I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Ro
bo
t
ics a
nd
Aut
o
m
a
t
io
n
(
I
J
RA)
Vo
l.
1
4
,
No
.
2
,
J
u
n
e
20
2
5
,
p
p
.
2
73
~
2
80
I
SS
N:
2722
-
2
5
8
6
,
DOI
:
1
0
.
1
1
5
9
1
/i
jr
a
.
v
14
i
2
.
pp
2
73
-
2
80
273
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
r
a
.
ia
esco
r
e.
co
m
Experime
ntal ev
a
lua
tion o
f
bidi
rect
io
na
l encoder
represen
tatio
ns fr
o
m trans
formers
mo
dels for de
-
ide
ntif
ica
tion
o
f
clinica
l docum
ent
ima
g
es
Ra
v
icha
nd
ra
Srira
m
,
Siv
a
S
a
t
hy
a
Su
nd
a
ra
m
,
S.
L
o
urdu
M
a
rie
So
ph
ie
D
e
p
a
r
t
me
n
t
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
,
P
o
n
d
i
c
h
e
r
r
y
U
n
i
v
e
r
si
t
y
,
P
o
n
d
i
c
h
e
r
r
y
,
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
Ap
r
2
6
,
2
0
2
4
R
ev
is
ed
J
an
2
8
,
2
0
2
5
Acc
ep
ted
Ma
r
5
,
2
0
2
5
M
a
n
y
h
e
a
lt
h
in
sti
tu
tes
m
a
in
tai
n
p
a
ti
e
n
ts
’
d
iag
n
o
sis
a
n
d
trea
tme
n
t
re
p
o
rts
a
s
sc
a
n
n
e
d
ima
g
e
s.
F
o
r
h
e
a
lt
h
c
a
re
a
n
a
ly
ti
c
s
a
n
d
re
se
a
rc
h
,
larg
e
v
o
lu
m
e
s
o
f
d
ig
it
a
l
ly
sto
re
d
p
a
ti
e
n
t
in
f
o
rm
a
ti
o
n
h
a
v
e
to
b
e
a
c
c
e
ss
e
d
,
b
u
t
t
h
e
p
riv
a
c
y
re
q
u
irem
e
n
ts
o
f
p
r
o
tec
ted
h
e
a
l
th
in
f
o
rm
a
ti
o
n
(P
HI)
li
m
it
t
h
e
re
se
a
rc
h
o
p
p
o
rt
u
n
it
ies
.
P
a
rti
c
u
larly
in
t
h
is
a
rti
ficia
l
in
telli
g
e
n
c
e
(AI)
e
ra
,
d
e
e
p
lea
rn
in
g
m
o
d
e
ls
re
q
u
ire
lar
g
e
d
a
t
a
se
ts
fo
r
train
in
g
p
u
r
p
o
se
s,
wh
ic
h
h
o
sp
i
tals
c
a
n
n
o
t
sh
a
re
u
n
les
s
th
e
P
HI
field
s
a
re
d
e
-
i
d
e
n
ti
f
ied
.
M
a
n
u
a
l
d
e
-
id
e
n
ti
fica
ti
o
n
is
b
e
y
o
n
d
p
o
ss
ib
le,
wit
h
m
il
li
o
n
s
o
f
p
a
ti
e
n
t
re
c
o
rd
s
g
e
n
e
ra
ted
i
n
h
o
s
p
it
a
ls
e
v
e
ry
d
a
y
.
He
n
c
e
,
th
is
wo
rk
a
ims
to
a
u
to
m
a
te
th
e
d
e
-
i
d
e
n
ti
fi
c
a
ti
o
n
o
f
c
li
n
ica
l
d
o
c
u
m
e
n
t
ima
g
e
s
u
t
il
i
z
in
g
AI
m
o
d
e
ls,
p
a
rti
c
u
larly
p
re
-
train
e
d
b
id
irec
ti
o
n
a
l
e
n
c
o
d
e
r
re
p
re
se
n
tati
o
n
s
fro
m
tran
sf
o
rm
e
rs
(BERT
)
m
o
d
e
ls.
F
o
r
th
e
p
u
rp
o
se
o
f
e
x
p
e
rime
n
tatio
n
,
a
sy
n
t
h
e
ti
c
d
a
tas
e
t
o
f
5
5
0
c
li
n
ica
l
d
o
c
u
m
e
n
t
ima
g
e
s
wa
s
g
e
n
e
ra
ted
,
e
n
c
o
m
p
a
ss
in
g
d
a
ta
o
b
tain
e
d
fr
o
m
d
iv
e
rs
e
p
a
ti
e
n
ts
a
c
ro
ss
m
u
lt
ip
le
h
o
sp
it
a
ls.
T
h
is
wo
rk
p
re
se
n
ts
a
two
-
sta
g
e
tran
sf
e
r
lea
rn
in
g
a
p
p
ro
a
c
h
,
in
i
ti
a
ll
y
e
m
p
l
o
y
in
g
Tes
se
ra
c
t
c
h
a
ra
c
ter
re
c
o
g
n
it
io
n
(OCR)
to
c
o
n
v
e
rt
c
li
n
ica
l
d
o
c
u
m
e
n
t
ima
g
e
s
in
to
tex
t.
S
u
b
se
q
u
e
n
t
ly
,
it
e
x
tr
a
c
ts
P
HI
field
s
fr
o
m
th
e
tex
t
f
o
r
d
e
-
id
e
n
ti
fica
ti
o
n
.
F
o
r
th
e
p
u
rp
o
se
o
f
e
x
trac
ti
o
n
,
BERT
m
o
d
e
ls
we
re
u
ti
l
ize
d
;
i
n
th
is
wo
r
k
,
we
c
o
n
tras
ted
six
p
re
-
train
e
d
v
e
rsio
n
s
o
f
su
c
h
m
o
d
e
ls
t
o
e
x
a
m
in
e
th
e
ir
e
ffe
c
ti
v
e
n
e
ss
a
n
d
a
c
h
i
e
v
e
th
e
F
1
sc
o
re
o
f
9
2
.
4
5
%
,
t
h
u
s
s
h
o
wi
n
g
b
e
tt
e
r
p
o
ten
ti
a
l
f
o
r
d
e
-
id
e
n
ti
f
y
i
n
g
P
HI
d
a
ta
in
c
li
n
ica
l
d
o
c
u
m
e
n
ts.
K
ey
w
o
r
d
s
:
C
lin
ical
d
e
-
id
en
tific
atio
n
Dee
p
lear
n
in
g
Natu
r
al
lan
g
u
ag
e
p
r
o
ce
s
s
in
g
Pro
tecte
d
h
ea
lth
in
f
o
r
m
atio
n
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es
T
ess
er
ac
t c
h
ar
ac
ter
r
ec
o
g
n
itio
n
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
:
R
av
ich
an
d
r
a
Srir
am
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
,
Po
n
d
ic
h
er
r
y
Un
iv
er
s
ity
Po
n
d
ich
er
r
y
,
I
n
d
ia
E
m
ail: r
av
ich
an
d
r
asrir
am
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Patien
ts
’
clin
ical
f
iles
co
n
ta
in
r
ec
o
r
d
s
o
f
m
e
d
icatio
n
s
an
d
th
e
r
ap
ies
th
at
h
av
e
alr
ea
d
y
b
ee
n
ad
m
in
is
ter
ed
to
th
em
.
T
h
ese
r
ec
o
r
d
s
p
lay
an
im
p
o
r
ta
n
t
r
o
le
in
im
p
r
o
v
in
g
th
e
ac
cu
r
ac
y
o
f
p
atien
t
d
iag
n
o
s
es
an
d
o
v
e
r
all
ca
r
e.
Fu
r
th
e
r
m
o
r
e
,
th
ey
ar
e
v
alu
ab
le
r
eso
u
r
ce
s
in
m
ed
ical
r
esear
ch
,
p
r
o
v
id
in
g
im
p
o
r
tan
t
in
s
ig
h
t
s
in
to
th
e
p
atien
t
’
s
s
tate.
Prio
r
to
s
h
ar
in
g
th
ese
m
ed
ical
r
e
co
r
d
s
with
r
esear
c
h
er
s
f
o
r
cl
in
ical
s
tu
d
ies,
th
e
p
r
o
tecte
d
h
ea
lth
in
f
o
r
m
atio
n
(
PHI
)
m
u
s
t
b
e
r
em
o
v
e
d
o
r
an
o
n
y
m
ized
.
T
h
is
co
m
p
r
is
es
th
e
p
atien
t
’
s
n
am
e,
ag
e,
an
d
co
n
tact
in
f
o
r
m
atio
n
.
T
h
is
d
e
-
id
en
tific
atio
n
n
ee
d
s
tem
s
f
r
o
m
th
e
s
en
s
itiv
e
n
atu
r
e
o
f
p
atien
t
d
ata.
I
n
th
e
Un
ited
States
,
HI
PA
A
[
1
]
,
o
r
th
e
Hea
lth
I
n
s
u
r
an
ce
Po
r
ta
b
ilit
y
an
d
Acc
o
u
n
ta
b
ilit
y
Act,
s
p
ec
if
ies
eig
h
teen
d
if
f
er
en
t
ty
p
es
o
f
PHI
th
at
m
u
s
t
b
e
ca
r
ef
u
lly
d
e
-
id
e
n
t
if
ied
b
ef
o
r
e
b
ein
g
s
h
ar
ed
with
th
ir
d
p
ar
ties
.
Fu
r
th
er
m
o
r
e
,
s
ev
er
al
ar
tific
ial
in
tellig
en
ce
(
AI
)
m
o
d
els
u
s
ed
to
an
aly
ze
h
ea
lth
ca
r
e
r
ec
o
r
d
s
r
eq
u
ir
e
lar
g
e
am
o
u
n
ts
o
f
d
ata
f
o
r
tr
ain
in
g
.
Ho
s
p
itals
d
o
n
o
t
s
h
ar
e
clin
i
ca
l
d
ata
d
u
e
to
p
r
iv
ac
y
p
o
lic
ies,
an
d
all
p
atien
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
7
2
2
-
2
5
8
6
I
AE
S
I
n
t
J
R
o
b
&
A
u
to
m
,
Vo
l
.
14
,
No
.
2
,
J
u
n
e
20
2
5
:
2
73
-
2
80
274
r
ec
o
r
d
s
co
n
tain
PHI
elem
en
ts
.
T
h
ese
PHI
co
m
p
o
n
e
n
ts
m
u
s
t
b
e
an
o
n
y
m
ized
o
r
d
e
-
i
d
en
t
if
ied
b
ef
o
r
e
b
ein
g
p
r
o
v
id
e
d
f
o
r
r
esear
ch
p
u
r
p
o
s
es.
Ho
wev
er
,
o
n
ly
a
lim
ited
n
u
m
b
er
o
f
h
ea
lth
ca
r
e
p
r
o
f
ess
io
n
als
m
an
u
ally
d
e
-
id
en
tify
clin
ical
d
o
c
u
m
en
t
im
ag
es
u
s
in
g
s
p
ec
ialized
to
o
ls
f
o
r
ed
itin
g
.
T
h
is
p
r
o
ce
d
u
r
e
is
in
ef
f
ec
tiv
e
f
o
r
b
o
t
h
tim
e
an
d
ex
p
en
s
e.
C
o
n
s
eq
u
en
t
ly
,
m
an
y
d
ig
ital c
o
p
ies o
f
th
ese
h
ea
lth
ca
r
e
d
o
cu
m
e
n
ts
r
em
ain
u
n
u
s
ed
.
T
h
is
wo
r
k
p
r
esen
ts
a
m
eth
o
d
th
at
u
tili
ze
s
p
r
e
-
t
r
ain
ed
b
i
d
ir
ec
tio
n
al
e
n
co
d
e
r
r
e
p
r
esen
t
atio
n
s
f
r
o
m
tr
an
s
f
o
r
m
er
s
(
B
E
R
T
)
m
o
d
el
s
f
o
r
th
e
au
to
m
ated
d
etec
ti
o
n
a
n
d
elim
in
atio
n
o
f
p
e
r
s
o
n
ally
i
d
en
tify
in
g
in
f
o
r
m
atio
n
in
clin
ical
d
o
cu
m
en
t
im
ag
es.
Me
d
ical
d
o
cu
m
en
ts
ex
h
ib
it
u
n
ev
e
n
o
r
g
a
n
i
za
tio
n
al
f
r
am
ewo
r
k
s
,
in
teg
r
atin
g
v
ar
i
o
u
s
d
ata
f
o
r
m
s
in
clu
d
in
g
lar
g
e
h
ea
d
i
n
g
s
,
lo
g
o
s
,
s
in
g
le
-
co
lu
m
n
an
d
d
u
al
-
co
lu
m
n
k
e
y
-
v
al
u
e
p
air
s
,
as
well
a
s
m
u
lti
-
co
lu
m
n
tab
les
with
an
d
with
o
u
t
b
o
r
d
er
s
.
T
h
is
d
iv
er
s
ity
in
d
o
cu
m
en
t
lay
o
u
ts
p
r
esen
ts
ch
allen
g
es
in
ac
cu
r
ately
d
etec
tin
g
PHI
d
etails
with
in
th
e
d
o
cu
m
en
ts
.
T
h
e
lan
g
u
ag
e
r
ep
r
es
en
tatio
n
p
ar
a
d
ig
m
,
B
E
R
T
[
2
]
,
is
s
p
ec
if
ically
d
ev
elo
p
ed
to
p
r
e
-
tr
ain
d
ee
p
b
id
ir
ec
tio
n
al
r
ep
r
esen
tatio
n
s
f
r
o
m
u
n
lab
eled
tex
t.
T
h
e
p
r
e
-
tr
ain
in
g
p
r
o
ce
s
s
in
v
o
lv
es
s
im
u
ltan
eo
u
s
co
n
s
id
er
atio
n
o
f
b
o
th
t
h
e
lef
t
an
d
r
ig
h
t
co
n
tex
ts
th
r
o
u
g
h
o
u
t
all
lay
er
s
.
As
a
r
esu
lt,
th
e
B
E
R
T
m
o
d
el
ca
n
b
e
ef
f
ec
tiv
ely
f
in
e
-
tu
n
ed
b
y
in
co
r
p
o
r
atin
g
a
s
in
g
u
lar
o
u
t
p
u
t
lay
er
.
T
h
is
en
ab
les
th
e
d
ev
elo
p
m
en
t
o
f
s
tate
-
of
-
th
e
-
ar
t
m
o
d
els
f
o
r
v
ar
io
u
s
task
s
,
s
u
ch
as
q
u
esti
o
n
an
s
wer
in
g
an
d
lan
g
u
ag
e
in
f
er
en
ce
,
with
o
u
t
n
ec
ess
itatin
g
s
u
b
s
tan
tial
m
o
d
if
icatio
n
s
to
th
e
u
n
d
e
r
ly
in
g
a
r
ch
itectu
r
e
s
p
ec
if
ic
to
ea
ch
task
.
T
h
e
liter
atu
r
e
r
ev
iew
co
n
d
u
ct
ed
s
o
f
ar
d
id
n
o
t
r
ev
ea
l
an
y
p
r
ev
io
u
s
s
tu
d
ies
o
n
d
e
-
id
en
tify
i
n
g
clin
ical
d
o
cu
m
e
n
ts
s
to
r
ed
as
im
a
g
es.
T
h
e
af
o
r
em
en
tio
n
ed
m
o
d
els
h
av
e
b
ee
n
s
u
b
jecte
d
to
test
in
g
u
tili
zin
g
p
u
b
licly
av
ailab
le
clin
ical
r
ec
o
r
d
s
d
atasets
,
in
clu
d
in
g
MI
M
I
C
–
I
I
[
3
]
,
MI
M
I
C
–
I
I
I
[
4
]
,
i
2
b
2
C
o
r
p
u
s
2
0
0
6
[
5
]
,
i2
b
2
C
o
r
p
u
s
2
0
1
4
[
6
]
,
a
n
d
C
E
GS
N
-
GR
I
D
C
o
r
p
u
s
2
0
1
6
[
7
]
.
T
h
e
d
atasets
in
q
u
esti
o
n
co
n
s
is
t
o
f
v
ar
io
u
s
f
r
ee
-
tex
t
an
d
elec
tr
o
n
ic
h
ea
lth
r
ec
o
r
d
s
(
E
HR
)
in
f
o
r
m
atio
n
,
u
n
lik
e
th
e
d
o
cu
m
en
ts
we
h
av
e
c
o
n
s
id
er
ed
in
t
h
is
r
esear
ch
,
wh
ich
ar
e
p
u
r
e
im
a
g
es.
T
h
is
r
esear
ch
f
o
c
u
s
es
o
n
th
e
d
e
-
id
en
tific
atio
n
a
n
d
a
n
o
n
y
m
iz
atio
n
o
f
PHI
f
ield
s
p
r
esen
t
in
p
atien
ts
’
clin
ical
d
o
cu
m
en
ts
.
Giv
en
th
e
ex
is
tin
g
g
ap
s
in
th
e
liter
atu
r
e,
th
e
r
esear
ch
ad
d
r
ess
es
th
e
f
o
llo
win
g
k
e
y
s
cien
tific
ch
all
en
g
es
to
tack
le
th
is
p
r
o
b
lem
ef
f
ec
tiv
ely
:
i)
ex
tr
ac
tio
n
o
f
co
n
ten
t
f
r
o
m
s
ca
n
n
ed
im
ag
es,
ii)
id
en
tify
in
g
th
e
PHI
f
ield
s
an
d
th
eir
lo
ca
tio
n
with
in
th
e
s
ca
n
n
ed
clin
ical
d
o
cu
m
en
ts
/im
ag
e,
an
d
iii)
de
-
id
en
tific
atio
n
/an
o
n
y
m
izin
g
th
e
PHI
f
ield
s
.
Sin
ce
1
9
9
6
,
n
u
m
er
o
u
s
r
esear
ch
er
s
h
av
e
c
o
m
m
itted
to
i
n
v
esti
g
atin
g
th
e
p
o
ten
tial
f
o
r
au
to
m
atin
g
th
e
p
r
o
ce
s
s
o
f
d
e
-
id
en
tif
y
in
g
h
e
alth
ca
r
e
r
ec
o
r
d
s
.
Sev
er
al
co
m
p
u
tatio
n
al
m
o
d
els
h
av
e
b
e
en
d
ev
is
ed
,
s
u
ch
as
co
n
d
itio
n
al
r
an
d
o
m
f
ield
s
(
C
R
Fs
)
,
h
id
d
en
Ma
r
k
o
v
m
o
d
els
(
HM
M)
,
d
ec
is
io
n
tr
ee
s
(
DT
s
)
,
an
d
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es
(
SVMs).
T
h
ese
s
y
s
tem
s
u
tili
ze
m
ac
h
in
e
-
lear
n
in
g
m
eth
o
d
o
lo
g
ies
t
h
at
ar
e
r
u
le
-
b
ased
,
k
n
o
wled
g
e
-
b
ased
,
o
r
d
ata
-
d
r
iv
en
.
Dee
p
l
ea
r
n
in
g
m
o
d
els
s
u
ch
as
lo
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
(
L
STM
)
,
g
ated
r
ec
u
r
r
en
t
u
n
it
(
GR
U)
,
an
d
b
id
ir
ec
tio
n
al
en
co
d
er
r
ep
r
esen
tatio
n
s
f
r
o
m
t
r
an
s
f
o
r
m
er
s
(
B
E
R
T
)
h
av
e
d
em
o
n
s
tr
ated
m
o
r
e
f
av
o
r
a
b
le
o
u
tco
m
es.
T
h
is
s
ec
tio
n
d
elv
es
in
to
v
ar
io
u
s
d
e
-
id
en
tific
atio
n
m
o
d
els
em
p
lo
y
ed
o
n
p
u
b
licl
y
ac
ce
s
s
ib
le
d
atasets
.
R
ef
er
en
ce
[
8
]
th
o
r
o
u
g
h
ly
ex
am
in
es
th
e
u
s
e
o
f
d
ee
p
lear
n
in
g
m
o
d
els
in
clin
ical
tex
t
d
e
-
id
en
tific
atio
n
,
o
f
f
er
i
n
g
a
co
m
p
r
eh
en
s
iv
e
ass
ess
m
en
t
o
f
th
e
to
p
ic.
T
h
is
s
u
r
v
ey
p
r
esen
ts
a
s
elec
tio
n
o
f
liter
ar
y
wo
r
k
s
f
o
r
e
x
am
in
atio
n
.
T
h
e
a
u
th
o
r
s
in
[
9
]
u
tili
ze
d
a
b
id
ir
e
ctio
n
al
L
STM
m
o
d
el
th
at
s
tr
o
n
g
ly
r
esem
b
les
th
e
m
o
d
el
estab
lis
h
ed
b
y
[
1
0
]
t
o
ac
h
iev
e
d
e
-
i
d
en
tific
atio
n
.
T
h
e
b
id
ir
ec
tio
n
al
L
STM
m
o
d
el
is
h
ig
h
ly
p
r
e
v
alen
t
in
n
atu
r
al
lan
g
u
ag
e
p
r
o
ce
s
s
in
g
(
NL
P)
an
d
d
em
o
n
s
tr
ates
ex
ce
p
tio
n
al
p
er
f
o
r
m
a
n
ce
in
a
r
an
g
e
o
f
n
am
ed
e
n
tity
r
ec
o
g
n
itio
n
(
NE
R
)
task
s
.
Pre
v
io
u
s
r
esear
ch
in
v
esti
g
atio
n
s
[
1
1
]
,
[
1
2
]
h
av
e
also
u
tili
ze
d
B
i
-
L
STM
f
o
r
th
e
p
u
r
p
o
s
e
o
f
clin
ical
d
e
-
id
en
tifi
ca
tio
n
.
T
h
is
ap
p
r
o
ac
h
allo
ws f
o
r
esti
m
atin
g
p
r
o
b
a
b
ilit
ies ass
o
ciate
d
with
s
ev
er
al
p
o
ten
tial la
b
els f
o
r
p
er
s
o
n
ally
id
en
tifia
b
le
h
ea
lth
in
f
o
r
m
atio
n
(
PHI
)
f
o
r
ev
er
y
to
k
en
in
th
e
s
eq
u
en
ce
.
Fu
r
th
er
m
o
r
e
,
th
e
au
th
o
r
s
in
[
2
]
p
r
o
p
o
s
ed
a
p
r
e
-
tr
ain
ed
d
ee
p
b
id
ir
ec
tio
n
al
tr
a
n
s
f
o
r
m
er
f
o
r
th
e
lan
g
u
ag
e
m
o
d
el,
r
e
f
er
r
ed
t
o
as
th
e
B
E
R
T
m
o
d
el.
T
h
is
m
o
d
el
ca
n
s
im
u
ltan
eo
u
s
ly
co
n
tex
tu
alize
wo
r
d
em
b
ed
d
in
g
s
b
y
in
co
r
p
o
r
atin
g
all
ad
jace
n
t
co
n
te
x
ts
,
th
er
eb
y
m
ak
in
g
m
ajo
r
c
o
n
tr
ib
u
tio
n
s
t
o
th
e
f
ield
o
f
NL
P
ac
r
o
s
s
m
u
ltip
le
task
s
.
I
n
an
o
th
er
wo
r
k
b
y
[
1
3
]
,
f
i
n
e
-
tu
n
i
n
g
o
f
a
b
id
ir
ec
tio
n
al
en
co
d
er
r
ep
r
e
s
en
tatio
n
m
o
d
el
to
attain
th
e
m
o
s
t
ad
v
an
ce
d
lev
el
o
f
d
e
-
id
en
tific
atio
n
p
er
f
o
r
m
an
ce
f
o
r
elec
tr
o
n
ic
h
ea
lth
d
ata
was
p
er
f
o
r
m
ed
.
T
h
is
m
o
d
el
d
em
o
n
s
tr
ates
a
h
ig
h
lev
el
o
f
ef
f
icac
y
i
n
r
em
o
v
i
n
g
p
atien
t
id
e
n
tifie
r
s
,
in
clu
d
in
g
b
u
t
n
o
t
lim
ited
to
n
am
e,
ag
e,
an
d
s
o
cial
s
ec
u
r
ity
n
u
m
b
er
.
T
h
o
u
g
h
v
ar
io
u
s
tech
n
iq
u
es
h
av
e
b
ee
n
ad
o
p
ted
,
n
o
s
in
g
le
m
o
d
el
co
u
ld
de
-
id
en
tify
all
th
e
PHI
f
ield
s
,
p
ar
ticu
lar
ly
f
r
o
m
C
lin
ical
d
o
cu
m
en
ts
th
at
ar
e
in
th
e
f
o
r
m
o
f
im
ag
es.
T
h
is
f
o
r
m
s
th
e
b
asis
o
f
th
is
r
esear
ch
,
w
h
ich
tr
ies
to
ex
p
er
im
en
tally
ev
alu
ate
s
o
m
e
o
f
th
e
p
o
p
u
lar
p
r
e
-
tr
ain
e
d
B
E
R
T
m
o
d
els
p
r
ev
io
u
s
ly
u
s
ed
f
o
r
d
ata
ex
tr
ac
tio
n
.
T
h
is
r
esear
ch
aim
s
to
u
s
e
B
E
R
T
f
o
r
th
e
p
u
r
p
o
s
e
o
f
PHI
id
en
tific
atio
n
an
d
an
o
n
y
m
iza
tio
n
,
wh
ich
is
n
o
t
f
o
u
n
d
in
t
h
e
ex
is
tin
g
liter
atu
r
e.
Sin
ce
s
ev
er
al
v
er
s
io
n
s
o
f
B
E
R
T
ar
e
av
ailab
le,
it
is
n
ec
e
s
s
ar
y
to
co
m
p
ar
e
t
h
em
with
p
r
o
p
er
ex
p
e
r
im
en
tatio
n
an
d
co
n
clu
d
e
th
e
ef
f
icac
y
o
f
th
e
b
est
o
n
e
s
o
th
at
f
u
tu
r
e
r
esear
ch
ca
n
r
ely
o
n
it
f
o
r
th
e
d
e
-
id
en
tific
atio
n
p
r
o
ce
s
s
.
T
h
e
p
er
f
o
r
m
an
ce
o
f
th
e
m
eth
o
d
s
is
ev
alu
ated
u
s
in
g
g
e
n
er
ally
estab
lis
h
ed
m
etr
ics s
u
ch
as p
o
s
itiv
e
p
r
ed
ictiv
ity
a
n
d
F1
S
co
r
e.
T
h
e
s
u
b
s
eq
u
en
t
s
ec
tio
n
s
o
f
th
e
p
ap
er
ar
e
s
tr
u
ctu
r
e
d
in
th
e
f
o
llo
win
g
m
an
n
e
r
:
s
ec
tio
n
2
.
i
n
tr
o
d
u
ce
s
th
e
p
r
o
p
o
s
ed
m
eth
o
d
,
s
ec
tio
n
3
o
u
tlin
es
th
e
ex
p
er
im
en
t
al
s
etu
p
an
d
d
is
p
lay
s
th
e
o
b
tain
ed
r
esu
lts
,
an
d
s
ec
tio
n
4
co
n
clu
d
es th
e
wo
r
k
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
AE
S
I
n
t
J
R
o
b
&
A
u
to
m
I
SS
N:
2722
-
2
5
8
6
E
xp
erimen
ta
l e
va
lu
a
tio
n
o
f b
i
d
ir
ec
tio
n
a
l e
n
co
d
er rep
r
esen
ta
tio
n
s
…
(
R
a
vich
a
n
d
r
a
S
r
ir
a
m
)
275
2.
M
E
T
H
O
D
T
h
e
p
r
o
p
o
s
ed
s
y
s
tem
lo
o
k
s
f
o
r
p
r
o
s
p
ec
tiv
e
tech
n
i
q
u
es
f
o
r
au
to
m
atica
lly
d
e
-
id
en
tif
y
in
g
clin
ical
d
o
cu
m
e
n
ts
s
to
r
ed
as
im
ag
es.
I
n
th
is
r
eg
ar
d
,
s
ev
er
al
tech
n
i
q
u
es
wer
e
an
aly
ze
d
.
Am
o
n
g
th
em
,
co
n
v
en
tio
n
al
wo
r
d
-
lev
el
v
ec
to
r
r
ep
r
esen
tatio
n
s
,
lik
e
w
o
r
d
2
v
ec
[
1
4
]
,
Glo
Ve
[
1
5
]
,
a
n
d
f
astTex
t
[
1
6
]
,
r
ed
u
ce
all
p
o
ten
tial
s
em
an
tics
o
f
a
wo
r
d
in
to
a
s
in
g
le
v
ec
to
r
,
u
n
ab
le
to
d
is
ce
r
n
d
if
f
er
en
t
wo
r
d
s
en
s
es
d
ep
en
d
in
g
o
n
th
e
co
n
tex
t
in
wh
ich
th
ey
ap
p
ea
r
.
E
L
Mo
[
1
7
]
an
d
B
E
R
T
[
2
]
o
f
f
er
r
o
b
u
s
t
s
o
lu
tio
n
s
f
o
r
co
n
tex
t
u
alize
d
w
o
r
d
r
ep
r
esen
tatio
n
s
co
m
p
ar
ed
t
o
o
th
er
lan
g
u
ag
e
m
o
d
els.
E
L
Mo
ac
h
iev
es
co
n
t
ex
t
s
en
s
itiv
ity
b
y
p
r
e
-
tr
ain
in
g
o
n
ex
ten
s
iv
e
tex
t
co
r
p
o
r
a,
g
e
n
er
atin
g
u
n
iq
u
e
em
b
ed
d
in
g
s
f
o
r
ea
ch
wo
r
d
in
a
g
iv
e
n
s
en
ten
ce
.
T
h
ese
em
b
ed
d
i
n
g
s
ar
e
s
u
b
s
eq
u
en
tly
u
tili
ze
d
in
d
o
wn
s
tr
ea
m
task
s
.
I
n
co
n
tr
ast,
B
E
R
T
i
s
a
m
o
r
e
co
m
p
lex
m
o
d
el
with
m
o
r
e
p
ar
am
eter
s
,
af
f
o
r
d
i
n
g
it
en
h
a
n
ce
d
r
ep
r
esen
tatio
n
al
ca
p
a
b
ilit
ies.
No
tab
ly
,
B
E
R
T
g
o
es
b
ey
o
n
d
p
r
o
v
i
d
in
g
m
er
e
wo
r
d
em
b
ed
d
i
n
g
s
;
it
ca
n
b
e
s
ea
m
less
ly
in
teg
r
ated
i
n
to
s
p
ec
if
ic
task
s
an
d
f
in
e
-
tu
n
e
d
f
o
r
e
n
h
an
ce
d
task
-
s
p
ec
if
ic
p
er
f
o
r
m
a
n
ce
.
Ov
er
all
,
B
E
R
T
h
as
d
em
o
n
s
tr
ated
s
u
p
er
io
r
p
er
f
o
r
m
an
ce
co
m
p
ar
e
d
to
E
L
Mo
an
d
n
o
n
-
co
n
tex
tu
al
em
b
ed
d
i
n
g
s
ac
r
o
s
s
v
ar
io
u
s
ap
p
licatio
n
s
,
in
clu
d
in
g
th
o
s
e
in
th
e
m
ed
ical
f
iel
d
[
1
8
]
.
W
ith
th
ese
ad
v
an
ce
m
e
n
ts
,
we
f
o
cu
s
e
x
clu
s
iv
ely
o
n
B
E
R
T
[
2
]
in
th
is
wo
r
k
,
o
m
itti
n
g
E
L
M
o
[
1
7
]
an
d
n
o
n
-
c
o
n
tex
t
u
al
em
b
ed
d
in
g
m
eth
o
d
s
.
T
h
e
f
r
a
m
ewo
r
k
f
o
r
au
to
m
atic
d
e
-
id
en
tific
atio
n
o
f
clin
ical
d
o
cu
m
en
t
im
ag
es
is
s
h
o
wn
in
Fig
u
r
e
1
,
c
o
m
p
r
is
in
g
s
ev
e
r
al
s
tep
s
wh
ich
ar
e
ex
p
lain
ed
in
t
h
is
s
ec
tio
n
.
Fig
u
r
e
1
.
Fra
m
ewo
r
k
o
f
clin
ic
al
d
o
cu
m
e
n
t im
ag
e
d
e
-
id
en
tifi
ca
tio
n
2
.
1
.
I
ma
g
e
-
to
-
t
e
x
t
c
o
nv
er
s
io
n
B
E
R
T
m
o
d
els
p
r
o
ce
s
s
o
n
ly
t
ex
t
d
ata,
s
o
th
e
clin
ical
d
o
cu
m
en
t
im
ag
e
m
u
s
t
b
e
co
n
v
er
t
ed
to
tex
t
u
s
in
g
T
ess
er
ac
t
OC
R
.
T
ess
er
a
ct
OC
R
co
n
v
er
ts
th
e
im
ag
e
to
tex
t
an
d
p
r
o
v
id
es
tex
t
as
to
k
e
n
s
,
p
r
o
v
id
in
g
ea
ch
to
k
en
l
o
ca
tio
n
in
t
h
e
in
p
u
t
im
ag
e.
T
h
ese
to
k
e
n
lo
ca
tio
n
s
ar
e
u
s
ef
u
l
wh
ile
r
em
o
v
in
g
to
k
en
s
f
r
o
m
th
e
im
ag
e
if
to
k
en
s
ar
e
id
e
n
tifie
d
as PHI
to
k
en
s
.
2
.
2
.
P
re
-
pro
ce
s
s
ing
Var
io
u
s
tex
t
p
r
e
-
p
r
o
ce
s
s
in
g
t
ask
s
ar
e
p
er
f
o
r
m
e
d
in
th
is
s
tep
to
clea
n
th
e
co
n
v
er
ted
tex
t.
T
h
e
task
s
p
er
f
o
r
m
ed
ar
e
r
em
o
v
i
n
g
p
u
n
ctu
atio
n
m
ar
k
s
f
r
o
m
th
e
tex
t,
s
u
ch
as
co
m
m
as,
p
er
io
d
s
,
q
u
esti
o
n
m
ar
k
s
,
etc.
Pu
n
ctu
atio
n
r
em
o
v
al
h
elp
s
to
elim
in
ate
u
n
n
ec
ess
ar
y
s
y
m
b
o
ls
an
d
m
ak
e
th
e
te
x
t
m
o
r
e
clea
n
an
d
s
im
p
le
.
No
r
m
aliza
tio
n
o
f
te
x
t
b
y
p
e
r
f
o
r
m
in
g
ca
s
e
co
n
v
er
s
io
n
h
e
lp
s
r
ed
u
ce
tex
t
v
ar
iab
ilit
y
a
n
d
m
a
k
e
it
m
o
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
7
2
2
-
2
5
8
6
I
AE
S
I
n
t
J
R
o
b
&
A
u
to
m
,
Vo
l
.
14
,
No
.
2
,
J
u
n
e
20
2
5
:
2
73
-
2
80
276
co
n
s
is
ten
t.
C
o
r
r
ec
tin
g
s
p
elli
n
g
er
r
o
r
s
o
r
ty
p
o
s
h
el
p
s
im
p
r
o
v
e
tex
t
q
u
ality
a
n
d
r
ea
d
ab
ilit
y
an
d
av
o
id
s
co
n
f
u
s
io
n
o
r
m
is
u
n
d
er
s
tan
d
in
g
.
2
.
3
.
De
-
identif
ica
t
io
n
Her
e,
we
ad
o
p
ted
t
h
e
m
o
d
el
ar
ch
itectu
r
e
o
f
B
E
R
T
with
ad
d
itio
n
al
co
m
p
o
n
e
n
ts
.
T
o
id
en
tify
PHI
to
k
en
s
,
th
e
r
esu
lt
o
f
t
h
e
f
i
n
al
l
ay
er
o
f
B
E
R
T
is
g
iv
e
n
to
a
f
u
lly
co
n
n
ec
ted
d
en
s
e
lay
e
r
wit
h
s
ev
en
o
u
t
p
u
ts
f
o
r
p
er
f
o
r
m
in
g
n
am
ed
en
tity
r
ec
o
g
n
itio
n
,
wh
er
e
s
ev
e
n
in
cl
u
d
e
t
h
e
s
ix
PHI
ca
te
g
o
r
ies
as
s
p
ec
i
f
ied
in
T
ab
le
2
p
lu
s
o
n
e
n
o
n
-
PHI
ty
p
e.
Ou
r
ex
p
er
i
m
en
t
aim
s
to
ev
alu
ate
th
e
p
er
f
o
r
m
an
ce
o
f
v
ar
io
u
s
p
r
e
-
tr
ain
ed
B
E
R
T
m
o
d
els
in
id
en
tify
in
g
PHI
T
o
k
e
n
s
o
f
clin
ical
d
o
cu
m
en
ts
.
T
ab
le
1
lis
ts
th
e
v
a
r
io
u
s
B
E
R
T
m
o
d
els
u
ti
lized
in
th
is
wo
r
k
.
T
h
e
s
eq
u
en
ce
o
f
p
r
e
-
p
r
o
ce
s
s
ed
tex
t
to
k
en
s
is
f
e
d
to
th
ese
p
r
e
-
tr
ain
ed
B
E
R
T
m
o
d
els
to
r
e
co
g
n
ize
wh
et
h
er
a
to
k
en
is
o
n
e
o
f
th
e
s
ix
PHI
ca
teg
o
r
ies
o
r
a
n
o
n
-
PHI
ty
p
e.
I
f
a
to
k
en
is
id
en
tifie
d
as
PHI
,
it
is
r
em
o
v
ed
f
r
o
m
th
e
clin
ical
d
o
cu
m
en
t
im
a
g
e
with
th
e
h
elp
o
f
th
e
T
o
k
e
n
lo
ca
tio
n
s
to
r
ed
at
th
e
tim
e
o
f
im
ag
e
-
to
-
tex
t
co
n
v
er
s
io
n
.
T
h
ese
id
en
tifie
d
PHI
to
k
en
s
ar
e
s
to
r
e
d
in
a
tex
t
f
ile
f
o
r
f
u
t
u
r
e
r
e
-
id
en
tific
atio
n
.
T
ab
le
1
.
Pre
-
tr
ai
n
ed
m
o
d
els em
p
lo
y
ed
in
th
is
wo
r
k
Pre
-
t
r
a
i
n
e
d
m
o
d
e
l
D
e
scri
p
t
i
o
n
B
ER
T
ba
s
e
[
2
]
B
a
se
B
E
R
T
i
s
p
r
e
-
t
r
a
i
n
e
d
w
i
t
h
1
1
0
mi
l
l
i
o
n
p
a
r
a
me
t
e
r
s
w
i
t
h
u
n
c
a
se
d
t
o
k
e
n
s,
a
l
so
p
r
e
-
t
r
a
i
n
e
d
w
i
t
h
B
o
o
k
s C
o
r
p
u
s
(
8
0
0
M
w
o
r
d
s)
[
1
9
]
a
n
d
En
g
l
i
s
h
W
i
k
i
p
e
d
i
a
(
2
.
5
B
w
o
r
d
s)
B
i
o
B
E
R
T
[
2
0
]
I
t
i
s p
r
e
-
t
r
a
i
n
e
d
w
i
t
h
P
u
b
M
e
d
a
b
st
r
a
c
t
s a
n
d
C
e
n
t
r
a
l
a
r
t
i
c
l
e
s
S
c
i
B
ER
T
[
2
1
]
I
t
i
s p
r
e
-
t
r
a
i
n
e
d
w
i
t
h
a
c
a
d
e
mi
c
c
o
r
p
u
s
f
r
o
m
t
h
e
se
ma
n
t
i
c
s
c
h
o
l
a
r
C
l
i
n
i
c
a
l
B
ER
T
[
2
2
]
I
t
i
s p
r
e
-
t
r
a
i
n
e
d
w
i
t
h
M
I
M
I
C
-
I
I
I
v
1
.
4
d
a
t
a
b
a
s
e
R
o
B
E
R
Ta
[
2
3
]
Pre
-
t
r
a
i
n
e
d
w
i
t
h
C
C
-
N
EWS
[
2
4
]
,
O
p
e
n
W
e
b
Te
x
t
,
a
n
d
S
t
o
r
i
e
s
[
2
5
]
B
ER
T
b
y
Jo
h
n
so
n
[
1
3
]
Pre
-
t
r
a
i
n
e
d
w
i
t
h
i
2
b
2
2
0
0
6
C
o
r
p
u
s
[
5
]
,
i
2
b
2
2
0
1
4
C
o
r
p
u
s
[
6
]
,
P
h
y
si
o
N
e
t
C
o
r
p
u
s
[
2
6
]
,
a
n
d
D
e
r
n
o
n
c
o
u
r
t
_
l
e
e
C
o
r
p
u
s
d
a
t
a
se
t
s
[
9
]
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
3
.
1
.
Clini
ca
l
do
cum
ent
im
a
g
e
da
t
a
s
et
A
s
y
n
th
etic
d
ataset
h
as
b
ee
n
cr
ea
ted
f
o
r
ex
p
e
r
im
en
tal
p
u
r
p
o
s
es,
co
n
s
is
tin
g
o
f
clin
ical
d
o
cu
m
e
n
t
im
ag
es
o
b
tain
e
d
f
r
o
m
v
ar
io
u
s
h
o
s
p
ital
p
atien
ts
.
T
h
e
r
e
p
o
r
ts
wer
e
a
cq
u
ir
e
d
with
th
e
p
atien
t
’
s
ex
p
licit
p
er
m
is
s
io
n
;
n
o
n
eth
eless
,
th
e
d
ataset
r
em
ain
s
in
ac
ce
s
s
ib
le
to
th
e
g
en
er
al
p
u
b
lic
o
win
g
to
th
e
co
n
f
id
en
tial
n
atu
r
e
o
f
th
e
d
ata.
T
h
e
d
ataset
u
s
ed
in
th
is
wo
r
k
is
a
s
y
n
t
h
etic
co
llectio
n
c
o
n
s
is
tin
g
o
f
5
5
0
im
a
g
es
o
b
tai
n
ed
f
r
o
m
4
6
in
d
iv
id
u
als.
T
h
is
d
ataset
en
co
m
p
ass
es
a
v
ar
iety
o
f
d
o
cu
m
en
t
ty
p
es,
s
u
ch
as
d
i
s
ch
ar
g
e
s
u
m
m
ar
ies,
r
ad
io
lo
g
y
r
ep
o
r
ts
,
b
io
ch
em
is
tr
y
,
h
em
ato
lo
g
y
,
lab
o
r
ato
r
y
f
in
d
in
g
s
,
an
d
m
o
r
e.
T
h
e
f
iles
ar
e
a
cc
ess
ib
le
as
J
PEG
im
ag
es,
em
p
lo
y
i
n
g
t
h
e
R
GB
co
lo
r
s
y
s
tem
.
T
h
e
s
tr
u
ct
u
r
e
o
f
th
e
clin
ical
d
o
c
u
m
en
t
im
ag
e
ex
h
ib
its
a
co
n
s
id
er
ab
le
d
eg
r
ee
o
f
co
m
p
le
x
ity
.
T
h
e
lay
o
u
t
o
f
ea
ch
im
ag
e
is
d
is
tin
ct
an
d
c
o
n
tin
g
e
n
t
u
p
o
n
th
e
s
p
ec
if
ic
ty
p
e
o
f
r
ep
o
r
t,
with
r
ep
o
r
ts
o
r
i
g
in
atin
g
f
r
o
m
th
e
s
am
e
o
r
g
an
iz
atio
n
ex
h
ib
itin
g
d
iv
er
s
e
p
atte
r
n
s
.
I
m
p
lem
en
tin
g
HI
PAA
-
co
m
p
lian
t
s
u
b
s
titu
tio
n
s
an
d
d
eletin
g
p
er
s
o
n
ally
i
d
en
tifia
b
le
h
ea
lth
in
f
o
r
m
atio
n
(
PHI
)
id
e
n
tifie
r
s
in
s
id
e
th
e
d
ataset
en
h
an
ce
s
i
ts
ap
p
r
o
p
r
iaten
ess
f
o
r
au
to
m
ated
d
e
-
id
e
n
tific
atio
n
in
NL
P
r
esear
ch
.
T
ab
le
2
p
r
esen
ts
th
e
d
ataset
’
s
p
r
o
b
a
b
ilit
y
d
is
tr
ib
u
tio
n
o
f
PHI
.
T
ab
le
2
.
PHI
d
is
tr
ib
u
tio
n
s
o
f
c
lin
ical
d
o
cu
m
en
t im
a
g
e
d
atase
t
C
a
t
e
g
o
r
y
S
u
b
-
c
a
t
e
g
o
r
y
N
o
o
f
t
o
k
e
n
s
C
a
t
e
g
o
r
y
S
u
b
-
c
a
t
e
g
o
r
y
N
o
o
f
t
o
k
e
n
s
N
a
me
P
a
t
i
e
n
t
5
5
0
A
g
e
A
g
e
5
5
0
D
o
c
t
o
r
4
9
5
D
a
t
e
D
a
t
e
7
8
4
H
o
sp
i
t
a
l
3
7
8
C
o
n
t
a
c
t
P
h
o
n
e
2
4
5
O
r
g
a
n
i
z
a
t
i
o
n
1
7
2
M
o
b
i
l
e
4
3
8
Lo
c
a
t
i
o
n
D
n
o
1
2
6
F
a
x
47
S
t
r
e
e
t
4
1
2
Emai
l
3
8
9
C
i
t
y
5
5
0
ID
P
a
t
i
e
n
t
I
D
2
4
1
S
t
a
t
e
3
1
6
UHID
2
4
3
C
o
u
n
t
r
y
1
4
9
I
P
N
O
3
8
8
P
i
n
c
o
d
e
4
7
8
3
.
2
.
E
v
a
lua
t
i
o
n
T
h
e
p
er
f
o
r
m
an
ce
o
f
th
e
d
if
f
er
en
t
m
o
d
els
is
ass
es
s
ed
b
y
co
m
p
u
tin
g
ess
en
tial
m
etr
ics
s
u
ch
as
F1
Sco
r
e,
p
r
ec
is
io
n
,
an
d
r
ec
all
.
Of
th
e
m
etr
ics
c
o
n
s
id
er
ed
,
s
e
n
s
itiv
ity
is
r
eg
ar
d
e
d
as
t
h
e
m
o
s
t
cr
itical
in
d
icato
r
f
o
r
p
atien
t
d
e
-
id
en
tific
atio
n
s
in
ce
it
s
h
o
ws
th
e
p
r
o
p
o
r
tio
n
o
f
ac
cu
r
ately
id
en
tifie
d
in
d
iv
id
u
als
th
at
th
e
m
o
d
el
ap
p
r
o
p
r
iately
an
n
o
tates.
T
h
e
p
r
e
-
p
r
o
ce
s
s
in
g
an
d
e
v
alu
atio
n
o
f
in
d
iv
id
u
al
p
r
e
-
tr
ain
e
d
B
E
R
T
m
o
d
els
wer
e
co
n
d
u
cte
d
u
s
in
g
th
e
C
lin
ical
Do
cu
m
en
t
I
m
ag
es
d
ataset,
with
all
co
m
p
u
tatio
n
s
p
er
f
o
r
m
ed
o
n
a
s
in
g
le
Evaluation Warning : The document was created with Spire.PDF for Python.
I
AE
S
I
n
t
J
R
o
b
&
A
u
to
m
I
SS
N:
2722
-
2
5
8
6
E
xp
erimen
ta
l e
va
lu
a
tio
n
o
f b
i
d
ir
ec
tio
n
a
l e
n
co
d
er rep
r
esen
ta
tio
n
s
…
(
R
a
vich
a
n
d
r
a
S
r
ir
a
m
)
277
NVI
DI
A
Qu
ad
r
o
GV
1
0
0
2
4
GB
GP
U.
T
h
e
ev
alu
atio
n
o
f
v
ar
io
u
s
p
r
e
-
t
r
ain
ed
m
o
d
els,
in
clu
d
in
g
B
E
R
T
base
,
B
io
B
E
R
T
,
Sci
B
E
R
T
,
C
lin
ical
B
E
R
T
,
R
o
B
E
R
T
a,
an
d
th
e
B
E
R
T
-
b
ased
m
o
d
el
p
r
o
p
o
s
ed
b
y
[
1
3
]
,
is
co
n
d
u
cted
u
s
in
g
a
d
ataset
o
f
clin
ical
d
o
cu
m
en
t
im
ag
es.
T
h
e
f
in
d
in
g
s
o
f
clin
ical
d
o
cu
m
en
t
im
ag
e
d
e
-
id
en
tific
atio
n
ar
e
d
ep
icted
i
n
Fig
u
r
e
2
.
T
h
e
ta
b
u
latio
n
o
f
t
h
e
p
er
f
o
r
m
an
ce
o
u
tco
m
es
o
f
th
e
d
e
-
id
e
n
tific
atio
n
ap
p
r
o
ac
h
m
a
y
b
e
s
ee
n
in
T
a
b
le
3
.
B
E
R
T
b
y
J
o
h
o
n
s
o
n
[
1
3
]
m
o
d
el
ac
h
iev
ed
F1
S
co
r
e
o
f
9
2
.
4
.
As
s
tated
i
n
th
e
liter
atu
r
e
r
ev
iew,
n
o
r
esear
ch
er
h
as
y
et
to
d
e
-
id
en
tify
clin
ical
d
o
cu
m
en
t
i
m
ag
es.
Pre
v
io
u
s
ly
,
all
r
esear
ch
f
o
cu
s
ed
o
n
d
e
-
id
en
tify
in
g
clin
ical
f
r
ee
tex
t
o
r
elec
tr
o
n
ic
r
ec
o
r
d
s
.
W
ith
th
is
p
r
o
p
o
s
ed
m
o
d
el,
we
ca
n
d
e
-
id
en
tify
clin
ical
d
o
cu
m
e
n
t im
ag
es a
n
d
u
s
e
th
e
m
f
o
r
clin
ical
r
esear
c
h
.
Fig
u
r
e
2
.
Sam
p
le
o
u
tp
u
ts
o
f
cl
in
ical
d
o
cu
m
e
n
t im
ag
e
de
-
id
e
n
tific
atio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
7
2
2
-
2
5
8
6
I
AE
S
I
n
t
J
R
o
b
&
A
u
to
m
,
Vo
l
.
14
,
No
.
2
,
J
u
n
e
20
2
5
:
2
73
-
2
80
278
T
ab
le
3
.
Per
f
o
r
m
an
ce
o
f
de
-
id
en
tific
atio
n
o
n
cli
n
ical
d
o
cu
m
en
t im
ag
es d
ataset
Pre
-
t
r
a
i
n
e
d
m
o
d
e
l
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
S
c
o
r
e
B
ER
T
ba
s
e
[
2
]
8
4
.
5
6
8
3
.
1
5
8
4
.
7
8
B
i
o
B
E
R
T
[
2
0
]
8
6
.
8
7
8
5
.
2
1
8
6
.
0
5
S
c
i
B
ER
T
[
2
1
]
9
1
.
4
8
7
.
4
8
9
.
3
6
C
l
i
n
i
c
a
l
B
ER
T
[
2
2
]
8
8
.
9
9
2
.
1
9
0
.
4
5
R
o
B
E
R
Ta
[
2
3
]
9
0
.
7
5
8
9
.
5
9
0
.
1
2
B
ER
T
b
y
Jo
h
n
so
n
[
1
3
]
9
3
.
8
9
9
1
.
0
6
9
2
.
4
5
3
.
4
.
Dis
cus
s
io
n
T
h
e
o
v
er
all
p
er
f
o
r
m
a
n
ce
s
o
f
all
p
r
e
-
tr
ain
e
d
m
o
d
els
ex
ce
p
t
B
E
R
T
b
y
J
o
h
o
n
s
o
n
[
1
3
]
ar
e
n
ea
r
b
y
an
d
h
av
e
a
d
if
f
e
r
en
ce
o
f
1
%
with
r
esp
ec
t
to
all
m
etr
ics.
B
E
R
T
b
y
J
o
h
n
s
o
n
[
1
3
]
m
o
d
el
h
as
a
h
ig
h
F1
Sco
r
e
o
f
9
2
.
4
5
%.
C
o
m
p
ar
e
d
to
o
th
e
r
p
r
e
-
tr
ain
ed
m
o
d
els,
B
E
R
T
b
y
J
o
h
n
s
o
n
[
1
3
]
is
a
tr
ain
ed
m
o
d
el
o
n
to
p
o
f
B
E
R
T
base
by
[
2
]
with
d
e
-
id
en
tific
atio
n
d
atasets
o
f
i2
b
2
2
0
0
6
co
r
p
u
s
[
5
]
,
i2
b
2
2
0
1
4
co
r
p
u
s
[
6
]
,
Ph
y
s
io
n
et
co
r
p
u
s
[
2
6
]
an
d
Der
n
o
n
co
u
r
t
-
lee
c
o
r
p
u
s
[
9
]
.
T
h
is
clin
ical
-
r
elate
d
s
p
ec
if
ic
tr
ain
in
g
g
av
e
t
h
e
b
est
p
e
r
f
o
r
m
an
ce
.
B
u
t
B
E
R
T
by
[
1
3
]
h
ad
ac
h
iev
ed
an
F1
S
co
r
e
o
f
9
8
.
8
2
o
n
th
e
i2
b
2
2
0
1
4
ch
allen
g
e
[
6
]
test
s
et.
T
h
e
clin
ical
in
f
o
r
m
atio
n
o
n
th
ese
d
o
cu
m
en
t
im
a
g
es
is
n
o
t
u
n
if
o
r
m
ly
s
tr
u
ctu
r
ed
.
T
h
e
lay
o
u
ts
o
f
th
ese
d
o
c
u
m
en
ts
ch
an
g
e
f
r
o
m
o
n
e
h
ea
lth
in
s
titu
te
to
a
n
o
th
er
,
an
d
s
o
d
o
th
e
ty
p
es
o
f
r
e
p
o
r
ts
.
W
h
en
im
ag
es
ar
e
c
o
n
v
er
te
d
to
t
ex
t,
th
e
i
n
f
o
r
m
atio
n
is
n
o
t
ex
ac
tly
tr
an
s
f
er
r
e
d
.
T
h
e
tex
t
in
th
ese
d
o
c
u
m
en
ts
m
a
y
h
av
e
d
i
f
f
er
en
t
f
o
n
t
s
ty
les
an
d
s
izes.
Als
o
,
th
e
tex
t
is
n
o
t u
n
if
o
r
m
ly
alig
n
e
d
.
So
m
e
is
s
u
es
in
ca
p
tu
r
in
g
p
h
y
s
ical
d
o
cu
m
e
n
ts
in
to
im
ag
es in
clu
d
e
th
e
d
eg
r
ad
atio
n
o
f
a
f
ew
d
o
cu
m
en
ts
d
u
e
to
ag
e.
T
h
e
h
ea
d
er
an
d
f
o
o
ter
o
f
th
ese
d
o
cu
m
en
t
im
ag
es
h
av
e
th
e
h
ea
lth
in
s
titu
te
’
s
n
am
e,
ad
d
r
ess
,
co
n
tact
d
etails,
an
d
lo
g
o
,
with
d
if
f
er
en
t
f
o
n
t
s
ty
les
an
d
s
izes.
C
ap
tu
r
in
g
in
f
o
r
m
atio
n
f
r
o
m
th
ese
lo
ca
tio
n
s
is
th
e
m
ajo
r
c
h
allen
g
e.
T
h
e
p
er
f
o
r
m
an
ce
o
f
B
E
R
T
b
y
J
o
h
n
s
o
n
[
1
3
]
f
o
r
i
n
d
iv
id
u
al
en
titi
es
o
f
PHI
ca
teg
o
r
ies
is
in
T
ab
le
4
.
I
t
s
h
o
ws
th
e
r
ea
s
o
n
f
o
r
th
e
lo
w
p
er
f
o
r
m
an
ce
o
f
B
E
R
T
b
y
J
o
h
n
s
o
n
[
1
3
]
co
m
p
ar
ed
to
t
h
e
i2
b
2
2
0
1
4
ch
allen
g
e.
I
d
en
tify
in
g
lo
ca
tio
n
a
n
d
co
n
tact
en
tity
t
y
p
es
is
less
ac
cu
r
ate
d
u
e
to
th
e
n
atu
r
e
o
f
th
ese
clin
ical
d
o
c
u
m
en
t im
ag
es.
T
ab
le
4
.
PHI
ca
teg
o
r
y
-
wis
e
p
e
r
f
o
r
m
a
n
ce
o
f
p
r
e
-
tr
ai
n
ed
B
E
R
T
b
y
J
o
h
n
s
o
n
[
1
3
]
En
t
i
t
y
Ty
p
e
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F
1
S
c
o
r
e
N
a
me
9
6
.
4
9
8
.
5
9
7
.
4
Lo
c
a
t
i
o
n
8
3
.
5
6
7
8
.
7
1
8
1
.
0
5
A
g
e
9
7
.
2
5
9
5
.
5
5
9
6
.
3
8
D
a
t
e
9
6
.
4
5
9
8
.
6
5
9
7
.
5
C
o
n
t
a
c
t
8
6
.
5
8
3
.
3
5
8
4
.
9
ID
9
8
.
4
9
6
.
6
7
9
7
.
5
4.
CO
NCLU
SI
O
N
T
h
is
wo
r
k
in
tr
o
d
u
ce
s
a
co
m
p
r
eh
en
s
iv
e
f
r
am
ewo
r
k
f
o
r
d
e
-
id
en
tify
in
g
clin
ical
d
o
c
u
m
e
n
t
im
ag
es
u
tili
zin
g
m
u
ltip
le
p
r
e
-
tr
ain
e
d
B
E
R
T
m
o
d
els.
T
h
e
m
eth
o
d
o
lo
g
y
c
o
n
s
is
ts
o
f
th
r
ee
ess
en
tial
s
tep
s
:
f
ir
s
t,
tr
an
s
f
o
r
m
in
g
im
ag
es
i
n
to
tex
t;
s
ec
o
n
d
,
p
r
ep
a
r
in
g
t
h
e
tex
t
to
en
h
an
ce
its
f
o
r
m
at
f
o
r
a
n
aly
s
is
;
an
d
th
ir
d
,
em
p
lo
y
in
g
a
p
r
e
-
tr
ain
e
d
B
E
R
T
m
o
d
el
t
o
d
etec
t
a
n
d
r
e
d
ac
t
PHI
to
k
en
s
.
W
e
g
en
er
ated
a
s
y
n
th
etic
d
ataset
o
f
5
5
0
clin
ical
d
o
cu
m
e
n
t
im
a
g
es
f
r
o
m
m
an
y
h
ea
lth
ca
r
e
o
r
g
an
izatio
n
s
to
v
alid
ate
th
e
ex
p
er
im
e
n
t.
Ou
r
co
m
p
ar
ativ
e
an
aly
s
is
o
f
p
u
b
licly
av
ailab
le
p
r
e
-
tr
ain
ed
B
E
R
T
m
o
d
els
r
ev
ea
led
n
o
tab
le
v
ar
iatio
n
s
in
p
er
f
o
r
m
an
ce
,
with
th
e
B
E
R
T
b
y
J
o
h
n
s
o
n
m
o
d
el
ac
h
iev
i
n
g
t
h
e
h
ig
h
est
F1
S
co
r
e
o
f
9
2
.
4
5
%.
T
h
is
p
er
f
o
r
m
an
ce
is
p
ar
ticu
lar
ly
s
ig
n
if
ican
t
wh
e
n
co
n
tr
asted
with
r
esu
lts
f
r
o
m
estab
lis
h
ed
d
atasets
u
s
ed
in
th
e
i2
b
2
2
0
0
6
an
d
i2
b
2
2
0
1
4
d
e
-
id
e
n
tific
atio
n
c
h
allen
g
es.
T
h
e
s
u
p
er
io
r
p
er
f
o
r
m
an
ce
o
f
th
ese
d
atasets
s
u
g
g
ests
th
at
th
e
B
E
R
T
b
y
J
o
h
n
s
o
n
m
o
d
el
m
ay
o
f
f
er
en
h
an
ce
d
ca
p
ab
ilit
ies
f
o
r
h
a
n
d
lin
g
th
e
s
p
ec
if
ic
ch
alle
n
g
es
p
r
esen
ted
b
y
clin
ical
d
o
cu
m
e
n
t
im
ag
es.
Ho
wev
er
,
t
h
e
co
m
p
le
x
lay
o
u
ts
ty
p
ical
o
f
clin
ical
d
o
cu
m
e
n
ts
h
in
d
er
e
d
co
n
v
er
tin
g
im
ag
es
to
tex
t,
wh
ich
o
f
ten
led
to
in
c
o
m
p
lete
o
r
in
ac
cu
r
ate
tex
t
ex
t
r
ac
tio
n
.
T
h
is
lim
itatio
n
u
n
d
er
s
co
r
es
a
cr
itical
ar
ea
f
o
r
f
u
tu
r
e
r
esear
ch
.
Su
b
s
eq
u
en
t
ef
f
o
r
ts
s
h
o
u
ld
aim
to
en
h
an
c
e
th
e
OC
R
p
h
ase
to
b
etter
m
a
n
ag
e
th
e
in
tr
icac
ies
o
f
v
ar
ied
d
o
cu
m
e
n
t
lay
o
u
ts
,
e
n
s
u
r
in
g
a
m
o
r
e
ac
cu
r
ate
tex
t
e
x
tr
ac
tio
n
.
Ad
d
itio
n
ally
,
f
u
t
u
r
e
wo
r
k
will
ex
p
lo
r
e
ad
v
an
ce
d
tech
n
iq
u
es
f
o
r
au
to
m
atica
lly
ex
tr
ac
tin
g
an
d
d
e
-
i
d
en
tify
in
g
PHI
f
r
o
m
th
ese
c
o
m
p
lex
d
o
c
u
m
en
ts
,
p
o
ten
tially
in
v
o
lv
in
g
in
teg
r
atin
g
AI
-
d
r
iv
en
lay
o
u
t
a
n
aly
s
is
to
o
ls
with
B
E
R
T
-
b
ase
d
tex
t
p
r
o
ce
s
s
in
g
m
eth
o
d
o
l
o
g
ies.
B
y
ad
v
an
cin
g
th
ese
ar
ea
s
,
th
e
f
ield
ca
n
m
o
v
e
clo
s
er
to
f
u
lly
a
u
to
m
at
ed
,
h
ig
h
ly
ac
c
u
r
ate
s
y
s
tem
s
f
o
r
th
e
s
ec
u
r
e
h
an
d
l
in
g
o
f
s
en
s
itiv
e
h
ea
lth
in
f
o
r
m
atio
n
,
th
er
eb
y
ex
p
a
n
d
in
g
t
h
e
p
o
s
s
ib
ilit
ies
f
o
r
clin
ical
r
esear
ch
wh
ile
r
ig
o
r
o
u
s
ly
s
af
eg
u
ar
d
in
g
p
atien
t p
r
iv
a
cy
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
AE
S
I
n
t
J
R
o
b
&
A
u
to
m
I
SS
N:
2722
-
2
5
8
6
E
xp
erimen
ta
l e
va
lu
a
tio
n
o
f b
i
d
ir
ec
tio
n
a
l e
n
co
d
er rep
r
esen
ta
tio
n
s
…
(
R
a
vich
a
n
d
r
a
S
r
ir
a
m
)
279
F
UNDING
I
NF
O
R
M
A
T
I
O
N
No
f
u
n
d
in
g
was r
ec
eiv
e
d
.
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
R
av
ich
an
d
r
a
Srir
am
✓
✓
✓
✓
Siv
a
Sath
y
a
Su
n
d
ar
am
✓
✓
✓
✓
✓
S.
L
o
u
r
d
u
Ma
r
ie
So
p
h
ie
✓
✓
✓
✓
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
:
V
a
l
i
d
a
t
i
o
n
Fo
:
Fo
r
ma
l
a
n
a
l
y
si
s
I
:
I
n
v
e
st
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
s
u
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
si
o
n
P
:
P
r
o
j
e
c
t
a
d
m
i
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
s
i
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
in
d
in
g
s
o
f
th
is
s
tu
d
y
ar
e
av
aila
b
le
o
n
r
eq
u
est
f
r
o
m
th
e
co
r
r
esp
o
n
d
in
g
au
th
o
r
,
R
av
ich
an
d
r
a
Srir
am
.
T
h
e
d
ata,
wh
ich
co
n
tain
in
f
o
r
m
atio
n
th
at
co
u
ld
co
m
p
r
o
m
is
e
th
e
p
r
iv
ac
y
o
f
r
esear
ch
p
ar
ticip
an
ts
,
ar
e
n
o
t
p
u
b
licly
av
ailab
le
d
u
e
to
ce
r
tain
r
estrictio
n
s
.
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
i
n
d
in
g
s
o
f
t
h
is
s
tu
d
y
ar
e
av
ailab
le
f
r
o
m
th
e
c
o
r
r
esp
o
n
d
i
n
g
au
th
o
r
,
R
av
ich
an
d
r
a
Srir
am
,
u
p
o
n
r
ea
s
o
n
ab
le
r
eq
u
est.
RE
F
E
R
E
NC
E
S
[
1
]
G
o
v
I
n
f
o
,
“
H
e
a
l
t
h
i
n
s
u
r
a
n
c
e
p
o
r
t
a
b
i
l
i
t
y
a
n
d
a
c
c
o
u
n
t
a
b
i
l
i
t
y
a
c
t
o
f
1
9
9
6
.
P
u
b
l
i
c
L
a
w
1
0
4
-
1
9
1
,
”
1
9
9
6
.
[
2
]
J.
D
e
v
l
i
n
,
M
.
-
W
.
C
h
a
n
g
,
K
.
L
e
e
,
a
n
d
K
.
T
o
u
t
a
n
o
v
a
,
“
B
ER
T
:
p
r
e
-
t
r
a
i
n
i
n
g
o
f
d
e
e
p
b
i
d
i
r
e
c
t
i
o
n
a
l
t
r
a
n
sf
o
r
m
e
r
s
f
o
r
l
a
n
g
u
a
g
e
u
n
d
e
r
s
t
a
n
d
i
n
g
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
2
0
1
9
C
o
n
f
e
r
e
n
c
e
o
f
t
h
e
N
o
rt
h
,
2
0
1
9
,
p
p
.
4
1
7
1
–
4
1
8
6
.
d
o
i
:
1
0
.
1
8
6
5
3
/
v
1
/
N
1
9
-
1
4
2
3
.
[
3
]
M
.
S
a
e
e
d
e
t
a
l
.
,
“
M
u
l
t
i
p
a
r
a
me
t
e
r
i
n
t
e
l
l
i
g
e
n
t
m
o
n
i
t
o
r
i
n
g
i
n
i
n
t
e
n
si
v
e
c
a
r
e
I
I
:
a
p
u
b
l
i
c
-
a
c
c
e
ss
i
n
t
e
n
s
i
v
e
c
a
r
e
u
n
i
t
d
a
t
a
b
a
se
,
”
C
r
i
t
i
c
a
l
C
a
r
e
M
e
d
i
c
i
n
e
,
v
o
l
.
3
9
,
n
o
.
5
,
p
p
.
9
5
2
–
9
6
0
,
M
a
y
2
0
1
1
,
d
o
i
:
1
0
.
1
0
9
7
/
C
C
M
.
0
b
0
1
3
e
3
1
8
2
0
a
9
2
c
6
.
[
4
]
A
.
E
.
W
.
J
o
h
n
s
o
n
e
t
a
l
.
,
“
M
I
M
I
C
-
I
I
I
,
a
f
r
e
e
l
y
a
c
c
e
ss
i
b
l
e
c
r
i
t
i
c
a
l
c
a
r
e
d
a
t
a
b
a
s
e
,
”
S
c
i
e
n
t
i
f
i
c
D
a
t
a
,
v
o
l
.
3
,
n
o
.
1
,
p
.
1
6
0
0
3
5
,
M
a
y
2
0
1
6
,
d
o
i
:
1
0
.
1
0
3
8
/
sd
a
t
a
.
2
0
1
6
.
3
5
.
[
5
]
O
.
U
z
u
n
e
r
,
Y
.
Lu
o
,
a
n
d
P
.
S
z
o
l
o
v
i
t
s
,
“
Ev
a
l
u
a
t
i
n
g
t
h
e
s
t
a
t
e
-
of
-
t
h
e
-
a
r
t
i
n
a
u
t
o
ma
t
i
c
d
e
-
i
d
e
n
t
i
f
i
c
a
t
i
o
n
,
”
J
o
u
r
n
a
l
o
f
t
h
e
Am
e
ri
c
a
n
Me
d
i
c
a
l
I
n
f
o
rm
a
t
i
c
s
Ass
o
c
i
a
t
i
o
n
,
v
o
l
.
1
4
,
n
o
.
5
,
p
p
.
5
5
0
–
5
6
3
,
S
e
p
.
2
0
0
7
,
d
o
i
:
1
0
.
1
1
9
7
/
j
a
mi
a
.
M
2
4
4
4
.
[
6
]
A
.
S
t
u
b
b
s
a
n
d
Ö
.
U
z
u
n
e
r
,
“
A
n
n
o
t
a
t
i
n
g
l
o
n
g
i
t
u
d
i
n
a
l
c
l
i
n
i
c
a
l
n
a
r
r
a
t
i
v
e
s
f
o
r
d
e
-
i
d
e
n
t
i
f
i
c
a
t
i
o
n
:
Th
e
2
0
1
4
i
2
b
2
/
U
TH
e
a
l
t
h
c
o
r
p
u
s,
”
J
o
u
rn
a
l
o
f
Bi
o
m
e
d
i
c
a
l
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
5
8
,
p
p
.
S
2
0
--
S
2
9
,
D
e
c
.
2
0
1
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
b
i
.
2
0
1
5
.
0
7
.
0
2
0
.
[
7
]
A
.
S
t
u
b
b
s,
M
.
F
i
l
a
n
n
i
n
o
,
a
n
d
Ö
.
U
z
u
n
e
r
,
“
De
-
i
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
p
s
y
c
h
i
a
t
r
i
c
i
n
t
a
k
e
r
e
c
o
r
d
s
:
o
v
e
r
v
i
e
w
o
f
2
0
1
6
C
EG
S
N
-
G
R
I
D
sh
a
r
e
d
t
a
s
k
s
Tr
a
c
k
1
,
”
J
o
u
rn
a
l
o
f
Bi
o
m
e
d
i
c
a
l
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
7
5
,
p
p
.
S
4
--
S
1
8
,
N
o
v
.
2
0
1
7
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
b
i
.
2
0
1
7
.
0
6
.
0
1
1
.
[
8
]
R
.
S
r
i
r
a
m,
S
.
S
.
S
u
n
d
a
r
a
m,
a
n
d
S
.
L.
S
o
p
h
i
e
,
“
D
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
s
f
o
r
a
u
t
o
ma
t
i
c
d
e
-
i
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
c
l
i
n
i
c
a
l
t
e
x
t
,
”
i
n
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
e
r,
C
o
m
m
u
n
i
c
a
t
i
o
n
,
a
n
d
S
i
g
n
a
l
Pr
o
c
e
ssi
n
g
,
2
0
2
3
,
p
p
.
1
1
6
–
1
2
7
.
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
1
-
3
9
8
1
1
-
7
_
1
0
.
[
9
]
F
.
D
e
r
n
o
n
c
o
u
r
t
,
J.
Y
.
L
e
e
,
O
.
U
z
u
n
e
r
,
a
n
d
P
.
S
z
o
l
o
v
i
t
s
,
“
De
-
i
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
p
a
t
i
e
n
t
n
o
t
e
s
w
i
t
h
r
e
c
u
r
r
e
n
t
n
e
u
r
a
l
n
e
t
w
o
r
k
s,
”
J
o
u
rn
a
l
o
f
t
h
e
Am
e
ri
c
a
n
Me
d
i
c
a
l
I
n
f
o
rm
a
t
i
c
s
Asso
c
i
a
t
i
o
n
,
v
o
l
.
2
4
,
n
o
.
3
,
p
p
.
5
9
6
–
6
0
6
,
M
a
y
2
0
1
7
,
d
o
i
:
1
0
.
1
0
9
3
/
j
a
mi
a
/
o
c
w
1
5
6
.
[
1
0
]
G
.
La
m
p
l
e
,
M
.
B
a
l
l
e
st
e
r
o
s,
S
.
S
u
b
r
a
man
i
a
n
,
K
.
K
a
w
a
k
a
mi
,
a
n
d
C
.
D
y
e
r
,
“
N
e
u
r
a
l
a
r
c
h
i
t
e
c
t
u
r
e
s
f
o
r
n
a
m
e
d
e
n
t
i
t
y
r
e
c
o
g
n
i
t
i
o
n
,
”
i
n
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
2
0
1
6
C
o
n
f
e
re
n
c
e
o
f
t
h
e
N
o
r
t
h
Am
e
r
i
c
a
n
C
h
a
p
t
e
r
o
f
t
h
e
Ass
o
c
i
a
t
i
o
n
f
o
r
C
o
m
p
u
t
a
t
i
o
n
a
l
L
i
n
g
u
i
st
i
c
s:
H
u
m
a
n
L
a
n
g
u
a
g
e
T
e
c
h
n
o
l
o
g
i
e
s
,
2
0
1
6
,
p
p
.
2
6
0
–
2
7
0
.
d
o
i
:
1
0
.
1
8
6
5
3
/
v
1
/
N
1
6
-
1
0
3
0
.
[
1
1
]
T.
A
h
m
e
d
,
M
.
M
.
A
l
A
z
i
z
,
a
n
d
N
.
M
o
h
a
mm
e
d
,
“
De
-
i
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
e
l
e
c
t
r
o
n
i
c
h
e
a
l
t
h
r
e
c
o
r
d
u
si
n
g
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
S
c
i
e
n
t
i
f
i
c
Re
p
o
r
t
s
,
v
o
l
.
1
0
,
n
o
.
1
,
p
.
1
8
6
0
0
,
O
c
t
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
3
8
/
s4
1
5
9
8
-
0
2
0
-
7
5
5
4
4
-
1.
[
1
2
]
R
.
C
a
t
e
l
l
i
,
V
.
C
a
s
o
l
a
,
G
.
D
e
P
i
e
t
r
o
,
H
.
F
u
j
i
t
a
,
a
n
d
M
.
Es
p
o
s
i
t
o
,
“
C
o
m
b
i
n
i
n
g
c
o
n
t
e
x
t
u
a
l
i
z
e
d
w
o
r
d
r
e
p
r
e
se
n
t
a
t
i
o
n
a
n
d
s
u
b
-
d
o
c
u
me
n
t
l
e
v
e
l
a
n
a
l
y
s
i
s
t
h
r
o
u
g
h
B
i
-
LST
M
+
C
R
F
a
r
c
h
i
t
e
c
t
u
r
e
f
o
r
c
l
i
n
i
c
a
l
d
e
-
i
d
e
n
t
i
f
i
c
a
t
i
o
n
,
”
K
n
o
w
l
e
d
g
e
-
Ba
s
e
d
S
y
st
e
m
s
,
v
o
l
.
2
1
3
,
p
.
1
0
6
6
4
9
,
F
e
b
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
k
n
o
s
y
s.
2
0
2
0
.
1
0
6
6
4
9
.
[
1
3
]
A
.
E.
W
.
J
o
h
n
s
o
n
,
L.
B
u
l
g
a
r
e
l
l
i
,
a
n
d
T.
J.
P
o
l
l
a
r
d
,
“
D
e
i
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
f
r
e
e
-
t
e
x
t
me
d
i
c
a
l
r
e
c
o
r
d
s
u
si
n
g
p
r
e
-
t
r
a
i
n
e
d
b
i
d
i
r
e
c
t
i
o
n
a
l
t
r
a
n
sf
o
r
mers
,
”
i
n
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
A
C
M
C
o
n
f
e
re
n
c
e
o
n
H
e
a
l
t
h
,
I
n
f
e
ren
c
e
,
a
n
d
L
e
a
rn
i
n
g
,
A
p
r
.
2
0
2
0
,
p
p
.
2
1
4
–
2
2
1
.
d
o
i
:
1
0
.
1
1
4
5
/
3
3
6
8
5
5
5
.
3
3
8
4
4
5
5
.
[
1
4
]
T.
M
i
k
o
l
o
v
,
K
.
C
h
e
n
,
G
.
C
o
r
r
a
d
o
,
a
n
d
J.
D
e
a
n
,
“
D
i
s
t
r
i
b
u
t
e
d
r
e
p
r
e
s
e
n
t
a
t
i
o
n
s
o
f
w
o
r
d
s
a
n
d
p
h
r
a
ses
a
n
d
t
h
e
i
r
c
o
mp
o
s
i
t
i
o
n
a
l
i
t
y
,
”
N
e
u
ra
l
i
n
f
o
rm
a
t
i
o
n
p
r
o
c
e
ssi
n
g
sy
st
e
m
s
,
v
o
l
.
1
,
p
p
.
1
–
9
,
2
0
0
6
.
[
1
5
]
J.
P
e
n
n
i
n
g
t
o
n
,
R
.
S
o
c
h
e
r
,
a
n
d
C
.
M
a
n
n
i
n
g
,
“
G
l
o
v
e
:
g
l
o
b
a
l
v
e
c
t
o
r
s
f
o
r
w
o
r
d
r
e
p
r
e
s
e
n
t
a
t
i
o
n
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
2
0
1
4
C
o
n
f
e
re
n
c
e
o
n
Em
p
i
ri
c
a
l
Me
t
h
o
d
s
i
n
N
a
t
u
r
a
l
L
a
n
g
u
a
g
e
Pro
c
e
ss
i
n
g
(
EM
N
L
P)
,
2
0
1
4
,
p
p
.
1
5
3
2
–
1
5
4
3
.
d
o
i
:
1
0
.
3
1
1
5
/
v
1
/
D
1
4
-
1
1
6
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
7
2
2
-
2
5
8
6
I
AE
S
I
n
t
J
R
o
b
&
A
u
to
m
,
Vo
l
.
14
,
No
.
2
,
J
u
n
e
20
2
5
:
2
73
-
2
80
280
[
1
6
]
P
.
B
o
j
a
n
o
w
s
k
i
,
E.
G
r
a
v
e
,
A
.
J
o
u
l
i
n
,
a
n
d
T
.
M
i
k
o
l
o
v
,
“
E
n
r
i
c
h
i
n
g
w
o
r
d
v
e
c
t
o
r
s
w
i
t
h
su
b
w
o
r
d
i
n
f
o
r
ma
t
i
o
n
,
”
T
r
a
n
s
a
c
t
i
o
n
s
o
f
t
h
e
Asso
c
i
a
t
i
o
n
f
o
r
C
o
m
p
u
t
a
t
i
o
n
a
l
L
i
n
g
u
i
st
i
c
s
,
v
o
l
.
5
,
p
p
.
1
3
5
–
1
4
6
,
D
e
c
.
2
0
1
7
,
d
o
i
:
1
0
.
1
1
6
2
/
t
a
c
l
_
a
_
0
0
0
5
1
.
[
1
7
]
M
.
P
e
t
e
r
s
e
t
a
l
.
,
“
D
e
e
p
c
o
n
t
e
x
t
u
a
l
i
z
e
d
w
o
r
d
r
e
p
r
e
s
e
n
t
a
t
i
o
n
s
,
”
i
n
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
2
0
1
8
C
o
n
f
e
re
n
c
e
o
f
t
h
e
N
o
r
t
h
Am
e
ri
c
a
n
C
h
a
p
t
e
r
o
f
t
h
e
Ass
o
c
i
a
t
i
o
n
f
o
r
C
o
m
p
u
t
a
t
i
o
n
a
l
L
i
n
g
u
i
st
i
c
s:
H
u
m
a
n
L
a
n
g
u
a
g
e
T
e
c
h
n
o
l
o
g
i
e
s,
V
o
l
u
m
e
1
(
L
o
n
g
Pa
p
e
rs)
,
2
0
1
8
,
p
p
.
2
2
2
7
–
2
2
3
7
.
d
o
i
:
1
0
.
1
8
6
5
3
/
v
1
/
N
1
8
-
1
2
0
2
.
[
1
8
]
Y
.
S
i
,
J
.
W
a
n
g
,
H
.
X
u
,
a
n
d
K
.
R
o
b
e
r
t
s,
“
E
n
h
a
n
c
i
n
g
c
l
i
n
i
c
a
l
c
o
n
c
e
p
t
e
x
t
r
a
c
t
i
o
n
w
i
t
h
c
o
n
t
e
x
t
u
a
l
e
m
b
e
d
d
i
n
g
s,
”
J
o
u
r
n
a
l
o
f
t
h
e
Am
e
ri
c
a
n
Me
d
i
c
a
l
I
n
f
o
rm
a
t
i
c
s A
ss
o
c
i
a
t
i
o
n
,
v
o
l
.
2
6
,
n
o
.
1
1
,
p
p
.
1
2
9
7
–
1
3
0
4
,
N
o
v
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
9
3
/
j
a
mi
a
/
o
c
z
0
9
6
.
[
1
9
]
Y
.
Zh
u
e
t
a
l
.
,
“
A
l
i
g
n
i
n
g
b
o
o
k
s
a
n
d
mo
v
i
e
s
:
t
o
w
a
r
d
s
st
o
r
y
-
l
i
k
e
v
i
su
a
l
e
x
p
l
a
n
a
t
i
o
n
s
b
y
w
a
t
c
h
i
n
g
m
o
v
i
e
s
a
n
d
r
e
a
d
i
n
g
b
o
o
k
s
,
”
i
n
2
0
1
5
I
EE
E
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
Vi
s
i
o
n
(
I
C
C
V)
,
D
e
c
.
2
0
1
5
,
p
p
.
1
9
–
2
7
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
V
.
2
0
1
5
.
1
1
.
[
2
0
]
I
.
B
e
l
t
a
g
y
,
K
.
L
o
,
a
n
d
A
.
C
o
h
a
n
,
“
S
c
i
B
ER
T
:
a
p
r
e
t
r
a
i
n
e
d
l
a
n
g
u
a
g
e
mo
d
e
l
f
o
r
s
c
i
e
n
t
i
f
i
c
t
e
x
t
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
2
0
1
9
C
o
n
f
e
re
n
c
e
o
n
Em
p
i
r
i
c
a
l
M
e
t
h
o
d
s
i
n
N
a
t
u
r
a
l
L
a
n
g
u
a
g
e
Pr
o
c
e
ssi
n
g
a
n
d
t
h
e
9
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
i
n
t
C
o
n
f
e
r
e
n
c
e
o
n
N
a
t
u
r
a
l
L
a
n
g
u
a
g
e
Pr
o
c
e
ssi
n
g
(
EM
N
L
P
-
I
J
C
N
L
P)
,
2
0
1
9
,
p
p
.
3
6
1
3
–
3
6
1
8
.
d
o
i
:
1
0
.
1
8
6
5
3
/
v
1
/
D
1
9
-
1
3
7
1
.
[
2
1
]
J.
Le
e
e
t
a
l
.
,
“
B
i
o
B
ER
T:
a
p
r
e
-
t
r
a
i
n
e
d
b
i
o
m
e
d
i
c
a
l
l
a
n
g
u
a
g
e
r
e
p
r
e
se
n
t
a
t
i
o
n
m
o
d
e
l
f
o
r
b
i
o
me
d
i
c
a
l
t
e
x
t
m
i
n
i
n
g
,
”
B
i
o
i
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
3
6
,
n
o
.
4
,
p
p
.
1
2
3
4
–
1
2
4
0
,
F
e
b
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
9
3
/
b
i
o
i
n
f
o
r
ma
t
i
c
s/
b
t
z
6
8
2
.
[
2
2
]
E.
A
l
s
e
n
t
z
e
r
e
t
a
l
.
,
“
P
u
b
l
i
c
l
y
a
v
a
i
l
a
b
l
e
c
l
i
n
i
c
a
l
B
E
R
T
e
m
b
e
d
d
i
n
g
s
,
”
a
rX
i
v
p
rep
r
i
n
t
a
rX
i
v
:
1
9
0
4
.
0
3
3
2
3
,
2
0
1
9
.
[
2
3
]
Y
.
Li
u
e
t
a
l
.
,
“
R
o
B
ER
Ta
:
a
r
o
b
u
st
l
y
o
p
t
i
m
i
z
e
d
B
ER
T
p
r
e
t
r
a
i
n
i
n
g
a
p
p
r
o
a
c
h
,
”
a
rXi
v
p
r
e
p
r
i
n
t
a
rX
i
v
:
1
9
0
7
.
1
1
6
9
2
,
2
0
1
9
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
:
/
/
a
r
x
i
v
.
o
r
g
/
a
b
s
/
1
9
0
7
.
1
1
6
9
2
[
2
4
]
J.
M
a
c
k
e
n
z
i
e
,
R
.
B
e
n
h
a
m
,
M
.
P
e
t
r
i
,
J.
R
.
Tr
i
p
p
a
s,
J
.
S
.
C
u
l
p
e
p
p
e
r
,
a
n
d
A
.
M
o
f
f
a
t
,
“
CC
-
n
e
w
s
-
e
n
,
”
i
n
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
2
9
t
h
AC
M
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
I
n
f
o
rm
a
t
i
o
n
&
K
n
o
w
l
e
d
g
e
Ma
n
a
g
e
m
e
n
t
,
O
c
t
.
2
0
2
0
,
p
p
.
3
0
7
7
–
3
0
8
4
.
d
o
i
:
1
0
.
1
1
4
5
/
3
3
4
0
5
3
1
.
3
4
1
2
7
6
2
.
[
2
5
]
A
.
G
o
k
a
s
l
a
n
a
n
d
V
.
C
o
h
e
n
,
“
O
p
e
n
w
e
b
t
e
x
t
c
o
r
p
u
s,
”
G
i
t
H
u
b
,
2
0
1
9
.
h
t
t
p
s:
/
/
sk
y
l
i
o
n
0
0
7
.
g
i
t
h
u
b
.
i
o
/
O
p
e
n
W
e
b
Te
x
t
C
o
r
p
u
s
/
(
a
c
c
e
sse
d
Jan
.
1
6
,
2
0
2
4
)
.
[
2
6
]
I
.
N
e
a
mat
u
l
l
a
h
e
t
a
l
.
,
“
A
u
t
o
ma
t
e
d
d
e
-
i
d
e
n
t
i
f
i
c
a
t
i
o
n
o
f
f
r
e
e
-
t
e
x
t
m
e
d
i
c
a
l
r
e
c
o
r
d
s,
”
BM
C
Me
d
i
c
a
l
I
n
f
o
rm
a
t
i
c
s
a
n
d
D
e
c
i
si
o
n
Ma
k
i
n
g
,
v
o
l
.
8
,
n
o
.
1
,
p
.
3
2
,
D
e
c
.
2
0
0
8
,
d
o
i
:
1
0
.
1
1
8
6
/
1
4
7
2
-
6
9
4
7
-
8
-
3
2
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Ra
v
icha
n
d
r
a
S
r
ira
m
is
p
r
e
se
n
tl
y
a
re
se
a
rc
h
sc
h
o
lar
i
n
th
e
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
t
P
o
n
d
ich
e
r
ry
Un
i
v
e
rsity
,
P
u
d
u
c
h
e
rry
.
He
c
o
m
p
lete
d
h
is
B.
Tec
h
in
In
fo
rm
a
ti
o
n
Tec
h
n
o
l
o
g
y
fr
o
m
S
wa
rn
a
n
d
h
ra
Co
ll
e
g
e
o
f
En
g
i
n
e
e
rin
g
a
n
d
Tec
h
n
o
lo
g
y
,
Na
rsa
p
u
r
in
2
0
0
7
a
n
d
M
.
Tec
h
i
n
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
Tec
h
n
o
lo
g
y
a
t
An
d
h
ra
Un
i
v
e
rsity
,
Visa
k
h
a
p
a
tn
a
m
i
n
2
0
1
0
.
He
wo
rk
e
d
a
s
a
n
a
ss
istan
t
p
ro
fe
ss
o
r
a
t
S
h
ri
Vis
h
n
u
En
g
in
e
e
rin
g
Co
ll
e
g
e
fo
r
Wo
m
e
n
,
B
h
ima
v
a
ra
m
fro
m
2
0
1
1
t
o
2
0
1
9
.
His
re
se
a
rc
h
in
tere
sts
in
c
l
u
d
e
c
o
m
p
u
ter
v
isio
n
,
n
a
tu
ra
l
lan
g
u
a
g
e
p
ro
c
e
ss
in
g
,
a
n
d
d
e
e
p
lea
rn
i
n
g
.
He
h
a
s
a
u
th
o
re
d
a
n
d
c
o
-
a
u
t
h
o
re
d
so
m
e
p
u
b
li
c
a
ti
o
n
s
wh
ic
h
i
n
c
lu
d
e
jo
u
r
n
a
ls
a
n
d
i
n
tern
a
ti
o
n
a
l
c
o
n
fe
re
n
c
e
s
in
t
h
e
fiel
d
o
f
c
o
m
p
u
ter sc
ien
c
e
.
He
c
a
n
b
e
c
o
n
t
a
c
ted
a
t
ra
v
ich
a
n
d
ra
srira
m
@g
m
a
i
l.
c
o
m
.
S
iv
a
S
a
t
h
y
a
S
u
n
d
a
r
a
m
c
o
m
p
lete
d
h
e
r
M
.
Tec
h
a
n
d
P
h
.
D.
i
n
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
rin
g
fr
o
m
P
o
n
d
ich
e
r
ry
Un
iv
e
rsit
y
.
S
h
e
h
a
s
2
5
y
e
a
rs
o
f
tea
c
h
i
n
g
e
x
p
e
rie
n
c
e
a
n
d
sp
e
c
ialize
s
in
e
v
o
lu
ti
o
n
a
r
y
a
n
d
b
i
o
-
in
sp
ired
c
o
m
p
u
ti
n
g
,
sp
a
ti
o
-
tem
p
o
ra
l
d
a
ta
m
in
i
n
g
,
VA
NET
a
n
d
n
a
tu
ra
l
lan
g
u
a
g
e
p
ro
c
e
ss
in
g
.
S
h
e
is
UG
C
NET
q
u
a
li
fied
a
n
d
h
a
s
p
u
b
l
ish
e
d
se
v
e
ra
l
re
se
a
rc
h
a
rti
c
les
in
re
p
u
ted
j
o
u
r
n
a
ls.
S
h
e
is
t
h
e
re
c
ip
ien
t
o
f
th
e
Na
a
ri
S
h
a
k
th
i
Aw
a
rd
2
0
1
7
fro
m
th
e
P
re
sid
e
n
t
o
f
I
n
d
ia
fo
r
h
e
r
m
o
b
il
e
In
n
o
v
a
ti
o
n
“
M
I
TRA
”
f
o
r
W
o
m
e
n
sa
fe
ty
.
S
h
e
h
a
s
a
lso
re
c
e
iv
e
d
th
e
Ch
a
irma
n
’
s Distin
c
ti
o
n
a
wa
rd
in
S
o
u
t
h
As
ia
’
s m
Bil
l
io
n
th
M
o
b
il
e
In
n
o
v
a
ti
o
n
c
o
n
tes
t.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
s
siv
a
sa
th
y
a
@p
o
n
d
iu
n
i.
a
c
.
in
o
r
ss
i
v
a
sa
th
y
a
@g
m
a
il
.
c
o
m
.
S
.
Lo
u
r
d
u
Ma
r
ie
S
o
p
h
ie
is
p
re
se
n
tl
y
a
re
se
a
rc
h
sc
h
o
lar
in
th
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
i
n
e
e
rin
g
a
t
P
o
n
d
ich
e
rry
Un
i
v
e
rsity
,
P
u
d
u
c
h
e
rry
.
S
h
e
c
o
m
p
lete
d
h
e
r
B.
Tec
h
in
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
fr
o
m
M
a
n
a
k
u
la
Vin
a
y
a
g
a
r
In
stit
u
te
o
f
Tec
h
n
o
l
o
g
y
,
P
u
d
u
c
h
e
rry
i
n
2
0
1
5
a
n
d
h
e
r
M
.
Tec
h
.
in
C
o
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
a
t
P
o
n
d
ich
e
rry
Un
iv
e
rsity
,
P
u
d
u
c
h
e
rry
i
n
2
0
1
9
.
S
h
e
wo
rk
e
d
a
s
a
n
A
ss
istan
t
S
y
ste
m
En
g
in
e
e
r
a
t
Tata
Co
n
s
u
lt
a
n
c
y
S
e
rv
ice
,
Ch
e
n
n
a
i
fro
m
2
0
1
5
t
o
2
0
1
7
.
S
h
e
a
l
so
h
a
s
a
y
e
a
r
o
f
tea
c
h
in
g
e
x
p
e
rien
c
e
a
s
a
g
u
e
st
fa
c
u
lt
y
at
P
o
n
d
ich
e
rry
Un
iv
e
rsity
i
n
2
0
1
9
-
2
0
2
0
.
S
h
e
h
a
s
q
u
a
li
fied
f
o
r
th
e
UG
C
NET
e
x
a
m
in
a
ti
o
n
in
2
0
1
9
.
He
r
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
n
a
tu
ra
l
lan
g
u
a
g
e
p
ro
c
e
ss
in
g
,
m
a
c
h
in
e
lea
rn
i
n
g
,
a
n
d
d
e
e
p
lea
rn
in
g
.
S
h
e
h
a
s
a
u
t
h
o
re
d
a
n
d
c
o
-
a
u
th
o
re
d
m
o
re
t
h
a
n
1
0
p
u
b
li
c
a
ti
o
n
s
wh
ich
i
n
c
lu
d
e
jo
u
rn
a
ls
a
n
d
in
ter
n
a
ti
o
n
a
l
c
o
n
fe
re
n
c
e
s
in
t
h
e
fiel
d
o
f
c
o
m
p
u
ter
sc
ien
c
e
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
l
o
u
r
d
u
m
a
ries
o
p
h
ie1
5
@
g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.