Internati
o
nal
Journal of Ele
c
trical
and Computer
Engineering
(IJE
CE)
V
o
l.
6, N
o
. 2
,
A
p
r
il
201
6, p
p
.
71
7
~
72
4
I
S
SN
: 208
8-8
7
0
8
,
D
O
I
:
10.115
91
/ij
ece.v6
i
2.9
029
7
17
Jo
urn
a
l
h
o
me
pa
ge
: h
ttp
://iaesjo
u
r
na
l.com/
o
n
lin
e/ind
e
x.ph
p
/
IJECE
Advan
cement in Research T
echniques on Medical Imaging
Processing for Breast
Cancer Detection
Sushm
a S J*, S
C
Pr
asann
a
Kum
a
r**
* Instrumentatio
n Technolog
y
,
V
i
svesvaray
a
Technological Univ
ersity
, Belg
avi, I
ndia
** Departmen
t
o
f
Instrumentatio
n
Technolog
y
,
R
V
CE Bangalore, India
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Sep 14, 2015
Rev
i
sed
No
v
30
, 20
15
Accepted Dec 18, 2015
W
ith the adva
ncem
ent of m
e
dica
l im
age pr
oces
s
i
ng, the
a
r
ea of the
heal
thcar
e s
e
cto
r
has
s
t
art
e
d r
e
c
e
iving
the b
e
nef
its
of th
e m
oder
n
aren
a of
diagnostic tools
to iden
tif
y
th
e diseases
eff
ectively
.
C
a
ncer is
one of th
e
dreaded
dis
e
as
es
, where
s
u
cc
es
s
factor
of tr
ea
tm
ent offer
e
d b
y
m
e
dica
l s
e
c
t
o
r
is still
an
unsolv
e
d probl
em
. Hen
ce,
th
e succ
ess f
actor
of th
e
tre
a
t
m
e
nt li
es in
e
a
r
ly
st
age
of the
di
se
a
s
e
or time
ly
det
e
c
t
i
on
of t
h
e
di
se
a
s
e.
T
h
i
s
pa
pe
r
discusses about the adv
a
ncement be
ing mad
e
in th
e medical
image
processing towards an effective diagnosis of the breast cancer from the
mammogram i
m
age in radiolog
y. Ther
e has been enough resear
ch activ
ity
with various so
rts of advan
ces
tech
n
i
ques being implemen
ted
in th
e past
decad
e.
The pr
im
e contribu
tio
n of this
m
a
nus
cript is
to s
h
owcas
e the
advancem
ent of
the technolog
y
along with
illustr
a
tion of th
e effectiv
eness of
the
existing
li
ter
a
tures wi
th r
e
spe
c
t
to r
e
sear
ch ga
p.
Keyword:
Breast Can
c
er
Detectio
n
Cancer Detection
M
a
m
m
ogram
R
a
di
ol
o
g
y
Copyright ©
201
6 Institut
e
o
f
Ad
vanced
Engin
eer
ing and S
c
i
e
nce.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
Sus
h
m
a
S J
Research Sc
holar
Inst
rum
e
nt
at
i
on Tec
h
nol
ogy
Vi
sves
va
ray
a
Tech
nol
ogi
cal
Un
i
v
ersity, Bel
g
avi, India
E-Mail: su
sh
jgo
w
d
a
@g
m
a
il.c
o
m
1.
INTRODUCTION
In t
h
e rece
nt
days the early
detection
of
cancer
i
s
very
m
u
ch i
n
di
spe
n
sa
bl
e fo
r i
m
pr
o
v
i
n
g t
h
e
sur
v
i
v
al
rat
e
s,
as i
t
i
s
one
of
t
h
e m
o
st
chal
l
e
ngi
ng
an
d
del
i
b
erat
ed t
opi
cs
whi
c
h are
bei
ng
di
sc
usse
d a
m
ong
th
e research
scien
tists. Can
cer refers to
so
m
e
m
o
l
ecu
lar even
ts wh
ich
are n
o
t
h
i
n
g
b
u
t
mu
ltip
licatio
n
o
f
a set
of
cel
l
s
i
n
a
part
i
c
ul
a
r
area
of
o
u
r
bo
dy
w
h
ere
t
h
e u
n
co
nt
r
o
l
l
a
bl
e gr
owt
h
of cells can
form
micro-
calcificatio
n
s
an
d lu
m
p
th
is typ
e
o
f
stru
ctural d
i
stortio
n
of c
e
l
l
s
are m
e
nt
i
oned
t
o
as m
a
l
i
gna
nt
t
u
m
o
rs
[
1
]
.
I
n
case of a
canc
e
r cell it is
observe
d
t
h
at the
cell overgr
o
w
t
h
ca
nn
ot
be
d
i
sal
l
o
wed
by
t
h
e c
o
nt
rol
sy
st
em
as
th
eir fun
c
tion
a
lity
b
eco
m
e
d
i
sab
l
ed, it h
a
s been
fo
und
th
at th
e can
cerou
s
cells g
r
ow and sp
lit in
th
e p
r
esen
ce
of a si
g
n
al
wh
i
c
h i
s
norm
a
l
l
y
respo
n
si
bl
e f
o
r p
r
e
v
ent
i
n
g t
h
e gr
o
w
t
h
o
f
any
m
a
l
i
gnant
t
u
m
o
r or cel
l
s
. T
h
e
d
e
v
e
l
o
p
m
en
t of th
ese cells g
e
n
e
rates
n
e
w characteristic
s which i
n
cludes t
h
e form
ation
of ne
w cell structure
and produce
new e
n
zym
e
s which all
o
ws
cells to
divide
and
g
r
ow.
Ab
no
rmali
ties in
th
e
cells wh
ich
cau
s
e t
h
e
cancer m
a
i
n
l
y
devel
o
p
f
r
om
the m
u
t
a
t
i
on w
h
i
c
h ca
n
occu
r
i
n
t
h
e
pr
ot
ei
n
-
enc
o
ded
ge
ne
s t
h
at
i
s
res
p
o
n
si
bl
e
for con
t
ro
llin
g th
e abn
o
rm
al
g
r
o
w
t
h
of th
e cells. And
the g
e
n
e
s
wh
ich are o
f
ten
work
to
rep
a
ir the DNA
d
a
m
a
g
e
, so
m
e
t
i
m
es d
o
n
’
t
work
p
r
o
p
erly and resu
lt th
e
furth
e
r abno
rm
alit
ies lik
e it allo
ws th
e ab
no
rm
al cells
t
o
di
vi
de
m
o
re ra
pi
dl
y
as c
o
m
p
ared t
o
t
h
e
no
rm
al
cel
l
s
. This study
gives
a better ove
r
vi
ew
of cance
rous cells
wh
ere it is also
said
t
h
at if th
ese e
nha
nce
growt
h
of the t
i
ssues are
found
in
th
ei
r orig
i
n
al lo
cation
then
th
ey
are conside
r
ed
as beni
gn else if those cells are found to
be i
nvasi
ve i
n
t
h
a
t
case t
h
ey
spread ve
ry
qui
c
k
l
y
and
un
desi
ra
bl
y
t
h
e
n
t
h
ose
t
i
ssues
are c
onsi
d
ere
d
as m
a
li
gnant
t
i
ssues.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
6, No
. 2, A
p
ri
l
20
16
:
71
7 – 7
2
4
71
8
Sect
i
on
1.
1 di
s
c
usses a
b
o
u
t
t
h
e back
gr
o
u
n
d
of t
h
e st
udy
fo
l
l
o
wed
by
p
r
o
b
l
em
descri
pt
i
o
n i
n
Sect
i
o
n
1.
2.
Sect
i
on
2
di
scuss
e
s a
b
o
u
t
t
h
e exi
s
t
i
n
g r
e
search
w
o
r
k
f
o
l
l
o
we
d
by
di
s
c
ussi
o
n
of
rese
arch
ga
p i
n
Se
ct
i
on
3
.
C
oncl
u
si
o
n
i
s
di
scuss
e
d
i
n
Se
ct
i
on 4.
1.
1. B
a
ck
gr
ou
nd
The st
udy
[2]
[
3
]
say
s
t
h
e
r
e c
a
n
be m
o
re t
h
a
n
20
0
di
f
f
ere
n
t
t
y
pes o
f
ca
nce
r
, a
n
d
cance
r
o
us cel
l
s
ca
n
be de
vel
o
ped i
n
any
or
ga
ns o
f
a bo
dy
, i
t
i
s
f
o
u
n
d
t
h
at
t
h
ere
are 60 di
ffe
re
nt
or
ga
ns w
h
i
c
h can be a
ffect
ed by
the seve
re attacks
of ca
nce
r
. Som
e
of the
vital states of
Breast cance
rs
are i)
Ductal
Carcinom
a a
nd ii
)
In
vasi
ve
D
u
ct
al
C
a
rci
n
om
a. D
u
ct
al
C
a
rci
nom
a i
s
al
so
r
e
fer
r
ed
t
o
t
h
e
no
n
-
i
n
vasi
ve
o
r
pre
-
i
n
vasi
ve
breast
cancer.
The Invasive Ductal
C
a
rci
nom
a i
s
a very
com
m
on t
y
pe of
brea
st
cancer w
h
i
c
h us
ual
l
y
ori
g
i
n
at
es
fr
om
t
h
e
m
i
l
k
duct
of
t
h
e
br
east
s
an
d I
n
va
si
ve d
u
ct
al carcinom
a sprea
d
through
th
e
wall o
f
th
e
ducts an
d
d
e
v
e
l
o
ps in
to
th
e fatty tissu
es o
f
t
h
e
b
r
east
.
Variou
s ot
he
r f
o
rm
s of i
n
v
a
si
ve b
r
east
ca
ncer a
r
e m
e
du
l
l
a
ry
carcino
m
a
, ad
en
osqu
am
o
u
s
carcino
m
a
, p
a
p
illary carcinoma,
Co
llo
id
carcino
m
a, cysti
c
carcino
m
a
et
c. Th
is
pape
r em
phasi
zes o
n
t
h
e
det
a
i
l
di
scussi
o
n
of t
h
e va
ri
o
u
s t
echni
q
u
es w
h
i
c
h
we
re devel
ope
d fo
r
a
n
ef
fect
i
v
e
Breast Can
cer
Detectio
n
an
d
th
e classificatio
n. Th
is st
u
d
y
ai
m
s
to
fill th
e
g
a
p, wh
ich
are asso
ciated
with
th
e
devel
opm
ent
o
f
t
h
e
p
r
evi
o
u
s
pr
o
pose
d
t
e
c
h
n
i
ques
.
1.
2. T
h
e Pr
obl
em
Medical im
ag
e processing has pl
ayed
a sig
n
i
fican
t ro
le in
d
i
g
ital i
m
ag
in
g
th
at is u
s
ed
for
in
v
e
stig
ation
of th
e m
a
mm
o
g
ram
s
. At presen
t, th
ere ex
i
s
t v
a
riou
s
d
i
ag
no
stic too
l
s th
at is u
s
ed
in d
i
g
ital
i
m
ag
in
g
o
f
b
r
east reg
i
o
n
fo
r id
en
tification
o
f
an
y abno
rm
alities o
f
can
cerou
s
typ
e
. This sectio
n
d
i
scu
sses
ab
ou
t th
e
v
a
rio
u
s
im
ag
in
g
tech
n
i
q
u
e
s and to
o
l
s
wh
ich
were
u
tilized
fo
r t
h
e
b
r
east
can
cer
d
e
tectio
n, it i
s
concl
ude
d i
n
t
h
e st
u
d
y
of
[
3
]
[4]
t
h
at
t
h
ere
are
very
p
o
or ev
id
en
ces th
at
sup
port th
e st
ate
m
ent that clinical
b
r
east ex
am
in
atio
n
s
alon
g
with
screen
ing
ma
mm
o
g
r
a
m
s
can
min
i
m
i
ze
th
e
m
o
rtality r
a
tes fro
m
th
e
b
r
east
cancer
. S
o
m
e
of t
h
e
i
m
port
a
nt
b
r
east
canc
e
r i
m
agi
ng t
e
c
h
ni
que
s i
s
di
scus
s
e
d
ove
r
here
.
1.
2.
1. M
a
mm
o
g
ra
ph
y
Ma
mm
o
g
r
aphy is an
X
-
r
a
y tech
n
i
q
u
e
th
at
w
a
s in
tro
d
u
c
ed sp
ecif
i
cally to
ex
amin
e th
e b
r
east lesio
n
s
[5]
.
Di
a
g
n
o
si
s,
eval
uat
i
on a
n
d t
h
e det
e
rm
i
n
at
i
on o
f
t
h
e det
ect
ed im
ages and t
h
e re
sul
t
s
are do
ne ba
sed
on t
h
e
di
ffe
re
nt
abs
o
r
p
t
i
on sc
hem
e
s
of t
h
e
X-
ray
s
whi
c
h can
be o
ccur
bet
w
ee
n d
i
ffere
nt
t
y
pes o
f
bre
a
st
t
i
ssues
suc
h
as fat
,
fi
br
o
gl
an
dul
a
r
t
i
ssu
es, cy
st
s ,
t
u
m
o
rs and
va
ri
ous
m
i
cro an
d
m
acro cal
ci
fi
cat
i
ons.
It
ha
s bee
n
diagnosed and seen that the
macro
calcifications are not usually relate
d to the cancer. Figure 1 shows the
visuals of normal
m
a
mmogra
m
im
age.
Fi
gu
re
1.
Exa
m
pl
e of M
a
m
m
ogram
1.
2.
2. C
o
mpu
t
ed
T
o
m
o
gra
p
hy (CT
)
There
are s
o
few be
ne
fits are associated
with th
e Com
p
uted Tom
o
graphy (Fi
g
ure
2)
as high c
o
st
effi
ci
ency
i
s
t
h
ere t
o
pe
rf
or
m
t
h
i
s
t
echni
q
u
e i
n
a
hum
an
breast
a
nd
hi
gh e
x
po
su
re o
f
ra
di
at
i
on i
s
anot
he
r
dem
e
ri
t
of t
h
i
s
t
echni
que
.
It
i
s
n
o
t
s
o
m
u
ch
usef
ul
as i
ndi
c
a
t
i
ons
of
t
h
i
s
t
echni
que
res
u
l
t
som
e
l
i
m
i
t
a
t
i
ons
i
n
the real tim
e scenari
o
[5].
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Adv
ance
m
e
n
t
i
n
Rese
arc
h
Te
chni
que
s
on
M
e
di
cal
Im
a
g
i
n
g
Proce
ssi
n
g
f
o
r
Brea
st
C
a
ncer
…
(
Sus
hm
a S J)
71
9
Fi
gu
re
2.
Exa
m
pl
e of C
T
sca
n
i
m
age of
b
r
e
a
st
1.
2.
3. Ul
tr
as
o
n
o
g
ra
ph
y
Ul
t
r
aso
u
nd
use
s
hi
g
h
f
r
e
que
n
c
y
sou
n
d
t
o
i
n
vest
i
g
at
e t
h
e i
n
t
e
rnal
st
r
u
ct
u
r
e
of t
h
e t
i
ssues
or
n
odes
o
r
m
a
sses inside the breast (Fi
g
ure 3). It
is on
e
o
f
th
e co
st effectiv
e an
d
frequen
tly u
s
ed
im
a
g
ing
techn
i
qu
es for
i
nvest
i
g
at
i
o
n o
f
di
sease resi
di
ng i
n
breast
.
N
o
rm
al
l
y
used i
n
vi
s
u
al
i
z
i
ng p
r
eg
na
ncy
st
age
s
, ul
t
r
as
ou
n
d
i
s
al
so
use
d
in breast
cancer. Ultras
ound is one of the m
o
st efficient br
east ca
ncer im
aging t
echni
que
s whi
c
h is
appl
i
e
d
t
o
t
h
e
m
a
m
m
ography
an
d t
h
e
phy
si
cal
exam
i
n
at
i
on
of
t
h
e
pat
i
e
nt
, B
r
east
ul
t
r
aso
u
n
d
i
s
ap
pl
i
e
d t
o
discrim
i
nate a cyst and a s
o
lid m
a
ss, it is us
ed to re
veal
t
h
e p
a
lpab
le i
rreg
u
l
arity wh
ich
is no
t easy to detect in
prese
n
t of noisy
m
easurem
e
n
ts of t
h
e im
a
g
e. Ultra
s
oun
d is n
o
t
v
e
ry
m
u
ch
effective in
d
e
tectin
g
m
i
cro
-
calcificatio
n
s
,
b
u
t
it is
u
s
efu
l
for th
e n
e
ed
le l
o
caliza
tion
under s
o
m
e
specific guid
elines
. C
o
m
p
licated imaging
and thei
r res
p
e
c
tive analysis cannot be
perform
ed on m
a
m
m
ogram
s [5]
.
At
prese
n
t
,
we
have
bot
h 3
D
and
4D
ul
t
r
aso
u
n
d
t
ech
nol
ogi
es
t
o
sca
n
t
h
e
o
n
c
o
l
o
gi
cal
regi
on
[
6
]
.
Fig
u
r
e
3
.
Ex
am
p
l
e o
f
U
ltr
aso
und
1.2.4. Posi
tion
Em
ission T
o
mographic Sc
reening
(PET)
In t
h
is technique a
radi
oactive c
h
em
ical
is adm
i
ni
st
ered
t
o
t
h
e
pat
i
e
nt
bef
o
re
pe
rf
o
r
m
i
ng sca
n
i
n
order to see t
h
e
hotspot areas under the
scan. T
h
e
hotspot a
r
eas a
r
e those
ar
eas
of c
r
itical clinical
sig
n
i
fican
ce. Early stu
d
y
of [5
] sugg
ested
th
at
m
e
tab
o
lic activ
ities rais
ed
in
presen
ce o
f
can
cer
which
is
det
ect
abl
e
u
s
i
n
g fl
uo
ri
ne
1
8
-l
abel
ed
gl
uc
ose
.
PET
(Fi
g
u
r
e
4) i
s
use
d
t
o
d
i
agn
o
se t
h
e
be
ni
g
n
f
r
o
m
m
a
l
i
gna
nt
l
e
si
ons.
It
i
s
al
so
ob
ser
v
e
d
t
h
at
PET m
a
y
be
a technique
for the
staging
of breast canc
e
rs.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
6, No
. 2, A
p
ri
l
20
16
:
71
7 – 7
2
4
72
0
Fi
gu
re
4.
Exa
m
pl
e of PET
I
m
age
1.2.5. Nucle
a
r medicine
Breast
Im
aging
C
once
n
t
r
at
e
d
Tech
net
i
u
m
-
99
sest
am
i
b
i
has
bee
n
det
ected in s
o
m
e
breas
t cancer
s. Th
e ro
le
of th
is
tech
n
i
qu
e is
n
o
t well d
e
m
a
rcated
as it is no
t
su
fficien
t
t
o
di
ffe
rent
i
a
t
e
l
e
si
ons
f
r
o
m
m
a
ligna
nt
t
u
m
o
rs (
F
i
g
u
r
e
5)
. It
i
s
consi
d
ere
d
as a l
e
ss effi
ci
ent
t
echni
que i
n
t
h
e
fi
el
d of C
a
nce
r
Det
ect
i
on t
echni
que
s usi
n
g
im
age
pr
ocessi
ng
[
7
]
.
Fi
gu
re
5.
Exa
m
pl
e of N
u
cl
e
a
r m
e
di
ci
nes
1.
2.
6.
M
a
g
n
eti
c
Res
o
n
a
nce
I
m
a
g
i
n
g
(
M
R
I
)
The s
p
eci
fi
ci
t
y
o
f
M
R
I t
e
c
h
ni
que
i
s
not
so
h
i
gh,
as i
t
i
s
a
p
pl
i
cabl
e
i
n
va
ri
ous
di
sease
det
ect
i
ons.
A
s
M
R
I
doe
s
no
t
gi
ve
a
bet
t
e
r
scan
re
p
o
rt
o
f
t
h
e
m
i
cro cal
ci
fi
cat
i
ons
but
som
e
few
t
y
p
e
s o
f
M
R
I
s
u
c
h
as
Dy
nam
i
c C
ont
rast
En
ha
ncem
ent
M
R
I’s ca
n
be u
s
ef
ul
t
o
d
e
t
ect
t
h
e
m
a
l
i
g
nancy
of t
h
e n
o
n
-
pal
p
a
b
l
e
l
e
si
on
s
(Figure
6). Breast cance
r rec
u
rrence
ve
rs
us
fibrosis
p
r
esent
s
t
h
e best
i
n
di
c
a
t
i
on fo
r breast
M
R
I
[
5
]
.
Fi
gu
re
6.
Exa
m
pl
e of M
R
I
I
m
agi
n
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Adv
ance
m
e
n
t
i
n
Rese
arc
h
Te
chni
que
s
on
M
e
di
cal
Im
a
g
i
n
g
Proce
ssi
n
g
f
o
r
Brea
st
C
a
ncer
…
(
Sus
hm
a S J)
72
1
1.
2.
7. Prepr
o
c
e
ssi
ng of
Ma
mmo
gra
m
s
Fro
m
th
e in
fo
rmatio
n
h
i
gh
ligh
t
ed
in
t
h
e stud
y o
f
[7
] it ca
n
b
e
seen
that cu
rren
t ev
id
ence sup
p
o
r
t
s
th
at in
th
e recen
t
ti
m
e
s th
e a
p
p
lication
of ma
mm
o
g
r
a
m
s
are m
u
ch beneficial than
in
the past for t
h
e early
breast
ca
nce
r
d
e
t
ect
i
on i
n
w
o
m
e
n. M
a
m
m
o
g
ram
s
sh
oul
d
be a
ppl
i
c
a
b
l
e
f
o
r
t
h
e
w
o
m
e
n of
al
l
ages
as l
o
n
g
a
s
she
get
s
af
fect
ed
by
any
t
y
pe
of
seri
ous
, c
h
ro
ni
c h
ealt
h
di
sease. T
h
e c
u
rrent e
v
ide
n
ce
gives a
confi
r
mation
ab
ou
t th
e sub
s
t
a
n
tial b
e
n
e
fits
o
f
m
a
mm
o
g
r
am
s fo
r th
e
women
o
f
40
’s. Sectio
n
IV
will d
i
scu
s
s abou
t th
e
work
of
past
t
e
n y
e
a
r
s w
h
e
r
e di
f
f
e
r
ent
t
y
pes o
f
t
e
chni
que
s w
h
i
c
h ha
ve
bee
n
de
vel
o
ped t
o
i
m
pro
v
e t
h
e c
o
nt
ra
st
and
th
e q
u
a
lity o
f
th
e m
a
mmo
g
r
am
s
fo
r th
e breast can
cer d
e
tec
tio
n
h
a
v
e
b
e
en
d
i
scu
s
sed, wh
ich
will b
e
v
e
ry
m
u
ch
bene
ficial fo
r the f
u
rt
her sta
g
es of
brea
st
cancer detection pipeline.
Ca
nc
er
in
vo
l
v
es th
e ab
no
r
m
al g
r
ow
th
or
u
n
c
on
tro
lled
mu
ltip
licatio
n
of cells in
a p
a
rticu
l
ar reg
i
on
of an
y b
o
d
y
. Breast Can
cer is also
a typ
e
o
f
can
cer
refe
rs t
o
a
f
o
r
m
of
m
a
l
i
gnan
t
t
u
m
o
r, us
ual
l
y
gr
ow
n
by
t
h
e rapi
d
di
vi
si
o
n
of
brea
st
cel
l
s
. It
i
s
ve
ry
m
u
ch
essent
i
a
l
t
o
ap
pl
y
t
h
e de-
n
oi
s
i
ng a
nd C
ont
ra
st
enha
ncem
ent
t
echni
q
u
es i
n
t
h
e m
a
mm
ogram
s, as t
h
ey
d
o
n
o
t
pr
o
v
i
d
e
a
very
go
od co
nt
rast
bet
w
een t
h
e
no
rm
al gl
and
u
l
ar t
i
ssues and
t
h
e
m
a
l
i
gnant
t
u
m
o
r t
i
ssues ,t
hi
s
happe
n
s beca
use The X-ray attenuation bet
w
een the
s
e two
tissu
es rep
r
esen
ts v
e
ry few d
i
ssi
m
ilarit
i
es. So
th
e
radi
ologists face problem
when they
m
a
nually differen
tiate
the norm
a
l and cancerous tissues. T
h
is lim
i
tations
associ
at
ed
wi
t
h
t
h
e
breast
ca
ncer
det
ect
i
on
fr
om
t
h
e
m
a
l
i
gnant
t
i
ssue
s
ca
n be
red
u
ce
d b
y
un
derst
a
ndi
n
g
t
h
e
linear a
b
s
o
rption coefficients
o
f
vari
ous
t
i
s
s
u
es
w
h
i
c
h
di
sc
usses
wel
l
a
b
o
u
t
t
h
e
bet
t
e
r
c
ont
rast
of a
n
i
m
age.
Im
ag
e Co
n
t
rast wh
ich
is acqu
ired
throug
h th
e lin
ear ab
so
r
p
tion
coe
fficie
n
ts is
defi
ned
by
the Bee
r-L
am
bart
law.
x
e
I
I
0
(
1
)
Th
e ab
ov
e equ
a
tio
n d
e
fin
e
s th
e in
tensity o
f
t
h
e Electromag
n
e
tic (EM
)
wav
e
(I
o
) the atten
u
a
tion
coefficient of t
h
e m
a
teria
l
is
defi
ned as
β
and
th
e leng
th
of th
e
m
a
terial
is
x
t
h
ro
ug
h w
h
i
c
h t
h
e wa
ve i
s
bei
n
g
transm
itted and here t
h
e (I) represe
n
ts the electro
m
a
gn
et
i
c
wave.
Th
e n
o
i
s
e i
s
u
n
w
ant
e
d a
nd
re
du
n
d
ant
inform
ation present in a
n
im
age and makes the de
te
ction of the cancerous m
a
ligna
nt tissues
m
o
re
challengi
ng. It
has
been
seen that
th
e no
ise p
r
esen
t in
an
i
m
ag
e in
crease with
th
e
p
i
x
e
l
in
ten
s
ity wh
ere th
e
lo
cal co
n
t
rast an
d
th
e im
ag
e in
ten
s
ity
are considere
d
to
be inde
pe
nden
t. Th
is affects
th
e
m
a
mm
o
g
r
a
m
s, a
sol
u
t
i
o
n
fo
r t
h
i
s
pr
o
b
l
e
m
has bee
n
pr
o
pose
d
by
va
ri
o
u
s s
t
udi
es
w
h
ere t
h
e
noi
se
eq
ual
i
zat
i
on t
ech
ni
q
u
e
has
b
een app
lied
t
o
ob
tain
t
h
e i
m
ag
es wh
ere t
h
e lo
cal co
n
t
rast is al
m
o
st same all th
e i
m
a
g
e in
ten
s
ities. Fo
r th
e
im
pro
v
em
ent
of
t
h
i
s
t
e
c
hni
que
va
ri
o
u
s
a
u
t
h
ors
di
sc
uss
e
d t
h
e
e
nha
n
c
em
ent
of
t
h
e
n
o
i
s
e e
qual
i
zat
i
on
t
echni
q
u
e.
In t
h
i
s
t
echni
que a
poi
nt
i
s
consi
d
ere
d
as a nei
g
hb
o
r
h
o
od
of a
n
im
age l
o
cat
i
on
(
x, y
). The l
o
cal
co
n
t
r
a
st of
t
h
is n
e
ighb
orh
ood
is calcu
lated
as
)
,
(
)
,
(
)
,
(
y
x
median
y
x
f
y
x
C
(
2
)
Whe
r
e C(x, y)
is the calculated as the
local c
ont
rast, f(x,y) is the im
age grey
l
e
vel
at
(x, y
)
and m
e
di
an (x
, y
)
is th
e m
e
d
i
an
gr
ey lev
e
l
o
f
t
h
e n
e
ighb
orh
ood
.
Fi
gu
re 7.
I
d
ent
i
fi
cat
i
on of
Microcalcification
Fig
u
re 7
sh
ows th
e d
e
tectio
n o
f
th
e m
i
cro
calcificatio
n
,
wh
ich
is b
a
sical
ly a p
o
r
tio
n
iden
tified
to
pos
ses m
i
neral
dep
o
si
t
sprea
d
o
v
er s
p
eci
fi
c
po
rt
i
on
of t
h
e
m
a
m
m
ogram
. The speci
fi
c i
d
ent
i
f
i
e
d p
o
si
t
i
on i
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
6, No
. 2, A
p
ri
l
20
16
:
71
7 – 7
2
4
72
2
l
o
cat
ed by
usi
ng
b
o
u
n
d
i
n
g b
o
x
.
Thi
s
i
s
t
h
e
no
rm
al
wa
y t
o
id
en
tify th
e
p
o
rtion
su
s
p
ec
ted of the ca
ncer or
certain
fo
rm
o
f
ab
no
rm
ality
i
n
breast. Th
e nex
t
sec
tio
n
will d
i
scu
ss ab
ou
t
th
e ex
istin
g
tech
n
i
q
u
e
s in
med
i
cal
im
age processi
ng towards
bre
a
st cancer detection.
2.
E
X
ISTI
NG R
E
SEAR
CH
M
ETHODOL
O
G
IES
Th
ere is a
sign
ifican
t con
t
ri
b
u
tion
of th
e
med
i
cal
im
age proces
sing in th
e a
r
ea
of breast ca
ncer
d
e
tectio
n. Majo
rity of th
e ex
istin
g
sy
st
em
co
nsi
s
t
s
o
f
pre
p
r
o
cessi
ng
,
feat
u
r
e e
x
t
r
a
c
t
i
on,
segm
ent
a
t
i
on,
fo
llowed
b
y
detectio
n
o
f
th
e can
cero
u
s
p
o
rtio
n
.
Howev
e
r,
we will atte
mp
t d
i
fferen
tly t
o
un
d
e
rstand
if th
ere
are an
y fo
rm
s o
f
op
timizatio
n
tech
n
i
q
u
e
s
bein
g im
p
l
e
m
en
ted
in th
e
p
a
st i
n
o
r
d
e
r to enh
a
n
ce t
h
e
p
r
ob
abilit
y o
f
t
h
e det
ect
i
on r
a
t
e
. Thi
s
sect
i
on
di
scusse
s abo
u
t
t
h
e t
echni
que
s ad
opt
e
d
by
t
h
e researc
h
ers i
n
or
der t
o
det
ect
the brea
st canc
e
r from
the
m
a
mmogram
.
There are va
rious
studies in the
pa
st exploring an effective process
of
detection m
echanism
of breast cance
r.
Hence
,
in
or
d
e
r to
show b
e
tt
er ou
tco
m
es, we fo
cus on
on
ly th
e
significa
nt studies ca
rried
out towards
investigati
on of ma
mm
ogra
m
s
only
from
the m
o
st recent
studies
carried ou
t i
n
l
a
st 5
years. Tab
l
e 1 will sh
ow th
e m
o
st sig
n
i
fican
t
work toward
s
b
r
east can
cer d
e
tection.
Tabl
e
1.
Su
rve
y
of
Exi
s
t
i
n
g T
echni
que
s
Appr
oach
Author
M
e
thodology
Outco
m
e
I
n
fer
e
nce
1
Conventio
nal
T
echniques
Zheng et al. [8]
Gabor
Filter,
Edge
Histogr
am
Descr
i
p
t
or
90% accuracy
Not
f
o
cused
on
classif
i
cation, no
bench
m
ar
king
2
Nagi et al.
[9]
Region of I
n
ter
e
st
E
ffective
seg
m
entation
-Do
-
3
Ganesan [10]
CAD based detection
Good
theor
e
tic
al
discussion
-N/A-
4 Hefnawy
[11]
W
a
ter
s
hed
Seg
m
e
n
tation Detection
of
abnorm
a
l m
a
sses
Not focused on
classification,
no
bench
m
ar
king
5 M
a
s
m
oudi
et
al.
[12]
Local Binar
y
Patt
e
r
n
Does Cl
assification no
bench
m
ar
king
6
Shareef
[
13]
Wate
rshed Tran
sf
or
m
a
tion
90.47%
accuracy
-Do-
7
Fleet et
al. [
14]
Haralick F
eature
88.9% accuracy
-Do-
8
ACO-Based
Opti
m
i
zation
Sivaku
m
a
r
[15]
Bilater
a
l subtr
action,
ACO,
GA
Bench
m
ar
ked study
with 91.
9%
accurac
y
Not focused on
classif
i
cation,
9
Machraoui et al.
[16]
ACO,
Otsu
m
e
thod o
f
seg
m
entation
Lesser processi
ng
ti
m
e
-Do
-
10
Bacha et al
. [
17]
ROI,
AC
O,
Marko
v
Rando
m
Field
Ef
f
ective
seg
m
entation
Not focused on
classification,
no
bench
m
ar
king
11
PSO-Based
Opti
m
i
zation
Al-
F
ar
is
[18]
ROI
,
PSO, clustering,
L
e
vel
Set Active Contour
Im
pr
oved
seg
m
entation
per
f
orm
a
nce
-Do
-
12 Mohe
mm
ed
et
al.
[19]
PSO, outlier detect
ion
Effective
seg
m
entation
Algor
ith
m
have tim
e
co
m
p
lexit
y
13
Dheeba et al
. [
20]
Wa
velet neur
al Networ
k,
CAD
93.67& accurac
y
,
bench
m
ar
ked study
Not focused on
classif
i
cation,
14
GA-based
Opti
m
i
zation
Zadeh et al
. [
21]
GA, Fuzz
y
Logic
93% accuracy
-Do-
15
Ah
m
a
d et
al. [
22]
GA-A
N
N
97-98% accuracy
-D0-
16
Yang et al. [23]
GA
Better
SNP
interaction
-Do
-
17
Fuzzy
L
ogic
based
optim
ization
Keles et
al. [
24]
Neuro-f
u
zzy classi
f
i
cation
93% accuracy
-Do-
18
Singh et al.
[25]
Fuzzy
C-
m
e
ans
cluster
i
ng,
k-m
eans cluster
i
ng
Good detection
Not focused on
classification,
no
bench
m
ar
king
19
Balanica et al
. [
26]
Fuzzy
L
ogic
Pr
ediction of disease
-
D
o-
3.
RESEA
R
C
H AN
D DIS
C
US
SION
Thi
s
sect
i
on
hi
ghl
i
g
ht
s ab
o
u
t
t
h
e researc
h
ga
p o
f
the breast cancer
detection
t
ech
ni
q
u
es u
s
i
ng
di
gi
t
a
l
i
m
ag
e p
r
o
c
essin
g
till d
a
te. The fo
llowing
research
g
a
p
s
are fou
n
d
after rev
i
ewing
so
m
e
stu
d
i
es don
e in
last
t
e
n y
ears an
d t
h
e s
u
m
m
e
ry
of
t
h
e past
t
e
n y
e
ars resea
r
c
h
w
o
r
k
s a
r
e p
r
ese
n
t
e
d i
n
t
h
e Ta
bl
e-I
whi
c
h i
s
i
n
cl
ude
d
in
th
e
pr
opo
sed
su
rv
ey st
u
d
y
.
Tabl
e
1 b
r
i
e
fl
y
sh
ows
t
h
e e
x
i
s
t
i
ng t
ech
ni
q
u
es
t
hose
are
use
d
for the early
detection of
bre
a
st cancers
fro
m
ma
mmo
g
r
am
s. It h
a
s b
een
seen
in
th
e
m
a
x
i
m
u
m ex
istin
g
stud
ies th
at th
e
maj
o
rity o
f
th
e
works
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Adv
ance
m
e
n
t
i
n
Rese
arc
h
Te
chni
que
s
on
M
e
di
cal
Im
a
g
i
n
g
Proce
ssi
n
g
f
o
r
Brea
st
C
a
ncer
…
(
Sus
hm
a S J)
72
3
d
i
scu
s
ses less ab
ou
t th
e practical ap
p
licatio
n
s
of th
e
propo
sed
tech
n
i
q
u
e
s rath
er th
an
th
e m
a
th
em
a
tic
al an
d
th
eoretical ex
plan
atio
n
s
. It has also
b
e
en
ob
serv
ed
th
at maj
o
rity o
f
th
e
work
wh
ich
are p
u
b
lish
e
d
til
l n
o
w
have
m
o
re em
pha
si
zed i
n
t
h
e
det
ect
i
o
n
o
f
m
i
cro cal
ci
fi
cat
i
ons,
A
r
chi
t
e
c
t
ural
di
st
ort
i
o
n
s
an
d m
a
sses,
but
t
h
e
n
a
ture
o
f
th
e ex
istin
g techn
i
qu
es are
rep
e
titiv
e in
n
a
tu
re.
As th
e ev
id
en
ce
says th
at th
e can
cer d
e
tection u
s
i
n
g
m
a
m
m
ogram
s
i
s
one
of t
h
e e
ffi
ci
ent
t
ech
ni
que
s b
u
t
t
h
e p
r
esence
of
bet
t
e
r o
r
ga
ni
zat
i
o
n o
f
st
u
d
i
e
s and t
h
e
per
f
o
r
m
a
nce p
a
ram
e
t
e
rs t
o
re
prese
n
t
t
h
e
res
u
l
t
anal
y
s
i
s
o
f
t
h
e p
r
op
ose
d
t
echni
que
s are
poi
nt
edl
y
m
i
ssi
ng i
n
m
o
st
of
t
h
e rec
e
nt
st
u
d
i
e
s. Th
ese
na
rr
o
w
ed
r
e
vi
ew w
o
r
k
s d
o
not
gi
ve
bet
t
er out
c
o
m
e
s
whi
c
h
ca
n be us
ed fo
r
fut
u
re pi
pel
i
n
e
of
t
h
e researc
h
wo
rk
as
t
h
e m
onot
on
o
u
s
t
y
pes of
t
e
c
hni
q
u
es do
n
o
t
p
r
o
v
i
d
e bet
t
e
r
s
o
l
u
t
i
o
n
s
t
o
com
pute a part
icular task.
Few
be
nchm
arked
st
u
d
i
e
s
ha
ve
been
f
o
un
d
i
n
som
e
of
t
h
e
pa
pers
as
ben
c
hm
arki
ng
m
a
kes a
rea
d
e
r
to understa
n
d t
h
at
under whic
h circ
um
stances their expe
ri
m
e
nt
al
res
u
l
t
s
or
o
u
t
c
om
es are
bet
t
e
r a
n
d
usef
ul
f
o
r
t
h
e f
u
t
u
re i
m
pl
em
ent
a
t
i
on and
a
d
o
p
t
i
o
n
of
t
h
e
w
o
r
k
.
It
has
been
f
o
u
n
d
t
h
at
fe
w
pa
pers
descri
bed
t
h
e
rep
e
titiv
e
n
a
ture
o
f
im
p
l
e
m
e
n
tatio
n of th
e
p
r
op
o
s
ed
system
an
d
th
e
resu
lt an
alysis
d
o
es no
t
g
i
v
e
a
b
e
tter
so
lu
tion
t
o
read
er to
un
d
e
rstan
d
th
e con
cep
t
.
4.
CO
NCL
USI
O
N
The p
r
ese
n
t
e
d
pape
r bri
e
fs abo
u
t the advancem
ent of the m
e
dical
image proces
sing towards the
detection
of breast cancer. T
h
e pa
per
di
scuss
e
s that at prese
n
t im
aging tech
n
o
l
o
gi
es e.
g.
ul
t
r
asoun
d, CT scan
,
PET scan, MRI etc are already in
u
s
e. Howev
e
r, m
a
mm
o
g
r
a
m
s are still
asso
ciated
with
variou
s prob
lems th
at
eith
er
resu
lts i
n
in
effectiv
e
d
i
ag
no
stics
o
r
resu
lts in
t
o
o
m
a
n
y
false po
sitiv
es, we h
a
v
e
also
seen
t
h
at there are
en
oug
h stud
ies con
d
u
c
ted to
w
a
rd
s detectio
n
of
can
cer
i
n
breast. Majority o
f
th
e ex
i
s
tin
g
techn
i
ques are
foc
u
se
d o
n
e
n
hanci
ng t
h
e c
o
n
v
e
n
t
i
onal
t
e
chni
que
s i
t
s
el
f t
h
at
refl
ect
s l
e
ss n
ovel
t
y
i
n
t
h
i
s
area.
He
nce t
h
e
out
c
o
m
e
s of
suc
h
i
m
pl
em
ent
a
t
i
ons
d
o
est
y
i
el
d m
u
ch
brea
k t
h
r
o
ug
h
fi
n
d
i
n
gs.
I
n
case o
f
o
p
t
i
m
i
zat
i
on
techniques
, there are
existi
ng va
ri
o
u
s
o
p
t
im
i
zat
i
on t
echni
que
s e.
g.
ACO,
PS
O,
GA
, F
u
zzy logic etc.
Ho
we
ver
,
al
l
t
h
ese t
e
c
hni
que
s are
n
o
t
f
o
un
d t
o
f
o
c
u
s
o
n
classification
of the
cance
r
stage. In fact
the
m
a
jor
fi
n
d
i
n
gs of
ou
r
i
nvest
i
g
at
i
o
n i
s
t
h
at
i
)
m
a
jori
t
y
of t
h
e st
udi
es are not
be
nch
m
arked
,
i
i
)
neg
l
i
g
ence o
n
foc
u
si
n
g
classificatio
n
t
ech
n
i
q
u
e
s
o
f
can
cer stag
es or criticality
, and
iii) algorithm
efficien
cy ev
alu
a
tion
s
are
al
m
o
st
n
o
n
e
to
b
e
foun
d
i
n
ex
isting
syste
m
. Hen
c
e, ou
r fu
ture
work
will b
e
to
ad
dress su
ch
issu
es in
real sense.
W
e
find that enha
ncem
ent of the
m
a
mm
ogram is one of
t
h
e significant
factor t
h
at wa
s not foc
u
se
d by any
research
ers till
d
a
te. A m
o
d
e
l fo
r en
h
a
n
c
i
n
g th
e co
n
t
rast
of m
a
mm
o
g
r
am fo
r sup
e
rior reso
lu
tion
will lead
to
bet
t
e
r preci
si
o
n
of det
ect
i
o
n
of
ca
nce
r
o
u
s p
o
rt
i
o
n. Exi
s
t
i
n
g opt
i
m
i
z
at
i
on
t
echni
que
s
co
ul
d be re-e
n
h
a
n
ced
t
o
su
it th
e
requ
iremen
t o
f
preci
s
e
breast cancer detection.
REFERE
NC
ES
[1]
H.
U.
Lemke,
K.
Inamura,
C.
C.
Jaffe,
R
.
Felix
, Co
mputer Assisted Radiolog
y
,
Springer Scien
ce
&
Business Media,
29-Jun-2013.
[2]
Srivastava, Rajeev, R
e
sear
ch D
e
velopments
in
Computer Visi
on and Image
Processi
ng: Methodologies and
Applications,
IGI Global, 30-Sep
-
2013
[3]
U. Bick
, F. Diek
mann, Digita
l Mammograph
y
, Springer Scien
ce
& Business Med
i
a, 11-Mar-2010.
[4]
G.R. Sinha, B
.
C. Patel, Medical I
m
age Pro
cessing
, PHI Learning
P
v
t.
Ltd
.
, 20-Jan-
2014.
[5]
D.M. Klieger
,
S
a
unders Essentials
of Medical Assisting,
Elsevi
er
Health
Scien
ces, 07-Aug-2013.
[6]
T.M. Deserno, B
i
omedical Image Processing,
Springer Science &
Busi
ness Media, 01-Mar-2011.
[7]
C. Aktolun
,
W.N. Taux
e, Nuclear Oncolog
y
,
Sp
ringer Scien
ce
& Busi
ness Media, 06-Dec-2012.
[8]
Y. Zheng
,
"Br
e
a
s
t can
cer
det
e
c
t
i
on with Gabor
f
eatur
es
from dig
ital mammograms",
algorithms
Vol.3, No
. 1
,
pp
.
44-62, 2010
.
[9]
J.
Nagi,
S.
A.
Kareem,
F.
Nagi,
S.
K.
Ahm
e
d, “Autom
ated Brea
s
t
P
r
ofile
Segmentation for ROI Detection Using
Digital Mammo
grams”, IEEE
EMBS Conference on Biomedical Engine
ering
&
Sciences (IECB
ES 2010), Kuala
Lumpur, Malay
s
ia, 30th
November, 2010
.
[10]
K.
Ganesan,
U. Ac
har
y
a,
C
.
K. Chua, L.C
.
Min
,
K.
T.
Abraham,
and K.
H.
Ng,
“
C
om
puter-Aided Breas
t C
a
nc
er
Detection Using
Mammograms: A Re
view”, I
E
EE REVIEWS IN
BIOMED
ICAL ENGINEERING, VOL. 6, 2013
[11]
A.A. Hefnaw
y
,
“An Improved Approach
for
Breast C
a
ncer
Detection
in M
a
mmogram based on Watershed
Segmentation
”
, I
n
ternational Jour
na
l of
Computer
Applications (09
75 Volume 75–
No.15, August 2
013.
[12]
A.D. Masmoudi, N. G. B. A
y
ed
,
D.
S. Masmoudi, and R
.
Abid, "
L
BPV desc
riptor
s-based automatic ACR/BIRADS
classification ap
proach",
EURASI
P Journal on I
m
age and
Video
Processing, No.
1, pp
. 1-9
,
2013
.
[13]
S
.
R. S
h
aree
f, “
B
reas
t Can
cer
Detec
tion Bas
e
d on W
a
te
rshed
Transform
a
tion
”
, IJCSI Interna
tional Journ
a
l o
f
Computer Scien
ce Issues, Vol. 1
1
, Issue 1
,
No 1
,
Januar
y
2014
.
[14]
B.D.Fleet, J. Y
a
n, D.B
.
Knoes
t
er, M. Yao
,
J.R. De
ller and E.D.
Goodman,
"Br
east C
a
ncer
Detection
Using
Harali
ck F
eatur
es
of Im
ages
Recons
truc
ted fro
m
Ultr
a W
i
deband M
i
crowave
S
cans
"
, In
Clin
ica
l
Im
age-Bas
e
d
Procedures. Translational R
e
search in
Medical I
m
agi
ng, Springer Intern
ation
a
l Publishing, pp. 9-
16, 2014
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
6, No
. 2, A
p
ri
l
20
16
:
71
7 – 7
2
4
72
4
[15]
R. Sivakum
ar,
M. Karnan, “Int
ellig
ent op
tim
ization
tec
hniques
for m
a
m
m
ogram
Im
age anal
y
s
i
s
through bil
a
ter
a
l
subtraction”, IEEE In
tern
ation
a
l Confer
en
ce
of
Com
putational
I
n
tell
igen
ce
and
Computing Research, pp.1-4
,
201
0.
[16]
A.N. Machraoui, M.A. Cherni, and
M. Sayad
i
, "Ant Colon
y
optim
ization a
l
gorithm
for breast can
cer c
e
ll
s
clas
s
i
fi
cat
ion",
In Ele
c
tri
c
a
l
En
gin
eer
ing and S
o
ftware Applications (ICEESA)
,
International Co
nference, pp
. 1-
6,
2013.
[17]
A. Bach
a, K.
Ka
lti,
N.E
.
B. Am
ar
a, and
B. Sol
a
im
an,
"Micro
cal
cif
i
ca
tions det
e
c
tio
n in m
a
m
m
ogram
s based on Ant
Colon
y
Optimization
and Mark
ov Random
Field", In
Soft Co
mputing a
nd Pattern R
ecognitio
n (SoCPaR), 6th
International Co
nference of
, pp
.
191-196, 2014
.
[18]
A.
Q.
A.
Faris,
U.
K.
Ngah,
N.A.
M.
Isa,
and
I.
L.
S
hua
ib, "B
reast MRI tum
our se
gmentatio
n using modified
automatic seed
ed region growin
g based on par
t
icle swarm
optimization
image
cluster
i
ng", In
Soft Computing in
Industrial Applications, Springer
Inte
rnational Pu
blishing, pp. 49-
60, 2014
.
[19]
A.W. Mohemmed, M. Zhang
,
and W.
N. Browne. "Particl
e
swarm optim
is
ation for outli
er dete
ction
.
" In
Proceedings of
t
h
e 12th
annu
al
c
onferenc
e
on
Ge
neti
c and
evo
l
uti
onar
y
com
put
ati
on, pp
. 83-84
. A
C
M, 2010.
[20]
J. Dheeba, N.A. Singh, S.
T. Selvi, “Computer-aided d
e
tection
of
breast cancer on mammog
r
ams: A swar
m
intel
ligen
ce
opti
m
i
zed wave
let
n
e
ural
ne
twork
approach”, Journ
a
l of
Biomedi
cal Informatics, V
o
l. 49
, pp
. 45-5
2
,
2014.
[21]
H.G. Zadeh
,
O. Pakdelazar,
J.
Haddadnia, G.R
.
Rad,
“Diagnosing Breast Cancer with
the Aid
of Fuzzy
Logic
Based on Data Mining of a Genetic Algor
ithm in Infrared Images”, Middle East
Journal of Cancer vol.3
, Iss.4, pp.
119-129, 2011
.
[22]
A.M. Ahm
a
d, G. Muham
m
a
d,
J.F. Miller, “
B
reast Canc
er
Detec
tion Using Cartesian Gen
e
ti
c Program
m
i
ng
evolved Artif
ici
a
l Neural Netwo
r
ks”, ACM-Proceedings
of 14th Annula Confere
n
ce on genet
i
c a
nd Evolutiona
r
y
Computation, pp
. 1031-1039
, 20
12.
[23]
CH Yang, YD
Lin, LY Chuan
g
, and HW
Cha
ng, “Evaluation
of Breast Ca
ncer Susceptibility Using I
m
proved
Genetic Algorithms to
Gene
rate Genoty
p
e SNP Barcodes”
, IEEE/ACM Trans
actions on Computation
a
l Biolo
g
y
and Bio
i
nformatics, vo
l. 10, no.
2, 2013
.
[24]
A. Keles, A. Keles,
“Extr
actin
g fuzzy
ru
les f
o
r the diagnosis
of
breast cancer”,
Turkish J
ournal of Electr
ical
Engineering
&
Computer Scien
ces, vo
l.21
, pp
. 1
495-1503, 2013
.
[25]
N. Singh, A.G Mohapatra,
G.
Kanungo, “Breast Cancer Mass Detect
ion in
Mammograms u
s
ing K-means and
F
u
zz
y
C-m
eans
Clus
tering”
,
International Journal of Computer
Applica
tions
(0975 – 8887)
Volume 22– No. 2,
May
2011.
[26]
V. Bal
a
nic
a
,
I.
Dum
itrache,
M
.
Caram
i
ha
i, W
.
Rae,
C.
Herbst,
“Evaluation of
breast
cancer r
i
sk b
y
using Fu
zzy
logic”, U.P.B. Sci. Bull., Seri
es
C, Vol. 73, Iss. 1
,
2011
.
BIOGRAP
HI
ES OF
AUTH
ORS
Sushma SJ is
working as Associate Professo
r, Department of ECE, GSSS Institute of
Engine
ering and
Technolog
y
for
wom
e
n, My
sur
u
.
She has got 14
y
e
ars of teach
i
ng experien
ce
.
She has obtain
e
d Bachelor of Engineer
ing from
Manglore Univ
ersity
in th
e
y
ear 2001. In 2007
dshe obtained
Master of Tech
nolog
y
from Visv
eswaray
a
Technological Univ
ersity
, Belag
a
vi.
Currently
pursuing Ph.D. at Vis
v
eswaray
a
Tech
nological Univ
ersity
, Belag
a
vi,
India. She has
published 2 papers nation
a
l conferences Her
ar
ea of
interests include Image Processing,
Com
putational
I
n
tell
igen
ce
and
Computer Netw
orks.
Dr Prasanna Kumar S C is wo
rking as Profe
ssor and Head
, Department of In
strumentation
Techno
log
y
, RV colleg
e
of Engineering
,
Bangalore
. He has got 18
y
e
ars of teaching, 01
y
ear of
industr
y
and 10
y
e
ars of resear
ch experience. He di
d his Bachelo
r
of Engineering
and master of
Engineering fro
m My
sore Un
iversity
. He
was awarded Ph
D in the
y
e
ar
2009 from
Avinashlingham University
, Tamilnadu. He ha
s published ov
er 14 p
a
pers in
nation
a
l
and
intern
ation
a
l co
nferences and around 24 papers
in
the
intern
ational journ
a
ls. H
e
has receiv
e
d
acad
em
ic exc
e
ll
ence
award
for t
h
e y
e
ars
2008 an
d
2009.
Evaluation Warning : The document was created with Spire.PDF for Python.