I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
pu
t
er
E
ng
ineering
(
I
J
E
CE
)
Vo
l.
15
,
No
.
1
,
Feb
r
u
ar
y
20
25
,
p
p
.
958
~
969
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
15
i
1
.
pp
9
5
8
-
9
6
9
958
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Berkeley
wav
elet
trans
form
a
nd
im
pro
v
ed
YO
L
O
v
7
-
ba
se
d
cla
ss
ificatio
n t
ec
h
nique f
o
r
brain
tu
mo
r sev
erity pre
d
iction
Nilesh
B
ha
s
k
a
rr
a
o
B
a
ha
du
re
1
,
Sid
hes
wa
r
Ro
utr
a
y
2
,
J
a
g
dis
h C
ha
nd
ra
P
a
t
ni
3
,
Na
g
ra
j
a
n Ra
j
u
4
,
P
ra
s
enj
ee
t
Da
m
o
da
r
P
a
t
il
5
1
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
,
S
c
h
o
o
l
o
f
Te
c
h
n
o
l
o
g
y
,
G
S
F
C
U
n
i
v
e
r
si
t
y
,
V
a
d
o
d
a
r
a
,
I
n
d
i
a
2
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
,
S
c
h
o
o
l
o
f
Te
c
h
n
o
l
o
g
y
,
P
a
n
d
i
t
D
e
e
n
d
a
y
a
l
En
e
r
g
y
U
n
i
v
e
r
s
i
t
y
,
G
a
n
d
h
i
n
a
g
a
r
,
I
n
d
i
a
3
S
c
h
o
o
l
o
f
A
d
v
a
n
c
e
C
o
mp
u
t
i
n
g
,
A
l
l
i
a
n
c
e
U
n
i
v
e
r
si
t
y
,
B
e
n
g
a
l
u
r
u
,
I
n
d
i
a
4
S
c
h
o
o
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
s
En
g
i
n
e
e
r
i
n
g
,
S
A
S
TR
A
D
e
e
m
e
d
U
n
i
v
e
r
si
t
y
,
Th
a
n
j
a
v
u
r
,
I
n
d
i
a
5
M
I
T
S
c
h
o
o
l
o
f
E
n
g
i
n
e
e
r
i
n
g
,
M
I
T
A
r
t
D
e
si
g
n
a
n
d
T
e
c
h
n
o
l
o
g
y
U
n
i
v
e
r
si
t
y
,
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
Ap
r
9
,
2
0
2
4
R
ev
is
ed
Au
g
2
2
,
2
0
2
4
Acc
ep
ted
Sep
3
,
2
0
2
4
Ab
n
o
rm
a
li
t
y
i
n
b
ra
in
ti
ss
u
e
s
is
a
li
fe
-
th
re
a
ten
i
n
g
il
l
n
e
ss
i
n
h
u
m
a
n
s
Un
-
b
ias
to
g
e
n
d
e
r
a
n
d
a
g
e
if
it
is u
n
re
c
o
g
n
ize
d
a
n
d
u
n
trea
ted
with
i
n
ti
m
e
,
will
lea
d
to
se
v
e
re
c
o
m
p
li
c
a
ti
o
n
s
a
n
d
e
x
tre
m
e
c
o
n
d
it
i
o
n
s.
T
h
e
b
ra
i
n
tu
m
o
r
is
m
a
in
ly
in
flu
e
n
c
e
d
b
y
a
v
a
riety
o
f
u
n
p
re
d
icte
d
a
n
d
u
n
a
v
o
id
a
b
le
re
a
so
n
s.
Its
e
v
a
lu
a
ti
o
n
,
s
p
re
a
d
p
a
tt
e
rn
,
a
n
d
i
d
e
n
ti
fica
ti
o
n
in
v
o
l
v
e
s
c
o
m
p
le
x
a
ss
ig
n
m
e
n
t.
Its
e
a
rly
g
ra
d
i
n
g
a
n
d
t
h
e
p
ro
p
e
r
c
l
a
ss
ifi
c
a
ti
o
n
e
n
s
u
re
e
ffe
c
ti
v
e
trea
t
m
e
n
t.
T
h
e
p
ro
p
o
se
d
w
o
rk
a
tt
e
m
p
ts
to
e
x
tra
c
t
a
n
d
c
las
sify
th
e
t
u
m
o
r
re
g
io
n
u
sin
g
a
n
a
u
to
m
a
ti
c
c
las
sifica
ti
o
n
sy
ste
m
f
o
r
m
a
g
n
e
ti
c
re
so
n
a
n
c
e
ima
g
in
g
(
M
RI)
b
ra
i
n
tu
m
o
rs.
A
d
e
e
p
lea
rn
i
n
g
c
o
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
two
rk
-
b
a
se
d
a
rc
h
it
e
c
tu
re
YO
LO
is
e
m
p
lo
y
e
d
to
c
las
sify
a
n
d
d
e
tec
t
t
h
e
t
u
m
o
r
fro
m
b
ra
in
M
R
ima
g
e
s.
Th
e
p
ro
p
o
se
d
m
e
th
o
d
re
su
l
ted
i
n
su
p
e
rio
r
se
g
m
e
n
tatio
n
,
a
n
d
c
la
ss
ifi
c
a
ti
o
n
p
e
rfo
rm
a
n
c
e
in
term
s
o
f
su
b
jec
t
iv
e
v
isu
a
li
z
a
ti
o
n
a
n
d
o
b
jec
ti
v
e
m
e
tri
c
s
a
s
c
o
m
p
a
re
d
to
sta
te
o
f
a
rt
a
p
p
ro
a
c
h
e
s.
Th
e
p
ro
p
o
se
d
YO
LO
-
b
a
se
d
m
e
th
o
d
c
o
ll
e
c
ti
v
e
ly
a
c
h
iev
e
d
9
8
.
8
9
%
c
la
ss
ifi
c
a
ti
o
n
a
c
c
u
ra
c
y
o
n
t
h
e
BRA
INIX
a
n
d
Ka
g
g
le d
a
tas
e
ts.
K
ey
w
o
r
d
s
:
B
er
k
eley
wav
elet
tr
an
s
f
o
r
m
C
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
Ma
g
n
etic
r
eso
n
an
ce
im
a
g
in
g
W
av
elet
tr
an
s
f
o
r
m
s
YOL
O
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
:
Nilesh
B
h
ask
ar
r
ao
B
ah
ad
u
r
e
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
in
ee
r
in
g
,
Sch
o
o
l o
f
T
ec
h
n
o
lo
g
y
,
GSFC
Un
iv
er
s
i
ty
Vad
o
d
ar
a,
G
u
jar
at,
I
n
d
ia
E
m
ail: n
b
ah
ad
u
r
e@
g
m
ail.
co
m
,
n
iles
h
.
b
ah
ad
u
r
e@
g
s
f
cu
n
iv
e
r
s
ity
.
ac
.
in
1.
I
NT
RO
D
UCT
I
O
N
L
o
ts
o
f
r
esear
ch
h
as
b
ee
n
in
v
esti
g
ated
in
th
e
last
d
ec
ad
e
o
n
tu
m
o
r
d
etec
tio
n
.
On
e
o
f
th
e
s
ig
n
if
ican
t
d
is
ea
s
es wh
er
e
tu
m
o
r
d
etec
tio
n
is
ad
d
r
ess
ed
is
th
e
b
r
ain
tu
m
o
r
.
E
v
en
th
o
u
g
h
th
e
tu
m
o
r
is
n
o
t lis
ted
in
th
e
f
ir
s
t
p
o
s
itio
n
r
eg
a
r
d
in
g
d
ea
th
s
,
r
o
a
d
ac
cid
en
ts
tak
e
m
o
r
e
life
th
an
tu
m
o
r
s
,
b
u
t
t
h
e
s
itu
atio
n
i
n
b
o
th
s
ce
n
ar
io
s
is
d
if
f
er
en
t.
R
o
ad
ac
cid
en
ts
ar
e
p
r
ev
en
tiv
e,
a
n
d
m
o
s
t
o
f
th
e
tim
e,
th
ey
h
ap
p
e
n
d
u
e
to
h
u
m
a
n
n
eg
lig
en
ce
.
So
m
e
u
n
av
o
i
d
ab
le
s
itu
atio
n
s
a
r
e
als
o
witn
ess
ed
f
o
r
r
o
ad
ca
lam
ities
,
s
u
ch
as
s
u
d
d
en
o
b
s
tacle
s
,
f
ailu
r
e
in
m
ec
h
a
n
is
m
,
an
d
m
a
n
y
o
th
er
s
[
1
]
.
I
n
th
is
c
o
n
tex
t,
b
r
ain
t
u
m
o
r
s
ca
n
h
a
p
p
en
d
u
e
to
v
ar
io
u
s
r
ea
s
o
n
s
s
u
ch
as
p
o
llu
tio
n
,
s
tr
ess
,
f
am
ily
b
ac
k
g
r
o
u
n
d
,
an
d
ea
tin
g
h
ab
its
.
T
h
is
s
itu
atio
n
is
u
n
av
o
id
ab
le
ev
en
t
h
o
u
g
h
m
a
n
y
p
r
ec
au
tio
n
s
ar
e
tak
en
.
T
h
is
ty
p
e
o
f
d
i
s
ea
s
e
ca
n
ca
u
s
e
m
u
ch
p
ain
to
th
e
s
u
f
f
e
r
er
.
Ho
wev
er
,
r
ec
e
n
t
tech
n
o
lo
g
ical
ad
v
a
n
ce
m
en
t
m
ak
es
it
p
o
s
s
ib
le
to
cu
r
e
it,
if
d
etec
ted
at
a
n
ea
r
ly
s
tag
e.
I
t
is
im
p
o
r
tan
t
to
n
o
te
th
at
it
s
h
o
u
ld
b
e
d
etec
ted
in
its
ea
r
ly
s
tag
e
to
o
f
f
er
th
e
p
r
o
p
e
r
tr
ea
tm
en
t,
wh
ich
is
o
n
e
o
f
th
e
m
ain
r
ea
s
o
n
s
wh
y
lo
ts
o
f
r
esear
ch
is
in
v
esti
g
ated
in
th
i
s
f
ield
.
I
n
th
e
last
d
ec
ad
e,
a
3
0
0
%
in
cr
ea
s
e
in
b
r
ain
tu
m
o
r
ca
s
es
h
as
b
ee
n
n
o
ticed
,
an
d
ac
c
o
r
d
in
g
to
th
e
d
at
a
o
f
GL
OB
OC
ON
2023
[
2
]
–
[
5
]
,
m
o
r
e
th
an
1
7
.
7
m
illi
o
n
d
ea
th
s
h
av
e
b
ee
n
r
eg
is
ter
ed
d
u
e
to
tu
m
o
r
in
f
ec
tio
n
s
.
Fig
u
r
e
1
s
h
o
ws
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
B
erkeley
w
a
ve
let
tr
a
n
s
fo
r
m
a
n
d
I
mp
r
o
ve
d
YOLOv7
-
b
a
s
ed
…
(
N
iles
h
B
h
a
s
ka
r
r
a
o
B
a
h
a
d
u
r
e
)
959
ca
s
es
an
d
d
ea
t
h
s
f
r
o
m
2
0
1
0
t
o
2
0
2
2
an
d
th
e
esti
m
ated
v
o
l
u
m
e
f
r
o
m
2
0
2
3
to
2
0
4
0
.
T
h
e
d
ea
th
r
ate
will
b
e
co
n
tr
o
lled
an
d
r
ed
u
ce
d
to
s
o
m
e
ex
ten
t,
an
d
th
e
s
u
r
v
iv
al
r
ate
ca
n
in
cr
ea
s
e.
I
f
t
h
e
in
f
e
ctio
n
in
tis
s
u
es
ca
n
b
e
d
etec
ted
in
its
ea
r
ly
s
tag
e,
it
is
p
o
s
s
ib
le
to
co
n
tr
o
l
a
n
d
r
ed
u
c
e
th
e
d
ea
th
r
ate
d
u
e
t
o
b
r
ai
n
t
u
m
o
r
s
[
6
]
,
[
7
]
.
New
tech
n
o
lo
g
ical
e
v
o
lu
tio
n
s
o
p
e
n
n
ew
f
in
d
in
g
s
an
d
h
elp
d
o
cto
r
s
o
f
f
er
p
atien
ts
m
o
r
e
c
o
n
v
en
ien
t
a
n
d
s
o
u
n
d
h
ea
lth
ca
r
e.
Fig
u
r
e
1
.
E
s
tim
ated
n
u
m
b
e
r
o
f
ca
s
es a
n
d
d
ea
th
s
[
4
]
,
[
5
]
T
h
e
t
u
m
o
r
i
s
a
n
a
b
n
o
r
m
a
l
i
t
y
g
r
o
w
i
n
g
t
i
s
s
u
es
,
a
n
d
a
n
a
b
n
o
r
m
a
l
e
n
l
a
r
g
e
m
e
n
t
o
f
t
is
s
u
e
s
i
n
b
r
a
i
n
c
a
u
s
es
t
u
m
o
r
i
n
b
r
a
i
n
.
T
h
e
s
e
t
i
s
s
u
es
h
a
v
e
n
o
p
h
y
s
i
o
l
o
g
i
c
a
l
f
u
n
c
t
i
o
n
a
n
d
p
e
r
f
o
r
m
n
o
a
c
t
i
v
i
t
y
i
n
t
h
e
b
r
a
i
n
.
H
o
w
e
v
e
r
,
a
s
t
h
e
y
i
n
c
r
e
a
s
e
t
h
e
v
o
l
u
m
e
i
n
t
h
e
b
r
a
i
n
,
i
t
c
r
e
a
t
e
s
u
n
d
u
e
p
r
e
s
s
u
r
e
o
n
t
h
e
b
r
a
i
n
,
a
n
d
s
o
i
t
a
f
f
e
c
t
s
t
h
e
o
v
e
r
a
ll
f
u
n
c
t
i
o
n
i
n
g
o
f
t
h
e
b
r
a
i
n
a
n
d
c
a
u
s
es
i
r
r
e
g
u
l
a
r
n
e
r
v
e
s
y
m
p
to
m
s
[
8
]
.
O
n
c
e
t
h
e
t
u
m
o
r
s
t
a
r
t
s
d
e
v
e
l
o
p
i
n
g
,
t
h
e
y
o
v
e
r
g
r
o
w
.
I
f
t
h
e
p
r
o
p
e
r
t
r
e
a
t
m
e
n
t
i
s
n
o
t
o
f
f
e
r
e
d
i
n
t
i
m
e
,
it
b
e
c
o
m
e
s
d
a
n
g
e
r
o
u
s
a
n
d
c
a
n
t
a
k
e
a
l
i
f
e
.
M
a
n
u
al
d
e
t
e
c
t
i
o
n
f
r
o
m
r
a
d
i
o
l
o
g
is
ts
r
eq
u
i
r
e
s
l
o
ts
o
f
i
n
v
e
s
ti
g
a
t
i
o
n
s
a
n
d
l
o
t
s
o
f
a
n
a
l
y
s
i
s
.
I
n
t
h
e
tu
m
o
r
i
d
e
n
t
i
f
i
c
a
ti
o
n
p
r
o
c
e
s
s
,
e
v
e
r
y
c
l
o
c
k
o
f
t
i
m
e
i
s
e
s
s
e
n
t
i
al
.
I
f
d
e
t
e
c
t
e
d
a
n
d
a
n
a
l
y
z
e
d
w
i
t
h
i
n
t
h
e
t
i
m
e
f
r
a
m
e
,
i
t
b
e
c
o
m
e
s
e
as
i
e
r
t
o
o
f
f
e
r
t
h
e
p
r
o
p
e
r
t
r
e
a
t
m
e
n
t
t
o
t
h
e
p
a
t
i
e
n
t
[
9
]
,
[
1
0
]
.
T
h
i
s
d
i
s
e
as
e
i
s
e
n
t
i
r
el
y
c
u
r
a
t
i
v
e
,
o
n
l
y
t
i
m
ely
i
d
e
n
t
i
f
i
c
a
ti
o
n
i
s
a
s
i
g
n
i
f
i
c
a
n
t
c
o
n
c
e
r
n
,
a
n
d
h
e
n
c
e
m
a
n
y
r
e
s
e
a
r
c
h
e
r
s
a
r
e
e
m
p
l
o
y
in
g
t
h
e
i
r
s
k
i
l
ls
t
o
d
e
v
e
l
o
p
n
e
w
a
d
v
a
n
c
e
d
m
e
t
h
o
d
s
f
o
r
i
d
e
n
t
i
f
y
i
n
g
t
h
e
t
u
m
o
r
w
i
t
h
im
p
r
o
v
i
s
i
n
g
p
e
r
f
o
r
m
a
n
c
e
p
a
r
a
m
e
t
e
r
s
[
1
1
]
–
[
1
3
]
.
Var
io
u
s
im
ag
in
g
m
o
d
alities
ar
e
av
ailab
le
f
o
r
ca
p
tu
r
in
g
tu
m
o
r
in
f
o
r
m
atio
n
f
r
o
m
th
e
b
r
ain
;
th
e
m
o
s
t
p
o
p
u
lar
ar
e
o
n
e
o
f
th
em
is
co
m
p
u
ted
to
m
o
g
r
ap
h
y
(
C
T
)
an
d
th
e
o
th
er
o
n
e
m
ag
n
etic
r
eso
n
an
ce
im
ag
in
g
(
MRI)
.
T
h
e
MRI
is
h
ig
h
l
y
p
r
e
f
er
ab
le
b
ec
au
s
e
o
f
its
n
o
n
-
in
v
asiv
e
n
atu
r
e,
n
o
r
ad
iatio
n
,
an
d
n
o
h
ar
m
[
1
4
]
,
[
1
5
]
.
Fo
r
id
en
tific
atio
n
an
d
d
etec
tio
n
o
f
lo
ca
tio
n
an
d
ex
tr
ac
t
m
ea
n
in
g
f
u
l
in
f
o
r
m
atio
n
f
r
o
m
th
e
b
r
ain
tu
m
o
r
,
r
ad
i
o
lo
g
is
ts
p
er
f
o
r
m
two
m
ain
ac
tiv
ities
:
i)
Dif
f
er
en
tiatio
n
o
f
th
e
b
r
ain
M
R
im
ag
e
ch
ar
ac
ter
is
tics
,
i.e
.
,
u
n
u
s
u
al
(
ab
n
o
r
m
al
)
o
r
ty
p
ical
(
n
o
r
m
al)
;
an
d
ii)
C
lass
if
y
in
g
ab
n
o
r
m
al
b
r
ain
tis
s
u
es
in
to
v
ar
io
u
s
t
u
m
o
r
ty
p
es
(
g
r
ad
es)
b
ased
o
n
th
eir
s
p
r
ea
d
.
B
as
e
d
o
n
t
h
e
s
e
a
c
t
i
v
it
i
es
,
b
r
a
i
n
b
e
h
a
v
i
o
r
c
a
n
b
e
s
t
u
d
i
e
d
a
n
d
a
n
a
l
y
z
e
d
e
f
f
e
c
t
i
v
e
l
y
.
T
h
e
p
a
p
e
r
b
r
i
e
f
s
,
as
i
l
l
u
s
t
r
a
t
e
d
,
a
r
e
t
h
e
li
t
e
r
a
t
u
r
e
r
ev
i
e
w
d
i
s
c
u
s
s
e
d
i
n
s
e
c
ti
o
n
2
.
T
h
e
p
r
o
p
o
s
e
d
m
e
t
h
o
d
i
n
s
e
c
t
i
o
n
3
a
n
d
m
a
t
e
r
i
a
l
s
i
n
s
e
c
ti
o
n
4
. R
es
u
l
ts
a
n
d
d
i
s
c
u
s
s
io
n
i
n
s
e
c
t
i
o
n
5
. F
i
n
a
l
l
y
,
s
e
c
ti
o
n
6
g
i
v
e
s
t
h
e
c
o
n
c
l
u
s
i
o
n
o
f
t
h
e
p
r
o
p
o
s
e
d
s
y
s
te
m
.
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
B
ased
o
n
th
e
s
ev
er
ity
,
i.e
.
,
wh
eth
er
th
e
tu
m
o
r
b
elo
n
g
s
to
m
a
lig
n
an
cy
o
r
b
en
i
g
n
ity
,
b
r
ain
t
u
m
o
r
s
ar
e
g
r
ad
ed
f
r
o
m
I
-
I
V
ty
p
ec
asts
ca
teg
o
r
ies b
y
th
e
W
o
r
ld
Hea
lth
Or
g
an
izatio
n
(
W
HO)
.
T
h
e
g
r
o
wth
r
ate
o
f
Gr
ad
e
I
I
an
d
I
V
m
alig
n
an
t
ty
p
es
o
f
b
r
ain
tu
m
o
r
s
h
av
e
h
a
v
in
g
f
ast
er
g
r
o
wth
r
ate,
co
m
p
ar
ed
to
o
th
er
g
r
ad
es.
T
h
e
y
s
p
r
ea
d
at
a
h
ig
h
er
r
ate
to
o
th
e
r
b
o
d
y
p
ar
ts
an
d
im
p
in
g
e
o
n
h
ea
lth
y
ce
lls
[
1
6
]
–
[
1
9
]
.
R
am
am
o
o
r
th
y
et
a
l.
[
2
0
]
in
v
esti
g
ated
an
d
p
r
o
p
o
s
ed
a
tech
n
iq
u
e,
co
n
s
is
tin
g
o
f
p
r
e
-
p
r
o
ce
s
s
in
g
u
s
in
g
h
is
to
g
r
am
e
q
u
aliza
tio
n
with
a
lear
n
in
g
-
b
ased
q
u
an
tizatio
n
m
o
d
el
f
o
r
d
etec
tin
g
tu
m
o
r
s
in
th
e
b
r
ain
.
W
ith
th
is
tech
n
iq
u
e,
t
h
ey
h
av
e
ac
h
iev
ed
,
m
o
d
er
ate
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
an
d
s
p
ec
if
icity
with
9
3
%,
9
2
%,
an
d
9
4
%,
r
esp
ec
tiv
el
y
.
Acc
o
r
d
in
g
to
Asaf
R
az
a
et
a
l.
[
2
1
]
,
tr
ad
itio
n
al
m
ac
h
in
e
lear
n
in
g
-
b
ased
class
if
ier
r
eq
u
ir
es
h
a
n
d
-
c
r
af
ted
f
e
atu
r
es,
wh
ich
r
eq
u
ir
e
a
lo
t
o
f
tim
e.
So
,
th
ey
in
v
esti
g
a
ted
an
d
p
r
o
p
o
s
ed
a
tech
n
iq
u
e
th
at
em
p
l
o
y
s
a
h
y
b
r
id
d
ee
p
l
ea
r
n
in
g
m
o
d
el
th
at
ef
f
ec
tiv
ely
d
etec
ts
an
d
class
if
ies
th
r
ee
ty
p
es
o
f
b
r
ain
tu
m
o
r
,
n
am
ely
,
g
lio
m
a,
m
en
i
n
g
i
o
m
a,
a
n
d
p
itu
itar
y
tu
m
o
r
.
Acc
o
r
d
in
g
to
th
em
,
t
h
eir
tech
n
iq
u
e
ad
o
p
ts
a
p
r
im
ar
y
co
n
v
o
lu
tio
n
n
e
u
r
al
n
etwo
r
k
-
b
ased
ar
ch
itectu
r
e,
an
d
with
th
is
,
th
ey
claim
h
ig
h
er
p
r
ec
is
io
n
an
d
ac
cu
r
ac
y
as we
ll a
s
1
0
0
% r
ec
all
an
d
an
F1
s
co
r
e
o
f
ab
o
v
e
9
9
%.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
1
,
Feb
r
u
ar
y
20
25
:
9
5
8
-
969
960
Salm
an
et
a
l.
[
2
2
]
in
v
esti
g
at
ed
th
e
h
y
b
r
id
m
et
h
o
d
o
lo
g
y
o
f
im
ag
e
p
r
o
ce
s
s
in
g
tech
n
iq
u
es
an
d
claim
ed
an
ac
cu
r
ac
y
r
ate
o
f
u
p
to
9
5
% f
o
r
s
eg
m
en
tatio
n
an
d
s
ep
ar
atio
n
o
f
an
a
r
ea
o
f
i
n
ter
est in
b
r
ain
tu
m
o
r
s
.
R
aso
o
l
et
a
l.
[
2
3
]
p
r
esen
ted
a
h
ig
h
ly
e
f
f
icien
t
h
y
b
r
i
d
d
ee
p
l
ea
r
n
in
g
m
o
d
el,
wh
ich
is
ca
n
b
e
u
s
ed
f
o
r
th
e
class
if
icatio
n
o
f
b
r
ain
tu
m
o
r
s
.
T
h
is
m
eth
o
d
u
tili
ze
s
a
n
ew
h
y
b
r
id
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
(
C
NN)
-
b
ased
ar
ch
itectu
r
e
to
class
if
y
b
r
ain
tu
m
o
r
s
.
T
h
e
p
r
o
p
o
s
ed
te
ch
n
iq
u
e
co
m
p
r
o
m
is
es
o
n
two
d
if
f
er
en
t
s
ce
n
ar
io
s
,
th
e
f
ir
s
t
is
b
ased
o
n
a
p
r
e
-
tr
ain
ed
Go
o
g
le
-
N
et
m
o
d
el
with
a
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
f
o
r
p
atter
n
class
if
icatio
n
,
an
d
th
e
s
ec
o
n
d
in
teg
r
ates
a
f
in
ely
t
u
n
ed
G
o
o
g
le
-
Net.
T
h
e
f
ir
s
t
ap
p
r
o
ac
h
tu
n
ed
th
e
p
r
o
p
o
s
ed
alg
o
r
ith
m
with
im
p
r
o
v
e
d
ac
cu
r
ac
y
to
9
8
.
1
%,
wh
e
r
ea
s
th
e
s
ec
o
n
d
a
p
p
r
o
ac
h
p
r
o
d
u
ce
d
an
ac
cu
r
ac
y
o
f
9
3
.
1
%.
Ali
et
a
l.
[
2
4
]
p
r
esen
ted
a
n
atten
tio
n
-
b
ased
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
ap
p
r
o
ac
h
f
o
r
th
e
s
eg
m
en
tatio
n
o
f
b
r
ai
n
tu
m
o
r
s
.
T
h
is
tech
n
iq
u
e
is
em
p
lo
y
ed
w
ith
a
m
ec
h
an
is
m
f
o
r
av
o
id
i
n
g
o
v
er
f
itti
n
g
.
R
esu
lts
o
f
th
eir
wo
r
k
h
a
v
e
b
ee
n
co
m
p
ar
ed
with
s
o
m
e
o
f
th
e
r
ec
en
t
ex
is
tin
g
p
r
o
v
e
n
tech
n
i
q
u
es
an
d
h
av
e
p
r
esen
ted
q
u
an
titativ
e
m
ea
s
u
r
es
with
9
8
%,
9
8
.
1
%,
9
9
%,
an
d
9
9
.
3
%
o
f
s
en
s
itiv
ity
,
s
p
ec
if
icity
,
ac
cu
r
ac
y
,
an
d
p
r
ec
is
io
n
,
r
esp
ec
tiv
ely
.
Acc
o
r
d
in
g
to
[
2
5
]
,
[
2
6
]
,
th
e
p
r
esen
t
m
an
u
al
m
e
th
o
d
o
lo
g
y
em
p
lo
y
ed
b
y
r
ad
i
o
l
o
g
is
ts
o
r
e
x
p
er
ts
to
d
etec
t
in
f
ec
tio
n
in
t
h
e
b
r
ai
n
c
o
n
s
u
m
es
a
lo
t
o
f
tim
e
an
d
is
p
r
o
n
e
t
o
h
u
m
an
er
r
o
r
s
.
T
h
is
h
ap
p
en
s
d
u
e
to
th
e
en
o
r
m
o
u
s
v
o
l
u
m
e
o
f
ca
s
es
an
d
d
ep
e
n
d
s
o
n
th
e
ex
p
er
ien
ce
o
f
th
e
ex
p
e
r
t.
So
,
th
ey
in
v
esti
g
ated
an
d
ass
ess
ed
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
ar
tific
ial
b
ee
c
o
lo
n
y
(
AB
C
)
alg
o
r
ith
m
f
o
r
a
d
ap
tiv
e
g
lio
b
last
o
m
a
d
etec
tio
n
,
a
n
d
th
eir
m
eth
o
d
r
ea
c
h
ed
a
n
ac
cu
r
ac
y
f
o
r
g
lio
b
last
o
m
a
d
etec
tio
n
u
p
t
o
9
3
.
6
7
%.
Fals
e
d
etec
tio
n
o
f
b
r
ain
t
u
m
o
r
s
lead
s
to
wr
o
n
g
m
ed
ic
al
in
ter
v
en
tio
n
,
w
h
ich
r
e
d
u
c
es
p
atien
ts
’
ch
an
ce
s
o
f
s
u
r
v
iv
al.
Sen
a
n
et
a
l.
[
2
7
]
h
av
e
p
r
o
p
o
s
ed
a
n
in
d
ig
en
o
u
s
m
eth
o
d
f
o
r
d
etec
tin
g
b
r
ain
tu
m
o
r
f
r
o
m
MRI
im
ag
es
u
s
in
g
a
h
y
b
r
id
t
ec
h
n
iq
u
e
to
f
asten
th
e
au
to
m
atic
d
etec
tio
n
o
f
b
r
ai
n
tu
m
o
r
s
.
I
n
th
eir
r
esear
ch
,
th
ey
in
co
r
p
o
r
ated
d
ee
p
lear
n
in
g
,
tr
ad
itio
n
al
m
ac
h
in
e
lear
n
in
g
tech
n
i
q
u
es,
an
d
SVM
as
a
class
if
icat
io
n
.
Acc
o
r
d
in
g
t
o
th
e
a
u
th
o
r
s
,
t
h
e
d
ee
p
lea
r
n
in
g
tech
n
iq
u
e
b
ased
o
n
th
e
Alex
Net
alg
o
r
ith
m
an
d
SVM
class
if
icatio
n
ex
h
ib
its
ac
cu
r
ac
y
,
s
en
s
itiv
ity
,
an
d
s
p
ec
if
icity
o
f
9
5
.
1
0
%,
9
5
.
2
5
%,
a
n
d
9
8
.
5
0
% r
esp
ec
tiv
ely
.
Hash
em
ze
h
i
et
a
l.
[
2
8
]
h
av
e
in
v
esti
g
ated
a
n
ew
tech
n
iq
u
e
f
o
r
MRI
b
ased
b
r
ain
tu
m
o
r
d
etec
tio
n
,
wh
ich
is
b
ased
o
n
d
ee
p
lea
r
n
in
g
u
s
in
g
t
h
e
in
teg
r
atio
n
o
f
C
NN
an
d
n
eu
r
al
au
to
-
r
eg
r
ess
iv
e
d
is
tr
ib
u
tio
n
esti
m
atio
n
(
NADE
)
.
An
ef
f
ec
tiv
e
s
tr
ateg
y
h
as
b
ee
n
p
lan
n
e
d
an
d
s
h
o
wca
s
ed
b
y
Mittal
et
a
l.
[
2
9
]
f
o
r
b
r
ain
tu
m
o
r
d
etec
tio
n
a
n
d
s
eg
m
en
ta
tio
n
o
f
tu
m
o
r
,
u
s
in
g
a
n
e
n
h
an
ce
d
d
ee
p
lear
n
i
n
g
c
o
n
ce
p
t,
w
h
ich
is
in
ter
r
o
g
ated
b
ased
o
n
s
tatio
n
ar
y
wav
elet
tr
an
s
f
o
r
m
(
SW
T
)
an
d
th
e
n
ew
g
r
o
win
g
co
n
v
o
lu
tio
n
n
eu
r
al
n
et
wo
r
k
(
GC
NN)
.
T
o
p
r
o
v
e
th
e
v
alid
ity
o
f
t
h
e
p
r
o
p
o
s
ed
s
y
s
tem
,
th
e
au
t
h
o
r
s
h
av
e
p
r
esen
te
d
p
e
r
f
o
r
m
an
ce
m
etr
ics
in
ter
m
s
o
f
ac
cu
r
ac
y
,
p
ea
k
-
s
ig
n
al
-
to
-
n
o
is
e
r
atio
(
PS
NR
)
,
an
d
m
ea
n
s
q
u
ar
ed
er
r
o
r
(
MSE
)
.
Acc
o
r
d
in
g
ly
,
th
e
co
m
b
in
atio
n
o
f
SW
T
an
d
GC
NN
s
h
o
ws a
s
ig
n
if
ican
t im
p
r
o
v
em
e
n
t in
th
e
s
eg
m
en
tatio
n
au
to
m
atio
n
p
r
o
c
ess
an
d
co
n
tr
ib
u
tes
to
th
e
r
ed
u
ctio
n
o
f
m
ea
n
s
q
u
ar
e
er
r
o
r
co
m
p
a
r
ed
to
co
n
v
en
tio
n
al
C
NN
m
eth
o
d
o
lo
g
y
.
Ar
u
n
ac
h
alam
an
d
Seth
u
m
ath
av
an
[
3
0
]
i
n
tr
o
d
u
ce
d
an
im
p
r
o
v
ed
YOL
O5
-
b
ased
tech
n
iq
u
e
f
o
r
b
r
ai
n
tu
m
o
r
d
etec
tio
n
.
T
h
e
s
eg
m
en
tatio
n
is
ac
h
iev
ed
u
s
i
n
g
th
e
Mc
C
u
llo
ch
m
eth
o
d
.
T
h
e
s
y
s
tem
g
i
v
es
an
ac
cu
r
ac
y
o
f
9
9
.
3
2
%
an
d
an
F1
s
co
r
e
o
f
9
1
.
2
6
%.
B
ab
u
et
a
l.
[
3
1
]
p
r
esen
ted
a
f
u
lly
au
to
m
ated
s
y
s
tem
u
s
in
g
f
o
u
r
f
o
u
r
-
s
tag
e
p
r
o
ce
s
s
es.
T
h
e
C
u
r
v
elet
tr
an
s
f
o
r
m
atio
n
is
u
s
ed
i
n
th
e
f
ir
s
t
s
tep
f
o
r
im
a
g
e
d
e
-
n
o
is
i
n
g
.
Ar
tific
ial
b
ee
c
o
lo
n
y
(
A
B
C
)
o
p
tim
izatio
n
is
ap
p
lied
to
r
em
o
v
e
in
f
ec
te
d
ar
ea
s
f
r
o
m
MRI
s
ca
n
s
in
t
h
e
n
e
x
t
s
tag
e.
I
n
th
e
th
ir
d
s
tag
e,
to
r
ec
o
v
er
t
h
e
lear
n
i
n
g
r
ate,
an
o
th
er
o
p
tim
izatio
n
b
ased
o
n
C
NN
is
u
s
ed
.
T
h
e
en
tire
s
y
s
tem
ex
p
er
im
en
ted
o
n
B
R
AT
S
2
0
1
3
an
d
2
0
1
5
d
atasets
an
d
ac
h
iev
ed
a
class
if
icatio
n
ac
cu
r
ac
y
o
f
9
8
.
5
%
an
d
9
9
.
0
%
f
o
r
b
o
t
h
d
atasets
,
r
esp
ec
tiv
ely
.
T
h
e
co
m
b
in
atio
n
o
f
C
NN
an
d
Ha
ar
wav
elet
f
ea
tu
r
es
h
as
b
ee
n
i
n
v
esti
g
ated
b
y
Dh
ee
p
a
an
d
S
h
an
k
ar
i
[
3
2
]
f
o
r
th
e
au
to
m
atic
id
en
tific
atio
n
o
f
in
f
ec
ted
ar
ea
s
f
r
o
m
MR
im
ag
es.
T
h
e
v
alid
atio
n
o
f
th
e
alg
o
r
ith
m
is
ex
p
er
im
en
ted
with
u
s
in
g
th
e
B
R
AT
S
2
0
1
8
d
ataset
an
d
ac
h
iev
e
d
an
F1
s
c
o
r
e
o
f
9
7
%,
p
r
ec
is
io
n
o
f
9
7
%,
s
en
s
itiv
ity
o
f
9
6
%,
s
p
ec
if
icity
o
f
9
7
%,
an
d
ac
c
u
r
ac
y
o
f
9
6
%.
Ma
n
y
r
esear
ch
er
s
witn
ess
th
e
C
NN
-
b
ased
t
ec
h
n
iq
u
es
f
o
r
b
r
ain
tu
m
o
r
class
if
icatio
n
,
b
u
t
l
o
ca
l
b
ac
k
g
r
o
u
n
d
in
f
o
r
m
atio
n
is
r
estricte
d
in
lo
ca
l
C
NN.
T
h
is
p
r
o
b
lem
is
ad
d
r
ess
e
d
b
y
Sil
le
et
a
l.
[
3
3
]
b
y
i
n
v
esti
g
atin
g
a
d
ee
p
co
n
v
o
lu
tio
n
al
g
e
n
er
ativ
e
ad
v
er
s
ar
ial
n
etwo
r
k
.
Fo
r
v
alid
atio
n
o
f
th
e
alg
o
r
ith
m
’
s
p
e
r
f
o
r
m
an
ce
,
t
h
e
d
ice
s
co
r
e
co
e
f
f
icien
t
(
DSC
)
,
p
ea
k
-
s
ig
n
al
-
to
-
n
o
is
e
r
atio
(
PS
NR
)
,
an
d
s
tr
u
ctu
r
al
in
d
ex
s
im
ilar
ity
(
SS
I
M)
ar
e
ca
lcu
lated
an
d
attain
ed
9
7
%
a
cc
u
r
ac
y
with
lo
s
s
r
ed
u
ce
d
to
0
.
0
1
2
.
R
ed
d
y
an
d
Dh
u
li
[
3
4
]
p
r
o
p
o
s
ed
a
s
eg
m
e
n
tatio
n
o
f
b
r
ain
tu
m
o
r
in
f
ec
ti
o
n
f
r
o
m
MR
im
ag
es
u
s
in
g
a
f
ast
-
lin
k
in
g
m
o
d
if
ied
s
p
ik
in
g
co
r
tical
m
o
d
el
(
FL
-
M
SC
M)
an
d
p
er
f
o
r
m
ed
th
e
class
if
icatio
n
u
s
in
g
a
lig
h
tweig
h
t c
o
n
v
o
lu
ti
o
n
al
n
eu
r
al
n
etwo
r
k
(
lig
h
tweig
h
t
C
NN)
m
o
d
el.
T
h
e
ex
p
e
r
im
en
tal
an
aly
s
is
ac
h
iev
ed
d
ice
s
im
ilar
ity
co
e
f
f
icien
t
(
DSC
)
an
d
ac
cu
r
ac
y
in
class
if
icatio
n
o
f
9
5
.
7
% a
n
d
9
9
.
5
8
%.
I
n
th
e
liter
atu
r
e
r
e
v
iew
d
is
cu
s
s
ed
ab
o
v
e,
we
h
av
e
s
ee
n
v
a
r
io
u
s
tu
m
o
r
d
etec
tio
n
an
d
cla
s
s
if
icatio
n
m
eth
o
d
o
l
o
g
ies,
m
o
s
tly
b
ased
o
n
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
S
VM
)
,
ar
tific
ial
n
eu
r
al
n
etwo
r
k
(
ANN)
,
k
-
n
ea
r
est
n
eu
r
al
n
etwo
r
k
(
K
-
NN)
,
an
d
co
n
v
o
l
u
tio
n
al
n
eu
r
al
n
etwo
r
k
(
C
NN)
.
T
o
im
p
r
o
v
e
t
h
e
s
u
r
v
iv
al
o
f
th
e
p
atien
ts
,
Alg
an
i
et
a
l.
[
3
5
]
e
m
p
lo
y
e
d
b
in
ar
y
g
r
ay
wo
lf
o
p
tim
izatio
n
-
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
-
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
(
B
GW
O
-
C
NN
-
L
ST
M
)
-
b
ased
tech
n
iq
u
e
a
n
d
ac
h
ie
v
ed
a
s
p
ec
if
icity
o
f
9
9
.
5
4
%,
r
ec
all
o
f
9
9
.
2
3
%,
an
d
ac
cu
r
ac
y
o
f
9
9
.
7
4
%.
T
h
e
f
iv
e
-
s
tag
e
m
o
d
el
f
o
r
th
e
i
d
en
tific
atio
n
o
f
in
f
ec
ted
b
r
ain
ar
ea
s
f
r
o
m
MR
im
ag
es
is
s
u
g
g
ested
b
y
R
am
tek
k
ar
et
a
l
.
[
3
6
]
.
I
n
th
e
f
ir
s
t
s
tag
e,
p
r
e
-
p
r
o
ce
s
s
in
g
is
p
er
f
o
r
m
ed
u
s
in
g
a
C
o
m
p
o
u
n
d
f
ilter
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
B
erkeley
w
a
ve
let
tr
a
n
s
fo
r
m
a
n
d
I
mp
r
o
ve
d
YOLOv7
-
b
a
s
ed
…
(
N
iles
h
B
h
a
s
ka
r
r
a
o
B
a
h
a
d
u
r
e
)
961
I
n
th
e
s
ec
o
n
d
s
tag
e,
s
eg
m
e
n
tatio
n
is
ac
h
iev
e
d
u
s
in
g
th
r
esh
o
ld
an
d
h
is
to
g
r
am
tech
n
i
q
u
es.
Featu
r
es
ar
e
ex
tr
ac
ted
,
a
n
d
o
p
tim
izatio
n
o
f
b
est
f
ea
tu
r
es
is
p
er
f
o
r
m
ed
in
th
e
th
ir
d
an
d
f
o
u
r
t
h
s
tag
e
s
u
s
in
g
g
r
ey
lev
el
co
-
o
cc
u
r
r
en
ce
m
atr
ix
(
GL
C
M)
,
wh
ale
o
p
tim
izatio
n
,
an
d
g
r
e
y
wo
lf
o
p
tim
izatio
n
[
3
7
]
.
Fin
a
lly
,
C
NN
is
ap
p
lied
f
o
r
th
e
class
if
icatio
n
an
d
attain
s
9
8
.
2
% d
etec
tio
n
ac
cu
r
ac
y
.
Desp
ite
m
u
ch
r
esear
ch
,
th
e
ex
is
tin
g
r
esear
ch
wo
r
k
ca
n
n
o
t
m
ee
t
th
e
ex
p
ec
tatio
n
s
.
Hen
ce
a
n
ew
m
eth
o
d
o
l
o
g
y
is
ex
ig
en
tly
r
eq
u
ir
ed
b
y
in
tr
o
d
u
cin
g
a
n
ew
alg
o
r
ith
m
f
o
r
b
r
ain
tu
m
o
r
d
etec
tio
n
.
T
h
e
o
u
tco
m
es
o
f
o
u
r
p
r
o
p
o
s
ed
tech
n
iq
u
e
d
e
m
o
n
s
tr
ate
th
e
ef
f
ec
tiv
en
ess
an
d
h
ig
h
p
e
r
f
o
r
m
an
ce
o
v
er
th
e
o
th
er
r
ec
en
t
wo
r
k
m
en
tio
n
ed
in
th
e
liter
atu
r
e.
I
n
ad
d
itio
n
to
th
is
,
th
e
p
r
esen
t
wo
r
k
will
ass
i
s
t
d
o
cto
r
s
an
d
r
ad
io
lo
g
is
ts
in
th
e
d
iag
n
o
s
is
o
f
b
r
ain
tu
m
o
r
s
in
p
atien
t
s
tu
d
ies
in
th
e
liter
atu
r
e.
I
n
ad
d
itio
n
,
th
is
r
esear
ch
will
u
p
h
o
ld
d
o
cto
r
s
an
d
r
ad
io
lo
g
is
ts
f
o
r
a
u
to
m
atic
d
iag
n
o
s
is
o
f
b
r
ain
t
u
m
o
r
d
is
ea
s
e.
3.
P
RO
P
O
SE
D
M
E
T
H
O
D
Fig
u
r
e
2
s
h
o
ws
v
ar
io
u
s
s
tag
es
u
s
ed
in
th
e
p
r
o
p
o
s
ed
s
y
s
tem
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
tem
u
tili
ze
s
p
r
e
-
p
r
o
ce
s
s
in
g
,
e
n
h
an
ce
m
en
t,
s
k
u
ll
s
tr
ip
p
in
g
,
s
eg
m
en
tatio
n
,
ex
tr
ac
tio
n
o
f
tu
m
o
r
ar
ea
,
f
ea
tu
r
e
ex
tr
ac
tio
n
a
n
d
s
elec
tio
n
o
f
r
elev
a
n
t
f
ea
tu
r
e
s
,
an
d
f
in
ally
,
class
if
icatio
n
b
ased
o
n
im
p
r
o
v
ed
d
ee
p
lea
r
n
in
g
C
NN
b
ased
ar
ch
itectu
r
e
YOL
Ov
7
.
T
h
e
r
a
w
im
ag
es
m
ig
h
t
h
av
e
u
n
wan
t
ed
n
o
is
e
an
d
u
n
wan
te
d
p
ar
ts
,
wh
ich
ar
e
r
em
o
v
e
d
at
th
e
p
r
e
-
p
r
o
ce
s
s
in
g
s
tag
e.
T
h
e
p
r
e
-
p
r
o
ce
s
s
in
g
h
elp
s
i
n
im
p
r
o
v
in
g
th
e
s
ig
n
al
-
to
-
n
o
is
e
r
atio
,
with
o
u
t
h
am
p
er
in
g
th
e
o
th
er
r
e
g
io
n
s
o
f
th
e
im
ag
e.
T
h
e
o
v
er
all
clar
it
y
o
f
th
e
r
aw
MR
im
ag
es
is
en
h
an
ce
d
b
y
ap
p
ly
in
g
th
e
en
h
an
ce
m
en
t th
r
o
u
g
h
ad
ap
tiv
e
co
n
tr
ast en
h
a
n
ce
m
en
t b
as
ed
o
n
t
h
e
m
o
d
if
ied
s
ig
m
o
id
f
u
n
ctio
n
[
3
8
]
,
[
3
9
]
.
Fig
u
r
e
2
.
Step
s
u
s
ed
in
th
e
p
r
o
p
o
s
ed
alg
o
r
it
h
m
B
r
ain
MR
im
ag
es
h
av
e
ad
d
itio
n
al
tis
s
u
es,
s
u
ch
as
th
e
s
k
u
ll,
s
k
in
,
an
d
f
at,
wh
ich
s
h
o
u
ld
b
e
r
em
o
v
ed
b
ef
o
r
e
th
e
s
eg
m
e
n
tatio
n
;
o
th
er
wis
e,
it
will
af
f
ec
t
th
e
p
e
r
f
o
r
m
a
n
ce
o
f
th
e
s
eg
m
en
tati
o
n
.
Sk
u
ll
-
s
tr
ip
p
in
g
o
p
er
atio
n
[
4
0
]
is
u
s
ed
to
eli
m
in
ate
th
ese
tis
s
u
es.
Fo
r
ef
f
ec
tiv
e
s
k
u
ll
-
s
tr
ip
p
in
g
,
s
ev
er
al
m
eth
o
d
o
l
o
g
ies
ar
e
av
ailab
le.
W
e
em
p
lo
y
ed
a
th
r
esh
o
ld
-
b
ased
[
4
1
]
s
k
u
ll
-
s
tr
ip
p
in
g
m
eth
o
d
in
o
u
r
s
y
s
tem
.
T
h
r
esh
o
ld
-
b
ased
s
k
u
ll
s
tr
ip
p
in
g
is
s
tr
aig
h
tf
o
r
war
d
.
I
t
s
im
p
ly
r
em
o
v
es
th
e
s
k
u
ll
f
r
o
m
th
e
b
r
ai
n
MRI
b
y
s
ettin
g
th
e
th
r
esh
o
ld
v
alu
e.
I
n
o
u
r
ca
s
e,
we
s
et
th
e
th
r
esh
o
ld
v
alu
e
at
0
.
7
,
wh
ich
wo
r
k
s
f
a
n
tast
ically
.
E
v
en
th
o
u
g
h
th
is
ad
d
itio
n
al
ce
r
e
b
r
al
tis
s
u
e
(
s
k
u
ll)
will n
o
t a
f
f
ec
t th
e
r
esu
lt r
em
o
v
in
g
th
is
en
s
u
r
es
m
o
r
e
clar
ity
in
th
e
s
eg
m
en
tati
o
n
o
p
e
r
atio
n
.
F
e
a
t
u
r
e
e
x
t
r
a
ct
i
o
n
is
u
s
e
d
t
o
r
e
d
u
c
e
i
m
a
g
e
c
o
m
p
l
e
x
i
t
y
a
n
d
r
e
d
u
c
e
t
h
e
b
u
r
d
e
n
o
n
t
h
e
c
l
a
s
s
i
f
i
c
a
ti
o
n
s
y
s
te
m
.
W
i
t
h
f
e
at
u
r
e
e
x
t
r
a
ct
i
o
n
,
t
u
m
o
r
d
e
t
e
c
ti
o
n
,
a
n
d
i
t
s
c
l
as
s
i
f
i
c
at
i
o
n
p
r
o
c
e
s
s
a
r
e
m
o
r
e
e
f
f
i
c
i
e
n
t,
s
t
a
te
-
f
o
r
w
a
r
d
,
a
c
c
u
r
a
t
e
,
a
n
d
e
f
f
i
c
i
e
n
t
.
W
i
t
h
f
e
a
t
u
r
e
e
x
t
r
a
ct
i
o
n
,
h
i
g
h
-
d
i
m
e
n
s
i
o
n
s
p
a
c
e
ca
n
b
e
m
a
p
p
e
d
t
o
l
e
s
s
-
d
i
m
e
n
s
i
o
n
s
p
a
c
e
i
n
li
n
e
ar
o
r
n
o
n
-
l
i
n
e
a
r
s
y
s
t
e
m
s
.
I
n
im
a
g
e
p
r
o
c
e
s
s
i
n
g
,
e
s
p
e
ci
a
ll
y
in
m
e
d
i
c
a
l
i
m
a
g
i
n
g
,
e
s
s
e
n
t
i
a
l
f
ea
t
u
r
e
s
s
u
c
h
a
s
c
o
n
t
r
a
s
t
,
s
h
a
p
e
,
t
e
x
t
u
r
e
,
a
n
d
c
o
l
o
r
a
r
e
e
x
t
r
a
c
t
e
d
t
o
r
e
d
u
c
e
i
m
a
g
e
c
o
m
p
l
e
x
i
t
y
[
4
2
]
.
G
r
a
y
l
e
v
e
l
c
o
-
o
c
c
u
r
r
e
n
c
e
m
a
t
r
i
x
(
GL
C
M
)
a
n
d
t
e
x
t
u
r
e
f
e
a
t
u
r
e
a
r
e
t
h
e
m
o
s
t
w
i
d
e
l
y
u
s
e
d
i
m
a
g
e
a
n
a
l
y
s
is
a
p
p
l
i
c
a
ti
o
n
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
1
,
Feb
r
u
ar
y
20
25
:
9
5
8
-
969
962
i
n
t
r
o
d
u
c
e
d
b
y
H
a
r
a
l
i
c
k
e
t
a
l
.
[
4
3
]
.
F
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
i
s
u
s
e
d
t
o
r
e
d
u
c
e
t
h
e
c
l
a
s
s
i
f
ie
r
’
s
c
o
m
p
l
e
x
i
t
y
s
o
c
l
a
s
s
i
f
ic
a
t
i
o
n
o
f
a
n
i
m
a
g
e
i
s
s
i
m
p
l
i
f
i
e
d
a
n
d
F
a
s
tl
y
p
r
o
c
e
s
s
e
d
.
W
i
t
h
o
u
t
f
e
a
t
u
r
e
e
x
t
r
a
c
ti
o
n
a
n
d
o
p
t
i
m
i
z
a
t
i
o
n
t
h
e
r
e
o
f
w
i
ll
a
f
f
e
c
t
t
h
e
r
e
s
u
l
t
s
b
a
d
l
y
.
T
h
e
e
x
p
e
r
i
m
e
n
t
a
l
a
n
a
l
y
s
is
is
a
ls
o
p
e
r
f
o
r
m
e
d
w
i
t
h
o
u
t
f
ea
t
u
r
e
o
p
t
i
m
i
z
at
i
o
n
,
a
n
d
t
h
e
r
e
s
u
l
t
s
y
i
e
l
d
i
s
n
o
t
a
p
p
r
e
c
i
a
b
l
e
.
T
h
e
f
e
a
t
u
r
e
e
x
t
r
a
c
ti
o
n
p
r
o
c
e
s
s
e
x
t
r
a
c
ts
t
h
e
r
e
l
e
v
a
n
t i
n
f
o
r
m
a
t
i
o
n
f
r
o
m
t
h
e
i
n
f
e
c
t
e
d
i
m
a
g
e
a
n
d
g
i
v
e
s
p
r
e
c
is
e
i
n
f
o
r
m
a
t
i
o
n
t
o
t
h
e
c
la
s
s
i
f
i
e
r
t
o
c
l
a
s
s
i
f
y
t
h
e
m
e
f
f
e
c
ti
v
e
l
y
.
I
n
t
h
e
f
i
n
a
l
r
e
s
u
l
ts
o
f
t
h
e
c
l
a
s
s
i
f
i
e
r
,
5
%
t
o
1
0
%
o
f
t
h
e
d
e
v
i
a
t
i
o
n
i
s
n
o
t
i
c
e
d
w
h
e
n
p
e
r
f
o
r
m
e
d
w
i
t
h
o
u
t
t
h
e
r
e
l
e
v
a
n
t
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
.
I
n
l
i
f
e
-
t
h
r
e
a
t
e
n
i
n
g
d
i
s
e
as
es
,
v
a
r
i
a
ti
o
n
i
n
p
r
o
c
e
s
s
i
n
g
t
i
m
e
b
y
s
o
m
e
m
i
n
u
t
e
s
m
a
y
p
l
a
y
a
s
i
g
n
i
f
i
c
an
t
a
n
d
d
e
c
i
s
i
v
e
r
o
l
e,
b
u
t
m
o
r
e
t
h
a
n
t
h
i
s
e
f
f
e
ct
i
v
e
n
ess
a
n
d
a
c
c
u
r
a
c
y
o
f
t
h
e
c
l
ass
i
f
i
e
r
m
a
t
t
e
r
an
i
n
t
e
r
n
et
o
f
t
h
i
n
g
s
(
I
o
T
)
.
3
.
1
.
Seg
m
ent
a
t
i
o
n a
nd
m
o
rp
ho
lo
g
ica
l o
pera
t
io
n
MRI
im
ag
e,
as
a
two
-
d
im
en
s
io
n
al
s
ig
n
al
ca
n
b
e
p
r
o
ce
s
s
ed
b
y
p
o
p
u
lar
B
er
k
eley
wav
elet
tr
an
s
f
o
r
m
(
B
W
T
)
,
also
d
escr
ib
ed
as
2
D
tr
iad
ic
wav
elet
tr
an
s
f
o
r
m
.
I
n
B
W
T
tr
an
s
f
o
r
m
atio
n
,
th
e
m
a
r
k
an
d
s
elec
tio
n
o
f
th
e
th
r
esh
o
ld
r
eq
u
ir
e
d
b
y
th
e
s
ee
d
p
o
in
t
is
ea
s
ily
lo
ca
ted
.
B
W
T
s
m
o
o
th
en
s
th
e
b
o
u
n
d
ar
y
ed
g
es
an
d
p
r
eser
v
es
th
e
in
f
o
r
m
atio
n
with
o
u
t
h
am
p
er
in
g
th
e
o
u
ter
lay
er
.
I
t
i
s
ea
s
y
o
u
tf
itti
n
g
ch
ar
ac
ter
is
tics
m
ak
e
B
W
T
th
e
m
o
s
t
ad
o
p
tab
le
wav
elet
tr
a
n
s
f
o
r
m
a
tio
n
f
o
r
s
eg
m
e
n
tatio
n
o
p
er
ati
o
n
.
T
h
e
s
eg
m
en
tatio
n
p
r
o
ce
s
s
is
p
er
f
o
r
m
e
d
lay
er
wis
e;
p
r
ec
is
e
in
f
o
r
m
atio
n
f
r
o
m
th
e
in
f
ec
ted
tu
m
o
r
im
ag
e
m
u
s
t
b
e
co
n
n
ec
ted
a
n
d
r
ec
o
g
n
ized
c
o
r
r
ec
tly
.
T
h
e
B
W
T
co
n
d
u
cts
its
o
p
er
atio
n
lay
er
-
wis
e,
b
ec
o
m
in
g
o
n
e
o
f
th
e
b
est
m
eth
o
d
o
lo
g
i
es
f
o
r
p
er
f
o
r
m
in
g
s
eg
m
en
tatio
n
.
T
h
e
im
a
g
e
tr
an
s
f
o
r
m
atio
n
is
ea
s
ily
r
ep
r
esen
t
ed
in
B
W
T
an
d
is
f
u
lly
o
r
t
h
o
n
o
r
m
al.
T
h
is
f
ea
tu
r
e
en
co
u
r
a
g
es e
asy
s
eg
m
en
tatio
n
o
f
MR im
ag
es in
v
o
lv
i
n
g
co
m
p
lex
ity
.
T
h
e
b
asis
o
f
th
e
tr
an
s
f
o
r
m
atio
n
o
p
er
atio
n
lies
with
in
its
m
o
th
er
wav
el
et
tr
an
s
f
o
r
m
atio
n
.
As
s
h
o
wn
in
(
1
)
,
s
u
b
s
titu
te
wav
elets
ar
e
p
r
o
d
u
ce
d
a
t
v
ar
io
u
s
p
ix
el
p
o
s
itio
n
s
in
a
tw
o
-
d
im
en
s
io
n
al
p
lan
e,
b
y
s
ca
lin
g
an
d
tr
a
n
s
latin
g
th
e
m
o
th
er
w
av
elet.
(
,
)
=
1
2
(
3
(
−
)
,
3
(
−
)
)
(
1
)
W
h
er
e
an
d
ar
e
u
s
ed
as
s
ca
lin
g
an
d
tr
an
s
latio
n
p
a
r
am
eter
s
o
f
th
e
tr
an
s
f
o
r
m
an
d
is
th
e
tr
an
s
f
o
r
m
in
g
f
u
n
ctio
n
.
T
h
e
d
etailed
al
g
o
r
it
h
m
f
o
r
th
e
b
r
ai
n
tu
m
o
r
s
eg
m
en
tatio
n
is
im
p
lem
en
ted
u
s
in
g
B
W
T
an
d
it
is
d
escr
ib
ed
in
alg
o
r
ith
m
1
.
T
h
e
b
o
u
n
d
ar
y
ex
tr
ac
tio
n
f
r
o
m
th
e
b
r
ain
im
ag
e
ar
ea
is
d
o
n
e
u
s
in
g
m
o
r
p
h
o
lo
g
ical
o
p
er
atio
n
.
T
h
e
th
r
esh
o
ld
is
u
s
ed
to
d
ec
id
e
th
e
b
o
u
n
d
ar
y
b
etwe
en
th
e
p
ix
el
v
alu
es;
First,
th
e
im
ag
e
is
b
in
ar
ized
,
i.e
.
,
c
o
n
v
e
r
ted
i
n
to
0
’
s
an
d
1
’
s
b
y
s
elec
tin
g
a
t
h
r
e
s
h
o
ld
v
alu
e
.
An
y
th
in
g
ab
o
v
e
t
h
e
th
r
esh
o
ld
v
alu
e
will
b
e
co
n
v
er
ted
in
to
wh
ite
p
ix
el
o
r
1
an
d
b
elo
w
it
will
b
e
co
n
v
er
ted
in
t
o
0
f
r
o
m
th
e
in
p
u
t
MRI
g
r
ay
im
ag
e
h
av
in
g
0
-
2
5
5
lev
els.
T
h
u
s
,
s
ep
ar
ate
r
eg
io
n
s
ar
e
f
o
r
m
e
d
in
th
e
im
ag
e,
s
ep
ar
atin
g
th
e
in
f
ec
ted
tis
s
u
e
in
th
e
im
ag
e.
T
h
e
in
f
ec
ted
tis
s
u
e
is
ex
tr
ac
ted
f
r
o
m
th
e
im
ag
e
an
d
th
e
im
ag
e
is
er
o
d
ed
to
r
e
m
o
v
e
u
n
n
ec
ess
ar
y
p
ix
els
f
r
o
m
th
e
tis
s
u
e.
Alg
o
r
ith
m
1
.
Seg
m
en
tatio
n
u
s
in
g
B
W
T
1
.
Img = Input image;
2
.
Get size of the image (size)
3
.
Find size1 = size/3
4
.
Img1 = Convert Img to square image
5
.
Img2 = double(Img);
6
.
Calculate mean of Img2
7
.
Calculate miniature component of Img2
8
.
Select Img3 = Img2(1 : size,1 : size) as image with coordinate of miniature component
9
.
Find Img4 =
3
(
l
o
g
(
)
l
o
g
(
3
)
)
10
.
Apply resizing as Img5 = imresize(Img,Img4/size);
11
.
Img6 = Apply decomposition on image Img5
12
.
Pick decomposition parameters (CON) i = 1 : size(CON,1)
13
.
Create coefficients y using the equation y = floor((i − 1)/3);
14
.
Create coefficients x using the equation x = mod(i − 1,3);
15
.
Create coefficients bw using the equation bw = makebw(CON(i,1),CON(i,2));
16
.
Decompose coefficients using the equation
(
∗
1
+
1
:
∗
1
+
1
,
∗
1
+
1
:
∗
1
+
1
)
=
(
6
,
)
;
17
.
Performs BWT decomposition
3
.
2
.
Cla
s
s
if
ica
t
io
n
C
las
s
if
icatio
n
is
ex
ec
u
ted
to
ex
tr
ac
t
v
ital
in
f
o
r
m
atio
n
an
d
f
in
d
i
n
g
s
f
r
o
m
m
ed
ical
im
ag
es.
T
h
e
class
if
icatio
n
ac
h
iev
es
h
ig
h
er
ac
cu
r
ac
y
an
d
g
iv
es
v
al
u
ed
in
f
o
r
m
atio
n
a
b
o
u
t
th
e
af
f
ec
ted
a
r
ea
b
y
th
e
d
is
ea
s
es
[
4
4
]
.
T
h
e
class
if
icatio
n
co
m
p
lex
ity
r
ed
u
ctio
n
an
d
im
p
r
o
v
e
m
en
t
in
ac
cu
r
ac
y
ar
e
n
o
ticed
with
th
e
h
elp
o
f
p
r
o
p
er
ac
q
u
is
itio
n
,
en
h
a
n
ce
m
en
t,
f
ea
tu
r
e
ex
tr
ac
tio
n
,
an
d
f
e
atu
r
e
o
p
tim
izatio
n
o
f
th
e
im
a
g
e.
T
h
e
s
u
g
g
ested
im
ag
e
class
if
icatio
n
p
r
o
ce
s
s
is
s
h
o
wn
in
Fig
u
r
e
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
B
erkeley
w
a
ve
let
tr
a
n
s
fo
r
m
a
n
d
I
mp
r
o
ve
d
YOLOv7
-
b
a
s
ed
…
(
N
iles
h
B
h
a
s
ka
r
r
a
o
B
a
h
a
d
u
r
e
)
963
Fig
u
r
e
3
.
Pro
ce
s
s
o
f
im
ag
e
cla
s
s
if
icatio
n
Po
p
u
lar
class
if
icatio
n
tech
n
iq
u
es
s
u
ch
as
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
,
r
an
d
o
m
f
o
r
est
(
R
F),
s
elf
-
o
r
g
an
izin
g
m
ap
(
SOM)
,
an
d
p
r
in
cip
al
c
o
m
p
o
n
en
t
a
n
aly
s
is
(
PC
A)
clas
s
if
ier
s
ar
e
u
n
a
b
le
to
s
u
p
p
o
r
t
l
o
w
-
r
eso
lu
tio
n
im
ag
es.
E
ar
lier
wo
r
k
h
ad
s
h
o
wn
th
at
th
ese
clas
s
if
ier
s
ar
e
co
m
p
u
tatio
n
ally
co
m
p
lex
an
d
r
eq
u
ir
e
lar
g
e
am
o
u
n
t
o
f
tim
e
f
o
r
c
o
n
v
er
g
en
ce
wh
e
n
wo
r
k
i
n
g
o
n
l
ar
g
er
d
atasets
.
T
h
ese
lim
itati
o
n
s
ar
e
r
eso
lv
ed
b
y
u
s
in
g
th
e
YOL
Ov
7
-
b
ased
class
if
icatio
n
m
eth
o
d
[
4
5
]
,
wh
ich
is
p
r
o
p
o
s
ed
in
th
is
p
ap
er
.
Ma
ch
in
e
lear
n
in
g
an
d
d
ee
p
lear
n
in
g
tech
n
iq
u
es
h
av
e
g
ain
ed
s
ig
n
if
ican
t
p
o
p
u
la
r
ity
f
o
r
tu
m
o
r
class
if
icatio
n
.
T
h
e
m
ajo
r
ity
o
f
th
ese
m
eth
o
d
s
in
v
o
lv
e
th
e
f
ir
s
t
s
tep
o
f
lear
n
in
g
f
r
o
m
tr
ain
in
g
m
o
d
els
d
ev
elo
p
e
d
f
r
o
m
an
n
o
tated
im
ag
es
o
f
lar
g
e
d
ataset,
wh
er
e
t
h
ey
lear
n
ab
o
u
t
f
ea
tu
r
es
an
d
p
atter
n
s
o
f
in
f
ec
ted
tis
s
u
es.
C
NN
-
b
ased
ar
ch
itect
u
r
es
s
u
ch
as
YOL
O
an
d
s
in
g
le
s
h
o
t
d
etec
to
r
(
SSD
)
h
av
e
s
h
o
wn
p
r
o
m
is
in
g
r
esu
lts
in
b
r
ain
tu
m
o
r
d
etec
tio
n
.
C
o
m
p
u
ter
v
is
io
n
a
p
p
licatio
n
s
u
s
e
th
e
well
-
k
n
o
wn
o
b
ject
d
etec
tio
n
alg
o
r
ith
m
YOL
O.
I
t
is
r
en
o
wn
ed
f
o
r
its
r
ea
l
-
tim
e
p
er
f
o
r
m
a
n
ce
a
n
d
s
p
ee
d
.
YOL
O
b
r
ea
k
s
u
p
a
n
in
p
u
t
im
ag
e
in
t
o
a
g
r
id
o
f
ce
lls
,
u
s
in
g
wh
ich
m
u
ltip
le
b
o
u
n
d
i
n
g
b
o
x
es
an
d
class
p
r
o
b
ab
ilit
ies
f
o
r
th
e
o
b
jects
in
ea
ch
ce
ll
ar
e
p
r
ed
icted
.
Step
s
to
d
etec
t
o
b
j
ec
ts
u
s
in
g
YOL
O:
i)
Ob
tain
a
b
lo
b
f
r
o
m
th
e
im
ag
e
s
in
ce
we
r
eq
u
ir
e
f
ix
e
d
-
s
ize
in
p
u
t.
ii)
Sto
r
e
th
e
v
ar
io
u
s
lay
er
s
ex
tr
ac
ted
u
s
in
g
YOL
O
in
a
v
ar
iab
le.
iii)
Fo
r
war
d
th
e
v
a
r
iab
le
to
th
e
YOL
O
n
etwo
r
k
an
d
th
en
r
ec
eiv
e
th
e
o
u
tp
u
t.
An
d
iv
)
Sto
r
e
th
e
o
u
tp
u
t in
th
e
lay
er
o
u
tp
u
t v
ar
ia
b
le.
T
h
e
d
ataset
is
tr
ain
ed
f
o
r
1
6
0
E
p
o
ch
s
with
th
e
in
p
u
t
im
a
g
e
s
ize
2
2
4
×2
2
4
an
d
0
.
1
as
th
e
in
itial
lear
n
in
g
r
ate
f
o
r
th
e
tr
ain
i
n
g
p
u
r
p
o
s
e.
Du
r
in
g
t
h
e
tr
ain
in
g
p
r
o
ce
s
s
,
s
tan
d
ar
d
d
ata
in
c
r
em
en
t
m
eth
o
d
s
ar
e
u
s
ed
.
T
h
en
th
e
f
in
e
-
tu
n
in
g
o
f
th
e
n
etwo
r
k
is
co
n
s
id
er
ed
u
s
in
g
a
4
4
8
×4
4
8
i
m
ag
e
s
ize
with
th
e
in
itial
lear
n
in
g
r
ate
ch
an
g
ed
t
o
0
.
0
0
1
f
o
r
3
0
e
p
o
ch
s
,
an
d
th
e
tr
ain
in
g
is
p
er
f
o
r
m
ed
ten
tim
es.
T
h
e
d
etec
tio
n
an
d
id
en
tific
atio
n
o
f
ten
r
eq
u
ir
e
f
in
e
-
g
r
ain
ed
v
is
u
al
in
f
o
r
m
atio
n
;
f
o
r
th
is
p
u
r
p
o
s
e,
th
e
n
etwo
r
k
’
s
in
p
u
t
r
eso
lu
tio
n
h
as
b
ee
n
in
cr
ea
s
ed
f
r
o
m
2
2
4
×2
2
4
to
4
4
8
×4
4
8
.
Ou
r
f
in
al
lay
e
r
ef
f
ec
tiv
ely
f
o
r
ec
asts
b
o
th
class
p
r
o
b
ab
ilit
ies
an
d
b
o
u
n
d
in
g
b
o
x
co
o
r
d
in
ates.
A
lin
ea
r
ac
tiv
atio
n
f
u
n
ctio
n
is
em
p
lo
y
ed
f
o
r
th
e
f
in
al
lay
e
r
,
an
d
leak
y
r
ec
tifie
d
lin
ea
r
ac
tiv
atio
n
s
h
o
wn
in
(
2
)
is
u
s
ed
f
o
r
all
o
th
e
r
lay
er
s
.
∅
(
)
=
{
,
>
0
0
.
1
,
ℎ
(
2
)
T
h
e
YOL
O
alg
o
r
it
h
m
s
ar
e
s
tr
o
n
g
e
n
o
u
g
h
t
o
h
a
n
d
le
m
u
lti
-
class
clas
s
if
icatio
n
.
I
m
ag
e
o
r
o
b
ject
d
etec
tio
n
co
n
s
is
ts
o
f
two
task
s
:
i)
im
ag
e
class
if
icatio
n
an
d
ii)
o
b
ject
lo
ca
lizatio
n
.
T
h
r
o
u
g
h
t
h
e
i
m
a
g
e
cl
a
s
s
i
f
i
c
a
t
io
n
a
l
g
o
r
i
t
h
m
s
,
t
h
e
t
y
p
e
o
r
cl
as
s
o
f
a
n
o
b
j
e
ct
i
s
p
r
e
d
i
c
t
e
d
.
I
n
c
o
n
t
r
a
s
t,
o
b
j
e
c
t
l
o
c
a
li
z
a
ti
o
n
a
l
g
o
r
i
t
h
m
s
f
i
n
d
t
h
e
o
b
j
e
c
t
i
n
t
h
e
i
m
a
g
e
a
n
d
r
e
p
r
e
s
e
n
t
it
w
it
h
a
b
o
u
n
d
i
n
g
b
o
x
.
F
i
g
u
r
e
4
s
h
o
w
s
t
h
e
c
l
as
s
i
f
i
c
a
ti
o
n
,
l
o
c
a
l
iz
a
t
i
o
n
,
a
n
d
d
e
t
e
c
t
i
o
n
o
p
e
r
a
t
i
o
n
o
f
t
h
e
o
b
j
e
c
t
o
r
c
l
as
s
f
r
o
m
a
s
a
m
p
l
e
in
p
u
t
i
m
a
g
e
.
YOL
O
u
s
es
o
n
e
o
f
th
e
b
est
ar
ch
itectu
r
es
o
f
n
eu
r
al
n
etwo
r
k
s
.
Du
e
to
its
s
im
p
licity
,
h
ig
h
ac
cu
r
ac
y
,
an
d
h
ig
h
p
r
o
ce
s
s
in
g
s
p
ee
d
,
Y
OL
O
h
as
b
ec
o
m
e
a
h
ig
h
ly
p
r
ef
er
r
ed
o
b
ject
d
etec
tio
n
m
o
d
e
l.
I
t
p
r
ed
icts
a
clas
s
an
d
th
e
b
o
u
n
d
i
n
g
b
o
x
th
at
d
e
f
in
es
th
e
o
b
ject’
s
lo
ca
tio
n
o
n
th
e
in
p
u
t
im
ag
e.
E
ac
h
b
o
u
n
d
i
n
g
b
o
x
r
ec
o
g
n
izes
f
o
u
r
m
em
b
er
s
:
−
(
,
)
as th
e
ce
n
ter
o
f
th
e
b
o
u
n
d
in
g
b
o
x
−
(
)
as th
e
wid
th
o
f
t
h
e
b
o
x
−
(
ℎ
)
as th
e
b
o
x
h
eig
h
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
1
,
Feb
r
u
ar
y
20
25
:
9
5
8
-
969
964
Fig
u
r
e
4
.
C
lass
if
icatio
n
,
lo
ca
lizatio
n
,
an
d
d
etec
tio
n
p
r
o
ce
s
s
o
f
th
e
o
b
ject
I
n
a
d
d
itio
n
to
t
h
is
,
it
p
r
ed
icts
t
h
e
co
r
r
esp
o
n
d
in
g
n
u
m
b
er
c
f
o
r
th
e
p
r
e
d
icted
class
an
d
th
e
p
r
o
b
ab
ilit
y
o
f
th
e
p
r
ed
ictio
n
(
)
.
T
h
e
en
tire
im
ag
e
is
d
iv
id
ed
in
to
a
g
r
i
d
,
f
o
r
ex
am
p
le,
a
3
×
3
g
r
id
.
T
h
r
o
u
g
h
th
e
g
r
id
,
it
b
ec
o
m
es
ea
s
y
t
o
d
etec
t
o
n
e
o
b
ject
p
er
g
r
i
d
ce
ll
c
o
m
p
ar
e
d
t
o
o
n
e
o
b
ject
p
e
r
im
a
g
e.
I
n
t
h
e
n
ex
t
s
tep
,
ea
ch
g
r
id
ce
ll
is
d
escr
ib
ed
b
y
a
v
ec
to
r
.
Fo
r
ex
am
p
le,
in
th
e
ca
s
e
o
f
b
r
ain
MRI
im
ag
e,
two
class
es
ar
e
d
ef
in
e
d
s
u
ch
as
No
r
m
al
an
d
A
b
n
o
r
m
al,
th
en
it is
d
escr
ib
ed
as:
,
=
(
,
,
,
,
ℎ
,
1
,
2
)
w
h
er
e
,
is
r
ep
r
esen
ted
th
e
co
r
r
esp
o
n
d
in
g
g
r
id
ce
ll,
f
o
r
ex
am
p
le,
th
e
f
ir
s
t
ce
ll
f
r
o
m
th
e
3
×3
g
r
id
is
r
ep
r
esen
ted
as
1
,
1
.
is
th
e
p
r
o
b
ab
ilit
y
o
f
th
e
o
b
ject
class
,
an
d
ar
e
th
e
co
o
r
d
in
ates
o
f
th
e
ce
n
t
er
o
f
th
e
b
o
u
n
d
in
g
b
o
x
,
ℎ
,
an
d
ar
e
th
e
h
eig
h
t
an
d
wid
th
o
f
th
e
b
o
u
n
d
i
n
g
b
o
x
r
elativ
e
to
th
e
e
n
tire
i
m
ag
e,
an
d
1
an
d
2
ar
e
r
ep
r
esen
ted
f
o
r
th
e
class
,
i.e
.
1
f
o
r
th
e
“No
r
m
al”
an
d
2
f
o
r
th
e
“Ab
n
o
r
m
al”.
T
h
e
v
al
u
e
o
f
1
an
d
2
is
0
a
n
d
1
,
d
ep
e
n
d
in
g
o
n
w
h
ich
class
r
ep
r
esen
ts
t
h
e
b
o
u
n
d
in
g
b
o
x
.
Alg
o
r
ith
m
2
,
en
li
s
ts
v
ar
io
u
s
s
tep
s
in
v
o
lv
ed
in
th
e
im
p
lem
en
tatio
n
o
f
YOL
Ov
7
,
f
o
r
th
e
d
etec
tio
n
an
d
class
if
icatio
n
o
f
b
r
ain
t
u
m
o
r
s
.
Alg
o
r
ith
m
2
.
C
lass
if
icatio
n
u
s
in
g
YOL
O
1
.
I
m
p
o
r
t th
e
r
eq
u
ir
e
d
p
ac
k
ag
es a
n
d
lib
r
ar
ies
2
.
Select
th
r
esh
o
ld
v
alu
e
(
0
.
5
)
,
b
o
x
c
o
n
f
id
e
n
ce
s
co
r
e,
a
n
d
b
o
x
class
p
r
o
b
ab
ilit
y
3
.
C
alcu
late
s
co
r
e,
b
o
x
es,
an
d
class
es
4
.
C
alcu
late
I
o
U
b
etwe
en
two
b
o
x
es
=
5
.
Select
n
o
n
-
m
ax
s
u
p
p
r
ess
io
n
6
.
Select
th
e
v
alu
e
o
f
s
h
ap
e
(
1
9
,
1
9
,
5
,
7
)
r
a
n
d
o
m
l
y
an
d
th
en
p
r
ed
ict
th
e
b
o
u
n
d
in
g
b
o
x
es
=
[
ℎ
1
2
]
7
.
Gen
er
ate
s
u
p
p
r
ess
ed
b
o
x
es f
r
o
m
th
e
o
u
tp
u
t o
f
C
NN
8
.
Fin
d
th
e
p
r
ed
ictio
n
f
o
r
a
r
an
d
o
m
v
o
lu
m
e
9
.
Ap
p
ly
p
r
e
-
tr
ai
n
ed
YOL
O
al
g
o
r
ith
m
o
n
n
ew
im
a
g
es
10
.
Gen
er
ate
t
h
e
p
r
e
d
ictio
n
o
f
b
o
u
n
d
in
g
b
o
x
es a
n
d
s
av
e
th
e
i
m
ag
es (
I
m
1
)
11
.
Get
an
im
a
g
e
an
d
m
ak
e
p
r
ed
ictio
n
s
u
s
in
g
th
e
p
r
ed
ict
f
u
n
ctio
n
12
.
Plo
t th
e
p
r
ed
ictio
n
s
3
.
2
.
1
.
I
nte
rsect
io
n
o
v
er
un
io
n
I
n
ter
s
ec
tio
n
o
v
er
u
n
io
n
(
I
o
U)
is
a
p
er
f
o
r
m
an
ce
p
ar
am
eter
t
o
ev
alu
ate
h
o
w
ef
f
ec
tiv
ely
th
e
o
b
ject
is
d
etec
ted
.
I
t
is
th
e
r
atio
b
etwe
e
n
th
e
g
r
o
u
n
d
tr
u
th
an
d
th
e
p
r
e
d
icted
b
o
u
n
d
in
g
b
o
x
i
n
Fig
u
r
e
5
.
An
I
OU
v
al
u
e
g
r
ea
ter
th
an
0
.
5
,
ca
lled
th
r
esh
o
ld
v
alu
e,
is
r
ec
o
m
m
e
n
d
ed
.
I
o
U
v
alu
e
less
th
an
th
e
th
r
esh
o
ld
v
alu
e
in
d
icate
s
f
alse
d
etec
tio
n
.
T
h
e
l
o
wer
th
e
v
alu
e,
th
e
h
ig
h
er
th
e
f
alse
d
etec
tio
n
r
ate.
Fo
r
a
p
o
s
itiv
e
p
r
ed
ictio
n
,
t
h
e
I
o
U
v
alu
e
s
h
o
u
ld
b
e
>
0
.
5
,
a
n
d
f
o
r
n
eg
ativ
e
p
r
e
d
ictio
n
s
,
an
I
o
U
v
alu
e
s
h
o
u
ld
b
e
<
0
.
5
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
B
erkeley
w
a
ve
let
tr
a
n
s
fo
r
m
a
n
d
I
mp
r
o
ve
d
YOLOv7
-
b
a
s
ed
…
(
N
iles
h
B
h
a
s
ka
r
r
a
o
B
a
h
a
d
u
r
e
)
965
Fig
u
r
e
5
.
C
alcu
latio
n
o
f
I
o
U
4.
P
RO
P
O
SE
D
M
E
T
H
O
D
I
n
th
is
p
a
p
er
,
two
d
atasets
,
n
am
ely
B
R
AI
NI
X
(
DI
C
OM
)
[
4
6
]
a
n
d
Kag
g
le
[
4
7
]
,
ar
e
ev
al
u
ated
.
F
o
r
v
alid
atin
g
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
,
1
6
5
n
o
r
m
al
MR
b
r
ain
im
ag
es
an
d
2
8
9
m
en
in
g
io
m
a
MR
im
ag
es
ar
e
u
s
ed
.
T
h
e
r
atio
o
f
tr
ain
in
g
an
d
test
in
g
will
b
e
m
ain
tain
ed
at
8
0
:2
0
f
o
r
th
e
ev
alu
atio
n
o
f
th
e
p
r
o
p
o
s
ed
s
y
s
tem
.
5.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
MA
T
L
AB
R
2
0
2
0
an
d
Py
th
o
n
3
.
7
s
o
f
twar
e
ar
e
u
s
ed
f
o
r
s
im
u
latio
n
p
u
r
p
o
s
es.
T
h
e
p
r
o
p
o
s
ed
s
y
s
tem
i
s
ev
alu
ated
o
n
two
d
i
f
f
er
en
t
d
atasets
[
4
6
]
,
[
4
7
]
.
T
h
e
d
ataset
co
m
p
r
is
es
n
o
n
-
ca
n
ce
r
o
u
s
(
n
o
n
-
m
en
i
n
g
io
m
a
o
r
n
o
r
m
al)
an
d
ca
n
ce
r
o
u
s
(
m
en
i
n
g
io
m
a)
b
r
ain
MR
im
ag
es.
Fro
m
B
AI
NI
X,
1
3
4
m
en
in
g
io
m
a
b
r
ain
im
ag
es
an
d
6
7
n
o
n
-
ca
n
ce
r
o
u
s
b
r
ain
im
a
g
es
ar
e
u
s
ed
an
d
f
r
o
m
Kag
g
l
e,
1
5
5
m
en
in
g
io
m
a
b
r
ain
im
ag
es
an
d
9
8
n
o
n
-
ca
n
ce
r
o
u
s
b
r
ain
im
ag
es
a
r
e
u
s
ed
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
tem
co
r
r
ec
tly
d
etec
ted
2
8
6
o
u
t
o
f
2
8
9
ca
n
ce
r
o
u
s
im
a
g
es
an
d
th
u
s
g
a
v
e
a
9
9
%
class
if
icatio
n
r
ate.
T
h
e
p
r
o
p
o
s
ed
s
y
s
tem
co
r
r
ec
tly
d
etec
ted
1
6
3
n
o
n
-
ca
n
ce
r
o
u
s
im
ag
es
o
u
t
o
f
1
6
5
im
ag
es
an
d
th
u
s
o
f
f
er
s
a
9
8
.
7
8
%
class
if
icatio
n
r
ate.
Hen
ce
th
e
p
r
o
p
o
s
ed
s
y
s
tem
ac
h
iev
ed
9
8
.
8
9
%
o
f
th
e
av
e
r
ag
e
class
if
icatio
n
r
a
te.
T
h
e
d
ice
co
e
f
f
icien
t
is
an
im
p
o
r
tan
t
p
er
f
o
r
m
a
n
ce
p
a
r
am
eter
wh
o
s
e
d
ef
au
lt
v
al
u
e
is
b
etwe
en
0
an
d
1
.
I
ts
v
alu
e
is
d
ir
ec
tly
ca
lcu
lated
u
s
in
g
th
e
f
o
r
m
u
las
an
d
ca
n
also
b
e
ca
lcu
lated
u
s
in
g
th
e
J
ac
ca
r
d
co
ef
f
icien
t
in
d
ex
(
J
C
I
)
.
T
h
e
d
e
v
iatio
n
b
e
twee
n
th
e
s
o
u
r
ce
im
ag
e
a
n
d
t
h
e
s
eg
m
en
ted
o
u
tp
u
t
im
ag
e
is
m
ea
s
u
r
ed
t
h
r
o
u
g
h
MSE
.
I
n
o
u
r
ca
s
e,
MSE
i
s
0
.
0
0
9
.
T
h
e
q
u
ality
o
f
th
e
s
eg
m
en
ted
im
ag
e
is
m
ea
s
u
r
ed
b
ased
o
n
im
ag
e
in
ten
s
ity
.
T
h
e
ca
lcu
latio
n
o
f
im
ag
e
in
te
n
s
ity
b
etwe
en
e
x
p
er
t
s
eg
m
e
n
t
atio
n
,
i.e
.
,
m
an
u
al
s
eg
m
en
tati
o
n
an
d
s
eg
m
e
n
ted
o
u
tp
u
t
u
s
in
g
a
m
ath
em
atica
l
n
o
tio
n
is
k
n
o
wn
as
PS
NR
.
T
h
e
r
ec
o
m
m
e
n
d
ed
v
alu
e
o
f
PS
NR
f
o
r
b
etter
q
u
ality
in
ter
m
s
o
f
n
o
is
e
is
4
0
d
ec
ib
els
(
d
B
)
.
Ou
r
p
r
o
p
o
s
ed
m
eth
o
d
o
b
tain
e
d
5
4
.
3
d
B
PS
NR
.
T
h
e
co
m
p
ar
ativ
e
an
aly
s
is
o
f
MSE
an
d
PS
NR
is
s
h
o
wn
in
Fig
u
r
e
5
.
T
h
e
co
m
p
ar
ativ
e
an
aly
s
is
o
f
MSE
an
d
P
SNR
,
as
s
h
o
wn
in
Fig
u
r
e
6
,
d
ep
icted
th
at
o
u
r
p
r
o
p
o
s
ed
s
y
s
tem
s
eg
m
en
ts
th
e
M
R
im
ag
es w
ith
les
s
n
o
is
e.
W
e
also
an
aly
ze
d
MSE
an
d
PS
NR
w
ith
o
u
t
a
n
o
is
e
e
lim
in
atio
n
s
tep
(
i.e
.
,
p
r
e
-
p
r
o
c
ess
in
g
)
,
an
d
th
e
r
esu
lts
wer
e
n
o
t
en
co
u
r
a
g
in
g
.
Fig
u
r
e
7
s
h
o
ws
th
e
co
m
p
ar
is
o
n
s
o
f
MSE
a
n
d
PS
NR
with
o
u
t
p
r
e
-
p
r
o
ce
s
s
in
g
.
T
h
e
ef
f
ec
t
o
f
th
e
p
r
e
-
p
r
o
ce
s
s
in
g
an
d
en
h
an
ce
m
en
t step
is
r
ef
lec
ted
in
th
e
o
v
er
all
p
er
f
o
r
m
a
n
ce
,
d
ep
icted
in
Fig
u
r
es 6
a
n
d
7.
Fig
u
r
e
6
.
C
o
m
p
a
r
is
o
n
o
f
MSE
an
d
PS
NR
(
with
p
r
e
-
p
r
o
ce
s
s
in
g
)
PSN
R
Me
th
o
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
1
,
Feb
r
u
ar
y
20
25
:
9
5
8
-
969
966
Fig
u
r
e
7
.
C
o
m
p
a
r
is
o
n
o
f
MSE
an
d
PS
NR
(
with
o
u
t p
r
e
-
p
r
o
ce
s
s
in
g
)
T
h
e
to
tal
n
u
m
b
er
o
f
c
o
r
r
ec
t
n
eg
ativ
e
p
r
ed
ictio
n
s
o
v
er
th
e
to
tal
n
eg
ativ
e
ca
s
es
is
k
n
o
wn
as
th
e
s
p
ec
if
icity
[
4
6
]
.
T
h
e
t
o
tal
n
u
m
b
er
o
f
co
r
r
ec
t
p
o
s
itiv
e
p
r
ed
ictio
n
s
o
v
er
th
e
to
tal
n
u
m
b
e
r
o
f
p
o
s
itiv
e
ca
s
es
is
k
n
o
wn
as
s
en
s
itiv
ity
(
Sen
s
)
[
4
8
]
.
I
n
o
th
er
wo
r
d
s
,
th
e
n
u
m
b
er
o
f
ca
s
es
d
etec
ted
co
r
r
e
ctly
as
p
o
s
itiv
e
is
m
ea
s
u
r
ed
as
s
en
s
itiv
ity
f
r
o
m
th
e
to
tal
p
o
s
itiv
e
ca
s
es.
Fo
r
ex
am
p
le,
s
u
p
p
o
s
e
s
en
s
itiv
ity
is
9
0
%
o
u
t
o
f
1
0
0
p
o
s
itiv
e
ca
s
es.
T
ab
le
1
e
n
lis
ts
v
ar
io
u
s
f
o
r
m
u
lae
wh
ich
ar
e
u
s
ed
f
o
r
ca
lcu
latio
n
s
o
f
p
ar
a
m
eter
s
o
f
in
ter
est.
T
ab
le
2
s
h
o
ws
a
co
m
p
ar
is
o
n
o
f
th
e
test
p
er
f
o
r
m
an
ce
o
f
p
r
o
p
o
s
ed
YOL
Ov
7
b
ased
class
if
ier
with
d
i
f
f
er
en
t
class
if
ier
s
s
u
ch
as
an
ad
a
p
tiv
e
n
eu
r
o
-
f
u
zz
y
in
f
er
e
n
ce
s
y
s
tem
(
ANFI
S),
g
e
n
etic
alg
o
r
ith
m
(
GA)
,
an
d
K
-
n
ea
r
est
n
eig
h
b
o
r
s
(
KNN)
.
T
ab
le
3
s
h
o
ws
th
e
co
m
p
ar
is
o
n
o
f
th
e
p
r
o
p
o
s
ed
m
eth
o
d
with
o
th
er
p
r
o
v
en
m
et
h
o
d
o
lo
g
ies.
E
v
en
th
o
u
g
h
th
e
r
esu
lts
o
f
s
o
m
e
o
f
th
e
p
r
o
v
en
r
esear
ch
an
d
p
r
o
p
o
s
ed
m
eth
o
d
lo
o
k
v
er
y
clo
s
e,
th
e
ap
p
r
o
ac
h
o
f
YOL
O
is
co
m
p
letely
d
if
f
er
en
t,
an
d
it
p
r
o
ce
s
s
es lar
g
e
d
ata
s
ets ef
f
icien
tly
.
T
ab
le
1
.
Per
f
o
r
m
an
ce
p
ar
am
et
er
m
atr
ices
Q
u
a
l
i
t
y
p
a
r
a
me
t
e
r
(
a
l
l
v
a
l
u
e
i
n
%)
F
o
r
mu
l
a
A
c
c
u
r
a
c
y
(
A
c
c
)
(
)
+
(
)
+
×
100
S
e
n
s
i
t
i
v
i
t
y
(
S
e
n
s)
(
)
(
+
)
×
100
S
p
e
c
i
f
i
c
i
t
y
(
S
p
e
c
)
(
)
(
+
)
×
100
P
r
e
c
i
s
i
o
n
o
r
p
o
s
i
t
i
v
e
p
r
e
d
i
c
t
i
v
e
v
a
l
u
e
(PPV)
+
×
100
R
e
c
a
l
l
o
r
n
e
g
a
t
i
v
e
p
r
e
d
i
c
t
i
v
e
v
a
l
u
e
(
N
P
V
)
+
×
100
R
e
l
e
v
a
n
c
e
f
a
c
t
o
r
×
100
F
a
l
se
n
e
g
a
t
i
v
e
r
a
t
e
(
F
N
R
)
+
×
100
F
a
l
se
p
o
si
t
i
v
e
r
a
t
e
(
F
P
R
)
+
×
100
D
i
c
e
si
m
i
l
a
r
i
t
y
c
o
e
f
f
i
c
i
e
n
t
(
D
S
C
)
2
2
+
+
×
100
T
ab
le
2
.
C
o
m
p
a
r
is
o
n
o
f
ac
cu
r
ac
ies in
d
if
f
er
en
t c
lass
if
ier
s
N
u
mb
e
r
o
f
t
e
st
i
ma
g
e
s (n
o
r
ma
l
=
1
6
5
,
a
b
n
o
r
m
a
l
=
2
8
9
)
Ev
a
l
u
a
t
i
o
n
p
a
r
a
me
t
e
r
A
N
F
I
S
GA
K
-
NN
Y
O
LO
(
P
r
o
p
o
se
d
)
Tr
u
e
n
e
g
a
t
i
v
e
72
74
69
77
F
a
l
se
p
o
si
t
i
v
e
9
5
12
3
Tr
u
e
p
o
si
t
i
v
e
3
4
0
3
5
0
3
3
8
3
7
2
F
a
l
se
n
e
g
a
t
i
v
e
22
25
35
2
S
p
e
c
i
f
i
c
i
t
y
(
%)
8
8
.
8
8
9
3
.
6
7
8
5
.
1
8
9
6
.
2
5
S
e
n
s
i
t
i
v
i
t
y
(
%)
9
1
.
1
5
9
3
.
3
3
9
0
.
6
1
9
9
.
4
6
A
c
c
u
r
a
c
y
(
%)
9
0
.
7
4
9
3
.
3
9
8
9
.
6
4
9
8
.
8
9
P
r
e
c
i
s
i
o
n
o
r
P
P
V
(
%)
9
7
.
4
2
9
8
.
5
9
9
6
.
5
7
9
9
.
2
R
e
c
a
l
l
o
r
N
P
V
(
%)
6
8
.
5
7
7
4
.
7
4
6
6
.
3
4
9
7
.
4
6
F
a
l
se
n
e
g
a
t
i
v
e
r
a
t
e
(
F
N
R
)
(
%)
8
.
8
4
6
.
6
7
9
.
3
8
0
.
5
3
F
a
l
se
p
o
si
t
i
v
e
r
a
t
e
(
F
P
R
)
(
%)
1
1
.
1
1
6
.
3
2
1
4
.
8
1
3
.
7
5
A
v
e
r
a
g
e
d
i
c
e
c
o
e
f
f
i
c
i
e
n
t
i
n
d
e
x
(
%)
9
4
.
1
8
9
5
.
8
9
9
9
.
4
9
9
9
.
3
3
I
o
U
--
--
--
0
.
8
9
PS
N
R
M
et
h
o
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
B
erkeley
w
a
ve
let
tr
a
n
s
fo
r
m
a
n
d
I
mp
r
o
ve
d
YOLOv7
-
b
a
s
ed
…
(
N
iles
h
B
h
a
s
ka
r
r
a
o
B
a
h
a
d
u
r
e
)
967
T
ab
le
3
.
C
o
m
p
a
r
is
o
n
o
f
p
er
f
o
r
m
an
ce
p
ar
am
ete
r
s
with
p
r
o
v
e
n
m
eth
o
d
s
R
e
f
.
M
e
t
h
o
d
Y
e
a
r
A
c
c
.
(
%)
S
e
n
s.
(
%)
S
p
e
c
.
(
%)
I
o
U
[
3
2
]
D
e
e
p
c
o
n
v
o
l
u
t
i
o
n
a
l
g
e
n
e
r
a
t
i
v
e
a
d
v
e
r
s
a
r
i
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
2
0
2
3
97
NA
NA
NA
[
3
5
]
G
LC
M
+
C
N
N
2
0
2
3
9
8
.
2
NA
NA
NA
[
2
0
]
H
i
st
o
g
r
a
m e
q
u
a
l
i
z
a
t
i
o
n
+
l
e
a
r
n
i
n
g
-
b
a
sed
n
e
u
r
a
l
n
e
t
w
o
r
k
2
0
2
2
93
92
94
NA
[
2
2
]
H
y
b
r
i
d
i
ma
g
e
p
r
o
c
e
ssi
n
g
2
0
2
2
95
NA
NA
NA
[
2
3
]
Pre
-
t
r
a
i
n
e
d
G
o
o
g
l
e
N
e
t
w
i
t
h
S
V
M
2
0
2
2
9
8
.
1
NA
NA
NA
[
2
3
]
F
i
n
e
l
y
t
u
n
e
e
d
G
o
o
g
l
e
N
e
t
2
0
2
2
9
3
.
1
NA
NA
NA
[
2
5
]
A
B
C
2
0
2
2
9
3
.
6
7
NA
NA
NA
[
2
6
]
A
l
e
x
N
e
t
+
S
V
M
2
0
2
2
9
5
.
1
0
9
5
.
2
5
NA
NA
[
4
7
]
M
o
d
i
f
i
e
d
A
B
C
2
0
2
2
96
9
8
.
9
64
NA
[
4
8
]
G
LC
M
+
S
e
g
N
e
t
+
D
T
2
0
2
2
98
NA
NA
NA
[
2
9
]
Y
O
LO
v
7
+
M
c
C
u
l
l
o
c
h
2
0
2
2
9
9
.
3
2
NA
NA
NA
[
3
1
]
C
N
N
+
h
a
a
r
w
a
v
e
l
e
t
2
0
2
2
96
96
97
NA
P
r
o
p
o
se
d
B
W
T
+
Y
O
LO
v
7
--
9
8
.
8
9
9
9
.
4
6
9
6
.
2
5
0
.
8
9
6.
CO
NCLU
SI
O
N
So
m
e
o
f
th
e
e
x
is
tin
g
r
esear
ch
p
r
o
p
o
s
ed
f
o
r
class
if
y
in
g
a
n
d
d
etec
tin
g
b
r
ain
tu
m
o
r
s
ar
e
p
r
o
m
is
in
g
b
u
t
o
n
ly
p
ar
tially
ac
c
u
r
ate,
an
d
co
m
p
lete
au
to
m
atio
n
s
till
n
ee
d
s
to
b
e
in
clu
d
e
d
.
T
h
e
y
v
ar
y
f
r
o
m
im
ag
e
p
r
o
ce
s
s
in
g
an
d
s
o
f
t
c
o
m
p
u
tin
g
t
o
d
ee
p
le
ar
n
in
g
-
b
ased
m
eth
o
d
o
l
o
g
ies.
So
m
e
r
esear
ch
er
s
p
r
esen
ted
p
er
f
o
r
m
a
n
ce
m
atr
ices
s
u
ch
as
ac
cu
r
ac
y
,
s
p
ec
if
icity
,
an
d
s
en
s
itiv
ity
f
o
r
v
alid
atio
n
p
u
r
p
o
s
es
o
f
th
e
al
g
o
r
ith
m
o
r
m
eth
o
d
o
lo
g
ies.
T
h
r
o
u
g
h
ac
cu
r
ac
y
,
t
h
e
to
tal
co
r
r
ec
t
p
r
ed
ictio
n
s
o
v
e
r
th
e
to
tal
u
s
ed
o
r
av
ailab
le
d
ata
ar
e
p
r
esen
ted
.
T
h
e
s
p
ec
if
icity
m
atr
ix
ca
n
ch
ar
a
cter
ize
h
ea
lth
y
p
atien
ts
,
wh
er
ea
s
s
en
s
itiv
ity
g
iv
es
a
ch
ar
ac
ter
i
za
tio
n
o
f
u
n
h
ea
lth
y
p
atien
ts
.
T
h
ese
p
er
f
o
r
m
an
ce
m
atr
ices
p
lay
a
s
ig
n
if
ican
t
ju
s
tific
atio
n
f
o
r
th
e
o
v
er
all
ass
es
s
m
en
t
o
f
an
y
p
r
o
p
o
s
ed
alg
o
r
ith
m
.
Alm
o
s
t
ev
er
y
r
esear
ch
e
r
h
as
p
r
ese
n
ted
an
d
an
al
y
ze
d
th
eir
wo
r
k
s
th
r
o
u
g
h
s
o
m
e
p
er
f
o
r
m
an
ce
m
atr
ices,
s
o
th
e
y
ar
e
u
s
ed
to
c
o
m
p
ar
e
th
e
r
esea
r
ch
er
’
s
wo
r
k
.
Ou
r
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
u
s
in
g
YOL
Ov
7
ex
tr
ac
ts
a
n
d
class
if
ies
h
ea
lth
y
a
n
d
a
b
n
o
r
m
al
b
r
ain
tis
s
u
es
f
r
o
m
MR
im
ag
es
with
h
ig
h
class
if
icatio
n
ac
cu
r
ac
y
,
an
d
tak
es
les
s
co
m
p
u
tatio
n
al
tim
e.
T
h
e
co
m
p
lete
s
y
s
tem
is
f
u
lly
au
to
m
ated
.
T
h
is
au
to
m
atio
n
s
u
p
p
o
r
ts
p
ath
o
lo
g
is
ts
in
th
e
d
etec
tio
n
o
f
th
e
tu
m
o
r
r
eg
io
n
an
d
also
h
elp
s
th
em
to
p
r
o
v
id
e
ea
r
ly
d
ia
g
n
o
s
i
s
s
u
g
g
esti
o
n
s
with
co
m
p
lete
c
o
n
f
id
en
ce
.
RE
F
E
R
E
NC
E
S
[
1
]
A
.
V
e
r
ma
,
S
.
N
.
S
h
i
v
h
a
r
e
,
S
.
P
.
S
i
n
g
h
,
N
.
K
u
mar,
a
n
d
A
.
N
a
y
y
a
r
,
“
C
o
m
p
r
e
h
e
n
s
i
v
e
r
e
v
i
e
w
o
n
M
R
I
-
b
a
se
d
b
r
a
i
n
t
u
m
o
r
seg
m
e
n
t
a
t
i
o
n
:
A
c
o
m
p
a
r
a
t
i
v
e
st
u
d
y
f
r
o
m
2
0
1
7
o
n
w
a
r
d
s,
”
Arc
h
i
v
e
s
o
f
C
o
m
p
u
t
a
t
i
o
n
a
l
Me
t
h
o
d
s
i
n
En
g
i
n
e
e
ri
n
g
,
M
a
y
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
8
3
1
-
0
2
4
-
1
0
1
2
8
-
0.
[
2
]
W
H
O
,
“
I
n
t
e
r
n
a
t
i
o
n
a
l
a
g
e
n
c
y
f
o
r
r
e
s
e
a
r
c
h
o
n
c
a
n
c
e
r
:
c
e
n
c
e
r
t
o
m
o
r
r
o
w
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
Ag
e
n
c
y
f
o
r
R
e
se
a
rc
h
o
n
C
a
n
c
e
r
(
I
ARC
)
,
A
c
c
e
ss
e
d
:
J
u
n
.
0
1
,
2
0
2
3
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
g
c
o
.
i
a
r
c
.
f
r
/
t
o
mo
r
r
o
w
/
e
n
/
d
a
t
a
v
i
z
/
.
[
3
]
Y
.
F
a
n
e
t
a
l
.
,
“
B
u
r
d
e
n
a
n
d
t
r
e
n
d
s
o
f
b
r
a
i
n
a
n
d
c
e
n
t
r
a
l
n
e
r
v
o
u
s
s
y
st
e
m
c
a
n
c
e
r
f
r
o
m
1
9
9
0
t
o
2
0
1
9
a
t
t
h
e
g
l
o
b
a
l
,
r
e
g
i
o
n
a
l
,
a
n
d
c
o
u
n
t
r
y
l
e
v
e
l
s,
”
Arc
h
i
v
e
s
o
f
P
u
b
l
i
c
H
e
a
l
t
h
,
v
o
l
.
8
0
,
n
o
.
1
,
S
e
p
.
2
0
2
2
,
d
o
i
:
1
0
.
1
1
8
6
/
s
1
3
6
9
0
-
0
2
2
-
0
0
9
6
5
-
5.
[
4
]
F
.
B
r
a
y
,
J.
F
e
r
l
a
y
,
I
.
S
o
e
r
j
o
m
a
t
a
r
a
m
,
R
.
L
.
S
i
e
g
e
l
,
L.
A
.
T
o
r
r
e
,
a
n
d
A
.
Jema
l
,
“
G
l
o
b
a
l
c
a
n
c
e
r
st
a
t
i
st
i
c
s
2
0
1
8
:
G
LO
B
O
C
A
N
e
st
i
mat
e
s
o
f
i
n
c
i
d
e
n
c
e
a
n
d
m
o
r
t
a
l
i
t
y
w
o
r
l
d
w
i
d
e
f
o
r
3
6
c
a
n
c
e
r
s
i
n
1
8
5
c
o
u
n
t
r
i
e
s,
”
C
A:
A
C
a
n
c
e
r
J
o
u
r
n
a
l
f
o
r
C
l
i
n
i
c
i
a
n
s
,
v
o
l
.
6
8
,
n
o
.
6
,
p
p
.
3
9
4
–
4
2
4
,
N
o
v
.
2
0
1
8
,
d
o
i
:
1
0
.
3
3
2
2
/
c
a
a
c
.
2
1
4
9
2
.
[
5
]
H
.
S
u
n
g
e
t
a
l
.
,
“
G
l
o
b
a
l
c
a
n
c
e
r
s
t
a
t
i
s
t
i
c
s 2
0
2
0
:
G
LO
B
O
C
A
N
e
s
t
i
m
a
t
e
s o
f
i
n
c
i
d
e
n
c
e
a
n
d
m
o
r
t
a
l
i
t
y
w
o
r
l
d
w
i
d
e
f
o
r
3
6
c
a
n
c
e
r
s
i
n
1
8
5
c
o
u
n
t
r
i
e
s,”
C
A:
A
C
a
n
c
e
r
J
o
u
r
n
a
l
f
o
r
C
l
i
n
i
c
i
a
n
s
,
v
o
l
.
7
1
,
n
o
.
3
,
p
p
.
2
0
9
–
2
4
9
,
F
e
b
.
2
0
2
1
,
d
o
i
:
1
0
.
3
3
2
2
/
c
a
a
c
.
2
1
6
6
0
.
[
6
]
L.
K
i
r
a
n
e
t
a
l
.
,
“
A
n
e
n
h
a
n
c
e
d
p
a
t
t
e
r
n
d
e
t
e
c
t
i
o
n
a
n
d
se
g
me
n
t
a
t
i
o
n
o
f
b
r
a
i
n
t
u
mo
r
s
i
n
M
R
I
i
mag
e
s
u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
,
”
Fro
n
t
i
e
rs
i
n
C
o
m
p
u
t
a
t
i
o
n
a
l
N
e
u
r
o
sci
e
n
c
e
,
v
o
l
.
1
8
,
J
u
n
.
2
0
2
4
,
d
o
i
:
1
0
.
3
3
8
9
/
f
n
c
o
m.
2
0
2
4
.
1
4
1
8
2
8
0
.
[
7
]
R
.
S
o
l
o
h
,
H
.
A
l
a
b
b
o
u
d
,
A
.
S
h
a
h
i
n
,
A
.
Y
a
ssi
n
e
,
a
n
d
A
.
El
C
h
a
k
i
k
,
“
B
r
a
i
n
t
u
mo
r
s
e
g
m
e
n
t
a
t
i
o
n
b
a
se
d
o
n
α
-
e
x
p
a
n
si
o
n
g
r
a
p
h
c
u
t
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
m
a
g
i
n
g
S
y
s
t
e
m
s a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
3
4
,
n
o
.
4
,
J
u
n
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
2
/
i
m
a
.
2
3
1
3
2
.
[
8
]
N
.
B
.
B
a
h
a
d
u
r
e
,
A
.
K
.
R
a
y
,
a
n
d
H
.
P
.
T
h
e
t
h
i
,
“
I
mag
e
a
n
a
l
y
si
s
f
o
r
b
r
a
i
n
t
u
m
o
u
r
d
e
t
e
c
t
i
o
n
u
si
n
g
G
A
-
S
V
M
w
i
t
h
a
u
t
o
-
r
e
p
o
r
t
g
e
n
e
r
a
t
i
o
n
t
e
c
h
n
i
q
u
e
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Bi
o
m
e
d
i
c
a
l
E
n
g
i
n
e
e
ri
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
3
2
,
n
o
.
3
,
p
p
.
2
4
5
–
2
6
6
,
2
0
2
0
,
d
o
i
:
1
0
.
1
5
0
4
/
I
JB
ET.
2
0
2
0
.
1
0
6
0
3
4
.
[
9
]
V
i
mal
G
u
p
t
a
,
“
B
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
a
n
d
se
g
m
e
n
t
a
t
i
o
n
u
s
i
n
g
i
m
p
r
o
v
e
d
b
a
t
a
l
g
o
r
i
t
h
m
w
i
t
h
i
m
p
r
o
v
e
d
i
n
v
a
s
i
v
e
w
e
e
d
o
p
t
i
m
i
z
a
t
i
o
n
,
”
J
o
u
r
n
a
l
o
f
El
e
c
t
ri
c
a
l
S
y
st
e
m
s
,
v
o
l
.
2
0
,
n
o
.
7
s,
p
p
.
7
9
2
–
8
0
3
,
M
a
y
2
0
2
4
,
d
o
i
:
1
0
.
5
2
7
8
3
/
j
e
s
.
3
4
5
4
.
[
1
0
]
J.
S
h
i
,
L.
D
e
n
g
,
C
.
Li
a
o
,
a
n
d
B
.
R
e
n
,
“
Ef
f
i
c
i
e
n
t
d
i
g
i
t
a
l
c
a
mo
u
f
l
a
g
e
p
a
t
t
e
r
n
g
e
n
e
r
a
t
i
o
n
a
l
g
o
r
i
t
h
m
b
a
s
e
d
o
n
i
m
p
r
o
v
e
d
G
A
-
K
-
mea
n
s
c
l
u
st
e
r
i
n
g
a
l
g
o
r
i
t
h
m
,
”
i
n
T
h
i
r
d
I
n
t
e
r
n
a
t
i
o
n
a
l
S
y
m
p
o
si
u
m
o
n
C
o
m
p
u
t
e
r
Ap
p
l
i
c
a
t
i
o
n
s
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
y
st
e
m
s
(
I
S
C
AI
S
2
0
2
4
)
,
Ju
l
.
2
0
2
4
,
v
o
l
.
1
0
,
d
o
i
:
1
0
.
1
1
1
7
/
1
2
.
3
0
3
4
8
2
7
.
[
1
1
]
K
.
A
.
S
u
l
t
a
n
p
u
r
e
,
J.
B
a
g
a
d
e
,
S
.
L.
B
a
n
g
a
r
e
,
M
.
L.
B
a
n
g
a
r
e
,
K
.
D
.
B
a
m
a
n
e
,
a
n
d
A
.
J.
P
a
t
a
n
k
a
r
,
“
I
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
b
a
se
d
d
i
g
i
t
a
l
t
w
i
n
s
f
o
r
d
i
a
g
n
o
si
s
o
f
b
r
a
i
n
t
u
m
o
r
b
y
a
n
a
l
y
z
i
n
g
M
R
I
i
mag
e
s,
”
M
e
a
s
u
rem
e
n
t
:
S
e
n
so
rs
,
v
o
l
.
3
3
,
Ju
n
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
m
e
a
s
e
n
.
2
0
2
4
.
1
0
1
2
2
0
.
[
1
2
]
N
.
I
.
M
d
.
A
sh
a
f
u
d
d
u
l
a
a
n
d
R
.
I
sl
a
m
,
“
C
o
n
t
o
u
r
TL
-
N
e
t
:
C
o
n
t
o
u
r
-
b
a
se
d
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
m
f
o
r
e
a
r
l
y
-
s
t
a
g
e
b
r
a
i
n
t
u
mo
r
d
e
t
e
c
t
i
o
n
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
B
i
o
m
e
d
i
c
a
l
I
m
a
g
i
n
g
,
v
o
l
.
2
0
2
4
,
p
p
.
1
–
2
0
,
A
p
r
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
4
/
6
3
4
7
9
2
0
.
[
1
3
]
S
.
M
.
A
l
q
h
t
a
n
i
e
t
a
l
.
,
“
I
mp
r
o
v
e
d
b
r
a
i
n
t
u
m
o
r
s
e
g
me
n
t
a
t
i
o
n
a
n
d
c
l
a
ss
i
f
i
c
a
t
i
o
n
i
n
b
r
a
i
n
M
R
I
w
i
t
h
F
C
M
-
S
V
M
:
A
d
i
a
g
n
o
st
i
c
a
p
p
r
o
a
c
h
,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
1
2
,
p
p
.
6
1
3
1
2
–
6
1
3
3
5
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
4
.
3
3
9
4
5
4
1
.
[
1
4
]
M
.
B
h
a
si
n
e
t
a
l
.
,
“
U
n
v
e
i
l
i
n
g
t
h
e
h
i
d
d
e
n
:
Le
v
e
r
a
g
i
n
g
m
e
d
i
c
a
l
i
m
a
g
i
n
g
d
a
t
a
f
o
r
e
n
h
a
n
c
e
d
b
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
u
s
i
n
g
C
N
N
a
r
c
h
i
t
e
c
t
u
r
e
s
,
”
T
ra
i
t
e
m
e
n
t
d
u
S
i
g
n
a
l
,
v
o
l
.
4
1
,
n
o
.
3
,
p
p
.
1
5
7
5
–
1
5
8
2
,
J
u
n
.
2
0
2
4
,
d
o
i
:
1
0
.
1
8
2
8
0
/
t
s.
4
1
0
3
4
5
.
[
1
5
]
R
.
F
a
r
n
o
o
s
h
a
n
d
H
.
N
o
u
sh
k
a
r
a
n
,
“
D
e
v
e
l
o
p
me
n
t
o
f
a
n
u
n
s
u
p
e
r
v
i
s
e
d
p
se
u
d
o
-
d
e
e
p
a
p
p
r
o
a
c
h
f
o
r
b
r
a
i
n
t
u
m
o
r
d
e
t
e
c
t
i
o
n
i
n
ma
g
n
e
t
i
c
r
e
so
n
a
n
c
e
i
m
a
g
e
s,
”
K
n
o
w
l
e
d
g
e
-
B
a
se
d
S
y
s
t
e
m
s
,
v
o
l
.
3
0
0
,
S
e
p
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
k
n
o
s
y
s.
2
0
2
4
.
1
1
2
1
7
1
.
[
1
6
]
E.
B
a
şar
a
n
,
“
A
n
e
w
b
r
a
i
n
t
u
mo
r
d
i
a
g
n
o
s
t
i
c
m
o
d
e
l
:
S
e
l
e
c
t
i
o
n
o
f
t
e
x
t
u
r
a
l
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
a
l
g
o
r
i
t
h
ms
a
n
d
c
o
n
v
o
l
u
t
i
o
n
n
e
u
r
a
l
Evaluation Warning : The document was created with Spire.PDF for Python.