I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
9
,
No
.
4
,
A
u
g
u
s
t
201
9
,
p
p
.
2
9
4
1
~2
9
4
9
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v9
i
4
.
p
p
2
9
4
1
-
2949
2941
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e
.
co
m/
jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
I
JE
C
E
PSO
-
SVM
hy
brid
sy
ste
m
f
o
r
m
e
la
n
o
m
a
d
e
tect
io
n
fro
m
h
isto
-
pa
tholo
g
ica
l
i
m
a
g
es
M
a
en
T
a
k
rur
i
1
,
M
o
ha
m
ed
K
ha
led A
bu
M
a
h
m
o
ud
2
,
A
d
el
Al
-
J
u
m
a
ily
3
1
De
p
a
rte
m
e
n
t
o
f
El
e
c
tri
c
a
l
,
El
e
c
tro
n
ics
a
n
d
Co
m
m
u
n
ica
ti
o
n
s
En
g
in
e
e
rin
g
,
Am
e
rica
n
Un
iv
e
rsit
y
o
f
Ra
s
A
l
Kh
a
im
a
h
,
Un
it
e
d
A
ra
b
Em
irate
s
2,
3
F
a
c
u
lt
y
o
f
En
g
in
e
e
rin
g
a
n
d
I
n
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
,
Un
iv
e
rsit
y
o
f
T
e
c
h
n
o
l
o
g
y
,
S
y
d
n
e
y
,
A
u
stra
li
a
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
28
,
2
0
1
7
R
ev
i
s
ed
J
u
n
9
,
2
0
1
8
A
cc
ep
ted
Mar
4
,
2
0
1
9
T
h
is
p
a
p
e
r
in
tr
o
d
u
c
e
s
a
n
a
u
t
o
m
a
ted
s
y
ste
m
f
o
r
sk
in
c
a
n
c
e
r
(m
e
lan
o
m
a
)
d
e
tec
ti
o
n
f
ro
m
Histo
-
p
a
th
o
l
o
g
ica
l
im
a
g
e
s
sa
m
p
led
f
ro
m
m
icro
sc
o
p
ic
slid
e
s
o
f
sk
in
b
io
p
sy
.
T
h
e
p
ro
p
o
se
d
sy
ste
m
is
a
h
y
b
rid
s
y
ste
m
b
a
se
d
o
n
P
a
rti
c
le
S
w
a
r
m
Op
ti
m
iza
ti
o
n
a
n
d
S
u
p
p
o
r
t
V
e
c
to
r
M
a
c
h
in
e
(
P
S
O
-
S
V
M
).
T
h
e
f
e
a
tu
re
s
u
se
d
a
re
e
x
trac
t
e
d
f
ro
m
th
e
g
r
a
y
sc
a
le
i
m
a
g
e
h
isto
g
ra
m
,
th
e
c
o
-
o
c
c
u
rre
n
c
e
m
a
tri
x
a
n
d
th
e
e
n
e
rg
y
o
f
th
e
w
a
v
e
let
c
o
e
ff
icie
n
ts
re
su
lt
in
g
f
ro
m
th
e
wa
v
e
let
p
a
c
k
e
t
d
e
c
o
m
p
o
siti
o
n
.
T
h
e
P
S
O
-
S
V
M
sy
ste
m
se
le
c
ts
th
e
b
e
st
f
e
a
tu
re
se
t
a
n
d
th
e
b
e
st
v
a
lu
e
s
f
o
r
th
e
S
VM
p
a
r
a
m
e
ters
(C
a
n
d
γ)
th
a
t
o
p
ti
m
ize
th
e
p
e
rf
o
rm
a
n
c
e
o
f
th
e
S
V
M
c
las
sif
ier.
T
h
e
s
y
ste
m
p
e
r
f
o
r
m
a
n
c
e
is
tes
ted
o
n
a
re
a
l
d
a
tas
e
t
o
b
tain
e
d
f
ro
m
th
e
S
o
u
t
h
e
rn
P
a
th
o
lo
g
y
L
a
b
o
ra
to
ry
in
W
o
ll
o
n
g
o
n
g
NSW
,
A
u
stra
li
a
.
Ev
a
lu
a
ti
o
n
re
su
l
ts
sh
o
w
a
c
las
sif
ic
a
ti
o
n
a
c
c
u
ra
c
y
o
f
8
7
.
1
3
%
,
a
se
n
siti
v
i
ty
o
f
9
4
.
1
%
a
n
d
a
sp
e
c
i
f
icit
y
o
f
8
0
.
2
2
%
.
T
h
e
se
n
siti
v
it
y
a
n
d
sp
e
c
if
icity
re
su
lt
s
a
re
c
o
m
p
a
ra
b
le
to
t
h
o
se
o
b
tain
e
d
b
y
d
e
r
m
a
to
lo
g
ists.
K
ey
w
o
r
d
s
:
H
is
to
-
p
at
h
o
lo
g
ical
i
m
a
g
es
Me
lan
o
m
a
Sk
i
n
c
an
ce
r
d
iag
n
o
s
tic
s
y
s
te
m
Sk
i
n
l
esio
n
Co
p
y
rig
h
t
©
2
0
1
9
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Ma
en
T
ak
r
u
r
i,
Dep
ar
te
m
en
t o
f
E
lectr
ical,
E
lectr
o
n
ics an
d
C
o
m
m
u
n
icat
io
n
s
E
n
g
i
n
ee
r
i
n
g
,
Am
er
ica
n
Un
i
v
er
s
i
t
y
o
f
R
as
Al
Kh
ai
m
ah
,
P
.
O.
B
o
x
1
0
0
2
1
,
R
as A
l K
h
a
m
ah
,
U
n
ited
A
r
ab
E
m
ir
ates
.
E
m
ail:
m
ae
n
.
ta
k
r
u
r
i@
a
u
r
ak
.
ac
.
ae
1.
I
NT
RO
D
UCT
I
O
N
Sk
i
n
ca
n
ce
r
is
in
cr
ea
s
in
g
l
y
b
e
co
m
in
g
m
o
r
e
co
m
m
o
n
g
lo
b
al
l
y
.
I
t
is
s
p
r
ea
d
in
g
in
Au
s
tr
alia
at
h
ig
h
e
r
r
ates
th
an
o
th
er
ca
n
ce
r
t
y
p
es
[1
,
2]
.
T
h
e
d
ea
d
lies
t
f
o
r
m
o
f
s
k
in
ca
n
ce
r
is
Me
lan
o
m
a.
Me
lan
o
m
a
ca
s
e
s
r
ep
o
r
ted
in
Au
s
tr
alia
ar
e
th
e
h
ig
h
est
g
lo
b
all
y
at
al
m
o
s
t
f
o
u
r
ti
m
es
t
h
e
r
ates
s
ee
n
in
C
an
ad
a,
th
e
U
n
ited
Kin
g
d
o
m
an
d
th
e
U
n
ited
Stat
es
[
3
]
.
R
ep
o
r
ts
h
av
e
s
h
o
w
n
t
h
at
m
ela
n
o
m
a
h
as
ac
co
u
n
ted
f
o
r
7
5
%
o
f
ca
n
ce
r
d
ea
th
s
i
n
A
u
s
tr
alia
[
4
]
.
E
ar
ly
d
iag
n
o
s
is
o
f
Me
la
n
o
m
a
,
w
h
ic
h
is
o
b
v
io
u
s
l
y
d
ep
en
d
e
n
t
u
p
o
n
p
atien
t
’
s
atte
n
tio
n
a
n
d
ac
cu
r
ate
ass
es
s
m
en
t
b
y
a
m
ed
ical
p
r
ac
t
itio
n
er
,
is
cr
u
cial.
S
tu
d
ies
h
a
v
e
s
h
o
w
ed
t
h
at
m
ela
n
o
m
a
ca
n
b
e
cu
r
ed
at
a
r
ate
o
f
9
5
%
if
d
ia
g
n
o
s
ed
an
d
tr
ea
ted
in
ea
r
l
y
s
ta
g
es
[
5
]
.
I
t
ca
n
b
e
r
e
m
o
v
ed
b
y
s
i
m
p
le
s
u
r
g
er
y
i
f
it
h
as
n
o
t
e
n
ter
ed
th
e
b
lo
o
d
s
tr
ea
m
.
Me
la
n
o
m
a
ca
n
b
e
d
iag
n
o
s
ed
v
is
u
all
y
in
a
n
o
n
in
v
a
s
iv
e
f
as
h
io
n
b
u
t
th
is
ca
n
lead
to
in
ac
cu
r
ate
j
u
d
g
m
e
n
ts
a
s
it
is
v
is
u
all
y
h
ar
d
f
o
r
m
ed
ical
p
r
o
f
es
s
io
n
al
s
to
d
if
f
er
e
n
tiate
n
o
r
m
al
f
r
o
m
ab
n
o
r
m
a
l
m
o
le.
I
t h
as b
ee
n
r
ep
o
r
ted
th
at
th
e
ac
cu
r
a
c
y
o
f
s
p
ec
ialized
Der
m
ato
lo
g
i
s
t
’
s
ce
n
ter
s
is
o
n
l
y
ab
o
u
t 6
0
% [
6
]
.
A
m
o
r
e
r
eliab
le
w
a
y
f
o
r
s
k
i
n
ca
n
ce
r
d
iag
n
o
s
is
i
s
b
ased
o
n
th
e
s
t
u
d
y
o
f
s
k
i
n
lesi
o
n
p
ath
o
lo
g
y
.
P
ath
o
lo
g
is
t
s
h
a
v
e,
tr
ad
itio
n
all
y
,
u
s
ed
Hi
s
to
‐
p
ath
o
lo
g
ical
i
m
ag
es
o
f
b
io
p
s
y
s
a
m
p
les
r
e
m
o
v
ed
f
r
o
m
p
atie
n
ts
an
d
m
ad
e
j
u
d
g
m
e
n
t
s
b
ased
o
n
t
h
e
d
e
v
iatio
n
s
i
n
t
h
e
ce
ll
s
t
r
u
ctu
r
es
an
d
/o
r
t
h
e
c
h
a
n
g
e
s
i
n
t
h
e
d
is
tr
ib
u
tio
n
o
f
th
e
ce
ll
s
ac
r
o
s
s
t
h
e
tis
s
u
e
u
n
d
er
ex
a
m
i
n
atio
n
.
Ho
w
e
v
er
,
th
ese
j
u
d
g
m
e
n
t
s
c
an
s
u
b
j
ec
tiv
e
as
th
e
y
d
ep
en
d
o
n
th
e
ex
p
er
ie
n
ce
o
f
P
ath
o
lo
g
i
s
ts
an
d
o
f
ten
lead
to
co
n
s
id
er
ab
le
v
ar
iab
ilit
y
[
7
,
8
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
9
,
No
.
4
,
A
u
g
u
s
t 2
0
1
9
:
2
9
4
1
-
29
4
9
2942
I
n
o
r
d
er
to
im
p
r
o
v
e
th
e
r
eliab
ilit
y
o
f
s
k
in
ca
n
ce
r
d
iag
n
o
s
i
s
,
r
esear
ch
er
s
h
a
v
e
t
h
o
u
g
h
t
o
f
d
ev
elo
p
in
g
co
m
p
u
tatio
n
al
to
o
ls
f
o
r
au
to
m
ated
ca
n
ce
r
d
ia
g
n
o
s
is
t
h
at
o
p
er
ate
o
n
q
u
an
titati
v
e
m
ea
s
u
r
es.
B
ec
au
s
e
o
f
it
s
p
r
o
m
i
s
in
g
b
en
e
f
it
s
i
n
r
ed
u
ci
n
g
t
h
e
n
u
m
b
er
o
f
f
a
talit
y
o
f
s
k
i
n
ca
n
ce
r
p
atie
n
ts
,
a
u
to
m
ated
ca
n
ce
r
d
iag
n
o
s
is
h
a
s
b
ec
o
m
e
an
i
m
p
o
r
tan
t
r
esear
ch
to
p
ic.
M
o
s
t
o
f
th
e
w
o
r
k
i
n
t
h
at
f
ie
ld
is
b
ased
o
n
ap
p
ly
i
n
g
i
m
ag
e
p
r
o
ce
s
s
i
n
g
an
d
m
ac
h
in
e
lear
n
in
g
tec
h
n
iq
u
es
o
n
ex
ter
n
al
s
k
i
n
m
ela
n
o
m
a
i
m
ag
e
s
to
d
iag
n
o
s
e
w
h
eth
er
th
e
y
ar
e
b
en
ig
n
o
r
m
ali
g
n
a
n
t.
So
m
e
o
f
t
h
e
ap
p
r
o
ac
h
es
w
o
r
k
o
n
a
n
al
y
zin
g
e
x
ter
n
al
s
k
i
n
i
m
a
g
es [
9
-
1
1
]
.
Oth
er
ap
p
r
o
ac
h
es
w
o
r
k
o
n
d
et
ec
tin
g
s
k
i
n
ca
n
ce
r
b
ased
o
n
H
is
to
-
p
at
h
o
lo
g
ical
i
m
a
g
e
s
o
b
tai
n
ed
f
r
o
m
s
k
i
n
lesi
o
n
s
[
1
2
,
1
3
]
an
d
th
ey
f
o
cu
s
o
n
q
u
a
n
ti
f
y
i
n
g
b
io
m
ar
k
er
s
o
n
a
p
ix
el
‐
by
‐
p
ix
e
l
b
asis
o
r
a
r
eg
io
n
al
b
asis
[
1
4
,
1
5
]
.
T
h
ese
ap
p
r
o
ac
h
es
ar
e
in
s
p
ir
ed
b
y
th
e
w
a
y
t
h
at
P
ath
o
lo
g
is
ts
f
o
llo
w
to
d
iag
n
o
s
e
s
k
i
n
ca
n
ce
r
.
T
h
ey
ar
e
m
o
r
e
b
io
lo
g
icall
y
r
el
ev
an
t
m
ea
s
u
r
e
a
n
d
ca
n
b
e
m
o
r
e
u
s
e
f
u
l i
n
m
ed
ical
d
ia
g
n
o
s
is
as th
e
y
ca
n
p
r
o
v
id
e
a
s
ec
o
n
d
o
p
in
io
n
f
o
r
th
e
p
at
h
o
lo
g
is
t b
ef
o
r
e
ta
k
in
g
t
h
e
f
i
n
al
d
ec
is
io
n
.
An
au
to
m
ated
s
k
in
ca
n
ce
r
d
etec
tio
n
m
e
th
o
d
th
at
i
s
b
ased
o
n
h
is
to
-
p
at
h
o
lo
g
ical
i
m
a
g
es
o
f
s
k
i
n
lesi
o
n
s
w
as
in
tr
o
d
u
ce
d
i
n
[
1
3
]
.
I
t a
i
m
ed
to
e
n
ab
le
th
e
d
is
cr
i
m
i
n
atio
n
b
et
w
ee
n
m
ela
n
o
c
y
t
i
c
n
e
v
i a
n
d
m
ali
g
n
an
t
m
ela
n
o
m
a.
Af
ter
f
ilter
i
n
g
th
e
i
m
a
g
e
u
s
i
n
g
s
p
atiall
y
ad
ap
tiv
e
co
lo
r
m
ed
ia
n
f
ilter
an
d
ap
p
ly
i
n
g
K
-
m
ea
n
s
clu
s
ter
i
n
g
f
o
r
s
e
g
m
en
tatio
n
,
i
m
ag
e
f
ea
t
u
r
es
w
er
e
o
b
tain
ed
f
r
o
m
t
h
e
h
is
to
g
r
a
m
a
n
d
t
h
e
co
-
o
cc
u
r
r
en
ce
m
atr
i
x
.
T
h
e
ex
tr
ac
ted
f
ea
t
u
r
es
w
er
e
th
en
r
ed
u
ce
d
u
s
i
n
g
s
eq
u
e
n
tia
l
f
e
atu
r
e
s
elec
t
io
n
a
n
d
t
h
en
w
er
e
f
ed
to
a
n
SVM
f
o
r
tr
ain
i
n
g
an
d
test
i
n
g
.
T
h
e
d
ata
s
et
u
s
ed
f
o
r
t
h
e
e
v
al
u
atio
n
o
f
th
e
alg
o
r
it
h
m
w
a
s
o
b
tai
n
ed
f
r
o
m
th
e
So
u
th
er
n
P
ath
o
lo
g
y
L
ab
o
r
ato
r
y
i
n
W
o
ll
o
n
g
o
n
g
N
SW
,
Au
s
tr
alia.
I
t
in
clu
d
ed
4
2
His
to
-
p
ath
o
lo
g
ical
i
m
a
g
es
(
2
8
b
en
ig
n
an
d
1
4
m
e
lan
o
m
a)
.
T
h
e
d
atas
et
w
as
d
iv
id
ed
i
n
to
tr
ai
n
i
n
g
s
e
t
(
6
0
%),
v
alid
atio
n
s
et
(
2
0
%),
an
d
te
s
t
s
et
(
2
0
%).
T
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
ac
h
ie
v
e
d
a
class
if
icatio
n
ac
cu
r
ac
y
o
f
8
8
.
9
%,
s
en
s
iti
v
it
y
o
f
8
7
.
5
%
an
d
s
p
ec
if
icit
y
o
f
1
0
0
%.
W
h
ile
th
e
ac
h
ie
v
ed
`
r
e
s
u
lt
s
w
er
e
v
er
y
g
o
o
d
,
th
e
m
et
h
o
d
h
ad
h
ig
h
co
m
p
u
tatio
n
al
ti
m
e
co
s
t.
T
h
e
p
ap
er
co
n
clu
d
ed
th
at
m
et
h
o
d
s
h
o
u
ld
b
e
f
u
r
th
er
te
s
ted
o
n
lar
g
er
d
ataset
to
v
er
if
y
t
h
e
r
eliab
ilit
y
its
r
eliab
ilit
y
.
An
o
t
h
er
ap
p
r
o
ac
h
f
o
r
au
to
m
ated
s
k
in
ca
n
ce
r
d
etec
tio
n
f
r
o
m
Hi
s
to
-
p
at
h
o
lo
g
ical
o
f
s
k
i
n
lesi
o
n
s
i
m
a
g
es
w
as
i
n
tr
o
d
u
ce
d
i
n
[
1
6
]
.
Un
li
k
e
[
1
3
]
,
in
th
e
p
r
e
-
p
r
o
ce
s
s
i
n
g
s
ta
g
e,
t
h
e
i
m
ag
e
w
as
p
a
s
s
ed
t
h
r
o
u
g
h
th
r
ee
f
ilter
i
n
g
s
ta
g
es,
n
a
m
el
y
,
W
ien
er
f
ilter
,
ad
ap
tiv
e
m
ed
ia
n
f
ilter
an
d
Gab
o
r
f
ilter
to
i
m
p
r
o
v
e
d
iag
n
o
s
tic
ac
cu
r
ac
y
.
His
to
g
r
a
m
eq
u
aliza
tio
n
w
as
u
s
ed
to
en
h
an
ce
t
h
e
co
n
tr
ast
o
f
th
e
i
m
a
g
es
p
r
io
r
to
s
eg
m
e
n
ta
tio
n
.
Seg
m
e
n
tatio
n
w
a
s
i
m
p
le
m
en
ted
th
r
o
u
g
h
E
d
g
e
Dete
c
tio
n
Op
er
atio
n
s
.
I
m
a
g
e
f
ea
t
u
r
es
w
er
e
o
b
tain
ed
f
r
o
m
t
h
e
h
i
s
to
g
ra
m
a
n
d
th
e
co
-
o
cc
u
r
r
en
ce
m
a
tr
ix
.
T
h
e
ex
tr
ac
ted
f
ea
t
u
r
es
w
er
e
t
h
e
n
r
ed
u
ce
d
u
s
in
g
s
eq
u
en
tial
f
e
atu
r
e
s
elec
tio
n
an
d
th
en
w
er
e
f
ed
to
an
SVM
f
o
r
t
r
ain
in
g
a
n
d
test
in
g
.
T
h
e
al
g
o
r
ith
m
w
as
test
ed
o
n
t
h
e
s
a
m
e
d
ataset
u
s
ed
i
n
[
1
3
]
.
I
t
ac
h
iev
ed
a
class
i
f
icat
io
n
ac
cu
r
ac
y
o
f
8
1
%,
a
s
en
s
i
tiv
it
y
o
f
7
6
%
an
d
a
s
p
ec
if
icit
y
o
f
1
0
0
%
w
h
ic
h
ar
e
less
th
an
t
h
e
r
es
u
lt
s
o
b
tain
ed
in
[
1
3
]
.
Ho
w
e
v
er
,
th
e
al
g
o
r
ith
m
w
a
s
1
7
ti
m
es
f
aster
t
h
an
t
h
at
i
n
[
1
3
]
.
I
n
th
eir
e
f
f
o
r
ts
to
ad
d
r
ess
th
e
s
a
m
e
p
r
o
b
le
m
,
th
e
a
u
th
o
r
s
i
n
[
1
7
]
u
s
ed
ad
d
itio
n
al
f
ea
tu
r
e
s
ex
tr
ac
te
d
f
r
o
m
t
h
e
W
av
elet
P
ac
k
et
T
r
a
n
s
f
o
r
m
atio
n
(
W
PT
)
o
f
th
e
p
r
e
-
p
r
o
ce
s
s
ed
His
to
-
p
at
h
o
lo
g
ica
l
i
m
a
g
e
alo
n
g
w
i
t
h
th
e
f
ea
tu
r
e
s
o
b
tain
ed
f
r
o
m
th
e
h
is
to
g
r
a
m
a
n
d
th
e
co
-
o
cc
u
r
r
e
n
ce
m
a
tr
ix
u
s
ed
i
n
[
1
6
]
.
T
h
e
p
ap
er
in
tr
o
d
u
ce
d
a
P
SO
-
SVM
f
r
a
m
e
w
o
r
k
t
h
at
en
ab
led
s
i
m
u
lta
n
eo
u
s
f
ea
t
u
r
e
s
elec
tio
n
an
d
SVM
p
ar
am
eter
s
o
p
ti
m
izat
io
n
.
T
h
e
ev
alu
atio
n
s
w
er
e
co
n
d
u
c
ted
o
n
a
r
ea
l
d
ataset
d
if
f
er
en
t
f
r
o
m
th
at
u
s
ed
in
[
1
3
]
an
d
[
1
6
]
.
I
t
in
clu
d
es
7
9
His
to
-
p
at
h
o
lo
g
ical
i
m
ag
e
s
.
T
h
e
s
y
s
te
m
ac
h
ie
v
ed
a
cla
s
s
i
f
ica
tio
n
ac
c
u
r
ac
y
o
f
8
7
.
1
3
%,
a
s
e
n
s
it
iv
i
t
y
o
f
9
4
.
1
%
an
d
a
s
p
ec
if
icit
y
o
f
8
0
.
2
2
%.
I
n
t
h
is
s
t
u
d
y
,
w
e
e
x
te
n
d
o
u
r
w
o
r
k
in
[
1
7
]
an
d
p
r
o
v
id
e
a
f
u
ll
e
x
p
la
n
atio
n
o
f
th
e
P
SO
-
S
VM
n
o
v
el
m
ela
n
o
m
a
d
etec
t
io
n
s
tr
ateg
y
i
n
tr
o
d
u
ce
d
th
er
e.
I
t
is
b
ased
o
n
a
h
y
b
r
id
P
ar
ticle
S
w
ar
m
Op
ti
m
izat
io
n
-
Su
p
p
o
r
t
Vec
to
r
Ma
ch
in
e
(
P
SO
-
SV
M)
f
r
a
m
e
w
o
r
k
t
h
at
ai
m
s
to
en
ab
le
i
m
p
r
o
v
in
g
i
m
ag
e
f
ea
tu
r
e
s
s
elec
tio
n
an
d
SV
M
p
ar
am
eter
s
o
p
ti
m
izat
io
n
s
i
m
u
ltan
eo
u
s
l
y
.
W
e
i
n
cl
u
d
e
d
etail
ed
an
d
f
air
co
m
p
ar
is
o
n
s
w
i
th
th
e
w
o
r
k
i
n
[
1
3
]
u
s
i
n
g
th
e
s
a
m
e
d
ataset.
W
e
clea
r
l
y
s
h
o
w
th
a
t
th
e
p
r
o
p
o
s
ed
w
o
r
k
ac
h
ie
v
ed
r
esu
lt
s
ar
e
b
en
ch
m
ar
k
w
it
h
t
h
o
s
e
ac
h
iev
ed
b
y
GP
s
an
d
Der
m
ato
lo
g
is
t
s
to
s
h
o
w
s
u
p
er
io
r
it
y
o
f
o
u
r
s
o
lu
tio
n
s
.
T
h
e
ev
al
u
atio
n
s
ar
e
co
n
d
u
cted
o
n
a
r
ea
l
d
ataset
o
b
tain
ed
f
r
o
m
th
e
So
u
t
h
er
n
P
ath
o
lo
g
y
L
ab
o
r
ato
r
y
in
W
o
llo
n
g
o
n
g
NSW
,
A
u
s
tr
alia.
I
t
i
n
clu
d
es
7
9
His
to
-
p
at
h
o
lo
g
ical
i
m
a
g
es.
T
h
e
p
ap
er
is
o
r
g
an
ized
as
f
o
ll
o
w
s
:
Sectio
n
s
2
ex
p
lai
n
s
t
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
a
n
d
t
h
e
s
ta
g
es
f
o
llo
w
ed
to
r
ea
ch
th
e
d
ec
is
io
n
.
T
h
e
ev
alu
a
tio
n
s
o
f
o
u
r
p
r
o
p
o
s
ed
s
y
s
te
m
ar
e
p
r
esen
ted
in
S
ec
tio
n
3
f
o
llo
w
ed
i
n
S
ec
tio
n
4
b
y
c
o
n
cl
u
s
io
n
s
an
d
d
ir
ec
tio
n
s
f
o
r
f
u
t
u
r
e
w
o
r
k
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
T
h
e
b
lo
ck
d
iag
r
am
i
n
F
ig
u
r
e
1
d
escr
ib
es
th
e
p
r
o
p
o
s
ed
s
k
in
ca
n
ce
r
d
iag
n
o
s
tic
s
y
s
te
m
.
I
t
co
n
s
is
t
o
f
th
e
f
o
llo
w
i
n
g
s
ta
g
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:
2
0
8
8
-
8708
P
S
O
-
S
V
M h
yb
r
id
s
ystem
fo
r
mela
n
o
ma
d
etec
tio
n
fr
o
m
h
is
to
-
p
a
th
o
lo
g
ica
l ima
g
es (
Ma
en
Ta
kru
r
i)
2943
H
i
s
t
o
-
P
a
t
h
o
l
o
g
i
c
a
l
D
i
g
i
t
a
l
I
m
a
g
e
P
r
e
-
p
r
o
c
e
s
s
i
n
g
:
W
i
e
n
e
r
F
i
l
t
e
r
G
a
b
o
r
F
i
l
t
e
r
M
e
d
i
a
n
F
i
l
t
e
r
I
m
a
g
e
E
n
h
a
n
c
e
m
e
n
t
:
H
i
s
t
o
g
r
a
m
E
q
a
l
i
z
a
t
i
o
n
S
e
g
m
e
n
t
a
t
i
o
n
:
E
d
g
e
D
e
t
e
c
t
i
o
n
(
S
o
b
e
l
)
F
e
a
t
u
r
e
E
x
t
r
a
c
t
i
o
n
F
o
r
m
:
I
m
a
g
e
H
i
s
t
o
g
r
a
m
C
o
n
f
u
s
i
o
n
M
a
t
r
i
x
W
a
v
e
l
e
t
e
P
a
c
k
e
t
T
r
a
n
f
o
r
m
F
e
a
t
u
r
e
S
e
l
e
c
t
i
o
n
:
P
S
O
C
l
a
s
s
i
f
i
c
a
t
i
o
n
:
S
V
M
Fig
u
r
e
1
.
B
lo
ck
d
iag
r
a
m
f
o
r
th
e
p
r
o
p
o
s
ed
s
k
in
le
s
io
n
clas
s
i
f
i
ca
tio
n
s
y
s
te
m
2
.
1
.
P
re
-
pro
ce
s
s
ing
s
t
a
g
e
P
r
e
-
P
r
o
ce
s
s
in
g
i
s
m
ea
n
t
to
f
a
cilitate
i
m
a
g
e
s
e
g
m
en
tatio
n
b
y
f
ilter
i
n
g
th
e
n
o
i
s
e
f
r
o
m
t
h
e
i
m
a
g
e
a
n
d
en
h
a
n
ci
n
g
its
i
m
p
o
r
tan
t
f
ea
t
u
r
es
[
1
8
]
.
I
n
th
is
w
o
r
k
,
W
ien
er
f
ilter
is
u
s
ed
as
an
o
p
ti
m
al
w
a
y
f
o
r
ac
co
u
n
ti
n
g
f
o
r
th
e
n
o
is
y
co
m
p
o
n
e
n
ts
to
r
esu
l
t
in
th
e
b
est
r
ec
o
n
s
tr
u
c
tio
n
o
f
th
e
o
r
ig
in
a
l
i
m
a
g
e.
I
t
is
b
asicall
y
co
n
s
id
er
ed
as
a
lo
w
p
a
s
s
f
i
lter
to
r
ed
u
ce
co
n
s
tan
t
p
o
w
er
ad
d
itiv
e
Gau
s
s
ia
n
n
o
is
e.
I
n
ad
d
itio
n
,
Gab
o
r
f
ilter
w
h
ich
is
w
el
l
s
u
ited
f
o
r
tex
t
u
r
e
s
e
g
m
en
ta
tio
n
p
r
o
b
lem
s
[
1
9
-
21
]
is
e
m
p
lo
y
ed
o
n
th
e
u
n
s
eg
m
e
n
ted
His
to
-
p
ath
o
lo
g
ical
i
m
a
g
e
s
th
at
r
eq
u
ir
e
C
ell
a
n
d
T
ex
tu
r
e
p
r
o
p
er
ties
an
al
y
s
is
to
i
m
p
r
o
v
e
th
e
s
e
g
m
e
n
tat
io
n
p
r
o
ce
s
s
.
Fi
n
all
y
,
Me
d
ian
f
ilter
i
n
g
w
h
ic
h
is
a
n
o
n
li
n
ea
r
o
p
er
atio
n
is
u
s
ed
to
r
ed
u
ce
n
o
is
e
an
d
p
r
eser
v
e
ed
g
es
[
2
2
,
23
]
.
2
.
2
.
I
m
a
g
e
e
nh
a
nce
m
ent
I
m
ag
e
e
n
h
a
n
ce
m
e
n
t
m
e
th
o
d
s
ai
m
to
i
m
p
r
o
v
e
t
h
e
co
n
tr
a
s
t
a
n
d
v
is
ib
ilit
y
o
f
t
h
e
i
m
a
g
e
f
ea
tu
r
es
t
h
at
d
ep
en
d
o
n
th
e
im
a
g
i
n
g
m
o
d
alit
y
as
w
ell
as
th
e
a
n
ato
m
ica
l
r
eg
io
n
s
[
1
9
,
23
,
24
]
.
I
n
th
is
w
o
r
k
,
His
to
g
r
a
m
E
q
u
aliza
tio
n
is
u
s
ed
f
o
r
i
m
a
g
e
en
h
a
n
ce
m
e
n
t.
I
t
is
b
ased
o
n
ad
j
u
s
tin
g
i
m
a
g
e
in
te
n
s
ities
Hi
s
to
g
r
a
m
to
en
h
a
n
ce
th
e
i
m
a
g
e
co
n
tr
a
s
t.
W
e
u
s
e
i
t
in
th
i
s
w
o
r
k
as
it
allo
w
s
f
o
r
ar
ea
s
o
f
lo
w
er
lo
ca
l
co
n
tr
ast
to
g
ain
a
h
ig
h
er
co
n
tr
ast
w
h
ich
in
t
u
r
n
r
es
u
lt
s
in
m
a
k
in
g
s
o
m
e
i
m
p
o
r
tan
t
tex
t
u
r
al
p
r
o
p
er
ties
in
His
to
-
p
ath
o
lo
g
ical
i
m
a
g
e
s
m
o
r
e
v
is
ib
le
a
n
d
i
m
p
r
o
v
es t
h
e
f
ea
tu
r
e
e
x
tr
ac
tio
n
p
r
o
ce
s
s
.
2
.
3
.
I
m
a
g
e
s
eg
m
ent
a
t
io
n
I
n
t
h
is
s
t
u
d
y
So
b
el
o
p
er
ato
r
b
ased
E
d
g
e
Dete
ct
io
n
is
u
s
ed
f
o
r
t
h
e
p
u
r
p
o
s
e
o
f
s
e
g
m
en
t
i
n
g
i
m
ag
e
s
.
T
h
e
So
b
el
o
p
e
r
ato
r
is
an
ex
a
m
p
le
o
f
t
h
e
g
r
ad
ie
n
t
ed
g
e
d
e
tectio
n
m
eth
o
d
s
.
I
t
is
a
d
is
cr
ete
d
if
f
er
e
n
tiatio
n
o
p
er
ato
r
,
co
m
p
u
ti
n
g
an
ap
p
r
o
x
i
m
atio
n
o
f
th
e
g
r
ad
ien
t o
f
t
h
e
i
m
ag
e
i
n
te
n
s
it
y
f
u
n
ct
io
n
[
2
5
,
2
6
]
.
T
h
e
g
r
ad
ien
t
m
a
g
n
it
u
d
e
an
d
d
ir
ec
tio
n
al
in
f
o
r
m
atio
n
ar
e
f
o
u
n
d
as
f
o
llo
w
s
:
T
h
e
i
m
ag
e
i
s
co
n
v
o
lv
ed
w
it
h
th
e
So
b
el
h
o
r
izo
n
tal
an
d
v
er
tical
d
ir
ec
tio
n
s
m
a
s
k
s
(
o
p
er
ato
r
s
)
to
r
esu
lt
i
n
th
e
h
o
r
izo
n
tal
a
n
d
v
er
tica
l
d
er
iv
ativ
e
(
g
r
ad
ien
t)
ap
p
r
o
x
i
m
atio
n
s
Gx
an
d
Gy
r
esp
ec
t
iv
el
y
.
T
h
e
g
r
ad
ien
t
m
ag
n
it
u
d
e
M
is
tak
e
n
a
s
th
e
ab
s
o
lu
te
s
u
m
o
f
v
al
u
es
o
f
th
e
h
o
r
izo
n
ta
l
an
d
v
er
tica
l
g
r
a
d
ien
t
ap
p
r
o
x
i
m
atio
n
s
[
2
7
,
2
8
]
.
T
h
e
p
o
in
t
th
at
h
a
s
h
ig
h
v
alu
e
o
f
M
w
ill ap
p
ea
r
as a
n
ed
g
e
p
o
in
t in
t
h
e
r
es
u
lti
n
g
i
m
a
g
e
[
2
5
]
.
2
.
4
.
F
ea
t
ure
e
x
t
ra
ct
io
n
Si
m
i
lar
to
[
1
6
]
,
f
o
r
ea
ch
His
to
-
p
ath
o
lo
g
ica
l
i
m
a
g
e,
s
i
x
f
e
atu
r
es
ar
e
ex
tr
ac
ted
f
r
o
m
t
h
e
g
r
a
y
s
ca
le
i
m
a
g
e
h
i
s
to
g
r
a
m
(
Me
an
,
Var
i
an
ce
,
Sk
e
w
n
e
s
s
,
K
u
r
to
s
is
,
E
n
er
g
y
,
an
d
E
n
tr
o
p
y
)
a
n
d
t
w
en
t
y
o
n
e
f
ea
tu
r
es
ar
e
ex
tr
ac
ted
f
r
o
m
th
e
co
-
o
cc
u
r
r
en
ce
.
I
n
th
is
p
ap
er
,
w
e
u
s
e
W
av
elet
P
ac
k
et
T
r
an
s
f
o
r
m
(
W
PT
)
o
f
th
e
i
m
ag
e
to
p
r
o
v
id
e
u
s
w
i
t
h
a
s
e
t
o
f
ad
d
itio
n
al
f
ea
tu
r
e
s
.
W
P
T
is
a
g
en
er
alize
d
v
er
s
io
n
o
f
th
e
W
av
elet
T
r
an
s
f
o
r
m
i
n
w
h
ic
h
th
e
h
i
g
h
-
f
r
eq
u
e
n
c
y
p
ar
t
is
als
o
s
p
lit
in
to
a
lo
w
a
n
d
a
h
ig
h
f
r
eq
u
en
c
y
p
ar
ts
an
d
s
o
o
n
[
2
9
]
.
T
h
is
p
r
o
d
u
ce
s
a
d
ec
o
m
p
o
s
itio
n
tr
ee
.
W
e
w
o
r
k
d
o
w
n
to
7
d
ec
o
m
p
o
s
itio
n
lev
els
o
f
W
P
T
r
esu
ltin
g
i
n
2
5
5
co
m
p
o
n
e
n
ts
.
T
h
e
f
ea
t
u
r
es a
r
e
g
e
n
er
ated
b
y
tak
i
n
g
t
h
e
e
n
er
g
y
o
f
t
h
e
w
a
v
el
et
co
ef
f
icie
n
ts
in
t
h
e
S
u
b
b
an
d
[
30
]
as seen
in
(
1
)
.
T
h
is
r
esu
lt
s
in
2
5
5
f
ea
tu
r
e
s
.
T
h
er
ef
o
r
e,
o
u
r
f
u
ll f
e
a
tu
r
e
v
ec
to
r
w
ill i
n
cl
u
d
e
2
8
2
f
ea
tu
r
es.
(
)
=
∑
2
(
,
)
∞
=
1
(
1
)
w
h
er
e
th
e
i
s
W
P
T
o
f
s
ig
n
al
,
is
t
h
e
S
u
b
b
an
d
f
r
eq
u
en
c
y
i
n
d
ex
a
n
d
is
th
e
n
u
m
b
er
o
f
W
av
ele
t
co
ef
f
icie
n
t
s
in
t
h
e
lt
h
S
u
b
b
an
d
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
9
,
No
.
4
,
A
u
g
u
s
t 2
0
1
9
:
2
9
4
1
-
29
4
9
2944
W
PT
p
r
o
v
id
es
a
h
i
g
h
d
i
m
en
s
io
n
al
f
ea
t
u
r
e
v
ec
to
r
th
u
s
p
r
o
v
id
in
g
m
o
r
e
i
n
f
o
r
m
ati
o
n
ab
o
u
t
th
e
i
m
a
g
es
[
30
]
.
T
h
e
v
ar
iab
ilit
y
i
n
t
h
e
te
x
t
u
r
e
o
f
a
H
is
t
o
-
p
ath
o
lo
g
ical
i
m
a
g
e
ap
p
ea
r
s
to
b
e
w
h
at
m
o
s
t
s
ep
ar
ates
m
al
ig
n
a
n
t
m
elan
o
m
a
f
r
o
m
b
en
i
g
n
n
e
v
i,
t
h
er
ef
o
r
e
th
e
b
est
ap
p
r
o
ac
h
at
f
ea
t
u
r
e
ex
tr
ac
tio
n
lev
e
l
w
o
u
ld
b
e
to
r
etain
as
m
u
c
h
o
f
th
e
d
ata
v
ar
iab
ilit
y
as
p
o
s
s
ib
le
[3
1
]
.
T
h
is
is
ac
h
ie
v
ed
b
y
W
PT
.
W
av
elet
p
ac
k
et
an
al
y
s
is
lo
o
k
s
at
th
ese
c
h
an
g
e
s
o
v
er
d
if
f
er
e
n
t
s
ca
les
w
h
ic
h
s
h
o
u
ld
d
escr
ib
e
th
e
w
h
o
le
les
io
n
p
r
o
p
er
ties
s
u
ch
as tex
t
u
r
e,
co
lo
r
,
an
d
lo
ca
l c
h
an
g
e
s
li
k
e
g
r
an
u
lar
it
y
.
2
.
5
.
F
ea
t
ure
s
elec
t
io
n
Sin
ce
t
h
e
n
u
m
b
er
o
f
f
ea
t
u
r
es
ex
tr
ac
ted
f
r
o
m
t
h
e
i
m
ag
e
is
h
i
g
h
a
n
d
to
ch
o
o
s
e
t
h
e
m
o
s
t
r
elev
an
t
f
ea
t
u
r
es
th
a
t
w
o
u
ld
i
m
p
r
o
v
e
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
cla
s
s
if
ier
,
it
i
s
i
m
p
o
r
tan
t
to
u
s
e
an
o
p
ti
m
al
f
ea
tu
r
e
s
elec
tio
n
m
et
h
o
d
.
So
m
e
o
f
th
e
p
o
p
u
lar
f
ea
tu
r
e
s
s
elec
tio
n
m
et
h
o
d
s
ar
e
Seq
u
e
n
tial
F
o
r
w
ar
d
Selectio
n
(
SF
S)
[
1
6
]
,
Seq
u
en
tia
l
B
ac
k
w
ar
d
Select
io
n
(
SB
S)
[
3
2
]
,
Gen
etic
alg
o
r
it
h
m
s
[
3
3
]
an
d
P
ar
ticle
S
w
ar
m
Op
ti
m
izatio
n
(
P
SO)
.
I
n
ad
d
iti
o
n
to
f
ea
t
u
r
e
s
elec
tio
n
[
3
4
,
3
5
]
,
P
SO
h
as
b
ee
n
u
s
ed
f
o
r
t
h
e
o
p
tim
izatio
n
SVM
p
ar
am
eter
s
[
3
6
]
,
r
a
d
ial
b
asis
f
u
n
ctio
n
ex
tr
e
m
e
lear
n
i
n
g
m
ac
h
in
e
p
ar
a
m
eter
s
[
3
7
]
,
an
d
f
o
r
b
o
th
f
ea
tu
r
e
s
elec
tio
n
a
n
d
p
ar
am
e
ter
o
p
ti
m
izatio
n
[
3
8
-
4
2
]
.
P
SO
is
a
p
o
p
u
latio
n
-
b
ased
s
t
o
ch
asti
c
o
p
ti
m
iza
tio
n
tec
h
n
iq
u
e
th
at
m
i
m
ics
t
h
e
m
o
v
e
m
en
t
o
f
s
w
ar
m
s
an
d
is
in
s
p
ir
ed
b
y
s
o
cial
b
eh
av
io
r
o
f
b
i
r
d
s
o
r
f
is
h
e
s
.
I
t
w
o
r
k
s
b
y
h
a
v
i
n
g
a
p
o
p
u
latio
n
(
ca
lled
a
s
w
ar
m
)
o
f
ca
n
d
id
ate
s
o
lu
tio
n
s
(
ca
lled
p
ar
ticles).
E
ac
h
p
ar
ticle
is
m
o
v
ed
ar
o
u
n
d
in
th
e
s
ea
r
ch
-
s
p
ac
e
ac
co
r
d
in
g
to
(2
)
an
d
(
3)
g
u
id
ed
b
y
its
o
w
n
b
e
s
t
k
n
o
w
n
p
o
s
i
tio
n
in
th
e
s
ea
r
ch
-
s
p
ac
e
as
w
ell
as
t
h
e
en
tire
s
w
ar
m
's
b
est
k
n
o
w
n
p
o
s
itio
n
an
d
its
o
w
n
v
e
lo
cit
y
(
b
o
u
n
d
e
d
w
it
h
a
m
a
x
i
m
u
m
v
al
u
e
).
T
h
e
p
r
o
ce
s
s
is
iter
ati
v
e.
E
ac
h
iter
atio
n
,
,
an
d
ar
e
u
p
d
ated
.
T
h
er
ef
o
r
e,
w
h
e
n
i
m
p
r
o
v
ed
p
o
s
itio
n
s
ar
e
b
ein
g
d
is
co
v
er
ed
th
ese
w
il
l th
e
n
co
m
e
to
g
u
id
e
th
e
m
o
v
e
m
en
t
s
o
f
t
h
e
s
w
ar
m
.
+
1
=
+
1
1
(
−
)
+
2
2
(
−
)
(
2
)
+
1
=
+
+
1
(
3
)
W
h
er
e
k
th
e
cu
r
r
en
t
g
en
er
ati
o
n
(
i
ter
atio
n
)
,
1
an
d
2
ar
e
p
er
s
o
n
al
an
d
s
o
cial
lear
n
in
g
f
ac
to
r
s
an
d
ar
e
tak
en
h
er
e
as p
o
s
itiv
e
co
n
s
ta
n
ts
,
1
an
d
2
ar
e
r
an
d
o
m
n
u
m
b
er
s
f
r
o
m
t
h
e
in
ter
v
al
[
0
,
1
]
.
I
n
[
1
6
]
,
f
ea
tu
r
es
w
er
e
s
elec
te
d
u
s
in
g
th
e
S
FS
m
e
th
o
d
.
T
h
e
u
s
e
o
f
SF
S
r
es
u
lted
in
a
co
n
s
id
er
ab
l
y
i
m
p
r
o
v
ed
clas
s
i
f
icatio
n
r
ate.
I
n
th
is
s
tu
d
y
w
e
u
s
e
P
SO
to
s
elec
t
t
h
e
m
o
s
t
r
ele
v
a
n
t
f
ea
t
u
r
es
a
n
d
o
p
ti
m
ize
th
e
cla
s
s
i
f
ier
p
ar
a
m
eter
s
(
S
VM
)
th
r
o
u
g
h
t
h
e
u
s
e
o
f
a
P
SO
-
S
VM
al
g
o
r
ith
m
as
t
y
p
e
o
f
lear
n
in
g
f
o
r
th
e
class
if
ier
[
4
2
]
.
T
h
e
s
tep
s
o
f
th
e
p
r
o
p
o
s
ed
s
y
s
te
m
a
r
e
d
is
cu
s
s
ed
b
el
o
w
.
T
h
e
y
ar
e
also
s
u
m
m
ar
ized
in
th
e
f
lo
w
c
h
ar
t o
f
F
ig
u
r
e
2
.
P
SO
-
SVM
s
y
s
te
m
f
o
r
o
p
ti
m
al
s
elec
tio
n
o
f
f
ea
t
u
r
e
an
d
SVM
p
ar
am
eter
s
:
1)
I
n
itializatio
n
E
ac
h
p
ar
ticle
i
s
d
e
f
i
n
ed
as
an
ar
r
ay
w
it
h
t
w
o
ce
lls
f
o
r
th
e
S
VM
p
ar
a
m
eter
s
(
C
a
n
d
γ
)
a
n
d
2
8
2
ce
lls
co
r
r
esp
o
n
d
in
g
f
o
r
ea
ch
o
f
t
h
e
2
8
2
f
ea
tu
r
es.
T
h
e
f
ea
tu
r
e
ce
lls
co
n
tain
w
ei
g
h
ts
i
n
r
an
g
e
o
f
0
to
1
w
h
ic
h
r
ef
lec
t
th
e
s
i
g
n
i
f
ican
ce
o
f
co
r
r
esp
o
n
d
in
g
f
ea
t
u
r
es.
I
n
f
ir
s
t p
o
p
u
latio
n
,
P
SO f
ir
s
t p
ar
ticle
is
i
n
itia
liz
ed
u
s
i
n
g
SF
S
w
h
ile
t
h
e
o
th
er
p
ar
ticles ar
e
in
i
tializ
ed
b
y
r
an
d
o
m
n
u
m
b
er
s
r
an
g
i
n
g
f
r
o
m
0
to
1
.
2)
Selectio
n
o
f
f
ea
t
u
r
es i
n
a
p
ar
ticle
I
n
ea
ch
iter
atio
n
,
f
ea
t
u
r
es
with
w
e
ig
h
t
s
m
o
r
e
th
a
n
a
s
p
e
cif
ied
t
h
r
es
h
o
ld
(
h
er
e
0
.
5
)
ar
e
ch
o
s
en
.
A
ll t
h
e
p
ar
ticles ar
e
th
e
n
s
e
n
t t
o
SVM
f
o
r
ca
lcu
lati
n
g
th
eir
ac
cu
r
ac
y
b
ased
o
n
th
e
c
h
o
s
e
n
f
e
atu
r
es.
3)
Fit
n
e
s
s
f
u
n
ctio
n
T
o
c
o
m
p
u
te
th
e
f
i
tn
e
s
s
(
ac
cu
r
ac
y
)
co
r
r
esp
o
n
d
in
g
to
ea
c
h
p
ar
ticle;
a
S
VM
i
s
lo
ad
ed
w
it
h
t
h
e
s
elec
te
d
f
ea
t
u
r
es
o
f
t
h
e
p
ar
ticle
to
g
eth
e
r
w
it
h
t
h
e
C
an
d
γ
p
ar
a
m
eter
s
.
C
r
o
s
s
v
alid
atio
n
i
s
u
s
ed
to
g
e
n
er
ate
tr
ain
i
n
g
a
n
d
v
alid
atio
n
s
et
s
.
T
h
e
S
VM
i
s
tr
ain
ed
on
th
e
tr
ai
n
in
g
s
et
.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
SVM
is
t
h
e
n
v
er
if
ied
on
th
e
v
ali
d
atio
n
s
et
a
n
d
th
e
co
r
r
esp
o
n
d
in
g
f
ea
tu
r
e
s
et
is
co
n
s
i
d
er
ed
as
th
e
f
i
tn
e
s
s
o
f
t
h
at
p
ar
ticle.
T
h
is
d
o
n
e
f
o
r
ev
er
y
p
ar
ticle.
Af
ter
ev
a
lu
at
in
g
t
h
e
ac
cu
r
ac
y
o
f
p
ar
ticles,
th
e
b
est
ac
c
u
r
ac
y
i
n
a
p
o
p
u
latio
n
is
co
n
s
id
er
ed
as
an
d
th
e
b
est
ac
cu
r
ac
y
i
n
th
e
h
is
to
r
y
o
f
ea
ch
p
ar
ticle
is
co
r
r
esp
o
n
d
in
g
to
th
at
p
ar
ticle.
T
h
e
p
ar
ticles
in
n
ex
t
p
o
p
u
latio
n
s
ar
e
g
en
er
ate
d
ac
co
r
d
in
g
to
E
q
u
atio
n
s
(
2
)
an
d
(
3
)
.
C
lear
l
y
,
in
th
e
f
ir
s
t
p
o
p
u
latio
n
an
d
w
i
l
l
b
e
s
a
m
e.
T
h
e
p
r
o
ce
s
s
is
s
to
p
p
ed
o
n
ce
th
e
m
ax
i
m
u
m
n
u
m
b
er
o
f
iter
atio
n
i
s
r
ea
ch
ec
d
.
T
h
e
p
ar
ticle
co
r
r
esp
o
n
d
in
g
to
w
ill
th
e
n
c
o
n
tain
t
h
e
b
est
f
ea
tu
r
e
s
an
d
th
e
b
est
SVM
p
ar
am
eter
s
.
T
h
is
is
u
s
ed
to
s
i
m
u
lta
n
eo
u
s
l
y
o
p
ti
m
ize
t
h
e
S
VM
p
ar
am
eter
s
(
C
a
n
d
γ
)
an
d
s
elec
t th
e
m
o
s
t r
ele
v
an
t
f
ea
t
u
r
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:
2
0
8
8
-
8708
P
S
O
-
S
V
M h
yb
r
id
s
ystem
fo
r
mela
n
o
ma
d
etec
tio
n
fr
o
m
h
is
to
-
p
a
th
o
lo
g
ica
l ima
g
es (
Ma
en
Ta
kru
r
i)
2945
Fig
u
r
e
2
.
A
b
lo
ck
d
iag
r
a
m
d
es
cr
ib
in
g
t
h
e
P
SO
-
SVM
s
y
s
te
m
2
.
6
.
Cla
s
s
if
ica
t
io
n
T
h
e
class
if
ier
u
s
ed
in
t
h
is
w
o
r
k
i
s
th
e
SVM
cla
s
s
i
f
ier
th
at
is
w
id
el
y
u
s
ed
d
u
e
t
o
its
h
ig
h
class
i
f
icatio
n
ac
cu
r
ac
y
a
n
d
ab
ilit
y
to
d
ea
l
w
it
h
h
ig
h
-
d
i
m
en
s
io
n
al
d
ata
[
3
2
,
4
3
].
A
f
ter
th
e
o
p
ti
m
al
SV
M
p
ar
am
eter
s
h
a
v
e
b
ee
n
s
elec
te
d
in
t
h
e
p
r
ev
io
u
s
s
tep
,
SVM
i
s
th
e
n
u
s
ed
to
c
lass
if
y
t
h
e
h
is
to
p
ath
o
lo
g
y
i
m
a
g
es
as
m
ali
g
n
an
t o
r
b
en
ig
n
.
3.
RE
SU
L
T
S
A
ND
D
I
SCU
SS
I
O
N
T
h
e
in
tr
o
d
u
ce
d
s
y
s
te
m
ab
ilit
y
to
d
ec
i
d
e
w
h
et
h
er
a
s
k
i
n
lesi
o
n
His
to
-
p
at
h
o
lo
g
ical
i
m
ag
e
i
s
b
en
ig
n
o
r
m
ali
g
n
a
n
t
is
e
v
al
u
ated
in
th
i
s
s
ec
tio
n
u
s
i
n
g
a
d
ataset
o
b
tain
ed
f
r
o
m
th
e
So
u
th
er
n
P
ath
o
lo
g
y
L
ab
o
r
ato
r
y
in
W
o
llo
n
g
o
n
g
NSW
,
Au
s
tr
alia.
I
t
in
clu
d
es
7
9
His
to
-
pa
t
h
o
lo
g
i
ca
l
i
m
ag
e
s
(
2
9
b
en
ig
n
i
m
a
g
e
s
an
d
5
0
m
ela
n
o
m
a
i
m
a
g
es).
T
w
o
ex
p
er
i
m
en
t
s
w
er
e
co
n
d
u
cted
.
T
h
ey
d
i
f
f
er
f
r
o
m
ea
ch
o
th
er
in
th
e
f
ea
tu
r
e
s
elec
t
io
n
m
et
h
o
d
e
m
p
lo
y
ed
a
n
d
th
e
w
a
y
t
h
e
o
p
ti
m
al
SVM
p
a
r
a
m
e
ter
s
w
er
e
p
ick
ed
(
C
a
n
d
γ
)
.
T
h
e
clas
s
i
f
ier
s
u
s
ed
in
all
o
f
th
e
ex
p
er
i
m
e
n
t
s
w
er
e
SVM
s
w
it
h
R
ad
ial
B
asis
F
u
n
ctio
n
K
er
n
el
[
4
4
]
im
p
le
m
e
n
ted
u
s
i
n
g
L
I
B
SVM
to
o
lb
o
x
f
o
r
Ma
t
L
ab
[
4
5
]
.
A
ll
th
e
al
g
o
r
ith
m
s
w
er
e
i
m
p
le
m
e
n
ted
u
s
i
n
g
M
A
T
L
A
B
R
2
0
1
3
b
.
6
0
%
o
f
t
h
e
i
m
a
g
es
w
er
e
u
s
ed
f
o
r
tr
ain
i
n
g
,
2
0
% f
o
r
v
al
id
atio
n
an
d
th
e
r
e
m
ain
in
g
2
0
%
w
er
e
u
s
ed
f
o
r
th
e
p
u
r
p
o
s
e
o
f
test
i
n
g
.
Fo
llo
w
i
n
g
th
e
i
m
a
g
e
clas
s
lab
elin
g
o
f
[4
6
]
,
th
e
im
a
g
es
t
h
at
w
er
e
co
n
f
ir
m
ed
b
y
t
h
e
p
at
h
o
lo
g
is
t
as
m
ela
n
o
m
a
i
m
ag
e
s
w
er
e
co
n
s
i
d
er
ed
as
n
eg
ativ
e
clas
s
i
m
a
g
e
s
,
w
h
ile
th
e
n
ev
u
s
o
n
e
s
w
er
e
tr
ea
ted
as
p
o
s
itiv
e
S
t
a
r
t
P
a
r
t
i
c
l
e
s
I
n
i
t
i
a
l
i
z
a
t
i
o
n
:
2
c
e
l
l
s
f
o
r
t
h
e
S
V
M
P
a
r
a
m
e
t
e
r
s
(
C
,
g
)
a
n
d
2
8
2
c
e
l
l
s
c
o
r
r
e
s
p
o
n
d
i
n
g
t
o
t
h
e
f
e
a
t
u
r
e
s
S
V
M
:
C
o
m
p
u
t
e
t
h
e
f
i
t
n
e
s
s
(
A
c
c
u
r
a
c
y
)
f
o
r
E
a
c
h
P
a
r
t
i
c
l
e
U
p
d
a
t
i
n
g
P
b
e
s
t
a
n
d
g
b
e
s
t
U
p
d
a
t
i
n
g
e
a
c
h
p
a
r
t
i
c
l
e
b
a
s
e
d
o
n
t
h
e
v
e
l
o
c
i
t
y
,
P
b
e
s
t
a
n
d
g
b
e
s
t
S
t
o
p
i
n
g
C
o
n
d
i
t
i
o
n
m
e
t
?
A
l
l
F
e
a
t
u
r
e
s
T
h
e
f
e
a
t
u
r
e
s
C
o
r
r
e
s
p
o
n
d
i
n
g
t
o
P
b
e
s
t
a
r
e
s
e
l
e
c
t
e
d
t
h
e
n
t
a
k
e
n
t
o
t
h
e
T
e
s
t
i
n
g
s
t
a
g
e
E
n
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
9
,
No
.
4
,
A
u
g
u
s
t 2
0
1
9
:
2
9
4
1
-
29
4
9
2946
class
i
m
a
g
es.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
alg
o
r
it
h
m
s
w
a
s
ev
alu
a
ted
b
y
co
m
p
u
t
in
g
t
h
e
Sen
s
i
tiv
it
y
(
SE
)
,
Sp
ec
if
icit
y
(
SP
)
an
d
A
cc
u
r
ac
y
(
AC
)
u
s
i
n
g
E
q
u
at
io
n
s
(
4
)
,
(
5
)
,
an
d
(
6
)
,
r
esp
ec
tiv
ely
.
=
(
+
)
100
(
4
)
=
(
+
)
100
(
5
)
=
(
+
)
(
+
+
+
)
100
(
6
)
w
h
er
e
TP
is
t
h
e
n
u
m
b
er
o
f
t
r
u
e
p
o
s
itiv
e
s
,
TN
is
t
h
e
n
u
m
b
er
o
f
tr
u
e
n
e
g
ati
v
es,
FN
i
s
t
h
e
n
u
m
b
er
o
f
f
al
s
e
n
eg
at
iv
e
s
,
an
d
FP
is
th
e
n
u
m
b
er
o
f
f
alse p
o
s
iti
v
es.
I
n
th
e
f
ir
s
t
ex
p
er
i
m
e
n
t,
SF
S
w
a
s
u
s
ed
to
ch
o
o
s
e
th
e
m
o
s
t
r
elev
an
t
f
ea
t
u
r
es
t
h
at
r
esu
lt
i
n
th
e
b
est
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
SVM.
5
-
f
o
ld
cr
o
s
s
v
alid
atio
n
w
a
s
u
s
ed
to
p
ick
th
e
b
est
R
B
F
k
er
n
el
p
ar
a
m
eter
s
(
C
=1
0
an
d
γ
=0
.
1
2
5
)
.
T
h
e
r
esu
lt
s
o
f
ex
p
er
i
m
e
n
t
1
ar
e
s
h
o
w
n
i
n
T
ab
le
1.
T
h
e
m
et
h
o
d
is
d
e
n
o
ted
in
T
a
b
le
1
by
W
P
T
-
SFS
-
SVM
i
n
d
icatin
g
t
h
at
W
P
T
f
ea
tu
r
es,
SV
M
clas
s
if
ier
an
d
SF
S
f
ea
t
u
r
e
s
elec
tio
n
m
eth
o
d
ar
e
u
s
ed
b
y
t
h
is
m
et
h
o
d
.
T
ab
le
1
.
R
esu
lts
o
f
W
PT
-
SFS
-
SVM
(
E
x
p
er
i
m
e
n
t 1
)
N
o
o
f
i
mag
e
s
S
e
n
si
t
i
v
i
t
y
S
p
e
c
i
f
i
c
i
t
y
A
c
c
u
r
a
c
y
79
8
3
.
6
7
0
.
7
7
7
.
4
P
SO
-
SVM
ar
r
an
g
e
m
en
t
w
a
s
u
s
ed
in
th
e
s
ec
o
n
d
ex
p
er
i
m
en
t
t
o
ch
o
o
s
e
th
e
m
o
s
t
r
elev
a
n
t
f
ea
tu
r
es
an
d
th
e
o
p
ti
m
al
R
B
F
k
er
n
el
p
ar
a
m
eter
s
th
at
o
p
ti
m
ize
th
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
S
VM
.
5
-
f
o
ld
cr
o
s
s
v
al
id
atio
n
w
a
s
u
s
ed
w
it
h
i
n
t
h
e
P
SO
-
SVM
ar
r
an
g
e
m
en
t.
T
h
e
r
es
u
lti
n
g
R
B
F
k
er
n
el
p
ar
a
m
eter
s
ar
e
C
=3
5
2
5
.
0
0
5
1
an
d
γ
=0
.
0
0
8
4
7
3
2
)
.
T
h
e
r
esu
lts
o
f
ex
p
er
i
m
en
t
2
ar
e
s
h
o
w
n
i
n
T
ab
le
2
.
T
h
e
m
e
th
o
d
is
d
en
o
ted
in
T
ab
le
2
by
W
P
T
-
P
SO
-
SVM.
T
ab
le
2
.
R
esu
lts
o
f
W
PT
-
P
SO
-
SVM
(E
x
p
er
i
m
e
n
t 2
)
N
o
o
f
i
mag
e
s
S
e
n
si
t
i
v
i
t
y
S
p
e
c
i
f
i
c
i
t
y
A
c
c
u
r
a
c
y
79
9
4
.
1
8
0
.
2
8
7
.
1
C
o
m
p
ar
in
g
t
h
e
r
es
u
lt
s
in
T
a
b
le
1
w
it
h
t
h
o
s
e
i
n
T
ab
le
2
,
it
ca
n
b
e
s
ee
n
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
P
SO
-
SVM
ar
r
a
n
g
e
m
e
n
t
f
o
r
f
ea
tu
r
e
s
elec
tio
n
an
d
S
VM
p
ar
a
m
eter
o
p
ti
m
izatio
n
i
s
co
n
s
i
d
er
ab
ly
b
etter
t
h
an
th
at
o
b
tain
ed
w
h
en
j
u
s
t
u
s
in
g
SF
S
f
o
r
f
ea
t
u
r
e
s
elec
tio
n
.
An
i
m
p
r
o
v
e
m
e
n
t o
f
1
0
%
ca
n
b
e
n
o
ticed
in
th
e
v
alu
e
s
o
f
th
e
Sen
s
iti
v
it
y
,
t
h
e
Sp
ec
i
f
icit
y
a
n
d
th
e
Acc
u
r
ac
y
.
T
h
is
h
i
g
h
l
ig
h
ts
th
e
s
u
cc
e
s
s
o
f
t
h
e
W
PT
-
P
SO
-
SVM
s
y
s
te
m
.
T
h
e
p
r
esen
ted
W
PT
-
P
SO
-
SV
M
s
y
s
te
m
r
es
u
lted
i
n
a
s
e
n
s
i
ti
v
it
y
o
f
9
4
.
1
%
a
s
p
ec
i
f
icit
y
o
f
8
0
.
2
%
an
d
an
ac
cu
r
ac
y
o
f
8
7
.
1
%.
T
h
e
o
b
tain
ed
s
en
s
it
iv
i
t
y
a
n
d
s
p
ec
if
icit
y
r
es
u
lts
a
r
e
co
m
p
ar
ab
le
to
th
o
s
e
o
b
tain
ed
b
y
Der
m
ato
lo
g
is
t
s
an
d
co
n
s
i
d
er
ab
ly
b
etter
th
a
n
t
h
o
s
e
o
b
t
ain
ed
b
y
les
s
tr
ai
n
ed
d
o
cto
r
s
as
s
ee
n
in
T
ab
le
3
(
q
u
o
ted
f
r
o
m
[
4
6
]
)
.
C
o
n
s
eq
u
e
n
tl
y
,
t
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
ca
n
b
e
co
n
s
id
er
ed
as
a
p
r
o
m
i
s
i
n
g
m
e
th
o
d
to
b
e
u
s
ed
b
y
p
at
h
o
lo
g
is
ts
f
o
r
s
k
i
n
ca
n
ce
r
d
iag
n
o
s
tic.
T
ab
le
3
.
Sen
s
itiv
it
y
a
n
d
s
p
ec
if
icit
y
s
tatis
t
ics [
4
4
]
S
e
n
si
t
i
v
i
t
y
S
p
e
c
i
f
i
c
i
t
y
Ex
p
e
r
t
s
9
0
%
5
9
%
D
e
r
mat
o
l
o
g
i
st
s
8
1
%
6
0
%
T
r
a
i
n
e
e
s
8
5
%
3
6
%
G
e
n
e
r
a
l
p
r
a
c
t
i
t
i
o
n
e
r
s
6
2
%
6
3
%
Fo
r
th
e
s
a
k
e
o
f
f
air
co
m
p
ar
is
io
n
b
et
w
ee
n
th
e
W
PT
-
P
SO
-
S
VM
f
r
a
m
e
w
o
r
k
a
n
d
t
h
e
m
eth
o
d
o
f
[
1
3
]
(
d
en
o
ted
h
er
as
SF
S
-
S
VM
)
,
wh
ich
s
h
o
w
ed
i
m
p
r
e
s
s
i
v
e
r
es
u
lt
s
o
n
a
s
m
aller
d
ata
s
et
(
4
2
i
m
g
es)
,
w
e
h
a
v
e
test
ed
th
e
S
FS
-
SVM
m
et
h
o
d
o
n
t
h
e
s
a
m
e
d
ata
s
et
u
s
ed
h
er
e
(
7
9
i
m
a
g
es)
as
r
ep
o
r
ted
in
T
ab
le
4
.
T
h
e
SF
S
-
SVM
m
et
h
o
d
u
s
ed
in
[
1
3
]
im
p
le
m
e
n
ted
SF
S
o
n
f
ea
t
u
r
es
o
b
tain
ed
f
r
o
m
th
e
h
i
s
to
g
r
a
m
a
n
d
th
e
c
o
-
o
cc
u
r
r
en
ce
m
atr
i
x
o
n
l
y
.
C
o
m
p
ar
i
n
g
t
h
e
r
es
u
lt
s
o
f
T
ab
le
4
(
SF
S
-
S
VM
)
w
it
h
th
o
s
e
in
T
ab
le
1
(
W
PT
-
SFS
-
SV
M
)
,
it
ca
n
b
e
s
ee
n
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:
2
0
8
8
-
8708
P
S
O
-
S
V
M h
yb
r
id
s
ystem
fo
r
mela
n
o
ma
d
etec
tio
n
fr
o
m
h
is
to
-
p
a
th
o
lo
g
ica
l ima
g
es (
Ma
en
Ta
kru
r
i)
2947
th
at
alt
h
o
u
g
h
b
o
th
m
et
h
o
d
s
u
s
e
t
h
e
s
a
m
e
f
ea
tu
r
e
s
elec
tio
n
m
et
h
o
d
(
SF
S),
t
h
e
u
s
e
o
f
ad
d
itio
n
al
f
ea
t
u
r
es
ex
tr
ac
ted
f
r
o
m
t
h
e
W
av
elet
P
ac
k
et
T
r
an
s
f
o
r
m
atio
n
(
W
PT
)
f
o
r
th
e
ca
s
e
o
f
W
P
T
-
SFS
-
SV
M
h
a
s
r
es
u
lted
i
n
a
b
etter
r
ec
o
g
n
itio
n
ac
c
u
r
a
c
y
,
s
en
s
iti
v
it
y
an
d
s
p
ec
i
f
icit
y
as
co
m
p
ar
ed
w
it
h
SF
S
-
SVM
.
On
t
h
e
o
th
er
h
a
n
d
,
t
h
e
r
esu
lts
d
is
p
la
y
ed
in
T
ab
le
2
,
w
h
ic
h
u
s
e
s
W
PT
-
P
SO
-
SV
M
,
ar
e
clea
r
ly
m
u
c
h
b
etter
th
an
th
o
s
e
i
n
T
ab
le
4
(
SF
S
-
SVM)
i
n
ter
m
s
o
f
ac
cu
r
ac
y
,
s
e
n
s
iti
v
it
y
a
n
d
s
p
ec
i
f
icit
y
w
h
ic
h
e
m
p
h
asize
s
t
h
e
ad
v
a
n
t
ag
es
o
b
tai
n
ed
f
r
o
m
u
s
i
n
g
th
e
P
SO
-
SVM
f
r
a
m
e
w
o
r
k
f
o
r
f
ea
t
u
r
e
s
elec
tio
n
an
d
S
VM
p
ar
am
eter
o
p
ti
m
izatio
n
.
Ta
b
le
4
.
R
esu
lts
o
f
SF
S
-
SVM
[
1
3
]
N
o
o
f
i
mag
e
s
S
e
n
si
t
i
v
i
t
y
S
p
e
c
i
f
i
c
i
t
y
A
c
c
u
r
a
c
y
79
7
1
.
4
7
6
.
9
7
5
.
0
Fig
u
r
e
3
s
h
o
w
s
t
h
e
a
v
er
ag
e
te
s
tin
g
ac
c
u
r
ac
y
o
v
er
1
0
0
0
r
u
n
s
o
f
th
e
t
h
r
ee
m
eth
o
d
s
in
t
h
e
f
o
r
m
o
f
a
b
ar
g
r
ap
h
w
it
h
t
h
e
co
n
f
id
en
c
e
in
ter
v
als
in
d
icate
d
.
I
t
ca
n
b
e
s
ee
n
t
h
at
i
n
ad
d
itio
n
to
t
h
e
i
m
p
r
o
v
ed
ac
cu
r
ac
y
ac
h
iv
ed
b
y
W
PT
-
P
SO
-
SVM,
it
p
r
o
v
id
es
m
o
r
e
s
tab
le
ac
cu
r
ac
y
p
er
f
o
r
m
an
ce
as
it
h
a
s
a
s
m
aller
er
r
o
r
r
an
g
e
as
co
m
p
ar
ed
w
it
h
W
PT
-
P
SO
-
SV
M
an
d
SF
S
-
S
VM
.
I
n
g
e
n
er
al
i
t
ca
n
b
e
s
aid
th
at,
f
o
r
th
e
s
a
m
e
av
er
ag
e
ac
cu
r
ac
y
,
a
class
if
icatio
n
s
y
s
te
m
w
it
h
n
a
r
r
o
w
er
co
n
f
id
en
ce
in
ter
v
al
ca
n
b
e
co
n
s
id
er
ed
as
m
o
r
e
r
eliab
l
e
as
its
r
esu
lts
w
ill
b
e
clo
s
er
to
th
e
ex
p
ec
ted
ac
cu
r
ac
y
.
Fig
u
r
e
3
.
Av
er
ag
e
clas
s
i
f
icatio
n
ac
cu
r
ac
y
f
o
r
w
it
h
co
n
f
id
en
c
e
in
ter
v
al
s
in
d
icate
d
4.
CO
NCLU
SI
O
N
T
h
is
p
ap
er
h
as
p
r
o
p
o
s
ed
an
au
to
m
ated
s
y
s
te
m
f
o
r
s
k
i
n
ca
n
ce
r
(
m
ela
n
o
m
a)
d
etec
tio
n
f
r
o
m
H
is
to
-
p
ath
o
lo
g
ical
i
m
a
g
es
s
a
m
p
led
f
r
o
m
m
icr
o
s
co
p
ic
s
lid
es
o
f
s
k
in
b
io
p
s
y
.
I
t
is
a
h
y
b
r
id
s
y
s
te
m
b
ased
o
n
P
ar
ticle
S
w
ar
m
Op
ti
m
iza
tio
n
a
n
d
S
u
p
p
o
r
t
Vec
to
r
Ma
ch
in
e
(
P
SO
-
SVM)
.
T
h
e
f
ea
t
u
r
es
u
s
ed
a
r
e
ex
tr
ac
ted
f
r
o
m
th
e
g
r
a
y
s
ca
le
i
m
ag
e
h
i
s
to
g
r
a
m
,
t
h
e
co
-
o
cc
u
r
r
en
ce
m
a
tr
ix
a
n
d
th
e
e
n
er
g
y
o
f
th
e
w
av
ele
t
co
ef
f
icie
n
t
s
r
es
u
lti
n
g
f
r
o
m
th
e
w
a
v
elet
p
ac
k
et
d
ec
o
m
p
o
s
itio
n
o
f
t
h
e
i
m
ag
e.
T
h
e
P
SO
-
SVM
s
y
s
te
m
s
elec
ts
t
h
e
b
est
f
ea
tu
r
e
s
et
a
n
d
th
e
b
est v
a
lu
e
s
f
o
r
th
e
S
VM
p
ar
a
m
eter
s
(
C
a
n
d
γ
)
th
at
o
p
ti
m
ize
th
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
SV
M
class
i
f
ier
.
E
v
alu
a
tio
n
s
h
av
e
b
ee
n
m
ad
e
o
n
a
d
ataset
o
b
tai
n
ed
f
r
o
m
th
e
So
u
t
h
er
n
P
ath
o
lo
g
y
L
a
b
o
r
ato
r
y
i
n
W
o
llo
n
g
o
n
g
NSW
,
Au
s
tr
alia.
I
t
in
clu
d
es
7
9
His
to
-
p
at
h
o
lo
g
i
ca
l
i
m
ag
e
s
(
2
9
b
en
ig
n
i
m
a
g
e
s
an
d
5
0
m
ela
n
o
m
a
i
m
a
g
es).
T
h
e
r
ec
o
g
n
itio
n
ac
cu
r
ac
y
o
b
tai
n
ed
b
y
t
h
e
P
SO
-
SV
M
s
y
s
te
m
is
8
7
.
7
.
1
%
w
h
er
ea
s
th
e
s
e
n
s
itiv
it
y
a
n
d
s
p
ec
if
icit
y
ar
e
9
4
.
1
%
an
d
8
0
.
2
%,
r
esp
ec
tiv
el
y
.
T
h
e
o
b
ta
in
ed
r
esu
lt
s
s
h
o
w
s
t
h
at
th
e
P
SO
-
SVM
s
y
s
te
m
o
u
tp
er
f
o
r
m
s
o
th
er
e
x
i
s
ti
n
g
s
y
s
te
m
s
.
T
h
e
s
en
s
iti
v
it
y
a
n
d
s
p
ec
if
icit
y
r
e
s
u
l
ts
ar
e
co
m
p
ar
ab
le
to
t
h
o
s
e
o
b
tain
e
d
b
y
d
er
m
a
to
lo
g
is
t
s
an
d
e
x
p
er
ts
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
,
as
a
r
esu
lt,
ca
n
b
e
t
h
o
u
g
h
t
o
f
a
s
an
en
co
u
r
ag
i
n
g
m
e
th
o
d
to
w
ar
d
s
th
e
au
to
m
atio
n
an
d
ea
r
l
y
d
etec
tio
n
o
f
s
k
i
n
ca
n
c
er
.
I
t
ca
n
s
er
v
e
f
o
r
m
ed
ical
p
r
ac
titi
o
n
er
s
,
af
ter
f
u
r
t
h
er
i
m
p
r
o
v
e
m
en
t
s
,
as
a
s
ec
o
n
d
o
p
in
io
n
in
t
h
e
s
k
i
n
ca
n
ce
r
d
iag
n
o
s
i
s
p
r
o
ce
s
s
.
Ho
w
e
v
er
,
m
u
c
h
m
o
r
e
Hi
s
to
-
p
ath
o
lo
g
ical
s
k
i
n
i
m
a
g
es
s
h
o
u
ld
b
e
o
b
tain
ed
f
r
o
m
h
o
s
p
ital
s
to
b
e
u
s
ed
f
o
r
tr
ai
n
i
n
g
an
d
t
esti
n
g
a
n
d
r
e
s
u
lt
i
n
m
o
r
e
r
eliab
le
s
y
s
te
m
.
I
n
f
u
t
u
r
e,
w
e
in
te
n
d
to
w
id
en
o
u
r
d
ata
b
ase
o
f
ca
r
ef
u
ll
y
lab
elled
Hi
s
t
o
-
p
ath
o
lo
g
ical
s
k
in
i
m
a
g
es
a
n
d
ex
p
lo
r
e
f
u
r
t
h
er
d
if
f
er
e
n
t
o
p
ti
m
al
f
ea
t
u
r
e
s
el
ec
tio
n
m
et
h
o
d
s
.
W
e
also
in
t
en
d
to
test
v
ar
io
u
s
class
i
f
icatio
n
tech
n
iq
u
e
s
s
u
c
h
N
eu
r
o
-
F
u
zz
y
alg
o
r
it
h
m
s
to
i
m
p
r
o
v
e
th
e
class
i
f
icat
io
n
ac
cu
r
a
c
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
9
,
No
.
4
,
A
u
g
u
s
t 2
0
1
9
:
2
9
4
1
-
29
4
9
2948
ACK
NO
WL
E
D
G
E
M
E
NT
S
T
h
is
w
o
r
k
h
as
b
ee
n
p
ar
tiall
y
s
u
p
p
o
r
ted
b
y
a
Seed
Gr
a
n
t p
r
o
v
id
ed
b
y
t
h
e
Am
er
ica
n
Un
i
v
er
s
it
y
o
f
R
a
s
A
l K
h
ai
m
a
h
.
RE
F
E
R
E
NC
E
S
[1
]
A
.
B.
O.
S.,
“
Ca
u
se
s o
f
d
e
a
th
2
0
1
0
,
”
Co
m
m
o
n
w
e
a
lt
h
o
f
A
u
stra
li
a
,
2
0
1
2
.
[2
]
A
.
I.
o
.
H.
a
.
W
e
lfare
,
“
Ca
n
c
e
r
i
n
A
u
stra
li
a
:
a
n
o
v
e
rv
ie
w
,
”
Ca
n
b
e
rra
,
A
u
stra
li
a
:
A
u
stra
li
a
n
In
stit
u
t
e
o
f
He
a
lt
h
a
n
d
W
e
l
f
a
re
,
2
0
0
6
.
[3
]
P
.
Ba
a
d
e
a
n
d
M
.
Co
o
ry
,
“
T
re
n
d
s
in
m
e
lan
o
m
a
m
o
rtalit
y
in
Au
stra
li
a
:
1
9
5
0
–
2
0
0
2
a
n
d
t
h
e
ir
i
m
p
l
ica
ti
o
n
s
f
o
r
m
e
lan
o
m
a
c
o
n
tro
l,
”
Au
str
a
li
a
n
a
n
d
Ne
w
Z
e
a
la
n
d
J
o
u
r
n
a
l
o
f
Pu
b
li
c
He
a
lt
h
,
v
o
l
.
2
9
,
p
p
.
3
8
3
-
3
8
6
,
2
0
0
7
.
[4
]
V
icto
r
ia,
C.
C.
O
.
,
“
S
k
in
c
a
n
c
e
r
fa
c
ts an
d
sta
ts
,”
2
0
1
2
.
A
v
a
il
a
b
le
:
h
tt
p
:/
/w
ww
.
su
n
s
m
a
rt.
c
o
m
.
a
u
/f
a
q
s/
fa
c
ts_
a
n
d
_
sta
ts
[5
]
A
u
stra
li
a
n
,
C.
C.
“
S
k
in
Ca
n
c
e
r
F
a
c
ts
a
n
d
F
ig
u
re
s
,”
2
0
1
0
.
A
v
a
il
a
b
le
h
tt
p
:
//
ww
w
.
c
a
n
c
e
r.
o
rg
.
a
u
/ca
n
c
e
rs
m
a
rtl
if
e
st
y
le/S
u
n
S
m
a
r
t/
S
k
in
c
a
n
c
e
rfa
c
tsa
n
d
f
ig
u
re
s.h
t
m
[6
]
J.
S
ik
o
rsk
i,
“
Id
e
n
ti
f
ica
ti
o
n
o
f
m
a
li
g
n
a
n
t
m
e
lan
o
m
a
b
y
w
a
v
e
let
a
n
a
ly
sis
,”
S
tu
d
e
n
t/
F
a
c
u
lt
y
Re
se
a
r
c
h
Da
y
,
CS
IS
,
P
a
c
e
Un
iv
e
rsity
,
20
04
.
[7
]
A
.
A
n
d
rio
n
,
e
t
a
l
.,
“
M
a
li
g
n
a
n
t
m
e
so
th
e
li
o
m
a
o
f
th
e
p
leu
ra
:
in
tero
b
se
rv
e
r
v
a
riab
il
it
y
,
”
J
.
Cli
n
.
Pa
th
o
l
,
v
o
l.
48
,
p
p
.
856
-
8
6
0
,
1
9
9
5
.
[8
]
S
.
M
.
Ism
a
il
,
e
t
a
l
.,
“
Ob
se
rv
e
r
v
a
riatio
n
in
h
isto
p
a
th
o
lo
g
ica
l
d
i
a
g
n
o
sis
a
n
d
g
ra
d
in
g
o
f
c
e
r
v
ica
l
in
trae
p
it
h
e
li
a
l
n
e
o
p
las
ia
,
”
Br.
M
e
d
.
J
.,
v
o
l
.
2
9
8
,
p
p
.
7
0
7
-
710
,
1
9
8
9
.
[9
]
M
.
A
.
M
a
h
m
o
u
d
,
e
t
a
l
.
,
“
T
h
e
Au
to
m
a
ti
c
Id
e
n
ti
f
ica
ti
o
n
o
f
M
e
lan
o
m
a
b
y
Wav
e
l
e
t
a
n
d
Cu
rv
e
let
A
n
a
ly
sis:
S
tu
d
y
Ba
se
d
o
n
Ne
u
ra
l
Ne
tw
o
rk
Clas
sif
ica
ti
o
n
,
”
Pro
c
e
e
d
in
g
s
o
f
t
h
e
El
e
v
e
n
th
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Hy
b
ri
d
In
telli
g
e
n
t
S
y
ste
ms
,
2
0
1
1
.
[1
0
]
M
.
A
.
M
a
h
m
o
u
d
,
e
t
a
l.
,
“
W
a
v
e
let
a
n
d
Cu
rv
e
let
A
n
a
l
y
sis
f
o
r
Au
to
m
a
ti
c
Id
e
n
ti
f
ica
ti
o
n
o
f
M
e
l
a
n
o
m
a
Ba
se
d
o
n
Ne
u
ra
l
Ne
t
w
o
rk
Clas
si
f
ic
a
ti
o
n
,
”
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
mp
u
ter
In
fo
rm
a
ti
o
n
S
y
ste
ms
a
n
d
I
n
d
u
stria
l
M
a
n
a
g
e
me
n
t
,
v
o
l.
5
,
p
p
.
6
0
6
-
6
1
4
,
2
0
1
3
.
[1
1
]
M
.
T
a
k
ru
ri,
e
t
a
l.
,
“
A
u
to
m
a
ti
c
Re
c
o
g
n
it
io
n
o
f
M
e
lan
o
m
a
Us
in
g
S
u
p
p
o
r
t
V
e
c
to
r
M
a
c
h
i
n
e
s:
A
S
tu
d
y
Ba
se
d
o
n
W
a
v
e
let,
Cu
rv
e
let
a
n
d
C
o
lo
r
F
e
a
tu
re
s,”
Pro
c
e
e
d
in
g
s
o
f
t
h
e
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
In
d
u
str
ia
l
A
u
t
o
ma
ti
o
n
,
In
fo
rm
a
t
io
n
a
n
d
Co
mm
u
n
ica
ti
o
n
s
T
e
c
h
n
o
lo
g
y
(
IAI
CT
2
0
1
4
)
,
2
0
1
4
.
[1
2
]
C.
D
e
m
ir
a
n
d
B.
Ye
n
e
r,
“
A
u
to
m
a
ted
c
a
n
c
e
r
d
iag
n
o
sis
b
a
se
d
o
n
h
isto
-
p
a
t
h
o
l
o
g
ica
l
im
a
g
e
s:
a
s
y
ste
m
a
ti
c
su
rv
e
y
,
”
T
e
c
h
n
ica
l
Re
p
o
rt,
D.o
.
C.
S
.
P
o
ly
tec
h
n
ic
In
stit
u
te,
a
n
d
E
d
it
o
r.
Re
n
ss
e
lae
r
P
o
ly
tec
h
n
ic
In
stit
u
te:
T
ro
y
,
NY
1
2
1
8
0
,
USA
,
2009
.
[1
3
]
M
.
A
.
M
a
h
m
o
u
d
,
e
t
a
l.
,
“
Clas
sifica
ti
o
n
o
f
M
a
li
g
n
a
n
t
M
e
lan
o
m
a
a
n
d
Be
n
ig
n
Ne
v
i
f
ro
m
S
k
in
L
e
s
io
n
s
Ba
se
d
o
n
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
i
n
e
,
”
Fi
ft
h
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
Co
mp
u
t
a
ti
o
n
a
l
In
telli
g
e
n
c
e
,
M
o
d
e
ll
in
g
a
n
d
S
imu
l
a
ti
o
n
(
CIM
S
im)
,
2
0
1
3
.
[1
4
]
T
.
Do
n
a
d
e
y
,
“
Bo
u
n
d
a
ry
d
e
tec
ti
o
n
o
f
b
lac
k
sk
in
t
u
m
o
rs
u
sin
g
a
n
a
d
a
p
ti
v
e
ra
d
ial
-
b
a
se
d
a
p
p
ro
a
c
h
,
”
S
PIE
M
e
d
ica
l
Ima
g
i
n
g
,
v
o
l.
3
3
7
9
,
p
p
.
8
1
0
-
8
1
6
,
2
0
0
0
.
[1
5
]
R.
P
.
Bra
u
n
,
e
t
a
l.
,
“
De
r
m
o
sc
o
p
y
o
f
P
ig
m
e
n
ted
Les
io
n
s:
A
v
a
lu
a
b
le
to
o
l
i
n
th
e
d
iag
n
o
sis
o
f
m
e
lan
o
m
a
,”
S
wiss
M
e
d
ica
l
W
e
e
k
ly
,
v
o
l/
issu
e
:
1
3
4
(7
-
8)
,
p
p
.
8
3
-
90
,
2
0
0
4
.
[1
6
]
M
.
A
.
M
a
h
m
o
u
d
a
n
d
A
.
A
.
Ju
m
a
i
ly
,
“
No
v
e
l
f
e
a
tu
re
e
x
tra
c
ti
o
n
m
e
th
o
d
o
l
o
g
y
b
a
se
d
o
n
h
isto
p
a
th
a
l
o
g
ica
l
im
a
g
e
s
a
n
d
su
b
se
q
u
e
n
t
c
las
sif
ic
a
ti
o
n
b
y
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
i
n
e
,
”
Pro
c
e
e
d
in
g
s
o
f
th
e
W
o
rld
S
y
mp
o
siu
m
o
n
Co
mp
u
ter
Ap
p
li
c
a
ti
o
n
s &
Res
e
a
rc
h
(
W
S
CA
R)
,
2
0
1
4
.
[1
7
]
M.
A.
M
a
h
m
o
u
d
a
n
d
A
.
A
.
Ju
m
a
il
y
,
“
A
H
y
b
rid
S
y
ste
m
f
o
r
S
k
in
L
e
sio
n
De
tec
ti
o
n
:
Ba
se
d
o
n
G
a
b
o
r
W
a
v
e
let
a
n
d
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
i
n
e
,
”
Pro
c
e
e
d
in
g
s o
f
th
e
7
th
I
n
ter
n
a
ti
o
n
a
l
C
o
n
g
re
ss
o
n
Ima
g
e
a
n
d
S
ig
n
a
l
Pr
o
c
e
ss
in
g
CIS
P’1
4
and
7
t
h
I
n
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
Bi
o
M
e
d
ica
l
E
n
g
i
n
e
e
rin
g
a
n
d
In
fo
rm
a
t
ics
BM
EI'1
4
,
2
0
1
4
.
[1
8
]
M
.
Ja
y
a
m
a
n
m
a
d
h
a
ra
o
,
e
t
a
l.
,
“
Im
p
u
lse
No
ise
Re
m
o
v
a
l
f
ro
m
Dig
it
a
l
Im
a
g
e
s
-
A
Co
m
p
u
tatio
n
a
l
Hy
b
rid
A
p
p
ro
a
c
h
,
”
Glo
b
a
l
J
o
u
rn
a
l
o
f
Co
m
p
u
ter
S
c
ie
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
Gr
a
p
h
ics
&
V
isio
n
,
v
o
l.
1
3
,
2
0
1
3
.
[1
9
]
A
.
K.
Ja
in
,
“
F
u
n
d
a
m
e
n
tals o
f
Dig
it
a
l
Im
a
g
e
P
ro
c
e
ss
in
g
,
”
En
g
lew
o
o
d
Cli
f
f
s,
1
9
8
9
.
[2
0
]
A
.
Ja
in
a
n
d
S
.
B
h
a
tt
a
c
h
a
rjee
,
“
A
d
d
re
ss
b
lo
c
k
lo
c
a
ti
o
n
o
n
e
n
v
e
lo
p
e
s
u
sin
g
g
a
b
o
r
f
il
ters
,”
V
ol
/i
ss
u
e
:
25
(
12
)
,
1
9
9
2
.
[2
1
]
A
.
Ja
in
,
e
t
a
l.
,
“
Ob
jec
t
d
e
tec
ti
o
n
u
sin
g
g
a
b
o
r
f
il
ters
,
”
Pa
tt
e
rn
Rec
o
g
n
it
io
n
,
v
o
l.
3
0
,
p
p
.
2
9
5
-
3
0
9
,
1
9
9
7
.
[2
2
]
Q
.
L
i,
“
D
a
rk
li
n
e
d
e
tec
ti
o
n
w
it
h
li
n
e
w
id
th
e
x
trac
ti
o
n
,
”
1
5
th
IEE
E
In
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
Ima
g
e
Pro
c
e
ss
in
g
,
ICIP
,
2
0
0
8
.
[2
3
]
A
.
Cri
m
in
isi,
e
t
a
l
.
,
“
Re
g
io
n
f
il
li
n
g
a
n
d
o
b
jec
t
re
m
o
v
a
l
b
y
e
x
e
m
p
lar
-
b
a
se
d
im
a
g
e
in
p
a
in
ti
n
g
,
”
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
Im
a
g
e
Pr
o
c
e
ss
in
g
,
v
o
l
.
1
3
,
p
p
.
1
2
0
0
-
1
2
1
2
,
2
0
0
4
.
[2
4
]
M
.
R.
S
w
e
e
t,
“A
d
a
p
ti
v
e
a
n
d
re
c
u
rsiv
e
m
e
d
ian
f
il
terin
g
.”
[2
5
]
O.
R.
V
i
n
c
e
n
t
a
n
d
O.
F
o
l
o
ru
n
so
,
“
A
De
sc
rip
ti
v
e
A
l
g
o
rit
h
m
f
o
r
S
o
b
e
l
Im
a
g
e
Ed
g
e
De
t
e
c
ti
o
n
,
”
Pro
c
e
e
d
in
g
s
o
f
In
fo
rm
in
g
S
c
ien
c
e
&
IT
Ed
u
c
a
t
io
n
Co
n
fer
e
n
c
e
(
In
S
IT
E)
,
2
0
0
9
.
[2
6
]
I.
S
o
b
e
l
a
n
d
G
.
F
e
ld
m
a
n
,
“
A
3
×
3
iso
tro
p
ic
g
ra
d
ien
t
o
p
e
ra
to
r
f
o
r
im
a
g
e
p
ro
c
e
ss
in
g
,”
P
re
se
n
ted
a
t
a
talk
a
t
th
e
S
tan
f
o
rd
A
rti
f
icia
l
P
ro
jec
t
,
1
9
6
8
.
[2
7
]
A
.
M
e
rrit
t,
e
t
a
l.
,
“
P
a
ra
ll
e
l
Ed
g
e
De
tec
ti
o
n
,
”
2
0
1
1
.
[2
8
]
Q.
A
b
b
a
s,
“
Les
io
n
b
o
rd
e
r
d
e
tec
ti
o
n
in
d
e
rm
o
sc
o
p
y
i
m
a
g
e
s
u
sin
g
d
y
n
a
m
i
c
p
ro
g
ra
m
m
in
g
,
”
S
k
in
Res
e
a
rc
h
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l.
1
7
,
p
p
.
9
1
-
1
0
0
,
2
0
1
1
.
[2
9
]
I
.
Be
n
y
a
h
ia,
e
t
a
l.
,
“
Ev
a
lu
a
ti
o
n
o
f
th
e
M
e
d
ica
l
Im
a
g
e
Co
m
p
re
ss
i
o
n
u
sin
g
W
a
v
e
let
P
a
c
k
e
t
T
r
a
n
sfo
rm
a
n
d
S
P
IHT
Co
d
i
n
g
,
”
In
ter
n
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
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
ol
/i
ss
u
e
:
8
(
4
)
,
p
p
.
2
1
3
9
-
2
1
4
7
,
2
0
1
8
.
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:
2
0
8
8
-
8708
P
S
O
-
S
V
M h
yb
r
id
s
ystem
fo
r
mela
n
o
ma
d
etec
tio
n
fr
o
m
h
is
to
-
p
a
th
o
lo
g
ica
l ima
g
es (
Ma
en
Ta
kru
r
i)
2949
[3
0
]
R.
N.
Kh
u
sh
a
b
a
,
e
t
a
l.
,
“
No
v
e
l
F
e
a
tu
re
Ex
trac
ti
o
n
M
e
th
o
d
b
a
se
d
o
n
F
u
z
z
y
En
tro
p
y
a
n
d
W
a
v
e
let
P
a
c
k
e
t
T
ra
n
s
f
o
r
m
f
o
r
M
y
o
e
lec
tri
c
Co
n
tro
l
,
”
7
th
In
ter
n
a
t
io
n
a
l
S
y
mp
o
siu
m
o
n
C
o
mm
u
n
ic
a
ti
o
n
s
a
n
d
In
fo
rm
a
t
io
n
T
e
c
h
n
o
lo
g
ies
IS
CIT
2
0
0
7
,
S
y
d
n
e
y
,
Au
stra
li
a
,
p
p
.
3
5
2
-
3
5
7
,
2
0
0
7
.
[3
1
]
J.
S
ik
o
rsk
i,
“
Id
e
n
ti
f
ica
ti
o
n
o
f
m
a
li
g
n
a
n
t
m
e
lan
o
m
a
b
y
w
a
v
e
let
a
n
a
ly
sis,
”
S
tu
d
e
n
t/
F
a
c
u
lt
y
Re
se
a
r
c
h
Da
y
,
CS
IS
,
P
a
c
e
Un
iv
e
rsity
,
2
0
0
4
.
[3
2
]
B.
S
c
h
o
lk
o
p
f
,
e
t
a
l
.,
“K
e
rn
e
l
M
e
t
h
o
d
s
in
c
o
m
p
u
tatio
n
a
l
b
i
o
lo
g
y
,
”
M
IT
P
re
ss
,
2
0
0
4
.
[3
3
]
M
.
P
e
i
,
e
t
a
l.
,
“
F
e
a
tu
re
e
x
trac
ti
o
n
u
sin
g
g
e
n
e
ti
c
a
lg
o
rit
h
m
s,”
Pro
c
e
e
d
in
g
o
f
In
ter
n
a
ti
o
n
a
l
S
y
m
p
o
si
u
m
o
n
In
telli
g
e
n
t
Da
ta
E
n
g
in
e
e
rin
g
a
n
d
L
e
a
rn
i
n
g
9
8
(
IDEA
L
9
8
)
,
Ho
n
g
K
o
n
g
,
1
9
9
8
.
[3
4
]
H.
C.
Ya
n
g
,
e
t
a
l.
,
“
Re
se
a
rc
h
in
t
o
a
F
e
a
tu
re
S
e
lec
ti
o
n
M
e
th
o
d
f
o
r
Hy
p
e
rsp
e
c
tral
I
m
a
g
e
r
y
Us
in
g
P
S
O
a
n
d
S
V
M
,
”
J
o
u
rn
a
l
o
f
Ch
i
n
a
Un
ive
rs
it
y
o
f
M
i
n
in
g
&
T
e
c
h
n
o
lo
g
y
,
v
o
l
/i
ss
u
e
:
17
(
4
)
,
p
p
.
0
4
7
3
-
0
4
7
8
,
2
0
0
7
.
[3
5
]
C
.
J
.
T
u
,
e
t
a
l.
,
“
F
e
a
t
u
re
S
e
lec
ti
o
n
u
sin
g
P
S
O
-
S
V
M
,
”
IAE
NG I
n
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
Co
mp
u
ter
S
c
ie
n
c
e
,
2
0
0
7
.
[3
6
]
X
.
Z
h
a
n
g
a
n
d
Y
.
G
u
o
,
“
Op
ti
m
iz
a
ti
o
n
o
f
S
VM
P
a
ra
m
e
ter
s
Ba
se
d
o
n
P
S
O
A
lg
o
rit
h
m
,
”
Fi
ft
h
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Na
t
u
ra
l
Co
mp
u
t
a
t
io
n
,
2
0
0
9
.
[3
7
]
K
.
A
n
a
m
a
n
d
A
.
A
.
Ju
m
a
il
y
,
“
Op
ti
m
ize
d
K
e
rn
e
l
Ex
tre
m
e
Lea
rn
in
g
M
a
c
h
in
e
f
o
r
M
y
o
e
le
c
tri
c
P
a
tt
e
rn
Re
c
o
g
n
it
io
n
,
”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
E
lec
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
ol
/i
ss
u
e
:
8
(
1
),
p
p
.
4
8
3
-
4
9
6
,
2
0
1
8
.
[3
8
]
S
.
L
in
,
e
t
a
l
.
,
“
P
a
ra
m
e
ter
d
e
ter
m
i
n
a
ti
o
n
o
f
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
a
n
d
f
e
a
tu
re
se
lec
ti
o
n
u
sin
g
sim
u
late
d
a
n
n
e
a
li
n
g
a
p
p
ro
a
c
h
,
”
Ap
p
l.
S
o
ft
Co
m
p
u
t
.
,
v
o
l/
issu
e
:
8
(4
)
,
p
p
.
1
5
0
5
-
1
5
1
2
,
2
0
0
8
.
[3
9
]
S
.
W
.
L
in
,
e
t
a
l.
,
“
P
a
rti
c
le
sw
a
r
m
o
p
ti
m
iza
ti
o
n
f
o
r
p
a
ra
m
e
t
e
r
d
e
ter
m
in
a
ti
o
n
a
n
d
f
e
a
tu
re
s
e
lec
ti
o
n
o
f
su
p
p
o
r
t
v
e
c
to
r
m
a
c
h
in
e
s
,
”
Exp
e
rt S
y
ste
ms
wit
h
A
p
p
li
c
a
ti
o
n
s
,
v
o
l/
issu
e
:
3
5
(4
)
,
p
p
.
1
8
1
7
-
1
8
2
4
,
2
0
0
8
.
[4
0
]
Hu
a
n
g
C
.
L
.
a
n
d
D
u
n
J
.
F
.
,
“
A
d
istri
b
u
ted
P
S
O
-
S
VM
h
y
b
rid
sy
st
e
m
w
it
h
f
e
a
tu
re
se
lec
ti
o
n
a
n
d
p
a
ra
m
e
ter
o
p
ti
m
iza
ti
o
n
,
”
A
p
p
l
ied
S
o
ft
Co
mp
u
ti
n
g
,
v
o
l
.
8,
p
p
.
1
3
8
1
-
1
3
9
1
,
2
0
0
8
.
[4
1
]
Y
.
M
a
a
li
a
n
d
A
.
A
.
Ju
m
a
il
y
,
“
S
e
l
f
-
a
d
v
isin
g
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
,
”
Kn
o
wled
g
e
-
B
a
se
d
S
y
ste
ms
,
v
o
l.
52
,
p
p
.
2
1
4
-
2
2
2
,
2
0
1
3
.
[4
2
]
Y
.
M
a
a
li
a
n
d
A
.
A
.
Ju
m
a
il
y
,
“
Hie
ra
rc
h
ica
l
P
a
ra
ll
e
l
P
S
O
-
S
V
M
Ba
se
d
S
u
b
jec
t
In
d
e
p
e
n
d
e
n
t
S
lee
p
A
p
n
e
a
Clas
sif
ic
a
ti
o
n
,
”
L
e
c
tu
re
No
tes
i
n
Co
m
p
u
ter
S
c
ien
c
e
(
L
NCS
)
o
n
Ne
u
ra
l
I
n
fo
rm
a
ti
o
n
Pro
c
e
ss
in
g
(
ICONIP)
Vo
l.
7
6
6
6
,
E
d
it
o
rs
Hu
a
n
g
,
T
in
g
we
n
a
n
d
Z
e
n
g
,
Z
h
ig
a
n
g
a
n
d
L
i,
C
h
u
a
n
d
o
n
g
a
n
d
L
e
u
n
g
,
Ch
i
S
in
g
,
S
p
rin
g
e
r
Ber
li
n
He
id
e
lb
e
rg
,
p
p
.
5
0
0
-
5
0
7
,
2
0
1
2
.
[4
3
]
L
.
P
.
W
a
n
g
,
e
t
a
l.
,
“
Clas
sif
ica
ti
o
n
u
si
n
g
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
s
w
it
h
g
ra
d
in
g
re
so
lu
ti
o
n
,
”
IEE
E
In
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
Gr
a
n
u
l
a
r Co
mp
u
ti
n
g
,
v
o
l.
2
,
p
p
.
6
6
6
-
6
7
0
,
2
0
0
5
.
[4
4
]
M
.
Ha
m
ian
e
a
n
d
F.
S
a
e
e
d
,
“
S
V
M
Clas
sif
ica
ti
o
n
o
f
M
RI
Bra
in
Im
a
g
e
s
f
o
r
Co
m
p
u
ter
-
A
ss
ist
e
d
Dia
g
n
o
sis,
”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
E
lec
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
ol
/i
ss
u
e
:
7
(
5
),
p
p
.
2
5
5
5
-
2
5
6
4
,
2
0
1
7
.
[4
5
]
C
.
C.
Ch
a
n
g
a
n
d
C
.
J.
L
in
,
“
L
IBS
V
M
:
A
L
ib
ra
r
y
f
o
r
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
i
n
e
s
A
CM
T
r
a
n
s
,”
In
tell.
S
y
st.
T
e
c
h
n
o
l
.,
v
ol
/i
ss
u
e
:
2
(
3
),
2
0
1
1
.
[4
6
]
S
.
W
.
M
e
n
z
ies
,
e
t
a
l
.
,
“
T
h
e
P
e
rf
o
rm
a
n
c
e
o
f
S
o
larSca
n
:
A
n
A
u
to
m
a
ted
De
r
m
o
sc
o
p
y
I
m
a
g
e
A
n
a
l
y
sis
In
stru
m
e
n
t
f
o
r
th
e
Dia
g
n
o
sis o
f
P
rim
a
r
y
M
e
lan
o
m
a
,
”
Arc
h
ive
s o
f
De
rm
a
to
lo
g
y
,
v
ol
/i
ss
u
e
:
1
4
1
(
11
)
,
p
p
.
1
3
8
8
-
1
3
9
6
,
2
0
0
5
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
M
a
e
n
Ta
k
r
u
r
i
re
c
e
iv
e
d
h
is
P
h
D
in
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
f
ro
m
Un
iv
e
rsit
y
o
f
Tec
h
n
o
lo
g
y
(U
T
S
)
,
S
y
d
n
e
y
in
2
0
1
0
.
He
a
lso
re
c
e
iv
e
d
h
is
BS
c
a
n
d
M
S
c
d
e
g
re
e
s
in
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
f
r
om
th
e
Un
iv
e
rsity
o
f
Jo
rd
a
n
.
Cu
rr
e
n
tl
y
,
h
e
is
a
n
A
ss
o
c
iat
e
P
ro
f
e
ss
o
r
a
n
d
th
e
C
h
a
irm
a
n
o
f
th
e
De
p
a
rtm
e
n
t
o
f
El
e
c
tri
c
a
l,
El
e
c
tro
n
ics
a
n
d
Co
m
m
u
n
ica
ti
o
n
s
En
g
i
n
e
e
rin
g
a
t
th
e
A
m
e
rica
n
Un
iv
e
rsity
of
Ra
s
A
l
Kh
a
i
m
a
h
(
A
UR
A
K)
.
He
h
a
s
a
w
id
e
sp
e
c
tru
m
o
f
r
e
se
a
rc
h
in
tere
sts
th
a
t
in
c
l
u
d
e
sig
n
a
l
p
ro
c
e
ss
in
g
a
n
d
d
a
ta
f
u
sio
n
,
e
stim
a
ti
o
n
th
e
o
ry
a
n
d
targ
e
t
tr
a
c
k
i
n
g
,
b
io
m
e
d
ic
a
l
s
y
ste
m
s
,
m
a
c
h
in
e
lea
rn
in
g
a
n
d
im
a
g
e
p
ro
c
e
ss
in
g
.
M
o
h
a
m
e
d
K
h
a
le
d
A
b
u
M
a
h
m
o
u
d
re
c
e
iv
e
d
h
is
Ba
c
h
e
lo
r
o
f
S
c
ien
c
e
d
e
g
re
e
in
T
e
lec
o
m
m
u
n
ica
ti
o
n
s
a
n
d
El
e
c
tro
n
ics
En
g
i
n
e
e
rin
g
f
ro
m
th
e
f
a
c
u
lt
y
o
f
e
n
g
in
e
e
rin
g
a
t
M
o
n
o
u
f
ia
Un
iv
e
rsit
y
,
Eg
y
p
t
1
9
6
9
a
n
d
h
is
M
a
ste
rs
o
f
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
f
ro
m
Ch
a
rles
S
tu
a
rt
Un
iv
e
rsity
,
A
u
stra
li
a
1
9
9
8
.
He
re
c
e
iv
e
d
h
is
P
h
D
f
ro
m
th
e
Un
iv
e
rsit
y
o
f
T
e
c
h
n
o
lo
g
y
,
S
y
d
n
e
y
(U
T
S
)
in
2
0
1
5
.
His
re
se
a
rc
h
in
tere
sts
a
re
c
e
n
tere
d
in
th
e
a
re
a
o
f
He
a
lt
h
T
e
c
h
n
o
l
o
g
y
a
n
d
S
k
in
C
a
n
c
e
r
(
m
e
lan
o
m
a
)
d
e
tec
ti
o
n
.
He
i
s
a
m
e
m
b
e
r
o
f
th
e
In
stit
u
ti
o
n
o
f
El
e
c
tri
c
a
l
a
n
d
El
e
c
tro
n
ic
E
n
g
in
e
e
rs
(IE
EE
)
a
n
d
th
e
In
sti
tu
ti
o
n
o
f
En
g
in
e
e
rs,
A
u
stra
li
a
.
Ad
e
l
Ali
Al
-
J
u
m
a
il
y
re
c
e
iv
e
d
h
is
P
h
D
in
A
rti
f
icia
l
In
telli
g
e
n
c
e
.
He
is
c
u
rre
n
tl
y
a
n
A
ss
o
c
iat
e
P
r
o
f
e
ss
o
r
in
F
a
c
u
lt
y
o
f
En
g
in
e
e
rin
g
a
n
d
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
l
o
g
y
a
t
th
e
Un
iv
e
rsity
o
f
Tec
h
n
o
lo
g
y
,
S
y
d
n
e
y
(U
T
S
).
His res
e
a
rc
h
in
tere
sts in
c
lu
d
e
Bi
o
m
e
d
ica
l;
in
telli
g
e
n
t
sy
ste
m
s; an
d
b
i
o
-
m
e
c
h
a
tro
n
ics
.
Evaluation Warning : The document was created with Spire.PDF for Python.