T
E
L
KO
M
N
I
KA
T
e
lec
om
m
u
n
icat
ion
,
Com
p
u
t
i
n
g,
E
lec
t
r
on
ics
an
d
Cont
r
ol
Vol.
18
,
No.
4
,
Augus
t
2020
,
pp.
1784
~
1794
I
S
S
N:
1693
-
6930,
a
c
c
r
e
dit
e
d
F
ir
s
t
G
r
a
de
by
Ke
me
nr
is
tekdikti
,
De
c
r
e
e
No:
21/E
/KP
T
/2018
DO
I
:
10.
12928/
T
E
L
KO
M
NI
KA
.
v18i4.
14718
1784
Jou
r
n
al
h
omepage
:
ht
tp:
//
jour
nal.
uad
.
ac
.
id/
index
.
php/T
E
L
K
OM
N
I
K
A
E
ar
ly
d
e
t
e
c
t
io
n
of
b
r
e
ast
c
a
n
c
e
r
u
si
n
g m
am
m
ogr
a
p
h
y
image
s an
d
sof
t
w
a
r
e
e
n
gi
n
e
e
r
in
g p
r
o
c
e
ss
M
u
ayad
S
ad
ik
Cr
ooc
k
1
,
S
aj
a
Dhyaa
Kh
u
d
e
r
2
,
A
yad
E
s
h
o
Korial
3
,
S
ah
ar
S
al
m
an
M
ah
m
ood
4
1,
2,
3
Co
m
p
u
t
er
E
n
g
i
n
eer
i
n
g
D
e
p
art
me
n
t
,
U
n
i
v
ers
i
t
y
o
f
T
e
ch
n
o
l
o
g
y
,
Iraq
4
Ci
v
i
l
E
n
g
i
n
eer
i
n
g
D
e
p
art
me
n
t
,
U
n
i
v
ers
i
t
y
o
f
T
ec
h
n
o
l
o
g
y
,
Iraq
Ar
t
icle
I
n
f
o
AB
S
T
RA
CT
A
r
ti
c
le
h
is
tor
y
:
R
e
c
e
ived
Nov
29
,
2019
R
e
vis
e
d
M
a
r
26,
2020
Ac
c
e
pted
Apr
12,
2020
T
h
e
b
rea
s
t
can
cer
h
a
s
affect
ed
a
w
i
d
e
reg
i
o
n
o
f
w
o
me
n
as
a
p
art
i
cu
l
ar
cas
e.
T
h
eref
o
re,
d
i
ffere
n
t
re
s
earch
er
s
h
a
v
e
fo
c
u
s
e
d
o
n
t
h
e
e
arl
y
d
e
t
ec
t
i
o
n
o
f
t
h
i
s
d
i
s
eas
e
t
o
o
v
erco
me
i
t
i
n
effi
c
i
en
t
w
ay
.
In
t
h
i
s
p
a
p
er,
an
earl
y
b
reas
t
can
cer
d
et
ec
t
i
o
n
s
y
s
t
em
h
a
s
b
ee
n
p
r
o
p
o
s
e
d
b
a
s
ed
o
n
ma
mmo
g
ra
p
h
y
i
mag
es
.
T
h
e
p
r
o
p
o
s
e
d
s
y
s
t
em
a
d
o
p
t
s
d
eep
-
l
earn
i
n
g
t
ec
h
n
i
q
u
e
t
o
i
n
crea
s
e
t
h
e
acc
u
racy
o
f
d
et
ec
t
i
o
n
.
T
h
e
co
n
v
o
l
u
t
i
o
n
a
l
n
e
u
ral
n
e
t
w
o
rk
(C
N
N
)
m
o
d
e
l
i
s
co
n
s
i
d
ere
d
fo
r
p
rep
ar
i
n
g
t
h
e
d
a
t
as
e
t
s
o
f
t
ra
i
n
i
n
g
an
d
t
es
t
.
It
i
s
i
m
p
o
r
t
an
t
t
o
n
o
t
e
t
h
at
t
h
e
s
o
ft
w
are
en
g
i
n
eer
i
n
g
p
ro
ce
s
s
mo
d
el
h
a
s
b
een
ad
o
p
t
e
d
i
n
co
n
s
t
ru
c
t
i
n
g
t
h
e
p
r
o
p
o
s
e
d
al
g
o
ri
t
h
m.
T
h
i
s
i
s
t
o
i
n
crea
s
e
t
h
e
rel
i
ab
l
y
,
fl
ex
i
b
i
l
i
t
y
an
d
ex
t
en
d
i
b
i
l
i
t
y
o
f
t
h
e
s
y
s
t
em.
T
h
e
u
s
er
i
n
t
erface
s
o
f
t
h
e
s
y
s
t
em
are
d
es
i
g
n
e
d
as
a
w
eb
s
i
t
e
u
s
e
d
at
co
u
n
t
r
y
s
i
d
e
g
en
era
l
p
u
r
p
o
s
e
(G
P)
h
eal
t
h
cen
t
er
s
fo
r
earl
y
d
et
ec
t
i
o
n
t
o
t
h
e
d
i
s
ea
s
e
u
n
d
er
l
ack
i
n
g
i
n
s
p
ec
i
al
i
s
t
med
i
c
al
s
t
aff.
T
h
e
o
b
t
ai
n
ed
res
u
l
t
s
s
h
o
w
t
h
e
effi
c
i
en
c
y
o
f
t
h
e
p
ro
p
o
s
ed
s
y
s
t
em
i
n
t
erms
o
f
acc
u
racy
u
p
t
o
mo
re
t
h
an
9
0
%
an
d
d
ecreas
e
t
h
e
effo
r
t
s
o
f
med
i
ca
l
s
t
aff
as
w
el
l
as
h
el
p
i
n
g
t
h
e
p
a
t
i
en
t
s
.
A
s
a
co
n
cl
u
s
i
o
n
,
t
h
e
p
ro
p
o
s
ed
s
y
s
t
em
ca
n
h
el
p
p
at
i
e
n
t
s
b
y
ear
l
y
d
et
ec
t
i
n
g
t
h
e
b
rea
s
t
can
cer
at
far
p
l
aces
fro
m
h
o
s
p
i
t
a
l
an
d
referri
n
g
t
h
em
t
o
n
eare
s
t
s
p
ec
i
al
i
s
t
cen
t
er.
K
e
y
w
o
r
d
s
:
B
r
e
a
s
t
c
a
nc
e
r
De
e
p
-
lea
r
ning
P
a
tt
e
r
n
r
e
c
ognit
ion
S
of
twa
r
e
e
ng
inee
r
ing
W
e
bs
it
e
de
s
ign
Th
i
s
i
s
a
n
o
p
en
a
c
ces
s
a
r
t
i
c
l
e
u
n
d
e
r
t
h
e
CC
B
Y
-
SA
l
i
ce
n
s
e
.
C
or
r
e
s
pon
din
g
A
u
th
or
:
M
ua
ya
d
S
a
dik
C
r
ooc
k,
De
pa
r
tm
e
nt
of
C
omput
e
r
E
nginee
r
ing,
Unive
r
s
it
y
of
T
e
c
hnology
-
I
r
a
q,
Als
inaa
S
tr
e
e
t,
B
a
ghda
d,
I
r
a
q
.
E
mail:
M
ua
ya
d.
S
.
C
r
ooc
k@uotec
hnology.
e
du.
iq
1.
I
NT
RODU
C
T
I
ON
R
e
c
e
ntl
y,
the
b
r
e
a
s
t
c
a
nc
e
r
is
c
ons
ider
e
d
a
s
the
m
os
t
da
nge
r
ous
r
is
k
that
a
tt
a
c
ks
the
li
f
e
o
f
wome
n.
T
his
dis
e
a
s
e
is
the
r
e
s
ult
of
dif
f
e
r
e
nt
r
e
a
s
ons
,
s
uc
h
a
s
the
li
f
e
s
tyl
e
a
nd
inhe
r
it
a
nc
e
e
f
f
e
c
ts
.
T
he
de
tec
ti
on
of
thi
s
dis
e
a
s
e
is
ba
s
e
d
on
a
ll
oc
a
ti
ng
the
c
ha
nging
in
the
s
of
t
ti
s
s
ue
of
the
br
e
a
s
t
in
e
a
r
ly
leve
l.
X
-
r
a
y
ba
s
e
d
mammogr
a
phy
im
a
ge
s
is
nor
mally
a
dopted
f
or
br
e
a
s
t
c
a
nc
e
r
de
tec
ti
on.
T
he
s
e
im
a
ge
s
h
a
ve
be
e
n
take
n
in
dif
f
e
r
e
nt
a
ngles
to
c
ove
r
a
ll
pa
r
ts
of
the
dis
e
a
s
e
.
I
t
is
we
ll
known
that
the
X
-
r
a
y
im
a
ge
s
s
uf
f
e
r
s
f
r
om
low
c
ontr
a
s
ti
ng
due
to
low
vo
lum
e
of
r
a
diation
f
or
he
a
lt
h
r
e
a
s
on.
T
hus
,
dif
f
e
r
e
nt
methods
a
r
e
us
e
d
f
or
i
mpl
e
menting
the
im
a
g
e
e
nha
nc
e
ment
including
a
r
ti
f
icia
l
int
e
ll
i
ge
nt
s
tr
a
tegie
s
a
nd
de
e
p
lea
r
ning
[
1
,
2]
.
T
he
de
e
p
-
lea
r
ning
tec
hnology
ha
s
be
e
n
c
ons
ider
e
d
in
de
tec
ti
ng
of
di
f
f
e
r
e
nt
dis
e
a
s
e
s
.
I
n
thi
s
wor
k
,
we
a
dopt
the
c
onv
olut
ional
ne
ur
a
l
ne
twor
k
(
C
NN
)
ba
s
e
d
de
e
p
-
lea
r
ning
metho
d
f
or
de
tec
ti
ng
the
dis
e
a
s
e
.
I
t
include
s
the
pr
e
-
pr
oc
e
s
s
ing
s
tage
that
e
nha
nc
e
the
c
ons
ider
e
d
im
a
ge
s
to
inc
r
e
a
s
e
the
c
ontr
a
s
t
a
nd
how
the
ti
s
s
ue
s
if
br
e
a
s
t
c
lea
r
ly.
T
he
C
NN
is
us
e
d
f
or
c
ons
tr
uc
ti
ng
the
model
of
e
x
tr
a
c
ti
ng
the
f
e
a
tur
e
s
of
include
d
im
a
ge
s
.
T
he
s
e
f
e
a
tur
e
s
a
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
E
ar
ly
de
tec
ti
on
of
br
e
as
t
c
anc
e
r
us
ing
mam
mogr
aphy
image
s
and
…
(
M
uay
ad
Sadik
C
r
ooc
k
)
1785
us
e
d
to
buil
d
the
tr
a
ini
ng
da
tas
e
t
a
nd
de
tec
ti
ng
the
dis
e
a
s
e
of
tes
t
s
a
mpl
e
s
.
As
a
r
e
s
ult
,
the
pr
oc
e
s
s
ing
ti
me
is
r
e
duc
e
d
a
s
we
ll
in
e
f
f
icie
nt
wa
y
due
to
low
s
ize
of
unde
r
lyi
ng
im
a
ge
s
[
2
-
5]
.
As
mentioned
e
a
r
li
e
r
,
the
r
e
s
e
a
r
c
he
r
s
int
e
r
e
s
ted
i
n
pr
e
vious
wor
k
a
bout
de
tec
ti
ng
dif
f
e
r
e
nt
types
of
dis
e
a
s
e
us
ing
de
e
p
-
lea
r
ning.
I
t
ba
s
e
d
on
c
las
s
if
ying
thes
e
dis
e
a
s
e
s
int
o
c
las
s
e
s
ba
s
e
d
on
the
f
e
a
tur
e
s
of
unde
r
lyi
ng
im
a
ge
s
.
I
n
[
1]
,
a
ne
w
de
e
p
lea
r
ning
c
las
s
if
ier
ha
s
be
e
n
pr
opos
e
d
ba
s
e
d
on
digi
tal
mammogr
a
phy
im
a
ge
s
.
T
he
a
uthor
s
in
tr
oduc
e
d
a
c
las
s
if
ier
f
o
r
de
tec
ti
ng
the
tum
o
r
t
is
s
ue
s
,
in
a
ddit
ion
to
ov
e
r
c
omi
ng
the
pr
oblem
o
f
the
low
c
ont
r
a
s
t
im
a
ge
s
.
T
he
c
onto
ur
f
unc
ti
on
wa
s
us
e
d
ba
s
e
d
on
C
ha
n
-
Ve
s
e
leve
l
s
e
t
method.
M
or
e
ove
r
,
the
r
e
quir
e
d
f
e
a
tu
r
e
s
we
r
e
e
xtr
a
c
ted
us
ing
de
e
p
lea
r
ning
ba
s
e
d
C
NN
.
the
f
a
ls
e
r
e
s
ult
s
ha
ve
be
e
n
r
e
duc
e
d
by
a
dding
a
c
ompl
e
x
va
lued
r
e
laxa
ti
on
to
the
c
las
s
if
ier
,
while
the
a
c
c
ur
a
c
y
is
incr
e
a
s
e
d
up
to
99%
.
I
n
[
2
]
,
the
a
utho
r
s
pr
e
s
e
nted
a
method
of
lea
r
nin
g
f
or
a
f
e
a
tur
e
hie
r
a
r
c
h
y
o
f
unlabe
led
da
tas
e
t.
T
h
e
da
tas
e
t
wa
s
e
nter
e
d
to
the
c
las
s
if
ier
f
o
r
s
e
gmenting
th
e
br
e
a
s
t
de
ns
it
y
a
nd
s
c
or
ing
the
mammog
r
a
phy
textur
e
.
B
oth
of
li
f
e
ti
me
a
nd
population
s
pa
r
s
it
y
we
r
e
c
ons
ider
e
d
in
the
pr
opos
e
d
r
e
gular
ize
r
,
us
e
d
f
or
c
ontr
oll
ing
the
e
xten
dibi
li
ty
o
f
the
p
r
e
s
e
nted
model.
T
his
method
wa
s
pr
e
s
e
nted
to
e
a
s
e
the
im
pleme
ntat
ion
a
nd
the
obtaine
d
r
e
s
ult
s
e
ns
ur
e
d
the
high
a
c
c
ur
a
c
y.
I
n
[
3]
,
the
a
uthor
s
s
olved
the
pr
oblem
of
the
r
is
ky
de
ve
lopm
e
nt
of
thi
s
dis
e
a
s
e
,
a
ppe
a
r
e
d
in
the
inves
ti
ga
ted
im
a
g
e
s
us
ing
c
r
a
nio
-
c
a
uda
l
(
C
C
)
a
nd
mediolate
r
a
l
obli
que
(
M
L
O)
.
A
de
e
p
lea
r
ning
model
wa
s
us
e
d
f
or
tac
kli
ng
the
pr
oblem
of
unr
e
gis
ter
e
d
br
e
a
s
t
im
a
ge
s
a
nd
r
e
late
d
s
e
gmenta
ti
ons
.
T
he
s
e
pa
r
a
mete
r
s
c
a
n
a
f
f
e
c
t
the
pe
r
f
or
manc
e
o
f
the
p
r
opos
e
d
method
in
ba
d
wa
y.
T
he
a
ut
hor
s
of
[
4]
,
a
dopted
dif
f
e
r
e
nt
de
e
p
lea
r
ning
a
ppr
oa
c
he
s
f
or
de
tec
ti
ng
a
nd
inves
ti
ga
ted
o
f
br
e
a
s
t
c
a
nc
e
r
us
ing
ult
r
a
s
ound
s
e
s
s
ion.
T
he
a
ppr
oa
c
he
s
of
P
a
tch
-
ba
s
e
d
L
e
Ne
t,
a
U
-
Ne
t,
a
nd
a
t
r
a
ns
f
e
r
lea
r
ning
in
c
ombi
na
t
ion
with
a
pe
r
taine
d
F
C
N
-
Ale
xNe
t
ha
d
be
e
n
uti
li
z
e
d
f
o
r
a
c
h
ieving
the
objec
ti
ve
o
f
the
p
r
e
s
e
nted
a
ppr
oa
c
he
s
.
T
he
obtaine
d
r
e
s
ult
s
s
howe
d
the
high
a
c
c
ur
a
c
y
in
c
ompar
is
on
with
the
tr
a
dit
ional
methods
.
I
n
[
5]
,
a
tom
os
ynthes
is
c
las
s
if
ica
ti
on
method
wa
s
pr
opos
e
d
us
ing
C
NN
ba
s
e
d
de
e
p
lea
r
ning.
M
or
e
t
ha
n
3
00
mammogr
a
phy
im
a
ge
s
we
r
e
c
oll
e
c
ted
f
r
om
Unive
r
s
it
y
of
Ke
ntucky.
T
he
uti
li
z
e
d
of
de
e
p
lea
r
n
ing
wa
s
buil
t
to
de
s
ign
a
c
las
s
if
ier
f
or
wor
king
on
2D
a
nd
3D
im
a
g
e
s
.
T
he
a
c
hieve
d
r
e
s
ult
s
e
xplaine
d
the
s
upe
r
ior
pe
r
f
or
manc
e
of
the
pr
opos
e
d
method
.
I
n
[
6]
,
the
a
u
thor
s
int
r
oduc
e
d
a
r
e
view
r
e
s
e
a
r
c
h
wor
k
that
tac
kled
the
uti
li
z
e
d
tec
hniques
,
us
e
d
f
or
br
e
a
s
t
c
a
nc
e
r
de
tec
ti
on
us
ing
in
mammogr
a
phy
s
a
mpl
e
s
.
Dif
f
e
r
e
nt
types
of
ne
ur
a
l
models
we
r
e
r
e
view
e
d,
s
uc
h
a
s
the
hybr
id
a
da
ptation
in
br
e
a
s
t
c
a
nc
e
r
de
tec
ti
on.
I
n
a
ddit
i
on,
nume
r
ous
a
r
ti
f
icia
l
ne
ur
a
l
ne
twor
ks
we
r
e
uti
li
z
e
d
f
or
de
tec
ti
ng
a
nd
diagnos
in
g
the
br
e
a
s
t
c
a
nc
e
r
in
[
7
-
9]
.
T
he
pr
e
s
e
nted
a
ppr
oa
c
he
s
we
r
e
us
e
d
f
or
e
nha
nc
ing
the
mi
c
r
o
-
c
a
lcif
ica
ti
on
ba
s
e
d
on
il
lum
ination
a
nd
non
-
r
e
gular
it
y.
T
he
a
uthor
s
a
ll
oc
a
ted
the
inf
e
c
ted
a
r
e
a
s
us
ing
it
e
r
a
ti
ve
s
e
lec
ti
on
of
thr
e
s
hold
leve
l
method.
T
his
wa
s
done
by
r
e
buil
ding
t
he
s
ha
pe
of
im
a
ge
s
a
nd
r
e
movi
ng
the
r
e
dunda
nt
pixels
.
I
n
a
ddit
ion,
the
int
r
oduc
e
d
a
ppr
oa
c
he
s
e
xtr
a
c
ted
the
f
e
a
tur
e
s
of
thes
e
im
a
ge
s
f
or
de
tec
ti
ng
the
b
r
e
a
s
t
c
a
n
c
e
r
.
T
he
obtaine
d
r
e
s
ult
s
e
xpr
e
s
s
e
d
the
high
a
c
c
ur
a
c
y
of
pe
r
f
o
r
manc
e
in
c
ompar
is
on
with
the
pr
e
vious
a
ppr
oa
c
he
s
.
T
he
s
a
me
a
ppr
oa
c
h
wa
s
a
dopted
b
y
a
uthor
s
of
the
r
e
s
e
a
r
c
h
wor
k
of
[
10
-
17]
that
we
r
e
f
oc
us
e
d
on
the
de
e
p
lea
r
ning
tec
hniques
.
T
he
a
uthor
s
o
f
[
1
8
-
23]
tac
kled
the
pr
oblem
of
a
pplyi
ng
the
s
of
twa
r
e
e
ng
inee
r
ing
tec
hnology
in
c
oope
r
a
ti
on
with
the
de
e
p
lea
r
ning
tec
hnology.
M
os
t
of
the
pr
e
vious
wor
k
c
ons
ider
the
Glopa
l
P
os
it
ioni
ng
S
ys
tem
(
GPS
)
a
nd
we
b
a
ppli
c
a
ti
ons
to
f
inalize
the
outcome
p
r
oduc
ti
ons
,
pa
r
ti
c
ular
ly
in
a
l
loca
ti
on
ter
ms
[
24
-
26]
.
T
his
pa
pe
r
pr
e
s
e
nts
a
C
NN
ba
s
e
d
de
e
p
-
lea
r
ning
model
f
or
buil
ding
a
n
e
a
r
ly
br
e
a
s
t
c
a
nc
e
r
de
tec
ti
on
s
ys
tem.
T
he
pr
opos
e
d
s
ys
tem
us
e
s
the
digi
tal
ma
mm
ogr
a
phy
im
a
ge
s
a
f
ter
a
pplyi
ng
the
p
r
e
-
pr
oc
e
s
s
ing
s
tage
.
T
he
pr
opos
e
d
a
lgor
it
hm
of
de
tec
ti
ng
the
br
e
a
s
t
c
a
nc
e
r
ba
s
e
d
on
the
c
ha
nge
s
of
s
of
t
ti
s
s
ue
s
is
buil
t
ba
s
e
d
on
s
of
twa
r
e
e
nginee
r
ing
pr
oc
e
s
s
model.
T
his
model
ta
c
kles
the
pr
oblem
of
r
e
li
a
bil
it
y
,
f
lexibil
it
y
a
nd
e
xtendibil
it
y
of
the
de
s
igned
a
ppr
oa
c
h.
I
t
is
im
po
r
tant
to
no
te
t
ha
t
the
p
r
opos
e
d
s
ys
tem
a
dopts
a
we
bs
it
e
de
s
ign
f
or
e
a
s
ing
the
a
c
c
e
s
s
of
the
s
ys
tem
in
the
c
ount
r
y
s
ide
plac
e
s
.
T
his
s
ys
tem
is
de
s
igned
f
or
thes
e
plac
e
s
a
s
they
s
uf
f
e
r
f
r
om
lac
k
in
s
pe
c
ialis
t
medic
a
l
s
taf
f
.
T
he
r
e
f
o
r
e
,
the
s
ys
tem
c
a
n
de
tec
t
the
dis
e
a
s
e
in
e
a
r
ly
s
tage
s
f
r
om
th
e
im
a
ge
s
a
nd
r
e
f
e
r
r
ing
the
pa
ti
e
nt
to
the
c
e
ntr
a
l
hos
pit
a
ls
f
or
p
r
ovidi
ng
the
r
e
quir
e
d
tr
e
a
tm
e
nts
.
I
t
a
ls
o
c
a
n
r
e
duc
e
the
load
on
the
c
e
ntr
a
l
hos
pit
a
l
by
li
mi
tation
th
e
number
of
r
e
f
e
r
r
ing
c
a
s
e
s
.
T
he
obtaine
d
r
e
s
ult
s
s
how
the
e
f
f
icie
nc
y
of
the
pr
opos
e
d
s
ys
tem
in
ter
ms
o
f
a
c
c
ur
a
c
y
up
to
90%
,
r
e
duc
ing
the
load
on
c
e
ntr
a
l
hos
pit
a
ls
a
nd
s
a
ving
li
f
e
of
pa
ti
e
nts
.
2.
P
ROP
OS
E
D
S
Y
S
T
E
M
T
he
pr
opos
e
d
s
ys
tem
is
ba
s
e
d
on
de
s
igni
ng
a
n
e
lec
tr
onic
s
it
e
f
or
de
tec
ti
ng
the
br
e
a
s
t
c
a
nc
e
r
a
t
the
e
a
r
ly
s
tage
s
.
T
he
s
ys
tem
is
mana
ge
d
by
p
r
of
e
s
s
ional
Ge
ne
r
a
l
P
u
r
pos
e
s
(
GP)
he
a
lt
h
c
e
nter
s
a
t
the
c
ountr
y
s
ides
of
c
ount
r
ies
.
T
his
is
due
to
the
lac
k
i
n
s
pe
c
ialis
t
doc
tor
s
a
s
we
ll
a
s
r
e
duc
e
the
wa
it
ing
q
ue
ue
f
or
pa
ti
e
nts
a
t
the
c
e
ntr
a
l
b
r
e
a
s
t
c
a
nc
e
r
hos
pit
a
ls
.
T
hi
s
s
e
c
ti
on
is
divi
de
d
int
o
numer
ous
s
ubs
e
c
ti
ons
f
o
r
e
a
s
ing
the
r
e
a
ding
f
low
.
2.
1
.
S
ys
t
e
m
b
lock
d
iagram
F
ig
ur
e
1
i
ll
us
tr
a
tes
the
ge
ne
r
a
l
bloc
k
diag
r
a
m
of
the
pr
opos
e
d
s
ys
tem.
T
his
f
igur
e
e
xplains
the
wor
king
s
teps
of
the
pr
opos
e
d
s
ys
tem
in
ter
ms
of
us
e
r
a
nd
pr
of
e
s
s
ional
r
e
gis
tr
a
ti
on
a
nd
f
e
e
ding
the
pa
tent
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
4
,
Augus
t
2020
:
1784
-
1794
1786
im
a
ge
s
to
the
s
ys
tem
w
e
bs
it
e
.
T
he
s
e
r
ve
r
s
ide
of
the
s
ys
tem
pe
r
f
or
mi
ng
the
de
s
igned
de
e
p
-
lea
r
ning
model
f
or
diagnos
ing
the
br
e
a
s
t
c
a
nc
e
r
.
T
he
e
nter
e
d
im
a
ge
is
diagnos
e
d
to
inf
e
c
ted
or
non
-
inf
e
c
ted,
in
whic
h
th
e
pa
ti
e
nt
e
it
he
r
dis
c
ha
r
ge
d
f
r
om
the
GP
or
r
e
f
e
r
e
e
ing
to
c
e
nt
r
a
l
hos
pit
a
l.
F
igur
e
2
s
hows
the
de
s
igni
ng
model
of
the
pr
opos
e
d
de
e
p
-
lea
r
ning
ba
s
e
d
br
e
a
s
t
c
a
n
c
e
r
.
T
he
c
onvolut
ional
ne
ur
a
l
ne
twor
k
(
C
NN
)
is
a
dopt
e
d
f
or
p
r
oc
e
s
s
ing
the
matc
hing
a
nd
pr
e
pa
r
ing
the
tr
a
ini
ng
da
tas
e
t.
T
he
s
ys
tem
is
de
s
igned
ba
s
e
d
on
two
c
la
s
s
e
s
;
inf
e
c
ted
a
nd
non
-
inf
e
c
ted.
T
he
c
las
s
e
s
a
nd
a
ppoint
e
d
labe
ls
f
or
r
e
c
e
ivi
ng
da
ta
is
e
nter
e
d
to
the
de
e
p
-
lea
r
ning
model.
I
n
a
ddit
ion,
the
tr
a
ini
ng
im
a
ge
s
a
r
e
f
e
d
to
the
s
a
me
model
f
or
pe
r
f
o
r
mi
ng
the
tr
a
ini
ng
mo
de
l
us
ing
C
NN
.
T
he
outcome
t
r
a
ined
model
is
us
e
d
f
or
diagn
os
ing
the
tes
t
im
a
ge
s
int
o
in
f
e
c
ted
or
non
-
inf
e
c
ted.
F
igur
e
1.
Ge
ne
r
a
l
block
diagr
a
m
o
f
the
pr
opos
e
d
s
ys
tem
F
igur
e
2.
B
lock
diagr
a
m
of
the
pr
opos
e
d
de
e
p
-
lea
r
ning
model
2.
2
.
De
s
ign
e
d
s
of
t
war
e
e
n
gi
n
e
e
r
in
g
p
r
oc
e
s
s
m
od
e
l
T
he
s
of
twa
r
e
e
nginee
r
ing
pr
oc
e
s
s
model
is
a
dopted
in
de
s
igni
ng
the
pr
opos
e
d
a
lgor
it
hms
of
the
pr
e
s
e
nted
s
ys
tem.
T
he
r
e
a
s
on
be
hind
us
ing
the
tec
hnique
of
s
of
twa
r
e
e
nginee
r
ing
is
f
o
r
in
c
r
e
a
s
ing
the
r
e
li
a
bil
it
y
of
the
pr
opos
e
d
s
ys
tem
a
nd
taking
to
the
c
ons
i
de
r
a
ti
on
a
ny
f
utur
e
de
ve
lopm
e
nts
.
T
he
s
e
de
ve
lopm
e
nts
include
the
e
xpa
nda
bil
it
y
a
nd
f
lexibil
it
y
in
ter
ms
of
incr
e
a
s
ing
the
s
ize
of
invol
ve
d
GPs
a
nd
number
of
us
e
r
s
.
F
ig
u
r
e
3
e
xplains
the
de
s
igned
s
of
twa
r
e
e
nginee
r
ing
pr
oc
e
s
s
model,
us
e
d
f
or
c
on
s
tr
uc
ti
ng
t
he
pr
opos
e
d
a
lgo
r
it
hms
.
I
t
is
we
ll
s
hown
that
the
r
e
qui
r
e
ments
of
the
pr
opos
e
d
a
lgo
r
it
hms
p
lay
a
s
the
c
or
e
of
de
s
igni
ng
the
s
of
twa
r
e
e
nginee
r
ing
pr
oc
e
s
s
model.
T
he
f
i
r
s
t
pha
s
e
take
s
c
a
r
e
f
r
om
c
oll
e
c
ti
ng
thes
e
r
e
quir
e
ments
a
nd
c
las
s
if
ies
them
int
o
two
main
c
las
s
e
s
,
whic
h
a
r
e
inf
e
c
ted
a
nd
non
-
inf
e
c
ted.
W
hil
e
,
th
e
s
e
c
ond
pha
s
e
de
s
ign
s
the
ini
ti
a
l
ve
r
s
ion
of
the
pr
opos
e
d
a
l
gor
it
hm
c
ons
ider
ing
the
r
e
quir
e
ments
.
T
he
r
e
is
a
f
e
e
dba
c
k
be
twe
e
n
the
pha
s
e
one
a
nd
two
f
or
c
onf
i
r
mi
ng
that
the
ini
ti
a
l
de
s
ign
is
done
a
c
c
o
r
ding
to
the
r
e
quir
e
d
li
mi
tations
.
T
he
thi
r
d
pha
s
e
de
ve
lops
the
de
s
igned
a
lgor
it
hm
in
it
s
ini
ti
a
l
s
tep
to
r
e
c
ove
r
a
ny
d
r
a
wba
c
k
a
ppe
a
r
e
d
thr
oughout
the
c
ompl
e
ti
on
p
r
oc
e
s
s
.
T
he
f
inal
v
e
r
s
ion
of
the
de
s
igned
a
lgor
it
hm
is
im
pleme
nte
d
us
ing
the
de
e
p
-
lea
r
ning
m
e
thod.
T
he
im
planta
ti
on
is
e
va
l
ua
ted
by
tes
ti
ng
the
pr
opos
e
d
a
lgor
it
hm
ove
r
dif
f
e
r
e
nt
c
a
s
e
s
tudi
e
s
of
the
da
tas
e
t
that
include
s
im
a
ge
s
of
pa
ti
e
nts
.
2.
3
.
T
h
e
p
r
op
os
e
d
d
e
e
p
-
lear
n
in
g
a
lgorit
h
m
I
t
is
we
ll
known
that
the
de
e
p
-
lea
r
ning
model
is
ba
s
e
d
on
buil
d
ing
a
tr
a
ined
da
tas
e
t
us
ing
the
t
r
a
ini
ng
model
a
nd
diagnos
ing
the
tes
t
i
mage
s
us
ing
tes
t
m
ode
l.
2.
3.
1
.
T
r
ain
i
n
g
m
o
d
e
l
T
he
tr
a
ini
ng
model
us
e
s
the
pr
opos
e
d
t
r
a
ined
a
lgor
it
hm
that
c
a
n
be
s
umm
a
r
ize
d
a
s
s
teps
f
low:
−
Appointi
ng
the
a
dopted
c
las
s
e
s
a
nd
labe
li
ng
the
da
ta.
−
C
las
s
if
ying
the
tr
a
ini
ng
da
tas
e
t.
−
E
xtr
a
c
ted
the
a
dopted
f
e
a
tur
e
s
f
r
om
t
r
a
ini
ng
da
tas
e
t.
−
C
ons
tr
uc
ti
ng
the
gr
a
ph
o
f
C
NN
method
.
−
C
he
c
king
the
va
li
dit
y
of
c
las
s
e
s
a
nd
labe
led
da
tas
e
t.
−
Doing
the
pr
e
pr
oc
e
s
s
ing
ope
r
a
ti
ons
on
the
tr
a
ini
ng
im
a
ge
s
.
−
D
e
tec
ti
ng
a
ny
pos
s
ibl
e
dis
tor
ti
on
f
o
r
a
pplyi
ng
the
dis
tor
ti
ons
pr
oc
e
s
s
e
s
.
−
E
va
luating
the
'bot
tl
e
ne
c
k'
im
a
ge
f
o
r
pos
s
ibl
e
s
a
ving.
−
C
r
e
a
ti
ng
the
r
e
quir
e
d
pr
oc
e
s
s
ing
laye
r
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
E
ar
ly
de
tec
ti
on
of
br
e
as
t
c
anc
e
r
us
ing
mam
mogr
aphy
image
s
and
…
(
M
uay
ad
Sadik
C
r
ooc
k
)
1787
−
E
va
luate
the
a
c
c
ur
a
c
y
of
the
c
r
e
a
ted
laye
r
.
−
S
e
tt
ing
up
the
r
e
quir
e
d
we
ight
s
to
thei
r
in
it
ial
de
f
a
ult
va
lues
.
−
C
ons
tr
uc
ti
ng
the
r
e
quir
e
d
f
e
a
tur
e
s
in
it
e
r
a
ti
on
met
hod.
−
L
oc
a
ti
ng
the
input
bott
lene
c
k
va
lues
by
f
r
e
que
mt
ly
f
r
e
s
hing
with
dis
tor
ti
ons
,
or
f
e
tching
f
r
om
c
a
c
he
.
−
R
unning
a
tr
a
ini
ng
s
teps
.
−
C
a
ptur
ing
tr
a
ini
ng
s
umm
a
r
ies
f
or
T
e
ns
or
B
oa
r
d
wit
h
the
`
me
r
ge
d`
ope
r
a
ti
on
.
−
Ope
r
a
ti
ng
the
va
li
da
ti
on
s
tep
a
nd
s
tor
ing
int
e
r
medi
a
te
r
e
s
ult
s
.
−
W
r
it
ing
out
the
tr
a
ined
a
nd
labe
ls
da
tas
e
t.
F
igur
e
3.
De
s
igned
s
of
twa
r
e
e
nginee
r
ing
p
r
oc
e
s
s
model
2.
3.
2.
T
e
s
t
in
g
m
o
d
e
l
T
he
obtaine
d
tr
a
ini
ng
da
tas
e
t
f
r
om
the
tr
a
ined
mo
de
l,
the
tes
ted
im
a
ge
s
a
r
e
e
nter
e
d
to
the
s
ys
tem
f
o
r
diagnos
ing.
I
t
is
dr
a
wn
a
s
s
teps
f
low:
−
E
nter
ing
the
im
a
ge
f
il
e
−
F
e
e
ding
the
im
a
ge
int
o
the
loade
d
gr
a
ph
a
s
input
o
f
it
.
−
Obta
ini
ng
the
pr
e
diction
s
e
t
to
s
how
labe
ls
of
f
i
r
s
t
pr
e
diction
in
o
r
de
r
o
f
c
onf
idenc
e
.
−
Obta
ini
ng
the
r
e
s
ult
s
.
2.
4
.
T
h
e
p
r
op
os
e
d
GUI
a
n
d
algori
t
h
m
s
Vis
ua
l
S
tudi
o
C
ode
(
VSC
)
e
nvir
onment
is
uti
li
z
e
d
to
de
s
ign
a
nd
im
pleme
nt
the
GU
I
of
the
p
r
opos
e
d
s
ys
tem's
we
b
a
ppli
c
a
ti
on.
F
ig
ur
e
4
s
hows
the
ho
me
pa
ge
of
the
p
r
opos
e
d
we
b
a
ppli
c
a
ti
on
whic
h
pr
ovides
a
us
e
r
with
a
us
e
f
ul
li
nks
a
nd
inf
or
mation
a
bout
th
e
br
e
a
s
t
c
a
nc
e
r
.
T
his
pa
ge
a
ll
ows
the
a
uthor
ize
d
us
e
r
to
us
e
the
whole
f
unc
ti
ons
of
the
s
ys
tem
a
f
ter
the
log
in
pr
oc
e
s
s
done
s
uc
c
e
s
s
f
ul
ly.
T
h
e
m
a
i
n
p
r
o
c
e
s
s
o
f
t
h
i
s
p
a
g
e
i
s
t
o
e
n
a
b
l
e
t
h
e
a
u
t
h
o
r
i
z
e
d
u
s
e
r
s
t
o
t
a
k
e
a
d
v
a
n
t
a
g
e
o
f
a
l
l
s
y
s
t
e
m
a
c
t
i
v
i
t
i
e
s
a
f
t
e
r
t
h
e
l
o
g
i
n
p
r
o
c
e
s
s
c
o
m
p
l
e
t
e
s
c
o
r
r
e
c
t
l
y
.
N
e
w
o
r
u
n
r
e
g
i
s
t
e
r
e
d
u
s
e
r
c
a
n
t
a
k
e
a
d
v
a
n
t
a
ge
o
n
l
y
f
r
o
m
t
h
e
i
n
f
o
r
m
a
t
i
o
n
p
o
s
t
e
d
o
n
t
h
i
s
p
a
g
e
i
n
a
d
d
i
t
i
o
n
t
o
t
h
e
u
s
e
f
u
l
l
i
n
k
s
.
F
i
g
u
r
e
5
s
h
o
w
s
t
h
e
f
l
o
w
c
h
a
r
t
o
f
t
h
e
h
o
m
e
p
a
g
e
.
F
i
g
u
r
e
6
s
h
o
w
s
t
h
e
r
e
g
i
s
t
r
a
t
i
o
n
p
a
g
e
t
h
a
t
a
l
l
o
w
s
t
h
e
u
s
e
r
t
o
a
d
d
n
e
w
e
m
p
l
o
y
e
e
o
r
n
e
w
p
a
t
i
e
n
t
t
o
t
h
e
s
y
s
t
e
m
'
s
d
a
t
a
b
a
s
e
f
r
o
m
s
e
p
a
r
a
t
e
d
p
a
g
e
s
.
T
h
e
R
e
g
i
s
t
e
r
i
n
g
N
e
w
E
m
p
l
o
y
e
e
p
a
g
e
i
s
s
h
o
w
n
i
n
F
i
g
u
r
e
7
.
T
h
r
o
u
g
h
t
h
i
s
p
a
g
e
,
t
h
e
u
s
e
r
m
u
s
t
e
n
t
e
r
a
l
l
t
h
e
r
e
q
u
i
r
e
d
i
n
f
o
r
m
a
t
i
o
n
i
n
t
h
e
a
s
s
o
c
i
a
t
e
d
f
i
e
l
d
s
.
W
h
e
n
t
h
e
s
u
b
m
i
t
b
u
t
t
o
n
i
s
p
r
e
s
s
e
d
,
a
c
o
m
p
a
r
i
s
o
n
p
r
o
c
e
s
s
w
i
l
l
b
e
d
o
n
e
b
e
t
w
e
e
n
t
h
e
e
n
t
e
r
e
d
i
n
f
o
r
m
a
t
i
o
n
a
n
d
a
l
l
i
n
f
o
r
m
a
t
i
o
n
s
t
o
r
e
d
i
n
t
h
e
e
m
p
l
o
y
e
e
'
s
t
a
b
l
e
o
f
t
h
e
s
y
s
t
e
m
'
s
d
a
t
a
b
a
s
e
.
I
f
t
h
i
s
e
m
p
l
o
y
e
e
h
a
s
b
e
e
n
r
e
g
i
s
t
e
r
e
d
i
n
a
d
v
a
n
c
e
,
a
w
a
r
n
i
n
g
m
e
s
s
a
g
e
w
i
l
l
a
p
p
e
a
r
w
h
i
c
h
t
e
l
l
s
u
s
e
r
t
h
a
t
t
h
e
r
e
g
i
s
t
r
a
t
i
o
n
p
r
o
c
e
s
s
i
s
n
o
t
d
o
n
e
.
H
o
w
e
v
e
r
,
i
f
t
h
e
e
n
t
e
r
e
d
i
n
f
o
r
m
a
t
i
o
n
d
o
e
s
n
’
t
m
a
t
c
h
a
n
y
e
m
p
l
o
y
e
e
'
s
i
n
f
o
r
m
a
t
i
o
n
i
n
t
h
e
d
a
t
a
b
a
s
e
,
t
h
e
r
e
g
i
s
t
r
a
t
i
o
n
p
r
o
c
e
s
s
i
s
d
o
n
e
s
u
c
c
e
s
s
f
u
l
l
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
4
,
Augus
t
2020
:
1784
-
1794
1788
B
e
s
i
d
e
s
,
t
h
e
r
e
g
i
s
t
e
r
i
n
g
n
e
w
p
a
t
i
e
n
t
p
a
g
e
i
s
s
h
o
w
n
i
n
F
i
g
u
r
e
8
.
A
t
t
h
i
s
p
a
g
e
,
t
h
e
r
e
g
i
s
t
r
a
t
i
o
n
p
r
o
c
e
s
s
i
s
c
a
r
r
i
e
d
o
u
t
i
n
t
h
e
s
a
m
e
m
a
n
n
e
r
a
s
t
h
e
r
e
g
i
s
t
e
r
i
n
g
a
n
e
w
e
m
p
l
o
y
e
e
p
r
o
c
e
s
s
b
y
c
o
n
s
i
d
e
r
i
n
g
t
h
e
d
i
f
f
e
r
e
n
c
e
s
b
e
t
w
e
e
n
t
h
e
r
e
q
u
i
r
e
d
i
n
f
o
r
m
a
t
i
o
n
.
A
s
w
e
l
l
a
s
a
d
d
i
t
i
o
n
a
l
i
n
f
o
r
m
a
t
i
o
n
a
r
e
r
e
q
u
i
r
e
d
w
h
i
c
h
a
r
e
:
t
h
e
g
e
n
e
t
i
c
i
n
f
o
r
m
a
t
i
o
n
.
T
h
i
s
i
n
f
o
r
m
a
t
i
o
n
i
s
c
o
m
p
o
s
e
d
t
h
e
g
e
n
e
t
i
c
h
i
s
t
o
r
y
o
f
t
h
e
d
i
s
e
a
s
e
i
n
t
h
e
p
a
t
i
e
n
t
'
s
f
a
m
i
l
y
w
h
i
c
h
a
r
e
t
a
k
e
n
u
n
d
e
r
c
o
n
s
i
d
e
r
a
t
i
o
n
d
u
r
i
n
g
t
h
e
d
i
a
g
n
o
s
i
s
p
r
o
c
e
s
s
.
F
i
g
u
r
e
9
s
h
o
w
s
t
h
e
f
l
o
w
c
h
a
r
t
o
f
t
h
e
r
e
g
i
s
t
r
a
t
i
o
n
p
r
o
c
e
s
s
e
s
.
F
igur
e
4.
Home
pa
ge
F
igur
e
5.
F
lowc
ha
r
t
of
the
home
pa
ge
pr
oc
e
s
s
F
igur
e
6.
R
e
gis
tr
a
ti
on
pa
ge
F
igur
e
7.
R
e
gis
ter
ing
ne
w
e
mpl
oye
e
pa
ge
F
igur
e
8.
R
e
gis
ter
ing
ne
w
pa
ti
e
nt
pa
ge
F
i
g
u
r
e
1
0
s
h
o
w
s
t
h
e
d
i
a
g
n
o
s
i
s
p
a
ge
w
h
i
c
h
c
o
m
p
o
s
e
s
t
h
e
m
a
i
n
f
u
n
c
t
i
o
n
o
f
o
u
r
s
y
s
t
e
m
.
F
r
o
m
t
h
i
s
p
a
g
e
,
t
h
e
u
s
e
r
s
e
l
e
c
t
s
t
h
e
p
a
t
i
e
n
t
(
w
h
o
i
s
a
l
r
e
a
d
y
s
t
o
r
e
d
i
n
t
h
e
d
a
t
a
b
a
s
e
)
t
h
e
n
t
h
e
m
a
m
m
o
g
r
a
p
h
y
b
r
e
a
s
t
x
-
r
a
y
f
o
r
t
h
i
s
p
a
t
i
e
n
t
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
E
ar
ly
de
tec
ti
on
of
br
e
as
t
c
anc
e
r
us
ing
mam
mogr
aphy
image
s
and
…
(
M
uay
ad
Sadik
C
r
ooc
k
)
1789
w
i
l
l
u
p
l
o
a
d
e
d
t
o
t
h
e
s
y
s
t
e
m
t
o
b
e
m
a
n
i
p
u
l
a
t
e
d
u
s
i
n
g
o
u
r
m
o
d
e
l
.
T
h
e
m
o
d
e
l
o
u
t
p
u
t
s
t
h
e
r
e
s
u
l
t
w
h
i
c
h
w
i
l
l
e
i
t
h
e
r
i
n
f
e
c
t
e
d
o
r
u
n
i
n
f
e
c
t
e
d
.
I
f
t
h
e
u
n
i
n
f
e
c
t
e
d
r
e
s
u
l
t
i
s
s
h
o
w
n
,
t
h
e
r
i
s
k
f
a
c
t
o
r
f
o
r
t
h
i
s
p
a
t
i
e
n
t
w
i
l
l
b
e
c
a
l
c
u
l
a
t
e
d
.
O
t
h
e
r
w
i
s
e
,
i
f
t
h
e
i
n
f
e
c
t
e
d
r
e
s
u
l
t
i
s
s
h
o
w
n
,
a
d
r
o
p
-
d
o
w
n
l
i
s
t
w
i
l
l
a
p
p
e
a
r
t
h
a
t
a
l
l
o
w
s
u
s
e
r
t
o
s
e
l
e
c
t
a
n
y
p
r
e
-
r
e
g
i
s
t
e
r
e
d
h
o
s
p
i
t
a
l
t
h
a
t
t
h
e
r
e
s
u
l
t
w
i
l
l
s
e
n
d
t
o
F
i
g
u
r
e
1
1
s
h
o
w
s
t
h
e
f
l
o
w
c
h
a
r
t
o
f
t
h
e
d
i
a
g
n
o
s
i
s
p
r
o
c
e
s
s
.
W
h
i
l
e
,
F
i
g
u
r
e
1
2
s
h
o
w
s
t
h
e
r
e
p
o
r
t
i
n
g
p
a
g
e
t
h
a
t
p
r
o
v
i
d
e
t
h
e
u
s
e
r
w
i
t
h
a
w
h
o
l
e
i
n
f
o
r
m
a
t
i
o
n
a
b
o
u
t
t
h
e
e
m
p
l
o
y
e
e
s
,
p
a
t
i
e
n
t
s
a
n
d
i
n
f
e
c
t
e
d
a
n
d
u
n
i
n
f
e
c
t
e
d
p
a
t
i
e
n
t
s
.
T
h
i
s
p
r
o
c
e
s
s
i
s
d
o
n
e
w
h
e
n
u
s
e
r
c
l
i
c
k
s
o
n
t
h
e
c
o
r
r
e
s
p
o
n
d
i
n
g
b
u
t
t
o
n
a
s
s
h
o
w
n
i
n
t
h
e
F
i
g
u
r
e
1
3
.
M
o
r
e
o
v
e
r
,
a
c
o
n
t
a
c
t
p
a
g
e
w
a
s
p
r
o
v
i
d
e
d
i
n
o
u
r
s
y
s
t
e
m
i
n
o
r
d
e
r
t
o
a
l
l
o
w
u
s
e
r
f
o
r
s
e
n
d
i
n
g
a
n
y
m
e
s
s
a
g
e
t
o
u
s
us
i
n
g
h
i
s
e
m
a
i
l
a
s
s
h
o
w
n
i
n
F
i
g
u
r
e
1
4
.
T
o
c
o
m
p
l
e
t
e
t
h
i
s
p
r
o
c
e
s
s
,
t
h
e
u
s
e
r
m
u
s
t
f
i
l
l
h
i
s
n
a
m
e
a
n
d
a
c
o
r
r
e
c
t
e
m
a
i
l
a
t
t
h
e
c
o
r
r
e
s
p
o
n
d
i
n
g
f
i
e
l
d
s
a
s
r
e
q
u
i
r
e
d
.
W
h
e
n
t
h
e
m
e
s
s
a
g
e
i
s
s
e
n
t
s
u
c
c
e
s
s
f
u
l
l
y
,
a
c
o
n
f
i
r
m
m
e
s
s
a
g
e
w
i
l
l
b
e
s
e
n
t
t
o
t
h
e
s
e
n
d
e
r
.
F
i
g
u
r
e
1
5
s
h
o
w
s
t
h
e
f
l
o
w
c
h
a
r
t
o
f
t
h
e
c
o
n
t
a
c
t
u
s
p
r
o
c
e
s
s
.
F
igur
e
9.
F
lowc
ha
r
t
of
the
r
e
gis
tr
a
ti
on
pr
oc
e
s
s
e
s
F
igur
e
10.
Dia
gnos
is
pa
ge
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
4
,
Augus
t
2020
:
1784
-
1794
1790
F
i
g
u
r
e
1
1
.
F
l
o
w
c
h
a
r
t
o
f
t
h
e
d
i
a
g
n
o
s
i
s
p
r
o
c
e
s
s
F
igur
e
12.
R
e
por
ti
ng
pa
ge
F
igur
e
13.
F
lowc
ha
r
t
of
the
r
e
por
ti
ng
p
r
oc
e
s
s
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
E
ar
ly
de
tec
ti
on
of
br
e
as
t
c
anc
e
r
us
ing
mam
mogr
aphy
image
s
and
…
(
M
uay
ad
Sadik
C
r
ooc
k
)
1791
F
igur
e
14.
C
ontac
t
us
pa
ge
F
igur
e
15.
F
lowc
ha
r
t
of
the
c
ontac
t
us
p
r
oc
e
s
s
3.
E
XP
E
RM
E
NT
AL
RE
S
UL
T
S
AN
D
AN
AL
YS
I
S
T
he
pr
opos
e
d
s
ys
tem
is
tes
ted
ove
r
da
ta
s
e
t
of
500
im
a
ge
s
of
mammogr
a
phy
types
.
T
a
ble
1
e
xplains
the
c
las
s
if
ica
ti
on
of
the
uti
li
z
e
d
da
tas
e
t
ba
s
e
d
on
the
r
a
ti
os
of
e
a
c
h
c
a
tegor
y.
T
he
tes
ti
ng
s
e
t
c
a
tegor
y
r
e
pr
e
s
e
nts
30%
o
f
th
e
tot
a
l
da
tas
e
t.
W
hil
e
,
70%
of
the
da
tas
e
t
is
a
ll
oc
a
ted
a
s
tr
a
ini
ng
da
tas
e
t.
T
he
r
e
s
ult
s
a
r
e
obtaine
d
us
ing
HP
laptop
with
2.
4
GH
z
pr
oc
e
s
s
or
,
4GB
R
AM
s
uppor
ted
with
de
dica
ted
dis
play
a
da
pter
of
(
2
GB
)
a
nd
unde
r
ope
r
a
ti
ng
s
ys
tem
of
W
indows
10
pr
o
.
W
it
h
t
he
s
e
s
pe
c
if
ica
ti
ons
,
the
pr
opos
e
d
method
is
r
un
in
e
f
f
icie
nt
wa
y
with
pr
oc
e
s
s
ing
ti
me
up
to
ha
lf
hour
f
r
om
the
ini
ti
a
l
point
.
T
he
r
e
s
ult
s
a
r
e
divi
de
d
int
o
t
wo
pa
r
ts
:
de
e
p
-
lea
r
ning
a
nd
we
bs
it
e
.
T
he
de
e
p
-
lea
r
ning
r
e
s
ult
s
e
xplain
the
pe
r
f
o
r
manc
e
o
f
the
pr
opos
e
d
a
lgor
it
hm
with
the
a
dopted
da
tas
e
t.
W
hil
e
,
the
we
bs
it
e
r
e
s
ult
s
s
ho
w
the
be
ha
vior
of
the
pr
opos
e
d
s
ys
tem
with
the
tes
t
ing
c
a
s
e
s
that
r
e
quir
e
s
f
r
om
the
s
ys
tem
to
diagnos
e
the
in
f
e
c
ti
on.
3.
1.
De
e
p
-
lear
n
in
g
r
e
s
u
lt
s
F
ig
ur
e
16
s
hows
the
c
omput
e
d
a
c
c
ur
a
c
y
of
tr
a
i
ning
pr
oc
e
s
s
.
T
h
is
a
c
c
ur
a
c
y
is
c
a
lcula
ted
f
r
om
the
e
nter
e
d
tr
a
ini
ng
da
tas
e
t.
I
t
is
view
e
d
f
r
om
thi
s
f
igur
e
that
the
a
c
c
ur
a
c
y
is
im
pr
ove
d
with
the
inc
r
e
a
s
ing
of
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
4
,
Augus
t
2020
:
1784
-
1794
1792
a
dopted
s
teps
.
T
his
is
be
c
a
us
e
of
the
e
nlar
ging
i
n
the
s
tor
e
d
da
tas
e
t
of
t
r
a
ini
ng
a
nd
f
e
a
tur
e
s
f
r
o
m
C
NN
.
W
he
n
the
s
ys
tem
c
r
os
s
e
s
the
s
tep
of
2000,
the
a
c
c
ur
a
c
y
r
e
a
c
he
s
a
r
a
ti
o
of
100%
.
I
t
is
c
onc
luded
f
r
om
th
is
f
igur
e
that
the
tr
a
ini
ng
pr
oc
e
s
s
a
c
hieve
s
th
e
a
c
c
e
ptabl
e
r
a
ti
o
of
the
r
e
quir
e
d
a
c
c
ur
a
c
y.
T
he
a
c
c
ur
a
c
y
is
a
dopted
a
s
im
por
tant
f
a
c
tor
f
or
e
va
luating
the
e
f
f
icie
nc
y
of
the
p
r
opos
e
d
de
e
p
-
lea
r
ning
a
lgor
it
hm.
M
or
e
ove
r
,
the
pr
e
-
pr
oc
e
s
s
ing
ope
r
a
ti
ons
,
pe
r
f
or
med
in
ter
ms
of
im
a
ge
pr
oc
e
s
s
ing,
e
nha
nc
e
the
in
it
ial
im
a
ge
s
to
be
r
e
a
dy
f
or
f
e
a
tur
e
e
xt
r
a
c
ti
on
in
C
NN
tr
a
ined
model
.
T
his
is
to
incr
e
a
s
e
the
e
f
f
icie
nc
y
of
the
p
r
opos
e
d
s
ys
te
m
a
s
the
nois
e
d
a
nd
blur
r
y
im
a
ge
s
c
a
n
a
f
f
e
c
t
the
pe
r
f
o
r
manc
e
,
ha
r
s
hly.
I
n
o
r
de
r
to
tes
t
the
va
li
da
ti
on
a
c
c
ur
a
c
y
of
the
pr
opo
s
e
d
m
e
thod,
F
ig
ur
e
17
de
s
c
r
ibes
thi
s
va
li
da
ti
on
a
s
a
r
e
s
ult
of
de
tec
ti
ng
the
br
e
a
s
t
c
a
nc
e
r
o
f
the
tes
ti
ng
da
tas
e
t.
T
his
f
igur
e
p
r
ove
s
the
high
va
li
da
ti
on
of
th
e
r
e
s
ult
s
of
the
p
r
opos
e
d
method
in
tr
a
ini
ng
a
nd
tes
ti
ng
pha
s
e
s
.
F
igur
e
9
s
hows
the
va
li
da
ti
on
a
c
c
ur
a
c
y
of
a
lm
os
t
90%
a
t
the
pr
oc
e
s
s
ing
s
tep
2000
a
nd
ove
r
.
I
t
is
highl
igh
ted
f
r
om
thi
s
f
igur
e
that
the
a
c
c
ur
a
c
y
is
va
r
ied
f
r
o
m
50%
a
t
the
lowe
r
p
r
oc
e
s
s
ing
s
tep
a
nd
r
e
a
c
he
d
up
to
90%
ove
r
s
tep
2000.
T
he
va
li
da
ti
on
a
c
c
ur
a
c
y
is
be
ing
in
the
a
c
c
e
ptable
leve
l
a
f
ter
s
tep
2000
f
or
the
s
a
me
r
e
a
s
ons
of
incr
e
a
s
ing
the
t
r
a
ini
ng
a
c
c
ur
a
c
y
a
nd
r
e
duc
ing
the
c
r
os
s
e
ntr
opy.
As
a
r
e
s
ult
o
f
the
tes
ti
ng
ou
tcome
,
the
pr
opos
e
d
method
pr
ove
s
it
s
e
f
f
icie
nc
y
in
ter
ms
of
tr
a
ini
ng
a
c
c
ur
a
c
y,
c
r
os
s
e
ntr
opy
a
nd
va
li
da
ti
on
a
c
c
ur
a
c
y.
Althoug
h,
the
c
oll
e
c
ted
da
tas
e
t
is
not
pr
e
p
a
r
e
d
f
or
c
omput
e
r
pr
ogr
a
mm
ing
us
e
,
the
pr
e
p
r
oc
e
s
s
ing
f
unc
ti
ons
pe
r
f
or
med
by
the
pr
opos
e
d
method
r
e
duc
e
s
thes
e
e
f
f
e
c
ts
to
ve
r
y
mi
nim
um
va
lue
of
e
r
r
o
r
r
a
ti
o
.
T
a
ble
1.
T
he
c
las
s
if
ica
ti
ons
of
da
tas
e
t
D
a
ta
s
e
t
No
C
la
s
s
if
ic
a
ti
on
R
a
ti
o
I
nf
e
c
te
d
N
on
-
I
nf
e
c
te
d
T
e
s
ti
ng s
e
t
150
70
80
30%
T
r
a
in
in
g s
e
t
350
200
150
70%
T
ot
a
l
da
ta
s
e
t
500
270
230
100%
0
2
0
0
0
4
0
0
0
6
0
0
0
2
0
4
0
6
0
8
0
1
0
0
T
a
i
n
n
i
n
g
A
c
c
u
r
a
c
y
S
t
e
p
T
r
a
i
n
a
c
c
u
r
a
c
y
0
2
0
0
0
4
0
0
0
6
0
0
0
0
2
0
4
0
6
0
8
0
1
0
0
V
a
l
i
d
a
t
i
o
n
A
c
c
u
r
a
c
y
S
t
e
p
V
a
l
i
d
a
t
i
o
n
A
c
c
u
r
a
c
y
F
igur
e
16.
T
r
a
ini
ng
a
c
c
ur
a
c
y
F
igur
e
17.
T
he
c
omput
e
d
va
li
da
ti
on
a
c
c
ur
a
c
y
3.
2.
We
b
s
it
e
r
e
s
u
lt
s
T
hr
oughout
the
s
ys
tem
ope
r
a
ti
ng,
F
ig
u
r
e
18
e
xplains
the
tes
t
r
e
s
ult
s
of
the
whole
s
y
s
tem
a
s
a
we
bs
it
e
r
e
pr
e
s
e
ntation.
I
n
thi
s
f
igur
e
,
the
tes
t
s
a
mpl
e
,
w
hich
is
the
mammogr
a
phy
im
a
ge
,
ha
s
be
e
n
s
ubmi
tt
e
d
to
the
pr
opos
e
d
s
ys
tem
a
nd
the
obtaine
d
r
e
s
ult
s
s
how
the
pa
ti
e
nt
is
inf
e
c
ted
.
T
he
s
e
r
e
s
ult
s
a
r
e
a
c
h
ieve
d
by
s
e
lec
ti
ng
the
diagnos
is
butt
on.
Nor
mally
,
the
pr
op
os
e
d
s
ys
tem
r
e
f
e
r
e
e
s
the
pa
ti
e
nt
to
the
s
pe
c
ial
he
a
l
th
c
e
nter
a
t
the
big
hos
pit
a
l
f
or
ne
xt
s
tep
of
tr
e
a
tm
e
nts
.
T
he
C
L
E
AR
butt
on
is
us
e
d
f
or
e
r
a
s
ing
the
r
e
s
ult
s
a
nd
looki
ng
f
or
the
ne
xt
c
a
s
e
s
tudy.
At
the
other
s
ide,
F
ig
u
r
e
19
s
hows
the
s
ys
t
e
m
r
e
s
ult
s
of
uninf
e
c
ted
c
a
s
e
.
I
n
thi
s
f
igur
e
,
the
mammogr
a
phy
im
a
ge
of
a
pa
ti
e
nt
is
s
ubmi
tt
e
d
to
the
pr
opos
e
d
s
ys
tem
f
o
r
tes
ti
ng
a
nd
dia
gnos
ing.
T
he
obtaine
d
r
e
s
ult
s
s
how
that
the
pa
ti
e
nt
is
not
inf
e
c
ted
,
but
the
other
f
a
c
tor
is
the
r
is
k.
T
he
r
i
s
k
f
a
c
tor
e
xpr
e
s
s
e
s
the
pr
oba
bil
it
y
o
f
in
f
e
c
ti
on
f
o
r
pa
ti
e
nts
.
I
t
c
a
n
be
e
va
luate
d
a
s
:
=
(
.
×
)
+
(
.
×
)
+
(
.
×
)
(
1)
whe
r
e
,
ℎ
is
the
mot
he
r
inf
e
c
ti
on
,
it
is
e
it
he
r
1
o
r
0
.
is
the
gr
a
ndmot
he
r
in
f
e
c
ti
on,
i
t
is
e
it
he
r
1
o
r
0
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
E
ar
ly
de
tec
ti
on
of
br
e
as
t
c
anc
e
r
us
ing
mam
mogr
aphy
image
s
and
…
(
M
uay
ad
Sadik
C
r
ooc
k
)
1793
is
the
s
is
ter
inf
e
c
ti
on,
it
is
e
it
he
r
1
or
0.
is
the
number
of
inf
e
c
ted
s
is
ter
s
.
I
t
is
im
por
tant
to
note
that
the
r
is
k
f
a
c
tor
is
a
n
indi
c
a
ti
on
to
moni
tor
the
s
ti
ll
not
in
f
e
c
ted
pa
ti
e
nt
with
inher
it
e
d
c
a
s
e
s
including
mot
he
r
,
g
r
a
ndmot
he
r
a
nd
s
is
ter
s
.
W
e
give
a
high
r
a
ti
o
r
is
k
to
the
in
f
e
c
ted
mot
he
r
a
n
d
les
s
f
or
other
s
.
T
his
is
f
or
inher
it
a
nc
e
r
e
a
s
ons
.
F
igur
e
18.
W
e
bs
it
e
r
e
s
ult
s
of
inf
e
c
ted
pa
ti
e
nt
F
igur
e
19.
W
e
bs
it
e
r
e
s
ult
s
of
un
inf
e
c
ted
pa
ti
e
nt
a
nd
r
is
k
f
a
c
tor
4.
CONC
L
USI
ON
I
n
thi
s
pa
pe
r
,
a
n
e
a
r
ly
de
tec
ti
on
of
br
e
a
s
t
c
a
nc
e
r
s
ys
tem
ba
s
e
d
on
mammogr
a
phy
im
a
ge
s
wa
s
pr
opos
e
d.
T
he
pr
opos
e
d
a
lgor
it
hm
wa
s
f
or
mul
a
te
d
de
pe
nding
on
the
s
of
twa
r
e
e
nginee
r
ing
model
t
o
gr
a
ntee
the
s
c
a
labili
ty,
f
lexibil
it
y
a
nd
r
e
li
a
bil
it
y
.
T
he
d
e
e
p
-
lea
r
ning
tec
hnology
ha
s
be
e
n
uti
l
ize
d
f
o
r
d
e
tec
ti
ng
the
c
ha
nge
s
in
the
s
of
t
ti
s
s
ue
s
a
t
the
inves
ti
ga
ted
mammogr
a
phy
im
a
ge
s
.
T
he
pr
opos
e
d
s
ys
tem
a
dopted
a
we
bs
it
e
f
o
r
GU
I
de
s
ign.
T
he
we
bs
it
e
a
ll
owe
d
the
doc
tor
s
a
nd
pa
ti
e
nts
to
a
c
c
e
s
s
the
s
ys
tem
r
e
ga
r
dles
s
the
di
s
tanc
e
s
a
nd
plac
e
s
.
At
the
other
ha
nd,
the
pr
opos
e
d
s
ys
t
e
m
c
ons
ider
e
d
the
c
omput
ing
of
r
is
k
f
a
c
tor
of
uninf
e
c
ted
pa
ti
e
nts
.
T
his
r
is
k
f
a
c
tor
of
f
e
r
e
d
a
moni
t
or
ing
indi
c
a
tor
f
o
r
pa
ti
e
nts
unde
r
r
is
k.
T
he
pr
opos
e
d
s
ys
tem
wa
s
tes
ted
in
two
c
a
tegor
ies
.
T
he
f
ir
s
t
one
tes
ted
the
a
c
c
ur
a
c
y
of
the
de
s
igned
de
e
p
-
lea
r
ning
a
lgor
it
hm.
W
hil
e
the
other
one
c
ons
ider
e
d
whole
s
ys
tem
tes
ti
ng
r
e
pr
e
s
e
nti
ng
a
s
we
bs
it
e
r
e
s
ult
s
.
T
he
obtaine
d
r
e
s
ult
s
s
howe
d
the
e
f
f
icie
nc
y
of
the
pr
opos
e
d
s
ys
tem
in
ter
ms
of
a
c
c
ur
a
c
y
a
nd
e
a
r
ly
de
tec
ti
on
o
f
b
r
e
a
s
t
c
a
nc
e
r
.
RE
F
E
RE
NC
E
S
[1
]
Saras
w
a
t
h
i
D
u
rai
s
amy
,
Sr
i
n
i
v
a
s
an
E
m
p
eru
ma
l
,
"
Co
mp
u
t
er
-
ai
d
e
d
mammo
g
ram
d
i
a
g
n
o
s
i
s
s
y
s
t
em
u
s
i
n
g
d
eep
l
ear
n
i
n
g
co
n
v
o
l
u
t
i
o
n
a
l
fu
l
l
y
co
mp
l
e
x
-
v
a
l
u
e
d
rel
a
x
at
i
o
n
n
e
u
ral
n
et
w
o
r
k
cl
a
s
s
i
fi
er
,
"
IE
T
Co
m
p
u
t
e
r
V
i
s
i
o
n
,
v
o
l
.
1
1
,
n
o
.
8
,
p
p
.
6
5
6
-
6
6
2
,
2
0
1
7
.
[2
]
Mi
ch
i
el
K
a
l
l
e
n
b
er
g
,
K
ers
t
en
Pe
t
ers
e
n
,
Mad
s
N
i
el
s
en
,
A
n
d
rew
Y
.
N
g
.
,
Pen
g
fe
i
D
i
a
o
,
Ch
ri
s
t
i
an
I
g
el
,
Cel
i
n
e
M.
V
ach
o
n
,
K
a
t
h
a
ri
n
a
H
o
l
l
an
d
,
Ri
k
k
e
Ra
s
s
W
i
n
k
e
l
,
N
i
co
K
ars
s
emei
j
er,
an
d
Mart
i
n
L
i
l
l
h
o
l
m,
"
U
n
s
u
p
erv
i
s
ed
d
eep
l
earn
i
n
g
ap
p
l
i
ed
t
o
b
rea
s
t
d
en
s
i
t
y
s
eg
me
n
t
a
t
i
o
n
a
n
d
ma
mmo
g
ra
p
h
i
c
ri
s
k
s
c
o
ri
n
g
,
"
IE
E
E
T
r
a
n
s
a
ct
i
o
n
s
o
n
M
e
d
i
ca
l
Im
a
g
i
n
g
,
v
o
l
.
3
5
,
n
o
.
5
,
p
p
.
1
3
2
2
-
1
3
3
1
,
2
0
1
8
.
[3
]
G
u
s
t
a
v
o
Carn
e
i
ro
,
J
ac
i
n
t
o
N
a
s
ci
me
n
t
o
,
an
d
A
n
d
rew
P.
Brad
l
e
y
,
"
A
u
t
o
mat
e
d
A
n
a
l
y
s
i
s
o
f
U
n
reg
i
s
t
ered
Mu
l
t
i
-
V
i
e
w
Mammo
g
ram
s
w
i
t
h
D
eep
L
earn
i
n
g
,
"
IE
E
E
Tr
a
n
s
ect
i
o
n
s
o
n
M
e
d
i
c
a
l
Im
a
g
i
n
g
,
v
o
l
.
3
6
,
n
o
.
1
1
,
p
p
.
2
3
5
5
-
2
3
6
5
,
2
0
1
7
.
[4
]
Mo
i
H
o
o
n
Y
a
p
,
G
erard
Po
n
s
,
J
o
an
Mar
t
´ı
,
Serg
i
G
an
a
u
,
Mel
ci
o
r
Sen
t
´ı
s
,
Rey
er
Z
w
i
g
g
e
l
aar,
A
d
ri
a
n
K
.
D
a
v
i
s
o
n
,
an
d
Ro
b
er
t
Mart
´,
"
A
u
t
o
mat
e
d
Breas
t
U
l
t
ras
o
u
n
d
L
es
i
o
n
s
D
e
t
ect
i
o
n
U
s
i
n
g
Co
n
v
o
l
u
t
i
o
n
a
l
N
eu
ra
l
N
et
w
o
r
k
s
,
"
IE
E
E
Jo
u
r
n
a
l
o
f
B
i
o
m
ed
i
ca
l
a
n
d
H
e
a
l
t
h
I
n
f
o
r
m
a
t
i
c
s
,
v
o
l
.
2
2
,
n
o
.
4
,
p
p
.
1
2
1
8
-
1
2
2
6
,
2
0
1
8
.
[5
]
X
i
a
o
fei
Z
h
a
n
g
,
Y
i
Z
h
an
g
,
E
ri
k
Y
.
H
a
n
,
N
a
t
h
a
n
J
aco
b
s
,
Q
i
o
n
g
H
an
,
X
i
ao
q
i
n
W
a
n
g
,
J
i
n
ze
L
i
u
,
"
Cl
as
s
i
f
i
ca
t
i
o
n
o
f
W
h
o
l
e
Mammo
g
ram
an
d
T
o
m
o
s
y
n
t
h
e
s
i
s
Imag
es
U
s
i
n
g
D
eep
Co
n
v
o
l
u
t
i
o
n
a
l
N
eu
ra
l
N
et
w
o
r
k
s
,
"
IE
E
E
Tr
a
n
s
a
ct
i
o
n
s
o
n
Na
n
o
B
i
o
s
c
i
en
ce
,
v
o
l
.
1
7
,
n
o
.
3
,
p
p
.
2
3
7
-
2
4
2
,
2
0
1
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.