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.
13
,
No
.
4
,
Dec
em
b
er
2
0
2
4
,
p
p
.
4
9
3
9
~
4
9
4
7
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
13
.i
4
.
p
p
4
9
3
9
-
4
9
4
7
4939
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Enha
ncing
ultras
o
und
-
g
uided
b
ra
chia
l
plex
us
nerv
e
lo
ca
liza
tion
with
ResNe
t50
a
nd
supp
o
rt
v
e
ctor
ma
chine
So
bh
a
na
M
um
m
a
neni
,
K
us
h
a
l
K
um
a
r
Chin
t
a
ka
y
a
l
a
,
L
a
l
it
h
Sa
i
M
uk
un
d
Ya
rla
g
a
dd
a
,
Venk
a
t
a
Siv
a
Na
g
a
R
a
j
u
Ala
,
Nihitha
Vem
ula
pa
lli
D
e
p
a
r
t
me
n
t
of
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
a
n
d
En
g
i
n
e
e
r
i
n
g
,
V
e
l
a
g
a
p
u
d
i
R
a
m
a
k
r
i
sh
n
a
S
i
d
d
h
a
r
t
h
a
E
n
g
i
n
e
e
r
i
n
g
C
o
l
l
e
g
e
,
V
i
j
a
y
a
w
a
d
a
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
1
5
,
2
0
2
3
R
ev
is
ed
Ma
r
1
,
2
0
2
4
Acc
ep
ted
Ma
r
2
1
,
2
0
2
4
M
e
d
ica
l
ima
g
e
se
g
m
e
n
tatio
n
a
n
d
c
las
sifica
ti
o
n
p
lay
s
a
v
it
a
l
r
o
l
e
in
n
e
r
v
e
b
lo
c
k
/reg
io
n
id
e
n
ti
fica
ti
o
n
,
p
a
rt
icu
larly
fo
r
a
n
e
sth
e
sio
lo
g
ists
r
e
ly
in
g
on
in
stin
c
t
u
a
l
ju
d
g
m
e
n
ts.
Ho
we
v
e
r,
due
to
p
a
ti
e
n
t
-
sp
e
c
ifi
c
a
n
a
to
m
ica
l
v
a
riatio
n
s,
th
e
se
m
e
th
o
d
s
so
m
e
ti
m
e
s
lac
k
p
re
c
isio
n
.
Th
is
re
se
a
rc
h
fo
c
u
se
s
on
a
d
d
re
ss
in
g
th
e
p
r
o
b
lem
,
by
in
c
o
rp
o
ra
ti
n
g
n
o
v
e
l
e
n
se
m
b
li
n
g
m
e
th
o
d
of
Re
sN
e
t
-
50
a
n
d
s
u
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
(S
VM)
to
a
c
h
ie
v
e
se
g
m
e
n
tatio
n
of
d
a
tas
e
t
ima
g
e
s
a
n
d
c
las
sifica
ti
o
n
of
n
e
rv
e
b
l
o
c
k
s
re
sp
e
c
ti
v
e
ly
.
Th
e
sa
id
n
o
v
e
l
e
n
se
m
b
le
m
o
d
e
l
is
train
e
d
on
a
p
u
b
li
c
l
y
a
v
a
il
a
b
le
d
a
tas
e
t
c
o
n
sistin
g
of
m
o
re
th
a
n
1
6
,
8
0
0
ima
g
e
s.
T
h
e
s
o
le
p
u
rp
o
se
of
th
is
wo
r
k
is
to
a
d
d
re
ss
th
e
p
ro
b
lem
of
p
e
ri
p
h
e
ra
l
n
e
rv
e
b
l
o
c
k
in
g
(P
NB)
with
t
h
e
u
sa
g
e
of
e
n
se
m
b
le
m
o
d
e
ll
in
g
,
w
h
il
e
a
c
h
iev
i
n
g
th
e
h
i
g
h
e
st
p
o
ss
ib
le
a
c
c
u
ra
c
y
.
T
h
is
re
s
e
a
rc
h
will
h
e
lp
p
ra
c
ti
ti
o
n
e
rs
in
a
c
c
u
ra
tely
i
d
e
n
ti
f
y
in
g
t
h
e
l
o
c
a
ti
o
n
of
b
ra
c
h
i
a
l
p
lex
u
s
a
n
d
d
isti
n
g
u
is
h
in
g
t
h
e
t
y
p
e
of
n
e
rv
e
b
l
o
c
k
to
be
in
jec
ted
–
in
ters
c
a
len
e
a
n
d
su
p
ra
c
lav
i
c
u
lar.
T
h
e
m
o
d
e
l,
wh
ic
h
in
teg
ra
tes
Re
sN
e
t5
0
a
n
d
S
VM
c
las
sifier
,
a
c
h
iev
e
d
a
c
o
m
m
e
n
d
a
b
le
9
9
.
27
%
a
c
c
u
ra
c
y
in
i
d
e
n
ti
f
y
i
n
g
a
n
d
c
las
sify
in
g
t
h
e
b
ra
c
h
ial
p
le
x
u
s
re
g
io
n
.
K
ey
w
o
r
d
s
:
B
r
ac
h
ial
p
lex
u
s
Per
ip
h
er
al
n
er
v
e
b
lo
c
k
in
g
R
esNet5
0
Seg
m
en
tatio
n
S
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
class
if
ier
T
h
is
is
an
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r
th
e
CC
BY
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
So
b
h
an
a
Mu
m
m
an
en
i
D
e
p
a
r
t
m
e
n
t
of
C
o
m
p
u
t
e
r
S
c
ie
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
V
e
l
a
g
a
p
u
d
i
R
a
m
a
k
r
is
h
n
a
Si
d
d
h
a
r
t
h
a
E
n
g
in
e
e
r
i
n
g
C
o
l
l
e
g
e
Vijay
awa
d
a
5
2
0
0
0
7
,
I
n
d
ia
E
m
ail:
s
o
b
h
an
a
@
v
r
s
id
d
h
ar
t
h
a
.
ac
.
in
1.
I
NT
RO
D
UCT
I
O
N
An
esth
esio
lo
g
is
ts
h
av
e
alwa
y
s
in
tu
itiv
ely
r
ec
o
g
n
ized
n
er
v
e
b
lo
c
k
s
an
d
r
eg
io
n
s
th
r
o
u
g
h
th
e
s
eg
m
en
tatio
n
of
m
ed
ical
im
a
g
es
[
1
]
.
W
h
en
it
co
m
es
to
p
r
ec
is
ely
wh
er
e
to
d
eliv
er
an
esth
esia,
th
ey
ty
p
ically
r
ely
on
th
ei
r
in
t
u
itio
n
a
n
d
ex
p
er
ien
ce
.
Ho
wev
er
,
b
ec
a
u
s
e
each
p
atien
t'
s
ar
ea
of
a
d
m
in
i
s
tr
atio
n
h
as
u
n
iq
u
e
d
y
n
am
ics,
th
is
p
r
o
ce
s
s
m
ay
n
o
t
alwa
y
s
p
r
o
d
u
ce
1
0
0
%
ac
cu
r
ac
y
.
As
a
r
esu
lt,
b
l
o
o
d
p
r
ess
u
r
e
(
BP
)
n
er
v
es
can
be
f
o
u
n
d
on
u
ltra
s
o
n
o
g
r
ap
h
y
p
ictu
r
es
th
at
id
en
tify
n
e
r
v
e
s
tr
u
ctu
r
es.
I
d
e
n
tific
atio
n
can
be
d
if
f
icu
lt
ev
en
f
o
r
s
p
ec
ialis
t
s
in
th
e
f
ield
s
in
ce
s
p
ec
k
le
n
o
is
e
an
d
ec
h
o
d
is
r
u
p
tio
n
s
[
2
]
ca
n
ca
u
s
e
th
e
u
ltr
aso
u
n
d
p
ictu
r
es
to
b
ec
o
m
e
d
is
to
r
ted
.
By
u
s
in
g
p
ain
m
a
n
ag
em
e
n
t
ca
th
eter
s
,
th
is
tr
ea
tm
en
t
ac
ts
as
a
h
elp
an
d
f
ac
ilit
ates
p
r
ac
titi
o
n
er
s
'
ab
ilit
y
to
o
p
er
at
e
on
n
er
v
es
ef
f
icien
tly
,
wh
ic
h
ea
s
es
p
r
o
ce
s
s
es
lik
e
p
er
ip
h
er
al
n
er
v
e
b
l
o
ck
ag
e
(
PNB
)
an
d
r
eg
io
n
al
a
n
esth
etic
[
3
]
,
[
4
]
.
T
h
er
e
h
a
v
e
b
ee
n
s
ev
er
al
w
o
r
k
s
th
at
f
o
c
u
s
ed
on
th
e
d
etec
tio
n
of
b
r
ac
h
ial
p
lex
u
s
r
eg
io
n
[
5
]
in
th
e
in
p
u
tted
u
ltra
s
o
u
n
d
im
ag
es;
but
th
is
ap
p
r
o
ac
h
in
v
o
lv
es
d
etec
tio
n
as
well
as
s
eg
m
en
tatio
n
of
th
e
b
r
ac
h
ial
p
lex
u
s
r
eg
i
o
n
,
in
to
s
u
p
r
ac
lav
i
cu
lar
an
d
in
ter
s
ca
len
e
[
6
]
,
an
d
th
is
wo
r
k
wo
u
l
d
be
th
e
f
ir
s
t
of
its
k
in
d
.
T
h
e
co
n
ce
p
ts
of
s
u
p
r
ac
la
v
icu
lar
an
d
in
ter
s
ca
len
e
n
er
v
e
b
l
o
ck
s
ar
e
m
ed
ically
co
m
p
lex
,
a
n
d
a
p
r
ac
titi
o
n
er
n
ee
d
s
to
be
ab
le
to
th
o
r
o
u
g
h
ly
d
if
f
er
en
tiate
b
etwe
en
th
em
.
An
in
ter
s
ca
len
e
n
er
v
e
b
lo
ck
is
in
jecte
d
b
etwe
en
th
e
m
u
s
cles
of
th
e
n
ec
k
r
eg
io
n
,
w
h
ich
is
in
b
etwe
en
th
e
s
ca
len
e
m
u
s
cles.
T
h
is
n
er
v
e
b
lo
c
k
s
p
e
cif
ically
af
f
ec
ts
th
e
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.
13
,
No
.
4
,
Dec
em
b
er
2
0
2
4
:
4
9
3
9
-
4
9
4
7
4940
n
er
v
es
th
at
ex
ten
d
to
th
e
s
h
o
u
ld
er
s
an
d
t
h
e
u
p
p
er
ar
m
a
n
d
p
r
o
v
id
es
ef
f
ec
tiv
e
a
n
esth
esia
f
o
r
o
p
er
atio
n
s
p
er
f
o
r
m
ed
on
th
e
s
am
e.
On
th
e
co
n
tr
a
r
y
,
a
s
u
p
r
ac
lav
icu
lar
is
in
jecte
d
ab
o
v
e
th
e
co
llar
b
o
n
e,
in
th
e
r
eg
io
n
ju
s
t
ab
o
v
e
t
h
e
s
h
o
u
ld
er
,
an
d
it
af
f
ec
ts
th
e
b
r
ac
h
ial
p
lex
u
s
n
er
v
e
s
in
a
s
im
ilar
m
a
n
n
er
,
ex
ce
p
t
f
o
r
th
e
f
ac
t
th
at
it
p
r
o
v
id
es
an
esth
esia
f
o
r
th
e
en
t
ir
e
ar
m
,
an
d
h
en
ce
is
u
s
ed
f
o
r
o
p
er
atio
n
s
t
h
at
in
v
o
lv
e
h
an
d
,
f
o
r
ea
r
m
an
d
elb
o
w
r
eg
io
n
s
.
R
esid
u
al
n
etwo
r
k
with
5
0
lay
er
s
(
R
esNet5
0
)
an
d
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM
)
h
y
b
r
id
m
o
d
el
co
m
b
in
es
th
e
s
tr
en
g
th
s
of
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
s
(
C
NNs)
an
d
tr
ad
itio
n
al
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
to
cr
ea
te
a
p
o
wer
f
u
l
class
if
icatio
n
s
y
s
tem
.
R
esNet5
0
,
a
v
a
r
ian
t
of
th
e
d
ee
p
r
esid
u
al
n
et
wo
r
k
ar
c
h
itectu
r
e,
s
er
v
es
as
th
e
f
ea
tu
r
e
e
x
tr
ac
to
r
,
lev
er
ag
in
g
its
ab
ilit
y
to
ca
p
t
u
r
e
co
m
p
lex
h
ier
ar
ch
ical
p
atter
n
s
in
im
ag
es.
T
h
e
p
r
etr
ain
ed
R
esNet5
0
m
o
d
el
is
u
tili
ze
d
to
ex
t
r
ac
t
h
ig
h
-
lev
el
f
ea
tu
r
es
[
7
]
f
r
o
m
in
p
u
t
im
ag
e
s
,
p
r
eser
v
in
g
s
p
atial
r
elatio
n
s
h
ip
s
an
d
s
em
an
tic
in
f
o
r
m
atio
n
.
T
h
ese
e
x
tr
ac
ted
f
ea
tu
r
es
ar
e
th
e
n
f
ed
i
n
to
a
SVM
class
if
ier
,
wh
ich
ex
ce
ls
in
h
an
d
lin
g
h
ig
h
-
d
im
e
n
s
io
n
al
f
ea
tu
r
e
s
p
ac
es
an
d
f
in
d
i
n
g
o
p
tim
al
d
ec
is
io
n
b
o
u
n
d
a
r
ie
s
.
T
h
e
d
ec
is
io
n
of
ch
o
o
s
in
g
b
e
twee
n
th
e
r
i
g
h
t
n
er
v
e
b
lo
c
k
d
ep
en
d
s
on
th
e
c
o
m
p
lex
ity
of
s
u
r
g
ical
p
r
o
ce
d
u
r
e
an
d
th
e
co
n
s
id
er
ati
o
n
s
of
t
h
e
an
esth
esia
p
r
o
v
id
e
r
.
An
esth
esio
lo
g
is
ts
ca
r
ef
u
lly
ass
es
s
th
e
p
atien
t'
s
m
ed
ical
h
is
to
r
y
,
c
u
r
r
en
t
h
ea
lth
s
tatu
s
,
an
d
th
e
n
at
u
r
e
of
th
e
s
u
r
g
ical
p
r
o
ce
d
u
r
e
to
d
eter
m
in
e
th
e
ap
p
r
o
p
r
iate
ty
p
e
an
d
a
m
o
u
n
t
of
a
n
esth
esia.
Pra
ctitio
n
er
’
s
tailo
r
th
e
d
o
s
ag
e
to
each
p
atien
t'
s
s
p
ec
if
ic
r
eq
u
ir
em
en
ts
,
en
s
u
r
in
g
o
p
tim
al
an
esth
etic
ef
f
ec
ts
an
d
m
in
im
izin
g
r
is
k
s
.
In
th
e
e
v
en
t
of
u
n
e
x
p
ec
te
d
co
m
p
licatio
n
s
or
em
er
g
en
cies
d
u
r
in
g
s
u
r
g
e
r
y
,
th
ey
m
u
s
t
be
p
r
ep
a
r
ed
to
r
esp
o
n
d
q
u
ick
ly
an
d
ad
j
u
s
t
th
e
a
n
esth
esia
d
o
s
ag
e
to
ad
d
r
ess
th
e
s
itu
atio
n
wh
ile
m
a
in
tain
in
g
p
atien
t
s
af
ety
.
T
h
e
p
ap
e
r
is
o
r
g
an
ized
i
n
to
f
i
v
e
s
ec
tio
n
s
.
T
h
e
f
ir
s
t
s
ec
tio
n
s
h
o
ws
th
e
in
tr
o
d
u
ctio
n
p
ar
t.
T
h
e
s
ec
o
n
d
s
ec
tio
n
is
liter
atu
r
e
r
ev
iew
,
f
o
c
u
s
ed
on
e
x
is
tin
g
r
esear
ch
r
elate
d
to
o
u
r
s
tu
d
y
.
T
h
e
th
ir
d
s
ec
tio
n
is
m
e
t
h
o
d
o
l
o
g
y
,
w
h
i
c
h
d
e
t
ai
ls
o
u
r
a
p
p
r
o
a
c
h
a
n
d
t
e
c
h
n
i
q
u
es
a
p
p
l
i
e
d
.
T
h
e
f
o
u
r
t
h
s
e
ct
i
o
n
is
r
es
u
l
t
s
a
n
d
d
is
c
u
s
s
i
o
n
,
w
h
i
c
h
p
r
e
s
e
n
ts
our
f
i
n
d
i
n
g
s
.
F
in
a
l
l
y
,
t
h
e
c
o
n
c
l
u
s
i
o
n
s
e
ct
i
o
n
s
u
m
m
a
r
i
z
e
s
t
h
e
f
i
n
d
i
n
g
s
a
n
d
f
u
t
u
r
e
w
o
r
k
.
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
Ka
y
e
et
a
l.
[
8
]
clea
r
l
y
an
d
co
n
cisely
d
elin
ea
ted
t
h
e
d
i
s
tin
ctio
n
s
b
etwe
en
s
u
p
r
ac
la
v
icu
lar
an
d
in
ter
s
ca
len
e
n
er
v
e
b
lo
c
k
s
.
T
h
ey
p
r
esen
ted
ev
i
d
en
ce
f
r
o
m
s
ev
er
al
s
tu
d
ies
in
d
icatin
g
th
at,
f
r
o
m
a
s
af
ety
s
tan
d
p
o
in
t,
i
n
ter
s
ca
len
e
b
lo
c
k
s
ar
e
o
f
te
n
p
r
ef
er
r
e
d
.
H
o
wev
er
,
b
o
th
in
ter
s
ca
len
e
an
d
s
u
p
r
ac
l
av
icu
lar
b
l
o
ck
s
ar
e
en
d
o
r
s
ed
as
h
ig
h
ly
s
af
e
m
et
h
o
d
s
f
o
r
ad
m
in
is
ter
in
g
r
e
g
io
n
al
an
esth
esia,
p
ar
ticu
lar
ly
wh
en
ap
p
lie
d
with
u
ltra
s
o
u
n
d
g
u
id
an
ce
.
T
h
is
ass
ess
m
en
t
u
n
d
er
s
co
r
es
th
e
im
p
o
r
tan
ce
of
ch
o
o
s
in
g
th
e
ap
p
r
o
p
r
iate
n
er
v
e
b
lo
ck
tech
n
iq
u
e
b
ased
on
s
af
ety
,
e
f
f
icac
y
,
an
d
th
e
s
p
ec
if
ic
clin
ical
co
n
tex
t,
h
i
g
h
lig
h
tin
g
th
e
ad
v
an
ce
m
e
n
ts
in
u
ltra
s
o
u
n
d
-
g
u
id
e
d
r
eg
i
o
n
al
an
esth
esia
p
r
ac
tices.
Gao
et
a
l.
[
9
]
co
n
d
u
cted
a
s
u
r
v
ey
on
UT
Net
,
a
h
y
b
r
id
tr
a
n
s
f
o
r
m
er
f
o
r
m
ed
ical
im
a
g
e
s
eg
m
en
tatio
n
,
wh
ich
co
m
b
in
es
co
n
v
o
l
u
tio
n
a
l
n
etwo
r
k
s
with
s
elf
-
atten
tio
n
m
o
d
u
les.
It
d
e
m
o
n
s
tr
ates
s
u
p
er
io
r
p
e
r
f
o
r
m
an
ce
an
d
r
o
b
u
s
tn
ess
,
s
h
o
win
g
p
r
o
m
is
e
f
o
r
m
ed
ical
v
is
io
n
,
a
lth
o
u
g
h
th
is
f
ield
r
em
ai
n
s
lar
g
ely
u
n
e
x
p
lo
r
e
d
.
Dh
ab
h
ai
an
d
Gu
p
ta
[
1
0
]
e
x
am
in
ed
m
e
d
ical
d
ata
g
r
o
wth
,
em
p
h
asizin
g
SVM'
s
e
f
f
icien
cy
in
im
a
g
e
class
if
icatio
n
.
T
h
eir
r
ev
iew
h
ig
h
lig
h
ts
SVM
'
s
ac
cu
r
ac
y
an
d
ef
f
ec
tiv
en
ess
f
o
r
m
an
ag
in
g
ex
p
an
d
in
g
m
ed
ical
im
ag
e
d
atasets
.
Ma
r
u
y
am
a
et
a
l.
[
1
1
]
co
m
p
a
r
ed
m
ed
ical
im
a
g
e
class
if
icatio
n
ac
cu
r
ac
y
ac
r
o
s
s
SVM,
a
r
tifi
cial
n
eu
r
al
n
etwo
r
k
(
ANN
)
,
an
d
C
NN
m
o
d
els.
T
h
ey
in
v
esti
g
ated
th
e
i
m
p
ac
t
of
lo
w
-
q
u
ality
im
ag
es
on
ac
cu
r
ac
y
u
s
in
g
J
PEG
an
d
DI
C
OM
d
atasets
,
f
in
d
in
g
C
NN
to
o
u
tp
er
f
o
r
m
o
th
er
s
in
ef
f
icien
cy
an
d
ac
c
u
r
a
cy
.
C
ai
et
a
l.
[
1
2
]
an
aly
ze
d
b
ig
m
e
d
ical
d
ata'
s
r
o
le
in
m
e
d
ical
im
ag
in
g
with
d
ee
p
lear
n
in
g
,
h
ig
h
lig
h
tin
g
its
s
u
cc
ess
in
im
ag
e
d
etec
tio
n
,
s
eg
m
en
tatio
n
,
an
d
c
lass
if
icatio
n
.
T
h
ey
d
is
cu
s
s
ed
c
h
allen
g
es,
5
G'
s
im
p
ac
t,
an
d
p
r
ed
icted
in
tellig
en
t
m
ed
ical
d
ev
ice
ad
v
an
ce
m
e
n
ts
.
Z
h
o
u
et
a
l.
[
1
3
]
in
tr
o
d
u
ce
d
a
n
o
v
el
ac
tiv
e
co
n
to
u
r
m
o
d
el
f
o
r
m
ed
ical
im
ag
e
s
eg
m
en
tatio
n
,
u
tili
zin
g
a
u
n
if
ied
Gau
s
s
ian
d
is
tr
ib
u
tio
n
f
r
am
ewo
r
k
.
It
co
m
b
in
es
g
lo
b
al
an
d
lo
ca
l
en
er
g
y
ter
m
s
to
en
h
an
ce
co
n
t
o
u
r
ev
o
lu
tio
n
a
n
d
p
r
ec
is
io
n
.
Ad
ap
tiv
e
weig
h
tin
g
im
p
r
o
v
es
h
an
d
lin
g
of
in
ten
s
ity
in
h
o
m
o
g
en
eities,
s
h
o
win
g
s
u
p
er
io
r
e
f
f
icien
cy
an
d
ac
cu
r
ac
y
in
e
x
ten
s
iv
e
s
y
n
th
etic
an
d
r
ea
l
im
a
g
e
test
s
.
Sh
ah
et
a
l
.
[
1
4
]
ex
p
lo
r
e
d
r
ice
d
is
ea
s
e
d
etec
tio
n
u
s
in
g
d
ee
p
lear
n
in
g
an
d
tr
an
s
f
er
lear
n
in
g
with
m
o
d
els
lik
e
R
esNe
t
5
0
,
ac
h
iev
in
g
up
to
9
9
.
7
5
%
ac
cu
r
ac
y
.
T
h
is
r
esear
ch
s
er
v
es
as
a
r
ef
er
en
ce
f
o
r
u
n
d
er
s
tan
d
i
n
g
t
h
e
p
er
f
o
r
m
an
c
e
of
I
n
c
ep
tio
n
V3
,
VGG
16,
C
NN
,
an
d
R
esNet5
0
f
r
am
ewo
r
k
s
f
o
r
class
ic
d
etec
tio
n
p
r
o
b
lem
s
.
I
k
ec
h
u
k
wu
et
a
l.
[
1
5
]
c
o
m
p
ar
ed
R
esNet
-
50,
VGG
-
19,
an
d
tr
ad
itio
n
al
tr
ain
in
g
f
o
r
class
if
y
in
g
p
n
eu
m
o
n
ia
f
r
o
m
c
h
est
x
-
r
a
y
s
.
Hig
h
lig
h
tin
g
d
ee
p
lear
n
in
g
'
s
h
ig
h
c
o
m
p
u
tatio
n
al
d
em
a
n
d
s
,
th
ey
n
o
ted
d
r
o
p
o
u
t
r
eg
u
lar
izatio
n
'
s
r
o
le
in
ef
f
icien
cy
.
T
h
eir
ap
p
r
o
ac
h
ac
h
ie
v
ed
a
9
2
.
0
3
%
r
ec
all,
m
atch
in
g
f
i
n
e
-
tu
n
ed
p
r
e
-
tr
ain
e
d
m
o
d
els.
Kh
air
a
n
d
is
h
et
a
l.
[
1
6
]
ex
p
lo
r
e
d
m
a
g
n
etic
r
eso
n
an
ce
im
ag
in
g
(
MRI
)
b
r
ain
im
ag
e
class
if
icatio
n
an
d
tu
m
o
r
d
etec
tio
n
u
s
in
g
a
C
NN
-
SVM
h
y
b
r
i
d
.
Ach
ie
v
in
g
9
8
.
4
9
5
9
%
ac
c
u
r
ac
y
,
th
is
m
et
h
o
d
o
lo
g
y
d
em
o
n
s
tr
ates
th
e
h
y
b
r
i
d
'
s
ef
f
ec
tiv
en
ess
,
u
tili
zin
g
s
tatis
tical
f
u
n
ctio
n
s
an
d
th
e
B
R
AT
S
2015
d
ataset,
d
er
iv
ed
f
r
o
m
ea
r
lier
v
er
s
io
n
s
,
f
o
r
co
m
p
r
eh
en
s
iv
e
a
n
aly
s
is
an
d
s
ig
n
if
ican
t
tu
m
o
r
d
etec
tio
n
ad
v
a
n
ce
m
en
ts
.
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
E
n
h
a
n
ci
n
g
u
ltr
a
s
o
u
n
d
-
g
u
id
ed
b
r
a
ch
ia
l p
lexu
s
n
erve
lo
c
a
liz
a
tio
n
…
(
S
o
b
h
a
n
a
Mu
mma
n
en
i
)
4941
Din
g
et
a
l.
[
1
7
]
s
eg
m
e
n
ted
th
e
b
r
ac
h
ial
p
lex
u
s
in
u
ltra
s
o
u
n
d
im
ag
es
u
s
in
g
Ma
lles
Net,
b
ased
on
R
-
C
NN
an
d
en
h
an
ce
d
with
s
h
o
r
t
-
liv
ed
clim
ate
f
o
r
ce
r
s
(
SLCF
)
an
d
s
to
ch
asti
c
av
er
ag
e
g
r
ad
ien
t
(
SAG)
f
o
r
d
etailed
co
n
tr
ast
an
d
s
p
atial
f
ea
tu
r
es.
A
lo
ca
l
d
ataset
of
1
0
1
p
atien
t
im
ag
es
v
alid
ated
th
e
m
eth
o
d
'
s
ef
f
ec
tiv
en
ess
.
B
o
wn
ess
et
a
l.
[
1
8
]
c
o
m
p
ar
e
d
an
esth
esio
lo
g
is
t
ass
ess
m
en
ts
an
d
AI
-
g
en
er
ate
d
co
lo
r
o
v
e
r
lay
s
by
Scan
Nav
on
u
ltra
s
o
u
n
d
im
a
g
es
f
o
r
u
ltra
s
o
u
n
d
-
g
u
id
e
d
r
e
g
io
n
al
a
n
ae
s
th
esia
(
UGRA)
.
Utilizin
g
d
ice
an
d
Hau
s
d
o
r
f
f
co
ef
f
icien
ts
,
th
ey
an
aly
ze
d
v
ar
io
u
s
n
er
v
e
b
lo
ck
s
,
f
in
d
in
g
b
o
th
h
u
m
an
an
d
AI
ap
p
r
o
ac
h
es
h
ig
h
ly
ac
cu
r
ate,
with
h
u
m
an
s
s
h
o
win
g
co
n
s
is
ten
t
s
co
r
es
an
d
AI
e
x
h
ib
itin
g
m
in
im
al
v
a
r
iatio
n
.
Jo
et
a
l.
[
1
9
]
ex
p
lo
r
ed
s
u
p
r
ac
l
av
icu
lar
b
l
o
ck
s
eg
m
e
n
tatio
n
u
s
in
g
co
m
p
u
ter
-
aid
e
d
d
iag
n
o
s
i
s
(
C
ADx
)
an
d
C
NNs,
co
n
tr
asti
n
g
tr
ad
iti
o
n
al
im
ag
e
d
atasets
with
v
id
eo
d
ata
f
o
r
en
h
an
ce
d
Stan
d
ar
d
C
h
ar
ter
ed
B
an
k
(
SC
B
)
ad
m
in
is
tr
atio
n
.
T
h
ey
in
teg
r
ated
R
esNet
f
o
r
class
i
f
icatio
n
an
d
U
-
Net
f
o
r
s
eg
m
en
tatio
n
with
in
a
ca
s
ca
d
in
g
f
r
am
ew
o
r
k
,
o
f
f
e
r
in
g
an
in
-
d
e
p
th
co
m
p
ar
ativ
e
an
aly
s
is
of
v
ar
i
o
u
s
C
NN
m
o
d
els,
en
h
a
n
cin
g
d
etec
tio
n
a
n
d
s
eg
m
en
tatio
n
ac
cu
r
ac
y
.
T
ian
et
a
l.
[
2
0
]
ev
alu
ated
12
d
ee
p
lea
r
n
in
g
m
o
d
els
f
o
r
b
r
ac
h
ial
p
lex
u
s
s
eg
m
en
tatio
n
in
u
ltra
s
o
u
n
d
i
m
ag
es,
cr
itical
f
o
r
an
esth
esia.
U
-
Net
e
x
ce
lled
with
6
8
.
5
0
%
in
ter
s
ec
tio
n
o
v
er
u
n
io
n
(
I
o
U
)
but
s
lo
w
p
r
o
ce
s
s
in
g
,
wh
er
ea
s
L
in
k
Net
b
alan
ce
d
s
p
ee
d
an
d
ac
cu
r
ac
y
,
id
ea
l
f
o
r
r
ea
l
-
tim
e
u
s
e.
C
h
allen
g
es
in
clu
d
e
a
s
m
all
d
at
aset
an
d
th
e
n
ee
d
f
o
r
f
u
tu
r
e
ad
v
an
ce
m
en
ts
.
3.
M
E
T
H
O
DO
L
O
G
Y
T
h
e
f
o
llo
win
g
s
y
s
tem
ar
ch
itectu
r
e
in
Fig
u
r
e
1
clea
r
ly
p
o
r
tr
ay
s
th
e
wo
r
k
in
g
m
eth
o
d
o
lo
g
y
an
d
r
eq
u
ir
ed
p
ar
am
eter
s
.
Fo
r
a
b
et
ter
u
n
d
er
s
tan
d
i
n
g
.
T
h
e
p
h
ases
in
th
e
s
u
g
g
ested
m
o
d
el
ar
e
s
h
o
wn
in
th
is
p
r
o
ce
s
s
f
lo
w
d
iag
r
a
m
.
Fo
llo
win
g
ar
e
th
e
m
ain
m
o
d
u
les
of
t
h
e
f
lo
w
d
iag
r
am
:
u
ltra
s
o
u
n
d
im
ag
e
d
ataset
,
d
ata
p
r
ep
r
o
ce
s
s
in
g
m
o
d
u
le
,
m
o
d
el
tr
ain
in
g
,
m
o
d
el
ev
alu
atio
n
,
an
d
r
e
g
io
n
al
class
if
icatio
n
.
E
ac
h
m
o
d
u
le
h
as
its
o
wn
u
s
ag
e,
to
p
r
o
d
u
ce
t
h
e
f
in
al
m
ask
ed
im
ag
e
of
th
e
u
ltra
s
o
u
n
d
i
m
ag
es.
I
n
c
o
r
p
o
r
atin
g
s
ep
ar
ate
m
o
d
els
f
o
r
s
eg
m
en
tatio
n
an
d
class
if
icatio
n
h
elp
s
in
ac
h
ie
v
in
g
r
esu
lts
with
h
ig
h
e
r
ac
cu
r
ac
y
.
SVM
alg
o
r
ith
m
s
ar
e
wid
ely
k
n
o
wn
f
o
r
th
ei
r
ex
ce
llen
t
p
e
r
f
o
r
m
an
ce
in
im
ag
e
class
if
icatio
n
.
Fig
u
r
e
1.
Pro
ce
s
s
f
lo
w
d
iag
r
a
m
of
th
e
p
r
o
p
o
s
ed
s
eg
m
en
tati
o
n
an
d
class
if
icatio
n
m
o
d
el
3
.
1
.
Ult
r
a
s
o
un
d
im
a
g
e
da
t
a
s
et
A
d
ataset
h
as
b
ee
n
u
s
ed
,
wh
ich
is
av
ailab
le
on
Kag
g
le
[
2
1
]
,
f
o
r
an
e
x
ten
s
iv
e
d
ataset
is
n
ee
d
ed
to
tr
ain
b
o
th
R
esNet
an
d
SVM
m
o
d
els.
T
h
is
d
ataset
co
n
s
is
ts
of
r
aw
im
ag
es
of
b
r
ac
h
ial
p
lex
u
s
th
at
wer
e
m
an
u
ally
an
n
o
tated
by
h
u
m
a
n
s
.
It
is
n
o
tewo
r
th
y
th
at
i
n
d
i
v
id
u
als
wh
o
an
n
o
tated
th
ese
im
ag
es
ar
e
tr
ain
ed
p
r
o
f
ess
io
n
als
an
d
ar
e
ask
ed
to
an
n
o
tate
th
e
im
a
g
es
wh
er
e
th
ey
ar
e
co
n
f
id
en
t
of
id
en
tify
in
g
th
e
p
r
esen
ce
of
b
r
ac
h
ial
p
lex
u
s
.
Giv
en
th
at
it
is
all
h
u
m
an
an
n
o
tated
d
at
a,
th
er
e
will
u
n
d
en
iab
ly
be
p
r
esen
ce
of
No
is
e,
Ar
tifa
cts,
an
d
ev
e
n
p
o
ten
tial
m
is
tak
es.
E
lim
in
atio
n
of
th
ese
ir
r
eg
u
lar
ities
n
ee
d
s
to
be
ca
r
r
ied
out
so
th
at
th
e
y
wo
n
’
t
h
in
d
er
th
e
r
esu
lt.
3
.
2
.
D
a
t
a
prepro
ce
s
s
ing
Data
p
r
ep
r
o
ce
s
s
in
g
[
2
2
]
in
im
ag
e
s
eg
m
en
tatio
n
d
en
o
tes
th
e
s
et
of
tech
n
i
q
u
es
ap
p
lied
on
r
aw
im
ag
e
d
ata,
b
ef
o
r
e
b
ein
g
g
iv
en
as
an
in
p
u
t
to
a
m
ac
h
in
e
lea
r
n
in
g
m
o
d
el
f
o
r
s
eg
m
e
n
tatio
n
task
s
.
I
m
ag
e
s
eg
m
en
tatio
n
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.
13
,
No
.
4
,
Dec
em
b
er
2
0
2
4
:
4
9
3
9
-
4
9
4
7
4942
in
v
o
lv
es
d
iv
i
d
in
g
an
im
a
g
e
i
n
to
m
ea
n
i
n
g
f
u
l
an
d
d
is
tin
ct
r
eg
io
n
s
or
o
b
jects.
To
im
p
r
o
v
e
th
e
s
eg
m
en
tatio
n
m
o
d
el'
s
p
er
f
o
r
m
an
ce
an
d
th
e
q
u
ality
of
th
e
in
p
u
t
d
ata,
ef
f
e
ctiv
e
d
ata
p
r
e
p
r
o
ce
s
s
in
g
is
ess
en
tial.
So
m
e
of
th
e
co
m
m
o
n
d
ata
p
r
ep
r
o
ce
s
s
in
g
tech
n
iq
u
es
with
r
esp
ec
t
to
i
m
ag
e
s
eg
m
en
tatio
n
a
n
d
class
if
icatio
n
ar
e
im
ag
e
r
esizin
g
,
n
o
r
m
aliza
tio
n
,
n
o
is
e
r
ed
u
ctio
n
an
d
m
ask
g
en
e
r
atio
n
.
Sev
er
al
p
r
ep
r
o
ce
s
s
in
g
tech
n
iq
u
es
h
a
v
e
b
ee
n
im
p
o
s
ed
on
th
e
in
p
u
t
r
aw
im
a
g
es,
wh
ich
ar
e
d
escr
ib
ed
as
f
o
l
lo
ws
.
3.
2
.
1
.
I
m
a
g
e
re
s
izing
I
m
ag
e
r
esizin
g
[
2
3
]
,
as
a
p
r
e
p
r
o
ce
s
s
in
g
tech
n
iq
u
e
,
p
r
o
v
id
e
s
u
n
if
o
r
m
ity
,
co
m
p
u
tatio
n
al
e
f
f
icien
cy
,
m
o
d
el
co
m
p
atib
ilit
y
,
g
en
e
r
aliza
tio
n
,
o
v
er
f
itti
n
g
p
r
ev
en
tio
n
,
r
ed
u
ce
d
tr
ain
in
g
tim
e,
d
ata
a
u
g
m
en
tatio
n
.
I
m
ag
e
r
esizin
g
is
a
f
u
n
d
am
e
n
tal
p
r
ep
r
o
ce
s
s
in
g
s
tep
th
at
a
d
d
r
ess
es
is
s
u
es
r
elate
d
to
d
ata
co
n
s
is
ten
cy
,
co
m
p
u
tatio
n
a
l
ef
f
icien
cy
,
an
d
m
o
d
el
co
m
p
a
tib
ilit
y
.
It
p
lay
s
a
cr
u
cial
r
o
le
in
p
r
ep
ar
in
g
im
ag
e
d
ata
f
o
r
v
ar
io
u
s
m
ac
h
in
e
lear
n
in
g
task
s
,
in
clu
d
i
n
g
im
ag
e
s
eg
m
en
tatio
n
,
by
cr
ea
tin
g
a
s
tan
d
ar
d
ized
in
p
u
t
f
o
r
m
at
f
o
r
t
h
e
m
o
d
el.
3.
2
.
2
.
D
a
t
a
no
rm
a
liza
t
io
n
Data
n
o
r
m
aliza
tio
n
[
2
4
]
is
a
p
r
ep
r
o
ce
s
s
in
g
tech
n
iq
u
e
u
s
ed
to
s
ca
le
an
d
tr
an
s
f
o
r
m
in
p
u
t
d
ata
to
a
s
tan
d
ar
d
r
an
g
e,
m
ak
in
g
it
ea
s
ier
f
o
r
m
ac
h
i
n
e
lear
n
in
g
m
o
d
els
to
lear
n
p
atter
n
s
an
d
g
e
n
e
r
alize
well
to
n
ew,
u
n
s
ee
n
d
ata.
No
r
m
aliza
tio
n
p
r
o
ce
s
s
ty
p
ically
in
v
o
lv
es
ad
ju
s
tin
g
th
e
s
ca
le
an
d
d
is
tr
ib
u
tio
n
of
th
e
d
ata
with
o
u
t
ch
an
g
in
g
its
in
h
er
e
n
t
s
tr
u
ctu
r
e.
T
h
e
k
ey
asp
ec
ts
in
clu
d
e
c
o
n
s
is
ten
t
s
ca
lin
g
,
co
n
v
e
r
g
en
c
e
s
p
ee
d
,
m
itig
atin
g
s
en
s
itiv
ity
an
d
s
atu
r
atio
n
av
o
id
an
ce
.
T
h
is
is
a
cr
u
cial
p
r
ep
r
o
ce
s
s
in
g
s
tep
in
im
ag
e
s
eg
m
en
tatio
n
an
d
class
if
icatio
n
ac
tiv
ities
an
d
en
s
u
r
es
th
at
th
e
in
p
u
t
d
ata
is
in
a
s
u
itab
le
f
o
r
m
f
o
r
th
e
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
.
3.
2
.
3
.
G
a
us
s
ia
n
f
ilte
ring
Gau
s
s
ian
f
ilter
in
g
[
2
5
]
is
an
im
ag
e
f
ilter
in
g
tech
n
i
q
u
e
t
h
at
in
v
o
l
v
es
co
n
v
o
lv
in
g
an
im
a
g
e
with
a
Gau
s
s
ian
f
u
n
ctio
n
,
wh
ic
h
in
t
u
r
n
,
is
a
m
ath
em
atica
l
f
u
n
ctio
n
th
at
r
ep
r
esen
ts
a
n
o
r
m
al
d
is
tr
ib
u
tio
n
,
o
f
ten
u
s
ed
in
im
ag
e
p
r
o
ce
s
s
in
g
f
o
r
its
s
m
o
o
th
in
g
p
r
o
p
e
r
ties
.
T
h
is
p
r
o
ce
s
s
in
v
o
lv
es
s
l
id
in
g
th
e
Gau
s
s
i
an
K
er
n
el
o
v
er
th
e
im
ag
e
an
d
co
m
p
u
tin
g
a
weig
h
ted
av
e
r
ag
e
of
t
h
e
p
i
x
el
v
al
u
es
with
in
th
e
k
er
n
el'
s
n
eig
h
b
o
r
h
o
o
d
.
Gau
s
s
ian
f
ilter
is
a
b
ell
-
s
h
ap
ed
cu
r
v
e
th
at
ass
ig
n
s
weig
h
ts
to
th
e
ce
n
tr
al
p
ix
els
wh
ile
g
r
ad
u
ally
d
ec
r
ea
s
in
g
weig
h
ts
as
you
m
o
v
e
away
f
r
o
m
t
h
e
ce
n
ter
.
Key
asp
ec
ts
of
Gau
s
s
ian
f
ilter
in
g
in
clu
d
e
s
m
o
o
th
in
g
,
e
d
g
e
d
etec
tio
n
a
n
d
f
ea
tu
r
e
ex
tr
ac
tio
n
.
3.
2
.
4
.
B
ina
ry
m
a
s
k
ing
B
in
ar
y
m
ask
in
g
[
2
6
]
is
a
p
r
ep
r
o
ce
s
s
in
g
tech
n
iq
u
e
co
m
m
o
n
ly
u
s
ed
in
im
a
g
e
s
eg
m
en
t
atio
n
an
d
class
if
icatio
n
ta
s
k
s
.
It
in
v
o
lv
es
cr
ea
tin
g
a
b
in
ar
y
im
ag
e,
also
k
n
o
wn
as
a
m
ask
,
th
at
h
i
g
h
lig
h
ts
or
is
o
lates
s
p
ec
if
ic
r
eg
io
n
s
or
o
b
jects
of
i
n
ter
est
with
in
th
e
o
r
ig
in
al
im
a
g
e.
It
ty
p
ically
co
n
tain
s
v
alu
es
of
0
an
d
1,
wh
er
e
0
r
ep
r
esen
ts
th
e
r
e
g
io
n
s
n
o
t
of
in
ter
est,
an
d
1
r
ep
r
esen
ts
th
e
r
eg
io
n
s
of
in
ter
est.
B
in
ar
y
m
ask
in
g
is
u
s
ed
f
o
r
s
ev
er
al
task
s
s
u
ch
as
r
eg
io
n
of
in
ter
est
(
R
OI
)
ex
tr
ac
tio
n
,
f
o
r
eg
r
o
u
n
d
-
b
a
ck
g
r
o
u
n
d
s
ep
a
r
atio
n
an
d
tr
ain
i
n
g
d
ata
m
o
d
if
icatio
n
.
Fig
u
r
e
2
illu
s
tr
ates
th
e
r
esu
ltan
t
p
r
ep
r
o
ce
s
s
ed
im
ag
e
f
o
r
an
e
x
em
p
lar
u
ltra
s
o
u
n
d
im
ag
e.
Fig
u
r
e
2.
Ultr
aso
u
n
d
im
ag
e
b
e
f
o
r
e
an
d
af
ter
p
r
ep
r
o
ce
s
s
in
g
3
.
3
.
S
pli
t
t
ing
t
he
da
t
a
Data
s
p
litt
in
g
,
or
d
iv
id
in
g
th
e
d
ata
in
to
tr
ain
in
g
an
d
test
in
g
s
ets,
is
an
im
p
o
r
tan
t
s
tag
e
in
m
ac
h
in
e
lear
n
in
g
t
h
at
allo
ws
y
o
u
to
as
s
ess
a
m
o
d
el'
s
p
er
f
o
r
m
an
ce
.
Data
s
p
litt
in
g
is
p
er
f
o
r
m
ed
to
ev
alu
ate
how
well
th
e
m
o
d
el
g
e
n
er
alize
s
to
n
ew,
u
n
o
b
s
er
v
ed
d
ata.
W
ith
r
esp
e
ct
to
th
e
p
r
o
p
o
s
ed
a
p
p
r
o
ac
h
,
th
e
d
ataset
h
as
b
ee
n
s
p
lit
in
to
a
r
atio
of
8
0
:2
0
.
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
E
n
h
a
n
ci
n
g
u
ltr
a
s
o
u
n
d
-
g
u
id
ed
b
r
a
ch
ia
l p
lexu
s
n
erve
lo
c
a
liz
a
tio
n
…
(
S
o
b
h
a
n
a
Mu
mma
n
en
i
)
4943
3
.
4
.
Resid
ua
l
net
wo
rk
wit
h
50
la
y
er
s
R
esNet
-
50
[
2
7
]
,
o
f
te
n
r
e
f
er
r
ed
to
as
r
esid
u
al
n
etwo
r
k
with
50
lay
er
s
,
is
a
d
ee
p
C
NN
ar
ch
it
ec
tu
r
e
th
at
h
as
s
h
o
wn
im
p
r
ess
iv
e
p
er
f
o
r
m
an
ce
s
eg
m
en
tatio
n
,
o
b
ject
r
ec
o
g
n
itio
n
,
an
d
p
h
o
to
class
if
icatio
n
.
R
esNet
-
50
b
elo
n
g
s
to
a
f
am
ily
of
R
esN
et
ar
ch
itectu
r
es
in
tr
o
d
u
ce
d
to
p
r
o
v
id
e
way
s
of
tr
ain
in
g
d
e
ep
n
eu
r
al
n
etwo
r
k
s
.
R
esNet
50
is
u
s
ed
f
o
r
s
eg
m
en
tatio
n
,
b
ec
a
u
s
e
it
c
o
n
s
is
ts
of
s
k
ip
co
n
n
ec
tio
n
s
,
u
s
ed
to
f
ac
ilit
ate
d
ee
p
lear
n
in
g
n
etwo
r
k
s
wh
ich
m
itig
ate
th
e
m
o
s
t
co
m
m
o
n
v
an
is
h
in
g
g
r
ad
ien
t
p
r
o
b
lem
.
Alg
o
r
ith
m
1
d
e
s
cr
ib
es
th
e
wo
r
k
in
g
p
r
o
ce
s
s
of
R
esNet5
0
f
o
r
im
a
g
e
class
if
icatio
n
.
Alg
o
r
ith
m
1
:
R
es
N
et5
0
alg
o
r
ith
m
I
n
p
u
t: I
n
p
u
t im
a
g
e
X,
R
esid
u
a
l b
lo
ck
f
u
n
ctio
n
s
F
i
f
o
r
i =
1
,
2
,
n
,
L
ea
r
n
in
g
r
ate
α
Ou
tp
u
t: C
lass
if
icatio
n
r
esu
lt
1.
I
n
itialize
th
e
co
n
v
o
lu
tio
n
al
a
n
d
b
atch
n
o
r
m
aliza
tio
n
p
ar
am
et
er
s
2.
f
o
r
i =
1
to
n
d
o
3.
Fo
r
war
d
p
ass
th
r
o
u
g
h
t
h
e
r
esid
u
al
b
lo
c
k
:
4.
y
=
F
i
(
X)
5.
Up
d
ate
X
=
y
6.
en
d
f
o
r
7.
Glo
b
al
av
er
ag
e
p
o
o
lin
g
: X
=
Glo
b
alAv
g
Po
o
l(
X)
8.
Fu
lly
co
n
n
ec
ted
lay
er
: y
=
FC
(
X)
9.
So
f
tm
ax
ac
tiv
atio
n
: y
=
So
f
tm
ax
(
y
)
10.
R
etu
r
n
y
3
.
5
.
Su
pp
o
rt
v
ec
t
o
r
ma
chine
cla
s
s
if
ier
T
h
e
a
i
m
of
S
VM
[
2
8
]
is
to
f
in
d
a
h
y
p
e
r
p
l
a
n
e
w
h
i
c
h
s
e
p
a
r
at
e
s
d
i
f
f
e
r
e
n
t
c
l
as
s
e
s
in
t
h
e
f
e
at
u
r
e
s
p
a
c
e
.
U
l
t
r
as
o
u
n
d
i
m
a
g
i
n
g
of
t
h
e
b
r
a
c
h
i
a
l
p
l
e
x
u
s
is
a
m
e
d
ic
a
l
im
a
g
i
n
g
m
o
d
a
l
i
t
y
u
s
e
d
to
v
i
s
u
a
l
i
z
e
n
e
r
v
es
in
t
h
e
s
h
o
u
l
d
e
r
r
e
g
i
o
n
.
S
V
M
c
l
ass
i
f
i
er
can
be
e
m
p
l
o
y
e
d
f
o
r
t
h
e
c
l
ass
i
f
i
c
at
i
o
n
of
i
n
p
u
t
i
m
a
g
es
to
d
if
f
e
r
e
n
t
i
a
t
e
b
e
t
w
ee
n
d
i
f
f
e
r
e
n
t
s
t
r
u
c
t
u
r
es
or
c
o
n
d
i
t
i
o
n
s
.
It
is
a
d
v
a
n
t
a
g
e
o
u
s
f
o
r
c
l
a
s
s
i
f
y
i
n
g
t
h
e
b
r
a
c
h
i
a
l
p
l
e
x
u
s
in
u
l
t
r
a
s
o
u
n
d
i
m
a
g
e
s
by
l
e
a
r
n
i
n
g
a
n
d
r
e
c
o
g
n
i
z
i
n
g
p
a
t
t
e
r
n
s
in
t
h
e
e
x
t
r
a
c
t
e
d
f
e
a
t
u
r
e
s
,
u
l
t
i
m
a
t
e
l
y
as
s
i
s
ti
n
g
h
e
a
l
t
h
c
ar
e
p
r
o
f
e
s
s
i
o
n
a
ls
in
m
a
k
i
n
g
a
c
c
u
r
a
t
e
d
i
a
g
n
o
s
es
.
Alg
o
r
i
t
h
m
2
d
e
s
c
r
i
b
e
s
t
h
e
w
o
r
k
i
n
g
p
r
o
c
e
s
s
of
SV
M
f
o
r
i
m
a
g
e
c
la
s
s
i
f
i
c
at
i
o
n
.
Alg
o
r
ith
m
2:
I
m
ag
e
class
if
icatio
n
u
s
in
g
s
v
m
I
n
p
u
t:
T
r
ai
n
in
g
d
ata
(X
i
,
y
i
),
K
er
n
el
f
u
n
ctio
n
K,
R
eg
u
lar
izati
o
n
p
ar
a
m
eter
C,
Un
lab
eled
im
ag
e
X
test
Ou
tp
u
t:
Pre
d
icted
class
lab
el
f
o
r
X
test
1.
Pre
p
r
o
ce
s
s
th
e
d
ata:
Featu
r
e
e
x
tr
ac
tio
n
,
an
d
n
o
r
m
aliza
tio
n
2.
I
n
itialize
th
e
p
ar
am
eter
s
:
w
=
0,
b
=
0
3.
C
h
o
o
s
e
a
k
er
n
el
f
u
n
ctio
n
K
4.
C
o
n
s
tr
u
ct
th
e
k
er
n
el
m
atr
ix
H
u
s
in
g
K:
H
ij
=
y
i
y
j
K(
X
i
,
X
j
)
5.
I
n
itialize
L
ag
r
an
g
e
m
u
ltip
lier
s
α
i
=
0
f
o
r
i
=
1,
2,
.
.
.
,
N
6.
f
o
r
t
=
1
to
N
iter
do
7.
C
o
m
p
u
te
th
e
SVM
d
ec
is
io
n
f
u
n
ctio
n
:
f
(
X)
=
∑
N
i=
1
α
i
y
i
K(
X
i
,
X
j
)
+
b
8.
C
o
m
p
u
te
th
e
h
in
g
e
lo
s
s
:
L
=
1
/
2
∥
w
∥
2
+
C
∑
N
i=
1
m
ax
(
0
,
1
−
y
i
f
(X
i
))
9.
C
o
m
p
u
te
th
e
g
r
ad
ien
ts
of
L
with
r
esp
ec
t
to
w
an
d
b
10.
Up
d
ate
th
e
p
ar
a
m
eter
s
u
s
in
g
g
r
ad
ien
t
d
escen
t
11.
Pro
ject
α
o
n
to
th
e
f
ea
s
ib
le
s
et
12.
en
d
f
o
r
13.
C
las
s
if
y
th
e
test
im
ag
e:
14.
C
o
m
p
u
te
th
e
d
ec
is
io
n
f
u
n
ctio
n
f
o
r
X
test
:
f
(X
test
)
=
∑
N
i=
1
α
i
y
i
K(
X
test
,
X
i
)
+
b
15.
Pre
d
icted
class
lab
el:
y
pred
=
s
i
g
n
(
f
(X
test
))
16.
R
etu
r
n
y
pred
3
.
6
.
E
ns
em
ble
m
o
del
E
n
s
em
b
lin
g
R
esNet5
0
with
a
SVM
[
2
9
]
can
cr
ea
te
a
r
o
b
u
s
t
an
d
ef
f
ec
ti
v
e
m
o
d
el
f
o
r
v
ar
i
o
u
s
m
ac
h
in
e
lear
n
in
g
task
s
.
R
esNet5
0
ex
c
els
at
f
ea
tu
r
e
ex
tr
ac
tio
n
f
r
o
m
co
m
p
lex
d
ata,
p
ar
ticu
lar
ly
in
im
ag
e
r
ec
o
g
n
itio
n
task
s
.
By
co
m
b
in
in
g
its
d
ee
p
lear
n
in
g
ca
p
a
b
ilit
ies
with
an
S
VM
class
if
ier
,
th
e
en
s
em
b
le
m
o
d
el
lev
er
ag
es
th
e
s
tr
en
g
th
s
of
b
o
th
m
o
d
els.
R
esNet5
0
ex
tr
ac
ts
in
tr
icate
h
ie
r
ar
ch
ical
f
ea
tu
r
es,
an
d
SVM
r
ef
in
es
th
e
d
ec
is
io
n
b
o
u
n
d
ar
ies
b
ased
on
th
ese
f
ea
tu
r
es,
r
esu
ltin
g
in
a
more
ac
c
u
r
ate
m
o
d
el.
T
h
is
ap
p
r
o
ac
h
le
ad
s
to
an
im
p
r
o
v
ed
p
er
f
o
r
m
an
ce
,
as
t
h
e
en
s
em
b
l
e
b
en
ef
its
f
r
o
m
th
e
co
m
p
lem
en
tar
y
s
tr
en
g
t
h
s
of
d
ee
p
lea
r
n
in
g
a
n
d
tr
a
d
itio
n
al
m
ac
h
in
e
lear
n
in
g
tech
n
iq
u
es.
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.
13
,
No
.
4
,
Dec
em
b
er
2
0
2
4
:
4
9
3
9
-
4
9
4
7
4944
4.
RE
SU
L
T
S
AND
D
I
SCU
SS
I
O
N
T
h
e
f
o
llo
win
g
s
ec
tio
n
d
ep
icts
th
e
p
er
f
o
r
m
a
n
ce
of
th
e
p
r
o
p
o
s
ed
R
esNet5
0
+SVM
en
s
em
b
le
m
o
d
el
in
class
if
y
in
g
th
e
in
p
u
t
u
ltra
s
o
u
n
d
im
ag
es.
W
h
ile
p
r
ev
io
u
s
wo
r
k
s
ad
d
r
ess
ed
th
e
u
s
ag
e
of
p
r
e
-
tr
ain
ed
m
o
d
els
in
ac
h
iev
in
g
th
e
s
am
e,
n
o
n
e
of
th
e
w
o
r
k
s
e
m
p
h
asized
th
e
u
s
ag
e
of
en
s
em
b
le
ap
p
r
o
ac
h
es,
wh
ich
p
r
o
v
ed
th
em
s
elv
es
to
be
more
ef
f
ici
en
t
th
an
tr
ad
itio
n
al
m
o
d
els,
e
s
p
ec
ially
in
class
if
icatio
n
p
r
o
b
lem
s
.
T
h
e
r
esu
lts
m
en
tio
n
ed
ea
r
lier
can
be
d
is
p
lay
ed
as
tab
les,
g
r
ap
h
s
,
or
f
ig
u
r
es.
T
h
e
r
esear
c
h
d
is
cu
s
s
io
n
is
m
ad
e
in
t
o
s
ev
er
a
l
s
u
b
-
s
ec
tio
n
s
,
n
am
ely
co
n
f
u
s
io
n
m
atr
ix
,
r
ec
eiv
e
r
-
o
p
er
atin
g
ch
ar
ac
ter
is
tic
cu
r
v
e
(
R
OC
)
-
a
r
ea
u
n
d
er
th
e
cu
r
v
e
(
AUC
)
,
ac
cu
r
ac
y
cu
r
v
e,
lo
s
s
c
u
r
v
e
,
a
n
d
u
s
er
i
n
ter
f
ac
e.
4
.
1
.
Co
nfusi
o
n
m
a
t
rix
a
nd
RO
C
-
AUC
curv
e
In
th
e
f
o
llo
win
g
co
n
f
u
s
io
n
m
atr
ix
in
Fig
u
r
e
3,
a
l
o
wer
n
u
m
b
er
of
f
alse
p
o
s
itiv
es
in
d
icate
s
th
at
th
e
m
o
d
el
is
m
ak
in
g
f
ewe
r
m
is
tak
es
in
p
r
ed
ictin
g
p
o
s
itiv
e
i
n
s
tan
ce
s
wh
en
th
ey
ar
e
not
ac
tu
ally
p
o
s
itiv
e.
Similar
ly
,
a
less
er
n
u
m
b
er
of
f
alse
n
eg
ativ
es
in
d
icate
s
th
at
th
at
th
e
m
o
d
el
is
m
a
k
in
g
f
ewe
r
m
is
tak
es
in
f
ailin
g
to
id
en
tify
p
o
s
itiv
e
in
s
tan
ce
s
.
Fig
u
r
e
4
illu
s
tr
ates
th
e
R
OC
-
AUC
cu
r
v
e
with
a
h
ig
h
v
alu
e
of
tr
u
e
p
o
s
itiv
e
r
ate
(
T
PR
)
.
W
h
en
th
e
T
PR
,
also
k
n
o
wn
as
s
en
s
itiv
ity
,
ap
p
r
o
ac
h
es
0
.
9
9
,
it
in
d
icate
s
th
at
th
e
m
o
d
el
is
p
er
f
o
r
m
in
g
ex
ce
p
tio
n
ally
well
in
c
o
r
r
ec
tly
id
en
tify
in
g
p
o
s
itiv
e
in
s
tan
ce
s
.
Fig
u
r
e
3.
C
o
n
f
u
s
io
n
m
atr
i
x
of
en
s
em
b
le
m
o
d
el
Fig
u
r
e
4.
R
OC
-
AUC
cu
r
v
e
of
en
s
em
b
le
m
o
d
el
4
.
2
.
Acc
ura
cy
curv
e
a
nd
lo
s
s
curv
e
T
h
e
f
o
llo
win
g
Fig
u
r
e
5
clea
r
l
y
in
d
icate
s
t
h
at
th
e
en
s
em
b
le
m
o
d
el’
s
ac
cu
r
ac
y
is
r
ea
ch
in
g
n
ea
r
-
1
0
0
s
co
r
es
(
9
9
.
2
7
)
.
In
an
en
s
em
b
le
m
o
d
el,
th
e
tr
ain
in
g
d
ata
i
n
f
lu
en
ce
s
th
e
m
o
d
el'
s
lear
n
in
g
p
r
o
ce
s
s
,
an
d
th
is
lear
n
ed
k
n
o
wled
g
e
is
th
en
ap
p
lied
to
m
ak
e
p
r
ed
ictio
n
s
on
th
e
test
in
g
d
ata.
C
r
o
s
s
-
en
tr
o
p
y
lo
s
s
f
u
n
ctio
n
is
illu
s
tr
ated
in
Fig
u
r
e
6.
T
r
ain
in
g
lo
s
s
m
ea
s
u
r
es
how
well
th
e
m
o
d
el
f
its
tr
ai
n
in
g
d
ata,
wh
ile
test
in
g
lo
s
s
m
ea
s
u
r
es
how
ef
f
icien
tly
th
e
m
o
d
el
will
g
en
e
r
alize
n
ew
a
n
d
u
n
r
ev
ea
led
d
ata,
a
n
d
h
en
ce
t
esti
n
g
lo
s
s
n
ee
d
s
to
be
ev
alu
ated
on
a
s
ep
a
r
ated
d
a
taset.
Fig
u
r
e
5.
Acc
u
r
ac
y
c
u
r
v
e
of
th
e
en
s
em
b
le
m
o
d
el
F
i
g
u
r
e
6
.
C
r
o
s
s
e
n
t
r
o
p
y
l
o
s
s
f
u
n
c
t
i
o
n
o
f
t
h
e
e
n
s
e
m
b
l
e
m
o
d
e
l
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
E
n
h
a
n
ci
n
g
u
ltr
a
s
o
u
n
d
-
g
u
id
ed
b
r
a
ch
ia
l p
lexu
s
n
erve
lo
c
a
liz
a
tio
n
…
(
S
o
b
h
a
n
a
Mu
mma
n
en
i
)
4945
4
.
3
.
User
inte
rf
a
ce
T
h
e
r
esu
lt
o
f
class
if
icatio
n
is
to
b
e
s
h
o
wn
u
s
in
g
a
g
r
ap
h
ic
al
u
s
er
in
te
r
f
ac
e
(
GUI
)
.
T
h
e
f
o
llo
win
g
f
ig
u
r
es
d
ep
ict
th
e
GUI
“
im
ag
e
class
if
icatio
n
Ap
p
”,
wh
ich
tak
es
a
r
aw
u
s
er
-
s
elec
ted
u
ltra
s
o
u
n
d
im
ag
e
as
a
n
in
p
u
t.
T
h
e
o
u
tp
u
t,
wh
ich
is
a
b
in
ar
y
m
ask
,
is
r
ep
r
esen
ted
in
th
e
s
am
e
GUI
.
Fig
u
r
es
7
an
d
8
d
ep
ict
th
e
class
es
u
n
d
er
w
h
ich
th
e
p
r
ep
r
o
ce
s
s
ed
im
ag
es a
r
e
class
if
ied
,
n
am
ely
in
ter
s
ca
len
e
an
d
s
u
p
r
ac
lav
icu
l
ar
.
I
n
ter
s
ca
len
e
Fig
u
r
e
7.
I
m
ag
e
class
if
ied
as
i
n
ter
s
ca
len
e
Su
p
r
ac
lav
icu
lar
Fig
u
r
e
8.
I
m
ag
e
class
if
ied
as
s
u
p
r
ac
lav
icu
lar
5.
CO
NCLU
SI
O
N
In
s
u
m
m
a
r
y
,
t
h
e
p
r
im
ar
y
o
b
jectiv
e
of
t
h
is
wo
r
k
was
to
d
ev
elo
p
a
h
ig
h
l
y
ac
cu
r
ate
m
o
d
el
f
o
r
p
r
ed
ictin
g
th
e
p
r
esen
ce
of
th
e
b
r
ac
h
ial
p
lex
u
s
n
er
v
e
r
eg
io
n
in
lim
b
s
an
d
class
if
y
in
g
it
in
to
"
in
ter
s
ca
len
e
"
an
d
"
s
u
p
r
ac
lav
icu
lar
"
n
er
v
e
b
l
o
ck
s
.
T
h
is
class
if
icatio
n
is
ess
en
tial,
s
in
ce
it
h
el
p
s
an
esth
esio
lo
g
is
ts
d
eter
m
in
e
th
e
ty
p
e
of
n
er
v
e
b
lo
ck
,
th
ey
n
ee
d
to
in
ject,
f
o
r
a
p
ar
ticu
lar
p
atien
t.
Alth
o
u
g
h
r
esu
lts
f
o
r
p
r
ev
io
u
s
ly
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
es
wer
e
p
r
o
m
is
in
g
,
b
r
ac
h
ial
p
le
x
u
s
an
ato
m
y
ca
n
v
ar
y
s
ig
n
if
ican
tly
am
o
n
g
in
d
iv
id
u
als,
in
clu
d
in
g
v
ar
iatio
n
s
in
b
r
a
n
ch
in
g
p
att
er
n
s
an
d
p
o
s
itio
n
al
r
elatio
n
s
h
ip
s
with
s
u
r
r
o
u
n
d
in
g
s
tr
u
ctu
r
es.
In
a
d
d
itio
n
,
ex
tr
ac
tin
g
r
elev
an
t
f
ea
tu
r
es
f
r
o
m
u
ltra
s
o
u
n
d
im
ag
es
to
r
ep
r
esen
t
th
e
an
ato
m
ical
s
tr
u
ctu
r
es
of
th
e
b
r
ac
h
ial
p
lex
u
s
ac
cu
r
ately
ca
n
be
c
h
allen
g
in
g
,
s
in
ce
it
is
ess
en
ti
al
f
o
r
class
if
icatio
n
ac
cu
r
ac
y
.
Owin
g
to
th
ese
ch
allen
g
es,
em
p
lo
y
in
g
R
esn
et5
0
alo
n
g
with
an
SVM
class
if
ier
aim
s
at
p
r
o
v
id
in
g
p
r
ec
is
e
o
u
tp
u
ts
wh
ich
ar
e
to
be
u
s
ed
as
an
aid
by
p
r
ac
titi
o
n
er
s
.
W
h
ile
th
e
cu
r
r
e
n
t
f
o
c
u
s
is
on
th
e
b
r
ac
h
ial
p
le
x
u
s
an
d
u
ltra
s
o
u
n
d
im
ag
e
in
p
u
ts
,
th
e
tr
ain
ed
m
o
d
el
will
be
ad
a
p
ted
a
n
d
e
x
p
an
d
ed
to
e
n
co
m
p
ass
v
ar
i
o
u
s
o
th
e
r
a
n
ato
m
ical
r
eg
io
n
s
with
ap
p
r
o
p
r
iate
m
o
d
if
icatio
n
s
to
th
e
alg
o
r
ith
m
s
an
d
d
iv
e
r
s
e
im
a
g
e
d
atasets
th
r
o
u
g
h
d
ata
au
g
m
en
tatio
n
.
T
h
is
wo
r
k
s
ets
th
e
f
o
u
n
d
atio
n
f
o
r
b
r
o
ad
e
r
ap
p
licatio
n
s
a
n
d
co
llab
o
r
ati
o
n
s
with
s
im
ilar
p
r
o
jects,
lay
i
n
g
th
e
g
r
o
u
n
d
wo
r
k
f
o
r
ad
v
a
n
ce
m
en
ts
in
m
ed
ical
im
ag
e
an
aly
s
is
.
L
o
o
k
in
g
ah
ea
d
,
th
er
e
ar
e
s
ev
er
al
p
r
o
m
is
in
g
av
en
u
es
f
o
r
f
u
t
u
r
e
r
esear
ch
an
d
d
ev
elo
p
m
en
t.
RE
F
E
R
E
NC
E
S
[
1
]
M
.
H
.
H
e
sa
mi
a
n
,
W
.
J
i
a
,
X
.
H
e
,
a
n
d
P
.
K
e
n
n
e
d
y
,
“
D
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
f
o
r
m
e
d
i
c
a
l
i
m
a
g
e
se
g
me
n
t
a
t
i
o
n
:
a
c
h
i
e
v
e
m
e
n
t
s
a
n
d
c
h
a
l
l
e
n
g
e
s
,
”
J
o
u
r
n
a
l
o
f
D
i
g
i
t
al
I
m
a
g
i
n
g
,
v
o
l
.
3
2
,
n
o
.
4
,
p
p
.
5
8
2
-
5
9
6
,
A
u
g
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
2
7
8
-
0
1
9
-
0
0
2
2
7
-
x.
[
2
]
Y
.
El
i
e
z
e
r
e
t
a
l
.,
“
S
u
p
p
r
e
s
si
n
g
m
e
t
a
-
h
o
l
o
g
r
a
p
h
i
c
a
r
t
i
f
a
c
t
s b
y
l
a
s
e
r
c
o
h
e
r
e
n
c
e
t
u
n
i
n
g
,
”
L
i
g
h
t
-
S
c
i
e
n
c
e
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
0
,
n
o
.
1
,
M
a
y
2
0
2
1
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
3
7
7
-
021
-
0
0
5
4
7
-
0.
[
3
]
G
.
C
.
F
e
i
g
l
,
R
.
J.
Li
t
z
,
a
n
d
P
.
M
a
r
h
o
f
e
r
,
“
A
n
a
t
o
my
o
f
t
h
e
b
r
a
c
h
i
a
l
p
l
e
x
u
s
a
n
d
i
t
s
i
mp
l
i
c
a
t
i
o
n
s
f
o
r
d
a
i
l
y
c
l
i
n
i
c
a
l
p
r
a
c
t
i
c
e
:
r
e
g
i
o
n
a
l
a
n
e
st
h
e
s
i
a
i
s
a
p
p
l
i
e
d
a
n
a
t
o
m
y
,
”
R
e
g
i
o
n
a
l
A
n
e
s
t
h
e
si
a
a
n
d
P
a
i
n
Me
d
i
c
i
n
e
,
v
o
l
.
4
5
,
n
o
.
8
,
p
p
.
6
2
0
-
6
2
7
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
3
6
/
r
a
p
m
-
2
0
2
0
-
1
0
1
4
3
5
.
[
4
]
G
.
M
.
H
a
m
i
l
t
o
n
e
t
a
l
.
,
“
P
e
r
i
p
h
e
r
a
l
n
e
r
v
e
b
l
o
c
k
s
f
o
r
a
m
b
u
l
a
t
o
r
y
s
h
o
u
l
d
e
r
s
u
r
g
e
r
y
:
A
p
o
p
u
l
a
t
i
o
n
-
b
a
se
d
c
o
h
o
r
t
st
u
d
y
o
f
o
u
t
c
o
mes
a
n
d
r
e
s
o
u
r
c
e
u
t
i
l
i
z
a
t
i
o
n
,
”
A
n
e
s
t
h
e
si
o
l
o
g
y
,
v
o
l
.
1
3
1
,
n
o
.
6
,
p
p
.
1
2
5
4
-
1
2
6
3
,
D
e
c
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
9
7
/
A
LN
.
0
0
0
0
0
0
0
0
0
0
0
0
2
8
6
5
.
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.
13
,
No
.
4
,
Dec
em
b
er
2
0
2
4
:
4
9
3
9
-
4
9
4
7
4946
[
5
]
M
.
S
o
b
h
a
n
a
,
L.
S
.
M
.
Y
a
r
l
a
g
a
d
d
a
,
K
.
K
.
C
h
i
n
t
a
k
a
y
a
l
a
,
a
n
d
V
.
S
.
N
.
R
.
A
l
a
,
“
B
r
a
c
h
i
a
l
P
l
e
x
u
s
se
g
me
n
t
a
t
i
o
n
a
n
d
d
e
t
e
c
t
i
o
n
t
h
r
o
u
g
h
u
l
t
r
a
s
o
u
n
d
i
ma
g
e
a
n
a
l
y
s
i
s
u
si
n
g
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s,
”
2
0
2
3
4
t
h
I
EEE
G
l
o
b
a
l
C
o
n
f
e
re
n
c
e
f
o
r
Ad
v
a
n
c
e
m
e
n
t
i
n
T
e
c
h
n
o
l
o
g
y
(
G
C
AT)
,
B
a
n
g
a
l
o
r
e
,
I
n
d
i
a
,
2
0
2
3
,
p
p
.
1
-
6
,
d
o
i
:
1
0
.
1
1
0
9
/
G
C
A
T5
9
9
7
0
.
2
0
2
3
.
1
0
3
5
3
4
1
8
.
[
6
]
H
.
Zh
a
n
g
,
Z.
Q
u
,
Y
.
M
i
a
o
,
R
.
Ji
a
,
F
.
Li
,
a
n
d
Z.
H
u
a
,
“
C
o
m
p
a
r
i
s
o
n
b
e
t
w
e
e
n
s
u
b
p
a
r
a
n
e
u
r
a
l
u
p
p
e
r
t
r
u
n
k
a
n
d
c
o
n
v
e
n
t
i
o
n
a
l
i
n
t
e
r
sc
a
l
e
n
e
b
l
o
c
k
s
f
o
r
a
r
t
h
r
o
sc
o
p
i
c
s
h
o
u
l
d
e
r
su
r
g
e
r
y
:
a
r
a
n
d
o
mi
z
e
d
n
o
n
i
n
f
e
r
i
o
r
i
t
y
t
r
i
a
l
,
”
A
n
e
s
t
h
e
si
a
a
n
d
A
n
a
l
g
e
si
a
,
v
o
l
.
1
3
4
,
n
o
.
6
,
p
p
.
1
3
0
8
-
1
3
1
7
,
2
0
2
2
,
d
o
i
:
1
0
.
1
2
1
3
/
A
N
E.
0
0
0
0
0
0
0
0
0
0
0
0
5
9
9
0
.
[
7
]
M
.
S
u
h
a
i
d
i
,
R
.
A
.
K
a
d
i
r
,
a
n
d
S
.
Ti
u
n
,
“
A
r
e
v
i
e
w
o
f
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
m
e
t
h
o
d
s
o
n
mac
h
i
n
e
l
e
a
r
n
i
n
g
,
”
J
o
u
rn
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
S
y
s
t
e
m
a
n
d
T
e
c
h
n
o
l
o
g
y
M
a
n
a
g
e
m
e
n
t
,
v
o
l
.
6
,
n
o
.
2
2
,
p
p
.
5
1
–
5
9
,
S
e
p
.
2
0
2
1
,
d
o
i
:
1
0
.
3
5
6
3
1
/
j
i
s
t
m.
6
2
2
0
0
5
.
[
8
]
D
.
K
a
y
e
e
t
a
l
.
,
“
S
u
p
r
a
c
l
a
v
i
c
u
l
a
r
v
s
.
i
n
f
r
a
c
l
a
v
i
c
u
l
a
r
b
r
a
c
h
i
a
l
p
l
e
x
u
s
n
e
r
v
e
b
l
o
c
k
s:
c
l
i
n
i
c
a
l
,
p
h
a
r
m
a
c
o
l
o
g
i
c
a
l
,
a
n
d
a
n
a
t
o
mi
c
a
l
c
o
n
si
d
e
r
a
t
i
o
n
s
,
”
A
n
e
s
t
h
e
si
o
l
o
g
y
a
n
d
Pa
i
n
Me
d
i
c
i
n
e
,
v
o
l
.
1
1
,
n
o
.
5
,
d
o
i
:
1
0
.
5
8
1
2
/
a
a
p
m.
1
2
0
6
5
8
.
[
9
]
Y
.
G
a
o
,
M
.
Zh
o
u
,
a
n
d
D
.
N
.
M
e
t
a
x
a
s,
“
U
TN
e
t
:
A
h
y
b
r
i
d
t
r
a
n
sf
o
r
mer
a
r
c
h
i
t
e
c
t
u
r
e
f
o
r
m
e
d
i
c
a
l
i
m
a
g
e
se
g
me
n
t
a
t
i
o
n
,
”
i
n
Me
d
i
c
a
l
I
m
a
g
e
C
o
m
p
u
t
i
n
g
a
n
d
C
o
m
p
u
t
e
r
Ass
i
st
e
d
I
n
t
e
r
v
e
n
t
i
o
n
–
MIC
C
AI
2
0
2
1
,
S
p
r
i
n
g
e
r
,
C
h
a
m
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
030
-
8
7
1
9
9
-
4
_
6
.
[
1
0
]
D
h
a
b
h
a
i
a
n
d
Y
.
G
u
p
t
a
,
“
E
mp
i
r
i
c
a
l
st
u
d
y
o
f
i
ma
g
e
c
l
a
ssi
f
i
c
a
t
i
o
n
t
e
c
h
n
i
q
u
e
s
t
o
c
l
a
s
si
f
y
t
h
e
i
ma
g
e
u
s
i
n
g
S
V
M
:
A
r
e
v
i
e
w
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
I
n
n
o
v
a
t
i
v
e
Re
se
a
rc
h
i
n
C
o
m
p
u
t
e
r
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
En
g
i
n
e
e
r
i
n
g
,
v
o
l
.
4
,
n
o
.
1
0
,
p
p
.
1
7
1
2
8
-
1
7
1
3
3
,
2
0
1
6
,
d
o
i
:
1
0
.
1
5
6
8
0
/
I
JI
R
C
C
E.
2
0
1
6
.
0
4
1
0
0
0
4
.
[
1
1
]
T.
M
a
r
u
y
a
m
a
e
t
a
l
.
,
“
C
o
m
p
a
r
i
s
o
n
o
f
med
i
c
a
l
i
m
a
g
e
c
l
a
ssi
f
i
c
a
t
i
o
n
a
c
c
u
r
a
c
y
a
m
o
n
g
t
h
r
e
e
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
m
e
t
h
o
d
s,
”
J
o
u
r
n
a
l
o
f
X
-
ra
y
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
2
6
,
n
o
.
6
,
p
p
.
8
8
5
-
8
9
3
,
2
0
1
8
,
d
o
i
:
1
0
.
3
2
3
3
/
X
S
T
-
1
8
3
8
6
.
[
1
2
]
L.
C
a
i
,
J.
G
a
o
,
a
n
d
D
.
Zh
a
o
,
“
A
r
e
v
i
e
w
o
f
t
h
e
a
p
p
l
i
c
a
t
i
o
n
o
f
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
ss
i
f
i
c
a
t
i
o
n
a
n
d
se
g
me
n
t
a
t
i
o
n
,
”
An
n
a
l
s
o
f
T
ra
n
sl
a
t
i
o
n
M
e
d
i
c
i
n
e
,
v
o
l
.
8
,
n
o
.
1
1
,
J
u
n
.
2
0
2
0
,
d
o
i
:
1
0
.
2
1
0
3
7
/
a
t
m.2
0
2
0
.
0
2
.
4
4
.
[
1
3
]
S
.
Zh
o
u
,
J.
W
a
n
g
,
S
.
Z
h
a
n
g
,
Y
.
L
i
a
n
g
,
a
n
d
Y
.
G
o
n
g
,
“
A
c
t
i
v
e
c
o
n
t
o
u
r
m
o
d
e
l
b
a
se
d
o
n
l
o
c
a
l
a
n
d
g
l
o
b
a
l
i
n
t
e
n
s
i
t
y
i
n
f
o
r
m
a
t
i
o
n
f
o
r
med
i
c
a
l
i
ma
g
e
se
g
me
n
t
a
t
i
o
n
,
”
N
e
u
r
o
c
o
m
p
u
t
i
n
g
,
v
o
l
.
1
8
6
,
p
p
.
1
0
7
-
1
1
8
,
2
0
1
6
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
n
e
u
c
o
m
.
2
0
1
5
.
1
2
.
0
7
3
.
[
1
4
]
S
.
R
.
S
h
a
h
,
S
.
Q
a
d
r
i
,
H
.
B
i
b
i
,
S
.
M
.
W
.
S
h
a
h
,
M
.
I
.
S
h
a
r
i
f
,
a
n
d
F
.
M
a
r
i
n
e
l
l
o
,
“
C
o
m
p
a
r
i
n
g
i
n
c
e
p
t
i
o
n
V
3
,
V
G
G
1
6
,
V
G
G
1
9
,
C
N
N
,
a
n
d
R
e
sN
e
t
5
0
:
A
c
a
se
s
t
u
d
y
o
n
e
a
r
l
y
d
e
t
e
c
t
i
o
n
o
f
a
r
i
c
e
d
i
sea
se
,
”
A
g
r
o
n
o
m
y
,
v
o
l
.
1
3
,
n
o
.
6
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
a
g
r
o
n
o
my
1
3
0
6
1
6
3
3
.
[
1
5
]
V
.
I
k
e
c
h
u
k
w
u
,
S
.
M
u
r
a
l
i
,
R
.
D
e
e
p
u
,
a
n
d
R
.
C
.
S
h
i
v
a
m
u
r
t
h
y
,
“
R
e
sN
e
t
-
5
0
v
s
V
G
G
-
1
9
v
s
t
r
a
i
n
i
n
g
f
r
o
m
s
c
r
a
t
c
h
:
A
c
o
m
p
a
r
a
t
i
v
e
a
n
a
l
y
si
s
o
f
t
h
e
s
e
g
m
e
n
t
a
t
i
o
n
a
n
d
c
l
a
s
si
f
i
c
a
t
i
o
n
o
f
P
n
e
u
m
o
n
i
a
f
r
o
m
c
h
e
st
X
-
r
a
y
i
ma
g
e
s
,
”
G
l
o
b
a
l
T
r
a
n
si
t
i
o
n
s
Pr
o
c
e
e
d
i
n
g
s
,
v
o
l
.
2
,
n
o
.
2
,
p
p
.
3
7
5
-
3
8
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
g
l
t
p
.
2
0
2
1
.
0
8
.
0
2
7
.
[
1
6
]
M
.
O
.
K
h
a
i
r
a
n
d
i
s
h
,
M
.
S
h
a
r
m
a
,
V
.
Jai
n
,
J.
M
.
C
h
a
t
t
e
r
j
e
e
,
a
n
d
N
.
Z.
J
h
a
n
j
h
i
,
“
A
H
y
b
r
i
d
C
N
N
-
S
V
M
t
h
r
e
s
h
o
l
d
se
g
m
e
n
t
a
t
i
o
n
a
p
p
r
o
a
c
h
f
o
r
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
o
f
M
R
I
b
r
a
i
n
i
m
a
g
e
s
,
”
I
RBM
,
v
o
l
.
4
3
,
n
o
.
4
,
p
p
.
2
9
0
-
2
9
9
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
r
b
m.
2
0
2
1
.
0
6
.
0
0
3
.
[
1
7
]
Y
.
D
i
n
g
,
Q
.
Y
a
n
g
,
Y
.
W
a
n
g
,
D
.
C
h
e
n
,
Z.
Q
i
n
,
a
n
d
J
.
Z
h
a
n
g
,
“
M
a
l
l
e
sN
e
t
:
A
mu
l
t
i
-
o
b
j
e
c
t
a
ssi
s
t
a
n
c
e
b
a
s
e
d
n
e
t
w
o
r
k
f
o
r
b
r
a
c
h
i
a
l
p
l
e
x
u
s s
e
g
me
n
t
a
t
i
o
n
i
n
u
l
t
r
a
s
o
u
n
d
i
m
a
g
e
s,
”
Me
d
i
c
a
l
I
m
a
g
e
A
n
a
l
y
si
s
,
v
o
l
.
8
0
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
me
d
i
a
.
2
0
2
2
.
1
0
2
5
1
1
.
[
1
8
]
J.
S
.
B
o
w
n
e
s
s
e
t
a
l
.
,
“
V
a
r
i
a
b
i
l
i
t
y
b
e
t
w
e
e
n
h
u
m
a
n
e
x
p
e
r
t
s
a
n
d
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
i
n
i
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
a
n
a
t
o
mi
c
a
l
st
r
u
c
t
u
r
e
s
b
y
u
l
t
r
a
s
o
u
n
d
i
n
r
e
g
i
o
n
a
l
a
n
a
e
s
t
h
e
si
a
:
a
f
r
a
mew
o
r
k
f
o
r
e
v
a
l
u
a
t
i
o
n
o
f
a
ss
i
st
i
v
e
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
,
”
Br
i
t
i
s
h
J
o
u
r
n
a
l
An
a
e
st
h
e
s
i
a
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
b
j
a
.
2
0
2
3
.
0
9
.
0
2
3
.
[
1
9
]
Y
.
Jo
e
t
a
l
.
,
“
O
p
t
i
ma
l
v
i
e
w
d
e
t
e
c
t
i
o
n
f
o
r
u
l
t
r
a
s
o
u
n
d
-
g
u
i
d
e
d
s
u
p
r
a
c
l
a
v
i
c
u
l
a
r
b
l
o
c
k
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s,
”
S
c
i
e
n
t
i
f
i
c
Re
p
o
r
t
s
,
v
o
l
.
1
3
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
8
-
0
2
3
-
4
4
1
7
0
-
y.
[
2
0
]
D
.
T
i
a
n
e
t
a
l
.
,
“
B
r
a
c
h
i
a
l
p
l
e
x
u
s
n
e
r
v
e
t
r
u
n
k
r
e
c
o
g
n
i
t
i
o
n
f
r
o
m
u
l
t
r
a
s
o
u
n
d
i
m
a
g
e
s:
a
c
o
m
p
a
r
a
t
i
v
e
st
u
d
y
o
f
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
s
,
”
i
n
I
EEE
A
c
c
e
ss
,
v
o
l
.
1
0
,
p
p
.
8
2
0
0
3
-
8
2
0
1
4
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
2
.
3
1
9
6
3
5
6
.
[
2
1
]
L.
C
a
i
,
Q
.
Li
,
J
.
Zh
a
n
g
,
Z
.
Zh
a
n
g
,
R
.
Y
a
n
g
,
a
n
d
L
.
Zh
a
n
g
,
“
U
l
t
r
a
s
o
u
n
d
i
m
a
g
e
se
g
me
n
t
a
t
i
o
n
b
a
se
d
o
n
t
r
a
n
sf
o
r
mer
a
n
d
U
-
N
e
t
w
i
t
h
j
o
i
n
t
l
o
ss
,
”
Pe
e
rJ
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
9
,
p
p
.
e
1
6
3
8
–
e
1
6
3
8
,
O
c
t
.
2
0
2
3
,
d
o
i
:
1
0
.
7
7
1
7
/
p
e
e
r
j
-
c
s.
1
6
3
8
.
[
2
2
]
G
.
R
a
n
g
a
n
a
t
h
a
n
,
“
A
st
u
d
y
t
o
f
i
n
d
f
a
c
t
s
b
e
h
i
n
d
p
r
e
p
r
o
c
e
ss
i
n
g
o
n
d
e
e
p
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms
,
”
J
o
u
rn
a
l
o
f
I
n
n
o
v
a
t
i
v
e
I
m
a
g
e
Pro
c
e
ssi
n
g
,
v
o
l
.
3
,
n
o
.
1
,
p
p
.
6
6
-
7
4
,
2
0
2
1
,
d
o
i
:
1
0
.
3
6
5
4
8
/
j
i
i
p
.
2
0
2
1
.
1
.
0
0
6
.
[
2
3
]
D
.
D
a
n
o
n
,
M
.
A
r
a
r
,
D
.
C
o
h
e
n
-
O
r
,
a
n
d
A
.
S
h
a
m
i
r
,
“
I
mag
e
r
e
s
i
z
i
n
g
b
y
r
e
c
o
n
st
r
u
c
t
i
o
n
f
r
o
m
d
e
e
p
f
e
a
t
u
r
e
s,
”
C
o
m
p
u
t
a
t
i
o
n
a
l
Vi
s
u
a
l
Me
d
i
a
,
v
o
l
.
7
,
p
p
.
4
5
3
–
4
6
6
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s4
1
0
9
5
-
021
-
0
2
1
6
-
x.
[
2
4
]
D
.
S
i
n
g
h
a
n
d
B
.
S
i
n
g
h
,
“
I
n
v
e
s
t
i
g
a
t
i
n
g
t
h
e
i
m
p
a
c
t
o
f
d
a
t
a
n
o
r
ma
l
i
z
a
t
i
o
n
o
n
c
l
a
ss
i
f
i
c
a
t
i
o
n
p
e
r
f
o
r
ma
n
c
e
,
”
A
p
p
l
i
e
d
S
o
f
t
C
o
m
p
u
t
i
n
g
,
v
o
l
.
9
7
,
p
p
.
1
0
5
5
2
4
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
s
o
c
.
2
0
1
9
.
1
0
5
5
2
4
.
[
2
5
]
F
.
D
i
n
g
,
Y
.
S
h
i
,
G
.
Zh
u
,
a
n
d
Y
.
-
Q
.
S
h
i
.,
“
R
e
a
l
-
t
i
me
e
st
i
ma
t
i
o
n
f
o
r
t
h
e
p
a
r
a
met
e
r
s
o
f
G
a
u
ss
i
a
n
f
i
l
t
e
r
i
n
g
v
i
a
d
e
e
p
l
e
a
r
n
i
n
g
,
”
J
o
u
r
n
a
l
o
f
Re
a
l
-
T
i
m
e
I
m
a
g
e
Pr
o
c
e
ssi
n
g
,
v
o
l
.
1
7
,
p
p
.
1
7
–
2
7
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
0
7
/
s1
1
5
5
4
-
019
-
0
0
9
0
7
-
5.
[
2
6
]
S
.
S
o
m
a
su
n
d
e
r
a
n
d
F
.
Y
.
S
h
i
h
,
“
La
n
d
c
o
v
e
r
i
ma
g
e
se
g
me
n
t
a
t
i
o
n
b
a
s
e
d
o
n
i
n
d
i
v
i
d
u
a
l
c
l
a
ss
b
i
n
a
r
y
m
a
s
k
s
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Pa
t
t
e
r
n
Re
c
o
g
n
i
t
i
o
n
a
n
d
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
3
5
,
n
o
.
1
6
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
4
2
/
s
0
2
1
8
0
0
1
4
2
1
5
4
0
3
4
3
.
[
2
7
]
B
.
K
o
o
n
c
e
,
C
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
rks
w
i
t
h
s
w
i
f
t
f
o
r
t
e
n
so
r
f
l
o
w
:
i
m
a
g
e
r
e
c
o
g
n
i
t
i
o
n
a
n
d
d
a
t
a
se
t
c
a
t
e
g
o
ri
za
t
i
o
n
,
B
e
r
k
e
l
e
y
,
C
A
:
A
p
r
e
ss,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
1
-
4
8
4
2
-
6
1
6
8
-
2.
[
2
8
]
M
.
A
.
C
h
a
n
d
r
a
a
n
d
S
.
S
.
B
e
d
i
,
“
S
u
r
v
e
y
o
n
s
u
p
p
o
r
t
v
e
c
t
o
r
mac
h
i
n
e
s
a
n
d
t
h
e
i
r
a
p
p
l
i
c
a
t
i
o
n
i
n
i
m
a
g
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
3
,
p
p
.
1
–
1
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
4
1
8
7
0
-
0
1
7
-
0
0
8
0
-
1.
[
2
9
]
A
.
M
o
h
a
mm
e
d
a
n
d
R
.
K
o
r
a
,
“
A
c
o
mp
r
e
h
e
n
si
v
e
r
e
v
i
e
w
o
n
e
n
s
e
mb
l
e
d
e
e
p
l
e
a
r
n
i
n
g
:
O
p
p
o
r
t
u
n
i
t
i
e
s
a
n
d
c
h
a
l
l
e
n
g
e
s,
”
J
o
u
rn
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
5
,
n
o
.
2
,
p
p
.
7
5
7
-
7
7
4
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
k
su
c
i
.
2
0
2
3
.
0
1
.
0
1
4
.
B
I
O
G
RAP
H
I
E
S
OF
AUTH
O
RS
Dr
.
S
o
b
h
a
n
a
Mu
m
m
a
n
e
n
i
is
c
u
rre
n
tl
y
w
o
rk
i
n
g
as
an
a
ss
o
c
iate
p
ro
fe
ss
o
r
in
th
e
De
p
a
rtme
n
t
of
C
o
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
V.
R.
S
id
d
h
a
rth
a
En
g
in
e
e
rin
g
C
o
ll
e
g
e
,
Vijay
a
wa
d
a
,
In
d
ia.
S
h
e
re
c
e
iv
e
d
P
h
.
D.
d
e
g
re
e
in
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
in
2
0
1
8
fro
m
Krish
n
a
Un
iv
e
rsity
.
S
h
e
h
a
s
16
y
e
a
rs
of
tea
c
h
in
g
e
x
p
e
rien
c
e
.
He
r
re
se
a
rc
h
in
tere
sts
lie
in
a
re
a
s
su
c
h
as
a
rti
ficia
l
in
telli
g
e
n
c
e
,
m
a
c
h
in
e
lea
rn
in
g
,
d
a
ta
a
n
a
ly
ti
c
s,
c
y
b
e
r
se
c
u
rit
y
,
a
n
d
so
ftwa
re
e
n
g
i
n
e
e
rin
g.
S
h
e
p
u
b
li
sh
e
d
35
p
a
p
e
rs
in
n
a
ti
o
n
a
l
a
n
d
in
ter
n
a
ti
o
n
a
l
j
o
u
r
n
a
ls
a
n
d
p
u
b
li
sh
e
d
7
p
a
ten
ts.
S
h
e
can
be
c
o
n
tac
ted
at
e
m
a
il
:
so
b
h
a
n
a
@v
rsid
d
h
a
rth
a
.
a
c
.
i
n
.
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
E
n
h
a
n
ci
n
g
u
ltr
a
s
o
u
n
d
-
g
u
id
ed
b
r
a
ch
ia
l p
lexu
s
n
erve
lo
c
a
liz
a
tio
n
…
(
S
o
b
h
a
n
a
Mu
mma
n
en
i
)
4947
K
u
sha
l
K
u
m
a
r
Chi
n
t
a
k
a
y
a
la
is
a
fi
n
a
l
-
y
e
a
r
B.
Tec
h
.
stu
d
e
n
t,
sp
e
c
ializin
g
in
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
r
in
g
at
V.
R.
S
id
d
h
a
rth
a
E
n
g
in
e
e
rin
g
Co
l
leg
e
,
Vijay
a
wa
d
a
,
In
d
ia.
He
is
p
a
ss
io
n
a
te
a
b
o
u
t
a
rti
ficia
l
in
telli
g
e
n
c
e
a
n
d
m
a
c
h
in
e
lea
rn
in
g
.
He
is
IEE
E
c
o
re
m
e
m
b
e
r
of
Co
m
p
u
ter
S
o
c
iet
y
(CS
)
stu
d
e
n
t
c
h
a
p
ter.
He
c
a
n
be
c
o
n
tac
ted
at
e
m
a
il
:
k
u
sh
a
l
k
u
m
a
rc
h
in
tak
a
y
a
la1
2
@g
m
a
il
.
c
o
m
.
La
li
th
S
a
i
M
u
k
u
n
d
Ya
r
l
a
g
a
d
d
a
is
a
fin
a
l
-
y
e
a
r
B.
Tec
h
.
stu
d
e
n
t,
sp
e
c
ializin
g
in
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
r
in
g
at
V.
R.
S
id
d
h
a
rth
a
E
n
g
in
e
e
rin
g
Co
l
leg
e
,
Vijay
a
wa
d
a
,
In
d
ia.
He
is
p
a
ss
io
n
a
te
a
b
o
u
t
a
rti
ficia
l
in
telli
g
e
n
c
e
a
n
d
m
a
c
h
in
e
le
a
rn
in
g
.
He
can
be
c
o
n
tac
ted
at
e
m
a
il
:
m
u
k
u
n
d
y
g
2
0
0
2
@g
m
a
il
.
c
o
m
.
Ve
n
k
a
ta
S
iv
a
N
a
g
a
Ra
j
u
Al
a
is
a
fin
a
l
-
y
e
a
r
B.
Tec
h
.
stu
d
e
n
t,
sp
e
c
ializin
g
in
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
r
in
g
at
V.
R.
S
id
d
h
a
rth
a
E
n
g
in
e
e
rin
g
Co
l
leg
e
,
Vijay
a
wa
d
a
,
In
d
ia.
He
is
p
a
ss
io
n
a
te
a
b
o
u
t
a
rti
ficia
l
in
telli
g
e
n
c
e
a
n
d
m
a
c
h
in
e
le
a
rn
in
g
.
He
can
be
c
o
n
tac
ted
at
e
m
a
il
:
a
v
sn
a
g
a
ra
ju
y
a
d
a
v
1
9
9
9
@g
m
a
il
.
c
o
m
.
Nih
ith
a
Ve
m
u
la
p
a
l
li
is
a
fi
n
a
l
-
y
e
a
r
B.
Tec
h
.
st
u
d
e
n
t
sp
e
c
iali
z
in
g
in
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
at
V.
R.
S
i
d
d
h
a
rth
a
E
n
g
i
n
e
e
rin
g
Co
ll
e
g
e
,
Vijay
a
wa
d
a
,
I
n
d
ia.
S
h
e
is
p
a
ss
io
n
a
te
a
b
o
u
t
d
e
e
p
lea
rn
in
g
a
n
d
h
a
s
b
e
e
n
re
c
o
g
n
ize
d
as
a
G
o
o
g
le
Wo
m
e
n
E
n
g
i
n
e
e
rs
S
c
h
o
lar.
A
d
d
it
io
n
a
ll
y
,
sh
e
h
o
ld
s
th
e
p
o
siti
o
n
of
IEE
E
Ch
a
ir
at
t
h
e
Wo
m
e
n
i
n
E
n
g
i
n
e
e
rin
g
(
WI
E)
stu
d
e
n
t
c
h
a
p
ter.
S
h
e
can
be
c
o
n
tac
ted
at
e
m
a
il
:
n
i
h
it
h
a
v
e
m
u
lap
a
ll
i@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.