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.
2
,
Apr
il
2020
,
pp.
928
~
935
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
.
v18i2.
14085
928
Jou
r
n
al
h
omepage
:
ht
tp:
//
jour
nal.
uad
.
ac
.
id/
index
.
php/T
E
L
K
OM
N
I
K
A
Qu
al
itat
iv
e
asse
ssm
e
n
t
o
f
i
m
a
ge
e
n
h
a
n
c
e
m
e
n
t
al
gor
i
t
h
m
s
f
or
m
am
m
og
r
am
s b
as
e
d
o
n
m
in
i
m
u
m
E
D
V
M
az
in
N.
F
ar
h
an
1
,
M
oh
am
m
e
d
G
.
Ayoub
2
,
Has
s
an
M
.
Qas
s
im
3
,
A
b
d
u
lrahm
a
n
K
.
E
e
s
e
e
4
1,
3,
4
D
ep
ar
t
men
t
o
f
Med
i
cal
In
s
t
r
u
men
t
at
i
o
n
T
ec
h
n
o
l
o
g
y
E
n
g
i
n
eeri
n
g
,
N
o
rt
h
ern
T
ec
h
n
i
cal
U
n
i
v
er
s
i
t
y
,
Iraq
2
D
ep
ar
t
men
t
o
f
Co
m
p
u
t
er
T
ech
n
o
l
o
g
y
E
n
g
i
n
eeri
n
g
,
N
o
r
t
h
er
n
T
ech
n
i
ca
l
U
n
i
v
ers
i
t
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
S
e
p
12
,
2019
R
e
vis
e
d
J
a
n
2
,
2020
Ac
c
e
pted
F
e
b
19
,
2020
Breas
t
ca
n
cer
i
s
o
n
e
o
f
t
h
e
l
ead
i
n
g
reas
o
n
o
f
d
e
at
h
am
o
n
g
w
o
men
.
N
ev
er
t
h
e
l
es
s
,
med
i
cat
i
o
n
s
fo
r
t
h
i
s
fat
a
l
d
i
s
eas
e
are
s
t
i
l
l
aw
ay
o
f
amb
i
t
i
o
n
s
.
Pat
i
e
n
t
s
(t
h
o
u
g
h
t
t
o
h
av
e
b
reas
t
can
cer)
s
h
o
u
l
d
g
o
t
h
ro
u
g
h
s
ev
era
l
ad
v
an
ce
d
med
i
ca
l
d
i
a
g
n
o
s
t
i
c
p
ro
ce
d
u
re
s
l
i
k
e
ma
mm
o
g
ra
p
h
y
,
b
i
o
p
s
y
an
al
y
s
i
s
,
u
l
t
ras
o
u
n
d
i
mag
i
n
g
,
et
c.
Mamm
o
g
ra
p
h
y
i
s
o
n
e
o
f
t
h
e
med
i
cal
i
ma
g
i
n
g
t
ech
n
i
q
u
e
s
u
s
e
d
fo
r
d
et
ec
t
i
n
g
b
rea
s
t
ca
n
cer.
H
o
w
e
v
er,
i
t
s
res
u
l
t
ed
i
ma
g
es
may
n
o
t
b
e
cl
ear
en
o
u
g
h
o
r
h
e
l
p
f
u
l
fo
r
p
h
y
s
i
ci
an
t
o
d
i
ag
n
o
s
e
each
cas
e
co
rrect
l
y
.
T
h
i
s
fact
h
as
p
u
s
h
e
d
res
earch
er
s
t
o
w
ard
s
d
ev
e
l
o
p
i
n
g
effect
i
v
e
w
ay
s
t
o
en
h
an
ce
i
mag
e
s
t
h
ro
u
g
h
o
u
t
u
s
i
n
g
v
ar
i
o
u
s
en
h
a
n
ceme
n
t
al
g
o
ri
t
h
ms
.
In
t
h
i
s
p
a
p
er,
a
co
mp
ari
s
o
n
amo
n
g
s
t
t
h
es
e
ap
p
l
i
e
d
al
g
o
ri
t
h
m
s
w
as
d
o
n
e
t
o
ev
al
u
a
t
e
t
h
e
o
p
t
i
m
u
m
en
h
a
n
cemen
t
t
ech
n
i
q
u
e.
A
mo
rp
h
o
l
o
g
y
e
n
h
a
n
cemen
t
,
w
h
i
c
h
i
s
res
u
l
t
e
d
fro
m
co
m
b
i
n
i
n
g
t
o
p
-
h
at
o
p
era
t
i
o
n
a
n
d
b
o
t
t
o
m
-
h
at
o
p
era
t
i
o
n
,
w
a
s
u
s
ed
as
a
p
ro
p
o
s
ed
en
h
an
cemen
t
al
g
o
ri
t
h
m.
T
h
e
p
r
o
p
o
s
e
d
en
h
an
ceme
n
t
al
g
o
r
i
t
h
m
w
as
c
o
mp
are
d
t
o
t
h
ree
o
t
h
er
w
el
l
-
k
n
o
w
n
e
n
h
a
n
cemen
t
al
g
o
r
i
t
h
ms
,
s
p
eci
f
i
ca
l
l
y
h
i
s
t
o
g
ram
eq
u
al
i
zat
i
o
n
,
l
o
g
ar
i
t
h
mi
c
t
ra
n
s
f
o
rm,
an
d
g
a
mma
co
rrect
i
o
n
w
i
t
h
d
i
ffere
n
t
g
amma
v
al
u
es
.
T
w
e
n
t
y
-
fi
v
e
mammo
g
rap
h
i
c
i
m
ag
es
w
er
e
t
ak
e
n
fro
m
t
h
e
mammo
g
ra
p
h
y
i
mag
e
an
a
l
y
s
i
s
s
o
c
i
et
y
(MIA
S)
d
at
ab
as
e
s
am
p
l
e
s
.
T
h
e
mi
n
i
m
u
m
en
t
r
o
p
y
d
i
fferen
ce
v
a
l
u
e
(
E
D
V
)
w
as
u
s
e
d
as
met
ri
c
t
o
ev
al
u
at
e
the
b
e
s
t
en
h
a
n
cemen
t
al
g
o
ri
t
h
m.
Res
u
l
t
s
h
as
ap
p
ro
v
e
d
t
h
a
t
t
h
e
p
ro
p
o
s
ed
en
h
a
n
cemen
t
al
g
o
r
i
t
h
m
g
av
e
t
h
e
b
es
t
-
e
n
h
a
n
ced
i
ma
g
es
i
n
co
mp
ar
i
s
o
n
t
o
t
h
e
afo
reme
n
t
i
o
n
ed
a
l
g
o
ri
t
h
m
s
.
K
e
y
w
o
r
d
s
:
E
ntr
opy
d
if
f
e
r
e
nc
e
va
lue
(
E
DV
)
Ga
mm
a
c
or
r
e
c
ti
on
His
togr
a
m
e
qua
li
z
a
ti
on
L
oga
r
it
hmi
c
tr
a
ns
f
or
ma
ti
on
M
or
phology
e
nha
nc
e
ment
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
oha
mm
e
d
G.
Ayoub,
De
pa
r
tm
e
nt
of
C
omput
e
r
T
e
c
hnology
E
nginee
r
ing
T
e
c
hnica
l
E
nginee
r
ing
C
oll
e
ge
/
M
os
ul,
Nor
ther
n
T
e
c
hnica
l
Unive
r
s
it
y,
Al
-
mi
na
s
a
s
tr
e
e
t,
a
c
r
os
s
f
r
om
I
bn
Ala
thee
r
hos
pit
a
l
,
M
os
ul,
I
r
a
q.
E
mail:
mohammed
.
gha
nim
@ntu.
e
du
.
iq
1.
I
NT
RODU
C
T
I
ON
C
a
nc
e
r
is
a
n
unc
onf
ined
gr
owth
of
malignant
c
e
ll
s
i
n
a
c
e
r
tain
loca
ti
on
of
human
body
[
1
-
3]
.
I
t
is
f
a
tal
dis
e
a
s
e
a
nd
thos
e
s
uf
f
e
r
ing
c
a
nc
e
r
s
mos
t
li
ke
ly
f
a
c
ing
de
a
th
[
4]
.
T
he
W
or
ld
He
a
lt
h
Or
ga
niza
ti
on
(
W
HO
)
r
e
por
ted
in
2014
that
c
a
nc
e
r
is
the
s
e
c
ond
c
a
us
e
of
de
a
ths
in
the
wo
r
ld
[
5]
.
B
r
e
a
s
t
c
a
nc
e
r
is
one
o
f
t
he
mos
t
he
a
lt
h
pr
oblems
in
the
wor
ld.
M
or
e
ove
r
,
it
is
th
e
main
c
a
us
e
of
c
a
nc
e
r
de
a
th
a
mongs
t
wome
n
e
s
pe
c
ially
in
de
ve
lopi
ng
c
ountr
ies
[
6]
.
Ac
c
or
ding
to
a
s
tudy
c
onduc
ted
by
R
a
jar
a
man
e
t
a
l.
in
2015,
the
r
a
nge
of
wome
n
with
br
e
a
s
t
c
a
nc
e
r
in
I
nd
ia
wa
s
a
r
ound
36
pe
r
100,
000
whils
t
in
E
ur
ope
a
nd
No
r
th
Ame
r
ica
wa
s
92
to
112
pe
r
100,
000
wome
n
[
7
]
.
T
his
s
tudy
a
ls
o
f
ound
that
the
mor
tali
ty
o
f
br
e
a
s
t
c
a
nc
e
r
in
I
ndia
is
r
e
lativ
e
ly
high
(
12.
7
pe
r
100,
000
wome
n)
wh
ich
is
s
im
il
a
r
to
the
r
a
tes
in
de
ve
lopi
ng
c
ountr
ies
[
7
]
.
Ac
c
or
ding
to
the
e
s
ti
mation
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
Qualit
ati
v
e
as
s
e
s
s
me
nt
of
image
e
nhanc
e
me
nt
algo
r
it
hms
for
mam
mogr
am
s
…
(
M
az
in
N
.
F
ar
han
)
929
of
the
Ame
r
ica
n
C
a
nc
e
r
S
oc
iety
(
AC
S
)
in
2016
,
w
omen
de
a
ths
r
e
s
ult
e
d
f
r
om
b
r
e
a
s
t
c
a
nc
e
r
wa
s
a
bout
40,
290
in
the
United
S
tate
s
[
8]
.
A
s
tudy
c
onduc
ted
by
M
a
lvi
a
e
t
a
l.
in
201
7
de
c
lar
e
d
that
1,
797
,
900
wome
n
s
uf
f
e
r
e
d
f
r
om
br
e
a
s
t
c
a
nc
e
r
in
I
ndia
by
2020
[
9]
.
T
he
las
t
e
s
ti
mation
o
f
the
Ame
r
ica
n
c
a
nc
e
r
s
oc
iety
in
2019
s
tate
d
that
a
bout
268,
600
wome
n
will
be
c
ons
ider
e
d
to
ha
ve
b
r
e
a
s
t
c
a
nc
e
r
a
nd
41,
760
wome
n
will
f
a
c
e
de
a
th
due
to
br
e
a
s
t
gr
owth
[
10
,
11
]
.
T
he
p
r
e
ve
nti
on
of
br
e
a
s
t
c
a
nc
e
r
is
s
ti
ll
i
mpos
s
ibl
e
due
to
the
unidentif
ied
c
a
us
e
of
it
.
T
hus
,
ther
e
is
no
e
f
f
e
c
ti
ve
wa
y
to
a
void
b
r
e
a
s
t
c
a
nc
e
r
but
s
c
r
e
e
ning
to
de
tec
t
it
in
e
a
r
ly
s
tage
s
[
12]
.
M
a
ny
or
ga
niza
ti
ons
in
the
he
a
lt
h
s
e
c
tor
ha
ve
r
a
is
e
d
the
a
wa
r
e
ne
s
s
to
br
e
a
s
t
c
a
nc
e
r
by
e
nc
our
a
ging
w
omen
to
pa
r
ti
c
ipate
in
diagnos
ti
c
inves
ti
ga
ti
ons
s
uc
h
a
s
ti
s
s
u
e
s
a
mpl
ing,
c
li
nica
l
br
e
a
s
t
e
xa
mi
na
ti
on
(
C
B
E
)
a
nd
im
a
ging
f
or
e
a
r
ly
de
tec
ti
on
o
f
br
e
a
s
t
c
a
nc
e
r
[
13]
.
I
maging
i
s
s
ti
ll
the
mos
t
popula
r
wa
y
in
br
e
a
s
t
s
c
r
e
e
ning.
T
he
X
-
r
a
y
mammogr
a
phy
is
us
e
d
f
o
r
int
e
r
pr
e
ti
ng
b
r
e
a
s
t
c
a
nc
e
r
.
Ac
c
or
ding
to
Vikhe
a
nd
T
hool
in
2016
,
mamm
ogr
a
phy
is
the
mos
t
e
f
f
e
c
ti
ve
wa
y
f
or
b
r
e
a
s
t
s
c
r
e
e
ning.
T
his
tec
hnique
ha
s
led
to
de
c
r
e
a
s
e
br
e
a
s
t
c
a
nc
e
r
de
a
ths
by
25%
[
14
]
.
M
a
mm
ogr
a
phy
is
the
ini
ti
a
l
s
tep
f
or
b
r
e
a
s
t
c
a
nc
e
r
r
e
c
ognit
ion
[
15
,
16]
.
C
a
lcium
de
pos
it
io
ns
c
r
e
a
te
c
a
lcif
ica
ti
ons
that
us
ua
ll
y
a
ppe
a
r
a
s
s
pots
in
mam
mogr
a
phic
im
a
ge
s
indi
c
a
ti
ng
a
potential
a
bnor
mal
it
y
that
is
mos
tl
y
c
ons
ider
e
d
a
s
a
s
ign
of
c
a
nc
e
r
[
1]
.
I
n
s
ome
c
a
s
e
s
of
mi
c
r
o
-
c
a
lcif
ica
ti
ons
in
br
e
a
s
t
ti
s
s
ue
s
,
low
c
ontr
a
s
t
a
f
f
e
c
ts
the
r
a
diol
ogis
t’
s
a
bil
it
y
in
diagnos
ing
b
r
e
a
s
t
c
a
nc
e
r
by
digi
tal
mammog
r
a
phic
im
a
ge
s
[
1]
.
Gr
a
y
s
ha
de
va
r
iation
in
mammog
r
a
ms
r
e
duc
e
s
c
ontr
a
s
t
leve
l
due
to
s
c
a
tt
e
r
e
d
X
-
r
a
diation,
high
ly
powe
r
e
d
X
-
r
a
y
pe
ne
tr
a
ti
on
a
nd
the
f
il
m
li
mi
ted
c
a
pa
c
it
y
whic
h
l
e
a
ds
to
f
a
ls
e
pos
it
ive
r
e
s
ult
s
[
17]
.
He
nc
e
,
e
nha
nc
ing
thes
e
im
a
ge
s
will
s
igni
f
ica
ntl
y
i
mpr
ove
the
diagnos
ing
c
a
pa
bil
it
y.
T
he
f
uz
z
y
na
tu
r
e
of
mammog
r
a
phic
im
a
ge
a
nd
it
s
low
dif
f
e
r
e
nti
a
bil
it
y
f
r
om
the
ba
c
kgr
ound
make
it
t
oo
dif
f
icult
to
a
na
lyze
them
.
T
hus
,
it
is
e
s
s
e
nti
a
l
to
s
uppr
e
s
s
the
nois
e
a
nd
e
nha
nc
e
the
r
e
gion
of
int
e
r
e
s
t
(
R
OI
)
[
18
,
19
]
.
I
mage
e
nha
nc
e
ment
tec
hniques
ha
ve
b
e
e
n
us
e
d
wide
ly
f
or
diagnos
ing
a
s
pe
c
ts
.
Algor
it
hms
s
uc
h
a
s
his
togr
a
m
e
qua
li
z
a
ti
on,
media
n
f
il
ter
s
,
ga
mm
a
c
o
r
r
e
c
ti
on,
Ga
us
s
ian
f
il
ter
s
,
logar
i
thm
ic
t
r
a
ns
f
or
mation
a
nd
m
or
phologi
c
a
l
f
il
ter
ha
ve
be
e
n
a
ppli
e
d
to
medic
a
l
i
mage
s
to
e
nha
nc
e
the
r
e
gion
o
f
int
e
r
e
s
t
(
R
OI
)
[
20
]
.
I
n
thi
s
pa
pe
r
,
f
our
e
nha
nc
e
ment
tec
hniques
ha
v
e
be
e
n
a
p
pli
e
d
on
digi
tal
mammog
r
a
phic
im
a
ge
s
:
His
togr
a
m
E
qua
li
z
a
ti
on
(
HE
)
,
L
oga
r
it
hmi
c
tr
a
ns
f
or
mation
(
L
og
T
r
a
ns
f
or
m)
,
Ga
mm
a
C
or
r
e
c
ti
on
(
GC
)
a
nd
f
inally
M
or
phology
e
nha
nc
e
ment
(
the
pr
opos
e
d
e
nha
nc
e
ment
a
lgor
it
h
m)
.
T
he
r
e
s
t
s
e
c
ti
ons
a
r
e
a
r
r
a
nge
d
a
s
f
oll
ows
:
ne
xt
s
e
c
ti
on
gives
a
br
ief
e
xplana
ti
on
a
bout
the
us
e
d
e
nha
nc
e
ment
a
lgor
it
hms
.
M
a
ter
ials
a
nd
method
s
s
e
c
ti
on
dis
c
us
s
e
s
the
r
e
s
e
a
r
c
h
methodology
a
nd
how
the
r
e
s
e
a
r
c
h
wa
s
pe
r
f
or
med.
R
e
s
ult
s
s
e
c
ti
on
c
lar
i
f
ies
the
e
nha
nc
e
d
mammog
r
a
phic
im
a
ge
s
r
e
s
ult
e
d
f
r
om
a
ll
us
e
d
a
lgor
it
hms
.
Dis
c
us
s
ion
s
e
c
ti
on
e
xplains
other
a
s
pe
c
ts
that
a
c
c
ompany
e
nha
nc
e
ment
a
lgor
it
hms
.
F
inally,
c
onc
lus
ion
s
e
c
ti
on
s
umm
a
r
ize
s
the
us
e
d
e
nha
nc
e
m
e
nt
a
lgor
it
hms
a
nd
de
c
lar
e
s
the
opti
mum
one
.
Digit
a
l
mammogr
a
phic
im
a
ge
s
a
r
e
us
ua
ll
y
a
c
c
ompanie
d
wi
th
high
nois
e
a
nd
low
c
ont
r
a
s
t
[
19]
.
S
e
ve
r
a
l
e
nha
nc
e
ment
tec
hniques
c
a
n
be
a
ppli
e
d
t
o
thes
e
im
a
ge
s
to
im
pr
ove
their
c
lar
it
y
.
T
hus
,
e
n
ha
nc
ing
c
ontr
a
s
t,
r
e
movi
ng
nois
e
,
s
uppr
e
s
s
ing
ba
c
kgr
oun
d
a
nd
e
nha
nc
ing
e
dge
s
a
r
e
e
xa
mpl
e
s
of
c
omm
o
n
im
a
ge
e
nha
nc
e
ments
[
20]
.
I
n
digi
tal
im
a
ge
s
,
his
togr
a
m
gi
ve
s
a
gr
a
phica
l
r
e
pr
e
s
e
ntation
of
gr
a
y
leve
ls
dis
tr
ibut
ion
in
the
im
a
ge
.
He
nc
e
,
the
f
r
e
que
nc
y
o
f
a
ny
indi
v
idual
gr
a
y
leve
l
in
the
im
a
ge
c
a
n
be
e
s
ti
mate
d
a
nd
a
na
lyz
e
d
e
a
s
il
y
by
obs
e
r
ving
the
im
a
ge
his
togr
a
m.
His
togr
a
m
e
q
ua
li
z
a
ti
on
is
a
c
omm
on
method
us
e
d
f
or
e
nha
nc
i
ng
im
a
ge
c
ontr
a
s
t
by
s
pr
e
a
ding
out
the
int
e
ns
it
y
va
lues
a
long
the
e
nti
r
e
r
a
nge
of
va
lues
.
His
togr
a
m
e
qua
li
z
a
ti
on
t
e
c
hnique
is
e
f
f
icie
nt
in
c
ontr
a
s
t
e
nha
nc
e
ment
whe
r
e
ve
r
th
e
r
e
pr
e
s
e
nted
im
a
ge
ha
s
c
los
e
d
c
ontr
a
s
t
va
l
ue
s
.
I
ntens
it
y
tr
a
ns
f
or
mation
is
a
nother
wa
y
of
im
a
ge
e
nha
nc
e
ment,
i
t
ha
s
be
e
n
c
ons
ider
e
d
a
s
the
s
im
ples
t
tec
hnique
us
e
d
in
the
f
ield
of
im
a
ge
pr
oc
e
s
s
ing
[
21]
.
I
ntens
it
y
tr
a
ns
f
or
mations
include
im
a
ge
ne
ga
ti
ve
,
log
a
r
it
hmi
c
tr
a
ns
f
or
mation,
ga
mm
a
c
or
r
e
c
ti
on
a
nd
piec
e
wis
e
li
ne
a
r
s
tr
e
tching.
L
oga
r
it
hmi
c
tr
a
ns
f
or
mation
maps
t
he
va
lues
of
low
int
e
ns
it
y
pixels
in
the
input
int
o
wide
r
outp
ut
va
lues
a
nd
vice
ve
r
s
a
.
T
his
tec
hnique
is
us
e
d
to
e
xpa
nd
da
r
k
pixels
va
lues
in
the
input
im
a
ge
s
a
nd
c
ompr
e
s
s
the
va
lue
s
of
br
ig
ht
one
s
.
S
im
i
lar
ly,
ga
mm
a
c
or
r
e
c
ti
on
maps
da
r
k
pixels
in
the
input
im
a
ge
s
int
o
br
igh
ter
one
s
a
nd
vice
ve
r
s
a
de
pe
nding
on
(
γ
)
va
lue.
T
he
mor
phology
e
nha
nc
e
ment
(
the
p
r
opos
e
d
E
nha
nc
e
ment
a
lgor
it
hm)
de
pe
nds
on
top
-
ha
t
ope
r
a
ti
on
a
nd
bott
om
-
ha
t
ope
r
a
ti
on
to
e
nh
a
nc
e
mammogr
a
phic
im
a
ge
in
ter
ms
of
a
c
c
e
ntuating
high
-
int
e
ns
it
y
a
nd
low
-
int
e
ns
it
y
mor
phologi
c
a
l
s
tr
uc
tur
e
s
.
Ope
ning
a
nd
c
los
ing
ope
r
a
ti
ons
a
r
e
buil
t
f
r
o
m
e
r
os
ion
a
nd
di
lation
a
s
f
oll
ows
[
22,
23]
.
Ope
ning
ope
r
a
ti
on:
○
=
(
⊖
)
⊕
(
1)
C
los
ing
ope
r
a
ti
on:
•
=
(
⊕
)
⊖
(
2)
w
he
r
e
,
OI
is
the
or
igi
nal
image
SI
is
the
s
tr
uc
tur
al
e
lem
e
nt
⊖
R
e
fer
s
to
e
r
os
ion
ope
r
ati
on
⊕
R
e
fer
s
to
dil
ati
on
ope
r
ati
on
.
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
.
2
,
Ap
r
il
2020:
928
-
935
930
T
he
n,
top
-
ha
t
(
T
H)
a
nd
bott
om
-
ha
t
(
B
H)
f
il
ter
ing
a
r
e
r
e
s
pe
c
ti
ve
ly
e
xpr
e
s
s
e
d
a
s
s
hown
in
(
3
)
a
nd
(
4)
.
=
−
○
(
3)
=
•
−
(
4)
F
inally,
mo
r
phology
e
nha
nc
e
ment
a
lgor
i
thm
is
obt
a
ind
by
a
dding
the
top
-
ha
t
ope
r
a
ti
on
to
the
o
r
igi
na
l
im
a
ge
,
then
the
bott
om
-
ha
t
ope
r
a
ti
on
is
s
ubtr
a
c
ted
f
r
o
m
th
e
r
e
s
ult
a
s
s
hown
in
the
f
oll
owing
(
5)
.
ℎ
=
+
−
(
5)
P
r
opos
e
d
e
nha
nc
e
ment
a
lgor
it
hm
will
r
e
duc
e
s
the
nois
e
in
a
ddit
ion
to
e
nha
nc
ing
the
low
a
nd
high
int
e
ns
it
y
r
e
gions
.
2.
M
AT
E
R
I
AL
S
AN
D
M
E
T
HO
D
M
AT
L
AB
ve
r
s
ion
R
2015b
wa
s
us
e
d
to
pe
r
f
or
m
a
ll
the
e
nha
nc
e
ment
a
lgor
it
hms
.
T
he
mammogr
a
phic
im
a
ge
s
we
r
e
take
n
f
r
om
the
M
a
mm
ogr
a
phy
I
ma
ge
Ana
lys
is
S
oc
iety
(
M
I
AS)
Da
taba
s
e
s
a
mpl
e
s
.
M
I
AS’
s
s
a
mpl
e
s
a
r
e
c
ha
r
a
c
ter
ize
d
by
8
-
bit
s
of
gr
a
ys
c
a
le
le
ve
l
a
nd
s
ize
of
1024
x1024.
T
he
M
I
AS’
s
s
a
mpl
e
s
pr
ovide
a
n
inf
or
mation
a
bout
the
s
pe
c
if
ic
loca
ti
on
of
the
a
bnor
mal
ti
s
s
ue
thr
oughout
the
mammog
r
a
phy
[
24]
.
C
ons
e
que
ntl
y,
M
I
AS’
s
a
r
e
c
ons
ider
e
d
the
mos
t
s
uit
a
ble
da
taba
s
e
a
nd
ha
s
be
e
n
c
hos
e
n
a
s
the
s
our
c
e
s
o
f
im
a
ge
s
f
or
tes
ti
ng
va
r
ious
im
a
ge
s
e
nha
nc
ing
methods
a
nd
identif
ying
the
opti
mum
one
ther
e
a
f
ter
.
M
I
AS’
s
mammogr
a
phy
s
a
mpl
e
s
we
r
e
labe
led
by
number
s
a
nd
letter
s
to
be
dif
f
e
r
e
nt
f
r
om
othe
r
s
a
mpl
e
s
.
F
ur
ther
mo
r
e
,
M
I
AS’
s
im
a
ge
s
ha
ve
two
r
e
gions
,
f
i
r
s
tl
y
the
ba
c
kgr
ound
,
whic
h
f
or
ms
th
e
who
le
im
a
ge
e
xc
e
pt
the
b
r
e
a
s
t
pa
r
t,
s
e
c
ondly
the
f
or
e
gr
ound
,
whic
h
r
e
p
r
e
s
e
nts
the
br
e
a
s
t
ti
s
s
ue
.
T
r
ying
to
e
nha
nc
e
the
mammogr
a
phy
with
their
labe
ls
a
nd
ba
c
kgr
ounds
ha
s
led
to
dis
tr
a
c
t
the
e
f
f
e
c
ts
of
the
e
nh
a
nc
e
ment
a
lgor
it
hms
on
the
e
nti
r
e
im
a
ge
.
T
he
r
e
f
or
e
,
r
e
gion
of
the
int
e
r
e
s
t
(
B
r
e
a
s
t
ti
s
s
ue
)
wa
s
f
ir
s
tl
y
e
xtr
a
c
ted
a
s
s
hown
in
F
igu
r
e
1
us
ing
a
lgor
it
hm
de
s
igned
f
or
that
pur
pos
e
,
then
a
ll
the
e
nha
nc
e
ment
a
lgor
it
hms
we
r
e
a
ppli
e
d
to
r
e
gion
of
int
e
r
e
s
t
(
R
oI
)
.
I
n
thi
s
pa
pe
r
,
the
qua
li
tati
ve
a
s
s
e
s
s
ment
pr
oc
e
s
s
is
done
thr
ough
the
f
ol
lowin
g
s
tage
s
:
-
I
mage
a
c
quis
it
ion.
-
P
r
e
pr
oc
e
s
s
ing
(
R
oI
e
xtr
a
c
ti
on)
.
-
Applying
the
e
nha
nc
e
ment
a
lgor
it
hms
.
-
Doing
E
ntr
opy
a
na
lys
is
to
the
r
e
s
ult
s
.
-
De
c
lar
ing
the
opti
mum
a
lgor
it
hm
.
F
our
e
nha
nc
e
ment
a
lgor
it
hms
we
r
e
a
ppl
ied
on
25
mammogr
a
phic
im
a
ge
s
take
n
f
r
om
M
I
AS
’
s
da
taba
s
e
.
T
he
a
ppli
e
d
e
nha
nc
e
ment
a
lgor
it
hms
w
e
r
e
his
togr
a
m
e
qua
li
z
a
ti
on
,
logar
it
hm
tr
a
ns
f
or
mat
ion,
a
nd
ga
mm
a
c
or
r
e
c
ti
on
f
or
dif
f
e
r
e
nt
va
lues
s
pe
c
if
ica
ll
y
(
0.
3,
0
.
45,
0.
6
,
0
.
75
a
nd
0.
9
)
a
nd
mor
pho
logy
e
nha
nc
e
ment
(
the
p
r
opos
e
d
one
)
.
T
he
opti
mum
e
nha
nc
e
ment
a
lgor
it
hm
wa
s
c
hos
e
n
de
pe
nding
on
the
m
ini
mum
e
ntr
opy
dif
f
e
r
e
nc
e
va
lue
(
E
DV
)
[
25]
,
looki
ng
f
o
r
e
ntr
opy
n
e
a
r
or
s
im
il
a
r
to
the
or
ig
inal
e
ntr
opy
.
F
igur
e
1.
T
wo
or
igi
na
l
s
a
mpl
e
s
f
r
om
M
I
AS’
s
da
taba
s
e
(
on
the
lef
t)
we
r
e
p
r
e
pr
oc
e
s
s
e
d
to
e
xtr
a
c
t
the
r
e
gion
o
f
int
e
r
e
s
t
(
R
oI
)
on
the
r
ight
F
igur
e
2.
T
he
block
diagr
a
m
of
a
s
s
e
s
s
ment
pr
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
Qualit
ati
v
e
as
s
e
s
s
me
nt
of
image
e
nhanc
e
me
nt
algo
r
it
hms
for
mam
mogr
am
s
…
(
M
az
in
N
.
F
ar
han
)
931
3.
RE
S
UL
T
S
A
ND
DI
S
CU
S
S
I
ON
T
he
e
ntr
opy
of
the
e
nha
nc
e
d
im
a
ge
s
is
de
pe
n
di
ng
on
the
us
e
d
a
lgor
it
hms
a
s
s
hown
in
T
a
ble
1.
T
a
ble
1
s
hows
that
dif
f
e
r
e
nt
a
lgor
it
h
ms
modi
f
y
the
or
igi
na
l
e
nt
r
opy
in
di
f
f
e
r
e
nt
wa
ys
.
His
togr
a
m
e
qu
a
li
z
a
ti
on
a
lgor
it
hm
s
igni
f
ica
ntl
y
de
c
r
e
a
s
e
d
the
o
r
igi
na
l
e
ntr
o
py,
whi
le
the
p
r
opos
e
d
e
nha
nc
e
ment
a
lgor
i
thm
ha
ve
a
lm
os
t
the
s
a
me
e
ntr
opy
a
s
the
or
igi
na
l
.
T
he
L
oga
r
it
hmi
c
tr
a
ns
f
or
mation
a
lgor
it
hm
ha
s
s
li
ghtl
y
modi
f
ied
the
or
ig
inal
e
ntr
opy.
W
he
r
e
a
s
the
Ga
mm
a
c
or
r
e
c
ti
on
a
lgor
it
hm
modi
f
ied
the
o
r
igi
na
l
e
ntr
opy
a
c
c
or
ding
to
the
us
e
d
f
a
c
tor
,
thi
s
a
lgor
it
hm
(
i
.
e
GC
)
s
howe
d
that
the
lowe
s
t
f
a
c
t
or
be
ing
us
e
d
gives
the
highes
t
c
ha
nge
in
e
ntr
opy
a
nd
vice
ve
r
s
a
(
r
e
f
e
r
to
F
igu
r
e
3
a
nd
F
igur
e
4
)
.
T
he
d
if
f
e
r
e
n
c
e
s
be
twe
e
n
the
or
igi
na
l
e
ntr
opy
a
nd
the
modi
f
ied
e
ntr
opies
that
r
e
s
ult
e
d
f
r
om
the
us
e
d
e
nha
nc
e
ment
a
lgor
it
h
ms
ha
ve
be
e
n
e
xa
mi
ne
d
to
identif
y
the
mi
nim
um
E
DV
a
nd
e
ve
ntually
the
opti
mum
e
nha
nc
e
ment
a
lgor
it
hm
is
c
hos
e
n.
I
t
is
c
lea
r
ly
s
e
e
n
that
the
mi
nim
um
E
DV
is
obtaine
d
whe
n
us
ing
the
p
r
opos
e
d
e
nha
nc
e
ment
a
lgor
it
hm
(
mor
pho
logy
e
na
ha
nc
e
ment)
.
Ga
mm
a
c
or
r
e
c
ti
on
of
0.
9
,
howe
ve
r
,
wa
s
r
a
nke
d
s
e
c
ondly
in
pr
oduc
ing
mi
nim
um
E
DV
.
F
igu
r
e
s
(
3
a
nd
4
)
s
how
how
the
mam
mogr
a
phy
im
a
ge
is
e
nha
nc
e
d
a
c
c
or
ding
to
us
e
d
a
lgor
i
thm
s
.
T
his
r
e
s
e
a
r
c
h
ha
s
pa
s
s
e
d
thr
ough
s
e
ve
r
a
l
s
t
a
ge
s
;
s
t
a
r
ti
ng
f
r
om
c
hoos
ing
the
r
ight
da
taba
s
e
a
s
a
s
our
c
e
of
the
im
a
ge
s
unt
il
the
f
inal
s
tage
of
pr
oduc
ing
e
n
ha
nc
e
d
im
a
ge
s
.
As
mentioned
in
the
pr
e
vious
s
e
c
ti
ons
,
f
ou
r
e
nha
nc
ing
a
lgor
it
hms
ha
ve
be
e
n
us
e
d.
T
he
pr
opos
e
d
e
nha
nc
e
ment
a
lgor
it
hm
(
mo
r
phology
e
nha
nc
e
ment)
,
HE
,
L
og
T
r
a
ns
f
or
m
a
nd
GC
with
dif
f
e
r
e
nt
va
lues
(
0.
3
,
0.
45,
0.
6
,
0
.
75,
0
.
9)
.
T
he
r
e
a
s
on
f
or
us
ing
thes
e
a
lgor
it
hms
in
a
c
c
ompanying
with
the
pr
opos
e
d
one
is
to
c
ompar
e
their
output
s
.
M
I
AS
wa
s
c
hos
e
n
a
s
a
s
our
c
e
of
the
mammogr
a
phy
im
a
ge
s
due
to
the
good
im
a
g
e
s
’
r
e
s
olut
ion,
in
a
ddit
ion
to
the
diagnos
ing
inf
o
r
mation
s
upp
li
e
d
with
them.
T
he
goa
l
of
the
pr
opos
e
d
e
nha
nc
e
ment
a
lgor
it
hm
is
to
r
e
a
c
h
the
be
s
t
pos
s
ibl
e
e
nha
nc
e
ment
of
the
int
e
r
e
s
ted
a
r
e
a
.
T
he
labe
ls
a
nd
the
ba
c
kgr
o
und
of
the
br
e
a
s
t
a
r
e
c
ons
ider
e
d
a
s
lowe
r
im
por
ta
nc
e
than
the
br
e
a
s
t
it
s
e
lf
.
T
he
r
e
f
o
r
e
,
the
r
e
gion
o
f
int
e
r
e
s
t
(
B
r
e
a
s
t
a
r
e
a
)
wa
s
identi
f
ied
a
nd
c
r
oppe
d
be
f
or
e
a
pp
lyi
ng
a
ny
of
the
e
nha
nc
ing
a
lgor
i
thm
s
.
Appr
oving
the
be
s
t
e
nha
nc
ing
a
lgor
it
hm
ha
s
ba
s
ica
ll
y
ba
s
e
d
on
the
s
tatis
ti
c
a
l
f
a
c
tor
mi
nim
um
E
DV
.
T
he
E
DV
ha
s
be
e
n
us
e
d
to
c
ompar
e
the
e
ntr
opy
r
e
s
ult
e
d
f
r
o
m
e
a
c
h
a
ppli
e
d
a
lgor
it
hm
with
that
o
f
the
or
igi
na
l
im
a
ge
s
.
T
hus
,
the
c
los
e
r
r
e
s
ult
e
d
e
ntr
opy
to
the
or
igi
na
l
one
wa
s
identi
f
ied
a
s
the
be
s
t
e
nha
nc
ing
a
lgor
it
hm.
T
a
ble
(
1)
s
howe
d
the
e
ntr
opi
e
s
r
e
s
ult
e
d
f
r
om
us
e
d
e
nha
nc
e
ments
a
lgor
it
ms
.
I
t
i
s
c
lea
r
ly
s
e
e
n
th
a
t
the
pr
opos
e
d
a
lgor
it
h
m
ha
s
pr
ovided
the
ne
a
r
e
s
t
e
ntr
opy
va
lue
to
the
o
r
igi
na
l
e
ntr
opy
.
T
he
qua
li
ty
of
im
a
ge
s
e
nha
nc
e
d
by
the
GC
is
ba
s
ica
ll
y
de
pe
nding
on
the
va
lue
o
f
(
γ)
.
I
n
thi
s
pa
pe
r
,
f
ive
va
lues
of
(
γ)
we
r
e
us
e
d,
the
maximum
us
e
d
va
lue
wa
s
(
0.
9)
.
E
x
c
e
e
ding
(
γ
=
0.
9)
to
(
γ
=
1)
will
pr
oduc
e
a
n
identica
l
i
mage
to
the
or
igi
na
l
one
.
I
n
a
ddit
ion,
mul
ti
ply
ing
L
og
T
r
a
ns
f
or
m
a
nd
GC
by
a
f
a
c
tor
will
modi
f
y
the
br
igh
tnes
s
of
the
e
nha
nc
e
d
im
a
ge
,
de
pe
nding
on
the
va
lue
of
th
a
t
f
a
c
tor
.
Us
ing
f
a
c
tor
les
s
than
on
e
de
c
r
e
a
s
e
the
br
ight
ne
s
s
a
nd
vice
ve
r
s
a
.
T
a
ble
1.
C
ompar
is
on
a
mongs
t
the
a
ppli
e
d
e
nha
nc
e
ment
a
lgor
it
hm
on
mammog
r
a
ms
f
or
E
nt
r
opy
a
na
l
ys
is
I
ma
ge
s
s
e
que
n
c
e
s
O
r
ig
in
a
l
E
nt
r
opy
P
r
opos
e
d
E
nha
nc
e
me
nt
E
nt
r
opy
HE
E
nt
r
opy
L
og.
T
r
a
ns
f
or
m
E
nt
r
opy
G
C
0.3
E
nt
r
opy
G
C
0.45
E
nt
r
opy
G
C
0.6
E
nt
r
opy
G
C
0.75
E
nt
r
opy
G
C
0.9
E
nt
r
opy
1
6.8974
6.9177
5.3222
6.4244
6.1989
6.4721
6.6368
6.7503
6.846
2
6.9393
6.9688
5.318
6.4318
6.1691
6.4694
6.6497
6.7763
6.881
3
6.607
6.6775
5.4783
6.0771
5.807
6.0958
6.2891
6.4416
6.5594
4
6.529
6.5814
5.1717
6.0034
5.7121
6.0118
6.1972
6.339
6.4601
5
5.6535
5.7205
4.5656
5.2023
4.9535
5.2111
5.3692
5.4912
5.5947
6
6.4031
6.4322
5.1507
5.9051
5.6415
5.9216
6.1044
6.2359
6.3463
7
6.4192
6.4398
5.0577
5.8848
5.5707
5.8837
6.0675
6.2172
6.3468
8
5.4226
5.4507
4.3588
4.9726
4.7036
4.9639
5.1273
5.2535
5.3598
9
6.1838
6.1987
4.8155
5.679
5.3886
5.6813
5.8511
5.9933
6.1117
10
6.7011
6.7298
5.3856
6.1186
5.7722
6.1085
6.3201
6.4835
6.6185
11
6.874
6.91
5.3906
6.3177
6.0029
6.3262
6.5194
6.6675
6.7949
12
6.1148
6.1507
4.9428
5.591
5.2726
5.5831
5.7603
5.9131
6.0407
13
6.9124
6.9429
5.456
6.4246
6.203
6.4756
6.6501
6.774
6.8652
14
7.0354
7.0564
5.6098
6.5593
6.3579
6.6344
6.8056
6.9135
6.9935
15
6.6299
6.6591
5.4398
5.9932
5.5986
5.9751
6.1874
6.3675
6.5279
16
6.6299
6.6591
5.4398
5.9932
5.5986
5.9751
6.1874
6.3675
6.5279
17
6.3156
6.3584
5.1258
5.7748
5.4529
5.7668
5.9625
6.115
6.2387
18
7.0258
7.0407
5.4752
6.4713
6.168
6.4888
6.6809
6.8312
6.9542
19
6.7645
6.7884
5.3806
6.3053
6.1011
6.3615
6.5275
6.6403
6.722
20
6.6236
6.6449
5.0616
6.1413
5.8861
6.1644
6.33
6.4567
6.5651
21
6.0842
6.1467
4.9165
5.5847
5.2929
5.581
5.7632
5.9073
6.0221
22
6.7434
6.7774
5.4621
6.1796
5.8593
6.183
6.3867
6.5374
6.6723
23
6.8611
6.8811
5.3794
6.3376
6.062
6.3556
6.5461
6.6899
6.8002
24
6.7768
6.8079
5.4936
6.1964
5.8702
6.2013
6.4133
6.5697
6.7018
25
6.8256
6.8473
5.3021
6.2644
5.9314
6.2601
6.4543
6.6104
6.7444
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
.
2
,
Ap
r
il
2020:
928
-
935
932
Or
igi
na
l
I
mage
HE
L
og
T
r
a
ns
f
or
m
GC
0.
3
GC
0.
45
GC
0.
6
GC
0.
75
GC
0.
9
P
r
opos
e
d
Algor
it
hm
(
a
)
Or
igi
na
l
I
mage
HE
L
og.
T
r
a
ns
f
or
m
GC
0.
3
GC
0.
45
GC
0.
6
GC
0.
75
GC
0.
9
P
r
opos
e
d
Algor
it
hm
(
b)
F
igur
e
3
.
(
a
)
His
togr
a
ms
of
the
o
r
igi
na
l
mammog
r
a
m
with
the
ou
tput
o
f
dif
f
e
r
e
nt
e
nha
nc
e
ment
a
lgor
i
thm
s
,
(
b)
T
he
e
f
f
e
c
t
of
a
ppli
e
d
e
nha
nc
e
ment
a
lgor
it
hms
on
mammogr
a
ms
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
Qualit
ati
v
e
as
s
e
s
s
me
nt
of
image
e
nhanc
e
me
nt
algo
r
it
hms
for
mam
mogr
am
s
…
(
M
az
in
N
.
F
ar
han
)
933
Or
igi
na
l
I
mage
HE
L
og
T
r
a
ns
f
or
m
GC
0.
3
GC
0.
45
GC
0.
6
GC
0.
75
GC
0.
9
P
r
opos
e
d
Algor
it
hm
(
a
)
Or
igi
na
l
I
mage
HE
L
og.
T
r
a
ns
f
or
m
GC
0.
3
GC
0.
45
GC
0.
6
GC
0.
75
GC
0.
9
P
r
opos
e
d
Algor
it
hm
(
b)
F
igur
e
4.
(
a
)
His
togr
a
ms
of
the
o
r
igi
na
l
mammog
r
a
m
with
the
ou
tput
o
f
dif
f
e
r
e
nt
e
nha
nc
e
ment
a
lgor
i
thm
s
,
(
b)
T
he
e
f
f
e
c
t
of
a
ppli
e
d
e
nha
nc
e
ment
a
lgor
it
hms
on
mammogr
a
ms
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
.
2
,
Ap
r
il
2020:
928
-
935
934
4.
CONC
L
USI
ON
S
e
ve
r
a
l
popular
e
nha
nc
e
ment
tec
hniques
including
the
pr
opos
e
d
e
nha
nc
e
ment
a
lgor
it
hm
ha
ve
be
e
n
a
ppli
e
d
to
the
s
e
lec
ted
mammogr
a
phic
im
a
ge
s
.
M
ini
mum
(
E
DV
)
wa
s
us
e
d
a
s
a
metr
ic
to
e
va
luate
the
e
f
f
icie
nc
y
of
pr
opos
e
d
e
nha
nc
e
ment
a
lgor
it
hm
in
c
ompar
is
o
n
with
other
e
nha
nc
e
ment
a
lgor
it
hms
.
F
o
r
e
ve
r
y
s
e
lec
ted
im
a
ge
,
the
e
ntr
opy
ha
s
be
e
n
c
a
lcula
ted
be
f
or
e
a
nd
a
f
ter
a
pplyi
ng
the
a
f
o
r
e
mentioned
a
lgor
it
hms
.
T
he
n,
a
br
ief
c
ompar
is
on
wa
s
done
a
mongs
t
thes
e
a
lgor
it
h
ms
a
c
c
or
ding
to
the
r
e
s
ult
e
d
e
ntr
opy
dif
f
e
r
e
nc
e
s
.
T
hus
,
o
ur
s
tudy
de
c
lar
e
d
that
the
be
s
t
e
nha
nc
e
ment
wa
s
done
by
the
mathe
matica
l
mo
r
phology
e
nha
nc
e
ment
(
the
pr
opos
e
d
e
nha
nc
e
ment
a
lgor
it
hm)
in
c
ompar
is
on
with
other
e
nha
nc
e
ment
a
lgor
it
hms
.
AC
KNOW
L
E
DGE
M
E
NT
S
W
e
take
thi
s
oppor
tuni
ty
to
e
xp
r
e
s
s
our
pr
of
ound
g
r
a
ti
tude
a
nd
de
e
p
r
e
ga
r
ds
to
the
pr
e
s
idenc
y
of
the
Nor
ther
n
T
e
c
hnica
l
Unive
r
s
it
y
f
or
their
c
ons
ta
nt
s
c
ientif
ic
e
nc
our
a
ge
ment.
RE
F
E
RE
NC
E
S
[1
]
V
i
k
h
e,
P.
S.
,
T
h
o
o
l
,
V
.
R
.
,
“
Co
n
t
ras
t
e
n
h
a
n
cemen
t
i
n
mammo
g
ram
s
u
s
i
n
g
h
o
mo
m
o
rp
h
i
c
f
i
l
t
er
t
ech
n
i
q
u
e,
”
2
0
1
6
In
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
er
e
n
ce
o
n
S
i
g
n
a
l
a
n
d
In
f
o
r
m
a
t
i
o
n
P
r
o
ces
s
i
n
g
,
p
p
.
1
-
5
,
2
0
1
6
.
[2
]
A
meri
ca
n
Can
cer
So
c
i
et
y
,
“Can
cer
fact
s
&
fi
g
u
res
2
0
1
1
,
”
A
t
l
a
n
t
a
:
A
mer
i
can
Can
cer
S
o
ci
e
t
y
,
2
0
1
1
.
[3
]
Can
cer
In
s
t
i
t
u
t
e
at
t
h
e
N
a
t
i
o
n
a
l
In
s
t
i
t
u
t
es
o
f
H
ea
l
t
h
.
[O
n
l
i
n
e]
.
A
v
ai
l
ab
l
e:
w
w
w
.
ca
n
cer.
g
o
v
.
[4
]
T
ab
ár
L
.
,
et
al
,
“
T
h
e
i
n
c
i
d
e
n
ce
o
f
fat
a
l
b
rea
s
t
ca
n
cer
meas
u
re
s
t
h
e
i
n
creas
e
d
effec
t
i
v
en
e
s
s
o
f
t
h
erap
y
i
n
w
o
men
p
art
i
ci
p
at
i
n
g
i
n
mamm
o
g
ra
p
h
y
s
creen
i
n
g
,
”
Ca
n
cer
,
v
o
l
.
1
2
5
,
n
o
.
4
,
p
p
.
5
1
5
-
5
2
3
,
2
0
1
9
.
[5
]
D
eSan
t
i
s
C.
,
Ma
J
.
,
Bry
a
n
L
.
,
&
J
emal
,
A
.
,
“
Breas
t
can
cer
s
t
a
t
i
s
t
i
cs
,
2
0
1
3
,”
C
A
:
A
Ca
n
cer
J
o
u
r
n
a
l
f
o
r
C
l
i
n
i
c
i
a
n
s
,
v
o
l
.
64
,
n
o
.
1
,
p
p
.
52
–
62
,
2
0
1
3
.
[6
]
D
u
X
.
L
.
,
Fo
x
E
.
E
.
,
L
ai
D
.
,
“
Co
mp
et
i
n
g
c
au
s
e
s
o
f
d
eat
h
fo
r
w
o
men
w
ith
b
rea
s
t
c
an
cer
an
d
c
h
an
g
e
o
v
er
t
i
me
f
r
o
m
1
9
7
5
t
o
2
0
0
3
,”
A
m
e
r
i
c
a
n
Jo
u
r
n
a
l
o
f
Cl
i
n
i
c
a
l
O
n
c
o
l
o
g
y
,
v
o
l
.
31
,
n
o
.
2
,
p
p
.
1
0
5
–
1
1
6
,
2
0
0
8
.
[7
]
Raj
araman
P
.
,
A
n
d
er
s
o
n
B
.
O
.
,
Bas
u
P
.
,
Bel
i
n
s
o
n
J
.
L
.
,
Cru
z
A
.
D
.
,
D
h
i
l
l
o
n
P
.
K
.,
“Reco
mmen
d
at
i
o
n
s
fo
r
s
cree
n
i
n
g
an
d
earl
y
d
e
t
ect
i
o
n
o
f
co
mm
o
n
ca
n
cers
i
n
In
d
i
a
,
”
Th
e
L
a
n
ce
t
O
n
c
o
l
ogy
,
v
o
l
.
1
6
,
n
o
.
7
,
p
p
.
e3
5
2
-
e3
6
1
,
2
0
1
5
.
[8
]
A
meri
ca
n
Can
cer
So
c
i
et
y
.
“Can
cer
fact
s
&
fi
g
u
res
2
0
1
5
,
”
A
t
l
a
n
t
a
:
A
mer
i
can
Can
cer
S
o
ci
e
t
y
;
2
0
1
5
[9
]
Mal
v
i
a
S.
,
Bag
ad
i
S.
A
.
,
D
u
b
ey
U
.
S.
,
Sax
en
a
S.
,
“
E
p
i
d
emi
o
l
o
g
y
o
f
b
reas
t
can
cer
i
n
In
d
i
a
n
w
o
me
n,
”
A
s
i
a
-
P
a
c
i
f
i
c
Jo
u
r
n
a
l
o
f
Cl
i
n
i
ca
l
O
n
co
l
o
g
y
,
v
o
l
.
13
,
n
o
.
4
,
p
p
.
289
–
2
9
5
,
2
0
1
7
.
[1
0
]
Si
eg
e
l
R.
L
.
,
Mi
l
l
er
K
.
D
.
,
J
emal
A
.,
“
Can
cer
s
t
at
i
s
t
i
c
s
,
2
0
1
9
,”
CA
:
A
C
a
n
ce
r
Jo
u
r
n
a
l
f
o
r
Cl
i
n
i
c
i
a
n
s
,
v
o
l
.
6
9
,
n
o
.
1
,
p
p
.
7
-
3
4
,
2
0
1
9
.
[1
1
]
Smi
t
h
R.
A
.
,
et
a
l
.
,
“
Can
cer
s
cree
n
i
n
g
i
n
t
h
e
U
n
i
t
ed
S
t
at
es
,
2
0
1
9
:
A
re
v
i
e
w
o
f
c
u
rren
t
A
meri
ca
n
Can
cer
S
o
ci
et
y
g
u
i
d
e
l
i
n
es
a
n
d
c
u
rren
t
i
s
s
u
es
i
n
can
cer
s
creen
i
n
g
,
”
CA
:
A
Ca
n
cer
J
o
u
r
n
a
l
f
o
r
Cl
i
n
i
ci
a
n
s
,
v
o
l
.
6
9
,
n
o
.
3
,
p
p
.
1
8
4
-
2
1
0
,
M
ay
2
0
1
9
.
[1
2
]
J
.
T
an
g
,
R.
M.
Ran
g
ay
y
a
n
,
J
.
X
u
,
I.
E
l
N
aq
a,
Y
.
Y
an
g
.
,
“
Co
mp
u
t
er
-
ai
d
ed
d
e
t
ect
i
o
n
an
d
d
i
a
g
n
o
s
i
s
o
f
b
rea
s
t
can
cer
w
i
t
h
mammo
g
rap
h
y
:
recen
t
ad
v
an
ce
s
,
”
IE
E
E
Tr
a
n
s
a
ct
i
o
n
s
o
n
In
f
o
r
m
a
t
i
o
n
Tech
n
o
l
o
g
y
i
n
B
i
o
m
ed
i
ci
n
e
,
v
o
l
.
1
3
,
n
o
.
2
,
p
p
.
2
3
6
–
2
5
1
,
2
0
0
9
.
[1
3
]
G
ad
g
i
l
A
,
Sau
v
a
g
et
C,
Ro
y
N
.,
et
al
.,
“Can
cer
ear
l
y
d
et
ec
t
i
o
n
p
r
o
g
ram
b
as
e
d
o
n
aw
are
n
es
s
an
d
c
l
i
n
i
ca
l
b
r
eas
t
ex
ami
n
at
i
o
n
:
I
n
t
eri
m
re
s
u
l
t
s
fro
m
an
u
r
b
an
c
o
mmu
n
i
t
y
i
n
Mu
mb
a
i
,
In
d
i
a
,”
Th
e
B
r
ea
s
t
,
v
o
l
.
31
,
p
p
.
85
–
89
,
2
0
1
6
.
[1
4
]
P.
S.
V
i
k
h
e
,
V
.
R.
T
h
o
o
l
,
“Mas
s
d
et
ec
t
i
o
n
i
n
mammo
g
ra
p
h
i
c
i
mag
es
u
s
i
n
g
w
a
v
el
e
t
p
ro
ce
s
s
i
n
g
an
d
ad
a
p
t
i
v
e
t
h
res
h
o
l
d
t
ech
n
i
q
u
e
,
”
Jo
u
r
n
a
l
o
f
M
ed
i
ca
l
S
ys
t
em
s
,
v
o
l
.
4
0
,
n
o
.
4
,
p
p
.
1
–
16
,
2
0
1
6
.
[1
5
]
Seel
y
J
.
M.
,
A
l
h
as
s
a
n
T.
,
“
Screen
i
n
g
fo
r
b
reas
t
can
cer
i
n
2
0
1
8
-
w
h
a
t
s
h
o
u
l
d
w
e
b
e
d
o
i
n
g
t
o
d
a
y
?
,”
Cu
r
r
e
n
t
O
n
co
l
o
g
y
,
v
o
l
.
2
5
,
n
o
.
1
,
p
p
.
s
1
1
5
-
s
1
2
4
,
2
0
1
8
.
[1
6
]
E
l
m
o
re
J
.
G
.
, “
Screen
i
n
g
fo
r
Brea
s
t
Can
c
er
,”
J
A
M
A
,
v
o
l
.
2
9
3
,
n
o
.
1
0
,
p
p
.
1
2
4
5
-
1
2
5
6
,
2
0
0
5
.
[1
7
]
Bh
at
e
j
a
V
.
,
Mi
s
ra
M.
,
U
ro
o
j
S.
,
“
H
u
ma
n
v
i
s
u
a
l
s
y
s
t
em
b
as
e
d
u
n
s
h
ar
p
mas
k
i
n
g
fo
r
en
h
an
ceme
n
t
o
f
mammo
g
ra
p
h
i
c
i
mag
e
s
,
”
Jo
u
r
n
a
l
o
f
Co
m
p
u
t
a
t
i
o
n
a
l
S
ci
e
n
ce
,
v
o
l
.
2
1
,
p
p
.
3
8
7
-
3
9
3
,
J
u
l
y
2
0
1
7
.
[1
8
]
W
.
Pen
g
,
R.
May
o
rg
a,
E
.
H
u
s
s
ei
n
,
“A
n
a
u
t
o
mat
e
d
co
n
f
i
rmat
o
ry
s
y
s
t
em
fo
r
an
a
l
y
s
i
s
o
f
mammo
g
rams
,”
Co
m
p
u
t
er
M
et
h
o
d
s
a
n
d
P
r
o
g
r
a
m
s
i
n
B
i
o
m
ed
i
ci
n
e
,
v
o
l
.
1
2
5
,
p
p
.
1
3
4
-
1
4
4
,
2
0
1
6
.
[1
9
]
V
i
j
ay
a
K
.
G
.
,
A
mb
al
i
k
a
S
.
,
“
Co
n
t
ra
s
t
e
n
h
a
n
cemen
t
o
f
mammo
g
ra
p
h
i
c
i
ma
g
es
u
s
i
n
g
w
a
v
el
e
t
t
ra
n
s
f
o
rm
,”
2
0
1
0
3
rd
In
t
er
n
a
t
i
o
n
a
l
Co
n
f
e
r
en
ce
o
n
Co
m
p
u
t
e
r
S
c
i
en
ce
a
n
d
In
f
o
r
m
a
t
i
o
n
Tech
n
o
l
o
g
y
,
p
p
.
3
2
3
–
3
2
7
,
2
0
1
0
.
[2
0
]
Q
as
s
i
m
H
.
,
Bas
h
ee
r
N
.
,
F
arh
an
M
.
,
“
Br
i
g
h
t
n
es
s
p
re
s
er
v
i
n
g
e
n
h
an
ceme
n
t
f
o
r
d
e
n
t
a
l
d
i
g
i
t
al
X
-
ray
i
ma
g
e
s
b
as
e
d
o
n
e
n
t
r
o
p
y
an
d
h
i
s
t
o
g
ram
a
n
a
l
y
s
i
s
,”
Jo
u
r
n
a
l
o
f
A
p
p
l
i
e
d
S
c
i
en
ces
a
n
d
E
n
g
i
n
eer
i
n
g
,
v
o
l
.
2
2
,
n
o
.
1
,
p
p
.
1
8
7
-
1
9
4
.
2
0
1
9
.
[2
1
]
G
o
n
zal
ez
R.
C.
,
R.
E
.
W
o
o
d
s
,
S.
L
.
E
d
d
i
n
s
,
“D
i
g
i
t
a
l
Imag
e
Pro
ces
s
i
n
g
U
s
i
n
g
MA
T
L
A
B,
Seco
n
d
E
d
i
t
i
o
n
,”
G
a
t
e
s
m
a
r
k
P
u
b
l
i
s
h
i
n
g
,
2
0
0
9
.
[2
2
]
McA
n
d
rew
A
.
,
“A
co
m
p
u
t
at
i
o
n
al
i
n
t
ro
d
u
c
t
i
o
n
t
o
d
i
g
i
t
a
l
i
mag
e
p
r
o
ces
s
i
ng
,
”
Tyl
o
r
a
n
d
F
r
a
n
c
i
s
G
r
o
u
p
,
2
0
1
5
.
[2
3
]
Al
-
N
i
ma
R.
R.
O
.
,
D
l
a
y
S.
S.
,
Ch
amb
ers
J
.
A
.
,
W
o
o
W
.
L
.
,
“H
u
man
au
t
h
e
n
t
i
cat
i
o
n
w
i
t
h
fi
n
g
er
t
ex
t
u
re
s
b
a
s
ed
o
n
i
ma
g
e
feat
u
re
en
h
an
ceme
n
t
,
”
2
nd
I
E
T
In
t
er
n
a
t
i
o
n
a
l
Co
n
f
e
r
en
ce
o
n
I
n
t
e
l
l
i
g
e
n
t
S
i
g
n
a
l
P
r
o
ce
s
s
i
n
g
2
0
1
5
(IS
P
)
,
p
p
.
1
-
6
,
2
0
1
5
.
[2
4
]
J
.
Su
ck
l
i
n
g
,
“
T
h
e
Mi
n
-
MA
IS
d
a
t
ab
a
s
e
o
f
mammo
g
rams
,
”
[O
n
l
i
n
e],
A
v
ai
l
ab
l
e
:
h
t
t
p
:
/
/
p
ei
p
a.
es
s
ex
.
ac.
u
k
/
i
n
fo
/
mi
as
.
h
t
ml
,
2
0
1
3
.
[2
5
]
Men
o
n
H
.
P.
,
Raj
es
h
w
ar
i
B.
,
“
E
n
h
an
ceme
n
t
o
f
D
en
t
al
D
i
g
i
t
al
X
-
Ray
Ima
g
es
b
as
e
d
O
n
t
h
e
Imag
e
Q
u
a
l
i
t
y
,”
I
n
t
e
l
l
i
g
en
t
S
ys
t
em
s
Tech
n
o
l
o
g
i
es
a
n
d
A
p
p
l
i
ca
t
i
o
n
s
,
v
o
l
.
5
3
0
,
p
p
.
33
-
4
5
,
2
0
1
6
.
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
Qualit
ati
v
e
as
s
e
s
s
me
nt
of
image
e
nhanc
e
me
nt
algo
r
it
hms
for
mam
mogr
am
s
…
(
M
az
in
N
.
F
ar
han
)
935
B
I
OG
RA
P
H
I
E
S
OF
AU
T
HO
RS
M
a
zi
n
N
.
F
a
rha
n
i
s
aca
d
emi
c
l
ect
u
rer
w
o
rk
s
at
t
h
e
N
o
rt
h
ern
T
ech
n
i
ca
l
U
n
i
v
ers
i
t
y
(N
T
U
).
H
e
recei
v
e
d
h
i
s
Bach
e
l
o
r’
s
d
eg
ree
i
n
Med
i
cal
I
n
s
t
ru
me
n
t
at
i
o
n
T
ech
n
o
l
o
g
y
E
n
g
i
n
eer
i
n
g
fro
m
t
h
e
N
o
rt
h
ern
T
ec
h
n
i
cal
U
n
i
v
ers
i
t
y
i
n
2
0
0
7
an
d
a
Mas
t
e
r’s
d
eg
ree
i
n
Med
i
ca
l
D
ev
i
ces
fro
m
t
h
e
U
n
i
v
er
s
i
t
y
o
f
S
t
rat
h
cl
y
d
e,
G
l
a
s
g
o
w
,
U
K
i
n
2
0
1
4
.
H
e
h
av
e
s
t
u
d
i
ed
an
d
w
o
rk
e
d
w
i
t
h
med
i
ca
l
In
s
t
ru
me
n
t
a
t
i
o
n
a
n
d
h
av
e
h
ad
t
h
e
ch
a
n
ce
t
o
g
e
t
an
i
n
d
ep
t
h
u
n
d
ers
t
an
d
i
n
g
o
f
me
d
i
ca
l
d
ev
i
ces
an
d
t
h
e
i
r
en
v
i
ro
n
men
t
.
H
e
i
s
l
o
o
k
i
n
g
t
o
u
s
e
h
i
s
en
g
i
n
eeri
n
g
s
k
i
l
l
s
i
n
i
m
p
ro
v
i
n
g
t
h
e
cu
rren
t
s
t
a
t
e
o
f
me
d
i
ca
l
d
e
v
i
ce
s
.
H
i
s
t
o
p
re
s
earch
i
n
t
eres
t
t
o
p
i
c
s
are
med
i
ca
l
i
ma
g
e
p
r
o
ces
s
i
n
g
,
b
i
o
s
i
g
n
al
p
ro
ce
s
s
i
n
g
an
d
med
i
cal
s
en
s
o
r
s
.
Pl
ea
s
e
emai
l
mazi
n
.
n
a
d
h
eer@
n
t
u
.
ed
u
.
i
q
o
r
g
o
t
o
h
t
t
p
s
:
/
/
w
w
w
.
l
i
n
k
e
d
i
n
.
co
m
/
i
n
/
maz
i
n
-
n
-
farh
a
n
-
2
a
1
3
6
7
a
7
/
t
o
co
n
t
ac
t
Mazi
n
.
M
o
ha
m
m
ed
Gha
ni
m
A
y
o
ub
i
s
a
L
ect
u
rer
at
t
h
e
T
ech
n
i
ca
l
E
n
g
i
n
eeri
n
g
Co
l
l
eg
e
o
f
Mo
s
u
l
-
N
o
r
t
h
er
n
T
ec
h
n
i
cal
U
n
i
v
ers
i
t
y
(N
T
U
)
s
i
n
ce
2
0
1
5
.
Mo
h
amme
d
co
m
p
l
e
t
ed
h
i
s
Bac
h
el
o
r
o
f
E
n
g
i
n
eeri
n
g
i
n
Co
m
p
u
t
er
T
ech
n
o
l
o
g
y
fr
o
m
N
o
r
t
h
ern
T
ech
n
i
ca
l
U
n
i
v
ers
i
t
y
,
i
n
2
0
0
6
,
an
d
co
m
p
l
et
e
d
h
i
s
MSc
i
n
E
l
ect
r
o
n
i
c
Co
mm
u
n
i
cat
i
o
n
s
a
n
d
C
o
mp
u
t
er
E
n
g
i
n
eeri
n
g
a
t
U
n
i
v
er
s
i
t
y
o
f
N
o
t
t
i
n
g
h
am
i
n
2
0
1
5
.
H
i
s
p
r
i
mary
res
earch
i
n
t
er
es
t
s
rev
o
l
v
e
aro
u
n
d
t
h
e
ap
p
l
i
cat
i
o
n
o
f
s
i
g
n
al
p
r
o
ces
s
i
n
g
an
d
s
en
s
o
r
d
a
t
a
t
ech
n
i
q
u
es
t
o
s
o
l
v
e
o
r
u
n
d
ers
t
an
d
b
i
o
m
ed
i
cal
en
g
i
n
eeri
n
g
p
ro
b
l
em
s
.
Ha
s
s
a
n
M
es
s
a
r
Qa
s
s
i
m
i
s
a
l
ect
u
rer
at
N
o
rt
h
ern
T
e
ch
n
i
cal
U
n
i
v
ers
i
t
y
(N
T
U
)
w
i
t
h
t
w
o
cert
i
f
i
cat
e
s
i
n
Me
d
i
ca
l
E
n
g
i
n
eer
i
n
g
,
BE
n
g
an
d
MSc.
H
e
o
b
t
a
i
n
e
d
h
i
s
mas
t
er
d
eg
ree
fr
o
m
U
n
i
v
er
s
i
t
y
o
f
St
rat
h
cl
y
d
e
i
n
2
0
1
3
.
H
i
s
res
earc
h
fi
el
d
i
s
t
h
e
med
i
ca
l
i
mag
e
p
ro
ce
s
s
i
n
g
an
d
med
i
ca
l
s
en
s
o
r
s
.
A
bdul
ra
h
m
a
n
K.
Ees
ee
i
s
a
L
ect
u
rer
i
n
t
h
e
T
ech
n
i
ca
l
E
n
g
i
n
eer
i
n
g
Co
l
l
eg
e
o
f
Mo
s
u
l
/
N
o
r
t
h
ern
T
ech
n
i
ca
l
U
n
i
v
er
s
i
t
y
,
Ira
q
.
H
o
l
d
i
n
g
M
as
t
er
d
eg
ree
MSc
i
n
t
h
e
B
i
o
me
d
i
ca
l
E
n
g
i
n
eeri
n
g
fro
m
t
h
e
U
n
i
v
ers
i
t
y
o
f
St
rat
h
c
l
y
d
e/
U
n
i
t
e
d
K
i
n
g
d
o
m
i
n
2
0
1
5
t
h
r
o
u
g
h
a
fu
l
l
y
fu
n
d
ed
s
c
h
o
l
ars
h
i
p
fro
m
t
h
e
Ira
q
i
g
o
v
er
n
men
t
,
an
d
h
o
l
d
i
n
g
Bach
e
l
o
r
o
f
E
n
g
i
n
eer
i
n
g
B.
E
n
g
i
n
t
h
e
Med
i
cal
In
s
t
r
u
men
t
at
i
o
n
T
ech
n
o
l
o
g
y
E
n
g
i
n
eer
i
n
g
.
T
h
e
mai
n
res
earch
i
n
g
fi
el
d
i
s
t
h
e
Bi
o
-
s
i
g
n
al
p
ro
ce
s
s
i
n
g
,
med
i
ca
l
s
en
s
o
r
s
an
d
i
mag
e
p
r
o
ces
s
i
n
g
.
Evaluation Warning : The document was created with Spire.PDF for Python.