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.
15
,
No
.
4
,
A
u
g
u
s
t
20
26
,
p
p
.
3
7
2
2
~
3
7
3
1
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
15
.i
4
.
p
p
3
7
2
2
-
3
7
3
1
3722
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
An optimiz
ed de
e
p learning
f
ra
mewo
rk f
o
r b
ra
in t
u
mo
r
cla
ss
ificatio
n
usin
g
ma
g
netic
reso
n
a
nce ima
g
ing
K
o
ma
l
K
u
ma
r
N
a
p
a
1
,
R
a
j
ku
ma
r
G
o
v
i
n
d
a
r
a
j
a
n
2
,
Se
n
t
h
i
l
M
u
ru
g
a
n
J
a
n
a
ki
r
a
ma
n
3
,
J
a
y
a
n
t
h
i
Ar
u
mu
g
a
m
4
1
D
e
p
a
r
t
me
n
t
o
f
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
D
a
t
a
S
c
i
e
n
c
e
,
S
a
v
e
e
t
h
a
E
n
g
i
n
e
e
r
i
n
g
C
o
l
l
e
g
e
,
C
h
e
n
n
a
i
,
I
n
d
i
a
2
D
e
p
a
r
t
m
e
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
&
E
n
g
i
n
e
e
r
i
n
g
(
D
a
t
a
S
c
i
e
n
c
e
)
,
M
a
d
a
n
a
p
a
l
l
e
I
n
s
t
i
t
u
t
e
o
f
T
e
c
h
n
o
l
o
g
y
&
S
c
i
e
n
c
e
D
e
e
m
e
d
t
o
b
e
U
n
i
v
e
r
s
i
t
y
,
M
a
d
a
n
a
p
a
l
l
e
,
I
n
d
i
a
3
C
e
n
t
r
e
of
E
x
c
e
l
l
e
n
c
e
,
R
a
j
a
l
a
k
s
h
m
i
E
n
g
i
n
e
e
r
i
n
g
C
o
l
l
e
g
e
,
Th
a
n
d
a
l
a
m,
C
h
e
n
n
a
i
,
I
n
d
i
a
4
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
,
V
e
l
a
mm
a
l
En
g
i
n
e
e
r
i
n
g
C
o
l
l
e
g
e
,
C
h
e
n
n
a
i
,
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
J
u
n
18
,
2
0
2
5
R
ev
is
ed
J
u
l
13
,
2
0
2
6
Acc
ep
ted
J
u
l
21
,
2
0
2
6
Ac
c
u
ra
te
a
n
d
i
n
terp
re
tab
le
c
la
ss
ifi
c
a
ti
o
n
o
f
b
ra
in
tu
m
o
rs
in
m
a
g
n
e
ti
c
re
so
n
a
n
c
e
ima
g
in
g
(M
RI)
sc
a
n
s
p
lay
s
a
c
ru
c
ial
r
o
le
i
n
e
a
rly
d
iag
n
o
sis
a
n
d
e
ffe
c
ti
v
e
trea
tme
n
t
p
lan
n
in
g
.
Th
is
stu
d
y
in
tr
o
d
u
c
e
s
a
d
e
e
p
lea
rn
in
g
(DL)
fra
m
e
wo
rk
b
a
se
d
o
n
a
c
u
st
o
m
ize
d
YO
LOv
5
m
a
rc
h
i
tec
tu
re
in
teg
ra
ted
with
a
b
id
irec
ti
o
n
a
l
fe
a
tu
re
p
y
ra
m
i
d
n
e
t
wo
rk
(BiF
P
N)
f
o
r
m
u
lt
i
-
c
las
s
b
r
a
in
tu
m
o
r
c
las
sifica
ti
o
n
.
Th
e
i
n
teg
ra
ti
o
n
o
f
BiF
P
N
e
n
h
a
n
c
e
s
m
u
lt
i
-
sc
a
le
fe
a
t
u
re
fu
sio
n
,
imp
ro
v
i
n
g
d
e
tec
ti
o
n
a
c
ro
ss
v
a
ried
tu
m
o
r
t
y
p
e
s,
wh
il
e
YO
LOv
5
m
e
n
su
re
s
re
a
l
-
ti
m
e
in
fe
re
n
c
e
c
a
p
a
b
il
it
ies
.
To
m
it
ig
a
te
c
las
s
imb
a
lan
c
e
,
a
c
las
s
-
we
ig
h
ted
c
ro
ss
-
e
n
tro
p
y
l
o
ss
is
a
d
o
p
ted
.
T
h
e
m
o
d
e
l
is
e
v
a
l
u
a
ted
o
n
m
u
l
ti
p
le
p
e
rfo
rm
a
n
c
e
m
e
tri
c
s,
a
c
h
iev
in
g
a
tes
t
a
c
c
u
ra
c
y
o
f
8
8
.
8
6
%
,
p
r
e
c
isio
n
o
f
8
8
.
7
0
%
,
re
c
a
ll
o
f
8
8
.
2
0
%
,
a
n
d
F
1
-
sc
o
re
o
f
8
8
.
2
5
%
.
I
t
a
lso
re
p
o
rt
s
a
m
e
a
n
a
v
e
ra
g
e
p
re
c
isio
n
(m
AP@0
.
5
)
o
f
9
4
.
3
6
%
,
wit
h
h
ig
h
c
las
s
-
wis
e
a
v
e
ra
g
e
p
re
c
isio
n
(
AP
)
f
o
r
g
li
o
m
a
,
m
e
n
i
n
g
i
o
m
a
,
p
i
tu
it
a
r
y
,
a
n
d
n
o
-
tu
m
o
r
c
a
teg
o
ries
.
Co
m
p
u
tati
o
n
a
l
ti
m
e
fo
r
train
in
g
(1
2
4
8
4
.
8
1
se
c
o
n
d
s)
a
n
d
tes
ti
n
g
(1
4
6
.
8
2
se
c
o
n
d
s)
c
o
n
firms
t
h
e
m
o
d
e
l’s
fe
a
sib
il
it
y
f
o
r
re
a
l
-
ti
m
e
c
li
n
ica
l
d
e
p
lo
y
m
e
n
t
.
To
su
p
p
o
rt
in
ter
p
re
tab
il
it
y
,
g
ra
d
ien
t
-
we
ig
h
ted
c
las
s
a
c
ti
v
a
ti
o
n
m
a
p
p
in
g
(G
ra
d
-
CAM)
is
in
teg
ra
ted
fo
r
v
is
u
a
li
z
in
g
c
las
s
-
d
isc
rimin
a
ti
v
e
re
g
io
n
s,
h
e
lp
i
n
g
c
li
n
icia
n
s
u
n
d
e
rsta
n
d
th
e
m
o
d
e
l’s
p
re
d
icti
o
n
s
.
A
g
ra
d
i
o
-
b
a
se
d
u
se
r
in
terfa
c
e
is also
d
e
v
e
l
o
p
e
d
,
e
n
a
b
l
i
n
g
i
n
t
u
it
i
v
e
in
tera
c
ti
o
n
wit
h
t
h
e
sy
ste
m
.
K
ey
w
o
r
d
s
:
B
r
ain
tu
m
o
r
class
if
icatio
n
C
lin
ical
d
ec
is
io
n
s
u
p
p
o
r
t
Ma
g
n
etic
r
eso
n
an
ce
im
a
g
in
g
Me
d
ical
im
ag
e
an
aly
s
is
Op
tim
ized
d
ee
p
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
:
Ko
m
al
Ku
m
ar
Nap
a
Dep
ar
tm
en
t o
f
Ar
tific
ial
I
n
tellig
en
ce
an
d
Data
Scien
ce
,
Sav
e
eth
a
E
n
g
in
ee
r
in
g
C
o
lleg
e
C
h
en
n
ai,
T
am
il Na
d
u
,
I
n
d
ia
E
m
ail:
k
o
m
alk
u
m
ar
n
ap
a@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
B
r
ain
tu
m
o
r
s
r
em
ain
a
m
o
n
g
th
e
m
o
s
t
cr
itical
an
d
li
f
e
-
th
r
ea
ten
in
g
n
e
u
r
o
l
o
g
ical
d
is
o
r
d
er
s
,
n
ec
ess
itatin
g
ea
r
ly
an
d
ac
cu
r
a
te
d
iag
n
o
s
is
to
im
p
r
o
v
e
p
atie
n
t
o
u
tco
m
es.
Ma
g
n
etic
r
eso
n
a
n
ce
im
ag
in
g
(
MRI)
is
th
e
s
tan
d
ar
d
n
o
n
-
in
v
asiv
e
m
o
d
ality
f
o
r
b
r
ain
tu
m
o
r
d
ete
ctio
n
d
u
e
to
its
h
ig
h
s
p
atial
r
eso
lu
tio
n
an
d
tis
s
u
e
co
n
tr
ast.
Ho
wev
er
,
m
a
n
u
al
i
n
ter
p
r
etatio
n
o
f
MRI
s
ca
n
s
is
t
im
e
-
co
n
s
u
m
in
g
an
d
s
u
s
ce
p
tib
le
to
in
ter
-
o
b
s
er
v
e
r
v
ar
iab
ilit
y
.
T
o
ad
d
r
ess
th
ese
ch
allen
g
es,
d
ee
p
lear
n
in
g
(
D
L
)
tech
n
i
q
u
es
h
a
v
e
em
e
r
g
ed
as
p
o
wer
f
u
l
to
o
ls
f
o
r
au
to
m
atin
g
b
r
ain
tu
m
o
r
class
if
icatio
n
an
d
s
eg
m
e
n
tatio
n
task
s
.
R
ec
en
t
ad
v
an
ce
m
en
ts
in
h
y
b
r
id
an
d
tr
an
s
f
o
r
m
er
-
b
ased
m
o
d
els
h
av
e
s
ig
n
if
ican
tly
im
p
r
o
v
ed
class
if
icatio
n
ac
cu
r
ac
y
an
d
in
ter
p
r
etab
ilit
y
.
Ah
m
ed
et
a
l.
[
1
]
p
r
o
p
o
s
ed
a
h
y
b
r
id
v
is
io
n
tr
a
n
s
f
o
r
m
er
(
ViT
)
an
d
g
ated
r
ec
u
r
r
en
t
u
n
it
(
GR
U
)
ar
ch
itectu
r
e
f
o
r
b
r
ain
tu
m
o
r
d
etec
tio
n
in
So
u
th
er
n
B
an
g
la
d
esh
,
em
p
h
asizin
g
ex
p
lain
ab
le
a
r
tific
ial
in
tellig
e
n
ce
(
XAI
)
f
o
r
clin
ical
tr
a
n
s
p
ar
en
cy
.
Su
n
et
a
l.
[
2
]
ex
p
lo
r
e
d
a
c
o
m
b
in
atio
n
o
f
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
n
o
p
timiz
ed
d
ee
p
lea
r
n
in
g
fr
a
mewo
r
k
fo
r
b
r
a
in
tu
mo
r
cla
s
s
ifica
tio
n
u
s
in
g
…
(
K
o
ma
l Ku
ma
r
N
a
p
a
)
3723
YOL
Ov
8
an
d
d
ata
ef
f
icien
t
i
m
ag
e
tr
a
n
s
f
o
r
m
er
(
DeiT
)
to
ac
h
iev
e
an
o
m
aly
-
awa
r
e
d
ia
g
n
o
s
is
,
d
em
o
n
s
tr
atin
g
th
e
p
o
ten
tial
o
f
r
ea
l
-
tim
e
o
b
j
ec
t
d
etec
tio
n
m
o
d
els
in
m
ed
ical
im
ag
in
g
.
T
h
ese
in
n
o
v
atio
n
s
r
ef
lect
a
g
r
o
win
g
in
ter
est in
lev
er
ag
in
g
o
b
ject
d
etec
tio
n
f
r
am
ewo
r
k
s
lik
e
YOL
O
f
o
r
tu
m
o
r
lo
ca
lizatio
n
a
n
d
c
lass
if
icatio
n
.
T
r
ad
itio
n
al
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
s
(
C
NNs)
h
av
e
d
o
m
in
ated
th
e
liter
atu
r
e
f
r
o
m
2
0
1
5
to
2
0
2
0
,
as
r
ev
iewe
d
b
y
Naz
ir
et
a
l.
[
3
]
,
b
u
t
m
o
r
e
r
ec
e
n
t
ar
ch
it
ec
tu
r
es
in
co
r
p
o
r
ate
atten
tio
n
m
ec
h
an
is
m
s
an
d
m
u
ltis
ca
le
f
ea
tu
r
es.
Fo
r
ex
am
p
le,
I
R
Netv
[
4
]
an
d
Ar
m
-
Net
[
5
]
u
s
e
atten
tio
n
-
g
u
id
ed
m
o
d
u
les
an
d
r
esid
u
al
co
n
n
ec
tio
n
s
to
ex
tr
ac
t
r
ich
e
r
r
ep
r
esen
tatio
n
s
o
f
tu
m
o
r
r
eg
io
n
s
.
So
lan
k
i
et
a
l.
[
6
]
p
r
esen
te
d
a
co
m
p
r
eh
e
n
s
iv
e
o
v
er
v
iew
o
f
i
n
tellig
en
t
tech
n
iq
u
es
f
o
r
b
r
ain
tu
m
o
r
c
lass
if
icatio
n
,
h
ig
h
lig
h
tin
g
th
e
cr
itical
r
o
le
o
f
co
m
p
u
tatio
n
al
ef
f
icien
c
y
a
n
d
d
iag
n
o
s
tic
r
eliab
ilit
y
.
T
al
u
k
d
er
et
a
l.
[
7
]
p
r
o
p
o
s
ed
a
tr
a
n
s
f
er
lear
n
i
n
g
-
b
ased
ap
p
r
o
ac
h
f
o
r
b
r
ai
n
tu
m
o
r
ca
teg
o
r
izatio
n
.
Fu
r
th
er
m
o
r
e,
d
ee
p
s
em
an
tic
s
eg
m
e
n
tatio
n
f
r
am
ewo
r
k
s
lik
e
Dee
p
L
ab
V3
[
8
]
an
d
o
p
tim
ize
d
C
NN
v
ar
ian
ts
s
u
ch
as
E
S
R
Net
[
9
]
h
av
e
p
u
s
h
ed
th
e
b
o
u
n
d
ar
ies
o
f
au
to
m
ated
b
r
ain
tu
m
o
r
g
r
ad
in
g
.
Desp
ite
th
ese
ad
v
an
ce
m
en
ts
,
a
g
ap
r
em
ain
s
in
ac
h
iev
in
g
a
b
ala
n
ce
b
etwe
en
d
etec
tio
n
s
p
ee
d
,
in
ter
p
r
etab
ilit
y
,
an
d
class
if
icat
io
n
p
r
ec
is
io
n
.
T
o
b
r
id
g
e
th
is
g
ap
,
o
u
r
wo
r
k
in
te
g
r
ates
th
e
YOL
Ov
5
m
b
ac
k
b
o
n
e
with
a
b
id
ir
ec
tio
n
al
f
ea
tu
r
e
p
y
r
am
id
n
etwo
r
k
(
B
iFP
N)
f
o
r
en
h
an
ce
d
f
ea
t
u
r
e
f
u
s
io
n
an
d
tu
m
o
r
lo
ca
lizatio
n
.
T
h
e
class
if
icatio
n
h
ea
d
is
d
es
ig
n
ed
to
h
an
d
le
m
u
lti
-
class
b
r
ain
tu
m
o
r
r
ec
o
g
n
itio
n
,
wh
ile
g
r
ad
ien
t
-
weig
h
ted
class
ac
tiv
atio
n
m
ap
p
in
g
(
Gr
ad
-
C
AM
)
is
em
p
lo
y
ed
f
o
r
v
is
u
al
ex
p
lan
atio
n
o
f
m
o
d
el
d
ec
is
io
n
s
,
alig
n
in
g
with
th
e
g
r
o
win
g
d
em
a
n
d
f
o
r
X
AI
i
n
h
ea
lth
ca
r
e
[
1
0
]
.
T
h
is
s
tu
d
y
f
u
r
th
e
r
in
c
o
r
p
o
r
ate
s
weig
h
ted
lo
s
s
f
u
n
ctio
n
s
to
a
d
d
r
ess
class
im
b
alan
ce
.
I
ts
p
e
r
f
o
r
m
a
n
ce
is
ev
alu
ated
u
s
in
g
m
etr
ics
s
u
ch
as
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
F1
-
s
co
r
e,
an
d
m
ea
n
av
er
ag
e
p
r
ec
is
io
n
(
m
AP
@
0
.
5
)
.
C
o
m
p
ar
e
d
to
e
x
is
tin
g
m
o
d
els,
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
o
f
f
er
s
an
ef
f
ec
tiv
e
tr
ad
e
-
o
f
f
b
etwe
en
co
m
p
u
tatio
n
al
s
p
ee
d
an
d
cla
s
s
if
icatio
n
ac
cu
r
ac
y
,
m
ak
in
g
it
a
v
iab
le
s
o
lu
tio
n
f
o
r
r
ea
l
-
tim
e
b
r
ain
tu
m
o
r
d
iag
n
o
s
is
in
clin
ical
s
ettin
g
s
.
T
h
e
p
a
p
er
is
s
tr
u
ctu
r
ed
as
f
o
llo
ws
.
S
ec
tio
n
2
d
is
cu
s
s
es
r
elate
d
wo
r
k
s
.
S
ec
tio
n
3
o
u
t
lin
es
th
e
p
r
o
p
o
s
ed
m
eth
o
d
o
lo
g
y
,
s
ec
tio
n
4
p
r
esen
ts
ex
p
er
im
en
tal
a
n
aly
s
is
an
d
f
in
d
in
g
s
,
an
d
s
ec
tio
n
5
co
n
clu
d
es st
u
d
y
.
2.
RE
L
AT
E
D
WO
RK
S
B
r
ain
tu
m
o
r
d
etec
tio
n
an
d
class
if
icatio
n
u
s
in
g
MRI
im
ag
es
h
as
g
ar
n
er
e
d
s
ig
n
if
ica
n
t
a
tten
tio
n
in
r
ec
en
t
y
ea
r
s
,
esp
ec
ially
with
t
h
e
ad
v
e
n
t
o
f
DL
tec
h
n
iq
u
es.
Nu
m
er
o
u
s
s
tu
d
ies h
av
e
p
r
o
p
o
s
ed
ad
v
an
ce
d
m
o
d
els
f
o
r
im
p
r
o
v
i
n
g
d
iag
n
o
s
tic
ac
cu
r
ac
y
,
ef
f
icien
cy
,
an
d
in
ter
p
r
et
ab
ilit
y
.
Ali
et
a
l.
[
1
1
]
in
tr
o
d
u
ce
d
an
e
v
o
lu
tio
n
ar
y
C
NN
to
o
p
tim
ize
f
ea
tu
r
e
le
ar
n
in
g
f
o
r
tu
m
o
r
d
etec
tio
n
,
o
u
tp
er
f
o
r
m
in
g
co
n
v
en
tio
n
al
C
NN
ar
ch
itectu
r
es.
Similar
ly
,
Ku
m
ar
et
a
l.
[
1
2
]
p
r
esen
ted
a
h
y
b
r
id
b
o
o
s
ted
e
n
s
em
b
le
lear
n
in
g
a
p
p
r
o
ac
h
t
h
at
co
m
b
in
ed
m
u
ltip
le
d
ee
p
lear
n
er
s
to
en
h
an
ce
th
e
r
o
b
u
s
tn
ess
o
f
Alzh
eim
er
’
s
an
aly
s
is
.
T
h
e
ap
p
licatio
n
o
f
p
u
r
e
DL
s
tr
ateg
ies
h
as
also
b
ee
n
ex
p
l
o
r
ed
ex
ten
s
iv
el
y
.
L
iu
an
d
W
an
g
[
1
3
]
d
em
o
n
s
tr
ated
th
e
im
p
ac
t
o
f
ad
v
a
n
ce
d
C
NN
ar
ch
itectu
r
es
in
MRI
-
b
ased
b
r
ain
tu
m
o
r
cl
ass
if
icatio
n
,
s
h
o
win
g
th
at
m
o
d
el
d
ep
th
a
n
d
d
ata
q
u
ality
s
ig
n
if
ican
tly
af
f
ec
t
p
er
f
o
r
m
an
ce
.
Po
k
h
r
el
et
a
l.
[
1
4
]
u
tili
ze
d
d
ee
p
C
NNs
s
p
ec
if
ically
tr
ain
ed
o
n
MRI
im
ag
e
s
f
o
r
class
if
icatio
n
task
s
,
ac
h
iev
in
g
h
ig
h
ac
cu
r
ac
y
ac
r
o
s
s
m
u
ltip
le
tu
m
o
r
ty
p
es.
B
o
u
h
af
r
a
an
d
B
ah
i
[
1
5
]
co
n
d
u
cted
a
s
y
s
tem
atic
r
ev
iew
o
f
DL
tech
n
iq
u
es
f
r
o
m
2
0
2
0
to
2
0
2
4
,
s
u
m
m
ar
izin
g
p
r
o
g
r
ess
in
c
lass
if
icatio
n
,
s
eg
m
en
tatio
n
,
an
d
g
e
n
er
aliza
tio
n
ac
r
o
s
s
m
u
ltip
le
d
atasets
.
C
o
m
p
lem
en
tar
ily
,
Ku
m
ar
et
a
l.
[
1
6
]
r
e
v
iewe
d
a
b
r
o
a
d
er
r
an
g
e
o
f
in
tellig
en
t
tech
n
i
q
u
es,
in
clu
d
in
g
C
NNs,
en
s
em
b
le
m
eth
o
d
s
,
an
d
h
y
b
r
id
ap
p
r
o
ac
h
es.
A
v
a
r
iety
o
f
n
o
v
el
ar
c
h
itectu
r
es
h
a
v
e
b
ee
n
p
r
o
p
o
s
ed
to
tack
le
m
u
lti
-
class
b
r
ain
tu
m
o
r
class
if
icatio
n
ch
allen
g
es.
B
ey
o
n
d
DL
,
ea
r
lier
m
eth
o
d
s
u
tili
ze
d
tr
ad
itio
n
al
m
ac
h
i
n
e
lear
n
in
g
with
h
an
d
cr
af
ted
f
e
atu
r
es
an
d
o
p
tim
izatio
n
s
tr
ateg
ies.
Fo
r
in
s
tan
ce
,
th
e
au
th
o
r
s
[
1
7
]
,
[
1
8
]
ap
p
lied
p
ar
ticle
s
war
m
o
p
tim
izatio
n
an
d
ar
tific
ial
b
ee
c
o
lo
n
y
alg
o
r
ith
m
s
f
o
r
MRI
class
if
icatio
n
.
Oth
er
n
o
ta
b
le
ap
p
r
o
ac
h
es
in
clu
d
e
wav
elet
-
b
ased
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM
)
class
if
icatio
n
[
1
9
]
,
s
p
id
e
r
-
web
p
l
o
ttin
g
tech
n
iq
u
es
[
2
0
]
,
g
en
etic
p
atter
n
s
ea
r
ch
alg
o
r
ith
m
s
[
2
1
]
,
an
d
h
y
b
r
i
d
m
o
d
els
co
m
b
i
n
in
g
o
p
tim
izatio
n
an
d
n
eu
r
a
l
n
etwo
r
k
s
[
2
2
]
.
Giv
ian
an
d
C
alb
im
o
n
te
[
2
3
]
r
ev
iewe
d
a
n
o
n
-
in
v
asiv
e
MRI
-
b
ased
ap
p
r
o
ac
h
f
o
r
d
etec
t
in
g
n
eu
r
o
co
g
n
itiv
e
d
is
o
r
d
er
s
b
y
e
x
tr
ac
tin
g
d
is
ea
s
e
-
s
p
ec
if
ic
f
ea
tu
r
e
v
e
cto
r
s
.
Nap
a
et
a
l.
[
2
4
]
in
tr
o
d
u
ce
d
a
m
eta
-
en
s
em
b
le
f
r
am
ewo
r
k
f
o
r
b
r
ain
tu
m
o
r
u
s
in
g
in
d
e
p
en
d
e
n
t
r
eg
io
n
-
of
-
in
ter
est
(
R
OI
)
f
ea
tu
r
es
e
x
tr
ac
ted
f
r
o
m
T
1
-
weig
h
ted
MRI
s
ca
n
s
.
B
y
co
m
b
in
in
g
m
u
ltip
le
class
if
ier
en
s
em
b
les
tr
ain
ed
o
n
a
n
ato
m
i
ca
lly
d
ef
in
ed
R
OI
s
,
th
eir
ap
p
r
o
ac
h
ac
h
iev
e
d
im
p
r
o
v
ed
d
iag
n
o
s
tic
p
er
f
o
r
m
an
ce
an
d
h
ig
h
lig
h
ted
t
h
e
m
o
s
t
d
is
c
r
im
in
ativ
e
b
r
ain
r
eg
io
n
s
.
Mo
r
en
o
et
a
l.
[
2
5
]
p
r
o
p
o
s
ed
a
n
en
s
em
b
le
-
b
ased
b
r
ain
tu
m
o
r
class
if
icatio
n
u
s
in
g
tr
a
n
s
f
er
lear
n
in
g
a
n
d
a
n
en
s
em
b
le
s
tr
ateg
y
.
C
h
au
h
an
et
a
l.
[
2
6
]
p
r
o
p
o
s
e
d
a
h
y
b
r
id
lear
n
in
g
f
r
am
ew
o
r
k
f
o
r
ce
r
v
ical
ca
n
ce
r
d
ia
g
n
o
s
is
th
at
in
teg
r
ates
p
r
o
g
r
ess
iv
e
im
a
g
e
r
esizin
g
,
d
ee
p
tr
an
s
f
er
lear
n
in
g
,
an
d
p
r
in
cip
al
co
m
p
o
n
e
n
t
an
aly
s
is
f
o
r
f
ea
tu
r
e
r
ed
u
ctio
n
.
T
h
eir
a
p
p
r
o
ac
h
co
m
b
in
es
f
in
e
-
tu
n
ed
C
NN
m
o
d
e
ls
with
class
ical
m
ac
h
in
e
lear
n
in
g
class
if
ier
s
u
s
in
g
en
s
em
b
le
v
o
tin
g
,
ac
h
ie
v
in
g
v
e
r
y
h
ig
h
class
if
icatio
n
ac
cu
r
ac
y
o
n
wh
o
le
s
lid
e
im
ag
es.
T
h
e
s
t
u
d
y
h
ig
h
lig
h
ts
th
e
ef
f
ec
tiv
en
ess
o
f
h
y
b
r
id
d
ee
p
–
m
ac
h
in
e
lear
n
in
g
p
ip
elin
es
in
en
h
an
cin
g
d
iag
n
o
s
tic
p
er
f
o
r
m
an
ce
,
p
ar
ticu
la
r
ly
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
4
,
Au
g
u
s
t
20
26
:
3
7
2
2
-
3
7
3
1
3724
wh
en
o
p
tim
ized
f
ea
tu
r
e
s
elec
tio
n
an
d
en
s
em
b
le
s
tr
ateg
ies
ar
e
em
p
l
o
y
ed
.
Var
io
u
s
en
s
e
m
b
le
m
eth
o
d
s
h
a
v
e
also
s
h
o
wn
p
r
o
m
is
e
in
m
ed
ic
al
im
ag
e
a
n
aly
s
is
.
T
h
e
a
u
th
o
r
s
[
2
7
]
,
[
2
8
]
p
r
o
p
o
s
ed
f
u
zz
y
r
a
n
k
-
b
ased
en
s
em
b
les
o
f
C
NNs
f
o
r
cy
to
lo
g
y
an
d
M
R
I
s
eg
m
en
tatio
n
.
An
o
th
e
r
au
t
h
o
r
s
[
2
9
]
,
[
3
0
]
e
x
p
lo
r
e
d
tr
ip
la
n
ar
en
s
em
b
le
m
o
d
els
f
o
r
b
r
ain
tu
m
o
r
s
eg
m
en
tatio
n
an
d
wh
ite
m
atter
h
y
p
er
i
n
ten
s
ities
.
T
h
e
s
u
cc
ess
o
f
th
ese
ap
p
r
o
a
ch
es
h
ig
h
lig
h
ts
th
e
tr
an
s
itio
n
f
r
o
m
h
a
n
d
cr
af
te
d
f
ea
tu
r
e
e
x
tr
ac
tio
n
to
au
to
m
ated
,
h
ig
h
-
d
im
e
n
s
io
n
al
DL
m
eth
o
d
s
.
T
h
is
tr
an
s
itio
n
h
as
also
led
to
a
g
r
o
win
g
e
m
p
h
asis
o
n
m
o
d
el
in
ter
p
r
etab
ilit
y
,
co
m
p
u
tatio
n
a
l
ef
f
icien
cy
,
an
d
m
u
lti
-
class
class
if
icatio
n
ca
p
ab
ilit
y
.
T
h
is
s
tu
d
y
b
u
ild
s
u
p
o
n
th
ese
in
s
ig
h
ts
b
y
p
r
o
p
o
s
in
g
a
YOL
Ov
5
m
-
B
iFP
N
-
b
ased
b
r
ain
tu
m
o
r
class
if
icatio
n
m
o
d
el
in
teg
r
ated
with
Gr
ad
-
C
AM
f
o
r
v
is
u
al
ex
p
lan
a
tio
n
,
em
p
h
asizin
g
b
o
th
ac
cu
r
a
cy
an
d
e
x
p
lain
ab
ilit
y
in
a
r
ea
l
-
tim
e
s
ettin
g
.
3.
M
E
T
H
O
D
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
o
lo
g
y
in
tr
o
d
u
ce
s
a
r
o
b
u
s
t
DL
f
r
am
ewo
r
k
f
o
r
m
u
lti
-
class
b
r
ain
tu
m
o
r
class
if
icatio
n
u
s
in
g
a
m
o
d
if
ie
d
YOL
Ov
5
m
ar
ch
itectu
r
e
in
te
g
r
ated
with
a
B
iFP
N
,
as
s
h
o
wn
in
Fig
u
r
e
1
.
T
h
e
YOL
Ov
5
m
b
ac
k
b
o
n
e
ef
f
icien
tly
ex
tr
ac
ts
r
ich
s
p
atial
an
d
s
e
m
an
tic
f
ea
tu
r
es
f
r
o
m
MRI
im
ag
es,
wh
ile
B
i
FP
N
r
ef
in
es
an
d
f
u
s
es
m
u
lti
-
s
ca
le
f
ea
tu
r
es
th
r
o
u
g
h
b
id
ir
ec
ti
o
n
al
p
ath
way
s
.
T
h
is
d
esig
n
en
h
an
ce
s
f
ea
t
u
r
e
r
ep
r
esen
tatio
n
a
n
d
e
n
ab
les
ef
f
ec
tiv
e
h
a
n
d
lin
g
o
f
tu
m
o
r
s
ize
v
ar
iab
ilit
y
,
co
m
p
lex
s
tr
u
ctu
r
es,
an
d
h
eter
o
g
en
e
o
u
s
tex
tu
r
es in
b
r
ai
n
MRI
s
ca
n
s
.
Fig
u
r
e
1
.
YOL
Ov
5
m
with
B
iFP
N
an
d
Gr
ad
-
C
AM
v
is
u
al
in
te
r
p
r
etab
ilit
y
f
r
a
m
ewo
r
k
T
h
e
ex
p
er
im
e
n
ts
u
tili
ze
d
th
e
p
u
b
licly
av
ailab
le
b
r
ain
t
u
m
o
r
MRI
d
atas
et
[
3
1
]
f
r
o
m
Kag
g
le,
co
n
s
is
tin
g
o
f
two
-
d
im
en
s
io
n
a
l
MRI
im
ag
es
ca
teg
o
r
ized
in
to
f
o
u
r
class
es:
g
lio
m
a,
m
en
in
g
io
m
a,
p
itu
itar
y
tu
m
o
r
,
a
n
d
n
o
t
u
m
o
r
.
T
h
e
d
at
aset
in
clu
d
es
7
,
0
2
3
im
a
g
es,
d
iv
id
ed
u
s
in
g
an
8
0
%
–
2
0
%
s
p
lit
f
o
r
tr
ain
in
g
an
d
test
in
g
.
T
h
e
tr
ain
in
g
s
et
co
n
tai
n
s
5
,
7
1
2
im
a
g
es,
wh
ile
th
e
test
s
et
in
clu
d
es
1
,
3
1
1
im
ag
es.
T
ab
le
1
s
u
m
m
ar
izes
th
e
d
ataset
f
ea
tu
r
es,
d
escr
ip
tio
n
s
,
an
d
class
-
wis
e
d
is
tr
ib
u
tio
n
.
T
ab
le
1
.
Data
s
et
f
ea
tu
r
es,
d
esc
r
ip
tio
n
s
,
an
d
r
an
g
es
F
e
a
t
u
r
e
D
e
scri
p
t
i
o
n
R
a
n
g
e
/
v
a
l
u
e
s
Tr
a
i
n
i
n
g
A
x
i
a
l
T
1
-
w
e
i
g
h
t
e
d
c
o
n
t
r
a
s
t
-
e
n
h
a
n
c
e
d
M
R
I
sca
n
o
f
t
h
e
b
r
a
i
n
G
r
a
y
sc
a
l
e
o
r
R
G
B
,
r
e
so
l
u
t
i
o
n
v
a
r
i
e
s (r
e
s
i
z
e
d
t
o
2
2
4
×
2
2
4
)
C
l
a
s
s
l
a
b
e
l
Tu
m
o
r
c
a
t
e
g
o
r
y
a
ss
i
g
n
e
d
t
o
e
a
c
h
i
ma
g
e
g
l
i
o
m
a
,
m
e
n
i
n
g
i
o
ma,
p
i
t
u
i
t
a
r
y
,
n
o
t
u
mo
r
D
a
t
a
s
e
t
s
p
l
i
t
I
n
d
i
c
a
t
e
s
w
h
e
t
h
e
r
t
h
e
i
ma
g
e
i
s
p
a
r
t
o
f
t
h
e
t
r
a
i
n
i
n
g
o
r
t
e
st
i
n
g
s
e
t
Tr
a
i
n
i
n
g
,
t
e
s
t
i
n
g
F
i
l
e
f
o
r
ma
t
I
mag
e
f
o
r
m
a
t
o
f
M
R
I
sca
n
s
.
j
p
g
,
.
p
n
g
I
mag
e
si
z
e
D
i
me
n
si
o
n
s
o
f
t
h
e
o
r
i
g
i
n
a
l
i
ma
g
e
s (v
a
r
i
e
s a
c
r
o
ss
sam
p
l
e
s)
V
a
r
i
e
s
(
st
a
n
d
a
r
d
i
z
e
d
t
o
2
2
4
×
2
2
4
d
u
r
i
n
g
p
r
e
p
r
o
c
e
ss
i
n
g
)
I
n
t
e
n
s
i
t
y
v
a
l
u
e
s
P
i
x
e
l
i
n
t
e
n
si
t
y
v
a
l
u
e
s
o
f
M
R
I
i
ma
g
e
s
0
–
2
5
5
(
n
o
r
m
a
l
i
z
e
d
t
o
-
1
t
o
1
d
u
r
i
n
g
p
r
e
p
r
o
c
e
ss
i
n
g
)
T
o
en
h
a
n
ce
g
e
n
er
aliza
tio
n
an
d
r
e
d
u
ce
o
v
er
f
itti
n
g
,
c
o
m
p
r
eh
e
n
s
iv
e
d
ata
p
r
ep
r
o
c
ess
in
g
an
d
au
g
m
en
tatio
n
s
tr
ateg
ies
wer
e
ap
p
lied
.
All
MRI
im
ag
es
wer
e
r
esized
to
2
2
4
×2
2
4
p
ix
els
t
o
s
atis
f
y
th
e
in
p
u
t
r
eq
u
ir
em
e
n
ts
o
f
th
e
YOL
Ov
5
m
b
ac
k
b
o
n
e
.
Du
r
in
g
tr
ain
in
g
,
d
ata
a
u
g
m
e
n
tatio
n
tec
h
n
iq
u
e
s
in
clu
d
in
g
r
a
n
d
o
m
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
n
o
p
timiz
ed
d
ee
p
lea
r
n
in
g
fr
a
mewo
r
k
fo
r
b
r
a
in
tu
mo
r
cla
s
s
ifica
tio
n
u
s
in
g
…
(
K
o
ma
l Ku
ma
r
N
a
p
a
)
3725
h
o
r
izo
n
tal
f
lip
p
in
g
,
r
o
tatio
n
s
with
in
±
1
5
°,
an
d
b
r
i
g
h
tn
es
s
an
d
co
n
tr
ast
ad
ju
s
tm
en
ts
wer
e
em
p
lo
y
e
d
to
s
im
u
late
v
ar
iatio
n
s
c
o
m
m
o
n
ly
o
b
s
er
v
ed
in
MRI
ac
q
u
is
itio
n
.
Pix
el
v
alu
es
wer
e
n
o
r
m
a
lized
to
t
h
e
r
a
n
g
e
[
–
1
,
1
]
u
s
in
g
a
m
ea
n
a
n
d
s
tan
d
ar
d
d
ev
iatio
n
o
f
0
.
5
,
w
h
ich
e
n
s
u
r
ed
n
u
m
er
ical
s
tab
ilit
y
an
d
f
aster
co
n
v
er
g
en
ce
.
Fo
r
th
e
test
s
et,
o
n
ly
r
esizin
g
an
d
n
o
r
m
aliza
tio
n
wer
e
a
p
p
lie
d
to
m
ain
tain
e
v
alu
atio
n
co
n
s
i
s
ten
cy
.
T
o
ad
d
r
ess
class
im
b
alan
ce
,
a
class
-
weig
h
ted
cr
o
s
s
-
en
tr
o
p
y
lo
s
s
b
ased
o
n
in
v
e
r
s
e
class
f
r
eq
u
en
cies
was
u
s
ed
.
T
h
e
m
o
d
el
was
o
p
tim
ized
u
s
in
g
th
e
A
d
a
m
o
p
tim
izer
with
a
n
in
itial
lear
n
in
g
r
ate
o
f
1
×1
0
⁻⁴,
an
d
a
s
tep
-
b
ased
lear
n
in
g
r
ate
s
ch
ed
u
ler
th
at
r
e
d
u
ce
d
t
h
e
lear
n
in
g
r
ate
b
y
h
alf
ev
er
y
f
iv
e
ep
o
ch
s
to
p
r
o
m
o
te
s
ta
b
le
tr
ain
in
g
.
B
atch
n
o
r
m
aliza
tio
n
lay
e
r
s
f
u
r
th
e
r
c
o
n
tr
ib
u
ted
to
r
eg
u
lar
izatio
n
a
n
d
im
p
r
o
v
ed
c
o
n
v
er
g
en
ce
.
T
o
ev
alu
ate
r
o
b
u
s
tn
ess
an
d
g
e
n
er
aliza
b
ilit
y
,
1
0
-
f
o
ld
c
r
o
s
s
-
v
alid
atio
n
was
p
e
r
f
o
r
m
ed
,
an
d
p
er
f
o
r
m
a
n
ce
m
et
r
ics
wer
e
av
er
a
g
ed
ac
r
o
s
s
f
o
ld
s
.
Mo
d
el
ev
alu
atio
n
wa
s
co
n
d
u
cte
d
u
s
in
g
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
an
d
F1
-
s
co
r
e,
alo
n
g
with
co
m
p
u
tatio
n
al
tim
e
a
n
aly
s
is
.
T
h
e
d
ataset
was
l
o
ad
ed
u
s
in
g
Py
T
o
r
c
h
’
s
I
m
ag
eFo
ld
er
u
tili
ty
an
d
p
r
o
ce
s
s
ed
i
n
m
in
i
-
b
atch
es to
en
a
b
le
ef
f
icie
n
t
g
r
ap
h
ics p
r
o
ce
s
s
in
g
u
n
it (
GPU
)
-
ac
ce
ler
ated
tr
ain
in
g
.
C
o
m
p
u
tatio
n
al
ef
f
icien
cy
was
ass
ess
ed
o
n
a
Go
o
g
le
C
o
lab
en
v
ir
o
n
m
en
t
e
q
u
ip
p
e
d
with
a
n
NVI
DI
A
T
esla
T
4
GPU
an
d
1
6
GB
R
A
M.
T
h
e
m
o
d
el
was
tr
ain
ed
f
o
r
2
0
ep
o
ch
s
with
a
b
atch
s
ize
o
f
3
2
,
ac
h
iev
in
g
an
av
er
ag
e
tr
ai
n
in
g
tim
e
o
f
6
2
4
.
2
4
s
ec
o
n
d
s
p
er
e
p
o
ch
an
d
a
t
o
tal
tr
ain
in
g
tim
e
o
f
1
2
,
4
8
4
.
8
1
s
ec
o
n
d
s
.
Du
r
in
g
in
f
er
en
ce
,
th
e
m
o
d
el
r
eq
u
ir
ed
1
4
6
.
8
2
s
ec
o
n
d
s
to
p
r
o
ce
s
s
th
e
f
u
ll
test
s
et
o
f
1
,
3
1
1
im
ag
es,
c
o
r
r
esp
o
n
d
in
g
to
a
n
av
er
ag
e
in
f
er
e
n
ce
tim
e
o
f
0
.
1
1
s
ec
o
n
d
s
p
er
im
ag
e.
Ad
d
itio
n
ally
,
Gr
a
d
-
C
AM
was
in
teg
r
a
ted
to
p
r
o
v
id
e
v
is
u
al
in
ter
p
r
etab
ilit
y
b
y
h
ig
h
lig
h
ti
n
g
d
is
cr
im
in
ativ
e
tu
m
o
r
r
e
g
io
n
s
in
f
lu
e
n
cin
g
p
r
ed
ictio
n
s
.
A
g
r
ad
io
-
b
ased
in
ter
f
ac
e
was
d
ev
elo
p
ed
to
en
ab
le
r
ea
l
-
tim
e
tu
m
o
r
class
if
icatio
n
an
d
v
is
u
aliza
tio
n
,
s
u
p
p
o
r
tin
g
clin
ical
u
s
ab
ilit
y
.
Ov
er
all,
th
e
p
r
o
p
o
s
e
d
f
r
a
m
ewo
r
k
ac
h
iev
es
a
s
tr
o
n
g
b
alan
ce
b
etwe
en
ac
cu
r
ac
y
,
i
n
ter
p
r
etab
ilit
y
,
an
d
co
m
p
u
tatio
n
al
e
f
f
icien
cy
,
m
ak
in
g
it su
itab
le
f
o
r
clin
ical
d
ec
i
s
io
n
s
u
p
p
o
r
t a
p
p
licatio
n
s
.
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
4
.
1
.
P
r
o
po
s
ed
m
o
del us
ing
YO
L
O
v
5
m
wit
h B
iFP
N
T
h
is
s
tu
d
y
p
r
esen
ts
an
ef
f
ic
ien
t
DL
f
r
a
m
ewo
r
k
f
o
r
m
u
l
ti
-
class
b
r
ain
tu
m
o
r
class
if
icatio
n
th
at
in
teg
r
ates
th
e
YOL
Ov
5
m
ar
ch
itectu
r
e
with
a
B
iFP
N.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
is
d
esig
n
ed
to
ad
d
r
ess
k
ey
ch
allen
g
es
in
MRI
-
b
ased
tu
m
o
r
an
al
y
s
is
,
in
clu
d
in
g
s
m
all
lesi
o
n
s
izes,
h
eter
o
g
en
eo
u
s
tu
m
o
r
m
o
r
p
h
o
lo
g
y
,
lo
w
co
n
tr
ast,
an
d
class
im
b
alan
ce
.
YOL
Ov
5
m
s
er
v
es
as
a
lig
h
tweig
h
t
y
et
p
o
wer
f
u
l
b
ac
k
b
o
n
e,
en
ab
lin
g
e
f
f
ec
tiv
e
f
ea
tu
r
e
ex
tr
ac
tio
n
an
d
r
ea
l
-
tim
e
in
f
er
en
ce
,
wh
ile
B
iFP
N
en
h
an
ce
s
m
u
lti
-
s
ca
le
f
ea
tu
r
e
f
u
s
io
n
to
im
p
r
o
v
e
lo
ca
lizatio
n
an
d
d
is
cr
im
in
atio
n
o
f
c
o
m
p
lex
t
u
m
o
r
s
tr
u
ctu
r
es
ac
r
o
s
s
v
ar
y
in
g
r
eso
lu
tio
n
s
.
Un
lik
e
tr
ad
itio
n
al
C
NN
o
r
p
atch
-
b
ased
a
p
p
r
o
ac
h
es
th
at
m
ay
lo
s
e
c
o
n
tex
tu
al
in
f
o
r
m
atio
n
,
th
e
YOL
Ov
5
m
–
B
iFP
N
f
r
am
ewo
r
k
p
r
o
ce
s
s
es
f
u
ll
MRI
im
ag
es
an
d
lev
er
ag
es
h
ier
a
r
ch
ical
f
ea
tu
r
e
r
ep
r
esen
tatio
n
s
to
p
r
eser
v
e
b
o
t
h
g
l
o
b
al
a
n
d
lo
ca
l
co
n
tex
t.
T
h
e
weig
h
ted
f
u
s
io
n
m
ec
h
an
is
m
in
B
iFP
N
f
u
r
th
e
r
im
p
r
o
v
es
r
o
b
u
s
tn
ess
to
s
ca
le
v
ar
iatio
n
an
d
tu
m
o
r
h
eter
o
g
en
eity
.
E
x
p
e
r
im
en
tal
r
esu
lts
d
em
o
n
s
tr
ate
s
tr
o
n
g
p
e
r
f
o
r
m
an
ce
,
ac
h
iev
in
g
a
test
ac
cu
r
ac
y
o
f
8
8
.
8
6
%,
a
n
F1
-
s
co
r
e
o
f
8
8
.
2
5
%,
an
d
a
m
AP@
0
.
5
o
f
9
4
.
3
6
%.
T
h
e
lo
w
in
f
e
r
en
ce
laten
cy
an
d
m
o
d
er
ate
co
m
p
u
tatio
n
al
co
s
t
co
n
f
ir
m
th
e
m
o
d
el’
s
s
u
itab
ilit
y
f
o
r
r
ea
l
-
tim
e
clin
ical
d
ec
is
io
n
s
u
p
p
o
r
t sy
s
tem
s
.
As
s
h
o
wn
in
T
ab
le
2
,
th
e
Y
OL
Ov
5
m
+
B
iFP
N
m
o
d
el
ex
h
ib
its
b
alan
ce
d
class
-
wis
e
p
er
f
o
r
m
a
n
ce
.
I
t
ac
h
iev
es
ex
ce
llen
t
r
esu
lts
f
o
r
h
ea
lth
y
ca
s
es
(
p
r
ec
is
io
n
=
0
.
9
4
,
r
ec
all
=
0
.
9
6
)
an
d
p
itu
itar
y
tu
m
o
r
s
(
p
r
ec
is
io
n
=
0
.
8
8
,
r
ec
all
=
0
.
9
7
)
,
in
d
icatin
g
h
ig
h
r
eliab
ilit
y
an
d
m
in
im
al
m
is
s
ed
d
etec
tio
n
s
.
Me
n
in
g
io
m
a
s
h
o
ws
m
o
d
er
ate
p
e
r
f
o
r
m
an
ce
(
p
r
ec
is
io
n
=
0
.
7
5
,
r
ec
all
=
0
.
8
3
)
,
lik
ely
d
u
e
to
v
is
u
al
s
im
ilar
ity
with
o
th
er
tu
m
o
r
s
.
Glio
m
a
ac
h
iev
es
h
i
g
h
p
r
ec
is
io
n
(
0
.
9
5
)
b
u
t
co
m
p
ar
at
iv
ely
lo
wer
r
ec
all
(
0
.
7
3
)
,
in
d
i
ca
tin
g
th
e
p
r
esen
ce
o
f
o
cc
asio
n
al
f
alse
n
eg
ativ
es.
Fro
m
a
clin
ical
p
er
s
p
ec
tiv
e,
m
is
s
ed
g
lio
m
a
ca
s
es
ar
e
p
ar
ti
cu
lar
ly
cr
itical,
as
d
elay
ed
d
etec
tio
n
m
ay
a
f
f
ec
t
t
r
ea
tm
en
t
p
lan
n
in
g
an
d
p
atien
t
o
u
tco
m
es.
Ho
we
v
er
,
t
h
e
h
i
g
h
p
r
ec
is
io
n
in
d
icate
s
th
at
g
lio
m
a
p
r
ed
ictio
n
s
m
ad
e
b
y
th
e
m
o
d
el
ar
e
h
ig
h
ly
r
eliab
le
wh
en
d
etec
ted
.
T
h
er
e
f
o
r
e,
th
e
p
r
o
p
o
s
ed
f
r
am
ewo
r
k
is
b
est
s
u
ited
as
a
clin
ical
d
ec
is
io
n
-
s
u
p
p
o
r
t
to
o
l,
ass
is
t
in
g
r
ad
io
lo
g
is
ts
b
y
h
ig
h
lig
h
tin
g
s
u
s
p
icio
u
s
r
eg
io
n
s
r
ath
er
th
an
r
ep
lacin
g
ex
p
e
r
t
ju
d
g
m
en
t.
m
ac
r
o
-
an
d
weig
h
ted
-
av
er
a
g
e
F1
-
s
co
r
e
s
(
0
.
8
7
a
n
d
0
.
8
8
)
co
n
f
ir
m
b
alan
ce
d
lea
r
n
in
g
ac
r
o
s
s
class
e
s
.
T
ab
le
2
.
Per
f
o
r
m
an
ce
m
etr
ics
C
l
a
s
s
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
s
c
o
r
e
S
u
p
p
o
r
t
G
l
i
o
ma
0
.
9
5
0
.
7
3
0
.
8
2
3
0
0
M
e
n
i
n
g
i
o
ma
0
.
7
5
0
.
8
3
0
.
7
9
3
0
6
N
o
t
u
m
o
r
0
.
9
4
0
.
9
6
0
.
9
5
4
0
5
P
i
t
u
i
t
a
r
y
0
.
8
8
0
.
9
7
0
.
9
2
3
0
0
A
c
c
u
r
a
c
y
0
.
8
8
1
3
1
1
M
a
c
r
o
av
g
0
.
8
8
0
.
8
7
0
.
8
7
1
3
1
1
W
e
i
g
h
t
e
d
a
v
g
0
.
8
9
0
.
8
8
0
.
8
8
1
3
1
1
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
4
,
Au
g
u
s
t
20
26
:
3
7
2
2
-
3
7
3
1
3726
T
o
v
alid
ate
th
e
p
r
o
p
o
s
ed
f
r
a
m
ewo
r
k
,
a
co
m
p
a
r
ativ
e
an
al
y
s
is
(
T
ab
le
3
)
was
p
er
f
o
r
m
ed
ag
ain
s
t
C
NN
–
tr
an
s
f
o
r
m
er
b
aselin
es,
in
clu
d
in
g
Den
s
eNe
t1
2
1
+
ViT
,
Den
s
eNe
t2
0
1
+
ViT
,
I
n
ce
p
t
io
n
V3
+
ViT
,
an
d
Xce
p
tio
n
+
ViT
,
u
s
in
g
id
e
n
tical
d
ata
s
p
lits
.
Den
s
eNe
t
-
b
ased
m
o
d
els
ac
h
iev
e
d
th
e
lo
wer
ac
c
u
r
ac
ies
(
6
9
%
an
d
7
4
%),
wh
ile
I
n
ce
p
ti
o
n
V3
+
ViT
an
d
Xce
p
tio
n
+
ViT
r
ea
ch
ed
8
2
%
a
n
d
8
7
%
ac
cu
r
ac
y
,
r
esp
ec
tiv
ely
.
T
h
e
p
r
o
p
o
s
ed
YOL
Ov
5
m
-
B
iFP
N
o
u
tp
er
f
o
r
m
ed
all
b
aselin
es
with
8
8
%
ac
c
u
r
ac
y
an
d
a
n
F1
-
s
co
r
e
o
f
0
.
8
8
,
attr
ib
u
ted
to
ef
f
ec
tiv
e
m
u
lti
-
s
c
ale
f
ea
tu
r
e
f
u
s
io
n
.
T
ab
le
3
.
Per
f
o
r
m
an
ce
co
m
p
a
r
is
o
n
with
b
aselin
e
DL
m
o
d
els
M
o
d
e
l
A
c
c
u
r
a
c
y
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
sc
o
r
e
D
e
n
seN
e
t
1
2
1
+
V
i
T
0
.
6
9
0
.
8
1
0
.
6
7
0
.
6
1
D
e
n
seN
e
t
2
0
1
+
V
i
T
0
.
7
4
0
.
8
3
0
.
7
1
0
.
6
4
I
n
c
e
p
t
i
o
n
V
3
+
V
i
T
0
.
8
2
0
.
8
8
0
.
8
0
0
.
8
2
X
c
e
p
t
i
o
n
+
V
i
T
0
.
8
7
0
.
8
8
0
.
8
6
0
.
8
7
Y
O
LO
v
5
m
–
B
i
F
P
N
(
P
r
o
p
o
se
d
)
0
.
8
8
0
.
8
9
0
.
8
8
0
.
8
8
T
o
v
e
r
if
y
th
e
s
t
a
t
i
s
t
i
c
a
l
s
i
g
n
i
f
i
c
a
n
ce
o
f
th
e
p
r
o
p
o
s
e
d
YO
L
O
v
5
m
–
B
i
F
P
N
m
o
d
e
l
’
s
p
e
r
f
o
r
m
an
c
e
g
a
i
n
s
,
a
p
a
ir
e
d
t
-
t
e
s
t
w
a
s
co
n
d
u
c
t
ed
u
s
in
g
r
e
s
u
l
t
s
f
r
o
m
1
0
-
f
o
ld
cr
o
s
s
-
v
a
l
i
d
a
t
io
n
.
Me
a
n
a
c
cu
r
a
c
y
a
n
d
F1
-
s
c
o
r
e
v
a
lu
e
s
a
c
r
o
s
s
f
o
ld
s
w
e
r
e
c
o
m
p
ar
e
d
a
t
a
s
ig
n
i
f
ic
a
n
c
e
l
ev
e
l
o
f
p
<
0
.
0
5
.
T
h
e
p
r
o
p
o
s
ed
m
o
d
e
l
a
c
h
i
ev
e
d
s
t
a
t
i
s
t
i
c
a
l
ly
s
i
g
n
if
ic
a
n
t
i
m
p
r
o
v
em
e
n
t
s
i
n
ac
c
u
r
a
c
y
o
v
e
r
D
en
s
e
N
e
t
1
2
1
–
V
i
T
(
p
=
0
.
0
0
3
)
a
n
d
I
n
c
ep
t
i
o
n
V
3
–
V
i
T
(
p
=
0
.
0
1
2
)
,
w
i
t
h
c
o
r
r
e
s
p
o
n
d
in
g
s
i
g
n
i
f
i
ca
n
t
g
a
in
s
i
n
F
1
-
s
c
o
r
e
(
p
=
0
.
0
0
4
an
d
p
=
0
.
0
1
8
)
.
P
e
r
f
o
r
m
an
c
e
i
m
p
r
o
v
e
m
e
n
t
s
o
v
e
r
X
c
ep
t
i
o
n
–
V
i
T
w
e
r
e
a
l
s
o
s
i
g
n
i
f
i
ca
n
t
(
p
=
0
.
0
4
1
)
,
co
n
f
i
r
m
i
n
g
th
a
t
t
h
e
g
a
i
n
s
s
t
e
m
f
r
o
m
t
h
e
ef
f
e
c
t
i
v
en
e
s
s
o
f
t
h
e
B
i
F
P
N
-
en
h
an
c
ed
m
u
l
t
i
-
s
c
a
le
f
e
a
tu
r
e
f
u
s
i
o
n
r
a
th
er
th
a
n
r
a
n
d
o
m
v
ar
i
a
t
io
n
.
T
h
e
d
etec
tio
n
p
er
f
o
r
m
an
ce
is
f
u
r
t
h
er
s
u
p
p
o
r
ted
b
y
th
e
s
tr
o
n
g
m
AP@
0
.
5
r
esu
lts
s
h
o
wn
i
n
T
ab
le
4
.
T
h
e
m
o
d
el
ac
h
iev
ed
a
n
o
v
er
all
m
AP@
0
.
5
o
f
0
.
9
4
,
in
d
icatin
g
h
ig
h
lo
ca
lizatio
n
an
d
class
if
icatio
n
ca
p
ab
ilit
y
ac
r
o
s
s
tu
m
o
r
class
es.
T
h
e
n
o
-
tu
m
o
r
class
r
ec
o
r
d
ed
th
e
h
ig
h
est
a
v
er
ag
e
p
r
ec
is
io
n
(
AP
)
(
0
.
9
8
)
,
f
o
llo
wed
b
y
p
itu
itar
y
tu
m
o
r
s
(
0
.
9
7
)
,
r
e
f
lectin
g
co
n
s
is
ten
t
s
tr
u
ctu
r
al
p
atter
n
s
.
Glio
m
a
ac
h
ie
v
ed
a
n
AP
o
f
0
.
9
4
,
wh
ile
m
en
in
g
io
m
a
s
h
o
wed
a
lo
wer
AP
(
0
.
8
6
)
,
lik
el
y
d
u
e
to
h
ig
h
e
r
in
tr
a
-
class
v
ar
iab
il
ity
an
d
o
v
er
lap
p
in
g
v
is
u
al
f
ea
tu
r
es.
T
ab
le
4
.
m
AP
a
t I
o
U
th
r
esh
o
l
d
0
.
5
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el
I
o
U
t
h
r
e
s
h
o
l
d
0
.
5
o
f
t
h
e
p
r
o
p
o
se
d
mo
d
e
l
AP
mA
P
@
0
.
5
(
mu
l
t
i
-
c
l
a
ss)
0
.
9
4
A
P
@
0
.
5
f
o
r
g
l
i
o
m
a
0
.
9
4
A
P
@
0
.
5
f
o
r
m
e
n
i
n
g
i
o
m
a
0
.
8
6
A
P
@
0
.
5
f
o
r
n
o
t
u
mo
r
0
.
9
8
A
P
@
0
.
5
f
o
r
p
i
t
u
i
t
a
r
y
0
.
9
7
O
v
e
r
a
l
l,
th
e
co
n
s
i
s
t
en
t
l
y
h
ig
h
p
e
r
f
o
r
m
a
n
c
e
ac
r
o
s
s
c
l
a
s
s
i
f
i
c
a
t
i
o
n
an
d
d
e
t
ec
t
i
o
n
m
e
t
r
i
c
s
v
a
l
id
a
t
e
s
t
h
e
e
f
f
e
c
t
i
v
en
e
s
s
o
f
t
h
e
B
i
FP
N
-
e
n
h
a
n
c
ed
Y
O
L
O
v
5
m
f
r
am
e
w
o
r
k
i
n
e
x
tr
a
c
t
in
g
c
l
a
s
s
-
s
p
e
c
i
f
i
c
s
p
a
t
i
a
l
a
n
d
s
e
m
a
n
t
i
c
cu
e
s
f
r
o
m
b
r
ai
n
MR
I
s
c
a
n
s
,
h
ig
h
l
ig
h
t
in
g
i
t
s
r
o
b
u
s
t
n
e
s
s
,
r
e
l
ia
b
i
l
i
ty
,
an
d
c
l
in
i
ca
l
a
p
p
l
ic
a
b
i
l
i
ty
.
An
e
r
r
o
r
a
n
a
ly
s
i
s
w
a
s
co
n
d
u
c
t
ed
b
y
r
ev
i
e
w
i
n
g
r
e
p
r
e
s
e
n
ta
t
i
v
e
m
i
s
c
l
a
s
s
i
f
i
ed
M
R
I
s
a
m
p
l
e
s
f
r
o
m
t
h
e
t
e
s
t
s
e
t.
M
o
s
t
e
r
r
o
r
s
o
c
cu
r
r
e
d
b
e
t
w
e
en
g
l
i
o
m
a
a
n
d
m
en
i
n
g
i
o
m
a
d
u
e
t
o
o
v
e
r
la
p
p
i
n
g
v
i
s
u
a
l
ch
ar
a
c
t
e
r
i
s
t
i
c
s
s
u
ch
a
s
s
i
m
i
l
a
r
i
n
te
n
s
i
t
y
p
a
t
t
er
n
s
,
t
u
m
o
r
lo
c
a
t
io
n
s
,
a
n
d
i
r
r
e
g
u
l
ar
b
o
u
n
d
a
r
i
e
s
.
G
l
i
o
m
a
e
x
h
i
b
i
ted
h
ig
h
p
r
ec
i
s
i
o
n
b
u
t
l
o
w
e
r
r
e
c
a
l
l
,
in
d
i
c
a
t
in
g
th
a
t
c
a
s
e
s
w
i
t
h
h
e
t
e
r
o
g
en
eo
u
s
te
x
t
u
r
e
s
o
r
d
i
f
f
u
s
e
m
ar
g
in
s
w
e
r
e
o
c
ca
s
i
o
n
a
l
l
y
m
i
s
s
e
d
.
S
o
m
e
m
e
n
in
g
io
m
a
ca
s
e
s
w
e
r
e
co
n
f
u
s
e
d
w
i
t
h
p
i
t
u
i
t
a
r
y
tu
m
o
r
s
b
e
c
a
u
s
e
o
f
c
o
m
p
a
r
ab
l
e
s
h
a
p
e
a
n
d
c
o
n
tr
a
s
t
en
h
an
c
e
m
e
n
t
.
Gr
ad
-
C
AM
an
aly
s
is
o
f
m
is
class
if
ied
s
am
p
les
s
h
o
wed
th
at
th
e
m
o
d
el
co
n
s
is
ten
tly
f
o
cu
s
ed
o
n
clin
ically
r
elev
an
t
tu
m
o
r
r
eg
i
o
n
s
,
s
u
g
g
esti
n
g
t
h
at
er
r
o
r
s
wer
e
p
r
im
ar
ily
d
u
e
to
in
ter
-
class
f
ea
tu
r
e
am
b
ig
u
ity
r
ath
er
th
an
in
co
r
r
ec
t
s
p
atial
atten
tio
n
.
Qu
alitativ
e
ev
alu
atio
n
f
u
r
th
e
r
r
ev
ea
led
s
tr
o
n
g
alig
n
m
en
t
b
etwe
en
Gr
ad
-
C
AM
h
ea
tm
ap
s
an
d
v
is
i
b
le
tu
m
o
r
r
eg
io
n
s
,
p
ar
ticu
lar
ly
h
ig
h
lig
h
tin
g
h
eter
o
g
en
e
o
u
s
co
r
es a
n
d
in
f
iltra
tiv
e
m
ar
g
in
s
in
g
lio
m
a
ca
s
es.
E
v
e
n
wh
en
m
is
class
if
ied
,
atten
tio
n
r
em
ai
n
ed
o
n
p
lau
s
ib
le
p
ath
o
lo
g
ical
ar
ea
s
.
T
h
ese
o
b
s
er
v
atio
n
s
in
d
icate
th
at
in
teg
r
atin
g
m
u
lti
-
m
o
d
al
MRI
s
eq
u
en
ce
s
o
r
3
D
s
p
atial
co
n
tex
t
co
u
ld
f
u
r
t
h
er
r
ed
u
ce
m
is
class
if
icatio
n
an
d
im
p
r
o
v
e
r
o
b
u
s
tn
ess
.
T
ab
le
5
d
e
p
icts
th
e
co
m
p
ar
is
o
n
o
f
t
h
e
p
r
o
p
o
s
ed
m
o
d
el
with
p
r
ev
io
u
s
s
tu
d
ies.
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
n
o
p
timiz
ed
d
ee
p
lea
r
n
in
g
fr
a
mewo
r
k
fo
r
b
r
a
in
tu
mo
r
cla
s
s
ifica
tio
n
u
s
in
g
…
(
K
o
ma
l Ku
ma
r
N
a
p
a
)
3727
T
ab
le
5
.
Per
f
o
r
m
an
ce
m
etr
ics co
m
p
ar
is
o
n
with
p
r
ev
io
u
s
s
tu
d
ies
S
t
u
d
y
D
a
t
a
s
e
t
C
l
a
s
si
f
i
e
r
A
c
c
u
r
a
c
y
(
%)
Tu
h
i
n
e
t
a
l
.
[
3
2
]
R
e
a
l
-
t
i
me
M
R
I
d
a
t
a
se
t
C
N
N
a
n
d
3
D
C
N
N
8
5
.
0
S
u
t
e
r
e
t
a
l
.
[
3
3
]
B
r
a
TS
2
0
1
8
S
V
M
w
i
t
h
C
N
N
7
2
.
2
0
H
u
a
n
d
X
i
a
[
3
4
]
B
r
a
TS
2
0
1
7
c
h
a
l
l
e
n
g
e
3D
-
b
a
s
e
d
C
N
N
8
1
.
4
0
P
r
e
sen
t
s
t
u
d
y
K
a
g
g
l
e
M
R
I
Y
O
LO
v
5
m
w
i
t
h
B
i
F
P
N
8
8
.
8
6
4
.
2
.
G
ra
dient
-
weig
hte
d c
la
s
s
a
ct
iv
a
t
io
n
m
a
pp
ing
v
is
ua
l int
er
pret
a
bil
it
y
I
n
ter
p
r
etab
ilit
y
is
ess
en
tial
in
m
ed
ical
im
a
g
e
an
al
y
s
is
to
b
u
ild
clin
ician
tr
u
s
t
an
d
en
s
u
r
e
tr
an
s
p
ar
en
cy
in
ar
tific
ial
in
te
llig
en
ce
(
AI
)
-
ass
is
ted
d
iag
n
o
s
is
.
T
o
ad
d
r
ess
th
is
,
th
e
p
r
o
p
o
s
ed
YOL
Ov
5
m
+
B
iFP
N
m
o
d
el
in
co
r
p
o
r
ates
Gr
ad
-
C
AM
to
p
r
o
v
id
e
v
is
u
al
ex
p
lan
atio
n
s
o
f
its
p
r
ed
i
ctio
n
s
.
Gr
ad
-
C
AM
g
en
er
ates
h
ea
tm
ap
s
th
at
h
ig
h
l
ig
h
t
th
e
m
o
s
t
in
f
lu
en
tial
r
eg
i
o
n
s
in
an
in
p
u
t
MRI
s
ca
n
,
allo
win
g
clin
ician
s
to
v
er
if
y
wh
eth
er
th
e
m
o
d
el’
s
d
ec
is
io
n
s
ar
e
b
ased
o
n
clin
ically
m
ea
n
in
g
f
u
l
f
ea
tu
r
es.
As
illu
s
tr
ated
in
Fig
u
r
e
2
,
Gr
ad
-
C
AM
is
ap
p
lied
to
th
e
B
iFP
N
lay
er
,
wh
ich
ca
p
tu
r
es
h
ig
h
-
lev
el
m
u
lti
-
s
ca
le
r
ep
r
esen
tatio
n
s
ac
r
o
s
s
d
if
f
er
en
t sp
atial
r
eso
lu
tio
n
s
.
Fig
u
r
e
2
.
Pro
ce
s
s
o
f
Gr
ad
-
C
AM
X
AI
Du
r
in
g
i
n
f
er
en
ce
,
g
r
ad
ien
ts
o
f
th
e
p
r
ed
icted
class
with
r
esp
ec
t
to
B
iFP
N
f
ea
tu
r
e
m
a
p
s
ar
e
av
er
ag
e
d
to
co
m
p
u
te
f
ea
tu
r
e
im
p
o
r
tan
c
e,
p
r
o
d
u
cin
g
class
-
s
p
ec
if
ic
Gr
ad
-
C
AM
h
ea
tm
ap
s
o
v
e
r
laid
o
n
MRI
s
ca
n
s
.
T
h
ese
v
is
u
aliza
tio
n
s
alig
n
well
wit
h
clin
ically
r
elev
an
t
tu
m
o
r
r
eg
io
n
s
,
esp
ec
ially
in
g
lio
m
a
an
d
p
itu
itar
y
ca
s
es,
co
n
f
ir
m
in
g
an
at
o
m
ically
m
ea
n
in
g
f
u
l
atten
tio
n
.
A
g
r
ad
io
in
ter
f
ac
e
f
u
r
th
er
en
a
b
les
r
ea
l
-
t
im
e
p
r
ed
ictio
n
an
d
v
is
u
aliza
tio
n
.
Fig
u
r
e
3
s
h
o
ws th
e
in
p
u
t M
R
I
s
ca
n
o
f
a
g
lio
m
a
,
m
en
in
g
i
o
m
a
an
d
p
itu
itar
y
ca
s
es
.
Fig
u
r
e
3
(
a)
s
h
o
ws
th
e
in
p
u
t
M
R
I
s
ca
n
o
f
a
g
lio
m
a
ca
s
e,
wh
il
e
Fig
u
r
e
4
p
r
esen
ts
th
e
co
r
r
es
p
o
n
d
in
g
p
r
ed
ictio
n
an
d
Gr
ad
-
C
AM
v
is
u
aliza
tio
n
g
en
er
ate
d
b
y
t
h
e
YOL
Ov
5
m
+
B
iFP
N
m
o
d
el.
T
h
e
m
o
d
el
p
r
ed
icts
g
lio
m
a
with
9
5
%
co
n
f
id
en
ce
,
an
d
th
e
Gr
ad
-
C
AM
h
ea
tm
ap
h
ig
h
lig
h
ts
th
e
ce
n
tr
al
tu
m
o
r
r
eg
io
n
,
c
o
n
f
ir
m
i
n
g
atten
tio
n
to
clin
ically
r
elev
an
t
f
ea
tu
r
es.
Fig
u
r
e
3
(
b
)
s
h
o
ws
th
e
in
p
u
t
MRI
s
ca
n
o
f
a
m
e
n
in
g
io
m
a
ca
s
e,
w
h
ile
Fig
u
r
e
5
p
r
esen
ts
th
e
co
r
r
esp
o
n
d
in
g
p
r
ed
ictio
n
s
an
d
Gr
ad
-
C
AM
v
is
u
aliza
tio
n
s
.
T
h
e
YO
L
Ov
5
m
+
B
iFP
N
m
o
d
el
p
r
ed
icts
m
e
n
in
g
io
m
a
with
6
8
%
c
o
n
f
id
e
n
ce
,
a
n
d
th
e
Gr
ad
-
C
AM
h
ea
tm
a
p
h
ig
h
lig
h
ts
th
e
well
-
d
ef
i
n
ed
lesi
o
n
in
th
e
r
ig
h
t
h
em
is
p
h
e
r
e
,
co
n
f
ir
m
in
g
atten
tio
n
to
clin
i
ca
lly
r
elev
an
t
tu
m
o
r
r
e
g
io
n
s
.
Fig
u
r
e
3
(
c)
s
h
o
ws
th
e
in
p
u
t
MRI
s
ca
n
o
f
a
p
itu
itar
y
tu
m
o
r
,
wh
ile
Fig
u
r
e
6
p
r
esen
ts
th
e
co
r
r
esp
o
n
d
i
n
g
p
r
e
d
ictio
n
s
an
d
Gr
ad
-
C
AM
v
is
u
aliza
tio
n
s
.
T
h
e
YOL
Ov
5
m
+
B
iFP
N
m
o
d
el
class
if
ies
th
e
s
ca
n
as
a
p
itu
itar
y
tu
m
o
r
with
6
4
%
co
n
f
id
en
ce
,
an
d
th
e
Gr
a
d
-
C
AM
h
ea
tm
ap
h
i
g
h
lig
h
ts
th
e
lo
wer
ce
n
tr
al
b
r
ain
r
eg
io
n
,
alig
n
in
g
with
th
e
an
ato
m
ical
lo
ca
tio
n
o
f
th
e
p
itu
itar
y
g
lan
d
.
T
h
ese
r
esu
lts
em
p
h
asize
th
e
m
o
d
el’
s
s
en
s
itiv
ity
to
b
o
th
th
e
p
r
esen
ce
an
d
a
b
s
en
ce
o
f
p
at
h
o
lo
g
y
,
a
n
d
d
e
m
o
n
s
tr
ate
th
at
e
v
en
in
am
b
i
g
u
o
u
s
s
ce
n
ar
io
s
,
it
ca
n
r
eliab
ly
r
ejec
t
tu
m
o
r
p
r
ed
ictio
n
s
u
n
less
s
ig
n
if
ican
t
p
atter
n
s
ar
e
d
etec
ted
.
T
h
e
v
is
u
al
e
x
p
lan
atio
n
s
p
r
o
v
i
d
ed
b
y
Gr
a
d
-
C
AM
s
u
p
p
o
r
t
th
e
clin
ical
in
ter
p
r
eta
b
ilit
y
o
f
s
u
ch
o
u
tc
o
m
es,
allo
w
in
g
r
a
d
io
lo
g
is
ts
to
v
alid
ate
th
e
m
o
d
el
’
s
r
ea
s
o
n
in
g
an
d
f
u
r
th
er
s
tr
en
g
th
en
in
g
its
p
o
ten
tial f
o
r
a
d
o
p
tio
n
in
r
ea
l
-
w
o
r
ld
d
iag
n
o
s
tic
wo
r
k
f
l
o
ws.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
4
,
Au
g
u
s
t
20
26
:
3
7
2
2
-
3
7
3
1
3728
(
a)
(
b
)
(
c)
Fig
u
r
e
3
.
I
n
p
u
t M
R
I
im
a
g
es o
f
(
a)
g
lio
m
a
,
m
en
in
g
io
m
a,
a
n
d
(
c)
p
itu
itar
y
Fig
u
r
e
4
.
B
r
ain
tu
m
o
r
m
o
d
el
p
r
ed
ictio
n
f
o
r
g
lio
m
a
Fig
u
r
e
5
.
B
r
ain
tu
m
o
r
m
o
d
el
p
r
ed
ictio
n
f
o
r
m
en
i
n
g
io
m
a
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
n
o
p
timiz
ed
d
ee
p
lea
r
n
in
g
fr
a
mewo
r
k
fo
r
b
r
a
in
tu
mo
r
cla
s
s
ifica
tio
n
u
s
in
g
…
(
K
o
ma
l Ku
ma
r
N
a
p
a
)
3729
Fig
u
r
e
6
.
B
r
ain
tu
m
o
r
m
o
d
el
p
r
ed
ictio
n
f
o
r
p
itu
itar
y
5.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
p
r
esen
ts
a
r
o
b
u
s
t
an
d
in
ter
p
r
etab
le
DL
f
r
a
m
ewo
r
k
f
o
r
m
u
lti
-
class
b
r
a
in
tu
m
o
r
class
if
icatio
n
b
ased
o
n
an
en
h
an
ce
d
YOL
Ov
5
m
ar
c
h
itectu
r
e
in
teg
r
ated
with
a
B
iFP
N
.
T
h
e
m
o
d
el
ef
f
ec
tiv
ely
ad
d
r
ess
es
k
ey
ch
allen
g
es
in
b
r
ain
MRI
an
aly
s
is
,
in
clu
d
in
g
tu
m
o
r
h
eter
o
g
en
eity
,
s
ize
v
ar
iatio
n
,
an
d
class
im
b
alan
ce
.
B
iFP
N
f
ac
ilit
ates
ef
f
icien
t
m
u
lti
-
s
ca
le
f
ea
tu
r
e
f
u
s
io
n
,
en
ab
lin
g
th
e
n
etwo
r
k
to
ca
p
tu
r
e
b
o
th
g
lo
b
al
co
n
tex
t
an
d
f
in
e
-
g
r
ain
ed
tu
m
o
r
d
etails,
wh
ile
a
class
-
we
ig
h
ted
lo
s
s
f
u
n
ctio
n
en
s
u
r
es
b
alan
ce
d
lear
n
in
g
ac
r
o
s
s
all
tu
m
o
r
class
es.
T
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
ac
h
iev
ed
a
test
ac
cu
r
ac
y
o
f
8
8
.
8
6
%
an
d
d
e
m
o
n
s
tr
ated
r
elia
b
le
class
-
wi
s
e
p
er
f
o
r
m
an
ce
,
with
an
o
v
er
all
m
AP@
0
.
5
o
f
9
4
.
3
6
%,
r
ef
lectin
g
s
tr
o
n
g
m
u
lti
-
class
d
etec
tio
n
ca
p
ab
ilit
y
.
T
o
im
p
r
o
v
e
tr
a
n
s
p
ar
en
cy
a
n
d
clin
ical
tr
u
s
t,
Gr
ad
-
C
AM
-
b
ased
v
is
u
al
ex
p
lan
atio
n
s
wer
e
in
co
r
p
o
r
ated
t
o
h
i
g
h
lig
h
t
d
i
s
cr
im
in
ativ
e
r
eg
io
n
s
in
f
lu
e
n
cin
g
p
r
ed
ictio
n
s
.
Ad
d
itio
n
all
y
,
a
g
r
ad
i
o
-
b
ased
in
ter
ac
tiv
e
in
ter
f
ac
e
was
d
e
v
e
lo
p
ed
to
en
ab
le
r
ea
l
-
tim
e
p
r
e
d
ictio
n
an
d
v
is
u
aliza
tio
n
.
Ov
er
all,
th
e
f
r
am
ewo
r
k
o
f
f
er
s
a
b
alan
ce
d
co
m
b
in
atio
n
o
f
ac
cu
r
ac
y
,
ef
f
icien
cy
,
a
n
d
ex
p
lain
ab
ilit
y
,
m
ak
in
g
it
s
u
itab
le
f
o
r
clin
ical
d
ec
is
io
n
s
u
p
p
o
r
t.
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
Ko
m
al
Ku
m
ar
Nap
a
✓
✓
✓
✓
✓
✓
✓
✓
✓
R
ajk
u
m
ar
Go
v
in
d
ar
ajan
✓
✓
✓
✓
✓
✓
Sen
th
il M
u
r
u
g
an
J
an
ak
ir
am
an
✓
✓
✓
✓
✓
✓
✓
J
ay
an
th
i A
r
u
m
u
g
am
✓
✓
✓
✓
✓
✓
✓
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.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
4
,
Au
g
u
s
t
20
26
:
3
7
2
2
-
3
7
3
1
3730
DATA AV
AI
L
AB
I
L
I
T
Y
Data
will b
e
av
ailab
le
u
p
o
n
r
e
aso
n
ab
le
r
eq
u
est.
RE
F
E
R
E
NC
E
S
[
1
]
M
.
M
.
A
h
me
d
e
t
a
l
.
,
“
B
r
a
i
n
t
u
mo
r
d
e
t
e
c
t
i
o
n
a
n
d
c
l
a
ssi
f
i
c
a
t
i
o
n
i
n
M
R
I
u
si
n
g
h
y
b
r
i
d
V
i
T
a
n
d
G
R
U
mo
d
e
l
w
i
t
h
e
x
p
l
a
i
n
a
b
l
e
A
I
i
n
S
o
u
t
h
e
r
n
B
a
n
g
l
a
d
e
s
h
,
”
S
c
i
e
n
t
i
f
i
c
R
e
p
o
rt
s
,
v
o
l
.
1
4
,
n
o
.
1
,
p
p
.
1
–
1
2
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
8
-
0
2
4
-
7
1
8
9
3
-
3.
[
2
]
L.
S
u
n
,
L.
Z
h
e
n
g
,
a
n
d
Y
.
X
i
n
,
“
F
A
LS
-
Y
O
LO
:
a
n
e
f
f
i
c
i
e
n
t
a
n
d
l
i
g
h
t
w
e
i
g
h
t
me
t
h
o
d
f
o
r
a
u
t
o
mat
i
c
b
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
a
n
d
seg
m
e
n
t
a
t
i
o
n
,
”
S
e
n
s
o
rs
,
v
o
l
.
2
5
,
n
o
.
1
9
,
2
0
2
5
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
5
1
9
5
9
9
3
.
[
3
]
M
.
N
a
z
i
r
,
S
.
S
h
a
k
i
l
,
a
n
d
K
.
K
h
u
r
sh
i
d
,
“
R
o
l
e
o
f
d
e
e
p
l
e
a
r
n
i
n
g
i
n
b
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
a
n
d
c
l
a
ssi
f
i
c
a
t
i
o
n
(
2
0
1
5
t
o
2
0
2
0
)
:
a
r
e
v
i
e
w
,
”
C
o
m
p
u
t
e
ri
ze
d
M
e
d
i
c
a
l
I
m
a
g
i
n
g
a
n
d
G
r
a
p
h
i
c
s
,
v
o
l
.
9
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
mp
me
d
i
m
a
g
.
2
0
2
1
.
1
0
1
9
4
0
.
[
4
]
C
h
a
n
d
n
i
,
M
.
S
a
c
h
d
e
v
a
,
a
n
d
A
.
K
.
S
.
K
u
s
h
w
a
h
a
,
“
I
R
N
e
t
v
:
a
d
e
e
p
l
e
a
r
n
i
n
g
f
r
a
mew
o
r
k
f
o
r
a
u
t
o
m
a
t
e
d
b
r
a
i
n
t
u
mo
r
d
i
a
g
n
o
s
i
s,”
Bi
o
m
e
d
i
c
a
l
S
i
g
n
a
l
Pr
o
c
e
s
si
n
g
a
n
d
C
o
n
t
r
o
l
,
v
o
l
.
8
7
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
b
s
p
c
.
2
0
2
3
.
1
0
5
4
5
9
.
[
5
]
T.
K
.
D
u
t
t
a
,
D
.
R
.
N
a
y
a
k
,
a
n
d
Y
.
D
.
Z
h
a
n
g
,
“
A
R
M
-
N
e
t
:
a
t
t
e
n
t
i
o
n
-
g
u
i
d
e
d
r
e
s
i
d
u
a
l
mu
l
t
i
sc
a
l
e
C
N
N
f
o
r
m
u
l
t
i
c
l
a
ss
b
r
a
i
n
t
u
mo
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
s
i
n
g
M
R
i
ma
g
e
s,
”
Bi
o
m
e
d
i
c
a
l
S
i
g
n
a
l
P
ro
c
e
ss
i
n
g
a
n
d
C
o
n
t
r
o
l
,
v
o
l
.
8
7
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
b
sp
c
.
2
0
2
3
.
1
0
5
4
2
1
.
[
6
]
S
.
S
o
l
a
n
k
i
,
U
.
P
.
S
i
n
g
h
,
S
.
S
.
C
h
o
u
h
a
n
,
a
n
d
S
.
Ja
i
n
,
“
B
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
a
n
d
c
l
a
ss
i
f
i
c
a
t
i
o
n
u
s
i
n
g
i
n
t
e
l
l
i
g
e
n
c
e
t
e
c
h
n
i
q
u
e
s
:
a
n
o
v
e
r
v
i
e
w
,
”
I
EE
E
A
c
c
e
ss
,
v
o
l
.
1
1
,
p
p
.
1
2
8
7
0
–
1
2
8
8
6
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
3
.
3
2
4
2
6
6
6
.
[
7
]
M
.
A
.
T
a
l
u
k
d
e
r
e
t
a
l
.
,
“
A
n
e
f
f
i
c
i
e
n
t
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
t
o
c
a
t
e
g
o
r
i
z
e
b
r
a
i
n
t
u
m
o
r
u
si
n
g
r
e
c
o
n
st
r
u
c
t
i
o
n
a
n
d
f
i
n
e
-
t
u
n
i
n
g
,
”
E
x
p
e
rt
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
3
0
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
3
.
1
2
0
5
3
4
.
[
8
]
M
.
Y
a
q
u
b
e
t
a
l
.
,
“
D
e
e
p
La
b
V
3
,
I
B
C
O
-
b
a
se
d
A
LC
R
e
sN
e
t
:
a
f
u
l
l
y
a
u
t
o
m
a
t
e
d
c
l
a
ssi
f
i
c
a
t
i
o
n
a
n
d
g
r
a
d
i
n
g
sy
s
t
e
m
f
o
r
b
r
a
i
n
t
u
m
o
r
,
”
Al
e
x
a
n
d
ri
a
E
n
g
i
n
e
e
ri
n
g
J
o
u
rn
a
l
,
v
o
l
.
7
6
,
p
p
.
6
0
9
–
6
2
7
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
e
j
.
2
0
2
3
.
0
6
.
0
6
2
.
[
9
]
B
.
A
s
h
w
i
n
i
,
M
.
K
a
u
r
,
D
.
S
i
n
g
h
,
S
.
R
o
y
,
a
n
d
M
.
A
m
o
o
n
,
“
Ef
f
i
c
i
e
n
t
s
k
i
p
c
o
n
n
e
c
t
i
o
n
s
-
b
a
se
d
r
e
si
d
u
a
l
n
e
t
w
o
r
k
(
E
S
R
N
e
t
)
f
o
r
b
r
a
i
n
t
u
m
o
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
,
”
D
i
a
g
n
o
s
t
i
c
s
,
v
o
l
.
1
3
,
n
o
.
2
0
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
d
i
a
g
n
o
st
i
c
s
1
3
2
0
3
2
3
4
.
[
1
0
]
A
.
B
.
A
b
d
u
sa
l
o
m
o
v
,
M
.
M
u
k
h
i
d
d
i
n
o
v
,
a
n
d
T
.
K
.
W
h
a
n
g
b
o
,
“
B
r
a
i
n
t
u
mo
r
d
e
t
e
c
t
i
o
n
b
a
s
e
d
o
n
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s
a
n
d
mag
n
e
t
i
c
r
e
s
o
n
a
n
c
e
i
m
a
g
i
n
g
,
”
C
a
n
c
e
r
s
,
v
o
l
.
1
5
,
n
o
.
1
6
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
c
a
n
c
e
r
s
1
5
1
6
4
1
7
2
.
[
1
1
]
A
.
A
.
A
l
i
,
M
.
T.
H
a
mm
a
d
,
a
n
d
H
.
S
.
H
a
ssan
,
“
A
c
o
-
e
v
o
l
u
t
i
o
n
a
r
y
g
e
n
e
t
i
c
a
l
g
o
r
i
t
h
m a
p
p
r
o
a
c
h
t
o
o
p
t
i
m
i
z
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
b
r
a
i
n
t
u
m
o
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
,
”
I
E
EE
Ac
c
e
ss
,
v
o
l
.
1
3
,
p
p
.
2
1
2
2
9
–
2
1
2
4
8
,
2
0
2
5
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
5
.
3
5
3
5
8
4
4
.
[
1
2
]
N
.
K
.
K
u
m
a
r
,
V
.
D
.
A
.
K
u
m
a
r
,
D
.
Q
u
a
m
a
r
,
B
.
S
.
K
.
R
e
d
d
y
,
R
.
Y
o
g
e
n
d
r
a
,
a
n
d
N
.
P
o
o
j
i
t
h
a
,
“
E
n
h
a
n
c
e
d
A
l
z
h
e
i
m
e
r
’
s
d
i
s
e
a
s
e
p
r
e
d
i
c
t
i
o
n
t
h
r
o
u
g
h
a
d
v
a
n
c
e
d
i
m
a
g
i
n
g
:
a
s
t
u
d
y
o
f
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s
,
”
i
n
5
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
S
m
a
r
t
E
l
e
c
t
r
o
n
i
c
s
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
,
I
C
O
S
E
C
2
0
2
4
,
2
0
2
4
,
p
p
.
1
1
7
7
–
1182
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
O
S
E
C
6
1
5
8
7
.
2
0
2
4
.
1
0
7
2
2
1
1
6
.
[
1
3
]
X
.
Li
u
a
n
d
Z.
W
a
n
g
,
“
D
e
e
p
l
e
a
r
n
i
n
g
i
n
m
e
d
i
c
a
l
i
ma
g
e
c
l
a
ssi
f
i
c
a
t
i
o
n
f
r
o
m
M
R
I
-
b
a
s
e
d
b
r
a
i
n
t
u
mo
r
i
ma
g
e
s
,
”
i
n
2
0
2
4
I
E
EE
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
P
o
w
e
r,
I
n
t
e
l
l
i
g
e
n
t
C
o
m
p
u
t
i
n
g
a
n
d
S
y
s
t
e
m
s,
I
C
PI
C
S
2
0
2
4
,
2
0
2
4
,
p
p
.
8
4
0
–
8
4
4
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
P
I
C
S
6
2
0
5
3
.
2
0
2
4
.
1
0
7
9
6
1
0
8
.
[
1
4
]
S
.
P
o
k
h
r
e
l
,
L
.
K
.
D
a
h
a
l
,
N
.
G
u
p
t
a
,
R
.
S
h
r
e
st
h
a
,
A
.
S
r
i
v
a
s
t
a
v
a
,
a
n
d
A
.
B
h
a
sn
e
y
,
“
B
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
a
p
p
l
i
c
a
t
i
o
n
b
a
se
d
o
n
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
2
0
2
2
2
n
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
I
n
t
e
l
l
i
g
e
n
t
T
e
c
h
n
o
l
o
g
i
e
s,
C
O
N
I
T
2
0
2
2
,
Ju
n
.
2
0
2
2
,
p
p
.
1
–
5
,
d
o
i
:
1
0
.
1
1
0
9
/
C
O
N
I
T5
5
0
3
8
.
2
0
2
2
.
9
8
4
8
1
7
7
.
[
1
5
]
S
.
B
o
u
h
a
f
r
a
a
n
d
H
.
E
.
B
a
h
i
,
“
D
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s
f
o
r
b
r
a
i
n
t
u
mo
r
d
e
t
e
c
t
i
o
n
a
n
d
c
l
a
ssi
f
i
c
a
t
i
o
n
u
si
n
g
M
R
I
i
ma
g
e
s
(
2
0
2
0
t
o
2
0
2
4
)
:
a
sy
s
t
e
m
a
t
i
c
r
e
v
i
e
w
,
”
J
o
u
rn
a
l
o
f
I
m
a
g
i
n
g
I
n
f
o
rm
a
t
i
c
s
i
n
Me
d
i
c
i
n
e
,
v
o
l
.
3
8
,
n
o
.
3
,
p
p
.
1
4
0
3
–
1
4
3
3
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
2
7
8
-
024
-
0
1
2
8
3
-
8.
[
1
6
]
N
.
K
.
K
u
mar,
A
.
K
.
K
u
m
a
r
,
G
.
R
a
j
k
u
mar,
S
.
R
.
H
i
ma
j
a
sr
e
e
,
P
.
S
.
V
y
s
h
n
a
v
i
,
a
n
d
A
.
B
.
D
a
k
k
a
,
“
En
h
a
n
c
i
n
g
A
l
z
h
e
i
m
e
r
’
s
d
i
se
a
se
d
e
t
e
c
t
i
o
n
:
a
c
o
mp
a
r
a
t
i
v
e
s
t
u
d
y
o
f
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
w
i
t
h
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
a
n
d
c
u
s
t
o
m
C
N
N
m
o
d
e
l
s,
”
i
n
C
I
N
S
2
0
2
4
-
2
n
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
a
t
i
o
n
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
N
e
t
w
o
rk
S
y
s
t
e
m
s
,
2
0
2
4
,
p
p
.
1
–
4
,
d
o
i
:
1
0
.
1
1
0
9
/
C
I
N
S
6
3
8
8
1
.
2
0
2
4
.
1
0
8
6
2
9
9
3
.
[
1
7
]
R
.
S
.
S
a
l
a
r
i
a
a
n
d
N
.
M
o
h
a
n
,
“
A
r
t
i
f
i
c
i
a
l
b
e
e
c
o
l
o
n
y
a
l
g
o
r
i
t
h
m
f
o
r
e
f
f
i
c
i
e
n
t
h
y
p
e
r
p
a
r
a
m
e
t
e
r
t
u
n
i
n
g
i
n
A
l
z
h
e
i
m
e
r
’
s
d
i
s
e
a
s
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
,
”
i
n
M
o
d
e
r
n
P
r
a
c
t
i
c
e
s
a
n
d
T
r
e
n
d
s
i
n
E
x
p
e
r
t
A
p
p
l
i
c
a
t
i
o
n
s
a
n
d
S
e
c
u
r
i
t
y
.
S
i
n
g
a
p
o
r
e
:
S
p
r
i
n
g
e
r
,
2
0
2
5
,
p
p
.
3
3
7
–
3
4
6
,
doi
:
1
0
.
1
0
0
7
/
9
7
8
-
9
8
1
-
96
-
5
7
8
1
-
0
_
2
8
.
[
1
8
]
C
.
M
o
k
h
t
a
r
i
,
M
.
D
e
b
a
k
l
a
,
a
n
d
B
.
M
e
f
t
a
h
,
“
F
u
z
z
y
c
l
u
st
e
r
i
n
g
o
p
t
i
m
i
z
a
t
i
o
n
b
a
se
d
a
r
t
i
f
i
c
i
a
l
b
e
e
c
o
l
o
n
y
a
l
g
o
r
i
t
h
m
f
o
r
b
r
a
i
n
ma
g
n
e
t
i
c
r
e
so
n
a
n
c
e
i
ma
g
i
n
g
i
ma
g
e
s
e
g
m
e
n
t
a
t
i
o
n
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
E
l
e
c
t
ri
c
a
l
a
n
d
C
o
m
p
u
t
e
r
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
1
5
,
n
o
.
5
,
2
0
2
5
,
pp
.
4
9
1
6
-
4
9
3
2
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
e
c
e
.
v
1
5
i
5
.
p
p
4
9
1
6
-
4
9
3
2
.
[
1
9
]
A
.
P
a
n
d
e
y
a
n
d
V
.
K
.
P
a
n
d
e
y
,
“
W
a
v
e
l
e
t
-
b
a
se
d
c
l
a
ssi
f
i
c
a
t
i
o
n
u
s
i
n
g
me
t
a
-
h
e
u
r
i
st
i
c
a
l
g
o
r
i
t
h
m w
i
t
h
d
e
e
p
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
,
”
S
N
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
6
,
n
o
.
3
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s
4
2
9
7
9
-
0
2
5
-
0
3
6
8
3
-
1.
[
2
0
]
S
.
A
n
j
a
n
a
,
S
.
J
.
S
u
s
mi
,
a
n
d
P
.
M
.
S
.
R
a
j
a
,
“
P
e
r
f
o
r
m
a
n
c
e
me
a
su
r
e
o
f
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
b
r
i
a
n
t
u
m
o
r
i
m
a
g
e
a
n
a
l
y
si
s
u
si
n
g
d
o
m
a
i
n
i
n
f
o
r
mat
i
o
n
,
”
i
n
2
n
d
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
a
n
d
Au
t
o
n
o
m
o
u
s
S
y
s
t
e
m
s,
I
C
M
L
AS
2
0
2
5
,
2
0
2
5
,
p
p
.
2
5
9
–
264
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
M
LA
S
6
4
5
5
7
.
2
0
2
5
.
1
0
9
6
8
4
1
2
.
[
2
1
]
A
.
Y
a
q
o
o
b
,
N
.
K
.
V
e
r
ma
,
a
n
d
R
.
M
.
A
z
i
z
,
“
O
p
t
i
mi
z
i
n
g
g
e
n
e
se
l
e
c
t
i
o
n
a
n
d
c
a
n
c
e
r
c
l
a
ssi
f
i
c
a
t
i
o
n
w
i
t
h
h
y
b
r
i
d
si
n
e
c
o
s
i
n
e
a
n
d
c
u
c
k
o
o
se
a
r
c
h
a
l
g
o
r
i
t
h
m,”
J
o
u
rn
a
l
o
f
Me
d
i
c
a
l
S
y
st
e
m
s
,
v
o
l
.
4
8
,
n
o
.
1
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
s1
0
9
1
6
-
0
2
3
-
0
2
0
3
1
-
1.
[
2
2
]
M
.
C
e
l
i
k
a
n
d
O
.
I
n
i
k
,
“
D
e
v
e
l
o
p
m
e
n
t
o
f
h
y
b
r
i
d
m
o
d
e
l
s
b
a
s
e
d
o
n
d
e
e
p
l
e
a
r
n
i
n
g
a
n
d
o
p
t
i
m
i
z
e
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
b
r
a
i
n
t
u
mo
r
mu
l
t
i
-
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
Ex
p
e
rt
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
3
8
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
3
.
1
2
2
1
5
9
.
[
2
3
]
H
.
G
i
v
i
a
n
a
n
d
J
.
P
.
C
a
l
b
i
m
o
n
t
e
,
“
A
r
e
v
i
e
w
o
n
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s
f
o
r
d
i
a
g
n
o
s
i
s
o
f
A
l
z
h
e
i
m
e
r
’
s
d
i
s
e
a
s
e
a
n
d
m
i
l
d
c
o
g
n
i
t
i
v
e
i
m
p
a
i
r
m
e
n
t
b
a
s
e
d
o
n
b
r
a
i
n
M
R
I
,
”
I
E
E
E
A
c
c
e
s
s
,
v
o
l
.
1
2
,
p
p
.
1
0
9
9
1
2
–
1
0
9
9
2
9
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
E
S
S
.
2
0
2
4
.
3
4
3
8
0
8
1
.
[
2
4
]
K
.
K
.
N
a
p
a
,
S
.
M
u
r
u
g
a
n
,
J
.
S
.
M
u
r
u
g
a
n
,
a
n
d
A
.
J
a
y
a
n
t
h
i
,
“
B
R
A
I
N
-
M
ETA
:
a
r
e
p
r
o
d
u
c
i
b
l
e
C
N
N
–
v
i
s
i
o
n
t
r
a
n
sf
o
r
mer
m
e
t
a
-
e
n
s
e
mb
l
e
p
i
p
e
l
i
n
e
f
o
r
e
x
p
l
a
i
n
a
b
l
e
b
r
a
i
n
t
u
m
o
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
,
”
Me
t
h
o
d
sX
,
v
o
l
.
1
6
,
2
0
2
6
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
me
x
.
2
0
2
5
.
1
0
3
7
6
9
.
[
2
5
]
L.
S
.
-
M
o
r
e
n
o
,
A
.
P
.
-
P
e
ñ
a
,
L
.
D
.
-
L
o
p
e
z
,
a
n
d
J.
P
.
D
.
-
M
o
r
a
l
e
s,
“
En
s
e
m
b
l
e
-
b
a
s
e
d
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
s
f
o
r
b
r
a
i
n
t
u
mo
r
c
l
a
ss
i
f
i
c
a
t
i
o
n
i
n
M
R
I
:
e
n
h
a
n
c
i
n
g
a
c
c
u
r
a
c
y
a
n
d
i
n
t
e
r
p
r
e
t
a
b
i
l
i
t
y
u
si
n
g
e
x
p
l
a
i
n
a
b
l
e
A
I
,
”
C
o
m
p
u
t
e
rs
i
n
Bi
o
l
o
g
y
a
n
d
Me
d
i
c
i
n
e
,
v
o
l
.
1
9
5
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
m
p
b
i
o
me
d
.
2
0
2
5
.
1
1
0
5
5
5
.
[
2
6
]
N
.
K
.
C
h
a
u
h
a
n
e
t
a
l
.
,
“
A
h
y
b
r
i
d
l
e
a
r
n
i
n
g
n
e
t
w
o
r
k
w
i
t
h
p
r
o
g
r
e
ss
i
v
e
r
e
si
z
i
n
g
a
n
d
P
C
A
f
o
r
d
i
a
g
n
o
s
i
s
o
f
c
e
r
v
i
c
a
l
c
a
n
c
e
r
o
n
W
S
I
sl
i
d
e
s,
”
S
c
i
e
n
t
i
f
i
c
Re
p
o
rt
s
,
v
o
l
.
1
5
,
n
o
.
1
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
8
-
025
-
9
7
7
1
9
-
4.
[
2
7
]
A
.
H
a
l
d
e
r
,
A
.
D
a
l
a
l
,
S
.
G
h
a
r
a
mi
,
M
.
W
o
z
n
i
a
k
,
M
.
F
.
I
j
a
z
,
a
n
d
P
.
K
.
S
i
n
g
h
,
“
A
f
u
z
z
y
r
a
n
k
-
b
a
se
d
d
e
e
p
e
n
s
e
mb
l
e
met
h
o
d
o
l
o
g
y
f
o
r
mu
l
t
i
-
c
l
a
ss
sk
i
n
c
a
n
c
e
r
c
l
a
ssi
f
i
c
a
t
i
o
n
,
”
S
c
i
e
n
t
i
f
i
c
R
e
p
o
r
t
s
,
v
o
l
.
1
5
,
n
o
.
1
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
8
-
025
-
9
0
4
2
3
-
3.
[
2
8
]
D
.
V
a
l
e
n
k
o
v
a
,
A
.
L
y
a
n
o
v
a
,
A
.
S
i
n
i
t
c
a
,
R
.
S
a
r
k
a
r
,
a
n
d
D
.
K
a
p
l
u
n
,
“
A
f
u
z
z
y
r
a
n
k
-
b
a
se
d
e
n
sem
b
l
e
o
f
C
N
N
mo
d
e
l
s
f
o
r
M
R
I
seg
m
e
n
t
a
t
i
o
n
,
”
Bi
o
m
e
d
i
c
a
l
S
i
g
n
a
l
P
r
o
c
e
ssi
n
g
a
n
d
C
o
n
t
r
o
l
,
v
o
l
.
1
0
2
,
A
p
r
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
b
s
p
c
.
2
0
2
4
.
1
0
7
3
4
2
.
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
n
o
p
timiz
ed
d
ee
p
lea
r
n
in
g
fr
a
mewo
r
k
fo
r
b
r
a
in
tu
mo
r
cla
s
s
ifica
tio
n
u
s
in
g
…
(
K
o
ma
l Ku
ma
r
N
a
p
a
)
3731
[
2
9
]
S
.
R
a
j
p
u
t
,
R
.
K
a
p
d
i
,
M
.
R
o
y
,
a
n
d
M
.
S
.
R
a
v
a
l
,
“
A
t
r
i
p
l
a
n
a
r
e
n
sem
b
l
e
mo
d
e
l
f
o
r
b
r
a
i
n
t
u
m
o
r
se
g
me
n
t
a
t
i
o
n
w
i
t
h
v
o
l
u
m
e
t
r
i
c
mu
l
t
i
p
a
r
a
me
t
r
i
c
ma
g
n
e
t
i
c
r
e
so
n
a
n
c
e
i
mag
e
s,”
H
e
a
l
t
h
c
a
r
e
A
n
a
l
y
t
i
c
s
,
v
o
l
.
5
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
h
e
a
l
t
h
.
2
0
2
4
.
1
0
0
3
0
7
.
[
3
0
]
S
.
S
.
R
a
n
i
,
K
.
S
u
b
r
a
m
a
n
i
a
n
,
V
.
R
a
j
i
n
i
k
a
n
t
h
,
a
n
d
R
.
S
e
l
v
a
r
a
j
a
n
,
“
U
-
N
e
t
e
n
s
e
mb
l
e
b
a
se
d
s
e
g
me
n
t
a
t
i
o
n
a
n
d
R
ESN
ET
f
o
r
b
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
o
f
m
a
g
n
e
t
i
c
r
e
so
n
a
n
c
e
i
ma
g
i
n
g
(
M
R
I
)
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
S
y
s
t
e
m
Ass
u
ra
n
c
e
En
g
i
n
e
e
ri
n
g
a
n
d
Ma
n
a
g
e
m
e
n
t
,
p
p
.
1
–
2
0
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
3
1
9
8
-
0
2
5
-
0
2
9
4
4
-
9.
[
3
1
]
M
.
N
i
c
k
p
a
r
v
a
r
,
“
B
r
a
i
n
t
u
m
o
r
M
R
I
d
a
t
a
s
e
t
,
”
k
a
g
g
l
e
.
c
o
m
.
A
c
c
e
ss
e
d
:
M
a
y
8
,
2
0
2
5
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
w
w
w
.
k
a
g
g
l
e
.
c
o
m
/
d
a
t
a
se
t
s/
mas
o
u
d
n
i
c
k
p
a
r
v
a
r
/
b
r
a
i
n
-
t
u
mo
r
-
mr
i
-
d
a
t
a
s
e
t
[
3
2
]
M
.
A
.
H
.
Tu
h
i
n
,
T.
P
r
a
ma
n
i
c
k
,
H
.
K
.
Em
o
n
,
W
.
R
a
h
m
a
n
,
M
.
M
.
I
.
R
a
h
i
,
a
n
d
M
.
A
.
A
l
a
m,
“
D
e
t
e
c
t
i
o
n
a
n
d
3
D
v
i
s
u
a
l
i
z
a
t
i
o
n
o
f
b
r
a
i
n
t
u
m
o
r
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
a
n
d
p
o
l
y
n
o
mi
a
l
i
n
t
e
r
p
o
l
a
t
i
o
n
,
”
i
n
2
0
2
0
I
E
EE
Asi
a
-
Pa
c
i
f
i
c
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
D
a
t
a
E
n
g
i
n
e
e
ri
n
g
,
C
S
D
E
2
0
2
0
,
2
0
2
0
,
p
p
.
1
–
6
,
d
o
i
:
1
0
.
1
1
0
9
/
C
S
D
E
5
0
8
7
4
.
2
0
2
0
.
9
4
1
1
5
9
5
.
[
3
3
]
Y
.
S
u
t
e
r
e
t
a
l
.
,
“
D
e
e
p
l
e
a
r
n
i
n
g
v
e
r
su
s
c
l
a
s
si
c
a
l
r
e
g
r
e
ss
i
o
n
f
o
r
b
r
a
i
n
t
u
m
o
r
p
a
t
i
e
n
t
su
r
v
i
v
a
l
p
r
e
d
i
c
t
i
o
n
,
”
i
n
Bra
i
n
l
e
s
i
o
n
:
G
l
i
o
m
a
,
M
u
l
t
i
p
l
e
S
c
l
e
r
o
s
i
s,
S
t
ro
k
e
a
n
d
T
r
a
u
m
a
t
i
c
Br
a
i
n
I
n
j
u
ri
e
s
.
C
h
a
m,
S
w
i
t
z
e
r
l
a
n
d
:
S
p
r
i
n
g
e
r
,
2
0
1
9
,
p
p
.
4
2
9
–
4
4
0
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
030
-
1
1
7
2
6
-
9
_
3
8
.
[
3
4
]
Y
.
H
u
a
n
d
Y
.
X
i
a
,
“
3
D
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
-
b
a
s
e
d
b
r
a
i
n
t
u
m
o
r
s
e
g
m
e
n
t
a
t
i
o
n
u
si
n
g
m
u
l
t
i
m
o
d
a
l
i
t
y
ma
g
n
e
t
i
c
r
e
so
n
a
n
c
e
s
e
q
u
e
n
c
e
s,
”
i
n
3
D
D
e
e
p
N
e
u
r
a
l
N
e
t
w
o
r
k
-
B
a
se
d
Bra
i
n
T
u
m
o
r
S
e
g
m
e
n
t
a
t
i
o
n
U
s
i
n
g
M
u
l
t
i
m
o
d
a
l
i
t
y
M
a
g
n
e
t
i
c
Re
s
o
n
a
n
c
e
S
e
q
u
e
n
c
e
s
.
C
h
a
m,
S
w
i
t
z
e
r
l
a
n
d
:
S
p
r
i
n
g
e
r
,
2
0
1
8
,
p
p
.
4
2
3
–
4
3
4
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
3
1
9
-
7
5
2
3
8
-
9
_
3
6
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Dr
.
K
o
m
a
l
K
u
m
a
r
Na
p
a
is
c
u
rre
n
tl
y
w
o
rk
i
n
g
a
s
a
ss
istan
t
p
ro
f
e
ss
o
r
(S
G
)
in
th
e
De
p
a
rtme
n
t
o
f
Artifi
c
ial
I
n
telli
g
e
n
c
e
a
n
d
Da
ta
S
c
ien
c
e
,
S
a
v
e
e
th
a
En
g
in
e
e
rin
g
Co
l
leg
e
,
Ch
e
n
n
a
i,
Tam
il
Na
d
u
,
I
n
d
ia.
His res
e
a
rc
h
in
tere
sts
in
c
lu
d
e
m
a
c
h
in
e
lea
rn
in
g
,
d
a
ta
m
in
in
g
,
a
n
d
c
lo
u
d
c
o
m
p
u
ti
n
g
.
He
c
a
n
b
e
c
o
n
t
a
c
ted
a
t
e
m
a
il
:
k
o
m
a
lk
u
m
a
rn
a
p
a
@g
m
a
il
.
c
o
m
.
Mr.
Ra
jk
u
m
a
r
G
o
v
in
d
a
r
a
ja
n
is
c
u
rre
n
tl
y
wo
rk
i
n
g
a
s
a
ss
istan
t
p
ro
fe
ss
o
r
in
t
h
e
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
(Da
ta
S
c
ien
c
e
)
a
t
M
a
d
a
n
a
p
a
ll
e
I
n
stit
u
te
o
f
Tec
h
n
o
l
o
g
y
a
n
d
S
c
ien
c
e
De
e
m
e
d
to
b
e
Un
i
v
e
rsity
,
M
a
d
a
n
a
p
a
ll
e
,
An
d
h
ra
P
ra
d
e
sh
.
His
re
se
a
rc
h
in
tere
sts
in
c
l
u
d
e
m
a
c
h
in
e
lea
rn
in
g
,
a
n
d
d
a
ta
m
in
i
n
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
k
u
m
a
r3
5
4
4
@
g
m
a
il
.
c
o
m
.
Dr
.
S
e
n
th
il
Mu
r
u
g
a
n
J
a
n
a
k
ira
m
a
n
is
c
u
rre
n
tl
y
w
o
rk
i
n
g
a
s
p
ro
fe
ss
o
r
in
th
e
Ce
n
tre
of
Ex
c
e
ll
e
n
c
e
,
Ra
jala
k
sh
m
i
En
g
i
n
e
e
rin
g
C
o
ll
e
g
e
,
Ch
e
n
n
a
i
In
d
ia.
His
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
m
a
c
h
in
e
lea
rn
i
n
g
,
a
n
d
ima
g
e
p
ro
c
e
ss
in
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
d
rjse
n
th
il
m
u
ru
g
a
n
m
tec
h
@g
m
a
il
.
c
o
m
.
Mrs
.
J
a
y
a
n
th
i
Ar
u
m
u
g
a
m
is
c
u
rre
n
tl
y
wo
r
k
in
g
a
s
a
ss
istan
t
p
ro
fe
ss
o
r
in
th
e
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ie
n
c
e
a
n
d
E
n
g
i
n
e
e
rin
g
a
t
Ve
lam
m
a
l
En
g
in
e
e
rin
g
Co
ll
e
g
e
,
Ch
e
n
n
a
i,
I
n
d
ia.
He
r
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
d
a
ta
m
in
in
g
a
n
d
m
a
c
h
in
e
lea
rn
in
g
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
jay
a
n
th
iaru
m
u
g
a
m
k
@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.