Method and system for substantially reducing streak artifacts in helical cone beam computer tomography (CT)
View Patent ↗Streak artifacts arise in helical CT reconstruction with cone beam weighting (CBW) with helical pitch ratio between 0.5 and 1.0 in a prevalent 2PI mode. The sreak artifacts are substantially removed by applying upsampling to the measured data in the segment direction before weighting. Furthermore, by making the upsampling adaptive to the view Z-position, an amount of extra processing is greatly reduced to near 1%.
1. A method of weighting measured helical cone beam data in computed tomography, comprising the steps of:
a) upsampling the measured helical cone beam data in a segment direction to obtain interpolated data with a reduced sampling pitch near a predetermined segment;
b) determining weights based upon a predetermined cone beam weight function corresponding to the measured helical cone beam data and the interpolated data; and
c) reconstructing an image using the weights, the measured data and the interpolated data.
2. The method of weighting measured helical cone beam data in computed tomography according to claim 1 wherein the reconstruction is between one and two rotations.
3. The method of weighting measured helical beam data in computed tomography according to claim 1 wherein the upsampling is performed with a fixed upsampling factor.
4. The method of weighting measured helical cone beam data in computed tomography according to claim 1 wherein the upsampling is performed with an adaptive upsampling factor.
5. The method of weighting measured helical cone beam data in computed tomography according to claim 4 wherein the adaptive upsampling factor depends on a view position.
6. The method of weighting measured helical cone beam data in computed tomography according to claim 5 wherein a number of the segments to be processed during the reconstruction is relatively constant.
7. The method of weighting measured helical cone beam data in computed tomography according to claim 5 wherein the adaptive upsampling factor U′ is determined by
U
′
=
(
(
Tseg
MIN
-
2
)
Δ
view
×
ViewRev
HP
)
where Δview is a distance from the central view, Tseg MIN is a minimal number of total segments to add, ViewRev is a number of views per revolution and HP is a helical pitch.
8. The method of weighting measured helical cone beam data in computed tomography according to claim 1 wherein the upsampling is performed in a fixed number of views.
9. The method of weighting measured helical cone beam data in computed tomography according to claim 1 wherein the upsampling is performed in an adaptive number of views (view positions).
10. The method of weighting measured helical cone beam data in computed tomography according to claim 9 wherein the adaptive number of views Uview is determined by
Uview
=
2
×
ceil
(
(
Tseg
MIN
-
2
)
ViewRev
HP
)
where ceil is a predetermined function, Tseg MIN is a minimal number of total segments to add, ViewRev is a number of views per revolution and HP is a helical pitch.
11. The method of weighting measured helical cone beam data in computed tomography according to claim 1 wherein the predetermined cone beam weight function is W CBW , which is a non-linear function as defined by
W
CBW
(
β
,
γ
,
v
)
=
u
CBW
(
β
,
γ
,
v
)
∑
n
=
-
N
n
=
N
u
CBW
(
β
n
C
(
β
,
γ
)
,
γ
n
C
(
γ
)
,
v
n
C
(
γ
,
v
)
)
wherein β is the view angle, (γ, v) are the detector coordinates respectively in fan angle and vertical coordinates, and superscript C stands for complementary.
12. A system for weighting measured helical cone beam data in computed tomography, comprising:
an upsampling unit for upsampling the measured helical cone beam data in a segment direction to obtain interpolated data with a reduced sampling pitch near a predetermined segment;
a weight determining unit connected to said upsampling unit for determining weights based upon a predetermined cone beam weight function corresponding to the measured helical cone beam data and the interpolated data; and
a reconstruction unit connected to said upsampling unit and said weight determining unit for reconstructing an image using the weights, the measured data and the interpolated data.
13. The system for weighting measured helical cone beam data in computed tomography according to claim 12 wherein the reconstruction is between one and two rotations.
14. The system for weighting measured helical beam data in computed tomography according to claim 12 wherein said upsampling unit utilizes a fixed upsampling factor.
15. The system for weighting measured helical cone beam data in computed tomography according to claim 12 wherein said upsampling unit utilizes an adaptive upsampling factor.
16. The system for weighting measured helical cone beam data in computed tomography according to claim 15 wherein the adaptive upsampling factor depends on a view position.
17. The system for weighting measured helical cone beam data in computed tomography according to claim 16 wherein a number of the segments to be processed during the reconstruction is relatively constant.
18. The system for weighting measured helical cone beam data in computed tomography according to claim 16 wherein the adaptive upsampling factor U′ is determined by
U
′
=
(
(
Tseg
MIN
-
2
)
Δ
view
×
ViewRev
HP
)
where Δview is a distance from the central view, Tseg MIN is a minimal number of total segments, ViewRev is a number of views per revolution and HP is a helical pitch.
19. The system for weighting measured helical cone beam data in computed tomography according to claim 12 wherein said upsampling unit utilizes a fixed number of views.
20. The system for weighting measured helical cone beam data in computed tomography according to claim 12 wherein said upsampling unit utilizes an adaptive number of views (view positions).
21. The system for weighting measured helical cone beam data in computed tomography according to claim 20 wherein the adaptive number of views Uview is determined by
Uview
=
2
×
ceil
(
(
Tseg
MIN
-
2
)
ViewRev
HP
)
where ceil is a predetermined function, Tseg MIN is a minimal number of total segments, ViewRev is a number of views per revolution and HP is a helical pitch.
22. The system for weighting measured helical cone beam data in computed tomography according to claim 12 wherein the predetermined cone beam weight function is W CBW , which is a non-linear function as defined by
W
CBW
(
β
,
γ
,
v
)
=
u
CBW
(
β
,
γ
,
v
)
∑
n
=
-
N
n
=
N
u
CBW
(
β
n
C
(
β
,
γ
)
,
γ
n
C
(
γ
)
,
v
n
C
(
γ
,
v
)
)
wherein β is the view angle, (γ, v) are the detector coordinates respectively in fan angle and vertical coordinates, and superscript C stands for complementary.