I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
3
7
,
No
.
3
,
Ma
r
ch
20
2
5
,
p
p
.
1
92
7
~
1
93
5
I
SS
N:
2
502
-
4
7
52
,
DOI
: 1
0
.
1
1
5
9
1
/ijee
cs
.v
3
7
.
i
3
.
pp
1
92
7
-
1
93
5
1927
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs
.
ia
esco
r
e.
co
m
Em
po
wer
Brea
st
Net:
b
reas
t
c
a
nce
r det
e
ction
with
trans
fer
l
ea
rning
VG
G
N
et
-
19
Va
is
ha
li M
.
J
o
s
hi
1
,
P
ra
j
k
t
a
P
.
Da
nd
a
v
a
t
e
2
,
R
a
s
hm
i R
a
ma
m
urt
hy
3
,
Ridd
hi M
ira
j
k
a
r
4
,
Nee
t
a
N.
T
hu
ne
5
,
G
it
a
nja
li R.
Sh
ind
e
6
1
D
e
p
a
r
t
me
n
t
o
f
E
a
n
d
TC
En
g
i
n
e
e
r
i
n
g
,
K
JEI
’
s T
r
i
n
i
t
y
A
c
a
d
e
m
y
o
f
E
n
g
i
n
e
e
r
i
n
g
,
P
u
n
e
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
En
g
i
n
e
e
r
i
n
g
,
V
i
sh
w
a
k
a
r
ma
I
n
st
i
t
u
t
e
o
f
T
e
c
h
n
o
l
o
g
y
,
P
u
n
e
,
I
n
d
i
a
3
D
e
p
a
r
t
me
n
t
o
f
E
a
n
d
TC
En
g
i
n
e
e
r
i
n
g
,
P
a
d
ma
b
h
o
o
s
h
a
n
V
a
s
a
n
t
d
a
d
a
P
a
t
i
l
I
n
st
i
t
u
t
e
o
f
T
e
c
h
n
o
l
o
g
y
,
P
u
n
e
,
I
n
d
i
a
4
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
Te
c
h
n
o
l
o
g
y
,
V
i
s
h
w
a
k
a
r
m
a
I
n
st
i
t
u
t
e
o
f
I
n
f
o
r
ma
t
i
o
n
Te
c
h
n
o
l
o
g
y
,
K
o
n
d
h
w
a
(
b
k
)
,
P
u
n
e
,
I
n
d
i
a
5
D
e
p
a
r
t
me
n
t
o
f
E
a
n
d
TC
En
g
i
n
e
e
r
i
n
g
,
M
a
r
a
t
h
w
a
d
a
C
o
l
l
e
g
e
o
f
E
n
g
i
n
e
e
r
i
n
g
,
P
u
n
e
,
I
n
d
i
a
6
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
En
g
i
n
e
e
r
i
n
g
(AI
-
M
L)
,
V
i
s
h
w
a
k
a
r
ma
I
n
st
i
t
u
t
e
o
f
I
n
f
o
r
mat
i
o
n
T
e
c
h
n
o
l
o
g
y
,
K
o
n
d
h
w
a
(
b
k
)
,
P
u
n
e
,
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
ul
1
,
2
0
2
4
R
ev
is
ed
Sep
26
,
2
0
2
4
Acc
ep
ted
Oct
7
,
2
0
2
4
Bre
a
st
c
a
n
c
e
r
is
a
m
a
jo
r
c
a
u
se
o
f
d
e
a
th
a
m
o
n
g
wo
m
e
n
g
l
o
b
a
ll
y
,
m
a
k
in
g
e
a
rly
d
e
tec
ti
o
n
c
r
u
c
ial
fo
r
e
ffe
c
ti
v
e
trea
tme
n
t.
Th
is
st
u
d
y
in
tro
d
u
c
e
s
a
n
e
w
d
e
e
p
lea
rn
i
n
g
(DL)
m
e
th
o
d
u
si
n
g
tran
sfe
r
lea
rn
in
g
(
TL
)
to
a
u
t
o
m
a
ti
c
a
ll
y
d
e
tec
t
a
n
d
d
iag
n
o
se
b
re
a
st
c
a
n
c
e
r.
TL
imp
r
o
v
e
s
p
e
rf
o
rm
a
n
c
e
o
n
n
e
w
tas
k
s
b
y
u
si
n
g
k
n
o
wle
d
g
e
fro
m
p
re
v
i
o
u
s
tas
k
s.
I
n
t
h
is
stu
d
y
,
we
u
se
p
re
-
train
e
d
c
o
n
v
o
lu
ti
o
n
a
l
n
e
u
ra
l
n
e
two
r
k
s
(CNN
s)
li
k
e
Ale
x
Ne
t,
Re
sN
e
t5
0
,
v
isu
a
l
g
e
o
m
e
try
g
r
o
u
p
(VG
G
)
-
1
6
,
a
n
d
VG
G
-
1
9
to
e
x
trac
t
fe
a
tu
re
s
fro
m
th
e
b
re
a
st
c
a
n
c
e
r
w
isc
o
n
sin
(
BCW)
d
iag
n
o
stic
d
a
tas
e
t.
We
m
e
a
su
re
th
e
m
o
d
e
l
’
s
su
c
c
e
ss
with
a
c
c
u
ra
c
y
,
se
n
siti
v
it
y
,
sp
e
c
ifi
c
it
y
,
p
re
c
isio
n
,
a
n
d
F
-
sc
o
re
.
Th
e
re
su
lt
s
sh
o
w
t
h
a
t
t
h
e
VG
G
-
1
9
m
o
d
e
l,
wh
e
n
a
p
p
li
e
d
with
TL
,
p
e
rfo
rm
s
b
e
st
fo
r
d
iag
n
o
si
n
g
b
re
a
st
c
a
n
c
e
r
,
a
c
h
iev
i
n
g
a
n
o
v
e
ra
ll
a
c
c
u
ra
c
y
o
f
9
8
.
7
5
%
,
se
n
siti
v
it
y
o
f
9
7
.
3
8
%
,
sp
e
c
ifi
c
it
y
o
f
9
8
.
3
5
%
,
p
re
c
is
io
n
o
f
9
7
.
3
5
%
,
a
n
d
a
n
F
-
sc
o
re
o
f
9
7
.
6
6
%
.
K
ey
w
o
r
d
s
:
B
r
ea
s
t c
an
ce
r
Dee
p
lear
n
in
g
T
r
an
s
f
er
lear
n
i
n
g
C
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
s
VGG
-
19
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
:
Vaish
ali
M.
J
o
s
h
i
Dep
ar
tm
en
t o
f
E
an
d
T
C
E
n
g
i
n
ee
r
in
g
,
KJEI
’
s
T
r
i
n
ity
Aca
d
e
m
y
o
f
E
n
g
in
ee
r
in
g
Pu
n
e,
I
n
d
ia
E
m
ail:
jo
s
h
iv
aish
ali8
6
7
5
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
B
r
ea
s
t
ca
n
ce
r
co
n
tin
u
es
to
p
o
s
e
a
s
ig
n
if
ican
t
h
ea
lth
ch
all
en
g
e
g
lo
b
ally
,
r
ep
r
esen
tin
g
a
s
u
b
s
tan
tial
p
o
r
tio
n
o
f
ca
n
ce
r
d
iag
n
o
s
es
an
d
m
o
r
tality
r
ates,
p
ar
ticu
la
r
ly
am
o
n
g
wo
m
e
n
.
Desp
ite
n
o
tab
le
p
r
o
g
r
ess
in
s
cr
ee
n
in
g
p
r
o
g
r
a
m
s
an
d
tr
ea
t
m
en
t
s
tr
ateg
ies,
th
e
tim
ely
d
etec
tio
n
an
d
ac
c
u
r
ate
d
ia
g
n
o
s
is
o
f
b
r
ea
s
t
ca
n
ce
r
r
em
ain
p
iv
o
tal
f
o
r
im
p
r
o
v
in
g
p
atien
t
o
u
tco
m
es
an
d
s
u
r
v
iv
al
r
ates
[
1
]
.
Ho
wev
er
,
ex
is
tin
g
d
iag
n
o
s
tic
m
eth
o
d
s
f
ac
e
in
h
er
en
t
lim
itatio
n
s
s
te
m
m
in
g
f
r
o
m
th
e
co
m
p
lex
ity
an
d
v
ar
iab
ilit
y
o
f
b
r
ea
s
t
ca
n
ce
r
m
an
if
estatio
n
s
,
co
m
p
o
u
n
d
e
d
b
y
th
e
s
u
b
jectiv
e
n
atu
r
e
o
f
in
ter
p
r
etatio
n
in
tr
ad
itio
n
al
ap
p
r
o
ac
h
es.
T
h
ese
ch
allen
g
es
co
n
tr
ib
u
te
to
d
iag
n
o
s
tic
er
r
o
r
s
,
tr
ea
tm
en
t
d
elay
s
,
an
d
s
u
b
o
p
tim
al
p
atien
t
ca
r
e.
I
n
th
is
c
o
n
tex
t,
th
e
e
m
er
g
en
c
e
o
f
d
ee
p
lear
n
i
n
g
(
DL
)
,
a
s
u
b
s
et
o
f
ar
tific
ial
in
tellig
en
ce
(
AI
)
th
at
ex
ce
ls
in
lear
n
in
g
in
tr
icate
p
a
tter
n
s
an
d
f
ea
tu
r
es
f
r
o
m
v
ast
am
o
u
n
ts
o
f
d
ata,
o
f
f
e
r
s
n
ew
av
en
u
es
to
en
h
a
n
ce
b
r
ea
s
t
ca
n
ce
r
d
etec
tio
n
an
d
d
iag
n
o
s
is
[
2
]
,
[
3
]
.
C
o
n
v
o
lu
ti
o
n
al
n
eu
r
al
n
etwo
r
k
s
(
C
NNs),
a
p
r
o
m
in
e
n
t
DL
ar
ch
itectu
r
e,
h
av
e
d
em
o
n
s
tr
at
ed
r
em
ar
k
a
b
le
ca
p
ab
ilit
ies
in
au
to
m
atica
lly
ex
tr
ac
tin
g
r
ele
v
an
t
f
ea
tu
r
es
f
r
o
m
m
ed
ical
im
ag
es,
in
clu
d
in
g
m
am
m
o
g
r
a
m
s
.
B
y
lev
er
ag
in
g
DL
,
r
esear
ch
er
s
an
d
clin
ician
s
ca
n
p
o
ten
tially
o
v
er
co
m
e
th
e
l
im
itatio
n
s
o
f
co
n
v
en
tio
n
al
d
iag
n
o
s
tic
m
eth
o
d
s
.
DL
m
o
d
els
tr
ain
ed
o
n
lar
g
e
d
atasets
o
f
m
am
m
o
g
r
a
m
im
a
g
es
ca
n
lear
n
to
r
ec
o
g
n
ize
s
u
b
tle
p
atter
n
s
in
d
icativ
e
o
f
b
r
ea
s
t
ca
n
ce
r
,
en
a
b
lin
g
ea
r
lie
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
7
,
No
.
3
,
Ma
r
ch
20
2
5
:
1
92
7
-
1
93
5
1928
d
etec
tio
n
an
d
in
te
r
v
en
tio
n
.
Mo
r
eo
v
er
,
DL
f
ac
ilit
ates
th
e
d
ev
elo
p
m
e
n
t
o
f
a
u
to
m
ated
s
y
s
tem
s
ca
p
ab
le
o
f
tr
iag
in
g
im
a
g
es,
p
r
i
o
r
itizin
g
ca
s
es
f
o
r
f
u
r
th
er
r
ev
iew,
a
n
d
ass
is
tin
g
r
ad
io
lo
g
is
ts
in
in
ter
p
r
etin
g
c
o
m
p
le
x
f
in
d
in
g
s
,
th
e
r
eb
y
im
p
r
o
v
in
g
w
o
r
k
f
lo
w
e
f
f
icien
cy
a
n
d
d
iag
n
o
s
tic
ac
cu
r
ac
y
.
Desp
ite
th
e
p
r
o
m
is
in
g
p
r
o
s
p
e
cts
o
f
f
er
ed
b
y
DL
,
its
ef
f
ec
tiv
e
ap
p
licatio
n
in
b
r
ea
s
t
ca
n
ce
r
d
iag
n
o
s
is
f
ac
es
s
ev
er
al
ch
allen
g
es.
On
e
s
ig
n
if
ican
t
h
u
r
d
le
is
th
e
r
eq
u
ir
em
en
t
f
o
r
lar
g
e
an
n
o
tated
d
atasets
,
wh
ich
ar
e
o
f
ten
s
ca
r
ce
an
d
co
s
tly
to
o
b
t
ain
in
m
ed
ical
im
ag
in
g
d
o
m
a
in
s
.
Ad
d
itio
n
ally
,
DL
m
o
d
els
m
u
s
t
d
em
o
n
s
tr
ate
r
o
b
u
s
tn
ess
,
g
en
er
aliza
b
ilit
y
,
a
n
d
in
ter
p
r
etab
ilit
y
to
g
ain
ac
c
ep
tan
ce
an
d
tr
u
s
t
am
o
n
g
h
ea
lth
ca
r
e
p
r
o
f
ess
io
n
als
an
d
r
eg
u
lato
r
y
b
o
d
ies.
I
n
r
esp
o
n
s
e
to
th
ese
ch
allen
g
es,
r
esear
ch
er
s
ar
e
e
x
p
lo
r
i
n
g
in
n
o
v
ativ
e
ap
p
r
o
ac
h
es
to
h
ar
n
ess
th
e
p
o
ten
tial
o
f
D
L
in
b
r
ea
s
t
ca
n
ce
r
d
iag
n
o
s
is
.
T
r
an
s
f
er
lear
n
i
n
g
(
T
L
)
h
as
em
er
g
ed
as
a
v
alu
ab
le
tech
n
iq
u
e
to
ad
d
r
ess
th
e
d
ata
s
ca
r
city
is
s
u
e
b
y
tr
an
s
f
er
r
in
g
k
n
o
wled
g
e
lear
n
ed
f
r
o
m
p
r
e
-
tr
ai
n
ed
m
o
d
els
o
n
lar
g
e
n
o
n
-
m
ed
ical
d
atasets
to
s
p
ec
if
ic
m
ed
ical
im
ag
in
g
task
s
.
T
L
en
ab
le
s
DL
m
o
d
els
to
lev
er
ag
e
f
ea
tu
r
es
lear
n
ed
f
r
o
m
d
iv
er
s
e
d
atasets
,
en
h
an
cin
g
th
eir
p
er
f
o
r
m
an
ce
o
n
task
s
with
lim
ited
d
ata
av
ailab
ilit
y
.
I
n
ad
d
itio
n
to
DL
tech
n
iq
u
es,
v
ar
io
u
s
m
ac
h
in
e
lear
n
in
g
(
ML
)
m
eth
o
d
o
lo
g
ies
h
av
e
b
ee
n
in
v
esti
g
ated
f
o
r
ca
n
ce
r
d
etec
tio
n
,
p
ar
t
ic
u
lar
ly
in
b
r
ea
s
t
ca
n
ce
r
d
iag
n
o
s
is
[
4
]
-
[
6
]
.
T
h
ese
m
eth
o
d
s
en
c
o
m
p
ass
a
s
p
ec
tr
u
m
f
r
o
m
co
n
v
en
tio
n
al
ML
alg
o
r
ith
m
s
to
s
o
p
h
is
ticated
DL
a
r
ch
itectu
r
es.
C
o
n
v
en
tio
n
al
M
L
tech
n
iq
u
es
lik
e
s
u
p
p
o
r
t
v
ec
t
o
r
m
ac
h
i
n
es
(
SVM)
,
r
an
d
o
m
f
o
r
ests
,
an
d
lo
g
is
tic
r
eg
r
ess
io
n
h
av
e
f
o
u
n
d
ex
te
n
s
iv
e
ap
p
licatio
n
in
task
s
s
u
ch
as
class
if
icatio
n
an
d
p
atter
n
r
e
co
g
n
itio
n
with
in
m
ed
ical
im
ag
in
g
d
atasets
[
7
]
,
[
8
]
.
Ho
wev
er
,
wh
ile
th
ese
ML
tech
n
iq
u
es
h
av
e
s
h
o
wn
p
r
o
m
is
e,
th
ey
o
f
ten
r
eq
u
ir
e
h
an
d
c
r
af
ted
f
ea
tu
r
e
e
x
tr
ac
tio
n
an
d
s
elec
tio
n
,
wh
ich
ca
n
b
e
tim
e
-
c
o
n
s
u
m
in
g
an
d
m
ay
n
o
t
f
u
lly
ca
p
tu
r
e
th
e
co
m
p
lex
ity
o
f
m
ed
ical
im
a
g
e
s
.
Fu
r
th
er
m
o
r
e,
th
e
p
er
f
o
r
m
an
ce
o
f
tr
a
d
itio
n
al
M
L
alg
o
r
ith
m
s
is
h
ig
h
ly
d
ep
en
d
en
t
o
n
th
e
q
u
ality
an
d
r
ele
v
an
ce
o
f
th
e
f
ea
t
u
r
es
p
r
o
v
id
e
d
as
in
p
u
t.
C
o
n
v
er
s
ely
,
DL
m
eth
o
d
o
lo
g
ies,
n
o
tab
l
y
C
NNs,
h
av
e
g
ar
n
er
ed
co
n
s
id
er
ab
le
atten
tio
n
in
r
ec
en
t
tim
es
d
u
e
to
th
eir
ca
p
ac
ity
to
au
to
n
o
m
o
u
s
ly
ac
q
u
ir
e
h
ier
ar
c
h
ical
r
ep
r
esen
ta
tio
n
s
d
ir
ec
tly
f
r
o
m
u
n
p
r
o
ce
s
s
ed
d
ata,
s
u
ch
as
m
a
m
m
o
g
r
am
im
ag
es.
C
NNs
ca
n
ef
f
ec
tiv
ely
ca
p
t
u
r
e
in
tr
icate
p
atter
n
s
an
d
f
ea
t
u
r
es
at
d
if
f
er
en
t
lev
els
o
f
ab
s
tr
ac
tio
n
,
m
ak
in
g
th
em
well
-
s
u
ited
f
o
r
m
ed
ical
im
ag
e
an
aly
s
is
tas
k
s
,
in
clu
d
in
g
ca
n
ce
r
d
etec
tio
n
.
T
L
em
er
g
es
as
a
p
o
wer
f
u
l
ap
p
r
o
ac
h
to
en
h
an
c
e
th
e
p
er
f
o
r
m
a
n
ce
o
f
DL
m
o
d
els,
esp
ec
ially
in
s
ce
n
ar
io
s
wh
er
e
lab
eled
d
ata
i
s
lim
ited
[
9
]
,
[
1
0
]
.
T
L
lev
e
r
ag
es
k
n
o
wled
g
e
lear
n
ed
f
r
o
m
p
r
e
-
tr
ain
ed
m
o
d
els o
n
lar
g
e
d
atasets
(
e.
g
.
,
I
m
ag
eNe
t
)
an
d
a
p
p
lies
it
to
r
elate
d
task
s
with
s
m
aller
d
atasets
,
s
u
c
h
as
m
ed
ical
im
ag
e
an
aly
s
is
.
B
y
in
itializin
g
DL
m
o
d
els
with
weig
h
ts
lear
n
ed
f
r
o
m
p
r
e
-
t
r
ain
ed
n
etwo
r
k
s
,
T
L
en
ab
les
th
e
tr
an
s
f
er
o
f
k
n
o
wled
g
e
a
b
o
u
t
l
o
w
-
lev
el
f
ea
tu
r
es,
ed
g
e
d
etec
tio
n
,
an
d
tex
tu
r
e
r
ec
o
g
n
itio
n
,
wh
ich
a
r
e
o
f
ten
tr
a
n
s
f
er
r
ab
le
ac
r
o
s
s
d
o
m
ain
s
.
I
n
th
e
c
o
n
tex
t
o
f
b
r
ea
s
t
ca
n
ce
r
d
etec
tio
n
,
T
L
allo
ws
DL
m
o
d
el
s
to
lev
er
a
g
e
th
e
r
ep
r
esen
tatio
n
s
lear
n
ed
f
r
o
m
n
o
n
-
m
e
d
ical
im
ag
e
d
atasets
to
im
p
r
o
v
e
th
e
p
er
f
o
r
m
a
n
ce
o
f
ca
n
ce
r
d
etec
tio
n
m
o
d
els.
T
h
is
ap
p
r
o
ac
h
ad
d
r
es
s
es
th
e
ch
allen
g
e
o
f
d
ata
s
ca
r
city
in
m
ed
ical
im
ag
in
g
b
y
p
r
o
v
id
in
g
a
m
ea
n
s
to
tr
an
s
f
er
k
n
o
wled
g
e
f
r
o
m
d
o
m
ain
s
with
ab
u
n
d
an
t
d
ata
to
d
o
m
ain
s
with
lim
ite
d
d
ata
a
v
ailab
ilit
y
.
B
y
f
in
e
-
tu
n
in
g
p
r
e
-
t
r
ain
ed
C
NN
ar
ch
it
ec
tu
r
es
o
n
s
p
ec
if
ic
m
ed
ical
i
m
ag
in
g
d
atasets
,
T
L
en
ab
les
DL
m
o
d
els
t
o
a
d
ap
t
to
th
e
in
tr
icac
ies
o
f
b
r
ea
s
t
ca
n
ce
r
d
etec
tio
n
task
s
wh
ile
b
en
ef
itin
g
f
r
o
m
th
e
g
e
n
er
aliza
tio
n
ca
p
ab
ilit
ies
lear
n
ed
f
r
o
m
d
i
v
er
s
e
d
atasets
.
As
a
r
esu
lt,
T
L
s
er
v
es
as
a
v
alu
ab
le
to
o
l
f
o
r
im
p
r
o
v
in
g
t
h
e
ac
cu
r
a
cy
,
ef
f
icien
c
y
,
an
d
r
o
b
u
s
tn
ess
o
f
DL
-
b
ased
ca
n
c
er
d
etec
tio
n
s
y
s
tem
s
,
u
ltima
tely
ad
v
an
cin
g
th
e
f
ield
o
f
m
ed
ical
im
ag
e
an
aly
s
is
an
d
co
n
tr
ib
u
tin
g
to
im
p
r
o
v
e
d
p
atien
t o
u
tco
m
es.
Th
is
s
tu
d
y
en
d
ea
v
o
r
s
to
co
n
s
tr
u
ct
a
r
o
b
u
s
t
DL
f
r
am
ewo
r
k
f
o
r
th
e
au
to
m
atic
d
etec
tio
n
an
d
d
iag
n
o
s
is
o
f
b
r
ea
s
t
ca
n
ce
r
th
r
o
u
g
h
th
e
u
tili
za
tio
n
o
f
T
L
m
eth
o
d
o
lo
g
ie
s
.
Ou
r
s
tr
ateg
y
e
n
tails
h
ar
n
ess
in
g
th
e
ca
p
ab
ilit
ies
o
f
p
r
e
-
tr
ain
e
d
C
NN
ar
ch
itec
tu
r
es,
en
co
m
p
ass
in
g
r
e
n
o
wn
ed
m
o
d
els
s
u
ch
as
Alex
Net,
R
esNet5
0
,
v
is
u
al
g
eo
m
etr
y
g
r
o
u
p
(
VGG
)
-
1
6
,
a
n
d
VGG
-
1
9
[
1
1
]
.
T
h
e
u
s
e
o
f
TL
in
th
is
r
esear
ch
en
a
b
les
u
s
to
o
v
er
co
m
e
d
ata
s
ca
r
city
,
lev
er
ag
e
p
r
e
-
lear
n
e
d
f
ea
tu
r
es,
im
p
r
o
v
e
g
e
n
er
ali
za
tio
n
,
an
d
en
h
an
ce
co
m
p
u
t
atio
n
al
ef
f
icien
cy
.
T
h
e
s
e
ad
v
an
tag
es
co
llectiv
ely
co
n
tr
ib
u
te
to
th
e
d
ev
elo
p
m
e
n
t
o
f
m
o
r
e
ac
c
u
r
ate,
ef
f
icien
t
,
an
d
s
ca
lab
le
DL
-
b
ased
s
o
lu
tio
n
s
f
o
r
b
r
ea
s
t
ca
n
ce
r
d
etec
tio
n
,
u
ltima
tely
b
e
n
ef
itin
g
p
atien
ts
,
clin
ician
s
,
an
d
h
ea
lth
ca
r
e
s
y
s
tem
s
.
B
y
f
in
e
-
tu
n
in
g
th
ese
p
r
e
-
t
r
ain
ed
m
o
d
els
o
n
a
s
p
ec
if
ic
m
e
d
ical
im
ag
in
g
d
ataset,
s
u
ch
a
s
th
e
b
r
ea
s
t
ca
n
ce
r
w
is
co
n
s
in
(
B
C
W
)
d
iag
n
o
s
tic
d
ataset,
we
s
ee
k
to
i
m
p
r
o
v
e
th
e
ac
c
u
r
ac
y
,
ef
f
icien
c
y
,
an
d
g
en
er
aliza
tio
n
ca
p
ab
ilit
ies
o
f
o
u
r
DL
f
r
am
e
wo
r
k
f
o
r
b
r
ea
s
t
ca
n
ce
r
d
etec
ti
o
n
[
1
2
]
.
T
h
e
p
r
im
a
r
y
o
b
jectiv
e
o
f
th
is
s
tu
d
y
is
to
ev
alu
ate
th
e
ef
f
icac
y
o
f
T
L
in
en
h
an
cin
g
b
r
ea
s
t
ca
n
ce
r
d
iag
n
o
s
is
b
y
ad
d
r
ess
in
g
th
e
ch
allen
g
es
o
f
d
ata
s
ca
r
city
,
f
ea
tu
r
e
lear
n
in
g
,
g
en
er
aliza
tio
n
,
an
d
co
m
p
u
tatio
n
al
ef
f
icien
cy
.
T
h
r
o
u
g
h
c
o
m
p
r
e
h
en
s
iv
e
ex
p
er
im
en
tatio
n
a
n
d
ev
alu
at
io
n
u
s
in
g
esta
b
lis
h
ed
p
er
f
o
r
m
an
ce
m
etr
ics
s
u
ch
as
ac
c
u
r
ac
y
,
s
en
s
itiv
ity
,
s
p
ec
if
icity
,
p
r
ec
is
io
n
,
a
n
d
F
-
s
co
r
e,
we
aim
to
d
em
o
n
s
tr
ate
th
e
s
u
p
er
io
r
ity
o
f
o
u
r
p
r
o
p
o
s
ed
DL
f
r
am
ewo
r
k
o
v
er
tr
ad
itio
n
al
m
eth
o
d
s
in
b
r
ea
s
t
ca
n
ce
r
d
etec
tio
n
.
T
h
r
o
u
g
h
ad
v
a
n
cin
g
th
e
f
r
o
n
tier
s
o
f
DL
an
d
T
L
with
in
b
r
ea
s
t
ca
n
ce
r
d
etec
tio
n
,
th
is
in
v
esti
g
atio
n
s
ee
k
s
to
m
a
k
e
s
ig
n
if
ican
t
co
n
tr
ib
u
tio
n
s
to
th
e
c
o
n
tin
u
o
u
s
e
n
d
ea
v
o
r
s
d
ir
ec
ted
to
war
d
s
en
h
an
cin
g
ea
r
ly
d
ia
g
n
o
s
is
,
tr
ea
tm
en
t
ef
f
icac
y
,
a
n
d
p
atien
t
s
u
r
v
iv
al
r
ates.
Ultim
ately
,
th
e
r
ef
in
em
e
n
t
o
f
m
o
r
e
p
r
ec
i
s
e
an
d
ef
f
ec
tiv
e
DL
-
d
r
iv
e
n
s
o
lu
tio
n
s
f
o
r
b
r
ea
s
t
ca
n
ce
r
d
et
ec
tio
n
h
ar
b
o
r
s
th
e
p
o
ten
tial
to
tr
an
s
f
o
r
m
h
ea
lth
ca
r
e
d
eliv
er
y
p
r
o
f
o
u
n
d
ly
an
d
b
r
in
g
a
b
o
u
t
p
o
s
itiv
e
ch
an
g
es
in
th
e
liv
es
o
f
co
u
n
tles
s
in
d
iv
id
u
als af
f
licted
b
y
th
is
ailm
en
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
mp
o
w
er B
r
ea
s
tN
e
t:
b
r
ea
s
t c
a
n
ce
r
d
etec
tio
n
w
ith
tr
a
n
s
fer lea
r
n
in
g
…
(
V
a
is
h
a
li M.
Jo
s
h
i
)
1929
I
n
o
r
d
er
to
en
h
a
n
ce
th
e
id
en
tific
atio
n
,
lo
ca
lizatio
n
,
r
is
k
a
s
s
es
s
m
en
t,
an
d
ca
teg
o
r
izatio
n
o
f
b
r
ea
s
t
lesi
o
n
s
,
Ma
h
m
o
o
d
et
a
l.
[
1
3
]
h
av
e
cr
ea
ted
s
o
p
h
is
ticated
DL
alg
o
r
ith
m
s
,
with
a
f
o
cu
s
o
n
r
ed
u
cin
g
f
alse
p
o
s
itiv
es
an
d
r
eso
lv
in
g
p
r
o
b
lem
s
ass
o
ciate
d
with
s
lo
w
c
o
n
v
er
g
en
ce
r
ates.
T
h
ey
u
s
e
a
co
m
b
in
atio
n
o
f
ad
v
an
ce
d
f
ilter
in
g
tech
n
iq
u
es
,
p
r
ep
r
o
ce
s
s
in
g
ap
p
r
o
ac
h
es,
a
n
d
d
ata
au
g
m
en
tatio
n
tactics
to
im
p
r
o
v
e
m
o
d
el
p
er
f
o
r
m
an
ce
an
d
p
r
e
v
en
t
b
o
th
u
n
d
er
f
itti
n
g
an
d
o
v
er
f
itti
n
g
.
On
e
s
ig
n
if
ican
t
b
r
ea
k
t
h
r
o
u
g
h
in
th
eir
r
esear
ch
is
th
e
ef
f
icien
t
d
etec
tio
n
o
f
d
en
s
e
b
r
ea
s
t
lesi
o
n
s
with
th
e
ap
p
licatio
n
o
f
c
h
ao
tic
lead
er
s
e
lectiv
e
f
iller
s
war
m
o
p
tim
izatio
n
(
cL
SF
SO)
.
Fu
r
th
er
m
o
r
e,
t
h
ey
e
n
h
an
ce
t
h
e
ca
p
ac
ity
o
f
DL
m
o
d
els
s
u
ch
m
o
d
if
ied
VGGN
et
an
d
SE
-
R
esNet1
5
2
to
d
is
tin
g
u
is
h
b
etwe
en
n
o
r
m
al
an
d
q
u
esti
o
n
ab
le
r
eg
io
n
s
in
m
am
m
o
g
r
a
m
s
b
y
u
tili
zin
g
TL
.
Fu
r
th
er
im
p
r
o
v
in
g
t
h
e
an
aly
s
i
s
,
th
e
s
tu
d
y
s
u
g
g
ests
h
y
b
r
id
d
ee
p
n
eu
r
al
n
etwo
r
k
m
eth
o
d
s
f
o
r
id
en
tify
i
n
g
an
d
class
if
y
in
g
m
alig
n
an
t
p
o
ly
p
s
u
s
in
g
p
r
e
-
s
eg
m
e
n
t
ed
r
eg
io
n
s
o
f
in
ter
est
(
R
OI
s
)
,
s
u
ch
as
C
NN+
L
STM
an
d
C
NN+
SV
M.
B
r
ea
s
t
ab
n
o
r
m
alities
ca
n
n
o
w
b
e
d
iag
n
o
s
ed
m
o
r
e
ac
cu
r
ately
th
an
k
s
to
th
e
ap
p
licatio
n
o
f
g
r
a
d
-
C
AM
tech
n
iq
u
es.
An
aly
s
es
o
n
p
u
b
lic
an
d
p
r
iv
ate
d
at
asets
d
em
o
n
s
tr
ate
r
em
ar
k
a
b
le
im
p
r
o
v
em
en
ts
in
m
am
m
o
g
r
a
p
h
y
an
al
y
s
is
,
with
r
em
ar
k
ab
le
s
en
s
itiv
ity
(
0
.
9
9
)
an
d
an
o
v
er
all
ar
ea
u
n
d
er
th
e
cu
r
v
e
(
AUC)
o
f
0
.
9
9
.
W
ith
m
u
ltip
le
n
o
tewo
r
th
y
a
d
v
an
ce
s
,
Sah
u
et
a
l.
[
1
4
]
h
a
v
e
p
r
o
p
o
s
ed
an
i
n
v
en
tiv
e
DL
b
ased
en
s
em
b
le
class
if
ier
s
p
ec
if
ically
d
esig
n
ed
f
o
r
b
r
ea
s
t
ca
n
ce
r
d
ia
g
n
o
s
is
.
R
e
s
u
lts
f
r
o
m
ex
p
er
im
en
ts
s
h
o
w
th
at
th
eir
s
u
g
g
ested
s
tr
ateg
y
p
r
o
d
u
ce
s
r
em
ar
k
ab
le
ca
teg
o
r
izatio
n
o
u
tco
m
es.
On
th
e
m
in
i
-
DDSM
an
d
u
ltra
s
o
u
n
d
d
ataset
(
B
USI
)
d
ataset,
i
t
s
p
ec
if
icall
y
ac
h
iev
es
an
ac
cu
r
ac
y
o
f
9
9
.
1
7
%
f
o
r
ab
n
o
r
m
ality
d
etec
ti
o
n
an
d
9
7
.
7
5
%
f
o
r
m
al
ig
n
an
cy
d
etec
tio
n
an
d
9
6
.
9
2
%,
an
d
9
4
.
6
2
%
ac
cu
r
ac
y
f
o
r
ab
n
o
r
m
ality
an
d
m
alig
n
an
cy
d
iag
n
o
s
is
,
r
esp
ec
tiv
ely
.
Fu
r
th
e
r
m
o
r
e,
it
ac
h
iev
es
9
7
.
5
0
%
ac
cu
r
ac
y
o
n
th
e
B
US2
u
ltra
s
o
u
n
d
d
ata
s
et.
T
h
e
s
u
g
g
ested
ap
p
r
o
ac
h
s
h
o
ws
p
o
ten
tial
f
o
r
b
r
ea
s
t
ca
n
ce
r
d
iag
n
o
s
is
ac
r
o
s
s
m
u
l
tim
o
d
al
d
atasets
d
u
e
to
its
ad
ap
tab
ilit
y
an
d
d
ep
en
d
a
b
ilit
y
.
Ab
u
n
ass
er
et
a
l.
[
1
5
]
p
r
o
p
o
s
ed
a
s
tr
ateg
y
f
o
r
class
if
y
in
g
b
r
ea
s
t
ca
n
ce
r
MRI
p
ictu
r
es
in
to
eig
h
t
g
r
o
u
p
s
.
T
h
eir
ap
p
r
o
ac
h
co
m
b
i
n
es
a
u
n
iq
u
e
DL
m
o
d
el
with
f
iv
e
f
in
e
-
tu
n
ed
m
o
d
els,
all
o
f
wh
ich
wer
e
tr
a
in
e
d
o
n
th
e
I
m
a
g
eNe
t
d
ata
b
ase.
T
h
ey
u
s
ed
a
g
en
e
r
ativ
e
a
d
v
er
s
ar
ial
n
etwo
r
k
(
GAN)
t
o
e
n
h
an
ce
th
e
d
ataset.
T
h
e
au
th
o
r
s
m
et
h
o
d
ically
an
aly
ze
d
th
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
eir
p
r
o
p
o
s
ed
DL
m
o
d
el
a
n
d
th
e
f
iv
e
p
r
e
-
tr
ain
ed
m
o
d
els
o
n
ea
ch
d
ataset
s
ep
ar
ately
.
T
h
e
class
if
ica
tio
n
ac
c
u
r
ac
ies
f
o
r
t
h
eir
p
r
o
p
o
s
ed
m
o
d
el
(
B
C
C
NN)
wer
e
9
8
.
2
8
%.
Fo
r
th
e
id
en
tific
atio
n
an
d
cl
ass
if
icatio
n
o
f
b
r
ea
s
t
ca
n
ce
r
,
R
az
a
et
a
l.
[
1
6
]
p
r
esen
ted
d
ee
p
b
r
ea
s
t
C
an
ce
r
Net,
an
in
n
o
v
ativ
e
DL
m
o
d
el.
As
its
two
n
o
r
m
aliza
tio
n
p
r
o
ce
s
s
es,
th
e
m
o
d
el
co
m
b
in
es
th
e
c
lip
p
ed
r
ec
tifie
d
lin
ea
r
u
n
it
(
R
eL
U
)
ac
tiv
atio
n
f
u
n
ctio
n
,
th
e
leak
y
R
eL
U
ac
tiv
atio
n
f
u
n
ctio
n
,
b
atch
n
o
r
m
aliza
tio
n
,
an
d
cr
o
s
s
-
ch
an
n
el
n
o
r
m
aliza
tio
n
.
T
h
eir
test
in
g
r
ev
ea
led
th
at
th
e
s
u
g
g
ested
m
o
d
el
h
as
an
asto
u
n
d
in
g
9
9
.
3
5
%
class
if
icatio
n
ac
cu
r
ac
y
.
T
wo
ex
tr
em
ely
s
u
cc
ess
f
u
l
d
ee
p
TL
b
ased
m
o
d
els
ar
e
p
r
esen
ted
b
y
Yar
i
et
a
l.
[
1
7
]
in
an
ef
f
o
r
t
to
en
h
an
ce
th
e
s
tate
-
of
-
th
e
-
a
r
t
m
eth
o
d
s
th
at
ar
e
c
u
r
r
en
tly
i
n
u
s
e
f
o
r
th
e
d
etec
tio
n
o
f
b
r
ea
s
t
ca
n
ce
r
in
b
o
th
b
i
n
ar
y
a
n
d
m
u
lticlas
s
cl
ass
if
icatio
n
.
T
h
ese
m
o
d
els
m
ak
e
u
s
e
o
f
d
ee
p
c
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
s
(
DC
NN)
th
at
h
av
e
alr
ea
d
y
b
ee
n
tr
ain
ed
u
s
in
g
a
s
izab
le
im
ag
e
d
ataset
f
r
o
m
th
e
I
m
ag
eNe
t
d
ataset.
R
eg
ar
d
in
g
m
u
lticlas
s
clas
s
if
icatio
n
task
s
,
th
e
ac
cu
r
ac
y
o
f
t
h
e
p
r
o
p
o
s
ed
s
y
s
tem
is
im
p
r
ess
iv
e.
T
h
e
cu
r
r
en
t
s
tatu
s
o
f
D
L
r
esear
ch
in
b
r
ea
s
t
ca
n
ce
r
im
ag
in
g
is
th
o
r
o
u
g
h
ly
r
ev
iewe
d
b
y
B
alk
en
en
d
e
et
a
l.
[
1
8
]
.
T
h
ey
em
p
h
asize
h
o
w
im
p
o
r
tan
t b
r
ea
s
t im
ag
in
g
is
f
o
r
ea
r
ly
b
r
ea
s
t c
an
ce
r
d
etec
tio
n
,
m
o
n
ito
r
in
g
,
an
d
ass
ess
m
en
t d
u
r
in
g
th
e
r
ap
y
.
T
h
e
in
teg
r
atio
n
o
f
DL
,
a
s
u
b
s
et
o
f
AI
,
b
ec
o
m
e
s
p
o
s
s
ib
le
with
th
e
au
to
m
atio
n
o
f
v
ar
io
u
s
im
ag
in
g
tech
n
iq
u
es.
I
n
b
r
ea
s
t
im
ag
in
g
,
DL
is
b
ein
g
u
s
ed
f
o
r
a
v
ar
iety
o
f
task
s
,
in
clu
d
in
g
ca
n
ce
r
r
is
k
p
r
ed
ictio
n
,
th
er
ap
eu
tic
r
esp
o
n
s
e
p
r
e
d
ictio
n
an
d
ev
alu
atio
n
,
lesi
o
n
class
if
icatio
n
an
d
s
eg
m
en
t
atio
n
,
an
d
p
ictu
r
e
r
ec
o
n
s
tr
u
ctio
n
an
d
g
e
n
er
atio
n
.
Stu
d
ies
s
h
o
w
t
h
at
DL
alg
o
r
it
h
m
s
p
er
f
o
r
m
as
well
as
o
r
b
etter
th
an
r
a
d
io
lo
g
is
ts
o
n
a
f
ew
task
s
.
Ho
wev
er
,
lar
g
er
s
tu
d
ies
ar
e
r
eq
u
ir
ed
to
p
r
o
p
er
ly
d
eter
m
i
n
e
th
e
ad
d
ed
v
alu
e
o
f
DL
in
b
r
ea
s
t
ca
n
ce
r
im
ag
in
g
,
esp
ec
ially
in
u
ltra
s
o
u
n
d
a
n
d
MRI.
T
h
e
liter
atu
r
e
r
e
v
iew
h
ig
h
lig
h
ts
th
e
v
ital
r
o
le
th
at
TL
p
lay
s
in
im
p
r
o
v
in
g
b
r
ea
s
t
ca
n
ce
r
d
e
tectio
n
b
y
lev
er
ag
in
g
DL
m
o
d
els
’
en
h
a
n
ce
d
ac
cu
r
ac
y
,
ef
f
icien
cy
,
an
d
g
en
er
aliza
tio
n
ca
p
ac
ities
.
TL
,
in
th
e
f
ig
h
t
ag
ain
s
t
b
r
ea
s
t
ca
n
ce
r
,
p
r
o
v
id
es
a
r
ea
lis
ti
c
tech
n
iq
u
e
to
im
p
r
o
v
e
p
atien
t
ca
r
e
b
y
lev
er
ag
in
g
k
n
o
wl
ed
g
e
lear
n
ed
f
r
o
m
p
r
e
-
tr
ain
ed
m
o
d
els to
im
p
r
o
v
e
d
iag
n
o
s
tic
o
u
tc
o
m
es.
2.
M
E
T
H
O
D
I
n
t
h
i
s
p
a
p
e
r
,
w
e
e
x
p
l
o
r
e
d
v
a
r
i
o
u
s
c
l
as
s
i
f
i
c
a
ti
o
n
m
o
d
e
l
s
f
o
r
t
h
e
d
i
a
g
n
o
s
i
s
o
f
b
r
e
as
t
c
a
n
c
e
r
,
i
n
c
l
u
d
i
n
g
A
l
e
x
N
e
t
,
R
e
s
N
e
t
5
0
,
VG
G
-
1
6
,
a
n
d
V
GG
-
1
9
.
O
u
r
p
r
o
p
o
s
e
d
m
e
t
h
o
d
o
l
o
g
y
i
s
il
l
u
s
t
r
a
t
e
d
i
n
F
i
g
u
r
e
1
.
T
h
is
f
i
g
u
r
e
d
e
p
i
c
t
s
t
h
e
p
r
o
c
e
s
s
s
t
a
r
ti
n
g
f
r
o
m
d
a
t
a
a
c
q
u
is
i
ti
o
n
f
r
o
m
t
h
e
d
a
t
a
s
et
.
T
h
e
d
a
ta
p
r
e
p
a
r
a
t
i
o
n
s
te
p
i
n
v
o
l
v
e
s
l
o
a
d
i
n
g
t
h
e
b
r
e
a
s
t
c
a
n
c
e
r
d
at
a
s
e
t
a
n
d
en
s
u
r
i
n
g
t
h
e
f
e
a
t
u
r
es
a
r
e
n
o
r
m
a
li
z
e
d
o
r
s
t
a
n
d
a
r
d
i
z
e
d
t
o
m
a
i
n
t
a
in
u
n
i
f
o
r
m
i
n
f
l
u
e
n
c
e
a
c
r
o
s
s
t
h
e
a
l
g
o
r
it
h
m
s
.
N
e
x
t
,
a
t
r
a
i
n
i
n
g
s
e
t
a
n
d
a
t
es
t
s
et
a
r
e
c
r
e
a
t
e
d
f
r
o
m
t
h
e
d
a
t
a
s
e
t
,
u
s
u
al
ly
w
i
t
h
a
7
0
-
3
0
s
p
l
i
t
f
o
r
e
a
c
h
.
Us
i
n
g
p
r
e
d
e
te
r
m
i
n
ed
c
r
i
t
e
r
ia
,
w
e
e
v
a
l
u
at
e
d
e
ac
h
m
o
d
e
l
’
s
p
e
r
f
o
r
m
a
n
c
e
u
s
i
n
g
A
l
e
x
N
e
t
,
R
e
s
N
e
t
5
0
,
VGG
-
1
6
,
a
n
d
V
GG
-
1
9
f
o
r
t
r
a
in
i
n
g
.
T
h
e
b
e
s
t
a
l
g
o
r
it
h
m
f
o
r
i
d
e
n
t
i
f
y
i
n
g
b
r
e
a
s
t
c
a
n
c
e
r
w
as
t
o
b
e
f
o
u
n
d
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
7
,
No
.
3
,
Ma
r
ch
20
2
5
:
1
92
7
-
1
93
5
1930
Fig
u
r
e
1
.
B
lo
ck
d
iag
r
am
o
f
b
r
ea
s
t c
an
ce
r
d
etec
tio
n
2.
1
.
Da
t
a
s
et
T
h
e
w
is
co
n
s
in
b
r
ea
s
t
ca
n
ce
r
d
ataset
(
d
iag
n
o
s
is
)
,
wh
ich
we
d
o
wn
lo
ad
e
d
f
r
o
m
th
e
Kag
g
le
ML
an
d
d
ata
s
cien
ce
co
m
m
u
n
ity
web
s
ite,
h
ttp
s
://ar
ch
iv
e.
ics.u
ci.
ed
u
/d
ataset/1
7
/b
r
ea
s
t+c
an
ce
r
+w
is
co
n
s
in
+
d
iag
n
o
s
tic
[
1
9
]
was
u
s
ed
in
th
is
s
tu
d
y
.
T
h
er
e
ar
e
5
6
9
in
s
tan
ce
s
in
th
i
s
co
llectio
n
,
an
d
ea
ch
i
n
s
tan
ce
h
as
3
0
p
r
o
p
er
ties
.
3
5
7
o
f
th
em
a
r
e
b
e
n
ig
n
,
an
d
2
1
2
ar
e
m
alig
n
a
n
t.
B
elo
w
a
r
e
th
e
f
ea
tu
r
es
o
f
th
e
d
ataset
.
T
en
r
ea
l
-
v
alu
e
d
ch
ar
ac
ter
is
tics
ar
e
u
s
ed
to
ev
al
u
ate
ea
ch
ce
ll n
u
cle
u
s
:
−
R
ad
iu
s
(
m
ea
n
o
f
d
is
tan
ce
s
f
r
o
m
ce
n
ter
to
p
o
in
ts
o
n
th
e
p
er
i
m
eter
)
.
−
T
ex
tu
r
e
(
s
tan
d
a
r
d
d
e
v
iatio
n
o
f
g
r
ay
-
s
ca
le
v
alu
es)
.
−
P
er
im
eter
.
−
A
r
ea
.
−
S
m
o
o
th
n
ess
(
lo
ca
l v
ar
iatio
n
in
r
ad
iu
s
len
g
th
s
)
.
−
C
o
m
p
ac
tn
ess
(
p
er
im
eter
^2
/ a
r
ea
-
1
.
0
)
.
−
C
o
n
ca
v
ity
(
s
ev
er
ity
o
f
co
n
ca
v
e
p
o
r
tio
n
s
o
f
th
e
c
o
n
to
u
r
)
.
−
C
o
n
ca
v
e
p
o
in
ts
(
n
u
m
b
e
r
o
f
co
n
ca
v
e
p
o
r
tio
n
s
o
f
th
e
co
n
to
u
r
)
.
−
S
y
m
m
etr
y
.
−
F
r
ac
tal
d
im
en
s
io
n
(
“
co
astl
in
e
ap
p
r
o
x
im
atio
n
”
-
1)
.
2.
2
.
Da
t
a
pre
-
pro
ce
s
s
ing
T
o
en
s
u
r
e
p
r
o
p
e
r
r
elev
a
n
ce
ass
ig
n
m
en
t [
2
0
]
,
th
e
d
ataset
u
n
d
er
wen
t stan
d
ar
d
izatio
n
u
s
in
g
i
n
th
e
(
1
)
.
−
−
(
1
)
T
h
e
s
tan
d
ar
d
d
ev
iatio
n
o
f
th
e
n
o
r
m
aliza
tio
n
f
ea
tu
r
e
is
g
iv
e
n
b
y
an
d
th
e
m
ea
n
b
y
.
Pre
p
r
o
ce
s
s
in
g
aim
ed
to
r
ep
la
ce
d
u
p
licate
v
alu
es
a
n
d
b
ala
n
ce
th
e
d
ataset
u
s
in
g
a
d
ata
b
alan
ce
r
,
f
o
llo
wed
b
y
f
ea
tu
r
e
ex
tr
ac
tio
n
.
T
h
e
p
r
o
ce
s
s
ed
d
ata
was
th
en
p
ar
titi
o
n
ed
in
to
tr
ain
in
g
a
n
d
t
esti
n
g
s
ets,
with
a
70
-
3
0
s
p
lit.
Sev
er
al
ML
clas
s
if
ier
s
,
in
clu
d
in
g
Ale
x
Net,
R
esNet5
0
,
VGG
-
1
6
,
an
d
VGG
-
1
9
,
wer
e
u
s
ed
to
ass
es
s
h
o
w
ac
cu
r
ate
th
ey
w
er
e
at
d
etec
tin
g
an
d
p
r
e
v
en
ti
n
g
b
r
ea
s
t
ca
n
ce
r
[
2
1
]
.
C
NNs
wer
e
p
ar
ticu
lar
ly
lev
er
ag
ed
d
u
e
to
t
h
eir
ad
ep
t
n
ess
in
lear
n
in
g
s
ig
n
if
ican
t
f
ea
tu
r
es
f
r
o
m
in
p
u
t
d
ata,
m
ak
in
g
th
em
h
ig
h
ly
ef
f
e
ctiv
e
f
o
r
im
ag
e
-
r
elate
d
ap
p
licatio
n
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
mp
o
w
er B
r
ea
s
tN
e
t:
b
r
ea
s
t c
a
n
ce
r
d
etec
tio
n
w
ith
tr
a
n
s
fer lea
r
n
in
g
…
(
V
a
is
h
a
li M.
Jo
s
h
i
)
1931
2.
3
.
Cla
s
s
if
ica
t
io
n
C
las
s
if
ier
s
ar
e
alg
o
r
ith
m
s
th
a
t
lear
n
to
ass
ig
n
class
lab
els
to
in
p
u
t
d
ata
b
ased
o
n
p
atter
n
s
lear
n
e
d
f
r
o
m
lab
ele
d
tr
ain
in
g
d
ata.
I
n
th
e
co
n
tex
t
o
f
im
ag
e
class
if
icatio
n
,
class
if
ier
s
an
aly
ze
th
e
f
ea
tu
r
es
ex
tr
ac
ted
f
r
o
m
im
a
g
es
an
d
m
ak
e
p
r
e
d
ictio
n
s
ab
o
u
t
th
eir
class
lab
els.
T
h
ese
p
r
ed
ictio
n
s
h
elp
i
n
task
s
s
u
ch
as
id
en
tify
i
n
g
o
b
jects,
r
ec
o
g
n
izin
g
p
atter
n
s
,
an
d
d
etec
tin
g
an
o
m
alies with
i
n
im
ag
es.
I
n
th
is
r
esear
ch
,
class
if
ier
s
s
u
ch
as
Alex
Net,
R
e
s
Net5
0
,
VGG
-
1
6
,
an
d
VGG
-
1
9
ar
e
e
m
p
lo
y
ed
to
class
if
y
im
ag
es
o
f
b
r
ea
s
t
tis
s
u
e
as
eith
er
b
en
ig
n
o
r
m
alig
n
an
t.
E
ac
h
class
if
ier
o
f
f
e
r
s
u
n
iq
u
e
ar
c
h
itectu
r
al
ch
ar
ac
ter
is
tics
an
d
ca
p
ab
ilit
ies,
m
ak
in
g
th
em
s
u
itab
le
f
o
r
an
aly
zin
g
m
e
d
ical
im
ag
es
an
d
ass
is
tin
g
in
th
e
d
iag
n
o
s
is
o
f
b
r
ea
s
t
ca
n
ce
r
.
A
lex
Net,
R
esNet5
0
,
VGG
-
1
6
,
an
d
VGG
-
1
9
ar
e
a
d
v
an
ce
d
DL
m
o
d
els
th
at
p
lay
a
cr
u
cial
r
o
le
in
b
r
ea
s
t
ca
n
ce
r
d
etec
tio
n
b
y
an
aly
zi
n
g
m
ed
ic
al
im
ag
es
to
id
en
tify
ca
n
ce
r
o
u
s
ce
lls
.
Alex
Net,
R
esNet5
0
,
VGG
-
1
6
,
an
d
VG
G
-
1
9
ar
e
s
o
p
h
is
ticated
DL
m
o
d
els
th
at
h
av
e
b
ee
n
a
d
ap
ted
f
o
r
b
r
ea
s
t
ca
n
ce
r
d
etec
tio
n
th
r
o
u
g
h
th
e
an
al
y
s
is
o
f
m
ed
ical
im
ag
es.
Her
e
’
s
a
d
etailed
lo
o
k
at
ea
ch
m
o
d
el
an
d
its
ap
p
licatio
n
in
th
is
co
n
tex
t:
-
Alex
Net
a)
Stru
ctu
r
e:
Alex
Net
co
m
p
r
is
es
8
lay
er
s
,
in
clu
d
i
n
g
5
c
o
n
v
o
lu
ti
o
n
al
lay
er
s
f
o
llo
wed
b
y
3
f
u
ll
y
co
n
n
ec
te
d
lay
er
s
.
I
t
u
tili
ze
s
R
eL
U
ac
tiv
atio
n
f
u
n
ctio
n
s
to
in
t
r
o
d
u
c
e
n
o
n
-
lin
ea
r
ity
,
m
ak
in
g
t
h
e
m
o
d
el
m
o
r
e
ca
p
ab
le
o
f
ca
p
tu
r
in
g
co
m
p
lex
p
atter
n
s
.
b)
Ap
p
licatio
n
in
b
r
ea
s
t
ca
n
ce
r
d
etec
tio
n
:
Alex
Net
ca
n
p
r
o
ce
s
s
lar
g
e
m
ed
ical
im
ag
es
an
d
id
en
tify
f
ea
tu
r
es
in
d
icativ
e
o
f
ca
n
ce
r
o
u
s
ch
an
g
es
in
b
r
ea
s
t
tis
s
u
e.
I
t
s
ar
ch
itectu
r
e
allo
ws
it
to
d
et
ec
t
v
ar
y
in
g
tex
tu
r
es
an
d
s
h
ap
es
ass
o
ciate
d
with
m
alig
n
an
t
tu
m
o
r
s
,
p
r
o
v
id
i
n
g
a
r
o
b
u
s
t
in
itial
f
r
am
ewo
r
k
f
o
r
m
ed
ical
im
ag
e
an
aly
s
is
.
-
R
esNet5
0
a)
Stru
ctu
r
e:
Alex
Net
co
m
p
r
is
es
8
lay
er
s
,
in
clu
d
i
n
g
5
c
o
n
v
o
lu
ti
o
n
al
lay
er
s
f
o
llo
wed
b
y
3
f
u
ll
y
co
n
n
ec
te
d
lay
er
s
.
I
t
u
tili
ze
s
R
eL
U
ac
tiv
atio
n
f
u
n
ctio
n
s
to
in
t
r
o
d
u
c
e
n
o
n
-
lin
ea
r
ity
,
m
ak
in
g
t
h
e
m
o
d
el
m
o
r
e
ca
p
ab
le
o
f
ca
p
tu
r
in
g
co
m
p
lex
p
atter
n
s
.
b)
Ap
p
licatio
n
in
b
r
ea
s
t
ca
n
ce
r
d
etec
tio
n
:
Alex
Net
ca
n
p
r
o
ce
s
s
lar
g
e
m
ed
ical
im
ag
es
an
d
id
en
tify
f
ea
tu
r
es
in
d
icativ
e
o
f
ca
n
ce
r
o
u
s
ch
an
g
es
in
b
r
ea
s
t
tis
s
u
e.
I
t
s
ar
ch
itectu
r
e
allo
ws
it
to
d
et
ec
t
v
ar
y
in
g
tex
tu
r
es
an
d
s
h
ap
es
ass
o
ciate
d
with
m
alig
n
an
t
t
u
m
o
r
s
,
p
r
o
v
id
i
n
g
a
r
o
b
u
s
t
in
itial
f
r
am
ewo
r
k
f
o
r
m
ed
ical
im
ag
e
an
aly
s
is
.
-
VGG
-
1
6
an
d
VGG
-
19
a)
Stru
ctu
r
e:
VGG
-
1
6
an
d
VGG
-
1
9
ar
e
k
n
o
w
n
f
o
r
th
eir
s
im
p
licity
an
d
d
ep
t
h
,
with
1
6
an
d
1
9
lay
er
s
,
r
esp
ec
tiv
ely
.
B
o
th
m
o
d
els
u
s
e
s
m
all
(
3
×
3
)
c
o
n
v
o
lu
tio
n
f
ilter
s
th
r
o
u
g
h
o
u
t
t
h
eir
ar
c
h
itectu
r
e,
wh
ic
h
h
elp
s
in
ca
p
tu
r
i
n
g
d
etailed
a
n
d
lo
ca
lized
f
ea
tu
r
es with
in
im
a
g
es.
b)
Ap
p
licatio
n
in
b
r
ea
s
t
ca
n
ce
r
d
etec
tio
n
:
t
h
ese
m
o
d
els
ex
c
el
in
id
en
tify
in
g
f
in
e
-
g
r
ain
ed
p
atter
n
s
in
b
r
ea
s
t
tis
s
u
e
im
ag
es.
VGG
-
1
6
an
d
VGG
-
1
9
ar
e
u
s
ef
u
l
d
iag
n
o
s
tic
to
o
ls
b
ec
a
u
s
e
o
f
th
eir
d
ep
th
,
wh
ic
h
en
ab
les th
em
to
d
is
tin
g
u
is
h
b
e
twee
n
b
en
ig
n
an
d
m
alig
n
an
t le
s
io
n
s
with
ex
tr
em
e
p
r
ec
is
io
n
.
-
Use o
f
p
re
-
tr
ain
e
d
m
o
d
els an
d
TL
a)
Pre
-
tr
ain
in
g
:
a
ll
th
ese
m
o
d
els
ar
e
in
itially
tr
ain
ed
o
n
lar
g
e
d
atasets
lik
e
I
m
ag
eNe
t,
wh
ich
co
n
s
is
ts
o
f
m
illi
o
n
s
o
f
d
iv
e
r
s
e
im
ag
es.
T
h
is
p
r
e
-
tr
ain
in
g
allo
ws
th
e
m
o
d
els
to
lear
n
a
b
r
o
a
d
s
et
o
f
f
e
atu
r
es
u
s
ef
u
l
f
o
r
v
a
r
io
u
s
im
ag
e
r
ec
o
g
n
itio
n
task
s
.
b)
TL
:
f
o
r
b
r
ea
s
t
ca
n
ce
r
d
ete
ctio
n
,
th
ese
p
r
e
-
tr
ain
ed
m
o
d
els
ar
e
f
in
e
-
tu
n
ed
o
n
s
p
ec
if
ic
m
ed
ical
im
ag
e
d
atasets
,
s
u
ch
as
th
e
B
C
W
d
iag
n
o
s
tic
d
ataset.
T
h
is
p
r
o
ce
s
s
in
v
o
l
v
es
r
etr
ain
in
g
th
e
m
o
d
els
o
n
b
r
ea
s
t
ca
n
ce
r
im
ag
es
s
o
th
ey
ca
n
ad
ap
t
th
eir
lear
n
e
d
f
e
atu
r
es
to
th
e
s
p
ec
if
ic
task
o
f
id
en
tify
in
g
ca
n
ce
r
o
u
s
ce
lls
.
T
h
ese
m
o
d
els
ar
e
o
f
ten
p
r
e
-
tr
ain
ed
o
n
lar
g
e
d
atasets
,
s
u
ch
as
I
m
ag
eNe
t,
wh
ich
h
elp
s
th
e
m
lear
n
a
wid
e
r
an
g
e
o
f
f
ea
tu
r
es.
Fo
r
b
r
ea
s
t
ca
n
ce
r
d
etec
tio
n
,
t
h
ese
p
r
e
-
tr
ain
e
d
m
o
d
els
ar
e
f
in
e
-
tu
n
e
d
u
s
in
g
m
ed
ical
im
ag
e
d
atasets
lik
e
th
e
B
C
W
d
iag
n
o
s
tic
d
ataset.
T
h
i
s
f
in
e
-
tu
n
in
g
p
r
o
ce
s
s
h
elp
s
th
e
m
o
d
els
to
b
etter
r
ec
o
g
n
ize
p
atter
n
s
s
p
ec
if
ic
t
o
b
r
e
ast
ca
n
ce
r
,
lead
in
g
to
h
ig
h
er
ac
cu
r
ac
y
,
s
en
s
itiv
ity
,
an
d
s
p
ec
if
icity
in
d
iag
n
o
s
is
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
e
p
r
o
p
o
s
ed
class
if
ier
s
wer
e
ev
alu
ated
u
s
in
g
s
tan
d
ar
d
v
alid
atio
n
m
etr
ics
lik
e
ac
cu
r
a
cy
,
r
ec
all
,
p
r
ec
is
io
n
,
an
d
F
-
m
ea
s
u
r
e.
Per
f
o
r
m
an
ce
m
et
r
ics
−
Acc
u
r
ac
y
:
m
ea
s
u
r
es
th
e
p
r
o
p
o
r
tio
n
o
f
c
o
r
r
ec
tly
i
d
en
tifie
d
ca
s
es
(
b
o
th
p
o
s
itiv
e
an
d
n
eg
a
tiv
e)
o
u
t
o
f
t
h
e
to
tal
ca
s
es.
−
Sen
s
itiv
ity
(
r
ec
all)
:
in
d
icate
s
th
e
m
o
d
el
’
s
ab
ilit
y
to
co
r
r
ec
tly
id
en
tify
p
atien
ts
with
b
r
ea
s
t
ca
n
ce
r
(
tr
u
e
p
o
s
itiv
e
r
ate)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
7
,
No
.
3
,
Ma
r
ch
20
2
5
:
1
92
7
-
1
93
5
1932
−
Sp
ec
if
icity
:
m
ea
s
u
r
es th
e
m
o
d
el
’
s
ab
ilit
y
to
co
r
r
ec
tly
id
en
tify
p
atien
ts
with
o
u
t b
r
ea
s
t c
an
ce
r
(
tr
u
e
n
eg
ativ
e
r
ate)
.
−
Pre
cisi
o
n
:
r
ef
lects
th
e
ac
cu
r
a
cy
o
f
p
o
s
itiv
e
p
r
e
d
ictio
n
s
,
s
h
o
win
g
h
o
w
m
an
y
o
f
th
e
p
o
s
itiv
ely
id
en
tifie
d
ca
s
es a
r
e
ac
tu
ally
co
r
r
ec
t.
−
F
-
Sco
r
e:
a
h
a
r
m
o
n
ic
m
ea
n
o
f
p
r
ec
is
io
n
an
d
r
ec
all,
p
r
o
v
id
in
g
a
s
in
g
le
m
etr
ic
t
h
at
b
ala
n
ce
s
b
o
th
f
alse
p
o
s
itiv
es a
n
d
f
alse n
eg
ativ
es.
Th
e
ex
p
er
im
en
t
u
s
ed
th
e
BCW
d
iag
n
o
s
tic
d
ataset
to
d
ia
g
n
o
s
e
b
r
ea
s
t
ca
n
ce
r
u
s
in
g
s
ev
e
r
a
l
class
if
ier
s
,
in
clu
d
in
g
Alex
Net,
R
esNet5
0
,
VGG
-
1
6
,
an
d
VGG
-
1
9
.
As
s
h
o
wn
in
Fig
u
r
e
2
,
th
e
o
u
t
co
m
es
o
f
d
if
f
er
en
t
class
if
ier
s
d
if
f
er
ed
.
VGG
-
1
9
h
ad
th
e
b
est
class
if
icatio
n
ac
c
u
r
ac
y
o
f
9
7
.
9
5
%,
wh
ile
R
esN
et5
0
h
ad
th
e
lo
west,
at
9
4
%.
T
h
e
F1
-
s
co
r
e,
w
h
ich
is
a
co
m
b
in
atio
n
o
f
p
r
ec
is
io
n
an
d
r
ec
all
r
atio
s
,
is
in
d
icativ
e
o
f
a
m
o
d
el
’
s
class
if
icatio
n
p
er
f
o
r
m
an
ce
.
Su
r
p
r
is
in
g
ly
,
all
f
o
u
r
class
if
ier
s
(
Alex
Net,
R
esNe
t5
0
,
VGG
-
1
6
,
an
d
VGG
-
1
9
)
pr
o
d
u
ce
d
eq
u
al
F1
-
s
co
r
e
v
alu
es o
f
9
6
.
6
6
%.
It
’
s
n
o
t
e
w
o
r
t
h
y
t
h
a
t
a
lt
h
o
u
g
h
V
G
G
-
1
6
a
n
d
V
G
G
-
1
9
s
h
o
we
d
s
u
p
e
r
i
o
r
s
e
n
s
it
i
v
it
y
a
n
d
a
c
c
u
r
a
c
y
i
n
F
i
g
u
r
e
2
,
V
G
G
-
19
o
u
t
p
e
r
f
o
r
m
e
d
i
n
t
e
r
m
s
o
f
F
1
-
s
c
o
r
e
a
n
d
p
r
e
c
i
s
i
o
n
.
C
o
n
s
e
q
u
e
n
tl
y
,
b
o
t
h
VG
G
-
1
6
a
n
d
V
G
G
-
19
c
l
a
s
s
i
f
ie
r
s
,
h
a
v
i
n
g
b
e
e
n
t
r
a
i
n
ed
o
n
t
h
e
d
a
t
a
s
e
t
,
e
m
e
r
g
e
d
a
s
t
h
e
t
o
p
-
p
e
r
f
o
r
m
i
n
g
c
l
a
s
s
i
f
i
e
r
s
f
o
r
p
r
e
d
i
c
t
i
n
g
n
e
w
c
a
s
es
a
m
o
n
g
a
l
l
m
o
d
e
ls
c
o
n
s
i
d
e
r
e
d
i
n
t
h
i
s
c
o
m
p
a
r
i
s
o
n
.
T
a
b
l
e
1
s
h
o
w
s
p
r
o
p
o
s
e
d
r
es
u
l
t
wi
t
h
s
ta
t
e
o
f
a
r
e
m
e
t
h
o
d
s
.
Fig
u
r
e
2
.
Per
f
o
r
m
an
c
e
ev
alu
at
io
n
o
f
class
if
ier
s
T
ab
le
1
.
C
o
m
p
a
r
is
o
n
o
f
th
e
p
r
o
p
o
s
ed
wo
r
k
with
s
tate
-
of
-
th
e
-
ar
t w
o
r
k
s
A
u
t
h
o
r
M
e
t
h
o
d
A
c
c
u
r
a
c
y
(
%)
P
r
e
c
i
s
i
o
n
(
%)
S
e
n
s
i
t
i
v
i
t
y
(
%)
S
p
e
c
i
f
i
c
i
t
y
(
%)
F
-
s
c
o
r
e
(
%)
M
a
g
b
o
o
e
t
a
l
.
[
2
2
]
Lo
g
i
s
t
i
c
r
e
g
r
e
ssi
o
n
80
80
-
--
76
N
a
j
i
e
t
a
l
.
[
2
3
]
S
V
M
9
7
.
2
9
7
.
5
-
-
-
A
l
w
o
h
a
i
b
i
e
t
a
l
.
[
2
4
]
M
u
l
t
i
-
st
a
g
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
82
81
82
96
--
S
a
h
o
o
e
t
a
l
.
[
2
5
]
S
V
M
8
4
.
8
5
9
0
.
5
9
8
4
.
8
1
8
4
.
9
3
8
4
.
8
1
A
l
su
d
a
n
i
e
t
a
l
.
[
2
6
]
S
V
M
9
8
.
6
9
7
.
5
9
6
.
9
-
-
P
r
o
p
o
se
d
V
G
G
N
e
t
-
19
9
8
.
7
5
9
7
.
3
5
9
7
.
3
8
9
8
.
3
5
9
7
.
6
6
4.
CO
NCLU
SI
O
N
On
e
m
ajo
r
g
lo
b
al
h
ea
lth
co
n
c
er
n
is
ca
n
ce
r
,
a
d
is
ea
s
e
with
t
er
r
ib
le
im
p
licatio
n
s
.
Of
all
th
e
ca
n
ce
r
s
,
b
r
ea
s
t
ca
n
ce
r
is
th
e
m
o
s
t
well
-
k
n
o
wn
.
No
t
o
n
ly
m
ay
ea
r
ly
id
en
tific
atio
n
o
f
ca
n
ce
r
s
av
e
liv
es,
b
u
t
it
also
lo
wer
s
th
e
co
s
t
o
f
tr
ea
tm
en
t.
T
h
er
ef
o
r
e,
it
is
cr
u
cial
to
d
es
ig
n
a
tr
u
s
two
r
th
y
p
r
e
d
ictio
n
s
y
s
tem
.
T
h
is
s
tu
d
y
ev
alu
ated
f
o
u
r
d
is
tin
ct
ML
alg
o
r
ith
m
s
in
ter
m
s
o
f
F1
-
s
co
r
e,
r
ec
all
(
s
en
s
itiv
ity
)
,
ac
cu
r
ac
y
,
an
d
p
r
ec
is
io
n
to
f
in
d
th
e
b
est
m
o
d
el
f
o
r
p
r
ed
ic
tin
g
b
r
ea
s
t
ca
n
ce
r
d
is
ea
s
e
(
B
C
D)
.
T
h
e
r
esu
lts
o
f
th
e
an
aly
s
is
co
n
d
u
cted
o
n
th
e
B
C
W
d
iag
n
o
s
tic
d
ataset
in
d
icate
d
th
at
th
e
p
er
ce
p
tr
o
n
an
d
VGG
-
1
9
m
eth
o
d
s
h
ad
s
u
p
er
io
r
s
en
s
itiv
ity
an
d
ac
cu
r
ac
y
.
On
t
h
e
o
th
er
h
an
d
,
R
esNet5
0
s
h
o
wed
less
er
s
en
s
itiv
ity
an
d
ac
cu
r
ac
y
,
w
h
er
ea
s
VGG
-
1
6
s
h
o
wed
b
et
ter
p
r
ec
is
io
n
an
d
F1
-
s
co
r
e
.
Ou
r
f
in
d
in
g
s
alig
n
with
p
r
e
v
io
u
s
r
esear
ch
h
ig
h
lig
h
tin
g
t
h
e
p
o
ten
tial
o
f
DL
m
o
d
els
in
m
ed
ical
d
iag
n
o
s
tics
.
Ho
wev
er
,
o
u
r
s
tu
d
y
s
p
ec
if
ically
f
o
cu
s
es
o
n
b
r
e
ast
ca
n
ce
r
d
etec
tio
n
an
d
p
r
o
v
i
d
es
a
co
m
p
ar
ativ
e
an
aly
s
is
o
f
d
if
f
er
en
t
ML
m
o
d
els,
ad
d
in
g
to
t
h
e
ex
is
tin
g
b
o
d
y
o
f
k
n
o
wled
g
e
b
y
h
ig
h
li
g
h
tin
g
th
e
s
p
ec
if
ic
s
tr
en
g
th
s
an
d
wea
k
n
ess
es o
f
ea
ch
m
o
d
el
i
n
th
is
co
n
tex
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
mp
o
w
er B
r
ea
s
tN
e
t:
b
r
ea
s
t c
a
n
ce
r
d
etec
tio
n
w
ith
tr
a
n
s
fer lea
r
n
in
g
…
(
V
a
is
h
a
li M.
Jo
s
h
i
)
1933
Fu
tu
r
e
r
esear
ch
s
h
o
u
l
d
ex
p
lo
r
e
th
e
i
n
teg
r
atio
n
o
f
th
ese
M
L
m
o
d
els
with
o
th
er
d
iag
n
o
s
t
ic
to
o
ls
to
en
h
an
ce
th
eir
p
r
ed
ictiv
e
p
o
w
e
r
.
Ad
d
itio
n
ally
,
e
x
am
in
in
g
th
e
p
er
f
o
r
m
an
ce
o
f
th
ese
m
o
d
els
o
n
lar
g
er
an
d
m
o
r
e
d
iv
er
s
e
d
atasets
co
u
ld
p
r
o
v
id
e
f
u
r
th
er
in
s
ig
h
ts
in
to
th
eir
g
en
er
aliza
b
ilit
y
.
I
n
v
esti
g
atin
g
t
h
e
co
m
b
in
atio
n
o
f
d
if
f
er
en
t
ML
m
o
d
els
th
r
o
u
g
h
en
s
em
b
le
m
eth
o
d
s
m
ig
h
t
a
ls
o
im
p
r
o
v
e
p
r
e
d
ictio
n
ac
cu
r
ac
y
an
d
r
eliab
ilit
y
.
Mo
r
eo
v
er
,
in
co
r
p
o
r
atin
g
p
atien
t
d
em
o
g
r
a
p
h
ics
an
d
g
e
n
etic
in
f
o
r
m
atio
n
co
u
ld
lea
d
to
m
o
r
e
p
er
s
o
n
alize
d
an
d
p
r
ec
is
e
b
r
ea
s
t c
an
ce
r
d
etec
tio
n
s
y
s
tem
s
.
I
n
s
u
m
m
a
r
y
,
o
u
r
s
tu
d
y
u
n
d
e
r
s
co
r
es
th
e
im
p
o
r
tan
ce
o
f
u
s
in
g
ad
v
an
ce
d
DL
m
o
d
els
,
p
ar
ticu
lar
ly
VGG
-
1
9
,
f
o
r
th
e
ea
r
ly
a
n
d
ac
cu
r
ate
d
etec
tio
n
o
f
b
r
ea
s
t
ca
n
ce
r
.
B
y
lev
er
a
g
in
g
th
es
e
m
o
d
els,
m
ed
ical
p
r
o
f
ess
io
n
als
ca
n
s
ig
n
if
ican
tly
im
p
r
o
v
e
d
iag
n
o
s
is
ac
cu
r
ac
y
a
n
d
ef
f
icien
c
y
,
u
ltima
tely
lead
i
n
g
to
b
etter
p
atien
t
o
u
tco
m
es th
r
o
u
g
h
ea
r
lier
an
d
m
o
r
e
r
eliab
le
d
etec
tio
n
.
RE
F
E
R
E
NC
E
S
[
1
]
L.
L
u
o
e
t
a
l
.
,
“
D
e
e
p
l
e
a
r
n
i
n
g
i
n
b
r
e
a
st
c
a
n
c
e
r
i
ma
g
i
n
g
:
a
d
e
c
a
d
e
o
f
p
r
o
g
r
e
ss
a
n
d
f
u
t
u
r
e
d
i
r
e
c
t
i
o
n
s
,
”
I
EEE
R
e
v
i
e
w
s i
n
Bi
o
m
e
d
i
c
a
l
E
n
g
i
n
e
e
ri
n
g
,
p
p
.
1
–
2
0
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
R
B
M
E.
2
0
2
4
.
3
3
5
7
8
7
7
.
[
2
]
Q
.
D
a
n
,
Z
.
X
u
,
H
.
B
u
r
r
o
w
s,
J.
B
i
ssr
a
m,
J.
S
.
A
.
S
t
r
i
n
g
e
r
,
a
n
d
Y
.
Li
,
“
D
i
a
g
n
o
st
i
c
p
e
r
f
o
r
m
a
n
c
e
o
f
d
e
e
p
l
e
a
r
n
i
n
g
i
n
u
l
t
r
a
s
o
u
n
d
d
i
a
g
n
o
si
s
o
f
b
r
e
a
s
t
c
a
n
c
e
r
:
a
sy
st
e
ma
t
i
c
r
e
v
i
e
w
,
”
N
J
P
Pre
c
i
s
i
o
n
O
n
c
o
l
o
g
y
,
v
o
l
.
8
,
n
o
.
1
,
p
.
2
1
,
Ja
n
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
6
9
8
-
0
2
4
-
0
0
5
1
4
-
z.
[
3
]
M
.
N
a
ss
e
r
a
n
d
U
.
K
.
Y
u
s
o
f
,
“
D
e
e
p
l
e
a
r
n
i
n
g
b
a
se
d
me
t
h
o
d
s
f
o
r
b
r
e
a
s
t
c
a
n
c
e
r
d
i
a
g
n
o
si
s
:
a
s
y
st
e
ma
t
i
c
r
e
v
i
e
w
a
n
d
f
u
t
u
r
e
d
i
r
e
c
t
i
o
n
,
”
D
i
a
g
n
o
s
t
i
c
s
,
v
o
l
.
1
3
,
n
o
.
1
,
p
.
1
6
1
,
Ja
n
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
d
i
a
g
n
o
st
i
c
s
1
3
0
1
0
1
6
1
.
[
4
]
C
.
H
.
S
a
l
h
a
n
d
A
.
M
.
A
l
i
,
“
B
r
e
a
st
c
a
n
c
e
r
r
e
c
o
g
n
i
t
i
o
n
b
a
se
d
o
n
p
e
r
f
o
r
man
c
e
e
v
a
l
u
a
t
i
o
n
o
f
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
,
”
I
n
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
El
e
c
t
r
i
c
a
l
E
n
g
i
n
e
e
ri
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
I
J
EE
C
S
)
,
v
o
l
.
2
7
,
n
o
.
2
,
p
p
.
9
8
0
–
9
8
9
,
A
u
g
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s
.
v
2
7
.
i
2
.
p
p
9
8
0
-
9
8
9
.
[
5
]
S
.
S
.
R
e
d
d
y
,
N
.
P
i
l
l
i
,
P
.
V
o
o
s
a
l
a
,
a
n
d
S
.
R
.
C
h
i
g
u
r
u
p
a
t
i
,
“
A
c
o
m
p
a
r
a
t
i
v
e
st
u
d
y
t
o
p
r
e
d
i
c
t
b
r
e
a
s
t
c
a
n
c
e
r
u
si
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,
”
I
n
d
o
n
e
s
i
a
n
J
o
u
r
n
a
l
o
f
El
e
c
t
ri
c
a
l
En
g
i
n
e
e
ri
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
I
J
EE
C
S
)
,
v
o
l
.
2
7
,
n
o
.
1
,
p
p
.
1
7
1
–
1
8
0
,
J
u
l
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s.
v
2
7
.
i
1
.
p
p
1
7
1
-
1
8
0
.
[
6
]
A
.
A
.
J
a
si
m
,
A
.
A
.
J
a
l
a
l
,
N
.
M
.
A
b
d
u
l
a
t
e
e
f
,
a
n
d
N
.
A
.
T
a
l
i
b
,
“
Ef
f
e
c
t
i
v
e
n
e
s
s
e
v
a
l
u
a
t
i
o
n
o
f
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
f
o
r
b
r
e
a
st
c
a
n
c
e
r
p
r
e
d
i
c
t
i
o
n
,
”
Bu
l
l
e
t
i
n
o
f
E
l
e
c
t
r
i
c
a
l
E
n
g
i
n
e
e
ri
n
g
a
n
d
I
n
f
o
rm
a
t
i
c
s
(
B
EEI)
,
v
o
l
.
1
1
,
n
o
.
3
,
p
p
.
1
5
1
6
–
1
5
2
5
,
Ju
n
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
9
1
/
e
e
i
.
v
1
1
i
3
.
3
6
2
1
.
[
7
]
S
.
C
h
i
t
t
i
n
e
n
i
a
n
d
S
.
S
.
E
d
a
r
a
,
“
A
u
t
o
mat
e
d
b
r
e
a
st
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
s
y
s
t
e
m
f
r
o
m
b
r
e
a
s
t
mam
mo
g
r
a
m
u
s
i
n
g
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
I
n
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
El
e
c
t
r
i
c
a
l
E
n
g
i
n
e
e
ri
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
I
J
EE
C
S
)
,
v
o
l
.
2
5
,
n
o
.
1
,
p
p
.
5
8
0
–
5
8
8
,
J
a
n
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s
.
v
2
5
.
i
1
.
p
p
5
8
0
-
5
8
8
.
[
8
]
M
.
H
.
A
l
sh
a
y
e
j
i
,
H
.
El
l
e
t
h
y
,
S
.
A
b
e
d
,
a
n
d
R
.
G
u
p
t
a
,
“
C
o
m
p
u
t
e
r
-
a
i
d
e
d
d
e
t
e
c
t
i
o
n
o
f
b
r
e
a
s
t
c
a
n
c
e
r
o
n
t
h
e
W
i
sc
o
n
si
n
d
a
t
a
se
t
:
A
n
a
r
t
i
f
i
c
i
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s
a
p
p
r
o
a
c
h
,
”
Bi
o
m
e
d
i
c
a
l
S
i
g
n
a
l
Pro
c
e
ss
i
n
g
a
n
d
C
o
n
t
r
o
l
,
v
o
l
.
7
1
,
p
.
1
0
3
1
4
1
,
Ja
n
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
b
s
p
c
.
2
0
2
1
.
1
0
3
1
4
1
.
[
9
]
S
.
R
.
S
a
n
n
a
si
C
h
a
k
r
a
v
a
r
t
h
y
,
N
.
B
h
a
r
a
n
i
d
h
a
r
a
n
,
V
.
V
i
n
o
t
h
K
u
m
a
r
,
T
.
R
.
M
a
h
e
s
h
,
M
.
S
.
A
l
q
a
h
t
a
n
i
,
a
n
d
S
.
G
u
l
u
w
a
d
i
,
“
D
e
e
p
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
w
i
t
h
f
u
z
z
y
e
n
s
e
m
b
l
e
a
p
p
r
o
a
c
h
f
o
r
t
h
e
e
a
r
l
y
d
e
t
e
c
t
i
o
n
o
f
b
r
e
a
st
c
a
n
c
e
r
,
”
BM
C
Me
d
i
c
a
l
I
m
a
g
i
n
g
,
v
o
l
.
2
4
,
n
o
.
1
,
p
.
8
2
,
A
p
r
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
8
6
/
s
1
2
8
8
0
-
0
2
4
-
0
1
2
6
7
-
8.
[
1
0
]
M
.
I
.
M
a
h
mu
d
,
M
.
M
a
m
u
n
,
a
n
d
A
.
A
b
d
e
l
g
a
w
a
d
,
“
A
d
e
e
p
a
n
a
l
y
si
s
o
f
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
b
a
se
d
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
h
i
s
t
o
p
a
t
h
o
l
o
g
y
i
ma
g
e
s
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
1
0
t
h
I
n
t
e
rn
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
s
si
n
g
a
n
d
I
n
t
e
g
r
a
t
e
d
N
e
t
w
o
rks
,
S
PI
N
2
0
2
3
,
I
EEE,
M
a
r
.
2
0
2
3
,
p
p
.
1
9
8
–
2
0
4
.
d
o
i
:
1
0
.
1
1
0
9
/
S
P
I
N
5
7
0
0
1
.
2
0
2
3
.
1
0
1
1
7
1
1
0
.
[
1
1
]
G
.
C
h
u
g
h
,
S
.
K
u
m
a
r
,
a
n
d
N
.
S
i
n
g
h
,
“
Tr
a
n
sN
e
t
:
a
c
o
mp
a
r
a
t
i
v
e
st
u
d
y
o
n
b
r
e
a
st
c
a
r
c
i
n
o
m
a
d
i
a
g
n
o
s
i
s
w
i
t
h
c
l
a
ssi
c
a
l
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
p
a
r
a
d
i
g
m,
”
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
3
,
n
o
.
1
1
,
p
p
.
3
3
8
5
5
–
3
3
8
7
7
,
S
e
p
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
0
4
2
-
0
2
3
-
1
6
9
3
8
-
x.
[
1
2
]
S
.
S
.
B
o
u
d
o
u
h
a
n
d
M
.
B
o
u
a
k
k
a
z
,
“
N
e
w
e
n
h
a
n
c
e
d
b
r
e
a
s
t
t
u
m
o
r
d
e
t
e
c
t
i
o
n
a
p
p
r
o
a
c
h
i
n
mamm
o
g
r
a
m
s
c
a
n
s b
a
s
e
d
o
n
p
r
e
-
p
r
o
c
e
ssi
n
g
a
n
d
d
e
e
p
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,
”
Mu
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
3
,
n
o
.
9
,
p
p
.
2
7
3
5
7
–
2
7
3
7
8
,
S
e
p
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
0
4
2
-
0
2
3
-
1
6
5
4
5
-
w.
[
1
3
]
T.
M
a
h
m
o
o
d
,
T.
S
a
b
a
,
A
.
R
e
h
m
a
n
,
a
n
d
F
.
S
.
A
l
a
mr
i
,
“
H
a
r
n
e
ssi
n
g
t
h
e
p
o
w
e
r
o
f
r
a
d
i
o
mi
c
s
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
i
m
p
r
o
v
e
d
b
r
e
a
st
c
a
n
c
e
r
d
i
a
g
n
o
si
s
w
i
t
h
mu
l
t
i
p
a
r
a
m
e
t
r
i
c
b
r
e
a
st
mam
mo
g
r
a
p
h
y
,
”
E
x
p
e
rt
S
y
st
e
m
s
w
i
t
h
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
4
9
,
p
.
1
2
3
7
4
7
,
S
e
p
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
4
.
1
2
3
7
4
7
.
[
1
4
]
A
.
S
a
h
u
,
P
.
K
.
D
a
s,
a
n
d
S
.
M
e
h
e
r
,
“
A
n
e
f
f
i
c
i
e
n
t
d
e
e
p
l
e
a
r
n
i
n
g
sc
h
e
me
t
o
d
e
t
e
c
t
b
r
e
a
st
c
a
n
c
e
r
u
si
n
g
m
a
mm
o
g
r
a
m
a
n
d
u
l
t
r
a
so
u
n
d
b
r
e
a
s
t
i
ma
g
e
s,
”
B
i
o
m
e
d
i
c
a
l
S
i
g
n
a
l
Pro
c
e
ss
i
n
g
a
n
d
C
o
n
t
ro
l
,
v
o
l
.
8
7
,
p
.
1
0
5
3
7
7
,
Ja
n
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
b
s
p
c
.
2
0
2
3
.
1
0
5
3
7
7
.
[
1
5
]
B
.
S
.
A
b
u
n
a
ss
e
r
,
M
.
R
.
J.
A
L
-
H
i
e
a
l
y
,
I
.
S
.
Za
q
o
u
t
,
a
n
d
S
.
S
.
A
b
u
-
N
a
s
e
r
,
“
C
o
n
v
o
l
u
t
i
o
n
n
e
u
r
a
l
n
e
t
w
o
r
k
f
o
r
b
r
e
a
s
t
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
d
e
e
p
l
e
a
r
n
i
n
g
,
”
As
i
a
n
Pa
c
i
f
i
c
J
o
u
r
n
a
l
o
f
C
a
n
c
e
r
P
rev
e
n
t
i
o
n
,
v
o
l
.
2
4
,
n
o
.
2
,
p
p
.
5
3
1
–
5
4
4
,
F
e
b
.
2
0
2
3
,
d
o
i
:
1
0
.
3
1
5
5
7
/
A
P
J
C
P
.
2
0
2
3
.
2
4
.
2
.
5
3
1
.
[
1
6
]
A
.
R
a
z
a
,
N
.
U
l
l
a
h
,
J.
A
.
K
h
a
n
,
M
.
A
ssam,
A
.
G
u
z
z
o
,
a
n
d
H
.
A
l
j
u
a
i
d
,
“
D
e
e
p
B
r
e
a
s
t
C
a
n
c
e
r
N
e
t
:
A
n
o
v
e
l
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
f
o
r
b
r
e
a
s
t
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
u
s
i
n
g
u
l
t
r
a
s
o
u
n
d
i
m
a
g
e
s
,
”
Ap
p
l
i
e
d
S
c
i
e
n
c
e
s
(
S
w
i
t
zerl
a
n
d
)
,
v
o
l
.
1
3
,
n
o
.
4
,
p
.
2
0
8
2
,
F
e
b
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
1
3
0
4
2
0
8
2
.
[
1
7
]
Y
.
Y
a
r
i
,
T.
V
.
N
g
u
y
e
n
,
a
n
d
H
.
T.
N
g
u
y
e
n
,
“
D
e
e
p
l
e
a
r
n
i
n
g
a
p
p
l
i
e
d
f
o
r
h
i
s
t
o
l
o
g
i
c
a
l
d
i
a
g
n
o
si
s
o
f
b
r
e
a
st
c
a
n
c
e
r
,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
8
,
p
p
.
1
6
2
4
3
2
–
1
6
2
4
4
8
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
0
.
3
0
2
1
5
5
7
.
[
1
8
]
L.
B
a
l
k
e
n
e
n
d
e
,
J.
Te
u
w
e
n
,
a
n
d
R
.
M
.
M
a
n
n
,
“
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
b
r
e
a
s
t
c
a
n
c
e
r
i
ma
g
i
n
g
,
”
S
e
m
i
n
a
rs
i
n
N
u
c
l
e
a
r
Me
d
i
c
i
n
e
,
v
o
l
.
5
2
,
n
o
.
5
,
p
p
.
5
8
4
–
5
9
6
,
S
e
p
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
5
3
/
j
.
s
e
mn
u
c
l
m
e
d
.
2
0
2
2
.
0
2
.
0
0
3
.
[
1
9
]
W
.
W
o
l
b
e
r
g
,
O
.
M
a
n
g
a
sar
i
a
n
,
N
.
S
t
r
e
e
t
,
a
n
d
W
.
S
t
r
e
e
t
,
“
B
r
e
a
s
t
c
a
n
c
e
r
W
i
sc
o
n
s
i
n
(
d
i
a
g
n
o
st
i
c
)
d
a
t
a
set
.
”
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
a
r
c
h
i
v
e
.
i
c
s.
u
c
i
.
e
d
u
/
d
a
t
a
s
e
t
/
1
7
/
b
r
e
a
s
t
+
c
a
n
c
e
r
+
w
i
sc
o
n
si
n
+
d
i
a
g
n
o
st
i
c
[
2
0
]
M
.
M
.
H
o
ssi
n
,
F
.
M
.
J
.
M
.
S
h
a
mr
a
t
,
M
.
R
.
B
h
u
i
y
a
n
,
R
.
A
.
H
i
r
a
,
T.
K
h
a
n
,
a
n
d
S
.
M
o
l
l
a
,
“
B
r
e
a
s
t
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
:
a
n
e
f
f
e
c
t
i
v
e
c
o
m
p
a
r
i
so
n
o
f
d
i
f
f
e
r
e
n
t
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
o
n
t
h
e
W
i
s
c
o
n
s
i
n
d
a
t
a
s
e
t
,
”
Bu
l
l
e
t
i
n
o
f
E
l
e
c
t
ri
c
a
l
En
g
i
n
e
e
r
i
n
g
a
n
d
I
n
f
o
rm
a
t
i
c
s
(
B
EEI)
,
v
o
l
.
1
2
,
n
o
.
4
,
p
p
.
2
4
4
6
–
2
4
5
6
,
A
u
g
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
5
9
1
/
e
e
i
.
v
1
2
i
4
.
4
4
4
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
In
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
3
7
,
No
.
3
,
Ma
r
ch
20
2
5
:
1
92
7
-
1
93
5
1934
[
2
1
]
S
.
M
o
h
a
p
a
t
r
a
,
S
.
M
u
d
u
l
y
,
S
.
M
o
h
a
n
t
y
,
J.
V
.
R
.
R
a
v
i
n
d
r
a
,
a
n
d
S
.
N
.
M
o
h
a
n
t
y
,
“
E
v
a
l
u
a
t
i
o
n
o
f
d
e
e
p
l
e
a
r
n
i
n
g
mo
d
e
l
s
f
o
r
d
e
t
e
c
t
i
n
g
b
r
e
a
s
t
c
a
n
c
e
r
u
si
n
g
h
i
s
t
o
p
a
t
h
o
l
o
g
i
c
a
l
mammo
g
r
a
ms
I
mag
e
s,
”
S
u
s
t
a
i
n
a
b
l
e
O
p
e
r
a
t
i
o
n
s
a
n
d
C
o
m
p
u
t
e
r
s
,
v
o
l
.
3
,
p
p
.
2
9
6
–
3
0
2
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
s
u
s
o
c
.
2
0
2
2
.
0
6
.
0
0
1
.
[
2
2
]
V
.
P
.
C
.
M
a
g
b
o
o
a
n
d
M
.
S
.
M
a
g
b
o
o
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
c
l
a
ss
i
f
i
e
r
s
o
n
b
r
e
a
st
c
a
n
c
e
r
r
e
c
u
r
r
e
n
c
e
s
,
”
Pro
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
1
9
2
,
p
p
.
2
7
4
2
–
2
7
5
2
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s.
2
0
2
1
.
0
9
.
0
4
4
.
[
2
3
]
M
.
A
.
N
a
j
i
,
S
.
E
.
F
i
l
a
l
i
,
K
.
A
a
r
i
k
a
,
E
.
H
.
B
e
n
l
a
h
mar,
R
.
A
.
A
b
d
e
l
o
u
h
a
h
i
d
,
a
n
d
O
.
D
e
b
a
u
c
h
e
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
m
s
f
o
r
b
r
e
a
s
t
c
a
n
c
e
r
p
r
e
d
i
c
t
i
o
n
a
n
d
d
i
a
g
n
o
si
s,
”
Pr
o
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
1
9
1
,
p
p
.
4
8
7
-
4
9
2
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s
.
2
0
2
1
.
0
7
.
0
6
2
.
[2
4
]
M
.
A
l
w
o
h
a
i
b
i
,
M
.
A
l
z
a
q
e
b
a
h
,
N
.
M
.
A
l
o
t
a
i
b
i
,
A
.
M
.
A
l
z
a
h
r
a
n
i
,
a
n
d
M
.
Zo
u
c
h
,
“
A
h
y
b
r
i
d
m
u
l
t
i
-
s
t
a
g
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
b
a
se
d
o
n
b
r
a
i
n
s
t
o
r
m
i
n
g
o
p
t
i
mi
z
a
t
i
o
n
a
l
g
o
r
i
t
h
m
f
o
r
b
r
e
a
s
t
c
a
n
c
e
r
r
e
c
u
r
r
e
n
c
e
p
r
e
d
i
c
t
i
o
n
,
”
J
o
u
r
n
a
l
o
f
K
i
n
g
S
a
u
d
U
n
i
v
e
rsi
t
y
-
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
3
4
,
n
o
.
8
,
p
p
.
5
1
9
2
–
5
2
0
3
,
S
e
p
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
k
s
u
c
i
.
2
0
2
1
.
0
5
.
0
0
4
.
[2
5
]
G
.
S
a
h
o
o
,
A
.
K
.
N
a
y
a
k
,
P
.
K
.
Tr
i
p
a
t
h
y
,
a
n
d
J
.
Tr
i
p
a
t
h
y
,
“
A
n
o
v
e
l
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
b
a
se
d
h
y
b
r
i
d
a
p
p
r
o
a
c
h
f
o
r
b
r
e
a
s
t
c
a
n
c
e
r
r
e
l
a
p
se
p
r
e
d
i
c
t
i
o
n
,
”
I
n
d
o
n
e
s
i
a
n
J
o
u
rn
a
l
o
f
El
e
c
t
ri
c
a
l
E
n
g
i
n
e
e
r
i
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
I
J
EE
C
S
)
,
v
o
l
.
3
2
,
n
o
.
3
,
p
p
.
1
6
5
5
–
1
6
6
3
,
D
e
c
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s.
v
3
2
.
i
3
.
p
p
1
6
5
5
-
1
6
6
3
.
[2
6
]
M
.
Q
.
A
l
s
u
d
a
n
i
,
H
.
F
.
F
a
k
h
r
u
l
d
e
e
n
,
a
n
d
I
.
F
.
Y
o
u
si
f
,
“
C
o
mp
a
r
a
t
i
v
e
e
v
a
l
u
a
t
i
o
n
o
f
d
a
t
a
mi
n
i
n
g
a
l
g
o
r
i
t
h
ms
i
n
b
r
e
a
s
t
c
a
n
c
e
r
,
”
I
n
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
El
e
c
t
r
i
c
a
l
E
n
g
i
n
e
e
ri
n
g
a
n
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
I
J
EE
C
S
)
,
v
o
l
.
3
1
,
n
o
.
2
,
p
p
.
7
7
7
–
7
8
4
,
A
u
g
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
e
c
s
.
v
3
1
.
i
2
.
p
p
7
7
7
-
7
8
4
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Dr
.
Va
ish
a
l
i
M.
J
o
shi
is
a
ss
o
c
iate
p
ro
fe
ss
o
r
a
t
KJ
EI
’
s
Tri
n
it
y
Ac
a
d
e
m
y
o
f
En
g
i
n
e
e
rin
g
,
P
u
n
e
I
n
d
ia.
S
h
e
h
o
ld
s
a
P
h
.
D
.
d
e
g
re
e
in
El
e
c
tro
n
ics
En
g
in
e
e
rin
g
.
He
r
re
se
a
rc
h
a
re
a
s
a
re
im
a
g
e
/sig
n
a
l
p
r
o
c
e
ss
in
g
,
VLS
I
,
AI
-
ML
,
a
n
d
p
a
tt
e
r
n
re
c
o
g
n
it
io
n
.
Dr
.
Jo
sh
i
h
a
s
fi
led
se
v
e
ra
l
p
a
ten
ts.
He
r
re
se
a
r
c
h
in
tere
sts
in
c
lu
d
e
ima
g
e
/sig
n
a
l
p
r
o
c
e
ss
in
g
,
b
i
o
m
e
tri
c
s,
m
e
d
ica
l
ima
g
e
a
n
d
a
n
a
ly
sis,
a
n
d
p
a
tt
e
rn
re
c
o
g
n
it
io
n
.
S
h
e
h
a
s
p
u
b
li
sh
e
d
ten
re
se
a
rc
h
a
rti
c
les
in
in
tern
a
ti
o
n
a
l
a
n
d
n
a
ti
o
n
a
l
jo
u
rn
a
l
s
a
n
d
c
o
n
fe
re
n
c
e
s.
S
h
e
p
u
b
li
sh
e
d
th
re
e
p
a
p
e
rs
i
n
S
CI
i
n
d
e
x
e
d
jo
u
r
n
a
l.
S
h
e
h
a
s
2
3
y
e
a
rs
o
f
tea
c
h
in
g
a
n
d
re
se
a
rc
h
e
x
p
e
rien
c
e
in
th
e
fiel
d
o
f
VLS
I,
c
o
m
m
u
n
ica
ti
o
n
,
AI
-
M
L
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
j
o
sh
i
v
a
ish
a
li
8
6
7
5
@
g
m
a
il
.
c
o
m
.
Pra
jk
ta
P
.
Da
n
d
a
v
a
te
is
a
ss
istan
t
p
ro
fe
ss
o
r
a
t
Vish
wa
k
a
rm
a
In
stit
u
te
o
f
Tec
h
n
o
l
o
g
y
,
P
u
n
e
.
He
r
re
se
a
rc
h
a
re
s
is
m
a
c
h
in
e
lea
rn
i
n
g
,
s
h
e
h
a
s
p
u
b
li
s
h
e
d
m
a
n
y
p
a
p
e
rs
in
n
a
ti
o
n
a
l
a
n
d
i
n
tern
a
ti
o
n
a
l
j
o
u
r
n
a
ls.
S
h
e
h
a
s
p
u
b
li
s
h
e
d
2
p
a
ten
ts.
S
h
e
h
a
s
to
tal
1
3
y
e
a
rs
o
f
e
x
p
e
rien
c
e
in
tea
c
h
in
g
.
S
h
e
p
u
b
li
sh
e
d
p
a
p
e
rs
in
UG
C
a
n
d
S
c
o
p
u
s
j
o
u
r
n
a
ls
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
p
ra
j
k
ta.sa
m
b
a
re
@g
m
a
il
.
c
o
m
.
Dr
.
Ra
shm
i
Ra
m
a
m
u
r
th
y
is
d
e
a
n
a
c
a
d
e
m
ics
/as
so
c
iate
p
ro
fe
ss
o
r
a
t
P
VPI
T
Ba
v
d
h
a
n
E
n
g
i
n
e
e
rin
g
c
o
ll
e
g
e
P
u
n
e
I
n
d
ia.
S
h
e
h
o
l
d
s
a
P
h
.
D
.
d
e
g
re
e
i
n
El
e
c
tri
c
a
l
a
n
d
El
e
c
tro
n
ics
En
g
i
n
e
e
rin
g
.
He
r
re
s
e
a
rc
h
a
re
a
s
a
re
Au
to
m
o
ti
v
e
El
e
c
tro
n
ics
,
P
o
we
r
El
e
c
tro
n
ics
a
n
d
Ve
h
icle
sa
fe
ty
.
Dr.
Ra
sh
m
i
h
a
s
fil
e
d
se
v
e
ra
l
p
a
ten
ts.
S
h
e
h
a
s
p
u
b
li
sh
e
d
e
i
g
h
t
re
se
a
rc
h
a
rti
c
les
in
in
tern
a
ti
o
n
a
l
a
n
d
n
a
t
io
n
a
l
j
o
u
r
n
a
ls
a
n
d
c
o
n
fe
re
n
c
e
s.
S
h
e
p
u
b
li
sh
e
d
p
a
p
e
rs
in
UG
C
a
n
d
S
c
o
p
u
s
jo
u
rn
a
ls.
S
h
e
h
a
s
1
7
y
e
a
rs
o
f
tea
c
h
in
g
a
n
d
re
se
a
rc
h
e
x
p
e
rien
c
e
in
t
h
e
field
o
f
a
u
to
m
o
ti
v
e
a
n
d
d
i
g
it
a
l
e
lec
tro
n
ics
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
d
e
a
n
a
c
a
d
e
m
ics
0
9
@g
m
a
il
.
c
o
m
.
Rid
d
h
i
Mi
r
a
jk
a
r
p
u
rsin
g
a
Do
c
t
o
r
o
f
P
h
il
o
so
p
h
y
d
e
g
re
e
in
C
o
m
p
u
ter
En
g
i
n
e
e
rin
g
fro
m
S
a
v
it
r
ib
a
i
P
h
u
l
e
P
u
n
e
Un
i
v
e
rsity
,
P
u
n
e
,
In
d
ia.
S
h
e
h
a
s
re
c
e
iv
e
d
h
e
r
m
a
ste
r’s
(M
.
E.
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
i
n
e
e
rin
g
)
d
e
g
re
e
fro
m
S
a
v
it
ri
b
a
i
P
h
u
le
P
u
n
e
Un
iv
e
rsit
y
,
P
u
n
e
in
2
0
1
6
.
S
h
e
is
wo
rk
i
n
g
a
s
a
n
a
ss
istan
t
p
r
o
fe
ss
o
r
i
n
I
n
fo
rm
a
t
io
n
Tec
h
n
o
lo
g
y
De
p
a
rtme
n
t
o
f
Vish
wa
k
a
rm
a
In
stit
u
te
o
f
I
n
fo
rm
a
ti
o
n
Tec
h
n
o
l
o
g
y
,
K
o
n
d
h
wa
(
b
k
),
P
u
n
e
.
S
h
e
is
h
a
v
i
n
g
1
0
y
e
a
rs
o
f
wo
rk
i
n
g
e
x
p
e
rien
c
e
.
H
e
r
a
re
a
o
f
in
tere
st
is
a
rti
ficia
l
in
telli
g
e
n
c
e
a
n
d
Co
m
p
u
ter
Ne
two
rk
s.
S
h
e
h
a
s
p
u
b
li
sh
e
d
o
v
e
r
2
0
p
a
p
e
rs
i
n
n
a
ti
o
n
a
l
a
n
d
in
t
e
rn
a
ti
o
n
a
l
c
o
n
fe
re
n
c
e
s
a
n
d
jo
u
r
n
a
ls.
S
h
e
is
a
li
fe
t
ime
m
e
m
b
e
r
o
f
I
S
TE
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
rid
d
h
i.
m
irajk
a
r@v
ii
t.
a
c
.
i
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
E
mp
o
w
er B
r
ea
s
tN
e
t:
b
r
ea
s
t c
a
n
ce
r
d
etec
tio
n
w
ith
tr
a
n
s
fer lea
r
n
in
g
…
(
V
a
is
h
a
li M.
Jo
s
h
i
)
1935
Dr
.
Ne
e
ta
N.
Th
u
n
e
is
a
ss
o
c
i
a
te
p
ro
fe
ss
o
r
a
t
M
a
ra
th
wa
d
a
c
o
ll
e
g
e
o
f
e
n
g
in
e
e
ri
n
g
P
u
n
e
,
sh
e
is
P
h
.
D
.
in
e
lec
tro
n
ics
fro
m
RTM
NU
Na
g
p
u
r
u
n
i
v
e
rsity
.
He
r
re
se
a
rc
h
a
re
a
s
VLS
I,
AI/M
L,
R
o
b
o
ti
c
s
,
a
n
d
Au
to
m
a
ti
o
n
.
Dr.
T
h
u
n
e
h
a
v
e
p
u
b
li
s
h
e
d
v
a
rio
u
s
p
a
p
e
rs
in
re
m
a
rk
a
b
le
in
d
e
x
in
g
j
o
u
r
n
a
ls
li
k
e
,
ES
CI,
W
o
S
,
S
c
o
p
u
s,
a
n
d
UG
C
c
a
re
.
S
h
e
p
u
b
li
sh
e
d
h
e
r
o
w
n
b
o
o
k
b
a
se
d
o
n
h
e
r
re
se
a
rc
h
a
re
a
.
Ha
v
i
n
g
v
a
rio
u
s
p
a
ten
ts
a
n
d
c
o
p
y
ri
g
h
ts.
M
a
jo
rly
sh
e
is
wo
r
k
i
n
g
o
n
in
d
u
stry
-
b
a
se
d
p
r
o
jec
ts
o
n
M
o
d
e
l
b
a
se
d
d
e
sig
n
wit
h
KPIT
,
P
u
n
e
a
n
d
a
s
a
d
a
ta
sc
ien
c
e
c
o
n
su
lt
a
n
t
wit
h
W
o
rk
o
p
o
li
s
.
He
r
a
re
a
o
f
in
tere
st
is
AI/M
L
a
n
d
a
u
t
o
m
a
ti
o
n
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
n
e
e
ta.t
h
u
n
e
@g
m
a
il
.
c
o
m
.
Dr
.
G
ita
n
ja
li
R.
S
h
in
d
e
h
a
s
o
v
e
ra
ll
1
5
y
e
a
rs
o
f
e
x
p
e
rien
c
e
,
p
re
se
n
tl
y
w
o
rk
i
n
g
a
s
h
e
a
d
a
n
d
a
ss
o
c
iate
p
ro
fe
ss
o
r
i
n
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ie
n
c
e
a
n
d
E
n
g
i
n
e
e
rin
g
(AI
a
n
d
M
L),
Vish
wa
k
a
rm
a
In
stit
u
te
o
f
I
n
fo
rm
a
ti
o
n
Tec
h
n
o
l
o
g
y
,
P
u
n
e
,
I
n
d
ia.
S
h
e
h
a
s
d
o
n
e
P
h
.
D.
in
Wi
re
les
s
Co
m
m
u
n
ica
ti
o
n
fro
m
C
M
I,
Aa
lb
o
r
g
Un
iv
e
rsity
,
Co
p
e
n
h
a
g
e
n
,
De
n
m
a
rk
o
n
Re
se
a
rc
h
P
ro
b
lem
S
tate
m
e
n
t
“
Clu
ste
r
F
ra
m
e
wo
rk
fo
r
I
n
tern
e
t
o
f
P
e
o
p
le,
T
h
in
g
s
a
n
d
S
e
rv
ice
s
”
–
P
h
.
D
.
a
wa
rd
e
d
o
n
8
t
h
M
a
y
2
0
1
8
.
S
h
e
o
b
tain
e
d
M
.
E.
(Co
m
p
u
ter
En
g
in
e
e
rin
g
)
d
e
g
re
e
fro
m
th
e
Un
iv
e
rsity
o
f
P
u
n
e
,
P
u
n
e
in
2
0
1
2
a
n
d
B
.
E.
(C
o
m
p
u
ter
En
g
i
n
e
e
rin
g
)
d
e
g
re
e
fro
m
th
e
Un
iv
e
rsity
o
f
P
u
n
e
,
P
u
n
e
in
2
0
0
6
.
S
h
e
h
a
s
re
c
e
iv
e
d
re
se
a
rc
h
f
u
n
d
in
g
f
o
r
th
e
p
ro
jec
t
“
Li
g
h
twe
ig
h
t
g
r
o
u
p
a
u
th
e
n
ti
c
a
ti
o
n
fo
r
I
o
T
”
b
y
S
P
P
U,
P
u
n
e
.
S
h
e
h
a
s
p
re
se
n
ted
a
re
se
a
r
c
h
a
rti
c
le
in
t
h
e
wo
rl
d
wire
l
e
ss
re
se
a
rc
h
fo
ru
m
(W
WRF
)
m
e
e
ti
n
g
,
Be
ij
in
g
Ch
i
n
a
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
g
i
tan
jali.
s
h
in
d
e
@v
ii
t.
a
c
.
in
.
Evaluation Warning : The document was created with Spire.PDF for Python.