hi
ghl
y
c
om
pr
e
he
ns
iv
e
a
nd
e
nr
ic
he
d
m
e
di
c
a
l
da
t
a
s
e
t
th
a
t
is
uni
ve
r
s
a
ll
y
a
c
c
e
pt
e
d
a
nd
h
a
ve
di
ve
r
s
if
ie
d
in
f
or
m
a
ti
on
th
a
t
c
a
n
a
s
s
is
t
s
th
e
r
e
s
e
a
r
c
he
r
to
de
ve
lo
p
a
m
ode
l
a
nd
w
or
k
on
f
la
w
s
on
m
a
c
hi
ne
l
e
a
r
ni
ng
m
ode
ls
.
T
hi
s
r
e
s
e
a
r
c
h
w
or
k
u
s
e
s
m
e
di
c
a
l
in
f
or
m
a
ti
on
m
a
r
t
f
or
in
te
ns
iv
e
c
a
r
e
(
M
I
M
I
C
)
da
ta
s
e
t
of
ve
r
s
io
n
la
te
s
t
ve
r
s
io
n
4
s
ig
ni
f
ie
d
a
s
M
I
M
I
C
-
I
V
da
ta
s
e
t
[
7]
.
T
hi
s
be
nc
hm
a
r
ke
d
da
ta
s
e
t
c
on
s
is
ts
of
r
e
a
l
-
w
or
ld
da
ta
f
r
om
ove
r
4
0,000
I
C
U
a
dm
it
te
d
pa
ti
e
nt
s
c
ha
r
e
c
te
r
iz
e
d
by
di
ve
r
s
e
c
li
ni
c
a
l
da
ta
, t
e
m
por
a
l
da
ta
, a
nd mul
ti
m
oda
l
da
ta
. A
p
a
r
t
f
r
om
t
hi
s
, M
I
M
I
C
-
I
V
i
s
ope
n a
c
c
e
s
s
da
ta
t
ha
t
e
nc
our
a
ge
s
c
ol
la
bor
a
ti
on
w
it
h
hi
ghe
r
s
uppor
ta
bi
li
ty
f
or
r
e
s
e
a
r
c
h
w
or
k
in
m
a
c
hi
ne
le
a
r
ni
ng
id
e
a
l
f
or
m
ode
l
de
ve
lo
pm
e
nt
a
nd
e
v
a
lu
a
ti
on
of
m
ode
l
to
w
a
r
ds
s
tr
a
ti
f
ic
a
ti
on
a
n
d
pr
e
di
c
ti
on
of
r
is
j,
e
a
r
ly
de
te
c
ti
on
of
c
r
it
ic
a
l
c
ondi
ti
on,
a
nd
m
or
ta
li
ty
pr
e
di
c
ti
on.
A
pa
r
t
f
r
om
th
is
,
it
a
ls
o
of
f
e
r
s
c
li
ni
c
a
l
te
xt
m
in
in
g
th
a
t
c
ont
r
ib
ut
e
s
to
w
a
r
ds
e
nha
nc
e
d
di
a
gno
s
is
a
nd
pr
ognos
i
s
.
T
he
in
s
ig
ht
e
xt
r
a
c
te
d
f
r
om
M
I
M
I
C
-
I
V
da
ta
a
s
s
is
t
in
id
e
nt
if
yi
ng
f
a
c
to
r
s
th
a
t
ha
s
pot
e
nt
ia
l
im
pa
c
t
f
or
r
e
a
dm
is
s
io
n
to
ho
s
pi
ta
l
a
n
d
c
a
pa
c
it
y
of
I
C
U
.
I
t
a
l
s
o
a
s
s
i
s
ts
in
e
s
ti
m
a
ti
ng
ove
r
a
ll
c
os
t
of
he
a
lt
hc
a
r
e
uni
t
a
s
s
is
ti
ng t
o s
ha
p
e
pol
ic
y de
c
is
io
n
f
or
be
tt
e
r
c
os
t
-
e
f
f
e
c
ti
ve
c
a
r
e
.
I
n
or
de
r
to
a
s
c
e
r
ta
in
th
e
im
pl
ic
a
ti
ons
of
m
a
c
hi
ne
le
a
r
ni
ng
a
ppr
oa
c
he
s
,
va
r
io
us
r
e
la
te
d
w
or
k
ha
s
be
e
n
r
e
vi
e
w
e
d.
P
a
ng
e
t
al
.
[
8]
ha
v
e
us
e
d
m
a
c
hi
ne
le
a
r
ni
ng
f
or
pr
e
di
c
ti
ng
m
or
ta
li
ty
r
is
k
f
or
pa
ti
e
nt
a
dm
it
te
d
in
I
C
U
.
T
he
s
tu
dy
ha
s
us
e
d
M
I
M
I
C
-
I
V
da
ta
s
e
t
w
he
r
e
m
ul
ti
pl
e
le
a
r
ni
ng
m
ode
ls
vi
z
.
d
e
c
is
io
n
t
r
e
e
(
D
T
)
,
s
uppor
t
ve
c
to
r
m
a
c
hi
ne
(
S
V
M
)
,
lo
gi
s
ti
c
r
e
gr
e
s
s
io
n
(
L
R
)
,
a
nd
e
xt
r
e
m
e
gr
a
di
e
nt
boos
ti
ng
(
X
G
B
oos
t
)
ha
s
be
e
n
us
e
d.
S
im
il
a
r
di
r
e
c
ti
on
of
m
or
ta
li
ty
pr
e
di
c
ti
on
is
a
ls
o
c
a
r
r
ie
d
out
by
C
hi
u
e
t
al
.
[
9]
c
ons
id
e
r
in
g
to
pi
c
m
ode
ll
in
g
a
ppr
oa
c
h.
T
he
m
ode
l
ha
s
u
s
e
d
la
te
nt
di
r
ic
hl
e
t
a
ll
oc
a
ti
on
(
L
D
A
)
f
or
te
xt
c
la
s
s
if
ic
a
ti
on
c
ons
id
e
r
in
g
M
I
M
I
C
-
I
I
I
da
ta
s
e
t
to
f
in
d
gr
a
di
e
nt
boos
ti
ng
to
e
xc
e
l
be
tt
e
r
pr
e
di
c
ti
ve
pe
r
f
or
m
a
nc
e
.
C
ogol
lo
e
t
al
.
[
10]
ha
ve
de
s
ig
n
e
d
a
m
ode
l
f
or
pr
e
di
c
ti
ng
s
e
ps
i
s
on
e
a
r
ly
s
ta
ge
u
s
in
g
M
I
M
I
C
-
I
I
I
da
ta
s
e
t
c
ons
id
e
r
in
g
m
ut
li
pl
e
m
a
c
hi
ne
le
a
r
ni
ng
m
ode
l
s
to
w
a
r
ds
a
s
s
e
s
s
in
g
or
ga
n
f
a
il
ur
e
po
s
s
ib
il
it
ie
s
.
A
d
e
n
e
t
al
.
[
11]
ha
v
e
pr
e
s
e
nt
e
d
a
c
la
s
s
if
ic
a
ti
oi
n
f
r
a
m
e
w
or
k
us
in
g
m
ul
ti
pl
e
de
e
p
le
a
r
ni
ng
m
ode
ls
a
nd
na
tu
r
a
l
la
ngua
ge
pr
oc
e
s
s
in
g
(
N
L
P
)
.
T
he
f
r
a
m
e
w
or
k
ha
s
us
e
d
v
a
r
io
us
c
om
bi
na
ti
on
of
lo
ng
s
hor
t
-
te
r
m
m
e
m
or
y
(
L
S
T
M
)
a
nd
bi
di
r
e
c
ti
ona
l
e
nc
ode
r
r
e
pr
e
s
e
nt
a
ti
ons
f
r
om
tr
a
ns
f
or
m
s
(
B
E
R
T
)
to
im
pr
ove
upon
th
e
ir
pr
e
di
c
ti
ve
a
c
c
ur
a
c
y
on
M
I
M
I
C
-
I
I
I
da
ta
s
e
t.
B
oz
kur
t
a
nd
A
ş
ur
oğl
u
[
12]
ha
ve
pr
e
s
e
nt
e
d
a
pr
e
di
c
ti
ve
m
ode
l
us
in
g
m
a
c
hi
ne
le
a
r
ni
ng
a
nd
f
e
a
tu
r
e
a
na
ly
s
is
to
w
a
r
ds
pa
ti
e
nt
s
s
uf
f
e
r
in
g
f
r
om
c
a
nc
e
r
us
in
g
M
I
M
I
C
-
I
V
da
ta
s
e
t.
A
uni
que
m
ode
ll
in
g
to
w
a
r
ds
r
is
k
pr
e
di
c
ti
on
is
di
s
c
u
s
s
e
d
by
O
ga
s
a
w
a
r
a
e
t
al
.
[
13]
f
or
a
s
s
e
s
in
g
pa
ti
e
nt
und
e
r
goi
ng
s
ur
ge
r
y.
T
he
m
od
e
l
ha
s
us
e
d
a
br
ut
e
f
or
c
e
m
e
c
ha
ni
s
m
to
w
a
r
ds
th
e
de
te
r
m
in
a
ti
on
of
ne
w
di
s
e
a
s
e
.
C
a
o
e
t
al
.
[
14]
ha
ve
di
s
c
us
s
e
d
a
s
pe
c
if
ic
va
li
da
ti
on
f
r
a
m
e
w
or
k
f
or
X
G
B
oos
t
a
lg
or
it
hm
to
w
a
r
ds
f
o
r
e
c
a
s
ti
ng
m
or
ta
li
ty
r
a
te
s
in
hos
pi
ta
l
us
in
g
r
e
gr
e
s
s
io
n
m
ode
l.
A
dopt
io
n
of
s
im
il
a
r
X
G
B
oos
t
is
a
ls
o
not
e
d
in
s
tu
dy
of
H
id
a
ya
tu
r
r
ohm
a
n
a
nd
H
a
na
da
[
15]
w
he
r
e
th
e
id
e
a
i
s
t
o a
s
s
e
s
s
t
he
pr
e
di
c
ti
ve
pe
r
f
or
m
a
nc
e
i
m
pr
ove
m
e
nt
us
in
g
pr
e
pr
oc
e
s
s
in
g. K
o
e
t
al
.
[
16]
ha
ve
pr
e
s
e
nt
e
d
a
m
or
ta
li
ty
pr
e
di
c
ti
on
m
ode
l
f
or
te
r
m
in
a
ll
y
il
l
pa
ti
e
nt
c
ons
id
e
r
in
g
m
ul
ti
pl
e
da
ta
s
e
t.
C
hung
e
t
al
.
[
17]
ha
ve
u
s
e
d
X
G
B
oos
t
m
ode
l
f
or
pr
e
di
c
ti
ve
a
na
ly
s
is
of
he
a
r
t
a
tt
a
c
k.
H
e
nc
e
,
c
onvolut
io
n
ne
ur
a
l
ne
twor
k
(
C
N
N
)
,
L
S
T
M
,
DT
,
S
V
M
,
L
R
,
a
nd,
X
G
B
oos
t
a
r
e
id
e
nt
if
ie
d
to
be
f
r
e
que
nl
y
a
dopt
e
d
m
a
c
hi
ne
le
a
r
ni
ng
to
w
a
r
ds
s
ol
vi
ng
pr
e
di
c
ti
on
pr
obl
e
m
s
a
s
s
oc
ia
te
d
w
it
h
c
r
it
ic
a
l
pa
ti
e
nt
s
[
18]
–
[
24]
.
A
f
te
r
r
e
vi
e
w
in
g
th
e
r
e
la
te
d
w
or
k,
th
e
r
e
a
r
e
va
r
io
us
i
de
nt
if
ic
a
ti
on of
r
e
s
e
a
r
c
h pr
obl
e
m
s
:
i)
E
xi
s
ti
ng
s
ys
te
m
ha
s
id
e
nt
if
ie
d
pot
e
nt
ia
l
of
C
N
N
a
s
w
e
ll
a
s
L
S
T
M
to
e
xc
e
l
be
tt
e
r
pr
e
di
c
ti
ve
pe
r
f
or
m
a
nc
e
.
H
ow
e
ve
r
,
th
e
y
c
a
nnot
be
c
ons
id
e
r
e
d
a
s
opt
i
m
a
ll
y
pr
e
f
e
r
r
e
d
s
ol
ut
io
n
f
or
di
a
gnos
is
pr
obl
e
m
s
.
ii)
T
he
r
e
a
r
e
f
e
w
r
e
s
e
a
r
c
h
m
ode
l
s
w
hi
c
h
ha
s
a
c
tu
a
ll
y
f
oc
us
s
e
d
on
c
hr
oni
c
di
s
e
a
s
e
a
s
th
e
c
or
e
pa
r
t
of
r
e
s
e
a
r
c
h e
m
pha
s
is
i
s
m
a
in
ly
gi
ve
n t
o c
r
it
ic
a
l
di
s
e
a
s
e
s
onl
y
.
iii)
E
xi
s
ti
ng
s
tu
dy
ha
s
m
a
in
ly
a
dopt
e
d
th
e
d
a
ta
s
e
t
a
s
a
w
hol
e
;
how
e
ve
r
,
le
s
s
e
m
pha
s
is
i
s
gi
ve
n
to
e
xt
r
a
c
t
th
e
c
om
m
ona
li
ti
e
s
,
uni
que
tr
a
it
s
,
a
nd
pa
ti
e
nt
-
s
pe
c
if
ic
f
e
a
tu
r
e
s
t
ha
t
c
oul
d
a
tt
r
ib
ut
e
to
hi
gh
pe
r
f
o
r
m
in
g
pr
e
di
c
ti
ve
m
ode
ll
in
g
.
iv
)
E
xi
s
ti
ng
s
ys
te
m
is
a
ls
o
w
it
ne
s
s
e
d
to
e
m
pha
s
iz
e
m
or
e
on
in
tr
oduc
in
g
m
a
c
hi
ne
le
a
r
ni
ng
a
lg
or
it
hm
s
in
it
s
s
ophi
s
ti
c
a
te
d
f
or
m
a
nd
he
nc
e
th
e
hi
ghe
r
a
c
c
ur
a
c
y
ha
s
c
om
e
u
p
w
it
h
in
c
r
e
a
s
e
d
c
om
put
a
ti
ona
l
c
os
t
th
a
t
ha
s
be
e
n not m
uc
h e
m
ph
a
s
iz
e
d upon.
H
e
nc
e
,
th
e
a
im
of
th
e
pr
opo
s
e
d
s
t
udy
i
s
to
pr
e
s
e
nt
a
n
ove
l
pr
e
di
c
ti
ve
m
ode
l
th
a
t
c
a
n
r
e
a
l
iz
e
th
e
s
e
v
e
r
it
y
de
gr
e
e
f
r
om
c
hr
oni
c
to
c
r
it
i
c
a
l
di
s
e
a
s
e
of
a
pa
ti
e
nt
w
i
th
opt
im
a
ll
y
hi
gh
e
r
pr
e
di
c
t
iv
e
a
c
c
ur
a
c
y.
T
hi
s
m
ode
l
i
s
a
n
e
xt
e
ns
i
on
of
our
pr
i
or
w
or
k
[
25]
–
[
2
7]
.
T
he
v
a
lu
e
-
a
dde
d
c
ont
r
ib
ut
io
n
of
th
i
s
s
tu
dy
a
r
e
a
s
f
ol
l
ow
s
:
i)
th
e
pr
o
pos
e
d
s
c
he
m
e
in
tr
od
uc
e
s
a
s
im
pl
if
i
e
d
hybr
id
i
z
a
ti
on
of
two
d
e
e
p
l
e
a
r
ni
n
g
m
ode
l
s
(
C
N
N
a
nd
L
S
T
M
)
to
of
f
e
r
b
e
tt
e
r
pr
e
di
c
ti
ve
s
c
or
e
,
ii
)
a
n
ove
l
r
e
c
om
m
e
d
e
r
m
od
ul
e
i
s
in
tr
od
uc
e
d
w
h
ic
h
c
om
pu
te
s
a
n
d
a
s
s
ig
n
s
e
v
e
r
it
y
s
c
or
e
f
ol
lo
w
e
d
by
f
ur
th
e
r
o
pt
im
iz
a
ti
on
u
s
in
g
de
e
p
le
a
r
ni
ng,
ii
i)
th
e
s
t
udy
u
s
e
s
M
I
M
I
C
-
I
V
d
a
ta
s
e
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 5, O
c
to
be
r
2025
:
3926
-
3933
3928
w
he
r
e
d
e
m
ogr
a
phi
c
in
f
or
m
a
ti
on
,
c
ur
r
e
nt
s
ta
tu
s
,
bi
ol
ogi
c
a
l
s
t
a
te
of
p
a
ti
e
nt
,
a
r
e
e
v
a
lu
a
te
d
u
s
in
g
s
im
pl
if
ie
d
a
na
ly
t
ic
a
l
m
ode
l,
a
n
d
iv
)
th
e
s
tu
dy
out
c
om
e
ha
s
be
e
n
b
e
nc
hm
a
r
ke
d
w
it
h
e
xi
s
ti
ng
m
e
th
od
s
to
pr
ove
t
he
e
f
f
e
c
ti
v
e
n
e
s
s
of
s
t
udy
m
ode
l
.
T
he
ne
xt
s
e
c
ti
on
pr
e
s
e
n
ts
di
s
c
u
s
s
i
on
of
th
e
s
ol
ut
io
n
in
th
e
f
or
m
of
a
do
pt
e
d
r
e
s
e
a
r
c
h
m
e
th
od.
2.
M
E
T
H
O
D
T
he
pr
im
e
a
im
of
th
e
pr
opos
e
d
s
tu
dy
i
s
to
d
e
ve
lo
p
a
s
im
pl
if
ie
d
a
nd
ye
t
e
f
f
ic
ie
nt
pr
e
di
c
ti
ve
m
ode
l
to
w
a
r
ds
a
na
ly
z
in
g
s
ta
te
of
c
r
it
ic
a
li
ty
of
a
pa
ti
e
nt
on t
he
ba
s
is
of
a
na
ly
z
e
d s
c
or
e
of
bi
ol
ogi
c
a
l
s
ta
te
c
ons
id
e
r
in
g
M
I
M
I
C
-
I
V
da
ta
s
e
t.
F
ig
ur
e
1
s
how
c
a
s
e
th
e
a
dopt
e
d
a
r
c
hi
te
c
tu
r
e
f
or
th
is
pur
pos
e
w
hi
c
h
s
how
s
m
ul
ti
pl
e
unde
r
ly
in
g
pr
oc
e
s
s
e
s
in
vol
ve
d
in
pr
opos
e
d
s
tu
dy.
T
he
pr
im
a
r
y
s
te
p
of
th
is
m
e
c
ha
ni
s
m
is
to
obt
a
in
th
e
in
f
or
m
a
ti
on
of
hos
pi
ta
l
-
ba
s
e
d
da
ta
a
s
s
o
c
ia
te
d
w
it
h
th
e
pa
ti
e
nt
to
unde
r
s
ta
nd
th
e
vi
s
it
in
f
or
m
a
ti
on
of
a
ll
th
e
pa
ti
e
nt
.
T
he
id
e
a
i
s
to
pe
r
f
or
m
f
or
e
c
a
s
ti
ng
of
e
va
lu
a
te
d
va
lu
e
o
f
bi
ol
ogi
c
a
l
s
ta
te
a
s
s
o
c
ia
te
d
in
up
c
om
in
g
vi
s
it
.
T
he
s
ys
te
m
th
e
n
a
c
qui
r
e
s
th
e
a
c
tu
a
l
in
f
or
m
a
ti
on
of
th
e
pa
t
ie
nt
’
s
bi
ol
ogi
c
a
l
s
t
a
te
in
th
e
ir
n
e
xt
vi
s
it
by
pr
e
di
c
ti
ng
th
e
bi
ol
ogi
c
a
l
s
ta
t
e
of
a
ll
th
e
pr
io
r
vi
s
it
s
.
T
he
c
ons
e
c
ut
iv
e
s
te
p
p
e
r
f
or
m
s
c
onc
a
te
na
ti
on
of
in
f
or
m
a
ti
on o
f
pa
ti
e
nt
’
s
bi
ol
ogi
c
a
l
s
ta
te
not
e
d i
n
i
ni
ti
a
l
a
nd c
ur
r
e
nt
vi
s
it
i
n or
de
r
t
o a
c
qui
r
e
t
he
r
e
s
ul
ta
nt
s
c
or
e
of
pr
e
di
c
ti
on.
T
he
ne
xt
ope
r
a
ti
on
in
a
r
c
hi
te
c
tu
r
e
of
F
ig
ur
e
1
is
a
s
s
oc
ia
te
d
w
it
h
di
a
gnos
i
s
w
he
r
e
a
s
im
pl
if
ie
d
r
e
c
om
m
e
nde
r
s
ys
te
m
ha
s
be
e
n
de
s
ig
ne
d
th
a
t
ge
ne
r
a
te
s
r
a
nks
a
s
s
o
c
ia
te
d
to
c
r
it
ic
a
l
s
ta
te
of
pa
ti
e
nt
.
T
hi
s
r
e
c
om
m
e
nde
r
s
ys
te
m
is
in
it
ia
ll
y
tr
a
in
e
d
by
obt
a
in
in
g
in
f
or
m
a
ti
on
of
bi
ol
ogi
c
a
l
s
ta
te
a
nd
de
m
ogr
a
phi
c
in
f
or
m
a
ti
on
f
or
th
e
pa
ti
e
nt
w
it
ne
s
s
e
d
w
it
h
bot
h
s
e
ve
r
e
c
r
it
ic
a
l
di
s
e
a
s
e
a
s
w
e
ll
a
s
c
hr
oni
c
di
s
e
a
s
e
.
T
he
tr
a
in
e
d
r
e
c
om
m
e
nde
r
s
ys
t
e
m
f
ur
th
e
r
ta
ke
s
th
e
in
put
of
pr
e
di
c
te
d
bi
ol
ogi
c
a
l
s
ta
te
a
nd
d
e
m
ogr
a
phi
c
in
f
or
m
a
ti
on
in
or
de
r
t
o ge
ne
r
a
te
t
he
f
in
a
l
out
c
om
e
of
pr
e
di
c
ti
on.
F
ig
ur
e
1. A
r
c
hi
te
c
tu
r
e
of
pr
opos
e
d s
ys
te
m
A
dopt
io
n
of
M
I
M
I
C
-
I
V
da
ta
s
e
t
of
f
e
r
s
a
c
c
e
s
s
to
va
r
io
us
e
s
s
e
nt
ia
l
in
f
or
m
a
ti
on
e
.g.
pa
ti
e
nt
de
m
ogr
a
phi
c
s
, c
li
ni
c
a
l
not
e
s
, m
e
di
c
a
ti
on,
a
nd
la
b r
e
s
ul
ts
. H
e
nc
e
, a
be
tt
e
r
f
or
m
of
p
r
e
di
c
ti
ve
m
ode
ll
in
g c
a
n be
c
a
r
r
ie
d
out
a
s
th
is
da
ta
s
e
t
of
f
e
r
s
be
tt
e
r
li
ke
li
hood
e
s
ti
m
a
ti
on
of
m
or
ta
li
ty
a
s
w
e
ll
a
s
ot
he
r
c
r
it
ic
a
l
c
om
pl
ic
a
ti
ons
.
A
pa
r
t
f
r
om
th
is
,
be
in
g
a
p
a
r
t
of
ope
n
a
c
c
e
s
s
c
om
m
uni
ty
,
it
of
f
e
r
s
w
id
e
r
s
uppor
ta
bi
li
ty
of
a
s
s
e
s
s
in
g A
I
m
ode
l
to
w
a
r
d s
tu
dyi
ng dis
e
a
s
e
pr
ogr
e
s
s
io
n, m
or
ta
li
ty
r
a
te
s
, a
nd c
r
it
ic
a
l
c
a
r
e
. H
e
nc
e
, a
dopt
io
n of
M
I
M
I
C
-
I
V
da
ta
s
e
t
of
f
e
r
a
w
id
e
r
s
c
ope
of
e
va
lu
a
ti
on f
o
r
t
he
pr
o
pos
e
d s
tu
dy mode
l
to
w
a
r
ds
pr
e
di
c
ti
ng s
ta
te
of
c
r
it
ic
a
li
ty
a
dopt
in
g de
e
p l
e
a
r
ni
ng mode
l.
T
he
e
la
bor
a
te
d op
e
r
a
ti
on of
t
he
a
r
c
hi
te
c
tu
r
e
a
r
e
a
s
f
ol
lo
w
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
H
y
br
id
iz
e
d de
e
p l
e
ar
ni
ng m
od
e
l
w
it
h nov
e
l
r
e
c
om
m
e
nd
e
r
f
or
p
r
e
di
c
ti
ng c
r
it
ic
al
it
y
s
ta
te
…
(
Sar
ik
a
K
hope
)
3929
2.1
.
I
n
t
e
gr
at
e
d
d
e
e
p
l
e
ar
n
in
g m
od
u
le
F
r
om
th
e
pr
a
c
ti
c
a
l
s
c
e
na
r
io
,
a
r
e
a
l
-
s
ta
te
of
th
e
p
a
ti
e
nt
is
a
na
ly
z
e
d
by
a
phys
i
c
ia
n
w
ho
r
e
c
om
m
e
nds
f
or
f
u
r
th
e
r
s
ta
ge
s
of
tr
e
a
tm
e
nt
.
T
hi
s
pr
oc
e
s
s
is
a
ut
om
a
te
d
by
a
dopt
in
g
C
N
N
w
he
r
e
th
e
M
I
M
I
C
-
I
V
da
ta
s
e
t
is
c
ons
id
e
r
e
d
a
s
a
n
in
put
a
r
gum
e
nt
f
ol
lo
w
e
d
by
e
xt
r
a
c
ti
ng
th
e
la
te
nt
a
tt
r
ib
ut
e
s
in
i)
da
ta
r
e
la
te
d
to
m
e
di
c
a
ti
on
a
nd
ii
)
da
ta
r
e
la
te
d
to
phys
ic
a
l
s
ym
pt
om
s
.
T
he
f
in
a
l
s
c
or
e
o
f
a
tt
r
ib
ut
e
s
a
r
e
obt
a
in
e
d
f
r
om
c
onc
a
te
na
ti
ng
va
r
io
us
la
ye
r
s
in
or
de
r
to
a
c
qui
r
e
a
hi
gh
-
le
ve
l
a
bs
tr
a
c
ti
ve
f
or
m
of
a
tt
r
ib
ut
e
s
.
T
hi
s
f
in
a
ll
y
obt
a
in
e
d
a
tt
r
ib
ut
e
s
unde
r
go
f
ur
th
e
r
pr
oc
e
s
s
in
g
us
in
g
L
S
T
M
us
in
g
th
r
e
e
di
s
c
r
e
te
g
a
te
s
ys
te
m
(
f
or
ge
t,
in
put
,
a
nd
r
e
s
ul
t)
.
F
ig
ur
e
2
s
how
c
a
s
e
th
e
f
lo
w
of
th
e
in
te
r
na
l
ope
r
a
ti
on
of
th
is
m
odul
e
,
w
he
r
e
it
c
a
n
be
s
e
e
n
th
a
t
in
f
o
r
m
a
ti
on
f
r
om
bot
h
in
di
vi
dua
l
a
nd
hos
pi
ta
l
a
s
s
oc
ia
t
e
d
w
it
h
c
li
ni
c
a
l
s
ta
te
of
a
n
in
di
vi
dua
l
is
c
ons
id
e
r
e
d
w
hi
c
h
a
r
e
s
ubj
e
c
te
d
to
C
N
N
la
ye
r
s
to
ge
ne
r
a
te
f
e
a
tu
r
e
s
.
T
he
ge
ne
r
a
te
d
f
e
a
tu
r
e
s
a
r
e
th
e
n
s
ubj
e
c
te
d
to
L
S
T
M
f
ol
lo
w
e
d
b
y
c
onc
a
te
na
ti
ng
th
e
m
w
he
r
e
th
e
c
onc
a
te
na
t
e
d
f
e
a
tu
r
e
s
a
r
e
f
ur
th
e
r
s
ubj
e
c
te
d
to
de
n
s
e
l
a
ye
r
s
.
T
he
de
t
e
r
m
in
a
ti
on
of
de
m
a
nds
of
f
or
ge
tt
in
g
a
s
s
oc
ia
te
d
w
it
h
pr
e
vi
ous
c
e
ll
is
c
a
r
r
i
e
d
out
by
f
or
ge
t
ga
te
f
ol
lo
w
e
d
by
de
te
r
m
in
in
g
th
e
ne
c
e
s
s
a
r
y
f
e
a
tu
r
e
s
to
te
h c
e
ll
f
or
upda
ti
ng
s
ta
t
e
of
c
e
ll
in
in
put
ga
te
.
A
t
th
e
e
nd,
th
e
e
xa
c
t
s
ta
te
of
r
e
s
ul
t
is
de
c
id
e
d a
t
out
put
ga
te
of
L
S
T
M
.
F
ig
ur
e
2
.
I
nt
e
gr
a
te
d de
e
p l
e
a
r
ni
ng modul
e
2.2
.
R
e
c
om
m
e
n
d
e
r
m
od
u
le
T
hi
s
m
odul
e
is
r
e
s
pons
ib
le
f
or
de
te
r
m
in
in
g
th
e
li
kl
ih
ood
of
oc
c
ur
a
nc
e
of
a
ny
c
r
it
ic
a
l
s
ta
te
of
a
pa
ti
e
nt
on t
he
ba
s
is
of
t
he
ir
de
m
ogr
a
phi
c
da
ta
a
s
w
e
ll
a
s
e
va
lu
a
t
e
d i
nf
or
m
a
ti
on of
phys
ic
a
l
a
tt
r
ib
ut
e
s
F
ig
ur
e
3.
A
dopt
io
n
of
M
I
M
I
C
-
I
V
da
ta
s
e
t
in
pr
opos
e
d
s
tu
dy
de
f
in
e
s
pa
r
ti
c
ul
a
r
di
s
e
a
s
e
th
a
t
is
f
ound
to
be
e
vol
vi
ng
a
s
w
e
ll
a
s
a
s
e
t
of
r
e
gul
a
r
c
hr
oni
c
di
s
e
a
s
e
. F
or
t
hi
s
pur
pos
e
, a
n obj
e
c
ti
ve
f
unc
ti
on i
s
e
m
pi
r
ic
a
ll
y s
e
t
a
s
(
1)
.
=
(
1)
I
n
e
xpr
e
s
s
io
n
(
1)
,
th
e
va
r
ia
bl
e
Φ
r
e
pr
e
s
e
nt
s
a
n
obj
e
c
ti
ve
f
unc
ti
on
de
pi
c
ti
ng
pos
s
ib
le
v
a
lu
e
of
s
pe
c
if
ic
num
be
r
of
pa
r
ie
nt
pos
s
e
s
s
in
g
m
a
ni
f
ol
d
di
s
e
a
s
e
,
va
r
ia
bl
e
π
r
e
pr
e
s
e
nt
s
w
e
ig
ht
e
d
m
a
tr
ix
,
a
nd
va
r
ia
bl
e
λ
r
e
pr
e
s
e
nt
s
s
ta
te
m
a
tr
ix
c
ons
id
e
r
in
g
d
dur
a
ti
on of
t
im
e
. T
he
w
e
ig
ht
e
d m
a
tr
ix
π
i
s
obt
a
in
e
d by pr
oduc
t
of
di
s
e
a
s
e
i
nf
or
m
a
ti
on i
n
M
I
M
I
C
-
I
V
da
ta
s
e
t
a
nd
c
ur
r
e
nt
s
t
a
te
w
hi
le
s
ta
te
m
a
tr
ix
λ
i
s
obt
a
in
e
d
f
r
om
pr
oduc
t
of
in
f
or
m
a
ti
on
of
pa
ti
e
nt
in
M
I
M
I
C
-
I
V
da
ta
s
e
t
a
nd
c
ur
r
e
nt
s
ta
te
.
I
t
is
ne
c
e
s
s
a
r
y
th
a
t
w
e
ig
ht
e
d
m
a
tr
ix
m
us
t
c
a
te
r
up
obj
e
c
ti
ve
f
unc
ti
on
be
a
r
in
g
s
ta
te
m
a
tr
ix
in
or
de
r
to
pr
e
di
c
t
th
e
r
a
nk
of
c
r
it
ic
a
li
ty
m
a
tc
hi
ng
w
it
h
pa
ti
e
nt
in
f
or
m
a
ti
on
a
nd
di
s
e
a
s
e
in
f
or
m
a
ti
on.
T
hi
s
c
ha
ll
e
nge
of
pr
e
di
c
ti
on
is
s
ol
ve
d
us
in
g
a
s
im
pl
if
ie
d
r
e
c
o
m
m
e
nde
r
s
ys
te
m
w
hi
c
h
is
c
a
pa
bl
e
of
ge
ne
r
a
ti
ng
a
r
a
nk
of
th
e
c
r
it
ic
a
li
ty
s
ta
te
by
le
a
r
ni
ng
th
e
m
od
e
l
us
in
g
s
to
c
ha
s
ti
c
gr
a
di
e
nt
de
s
c
e
nt
f
or
be
tt
e
r
opt
im
iz
a
ti
on.
T
hi
s
m
e
c
h
a
ni
s
m
is
u
s
e
d
f
or
c
om
put
in
g
th
e
w
e
ig
ht
e
d
m
a
tr
ix
.
F
or
th
is
pur
pos
e
,
th
is
m
odul
e
c
ons
tr
uc
ts
a
m
a
tr
ix
(
δ
,
P
1
,
P
2
)
w
he
r
e
δ
r
e
pr
e
s
e
nt
s
di
s
e
a
s
e
to
c
ons
tr
uc
t
a
lo
gi
c
th
a
t
pa
ti
e
nt
P
1
ha
s
hi
ghe
r
pr
oba
bi
li
ty
to
de
m
a
nd
c
r
it
ic
a
li
ty
s
c
r
e
e
ni
ng
in
c
ont
r
a
s
t
to
pa
ti
e
nt
P
2
.
T
he
c
om
put
a
ti
on
of
th
is
pr
oba
bi
li
ty
Pr
is
c
a
r
r
ie
d out e
m
pi
r
ic
a
ll
y a
s
(
2)
.
Pr
(
hp
)
=
∑
1
=
1
(
2)
I
n
(
2)
,
pr
oba
bi
li
ty
Pr
is
c
om
put
e
d
w
it
h
r
e
s
pe
c
t
to
hype
r
pa
r
a
m
e
te
r
s
hp
a
nd
is
obt
a
in
e
d
by
s
um
m
a
ti
on
of
pa
r
a
m
e
te
r
A
1
w
he
r
e
A
1
r
e
pr
e
s
e
nt
s
pr
oduc
t
of
in
di
vi
dua
l
w
e
ig
ht
a
nd
c
um
ul
a
ti
ve
in
f
or
m
a
ti
on
f
r
om
M
I
M
I
C
-
I
V
da
ta
s
e
t
pe
r
ta
in
in
g
to
s
ym
pt
om
s
of
pa
ti
e
nt
s
a
nd
th
e
ir
a
s
s
oc
i
a
te
d
de
m
ogr
a
phi
c
s
.
I
t
w
il
l
e
ve
nt
ua
ll
y
m
e
a
n
th
a
t
if
pa
ti
e
nt
P
1
ha
s
c
onf
ir
m
e
d
a
nd
in
c
r
e
a
s
in
g
va
lu
e
of
e
vol
vi
ng
di
s
e
a
s
e
s
c
om
pa
r
e
d
to
a
lr
e
a
dy
e
xi
s
ti
ng
va
lu
e
s
of
di
s
e
a
s
e
s
in
M
I
M
I
C
I
V
da
ta
s
e
t
th
a
n
th
is
r
e
s
pe
c
ti
ve
pa
ti
e
nt
in
f
or
m
a
ti
on
is
pr
e
di
c
te
d
to
e
xhi
bi
t
th
a
t
s
pe
c
if
ic
di
s
e
a
s
e
i
n f
ut
ur
e
. F
ig
ur
e
3 s
how
c
a
s
e
t
he
m
e
c
ha
ni
s
m
of
pr
opos
e
d r
e
c
om
m
e
nde
r
m
odul
e
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 5, O
c
to
be
r
2025
:
3926
-
3933
3930
F
ig
ur
e
3. P
r
opos
e
d r
e
c
om
m
e
nde
r
m
odul
e
2.3
.
S
ym
p
t
om
p
r
e
d
ic
t
io
n
m
od
u
le
T
he
out
c
om
e
of
th
e
pr
e
vi
ous
m
odul
e
a
c
ts
a
s
a
n
in
put
f
or
th
e
c
ur
r
e
nt
m
odul
e
to
pr
e
di
c
t
s
e
ve
r
it
y.
V
a
r
io
us
la
ye
r
w
is
e
ope
r
a
ti
on
is
c
a
r
r
ie
d
out
c
ons
id
e
r
in
g
c
om
bi
ne
d
C
N
N
a
nd
L
S
T
M
a
s
s
how
n
in
F
ig
ur
e
4.
T
he
M
I
M
I
C
-
I
V
da
ta
s
e
t
i
s
us
e
d
w
he
r
e
th
e
hos
pi
ta
l
d
a
ta
a
nd
in
di
vi
dua
l
pa
ti
e
nt
in
f
or
m
a
ti
on
w
e
r
e
obt
a
in
e
d
u
s
in
g
in
te
gr
a
te
d
de
e
p
le
a
r
ni
ng
m
odul
e
.
T
he
f
e
a
tu
r
e
s
a
s
s
o
c
ia
te
d
w
i
th
bot
h
in
di
vi
dua
l
in
f
or
m
a
ti
on
a
nd
c
om
m
on
in
f
or
m
a
ti
on
w
e
r
e
a
ppe
nde
d
to
ge
th
e
r
in
c
onc
a
te
na
ti
on
l
a
ye
r
.
T
h
e
a
c
c
om
pl
is
h
e
d
out
c
om
e
i
s
th
e
n
f
or
w
a
r
de
d
to
f
ul
ly
c
onne
c
te
d l
a
ye
r
i
n o
r
de
r
t
o
f
or
e
c
a
s
t
th
e
bi
ol
ogi
c
a
l
s
ta
te
of
th
e
pa
ti
e
nt
w
hi
c
h a
r
e
l
ik
e
ly
t
o be
e
xhi
bi
te
d
by
th
e
pa
ti
e
nt
in
th
e
ir
upc
om
in
g
vi
s
it
or
f
ol
lo
w
-
ups
.
T
he
pr
e
di
c
te
d
va
lu
e
s
of
bi
ol
ogi
c
a
l
s
ta
te
is
f
e
d
to
th
e
pr
io
r
r
e
c
om
m
e
nde
r
m
odul
e
i
n or
de
r
t
o i
de
nt
if
y a
nd a
s
s
e
s
s
t
he
l
ik
li
ho
od of
s
e
ve
r
e
c
r
it
ic
a
l
s
ta
te
i
n upc
om
in
g vi
s
it
f
or
th
e
pa
r
ti
c
ul
a
r
pa
ti
e
nt
. T
hi
s
i
s
m
or
e
va
li
d f
or
bot
h pa
ti
e
nt
w
it
h c
hr
oni
c
a
nd e
vol
vi
ng dis
e
a
s
e
s
. I
t
is
ne
c
e
s
s
a
r
y t
o
unde
r
s
ta
nd t
ha
t
va
lu
e
s
of
c
li
ni
c
a
l
s
e
tt
in
gs
f
or
hos
pi
ta
l
-
ba
s
e
d i
nf
or
m
a
ti
on w
il
l
di
f
f
e
r
f
r
om
t
ha
t
of
pa
ti
e
nt
-
ba
s
e
d
in
f
or
m
a
ti
on
a
s
w
e
ll
a
s
c
a
r
di
na
li
ty
of
bi
ol
ogi
c
a
l
s
ta
te
of
f
e
a
tu
r
e
s
.
T
h
e
pr
opos
e
d
s
ys
t
e
m
c
on
s
id
e
r
s
it
e
m
s
of
bi
ol
ogi
c
a
l
s
ta
te
a
s
w
e
ll
a
s
it
e
m
s
of
m
e
di
c
in
e
a
s
s
oc
ia
te
d w
it
h
va
r
io
us
c
r
it
ic
a
l
c
ondi
ti
ons
r
e
la
te
d
to
ke
oa
c
id
o
s
is
(
X
1
)
,
r
e
s
pi
r
a
to
r
y
f
a
il
ur
e
(
X
2
)
,
pa
nc
r
e
a
ti
ti
s
(
X
3
)
,
r
e
na
l
f
a
il
ur
e
s
(
X
4
)
,
he
a
r
t
f
a
il
ur
e
(
X
5
)
,
a
nd
c
e
r
e
br
a
l
he
m
or
r
ha
ge
(
X
6
)
.
A
pa
r
t
f
r
om
th
is
,
th
e
s
ys
te
m
a
ls
o
c
ons
id
e
r
s
va
r
io
us
pr
e
e
xi
s
ti
ng
s
e
t
of
di
s
e
a
s
e
s
of
c
hr
oni
c
f
or
m
a
s
s
oc
ia
te
d
w
it
h
e
a
c
h
c
r
it
ic
a
l
c
ondi
ti
ons
.
T
he
f
in
a
l
out
c
om
e
of
th
is
pr
e
di
c
ti
on
m
odul
e
is
th
e
opt
im
a
l
r
a
nki
ng
s
c
or
e
of
bi
ol
ogi
c
a
l
s
ta
te
of
pa
ti
e
nt
w
hi
c
h
is
a
nt
ic
ip
a
te
d
to
be
in
hi
ghe
r
pr
oxi
m
it
y
to
th
e
a
c
tu
a
l
s
ta
t
e
w
he
n
th
e
pa
ti
e
nt
vi
s
it
s
ne
xt
f
or
f
ol
lo
w
up.
F
ig
ur
e
4. S
e
ve
r
it
y pr
e
di
c
ti
on modul
e
3.
R
E
S
U
L
T
T
he
s
c
r
ip
ti
ng
of
th
e
pr
opos
e
d
m
ode
l
is
c
a
r
r
ie
d
out
in
pyt
hon
in
J
upyt
e
r
e
nvi
r
onm
e
nt
c
ons
i
s
ts
of
dua
l
c
onvolut
io
n
la
ye
r
s
w
he
r
e
th
e
r
e
a
r
e
32
f
il
te
r
s
f
or
f
ir
s
t
la
ye
r
a
nd
64
f
il
te
r
s
f
or
s
e
c
ond
la
ye
r
.
T
he
le
ngt
h
of
f
il
te
r
is
3
w
hi
le
r
e
c
ti
f
ie
d
li
ne
a
r
uni
t
is
us
e
d
a
s
a
c
ti
va
ti
on
f
unc
ti
o
n.
T
he
le
ngt
h
of
pool
is
c
ons
id
e
r
e
d
a
s
2
in
M
a
xP
ool
in
g
la
ye
r
of
C
N
N
w
hi
le
th
r
e
e
di
f
f
e
r
e
nt
la
ye
r
s
of
L
S
T
M
is
c
on
s
id
e
r
e
d.
T
he
f
ir
s
t
a
nd
s
e
c
ond
la
ye
r
of
L
S
T
M
ha
s
64
e
a
c
h
a
s
hi
dd
e
n
uni
ts
w
hi
le
th
e
th
ir
d
L
S
T
M
l
a
ye
r
ha
s
32
hi
dde
n
uni
ts
.
T
he
s
tu
dy
c
hoo
s
e
s
24
bi
ol
ogi
c
a
l
s
ta
te
s
f
r
om
M
I
M
I
C
-
I
V
da
t
a
s
e
t.
T
he
num
e
r
ic
a
l
out
c
om
e
of
s
tu
dy
i
s
a
na
ly
z
e
d
u
s
in
g
s
ta
nd
a
r
d
pe
r
f
or
m
a
nc
e
m
e
tr
ic
s
of
a
c
c
ur
a
c
y,
pr
e
c
i
s
io
n
,
r
e
c
a
ll
,
a
nd
F
1
-
s
c
or
e
a
s
e
xhi
bi
te
d
in
T
a
bl
e
1.
T
he
out
c
om
e
s
how
c
a
s
e
s
s
im
il
a
r
c
ons
is
te
nc
y f
or
a
lm
os
t
a
ll
t
he
t
a
r
ge
t
s
c
r
e
e
ni
n
g a
s
s
oc
ia
t
e
d w
it
h c
r
it
ic
a
l
di
s
e
a
s
e
s
.
T
he
pr
opos
e
d
s
tu
dy
m
ode
l
(
P
r
op)
ha
s
a
ls
o
unde
r
w
e
nt
a
c
om
pa
r
it
iv
e
a
na
ly
s
is
c
ons
id
e
r
in
g
th
e
f
r
e
que
nt
ly
a
dopt
e
d
m
a
c
hi
ne
le
a
r
ni
ng
m
ode
ls
a
s
s
how
n
in
F
ig
ur
e
5
w
it
h
r
e
s
pe
c
t
to
pr
e
di
c
ti
ve
a
c
c
ur
a
c
y
a
nd
pr
oc
e
s
s
in
g
ti
m
e
.
T
he
out
c
om
e
in
F
ig
ur
e
5(
a
)
s
how
c
a
s
e
s
th
a
t
pr
opos
e
d
s
tu
dy
m
ode
l
of
f
e
r
s
a
ppr
oxi
m
a
te
ly
54.7%
of
hi
ghe
r
a
c
c
ur
a
c
y
in
c
ont
r
a
s
t
to
e
xi
s
ti
ng
m
a
c
hi
ne
le
a
r
ni
ng
m
ode
ls
.
I
nt
e
r
e
s
ti
ngl
y,
pr
opos
e
d
m
ode
l
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
H
y
br
id
iz
e
d de
e
p l
e
ar
ni
ng m
od
e
l
w
it
h nov
e
l
r
e
c
om
m
e
nd
e
r
f
or
p
r
e
di
c
ti
ng c
r
it
ic
al
it
y
s
ta
te
…
(
Sar
ik
a
K
hope
)
3931
us
in
g
C
N
N
a
nd
L
S
T
M
is
f
ound
to
e
xc
e
l
a
ppr
oxi
m
a
te
ly
5
%
a
nd
6%
of
in
c
r
e
a
s
e
d
a
c
c
ur
a
c
y
in
c
ont
r
a
s
t
to
s
ta
nda
lo
ne
ve
r
s
io
n
of
C
N
N
a
nd
L
S
T
M
r
e
s
pe
c
ti
ve
ly
.
I
n
p
e
r
s
pe
c
ti
ve
of
pr
oc
e
s
s
in
g
ti
m
e
a
s
s
how
n
in
F
ig
ur
e
5(
b
)
, pr
opos
e
d s
ys
te
m
i
s
w
it
ne
s
s
e
d t
o s
how
a
ppr
oxi
m
a
te
ly
70%
r
e
duc
e
d pr
oc
e
s
s
in
g t
im
e
i
n c
ont
r
a
s
t
t
o
m
e
a
n
va
lu
e
of
a
ll
e
xi
s
ti
ng
m
ode
ls
.
T
he
s
tu
dy
m
ode
l
is
f
ound
to
e
xhi
bi
t
19
a
nd
45%
of
r
e
du
c
e
d
pr
oc
e
s
s
in
g
ti
m
e
in
c
ont
r
a
s
t
to
s
ta
nda
lo
ne
C
N
N
a
nd
L
S
T
M
r
e
s
pe
c
ti
ve
ly
.
O
th
e
r
m
ode
ls
li
ke
X
G
B
oos
t,
L
R
,
a
nd
S
V
M
is
f
ound with nea
r
ly
s
im
il
a
r
t
r
e
nd w
it
h r
e
s
pe
c
t
to
bot
h t
he
pe
r
f
or
m
a
nc
e
m
e
tr
ic
.
T
he
a
na
ly
s
is
of
f
e
r
s
va
r
io
us
m
a
ni
f
ol
d
le
a
r
ni
ng
out
c
om
e
s
.
T
he
pr
im
a
r
y
le
a
r
ni
ng
out
c
om
e
is
t
o
unde
r
s
ta
nd
th
a
t
a
be
tt
e
r
pr
e
di
c
ti
ve
m
ode
ll
in
g
de
m
a
nds
hybr
id
m
ode
ll
in
g
w
hi
c
h
c
oul
d
a
c
tu
a
ll
y
c
ur
ta
in
th
e
ope
r
a
ti
ona
l
a
nd r
e
s
our
c
e
de
m
a
nd c
os
t
in
pe
r
s
pe
c
ti
ve
of
m
a
c
hi
n
e
l
e
a
r
ni
ng a
ppr
oa
c
he
s
. T
he
s
e
c
ond
a
r
y l
e
a
r
ni
ng
out
c
om
e
is
to
r
e
a
li
z
e
th
a
t
M
I
M
I
C
-
I
V
da
ta
s
e
t
ha
s
unde
ni
a
bl
y
e
nr
ic
he
d
s
e
t
of
in
f
or
m
a
ti
oni
;
how
e
ve
r
,
not
a
ll
th
e
in
f
or
m
a
ti
ons
a
r
e
de
m
a
nde
d
f
o
r
pe
r
f
o
r
m
in
g
p
r
e
di
c
ti
ve
m
ode
ll
in
g.
A
s
th
e
pr
opos
e
d
s
ys
te
m
ta
r
ge
ts
to
w
a
r
ds
m
in
im
iz
in
g
th
e
e
f
f
or
t
of
phys
ic
ia
n
to
di
a
gno
s
e
th
e
pa
ti
e
nt
in
e
ve
r
y
upc
om
in
g
vi
s
it
s
,
s
o
th
e
de
m
ogr
a
phi
c
in
f
or
m
a
ti
on
a
nd
a
c
tu
a
l
s
ta
te
in
f
or
m
a
ti
on
obt
a
in
e
d
f
r
om
pr
e
di
c
ti
ve
ope
r
a
ti
on
in
ju
s
t
th
r
e
e
s
te
ps
c
a
n
ov
e
r
c
om
e
th
e
c
ha
ll
e
nge
.
H
e
nc
e
,
pr
opos
e
d
s
tu
dy
m
ode
l
i
s
pr
ove
n
to
of
f
e
r
a
c
o
s
t
-
e
f
f
e
c
ti
ve
a
r
c
hi
te
c
tu
r
a
l
d
e
s
ig
n
th
a
t
not
onl
y c
us
to
m
iz
e
t
he
c
li
ni
c
a
l
s
e
tt
in
gs
f
or
c
r
it
ic
a
l
pa
ti
e
nt
c
a
r
e
but
a
ls
o of
f
e
r
f
a
s
te
r
a
nd s
im
pl
if
e
d ope
r
a
ti
ons
.
T
a
bl
e
1. N
um
e
r
ic
a
l
out
c
om
e
f
or
va
r
io
us
di
s
e
a
s
e
t
a
r
ge
ts
T
a
r
ge
t
s
c
r
e
e
ni
ng
A
c
c
ur
a
c
y
P
r
e
c
i
s
i
on
R
e
c
a
l
l
F1
-
s
c
or
e
X
1
97.9
0.993
0.951
0.978
X
2
98.6
0.899
0.992
0.991
X
3
96.9
0.942
0.975
0.961
X
4
98.8
0.984
0.984
0.987
X
5
93.1
0.992
0.928
0.933
X
6
96.3
0.811
0.935
0.999
A
ve
r
a
ge
96.93
0.9368
0.9608
0.9748
(
a
)
(
b)
F
ig
ur
e
5. R
e
s
ul
t
of
c
om
pa
r
a
ti
ve
a
na
ly
s
is
of
(
a
)
a
c
c
ur
a
c
y
a
nd
(
b)
pr
oc
e
s
s
in
g t
im
e
4.
C
O
N
C
L
U
S
I
O
N
T
he
pr
opo
s
e
d
s
tu
d
y
in
ve
s
ti
g
a
te
s
di
f
f
e
r
e
nt
c
ont
r
ib
ut
i
on
s
of
m
a
c
hi
n
e
le
a
r
ni
ng
a
ppr
o
a
c
h
e
s
t
ow
a
r
d
s
c
r
it
ic
a
l
c
a
r
e
of
p
a
ti
e
nt
s
s
uf
f
e
r
in
g
f
r
om
e
it
he
r
c
hr
oni
c
d
is
e
a
s
e
a
s
w
e
ll
a
s
ne
w
ly
e
vol
vi
ng dis
e
a
s
e
s
.
T
he
pr
opo
s
e
d
s
c
h
e
m
e
of
f
e
r
s
a
hybr
id
i
z
a
ti
o
n
of
C
N
N
a
n
d
L
S
T
M
,
th
e
t
w
o
po
w
e
r
f
ul
de
e
p
l
e
a
r
ni
ng
m
ode
l
,
in
or
d
e
r
t
o
pr
e
di
c
t
th
e
p
os
s
ib
l
e
s
ta
t
e
of
c
r
it
i
c
a
l
it
y
i
n
up
c
om
in
g
vi
s
it
s
.
U
nl
ik
e
e
xi
s
ti
ng
a
ppr
oa
c
he
s
,
pr
opo
s
e
d
s
c
h
e
m
e
do
e
s
n’
t
de
m
a
n
d
a
ny
it
e
r
a
ti
v
e
tr
a
in
in
g
ope
r
a
ti
on
,
nor
doe
s
it
in
vol
ve
in
te
ns
i
ve
a
na
ly
ti
c
a
l
pr
oc
e
s
s
in
g
th
a
t
is
w
it
n
e
s
s
e
d
f
r
om
it
s
a
c
c
om
pl
is
he
d
num
e
r
ic
a
l
out
c
om
e
s
.
T
he
s
c
he
m
e
ha
s
in
tr
odu
c
e
d
a
nov
e
l
r
e
c
om
m
e
nd
a
ti
on
s
ys
t
e
m
w
hi
c
h
w
or
ks
i
n
two
m
o
de
s
—
pr
io
r
p
e
r
f
or
m
in
g
de
e
p
l
e
a
r
ni
n
g
ope
r
a
ti
o
n
a
nd
a
f
te
r
p
e
r
f
or
m
in
g
d
e
e
p
l
e
a
r
ni
ng
ope
r
a
ti
on
—
t
he
r
e
by
e
n
s
ur
in
g
hi
g
hl
y
va
li
da
t
e
d
c
r
it
ic
a
li
ty
s
c
or
e
a
s
pr
e
di
c
ti
ve
o
ut
c
om
e
.
T
he
s
t
udy
out
c
om
e
i
s
a
s
s
e
s
s
e
d
on
s
t
a
nd
a
r
d
p
e
r
f
or
m
a
n
c
e
m
e
tr
i
c
t
o
f
in
d
pr
op
os
e
d
s
c
he
m
e
to
e
xc
e
l
b
e
tt
e
r
pe
r
f
or
m
a
nc
e
on
m
ul
ti
pl
e
s
e
t
of
c
r
it
ic
a
l
di
s
e
a
s
e
a
tt
r
ib
ut
e
s
in
c
ont
r
a
s
t
to
e
xi
s
ti
n
g f
r
e
qu
e
nt
ly
a
do
pt
e
d m
a
c
hi
ne
l
e
a
r
ni
ng a
ppr
oa
c
he
s
.
F
U
N
D
I
N
G
I
N
F
O
R
M
A
T
I
O
N
A
ut
hor
s
s
ta
te
no f
undi
ng i
nvol
ve
d.
A
U
T
H
O
R
C
O
N
T
R
I
B
U
T
I
O
N
S
S
T
A
T
E
M
E
N
T
T
hi
s
jo
ur
na
l
us
e
s
th
e
C
ont
r
ib
ut
or
R
ol
e
s
T
a
xonomy
(
C
R
e
d
iT
)
to
r
e
c
ogni
z
e
in
di
vi
dua
l
a
ut
hor
c
ont
r
ib
ut
io
ns
, r
e
duc
e
a
ut
hor
s
hi
p di
s
put
e
s
,
a
nd f
a
c
il
it
a
te
c
ol
la
bo
r
a
ti
on.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
. 14, No. 5, O
c
to
be
r
2025
:
3926
-
3933
3932
N
am
e
o
f
A
u
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
S
a
r
ik
a
K
hope
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
D
e
e
pa
li
K
ot
a
m
bka
r
✓
✓
✓
✓
✓
✓
R
a
m
a
V
a
s
a
nt
h
a
A
d
ir
a
ju
✓
✓
✓
✓
✓
S
m
it
a
S
uha
s
B
a
tt
a
lwa
r
✓
✓
✓
✓
✓
✓
✓
✓
C
:
C
onc
e
pt
ua
l
i
z
a
t
i
on
M
:
M
e
t
hodol
ogy
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
da
t
i
on
Fo
:
Fo
r
m
a
l
a
na
l
ys
i
s
I
:
I
nve
s
t
i
ga
t
i
on
R
:
R
e
s
our
c
e
s
D
:
D
a
t
a
C
ur
a
t
i
on
O
:
W
r
i
t
i
ng
-
O
r
i
gi
na
l
D
r
a
f
t
E
:
W
r
i
t
i
ng
-
R
e
vi
e
w
&
E
di
t
i
ng
Vi
:
Vi
s
ua
l
i
z
a
t
i
on
Su
:
Su
pe
r
vi
s
i
on
P
:
P
r
oj
e
c
t
a
dm
i
ni
s
t
r
a
t
i
on
Fu
:
Fu
ndi
ng a
c
qui
s
i
t
i
on
C
O
N
F
L
I
C
T
O
F
I
N
T
E
R
E
S
T
S
T
A
T
E
M
E
N
T
A
ut
hor
s
s
ta
te
no c
onf
li
c
t
of
i
nt
e
r
e
s
t.
D
A
T
A
A
V
A
I
L
A
B
I
L
I
T
Y
T
he
da
ta
th
a
t
s
uppor
t
th
e
f
in
di
ngs
of
th
is
s
tu
dy
a
r
e
a
va
il
a
bl
e
f
r
om
th
e
c
or
r
e
s
ponding
a
ut
hor
,
[
S
K
]
,
upon r
e
a
s
ona
bl
e
r
e
que
s
t.
R
E
F
E
R
E
N
C
E
S
[
1]
S
.
G
ow
t
ha
m
i
,
R
.
V
.
S
.
R
e
ddy,
a
nd
M
.
R
.
A
hm
e
d,
“
P
e
r
s
ona
l
i
z
e
d
di
a
be
t
e
s
di
a
gnos
i
s
us
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng
a
nd
e
l
e
c
t
r
oni
c
he
a
l
t
h
r
e
c
or
ds
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
E
l
e
c
t
r
i
c
al
and
C
om
put
e
r
E
ngi
ne
e
r
i
ng
,
vol
.
14,
no.
4,
pp.
4791
–
4801,
2024,
doi
:
10.11591/
i
j
e
c
e
.v14i
4.pp4791
-
4801.
[
2]
A
.
A
ba
t
a
l
e
t
al
.
,
“
H
ybr
i
d
l
ong
s
hor
t
-
t
e
r
m
m
e
m
or
y
a
nd
de
c
i
s
i
on
t
r
e
e
m
od
e
l
f
or
opt
i
m
i
z
i
ng
pa
t
i
e
nt
vol
um
e
pr
e
di
c
t
i
ons
i
n
e
m
e
r
ge
nc
y
de
pa
r
t
m
e
nt
s
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
E
l
e
c
t
r
i
c
al
and
C
om
put
e
r
E
ngi
ne
e
r
i
ng
,
vol
.
15,
no.
1,
pp.
669
–
676,
2025,
doi
:
10.11591/
i
j
e
c
e
.v15i
1.pp669
-
676.
[
3]
A
. O
.
-
A
c
e
da
ńs
ka
, “
A
c
c
ur
a
c
y of
s
m
a
l
l
a
r
e
a
m
or
t
a
l
i
t
y pr
e
di
c
t
i
on m
e
t
hods
:
e
vi
de
nc
e
f
r
om
P
ol
a
nd,”
J
our
nal
of
P
opul
at
i
on R
e
s
e
ar
c
h
,
vol
. 41, no. 1, 2024, doi
:
10.1007/
s
12546
-
023
-
09326
-
7.
[
4]
L
.
B
ohl
e
n,
J
.
R
os
e
nbe
r
ge
r
,
P
.
Z
s
c
he
c
h,
a
nd
M
.
K
r
a
us
,
“
L
e
ve
r
a
gi
ng
i
nt
e
r
pr
e
t
a
bl
e
m
a
c
hi
ne
l
e
a
r
ni
ng
i
n
i
nt
e
ns
i
ve
c
a
r
e
,”
A
nnal
s
of
O
pe
r
at
i
ons
R
e
s
e
a
r
c
h
, vol
. 347, no. 2, pp. 1093
–
1132, 2025, doi
:
10.1007/
s
1047
9
-
024
-
06226
-
8.
[
5]
A
.
T
.
A
s
l
a
n,
B
.
P
e
r
m
a
na
,
P
.
N
.
A
.
H
a
r
r
i
s
,
K
.
D
.
N
a
i
doo,
M
.
A
.
P
i
e
na
a
r
,
a
nd
A
.
D
.
I
r
w
i
n,
“
T
he
oppor
t
uni
t
i
e
s
a
nd
c
ha
l
l
e
nge
s
f
or
a
r
t
i
f
i
c
i
a
l
i
nt
e
l
l
i
ge
nc
e
t
o
i
m
pr
ove
s
e
p
s
i
s
out
c
om
e
s
i
n
t
he
P
a
e
di
a
t
r
i
c
i
nt
e
n
s
i
v
e
c
a
r
e
uni
t
,”
C
ur
r
e
nt
I
nf
e
c
t
i
ous
D
i
s
e
a
s
e
R
e
po
r
t
s
,
vol
. 25, no. 11, pp. 243
–
253, 2023, doi
:
10.1007/
s
11908
-
023
-
00818
-
4.
[
6]
M
.
R
.
P
i
ns
ky
e
t
al
.
,
“
U
s
e
of
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
ge
nc
e
i
n
c
r
i
t
i
c
a
l
c
a
r
e
:
oppor
t
uni
t
i
e
s
a
nd
obs
t
a
c
l
e
s
,”
C
r
i
t
i
c
al
C
ar
e
,
vol
.
28,
no.
1,
2024,
doi
:
10.1186/
s
13054
-
024
-
04860
-
z.
[
7]
A
.
E
.
W
.
J
ohns
on
e
t
al
.
,
“
M
I
M
I
C
-
I
V
,
a
f
r
e
e
l
y
a
c
c
e
s
s
i
bl
e
e
l
e
c
t
r
oni
c
h
e
a
l
t
h
r
e
c
or
d
da
t
a
s
e
t
,”
Sc
i
e
nt
i
f
i
c
D
at
a
,
vol
.
10,
no.
1,
2023
,
doi
:
10.1038/
s
41597
-
022
-
01899
-
x.
[
8]
K
.
P
a
ng,
L
.
L
i
,
W
.
O
uy
a
ng,
X
.
L
i
u,
a
nd
Y
.
T
a
ng,
“
E
s
t
a
bl
i
s
hm
e
nt
of
I
C
U
m
or
t
a
l
i
t
y
r
i
s
k
pr
e
di
c
t
i
on
m
ode
l
s
w
i
t
h
m
a
c
hi
ne
l
e
a
r
ni
ng
a
l
gor
i
t
hm
us
i
ng M
I
M
I
C
-
I
V
da
t
a
ba
s
e
,”
D
i
agno
s
t
i
c
s
, vol
. 12, no. 5, 2022, doi
:
1
0.3390/
di
a
gnos
t
i
c
s
12051068.
[
9]
C
.
C
.
C
hi
u,
C
.
M
.
W
u,
T
.
N
.
C
hi
e
n,
L
.
J
.
K
a
o,
a
nd
J
.
T
.
Q
i
u,
“
P
r
e
di
c
t
i
ng
t
he
m
or
t
a
l
i
t
y
of
I
C
U
pa
t
i
e
nt
s
by
t
opi
c
m
ode
l
w
i
t
h
m
a
c
hi
ne
-
l
e
a
r
ni
ng t
e
c
hni
que
s
,”
H
e
al
t
hc
a
r
e
, vol
. 10, no. 6, 2022, doi
:
10.3390/
he
a
l
t
hc
a
r
e
10061087.
[
10]
J
.
E
.
C
.
-
C
ogol
l
o,
I
.
B
one
t
,
B
.
G
i
l
,
a
nd
E
.
I
a
da
nz
a
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
m
ode
l
s
f
or
e
a
r
l
y
pr
e
di
c
t
i
on
of
s
e
ps
i
s
on
l
a
r
ge
h
e
a
l
t
hc
a
r
e
da
t
a
s
e
t
s
,
”
E
l
e
c
t
r
oni
c
s
, vol
. 11, no. 9, 2022, doi
:
10.3390/
e
l
e
c
t
r
oni
c
s
11091507.
[
11]
I
.
A
de
n,
C
.
H
.
T
.
C
hi
l
d,
a
nd
C
.
C
.
R
.
-
A
l
da
s
or
o,
“
I
nt
e
r
na
t
i
ona
l
c
l
a
s
s
i
f
i
c
a
t
i
on
of
di
s
e
a
s
e
s
pr
e
di
c
t
i
on
f
r
om
M
I
M
I
I
C
-
I
I
I
c
l
i
ni
c
a
l
t
e
xt
us
i
ng
pr
e
-
t
r
a
i
ne
d
C
l
i
ni
c
a
l
B
E
R
T
a
nd
N
L
P
de
e
p
l
e
a
r
ni
ng
m
ode
l
s
a
c
hi
e
vi
ng
s
t
a
t
e
of
t
he
a
r
t
,”
B
i
g
D
at
a
and
C
ogni
t
i
v
e
C
om
put
i
ng
,
vol
. 8, no. 5, 2024, doi
:
10.3390/
bdc
c
8050047.
[
12]
C
.
B
oz
kur
t
a
nd
T
.
A
ş
ur
oğl
u,
“
M
or
t
a
l
i
t
y
pr
e
di
c
t
i
on
of
va
r
i
ous
c
a
n
c
e
r
pa
t
i
e
nt
s
v
i
a
r
e
l
e
va
nt
f
e
a
t
ur
e
a
na
l
ys
i
s
a
nd
m
a
c
hi
ne
l
e
a
r
ni
ng,”
SN
C
om
put
e
r
S
c
i
e
nc
e
, vol
. 4, no. 3, 2023, doi
:
10.1007/
s
42979
-
023
-
01720
-
5.
[
13]
K
.
O
ga
s
a
w
a
r
a
e
t
al
.
,
“
A
l
ogi
s
t
i
c
r
e
gr
e
s
s
i
on
m
ode
l
f
or
pr
e
di
c
t
i
ng
t
he
r
i
s
k
of
s
ubs
e
que
nt
s
ur
ge
r
y
a
m
ong
pa
t
i
e
nt
s
w
i
t
h
ne
w
l
y
di
a
gnos
e
d
C
r
ohn’
s
di
s
e
a
s
e
u
s
i
ng a
br
ut
e
f
or
c
e
m
e
t
hod,”
D
i
agnos
t
i
c
s
, vol
. 13, n
o. 23, 2023, doi
:
10.3390/
di
a
gnos
t
i
c
s
13233587.
[
14]
Y
.
C
a
o,
M
.
P
.
F
or
s
s
t
e
n,
B
.
S
a
r
a
ni
,
S
.
M
ont
gom
e
r
y,
a
nd
S
.
M
ohs
e
ni
,
“
D
e
ve
l
opm
e
nt
a
nd
va
l
i
da
t
i
on
of
a
n
X
G
B
oo
s
t
-
a
l
gor
i
t
hm
-
pow
e
r
e
d
s
ur
vi
va
l
m
ode
l
f
or
pr
e
di
c
t
i
ng
i
n
-
hos
pi
t
a
l
m
o
r
t
a
l
i
t
y
ba
s
e
d
on
545,388
i
s
ol
a
t
e
d
s
e
ve
r
e
t
r
a
um
a
t
i
c
br
a
i
n
i
nj
ur
y
pa
t
i
e
nt
s
f
r
om
t
he
T
Q
I
P
da
t
a
ba
s
e
,”
J
our
nal
of
P
e
r
s
onal
i
z
e
d M
e
di
c
i
ne
, vol
. 13, no. 9, 2023, doi
:
10.3390/
j
pm
13091401.
[
15]
Q
. A
. H
i
da
y
a
t
ur
r
ohm
a
n a
nd E
. H
a
na
d
a
, “
I
m
pa
c
t
of
da
t
a
pr
e
-
pr
oc
e
s
s
i
ng t
e
c
hni
que
s
on
X
G
B
oo
s
t
m
ode
l
p
e
r
f
or
m
a
nc
e
f
or
pr
e
di
c
t
i
ng
a
l
l
-
c
a
us
e
r
e
a
dm
i
s
s
i
on
a
nd
m
or
t
a
l
i
t
y
a
m
ong
pa
t
i
e
nt
s
w
i
t
h
he
a
r
t
f
a
i
l
u
r
e
,”
B
i
oM
e
dI
nf
or
m
at
i
c
s
,
vol
.
4,
no.
4,
pp.
2201
–
2212,
2024,
doi
:
10.3390/
bi
om
e
di
nf
or
m
a
t
i
c
s
4040118.
[
16]
R
.
E
.
K
o
e
t
a
l
.
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
-
ba
s
e
d
m
or
t
a
l
i
t
y
p
r
e
di
c
t
i
on
m
ode
l
f
or
c
r
i
t
i
c
a
l
l
y
I
l
l
c
a
nc
e
r
pa
t
i
e
nt
s
a
dm
i
t
t
e
d
t
o
t
he
i
nt
e
ns
i
ve
c
a
r
e
uni
t
(
C
a
nI
C
U
)
,”
C
anc
e
r
s
, vol
. 15, no. 3, 2023, doi
:
10.3390/
c
a
nc
e
r
s
15030569.
[
17]
C
.
C
.
C
hung,
E
.
C
.
Y
.
S
u,
J
.
H
.
C
he
n,
Y
.
T
.
C
he
n,
a
nd
C
.
Y
.
K
uo,
“
X
G
B
oos
t
-
ba
s
e
d
s
i
m
pl
e
t
hr
e
e
-
i
t
e
m
m
ode
l
a
c
c
ur
a
t
e
l
y
pr
e
di
c
t
s
out
c
om
e
s
of
a
c
ut
e
i
s
c
h
e
m
i
c
s
t
r
oke
,”
D
i
agnos
t
i
c
s
, vol
. 13, no. 5, 2023, doi
:
10.3
390/
di
a
gnos
t
i
c
s
13050842.
[
18]
F
.
M
uș
a
t
e
t
al
.
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
m
ode
l
s
i
n
s
e
ps
i
s
out
c
om
e
pr
e
di
c
t
i
on
f
or
I
C
U
pa
t
i
e
nt
s
:
i
nt
e
gr
a
t
i
ng
r
out
i
ne
l
a
bor
a
t
or
y
t
e
s
t
s
-
a
s
ys
t
e
m
a
t
i
c
r
e
vi
e
w
,”
B
i
om
e
di
c
i
ne
s
, vol
. 12, no. 12, 2024, doi
:
10.3390/
bi
om
e
di
c
i
ne
s
12122892.
[
19]
N
.
M
.
E
l
s
he
nna
w
y,
D
.
M
.
I
br
a
hi
m
,
A
.
M
.
S
a
r
ha
n,
a
nd
M
.
A
r
a
f
a
,
“
D
e
e
p
-
R
i
s
k:
de
e
p
l
e
a
r
ni
ng
-
ba
s
e
d
m
or
t
a
l
i
t
y
r
i
s
k
pr
e
di
c
t
i
ve
m
ode
l
s
f
or
C
O
V
I
D
-
19,”
D
i
agnos
t
i
c
s
, vol
. 12, no. 8, 2022, doi
:
10.3390/
di
a
gnos
t
i
c
s
12081847.
[
20]
K
.
M
.
M
um
e
ni
n,
P
.
B
i
s
w
a
s
,
M
.
A
.
M
.
K
ha
n,
A
.
S
.
A
l
a
m
m
a
r
y,
a
nd
A
.
A
l
N
a
hi
d,
“
A
m
odi
f
i
e
d
a
qui
l
a
-
ba
s
e
d
opt
i
m
i
z
e
d
X
G
B
oos
t
f
r
a
m
e
w
or
k f
or
de
t
e
c
t
i
ng pr
oba
bl
e
s
e
i
z
ur
e
s
t
a
t
us
i
n ne
ona
t
e
s
,”
S
e
ns
o
r
s
, vol
. 23,
no. 16, 2023, doi
:
10.3390/
s
23167037.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
H
y
br
id
iz
e
d de
e
p l
e
ar
ni
ng m
od
e
l
w
it
h nov
e
l
r
e
c
om
m
e
nd
e
r
f
or
p
r
e
di
c
ti
ng c
r
it
ic
al
it
y
s
ta
te
…
(
Sar
ik
a
K
hope
)
3933
[
21]
A
. O
gunpol
a
, F
. S
a
e
e
d,
S
. B
a
s
ur
r
a
, A
.
M
. A
l
ba
r
r
a
k, a
nd
S
. N
. Q
a
s
e
m
,
“
M
a
c
hi
n
e
l
e
a
r
ni
ng
-
ba
s
e
d pr
e
di
c
t
i
ve
m
od
e
l
s
f
or
de
t
e
c
t
i
on o
f
c
a
r
di
ova
s
c
ul
a
r
di
s
e
a
s
e
s
,
”
D
i
agnos
t
i
c
s
, vol
. 14, no. 2, 2024, doi
:
10.3390/
di
a
gnos
t
i
c
s
14020144.
[
22]
S
.
G
.
P
a
ul
e
t
al
.
,
“
C
om
ba
t
i
ng
C
O
V
I
D
-
19
u
s
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng
a
nd
de
e
p
l
e
a
r
ni
ng:
a
ppl
i
c
a
t
i
ons
,
c
ha
l
l
e
nge
s
,
a
nd
f
ut
ur
e
pe
r
s
pe
c
t
i
ve
s
,”
A
r
r
ay
, vol
. 17, 2023, doi
:
10.1016/
j
.a
r
r
a
y.2022.100271.
[
23]
H
.
E
l
m
a
nna
i
e
t
al
.
,
“
D
i
a
gnos
i
s
m
yoc
a
r
di
a
l
i
nf
a
r
c
t
i
on
ba
s
e
d
on
s
t
a
c
ki
ng
e
ns
e
m
bl
e
of
c
onvol
ut
i
ona
l
ne
ur
a
l
ne
t
w
or
k,”
E
l
e
c
t
r
oni
c
s
,
vol
. 11, no. 23, 2022, doi
:
10.3390/
e
l
e
c
t
r
oni
c
s
11233976.
[
24]
M
.
C
ha
e
,
S
.
H
a
n,
H
.
G
i
l
,
N
.
C
ho,
a
nd
H
.
L
e
e
,
“
P
r
e
di
c
t
i
on
of
i
n
-
hos
pi
t
a
l
c
a
r
di
a
c
a
r
r
e
s
t
us
i
ng
s
ha
l
l
ow
a
nd
de
e
p
l
e
a
r
ni
ng,”
D
i
agnos
t
i
c
s
, vol
. 11, no. 7, 2021, doi
:
10.3390/
di
a
gnos
t
i
c
s
11071255.
[
25]
S
.
R
.
K
hope
a
nd
S
.
E
l
i
a
s
,
“
S
i
m
pl
i
f
i
e
d
a
nd
nove
l
pr
e
di
c
t
i
ve
m
ode
l
u
s
i
ng
f
e
a
t
u
r
e
e
ngi
ne
e
r
i
ng
ove
r
M
I
M
I
C
-
I
I
I
da
t
a
s
e
t
,”
P
r
oc
e
di
a
C
om
put
e
r
Sc
i
e
n
c
e
, vol
. 218, pp. 1968
–
1976, 2022, doi
:
10.1016/
j
.pr
oc
s
.2023.01.173.
[
26]
S
.
R
.
K
hope
a
nd
S
.
E
l
i
a
s
,
“
S
t
r
a
t
e
gi
e
s
of
pr
e
di
c
t
i
ve
s
c
he
m
e
s
a
nd
c
l
i
ni
c
a
l
di
a
gn
os
i
s
f
or
pr
ognos
i
s
u
s
i
ng
M
I
M
I
C
-
I
I
I
:
A
s
ys
t
e
m
a
t
i
c
r
e
vi
e
w
,”
H
e
al
t
hc
ar
e
, vol
. 11, no. 5, 2023, doi
:
10.3390/
he
a
l
t
hc
a
r
e
11050710.
[
27]
S
.
R
.
K
hope
a
nd
S
.
E
l
i
a
s
,
“
C
r
i
t
i
c
a
l
c
or
r
e
l
a
t
i
on
of
pr
e
di
c
t
or
s
f
or
a
n
e
f
f
i
c
i
e
nt
r
i
s
k
pr
e
di
c
t
i
on
f
r
a
m
e
w
or
k
of
I
C
U
pa
t
i
e
nt
us
i
n
g
c
or
r
e
l
a
t
i
on
a
nd
t
r
a
ns
f
or
m
a
t
i
on
of
M
I
M
I
C
-
I
I
I
da
t
a
s
e
t
,”
D
at
a
Sc
i
e
nc
e
and
E
ngi
ne
e
r
i
ng
,
vol
.
7,
no.
1,
pp.
71
–
86,
2022,
doi
:
10.1007/
s
41019
-
022
-
00176
-
6.
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Sarika
Khope
is
an
Assistant
Professor
in
the
Department
of
Electronics
and
Telecommunication
at G.H. R
aisoni
College
of Engin
eering and
Mana
gement,
Pune, wi
th over
two
decades
of
academic
experience
since
2001.
She
specializes
i
n
artificial
intell
igence,
machine
learning,
and
deep
learning
and
holds
an
M
.
Tech
.
in
Elect
ronics
from
Rashtrasant
Tukadoji
Maharaj
Nagpur
University
and
a
Ph.D.
from
VIT
Chennai
.
She
is
the
Institution’s
Innovation
Council
(IIC)
Convener
since
2019,
Innovation
Activity
C
oordinator
at
GHRCEM,
and the N
odal Offi
cer for NIRF
.
She
has secured two Indian pate
n
ts
a
nd one Australian patent,
registered
over
95
literary
copyrights
,
and
published
more
than
15
research
articles.
As
a
passionate
educator
and
researcher
,
she
actively
fosters
innovation
a
nd
academic
excellence.
She ca
n be c
ontact
ed at
email:
sarika
.khope
@
gmail.c
om
.
Deepali Kotambk
ar
is curren
tly working a
s
Assistant Prof
essor in
Departme
nt of
Electronics
Engineering,
Ramdeobaba
University,
Nagpur,
India
(Formerly,
Shri
Ramdeobaba
College
of
Enginee
ring
and
Manage
ment
,
Nagpur).
She
has
done
her
M.Tech.
in
Electronics
and
carried
out
her
doctoral
research
in
Wireless
Communication.
Her
area
of
research
are
digital
image
processing
and
wireless
communi
cation.
She
has
p
ublicati
ons
in
national
-
internationa
l
confere
nces
and
reputed
journals.
She
can
be
contacted
at
email:
shelkedt@
rknec.edu
.
Rama
Vasantha
Adiraju
is
an
Assistant
Professor
in
the
Department
of
Electronics
and
Communication
Engineering
at
Aditya
University,
Surampalem,
Kakinada,
with
over
thirteen
years
of
experience
in
teaching.
She
specializ
es
in
medical
imaging,
artificial
intell
igence,
machine
learning,
and
deep
learning
and
hold
s
an
M
.
Tech
.
in
System
and
Signal
Processing
from
JNTU
University,
Hyderabad
and
a
Ph.D.
from
VIT
Chennai.
She
is
the
Internal
Compli
ance
Commit
tee
(
ICC
)
Convener
during
2016
t
o
2023.
She
has
secured
published
more
than
15
research
articles.
As
a
passionate
educator
and
researcher
,
she
activel
y
fosters
innovatio
n
and
academic
excellence.
She
can
be
contacted
at
email:
vasantha.adiraju@acet.ac.in
.
Smita
Suhas
Battalwar
is
Assistant
Professor
at
the
D
epartment
of
Artifici
al
Intelligen
ce
and
Machine
Learning
of
the
G
.
H
.
Raisoni
Colle
ge
of
Enginee
ring
and
Management,
Pune.
Her
researc
h
is
situated
in
the
field
of
database
s
ystems
with
a
special
focus
on
data
mining
and
machine
learning.
She
can
be
contacted
at
email:
smitabattalwar@gmail.com
.
Evaluation Warning : The document was created with Spire.PDF for Python.