I
nte
rna
t
io
na
l J
o
urna
l o
f
I
nfo
rm
a
t
ics a
nd
Co
m
m
un
ica
t
io
n T
ec
hn
o
lo
g
y
(
I
J
-
I
CT
)
Vo
l.
1
4
,
No
.
3
,
Dec
em
b
er
20
2
5
,
p
p
.
8
3
7
~
8
4
4
I
SS
N:
2252
-
8
7
7
6
,
DOI
:
1
0
.
1
1
5
9
1
/iji
ct
.
v
1
4
i
3
.
pp
837
-
8
4
4
837
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ict.
ia
esco
r
e.
co
m
No
v
el multilevel l
o
ca
l bina
ry
t
ex
tur
e descrip
tor for
ora
l cancer
detec
tion
Vij
a
y
a
Ya
du
v
a
ns
hi
1
,
R
a
m
a
n
M
urug
a
n
2
1
P
i
mp
r
i
C
h
i
n
c
h
w
a
d
C
o
l
l
e
g
e
o
f
En
g
i
n
e
e
r
i
n
g
a
n
d
R
e
se
a
r
c
h
,
P
u
n
e
,
I
n
d
i
a
2
B
i
o
-
M
e
d
i
c
a
l
I
mag
i
n
g
La
b
(
B
I
O
M
I
L)
,
D
e
p
a
r
t
m
e
n
t
o
f
E
l
e
c
t
r
o
n
i
c
s
a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
En
g
i
n
e
e
r
i
n
g
,
N
a
t
i
o
n
a
l
I
n
st
i
t
u
t
e
o
f
Te
c
h
n
o
l
o
g
y
S
i
l
c
h
a
r
,
S
i
l
c
h
a
r
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
l 1
,
2
0
2
4
R
ev
is
ed
Ma
r
2
0
,
2
0
2
5
Acc
ep
ted
J
u
n
9
,
2
0
2
5
Ca
teg
o
rizin
g
tex
tu
re
m
e
d
ica
l
ima
g
e
s
is
a
n
e
x
ten
si
v
e
jo
b
in
m
o
st
o
f
th
e
fiel
d
s
o
f
c
o
m
p
u
ter
v
isio
n
,
p
a
tt
e
rn
re
c
o
g
n
it
io
n
a
n
d
b
i
o
m
e
d
ica
l
ima
g
in
g
.
F
o
r
t
h
e
p
a
st
fe
w
y
e
a
rs,
th
e
tex
t
u
re
a
n
a
l
y
sis
sy
ste
m
m
o
d
e
l,
e
sp
e
c
iall
y
fo
r
b
io
l
o
g
ica
l
ima
g
e
s,
h
a
s
b
e
e
n
b
ro
u
g
h
t
t
o
a
tt
e
n
ti
o
n
b
e
c
a
u
se
o
f
it
s
e
v
e
r
-
g
ro
wi
n
g
re
q
u
irem
e
n
ts
a
n
d
c
h
a
ra
c
teristics
.
Th
is
re
se
a
rc
h
sh
o
ws
it
s
n
o
v
e
lt
y
b
y
u
sin
g
a
m
u
lt
il
e
v
e
l
l
o
c
a
l
b
in
a
ry
tex
t
u
re
d
e
sc
rip
to
r
(
M
LBT
D)
a
l
g
o
rit
h
m
wit
h
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
(
S
VM
)
,
k
-
n
e
ar
e
s
t
n
eig
h
b
o
r
(
KNN
)
,
a
n
d
CT
Cla
ss
ifi
e
rs
t
o
in
v
e
stig
a
te
th
e
tex
t
u
re
fe
a
tu
re
s
o
f
th
e
o
ra
l
c
a
n
c
e
r
sa
m
p
les
.
Th
e
s
imu
latio
n
wo
rk
is
d
o
n
e
in
M
ATLAB
2021a
e
n
v
ir
o
n
m
e
n
t
b
y
e
m
p
l
o
y
in
g
th
e
M
LBT
D
a
lg
o
rit
h
m
.
A
M
e
n
d
e
ley
d
a
tas
e
t,
c
o
n
tain
i
n
g
8
9
o
ra
l
c
a
v
it
y
h
isto
p
a
th
o
l
o
g
ica
l
ima
g
e
s
a
n
d
4
3
9
OSCC
ima
g
e
s
i
n
1
0
0
x
m
a
g
n
ifi
c
a
ti
o
n
,
is
u
se
d
.
A
sta
ti
stica
l
c
o
m
p
a
ra
ti
v
e
stu
d
y
o
f
lo
c
a
l
b
in
a
r
y
p
a
tt
e
rn
(
LBP
)
a
n
d
M
LBT
D
w
it
h
li
n
e
a
r
S
VM,
KN
N,
CT
c
las
sifier
is
p
e
rfo
rm
e
d
i
n
w
h
ich
re
su
lt
s
s
h
o
w
th
e
b
e
t
ter
p
e
rfo
rm
a
n
c
e
o
f
M
LBT
D
a
n
d
li
n
e
a
r
S
VM
with
8
9
.
9
4
%
o
f
a
c
c
u
ra
c
y
a
n
d
b
y
a
p
p
ly
in
g
M
LBT
D
a
lg
o
r
it
h
m
o
v
e
r
9
0
.
5
7
%
a
c
c
u
ra
c
y
is o
b
tai
n
e
d
w
h
e
re
a
s
LBP
a
lg
o
rit
h
m
o
n
ly
p
r
o
v
i
d
e
s 8
6
.
1
6
%
o
f
a
c
c
u
ra
c
y
.
K
ey
w
o
r
d
s
:
C
las
s
if
ier
KNN
L
o
ca
l b
in
ar
y
p
atter
n
Mu
ltil
ev
el
lo
ca
l b
i
n
ar
y
p
atter
n
Or
al
ca
n
ce
r
SVM
T
ex
tu
r
e
an
al
y
s
is
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
:
R
am
an
Mu
r
u
g
a
n
Dep
ar
tm
en
t o
f
E
lectr
o
n
ics an
d
C
o
m
m
u
n
icatio
n
E
n
g
in
ee
r
i
n
g
,
Natio
n
al
I
n
s
titu
te
o
f
T
ec
h
n
o
l
o
g
y
Sil
ch
ar
Asaam
-
7
8
8
0
1
0
,
I
n
d
ia
E
m
ail:
Mu
r
u
g
an
.
r
m
n
@
ec
e.
n
it
s
.
ac
.
in
1.
I
NT
RO
D
UCT
I
O
N
Or
a
l c
an
ce
r
is
o
n
e
o
f
t
h
e
d
is
e
a
s
es t
h
at
is
k
n
o
w
n
t
o
s
p
r
e
ad
r
ap
id
l
y
all
o
v
e
r
t
h
e
g
lo
b
e.
T
h
e
r
e
a
r
e
r
o
u
g
h
l
y
1
7
7
,
7
5
7
f
at
alit
ies
t
h
at
ar
e
p
r
e
d
icte
d
t
o
b
e
ca
u
s
e
d
b
y
o
r
al
c
an
ce
r
,
a
n
d
th
e
r
e
a
r
e
3
5
4
,
8
6
4
n
e
w
c
ases
p
e
r
y
ea
r
[
1
]
.
Ho
w
ev
er
,
d
u
e
to
a
lac
k
o
f
i
n
f
o
r
m
ati
o
n
a
b
o
u
t
t
h
e
s
i
g
n
s
o
f
m
o
u
t
h
c
a
n
ce
r
,
t
h
e
m
aj
o
r
it
y
o
f
ca
s
es
a
r
e
d
is
c
o
v
er
e
d
i
n
th
e
l
ate
r
s
ta
g
es
o
f
t
h
e
d
is
e
as
e.
T
h
e
s
u
r
v
i
v
a
l
r
at
e
f
o
r
o
r
al
ca
n
ce
r
m
ay
r
a
n
g
e
f
r
o
m
7
5
t
o
9
0
p
er
ce
n
t
i
f
i
t
is
r
e
co
g
n
iz
e
d
i
n
its
ea
r
l
y
s
t
ag
es.
T
h
e
im
p
r
o
v
em
en
t
o
f
v
is
i
o
n
-
b
as
ed
a
d
j
u
n
ct
iv
e
t
ec
h
n
i
q
u
es
t
h
at
c
an
d
if
f
er
en
tia
te
b
et
we
en
o
r
al
an
d
p
o
t
e
n
tia
ll
y
m
al
ig
n
an
t
i
ll
n
ess
es
is
o
n
e
o
f
t
h
e
m
o
s
t
e
f
f
ec
ti
v
e
s
tr
ate
g
i
es
f
o
r
m
ed
ica
l
p
r
o
f
ess
i
o
n
a
ls
t
o
i
d
e
n
ti
f
y
m
o
u
t
h
ca
n
c
er
.
S
u
r
g
e
r
y
is
th
e
m
o
s
t
f
u
n
d
am
e
n
tal
k
i
n
d
o
f
m
ed
ic
al
th
e
r
a
p
y
b
ec
a
u
s
e
it
g
u
a
r
a
n
te
es
a
h
ig
h
li
k
eli
h
o
o
d
o
f
s
u
cc
ess
a
n
d
o
f
f
e
r
s
a
s
u
r
v
i
v
al
r
at
e
b
etw
ee
n
7
5
a
n
d
9
0
p
er
ce
n
t
i
n
t
h
e
e
ar
ly
p
h
ase
[
2
]
.
Des
p
it
e
th
is
,
s
i
x
t
y
-
f
iv
e
t
o
s
ix
t
y
-
f
iv
e
p
er
ce
n
t
o
f
ca
s
es
ar
e
d
is
c
o
v
e
r
e
d
i
n
th
e
la
te
s
ta
g
e,
wit
h
a
c
o
n
s
id
er
a
b
le
r
at
io
o
f
f
at
alit
ies
[
3
]
.
I
n
t
h
e
c
o
n
te
x
t
o
f
a
n
y
we
ll
n
ess
p
r
o
g
r
a
m
t
h
a
t
s
e
ek
s
to
u
n
d
e
r
s
ta
n
d
t
h
e
m
o
r
t
alit
y
r
a
ti
o
s
a
n
d
th
e
s
e
v
er
it
y
o
f
t
h
e
co
n
d
it
io
n
,
t
h
e
d
ia
g
n
o
s
is
o
f
o
r
al
c
an
ce
r
b
e
c
o
m
es
v
e
r
y
im
p
o
r
ta
n
t
.
Or
a
l
s
q
u
a
m
o
u
s
c
ell
ca
r
ci
n
o
m
a,
als
o
k
n
o
w
n
as
OS
C
C
,
is
r
es
p
o
n
s
i
b
le
f
o
r
m
o
r
e
t
h
an
n
i
n
e
ty
p
e
r
ce
n
t
o
f
all
o
c
c
u
r
r
en
ce
s
o
f
o
r
al
ca
n
c
er
,
in
c
lu
d
i
n
g
l
eu
k
o
p
la
k
i
a
a
n
d
er
y
t
h
r
o
p
l
ak
ia
,
wh
ic
h
a
r
e
p
r
e
ce
d
e
d
b
y
o
r
al
p
o
t
en
tial
ly
m
a
li
g
n
a
n
t
d
is
e
ases
(
OPM
D)
.
T
h
e
d
i
a
g
n
o
s
is
o
f
OP
MD
is
ass
o
ci
at
ed
wi
th
a
h
i
g
h
p
r
o
b
ab
ilit
y
o
f
f
at
al
c
o
n
v
er
s
i
o
n
,
as
w
ell
as
a
r
e
d
u
cti
o
n
in
t
h
e
f
at
ali
ty
r
a
ti
o
o
f
o
r
a
l
ca
n
ce
r
a
n
d
th
e
n
ee
d
t
o
p
a
y
att
e
n
ti
o
n
t
o
s
c
r
e
en
in
g
p
r
o
g
r
am
s
.
D
u
e
t
o
t
h
e
f
a
ct
t
h
at
th
ese
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
4
,
No
.
3
,
Dec
em
b
er
20
2
5
:
837
-
8
4
4
838
s
cr
ee
n
i
n
g
p
r
o
g
r
a
m
s
a
r
e
o
n
l
y
d
e
p
e
n
d
e
n
t
o
n
o
cu
la
r
a
n
a
ly
s
is
an
d
a
r
e
ca
r
r
ie
d
o
u
t
b
y
p
r
im
a
r
y
ca
r
e
h
ea
l
th
p
r
ac
titi
o
n
e
r
s
,
w
h
o
a
r
e
o
f
te
n
n
o
t
c
a
p
a
b
l
e
o
f
i
d
e
n
ti
f
y
i
n
g
o
r
al
ca
n
c
er
t
u
m
o
r
s
[
4
]
,
[
5
]
,
it
is
u
n
ce
r
tai
n
wh
et
h
e
r
o
r
n
o
t
th
e
y
wi
ll
b
e
e
f
f
ec
ti
v
e
i
n
t
h
ei
r
im
p
le
m
e
n
t
ati
o
n
.
T
h
er
e
is
a
g
r
ea
t
a
m
o
u
n
t
o
f
v
ar
i
ati
o
n
i
n
th
e
s
y
m
p
t
o
m
s
t
h
at
a
r
e
ass
o
cia
te
d
wit
h
o
r
a
l
ca
n
c
er
m
ali
g
n
a
n
ci
es,
w
h
i
ch
m
a
k
es
it
v
er
y
d
if
f
i
cu
lt
f
o
r
m
e
d
i
ca
l
p
r
o
f
e
s
s
io
n
als
t
o
m
a
k
e
a
d
ia
g
n
o
s
is
.
As
a
r
esu
lt,
t
h
is
v
a
r
iati
o
n
is
c
o
n
s
id
er
e
d
o
n
e
o
f
th
e
p
r
i
m
a
r
y
r
ea
s
o
n
s
p
atie
n
ts
'
r
ef
e
r
r
als
t
o
o
r
a
l
ca
n
c
e
r
s
p
e
cia
lis
ts
ar
e
d
e
la
y
e
d
[
5
]
.
On
th
e
o
t
h
e
r
h
an
d
,
i
n
i
tial
-
s
t
ag
e
o
es
o
p
h
a
g
ea
l
s
q
u
a
m
o
u
s
ce
ll
c
ar
ci
n
o
m
a
a
n
d
o
es
o
p
h
a
g
ea
l
s
q
u
am
o
u
s
ce
ll
c
a
r
cin
o
m
a
m
al
ig
n
an
ci
es
a
r
e
o
f
te
n
as
y
m
p
to
m
atic
a
n
d
m
a
n
i
f
es
t
as
b
e
n
i
g
n
,
n
o
n
-
t
o
x
i
c
m
al
ig
n
a
n
cies
,
t
y
p
i
ca
l
ly
d
ela
y
i
n
g
s
u
b
s
eq
u
en
t
i
d
e
n
t
if
ica
ti
o
n
.
T
h
e
d
ev
el
o
p
m
en
t o
f
c
o
n
te
m
p
o
r
ar
y
co
m
p
u
te
r
v
is
i
o
n
m
et
h
o
d
s
a
n
d
m
ac
h
i
n
e
l
ea
r
n
i
n
g
t
ec
h
n
i
q
u
es
h
as
r
es
u
lt
e
d
i
n
s
t
r
o
n
g
m
o
d
els
t
h
at
c
an
p
er
f
o
r
m
a
u
t
o
m
ate
d
s
cr
ee
n
i
n
g
o
f
o
r
al
les
io
n
s
a
n
d
p
r
o
v
i
d
e
m
e
d
ic
al
e
x
p
e
r
ts
wit
h
t
h
e
m
o
s
t
e
f
f
ec
ti
v
e
t
h
er
ap
y
f
o
r
th
ese
l
esi
o
n
s
.
Ma
c
h
i
n
e
le
ar
n
i
n
g
m
o
d
els
ar
e
p
r
ese
n
te
d
to
p
r
e
cis
el
y
c
o
m
p
ar
e
h
i
g
h
l
y
d
is
ti
n
g
u
is
h
e
d
OSC
C
s
an
d
s
o
m
ew
h
at
d
i
f
f
e
r
e
n
ti
ate
d
OSC
C
s
[
6
]
.
M
ac
h
i
n
e
lea
r
n
i
n
g
m
o
d
els
m
a
y
p
r
ed
ict
t
h
e
f
i
r
s
t
p
h
ase
o
f
l
y
m
p
h
n
o
d
e
m
et
astas
is
ca
u
s
ed
b
y
o
r
al
t
o
n
g
u
e
s
q
u
a
m
o
u
s
ce
ll
c
ar
ci
n
o
m
a
[
7
]
,
a
n
d
th
ese
m
o
d
els
a
ls
o
c
o
n
tr
ib
u
t
e
t
o
d
et
e
r
m
i
n
i
n
g
t
h
e
i
ll
n
ess
'
s
o
u
tc
o
m
es
[
8
]
.
Us
in
g
m
ac
h
i
n
e
le
a
r
n
in
g
m
o
d
els
g
r
e
atl
y
a
s
s
is
ts
t
h
e
i
n
v
est
ig
ati
o
n
o
f
v
a
r
i
o
u
s
ca
n
ce
r
o
u
s
t
u
m
o
r
s
.
T
h
e
d
is
s
em
in
ati
o
n
o
f
M
L
a
p
p
s
s
o
le
ly
d
ep
en
d
s
o
n
cli
n
i
ca
l
d
o
c
u
m
e
n
ta
ti
o
n
o
f
ill
n
ess
an
d
th
e
in
te
r
p
r
eta
ti
o
n
a
n
d
p
r
e
v
en
t
io
n
o
f
p
o
t
e
n
ti
all
y
m
a
li
g
n
a
n
t
o
r
al
c
o
n
tu
s
io
n
s
[
9
]
.
T
h
e
a
u
to
m
ated
d
etec
tio
n
o
f
o
r
al
m
alig
n
an
cies,
b
en
ig
n
co
n
tu
s
io
n
,
an
d
o
r
al
p
o
te
n
tially
m
alig
n
an
t
d
is
o
r
d
er
s
(
OPMDs)
b
r
o
ad
ly
d
ep
en
d
s
o
n
th
e
m
icr
o
s
co
p
ic
r
e
p
r
esen
tatio
n
o
f
th
e
im
a
g
es
[
1
0
]
-
[
1
3
]
.
So
m
e
o
th
er
s
tu
d
ies
in
clu
d
e
t
h
e
im
p
lem
en
tatio
n
o
f
m
u
lti
-
d
im
en
s
io
n
al
h
y
p
er
s
p
ec
t
r
al
im
ag
es
o
f
th
e
ca
v
ity
[
1
4
]
,
th
e
ap
p
licatio
n
o
f
co
m
p
u
ted
to
m
o
g
r
a
p
h
y
(
C
T
)
im
ag
es
[
1
5
]
,
th
e
a
p
p
licatio
n
o
f
au
to
f
lu
o
r
escen
ce
[
1
6
]
,
[
1
7
]
a
n
d
f
lu
o
r
escen
ce
im
ag
in
g
[
1
8
]
,
wh
ich
em
p
h
asizes
o
n
th
e
c
o
m
p
ar
ativ
e
v
iew
o
f
o
r
al
m
alig
n
an
cies a
n
d
wh
ite
lig
h
t im
ag
es f
o
r
o
r
al
ca
v
ity
tex
t
u
r
e
[
1
9
]
-
[
2
1
]
.
I
n
th
e
b
eg
in
n
in
g
o
f
th
is
ar
ea
,
m
ain
ly
ch
ar
ac
te
r
is
tics
r
elate
d
to
tex
tu
r
e
h
av
e
b
ee
n
d
ir
ec
ted
.
T
h
e
g
r
ay
lev
el
co
-
o
cc
u
r
r
en
ce
m
atr
ix
an
d
g
r
e
y
lev
el
r
u
n
-
le
n
g
th
a
r
e
o
p
er
ated
b
y
T
h
o
m
as
et
al.
[
19
]
,
L
PB
(
L
o
ca
l
B
in
ar
y
Patter
n
)
,
laws
tex
tu
r
e
en
er
g
y
,
an
d
h
ig
h
er
o
r
d
er
s
p
ec
tr
a
ar
e
u
tili
ze
d
b
y
Kr
is
h
n
an
et
a
l
.
[
1
0
]
.
T
h
e
r
ec
en
t
s
tu
d
ies
[
1
1
]
-
[
1
8
]
,
[
2
0
]
,
[
2
1
]
g
av
e
a
b
o
o
m
to
d
ee
p
lear
n
in
g
,
i.e
.
,
ar
t
if
icial
n
eu
r
al
n
etwo
r
k
s
(
ANN)
,
wh
ich
r
esid
es
in
m
u
ltip
le
lay
er
s
o
f
n
eu
r
o
n
s
an
d
r
eq
u
ir
es
h
u
g
e
d
atasets
.
T
h
ese
r
ec
en
t
tech
n
iq
u
es
o
f
f
er
f
as
t
co
m
p
u
tin
g
s
p
ee
d
,
wh
ich
h
elp
s
in
v
esti
g
ato
r
s
in
v
esti
g
ate
an
d
s
o
lv
e
cr
itical
p
r
o
b
lem
s
.
Ad
v
an
ce
m
e
n
ts
in
th
is
f
ield
h
av
e
p
r
o
v
i
d
ed
an
ef
f
icien
t
ap
p
licatio
n
o
f
d
ee
p
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
(
DC
NN)
.
Af
ter
co
n
q
u
er
in
g
th
e
I
m
ag
eNe
t
[
2
2
]
im
ag
e
class
if
i
ca
tio
n
co
n
test
i
n
th
e
y
ea
r
2
0
1
2
alo
n
g
with
Alex
Net
[
2
3
]
,
th
e
c
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
s
b
ec
am
e
f
am
o
u
s
in
th
e
c
o
m
p
u
t
er
v
is
io
n
d
o
m
ain
.
I
n
th
e
ea
r
ly
s
tag
es,
co
n
v
o
lu
ti
o
n
al
n
eu
r
al
n
etwo
r
k
s
(
C
NNs)
ar
e
o
p
er
ated
o
n
im
ag
e
ch
ar
a
cter
izatio
n
(
C
lass
if
icatio
n
o
f
im
ag
e
in
a
s
p
ec
if
ied
d
o
m
ain
)
.
C
NN
-
b
ased
f
r
a
m
ewo
r
k
d
esig
n
i
n
g
h
as
b
ee
n
an
n
o
u
n
ce
d
as
a
s
ig
n
if
ican
t
b
r
ea
k
t
h
r
o
u
g
h
in
th
e
o
b
ject
d
etec
tio
n
f
ield
,
lik
e
PASC
AL
v
is
u
al
o
b
ject
class
es
[2
4
]
a
n
d
c
o
m
m
o
n
o
b
jects
in
co
n
tex
t
(
C
OC
O)
[
2
5
]
.
T
h
e
m
a
x
im
u
m
ac
cu
r
ac
y
is
ac
h
iev
ed
b
y
t
h
e
R
-
C
NN
g
r
o
u
p
(
r
e
g
io
n
-
b
ased
C
NN
tech
n
iq
u
e)
[
2
6
]
,
Fas
t
R
-
C
NN
[
2
7
]
,
Fas
ter
R
-
C
NN
[
2
8
]
,
an
d
th
e
r
ec
e
n
t
Ma
s
k
R
-
C
N
N
[
29
]
.
Sin
g
le
-
lev
el
d
etec
to
r
s
lik
e
y
o
u
o
n
ly
lo
o
k
o
n
ce
(
YOL
O)
[
3
0
]
an
d
s
in
g
le
s
h
o
t
d
etec
to
r
(
SS
D)
[
3
1
]
ar
e
th
e
f
aster
tech
n
iq
u
es
to
ac
h
iev
e
g
o
o
d
ac
cu
r
ac
y
.
I
n
th
e
m
e
d
ical
im
ag
in
g
f
ield
,
o
b
ject
d
etec
tio
n
f
r
am
ewo
r
k
s
h
av
e
b
ee
n
em
p
lo
y
ed
alo
n
g
with
Fas
ter
R
-
C
NN
f
o
r
co
lo
n
p
o
ly
p
d
etec
tio
n
[
3
2
]
a
s
well
as
th
e
ch
ar
ac
ter
izatio
n
o
f
m
alig
n
a
n
cies
in
m
am
m
o
g
r
a
m
s
[
3
3
]
.
An
an
th
ar
am
an
et
a
l.
[2
1
]
h
a
v
e
wo
r
k
e
d
o
n
o
r
al
ca
n
ce
r
im
ag
es
u
s
in
g
th
e
Ma
s
k
R
-
C
NN
tech
n
iq
u
e
u
s
in
g
4
0
4
0
-
im
ag
e
d
ataset.
T
h
eir
s
tu
d
y
was
to
d
iag
n
o
s
e
b
en
ig
n
o
r
al
ca
v
ities
(
h
er
p
eslab
ialis
an
d
ap
h
th
o
u
s
u
lcer
s
u
s
in
g
in
s
tan
ce
s
eg
m
en
tatio
n
.
An
au
th
en
tic
clin
ically
test
ed
d
ataset
is
r
eq
u
ir
ed
f
o
r
th
e
m
o
s
t
ac
cu
r
ate
a
u
t
o
m
atic
d
ia
g
n
o
s
is
o
f
ea
r
l
y
o
r
al
m
alig
n
an
cy
.
B
y
m
ak
in
g
u
s
e
o
f
d
ee
p
lear
n
in
g
a
lg
o
r
ith
m
s
,
ac
c
u
r
ac
y
,
an
d
e
f
f
icien
cy
ca
n
b
e
en
h
a
n
c
ed
to
th
e
b
r
o
a
d
est
p
o
s
s
ib
le
d
a
ta.
I
n
2
0
2
1
,
to
ac
ce
s
s
th
e
f
o
u
r
d
atasets
:
E
B
SC
O,
Pu
b
Me
d
,
OVI
D,
an
d
Sco
p
u
s
,
th
e
Un
iv
er
s
ity
o
f
Sh
ar
jah
L
ib
r
ar
y
was
u
tili
ze
d
to
m
an
ag
e
th
e
in
v
esti
g
atio
n
.
T
h
e
d
is
co
v
er
ies
wer
e
r
elea
s
ed
in
th
e
y
ea
r
2
0
0
0
-
2
0
2
1
an
d
h
a
v
e
p
r
esen
ted
a
r
o
b
u
s
t
im
p
r
o
v
em
en
t
in
th
e
d
etec
tio
n
an
d
tr
ea
tm
en
t
o
f
o
r
al
ca
n
ce
r
u
s
in
g
AI
,
ML
,
DL
,
an
d
n
eu
r
al
n
etwo
r
k
s
.
T
o
f
in
d
th
e
r
esear
c
h
,
a
s
et
o
f
k
ey
wo
r
d
s
lik
e
“m
ac
h
in
e
lear
n
in
g
”
[
Me
S
H
ter
m
]
OR
“n
eu
r
al
n
etwo
r
k
”
[
Me
SH te
r
m
]
)
wer
e
u
tili
ze
d
to
f
in
d
th
e
ar
ticles in
all
f
o
u
r
d
atab
ases
f
o
r
th
e
a
p
p
r
o
p
r
iate
s
cr
ee
n
in
g
o
f
ar
ticles.
T
h
ese
Den
tal
J
o
u
r
n
als
ca
n
b
e
f
o
u
n
d
u
s
in
g
th
e
m
an
u
al
s
ea
r
ch
o
p
tio
n
s
,
wh
ich
a
r
e:
J
o
u
r
n
al
o
f
On
c
o
lo
g
y
,
J
o
u
r
n
al
o
f
Or
al
Dis
ea
s
es,
J
o
u
r
n
al
o
f
Or
al
Path
o
lo
g
y
an
d
Me
d
icin
e
a
n
d
Or
al
Su
r
g
er
y
Or
al
Me
d
icin
e,
Or
al
Path
o
lo
g
y
Or
al
R
ad
io
lo
g
y
,
I
n
te
r
n
atio
n
al
J
o
u
r
n
al
o
f
Or
al
an
d
Ma
x
illo
f
ac
ial
Su
r
g
er
y
,
E
u
r
o
p
ea
n
J
o
u
r
n
al
o
f
C
r
an
io
m
a
x
illo
f
ac
ial
Su
r
g
e
r
y
,
B
r
itis
h
J
o
u
r
n
al
o
f
Or
al
a
n
d
Ma
x
illo
f
ac
ial
Su
r
g
er
y
,
a
n
d
J
o
u
r
n
al
o
f
C
r
an
io
f
ac
ial
Su
r
g
e
r
y
.
W
ith
th
e
h
elp
o
f
t
h
e
f
o
llo
win
g
r
esear
ch
r
ef
er
en
ce
lis
ts
,
m
an
y
s
tu
d
ies
ca
n
b
e
m
an
ag
ed
.
Ad
d
itio
n
ally
,
He
et
a
l
.
[3
4
]
p
r
esen
ted
d
ee
p
r
esid
u
al
lear
n
in
g
f
o
r
im
ag
e
r
ec
o
g
n
i
tio
n
.
I
n
wh
ich
Fas
ter
R
-
C
NN
i
s
ad
o
p
ted
as
a
d
etec
ti
o
n
m
eth
o
d
.
Yad
u
v
an
s
h
i
et
a
l
.
[
3
5
]
is
d
is
cu
s
s
ed
an
au
to
m
atic
class
if
icatio
n
m
et
h
o
d
s
in
o
r
al
ca
n
ce
r
d
etec
tio
n
.
P
r
e
v
i
o
u
s
r
ese
ar
ch
h
as
s
h
o
wn
n
o
te
wo
r
t
h
y
im
p
r
o
v
e
m
e
n
t
in
ca
n
c
er
d
ete
ct
io
n
;
h
o
we
v
e
r
,
t
h
e
p
e
r
f
o
r
m
a
n
c
e
o
f
t
h
e
s
y
s
t
em
s
is
c
h
a
lle
n
g
i
n
g
b
ec
au
s
e
o
f
lo
w
f
e
at
u
r
e
d
is
ti
n
c
ti
v
e
n
ess
an
d
p
o
o
r
c
o
r
r
e
lat
io
n
b
e
twe
en
g
l
o
b
a
l
a
n
d
l
o
c
al
ch
ar
ac
t
e
r
is
ti
cs
o
f
t
h
e
ca
n
c
er
im
ag
es.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
N
o
ve
l m
u
ltil
ev
el
lo
ca
l b
in
a
r
y
textu
r
e
d
escr
ip
to
r
fo
r
o
r
a
l c
a
n
ce
r
…
(
V
ija
ya
Ya
d
u
v
a
n
s
h
i
)
839
Vija
y
a
Ya
d
u
v
a
n
s
h
i
et
a
l
.
[
3
6
]
h
as
g
i
v
e
n
a
n
a
u
t
o
m
at
ic
o
r
a
l
c
an
ce
r
d
e
tec
ti
o
n
a
n
d
class
if
ica
t
io
n
u
s
i
n
g
m
o
d
i
f
i
e
d
lo
c
al
t
e
x
t
u
r
e
d
esc
r
i
p
t
o
r
a
n
d
m
a
ch
i
n
e
le
ar
n
i
n
g
a
lg
o
r
it
h
m
s
in
w
h
ic
h
co
n
v
o
l
u
ti
o
n
al
n
e
u
r
a
l n
et
wo
r
k
(
C
NN
)
is
u
s
e
d
f
o
r
b
et
te
r
te
x
t
u
r
e
f
ea
tu
r
e
r
e
p
r
e
s
en
t
ati
o
n
.
T
r
a
d
iti
o
n
a
l
te
x
t
u
r
e
d
es
cr
ip
to
r
te
c
h
n
iq
u
es
ar
e
c
h
al
l
en
g
i
n
g
d
u
e
t
o
s
c
al
e
v
a
r
i
an
ce
,
l
ar
g
er
f
ea
t
u
r
e
v
ec
to
r
s
,
r
o
tat
io
n
i
n
v
ar
ia
n
c
e,
a
n
d
s
e
n
s
i
tiv
it
y
t
o
u
n
e
v
e
n
ill
u
m
i
n
at
io
n
,
n
o
is
e
,
a
n
d
b
lu
r
.
T
h
e
o
u
tc
o
m
es
o
f
t
h
e
ML
c
lass
i
f
i
er
ar
e
h
u
g
el
y
af
f
ec
t
ed
b
y
t
h
e
s
i
ze
an
d
le
n
g
t
h
o
f
th
e
f
e
at
u
r
es.
T
h
e
r
esear
ch
p
ap
e
r
r
ep
r
esen
ts
a
m
u
ltil
ev
el
lo
ca
l
b
in
ar
y
p
att
er
n
tech
n
i
q
u
e
t
o
in
v
esti
g
ate
t
h
e
tex
tu
r
e
f
ea
tu
r
es
in
o
r
al
ca
n
ce
r
im
ag
es
.
T
h
is
is
a
p
r
o
v
e
n
n
o
v
el
tech
n
iq
u
e
to
d
if
f
er
en
tiate
tex
tu
r
e
o
r
al
ca
n
ce
r
s
am
p
les
with
s
ig
n
if
ican
tly
g
o
o
d
ac
c
u
r
ac
y
.
I
n
th
e
p
r
o
p
o
s
ed
wo
r
k
,
a
m
u
ltil
ev
el
lo
ca
l
b
in
ar
y
p
atter
n
is
u
s
ed
to
in
v
esti
g
ate
th
e
tex
tu
r
e
f
ea
tu
r
es
o
f
th
e
o
r
al
ca
n
ce
r
s
am
p
l
es.
T
h
e
r
esu
lts
estab
lis
h
th
e
ad
v
an
tag
es
o
f
t
h
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
i
n
ch
ar
ac
t
er
izin
g
th
e
tex
tu
r
e
o
f
o
r
al
ca
n
ce
r
im
ag
es.
al
s
o
,
it
p
r
o
v
id
es
m
o
r
e
p
r
ec
is
e
f
ea
tu
r
es
f
o
r
tex
tu
r
e
in
o
r
al
ca
n
ce
r
I
m
a
g
es.
I
t
is
p
r
o
v
ed
th
at
th
e
ML
B
T
D
alg
o
r
ith
m
g
iv
es
b
etter
ac
cu
r
ac
y
th
an
th
e
L
B
P
alg
o
r
ith
m
,
with
9
0
.
5
7
%
ac
cu
r
ac
y
.
Als
o
,
th
e
o
v
er
all
p
er
f
o
r
m
an
ce
o
f
ML
B
T
D
with
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
clas
s
if
ier
o
u
tp
er
f
o
r
m
s
in
co
n
tr
ast
with
ML
B
T
D
wi
th
k
-
n
ea
r
est
n
eig
h
b
o
r
(
KNN)
an
d
ML
B
T
D
with
class
if
icatio
n
tr
ee
(
C
T
)
alg
o
r
ith
m
s
.
I
n
t
h
e
p
r
ese
n
t
s
c
e
n
a
r
i
o
,
o
r
al
c
a
n
c
er
c
h
ar
ac
t
er
is
ti
cs,
b
y
its
o
c
c
u
r
r
en
ce
r
ati
o
,
th
e
r
e
q
u
i
r
e
m
e
n
t
to
b
o
o
s
t
its
p
r
e
v
e
n
ti
o
n
m
et
h
o
d
s
,
a
n
d
m
an
y
s
t
u
d
ies
ar
e
p
r
esc
r
i
b
e
d
f
o
r
a
p
p
l
y
i
n
g
M
L
m
o
d
els
.
Fo
r
th
ese
ty
p
es
o
f
m
al
ig
n
a
n
cies
,
e
x
te
n
s
i
v
e
r
es
ea
r
c
h
was
c
ar
r
i
ed
o
u
t
to
es
ta
b
li
s
h
a
n
ef
f
i
cie
n
t
a
p
p
li
ca
t
io
n
o
f
ML
a
lg
o
r
it
h
m
s
in
d
et
ec
ti
n
g
o
r
al
ca
n
ce
r
.
[
9
]
T
h
e
p
a
p
e
r
'
s
a
r
r
a
n
g
e
m
e
n
t
i
s
as
f
o
l
l
o
w
s
:
S
e
ct
i
o
n
2
r
e
p
r
e
s
e
n
t
s
t
h
e
b
a
c
k
g
r
o
u
n
d
a
n
d
t
h
e
o
r
y
o
f
t
h
e
p
r
o
p
o
s
e
d
w
o
r
k
.
S
e
c
t
i
o
n
3
r
e
p
r
e
s
e
n
ts
th
e
p
r
o
p
o
s
e
d
d
e
s
i
g
n
a
n
d
o
p
e
r
a
t
i
n
g
p
r
i
n
c
i
p
l
e
.
S
ec
t
i
o
n
4
s
h
o
w
s
t
h
e
r
e
s
u
l
t
s
a
n
d
d
i
s
c
u
s
s
es
t
h
e
p
r
o
p
o
s
e
d
a
l
g
o
r
it
h
m
.
S
e
c
t
i
o
n
5
d
es
c
r
i
b
es
t
h
e
c
o
n
c
l
u
s
i
o
n
a
n
d
f
u
t
u
r
e
s
c
o
p
e
o
f
t
h
e
w
o
r
k
.
2.
M
E
T
H
O
D
A
m
ac
h
in
e
lear
n
in
g
-
b
ased
ap
p
r
o
ac
h
u
s
in
g
c
o
llab
o
r
ativ
e
tex
tu
r
e
an
d
co
lo
r
f
ea
tu
r
es
is
p
r
o
p
o
s
ed
.
T
h
e
p
r
o
p
o
s
ed
m
u
ltil
ev
el
lo
ca
l
b
in
ar
y
tex
tu
r
e
d
escr
ip
to
r
(
ML
B
T
D)
co
n
s
id
er
s
m
o
r
e
th
an
o
n
e
lab
el
f
o
r
th
e
b
in
a
r
y
v
alu
e
co
m
p
u
tatio
n
to
im
p
r
o
v
e
th
e
tex
tu
r
e
in
f
o
r
m
atio
n
o
f
th
e
o
r
al
ca
v
ity
im
a
g
es
(
Fig
u
r
e
1
)
.
Fig
u
r
e
2
r
ep
r
esen
ts
th
e
ML
B
T
D
alg
o
r
i
th
m
,
wh
ich
illu
s
tr
ates
th
e
co
n
n
ec
tio
n
o
f
th
e
p
ix
el
to
th
eir
lo
ca
l
n
eig
h
b
o
r
h
o
o
d
with
tex
tu
r
e
f
ea
tu
r
e
r
ep
r
esen
ta
tio
n
in
th
e
ex
is
tin
g
a
n
d
p
r
o
p
o
s
ed
m
eth
o
d
o
lo
g
y
.
I
n
th
e
p
r
e
-
p
r
o
ce
s
s
in
g
s
tag
e,
th
e
R
GB
p
ictu
r
e
s
am
p
le
is
r
ef
o
r
m
ed
to
a
g
r
ay
s
ca
le
p
ictu
r
e.
Fig
u
r
e
1
.
Mu
ltil
ev
el
lo
ca
l
b
in
a
r
y
p
atter
n
(
ML
B
T
D)
ML
B
T
D
alg
o
r
ith
m
is
p
r
o
p
o
s
ed
to
o
f
f
er
an
ad
eq
u
ate
tex
tu
r
e
d
escr
ip
to
r
.
L
B
P
h
is
to
g
r
am
is
g
en
er
ate
d
f
o
r
d
if
f
er
en
t
b
l
o
ck
s
izes
(
N)
.
Su
p
p
o
r
t
v
ec
t
o
r
m
ac
h
i
n
e,
k
-
n
e
ar
est
n
eig
h
b
o
r
,
an
d
class
i
f
icatio
n
T
r
ee
class
if
ier
s
ar
e
u
s
ed
.
Acc
u
r
ac
y
will su
f
f
er
b
ec
au
s
e
o
f
th
e
class
im
b
alan
ce
p
r
o
b
lem
,
s
o
s
y
n
th
etic
d
ata
is
g
en
er
ated
with
th
e
ex
is
ti
n
g
d
ata.
Gen
e
r
ally
,
Gab
o
r
tr
an
s
f
o
r
m
is
u
s
ed
,
b
u
t
th
is
r
e
s
ea
r
ch
is
d
o
n
e
o
n
tex
tu
r
e
an
d
co
lo
r
f
ea
t
u
r
es.
T
h
e
tex
tu
r
al
s
tr
u
ctu
r
e
o
f
an
y
im
ag
e
d
ep
en
d
s
u
p
o
n
th
e
m
o
r
e
m
in
o
r
ch
an
g
es
in
th
e
in
ten
s
ity
,
an
d
ca
n
ce
r
o
u
s
ce
lls
in
th
e
o
r
al
ca
v
ity
b
r
in
g
th
e
n
o
n
-
h
o
m
o
g
en
eity
in
th
e
te
x
tu
r
e.
T
h
e
lo
ca
l
b
in
ar
y
p
atter
n
is
co
m
p
u
ted
u
s
in
g
(
1
)
a
n
d
(
2
)
.
T
h
e
h
is
to
g
r
am
o
f
t
h
e
ML
B
T
D
d
escr
ip
to
r
is
co
m
p
u
ted
to
m
in
im
ize
th
e
f
ea
tu
r
e
v
e
cto
r
.
I
t
r
ep
r
esen
ts
th
at
h
ig
h
er
im
p
o
r
tan
ce
is
g
iv
en
to
th
e
im
m
ed
iate
n
eig
h
b
o
r
an
d
l
o
wer
im
p
o
r
tan
ce
is
g
iv
en
to
t
h
e
f
ar
th
er
n
eig
h
b
o
r
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
4
,
No
.
3
,
Dec
em
b
er
20
2
5
:
837
-
8
4
4
840
I
t
is
a
r
o
tatio
n
in
v
ar
ian
t
an
d
s
ca
le
v
ar
ian
t.
I
t
s
h
o
ws
b
etter
p
er
f
o
r
m
a
n
ce
f
o
r
u
n
ev
en
illu
m
i
n
atio
n
ch
an
g
es
an
d
b
lu
r
r
ed
co
n
d
itio
n
s
.
(
)
=
{
1
(
)
≥
0
0
ℎ
(
1
)
(
)
=
(
1
−
0
)
∗
+
(
2
−
0
)
∗
(
−
1
)
+
(
3
−
0
)
∗
(
−
2
)
+
⋯
(
2
)
(
)
=
∑
(
−
0
)
∗
(
−
−
1
)
=
1
(
4
)
W
h
er
e,
th
e
win
d
o
w
r
ad
iu
s
is
R
,
ce
n
ter
p
ix
el
in
th
e
win
d
o
w
is
Ao
,
an
d
lo
ca
tio
n
o
f
t
h
e
n
eig
h
b
o
r
in
g
p
ix
el
is
i.
Fig
u
r
e
2
.
Flo
w
d
ia
g
r
am
o
f
th
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
I
n
th
e
ex
p
er
im
e
n
tatio
n
,
1
0
0
x
r
eso
lu
tio
n
b
e
n
ig
n
h
is
to
p
ath
o
l
o
g
ical
im
ag
e
is
tak
en
f
r
o
m
th
e
p
u
b
licly
av
ailab
le
Me
n
d
eley
d
ataset
,
wh
ich
co
n
s
is
ts
o
f
a
T
o
tal
o
f
1
2
2
4
im
ag
es
[3
7
]
.
I
m
ag
es
ar
e
d
if
f
er
en
tiated
in
to
tw
o
ca
teg
o
r
ies
with
two
d
if
f
er
en
t
r
eso
lu
tio
n
s
.
T
h
e
f
ir
s
t
ca
teg
o
r
y
co
n
s
is
ts
o
f
8
9
h
is
to
p
ath
o
lo
g
ical
im
ag
es
with
th
e
n
o
r
m
al
ep
ith
eliu
m
o
f
th
e
o
r
al
ca
v
ity
an
d
4
3
9
im
ag
es
o
f
o
r
al
s
q
u
am
o
u
s
ce
ll
ca
r
cin
o
m
a
(
OSC
C
)
with
1
0
0
x
m
ag
n
if
icatio
n
.
T
h
e
s
ec
o
n
d
ca
t
eg
o
r
y
c
o
n
s
is
ts
o
f
2
0
1
im
ag
es
with
th
e
n
o
r
m
al
ep
ith
eliu
m
o
f
th
e
o
r
al
ca
v
ity
an
d
4
9
5
h
is
to
p
ath
o
lo
g
ical
im
ag
es
o
f
OSC
C
wi
th
4
0
0
x
m
a
g
n
if
ic
atio
n
s
.
T
h
e
im
ag
es
wer
e
ca
p
t
u
r
ed
u
s
in
g
a
L
eica
I
C
C
5
0
HD
m
icr
o
s
co
p
e
f
r
o
m
H&
E
s
tain
ed
tis
s
u
e
s
lid
es
co
llected
,
d
esig
n
e
d
,
a
n
d
ca
talo
g
ed
b
y
m
ed
ical
e
x
p
e
r
ts
f
r
o
m
2
3
0
p
atien
ts
.
Fig
u
r
e
3
r
e
p
r
esen
ts
th
e
d
if
f
er
en
t
v
e
r
s
io
n
s
o
f
th
e
s
am
p
le
im
ag
e.
Fig
u
r
e
3
(
a)
s
h
o
ws
o
r
i
g
in
al
s
am
p
le
im
ag
e,
Fig
u
r
e
3
(
b
)
ill
u
s
tr
ates
th
e
g
r
ay
co
n
v
er
ted
i
m
ag
e,
Fig
u
r
e
3
(
c)
r
e
p
r
esen
ts
ML
B
T
D
f
o
r
R
=1
,
Fig
u
r
e
3
(
d
)
d
is
p
lay
s
ML
B
T
D
f
o
r
R
=2
,
Fig
u
r
e
3
(
e)
d
e
p
icts
ML
B
T
D
f
o
r
R
=3
,
Fig
u
r
e
3
(
f
)
s
h
o
ws
L
B
P
d
escr
ip
to
r
f
o
r
R
=1
,
Fig
u
r
e
3
(
g
)
d
is
p
lay
s
ML
B
T
D
h
is
to
g
r
am
N=
1
an
d
Fig
u
r
e
3
(
h
)
v
is
u
alize
s
ML
B
T
D
h
is
to
g
r
am
f
o
r
N=
2
.
I
t
ca
n
b
e
o
b
s
er
v
ed
th
at
ML
B
T
D
d
escr
ip
to
r
s
h
o
ws
g
o
o
d
tex
tu
r
e
f
ea
tu
r
e
r
ep
r
esen
tatio
n
i
n
co
n
t
r
ast
with
th
e
co
n
v
e
n
tio
n
a
l
L
B
P
tex
tu
r
e
d
escr
ip
to
r
.
ML
B
T
D
d
escr
ip
to
r
f
o
r
v
a
r
io
u
s
b
lo
ck
s
ize
(
N)
an
d
r
ad
iu
s
(
R
)
ar
e
s
h
o
wn
in
Fig
u
r
e
3
.
R
esu
lts
ar
e
s
h
o
wn
u
s
in
g
th
e
OSC
C
d
ataset,
wh
ich
co
n
s
i
s
ts
o
f
4
3
9
p
h
o
to
s
o
f
OSC
C
at
1
0
0
x
m
ag
n
if
icatio
n
an
d
8
9
h
is
to
lo
g
ical
im
ag
es
s
h
o
win
g
th
e
o
r
al
ca
v
ity
'
s
n
o
r
m
al
ep
ith
eliu
m
.
T
ab
le
1
c
o
m
p
a
r
es
th
e
ML
B
T
D
m
eth
o
d
an
d
th
e
l
o
ca
l
b
in
ar
y
p
atter
n
(
L
B
P)
ap
p
r
o
a
ch
u
s
in
g
a
lin
ea
r
s
u
p
p
o
r
t
v
ec
t
o
r
m
ac
h
i
n
e
(
SVM)
f
o
r
b
lo
c
k
s
izes
N
=
1
,
2
,
3
,
a
n
d
4
.
Fo
r
N=
3
with
2
2
9
5
to
tal
n
o
.
o
f
f
ea
tu
r
es,
8
6
.
1
6
%
ac
cu
r
a
cy
,
0
.
5
7
r
ec
all,
0
.
7
8
p
r
ec
is
io
n
an
d
0
.
6
6
F1
-
s
co
r
e
is
ac
h
iev
ed
in
L
B
P+SVM
an
d
9
0
.
5
7
%
ac
c
u
r
ac
y
,
0
.
6
8
r
ec
all,
0
.
8
5
p
r
ec
is
io
n
an
d
0
.
7
5
F1
-
s
co
r
e
is
ac
h
iev
ed
in
ML
B
T
D+
SVM,
th
e
L
B
P
an
d
ML
B
T
D
co
m
p
ar
is
o
n
is
d
o
n
e
with
th
e
k
-
n
ea
r
est
Neig
h
b
o
u
r
(
KNN)
f
o
r
B
lo
ck
s
i
ze
N=
1
,
2
,
3
,
4
.
Fo
r
N=
3
with
2
2
9
5
to
tal
n
o
.
o
f
f
ea
tu
r
es,
8
2
.
2
5
%
ac
cu
r
ac
y
,
0
.
4
6
r
ec
all,
0
.
7
0
p
r
ec
is
io
n
an
d
0
.
5
6
F1
-
s
co
r
e
is
ac
h
iev
ed
in
L
B
P+KNN
an
d
8
8
.
6
8
%
ac
c
u
r
ac
y
,
0
.
6
3
r
ec
all,
0
.
8
1
Pre
cisi
o
n
an
d
0
.
7
1
F1
-
s
co
r
e
is
ac
h
iev
ed
in
ML
B
T
D+
KNN,
th
e
L
B
P
an
d
ML
B
T
D
alg
o
r
ith
m
s
with
B
lo
ck
s
ize
N=
1
,
2
,
3
,
4
an
d
co
m
p
ar
ed
with
th
e
class
if
icatio
n
tr
ee
(
C
T
)
.
Fo
r
N=
3
with
2
2
9
5
to
tal
n
o
.
o
f
f
ea
tu
r
es,
8
2
.
8
4
%
ac
cu
r
ac
y
,
0
.
4
8
r
ec
all,
0
.
7
0
Pr
ec
is
io
n
,
an
d
0
.
5
7
F1
-
s
co
r
e
is
ac
h
iev
ed
i
n
L
B
P+C
T
an
d
8
9
.
9
4
%
ac
cu
r
ac
y
,
0
.
6
7
r
ec
all,
0
.
8
1
Pre
cisi
o
n
,
an
d
0
.
7
3
F1
-
s
co
r
e
h
as
ac
h
ie
v
ed
i
n
M
L
B
T
D+
C
T
T
h
e
f
in
d
in
g
s
s
h
o
w
th
at
u
s
in
g
an
SVM
class
if
ier
in
co
n
ju
n
ctio
n
wit
h
th
e
ML
B
T
D
m
eth
o
d
im
p
r
o
v
es
ac
cu
r
ac
y
an
d
o
th
er
m
e
asu
r
em
en
t
m
etr
ics.
ML
B
T
D
tex
tu
r
e
d
escr
ip
to
r
p
r
o
v
id
es
a
g
o
o
d
tex
tu
r
e
r
e
p
r
esen
tatio
n
o
f
th
e
h
is
to
p
a
th
o
lo
g
ical
im
ag
es
a
n
d
im
p
r
o
v
es
th
e
ac
cu
r
ac
y
o
f
th
e
ev
alu
atio
n
o
f
m
alig
n
an
t
r
e
g
io
n
s
in
OSC
C
im
ag
es.
Up
to
b
lo
ck
s
ize,
N=
4
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
N
o
ve
l m
u
ltil
ev
el
lo
ca
l b
in
a
r
y
textu
r
e
d
escr
ip
to
r
fo
r
o
r
a
l c
a
n
ce
r
…
(
V
ija
ya
Ya
d
u
v
a
n
s
h
i
)
841
co
n
tr
ib
u
tes
g
o
o
d
r
esu
lts
,
b
u
t
o
n
f
u
r
th
e
r
in
cr
ea
s
in
g
b
lo
c
k
s
ize,
th
e
ac
cu
r
ac
y
wo
u
ld
b
e
af
f
ec
ted
b
ec
au
s
e
o
f
th
e
r
ed
u
n
d
an
cy
i
n
th
e
f
ea
t
u
r
e
s.
(
a)
(
b)
(
c)
(
d)
(
e)
(
f)
(
g
)
(
h
)
Fig
u
r
e
3
.
Vis
u
aliza
tio
n
o
f
th
e
ML
B
T
D
r
esu
lts
(
a)
co
lo
r
ed
im
ag
e
(
b
)
g
r
ey
s
ca
le
im
ag
e
(
c)
M
L
B
T
D
f
o
r
R
=1
,
(
d
)
ML
B
T
D
f
o
r
R
=2
,
(
e)
ML
B
T
D
f
o
r
R
=3
,
(
f
)
L
B
P f
o
r
R
=1
,
(
g
)
ML
B
T
D
h
is
to
g
r
am
N=
1
,
(
h
)
ML
B
T
D
h
is
to
g
r
am
f
o
r
N=
2
T
h
e
co
m
p
ar
ativ
e
r
esu
lts
o
f
t
h
e
ML
B
T
D
ar
e
p
r
o
v
id
ed
in
T
ab
le
2
wh
er
e
r
esu
lts
o
f
M
L
B
T
D
ar
e
co
m
p
ar
ed
with
h
is
to
g
r
am
o
f
o
r
ien
ted
g
r
a
d
ien
t,
g
r
e
y
lev
e
l
co
-
o
cc
u
r
r
en
ce
m
atr
ix
(
GL
C
M)
an
d
L
B
P.
I
t
is
o
b
s
er
v
ed
th
at
t
h
e
ML
B
T
D
o
f
f
er
s
s
u
p
er
io
r
r
esu
lts
th
an
HOG,
L
B
O
an
d
GL
C
M.
GL
C
M,
a
tex
tu
r
e
-
b
ased
m
eth
o
d
,
ac
h
iev
ed
r
elativ
ely
lo
wer
ac
cu
r
ac
y
ac
r
o
s
s
all
class
if
ier
s
,
with
th
e
h
ig
h
est
b
ein
g
6
8
.
3
0
%
u
s
in
g
SVM.
HOG,
wh
ich
f
o
cu
s
es
o
n
o
b
jec
t
s
h
ap
e
an
d
ed
g
e
in
f
o
r
m
atio
n
,
p
er
f
o
r
m
e
d
b
etter
,
with
a
m
ax
im
u
m
ac
cu
r
ac
y
o
f
7
8
.
8
5
%
(
SVM)
.
L
B
P,
ca
p
tu
r
in
g
lo
ca
l
tex
tu
r
e
p
atter
n
s
,
s
h
o
wed
s
ig
n
if
ican
t
im
p
r
o
v
em
e
n
t,
r
ea
ch
in
g
8
6
.
1
6
%
with
SVM.
T
h
e
m
o
s
t
ef
f
ec
tiv
e
s
ch
em
e,
ML
B
T
D,
co
m
b
in
in
g
m
u
ltip
le
tex
tu
r
e
d
escr
ip
to
r
s
,
ac
h
iev
ed
th
e
h
ig
h
est
o
v
er
all
ac
cu
r
ac
y
,
with
SVM
r
ea
ch
in
g
9
0
.
5
7
%,
f
o
ll
o
wed
clo
s
ely
b
y
C
T
(
8
9
.
9
4
%)
an
d
KNN
(
8
8
.
5
6
%).
T
h
is
s
h
o
ws
a
clea
r
tr
en
d
o
f
im
p
r
o
v
in
g
ac
cu
r
ac
y
f
r
o
m
b
as
ic
tex
tu
r
e
m
eth
o
d
s
(
GL
C
M)
to
m
o
r
e
ad
v
an
ce
d
tex
tu
r
e
-
d
escr
ib
in
g
tech
n
i
q
u
es
(
ML
B
T
D)
,
with
SVM
co
n
s
is
ten
tly
o
u
tp
er
f
o
r
m
in
g
th
e
o
t
h
er
class
if
ier
s
ac
r
o
s
s
all
s
ch
em
es.
T
h
e
ML
B
T
D
o
f
f
er
s
a
s
ig
n
if
i
ca
n
t
im
p
r
o
v
em
en
t
in
t
h
e
lo
ca
l
f
ea
tu
r
e
d
ep
ictio
n
th
an
t
h
e
c
o
n
v
en
tio
n
L
B
P
s
ch
em
e
b
y
p
r
o
v
id
in
g
th
e
m
u
lti
-
n
eig
h
b
o
r
r
elatio
n
s
h
ip
b
etwe
en
th
e
ad
jace
n
t
f
ea
tu
r
es.
T
h
e
f
ea
tu
r
es
s
h
o
ws
th
e
s
ca
le
in
v
ar
ian
c
e
an
d
r
o
tatio
n
in
v
ar
ia
n
ce
b
ec
au
s
e
o
f
t
h
e
h
is
to
g
r
am
f
ea
tu
r
es
wh
ich
a
r
e
in
d
ep
en
d
en
t
o
n
t
h
e
s
ca
le
an
d
r
o
tatio
n
o
f
th
e
im
ag
es.
T
h
e
ML
T
B
D
h
is
to
g
r
am
p
r
o
v
id
es
im
p
er
ativ
e
r
ed
u
ctio
n
in
th
e
f
ea
tu
r
e
v
ec
to
r
len
g
th
wh
ich
h
elp
s
to
ac
h
iev
e
th
e
lo
wer
r
ec
o
g
n
itio
n
tim
e
(
0
.
1
2
s
ec
p
er
s
am
p
le)
.
T
h
is
s
tu
d
y
p
r
o
v
id
e
d
ex
ce
llen
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
4
,
No
.
3
,
Dec
em
b
er
20
2
5
:
837
-
8
4
4
842
r
esu
lts
f
o
r
t
h
e
v
ar
io
u
s
ML
cla
s
s
if
ier
s
h
o
wev
er
th
e
p
e
r
f
o
r
m
a
n
ce
o
f
th
e
s
y
s
tem
is
lim
ited
f
o
r
th
e
lar
g
er
d
atasets
an
d
m
u
lti
-
lev
el
ca
n
ce
r
d
etec
tio
n
s
y
s
tem
.
T
h
e
s
y
s
tem
lack
s
i
n
p
r
o
v
id
i
n
g
t
h
e
in
te
r
p
r
etab
ilit
y
an
d
e
x
p
lain
ab
ilit
y
f
o
r
ca
n
ce
r
d
etec
tio
n
.
T
ab
le
1
.
Per
f
o
r
m
an
ce
m
etr
ics co
m
p
ar
is
o
n
o
f
p
r
o
p
o
s
ed
ML
B
T
D
an
d
L
B
P
f
ea
tu
r
es with
v
ar
io
u
s
ML
class
if
ier
s
b
y
v
ar
y
in
g
b
lo
ck
s
ize
A
l
g
o
r
i
t
h
m
N
o
o
f
B
l
o
c
k
s
N
u
mb
e
r
o
f
F
e
a
t
u
r
e
s
A
c
c
u
r
a
c
y
R
e
c
a
l
l
P
r
e
c
i
s
i
o
n
F1
-
sc
o
r
e
LB
P
+
Li
n
e
a
r
S
V
M
N
=
1
2
5
5
8
3
.
0
2
0
.
5
0
0
.
7
4
0
.
6
0
N
=
2
1
0
2
0
8
4
.
2
8
0
.
5
3
0
.
7
4
0
.
6
2
N
=
3
2
2
9
5
8
6
.
1
6
0
.
5
7
0
.
7
8
0
.
6
6
N
=
4
4
0
8
0
8
5
.
5
3
0
.
5
6
0
.
7
4
0
.
6
3
M
L
B
TD
+
Li
n
e
a
r
S
V
M
N
=
1
2
5
5
8
8
.
0
5
0
.
6
1
0
.
8
1
0
.
7
0
N
=
2
1
0
2
0
8
9
.
3
1
0
.
6
5
0
.
8
1
0
.
7
2
N
=
3
2
2
9
5
9
0
.
5
7
0
.
6
8
0
.
8
5
0
.
7
5
N
=
4
4
0
8
0
8
9
.
9
4
0
.
6
7
0
.
8
1
0
.
7
3
LB
P
+
K
N
N
N
=
1
2
5
5
7
6
.
7
3
0
.
3
8
0
.
5
9
0
.
4
6
N
=
2
1
0
2
0
7
9
.
2
5
0
.
4
3
0
.
6
7
0
.
5
2
N
=
3
2
2
9
5
8
2
.
2
5
0
.
4
6
0
.
7
0
0
.
5
6
N
=
4
4
0
8
0
7
9
.
8
7
0
.
4
4
0
.
6
7
0
.
5
3
M
L
B
TD
+
K
N
N
N
=
1
2
5
5
8
5
.
5
3
0
.
5
6
0
.
7
4
0
.
6
3
N
=
2
1
0
2
0
8
7
.
4
2
0
.
6
0
0
.
7
8
0
.
6
8
N
=
3
2
2
9
5
8
8
.
6
8
0
.
6
3
0
.
8
1
0
.
7
1
N
=
4
4
0
8
0
8
6
.
7
9
0
.
5
8
0
.
7
8
0
.
6
7
LB
P
+
C
T
N
=
1
2
5
5
7
6
.
7
3
0
.
3
9
0
.
6
3
0
.
4
8
N
=
2
1
0
2
0
7
9
.
5
0
0
.
4
3
0
.
7
0
0
.
5
4
N
=
3
2
2
9
5
8
2
.
8
4
0
.
4
8
0
.
7
0
0
.
5
7
N
=
4
4
0
8
0
8
0
.
5
0
0
.
4
5
0
.
7
0
0
.
5
5
M
L
B
TD
+
C
T
N
=
1
2
5
5
8
6
.
1
6
0
.
5
7
0
.
7
8
0
.
6
6
N
=
2
1
0
2
0
8
8
.
0
5
0
.
6
1
0
.
8
1
0
.
7
0
N
=
3
2
2
9
5
8
9
.
9
4
0
.
6
7
0
.
8
1
0
.
7
3
N
=
4
4
0
8
0
8
7
.
4
2
0
.
5
9
0
.
8
1
0
.
6
9
T
ab
le
2
: Co
m
p
ar
ativ
e
r
esu
lts
o
f
ML
B
T
D
with
tr
ad
itio
n
al
m
eth
o
d
s
[
38
]
F
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
S
c
h
e
me
%
A
c
c
u
r
a
c
y
C
l
a
s
si
f
i
e
r
K
N
N
CT
S
V
M
G
LC
M
6
3
.
4
5
6
6
.
5
0
6
8
.
3
0
HOG
7
2
.
4
0
7
4
.
8
0
7
8
.
8
5
LB
P
8
2
.
2
5
8
2
.
8
4
8
6
.
1
6
M
L
B
TD
8
8
.
5
6
8
9
.
9
4
9
0
.
5
7
4.
CO
NCLU
SI
O
N
T
h
is
p
ap
er
u
s
es
a
m
u
ltil
ev
el
l
o
ca
l
b
in
ar
y
tex
tu
r
e
d
escr
ip
to
r
to
id
en
tif
y
o
r
al
ca
n
ce
r
.
T
h
e
class
if
ier
s
u
tili
ze
d
f
o
r
th
e
class
if
icatio
n
i
n
clu
d
e
SVM,
KNN,
an
d
C
T
.
Dif
f
er
en
t
b
lo
ck
s
izes
an
d
r
ad
i
i
ar
e
u
s
ed
to
ass
e
s
s
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
lo
ca
l b
in
ar
y
p
atter
n
(
L
B
P)
an
d
ML
B
T
D.
8
8
.
0
5
%
ac
cu
r
ac
y
,
0
.
8
1
p
r
ec
is
io
n
,
0
.
6
1
r
e
ca
ll,
an
d
0
.
7
0
f
1
-
s
co
r
e
ar
e
o
b
tain
e
d
u
s
in
g
t
h
e
ML
B
T
D
m
eth
o
d
with
b
lo
ck
s
ize
N=
1
an
d
2
5
5
f
ea
tu
r
es.
R
esu
lts
in
clu
d
e
8
9
.
3
1
%
ac
cu
r
ac
y
,
0
.
8
1
p
r
ec
is
io
n
,
0
.
6
5
r
ec
all,
an
d
0
.
7
2
f
1
-
s
co
r
e
f
o
r
b
lo
ck
s
ize
N=
2
wit
h
1
0
2
0
f
ea
tu
r
es.
I
t
is
p
o
s
s
ib
le
to
g
et
9
0
.
5
7
%
ac
cu
r
ac
y
,
0
.
8
5
p
r
ec
is
io
n
,
0
.
6
8
r
e
ca
ll,
an
d
0
.
7
5
f
1
-
s
co
r
e
with
b
lo
c
k
s
ize
N=
3
an
d
2
2
9
5
f
ea
t
u
r
es.
B
lo
ck
s
ize
N=
4
with
4
0
8
0
f
ea
t
u
r
es
y
ield
s
8
9
.
9
4
%
ac
cu
r
ac
y
,
0
.
8
1
p
r
ec
is
io
n
,
0
.
6
7
r
ec
all,
an
d
0
.
7
3
f
1
-
s
co
r
e.
C
o
n
s
eq
u
e
n
tly
,
it
is
d
is
co
v
er
ed
th
a
t
f
o
r
an
eq
u
iv
alen
t
n
u
m
b
er
o
f
f
ea
tu
r
es,
ML
B
T
D
o
u
tp
er
f
o
r
m
s
L
B
P
b
y
o
b
s
er
v
in
g
th
e
p
e
r
f
o
r
m
an
ce
p
ar
am
e
ter
s
:
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
e
ca
ll,
an
d
f
1
-
s
co
r
e.
Fu
r
th
er
m
o
r
e
,
co
m
p
a
r
ed
to
t
h
e
KNN
an
d
C
T
class
if
ier
s
,
ML
B
T
D
with
th
e
SVM
cla
s
s
if
ier
p
er
f
o
r
m
s
b
etter
.
T
h
u
s
,
in
th
e
f
u
tu
r
e
,
th
e
f
o
cu
s
c
an
b
e
g
iv
e
n
to
im
p
r
o
v
in
g
th
e
f
ea
tu
r
e
r
e
p
r
esen
tatio
n
u
s
in
g
D
L
-
b
ased
alg
o
r
ith
m
s
an
d
e
n
h
an
ci
n
g
t
h
e
i
n
ter
p
r
etab
ilit
y
an
d
ex
p
lain
a
b
ilit
y
o
f
t
h
e
s
y
s
tem
s
.
T
h
e
ef
f
ec
ti
v
en
ess
o
f
th
e
class
if
ier
s
ca
n
b
e
b
o
o
s
ted
b
y
c
o
m
b
i
n
i
n
g
t
h
e
co
lo
r
a
n
d
s
h
a
p
e
f
ea
tu
r
es
with
th
e
tex
t
u
r
e
f
ea
tu
r
es.
I
n
t
h
e
f
u
tu
r
e,
d
ee
p
lear
n
in
g
alg
o
r
ith
m
s
ca
n
im
p
r
o
v
e
f
ea
tu
r
e
d
ep
ictio
n
a
n
d
ad
d
r
ess
class
i
m
b
alan
ce
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
s
tate
n
o
f
u
n
d
in
g
in
v
o
lv
ed
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
N
o
ve
l m
u
ltil
ev
el
lo
ca
l b
in
a
r
y
textu
r
e
d
escr
ip
to
r
fo
r
o
r
a
l c
a
n
ce
r
…
(
V
ija
ya
Ya
d
u
v
a
n
s
h
i
)
843
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Vijay
a
Yad
u
v
an
s
h
i
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
R
am
an
Mu
r
u
g
a
n
✓
✓
✓
✓
C
:
C
o
n
c
e
p
tu
a
li
z
a
ti
o
n
M
:
M
e
th
o
d
o
l
o
g
y
So
:
So
ftwa
re
Va
:
Va
li
d
a
ti
o
n
Fo
:
Fo
rm
a
l
a
n
a
ly
sis
I
:
I
n
v
e
stig
a
ti
o
n
R
:
R
e
so
u
rc
e
s
D
:
D
a
ta Cu
ra
ti
o
n
O
:
Wr
it
in
g
-
O
ri
g
in
a
l
Dra
ft
E
:
Wr
it
in
g
-
Re
v
iew
&
E
d
it
i
n
g
Vi
:
Vi
su
a
li
z
a
ti
o
n
Su
:
Su
p
e
rv
isi
o
n
P
:
P
ro
jec
t
a
d
m
in
istrati
o
n
Fu
:
Fu
n
d
in
g
a
c
q
u
isi
ti
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.
DATA AV
AI
L
AB
I
L
I
T
Y
Data
av
ailab
ilit
y
is
n
o
t
ap
p
li
ca
b
le
to
th
is
p
ap
er
as
n
o
n
e
w
d
ata
wer
e
cr
ea
ted
o
r
an
aly
ze
d
in
th
is
s
tu
d
y
.
RE
F
E
R
E
NC
E
S
[
1
]
I
n
t
e
r
n
a
t
i
o
n
a
l
a
g
e
n
c
y
f
o
r
r
e
s
e
a
r
c
h
on
c
a
n
c
e
r
.
9
0
0
W
o
r
l
d
F
a
c
t
S
h
e
e
t
s.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
g
c
o
.
i
a
r
c
.
f
r
/
t
o
d
a
y
/
d
a
t
a
/
f
a
c
t
s
h
e
e
t
s
/
p
o
p
u
l
a
t
i
o
n
s/
9
0
0
-
w
o
r
l
d
-
f
a
c
t
-
s
h
e
e
t
s
.
p
d
f
(
A
c
c
e
sse
d
:
27
A
u
g
u
s
t
2
0
2
0
)
.
[
2
]
C
.
G
.
-
C
l
a
r
k
e
,
K
.
W
.
C
h
e
n
,
a
n
d
J.
W
i
l
c
o
c
k
,
"
D
i
a
g
n
o
si
s
a
n
d
r
e
f
e
r
r
a
l
d
e
l
a
y
s
i
n
p
r
i
m
a
r
y
c
a
r
e
f
o
r
o
r
a
l
s
q
u
a
m
o
u
s
c
e
l
l
c
a
n
c
e
r
:
a
sy
st
e
ma
t
i
c
r
e
v
i
e
w
,
”
Bri
t
i
s
h
J
o
u
r
n
a
l
o
f
G
e
n
e
ra
l
P
ra
c
t
i
c
e
,
v
o
l
.
6
9
,
n
o
.
6
7
9
,
p
p
.
e
1
1
2
-
e
1
2
6
,
2
0
1
9
,
d
o
i
:
1
0
.
3
3
9
9
/
b
j
g
p
1
8
X
7
0
0
2
0
5
.
[
3
]
J.
S
e
o
a
n
e
e
t
a
l
.
,
"
Ea
r
l
y
o
r
a
l
c
a
n
c
e
r
d
i
a
g
n
o
s
i
s:
T
h
e
A
a
r
h
u
s
s
t
a
t
e
me
n
t
p
e
r
s
p
e
c
t
i
v
e
.
A
sy
s
t
e
m
a
t
i
c
r
e
v
i
e
w
a
n
d
me
t
a
‐
a
n
a
l
y
s
i
s
,
"
H
e
a
d
&
n
e
c
k
,
v
o
l
.
38
,
p
p
.
E
2
1
8
2
-
E
2
1
8
9
,
2
0
1
6
,
d
o
i
:
1
0
.
1
0
0
2
/
h
e
d
.
2
4
0
5
0
.
[
4
]
S
.
W
a
r
n
a
k
u
l
a
su
r
i
y
a
a
n
d
J.
S
.
G
r
e
e
n
s
p
a
n
,
“
T
e
x
t
b
o
o
k
o
f
o
r
a
l
c
a
n
c
e
r:
Pre
v
e
n
t
i
o
n
,
d
i
a
g
n
o
s
i
s
a
n
d
m
a
n
a
g
e
m
e
n
t
,”
v
o
l
.
1
.
N
e
w
Y
o
r
k
,
N
Y
,
U
S
A
:
S
p
r
i
n
g
e
r
,
2
0
2
0
.
[
5
]
C
.
S
c
u
l
l
y
,
J.
B
a
g
a
n
,
C
.
H
o
p
p
e
r
a
n
d
J
.
B
.
Ep
st
e
i
n
,
"
O
r
a
l
c
a
n
c
e
r
:
c
u
r
r
e
n
t
a
n
d
f
u
t
u
r
e
d
i
a
g
n
o
s
t
i
c
t
e
c
h
n
i
q
u
e
s,"
Am
J
D
e
n
t
,
v
o
l
.
2
1
,
n
o
.
4
,
p
p
.
1
9
9
-
2
0
9
,
2
0
0
8
.
[
6
]
J.
R
e
n
,
M
.
Q
i
,
Y
.
Y
u
a
n
,
S
.
D
u
a
n
a
n
d
X
.
Ta
o
,
"
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
-
b
a
s
e
d
M
R
I
t
e
x
t
u
r
e
a
n
a
l
y
s
i
s t
o
p
r
e
d
i
c
t
t
h
e
h
i
s
t
o
l
o
g
i
c
g
r
a
d
e
o
f
o
r
a
l
sq
u
a
m
o
u
s
c
e
l
l
c
a
r
c
i
n
o
ma,
"
Am
e
r
i
c
a
n
J
o
u
rn
a
l
o
f
Ro
e
n
t
g
e
n
o
l
o
g
y
,
v
o
l
.
2
1
5
,
n
o
.
5
,
p
p
.
1
1
8
4
-
1
1
9
0
,
2
0
2
0
,
d
o
i
:
1
0
.
2
2
1
4
/
A
JR
.
1
9
.
2
2
5
9
3
.
[
7
]
J.
S
h
a
n
e
t
a
l
.
,
"
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
p
r
e
d
i
c
t
s
l
y
m
p
h
n
o
d
e
m
e
t
a
s
t
a
s
i
s
i
n
e
a
r
l
y
-
s
t
a
g
e
o
r
a
l
t
o
n
g
u
e
s
q
u
a
m
o
u
s
c
e
l
l
c
a
r
c
i
n
o
ma
.
"
J
o
u
r
n
a
l
o
f
O
ra
l
a
n
d
M
a
x
i
l
l
o
f
a
c
i
a
l
S
u
r
g
e
r
y
,
v
o
l
.
7
8
,
n
o
.
1
2
,
p
p
.
2
2
0
8
-
2
2
1
8
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
o
ms.
2
0
2
0
.
0
6
.
0
1
5
.
[
8
]
C
.
S
.
C
h
u
,
N
.
P
.
Le
e
,
J.
A
d
e
o
y
e
,
P
.
T
h
o
ms
o
n
a
n
d
S
.
-
W
.
C
h
o
i
,
"
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
t
r
e
a
t
m
e
n
t
o
u
t
c
o
m
e
p
r
e
d
i
c
t
i
o
n
f
o
r
o
r
a
l
c
a
n
c
e
r
,
"
J
o
u
r
n
a
l
o
f
O
r
a
l
Pa
t
h
o
l
o
g
y
&
Me
d
i
c
i
n
e
,
v
o
l
.
4
9
,
n
o
.
1
0
,
p
p
.
9
7
7
-
9
8
5
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
1
1
/
j
o
p
.
1
3
0
8
9
.
[
9
]
X
.
A
.
L.
-
C
o
r
t
é
s,
F
.
M
a
t
a
ma
l
a
,
B
.
V
e
n
e
g
a
s
a
n
d
C
.
R
i
v
e
r
a
,
"
M
a
c
h
i
n
e
-
l
e
a
r
n
i
n
g
a
p
p
l
i
c
a
t
i
o
n
s
i
n
o
r
a
l
c
a
n
c
e
r
:
a
sy
s
t
e
ma
t
i
c
r
e
v
i
e
w
,
"
Ap
p
l
i
e
d
S
c
i
e
n
c
e
s
,
v
o
l
.
1
2
,
n
o
.
1
1
,
p
.
5
7
1
5
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
1
2
1
1
5
7
1
5
.
[
1
0
]
M
.
M
.
R
.
K
r
i
s
h
n
a
n
e
t
a
l
.
,
"
A
u
t
o
m
a
t
e
d
o
r
a
l
c
a
n
c
e
r
i
d
e
n
t
i
f
i
c
a
t
i
o
n
u
s
i
n
g
h
i
s
t
o
p
a
t
h
o
l
o
g
i
c
a
l
i
ma
g
e
s
:
a
h
y
b
r
i
d
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
p
a
r
a
d
i
g
m
,
"
M
i
c
r
o
n
,
v
o
l
.
4
3
,
n
o
.
2
-
3
,
p
p
.
3
5
2
-
3
6
4
,
2
0
1
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
m
i
c
r
o
n
.
2
0
1
1
.
0
9
.
0
1
6
.
[
1
1
]
M
.
A
u
b
r
e
v
i
l
l
e
e
t
a
l
.
,
"
A
u
t
o
ma
t
i
c
c
l
a
s
si
f
i
c
a
t
i
o
n
o
f
c
a
n
c
e
r
o
u
s
t
i
ss
u
e
i
n
l
a
ser
e
n
d
o
mi
c
r
o
s
c
o
p
y
i
m
a
g
e
s
o
f
t
h
e
o
r
a
l
c
a
v
i
t
y
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
"
S
c
i
e
n
t
i
f
i
c
Re
p
o
r
t
s
,
p
p
.
1
-
1
0
,
2
0
1
7
.
[
1
2
]
J.
F
o
l
ms
b
e
e
,
X
.
L
i
u
,
M
.
B
r
a
n
d
w
e
i
n
-
W
e
b
e
r
a
n
d
S
.
D
o
y
l
e
,
"
A
c
t
i
v
e
d
e
e
p
l
e
a
r
n
i
n
g
:
I
mp
r
o
v
e
d
t
r
a
i
n
i
n
g
e
f
f
i
c
i
e
n
c
y
o
f
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s
f
o
r
t
i
ssu
e
c
l
a
ssi
f
i
c
a
t
i
o
n
i
n
o
r
a
l
c
a
v
i
t
y
c
a
n
c
e
r
,
"
2
0
1
8
I
EEE
1
5
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
S
y
m
p
o
si
u
m
o
n
Bi
o
m
e
d
i
c
a
l
I
m
a
g
i
n
g
(
I
S
BI
2
0
1
8
)
,
W
a
s
h
i
n
g
t
o
n
,
D
C
,
U
S
A
,
2
0
1
8
,
p
p
.
7
7
0
-
7
7
3
,
d
o
i
:
1
0
.
1
1
0
9
/
I
S
B
I
.
2
0
1
8
.
8
3
6
3
6
8
6
.
[
1
3
]
R
.
K
.
G
u
p
t
a
,
M
.
K
a
u
r
a
n
d
J.
M
a
n
h
a
s,
"
T
i
s
s
u
e
l
e
v
e
l
b
a
se
d
d
e
e
p
l
e
a
r
n
i
n
g
f
r
a
m
e
w
o
r
k
f
o
r
e
a
r
l
y
d
e
t
e
c
t
i
o
n
o
f
d
y
s
p
l
a
si
a
i
n
o
r
a
l
sq
u
a
m
o
u
s
e
p
i
t
h
e
l
i
u
m
,
"
J
o
u
r
n
a
l
o
f
Mu
l
t
i
m
e
d
i
a
I
n
f
o
rm
a
t
i
o
n
S
y
s
t
e
m
,
v
o
l
.
6
,
n
o
.
2
,
p
p
.
8
1
-
8
6
,
2
0
1
9
,
d
o
i
:
1
0
.
3
3
8
5
1
/
JM
I
S
.
2
0
1
9
.
6
.
2
.
8
1
.
[
1
4
]
P
.
R
.
J
e
y
a
r
a
j
a
n
d
E.
R
.
S
.
N
a
d
a
r
,
"
C
o
mp
u
t
e
r
-
a
ss
i
st
e
d
me
d
i
c
a
l
i
ma
g
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
f
o
r
e
a
r
l
y
d
i
a
g
n
o
s
i
s
o
f
o
r
a
l
c
a
n
c
e
r
e
m
p
l
o
y
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
m,"
J
o
u
r
n
a
l
o
f
c
a
n
c
e
r
r
e
se
a
rc
h
a
n
d
c
l
i
n
i
c
a
l
o
n
c
o
l
o
g
y
,
v
o
l
.
1
4
5
,
n
o
.
4
,
p
p
.
8
2
9
-
8
3
7
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/
s
0
0
4
3
2
-
0
1
8
-
0
2
8
3
4
-
7.
[
1
5
]
S
.
X
u
e
t
a
l
.
,
"
A
n
Ea
r
l
y
D
i
a
g
n
o
si
s
o
f
o
r
a
l
c
a
n
c
e
r
b
a
s
e
d
o
n
t
h
r
e
e
-
d
i
m
e
n
s
i
o
n
a
l
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s,"
i
n
I
EEE
Ac
c
e
ss
,
v
o
l
.
7
,
p
p
.
1
5
8
6
0
3
-
1
5
8
6
1
1
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
1
9
.
2
9
5
0
2
8
6
.
[
1
6
]
B
.
S
o
n
g
e
t
a
l
.
,
"
A
u
t
o
m
a
t
i
c
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
d
u
a
l
-
mo
d
a
l
i
l
t
y
,
s
mart
p
h
o
n
e
-
b
a
s
e
d
o
r
a
l
d
y
s
p
l
a
si
a
a
n
d
mal
i
g
n
a
n
c
y
i
mag
e
s
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
"
B
i
o
m
e
d
i
c
a
l
O
p
t
i
c
s E
x
p
r
e
ss
,
v
o
l
.
9
,
n
o
.
1
1
,
p
p
.
5
3
1
8
–
5
3
2
9
,
2
0
1
8
,
d
o
i
:
1
0
.
1
3
6
4
/
B
O
E.
9
.
0
0
5
3
1
8
.
[
1
7
]
R
o
ss
D
U
t
h
o
f
f
e
t
a
l
.
,
"
P
o
i
n
t
-
of
-
c
a
r
e
,
smar
t
p
h
o
n
e
-
b
a
se
d
,
d
u
a
l
-
m
o
d
a
l
i
t
y
,
d
u
a
l
-
v
i
e
w
,
o
r
a
l
c
a
n
c
e
r
scre
e
n
i
n
g
d
e
v
i
c
e
w
i
t
h
n
e
u
r
a
l
n
e
t
w
o
r
k
c
l
a
ss
i
f
i
c
a
t
i
o
n
f
o
r
l
o
w
-
r
e
s
o
u
r
c
e
c
o
mm
u
n
i
t
i
e
s
,
"
P
l
o
S
o
n
e
,
v
o
l
.
1
3
,
n
o
.
1
2
,
2
0
1
8
,
d
o
i
:
0
.
1
3
7
1
/
j
o
u
r
n
a
l
.
p
o
n
e
.
0
2
0
7
4
9
3
.
[
1
8
]
A
.
R
a
n
a
,
G
.
Y
a
u
n
e
y
,
L
.
C
.
W
o
n
g
,
O
.
G
u
p
t
a
,
A
.
M
u
f
t
u
a
n
d
P
.
S
h
a
h
,
"
A
u
t
o
ma
t
e
d
se
g
me
n
t
a
t
i
o
n
o
f
g
i
n
g
i
v
a
l
d
i
s
e
a
s
e
s
f
r
o
m
o
r
a
l
i
ma
g
e
s,"
2
0
1
7
I
EE
E
H
e
a
l
t
h
c
a
re
I
n
n
o
v
a
t
i
o
n
s
a
n
d
Po
i
n
t
o
f
C
a
r
e
T
e
c
h
n
o
l
o
g
i
e
s
(
H
I
-
PO
C
T
)
,
B
e
t
h
e
s
d
a
,
M
D
,
U
S
A
,
2
0
1
7
,
p
p
.
1
4
4
-
1
4
7
,
d
o
i
:
1
0
.
1
1
0
9
/
H
I
C
.
2
0
1
7
.
8
2
2
7
6
0
5
.
[
1
9
]
B
.
T
h
o
mas
,
V
.
K
u
mar a
n
d
S
.
S
a
i
n
i
,
"
Te
x
t
u
r
e
a
n
a
l
y
si
s b
a
s
e
d
se
g
me
n
t
a
t
i
o
n
a
n
d
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
o
r
a
l
c
a
n
c
e
r
l
e
si
o
n
s
i
n
c
o
l
o
r
i
m
a
g
e
s
u
si
n
g
A
N
N
,
"
2
0
1
3
I
E
EE
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
S
i
g
n
a
l
Pro
c
e
ss
i
n
g
,
C
o
m
p
u
t
i
n
g
a
n
d
C
o
n
t
r
o
l
(
I
S
P
C
C
)
,
S
o
l
a
n
,
I
n
d
i
a
,
2
0
1
3
,
p
p
.
1
-
5
,
d
o
i
:
1
0
.
1
1
0
9
/
I
S
P
C
C
.
2
0
1
3
.
6
6
6
3
4
0
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
4
,
No
.
3
,
Dec
em
b
er
20
2
5
:
837
-
8
4
4
844
[
2
0
]
R
.
A
n
a
n
t
h
a
r
a
m
a
n
,
V
.
A
n
a
n
t
h
a
r
a
ma
n
a
n
d
Y
.
Le
e
,
"
O
r
o
v
i
s
i
o
n
:
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
c
l
a
ss
i
f
y
i
n
g
o
r
o
f
a
c
i
a
l
d
i
sea
ses,
"
2
0
1
7
I
EEE
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
H
e
a
l
t
h
c
a
re
I
n
f
o
rm
a
t
i
c
s
(
I
C
H
I
)
,
P
a
r
k
C
i
t
y
,
U
T,
U
S
A
,
2
0
1
7
,
p
p
.
3
9
-
4
5
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
H
I
.
2
0
1
7
.
6
9
.
[
2
1
]
R
.
A
n
a
n
t
h
a
r
a
ma
n
,
M
.
V
e
l
a
z
q
u
e
z
a
n
d
Y
.
Le
e
,
"
U
t
i
l
i
z
i
n
g
mas
k
R
-
C
N
N
f
o
r
d
e
t
e
c
t
i
o
n
a
n
d
se
g
me
n
t
a
t
i
o
n
o
f
o
r
a
l
d
i
se
a
ses
,
"
2
0
1
8
I
EEE
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
Bi
o
i
n
f
o
rm
a
t
i
c
s
a
n
d
B
i
o
m
e
d
i
c
i
n
e
(
BI
BM
)
,
M
a
d
r
i
d
,
S
p
a
i
n
,
2
0
1
8
,
p
p
.
2
1
9
7
-
2
2
0
4
,
d
o
i
:
1
0
.
1
1
0
9
/
B
I
B
M
.
2
0
1
8
.
8
6
2
1
1
1
2
.
[
2
2
]
J.
D
e
n
g
,
W
.
D
o
n
g
,
R
.
S
o
c
h
e
r
,
L
.
-
J.
L
i
,
K
a
i
L
i
a
n
d
Li
F
e
i
-
F
e
i
,
"
I
mag
e
N
e
t
:
A
l
a
r
g
e
-
s
c
a
l
e
h
i
e
r
a
r
c
h
i
c
a
l
i
ma
g
e
d
a
t
a
b
a
s
e
,
"
2
0
0
9
I
EE
E
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
e
r
V
i
si
o
n
a
n
d
P
a
t
t
e
r
n
Re
c
o
g
n
i
t
i
o
n
,
M
i
a
mi
,
F
L
,
U
S
A
,
2
0
0
9
,
p
p
.
2
4
8
-
2
5
5
,
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
.
2
0
0
9
.
5
2
0
6
8
4
8
.
[
2
3
]
A
.
K
r
i
z
h
e
v
s
k
y
,
I
.
S
u
t
s
k
e
v
e
r
a
n
d
G
.
E.
H
i
n
t
o
n
,
"
I
mag
e
N
e
t
c
l
a
ssi
f
i
c
a
t
i
o
n
w
i
t
h
d
e
e
p
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s
,
"
A
d
v
a
n
c
e
s
i
n
N
e
u
ra
l
I
n
f
o
rm
a
t
i
o
n
Pr
o
c
e
ssi
n
g
S
y
s
t
e
m
s
,
v
o
l
.
2
5
,
p
p
.
1
0
9
7
-
1
1
0
5
,
2
0
1
2
.
[
2
4
]
M
.
E
v
e
r
i
n
g
h
a
m,
S
.
M
.
A
.
Es
l
a
m
i
,
L.
V
.
G
o
o
l
,
C
h
.
K
.
I
.
W
i
l
l
i
a
ms
,
J
.
W
i
n
n
a
n
d
A
.
Z
i
sserm
a
n
,
"
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
C
o
m
p
u
t
e
r
Vi
si
o
n
,
v
o
l
.
1
1
1
,
p
p
.
9
8
-
1
3
6
,
2
0
1
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
2
6
3
-
0
1
4
-
0
7
3
3
-
5
.
[
2
5
]
T.
-
Y
.
L
i
n
,
"
M
i
c
r
o
s
o
f
t
C
O
C
O
:
c
o
mm
o
n
o
b
j
e
c
t
s
i
n
c
o
n
t
e
x
t
,
"
Eu
r
o
p
e
a
n
c
o
n
f
e
ren
c
e
o
n
c
o
m
p
u
t
e
r
v
i
si
o
n
,
S
p
r
i
n
g
e
r
,
C
h
a
m,
2
0
1
4
,
d
o
i
:
1
0
.
4
8
5
5
0
/
a
r
X
i
v
.
1
4
0
5
.
0
3
1
2
.
[
2
6
]
R
.
G
i
r
s
h
i
c
k
,
J.
D
o
n
a
h
u
e
,
T.
D
a
r
r
e
l
l
a
n
d
J.
M
a
l
i
k
,
"
R
i
c
h
f
e
a
t
u
r
e
h
i
e
r
a
r
c
h
i
e
s
f
o
r
a
c
c
u
r
a
t
e
o
b
j
e
c
t
d
e
t
e
c
t
i
o
n
a
n
d
s
e
ma
n
t
i
c
seg
m
e
n
t
a
t
i
o
n
,
"
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
I
E
EE
c
o
n
f
.
o
n
c
o
m
p
u
t
e
r v
i
si
o
n
a
n
d
p
a
t
t
e
rn
rec
o
g
n
i
t
i
o
n
,
2
0
1
4
,
d
o
i
:
1
0
.
4
8
5
5
0
/
a
r
X
i
v
.
1
3
1
1
.
2
5
2
4
[
2
7
]
R
.
G
i
r
s
h
i
c
k
,
"
F
a
s
t
R
-
C
N
N
,
"
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
I
E
EE
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
e
r
Vi
si
o
n
(
I
C
C
V)
,
2
0
1
5
,
p
p
.
1
4
4
0
-
1
4
4
8
.
[
2
8
]
S
.
R
e
n
,
K
.
H
e
,
R
.
G
i
r
sh
i
c
k
a
n
d
J
.
S
u
n
,
"
F
a
s
t
e
r
R
-
C
N
N
:
t
o
w
a
r
d
s
r
e
a
l
-
t
i
me
o
b
j
e
c
t
d
e
t
e
c
t
i
o
n
w
i
t
h
r
e
g
i
o
n
p
r
o
p
o
s
a
l
n
e
t
w
o
r
k
s
,
"
Ad
v
a
n
c
e
s i
n
n
e
u
r
a
l
i
n
f
o
rm
a
t
i
o
n
p
ro
c
e
ssi
n
g
s
y
st
e
m
s
,
v
o
l
2
8
,
p
p
.
9
1
-
9
9
,
2
0
1
5
,
d
o
i
:
1
0
.
4
8
5
5
0
/
a
r
X
i
v
.
1
5
0
6
.
0
1
4
9
7
.
[
2
9
]
K
.
H
e
,
G
.
G
k
i
o
x
a
r
i
,
P
.
D
o
l
l
á
r
a
n
d
R
.
G
i
r
sh
i
c
k
,
‘
‘
M
a
s
k
R
-
C
N
N
,
’
’
Pr
o
c
.
I
E
EE
I
n
t
.
C
o
n
f
.
C
o
m
p
u
t
.
V
i
s.
,
p
p
.
2
9
6
1
–
2
9
6
9
,
2
0
1
7
.
[
3
0
]
W
.
Li
u
e
t
a
l
.
,
"
S
S
D
:
S
i
n
g
l
e
S
h
o
t
M
u
l
t
i
B
o
x
D
e
t
e
c
t
o
r
,
"
E
u
r
o
p
e
a
n
c
o
n
f
e
r
e
n
c
e
o
n
c
o
m
p
u
t
e
r
v
i
s
i
o
n
,
S
p
r
i
n
g
e
r
,
C
h
a
m,
2
0
1
6
.
[
3
1
]
Y
.
S
h
i
n
,
H
.
A
.
Q
a
d
i
r
,
L
.
A
a
b
a
k
k
e
n
,
J
.
B
e
r
g
sl
a
n
d
a
n
d
I
.
B
a
l
a
si
n
g
h
a
m,
"
A
u
t
o
ma
t
i
c
c
o
l
o
n
p
o
l
y
p
d
e
t
e
c
t
i
o
n
u
si
n
g
r
e
g
i
o
n
b
a
se
d
d
e
e
p
C
N
N
a
n
d
p
o
st
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s,"
i
n
I
EE
E
A
c
c
e
ss
,
v
o
l
.
6
,
p
p
.
4
0
9
5
0
-
4
0
9
6
2
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
1
8
.
2
8
5
6
4
0
2
.
[
3
2
]
D
.
R
i
b
l
i
,
A
.
H
o
r
v
á
t
h
,
Z
.
a
U
n
g
e
r
,
P
.
P
o
l
l
n
e
r
a
n
d
I
.
C
sa
b
a
i
,
"
D
e
t
e
c
t
i
n
g
a
n
d
c
l
a
ssi
f
y
i
n
g
l
e
s
i
o
n
s i
n
m
a
mm
o
g
r
a
ms w
i
t
h
d
e
e
p
l
e
a
r
n
i
n
g
,
"
S
c
i
e
n
t
i
f
i
c
Re
p
o
r
t
s
,
2
0
1
8
.
[
3
3
]
N
.
A
.
-
R
a
w
i
e
t
a
l
.
,
"
Th
e
e
f
f
e
c
t
i
v
e
n
e
ss
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
d
e
t
e
c
t
i
o
n
o
f
o
r
a
l
C
a
n
c
e
r
,
"
I
n
t
e
r
n
a
t
i
o
n
a
l
D
e
n
t
a
l
J
o
u
r
n
a
l
,
v
o
l
.
7
2
,
n
o
.
4
,
p
p
.
4
3
6
-
4
4
7
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
d
e
n
t
j
.
2
0
2
2
.
0
3
.
0
0
1
.
[
3
4
]
K
.
H
e
,
X
.
Z
h
a
n
g
,
S
.
R
e
n
a
n
d
J.
S
u
n
,
"
D
e
e
p
r
e
s
i
d
u
a
l
l
e
a
r
n
i
n
g
f
o
r
i
ma
g
e
r
e
c
o
g
n
i
t
i
o
n
,
"
2
0
1
6
I
EEE
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
e
r
Vi
s
i
o
n
a
n
d
Pa
t
t
e
rn
Re
c
o
g
n
i
t
i
o
n
(
C
V
PR)
,
La
s
V
e
g
a
s,
N
V
,
U
S
A
,
2
0
1
6
,
p
p
.
7
7
0
-
7
7
8
,
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
.
2
0
1
6
.
9
0
.
[
3
5
]
V
.
Y
a
d
u
v
a
n
s
h
i
,
R
.
M
u
r
u
g
a
n
a
n
d
Tr
i
p
t
i
G
o
e
l
,
"
A
n
a
u
t
o
ma
t
i
c
c
l
a
ssi
f
i
c
a
t
i
o
n
met
h
o
d
s
i
n
o
r
a
l
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
,
"
H
e
a
l
t
h
I
n
f
o
rm
a
t
i
c
s:
A
C
o
m
p
u
t
a
t
i
o
n
a
l
P
e
rsp
e
c
t
i
v
e
i
n
H
e
a
l
t
h
c
a
re
,
p
p
.
1
3
3
-
1
5
8
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
9
8
1
-
15
-
9
7
3
5
-
0
_
8
.
[
3
6
]
V
.
Y
a
d
u
v
a
n
s
h
i
,
R
.
M
u
r
u
g
a
n
a
n
d
T.
G
o
e
l
,
"
A
u
t
o
ma
t
i
c
o
r
a
l
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
a
n
d
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
s
i
n
g
m
o
d
i
f
i
e
d
l
o
c
a
l
t
e
x
t
u
r
e
d
e
s
c
r
i
p
t
o
r
a
n
d
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
,
"
M
u
l
t
i
m
e
d
i
a
T
o
o
l
s
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
8
4
,
p
p
.
1
0
3
1
-
1
0
5
5
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
0
4
2
-
0
2
4
-
1
9
0
4
0
-
y
.
[
3
7
]
M
e
n
d
e
l
e
y
D
a
t
a
se
t
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
d
a
t
a
.
m
e
n
d
e
l
e
y
.
c
o
m/
d
a
t
a
s
e
t
s
/
f
t
mp
4
c
v
t
m
b
/
1
(
A
c
c
e
s
s
d
a
t
e
:
0
1
J
u
n
e
,
2
0
2
3
)
[
3
8
]
V
.
L.
K
o
u
z
n
e
t
so
v
a
,
J.
L
i
,
E.
R
o
mm
a
n
d
I
g
o
r
F
.
Tsi
g
e
l
n
y
,
"
F
i
n
d
i
n
g
d
i
st
i
n
c
t
i
o
n
s
b
e
t
w
e
e
n
o
r
a
l
c
a
n
c
e
r
a
n
d
p
e
r
i
o
d
o
n
t
i
t
i
s
u
si
n
g
sa
l
i
v
a
met
a
b
o
l
i
t
e
s a
n
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
.
"
O
ra
l
d
i
se
a
ses
,
v
o
l
.
2
7
,
n
o
.
3
,
p
p
.
4
8
4
-
4
9
3
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
1
1
/
o
d
i
.
1
3
5
9
1
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Dr
.
Vija
y
a
Ya
d
u
v
a
n
shi
s
h
e
h
a
s
c
o
m
p
lete
d
h
e
r
P
h
.
D.
d
e
g
re
e
i
n
E
lec
tro
n
ics
a
n
d
Co
m
m
u
n
ica
ti
o
n
En
g
i
n
e
e
rin
g
fr
o
m
th
e
Na
ti
o
n
a
l
In
st
it
u
te
o
f
Tec
h
n
o
lo
g
y
(NIT)
,
S
il
c
h
a
r,
As
sa
m
,
In
d
ia.
He
r
m
a
in
re
se
a
rc
h
in
ter
e
sts
fo
c
u
s
o
n
m
e
d
ica
l
ima
g
e
p
ro
c
e
ss
in
g
,
d
ig
it
a
l
ima
g
e
p
ro
c
e
ss
in
g
,
m
a
c
h
i
n
e
lea
rn
in
g
,
a
n
d
d
e
e
p
lea
rn
i
n
g
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
v
ij
a
y
a
_
rs@
e
c
e
.
n
it
s.a
c
.
i
n
.
Dr
.
R.
Mu
r
u
g
a
n
h
e
is
w
o
rk
i
n
g
a
s
a
n
As
sista
n
t
P
r
o
fe
ss
o
r
in
t
h
e
De
p
a
rtme
n
t
o
f
El
e
c
tro
n
ics
a
n
d
C
o
m
m
u
n
ica
ti
o
n
En
g
i
n
e
e
rin
g
,
Na
ti
o
n
a
l
I
n
stit
u
te
o
f
Tec
h
n
o
lo
g
y
(NIT),
S
il
c
h
a
r
,
As
sa
m
,
In
d
ia,
sin
c
e
1
5
J
u
n
e
2
0
1
8
.
His
m
a
in
re
se
a
rc
h
in
tere
sts
fo
c
u
s
o
n
m
e
d
ica
l
ima
g
e
p
ro
c
e
ss
in
g
,
d
ig
i
ta
l
ima
g
e
p
r
o
c
e
ss
in
g
,
i
n
tern
e
t
o
f
th
in
g
s/m
e
d
ica
l
t
h
in
g
s,
a
u
to
m
a
ti
c
d
ri
v
e
rles
s
a
ss
istan
c
e
sy
ste
m
s,
d
e
sig
n
o
f
e
x
o
sk
e
let
o
n
s,
m
u
lt
ime
d
ia
c
o
m
p
re
ss
io
n
tec
h
n
i
q
u
e
s,
m
a
c
h
in
e
lea
rn
in
g
,
d
e
e
p
lea
rn
in
g
,
VLS
I
,
e
m
b
e
d
d
e
d
sig
n
a
l
p
ro
c
e
ss
in
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
m
u
ru
g
a
n
.
rm
n
@e
c
e
.
n
it
s.a
c
.
in
.
Evaluation Warning : The document was created with Spire.PDF for Python.