Renormalization of dual-energy images
View Patent ↗A method is disclosed for generating a composite image from dual-energy projection radiographic image data and renormalizing the composite image such that the pixel values for the composite image have the same relationship to the detected x-ray energy as the original high- and low-energy images. The method disclosed further provides for renormalization of material specific decomposition images formed form dual energy projection radiographic image data such that the decomposition images are on a common scale with either the high-energy, low-energy or composite image.
1. A computer-implemented method of producing a composite image from dual energy image data comprising:
calculating a weighted sum of image data for a high-energy radiographic image and a low-energy radiographic image, wherein the image data for the low-energy radiographic image is weighted by a weighting factor.
2. The method of claim 1 wherein the weighting factor is a weighting factor, w c .
3. The method of claim 2 wherein composite image data is renormalized such that the composite image data has a same relationship to detected x-ray energy as the high- and low-energy radiographic images.
4. A method of producing a composite image from dual energy image data comprising:
calculating a weighted sum of image data for a high-energy radiographic image and a low-energy radiographic image,
wherein the sum is weighted by a weighting factor, w c , and
wherein the composite image is normalized by a factor {1+w c } such that composite image data has a same relationship to detected x-ray energy as the high- and low-energy radiographic images.
5. The method of claim 4 wherein pixel values of an exposure independent composite image are computed as:
PV
C
*
=
PV
H
*
+
w
C
PV
L
*
[
1
+
w
C
]
=
-
μ
BC
t
B
-
μ
SC
t
S
wherein μ BC represents a bone attenuation coefficient of the composite image, μ SC represents a soft-tissue attenuation coefficient of the composite image, t B is a thickness of bone, and t S is a thickness of soft-tissue.
6. The method of claim 5 wherein the attenuation coefficients μ BC and μ SC are given by a weighted average of the bone and soft-tissue coefficients for the high- and low-energy radiographic images.
7. The method of claim 5 wherein a value for an unattenuated exposure for the composite image is added as an offset.
8. The method of claim 7 wherein pixel values of an exposure dependent composite image are computed as:
PV
C
#
=
PV
H
*
+
w
C
PV
L
*
[
1
+
w
C
]
+
log
(
I
C
0
)
=
log
(
I
C
0
)
-
μ
BC
t
B
-
μ
SC
t
S
.
9. The method of claim 8 wherein the unattenuated exposure for the composite image is given by the equation:
log
(
I
C
0
)
=
[
log
(
I
H
0
)
+
w
C
log
(
I
L
0
)
]
[
1
+
w
C
]
.
10. The method of claim 8 wherein the unattenuated exposure for the composite image is given by the equation:
log( I C0 )=log( I H0 +I L0 ).
11. A computer-implemented method for renormalizing dual-energy decomposition images comprising:
renormalizing the decomposition images such that a relationship between pixel values of at least one of the decomposition image to an original detected x-ray energy is well-defined.
12. The method of claim 11 wherein the decomposition images consist of a bone image and a soft tissue image.
13. The method of claim 11 wherein the relationship between the pixel values of the decomposition images to the original detected x-ray energy is the same as in an original high-energy image.
14. The method of claim 11 wherein the relationship between the pixel values of the decomposition images to the original detected x-ray energy is the same as in an original low-energy image.
15. The method of claim 11 wherein the relationship between the pixel values of the decomposition image to the original detected x-ray energy is the same as in a composite image formed from high- and low-energy images.
16. The method of claim 15 wherein pixel values of an exposure independent soft-tissue image are computed as:
PV
S
″
=
[
w
B
+
w
C
]
[
w
B
-
w
S
]
PV
H
*
-
w
S
PV
L
*
1
+
w
C
=
-
μ
SH
+
w
C
μ
SL
[
1
+
w
C
]
t
S
=
-
μ
SC
t
S
.
17. The method of claim 16 wherein a value for an unattenuated exposure for the composite image is added as an offset.
18. The method of claim 17 wherein pixel values of an exposure dependent composite image are computed as:
PV S # =log( I C0 )+ PV S ′″=log( I C0 )−μ SC t S .
19. The method of claim 18 wherein the unattenuated exposure for the composite image is given by the equation:
log
(
I
C
0
)
=
[
log
(
I
H
0
)
+
w
C
log
(
I
L
0
)
]
[
1
+
w
C
]
.
20. The method of claim 18 wherein the unattenuated exposure for the composite image is given by the equation:
log( I C0 )=log( I H0 +I L0 ).
21. The method of claim 15 wherein pixel values of an exposure independent bone image are computed as:
PV
B
′′′
=
[
w
S
+
w
C
]
[
w
S
-
w
B
]
PV
H
*
-
w
B
PV
L
*
1
+
w
C
=
-
μ
BH
+
w
C
μ
BL
[
1
+
w
C
]
t
B
=
-
μ
BC
t
B
.
22. The method of claim 21 wherein a value for an unattenuated exposure for the composite image is added as an offset.
23. The method of claim 22 wherein pixel values of an exposure dependent bone image are computed as:
PV B # =log( I C0 )+ PV B ′″=log( I C0 )−μ BC t B .
24. The method of claim 23 wherein the unattenuated exposure for the compose image is given by the equation:
log
(
I
C
0
)
=
[
log
(
I
H
0
)
+
w
C
log
(
I
L
0
)
]
[
1
+
w
C
]
.
25. The method of claim 23 wherein the unattenuated exposure for the composite image is given by the equation:
log( I C0 )=log( I H0 +I L0 ).