I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
pu
t
er
E
ng
ineering
(
I
J
E
CE
)
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
,
p
p
.
2845
~
2
8
5
6
I
SS
N:
2088
-
8
7
0
8
,
DOI
:
1
0
.
1
1
5
9
1
/ijece.
v
15
i
3
.
pp
2
8
4
5
-
2
8
5
6
2845
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Dev
elo
pment of
w
a
tershed alg
o
rith
m for ide
ntif
i
ca
ti
o
n of
dia
betic
r
etinopa
t
hy
bas
ed on fun
d
us ima
g
es
Su
rm
a
y
a
nti
1
,
Su
m
ij
a
n
1
,
Sa
if
ul B
uk
ho
ri
2
1
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
c
s
M
a
n
a
g
e
m
e
n
t
,
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
F
a
c
u
l
t
y
,
U
n
i
v
e
r
si
t
a
s
P
u
t
r
a
I
n
d
o
n
e
s
i
a
Y
P
T
K
,
P
a
d
a
n
g
,
I
n
d
o
n
e
s
i
a
1
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
Te
c
h
n
o
l
o
g
y
,
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
F
a
c
u
l
t
y
,
U
n
i
v
e
r
si
t
a
s
P
u
t
r
a
I
n
d
o
n
e
s
i
a
Y
P
T
K
,
P
a
d
a
n
g
,
I
n
d
o
n
e
s
i
a
2
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
Te
c
h
n
o
l
o
g
y
,
U
n
i
v
e
r
si
t
a
s
Jem
b
e
r
,
Jem
b
e
r
,
I
n
d
o
n
e
s
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
No
v
2
6
,
2
0
2
4
R
ev
is
ed
Ma
r
1
2
,
2
0
2
5
Acc
ep
ted
Ma
r
2
0
,
2
0
2
5
Dia
b
e
ti
c
re
ti
n
o
p
a
th
y
(DR)
is
a
se
rio
u
s
c
o
m
p
li
c
a
ti
o
n
o
f
d
iab
e
tes
th
a
t
c
a
n
lea
d
to
b
li
n
d
n
e
ss
if
n
o
t
d
e
tec
ted
e
a
rly
.
Th
is
re
se
a
rc
h
p
re
se
n
ts
a
n
o
v
e
l
m
e
th
o
d
f
o
r
th
e
i
d
e
n
ti
fica
ti
o
n
o
f
DR
u
sin
g
fu
n
d
u
s
ima
g
e
s,
e
m
p
lo
y
i
n
g
th
e
Wate
rsh
e
d
Alg
o
rit
h
m
fo
r
a
c
c
u
ra
te
ima
g
e
se
g
m
e
n
tatio
n
a
n
d
th
e
g
ra
y
lev
e
l
c
o
-
o
c
c
u
rre
n
c
e
m
a
tri
x
(G
LCM
)
fo
r
tex
t
u
re
fe
a
tu
r
e
e
x
trac
ti
o
n
.
Th
e
ima
g
e
p
ro
c
e
ss
in
g
p
i
p
e
li
n
e
in
v
o
lv
e
s
se
v
e
ra
l
sta
g
e
s,
i
n
c
lu
d
i
n
g
g
ra
y
sc
a
le
c
o
n
v
e
rsio
n
,
n
o
ise
re
d
u
c
ti
o
n
th
ro
u
g
h
G
a
u
ss
ian
a
n
d
m
e
d
ian
fil
ters
,
a
n
d
Otsu
'
s
th
re
sh
o
l
d
in
g
t
o
i
so
late
k
e
y
fe
a
tu
re
s
su
c
h
a
s
re
ti
n
a
l
les
io
n
s
.
Th
e
wa
ters
h
e
d
a
lg
o
rit
h
m
is
a
p
p
li
e
d
t
o
d
e
li
n
e
a
te
th
e
b
o
u
n
d
a
ries
o
f
a
b
n
o
rm
a
l
re
g
io
n
s,
wh
i
le
th
e
G
LCM
m
e
th
o
d
e
x
trac
ts
tex
tu
re
fe
a
tu
re
s
li
k
e
c
o
n
t
ra
st,
c
o
rre
latio
n
,
e
n
e
rg
y
,
a
n
d
h
o
m
o
g
e
n
e
it
y
,
wh
ich
a
re
e
ss
e
n
ti
a
l
f
o
r
d
iag
n
o
sin
g
re
ti
n
a
l
a
b
n
o
rm
a
li
ti
e
s.
Th
e
p
ro
p
o
se
d
a
p
p
ro
a
c
h
d
e
m
o
n
stra
tes
a
h
ig
h
a
c
c
u
ra
c
y
ra
te
o
f
9
2
%
,
su
c
c
e
ss
fu
ll
y
i
d
e
n
ti
f
y
in
g
a
b
n
o
rm
a
li
ti
e
s
i
n
4
6
o
u
t
o
f
5
0
f
u
n
d
u
s
ima
g
e
s.
T
h
e
m
e
th
o
d
s
h
o
ws
s
ig
n
ifi
c
a
n
t
p
o
ten
ti
a
l
f
o
r
e
n
h
a
n
c
i
n
g
e
a
rly
d
e
tec
ti
o
n
o
f
DR,
p
ro
v
i
d
in
g
a
c
c
u
ra
te
se
g
m
e
n
tatio
n
a
n
d
tex
t
u
re
a
n
a
ly
s
is,
m
a
k
in
g
it
a
v
a
l
u
a
b
le
t
o
o
l
f
o
r
m
e
d
ica
l
p
ro
fe
ss
io
n
a
ls i
n
d
iag
n
o
si
n
g
re
ti
n
a
l
d
ise
a
se
s.
K
ey
w
o
r
d
s
:
Diab
etic
r
etin
o
p
ath
y
E
x
tr
ac
tio
n
m
eth
o
d
Fu
n
d
u
s
im
ag
e
I
d
en
tific
atio
n
W
ater
s
h
ed
alg
o
r
ith
m
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
:
Su
r
m
ay
an
ti
Dep
ar
tm
en
t o
f
I
n
f
o
r
m
atics M
an
ag
em
en
t,
Facu
lty
o
f
C
o
m
p
u
t
er
Scien
ce
,
Un
iv
er
s
itas
Pu
tr
a
I
n
d
o
n
esia Y
PTK
Pad
an
g
2
5
2
2
1
,
Pad
an
g
,
W
est Su
m
ater
a,
I
n
d
o
n
esia
E
m
ail:
s
u
r
m
ay
an
ti@
u
p
iy
p
tk
.
a
c.
id
1.
I
NT
RO
D
UCT
I
O
N
Diab
etic
r
etin
o
p
ath
y
(
DR
)
is
a
lead
in
g
ca
u
s
e
o
f
b
lin
d
n
ess
wo
r
ld
wid
e,
p
r
im
ar
ily
af
f
ec
tin
g
i
n
d
iv
id
u
als
with
d
iab
etes d
u
e
to
u
n
c
o
n
tr
o
l
led
b
lo
o
d
s
u
g
ar
lev
els th
at
d
a
m
ag
e
th
e
r
etin
al
b
lo
o
d
v
ess
els
[
1
]
.
E
ar
ly
d
iag
n
o
s
is
an
d
tr
ea
tm
en
t
o
f
DR
ar
e
cr
u
cial
to
p
r
ev
en
t
ir
r
ev
er
s
ib
le
v
is
io
n
lo
s
s
;
h
o
wev
er
,
ex
is
tin
g
m
an
u
al
s
cr
ee
n
in
g
m
eth
o
d
s
ar
e
tim
e
-
in
ten
s
iv
e,
r
eq
u
ir
e
s
p
ec
ialized
ex
p
er
tis
e
f
r
o
m
o
p
h
t
h
alm
o
lo
g
is
ts
,
an
d
ar
e
im
p
r
ac
t
ical
f
o
r
lar
g
e
-
s
ca
le
im
p
lem
en
tatio
n
[
2
]
.
Pre
v
io
u
s
s
o
lu
tio
n
s
,
s
u
ch
as
th
o
s
e
em
p
lo
y
in
g
b
asic
m
o
r
p
h
o
lo
g
i
ca
l
o
p
er
atio
n
s
o
r
co
n
v
o
l
u
tio
n
al
n
eu
r
al
n
etwo
r
k
s
(
C
NN
s
)
,
h
av
e
m
ad
e
s
ig
n
if
ican
t
s
tr
id
es
in
au
to
m
atin
g
DR
d
etec
tio
n
b
u
t
f
ac
e
lim
itatio
n
s
[
3
]
.
T
h
ese
in
clu
d
e
ch
allen
g
es
in
ac
cu
r
ately
s
eg
m
en
tin
g
p
ath
o
lo
g
ical
f
ea
tu
r
es
lik
e
le
s
io
n
s
an
d
ex
u
d
ates
in
n
o
is
y
o
r
co
m
p
lex
f
u
n
d
u
s
im
ag
es,
h
ig
h
c
o
m
p
u
t
atio
n
al
co
s
ts
,
an
d
lim
ited
g
en
er
aliza
b
ilit
y
ac
r
o
s
s
d
iv
er
s
e
d
atasets
[
4
]
.
O
u
r
r
esear
ch
aim
s
t
o
o
v
er
c
o
m
e
t
h
ese
c
o
n
s
tr
ain
ts
b
y
in
teg
r
atin
g
t
h
e
wate
r
s
h
ed
alg
o
r
ith
m
f
o
r
p
r
ec
is
e
im
ag
e
s
eg
m
en
tati
o
n
an
d
th
e
g
r
ay
lev
el
co
-
o
cc
u
r
r
en
ce
m
atr
ix
(
GL
C
M)
f
o
r
r
o
b
u
s
t
tex
tu
r
e
f
ea
tu
r
e
ex
tr
ac
tio
n
[
5
]
.
T
h
is
n
o
v
el
a
p
p
r
o
ac
h
s
ee
k
s
to
e
n
h
an
ce
s
eg
m
en
tatio
n
ac
cu
r
ac
y
,
im
p
r
o
v
e
f
e
atu
r
e
an
aly
s
is
,
an
d
p
r
o
v
id
e
an
ef
f
icien
t,
s
ca
lab
le
s
o
lu
tio
n
f
o
r
ea
r
ly
DR
d
etec
tio
n
,
u
ltima
tely
a
d
d
r
ess
in
g
th
e
lim
itatio
n
s
o
f
ex
is
tin
g
m
eth
o
d
o
l
o
g
ies an
d
s
u
p
p
o
r
tin
g
tim
ely
d
iag
n
o
s
is
an
d
in
ter
v
en
tio
n
[
6
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
8
4
5
-
2
8
5
6
2846
Mo
r
p
h
o
l
o
g
ical
o
p
er
atio
n
s
in
th
e
f
ield
o
f
d
ig
ital
im
ag
e
p
r
o
ce
s
s
in
g
ar
e
tech
n
iq
u
es
th
at
u
s
e
s
tr
u
ctu
r
in
g
elem
en
ts
to
p
r
o
ce
s
s
an
d
an
aly
ze
s
h
ap
es
an
d
s
tr
u
ctu
r
es
with
i
n
im
ag
es
[
7
]
.
I
n
th
e
co
n
tex
t
o
f
f
u
n
d
u
s
im
ag
es
f
o
r
th
e
id
en
tific
atio
n
o
f
DR
,
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
p
lay
a
cr
u
cial
r
o
le
in
ex
t
r
ac
tin
g
a
n
d
id
en
t
if
y
in
g
p
ath
o
l
o
g
ical
f
ea
tu
r
es
s
u
ch
as
m
icr
o
an
e
u
r
y
s
m
s
,
h
ar
d
ex
u
d
ates,
an
d
h
em
o
r
r
h
ag
es
[
8
]
.
B
asic
m
o
r
p
h
o
l
o
g
i
ca
l
tech
n
iq
u
es
lik
e
d
ilatio
n
an
d
er
o
s
io
n
h
elp
en
la
r
g
e
an
d
s
h
r
in
k
o
b
ject
ar
ea
s
to
f
ill
g
ap
s
o
r
r
e
m
o
v
e
n
o
is
e.
T
h
e
co
m
b
in
atio
n
o
f
d
ilatio
n
an
d
er
o
s
io
n
,
k
n
o
w
n
as o
p
en
in
g
a
n
d
clo
s
in
g
o
p
er
atio
n
s
,
is
u
s
ed
to
im
p
r
o
v
e
im
ag
e
q
u
ality
b
y
r
em
o
v
in
g
s
m
a
ll
n
o
is
e
an
d
s
m
o
o
th
in
g
o
b
j
ec
t
co
n
to
u
r
s
[
9
]
,
[
1
0
]
.
Fo
r
in
s
tan
ce
,
th
e
o
p
en
i
n
g
o
p
e
r
atio
n
ca
n
b
e
u
s
ed
to
r
em
o
v
e
n
o
is
e
f
r
o
m
th
e
b
ac
k
g
r
o
u
n
d
o
f
f
u
n
d
u
s
im
a
g
es,
wh
ile
th
e
clo
s
in
g
o
p
e
r
atio
n
ca
n
h
elp
f
ill
g
ap
s
i
n
d
am
a
g
ed
v
ascu
lar
s
tr
u
ctu
r
es.
Ad
d
itio
n
ally
,
o
p
er
a
tio
n
s
s
u
ch
as
to
p
-
h
at
a
n
d
b
o
tto
m
-
h
at
tr
an
s
f
o
r
m
atio
n
s
a
r
e
u
s
e
d
to
h
ig
h
lig
h
t
b
r
i
g
h
t
o
r
d
ar
k
f
ea
tu
r
es
o
n
a
n
u
n
e
v
en
b
ac
k
g
r
o
u
n
d
,
ai
d
in
g
in
t
h
e
d
etec
tio
n
o
f
m
icr
o
an
e
u
r
y
s
m
s
an
d
ex
u
d
ates.
Sk
eleto
n
izatio
n
is
an
o
th
e
r
tec
h
n
iq
u
e
t
h
at
r
ed
u
ce
s
v
ascu
lar
o
b
jects
to
th
in
lin
es,
m
ain
tain
in
g
th
e
to
p
o
lo
g
ical
s
tr
u
ctu
r
e
f
o
r
f
u
r
th
e
r
an
al
y
s
is
[
1
1
]
.
T
h
e
im
p
lem
en
tatio
n
o
f
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
a
llo
ws
f
o
r
en
h
an
ce
d
co
n
tr
ast
an
d
v
is
ib
ilit
y
o
f
p
at
h
o
lo
g
ical
f
ea
tu
r
es,
f
ac
ilit
atin
g
au
to
m
atic
d
etec
tio
n
an
d
an
aly
s
is
o
f
DR
u
s
in
g
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
.
B
y
im
p
r
o
v
in
g
th
e
q
u
ality
o
f
f
u
n
d
u
s
im
ag
es
an
d
h
ig
h
lig
h
tin
g
cr
itical
f
ea
tu
r
es,
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
s
ig
n
if
ican
tly
co
n
tr
i
b
u
te
to
th
e
ac
c
u
r
ac
y
an
d
ef
f
icien
cy
o
f
ea
r
l
y
D
R
d
etec
tio
n
s
y
s
tem
s
,
wh
ich
ar
e
ess
en
tial f
o
r
p
r
ev
en
t
in
g
b
lin
d
n
ess
in
d
iab
etic
p
atie
n
ts
.
R
esear
ch
o
n
th
is
to
p
ic
h
as
p
r
ev
io
u
s
ly
b
ee
n
co
n
d
u
cted
b
y
Sam
an
et
a
l
.
[
1
2
]
,
f
o
cu
s
in
g
o
n
th
e
au
to
m
atic
d
etec
tio
n
an
d
s
ev
er
ity
class
if
i
ca
tio
n
o
f
DR
.
T
h
e
aim
was
to
d
ev
elo
p
a
m
et
h
o
d
f
o
r
th
e
a
u
t
o
m
atic
d
etec
tio
n
o
f
DR
an
d
p
r
o
p
o
s
e
a
m
o
d
el
to
d
et
er
m
in
e
its
p
r
o
g
r
ess
io
n
/s
ev
er
ity
u
s
in
g
f
u
n
d
u
s
im
a
g
es.
T
h
is
m
e
th
o
d
was
d
esig
n
ed
to
d
etec
t
DR
ef
f
ec
tiv
ely
an
d
ef
f
icien
tly
b
e
f
o
r
e
ca
u
s
in
g
e
y
e
d
am
ag
e,
with
o
u
t
r
eq
u
ir
i
n
g
a
n
o
p
h
th
alm
o
lo
g
is
t's
p
r
esen
ce
.
Ma
n
u
al
s
cr
ee
n
in
g
n
ec
ess
itate
s
o
p
h
th
alm
o
lo
g
is
ts
an
d
s
u
b
s
tan
tial
tim
e
r
eso
u
r
ce
s
.
Dete
ctin
g
ex
u
d
ates
is
cr
u
cial
f
o
r
d
iag
n
o
s
in
g
DR
.
T
h
e
ap
p
r
o
ac
h
in
clu
d
e
d
:
i
)
ex
tr
ac
tin
g
s
ig
n
if
ican
t
f
ea
tu
r
es
f
r
o
m
th
e
im
ag
es,
s
u
ch
as
b
lo
o
d
v
ess
els,
th
e
o
p
tic
d
is
c
(
OD)
,
ex
u
d
ates,
an
d
m
icr
o
a
n
eu
r
y
s
m
s
u
s
in
g
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
,
a
n
d
ii)
class
if
y
in
g
th
e
s
ev
er
ity
as
m
ild
o
r
m
o
d
e
r
ate
u
s
in
g
a
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
class
if
ier
to
ass
is
t
o
p
h
th
alm
o
lo
g
is
ts
.
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
'
s
p
er
f
o
r
m
an
ce
was
ev
alu
ated
an
d
ap
p
r
o
v
e
d
b
y
o
p
h
th
alm
o
lo
g
is
ts
.
T
h
is
p
ap
er
co
n
tr
i
b
u
tes
to
th
e
f
ield
o
f
au
to
m
atic
d
etec
tio
n
o
f
ab
n
o
r
m
al
s
tr
u
ctu
r
es
an
d
th
eir
s
ev
er
ity
.
An
o
th
er
s
tu
d
y
b
y
Sar
k
i
et
a
l.
[
1
3
]
ex
am
in
e
d
im
ag
e
p
r
ep
r
o
ce
s
s
in
g
in
th
e
class
if
icatio
n
an
d
id
en
tific
atio
n
o
f
DR
.
T
h
is
p
ap
er
s
y
s
tem
atica
lly
s
tu
d
ied
th
e
im
p
o
r
tan
ce
o
f
im
ag
e
p
r
o
ce
s
s
in
g
f
o
r
DR
class
if
icatio
n
.
T
h
e
p
r
o
p
o
s
ed
au
to
m
ati
c
class
if
icatio
n
f
r
am
ewo
r
k
f
o
r
DR
wa
s
ac
h
iev
ed
th
r
o
u
g
h
s
ev
er
al
s
tep
s
:
im
ag
e
q
u
ality
en
h
an
ce
m
en
t,
im
ag
e
s
eg
m
en
tatio
n
(
r
eg
io
n
o
f
in
ter
e
s
t)
,
im
a
g
e
au
g
m
en
tatio
n
(
g
eo
m
etr
ic
tr
an
s
f
o
r
m
atio
n
)
,
an
d
cl
ass
if
icatio
n
.
Op
tim
al
r
esu
lts
wer
e
o
b
tain
ed
u
s
in
g
tr
ad
itio
n
al
im
ag
e
p
r
o
ce
s
s
in
g
m
eth
o
d
s
co
m
b
in
e
d
with
a
n
ewly
co
n
s
tr
u
cted
C
NN
ar
ch
i
tectu
r
e.
T
h
e
n
ewly
b
u
ilt
C
NN,
co
m
b
in
ed
with
tr
ad
itio
n
al
im
ag
e
p
r
o
ce
s
s
in
g
ap
p
r
o
ac
h
es,
s
h
o
wed
th
e
b
est
p
er
f
o
r
m
an
ce
with
h
i
g
h
ac
cu
r
a
cy
in
DR
class
if
icatio
n
.
E
x
p
er
im
en
tal
r
esu
lts
d
em
o
n
s
tr
ated
ad
eq
u
ate
ac
cu
r
ac
y
,
s
p
ec
if
icity
,
an
d
s
en
s
itiv
ity
.
Fu
r
th
er
m
o
r
e
,
L
o
n
g
et
a
l.
[
1
4
]
co
n
d
u
cted
r
esear
ch
o
n
m
ac
h
in
e
lear
n
in
g
-
b
ased
d
ete
ctio
n
an
d
class
if
icatio
n
o
f
DR
.
Fu
n
d
u
s
i
m
ag
in
g
to
o
ls
an
d
o
p
en
-
d
o
m
ai
n
d
atasets
f
o
r
DR
-
r
elate
d
r
esea
r
ch
wer
e
in
tr
o
d
u
ce
d
.
Var
io
u
s
d
iag
n
o
s
tic
m
eth
o
d
s
b
ased
o
n
m
ac
h
in
e
lear
n
in
g
wer
e
also
in
tr
o
d
u
ce
d
,
in
cl
u
d
in
g
ex
u
d
ate
an
d
m
icr
o
an
eu
r
y
s
m
d
etec
tio
n
m
eth
o
d
s
,
DR
class
if
icatio
n
m
eth
o
d
s
with
d
if
f
er
en
t
s
ev
er
ity
s
ca
le
s
,
an
d
s
eg
m
en
tatio
n
m
eth
o
d
s
f
o
r
th
e
o
p
tic
d
is
c
an
d
b
lo
o
d
v
ess
els.
T
h
e
ap
p
licatio
n
o
f
m
ac
h
in
e
lear
n
in
g
m
o
d
els
in
r
ea
l
-
life
s
ce
n
ar
io
s
was a
ls
o
d
is
cu
s
s
ed
.
Mo
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
in
d
i
g
ital im
ag
e
p
r
o
ce
s
s
in
g
h
av
e
s
ee
n
s
ig
n
if
ican
t a
d
v
an
ce
m
en
ts
in
r
ec
en
t
y
ea
r
s
,
esp
ec
ially
with
th
e
in
teg
r
atio
n
o
f
n
ew
tech
n
iq
u
es
an
d
co
m
p
u
tatio
n
al
im
p
r
o
v
em
en
ts
.
C
lass
ical
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
lik
e
d
ilatio
n
,
er
o
s
io
n
,
o
p
e
n
in
g
,
a
n
d
clo
s
in
g
h
av
e
b
ee
n
ex
p
an
d
e
d
with
m
o
r
e
co
m
p
lex
tech
n
iq
u
es.
Fo
r
ex
am
p
le,
to
p
-
h
at
an
d
b
o
tto
m
-
h
at
t
r
an
s
f
o
r
m
a
tio
n
s
ar
e
n
o
w
co
m
m
o
n
l
y
u
s
ed
to
h
ig
h
lig
h
t
s
p
ec
if
ic
f
ea
tu
r
es
in
im
ag
es.
Ad
d
itio
n
a
lly
,
tech
n
iq
u
es
s
u
ch
as
s
k
eleto
n
izatio
n
h
a
v
e
b
ee
n
im
p
r
o
v
e
d
to
p
r
o
v
id
e
m
o
r
e
ac
cu
r
ate
r
ep
r
esen
tatio
n
s
o
f
t
h
e
m
o
r
p
h
o
lo
g
ical
s
tr
u
ctu
r
e
o
f
o
b
jects
in
im
ag
es.
I
n
teg
r
at
in
g
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
with
m
ac
h
in
e
lear
n
in
g
,
p
ar
ticu
lar
ly
n
e
u
r
al
n
etwo
r
k
s
,
h
as
p
av
e
d
th
e
way
f
o
r
m
o
r
e
ac
c
u
r
ate
a
n
d
ef
f
icien
t
d
etec
tio
n
an
d
class
if
icatio
n
.
Fo
r
in
s
tan
ce
,
C
NN
o
f
ten
u
s
e
m
o
r
p
h
o
lo
g
ical
o
p
er
a
tio
n
s
as
p
ar
t
o
f
th
e
p
r
ep
r
o
ce
s
s
in
g
s
tag
e
to
im
p
r
o
v
e
im
ag
e
q
u
ality
b
ef
o
r
e
f
ea
tu
r
e
ex
tr
ac
tio
n
an
d
f
u
r
th
er
class
if
icatio
n
.
Stu
d
ies
h
av
e
s
h
o
wn
th
at
th
is
co
m
b
in
atio
n
c
an
en
h
a
n
ce
th
e
ac
c
u
r
ac
y
o
f
d
e
tectin
g
p
ath
o
l
o
g
ical
f
ea
tu
r
e
s
lik
e
m
icr
o
a
n
eu
r
y
s
m
s
an
d
ex
u
d
ates in
m
ed
ical
im
a
g
es,
s
p
ec
if
ically
in
f
u
n
d
u
s
im
ag
es.
DR
r
em
ain
s
a
s
ig
n
if
ican
t
ca
u
s
e
o
f
b
lin
d
n
ess
,
p
ar
ticu
lar
l
y
i
n
d
iab
etic
p
o
p
u
latio
n
s
,
d
u
e
t
o
d
elay
e
d
d
etec
tio
n
an
d
d
iag
n
o
s
is
.
E
x
is
tin
g
m
eth
o
d
o
l
o
g
ies
f
o
r
DR
d
ete
ctio
n
,
s
u
ch
as
t
h
o
s
e
b
y
Sam
an
et
a
l.
[
1
2
]
,
em
p
lo
y
ed
im
ag
e
p
r
o
ce
s
s
in
g
with
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
an
d
m
ac
h
in
e
lear
n
in
g
m
o
d
els
to
class
if
y
s
ev
er
ity
b
u
t
f
ac
e
d
ch
allen
g
es
in
ac
h
iev
in
g
p
r
ec
is
io
n
u
n
d
e
r
v
ar
y
in
g
im
ag
e
q
u
ali
ties
.
Similar
ly
,
Sar
k
i
et
a
l.
[
1
3
]
d
ev
elo
p
e
d
C
NNs
co
m
b
in
ed
with
p
r
e
p
r
o
ce
s
s
in
g
tech
n
iq
u
es,
y
ield
in
g
h
ig
h
ac
cu
r
ac
y
b
u
t
r
e
q
u
ir
in
g
s
u
b
s
ta
n
tial
co
m
p
u
tatio
n
al
r
eso
u
r
ce
s
.
Desp
ite
th
ese
ad
v
an
ce
m
en
ts
,
u
n
r
eso
lv
ed
ch
alle
n
g
es
in
clu
d
e
th
e
o
p
tim
izatio
n
o
f
s
eg
m
e
n
tatio
n
tech
n
iq
u
es
an
d
r
o
b
u
s
t
f
ea
tu
r
e
ex
tr
ac
tio
n
f
o
r
d
i
v
er
s
e
d
atasets
.
Ou
r
r
esear
ch
in
tr
o
d
u
ce
s
a
n
o
v
el
in
teg
r
atio
n
o
f
th
e
wate
r
s
h
ed
alg
o
r
ith
m
an
d
th
e
GL
C
M
f
o
r
p
r
ec
is
e
s
eg
m
en
tat
io
n
an
d
tex
tu
r
e
f
ea
tu
r
e
ex
tr
a
ctio
n
,
r
esp
ec
tiv
ely
.
T
h
ese
co
n
tr
ib
u
tio
n
s
ad
d
r
ess
g
ap
s
in
ef
f
icie
n
cy
a
n
d
r
eliab
ilit
y
,
d
em
o
n
s
tr
at
in
g
a
r
o
b
u
s
t a
p
p
r
o
ac
h
f
o
r
ea
r
ly
-
s
tag
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Dev
elo
p
men
t o
f wa
ters
h
ed
a
l
g
o
r
ith
m
fo
r
id
en
tifi
ca
tio
n
o
f
d
ia
b
etic
r
etin
o
p
a
th
y
b
a
s
ed
…
(
S
u
r
ma
ya
n
ti
)
2847
DR
d
iag
n
o
s
is
.
T
h
e
s
u
b
s
eq
u
e
n
t
s
ec
tio
n
s
d
etail
o
u
r
m
eth
o
d
o
lo
g
ical
in
n
o
v
atio
n
s
,
v
alid
ate
o
u
r
f
in
d
in
g
s
,
an
d
co
m
p
ar
e
th
e
m
ag
ain
s
t p
r
i
o
r
b
e
n
ch
m
ar
k
s
.
Desp
ite
s
ig
n
if
ican
t
ad
v
a
n
ce
m
e
n
ts
in
d
i
g
ital
im
ag
e
p
r
o
ce
s
s
in
g
an
d
m
ac
h
in
e
lear
n
i
n
g
,
ex
is
tin
g
m
eth
o
d
s
o
f
ten
lack
th
e
p
r
ec
is
io
n
an
d
e
f
f
icien
cy
n
ee
d
ed
f
o
r
ea
r
ly
-
s
tag
e
d
etec
tio
n
o
f
DR
.
Mo
r
eo
v
er
,
m
an
y
ap
p
r
o
ac
h
es
d
o
n
o
t
ad
eq
u
ately
in
teg
r
ate
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
to
en
h
an
ce
th
e
v
is
ib
ilit
y
an
d
id
en
tific
a
tio
n
o
f
p
ath
o
lo
g
ical
f
ea
tu
r
es.
T
h
is
g
ap
h
ig
h
lig
h
ts
th
e
n
ee
d
f
o
r
n
o
v
el
m
eth
o
d
o
lo
g
ies
th
at
o
p
tim
ize
im
ag
e
p
r
ep
r
o
ce
s
s
in
g
,
im
p
r
o
v
e
f
ea
tu
r
e
ex
t
r
ac
tio
n
,
an
d
lev
er
a
g
e
ad
v
a
n
ce
d
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
to
p
r
o
v
id
e
ac
cu
r
ate,
au
to
m
ated
DR
d
etec
tio
n
.
Ad
d
r
ess
in
g
th
ese
ch
allen
g
es,
th
is
p
ap
er
p
r
o
p
o
s
es
th
e
d
ev
elo
p
m
en
t
o
f
en
h
an
ce
d
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
a
n
d
th
eir
in
teg
r
atio
n
with
m
ac
h
i
n
e
lear
n
in
g
tech
n
iq
u
es
to
cr
e
ate
a
r
o
b
u
s
t
an
d
ef
f
icien
t
s
y
s
tem
f
o
r
d
etec
tin
g
an
d
class
if
y
in
g
DR
with
h
ig
h
p
r
ec
is
io
n
,
en
a
b
lin
g
tim
ely
in
ter
v
e
n
tio
n
s
to
p
r
e
v
en
t
v
is
io
n
lo
s
s
.
T
h
is
r
esear
ch
d
ev
el
o
p
s
m
o
r
p
h
o
lo
g
i
ca
l
o
p
er
atio
n
m
eth
o
d
s
to
id
e
n
tify
DR
b
ased
o
n
f
u
n
d
u
s
im
a
g
es.
T
h
e
n
o
v
elty
o
f
th
is
r
esear
ch
lies
in
th
e
cr
ea
tio
n
o
f
a
n
ew
o
p
e
r
atio
n
m
et
h
o
d
r
esu
ltin
g
f
r
o
m
th
e
d
ev
elo
p
m
en
t
o
f
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
,
wh
ic
h
ca
n
d
etec
t D
R
b
ased
o
n
f
u
n
d
u
s
im
ag
es m
o
r
e
p
r
ec
is
ely
an
d
ac
cu
r
ately
.
2.
M
E
T
H
O
D
2
.
1
.
F
r
a
m
ewo
r
k
re
s
ea
rc
h
T
h
is
s
tu
d
y
co
n
d
u
cte
d
th
e
d
ev
elo
p
m
en
t
o
f
m
o
r
p
h
o
lo
g
y
o
p
er
atio
n
f
o
r
id
e
n
tific
atio
n
o
f
DR
b
ased
o
n
f
u
n
d
u
s
im
ag
es.
T
h
er
ef
o
r
e,
th
e
r
esear
ch
er
d
esig
n
e
d
a
r
esear
c
h
f
r
a
m
ewo
r
k
f
ir
s
t
b
ef
o
r
e
c
o
n
d
u
ctin
g
t
h
e
r
esear
c
h
.
T
h
e
f
o
llo
win
g
Fig
u
r
e
1
is
th
e
r
esear
ch
f
r
am
ewo
r
k
in
th
is
s
tu
d
y
.
T
h
e
p
r
o
v
id
e
d
r
esear
ch
f
r
am
e
wo
r
k
r
ep
r
esen
ts
a
co
m
p
r
eh
en
s
iv
e
im
ag
e
p
r
o
ce
s
s
in
g
p
ip
elin
e
ap
p
lied
to
f
u
n
d
u
s
im
ag
es
f
o
r
m
ed
ical
an
a
ly
s
is
,
co
n
s
is
tin
g
o
f
th
r
ee
m
ain
s
tag
es: in
p
u
t,
p
r
o
ce
s
s
in
g
,
an
d
r
esu
lt.
T
h
e
p
r
o
ce
s
s
b
eg
in
s
with
th
e
ac
q
u
is
itio
n
o
f
an
in
p
u
t
im
ag
e,
s
p
ec
i
f
ically
a
f
u
n
d
u
s
im
a
g
e,
wh
ic
h
is
co
m
m
o
n
ly
u
s
ed
in
m
e
d
ical
d
iag
n
o
s
is
to
d
etec
t
r
etin
al
d
is
ea
s
es.
T
h
is
im
ag
e
s
er
v
es
as
th
e
r
aw
d
ata
f
o
r
th
e
s
u
b
s
eq
u
e
n
t
p
r
o
ce
s
s
in
g
task
s
.
I
n
th
e
p
r
o
ce
s
s
in
g
s
tag
e,
two
k
ey
p
h
ases
ar
e
in
v
o
lv
ed
: p
r
e
-
p
r
o
ce
s
s
in
g
an
d
im
ag
e
s
eg
m
en
tatio
n
/ex
tr
ac
tio
n
.
Du
r
in
g
p
r
e
-
p
r
o
ce
s
s
in
g
,
th
e
im
a
g
e
is
en
h
an
ce
d
to
im
p
r
o
v
e
its
q
u
ality
b
y
r
e
d
u
cin
g
n
o
is
e
an
d
tr
an
s
f
o
r
m
in
g
it
in
to
a
m
o
r
e
s
u
itab
le
f
o
r
m
at
f
o
r
s
eg
m
en
tati
o
n
.
First,
th
e
r
ed
,
g
r
ee
n
,
an
d
b
l
u
e
(
R
GB
)
f
u
n
d
u
s
im
ag
e
is
co
n
v
er
ted
to
g
r
ay
s
ca
le
to
r
ed
u
ce
co
m
p
u
tatio
n
al
co
m
p
lex
ity
an
d
f
o
cu
s
o
n
in
ten
s
ity
v
ar
iatio
n
s
cr
u
cial
f
o
r
m
ed
ical
i
m
ag
e
an
aly
s
is
.
No
is
e
r
ed
u
ctio
n
is
th
en
c
o
n
d
u
cted
i
n
two
s
tag
es:
h
ar
d
n
o
is
e
r
e
d
u
ctio
n
u
s
in
g
a
Ga
u
s
s
ian
f
ilter
to
r
em
o
v
e
h
ig
h
-
fr
eq
u
e
n
cy
n
o
is
e,
f
o
llo
wed
b
y
s
o
f
t
n
o
is
e
r
ed
u
ctio
n
th
r
o
u
g
h
a
m
ed
ian
f
ilter
to
s
m
o
o
th
th
e
i
m
ag
e
an
d
p
r
eser
v
e
ed
g
es.
Ots
u
’
s
th
r
esh
o
ld
in
g
m
eth
o
d
is
th
en
em
p
lo
y
ed
t
o
b
i
n
ar
ize
th
e
im
ag
e
b
y
au
to
m
at
ically
s
elec
tin
g
an
o
p
tim
al
th
r
esh
o
ld
.
Af
ter
p
r
e
-
p
r
o
ce
s
s
in
g
,
th
e
n
ex
t
s
tep
in
v
o
lv
es
im
ag
e
s
eg
m
en
tatio
n
,
wh
ich
is
o
lates
r
eg
io
n
s
o
f
in
ter
est
th
at
m
ay
in
d
icate
s
p
ec
if
ic
f
ea
tu
r
es
s
u
ch
as
DR
.
A
d
is
tan
ce
tr
an
s
f
o
r
m
is
ap
p
l
ied
to
id
e
n
tify
th
e
f
o
r
eg
r
o
u
n
d
,
a
n
d
m
o
r
p
h
o
lo
g
ica
l o
p
er
atio
n
s
a
r
e
u
s
ed
t
o
d
is
tin
g
u
is
h
th
e
b
ac
k
g
r
o
u
n
d
.
T
h
e
wate
r
s
h
ed
alg
o
r
ith
m
is
th
en
ap
p
lied
t
o
s
eg
m
e
n
t
d
is
ti
n
c
t
o
b
jects
in
th
e
im
ag
e
b
ased
o
n
in
ten
s
ity
g
r
ad
ie
n
ts
.
On
ce
th
e
i
m
ag
e
is
s
eg
m
e
n
ted
,
im
ag
e
ex
tr
ac
tio
n
o
cc
u
r
s
th
r
o
u
g
h
th
e
u
s
e
o
f
t
h
e
GL
C
M
m
eth
o
d
,
wh
ich
e
x
tr
ac
ts
k
ey
tex
tu
r
e
f
ea
tu
r
es
in
clu
d
in
g
co
n
tr
ast,
co
r
r
elatio
n
,
en
e
r
g
y
,
an
d
h
o
m
o
g
en
eity
.
Fin
ally
,
th
e
R
esu
lt
s
tag
e
p
r
esen
ts
th
e
o
u
tco
m
e
o
f
th
e
s
eg
m
en
tatio
n
an
d
f
ea
t
u
r
e
e
x
t
r
ac
tio
n
p
r
o
ce
s
s
es.
T
h
e
o
u
tp
u
t
is
a
p
r
o
ce
s
s
ed
im
ag
e
with
d
is
tin
ct
s
eg
m
en
ted
r
eg
io
n
s
,
h
ig
h
lig
h
tin
g
ar
ea
s
th
a
t c
o
u
ld
b
e
o
f
clin
ical
in
ter
est,
s
u
ch
as a
b
n
o
r
m
al
tis
s
u
e
o
r
lesi
o
n
s
.
Fig
u
r
e
1
.
R
esear
ch
f
r
am
ewo
r
k
2
.
2
.
F
r
a
m
ewo
r
k
re
s
ea
rc
h de
t
a
ils
2
.
2
.
1
.
I
np
ut
im
a
g
e
(
f
un
du
s
im
a
g
e)
T
y
p
ically
,
a
h
ig
h
-
q
u
ality
f
u
n
d
u
s
ca
m
er
a
is
u
s
ed
to
ca
p
tu
r
e
im
ag
es
o
f
th
e
r
etin
a.
Op
h
th
a
lm
o
lo
g
is
ts
u
tili
ze
an
o
p
h
th
alm
o
s
co
p
e,
a
s
im
p
le
to
o
l
with
a
l
ig
h
t
s
o
u
r
ce
,
to
ex
am
in
e
th
e
ey
e’
s
in
ter
io
r
.
T
h
e
in
p
u
t
im
ag
e
r
ep
r
esen
ts
a
c
o
lo
r
im
a
g
e
o
f
t
h
e
r
etin
a.
Fu
n
d
u
s
c
a
m
er
as
ar
e
g
en
er
ally
u
s
ed
to
ca
p
tu
r
e
r
etin
al
im
ag
es,
r
esu
ltin
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
8
4
5
-
2
8
5
6
2848
in
a
b
in
ar
y
im
ag
e
as th
e
o
u
tp
u
t.
T
h
e
p
r
o
ce
s
s
b
eg
in
s
b
y
in
p
u
ttin
g
an
im
a
g
e
o
f
a
DR
f
u
n
d
u
s
.
An
ex
am
p
le
o
f
DR
f
u
n
d
u
s
im
ag
e
d
ata
is
s
h
o
wn
in
Fig
u
r
e
2
.
Fig
u
r
e
2
.
I
n
p
u
t
i
m
ag
e
(
f
u
n
d
u
s
im
ag
e)
2
.
2
.
2
.
P
re
-
pro
ce
s
s
ing
A.
R
GB
to
g
r
ay
s
ca
le
Gr
ay
s
ca
le
m
ig
h
t
b
e
en
o
u
g
h
to
id
en
tify
s
o
m
e
o
b
jects.
C
o
lo
r
p
h
o
to
s
ca
n
ad
d
e
x
tr
a
c
o
m
p
lex
it
y
an
d
n
ee
d
m
o
r
e
m
em
o
r
y
s
p
ac
e
s
in
ce
th
ey
ca
r
r
y
m
o
r
e
in
f
o
r
m
atio
n
th
an
b
lack
-
an
d
-
wh
ite
g
r
a
p
h
ics
[
1
5
]
.
Gr
ay
s
ca
le
co
n
v
er
s
io
n
o
f
co
lo
r
r
etin
a
im
a
g
es
lo
wer
s
th
e
am
o
u
n
t
o
f
p
ix
els
th
at
n
ee
d
to
b
e
p
r
o
ce
s
s
ed
.
An
p
ictu
r
e
is
s
aid
to
b
e
g
r
a
y
s
ca
le
if
o
n
ly
in
ten
s
ity
i
n
f
o
r
m
atio
n
ca
n
b
e
d
eter
m
in
e
d
f
r
o
m
a
s
in
g
le
s
am
p
le
p
ix
el
v
a
lu
e
[
1
6
]
.
Ma
p
p
in
g
d
if
f
er
en
t
c
o
lo
r
c
h
an
n
els
(
R
,
G
,
an
d
B
)
i
n
to
a
s
in
g
le
g
r
ay
s
ca
l
e
v
alu
e
in
ter
m
s
o
f
a
weig
h
ted
s
u
m
is
th
e
p
r
o
ce
s
s
o
f
g
r
a
y
s
ca
le
co
n
v
er
s
io
n
.
T
h
e
f
o
llo
win
g
is
th
e
f
o
r
m
u
la
f
o
r
co
n
v
er
tin
g
g
r
ay
s
ca
le
b
as
ed
o
n
illu
m
in
atio
n
ch
ar
ac
ter
is
tics
:
=
0
.
2989
∗
+
0
.
5870
∗
+
0
,
1140
∗
(
1
)
T
h
e
co
lo
r
o
f
a
p
ix
el
is
r
e
p
r
esen
ted
b
y
th
e
letter
s
R
,
G,
a
n
d
B
,
r
esp
ec
tiv
ely
.
Hu
m
an
p
er
ce
p
tio
n
s
o
f
c
o
lo
r
ar
e
r
ep
r
esen
ted
b
y
th
e
R
,
G,
an
d
B
co
ef
f
icien
ts
.
Gr
ay
s
ca
le
c
o
n
v
er
ts
R
GB
(
3
-
d
im
en
s
io
n
al
p
ix
el
v
alu
es)
in
to
1
D
v
alu
es,
wh
ich
ar
e
s
im
p
ler
.
Ma
k
e
u
s
e
o
f
th
e
r
g
b
2
g
r
a
y
(
im
g
)
m
eth
o
d
.
1)
Har
d
n
o
is
e
r
ed
u
ctio
n
:
g
a
u
s
s
ian
f
ilter
Har
d
n
o
is
e
r
ed
u
ctio
n
u
s
in
g
t
h
e
g
au
s
s
ian
f
ilt
er
in
d
ig
ital
im
a
g
es,
p
ar
ticu
lar
ly
in
f
u
n
d
u
s
im
ag
es,
r
ef
er
s
to
th
e
p
r
o
ce
s
s
o
f
s
m
o
o
th
i
n
g
th
e
im
ag
e
b
y
r
e
d
u
cin
g
h
ig
h
-
f
r
eq
u
en
cy
n
o
is
e
o
r
s
h
ar
p
in
ten
s
ity
v
ar
iatio
n
s
th
at
ca
n
in
ter
f
er
e
with
im
ag
e
an
aly
s
is
[
1
7
]
.
T
h
e
Gau
s
s
ian
Fil
ter
is
wid
ely
u
s
ed
b
ec
au
s
e
o
f
its
ab
ilit
y
to
p
r
eser
v
e
ed
g
es
an
d
s
m
o
o
th
e
n
o
u
t
n
o
is
e
ef
f
ic
ien
tly
[
1
8
]
.
T
h
is
h
elp
s
in
r
ed
u
cin
g
r
a
n
d
o
m
n
o
is
e
with
o
u
t
b
lu
r
r
in
g
th
e
ed
g
es
ex
ce
s
s
iv
ely
.
Fo
r
f
u
n
d
u
s
im
ag
e
s
,
th
is
m
ea
n
s
en
h
an
cin
g
v
is
ib
ilit
y
o
f
cr
itical
s
tr
u
ctu
r
es
wh
ile
r
em
o
v
in
g
ir
r
ele
v
an
t
n
o
is
e,
aid
in
g
in
clea
r
er
d
ia
g
n
o
s
tics
.
T
h
e
Gau
s
s
ian
Fil
ter
i
s
r
ep
r
esen
ted
m
ath
e
m
atica
lly
as
(
2
)
(
,
)
=
1
2
2
−
2
+
2
2
2
(
2
)
T
h
is
f
u
n
ctio
n
cr
ea
tes
a
f
ilter
w
h
er
e
v
al
u
es
clo
s
er
to
th
e
ce
n
ter
(
wh
er
e)
(
,
)
is
th
e
v
alu
e
o
f
th
e
Gau
s
s
ian
f
ilter
at
co
o
r
d
in
ates
(
,
)
,
is
th
e
s
tan
d
a
r
d
d
ev
iatio
n
,
wh
ic
h
co
n
tr
o
ls
th
e
am
o
u
n
t o
f
s
m
o
o
t
h
in
g
,
(
,
)
ar
e
th
e
p
ix
el
co
o
r
d
in
ates
r
elativ
e
to
th
e
ce
n
ter
o
f
th
e
k
er
n
el.
I
n
f
u
n
d
u
s
im
ag
in
g
,
u
s
in
g
th
e
Gau
s
s
ian
f
ilter
f
o
r
h
ar
d
n
o
is
e
r
ed
u
ctio
n
is
an
ess
en
tial
p
r
ep
r
o
ce
s
s
in
g
s
tep
.
I
t
h
elp
s
in
r
ed
u
c
in
g
u
n
n
ec
ess
ar
y
n
o
is
e
wh
ile
p
r
eser
v
in
g
im
p
o
r
tan
t
m
ed
ical
f
ea
tu
r
es th
at
ar
e
c
r
itical
f
o
r
d
ia
g
n
o
s
tic
p
u
r
p
o
s
es.
2)
So
f
t n
o
is
e
r
ed
u
ctio
n
:
m
ed
ian
f
ilter
So
f
t
n
o
is
e
r
ed
u
c
ti
o
n
u
s
in
g
t
h
e
m
ed
ia
n
f
il
ter
i
s
a
m
eth
o
d
co
m
m
o
n
ly
ap
p
li
ed
t
o
d
ig
i
ta
l
i
m
ag
es
,
in
c
lu
d
in
g
f
u
n
d
u
s
im
ag
e
s
,
to
r
e
d
u
ce
n
o
i
s
e
wh
il
e
p
r
e
s
er
v
in
g
i
m
p
o
r
t
an
t s
tr
u
ctu
r
al
d
e
ta
il
s
,
e
s
p
ec
i
al
ly
ed
g
e
s
[
1
9
]
.
T
h
e
m
ed
ian
f
i
lte
r
i
s
p
ar
ti
cu
lar
ly
ef
f
ec
tiv
e
in
ad
d
r
e
s
s
in
g
th
is
ty
p
e
o
f
n
o
is
e
w
ith
o
u
t
o
v
er
ly
b
lu
r
r
in
g
th
e
ed
g
e
s
o
f
k
ey
s
tr
u
ctu
r
e
s
[
2
0
]
.
T
h
e
p
r
im
ar
y
p
u
r
p
o
s
e
o
f
u
s
in
g
th
e
M
ed
ia
n
f
i
lt
er
i
s
to
r
e
m
o
v
e
im
p
u
l
s
e
n
o
is
e
(
s
u
ch
as
s
a
lt
-
an
d
-
p
ep
p
e
r
n
o
i
s
e)
f
r
o
m
t
h
e
im
ag
e
wh
il
e
m
a
in
tai
n
in
g
th
e
s
h
a
r
p
n
e
s
s
o
f
im
p
o
r
ta
n
t
f
ea
tu
r
e
s
.
W
h
i
le
th
e
m
ed
i
an
f
i
lt
er
d
o
e
s
n
o
t
h
av
e
a
m
a
th
em
at
ic
al
f
o
r
m
u
l
a
in
th
e
s
am
e
way
a
s
o
th
er
f
i
lt
er
s
,
i
t
s
o
p
er
at
io
n
ca
n
b
e
d
es
cr
i
b
ed
a
lg
o
r
i
th
m
ic
al
ly
a
s
f
o
llo
w
s
:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Dev
elo
p
men
t o
f wa
ters
h
ed
a
l
g
o
r
ith
m
fo
r
id
en
tifi
ca
tio
n
o
f
d
ia
b
etic
r
etin
o
p
a
th
y
b
a
s
ed
…
(
S
u
r
ma
ya
n
ti
)
2849
−
Fo
r
ea
ch
p
ix
el
in
t
h
e
im
ag
e
,
a
n
eig
h
b
o
r
h
o
o
d
(
u
s
u
ally
a
s
q
u
a
r
e
o
r
r
ec
tan
g
u
lar
win
d
o
w)
ar
o
u
n
d
th
e
p
ix
el
is
d
ef
in
ed
.
−
T
h
e
p
ix
el
v
al
u
es with
in
th
is
n
eig
h
b
o
r
h
o
o
d
ar
e
s
o
r
te
d
in
asce
n
d
in
g
o
r
d
er
.
−
T
h
e
p
ix
el
at
th
e
ce
n
ter
o
f
t
h
e
n
eig
h
b
o
r
h
o
o
d
is
r
ep
lace
d
with
th
e
m
ed
ian
v
alu
e
f
r
o
m
th
e
s
o
r
ted
lis
t
o
f
n
eig
h
b
o
r
in
g
p
ix
el
v
alu
es.
I
n
f
u
n
d
u
s
im
ag
in
g
,
ap
p
ly
in
g
th
e
Me
d
ian
Fil
ter
f
o
r
s
o
f
t
n
o
is
e
r
ed
u
ctio
n
is
ess
en
tial
f
o
r
im
p
r
o
v
in
g
im
ag
e
q
u
ality
,
p
a
r
ticu
lar
ly
wh
en
th
e
im
ag
e
is
af
f
ec
te
d
b
y
im
p
u
ls
e
n
o
is
e.
Un
lik
e
o
th
er
s
m
o
o
th
in
g
tech
n
i
q
u
es,
th
e
m
ed
ian
f
ilter
p
r
eser
v
es
ed
g
es,
wh
ich
is
cr
u
cia
l
f
o
r
m
ain
tain
i
n
g
th
e
in
teg
r
ity
o
f
im
p
o
r
tan
t
r
et
in
al
s
tr
u
ctu
r
es
s
u
ch
as
b
lo
o
d
v
ess
els
an
d
lesi
o
n
s
.
B
y
r
ed
u
cin
g
n
o
is
e
in
th
is
m
an
n
er
,
th
e
m
ed
ian
f
ilter
en
h
an
c
es
th
e
clar
ity
o
f
th
e
im
ag
e,
m
ak
in
g
it m
o
r
e
s
u
itab
le
f
o
r
s
u
b
s
eq
u
en
t p
r
o
ce
s
s
in
g
s
tag
es lik
e
s
eg
m
en
ta
tio
n
o
r
f
ea
t
u
r
e
ex
tr
ac
tio
n
.
3)
T
h
r
esh
o
ld
in
g
: O
ts
u
'
s
th
r
esh
o
l
d
in
g
T
h
r
esh
o
ld
in
g
is
a
f
u
n
d
am
en
tal
im
ag
e
s
eg
m
en
tatio
n
tech
n
iq
u
e
u
s
ed
to
d
if
f
er
en
tiate
o
b
jects o
f
in
ter
est
f
r
o
m
th
e
b
ac
k
g
r
o
u
n
d
in
d
ig
ital
im
ag
es,
in
clu
d
in
g
f
u
n
d
u
s
im
ag
es
[
2
1
]
.
T
h
r
esh
o
ld
in
g
s
im
p
lifie
s
im
ag
e
an
aly
s
is
b
y
co
n
v
er
tin
g
g
r
ay
s
ca
le
im
ag
es
in
to
b
in
ar
y
im
ag
es,
wh
er
e
t
h
e
p
ix
el
in
ten
s
ity
is
eith
er
s
et
to
a
b
ac
k
g
r
o
u
n
d
o
r
f
o
r
eg
r
o
u
n
d
v
alu
e,
b
ased
o
n
a
th
r
esh
o
ld
[
2
2
]
.
T
h
e
m
eth
o
d
wo
r
k
s
b
y
ex
h
a
u
s
tiv
ely
s
ea
r
ch
in
g
f
o
r
th
e
th
r
esh
o
ld
th
at
s
ep
ar
ates
th
e
f
o
r
eg
r
o
u
n
d
(
th
e
o
b
ject
o
f
in
ter
est,
lik
e
b
lo
o
d
v
ess
els
o
r
lesi
o
n
s
)
f
r
o
m
th
e
b
ac
k
g
r
o
u
n
d
.
T
h
e
f
o
r
m
u
la
f
o
r
Ots
u
'
s
m
eth
o
d
ca
n
b
e
ex
p
r
ess
ed
as
(
3
)
2
(
)
=
1
(
)
1
2
(
)
+
2
(
)
2
2
(
)
(
3
)
w
h
er
e:
is
th
e
th
r
esh
o
ld
v
alu
e,
1
(
)
an
d
2
(
)
ar
e
th
e
weig
h
ts
o
f
th
e
two
class
es
(
f
o
r
eg
r
o
u
n
d
an
d
b
ac
k
g
r
o
u
n
d
)
,
1
2
(
)
an
d
2
2
(
)
ar
e
th
e
v
ar
ian
ce
s
o
f
th
e
two
class
es.
T
h
e
p
u
r
p
o
s
e
o
f
u
s
in
g
Ots
u
'
s
th
r
es
h
o
ld
in
g
in
f
u
n
d
u
s
im
a
g
es
is
to
ac
c
u
r
a
tely
s
eg
m
en
t
im
p
o
r
tan
t
s
tr
u
ct
u
r
es,
s
u
ch
as
b
lo
o
d
v
ess
els
an
d
lesi
o
n
s
,
f
r
o
m
th
e
b
ac
k
g
r
o
u
n
d
f
o
r
f
u
r
th
er
an
al
y
s
is
.
I
t is p
ar
ticu
lar
ly
u
s
ef
u
l in
m
ed
ica
l d
iag
n
o
s
tics
,
wh
er
e
p
r
ec
is
e
s
eg
m
en
tatio
n
i
s
cr
itical
f
o
r
d
etec
tin
g
ab
n
o
r
m
a
liti
es.
O
ts
u
’
s
th
r
esh
o
ld
in
g
is
a
p
o
wer
f
u
l
to
o
l
f
o
r
s
eg
m
en
tin
g
f
u
n
d
u
s
im
ag
es
b
y
s
elec
tin
g
an
o
p
tim
al
th
r
esh
o
ld
v
alu
e,
lea
d
in
g
t
o
im
p
r
o
v
e
d
m
ed
ical
im
ag
e
a
n
aly
s
is
an
d
a
s
s
is
tin
g
in
th
e
ea
r
ly
d
etec
tio
n
o
f
e
y
e
d
is
ea
s
es.
B.
Pro
ce
s
s
in
g
1)
I
m
ag
e
s
eg
m
en
tatio
n
a.
Ma
r
k
er
f
o
r
eg
r
o
u
n
d
:
d
is
tan
ce
t
r
an
s
f
o
r
m
Ma
r
k
er
f
o
r
eg
r
o
u
n
d
with
d
is
tan
ce
tr
an
s
f
o
r
m
is
a
cr
u
cial
m
et
h
o
d
i
n
im
ag
e
s
eg
m
en
tatio
n
,
p
ar
ticu
lar
ly
f
o
r
m
ed
ical
im
a
g
in
g
s
u
ch
as
f
u
n
d
u
s
im
ag
es
[
2
3
]
.
Fu
n
d
u
s
im
a
g
es
ar
e
u
s
ed
to
ex
am
i
n
e
th
e
in
t
er
io
r
s
u
r
f
ac
e
o
f
th
e
ey
e,
in
clu
d
in
g
th
e
r
etin
a,
b
l
o
o
d
v
ess
els,
an
d
o
p
tic
d
is
c,
ess
e
n
tial
f
o
r
d
iag
n
o
s
in
g
v
ar
i
o
u
s
ey
e
d
is
ea
s
es
lik
e
DR
o
r
g
lau
c
o
m
a
[
2
4
]
.
T
h
e
d
is
tan
ce
tr
an
s
f
o
r
m
is
u
s
ed
in
s
eg
m
e
n
t
atio
n
to
co
m
p
u
te
t
h
e
m
in
im
u
m
d
is
tan
ce
f
r
o
m
ea
ch
p
ix
el
with
in
an
o
b
ject
(
e.
g
.
,
a
r
etin
al
b
lo
o
d
v
ess
el)
to
th
e
n
ea
r
est
b
o
u
n
d
ar
y
o
r
b
ac
k
g
r
o
u
n
d
p
ix
el.
I
t
h
elp
s
in
id
en
tify
in
g
d
is
tin
ct
r
eg
io
n
s
b
y
tr
an
s
f
o
r
m
in
g
th
e
b
in
ar
y
s
eg
m
en
ted
im
ag
e
in
to
a
d
is
tan
ce
m
ap
,
wh
er
e
h
ig
h
er
v
alu
es
r
ep
r
esen
t
p
ix
els
f
ar
th
er
f
r
o
m
th
e
o
b
ject'
s
ed
g
es.
T
h
e
d
i
s
tan
ce
tr
an
s
f
o
r
m
ca
n
b
e
m
ath
e
m
atica
lly
ex
p
r
ess
ed
as
(
4
)
.
(
)
=
∈
(
,
)
(
4
)
w
h
er
e
(
)
is
th
e
d
is
tan
ce
o
f
p
ix
el,
r
ep
r
esen
ts
th
e
b
o
u
n
d
ar
y
p
i
x
e
ls
,
(
,
)
i
s
th
e
E
u
clid
ea
n
d
is
t
an
ce
b
etwe
en
p
ix
el
.
On
ce
th
e
d
is
tan
ce
tr
a
n
s
f
o
r
m
is
ca
lcu
lated
,
m
ar
k
er
s
f
o
r
th
e
f
o
r
e
g
r
o
u
n
d
a
n
d
b
ac
k
g
r
o
u
n
d
ar
e
ass
ig
n
ed
,
wh
ich
ca
n
th
e
n
b
e
u
s
ed
in
wat
er
s
h
ed
s
eg
m
en
tatio
n
t
o
clea
r
ly
d
ef
in
e
o
b
ject
b
o
u
n
d
a
r
ies.
b.
W
ater
s
h
ed
alg
o
r
ith
m
T
h
e
wate
r
s
h
ed
alg
o
r
it
h
m
is
in
s
p
ir
ed
b
y
th
e
way
wate
r
f
lo
ws
o
v
er
a
lan
d
s
ca
p
e.
I
n
im
ag
e
s
eg
m
en
tatio
n
,
th
e
in
ten
s
ity
v
alu
es
o
f
p
ix
els
ar
e
tr
ea
ted
as
elev
atio
n
s
in
a
t
o
p
o
g
r
ap
h
ic
s
u
r
f
ac
e
[
2
5
]
.
T
h
e
alg
o
r
ith
m
s
im
u
lates
th
e
f
lo
o
d
in
g
o
f
th
is
s
u
r
f
ac
e,
w
h
er
e
wate
r
s
tar
ts
f
r
o
m
th
e
lo
west
in
ten
s
ity
p
o
in
ts
(
o
f
ten
co
r
r
esp
o
n
d
in
g
t
o
d
ar
k
p
ix
els
o
r
f
ea
tu
r
es
in
th
e
im
ag
e)
an
d
g
r
ad
u
ally
r
is
es
[
2
6
]
.
T
h
e
"wa
ter
s
h
ed
"
is
f
o
r
m
ed
wh
er
e
th
e
wate
r
f
r
o
m
d
if
f
er
en
t
ca
tch
m
e
n
t
b
asin
s
(
r
eg
io
n
s
)
m
ee
ts
,
ef
f
ec
tiv
ely
cr
ea
t
in
g
b
o
u
n
d
ar
ies
th
at
s
ep
ar
ate
d
is
tin
ct
r
eg
io
n
s
in
th
e
im
ag
e.
L
et
(
,
)
r
ep
r
esen
t
th
e
in
ten
s
ity
o
f
a
p
ix
el
at
co
o
r
d
in
ates
(
,
)
in
th
e
f
u
n
d
u
s
im
ag
e.
T
h
e
alg
o
r
ith
m
f
ir
s
t
id
en
tifie
s
lo
ca
l
m
in
im
a
(
wh
e
r
e
th
e
p
ix
el
in
te
n
s
ity
is
th
e
l
o
west
co
m
p
ar
ed
to
s
u
r
r
o
u
n
d
i
n
g
p
ix
els),
wh
ich
ac
t
as
m
ar
k
er
s
f
o
r
s
eg
m
en
tatio
n
.
T
h
e
f
lo
o
d
in
g
p
r
o
ce
s
s
is
co
n
tr
o
lle
d
b
y
r
is
in
g
th
r
esh
o
ld
s
,
an
d
t
h
e
s
eg
m
en
tatio
n
is
d
ef
in
ed
wh
e
n
two
r
e
g
io
n
s
m
ee
t.
An
ex
am
p
le
o
f
a
f
o
r
m
u
la
in
v
o
lv
ed
i
n
th
e
f
lo
o
d
in
g
p
r
o
ce
s
s
is
:
(
)
={
(
,
)
∈
∣
(
,
)
=m
in
(
)}
(
5
)
w
h
er
e
(
)
r
ep
r
esen
ts
th
e
wat
er
s
h
ed
r
eg
io
n
s
f
o
r
m
e
d
at
lo
ca
l
m
in
im
a
o
f
t
h
e
im
ag
e
i
n
ten
s
ity
f
u
n
ctio
n
(
,
).
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
8
4
5
-
2
8
5
6
2850
2)
I
m
ag
e
ex
tr
ac
tio
n
a.
GL
C
M
m
eth
o
d
T
h
e
GL
C
M
m
eth
o
d
is
a
wid
ely
-
u
s
ed
tech
n
iq
u
e
f
o
r
tex
tu
r
e
an
aly
s
is
in
d
ig
ital
im
ag
e
p
r
o
ce
s
s
in
g
,
p
ar
ticu
lar
ly
f
o
r
f
ea
tu
r
e
ex
tr
ac
tio
n
in
f
u
n
d
u
s
im
a
g
es
[
2
7
]
.
Fu
n
d
u
s
im
ag
es,
wh
ich
ca
p
tu
r
e
th
e
r
etin
a
an
d
its
s
tr
u
ctu
r
es,
ar
e
o
f
te
n
an
aly
ze
d
f
o
r
d
iag
n
o
s
in
g
m
ed
ical
co
n
d
itio
n
s
s
u
ch
as
DR
an
d
g
lau
co
m
a
[
2
8
]
.
T
e
x
tu
r
e
f
ea
tu
r
es
ex
tr
ac
ted
f
r
o
m
th
ese
im
ag
es
p
r
o
v
id
e
v
alu
ab
le
in
f
o
r
m
atio
n
ab
o
u
t
th
e
im
ag
e’
s
co
n
ten
t,
s
u
ch
as
th
e
p
atter
n
s
an
d
s
tr
u
ctu
r
es
o
f
b
lo
o
d
v
ess
els
o
r
lesi
o
n
s
.
T
h
e
GL
C
M
m
eth
o
d
wo
r
k
s
b
y
ca
lcu
la
tin
g
h
o
w
f
r
eq
u
en
tly
p
air
s
o
f
p
ix
el
in
ten
s
ity
v
alu
es
(
g
r
a
y
lev
els)
o
cc
u
r
in
a
s
p
ec
if
i
c
s
p
atial
r
elatio
n
s
h
ip
with
in
an
im
ag
e.
Fro
m
th
is
m
atr
ix
,
v
ar
i
o
u
s
s
tatis
tical
f
ea
t
u
r
es
lik
e
C
o
n
tr
ast,
C
o
r
r
elatio
n
,
E
n
er
g
y
,
a
n
d
H
o
m
o
g
e
n
eity
ca
n
b
e
c
o
m
p
u
te
d
to
ch
ar
ac
ter
ize
th
e
tex
tu
r
e
o
f
th
e
im
ag
e.
−
C
o
n
tr
ast:
Me
asu
r
es
th
e
in
ten
s
ity
d
if
f
er
en
ce
b
etwe
en
a
p
i
x
el
an
d
its
n
eig
h
b
o
r
o
v
er
th
e
en
tire
im
ag
e,
r
ef
lectin
g
th
e
d
eg
r
ee
o
f
lo
ca
l
v
ar
iatio
n
[
2
9
]
.
:
∑
(
−
)
2
(
,
)
(
6
)
wh
er
e
(
,
)
is
th
e
p
r
o
b
ab
ilit
y
o
f
o
cc
u
r
r
en
ce
o
f
p
ix
el
p
air
v
alu
es
an
d
.
−
C
o
r
r
elatio
n
:
R
ep
r
esen
ts
h
o
w
co
r
r
elate
d
a
p
ix
el
is
with
its
n
eig
h
b
o
r
s
,
in
d
icatin
g
th
e
lin
ea
r
d
ep
e
n
d
en
c
y
o
f
g
r
ay
lev
els
[
3
0
]
.
:
∑
(
−
)
(
−
)
(
,
)
,
(
7
)
wh
er
e
an
d
a
r
e
th
e
m
ea
n
s
an
d
s
tan
d
ar
d
d
ev
iatio
n
s
o
f
g
r
ay
lev
els.
−
E
n
er
g
y
:
R
ef
lects
th
e
u
n
if
o
r
m
i
ty
o
f
th
e
im
ag
e
tex
tu
r
e,
with
h
ig
h
er
en
er
g
y
in
d
icatin
g
a
m
o
r
e
u
n
if
o
r
m
o
r
s
m
o
o
th
tex
tu
r
e
[
3
1
]
.
:
∑
(
,
)
2
,
(
8
)
−
Ho
m
o
g
en
eity
:
Me
asu
r
es
th
e
cl
o
s
en
ess
o
f
th
e
d
is
tr
ib
u
tio
n
o
f
elem
en
ts
in
th
e
GL
C
M
to
t
h
e
d
iag
o
n
al,
wh
er
e
h
o
m
o
g
en
o
u
s
tex
t
u
r
es h
av
e
h
ig
h
v
alu
es
[
3
2
]
.
:
∑
(
,
)
1
+
|
,
|
,
(
9
)
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
3
.
1
.
Resul
t
o
f
inp
ut
im
a
g
e
(
f
un
du
s
im
a
g
e)
T
h
e
g
o
al
o
f
im
ag
e
in
p
u
t
is
to
p
r
o
v
id
e
v
is
u
al
d
ata
o
r
in
f
o
r
m
atio
n
f
o
r
f
u
r
t
h
er
p
r
o
ce
s
s
in
g
,
a
n
aly
s
is
,
o
r
p
r
esen
tatio
n
.
I
m
ag
e
ac
q
u
is
itio
n
d
u
r
i
n
g
a
m
ed
ical
ex
am
in
atio
n
,
s
u
ch
as
f
u
n
d
u
s
im
ag
in
g
f
o
r
DR
,
in
v
o
lv
es
em
p
lo
y
in
g
s
p
ec
ialis
t
eq
u
ip
m
e
n
t,
s
u
ch
as
a
f
u
n
d
u
s
ca
m
er
a
o
r
o
t
h
er
im
a
g
in
g
d
ev
ices,
to
t
ak
e
p
ictu
r
es
o
f
t
h
e
in
ter
n
al
s
tr
u
ctu
r
es
o
f
th
e
ey
e
.
Fu
n
d
u
s
p
ictu
r
es
in
p
atien
ts
with
DR
p
r
o
d
u
ce
in
t
r
icate
r
et
in
al
v
is
u
aliza
tio
n
s
.
Af
ter
war
d
s
,
b
ased
o
n
th
e
p
atie
n
t'
s
ey
e
co
n
d
itio
n
,
e
y
e
p
h
y
s
ician
s
u
s
e
th
ese
p
h
o
t
o
s
f
o
r
d
ia
g
n
o
s
is
,
f
o
llo
w
-
u
p
,
an
d
th
er
ap
y
p
lan
n
i
n
g
.
I
m
p
r
o
v
e
d
i
m
ag
e
q
u
ality
p
r
o
v
id
es
m
o
r
e
u
s
ef
u
l
d
ata
f
o
r
m
ed
ical
d
iag
n
o
s
is
an
d
th
er
a
p
y
.
W
e
u
tili
ze
d
5
0
f
u
n
d
u
s
p
ictu
r
es f
r
o
m
a
s
in
g
le
s
u
b
ject
f
o
r
o
u
r
s
tu
d
y
.
I
n
th
is
ar
ticle,
we
ju
s
t d
is
p
la
y
8
p
h
o
to
g
r
ap
h
s
as
an
ex
am
p
le.
T
h
e
in
p
u
t im
ag
e
is
d
is
p
lay
ed
in
T
ab
le
1.
T
ab
le
1
.
I
n
p
u
t
im
a
g
e
(
f
u
n
d
u
s
im
ag
e)
P
a
t
i
e
n
t
I
n
p
u
t
I
mag
e
(
F
u
n
d
u
s
I
mag
e
)
P
a
t
i
e
n
t
I
n
p
u
t
I
mag
e
(
F
u
n
d
u
s
I
mag
e
)
P
a
t
i
e
n
t
I
n
p
u
t
I
mag
e
(
F
u
n
d
u
s
I
mag
e
)
P
a
t
i
e
n
t
I
n
p
u
t
I
mag
e
(
F
u
n
d
u
s
I
mag
e
)
1
3
5
7
2
4
6
8
T
ab
le
1
s
h
o
ws
in
p
u
t
im
ag
es
o
f
f
u
n
d
u
s
s
ca
n
s
f
r
o
m
1
0
d
i
f
f
er
en
t
p
atien
ts
.
Fu
n
d
u
s
im
ag
es,
as s
ee
n
in
th
e
f
ig
u
r
e,
ar
e
cr
itical
in
an
aly
zin
g
th
e
r
etin
a,
o
p
tic
d
is
c,
an
d
b
lo
o
d
v
ess
els
f
o
r
ea
r
ly
d
etec
tio
n
o
f
r
etin
al
d
is
ea
s
es
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Dev
elo
p
men
t o
f wa
ters
h
ed
a
l
g
o
r
ith
m
fo
r
id
en
tifi
ca
tio
n
o
f
d
ia
b
etic
r
etin
o
p
a
th
y
b
a
s
ed
…
(
S
u
r
ma
ya
n
ti
)
2851
s
u
ch
as
DR
an
d
g
lau
c
o
m
a.
U
p
o
n
e
x
am
in
in
g
th
e
in
p
u
t
im
a
g
es,
n
o
tab
le
v
a
r
iatio
n
s
ca
n
b
e
o
b
s
er
v
ed
in
ter
m
s
o
f
clar
ity
,
b
r
ig
h
tn
ess
,
an
d
t
h
e
v
is
ib
ilit
y
o
f
im
p
o
r
tan
t
r
etin
al
s
tr
u
ctu
r
es lik
e
th
e
o
p
tic
d
is
c
an
d
b
lo
o
d
v
ess
els.
So
m
e
im
ag
es,
s
u
ch
as
th
o
s
e
f
r
o
m
Pa
tien
ts
1
an
d
6
,
d
is
p
lay
a
r
elativ
ely
clea
r
er
v
iew
o
f
th
e
o
p
tic
d
i
s
c
an
d
s
u
r
r
o
u
n
d
in
g
r
etin
al
v
ess
els,
wh
ile
o
th
e
r
s
,
s
u
ch
as
t
h
o
s
e
f
r
o
m
Patien
ts
3
an
d
8
,
s
h
o
w
s
ig
n
s
o
f
p
o
s
s
ib
le
ab
n
o
r
m
alities
lik
e
lesi
o
n
s
o
r
o
p
ac
ities
,
wh
ich
m
a
y
in
d
icate
t
h
e
p
r
esen
ce
o
f
d
is
e
ase.
T
h
e
in
p
u
t im
ag
es
v
ar
y
in
co
n
tr
ast,
with
s
o
m
e
im
ag
es a
p
p
ea
r
in
g
m
o
r
e
r
ed
d
is
h
o
r
y
ello
wis
h
,
wh
ich
co
u
ld
b
e
d
u
e
to
d
if
f
er
in
g
lig
h
tin
g
co
n
d
itio
n
s
o
r
v
ar
iatio
n
s
in
p
atien
t
ey
e
co
n
d
itio
n
s
.
T
h
i
s
v
ar
iab
ilit
y
in
im
ag
e
q
u
ality
p
r
esen
ts
a
ch
allen
g
e
f
o
r
i
m
ag
e
s
eg
m
en
tatio
n
an
d
f
u
r
th
er
a
n
aly
s
is
,
s
u
ch
as
lesi
o
n
d
etec
tio
n
o
r
tex
t
u
r
e
an
aly
s
is
,
w
h
ich
is
ess
en
tial
f
o
r
d
iag
n
o
s
in
g
r
etin
al
co
n
d
itio
n
s
.
I
n
s
u
m
m
ar
y
,
th
e
in
p
u
t
f
u
n
d
u
s
i
m
ag
es
h
ig
h
lig
h
t
d
iv
er
s
e
r
etin
al
co
n
d
itio
n
s
ac
r
o
s
s
d
if
f
er
en
t
p
at
ien
ts
.
So
m
e
im
ag
es
ap
p
ea
r
to
b
e
in
g
o
o
d
co
n
d
itio
n
f
o
r
an
aly
s
is
,
wh
ile
o
th
er
s
s
h
o
w
s
ig
n
s
o
f
p
o
ten
tial
ab
n
o
r
m
alities
,
in
d
icatin
g
th
e
n
ee
d
f
o
r
ad
v
an
ce
d
im
a
g
e
p
r
o
c
ess
in
g
tech
n
iq
u
es to
ex
tr
ac
t
v
a
lu
ab
le
f
ea
tu
r
es f
o
r
m
ed
ical
d
ia
g
n
o
s
is
.
3
.
2
.
Resul
t
o
f
prepro
ce
s
s
ing
B
ef
o
r
e
th
e
DR
f
u
n
d
u
s
im
ag
e
p
r
o
ce
s
s
is
ca
r
r
ied
o
u
t,
wh
ich
will
b
e
u
s
ed
in
th
e
p
r
o
ce
s
s
in
g
s
tag
e,
th
e
o
r
ig
in
al
im
ag
e
o
r
f
u
n
d
u
s
im
ag
e
is
f
ir
s
t
co
n
v
er
ted
in
to
a
g
r
ay
s
ca
le
im
ag
e,
th
e
n
th
e
im
ag
e
is
Har
d
No
is
e
R
ed
u
ctio
n
,
with
th
e
Ga
u
s
s
ian
Fil
ter
,
th
en
So
f
t
No
is
e
R
ed
u
c
tio
n
m
eth
o
d
with
th
e
Me
d
ian
Fil
ter
m
eth
o
d
,
an
d
T
h
r
esh
o
ld
in
g
with
th
e
Ots
u
'
s
T
h
r
esh
o
ld
in
g
m
eth
o
d
.
B
elo
w
t
h
e
r
esu
lt o
f
p
r
ep
r
o
ce
s
s
in
g
is
s
h
o
wn
in
T
a
b
le
2
.
T
ab
le
2
.
Pre
p
r
o
ce
s
s
in
g
r
esu
lt
P
a
t
i
e
n
t
I
n
p
u
t
I
mag
e
(
F
u
n
d
u
s
I
mag
e
)
R
G
B
t
o
G
r
a
y
sc
a
l
e
H
a
r
d
N
o
i
s
e
R
e
d
u
c
t
i
o
n
S
o
f
t
N
o
i
s
e
R
e
d
u
c
t
i
o
n
Th
r
e
s
h
o
l
d
i
n
g
1
2
3
4
5
6
7
8
T
ab
le
2
d
is
p
lay
s
th
e
r
esu
lts
o
f
th
e
p
r
ep
r
o
ce
s
s
in
g
s
tep
s
ap
p
li
ed
to
f
u
n
d
u
s
im
ag
es
f
r
o
m
eig
h
t
d
if
f
e
r
en
t
p
atien
ts
.
T
h
is
p
r
ep
r
o
ce
s
s
in
g
p
ip
elin
e
in
clu
d
es
th
e
co
n
v
er
s
io
n
f
r
o
m
R
GB
to
g
r
ay
s
ca
le,
f
o
llo
wed
b
y
n
o
is
e
r
ed
u
ctio
n
(
b
o
th
h
ar
d
an
d
s
o
f
t)
,
an
d
f
in
ally
,
a
th
r
esh
o
ld
i
n
g
o
p
er
atio
n
.
T
h
e
R
GB
to
g
r
ay
s
ca
le
co
n
v
er
s
io
n
ef
f
ec
tiv
ely
r
ed
u
ce
s
th
e
co
m
p
le
x
ity
o
f
th
e
im
ag
e
b
y
s
im
p
lify
i
n
g
th
e
co
lo
r
in
f
o
r
m
atio
n
in
to
i
n
ten
s
ity
lev
els.
T
h
is
s
tep
is
cr
u
cial
f
o
r
f
u
r
th
e
r
im
a
g
e
p
r
o
ce
s
s
in
g
task
s
as
it
f
o
cu
s
es
o
n
p
ix
el
in
ten
s
ity
v
ar
iatio
n
s
r
ath
er
th
a
n
co
lo
r
v
ar
iatio
n
s
,
wh
ich
ar
e
m
o
r
e
r
elev
an
t
in
m
ed
ical
im
a
g
in
g
f
o
r
an
aly
zin
g
r
etin
al
s
tr
u
ct
u
r
es.
T
h
e
g
r
ay
s
ca
le
im
ag
es
in
th
e
tab
le
r
etain
th
e
o
v
er
all
s
t
r
u
ctu
r
e
o
f
th
e
r
etin
a,
o
p
tic
d
is
c,
an
d
v
is
ib
le
ab
n
o
r
m
alities
,
p
r
o
v
id
in
g
a
f
o
u
n
d
atio
n
f
o
r
s
u
b
s
eq
u
e
n
t n
o
is
e
r
em
o
v
al
an
d
s
eg
m
en
tatio
n
s
tep
s
.
T
h
e
p
r
e
p
r
o
ce
s
s
in
g
p
ip
elin
e
in
c
lu
d
es
b
o
th
h
ar
d
n
o
is
e
r
ed
u
ctio
n
an
d
s
o
f
t
n
o
is
e
r
ed
u
ctio
n
to
i
m
p
r
o
v
e
th
e
im
ag
e
q
u
ality
a
n
d
s
u
p
p
r
ess
u
n
wan
ted
n
o
is
e
th
at
c
o
u
ld
af
f
e
ct
s
eg
m
en
tatio
n
ac
cu
r
ac
y
.
Ha
r
d
n
o
is
e
r
e
d
u
ctio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
8
4
5
-
2
8
5
6
2852
r
em
o
v
es h
ig
h
-
f
r
e
q
u
en
c
y
n
o
is
e
co
m
p
o
n
en
ts
,
wh
ich
m
a
y
r
ep
r
esen
t sm
all
an
d
ir
r
elev
an
t stru
ctu
r
es o
r
ar
tifa
cts in
th
e
im
ag
e.
As
s
h
o
wn
i
n
th
e
ta
b
le,
h
ar
d
n
o
is
e
r
ed
u
ctio
n
h
as
s
ig
n
if
ican
tly
r
ed
u
ce
d
th
ese
a
r
tifa
cts
in
th
e
im
ag
es,
th
o
u
g
h
s
o
m
e
im
a
g
e
s
tr
u
ctu
r
es
h
av
e
b
ee
n
o
v
er
-
s
im
p
lifie
d
,
lo
s
in
g
f
in
er
d
etails.
So
f
t
n
o
is
e
r
ed
u
ctio
n
,
o
n
t
h
e
o
th
er
h
an
d
,
r
etain
s
m
o
r
e
o
f
th
e
im
a
g
e'
s
s
tr
u
ctu
r
al
in
f
o
r
m
atio
n
wh
il
e
s
till
s
u
p
p
r
ess
in
g
n
o
is
e.
T
h
e
r
esu
lts
d
em
o
n
s
tr
ate
th
at
th
e
im
ag
es
m
ai
n
tain
m
o
r
e
f
in
e
d
etails
co
m
p
ar
ed
to
h
ar
d
n
o
is
e
r
e
d
u
ctio
n
,
wh
ile
s
till
s
m
o
o
th
in
g
o
u
t
less
r
elev
an
t
ar
ea
s
.
T
h
is
b
alan
ce
b
etwe
en
n
o
is
e
s
u
p
p
r
ess
io
n
an
d
s
tr
u
ctu
r
e
r
eten
tio
n
is
cr
u
cial
f
o
r
en
h
a
n
cin
g
s
ig
n
if
ican
t
f
ea
tu
r
es,
s
u
ch
as
lesi
o
n
s
o
r
v
ess
el
p
atter
n
s
,
in
f
u
n
d
u
s
im
ag
es.
T
h
r
esh
o
ld
i
n
g
is
th
e
f
in
al
s
tep
in
th
e
p
r
ep
r
o
ce
s
s
in
g
p
ip
elin
e,
co
n
v
e
r
tin
g
th
e
g
r
a
y
s
ca
le
im
ag
es
in
t
o
b
in
ar
y
im
a
g
es.
I
n
th
is
s
tep
,
ar
ea
s
o
f
in
te
r
est
ar
e
h
ig
h
lig
h
ted
in
wh
ite,
wh
ile
th
e
b
ac
k
g
r
o
u
n
d
is
r
ep
r
esen
ted
a
s
b
lack
.
T
h
is
tech
n
iq
u
e
ef
f
ec
t
iv
ely
s
eg
m
en
ts
o
u
t
im
p
o
r
tan
t
r
eg
io
n
s
s
u
c
h
as
lesi
o
n
s
o
r
o
th
er
ab
n
o
r
m
al
a
r
ea
s
f
r
o
m
th
e
b
ac
k
g
r
o
u
n
d
.
I
n
th
e
tab
le,
th
e
th
r
esh
o
ld
in
g
s
tep
h
as su
cc
ess
f
u
lly
is
o
lated
k
ey
f
ea
t
u
r
es f
r
o
m
th
e
b
ac
k
g
r
o
u
n
d
,
m
ak
in
g
it
ea
s
ier
to
i
d
en
tif
y
cr
itical
r
e
g
io
n
s
o
f
in
ter
est
in
th
e
im
ag
es.
T
h
e
r
e
s
u
lt
is
a
clea
n
s
eg
m
e
n
tatio
n
o
f
s
i
g
n
if
ican
t
a
b
n
o
r
m
alities
,
p
o
ten
tially
en
h
an
ci
n
g
th
e
ac
cu
r
ac
y
o
f
f
u
r
th
er
an
aly
s
i
s
an
d
f
ea
tu
r
e
e
x
tr
ac
tio
n
p
r
o
ce
s
s
es.
T
h
e
p
r
e
p
r
o
ce
s
s
in
g
r
esu
lts
d
e
m
o
n
s
tr
ate
an
e
f
f
ec
tiv
e
p
ip
elin
e
f
o
r
p
r
e
p
ar
in
g
f
u
n
d
u
s
im
a
g
es
f
o
r
f
u
r
t
h
er
m
ed
ical
an
aly
s
is
an
d
s
eg
m
en
tatio
n
task
s
.
T
h
e
g
r
ay
s
ca
le
co
n
v
er
s
io
n
s
im
p
lifie
s
th
e
im
ag
e,
n
o
is
e
r
ed
u
ctio
n
en
h
an
ce
s
im
ag
e
q
u
ality
b
y
s
u
p
p
r
ess
in
g
ir
r
ele
v
an
t
in
f
o
r
m
a
tio
n
,
an
d
t
h
r
esh
o
ld
in
g
is
o
lates
k
ey
f
ea
tu
r
es
f
o
r
m
ed
ical
d
iag
n
o
s
is
.
W
h
ile
h
ar
d
n
o
is
e
r
ed
u
ctio
n
m
ig
h
t
o
v
er
-
s
im
p
lify
s
o
m
e
im
ag
e
d
etails,
s
o
f
t
n
o
is
e
r
ed
u
ctio
n
s
tr
ik
es
a
g
o
o
d
b
alan
ce
b
y
r
e
tain
in
g
ess
en
tial
s
tr
u
ctu
r
al
in
f
o
r
m
atio
n
.
T
h
e
s
eg
m
en
tatio
n
ac
h
iev
ed
th
r
o
u
g
h
th
r
esh
o
ld
in
g
p
r
esen
ts
a
clea
r
b
in
ar
y
r
e
p
r
esen
tatio
n
o
f
th
e
r
eg
i
o
n
s
o
f
in
ter
est,
wh
ich
is
u
s
ef
u
l
f
o
r
f
u
r
th
e
r
an
aly
s
is
s
u
ch
as
th
e
d
etec
tio
n
o
f
r
etin
al
d
is
ea
s
es.
T
h
is
p
r
ep
r
o
ce
s
s
in
g
p
ip
elin
e,
as
s
h
o
wn
in
th
e
tab
le,
ef
f
ec
tiv
ely
en
h
an
ce
s
th
e
f
u
n
d
u
s
im
ag
es,
m
a
k
in
g
th
e
m
s
u
itab
le
f
o
r
th
e
n
e
x
t
s
tag
es
o
f
f
ea
tu
r
e
ex
tr
ac
tio
n
an
d
class
if
icatio
n
,
cr
u
cial
f
o
r
ac
cu
r
ate
d
iag
n
o
s
is
in
o
p
h
th
al
m
o
lo
g
y
.
3.
3
.
Resul
t
o
f
pro
ce
s
s
ing
T
h
e
r
esu
lts
p
r
esen
ted
in
T
ab
le
3
illu
s
tr
ate
th
e
im
ag
e
s
eg
m
en
tatio
n
an
d
f
ea
tu
r
e
ex
tr
ac
tio
n
p
r
o
ce
s
s
es
ap
p
lied
to
f
u
n
d
u
s
im
ag
es
f
r
o
m
eig
h
t
p
atien
ts
.
T
h
e
tab
le
is
d
iv
id
ed
in
to
two
m
ain
s
ec
tio
n
s
:
I
m
ag
e
Seg
m
en
tatio
n
an
d
I
m
ag
e
E
x
tr
ac
tio
n
u
s
in
g
t
h
e
GL
C
M
Me
th
o
d
.
T
h
e
s
eg
m
en
tatio
n
p
r
o
ce
s
s
in
clu
d
es
th
r
esh
o
ld
in
g
,
m
ar
k
e
r
f
o
r
eg
r
o
u
n
d
d
etec
tio
n
,
a
n
d
th
e
ap
p
licatio
n
o
f
th
e
W
ater
s
h
ed
Alg
o
r
ith
m
,
wh
ich
h
i
g
h
lig
h
ts
d
if
f
er
en
t
r
e
g
io
n
s
o
f
in
ter
est
in
th
e
f
u
n
d
u
s
im
a
g
es.
Ad
d
itio
n
ally
,
t
h
e
GL
C
M
m
eth
o
d
is
u
s
ed
t
o
ex
tr
ac
t
te
x
tu
r
e
f
ea
tu
r
es,
in
clu
d
i
n
g
co
n
tr
ast,
co
r
r
elatio
n
,
en
er
g
y
,
an
d
h
o
m
o
g
en
eity
,
p
r
o
v
id
in
g
q
u
an
tita
tiv
e
in
s
ig
h
ts
i
n
to
th
e
t
ex
tu
r
e
p
r
o
p
er
ties
o
f
th
e
s
eg
m
en
ted
r
e
g
io
n
s
.
T
h
ese
f
ea
tu
r
es
ar
e
cr
u
cial
f
o
r
d
iag
n
o
s
in
g
an
d
d
if
f
er
e
n
tiatin
g
r
etin
al
co
n
d
itio
n
s
b
ased
o
n
v
ar
iatio
n
s
in
th
e
f
u
n
d
u
s
im
ag
es.
B
elo
w
th
e
r
esu
lt o
f
p
r
ep
r
o
ce
s
s
in
g
is
s
h
o
wn
in
T
ab
le
3
.
T
h
e
im
ag
e
s
eg
m
en
tatio
n
an
d
f
ea
tu
r
e
ex
tr
ac
tio
n
p
ip
eli
n
e
p
r
esen
ted
in
T
ab
le
3
d
em
o
n
s
tr
a
te
a
r
o
b
u
s
t
ap
p
r
o
ac
h
to
a
n
a
l
y
z
i
n
g
f
u
n
d
u
s
i
m
a
g
e
s
.
T
h
e
W
at
e
r
s
h
e
d
A
l
g
o
r
i
th
m
s
u
c
c
es
s
f
u
ll
y
s
e
g
m
e
n
t
s
r
e
g
io
n
s
o
f
i
n
t
e
r
es
t
,
w
h
i
le
t
h
e
G
L
C
M
m
e
t
h
o
d
p
r
o
v
i
d
e
s
c
r
i
t
i
c
al
t
e
x
t
u
r
e
f
e
at
u
r
e
s
t
h
at
a
r
e
es
s
e
n
t
i
a
l
f
o
r
d
i
a
g
n
o
s
i
n
g
r
e
t
i
n
a
l
a
b
n
o
r
m
a
l
i
t
i
es
.
E
a
c
h
p
a
t
i
e
n
t'
s
f
u
n
d
u
s
i
m
a
g
e
r
e
v
e
als
d
is
t
i
n
ct
c
h
a
r
a
ct
e
r
is
t
i
cs
,
w
i
t
h
v
a
r
i
a
t
i
o
n
s
i
n
c
o
n
t
r
a
s
t
,
c
o
r
r
e
la
t
i
o
n
,
e
n
e
r
g
y
,
a
n
d
h
o
m
o
g
e
n
e
i
t
y
.
H
i
g
h
e
r
c
o
n
t
r
as
t
v
a
l
u
e
s
a
r
e
t
y
p
i
ca
l
l
y
a
s
s
o
c
i
at
e
d
w
i
t
h
m
o
r
e
p
r
o
n
o
u
n
c
e
d
l
es
i
o
n
s
o
r
a
b
n
o
r
m
a
li
t
i
es
,
w
h
i
l
e
h
i
g
h
e
r
h
o
m
o
g
e
n
e
i
t
y
a
n
d
e
n
e
r
g
y
v
a
l
u
e
s
s
u
g
g
e
s
t
s
m
o
o
t
h
er
a
n
d
m
o
r
e
u
n
i
f
o
r
m
r
e
t
i
n
a
l
s
t
r
u
c
t
u
r
e
s
.
T
h
is
a
n
al
y
s
is
h
i
g
h
l
i
g
h
t
s
h
o
w
a
c
o
m
b
i
n
at
i
o
n
o
f
i
m
a
g
e
s
e
g
m
e
n
t
at
i
o
n
t
e
c
h
n
i
q
u
e
s
a
n
d
t
e
x
t
u
r
e
a
n
a
l
y
s
is
c
a
n
e
f
f
e
c
t
i
v
e
l
y
is
o
l
at
e
a
n
d
q
u
a
n
t
i
f
y
a
b
n
o
r
m
a
l
i
t
ie
s
i
n
f
u
n
d
u
s
i
m
a
g
es
,
p
r
o
v
i
d
i
n
g
v
al
u
a
b
l
e
i
n
s
i
g
h
ts
f
o
r
m
e
d
i
c
al
p
r
o
f
e
s
s
io
n
a
l
s
i
n
d
i
a
g
n
o
s
i
n
g
c
o
n
d
i
t
i
o
n
s
s
u
c
h
a
s
DR
o
r
m
ac
u
l
a
r
d
e
g
e
n
e
r
a
t
i
o
n
.
T
h
e
r
es
u
l
ts
e
m
p
h
a
s
i
z
e
t
h
e
i
m
p
o
r
t
a
n
c
e
o
f
p
r
e
p
r
o
c
e
s
s
i
n
g
a
n
d
s
e
g
m
e
n
t
a
ti
o
n
i
n
e
n
h
a
n
c
i
n
g
t
h
e
c
l
a
r
i
t
y
an
d
p
r
ec
is
i
o
n
o
f
m
ed
ic
al
im
ag
e
an
aly
s
is
.
T
h
e
ac
cu
r
ac
y
o
f
th
e
im
ag
e
p
r
o
ce
s
s
in
g
an
d
f
ea
tu
r
e
ex
tr
a
ctio
n
m
eth
o
d
s
ap
p
lied
i
n
th
is
r
esear
ch
d
em
o
n
s
tr
ates
a
s
o
lid
p
er
f
o
r
m
a
n
ce
.
Fro
m
th
e
5
0
f
u
n
d
u
s
im
a
g
e
s
u
s
ed
as
in
p
u
t
d
ata,
th
e
s
eg
m
e
n
tatio
n
an
d
an
aly
s
is
tech
n
iq
u
es
wer
e
ab
le
to
ac
cu
r
ately
id
en
tify
a
b
n
o
r
m
ali
ties
in
4
6
im
ag
es,
r
esu
ltin
g
in
a
n
ac
cu
r
ac
y
r
ate
o
f
9
2
%.
T
h
is
h
ig
h
ac
cu
r
ac
y
co
n
f
ir
m
s
th
at
th
e
co
m
b
in
atio
n
o
f
th
e
W
ater
s
h
ed
Alg
o
r
ith
m
f
o
r
im
ag
e
s
eg
m
en
tatio
n
an
d
th
e
GL
C
M
m
eth
o
d
f
o
r
tex
tu
r
e
f
ea
tu
r
e
e
x
tr
ac
tio
n
is
ef
f
ec
tiv
e
in
d
etec
tin
g
k
e
y
p
ath
o
lo
g
ical
f
ea
tu
r
es,
s
u
ch
as
lesi
o
n
s
an
d
r
etin
al
ab
n
o
r
m
alities
,
with
m
in
im
al
m
is
class
if
icatio
n
.
T
h
e
r
o
b
u
s
t
s
eg
m
en
tatio
n
,
alo
n
g
with
th
e
r
eliab
le
tex
tu
r
e
an
aly
s
is
,
h
ig
h
lig
h
ts
th
e
ef
f
icac
y
o
f
th
is
ap
p
r
o
ac
h
in
p
r
o
v
id
in
g
c
o
n
s
is
ten
t
an
d
ac
cu
r
ate
d
iag
n
o
s
tic
s
u
p
p
o
r
t,
m
a
k
in
g
it a
v
alu
ab
le
t
o
o
l f
o
r
r
etin
al
d
is
ea
s
e
an
aly
s
is
.
Ou
r
r
esu
lts
s
u
b
s
tan
tiate
s
ig
n
if
i
ca
n
t
ad
v
a
n
ce
m
en
ts
o
v
er
p
r
io
r
s
tu
d
ies.
T
h
e
in
te
g
r
atio
n
o
f
th
e
W
ater
s
h
ed
Alg
o
r
ith
m
e
n
ab
led
m
o
r
e
p
r
ec
is
e
s
eg
m
en
tatio
n
co
m
p
ar
e
d
to
Sam
a
n
et
a
l.
[
1
2
]
wh
i
ch
r
elied
o
n
b
asic
m
o
r
p
h
o
lo
g
ical
o
p
e
r
atio
n
s
.
Ou
r
s
y
s
tem
ac
h
iev
ed
an
ac
cu
r
ac
y
o
f
9
2
%,
o
u
tp
e
r
f
o
r
m
in
g
th
e
8
7
%
ac
cu
r
ac
y
o
f
Sar
k
i
et
a
l.
[
1
3
]
p
ar
ticu
lar
ly
in
h
an
d
l
in
g
im
ag
es
with
n
o
is
e
an
d
ir
r
e
g
u
lar
ities
.
Fu
r
th
er
m
o
r
e,
o
u
r
G
L
C
M
-
b
ased
tex
tu
r
e
ex
tr
ac
tio
n
d
em
o
n
s
tr
ated
en
h
an
ce
d
d
if
f
er
en
tiatio
n
o
f
r
eti
n
al
f
ea
tu
r
es,
with
m
etr
ics
s
u
ch
as
c
o
n
tr
ast
an
d
co
r
r
elatio
n
s
h
o
win
g
co
n
s
is
ten
t
im
p
r
o
v
em
en
ts
ac
r
o
s
s
d
a
tasets
.
Fo
r
in
s
tan
ce
,
th
e
en
er
g
y
m
etr
ic
in
o
u
r
s
tu
d
y
ac
h
iev
ed
0
.
9
6
4
0
f
o
r
u
n
if
o
r
m
t
ex
tu
r
es,
s
u
r
p
ass
in
g
p
r
e
v
io
u
s
r
esu
lts
th
at
d
id
n
o
t
em
p
h
asize
tex
tu
r
e
u
n
if
o
r
m
ity
.
T
h
ese
f
in
d
in
g
s
u
n
d
er
s
co
r
e
o
u
r
f
r
am
ewo
r
k
'
s
r
o
b
u
s
tn
ess
,
s
ca
la
b
ilit
y
,
an
d
p
o
ten
tial
as
a
d
iag
n
o
s
t
ic
to
o
l
in
clin
ical
s
ettin
g
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Dev
elo
p
men
t o
f wa
ters
h
ed
a
l
g
o
r
ith
m
fo
r
id
en
tifi
ca
tio
n
o
f
d
ia
b
etic
r
etin
o
p
a
th
y
b
a
s
ed
…
(
S
u
r
ma
ya
n
ti
)
2853
Pre
v
io
u
s
s
tu
d
ies h
a
v
e
e
x
p
lo
r
e
d
v
ar
i
o
u
s
ap
p
r
o
ac
h
es to
d
etec
t
in
g
DR
,
b
u
t th
e
y
s
till
f
ac
ed
li
m
itatio
n
s
in
ce
r
tain
asp
ec
ts
.
Fo
r
in
s
tan
ce
,
S
am
an
et
a
l.
[
1
2
]
u
s
ed
b
asic
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
s
u
ch
as
d
ilatio
n
an
d
e
r
o
s
io
n
to
d
etec
t
an
d
class
if
y
DR
,
b
u
t
th
is
ap
p
r
o
ac
h
was
less
ef
f
ec
tiv
e
in
h
an
d
lin
g
o
v
er
lap
p
in
g
f
e
atu
r
es
an
d
n
o
is
e
in
im
ag
es.
On
th
e
o
th
er
h
a
n
d
,
Sa
r
k
i
et
a
l.
[
1
3
]
d
ev
elo
p
ed
a
f
r
a
m
ewo
r
k
b
ased
o
n
C
NNs
co
m
b
in
ed
with
tr
ad
itio
n
al
p
r
ep
r
o
ce
s
s
in
g
tech
n
iq
u
es.
Alth
o
u
g
h
ac
h
iev
in
g
h
ig
h
ac
cu
r
ac
y
,
th
is
m
eth
o
d
r
eq
u
ir
ed
s
ig
n
if
i
ca
n
t
co
m
p
u
tatio
n
al
r
eso
u
r
ce
s
an
d
was
less
f
lex
ib
le
f
o
r
d
i
v
er
s
e
d
atasets
.
R
esear
ch
b
y
L
o
n
g
et
a
l.
[
1
4
]
em
p
h
asized
m
ac
h
in
e
lear
n
i
n
g
-
b
ased
m
eth
o
d
s
f
o
r
s
eg
m
en
tati
o
n
an
d
class
if
icatio
n
;
h
o
we
v
e
r
,
th
e
i
n
teg
r
atio
n
o
f
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
s
f
o
r
im
p
r
o
v
e
d
s
eg
m
e
n
tatio
n
r
em
ai
n
ed
m
i
n
im
al.
Fu
r
t
h
er
m
o
r
e,
te
x
tu
r
e
a
n
aly
s
is
h
as
o
f
te
n
b
ee
n
o
v
er
l
o
o
k
e
d
as
a
n
im
p
o
r
tan
t
d
ia
g
n
o
s
t
ic
p
ar
am
et
er
.
B
y
in
teg
r
atin
g
te
x
tu
r
e
f
ea
tu
r
es
th
r
o
u
g
h
GL
C
M,
th
is
s
t
u
d
y
h
i
g
h
lig
h
ts
th
e
im
p
o
r
tan
ce
o
f
tex
t
u
r
e
u
n
if
o
r
m
ity
an
d
d
i
f
f
er
en
tiatio
n
in
d
ia
g
n
o
s
in
g
r
etin
al
a
b
n
o
r
m
alities
.
T
h
is
r
esear
ch
b
u
ild
s
a
n
ew
f
o
u
n
d
atio
n
f
o
r
m
o
r
e
p
r
e
cise a
n
d
ef
f
icien
t a
u
t
o
m
ated
D
R
d
iag
n
o
s
is
.
T
ab
le
3
.
Pro
ce
s
s
in
g
r
esu
lt
I
mag
e
S
e
g
me
n
t
a
t
i
o
n
I
mag
e
e
x
t
r
a
c
t
i
o
n
(
G
LC
M
M
e
t
h
o
d
)
P
a
t
i
e
n
t
Th
r
e
s
h
o
l
d
i
n
g
M
a
r
k
e
r
F
o
r
e
g
r
o
u
n
d
W
a
t
e
r
s
h
e
d
A
l
g
o
r
i
t
h
m
C
o
n
t
r
a
st
C
o
r
r
e
l
a
t
i
o
n
En
e
r
g
y
H
o
mo
g
e
n
e
i
t
y
1
0
.
2
1
7
1
0
.
9
3
7
3
0
.
9
2
4
8
0
.
9
9
6
1
2
0
.
1
6
1
1
0
.
9
5
9
6
0
.
9
1
5
2
0
.
9
9
7
1
3
0
.
1
0
3
4
0
.
9
3
7
7
0
.
9
6
4
0
0
.
9
9
8
1
4
0
.
0
8
0
6
0
.
9
5
4
6
0
.
9
6
2
1
0
.
9
9
8
5
5
0
.
3
3
9
0
0
.
9
3
2
8
0
.
8
8
9
9
0
.
9
9
3
9
6
0
.
1
6
8
3
0
.
9
7
7
2
0
.
8
4
5
5
0
.
9
9
7
0
7
0
.
1
0
6
1
0
.
9
6
5
6
0
.
9
3
4
8
0
.
9
9
8
1
8
0
.
0
5
2
6
0
.
9
9
1
5
0
.
8
7
2
4
0
.
9
9
8
6
4.
CO
NCLU
SI
O
N
T
h
is
r
esear
ch
d
em
o
n
s
tr
ates
th
e
ef
f
ec
tiv
e
n
ess
o
f
co
m
b
i
n
in
g
th
e
W
ater
s
h
ed
Alg
o
r
ith
m
f
o
r
p
r
e
cise
im
ag
e
s
eg
m
en
tatio
n
an
d
th
e
GL
C
M
f
o
r
tex
tu
r
e
f
ea
tu
r
e
ex
tr
ac
tio
n
in
d
etec
tin
g
DR
f
r
o
m
f
u
n
d
u
s
im
ag
es.
W
ith
an
ac
cu
r
ac
y
o
f
9
2
%,
o
u
r
ap
p
r
o
ac
h
s
ig
n
if
ican
tly
ad
v
an
ce
s
au
to
m
ated
DR
d
etec
tio
n
b
y
ad
d
r
e
s
s
in
g
lim
itatio
n
s
in
s
eg
m
en
tatio
n
p
r
ec
is
io
n
an
d
te
x
tu
r
e
an
aly
s
is
f
o
u
n
d
i
n
p
r
e
v
io
u
s
m
eth
o
d
o
l
o
g
ies.
T
h
ese
f
in
d
i
n
g
s
co
n
tr
ib
u
te
to
th
e
f
ield
b
y
o
f
f
e
r
in
g
a
s
ca
lab
le,
ef
f
icien
t,
an
d
r
eliab
le
d
iag
n
o
s
tic
to
o
l
th
at
ca
n
en
h
an
ce
ea
r
ly
d
etec
tio
n
an
d
tim
ely
in
ter
v
en
tio
n
,
u
ltima
tely
r
e
d
u
cin
g
th
e
r
is
k
o
f
b
lin
d
n
ess
in
d
iab
etic
p
atien
ts
.
Desp
ite
th
ese
ac
h
iev
em
en
ts
,
q
u
esti
o
n
s
r
em
ain
r
eg
ar
d
in
g
th
e
ap
p
licatio
n
o
f
th
is
m
eth
o
d
t
o
lar
g
er
,
m
o
r
e
d
iv
er
s
e
d
ataset
s
an
d
its
in
teg
r
atio
n
with
r
ea
l
-
tim
e
d
iag
n
o
s
tic
s
y
s
tem
s
in
clin
ical
s
ettin
g
s
.
Fu
tu
r
e
wo
r
k
s
h
o
u
ld
f
o
cu
s
o
n
en
h
a
n
cin
g
th
e
r
o
b
u
s
tn
ess
o
f
th
e
alg
o
r
ith
m
f
o
r
h
an
d
lin
g
v
ar
iatio
n
s
in
im
ag
e
q
u
ality
an
d
ex
p
an
d
in
g
th
e
m
eth
o
d
to
in
clu
d
e
m
u
lti
-
class
class
if
icatio
n
f
o
r
d
if
f
er
e
n
t
s
tag
es
o
f
DR
s
ev
er
ity
.
A
d
d
itio
n
ally
,
i
n
teg
r
atin
g
th
is
a
p
p
r
o
ac
h
with
a
d
v
an
ce
d
m
ac
h
in
e
lear
n
i
n
g
tech
n
iq
u
es
,
s
u
ch
as
d
ee
p
lear
n
in
g
-
b
as
ed
m
o
d
els,
ca
n
f
u
r
th
er
im
p
r
o
v
e
ac
c
u
r
ac
y
a
n
d
ad
ap
tab
ilit
y
.
E
x
p
l
o
r
in
g
its
ap
p
licatio
n
to
o
th
er
r
etin
al
d
is
ea
s
es,
s
u
ch
as
m
ac
u
lar
d
eg
en
er
atio
n
o
r
g
lau
c
o
m
a,
an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
8
4
5
-
2
8
5
6
2854
im
p
lem
en
tin
g
it
with
in
p
o
r
tab
le
d
iag
n
o
s
tic
d
ev
ices
o
r
telem
ed
icin
e
p
latf
o
r
m
s
,
co
u
l
d
s
ig
n
if
ican
tly
b
r
o
a
d
en
its
u
tili
ty
.
B
y
ad
d
r
ess
in
g
th
ese
ar
ea
s
,
th
is
r
esear
ch
lay
s
a
f
o
u
n
d
atio
n
f
o
r
d
ev
el
o
p
in
g
c
o
m
p
r
e
h
e
n
s
iv
e
an
d
im
p
ac
tf
u
l
au
to
m
ated
d
ia
g
n
o
s
tic
s
y
s
tem
s
in
o
p
h
t
h
alm
o
lo
g
y
,
s
u
p
p
o
r
tin
g
b
r
o
ad
e
r
ac
ce
s
s
to
h
ig
h
-
q
u
ality
ey
e
ca
r
e.
ACK
NO
WL
E
DG
M
E
N
T
S
T
h
e
au
th
o
r
s
wo
u
l
d
lik
e
to
e
x
p
r
ess
th
eir
s
in
ce
r
e
g
r
atitu
d
e
t
o
Pro
f
.
Dr
.
Sar
jo
n
Def
it,
S.Ko
m
.
,
M.
Ko
m
.
as
R
ec
to
r
o
f
Un
iv
er
s
itas
Pu
tr
a
I
n
d
o
n
esia
YPTK
Pad
an
g
an
d
Ass
o
c.
Pro
f
.
Dr
.
Mu
h
am
m
ad
R
id
wan
,
S.E
.
,
M.
M.
,
as
th
e
f
o
u
n
d
atio
n
'
s
p
r
esid
en
t
o
f
Yay
asan
Per
g
u
r
u
an
T
i
n
g
g
i
Ko
m
p
u
ter
(
YPTK
)
Pad
a
n
g
f
o
r
p
r
o
v
id
in
g
th
e
n
ec
ess
ar
y
f
ac
ilit
ies an
d
s
u
p
p
o
r
t th
r
o
u
g
h
o
u
t t
h
is
r
esear
ch
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
is
r
esear
ch
was
n
o
t
s
u
p
p
o
r
t
ed
b
y
an
y
g
r
an
ts
f
r
o
m
f
u
n
d
in
g
b
o
d
ies
in
th
e
p
u
b
lic,
p
r
iv
ate,
o
r
n
o
t
-
f
o
r
-
p
r
o
f
it secto
r
s
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Su
r
m
ay
an
ti
✓
✓
✓
✓
✓
✓
Su
m
ijan
✓
✓
✓
✓
✓
✓
Saif
u
l Bu
k
h
o
r
i
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
I
NF
O
RM
E
D
CO
NS
E
N
T
W
e
h
av
e
o
b
tain
ed
in
f
o
r
m
ed
c
o
n
s
en
t f
r
o
m
all
in
d
iv
id
u
als in
c
lu
d
ed
in
t
h
is
s
tu
d
y
.
E
T
H
I
CAL AP
P
RO
V
AL
T
h
e
r
esear
ch
r
elate
d
to
an
i
m
a
l
u
s
e
h
as
b
ee
n
co
m
p
lied
with
all
th
e
r
elev
an
t
n
atio
n
al
r
eg
u
l
atio
n
s
an
d
in
s
titu
tio
n
al
p
o
licies f
o
r
th
e
ca
r
e
an
d
u
s
e
o
f
an
im
als.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
in
d
i
n
g
s
o
f
th
is
s
tu
d
y
ar
e
av
ailab
le
f
r
o
m
th
e
c
o
r
r
esp
o
n
d
in
g
au
t
h
o
r
,
[
S],
u
p
o
n
r
ea
s
o
n
ab
le
r
eq
u
est.
RE
F
E
R
E
NC
E
S
[
1
]
Y
.
Le
e
e
t
a
l
.
,
“
P
e
r
f
o
r
ma
n
c
e
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
b
a
r
i
a
t
r
i
c
su
r
g
e
r
y
:
c
o
mp
a
r
a
t
i
v
e
a
n
a
l
y
s
i
s o
f
C
h
a
t
G
P
T
-
4
,
b
i
n
g
,
a
n
d
b
a
r
d
i
n
t
h
e
a
meri
c
a
n
so
c
i
e
t
y
f
o
r
me
t
a
b
o
l
i
c
a
n
d
b
a
r
i
a
t
r
i
c
su
r
g
e
r
y
t
e
x
t
b
o
o
k
o
f
b
a
r
i
a
t
r
i
c
s
u
r
g
e
r
y
q
u
e
st
i
o
n
s
,
”
S
u
r
g
e
r
y
f
o
r
O
b
e
si
t
y
a
n
d
Re
l
a
t
e
d
D
i
sea
ses
,
v
o
l
.
2
0
,
n
o
.
7
,
p
p
.
6
0
9
–
6
1
3
,
Ju
l
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
s
o
a
r
d
.
2
0
2
4
.
0
4
.
0
1
4
.
[
2
]
K
.
S
.
Q
a
d
e
r
a
n
d
K
.
C
e
k
,
“
I
n
f
l
u
e
n
c
e
o
f
b
l
o
c
k
c
h
a
i
n
a
n
d
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
o
n
a
u
d
i
t
q
u
a
l
i
t
y
:
E
v
i
d
e
n
c
e
f
r
o
m
T
u
r
k
e
y
,
”
H
e
l
i
y
o
n
,
v
o
l
.
1
0
,
n
o
.
9
,
M
a
y
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
h
e
l
i
y
o
n
.
2
0
2
4
.
e
3
0
1
6
6
.
[
3
]
A
.
K
h
a
m
a
j
,
A
.
M
.
A
l
i
,
R
.
S
a
mi
n
a
t
h
a
n
,
a
n
d
S
.
M
,
“
H
u
m
a
n
f
a
c
t
o
r
s
e
n
g
i
n
e
e
r
i
n
g
s
i
m
u
l
a
t
e
d
a
n
a
l
y
si
s
i
n
a
d
mi
n
i
s
t
r
a
t
i
v
e
,
o
p
e
r
a
t
i
o
n
a
l
a
nd
mai
n
t
e
n
a
n
c
e
l
o
o
p
s
o
f
n
u
c
l
e
a
r
r
e
a
c
t
o
r
c
o
n
t
r
o
l
u
n
i
t
u
si
n
g
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
a
n
d
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,
”
H
e
l
i
y
o
n
,
v
o
l
.
1
0
,
n
o
.
1
0
,
M
a
y
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
h
e
l
i
y
o
n
.
2
0
2
4
.
e
3
0
8
6
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.