I
nte
rna
t
io
na
l J
o
urna
l o
f
I
nfo
rm
a
t
ics a
nd
Co
m
m
un
ica
t
io
n T
ec
hn
o
lo
g
y
(
I
J
-
I
CT
)
Vo
l.
1
5
,
No
.
2
,
J
u
n
e
20
2
6
,
p
p
.
759
~
7
6
8
I
SS
N:
2252
-
8
7
7
6
,
DOI
:
1
0
.
1
1
5
9
1
/iji
ct
.
v
1
5
i
2
.
pp
759
-
7
6
8
759
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ict.
ia
esco
r
e.
co
m
Adv
a
nced
ma
chi
ne learning
for en
ha
nced
a
bdo
mina
l org
a
n
seg
menta
tion
Ro
hin
i P
a
wa
r
1
,
Ro
hin
i J
a
dh
a
v
2
,
Ro
hit
J
a
dh
a
v
3
1
D
e
p
t
a
r
t
m
e
n
t
o
f
C
o
m
p
u
t
e
r
E
n
g
i
n
e
e
r
i
n
g
,
B
h
a
r
a
t
i
V
i
d
y
a
p
e
e
t
h
(
D
e
e
me
d
t
o
b
e
U
n
i
v
e
r
si
t
y
)
C
o
l
l
e
g
e
o
f
En
g
i
n
e
e
r
i
n
g
,
P
u
n
e
,
I
n
d
i
a
2
D
e
p
t
a
r
t
m
e
n
t
o
f
I
n
f
o
r
ma
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
B
h
a
r
a
t
i
V
i
d
y
a
p
e
e
t
h
(
D
e
e
m
e
d
t
o
b
e
U
n
i
v
e
r
s
i
t
y
)
C
o
l
l
e
g
e
o
f
E
n
g
i
n
e
e
r
i
n
g
,
P
u
n
e
,
I
n
d
i
a
3
D
e
p
t
a
r
t
m
e
n
t
o
f
E
N
T
,
B
h
a
r
a
t
i
V
i
d
y
a
p
e
e
t
h
(
D
e
e
me
d
t
o
b
e
U
n
i
v
e
r
s
i
t
y
)
M
e
d
i
c
a
l
C
o
l
l
e
g
e
,
P
u
n
e
,
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
8
,
2
0
2
4
R
ev
is
ed
No
v
2
1
,
2
0
2
5
Acc
ep
ted
Dec
1
1
,
2
0
2
5
Th
is
re
se
a
rc
h
e
v
a
lu
a
tes
th
e
Re
sU
n
e
t
m
o
d
e
l’s
p
e
rfo
rm
a
n
c
e
in
u
si
n
g
c
o
m
p
u
ted
to
m
o
g
ra
p
h
y
(
CT
)
ima
g
e
s
to
se
g
m
e
n
t
v
a
rio
u
s
a
b
d
o
m
i
n
a
l
o
rg
a
n
s.
Wea
k
b
o
u
n
d
a
ries
,
c
o
m
p
u
t
in
g
e
fficie
n
c
y
,
a
n
d
a
n
a
to
m
ica
l
d
i
v
e
rsi
ty
a
re
th
e
c
u
rre
n
t
o
b
sta
c
les
in
a
b
d
o
m
in
a
l
m
u
lt
i
-
o
r
g
a
n
se
g
m
e
n
tatio
n
.
By
m
e
rg
in
g
re
sid
u
a
l
n
e
two
r
k
s
with
U
-
Ne
t,
Re
sU
n
e
t
o
v
e
rc
o
m
e
s
o
b
sta
c
les
b
y
i
n
c
re
a
sin
g
p
re
c
isio
n
a
n
d
e
ffe
c
ti
v
e
n
e
ss
,
w
h
i
c
h
q
u
a
li
fies
i
t
f
o
r
u
se
i
n
m
e
d
i
c
in
e
.
T
h
e
m
o
d
e
l’s
e
ffe
c
ti
v
e
n
e
ss
wa
s
a
ss
e
ss
e
d
o
n
a
n
u
m
b
e
r
o
f
o
r
g
a
n
s,
a
n
d
t
h
e
se
g
m
e
n
tatio
n
a
c
c
u
ra
c
y
wa
s
m
e
a
su
re
d
u
sin
g
t
h
e
d
ice
sim
il
a
rit
y
c
o
e
fficie
n
t
(DSC).
T
h
e
Re
sU
n
e
t
m
o
d
e
l’s
a
b
il
it
y
t
o
p
re
c
ise
ly
se
g
m
e
n
t
o
r
g
a
n
s
wit
h
d
isti
n
c
t
b
o
rd
e
rs
wa
s
p
r
o
v
e
d
b
y
it
s
e
x
c
e
p
ti
o
n
a
l
a
c
c
u
ra
c
y
in
se
g
m
e
n
ti
n
g
imp
o
rtan
t
o
rg
a
n
s,
su
c
h
a
s
th
e
li
v
e
r
(m
e
a
n
DSC:
0
.
8
8
0
)
a
n
d
sp
l
e
e
n
(m
e
a
n
DSC:
0
.
8
3
0
).
Ho
we
v
e
r,
th
e
m
o
d
e
l
stru
g
g
led
t
o
se
p
a
ra
te
th
e
e
so
p
h
a
g
u
s
c
o
rre
c
tl
y
(m
e
a
n
DSC:
0
.
0
0
0
)
a
n
d
stru
g
g
led
wit
h
sm
a
ll
e
r
a
n
d
m
o
re
c
o
m
p
le
x
o
rg
a
n
s
li
k
e
th
e
p
a
n
c
re
a
s
(m
e
a
n
DSC:
0
.
4
2
9
)
a
n
d
g
a
ll
b
lad
d
e
r
(m
e
a
n
DSC:
0
.
1
4
3
)
.
Th
e
se
re
su
lt
s
h
i
g
h
li
g
h
t
th
e
m
e
th
o
d
’s
li
m
it
a
ti
o
n
s
wh
e
n
h
a
n
d
li
n
g
o
rg
a
n
s wit
h
a
sy
m
m
e
tri
c
a
l
sh
a
p
e
s
o
r
h
a
z
y
b
o
r
d
e
rs.
K
ey
w
o
r
d
s
:
Ab
d
o
m
in
al
m
u
lti
-
o
r
g
a
n
s
eg
m
en
tatio
n
C
T
s
ca
n
s
Dice
s
im
ilar
ity
co
ef
f
icien
t
M
ed
ical
im
ag
e
s
eg
m
en
tatio
n
R
esUn
et
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
o
h
in
i Paw
ar
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
E
n
g
i
n
ee
r
in
g
B
h
ar
ati
Vid
y
ap
ee
th
(
Dee
m
ed
to
b
e
Un
iv
e
r
s
ity
)
C
o
lleg
e
o
f
E
n
g
in
ee
r
in
g
Pu
n
e,
I
n
d
ia
E
m
ail:
r
o
h
in
i.p
awa
r
@
b
h
a
r
ativ
id
y
ap
ee
th
.
e
d
u
1.
I
NT
RO
D
UCT
I
O
N
Ov
er
th
e
last
s
ev
er
al
d
ec
ad
es,
th
er
e
h
av
e
m
ajo
r
b
r
ea
k
t
h
r
o
u
g
h
s
with
in
th
e
d
o
m
ain
o
f
m
e
d
ical
im
ag
in
g
,
esp
ec
ially
in
th
e
d
ev
elo
p
m
en
t
o
f
to
o
ls
f
o
r
v
is
u
alizin
g
th
e
h
u
m
an
b
o
d
y
’
s
in
ter
io
r
ar
ch
itectu
r
e.
Am
o
n
g
th
ese
m
eth
o
d
s
,
ab
d
o
m
in
al
im
ag
in
g
is
ess
en
tial
f
o
r
th
e
d
etec
tio
n
an
d
m
a
n
ag
em
en
t
o
f
a
v
ar
iety
o
f
illn
ess
es,
s
u
ch
as
m
alig
n
an
cies,
liv
er
p
r
o
b
lem
s
,
an
d
g
astro
in
test
in
al
is
s
u
es.
Ho
wev
er
,
m
an
u
ally
s
ep
ar
atin
g
ab
d
o
m
in
al
o
r
g
an
s
f
r
o
m
m
e
d
i
ca
l
im
ag
es
-
lik
e
co
m
p
u
ted
to
m
o
g
r
ap
h
y
(
C
T
)
o
r
m
ag
n
etic
r
eso
n
an
ce
im
ag
in
g
(
MRI)
s
ca
n
s
-
r
em
ain
s
a
tim
e
-
co
n
s
u
m
in
g
an
d
lab
o
u
r
-
in
te
n
s
iv
e
p
r
o
ce
s
s
.
B
ec
au
s
e
r
ad
io
lo
g
is
ts
’
r
an
g
e
in
th
eir
s
k
ill
lev
els
an
d
in
ter
p
r
etatio
n
s
,
th
is
tech
n
iq
u
e
is
lik
ewise
p
r
o
n
e
to
f
lu
ctu
atio
n
.
An
i
n
cr
ea
s
in
g
n
u
m
b
er
o
f
p
eo
p
le
a
r
e
in
ter
ested
i
n
u
s
in
g
m
ac
h
in
e
lear
n
in
g
(
ML
)
a
p
p
r
o
ac
h
es,
esp
ec
ially
d
ee
p
lear
n
in
g
(
DL
)
,
to
a
u
to
m
ate
an
d
in
cr
ea
s
e
th
e
ac
c
u
r
ac
y
o
f
a
b
d
o
m
in
al
s
eg
m
e
n
tatio
n
as a
s
o
lu
tio
n
to
th
ese
p
r
o
b
lem
s
[
1
]
.
T
h
e
ar
ea
o
f
m
ed
ical
im
a
g
in
g
h
as
s
ee
n
a
r
ev
o
l
u
tio
n
ar
y
ch
an
g
e
b
ec
a
u
s
e
to
ML
,
p
a
r
ticu
lar
ly
DL
,
wh
ich
p
r
o
v
id
es
r
o
b
u
s
t
to
o
ls
f
o
r
au
to
n
o
m
o
u
s
im
ag
e
p
r
o
ce
s
s
in
g
.
C
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
et
wo
r
k
s
(
C
NNs)
h
av
e
s
h
o
wn
esp
ec
ially
g
o
o
d
r
esu
lts
in
task
s
in
clu
d
in
g
s
eg
m
en
tatio
n
,
o
b
ject
d
etec
tio
n
,
an
d
p
ictu
r
e
ca
teg
o
r
izatio
n
.
I
n
th
e
co
n
tex
t
o
f
ab
d
o
m
in
al
i
m
ag
in
g
,
th
ese
m
o
d
els
ca
n
l
ea
r
n
to
id
en
t
if
y
an
d
d
elin
ea
te
m
u
ltip
le
o
r
g
an
s
s
im
u
ltan
eo
u
s
ly
,
with
a
lev
el
o
f
p
r
ec
is
io
n
th
at
o
f
ten
s
u
r
p
ass
es
tr
ad
itio
n
al
m
eth
o
d
s
.
Fo
r
in
s
tan
ce
,
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
5
,
No
.
2
,
J
u
n
e
20
2
6
:
7
5
9
-
7
6
8
760
d
ev
elo
p
m
e
n
t
o
f
f
u
lly
a
u
to
m
a
ted
s
eg
m
en
tatio
n
tech
n
iq
u
es,
s
u
ch
as
th
e
AL
AM
O
m
o
d
el
,
h
as
d
em
o
n
s
tr
ated
h
ig
h
ac
c
u
r
ac
y
in
s
eg
m
e
n
tin
g
m
u
ltip
le
o
r
g
an
s
at
r
is
k
(
O
AR
s
)
in
ab
d
o
m
in
al
MRI
s
ca
n
s
,
ac
h
iev
in
g
d
ic
e
s
im
ilar
ity
co
ef
f
icien
ts
(
DSC
s
)
as
h
ig
h
as
0
.
9
6
f
o
r
ce
r
tain
o
r
g
an
s
.
Similar
ly
,
ap
p
r
o
ac
h
es
lik
e
d
ee
p
m
u
lti
-
p
lan
ar
co
-
tr
ain
in
g
(
DM
PC
T
)
h
av
e
b
ee
n
p
r
o
p
o
s
ed
to
lev
er
ag
e
co
n
s
en
s
u
s
in
f
o
r
m
a
tio
n
f
r
o
m
m
u
l
tip
le
p
lan
es
in
C
T
s
ca
n
s
,
f
u
r
th
er
e
n
h
an
cin
g
th
e
r
e
liab
ilit
y
o
f
s
eg
m
en
tatio
n
o
u
tco
m
es [
2
]
.
T
h
e
ad
o
p
tio
n
o
f
th
ese
ML
tech
n
iq
u
es
in
ab
d
o
m
in
al
im
ag
i
n
g
is
d
r
iv
en
b
y
th
eir
ab
ilit
y
to
p
r
o
ce
s
s
lar
g
e
d
atasets
,
lear
n
co
m
p
le
x
p
atter
n
s
,
an
d
g
en
e
r
alize
wel
l
to
n
ew
d
ata.
T
h
is
ca
p
ab
i
lity
is
p
ar
ticu
lar
ly
im
p
o
r
tan
t g
i
v
en
th
e
h
ig
h
a
n
ato
m
ical
v
ar
iab
ilit
y
an
d
th
e
p
r
ese
n
ce
o
f
wea
k
b
o
u
n
d
ar
ies b
etwe
en
ad
jace
n
t o
r
g
an
s
in
ab
d
o
m
in
al
im
ag
es.
Fo
r
ex
am
p
le,
th
e
u
s
e
o
f
d
e
n
s
e
V
-
n
etw
o
r
k
s
h
as
s
h
o
w
n
p
r
o
m
is
e
in
im
p
r
o
v
in
g
s
eg
m
en
ta
tio
n
ac
cu
r
ac
y
with
o
u
t th
e
n
ee
d
f
o
r
in
ter
-
s
u
b
ject
im
ag
e
r
eg
is
tr
atio
n
,
a
s
tep
th
at
is
o
f
ten
ch
allen
g
in
g
in
ab
d
o
m
in
al
im
a
g
es d
u
e
to
v
ar
ia
b
ilit
y
in
an
ato
m
y
[
3
]
.
Pre
cise
m
u
lti
-
o
r
g
an
d
iv
is
io
n
i
s
cr
itical
f
o
r
v
ar
io
u
s
clin
ical
a
p
p
licatio
n
s
,
in
clu
d
in
g
s
u
r
g
ical
p
lan
n
in
g
,
d
is
ea
s
e
d
iag
n
o
s
is
,
an
d
r
ad
iatio
n
th
er
ap
y
.
T
h
e
ab
ilit
y
to
p
r
ec
is
ely
d
elin
ea
te
m
u
ltip
le
o
r
g
an
s
with
in
th
e
ab
d
o
m
e
n
en
a
b
les
clin
ician
s
to
b
etter
ass
ess
th
e
ex
ten
t
o
f
d
is
ea
s
e,
p
lan
a
p
p
r
o
p
r
iate
in
ter
v
e
n
tio
n
s
,
an
d
m
o
n
ito
r
tr
ea
tm
en
t
o
u
tco
m
es.
I
n
r
a
d
iat
io
n
th
er
ap
y
,
f
o
r
e
x
am
p
le,
p
r
e
cise
o
r
g
an
s
eg
m
en
tatio
n
is
at
r
is
k
is
es
s
en
tial
to
m
in
im
ize
d
am
ag
e
t
o
h
ea
lth
y
tis
s
u
es wh
ile
d
eliv
er
in
g
th
e
m
a
x
im
u
m
th
er
a
p
eu
tic
d
o
s
e
to
th
e
tar
g
et
ar
ea
[
4
]
.
Mo
r
eo
v
er
,
au
to
m
ate
d
m
u
lti
-
o
r
g
an
s
eg
m
e
n
tatio
n
f
ac
ilit
at
es
th
e
an
aly
s
is
o
f
lar
g
e
-
s
ca
le
clin
ical
d
atasets
,
p
av
in
g
th
e
way
f
o
r
m
o
r
e
p
er
s
o
n
alize
d
an
d
d
ata
-
d
r
iv
en
ap
p
r
o
ac
h
es
to
h
ea
lth
ca
r
e.
T
ec
h
n
iq
u
es
s
u
ch
as
atlas
-
b
ased
m
eth
o
d
s
,
wh
er
e
s
eg
m
en
tatio
n
is
g
u
id
ed
b
y
p
r
e
-
lab
eled
an
ato
m
ical
tem
p
lates,
an
d
m
o
r
e
r
ec
en
t
tech
n
iq
u
es
o
f
DL
h
a
v
e
d
em
o
n
s
tr
ated
s
ig
n
if
ican
t
im
p
r
o
v
em
en
ts
in
h
an
d
lin
g
th
e
c
o
m
p
lex
ities
o
f
ab
d
o
m
in
al
an
ato
m
y
.
Fo
r
in
s
tan
ce
,
th
e
c
o
m
b
in
atio
n
o
f
atlas
-
b
ased
s
eg
m
en
tatio
n
with
co
n
tex
t
lear
n
i
n
g
h
as
b
ee
n
u
s
ed
to
en
h
an
ce
th
e
ac
cu
r
ac
y
o
f
m
u
l
ti
-
o
r
g
an
s
eg
m
en
tatio
n
o
n
clin
ically
ac
q
u
ir
ed
C
T
s
ca
n
s
,
d
em
o
n
s
tr
atin
g
r
o
b
u
s
t
p
er
f
o
r
m
an
ce
ac
r
o
s
s
d
if
f
er
en
t
p
atien
t p
o
p
u
latio
n
s
[
5
]
.
C
o
n
ze
et
a
l.
[
6
]
ex
p
lo
r
ed
a
ca
s
ca
d
ed
DL
m
o
d
el
with
ad
v
er
s
ar
ial
n
etwo
r
k
s
f
o
r
C
T
an
d
MRI
s
eg
m
en
tatio
n
.
T
h
is
m
eth
o
d
o
u
tp
er
f
o
r
m
ed
tr
a
d
itio
n
al
en
c
o
d
er
-
d
ec
o
d
er
ar
ch
itectu
r
es
f
o
r
t
h
e
s
p
leen
,
k
id
n
ey
s
,
an
d
liv
er
.
W
o
lz
et
a
l.
[
7
]
p
r
o
p
o
s
ed
a
weig
h
tin
g
s
ch
e
m
e
an
d
atlas
r
eg
is
tr
atio
n
f
o
r
f
u
lly
a
u
to
m
ated
s
eg
m
en
tatio
n
.
T
h
is
m
eth
o
d
d
em
o
n
s
tr
ated
co
m
p
etitiv
e
r
e
s
u
lts
f
o
r
liv
er
,
k
id
n
ey
s
,
p
an
cr
ea
s
,
an
d
s
p
leen
.
L
ee
et
a
l.
[
8
]
d
ev
elo
p
e
d
Sia
m
ese
lear
n
in
g
to
im
p
r
o
v
e
s
e
g
m
en
tatio
n
b
y
ca
p
tu
r
in
g
b
o
t
h
g
lo
b
al
an
d
lo
ca
l
co
n
tex
ts
.
T
h
is
m
eth
o
d
ac
h
iev
e
d
a
2
% im
p
r
o
v
em
e
n
t in
d
ice
s
co
r
e
co
ef
f
icien
ts
co
m
p
ar
e
d
to
ex
is
tin
g
m
eth
o
d
s
.
A
m
u
lti
-
s
eq
u
en
ce
DL
m
eth
o
d
was
p
r
esen
ted
b
y
Am
ja
d
et
a
l.
[
9
]
f
o
r
au
to
m
atic
ab
d
o
m
in
al
s
eg
m
en
tatio
n
in
MRIs
u
s
ed
t
o
s
ch
ed
u
le
r
ad
iatio
n
th
er
a
p
y
.
T
o
im
p
r
o
v
e
s
eg
m
e
n
tatio
n
ac
c
u
r
ac
y
,
th
er
e
m
o
d
el
m
ak
es u
s
e
o
f
d
ata
f
r
o
m
s
ev
er
a
l M
R
I
s
eq
u
en
ce
s
.
T
1
an
d
T
2
weig
h
ted
MRI
s
eq
u
en
ce
s
ar
e
u
s
ed
to
test
an
d
tr
ain
th
e
m
o
d
el,
wh
ich
wo
r
k
s
o
n
a
3
DR
esUn
et
n
etwo
r
k
.
T
h
e
r
esu
lts
o
f
th
e
in
v
esti
g
atio
n
s
h
o
wed
th
at
th
e
m
o
d
el
co
u
ld
ac
cu
r
ately
a
n
d
ef
f
ec
ti
v
ely
cr
ea
te
o
u
tlin
es
f
o
r
1
2
u
p
p
er
ab
d
o
m
in
al
o
r
g
an
s
,
b
y
th
e
DSC
o
f
0
.
8
7
.
T
h
e
s
tu
d
y
em
p
h
asizes
h
o
w
cr
u
cial
m
u
lti
-
s
eq
u
en
ce
MRI
is
f
o
r
en
h
an
ci
n
g
th
e
p
r
ec
is
io
n
o
f
o
r
g
an
s
eg
m
en
tatio
n
in
clin
ical
p
r
ac
tice,
esp
ec
iall
y
wh
en
it
co
m
es
to
tu
m
o
r
s
o
f
th
e
ab
d
o
m
e
n
.
T
o
ac
co
m
p
lis
h
s
eg
m
en
tatio
n
o
f
im
ag
es.
Kak
ey
a
et
a
l.
[
1
0
]
p
r
esen
ted
a
DL
s
ch
em
e,
3
D
U
-
J
APA
-
Net,
wh
ich
in
teg
r
ates
t
r
an
s
f
er
lear
n
i
n
g
with
a
p
r
o
b
ab
ilis
tic
atlas
o
f
o
r
g
an
s
f
o
r
C
T
-
b
ased
a
b
d
o
m
in
al
m
u
lti
-
o
r
g
an
s
eg
m
en
tatio
n
.
T
h
is
m
eth
o
d
im
p
r
o
v
ed
u
p
o
n
tr
ad
itio
n
al
3
D
U
-
n
et
b
y
ad
d
r
ess
in
g
er
r
o
r
s
ass
o
ciate
d
with
lo
ca
l
v
o
l
u
m
etr
ic
d
ata,
th
er
eb
y
en
h
an
cin
g
s
eg
m
en
tatio
n
ac
cu
r
ac
y
.
T
h
e
s
tu
d
y
ac
h
iev
ed
h
ig
h
er
d
ice
s
co
r
es
co
m
p
ar
ed
to
co
n
v
en
tio
n
al
2
D
an
d
3
D
U
-
n
ets,
u
n
d
er
s
co
r
i
n
g
th
e
ef
f
ec
tiv
e
n
e
s
s
o
f
in
co
r
p
o
r
atin
g
p
r
o
b
a
b
i
lis
tic
at
lases
in
to
DL
f
r
am
ewo
r
k
s
f
o
r
o
r
g
an
s
eg
m
en
tatio
n
.
W
an
g
et
a
l.
[
1
1
]
d
ev
el
o
p
ed
th
e
cr
o
s
s
-
co
n
v
o
lu
tio
n
al
tr
an
s
f
o
r
m
er
(
C
2
Fo
r
m
er
)
n
etwo
r
k
to
en
h
an
ce
th
e
g
e
n
er
aliza
tio
n
an
d
ac
cu
r
ac
y
o
f
o
r
g
an
s
eg
m
e
n
t
atio
n
ac
r
o
s
s
v
ar
io
u
s
m
ed
ical
i
m
ag
in
g
m
o
d
alities
.
T
h
e
s
tu
d
y
ap
p
lied
th
is
m
eth
o
d
to
C
T
im
ag
es
o
f
ab
d
o
m
in
al
o
r
g
an
s
,
MRI
o
f
ca
r
d
iac
s
tr
u
ctu
r
es,
an
d
s
k
in
ca
n
ce
r
im
ag
es,
ac
h
iev
in
g
h
i
g
h
p
er
f
o
r
m
an
ce
in
all
d
atasets
.
T
h
e
C
2
Fo
r
m
er
ef
f
ec
tiv
el
y
in
teg
r
at
es
lo
ca
l
an
d
g
lo
b
al
co
n
tex
ts
,
m
ak
in
g
it
a
r
o
b
u
s
t
s
o
lu
tio
n
f
o
r
d
iv
er
s
e
m
ed
ical
i
m
ag
in
g
c
h
allen
g
es.
I
n
o
r
d
er
t
o
en
h
a
n
ce
ab
d
o
m
in
al
s
eg
m
en
tatio
n
f
r
o
m
C
T
im
ag
e
s
.
I
r
s
h
ad
et
a
l.
[
1
2
]
s
u
g
g
ested
a
tech
n
iq
u
e
wh
ich
u
tili
ze
s
p
r
ed
ictio
n
o
f
o
r
g
a
n
-
b
o
u
n
d
ar
y
as
a
s
u
p
p
lem
e
n
tal
j
o
b
.
T
h
e
tech
n
i
q
u
e
d
em
o
n
s
tr
at
ed
th
e
b
en
ef
it
o
f
m
u
lti
-
task
lear
n
in
g
i
n
m
an
a
g
in
g
co
m
p
licated
ab
d
o
m
in
al
a
n
ato
m
y
b
y
ex
h
ib
itin
g
n
o
ta
b
le
g
ain
s
.
L
ian
g
et
a
l.
[
1
3
]
d
e
v
elo
p
e
d
a
h
y
b
r
id
m
o
d
el
(
HDM
)
c
o
m
b
in
ed
with
f
u
s
io
n
n
etwo
r
k
t
o
en
h
an
ce
s
eg
m
en
tatio
n
ac
cu
r
ac
y
o
f
ab
d
o
m
in
al
C
T
s
ca
n
s
.
T
h
e
HD
M
ad
d
r
ess
es
th
e
lim
itatio
n
s
o
f
r
ig
id
o
r
af
f
in
e
tr
an
s
f
o
r
m
atio
n
s
b
y
in
co
r
p
o
r
a
tin
g
b
o
th
in
ter
-
p
atien
t
an
d
in
tr
a
-
p
atien
t
d
ef
o
r
m
atio
n
s
,
th
u
s
ca
p
tu
r
in
g
s
u
b
tl
e
o
r
g
an
d
e
f
o
r
m
atio
n
s
m
o
r
e
ef
f
e
ctiv
ely
.
T
h
e
s
tr
ateg
y
b
ea
t
cu
r
r
en
t
cu
ttin
g
-
ed
g
e
m
eth
o
d
o
lo
g
ies
,
u
s
in
g
0
.
8
5
2
as
th
e
av
er
ag
e
DSC
ac
r
o
s
s
v
ar
io
u
s
o
r
g
an
s
.
Hu
et
a
l.
[
1
4
]
cr
ea
ted
a
co
m
p
letely
au
to
m
ated
m
u
lti
-
o
r
g
a
n
s
eg
m
en
tatio
n
ap
p
r
o
ac
h
t
h
at
c
o
m
b
in
es
d
ee
p
C
NNs
with
tim
e
-
im
p
licit
lev
el
s
ets.
T
h
is
ap
p
r
o
ac
h
was
h
ig
h
ly
ac
cu
r
ate,
with
d
ice
o
v
er
lap
r
atio
s
o
f
9
6
.
0
%,
9
4
.
2
%,
an
d
9
5
.
4
%
f
o
r
s
p
leen
,
k
id
n
ey
s
an
d
liv
er
r
esp
ec
tiv
ely
,
p
r
o
v
in
g
its
r
esil
ien
ce
an
d
e
f
f
i
cien
cy
in
clin
ical
s
itu
atio
n
s
.
W
an
g
et
a
l.
[
1
5
]
s
tu
d
ie
d
o
r
g
an
s
eg
m
en
tatio
n
o
f
ab
d
o
m
in
al
C
T
im
ag
es
b
ased
o
n
o
r
g
an
n
etwo
r
k
s
with
r
ev
er
s
e
co
n
n
ec
tio
n
s
.
T
h
is
m
eth
o
d
en
h
a
n
ce
s
o
r
g
an
d
is
cr
im
in
atio
n
b
y
f
o
cu
s
in
g
o
n
lo
ca
l
b
ac
k
g
r
o
u
n
d
r
e
d
u
ctio
n
an
d
s
tr
u
ctu
r
al
s
im
ilar
ity
,
w
h
ich
ar
e
cr
itical
in
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
A
d
va
n
ce
d
ma
ch
in
e
le
a
r
n
in
g
fo
r
en
h
a
n
ce
d
a
b
d
o
min
a
l
o
r
g
a
n
s
eg
men
ta
tio
n
(
R
o
h
in
i P
a
w
a
r
)
761
co
m
p
lex
an
at
o
m
ical
r
eg
io
n
s
.
T
h
e
ap
p
r
o
ac
h
o
u
tp
er
f
o
r
m
e
d
tr
ad
itio
n
al
2
D
an
d
3
D
p
atch
-
b
ased
m
eth
o
d
s
with
r
eg
ar
d
t
o
m
ea
n
s
u
r
f
ac
e
d
is
tan
ce
s
an
d
DSC
s
,
estab
lis
h
in
g
it
s
elf
as
a
lead
in
g
m
eth
o
d
in
a
cc
u
r
ate
ab
d
o
m
in
al
o
r
g
an
s
eg
m
en
tatio
n
.
I
n
o
r
d
er
t
o
en
h
an
ce
s
eg
m
en
tatio
n
,
T
o
n
g
et
a
l.
[
1
6
]
p
r
esen
ted
th
e
De
n
s
eNe
t
,
wh
ich
u
s
es
a
d
en
s
e
co
n
n
ec
tio
n
b
lo
ck
.
Sel
f
-
p
ac
ed
De
n
s
eNe
t
attain
ed
d
ice
co
ef
f
icien
t
o
f
8
4
.
4
6
%
ac
r
o
s
s
eig
h
t
o
r
g
an
s
,
s
u
r
p
ass
in
g
p
r
ev
io
u
s
m
eth
o
d
s
,
esp
ec
ially
f
o
r
c
h
allen
g
in
g
s
tr
u
ctu
r
es
lik
e
th
e
d
u
o
d
en
u
m
an
d
g
allb
lad
d
er
.
T
an
g
et
a
l.
[
1
7
]
d
ev
elo
p
ed
a
r
an
d
o
m
n
etwo
r
k
tech
n
i
q
u
e
f
o
r
r
eso
lu
tio
n
o
f
3
D
ab
d
o
m
in
al.
T
h
is
m
eth
o
d
ad
d
r
ess
es
th
e
m
em
o
r
y
lim
ita
tio
n
s
o
f
GPU
-
b
ased
n
etwo
r
k
s
b
y
u
s
in
g
a
p
atch
-
b
ased
ap
p
r
o
ac
h
.
T
h
e
g
iv
e
n
tech
n
iq
u
e
s
ig
n
if
ican
tly
im
p
r
o
v
es
ac
cu
r
ac
y
o
f
s
eg
m
en
tatio
n
,
ac
h
ie
v
in
g
d
ice
c
o
ef
f
icien
t
o
f
0
.
8
5
6
,
o
u
tp
er
f
o
r
m
in
g
co
n
v
en
tio
n
al
c
o
ar
s
e
b
aselin
e
tech
n
iq
u
e
.
Ab
d
o
m
in
al
m
u
lti
-
o
r
g
a
n
s
eg
m
en
tatio
n
h
as
s
ee
n
r
ec
en
t
p
r
o
g
r
ess
u
s
in
g
U
-
Net,
d
en
s
e
V
-
n
etwo
r
k
s
,
atlas
-
b
ased
m
eth
o
d
s
,
an
d
D
MPC
T
.
U
-
Net
an
d
d
en
s
e
V
-
n
etwo
r
k
s
ar
e
ab
le
to
s
eg
m
en
t
lar
g
e,
well
-
d
ef
in
ed
o
r
g
an
s
b
u
t
s
t
r
u
g
g
le
to
s
eg
m
e
n
t
s
m
aller
s
tr
u
ctu
r
es
with
we
ak
b
o
u
n
d
a
r
ies.
Me
th
o
d
s
th
at
r
ely
o
n
atlas
-
b
ased
ap
p
r
o
ac
h
es
d
o
well
wh
er
e
th
er
e
is
co
n
s
is
ten
cy
o
f
th
e
an
at
o
m
ical
s
h
ap
es,
b
u
t
ar
e
ch
alle
n
g
ed
b
y
v
ar
iab
ilit
y
.
Alth
o
u
g
h
DM
PC
T
r
eq
u
ir
es
h
ig
h
c
o
m
p
u
tatio
n
al
r
eso
u
r
ce
s
,
u
s
i
n
g
DM
PC
T
with
lim
ited
an
n
o
tatio
n
ca
n
s
ig
n
if
ican
tly
in
cr
ea
s
e
th
e
ac
cu
r
ac
y
o
f
th
e
d
etec
to
r
.
H
o
wev
er
,
th
ese
tech
n
iq
u
es
s
u
f
f
er
f
r
o
m
c
o
m
m
o
n
lim
itatio
n
s
;
th
ey
ca
n
o
p
er
ate
o
n
s
m
aller
o
r
g
an
s
,
h
an
d
le
wea
k
in
ter
-
o
r
g
an
b
o
u
n
d
ar
ies,
an
d
b
e
co
m
p
u
tatio
n
ally
ef
f
icien
t,
b
u
t t
h
er
e
is
a
n
ee
d
f
o
r
m
o
r
e
i
n
n
o
v
atio
n
[
1
8
]
.
T
h
e
cu
r
r
e
n
t
ch
allen
g
es
in
a
b
d
o
m
in
al
m
u
lti
-
o
r
g
a
n
s
eg
m
e
n
tatio
n
in
clu
d
e
m
a
n
ag
in
g
a
n
ato
m
ical
v
ar
iab
ilit
y
,
s
eg
m
en
tin
g
o
r
g
an
s
with
wea
k
b
o
u
n
d
a
r
ies,
an
d
e
n
s
u
r
in
g
co
m
p
u
tatio
n
al
ef
f
icien
cy
.
W
h
ile
ex
is
tin
g
m
o
d
els
h
av
e
m
ad
e
p
r
o
g
r
ess
,
g
ap
s
r
em
ain
,
p
ar
ticu
lar
ly
i
n
h
an
d
lin
g
c
o
m
p
lex
an
at
o
m
y
an
d
m
ain
tain
in
g
ef
f
icien
cy
.
T
h
e
R
esUn
et
m
o
d
el
ad
d
r
ess
es
th
ese
is
s
u
es
b
y
c
o
m
b
in
in
g
th
e
DL
s
tr
en
g
th
s
o
f
r
esid
u
al
n
etwo
r
k
s
with
U
-
Net’
s
ef
f
ec
tiv
e
lo
ca
lizatio
n
.
T
h
is
ar
ch
itectu
r
e
im
p
r
o
v
es
th
e
m
o
d
el’
s
ab
ilit
y
to
h
an
d
le
an
ato
m
ical
d
iv
er
s
ity
,
ac
cu
r
ately
s
eg
m
en
t
o
r
g
an
s
with
wea
k
b
o
u
n
d
ar
i
es,
an
d
r
em
ain
co
m
p
u
tatio
n
ally
ef
f
icien
t
;
th
is
r
en
d
er
s
it id
ea
l f
o
r
u
s
e
in
m
ed
i
ca
l c
lin
ics [
1
9
]
.
2.
M
E
T
H
O
D
2
.
1
.
Da
t
a
s
et
s
Data
s
ets
ar
e
ess
en
tial
f
o
r
th
e
ad
v
an
ce
m
e
n
t
an
d
ass
ess
m
en
t
o
f
ML
m
o
d
els,
esp
ec
ially
i
n
t
h
e
f
ield
o
f
m
ed
ical
im
ag
in
g
.
T
h
is
s
tu
d
y
ex
clu
s
iv
ely
u
tili
ze
d
s
tan
d
ar
d
,
p
u
b
licly
av
ailab
le
d
atasets
,
wid
ely
r
ec
o
g
n
ize
d
in
th
e
d
o
m
ain
o
f
m
u
lti
-
o
r
g
an
s
e
g
m
en
tatio
n
.
T
h
e
d
ata
s
ets
in
clu
d
e
s
tan
d
ar
d
ize
d
an
d
a
n
n
o
tat
ed
p
h
o
t
o
s
,
en
ab
lin
g
r
esear
ch
er
s
to
ev
alu
ate
an
d
v
alid
ate
th
eir
alg
o
r
ith
m
s
u
s
in
g
a
s
h
ar
ed
r
ef
er
e
n
ce
.
T
h
e
s
u
b
s
eq
u
en
t
d
atasets
ar
e
o
f
ten
em
p
lo
y
ed
in
th
e
d
o
m
ain
an
d
ar
e
cr
u
cial
f
o
r
ev
alu
atin
g
an
d
v
er
if
y
in
g
s
eg
m
en
tatio
n
m
o
d
e
ls
.
T
h
e
b
ey
o
n
d
th
e
cr
an
ial
v
au
lt
(
B
T
C
V)
d
at
aset
in
clu
d
es
3
0
f
u
lly
an
n
o
tated
C
T
s
ca
n
s
o
f
m
an
y
o
r
g
an
s
,
s
u
ch
as
th
e
liv
er
,
k
id
n
ey
s
,
p
an
c
r
ea
s
,
an
d
s
p
leen
.
T
h
e
co
m
p
lete
co
v
er
ag
e
o
f
ab
d
o
m
in
al
o
r
g
an
s
in
th
is
s
tu
d
y
is
g
en
er
ally
ac
k
n
o
wled
g
e
d
,
m
ak
in
g
it
a
ty
p
ical
r
e
f
er
en
ce
p
o
in
t
in
s
eg
m
en
tatio
n
r
esear
ch
[
2
0
]
.
T
h
e
co
m
b
in
ed
h
ea
lth
y
ab
d
o
m
in
al
o
r
g
an
s
eg
m
en
tatio
n
(
C
HAOS)
d
ataset
co
n
tai
n
s
4
0
a
n
n
o
tated
C
T
v
o
lu
m
es,
c
o
m
p
lem
en
ted
b
y
MRI
s
ca
n
s
f
o
r
m
u
lti
-
m
o
d
ality
s
eg
m
en
tatio
n
h
av
e
b
ee
n
a
n
n
o
tate
d
to
id
en
tify
o
r
g
an
s
s
u
ch
as
th
e
li
v
er
,
s
p
leen
,
a
n
d
k
id
n
ey
s
.
T
h
is
d
ataset
is
h
ig
h
ly
r
elev
an
t
f
o
r
in
v
esti
g
atio
n
s
in
v
o
lv
in
g
th
e
s
eg
m
en
tatio
n
o
f
m
an
y
m
o
d
alities
.
I
t
h
as
b
ee
n
wid
ely
u
tili
ze
d
in
ch
allen
g
es
an
d
co
n
test
s
to
ass
ess
th
e
p
er
f
o
r
m
an
ce
o
f
m
o
d
el
s
[
2
1
]
.
I
n
ad
d
itio
n
,
th
e
liv
er
tu
m
o
u
r
s
eg
m
en
tatio
n
(
L
iTS)
d
ataset
in
clu
d
es 1
3
0
a
n
n
o
tated
C
T
v
o
lu
m
es,
p
r
im
ar
i
ly
f
o
cu
s
in
g
o
n
liv
er
an
d
tu
m
o
r
s
eg
m
en
tatio
n
,
b
u
t
with
ad
d
itio
n
al
o
r
g
an
a
n
n
o
tat
io
n
s
p
r
im
ar
ily
f
o
cu
s
s
es
o
n
s
eg
m
en
tin
g
t
h
e
liv
er
an
d
tu
m
o
u
r
s
,
b
u
t
it
also
p
r
o
v
id
es
a
s
u
b
s
tan
tial
n
u
m
b
er
o
f
C
T
s
ca
n
s
th
at
h
av
e
b
ee
n
an
n
o
tated
.
T
h
e
d
ataset
is
o
f
g
r
ea
t
s
ig
n
if
ic
an
ce
f
o
r
tr
ain
in
g
m
o
d
els
th
at
u
s
e
liv
er
s
eg
m
en
tatio
n
as
a
co
m
p
o
n
e
n
t
o
f
a
m
u
lti
-
o
r
g
an
s
tr
ateg
y
[
2
2
]
.
T
h
e
to
tal
d
ataset
s
ize
f
o
r
t
h
is
s
tu
d
y
was
2
0
0
a
n
n
o
tated
C
T
v
o
lu
m
es.
T
h
ese
wer
e
d
i
v
id
ed
in
to
tr
ain
in
g
,
v
alid
atio
n
,
an
d
test
in
g
s
u
b
s
ets
u
s
in
g
a
7
0
:2
0
:1
0
s
p
lit
to
id
en
tify
s
ev
er
al
ab
d
o
m
in
al
o
r
g
an
s
,
s
u
ch
as
th
e
s
p
leen
,
k
id
n
e
y
s
,
g
allb
la
d
d
er
,
liv
er
,
s
to
m
ac
h
,
an
d
o
t
h
er
s
as
s
h
o
wn
in
T
ab
le
1
.
T
h
e
s
p
atial
r
eso
lu
tio
n
o
f
th
e
C
T
s
ca
n
s
v
ar
ies
ac
r
o
s
s
d
at
asets
b
u
t
was
s
tan
d
ar
d
ized
d
u
r
in
g
p
r
e
p
r
o
ce
s
s
in
g
t
o
en
s
u
r
e
u
n
if
o
r
m
in
p
u
t
d
im
en
s
io
n
s
.
Key
f
ea
tu
r
es
ex
tr
ac
ted
f
r
o
m
th
e
im
ag
es
in
clu
d
e
p
ix
el
in
ten
s
ities
n
o
r
m
alize
d
to
th
e
r
an
g
e
[
0
,
1
]
an
d
an
ato
m
ical
s
tr
u
ctu
r
e
b
o
u
n
d
ar
ies
r
ep
r
esen
ted
in
th
e
s
e
g
m
en
tatio
n
m
ask
s
.
T
h
e
im
ag
i
n
g
tech
n
iq
u
e
o
f
f
e
r
s
f
u
ll
v
is
u
aliza
tio
n
s
o
f
ea
c
h
o
r
g
an
,
i
n
clu
d
in
g
d
etailed
ax
ial
v
iews.
T
h
e
d
ataset’
s
co
m
p
r
e
h
en
s
iv
e
an
n
o
tatio
n
s
m
a
k
e
i
t
we
l
l
-
s
u
i
t
e
d
f
o
r
t
r
ai
n
i
n
g
a
n
d
a
s
s
ess
i
n
g
M
L
m
o
d
e
ls
th
a
t
s
p
e
ci
a
l
i
ze
i
n
m
u
l
ti
-
o
r
g
a
n
s
e
g
m
e
n
t
a
ti
o
n
.
T
h
is
g
u
a
r
a
n
t
e
e
s
p
r
e
c
is
i
o
n
a
n
d
d
e
p
e
n
d
a
b
i
l
i
t
y
i
n
a
cc
u
r
a
t
e
l
y
d
e
li
n
e
a
tin
g
o
r
g
a
n
s
a
c
r
o
s
s
s
e
v
e
r
a
l
a
n
at
o
m
i
c
a
l
p
l
a
n
es
[
2
3
]
.
2
.
2
.
M
o
del dev
elo
pm
ent
W
e
d
ev
elo
p
th
e
R
esUn
et
m
o
d
el
f
o
r
ab
d
o
m
in
al
o
r
g
a
n
s
e
g
m
en
tatio
n
in
a
s
y
s
tem
atic
f
ash
io
n
b
y
in
co
r
p
o
r
atin
g
n
o
v
el
p
r
e
p
r
o
c
ess
in
g
,
tr
ain
in
g
,
an
d
ev
alu
a
tio
n
s
tr
ateg
ies.
Per
f
o
r
m
an
ce
o
f
t
h
e
m
o
d
el
is
b
en
ch
m
ar
k
ed
u
s
in
g
ad
v
an
ce
d
m
etr
ics f
o
r
r
o
b
u
s
tn
ess
an
d
ac
c
u
r
ac
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
5
,
No
.
2
,
J
u
n
e
20
2
6
:
7
5
9
-
7
6
8
762
T
ab
le
1
.
Data
s
et
co
m
p
r
is
in
g
C
T
im
ag
es u
s
ed
f
o
r
s
ev
er
al
ab
d
o
m
in
al
o
r
g
an
s
N
a
me
o
f
t
h
e
o
r
g
a
n
A
x
i
a
l
v
i
e
w
S
p
l
e
e
n
R
i
g
h
t
k
i
d
n
e
y
Le
f
t
k
i
d
n
e
y
G
a
l
l
b
l
a
d
d
e
r
Eso
p
h
a
g
u
s
Li
v
e
r
S
t
o
m
a
c
h
A
o
r
t
a
P
o
o
r
v
e
n
a
c
a
v
a
S
p
l
e
n
i
c
v
e
i
n
a
n
d
p
o
r
t
a
l
v
e
i
n
P
a
n
c
r
e
a
s
R
i
g
h
t
a
d
r
e
n
a
l
g
l
a
n
d
Le
f
t
a
d
r
e
n
a
l
g
l
a
n
d
2
.
2
.
1
.
Da
t
a
pre
-
pro
ce
s
s
ing
Sev
er
al
p
r
ep
r
o
ce
s
s
in
g
s
tep
s
a
r
e
tak
en
to
s
tan
d
a
r
d
ize
in
p
u
t
d
ata
an
d
to
im
p
r
o
v
e
m
o
d
el
p
er
f
o
r
m
a
n
ce
p
r
io
r
to
u
s
in
g
th
e
C
T
d
ataset
f
o
r
m
o
d
el
tr
ain
in
g
.
T
o
r
escale
th
e
C
T
im
ag
e
in
ten
s
ity
v
al
u
es
b
etwe
en
0
to
1
u
n
if
o
r
m
ly
th
e
C
T
im
ag
e
in
te
n
s
ity
v
alu
es
ar
e
p
er
f
o
r
m
e
d
i
n
ten
s
ity
n
o
r
m
aliza
tio
n
an
d
th
e
s
am
e
is
ex
p
r
ess
ed
b
y
(
1
)
;
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
A
d
va
n
ce
d
ma
ch
in
e
le
a
r
n
in
g
fo
r
en
h
a
n
ce
d
a
b
d
o
min
a
l
o
r
g
a
n
s
eg
men
ta
tio
n
(
R
o
h
in
i P
a
w
a
r
)
763
=
−
−
(
1
)
wh
er
e
I
is
th
e
d
ef
au
lt
i
n
ten
s
ity
v
alu
e,
an
d
I
min
an
d
I
max
ar
e
th
e
m
ax
im
u
m
an
d
m
in
im
u
m
i
n
ten
s
ity
f
ig
u
r
es
in
d
atasets
.
I
n
o
r
d
er
to
m
ain
tain
c
o
n
s
tan
t
i
n
p
u
t
d
im
en
s
io
n
s
f
o
r
t
h
e
m
o
d
e
l,
all
p
h
o
to
s
in
th
e
d
ataset
ar
e
r
esam
p
led
to
a
u
n
if
o
r
m
r
eso
lu
tio
n
,
e
n
s
u
r
in
g
h
o
m
o
g
en
eity
th
r
o
u
g
h
o
u
t.
T
o
en
h
a
n
ce
th
e
v
ar
iety
a
n
d
r
esil
ien
c
e
o
f
th
e
m
o
d
el,
th
e
tr
ai
n
in
g
d
ata
u
n
d
er
g
o
es
s
ev
er
al
tech
n
iq
u
es
lik
e
r
o
tatio
n
,
s
ca
lin
g
,
f
lip
p
in
g
,
an
d
elastic
d
ef
o
r
m
atio
n
s
.
Data
au
g
m
e
n
tatio
n
ca
n
b
e
ex
p
r
ess
ed
q
u
an
titat
iv
ely
as [
2
4
]
,
(
)
=
(
)
+
∈
(
2
)
wh
er
e,
T
(x)
d
en
o
tes
tr
an
s
f
o
r
m
e
d
im
ag
e,
A
(x)
d
en
o
tes
au
g
m
en
t
ed
o
p
er
atio
n
an
d
∈
is
th
e
r
an
d
o
m
n
o
is
e
o
r
ca
n
b
e
ter
m
ed
as p
er
tu
r
b
atio
n
s
.
2
.
2
.
2
.
M
o
del t
ra
ini
ng
T
h
e
R
esUn
et
m
o
d
el
in
teg
r
ates
th
e
r
esid
u
al
lear
n
in
g
t
ec
h
n
iq
u
e
with
ar
c
h
itectu
r
e
o
f
U
-
Net.
T
h
e
co
n
v
en
tio
n
al
U
-
Net
is
en
h
an
ce
d
b
y
r
esid
u
al
c
o
n
n
ec
tio
n
s
,
w
h
ich
ca
n
b
e
f
o
r
m
ally
d
ef
in
ed
as,
=
(
,
{
}
)
+
(
3
)
wh
er
e,
x
i
n
d
icate
s
in
p
u
t
g
iv
e
n
to
lay
e
r
,
F
(
x
,
{W
i}
)
in
d
icate
s
o
u
tp
u
t
o
f
b
lo
c
k
a
n
d
y
d
en
o
te
s
lay
er
o
u
tp
u
t
p
o
s
t
r
esid
u
al
co
n
n
ec
ti
o
n
.
T
h
e
r
ea
s
o
n
th
e
l
o
s
s
f
u
n
ctio
n
is
u
s
ed
i
s
to
o
p
tim
izatio
n
wh
ic
h
is
m
o
r
e
co
m
p
atib
le
f
o
r
s
eg
m
en
tatio
n
task
.
T
h
e
c
o
ef
f
ic
ien
t o
f
d
ice
is
g
iv
e
n
as [
2
5
]
,
=
2
|
∩
|
|
|
+
|
|
(
4
)
wh
er
e,
P
u
s
ed
f
o
r
s
eg
m
e
n
tatio
n
an
d
G
is
tr
u
th
s
eg
m
e
n
tatio
n
.
Mo
d
el
m
i
n
im
ized
t
h
e
d
ice
lo
s
s
wh
ich
is
g
iv
en
b
y
,
=
1
–
(
5
)
th
e
in
p
u
t CT
im
ag
es we
r
e
n
o
r
m
alize
d
to
th
e
r
a
n
g
e
[
0
,
1
]
to
e
n
s
u
r
e
u
n
if
o
r
m
ity
i
n
in
ten
s
ity
v
alu
es,
as g
iv
en
b
y
,
=
−
−
(
6
)
wh
er
e,
I
r
ep
r
esen
t
th
e
in
te
n
s
ity
o
f
p
ix
el
an
d
I
min
an
d
I
max
d
e
n
o
te
th
e
m
in
im
u
m
an
d
m
ax
im
u
m
p
ix
el
in
ten
s
ities
in
th
e
d
ataset
r
esp
ec
tiv
ely
.
R
esam
p
lin
g
en
s
u
r
ed
th
at
all
im
ag
es
h
ad
c
o
n
s
is
ten
t
d
im
en
s
io
n
s
,
cr
itical
f
o
r
m
ain
tain
in
g
u
n
i
f
o
r
m
ity
in
in
p
u
t a
cr
o
s
s
th
e
n
etwo
r
k
.
T
h
e
s
to
ch
asti
c
g
r
a
d
ien
t
d
esc
en
t
(
SGD)
o
p
tim
izer
was
e
m
p
lo
y
ed
,
s
elec
ted
f
o
r
its
ef
f
icien
cy
an
d
ab
ilit
y
to
g
en
er
alize
well
o
n
s
eg
m
en
tatio
n
task
s
.
A
m
o
m
en
tu
m
f
ac
to
r
o
f
0
.
9
was
u
s
ed
to
ac
ce
le
r
ate
co
n
v
er
g
en
ce
an
d
r
ed
u
ce
o
s
cillatio
n
s
d
u
r
i
n
g
tr
ain
in
g
.
T
h
e
R
e
s
Un
et
m
o
d
el
was
tr
ain
ed
with
an
in
itial
lear
n
in
g
r
ate
o
f
0
.
0
1
t
h
at
was
d
y
n
am
ically
m
o
d
if
ie
d
u
s
in
g
a
s
tep
d
ec
ay
s
ch
ed
u
le
to
en
s
u
r
e
s
m
o
o
th
c
o
n
v
e
r
g
en
ce
.
I
n
p
ar
ticu
lar
,
t
h
e
lear
n
in
g
r
ate
was
d
ec
r
ea
s
ed
b
y
0
.
1
af
ter
ev
er
y
1
0
e
p
o
ch
s
,
wh
ich
e
n
a
b
led
th
e
m
o
d
el
to
im
p
r
o
v
e
its
p
er
f
o
r
m
a
n
ce
g
r
a
d
u
ally
as
tr
ain
in
g
p
r
o
g
r
ess
ed
.
I
n
o
r
d
er
to
o
p
tim
ize
th
e
m
o
d
el’
s
g
en
er
aliza
tio
n
ca
p
ab
ilit
ies
an
d
co
m
p
u
tatio
n
a
l
ef
f
icien
cy
,
a
s
am
p
le
s
ize
o
f
1
6
was
ch
o
s
en
.
T
h
e
h
ig
h
-
r
eso
lu
tio
n
ch
ar
ac
ter
o
f
th
e
C
T
im
ag
es
i
n
th
e
d
atase
t
also
in
f
lu
e
n
ce
d
th
i
s
d
ec
is
io
n
,
as
well
as
th
e
GPU
m
em
o
r
y
r
eq
u
i
r
em
en
ts
.
T
h
e
tr
ain
in
g
was
co
n
d
u
cted
o
v
er
1
0
0
ep
o
ch
s
,
wh
ich
al
lo
wed
f
o
r
a
s
u
f
f
icie
n
t
n
u
m
b
er
o
f
iter
atio
n
s
to
ac
co
m
p
lis
h
co
n
v
er
g
e
n
ce
wh
il
e
r
ed
u
cin
g
th
e
lik
elih
o
o
d
o
f
o
v
er
f
itti
n
g
.
I
n
ad
d
itio
n
,
a
n
ea
r
l
y
h
altin
g
m
ec
h
a
n
is
m
was
im
p
lem
en
ted
to
ter
m
in
at
e
tr
ain
in
g
if
th
e
v
alid
atio
n
lo
s
s
d
id
n
o
t
im
p
r
o
v
e
f
o
r
1
0
c
o
n
s
ec
u
tiv
e
ep
o
c
h
s
,
th
er
eb
y
p
r
ev
e
n
tin
g
u
n
n
ec
ess
ar
y
o
v
er
t
r
ain
in
g
a
n
d
e
n
s
u
r
in
g
t
h
e
ef
f
icien
t u
s
e
o
f
co
m
p
u
tatio
n
al
r
eso
u
r
ce
s
.
2
.
3
.
M
o
del v
a
lid
a
t
io
n
T
o
r
ig
o
r
o
u
s
ly
ev
alu
a
te
th
e
g
e
n
er
aliza
tio
n
ca
p
ab
ilit
ies
o
f
th
e
R
esUn
et
m
o
d
el,
we
ad
o
p
ted
a
5
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
p
r
o
to
co
l.
I
n
t
h
is
estab
lis
h
ed
ap
p
r
o
ac
h
,
th
e
en
tire
d
ataset
was
p
ar
titi
o
n
ed
in
to
f
iv
e
s
u
b
s
ets
(
o
r
‘
f
o
ld
s
’
)
o
f
co
m
p
a
r
ab
le
s
ize.
T
h
e
ex
p
e
r
im
en
tal
p
r
o
ce
d
u
r
e
was
r
ep
ea
ted
f
iv
e
tim
es:
d
u
r
in
g
ea
ch
iter
atio
n
,
o
n
e
s
u
b
s
et
was
r
eser
v
ed
as
th
e
h
o
ld
o
u
t
v
alid
atio
n
s
et,
wh
ile
th
e
o
th
er
f
o
u
r
s
u
b
s
ets
wer
e
u
tili
ze
d
f
o
r
m
o
d
el
tr
ai
n
in
g
.
T
h
is
m
eth
o
d
i
ca
l
r
o
tatio
n
en
s
u
r
e
d
t
h
at
ev
e
r
y
d
ata
p
o
i
n
t
was
in
cl
u
d
ed
in
t
h
e
v
alid
atio
n
p
h
ase
ex
ac
tl
y
o
n
ce
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
5
,
No
.
2
,
J
u
n
e
20
2
6
:
7
5
9
-
7
6
8
764
Fo
r
a
r
o
b
u
s
t
ass
ess
m
en
t
o
f
th
e
m
o
d
el
’
s
s
eg
m
en
tatio
n
p
er
f
o
r
m
an
ce
,
we
ca
lcu
lated
an
d
r
e
p
o
r
ted
th
e
av
er
ag
e
r
esu
lts
o
f
k
ey
p
er
f
o
r
m
an
ce
m
etr
ics
ac
r
o
s
s
all
f
i
v
e
f
o
ld
s
.
T
h
ese
m
etr
ics
in
clu
d
ed
th
e
DSC
,
th
e
Hau
s
d
o
r
f
f
d
is
tan
ce
(
HD)
,
an
d
th
e
av
er
ag
e
s
u
r
f
ac
e
d
is
tan
ce
(
ASD)
.
T
h
e
s
elec
tio
n
o
f
th
e
K=
5
co
n
f
ig
u
r
atio
n
was
s
tr
ateg
ically
m
ad
e
to
ac
h
iev
e
an
o
p
tim
al
b
ala
n
ce
b
e
twee
n
th
e
d
ep
t
h
o
f
m
o
d
el
e
v
alu
atio
n
an
d
th
e
co
n
s
tr
ain
ts
o
f
co
m
p
u
tatio
n
al
e
f
f
icien
cy
,
g
iv
en
th
e
in
h
er
e
n
t size
an
d
co
m
p
lex
ity
o
f
o
u
r
im
ag
e
d
ataset.
3.
RE
SU
L
T
S
AND
D
I
SCU
SS
I
O
N
T
h
is
s
ec
tio
n
in
tr
o
d
u
ce
s
th
e
q
u
an
titativ
e
r
esu
lts
d
er
iv
ed
f
r
o
m
th
e
s
eg
m
en
tatio
n
task
,
f
o
cu
s
in
g
o
n
t
h
e
R
esUn
et
m
o
d
el’
s
ab
ilit
y
to
ac
cu
r
ately
d
elin
ea
te
v
ar
io
u
s
ab
d
o
m
in
al
o
r
g
a
n
s
.
W
e
p
r
im
ar
ily
u
s
e
th
e
DS
C
as
o
u
r
m
ain
p
er
f
o
r
m
a
n
ce
in
d
icato
r
.
T
h
e
DSC
is
a
well
-
estab
lis
h
ed
,
s
tan
d
ar
d
m
etr
ic
with
in
m
ed
ical
im
ag
e
an
aly
s
is
,
d
esig
n
ed
to
p
r
ec
is
ely
q
u
an
tif
y
th
e
d
eg
r
ee
o
f
s
p
atia
l
o
v
er
lap
b
etwe
en
th
e
m
o
d
el’
s
p
r
ed
icte
d
s
eg
m
en
tatio
n
an
d
th
e
m
an
u
ally
an
n
o
tated
g
r
o
u
n
d
tr
u
th
.
I
t
is
in
ter
p
r
eted
s
u
ch
t
h
at
h
ig
h
e
r
v
alu
es
i
n
d
icate
s
u
p
er
io
r
s
eg
m
en
tatio
n
ac
cu
r
ac
y
,
with
a
p
er
f
ec
t
s
co
r
e
o
f
1
.
0
r
e
p
r
esen
tin
g
co
m
p
lete
an
d
p
er
f
ec
t
alig
n
m
e
n
t
w
ith
th
e
tr
u
e
o
r
g
an
b
o
u
n
d
ar
ies.
T
h
e
r
esu
lts
in
T
ab
le
2
clea
r
l
y
s
h
o
w
th
at
th
e
R
esUn
et
m
o
d
el’
s
ab
ilit
y
to
s
ep
ar
ate
o
r
g
a
n
s
is
v
er
y
d
if
f
er
en
t
f
r
o
m
o
n
e
an
o
t
h
er
.
T
h
e
m
o
d
el
d
id
a
g
r
ea
t
jo
b
o
f
s
ep
ar
atin
g
th
e
s
p
lee
n
in
t
o
its
p
ar
ts
.
I
t
g
o
t
a
h
ig
h
m
ea
n
DSC
o
f
0
.
8
2
9
9
,
wh
ich
m
ea
n
s
it
was
ac
cu
r
ate
in
all
o
f
its
test
s
.
T
h
is
g
o
o
d
r
esu
lt
is
p
r
o
b
ab
ly
b
ec
a
u
s
e
th
e
s
p
leen
is
b
ig
g
er
an
d
h
as
cle
ar
ed
g
es,
wh
ich
m
ak
es
it
ea
s
ier
f
o
r
th
e
m
o
d
el
to
f
i
n
d
a
n
d
d
iv
id
e
c
o
r
r
ec
tly
.
T
h
e
m
o
d
el
also
d
id
a
g
o
o
d
jo
b
with
th
e
k
id
n
ey
s
.
T
h
e
r
ig
h
t
k
id
n
ey
h
a
d
a
m
ea
n
DSC
v
alu
e
o
f
0
.
7
4
5
2
an
d
th
e
lef
t
k
id
n
ey
h
ad
a
v
al
u
e
o
f
0
.
7
1
2
4
.
T
h
er
e
was
s
o
m
e
v
ar
iat
io
n
in
th
e
r
esu
lts
,
esp
ec
ially
f
o
r
th
e
lef
t
k
i
d
n
ey
,
b
u
t
th
e
to
tal
p
er
f
o
r
m
an
ce
is
g
o
o
d
e
n
o
u
g
h
f
o
r
m
o
s
t
clin
ical
u
s
es,
lik
e
f
in
d
in
g
k
id
n
e
y
tu
m
o
u
r
s
o
r
m
a
k
in
g
p
lan
s
f
o
r
s
u
r
g
er
ies.
T
h
e
m
o
d
el
h
ad
a
h
ar
d
tim
e
with
th
e
g
allb
lad
d
er
,
th
o
u
g
h
.
I
ts
m
ea
n
DSC
v
al
u
e
was
o
n
ly
0
.
1
4
3
3
,
wh
ich
m
ea
n
s
it
wasn
’
t
v
er
y
g
o
o
d
at
s
eg
m
en
tin
g
.
T
h
is
p
r
o
b
l
em
p
r
o
b
a
b
ly
h
a
p
p
en
s
b
ec
au
s
e
th
e
g
allb
lad
d
er
is
s
m
all,
h
as
an
o
d
d
s
h
a
p
e,
an
d
d
o
esn
’
t
h
a
v
e
clea
r
ed
g
es,
wh
i
ch
m
ak
es
it
h
ar
d
f
o
r
th
e
m
o
d
e
l
to
tell
it
ap
ar
t
f
r
o
m
o
th
er
ce
lls
n
ea
r
b
y
.
A
m
ea
n
DS
C
v
alu
e
o
f
0
.
0
0
0
4
f
o
r
th
e
o
eso
p
h
ag
u
s
,
wh
ic
h
was th
e
h
ar
d
est
o
r
g
an
t
o
s
ep
ar
ate,
s
h
o
ws
th
at
th
e
m
o
d
el
d
id
n
o
t
d
o
a
g
o
o
d
jo
b
o
f
it.
T
h
is
b
ad
p
er
f
o
r
m
a
n
ce
m
a
y
h
av
e
b
ee
n
ca
u
s
ed
b
y
th
e
eso
p
h
ag
u
s
’
s
n
ar
r
o
w,
tu
b
u
lar
s
h
ap
e
an
d
its
clo
s
en
ess
to
o
th
er
b
o
d
y
p
ar
ts
.
T
h
is
s
u
g
g
ests
th
at
th
e
R
e
s
Un
et
m
o
d
el,
in
its
cu
r
r
en
t f
o
r
m
,
m
i
g
h
t n
o
t
b
e
g
o
o
d
f
o
r
task
s
th
at
r
eq
u
ir
e
s
eg
m
en
tin
g
th
e
o
eso
p
h
ag
u
s
.
T
ab
le
2
.
DSC
f
o
r
d
i
f
f
er
en
t c
it
y
s
ca
n
s
O
r
g
a
n
n
a
me
i
mg
0
0
3
7
.
n
i
i
i
mg
0
0
3
0
.
n
i
i
i
mg
0
0
0
7
.
n
i
i
i
mg
0
0
3
9
.
n
i
i
i
mg
0
0
2
4
.
n
i
i
S
p
l
e
e
n
0
.
8
0
7
3
0
.
8
6
5
8
0
.
6
3
6
8
0
.
9
2
8
8
0
.
9
1
0
6
R
i
g
h
t
k
i
d
n
e
y
0
.
7
9
1
3
0
.
7
2
8
3
0
.
7
0
3
8
0
.
6
9
2
1
0
.
8
1
0
6
Le
f
t
k
i
d
n
e
y
0
.
7
3
0
4
0
.
8
1
7
5
0
.
5
9
9
0
0
.
6
1
9
1
0
.
7
9
5
7
G
a
l
l
b
l
a
d
d
e
r
0
.
3
7
0
2
0
.
1
5
2
9
0
.
0
7
8
9
0
.
1
1
3
4
0
.
0
0
0
9
O
e
so
p
h
a
g
u
s
0
.
0
0
0
9
0
.
0
0
0
3
0
.
0
0
0
5
0
.
0
0
0
2
0
.
0
0
0
3
Li
v
e
r
0
.
8
6
2
3
0
.
8
2
4
5
0
.
8
7
2
2
0
.
9
1
7
2
0
.
9
2
1
5
S
t
o
m
a
c
h
0
.
4
0
4
1
0
.
7
6
5
5
0
.
5
7
5
4
0
.
8
1
0
8
0
.
7
1
5
4
A
o
r
t
a
0
.
7
6
1
9
0
.
7
6
5
6
0
.
5
0
8
9
0
.
5
7
8
1
0
.
8
1
3
4
I
n
f
e
r
i
o
r
v
e
n
a
c
a
v
a
0
.
6
6
6
3
0
.
7
6
1
4
0
.
6
5
7
4
0
.
5
5
4
3
0
.
3
8
4
9
P
o
r
t
a
l
v
e
i
n
a
n
d
sp
l
e
n
i
c
v
e
i
n
0
.
5
2
5
7
0
.
5
9
1
6
0
.
4
1
4
1
0
.
3
7
5
0
0
.
5
9
2
7
P
a
n
c
r
e
a
s
0
.
3
9
3
8
0
.
6
4
9
2
0
.
1
6
6
7
0
.
5
0
7
2
0
.
4
2
9
7
R
i
g
h
t
a
d
r
e
n
a
l
g
l
a
n
d
0
.
1
1
4
7
0
.
2
6
1
7
0
.
0
0
0
5
0
.
0
6
4
0
0
.
0
0
5
1
Le
f
t
a
d
r
e
n
a
l
g
l
a
n
d
0
.
0
2
0
5
0
.
2
8
0
1
0
.
0
0
0
0
0
.
0
4
8
7
0
.
2
0
4
0
T
h
e
Fig
u
r
e
s
1
an
d
2
d
em
o
n
s
tr
ates
h
o
w
th
e
wo
r
k
i
n
g
s
p
ee
d
o
f
th
e
R
esUn
et
m
o
d
el
c
h
an
g
e
s
with
th
e
n
u
m
b
er
o
f
s
lices
f
o
r
d
if
f
er
e
n
t
C
T
s
ca
n
f
iles
,
in
clu
d
i
n
g
im
g
0
0
3
7
.
n
ii,
im
g
0
0
3
9
.
n
ii,
im
g
0
0
0
7
.
n
ii,
im
g
0
0
2
4
.
n
ii,
an
d
im
g
0
0
3
0
.
n
ii.
T
h
e
s
p
ee
d
,
wh
ich
is
g
iv
en
in
s
lices
p
er
s
ec
o
n
d
,
t
ells
y
o
u
h
o
w
f
ast
th
e
m
o
d
el
p
r
o
ce
s
s
es
ea
ch
s
ca
n
,
an
d
th
e
f
o
r
m
tells
y
o
u
h
o
w
m
an
y
s
lices
ar
e
in
ea
ch
C
T
s
ca
n
.
T
h
e
r
esear
ch
s
h
o
ws
th
at
th
e
p
r
o
ce
s
s
in
g
s
p
ee
d
ch
an
g
es
a
lo
t
f
r
o
m
o
n
e
s
ca
n
to
th
e
n
ex
t.
As
an
ex
am
p
le,
im
g
0
0
0
7
.
n
ii
is
h
an
d
led
th
e
f
astes
t,
at
1
6
.
3
1
7
s
lices
p
er
s
ec
o
n
d
,
ev
e
n
th
o
u
g
h
it
h
as
th
e
m
o
s
t
s
lices
(
1
6
3
)
.
I
f
we
co
m
p
a
r
e
th
is
to
im
g
0
0
3
0
.
n
ii,
wh
ich
h
as
1
2
4
s
lices,
it
is
h
an
d
led
m
o
r
e
s
lo
wly
,
at
1
2
.
0
3
8
s
lices
p
er
s
ec
o
n
d
.
T
h
is
s
h
o
ws
th
at
th
e
p
r
o
ce
s
s
in
g
s
p
ee
d
is
n
o
t
d
ir
ec
tly
r
elate
d
to
th
e
n
u
m
b
e
r
o
f
s
lic
es.
I
n
s
tead
,
h
o
w
q
u
ick
l
y
th
e
m
o
d
el
ca
n
s
eg
m
en
t
th
e
im
ag
es
m
ay
d
e
p
en
d
o
n
o
th
er
f
ac
t
o
r
s
.
I
t’
s
in
ter
esti
n
g
th
at
im
g
0
0
2
4
.
n
ii,
th
e
s
ca
n
w
ith
th
e
f
ewe
s
t
s
lice
s
(
9
0
)
,
d
o
es
n
’
t
h
an
d
le
d
ata
th
e
f
astes
t.
I
n
s
tead
,
it
m
o
v
es
at
a
s
lo
wer
r
ate
o
f
1
2
.
0
7
5
s
lices
p
er
s
ec
o
n
d
.
T
h
is
s
u
g
g
ests
th
at
th
e
co
m
p
lex
ity
o
f
t
h
e
an
ato
m
ical
s
tr
u
ctu
r
es,
th
e
s
h
a
r
p
n
ess
o
f
th
e
o
r
g
a
n
b
o
u
n
d
ar
i
es,
an
d
m
ay
b
e
ev
e
n
h
o
w
wel
l
th
e
m
o
d
el
ca
n
b
e
co
m
p
u
ted
m
ig
h
t
h
av
e
a
b
ig
g
e
r
ef
f
ec
t
o
n
p
r
o
ce
s
s
in
g
s
p
ee
d
t
h
an
ju
s
t
th
e
n
u
m
b
e
r
o
f
s
lices.
T
h
e
o
v
er
all
r
esu
lts
s
h
o
w
th
at
th
er
e
is
s
o
m
e
in
ter
a
ctio
n
b
etwe
en
th
e
n
u
m
b
e
r
o
f
s
lices
an
d
th
e
w
o
r
k
in
g
s
p
ee
d
,
b
u
t
it’s
n
o
t
a
s
im
p
le
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
A
d
va
n
ce
d
ma
ch
in
e
le
a
r
n
in
g
fo
r
en
h
a
n
ce
d
a
b
d
o
min
a
l
o
r
g
a
n
s
eg
men
ta
tio
n
(
R
o
h
in
i P
a
w
a
r
)
765
o
n
e.
T
h
e
h
ig
h
p
r
o
ce
s
s
in
g
s
p
e
ed
o
f
th
e
s
ca
n
with
th
e
m
o
s
t
s
lices
s
h
o
ws
th
at
th
e
R
esUn
et
m
o
d
el
ca
n
h
an
d
le
b
ig
g
er
d
atasets
well.
B
u
t
th
e
m
o
d
el’
s
s
u
cc
ess
is
af
f
ec
ted
b
y
m
o
r
e
th
an
ju
s
t
th
e
s
ize
o
f
th
e
d
ataset.
T
h
is
s
h
o
ws
h
o
w
d
if
f
icu
lt it is to
s
ep
ar
ate
p
ar
ts
o
f
m
ed
ical
im
ag
es.
Fig
u
r
e
1
.
Me
an
d
ice
v
al
u
es p
lo
t f
o
r
d
if
f
er
e
n
t o
r
g
an
s
ch
o
s
en
in
th
e
s
tu
d
y
Fig
u
r
e
2
.
R
elatio
n
s
h
ip
b
etwe
e
n
s
h
ap
e
an
d
s
p
ee
d
f
o
r
d
i
f
f
er
en
t CT
s
ca
n
im
ag
es
4.
CO
NCLU
SI
O
N
AND
F
U
T
U
RE
DIR
E
C
T
I
O
N
T
h
e
s
tu
d
y
s
p
ec
if
ically
attem
p
ted
to
id
en
tify
th
e
m
o
d
el’
s
ca
p
ab
ilit
ies
an
d
co
n
s
tr
ain
ts
wh
i
le
d
ea
lin
g
with
v
ar
ied
an
ato
m
ical
s
tr
u
ct
u
r
es.
T
h
e
R
esUn
et
m
o
d
el
ex
h
ib
ited
ex
ce
p
tio
n
al
s
eg
m
en
tat
io
n
p
r
ec
is
i
o
n
wh
e
n
ap
p
lied
to
b
ig
g
er
o
r
g
a
n
s
ch
ar
a
cter
ized
b
y
well
d
elin
ea
ted
b
o
r
d
er
s
.
T
h
e
liv
er
attain
ed
t
h
e
h
i
g
h
est
av
er
ag
e
DSC
o
f
0
.
8
8
0
,
wh
er
ea
s
th
e
s
p
leen
o
b
tain
ed
an
a
v
er
ag
e
DSC
o
f
0
.
8
3
0
.
T
h
e
h
ig
h
s
co
r
es
s
u
g
g
e
s
t
th
at
th
e
m
o
d
el
is
v
er
y
s
u
itab
le
f
o
r
ac
cu
r
ately
d
i
v
i
d
in
g
b
ig
,
clea
r
ly
d
ef
in
e
d
o
r
g
an
s
in
to
s
ep
ar
ate
s
eg
m
en
ts
.
T
h
e
r
ig
h
t
k
id
n
ey
h
a
d
a
s
atis
f
ac
to
r
y
p
er
f
o
r
m
a
n
ce
,
with
a
m
ea
n
DSC
o
f
0
.
7
4
5
.
C
o
n
v
er
s
ely
,
th
e
lef
t
k
id
n
ey
e
x
h
ib
ited
a
s
lig
h
tly
lo
wer
m
ea
n
DSC
o
f
0
.
7
1
2
,
in
d
icatin
g
c
o
n
s
is
ten
t
p
er
f
o
r
m
an
ce
in
k
id
n
e
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
5
,
No
.
2
,
J
u
n
e
20
2
6
:
7
5
9
-
7
6
8
766
s
eg
m
en
tatio
n
d
esp
ite
m
in
o
r
v
ar
iatio
n
s
.
T
h
ese
f
in
d
i
n
g
s
in
d
icate
th
at
alth
o
u
g
h
th
e
m
o
d
el
h
as
ef
f
icac
y
i
n
s
eg
m
en
tin
g
lar
g
er
o
r
g
an
s
,
it
en
co
u
n
ter
s
d
if
f
icu
lties
wh
en
s
eg
m
en
tin
g
o
r
g
an
s
th
at
ar
e
en
co
m
p
ass
ed
b
y
in
tr
icate
an
ato
m
ical
s
tr
u
ctu
r
es
o
r
p
o
s
s
ess
v
ar
y
in
g
f
o
r
m
s
.
T
h
e
g
allb
la
d
d
er
h
ad
a
m
ea
n
DSC
o
f
0
.
1
4
3
,
in
d
icatin
g
a
lo
w
lev
el
o
f
s
im
ilar
ity
,
w
h
ile
th
e
p
an
c
r
ea
s
ac
h
iev
ed
a
DSC
s
co
r
e
o
f
0
.
4
2
9
.
T
h
e
p
r
o
ce
s
s
o
f
d
iv
i
d
in
g
th
e
r
ig
h
t
an
d
lef
t
ad
r
en
al
g
l
an
d
s
in
to
s
ep
ar
ate
s
eg
m
en
ts
was
f
o
u
n
d
to
b
e
d
i
f
f
icu
lt,
as
s
h
o
wn
b
y
th
e
m
ea
n
DSC
s
o
f
0
.
0
8
9
an
d
0
.
1
1
1
,
r
esp
ec
tiv
ely
.
T
h
e
s
eg
m
en
tatio
n
o
f
th
e
o
eso
p
h
a
g
u
s
ex
h
ib
ited
a
s
ig
n
if
ican
t
li
m
itatio
n
,
as
s
ee
n
b
y
a
m
ea
n
DSC
o
f
0
.
0
0
0
.
T
h
is
in
d
icate
s
th
at
th
e
m
o
d
el
was u
n
ab
le
to
ac
cu
r
ately
s
eg
m
en
t th
is
p
ar
ticu
lar
o
r
g
a
n
.
I
m
p
r
o
v
in
g
th
e
m
o
d
el’
s
c
o
n
ce
n
tr
atio
n
o
n
cr
itical
r
eg
io
n
s
,
s
u
ch
as
SE
b
lo
ck
s
o
r
s
elf
-
atten
ti
o
n
lay
e
r
s
,
ca
n
im
p
r
o
v
e
s
eg
m
en
tatio
n
a
c
cu
r
ac
y
f
o
r
s
m
aller
an
d
less
d
is
tin
ct
o
r
g
an
s
b
y
in
c
o
r
p
o
r
atin
g
atten
tio
n
m
o
d
u
les.
T
h
e
m
o
d
el’
s
g
en
er
aliza
tio
n
t
o
a
v
ar
iety
o
f
an
at
o
m
i
ca
l
s
tr
u
ctu
r
es
ca
n
b
e
en
h
a
n
ce
d
b
y
th
e
ex
p
an
s
io
n
o
f
au
g
m
en
tatio
n
tech
n
iq
u
es,
s
u
c
h
as
s
y
n
th
etic
d
ata
g
en
er
atio
n
th
r
o
u
g
h
GANs
,
wh
ich
ca
n
h
elp
r
eso
lv
e
d
ata
s
ca
r
city
.
T
h
e
ef
f
icac
y
o
f
co
m
p
lex
s
eg
m
en
tatio
n
task
s
co
u
ld
b
e
en
h
an
ce
d
b
y
in
te
g
r
atin
g
f
ea
tu
r
es
f
r
o
m
m
u
lti
-
s
ca
le
f
ea
tu
r
e
f
u
s
io
n
n
etwo
r
k
s
o
r
co
m
b
in
in
g
R
esUn
et
with
s
tate
-
of
-
th
e
-
a
r
t m
o
d
els s
u
ch
as t
r
an
s
f
o
r
m
er
s
.
ACK
NO
WL
E
DG
M
E
N
T
S
W
e
a
r
e
d
e
e
p
l
y
t
h
a
n
k
f
u
l
t
o
t
h
e
P
r
i
n
c
i
p
a
l
,
De
a
n
,
a
d
m
i
n
i
s
t
r
a
t
i
v
e
a
n
d
t
e
c
h
n
i
c
a
l
s
t
a
f
f
o
f
t
h
e
c
o
l
l
eg
e
f
o
r
t
h
e
i
r
s
e
a
m
l
ess
g
u
i
d
e
a
n
c
e
,
e
n
c
o
r
o
g
em
e
n
t
,
a
s
s
is
t
a
n
c
e
,
a
n
d
f
o
r
p
r
o
v
i
d
i
n
g
a
c
c
e
s
s
t
o
t
h
e
h
i
g
h
-
p
e
r
f
o
r
m
a
n
c
e
c
o
m
p
u
t
i
n
g
r
e
s
o
u
r
c
e
s
r
e
q
u
i
r
e
d
f
o
r
i
m
p
l
e
m
e
n
t
i
n
g
“
A
d
v
a
n
c
e
d
M
L
f
o
r
e
n
h
a
n
c
e
d
a
b
d
o
m
i
n
a
l
o
r
g
a
n
s
e
g
m
e
n
t
a
t
i
o
n
.
”
T
h
e
i
r
s
u
p
p
o
r
t
a
l
l
o
w
e
d
f
o
r
t
h
e
r
i
g
o
r
o
u
s
t
es
t
i
n
g
a
n
d
v
a
l
i
d
at
i
o
n
o
f
t
h
e
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
m
o
d
e
ls
p
r
e
s
en
t
e
d
i
n
t
h
i
s
p
a
p
e
r
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
W
e,
th
e
au
th
o
r
s
o
f
th
e
m
an
u
s
cr
ip
t
titl
ed
“Ad
v
a
n
ce
d
ML
f
o
r
en
h
a
n
ce
d
a
b
d
o
m
in
al
o
r
g
a
n
s
eg
m
en
tatio
n
”
(
Au
th
o
r
s
:
R
o
h
in
i
Pawar
,
Dr
.
R
o
h
in
i
J
ad
h
av
,
Dr
.
R
o
h
it
J
ad
h
a
v
)
,
h
er
e
b
y
d
ec
lar
e
th
at
n
o
ex
ter
n
a
l
f
u
n
d
in
g
was
r
ec
eiv
ed
f
o
r
t
h
is
r
esear
ch
.
T
h
e
wo
r
k
d
escr
ib
ed
in
th
e
p
ap
e
r
was
c
o
n
d
u
ct
ed
s
o
lely
u
s
in
g
th
e
in
s
titu
tio
n
al
r
eso
u
r
ce
s
a
v
ail
ab
l
e
at
th
e
au
t
h
o
r
s
'
r
esp
ec
tiv
e
d
e
p
ar
tm
en
ts
with
in
B
h
ar
ati
Vid
y
ap
ee
th
(
Dee
m
e
d
t
o
b
e
Un
iv
er
s
ity
)
,
Pu
n
e,
I
n
d
ia.
W
e
co
n
f
ir
m
th
at
th
is
d
ec
lar
atio
n
is
ac
cu
r
ate
an
d
co
m
p
lete.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
W
e,
th
e
au
th
o
r
s
,
c
o
n
f
ir
m
th
at
we
h
av
e
all
co
n
tr
ib
u
ted
s
ig
n
if
i
ca
n
tly
to
th
e
r
esear
ch
a
n
d
p
r
e
p
ar
atio
n
o
f
th
e
m
an
u
s
cr
ip
t
titl
ed
“Ad
v
an
ce
d
ML
f
o
r
en
h
an
ce
d
ab
d
o
m
in
al
o
r
g
a
n
s
eg
m
en
tatio
n
”
an
d
ag
r
ee
to
its
s
u
b
m
is
s
io
n
to
th
e
I
n
ter
n
atio
n
al
J
o
u
r
n
al
o
f
I
n
f
o
r
m
atic
s
an
d
C
o
m
m
u
n
icatio
n
T
ec
h
n
o
lo
g
y
(
I
J
-
I
C
T
)
.
T
h
e
f
o
llo
win
g
tab
le
d
etails th
e
s
p
ec
if
ic
co
n
tr
ib
u
tio
n
s
o
f
ea
c
h
au
th
o
r
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
R
o
h
in
i Paw
ar
✓
✓
✓
✓
✓
✓
✓
✓
✓
R
o
h
in
i Jad
h
av
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
R
o
h
it Jad
h
av
✓
✓
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
W
e,
th
e
au
th
o
r
s
o
f
th
e
m
an
u
s
cr
ip
t
titl
ed
“
Ad
v
a
n
ce
d
ML
f
o
r
en
h
a
n
ce
d
a
b
d
o
m
in
al
o
r
g
a
n
s
eg
m
en
tatio
n
”
(
R
o
h
i
n
i
Pawar
,
Dr
.
R
o
h
in
i
J
ad
h
av
,
an
d
Dr
.
R
o
h
it
J
ad
h
av
)
,
h
er
e
b
y
d
ec
lar
e
th
at
n
o
c
o
n
f
licts
o
f
in
ter
est
ex
is
t,
eith
er
f
in
an
cial
o
r
o
th
er
wis
e,
th
at
co
u
ld
b
e
co
n
s
tr
u
ed
as
in
f
lu
en
cin
g
th
e
r
esu
lts
,
in
teg
r
ity
,
o
r
in
ter
p
r
etatio
n
s
p
r
esen
ted
in
t
h
is
m
a
n
u
s
cr
ip
t.
Sp
ec
if
ically
,
th
e
au
th
o
r
s
af
f
ir
m
t
h
at,
we
h
av
e
n
o
co
m
m
e
r
cial
o
r
ass
o
ciativ
e
in
ter
est
th
at
r
ep
r
es
en
ts
a
co
n
f
lict
o
f
in
ter
est
in
co
n
n
ec
tio
n
with
th
e
wo
r
k
s
u
b
m
itted
.
W
e
h
av
e
n
o
t
r
ec
eiv
ed
an
y
f
u
n
d
s
,
g
r
an
ts
,
o
r
f
in
an
cial
s
u
p
p
o
r
t f
r
o
m
a
n
y
o
r
g
an
izatio
n
th
at
wo
u
ld
b
e
n
ef
it f
r
o
m
th
e
p
u
b
licatio
n
o
f
th
is
m
an
u
s
cr
ip
t.
All a
u
t
h
o
r
s
h
av
e
s
ee
n
an
d
ap
p
r
o
v
e
d
th
e
f
i
n
al
v
er
s
io
n
o
f
th
e
m
a
n
u
s
cr
ip
t
an
d
th
e
s
u
b
m
is
s
io
n
o
f
th
is
d
ec
lar
atio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
A
d
va
n
ce
d
ma
ch
in
e
le
a
r
n
in
g
fo
r
en
h
a
n
ce
d
a
b
d
o
min
a
l
o
r
g
a
n
s
eg
men
ta
tio
n
(
R
o
h
in
i P
a
w
a
r
)
767
I
NF
O
RM
E
D
CO
NS
E
N
T
R
eg
ar
d
in
g
o
u
r
m
an
u
s
cr
ip
t
tit
led
“Ad
v
an
ce
d
ML
f
o
r
en
h
a
n
ce
d
ab
d
o
m
in
al
o
r
g
an
s
eg
m
en
tatio
n
,
”
au
th
o
r
ed
b
y
R
o
h
in
i
Pawar
,
Dr
.
R
o
h
in
i
J
ad
h
av
,
an
d
Dr
.
R
o
h
it
J
ad
h
av
,
we
af
f
ir
m
o
u
r
co
m
m
itm
en
t
to
eth
ical
r
esear
ch
s
tan
d
ar
d
s
.
W
e
r
ec
o
g
n
ize
th
at
th
e
p
r
o
tectio
n
o
f
p
r
iv
ac
y
is
a
leg
al
r
ig
h
t
th
at
m
u
s
t
n
o
t
b
e
b
r
ea
ch
ed
with
o
u
t
in
d
i
v
id
u
al
i
n
f
o
r
m
ed
c
o
n
s
en
t.
Ou
r
s
tu
d
y
en
s
u
r
es
th
at
n
o
p
er
s
o
n
al
id
en
tif
y
in
g
in
f
o
r
m
atio
n
is
d
is
clo
s
ed
,
m
ain
tain
in
g
th
e
h
ig
h
est
lev
el
o
f
co
n
f
id
en
tiality
.
W
e
s
tr
ictly
f
o
llo
wed
th
e
in
s
titu
tio
n
al
g
u
id
elin
es
an
d
leg
al
f
r
am
ewo
r
k
s
r
e
q
u
ir
ed
f
o
r
th
e
u
s
e
o
f
d
atasets
in
s
cien
tific
r
e
s
ea
r
ch
.
E
T
H
I
CAL AP
P
RO
V
AL
R
eg
ar
d
in
g
o
u
r
m
an
u
s
cr
ip
t
tit
led
“Ad
v
an
ce
d
ML
f
o
r
en
h
a
n
ce
d
ab
d
o
m
in
al
o
r
g
an
s
eg
m
en
tatio
n
,
”
au
th
o
r
ed
b
y
R
o
h
in
i
Pawar
,
D
r
.
R
o
h
in
i
J
ad
h
av
,
an
d
Dr
.
R
o
h
it
J
ad
h
av
,
we
co
n
f
ir
m
t
h
at
t
h
is
r
esear
ch
s
tr
ictly
ad
h
er
es
to
all
eth
ical
g
u
id
eli
n
es
.
T
h
e
s
tu
d
y
was
co
n
d
u
cted
in
f
u
ll
co
m
p
lian
ce
with
th
e
r
elev
an
t
n
atio
n
al
r
eg
u
latio
n
s
an
d
i
n
s
titu
tio
n
al
p
o
licies
in
ac
co
r
d
an
ce
with
th
e
ten
ets
o
f
th
e
Helsin
k
i
Dec
lar
atio
n
.
W
e
f
o
r
m
ally
s
tate
th
at
th
e
r
esear
ch
h
as
b
ee
n
ap
p
r
o
v
ed
b
y
th
e
I
n
s
titu
tio
n
al
R
ev
iew
B
o
ar
d
o
f
B
h
ar
ati
Vid
y
ap
ee
th
(
Dee
m
ed
to
b
e
Un
iv
er
s
ity
)
C
o
lleg
e
o
f
E
n
g
in
ee
r
in
g
,
Pu
n
e.
DATA AV
AI
L
AB
I
L
I
T
Y
R
eg
ar
d
in
g
o
u
r
m
an
u
s
cr
ip
t
tit
led
“Ad
v
an
ce
d
ML
f
o
r
en
h
a
n
ce
d
ab
d
o
m
in
al
o
r
g
a
n
s
eg
m
en
tatio
n
,
”
we
ar
e
co
m
m
itted
to
p
r
o
m
o
tin
g
tr
an
s
p
ar
e
n
cy
an
d
r
ep
r
o
d
u
c
ib
ilit
y
in
o
u
r
r
esear
c
h
.
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
ail
ab
le
f
r
o
m
th
e
co
r
r
esp
o
n
d
in
g
a
u
th
o
r
,
u
p
o
n
r
ea
s
o
n
ab
le
r
eq
u
est.
W
e
m
ain
tain
th
e
s
u
p
p
o
r
tin
g
e
v
id
en
ce
an
d
r
aw
p
r
o
ce
s
s
ed
d
ata
s
ec
u
r
ely
with
in
o
u
r
in
s
titu
tio
n
al
r
ec
o
r
d
s
at
B
h
ar
ati
Vid
y
ap
ee
th
.
I
n
ter
ested
r
esear
ch
e
r
s
m
ay
co
n
tact
th
e
a
u
th
o
r
s
f
o
r
th
e
p
u
r
p
o
s
e
o
f
s
tu
d
y
.
T
h
is
av
ailab
ilit
y
is
s
u
b
ject
to
eth
ical
co
n
s
id
er
atio
n
s
an
d
th
e
r
estrictio
n
s
estab
lis
h
ed
d
u
r
in
g
th
e
in
itial
d
ata
co
llectio
n
p
h
ase.
W
e
af
f
ir
m
th
at
all
ev
id
en
ce
s
u
p
p
o
r
tin
g
th
e
f
in
d
in
g
s
p
r
esen
ted
in
th
is
ar
ticle
r
em
ain
s
ac
ce
s
s
ib
le
f
o
r
leg
itima
te
s
cien
tific
in
q
u
ir
y
.
RE
F
E
R
E
NC
E
S
[
1
]
Y
.
C
h
e
n
e
t
a
l
.
,
“
F
u
l
l
y
a
u
t
o
ma
t
e
d
m
u
l
t
i
o
r
g
a
n
se
g
m
e
n
t
a
t
i
o
n
i
n
a
b
d
o
m
i
n
a
l
ma
g
n
e
t
i
c
r
e
so
n
a
n
c
e
i
ma
g
i
n
g
w
i
t
h
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
s,”
Me
d
i
c
a
l
P
h
y
s
i
c
s
,
v
o
l
.
4
7
,
n
o
.
1
0
,
p
p
.
4
9
7
1
–
4
9
8
2
,
O
c
t
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
0
2
/
mp
.
1
4
4
2
9
.
[
2
]
Y
.
Zh
o
u
e
t
a
l
.
,
“
S
e
m
i
-
su
p
e
r
v
i
s
e
d
3
D
a
b
d
o
mi
n
a
l
m
u
l
t
i
-
o
r
g
a
n
se
g
me
n
t
a
t
i
o
n
v
i
a
d
e
e
p
mu
l
t
i
-
p
l
a
n
a
r
c
o
-
t
r
a
i
n
i
n
g
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
-
2
0
1
9
I
E
EE
Wi
n
t
e
r
C
o
n
f
e
r
e
n
c
e
o
n
A
p
p
l
i
c
a
t
i
o
n
s
o
f
C
o
m
p
u
t
e
r
V
i
si
o
n
,
WA
C
V
2
0
1
9
,
Ja
n
.
2
0
1
9
,
p
p
.
1
2
1
–
1
4
0
,
d
o
i
:
1
0
.
1
1
0
9
/
W
A
C
V
.
2
0
1
9
.
0
0
0
2
0
.
[
3
]
E.
G
i
b
s
o
n
et
a
l
.
,
“
A
u
t
o
mat
i
c
m
u
l
t
i
-
o
r
g
a
n
s
e
g
me
n
t
a
t
i
o
n
o
n
a
b
d
o
m
i
n
a
l
C
T
w
i
t
h
d
e
n
se
V
-
n
e
t
w
o
r
k
s
,
”
I
EEE
T
r
a
n
s
a
c
t
i
o
n
s
o
n
M
e
d
i
c
a
l
I
m
a
g
i
n
g
,
v
o
l
.
3
7
,
n
o
.
8
,
p
p
.
1
8
2
2
–
1
8
3
4
,
A
u
g
.
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
T
M
I
.
2
0
1
8
.
2
8
0
6
3
0
9
.
[
4
]
S
.
P
a
n
e
t
a
l
.
,
“
A
b
d
o
me
n
C
T
m
u
l
t
i
-
o
r
g
a
n
s
e
g
m
e
n
t
a
t
i
o
n
u
si
n
g
t
o
k
e
n
-
b
a
sed
M
LP
-
M
i
x
e
r
,
”
M
e
d
i
c
a
l
P
h
y
s
i
c
s
,
v
o
l
.
5
0
,
n
o
.
5
,
p
p
.
3
0
2
7
–
3
0
3
8
,
M
a
y
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
2
/
m
p
.
1
6
1
3
5
.
[
5
]
Z.
X
u
e
t
a
l
.
,
“
Ef
f
i
c
i
e
n
t
m
u
l
t
i
-
a
t
l
a
s
a
b
d
o
mi
n
a
l
s
e
g
m
e
n
t
a
t
i
o
n
o
n
c
l
i
n
i
c
a
l
l
y
a
c
q
u
i
r
e
d
C
T
w
i
t
h
S
I
M
P
LE
c
o
n
t
e
x
t
l
e
a
r
n
i
n
g
,
”
Me
d
i
c
a
l
I
m
a
g
e
A
n
a
l
y
s
i
s
,
v
o
l
.
2
4
,
n
o
.
1
,
p
p
.
1
8
–
2
7
,
A
u
g
.
2
0
1
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
m
e
d
i
a
.
2
0
1
5
.
0
5
.
0
0
9
.
[
6
]
P
.
H
.
C
o
n
z
e
e
t
a
l
.
,
“
A
b
d
o
m
i
n
a
l
mu
l
t
i
-
o
r
g
a
n
s
e
g
me
n
t
a
t
i
o
n
w
i
t
h
c
a
s
c
a
d
e
d
c
o
n
v
o
l
u
t
i
o
n
a
l
a
n
d
a
d
v
e
r
sari
a
l
d
e
e
p
n
e
t
w
o
r
k
s,
”
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
i
n
M
e
d
i
c
i
n
e
,
v
o
l
.
1
1
7
,
p
.
1
0
2
1
0
9
,
Ju
l
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
r
t
m
e
d
.
2
0
2
1
.
1
0
2
1
0
9
.
[
7
]
R
.
W
o
l
z
,
C
.
C
h
u
,
K
.
M
i
s
a
w
a
,
M
.
F
u
j
i
w
a
r
a
,
K
.
M
o
r
i
,
a
n
d
D
.
R
u
e
c
k
e
r
t
,
“
A
u
t
o
m
a
t
e
d
a
b
d
o
mi
n
a
l
m
u
l
t
i
-
o
r
g
a
n
s
e
g
m
e
n
t
a
t
i
o
n
w
i
t
h
su
b
j
e
c
t
-
s
p
e
c
i
f
i
c
a
t
l
a
s
g
e
n
e
r
a
t
i
o
n
,
”
I
EEE
T
r
a
n
s
a
c
t
i
o
n
s
o
n
M
e
d
i
c
a
l
I
m
a
g
i
n
g
,
v
o
l
.
3
2
,
n
o
.
9
,
p
p
.
1
7
2
3
–
1
7
3
0
,
S
e
p
.
2
0
1
3
,
d
o
i
:
1
0
.
1
1
0
9
/
TM
I
.
2
0
1
3
.
2
2
6
5
8
0
5
.
[
8
]
C
.
E.
L
e
e
,
M
.
C
h
u
n
g
,
a
n
d
Y
.
G
.
S
h
i
n
,
“
V
o
x
e
l
-
l
e
v
e
l
s
i
a
m
e
se
r
e
p
r
e
s
e
n
t
a
t
i
o
n
l
e
a
r
n
i
n
g
f
o
r
a
b
d
o
m
i
n
a
l
m
u
l
t
i
-
o
r
g
a
n
se
g
me
n
t
a
t
i
o
n
,
”
C
o
m
p
u
t
e
r
M
e
t
h
o
d
s
a
n
d
Pr
o
g
r
a
m
s
i
n
Bi
o
m
e
d
i
c
i
n
e
,
v
o
l
.
2
1
3
,
p
.
1
0
6
5
4
7
,
Ja
n
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
mp
b
.
2
0
2
1
.
1
0
6
5
4
7
.
[
9
]
A
.
A
mj
a
d
e
t
a
l
.
,
“
D
e
e
p
l
e
a
r
n
i
n
g
a
u
t
o
-
seg
me
n
t
a
t
i
o
n
o
n
mu
l
t
i
-
se
q
u
e
n
c
e
m
a
g
n
e
t
i
c
r
e
s
o
n
a
n
c
e
i
m
a
g
e
s
f
o
r
u
p
p
e
r
a
b
d
o
m
i
n
a
l
o
r
g
a
n
s,”
Fro
n
t
i
e
rs
i
n
O
n
c
o
l
o
g
y
,
v
o
l
.
1
3
,
J
u
l
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
8
9
/
f
o
n
c
.
2
0
2
3
.
1
2
0
9
5
5
8
.
[
1
0
]
H
.
K
a
k
e
y
a
,
T.
O
k
a
d
a
,
a
n
d
Y
.
O
s
h
i
r
o
,
“
3
D
U
-
JA
P
A
-
N
e
t
:
m
i
x
t
u
r
e
o
f
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
t
w
o
r
k
s
f
o
r
a
b
d
o
m
i
n
a
l
mu
l
t
i
-
o
r
g
a
n
C
T
seg
m
e
n
t
a
t
i
o
n
,
”
i
n
L
e
c
t
u
re
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
i
n
c
l
u
d
i
n
g
s
u
b
s
e
ri
e
s
L
e
c
t
u
r
e
N
o
t
e
s
i
n
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
L
e
c
t
u
r
e
N
o
t
e
s
i
n
B
i
o
i
n
f
o
rm
a
t
i
c
s)
,
v
o
l
.
1
1
0
7
3
LN
C
S
,
2
0
1
8
,
p
p
.
4
2
6
–
4
3
3
.
[
1
1
]
J
.
W
a
n
g
,
H
.
Z
h
a
o
,
W
.
L
i
a
n
g
,
S
.
W
a
n
g
,
a
n
d
Y
.
Z
h
a
n
g
,
“
C
r
o
s
s
-
c
o
n
v
o
l
u
t
i
o
n
a
l
t
r
a
n
s
f
o
r
m
e
r
f
o
r
a
u
t
o
m
a
t
e
d
m
u
l
t
i
-
o
r
g
a
n
s
s
e
g
m
e
n
t
a
t
i
o
n
i
n
a
v
a
r
i
e
t
y
o
f
m
e
d
i
c
a
l
i
m
a
g
e
s
,
”
P
h
y
s
i
c
s
i
n
M
e
d
i
c
i
n
e
a
n
d
B
i
o
l
o
g
y
,
v
o
l
.
6
8
,
n
o
.
3
,
p
.
0
3
5
0
0
8
,
F
e
b
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
8
8
/
1
3
6
1
-
6
5
6
0
/
a
c
b
1
9
a
.
[
1
2
]
S
.
I
r
sh
a
d
,
D
.
P
.
S
.
G
o
mes
,
a
n
d
S
.
T
.
K
i
m,
“
I
mp
r
o
v
e
d
a
b
d
o
m
i
n
a
l
mu
l
t
i
-
o
r
g
a
n
s
e
g
me
n
t
a
t
i
o
n
v
i
a
3
D
b
o
u
n
d
a
r
y
-
c
o
n
s
t
r
a
i
n
e
d
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
s
,
”
I
E
EE
Ac
c
e
ss
,
v
o
l
.
1
1
,
p
p
.
3
5
0
9
7
–
3
5
1
1
0
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ES
S
.
2
0
2
3
.
3
2
6
4
5
8
2
.
[
1
3
]
X
.
Li
a
n
g
,
N
.
L
i
,
Z
.
Z
h
a
n
g
,
J.
X
i
o
n
g
,
S
.
Z
h
o
u
,
a
n
d
Y
.
X
i
e
,
“
I
n
c
o
r
p
o
r
a
t
i
n
g
t
h
e
h
y
b
r
i
d
d
e
f
o
r
m
a
b
l
e
mo
d
e
l
f
o
r
i
mp
r
o
v
i
n
g
t
h
e
p
e
r
f
o
r
m
a
n
c
e
o
f
a
b
d
o
m
i
n
a
l
C
T
se
g
m
e
n
t
a
t
i
o
n
v
i
a
m
u
l
t
i
-
sca
l
e
f
e
a
t
u
r
e
f
u
si
o
n
n
e
t
w
o
r
k
,
”
M
e
d
i
c
a
l
I
m
a
g
e
A
n
a
l
y
s
i
s
,
v
o
l
.
7
3
,
p
.
1
0
2
1
5
6
,
O
c
t
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
me
d
i
a
.
2
0
2
1
.
1
0
2
1
5
6
.
[
1
4
]
P
.
H
u
,
F
.
W
u
,
J
.
P
e
n
g
,
Y
.
B
a
o
,
F
.
C
h
e
n
,
a
n
d
D
.
K
o
n
g
,
“
A
u
t
o
mat
i
c
a
b
d
o
mi
n
a
l
m
u
l
t
i
-
o
r
g
a
n
s
e
g
m
e
n
t
a
t
i
o
n
u
s
i
n
g
d
e
e
p
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
a
n
d
t
i
me
-
i
mp
l
i
c
i
t
l
e
v
e
l
set
s,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
C
o
m
p
u
t
e
r
Ass
i
st
e
d
R
a
d
i
o
l
o
g
y
a
n
d
S
u
rg
e
ry
,
v
o
l
.
1
2
,
n
o
.
3
,
p
p
.
3
9
9
–
4
1
1
,
M
a
r
.
2
0
1
7
,
d
o
i
:
1
0
.
1
0
0
7
/
s1
1
5
4
8
-
016
-
1
5
0
1
-
5.
[
1
5
]
Y
.
W
a
n
g
,
Y
.
Z
h
o
u
,
W
.
S
h
e
n
,
S
.
P
a
r
k
,
E.
K
.
F
i
s
h
ma
n
,
a
n
d
A
.
L.
Y
u
i
l
l
e
,
“
A
b
d
o
mi
n
a
l
m
u
l
t
i
-
o
r
g
a
n
se
g
me
n
t
a
t
i
o
n
w
i
t
h
o
r
g
a
n
-
a
t
t
e
n
t
i
o
n
n
e
t
w
o
r
k
s
a
n
d
s
t
a
t
i
s
t
i
c
a
l
f
u
si
o
n
,
”
Me
d
i
c
a
l
I
m
a
g
e
A
n
a
l
y
s
i
s
,
v
o
l
.
5
5
,
p
p
.
8
8
–
1
0
2
,
J
u
l
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
m
e
d
i
a
.
2
0
1
9
.
0
4
.
0
0
5
.
[
1
6
]
N
.
To
n
g
,
S
.
G
o
u
,
T.
N
i
u
,
S
.
Y
a
n
g
,
a
n
d
K
.
S
h
e
n
g
,
“
S
e
l
f
-
p
a
c
e
d
D
e
n
s
e
N
e
t
w
i
t
h
b
o
u
n
d
a
r
y
c
o
n
st
r
a
i
n
t
f
o
r
a
u
t
o
mat
e
d
mu
l
t
i
-
o
r
g
a
n
seg
m
e
n
t
a
t
i
o
n
o
n
a
b
d
o
mi
n
a
l
C
T
i
mag
e
s,”
P
h
y
si
c
s
i
n
M
e
d
i
c
i
n
e
a
n
d
Bi
o
l
o
g
y
,
v
o
l
.
6
5
,
n
o
.
1
3
,
p
.
1
3
5
0
1
1
,
Ju
l
.
2
0
2
0
,
d
o
i
:
1
0
.
1
0
8
8
/
1
3
6
1
-
6
5
6
0
/
a
b
9
b
5
7
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
5
,
No
.
2
,
J
u
n
e
20
2
6
:
7
5
9
-
7
6
8
768
[
1
7
]
Y
.
T
a
n
g
e
t
a
l
.
,
“
H
i
g
h
-
r
e
s
o
l
u
t
i
o
n
3
D
a
b
d
o
m
i
n
a
l
s
e
g
m
e
n
t
a
t
i
o
n
w
i
t
h
r
a
n
d
o
m
p
a
t
c
h
n
e
t
w
o
r
k
f
u
si
o
n
,
”
Me
d
i
c
a
l
I
m
a
g
e
A
n
a
l
y
s
i
s
,
v
o
l
.
6
9
,
p
.
1
0
1
8
9
4
,
A
p
r
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
m
e
d
i
a
.
2
0
2
0
.
1
0
1
8
9
4
.
[
1
8
]
P
.
B
i
l
i
c
e
t
a
l
.
,
“
Th
e
l
i
v
e
r
t
u
m
o
r
se
g
me
n
t
a
t
i
o
n
b
e
n
c
h
mar
k
(
L
i
TS)
,
”
Me
d
i
c
a
l
I
m
a
g
e
An
a
l
y
si
s
,
v
o
l
.
8
4
,
p
.
1
0
2
6
8
0
,
F
e
b
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
m
e
d
i
a
.
2
0
2
2
.
1
0
2
6
8
0
.
[
1
9
]
K
.
H
e
,
X
.
Z
h
a
n
g
,
S
.
R
e
n
,
a
n
d
J
.
S
u
n
,
“
D
e
e
p
r
e
s
i
d
u
a
l
l
e
a
r
n
i
n
g
f
o
r
i
ma
g
e
r
e
c
o
g
n
i
t
i
o
n
,
”
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
I
EEE
C
o
m
p
u
t
e
r
S
o
c
i
e
t
y
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
e
r
Vi
s
i
o
n
a
n
d
Pa
t
t
e
r
n
Re
c
o
g
n
i
t
i
o
n
,
v
o
l
.
2
0
1
6
-
D
e
c
e
mb
e
r
,
p
p
.
7
7
0
–
7
7
8
,
2
0
1
6
,
d
o
i
:
1
0
.
1
1
0
9
/
C
V
P
R
.
2
0
1
6
.
9
0
.
[
2
0
]
F
.
I
sen
see
e
t
a
l
.
,
“
A
b
st
r
a
c
t
:
n
n
U
-
N
e
t
:
se
l
f
-
a
d
a
p
t
i
n
g
f
r
a
m
e
w
o
r
k
f
o
r
U
-
N
e
t
-
b
a
s
e
d
me
d
i
c
a
l
i
m
a
g
e
se
g
me
n
t
a
t
i
o
n
,
”
i
n
I
n
f
o
rm
a
t
i
k
a
k
t
u
e
l
l
,
2
0
1
9
,
p
p
.
2
2
–
2
2
.
[
2
1
]
A
.
E.
K
a
v
u
r
e
t
a
l
.
,
“
C
H
A
O
S
c
h
a
l
l
e
n
g
e
-
c
o
m
b
i
n
e
d
(
C
T
-
M
R
)
h
e
a
l
t
h
y
a
b
d
o
mi
n
a
l
o
r
g
a
n
s
e
g
m
e
n
t
a
t
i
o
n
,
”
Me
d
i
c
a
l
I
m
a
g
e
An
a
l
y
si
s
,
v
o
l
.
6
9
,
p
.
1
0
1
9
5
0
,
A
p
r
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
m
e
d
i
a
.
2
0
2
0
.
1
0
1
9
5
0
.
[
2
2
]
B
.
La
n
d
ma
n
,
“
M
I
C
C
A
I
2
0
1
5
c
h
a
l
l
e
n
g
e
o
n
m
u
l
t
i
-
a
t
l
a
s
l
a
b
e
l
i
n
g
b
e
y
o
n
d
t
h
e
c
r
a
n
i
a
l
v
a
u
l
t
,
”
W
o
rks
h
o
p
Pr
o
c
e
e
d
i
n
g
s
,
2
0
1
5
.
h
t
t
p
s
:
/
/
w
w
w
.
l
i
n
k
e
d
i
n
.
c
o
m/
p
u
l
se
/
b
c
v
2
0
1
5
-
mi
c
c
a
i
-
m
u
l
t
i
-
a
t
l
a
s
-
l
a
b
e
l
i
n
g
-
c
h
a
l
l
e
n
g
e
-
w
o
r
k
s
h
o
p
-
l
a
n
d
m
a
n
/
.
[
2
3
]
F
.
M
i
l
l
e
t
a
r
i
,
N
.
N
a
v
a
b
,
a
n
d
S
.
-
A
.
A
h
ma
d
i
,
“
V
-
N
e
t
:
f
u
l
l
y
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s
f
o
r
v
o
l
u
me
t
r
i
c
med
i
c
a
l
i
ma
g
e
seg
m
e
n
t
a
t
i
o
n
,
”
i
n
2
0
1
6
F
o
u
r
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
3
D
V
i
si
o
n
(
3
D
V)
,
O
c
t
.
2
0
1
6
,
p
p
.
5
6
5
–
5
7
1
,
d
o
i
:
1
0
.
1
1
0
9
/
3
D
V
.
2
0
1
6
.
7
9
.
[
2
4
]
C
.
S
h
o
r
t
e
n
a
n
d
T.
M
.
K
h
o
s
h
g
o
f
t
a
a
r
,
“
A
su
r
v
e
y
o
n
i
ma
g
e
d
a
t
a
a
u
g
m
e
n
t
a
t
i
o
n
f
o
r
d
e
e
p
l
e
a
r
n
i
n
g
,
”
J
o
u
r
n
a
l
o
f
Bi
g
D
a
t
a
,
v
o
l
.
6
,
n
o
.
1
,
p
.
6
0
,
D
e
c
.
2
0
1
9
,
d
o
i
:
1
0
.
1
1
8
6
/
s
4
0
5
3
7
-
0
1
9
-
0
1
9
7
-
0.
[
2
5
]
F
.
S
h
a
ms
h
a
d
,
S
.
K
h
a
n
,
S
.
W
.
Z
a
mi
r
,
M
.
H
.
K
h
a
n
,
M
.
H
a
y
a
t
,
F
.
S
.
K
h
a
n
,
a
n
d
H
.
F
u
,
“
Tr
a
n
sf
o
r
m
e
r
s i
n
m
e
d
i
c
a
l
i
mag
i
n
g
:
a
s
u
r
v
e
y
,
”
med
i
c
a
l
i
ma
g
e
a
n
a
l
y
s
i
s
,
v
o
l
.
8
8
,
p
.
1
0
2
8
0
2
,
A
u
g
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
me
d
i
a
.
2
0
2
3
.
1
0
2
8
0
2
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Pro
f.
Ro
h
i
n
i
Pa
wa
r
re
c
e
iv
e
d
h
e
r
M
.
E.
d
e
g
re
e
s
in
c
o
m
p
u
ter
e
n
g
in
e
e
rin
g
fro
m
Bh
a
ra
ti
Vid
y
a
p
e
e
th
(De
e
m
e
d
to
b
e
Un
iv
e
rsity
)
P
u
n
e
,
I
n
d
ia
a
n
d
B.
E.
d
e
g
re
e
in
c
o
m
p
u
ter
e
n
g
in
e
e
rin
g
fro
m
Am
ru
t
v
a
h
i
n
i
C
o
ll
e
g
e
o
f
E
n
g
in
e
e
rin
g
(P
u
n
e
Un
i
v
e
rsity
)
S
a
n
g
a
m
n
e
r,
In
d
ia.
S
h
e
is
c
u
rre
n
tl
y
w
o
rk
i
n
g
a
s
a
S
r.
Lec
tu
re
re
in
th
e
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
Tec
h
n
o
l
o
g
y
a
t
Bh
a
ra
ti
Vid
y
a
p
e
e
th
Ja
wa
h
e
rlal
Ne
h
ru
I
n
stit
u
te
o
f
Tec
h
n
o
l
o
g
y
,
P
u
n
e
,
I
n
d
ia
.
S
h
e
h
a
s
p
u
b
li
s
h
e
d
1
1
+
re
se
a
rc
h
a
rti
c
les
in
S
c
o
p
u
s,
UG
C
Ca
r
e
jo
u
rn
a
ls
a
n
d
c
o
n
fe
re
n
c
e
s.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
1
5
ro
h
in
ip
a
wa
r@g
m
a
il
.
c
o
m
.
Dr
.
Ro
h
in
i
J
a
d
h
a
v
re
c
e
iv
e
d
h
e
r
P
h
.
D.
a
n
d
M
.
E.
d
e
g
re
e
s
in
c
o
m
p
u
ter
e
n
g
i
n
e
e
rin
g
fro
m
Bh
a
ra
ti
Vid
y
a
p
e
e
th
(De
e
m
e
d
to
b
e
Un
i
v
e
rsity
)
P
u
n
e
,
I
n
d
ia
a
n
d
B.
E.
d
e
g
re
e
in
c
o
m
p
u
ter
e
n
g
in
e
e
rin
g
fro
m
Bh
a
ra
ti
Vi
d
y
a
p
e
e
th
(De
e
m
e
d
to
b
e
Un
iv
e
rsity
)
P
u
n
e
,
In
d
ia.
S
h
e
is
c
u
rre
n
tl
y
wo
rk
i
n
g
a
s
a
n
a
ss
o
c
iate
p
ro
fe
ss
o
r
in
t
h
e
De
p
a
rtme
n
t
o
f
In
f
o
rm
a
ti
o
n
Tec
h
n
o
l
o
g
y
a
t
Bh
a
ra
ti
Vid
y
a
p
e
e
th
(De
e
m
e
d
to
b
e
Un
i
v
e
rsity
)
Co
l
leg
e
o
f
En
g
i
n
e
e
rin
g
,
P
u
n
e
,
I
n
d
ia.
He
r
a
re
a
s
o
f
e
x
p
e
r
ti
se
a
re
so
ftwa
re
e
n
g
in
e
e
rin
g
,
ima
g
e
a
n
d
v
id
e
o
p
r
o
c
e
ss
in
g
,
p
a
tt
e
rn
re
c
o
g
n
i
ti
o
n
,
M
L
,
a
n
d
AI.
S
h
e
h
a
s
p
u
b
li
sh
e
d
2
6
re
se
a
rc
h
a
rti
c
les
in
re
p
u
ted
sc
o
p
u
s
in
d
e
x
e
d
jo
u
rn
a
ls
a
n
d
c
o
n
fe
re
n
c
e
s.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
r
b
j
a
d
h
a
v
@b
v
u
c
o
e
p
.
e
d
u
.
i
n
.
Dr
.
Ro
h
it
J
a
d
h
a
v
is
c
u
rre
n
tl
y
wo
r
k
in
g
a
s
a
n
a
ss
istan
t
p
r
o
fe
ss
o
r
with
7
y
e
a
rs
o
f
tea
c
h
in
g
a
n
d
re
se
a
rc
h
e
x
p
e
rien
c
e
in
ENT
a
t
Bh
a
ra
ti
Vi
d
y
a
p
e
e
th
(
De
e
m
e
d
to
b
e
Un
i
v
e
rsity
)
M
e
d
ica
l
Co
ll
e
g
e
,
P
u
n
e
.
His
re
se
a
rc
h
c
o
n
tri
b
u
tes
i
n
G
e
n
e
ra
l
M
e
d
icin
e
,
ENT
a
n
d
p
u
b
l
ish
e
d
9
+
re
se
a
rc
h
a
rti
c
les
in
S
c
o
p
u
s
a
n
d
UG
C
Ca
r
e
jo
u
rn
a
ls.
He
h
a
s
c
o
m
p
lete
d
h
is
M
BBS
a
n
d
M
S
i
n
ENT
fro
m
B
h
a
ra
ti
Vid
y
a
p
e
e
th
(
De
e
m
e
d
to
b
e
U
n
iv
e
rsit
y
)
M
e
d
ica
l
Co
ll
e
g
e
,
P
u
n
e
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
d
r
.
ro
h
it
b
jad
h
a
v
@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.