Co
m
pu
t
er
Science
a
nd
I
nfo
r
m
a
t
io
n T
ec
hn
o
lo
g
ies
Vo
l.
6
,
No
.
2
,
J
u
ly
2
0
2
5
,
p
p
.
1
6
9
~
1
7
7
I
SS
N:
2722
-
3
2
2
1
,
DOI
:
1
0
.
1
1
5
9
1
/csi
t
.
v6
i
2
.
pp1
6
9
-
1
7
7
169
J
o
ur
na
l
ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
p
r
ime.
co
m/in
d
ex
.
p
h
p
/csi
t
H
epa
toSca
n:
Ens
emble cla
ss
ificati
o
n learning
mo
de
ls for liv
er
ca
ncer
disea
se
d
et
ection
T
ella
S
um
a
llik
a
1
,
Ra
a
v
i
Sa
t
y
a
P
ra
s
a
d
2
1
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
A
c
h
a
r
y
a
N
a
g
a
r
j
u
n
a
U
n
i
v
e
r
s
i
t
y
,
G
u
n
t
u
r
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
D
h
a
n
e
k
u
l
a
I
n
st
i
t
u
t
e
o
f
En
g
i
n
e
e
r
i
n
g
a
n
d
Te
c
h
n
o
l
o
g
y
,
V
i
j
a
y
a
w
a
d
a
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Oct
2
6
,
2
0
2
4
R
ev
is
ed
J
an
2
3
,
2
0
2
5
Acc
ep
ted
Feb
2
0
,
2
0
2
5
Li
v
e
r
c
a
n
c
e
r
is
a
d
a
n
g
e
ro
u
s
d
is
e
a
se
th
a
t
p
o
se
s
sig
n
ifi
c
a
n
t
risk
s
to
h
u
m
a
n
h
e
a
lt
h
.
Th
e
c
o
m
p
lex
it
y
o
f
e
a
rly
d
e
tec
ti
o
n
o
f
li
v
e
r
c
a
n
c
e
r
in
c
re
a
se
s
d
u
e
to
t
h
e
u
n
p
re
d
icta
b
le
g
r
o
wth
o
f
c
a
n
c
e
r
c
e
ll
s.
Th
is
p
a
p
e
r
in
tr
o
d
u
c
e
s
He
p
a
to
S
c
a
n
,
a
n
e
n
se
m
b
le
c
las
sifica
ti
o
n
to
d
e
tec
t
a
n
d
d
iag
n
o
se
li
v
e
r
c
a
n
c
e
r
tu
m
o
rs fro
m
li
v
e
r
c
a
n
c
e
r
d
a
tas
e
ts.
Th
e
p
ro
p
o
se
d
H
e
p
a
to
S
c
a
n
is
th
e
i
n
teg
ra
ted
a
p
p
r
o
a
c
h
th
a
t
c
las
sifies
th
e
th
re
e
ty
p
e
s
o
f
li
v
e
r
c
a
n
c
e
rs:
h
e
p
a
to
c
e
ll
u
lar
c
a
rc
in
o
m
a
,
c
h
o
lan
g
io
c
a
rc
in
o
m
a
,
a
n
d
a
n
g
i
o
s
a
rc
o
m
a
.
In
t
h
e
in
i
ti
a
l
sta
g
e
,
li
v
e
r
c
a
n
c
e
r
sta
rts
in
th
e
li
v
e
r,
wh
il
e
t
h
e
se
c
o
n
d
sta
g
e
sp
re
a
d
s
fr
o
m
t
h
e
l
iv
e
r
to
o
th
e
r
p
a
rts
o
f
th
e
b
o
d
y
.
De
e
p
lea
rn
i
n
g
is
a
n
e
m
e
rg
in
g
d
o
m
a
in
t
h
a
t
d
e
v
e
l
o
p
s
a
d
v
a
n
c
e
d
lea
rn
in
g
m
o
d
e
ls
to
d
e
tec
t
a
n
d
d
ia
g
n
o
se
li
v
e
r
c
a
n
c
e
rs
i
n
t
h
e
e
a
rly
st
a
g
e
s.
We
train
t
h
e
p
re
-
train
e
d
m
o
d
e
l
In
c
e
p
ti
o
n
V
3
o
n
l
iv
e
r
c
a
n
c
e
r
d
a
tas
e
ts
t
o
id
e
n
ti
f
y
a
d
v
a
n
c
e
d
p
a
tt
e
rn
s
a
ss
o
c
iate
d
w
it
h
c
a
n
c
e
r
tu
m
o
rs
o
r
c
e
ll
s.
F
o
r
a
c
c
u
ra
te
se
g
m
e
n
tatio
n
a
n
d
c
las
sifica
ti
o
n
o
f
li
v
e
r
les
io
n
s
in
c
o
m
p
u
ted
t
o
m
o
g
ra
p
h
y
(
CT
)
sc
a
n
s,
th
e
e
n
se
m
b
le m
u
lt
i
-
c
l
a
ss
c
las
sifica
ti
o
n
(EM
CC)
c
o
m
b
i
n
e
s U
-
Ne
t
a
n
d
m
a
sk
re
g
io
n
-
b
a
se
d
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
tw
o
rk
(
R
-
CNN
)
.
In
t
h
i
s
c
o
n
tex
t,
re
se
a
rc
h
e
rs
u
se
th
e
CT
sc
a
n
ima
g
e
s
fro
m
Ka
g
g
le
to
a
n
a
l
y
z
e
th
e
li
v
e
r
c
a
n
c
e
r
tu
m
o
rs
fo
r
e
x
p
e
rime
n
tal
a
n
a
ly
sis.
F
in
a
ll
y
,
q
u
a
n
ti
tativ
e
re
su
lt
s
sh
o
w
th
a
t
th
e
p
ro
p
o
se
d
a
p
p
r
o
a
c
h
o
b
tai
n
e
d
a
n
i
m
p
ro
v
e
d
d
ise
a
se
d
e
tec
ti
o
n
ra
te
with
m
e
a
n
sq
u
a
re
d
e
rr
o
r
(
MSE
)
-
1
1
.
3
4
a
n
d
p
e
a
k
sig
n
a
l
-
to
-
n
o
ise
ra
ti
o
(
P
S
NR
)
-
1
0
.
3
4
,
wh
ich
is
h
ig
h
c
o
m
p
a
re
d
with
e
x
is
ti
n
g
m
o
d
e
ls
su
c
h
a
s
fu
z
z
y
C
-
m
e
a
n
s
(
F
C
M
)
a
n
d
k
e
rn
e
l
fu
z
z
y
C
-
m
e
a
n
s
(
KFCM
)
.
Th
e
c
las
sifica
ti
o
n
re
su
lt
s
o
b
tain
e
d
b
a
se
d
o
n
d
e
tec
ti
o
n
ra
te
w
it
h
a
c
c
u
ra
c
y
-
0
.
9
7
%
,
s
p
e
c
ifi
c
it
y
-
0
.
9
9
%
,
re
c
a
ll
-
0
.
9
9
%
,
a
n
d
F
1
S
-
0
.
9
7
%
a
re
v
e
ry
h
i
g
h
c
o
m
p
a
re
d
wi
th
o
t
h
e
r
e
x
isti
n
g
m
o
d
e
ls.
K
ey
w
o
r
d
s
:
An
g
io
s
ar
co
m
a
C
h
o
lan
g
io
ca
r
cin
o
m
a
Dee
p
lear
n
in
g
E
n
s
em
b
le
class
if
icatio
n
Hep
ato
ce
llu
lar
ca
r
cin
o
m
a
Hep
ato
Scan
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
T
ella
Su
m
allik
a
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
in
ee
r
in
g
,
Ac
h
ar
y
a
Nag
ar
ju
n
a
Un
i
v
er
s
ity
Nag
ar
ju
n
a
Nag
a
r
-
5
2
2
5
1
0
,
An
d
h
r
a
Pra
d
esh
,
I
n
d
ia
E
m
ail: su
m
allik
a.
p
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
liv
er
is
a
ty
p
ical
o
r
g
an
in
th
e
h
u
m
an
b
o
d
y
th
at
r
eg
u
lates
ch
em
ical
lev
els
in
th
e
b
lo
o
d
an
d
ex
cr
etes
a
s
u
b
s
tan
ce
ca
lled
b
i
le.
T
h
e
p
r
im
ar
y
p
u
r
p
o
s
e
o
f
b
ile
is
to
tr
an
s
p
o
r
t
waste
m
ate
r
ial
f
r
o
m
th
e
liv
er
.
L
iv
er
ca
n
ce
r
is
a
s
ig
n
if
ica
n
t
c
o
n
tr
ib
u
to
r
to
g
lo
b
al
m
ali
g
n
a
n
c
y
m
o
r
tality
,
m
ak
in
g
u
p
a
s
ig
n
i
f
ican
t
p
o
r
tio
n
o
f
all
m
alig
n
an
t
d
ea
th
ca
teg
o
r
ies
[
1]
.
B
ec
au
s
e
liv
er
ca
n
ce
r
,
p
ar
ticu
lar
ly
h
ep
ato
ce
llu
lar
ca
r
cin
o
m
a
,
h
as
d
em
o
n
s
tr
ated
p
r
o
m
is
in
g
tr
ea
tm
en
t
e
f
f
icac
y
an
d
im
p
r
o
v
ed
s
u
r
v
iv
al
wh
en
d
iag
n
o
s
ed
at
a
n
ea
r
lier
s
tag
e,
el
u
cid
atin
g
m
o
lecu
lar
b
io
m
ar
k
er
s
o
r
m
ec
h
an
is
m
s
f
o
r
ea
r
ly
d
etec
tio
n
is
cr
u
cial
[
2
]
,
[
3
]
.
W
e
also
v
alid
ate
th
e
m
ea
s
u
r
em
en
ts
ag
ain
s
t
o
th
er
cl
in
ical
s
tan
d
ar
d
s
,
s
u
ch
as
b
io
p
s
y
,
wh
ich
r
em
ain
s
th
e
g
o
ld
s
tan
d
ar
d
f
o
r
d
etec
tio
n
,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
7
2
2
-
3
2
2
1
C
o
m
p
u
t Sci
I
n
f
T
ec
h
n
o
l
,
Vo
l.
6
,
No
.
2
,
J
u
ly
2
0
2
5
:
1
6
9
-
1
7
7
170
an
d
tr
ad
itio
n
al
im
a
g
in
g
tech
n
iq
u
es
lik
e
u
ltra
s
o
u
n
d
.
Ho
w
ev
er
,
b
o
t
h
o
f
th
ese
m
eth
o
d
s
h
av
e
lim
itatio
n
s
,
in
clu
d
in
g
l
o
w
ac
cu
r
ac
y
,
tim
e
-
co
n
s
u
m
in
g
n
atu
r
e,
p
o
ten
tial e
x
p
en
s
e,
an
d
th
e
p
o
s
s
ib
ilit
y
o
f
h
u
m
an
e
r
r
o
r
[
4
]
.
Ad
v
an
ce
s
in
m
ed
ical
im
ag
e
p
r
o
ce
s
s
in
g
an
d
d
ee
p
lear
n
in
g
al
g
o
r
ith
m
s
r
ep
r
esen
t
a
p
r
o
m
is
in
g
p
r
o
s
p
ec
t
f
o
r
o
v
er
h
a
u
lin
g
ca
n
ce
r
d
etec
t
io
n
as
we
cu
r
r
e
n
tly
k
n
o
w
it
[
5
]
.
T
h
ese
m
eth
o
d
s
co
u
ld
h
el
p
r
ad
io
lo
g
is
ts
an
d
o
n
co
lo
g
is
ts
to
b
etter
d
etec
t
ca
n
ce
r
o
u
s
tis
s
u
es
b
y
au
to
m
atin
g
th
e
a
n
aly
s
is
o
f
m
ed
ical
im
ag
es.
I
m
ag
e
p
r
o
ce
s
s
in
g
is
cr
u
cial
in
u
p
g
r
a
d
in
g
th
e
q
u
ali
ty
an
d
b
ea
u
tify
i
n
g
m
ed
ical
im
a
g
es
f
o
r
d
iag
n
o
s
is
[
6
]
.
Me
an
wh
ile,
d
ee
p
lear
n
in
g
alg
o
r
ith
m
s
,
p
ar
t
icu
lar
ly
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
s
(
C
NNs),
p
er
f
o
r
m
b
e
s
t
f
o
r
d
etec
tin
g
an
d
class
if
y
in
g
liv
er
ca
n
ce
r
p
ath
o
lo
g
y
o
n
m
ed
ical
im
ag
es
[
7
]
.
T
h
is
in
clu
d
es
u
s
in
g
v
ar
io
u
s
id
en
tifie
d
im
ag
e
m
o
d
alities
lik
e
co
m
p
u
ted
to
m
o
g
r
ap
h
y
(
C
T
)
,
m
a
g
n
etic
r
eso
n
an
ce
im
a
g
in
g
(
MRI
)
,
an
d
u
lt
r
aso
u
n
d
s
ca
n
s
.
W
e
p
r
o
ce
s
s
th
ese
im
ag
es
u
s
in
g
s
e
v
er
al
tech
n
iq
u
es
to
ex
tr
ac
t
th
e
r
eq
u
ir
ed
f
ea
t
u
r
es,
an
d
th
en
e
m
p
lo
y
d
ee
p
lear
n
in
g
m
o
d
els
to
d
eter
m
in
e
wh
eth
e
r
ab
n
o
r
m
al
g
r
o
wth
is
p
r
esen
t
in
h
ea
lth
y
tis
s
u
es
[
8
]
.
B
y
id
en
tify
in
g
p
atter
n
s
in
lar
g
e
d
atasets
o
f
lab
eled
im
ag
es,
d
ee
p
lear
n
in
g
m
o
d
els
ca
n
aid
in
ea
r
ly
d
iag
n
o
s
is
,
tu
m
o
r
s
eg
m
en
tatio
n
,
an
d
ca
n
ce
r
p
r
o
g
r
ess
io
n
p
r
e
d
ictio
n
.
Ho
wev
er
,
d
esp
ite
th
ese
s
etb
a
ck
s
,
ap
p
ly
i
n
g
im
a
g
e
p
r
o
ce
s
s
in
g
an
d
d
ee
p
lear
n
in
g
alg
o
r
ith
m
s
in
liv
er
ca
n
ce
r
d
etec
tio
n
is
h
ig
h
ly
p
r
o
m
is
in
g
[
9
]
,
[
1
0
]
.
I
n
th
is
r
esear
ch
w
o
r
k
,
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
is
th
e
co
m
b
in
atio
n
o
f
U
-
Net
an
d
Ma
s
k
r
eg
i
o
n
-
b
ased
co
n
v
o
l
u
tio
n
al
n
etwo
r
k
(
R
-
C
NN
)
.
U
-
Net
co
n
ce
n
tr
ated
o
n
s
eg
m
en
tatio
n
,
wh
ile
M
as
k
R
-
C
NN
h
an
d
led
class
if
icatio
n
.
W
e
u
s
e
I
n
ce
p
ti
o
n
V3
t
o
tr
ain
im
ag
es
o
f
liv
er
ca
n
ce
r
a
n
d
n
o
r
m
al
C
T
s
ca
n
s
.
W
e
u
tili
ze
tr
an
s
f
er
lear
n
in
g
to
tr
a
n
s
f
er
t
h
e
f
r
o
ze
n
lay
er
s
t
o
th
e
p
r
o
p
o
s
ed
en
s
em
b
le
m
u
lti
-
class
class
if
icati
o
n
(
E
MCC
)
,
wh
ich
en
h
an
ce
s
lesi
o
n
d
etec
tio
n
.
Fin
ally
,
th
e
p
r
o
p
o
s
ed
E
MCC
s
h
o
ws
th
e
p
er
f
o
r
m
a
n
ce
in
ter
m
s
o
f
d
etec
tin
g
th
e
ca
n
ce
r
-
af
f
ec
te
d
r
eg
i
o
n
s
an
d
cl
ass
if
icatio
n
o
f
im
ag
es.
Fig
u
r
e
1
s
h
o
ws th
e
m
eth
o
d
s
u
s
ed
in
t
h
is
wo
r
k
.
Fig
u
r
e
1
.
T
h
e
ar
ch
itectu
r
e
o
f
E
MCC
2.
L
I
T
E
R
AT
U
RE
SU
RVE
Y
Z
h
en
e
t a
l
.
[
1
1
]
in
tr
o
d
u
ce
d
a
n
ew
d
e
ep
l
ea
r
n
in
g
s
y
s
te
m
(
D
L
S
)
t
h
at
d
i
ag
n
o
s
es
a
n
d
d
e
te
cts li
v
er
t
u
m
o
r
s
in
t
h
e
ea
r
l
y
s
ta
g
es
u
s
in
g
MR
I
s
ca
n
i
m
a
g
es.
T
h
e
p
r
o
p
o
s
e
d
a
p
p
r
o
ac
h
m
a
in
ly
f
o
cu
s
e
d
o
n
c
las
s
if
y
i
n
g
li
v
er
t
u
m
o
r
s
in
t
o
s
ev
e
n
t
y
p
es
.
T
h
e
r
es
u
lts
s
h
o
w
t
h
a
t
t
h
e
p
r
o
p
o
s
e
d
a
p
p
r
o
a
ch
o
b
t
ai
n
e
d
a
h
i
g
h
ac
c
u
r
ac
y
o
f
9
2
.
3
4
%
b
ase
d
o
n
th
e
d
ete
cti
o
n
r
at
e.
T
h
e
r
e
is
t
h
e
l
ac
k
o
f
ac
c
u
r
ac
y
b
ase
d
o
n
t
h
e
d
is
ea
s
e
d
ete
cti
o
n
.
Ali
r
r
[
1
2
]
p
r
o
p
o
s
e
d
a
n
ad
v
a
n
ce
d
le
ar
n
i
n
g
a
p
p
r
o
a
ch
t
h
at
s
e
g
m
en
ts
th
e
li
v
e
r
s
am
p
l
es
f
r
o
m
t
h
e
C
T
s
c
an
i
m
a
g
es.
I
t
is
t
h
e
c
o
m
b
in
ati
o
n
o
f
FC
N
a
n
d
r
e
g
i
o
n
-
b
ase
d
le
v
e
l
t
ec
h
n
iq
u
es.
T
h
e
p
r
o
p
o
s
e
d
a
p
p
r
o
a
ch
s
e
g
m
e
n
ts
t
h
e
li
v
er
o
r
g
an
f
r
o
m
t
h
e
i
n
p
u
t
C
T
s
ca
n
i
m
a
g
e
t
o
f
in
d
th
e
t
u
m
o
r
s
.
I
n
t
h
e
f
i
n
al
s
t
e
p
,
th
e
r
e
g
i
o
n
-
b
ase
d
l
ev
el
r
e
f
i
n
e
d
th
e
s
e
g
m
e
n
t
ati
o
n
p
r
ed
icti
o
n
t
o
id
e
n
t
if
y
t
h
e
ac
c
u
r
a
te
f
in
al
s
e
g
m
e
n
ta
ti
o
n
.
T
h
e
r
es
u
l
ts
d
em
o
n
s
tr
a
te
t
h
a
t th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
p
r
o
d
u
ce
s
a
cc
u
r
at
e
r
es
u
lts
.
T
h
e
r
e
is
la
ck
o
f
s
eg
m
e
n
ta
ti
o
n
r
es
u
l
ts
in
th
is
a
p
p
r
o
ac
h
.
S
u
m
a
lli
k
a
a
n
d
P
r
asa
d
[
1
3
]
p
r
esen
ted
a
Evaluation Warning : The document was created with Spire.PDF for Python.
C
o
m
p
u
t Sci
I
n
f
T
ec
h
n
o
l
I
SS
N:
2722
-
3
2
2
1
Hep
a
to
S
ca
n
:
E
n
s
emb
le
cla
s
s
ifica
tio
n
lea
r
n
in
g
mo
d
els fo
r
liver c
a
n
ce
r
d
is
ea
s
es d
etec
tio
n
(
Tella
S
u
ma
llika
)
171
co
m
b
in
ed
m
o
d
el
f
o
r
d
etec
tin
g
liv
er
ca
n
ce
r
an
d
d
is
ea
s
es in
th
eir
ea
r
ly
s
tag
es.
I
n
th
is
ap
p
licatio
n
,
th
e
R
esNet
5
0
is
a
p
r
e
-
tr
ain
ed
m
o
d
el
th
at
ex
tr
ac
t
s
f
ea
tu
r
es
an
d
f
ilter
s
o
u
t
n
o
is
e
f
r
o
m
in
p
u
t
im
ag
es.
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
co
m
b
in
es
ex
tr
em
e
g
r
ad
ien
t
b
o
o
s
tin
g
(
E
GB
)
an
d
r
ec
u
r
r
en
t
n
eu
r
al
n
etwo
r
k
(
R
NN)
to
d
etec
t
u
n
u
s
u
al
ca
n
ce
r
ce
lls
an
d
ac
h
iev
es a
h
ig
h
er
ac
cu
r
ac
y
o
f
9
8
.
4
8
%.
R
ela
et
a
l
.
[
1
4
]
in
tr
o
d
u
ce
d
a
n
ew
o
p
tim
izatio
n
alg
o
r
ith
m
th
a
t a
d
o
p
ts
th
e
U
-
Net
f
ea
tu
r
es th
a
t c
o
m
b
in
e
with
g
r
ey
wo
l
f
-
class
to
p
p
er
o
p
tim
izatio
n
(
GW
-
C
T
O)
.
T
h
e
f
e
atu
r
es
ar
e
ex
tr
ac
ted
u
s
in
g
t
h
e
t
r
ain
in
g
m
o
d
el
th
at
s
elec
ts
th
e
f
ea
tu
r
es
b
elo
n
g
in
g
to
h
y
p
er
-
p
ar
am
eter
tu
n
e
d
im
p
r
o
v
e
d
-
d
ee
p
n
eu
r
al
n
etwo
r
k
(
HI
-
DNN)
en
h
an
ce
d
b
y
th
e
s
am
e
GW
-
C
T
O
alg
o
r
ith
m
.
I
n
th
e
f
in
al
s
tep
,
th
e
GW
-
C
T
O
-
HI
-
DNN
im
p
r
o
v
ed
th
e
ac
cu
r
ac
y
u
p
t
o
4
.
3
%,
2
.
4
%,
5
.
2
%,
a
n
d
4
.
3
%
f
o
r
all
th
e
o
th
er
m
o
d
els.
T
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
h
as
h
ig
h
a
cc
u
r
ac
y
co
m
p
ar
e
d
with
o
th
er
m
o
d
els.
L
iu
et
a
l
.
[
1
5
]
in
tr
o
d
u
ce
d
AI
-
b
ased
m
o
d
els
th
at
d
etec
t
liv
er
tu
m
o
r
s
u
s
in
g
ad
v
an
ce
d
s
eg
m
en
tatio
n
co
m
b
in
e
d
with
a
K
-
m
ea
n
s
clu
s
ter
in
g
(
KM
C
)
alg
o
r
ith
m
.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
d
iag
n
o
s
es
th
e
liv
er
tu
m
o
r
s
a
n
d
class
if
ies
th
e
n
o
r
m
al
a
n
d
t
u
m
o
r
C
T
im
ag
es.
T
h
e
ex
p
er
im
en
ts
s
h
o
w
t
h
at
th
e
liv
er
t
u
m
o
r
o
b
tain
s
an
ac
c
u
r
ac
y
o
f
9
3
.
2
3
%.
Di
et
a
l
.
[
1
6
]
d
em
o
n
s
tr
ated
an
a
u
to
m
ated
ap
p
r
o
ac
h
f
o
r
s
eg
m
en
tin
g
liv
er
C
T
s
ca
n
im
ag
es a
n
d
ex
tr
ac
tin
g
liv
er
tu
m
o
u
r
s
.
I
n
th
is
ca
s
e,
th
e
s
e
g
m
en
tatio
n
d
if
f
icu
lties
ar
e
m
it
ig
ated
b
y
u
s
in
g
th
e
3
D
U
-
Net
to
d
iv
id
e
th
e
h
ig
h
-
r
eso
lu
tio
n
p
ix
els
u
s
in
g
lo
ca
l
in
f
o
r
m
atio
n
b
ased
s
im
p
le
lin
ea
r
iter
ativ
e
clu
s
ter
in
g
(
LI
-
SLI
C
)
-
b
ased
h
ier
ar
ch
ical
iter
ativ
e
s
eg
m
en
tatio
n
.
Fin
all
y
,
th
e
v
o
tin
g
m
o
d
el
is
u
tili
s
ed
to
ex
tr
ac
t
tu
m
o
u
r
r
eg
io
n
s
f
r
o
m
h
ig
h
-
r
eso
lu
tio
n
p
ix
els,
id
en
tify
ab
n
o
r
m
al
a
r
e
as,
an
d
class
if
y
p
i
x
el
-
b
ased
r
esu
lts
.
T
h
e
r
esu
lts
d
em
o
n
s
tr
ate
th
at
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
p
r
o
d
u
ce
s
ac
cu
r
ate
r
e
s
u
lts
.
L
i
et
a
l
.
[
1
7
]
p
r
o
p
o
s
ed
a
d
ee
p
atten
tio
n
-
b
ased
n
e
u
r
al
n
etw
o
r
k
with
h
ig
h
-
r
eso
lu
tio
n
a
n
d
m
u
lti
-
s
ca
le
ch
ar
ac
ter
is
tics
f
o
r
liv
e
r
an
d
t
u
m
o
u
r
s
eg
m
en
tatio
n
in
C
T
s
ca
n
p
ictu
r
es.
T
h
e
m
u
lti
-
s
ca
le
f
ea
tu
r
es
alter
t
h
e
f
u
s
io
n
,
allo
win
g
f
ield
s
to
m
o
d
if
y
th
e
liv
er
an
d
tu
m
o
u
r
to
v
ar
io
u
s
f
o
r
m
s
an
d
s
izes.
Fin
ally
,
th
e
p
r
o
p
o
s
ed
s
tr
ateg
y
im
p
r
o
v
es p
er
f
o
r
m
an
c
e.
Gu
n
asek
h
ar
et
a
l
.
[
1
8
]
p
r
o
p
o
s
ed
a
n
ew
d
ee
p
lear
n
in
g
-
b
as
ed
m
o
d
el
co
m
b
i
n
ed
with
an
o
p
tim
izatio
n
alg
o
r
ith
m
.
T
h
is
wo
r
k
co
m
b
in
es
s
ix
f
ilter
s
with
f
ea
tu
r
e
s
elec
tio
n
t
o
im
p
r
o
v
e
th
e
f
in
al
o
u
tco
m
es.
Am
o
n
g
th
ese
f
ilter
s
,
two
h
ig
h
-
d
im
e
n
s
io
n
al
f
e
atu
r
es
ar
e
ex
tr
ac
ted
u
s
in
g
t
h
e
m
o
d
if
ied
s
o
cial
s
k
i
-
d
r
iv
er
o
p
tim
izatio
n
(
MSSO)
alg
o
r
ith
m
.
T
h
ese
ex
tr
ac
te
d
f
ea
tu
r
es,
also
ca
lled
h
ig
h
-
r
an
k
ed
f
ea
tu
r
es,
f
in
d
th
e
ac
cu
r
ate
f
ea
tu
r
es
f
r
o
m
t
h
e
liv
er
ca
n
ce
r
tis
s
u
es
d
etec
ted
b
y
th
e
s
u
n
f
lo
w
er
o
p
tim
izatio
n
-
b
ased
d
ee
p
n
eu
r
al
n
etwo
r
k
(
DSFNN)
ap
p
r
o
ac
h
.
T
h
e
e
x
p
er
im
e
n
tal
an
aly
s
is
was
ap
p
lied
to
th
e
N
atio
n
al
C
en
ter
f
o
r
B
io
tech
n
o
lo
g
y
I
n
f
o
r
m
atio
n
-
Gen
e
E
x
p
r
ess
io
n
Om
n
ib
u
s
(
NC
B
I
-
GE
O
)
d
atab
ase,
an
d
s
u
p
e
r
io
r
p
er
f
o
r
m
a
n
ce
was
o
b
tain
ed
in
class
if
y
in
g
ty
p
es
o
f
liv
er
ca
n
ce
r
s
.
B
aâ
za
o
u
i
e
t
a
l
.
[
1
9
]
s
tated
a
s
em
i
-
au
to
m
ated
s
eg
m
en
tatio
n
m
eth
o
d
f
o
r
n
u
m
e
r
o
u
s
lesi
o
n
s
in
th
e
liv
er
u
s
in
g
C
T
s
ca
n
d
ata.
T
h
e
r
ec
o
m
m
e
n
d
ed
ap
p
r
o
ac
h
r
em
o
v
es
liv
er
lesi
o
n
s
f
r
o
m
th
e
in
p
u
t
im
ag
es
an
d
class
if
ies
th
em
as
n
o
r
m
al
o
r
ab
n
o
r
m
al
im
ag
es.
T
h
e
r
esu
lts
s
u
g
g
est
th
at
th
e
p
r
o
p
o
s
ed
s
tr
ateg
y
o
u
tp
e
r
f
o
r
m
s
o
th
er
m
o
d
els
in
ter
m
s
o
f
cla
s
s
if
icatio
n
ac
cu
r
ac
y
.
Das
et
a
l
.
[
2
0
]
pr
o
p
o
s
ed
th
e
m
o
d
el
k
er
n
el
f
u
zz
y
C
-
m
ea
n
s
(
KFC
M)
clu
s
ter
in
g
co
m
b
i
n
e
d
with
a
d
ap
tiv
e
an
d
m
o
r
p
h
o
lo
g
ical
p
r
o
ce
s
s
in
g
m
o
d
els
th
at
s
eg
m
e
n
t
th
e
liv
e
r
C
T
s
ca
n
m
o
r
p
h
o
l
o
g
ical
im
a
g
es.
KFC
M
is
also
u
s
ed
to
r
e
d
u
ce
th
e
n
o
is
e
a
n
d
in
cr
ea
s
e
th
e
s
tr
en
g
th
o
f
th
e
clu
s
ter
in
g
.
T
h
e
q
u
an
titativ
e
r
esu
lt
s
s
h
o
w
th
at
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
o
b
tain
s
b
etter
p
ea
k
s
ig
n
al
-
to
-
n
o
is
e
r
atio
(
PS
NR
)
an
d
lo
w
m
ea
n
s
q
u
ar
ed
e
r
r
o
r
(
MSE
)
with
c
o
n
s
is
ten
t a
cc
u
r
ac
y
.
3.
DATAS
E
T
D
E
SC
RIP
T
I
O
N
L
iv
er
tu
m
o
u
r
s
eg
m
en
tatio
n
c
h
allen
g
e
2
0
1
7
(
L
iTS1
7
)
d
ataset
is
th
e
m
o
s
t
o
f
ten
u
s
ed
[
2
1
]
.
I
t
co
n
tain
s
C
T
s
ca
n
im
ag
es
o
f
liv
e
r
tu
m
o
u
r
s
f
r
o
m
a
v
ar
iet
y
o
f
p
eo
p
le.
E
ac
h
s
am
p
le
h
as
3
D
s
ca
n
s
eg
m
en
tatio
n
m
a
r
k
in
g
s
r
ep
r
esen
tin
g
liv
er
a
r
ea
s
an
d
li
v
er
tu
m
o
u
r
s
.
I
t
c
o
m
p
r
is
es
5
0
0
C
T
s
ca
n
liv
er
p
h
o
to
s
,
in
cl
u
d
i
n
g
3
0
0
tr
ain
in
g
an
d
2
0
0
C
T
s
ca
n
im
ag
es.
Fig
u
r
e
2
s
h
o
ws th
e
C
T
s
ca
n
im
ag
es f
r
o
m
th
e
d
ataset.
Fig
u
r
e
2
.
Sam
p
le
l
iv
e
r
C
T
i
m
a
g
es f
r
o
m
d
ataset
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
7
2
2
-
3
2
2
1
C
o
m
p
u
t Sci
I
n
f
T
ec
h
n
o
l
,
Vo
l.
6
,
No
.
2
,
J
u
ly
2
0
2
5
:
1
6
9
-
1
7
7
172
4.
P
RE
-
T
RA
I
N
E
D
M
O
D
E
L
I
NCEPT
I
O
NV3
T
h
e
ar
c
h
itectu
r
e
o
f
I
n
ce
p
tio
n
V3
is
p
o
p
u
lated
with
s
ev
er
al
m
an
if
o
ld
m
o
d
u
les
f
r
o
m
th
e
th
ir
d
-
o
r
d
er
m
o
m
en
t
s
eq
u
en
ce
p
o
o
l,
wh
i
c
h
ar
e
co
n
ce
p
tu
ally
s
u
itab
le
f
o
r
lear
n
in
g
c
o
ar
s
e
an
d
m
i
d
-
lev
el
ab
s
tr
ac
tio
n
s
.
An
aly
s
is
o
f
liv
er
tu
m
o
r
s
o
f
te
n
in
v
o
lv
es
d
is
tin
g
u
is
h
in
g
b
etwe
en
lo
w
-
co
n
tr
ast
s
u
b
tle
tex
tu
r
es
an
d
p
atter
n
s
th
at
co
u
ld
c
h
ar
ac
ter
ize
ca
n
ce
r
o
u
s
ce
lls
.
Hen
ce
,
h
ig
h
-
r
eso
lu
tio
n
f
ea
tu
r
e
ex
t
r
ac
tio
n
is
ess
en
tial.
Fu
r
th
er
f
i
n
e
-
tu
n
in
g
I
n
ce
p
tio
n
V3
o
n
liv
er
tu
m
o
r
d
ata
f
r
o
m
o
u
r
in
s
titu
tio
n
en
a
b
les
th
e
m
o
d
el
to
twea
k
its
lear
n
ed
f
ea
tu
r
es
an
d
r
ec
o
g
n
ize
an
y
liv
er
-
s
p
ec
if
ic
a
n
o
m
alies
o
r
d
if
f
er
e
n
ce
s
in
th
eir
m
o
r
p
h
o
lo
g
y
,
h
en
ce
allo
wi
n
g
b
etter
d
etec
tio
n
pe
r
f
o
r
m
an
ce
[
2
2
]
.
T
h
e
p
r
e
-
tr
ain
ed
lay
er
s
in
I
n
ce
p
tio
n
V3
h
av
e
alr
ea
d
y
lear
n
ed
to
d
etec
t
th
e
b
est
co
m
p
lex
tex
tu
r
es
an
d
p
atter
n
s
in
im
ag
es,
tr
ain
in
g
(
f
in
e
-
tu
n
i
n
g
)
th
e
m
o
d
el
o
n
m
ed
ical
im
ag
in
g
d
atasets
with
s
l
ig
h
t
ad
ap
tatio
n
.
I
t
m
ain
l
y
allo
ws
th
e
d
ee
p
an
d
d
iv
er
s
e
co
n
v
o
lu
ti
o
n
al
s
tr
u
ctu
r
es
to
ca
p
tu
r
e
t
h
e
s
ig
n
if
ican
t
f
ea
tu
r
es
cr
u
cial
f
o
r
d
if
f
e
r
en
tiatin
g
liv
er
tu
m
o
r
s
an
d
n
o
n
-
t
u
m
o
r
tis
s
u
es
[
2
3
]
.
T
h
is
m
o
d
el
m
ain
ly
r
ed
u
ce
s
th
e
co
m
p
u
tatio
n
al
d
em
an
d
s
an
d
i
m
p
r
o
v
es p
atter
n
d
etec
tio
n
with
r
ap
id
o
u
tco
m
es.
T
h
e
s
tr
u
ctu
r
e
o
f
I
n
ce
p
tio
n
V
3
m
ain
ly
co
n
s
is
ts
o
f
r
ich
f
ea
tu
r
e
ex
tr
ac
tio
n
lay
er
s
.
T
h
e
m
o
d
el
lo
ad
ed
t
h
e
weig
h
ts
f
o
r
liv
e
r
ca
n
ce
r
d
etec
tio
n
b
u
t
ig
n
o
r
ed
t
h
e
to
p
lay
er
,
wh
ich
allo
ws
cu
s
to
m
izatio
n
to
class
if
y
n
o
r
m
al
an
d
a
b
n
o
r
m
al
s
am
p
les.
T
h
e
c
lass
if
icatio
n
im
p
r
o
v
e
d
b
y
a
d
d
in
g
th
e
d
e
n
s
e
an
d
d
r
o
p
o
u
t
la
y
er
s
to
th
e
d
ef
au
lt
I
n
ce
p
tio
n
V3
,
wh
ich
in
v
o
lv
es
t
h
e
f
u
lly
c
o
n
n
ec
te
d
lay
er
u
s
in
g
s
o
f
tm
ax
ac
tiv
atio
n
f
o
r
m
u
lti
-
class
clas
s
if
icatio
n
,
r
ep
r
esen
ted
in
(
1
)
.
I
n
c
e
p
t
i
o
n
V
3
m
o
d
e
l
=
I
n
c
e
p
t
i
o
n
V
3
(
we
i
gh
t
s
=
’
i
m
a
ge
n
e
t
’
,
i
n
c
lud
e
to
p
=
Fa
lse
,
i
n
p
ut
s
h
a
p
e
=
(
299
,
299
,
3
)
)
(
1
)
T
r
an
s
f
er
lear
n
in
g
is
m
ain
ly
u
s
ed
to
f
in
e
-
tu
n
e
th
e
to
p
lay
er
s
o
f
I
n
ce
p
tio
n
V3
,
wh
ich
in
itially
k
ee
p
s
th
e
lo
wer
lay
er
s
f
r
o
ze
n
to
r
etr
ie
v
e
th
e
p
r
e
-
tr
ain
ed
f
ea
tu
r
es.
T
h
e
to
tal
n
u
m
b
er
o
f
tr
ai
n
ab
l
e
lay
er
s
L
t
r
ai
n
ab
l
e
is
r
ep
r
esen
ted
in
(
2
)
:
θ
=
{
w
i
ϵ
L
t
r
ai
n
ab
l
e
|
F
r
e
e
ze
L
f
r
o
zen
}
(
2
)
L
o
s
s
f
u
n
ctio
n
(
L
)
:
th
e
b
i
n
ar
y
c
r
o
s
s
-
en
tr
o
p
y
u
s
ed
f
o
r
b
in
a
r
y
cl
ass
if
icatio
n
f
o
r
m
u
lticlas
s
is
r
ep
r
esen
ted
in
(
3
)
:
L
=
−
1
N
∑
y
i
∙
l
og
(
y
̂
i
)
N
i
=
1
(
3
)
O
p
t
i
m
i
z
e
r
:
t
h
e
A
d
a
m
o
p
t
i
m
i
z
e
r
u
s
e
d
t
o
i
m
p
r
o
v
e
t
h
e
l
e
a
r
n
i
n
g
r
a
t
e
o
f
f
o
r
10
−
4
f
i
n
e
t
u
n
e
d
l
a
y
e
r
s
r
e
p
r
e
s
e
n
t
e
d
i
n
(
4
)
:
θ
←
θ
−
α
∙
∇
θ
L
(
θ
)
(
4
)
W
h
er
e,
α
is
lear
n
in
g
r
ate
a
n
d
L
is
lo
s
s
f
u
n
ctio
n
.
5.
E
NS
E
M
B
L
E
M
UL
T
I
-
C
L
AS
S CLAS
SI
F
I
CA
T
I
O
N
C
o
m
b
in
in
g
U
-
Net
an
d
M
ask
R
-
C
NN
f
o
r
liv
er
ca
n
ce
r
d
etec
tio
n
is
a
s
u
cc
ess
f
u
l
m
er
g
er
b
e
ca
u
s
e
th
ey
h
av
e
s
tr
en
g
th
s
th
at
m
ak
e
th
e
m
p
ar
ticu
l
ar
l
y
well
-
s
u
ited
to
m
ed
ical
im
a
g
in
g
task
s
.
U
-
Net,
f
am
o
u
s
f
o
r
its
p
er
f
o
r
m
an
ce
in
s
em
an
tic
s
eg
m
en
tatio
n
p
r
o
b
lem
s
,
d
o
m
in
ate
s
p
ix
el
-
lev
el
class
if
ica
tio
n
b
y
tak
in
g
ad
v
an
ta
g
e
o
f
im
ag
e
r
eso
lu
tio
n
s
at
d
if
f
er
e
n
t
s
ca
les.
T
h
e
p
r
o
p
o
s
ed
n
et
wo
r
k
u
s
es
a
s
y
m
m
etr
ic
en
c
o
d
er
-
d
ec
o
d
er
U
-
N
et
ar
ch
itectu
r
e
with
s
k
ip
c
o
n
n
ec
tio
n
s
,
allo
win
g
f
o
r
g
r
ea
t
lo
ca
l
izatio
n
an
d
m
a
k
in
g
it
well
-
s
u
ited
to
s
eg
m
e
n
tin
g
tu
m
o
r
s
o
r
o
th
er
ab
n
o
r
m
alit
ies
in
h
ig
h
-
co
m
p
lex
ity
an
at
o
m
ical
s
tr
u
ctu
r
es
lik
e
th
e
liv
er
.
I
n
c
o
n
tr
ast,
M
ask
R
-
C
NN
ex
ten
d
s
Fa
s
t
er
R
-
C
NN
b
y
ad
d
in
g
in
s
ta
n
ce
s
eg
m
en
tatio
n
(
l
o
ca
lized
p
ix
el
-
wis
e
im
ag
e
ca
teg
o
r
izatio
n
)
to
th
e
to
p
o
f
o
b
ject
d
etec
tio
n
.
T
h
is
m
ak
es
it
p
o
s
s
ib
le
to
p
er
f
o
r
m
tu
m
o
r
lo
ca
lizatio
n
an
d
liv
er
tu
m
o
r
d
elin
ea
tio
n
in
a
s
in
g
le
s
lice
co
o
r
d
i
n
ate
m
ap
p
air
e
d
with
th
e
in
p
u
t
C
T
s
ca
n
.
I
t
w
ill
u
s
e
a
two
-
s
tag
e
d
etec
tio
n
alg
o
r
ith
m
,
wh
er
e
th
e
r
eg
i
o
n
s
ar
e
in
itially
p
r
o
p
o
s
ed
.
T
h
en
,
th
ese
r
eg
io
n
s
ar
e
r
ef
i
n
ed
wh
ile
s
im
u
ltan
eo
u
s
ly
cr
ea
tin
g
o
b
jec
t
m
ask
s
,
b
o
u
n
d
in
g
b
o
x
es,
an
d
class
if
y
in
g
lab
els.
T
h
is
m
u
lti
-
task
p
r
o
ce
s
s
in
g
is
b
en
ef
icial
in
s
ep
ar
atin
g
a
n
d
s
eg
m
en
tin
g
in
d
iv
i
d
u
al
tu
m
o
r
s
in
th
e
ca
s
e
o
f
m
u
ltip
le
liv
er
lesi
o
n
s
.
U
-
Net
i
s
ty
p
ically
u
s
ed
as
a
p
r
ep
ar
ato
r
y
an
d
in
itial
liv
er
ar
ea
s
eg
m
en
tatio
n
o
f
th
e
f
r
am
ewo
r
k
with
in
an
in
teg
r
ated
m
o
d
el
f
o
r
d
etec
tin
g
liv
er
ca
n
c
er
.
I
t
is
b
e
n
ef
icial
to
M
ask
R
-
C
NN
b
y
en
s
u
r
i
n
g
it
ca
n
tar
g
et
liv
er
tu
m
o
r
s
m
o
r
e
d
ir
ec
tly
an
d
ac
c
u
r
ately
in
s
tea
d
o
f
a
m
o
r
e
s
ig
n
if
ica
n
t
p
o
r
tio
n
.
Alter
n
ativ
ely
,
th
e
s
eg
m
en
ta
tio
n
m
ap
o
f
U
-
Net
ca
n
g
u
id
e
M
ask
R
-
C
NN’
s
r
eg
io
n
p
r
o
p
o
s
als
o
n
tu
m
o
r
b
o
u
n
d
ar
ies
an
d
co
n
s
eq
u
en
tly
in
cr
ea
s
e
lo
ca
lizatio
n
ac
cu
r
ac
y
a
r
o
u
n
d
i
n
s
tan
ce
m
ask
s
.
Fin
ally
,
th
e
co
m
b
in
ed
m
o
d
el
im
p
r
o
v
es
th
e
d
etec
tio
n
r
ate
in
ter
m
s
o
f
ac
cu
r
ac
y
,
q
u
ality
o
f
s
eg
m
en
ta
tio
n
,
an
d
u
n
d
er
s
tan
d
in
g
,
wh
ic
h
in
tr
o
d
u
ce
s
a
m
o
r
e
r
o
b
u
s
t
m
o
d
el
f
o
r
d
ia
g
n
o
s
in
g
liv
er
ca
n
ce
r
.
Evaluation Warning : The document was created with Spire.PDF for Python.
C
o
m
p
u
t Sci
I
n
f
T
ec
h
n
o
l
I
SS
N:
2722
-
3
2
2
1
Hep
a
to
S
ca
n
:
E
n
s
emb
le
cla
s
s
ifica
tio
n
lea
r
n
in
g
mo
d
els fo
r
liver c
a
n
ce
r
d
is
ea
s
es d
etec
tio
n
(
Tella
S
u
ma
llika
)
173
Step
1
: I
n
p
u
t im
ag
e
p
r
o
ce
s
s
in
g
an
d
p
r
ep
r
o
ce
s
s
in
g
I
m
ag
e
n
o
r
m
aliza
tio
n
:
th
e
in
p
u
t
im
ag
e
I
is
n
o
r
m
alize
d
b
y
d
e
d
u
ctin
g
th
e
in
te
n
s
ity
o
f
th
e
m
ea
n
an
d
d
i
v
id
in
g
b
y
th
e
s
tan
d
ar
d
d
ev
iatio
n
σ
g
iv
en
i
n
(
5
)
.
I
n
o
r
m
=
1
−
μ
σ
(
5
)
R
esizin
g
:
t
h
e
in
p
u
t im
ag
e
is
r
e
s
ized
to
a
f
ix
ed
s
ize
th
at
is
co
m
p
atib
le
with
th
e
n
etwo
r
k
,
s
u
ch
as H
X
W
.
Step
2
: I
n
itial seg
m
en
tatio
n
wi
th
U
-
Net
T
h
e
U
-
Net
f
u
n
ctio
n
ality
is
m
ain
ly
u
s
ed
to
s
eg
m
en
t
th
e
liv
e
r
r
eg
i
o
n
b
y
is
o
latin
g
th
e
r
eg
i
o
n
o
f
in
ter
est
(
R
OI
)
th
at
is
m
o
s
t lik
ely
to
co
n
tain
t
h
e
liv
er
.
a)
Do
wn
s
am
p
lin
g
p
ath
(
en
co
d
er
):
−
T
h
e
s
p
atial
d
im
en
s
io
n
s
ar
e
r
ed
u
ce
d
b
y
u
s
in
g
th
e
co
n
v
o
lu
tio
n
al
lay
er
s
to
ca
p
tu
r
e
th
e
f
ea
tu
r
e
r
ep
r
esen
tatio
n
s
.
−
T
h
e
f
ilter
f
is
ap
p
lied
at
ev
er
y
co
n
v
o
l
u
tio
n
lay
er
,
an
d
X
as in
p
u
t,
f
o
llo
we
d
b
y
R
eL
U
ac
tiv
atio
n
:
X
co
n
v
=
R
e
L
U
(
f
∗
X
+
b
)
(
6
)
−
T
o
d
o
wn
s
am
p
le
t
h
e
f
ea
tu
r
e
m
ap
s
,
m
ax
-
p
o
o
in
g
is
ap
p
lied
af
t
er
co
n
v
o
lu
tio
n
:
X
p
o
o
l
=
M
a
xPo
ol
(
X
co
n
v
)
(
7
)
b)
B
o
ttlen
ec
k
lay
er
:
t
h
is
lay
er
ca
p
tu
r
es
th
e
co
r
e
f
ea
tu
r
es
b
y
co
m
b
in
in
g
th
e
h
ig
h
-
lev
el
s
e
m
an
tics
f
r
o
m
th
e
en
co
d
er
with
ac
c
u
r
ate
u
p
s
am
p
lin
g
d
etails.
c)
Up
s
am
p
lin
g
p
ath
(
d
ec
o
d
er
):
−
T
r
an
s
m
it th
e
co
n
v
o
lu
tio
n
s
u
p
s
am
p
le
th
e
f
ea
tu
r
e
m
ap
s
:
X
up
s
a
m
p
l
e
=
Tr
a
n
s
pose
C
on
v
(
X
p
o
o
l
)
(
8
)
−
T
h
e
s
k
ip
co
n
n
ec
tio
n
s
ar
e
i
n
teg
r
ated
with
th
e
u
p
s
am
p
led
f
ea
t
u
r
es to
r
etr
iev
e
t
h
e
s
p
atial
d
ata
.
d)
Ou
tp
u
t
s
eg
m
en
tatio
n
m
ask
(
li
v
er
R
OI
)
:
−
T
h
e
U
-
Net
o
u
tp
u
ts
a
b
in
ar
y
m
ask
,
M
Li
v
er
wh
er
e:
M
l
i
v
er
(
x
,
y
)
=
{
1
if
(
x
,
y
)
∈
l
ive
r
r
e
g
ion
0
othe
r
w
ise
(
9
)
Step
3
: Filter
ed
in
s
tan
ce
s
eg
m
en
tatio
n
with
M
ask
R
-
C
NN
I
n
th
is
s
tep
,
th
e
U
-
Net
p
er
f
o
r
m
ed
th
e
ty
p
ical
s
eg
m
en
tatio
n
t
o
f
in
d
a
n
y
ca
n
ce
r
o
u
s
r
eg
io
n
s
(
ce
lls
)
in
th
e
liv
er
.
a)
R
eg
io
n
p
r
o
p
o
s
al
n
etwo
r
k
:
−
Fo
r
ev
er
y
i
n
p
u
t im
a
g
e,
th
e
b
o
u
n
d
in
g
b
o
x
is
u
s
ed
to
s
et
th
e
r
eg
io
n
p
r
o
p
o
s
als.
−
Fo
r
ev
er
y
a
n
ch
o
r
b
o
x
,
p
r
ep
r
esen
ts
th
e
o
b
jectn
ess
s
co
r
e,
wh
i
ch
is
m
ea
s
u
r
ed
as:
p
=
σ
(
w
∙
f
(
X
)
+
b
)
(
1
0
)
−
All
th
e
b
o
x
es with
h
ig
h
o
b
ject
n
ess
s
co
r
es in
itialize
s
tr
o
n
g
ca
n
ce
r
lesi
o
n
s
.
b)
Ma
s
k
p
r
ed
ictio
n
:
f
o
r
ev
e
r
y
R
OI
,
a
b
in
ar
y
m
ask
M
L
es
i
o
n
is
co
n
s
tr
u
cted
f
o
r
t
h
e
lesi
o
n
o
cc
u
r
r
en
ce
s
.
M
L
es
i
o
n
(
x
,
y
)
=
{
1
if
(
x
,
y
)
∈
L
e
s
ion
r
e
gion
0
othe
r
wi
s
e
(
1
1
)
c)
Fil
ter
in
g
b
o
u
n
d
in
g
b
o
x
an
d
p
r
ed
ictio
n
o
f
c
lass
:
−
I
n
th
is
co
n
tex
t,
th
e
n
etwo
r
k
f
i
lter
s
th
e
b
o
u
n
d
in
g
b
o
x
co
o
r
d
i
n
ates
B
=
(
x
,
y
,
w
,
h
)
an
d
in
itializes
c
las
s
lab
els b
ased
o
n
th
e
p
r
o
b
ab
ilit
y
o
f
lesi
o
n
ex
is
ten
ce
.
Step
4
: Fin
al
lay
er
-
in
t
e
g
r
ated
m
ask
an
d
b
o
u
n
d
in
g
b
o
x
T
h
e
f
in
al
lay
er
co
n
tain
s
:
−
Fro
m
U
-
Net,
th
e
liv
er
m
ask
r
e
p
r
esen
ted
as
M
L
i
v
er
.
−
T
h
e
Ma
s
k
R
-
C
NN
M
L
es
i
o
n
an
d
B
o
b
ta
in
th
e
lesi
o
n
m
ask
s
an
d
b
o
u
n
d
in
g
b
o
x
es.
I
n
th
e
f
i
n
al
s
tep
,
th
e
e
n
s
em
b
le
ap
p
r
o
ac
h
u
s
ed
to
im
p
o
v
e
th
e
a
cc
u
r
ac
y
b
ased
o
n
t
h
e
d
etec
tio
n
r
ate.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
7
2
2
-
3
2
2
1
C
o
m
p
u
t Sci
I
n
f
T
ec
h
n
o
l
,
Vo
l.
6
,
No
.
2
,
J
u
ly
2
0
2
5
:
1
6
9
-
1
7
7
174
6.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
I
n
th
is
co
n
tex
t,
th
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
E
MCC
i
s
im
p
lem
en
ted
u
s
in
g
Py
th
o
n
with
b
etter
lib
r
ar
ies
s
u
ch
as Ke
r
as,
Pan
d
as,
an
d
Nu
m
p
y
.
T
h
e
tu
m
o
r
o
r
ca
n
ce
r
d
etec
tio
n
an
d
class
if
icatio
n
is
m
ea
s
u
r
ed
b
y
u
s
in
g
th
e
co
n
f
u
s
io
n
m
atr
ix
ap
p
lied
to
C
T
s
ca
n
im
ag
es.
T
h
e
co
m
b
i
n
ed
ap
p
r
o
ac
h
,
U
-
Net
a
n
d
M
ask
R
-
C
NN,
m
ain
ly
f
o
cu
s
ed
o
n
s
eg
m
en
tatio
n
a
n
d
f
in
d
in
g
ac
cu
r
ate
lesi
o
n
s
u
s
in
g
th
e
b
o
u
n
d
in
g
b
o
x
.
T
h
e
p
r
e
-
tr
ain
ed
m
o
d
el
I
n
ce
p
tio
n
V3
is
u
s
ed
t
o
ex
t
r
ac
t th
e
s
ig
n
if
ican
t f
ea
tu
r
es th
at
o
b
tain
th
e
ab
n
o
r
m
alities
p
r
esen
t in
th
e
in
p
u
t im
ag
e
.
T
h
e
f
o
llo
win
g
eq
u
atio
n
s
ar
e
u
s
ed
to
m
ea
s
u
r
e
th
e
d
etec
tio
n
a
n
d
class
if
icatio
n
r
ate:
MSE
:
T
h
e
MSE
m
ea
s
u
r
es
th
e
av
er
ag
e
s
q
u
a
r
ed
d
if
f
er
e
n
ce
b
e
twee
n
th
e
p
r
e
d
icted
an
d
o
r
i
g
in
al
v
alu
es.
I
n
th
is
co
n
tex
t,
th
e
MSE
m
ea
s
u
r
es
th
e
s
eg
m
en
tatio
n
o
u
tc
o
m
es
b
y
p
r
ed
ictin
g
th
e
tu
m
o
r
s
ize.
T
h
e
eq
u
atio
n
o
f
th
e
MSE
is
m
ea
s
u
r
ed
as
(
1
2
)
:
M
SE
=
1
n
∑
(
y
i
−
y
̂
i
n
i
=
1
)
2
(
12
)
PS
NR
:
T
h
is
r
atio
is
g
en
e
r
ally
u
s
ed
to
m
ea
s
u
r
e
t
h
e
q
u
ality
o
f
th
e
ac
tu
al
in
p
u
t
i
m
a
g
e
an
d
p
r
o
ce
s
s
ed
im
ag
e,
ty
p
ically
in
d
ec
ib
els
(
d
B
)
.
I
t
is
m
ain
ly
f
o
cu
s
ed
o
n
m
e
asu
r
in
g
h
o
w
th
e
ac
tu
al
im
ag
e
d
if
f
er
en
tiates
f
r
o
m
n
o
is
e
f
ilter
s
an
d
tr
an
s
m
is
s
io
n
.
PS
N
R
=
10
∙
l
og
10
(
MA
X
2
MS
E
)
(
1
3
)
T
h
e
p
er
f
o
r
m
an
ce
o
f
ex
is
tin
g
alg
o
r
ith
m
s
is
co
m
p
ar
ed
with
th
e
f
u
zz
y
C
-
m
ea
n
s
(
FC
M
)
an
d
KFC
M
s
h
o
wn
in
T
ab
le
1
.
T
h
ese
r
esu
lts
r
ep
r
ese
n
t
th
e
q
u
ality
o
f
th
e
f
in
al
o
u
tp
u
t.
Fig
u
r
e
3
s
h
o
ws
th
e
co
m
p
a
r
ativ
e
p
er
f
o
r
m
a
n
ce
b
etwe
en
MSE
an
d
PS
NR
.
T
h
e
E
MCC
o
b
tain
ed
th
e
MSE
o
f
1
1
.
3
4
an
d
PS
NR
with
1
0
.
3
4
%
wh
ich
is
h
ig
h
co
m
p
ar
e
with
e
x
is
tin
g
m
o
d
els.
T
ab
le
1
.
T
h
e
q
u
a
n
titativ
e
p
er
f
o
r
m
an
ce
o
f
alg
o
r
ith
m
s
F
C
M
K
F
C
M
[
2
0
]
EM
C
C
M
S
E
3
.
4
5
6
9
.
5
3
1
1
1
.
3
4
P
S
N
R
2
.
6
7
5
8
.
5
4
1
1
0
.
3
4
Fig
u
r
e
3
.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
alg
o
r
ith
m
s
b
ased
o
n
th
e
q
u
ality
o
f
d
etec
tin
g
th
e
ca
n
ce
r
r
e
g
io
n
s
in
th
e
in
p
u
t
im
ag
es
Am
o
n
g
all
th
e
m
o
d
els
th
e
p
r
o
p
o
s
ed
E
C
MM
s
h
o
ws
th
e
h
ig
h
-
q
u
ality
im
ag
es
af
ter
th
e
p
r
e
-
p
r
o
ce
s
s
in
g
an
d
f
ea
tu
r
e
e
x
tr
ac
tio
n
tech
n
iq
u
es.
Fig
u
r
e
4
s
h
o
ws
th
e
d
if
f
e
r
en
ce
b
etwe
en
o
r
ig
i
n
al
C
T
s
ca
n
im
ag
e
an
d
ca
n
ce
r
p
r
ed
ictio
n
im
ag
e.
T
h
e
cla
s
s
if
icatio
n
p
ar
a
m
eter
s
ar
e
s
h
o
wn
b
ased
o
n
th
e
s
tr
en
g
th
o
f
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
.
T
h
ese
p
ar
am
eter
s
ar
e
o
b
tain
ed
f
r
o
m
th
e
co
n
f
u
s
io
n
m
atr
ix
.
T
h
e
class
if
icatio
n
r
esu
lt
s
b
ase
d
o
n
th
e
d
etec
tio
n
r
ate
an
d
th
ese
ar
e
m
ea
s
u
r
ed
b
y
u
s
in
g
th
e
f
o
llo
win
g
:
A
c
c
ura
c
y
(
ACC
)
=
TP
+
TN
TP
+
TN
+
FP
+
FN
(
1
4
)
Sp
e
c
ifi
c
ity
(
Spc
)
=
No
of
TN
No
of
TN
+
No
of
FP
(
1
5
)
Evaluation Warning : The document was created with Spire.PDF for Python.
C
o
m
p
u
t Sci
I
n
f
T
ec
h
n
o
l
I
SS
N:
2722
-
3
2
2
1
Hep
a
to
S
ca
n
:
E
n
s
emb
le
cla
s
s
ifica
tio
n
lea
r
n
in
g
mo
d
els fo
r
liver c
a
n
ce
r
d
is
ea
s
es d
etec
tio
n
(
Tella
S
u
ma
llika
)
175
R
e
c
a
l
l
(
Re
)
=
TP
TP
+
FN
(
1
6
)
F1
−
s
c
or
e
(
F
1S
)
=
2
∗
(
Pr
ecis
i
o
n
∗
Recal
l
)
(
Pr
ecis
i
o
n
+
Recal
l
)
(
1
7
)
Fig
u
r
e
4
.
T
h
e
o
r
i
g
in
al
an
d
p
r
e
d
icted
ca
n
ce
r
r
eg
io
n
im
ag
e
T
ab
le
2
s
h
o
ws
th
e
class
if
icat
io
n
r
esu
lts
b
y
s
h
o
win
g
th
e
co
m
p
ar
is
o
n
b
etwe
en
v
ar
io
u
s
d
ee
p
lear
n
in
g
-
b
ased
alg
o
r
ith
m
.
T
h
e
p
r
o
p
o
s
ed
E
MCC
o
b
tain
s
t
h
e
h
ig
h
class
if
icatio
n
co
m
p
a
r
e
with
o
th
er
ex
is
tin
g
m
o
d
els
as
s
h
o
wn
in
Fig
u
r
e
5
.
T
h
ese
r
esu
lts
ar
e
o
b
tain
ed
b
a
s
ed
o
n
th
e
co
u
n
t
v
alu
es
o
f
th
e
co
n
f
u
s
io
n
m
atr
ix
attr
ib
u
tes.
I
n
th
is
co
m
p
a
r
is
o
n
,
th
e
lo
west
p
er
f
o
r
m
a
n
ce
is
s
h
o
wn
by
d
ee
p
n
e
u
r
al
n
et
wo
r
k
-
b
a
s
ed
Gab
o
r
f
ea
tu
r
es
(
DNN
-
GF
)
with
th
e
ac
cu
r
ac
y
o
f
0
.
8
0
%,
Sp
c
-
0
.
9
9
,
R
e
-
0
.
8
5
,
an
d
F1
S o
f
0
.
6
9
%.
T
ab
le
2
.
T
h
e
q
u
a
n
titativ
e
p
er
f
o
r
m
an
ce
o
f
liv
er
ca
n
ce
r
im
a
g
e
s
u
s
in
g
in
ter
m
s
o
f
class
if
icatio
n
with
E
MCC
A
C
C
S
p
c
Re
F
1
S
DNN
-
G
F
[
2
4
]
0
.
8
0
0
.
9
9
0
.
8
5
0
.
6
9
HI
-
DNN
[
2
5
]
0
.
8
3
0
.
9
8
0
.
9
3
0
.
8
5
C
o
o
t
o
p
t
i
m
i
z
a
t
i
o
n
a
l
g
o
r
i
t
h
m (CO
A
)
[
4
]
0
.
8
7
0
.
9
9
0
.
9
9
0
.
8
4
EM
C
C
0
.
9
7
0
.
9
9
0
.
9
9
0
.
9
7
Fig
u
r
e
5
.
T
h
e
co
m
p
ar
is
o
n
o
f
li
v
er
ca
n
ce
r
d
etec
tio
n
s
am
p
les i
n
ter
m
s
o
f
class
if
icatio
n
7.
CO
NCLU
SI
O
N
T
h
is
p
ap
er
p
r
esen
ted
Hep
ato
Scan
,
an
en
s
em
b
le
class
if
icatio
n
m
eth
o
d
f
o
r
th
e
d
etec
tio
n
an
d
d
iag
n
o
s
is
o
f
liv
er
ca
n
ce
r
tu
m
o
r
s
u
s
in
g
liv
er
ca
n
ce
r
d
atasets
.
T
h
e
p
r
o
p
o
s
ed
Hep
ato
Scan
was
an
in
te
g
r
ated
m
eth
o
d
th
a
t
ca
teg
o
r
izes
th
e
th
r
ee
t
y
p
es
o
f
l
iv
er
ca
n
ce
r
: h
ep
ato
ce
llu
lar
c
ar
cin
o
m
a,
ch
o
lan
g
io
ca
r
cin
o
m
a
,
an
d
an
g
io
s
ar
co
m
a
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
7
2
2
-
3
2
2
1
C
o
m
p
u
t Sci
I
n
f
T
ec
h
n
o
l
,
Vo
l.
6
,
No
.
2
,
J
u
ly
2
0
2
5
:
1
6
9
-
1
7
7
176
T
h
is
p
ap
er
d
em
o
n
s
tr
ated
th
at
th
e
in
teg
r
ated
m
eth
o
d
o
f
E
MCC
ex
h
ib
its
s
u
p
er
io
r
ef
f
icac
y
in
id
en
tify
in
g
lesi
o
n
s
an
d
class
if
y
in
g
liv
er
ca
n
ce
r
tu
m
o
r
s
with
in
th
e
p
r
o
v
id
ed
s
am
p
les.
T
h
e
p
r
e
-
tr
ain
e
d
m
o
d
el
id
en
tifie
s
p
r
ec
is
e
an
d
h
ig
h
-
d
im
e
n
s
io
n
al
f
ea
tu
r
es th
r
o
u
g
h
th
e
co
n
v
o
l
u
tio
n
al
lay
er
i
n
co
n
ju
n
ctio
n
with
th
e
tr
an
s
f
er
lear
n
in
g
lay
er
.
T
h
e
p
r
e
-
tr
ain
ed
m
o
d
el
d
em
o
n
s
tr
at
es
a
m
in
im
al
er
r
o
r
lo
s
s
in
th
e
lo
s
s
f
u
n
ctio
n
.
T
h
e
E
MCC
in
teg
r
ates
U
-
Net
an
d
M
ask
R
-
C
NN
to
id
en
tify
ab
n
o
r
m
al
(
ca
n
ce
r
o
u
s
ce
lls
o
r
tu
m
o
r
s
)
an
d
n
o
r
m
al
tis
s
u
es.
T
h
e
en
co
d
er
,
b
o
ttlen
ec
k
lay
er
,
an
d
d
ec
o
d
e
r
ex
e
cu
te
th
e
U
-
Net
f
u
n
ctio
n
.
W
e
em
p
lo
y
t
h
e
M
ask
R
-
C
NN
to
d
etec
t
th
e
an
o
m
alo
u
s
r
eg
io
n
s
in
th
e
p
r
o
v
id
e
d
in
p
u
t
im
ag
e.
T
h
e
f
in
al
lay
er
co
m
b
in
es
th
e
m
ask
an
d
b
o
u
n
d
in
g
b
o
x
t
o
d
elin
ea
te
p
r
ec
is
e
r
eg
io
n
s
.
T
h
e
class
if
icatio
n
o
u
tco
m
es
d
em
o
n
s
tr
ate
th
at
th
e
p
r
o
p
o
s
ed
m
eth
o
d
attain
s
an
ac
cu
r
ac
y
o
f
0
.
9
7
%,
s
p
ec
if
icity
o
f
0
.
9
9
%,
r
ec
all
o
f
0
.
9
9
%,
an
d
F1
s
co
r
e
o
f
0
.
9
7
%.
All
th
ese
v
alu
es
in
d
icate
a
h
ig
h
d
eg
r
ee
o
f
s
u
p
er
io
r
class
if
icatio
n
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
is
r
esear
ch
d
id
n
o
t
r
ec
eiv
e
an
y
s
p
ec
if
ic
g
r
an
t
f
r
o
m
f
u
n
d
i
n
g
ag
en
cies
in
th
e
p
u
b
lic,
co
m
m
er
cial,
o
r
not
-
f
o
r
-
p
r
o
f
it secto
r
s
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
T
ella
Su
m
allik
a
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
R
aa
v
i Saty
a
Pra
s
ad
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
No
n
ew
d
ata
wer
e
g
en
er
ate
d
o
r
an
aly
ze
d
d
u
r
in
g
th
is
s
tu
d
y
.
All
d
ata
u
s
ed
a
r
e
f
r
o
m
p
u
b
licly
av
ailab
le
s
o
u
r
ce
s
cited
in
th
e
m
a
n
u
s
cr
ip
t.
RE
F
E
R
E
NC
E
S
[
1
]
A
.
D
u
t
t
a
a
n
d
A
.
D
u
b
e
y
,
“
D
e
t
e
c
t
i
o
n
o
f
l
i
v
e
r
c
a
n
c
e
r
u
s
i
n
g
i
ma
g
e
p
r
o
c
e
ss
i
n
g
t
e
c
h
n
i
q
u
e
s
,
”
i
n
2
0
1
9
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
m
u
n
i
c
a
t
i
o
n
a
n
d
S
i
g
n
a
l
Pr
o
c
e
ssi
n
g
(
I
C
C
S
P)
,
A
p
r
.
2
0
1
9
,
p
p
.
3
1
5
–
3
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
S
P
.
2
0
1
9
.
8
6
9
8
0
3
3
.
[
2
]
S
.
C
h
a
r
a
n
,
G
.
U
g
a
n
y
a
,
a
n
d
M
.
N
.
K
u
mar,
“
C
o
m
p
a
r
i
s
o
n
o
f
a
c
c
u
r
a
c
y
a
n
d
se
n
si
t
i
v
i
t
y
i
n
l
i
v
e
r
c
a
n
c
e
r
se
g
me
n
t
a
t
i
o
n
o
f
m
a
g
n
e
t
i
c
r
e
so
n
a
n
c
e
i
m
a
g
e
s
u
si
n
g
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
i
n
c
o
mp
a
r
i
s
o
n
w
i
t
h
su
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
,
”
i
n
2
0
2
2
1
4
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Ma
t
h
e
m
a
t
i
c
s,
A
c
t
u
a
r
i
a
l
S
c
i
e
n
c
e
,
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
S
t
a
t
i
st
i
c
s
(
MA
C
S
)
,
2
0
2
2
,
p
p
.
1
–
9
,
d
o
i
:
1
0
.
1
1
0
9
/
M
A
C
S
5
6
7
7
1
.
2
0
2
2
.
1
0
0
2
3
0
4
8
.
[
3
]
G
.
L.
K
u
l
k
a
r
n
i
,
S
.
S
.
S
a
n
n
a
k
k
i
,
a
n
d
V
.
S
.
R
a
j
p
u
r
o
h
i
t
,
“
T
e
x
t
u
r
e
f
e
a
t
u
r
e
a
n
a
l
y
si
s
f
o
r
t
h
e
l
i
v
e
r
c
a
n
c
e
r
d
i
se
a
ses
u
si
n
g
st
a
t
i
st
i
c
a
l
b
a
se
d
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
t
e
c
h
n
i
q
u
e
,
”
i
n
2
0
2
0
F
o
u
r
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
I
-
S
MA
C
(
I
o
T
i
n
S
o
c
i
a
l
,
M
o
b
i
l
e
,
A
n
a
l
y
t
i
c
s
a
n
d
C
l
o
u
d
)
(I
-
S
MA
C
)
,
2
0
2
0
,
p
p
.
1
0
2
8
–
1
0
3
3
,
d
o
i
:
1
0
.
1
1
0
9
/
I
-
S
M
A
C
4
9
0
9
0
.
2
0
2
0
.
9
2
4
3
4
9
6
.
[
4
]
K
.
S
r
i
d
h
a
r
,
C
.
K
a
v
i
t
h
a
,
W
.
C
.
La
i
,
a
n
d
B
.
P
.
K
a
v
i
n
,
“
D
e
t
e
c
t
i
o
n
o
f
l
i
v
e
r
t
u
mo
u
r
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
b
a
se
d
s
e
g
m
e
n
t
a
t
i
o
n
w
i
t
h
c
o
o
t
e
x
t
r
e
me
l
e
a
r
n
i
n
g
mo
d
e
l
,
”
B
i
o
m
e
d
i
c
i
n
e
s
,
v
o
l
.
1
1
,
n
o
.
3
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
b
i
o
me
d
i
c
i
n
e
s1
1
0
3
0
8
0
0
.
[
5
]
B
.
L
a
k
s
h
m
i
p
r
i
y
a
,
B
.
P
o
t
t
a
k
k
a
t
,
a
n
d
G
.
R
a
m
k
u
mar,
“
D
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
i
n
l
i
v
e
r
t
u
mo
u
r
d
i
a
g
n
o
s
i
s
u
si
n
g
C
T
a
n
d
M
R
i
ma
g
i
n
g
-
A
sy
s
t
e
m
a
t
i
c
r
e
v
i
e
w
,
”
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
i
n
Me
d
i
c
i
n
e
,
v
o
l
.
1
4
1
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
r
t
m
e
d
.
2
0
2
3
.
1
0
2
5
5
7
.
[
6
]
M
.
C
h
e
n
e
t
a
l
.
,
“
C
l
a
ssi
f
i
c
a
t
i
o
n
a
n
d
m
u
t
a
t
i
o
n
p
r
e
d
i
c
t
i
o
n
b
a
se
d
o
n
h
i
st
o
p
a
t
h
o
l
o
g
y
H
&
E
i
m
a
g
e
s i
n
l
i
v
e
r
c
a
n
c
e
r
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
”
n
p
j
Pr
e
c
i
si
o
n
O
n
c
o
l
o
g
y
,
v
o
l
.
4
,
n
o
.
1
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
3
8
/
s4
1
6
9
8
-
0
2
0
-
0
1
2
0
-
3.
[
7
]
K
.
C
h
a
u
d
h
a
r
y
,
O
.
B
.
P
o
i
r
i
o
n
,
L
.
L
u
,
a
n
d
L
.
X
.
G
a
r
m
i
r
e
,
“
D
e
e
p
l
e
a
r
n
i
n
g
–
b
a
se
d
m
u
l
t
i
-
o
mi
c
s
i
n
t
e
g
r
a
t
i
o
n
r
o
b
u
st
l
y
p
r
e
d
i
c
t
s
s
u
r
v
i
v
a
l
i
n
l
i
v
e
r
c
a
n
c
e
r
,
”
C
l
i
n
i
c
a
l
C
a
n
c
e
r
Re
s
e
a
rc
h
,
v
o
l
.
2
4
,
n
o
.
6
,
p
p
.
1
2
4
8
–
1
2
5
9
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
5
8
/
1
0
7
8
-
0
4
3
2
.
C
C
R
-
17
-
0
8
5
3
.
[
8
]
J.
G
r
e
g
o
r
y
,
M
.
D
i
o
g
u
a
r
d
i
B
u
r
g
i
o
,
G
.
C
o
r
r
i
a
s,
V
.
V
i
l
g
r
a
i
n
,
a
n
d
M
.
R
o
n
o
t
,
“
E
v
a
l
u
a
t
i
o
n
o
f
l
i
v
e
r
t
u
m
o
u
r
r
e
s
p
o
n
se
b
y
i
ma
g
i
n
g
,
”
J
H
EP
R
e
p
o
r
t
s
,
v
o
l
.
2
,
n
o
.
3
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
h
e
p
r
.
2
0
2
0
.
1
0
0
1
0
0
.
[
9
]
C
.
J.
W
a
n
g
e
t
a
l
.
,
“
D
e
e
p
l
e
a
r
n
i
n
g
f
o
r
l
i
v
e
r
t
u
m
o
r
d
i
a
g
n
o
s
i
s
p
a
r
t
I
I
:
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
i
n
t
e
r
p
r
e
t
a
t
i
o
n
u
si
n
g
r
a
d
i
o
l
o
g
i
c
i
ma
g
i
n
g
f
e
a
t
u
r
e
s,
”
E
u
r
o
p
e
a
n
Ra
d
i
o
l
o
g
y
,
v
o
l
.
2
9
,
n
o
.
7
,
p
p
.
3
3
4
8
–
3
3
5
7
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/
s0
0
3
3
0
-
0
1
9
-
0
6
2
1
4
-
8.
[
1
0
]
V
.
V
e
k
a
r
i
y
a
,
T
.
G
o
s
w
a
m
i
,
S
.
S
i
n
g
h
,
K
.
G
h
o
d
k
e
,
I
.
S
.
A
b
d
u
l
r
a
h
m
a
n
,
a
n
d
A
.
J
a
i
n
,
“
D
e
t
e
r
m
i
n
a
t
i
o
n
o
f
a
c
c
u
r
a
c
y
o
f
n
e
u
r
a
l
n
e
t
w
o
r
k
m
e
t
h
o
d
u
s
i
n
g
m
a
g
n
e
t
i
c
r
e
s
o
n
a
n
c
e
i
m
a
g
e
s
i
n
f
i
n
d
i
n
g
l
i
v
e
r
c
a
n
c
e
r
l
e
v
e
l
,
”
i
n
2
0
2
3
3
r
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
A
d
v
a
n
c
e
C
o
m
p
u
t
i
n
g
a
n
d
I
n
n
o
v
a
t
i
v
e
T
e
c
h
n
o
l
o
g
i
e
s
i
n
E
n
g
i
n
e
e
r
i
n
g
(
I
C
A
C
I
T
E
)
,
2
0
2
3
,
p
p
.
7
0
2
–
7
0
5
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
A
C
I
T
E
5
7
4
1
0
.
2
0
2
3
.
1
0
1
8
2
9
0
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
C
o
m
p
u
t Sci
I
n
f
T
ec
h
n
o
l
I
SS
N:
2722
-
3
2
2
1
Hep
a
to
S
ca
n
:
E
n
s
emb
le
cla
s
s
ifica
tio
n
lea
r
n
in
g
mo
d
els fo
r
liver c
a
n
ce
r
d
is
ea
s
es d
etec
tio
n
(
Tella
S
u
ma
llika
)
177
[
1
1
]
S
.
H
.
Zh
e
n
e
t
a
l
.
,
“
D
e
e
p
l
e
a
r
n
i
n
g
f
o
r
a
c
c
u
r
a
t
e
d
i
a
g
n
o
si
s
o
f
l
i
v
e
r
t
u
mo
r
b
a
sed
o
n
ma
g
n
e
t
i
c
r
e
s
o
n
a
n
c
e
i
ma
g
i
n
g
a
n
d
c
l
i
n
i
c
a
l
d
a
t
a
,
”
Fro
n
t
i
e
rs
i
n
O
n
c
o
l
o
g
y
,
v
o
l
.
1
0
,
2
0
2
0
,
d
o
i
:
1
0
.
3
3
8
9
/
f
o
n
c
.
2
0
2
0
.
0
0
6
8
0
.
[
1
2
]
O
.
I
.
A
l
i
r
r
,
“
D
e
e
p
l
e
a
r
n
i
n
g
a
n
d
l
e
v
e
l
set
a
p
p
r
o
a
c
h
f
o
r
l
i
v
e
r
a
n
d
t
u
mo
r
s
e
g
men
t
a
t
i
o
n
f
r
o
m
C
T
sc
a
n
s,”
J
o
u
r
n
a
l
o
f
Ap
p
l
i
e
d
C
l
i
n
i
c
a
l
Me
d
i
c
a
l
P
h
y
s
i
c
s
,
v
o
l
.
2
1
,
n
o
.
1
0
,
p
p
.
2
0
0
–
2
0
9
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
0
2
/
a
c
m
2
.
1
3
0
0
3
.
[
1
3
]
T.
S
u
m
a
l
l
i
k
a
a
n
d
R
.
S
.
P
r
a
sa
d
,
“
A
c
o
mb
i
n
e
d
e
n
sem
b
l
e
m
o
d
e
l
(
C
E
M
)
f
o
r
a
l
i
v
e
r
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
sy
s
t
e
m
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Ad
v
a
n
c
e
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
5
,
n
o
.
2
,
p
p
.
1
6
3
–
1
7
2
,
2
0
2
4
,
d
o
i
:
1
0
.
1
4
5
6
9
/
I
JA
C
S
A
.
2
0
2
4
.
0
1
5
0
2
1
8
.
[
1
4
]
T.
S
u
ma
l
l
i
k
a
a
n
d
R
.
S
.
P
r
a
sa
d
,
“
Li
v
e
r
S
c
o
p
e
:
a
n
a
u
t
o
ma
t
e
d
d
e
e
p
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
l
i
v
e
r
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
,
”
2
0
2
5
6
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Mo
b
i
l
e
C
o
m
p
u
t
i
n
g
a
n
d
S
u
s
t
a
i
n
a
b
l
e
I
n
f
o
rm
a
t
i
c
s
(
I
C
MC
S
I
)
,
G
o
a
t
h
g
a
u
n
,
N
e
p
a
l
,
2
0
2
5
,
p
p
.
1
6
6
9
-
1
6
7
3
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
M
C
S
I
6
4
6
2
0
.
2
0
2
5
.
1
0
8
8
3
3
6
1
.
[
1
5
]
L.
Li
u
e
t
a
l
.
,
“
C
T
i
m
a
g
e
se
g
m
e
n
t
a
t
i
o
n
met
h
o
d
o
f
l
i
v
e
r
t
u
m
o
r
b
a
s
e
d
o
n
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
e
n
a
b
l
e
d
m
e
d
i
c
a
l
i
ma
g
i
n
g
,
”
Ma
t
h
e
m
a
t
i
c
a
l
Pr
o
b
l
e
m
s i
n
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
2
0
2
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
1
/
9
9
1
9
5
0
7
.
[
1
6
]
S
.
D
i
,
Y
.
Zh
a
o
,
M
.
Li
a
o
,
Z
.
Y
a
n
g
,
a
n
d
Y
.
Ze
n
g
,
“
A
u
t
o
ma
t
i
c
l
i
v
e
r
t
u
m
o
r
s
e
g
me
n
t
a
t
i
o
n
f
r
o
m C
T
i
m
a
g
e
s u
si
n
g
h
i
e
r
a
r
c
h
i
c
a
l
i
t
e
r
a
t
i
v
e
su
p
e
r
p
i
x
e
l
s
a
n
d
l
o
c
a
l
st
a
t
i
s
t
i
c
a
l
f
e
a
t
u
r
e
s,”
Ex
p
e
r
t
S
y
s
t
e
m
s
w
i
t
h
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
0
3
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
2
.
1
1
7
3
4
7
.
[
1
7
]
Y
.
L
i
,
B
.
Z
o
u
,
a
n
d
Q
.
L
i
u
,
“
A
d
e
e
p
a
t
t
e
n
t
i
o
n
n
e
t
w
o
r
k
v
i
a
h
i
gh
-
r
e
so
l
u
t
i
o
n
r
e
p
r
e
s
e
n
t
a
t
i
o
n
f
o
r
l
i
v
e
r
a
n
d
l
i
v
e
r
t
u
mo
r
se
g
me
n
t
a
t
i
o
n
,
”
Bi
o
c
y
b
e
r
n
e
t
i
c
s
a
n
d
Bi
o
m
e
d
i
c
a
l
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
4
1
,
n
o
.
4
,
p
p
.
1
5
1
8
–
1
5
3
2
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
b
b
e
.
2
0
2
1
.
0
8
.
0
1
0
.
[
1
8
]
P
.
G
u
n
a
se
k
h
a
r
a
n
d
S
.
V
i
j
a
y
a
l
a
k
s
h
mi
,
“
O
p
t
i
ma
l
b
i
o
mark
e
r
sel
e
c
t
i
o
n
u
s
i
n
g
a
d
a
p
t
i
v
e
s
o
c
i
a
l
sk
i
-
d
r
i
v
e
r
o
p
t
i
mi
z
a
t
i
o
n
f
o
r
l
i
v
e
r
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
,
”
Bi
o
c
y
b
e
r
n
e
t
i
c
s
a
n
d
B
i
o
m
e
d
i
c
a
l
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
4
0
,
n
o
.
4
,
p
p
.
1
6
1
1
–
1
6
2
5
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
b
b
e
.
2
0
2
0
.
1
0
.
0
0
5
.
[
1
9
]
A
.
B
a
â
z
a
o
u
i
,
W
.
B
a
r
h
o
u
mi
,
A
.
A
h
m
e
d
,
a
n
d
E
.
Za
g
r
o
u
b
a
,
“
S
e
mi
-
a
u
t
o
m
a
t
e
d
s
e
g
m
e
n
t
a
t
i
o
n
o
f
s
i
n
g
l
e
a
n
d
mu
l
t
i
p
l
e
t
u
m
o
r
s
i
n
l
i
v
e
r
C
T
i
ma
g
e
s
u
si
n
g
e
n
t
r
o
p
y
-
b
a
se
d
f
u
z
z
y
r
e
g
i
o
n
g
r
o
w
i
n
g
,
”
I
RBM
-
I
n
n
o
v
a
t
i
o
n
a
n
d
R
e
se
a
rc
h
i
n
B
i
o
Me
d
i
c
a
l
e
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
3
8
,
n
o
.
2
,
p
p
.
9
8
–
1
0
8
,
2
0
1
7
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
r
b
m.
2
0
1
7
.
0
2
.
0
0
3
.
[
2
0
]
A
.
D
a
s
a
n
d
S
.
K
.
S
a
b
u
t
,
“
K
e
r
n
e
l
i
z
e
d
f
u
z
z
y
c
-
mea
n
s
c
l
u
s
t
e
r
i
n
g
w
i
t
h
a
d
a
p
t
i
v
e
t
h
r
e
s
h
o
l
d
i
n
g
f
o
r
se
g
me
n
t
i
n
g
l
i
v
e
r
t
u
mo
r
s,”
Pr
o
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
9
2
,
p
p
.
3
8
9
–
3
9
5
,
2
0
1
6
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s.
2
0
1
6
.
0
7
.
3
9
5
.
[
2
1
]
S
.
K
o
l
l
i
,
B
.
R
.
P
a
r
v
a
t
h
a
l
a
,
a
n
d
A
.
V
.
P
.
K
r
i
s
h
n
a
,
“
A
n
o
v
e
l
l
i
v
e
r
t
u
mo
r
c
l
a
s
si
f
i
c
a
t
i
o
n
u
s
i
n
g
i
mp
r
o
v
e
d
p
r
o
b
a
b
i
l
i
st
i
c
n
e
u
r
a
l
n
e
t
w
o
r
k
s
w
i
t
h
B
a
y
e
si
a
n
o
p
t
i
mi
z
a
t
i
o
n
,
”
e
-
P
ri
m
e
-
A
d
v
a
n
c
e
s
i
n
El
e
c
t
ri
c
a
l
E
n
g
i
n
e
e
ri
n
g
,
E
l
e
c
t
r
o
n
i
c
s
a
n
d
En
e
r
g
y
,
v
o
l
.
8
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
i
m
e
.
2
0
2
4
.
1
0
0
5
1
4
.
[
2
2
]
J
.
K
a
u
r
a
n
d
P
.
K
a
u
r
,
“
P
S
O
-
PSP
-
N
e
t
+
I
n
c
e
p
t
i
o
n
V
3
:
A
n
o
p
t
i
m
i
z
e
d
h
y
p
e
r
-
p
a
r
a
m
e
t
e
r
t
u
n
e
d
c
o
m
p
u
t
e
r
-
a
i
d
e
d
d
i
a
g
n
o
s
t
i
c
m
o
d
e
l
f
o
r
l
i
v
e
r
t
u
m
o
r
d
e
t
e
c
t
i
o
n
u
s
i
n
g
C
T
s
c
a
n
s
l
i
c
e
s
,
”
B
i
o
m
e
d
i
c
a
l
S
i
g
n
a
l
P
r
o
c
e
s
s
i
n
g
a
n
d
C
o
n
t
r
o
l
,
v
o
l
.
9
5
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
b
s
p
c
.
2
0
2
4
.
1
0
6
4
4
2
.
[
2
3
]
N
.
K
h
a
l
e
d
,
H
.
Ta
r
e
k
,
M
.
M
a
k
r
a
m,
a
n
d
A
.
M
o
h
a
mm
e
d
,
“
E
n
h
a
n
c
e
d
d
i
a
g
n
o
s
i
s
o
f
l
i
v
e
r
c
a
n
c
e
r
su
b
t
y
p
e
s
t
h
r
o
u
g
h
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,
”
i
n
2
0
2
4
6
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
i
n
g
a
n
d
I
n
f
o
rm
a
t
i
c
s
(
I
C
C
I
)
,
2
0
2
4
,
p
p
.
1
8
9
–
1
9
4
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
I
6
1
6
7
1
.
2
0
2
4
.
1
0
4
8
5
1
7
8
.
[
2
4
]
B
.
A
sh
r
e
e
t
h
a
,
M
.
R
.
D
e
v
i
,
U
.
P
.
K
u
m
a
r
,
M
.
K
.
M
a
n
i
,
D
.
N
.
S
a
h
u
,
a
n
d
P
.
C
.
S
.
R
e
d
d
y
,
“
S
o
f
t
o
p
t
i
m
i
z
a
t
i
o
n
t
e
c
h
n
i
q
u
e
s
f
o
r
a
u
t
o
m
a
t
i
c
l
i
v
e
r
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
i
n
a
b
d
o
mi
n
a
l
l
i
v
e
r
i
ma
g
e
s,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
H
e
a
l
t
h
S
c
i
e
n
c
e
s
,
v
o
l
.
6
,
p
p
.
1
0
8
2
0
–
1
0
8
3
1
,
2
0
2
2
,
d
o
i
:
1
0
.
5
3
7
3
0
/
i
j
h
s.
v
6
n
s
1
.
7
5
9
7
.
[
2
5
]
M
.
R
e
l
a
,
N
.
R
.
S
u
r
y
a
k
a
r
i
,
a
n
d
R
.
R
.
P
a
t
i
l
,
“
A
d
i
a
g
n
o
si
s
sy
s
t
e
m
b
y
U
-
N
e
t
a
n
d
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
e
n
a
b
l
e
d
w
i
t
h
o
p
t
i
m
a
l
f
e
a
t
u
r
e
sel
e
c
t
i
o
n
f
o
r
l
i
v
e
r
t
u
mo
r
d
e
t
e
c
t
i
o
n
u
s
i
n
g
C
T
i
ma
g
e
s
,
”
M
u
l
t
i
m
e
d
i
a
T
o
o
l
s
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
8
2
,
n
o
.
3
,
p
p
.
3
1
8
5
–
3
2
2
7
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
0
4
2
-
022
-
1
3
3
8
1
-
2.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Te
ll
a
S
u
m
a
ll
ik
a
wo
r
k
i
n
g
a
s
a
n
a
ss
istan
t
p
r
o
fe
ss
o
r
i
n
De
p
a
rtme
n
t
o
f
I
n
fo
rm
a
ti
o
n
Tec
h
n
o
l
o
g
y
i
n
G
u
d
lav
a
ll
e
ru
En
g
i
n
e
e
rin
g
C
o
ll
e
g
e
,
h
a
v
i
n
g
1
2
y
e
a
rs
o
f
e
x
p
e
rien
c
e
.
S
h
e
p
u
rsu
i
n
g
P
h
.
D
.
i
n
c
o
m
p
u
ter
sc
ien
c
e
a
n
d
e
n
g
in
e
e
rin
g
,
Ac
h
a
r
y
a
Na
g
a
rju
n
a
Un
iv
e
rsity
.
S
h
e
r
e
c
e
iv
e
d
M
.
Tec
h
.
i
n
c
o
m
p
u
ter
sc
ien
c
e
e
n
g
in
e
e
r
in
g
fr
o
m
P
y
d
a
h
En
g
i
n
e
e
rin
g
C
o
ll
e
g
e
in
2
0
1
1
.
C
u
rre
n
t
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
b
ig
d
a
ta
a
n
a
ly
ti
c
s
,
Io
T
,
a
n
d
m
a
c
h
in
e
lea
rn
in
g
.
S
h
e
p
u
b
li
s
h
e
d
8
re
se
a
rc
h
p
a
p
e
rs
in
v
a
ri
o
u
s
jo
u
r
n
a
ls,
p
re
se
n
ted
4
p
a
p
e
rs
i
n
v
a
rio
u
s
c
o
n
fe
re
n
c
e
s,
a
p
p
li
e
d
re
se
a
rc
h
wo
rk
s
in
DST
,
S
ER
B.
S
h
e
r
e
c
e
iv
e
d
a
p
p
re
c
iatio
n
c
e
rti
fica
te
fo
r
th
e
b
e
st
p
e
rfo
rm
a
n
c
e
in
tea
c
h
in
g
.
S
h
e
r
e
c
e
iv
e
d
El
it
e
+
Go
ld
,
e
li
te
o
n
li
n
e
-
c
o
u
rse
c
e
rti
fica
tes
fo
r
v
a
ri
o
u
s
su
b
jec
ts
fro
m
NPT
EL
,
s
u
n
c
e
rti
fica
ti
o
n
i
n
J
AV
A
.
S
h
e
is
w
o
r
k
i
n
g
in
a
p
ro
g
re
ss
iv
e
a
n
d
c
h
a
ll
e
n
g
in
g
e
n
v
iro
n
m
e
n
t
,
wh
e
re
sh
e
c
a
n
e
n
h
a
n
c
e
sk
il
ls an
d
p
o
ten
ti
a
ls t
o
t
h
e
b
e
st an
d
e
x
p
l
o
it
th
e
m
to
c
re
a
te
n
e
w
a
v
e
n
u
e
s
in
th
e
field
o
f
i
n
fo
rm
a
ti
o
n
t
e
c
h
n
o
lo
g
y
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
su
m
a
ll
ik
a
.
p
@
g
m
a
il
.
c
o
m
.
Ra
a
v
i
S
a
ty
a
Pra
sa
d
o
b
tai
n
e
d
h
is
B.
S
c
.
(
1
9
8
1
)
a
n
d
M
.
S
c
.
wi
th
a
G
o
ld
M
e
d
a
l
(1
9
8
3
)
fro
m
Na
g
a
rju
n
a
Un
i
v
e
rsity
,
fo
l
lo
we
d
b
y
M
.
P
h
i
l.
stu
d
ies
a
t
th
e
sa
m
e
in
stit
u
ti
o
n
.
He
late
r
e
a
rn
e
d
a
n
M
.
Tec
h
.
in
c
o
m
p
u
ter
sc
ien
c
e
(F
irst
Clas
s)
fro
m
A
c
h
a
ry
a
Na
g
a
rju
n
a
Un
i
v
e
rsity
in
2
0
1
0
a
n
d
a
P
h
.
D.
i
n
h
a
lf
l
o
g
isti
c
so
ftwa
re
re
li
a
b
il
it
y
g
r
o
wth
m
o
d
e
l
fr
o
m
t
h
e
sa
m
e
u
n
i
v
e
rsity
.
Cu
rre
n
tl
y
,
h
e
is
a
P
ro
fe
ss
o
r
a
n
d
De
a
n
i
n
Dh
a
n
e
k
u
la
I
n
stit
u
te
o
f
En
g
i
n
e
e
rin
g
a
n
d
Tec
h
n
o
l
o
g
y
-
JN
TU
Ka
k
i
n
a
d
a
,
Vijay
a
wa
d
a
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
d
e
a
n
re
se
a
rc
h
@d
iet.
a
c
.
in
o
r
p
r
o
frs
p
@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.