I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
14
,
No
.
2
,
A
p
r
il
2
0
2
5
,
p
p
.
1
0
8
7
~
1
0
9
5
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
14
.i
2
.
p
p
1
0
8
7
-
1
0
9
5
1087
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Unv
eiling
precisi
o
n:
Ey
e cancer
de
tect
io
n
rede
fined
with
pa
rticle
swa
rm
optimiza
tion a
nd
genetic alg
o
rithms
S
a
n
v
e
d
N
a
r
w
a
d
k
a
r
1
,
P
r
a
d
n
y
a
S
a
m
i
t
M
e
h
t
a
2
,
R
u
t
u
j
a
R
a
j
e
n
d
r
a
P
a
t
i
l
3
,
K
a
l
y
a
n
i
K
a
d
a
m
4
,
V
i
j
a
y
k
u
m
a
r
B
i
d
v
e
5
1
D
e
p
a
r
t
m
e
n
t
o
f
I
n
f
o
r
m
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
V
i
s
h
w
a
k
a
r
m
a
I
n
s
t
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
m
e
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
-
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
V
i
s
h
w
a
k
a
r
m
a
I
n
s
t
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
3
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
-
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
& M
a
c
h
i
n
e
Le
a
r
n
i
n
g
,
V
i
s
h
w
a
k
a
r
ma
I
n
st
i
t
u
t
e
o
f
I
n
f
o
r
ma
t
i
o
n
Te
c
h
n
o
l
o
g
y
,
P
u
n
e
,
I
n
d
i
a
4
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
,
V
i
sh
w
a
k
a
r
ma
U
n
i
v
e
r
s
i
t
y
,
P
u
n
e
,
I
n
d
i
a
5
S
c
h
o
o
l
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
I
n
f
o
r
mat
i
o
n
Te
c
h
n
o
l
o
g
y
,
S
y
mb
i
o
s
i
s S
k
i
l
l
s a
n
d
P
r
o
f
e
ssi
o
n
a
l
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
Ma
y
6
,
2
0
2
4
R
ev
is
ed
No
v
7
,
2
0
2
4
Acc
ep
ted
No
v
1
4
,
2
0
2
4
Ey
e
c
a
n
c
e
r
d
e
tec
ti
o
n
is
ra
re
.
Th
e
stu
d
y
i
n
tro
d
u
c
e
s
a
h
o
li
sti
c
sw
a
rm
in
telli
g
e
n
c
e
m
e
th
o
d
fo
r
th
e
ti
m
e
l
y
id
e
n
t
ifi
c
a
ti
o
n
a
n
d
c
a
teg
o
riza
ti
o
n
o
f
th
re
e
sig
n
ifi
c
a
n
t
e
y
e
d
iso
r
d
e
rs:
g
lau
c
o
m
a
,
d
ia
b
e
ti
c
re
ti
n
o
p
a
t
h
y
,
a
n
d
c
a
tara
c
t.
G
lau
c
o
m
a
is
d
isti
n
g
u
ish
e
d
b
y
e
lev
a
ted
p
re
ss
u
re
with
i
n
th
e
e
y
e
a
n
d
h
a
rm
to
th
e
o
p
ti
c
n
e
rv
e
,
p
o
ten
ti
a
ll
y
lea
d
in
g
to
p
e
rm
a
n
e
n
t
l
o
ss
o
f
v
isio
n
.
Dia
b
e
ti
c
p
a
ti
e
n
ts
e
x
p
e
rien
c
e
d
iab
e
ti
c
re
ti
n
o
p
a
t
h
y
p
rima
ril
y
d
u
e
t
o
th
e
p
r
e
se
n
c
e
o
f
h
ig
h
b
lo
o
d
su
g
a
r
lev
e
ls.
T
h
e
e
a
rly
d
e
tec
ti
o
n
a
n
d
c
las
sifica
ti
o
n
o
f
c
a
t
a
ra
c
ts
c
a
n
b
e
a
c
h
ie
v
e
d
b
y
c
o
m
b
i
n
i
n
g
sw
a
rm
in
telli
g
e
n
c
e
a
lg
o
rit
h
m
s
su
c
h
a
s
p
a
rti
c
le
sw
a
rm
o
p
ti
m
iza
ti
o
n
(
P
S
O)
a
n
d
g
e
n
e
ti
c
a
l
g
o
rit
h
m
s
(G
A).
In
th
e
c
a
se
o
f
d
iab
e
ti
c
re
ti
n
o
p
a
th
y
d
iag
n
o
sis,
sw
a
rm
in
telli
g
e
n
c
e
is
e
m
p
lo
y
e
d
to
o
p
ti
m
ize
th
e
p
a
ra
m
e
ters
o
f
d
e
e
p
lea
rn
in
g
m
o
d
e
ls,
th
e
re
b
y
e
n
h
a
n
c
in
g
t
h
e
a
c
c
u
ra
c
y
o
f
les
io
n
se
g
m
e
n
tati
o
n
a
n
d
c
las
sifica
ti
o
n
.
Ca
tara
c
t
d
e
tec
ti
o
n
u
se
d
to
imp
ro
v
e
th
e
e
v
a
lu
a
ti
o
n
o
f
le
n
s
o
p
a
c
it
y
a
n
d
c
lo
u
d
in
e
ss
,
p
r
o
v
i
d
i
n
g
a
ro
b
u
st
d
iag
n
o
stic
m
e
c
h
a
n
ism
.
T
h
e
a
c
c
u
ra
c
y
o
b
tain
e
d
with
a
P
S
O
is
8
5
.
7
9
%
,
F
1
sc
o
re
8
3
.
4
5
%
,
a
n
d
re
c
a
ll
8
2
.
4
3
%
.
Th
e
a
c
c
u
ra
c
y
o
b
tai
n
e
d
wit
h
a
GA
is
8
2
.
1
0
%
,
F
1
s
c
o
re
8
1
.
1
6
%
,
a
n
d
r
e
c
a
ll
8
1
.
5
1
%
.
T
h
e
c
o
m
p
a
riso
n
o
f
G
A,
c
o
n
v
o
lu
ti
o
n
n
e
u
ra
l
n
e
two
rk
,
a
n
d
P
S
O
a
lg
o
ri
th
m
s
p
ro
v
e
s
th
a
t
th
e
a
c
c
u
ra
c
y
to
d
e
tec
t
th
e
e
y
e
c
a
n
c
e
r
is ac
h
iev
e
d
with
P
S
O
a
n
d
G
A alg
o
rit
h
m
.
K
ey
w
o
r
d
s
:
D
e
e
p
l
e
a
r
n
i
n
g
E
y
e
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
G
e
n
e
t
i
c
a
l
g
o
r
i
t
h
m
s
P
a
r
t
i
c
l
e
s
w
a
r
m
o
p
t
i
m
i
z
a
t
i
o
n
S
w
a
r
m
i
n
t
e
l
l
i
g
e
n
c
e
V
G
G
-
1
6
C
N
N
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Pra
d
n
y
a
Sam
it M
eh
ta
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
in
ee
r
in
g
-
Ar
tific
ial
I
n
tellig
en
ce
,
Sav
itrib
ai
Ph
u
le
Pu
n
e
Un
iv
er
s
ity
Ko
n
d
h
wa
(
B
u
d
r
u
k
)
,
Pu
n
e
–
4
1
1
0
4
8
,
Ma
h
ar
ash
tr
a
,
I
n
d
ia
E
m
ail:
p
r
ad
n
y
a1
7
.
m
eh
ta@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
h
ea
lth
ca
r
e
lan
d
s
ca
p
e
is
i
n
a
co
n
s
tan
t
s
tate
o
f
ev
o
lu
tio
n
,
d
r
iv
en
b
y
th
e
n
ee
d
f
o
r
ac
c
u
r
ate
an
d
ea
r
ly
d
etec
tio
n
m
eth
o
d
s
to
im
p
r
o
v
e
p
atien
t
o
u
tco
m
es.
I
n
th
e
d
o
m
i
n
io
n
o
f
o
p
to
m
etr
is
t,
th
e
id
en
tific
atio
n
an
d
v
er
d
ict
o
f
ey
e
ca
n
ce
r
s
ar
e
cr
u
cial
f
r
o
n
tier
s
th
at
d
em
an
d
tim
ely
in
ter
v
en
tio
n
an
d
p
e
r
s
o
n
alize
d
tr
ea
tm
en
t
ap
p
r
o
ac
h
es.
T
h
e
f
ield
o
f
ey
e
c
an
ce
r
,
wh
ich
in
clu
d
es
d
if
f
er
en
t
ty
p
es
o
f
m
alig
n
an
t
tu
m
o
r
s
af
f
ec
tin
g
th
e
tis
s
u
es
o
f
th
e
ey
e,
n
ec
ess
itates
a
p
r
ec
is
e
an
d
d
ef
in
ed
ap
p
r
o
ac
h
to
d
iag
n
o
s
is
.
T
r
ad
itio
n
al
m
eth
o
d
s
o
f
ten
s
tr
u
g
g
le
to
ac
h
iev
e
th
e
n
ec
ess
ar
y
lev
el
o
f
s
en
s
itiv
ity
an
d
s
p
ec
if
icity
f
o
r
ea
r
ly
d
etec
tio
n
.
B
y
in
c
o
r
p
o
r
ati
n
g
s
war
m
in
tellig
en
ce
,
s
p
ec
if
ically
p
ar
ticle
s
war
m
o
p
tim
izatio
n
(
PS
O)
an
d
g
e
n
etic
alg
o
r
ith
m
s
(
GA)
,
a
n
ew
co
m
p
u
tatio
n
al
f
r
am
ewo
r
k
em
er
g
es
th
at
h
as
th
e
p
o
ten
tial
to
co
m
p
letely
tr
an
s
f
o
r
m
th
e
d
iag
n
o
s
tic
lan
d
s
ca
p
e
b
y
o
p
tim
izin
g
th
e
d
etec
tio
n
p
r
o
c
ess
.
Fig
u
r
e
1
d
ep
icts
th
e
n
o
r
m
al
ey
e
r
etin
a
im
ag
e.
G
lau
co
m
a,
ca
tar
ac
t,
d
iab
etic
r
etin
o
p
ath
y
,
an
d
o
th
er
well
-
k
n
o
wn
o
c
u
lar
illn
ess
es
h
av
e
estab
lis
h
ed
th
eir
wo
r
ld
wid
e
m
an
if
estatio
n
.
T
h
e
wo
r
ld
is
co
n
f
r
o
n
ted
with
g
la
u
co
m
a
as
p
r
im
ar
y
f
ac
to
r
co
n
t
r
ib
u
tin
g
to
b
lin
d
n
ess
[
1
]
.
Glau
co
m
a
ca
n
lead
to
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
14
,
No
.
2
,
Ap
r
il 2
0
2
5
:
1
0
8
7
-
1
0
9
5
1088
d
am
ag
e
in
t
h
e
o
p
tic
n
er
v
e,
wh
ich
p
lay
s
a
cr
u
cial
r
o
le
in
tr
a
n
s
m
itti
n
g
v
is
u
al
in
f
o
r
m
atio
n
f
r
o
m
th
e
r
etin
a
to
th
e
b
r
ain
.
T
h
is
d
am
ag
e
ca
n
r
esu
lt
in
th
e
lo
s
s
o
f
th
e
v
is
u
al
f
iel
d
,
an
d
if
n
o
t
ad
d
r
ess
ed
,
p
o
te
n
tial
b
lin
d
n
ess
[
2
]
.
Glau
co
m
a
is
a
co
n
d
itio
n
t
h
at
in
clu
d
es
d
if
f
er
e
n
t
s
u
b
ty
p
es
with
p
r
im
ar
y
o
p
en
-
a
n
g
le
g
lau
co
m
a
(
POAG)
a
n
d
an
g
le
-
clo
s
u
r
e
g
lau
co
m
a
(
AC
G)
b
ein
g
t
h
e
m
o
s
t c
o
m
m
o
n
f
o
r
m
s
.
Fig
u
r
e
1
.
No
r
m
al
ey
e
r
atin
al
im
ag
e
POAG
d
ev
elo
p
s
g
r
ad
u
ally
,
wh
ile
AC
G
p
r
esen
t
s
m
o
r
e
ac
u
tely
d
u
e
to
a
s
u
d
d
en
b
lo
c
k
ag
e
o
f
th
e
ey
e'
s
d
r
ain
ag
e
an
g
le.
T
h
e
r
e
ar
e
s
ev
e
r
al
f
ac
to
r
s
t
h
at
ca
n
co
n
tr
ib
u
te
to
th
e
d
ev
elo
p
m
en
t
o
f
ca
tar
ac
t
s
,
in
clu
d
in
g
ag
i
n
g
,
s
m
o
k
in
g
,
r
ad
iatio
n
ex
p
o
s
u
r
e,
d
iab
etes,
an
d
o
th
er
ca
u
s
es
[
3
]
.
B
y
in
co
r
p
o
r
atin
g
PS
O
an
d
G
A
in
to
th
e
d
etec
tio
n
p
ip
elin
e,
th
e
c
o
m
p
u
tatio
n
al
f
r
am
ewo
r
k
g
ai
n
s
th
e
ca
p
ab
ilit
y
to
ad
ap
tiv
ely
o
p
tim
ize
b
o
th
f
e
atu
r
e
s
elec
tio
n
an
d
d
iag
n
o
s
tic
m
o
d
el
p
ar
am
eter
s
.
C
o
n
s
eq
u
en
tly
,
th
is
lead
s
to
a
s
ig
n
if
ican
t im
p
r
o
v
em
e
n
t in
th
e
o
v
er
all
ac
cu
r
ac
y
o
f
ey
e
ca
n
ce
r
d
etec
tio
n
,
ev
e
n
wh
en
d
ea
lin
g
with
th
e
ch
allen
g
in
g
s
ce
n
ar
io
o
f
ca
tar
ac
t
-
af
f
ec
ted
ey
es.
T
h
is
in
n
o
v
ativ
e
a
p
p
r
o
a
ch
h
o
ld
s
g
r
e
at
p
r
o
m
is
e
in
f
ac
ilit
atin
g
m
o
r
e
ef
f
ec
tiv
e
an
d
p
r
ec
is
e
ea
r
ly
d
i
ag
n
o
s
is
,
p
o
ten
tially
r
esu
ltin
g
in
e
n
h
an
ce
d
o
u
tco
m
e
s
f
o
r
in
d
iv
id
u
als
at
r
is
k
o
f
ey
e
ca
n
ce
r
with
c
o
n
cu
r
r
en
t
ca
tar
ac
ts
[
4
]
.
Hig
h
b
lo
o
d
s
u
g
ar
ca
n
b
e
attr
ib
u
ted
t
o
a
r
an
g
e
o
f
f
ac
to
r
s
,
s
u
ch
as
in
s
u
f
f
icien
t
in
s
u
lin
p
r
o
d
u
ctio
n
o
r
i
n
ad
eq
u
ate
ce
llu
lar
r
esp
o
n
s
e
to
in
s
u
lin
[
5
]
.
T
o
a
v
er
t
ey
esig
h
t
d
am
ag
e,
it
is
im
p
er
io
u
s
to
s
en
s
e
th
e
in
f
ec
tio
n
ti
m
ely
o
n
as
it
o
f
te
n
g
o
es
u
n
n
o
ticed
u
n
til
th
e
later
s
tag
es.
T
h
e
elev
ate
d
s
u
g
ar
le
v
els
h
av
e
a
d
et
r
im
en
tal
e
f
f
ec
t
o
n
th
e
b
lo
o
d
v
ess
els
with
in
th
e
r
etin
al
tis
s
u
es
[
6
]
.
T
h
e
class
if
icatio
n
is
s
h
o
wn
in
F
ig
u
r
e
2
f
o
r
d
iab
etic
r
etin
o
p
ath
y
.
I
t
in
cl
u
d
es
non
-
p
r
o
life
r
ativ
e
d
iab
etic
r
etin
o
p
ath
y
(
NPDR
)
an
d
p
r
o
life
r
ativ
e
d
iab
etic
r
etin
o
p
ath
y
(
P
DR
)
as
its
two
m
ain
ty
p
es
s
h
o
wn
in
F
ig
u
r
e
s
3
an
d
4
r
esp
ec
tiv
el
y
.
B
lo
ck
ag
es
o
f
s
ig
n
if
ican
ce
m
a
n
if
est
with
in
th
e
b
l
o
o
d
v
ess
els,
ca
u
s
in
g
a
d
ec
lin
e
in
th
e
s
u
p
p
ly
o
f
b
lo
o
d
to
th
e
m
em
b
r
a
n
e.
T
h
e
e
v
o
lu
tio
n
o
f
n
ewf
an
g
led
p
lasma
n
er
v
es,
k
n
o
wn
as
n
eo
v
ascu
lar
izatio
n
,
ca
n
o
cc
u
r
o
n
th
e
r
etin
a'
s
s
u
r
f
ac
e
o
r
with
in
th
e
v
itre
o
u
s
g
el.
T
h
e
v
itre
o
u
s
g
el
is
a
clea
r
,
g
el
-
lik
e
s
u
b
s
tan
ce
th
at
f
i
lls
th
e
ey
e
[
7
]
,
[
8
]
.
T
h
e
o
cc
u
r
r
en
ce
o
f
an
ey
e
ca
ta
r
ac
t
is
a
cr
itical
is
s
u
e
as
it
ca
u
s
es
th
e
len
s
to
b
ec
o
m
e
cl
o
u
d
y
o
r
l
o
s
e
its
tr
an
s
p
ar
en
cy
,
r
esu
ltin
g
in
v
is
u
al
im
p
air
m
en
t
as
d
e
p
icted
in
Fig
u
r
e
5
.
E
ar
l
y
an
d
p
r
ec
is
e
d
i
ag
n
o
s
is
o
f
ca
tar
ac
ts
ca
n
s
ig
n
if
ican
tly
en
h
a
n
ce
th
e
well
-
b
ein
g
o
f
p
atien
ts
with
PS
O
an
d
GA
alg
o
r
ith
m
[
9
]
,
[
1
0
]
.
As
g
lau
c
o
m
a
ad
v
an
ce
s
,
it
lead
s
to
ir
r
e
v
er
s
ib
le
b
lin
d
n
ess
an
d
b
r
in
g
s
ab
o
u
t
n
o
tab
le
s
tr
u
ct
u
r
al
alter
ati
o
n
s
.
W
ith
in
d
ig
ital
f
u
n
d
u
s
im
ag
es,
th
e
o
p
tic
d
is
c
ac
ts
as
th
e
g
atew
ay
f
o
r
b
l
o
o
d
v
ess
els
an
d
o
p
tic
n
e
r
v
e
f
i
b
er
s
to
en
ter
th
e
r
etin
a.
I
t
is
v
is
u
ally
d
is
tin
g
u
is
h
ab
le
as
a
lu
m
in
o
u
s
o
v
al
r
eg
io
n
.
Mo
r
eo
v
er
,
th
e
o
p
tic
c
u
p
ca
n
b
e
r
ec
o
g
n
ize
d
as
a
liv
elier
o
v
ate
r
an
g
e
p
o
s
itio
n
ed
at
th
e
ce
n
ter
o
f
th
e
o
p
tic
d
is
c
[
1
1
]
.
W
ith
in
d
i
g
ital
f
u
n
d
u
s
i
m
ag
es,
th
e
o
p
tic
d
is
c
ac
ts
as
th
e
g
atew
ay
f
o
r
p
lasma
v
ess
els
an
d
o
p
tic
v
ess
el
f
ib
er
s
to
p
ass
in
th
e
lay
er
o
f
ce
lls
.
I
t
is
v
is
u
ally
d
is
tin
g
u
is
h
ab
le
as
a
lu
m
in
o
u
s
o
v
al
r
eg
io
n
.
Mo
r
eo
v
e
r
,
th
e
o
p
tic
cu
p
ca
n
b
e
r
ec
o
g
n
ize
d
as
a
b
r
ig
h
ter
ellip
tical
r
eg
io
n
p
o
s
itio
n
ed
at
th
e
ce
n
te
r
o
f
th
e
o
p
tic
d
is
c
.
T
h
is
r
esear
ch
en
d
ea
v
o
r
s
ee
k
s
to
u
n
r
av
el
th
e
in
tr
icate
d
etails
o
f
ey
e
ca
n
ce
r
d
etec
tio
n
b
y
p
r
esen
tin
g
a
s
o
p
h
is
ticated
b
len
d
o
f
co
m
p
u
tatio
n
al
in
tellig
en
ce
an
d
o
p
h
th
alm
ic
ex
p
er
tis
e.
B
y
in
co
r
p
o
r
atin
g
PS
O
an
d
GA
in
to
th
e
d
iag
n
o
s
t
ic
p
ip
elin
e,
o
u
r
g
o
al
is
to
en
h
an
ce
th
e
p
r
e
cisi
o
n
,
ef
f
icien
cy
,
an
d
ea
r
ly
d
etec
tio
n
ca
p
ab
ilit
ies,
th
er
eb
y
m
a
k
in
g
s
ig
n
if
ican
t
co
n
tr
i
b
u
tio
n
s
to
th
e
well
-
b
ein
g
o
f
p
atien
ts
.
T
h
e
jo
u
r
n
e
y
to
war
d
s
ac
h
iev
in
g
ac
cu
r
ac
y
in
e
y
e
ca
n
ce
r
d
etec
tio
n
is
d
r
iv
en
b
y
th
e
in
teg
r
atio
n
o
f
cu
ttin
g
-
ed
g
e
tec
h
n
o
lo
g
ies,
an
d
th
is
s
tu
d
y
aim
s
to
s
h
ed
lig
h
t
o
n
a
p
ath
to
war
d
s
a
f
u
tu
r
e
wh
er
e
ea
r
ly
in
ter
v
en
tio
n
is
s
y
n
o
n
y
m
o
u
s
w
ith
im
p
r
o
v
ed
p
r
o
g
n
o
s
is
an
d
a
n
en
h
a
n
ce
d
q
u
ality
o
f
life
f
o
r
in
d
iv
id
u
als
at
r
is
k
o
f
o
r
af
f
ec
ted
b
y
e
y
e
ca
n
ce
r
.
Fig
u
r
e
2
.
T
y
p
es o
f
d
iab
etic
r
et
in
o
p
ath
y
Fig
u
r
e
3
.
Mild
NPDR
[
8
]
a
n
d
PDR
d
iab
etic
r
etin
o
p
ath
y
[
9
]
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Un
ve
ilin
g
p
r
ec
is
io
n
:
E
ye
ca
n
c
er d
etec
tio
n
r
ed
efin
ed
w
ith
p
a
r
ticle
s
w
a
r
m
…
(
S
a
n
ve
d
N
a
r
w
a
d
ka
r
)
1089
Fig
u
r
e
4
.
Diab
etic
r
etin
o
p
ath
y
(
NPDR
an
d
PDR
im
ag
es)
[
1
2
]
Fig
u
r
e
5
.
C
atar
ac
t im
ag
es
2.
L
I
T
E
R
AT
U
RE
SU
RVE
Y
T
h
e
p
r
im
ar
y
m
eth
o
d
o
lo
g
y
em
p
lo
y
ed
was
tr
an
s
f
er
lear
n
in
g
,
a
tech
n
iq
u
e
wh
er
e
p
r
e
-
tr
ai
n
ed
m
o
d
els
o
n
h
u
g
e
s
et
o
f
d
ata
ar
e
a
d
ap
ted
to
a
s
p
ec
if
ic
task
f
o
r
th
e
class
if
icatio
n
o
f
ey
e
ca
n
ce
r
.
T
h
ese
ar
ch
itectu
r
es
ar
e
r
en
o
wn
ed
f
o
r
th
eir
a
d
ep
tn
ess
in
ex
tr
ac
tin
g
in
t
r
icate
f
ea
tu
r
es
f
r
o
m
im
a
g
es,
r
e
n
d
er
in
g
t
h
em
id
ea
l
f
o
r
m
ed
ical
im
ag
e
an
aly
s
is
.
B
y
u
tili
zin
g
p
r
e
-
tr
ain
ed
m
o
d
els,
t
h
e
n
etwo
r
k
ca
n
ca
p
italize
o
n
th
e
k
n
o
wled
g
e
ac
q
u
ir
ed
f
r
o
m
ex
ten
s
iv
e
d
atasets
;
th
er
eb
y
en
h
an
cin
g
p
er
f
o
r
m
an
ce
ev
e
n
wh
en
co
n
f
r
o
n
ted
with
r
elativ
ely
s
m
aller
m
ed
ical
d
atasets
[
1
3
]
.
T
h
e
id
en
tific
ati
o
n
o
f
ey
e
s
k
in
is
m
a
d
e
ea
s
ier
with
th
e
in
tr
o
d
u
ctio
n
o
f
an
au
to
m
ated
tech
n
iq
u
e.
A
co
n
v
o
lu
tio
n
al
n
e
u
r
al
n
etwo
r
k
(
C
NN)
is
u
til
ized
,
alo
n
g
with
a
g
r
ey
v
ictim
izatio
n
co
n
v
er
s
io
n
; to
en
h
an
ce
th
e
r
eso
lu
tio
n
o
f
th
e
im
ag
es.
T
h
e
ac
cu
r
ac
y
was
9
2
.
5
%
[
1
4
]
.
T
h
is
m
o
d
el
was
s
p
ec
if
ically
tailo
r
ed
to
wo
r
k
with
s
eg
m
en
ted
th
e
s
p
h
er
ical
ca
p
s
u
le
th
at
en
clo
s
es
th
e
ey
e
o
f
a
v
er
teb
r
ate
im
ag
es.
B
y
lev
er
ag
in
g
th
e
h
o
u
g
h
cir
cle
tr
an
s
f
o
r
m
atio
n
,
ac
cu
r
ate
p
r
ed
i
ctio
n
s
o
f
b
o
th
th
e
s
p
h
er
ical
e
n
clo
s
u
r
e
o
f
e
y
e
an
d
ir
is
r
eg
io
n
s
wer
e
ac
h
ie
v
ed
.
T
h
e
ef
f
icac
io
u
s
test
in
g
o
f
th
is
p
r
ac
tice
y
ield
ed
an
im
p
r
ess
iv
e
ac
cu
r
ac
y
r
ate
o
f
9
5
%,
a
f
f
ir
m
in
g
its
v
iab
ilit
y
f
o
r
r
ea
l
-
tim
e
ap
p
licatio
n
s
[
1
5
]
.
B
y
an
aly
zin
g
im
ag
es
o
f
f
u
n
d
u
s
an
d
ex
tr
ac
tin
g
e
n
u
m
er
at
ed
im
ag
er
ies
o
p
ted
th
r
o
u
g
h
o
p
tical
c
o
h
er
en
ce
t
o
m
o
g
r
a
p
h
y
(
OC
T
)
d
ata,
w
h
ich
ev
er
i
n
d
iv
id
u
ally
o
r
else
in
g
r
o
u
p
in
g
,
th
e
r
esear
ch
er
s
d
ev
is
ed
an
m
ec
h
an
ized
,
o
b
jectiv
e,
m
ac
h
in
e
lear
n
in
g
tech
n
iq
u
e
f
o
r
d
ia
g
n
o
s
in
g
g
lau
c
o
m
a.
R
em
ar
k
ab
ly
,
th
e
g
r
o
u
p
in
g
m
eth
o
d
ex
h
ib
ited
an
ar
ea
u
n
d
er
th
e
cu
r
v
e
(
AUC
)
o
f
0
.
9
6
3
,
s
u
g
g
esti
n
g
its
h
eig
h
ten
ed
s
en
s
itiv
ity
in
d
ete
ctin
g
g
la
u
co
m
a
at
its
ea
r
l
y
s
t
ag
es
[
1
6
]
.
T
h
e
u
tili
za
tio
n
o
f
a
r
tific
ial
in
tellig
en
ce
(
AI
)
ap
p
licatio
n
s
in
g
lau
co
m
a
h
ad
r
ev
o
lu
tio
n
ized
th
e
f
ield
b
y
p
r
o
v
id
in
g
a
co
m
p
r
eh
en
s
i
v
e
o
v
er
v
iew
o
f
t
h
e
d
is
ea
s
e.
T
h
ese
AI
to
o
ls
h
av
e
p
r
o
v
e
n
to
b
e
in
s
tr
u
m
en
tal
in
ea
r
ly
d
etec
tio
n
,
p
er
s
o
n
alize
d
t
r
ea
tm
en
t
p
lan
n
in
g
,
an
d
en
h
an
cin
g
clin
ical
d
ec
is
io
n
-
m
ak
in
g
[
1
7
]
.
T
h
e
m
o
d
e
l
o
u
t
p
e
r
f
o
r
m
e
d
f
i
v
e
C
N
N
m
o
d
el
s
t
h
a
t
h
a
d
a
l
r
e
a
d
y
b
e
e
n
p
r
e
-
t
r
a
i
n
e
d
:
M
o
b
i
l
eN
e
t
,
V
G
G
-
1
6
,
VGG
-
1
9
,
I
n
c
e
p
ti
o
n
-
v
3
,
a
n
d
R
es
N
e
t
-
5
0
.
C
at
a
r
a
c
tN
e
t
a
s
s
a
u
l
t
e
d
t
h
e
c
u
r
r
e
n
t
s
ta
t
e
-
of
-
t
h
e
-
ar
t
c
a
t
a
r
a
c
t
d
et
e
c
ti
o
n
t
e
c
h
n
i
q
u
e
s
o
n
t
h
e
b
a
s
is
o
f
c
o
r
r
e
c
t
n
e
s
s
(
9
9
.
1
3
%
)
,
p
r
e
c
i
s
i
o
n
(
9
9
.
0
8
%
)
,
r
e
c
a
l
l
(
9
9
.
0
7
%
)
,
s
p
e
c
i
f
i
c
it
y
(
9
9
.
1
7
%
)
,
M
a
t
t
h
e
ws
c
o
r
r
e
la
t
i
o
n
c
o
e
f
f
ic
i
en
t
(
M
C
C
)
(
9
8
.
2
3
%
)
,
a
n
d
f
1
-
s
co
r
e
(
9
9
.
0
7
%
)
[
1
8
]
.
I
n
t
h
i
s
r
es
ea
r
c
h
,
D
e
n
s
eN
e
t
-
161,
R
esNet
-
1
5
2
,
an
d
R
esNet
-
1
0
1
m
o
d
els
wer
e
em
p
lo
y
e
d
.
E
v
er
y
o
r
d
er
in
g
n
etwo
r
k
in
c
o
r
p
o
r
ates
a
en
d
u
r
in
g
r
esp
o
n
s
iv
en
ess
s
ec
tio
n
.
T
h
e
ex
p
er
im
en
tal
o
u
tco
m
es,
o
b
tai
n
ed
f
r
o
m
th
e
b
e
n
ch
m
a
r
k
B
-
s
ca
n
ey
e
u
ltra
s
o
u
n
d
d
ataset,
r
ev
ea
l
th
at
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
h
as
th
e
ab
ilit
y
to
s
elec
tiv
ely
em
p
h
asis
o
n
th
e
tailo
r
ed
ar
ea
s
o
f
ca
tar
ac
t
in
th
e
e
y
eb
all,
lead
i
n
g
to
a
n
ac
cu
r
ac
y
o
f
9
7
.
5
%
[
1
9
]
.
I
n
th
is
ar
ticle,
th
e
co
n
s
u
m
p
tio
n
o
f
p
r
etr
ain
ed
m
o
d
els
s
u
ch
as
I
n
ce
p
tio
n
V
3
,
I
n
ce
p
tio
n
R
esNetV2
,
Xce
p
tio
n
,
an
d
Den
s
en
et1
2
1
en
ab
les
th
e
im
p
lem
en
tatio
n
o
f
a
co
m
p
u
ter
ized
ca
tar
ac
t
an
al
y
s
is
m
eth
o
d
.
W
ith
an
ex
ce
p
tio
n
al
test
ac
cu
r
ac
y
o
f
9
8
.
1
7
%,
a
s
en
s
itiv
ity
o
f
9
7
%,
an
d
a
s
p
ec
if
icity
o
f
1
0
0
%,
th
e
I
n
ce
p
tio
n
R
esNetV2
m
o
d
el
h
a
s
ex
ce
ed
ed
p
r
ev
io
u
s
b
e
n
ch
m
a
r
k
s
.
I
ts
o
u
ts
tan
d
in
g
p
er
f
o
r
m
an
ce
s
ets
a
n
ew
s
tan
d
ar
d
in
th
e
f
ield
o
f
ca
tar
ac
t
d
is
ea
s
e
id
en
tific
atio
n
,
s
o
lid
if
y
in
g
its
p
o
s
itio
n
as
th
e
s
tate
-
of
-
th
e
-
ar
t
s
o
lu
tio
n
[
2
0
]
.
T
h
e
au
th
o
r
p
r
o
p
o
s
ed
u
tili
za
tio
n
o
f
th
e
b
i
n
ar
y
class
if
icatio
n
m
eth
o
d
i.e
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
-
b
ased
class
if
icatio
n
wh
ich
ef
f
ec
tiv
ely
id
en
tifie
s
p
r
o
ce
s
s
es.
T
h
is
lead
s
to
a
p
er
f
ec
t
class
if
icatio
n
o
f
b
o
th
n
o
r
m
al
an
d
ca
tar
ac
t
ey
es
u
s
in
g
SVM.
Mo
r
eo
v
er
,
d
u
r
in
g
th
e
ey
e
d
at
a
test
in
g
s
tag
e,
th
e
an
aly
s
is
p
r
o
d
u
ce
s
a
class
if
ic
atio
n
r
ate
o
f
7
2
.
5
%
f
o
r
n
o
r
m
al
ey
es
an
d
8
2
.
5
%
f
o
r
ca
t
ar
ac
t
ey
es
[
2
1
]
.
T
h
e
ac
cu
r
aten
ess
o
f
ey
e
b
u
g
d
ia
g
n
o
s
is
h
as
b
ee
n
s
ig
n
i
f
ican
tly
im
p
r
o
v
ed
th
r
o
u
g
h
th
e
u
s
e
o
f
m
i
x
ed
ad
ap
tiv
e
tr
an
s
m
u
tatio
n
s
war
m
o
p
tim
iz
atio
n
an
d
r
eg
r
ess
io
n
n
eu
r
al
n
etwo
r
k
(
AE
D
-
HSR
)
tech
n
iq
u
es
,
wh
ich
h
av
e
b
ee
n
au
to
m
ated
b
y
th
e
r
esear
c
h
er
.
T
h
e
s
y
s
tem
h
as
ac
h
iev
ed
a
p
r
ed
ictio
n
ac
cu
r
ac
y
o
f
9
8
.
0
8
%,
alo
n
g
with
im
p
r
ess
iv
e
s
p
ec
if
icity
(
9
9
.
3
4
%)
an
d
s
en
s
itiv
ity
(
9
8
.
0
3
%)
r
ates.
Ad
d
itio
n
ally
,
th
e
p
o
s
itiv
e
p
r
ed
ictiv
e
v
alu
e
(
PP
V)
an
d
n
eg
ativ
e
p
r
e
d
ictiv
e
v
alu
e
(
NPV)
s
tan
d
at
9
8
.
0
3
%
an
d
9
9
.
3
4
%
r
esp
ec
tiv
ely
.
T
h
e
f
alse
p
o
s
itiv
e
r
ate
(
FP
R
)
is
o
n
ly
0
.
6
2
%,
wh
ile
th
e
f
alse
n
eg
ativ
e
r
ate
(
FNR
)
is
1
.
9
3
%.
Fu
r
th
er
m
o
r
e,
th
e
F
1
s
co
r
e
r
ea
ch
es
an
im
p
r
ess
iv
e
9
8
.
6
7
%,
with
a
f
alse
d
is
co
v
er
y
r
ate
(
FDR
)
o
f
1
.
9
6
%.
I
n
co
m
p
ar
is
o
n
to
o
th
er
tech
n
iq
u
es
lik
e
r
eg
r
ess
io
n
n
e
u
r
al
n
etwo
r
k
(
R
NN)
-
PS
O,
R
N
N
-
GA,
an
d
R
NN,
th
e
p
r
o
p
o
s
ed
m
eth
o
d
s
co
n
s
is
ten
tly
y
ield
s
u
p
er
io
r
r
esu
lts
.
T
h
e
p
r
o
p
o
s
e
d
m
o
d
el
o
u
t
p
er
f
o
r
m
s
th
e
p
ar
a
m
eter
s
lik
e
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
PP
V,
NPV,
FP
R
,
FNR
,
F1
s
co
r
e,
an
d
FDR
[
2
2
]
.
T
h
e
s
tate
-
of
-
th
e
-
a
r
t
n
u
clea
r
ca
tar
ac
t
class
if
icatio
n
r
esu
lts
ar
e
ac
h
iev
e
d
b
y
th
e
s
u
g
g
ested
f
ea
tu
r
e
ex
tr
ac
tio
n
-
b
ased
co
n
tex
t,
as
d
em
o
n
s
tr
ated
b
y
th
e
an
aly
s
is
o
f
th
e
a
n
ter
i
o
r
s
eg
m
en
t
o
p
tical
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
14
,
No
.
2
,
Ap
r
il 2
0
2
5
:
1
0
8
7
-
1
0
9
5
1090
co
h
er
en
ce
to
m
o
g
r
a
p
h
y
(
AS
-
OC
T
)
im
ag
e
d
ataset.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
th
is
f
r
am
ewo
r
k
s
u
r
p
ass
es
th
at
o
f
cu
ttin
g
-
ed
g
e
m
eth
o
d
s
an
d
d
ee
p
lear
n
in
g
m
eth
o
d
s
[
2
3
]
.
T
h
e
u
tili
za
tio
n
o
f
th
e
g
r
o
wth
r
eg
io
n
tech
n
i
q
u
e
in
a
s
eg
m
en
tatio
n
ap
p
r
o
ac
h
is
d
em
o
n
s
tr
ated
in
th
is
s
tu
d
y
.
T
h
e
m
o
d
el
in
co
r
p
o
r
ates
b
o
th
th
e
f
u
zz
y
C
-
m
ea
n
s
(
FC
M)
an
d
GA
m
eth
o
d
s
,
with
th
e
o
b
jectiv
e
o
f
d
iag
n
o
s
in
g
d
iab
etes
b
ased
o
n
an
g
io
g
r
a
p
h
y
im
ag
es
o
f
p
atien
ts
'
ey
es.
T
h
e
GA
class
if
icatio
n
m
eth
o
d
d
em
o
n
s
tr
ated
s
u
p
er
i
o
r
p
er
f
o
r
m
an
ce
in
ter
m
s
o
f
o
v
e
r
all
r
esu
lts
.
T
o
f
u
r
th
er
en
h
a
n
ce
GA
s
eg
m
en
tatio
n
,
it
is
n
ec
ess
ar
y
to
in
co
r
p
o
r
ate
ad
d
i
tio
n
al
alg
o
r
ith
m
s
wh
ile
p
r
ese
r
v
in
g
t
h
e
in
h
e
r
en
t
ch
a
r
ac
ter
is
tics
o
f
th
e
p
r
im
ar
y
m
ed
ical
an
g
io
g
r
ap
h
y
im
ag
es
[
2
4
]
.
T
h
e
GA
-
FC
M
m
eth
o
d
p
r
o
v
ed
t
o
b
e
s
u
p
er
io
r
t
o
th
e
h
a
n
d
m
eth
o
d
wh
e
n
it
ca
m
e
to
s
elec
tin
g
in
itial
p
o
in
ts
.
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
d
e
m
o
n
s
tr
ated
a
s
en
s
itiv
ity
o
f
0
.
7
8
.
C
o
m
p
ar
in
g
th
e
f
u
zz
y
f
itn
ess
f
u
n
ctio
n
in
GA
with
o
th
er
tech
n
iq
u
es,
it
was
ev
id
en
t
th
at
th
e
ap
p
r
o
ac
h
in
tr
o
d
u
ce
d
i
n
th
is
s
tu
d
y
is
h
ig
h
ly
s
u
itab
le
f
o
r
th
e
J
ac
c
ar
d
in
d
ex
.
T
h
is
ap
p
r
o
ac
h
n
o
t
o
n
ly
o
f
f
e
r
ed
th
e
lo
west
J
ac
ca
r
d
d
is
tan
ce
b
u
t
also
p
r
o
v
id
e
d
th
e
h
ig
h
est
J
ac
ca
r
d
v
alu
es
[
2
4
]
.
T
h
e
p
r
o
b
ab
ilis
tic
n
eu
r
al
n
etwo
r
k
(
PNN)
class
if
i
er
,
wh
en
s
u
b
jecte
d
to
th
r
ee
f
o
ld
cr
o
s
s
-
v
alid
atio
n
,
ex
h
ib
ited
an
a
v
er
ag
e
class
if
icatio
n
ac
cu
r
ac
y
o
f
9
6
.
1
5
%.
Ad
d
itio
n
ally
,
it
d
em
o
n
s
tr
ated
a
s
en
s
itiv
ity
o
f
9
6
.
2
7
% a
n
d
a
s
p
ec
if
icity
o
f
9
6
.
0
8
% f
o
r
σ
=
0
.
0
1
0
4
[
2
5
]
.
T
h
e
f
o
r
em
o
s
t g
o
al
o
f
th
e
r
esear
ch
er
an
d
r
esear
ch
was
to
p
r
esen
t
a
PS
O
m
o
d
el
d
ev
elo
p
ed
t
o
en
h
an
ce
th
e
o
p
t
im
izatio
n
o
f
h
y
p
e
r
p
ar
am
eter
s
,
esp
ec
ially
th
e
lear
n
in
g
r
ate
an
d
m
o
m
e
n
tu
m
d
u
r
i
n
g
th
e
tr
an
s
f
er
lear
n
in
g
.
T
h
e
f
o
cu
s
is
o
n
ap
p
ly
in
g
tr
an
s
f
er
lear
n
in
g
m
eth
o
d
o
lo
g
i
es
to
f
in
e
-
tu
n
e
Ma
s
k
r
eg
io
n
-
b
ased
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
et
wo
r
k
(
R
-
C
NN
)
f
o
r
o
b
ject
d
etec
tio
n
s
eg
m
e
n
tatio
n
,
with
a
p
ar
ticu
lar
em
p
h
asis
o
n
th
e
f
u
n
d
u
s
im
ag
e
d
atasets
[
2
6
]
.
T
h
e
d
ataset,
wh
ich
h
as
b
ee
n
s
o
u
r
ce
d
f
r
o
m
Kag
g
le
[
2
7
]
,
co
n
tain
s
a
to
tal
o
f
4
3
0
0
au
th
en
tic
co
m
p
u
ted
to
m
o
g
r
a
p
h
y
(
C
T
)
s
ca
n
im
a
g
es.
T
h
is
d
ataset
en
co
m
p
ass
es
r
etin
al
im
ag
es
th
at
s
h
o
wca
s
e
s
ev
er
al
ey
e
cir
cu
m
s
tan
ce
s
s
u
ch
as
n
o
r
m
al,
d
iab
etic
r
etin
o
p
ath
y
,
ca
tar
a
ct,
an
d
g
lau
co
m
a
.
T
h
e
ca
teg
o
r
izatio
n
o
f
th
ese
im
ag
es wa
s
d
o
n
e
b
y
co
n
s
id
er
i
n
g
f
ac
to
r
s
lik
e
C
T
s
ca
n
im
ag
e
s
an
d
th
e
s
p
ec
if
ic
ty
p
e
o
f
e
y
e
d
is
ea
s
e
d
etec
ted
.
3.
P
RO
P
O
SE
D
M
E
T
H
O
DO
L
O
G
Y
T
h
is
r
esear
ch
p
r
o
ject
u
tili
ze
s
co
n
tem
p
o
r
ar
y
d
ee
p
lear
n
in
g
,
s
o
f
t
co
m
p
u
tin
g
o
p
tim
izatio
n
,
an
d
im
a
g
e
p
r
o
ce
s
s
in
g
tech
n
i
q
u
es
to
d
ev
e
lo
p
an
in
n
o
v
ativ
e
ap
p
r
o
ac
h
f
o
r
au
to
m
atin
g
th
e
c
ateg
o
r
izatio
n
o
f
ey
e
d
is
ea
s
es.
T
h
e
p
r
o
p
o
s
ed
r
esear
ch
is
b
u
i
lt
u
p
o
n
a
m
eticu
lo
u
s
ly
ch
o
s
e
n
d
ataset
th
at
en
co
m
p
ass
es
a
wid
e
r
an
g
e
o
f
ey
e
co
n
d
itio
n
im
ag
es.
T
o
en
s
u
r
e
o
p
tim
al
f
ea
tu
r
e
ex
tr
ac
tio
n
,
ea
ch
im
ag
e
is
s
u
b
jecte
d
t
o
a
co
m
p
r
eh
e
n
s
iv
e
p
r
ep
r
o
ce
s
s
in
g
wo
r
k
f
l
o
w
th
at
en
co
m
p
ass
es
r
esizin
g
,
co
n
v
er
tin
g
c
o
lo
r
s
p
ac
es,
e
n
h
an
cin
g
co
n
t
r
ast,
an
d
d
etec
tin
g
ed
g
es.
B
y
in
teg
r
atin
g
th
ese
p
r
o
ce
d
u
r
es,
a
d
ataset
is
ab
u
n
d
a
n
t
in
f
ea
tu
r
es,
p
la
y
in
g
a
v
ital
r
o
le
in
tr
ain
in
g
a
r
o
b
u
s
t
class
if
ica
tio
n
m
o
d
el.
C
o
n
v
er
t
th
e
u
p
d
ate
d
p
o
s
itio
n
s
o
f
p
ar
ticles
in
to
b
in
ar
y
v
alu
es
u
s
in
g
b
in
ar
y
e
n
co
d
i
n
g
,
w
h
ich
will
e
f
f
ec
tiv
ely
r
e
p
r
esen
t
f
ea
tu
r
e
in
clu
s
io
n
o
r
e
x
clu
s
io
n
.
T
o
ac
h
iev
e
th
is
,
a
th
r
esh
o
ld
ca
n
b
e
estab
lis
h
ed
to
co
n
v
er
t
th
e
r
ea
l
-
v
alu
ed
p
o
s
itio
n
s
in
to
b
in
ar
y
f
o
r
m
at.
T
h
e
o
p
tim
ized
s
et
o
f
f
ea
tu
r
es
f
o
r
th
e
g
iv
en
o
b
jectiv
e
f
u
n
ctio
n
is
o
b
tain
ed
b
y
e
x
tr
ac
tin
g
th
e
f
in
al
f
ea
tu
r
e
s
u
b
s
et
f
r
o
m
th
e
p
ar
ticle
with
th
e
h
ig
h
est
ev
alu
atio
n
in
th
e
s
wa
r
m
.
Sp
litt
in
g
an
d
en
c
o
d
in
g
:
t
o
ac
cu
r
ately
ass
ess
th
e
m
o
d
el'
s
p
er
f
o
r
m
an
ce
,
we
em
p
lo
y
a
s
tr
atif
ied
tech
n
iq
u
e
t
o
d
iv
id
e
th
e
d
ataset
in
to
t
r
ain
i
n
g
an
d
test
in
g
s
ets,
p
r
eser
v
in
g
th
e
d
is
tr
ib
u
tio
n
o
f
d
if
f
er
en
t
class
es.
Fu
r
th
er
m
o
r
e
,
we
u
tili
ze
a
lab
el
en
co
d
er
to
co
n
v
er
t
ca
teg
o
r
ical
d
is
ea
s
e
la
b
els
in
to
n
u
m
er
ical
v
alu
es.
T
h
is
en
co
d
i
n
g
p
lay
s
a
cr
u
cial
r
o
le
in
t
r
ain
in
g
a
n
eu
r
a
l
n
etwo
r
k
as
it
en
ab
les
th
e
m
o
d
el
to
co
m
p
r
eh
e
n
d
an
d
lear
n
f
r
o
m
th
e
lab
eled
in
p
u
t e
f
f
ec
tiv
ely
.
3
.
1
.
Sy
s
t
e
m
d
esig
n
Me
th
o
d
o
lo
g
y
is
b
u
ilt
u
p
o
n
a
s
p
ec
ially
d
esig
n
ed
C
NN
as
d
ep
icted
in
F
ig
u
r
e
6
.
C
NNs
ex
ce
l
in
im
ag
e
class
if
icatio
n
d
u
e
to
th
eir
ab
i
lity
to
au
to
m
atica
lly
e
x
tr
ac
t
h
ier
ar
ch
ical
in
f
o
r
m
atio
n
f
r
o
m
im
ag
es.
T
h
e
f
o
ca
l
p
o
in
t
o
f
th
e
ey
e
d
is
ea
s
e
d
etec
tio
n
s
y
s
tem
ce
n
ter
s
o
n
a
b
esp
o
k
e
C
NN
m
o
d
el.
T
h
is
m
o
d
el
h
as
b
ee
n
in
tr
icate
ly
en
g
in
ee
r
ed
to
p
r
o
f
icien
tly
e
x
tr
ac
t
s
alien
t
f
ea
tu
r
es
f
r
o
m
ey
e
im
ag
es
an
d
p
r
ec
is
ely
ca
teg
o
r
ize
th
em
in
t
o
d
if
f
er
en
t
d
is
ea
s
e
class
if
icatio
n
s
.
T
h
e
m
o
d
el
ar
c
h
itectu
r
e
en
ta
ils
o
f
co
n
v
o
lu
tio
n
al
lay
er
s
f
o
r
f
ea
tu
r
e
ex
t
r
ac
tio
n
,
m
ax
-
p
o
o
lin
g
lay
er
s
f
o
r
d
o
w
n
s
am
p
lin
g
,
an
d
d
en
s
e
lay
er
s
f
o
r
class
if
icatio
n
.
T
h
e
q
u
an
tity
o
f
f
ilter
s
an
d
lay
er
s
f
in
e
-
tu
n
e
t
o
m
ax
im
ize
th
e
m
o
d
el'
s
p
er
f
o
r
m
an
ce
.
PS
O:
i
n
o
u
r
m
o
d
el
o
p
tim
izatio
n
,
we
lev
er
ag
e
PS
O.
T
h
is
tech
n
iq
u
e
aid
s
in
f
in
e
-
t
u
n
in
g
th
e
m
o
d
el'
s
p
ar
am
eter
s
b
y
s
im
u
latin
g
th
e
b
eh
a
v
io
r
o
f
a
s
war
m
o
f
p
ar
ticles.
PS
O
o
p
tim
izes
th
e
m
o
d
el'
s
p
er
f
o
r
m
an
c
e
b
y
iter
ativ
ely
ad
ju
s
tin
g
th
e
p
ar
am
eter
s
b
ased
o
n
th
eir
in
d
iv
id
u
al
an
d
co
llectiv
e
ex
p
e
r
ien
ce
s
,
lead
in
g
to
im
p
r
o
v
e
d
r
esu
lts
.
PS
O
o
b
jectiv
e
f
u
n
ctio
n
:
T
h
e
PS
O
o
b
jectiv
e
f
u
n
ctio
n
i
n
(
1
)
is
in
ten
d
ed
to
cu
r
tail
th
e
cla
s
s
if
icatio
n
er
r
o
r
.
I
t r
esh
ap
es a
n
d
s
ets th
e
weig
h
ts
o
f
th
e
C
NN
m
o
d
el
b
a
s
ed
o
n
p
ar
ticle
p
o
s
itio
n
s
.
(
,
,
)
=
1
−
(
1
)
W
h
er
e:
p
is
th
e
p
ar
ticle
p
o
s
itio
n
,
r
ep
r
esen
tin
g
f
latten
ed
w
eig
h
ts
o
f
th
e
C
NN
m
o
d
el,
X
is
th
e
in
p
u
t
d
ata
(
f
ea
tu
r
es)
,
an
d
y
is
th
e
tr
u
e
cla
s
s
lab
els.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Un
ve
ilin
g
p
r
ec
is
io
n
:
E
ye
ca
n
c
er d
etec
tio
n
r
ed
efin
ed
w
ith
p
a
r
ticle
s
w
a
r
m
…
(
S
a
n
ve
d
N
a
r
w
a
d
ka
r
)
1091
PS
O
u
p
d
ate
eq
u
atio
n
s
:
T
h
e
PS
O
u
p
d
ate
in
(
2
)
an
d
(
3
)
g
o
v
er
n
t
h
e
m
o
v
em
en
t
o
f
p
a
r
ticles
in
th
e
s
o
lu
tio
n
s
p
ac
e.
+
1
=
.
+
1
.
1
.
(
−
(
)
+
2
.
2
.
−
(
2
)
+
1
=
+
.
+
1
(
3
)
W
h
er
e
Vik
is
th
e
r
ap
id
ity
o
f
t
h
e
p
ar
ticle
in
d
im
en
s
io
n
i
at
r
eiter
atio
n
k
A=
πr
^2
,
Xik
is
th
e
p
o
in
t
o
f
th
e
p
ar
ticle
in
d
im
en
s
io
n
s
i
at
iter
atio
n
k
,
w
is
th
e
in
er
tia
weig
h
t,
C
1
a
n
d
C
2
ar
e
th
e
c
o
g
n
itiv
e
an
d
s
o
cial
co
ef
f
icien
ts
,
r
1
an
d
r
2
a
r
e
r
an
d
o
m
n
u
m
b
er
s
b
etwe
en
0
an
d
1
,
p
b
est
i
is
th
e
b
est
p
o
s
itio
n
o
f
th
e
p
ar
ticle
in
d
im
en
s
io
n
i
s
o
f
ar
,
g
b
est i is th
e
b
est p
o
s
itio
n
am
o
n
g
all
p
ar
ticles in
d
im
en
s
i
o
n
I
[
2
8
]
.
GA
a
lg
o
r
it
h
m
:
GA
,
w
h
i
c
h
is
a
p
o
we
r
f
u
l
o
p
ti
m
iz
ati
o
n
m
e
th
o
d
.
T
h
e
o
b
je
cti
v
e
o
f
u
til
izi
n
g
GAs
i
n
t
h
e
r
e
co
g
n
it
io
n
o
f
ey
e
ca
n
ce
r
is
t
o
a
u
t
o
m
at
ica
ll
y
el
ec
t
a
s
u
b
g
r
o
u
p
o
f
s
i
g
n
i
f
ic
an
t
s
tr
u
ct
u
r
es
f
r
o
m
a
la
r
g
e
r
c
o
ll
ec
ti
o
n
.
T
h
is
p
r
o
c
ess
e
n
h
a
n
c
es
t
h
e
ef
f
ic
ie
n
c
y
an
d
ef
f
e
cti
v
e
n
ess
o
f
t
h
e
m
a
ch
in
e
l
ea
r
n
in
g
m
o
d
el
i
n
a
cc
u
r
at
el
y
d
i
f
f
er
e
n
ti
ati
n
g
b
etw
ee
n
ca
n
c
er
o
u
s
a
n
d
n
o
n
-
c
an
ce
r
o
u
s
ca
s
es.
I
t
is
c
r
u
ci
al
t
o
f
i
n
e
-
t
u
n
e
th
e
GA
p
ar
a
m
et
er
s
a
n
d
v
al
id
at
e
t
h
e
m
o
d
el
u
s
in
g
d
iv
er
s
e
d
at
ase
ts
i
n
o
r
d
e
r
t
o
ac
h
ie
v
e
r
o
b
u
s
t
a
n
d
g
e
n
er
ali
za
b
le
o
u
tc
o
m
es
.
Ob
jectiv
e
f
u
n
ctio
n
:
T
h
e
p
r
im
ar
y
g
o
al
o
f
th
e
GA
o
b
jectiv
e
f
u
n
ctio
n
is
to
r
ed
u
ce
th
e
cla
s
s
if
icatio
n
er
r
o
r
b
y
m
o
d
if
y
in
g
th
e
C
NN
m
o
d
el'
s
weig
h
ts
u
s
in
g
GA
p
ar
am
eter
s
.
T
h
is
p
r
o
ce
s
s
in
v
o
lv
es
r
esh
ap
in
g
th
e
m
o
d
el
to
e
n
h
an
ce
its
p
er
f
o
r
m
a
n
ce
in
class
if
y
in
g
d
ata
ac
c
u
r
at
ely
.
(
)
=
1
−
(
4
)
W
h
er
e:
p
is
th
e
in
d
iv
id
u
al
in
th
e
GA
p
o
p
u
latio
n
,
r
e
p
r
esen
tin
g
f
latten
ed
weig
h
ts
o
f
th
e
C
NN
m
o
d
el
;
an
d
A
is
ac
cu
r
ac
y
.
GA
o
p
er
atio
n
s
: T
h
e
GA
in
v
o
l
v
es th
r
ee
m
ain
o
p
er
atio
n
s
: selec
tio
n
,
cr
o
s
s
o
v
er
,
an
d
m
u
tatio
n
.
Selectio
n
: Sele
ctio
n
is
ty
p
ically
b
ased
o
n
f
itn
ess
,
f
av
o
r
in
g
in
d
iv
id
u
als with
h
ig
h
e
r
f
itn
ess
v
alu
es.
C
r
o
s
s
o
v
er
:
C
r
o
s
s
o
v
er
is
a
r
ep
r
o
d
u
ctiv
e
m
ec
h
a
n
is
m
th
at
i
n
v
o
lv
es
th
e
m
er
g
in
g
o
f
g
en
etic
m
ater
ial
f
r
o
m
two
p
ar
en
ts
,
lead
in
g
to
th
e
g
en
er
at
io
n
o
f
o
f
f
s
p
r
i
n
g
.
Mu
tatio
n
:
Mu
tatio
n
in
tr
o
d
u
ce
s
s
m
all
r
an
d
o
m
ch
an
g
es
in
an
in
d
iv
id
u
al'
s
g
en
etic
m
ater
ial
to
m
ain
tain
d
iv
er
s
ity
in
th
e
p
o
p
u
latio
n
.
Fig
u
r
e
6
.
Sy
s
tem
ar
c
h
itectu
r
e
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
cu
lm
in
atio
n
o
f
o
u
r
r
es
ea
r
ch
en
d
ea
v
o
r
s
h
as
y
ield
e
d
p
r
o
m
is
in
g
o
u
tco
m
es
in
th
e
f
ield
o
f
au
to
m
ated
illn
ess
class
if
icat
io
n
f
o
r
ey
e
d
is
ea
s
es.
T
h
r
o
u
g
h
t
h
e
in
teg
r
atio
n
o
f
ad
v
a
n
ce
d
i
m
ag
e
p
r
o
ce
s
s
in
g
,
a
cu
s
to
m
-
b
u
ilt
C
NN,
an
d
o
p
ti
m
izatio
n
tech
n
iq
u
es,
o
u
r
co
m
p
r
eh
en
s
iv
e
s
o
l
u
tio
n
h
as
es
tab
lis
h
ed
a
r
o
b
u
s
t
f
r
am
ewo
r
k
f
o
r
ac
c
u
r
ate
d
is
ea
s
e
id
en
tific
atio
n
.
Up
o
n
t
h
e
co
n
clu
s
io
n
o
f
th
e
tr
ain
in
g
p
h
ase,
th
e
C
NN
s
h
o
wca
s
ed
o
u
ts
tan
d
in
g
p
er
f
o
r
m
an
ce
o
n
th
e
test
s
et.
T
h
e
ef
f
icac
y
o
f
t
h
e
m
o
d
el
is
ev
id
e
n
t
in
its
ab
ilit
y
to
g
en
er
alize
well
to
u
n
f
am
iliar
d
ata.
T
o
f
u
r
th
er
ev
alu
ate
it
s
p
er
f
o
r
m
an
ce
as
s
h
o
wn
i
n
T
ab
le
1
,
ad
d
itio
n
al
m
etr
ics
s
u
ch
as
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
s
co
r
e,
wh
ich
o
f
f
er
v
alu
ab
le
in
s
ig
h
ts
in
to
th
e
m
o
d
el'
s
p
er
f
o
r
m
a
n
ce
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
14
,
No
.
2
,
Ap
r
il 2
0
2
5
:
1
0
8
7
-
1
0
9
5
1092
ac
r
o
s
s
d
if
f
er
en
t
class
es.
T
h
ese
m
ea
s
u
r
em
en
ts
o
f
f
er
a
c
o
m
p
r
eh
en
s
iv
e
o
v
e
r
v
iew
o
f
th
e
m
o
d
el'
s
s
tr
en
g
th
s
an
d
wea
k
n
ess
es,
p
r
o
v
id
in
g
v
alu
ab
l
e
in
f
o
r
m
atio
n
f
o
r
f
u
t
u
r
e
en
h
an
ce
m
en
ts
.
T
ab
le
1
.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
C
NN
A
c
c
u
r
a
c
y
F
1
sc
o
r
e
R
e
c
a
l
l
0
.
6
9
4
3
1
2
0
.
6
7
9
7
2
3
0
.
6
8
9
6
4
7
T
h
e
tr
ain
in
g
d
y
n
am
ics
wer
e
v
is
u
alize
d
u
s
in
g
g
r
ap
h
s
th
at
d
ep
icted
th
e
ac
cu
r
ac
y
an
d
lo
s
s
v
er
s
u
s
ep
o
ch
i
n
F
ig
u
r
e
s
7
(
a)
to
7
(
d
)
.
T
h
ese
ch
ar
ts
ef
f
ec
tiv
el
y
s
h
o
w
ca
s
ed
th
e
d
y
n
am
ic
n
atu
r
e
o
f
th
e
m
o
d
el'
s
lear
n
in
g
p
r
o
ce
s
s
.
B
y
an
aly
zin
g
th
e
ac
cu
r
ac
y
cu
r
v
e,
we
co
u
ld
id
e
n
tify
s
ig
n
if
ican
t
p
er
io
d
s
o
f
lear
n
in
g
as
well
as
p
o
ten
tial
o
v
er
f
itti
n
g
.
C
o
n
v
er
s
ely
,
th
e
lo
s
s
cu
r
v
es
p
r
o
v
id
ed
v
alu
ab
le
in
s
ig
h
ts
in
to
th
e
m
o
d
el'
s
co
n
v
er
g
en
c
e
ten
d
en
cies a
n
d
its
ab
ilit
y
to
m
i
n
im
ize
class
if
icatio
n
er
r
o
r
s
.
T
h
e
co
n
f
u
s
io
n
m
at
r
ix
as
d
e
p
icted
in
Fig
u
r
e
8
u
s
ed
as
as
a
t
o
o
l
in
class
if
icatio
n
ev
alu
atio
n
,
wh
ich
is
to
b
e
u
tili
ze
d
to
illu
s
tr
ate
th
e
ca
teg
o
r
izatio
n
p
atter
n
s
o
f
th
e
m
o
d
el.
T
h
ese
m
atr
ices
d
is
p
la
y
th
e
co
u
n
ts
o
f
t
r
u
e
p
o
s
itiv
es,
tr
u
e
n
eg
ativ
es,
f
alse
p
o
s
itiv
es,
an
d
f
alse
n
e
g
ativ
es
f
o
r
ea
c
h
class
.
B
y
an
aly
zin
g
th
ese
m
atr
ices,
we
g
ain
ed
a
d
ee
p
e
r
u
n
d
er
s
tan
d
in
g
o
f
th
e
m
o
d
el'
s
p
er
f
o
r
m
a
n
ce
as
s
h
o
wn
in
Fig
u
r
e
9
,
p
ar
tic
u
lar
ly
in
in
s
tan
ce
s
wh
er
e
m
is
class
if
icat
io
n
s
o
cc
u
r
r
ed
.
T
h
is
v
alu
ab
le
d
ata
p
lay
s
a
cr
u
cial
r
o
le
in
en
h
an
c
in
g
th
e
m
o
d
el
an
d
ad
d
r
ess
in
g
s
p
ec
if
ic
c
h
allen
g
es
ass
o
ciate
d
with
d
if
f
er
e
n
t
ey
e
c
o
n
d
itio
n
s
[
2
9
]
.
Fig
u
r
e
9
s
p
ec
if
ies
th
e
co
m
p
ar
ativ
e
an
aly
s
is
o
f
all
t
h
e
u
s
ed
alg
o
r
ith
m
s
in
p
r
o
p
o
s
ed
s
y
s
tem
.
T
h
e
ac
cu
r
ac
y
o
f
ey
e
ca
n
ce
r
d
etec
tio
n
is
ac
h
iev
ed
b
y
PS
O
alg
o
r
ith
m
w
ith
a
g
r
ea
ter
ac
cu
r
ac
y
co
m
p
ar
ed
to
GA
an
d
C
NN.
T
h
is
ad
v
an
ce
m
en
t
is
cr
itical
in
clin
ical
s
ettin
g
s
wh
er
e
p
r
o
m
p
t
an
d
p
r
o
p
er
th
er
a
p
y
ca
n
e
n
h
an
ce
p
atien
t
o
u
tc
o
m
es
th
r
o
u
g
h
ea
r
ly
an
d
p
r
ec
is
e
id
en
tific
atio
n
o
f
ey
e
ca
n
ce
r
.
Fas
t
p
r
o
ce
s
s
in
g
r
ates
ca
n
s
p
e
ed
u
p
d
iag
n
o
s
is
an
d
tr
ea
tm
e
n
t,
th
er
ef
o
r
e
th
is
is
esp
ec
ially
cr
u
cial
in
clin
ical
s
ettin
g
s
.
PS
O
an
d
GA,
two
o
f
o
u
r
o
p
ti
m
izatio
n
tech
n
iq
u
es,
wer
e
ess
en
tial in
h
elp
in
g
to
r
e
f
in
e
th
e
m
o
d
el.
PS
O
d
y
n
am
ically
ch
an
g
ed
th
e
weig
h
ts
o
f
th
e
n
eu
r
al
n
etwo
r
k
,
ef
f
ec
tiv
ely
ex
p
lo
r
in
g
th
e
weig
h
t
s
p
ac
e
an
d
en
h
an
cin
g
p
e
r
f
o
r
m
an
ce
[
3
0
]
.
On
th
e
o
th
er
h
an
d
,
GA,
i
n
f
lu
e
n
ce
d
b
y
n
atu
r
al
s
elec
tio
n
,
cr
ea
ted
a
p
o
p
u
latio
n
o
f
v
iab
le
s
o
lu
tio
n
s
,
f
u
r
th
er
r
e
f
in
in
g
t
h
e
m
o
d
el'
s
ac
cu
r
ac
y
.
T
h
e
m
o
d
el'
s
class
if
icati
o
n
ab
ilit
ies
wer
e
s
y
n
er
g
is
tically
im
p
r
o
v
e
d
b
y
th
e
in
ter
ac
tio
n
b
etwe
en
th
ese
o
p
tim
izatio
n
m
eth
o
d
s
an
d
th
e
C
NN
d
esig
n
.
(
a)
(
b
)
(
c)
(
d
)
Fig
u
r
e
7
.
T
h
e
tr
ain
in
g
g
r
a
p
h
s
o
f
(
a)
ac
c
u
r
ac
y
v
s
ep
o
c
h
,
(
b
)
l
o
s
s
v
s
ep
o
ch
,
(
c
)
v
alid
atio
n
ac
cu
r
ac
y
v
s
e
p
o
ch
,
an
d
(
d
)
lo
s
s
v
s
ep
o
ch
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Un
ve
ilin
g
p
r
ec
is
io
n
:
E
ye
ca
n
c
er d
etec
tio
n
r
ed
efin
ed
w
ith
p
a
r
ticle
s
w
a
r
m
…
(
S
a
n
ve
d
N
a
r
w
a
d
ka
r
)
1093
Fig
u
r
e
8
.
C
o
n
f
u
s
io
n
m
atr
i
x
C
NN
Fig
u
r
e
9
.
C
o
m
p
a
r
is
o
n
of
ev
alu
atio
n
m
etr
ics o
f
C
NN,
PS
O,
GA
s
o
f
t o
p
tim
izatio
n
tech
n
iq
u
e
s
5.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
p
r
esen
ts
a
n
o
v
el
ap
p
r
o
ac
h
f
o
r
ac
cu
r
ate
an
d
ef
f
ici
en
t
d
etec
tio
n
o
f
cr
itical
ey
e
d
i
s
o
r
d
er
s
b
y
in
teg
r
atin
g
PS
O
an
d
GA
with
in
th
e
f
r
am
ewo
r
k
o
f
s
war
m
in
tellig
en
ce
.
T
h
e
h
y
b
r
id
PS
O
-
GA
ap
p
r
o
ac
h
d
em
o
n
s
tr
ated
s
u
p
er
io
r
p
er
f
o
r
m
an
ce
co
m
p
ar
e
d
to
s
tan
d
alo
n
e
GA
a
n
d
o
th
er
m
eth
o
d
s
,
a
ch
iev
in
g
s
ig
n
if
ican
t
im
p
r
o
v
em
e
n
ts
in
ac
cu
r
ac
y
,
F
1
-
s
co
r
e,
an
d
r
ec
all
f
o
r
g
lau
c
o
m
a,
d
iab
etic
r
etin
o
p
ath
y
,
an
d
ca
tar
ac
t
d
etec
tio
n
.
T
h
is
s
u
cc
ess
ca
n
b
e
attr
ib
u
ted
to
th
e
en
h
an
ce
d
f
ea
tu
r
e
s
elec
tio
n
an
d
p
ar
am
eter
o
p
tim
izatio
n
ca
p
ab
ilit
ies o
f
th
e
co
m
b
in
ed
ap
p
r
o
ac
h
.
T
h
e
s
tu
d
y
em
p
h
asizes
th
e
im
p
o
r
tan
ce
o
f
ea
r
ly
id
e
n
tific
atio
n
a
n
d
cl
ass
if
icatio
n
o
f
ey
e
co
n
d
itio
n
s
an
d
h
ig
h
lig
h
ts
th
e
p
o
ten
tial
o
f
s
war
m
in
tellig
en
c
e
in
ad
v
an
ci
n
g
d
iag
n
o
s
tic
to
o
l
s
in
o
p
h
th
alm
o
l
o
g
y
.
Fu
tu
r
e
r
esear
ch
s
h
o
u
ld
f
o
cu
s
o
n
f
u
r
t
h
er
o
p
tim
izatio
n
o
f
s
war
m
in
tellig
en
ce
alg
o
r
ith
m
s
,
t
h
eir
in
teg
r
atio
n
with
r
ea
l
-
tim
e
im
ag
in
g
tech
n
o
lo
g
ie
s
,
an
d
im
p
r
o
v
i
n
g
th
e
ex
p
lai
n
ab
ilit
y
o
f
th
e
d
iag
n
o
s
tic
m
o
d
els
to
en
s
u
r
e
wid
er
clin
ical
ad
o
p
tio
n
an
d
im
p
ac
t.
T
h
is
co
n
clu
s
io
n
s
u
m
m
ar
izes
t
h
e
k
ey
f
in
d
i
n
g
s
an
d
em
p
h
asizes
th
e
p
o
ten
tial
o
f
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
f
o
r
r
ev
o
lu
tio
n
izin
g
o
p
h
t
h
alm
ic
d
iag
n
o
s
tics
.
0
0.2
0.4
0.6
0.8
1
A
c
c
ur
ac
y
R
e
c
al
l
F
1
S
c
o
r
e
R
e
s
ul
t
s
P
e
r
fo
r
m
ance
m
e
asur
e
s
Eva
l
ua
ti
on
Metr
i
c
s
C
NN
P
S
O
GA
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
14
,
No
.
2
,
Ap
r
il 2
0
2
5
:
1
0
8
7
-
1
0
9
5
1094
RE
F
E
R
E
NC
E
S
[
1
]
Y
.
Q
i
n
a
n
d
A
.
H
a
w
b
a
n
i
,
“
A
n
o
v
e
l
s
e
g
me
n
t
a
t
i
o
n
me
t
h
o
d
f
o
r
o
p
t
i
c
d
i
sc
a
n
d
o
p
t
i
c
c
u
p
b
a
se
d
o
n
d
e
f
o
r
m
a
b
l
e
U
-
n
e
t
,
”
i
n
2
0
1
9
2
n
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
A
rt
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
Bi
g
D
a
t
a
(
I
C
AI
BD
)
,
2
0
1
9
,
p
p
.
3
9
4
–
3
9
9
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
A
I
B
D
.
2
0
1
9
.
8
8
3
7
0
2
5
.
[
2
]
A
.
S
h
o
u
k
a
t
,
S
.
A
k
b
a
r
,
S
.
A
.
H
a
ssa
n
,
S
.
I
q
b
a
l
,
A
.
M
e
h
mo
o
d
,
a
n
d
Q
.
M
.
I
l
y
a
s,
“
A
u
t
o
m
a
t
i
c
d
i
a
g
n
o
si
s
o
f
g
l
a
u
c
o
ma
f
r
o
m
r
e
t
i
n
a
l
i
ma
g
e
s
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
D
i
a
g
n
o
st
i
c
s
,
v
o
l
.
1
3
,
n
o
.
1
0
,
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
0
1
7
3
8
.
[
3
]
S
.
Y
a
d
a
v
a
n
d
J.
K
.
P
.
S
.
Y
a
d
a
v
,
“
A
u
t
o
ma
t
i
c
c
a
t
a
r
a
c
t
s
e
v
e
r
i
t
y
d
e
t
e
c
t
i
o
n
a
n
d
g
r
a
d
i
n
g
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
”
J
o
u
r
n
a
l
o
f
S
e
n
s
o
rs
,
v
o
l
.
2
0
2
3
,
n
o
.
1
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
3
/
2
9
7
3
8
3
6
.
[
4
]
P
.
V
a
sh
i
st
,
S
.
S
.
S
e
n
j
a
m,
V
.
G
u
p
t
a
,
N
.
G
u
p
t
a
,
a
n
d
A
.
K
u
mar,
“
D
e
f
i
n
i
t
i
o
n
o
f
b
l
i
n
d
n
e
ss
u
n
d
e
r
N
a
t
i
o
n
a
l
P
r
o
g
r
a
m
me
f
o
r
C
o
n
t
r
o
l
o
f
B
l
i
n
d
n
e
ss:
D
o
w
e
n
e
e
d
t
o
r
e
v
i
s
e
i
t
?
,
”
I
n
d
i
a
n
J
o
u
r
n
a
l
o
f
O
p
h
t
h
a
l
m
o
l
o
g
y
,
v
o
l
.
6
5
,
n
o
.
2
,
p
p
.
9
2
–
9
6
,
2
0
1
7
,
d
o
i
:
1
0
.
4
1
0
3
/
i
j
o
.
I
JO
_
8
6
9
_
1
6
.
[
5
]
M
.
M
.
B
u
t
t
,
D
.
N
.
F
.
A
.
I
sk
a
n
d
a
r
,
S
.
E
.
A
b
d
e
l
h
a
mi
d
,
G
.
L
a
t
i
f
,
a
n
d
R
.
A
l
g
h
a
z
o
,
“
D
i
a
b
e
t
i
c
r
e
t
i
n
o
p
a
t
h
y
d
e
t
e
c
t
i
o
n
f
r
o
m
f
u
n
d
u
s
i
ma
g
e
s
o
f
t
h
e
e
y
e
u
si
n
g
h
y
b
r
i
d
d
e
e
p
l
e
a
r
n
i
n
g
f
e
a
t
u
r
e
s
,
”
D
i
a
g
n
o
st
i
c
s
,
v
o
l
.
1
2
,
n
o
.
7
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
d
i
a
g
n
o
st
i
c
s1
2
0
7
1
6
0
7
.
[
6
]
Y
.
G
.
P
a
r
k
a
n
d
Y
.
J.
R
o
h
,
“
N
e
w
d
i
a
g
n
o
s
t
i
c
a
n
d
t
h
e
r
a
p
e
u
t
i
c
a
p
p
r
o
a
c
h
e
s
f
o
r
p
r
e
v
e
n
t
i
n
g
t
h
e
p
r
o
g
r
e
ss
i
o
n
o
f
d
i
a
b
e
t
i
c
r
e
t
i
n
o
p
a
t
h
y
,
”
J
o
u
rn
a
l
o
f
D
i
a
b
e
t
e
s
Re
s
e
a
rc
h
,
v
o
l
.
2
0
1
6
,
2
0
1
6
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
1
6
/
1
7
5
3
5
8
4
.
[
7
]
A
.
B
a
j
w
a
,
N
.
N
o
sh
e
e
n
,
K
.
I
.
Ta
l
p
u
r
,
a
n
d
S
.
A
k
r
a
m,
“
A
p
r
o
sp
e
c
t
i
v
e
st
u
d
y
o
n
d
i
a
b
e
t
i
c
r
e
t
i
n
o
p
a
t
h
y
d
e
t
e
c
t
i
o
n
b
a
s
e
d
o
n
m
o
d
i
f
y
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
u
s
i
n
g
f
u
n
d
u
s
i
m
a
g
e
s
a
t
s
i
n
d
h
i
n
st
i
t
u
t
e
o
f
o
p
h
t
h
a
l
mo
l
o
g
y
& v
i
su
a
l
sc
i
e
n
c
e
,
”
D
i
a
g
n
o
st
i
c
s
,
v
o
l
.
1
3
,
n
o
.
3
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
d
i
a
g
n
o
st
i
c
s
1
3
0
3
0
3
9
3
.
[
8
]
R
.
G
a
r
d
l
i
k
a
n
d
I
.
F
u
sek
o
v
a
,
“
P
h
a
r
m
a
c
o
l
o
g
i
c
t
h
e
r
a
p
y
f
o
r
d
i
a
b
e
t
i
c
r
e
t
i
n
o
p
a
t
h
y
,
”
S
e
m
i
n
a
rs
i
n
O
p
h
t
h
a
l
m
o
l
o
g
y
,
v
o
l
.
3
0
,
n
o
.
4
,
p
p
.
2
5
2
–
2
6
3
,
2
0
1
5
,
d
o
i
:
1
0
.
3
1
0
9
/
0
8
8
2
0
5
3
8
.
2
0
1
3
.
8
5
9
2
8
0
.
[
9
]
M
.
M
e
sq
u
i
d
a
,
F
.
D
r
a
w
n
e
l
,
a
n
d
S
.
F
a
u
ser,
“
Th
e
r
o
l
e
o
f
i
n
f
l
a
mm
a
t
i
o
n
i
n
d
i
a
b
e
t
i
c
e
y
e
d
i
se
a
se
,
”
S
e
m
i
n
a
rs
i
n
I
m
m
u
n
o
p
a
t
h
o
l
o
g
y
,
v
o
l
.
4
1
,
n
o
.
4
,
p
p
.
4
2
7
–
4
4
5
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/
s
0
0
2
8
1
-
0
1
9
-
0
0
7
5
0
-
7.
[
1
0
]
N
.
V
a
r
ma,
S
.
Y
a
d
a
v
,
a
n
d
J
.
K
.
P
.
S
.
Y
a
d
a
v
,
“
A
r
e
l
i
a
b
l
e
a
u
t
o
ma
t
i
c
c
a
t
a
r
a
c
t
d
e
t
e
c
t
i
o
n
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
S
y
s
t
e
m
A
ssu
r
a
n
c
e
E
n
g
i
n
e
e
r
i
n
g
a
n
d
M
a
n
a
g
e
m
e
n
t
,
v
o
l
.
1
4
,
n
o
.
3
,
p
p
.
1
0
8
9
–
1
1
0
2
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
3
1
9
8
-
023
-
0
1
9
2
3
-
2.
[
1
1
]
J.
Y
i
,
Y
.
R
a
n
,
a
n
d
G
.
Y
a
n
g
,
“
P
a
r
t
i
c
l
e
sw
a
r
m
o
p
t
i
m
i
z
a
t
i
o
n
-
b
a
se
d
a
p
p
r
o
a
c
h
f
o
r
o
p
t
i
c
d
i
sc
se
g
me
n
t
a
t
i
o
n
,
”
En
t
ro
p
y
,
v
o
l
.
2
4
,
n
o
.
6
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
e
2
4
0
6
0
7
9
6
.
[
1
2
]
“
G
l
a
u
c
o
ma
,
”
Ad
i
t
y
a
J
y
o
t
E
y
e
H
o
s
p
i
t
a
l
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
w
w
w
.
a
d
i
t
y
a
j
y
o
t
e
y
e
h
o
s
p
i
t
a
l
.
o
r
g
/
g
l
a
u
c
o
ma
.
h
t
m
l
[
1
3
]
D
.
F
.
S
a
n
t
o
s
-
B
u
st
o
s,
B
.
M
.
N
g
u
y
e
n
,
a
n
d
H
.
E.
Esp
i
t
i
a
,
“
T
o
w
a
r
d
s
a
u
t
o
m
a
t
e
d
e
y
e
c
a
n
c
e
r
c
l
a
ssi
f
i
c
a
t
i
o
n
v
i
a
V
G
G
a
n
d
R
e
sN
e
t
n
e
t
w
o
r
k
s
u
si
n
g
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
,
”
En
g
i
n
e
e
ri
n
g
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
a
n
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
,
v
o
l
.
3
5
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
e
st
c
h
.
2
0
2
2
.
1
0
1
2
1
4
.
[
1
4
]
S
.
D
e
g
a
d
w
a
l
a
,
D
.
V
y
a
s
,
H
.
S
.
D
a
v
e
,
V
.
P
a
t
e
l
,
a
n
d
J
.
N
.
M
e
h
t
a
,
“
E
y
e
mel
a
n
o
ma
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
a
n
d
c
l
a
ssi
f
i
c
a
t
i
o
n
u
si
n
g
C
N
N
,
”
i
n
S
e
c
o
n
d
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
I
m
a
g
e
Pr
o
c
e
ssi
n
g
a
n
d
C
a
p
su
l
e
N
e
t
w
o
rks
,
2
0
2
2
,
p
p
.
4
8
9
–
4
9
7
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
0
-
8
4
7
6
0
-
9
_
4
2
.
[
1
5
]
A
.
S
i
n
h
a
,
A
.
R
P
,
a
n
d
N
.
N
.
S
,
“
E
y
e
t
u
mo
u
r
d
e
t
e
c
t
i
o
n
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
”
i
n
2
0
2
1
S
e
v
e
n
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
c
o
n
f
e
r
e
n
c
e
o
n
B
i
o
S
i
g
n
a
l
s,
I
m
a
g
e
s,
a
n
d
I
n
s
t
ru
m
e
n
t
a
t
i
o
n
(
I
C
BS
I
I
)
,
2
0
2
1
,
p
p
.
1
–
5
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
B
S
I
I
5
1
8
3
9
.
2
0
2
1
.
9
4
4
5
1
7
2
.
[
1
6
]
G
.
A
n
e
t
a
l
.
,
“
G
l
a
u
c
o
ma
d
i
a
g
n
o
si
s
w
i
t
h
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
b
a
se
d
o
n
o
p
t
i
c
a
l
c
o
h
e
r
e
n
c
e
t
o
m
o
g
r
a
p
h
y
a
n
d
c
o
l
o
r
f
u
n
d
u
s
i
ma
g
e
s
,
”
J
o
u
rn
a
l
o
f
H
e
a
l
t
h
c
a
re
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
2
0
1
9
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
1
9
/
4
0
6
1
3
1
3
.
[
1
7
]
S
.
Y
o
u
sef
i
,
“
C
l
i
n
i
c
a
l
a
p
p
l
i
c
a
t
i
o
n
s
o
f
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
i
n
g
l
a
u
c
o
m
a
,
”
J
o
u
rn
a
l
o
f
O
p
h
t
h
a
l
m
i
c
a
n
d
Vi
s
i
o
n
Re
s
e
a
rc
h
,
v
o
l
.
1
8
,
n
o
.
1
,
p
p
.
9
7
–
1
1
2
,
2
0
2
3
,
d
o
i
:
1
0
.
1
8
5
0
2
/
J
O
V
R
.
V
1
8
I
1
.
1
2
7
3
0
.
[
1
8
]
M
.
S
.
J
u
n
a
y
e
d
,
M
.
B
.
I
sl
a
m,
A
.
S
a
d
e
g
h
z
a
d
e
h
,
a
n
d
S
.
R
a
h
m
a
n
,
“
C
a
t
a
r
a
c
t
N
e
t
:
A
n
a
u
t
o
ma
t
e
d
c
a
t
a
r
a
c
t
d
e
t
e
c
t
i
o
n
s
y
st
e
m
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
f
u
n
d
u
s
i
ma
g
e
s
,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
9
,
p
p
.
1
2
8
7
9
9
–
1
2
8
8
0
8
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
1
.
3
1
1
2
9
3
8
.
[
1
9
]
X
.
Zh
a
n
g
,
J
.
L
v
,
H
.
Z
h
e
n
g
,
a
n
d
Y
.
S
a
n
g
,
“
A
t
t
e
n
t
i
o
n
-
b
a
se
d
mu
l
t
i
-
mo
d
e
l
e
n
sem
b
l
e
f
o
r
a
u
t
o
m
a
t
i
c
c
a
t
a
r
a
c
t
d
e
t
e
c
t
i
o
n
i
n
B
-
s
c
a
n
e
y
e
u
l
t
r
a
s
o
u
n
d
i
m
a
g
e
s,
”
i
n
2
0
2
0
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
i
n
t
C
o
n
f
e
re
n
c
e
o
n
N
e
u
r
a
l
N
e
t
w
o
r
k
s
(
I
J
C
N
N
)
,
2
0
2
0
,
p
p
.
1
–
10
,
d
o
i
:
1
0
.
1
1
0
9
/
I
JC
N
N
4
8
6
0
5
.
2
0
2
0
.
9
2
0
7
6
9
6
.
[
2
0
]
M
.
K
.
H
a
s
a
n
e
t
a
l
.
,
“
C
a
t
a
r
a
c
t
d
i
se
a
se
d
e
t
e
c
t
i
o
n
b
y
u
s
i
n
g
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
-
b
a
s
e
d
i
n
t
e
l
l
i
g
e
n
t
me
t
h
o
d
s,
”
C
o
m
p
u
t
a
t
i
o
n
a
l
a
n
d
Ma
t
h
e
m
a
t
i
c
a
l
M
e
t
h
o
d
s
i
n
Me
d
i
c
i
n
e
,
v
o
l
.
2
0
2
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
1
/
7
6
6
6
3
6
5
.
[
2
1
]
L.
M
.
M
a
r
c
e
l
l
o
,
E
.
O
e
y
,
Z.
S
.
L
i
e
,
a
n
d
W
.
A
st
u
t
i
,
“
A
u
t
o
m
a
t
i
c
c
a
t
a
r
a
c
t
d
e
t
e
c
t
i
o
n
sy
s
t
e
m
b
a
s
e
d
o
n
su
p
p
o
r
t
v
e
c
t
o
r
mac
h
i
n
e
(
S
V
M
)
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
S
e
c
o
n
d
Asi
a
Pa
c
i
f
i
c
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
I
n
d
u
s
t
ri
a
l
En
g
i
n
e
e
ri
n
g
a
n
d
O
p
e
r
a
t
i
o
n
s
M
a
n
a
g
e
m
e
n
t
,
S
u
r
a
k
a
r
t
a
,
I
n
d
o
n
e
si
a
,
2
0
2
1
,
p
p
.
9
5
9
–
9
6
5
.
[
2
2
]
P
.
G
.
S
u
b
i
n
a
n
d
P
.
M
.
K
a
n
n
a
n
,
“
M
u
l
t
i
p
l
e
e
y
e
d
i
sea
s
e
d
e
t
e
c
t
i
o
n
u
si
n
g
h
y
b
r
i
d
a
d
a
p
t
i
v
e
mu
t
a
t
i
o
n
sw
a
r
m
o
p
t
i
mi
z
a
t
i
o
n
a
n
d
R
N
N
,
”
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
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
3
,
n
o
.
9
,
p
p
.
4
0
1
–
4
1
0
,
2
0
2
2
,
d
o
i
:
1
0
.
1
4
5
6
9
/
I
JA
C
S
A
.
2
0
2
2
.
0
1
3
0
9
4
6
.
[
2
3
]
D
.
El
t
i
g
a
n
i
a
n
d
S
.
M
a
sr
i
,
“
C
h
a
l
l
e
n
g
e
s
o
f
i
n
t
e
g
r
a
t
i
n
g
r
e
n
e
w
a
b
l
e
e
n
e
r
g
y
so
u
r
c
e
s
t
o
smar
t
g
r
i
d
s
:
a
r
e
v
i
e
w
,
”
Re
n
e
w
a
b
l
e
a
n
d
S
u
s
t
a
i
n
a
b
l
e
E
n
e
rg
y
R
e
v
i
e
w
s
,
v
o
l
.
5
2
,
p
p
.
7
7
0
–
7
8
0
,
2
0
1
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
r
se
r
.
2
0
1
5
.
0
7
.
1
4
0
.
[
2
4
]
S
.
J.
G
h
o
u
s
h
c
h
i
,
R
.
R
a
n
j
b
a
r
z
a
d
e
h
,
A
.
H
.
D
a
d
k
h
a
h
,
Y
.
P
o
u
r
a
sa
d
,
a
n
d
M
.
B
e
n
d
e
c
h
a
c
h
e
,
“
A
n
e
x
t
e
n
d
e
d
a
p
p
r
o
a
c
h
t
o
p
r
e
d
i
c
t
r
e
t
i
n
o
p
a
t
h
y
i
n
d
i
a
b
e
t
i
c
p
a
t
i
e
n
t
s
u
s
i
n
g
t
h
e
g
e
n
e
t
i
c
a
l
g
o
r
i
t
h
m a
n
d
f
u
z
z
y
c
-
me
a
n
s,
”
Bi
o
Me
d
Re
s
e
a
r
c
h
I
n
t
e
rn
a
t
i
o
n
a
l
,
v
o
l
.
2
0
2
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
1
/
5
5
9
7
2
2
2
.
[
2
5
]
M
.
R
.
K
.
M
o
o
k
i
a
h
e
t
a
l
.
,
“
E
v
o
l
u
t
i
o
n
a
r
y
a
l
g
o
r
i
t
h
m
b
a
s
e
d
c
l
a
ssi
f
i
e
r
p
a
r
a
me
t
e
r
t
u
n
i
n
g
f
o
r
a
u
t
o
mat
i
c
d
i
a
b
e
t
i
c
r
e
t
i
n
o
p
a
t
h
y
g
r
a
d
i
n
g
:
A
h
y
b
r
i
d
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
a
p
p
r
o
a
c
h
,
”
K
n
o
w
l
e
d
g
e
-
B
a
se
d
S
y
s
t
e
m
s
,
v
o
l
.
3
9
,
p
p
.
9
–
2
2
,
2
0
1
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
k
n
o
s
y
s.
2
0
1
2
.
0
9
.
0
0
8
.
[
2
6
]
L.
Z
h
a
n
g
a
n
d
C
.
P
.
Li
m
,
“
I
n
t
e
l
l
i
g
e
n
t
o
p
t
i
c
d
i
sc
s
e
g
me
n
t
a
t
i
o
n
u
si
n
g
i
m
p
r
o
v
e
d
p
a
r
t
i
c
l
e
sw
a
r
m
o
p
t
i
m
i
z
a
t
i
o
n
a
n
d
e
v
o
l
v
i
n
g
e
n
s
e
m
b
l
e
mo
d
e
l
s
,
”
Ap
p
l
i
e
d
S
o
f
t
C
o
m
p
u
t
i
n
g
J
o
u
rn
a
l
,
v
o
l
.
9
2
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
s
o
c
.
2
0
2
0
.
1
0
6
3
2
8
.
[
2
7
]
G
.
V
.
D
o
d
d
i
,
“
Ey
e
d
i
s
e
a
s
e
c
l
a
ssi
f
i
c
a
t
i
o
n
:
e
y
e
d
i
se
a
se
r
e
t
i
n
a
l
i
ma
g
e
s
,
”
K
a
g
g
l
e
.
2
0
2
2
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
w
w
w
.
k
a
g
g
l
e
.
c
o
m
/
d
a
t
a
se
t
s/
g
u
n
a
v
e
n
k
a
t
d
o
d
d
i
/
e
y
e
-
d
i
se
a
ses
-
c
l
a
ss
i
f
i
c
a
t
i
o
n
[
2
8
]
A
.
G
.
G
a
d
,
“
P
a
r
t
i
c
l
e
sw
a
r
m
o
p
t
i
mi
z
a
t
i
o
n
a
l
g
o
r
i
t
h
m
a
n
d
i
t
s
a
p
p
l
i
c
a
t
i
o
n
s:
a
s
y
st
e
mat
i
c
r
e
v
i
e
w
,
”
Arc
h
i
v
e
s
o
f
C
o
m
p
u
t
a
t
i
o
n
a
l
Me
t
h
o
d
s
i
n
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
2
9
,
n
o
.
5
,
p
p
.
2
5
3
1
–
2
5
6
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s1
1
8
3
1
-
0
2
1
-
0
9
6
9
4
-
4.
[
2
9
]
M
.
Z
o
l
f
p
o
u
r
-
A
r
o
k
h
l
o
,
A
.
S
e
l
a
ma
t
,
S
.
Z.
M
.
H
a
s
h
i
m
,
a
n
d
H
.
A
f
k
h
a
m
i
,
“
M
o
d
e
l
i
n
g
o
f
r
o
u
t
e
p
l
a
n
n
i
n
g
s
y
st
e
m
b
a
se
d
o
n
Q
v
a
l
u
e
-
b
a
s
e
d
d
y
n
a
mi
c
p
r
o
g
r
a
mm
i
n
g
w
i
t
h
mu
l
t
i
-
a
g
e
n
t
r
e
i
n
f
o
r
c
e
me
n
t
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms,
”
En
g
i
n
e
e
ri
n
g
A
p
p
l
i
c
a
t
i
o
n
s
o
f
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
2
9
,
p
p
.
1
6
3
–
1
7
7
,
2
0
1
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
n
g
a
p
p
a
i
.
2
0
1
4
.
0
1
.
0
0
1
.
[
3
0
]
J.
H
u
,
C
.
C
h
e
n
,
L.
C
a
i
,
M
.
R
.
K
h
o
sr
a
v
i
,
Q
.
P
e
i
,
a
n
d
S
.
W
a
n
,
“
U
A
V
-
a
ss
i
st
e
d
v
e
h
i
c
u
l
a
r
e
d
g
e
c
o
m
p
u
t
i
n
g
f
o
r
t
h
e
6
G
i
n
t
e
r
n
e
t
o
f
v
e
h
i
c
l
e
s:
a
r
c
h
i
t
e
c
t
u
r
e
,
i
n
t
e
l
l
i
g
e
n
c
e
,
a
n
d
c
h
a
l
l
e
n
g
e
s
,
”
I
EEE
C
o
m
m
u
n
i
c
a
t
i
o
n
s
S
t
a
n
d
a
rd
s
M
a
g
a
zi
n
e
,
v
o
l
.
5
,
n
o
.
2
,
p
p
.
1
2
–
1
8
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
M
C
O
M
S
TD
.
0
0
1
.
2
0
0
0
0
1
7
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Un
ve
ilin
g
p
r
ec
is
io
n
:
E
ye
ca
n
c
er d
etec
tio
n
r
ed
efin
ed
w
ith
p
a
r
ticle
s
w
a
r
m
…
(
S
a
n
ve
d
N
a
r
w
a
d
ka
r
)
1095
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
S
a
n
v
e
d
Na
r
wa
d
k
a
r
is
s
tu
d
e
n
t
a
t
Vis
h
wa
k
a
rm
a
In
stit
u
te
o
f
In
f
o
rm
a
ti
o
n
Tec
h
n
o
l
o
g
y
,
P
u
n
e
,
M
a
h
a
ra
sh
tra.
He
p
u
rsu
i
n
g
B
.T
ech
.
i
n
i
n
fo
rm
a
t
io
n
tec
h
n
o
lo
g
y
.
P
a
ss
io
n
a
te
a
b
o
u
t
a
rti
ficia
l
i
n
telli
g
e
n
c
e
,
d
a
ta
sc
ien
c
e
a
s
we
ll
a
s
m
a
c
h
in
e
lea
r
n
in
g
.
Als
o
e
n
g
a
g
e
d
with
a
re
se
a
rc
h
p
ro
jec
t
wit
h
Ad
h
y
a
y
a
n
Ac
a
d
e
m
y
P
u
n
e
,
M
a
h
a
ra
sh
tra.
Ke
e
n
i
n
tere
st
in
c
lu
d
e
s
lea
rn
i
n
g
tec
h
n
iq
u
e
s o
f
m
a
c
h
in
e
lea
rn
in
g
a
n
d
a
p
p
ly
i
n
g
th
e
m
to
so
l
v
e
re
a
l
li
fe
p
ro
b
lem
s a
n
d
a
p
p
li
c
a
ti
o
n
s
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
sa
n
v
e
d
.
2
2
2
1
1
5
3
9
@v
ii
t.
a
c
.
in
.
Pra
d
n
y
a
S
a
m
it
Meht
a
wo
r
k
in
g
a
t
t
h
e
Vish
wa
k
a
rm
a
In
st
it
u
te
o
f
I
n
fo
rm
a
ti
o
n
Tec
h
n
o
l
o
g
y
P
u
n
e
.
S
h
e
re
c
e
iv
e
d
t
h
e
B
.
E
.
&
M
.
E
.
d
e
g
re
e
in
c
o
m
p
u
ter
e
n
g
in
e
e
rin
g
fro
m
S
P
P
U
Un
iv
e
rsity
,
P
u
n
e
,
M
a
h
a
ra
sh
tra.
S
h
e
h
a
s
1
5
y
e
a
rs
tea
c
h
in
g
e
x
p
e
rien
c
e
a
n
d
2
y
e
a
rs
o
f
i
n
d
u
strial
e
x
p
e
rien
c
e
.
He
r
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
a
rti
ficia
l
in
telli
g
e
n
c
e
,
d
e
e
p
lea
rn
i
n
g
,
h
i
g
h
p
e
rfo
rm
a
n
c
e
c
o
m
p
u
ti
n
g
,
a
n
d
m
a
c
h
in
e
lea
rn
i
n
g
.
S
h
e
h
a
s
p
u
b
li
s
h
e
d
2
0
p
a
p
e
rs
in
n
a
ti
o
n
a
l
a
n
d
in
tern
a
ti
o
n
a
l
j
o
u
r
n
a
ls
a
n
d
c
o
n
fe
re
n
c
e
s,
7
b
o
o
k
c
h
a
p
ters
,
7
p
a
te
n
ts
,
a
n
d
2
p
a
ten
ts
a
re
g
ra
n
ted
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
p
r
a
d
n
y
a
1
7
.
m
e
h
ta@
g
m
a
il
.
c
o
m
.
Rutu
ja
Ra
je
n
d
r
a
Pa
til
re
c
e
iv
e
d
th
e
b
a
c
h
e
lo
r’s
d
e
g
re
e
a
s
we
ll
a
s
m
a
ste
r’s
d
e
g
re
e
in
in
f
o
rm
a
ti
o
n
tec
h
n
o
lo
g
y
fr
o
m
P
u
n
e
Un
i
v
e
rsity
,
P
u
n
e
,
In
d
ia,
in
2
0
0
6
a
n
d
2
0
1
5
,
re
sp
e
c
ti
v
e
ly
.
S
h
e
h
a
s
c
o
m
p
lete
d
h
e
r
P
h
.
D
.
d
e
g
re
e
with
t
h
e
S
y
m
b
i
o
sis
In
st
it
u
te
o
f
Tec
h
n
o
l
o
g
y
,
S
y
m
b
io
sis
In
tern
a
ti
o
n
a
l
(De
e
m
e
d
Un
iv
e
rsity
),
P
u
n
e
.
Cu
rre
n
tl
y
,
s
h
e
is
wo
rk
i
n
g
a
s
a
n
As
sista
n
t
P
ro
fe
ss
o
r
in
De
p
a
rtme
n
t
o
f
C
o
m
p
u
ter
S
c
ien
c
e
En
g
in
e
e
rin
g
-
Artifi
c
ial
In
telli
g
e
n
c
e
a
n
d
M
a
c
h
i
n
e
Lea
rn
in
g
at
Vish
wa
k
a
rm
a
In
stit
u
t
e
o
f
In
f
o
rm
a
ti
o
n
Tec
h
n
o
l
o
g
y
,
P
u
n
e
,
M
a
h
a
ra
sh
tra,
In
d
ia.
He
r
re
se
a
rc
h
in
tere
sts
in
c
l
u
d
e
m
a
c
h
i
n
e
lea
rn
in
g
,
d
e
e
p
lea
rn
in
g
,
an
d
m
u
lt
imo
d
a
l
fu
si
o
n
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
ru
tu
jap
a
t@
g
m
a
il
.
c
o
m
.
K
a
ly
a
n
i
K
a
d
a
m
re
c
e
iv
e
d
th
e
P
h
.
D.
d
e
g
re
e
fro
m
S
y
m
b
i
o
sis
In
ter
n
a
ti
o
n
a
l
(De
e
m
e
d
Un
iv
e
rsity
),
Lav
a
le,
P
u
n
e
,
M
a
h
a
ra
sh
tra,
I
n
d
ia.
S
h
e
is
c
u
r
re
n
tl
y
wo
r
k
in
g
a
s
As
sista
n
t
P
ro
fe
ss
o
r
with
S
IT
,
P
u
n
e
.
He
r
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
m
a
c
h
in
e
lea
rn
in
g
,
d
e
e
p
lea
rn
in
g
,
a
n
d
c
o
m
p
u
ter
v
isio
n
.
S
h
e
c
a
n
b
e
c
o
n
t
a
c
ted
a
t
e
m
a
il
:
h
u
law
a
lek
a
ly
a
n
i@
g
m
a
il
.
c
o
m
.
Vija
y
k
u
m
a
r
Bid
v
e
is
As
so
c
i
a
te
P
ro
fe
ss
o
r
a
t
S
y
m
b
i
o
sis
S
k
il
ls
a
n
d
P
r
o
fe
ss
io
n
a
l
Un
iv
e
rsity
,
Kiwa
le,
P
u
n
e
,
M
a
h
a
r
a
sh
tra,
In
d
ia.
He
Ho
ld
s
a
P
h
.
D
.
d
e
g
re
e
in
c
o
m
p
u
ter
sc
ien
c
e
a
n
d
e
n
g
in
e
e
rin
g
wit
h
sp
e
c
ializa
ti
o
n
i
n
so
ftwa
re
e
n
g
i
n
e
e
rin
g
.
His
re
se
a
rc
h
a
re
a
s
a
re
so
ftwa
re
e
n
g
in
e
e
rin
g
a
n
d
m
a
c
h
in
e
lea
rn
in
g
.
He
h
a
s
p
u
b
li
sh
e
d
ten
p
a
ten
ts
a
n
d
m
o
re
th
a
n
4
0
+
re
se
a
rc
h
a
rti
c
les
in
n
a
ti
o
n
a
l
a
n
d
in
tern
a
ti
o
n
a
l
jo
u
rn
a
ls.
He
is
a
li
fe
m
e
m
b
e
r
o
f
IS
TE
.
He
is
w
o
rk
i
n
g
a
s
a
n
e
x
p
e
rt
fo
r
v
a
ri
o
u
s
su
b
jec
ts.
Also
,
h
e
h
a
s
wo
rk
e
d
a
s
a
re
v
iew
e
r
f
o
r
v
a
rio
u
s
c
o
n
fe
re
n
c
e
s
a
n
d
jo
u
r
n
a
ls.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
v
ij
a
y
.
b
i
d
v
e
@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.