I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
23
,
No
.
2
,
A
u
g
u
s
t
2
0
2
1
,
p
p
.
8
3
7
~
8
4
6
I
SS
N:
2502
-
4
7
5
2
,
DOI
: 1
0
.
1
1
5
9
1
/ijeecs.v
23
.i
2
.
pp
837
-
8
4
6
837
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs.ia
esco
r
e.
co
m
A review
on sup
e
rv
ised lea
rning
m
ethodo
lo
g
ies for d
etec
tion
o
f
ex
uda
tes
in
dia
be
tic
re
tinopa
thy
Uj
wa
la
W.
Wa
s
ek
a
r,
R.
K
.
B
a
t
hla
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
,
De
sh
Bh
a
g
a
t
U
n
iv
e
rsit
y
,
M
a
n
d
i
G
o
b
in
d
g
a
r
h
,
P
u
n
jab
,
I
n
d
ia
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
r
1
8
,
2
0
2
1
R
ev
is
ed
J
u
l 1
2
,
2
0
2
1
Acc
ep
ted
J
u
l 1
8
,
2
0
2
1
Dia
b
e
ti
c
r
e
ti
n
o
p
a
t
h
y
h
a
s
b
e
c
o
m
e
o
n
e
o
f
t
h
e
m
a
jo
r
re
a
so
n
s
fo
r
b
l
in
d
n
e
ss
in
th
e
wo
rl
d
.
Early
a
n
d
p
re
c
ise
d
ia
g
n
o
sis o
f
t
h
e
d
ise
a
se
m
a
y
sa
v
e
o
n
e
’s
e
y
e
sig
h
t
fro
m
irrev
e
rsib
le d
a
m
a
g
e
.
M
a
n
u
a
l
d
e
tec
ti
o
n
o
f
les
io
n
s
is
ti
m
e
c
o
n
su
m
in
g
a
n
d
m
a
y
n
o
t
b
e
a
s
a
c
c
u
ra
te
a
s
d
e
sir
a
b
le.
M
a
n
y
a
u
to
m
a
ted
sy
ste
m
s
h
a
v
e
b
e
e
n
d
e
v
e
lo
p
e
d
re
c
e
n
tl
y
to
h
e
lp
o
p
h
th
a
lmo
lo
g
ists
in
t
h
e
ir
e
n
d
e
a
v
o
rs.
E
x
u
d
a
tes
a
re
o
n
e
o
f
t
h
e
e
a
rly
sig
n
s
o
f
m
a
n
ifes
tatio
n
o
f
d
iab
e
ti
c
re
ti
n
o
p
a
t
h
y
.
I
n
th
is
p
a
p
e
r,
th
e
m
e
th
o
d
o
lo
g
ies
d
e
tec
ti
n
g
e
x
u
d
a
tes
in
re
ti
n
a
l
fu
n
d
u
s
ima
g
e
s
we
re
re
v
iew
e
d
.
Th
e
se
m
e
th
o
d
s
we
re
c
a
teg
o
rize
d
in
to
d
e
e
p
lea
rn
in
g
,
m
a
c
h
in
e
lea
rn
in
g
a
n
d
m
e
th
o
d
s
p
rima
ril
y
fo
c
u
sin
g
o
n
ima
g
e
p
ro
c
e
ss
in
g
t
e
c
h
n
iq
u
e
s.
Th
e
c
o
m
p
re
h
e
n
siv
e
v
iew
o
f
t
h
e
p
e
rfo
rm
a
n
c
e
s
o
f
th
e
m
e
th
o
d
s
wa
s
g
iv
e
n
.
S
e
v
e
ra
l
d
a
tas
e
ts
we
re
d
e
sc
rib
e
d
b
riefly
.
M
o
st
o
f
th
e
re
se
a
rc
h
e
rs
p
re
fe
rre
d
c
o
m
b
in
a
ti
o
n
o
f
m
u
lt
ip
le
p
u
b
li
c
a
ll
y
a
v
a
il
a
b
le
d
a
tab
a
se
s.
Also
,
t
h
e
p
o
ten
ti
a
l
a
re
a
s
o
f
re
s
e
a
rc
h
we
r
e
d
isc
u
ss
e
d
.
It
wa
s
fo
u
n
d
t
h
a
t
se
n
siti
v
it
y
wh
ich
id
e
n
ti
fies
th
e
a
b
n
o
rm
a
l
ima
g
e
s
c
o
rre
c
tl
y
,
is
th
e
m
o
st
wid
e
ly
u
se
d
p
e
rfo
rm
a
n
c
e
m
e
a
su
re
.
Th
e
stu
d
y
will
b
e
h
e
l
p
fu
l
to
th
e
re
se
a
rc
h
e
rs
wa
n
ti
n
g
to
e
x
p
lo
re
m
o
re
in
th
is f
iel
d
.
K
ey
w
o
r
d
s
:
B
r
ig
h
t le
s
io
n
s
Dee
p
lear
n
in
g
Fu
n
d
u
s
i
m
ag
e
I
m
ag
e
p
r
o
ce
s
s
in
g
Ma
ch
in
e
lear
n
in
g
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
:
Ujwala
W
.
W
a
s
ek
ar
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
Desh
B
h
ag
at
Un
iv
er
s
ity
Ma
n
d
i G
o
b
in
d
g
ar
h
,
Dis
tr
ict
Fateh
g
ar
h
Sah
ib
,
Pu
n
jab
-
1
4
7
3
0
1
I
n
d
ia
E
m
ail:
u
jwalaz
ad
e@
r
ed
if
f
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
I
n
th
e
last
f
ew
y
ea
r
s
,
with
t
h
e
ad
v
e
n
t
o
f
tech
n
o
l
o
g
y
,
life
s
ty
le
o
f
p
eo
p
le
h
as
ch
a
n
g
ed
a
lo
t.
L
ess
p
h
y
s
ical
wo
r
k
o
u
t
an
d
u
n
h
ea
lth
y
ea
tin
g
h
ab
its
h
as
led
to
th
e
in
cr
ea
s
ed
lev
el
o
f
g
lu
co
s
e
i
n
th
e
b
lo
o
d
.
Sin
ce
1
9
9
0
,
n
u
m
b
er
o
f
d
ia
b
etic
p
ati
en
ts
h
as
in
cr
ea
s
ed
tr
e
m
en
d
o
u
s
ly
th
r
o
u
g
h
o
u
t
th
e
g
lo
b
e
[
1
]
.
Diab
etes
ca
n
g
iv
e
r
is
e
to
m
an
y
o
th
er
d
is
o
r
d
e
r
s
s
u
ch
as
d
iab
etic
r
etin
o
p
ath
y
(
DR
)
,
d
iab
etic
m
ac
u
la
r
ed
e
m
a
an
d
g
lau
c
o
m
a
.
DR
ca
n
d
ir
ec
tly
in
f
lu
en
ce
th
e
v
is
io
n
if
n
o
t
tr
ea
ted
at
a
n
ea
r
ly
s
t
ag
e.
Ma
n
u
al
d
iag
n
o
s
is
o
f
DR
m
ay
g
iv
e
i
n
ac
cu
r
ate
r
esu
lts
.
C
o
m
p
u
ter
aid
e
d
s
cr
e
en
in
g
o
f
DR
co
u
ld
b
e
h
el
p
f
u
l
f
o
r
o
p
h
th
alm
o
l
o
g
is
ts
in
p
r
o
v
id
in
g
q
u
ick
a
n
d
p
r
ec
is
e
d
iag
n
o
s
is
[
2
]
-
[
4
]
.
T
h
o
u
g
h
,
th
er
a
p
eu
tic
o
p
tio
n
s
s
u
c
h
as
p
h
a
r
m
ac
o
th
er
ap
y
ar
e
b
e
in
g
m
ad
e
av
ailab
le
alo
n
g
with
tr
ad
itio
n
al
laser
th
e
r
ap
y
[
5
]
.
W
ea
k
en
ed
b
lo
o
d
v
ess
els
in
s
i
d
e
th
e
r
etin
a
r
u
p
tu
r
e
r
esu
ltin
g
in
to
em
is
s
io
n
o
f
b
lo
o
d
a
n
d
lip
id
s
an
d
f
o
r
m
atio
n
o
f
lesi
o
n
s
[
6
]
.
T
h
e
ab
n
o
r
m
alities
th
at
ap
p
ea
r
o
n
th
e
r
etin
a
ar
e
m
icr
o
an
eu
r
y
s
m
s
(
MA
)
,
h
ae
m
o
r
r
h
ag
es
(
HM
)
an
d
ex
u
d
ates
(
E
X)
.
Neo
v
ascu
lar
iza
tio
n
(
n
ew
b
u
t
ab
n
o
r
m
al
v
ein
s
)
is
th
e
ap
ex
d
ef
o
r
m
ity
m
ak
in
g
p
r
ec
is
e
b
lo
o
d
v
ess
el
s
eg
m
en
tatio
n
in
e
v
itab
le
[
7
]
.
M
As
an
d
HM
s
ar
e
ca
lled
r
ed
les
io
n
s
wh
ile
E
Xs
ar
e
ter
m
ed
as
b
r
ig
h
t
lesi
o
n
s
.
E
x
u
d
ates
ar
e
ag
ain
ca
teg
o
r
ized
as
h
ar
d
ex
u
d
ates
(
HE
)
an
d
co
tt
o
n
wo
o
l
s
p
o
ts
,
also
k
n
o
wn
as
s
o
f
t
ex
u
d
ates
(
SE)
d
ep
en
d
in
g
u
p
o
n
th
eir
tex
tu
r
e
an
d
ap
p
ea
r
a
n
ce
.
So
m
e
o
f
t
h
e
ea
r
ly
s
ig
n
s
o
f
DR
ar
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
23
,
No
.
2
,
Au
g
u
s
t 2
0
2
1
:
837
-
8
4
6
838
m
icr
o
an
eu
r
y
s
m
s
an
d
ex
u
d
ates.
Fig
u
r
e
1
(
a
)
an
d
(
b
)
s
h
o
ws
t
h
e
n
o
r
m
al
r
etin
a
a
n
d
r
etin
a
h
av
in
g
e
x
u
d
ates
[
8
]
.
No
r
m
al
r
etin
a
im
ag
e
co
n
s
is
ts
o
f
o
p
t
ic
d
is
c
(
OD)
,
m
ac
u
la
a
n
d
b
lo
o
d
v
ess
els.
T
h
is
p
ap
er
m
ain
ly
f
o
cu
s
es
o
n
th
e
m
eth
o
d
s
av
ailab
le
in
th
e
liter
atu
r
e
f
o
r
t
h
e
d
et
ec
tio
n
an
d
class
if
icatio
n
o
f
ex
u
d
ates.
E
x
u
d
ates
ap
p
ea
r
as
b
r
ig
h
t
as
o
p
t
ic
d
is
c.
So
,
it
b
ec
o
m
es
ev
id
en
t
f
o
r
th
e
r
esear
ch
er
s
to
elim
in
ate
it
b
ef
o
r
e
d
etec
tin
g
th
e
ex
u
d
ates.
B
ein
g
in
d
icato
r
o
f
th
e
o
n
s
et
o
f
th
e
d
is
ea
s
e,
e
x
u
d
ates
h
a
v
e
to
b
e
d
etec
ted
ac
cu
r
ately
at
an
ea
r
l
y
s
tag
e
to
a
v
o
id
an
y
f
u
r
th
er
c
o
m
p
licatio
n
.
T
h
er
e
ar
e
m
a
n
y
t
ec
h
n
iq
u
es
p
r
o
p
o
s
ed
in
th
is
d
o
m
ain
f
o
r
s
eg
m
en
tatio
n
,
d
etec
tio
n
a
n
d
cl
ass
if
icatio
n
o
f
th
e
im
ag
es
as
n
o
r
m
al
a
n
d
ab
n
o
r
m
al.
I
n
th
is
p
ap
er
,
d
if
f
er
en
t
m
et
h
o
d
s
o
f
e
v
alu
atio
n
b
elo
n
g
in
g
to
d
ee
p
l
ea
r
n
in
g
,
m
ac
h
in
e
lear
n
i
n
g
an
d
im
ag
e
p
r
o
ce
s
s
in
g
,
h
av
e
b
ee
n
r
ev
iewe
d
.
E
ac
h
m
eth
o
d
h
as
its
u
n
iq
u
e
way
o
f
im
p
lem
en
tatio
n
,
b
u
t
s
o
m
e
lim
itatio
n
s
as
well.
Var
io
u
s
d
atab
ases
an
d
p
er
f
o
r
m
an
ce
m
etr
ics
h
av
e
b
ee
n
d
is
cu
s
s
ed
.
All
th
ese
asp
ec
ts
h
av
e
b
ee
n
p
r
esen
ted
in
a
co
m
p
r
eh
e
n
s
iv
e
m
an
n
er
.
R
est
o
f
th
e
p
ap
er
is
o
r
g
an
ize
d
as
s
h
o
wn
in
:
se
ctio
n
2
p
r
esen
ts
th
e
d
atab
ases
u
s
ed
in
ev
alu
atin
g
th
e
alg
o
r
ith
m
,
s
ec
tio
n
3
b
r
ief
ly
d
e
s
cr
ib
es
th
e
p
er
f
o
r
m
a
n
ce
m
ea
s
u
r
es,
s
ec
tio
n
4
p
r
o
v
id
es
th
e
d
i
f
f
er
en
t
ap
p
r
o
ac
h
es
b
ased
o
n
d
ee
p
lear
n
in
g
,
m
ac
h
in
e
lear
n
in
g
an
d
im
ag
e
p
r
o
c
ess
in
g
,
s
ec
tio
n
5
g
iv
es
t
h
e
o
u
tlo
o
k
o
f
th
e
s
tu
d
y
u
n
d
er
t
h
e
h
ea
d
i
n
g
‘
d
is
cu
s
s
io
n
’
,
f
in
ally
,
se
ctio
n
6
co
n
cl
u
d
es t
h
e
wo
r
k
with
c
r
is
p
in
f
er
e
n
ce
.
(
a)
(
b
)
Fig
u
r
e
1
.
Sh
o
w;
(
a)
No
r
m
al
r
e
tin
a
(
b
)
R
etin
a
with
ex
u
d
ates
2.
DATAS
E
T
S
Data
b
ase
is
a
co
llectio
n
o
f
r
eti
n
al
im
ag
es
th
at
p
r
o
v
id
e
a
co
m
p
etitiv
e
en
v
ir
o
n
m
e
n
t
f
o
r
th
e
r
esear
ch
er
s
to
co
n
d
u
ct
f
air
e
v
alu
atio
n
o
f
t
h
e
alg
o
r
ith
m
s
.
I
t
g
iv
es
an
u
n
a
m
b
ig
u
o
u
s
way
o
f
ass
ess
m
en
t
o
f
th
e
p
er
f
o
r
m
an
c
e
o
f
th
e
m
eth
o
d
s
.
T
h
er
e
ar
e
m
an
y
p
u
b
lically
av
ail
ab
le
d
atasets
with
g
r
o
u
n
d
tr
u
t
h
s
f
o
r
th
e
id
en
tific
atio
n
an
d
s
eg
m
en
tatio
n
o
f
v
ar
io
u
s
lesi
o
n
s
o
f
DR
.
So
m
e
o
f
th
em
ar
e
d
is
cu
s
s
ed
b
elo
w:
2
.
1
.
M
ess
ido
r
[
9
]
I
t
is
a
p
u
b
lically
d
is
tr
ib
u
ted
d
atab
ase
co
n
tain
in
g
1
2
0
0
f
u
n
d
u
s
im
ag
es
in
T
I
FF
f
o
r
m
at
ca
p
tu
r
ed
u
s
in
g
co
lo
r
v
id
eo
3
C
C
D
ca
m
er
a
wi
th
4
5
º
f
ield
o
f
v
iew
(
FOV)
,
8
b
its
p
er
co
lo
r
p
lan
e.
R
eso
lu
tio
n
s
o
f
th
e
im
ag
es
wer
e
s
et
at
1
4
4
0
*
9
6
0
,
2
2
4
0
*
1
4
8
8
o
r
2
3
0
4
*
1
5
3
6
p
i
x
els.
2
.
2
.
Dia
re
t
db
0
[
1
0
]
I
t
co
n
s
is
ts
o
f
1
3
0
im
ag
es
tak
en
at
Ku
o
p
io
u
n
iv
e
r
s
ity
h
o
s
p
ital
with
5
0
º
FOV
an
d
u
n
k
n
o
wn
ca
m
er
a
s
ettin
g
s
.
I
m
ag
es a
r
e
in
p
n
g
f
o
r
m
at
with
r
eso
lu
tio
n
o
f
1
5
0
0
*
1
1
5
2
p
i
x
els.
2
.
3
.
Dia
re
t
db
1
[
1
1
]
I
t
co
m
p
r
is
es
o
f
8
9
im
ag
es
in
p
n
g
f
o
r
m
at.
Ou
t
o
f
th
ese,
2
8
a
n
d
6
1
im
a
g
es
ar
e
f
o
r
tr
ain
in
g
an
d
test
in
g
p
u
r
p
o
s
e
r
esp
ec
tiv
ely
.
I
m
a
g
es
wer
e
tak
en
at
5
0
º
FOV
with
s
ize
o
f
1
5
0
0
*
1
1
5
2
p
i
x
els.
2
.
4
.
E
-
o
ph
t
ha
E
X
[
1
2
]
T
h
er
e
ar
e
8
2
im
ag
es
with
4
5
º
FOV
an
d
in
J
PEG
f
o
r
m
at.
D
ataset
co
n
tain
s
im
ag
es
with
4
d
if
f
er
en
t
s
izes r
an
g
in
g
f
r
o
m
1
4
4
0
*
9
6
0
t
o
2
5
4
4
*
1
6
9
6
p
ix
els.
2
.
5
.
I
DRiD
[
1
3
]
I
t
h
as
5
1
6
im
ag
es
ca
p
tu
r
e
d
with
Ko
wa
VX
-
1
0
α
d
ig
ital
c
am
er
a
with
5
0
º
FOV.
I
m
a
g
es
ar
e
in
jp
g
f
o
r
m
at
with
r
eso
lu
tio
n
o
f
4
2
8
8
*
2
8
4
8
p
ix
els.
I
t
co
n
s
is
ts
o
f
p
ix
el
lev
el
a
n
n
o
tatio
n
s
f
o
r
th
e
DR
lesi
o
n
s
an
d
DR
g
r
ad
i
n
g
at
im
ag
e
lev
el.
4
1
3
an
d
1
0
3
im
ag
es m
ak
e
u
p
th
e
tr
ai
n
in
g
an
d
test
in
g
s
et
r
esp
ec
tiv
e
ly
.
2
.
6
.
Driv
e
[
1
4
]
I
t
co
n
tain
s
4
0
im
ag
es
with
4
5
º
FOV
h
a
v
in
g
s
ize
o
f
7
6
8
*
5
8
4
p
ix
els
in
co
m
p
r
ess
ed
J
PEG
f
o
r
m
at.
I
m
ag
es we
r
e
ca
p
tu
r
ed
u
s
in
g
C
an
o
n
C
R
5
n
o
n
m
y
d
r
iatic
3
C
C
D
ca
m
er
a.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
A
r
ev
iew
o
n
s
u
p
ervis
ed
lea
r
n
in
g
m
eth
o
d
o
lo
g
ies fo
r
d
etec
tio
n
o
f …
(
Ujw
a
la
W.
Wa
s
ek
a
r
)
839
2
.
7
.
St
a
re
[
1
5
]
I
t
co
n
s
is
ts
o
f
2
0
im
ag
es
tak
e
n
b
y
T
o
p
C
o
n
T
R
V
-
5
0
f
u
n
d
u
s
c
am
er
a
at
3
5
º
FOV
with
s
ize
o
f
6
0
5
*
7
0
0
p
ix
els.
T
ab
le
1
s
h
o
ws m
o
r
e
d
e
s
cr
ip
tio
n
ab
o
u
t th
e
d
atab
ases
.
T
ab
le
1
.
Descr
ip
tio
n
o
f
th
e
d
at
ab
ases
D
a
t
a
b
a
s
e
N
o
.
o
f
i
ma
g
e
s
Le
v
e
l
o
f
d
e
t
e
c
t
i
o
n
Ty
p
e
s
o
f
l
e
s
i
o
n
s
d
e
t
e
c
t
e
d
N
o
r
mal
A
b
n
o
r
ma
l
M
e
ss
i
d
o
r
5
4
6
6
5
4
-
-
D
i
a
r
e
t
d
b
0
20
1
1
0
I
mag
e
M
A
,
H
M
,
H
E,
SE
D
i
a
r
e
t
d
b
1
5
84
I
mag
e
M
A
,
H
M
,
H
E,
SE
e
-
o
p
h
t
h
a
EX
35
47
P
i
x
e
l
Ex
u
d
a
t
e
s
I
D
R
i
D
1
6
4
(
p
i
x
e
l
l
e
v
e
l
)
8
1
(
p
i
x
e
l
l
e
v
e
l
)
P
i
x
e
l
,
i
m
a
g
e
M
A
,
H
M
,
H
E,
SE
D
r
i
v
e
33
7
P
i
x
e
l
B
l
o
o
d
v
e
s
sel
S
t
a
r
e
10
10
P
i
x
e
l
B
l
o
o
d
v
e
s
sel
3.
P
E
RF
O
RM
A
NCE
M
E
T
R
I
C
S
Per
f
o
r
m
an
ce
m
ea
s
u
r
es
a
r
e
th
e
ev
alu
atio
n
to
o
ls
th
at
ass
is
t
in
f
in
d
i
n
g
t
h
e
ef
f
icien
cy
o
f
th
e
m
eth
o
d
o
r
tech
n
iq
u
e.
So
m
e
wid
ely
u
s
ed
m
ea
s
u
r
es
a
r
e
s
en
s
itiv
ity
,
s
p
ec
if
icity
an
d
ac
cu
r
ac
y
[
2
]
.
Sen
s
itiv
ity
is
th
e
p
er
ce
n
tag
e
o
f
c
o
r
r
ec
tly
id
en
tif
ied
lesi
o
n
s
.
Sp
ec
if
icity
is
th
e
p
er
ce
n
tag
e
o
f
c
o
r
r
ec
tly
id
en
tif
ied
n
o
n
-
lesi
o
n
s
an
d
ac
cu
r
ac
y
is
th
e
av
e
r
ag
e
o
f
b
o
th
.
T
h
ey
ar
e
g
i
v
en
as
in
:
−
Sen
s
itiv
ity
=
+
−
Sp
ec
if
icity
=
+
−
Acc
u
r
ac
y
=
+
+
+
+
W
h
er
e,
T
P
is
co
r
r
ec
tly
id
en
tifie
d
lesi
o
n
s
,
T
N
is
co
r
r
ec
tly
id
en
tifie
d
n
o
n
-
lesi
o
n
s
,
FP
i
s
i
n
co
r
r
ec
tly
id
en
tifie
d
n
o
n
-
lesi
o
n
s
an
d
FN is in
co
r
r
ec
tly
id
en
tifie
d
lesi
o
n
s
.
4.
CO
M
P
UT
E
R
A
I
DE
D
SYS
T
E
M
S T
O
CL
ASS
I
F
Y
RE
T
I
NAL I
M
AG
E
S H
AVING
E
XUDA
T
E
S
T
h
er
e
ar
e
d
if
f
er
en
t
a
b
n
o
r
m
al
ities
th
at
d
escr
ib
e
th
e
d
iab
etic
r
etin
o
p
ath
y
s
u
ch
as
m
icr
o
a
n
eu
r
y
s
m
s
,
h
ae
m
o
r
r
h
ag
es,
ex
u
d
ates
an
d
n
eo
v
ascu
lar
izatio
n
.
I
n
th
is
p
ap
e
r
,
we
h
av
e
l
im
ited
o
u
r
wo
r
k
t
o
th
e
id
en
tific
atio
n
an
d
class
if
icatio
n
o
f
e
x
u
d
ate
s
.
Mo
s
t
o
f
th
e
tech
n
iq
u
es
g
o
f
o
r
wid
ely
f
o
llo
wed
c
h
ain
o
f
p
r
o
ce
s
s
es
v
iz.
p
r
ep
r
o
ce
s
s
in
g
,
s
eg
m
en
tatio
n
,
f
ea
tu
r
e
ex
tr
ac
tio
n
an
d
cl
ass
if
icatio
n
[
1
6
]
.
T
h
is
s
ec
tio
n
r
e
p
r
esen
ts
th
e
m
eth
o
d
o
l
o
g
ies b
ased
o
n
im
ag
e
p
r
o
ce
s
s
in
g
an
d
s
eg
m
en
tatio
n
,
d
ee
p
lear
n
in
g
a
n
d
m
ac
h
in
e
le
ar
n
in
g
.
4
.
1
.
I
ma
g
e
pro
ce
s
s
ing
a
nd
s
eg
m
ent
a
t
i
o
n m
et
ho
do
lo
g
ies
I
n
im
ag
e
ac
q
u
is
itio
n
,
s
o
m
e
o
f
th
e
n
o
is
e
m
ay
g
et
in
tr
o
d
u
ce
d
in
th
e
im
ag
e
i
n
th
e
f
o
r
m
o
f
u
n
wan
ted
p
ix
els,
u
n
e
v
en
illu
m
i
n
atio
n
.
T
o
g
et
th
e
m
o
s
t
o
u
t
o
f
th
e
im
a
g
e,
im
ag
e
s
h
o
u
ld
b
e
clea
r
a
n
d
h
ig
h
lig
h
t
th
e
o
b
jects
p
r
esen
t
in
th
e
im
ag
e.
T
h
o
u
g
h
,
u
ltra
-
wid
e
-
f
ield
im
ag
es
g
iv
e
wid
er
v
iew
o
f
r
etin
a
[
1
7
]
.
So
m
e
o
f
th
e
r
esear
ch
es
h
av
e
b
ee
n
d
is
cu
s
s
ed
wh
ich
p
r
im
ar
ily
f
o
c
u
s
o
n
th
e
im
a
g
e
p
r
o
ce
s
s
in
g
tech
n
i
q
u
es.
No
is
e
f
ilter
in
g
,
c
o
n
tr
ast
en
h
an
ce
m
e
n
t,
OD
lo
ca
lizatio
n
[
1
8
]
ar
e
s
o
m
e
o
f
t
h
e
asp
ec
ts
o
f
im
ag
e
p
r
o
ce
s
s
in
g
tech
n
iq
u
es.
Sin
ce
o
p
tic
d
is
c
r
esem
b
les
ex
u
d
ates
in
b
r
ig
h
t
n
ess
,
it
is
o
f
u
tm
o
s
t
im
p
o
r
ta
n
ce
to
elim
in
ate
it
[
1
9
]
.
OD
was
lo
ca
lized
u
s
in
g
wate
r
s
h
ed
tr
an
s
f
o
r
m
atio
n
[
2
0
]
an
d
ex
u
d
ates
wer
e
id
e
n
tifie
d
with
th
e
h
elp
o
f
m
o
r
p
h
o
l
o
g
ical
tech
n
iq
u
es.
M
eth
o
d
to
d
etec
t
ex
u
d
ates
in
n
o
n
-
d
ilated
r
etin
al
im
ag
es
u
s
in
g
m
ath
em
at
ical
m
o
r
p
h
o
lo
g
y
[
2
1
]
an
d
f
u
zz
y
c
-
m
ea
n
s
(
FC
M)
clu
s
ter
in
g
[
2
2
]
wer
e
p
r
o
p
o
s
ed
.
Pro
p
o
s
ed
s
y
s
t
em
s
m
ay
b
e
u
s
ef
u
l
in
r
u
r
al
a
r
ea
s
wh
er
e
m
ed
ical
f
ac
ilit
ies
ar
e
p
o
o
r
.
An
a
r
ea
-
b
ased
f
ea
tu
r
e
was
in
t
r
o
d
u
ce
d
[
2
3
]
ca
lle
d
as
v
ein
r
em
o
v
al
ter
m
(
VR
T
)
.
T
h
e
m
eth
o
d
was
v
alid
at
ed
o
n
d
ia
r
etd
b
1
d
atab
ase
at
im
ag
e
lev
el
an
d
o
n
e
-
o
p
h
th
a
E
X
d
atab
ase
at
le
s
io
n
lev
el.
Mid
p
o
in
t
cir
cle
alg
o
r
ith
m
w
as
u
s
ed
to
r
em
o
v
e
OD.
I
m
a
g
e
was
d
iv
id
ed
an
d
class
if
ied
in
to
ex
u
d
ate
an
d
ex
u
d
ate
f
r
ee
.
T
h
en
t
h
e
s
u
b
im
ag
e
was
s
eg
m
en
ted
u
s
in
g
s
alien
cy
m
eth
o
d
[
2
4
]
.
C
o
n
s
id
er
in
g
th
e
tim
e
co
n
s
tr
ain
t
an
d
ju
d
icio
u
s
u
s
e
o
f
r
eso
u
r
ce
s
,
MA
T
L
AB
an
d
FP
GA
b
ase
d
s
o
lu
tio
n
s
h
av
e
b
ee
n
p
r
o
p
o
s
ed
[
2
5
]
.
T
h
e
s
y
s
tem
wen
t
th
r
o
u
g
h
all
th
e
r
eq
u
ir
e
d
p
r
ep
r
o
ce
s
s
in
g
tech
n
iq
u
es.
W
it
h
th
e
u
s
e
o
f
s
en
s
o
r
-
b
ased
s
cr
ee
n
in
g
s
y
s
tem
s
an
d
clo
u
d
s
o
f
twar
e,
a
telem
ed
icin
e
m
eth
o
d
[
2
6
]
was
p
u
t
f
o
r
wa
r
d
to
id
en
tify
h
a
r
d
ex
u
d
ates
in
th
e
r
etin
al
im
ag
es.
Data
was
r
ea
d
b
y
d
-
E
y
e
s
e
n
s
o
r
.
An
o
th
er
m
eth
o
d
u
s
ed
v
ar
io
u
s
tech
n
iq
u
es
s
u
ch
as
Gam
m
a
co
r
r
ec
tio
n
,
C
L
AHE
,
p
r
in
cip
al
co
m
p
o
n
en
t
an
aly
s
is
an
d
co
n
v
e
x
h
u
ll
tr
an
s
f
o
r
m
[
2
7
]
f
o
r
lu
m
in
o
s
ity
,
co
n
tr
ast
en
h
an
ce
m
en
t,
v
ess
el
ex
tr
ac
tio
n
an
d
OD
d
ete
ctio
n
r
esp
ec
tiv
ely
.
Mo
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
wer
e
u
s
ed
to
d
etec
t
lesi
o
n
s
in
th
e
f
u
n
d
u
s
im
ag
e
an
d
class
if
icatio
n
was
d
o
n
e
b
ased
o
n
th
e
c
o
u
n
t
o
f
lesi
o
n
s
.
A
n
ew
alg
o
r
ith
m
[
2
8
]
ca
lled
as
‘
m
o
at
o
p
er
ato
r
’
b
ased
o
n
r
ec
u
r
s
iv
e
r
eg
io
n
g
r
o
win
g
s
eg
m
e
n
tatio
n
alg
o
r
ith
m
s
,
was
d
ev
el
o
p
ed
to
r
ec
o
g
n
ize
HE
s
,
MA
s
an
d
HM
s
.
Do
m
ain
k
n
o
wled
g
e
a
b
o
u
t
t
h
e
lesi
o
n
was
u
s
ed
to
co
r
r
ec
tly
id
en
tif
y
th
e
ex
u
d
ates
[
2
9
]
.
Af
ter
m
ed
ian
f
ilter
in
g
an
d
d
y
n
am
ic
clu
s
ter
in
g
,
d
o
m
ain
k
n
o
w
led
g
e
was a
p
p
lied
to
d
etec
t tr
u
e
h
ar
d
e
x
u
d
ates.
T
ab
le
2
d
e
p
icts
th
e
d
atab
ases
u
s
ed
an
d
p
er
f
o
r
m
an
ce
s
o
f
th
e
ab
o
v
e
m
et
h
o
d
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
23
,
No
.
2
,
Au
g
u
s
t 2
0
2
1
:
837
-
8
4
6
840
T
ab
le
2
.
Per
f
o
r
m
an
ce
s
o
f
m
et
h
o
d
s
in
d
etec
tin
g
ex
u
d
ates u
s
in
g
im
ag
e
p
r
o
ce
s
s
in
g
tech
n
iq
u
es
A
u
t
h
o
r
(
s)
M
e
t
h
o
d
o
l
o
g
y
D
a
t
a
b
a
s
e
P
e
r
f
o
r
ma
n
c
e
(
%)
V
e
n
k
a
t
a
l
a
k
s
h
m
i
e
t
a
l
.
[
1
8
]
G
r
a
p
h
i
c
a
l
U
s
e
r
I
n
t
e
r
f
a
c
e
D
r
i
v
e
D
i
a
r
e
t
d
b
0
S
e
n
s
i
t
i
v
i
t
y
8
2
S
a
r
a
n
y
a
a
n
d
U
mam
a
h
e
sw
a
r
i
[
1
9
]
O
D
e
l
i
mi
n
a
t
i
o
n
C
a
n
d
i
d
a
t
e
l
e
si
o
n
s
e
g
.
M
e
ss
i
d
o
r
A
c
c
u
r
a
c
y
9
7
.
5
4
S
e
n
s
i
t
i
v
i
t
y
9
0
.
3
4
S
p
e
c
i
f
i
c
i
t
y
9
8
.
2
4
W
a
l
t
e
r
e
t
a
l
.
[
2
0
]
M
o
r
p
h
o
l
o
g
i
c
a
l
t
e
c
h
n
i
q
u
e
,
W
a
t
e
r
sh
e
d
t
r
a
n
sf
o
r
m
P
r
i
v
a
t
e
S
e
n
s
i
t
i
v
i
t
y
9
2
.
8
M
e
a
n
p
r
e
d
i
c
t
i
v
e
v
a
l
u
e
9
2
.
4
S
o
p
h
a
r
a
k
e
t
a
l
.
[
2
1
]
N
o
n
-
d
i
l
a
t
e
d
i
m
a
g
e
s,
M
a
t
h
e
mat
i
c
a
l
m
o
r
p
h
o
l
o
g
y
P
r
i
v
a
t
e
S
e
n
s
i
t
i
v
i
t
y
8
0
S
p
e
c
i
f
i
c
i
t
y
9
9
.
5
S
o
p
h
a
r
a
k
e
t
a
l
.
[
2
2
]
N
o
n
-
d
i
l
a
t
e
d
i
m
a
g
e
s,
F
u
z
z
y
C
-
mea
n
s
C
l
u
st
e
r
i
n
g
P
r
i
v
a
t
e
S
e
n
s
i
t
i
v
i
t
y
8
7
.
2
8
S
p
e
c
i
f
i
c
i
t
y
9
9
.
2
4
A
c
c
u
r
a
c
y
9
9
.
1
1
P
P
V
4
2
.
7
7
P
LR
2
2
4
.
2
6
Jo
sh
i
a
n
d
K
a
r
u
l
e
[
2
3
]
M
o
r
p
h
o
l
o
g
i
c
a
l
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
(
V
R
T)
D
i
a
r
e
t
d
b
1
(
1
)
e
-
o
p
h
t
h
a
EX
(
2
)
S
e
n
s
i
t
i
v
i
t
y
9
6
.
7
(
1
)
S
p
e
c
i
f
i
c
i
t
y
8
5
.
4
(
1
)
A
c
c
u
r
a
c
y
9
1
(
1
)
S
e
n
s
i
t
i
v
i
t
y
8
0
.
5
(
2
)
A
c
c
u
r
a
c
y
9
5
(
2
)
N
u
r
a
n
d
Tj
a
n
d
r
a
s
a
[
2
4
]
S
a
l
i
e
n
c
y
me
t
h
o
d
b
a
s
e
d
o
n
r
e
g
i
o
n
D
i
a
r
e
t
d
b
1
S
e
n
s
i
t
i
v
i
t
y
8
1
.
6
5
S
p
e
c
i
f
i
c
i
t
y
9
9
.
4
2
A
c
c
u
r
a
c
y
9
9
.
3
3
S
a
t
y
a
n
a
n
d
a
e
t
a
l
.
[
2
5
]
F
P
G
A
a
n
d
M
A
TLA
B
b
a
se
d
S
o
l
u
t
i
o
n
D
i
a
r
e
t
d
b
0
D
i
a
r
e
t
d
b
1
I
D
R
i
D
M
e
ss
i
d
o
r
A
c
c
u
r
a
c
y
9
0
S
a
e
e
d
e
t
a
l
.
[
2
6
]
Te
l
e
me
d
i
c
i
n
e
S
y
s
t
e
m
b
a
se
d
o
n
d
-
E
y
e
sen
s
o
r
P
r
i
v
a
t
e
S
e
n
s
i
t
i
v
i
t
y
1
0
0
S
p
e
c
i
f
i
c
i
t
y
9
6
A
c
c
u
r
a
c
y
9
8
K
a
n
i
mo
z
h
i
e
t
a
l
.
[
2
7
]
G
a
mm
a
c
o
r
r
e
c
t
i
o
n
,
C
LA
H
E,
P
C
A
,
C
o
n
v
e
x
H
u
l
l
t
r
a
n
sf
o
r
m
D
r
i
v
e
S
t
a
r
e
D
i
a
r
e
t
d
b
1
M
e
ss
i
d
o
r
R
O
C
S
e
n
s
i
t
i
v
i
t
y
9
9
.
0
3
S
p
e
c
i
f
i
c
i
t
y
9
6
.
3
7
A
c
c
u
r
a
c
y
9
6
.
9
8
S
i
n
t
h
a
n
a
y
o
t
h
i
n
e
t
a
l
.
[
2
8
]
M
o
a
t
o
p
e
r
a
t
o
r
P
r
i
v
a
t
e
S
e
n
s
i
t
i
v
i
t
y
8
8
.
5
S
p
e
c
i
f
i
c
i
t
y
9
9
.
7
H
su
e
t
a
l
.
[
2
9
]
D
o
ma
i
n
k
n
o
w
l
e
d
g
e
P
r
i
v
a
t
e
S
e
n
s
i
t
i
v
i
t
y
1
0
0
S
p
e
c
i
f
i
c
i
t
y
7
4
4
.
2
.
Dee
p
lea
rning
t
ec
hn
iqu
es
R
e
c
e
n
t
l
y
,
d
e
e
p
l
e
a
r
n
i
n
g
(
D
L
)
i
s
g
a
i
n
i
n
g
p
o
p
u
l
a
r
i
t
y
b
e
c
a
u
s
e
o
f
i
t
s
f
u
n
c
t
i
o
n
a
l
i
t
y
a
n
d
a
p
p
l
i
c
a
b
i
l
i
t
y
.
D
e
e
p
l
e
a
r
n
i
n
g
i
s
a
c
o
m
p
u
t
e
r
-
a
i
d
e
d
m
e
d
i
c
a
l
d
i
a
g
n
o
s
i
s
s
y
s
t
e
m
[
3
0
]
.
I
t
i
s
a
b
r
a
n
c
h
o
f
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
t
h
a
t
c
o
n
s
i
s
t
s
o
f
n
o
n
-
l
i
n
e
a
r
i
n
f
o
r
m
a
t
i
o
n
p
r
o
c
e
s
s
i
n
g
a
r
c
h
i
t
e
c
t
u
r
e
s
.
I
t
i
s
d
e
s
i
g
n
e
d
f
o
r
t
h
e
p
u
r
p
o
s
e
o
f
u
n
s
u
p
e
r
v
i
s
e
d
f
e
a
t
u
r
e
l
e
a
r
n
i
n
g
a
n
d
p
a
t
t
e
r
n
a
n
a
l
y
s
i
s
[
3
1
]
.
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
(
C
N
N
)
i
s
t
h
e
m
o
s
t
w
i
d
e
l
y
u
s
e
d
m
e
t
h
o
d
o
f
D
L
[
3
2
]
.
T
h
e
t
h
r
e
e
m
a
i
n
l
a
y
e
r
s
o
f
C
N
N
i
n
c
l
u
d
e
c
o
n
v
o
l
u
t
i
o
n
l
a
y
e
r
,
p
o
o
l
i
n
g
l
a
y
e
r
a
n
d
f
u
l
l
y
c
o
n
n
e
c
t
e
d
l
a
y
e
r
[
3
3
]
.
M
u
c
h
e
m
p
h
a
s
i
s
h
a
s
b
e
e
n
g
i
v
e
n
t
o
t
h
e
a
u
t
o
m
a
t
i
c
d
e
t
e
c
t
i
o
n
o
f
d
i
s
e
a
s
e
s
u
s
i
n
g
D
L
m
e
t
h
o
d
s
r
e
c
e
n
t
l
y
.
A
n
i
m
a
g
e
d
i
c
t
i
o
n
a
r
y
w
a
s
b
u
i
l
t
u
p
w
i
t
h
e
a
c
h
i
m
a
g
e
o
f
t
h
e
s
i
z
e
1
0
*
1
0
p
i
x
e
l
s
[
3
4
]
c
u
t
o
u
t
o
f
t
h
e
d
a
t
a
b
a
s
e
i
m
a
g
e
s
.
T
h
e
i
m
a
g
e
s
w
e
r
e
d
i
v
i
d
e
d
i
n
t
o
t
r
a
i
n
i
n
g
,
t
e
s
t
i
n
g
a
n
d
v
a
l
i
d
a
t
i
o
n
g
r
o
u
p
s
.
U
s
i
n
g
T
e
n
s
o
r
F
l
o
w
a
n
d
K
e
r
a
s
l
i
b
r
a
r
i
e
s
,
a
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
t
w
o
r
k
i
n
P
y
t
h
o
n
w
a
s
b
u
i
l
t
t
o
d
e
t
e
c
t
t
h
e
e
x
u
d
a
t
e
s
.
C
N
N
s
c
a
n
a
l
s
o
b
e
p
r
e
t
r
a
i
n
e
d
t
o
s
o
l
v
e
t
h
e
s
p
e
c
i
f
i
c
p
r
o
b
l
e
m
.
V
a
r
i
o
u
s
p
r
e
t
r
a
i
n
e
d
C
N
N
m
o
d
e
l
s
s
u
c
h
a
s
I
n
c
e
p
t
i
o
n
-
v
3
,
r
e
s
i
d
u
a
l
n
e
t
w
o
r
k
-
5
0
a
n
d
v
i
s
u
a
l
g
e
o
m
e
t
r
y
g
r
o
u
p
n
e
t
w
o
r
k
-
1
9
w
e
r
e
u
s
e
d
f
o
r
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
a
n
d
c
l
a
s
s
i
f
i
c
a
t
i
o
n
[
3
5
]
.
B
a
s
e
d
o
n
t
r
a
n
s
f
e
r
l
e
a
r
n
i
n
g
,
t
h
e
f
e
a
t
u
r
e
s
f
r
o
m
a
l
l
t
h
e
t
h
r
e
e
m
o
d
e
l
s
w
e
r
e
c
o
m
b
i
n
e
d
t
o
f
o
r
m
a
s
i
n
g
l
e
f
e
a
t
u
r
e
v
e
c
t
o
r
(
f
e
a
t
u
r
e
s
f
u
s
i
o
n
)
a
n
d
f
e
d
t
o
S
o
f
t
m
a
x
c
l
a
s
s
i
f
i
e
r
.
I
m
a
g
e
p
r
o
c
e
s
s
i
n
g
i
s
t
h
e
f
i
r
s
t
a
n
d
b
a
s
i
c
r
e
q
u
i
r
e
m
e
n
t
i
n
a
n
y
p
a
t
t
e
r
n
d
e
t
e
c
t
i
o
n
a
n
d
c
l
a
s
s
i
f
i
c
a
t
i
o
n
s
e
t
u
p
.
W
e
h
a
v
e
d
i
s
c
u
s
s
e
d
e
f
f
e
c
t
i
v
e
i
m
a
g
e
p
r
o
c
e
s
s
i
n
g
t
e
c
h
n
i
q
u
e
s
i
n
t
h
e
s
e
c
t
i
o
n
‘
i
m
a
g
e
p
r
o
c
e
s
s
i
n
g
a
n
d
s
e
g
m
e
n
t
a
t
i
o
n
m
e
t
h
o
d
o
l
o
g
i
e
s
.
’
I
m
a
g
e
p
r
o
c
e
s
s
i
n
g
c
o
m
b
i
n
e
d
w
i
t
h
C
N
N
c
a
n
d
o
t
h
e
w
o
n
d
e
r
s
.
P
e
r
f
o
r
m
a
n
c
e
o
f
C
N
N
i
n
t
e
g
r
a
t
e
d
w
i
t
h
c
i
r
c
u
l
a
r
H
o
u
g
h
t
r
a
n
s
f
o
r
m
a
t
i
o
n
[
3
6
]
w
a
s
c
o
m
p
a
r
e
d
w
i
t
h
C
N
N
b
a
s
e
d
p
e
r
f
o
r
m
a
n
c
e
a
l
o
n
e
a
n
d
f
o
u
n
d
o
u
t
t
h
a
t
f
o
r
m
e
r
g
a
v
e
b
e
t
t
e
r
r
e
s
u
l
t
s
.
A
n
o
t
h
e
r
C
N
N
m
o
d
e
l
w
a
s
t
r
a
i
n
e
d
u
s
i
n
g
p
a
t
c
h
e
s
o
f
s
i
z
e
6
4
*
6
4
p
i
x
e
l
s
e
x
t
r
a
c
t
e
d
f
r
o
m
t
h
e
e
x
u
d
a
t
e
i
m
a
g
e
s
[
3
7
]
.
T
h
e
s
e
c
a
n
d
i
d
a
t
e
p
i
x
e
l
s
w
e
r
e
t
a
k
e
n
o
u
t
u
s
i
n
g
m
o
r
p
h
o
l
o
g
i
c
a
l
u
l
t
i
m
a
t
e
o
p
e
n
i
n
g
.
A
n
o
n
-
lin
ea
r
class
if
icatio
n
an
d
lear
n
in
g
ab
ilit
y
m
a
k
e
th
e
n
eu
r
al
n
etwo
r
k
m
o
r
e
ef
f
icien
t.
A
th
r
ee
-
lay
er
p
e
r
ce
p
tr
o
n
n
eu
r
al
n
etw
o
r
k
was
tr
ain
ed
to
d
is
tin
g
u
is
h
b
etwe
en
th
e
ex
u
d
ates
an
d
n
o
n
-
e
x
u
d
ates.
A
f
ter
p
r
ep
r
o
ce
s
s
in
g
,
FC
M
clu
s
ter
in
g
was
u
s
ed
f
o
r
s
eg
m
e
n
tatio
n
[
3
8
]
,
[
3
9
]
.
Osar
eh
et
a
l.
[
3
9
]
,
a
f
ea
tu
r
e
v
ec
to
r
was
g
en
er
ated
with
th
e
h
el
p
o
f
g
e
n
etic
alg
o
r
ith
m
.
A
n
o
t
h
e
r
A
N
N
w
a
s
u
s
e
d
t
o
d
is
c
r
i
m
i
n
a
t
e
b
et
we
e
n
l
e
s
i
o
n
s
w
i
t
h
t
h
e
h
e
l
p
o
f
f
e
a
t
u
r
e
s
s
u
c
h
a
s
c
o
l
o
r
,
s
i
z
e
,
s
h
a
p
e
a
n
d
t
e
x
t
u
r
e
a
p
p
l
y
i
n
g
b
a
c
k
p
r
o
p
a
g
a
t
i
o
n
a
l
g
o
r
i
t
h
m
[
4
0
]
.
A
f
u
z
z
y
n
e
u
r
a
l
n
e
t
w
o
r
k
w
a
s
p
r
o
p
o
s
e
d
wi
t
h
s
el
f
-
o
r
g
a
n
i
z
i
n
g
l
a
y
e
r
a
n
d
m
u
l
t
i
l
ay
e
r
p
e
r
c
e
p
t
r
o
n
(
M
L
P
)
[
4
1
]
.
S
e
l
f
-
o
r
g
a
n
i
z
i
n
g
l
a
y
e
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
A
r
ev
iew
o
n
s
u
p
ervis
ed
lea
r
n
in
g
m
eth
o
d
o
lo
g
ies fo
r
d
etec
tio
n
o
f …
(
Ujw
a
la
W.
Wa
s
ek
a
r
)
841
p
e
r
f
o
r
m
e
d
t
h
e
t
a
s
k
o
f
d
e
t
e
ct
i
n
g
t
h
e
l
e
s
i
o
n
s
a
n
d
g
r
o
u
p
i
n
g
t
h
e
m
a
c
c
o
r
d
i
n
g
t
o
t
h
e
i
r
m
e
m
b
er
s
h
i
p
v
a
l
u
e
s
,
w
h
i
le
M
L
P
n
e
t
w
o
r
k
c
l
as
s
i
f
i
e
d
t
h
e
m
.
M
L
P
a
g
a
i
n
p
r
o
v
e
d
i
t
s
s
u
p
e
r
i
o
r
i
t
y
i
n
[
4
2
]
a
s
c
o
m
p
a
r
e
d
t
o
o
t
h
e
r
t
h
r
e
e
c
l
a
s
s
i
f
i
e
r
s
.
Dec
o
m
p
o
s
in
g
f
u
n
d
u
s
im
ag
e
in
to
v
a
r
io
u
s
lay
e
r
s
,
ML
P
g
av
e
b
etter
r
esu
lts
o
n
d
iar
et
d
b
1
d
atab
ase
[
4
3
]
.
Ad
ap
tiv
e
n
eu
r
o
-
f
u
zz
y
in
f
er
en
ce
s
y
s
tem
(
ANFI
S)
was
u
s
e
d
to
d
is
tin
g
u
is
h
b
etwe
en
n
o
r
m
al
an
d
ab
n
o
r
m
a
l
im
ag
es [
4
4
]
with
tex
tu
r
e
en
h
a
n
ce
m
en
t u
s
in
g
m
o
r
p
h
o
lo
g
ical
co
m
p
o
n
en
t a
n
aly
s
is
(
MCA)
.
R
ec
en
tly
,
m
an
y
n
ew
an
d
u
p
g
r
a
d
ed
a
p
p
r
o
ac
h
es
em
e
r
g
ed
as
a
to
o
l
to
d
iag
n
o
s
e
lesi
o
n
s
s
u
ch
as
ch
a
n
n
el
atten
tio
n
m
eth
o
d
[
4
5
]
,
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
n
etwo
r
k
(
L
STM
)
[
4
6
]
an
d
ca
p
s
u
le
n
etwo
r
k
[
4
7
]
.
W
o
r
k
in
g
o
n
d
if
f
e
r
en
t
d
atab
ases
,
th
ese
m
eth
o
d
s
g
av
e
p
r
o
m
is
in
g
r
esu
l
ts
.
Mo
r
e
d
etailed
in
s
ig
h
t o
f
th
e
m
eth
o
d
o
lo
g
ies is
g
iv
en
in
T
a
b
le
3
.
T
ab
le
3
.
E
x
u
d
ate
d
etec
tio
n
u
s
i
n
g
d
ee
p
lear
n
in
g
A
u
t
h
o
r
s
M
e
t
h
o
d
o
l
o
g
y
D
a
t
a
b
a
s
e
C
l
a
s
si
f
i
e
r
F
e
a
t
u
r
e
s
P
e
r
f
o
r
ma
n
c
e
(
%)
F
u
t
u
r
e
S
c
o
p
e
A
u
c
c
a
h
u
a
si
e
t
al
.
[
3
4
]
I
mag
e
D
i
c
t
i
o
n
a
r
y
D
i
a
r
e
t
d
b
1
C
N
N
-
S
e
n
s
i
t
i
v
i
t
y
9
2
S
p
e
c
i
f
i
c
i
t
y
9
2
-
M
a
t
e
e
n
e
t
a
l
.
[
3
5
]
P
r
e
t
r
a
i
n
e
d
C
N
N
s,
F
e
a
t
u
r
e
s f
u
si
o
n
,
Tr
a
n
sf
e
r
Le
a
r
n
i
n
g
e
-
o
p
h
t
h
a
D
i
a
r
e
t
d
b
1
S
o
f
t
ma
x
-
A
c
c
u
r
a
c
y
9
8
.
4
3
(
e
-
o
p
h
t
h
a
)
A
c
c
u
r
a
c
y
9
8
.
9
1
(
D
i
a
r
e
t
d
b
1
)
To
d
i
s
c
r
i
m
i
n
a
t
e
b
e
t
w
e
e
n
h
a
r
d
a
n
d
s
o
f
t
e
x
u
d
a
t
e
s,
t
o
d
i
a
g
n
o
se
H
M
a
n
d
M
A
K
e
m
a
l
A
d
e
m
[
3
6
]
C
i
r
c
u
l
a
r
H
o
u
g
h
Tr
a
n
sf
o
r
m
D
i
a
r
e
t
d
b
0
D
i
a
r
e
t
d
b
1
D
r
i
md
b
C
N
N
-
S
e
n
s
i
t
i
v
i
t
y
1
0
0
S
p
e
c
i
f
i
c
i
t
y
9
8
.
4
1
TC
C
9
9
.
1
7
-
Y
u
e
t
a
l
.
[
3
7
]
Lo
c
a
l
p
h
a
s
e
sy
mm
e
t
r
y
a
l
g
o
r
i
t
h
m
f
o
r
O
D
d
e
t
e
c
t
i
o
n
e
-
o
p
h
t
h
a
EX
C
N
N
-
S
e
n
s
i
t
i
v
i
t
y
8
8
.
8
5
S
p
e
c
i
f
i
c
i
t
y
9
6
A
c
c
u
r
a
c
y
9
1
.
9
2
Te
st
i
n
g
o
n
M
e
ss
i
d
o
r
a
n
d
D
i
a
r
e
t
d
b
d
a
t
a
b
a
se
s
O
sareh
e
t
a
l
.
[
3
8
]
F
u
z
z
y
c
m
e
a
n
s f
o
r
seg
m
e
n
t
a
t
i
o
n
P
r
i
v
a
t
e
Th
r
e
e
-
l
a
y
e
r
p
e
r
c
e
p
t
r
o
n
n
e
u
r
a
l
n
e
t
w
o
r
k
18
S
e
n
s
i
t
i
v
i
t
y
9
5
(
i
ma
g
e
)
S
p
e
c
i
f
i
c
i
t
y
8
8
.
9
(
i
m
a
g
e
)
S
e
n
s
i
t
i
v
i
t
y
9
3
(
l
e
s
i
o
n
)
S
p
e
c
i
f
i
c
i
t
y
9
4
.
1
(
l
e
si
o
n
)
-
O
sareh
e
t
a
l
.
[
3
9
]
I
mag
e
se
g
me
n
t
a
t
i
o
n
b
a
s
e
d
o
n
F
C
M
,
F
e
a
t
u
r
e
s
sel
e
c
t
i
o
n
b
a
s
e
d
o
n
G
A
P
r
i
v
a
t
e
M
u
l
t
i
l
a
y
e
r
p
e
r
c
e
p
t
r
o
n
NN
65
S
e
n
s
i
t
i
v
i
t
y
9
6
(
i
ma
g
e
)
S
p
e
c
i
f
i
c
i
t
y
9
4
.
6
(
i
m
a
g
e
)
S
e
n
s
i
t
i
v
i
t
y
9
3
.
5
(
p
i
x
e
l
)
P
r
e
d
i
c
t
i
v
i
t
y
9
2
.
1
(
p
i
x
e
l
)
To
e
st
a
b
l
i
s
h
sp
a
t
i
a
l
r
e
l
a
t
i
o
n
s
h
i
p
b
e
t
w
e
e
n
e
x
u
d
a
t
e
s a
n
d
f
o
v
e
a
,
d
e
t
e
c
t
i
o
n
o
f
r
e
d
l
e
si
o
n
s
F
r
a
n
k
l
i
n
a
n
d
R
a
j
a
n
[
4
0
]
Le
si
o
n
b
a
se
d
e
v
a
l
u
a
t
i
o
n
D
i
a
r
e
t
d
b
1
M
u
l
t
i
l
a
y
e
r
p
e
r
c
e
p
t
r
o
n
n
e
u
r
a
l
n
e
t
w
o
r
k
15
S
e
n
s
i
t
i
v
i
t
y
9
6
.
3
S
p
e
c
i
f
i
c
i
t
y
9
9
.
8
A
c
c
u
r
a
c
y
9
9
.
7
P
r
e
d
i
c
t
i
v
e
v
a
l
u
e
9
3
.
7
I
n
c
l
u
s
i
o
n
o
f
M
A
,
H
M
i
n
t
h
e
sy
st
e
m
A
k
r
a
m
a
n
d
K
h
a
n
[
4
1
]
A
v
e
r
a
g
e
f
i
l
t
e
r
,
t
h
r
e
s
h
o
l
d
i
n
g
,
H
o
u
g
h
t
r
a
n
sf
o
r
m
D
r
i
v
e
S
t
a
r
e
D
i
a
r
e
t
d
b
0
D
i
a
r
e
t
d
b
1
F
u
z
z
y
n
e
u
r
a
l
n
e
t
w
o
r
k
6
A
c
c
u
r
a
c
y
9
4
.
7
3
-
Th
e
e
r
a
-
U
mp
o
n
e
t
a
l
.
[
4
2
]
H
E
d
e
t
e
c
t
i
o
n
u
s
i
n
g
4
d
i
f
f
e
r
e
n
t
c
l
a
ss
i
f
i
e
r
s
D
i
a
r
e
t
d
b
1
M
LP
S
V
M
H
i
e
r
a
r
c
h
i
c
a
l
A
N
F
I
S
C
N
N
1
4
(
f
o
r
M
LP,
S
V
M
a
n
d
H
A
N
F
I
S
)
A
U
C
0
.
9
9
8
(
M
LP)
0
.
9
9
7
(
S
V
M
)
0
.
9
8
7
(
H
A
N
F
I
S
)
0
.
9
5
1
(
C
N
N
)
-
R
o
m
e
r
o
-
O
r
a
a
e
t
a
l
.
[
4
3
]
I
mag
e
d
e
c
o
m
p
o
s
i
t
i
o
n
i
n
t
o
l
a
y
e
r
s
D
i
a
r
e
t
d
b
1
M
LP
34
S
e
n
s
i
t
i
v
i
t
y
9
5
S
p
e
c
i
f
i
c
i
t
y
9
0
.
2
4
A
c
c
u
r
a
c
y
9
1
.
8
0
S
u
p
e
r
v
i
se
d
c
l
a
ss
i
f
i
c
a
t
i
o
n
f
o
r
b
l
o
o
d
v
e
ss
e
l
seg
m
e
n
t
s
V
a
l
a
r
ma
t
h
i
a
n
d
S
a
r
a
v
a
n
a
n
[
4
4
]
M
C
A
b
a
se
d
seg
m
e
n
t
a
t
i
o
n
S
t
a
r
e
M
e
ss
i
d
o
r
A
N
F
I
S
12
A
v
g
e
r
r
o
r
r
a
t
e
0
.
0
1
5
F
-
sco
r
e
0
.
9
6
8
A
c
c
u
r
a
c
y
8
8
K
a
p
p
a
v
a
l
u
e
0
.
7
9
1
-
S
i
e
t
a
l
.
[
4
5
]
C
h
a
n
n
e
l
a
t
t
e
n
t
i
o
n
met
h
o
d
e
-
o
p
h
t
h
a
(
1
)
H
EI
-
M
ED
(
2
)
F
C
N
-
S
e
n
s
i
t
i
v
i
t
y
8
9
(
1
)
P
P
V
8
1
(
1
)
F
1
8
2
(
1
)
P
a
r
a
l
l
e
l
p
r
o
c
e
ss
i
n
g
t
o
sav
e
t
i
me
A
sh
i
r
e
t
a
l
.
[
4
6
]
H
a
r
a
l
i
c
k
f
e
a
t
u
r
e
s
D
i
a
r
e
t
d
b
1
LSTM
20
S
e
n
s
i
t
i
v
i
t
y
9
1
.
6
5
S
p
e
c
i
f
i
c
i
t
y
9
5
.
4
5
A
c
c
u
r
a
c
y
9
5
.
4
5
P
P
V
9
9
.
3
4
D
e
t
e
c
t
i
o
n
o
f
M
A
a
n
d
H
M
K
a
l
y
a
n
i
e
t
a
l
.
[
4
7
]
C
p
a
su
l
e
n
e
t
w
o
r
k
M
e
ss
i
d
o
r
S
o
f
t
ma
x
-
A
c
c
u
r
a
c
y
9
7
.
9
8
(
h
e
a
l
t
h
y
r
e
t
i
n
a
)
A
c
c
u
r
a
c
y
9
7
.
6
5
(
st
a
g
e
1
)
A
c
c
u
r
a
c
y
9
7
.
6
5
(
st
a
g
e
2
)
A
c
c
u
r
a
c
y
9
8
.
6
4
(
st
a
g
e
3
)
C
l
a
s
si
f
i
c
a
t
i
o
n
f
o
r
a
l
l
p
o
ssi
b
l
e
c
l
a
ss
e
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
23
,
No
.
2
,
Au
g
u
s
t 2
0
2
1
:
837
-
8
4
6
842
4
.
3
.
M
a
chine
lea
rning
a
pp
ro
a
ch
Ma
ch
in
e
lear
n
in
g
(
ML
)
is
a
p
ar
t
o
f
ar
tific
ial
in
tellig
en
ce
i
n
wh
ich
th
e
s
y
s
tem
lear
n
s
f
r
o
m
d
ata
with
th
e
in
ten
tio
n
o
f
a
n
aly
zin
g
t
h
e
p
atter
n
an
d
tak
e
d
ec
is
io
n
s
with
o
u
t
h
u
m
a
n
in
ter
v
en
tio
n
.
Dif
f
er
en
ce
b
etwe
en
m
ac
h
in
e
lear
n
in
g
an
d
d
ee
p
lea
r
n
in
g
is
ill
u
s
tr
ated
in
T
ab
le
4
[
3
3
]
.
T
ab
le
4
.
Dif
f
e
r
en
ce
b
etwe
en
m
ac
h
in
e
lear
n
in
g
an
d
d
ee
p
lea
r
n
in
g
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
D
e
e
p
l
e
a
r
n
i
n
g
H
a
n
d
-
d
r
a
w
n
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
R
e
q
u
i
r
e
d
N
o
t
r
e
q
u
i
r
e
d
Tr
a
i
n
i
n
g
d
a
t
a
La
r
g
e
d
a
t
a
n
o
t
r
e
q
u
i
r
e
d
La
r
g
e
d
a
t
a
r
e
q
u
i
r
e
d
R
esear
ch
in
th
e
f
ield
o
f
ML
in
clu
d
es
m
a
n
y
s
tate
-
of
-
th
e
-
a
r
t
ap
p
r
o
ac
h
es.
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
is
a
ML
m
eth
o
d
s
p
ec
if
ically
u
s
ed
f
o
r
b
in
ar
y
class
if
icatio
n
.
E
x
u
d
ates
wer
e
d
is
tin
g
u
is
h
ed
u
s
in
g
SVM
f
o
llo
win
g
s
o
m
e
p
r
ep
r
o
ce
s
s
in
g
tech
n
iq
u
es.
L
esio
n
s
wer
e
d
et
e
cted
with
th
e
h
elp
o
f
m
o
r
p
h
o
lo
g
ical
o
p
en
in
g
an
d
clo
s
in
g
[
4
8
]
.
SVM
b
ased
o
n
r
ad
ial
b
asis
f
u
n
ctio
n
(
R
B
F)
d
etec
ted
th
e
h
ar
d
ex
u
d
ates
[
4
9
]
.
I
t
is
a
co
m
m
o
n
p
r
ac
tice
to
f
o
llo
w
th
e
co
n
v
en
tio
n
al
p
r
o
ce
s
s
f
o
r
d
etec
tio
n
o
f
lesi
o
n
s
u
s
in
g
C
AD
s
y
s
tem
s
i.
e.
p
r
ep
r
o
ce
s
s
in
g
,
s
eg
m
en
tatio
n
,
f
ea
tu
r
e
ex
tr
ac
t
io
n
an
d
class
if
icatio
n
[
5
0
]
.
B
ased
o
n
m
o
r
p
h
o
lo
g
ical
o
p
er
ato
r
s
an
d
tex
tu
r
e
f
ea
tu
r
e
a
n
aly
s
is
,
SVM
was
tr
ain
ed
t
o
id
e
n
tify
t
h
e
im
a
g
e
s
as
n
o
r
m
al
an
d
ab
n
o
r
m
al
[
5
1
]
.
Dr
u
s
en
is
an
ab
n
o
r
m
ality
th
at
ap
p
ea
r
s
o
n
r
e
tin
a
an
d
r
esem
b
les
t
o
e
x
u
d
ate
s
.
I
t
is
th
e
lesi
o
n
r
elate
d
to
m
a
cu
lar
d
e
g
en
er
atio
n
.
I
ts
d
etec
tio
n
is
n
ec
ess
ar
y
in
o
r
d
er
to
i
d
en
tify
e
x
u
d
ates
co
r
r
ec
tly
.
SVM
with
R
B
F
was
tr
ain
ed
with
f
ea
tu
r
es
lik
e
s
ize,
ar
ea
,
s
h
ap
e,
co
lo
u
r
,
b
r
ig
h
tn
ess
an
d
co
n
te
x
tu
al
in
f
o
r
m
atio
n
t
o
g
r
ad
e
ca
n
d
id
ates
as
ex
u
d
ates,
d
r
u
s
en
o
r
b
ac
k
g
r
o
u
n
d
[
5
2
]
.
I
n
[
5
3
]
,
SVM
was
ta
ilo
r
ed
f
o
r
lesi
o
n
d
e
tectio
n
with
f
ea
tu
r
es
ex
tr
ac
ted
u
s
in
g
lo
ca
l
b
in
ar
y
p
atter
n
s
(
L
B
P).
L
B
P
an
d
g
r
an
u
lo
m
etr
ic
p
r
o
f
iles
wer
e
u
s
ed
[
5
4
]
to
tak
e
o
u
t
f
ea
tu
r
es
to
f
ee
d
to
r
an
d
o
m
f
o
r
est
(
R
F),
SVM
an
d
g
au
s
s
ian
p
r
o
c
ess
es f
o
r
class
if
icatio
n
(
GP
C
)
.
Featu
r
e
s
elec
tio
n
u
s
in
g
g
r
e
y
wo
lf
o
p
tim
izatio
n
was
p
e
r
f
o
r
m
ed
an
d
f
ed
to
k
NN
class
if
ier
to
d
is
cr
im
in
ate
th
e
ex
u
d
ates
b
et
wee
n
p
r
o
life
r
ativ
e
an
d
n
o
n
-
p
r
o
life
r
ativ
e
[
5
5
]
.
A
b
ag
o
f
w
o
r
d
s
ap
p
r
o
ac
h
was
p
r
o
p
o
s
ed
in
wh
ich
f
e
atu
r
es
f
r
o
m
im
ag
e
p
atch
es
wer
e
s
to
r
ed
to
cr
ea
te
th
e
d
ictio
n
ar
y
[
5
6
]
.
T
h
is
f
ea
tu
r
e
s
et
was
u
tili
ze
d
to
g
r
a
d
e
p
atch
es
b
et
wee
n
n
o
r
m
al,
ex
u
d
ate
o
r
d
r
u
s
en
.
A
f
u
zz
y
l
o
g
ic
-
b
ased
ca
t
eg
o
r
izatio
n
o
f
h
a
r
d
ex
u
d
ates
was
p
u
t
f
o
r
war
d
in
wh
ich
v
alu
es
o
f
h
ar
d
e
x
u
d
ates
in
R
G
B
co
lo
r
s
p
ac
e
wer
e
u
s
ed
to
f
o
r
m
th
e
f
u
zz
y
s
et
[
5
7
]
,
[
5
8
]
.
I
n
an
o
th
er
atte
m
p
t
to
class
if
y
b
etwe
en
ex
u
d
ates,
g
au
s
s
ian
m
ix
tu
r
e
m
o
d
el
(
GM
M)
was
u
s
ed
.
T
o
d
etec
t
all
p
o
s
s
ib
le
ca
n
d
id
ate
lesi
o
n
s
,
Gab
o
r
f
ilter
b
an
k
[
8
]
was
ap
p
lied
.
R
etin
a
o
f
y
o
u
n
g
er
p
atien
ts
s
h
o
ws
h
ig
h
er
r
ef
lecta
n
ce
b
ec
au
s
e
o
f
Ner
v
e
Fib
er
L
ay
er
.
T
h
is
is
s
u
e
was
tack
led
b
y
[
5
9
]
wh
er
ein
class
if
ier
(
Fis
h
er
’
s
lin
ea
r
d
is
cr
im
in
an
t
class
if
ier
)
was
tr
ain
ed
u
s
in
g
te
x
tu
r
e
d
escr
ip
to
r
s
.
I
n
2
0
2
0
,
W
an
g
,
et
a
l
.
[
6
0
]
,
a
m
eth
o
d
is
r
ep
r
esen
ted
th
at
co
m
b
i
n
es
d
e
ep
f
ea
tu
r
es
with
h
an
d
cr
af
ted
o
n
es
an
d
class
if
y
u
s
in
g
R
F.
T
ab
le
5
e
x
h
ib
its
th
e
p
er
f
o
r
m
an
ce
,
d
atab
ase
an
d
o
th
er
d
etails o
f
th
e
ab
o
v
e
-
m
en
tio
n
ed
m
eth
o
d
s
.
T
ab
le
5
.
Ma
ch
i
n
e
lear
n
in
g
m
e
th
o
d
s
f
o
r
ex
u
d
ate
d
etec
tio
n
A
u
t
h
o
r
(
s)
M
e
t
h
o
d
o
l
o
g
y
D
a
t
a
b
a
s
e
C
l
a
s
si
f
i
e
r
F
e
a
t
u
r
e
s
P
e
r
f
o
r
ma
n
c
e
(
%)
F
u
t
u
r
e
s
c
o
p
e
A
l
e
e
n
a
a
n
d
P
r
a
j
i
t
h
[
4
8
]
M
o
r
p
h
o
l
o
g
i
c
a
l
o
p
e
n
i
n
g
a
n
d
c
l
o
si
n
g
I
D
R
i
D
S
V
M
5
S
e
n
s
i
t
i
v
i
t
y
9
9
.
3
7
A
c
c
u
r
a
c
y
9
9
.
2
7
I
mp
r
o
v
e
m
e
n
t
i
n
so
f
t
e
x
u
d
a
t
e
d
e
t
e
c
t
i
o
n
Lo
n
g
e
t
a
l
.
[
4
9
]
D
y
n
a
mi
c
t
h
r
e
s
h
o
l
d
a
n
d
F
C
M
c
l
u
st
e
r
i
n
g
f
o
r
H
E
d
e
t
e
c
t
i
o
n
e
-
o
p
h
t
h
a
EX
(
1
)
D
i
a
r
e
t
d
b
1
(
2
)
S
V
M
8
S
e
n
s
i
t
i
v
i
t
y
7
6
.
5
(
1
)
P
P
V
8
2
.
7
(
1
)
F
S
c
o
r
e
7
6
.
7
(
1
)
S
e
n
s
i
t
i
v
i
t
y
9
7
.
5
(
2
)
S
p
e
c
i
f
i
c
i
t
y
9
7
.
8
(
2
)
A
c
c
u
r
a
c
y
9
7
.
7
(
2
)
I
mp
r
o
v
e
m
e
n
t
i
n
O
D
a
n
d
b
l
o
o
d
v
e
sse
l
d
e
t
e
c
t
i
o
n
B
h
a
r
g
a
v
i
a
n
d
S
e
n
a
p
a
t
i
[
5
0
]
Te
x
t
u
r
e
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
D
i
a
r
e
t
d
b
1
M
e
ss
i
d
o
r
S
V
M
20
S
e
n
s
i
t
i
v
i
t
y
1
0
0
S
p
e
c
i
f
i
c
i
t
y
9
4
.
6
A
c
c
u
r
a
c
y
9
6
.
6
6
H
y
b
r
i
d
c
l
a
ssi
f
i
e
r
,
d
e
t
e
c
t
i
o
n
o
f
l
e
si
o
n
s
l
i
k
e
M
A
,
H
M
a
n
d
c
o
t
t
o
n
w
o
o
l
sp
o
t
s
M
a
h
e
sw
a
r
i
a
n
d
P
u
n
n
o
l
i
l
[
5
1
]
M
o
r
p
h
o
l
o
g
i
c
a
l
o
p
e
r
a
t
o
r
s
D
r
i
v
e
S
t
a
r
e
S
V
M
7
S
e
n
s
i
t
i
v
i
t
y
8
7
.
6
5
S
p
e
c
i
f
i
c
i
t
y
9
1
.
4
5
A
c
c
u
r
a
c
y
9
3
.
5
U
se
o
f
mu
l
t
i
c
l
a
ss
S
V
M
a
n
d
m
o
r
e
f
e
a
t
u
r
e
s
F
l
e
mi
n
g
e
t
a
l
.
[
5
2
]
D
r
u
sen
i
d
e
n
t
i
f
i
c
a
t
i
o
n
P
r
i
v
a
t
e
S
V
M
8
S
e
n
s
i
t
i
v
i
t
y
9
5
S
p
e
c
i
f
i
c
i
t
y
8
4
.
6
El
i
m
i
n
a
t
i
o
n
o
f
o
t
h
e
r
b
r
i
g
h
t
f
e
a
t
u
r
e
s
R
a
t
h
o
r
e
e
t
a
l
.
[
5
3
]
LB
P
I
D
R
i
D
S
V
M
-
S
e
n
s
i
t
i
v
i
t
y
8
9
S
p
e
c
i
f
i
c
i
t
y
9
6
A
c
c
u
r
a
c
y
9
6
.
9
5
Le
si
o
n
d
e
t
e
c
t
i
o
n
u
si
n
g
N
N
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
A
r
ev
iew
o
n
s
u
p
ervis
ed
lea
r
n
in
g
m
eth
o
d
o
lo
g
ies fo
r
d
etec
tio
n
o
f …
(
Ujw
a
la
W.
Wa
s
ek
a
r
)
843
T
ab
le
5
.
Ma
ch
i
n
e
lear
n
in
g
m
e
th
o
d
s
f
o
r
ex
u
d
ate
d
etec
tio
n
(
c
o
n
tin
u
e)
A
u
t
h
o
r
(
s)
M
e
t
h
o
d
o
l
o
g
y
D
a
t
a
b
a
s
e
C
l
a
s
si
f
i
e
r
F
e
a
t
u
r
e
s
P
e
r
f
o
r
ma
n
c
e
(
%)
F
u
t
u
r
e
s
c
o
p
e
C
o
l
o
m
e
r
e
t
a
l
.
[
5
4
]
LB
P
&
g
r
a
n
u
l
o
m
e
t
r
i
c
p
r
o
f
i
l
e
s
e
-
o
p
h
t
h
a
D
i
a
r
e
t
d
b
1
R
F
,
S
V
M
,
G
P
C
-
S
e
n
s
i
t
i
v
i
t
y
0
.
8
1
8
4
±
0
.
0
3
2
4
S
p
e
c
i
f
i
c
i
t
y
0
.
8
1
8
3
±
0
.
0
3
2
4
P
P
V
0
.
4
3
7
3
±
0
.
1
3
7
4
I
d
e
n
t
i
f
y
d
i
f
f
e
r
e
n
t
st
a
g
e
s
o
f
D
R
K
a
d
a
n
a
n
d
S
u
b
b
i
a
n
[
5
5
]
G
r
e
y
w
o
l
f
o
p
t
i
m
i
z
a
t
i
o
n
a
l
g
o
r
i
t
h
m f
o
r
f
e
a
t
u
r
e
sel
e
c
t
i
o
n
D
i
a
r
e
t
d
b
1
D
r
i
v
e
k
N
N
5
S
e
n
s
i
t
i
v
i
t
y
9
9
.
4
2
S
p
e
c
i
f
i
c
i
t
y
9
9
.
8
7
A
c
c
u
r
a
c
y
9
9
.
3
4
F
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
w
i
t
h
b
e
t
t
e
r
o
p
t
i
m
i
z
a
t
i
o
n
a
n
d
u
s
e
o
f
d
e
e
p
l
e
a
r
n
i
n
g
V
a
n
G
r
i
n
s
v
e
n
[
5
6
]
B
a
g
o
f
w
o
r
d
s
a
p
p
r
o
a
c
h
S
t
a
r
e
M
e
ss
i
d
o
r
P
r
i
v
a
t
e
W
e
i
g
h
t
e
d
n
e
a
r
e
st
n
e
i
g
h
b
o
r
58
A
U
C
0
.
9
0
P
r
e
c
i
s
i
o
n
0
.
7
6
I
mp
l
e
me
n
t
a
t
i
o
n
o
n
l
a
r
g
e
r
d
a
t
a
se
t
R
a
n
a
m
u
k
a
a
n
d
M
e
e
g
a
ma
[
5
7
]
M
a
t
h
e
ma
t
i
c
a
l
mo
r
p
h
o
l
o
g
y
f
o
r
p
r
e
p
r
o
c
e
ssi
n
g
D
i
a
r
e
t
d
b
0
D
i
a
r
e
t
d
b
1
F
u
z
z
y
l
o
g
i
c
-
S
e
n
s
i
t
i
v
i
t
y
8
1
.
7
5
S
p
e
c
i
f
i
c
i
t
y
1
0
0
A
c
c
u
r
a
c
y
9
9
.
8
4
U
se
o
f
o
t
h
e
r
d
a
t
a
b
a
se
s
A
k
r
a
m
e
t
a
l
.
[
8
]
F
i
l
t
e
r
b
a
n
k
s
t
o
e
x
t
r
a
c
t
c
a
n
d
i
d
a
t
e
r
e
g
i
o
n
s
S
t
a
r
e
D
i
a
r
e
t
d
b
0
D
i
a
r
e
t
d
b
1
G
M
M
9
S
e
n
s
i
t
i
v
i
t
y
9
6
.
3
6
S
p
e
c
i
f
i
c
i
t
y
9
8
.
2
5
A
c
c
u
r
a
c
y
9
7
.
5
9
P
P
V
9
7
.
4
5
-
P
r
a
d
e
e
p
a
n
d
Jo
s
e
p
h
[
5
8
]
B
i
n
a
r
y
o
p
e
r
a
t
i
o
n
D
i
a
r
e
t
d
b
0
D
i
a
r
e
t
d
b
1
F
u
z
z
y
l
o
g
i
c
-
S
e
n
s
i
t
i
v
i
t
y
9
8
.
1
0
S
p
e
c
i
f
i
c
i
t
y
9
6
.
9
6
A
c
c
u
r
a
c
y
9
8
.
3
3
I
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
o
t
h
e
r
p
a
t
h
o
l
o
g
i
e
s
G
i
a
n
c
a
r
d
o
e
t
a
l
.
[
5
9
]
Te
x
t
u
r
e
d
e
s
c
r
i
p
t
o
r
s t
o
t
r
a
i
n
c
l
a
ssi
f
i
e
r
P
r
i
v
a
t
e
F
i
sh
e
r
’
s
LD
A
38
S
e
n
s
i
t
i
v
i
t
y
1
0
0
S
p
e
c
i
f
i
c
i
t
y
1
0
0
U
se
o
f
l
a
r
g
e
r
d
a
t
a
s
e
t
w
i
t
h
a
l
l
l
e
s
i
o
n
s
p
r
e
se
n
t
W
a
n
g
e
t
a
l
.
[
6
0
]
D
e
e
p
f
e
a
t
u
r
e
s
(
D
F
)
w
i
t
h
H
C
F
e
-
o
p
h
t
h
a
(
1
)
H
EI
-
M
ED
(
2
)
RF
D
F
& 1
2
8
H
C
F
F
S
c
o
r
e
8
9
.
2
9
(
1
)
A
U
C
9
6
.
4
4
(
1
)
F
S
c
o
r
e
9
3
.
2
6
(
2
)
A
U
C
9
3
.
2
3
(
2
)
C
a
sc
a
d
e
d
n
e
t
w
o
r
k
w
i
t
h
v
o
t
i
n
g
st
r
a
t
e
g
y
,
mo
r
e
l
e
s
i
o
n
s
l
i
k
e
H
M
&
n
e
o
-
v
a
s
c
u
l
a
r
i
z
a
t
i
o
n
5.
DIS
CU
SS
I
O
N
Diab
etic
r
etin
o
p
ath
y
ca
n
claim
a
p
e
r
m
an
en
t
v
is
io
n
l
o
s
s
in
ca
s
e
o
f
s
ev
er
ity
.
T
o
a
v
o
id
th
is
,
ea
r
ly
d
etec
tio
n
o
f
ex
u
d
ates
b
ec
o
m
e
o
f
p
r
im
e
co
n
ce
r
n
.
Ma
n
y
au
t
h
o
r
s
h
av
e
p
u
t
f
o
r
war
d
th
e
ap
p
r
ec
iab
le
au
to
m
ated
s
y
s
tem
s
th
at
co
u
ld
less
en
th
e
b
u
r
d
e
n
o
f
ey
e
c
ar
e
p
r
ac
titi
o
n
er
s
an
d
ex
te
n
d
th
e
g
r
ea
t
h
elp
to
th
e
r
esear
ch
co
m
m
u
n
ity
.
Fo
r
s
u
cc
ess
f
u
l
d
etec
tio
n
o
f
e
x
u
d
ates,
ev
er
y
f
ac
et
in
th
e
p
r
o
ce
s
s
s
h
o
u
ld
b
e
tak
en
in
t
o
co
n
s
id
er
atio
n
.
Ma
n
y
r
esear
ch
es
in
th
e
d
o
m
ain
o
f
m
ed
ical
d
iag
n
o
s
is
o
f
th
e
b
r
ig
h
t
lesi
o
n
s
in
DR
im
ag
es
h
av
e
b
ee
n
illu
s
tr
at
ed
.
Me
th
o
d
o
lo
g
ies
wer
e
ca
teg
o
r
ized
p
er
tain
in
g
to
d
ee
p
lear
n
in
g
,
m
ac
h
in
e
lear
n
in
g
an
d
th
o
s
e
m
ain
ly
f
o
cu
s
in
g
o
n
im
ag
e
p
r
o
ce
s
s
in
g
tech
n
iq
u
es.
Me
th
o
d
s
u
s
ed
f
o
r
p
r
ep
r
o
ce
s
s
in
g
p
lay
a
v
ital
r
o
le
in
r
ef
in
in
g
th
e
im
ag
e
in
tu
r
n
ex
tr
ac
tin
g
m
o
r
e
b
en
e
f
icial
in
f
o
r
m
atio
n
f
o
r
p
atter
n
r
ec
o
g
n
itio
n
.
Sin
ce
,
OD
is
a
s
b
r
ig
h
t
as
ex
u
d
ate,
its
elim
in
atio
n
en
s
u
r
es
s
m
o
o
th
id
en
tific
atio
n
o
f
ex
u
d
ate.
Var
io
u
s
ap
p
r
o
ac
h
es
ar
e
av
ailab
le
f
o
r
OD
d
etec
tio
n
s
u
ch
as
lo
ca
l
p
h
ase
s
y
m
m
etr
y
[
3
7
]
an
d
Ho
u
g
h
tr
an
s
f
o
r
m
[
3
6
]
,
[
4
1
]
.
Fu
zz
y
c
m
ea
n
s
f
o
r
s
eg
m
en
tatio
n
[
2
2
]
,
[
38
]
,
[
39
]
,
[
4
9
]
is
wid
ely
u
s
ed
.
Mo
r
p
h
o
lo
g
ical
o
p
er
ato
r
s
(
in
ass
o
ciatio
n
with
s
tr
u
ctu
r
in
g
elem
en
ts
)
m
ak
e
its
p
lace
in
al
m
o
s
t
ev
er
y
m
eth
o
d
o
l
o
g
y
[
2
0
]
,
[
48
]
,
[
51
]
,
[
5
7
]
b
ec
au
s
e
o
f
its
ab
ilit
y
to
s
eg
m
en
t
th
e
o
b
jects in
th
e
im
a
g
e
d
ep
e
n
d
in
g
o
n
th
eir
s
h
a
p
es.
Dee
p
lear
n
in
g
is
a
b
r
an
c
h
o
f
m
ac
h
in
e
lear
n
in
g
a
n
d
r
e
q
u
ir
es
lar
g
er
d
ata
to
tr
ai
n
th
e
s
y
s
tem
[
3
3
]
.
B
u
ild
in
g
a
n
eu
r
al
n
etwo
r
k
m
ay
b
e
tim
e
c
o
n
s
u
m
in
g
an
d
ef
f
o
r
t
d
em
a
n
d
in
g
.
I
n
s
tead
,
p
r
etr
ain
ed
s
y
s
tem
s
ca
n
b
e
u
tili
ze
d
[
3
5
]
.
Par
allel
p
r
o
ce
s
s
in
g
s
y
s
tem
s
m
ay
also
b
o
o
s
t
th
e
p
er
f
o
r
m
an
ce
b
y
s
av
in
g
tim
e
[
4
5
]
.
DL
co
m
b
in
ed
with
im
ag
e
p
r
o
ce
s
s
in
g
m
ay
p
r
o
v
e
ef
f
icien
t
s
et
-
u
p
[
3
6
]
.
R
elev
an
t
f
ea
tu
r
e
s
o
f
an
en
tity
ca
n
ac
cu
r
ately
id
en
tif
y
i
t.
E
x
u
d
ate
s
ca
n
b
e
g
r
ea
tly
r
ec
o
g
n
ized
b
y
its
s
h
ap
e,
s
ize,
co
lo
r
,
tex
t
u
r
e,
in
ten
s
ity
an
d
ed
g
e
s
tr
en
g
th
[
8
]
,
[
38
]
,
[
55
]
,
[
4
4
]
.
Dete
ctio
n
o
f
all
lesi
o
n
s
[
4
6
]
,
[
5
8
]
an
d
m
u
ltip
le
DR
s
tag
e
s
[
4
7
]
,
[
5
4
]
is
th
e
m
atter
o
f
co
n
ce
r
n
.
M
u
l
t
i
p
l
e
d
at
a
b
as
e
s
a
r
e
p
u
b
l
i
c
al
l
y
a
v
a
i
la
b
l
e
f
o
r
t
h
e
r
es
e
a
r
c
h
e
r
s
t
h
a
t
p
r
o
v
i
d
e
a
b
s
o
l
u
t
e
u
n
i
f
o
r
m
p
l
a
t
f
o
r
m
t
o
e
v
a
l
u
a
te
t
h
e
m
e
t
h
o
d
s
p
r
o
p
o
s
e
d
b
y
t
h
e
m
.
Se
v
e
r
a
l
a
u
t
h
o
r
s
o
p
t
f
o
r
p
r
i
v
a
t
e
d
a
t
a
i
.
e
.
d
a
ta
t
a
k
e
n
f
r
o
m
s
o
m
e
h
e
a
l
t
h
c
a
r
e
i
n
s
ti
t
u
ti
o
n
s
a
n
d
e
v
a
l
u
a
t
i
o
n
d
o
n
e
o
n
t
h
e
b
a
s
is
o
f
t
h
e
h
a
n
d
-
d
r
a
w
n
g
r
o
u
n
d
t
r
u
t
h
g
i
v
e
n
b
y
t
h
e
e
x
p
e
r
t
s
[
2
1
]
,
[
39
]
,
[
5
9
]
.
I
n
t
h
i
s
s
t
u
d
y
o
n
e
f
o
u
r
t
h
o
f
t
h
e
m
et
h
o
d
o
l
o
g
i
e
s
u
s
e
d
p
r
i
v
a
t
e
d
a
t
ase
t
s
.
F
i
g
u
r
e
2
d
e
p
ic
ts
t
h
e
d
i
s
t
r
i
b
u
t
i
o
n
o
f
d
a
t
a
b
a
s
es
ac
c
o
r
d
i
n
g
t
o
t
h
e
i
r
d
o
m
a
i
n
a
n
d
m
u
l
t
i
p
l
i
ci
t
y
.
I
t
i
s
c
l
e
a
r
t
h
at
m
a
n
y
a
u
t
h
o
r
s
p
r
e
f
e
r
t
h
e
c
o
m
b
i
n
a
t
i
o
n
o
f
m
u
l
t
i
p
l
e
p
u
b
l
i
c
d
a
t
a
b
a
s
es
t
o
e
n
h
a
n
c
e
t
h
e
p
e
r
f
o
r
m
a
n
c
e
.
A
l
s
o
,
p
u
b
l
i
c
d
a
t
a
b
a
s
es
g
i
v
e
m
o
r
e
c
o
m
p
e
t
i
t
i
v
e
e
n
v
i
r
o
n
m
e
n
t
f
o
r
e
v
a
l
u
a
t
i
o
n
.
P
e
r
f
o
r
m
a
n
c
e
m
e
a
s
u
r
e
s
o
f
f
e
r
d
i
f
f
e
r
e
n
t
a
n
g
l
e
s
o
f
a
s
s
es
s
m
e
n
t
o
f
t
h
e
s
y
s
te
m
a
n
d
p
r
o
v
i
d
e
i
n
c
e
n
t
i
v
e
t
o
p
e
r
f
o
r
m
b
e
t
t
e
r
.
F
i
g
u
r
e
3
s
h
o
w
s
t
h
a
t
s
e
n
s
i
t
i
v
i
t
y
,
s
p
e
ci
f
i
c
ity
a
n
d
a
c
c
u
r
a
c
y
a
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
23
,
No
.
2
,
Au
g
u
s
t 2
0
2
1
:
837
-
8
4
6
844
e
x
t
e
n
s
i
v
el
y
u
s
e
d
m
e
a
s
u
r
es
f
o
r
s
y
s
t
e
m
e
v
al
u
a
t
i
o
n
.
I
n
m
e
d
i
c
al
d
o
m
a
i
n
,
s
e
n
s
it
i
v
it
y
is
t
h
e
m
o
s
t
g
o
f
o
r
m
e
as
u
r
e
.
T
h
o
u
g
h
,
in
v
esti
g
atio
n
i
n
th
is
r
ea
lm
h
as
r
ea
ch
ed
a
f
a
r
way
.
Sti
ll,
th
er
e
is
a
n
ee
d
to
f
o
cu
s
o
n
s
o
m
e
ar
ea
s
lik
e
d
is
cr
im
in
atin
g
b
etwe
en
h
ar
d
an
d
s
o
f
t
ex
u
d
ates
[
3
5
]
,
b
etter
f
ea
tu
r
e
s
elec
tio
n
[
5
5
]
an
d
im
p
r
o
v
em
en
t
in
OD
an
d
b
lo
o
d
v
ess
el
d
etec
tio
n
[
4
9
]
.
Fig
u
r
e
2
.
Per
ce
n
ta
g
e
d
is
tr
ib
u
ti
o
n
o
f
d
atab
ases
Fig
u
r
e
3
.
Usag
e
o
f
p
er
f
o
r
m
a
n
ce
m
ea
s
u
r
es in
u
s
ed
in
th
e
s
tu
d
y
th
e
s
tu
d
y
6.
CO
NCLU
SI
O
N
D
i
a
b
e
t
i
c
R
e
t
i
n
o
p
a
t
h
y
h
a
s
b
e
c
o
m
e
t
h
e
m
a
j
o
r
c
o
n
c
e
r
n
f
o
r
v
i
s
u
a
l
i
m
p
a
i
r
m
e
n
t
w
o
r
l
d
w
i
d
e
.
R
e
c
e
n
t
l
y
c
o
m
p
u
t
e
r
a
i
d
e
d
d
i
a
g
n
o
s
i
s
(
C
A
D
)
s
y
s
t
e
m
s
a
l
o
n
g
w
i
t
h
i
m
a
g
e
p
r
o
c
e
s
s
i
n
g
t
e
c
h
n
i
q
u
e
s
h
a
v
e
e
m
e
r
g
e
d
o
u
t
t
o
b
e
b
e
n
e
f
i
c
i
a
l
i
n
t
e
r
m
s
o
f
c
o
s
t
,
t
i
m
e
a
n
d
a
c
c
u
r
a
c
y
.
M
a
n
y
s
t
a
t
e
-
of
-
t
h
e
-
a
r
t
a
p
p
r
o
a
c
h
e
s
h
a
v
e
b
e
e
n
p
r
o
p
o
s
e
d
i
n
t
h
i
s
r
e
g
a
r
d
.
A
c
c
o
r
d
i
n
g
t
o
B
r
i
t
i
s
h
D
i
a
b
e
t
i
c
A
s
s
o
c
i
a
t
i
o
n
,
m
i
n
i
m
u
m
8
0
%
s
e
n
s
i
t
i
v
i
t
y
a
n
d
9
5
%
s
p
e
c
i
f
i
c
i
t
y
i
s
d
e
s
i
r
a
b
l
e
.
V
a
r
i
o
u
s
m
e
t
h
o
d
o
l
o
g
i
e
s
p
e
r
t
a
i
n
i
n
g
t
o
c
l
a
s
s
i
f
i
c
a
t
i
o
n
m
o
d
u
l
e
s
s
u
c
h
a
s
d
e
e
p
l
e
a
r
n
i
n
g
a
n
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
h
a
v
e
b
e
e
n
b
r
i
e
f
l
y
d
i
s
c
u
s
s
e
d
.
I
t
i
s
o
b
s
e
r
v
e
d
t
h
a
t
S
V
M
i
s
t
h
e
p
o
p
u
l
a
r
a
p
p
r
o
a
c
h
i
n
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
.
M
e
t
h
o
d
s
s
u
g
g
e
s
t
e
d
b
y
t
h
e
r
e
s
e
a
r
c
h
e
r
s
a
r
e
e
f
f
i
c
i
e
n
t
b
u
t
h
a
v
e
p
o
t
e
n
t
i
a
l
a
r
e
a
s
f
o
r
e
x
p
e
r
i
m
e
n
t
a
t
i
o
n
.
D
e
e
p
l
e
a
r
n
i
n
g
w
i
t
h
i
m
a
g
e
p
r
o
c
e
s
s
i
n
g
c
o
u
l
d
b
e
a
c
o
n
s
i
d
e
r
a
b
l
e
a
r
e
a
t
o
e
x
p
l
o
r
e
.
S
i
n
c
e
d
e
e
p
l
e
a
r
n
i
n
g
r
e
q
u
i
r
e
s
a
l
a
r
g
e
a
m
o
u
n
t
o
f
d
a
t
a
a
n
d
s
i
g
n
i
f
i
c
a
n
t
t
i
m
e
t
o
b
u
i
l
d
,
s
t
e
p
c
a
n
b
e
p
u
t
f
o
r
w
a
r
d
t
o
a
d
d
r
e
s
s
t
h
e
s
e
i
s
s
u
e
s
.
S
e
v
e
r
a
l
p
u
b
l
i
c
a
l
l
y
a
v
a
i
l
a
b
l
e
d
a
t
a
b
a
s
e
s
w
e
r
e
b
r
i
e
f
l
y
d
e
s
c
r
i
b
e
d
.
S
e
l
e
c
t
i
o
n
o
f
m
u
l
t
i
p
l
e
d
a
t
a
s
e
t
s
i
s
a
c
o
m
m
o
n
c
h
o
i
c
e
f
o
r
m
a
n
y
r
e
s
e
a
r
c
h
e
r
s
.
I
t
c
o
u
l
d
b
e
i
n
f
e
r
r
e
d
t
h
a
t
s
e
n
s
i
t
i
v
i
t
y
i
s
t
h
e
w
i
d
e
l
y
u
s
e
d
m
e
a
s
u
r
e
.
C
o
m
p
a
r
i
n
g
t
h
e
p
e
r
f
o
r
m
a
n
c
e
s
o
f
t
h
e
a
l
g
o
r
i
t
h
m
s
i
n
s
t
i
g
a
t
e
t
o
p
e
r
f
o
r
m
b
e
t
t
e
r
t
o
g
i
v
e
m
a
x
i
m
u
m
t
h
r
o
u
g
h
p
u
t
.
H
o
w
e
v
e
r
,
t
h
e
m
o
s
t
e
f
f
i
c
i
e
n
t
a
n
d
c
a
p
a
b
l
e
a
u
t
o
m
a
t
e
d
s
y
s
t
e
m
i
s
s
t
i
l
l
o
n
t
h
e
w
a
y
.
RE
F
E
R
E
NC
E
S
[1
]
X.
Li
n
,
e
t
a
l
.
,
“
G
lo
b
a
l,
re
g
i
o
n
a
l,
a
n
d
n
a
ti
o
n
a
l
b
u
r
d
e
n
a
n
d
tre
n
d
o
f
d
iab
e
tes
in
1
9
5
c
o
u
n
tri
e
s
a
n
d
territo
ries
:
a
n
a
n
a
ly
sis fro
m
1
9
9
0
t
o
2
0
2
5
,
”
S
c
ien
ti
fi
c
Rep
o
rts
,
v
o
l
.
1
0
,
n
o
.
1
,
S
e
p
t
.
2
0
2
0
,
d
o
i:
1
0
.
1
0
3
8
/s4
1
5
9
8
-
0
2
0
-
7
1
9
0
8
-
9.
[2
]
R.
S
.
Biy
a
n
i
a
n
d
B.
M
.
P
a
tre,
“
Alg
o
rit
h
m
s fo
r
re
d
les
i
o
n
d
e
tec
ti
o
n
i
n
Dia
b
e
ti
c
Re
ti
n
o
p
a
th
y
:
A rev
iew
,
”
Bi
o
me
d
icin
e
&
Ph
a
rm
a
c
o
th
e
ra
p
y
,
v
o
l.
1
0
7
,
p
p
.
6
8
1
–
6
8
8
,
N
o
v
.
2
0
1
8
,
d
o
i
:
1
0
.
1
0
1
6
/j
.
b
i
o
p
h
a
.
2
0
1
8
.
0
7
.
1
7
5
.
[3
]
S
.
Jo
s
h
i
a
n
d
P
.
T
.
Ka
ru
le,
“
A
re
v
iew
o
n
e
x
u
d
a
tes
d
e
tec
ti
o
n
m
e
th
o
d
s
fo
r
d
ia
b
e
ti
c
re
ti
n
o
p
a
t
h
y
,
”
Bi
o
me
d
icin
e
&
Ph
a
rm
a
c
o
th
e
ra
p
y
,
v
o
l.
9
7
,
p
p
.
1
4
5
4
–
1
4
6
0
,
Ja
n
.
2
0
1
8
,
d
o
i:
1
0
.
1
0
1
6
/j
.
b
i
o
p
h
a
.
2
0
1
7
.
1
1
.
0
0
9
.
[4
]
H.
As
h
a
,
e
t
a
l
.
,
“
De
tec
ti
o
n
a
n
d
g
ra
d
in
g
o
f
d
iab
e
ti
c
re
ti
n
o
p
a
t
h
y
in
re
ti
n
a
l
ima
g
e
s
u
sin
g
d
e
e
p
in
telli
g
e
n
t
sy
ste
m
s:
a
c
o
m
p
re
h
e
n
siv
e
re
v
iew
,
”
Co
mp
u
ter
s,
M
a
ter
ia
ls
&
Co
n
ti
n
u
a
,
v
o
l
.
6
6
,
n
o
.
3
,
p
p
.
2
7
7
1
–
2
7
8
6
,
2
0
2
1
,
d
o
i:
1
0
.
3
2
6
0
4
/cm
c
.
2
0
2
1
.
0
1
2
9
0
7
.
[5
]
S
.
E.
M
a
n
so
u
r,
D.
J.
Bro
wn
i
n
g
,
K.
Wo
n
g
,
H.
W.
F
l
y
n
n
Jr,
a
n
d
A.
R
.
Bh
a
v
sa
r,
“
Th
e
Ev
o
l
v
i
n
g
Trea
tme
n
t
o
f
Dia
b
e
ti
c
Re
ti
n
o
p
a
t
h
y
,
”
Cli
n
ica
l
Op
h
t
h
a
lm
o
lo
g
y
,
v
o
l.
1
4
,
p
p
.
6
5
3
–
6
7
8
,
2
0
2
0
,
d
o
i:
1
0
.
2
1
4
7
/O
P
TH.
S
2
3
6
6
3
7
.
[6
]
A.
Iss
a
c
,
M
.
K.
Du
tt
a
,
a
n
d
C.
M
.
Trav
ies
o
,
“
Au
t
o
m
a
ti
c
c
o
m
p
u
ter
v
isio
n
-
b
a
se
d
d
e
tec
ti
o
n
a
n
d
q
u
a
n
ti
ta
ti
v
e
a
n
a
ly
sis
o
f
in
d
ica
ti
v
e
p
a
ra
m
e
ters
fo
r
g
ra
d
in
g
o
f
d
iab
e
ti
c
re
ti
n
o
p
a
t
h
y
,
”
N
e
u
ra
l
Co
mp
u
ti
n
g
a
n
d
Ap
p
li
c
a
t
io
n
s
,
v
o
l
.
3
2
,
p
p
.
1
5
6
8
7
–
1
5
6
9
7
,
2
0
2
0
,
d
o
i:
1
0
.
1
0
0
7
/s0
0
5
2
1
-
0
1
8
-
3
4
4
3
-
z
.
[7
]
M
.
U.
Ak
ra
m
,
I.
Ja
m
a
l
,
an
d
A.
Tariq
,
“
Blo
o
d
Ve
ss
e
l
En
h
a
n
c
e
m
e
n
t
a
n
d
S
e
g
m
e
n
tati
o
n
f
o
r
S
c
re
e
n
in
g
o
f
Dia
b
e
ti
c
Re
ti
n
o
p
a
t
h
y
,
”
T
EL
KOM
NIKA
T
e
lec
o
mm
u
n
ica
ti
o
n
,
Co
mp
u
ti
n
g
,
El
e
c
tro
n
ics
a
n
d
Co
n
tr
o
l
,
v
o
l.
1
0
,
n
o
.
2
,
p
p
.
3
2
7
-
3
3
4
,
Ju
n
2
0
1
2
,
d
o
i:
1
0
.
1
1
5
9
1
/t
e
lk
o
m
n
ik
a
.
v
1
0
i
2
.
6
8
6
.
[8
]
M
.
U.
Ak
ra
m
,
A.
Tariq
,
M
.
A.
An
ju
m
,
a
n
d
M
.
Y.
Ja
v
e
d
,
“
Au
to
m
a
ted
d
e
tec
ti
o
n
o
f
e
x
u
d
a
tes
in
c
o
lo
re
d
re
ti
n
a
l
ima
g
e
s
fo
r
d
iag
n
o
sis
o
f
d
iab
e
t
ic
re
ti
n
o
p
a
th
y
,
”
Ap
p
li
e
d
Op
ti
c
s
,
v
o
l.
5
1
,
n
o
.
2
0
,
p
p
.
4
8
5
8
-
4
8
6
6
,
Ju
l
.
2
0
1
2
,
d
o
i:
1
0
.
1
3
6
4
/AO.5
1
.
0
0
4
8
5
8
.
[9
]
E.
De
c
e
n
c
ière
,
e
t
a
l
.
,
“
F
e
e
d
b
a
c
k
o
n
a
p
u
b
li
c
ly
d
istri
b
u
ted
ima
g
e
d
a
tab
a
se
:
th
e
m
e
ss
id
o
r
d
a
tab
a
se
,”
Ima
g
e
An
a
lys
is
&
S
ter
e
o
lo
g
y
,
v
o
l.
3
3
,
n
o
.
3
,
p
p
.
2
3
1
-
2
3
4
,
2
0
1
4
,
d
o
i:
1
0
.
5
5
6
6
/
ias
.
1
1
5
5
.
[1
0
]
T.
Ka
u
p
p
i,
e
t
a
l
.
,
“
DIA
RET
DB0
:
Ev
a
lu
a
ti
o
n
Da
tab
a
se
a
n
d
M
e
t
h
o
d
o
lo
g
y
f
o
r
Dia
b
e
ti
c
Re
ti
n
o
p
a
th
y
Alg
o
rit
h
m
s,”
T
e
c
h
n
ica
l
re
p
o
rt
,
v
o
l.
7
3
,
p
p
.
1
-
1
7
,
2
0
0
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
A
r
ev
iew
o
n
s
u
p
ervis
ed
lea
r
n
in
g
m
eth
o
d
o
lo
g
ies fo
r
d
etec
tio
n
o
f …
(
Ujw
a
la
W.
Wa
s
ek
a
r
)
845
[1
1
]
T.
Ka
u
p
p
i
,
e
t
a
l
.
,
“
Dia
re
td
b
1
d
iab
e
ti
c
re
ti
n
o
p
a
t
h
y
d
a
tab
a
se
a
n
d
e
v
a
lu
a
ti
o
n
p
r
o
to
c
o
l,
”
Pr
o
c
e
e
d
in
g
s
M
e
d
ica
l
Ima
g
e
Un
d
e
rs
ta
n
d
in
g
a
n
d
A
n
a
lys
is (
M
I
UA)
,
v
o
l.
2
0
0
7
,
2
0
0
7
,
d
o
i:
1
0
.
5
2
4
4
/C.
2
1
.
1
5
.
[1
2
]
X.
Zh
a
n
g
,
e
t
a
l.
,
“
Ex
u
d
a
te
d
e
tec
ti
o
n
i
n
c
o
l
o
r
re
ti
n
a
l
ima
g
e
s
fo
r
m
a
ss
sc
re
e
n
in
g
o
f
d
iab
e
ti
c
re
ti
n
o
p
a
th
y
,
”
M
e
d
ica
l
Ima
g
e
An
a
lys
is
,
v
o
l
.
1
8
,
n
o
.
7
,
p
p
.
1
0
2
6
–
1
0
4
3
,
Oc
t.
2
0
1
4
,
d
o
i:
1
0
.
1
0
1
6
/j
.
m
e
d
ia.
2
0
1
4
.
0
5
.
0
0
4
.
[1
3
]
P.
P
o
rwa
l,
e
t
a
l
.
,
“
In
d
ian
Dia
b
e
ti
c
Re
ti
n
o
p
a
th
y
Im
a
g
e
Da
tas
e
t
(IDRiD):
A
Da
tab
a
se
fo
r
Dia
b
e
ti
c
Re
ti
n
o
p
a
t
h
y
S
c
re
e
n
in
g
Re
se
a
rc
h
,
”
Da
t
a
,
v
o
l
3
,
n
o
.
3
,
2
0
1
8
,
d
o
i
:
1
0
.
3
3
9
0
/
d
a
ta3
0
3
0
0
2
5
.
[1
4
]
J.
S
taa
l,
M
.
D.
Ab
ra
m
o
ff,
M
.
Nie
m
e
ij
e
r,
M
.
A.
Vie
rg
e
v
e
r
,
a
n
d
B
.
v
a
n
G
in
n
e
k
e
n
,
“
Rid
g
e
-
Ba
se
d
Ve
ss
e
l
S
e
g
m
e
n
tatio
n
in
Co
l
o
r
Im
a
g
e
s
o
f
th
e
Re
ti
n
a
,
”
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
M
e
d
ic
a
l
I
ma
g
in
g
,
v
o
l.
2
3
,
n
o
.
4
,
p
p
.
5
0
1
–
5
0
9
,
A
p
r.
2
0
0
4
,
d
o
i:
1
0
.
1
1
0
9
/T
M
I
.
2
0
0
4
.
8
2
5
6
2
7
.
[1
5
]
A.
Ho
o
v
e
r,
V.
K
o
u
z
n
e
tso
v
a
,
a
n
d
M
.
G
o
ld
b
a
u
m
,
“
L
o
c
a
ti
n
g
b
lo
o
d
v
e
ss
e
ls
in
re
ti
n
a
l
ima
g
e
s
b
y
p
iec
e
wise
th
re
sh
o
l
d
p
ro
b
in
g
o
f
a
m
a
t
c
h
e
d
fil
ter
re
sp
o
n
se
,
”
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
M
e
d
ica
l
Ima
g
i
n
g
,
v
o
l
.
1
9
,
n
o
.
3
,
p
p
.
2
0
3
–
2
1
0
,
2
0
0
0
,
d
o
i:
1
0
.
1
1
0
9
/4
2
8
4
5
1
7
8
.
[1
6
]
G
.
T.
Zag
o
,
R
.
V.
A
n
d
re
ã
o
,
B
.
Do
rizz
i
,
a
n
d
E.
O.
Tea
ti
n
i
S
a
ll
e
s,
“
Dia
b
e
ti
c
re
ti
n
o
p
a
th
y
d
e
tec
ti
o
n
u
sin
g
re
d
les
i
o
n
lo
c
a
li
z
a
ti
o
n
a
n
d
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
two
r
k
s,”
Co
mp
u
ter
s
in
Bi
o
l
o
g
y
a
n
d
M
e
d
ici
n
e
,
v
o
l
.
1
1
6
,
Ja
n
.
2
0
2
0
,
d
o
i:
1
0
.
1
0
1
6
/j
.
c
o
m
p
b
i
o
m
e
d
.
2
0
1
9
.
1
0
3
5
3
7
.
[1
7
]
K.
Oh
,
H.
M
.
Ka
n
g
,
D.
Lee
m
,
H.
Lee
,
K.
Y.
S
e
o
,
a
n
d
S
.
Yo
o
n
,
“
Early
d
e
tec
ti
o
n
o
f
d
iab
e
ti
c
re
ti
n
o
p
a
th
y
b
a
se
d
o
n
d
e
e
p
lea
rn
i
n
g
a
n
d
u
lt
ra
-
wi
d
e
-
field
f
u
n
d
u
s
ima
g
e
s,”
S
c
ien
ti
fi
c
Rep
o
rts
,
v
o
l.
11
,
no
.
1
,
1
8
9
7
,
Ja
n
.
2
0
2
1
,
d
o
i:
1
0
.
1
0
3
8
/s4
1
5
9
8
-
0
2
1
-
8
1
5
3
9
-
3
.
[1
8
]
B.
Ve
n
k
a
tala
k
sh
m
i,
V.
S
a
ra
v
a
n
a
n
,
a
n
d
G
.
J.
Niv
e
d
it
h
a
,
“
G
ra
p
h
ica
l
u
se
r
in
terfa
c
e
fo
r
e
n
h
a
n
c
e
d
re
ti
n
a
l
ima
g
e
a
n
a
ly
sis
fo
r
d
iag
n
o
sin
g
d
iab
e
ti
c
re
ti
n
o
p
a
t
h
y
,
”
2
0
1
1
I
EE
E
3
rd
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
Co
mm
u
n
ica
ti
o
n
S
o
ft
w
a
re
a
n
d
Ne
tw
o
rk
s
,
2
0
1
1
,
p
p
.
6
1
0
-
6
1
3
,
d
o
i:
1
0
.
1
1
0
9
/ICCS
N.
2
0
1
1
.
6
0
1
4
9
6
7
.
[1
9
]
P
.
S
a
ra
n
y
a
a
n
d
K
.
M
.
Um
a
m
a
h
e
sw
a
ri,
“
De
tec
ti
n
g
E
x
u
d
a
tes
i
n
Co
lo
r
F
u
n
d
u
s
Im
a
g
e
s
fo
r
Dia
b
e
ti
c
Re
ti
n
o
p
a
t
h
y
De
tec
ti
o
n
Us
in
g
De
e
p
Lea
rn
in
g
,
”
An
n
a
ls
o
f
t
h
e
Ro
ma
n
ia
n
S
o
c
iety
fo
r
Ce
ll
Bi
o
l
o
g
y
,
v
o
l.
2
5
,
n
o
.
5
,
p
p
.
5
3
6
8
–
5
3
7
5
,
2
0
2
1
.
[2
0
]
T.
Walter,
J.
Kle
i
n
,
P
.
M
a
ss
in
,
a
n
d
A.
Er
g
in
a
y
,
“
A
c
o
n
tr
ib
u
ti
o
n
o
f
ima
g
e
p
ro
c
e
ss
in
g
t
o
t
h
e
d
iag
n
o
si
s
o
f
d
iab
e
ti
c
re
ti
n
o
p
a
th
y
-
d
e
tec
ti
o
n
o
f
e
x
u
d
a
te
s
in
c
o
lo
r
fu
n
d
u
s
ima
g
e
s
o
f
th
e
h
u
m
a
n
re
ti
n
a
,
”
IEE
E
T
r
a
n
sa
c
ti
o
n
s
o
n
M
e
d
ica
l
Ima
g
i
n
g
,
v
o
l.
2
1
,
n
o
.
1
0
,
p
p
.
1
2
3
6
–
1
2
4
3
,
Oc
t.
2
0
0
2
,
d
o
i:
1
0
.
1
1
0
9
/T
M
I.
2
0
0
2
.
8
0
6
2
9
0
.
[2
1
]
A.
S
o
p
h
a
ra
k
,
B.
Uy
y
a
n
o
n
v
a
ra
,
S
.
Ba
rm
a
n
,
a
n
d
T.
H.
Wi
ll
iam
so
n
,
“
Au
t
o
m
a
ti
c
d
e
tec
ti
o
n
o
f
d
ia
b
e
ti
c
re
ti
n
o
p
a
th
y
e
x
u
d
a
tes
fr
o
m
n
o
n
-
d
il
a
ted
re
ti
n
a
l
ima
g
e
s
u
sin
g
m
a
th
e
m
a
ti
c
a
l
m
o
rp
h
o
lo
g
y
m
e
th
o
d
s,”
C
o
mp
u
t
e
rize
d
M
e
d
ica
l
Ima
g
i
n
g
a
n
d
Gr
a
p
h
ics
,
v
o
l
.
3
2
,
n
o
.
8
,
p
p
.
7
2
0
–
7
2
7
,
De
c
.
2
0
0
8
,
d
o
i
:
1
0
.
1
0
1
6
/j
.
c
o
m
p
m
e
d
im
a
g
.
2
0
0
8
.
0
8
.
0
0
9
.
[2
2
]
A.
S
o
p
h
a
ra
k
,
B.
U
y
y
a
n
o
n
v
a
ra
,
a
n
d
S
.
Ba
rm
a
n
,
“
Au
to
m
a
ti
c
Ex
u
d
a
te
De
tec
ti
o
n
fr
o
m
No
n
-
d
il
a
ted
Dia
b
e
ti
c
Re
ti
n
o
p
a
t
h
y
Re
ti
n
a
l
Im
a
g
e
s
Us
in
g
F
u
z
z
y
C
-
m
e
a
n
s
Clu
ste
rin
g
,
”
S
e
n
so
rs
,
v
o
l.
9
,
n
o
.
3
,
p
p
.
2
1
4
8
–
2
1
6
1
,
M
a
r.
2
0
0
9
,
d
o
i:
1
0
.
3
3
9
0
/s9
0
3
0
2
1
4
8
.
[2
3
]
S.
Jo
sh
i
a
n
d
P
.
T.
Ka
ru
le,
“
De
tec
ti
o
n
o
f
Ha
rd
Ex
u
d
a
tes
Ba
se
d
o
n
M
o
rp
h
o
lo
g
ica
l
F
e
a
tu
re
Ex
tra
c
ti
o
n
,
”
Bi
o
me
d
Ph
a
rm
a
c
o
l
J
,
v
o
l
.
1
1
,
n
o
.
1
,
p
p
.
2
1
5
-
2
2
5
,
2
0
1
8
,
d
o
i:
1
0
.
1
3
0
0
5
/b
p
j/
1
3
6
6
.
[2
4
]
N
.
Nu
r
a
n
d
H
.
Tj
a
n
d
ra
sa
,
“
Ex
u
d
a
te
S
e
g
m
e
n
tati
o
n
in
Re
ti
n
a
l
I
m
a
g
e
s
o
f
Dia
b
e
ti
c
Re
ti
n
o
p
a
t
h
y
Us
in
g
S
a
li
e
n
c
y
M
e
th
o
d
Ba
se
d
o
n
Re
g
i
o
n
,
”
J
.
o
f
Ph
y
s.
:
Co
n
f
.
S
e
rie
s
,
v
o
l.
1
1
0
8
,
2
0
1
8
,
d
o
i:
1
0
.
1
0
8
8
/
1
7
4
2
-
6
5
9
6
/1
1
0
8
/1
/0
1
2
1
1
0
.
[2
5
]
V.
S
a
ty
a
n
a
n
d
a
,
K.
V.
Na
ra
y
a
n
a
sw
a
m
y
,
a
n
d
Ka
rib
a
sa
p
p
a
,
“
F
P
G
A
a
n
d
M
ATLAB
Ba
se
d
S
o
l
u
ti
o
n
f
o
r
Re
ti
n
a
l
Ex
u
d
a
te
De
tec
ti
o
n
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Rec
e
n
t
T
e
c
h
n
o
lo
g
y
a
n
d
E
n
g
i
n
e
e
rin
g
,
v
o
l.
8
,
n
o
.
6
,
p
p
.
7
2
7
-
7
3
4
,
2
0
2
0
,
d
o
i:
1
0
.
3
5
9
4
0
/
ij
rte.
[2
6
]
E.
S
a
e
e
d
,
M
.
S
z
y
m
k
o
ws
k
i,
K.
S
a
e
e
d
,
a
n
d
Z
.
M
a
riak
,
“
An
Ap
p
ro
a
c
h
to
A
u
to
m
a
ti
c
Ha
rd
E
x
u
d
a
te
De
tec
ti
o
n
i
n
Re
ti
n
a
Co
l
o
r
Im
a
g
e
s
b
y
a
Tele
m
e
d
icin
e
S
y
ste
m
Ba
se
d
o
n
t
h
e
d
-
Ey
e
S
e
n
so
r
a
n
d
Im
a
g
e
P
ro
c
e
ss
in
g
Alg
o
rit
h
m
s,”
S
e
n
so
rs
,
v
o
l
.
1
9
,
n
o
.
3
,
2
0
1
9
,
d
o
i
:
1
0
.
3
3
9
0
/s1
9
0
3
0
6
9
5
.
[2
7
]
J.
Ka
n
imo
z
h
i,
P
.
Va
su
k
i
,
a
n
d
S
.
M
.
M
.
Ro
o
m
i,
“
F
u
n
d
u
s
ima
g
e
le
sio
n
d
e
tec
ti
o
n
a
lg
o
r
it
h
m
f
o
r
d
iab
e
ti
c
re
ti
n
o
p
a
th
y
sc
re
e
n
in
g
,
”
J
o
u
rn
a
l
o
f
Amb
ien
t
In
telli
g
e
n
c
e
a
n
d
Hu
ma
n
ize
d
Co
m
p
u
ti
n
g
,
v
o
l
.
1
2
,
p
p
.
7
4
0
7
–
7
4
1
6
,
2
0
2
0
,
d
o
i:
h
tt
p
s://
d
o
i
.
o
r
g
/1
0
.
1
0
0
7
/s1
2
6
5
2
-
0
2
0
-
0
2
4
1
7
-
w
.
[2
8
]
C.
S
in
th
a
n
a
y
o
t
h
in
,
e
t
a
l.
,
“
A
u
t
o
m
a
ted
d
e
tec
ti
o
n
o
f
d
ia
b
e
ti
c
re
ti
n
o
p
a
th
y
o
n
d
ig
it
a
l
fu
n
d
u
s
im
a
g
e
s,”
Dia
b
e
ti
c
M
e
d
icin
e
,
v
o
l.
1
9
,
n
o
.
2
,
p
p
.
1
0
5
–
1
1
2
,
F
e
b
.
2
0
0
2
,
d
o
i:
1
0
.
1
0
4
6
/
j.
1
4
6
4
-
5
4
9
1
.
2
0
0
2
.
0
0
6
1
3
.
x
.
[2
9
]
W.
Hs
u
,
P
.
M
.
D.
S
.
P
a
ll
a
wa
la,
M
.
Li
Lee
,
a
n
d
K
.
Au
Eo
n
g
,
“
T
h
e
ro
le
o
f
d
o
m
a
in
k
n
o
wle
d
g
e
i
n
th
e
d
e
tec
ti
o
n
o
f
re
ti
n
a
l
h
a
rd
e
x
u
d
a
tes
,
”
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
2
0
0
1
IEE
E
Co
m
p
u
te
r
S
o
c
iety
Co
n
fer
e
n
c
e
o
n
C
o
mp
u
ter
Vi
sio
n
a
n
d
Pa
tt
e
rn
Rec
o
g
n
it
io
n
.
CV
PR
2
0
0
1
,
2
0
0
1
,
p
p
.
II
-
II
,
d
o
i:
1
0
.
1
1
0
9
/CV
P
R.
2
0
0
1
.
9
9
0
9
6
7
.
[3
0
]
D.
L
in
,
A.
V.
Va
silak
o
s,
Y.
Tan
g
,
a
n
d
Y.
Ya
o
,
“
Ne
u
ra
l
n
e
two
r
k
s
fo
r
c
o
m
p
u
ter
-
a
id
e
d
d
iag
n
o
sis
i
n
m
e
d
icin
e
:
A
re
v
iew
,
”
Ne
u
ro
c
o
mp
u
ti
n
g
,
v
o
l.
2
1
6
,
p
p
.
7
0
0
-
7
0
8
,
De
c
.
2
0
1
6
,
d
o
i:
1
0
.
1
0
1
6
/
j.
n
e
u
c
o
m
.
2
0
1
6
.
0
8
.
0
3
9
.
[3
1
]
L.
De
n
g
,
“
A
tu
t
o
rial
su
r
v
e
y
o
f
a
r
c
h
it
e
c
tu
re
s,
a
lg
o
rit
h
m
s,
a
n
d
a
p
p
li
c
a
ti
o
n
s
fo
r
d
e
e
p
lea
r
n
i
n
g
,
”
AP
S
I
PA
T
ra
n
s
a
c
ti
o
n
s
o
n
S
ig
n
a
l
a
n
d
In
f
o
rm
a
ti
o
n
Pro
c
e
ss
in
g
,
v
o
l.
3
,
2
0
1
4
,
d
o
i:
1
0
.
1
0
1
7
/atsip
.
2
0
1
3
.
9
.
[3
2
]
M
.
Ba
k
a
t
o
r
a
n
d
D.
Ra
d
o
sa
v
,
“
De
e
p
Lea
rn
in
g
a
n
d
M
e
d
ica
l
Di
a
g
n
o
sis:
A
Re
v
iew
o
f
Li
tera
tu
r
e
,
”
M
u
lt
imo
d
a
l
T
e
c
h
n
o
l
o
g
ies
a
n
d
I
n
ter
a
c
ti
o
n
,
v
o
l
.
2
,
n
o
.
3
,
2
0
1
8
,
d
o
i:
1
0
.
3
3
9
0
/mti
2
0
3
0
0
4
7
.
[3
3
]
W.
L.
Al
y
o
u
b
i
,
W.
M
.
S
h
a
las
h
,
a
n
d
M
.
F
.
Ab
u
lk
h
a
ir,
“
Dia
b
e
t
ic
re
ti
n
o
p
a
th
y
d
e
tec
ti
o
n
t
h
ro
u
g
h
d
e
e
p
lea
rn
i
n
g
tec
h
n
iq
u
e
s: A
re
v
iew
,
”
In
fo
rm
a
t
i
c
s in
M
e
d
icin
e
Un
l
o
c
k
e
d
,
v
o
l.
2
0
,
2
0
2
0
,
d
o
i:
1
0
.
1
0
1
6
/j
.
imu
.
2
0
2
0
.
1
0
0
3
7
7
.
[3
4
]
W.
Au
c
c
a
h
u
a
si,
e
t
a
l
.
,
“
Re
c
ig
n
i
ti
o
n
o
f
h
a
rd
e
x
u
d
a
tes
u
si
n
g
De
e
p
Lea
rn
in
g
,
”
in
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Co
mp
u
t
a
ti
o
n
a
l
I
n
telli
g
e
n
c
e
a
n
d
Da
ta
S
c
ien
c
e
,
v
o
l.
1
6
7
,
2
0
1
9
,
p
p
.
2
3
4
3
-
2
3
5
3
,
d
o
i:
1
0
.
1
0
1
6
/
j.
p
r
o
c
s.2
0
2
0
.
0
3
.
2
8
7
.
[3
5
]
M
.
M
a
tee
n
,
J.
Wen
,
N.
N
a
sru
ll
a
h
,
S
.
S
u
n
,
a
n
d
S
.
Ha
y
a
t,
“
E
x
u
d
a
te
De
tec
ti
o
n
fo
r
Dia
b
e
ti
c
Re
ti
n
o
p
a
t
h
y
Us
i
n
g
P
re
train
e
d
Co
n
v
o
l
u
ti
o
n
a
l
Ne
u
ra
l
Ne
two
rk
s,”
Co
mp
lex
it
y
,
v
o
l.
2
0
2
0
,
p
p
.
1
–
1
1
,
2
0
2
0
,
d
o
i:
1
0
.
1
1
5
5
/
2
0
2
0
/5
8
0
1
8
7
0
.
[3
6
]
K.
Ad
e
m
,
“
Ex
u
d
a
te
d
e
tec
ti
o
n
f
o
r
d
iab
e
ti
c
re
ti
n
o
p
a
t
h
y
wit
h
c
ircu
la
r
Ho
u
g
h
tran
sf
o
rm
a
ti
o
n
a
n
d
c
o
n
v
o
lu
ti
o
n
a
l
n
e
u
ra
l
n
e
two
rk
s,”
Ex
p
e
rt S
y
ste
ms
wit
h
A
p
p
li
c
a
ti
o
n
s
,
v
o
l
.
1
1
4
,
p
p
.
2
8
9
–
2
9
5
,
De
c
.
2
0
1
8
,
d
o
i
:
1
0
.
1
0
1
6
/j
.
e
sw
a
.
2
0
1
8
.
0
7
.
0
5
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
23
,
No
.
2
,
Au
g
u
s
t 2
0
2
1
:
837
-
8
4
6
846
[3
7
]
S
.
Yu
,
D.
Xia
o
,
a
n
d
Y.
Ka
n
a
g
a
sin
g
a
m
,
“
Ex
u
d
a
te
d
e
tec
ti
o
n
fo
r
d
iab
e
ti
c
re
ti
n
o
p
a
th
y
wit
h
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
two
rk
s,”
i
n
2
0
1
7
3
9
th
A
n
n
u
a
l
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
f
t
h
e
IEE
E
En
g
i
n
e
e
rin
g
i
n
M
e
d
icin
e
a
n
d
Bi
o
l
o
g
y
S
o
c
iety
(EM
BC)
,
2
0
1
7
,
p
p
.
1
7
4
4
-
1
7
4
7
,
d
o
i:
1
0
.
1
1
0
9
/em
b
c
.
2
0
1
7
.
8
0
3
7
1
8
0
.
[3
8
]
A.
Os
a
re
h
,
M
.
M
irme
h
d
i
,
B.
T
h
o
m
a
s,
a
n
d
R.
M
a
rk
h
a
m
,
“
Au
t
o
m
a
ted
id
e
n
ti
fica
ti
o
n
o
f
d
i
a
b
e
ti
c
re
t
in
a
l
e
x
u
d
a
tes
i
n
d
ig
it
a
l
c
o
l
o
u
r
ima
g
e
s,”
Brit
is
h
J
o
u
rn
a
l
o
f
Op
h
th
a
lmo
l
o
g
y
,
v
o
l
.
8
7
,
n
o
.
1
0
,
p
p
.
1
2
2
0
–
1
2
2
3
,
2
0
0
3
,
d
o
i:
1
0
.
1
1
3
6
/b
jo
.
8
7
.
1
0
.
1
2
2
0
.
[3
9
]
A.
Os
a
re
h
,
B.
S
h
a
d
g
a
r
,
a
n
d
R
.
M
a
rk
h
a
m
,
“
A
Co
m
p
u
tati
o
n
a
l
-
In
telli
g
e
n
c
e
-
Ba
se
d
Ap
p
ro
a
c
h
f
o
r
De
tec
ti
o
n
o
f
Ex
u
d
a
tes
in
Dia
b
e
ti
c
Re
ti
n
o
p
a
t
h
y
Im
a
g
e
s,”
IEE
E
T
r
a
n
sa
c
ti
o
n
s
o
n
In
fo
rm
a
t
io
n
T
e
c
h
n
o
lo
g
y
i
n
Bi
o
me
d
icin
e
,
v
o
l.
1
3
,
n
o
.
4
,
p
p
.
5
3
5
–
5
4
5
,
Ju
l
.
2
0
0
9
,
d
o
i
:
1
0
.
1
1
0
9
/T
ITB.
2
0
0
8
.
2
0
0
7
4
9
3
.
[4
0
]
S
.
W.
F
ra
n
k
li
n
a
n
d
S
.
E
.
Ra
jan
,
“
Dia
g
n
o
sis
o
f
d
iab
e
t
ic
re
ti
n
o
p
a
th
y
b
y
e
m
p
lo
y
i
n
g
ima
g
e
p
ro
c
e
ss
in
g
tec
h
n
i
q
u
e
t
o
d
e
tec
t
e
x
u
d
a
tes
in
re
ti
n
a
l
ima
g
e
s,”
IET
Ima
g
e
Pr
o
c
e
ss
in
g
,
v
o
l.
8
,
n
o
.
1
0
,
p
p
.
6
0
1
–
6
0
9
,
Oc
t.
2
0
1
4
,
d
o
i:
1
0
.
1
0
4
9
/i
e
t
-
ip
r.
2
0
1
3
.
0
5
6
5
.
[4
1
]
U.
M
.
Ak
ra
m
a
n
d
S
.
A.
K
h
a
n
,
“
Au
to
m
a
ted
De
tec
ti
o
n
o
f
Da
rk
a
n
d
Bri
g
h
t
Les
io
n
s
i
n
Re
ti
n
a
l
I
m
a
g
e
s
fo
r
Early
De
tec
ti
o
n
o
f
Dia
b
e
ti
c
Re
ti
n
o
p
a
t
h
y
,
”
J
o
u
r
n
a
l
o
f
M
e
d
ica
l
S
y
ste
ms
,
v
o
l.
3
6
,
n
o
.
5
,
p
p
.
3
1
5
1
–
3
1
6
2
,
Oc
t.
2
0
1
2
,
d
o
i:
1
0
.
1
0
0
7
/s1
0
9
1
6
-
0
1
1
-
9
8
0
2
-
2.
[4
2
]
N.
Th
e
e
ra
-
Um
p
o
n
,
I.
P
o
o
n
k
a
se
m
,
S
.
Au
e
p
h
a
n
wir
iy
a
k
u
l
,
a
n
d
D.
P
a
ti
k
u
lsil
a
,
“
Ha
rd
e
x
u
d
a
te
d
e
t
e
c
ti
o
n
in
re
ti
n
a
l
fu
n
d
u
s
ima
g
e
s
u
sin
g
su
p
e
rv
ise
d
l
e
a
rn
in
g
,
”
Ne
u
ra
l
Co
mp
u
ti
n
g
a
n
d
Ap
p
li
c
a
ti
o
n
s
,
v
o
l.
3
2
,
p
p
.
1
3
0
7
9
–
1
3
0
9
6
,
2
0
2
0
,
d
o
i:
1
0
.
1
0
0
7
/s0
0
5
2
1
-
0
1
9
-
0
4
4
0
2
-
7.
[4
3
]
R.
Ro
m
e
ro
-
Ora
á
,
M
.
G
a
rc
ía,
J.
Ora
á
-
P
é
re
z
,
M
.
I.
Ló
p
e
z
-
G
á
lv
e
z
a
n
d
R.
Ho
rn
e
ro
,
“
Eff
e
c
ti
v
e
F
u
n
d
u
s
Im
a
g
e
De
c
o
m
p
o
siti
o
n
fo
r
th
e
De
tec
ti
o
n
o
f
Re
d
Les
io
n
s
a
n
d
Ha
rd
E
x
u
d
a
tes
t
o
Aid
in
th
e
Dia
g
n
o
s
is
o
f
Dia
b
e
ti
c
Re
ti
n
o
p
a
t
h
y
,
”
S
e
n
s
o
rs
,
v
o
l.
2
0
,
n
o
.
2
2
,
2
0
2
0
,
d
o
i:
1
0
.
3
3
9
0
/s2
0
2
2
6
5
4
9
.
[4
4
]
R.
Va
larm
a
th
i
a
n
d
S
.
S
a
ra
v
a
n
a
n
,
“
Ex
u
d
a
te
c
h
a
ra
c
teriz
a
ti
o
n
to
d
iag
n
o
se
d
iab
e
ti
c
re
ti
n
o
p
a
th
y
u
si
n
g
g
e
n
e
ra
li
z
e
d
m
e
th
o
d
,
”
J
o
u
rn
a
l
o
f
Amb
ien
t
I
n
telli
g
e
n
c
e
a
n
d
Hu
m
a
n
ize
d
C
o
mp
u
ti
n
g
,
v
o
l.
1
2
,
p
p
.
3
6
3
3
–
3
6
4
5
,
M
a
r.
2
0
2
1
,
d
o
i:
1
0
.
1
0
0
7
/s1
2
6
5
2
-
0
1
9
-
0
1
6
1
7
-
3.
[4
5
]
Z.
S
i,
D
.
F
u
,
Y.
L
iu
,
a
n
d
Z
.
Hu
a
n
g
,
“
Ha
rd
e
x
u
d
a
te
se
g
m
e
n
tatio
n
in
re
ti
n
a
l
ima
g
e
with
a
tt
e
n
ti
o
n
m
e
c
h
a
n
ism
,”
IET
Ima
g
e
Pro
c
e
ss
,
v
o
l.
1
5
,
n
o
.
3
,
p
p
.
587
–
5
9
7
,
F
e
b
.
2
0
2
1
,
d
o
i:
1
0
.
1
0
4
9
/i
p
r
2
.
1
2
0
0
7
.
[4
6
]
A.
M
.
As
h
ir
,
S
.
Ib
ra
h
im,
M
.
A
b
d
u
lg
h
a
n
i,
A.
A.
Ib
ra
h
im
,
a
n
d
M
.
S
.
An
wa
r,
“
Dia
b
e
ti
c
Re
ti
n
o
p
a
th
y
De
tec
ti
o
n
Us
in
g
Lo
c
a
l
Ex
trem
a
Qu
a
n
ti
z
e
d
Ha
ra
li
c
k
F
e
a
tu
re
s
with
Lo
n
g
S
h
o
rt
-
Term
M
e
m
o
ry
Ne
two
rk
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Bi
o
me
d
ica
l
Ima
g
in
g
,
v
o
l
.
2
0
2
1
,
2
0
2
1
,
d
o
i:
1
0
.
1
1
5
5
/2
0
2
1
/
6
6
1
8
6
6
6
.
[4
7
]
G
.
K
a
ly
a
n
i,
B.
Ja
n
a
k
iram
a
iah
,
A.
Ka
ru
n
a
,
a
n
d
L.
V.
Na
ra
sim
h
a
P
ra
sa
d
,
“
Dia
b
e
ti
c
re
ti
n
o
p
a
th
y
d
e
tec
ti
o
n
a
n
d
c
las
sifica
ti
o
n
u
sin
g
c
a
p
su
le n
e
two
rk
s,”
Co
mp
lex
&
I
n
telli
g
e
n
t
S
y
ste
ms
,
2
0
2
1
,
d
o
i:
1
0
.
1
0
0
7
/s4
0
7
4
7
-
0
2
1
-
0
0
3
1
8
-
9.
[4
8
]
S
.
L.
Ale
e
n
a
a
n
d
C.
A.
P
ra
ji
t
h
,
"
Re
ti
n
a
l
Les
io
n
s
De
tec
ti
o
n
f
o
r
S
c
re
e
n
in
g
o
f
Dia
b
e
ti
c
Re
ti
n
o
p
a
th
y
,
"
in
2
0
2
0
1
1
t
h
In
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
Co
mp
u
ti
n
g
,
C
o
mm
u
n
ic
a
ti
o
n
a
n
d
N
e
two
rk
in
g
T
e
c
h
n
o
l
o
g
ies
(ICCCN
T
)
,
Kh
a
ra
g
p
u
r,
In
d
ia,
2
0
2
0
,
p
p
.
1
-
6
,
d
o
i:
1
0
.
1
1
0
9
/ICCCNT4
9
2
3
9
.
2
0
2
0
.
9
2
2
5
6
1
7
.
[4
9
]
S
.
L
o
n
g
,
X.
Hu
a
n
g
,
Z.
C
h
e
n
,
S
.
P
a
rd
h
a
n
,
a
n
d
D.
Zh
e
n
g
,
"
Au
t
o
m
a
ti
c
De
tec
ti
o
n
o
f
Ha
rd
Ex
u
d
a
tes
i
n
Co
lo
r
Re
ti
n
a
l
Im
a
g
e
s
Us
in
g
Dy
n
a
m
ic
Th
re
sh
o
ld
a
n
d
S
VM
Clas
sifica
ti
o
n
:
Al
g
o
rit
h
m
De
v
e
lo
p
m
e
n
t
a
n
d
E
v
a
lu
a
ti
o
n
,
"
Bi
o
M
e
d
Res
e
a
rc
h
In
ter
n
a
ti
o
n
a
l
,
v
o
l
.
2
0
1
9
,
2
0
1
9
,
d
o
i:
1
0
.
1
1
5
5
/2
0
1
9
/
3
9
2
6
9
3
0
.
[5
0
]
R.
Bh
a
rg
a
v
i
a
n
d
R.
K.
S
e
n
a
p
a
t
i,
“
Brig
h
t
les
i
o
n
d
e
tec
ti
o
n
i
n
c
o
lo
r
f
u
n
d
u
s ima
g
e
s b
a
se
d
o
n
tex
t
u
re
fe
a
tu
re
s,”
Bu
ll
e
ti
n
o
f
e
lec
trica
l
e
n
g
in
e
e
rin
g
a
n
d
in
f
o
rm
a
ti
c
s,
v
o
l.
5
,
n
o
.
1
,
p
p
9
2
-
1
0
0
,
M
a
rc
h
2
0
1
6
,
d
o
i:
1
0
.
1
1
5
9
1
/ee
i.
v
5
i1
.
5
5
3
.
[5
1
]
M.
S
.
M
a
h
e
sw
a
ri
a
n
d
A.
P
u
n
n
o
li
l
,
“
A
No
v
e
l
A
p
p
r
o
a
c
h
fo
r
Re
ti
n
a
l
Les
io
n
De
tec
ti
o
n
Dia
b
e
ti
c
Re
ti
n
o
p
a
th
y
Im
a
g
e
s,”
in
2
0
1
4
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
In
n
o
v
a
ti
o
n
s
i
n
En
g
in
e
e
rin
g
a
n
d
T
e
c
h
n
o
lo
g
y
,
In
d
ia,
2
0
1
4
,
v
o
l.
3
,
n
o
.
3
,
p
p
.
1
1
0
9
-
1
1
1
4
.
[5
2
]
A.
D.
F
lem
in
g
,
e
t
a
l
.
,
“
Au
to
m
a
ted
d
e
tec
ti
o
n
o
f
e
x
u
d
a
tes
f
o
r
d
iab
e
ti
c
re
ti
n
o
p
a
t
h
y
sc
re
e
n
in
g
,
”
Ph
y
si
c
s
in
M
e
d
ici
n
e
a
n
d
Bi
o
lo
g
y
,
v
o
l.
5
2
,
n
o
.
2
4
,
p
p
.
7
3
8
5
–
7
3
9
6
,
De
c
.
2
0
0
7
,
d
o
i:
1
0
.
1
0
8
8
/
0
0
3
1
-
9
1
5
5
/
5
2
/
2
4
/0
1
2
.
[5
3
]
S
.
Ra
th
o
re
,
A.
As
wa
l
,
a
n
d
P
.
S
a
ra
n
y
a
,
“
Bri
g
h
t
Les
io
n
De
tec
ti
o
n
in
Re
ti
n
a
l
F
u
n
d
u
s
Im
a
g
e
s
fo
r
Dia
b
e
t
ic
Re
ti
n
o
p
a
th
y
De
tec
ti
o
n
Us
in
g
M
a
c
h
i
n
e
Lea
rn
i
n
g
A
p
p
r
o
a
c
h
,
”
A
n
n
a
ls
o
f
t
h
e
Ro
ma
n
ia
n
S
o
c
iety
f
o
r
Ce
ll
B
io
l
o
g
y
,
v
o
l
.
2
5
,
n
o
.
5
,
p
p
.
4
3
6
0
–
4
3
6
7
,
2
0
2
1
.
[5
4
]
A.
Co
l
o
m
e
r,
J.
Ig
u
a
l
,
a
n
d
V.
Na
ra
n
jo
,
“
De
tec
ti
o
n
o
f
Early
S
ig
n
s
o
f
Dia
b
e
ti
c
Re
ti
n
o
p
a
t
h
y
Ba
se
d
o
n
Te
x
tu
ra
l
a
n
d
M
o
rp
h
o
l
o
g
ica
l
In
f
o
rm
a
ti
o
n
in
F
u
n
d
u
s
Im
a
g
e
s,”
S
e
n
so
rs
,
v
o
l
.
2
0
,
n
o
.
4
,
F
e
b
.
2
0
2
0
,
d
o
i:
1
0
.
3
3
9
0
/s
2
0
0
4
1
0
0
5
.
[5
5
]
A.
B.
Ka
d
a
n
a
n
d
P
.
S
.
S
u
b
b
ian
,
“
De
tec
ti
o
n
o
f
Ha
rd
E
x
u
d
a
tes
Us
in
g
E
v
o
lu
ti
o
n
a
r
y
F
e
a
tu
re
S
e
lec
ti
o
n
in
Re
ti
n
a
l
F
u
n
d
u
s Im
a
g
e
s,”
J
o
u
rn
a
l
o
f
M
e
d
i
c
a
l
S
y
ste
ms
,
v
o
l.
4
3
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/s1
0
9
1
6
-
0
1
9
-
1
3
4
9
-
7.
[5
6
]
M
.
J.
J.
P
.
v
a
n
G
rin
sv
e
n
,
A
.
Ch
a
k
ra
v
a
rty
,
J.
S
i
v
a
sw
a
m
y
,
T.
T
h
e
e
len
,
B.
v
a
n
G
in
n
e
k
e
n
,
a
n
d
C.
I
.
S
á
n
c
h
e
z
,
“
A
Ba
g
o
f
Wo
r
d
s
a
p
p
r
o
a
c
h
fo
r
d
isc
rimi
n
a
ti
n
g
b
e
twe
e
n
re
ti
n
a
l
ima
g
e
s
c
o
n
tai
n
in
g
e
x
u
d
a
tes
o
r
d
r
u
se
n
,
”
2
0
1
3
IEE
E
1
0
t
h
In
ter
n
a
t
io
n
a
l
S
y
mp
o
si
u
m o
n
B
io
me
d
ica
l
Ima
g
in
g
,
2
0
1
3
,
p
p
.
1
4
4
4
-
1
4
4
7
,
d
o
i:
1
0
.
1
1
0
9
/I
S
BI.
2
0
1
3
.
6
5
5
6
8
0
6
.
[5
7
]
N.
G
.
Ra
n
a
m
u
k
a
a
n
d
R.
G
.
N.
M
e
e
g
a
m
a
,
“
De
tec
ti
o
n
o
f
h
a
rd
e
x
u
d
a
tes
fro
m
d
iab
e
t
ic
re
ti
n
o
p
a
t
h
y
im
a
g
e
s
u
sin
g
f
u
z
z
y
lo
g
ic,”
IE
T
Ima
g
e
Pro
c
e
ss
in
g
,
v
o
l.
7
,
n
o
.
2
,
p
p
.
1
2
1
–
1
3
0
,
M
a
r.
2
0
1
3
,
d
o
i:
1
0
.
1
0
4
9
/i
e
t
-
i
p
r.
2
0
1
2
.
0
1
3
4
.
[5
8
]
A.
P
ra
d
e
e
p
a
n
d
F
.
Jo
se
p
h
,
“
Bi
n
a
r
y
o
p
e
ra
ti
o
n
b
a
se
d
h
a
rd
e
x
u
d
a
te
d
e
tec
ti
o
n
a
n
d
fu
z
z
y
b
a
se
d
c
las
sifica
ti
o
n
in
d
iab
e
ti
c
re
ti
n
a
l
fu
n
d
u
s
ima
g
e
s
f
o
r
re
a
l
ti
m
e
d
iag
n
o
sis
a
p
p
li
c
a
ti
o
n
s,”
In
te
rn
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
Co
m
p
u
te
r
En
g
i
n
e
e
rin
g
(
IJ
ECE
)
,
v
o
l.
1
0
,
n
o
.
3
,
p
p
.
2
3
0
5
-
2
3
1
2
,
2
0
2
0
,
d
o
i:
1
0
.
1
1
5
9
1
/i
jec
e
.
v
1
0
i
3
.
p
p
2
3
0
5
-
2
3
1
2
.
[5
9
]
L.
G
ian
c
a
rd
o
,
e
t
a
l
.
,
“
Brig
h
t
Re
ti
n
a
l
Les
io
n
s
De
tec
ti
o
n
u
sin
g
C
o
l
o
r
F
u
n
d
u
s
Im
a
g
e
s
Co
n
tain
i
n
g
Re
flec
ti
v
e
F
e
a
tu
re
s,”
in
W
o
rld
C
o
n
g
re
ss
o
n
M
e
d
ica
l
Ph
y
sic
s
a
n
d
Bi
o
me
d
ic
a
l
En
g
i
n
e
e
rin
g
,
M
u
n
ich
,
G
e
rm
a
n
y
,
2
0
0
9
,
p
p
.
2
9
2
–
2
9
5
,
d
o
i:
1
0
.
1
0
0
7
/9
7
8
-
3
-
6
4
2
-
0
3
8
9
1
-
4
_
7
8
.
[6
0
]
H.
Wan
g
,
e
t
a
l
.
,
“
Ha
rd
e
x
u
d
a
t
e
d
e
tec
ti
o
n
b
a
se
d
o
n
d
e
e
p
m
o
d
e
l
lea
rn
e
d
in
f
o
rm
a
ti
o
n
a
n
d
m
u
lt
i
-
fe
a
tu
re
jo
in
t
re
p
re
se
n
tatio
n
f
o
r
d
ia
b
e
ti
c
re
ti
n
o
p
a
th
y
sc
re
e
n
in
g
,
”
Co
mp
u
ter
M
e
th
o
d
s
a
n
d
Pro
g
ra
ms
i
n
Bi
o
me
d
i
c
in
e
,
v
o
l
.
1
9
1
,
Ju
l.
2
0
2
0
,
d
o
i:
1
0
.
1
0
1
6
/
j.
c
m
p
b
.
2
0
2
0
.
1
0
5
3
9
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.