I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
1
4
,
No
.
6
,
Dec
em
b
er
2
0
2
5
,
p
p
.
4
7
3
1
~
4
7
3
8
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
14
.i
6
.
p
p
4
7
3
1
-
4
7
3
8
4731
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
A t
wo
-
s
tep
i
ntelli
g
ent
frame
wo
rk f
o
r gene e
x
press
io
n
-
ba
sed
ca
ncer diag
no
sis
Sa
ra
H
a
dd
o
u B
o
ua
zz
a
1
,
J
iha
d H
a
dd
o
u
B
o
ua
zz
a
2
1
LA
M
I
G
EP
La
b
o
r
a
t
o
r
y
,
E
M
S
I
M
o
r
o
c
c
a
n
S
c
h
o
o
l
o
f
E
n
g
i
n
e
e
r
i
n
g
,
M
a
r
r
a
k
e
s
h
,
M
o
r
o
c
c
o
2
I
G
A
-
I
n
st
i
t
u
t
S
u
p
é
r
i
e
u
r
d
u
G
é
n
i
e
A
p
p
l
i
q
u
é
,
C
a
sa
b
l
a
n
c
a
,
M
o
r
o
c
c
o
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ap
r
2
2
,
2
0
2
5
R
ev
is
ed
Sep
8
,
2
0
2
5
Acc
ep
ted
Oct
1
6
,
2
0
2
5
DN
A
m
icro
a
rra
y
tec
h
n
o
lo
g
y
h
a
s
a
d
v
a
n
c
e
d
c
a
n
c
e
r
d
iag
n
o
sis
b
y
e
n
a
b
li
n
g
larg
e
-
sc
a
le
g
e
n
e
e
x
p
re
ss
io
n
a
n
a
ly
sis,
y
e
t
c
h
a
ll
e
n
g
e
s
re
m
a
in
in
se
lec
ti
n
g
re
lev
a
n
t
g
e
n
e
s
a
n
d
a
c
h
iev
in
g
a
c
c
u
ra
te
c
las
sifica
ti
o
n
.
Th
is
st
u
d
y
in
tro
d
u
c
e
s
two
n
o
v
e
l
m
e
th
o
d
s:
th
e
th
re
e
-
sta
g
e
g
e
n
e
se
lec
ti
o
n
(3
S
G
S
)
m
e
th
o
d
a
n
d
th
e
sta
ti
stics
c
las
sifier
(S
C).
By
e
li
m
in
a
ti
n
g
re
d
u
n
d
a
n
t,
n
o
isy
,
a
n
d
les
s
in
fo
rm
a
ti
v
e
g
e
n
e
s,
th
e
3
S
G
S
m
e
th
o
d
e
ffe
c
ti
v
e
l
y
l
o
we
rs
th
e
d
ime
n
sio
n
a
li
ty
o
f
g
e
n
e
e
x
p
re
ss
io
n
d
a
ta,
wh
il
e
th
e
S
C
c
las
sifier
u
se
s
sta
ti
stica
l
m
e
a
su
re
s
o
f
ge
n
e
e
x
p
re
ss
io
n
to
c
las
sify
sa
m
p
les
with
h
i
g
h
a
c
c
u
ra
c
y
a
n
d
sp
e
e
d
.
Ev
a
lu
a
te
d
o
n
le
u
k
e
m
ia,
p
r
o
sta
te
c
a
n
c
e
r,
a
n
d
c
o
l
o
n
c
a
n
c
e
r
d
a
tas
e
ts,
t
h
e
3
S
G
S
m
e
th
o
d
e
ffe
c
ti
v
e
ly
id
e
n
ti
fied
m
in
ima
l
y
e
t
in
f
o
rm
a
ti
v
e
g
e
n
e
su
b
se
ts,
a
c
h
iev
in
g
1
0
0
%
a
c
c
u
ra
c
y
fo
r
leu
k
e
m
ia,
9
9
.
3
%
f
o
r
p
ro
sta
te
c
a
n
c
e
r,
a
n
d
9
7
%
fo
r
c
o
l
o
n
c
a
n
c
e
r.
Th
e
S
C
c
las
sifier
c
o
n
siste
n
tl
y
o
u
t
p
e
rfo
rm
e
d
trad
it
i
o
n
a
l
m
o
d
e
ls
in
b
o
t
h
a
c
c
u
ra
c
y
a
n
d
c
o
m
p
u
tati
o
n
a
l
e
ffi
c
ien
c
y
,
c
o
m
p
letin
g
p
re
d
ictio
n
s
i
n
u
n
d
e
r
2
se
c
o
n
d
s
p
e
r
d
a
tas
e
t.
Co
m
p
a
re
d
to
c
o
n
v
e
n
ti
o
n
a
l
c
las
sifiers
,
it
re
q
u
ires
n
o
p
a
ra
m
e
ter
tu
n
i
n
g
a
n
d
p
e
rfo
rm
s
re
li
a
b
ly
e
v
e
n
with
sm
a
ll
g
e
n
e
s
e
ts.
Wh
i
le
p
ro
m
isin
g
,
f
u
tu
re
w
o
rk
s
h
o
u
ld
a
d
d
re
ss
m
u
lt
ic
las
s
c
las
sifica
ti
o
n
a
n
d
c
li
n
ica
l
v
a
li
d
a
ti
o
n
t
o
b
ro
a
d
e
n
th
e
fra
m
e
wo
rk
’s ap
p
li
c
a
b
il
i
ty
.
To
g
e
th
e
r,
t
h
e
s
e
m
e
th
o
d
s
o
ffe
r
a
p
re
c
ise
a
n
d
ra
p
id
c
a
n
c
e
r
c
las
sifica
ti
o
n
fra
m
e
wo
rk
,
su
p
p
o
r
ti
n
g
e
a
rly
d
iag
n
o
sis a
n
d
p
e
rso
n
a
li
z
e
d
trea
tme
n
t
stra
teg
ies
a
c
ro
ss
d
i
v
e
rse
c
a
n
c
e
r
ty
p
e
s.
K
ey
w
o
r
d
s
:
C
an
ce
r
class
if
icatio
n
C
o
m
p
u
ter
s
cien
ce
Featu
r
e
s
elec
tio
n
I
m
ag
e
p
r
o
ce
s
s
in
g
Ma
ch
in
e
lear
n
in
g
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
:
Sar
a
Had
d
o
u
B
o
u
az
z
a
L
AM
I
GE
P
L
ab
o
r
ato
r
y
,
E
MSI
Mo
r
o
cc
an
Sch
o
o
l o
f
E
n
g
i
n
ee
r
in
g
Ma
r
r
ak
ec
h
,
Ma
r
o
cc
o
E
m
ail:
s
ar
a.
h
b
.
s
ar
a@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
DNA
m
icr
o
ar
r
ay
tech
n
o
lo
g
y
h
as
g
r
ea
tly
i
m
p
r
o
v
ed
ca
n
ce
r
d
iag
n
o
s
is
an
d
p
r
o
g
n
o
s
is
b
y
al
lo
win
g
th
e
p
ar
allel
ex
am
in
atio
n
o
f
th
o
u
s
an
d
s
o
f
g
en
e
e
x
p
r
ess
io
n
p
r
o
f
i
les
[
1
]
–
[
4
]
.
T
h
is
ad
v
an
ce
m
e
n
t
h
as
en
h
an
ce
d
o
u
r
k
n
o
wled
g
e
o
f
g
e
n
e
in
ter
ac
tio
n
s
an
d
th
eir
co
n
tr
ib
u
tio
n
to
c
an
ce
r
d
ev
elo
p
m
en
t.
Ho
we
v
er
,
a
m
ajo
r
ch
allen
g
e
s
tem
s
f
r
o
m
th
e
im
b
alan
ce
b
etwe
en
th
e
ex
tr
em
ely
lar
g
e
n
u
m
b
er
o
f
g
e
n
es
an
d
th
e
li
m
ited
av
ailab
ilit
y
o
f
s
am
p
les.
Sin
ce
n
o
t
all
g
en
es
ar
e
in
v
o
lv
ed
i
n
ca
n
ce
r
p
r
o
g
r
ess
io
n
an
d
m
an
y
ar
e
c
o
r
r
ela
ted
,
r
ely
in
g
o
n
th
e
co
m
p
lete
g
e
n
e
s
et
m
a
y
in
cr
ea
s
e
co
m
p
lex
ity
an
d
lo
wer
p
r
ed
ictio
n
ac
cu
r
ac
y
[
5
]
.
T
h
i
s
u
n
d
er
s
co
r
es
t
h
e
im
p
o
r
tan
ce
o
f
ap
p
ly
in
g
ef
f
ec
ti
v
e
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
s
t
o
en
h
a
n
ce
class
if
icatio
n
p
er
f
o
r
m
an
ce
.
Mic
r
o
ar
r
ay
g
e
n
e
class
if
icatio
n
,
a
s
u
p
er
v
is
ed
lear
n
in
g
task
,
r
elies
o
n
lab
eled
g
en
e
ex
p
r
ess
io
n
d
ata
to
p
r
ed
ict
d
is
ea
s
e
class
es.
I
ts
s
u
cc
ess
d
ep
en
d
s
h
ea
v
ily
o
n
s
elec
tin
g
th
e
m
o
s
t
r
elev
an
t
f
ea
t
u
r
es
[
6
]
,
[
7
]
.
W
h
ile
tr
ad
itio
n
al
s
tatis
tical
m
eth
o
d
s
h
av
e
b
ee
n
wid
el
y
u
s
ed
[
6
]
,
m
ac
h
in
e
lear
n
in
g
tech
n
iq
u
es
n
o
w
p
lay
a
v
ital
r
o
le
in
h
an
d
lin
g
co
m
p
le
x
m
icr
o
ar
r
ay
d
ata
[
8
]
,
[
9
]
.
Yet,
th
e
h
ig
h
d
im
en
s
io
n
ality
o
f
th
ese
d
at
asets
o
f
ten
lead
s
to
o
v
er
f
itti
n
g
,
em
p
h
asizin
g
t
h
e
i
m
p
o
r
tan
ce
o
f
d
im
en
s
io
n
ality
r
ed
u
ctio
n
th
r
o
u
g
h
f
ea
t
u
r
e
s
elec
tio
n
[
6
]
,
[
1
0
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2252
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
4
,
No
.
6
,
Dec
em
b
er
2
0
2
5
:
4
7
3
1
-
4
7
3
8
4732
Featu
r
e
s
elec
tio
n
aim
s
to
id
en
tify
g
en
es
th
at
s
h
o
w
s
ig
n
if
i
ca
n
t
d
if
f
er
en
ce
s
ac
r
o
s
s
d
is
ea
s
e
class
e
s
.
Ap
p
r
o
ac
h
es
in
cl
u
d
e
f
ilter
m
et
h
o
d
s
,
wh
ich
r
an
k
g
e
n
es
b
ased
o
n
s
tatis
tical
m
ea
s
u
r
es
lik
e
p
-
v
alu
es
[
1
1
]
,
[
1
2
]
,
s
ig
n
al
-
to
-
n
o
is
e
r
atio
(
SNR
)
,
m
R
MR,
R
elief
F,
an
d
p
er
f
o
r
m
an
ce
o
f
u
p
p
er
lim
b
(
PUL
)
s
co
r
es
[
1
3
]
,
an
d
wr
a
p
p
er
m
eth
o
d
s
,
wh
ich
u
s
e
class
if
ier
s
to
ev
alu
ate
g
en
e
s
u
b
s
ets
[
1
4
]
.
W
h
ile
f
ilter
s
ar
e
ef
f
icien
t,
th
ey
o
f
ten
ig
n
o
r
e
g
en
e
in
ter
ac
tio
n
s
,
wh
er
ea
s
wr
ap
p
er
s
o
f
f
er
g
r
ea
ter
ac
cu
r
ac
y
at
a
h
ig
h
er
co
m
p
u
tatio
n
al
co
s
t.
Hy
b
r
id
m
eth
o
d
s
co
m
b
in
e
b
o
th
s
tr
ateg
ies f
o
r
o
p
tim
al
r
esu
lts
.
T
o
ad
d
r
ess
th
e
lim
itatio
n
s
o
f
co
n
v
en
tio
n
al
ap
p
r
o
ac
h
es;
s
u
c
h
as
o
v
er
f
itti
n
g
,
lac
k
o
f
s
ca
la
b
ilit
y
,
an
d
co
m
p
u
tatio
n
al
b
u
r
d
en
;
we
p
r
o
p
o
s
e
a
two
-
s
tep
in
tellig
en
t
f
r
am
ewo
r
k
co
m
b
i
n
in
g
a
h
y
b
r
id
g
en
e
s
elec
tio
n
s
tr
ateg
y
an
d
a
s
tatis
tical
c
lass
if
icatio
n
m
ec
h
an
is
m
.
T
h
e
th
r
ee
-
s
tag
e
g
en
e
s
elec
tio
n
(
3
SGS
)
m
eth
o
d
s
eq
u
en
tially
f
ilter
s
,
ev
alu
ates,
an
d
co
m
p
r
ess
es
g
en
e
s
u
b
s
e
ts
to
en
h
an
ce
p
r
ed
ictiv
e
p
o
wer
wh
ile
r
ed
u
cin
g
d
im
en
s
io
n
ality
.
C
o
m
p
lem
en
ti
n
g
th
is
,
th
e
s
tati
s
tics
clas
s
if
ier
(
SC
)
u
s
e
s
in
ter
p
r
etab
le
s
tati
s
t
ical
b
o
u
n
d
ar
ies
f
o
r
class
if
icatio
n
,
en
ab
lin
g
f
ast an
d
p
r
ec
is
e
d
ec
is
io
n
s
with
m
in
i
m
al
p
ar
am
eter
tu
n
in
g
.
2.
M
E
T
H
O
D
T
h
is
s
tu
d
y
s
ee
k
s
to
d
esig
n
a
r
eliab
le
g
en
e
s
elec
tio
n
ap
p
r
o
ac
h
f
o
r
ac
c
u
r
ate
tu
m
o
r
class
if
icatio
n
b
ased
o
n
m
icr
o
a
r
r
ay
d
ata.
T
h
e
wo
r
k
f
lo
w
co
n
s
is
ts
o
f
s
ev
er
al
s
tag
es:
d
ata
p
r
ep
r
o
ce
s
s
in
g
to
im
p
r
o
v
e
q
u
ality
,
ad
v
an
ce
d
s
elec
tio
n
tech
n
iq
u
es
to
e
x
tr
a
ct
th
e
m
o
s
t
in
f
o
r
m
ativ
e
g
e
n
es,
an
d
th
e
a
p
p
licatio
n
o
f
r
e
f
in
ed
class
if
icatio
n
m
o
d
els.
T
h
e
f
o
llo
win
g
s
ec
tio
n
d
etails
th
e
m
ater
ials
an
d
m
e
th
o
d
s
em
p
l
o
y
ed
,
with
p
ar
ticu
l
ar
em
p
h
asis
o
n
th
e
s
tr
ateg
ies ad
o
p
ted
f
o
r
id
en
tif
y
in
g
r
elev
a
n
t g
en
es.
2
.
1
.
G
ene
s
elec
t
io
n
T
o
id
en
tif
y
g
e
n
es
r
elev
an
t
f
o
r
tu
m
o
r
class
if
icatio
n
f
r
o
m
m
icr
o
ar
r
ay
d
atasets
,
we
ad
o
p
te
d
a
th
r
ee
-
p
h
ase
s
elec
tio
n
s
tr
ateg
y
.
T
h
e
f
ir
s
t
p
h
ase
ap
p
lied
a
f
ilter
in
g
s
tep
to
d
is
ca
r
d
lar
g
ely
ir
r
elev
an
t
g
en
es,
th
er
eb
y
s
im
p
lify
in
g
th
e
d
ataset.
T
h
is
f
ilter
in
g
r
elied
o
n
th
r
ee
p
a
r
a
m
etr
ic
tech
n
iq
u
es
—
SNR
,
co
r
r
elatio
n
co
ef
f
icien
t
(
C
C
)
,
an
d
R
elief
F
—
to
h
ig
h
lig
h
t th
e
m
o
s
t in
f
o
r
m
ativ
e
g
e
n
es
f
o
r
s
u
b
s
eq
u
e
n
t a
n
aly
s
is
.
T
h
e
SNR
tech
n
iq
u
e
f
in
d
s
e
x
p
r
ess
io
n
p
atter
n
s
with
th
e
h
ig
h
est
m
ea
n
e
x
p
r
ess
io
n
d
if
f
er
e
n
c
e
b
etwe
en
two
g
r
o
u
p
s
an
d
th
e
least
f
lu
ctu
atio
n
with
in
ea
c
h
g
r
o
u
p
[
1
5
]
,
[
1
6
]
.
T
h
is
cr
iter
io
n
,
p
r
o
p
o
s
ed
b
y
[
1
7
]
,
r
ates
g
en
es a
cc
o
r
d
i
n
g
to
(
1
)
.
(
)
=
1
−
2
1
+
2
(
1
)
Her
e,
Mk
j
a
n
d
Sk
j
r
ep
r
esen
t
t
h
e
m
ea
n
an
d
s
tan
d
ar
d
d
ev
iatio
n
o
f
g
en
e
j
with
in
class
k
=1
,
2
.
L
ar
g
er
v
alu
es
o
f
∣
P(j)
∣
s
u
g
g
est
a
s
tr
o
n
g
er
ass
o
ciatio
n
b
etwe
en
th
e
g
en
e’
s
ex
p
r
ess
io
n
an
d
class
d
if
f
er
en
tiatio
n
.
T
h
e
Pear
s
o
n
CC
[
1
8
]
ev
al
u
ates
h
o
w
s
tr
o
n
g
ly
t
wo
g
en
es
ar
e
lin
ea
r
ly
r
elate
d
.
Valu
es
clo
s
e
to
+1
in
d
icate
a
d
ir
ec
t
r
elatio
n
s
h
ip
,
th
o
s
e
n
ea
r
-
1
r
e
f
lect
an
in
v
er
s
e
r
elatio
n
s
h
ip
,
an
d
v
al
u
es
ar
o
u
n
d
0
s
u
g
g
est
n
o
lin
ea
r
co
r
r
elatio
n
.
T
h
e
co
ef
f
icien
t f
o
r
g
e
n
e
j is co
m
p
u
ted
as
(
2
)
.
=
∑
(
−
)
(
−
Ῡ
)
=
1
√
∑
(
−
)
²
=
1
√
∑
(
−
Ῡ
)
²
=
1
(
2
)
W
h
er
e
r
is
th
e
Pear
s
o
n
co
r
r
elatio
n
s
co
r
e,
X
ij
is
th
e
i
th
s
am
p
le
v
alu
e
f
o
r
th
e
g
en
e
j,
Yi
is
th
e
co
r
r
esp
o
n
d
in
g
class
,
X
j
=
1
/
∑
=
1
,
an
d
Ῡ
ar
e
th
e
m
ea
n
s
f
o
r
g
en
e
an
d
th
e
class
es,
r
esp
ec
tiv
ely
.
R
elief
F
is
a
s
u
p
e
r
v
is
ed
f
ea
tu
r
e
-
weig
h
tin
g
tech
n
iq
u
e
d
ev
elo
p
e
d
in
[
1
9
]
an
d
f
u
r
th
er
en
h
a
n
ce
d
b
y
[
2
0
]
.
I
t
ass
ess
es
th
e
q
u
ality
o
f
q
u
alities
as
(
3
)
.
W
A
=
w
d
−
∑
d
i
f
f
(
A
i
,
X
i
,
hit
s
j
)
∗
=
1
+
∑
(
)
1
−
(
(
)
)
≠
(
)
∑
(
,
,
)
∗
=
1
(
3
)
W
h
er
e
th
e
d
is
tan
ce
u
s
ed
is
d
ef
in
ed
b
y
(
4
)
.
(
,
1
,
2
)
=
|
(
,
1
)
−
(
,
2
)
|
m
ax
(
)
−
m
i
n
(
)
(
4
)
Her
e,
X
i
is
an
i
n
s
tan
ce
d
escr
i
b
ed
b
y
th
e
v
ec
to
r
A
i
o
f
n
g
en
e
s
,
m
is
th
e
n
u
m
b
e
r
o
f
p
r
o
ce
s
s
r
ep
etitio
n
s
,
k
is
th
e
n
u
m
b
er
o
f
n
ea
r
est
m
is
s
es,
an
d
h
its
an
d
m
is
s
es
r
ef
er
to
n
ea
r
e
s
t
h
it
an
d
m
is
s
in
s
tan
ce
s
,
r
esp
ec
tiv
ely
.
T
h
e
f
ilter
s
elec
tio
n
ap
p
r
o
ac
h
ev
alu
ates
ea
ch
g
en
e
an
d
s
elec
ts
a
s
u
b
s
et
o
f
r
elev
an
t
o
n
es.
Ho
wev
er
,
s
o
m
e
n
o
is
y
g
en
es
m
ay
s
till
r
ed
u
ce
clas
s
if
icatio
n
ac
cu
r
ac
y
[
9
]
.
T
o
ad
d
r
ess
th
is
,
th
e
s
ec
o
n
d
s
tag
e
u
s
es a
wr
ap
p
er
s
tr
ateg
y
,
s
tar
tin
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
A
tw
o
-
s
tep
in
tellig
en
t fra
mew
o
r
k
fo
r
g
en
e
ex
p
r
ess
io
n
-
b
a
s
ed
…
(
S
a
r
a
Ha
d
d
o
u
B
o
u
a
z
z
a
)
4733
with
o
n
e
g
en
e
a
n
d
g
r
ad
u
all
y
ad
d
in
g
o
th
er
s
f
r
o
m
th
e
f
ilter
ed
s
u
b
s
et,
r
etain
in
g
o
n
ly
th
o
s
e
th
at
im
p
r
o
v
e
ac
cu
r
ac
y
.
T
h
e
f
ir
s
t
two
s
tep
s
id
en
tify
th
e
m
o
s
t
in
f
o
r
m
ativ
e
g
en
es.
T
h
e
f
in
al
s
tag
e
r
ef
i
n
es
th
is
s
elec
tio
n
,
ch
o
o
s
in
g
th
e
s
m
allest s
u
b
s
et
t
h
at
ac
h
iev
es th
e
h
ig
h
est ac
cu
r
ac
y
o
n
t
h
e
tr
ain
in
g
s
et.
2
.
2
.
Alg
o
rit
hm
o
f
o
ur
s
elec
t
i
o
n a
pp
ro
a
ch:
t
hree
-
s
t
a
g
e
g
e
ne
s
elec
t
io
n
T
h
e
3
SGS
ap
p
r
o
ac
h
is
in
tr
o
d
u
ce
d
to
im
p
r
o
v
e
b
o
th
t
h
e
ac
c
u
r
ac
y
a
n
d
r
eliab
ilit
y
o
f
g
en
e
s
elec
tio
n
in
tu
m
o
r
class
if
icatio
n
.
I
t
s
y
s
te
m
atica
lly
r
ed
u
ce
s
th
e
d
im
en
s
io
n
ality
o
f
h
ig
h
-
th
r
o
u
g
h
p
u
t
g
en
e
ex
p
r
ess
io
n
d
ata
wh
ile
m
ax
im
izin
g
p
r
e
d
ictiv
e
p
er
f
o
r
m
an
ce
.
T
h
e
al
g
o
r
ith
m
b
eg
in
s
with
a
d
ataset
co
m
p
o
s
ed
o
f
tr
ain
in
g
d
ata
(
X
train
,
Y
train
)
,
co
n
tain
in
g
n
g
e
n
es
ac
r
o
s
s
m
s
am
p
les
with
k
n
o
wn
class
lab
els,
an
d
a
test
d
ataset
X
test
f
ea
tu
r
in
g
th
e
s
am
e
g
en
es
b
u
t
u
n
k
n
o
w
n
lab
els.
T
h
e
u
s
er
also
d
ef
in
es
th
e
d
esire
d
n
u
m
b
er
k
o
f
to
p
-
r
an
k
ed
g
e
n
es
to
b
e
in
itially
s
elec
ted
,
as
wel
l
as
th
e
class
if
ier
to
b
e
em
p
lo
y
ed
(
e.
g
.
,
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
S
VM
)
an
d
k
-
n
ea
r
est
n
eig
h
b
o
r
s
(
KNN
)
).
i)
Step
1
:
f
ea
tu
r
e
s
elec
tio
n
v
ia
f
ilter
-
b
ased
r
an
k
in
g
,
th
e
p
r
o
ce
d
u
r
e
b
eg
i
n
s
with
th
e
ca
lcu
lati
o
n
o
f
r
an
k
in
g
s
co
r
es
f
o
r
ea
ch
g
en
e
in
th
e
tr
ain
in
g
d
ataset,
em
p
l
o
y
in
g
f
i
lter
-
b
ased
m
ea
s
u
r
es
s
u
ch
as
SNR
,
C
C
,
o
r
R
elief
F.
T
h
ese
s
co
r
es
ar
e
s
to
r
ed
in
a
lis
t
ter
m
ed
Gen
e
_
Sco
r
es.
I
f
m
u
ltip
le
m
etr
ics
ar
e
u
tili
ze
d
,
n
o
r
m
aliza
tio
n
is
p
er
f
o
r
m
e
d
t
o
s
ca
le
th
e
s
co
r
es
u
n
if
o
r
m
l
y
b
etwe
en
0
an
d
1
to
en
s
u
r
e
f
air
co
m
p
ar
is
o
n
.
T
h
e
g
en
es
ar
e
th
en
s
o
r
ted
in
d
escen
d
in
g
o
r
d
er
b
ased
o
n
th
ei
r
s
co
r
es,
an
d
th
e
to
p
k
g
en
es
a
r
e
s
elec
ted
to
f
o
r
m
a
n
in
itial su
b
s
et
r
ef
er
r
e
d
to
as T
o
p
_
R
an
k
e
d
_
Gen
es.
ii)
Step
2
:
r
ec
u
r
s
iv
e
s
u
b
s
et
r
ef
i
n
em
en
t
f
o
r
ac
cu
r
ac
y
m
ax
im
i
za
tio
n
,
th
e
n
e
x
t
s
tag
e
in
v
o
lv
es
r
ec
u
r
s
iv
ely
ev
alu
atin
g
g
en
e
s
u
b
s
ets
to
i
d
en
tify
t
h
e
c
o
m
b
in
atio
n
th
at
y
ield
s
th
e
h
ig
h
est
class
if
ica
tio
n
ac
cu
r
ac
y
.
Star
tin
g
f
r
o
m
an
em
p
t
y
s
et
S
={
},
ea
ch
g
en
e
in
T
o
p
_
R
an
k
e
d
_
Gen
es
is
iter
ativ
ely
ad
d
ed
to
a
tem
p
o
r
ar
y
s
et
Stem
p
=S
∪
{g
}.
T
h
e
class
i
f
ier
is
tr
ain
ed
o
n
th
is
s
u
b
s
et,
an
d
p
er
f
o
r
m
an
ce
is
ass
ess
ed
u
s
in
g
cr
o
s
s
-
v
alid
atio
n
.
I
f
th
e
ac
c
u
r
ac
y
i
m
p
r
o
v
es
o
r
r
em
ai
n
s
th
e
s
am
e,
S
is
u
p
d
ated
t
o
in
clu
d
e
g
;
o
th
er
wis
e,
g
is
d
is
ca
r
d
ed
.
T
h
e
p
r
o
ce
d
u
r
e
in
cl
u
d
es
an
ea
r
ly
s
to
p
p
i
n
g
c
r
iter
io
n
to
h
alt
th
e
iter
atio
n
o
n
ce
n
o
ac
cu
r
ac
y
g
ain
is
o
b
s
er
v
ed
o
v
er
s
ev
er
al
iter
at
io
n
s
,
th
er
eb
y
m
itig
atin
g
o
v
er
f
itti
n
g
an
d
r
ed
u
ci
n
g
c
o
m
p
u
tati
o
n
al
co
s
t.
T
h
e
f
in
al
s
u
b
s
et,
r
ef
er
r
ed
to
as
Hig
h
_
Acc
u
r
ac
y
_
Gen
es,
c
o
n
tain
s
th
e
g
en
es
th
at
p
r
o
v
i
d
e
th
e
g
r
ea
test
co
n
tr
ib
u
tio
n
to
class
if
icatio
n
a
cc
u
r
ac
y
.
iii)
Ste
p
3
:
r
ed
u
n
d
a
n
cy
r
ed
u
ctio
n
to
id
en
tify
m
ar
k
er
g
en
es,
to
f
u
r
th
er
r
e
f
in
e
th
e
g
e
n
e
s
et,
r
e
d
u
n
d
an
cy
a
m
o
n
g
g
en
es
in
Hig
h
_
Acc
u
r
ac
y
_
Gen
es
is
an
aly
ze
d
u
s
in
g
co
r
r
elatio
n
o
r
m
u
tu
al
in
f
o
r
m
atio
n
.
Ge
n
es
ex
h
ib
itin
g
h
ig
h
r
ed
u
n
d
an
c
y
o
r
m
i
n
im
al
co
n
tr
ib
u
tio
n
to
ac
c
u
r
ac
y
ar
e
p
r
u
n
ed
.
Op
tio
n
ally
,
p
r
in
cip
al
co
m
p
o
n
en
t
an
aly
s
is
(
PC
A)
m
ay
b
e
u
s
ed
t
o
aid
in
d
etec
tin
g
o
v
e
r
lap
p
in
g
ex
p
r
ess
io
n
p
atter
n
s
.
T
h
e
class
if
ier
is
th
en
r
etr
ain
ed
o
n
t
h
e
r
e
d
u
ce
d
g
en
e
s
et
to
en
s
u
r
e
class
if
icatio
n
p
e
r
f
o
r
m
a
n
ce
is
n
o
t
co
m
p
r
o
m
is
e
d
.
I
f
ac
cu
r
ac
y
d
r
o
p
s
,
p
r
e
v
io
u
s
ly
ex
clu
d
ed
g
en
es
m
ay
b
e
r
ec
o
n
s
id
er
ed
.
T
h
e
f
in
alize
d
,
n
o
n
-
r
e
d
u
n
d
a
n
t,
an
d
h
ig
h
ly
in
f
o
r
m
ativ
e
g
en
es a
r
e
r
etai
n
ed
as M
ar
k
er
_
Gen
es.
iv
)
Step
4
:
f
i
n
al
class
if
icatio
n
u
s
i
n
g
s
elec
ted
m
ar
k
er
g
e
n
es
in
t
h
e
f
in
al
s
tag
e,
th
e
class
if
ier
i
s
r
etr
ain
ed
o
n
th
e
co
m
p
lete
t
r
ain
in
g
d
ataset
u
s
in
g
o
n
ly
th
e
s
elec
ted
Ma
r
k
er
_
Gen
es.
Af
ter
v
alid
atin
g
t
h
e
m
o
d
el
v
ia
cr
o
s
s
-
v
alid
atio
n
to
en
s
u
r
e
g
en
er
aliza
b
ilit
y
,
it
is
ap
p
lied
t
o
th
e
test
d
ata
Xtest.
T
h
e
class
if
ier
p
r
e
d
icts
th
e
class
lab
els
f
o
r
th
e
u
n
s
ee
n
s
am
p
les,
p
r
o
d
u
cin
g
th
e
f
i
n
al
o
u
tp
u
t,
p
r
ed
icted
_
lab
els
,
wh
ich
r
ep
r
esen
t
th
e
ca
n
ce
r
class
p
r
ed
ictio
n
s
b
ased
o
n
a
m
in
im
al
y
et
in
f
o
r
m
ativ
e
g
en
e
s
et.
2
.
3
.
Cla
s
s
if
ica
t
io
n
m
et
ho
ds
W
e
ev
alu
ated
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
s
with
f
iv
e
class
i
f
ier
s
:
KNN,
SVM,
l
in
ea
r
d
is
cr
im
in
an
t
an
aly
s
is
(
L
DA
)
,
d
ec
is
io
n
tr
ee
(
DT
)
,
an
d
n
aiv
e
B
ay
es (
NB
)
.
i)
KNN
class
if
ies s
am
p
les b
ased
o
n
p
r
o
x
im
ity
,
u
s
in
g
E
u
clid
ea
n
d
is
tan
ce
to
id
en
tif
y
th
e
KN
N
[
2
1
]
,
[
2
2
]
.
ii)
SVM
co
n
s
tr
u
cts
a
m
ax
im
u
m
-
m
ar
g
in
h
y
p
er
p
lan
e
in
a
h
i
g
h
-
d
im
en
s
io
n
al
s
p
ac
e
v
ia
k
er
n
el
f
u
n
ctio
n
s
,
o
p
tim
izin
g
s
ep
ar
atio
n
b
etwe
en
class
es [
2
3
]
.
iii)
L
DA
id
en
tifie
s
a
lin
ea
r
co
m
b
in
atio
n
o
f
f
ea
t
u
r
es
th
at
im
p
r
o
v
es
clas
s
s
ep
ar
atio
n
b
y
m
ax
im
i
zin
g
b
etwe
en
-
class
v
ar
ian
ce
wh
ile
m
in
im
izin
g
with
in
-
class
v
ar
ian
ce
[
2
4
]
.
iv
)
DT
u
s
es
a
h
ier
ar
ch
ical
s
tr
u
ct
u
r
e
o
f
d
ec
is
io
n
r
u
les
to
m
o
d
e
l
o
u
tco
m
es,
m
ak
i
n
g
it
wid
ely
ap
p
licab
le
in
m
ac
h
in
e
lear
n
in
g
an
d
d
ata
m
i
n
in
g
[
2
5
]
,
[
2
6
]
.
v)
NB
is
a
p
r
o
b
ab
ilis
tic
m
o
d
el
b
ased
o
n
B
ay
es’
th
eo
r
em
,
ass
u
m
in
g
f
ea
tu
r
e
in
d
ep
en
d
en
ce
g
i
v
en
th
e
class
,
an
d
is
k
n
o
w
n
f
o
r
its
s
im
p
licity
an
d
ef
f
icien
cy
[
2
5
]
.
T
o
ass
ess
cla
s
s
if
ier
p
er
f
o
r
m
an
ce
,
we
u
s
e
class
if
icatio
n
ac
cu
r
ac
y
[
2
6
]
–
[
2
7
]
.
2
.
4
.
O
ur
pro
po
s
it
io
n f
o
r
g
en
e
cla
s
s
if
ica
t
io
n f
o
r
bin
a
ry
cla
s
s
pro
blem
s
T
h
e
SC
is
b
ased
o
n
lev
er
ag
in
g
s
tatis
tical
d
escr
ip
to
r
s
s
u
ch
as
m
in
im
u
m
,
m
ax
im
u
m
,
m
ea
n
,
an
d
s
tan
d
ar
d
d
e
v
iatio
n
o
f
g
en
e
e
x
p
r
ess
io
n
to
ass
ess
class
m
em
b
er
s
h
ip
.
Ou
r
n
o
v
el
g
en
e
class
if
icatio
n
ap
p
r
o
ac
h
,
b
ased
o
n
g
en
e
ex
p
r
ess
io
n
p
r
o
f
ilin
g
d
ata,
i
n
tr
o
d
u
ce
s
a
s
tr
ea
m
lin
ed
two
-
s
tep
p
r
o
ce
s
s
d
esig
n
ed
f
o
r
clar
ity
an
d
p
r
ec
is
io
n
.
T
h
e
f
ir
s
t
s
tep
in
v
o
lv
es
ca
lcu
latin
g
k
ey
s
tatis
tical
m
ea
s
u
r
es
f
o
r
ea
ch
s
elec
ted
g
en
e
with
in
th
e
tr
ain
in
g
s
am
p
les
ac
r
o
s
s
b
o
th
class
es.
T
h
ese
m
ea
s
u
r
es
co
m
p
r
is
e
th
e
m
in
im
u
m
(
m
in
)
,
m
a
x
im
u
m
(
m
ax
)
,
m
ea
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2252
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
4
,
No
.
6
,
Dec
em
b
er
2
0
2
5
:
4
7
3
1
-
4
7
3
8
4734
(
m
ea
n
)
,
an
d
s
tan
d
ar
d
d
e
v
iatio
n
(
Std
)
o
f
g
en
e
ex
p
r
ess
io
n
v
al
u
es,
wh
ich
t
o
g
eth
er
d
ef
i
n
e
th
e
ex
p
r
ess
io
n
p
r
o
f
ile
o
f
ea
ch
g
en
e
in
its
co
r
r
esp
o
n
d
in
g
class
.
I
n
th
e
s
ec
o
n
d
s
tep
,
test
s
am
p
les
ar
e
class
if
ied
b
y
c
o
m
p
ar
i
n
g
th
ei
r
g
en
e
ex
p
r
ess
io
n
lev
e
ls
ag
ain
s
t
th
ese
s
tatis
tica
l
r
an
g
es.
I
f
a
te
s
t
s
am
p
le’
s
ex
p
r
ess
io
n
v
alu
e
f
o
r
a
g
iv
en
g
en
e
f
alls
with
in
th
e
r
an
g
e
d
ef
in
e
d
b
y
m
in
an
d
m
ax
f
o
r
a
s
p
ec
if
ic
class
,
th
e
s
am
p
le
i
s
d
ir
ec
tly
ass
ig
n
ed
to
th
at
class
f
o
r
th
at
g
en
e.
W
h
en
th
e
ex
p
r
ess
io
n
v
alu
e
lies
o
u
ts
id
e
th
ese
b
o
u
n
d
ar
ies,
th
e
s
am
p
le
is
ass
ig
n
ed
to
th
e
class
wh
o
s
e
r
an
g
e
—
s
p
ec
if
ically
th
e
m
ea
n
±
Std
in
ter
v
al
is
clo
s
est
to
th
e
tes
t
v
alu
e,
en
s
u
r
in
g
a
cc
u
r
ate
class
if
icatio
n
ev
en
in
ca
s
es
o
f
o
u
tlier
s
o
r
v
ar
iab
ilit
y
.
T
o
f
in
alize
th
e
class
if
icatio
n
,
a
v
o
tin
g
m
ec
h
an
is
m
is
ap
p
lied
.
E
ac
h
s
elec
ted
g
en
e
ca
s
ts
a
"v
o
te"
b
ased
o
n
its
class
if
icat
io
n
r
esu
lt.
T
h
e
o
v
er
all
class
ass
ig
n
ed
to
th
e
s
am
p
le
is
d
eter
m
in
ed
b
y
th
e
m
ajo
r
ity
o
f
v
o
tes,
wh
ich
h
elp
s
b
ala
n
ce
in
d
iv
id
u
al
g
en
e
-
lev
el
v
ar
ian
ce
s
a
n
d
l
ea
d
s
to
a
m
o
r
e
r
eliab
le
p
r
e
d
i
ctio
n
.
T
h
is
p
r
o
ce
s
s
o
f
f
er
s
a
s
y
s
tem
atic
m
eth
o
d
f
o
r
g
en
e
-
b
ased
class
if
icatio
n
b
y
l
ev
er
ag
in
g
s
tatis
tical
b
o
u
n
d
a
r
ie
s
an
d
a
co
n
s
en
s
u
s
-
d
r
iv
en
v
o
tin
g
s
tr
ateg
y
,
en
s
u
r
i
n
g
b
o
th
r
o
b
u
s
tn
ess
an
d
in
ter
p
r
e
tab
ilit
y
.
2
.
5
.
Da
t
a
s
et
s
T
h
is
s
tu
d
y
ev
alu
ates
th
e
p
r
o
p
o
s
ed
m
eth
o
d
u
s
in
g
th
r
ee
p
u
b
licly
av
ailab
le
b
in
ar
y
-
class
m
icr
o
ar
r
ay
d
atasets
:
leu
k
em
ia,
p
r
o
s
tate
ca
n
ce
r
,
an
d
co
l
o
n
ca
n
c
er
.
E
ac
h
d
ataset
co
n
s
is
ts
o
f
g
en
e
ex
p
r
ess
io
n
p
r
o
f
iles
r
ep
r
esen
ted
as m
atr
ices,
with
r
o
ws co
r
r
esp
o
n
d
in
g
to
s
am
p
les an
d
co
l
u
m
n
s
to
g
e
n
e
f
ea
tu
r
es.
i)
T
h
e
leu
k
em
ia
d
ataset
co
n
s
is
ts
o
f
7
2
s
am
p
les,
with
3
8
u
s
ed
f
o
r
tr
ain
i
n
g
a
n
d
3
4
f
o
r
test
in
g
,
ca
teg
o
r
ized
in
to
ac
u
te
l
y
m
p
h
o
b
last
ic
leu
k
em
ia
(
AL
L
)
a
n
d
ac
u
te
m
y
elo
id
leu
k
em
ia
(
AM
L
)
.
E
ac
h
s
a
m
p
le
co
n
tain
s
7
,
1
2
9
g
en
e
ex
p
r
ess
io
n
f
ea
t
u
r
e
s
[
2
8
]
.
ii)
T
h
e
p
r
o
s
tate
can
ce
r
d
ataset
co
m
p
r
is
es
1
0
1
s
am
p
les
(
8
1
tr
ain
in
g
an
d
2
0
test
in
g
)
,
in
clu
d
in
g
5
2
tu
m
o
r
an
d
4
9
n
o
n
-
tu
m
o
r
ca
s
es.
Gen
e
ex
p
r
ess
io
n
was
m
ea
s
u
r
ed
u
s
in
g
o
lig
o
n
u
cleo
tid
e
m
icr
o
ar
r
ay
s
co
v
er
in
g
ap
p
r
o
x
im
ately
1
2
,
6
0
0
g
en
es [
2
9
]
.
iii)
T
h
e
co
lo
n
ca
n
ce
r
d
ataset
co
n
tain
s
6
2
s
am
p
les
(
4
8
tr
ain
in
g
a
n
d
1
4
test
in
g
)
,
with
4
0
t
u
m
o
r
an
d
2
2
n
o
r
m
a
l
tis
s
u
e
s
am
p
les.
Gen
e
ex
p
r
ess
i
o
n
was
r
ec
o
r
d
e
d
u
s
in
g
an
Af
f
y
m
etr
ix
o
lig
o
n
u
cleo
tid
e
a
r
r
ay
,
f
r
o
m
wh
ic
h
2
,
0
0
0
g
en
es we
r
e
s
elec
ted
b
as
ed
o
n
m
ea
s
u
r
em
en
t r
eliab
ilit
y
[
3
0
]
.
T
h
ese
d
atasets
p
r
o
v
id
e
a
r
eliab
le
b
en
ch
m
ar
k
f
o
r
ass
ess
in
g
th
e
g
e
n
er
aliza
b
ilit
y
an
d
p
e
r
f
o
r
m
a
n
ce
o
f
g
en
e
s
elec
tio
n
an
d
class
if
icatio
n
s
tr
ateg
ies in
ca
n
ce
r
d
iag
n
o
s
is
.
3.
E
XP
E
R
I
M
E
N
T
A
L
RE
SUL
T
S AN
D
D
I
SC
USS
I
O
N
I
n
th
is
s
ec
tio
n
,
we
o
u
tlin
e
th
e
o
u
tco
m
es
o
f
o
u
r
r
esear
c
h
,
co
v
er
in
g
ea
ch
s
tep
f
r
o
m
d
ata
p
r
ep
r
o
ce
s
s
in
g
to
th
e
im
p
lem
en
tatio
n
o
f
o
u
r
g
en
e
s
elec
tio
n
s
tr
ateg
y
an
d
th
e
f
in
al
class
if
icatio
n
.
W
e
ev
al
u
ated
o
u
r
m
eth
o
d
ac
r
o
s
s
m
u
ltip
le
d
atasets
to
ass
ess
its
ef
f
ec
tiv
en
ess
an
d
r
o
b
u
s
tn
ess
in
id
en
tify
in
g
th
e
m
o
s
t
s
ig
n
if
ican
t
g
e
n
es
f
o
r
tu
m
o
r
class
if
icatio
n
.
Ad
d
itio
n
ally
,
we
o
u
tlin
e
th
e
to
o
ls
an
d
tech
n
iq
u
es
em
p
lo
y
ed
an
d
p
r
o
v
id
e
an
in
-
d
ep
th
an
aly
s
is
o
f
th
e
class
if
icatio
n
r
esu
lts
o
b
tain
ed
u
s
in
g
v
a
r
io
u
s
c
lass
if
ier
s
.
3
.
1
.
Da
t
a
prepro
ce
s
s
ing
Pre
tr
ea
tm
en
t
f
ilter
s
o
u
t
n
o
n
-
i
n
f
o
r
m
ativ
e
g
e
n
es
with
co
n
s
is
t
en
t
ex
p
r
ess
io
n
ac
r
o
s
s
class
e
s
.
L
eu
k
em
ia
d
ata
is
ex
clu
d
ed
f
r
o
m
th
is
p
r
o
ce
s
s
.
T
h
e
m
eth
o
d
in
v
o
lv
es
th
r
esh
o
ld
in
g
,
f
ilter
in
g
,
an
d
lo
g
ar
ith
m
ic
tr
an
s
f
o
r
m
atio
n
[
3
1
]
.
T
h
r
esh
o
l
d
in
g
k
ee
p
s
v
alu
es
b
etwe
en
1
0
0
an
d
1
6
,
0
0
0
.
Fil
ter
in
g
r
em
o
v
es
g
en
es
with
lo
w
v
ar
iab
ilit
y
,
r
etain
i
n
g
th
o
s
e
w
h
er
e
th
e
r
atio
Sm
ax
/Smin
ex
c
ee
d
s
5
an
d
th
e
d
if
f
e
r
en
ce
Sm
ax
−Sm
in
is
g
r
ea
ter
th
an
5
0
0
.
A
lo
g
ar
ith
m
ic
tr
an
s
f
o
r
m
atio
n
th
e
n
n
o
r
m
alize
s
th
e
d
ata.
T
h
is
p
r
o
ce
s
s
r
ed
u
ce
s
th
e
d
ataset
f
r
o
m
7
,
1
2
9
to
3
,
0
5
1
g
e
n
es,
k
ee
p
in
g
th
e
m
o
s
t in
f
o
r
m
ativ
e
f
ea
tu
r
es f
o
r
an
aly
s
is
.
3
.
2
.
P
er
f
o
r
m
a
nce
a
na
ly
s
is
o
f
t
he
t
hree
-
s
t
a
g
e
g
ene
s
elec
t
io
n m
et
ho
d
T
h
e
p
r
o
p
o
s
ed
3
SGS
m
eth
o
d
was
ev
alu
ated
o
n
leu
k
em
ia,
p
r
o
s
tate,
a
n
d
c
o
lo
n
ca
n
ce
r
m
icr
o
ar
r
ay
d
atasets
.
I
n
ea
ch
ca
s
e,
tr
ain
i
n
g
d
atasets
wer
e
ap
p
lied
f
o
r
g
en
e
s
elec
tio
n
an
d
m
o
d
el
c
o
n
s
tr
u
ctio
n
,
an
d
test
d
atasets
a
s
s
es
s
ed
clas
s
if
icatio
n
p
er
f
o
r
m
an
ce
u
s
in
g
f
iv
e
class
if
ier
s
.
T
ab
le
1
p
r
esen
ts
th
e
c
lass
if
icatio
n
r
esu
lt
s
o
n
th
e
leu
k
em
ia
d
ataset
u
s
in
g
f
iv
e
class
if
ier
s
,
with
d
if
f
er
en
t
f
ea
tu
r
e
s
elec
tio
n
s
tr
ateg
ies.
T
h
e
3
SGS
-
en
h
an
ce
d
v
er
s
io
n
s
o
f
SNR
,
C
C
,
an
d
R
elief
F
co
n
s
is
ten
tly
ac
h
iev
ed
p
er
f
ec
t
ac
cu
r
ac
y
(
1
0
0
%)
u
s
in
g
as
f
ew
as
th
r
ee
to
f
o
u
r
g
en
es.
T
h
is
h
ig
h
lig
h
ts
t
h
e
ab
ilit
y
o
f
3
SGS
to
r
ed
u
ce
d
im
en
s
io
n
ality
w
h
ile
m
ain
ta
in
in
g
o
r
im
p
r
o
v
i
n
g
p
r
ed
ictiv
e
p
e
r
f
o
r
m
an
ce
.
T
h
e
m
o
s
t
r
elev
an
t
g
en
es
id
en
tifie
d
wer
e
Y0
0
7
8
7
,
M2
3
1
9
7
,
an
d
M2
7
8
9
1
ef
f
ec
tiv
el
y
d
if
f
er
en
tiatin
g
b
etwe
en
AL
L
a
n
d
AM
L
s
u
b
ty
p
es.
As
s
h
o
wn
in
T
ab
le
2
,
th
e
3
S
GS
m
eth
o
d
s
ig
n
if
ican
tly
im
p
r
o
v
ed
class
if
icatio
n
ac
cu
r
ac
y
ac
r
o
s
s
all
class
if
ier
s
o
n
th
e
p
r
o
s
tate
ca
n
ce
r
d
ataset.
I
t
ac
h
iev
ed
u
p
t
o
9
5
%
ac
cu
r
ac
y
u
s
in
g
o
n
ly
t
h
r
ee
to
f
o
u
r
g
en
es,
co
m
p
ar
ed
to
s
tan
d
ar
d
m
et
h
o
d
s
r
eq
u
ir
in
g
m
an
y
m
o
r
e
f
ea
tu
r
es.
T
h
e
k
ey
g
e
n
es
s
elec
ted
—
3
7
7
2
0
_
at,
3
7
6
3
9
_
at,
an
d
4
0
4
3
5
_
at
—
en
ab
led
r
o
b
u
s
t d
is
cr
im
in
atio
n
b
etwe
en
t
u
m
o
r
an
d
n
o
r
m
al
s
am
p
les.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
A
tw
o
-
s
tep
in
tellig
en
t fra
mew
o
r
k
fo
r
g
en
e
ex
p
r
ess
io
n
-
b
a
s
ed
…
(
S
a
r
a
Ha
d
d
o
u
B
o
u
a
z
z
a
)
4735
T
ab
le
3
h
ig
h
lig
h
ts
th
e
p
er
f
o
r
m
an
ce
o
n
th
e
co
lo
n
ca
n
ce
r
d
ataset.
On
ce
ag
ain
,
th
e
3
S
GS
m
eth
o
d
d
em
o
n
s
tr
ated
its
ef
f
ec
tiv
en
ess
b
y
en
h
an
ci
n
g
ac
cu
r
ac
y
wh
ile
s
im
u
ltan
eo
u
s
ly
r
ed
u
cin
g
t
h
e
n
u
m
b
er
o
f
s
elec
ted
g
en
es.
No
tab
ly
,
th
e
KNN
class
if
ier
r
ea
ch
ed
9
6
%
ac
cu
r
ac
y
u
s
in
g
o
n
ly
f
o
u
r
g
en
es
v
ia
SNR
_
3
SGS,
co
m
p
ar
ed
to
9
2
.
8
%
with
th
e
s
tan
d
ar
d
SNR
m
eth
o
d
.
T
h
e
s
elec
ted
g
en
es;
M6
3
3
9
1
,
H6
4
4
8
9
,
T
9
2
4
5
1
,
an
d
T
5
7
6
1
9
;
en
ab
led
p
r
ec
is
e
d
if
f
er
en
tiatio
n
b
etwe
en
ca
n
ce
r
o
u
s
an
d
n
o
r
m
al
tis
s
u
e
s
am
p
les.
T
ab
le
1
.
L
e
u
k
em
ia
d
ataset
–
p
e
r
f
o
r
m
a
n
ce
o
f
class
if
ier
s
with
g
en
e
s
elec
tio
n
ap
p
r
o
ac
h
es
F
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
met
h
o
d
s
K
N
N
S
V
M
LD
A
DT
NB
S
N
R
1
0
0
%
(
1
3
)
9
7
%
(
4
)
9
7
%
(
9
)
9
7
%
(
3
)
9
7
%
(
5
)
S
N
R
_
3
S
G
S
1
0
0
%
(
3
)
9
7
%
(
2
)
9
7
%
(
4
)
9
7
%
(
3
)
9
7
%
(
4
)
CC
1
0
0
%
(
5
0
)
9
7
%
(
3
)
1
0
0
%
(
9
3
)
9
7
%
(
4
)
9
7
%
(
6
)
C
C
_
3
S
G
S
1
0
0
%
(
4
)
9
7
%
(
3
)
1
0
0
%
(
5
)
1
0
0
%
(
4
)
1
0
0
%
(
4
)
R
e
l
i
e
f
F
9
7
%
(
4
1
)
9
7
%
(
2
)
9
7
%
(
6
9
)
9
4
%
(
1
1
)
9
4
%
(
5
)
R
e
l
i
e
f
F
_
3
S
G
S
1
0
0
%
(
4
)
9
7
%
(
1
)
1
0
0
%
(
4
)
9
7
%
(
4
)
9
7
%
(
4
)
T
ab
le
2
.
Pro
s
tate
ca
n
ce
r
d
ataset
–
p
er
f
o
r
m
an
ce
o
f
class
if
ier
s
with
g
en
e
s
elec
tio
n
ap
p
r
o
ac
h
e
s
F
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
met
h
o
d
s
K
N
N
S
V
M
LD
A
DT
NB
S
N
R
9
0
%
(
2
2
)
9
2
%
(
8
)
9
2
%
(
4
)
9
1
%
(
1
9
)
9
1
%
(
4
5
)
S
N
R
_
3
S
G
S
9
5
%
(
3
)
9
5
%
(
2
)
9
2
%
(
1
)
9
2
%
(
3
)
9
2
%
(
4
)
CC
8
5
%
(
6
)
9
2
%
(
4
4
)
9
2
%
(
6
)
9
2
%
(
4
6
)
9
1
%
(
6
5
)
C
C
_
3
S
G
S
9
2
%
(
4
)
9
5
%
(
3
)
9
5
%
(
3
)
9
5
%
(
4
)
9
2
%
(
3
)
R
e
l
i
e
f
F
9
0
%
(
3
2
)
9
2
%
(
3
4
)
9
1
%
(
7
5
)
9
0
%
(
3
6
)
9
2
%
(
5
0
)
R
e
l
i
e
f
F
_
3
S
G
S
9
5
%
(
3
)
9
5
%
(
3
)
9
1
%
(
1
)
9
1
%
(
4
)
9
5
%
(
4
)
T
ab
le
3
.
C
o
lo
n
ca
n
ce
r
d
ataset
–
p
er
f
o
r
m
an
ce
o
f
class
if
ier
s
with
g
en
e
s
elec
tio
n
ap
p
r
o
ac
h
es
F
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
met
h
o
d
s
K
N
N
S
V
M
LD
A
DT
NB
S
N
R
9
2
.
8
%
(
5
)
8
5
.
7
%
(
2
9
)
9
2
.
8
%
(
2
)
9
1
%
(
2
1
)
8
5
.
7
%
(
2
2
)
S
N
R
_
3
S
G
S
9
6
%
(
4
)
9
2
.
8
%
(
9
)
9
4
%
(
5
)
9
2
.
8
%
(
6
)
9
1
%
(
6
)
CC
9
2
.
8
%
(
7
)
8
5
.
7
%
(
2
)
9
2
.
8
%
(
2
7
)
9
2
.
8
%
(
2
1
)
8
5
.
7
%
(
5
)
C
C
_
3
S
G
S
9
6
%
(
5
)
9
5
%
(
4
)
9
5
%
(
4
)
9
5
%
(
5
)
9
1
%
(
4
)
R
e
l
i
e
f
F
8
5
.
7
%
(
4
0
)
8
5
.
7
%
(
1
1
)
7
8
.
5
%
(
7
8
)
9
0
%
(
2
6
)
8
5
.
7
%
(
6
4
)
R
e
l
i
e
f
F
_
3
S
G
S
9
1
%
(
5
)
9
2
.
8
%
(
4
)
9
1
%
(
4
)
9
4
%
(
5
)
9
2
.
8
%
(
5
)
3
.
3
.
R
esu
lt
s
o
f
t
he
pro
po
s
ed
s
t
a
t
is
t
ics cla
s
s
if
ier
T
h
is
s
u
b
s
ec
tio
n
co
m
p
ar
es
th
e
p
r
o
p
o
s
ed
SC
with
f
iv
e
co
n
v
en
tio
n
al
class
if
ier
s
,
o
n
leu
k
em
ia,
p
r
o
s
tate,
an
d
co
l
o
n
ca
n
ce
r
d
atasets
,
co
n
s
id
er
in
g
b
o
th
ac
c
u
r
ac
y
an
d
co
m
p
u
tatio
n
tim
e
All
m
o
d
els
wer
e
tr
ain
ed
u
s
in
g
g
en
es
s
elec
ted
v
ia
th
e
SNR
-
b
ased
SNR
_
3
SGS
m
eth
o
d
.
Fo
r
leu
k
em
ia
,
SC
ac
h
iev
ed
1
0
0
%
ac
cu
r
ac
y
u
s
in
g
th
r
ee
g
e
n
es,
m
atch
in
g
KNN
b
u
t
with
th
e
s
h
o
r
test
r
u
n
tim
e
(
1
.
9
s
ec
o
n
d
s
)
.
I
n
p
r
o
s
tate
ca
n
c
er
,
SC
r
ea
ch
ed
th
e
h
ig
h
est
ac
cu
r
ac
y
(
9
9
.
3
%)
with
th
e
s
am
e
m
in
im
al
tim
e,
o
u
tp
e
r
f
o
r
m
in
g
o
th
e
r
s
(
9
2
–
9
5
%).
Fo
r
co
lo
n
ca
n
ce
r
,
SC
attain
ed
9
7
%
ac
c
u
r
ac
y
,
ag
ai
n
s
u
r
p
ass
in
g
t
r
ad
itio
n
al
clas
s
if
ier
s
in
b
o
th
ac
cu
r
ac
y
a
n
d
s
p
ee
d
.
Acr
o
s
s
all
d
atasets
,
S
C
co
n
s
i
s
ten
tly
d
eliv
er
ed
to
p
p
er
f
o
r
m
an
ce
with
s
ig
n
if
ican
tly
lo
wer
co
m
p
u
tatio
n
al
co
s
t.
C
las
s
if
icatio
n
r
esu
lts
an
d
tim
in
g
ar
e
s
u
m
m
ar
ized
in
T
ab
le
4
.
T
ab
le
4
.
R
u
n
tim
e
f
o
r
ca
n
ce
r
cl
ass
if
icatio
n
S
e
l
e
c
t
i
o
n
m
e
t
h
o
d
K
N
N
S
V
M
LD
A
DT
NB
SC
Le
u
k
e
mi
a
S
N
R
_
3
S
G
S
(
%)
1
0
0
97
97
97
97
1
0
0
r
u
n
t
i
me
(
s)
2
.
3
2
.
4
3
.
1
3
.
3
2
.
7
1
.
9
P
r
o
st
a
t
e
c
a
n
c
e
r
S
N
R
_
3
S
G
S
(
%)
95
95
95
92
95
9
9
.
3
r
u
n
t
i
me
(
s)
2
.
3
2
.
4
3
.
1
3
.
3
2
.
7
1
.
9
C
o
l
o
n
c
a
n
c
e
r
S
N
R
_
3
S
G
S
(
%)
96
9
2
.
8
94
94
94
97
r
u
n
t
i
me
(
s)
2
.
3
2
.
5
3
.
1
3
.
4
2
.
7
1
.
9
3
.
4
.
Dis
cus
s
io
n
R
ec
en
t
s
tu
d
ies
h
av
e
in
tr
o
d
u
ce
d
h
y
b
r
id
g
e
n
e
s
elec
tio
n
s
tr
ate
g
ies
f
o
r
ca
n
ce
r
class
if
icatio
n
,
ac
h
iev
in
g
h
ig
h
ac
c
u
r
ac
y
with
s
m
all
g
e
n
e
s
u
b
s
ets.
Fo
r
in
s
tan
ce
,
g
e
n
etic
alg
o
r
ith
m
(
GA
)
–
I
s
o
m
a
p
r
ea
ch
e
d
1
0
0
%
in
leu
k
em
ia
(
4
3
g
en
es)
a
n
d
8
5
.
8
% in
co
lo
n
ca
n
ce
r
(
1
1
g
en
es)
[
3
1
]
,
e
x
tr
em
e
g
r
ad
ien
t b
o
o
s
tin
g
(
XGBo
o
s
t
)
–
m
u
lti
-
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2252
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
4
,
No
.
6
,
Dec
em
b
er
2
0
2
5
:
4
7
3
1
-
4
7
3
8
4736
o
b
jectiv
e
g
e
n
etic
alg
o
r
ith
m
(
MO
GA
)
o
b
tain
ed
1
0
0
%
i
n
l
eu
k
em
ia
(
7
g
e
n
es)
an
d
9
0
.
2
%
in
co
lo
n
ca
n
ce
r
(
6
2
g
e
n
es)
[
3
2
]
,
a
h
ier
ar
ch
ical
f
u
zz
y
–
a
n
aly
tic
h
ier
ar
ch
y
p
r
o
c
ess
(
AHP
)
ap
p
r
o
ac
h
ac
h
iev
ed
1
0
0
%
in
leu
k
em
ia
(
1
5
g
en
es)
an
d
9
6
%
in
p
r
o
s
t
ate
ca
n
ce
r
(
3
0
g
e
n
es)
[
3
3
]
,
w
h
ile
an
e
n
tr
o
p
y
-
b
ased
m
eth
o
d
r
ep
o
r
ted
1
0
0
%
in
leu
k
em
ia
(
1
0
g
en
es)
a
n
d
9
1
.
9
%
in
co
lo
n
ca
n
ce
r
(
9
g
e
n
es)
[
3
4
]
.
Alth
o
u
g
h
ef
f
ec
tiv
e,
t
h
ese
m
eth
o
d
s
ar
e
o
f
ten
co
m
p
u
tatio
n
ally
in
ten
s
iv
e,
p
a
r
am
eter
-
s
en
s
itiv
e,
an
d
less
g
en
er
aliza
b
le.
T
o
o
v
e
r
co
m
e
th
ese
c
h
allen
g
es,
we
p
r
o
p
o
s
ed
th
e
3
SGS
m
eth
o
d
a
n
d
th
e
SC
.
T
h
e
3
S
GS
m
eth
o
d
co
m
b
in
es
f
ilter
-
b
ased
r
a
n
k
in
g
(
SNR
,
C
C
,
an
d
R
elief
F),
r
ec
u
r
s
iv
e
ev
alu
atio
n
,
an
d
r
ed
u
n
d
an
c
y
r
ed
u
ctio
n
,
b
alan
cin
g
th
e
ef
f
icien
c
y
o
f
f
il
ter
s
with
th
e
ac
cu
r
ac
y
o
f
wr
a
p
p
er
s
.
T
h
e
SC
class
if
ier
ap
p
li
es
s
im
p
le
s
tatis
t
ical
b
o
u
n
d
ar
ies
(
m
i
n
,
m
a
x
,
m
ea
n
,
an
d
Std
)
with
a
v
o
tin
g
m
ec
h
an
is
m
,
en
a
b
lin
g
f
ast,
in
ter
p
r
etab
le,
an
d
r
o
b
u
s
t
p
r
ed
ictio
n
s
with
o
u
t
p
ar
am
eter
tu
n
in
g
,
m
ak
in
g
it su
itab
le
f
o
r
clin
ical
s
ettin
g
s
.
E
x
p
er
im
en
ts
co
n
f
ir
m
ed
th
e
f
r
am
ewo
r
k
’
s
ef
f
ec
tiv
e
n
ess
:
3
S
GS
ac
h
iev
ed
1
0
0
%
ac
cu
r
ac
y
in
leu
k
em
ia
with
th
r
ee
g
en
es
(
M2
7
8
9
1
,
M2
3
1
9
7
,
a
n
d
Y0
0
7
8
7
)
,
9
5
%
in
p
r
o
s
tate
ca
n
ce
r
(
3
7
7
2
0
_
at,
3
7
6
3
9
_
at,
an
d
4
0
4
3
5
_
at)
,
an
d
9
6
%
i
n
co
lo
n
ca
n
ce
r
(
M6
3
3
9
1
,
H6
4
4
8
9
,
T
9
2
4
5
1
,
a
n
d
T
5
7
6
1
9
)
.
T
h
e
SC
f
u
r
th
e
r
im
p
r
o
v
e
d
r
esu
lts
to
9
9
.
3
%
in
p
r
o
s
tate
a
n
d
9
7
%
in
co
lo
n
ca
n
ce
r
,
with
r
u
n
tim
es
as
lo
w
as
1
.
9
s
ec
o
n
d
s
.
T
h
ese
o
u
tco
m
es
d
em
o
n
s
tr
ate
th
at
th
e
3
SGS
–
S
C
f
r
am
ewo
r
k
o
f
f
er
s
p
r
ec
is
io
n
,
ef
f
icien
cy
,
an
d
in
ter
p
r
etab
ili
ty
,
s
h
o
win
g
s
tr
o
n
g
p
o
ten
tial f
o
r
p
er
s
o
n
alize
d
ca
n
ce
r
d
iag
n
o
s
is
an
d
clin
ical
d
ec
i
s
io
n
s
u
p
p
o
r
t.
4.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
p
r
o
p
o
s
ed
a
tw
o
-
s
tep
in
tellig
en
t
f
r
am
ewo
r
k
f
o
r
g
en
e
ex
p
r
ess
io
n
-
b
ased
ca
n
ce
r
class
if
icatio
n
,
in
teg
r
atin
g
th
e
3
SGS
m
eth
o
d
an
d
t
h
e
SC
.
T
h
e
3
SGS
ap
p
r
o
ac
h
ef
f
icien
tly
r
e
d
u
ce
d
d
im
en
s
io
n
ality
b
y
f
ilter
in
g
ir
r
elev
an
t
an
d
r
e
d
u
n
d
an
t
g
en
es
wh
ile
r
etain
in
g
th
e
m
o
s
t
in
f
o
r
m
ativ
e
o
n
es,
an
d
th
e
SC
clas
s
if
ier
co
m
p
lem
en
ted
th
is
b
y
ap
p
ly
i
n
g
s
im
p
le
s
tatis
tical
m
ea
s
u
r
es
(
m
in
,
m
ax
,
m
ea
n
,
an
d
Std
)
to
ac
h
iev
e
r
o
b
u
s
t,
in
ter
p
r
etab
le,
an
d
co
m
p
u
tatio
n
ally
ef
f
icien
t c
lass
if
icatio
n
.
E
x
p
er
im
en
ts
o
n
leu
k
em
ia,
p
r
o
s
tate,
an
d
co
lo
n
ca
n
ce
r
d
atasets
d
em
o
n
s
tr
ated
th
e
ef
f
ec
tiv
en
ess
o
f
th
e
f
r
a
m
ewo
r
k
,
with
h
ig
h
ac
cu
r
ac
y
,
m
i
n
im
al
g
en
e
s
u
b
s
ets,
a
n
d
r
ed
u
ce
d
r
u
n
tim
e,
co
n
f
ir
m
in
g
its
p
o
te
n
tial
f
o
r
r
eliab
le
ea
r
l
y
ca
n
ce
r
d
iag
n
o
s
is
.
No
n
eth
eless
,
th
e
f
r
am
ewo
r
k
was
test
ed
o
n
l
y
o
n
b
in
ar
y
-
class
p
r
o
b
lem
s
with
r
elativ
ely
s
m
all
s
am
p
le
s
izes,
an
d
f
u
tu
r
e
wo
r
k
s
h
o
u
ld
ad
d
r
ess
m
u
lticlas
s
cla
s
s
if
icatio
n
,
lar
g
er
an
d
m
o
r
e
h
et
er
o
g
en
e
o
u
s
d
atasets
,
an
d
in
teg
r
atio
n
with
clin
ical
m
etad
ata
an
d
e
x
p
lain
ab
ilit
y
to
o
ls
s
u
ch
as
s
h
ap
ley
ad
d
itiv
e
e
x
p
lan
atio
n
s
(
SHAP
)
o
r
lo
ca
l
i
n
ter
p
r
etab
le
m
o
d
el
-
ag
n
o
s
tic
ex
p
lan
atio
n
s
(
LIME
)
to
en
h
an
ce
r
ea
l
-
wo
r
ld
ap
p
lica
b
ilit
y
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
s
tate
n
o
f
u
n
d
in
g
in
v
o
lv
ed
.
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
Sar
a
Had
d
o
u
B
o
u
az
z
a
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
J
ih
ad
Had
d
o
u
B
o
u
az
za
✓
✓
✓
✓
✓
✓
✓
✓
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.
I
NF
O
RM
E
D
CO
NS
E
N
T
T
h
is
s
tu
d
y
d
i
d
n
o
t
in
v
o
lv
e
in
d
iv
id
u
als
n
o
r
an
y
p
er
s
o
n
al
id
en
tific
atio
n
in
f
o
r
m
atio
n
t
h
at
co
u
ld
r
eq
u
i
r
e
an
y
in
f
o
r
m
ed
co
n
s
en
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
A
tw
o
-
s
tep
in
tellig
en
t fra
mew
o
r
k
fo
r
g
en
e
ex
p
r
ess
io
n
-
b
a
s
ed
…
(
S
a
r
a
Ha
d
d
o
u
B
o
u
a
z
z
a
)
4737
E
T
H
I
CAL AP
P
RO
V
AL
T
h
is
p
ap
er
d
o
es
n
o
t
in
v
o
lv
e
p
eo
p
le
o
r
an
im
als;
n
o
in
v
e
s
tig
atio
n
h
as
in
v
o
lv
e
d
h
u
m
a
n
s
u
b
jects.
T
h
er
ef
o
r
e,
th
e
a
u
th
o
r
s
d
i
d
n
o
t
s
ee
k
ap
p
r
o
v
al
f
r
o
m
an
y
in
s
titu
tio
n
al
r
ev
iew
b
o
a
r
d
.
DATA AV
AI
L
AB
I
L
I
T
Y
Data
av
ailab
ilit
y
is
n
o
t
ap
p
li
ca
b
le
to
th
is
p
ap
er
as
n
o
n
e
w
d
ata
wer
e
cr
ea
ted
o
r
an
aly
ze
d
in
th
is
s
tu
d
y
.
RE
F
E
R
E
NC
E
S
[
1
]
H
.
Z.
A
l
marz
o
u
k
i
,
“
D
e
e
p
-
l
e
a
r
n
i
n
g
-
b
a
sed
c
a
n
c
e
r
p
r
o
f
i
l
e
s
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
s
i
n
g
g
e
n
e
e
x
p
r
e
ss
i
o
n
d
a
t
a
p
r
o
f
i
l
e
,
”
J
o
u
r
n
a
l
o
f
H
e
a
l
t
h
c
a
re
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
2
0
2
2
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
2
/
4
7
1
5
9
9
8
.
[
2
]
G
.
G
e
e
t
h
a
a
n
d
P
.
G
e
e
t
h
a
n
j
a
l
i
,
“
P
a
t
t
e
r
n
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
b
e
a
r
i
n
g
f
a
u
l
t
s
i
n
P
M
S
M
b
a
se
d
o
n
t
i
me
d
o
m
a
i
n
f
e
a
t
u
r
e
e
n
s
e
m
b
l
e
s,”
En
g
i
n
e
e
ri
n
g
Re
s
e
a
rc
h
E
x
p
r
e
ss
,
v
o
l
.
6
,
n
o
.
3
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
8
8
/
2
6
3
1
-
8
6
9
5
/
a
d
1
1
f
8
.
[
3
]
S
.
S
h
a
r
m
a
,
T
.
H
.
P
r
i
y
a
,
a
n
d
V
.
P
.
S
.
N
a
i
d
u
,
“
O
p
t
i
mi
z
i
n
g
b
e
a
r
i
n
g
h
e
a
l
t
h
c
o
n
d
i
t
i
o
n
m
o
n
i
t
o
r
i
n
g
:
e
x
p
l
o
r
i
n
g
c
o
r
r
e
l
a
t
i
o
n
f
e
a
t
u
r
e
sel
e
c
t
i
o
n
a
l
g
o
r
i
t
h
m,
”
E
n
g
i
n
e
e
ri
n
g
R
e
s
e
a
r
c
h
Ex
p
ress
,
v
o
l
.
6
,
n
o
.
2
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
8
8
/
2
6
3
1
-
8
6
9
5
/
a
d
0
8
3
d
.
[
4
]
R
.
K
u
m
a
r
a
n
d
R
.
S
.
A
n
a
n
d
,
“
A
me
t
h
o
d
o
l
o
g
i
c
a
l
i
n
t
e
g
r
a
t
i
o
n
o
f
f
i
sh
e
r
sco
r
e
t
e
c
h
n
i
q
u
e
w
i
t
h
i
n
t
e
l
l
i
g
e
n
t
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
met
h
o
d
s f
o
r
b
a
l
l
b
e
a
r
i
n
g
f
a
u
l
t
i
n
v
e
s
t
i
g
a
t
i
o
n
,
”
E
n
g
i
n
e
e
ri
n
g
Re
s
e
a
rc
h
E
x
p
r
e
ss
,
v
o
l
.
6
,
n
o
.
2
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
8
8
/
2
6
3
1
-
8
6
9
5
/
a
d
0
8
4
1
.
[
5
]
H
.
El
w
a
h
s
h
e
t
a
l
.
,
“
A
n
e
w
a
p
p
r
o
a
c
h
f
o
r
c
a
n
c
e
r
p
r
e
d
i
c
t
i
o
n
b
a
se
d
o
n
d
e
e
p
n
e
u
r
a
l
l
e
a
r
n
i
n
g
,
”
J
o
u
rn
a
l
o
f
K
i
n
g
S
a
u
d
U
n
i
v
e
rs
i
t
y
–
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
3
5
,
n
o
.
6
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
k
s
u
c
i
.
2
0
2
2
.
1
0
1
5
6
5
.
[
6
]
S
.
H
.
B
o
u
a
z
z
a
,
“
A
d
e
e
p
e
n
s
e
m
b
l
e
g
e
n
e
se
l
e
c
t
i
o
n
a
n
d
a
t
t
e
n
t
i
o
n
-
g
u
i
d
e
d
c
l
a
s
si
f
i
c
a
t
i
o
n
f
r
a
mew
o
r
k
f
o
r
r
o
b
u
st
c
a
n
c
e
r
d
i
a
g
n
o
s
i
s
f
r
o
m
mi
c
r
o
a
r
r
a
y
d
a
t
a
,
”
E
n
g
i
n
e
e
r
i
n
g
,
T
e
c
h
n
o
l
o
g
y
&
A
p
p
l
i
e
d
S
c
i
e
n
c
e
Re
s
e
a
rch
,
v
o
l
.
1
5
,
n
o
.
1
,
p
p
.
2
0
2
3
5
–
2
0
2
4
1
,
2
0
2
5
,
d
o
i
:
1
0
.
4
8
0
8
4
/
e
t
a
sr
.
9
4
7
6
.
[
7
]
D
.
Tr
i
p
a
t
h
i
,
S
.
K
.
B
i
sw
a
s,
a
n
d
B
.
B
a
r
u
a
h
,
“
D
a
t
a
a
n
a
l
y
t
i
c
s
i
n
e
n
sem
b
l
e
l
e
a
r
n
i
n
g
f
o
r
e
f
f
e
c
t
i
v
e
c
r
o
p
y
i
e
l
d
p
r
e
d
i
c
t
i
o
n
,
”
E
n
g
i
n
e
e
r
i
n
g
Re
se
a
rc
h
Ex
p
ress
,
v
o
l
.
6
,
n
o
.
3
,
n
o
.
0
3
5
2
3
7
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
8
8
/
2
6
3
1
-
8
6
9
5
/
a
d
1
1
f
b
.
[
8
]
W
.
A
l
i
a
n
d
F
.
S
a
e
e
d
,
“
H
y
b
r
i
d
f
i
l
t
e
r
a
n
d
g
e
n
e
t
i
c
a
l
g
o
r
i
t
h
m
-
b
a
se
d
f
e
a
t
u
r
e
sel
e
c
t
i
o
n
f
o
r
i
mp
r
o
v
i
n
g
c
a
n
c
e
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
i
n
h
i
g
h
-
d
i
m
e
n
s
i
o
n
a
l
mi
c
r
o
a
r
r
a
y
d
a
t
a
,
”
Pro
c
e
s
ses
,
v
o
l
.
1
1
,
n
o
.
2
,
p
p
.
5
6
2
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
p
r
1
1
0
2
0
5
6
2
.
[
9
]
S
.
H
.
B
o
u
a
z
z
a
a
n
d
J
.
H
.
B
o
u
a
z
z
a
,
“
R
e
v
o
l
u
t
i
o
n
i
z
i
n
g
c
a
n
c
e
r
c
l
a
s
s
i
f
i
c
a
t
i
o
n
:
t
h
e
S
N
R
-
O
G
S
C
C
m
e
t
h
o
d
f
o
r
i
m
p
r
o
v
e
d
g
e
n
e
s
e
l
e
c
t
i
o
n
a
n
d
c
l
u
s
t
e
r
i
n
g
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
1
4
,
n
o
.
1
,
p
p
.
4
6
6
–
4
7
2
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
a
i
.
v
1
4
.
i
1
.
p
p
4
6
6
-
4
7
2
.
[
1
0
]
S
.
H
.
B
o
u
a
z
z
a
,
“
O
p
t
i
m
i
z
e
d
mac
h
i
n
e
l
e
a
r
n
i
n
g
f
o
r
c
a
n
c
e
r
c
l
a
ssi
f
i
c
a
t
i
o
n
v
i
a
t
h
r
e
e
-
s
t
a
g
e
g
e
n
e
s
e
l
e
c
t
i
o
n
,
”
E
n
g
i
n
e
e
ri
n
g
,
T
e
c
h
n
o
l
o
g
y
&
Ap
p
l
i
e
d
S
c
i
e
n
c
e
Re
s
e
a
rc
h
,
v
o
l
.
1
5
,
n
o
.
2
,
p
p
.
2
1
0
9
3
–
2
1
0
9
9
,
2
0
2
5
,
d
o
i
:
1
0
.
4
8
0
8
4
/
e
t
a
sr
.
9
4
7
3
.
[
1
1
]
R
.
D
a
s
h
,
“
A
n
a
d
a
p
t
i
v
e
h
a
r
m
o
n
y
s
e
a
r
c
h
a
p
p
r
o
a
c
h
f
o
r
g
e
n
e
s
e
l
e
c
t
i
o
n
a
n
d
c
l
a
s
s
i
f
i
c
a
t
i
o
n
o
f
h
i
g
h
d
i
m
e
n
s
i
o
n
a
l
m
e
d
i
c
a
l
d
a
t
a
,
”
J
o
u
r
n
a
l
o
f
K
i
n
g
S
a
u
d
U
n
i
v
e
r
s
i
t
y
–
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
r
m
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
3
3
,
n
o
.
2
,
p
p
.
1
9
5
–
2
0
7
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
k
s
u
c
i
.
2
0
1
9
.
0
9
.
0
0
4
.
[
1
2
]
A
.
B
e
n
k
e
s
s
i
r
a
t
a
n
d
N
.
B
e
n
b
l
i
d
i
a
,
“
A
n
o
v
e
l
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
a
p
p
r
o
a
c
h
b
a
s
e
d
o
n
c
o
n
s
t
r
a
i
n
e
d
e
i
g
e
n
v
a
l
u
e
s
o
p
t
i
m
i
z
a
t
i
o
n
,
”
J
o
u
r
n
a
l
o
f
K
i
n
g
S
a
u
d
U
n
i
v
e
r
s
i
t
y
–
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
r
m
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
3
4
,
n
o
.
8
,
p
p
.
4
8
3
6
–
4
8
4
6
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
k
s
u
c
i
.
2
0
2
1
.
0
8
.
0
0
5
.
[
1
3
]
M
.
N
.
K
P
a
n
d
P
.
Th
i
y
a
g
a
r
a
j
a
n
,
“
F
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
u
si
n
g
e
f
f
i
c
i
e
n
t
f
u
si
o
n
o
f
f
i
sh
e
r
sc
o
r
e
a
n
d
g
r
e
e
d
y
se
a
r
c
h
i
n
g
f
o
r
A
l
z
h
e
i
m
e
r
’
s
c
l
a
ss
i
f
i
c
a
t
i
o
n
,
”
J
o
u
r
n
a
l
o
f
K
i
n
g
S
a
u
d
U
n
i
v
e
rsi
t
y
–
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
r
m
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
3
4
,
n
o
.
8
,
p
p
.
4
9
9
3
–
5
0
0
6
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
k
s
u
c
i
.
2
0
2
1
.
0
8
.
0
0
6
.
[
1
4
]
E.
H
e
g
a
z
y
,
M
.
A
.
M
a
k
h
l
o
u
f
,
a
n
d
G
.
S
.
E
l
-
Ta
w
e
l
,
“
I
mp
r
o
v
e
d
sa
l
p
sw
a
r
m
a
l
g
o
r
i
t
h
m
f
o
r
f
e
a
t
u
r
e
se
l
e
c
t
i
o
n
,
”
J
o
u
r
n
a
l
o
f
K
i
n
g
S
a
u
d
U
n
i
v
e
rsi
t
y
–
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
3
2
,
n
o
.
3
,
p
p
.
3
3
5
–
3
4
4
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
k
su
c
i
.
2
0
1
8
.
0
9
.
0
0
1
.
[
1
5
]
K
.
A
.
U
t
h
ma
n
,
F
.
M
.
B
a
-
A
l
w
i
,
a
n
d
S
.
M
.
O
t
h
ma
n
,
“
A
s
u
r
v
e
y
o
n
f
e
a
t
u
r
e
se
l
e
c
t
i
o
n
i
n
m
i
c
r
o
a
r
r
a
y
d
a
t
a
:
m
e
t
h
o
d
s,
a
l
g
o
r
i
t
h
ms
a
n
d
c
h
a
l
l
e
n
g
e
s,”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
8
,
n
o
.
1
0
,
p
p
.
1
0
6
–
1
1
6
,
2
0
2
0
,
d
o
i
:
1
0
.
2
6
4
3
8
/
i
j
c
s
e
/
v
8
i
1
0
.
1
0
6
1
1
6
[
1
6
]
D
.
Ji
a
n
g
,
C
.
Ta
n
g
,
a
n
d
A
.
Z
h
a
n
g
,
“
C
l
u
s
t
e
r
a
n
a
l
y
si
s
f
o
r
g
e
n
e
e
x
p
r
e
ssi
o
n
d
a
t
a
:
a
su
r
v
e
y
,
”
I
EE
E
T
r
a
n
s
a
c
t
i
o
n
s
o
n
K
n
o
w
l
e
d
g
e
a
n
d
D
a
t
a
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
1
6
,
n
o
.
1
1
,
p
p
.
1
3
7
0
–
1
3
8
6
,
2
0
0
4
,
d
o
i
:
1
0
.
1
1
0
9
/
T
K
D
E.
2
0
0
4
.
6
8
.
[
1
7
]
T.
R
.
G
o
l
u
b
e
t
a
l
.
,
“
M
o
l
e
c
u
l
a
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
o
f
c
a
n
c
e
r
:
c
l
a
ss
d
i
sc
o
v
e
r
y
a
n
d
c
l
a
ss
p
r
e
d
i
c
t
i
o
n
b
y
g
e
n
e
e
x
p
r
e
ss
i
o
n
mo
n
i
t
o
r
i
n
g
,
”
S
c
i
e
n
c
e
,
v
o
l
.
2
8
6
,
p
p
.
5
3
1
–
5
3
7
,
1
9
9
9
,
d
o
i
:
1
0
.
1
1
2
6
/
s
c
i
e
n
c
e
.
2
8
6
.
5
4
3
9
.
5
3
1
.
[
1
8
]
J.
H
o
u
e
t
a
l
.
,
“
D
i
st
a
n
c
e
c
o
r
r
e
l
a
t
i
o
n
a
p
p
l
i
c
a
t
i
o
n
t
o
g
e
n
e
c
o
-
e
x
p
r
e
ss
i
o
n
n
e
t
w
o
r
k
a
n
a
l
y
si
s
,
”
B
MC
B
i
o
i
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
2
3
,
n
o
.
1
,
p
p
.
8
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
8
6
/
s1
2
8
5
9
-
0
2
2
-
0
4
6
3
2
-
8.
[
1
9
]
R
.
K
o
h
a
v
i
a
n
d
G
.
H
.
J
o
h
n
,
“
W
r
a
p
p
e
r
s
f
o
r
f
e
a
t
u
r
e
su
b
se
t
se
l
e
c
t
i
o
n
,
”
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
9
7
,
n
o
.
1
–
2
,
p
p
.
2
7
3
–
3
2
4
,
1
9
9
7
,
d
o
i
:
1
0
.
1
0
1
6
/
S
0
0
0
4
-
3
7
0
2
(
9
7
)
0
0
0
4
3
-
X.
[
2
0
]
S
.
K
w
o
n
,
H
.
Le
e
,
a
n
d
S
.
L
e
e
,
“
I
ma
g
e
e
n
h
a
n
c
e
m
e
n
t
w
i
t
h
G
a
u
ssi
a
n
f
i
l
t
e
r
i
n
g
i
n
t
i
me
-
d
o
m
a
i
n
mi
c
r
o
w
a
v
e
i
m
a
g
i
n
g
sy
st
e
m
f
o
r
b
r
e
a
st
c
a
n
c
e
r
d
e
t
e
c
t
i
o
n
,
”
E
l
e
c
t
r
o
n
i
c
s Le
t
t
e
rs
,
v
o
l
.
5
2
,
n
o
.
5
,
p
p
.
3
4
2
–
3
4
4
,
2
0
1
6
,
d
o
i
:
1
0
.
1
0
4
9
/
e
l
.
2
0
1
6
.
0
2
1
6
.
[
2
1
]
P
.
D
h
a
k
a
r
,
B
.
S
i
n
g
h
,
a
n
d
P
.
G
u
p
t
a
,
“
C
o
m
p
a
r
a
t
i
v
e
p
e
r
f
o
r
ma
n
c
e
a
n
a
l
y
si
s o
f
d
i
f
f
e
r
e
n
t
t
y
p
e
s o
f
k
-
n
e
a
r
e
st
n
e
i
g
h
b
o
r
(
k
-
N
N
)
c
l
a
ss
i
f
i
e
r
s
f
o
r
f
a
u
l
t
d
i
a
g
n
o
s
i
s
o
f
a
i
r
c
o
m
p
r
e
ss
o
r
s
e
t
u
p
,
”
En
g
i
n
e
e
ri
n
g
R
e
se
a
rc
h
Ex
p
ress
,
v
o
l
.
6
,
n
o
.
2
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
8
8
/
2
6
3
1
-
8
6
9
5
/
a
d
0
8
4
4
.
[
2
2
]
M
.
A
l
w
o
h
a
i
b
i
e
t
a
l
.
,
“
A
h
y
b
r
i
d
mu
l
t
i
-
st
a
g
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
b
a
se
d
o
n
b
r
a
i
n
s
t
o
r
m
i
n
g
o
p
t
i
mi
z
a
t
i
o
n
a
l
g
o
r
i
t
h
m
f
o
r
b
r
e
a
s
t
c
a
n
c
e
r
r
e
c
u
r
r
e
n
c
e
p
r
e
d
i
c
t
i
o
n
,
”
J
o
u
r
n
a
l
o
f
K
i
n
g
S
a
u
d
U
n
i
v
e
rs
i
t
y
–
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
3
4
,
n
o
.
8
,
p
p
.
5
1
9
2
–
5
2
0
3
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
k
su
c
i
.
2
0
2
1
.
0
7
.
0
0
6
.
[
2
3
]
V
.
G
u
l
e
r
i
a
,
V
.
K
u
m
a
r
,
a
n
d
P
.
K
.
S
i
n
g
h
,
“
C
l
a
s
s
i
f
i
c
a
t
i
o
n
o
f
s
u
r
f
a
c
e
r
o
u
g
h
n
e
s
s
d
u
r
i
n
g
t
u
r
n
i
n
g
o
f
f
o
r
g
e
d
E
N
8
s
t
e
e
l
u
s
i
n
g
v
i
b
r
a
t
i
o
n
s
i
g
n
a
l
p
r
o
c
e
s
s
i
n
g
a
n
d
s
u
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
,
”
E
n
g
i
n
e
e
r
i
n
g
R
e
s
e
a
r
c
h
E
x
p
r
e
s
s
,
v
o
l
.
4
,
n
o
.
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
8
8
/
2
6
3
1
-
8
6
9
5
/
a
c
5
d
7
6
.
[
2
4
]
Y
.
E.
A
l
mal
k
i
e
t
a
l
.
,
“
L
B
P
–
b
i
l
a
t
e
r
a
l
b
a
s
e
d
f
e
a
t
u
r
e
f
u
s
i
o
n
f
o
r
b
r
e
a
st
c
a
n
c
e
r
d
i
a
g
n
o
si
s
,
”
C
o
m
p
u
t
e
r
M
o
d
e
l
i
n
g
i
n
E
n
g
i
n
e
e
r
i
n
g
&
S
c
i
e
n
c
e
s
,
v
o
l
.
7
3
,
p
p
.
4
1
0
3
–
4
1
2
1
,
2
0
2
2
,
d
o
i
:
1
0
.
3
2
6
0
4
/
c
mc
.
2
0
2
2
.
0
2
3
4
2
9
.
[
2
5
]
S
.
H
.
B
o
u
a
z
z
a
a
n
d
J.
H
.
B
o
u
a
z
z
a
,
“
A
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
a
p
p
l
i
c
a
t
i
o
n
f
o
r
t
h
e
c
l
a
ssi
f
i
c
a
t
i
o
n
o
f
c
e
n
t
r
a
l
n
e
r
v
o
u
s
sy
s
t
e
m
t
u
mo
r
s
b
a
s
e
d
o
n
b
l
o
o
d
b
i
o
m
a
r
k
e
r
s,”
i
n
2
0
2
4
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
G
l
o
b
a
l
Ae
r
o
n
a
u
t
i
c
s
En
g
i
n
e
e
ri
n
g
a
n
d
S
a
t
e
l
l
i
t
e
T
e
c
h
n
o
l
o
g
y
(
G
AS
T
)
,
2
0
2
4
,
p
p
.
1
–
5
,
d
o
i
:
1
0
.
1
1
0
9
/
G
A
S
T
2
0
2
4
.
9
8
4
5
6
3
2
.
[
2
6
]
H
.
S
.
P
o
k
h
a
r
i
y
a
,
D
.
P
.
S
i
n
g
h
,
a
n
d
R
.
P
r
a
k
a
s
h
,
“
Ev
a
l
u
a
t
i
o
n
o
f
d
i
f
f
e
r
e
n
t
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
f
o
r
LU
LC
c
l
a
ss
i
f
i
c
a
t
i
o
n
i
n
h
e
t
e
r
o
g
e
n
e
o
u
s
l
a
n
d
sc
a
p
e
b
y
u
s
i
n
g
r
e
mo
t
e
s
e
n
si
n
g
a
n
d
G
I
S
t
e
c
h
n
i
q
u
e
s,
”
E
n
g
i
n
e
e
r
i
n
g
R
e
se
a
rc
h
Ex
p
ress
,
v
o
l
.
5
,
n
o
.
4
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
8
8
/
2
6
3
1
-
8
6
9
5
/
a
c
f
a
6
4
.
[
2
7
]
M
.
Ç
a
k
i
r
,
M
.
Y
i
l
ma
z
,
M
.
A
.
O
r
a
l
,
H
.
Ö
.
K
a
z
a
n
c
i
,
a
n
d
O
.
O
r
a
l
,
“
A
c
c
u
r
a
c
y
a
sse
ssm
e
n
t
o
f
R
F
e
r
n
s
,
N
B
,
S
V
M
,
a
n
d
k
N
N
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
c
l
a
ssi
f
i
e
r
s
i
n
a
q
u
a
c
u
l
t
u
r
e
,
”
J
o
u
rn
a
l
o
f
K
i
n
g
S
a
u
d
U
n
i
v
e
rsi
t
y
–
S
c
i
e
n
c
e
,
v
o
l
.
3
5
,
n
o
.
6
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
k
su
s.
2
0
2
3
.
1
0
2
7
5
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2252
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
4
,
No
.
6
,
Dec
em
b
er
2
0
2
5
:
4
7
3
1
-
4
7
3
8
4738
[
2
8
]
R
.
K
o
h
a
v
i
a
n
d
F
.
P
r
o
v
o
st
,
“
G
l
o
ss
a
r
y
o
f
t
e
r
ms,”
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
,
v
o
l
.
3
0
,
p
p
.
2
7
1
–
2
7
4
,
1
9
9
8
,
d
o
i
:
1
0
.
1
0
2
3
/
A
:
1
0
1
7
1
8
1
8
2
6
8
9
9
.
[
2
9
]
T.
C
o
v
e
r
a
n
d
P
.
H
a
r
t
,
“
N
e
a
r
e
st
n
e
i
g
h
b
o
r
p
a
t
t
e
r
n
c
l
a
ss
i
f
i
c
a
t
i
o
n
,
”
I
EEE
T
ra
n
s
a
c
t
i
o
n
s
o
n
I
n
f
o
rm
a
t
i
o
n
T
h
e
o
ry
,
v
o
l
.
1
3
,
n
o
.
1
,
p
p
.
2
1
–
2
7
,
1
9
6
7
,
d
o
i
:
1
0
.
1
1
0
9
/
TI
T.
1
9
6
7
.
1
0
5
3
9
6
4
.
[
3
0
]
S
.
O
sama
,
H
.
S
h
a
b
a
n
,
a
n
d
A
.
A
.
A
l
i
,
“
G
e
n
e
r
e
d
u
c
t
i
o
n
a
n
d
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
f
o
r
c
a
n
c
e
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
b
a
s
e
d
o
n
mi
c
r
o
a
r
r
a
y
g
e
n
e
e
x
p
r
e
ss
i
o
n
d
a
t
a
:
a
c
o
m
p
r
e
h
e
n
s
i
v
e
r
e
v
i
e
w
,
”
Ex
p
e
r
t
S
y
st
e
m
s
w
i
t
h
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
1
3
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
2
.
1
1
8
9
4
6
.
[
3
1
]
Z.
W
a
n
g
e
t
a
l
.
,
“
G
e
n
e
t
i
c
a
l
g
o
r
i
t
h
m
-
b
a
sed
f
e
a
t
u
r
e
se
l
e
c
t
i
o
n
w
i
t
h
ma
n
i
f
o
l
d
l
e
a
r
n
i
n
g
f
o
r
c
a
n
c
e
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
s
i
n
g
mi
c
r
o
a
r
r
a
y
d
a
t
a
,
”
BM
C
Bi
o
i
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
2
4
,
n
o
.
1
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
8
6
/
s
1
2
8
5
9
-
0
2
3
-
0
5
2
8
5
-
9.
[
32
]
X
.
D
e
n
g
,
M
.
L
i
,
S
.
D
e
n
g
,
a
n
d
L.
W
a
n
g
,
“
H
y
b
r
i
d
g
e
n
e
s
e
l
e
c
t
i
o
n
a
p
p
r
o
a
c
h
u
si
n
g
X
G
B
o
o
st
a
n
d
m
u
l
t
i
-
o
b
j
e
c
t
i
v
e
g
e
n
e
t
i
c
a
l
g
o
r
i
t
h
m
f
o
r
c
a
n
c
e
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
,
”
Me
d
i
c
a
l
&
B
i
o
l
o
g
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
&
C
o
m
p
u
t
i
n
g
,
v
o
l
.
6
0
,
n
o
.
3
,
p
p
.
6
6
3
–
6
8
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
5
1
7
-
021
-
0
2
4
2
0
-
4.
[
3
3
]
T.
N
g
u
y
e
n
,
A
.
K
h
o
sr
a
v
i
,
D
.
C
r
e
i
g
h
t
o
n
,
a
n
d
S
.
N
a
h
a
v
a
n
d
i
,
“
H
i
e
r
a
r
c
h
i
c
a
l
g
e
n
e
s
e
l
e
c
t
i
o
n
a
n
d
g
e
n
e
t
i
c
f
u
z
z
y
s
y
st
e
m
f
o
r
c
a
n
c
e
r
mi
c
r
o
a
r
r
a
y
d
a
t
a
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
PLo
S
O
N
E
,
v
o
l
.
1
0
,
n
o
.
3
,
2
0
1
5
,
d
o
i
:
1
0
.
1
3
7
1
/
j
o
u
r
n
a
l
.
p
o
n
e
.
0
1
2
0
3
6
4
.
[
3
4
]
X
.
L
i
u
,
A
.
K
r
i
sh
n
a
n
,
a
n
d
A
.
M
o
n
d
r
y
,
“
A
n
e
n
t
r
o
p
y
-
b
a
se
d
g
e
n
e
se
l
e
c
t
i
o
n
m
e
t
h
o
d
f
o
r
c
a
n
c
e
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
s
i
n
g
mi
c
r
o
a
r
r
a
y
d
a
t
a
,
”
BM
C
Bi
o
i
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
6
,
2
0
0
5
,
d
o
i
:
1
0
.
1
1
8
6
/
1
4
7
1
-
2
1
0
5
-
6
-
1.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
S
a
r
a
H
a
d
d
o
u
Bo
u
a
z
z
a
h
o
ld
s
a
Do
c
t
o
ra
te
in
El
e
c
tri
c
a
l
En
g
i
n
e
e
rin
g
a
n
d
In
fo
rm
a
ti
c
s,
a
s
we
ll
a
s
a
m
a
ste
r'
s
in
El
e
c
tri
c
a
l
En
g
i
n
e
e
rin
g
fr
o
m
Ca
d
i
Ay
y
a
d
Un
iv
e
rsity
,
M
a
rra
k
e
c
h
.
S
h
e
a
lso
c
o
m
p
lete
d
h
e
r
Ba
c
h
e
lo
r'
s
in
P
h
y
sic
a
l
S
c
i
e
n
c
e
s.
Cu
rre
n
tl
y
,
sh
e
is
a
p
ro
fe
ss
o
r
a
n
d
re
se
a
rc
h
e
r
a
t
th
e
LAM
IG
EP
Lab
o
ra
to
r
y
,
EM
S
I
M
a
rra
k
e
c
h
.
He
r
re
se
a
rc
h
in
c
lu
d
e
s
AI
tec
h
n
iq
u
e
s
f
o
r
c
a
n
c
e
r
c
las
sifica
ti
o
n
,
g
e
n
e
e
x
p
re
ss
i
o
n
a
n
a
ly
sis,
a
n
d
se
c
u
rit
y
c
h
a
ll
e
n
g
e
s in
Io
T
e
n
v
ir
o
n
m
e
n
ts
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
sa
r
a
.
h
b
.
sa
ra
@g
m
a
il
.
c
o
m
.
J
ih
a
d
H
a
d
d
o
u
B
o
u
a
z
z
a
is
a
n
e
n
g
in
e
e
r
sp
e
c
ializin
g
i
n
so
ftwa
r
e
e
n
g
in
e
e
rin
g
a
n
d
ima
g
e
p
ro
c
e
ss
in
g
fr
o
m
IG
A
-
In
stit
u
t
S
u
p
é
rieu
r
d
u
G
é
n
ieA
p
p
li
q
u
é
,
M
a
rra
k
e
c
h
.
Cu
rre
n
tl
y
,
h
e
se
rv
e
s
a
s
a
se
n
io
r
f
u
ll
sta
c
k
d
e
v
e
l
o
p
e
r
a
n
d
tec
h
lea
d
a
t
Ne
x
u
lar
Co
r
p
.
He
is
c
e
rti
fied
i
n
P
y
t
h
o
n
,
m
a
c
h
in
e
lea
rn
in
g
,
a
n
d
as
a
c
e
rti
fi
e
d
n
e
two
rk
se
c
u
rit
y
sp
e
c
ialist
(CNSS
).
His
re
se
a
rc
h
in
c
lu
d
e
s
p
a
tt
e
rn
re
c
o
g
n
it
i
o
n
u
sin
g
a
rti
fici
a
l
in
telli
g
e
n
c
e
,
with
a
p
u
b
li
c
a
ti
o
n
p
re
se
n
ted
at
th
e
G
AST
2
4
c
o
n
g
re
ss
.
He
can
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
h
a
d
d
o
u
.
j
ih
a
d
@
g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.