I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
1
5
,
No
.
2
,
A
p
r
il 2
0
2
6
,
p
p
.
1
9
9
9
~
2
0
0
8
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
15
.i
2
.
p
p
1
9
9
9
-
2
0
0
8
1999
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Im
pro
v
ised ma
sk fas
ter
re
curre
nt
c
o
nv
o
lutiona
l neural network
for breas
t
ca
ncer
cla
ss
ificatio
n usin
g
histopa
tholo
g
y
ima
g
es
P
a
t
t
a
n M
.
D
.
Ali K
ha
n,
Xa
v
i
er
Arput
ha
Ra
t
hin
a
D
e
p
a
r
t
me
n
t
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
a
n
d
En
g
i
n
e
e
r
i
n
g
,
B
.
S
.
A
b
d
u
r
R
a
h
ma
n
C
r
e
sce
n
t
I
n
st
i
t
u
t
e
o
f
S
c
i
e
n
c
e
a
n
d
Te
c
h
n
o
l
o
g
y
,
C
h
e
n
n
a
i
,
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
an
2
5
,
2
0
2
4
R
ev
is
ed
J
u
l 2
7
,
2
0
2
5
Acc
ep
ted
Sep
8
,
2
0
2
5
De
sp
it
e
th
e
p
re
v
a
len
c
e
o
f
t
h
is
d
i
se
a
se
,
th
e
e
x
isti
n
g
m
e
th
o
d
f
o
r
o
b
tain
in
g
a
n
e
x
a
c
t
b
re
a
st
c
a
n
c
e
r
d
iag
n
o
sis
wo
u
ld
n
e
e
d
a
l
o
t
o
f
t
ime
a
n
d
lab
o
r.
It
n
e
e
d
s
a
q
u
a
li
fied
p
a
th
o
lo
g
ist
t
o
m
a
n
u
a
l
ly
p
ro
c
e
ss
a
n
d
re
v
iew
h
ist
o
p
a
t
h
o
l
o
g
ica
l
ima
g
e
s
to
d
isti
n
g
u
ish
t
h
e
c
h
a
ra
c
teristics
th
a
t
c
h
a
ra
c
teriz
e
d
iffere
n
t
c
a
n
c
e
r
se
v
e
rit
y
lev
e
ls.
Bu
il
d
i
n
g
a
m
o
d
e
l
fo
r
a
u
t
o
m
a
ti
c
a
ll
y
d
e
tec
ti
n
g
,
se
g
m
e
n
ti
n
g
,
a
n
d
c
las
sify
in
g
b
re
a
st
les
io
n
s
u
si
n
g
h
isto
p
a
th
o
lo
g
ica
l
ima
g
e
s
se
e
m
s
to
b
e
th
e
g
o
a
l
o
f
t
h
is
wo
r
k
.
Va
rio
u
s
d
e
e
p
lea
rn
in
g
m
e
th
o
d
s
h
a
v
e
b
e
e
n
u
se
d
in
c
o
m
p
u
tati
o
n
a
l
p
a
th
o
lo
g
y
fo
r
th
e
d
iag
n
o
sis
o
f
c
a
n
c
e
r.
Im
p
ro
v
e
d
fa
ste
r
re
c
u
rre
n
t
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
t
w
o
rk
(IM
F
RCNN
)
is
a
su
p
e
rv
ise
d
lea
rn
in
g
sy
ste
m
with
p
ro
p
o
se
d
fo
r
re
c
o
g
n
izin
g
sm
a
ll
it
e
m
s
li
k
e
m
it
o
ti
c
a
n
d
n
o
n
-
m
it
o
ti
c
n
u
c
lei.
T
o
p
r
o
tec
t
sm
a
ll
it
e
m
s
fro
m
v
a
n
ish
i
n
g
i
n
t
h
e
d
e
e
p
l
a
y
e
rs,
th
is
sy
ste
m
u
se
s
e
x
p
a
n
d
e
d
la
y
e
rs
in
t
h
e
sp
in
e
.
T
o
c
lo
se
ima
g
e
a
n
d
t
h
e
th
in
g
s
g
a
p
siz
e
in
c
lu
d
e
s,
t
h
is
a
p
p
ro
a
c
h
u
s
e
s
e
x
p
a
n
d
e
d
lay
e
rs.
T
h
e
re
g
i
o
n
p
ro
p
o
sa
l
n
e
two
rk
h
a
s
b
e
e
n
c
re
a
ted
fo
r
p
re
c
ise
ti
n
y
o
b
jec
t
i
d
e
n
ti
fica
ti
o
n
.
Re
se
a
rc
h
e
rs
e
x
a
m
in
e
d
ti
m
e
fo
r
train
in
g
a
n
d
tes
ti
n
g
ti
m
e
fo
r
v
a
rio
u
s
tec
h
n
iq
u
e
s
fo
r
id
e
n
ti
f
y
in
g
o
b
jec
ts.
Th
e
to
tal
a
c
c
u
ra
c
y
o
f
b
e
n
ig
n
/ma
li
g
n
a
n
t
c
a
teg
o
riza
ti
o
n
in
p
ro
p
o
se
d
sy
ste
m
re
a
c
h
e
s
9
6
.
5
%
.
Th
e
p
r
o
p
o
se
d
tec
h
n
i
q
u
e
o
ffe
rs
a
th
o
r
o
u
g
h
a
n
d
n
o
n
-
i
n
v
a
siv
e
m
e
th
o
d
fo
r
i
d
e
n
ti
fy
in
g
a
n
d
c
a
teg
o
rize
s a
n
a
re
a
o
f
a
b
n
o
rm
a
l
b
re
a
st t
issu
e
.
K
ey
w
o
r
d
s
:
B
r
ea
s
t le
s
io
n
s
C
las
s
if
icatio
n
Dee
p
lear
n
in
g
His
to
p
ath
o
lo
g
y
im
a
g
es
I
m
p
r
o
v
ed
f
aster
r
ec
u
r
r
e
n
t
co
n
v
o
l
u
tio
n
al
n
e
u
r
al
n
etwo
r
k
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
:
Pattan
M.
D.
Ali K
h
an
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
in
ee
r
in
g
B
.
S.
Ab
d
u
r
R
ah
m
an
C
r
escen
t
I
n
s
titu
te
o
f
Scien
ce
an
d
T
ec
h
n
o
lo
g
y
C
h
en
n
ai,
I
n
d
ia
E
m
ail:
alik
_
1
2
@
r
ed
if
f
m
ail.
c
o
m
1.
I
NT
RO
D
UCT
I
O
N
Mo
d
er
n
d
ig
ital
c
o
p
ies
o
f
g
lass
s
lid
es
wer
e
ty
p
ically
tab
le
-
t
o
p
e
q
u
ip
m
e
n
t
th
at
s
ca
n
n
ed
g
l
ass
s
lid
es
an
d
p
r
o
v
id
ed
wh
o
le
-
s
lid
e
im
ag
es
q
u
ick
ly
an
d
af
f
o
r
d
ab
l
y
,
f
r
eq
u
en
tly
au
to
m
atin
g
in
ter
m
ed
iar
y
s
tag
es
lik
e
tis
s
u
e
s
eg
m
en
tatio
n
an
d
f
o
cu
s
p
lan
e
ch
o
ice
[
1
]
.
T
h
e
cr
ea
tio
n
o
f
a
h
ig
h
-
r
eso
lu
tio
n
d
ig
ital
p
h
o
to
m
icr
o
g
r
a
p
h
f
o
r
an
en
tire
h
is
to
lo
g
ical
o
r
cy
to
lo
g
y
s
lid
e
was
r
e
f
er
r
ed
to
as
wh
o
le
-
s
lid
e
p
h
o
to
g
r
ap
h
y
[
2
]
.
Alth
o
u
g
h
tu
m
o
r
s
r
em
ain
th
e
m
o
s
t
co
m
m
o
n
ty
p
e
o
f
ca
n
ce
r
in
wo
m
en
,
a
s
izab
le
p
o
r
tio
n
o
f
th
e
s
am
p
les
ex
a
m
in
ed
in
p
ath
o
lo
g
y
lab
s
co
m
e
f
r
o
m
in
d
iv
id
u
als
with
th
e
co
n
d
itio
n
[
3
]
.
Hem
ato
x
y
lin
eo
s
in
s
tain
s
h
av
e
en
d
u
r
e
d
as
th
e
g
o
-
to
s
tain
f
o
r
h
is
to
lo
g
ical
ex
am
in
atio
n
o
f
in
d
iv
id
u
al
ce
lls
.
T
h
e
f
in
e
tis
s
u
e
an
d
ce
ll
f
ea
tu
r
es
ca
n
b
e
h
ig
h
lig
h
ted
b
y
th
is
s
tr
aig
h
tf
o
r
war
d
d
y
e
m
i
x
tu
r
e
[
4
]
.
Similar
to
th
is
,
o
u
r
m
eth
o
d
h
as
m
ad
e
s
u
itab
le
th
er
ap
y
r
ec
o
m
m
en
d
atio
n
s
ac
co
r
d
in
g
to
t
h
e
less
o
n
s
lear
n
ed
.
B
r
ea
s
t
ca
n
ce
r
r
ef
er
s
to
a
d
is
ea
s
e
o
f
th
e
b
r
ain
ce
ll
th
at
ca
u
s
es
th
e
ce
ll
to
d
eter
io
r
ate
wh
ile
b
ec
o
m
in
g
in
ac
tiv
e
[
5
]
.
T
h
is
wo
u
ld
c
o
n
tr
ib
u
te
to
d
e
m
en
tia.
Dete
r
io
r
atio
n
o
f
m
en
tal,
b
eh
a
v
i
o
r
al,
an
d
ef
f
ec
tiv
e
co
m
m
u
n
icatio
n
was
d
em
en
tia
s
y
m
p
to
m
s
th
at
im
p
air
p
eo
p
le'
s
ca
p
ac
ity
f
o
r
au
to
n
o
m
o
u
s
ac
tio
n
.
C
o
g
n
itiv
e
p
r
o
b
lem
s
m
ak
e
u
p
o
n
e
o
f
th
e
Alzh
eim
er
'
s
d
is
ea
s
e
's
s
ig
n
s
.
T
h
e
clien
t
m
i
g
h
t
f
o
r
g
et
cu
r
r
en
t
o
cc
u
r
r
e
n
ce
s
in
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
2
,
Ap
r
il 2
0
2
6
:
1
9
9
9
-
2
0
0
8
2000
in
itial
s
tag
es.
A
s
th
e
illn
es
s
wo
r
s
en
s
,
s
o
m
eo
n
e
wo
u
ld
g
r
ad
u
ally
lo
s
e
d
etails
[
6
]
.
S
en
tim
en
t
an
aly
s
is
,
elec
tr
o
m
y
o
g
r
ap
h
y
s
ig
n
als
d
ia
g
n
o
s
is
,
an
d
ca
r
d
io
v
ascu
lar
illn
ess
ca
teg
o
r
izatio
n
s
f
r
o
m
an
elec
tr
o
ca
r
d
io
g
r
am
(
E
KG)
s
ig
n
al
wer
e
s
o
m
e
we
ll
-
k
n
o
wn
m
ac
h
in
e
lear
n
in
g
a
p
p
licatio
n
s
in
th
e
b
io
m
ed
ical
f
ield
[
7
]
.
Usi
n
g
a
m
icr
o
s
co
p
e,
h
is
to
p
ath
o
l
o
g
ical
im
ag
es
wer
e
ac
q
u
i
r
ed
,
an
d
s
p
ec
im
en
s
o
f
th
e
b
r
ea
s
t
tis
s
u
e
f
r
o
m
th
e
af
f
licted
ar
ea
s
wer
e
co
llected
.
Hem
ato
x
y
lin
an
d
eo
s
in
h
av
e
b
ee
n
u
tili
ze
d
to
s
tain
th
e
tis
s
u
es
to
id
en
tify
tu
m
o
r
s
[
8
]
.
E
o
s
in
co
lo
r
s
th
e
r
em
ain
d
er
o
f
th
e
ce
lls
p
in
k
,
wh
ile
h
em
ato
x
y
lin
g
iv
es
th
e
n
u
cleu
s
a
d
a
r
k
p
u
r
p
le
co
lo
r
.
B
y
in
clu
d
in
g
a
s
tr
o
n
g
av
er
a
g
e
p
e
r
ce
n
tag
e
ac
cu
r
ac
y
an
d
r
ec
o
lle
ctio
n
,
tin
y
item
s
co
u
ld
b
e
d
etec
ted
an
d
class
if
ied
u
s
in
g
th
e
r
eg
i
o
n
-
b
ased
co
n
v
o
l
u
tio
n
al
n
eu
r
al
n
etwo
r
k
(
R
C
NN)
.
T
in
y
item
s
d
is
ap
p
ea
r
in
g
i
n
im
ag
es
wo
u
ld
b
e
a
p
r
ev
alen
t
p
r
o
b
lem
with
d
ee
p
l
ea
r
n
in
g
m
o
d
els
b
ec
au
s
e
s
m
all
th
in
g
s
h
av
e
a
h
ig
h
p
r
o
p
o
r
tio
n
v
ar
iatio
n
b
etwe
en
th
e
im
ag
e
an
d
th
e
o
b
ject.
B
y
n
ea
r
ly
m
ain
tain
in
g
th
e
g
eo
m
e
tr
y
o
f
th
e
n
u
cleu
s
,
t
h
e
R
C
NN
m
in
im
izes
th
e
lo
s
s
o
f
n
u
clei.
T
h
is
tech
n
iq
u
e
r
ed
u
ce
d
th
e
f
r
eq
u
en
cy
o
f
f
alse
n
eg
ativ
es
an
d
f
alse
p
o
s
itiv
es
wh
ile
aid
in
g
i
n
th
e
r
ec
o
v
er
y
o
f
n
u
cleu
s
ch
a
r
ac
ter
i
s
tics
f
r
o
m
th
e
b
asal la
y
er
.
I
m
ag
in
g
d
iag
n
o
s
tics
in
clu
d
e
b
r
ea
s
t
m
ag
n
et
r
eso
n
a
n
ce
im
a
g
in
g
(
MRI)
,
m
am
m
o
g
r
ap
h
y
,
an
d
b
r
ea
s
t
u
ltra
-
s
o
n
o
g
r
ap
h
y
h
as
b
ec
o
m
e
th
e
p
r
ef
er
r
ed
m
et
h
o
d
o
f
b
r
e
ast
s
cr
ee
n
in
g
.
Var
io
u
s
im
ag
in
g
m
o
d
alities
wer
e
lin
k
ed
to
v
a
r
io
u
s
in
d
icatio
n
s
[
9
]
,
[
1
0
]
.
So
f
t
tis
s
u
e
lesi
o
n
s
ca
n
ea
s
ily
b
e
d
etec
ted
with
MR
I
f
o
r
s
cr
ee
n
i
n
g
t
h
e
b
r
ea
s
t
ca
n
ce
r
.
Ho
wev
e
r
,
it'
s
e
x
p
en
s
iv
e
,
h
as
a
p
r
o
p
e
n
s
ity
f
o
r
f
alse
p
o
s
itiv
es
,
an
d
tak
es
a
wh
ile
to
s
ca
n
.
As
a
r
esu
lt,
b
r
ea
s
t
MRI
wa
s
p
r
im
ar
ily
ad
v
is
ed
f
o
r
wo
m
e
n
h
av
e
a
h
ig
h
r
is
k
o
f
d
ev
elo
p
in
g
b
r
ea
s
t
ca
n
ce
r
[
1
1
]
.
Ma
m
m
o
g
r
a
p
h
y
h
as
lim
its
f
o
r
th
o
s
e
with
d
en
s
e
b
r
ea
s
t
tis
s
u
e
b
ec
au
s
e
o
f
its
g
r
ea
t
s
en
s
itiv
ity
to
th
e
id
en
tific
atio
n
o
f
ab
n
o
r
m
al
ce
l
ls
.
T
h
e
tr
an
s
d
u
ce
r
tr
an
s
f
o
r
m
s
elec
tr
ical
s
ig
n
als
in
to
u
ltra
s
o
n
ic
wav
es
f
o
r
b
r
ea
s
t
im
ag
in
g
[
1
2
]
.
T
h
e
r
ef
lecte
d
s
o
u
n
d
wav
es c
o
u
ld
b
e
p
r
o
ce
s
s
ed
b
y
a
co
m
p
u
ter
to
p
r
o
d
u
ce
an
i
m
ag
e
b
ased
o
n
th
e
v
ar
y
in
g
u
ltra
s
o
n
ic
wav
e
am
p
litu
d
es
an
d
ec
h
o
es
tim
es
[
1
3
]
.
Ultr
a
s
o
n
o
g
r
a
p
h
y
h
as
th
e
b
en
ef
it
o
f
r
ea
l
-
tim
e
ex
am
in
atio
n
an
d
n
o
io
n
izin
g
r
ad
iatio
n
[
1
4
]
.
Ultr
aso
u
n
d
ca
n
b
e
u
tili
ze
d
in
m
e
d
icin
e
f
o
r
ec
h
o
-
g
u
id
ed
b
io
p
s
y
in
v
esti
g
atio
n
s
.
No
wa
d
ay
s
,
th
e
m
o
s
t
p
o
p
u
lar
tes
tin
g
m
eth
o
d
s
ar
e
m
a
m
m
o
g
r
ap
h
y
an
d
b
r
ea
s
t
u
ltra
-
s
o
n
o
g
r
ap
h
y
[
1
5
]
.
Als
o
,
it tak
es a
lo
t o
f
tim
e
an
d
h
ig
h
lev
el
o
f
ex
p
e
r
tis
e
to
d
if
f
er
en
tiat
e
b
etwe
en
d
if
f
er
e
n
t
s
u
b
ty
p
es o
f
b
r
ea
s
t c
an
ce
r
u
s
in
g
d
ig
ital p
ictu
r
es c
r
ea
te
d
f
r
o
m
b
io
p
s
y
s
am
p
les th
at
h
a
v
e
b
ee
n
co
llected
.
T
h
e
s
tain
in
g
p
r
o
ce
d
u
r
e,
lab
p
r
o
ce
d
u
r
es,
a
n
d
s
ca
n
n
er
b
r
ig
h
tn
ess
all
p
r
o
d
u
ce
s
ig
n
if
i
ca
n
t
co
lo
r
d
if
f
er
en
ce
s
wh
e
n
cr
ea
tin
g
h
i
s
to
p
ath
o
lo
g
y
im
ag
es,
m
ak
i
n
g
it
d
if
f
icu
lt
to
ef
f
ec
tiv
ely
tr
ain
a
m
u
lti
-
class
co
n
v
o
l
u
tio
n
al
n
e
u
r
al
n
etwo
r
k
(
C
NN)
m
o
d
el,
p
ar
ticu
lar
ly
i
n
lig
h
t
o
f
b
o
r
d
er
lin
e
in
s
tan
ce
s
[
1
6
]
.
Sin
ce
d
ig
ital
im
ag
es m
ay
well
h
av
e
clu
tter
an
d
in
s
u
f
f
icien
t b
r
ig
h
tn
ess
,
im
ag
e
au
g
m
en
tatio
n
a
n
d
p
r
e
p
r
o
c
ess
in
g
wer
e
cr
u
cial
p
h
ases
[
1
7
]
.
T
h
e
ef
f
ec
tiv
en
es
s
o
f
ex
tr
ac
tin
g
f
ea
tu
r
es
an
d
al
s
o
th
e
o
u
tco
m
es
o
f
im
ag
e
r
ec
o
g
n
itio
n
c
o
u
ld
b
o
th
b
e
d
r
am
atica
lly
im
p
r
o
v
e
d
b
y
p
ictu
r
e
p
r
ep
r
o
ce
s
s
in
g
.
T
h
e
m
ath
em
atica
l
s
tan
d
ar
d
izati
o
n
o
f
in
f
o
r
m
atio
n
g
ath
er
in
g
,
a
f
r
eq
u
en
t
s
tep
in
m
an
y
f
ea
tu
r
e
d
escr
ip
to
r
a
p
p
r
o
ac
h
es,
was
ak
in
to
p
r
etr
ea
t
m
en
t
[
1
8
]
.
T
o
s
o
lv
e
is
s
u
es
ca
u
s
ed
b
y
c
o
lo
r
d
is
cr
e
p
an
cies,
p
r
e
p
r
o
ce
s
s
in
g
m
u
s
t
f
o
llo
w
g
r
a
y
s
ca
le
p
r
in
cip
les.
I
m
ag
e
en
h
an
ce
m
e
n
t
aim
s
to
m
ak
e
p
h
o
t
o
g
r
a
p
h
s
ea
s
ier
f
o
r
v
iewe
r
s
to
u
n
d
er
s
tan
d
o
r
in
ter
p
r
et
o
r
to
g
iv
e
"b
etter
"
in
p
u
t
to
o
t
h
er
au
to
m
ated
[
1
9
]
.
T
h
e
r
e
h
av
e
b
ee
n
two
ca
teg
o
r
ies
o
f
im
ag
e
-
en
h
an
cin
g
tech
n
iq
u
es:
f
r
eq
u
en
cy
d
o
m
ain
an
d
s
p
atial
d
o
m
ain
.
T
h
e
p
r
ev
io
u
s
wo
r
k
s
with
in
d
iv
id
u
al
p
ix
els
d
ir
ec
tly
,
wh
er
ea
s
th
e
latter
u
s
es
th
e
im
ag
e
'
s
Fo
u
r
ier
tr
an
s
f
o
r
m
[
2
0
]
,
[
2
1
]
.
T
y
p
ically
s
p
ea
k
in
g
,
a
n
ac
tiv
e
co
n
to
u
r
s
y
s
tem
o
r
a
s
n
ak
e
m
o
d
el
b
ased
o
n
cu
r
v
e
ass
es
s
m
en
t
ap
p
r
o
ac
h
es
wo
u
ld
b
e
em
p
lo
y
ed
.
Mo
r
eo
v
er
,
a
s
ig
n
if
ican
t
d
r
awb
ac
k
o
f
ac
tiv
e
co
n
to
u
r
s
was
th
eir
f
ailu
r
e
to
d
ea
l w
ith
s
h
a
d
o
win
g
o
r
d
eter
m
in
e
th
e
b
o
u
n
d
ar
ies o
f
item
s
th
at
o
v
er
lap
t
h
em
[
2
2
]
–
[
2
5
]
.
Size
an
d
s
h
ap
e
wer
e
co
n
ce
p
ts
in
clu
d
ed
as
p
a
r
t
o
f
t
h
e
m
o
r
p
h
m
etr
ic
ass
ess
m
en
t.
An
aly
s
is
o
f
an
o
r
g
an
is
m
'
s
f
o
s
s
il
ev
id
en
ce
s
ee
m
s
to
b
e
a
co
m
m
o
n
p
r
o
ce
d
u
r
e
[
2
6
]
.
T
h
e
ef
f
ec
ts
o
f
m
u
tatio
n
s
o
n
d
esig
n
,
d
ev
elo
p
m
e
n
t
,
an
d
m
o
d
if
icatio
n
s
th
at
tak
e
p
lace
,
th
e
s
h
ap
e'
s
co
r
r
elatio
n
with
en
v
ir
o
n
m
en
t
al
p
ar
am
eter
s
,
an
d
m
eth
o
d
s
f
o
r
ass
ess
in
g
q
u
an
tit
ativ
e
f
ea
tu
r
es
o
f
a
ce
r
tain
s
h
ap
e
[
2
7
]
.
Fo
r
th
e
ce
ll
an
d
al
s
o
th
e
n
u
cleu
s
,
th
e
f
o
llo
win
g
m
o
r
p
h
m
etr
ic
f
ea
tu
r
es
—
in
cr
ea
s
ed
m
ea
n
v
alu
es
—
wer
e
s
tati
s
tically
s
ig
n
if
ican
t:
ar
ea
,
co
n
v
ex
ar
ea
,
an
d
o
u
tlin
e
[
2
8
]
.
I
d
e
n
tific
atio
n
o
f
item
s
o
r
r
eg
io
n
s
o
f
in
ter
e
s
t
in
an
im
ag
e
u
s
in
g
tex
t
u
r
al
clu
es.
B
y
ass
es
s
in
g
f
ea
tu
r
e
p
o
in
ts
at
ev
er
y
p
lace
i
n
th
e
im
ag
e
an
d
d
er
iv
in
g
a
c
o
llectio
n
o
f
s
tatis
tics
f
r
o
m
th
e
d
is
tr
ib
u
tio
n
s
o
f
th
e
f
ea
tu
r
e
p
o
in
ts
,
s
tatis
tical
tech
n
iq
u
es
m
ay
b
e
u
s
ed
in
tex
tu
r
e
an
aly
s
is
to
in
v
esti
g
ate
th
e
g
eo
g
r
ap
h
ical
ex
ten
t
o
f
g
r
ay
lev
els
[
2
9
]
.
Dis
cr
ete
wav
elet
tr
an
s
f
o
r
m
s
,
wh
ich
ty
p
ically
class
if
y
im
ag
es
o
f
m
alig
n
an
t
ce
lls
,
ar
e
b
ased
o
n
tex
tu
r
al
cu
es.
T
h
e
lo
ca
l
b
in
ar
y
p
atter
n
co
m
p
o
n
en
t,
a
p
o
ten
t
tech
n
iq
u
e
u
tili
ze
d
i
n
co
m
p
u
ter
v
is
io
n
an
d
ap
p
licatio
n
s
in
v
o
lv
in
g
p
atter
n
r
ec
o
g
n
itio
n
,
h
as
b
ee
n
p
r
esen
te
d
h
er
e
an
d
em
p
lo
y
ed
f
o
r
s
u
r
f
a
ce
r
ec
o
g
n
itio
n
an
d
ca
teg
o
r
izatio
n
[
3
0
]
.
C
o
-
o
cc
u
r
r
en
ce
at
th
e
g
r
ay
lev
el
,
th
e
c
h
ar
ac
ter
is
tics
o
f
th
e
m
atr
ix
w
av
elet
an
d
th
e
law's
tex
tu
r
e
wer
e
r
etr
ie
v
ed
.
Her
e
,
th
e
s
ig
n
if
ican
ce
of
th
e
wo
r
k
is
m
en
tio
n
ed
.
‒
B
y
ad
d
r
ess
in
g
t
h
e
in
ter
-
class
r
esem
b
lan
ce
o
f
tin
y
item
s
wh
ile
m
ain
tain
in
g
u
n
iq
u
e
f
o
r
m
s
,
th
e
im
p
r
o
v
e
d
f
aster
r
ec
u
r
r
en
t
c
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
(
I
MFR
C
NN)
wo
u
ld
b
e
a
u
s
ef
u
l
co
n
ce
p
t
f
o
r
e
n
h
an
ce
d
f
ea
tu
r
e
ex
tr
ac
tio
n
.
‒
A
cr
itical
ch
allen
g
e
wo
u
ld
b
e
to
co
u
n
t t
h
e
d
ilated
lay
e
r
s
an
d
u
tili
ze
th
e
d
ilatin
g
r
ate
in
ea
c
h
o
n
e.
‒
3
d
ilated
lay
er
s
with
a
d
ilatati
o
n
r
ate
o
f
2
wer
e
e
m
p
lo
y
e
d
i
n
th
e
p
r
o
p
o
s
ed
m
o
d
el
to
h
an
d
le
th
is
cr
u
cial
wo
r
k
with
o
u
t c
h
an
g
i
n
g
th
e
m
o
r
p
h
o
lo
g
y
o
f
m
ito
tic
n
u
clei.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
I
mp
r
o
vised
ma
s
k
fa
s
ter r
ec
u
r
r
en
t c
o
n
vo
l
u
tio
n
a
l
n
eu
r
a
l n
etw
o
r
k
fo
r
b
r
ea
s
t
…
(
P
a
tta
n
M.
D
.
A
li K
h
a
n
)
2001
‒
B
y
m
ak
in
g
tin
y
o
b
jects
lar
g
er
,
th
e
p
r
o
p
o
s
ed
m
o
d
el
co
r
r
ec
ts
th
e
m
is
m
atch
b
etwe
en
t
h
e
im
ag
e
an
d
o
b
ject
-
co
v
e
r
ed
ar
ea
s
in
d
ee
p
lay
er
s
.
As
a
r
esu
lt,
th
er
e
ca
n
b
e
less
s
ize
d
is
p
ar
ity
b
etwe
en
th
e
f
o
r
e
g
r
o
u
n
d
an
d
b
ac
k
d
r
o
p
.
‒
D
u
r
i
n
g
m
i
t
o
t
i
c
i
d
e
n
t
i
f
i
c
a
t
i
o
n
,
i
t
a
i
d
s
i
n
r
e
d
u
c
i
n
g
c
a
t
e
g
o
r
i
z
a
t
i
o
n
,
e
n
d
-
e
f
f
e
c
t
o
r
s
,
a
n
d
l
o
c
a
l
i
z
a
t
i
o
n
d
e
g
r
a
d
a
t
i
o
n
.
T
h
e
o
th
er
s
ec
tio
n
s
o
f
th
e
r
esear
ch
wer
e
s
tr
u
ctu
r
ed
s
o
th
at
s
ec
tio
n
2
ex
p
lain
s
th
e
d
atab
ase,
ch
ar
ac
ter
is
tics
,
an
d
ca
teg
o
r
iz
atio
n
o
f
th
e
I
MFR
C
NN.
W
h
ile
s
ec
tio
n
3
g
iv
es
a
r
e
v
iew
o
f
th
e
liter
atu
r
e
.
Ou
tco
m
es
ar
e
co
v
er
ed
in
s
ec
tio
n
4
.
L
astl
y
,
co
n
clu
s
io
n
an
d
r
ec
o
m
m
en
d
atio
n
s
f
o
r
f
u
r
t
h
er
r
e
s
ea
r
ch
ar
e
co
v
er
ed
in
s
ec
tio
n
5
.
2.
M
E
T
H
O
D
T
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
m
o
d
if
ies
th
e
I
MFR
C
N
N
ar
ch
itectu
r
e,
in
co
r
p
o
r
atin
g
R
es
N
et1
0
1
p
r
e
-
tr
ain
ed
weig
h
ts
f
r
o
m
th
e
co
m
m
o
n
o
b
jects
in
co
n
tex
t
(
C
O
C
O)
2
0
1
7
d
ataset.
I
n
ad
d
r
ess
in
g
th
e
ch
allen
g
es
p
o
s
ed
b
y
h
is
to
p
ath
o
lo
g
y
im
ag
es,
p
ar
tic
u
lar
ly
with
m
ito
tic
n
u
clei'
s
p
o
o
r
v
is
u
al
r
atio
,
an
d
d
ee
p
lear
n
in
g
m
o
d
els
u
n
d
er
g
o
cr
u
cial
p
r
ep
r
o
ce
s
s
in
g
s
tag
es.
Utilizin
g
R
esNet1
0
1
f
ea
tu
r
es
s
u
ch
as
p
o
o
lin
g
,
c
o
n
v
o
lu
tio
n
al
lay
er
s
with
s
tr
id
e
2
,
an
d
s
tr
ateg
ic
m
ax
-
p
o
o
lin
g
,
th
e
m
eth
o
d
aim
s
to
r
ed
u
ce
im
ag
e
s
ize,
p
r
eser
v
e
r
i
ch
in
f
o
r
m
atio
n
,
an
d
p
r
ev
en
t
t
h
e
d
is
ap
p
ea
r
an
ce
o
f
s
m
all
item
s
.
T
h
is
en
h
an
ce
s
th
e
f
ea
tu
r
e
e
x
tr
ac
to
r
'
s
ab
ilit
y
to
d
is
ce
r
n
m
ea
n
in
g
f
u
l
ch
ar
ac
ter
is
tics
with
in
h
is
to
p
ath
o
lo
g
ical
im
ag
es.
2
.
1
.
Da
t
a
s
et
des
cr
iptio
n
R
es
e
a
r
c
h
e
r
s
h
a
v
e
u
s
e
d
a
b
r
e
a
s
t
c
a
n
c
e
r
d
a
t
as
e
t
i
n
t
h
i
s
s
t
u
d
y
is
a
h
i
s
t
o
p
a
t
h
o
l
o
g
i
c
a
l
d
a
t
a
b
as
e
.
F
i
g
u
r
e
1
i
l
l
u
s
t
r
a
t
es
t
h
e
9
,
1
0
9
m
i
c
r
o
s
c
o
p
i
c
i
m
a
g
es
o
f
b
r
e
a
s
t
t
u
m
o
r
t
is
s
u
e
o
b
t
a
i
n
e
d
f
r
o
m
8
2
p
a
t
ie
n
t
s
u
t
i
li
z
i
n
g
v
a
r
i
o
u
s
m
a
g
n
i
f
i
c
a
t
i
o
n
v
a
r
i
a
b
l
es
t
h
a
t
co
m
p
r
i
s
e
t
h
e
b
r
e
a
s
t
c
a
n
ce
r
h
i
s
t
o
p
a
t
h
o
l
o
g
i
c
a
l
i
m
a
g
i
n
g
c
l
ass
i
f
i
c
a
t
i
o
n
.
I
t
n
o
w
h
a
s
2
,
4
8
0
b
e
n
i
g
n
s
a
m
p
l
e
s
a
n
d
5
,
4
2
9
c
a
n
c
e
r
o
u
s
o
n
e
s
.
I
n
a
d
d
i
t
i
o
n
t
o
t
h
e
P
a
n
d
D
L
a
b
o
r
a
t
o
r
y
f
o
r
P
a
t
h
o
l
o
g
i
c
a
l
A
n
a
t
o
m
y
a
n
d
C
y
t
o
p
a
t
h
o
l
o
g
y
i
n
P
a
r
a
n
a
,
t
h
i
s
d
a
t
a
b
as
e
w
as
d
e
v
e
l
o
p
e
d
.
Si
n
c
e
i
t
e
n
a
b
l
e
s
f
u
t
u
r
e
c
o
m
p
a
r
is
o
n
a
n
d
a
s
s
e
s
s
m
e
n
t
,
h
u
m
a
n
s
t
h
i
n
k
t
h
at
r
es
e
a
r
c
h
e
r
s
w
o
u
l
d
c
o
n
s
i
d
e
r
t
h
is
d
at
a
s
e
t
to
b
e
b
e
n
e
f
i
c
i
al
[
3
1
]
.
C
a
n
c
e
r
w
a
s
r
e
f
e
r
r
e
d
t
o
as
a
m
a
l
i
g
n
a
n
t
t
u
m
o
r
s
i
n
c
e
t
h
e
le
s
i
o
n
c
a
n
s
p
r
e
a
d
t
o
d
i
s
t
a
n
t
a
r
e
as
,
i
n
v
a
d
e
o
t
h
e
r
t
i
s
s
u
e
s
,
a
n
d
l
e
a
d
t
o
d
e
a
t
h
.
R
esear
ch
er
s
in
v
esti
g
ate
h
id
d
en
tr
en
d
s
in
th
e
in
f
o
r
m
atio
n
u
s
in
g
ad
v
an
c
ed
an
aly
tics
.
T
h
r
ee
f
iles
wer
e
p
r
o
d
u
ce
d
.
T
h
e
r
ea
l
b
o
u
n
d
i
n
g
b
o
x
es
f
o
r
ev
er
y
class
,
in
ad
d
itio
n
to
th
e
class
es.
Fo
llo
win
g
th
e
tr
ain
in
g
p
h
ase,
r
esear
ch
er
s
m
ak
e
p
r
ed
ictio
n
s
u
s
in
g
th
ese
im
ag
es.
T
h
e
n
am
e
o
f
th
e
im
ag
e,
t
h
e
class
,
an
d
th
e
b
o
u
n
d
i
n
g
b
o
x
co
o
r
d
in
ates
h
as
all
b
ee
n
s
to
r
e
d
in
th
is
f
o
ld
er
.
Mo
r
e
t
h
an
o
n
e
o
b
ject
m
ay
b
e
p
r
esen
t
in
a
s
in
g
le
p
h
o
to
g
r
ap
h
.
Hen
ce
,
th
er
e
co
u
ld
b
e
m
o
r
e
r
o
ws in
a
s
in
g
le
im
ag
e.
Fig
u
r
e
1
.
His
to
p
ath
o
l
o
g
ical
im
ag
e
d
ataset
f
o
r
b
r
ea
s
t c
an
ce
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
2
,
Ap
r
il 2
0
2
6
:
1
9
9
9
-
2
0
0
8
2002
2
.
2
.
M
et
ho
do
lo
g
y
T
h
e
I
MFR
C
NN
ar
ch
itectu
r
e
h
as
b
ee
n
twea
k
ed
to
cr
ea
te
t
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
.
T
h
e
I
MFR
C
N
N
with
R
esNe
t1
0
1
p
r
e
-
tr
ain
e
d
weig
h
ts
f
r
o
m
th
e
C
OC
O
2
0
1
7
d
ataset
wer
e
u
tili
ze
d
in
th
is
s
y
s
tem
.
On
e
o
f
th
e
cr
u
cial
s
tag
es
o
f
m
o
d
els
b
ased
o
n
d
ee
p
lear
n
in
g
was
p
r
e
-
p
r
o
ce
s
s
in
g
.
T
o
im
p
r
o
v
e
th
e
co
n
tr
asts
,
a
v
ar
iety
o
f
p
r
ep
r
o
ce
s
s
in
g
m
eth
o
d
s
h
av
e
b
ee
n
u
s
ed
,
in
clu
d
in
g
r
a
n
d
o
m
ad
ju
s
tm
en
ts
o
f
co
m
p
ar
is
o
n
,
b
r
ig
h
tn
ess
,
an
d
h
u
e.
Mito
tic
n
u
clei
h
av
e
a
p
o
o
r
v
i
s
u
al
r
atio
in
h
is
to
p
ath
o
lo
g
y
i
m
ag
es.
T
h
e
im
ag
e
-
to
-
o
b
ject
r
atio
was
s
tr
o
n
g
in
th
ese
k
in
d
s
o
f
p
h
o
to
s
s
in
ce
th
e
n
u
clei
wer
e
tin
y
[
2
0
]
.
Po
o
lin
g
a
n
d
a
f
ew
co
n
v
o
lu
tio
n
al
lay
er
s
with
s
tr
id
e
2
wer
e
f
ea
tu
r
es
o
f
th
e
R
esNet1
0
1
th
at
s
ee
m
to
b
e
u
s
ef
u
l
f
o
r
r
ed
u
cin
g
th
e
len
g
th
o
f
b
o
th
t
h
e
in
p
u
t
im
ag
e
an
d
th
e
o
b
jects
in
clu
d
e
d
with
in
it.
As
we
p
r
ev
io
u
s
ly
ex
p
lain
ed
,
o
n
e
g
o
al
o
f
em
p
l
o
y
in
g
a
p
o
o
lin
g
lay
er
wo
u
ld
b
e
to
d
ec
r
ea
s
e
im
ag
e
s
ize
wh
il
e
m
ain
tain
in
g
r
ich
in
f
o
r
m
atio
n
.
Af
ter
m
ax
-
p
o
o
lin
g
,
th
e
s
tr
etch
ed
lay
e
r
s
en
lar
g
e
o
b
jects
wh
ile
m
ain
tai
n
in
g
t
h
e
d
im
en
s
io
n
s
o
f
th
e
s
u
p
p
lied
im
ag
es
b
y
u
s
in
g
th
e
r
ich
i
n
f
o
r
m
atio
n
o
b
tain
ed
f
r
o
m
th
e
o
u
tp
u
t
o
f
m
ax
-
p
o
o
li
n
g
.
T
h
u
s
,
a
f
ter
th
e
m
ax
-
p
o
o
lin
g
la
y
er
,
a
s
tr
etch
e
d
lay
er
was
u
tili
ze
d
to
p
r
ev
en
t
th
e
d
is
ap
p
ea
r
a
n
ce
o
f
item
s
.
T
h
is
m
eth
o
d
lo
wer
s
th
e
lik
elih
o
o
d
o
f
s
m
all
item
s
d
is
ap
p
ea
r
i
n
g
,
as
illu
s
tr
ated
i
n
Fig
u
r
e
2
,
wh
ic
h
aid
s
th
e
f
ea
tu
r
e
ex
t
r
ac
to
r
in
lear
n
in
g
th
e
g
o
o
d
c
h
ar
ac
ter
is
tics
o
f
in
-
d
ep
th
b
lo
ck
s
.
Alth
o
u
g
h
b
ig
-
s
ca
le
in
cr
em
e
n
ts
at
o
n
ce
af
f
ec
t
th
e
s
h
ap
e
o
f
n
u
clei,
a
h
ig
h
d
ilated
r
ate
w
as a
d
v
an
tag
eo
u
s
f
o
r
i
n
cr
ea
s
in
g
o
b
ject
s
ize
[
2
4
]
.
T
h
e
m
o
r
p
h
o
l
o
g
y
o
f
m
it
o
ti
c
an
d
n
o
n
-
m
it
o
t
ic
n
u
cle
i
w
it
h
s
im
i
lar
s
h
a
p
es
a
n
d
s
iz
es
w
as
alt
er
e
d
b
y
m
o
r
e
d
il
ate
d
la
y
er
s
.
Usi
n
g
o
n
ly
3
e
x
p
a
n
d
e
d
l
e
v
els
o
u
t
o
f
R
es
Net
1
0
1
'
s
t
o
t
al
o
f
1
0
1
la
y
e
r
s
was
in
t
en
d
ed
t
o
p
r
e
v
e
n
t
s
m
al
l
i
te
m
s
f
r
o
m
d
is
s
i
p
at
in
g
i
n
d
ee
p
la
y
e
r
s
w
h
il
e
m
a
in
ta
in
i
n
g
th
e
s
h
ap
e
o
f
n
u
cle
i.
I
MFR
C
NN
h
as
b
e
en
g
e
n
e
r
a
te
d
in
c
o
n
v
3
b
l
o
c
k
1
,
c
o
n
v
4
b
l
o
c
k
1
,
a
n
d
co
n
v
3
b
l
o
c
k
2
,
as
s
h
o
w
n
i
n
F
ig
u
r
e
3
d
il
ate
d
c
o
n
v
o
l
u
t
io
n
a
l
n
o
d
e
.
Fig
u
r
e
2
.
Fil
ter
with
o
u
t
d
ilatio
n
r
ate
Fig
u
r
e
3
.
A
d
ilated
c
o
n
v
o
lu
tio
n
al
n
o
d
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
I
mp
r
o
vised
ma
s
k
fa
s
ter r
ec
u
r
r
en
t c
o
n
vo
l
u
tio
n
a
l
n
eu
r
a
l n
etw
o
r
k
fo
r
b
r
ea
s
t
…
(
P
a
tta
n
M.
D
.
A
li K
h
a
n
)
2003
R
eg
ar
d
less
o
f
th
e
s
ize
o
f
th
e
i
n
p
u
t,
th
e
r
eg
io
n
o
f
in
te
r
est
(
R
OI
)
p
o
o
lin
g
lay
er
s
ep
ar
ates
th
e
p
r
o
p
o
s
ed
r
eg
io
n
s
'
ch
ar
ac
ter
is
tics
in
to
s
m
aller
ar
ea
s
an
d
o
u
tp
u
ts
f
ix
ed
-
s
ize
ch
ar
ac
ter
is
tics
.
T
h
e
R
OI
p
o
o
lin
g
lay
er
'
s
r
esu
lt
in
th
e
p
r
o
p
o
s
ed
m
eth
o
d
was
7
×
7
.
T
h
e
n
e
x
t
f
u
lly
c
o
n
n
ec
ted
lay
er
s
,
So
f
tMa
x
,
i
n
clu
d
in
g
b
o
u
n
d
in
g
b
o
x
ex
tr
ap
o
latio
n
b
r
an
c
h
es
,
wo
u
ld
th
en
b
e
g
iv
en
th
e
r
esu
lt c
h
ar
a
cter
is
tics
o
f
th
e
R
OI
p
o
o
lin
g
l
ay
er
s
in
I
MFR
C
NN
as illu
s
tr
ated
in
Fig
u
r
e
4
.
Fig
u
r
e
4
.
R
OI
b
lo
c
k
f
o
r
lo
ca
li
za
tio
n
an
d
ca
te
g
o
r
izatio
n
o
n
I
MFR
C
NN
2
.
3
.
I
m
pro
v
ed
f
a
s
t
er
re
curr
ent
co
nv
o
lutio
na
l neura
l net
wo
rk
T
o
id
en
tify
d
is
tin
ct
item
s
in
an
im
ag
e
an
d
cr
ea
te
a
b
o
u
n
d
in
g
b
o
x
a
r
o
u
n
d
a
p
a
r
ticu
l
ar
o
b
ject,
tech
n
iq
u
es
ca
lled
o
b
ject
r
ec
o
g
n
itio
n
an
d
s
eg
m
en
tatio
n
h
av
e
b
ee
n
u
s
ed
.
On
e
tech
n
iq
u
e
f
o
r
o
b
ject
r
ec
o
g
n
itio
n
an
d
s
eg
m
en
tatio
n
was
a
c
o
n
v
o
lu
ti
o
n
al
n
eu
r
al
n
etwo
r
k
-
b
ased
m
u
lti
-
s
ca
le
lo
ca
l
a
n
d
g
lo
b
al
f
ea
tu
r
e
r
ep
r
esen
tatio
n
(
MFR
-
C
NN)
[
3
2
]
.
MFR
-
C
NN
s
er
v
es
a
s
th
e
f
o
u
n
d
atio
n
f
o
r
m
ask
R
-
C
NN,
wh
ich
ex
p
an
d
s
its
u
s
e
in
p
ictu
r
e
ca
teg
o
r
izatio
n
.
Fig
u
r
e
5
s
h
o
ws h
o
w
its
n
etwo
r
k
ar
ch
itectu
r
e
was stru
ctu
r
ed
.
T
h
e
p
r
o
ce
d
u
r
e
r
esu
lts
in
th
e
lo
s
s
o
f
d
ata
s
o
u
r
ce
s
,
ca
u
s
in
g
t
h
e
R
OI
an
d
ex
tr
ac
ted
c
h
ar
ac
t
er
is
tics
o
f
th
e
o
r
ig
in
al
im
ag
e
t
o
b
e
d
is
p
lace
d
.
Fo
llo
win
g
t
h
e
co
m
p
letio
n
o
f
th
e
n
etwo
r
k
d
esig
n
,
t
h
e
MFR
-
C
NN
wa
s
lear
n
ed
u
s
in
g
th
e
u
ltra
s
o
u
n
d
im
ag
es,
th
e
m
atch
in
g
b
io
p
s
y
in
f
o
r
m
atio
n
,
an
d
th
e
r
ad
io
lo
g
is
t
-
d
r
awn
tu
m
o
r
co
n
to
u
r
s
as
th
e
g
r
o
u
n
d
tr
u
th
[
2
2
]
.
T
h
e
in
f
o
r
m
atio
n
f
r
o
m
th
e
tr
ain
in
g
s
et
was
u
tili
ze
d
to
b
u
ild
a
m
o
d
el,
wh
ic
h
was
s
u
b
s
eq
u
en
tly
ev
alu
ated
ag
ain
s
t
th
e
v
alid
ity
co
llectio
n
to
m
ak
e
s
u
r
e
it
was
ac
cu
r
ate
an
d
s
tab
le.
T
h
e
tr
ain
in
g
tech
n
iq
u
e
d
iv
i
d
ed
r
an
d
o
m
ly
t
h
e
g
ath
er
e
d
in
s
tan
ce
s
in
to
a
tr
ain
in
g
d
ataset
an
d
test
in
g
s
et.
I
MFR
C
NN
'
s
m
o
d
el
f
u
n
ctio
n
l
o
s
s
was d
escr
ib
ed
as in
(
1
)
.
=
+
+
(
1
)
Fig
u
r
e
5
.
I
MFR
C
NN
'
s
s
tr
u
ctu
r
e
Me
an
av
er
a
g
e
p
r
ec
is
io
n
(
m
A
P
)
is
to
m
ea
s
u
r
e
th
e
lesi
o
n
d
etec
tio
n
/s
eg
m
en
tatio
n
ef
f
icien
cy
o
n
th
e
test
in
g
s
et
an
d
tr
ain
ed
MFR
-
C
NN
m
o
d
el
as
g
iv
en
in
(
2
)
.
W
h
er
e
A
is
s
eg
m
en
tatio
n
o
f
m
o
d
el
r
esu
lts
;
B
is
th
e
ap
p
r
o
p
r
iate
tu
m
o
r
o
u
tlin
e
ex
p
er
t
r
ad
io
lo
g
is
t
h
as
d
r
awn
a
s
a
r
ea
l
clin
ical
lesi
o
n
.
s
ee
m
s
to
b
e
th
e
o
v
er
lap
p
i
n
g
ar
ea
b
etwe
en
th
e
m
o
d
el
-
id
en
tifie
d
lesi
o
n
an
d
also
th
e
g
en
u
in
e
m
ed
ical
lesi
o
n
r
eg
io
n
s
h
av
e
r
ea
l
clin
ical
lesi
o
n
s
s
ize.
s
tan
d
s
f
o
r
th
e
n
u
m
b
er
o
f
im
a
g
es.
R
eliab
ilit
y
was
u
tili
ze
d
to
co
n
f
ir
m
th
e
p
r
o
p
o
s
ed
m
eth
o
d
'
s
o
v
er
all
lesi
o
n
ca
teg
o
r
izatio
n
ef
f
icien
cy
.
I
n
(
3
)
an
al
y
ze
th
e
ev
alu
atio
n
m
ea
s
u
r
es
s
u
ch
as
tr
u
e
p
o
s
itiv
e
(
T
P),
tr
u
e
n
eg
ativ
e
(
T
N)
,
f
alse p
o
s
itiv
e
(
FP
)
,
an
d
f
alse p
o
s
itiv
e
(
FN)
.
=
∩
∪
=
1
∑
(
)
=
1
(
2
)
=
(
+
)
/
(
+
+
+
)
(
3
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
2
,
Ap
r
il 2
0
2
6
:
1
9
9
9
-
2
0
0
8
2004
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
o
u
tco
m
es
o
f
th
e
r
ad
io
l
o
g
is
ts
'
tu
m
o
r
co
n
to
u
r
an
aly
s
is
h
av
e
b
ee
n
d
is
p
lay
ed
in
Fig
u
r
e
6
.
B
r
ea
s
t
u
ltra
s
o
u
n
d
im
a
g
es
o
f
two
s
ep
ar
ate
m
alig
n
an
t
tu
m
o
r
s
ca
n
b
e
s
ee
n
in
Fig
u
r
e
6
(
a
)
,
wh
ile
b
e
n
ig
n
tu
m
o
r
s
ca
n
b
e
s
ee
n
in
Fig
u
r
e
6
(
b
)
.
T
h
e
ac
tu
a
l
r
ef
er
en
ce
im
ag
e
o
cc
u
p
ies
th
e
lef
t
s
id
e
o
f
t
h
e
im
ag
e,
an
d
also
th
e
r
ea
l
m
ask
in
g
cr
ea
ted
b
y
a
q
u
alif
ied
r
ad
io
l
o
g
is
t u
s
in
g
th
e
o
r
ig
in
al
im
a
g
e
h
a
s
b
ee
n
s
h
o
wn
o
n
th
e
r
ig
h
t sid
e
.
(
a)
(
b
)
Fig
u
r
e
6
.
Dete
ctio
n
s
o
f
s
am
p
l
es (
a)
m
alig
n
an
t tu
m
o
r
a
n
d
(
b
)
b
en
ig
n
tu
m
o
r
T
h
e
lear
n
in
g
alg
o
r
ith
m
c
o
n
ti
n
u
o
u
s
ly
r
ed
u
ce
d
u
n
til
th
e
co
n
clu
s
io
n
o
f
2
5
K
iter
atio
n
s
,
a
f
ter
2
.
3
K
iter
atio
n
s
.
B
y
in
clu
d
in
g
a
to
tal
o
f
1
0
0
b
o
u
n
d
in
g
b
o
x
es
f
o
r
ev
er
y
ca
teg
o
r
y
,
th
e
lar
g
est
am
o
u
n
t
o
f
an
ch
o
r
b
o
x
es
in
r
eg
io
n
al
p
r
o
p
o
s
al
n
etwo
r
k
(
R
PN
)
wa
s
s
et
to
3
0
0
,
wh
er
ea
s
in
th
e
o
u
tco
m
e,
it
m
u
s
t
b
e
s
et
to
2
0
0
.
A
v
ar
iety
o
f
b
lo
ck
s
in
th
e
p
r
o
p
o
s
ed
I
MFR
C
NN
em
p
lo
y
v
ar
y
in
g
n
u
m
b
er
s
o
f
d
ilated
la
y
er
s
u
s
i
n
g
v
ar
iab
le
r
ates
o
f
d
ilatio
n
.
Ap
p
ly
i
n
g
d
ilated
lay
e
r
s
u
s
in
g
d
ilatio
n
r
ates
th
at
v
ar
y
b
etwe
en
2
to
5
.
All
v
ar
iatio
n
s
'
o
u
tco
m
es
ca
n
b
e
s
ee
n
in
Fig
u
r
e
7
.
Fig
u
r
e
7
.
R
ate
o
f
lea
r
n
in
g
L
o
s
s
b
eh
av
io
r
in
g
en
er
al
i
n
d
i
ca
tes
th
at
th
e
m
o
d
el
p
r
o
g
r
ess
iv
ely
ac
q
u
ir
ed
b
etter
f
ea
tu
r
es
f
o
r
r
e
g
io
n
s
u
g
g
esti
o
n
s
.
Ultim
ately
,
th
e
p
r
o
p
o
s
ed
m
o
d
el'
s
o
v
er
all
lo
s
s
h
a
s
b
ee
n
esti
m
ated
an
d
s
ee
m
s
to
b
e
0
.
1
5
,
as
s
h
o
wn
in
Fig
u
r
e
8
.
T
h
e
to
tal
l
o
s
s
was
0
.
5
2
a
f
ter
th
e
f
ir
s
t
iter
at
io
n
an
d
clim
b
e
d
to
0
.
7
af
ter
7
K
iter
atio
n
s
.
T
h
e
u
n
p
r
e
d
ictab
ilit
y
o
f
ca
te
g
o
r
izat
io
n
,
R
PN
o
b
jectless
an
d
lo
ca
li
za
tio
n
lo
s
s
u
p
to
2
.
1
K
iter
atio
n
s
s
ee
m
s
to
b
e
th
e
s
o
u
r
ce
o
f
th
e
o
v
er
all
lo
s
s
'
s
r
is
in
g
b
e
h
av
io
r
.
Af
ter
2
.
1
K
iter
a
tio
n
s
,
all
lo
s
s
es
d
r
o
p
p
ed
g
r
a
d
u
ally
,
r
esu
ltin
g
in
a
r
ed
u
ctio
n
in
th
e
o
v
er
all
l
o
s
s
.
I
t
was
f
o
u
n
d
th
at
th
e
lo
s
s
es
o
f
th
e
I
MFR
C
NN
d
r
o
p
p
e
d
as
th
e
n
u
m
b
er
o
f
iter
atio
n
s
in
cr
ea
s
ed
,
in
d
icatin
g
th
at
th
e
p
r
o
p
o
s
ed
m
o
d
el
l
ea
r
n
ed
m
o
r
e
ef
f
ec
tiv
ely
as
r
ep
ea
ts
wer
e
ad
d
ed
.
T
y
p
ical
o
b
ject
r
ec
o
g
n
itio
n
a
lg
o
r
ith
m
s
h
av
e
b
ee
n
h
o
n
e
d
f
o
r
2
5
K
s
tep
s
in
th
e
s
am
e
s
u
r
r
o
u
n
d
in
g
s
.
All
ex
p
er
im
en
tal
m
o
d
els’
o
u
tco
m
es
wer
e
co
n
tr
asted
with
th
o
s
e
o
f
th
e
p
r
o
p
o
s
ed
I
MFR
C
NN.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
o
u
tp
er
f
o
r
m
ed
th
e
o
p
p
o
s
itio
n
r
eg
ar
d
in
g
av
er
a
g
e
ac
c
u
r
ac
y
.
Acc
o
r
d
i
n
g
to
T
a
b
le
1
,
th
e
p
r
o
p
o
s
ed
m
eth
o
d
p
er
f
o
r
m
ed
b
etter
f
o
r
th
e
av
er
a
g
e
p
r
ec
is
io
n
(
AP)
a
n
d
m
AP a
c
cu
r
ac
y
m
ea
s
u
r
es.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
p
er
f
o
r
m
s
b
est
f
o
r
m
ed
ian
m
e
m
o
r
y
wh
en
co
m
p
ar
ed
to
ex
is
tin
g
co
n
v
en
tio
n
a
l
m
eth
o
d
s
f
o
r
id
e
n
tify
in
g
o
b
jec
ts
.
T
h
e
p
r
o
p
o
s
ed
f
r
am
ewo
r
k
p
er
f
o
r
m
ed
th
e
b
est
at
AP
an
d
m
AP
as
s
h
o
wn
in
T
ab
le
2
.
W
h
en
m
o
d
el
allo
w
id
en
tify
in
g
o
n
e
o
b
ject
in
a
g
i
v
en
im
ag
e
its
m
ea
n
r
ec
o
llectio
n
a
s
AP.
0
.
0
0
E
+
0
0
2
.
0
0
E
-
0
4
4
.0
0
E
-
0
4
6
.0
0
E
-
0
4
8
.0
0
E
-
0
4
1
.
0
0
E
-
0
3
1
.
2
0
E
-
0
3
0
4K
8K
1
2
K
1
6
K
2
0
K
2
4
K
3
0
k
L
e
a
r
n
i
n
g
r
a
t
e
It
e
r
a
t
i
o
n
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
I
mp
r
o
vised
ma
s
k
fa
s
ter r
ec
u
r
r
en
t c
o
n
vo
l
u
tio
n
a
l
n
eu
r
a
l n
etw
o
r
k
fo
r
b
r
ea
s
t
…
(
P
a
tta
n
M.
D
.
A
li K
h
a
n
)
2005
Fig
u
r
e
8
.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el'
s
o
v
er
all
lo
s
s
o
f
av
er
a
g
e
ac
cu
r
ac
y
T
ab
le
1
.
C
o
m
p
a
r
is
o
n
o
f
th
e
p
r
o
p
o
s
ed
an
d
ex
is
tin
g
s
y
s
tem
ac
cu
r
ac
y
Te
c
h
n
i
q
u
e
s
B
a
c
k
b
o
n
e
AP
mA
P
16
R
e
sN
e
t
1
0
1
1
9
.
7
8
2
1
.
8
9
18
Ef
f
i
c
i
e
n
t
N
e
t
24
2
4
.
6
8
21
R
e
sN
e
t
1
0
1
1
6
.
6
2
1
8
.
0
0
P
r
o
p
o
se
d
m
o
d
e
l
M
o
d
i
f
i
e
d
R
e
sN
e
t
1
0
1
2
5
.
3
8
2
6
.
1
8
T
ab
le
2
.
C
o
m
p
a
r
is
o
n
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
els b
ased
o
n
m
em
o
r
y
Te
c
h
n
i
q
u
e
s
B
a
c
k
b
o
n
e
AP
mA
P
16
R
e
sN
e
t
1
0
1
1
4
.
5
4
7
.
5
18
Ef
f
i
c
i
e
n
t
N
e
t
1
5
.
6
3
7
.
8
21
R
e
sN
e
t
1
0
1
1
2
.
3
4
1
.
5
P
r
o
p
o
se
d
m
o
d
e
l
M
o
d
i
f
i
e
d
R
e
sN
e
t
1
0
1
1
5
.
6
5
2
.
5
Ou
tco
m
es
f
r
o
m
o
b
ject
d
etec
tio
n
m
o
d
els
wer
e
p
r
esen
ted
as
b
o
u
n
d
in
g
b
o
x
es
u
s
in
g
cla
s
s
lab
els
.
Fig
u
r
e
9
d
is
p
lay
s
th
e
to
p
5
m
o
d
els'
r
esu
lt
s
.
I
n
p
r
e
d
icted
r
esu
l
ts
,
f
r
am
es
r
ep
r
esen
t
e
x
p
ec
ted
n
o
n
-
m
ito
tic
n
u
clei,
wh
ile
g
r
ee
n
b
o
u
n
d
i
n
g
b
o
x
es
r
ep
r
esen
t
p
r
o
jecte
d
m
ito
tic
n
u
c
lei.
Du
e
to
th
e
i
n
tr
icac
y
o
f
t
h
e
im
ag
es,
th
e
o
b
ject
r
ec
o
g
n
itio
n
p
er
f
o
r
m
an
ce
o
f
t
h
e
m
o
d
el
in
h
is
to
p
ath
o
lo
g
ical
im
ag
es
r
em
ain
s
v
er
y
lo
w.
T
h
e
o
th
er
ca
u
s
es
o
f
p
o
o
r
ac
h
iev
em
en
t
wer
e
in
tr
a
-
class
h
eter
o
g
en
eity
an
d
in
ter
-
class
h
o
m
o
g
en
eity
.
T
h
e
ab
s
e
n
ce
o
f
tin
y
item
s
,
m
o
r
eo
v
e
r
,
m
ea
n
t
th
at
t
h
e
f
ea
t
u
r
e
ex
tr
ac
t
o
r
'
s
ef
f
icien
cy
s
tay
ed
p
o
o
r
.
T
o
p
r
ev
e
n
t
s
m
all
item
s
f
r
o
m
v
an
is
h
in
g
,
v
ar
io
u
s
n
u
m
b
er
s
o
f
d
ilated
le
v
els
at
d
if
f
er
en
t
h
ier
ar
c
h
ical
lev
els
with
v
ar
ied
d
ilatatio
n
s
p
ee
d
s
wer
e
u
tili
ze
d
.
T
h
e
b
est
o
u
tco
m
es
wer
e
f
in
ally
o
b
tain
ed
with
3
d
ilated
lay
e
r
s
with
a
d
ilatatio
n
f
r
eq
u
en
cy
o
f
2
.
T
h
r
ee
c
r
u
cial
p
ar
am
eter
s
th
at
aid
th
e
f
ea
tu
r
e
ex
tr
ac
to
r
s
ee
m
to
b
e
th
e
l
o
ca
tio
n
o
f
d
ilated
lay
er
s
an
d
th
eir
r
ate
o
f
d
ilatatio
n
.
I
mag
e
1
I
mag
e
2
G
r
o
u
n
d
Tr
u
t
h
C
e
n
t
e
r
N
e
t
I
M
F
R
C
N
N
Fig
u
r
e
9
.
C
o
m
p
a
r
is
o
n
o
f
v
ar
io
u
s
m
o
d
els u
s
in
g
ac
tu
al
d
ata
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
2
,
Ap
r
il 2
0
2
6
:
1
9
9
9
-
2
0
0
8
2006
T
h
e
m
o
s
t
ad
v
an
ce
d
m
o
d
el
f
o
r
id
en
tify
in
g
tin
y
o
b
jects,
in
clu
d
in
g
s
u
ch
m
ito
tic
an
d
n
o
n
-
m
it
o
tic
n
u
clei
s
ee
m
s
to
b
e
th
e
I
MFR
C
NN.
Utilizin
g
d
o
m
ain
ad
ap
tab
ilit
y
an
d
m
u
lti
-
m
o
d
el
f
ea
tu
r
es
f
u
s
io
n
wo
u
l
d
h
elp
th
e
I
MFR
C
NN
p
er
f
o
r
m
b
etter
.
W
h
en
a
m
o
d
el
lear
n
s
f
r
o
m
im
ag
e
f
ea
tu
r
es
to
m
icr
o
s
co
p
ic
im
a
g
es,
its
p
er
f
o
r
m
an
ce
s
u
f
f
er
s
.
T
h
e
s
tr
u
ctu
r
al
d
if
f
e
r
e
n
ce
s
b
etwe
en
n
o
r
m
al
an
d
m
ic
r
o
s
co
p
ic
im
ag
es
s
ee
m
to
b
e
t
h
e
ca
u
s
e
o
f
in
f
er
io
r
p
er
f
o
r
m
an
ce
.
Usi
n
g
weig
h
ts
t
h
at
h
av
e
alr
ea
d
y
b
ee
n
tr
ain
e
d
in
th
e
s
am
e
d
o
m
ai
n
m
ig
h
t
im
p
r
o
v
e
p
er
f
o
r
m
a
n
ce
.
I
n
th
e
p
r
o
ce
s
s
o
f
tr
an
s
f
er
r
in
g
weig
h
ts
,
th
e
m
o
d
el
m
u
s
t
f
ir
s
t
b
e
tr
ain
ed
o
n
m
icr
o
s
co
p
ic
i
m
ag
es.
T
h
e
f
ea
tu
r
e
ex
tr
ac
to
r
d
ete
r
m
in
es
h
o
w
well
a
m
o
d
el
p
e
r
f
o
r
m
s
.
T
h
e
ar
ch
itectu
r
e
o
f
h
is
to
p
ath
o
lo
g
y
im
a
g
es
was
co
m
p
licated
,
s
tan
d
-
alo
n
e
p
att
er
n
ex
tr
ac
t
o
r
s
m
ig
h
t
well
h
a
v
e
tr
o
u
b
le
ex
tr
a
ctin
g
c
h
ar
ac
t
er
is
tics
.
A
d
if
f
er
en
t
ap
p
r
o
ac
h
to
ex
t
r
ac
tin
g
r
ic
h
f
ea
tu
r
es m
ig
h
t th
er
e
f
o
r
e
b
e
to
b
e
u
s
in
g
n
u
m
e
r
o
u
s
m
o
d
els f
o
r
e
x
tr
ac
tin
g
f
ea
tu
r
es.
I
m
p
lem
en
ted
b
o
th
a
n
o
r
m
al
f
aster
r
ec
u
r
r
en
t
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
(
n
o
r
m
al
F
R
C
NN
)
an
d
a
F
R
C
NN
alo
n
e
to
d
em
o
n
s
tr
at
e
th
e
ef
f
ec
tiv
en
ess
o
f
o
u
r
u
n
iq
u
e
two
-
s
tag
e
to
p
-
d
o
w
n
c
ascad
e
m
u
lti
-
s
ca
le
s
u
g
g
esti
o
n
-
g
en
e
r
atin
g
tech
n
i
q
u
e.
T
h
e
ef
f
ec
tiv
en
ess
o
f
o
u
r
p
r
o
p
o
s
ed
m
o
d
el
in
co
m
p
ar
is
o
n
to
th
e
o
th
er
two
b
en
ch
m
ar
k
m
o
d
els
ca
n
b
e
s
ee
n
in
T
ab
le
2
.
T
h
e
I
MFR
C
NN
alg
o
r
ith
m
was
en
h
a
n
ce
d
b
y
o
u
r
u
n
iq
u
e
p
r
o
p
o
s
al
n
etwo
r
k
f
o
r
r
eg
i
o
n
s
.
Ou
r
co
n
tr
ib
u
tio
n
r
aises
th
e
b
a
r
f
o
r
c
o
m
p
u
ter
ized
b
r
ea
s
t
ca
n
ce
r
p
r
o
g
n
o
s
is
,
p
o
te
n
tially
m
o
v
in
g
it c
lo
s
er
to
clin
ical
p
r
a
ctice
b
ec
o
m
in
g
e
n
tire
ly
au
t
o
m
ated
.
Fig
u
r
e
1
0
s
h
o
ws th
e
co
n
f
u
s
io
n
m
atr
ix
.
Fig
u
r
e
1
0
.
C
o
n
f
u
s
io
n
m
atr
ix
4.
CO
NCLU
SI
O
N
T
o
q
u
ick
ly
a
n
d
ac
cu
r
ately
id
en
tify
m
ito
tic
f
ig
u
r
es
in
b
r
ea
s
t
ca
n
ce
r
h
is
to
p
ath
o
lo
g
y
p
ictu
r
es,
r
esear
ch
er
s
s
u
g
g
est a
n
o
v
el
I
MFR
C
NN
s
tr
u
ctu
r
e.
I
n
co
m
p
a
r
is
o
n
to
th
e
ea
r
lier
p
r
o
p
o
s
ed
r
e
s
ea
r
ch
,
o
u
r
f
in
d
in
g
s
h
ig
h
lig
h
t
th
e
s
tr
en
g
th
o
f
o
u
r
p
r
o
p
o
s
ed
m
o
d
el.
T
h
e
h
ig
h
est
ac
cu
r
ac
y
to
d
ate
was
ac
h
iev
ed
b
y
o
u
r
p
r
o
p
o
s
ed
m
o
d
el,
w
h
ich
h
ad
a
n
F
-
s
co
r
e
v
alu
e
o
f
0
.
9
5
5
.
T
o
i
d
en
tify
m
i
to
tic
f
ig
u
r
es,
o
u
r
m
o
d
el
u
tili
ze
s
th
e
ch
ar
ac
te
r
is
tics
th
at
wer
e
lear
n
ed
,
d
e
m
o
n
s
tr
at
in
g
th
e
s
u
p
er
io
r
ity
o
f
lear
n
in
g
f
ea
tu
r
es
o
v
er
c
h
ar
ac
ter
is
tics
th
at
wer
e
cr
ea
ted
.
Fu
r
th
er
s
tu
d
ies
f
o
r
m
ito
tic
d
etec
tio
n
an
d
co
u
n
tin
g
u
s
in
g
wh
o
le‑
s
lid
e
im
ag
in
g
(
W
SI
)
co
u
ld
b
e
co
n
d
u
cted
.
Similar
ly
to
th
is
,
a
h
y
b
r
id
m
o
d
el
m
ig
h
t
b
e
ad
d
ed
t
o
th
e
o
b
ject
s
eg
m
en
tatio
n
m
o
d
el'
s
f
o
u
n
d
atio
n
to
ex
tr
ac
t
b
etter
f
ea
tu
r
es a
n
d
im
p
r
o
v
e
ef
f
icien
cy
.
ACK
NO
WL
E
DG
M
E
N
T
S
W
e
wo
u
ld
lik
e
to
ex
p
r
ess
o
u
r
h
ea
r
tf
elt
th
an
k
s
to
o
u
r
g
u
id
e
f
o
r
h
is
u
n
wav
er
in
g
g
u
id
a
n
ce
,
in
v
alu
ab
le
in
s
ig
h
ts
,
an
d
en
c
o
u
r
ag
e
m
en
t t
h
r
o
u
g
h
o
u
t th
e
r
esear
ch
p
r
o
ce
s
s
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
No
f
u
n
d
in
g
is
r
aised
f
o
r
th
is
r
e
s
ea
r
ch
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Pattan
M.
D.
Ali K
h
an
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
X
a
v
i
e
r
A
r
p
u
t
h
a
R
a
t
h
i
n
a
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
I
mp
r
o
vised
ma
s
k
fa
s
ter r
ec
u
r
r
en
t c
o
n
vo
l
u
tio
n
a
l
n
eu
r
a
l n
etw
o
r
k
fo
r
b
r
ea
s
t
…
(
P
a
tta
n
M.
D
.
A
li K
h
a
n
)
2007
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
I
NF
O
RM
E
D
CO
NS
E
N
T
W
e
h
av
e
o
b
tain
ed
in
f
o
r
m
ed
c
o
n
s
en
t f
r
o
m
all
in
d
iv
id
u
als in
c
lu
d
ed
in
t
h
is
s
tu
d
y
.
E
T
H
I
CAL AP
P
RO
V
AL
T
h
e
r
esear
ch
r
elate
d
to
h
u
m
a
n
u
s
e
h
as
b
ee
n
co
m
p
lie
d
with
all
th
e
r
elev
an
t
n
atio
n
al
r
eg
u
l
atio
n
s
an
d
in
s
titu
tio
n
al
p
o
licies
in
ac
co
r
d
an
ce
with
th
e
te
n
ets
o
f
t
h
e
He
ls
in
k
i
Dec
lar
atio
n
an
d
h
as
b
ee
n
ap
p
r
o
v
e
d
b
y
th
e
au
th
o
r
s
'
in
s
titu
tio
n
al
r
ev
iew
b
o
ar
d
o
r
eq
u
i
v
alen
t c
o
m
m
ittee.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
t
h
at
s
u
p
p
o
r
t
th
e
f
in
d
in
g
s
o
f
th
is
s
tu
d
y
ar
e
o
p
en
l
y
av
ailab
le
i
n
C
o
m
p
u
ter
s
&
E
lectr
ical
E
n
g
in
ee
r
in
g
at
h
ttp
://d
o
i.
o
r
g
/
1
0
.
1
0
1
6
/j.c
o
m
p
elec
en
g
.
2
0
2
1
.
1
0
7
0
3
8
,
r
ef
e
r
en
ce
n
u
m
b
e
r
[
3
1
]
.
RE
F
E
R
E
NC
E
S
[
1
]
L.
A
l
z
u
b
a
i
d
i
e
t
a
l
.
,
“
R
e
v
i
e
w
o
f
d
e
e
p
l
e
a
r
n
i
n
g
:
c
o
n
c
e
p
t
s,
C
N
N
a
r
c
h
i
t
e
c
t
u
r
e
s
,
c
h
a
l
l
e
n
g
e
s,
a
p
p
l
i
c
a
t
i
o
n
s
,
f
u
t
u
r
e
d
i
r
e
c
t
i
o
n
s,
”
J
o
u
rn
a
l
o
f
Bi
g
D
a
t
a
,
v
o
l
.
8
,
n
o
.
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
8
6
/
s
4
0
5
3
7
-
0
2
1
-
0
0
4
4
4
-
8.
[
2
]
T.
M
a
h
mo
o
d
,
M
.
A
r
sa
l
a
n
,
M
.
O
w
a
i
s
,
M
.
B
.
Le
e
,
a
n
d
K
.
R
.
P
a
r
k
,
“
A
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
-
b
a
se
d
m
i
t
o
si
s
d
e
t
e
c
t
i
o
n
i
n
b
r
e
a
s
t
c
a
n
c
e
r
h
i
s
t
o
p
a
t
h
o
l
o
g
y
i
ma
g
e
s
u
si
n
g
f
a
s
t
e
r
R
-
C
N
N
a
n
d
d
e
e
p
C
N
N
s,
”
J
o
u
rn
a
l
o
f
C
l
i
n
i
c
a
l
M
e
d
i
c
i
n
e
,
v
o
l
.
9
,
n
o
.
3
,
2
0
2
0
,
d
o
i
:
1
0
.
3
3
9
0
/
j
c
m
9
0
3
0
7
4
9
.
[
3
]
S
.
Y
.
S
i
d
d
i
q
u
i
e
t
a
l
.
,
“
I
o
M
T
c
l
o
u
d
-
b
a
se
d
i
n
t
e
l
l
i
g
e
n
t
p
r
e
d
i
c
t
i
o
n
o
f
b
r
e
a
st
c
a
n
c
e
r
s
t
a
g
e
s
e
mp
o
w
e
r
e
d
w
i
t
h
d
e
e
p
l
e
a
r
n
i
n
g
,
”
I
E
EE
Ac
c
e
ss
,
v
o
l
.
9
,
p
p
.
1
4
6
4
7
8
–
1
4
6
4
9
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
1
.
3
1
2
3
4
7
2
.
[
4
]
J.
V
.
D
.
La
a
k
,
G
.
Li
t
j
e
n
s
,
a
n
d
F
.
C
i
o
mp
i
,
“
D
e
e
p
l
e
a
r
n
i
n
g
i
n
h
i
st
o
p
a
t
h
o
l
o
g
y
:
t
h
e
p
a
t
h
t
o
t
h
e
c
l
i
n
i
c
,
”
N
a
t
u
r
e
M
e
d
i
c
i
n
e
,
v
o
l
.
2
7
,
n
o
.
5
,
p
p
.
7
7
5
–
7
8
4
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
1
-
0
2
1
-
0
1
3
4
3
-
4.
[
5
]
P
.
H
a
r
r
i
s
o
n
a
n
d
K
.
P
a
r
k
,
“
Tu
m
o
r
d
e
t
e
c
t
i
o
n
i
n
b
r
e
a
st
h
i
s
t
o
p
a
t
h
o
l
o
g
i
c
a
l
i
ma
g
e
s
u
si
n
g
f
a
s
t
e
r
R
-
C
N
N
,
”
i
n
2
0
2
1
I
n
t
e
rn
a
t
i
o
n
a
l
S
y
m
p
o
s
i
u
m
o
n
Me
d
i
c
a
l
R
o
b
o
t
i
c
s
,
2
0
2
1
,
p
p
.
1
–
7
,
d
o
i
:
1
0
.
1
1
0
9
/
I
S
M
R
4
8
3
4
6
.
2
0
2
1
.
9
6
6
1
4
8
3
.
[
6
]
A
.
D
.
P
r
a
t
i
w
i
a
n
d
I
.
P
.
S
a
r
i
,
“
R
o
b
u
s
t
b
r
e
a
s
t
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
u
si
n
g
f
a
s
t
e
r
R
-
C
N
N
a
l
g
o
r
i
t
ma,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
f
o
rm
a
t
i
c
s
a
n
d
C
o
m
p
u
t
a
t
i
o
n
,
v
o
l
.
4
,
n
o
.
1
,
2
0
2
2
,
d
o
i
:
1
0
.
3
5
8
4
2
/
i
j
i
c
o
m.
v
4
i
1
.
4
9
.
[
7
]
K
.
G
.
C
h
a
u
d
h
a
r
i
,
“
C
o
m
p
a
r
a
t
i
v
e
a
n
a
l
y
si
s o
f
C
N
N
m
o
d
e
l
s
t
o
d
i
a
g
n
o
se
b
r
e
a
s
t
c
a
n
c
e
r
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
n
o
v
a
t
i
v
e
Re
s
e
a
r
c
h
i
n
S
c
i
e
n
c
e
,
E
n
g
i
n
e
e
r
i
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
7
,
n
o
.
1
0
,
p
p
.
8
1
8
0
–
8
1
8
7
,
2
0
1
8
,
d
o
i
:
1
0
.
1
5
6
8
0
/
I
JI
R
S
ET.
2
0
1
8
.
0
7
1
0
0
7
4
.
[
8
]
H
.
H
u
a
n
g
,
X
.
F
e
n
g
,
J.
Ji
a
n
g
,
P
.
C
h
e
n
,
a
n
d
S
.
Z
h
o
u
,
“
M
a
sk
R
C
N
N
a
l
g
o
r
i
t
h
m
f
o
r
n
u
c
l
e
i
d
e
t
e
c
t
i
o
n
o
n
b
r
e
a
s
t
c
a
n
c
e
r
h
i
s
t
o
p
a
t
h
o
l
o
g
i
c
a
l
i
ma
g
e
s,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
m
a
g
i
n
g
S
y
s
t
e
m
s
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
3
2
,
n
o
.
1
,
p
p
.
2
0
9
–
2
1
7
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
2
/
i
ma
.
2
2
6
1
8
.
[
9
]
L.
Li
u
e
t
a
l
.
,
“
C
o
l
l
a
b
o
r
a
t
i
v
e
t
r
a
n
sf
e
r
n
e
t
w
o
r
k
f
o
r
m
u
l
t
i
-
c
l
a
ssi
f
i
c
a
t
i
o
n
o
f
b
r
e
a
st
c
a
n
c
e
r
h
i
s
t
o
p
a
t
h
o
l
o
g
i
c
a
l
i
m
a
g
e
s,
”
I
E
EE
J
o
u
r
n
a
l
o
f
Bi
o
m
e
d
i
c
a
l
a
n
d
H
e
a
l
t
h
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
2
8
,
n
o
.
1
,
p
p
.
1
1
0
–
1
2
1
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
JB
H
I
.
2
0
2
3
.
3
2
8
3
0
4
2
.
[
1
0
]
S
.
A
y
a
s
h
m
,
M
.
V
a
l
i
z
a
d
e
h
,
M
.
C
.
A
m
i
r
a
n
i
,
a
n
d
S
.
M
i
h
a
n
d
o
o
s
t
,
“
M
i
t
o
s
i
s
d
e
t
e
c
t
i
o
n
i
n
b
r
e
a
s
t
h
i
s
t
o
p
a
t
h
o
l
o
g
y
i
m
a
g
e
s
u
s
i
n
g
a
s
e
l
f
-
a
t
t
e
n
t
i
o
n
-
e
n
h
a
n
c
e
d
F
a
s
t
e
r
R
-
C
N
N
f
r
a
m
e
w
o
r
k
,
”
I
E
E
E
A
c
c
e
s
s
,
v
o
l
.
1
4
,
p
p
.
1
6
4
5
5
–
1
6
4
6
9
,
2
0
2
6
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
E
S
S
.
2
0
2
6
.
3
6
5
8
6
4
6
.
[
1
1
]
V
.
P
a
t
e
l
,
V
.
C
h
a
u
r
a
si
a
,
R
.
M
a
h
a
d
e
v
a
,
a
n
d
S
.
P
.
P
a
t
o
l
e
,
“
G
A
R
L
-
N
e
t
:
g
r
a
p
h
b
a
se
d
a
d
a
p
t
i
v
e
r
e
g
u
l
a
r
i
z
e
d
l
e
a
r
n
i
n
g
d
e
e
p
n
e
t
w
o
r
k
f
o
r
b
r
e
a
s
t
c
a
n
c
e
r
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
I
EE
E
A
c
c
e
ss
,
v
o
l
.
1
1
,
p
p
.
9
0
9
5
–
9
1
1
2
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
3
.
3
2
3
9
6
7
1
.
[
1
2
]
Y
.
K
a
w
a
z
o
e
e
t
a
l
.
,
“
F
a
st
e
r
R
-
C
N
N
-
b
a
se
d
g
l
o
m
e
r
u
l
a
r
d
e
t
e
c
t
i
o
n
i
n
m
u
l
t
i
-
st
a
i
n
e
d
h
u
ma
n
w
h
o
l
e
s
l
i
d
e
i
ma
g
e
s
,
”
J
o
u
rn
a
l
o
f
I
m
a
g
i
n
g
,
v
o
l
.
4
,
n
o
.
7
,
2
0
1
8
,
d
o
i
:
1
0
.
3
3
9
0
/
j
i
ma
g
i
n
g
4
0
7
0
0
9
1
.
[
1
3
]
S
.
A
.
S
á
n
c
h
e
z
,
A
.
D
.
M
o
r
a
l
e
s
,
M
.
F
.
A
r
r
o
y
o
,
a
n
d
y
.
H
.
D
.
A
s
í
s,
“
A
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
mo
d
e
l
f
o
r
t
h
e
p
r
e
d
i
c
t
i
o
n
o
f
m
a
l
i
g
n
a
n
t
t
u
m
o
r
s
u
si
n
g
a
se
t
o
f
me
d
i
c
a
l
i
m
a
g
e
s
f
r
o
m
mamm
o
g
r
a
p
h
y
s
t
u
d
i
e
s,
”
I
O
P
C
o
n
f
e
r
e
n
c
e
S
e
r
i
e
s:
Ma
t
e
r
i
a
l
s
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
1
1
5
4
,
n
o
.
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
8
8
/
1
7
5
7
-
8
9
9
X
/
1
1
5
4
/
1
/
0
1
2
0
1
3
.
[
1
4
]
N
.
D
i
f
a
n
d
Z
.
E
l
b
e
r
r
i
c
h
i
,
“
D
e
e
p
l
e
a
r
n
i
n
g
me
t
h
o
d
s
f
o
r
mi
t
o
si
s
d
e
t
e
c
t
i
o
n
i
n
b
r
e
a
st
c
a
n
c
e
r
h
i
s
t
o
p
a
t
h
o
l
o
g
i
c
a
l
i
m
a
g
e
s:
a
c
o
m
p
r
e
h
e
n
si
v
e
r
e
v
i
e
w
,
”
i
n
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
Ma
c
h
i
n
e
L
e
a
r
n
i
n
g
f
o
r
D
i
g
i
t
a
l
P
a
t
h
o
l
o
g
y
,
C
h
a
m,
S
w
i
t
z
e
r
l
a
n
d
:
S
p
r
i
n
g
e
r
,
2
0
2
0
,
p
p
.
2
7
9
–
3
0
6
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
0
-
5
0
4
0
2
-
1
_
1
7
.
[
1
5
]
I
.
S
i
r
a
z
i
t
d
i
n
o
v
,
M
.
K
h
o
l
i
a
v
c
h
e
n
k
o
,
T
.
M
u
s
t
a
f
a
e
v
,
Y
.
Y
i
x
u
a
n
,
R
.
K
u
l
e
e
v
,
a
n
d
B
.
I
b
r
a
g
i
m
o
v
,
“
D
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
e
n
sem
b
l
e
f
o
r
p
n
e
u
mo
n
i
a
l
o
c
a
l
i
z
a
t
i
o
n
f
r
o
m
a
l
a
r
g
e
-
sca
l
e
c
h
e
s
t
X
-
r
a
y
d
a
t
a
b
a
s
e
,
”
C
o
m
p
u
t
e
rs
&
El
e
c
t
r
i
c
a
l
En
g
i
n
e
e
r
i
n
g
,
v
o
l
.
7
8
,
p
p
.
3
8
8
–
3
9
9
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
mp
e
l
e
c
e
n
g
.
2
0
1
9
.
0
8
.
0
0
4
.
[
1
6
]
I
.
G
i
r
o
l
a
mi
e
t
a
l
.
,
“
A
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
a
p
p
l
i
c
a
t
i
o
n
s
f
o
r
p
r
e
-
i
m
p
l
a
n
t
a
t
i
o
n
k
i
d
n
e
y
b
i
o
p
s
y
p
a
t
h
o
l
o
g
y
p
r
a
c
t
i
c
e
:
a
sy
s
t
e
mat
i
c
r
e
v
i
e
w
,
”
J
o
u
r
n
a
l
o
f
N
e
p
h
r
o
l
o
g
y
,
v
o
l
.
3
5
,
n
o
.
7
,
p
p
.
1
8
0
1
–
1
8
0
8
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s
4
0
6
2
0
-
0
2
2
-
0
1
3
2
7
-
8.
[
1
7
]
E.
U
c
h
i
n
o
e
t
a
l
.
,
“
C
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
g
l
o
m
e
r
u
l
a
r
p
a
t
h
o
l
o
g
i
c
a
l
f
i
n
d
i
n
g
s
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
a
n
d
n
e
p
h
r
o
l
o
g
i
st
–
A
I
c
o
l
l
e
c
t
i
v
e
i
n
t
e
l
l
i
g
e
n
c
e
a
p
p
r
o
a
c
h
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Me
d
i
c
a
l
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
1
4
1
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
j
me
d
i
n
f
.
2
0
2
0
.
1
0
4
2
3
1
.
[
1
8
]
N
.
A
l
t
i
n
i
e
t
a
l
.
,
“
S
e
m
a
n
t
i
c
s
e
g
m
e
n
t
a
t
i
o
n
f
r
a
mew
o
r
k
f
o
r
g
l
o
mer
u
l
i
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
i
n
k
i
d
n
e
y
h
i
st
o
l
o
g
i
c
a
l
s
e
c
t
i
o
n
s,
”
El
e
c
t
r
o
n
i
c
s
,
v
o
l
.
9
,
n
o
.
3
,
2
0
2
0
,
d
o
i
:
1
0
.
3
3
9
0
/
e
l
e
c
t
r
o
n
i
c
s
9
0
3
0
5
0
3
.
[
1
9
]
E.
H
.
N
g
u
y
e
n
e
t
a
l
.
,
“
C
i
r
c
l
e
r
e
p
r
e
s
e
n
t
a
t
i
o
n
f
o
r
m
e
d
i
c
a
l
o
b
j
e
c
t
d
e
t
e
c
t
i
o
n
,
”
I
EEE
T
r
a
n
sa
c
t
i
o
n
s
o
n
Me
d
i
c
a
l
I
m
a
g
i
n
g
,
v
o
l
.
4
1
,
n
o
.
3
,
p
p
.
7
4
6
–
7
5
4
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
T
M
I
.
2
0
2
1
.
3
1
2
2
8
3
5
.
[
2
0
]
M
.
A
l
j
a
b
r
i
,
M
.
A
l
A
mi
r
,
M
.
A
l
G
h
a
m
d
i
,
M
.
A
.
-
M
o
t
t
a
l
e
b
,
a
n
d
F
.
C
.
-
M
e
sa,
“
To
w
a
r
d
s a
b
e
t
t
e
r
u
n
d
e
r
st
a
n
d
i
n
g
o
f
a
n
n
o
t
a
t
i
o
n
t
o
o
l
s
f
o
r
med
i
c
a
l
i
ma
g
i
n
g
:
a
su
r
v
e
y
,
”
M
u
l
t
i
m
e
d
i
a
T
o
o
l
s
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
8
1
,
n
o
.
1
8
,
p
p
.
2
5
8
7
7
–
2
5
9
1
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
0
4
2
-
0
2
2
-
1
2
1
0
0
-
1.
[
2
1
]
R
.
G
.
-
C
a
b
a
l
l
e
r
o
,
C
.
J.
G
.
-
O
r
e
l
l
a
n
a
,
A
.
G
.
-
M
a
n
s
o
,
H
.
M
.
G
.
-
V
e
l
a
sc
o
,
R
.
T.
-
M
o
l
i
n
a
,
a
n
d
M
.
M
.
-
M
a
c
í
a
s,
“
P
r
e
c
i
s
e
p
o
l
l
e
n
g
r
a
i
n
d
e
t
e
c
t
i
o
n
i
n
b
r
i
g
h
t
f
i
e
l
d
mi
c
r
o
s
c
o
p
y
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
,
”
S
e
n
s
o
rs
,
v
o
l
.
1
9
,
n
o
.
1
6
,
2
0
1
9
,
d
o
i
:
1
0
.
3
3
9
0
/
s1
9
1
6
3
5
8
3
.
[
2
2
]
Y
.
L
u
e
t
a
l
.
,
“
I
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
me
t
a
st
a
t
i
c
l
y
mp
h
n
o
d
e
s
i
n
M
R
i
m
a
g
i
n
g
w
i
t
h
f
a
st
e
r
r
e
g
i
o
n
-
b
a
se
d
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,
”
C
a
n
c
e
r R
e
se
a
rc
h
,
v
o
l
.
7
8
,
n
o
.
1
7
,
p
p
.
5
1
3
5
–
5
1
4
3
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
5
8
/
0
0
0
8
-
5
4
7
2
.
C
A
N
-
18
-
0
4
9
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
2
,
Ap
r
il 2
0
2
6
:
1
9
9
9
-
2
0
0
8
2008
[
2
3
]
H
.
S
o
l
t
a
n
i
,
M
.
A
mr
o
u
n
e
,
I
.
B
e
n
d
i
b
,
a
n
d
M
.
Y
.
H
a
o
u
a
m
,
“
B
r
e
a
st
c
a
n
c
e
r
l
e
si
o
n
d
e
t
e
c
t
i
o
n
a
n
d
s
e
g
me
n
t
a
t
i
o
n
b
a
sed
o
n
m
a
sk
R
-
C
N
N
,
”
i
n
2
0
2
1
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
ren
c
e
o
n
Re
c
e
n
t
A
d
v
a
n
c
e
s
i
n
M
a
t
h
e
m
a
t
i
c
s
a
n
d
I
n
f
o
rm
a
t
i
c
s
(
I
C
RAM
I
)
,
2
0
2
1
,
p
p
.
1
–
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
R
A
M
I
5
2
6
2
2
.
2
0
2
1
.
9
5
8
5
9
1
3
.
[
2
4
]
T.
P
.
La
t
c
h
o
u
m
i
,
G
.
K
a
l
u
s
u
r
a
m
a
n
,
J.
F
.
B
a
n
u
,
T.
L
.
Y
o
o
k
e
sh
,
T.
P
.
E
z
h
i
l
a
r
a
si
,
a
n
d
K
.
B
a
l
a
m
u
r
u
g
a
n
,
“
En
h
a
n
c
e
me
n
t
i
n
man
u
f
a
c
t
u
r
i
n
g
s
y
s
t
e
ms u
si
n
g
g
r
e
y
-
f
u
z
z
y
a
n
d
LK
-
S
V
M
a
p
p
r
o
a
c
h
,
”
i
n
2
0
2
1
I
EEE
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
I
n
t
e
l
l
i
g
e
n
t
S
y
st
e
m
s
,
S
m
a
r
t
a
n
d
G
r
e
e
n
T
e
c
h
n
o
l
o
g
i
e
s (I
C
I
S
S
G
T
)
,
2
0
2
1
,
p
p
.
7
2
–
78
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
I
S
S
G
T5
2
0
2
5
.
2
0
2
1
.
0
0
0
2
6
.
[
2
5
]
A
.
S
o
h
a
i
l
,
A
.
K
h
a
n
,
N
.
W
a
h
a
b
,
A
.
Z
a
mee
r
,
a
n
d
S
.
K
h
a
n
,
“
A
mu
l
t
i
-
p
h
a
s
e
d
e
e
p
C
N
N
b
a
se
d
mi
t
o
s
i
s
d
e
t
e
c
t
i
o
n
f
r
a
mew
o
r
k
f
o
r
b
r
e
a
st
c
a
n
c
e
r
h
i
st
o
p
a
t
h
o
l
o
g
i
c
a
l
i
m
a
g
e
s,”
S
c
i
e
n
t
i
f
i
c
Re
p
o
rt
s
,
v
o
l
.
1
1
,
n
o
.
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
3
8
/
s4
1
5
9
8
-
0
2
1
-
8
5
6
5
2
-
1.
[
2
6
]
B
.
S
h
e
n
g
,
M
.
Z
h
o
u
,
M
.
H
u
,
Q
.
Li
,
L.
S
u
n
,
a
n
d
Y
.
W
e
n
,
“
A
b
l
o
o
d
c
e
l
l
d
a
t
a
set
f
o
r
l
y
mp
h
o
ma
c
l
a
ssi
f
i
c
a
t
i
o
n
u
s
i
n
g
f
a
st
e
r
R
-
C
N
N
,
”
Bi
o
t
e
c
h
n
o
l
o
g
y
&
Bi
o
t
e
c
h
n
o
l
o
g
i
c
a
l
E
q
u
i
p
m
e
n
t
,
v
o
l
.
3
4
,
n
o
.
1
,
p
p
.
4
1
3
–
4
2
0
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
8
0
/
1
3
1
0
2
8
1
8
.
2
0
2
0
.
1
7
6
5
8
7
1
.
[
2
7
]
B
.
G
a
mi
,
K
.
C
h
a
u
h
a
n
,
a
n
d
B
.
Y
.
P
a
n
c
h
a
l
,
“
B
r
e
a
st
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
”
i
n
M
o
b
i
l
e
R
a
d
i
o
C
o
m
m
u
n
i
c
a
t
i
o
n
s
a
n
d
5
G
N
e
t
w
o
rks
,
2
0
2
3
,
p
p
.
8
5
–
95
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
9
8
1
-
19
-
7
9
8
2
-
8
_
8
.
[
2
8
]
X
.
L
i
,
Z.
X
u
,
X
.
S
h
e
n
,
Y
.
Z
h
o
u
,
B
.
X
i
a
o
,
a
n
d
T.
-
Q
.
L
i
,
“
D
e
t
e
c
t
i
o
n
o
f
c
e
r
v
i
c
a
l
c
a
n
c
e
r
c
e
l
l
s
i
n
w
h
o
l
e
sl
i
d
e
i
ma
g
e
s
u
si
n
g
d
e
f
o
r
ma
b
l
e
a
n
d
g
l
o
b
a
l
c
o
n
t
e
x
t
a
w
a
r
e
f
a
s
t
e
r
R
C
N
N
-
F
P
N
,
”
C
u
rre
n
t
O
n
c
o
l
o
g
y
,
v
o
l
.
2
8
,
n
o
.
5
,
p
p
.
3
5
8
5
–
3
6
0
1
,
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
c
u
r
r
o
n
c
o
l
2
8
0
5
0
3
0
7
.
[
2
9
]
S
.
Jar
u
j
u
n
a
w
o
n
g
a
n
d
P
.
H
o
r
k
a
e
w
,
“
En
h
a
n
c
i
n
g
A
I
-
D
r
i
v
e
n
d
i
a
g
n
o
si
s
o
f
i
n
v
a
s
i
v
e
d
u
c
t
a
l
c
a
r
c
i
n
o
ma
w
i
t
h
m
o
r
p
h
o
l
o
g
i
c
a
l
l
y
g
u
i
d
e
d
a
n
d
i
n
t
e
r
p
r
e
t
a
b
l
e
d
e
e
p
l
e
a
r
n
i
n
g
,
”
A
p
p
l
i
e
d
S
c
i
e
n
c
e
s
,
v
o
l
.
1
5
,
n
o
.
1
2
,
2
0
2
5
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
1
5
1
2
6
8
8
3
.
[
3
0
]
B
.
M
u
r
u
g
e
s
a
n
,
S
.
S
e
l
v
a
r
a
j
,
K
.
S
a
r
v
e
sw
a
r
a
n
,
K
.
R
a
m,
J.
Jo
se
p
h
,
a
n
d
M
.
S
i
v
a
p
r
a
k
a
s
a
m,
“
D
e
e
p
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
o
f
mi
t
o
t
i
c
f
i
g
u
r
e
s
,
”
i
n
M
e
d
i
c
a
l
I
m
a
g
i
n
g
2
0
1
9
:
D
i
g
i
t
a
l
P
a
t
h
o
l
o
g
y
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
1
7
/
1
2
.
2
5
0
8
7
7
0
.
[
3
1
]
X
.
P
a
n
e
t
a
l
.
,
“
M
i
t
o
s
i
s
d
e
t
e
c
t
i
o
n
t
e
c
h
n
i
q
u
e
s
i
n
H
a
n
d
E
st
a
i
n
e
d
b
r
e
a
st
c
a
n
c
e
r
p
a
t
h
o
l
o
g
i
c
a
l
i
ma
g
e
s
:
a
c
o
mp
r
e
h
e
n
s
i
v
e
r
e
v
i
e
w
,
”
C
o
m
p
u
t
e
rs
& E
l
e
c
t
r
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
9
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
m
p
e
l
e
c
e
n
g
.
2
0
2
1
.
1
0
7
0
3
8
.
[
3
2
]
Y
.
Zh
a
o
,
J.
Z
h
a
n
g
,
D
.
H
u
,
H
.
Q
u
,
Y
.
Ti
a
n
,
a
n
d
X
.
C
u
i
,
“
A
p
p
l
i
c
a
t
i
o
n
o
f
d
e
e
p
l
e
a
r
n
i
n
g
i
n
h
i
s
t
o
p
a
t
h
o
l
o
g
y
i
ma
g
e
s o
f
b
r
e
a
s
t
c
a
n
c
e
r
:
a
r
e
v
i
e
w
,
”
M
i
c
r
o
m
a
c
h
i
n
e
s
,
v
o
l
.
1
3
,
n
o
.
1
2
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
mi
1
3
1
2
2
1
9
7
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Pa
tta
n
M.
D.
Ali
K
h
a
n
e
a
rn
e
d
h
is
Ba
c
h
e
lo
r’s
o
f
Tec
h
n
o
l
o
g
y
B.
Tec
h
.
d
e
g
re
e
in
Co
m
p
u
ter
S
c
ien
c
e
E
n
g
i
n
e
e
rin
g
fr
o
m
JN
TUA
in
2
0
0
5
.
He
h
a
s
o
b
tain
e
d
h
is
M
a
ste
r’s
d
e
g
re
e
i
n
M.E
.
(
C
o
m
p
u
ter
S
c
ien
c
e
En
g
in
e
e
rin
g
)
fr
o
m
S
t.
P
e
ter’s
Un
iv
e
rsit
y
in
2
0
1
2
.
C
u
rre
n
tl
y
,
h
e
is
a
re
se
a
rc
h
sc
h
o
lar
a
t
B.
S
.
Ab
d
u
r
Ra
h
m
a
n
Cre
sc
e
n
t
In
sti
tu
te
o
f
S
c
ien
c
e
a
n
d
Tec
h
n
o
lo
g
y
(De
e
m
e
d
to
b
e
Un
iv
e
rsit
y
)
d
o
in
g
h
is
P
h
.
D
.
i
n
C
o
m
p
u
ter
S
c
ie
n
c
e
a
n
d
En
g
in
e
e
ri
n
g
.
He
h
a
s
a
tt
e
n
d
e
d
m
a
n
y
w
o
rk
s
h
o
p
s
a
n
d
i
n
d
u
c
t
io
n
p
r
o
g
ra
m
s
c
o
n
d
u
c
ted
b
y
v
a
rio
u
s
Un
i
v
e
rsiti
e
s.
His
a
re
a
s
o
f
in
tere
st
a
re
m
a
c
h
i
n
e
lea
rn
in
g
a
n
d
ima
g
e
p
r
o
c
e
ss
in
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
a
li
k
_
1
2
@re
d
iff
m
a
il
.
c
o
m
.
Xa
v
ier
Ar
p
u
th
a
Ra
th
i
n
a
is
a
n
a
ss
o
c
iate
p
ro
fe
ss
o
r
in
D
e
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
a
t
S
c
h
o
o
l
o
f
S
CIM
S
o
f
B.
S
.
A
b
d
u
r
Ra
h
m
a
n
Cre
sc
e
n
t
In
stit
u
te
o
f
S
c
ien
c
e
a
n
d
Tec
h
n
o
l
o
g
y
(De
e
m
e
d
t
o
b
e
U
n
iv
e
rsit
y
)
C
h
e
n
n
a
i
wit
h
a
n
e
x
p
e
rien
c
e
o
f
2
7
y
e
a
rs
i
n
tea
c
h
in
g
.
S
h
e
d
i
d
h
e
r
B.
E
.
in
El
e
c
tro
n
ics
Co
m
m
u
n
ica
ti
o
n
E
n
g
i
n
e
e
rin
g
a
n
d
M
a
ste
r’s
d
e
g
re
e
i
n
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
a
n
d
P
h
.
D
.
i
n
C
o
m
p
u
ter
S
c
i
e
n
c
e
a
n
d
En
g
in
e
e
rin
g
.
He
r
a
re
a
s
o
f
in
tere
st
a
re
ima
g
e
p
ro
c
e
s
sin
g
,
e
m
o
ti
o
n
a
l
in
telli
g
e
n
c
e
,
a
n
d
sp
e
e
c
h
p
ro
c
e
ss
in
g
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
th
is
e
m
a
il
:
x
a
ra
th
i
n
a
@c
re
sc
e
n
t.
e
d
u
c
a
ti
o
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.