I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
1
5
,
No
.
2
,
A
p
r
il 2
0
2
6
,
p
p
.
1
6
2
3
~
1
6
3
1
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
15
.i
2
.
p
p
1
6
2
3
-
1
6
3
1
1623
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Ana
ly
sis
of t
ube
r
culo
sis
det
ec
tion
using
deep
l
ea
rni
ng
t
echniq
ue
a
nd
ex
pla
ina
ble
artificial i
ntelligen
ce
Sh
a
s
hik
ira
n
Srin
iv
a
s
1
,
K
a
v
it
a
Av
ina
s
h P
a
t
il
2
,
K
us
ha
la
t
ha
M
o
na
pp
a
Ra
m
a
3
,
Su
dh
a
Venk
a
t
esh
lu
4
,
J
a
y
a
nthi Mu
t
hu
s
wa
m
y
1
,
Srini
v
a
s
B
a
bu
Na
ra
y
a
na
pp
a
1
1
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
o
n
i
c
s
a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
En
g
i
n
e
e
r
i
n
g
,
N
e
w
H
o
r
i
z
o
n
C
o
l
l
e
g
e
o
f
E
n
g
i
n
e
e
r
i
n
g
,
B
e
n
g
a
l
u
r
u
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
o
n
i
c
s
a
n
d
T
e
l
e
c
o
mm
u
n
i
c
a
t
i
o
n
E
n
g
i
n
e
e
r
i
n
g
,
R
a
ma
i
a
h
I
n
st
i
t
u
t
e
o
f
Te
c
h
n
o
l
o
g
y
,
B
e
n
g
a
l
u
r
u
,
I
n
d
i
a
3
D
e
p
a
r
t
m
e
n
t
o
f
El
e
c
t
r
o
n
i
c
s
a
n
d
C
o
mm
u
n
i
c
a
t
i
o
n
E
n
g
i
n
e
e
r
i
n
g
,
N
i
t
t
e
M
e
e
n
a
k
s
h
i
I
n
s
t
i
t
u
t
e
o
f
T
e
c
h
n
o
l
o
g
y
,
N
I
T
T
E
(
D
e
e
m
e
d
t
o
b
e
U
n
i
v
e
r
s
i
t
y
)
,
B
e
n
g
a
l
u
r
u
,
I
n
d
i
a
4
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
A
mi
t
y
U
n
i
v
e
r
si
t
y
,
B
e
n
g
a
l
u
r
u
,
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
Sep
2
6
,
2
0
2
4
R
ev
is
ed
J
an
3
0
,
2
0
2
6
Acc
ep
ted
Feb
6
,
2
0
2
6
Tu
b
e
rc
u
l
o
sis
(TB)
a
ffe
c
ts
th
e
h
e
a
lt
h
o
f
m
a
n
y
i
n
d
i
v
i
d
u
a
ls
a
n
d
is
sti
ll
a
p
rime
wo
rld
wi
d
e
h
e
a
lt
h
c
o
n
c
e
rn
d
e
sp
i
te
h
a
v
i
n
g
so
m
a
n
y
a
d
v
a
n
c
e
d
trea
tm
e
n
ts,
a
s
it
stil
l
lac
k
s
tec
h
n
ica
l
a
d
v
a
n
c
e
m
e
n
t
in
i
ts
trea
tme
n
t
a
n
d
d
iag
n
o
sis.
Ac
c
u
ra
c
y
in
id
e
n
ti
fica
ti
o
n
a
n
d
e
a
rly
d
e
tec
ti
o
n
is
e
ss
e
n
ti
a
l
to
re
d
u
c
e
th
e
sp
re
a
d
a
n
d
imp
ro
v
e
trea
tme
n
t
o
u
tco
m
e
s.
T
ra
d
it
io
n
a
l
m
e
th
o
d
s
o
f
d
iag
n
o
sis
,
su
c
h
a
s
sp
u
tu
m
m
icro
sc
o
p
y
a
n
d
c
u
lt
u
re
,
a
re
lab
o
r
-
d
e
p
e
n
d
e
n
t
a
n
d
s
u
b
jec
t
to
h
u
m
a
n
m
istak
e
s
a
s
it
is
d
o
n
e
b
y
lab
t
e
c
h
n
icia
n
s.
Re
c
e
n
t
imp
r
o
v
e
m
e
n
t
s
in
d
e
e
p
lea
rn
in
g
h
a
v
e
d
e
m
o
n
stra
ted
s
ig
n
ifi
c
a
n
t
p
o
ten
ti
a
l
fo
r
e
n
h
a
n
c
in
g
a
n
d
a
u
to
m
a
ti
n
g
d
iag
n
o
stic
a
c
c
u
ra
c
y
.
Ou
r
re
se
a
rc
h
p
r
o
p
o
se
s
a
d
e
e
p
lea
rn
in
g
-
b
a
se
d
tec
h
n
iq
u
e
th
a
t
d
e
tec
ts
TB
fro
m
c
h
e
st
X
-
ra
y
s
a
fter
im
a
g
e
p
ro
c
e
ss
in
g
tec
h
n
iq
u
e
s
li
k
e
a
u
g
m
e
n
tatio
n
.
Aft
e
r
train
in
g
o
n
b
ig
d
a
ta,
o
u
r
m
o
d
e
l
p
u
l
ls
o
ff
a
n
a
sto
n
is
h
in
g
a
c
c
u
ra
c
y
o
f
9
7
.
4
2
%
a
n
d
a
lo
ss
o
f
7
.
1
7
%
,
o
u
tp
e
rfo
rm
in
g
trad
it
io
n
a
l
m
e
th
o
d
s.
Th
e
m
o
d
e
l
u
se
s
c
o
n
v
o
lu
ti
o
n
a
l
n
e
u
ra
l
n
e
two
r
k
(CNN
)
a
s
a
b
a
se
a
n
d
tran
sfe
r
lea
rn
i
n
g
m
e
th
o
d
,
li
k
e
De
n
se
Ne
t
-
1
2
1
,
a
n
d
e
x
p
lai
n
a
b
le
a
rti
ficia
l
in
telli
g
e
n
c
e
(XA
I)
tec
h
n
iq
u
e
,
li
k
e
G
ra
d
-
CAM
,
to
re
c
o
g
n
ize
TB
-
re
late
d
p
a
tt
e
rn
s
e
ffe
c
ti
v
e
ly
a
n
d
with
l
o
w
fa
lse
p
o
siti
v
e
s.
T
h
is
a
p
p
ro
a
c
h
h
a
s
th
e
a
b
il
it
y
to
re
v
o
lu
t
io
n
ize
th
e
d
i
a
g
n
o
sis
o
f
TB
a
n
d
o
ffe
r
m
o
re
d
e
p
e
n
d
a
b
le,
sc
a
lab
le,
a
n
d
ti
m
e
ly
so
lu
ti
o
n
s t
o
h
e
a
lt
h
c
a
re
sy
ste
m
s wo
rld
wid
e
.
K
ey
w
o
r
d
s
:
Acc
u
r
ac
y
C
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
Den
s
eNe
t
-
121
Gr
ad
-
C
AM
T
r
an
s
f
er
lear
n
i
n
g
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Kav
ita
Av
in
ash
Patil
Dep
ar
tm
en
t o
f
E
lectr
o
n
ics an
d
T
elec
o
m
m
u
n
icatio
n
E
n
g
in
ee
r
in
g
,
R
am
aiah
I
n
s
titu
te
o
f
T
ec
h
n
o
lo
g
y
,
B
en
g
alu
r
u
,
I
n
d
ia.
E
m
ail:
k
a
v
it
a
m
a
lag
a
tt
i@g
m
a
il
.
c
o
m
1.
I
NT
RO
D
UCT
I
O
N
Myco
b
a
cteriu
m
tu
b
er
cu
l
o
s
is
(
Mtb
)
,
th
e
b
ac
te
r
ia
th
at
ca
u
s
es
tu
b
er
cu
l
o
s
is
(
TB
)
[
1
]
,
c
o
n
tin
u
es
to
b
e
a
m
ajo
r
p
u
b
lic
h
ea
lth
co
n
ce
r
n
,
with
m
an
y
n
ew
ca
s
es
an
d
d
ea
t
h
s
ea
ch
y
ea
r
.
Desp
ite
s
ig
n
if
ica
n
t
ad
v
an
ce
m
e
n
ts
in
p
u
b
lic
h
ea
lth
an
d
m
ed
icin
e,
T
B
r
em
ain
s
a
p
r
im
e
ca
u
s
e
o
f
d
e
ath
an
d
m
o
r
b
id
ity
,
p
ar
ticu
lar
ly
in
u
n
d
er
d
ev
elo
p
ed
co
u
n
tr
ies
wh
er
e
ac
ce
s
s
to
h
ea
lth
ca
r
e
u
n
its
is
lim
ited
.
Sp
u
tu
m
[
2
]
s
m
ea
r
m
icr
o
s
co
p
y
[
3
]
a
n
d
ch
est
X
-
r
ay
s
[
4
]
ar
e
two
ess
en
tial
tr
ad
itio
n
al
d
iag
n
o
s
tic
tech
n
i
q
u
es
[
5
]
f
o
r
T
B
.
E
v
en
with
th
ese
d
ev
el
o
p
m
en
ts
,
p
r
o
b
le
m
s
in
clu
d
in
g
lim
ited
s
en
s
itiv
ity
,
h
ig
h
f
alse
-
n
eg
ativ
e
r
ates,
an
d
in
co
n
s
is
ten
t
d
iag
n
o
s
tic
ac
cu
r
ac
y
m
ig
h
t
o
cc
asio
n
ally
lim
it
th
ese
tech
n
iq
u
es.
An
in
n
o
v
ativ
e
s
o
lu
tio
n
to
th
ese
is
s
u
es
is
o
f
f
er
ed
b
y
d
ee
p
lear
n
in
g
[
6
]
,
a
p
o
ten
t
b
r
an
ch
o
f
a
r
tific
ial
in
tellig
en
ce
[
7
]
.
B
y
in
cr
ea
s
in
g
th
e
p
r
ec
is
io
n
an
d
au
t
o
m
atio
n
o
f
c
o
m
p
licated
p
atter
n
d
etec
tio
n
in
im
ag
es
[
8
]
,
d
ee
p
lear
n
in
g
[
9
]
h
as
s
ig
n
if
ican
tly
ch
an
g
ed
th
e
ar
ea
o
f
m
ed
ical
im
ag
in
g
[
1
0
]
.
T
h
is
is
esp
ec
ially
tr
u
e
wh
e
n
u
s
in
g
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
s
(
C
NNs)
[
1
1
]
.
T
h
ese
d
ev
elo
p
m
en
ts
ar
e
p
ar
tic
u
lar
ly
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
2
,
Ap
r
il 2
0
2
6
:
1
6
2
3
-
1
6
3
1
1624
aid
s
in
th
e
id
en
tific
atio
n
o
f
T
B
[
1
2
]
,
s
in
ce
C
NN’
s
ar
e
g
r
o
o
m
e
d
to
id
e
n
tify
s
u
b
tle
ab
n
o
r
m
alities
th
at
tr
ad
itio
n
al
ap
p
r
o
ac
h
es
[
1
3
]
a
r
e
u
n
ab
le
to
id
en
tify
.
T
B
is
u
s
u
ally
cu
r
ab
le
an
d
p
r
ev
en
tab
le.
I
t
is
p
o
s
s
ib
le
to
s
u
cc
es
s
f
u
ll
y
m
o
d
if
y
p
r
etr
ain
e
d
s
tr
u
ctu
r
es
[
1
4
]
f
o
r
r
a
d
io
lo
g
ical
in
ter
p
r
etatio
n
p
er
tin
en
t
to
a
p
ar
ticu
lar
ar
ea
.
Den
s
eNe
t
-
121
[
1
5
]
is
esp
ec
ially
u
s
ef
u
l
f
o
r
T
B
d
etec
tio
n
[
1
6
]
d
u
e
to
its
d
en
s
ely
co
n
n
ec
ted
f
ea
tu
r
e
p
r
o
p
a
g
atio
n
,
wh
ich
im
p
r
o
v
es
g
r
ad
ien
t
s
tab
ilit
y
an
d
p
er
m
its
th
e
r
ec
o
v
er
y
o
f
f
in
e
-
s
ca
le
p
ar
en
ch
y
m
al
p
atter
n
s
[
1
7
]
li
n
k
ed
t
o
T
B
lesi
o
n
s
.
T
h
e
m
o
d
el'
s
[
1
8
]
ca
p
ac
ity
t
o
g
e
n
er
al
ize
is
en
h
an
ce
d
b
y
co
m
b
in
in
g
Den
s
eNe
t
-
121
with
p
ar
ticu
lar
im
b
alan
ce
-
h
an
d
lin
g
p
r
o
ce
s
s
es,
s
u
ch
as
m
in
o
r
ity
-
class
f
r
eq
u
en
cies
as
lo
w
as
1
0
–
2
0
%,
em
p
lo
y
in
g
s
y
n
th
etic
m
in
o
r
ity
o
v
er
s
am
p
lin
g
tech
n
iq
u
e
(
SMOT
E
)
[
1
9
]
an
d
ass
o
ciate
d
h
y
b
r
id
r
esam
p
lin
g
tech
n
i
q
u
es.
Den
s
eNe
t
-
121
,
a
d
ee
p
C
NN
[
2
0
]
,
is
wid
ely
u
s
ed
f
o
r
i
m
ag
e
class
if
icatio
n
ap
p
licatio
n
s
,
s
u
ch
as
T
B
id
en
t
if
icatio
n
f
r
o
m
ch
est
X
-
r
ay
p
ictu
r
es.
T
o
r
ed
u
ce
o
v
er
f
itti
n
g
a
n
d
en
h
a
n
ce
f
ea
tu
r
e
p
r
o
p
a
g
atio
n
,
d
e
n
s
ely
lin
k
ed
la
y
er
s
ar
e
em
p
lo
y
e
d
.
I
n
o
r
d
e
r
to
ef
f
ec
tiv
ely
id
en
tify
p
atter
n
s
a
s
s
o
ciate
d
with
T
B
,
Den
s
eNe
t
-
121
is
u
s
ed
o
n
a
lab
eled
d
ataset.
T
h
e
ar
ea
s
o
f
th
e
X
-
r
ay
[
2
1
]
t
h
at
ar
e
m
o
s
t
cr
u
cial
to
th
e
m
o
d
el'
s
co
m
p
letio
n
ar
e
t
h
en
d
is
p
lay
e
d
[
2
2
]
u
s
in
g
g
r
a
d
ien
t
-
weig
h
t
ed
class
ac
tiv
atio
n
m
ap
p
in
g
o
r
Gr
a
d
-
C
AM
[
2
3
]
.
Gr
ad
-
C
AM
d
r
aws
atten
tio
n
t
o
th
ese
ar
ea
s
to
m
ak
e
Den
s
eNe
t
-
121
'
s
f
o
r
ec
asts
ea
s
ier
to
u
n
d
er
s
tan
d
.
T
h
is
m
ac
h
in
e
f
ac
ilit
ates
p
r
ec
is
e
d
iag
n
o
s
is
[
2
4
]
,
b
o
o
s
ts
p
h
y
s
ician
tr
u
s
t
in
ar
tific
ial
in
tellig
en
ce
m
o
d
els,
an
d
e
n
ab
les
th
em
to
u
n
d
er
s
tan
d
t
h
e
lo
g
ic
u
n
d
er
ly
in
g
class
if
icatio
n
.
Fu
r
th
er
m
o
r
e
,
th
e
p
r
o
ce
s
s
o
f
d
etec
tin
g
an
d
d
iag
n
o
s
in
g
T
B
[
2
5
]
will
b
e
ac
ce
ler
ated
an
d
m
a
d
e
s
im
p
ler
b
y
u
s
in
g
d
ee
p
lear
n
i
n
g
a
n
d
ar
tific
ial
in
tellig
en
ce
.
T
h
e
m
ajo
r
ity
o
f
ea
r
lier
r
esear
ch
o
n
th
e
d
ia
g
n
o
s
is
an
d
d
etec
tio
n
o
f
T
B
[
2
6
]
was
co
n
ce
n
tr
ated
o
n
im
ag
e
p
r
o
ce
s
s
in
g
m
eth
o
d
s
,
u
s
in
g
MA
T
L
A
B
an
d
s
im
p
le
C
NN
n
etwo
r
k
s
as
a
d
etec
tio
n
m
o
d
el.
T
h
is
r
esear
ch
ar
ticle
is
d
i
v
id
ed
in
to
f
iv
e
s
ec
tio
n
s
:
s
ec
tio
n
1
in
tr
o
d
u
ce
s
th
e
d
is
ea
s
e
T
B
an
d
th
e
u
s
e
o
f
d
ee
p
lear
n
in
g
to
d
iag
n
o
s
e
it;
s
ec
tio
n
2
p
r
o
v
id
es
an
o
v
er
v
iew
o
f
ea
r
lier
r
esear
ch
th
a
t
is
r
elev
an
t
to
th
is
s
tu
d
y
;
s
ec
t
io
n
3
illu
s
tr
ates
th
e
s
u
g
g
este
d
m
eth
o
d
an
d
h
o
w
it
is
im
p
l
em
en
ted
;
s
ec
tio
n
4
s
h
o
ws
th
e
f
in
d
in
g
s
;
an
d
f
i
n
ally
,
s
ec
tio
n
5
p
r
o
v
id
es
th
e
s
u
m
m
ar
y
,
wh
ic
h
is
f
o
llo
wed
b
y
a
lis
t
o
f
r
ef
er
e
n
ce
s
.
T
h
e
r
esu
lts
o
f
th
is
wo
r
k
ar
e
as
f
o
llo
ws:
i)
T
B
is
d
etec
ted
f
r
o
m
ch
est
X
-
r
ay
i
m
ag
es
u
s
in
g
C
NN
,
ii)
th
e
tr
an
s
f
er
lear
n
in
g
tech
n
iq
u
e
k
n
o
w
n
as
Den
s
eNe
t
-
121
is
u
s
ed
,
wh
ich
im
p
r
o
v
es
class
if
icatio
n
ac
cu
r
ac
y
u
s
in
g
s
p
ec
if
ied
weig
h
ts
,
iii)
T
o
b
etter
d
is
p
lay
an
d
co
m
p
r
e
h
en
d
th
e
o
u
tp
u
t,
th
e
ex
p
lain
ab
le
ar
tific
ial
in
tellig
en
ce
(
XAI
)
[
2
7
]
ap
p
r
o
ac
h
k
n
o
wn
as
Gr
ad
-
C
A
M
is
em
p
lo
y
ed
.
2.
RE
L
AT
E
D
WO
RK
Pan
ick
er
et
a
l.
[
2
]
ex
p
lain
s
an
a
u
to
m
atic
m
eth
o
d
f
o
r
i
d
en
tify
in
g
T
B
b
ac
illi
f
r
o
m
p
ictu
r
es
o
f
m
icr
o
s
co
p
ic
s
p
u
tu
m
i
m
ag
es.
Acc
o
r
d
in
g
t
o
d
ata
f
r
o
m
t
h
e
W
o
r
ld
Hea
lth
Or
g
a
n
izatio
n
(
W
HO)
,
T
B
is
th
e
ten
th
m
o
s
t
p
r
ev
alen
t
ca
u
s
e
o
f
m
o
r
tality
g
lo
b
ally
.
Alth
o
u
g
h
th
e
r
e
a
r
e
s
ev
er
al
way
s
to
d
iag
n
o
s
e
TB
,
th
e
co
n
v
e
n
tio
n
al
m
icr
o
s
co
p
ic
a
n
aly
s
is
o
f
s
p
u
tu
m
s
m
ea
r
s
is
co
n
s
id
er
ed
th
e
s
tan
d
ar
d
tech
n
i
q
u
e.
T
h
e
d
iag
n
o
s
is
p
r
o
ce
d
u
r
e
is
lo
n
g
an
d
e
r
r
o
n
e
o
u
s
,
ev
en
wh
en
p
er
f
o
r
m
ed
b
y
p
r
o
f
ess
io
n
als.
T
h
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
f
o
r
T
B
id
e
n
tific
atio
n
ac
h
iev
es
8
6
.
7
6
%
-
F1
-
s
co
r
e
,
7
8
.
4
%
-
p
r
ec
is
io
n
,
an
d
9
7
.
1
3
%
-
r
e
ca
ll
b
ased
o
n
e
x
p
er
im
e
n
tal
d
a
ta.
T
h
is
au
t
o
m
atic
m
eth
o
d
tells
wh
eth
er
o
r
n
o
t th
e
s
p
u
tu
m
s
m
ea
r
p
ictu
r
es in
d
ic
ate
T
B
in
f
ec
tio
n
.
Kab
ir
et
a
l.
[
7
]
clar
if
ies
th
at
c
h
est
r
ad
io
g
r
ap
h
y
is
a
cr
u
cial
d
iag
n
o
s
tic
tech
n
iq
u
e
f
o
r
d
is
ea
s
es
lik
e
T
B
,
p
n
eu
m
o
n
ia,
an
d
C
OVI
D
-
1
9
b
ec
au
s
e
it
g
iv
es
a
r
ea
lis
tic
r
ep
r
esen
tatio
n
o
f
t
h
e
ar
ch
ite
ctu
r
e
o
f
th
e
ch
est.
Ho
wev
er
,
ef
f
ec
tiv
ely
i
d
en
tify
i
n
g
th
ese
illn
ess
es
f
r
o
m
r
ad
i
o
g
r
ap
h
s
is
a
c
h
allen
g
in
g
task
th
at
r
eq
u
ir
es
m
ed
ical
im
ag
in
g
tec
h
n
o
lo
g
y
.
C
o
n
v
en
t
io
n
al
d
ee
p
lear
n
i
n
g
m
o
d
els
o
f
f
er
a
p
r
ac
tical
au
t
o
m
ated
s
o
lu
tio
n
f
o
r
th
is
is
s
u
e.
Ho
wev
er
,
b
ec
au
s
e
th
ese
m
o
d
e
ls
ar
e
s
o
s
o
p
h
is
ticated
,
th
eir
p
r
ac
tical
d
ep
lo
y
m
e
n
t
in
m
ed
ical
ap
p
licatio
n
s
o
f
ten
en
co
u
n
ter
s
s
ig
n
if
ican
t
o
b
s
tacle
s
.
B
y
u
s
in
g
k
n
o
wled
g
e
d
is
till
atio
n
tech
n
i
q
u
es
(
KDT
)
to
les
s
en
th
e
co
m
p
lex
ity
o
f
C
NN,
th
is
wo
r
k
ad
d
r
ess
es a
n
d
r
eso
lv
es th
is
p
u
zz
le.
Hass
an
et
a
l.
[
1
9
]
ad
d
s
th
at
d
ee
p
lear
n
in
g
tech
n
i
q
u
es
f
o
r
k
n
ee
o
s
teo
ar
th
r
itis
(
KOA
)
d
ete
ctio
n
h
av
e
b
ee
n
m
o
r
e
p
r
o
m
in
en
t
i
n
r
ec
e
n
t
y
ea
r
s
.
A
ls
o
,
h
o
w
to
cr
ea
te
a
d
ee
p
lear
n
in
g
m
o
d
el
f
o
r
K
OA
d
etec
tio
n
u
s
in
g
k
n
ee
X
-
r
a
y
im
ag
es
an
d
th
e
K
ellg
r
en
–
L
awr
en
ce
(
KL
)
g
r
a
d
in
g
s
y
s
tem
.
T
h
e
k
n
ee
o
s
teo
ar
t
h
r
itis
class
if
icatio
n
n
etwo
r
k
(
KOC_
Net)
,
a
n
o
v
el
C
NN
-
b
ased
m
o
d
el,
is
p
r
o
p
o
s
ed
in
th
is
p
ap
er
.
T
wo
p
u
b
licly
ac
ce
s
s
ib
le
b
en
ch
m
ar
k
d
atasets
co
m
p
r
is
in
g
X
-
r
ay
im
ag
es
o
f
KOA
b
ased
o
n
th
e
KL
g
r
ad
in
g
s
y
s
tem
ar
e
u
s
ed
to
ass
es
s
th
e
KOC_
Net
m
o
d
el.
Ad
d
itio
n
ally
,
we
u
s
ed
SMOT
E
T
o
m
ek
to
s
o
lv
e
th
e
is
s
u
e
o
f
m
in
o
r
ity
cl
ass
es
an
d
co
n
tr
ast
-
lim
ited
ad
ap
tiv
e
h
is
to
g
r
am
eq
u
aliza
tio
n
(
C
L
AHE
)
tech
n
iq
u
es.
W
ith
an
ar
ea
u
n
d
e
r
th
e
cu
r
v
e
(
AUC
)
-
9
6
.
7
1
%,
ac
cu
r
ac
y
-
9
6
.
5
1
%,
r
ec
all
-
9
1
.
9
5
%,
p
r
ec
is
io
n
-
9
0
.
2
5
%,
an
d
F
1
-
s
co
r
e
o
f
9
6
.
7
0
%,
th
e
s
u
g
g
e
s
ted
KOC_
Net
was
ab
le
to
ca
teg
o
r
ize
KOA
in
to
f
i
v
e
d
if
f
er
e
n
t g
r
o
u
p
s
.
R
o
n
y
et
a
l.
[
2
7
]
b
r
ief
s
th
at
,
i
n
o
r
d
er
to
im
p
r
o
v
e
th
e
p
r
ec
is
io
n
an
d
d
ep
en
d
ab
ilit
y
o
f
au
tis
m
s
p
ec
tr
u
m
d
is
o
r
d
er
(
ASD
)
d
ia
g
n
o
s
is
,
th
i
s
s
tu
d
y
u
s
ed
s
o
p
h
is
ticated
m
ac
h
in
e
-
lear
n
in
g
alg
o
r
ith
m
s
.
W
e
u
s
ed
a
s
tan
d
ar
d
d
ataset
with
2
0
v
ar
iab
les
an
d
1
,
0
5
4
p
atien
t
s
am
p
les.
At
9
9
%,
th
e
s
u
g
g
ested
d
iab
etes
m
ellitu
s
an
d
lo
g
is
tic
r
eg
r
ess
io
n
with
Sh
ap
ley
ad
d
it
iv
e
ex
p
lan
atio
n
s
(
DM
L
R
S)
m
o
d
el
o
u
tp
e
r
f
o
r
m
ed
cu
ttin
g
-
e
d
g
e
tech
n
iq
u
es.
T
o
im
p
r
o
v
e
i
n
ter
p
r
etab
ilit
y
,
Sh
a
p
ley
ad
d
iti
v
e
ex
p
lan
atio
n
s
(
SHAP)
wer
e
u
s
ed
to
in
teg
r
ate
X
AI
.
E
ac
h
tech
n
i
q
u
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
A
n
a
lysi
s
o
f tu
b
ercu
lo
s
is
d
etec
tio
n
u
s
in
g
d
ee
p
lea
r
n
in
g
tech
n
i
q
u
e
a
n
d
ex
p
l
a
in
a
b
le
…
(
S
h
a
s
h
ikir
a
n
S
r
in
iva
s
)
1625
was
r
ef
in
ed
,
an
d
k
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
was
u
s
ed
to
co
n
f
i
r
m
p
e
r
f
o
r
m
an
ce
.
Ad
d
itio
n
all
y
,
a
r
ea
l
-
tim
e
we
b
ap
p
licatio
n
th
at
co
m
b
in
es th
e
Djan
g
o
f
r
a
m
ewo
r
k
f
o
r
ASD
d
etec
tio
n
with
th
e
DM
L
R
S m
o
d
el
was c
r
ea
ted
.
Sh
ah
s
h
ik
ir
an
et
a
l.
[
2
8
]
b
r
ie
f
s
th
at,
Myco
b
a
cteriu
m
ca
u
s
es
T
B
,
a
d
ea
d
ly
illn
ess
.
Ma
n
ag
in
g
T
B
in
f
ec
tio
n
s
r
eq
u
ir
es
ea
r
ly
d
ete
ctio
n
an
d
id
e
n
tific
atio
n
o
f
t
h
e
d
is
ea
s
e.
As
th
is
s
tu
d
y
s
h
o
ws,
r
ec
en
t
tech
n
ical
ad
v
an
ce
m
e
n
ts
em
p
lo
y
a
m
ac
h
in
e
lear
n
i
n
g
(
ML
)
-
b
ased
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM
)
a
n
d
C
NN
,
wh
ich
is
m
o
d
if
ied
to
m
o
r
e
p
r
ec
is
ely
d
etec
t
p
ar
ticu
lar
d
is
ea
s
es.
T
h
r
o
u
g
h
o
u
t
co
n
s
tr
u
ctio
n
,
th
e
en
h
an
ce
d
f
ea
tu
r
e
ex
tr
ac
tio
n
an
d
class
if
icatio
n
ac
cu
r
ac
y
o
f
th
e
u
p
d
ated
C
NN
ar
e
p
r
eser
v
e
d
.
T
h
e
T
B
X1
1
K
p
u
b
licly
av
ailab
l
e
d
ataset,
wh
ich
in
clu
d
es
1
1
,
0
0
0
im
ag
es
—
4
,
6
0
0
o
f
wh
ich
ar
e
ch
est
X
-
r
ay
im
ag
es
—
is
u
s
e
d
to
ac
h
iev
e
g
o
o
d
p
er
f
o
r
m
an
ce
,
an
d
th
e
p
r
o
p
o
s
e
d
m
o
d
el
is
v
alid
ated
.
T
h
e
ac
c
u
r
ac
y
o
f
T
B
id
en
tific
atio
n
u
s
in
g
SVM
is
9
3
.
1
4
%,
wh
ile
th
e
ac
cu
r
ac
y
u
s
in
g
m
o
d
i
f
ied
C
NN
is
9
6
.
7
2
%.
3.
M
E
T
H
O
D
Sev
er
al
ess
en
tial
p
r
o
ce
d
u
r
es
ar
e
ty
p
ically
in
v
o
lv
ed
in
d
ee
p
lear
n
in
g
-
b
ased
p
r
o
ject
-
r
elate
d
wo
r
k
f
o
r
T
B
d
iag
n
o
s
is
.
T
h
e
b
lo
ck
d
iag
r
am
f
o
r
th
e
ap
p
r
o
ac
h
em
p
lo
y
ed
is
s
h
o
wn
i
n
Fig
u
r
e
1
.
I
t
is
m
ad
e
u
p
o
f
s
ev
er
al
b
lo
ck
s
th
at
u
s
e
th
e
ch
e
s
t
X
-
r
a
y
d
ataset'
s
im
ag
es
as
in
p
u
t,
a
p
p
ly
a
tr
an
s
f
er
lear
n
i
n
g
m
o
d
e
l,
an
d
th
en
class
if
y
th
e
q
u
er
y
im
ag
e
u
s
in
g
th
e
in
f
o
r
m
atio
n
lear
n
e
d
f
r
o
m
th
e
m
o
d
el's tr
ain
in
g
.
Fig
u
r
e
1
.
B
lo
ck
d
iag
r
am
f
o
r
t
h
e
p
r
o
p
o
s
ed
tech
n
i
q
u
e
3
.
1
.
Da
t
a
s
et
a
v
a
ila
bil
it
y
a
nd
prepa
ra
t
io
n
L
o
ad
in
g
m
eta
d
ata
an
d
im
ag
es
:
in
itially
,
X
-
r
ay
im
ag
es
an
d
th
e
ass
o
ciate
d
m
etad
ata
wer
e
all
o
v
er
th
e
p
lace
.
T
h
e
co
llectio
n
c
o
n
tain
s
two
ty
p
es
o
f
c
h
est
X
-
r
ay
im
ag
es:
n
o
r
m
al
an
d
T
B
-
in
f
ec
ted
.
W
e
wer
e
ab
le
t
o
d
is
tin
g
u
is
h
b
etwe
en
h
ea
lth
y
p
eo
p
le
an
d
TB
p
atien
ts
s
in
ce
th
e
d
ataset
was
p
r
e
-
lab
eled
.
Fig
u
r
e
2
s
h
o
ws
th
e
s
am
p
le
o
f
T
B
im
ag
es
o
f
d
ataset,
wh
er
e
u
n
in
f
ec
ted
a
r
e
in
Fig
u
r
e
2
(
a
)
an
d
in
f
ec
te
d
ar
e
in
th
e
Fig
u
r
e
2
(
b
)
.
D
ata
s
eg
m
en
tatio
n
f
o
r
tr
ain
in
g
,
v
a
lid
atio
n
,
a
n
d
test
in
g
:
th
e
f
o
ll
o
win
g
im
a
g
e
d
is
tr
ib
u
tio
n
is
p
r
o
v
id
e
d
t
o
e
n
s
u
r
e
co
n
s
is
ten
t
m
o
d
el
p
er
f
o
r
m
an
c
e
:
th
e
n
atio
n
al
in
s
titu
te
o
f
all
er
g
y
a
n
d
in
f
ec
tio
u
s
d
is
ea
s
es
(
NI
AI
D)
T
B
p
o
r
tal
p
r
o
g
r
a
m
d
ataset
was
u
tili
ze
d
in
th
is
s
tu
d
y
f
o
r
t
h
e
class
if
icatio
n
s
tag
e
o
f
th
e
e
x
p
er
im
e
n
t.
T
h
e
"
tu
b
er
cu
lo
s
is
"
Kag
g
le
p
r
o
ject'
s
"c
h
es
t
X
-
r
ay
p
ictu
r
es
d
ataset",
wh
ich
co
n
ta
in
ed
7
0
0
T
B
im
ag
es
an
d
3
,
5
0
0
n
o
r
m
al
im
ag
es,
is
wh
er
e
th
is
d
ataset
o
r
ig
in
ated
.
All
7
0
0
T
B
o
f
th
e
co
n
tr
o
l
ch
est
X
-
r
ay
im
ag
es
wer
e
u
tili
ze
d
.
W
ith
to
tal
im
ag
es
b
ein
g
4
,
2
0
0
th
e
s
p
lit
r
atio
=0
.
2
.
T
h
er
e
f
o
r
e,
tr
ai
n
in
g
im
ag
es
=(
2
,
8
0
0
n
o
r
m
al+
5
6
0
T
B
im
ag
es)
an
d
test
im
ag
es =
(
7
0
0
n
o
r
m
al+
1
4
0
T
B
im
ag
es).
(
a)
(
b
)
Fig
u
r
e
2
.
Sam
p
le
im
a
g
es f
ig
u
r
e
f
o
r
(
a
)
T
B
im
ag
es
-
u
n
i
n
f
ec
te
d
an
d
(
b
)
T
B
im
ag
es
-
in
f
ec
te
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
2
,
Ap
r
il 2
0
2
6
:
1
6
2
3
-
1
6
3
1
1626
3
.
2
.
Da
t
a
prepro
ce
s
s
ing
3
.
2
.
1
.
G
ener
a
t
ing
t
r
a
in da
t
a
T
h
e
im
ag
es
wer
e
s
ca
led
,
n
o
r
m
alize
d
,
an
d
im
p
r
o
v
e
d
u
s
in
g
r
o
tatio
n
,
m
a
g
n
if
icatio
n
,
an
d
f
lip
p
i
n
g
alg
o
r
ith
m
s
to
im
p
r
o
v
e
g
e
n
er
a
lizatio
n
in
o
r
d
er
to
f
it
in
to
th
e
n
etwo
r
k
d
esig
n
.
A
tr
ain
g
e
n
er
ato
r
was
b
u
ilt
to
s
wif
tly
im
p
o
r
t th
e
tr
ain
in
g
d
at
a
an
d
p
e
r
f
o
r
m
o
n
-
th
e
-
s
p
o
t
p
r
e
p
r
o
ce
s
s
in
g
.
−
No
r
m
alize
th
e
m
ea
n
an
d
s
tan
d
ar
d
d
ev
iatio
n
(
SD)
o
f
ev
er
y
d
a
ta
s
et.
−
Sh
u
f
f
le
th
e
in
p
u
t a
f
ter
ea
ch
ep
o
ch
.
−
Ver
if
y
th
at
th
e
im
ag
e
is
3
2
0
b
y
3
2
0
p
ix
els.
T
o
b
alan
ce
co
m
p
u
tatio
n
al
s
p
ee
d
an
d
d
iag
n
o
s
tic
ac
cu
r
ac
y
,
all
ch
est X
-
r
ay
p
ictu
r
es we
r
e
m
o
d
if
ied
to
3
2
0
×3
2
0
p
ix
els.
−
I
m
p
lem
en
t
s
o
m
e
m
o
d
if
icatio
n
s
(
r
o
tatio
n
,
zo
o
m
,
wid
t
h
s
h
if
t,
an
d
h
eig
h
t
s
h
if
t
)
b
ased
o
n
p
o
s
itio
n
d
ev
iatio
n
,
wh
ic
h
m
ay
c
h
an
g
e
s
lig
h
tly
wh
en
r
ad
i
o
lo
g
is
ts
X
-
r
ay
p
atien
ts
.
−
I
n
th
e
n
e
x
t
s
tep
,
it
h
ap
p
en
s
in
s
id
e
th
e
g
en
e
r
ato
r
to
o
l
-
g
r
a
y
s
ca
le
X
-
r
ay
p
ictu
r
es
g
ain
d
e
p
th
w
h
en
th
eir
d
ata
f
ills
ea
ch
o
f
t
h
e
3
co
lo
r
p
ath
s
.
Sin
ce
th
e
r
ea
d
y
-
m
ad
e
m
o
d
e
l
o
n
ly
w
o
r
k
s
with
3
-
lay
er
in
p
u
ts
,
th
is
s
h
if
t
m
ak
es it r
u
n
.
3
.
2
.
2
.
G
ener
a
t
ing
t
est
a
nd
v
a
lid
a
t
io
n da
t
a
T
h
e
s
am
e
d
ata
wer
e
cr
ea
ted
f
o
r
th
e
test
an
d
v
alid
atio
n
d
atasets
wi
th
o
u
t
an
y
au
g
m
en
tatio
n
s
in
o
r
d
er
to
p
r
eser
v
e
th
e
in
teg
r
ity
o
f
th
e
ev
alu
atio
n
p
r
o
ce
s
s
.
T
o
v
e
r
if
y
th
at
th
e
p
r
ep
r
o
ce
s
s
in
g
was
d
o
n
e
co
r
r
ec
tly
,
a
n
o
r
m
alize
d
p
ictu
r
e
s
am
p
le
f
r
o
m
th
e
tr
ain
in
g
g
e
n
er
ato
r
was
d
is
p
lay
ed
.
T
h
e
i
m
ag
es
h
a
d
p
i
x
el
v
alu
es
b
etwe
en
0
an
d
1
,
an
d
th
ey
wer
e
r
esized
to
m
atch
th
e
m
o
d
el'
s
r
eq
u
ir
ed
in
p
u
t sh
ap
e.
−
W
e
n
o
r
m
alize
in
co
m
in
g
test
an
d
v
alid
atio
n
d
ata
u
s
in
g
th
e
s
tatis
tics
e
s
tim
ated
f
r
o
m
th
e
tr
ai
n
in
g
s
et.
−
T
o
s
av
e
co
m
p
u
tatio
n
al
tim
e,
we
co
m
p
u
ted
th
e
s
am
p
le
m
ea
n
(
SM)
an
d
s
am
p
le
SD
u
s
in
g
a
r
an
d
o
m
s
am
p
le
f
r
o
m
th
e
d
ataset
(
id
ea
lly
,
th
e
f
u
ll
tr
ain
in
g
s
et
s
h
o
u
l
d
b
e
u
s
ed
f
o
r
ca
lc
u
latin
g
th
e
SM
an
d
SD)
.
Fig
u
r
e
3
illu
s
tr
ates
th
e
u
n
ev
en
d
ata
d
is
tr
ib
u
tio
n
b
ef
o
r
e
im
ag
e
au
g
m
e
n
tatio
n
.
Ad
d
itio
n
ally
,
Fig
u
r
e
4
d
is
p
lay
s
a
b
alan
ce
d
d
ata
d
is
tr
ib
u
tio
n
f
o
llo
win
g
im
ag
e
a
u
g
m
en
tatio
n
.
T
h
e
s
am
p
le
weig
h
tin
g
f
ix
f
o
r
th
e
lo
s
s
f
u
n
ctio
n
.
C
r
ea
tin
g
+v
e
(
p
o
s
itiv
e)
an
d
-
v
e
(
n
eg
ativ
e
)
co
n
t
r
ib
u
to
r
s
.
Fig
u
r
e
3
.
I
m
b
alan
ce
d
ata
(
u
n
e
v
en
d
ata)
Fig
u
r
e
4
.
B
alan
ce
d
ata
(
ev
en
d
ata)
T
h
is
s
tu
d
y
'
s
weig
h
ted
lo
s
s
f
o
r
m
u
latio
n
b
ala
n
ce
s
p
o
s
itiv
e
an
d
n
eg
ativ
e
co
n
tr
i
b
u
tio
n
s
,
as
i
n
d
icate
d
in
(
1
)
an
d
(
2
)
.
T
h
e
f
r
eq
u
e
n
cy
o
f
n
eg
ativ
e
a
n
d
p
o
s
itiv
e
s
am
p
le
s
d
eter
m
in
es
th
e
p
o
s
itiv
e
-
class
weig
h
t
(
W
p
)
an
d
n
eg
ativ
e
-
class
weig
h
t
(
W
n
)
,
r
esp
ec
tiv
ely
.
T
h
ese
weig
h
ts
wer
e
th
en
ad
d
e
d
to
th
e
weig
h
ted
cr
o
s
s
-
en
tr
o
p
y
lo
s
s
d
escr
ib
ed
in
(
3
)
.
T
h
e
f
o
llo
wi
n
g
ca
lcu
latio
n
p
r
o
v
id
es
th
e
f
i
n
al
weig
h
t
lo
s
s
to
b
e
u
s
ed
in
th
e
Den
s
eNe
t
-
121
ar
ch
itectu
r
e
as in
(
3
)
.
×
=
×
(
1
)
=
&
=
(
2
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
A
n
a
lysi
s
o
f tu
b
ercu
lo
s
is
d
etec
tio
n
u
s
in
g
d
ee
p
lea
r
n
in
g
tech
n
i
q
u
e
a
n
d
ex
p
l
a
in
a
b
le
…
(
S
h
a
s
h
ikir
a
n
S
r
in
iva
s
)
1627
=
−
(
.
.
l
og
(
(
)
)
+
(
1
−
)
l
og
(
1
−
(
)
(
3
)
T
o
ad
d
r
ess
th
e
q
u
esti
o
n
co
n
ce
r
n
in
g
em
p
ir
ical
v
alid
atio
n
,
we
r
a
n
co
m
p
ar
is
o
n
test
s
o
n
m
o
d
el
p
er
f
o
r
m
an
ce
with
o
u
t
an
d
with
weig
h
ted
l
o
s
s
f
u
n
ctio
n
s
.
T
h
e
n
ew
d
ata
clea
r
l
y
ex
p
lai
n
th
at
th
e
weig
h
tin
g
tech
n
iq
u
e
im
p
r
o
v
e
s
s
en
s
itiv
ity
,
F1
-
s
co
r
e,
an
d
Ma
tth
ews
co
r
r
elatio
n
c
o
ef
f
ici
en
t
(
MCC
)
f
o
r
th
e
TB
-
p
o
s
itiv
e
class
wh
ile
al
s
o
lo
wer
in
g
th
e
f
alse
-
n
eg
ativ
e
r
ate,
wh
ich
is
an
im
p
o
r
tan
t
in
d
icato
r
o
f
r
ed
u
ce
d
m
ajo
r
ity
-
class
b
ias.
T
h
e
tr
an
s
f
o
r
m
er
-
d
ep
e
n
d
en
t
d
en
o
is
in
g
m
eth
o
d
o
l
o
g
y
[
2
7
]
tak
es
a
m
et
h
o
d
ical
ap
p
r
o
ac
h
to
p
r
eser
v
in
g
d
ia
g
n
o
s
tic
d
etail
wh
ile
r
ed
u
cin
g
ac
q
u
is
itio
n
-
r
e
lated
n
o
is
e,
p
o
ten
tially
in
cr
e
asin
g
th
e
q
u
ality
o
f
Den
s
eNe
t
-
121
'
s
ch
est X
-
r
ay
in
p
u
ts
.
3
.
3
.
T
ra
ns
f
er
lea
rning
inte
g
ra
t
io
n wit
h
deep
lea
rning
(
DenseNet
-
121
+
CNN)
Den
s
eNe
t
-
121
,
a
p
o
wer
f
u
l
C
NN
p
r
e
-
tr
ain
ed
o
n
th
e
I
m
ag
eNe
t
d
ataset,
f
o
r
tr
a
n
s
f
er
lea
r
n
in
g
.
B
y
lev
er
ag
in
g
p
r
e
-
lear
n
e
d
f
ea
tu
r
es,
th
e
m
o
d
el
g
ain
s
th
e
p
o
ten
tial
to
en
h
an
ce
ac
cu
r
ac
y
an
d
f
aster
co
n
v
er
g
e
n
ce
with
a
s
m
aller
d
ataset.
T
h
e
n
u
m
b
er
"
1
2
1
"
i
n
d
icate
s
th
at
th
e
tr
an
s
f
er
lear
n
i
n
g
m
o
d
el
in
i
s
s
u
e
h
as
1
2
1
lay
er
s
.
T
h
e
m
ath
em
atica
l
b
r
ea
k
d
o
wn
o
f
De
n
s
eNe
t
-
1
2
1
is
as
f
o
l
lo
ws:
co
m
b
in
ed
f
u
n
ctio
n
:
ea
ch
lay
er
a
p
p
lies
a
co
m
p
o
s
ite
f
u
n
ctio
n
th
at
in
clu
d
es
as in
(
4
)
.
(
)
=
(
[
0
,
1
.
.
−
1
]
)
(
4
)
W
h
er
e
(
)
–
ℎ
lay
er
o
u
tp
u
t
,
–
ℎ
lay
er
weig
h
ts
,
[
0
,
1
.
.
−
1
]
–
f
ea
tu
r
e
m
a
p
s
.
T
r
an
s
itio
n
lay
er
:
Den
s
eNe
t
u
s
es
a
tr
an
s
itio
n
lay
e
r
th
at
p
er
f
o
r
m
s
a
c
o
n
v
o
lu
tio
n
an
d
th
en
p
o
o
lin
g
to
r
eg
u
late
th
e
s
ize
o
f
f
ea
tu
r
e
m
a
p
s
.
T
h
e
f
o
llo
win
g
is
th
e
m
ath
e
m
atica
l e
x
p
r
ess
io
n
f
o
r
th
e
tr
an
s
itio
n
lay
er
as (
5
)
.
(
)
=
(
(
(
)
)
)
(
5
)
W
h
er
e
(
)
–
ℎ
lay
er
o
u
tp
u
t
,
–
ℎ
lay
er
weig
h
ts
,
–
b
atch
n
o
r
m
aliza
tio
n
.
A
s
in
g
le
lay
er
’
s
o
u
tp
u
t
s
ize
d
ep
en
d
s
o
n
h
o
w
m
an
y
f
ea
tu
r
e
m
ap
s
it
cr
ea
tes.
T
h
is
am
o
u
n
t
clim
b
s
with
ea
ch
s
tep
b
ec
au
s
e
o
f
a
s
ettin
g
ca
lled
th
e
g
r
o
wth
r
ate,
lab
eled
k
.
E
ac
h
n
ew
lay
er
ad
d
s
ex
ac
tly
k
m
o
r
e
m
ap
p
in
g
s
th
an
th
e
o
n
e
b
ef
o
r
e.
T
h
e
p
atter
n
co
n
tin
u
es c
o
n
s
is
ten
tly
th
r
o
u
g
h
th
e
n
etwo
r
k
a
n
d
is
g
iv
e
n
as
(
6
)
.
=
0
+
.
(
6
)
W
h
er
e:
–
th
lay
er
o
u
tp
u
t
f
ea
t
u
r
e
m
ap
s
,
0
–
1
st
f
ea
tu
r
e
m
a
p
s
.
Den
s
e
b
lo
ck
(
DB
)
:
Den
s
eNe
t
-
1
2
1
c
o
n
s
is
ts
o
f
f
o
u
r
DB
s
,
ea
ch
with
a
d
is
tin
ct
n
u
m
b
er
o
f
lay
er
s
.
T
r
an
s
itio
n
lay
er
f
o
llo
ws
ev
er
y
b
lo
ck
,
with
th
e
ex
ce
p
tio
n
o
f
th
e
last
o
n
e
.
T
h
e
ex
a
ct
lay
er
co
u
n
ts
o
f
Den
s
eNe
t
-
1
2
1
ar
e
as
f
o
llo
ws:
s
ix
lay
er
s
m
ak
e
u
p
DB
1
,
twe
lv
e
in
DB
2
,
twen
ty
-
f
o
u
r
in
D
B
3
,
an
d
s
ix
teen
in
DB
4
.
T
h
e
f
in
al
D
en
s
e
N
et
-
1
2
1
f
o
r
m
u
la
is
s
h
o
wn
as
(
7
)
,
with
f
–
DB
,
t
–
tr
an
s
itio
n
la
y
er
,
a
n
d
P
–
o
u
tp
u
t
f
o
llo
win
g
g
l
o
b
al
av
e
r
ag
e
p
o
o
l
in
g
.
=
121
(
)
=
4
(
4
(
3
(
3
(
2
(
1
(
1
(
)
)
)
)
)
)
(
7
)
3
.
4
.
E
x
pla
ina
ble
a
rt
if
icia
l in
t
ellig
ence
(
G
ra
d
-
CAM)
T
h
e
m
o
d
el
p
ay
s
atten
tio
n
to
a
T
B
im
ag
e
th
at
co
m
es
clea
r
th
r
o
u
g
h
Gr
ad
-
C
AM
.
On
e
lay
er
at
a
tim
e
g
ets p
ick
ed
b
y
th
e
s
y
s
tem
to
m
ap
o
u
t w
h
at
m
atter
s
f
o
r
s
o
r
tin
g
im
ag
es in
to
ty
p
es.
Me
d
ical
p
ictu
r
es g
ain
clar
ity
s
in
ce
th
ese
m
ap
s
h
ig
h
lig
h
t
s
p
o
ts
lik
ely
a
f
f
ec
ted
.
I
n
f
ec
tio
n
zo
n
es
s
tan
d
o
u
t
wh
er
e
co
lo
r
in
ten
s
if
ies
o
n
th
e
o
v
er
lay
.
Ou
r
s
etu
p
u
s
es
Gr
ad
-
C
AM
to
s
k
etch
wh
er
e
atten
tio
n
s
h
o
u
ld
g
o
,
s
h
ap
e
d
b
y
s
h
if
ts
ac
r
o
s
s
lay
er
s
.
E
m
er
g
i
n
g
XAI
m
eth
o
d
s
[
2
9
]
in
m
ed
ical
f
ield
s
.
T
h
eir
em
p
lo
y
m
en
t
o
f
c
o
m
p
lem
e
n
tar
y
e
x
p
lain
ab
ilit
y
m
ec
h
an
is
m
s
illu
s
tr
ates
th
e
p
o
te
n
tial
b
en
e
f
it
o
f
c
o
m
b
in
in
g
ad
v
a
n
ce
d
v
is
u
aliza
tio
n
tec
h
n
iq
u
es
lik
e
Gr
ad
-
C
AM
++
,
Sco
r
e
-
C
AM
,
o
r
atten
tio
n
-
b
ased
in
ter
p
r
etab
ili
ty
to
p
r
o
v
id
e
m
o
r
e
g
r
an
u
lar
a
n
d
clin
ically
u
s
ef
u
l
ex
p
lan
atio
n
s
in
TB
ca
teg
o
r
izatio
n
.
4.
RE
SU
L
T
S
AND
D
I
SCU
SS
I
O
N
C
las
s
if
icatio
n
r
ep
o
r
t:
−
Pre
cisi
o
n
:
th
e
tr
u
e
p
o
s
itiv
e
(
T
P)
class
if
icatio
n
r
atio
to
t
h
e
to
tal
n
u
m
b
e
r
o
f
TP
+f
alse
p
o
s
itiv
es
(
FP
)
is
k
n
o
wn
as p
r
ec
is
io
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
2
,
Ap
r
il 2
0
2
6
:
1
6
2
3
-
1
6
3
1
1628
=
+
(
8
)
−
R
ec
all
:
r
ec
all
is
e
s
s
en
tial
to
d
eter
m
in
e
ca
p
ac
ity
o
f
th
e
d
ee
p
lear
n
in
g
m
o
d
el
to
d
etec
t
p
o
s
i
tiv
e
s
am
p
les.
C
o
m
p
ar
ed
to
s
en
s
itiv
ity
,
r
ec
all
is
co
n
ce
r
n
ed
o
n
ly
with
h
o
w
p
o
s
itiv
es a
r
e
class
if
ied
.
=
+
(
9
)
−
F1
-
s
co
r
e:
an
o
th
e
r
co
m
m
o
n
n
a
m
e
f
o
r
F1
-
s
co
r
e
–
F
m
ea
s
u
r
e.
T
o
ac
h
iev
e
th
e
b
ala
n
ce
in
b
et
wee
n
p
r
ec
is
io
n
an
d
r
ec
all,
we
n
ee
d
a
lar
g
er
n
u
m
b
er
o
f
tr
u
e
n
eg
ativ
es
(
T
N)
an
d
TP
in
th
e
m
o
d
el,
a
n
d
th
e
F1
-
s
co
r
e
allo
ws u
s
to
ch
o
o
s
e
th
e
id
ea
l c
o
n
f
id
en
ce
lev
el.
1
−
=
2
(
∗
)
+
(
1
0
)
−
Acc
u
r
ac
y
:
th
e
s
u
m
o
f
TN
a
n
d
TP
class
if
icatio
n
r
atio
to
th
e
to
tal
in
s
tan
ts
.
=
+
+
+
+
(
1
1
)
W
i
t
h
i
n
t
h
e
c
o
n
f
u
s
i
o
n
m
a
t
r
i
x
(
C
M
)
a
r
e
f
o
u
r
v
a
l
u
e
s
:
TP
,
TN
,
FP
,
a
n
d
f
al
s
e
n
e
g
a
t
i
v
es
(
FN
)
.
W
i
t
h
c
o
u
n
ts
s
t
a
n
d
i
n
g
a
t
4
6
3
f
o
r
T
P
,
a
l
o
n
g
s
i
d
e
4
2
9
f
o
r
T
N
,
f
a
ls
e
o
u
t
c
o
m
e
s
a
p
p
e
a
r
l
ess
o
f
t
e
n
;
s
p
e
c
i
f
ic
a
l
l
y
,
FP
r
e
a
c
h
es
1
2
,
w
h
i
l
e
FN
s
e
tt
l
es
a
t
6
.
A
c
c
u
r
a
cy
m
e
a
s
u
r
e
s
9
7
.
4
2
p
e
r
c
e
n
t
,
s
i
n
ce
each
p
r
e
c
i
s
i
o
n
n
e
a
r
s
9
7
.
4
%
a
n
d
r
e
c
a
l
l
c
l
i
m
b
s
t
o
9
8
.
7
%
,
a
n
d
h
e
n
c
e
t
h
e
F
1
-
s
c
o
r
e
a
l
i
g
n
s
cl
o
s
el
y
a
t
9
8
.
0
%
.
At
t
h
e
s
e
v
e
n
t
e
e
n
t
h
t
r
a
i
n
i
n
g
c
y
c
le
,
F
i
g
u
r
e
5
d
i
s
p
l
a
y
s
C
N
N
a
c
h
i
e
v
i
n
g
e
x
a
ct
l
y
9
7
.
4
2
%
a
c
c
u
r
a
c
y
.
M
ea
n
w
h
i
l
e
,
l
o
s
s
d
r
o
p
s
n
o
t
a
b
l
y
b
y
t
h
a
t
s
t
a
g
e
–
F
ig
u
r
e
6
r
e
c
o
r
d
s
i
t
a
s
l
o
w
s
as
0
.
0
7
1
7
w
h
e
n
e
p
o
c
h
c
o
u
n
t
h
i
t
s
tw
e
n
t
y
.
Gr
ad
-
C
A
M
h
ea
tm
ap
s
d
eter
m
i
n
e
wh
ich
lu
n
g
r
eg
io
n
s
h
av
e
th
e
g
r
ea
test
im
p
ac
t
o
n
a
m
o
d
el'
s
d
iag
n
o
s
is
f
o
r
T
B
d
etec
tio
n
.
T
h
e
m
o
d
el'
s
f
o
cu
s
o
n
a
b
n
o
r
m
al
ar
ea
s
,
s
u
ch
as
lesi
o
n
s
o
r
s
ick
tis
s
u
e,
ca
n
b
e
clea
r
ly
s
ee
n
b
y
s
u
p
er
im
p
o
s
in
g
th
e
h
ea
tm
a
p
o
n
ch
est
X
-
r
ay
s
.
Fig
u
r
e
7
illu
s
tr
ates
th
e
Gr
ad
-
C
AM
o
v
er
lay
th
o
s
e
r
esu
lts
f
r
o
m
s
u
p
er
im
p
o
s
in
g
t
h
e
Gr
ad
-
C
AM
h
ea
tm
ap
o
v
er
th
e
T
B
s
am
p
le.
Fig
u
r
e
5
.
Acc
u
r
ac
y
v
s
.
e
p
o
ch
p
lo
t
Fig
u
r
e
6
.
L
o
s
s
v
s
.
ep
o
ch
p
lo
t
Fig
u
r
e
7
.
Gr
a
d
-
C
AM
h
ea
t m
a
p
with
r
esu
lt o
v
er
la
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
A
n
a
lysi
s
o
f tu
b
ercu
lo
s
is
d
etec
tio
n
u
s
in
g
d
ee
p
lea
r
n
in
g
tech
n
i
q
u
e
a
n
d
ex
p
l
a
in
a
b
le
…
(
S
h
a
s
h
ikir
a
n
S
r
in
iva
s
)
1629
5.
CO
NCLU
SI
O
N
A
tr
ain
ed
Den
s
eNe
t
-
1
2
1
f
o
r
m
s
th
e
b
ase
o
f
th
is
s
tu
d
y
’
s
ap
p
r
o
ac
h
to
id
e
n
tify
in
g
T
B
with
i
n
ch
est
X
-
r
ay
s
.
I
n
s
tead
o
f
b
u
ild
in
g
f
r
o
m
s
cr
atch
,
p
r
io
r
k
n
o
wl
ed
g
e
g
u
i
d
es
ad
ap
tatio
n
t
h
r
o
u
g
h
tr
an
s
f
er
lear
n
in
g
,
s
h
ap
in
g
d
etec
tio
n
ca
p
ab
ilit
y
.
Atten
tio
n
s
h
if
ts
to
war
d
k
ey
r
e
g
io
n
s
em
er
g
e
v
ia
Gr
ad
-
C
AM
,
r
ev
ea
lin
g
wh
er
e
i
n
ea
ch
im
ag
e
d
ec
is
io
n
s
o
r
ig
i
n
at
e.
Per
f
o
r
m
an
ce
s
tay
s
b
alan
ce
d
b
etwe
en
p
o
s
itiv
e
an
d
n
eg
ativ
e
ca
s
es,
ev
en
wh
en
s
am
p
le
n
u
m
b
er
s
d
if
f
er
.
Ad
j
u
s
tm
en
ts
s
u
ch
as
weig
h
ted
p
en
alties
an
d
ex
p
a
n
d
ed
d
ata
s
ets
h
elp
s
tab
ilize
o
u
tco
m
es.
Pre
cisi
o
n
h
o
ld
s
h
i
g
h
,
r
ec
all
f
o
llo
ws
clo
s
ely
,
s
h
o
win
g
co
n
s
is
ten
t
r
eliab
ilit
y
.
E
r
r
o
r
r
ate
s
ettles
at
7
.
1
7
%,
wh
ile
c
o
r
r
ec
t
class
if
icatio
n
s
r
ea
ch
9
7
.
4
2
%.
Vis
u
al
ex
p
lan
atio
n
s
g
r
o
w
clea
r
er
o
n
ce
h
ea
tm
ap
s
o
v
er
la
y
th
e
o
r
ig
in
al
s
ca
n
s
.
Un
d
er
s
tan
d
in
g
im
p
r
o
v
es
–
n
o
t
ju
s
t
f
o
r
m
ac
h
in
es,
b
u
t
f
o
r
th
o
s
e
wh
o
in
t
er
p
r
et
r
esu
lts
.
T
r
u
s
t
b
u
ild
s
s
lo
wly
wh
en
r
ea
s
o
n
in
g
b
ec
o
m
es
v
is
ib
le.
I
n
clin
ical
s
ettin
g
s
,
s
ee
in
g
wh
y
m
atter
s
as
m
u
ch
as
wh
at
is
s
ee
n
.
Fo
llo
win
g
tr
ain
in
g
,
ev
alu
atio
n
o
cc
u
r
r
ed
th
r
o
u
g
h
m
u
ltip
le
m
etr
ics.
Ov
er
all
s
u
cc
ess
em
er
g
ed
wh
en
Gr
ad
-
C
AM
p
air
ed
with
Den
s
eNe
t
-
121
,
p
o
in
tin
g
to
war
d
u
s
ef
u
ln
ess
in
d
etec
tin
g
T
B
au
to
m
atica
lly
with
in
m
ed
ical
im
ag
in
g
c
o
n
tex
ts
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
s
tate
n
o
f
u
n
d
in
g
in
v
o
lv
ed
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Sh
ash
ik
ir
an
Srin
iv
as
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Kav
ita
Av
in
ash
Patil
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Ku
s
h
alath
a
Mo
n
ap
p
a
R
am
a
✓
✓
✓
✓
✓
✓
✓
Su
d
h
a
Ven
k
atesh
lu
✓
✓
✓
✓
✓
J
ay
an
th
i M
u
th
u
s
wam
y
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Srin
iv
as B
ab
u
Nar
ay
an
ap
p
a
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
I
NF
O
RM
E
D
CO
NS
E
N
T
W
e
h
av
e
o
b
tain
ed
in
f
o
r
m
ed
c
o
n
s
en
t f
r
o
m
all
in
d
iv
id
u
als in
c
lu
d
ed
in
t
h
is
s
tu
d
y
.
E
T
H
I
CAL AP
P
RO
V
AL
T
h
e
r
esear
ch
p
er
tain
in
g
to
h
u
m
an
u
s
e
h
as
b
ee
n
au
t
h
o
r
iz
ed
b
y
t
h
e
in
s
titu
tio
n
al
r
ev
ie
w
b
o
ar
d
o
r
co
m
p
ar
ab
le
co
m
m
ittee
o
f
th
e
au
th
o
r
s
an
d
h
as
c
o
m
p
lied
w
ith
all
ap
p
lica
b
le
n
atio
n
al
r
u
les
an
d
in
s
titu
tio
n
al
p
o
licies in
ac
co
r
d
a
n
ce
with
th
e
ten
ets o
f
th
e
Helsin
k
i D
ec
lar
atio
n
.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
au
th
o
r
s
c
o
n
f
i
r
m
t
h
at
th
e
d
ata
s
u
p
p
o
r
ti
n
g
t
h
e
f
i
n
d
in
g
s
o
f
t
h
is
s
t
u
d
y
a
r
e
a
v
a
ila
b
l
e
wit
h
i
n
t
h
e
a
r
ti
cle
.
RE
F
E
R
E
NC
E
S
[
1
]
W
o
r
l
d
H
e
a
l
t
h
O
r
g
a
n
i
z
a
t
i
o
n
,
G
l
o
b
a
l
t
u
b
e
r
c
u
l
o
s
i
s
r
e
p
o
r
t
2
0
2
3
.
G
e
n
e
v
a
,
S
w
i
t
z
e
r
l
a
n
d
:
W
o
r
l
d
H
e
a
l
t
h
O
r
g
a
n
i
z
a
t
i
o
n
,
2
0
2
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
5
,
No
.
2
,
Ap
r
il 2
0
2
6
:
1
6
2
3
-
1
6
3
1
1630
[
2
]
R
.
O
.
P
a
n
i
c
k
e
r
,
K
.
S
.
K
a
l
m
a
d
y
,
J
.
R
a
j
a
n
,
a
n
d
M
.
K
.
S
a
b
u
,
“
A
u
t
o
m
a
t
i
c
d
e
t
e
c
t
i
o
n
o
f
t
u
b
e
r
c
u
l
o
s
i
s
b
a
c
i
l
l
i
f
r
o
m
m
i
c
r
o
s
c
o
p
i
c
s
p
u
t
u
m
s
m
e
a
r
i
m
a
g
e
s
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
m
e
t
h
o
d
s
,
”
B
i
o
c
y
b
e
r
n
e
t
i
c
s
a
n
d
B
i
o
m
e
d
i
c
a
l
En
g
i
n
e
e
r
i
n
g
,
v
o
l
.
3
8
,
n
o
.
3
,
p
p
.
6
9
1
–
6
9
9
,
2
0
1
8
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
b
b
e
.
2
0
1
8
.
0
5
.
0
0
7
.
[
3
]
T
.
F
.
M
.
C
a
r
v
a
l
h
o
e
t
a
l
.
,
“
A
s
y
s
t
e
m
a
t
i
c
r
e
v
i
e
w
a
n
d
r
e
p
e
a
t
a
b
i
l
i
t
y
s
t
u
d
y
o
n
t
h
e
u
s
e
o
f
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
c
l
a
s
s
i
f
y
i
n
g
a
n
d
d
e
t
e
c
t
i
n
g
t
u
b
e
r
c
u
l
o
s
i
s
b
a
c
i
l
l
i
i
n
mi
c
r
o
s
c
o
p
i
c
i
m
a
g
e
s
,
”
P
r
o
g
r
e
s
s
i
n
B
i
o
p
h
y
s
i
c
s
a
n
d
M
o
l
e
c
u
l
a
r
B
i
o
l
o
g
y
,
v
o
l
.
1
8
0
–
1
8
1
,
p
p
.
1
–
1
8
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
b
i
o
m
o
l
b
i
o
.
2
0
2
3
.
0
3
.
0
0
2
.
[
4
]
V
.
S
h
a
r
m
a
,
N
i
l
l
m
a
n
i
,
S
.
K
.
G
u
p
t
a
,
a
n
d
K
.
K
.
S
h
u
k
l
a
,
“
D
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
s
f
o
r
t
u
b
e
r
c
u
l
o
s
i
s
d
e
t
e
c
t
i
o
n
a
n
d
i
n
f
e
c
t
e
d
r
e
g
i
o
n
v
i
s
u
a
l
i
z
a
t
i
o
n
i
n
c
h
e
s
t
X
-
r
a
y
i
m
a
g
e
s
,
”
I
n
t
e
l
l
i
g
e
n
t
M
e
d
i
c
i
n
e
,
v
o
l
.
4
,
n
o
.
2
,
p
p
.
1
0
4
–
1
1
3
,
M
a
y
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
m
e
d
.
2
0
2
3
.
0
6
.
0
0
1
.
[
5
]
G
.
T
a
v
a
z
i
v
a
e
t
a
l
.
,
“
D
i
a
g
n
o
s
t
i
c
a
c
c
u
r
a
c
y
o
f
a
c
o
m
m
e
r
c
i
a
l
l
y
a
v
a
i
l
a
b
l
e
,
d
e
e
p
l
e
a
r
n
i
n
g
-
b
a
s
e
d
c
h
e
s
t
X
-
r
a
y
i
n
t
e
r
p
r
e
t
a
t
i
o
n
s
o
f
t
w
a
r
e
f
o
r
d
e
t
e
c
t
i
n
g
c
u
l
t
u
r
e
-
c
o
n
f
i
r
m
e
d
p
u
l
m
o
n
a
r
y
t
u
b
e
r
c
u
l
o
s
i
s
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
f
e
c
t
i
o
u
s
D
i
s
e
a
s
e
s
,
v
o
l
.
1
2
2
,
p
p
.
1
5
–
2
0
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
j
i
d
.
2
0
2
2
.
0
5
.
0
3
7
.
[
6
]
S
.
H
a
n
s
u
n
,
A
.
A
r
g
h
a
,
S
.
-
T
.
L
i
a
w
,
B
.
G
.
C
e
l
l
e
r
,
a
n
d
G
.
B
.
M
a
r
k
s
,
“
M
a
c
h
i
n
e
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
t
u
b
e
r
c
u
l
o
s
i
s
d
e
t
e
c
t
i
o
n
o
n
c
h
e
s
t
X
-
r
a
y
s
:
s
y
s
t
e
m
a
t
i
c
l
i
t
e
r
a
t
u
r
e
r
e
v
i
e
w
,”
J
o
u
r
n
a
l
o
f
M
e
d
i
c
a
l
I
n
t
e
r
n
e
t
R
e
s
e
a
r
c
h
,
v
o
l
.
2
5
,
Ju
l
.
2
0
2
3
,
d
o
i
:
1
0
.
2
1
9
6
/
4
3
1
5
4
.
[
7
]
M
.
M
.
K
a
b
i
r
,
M
.
F
.
M
r
i
d
h
a
,
A
.
R
a
h
m
a
n
,
M
.
A
.
H
a
m
i
d
,
a
n
d
M
.
M
.
M
o
n
o
w
a
r
,
“
D
e
t
e
c
t
i
o
n
o
f
C
O
V
I
D
-
1
9
,
p
n
e
u
mo
n
i
a
,
a
n
d
t
u
b
e
r
c
u
l
o
s
i
s
f
r
o
m
r
a
d
i
o
g
r
a
p
h
s
u
s
i
n
g
A
I
-
d
r
i
v
e
n
k
n
o
w
l
e
d
g
e
d
i
s
t
i
l
l
a
t
i
o
n
,
”
H
e
l
i
y
o
n
,
v
o
l
.
1
0
,
n
o
.
5
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
h
e
l
i
y
o
n
.
2
0
2
4
.
e
2
6
8
0
1
.
[
8
]
A
.
A
l
j
u
a
i
d
a
n
d
M
.
A
n
w
a
r
,
“
S
u
r
v
e
y
o
f
s
u
p
e
r
v
i
s
e
d
l
e
a
r
n
i
n
g
f
o
r
m
e
d
i
c
a
l
i
m
a
g
e
p
r
o
c
e
s
s
i
n
g
,
”
S
N
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
3
,
n
o
.
4
,
M
a
y
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s
4
2
9
7
9
-
0
2
2
-
0
1
1
6
6
-
1.
[
9
]
M
.
N
i
j
i
a
t
i
e
t
a
l
.
,
“
D
e
e
p
l
e
a
r
n
i
n
g
a
n
d
r
a
d
i
o
m
i
c
s
o
f
l
o
n
g
i
t
u
d
i
n
a
l
C
T
sc
a
n
s
f
o
r
e
a
r
l
y
p
r
e
d
i
c
t
i
o
n
o
f
t
u
b
e
r
c
u
l
o
s
i
s
t
r
e
a
t
m
e
n
t
o
u
t
c
o
m
e
s
,
”
E
u
r
o
p
e
a
n
J
o
u
r
n
a
l
o
f
R
a
d
i
o
l
o
g
y
,
v
o
l
.
1
6
9
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
j
r
a
d
.
2
0
2
3
.
1
1
1
1
8
0
.
[
1
0
]
S
.
V
a
t
s
e
t
a
l
.
,
“
I
n
c
r
e
m
e
n
t
a
l
l
e
a
r
n
i
n
g
-
b
a
s
e
d
c
a
s
c
a
d
e
d
m
o
d
e
l
f
o
r
d
e
t
e
c
t
i
o
n
a
n
d
l
o
c
a
l
i
z
a
t
i
o
n
o
f
t
u
b
e
r
c
u
l
o
s
i
s
f
r
o
m
c
h
e
s
t
X
-
r
a
y
i
m
a
g
e
s
,
”
E
x
p
e
r
t
S
y
s
t
e
m
s
w
i
t
h
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
3
8
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
s
w
a
.
2
0
2
3
.
1
2
2
1
2
9
.
[
1
1
]
R
.
D
u
w
a
i
r
i
a
n
d
A
.
M
e
l
h
e
m
,
“
A
d
e
e
p
l
e
a
r
n
i
n
g
-
b
a
s
e
d
f
r
a
m
e
w
o
r
k
f
o
r
a
u
t
o
m
a
t
i
c
d
e
t
e
c
t
i
o
n
o
f
d
r
u
g
r
e
s
i
s
t
a
n
c
e
i
n
t
u
b
e
r
c
u
l
o
s
i
s
p
a
t
i
e
n
t
s
,
”
E
g
y
p
t
i
a
n
I
n
f
o
r
m
a
t
i
c
s
J
o
u
rn
a
l
,
v
o
l
.
2
4
,
n
o
.
1
,
p
p
.
1
3
9
–
1
4
8
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
i
j
.
2
0
2
3
.
0
1
.
0
0
2
.
[
1
2
]
N
.
S
h
o
m
e
,
R
.
K
a
sh
y
a
p
,
a
n
d
R
.
H
.
La
s
k
a
r
,
“
D
e
t
e
c
t
i
o
n
o
f
t
u
b
e
r
c
u
l
o
s
i
s
u
s
i
n
g
c
u
s
t
o
m
i
z
e
d
M
o
b
i
l
e
N
e
t
a
n
d
t
r
a
n
s
f
e
r
l
e
a
r
n
i
n
g
f
r
o
m
c
h
e
s
t
X
-
r
a
y
i
m
a
g
e
,
”
I
m
a
g
e
a
n
d
V
i
s
i
o
n
C
o
m
p
u
t
i
n
g
,
v
o
l
.
1
4
7
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
m
a
v
i
s
.
2
0
2
4
.
1
0
5
0
6
3
.
[
1
3
]
M
.
A
l
s
h
u
t
b
i
,
Z
.
Li
,
M
.
A
l
r
i
f
a
e
y
,
M
.
A
h
m
a
d
i
p
o
u
r
,
a
n
d
M
.
M
.
O
t
h
m
a
n
,
“
A
h
y
b
r
i
d
c
l
a
s
s
i
f
i
e
r
b
a
s
e
d
o
n
s
u
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
a
n
d
J
a
y
a
a
l
g
o
r
i
t
h
m
f
o
r
b
r
e
a
s
t
c
a
n
c
e
r
c
l
a
s
s
i
f
i
c
a
t
i
o
n
,
”
N
e
u
ra
l
C
o
m
p
u
t
i
n
g
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
3
4
,
n
o
.
1
9
,
p
p
.
1
6
6
6
9
–
1
6
6
8
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s
0
0
5
2
1
-
0
2
2
-
0
7
2
9
0
-
6.
[
1
4
]
A
.
I
q
b
a
l
,
M
.
U
s
m
a
n
,
a
n
d
Z
.
A
h
m
e
d
,
“
A
n
e
f
f
i
c
i
e
n
t
d
e
e
p
l
e
a
r
n
i
n
g
-
b
a
s
e
d
f
r
a
m
e
w
o
r
k
f
o
r
t
u
b
e
r
c
u
l
o
s
i
s
d
e
t
e
c
t
i
o
n
u
s
i
n
g
c
h
e
st
X
-
r
a
y
i
m
a
g
e
s
,
”
T
u
b
e
r
c
u
l
o
s
i
s
,
v
o
l
.
1
3
6
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
t
u
b
e
.
2
0
2
2
.
1
0
2
2
3
4
.
[
1
5
]
E
.
T
.
H
a
s
t
u
t
i
,
A
.
B
u
st
a
m
a
m
,
P
.
A
n
k
i
,
R
.
A
m
a
l
i
a
,
a
n
d
A
.
S
a
l
m
a
,
“
P
e
r
f
o
r
m
a
n
c
e
o
f
t
r
u
e
t
r
a
n
s
f
e
r
l
e
a
r
n
i
n
g
u
s
i
n
g
C
N
N
D
e
n
s
e
N
e
t
-
121
f
o
r
C
O
V
I
D
-
1
9
d
e
t
e
c
t
i
o
n
f
r
o
m
c
h
e
s
t
X
-
r
a
y
i
m
a
g
e
s
,
”
i
n
I
n
H
e
N
c
e
2
0
2
1
-
2
0
2
1
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
H
e
a
l
t
h
,
I
n
s
t
r
u
m
e
n
t
a
t
i
o
n
a
n
d
M
e
a
s
u
r
e
m
e
n
t
,
a
n
d
N
a
t
u
r
a
l
S
c
i
e
n
c
e
s
,
2
0
2
1
,
p
p
.
1
–
5
.
d
o
i
:
1
0
.
1
1
0
9
/
I
n
H
e
N
c
e
5
2
8
3
3
.
2
0
2
1
.
9
5
3
7
2
6
1
.
[
1
6
]
P
.
K
.
D
a
s
,
S
.
S
r
e
e
v
a
t
s
a
v
,
a
n
d
A
.
A
b
r
a
h
a
m
,
“
A
n
e
f
f
i
c
i
e
n
t
d
e
e
p
l
e
a
r
n
i
n
g
n
e
t
w
o
r
k
w
i
t
h
o
r
t
h
o
g
o
n
a
l
s
o
f
t
m
a
x
l
a
y
e
r
f
o
r
a
u
t
o
m
a
t
i
c
d
e
t
e
c
t
i
o
n
o
f
t
u
b
e
r
c
u
l
o
s
i
s
,
”
E
n
g
i
n
e
e
r
i
n
g
A
p
p
l
i
c
a
t
i
o
n
s
o
f
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
1
3
3
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
n
g
a
p
p
a
i
.
2
0
2
4
.
1
0
8
1
1
6
.
[
1
7
]
S
.
H
.
A
b
d
u
l
l
a
h
,
W
.
M
.
S
.
A
b
e
d
i
,
a
n
d
R
.
M
.
H
a
d
i
,
“
E
n
h
a
n
c
e
d
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
a
l
g
o
r
i
t
h
m
f
o
r
p
n
e
u
mo
n
i
a
d
e
t
e
c
t
i
o
n
,
”
P
e
r
i
o
d
i
c
a
l
s
o
f
E
n
g
i
n
e
e
ri
n
g
a
n
d
N
a
t
u
r
a
l
S
c
i
e
n
c
e
s
,
v
o
l
.
1
0
,
n
o
.
6
,
p
p
.
1
6
8
–
1
8
0
,
2
0
2
2
,
d
o
i
:
1
0
.
2
1
5
3
3
/
p
e
n
.
v
1
0
i
6
.
3
3
9
7
.
[
1
8
]
S
.
C
h
a
t
t
o
r
a
j
,
B
.
R
e
d
d
y
,
M
.
T
a
d
e
p
a
l
l
i
,
a
n
d
P
.
P
u
t
h
a
,
“
C
o
m
p
a
r
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
mo
d
e
l
s
f
o
r
t
u
b
e
r
c
u
l
o
s
i
s
d
e
t
e
c
t
i
o
n
:
a
r
e
t
r
o
s
p
e
c
t
i
v
e
s
t
u
d
y
o
f
d
i
g
i
t
a
l
v
s
.
a
n
a
l
o
g
c
h
e
s
t
r
a
d
i
o
g
r
a
p
h
s
,
”
I
n
d
i
a
n
J
o
u
r
n
a
l
o
f
T
u
b
e
r
c
u
l
o
s
i
s
,
v
o
l
.
7
2
,
p
p
.
S
4
3
–
S
4
6
,
M
a
y
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
j
t
b
.
2
0
2
4
.
0
5
.
0
0
8
.
[
1
9
]
S
.
N
.
H
a
s
s
a
n
,
M
.
K
h
a
l
i
l
,
H
.
S
a
l
a
h
u
d
d
i
n
,
R
.
A
.
N
a
q
v
i
,
D
.
J
e
o
n
g
,
a
n
d
S
.
W
.
L
e
e
,
“
K
O
C
_
N
e
t
:
i
mp
a
c
t
o
f
t
h
e
sy
n
t
h
e
t
i
c
m
i
n
o
r
i
t
y
o
v
e
r
-
s
a
m
p
l
i
n
g
t
e
c
h
n
i
q
u
e
w
i
t
h
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
s
f
o
r
c
l
a
s
s
i
f
i
c
a
t
i
o
n
o
f
k
n
e
e
o
s
t
e
o
a
r
t
h
r
i
t
i
s
u
s
i
n
g
K
e
l
l
g
r
e
n
–
L
a
w
r
e
n
c
e
X
-
r
a
y
g
r
a
d
e
,”
Ma
t
h
e
m
a
t
i
c
s
,
v
o
l
.
1
2
,
n
o
.
2
2
,
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
m
a
t
h
1
2
2
2
3
5
3
4
.
[
2
0
]
B
.
N
.
S
r
i
n
i
v
a
s
,
S
.
S
h
a
s
h
i
k
i
r
a
n
,
M
.
J
a
y
a
n
t
h
i
,
N
.
R
a
j
a
n
i
,
K
.
M
.
P
a
l
a
n
i
s
w
a
m
y
,
a
n
d
M
.
R
.
K
u
s
h
a
l
a
t
h
a
,
“
T
u
b
e
r
c
u
l
o
s
i
s
c
l
a
s
s
i
f
i
c
a
t
i
o
n
u
s
i
n
g
S
V
M
a
n
d
m
o
d
i
f
i
e
d
C
N
N
,
”
I
r
a
n
i
a
n
J
o
u
r
n
a
l
o
f
E
l
e
c
t
r
i
c
a
l
a
n
d
E
l
e
c
t
r
o
n
i
c
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
2
0
,
n
o
.
4
,
p
p
.
1
2
6
–
1
3
3
,
2
0
2
4
,
d
o
i
:
1
0
.
2
2
0
6
8
/
I
J
E
E
E
.
2
0
.
4
.
3
4
6
3
.
[
2
1
]
M
.
N
i
j
i
a
t
i
e
t
a
l
.
,
“
D
e
e
p
l
e
a
r
n
i
n
g
o
n
l
o
n
g
i
t
u
d
i
n
a
l
C
T
s
c
a
n
s
:
a
u
t
o
m
a
t
e
d
p
r
e
d
i
c
t
i
o
n
o
f
t
r
e
a
t
m
e
n
t
o
u
t
c
o
m
e
s
i
n
h
o
s
p
i
t
a
l
i
z
e
d
t
u
b
e
r
c
u
l
o
s
i
s
p
a
t
i
e
n
t
s
,
”
i
S
c
i
e
n
c
e
,
v
o
l
.
2
6
,
n
o
.
1
1
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
s
c
i
.
2
0
2
3
.
1
0
8
3
2
6
.
[
2
2
]
T
.
R
a
h
m
a
n
e
t
a
l
.
,
“
R
e
l
i
a
b
l
e
t
u
b
e
r
c
u
l
o
s
i
s
d
e
t
e
c
t
i
o
n
u
s
i
n
g
c
h
e
s
t
X
-
r
a
y
w
i
t
h
d
e
e
p
l
e
a
r
n
i
n
g
,
s
e
g
m
e
n
t
a
t
i
o
n
a
n
d
v
i
s
u
a
l
i
z
a
t
i
o
n
,
”
I
E
E
E
A
c
c
e
s
s
,
v
o
l
.
8
,
p
p
.
1
9
1
5
8
6
–
1
9
1
6
0
1
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
E
S
S
.
2
0
2
0
.
3
0
3
1
3
8
4
.
[
2
3
]
R
.
L
o
k
a
r
e
,
J
.
W
a
d
m
a
r
e
,
S
.
R
.
P
a
t
i
l
,
G
.
W
a
d
m
a
r
e
,
a
n
d
D
.
P
a
t
i
l
,
“
T
r
a
n
s
p
a
r
e
n
t
p
r
e
c
i
s
i
o
n
:
e
x
p
l
a
i
n
a
b
l
e
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
e
m
p
o
w
e
r
e
d
b
r
e
a
s
t
c
a
n
c
e
r
r
e
c
o
mm
e
n
d
a
t
i
o
n
s
f
o
r
p
e
r
s
o
n
a
l
i
z
e
d
t
r
e
a
t
m
e
n
t
,
”
I
A
E
S
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
1
3
,
n
o
.
3
,
p
p
.
2
6
9
4
–
2
7
0
2
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
a
i
.
v
1
3
.
i
3
.
p
p
2
6
9
4
-
2
7
0
2
.
[
2
4
]
R
.
G
e
e
t
h
a
m
a
n
i
a
n
d
A
.
R
a
n
i
c
h
i
t
r
a
,
“
E
n
h
a
n
c
i
n
g
t
u
b
e
r
c
u
l
o
s
i
s
d
e
t
e
c
t
i
o
n
:
l
e
v
e
r
a
g
i
n
g
R
F
-
H
O
G
m
o
d
e
l
f
o
r
a
u
t
o
m
a
t
e
d
d
i
a
g
n
o
s
i
s
f
r
o
m
c
h
e
s
t
X
-
r
a
y
i
m
a
g
e
s
,”
P
r
o
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
3
0
,
p
p
.
2
1
–
3
2
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s
.
2
0
2
3
.
1
2
.
0
5
7
.
[
2
5
]
G
.
T
a
m
u
r
a
,
G
.
L
l
a
n
o
,
A
.
A
r
i
s
t
i
z
á
b
a
l
,
J
.
V
a
l
e
n
c
i
a
,
L.
S
u
a
,
a
n
d
L
.
F
e
r
n
a
n
d
e
z
,
“
M
a
c
h
i
n
e
-
l
e
a
r
n
i
n
g
m
e
t
h
o
d
s
f
o
r
d
e
t
e
c
t
i
n
g
t
u
b
e
r
c
u
l
o
s
i
s
i
n
Zi
e
h
l
-
N
e
e
l
s
e
n
s
t
a
i
n
e
d
s
l
i
d
e
s
:
a
s
y
s
t
e
m
a
t
i
c
l
i
t
e
r
a
t
u
r
e
r
e
v
i
e
w
,
”
I
n
t
e
l
l
i
g
e
n
t
S
y
s
t
e
m
s
w
i
t
h
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
2
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
s
w
a
.
2
0
2
4
.
2
0
0
3
6
5
.
[
2
6
]
C
.
Y
i
n
g
e
t
a
l
.
,
“
T
-
S
P
O
T
w
i
t
h
C
T
i
m
a
g
e
a
n
a
l
y
s
i
s
b
a
s
e
d
o
n
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
e
a
r
l
y
d
i
f
f
e
r
e
n
t
i
a
l
d
i
a
g
n
o
s
i
s
o
f
n
o
n
t
u
b
e
r
c
u
l
o
u
s
m
y
c
o
b
a
c
t
e
r
i
a
p
u
l
mo
n
a
r
y
d
i
s
e
a
s
e
a
n
d
p
u
l
mo
n
a
r
y
t
u
b
e
r
c
u
l
o
s
i
s
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
f
e
c
t
i
o
u
s
D
i
se
a
s
e
s
,
v
o
l
.
1
2
5
,
p
p
.
4
2
–
5
0
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
j
i
d
.
2
0
2
2
.
0
9
.
0
3
1
.
[
2
7
]
M
.
A
.
T
.
R
o
n
y
e
t
a
l
.
,
“
I
n
n
o
v
a
t
i
v
e
a
p
p
r
o
a
c
h
t
o
d
e
t
e
c
t
i
n
g
a
u
t
i
s
m
s
p
e
c
t
r
u
m
d
i
s
o
r
d
e
r
u
s
i
n
g
e
x
p
l
a
i
n
a
b
l
e
f
e
a
t
u
r
e
s
a
n
d
s
m
a
r
t
w
e
b
a
p
p
l
i
c
a
t
i
o
n
,
”
M
a
t
h
e
m
a
t
i
c
s
,
v
o
l
.
1
2
,
n
o
.
2
2
,
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
m
a
t
h
1
2
2
2
3
5
1
5
.
[
2
8
]
S
.
S
h
a
h
s
h
i
k
i
r
a
n
,
N
.
S
r
i
n
i
v
a
s
B
a
b
u
,
S
.
P
r
a
m
a
n
i
k
,
a
n
d
S
a
t
h
w
i
k
,
“
A
m
o
d
i
f
i
e
d
d
e
e
p
l
e
a
r
n
i
n
g
a
r
c
h
i
t
e
c
t
u
r
e
f
o
r
d
e
t
e
c
t
i
n
g
M
a
l
a
r
i
a
I
n
f
e
c
t
e
d
c
e
l
l
s
,
”
i
n
2
0
2
4
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
K
n
o
w
l
e
d
g
e
E
n
g
i
n
e
e
r
i
n
g
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
S
y
s
t
e
m
s
,
I
C
K
E
C
S
2
0
2
4
,
2
0
2
4
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
K
E
C
S
6
1
4
9
2
.
2
0
2
4
.
1
0
6
1
7
0
0
5
.
[
2
9
]
R
.
A
.
N
a
q
v
i
,
A
.
H
a
i
d
e
r
,
H
.
S
.
K
i
m
,
D
.
J
e
o
n
g
,
a
n
d
S
.
W
.
L
e
e
,
“
T
r
a
n
s
f
o
r
m
a
t
i
v
e
n
o
i
s
e
r
e
d
u
c
t
i
o
n
:
l
e
v
e
r
a
g
i
n
g
a
t
r
a
n
s
f
o
r
m
e
r
-
b
a
s
e
d
d
e
e
p
n
e
t
w
o
r
k
f
o
r
m
e
d
i
c
a
l
i
m
a
g
e
d
e
n
o
i
s
i
n
g
,
”
M
a
t
h
e
m
a
t
i
c
s
,
v
o
l
.
1
2
,
n
o
.
1
5
,
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
m
a
t
h
1
2
1
5
2
3
1
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
A
n
a
lysi
s
o
f tu
b
ercu
lo
s
is
d
etec
tio
n
u
s
in
g
d
ee
p
lea
r
n
in
g
tech
n
i
q
u
e
a
n
d
ex
p
l
a
in
a
b
le
…
(
S
h
a
s
h
ikir
a
n
S
r
in
iva
s
)
1631
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
S
h
a
shi
k
ira
n
S
r
in
i
v
a
s
in
a
ff
il
iatio
n
with
Vis
v
e
sv
e
ra
y
a
Tec
h
n
o
l
o
g
ica
l
Un
iv
e
rsity
,
Be
lag
a
v
i
-
5
9
0
0
1
8
.
He
is
a
re
se
a
rc
h
sc
h
o
lar
a
t
t
h
e
De
p
a
rtme
n
t
o
f
E
lec
tro
n
ics
a
n
d
Co
m
m
u
n
ica
ti
o
n
En
g
i
n
e
e
rin
g
,
R.
R.
In
st
it
u
te
o
f
Te
c
h
n
o
l
o
g
y
,
Ba
n
g
a
lo
re
,
In
d
ia,
a
n
d
a
fa
c
u
lt
y
a
t
th
e
De
p
a
rtme
n
t
o
f
El
e
c
tro
n
ics
a
n
d
Co
m
m
u
n
ica
ti
o
n
En
g
i
n
e
e
rin
g
,
Ne
w
H
o
rizo
n
C
o
ll
e
g
e
o
f
E
n
g
i
n
e
e
rin
g
,
Ba
n
g
a
lo
re
,
In
d
ia.
He
wa
s
a
wa
rd
e
d
t
h
e
B
.
E
.
in
ECE
.
Visv
e
sv
e
ra
y
a
Tec
h
n
o
lo
g
ica
l
U
n
iv
e
rsit
y
-
P
G
Ce
n
ter
Ba
n
g
a
lo
re
a
wa
rd
e
d
h
im
M
.
Tec
h
.
in
DECS
.
At
Visv
e
sv
e
ra
y
a
Tec
h
n
o
l
o
g
ica
l
Un
i
v
e
rsity
in
Be
lag
a
v
i,
h
e
is
wo
rk
i
n
g
t
o
wa
rd
a
P
h
.
D.
His
a
re
a
s
o
f
in
tere
sts
a
re
m
e
d
ica
l
ima
g
e
a
n
a
ly
sis,
a
rti
ficia
l
in
telli
g
e
n
c
e
,
a
n
d
ima
g
e
/sig
n
a
l
p
r
o
c
e
ss
in
g
.
Ba
se
d
o
n
h
is
c
re
a
ti
v
e
id
e
a
s,
h
e
h
a
s
p
u
b
l
ish
e
d
6
p
a
ten
ts
a
s
first
in
v
e
n
to
r
,
2
UG
C
c
a
re
jo
u
rn
a
ls,
6
IEE
E
S
c
o
p
u
s
i
n
d
e
x
e
d
c
o
n
fe
re
n
c
e
p
a
p
e
rs,
2
Q
3
jo
u
r
n
a
l,
1
S
CI
p
a
p
e
r
a
n
d
re
c
e
iv
e
d
th
e
NMIT
Co
n
fe
re
n
c
e
2
0
2
4
'
s
"
Be
st
p
a
p
e
r
a
wa
rd
"
fo
r
o
n
e
o
f
h
i
s
a
rti
c
les
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
sh
a
sh
ik
iran
0
6
7
@g
m
a
il
.
c
o
m
.
K
a
v
ita
Av
i
n
a
sh
P
a
til
re
c
e
iv
e
d
h
e
r
B
.
E
.
in
Tele
c
o
m
m
u
n
ica
ti
o
n
En
g
i
n
e
e
rin
g
fr
o
m
t
h
e
K
.
L
.
E
.
Co
ll
e
g
e
o
f
E
n
g
i
n
e
e
rin
g
Be
lg
a
u
m
in
2
0
1
0
a
n
d
th
e
M
.
Tec
h
.
in
VLS
I
De
sig
n
a
n
d
Emb
e
d
d
e
d
s
y
ste
m
fro
m
K
.
L
.
E
.
Co
ll
e
g
e
o
f
En
g
in
e
e
ri
n
g
Be
lg
a
u
m
in
2
0
1
2
.
S
h
e
re
c
e
iv
e
d
h
e
r
P
h
.
D.
d
e
g
re
e
in
El
e
c
tri
c
a
l
a
n
d
El
e
c
tro
n
ics
E
n
g
i
n
e
e
rin
g
fro
m
Visv
e
sv
a
ra
y
a
Tec
h
n
o
lo
g
ica
l
Un
iv
e
rsity
in
Ju
n
e
2
0
2
3
.
P
re
se
n
tl
y
sh
e
is
wo
r
k
in
g
a
s
a
n
a
ss
istan
t
p
ro
fe
ss
o
r
i
n
th
e
De
p
a
rtme
n
t
o
f
El
e
c
tro
n
ics
a
n
d
Tele
c
o
m
m
u
n
ica
ti
o
n
En
g
in
e
e
ri
n
g
,
a
t
Ra
m
a
iah
In
stit
u
te
o
f
Tec
h
n
o
l
o
g
y
,
Ba
n
g
a
lo
re
.
He
r
re
se
a
rc
h
in
tere
st
in
c
lu
d
e
s
b
io
m
e
d
ica
l
e
n
g
in
e
e
ri
n
g
,
m
a
c
h
in
e
lea
rn
in
g
,
d
a
ta
sc
ien
c
e
,
a
n
d
a
rti
ficia
l
in
telli
g
e
n
c
e
.
To
tal
1
3
y
e
a
rs
o
f
tea
c
h
in
g
e
x
p
e
rien
c
e
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
k
a
v
it
a
m
a
lag
a
tt
i@g
m
a
il
.
c
o
m
.
K
u
sha
la
t
h
a
Mo
n
a
p
p
a
Ra
m
a
re
c
e
iv
e
d
h
e
r
B.
E
.
i
n
El
e
c
tro
n
ics
a
n
d
C
o
m
m
u
n
ica
ti
o
n
En
g
i
n
e
e
rin
g
fro
m
P
A
Co
ll
e
g
e
o
f
En
g
in
e
e
rin
g
,
M
a
n
g
a
l
o
re
in
2
0
0
5
,
a
ffil
iate
d
to
VTU,
Be
lg
a
u
m
a
n
d
h
e
r
M
.
Tec
h
.
in
Dig
it
a
l
C
o
m
m
u
n
ica
ti
o
n
fro
m
BM
S
Co
l
leg
e
o
f
En
g
in
e
e
rin
g
,
Be
n
g
a
l
u
ru
,
a
ffil
iate
d
to
VTU,
Be
l
g
a
u
m
in
2
0
1
4
.
S
h
e
c
o
m
p
lete
d
h
e
r
P
h
.
D
.
fro
m
VTU.
P
re
se
n
tl
y
,
sh
e
is
wo
rk
i
n
g
a
s
a
ss
istan
t
p
ro
fe
ss
o
r
i
n
D
e
p
a
rtme
n
t
o
f
E
lec
tro
n
ics
a
n
d
Co
m
m
u
n
ica
ti
o
n
E
n
g
i
n
e
e
rin
g
in
Nitt
e
M
e
e
n
a
k
sh
i
In
stit
u
te
o
f
Te
c
h
n
o
l
o
g
y
(De
e
m
e
d
to
b
e
Un
iv
e
r
sity
),
Be
n
g
a
l
u
ru
.
He
r
re
se
a
rc
h
a
re
a
s
in
c
lu
d
e
sig
n
a
l
p
ro
c
e
ss
in
g
,
r
e
m
o
te
se
n
sin
g
,
m
a
c
h
i
n
e
lea
rn
i
n
g
,
a
n
d
d
a
ta
sc
ien
c
e
.
S
h
e
h
a
s
to
tal
o
f
1
6
y
e
a
rs o
f
tea
c
h
in
g
e
x
p
e
rien
c
e
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
k
u
sh
a
lat
h
a
.
m
r@n
m
it
.
a
c
.
in
.
S
u
d
h
a
Ve
n
k
a
te
shl
u
is
a
re
se
a
rc
h
sc
h
o
lar
a
t
th
e
De
p
a
rtm
e
n
t
o
f
El
e
c
tro
n
ics
a
n
d
Co
m
m
u
n
ica
ti
o
n
E
n
g
in
e
e
rin
g
,
S
JB
In
stit
u
te
o
f
Tec
h
n
o
l
o
g
y
,
Be
n
g
a
lu
ru
,
In
d
ia
a
n
d
fa
c
u
l
ty
a
t
t
h
e
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
ri
n
g
,
Am
it
y
Un
i
v
e
rsity
,
Be
n
g
a
lu
r
u
.
S
h
e
re
c
e
iv
e
d
th
e
B
.
E
.
d
e
g
re
e
RV
C
o
ll
e
g
e
o
f
E
n
g
i
n
e
e
rin
g
.
S
h
e
re
c
e
iv
e
d
th
e
M
.
Tec
h
.
d
e
g
re
e
in
RV
C
o
ll
e
g
e
o
f
En
g
i
n
e
e
rin
g
fr
o
m
Visv
e
sv
e
ra
y
a
Tec
h
n
o
l
o
g
ica
l
Un
i
v
e
rsity
–
P
G
Ce
n
ter
Ba
n
g
a
lo
re
.
S
h
e
is
p
u
rs
u
in
g
a
P
h
.
D.
a
t
Visv
e
sv
e
ra
y
a
Tec
h
n
o
lo
g
ica
l
Un
i
v
e
rsity
,
Be
lag
a
v
i.
He
r
re
se
a
rc
h
a
re
a
s
a
re
a
rti
ficia
l
in
telli
g
e
n
c
e
,
m
a
c
h
in
e
lea
rn
in
g
,
a
n
d
m
e
d
ica
l
ima
g
e
a
n
a
ly
sis
.
S
h
e
h
a
s
fil
e
d
3
p
a
ten
ts
o
n
h
e
r
in
n
o
v
a
ti
v
e
i
d
e
a
s.
S
h
e
h
a
s
p
u
b
li
s
h
e
d
2
IE
EE
S
c
o
p
u
s
i
n
d
e
x
e
d
c
o
n
fe
re
n
c
e
p
a
p
e
rs
,
2
UG
C
c
a
re
jo
u
r
n
a
ls
,
a
n
d
1
sc
o
p
u
s in
d
e
x
e
d
jo
u
rn
a
l.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
su
d
h
a
v
ij
u
1
@
g
m
a
il
.
c
o
m
.
J
a
y
a
n
t
h
i
Mu
th
u
sw
a
m
y
re
c
e
iv
e
d
h
e
r
P
h
.
D.
d
e
g
re
e
i
n
El
e
c
t
rica
l
a
n
d
El
e
c
tro
n
ics
En
g
i
n
e
e
rin
g
,
Visv
e
sv
a
ra
y
a
Tec
h
n
o
l
o
g
ica
l
Un
i
v
e
rsity
,
Be
lag
a
v
i,
In
d
ia,
2
0
1
9
.
S
h
e
is
c
u
rre
n
t
ly
wo
rk
i
n
g
a
s
a
ss
o
c
iate
p
ro
fe
ss
o
r
i
n
El
e
c
tro
n
ics
a
n
d
C
o
m
m
u
n
ica
ti
o
n
En
g
in
e
e
rin
g
,
Ne
w
Ho
rizo
n
Co
ll
e
g
e
o
f
En
g
in
e
e
rin
g
,
Ba
n
g
a
l
o
r
e
.
S
h
e
h
a
s
a
u
th
o
re
d
o
r
c
o
a
u
t
h
o
re
d
m
o
re
t
h
a
n
3
2
re
fe
re
e
d
j
o
u
r
n
a
l
a
n
d
c
o
n
fe
re
n
c
e
p
a
p
e
rs.
He
r
re
se
a
rc
h
in
tere
sts
in
c
l
u
d
e
d
ig
it
a
l
sig
n
a
l
p
ro
c
e
ss
in
g
,
b
io
m
e
d
ica
l
si
g
n
a
l
a
n
d
ima
g
e
p
ro
c
e
ss
in
g
,
Io
T,
a
n
d
m
a
c
h
in
e
lea
rn
in
g
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
jay
a
n
th
isa
t
h
ish
1
0
1
2
@g
m
a
il
.
c
o
m
.
S
r
in
iv
a
s
Ba
b
u
Na
r
a
y
a
n
a
p
p
a
is
a
n
a
ss
istan
t
p
r
o
fe
ss
o
r
a
t
th
e
De
p
a
rtme
n
t
o
f
El
e
c
tro
n
ics
a
n
d
Co
m
m
u
n
ica
ti
o
n
En
g
in
e
e
rin
g
,
Ne
w
Ho
r
izo
n
Co
l
leg
e
o
f
E
n
g
in
e
e
rin
g
a
ffil
iate
d
with
Visv
e
sw
a
ra
y
a
Tec
h
n
o
l
o
g
ic
a
l
Un
iv
e
rsity
,
Be
lag
a
v
i
5
9
0
0
1
8
In
d
ia.
He
h
o
ld
s
a
n
M
.
Tec
h
.
,
d
e
g
re
e
in
Dig
it
a
l
Co
m
m
u
n
ica
ti
o
n
a
n
d
Ne
two
r
k
i
n
g
wit
h
a
re
se
a
rc
h
sp
e
c
ializa
ti
o
n
i
n
m
e
d
ica
l
ima
g
e
a
n
a
ly
sis.
His
re
se
a
rc
h
a
re
a
s
a
re
ima
g
e
/sig
n
a
l
p
ro
c
e
ss
in
g
,
b
io
m
e
tri
c
s,
m
e
d
ica
l
ima
g
e
a
n
a
ly
sis,
a
n
d
p
a
tt
e
rn
re
c
o
g
n
it
io
n
.
He
h
a
s
fil
e
d
s
o
m
e
d
e
sig
n
p
a
ten
ts
o
n
h
is
i
n
n
o
v
a
ti
v
e
id
e
a
s.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
sri
b
a
b
u
6
8
8
7
@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.