SYSTEM AND METHOD FOR ESTIMATING AND MANIPULATING ESTIMATED RADIATION DOSE

• US 20130102830A1
• Filed: 06/22/2011
• Published: 04/25/2013
• Est. Priority Date: 06/22/2010
• Status: Active Grant

• Current Assignee: Siemens Healthineers International AG
• Original Assignee: Varian Medical Systems International AG (Siemens AG)
• Inventors: Otto, Karl
• Granted Patent: US 9,289,627 B2
• Time in Patent Office: Days
• US Class Current: 600/1
• CPC Class Codes: A61N 5/00 Radiation therapy ultrasound therapy A61N7/00; devices or apparatus applicable to both therapy and diagnosis A61B6/00
• CPC Class Codes: A61N 5/10 X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy A61N5/01 takes precedence
• CPC Class Codes: A61N 5/1031 using a specific method of dose optimization
• CPC Class Codes: G01T 1/29 Measurement performed on radiation beams, e.g. position or section of the beam; Measurement of spatial distribution of radiation
• CPC Class Codes: H05K 999/00 dummy group
• CPC Class Codes: H05K 999/99 dummy group

Abstract

Methods are provided for permitting manipulation of an achievable dose distribution estimate deliverable by a radiation delivery apparatus for proposed treatment of a subject. One such method comprises: determining a dose modification voxel for which it is desired to modify the dose value and a corresponding magnitude of desired dose modification; for each of a plurality of beams: (i) characterizing the beam as a two-dimensional array of beamlets, wherein each beamlet is associated with a corresponding intensity value and a ray line representing the projection of the beamlet into space; and (ii) identifying one or more dose-change beamlets which have associated ray lines that intersect the dose modification voxel; modifying the intensity values of at least one of the dose-change beamlets; and updating the achievable dose distribution estimate to account for the modified intensity values of the at least one of the dose-change beamlets.

Claims

1. A method for permitting manipulation of an achievable dose distribution estimate deliverable by a radiation delivery apparatus for proposed treatment of a subject, the achievable dose distribution estimate defined over a three-dimensional range of voxels with a dose value for each voxel, the method comprising:

(a) determining a dose modification voxel for which it is desired to modify the dose value and determining a corresponding magnitude of desired dose modification;

(b) for each of a plurality of beams;

(i) characterizing the beam as a two-dimensional array of beamlets, wherein each beamlet is associated with a corresponding intensity value and a ray line representing a projection of the beamlet into space; and

(ii) identifying one or more dose-change beamlets which have associated ray lines that intersect the dose modification voxel;

(c) modifying the intensity values of the identified dose-change beamlets based on the magnitude of desired dose modification, wherein modifying the intensity value of the identified dose-change beamlets comprises, for each of the identified dose-change beamlets, modifying the intensity value of the identified dose-change beamlet by a corresponding amount, the corresponding amount based on the magnitude of the desired dose modification; and

(d) updating the achievable dose distribution estimate to account for the modified intensity values of the identified dose-change beamlets.

2. A method according to claim 1 wherein determining the dose modification voxel comprises:

receiving, as input, a requested modification to a dose quality metric; and

using the requested modification to the dose quality metric to determine the dose modification voxel.

7. A method according to claim 2 wherein using the requested modification to the dose quality metric to determine the dose modification voxel comprises solving an inverted dose quality metric function that expresses coordinates and magnitudes of dose in terms of values of the dose quality metric.

8. A method according to claim 2 wherein the dose quality metric comprises a DVH curve associated with a tissue structure in the subject and wherein using the requested modification to the dose quality metric to determine the dose modification voxel comprises identifying a voxel in the tissue structure of the subject to be the dose modification voxel.

9. A method according to claim 8 wherein identifying the voxel in the tissue structure of the subject to be the dose modification voxel comprises identifying all of the voxels in the tissue structure of the subject to be dose modification voxels and performing steps (b), (c) and (d) for each of the dose modification voxels.

10. A method according to claim 8 wherein identifying the voxel in the tissue structure of the subject to be the dose modification voxel comprises:

receiving image data for the subject, the image data comprising an image of the tissue structure; and

associating particular voxels with the tissue structure based on the image data.

11. A method according to claim 2 wherein the dose quality metric comprises a DVH curve and wherein using the requested modification to the dose quality metric to determine the dose modification voxel comprises:

identifying a location of the requested modification on the DVH curve to have a dose value D_selected; and

identifying a voxel in the achievable dose distribution estimate having a dose value in a range D_selected±

Δ

to be the dose modification voxel, where Δ

is a range parameter.

12. A method according to claim 11 wherein using the requested modification to the dose quality metric to determine the dose modification voxel comprises:

if an attempt to identify the voxel in the achievable dose distribution estimate having the dose value in the range D_selected±

Δ

is not successful, increasing the range parameter Δ and

repeating the attempt to identify the voxel with the increased range parameter Δ

.

13. A method according to claim 11 wherein identifying the voxel in the achievable dose distribution estimate having the dose value in the range D_selected±

Δ

to be the dose modification voxel comprises identifying a plurality of candidate dose modification voxels having dose values in the range D_selected±

Δ

.

14. A method according to claim 13 wherein if the plurality of candidate dose modification voxels comprise a number of candidate dose modification voxels that is greater than a threshold number V_max, then the method comprises decreasing the range parameter Δ

and repeating the attempt to identify the voxel in the achievable dose distribution estimate having the dose value in the range D_selected±

Δ

to be the dose modification voxel with the decreased range parameter Δ

.

20. A method according to claim 1 comprising determining one or more secondary dose modification voxels and corresponding magnitudes of desired dose modification for the secondary dose modification voxels and treating the secondary dose modification values as dose modification voxels when performing steps (b), (c) and (d).

21. A method according to claim 20 wherein determining one or more secondary dose modification voxels comprises determining the one or more secondary dose modification voxels to be in a marginal region proximate to the dose modification voxel.

22. A method according to claim 20 wherein the corresponding magnitudes of desired dose modification for the secondary dose modification voxels are less than the magnitude of the desired dose modification for the dose modification voxel.

23. A method according to claim 1 wherein determining the magnitude of the desired dose modification comprises determining the magnitude to be one or more of:

a fraction f (0≦

f≦

1) of the dose value for the dose modification voxel;

a fraction f (0≦

f≦

1) of a maximum dose restriction for a tissue structure at the location of the dose modification voxel;

a fraction f (0≦

f≦

1) of a prescription dose amount for a target tissue structure at the location of the dose modification voxel;

an amount proportional to movement of a computer pointing device;

an amount input by an operator;

an operator-configurable fixed amount;

a fixed amount that is a parameter of a system on which the method is performed.

29. A method according to claim 1 wherein modifying the intensity values of one or more of the dose-change beamlets comprises determining magnitudes of intensity value changes for the one or more dose-change beamlets according to a function that is dependent on the magnitude of desired dose modification and the number of the one or more dose-change beamlets, and wherein the function is proportional to magnitude of desired dose modification and inversely proportional to the number of the one or more dose-change beamlets.

30. A method according to claim 1 wherein modifying the intensity values of one or more of the dose-change beamlets comprises subjecting proposed intensity value modifications to one or more intensity restrictions.

31. A method according to claim 30 wherein the one or more intensity restrictions comprise one or more of:

the intensity values of the one or more dose-change beamlets must be greater than a minimum threshold;

the intensity values of the one or more dose-change beamlets must be less than a maximum threshold;

a spatial variation between the intensity values of the one or more dose-change beamlets and the intensity values of one or more neighboring beamlets must be less than a maximum variation threshold.

33. A method according to claim 1 comprising, for each of the plurality of beams and for each of the one or more dose-change beamlets, determining one or more secondary dose-change beamlets and treating the secondary dose-change beamlets as dose-change beamlets when performing steps (c) and (d).

34. A method according to claim 33 wherein, for each of the plurality of beams and for each of the one or more dose-change beamlets, determining one or more secondary dose-change beamlets comprises determining the one or more secondary dose-change beamlets to be in a marginal region proximate to the dose-change beamlet.

35. A method according to claim 34 wherein modifying the intensity values of one or more of the dose change beamlets comprises, for each of the one or more dose change beamlets, modifying the intensity values of the secondary dose change beamlets in the marginal region proximate to the dose-change beamlet by magnitudes less than a magnitude of the change to the dose-change beamlet.

36. A method according to claim 1 wherein updating the achievable dose distribution estimate comprises using a dose estimation technique to determine dose contributions attributable to the modified intensity values of the one or more dose-change beamlets, the dose estimation technique comprising one or more of:

Monte Carlo, collapsed cone convolution, pencil beam, anisotropic analytical algorithm and Boltzman equation solvers.

39. A method according to claim 1 wherein updating the achievable dose distribution estimate comprises, for each of the plurality of beams comprising one or more dose change beamlets:

convolving the intensity values of the two-dimensional array of beamlets with a two-dimensional dose estimate kernel to obtain a two-dimensional convolved intensity distribution, the two-dimensional convolved intensity distribution comprising a convolved intensity value for each beamlet; and

for each beamlet in the two-dimensional array of beamlets;

identifying voxels that are intersected by the ray line associated with the beamlet; and

adding a dose contribution to the intersected voxels, the added dose contribution based on the convolved intensity value of the beamlet.

40. A method according to claim 39 wherein the dose estimate kernel comprises a two-dimensional point spread function.

41. A method according to claim 39 wherein the dose estimate kernel comprises a linear combination of two dimensional point spread functions.

43. A method according to claim 39 wherein convolving the intensity values of the two-dimensional array of beamlets with the two-dimensional dose estimate kernel comprises:

multiplying a Fourier transform of the intensity values of the two-dimensional array of beamlets with a Fourier transform of the two-dimensional dose estimate kernel to obtain a Fourier multiplication result; and

determining an inverse Fourier transform of the Fourier multiplication result to obtain the two-dimensional convolved intensity distribution.

45. A method according to claim 39 wherein, in cases where a plurality of ray lines from a single beam intersect a particular voxel, adding the dose contribution to the particular voxel comprises adding the dose contribution based on a linear combination of the convolved intensity values of the plurality of beamlets corresponding to the plurality of ray lines.

46. A method according to claim 39 wherein, for each beamlet in the two-dimensional array of beamlets and for each intersected voxel, the added dose contribution is based on the convolved intensity value of the beamlet multiplied by an attenuation factor, the attenuation factor decreasing with a depth of the intersected voxel within the subject.

49. A method according to claim 1 wherein updating the achievable dose distribution estimate comprises, for each of the plurality of beams comprising one or more dose change beamlets:

convolving the intensity values of the two-dimensional array of beamlets with a plurality of two-dimensional dose estimate kernels to obtain a corresponding plurality of two-dimensional convolved intensity distributions, each two-dimensional convolved intensity distribution comprising a convolved intensity value for each beamlet, wherein each dose estimate kernel is associated with a different range of tissue densities; and

for each beamlet in the two-dimensional array of beamlets;

identifying voxels that are intersected by the ray line associated with the beamlet; and

adding a dose contribution to the intersected voxels, the dose contribution for any one intersected voxel based on the convolved intensity value of the beamlet from one of the plurality of two-dimensional convolved intensity distributions, the one of the plurality of two-dimensional convolved intensity distributions selected based on tissue density of the one intersected voxel.

50. A method according to claim 1 wherein updating the achievable dose distribution estimate comprises scaling the updated achievable dose distribution estimate such that the updated achievable dose distribution estimate more closely approximates the magnitude of desired dose modification corresponding to the dose modification voxel.

51. A method according to claim 1 wherein updating the achievable dose distribution estimate comprises subjecting the updated achievable dose distribution to one or more dose restrictions.

52. A method according to claim 51 wherein the one or more dose restrictions comprise one or more of:

the updated achievable dose distribution must not reduce dose in voxels corresponding to target tissue structures in the subject below a minimum target structure threshold; and

the updated achievable dose distribution must not increase dose in voxels corresponding to healthy tissue structures in the subject above a maximum healthy structure threshold.

53. A method according to claim 51 wherein, if subjecting the updated achievable dose distribution to one or more dose restrictions concludes that there is a violation of any dose restriction, then, the method comprises addressing the violation of the dose restriction, wherein addressing the violation of the dose restriction comprises one or more of:

rejecting the update to the achievable dose distribution estimate; and

adjusting the magnitude of the desired dose modification corresponding to the dose modification voxel and repeating steps (b), (c) and (d) with the adjusted magnitude.

54. A method according to claim 51 wherein, if subjecting the updated achievable dose distribution to one or more dose restrictions concludes that there is a violation of any dose restriction, then, the method comprises:

addressing the violation of the dose restriction, wherein addressing the violation of the dose restriction comprises;

determining one or more restriction violating voxels which cause the updated dose distribution to violate the dose restriction and one or more corresponding desired dose change magnitudes;

repeating steps (b), (c) and (d) with the one or more restriction violating voxels and the one or more corresponding desired dose change magnitudes in place of the dose modification voxel and the corresponding magnitude of the desired dose modification;

re-subjecting the updated achievable dose distribution to the one or more dose restrictions; and

repeating addressing the violation of the dose restriction until subjecting the updated achievable dose distribution to one or more dose restrictions concludes that there is not a violation of any dose restriction.

55. A method according to claim 1 comprising, after updating the achievable dose distribution estimate, determining one or more updated dose quality metrics using the updated achievable dose distribution estimate.

56. A method according to claim 55 wherein determining one or more updated dose quality metrics comprises subjecting at least one updated dose quality metric to one or more dose quality metric restrictions.

57. A method according to claim 56 wherein the one or more dose quality metric restrictions comprise one or more of:

the at least one updated dose quality metric must have a minimum percentage volume of a target structure with dose above a minimum target structure threshold; and

the at least one updated dose quality metric must have a minimum percentage volume of a healthy tissue structure with dose lower than a maximum healthy tissue structure threshold.

60. A method according to claim 1 comprising, for each of the plurality of beams, initializing the intensity values of the beamlets in the two-dimensional array of beamlets, wherein initializing the intensity values of the beamlets in the two-dimensional array of beamlets comprises one or more of:

setting all intensity values to zero;

setting intensity values according to a random process;

setting intensity values such that an approximate sphere of dose encompasses target tissue structures;

setting intensity values for beamlets having ray lines that intersect target structures to be a first positive value and setting intensity values for other beamlets to have a lower value;

setting intensity values such that a resultant dose distribution estimate has a maximum dose higher than a threshold; and

setting intensity values for all or a subset of the beamlets based on a function of a prescription dose for a target structure.

61. A method according to claim 1 comprising, after updating the achievable dose distribution estimate, determining one or more radiation delivery parameters based on the updated achievable dose estimate, the one or more radiation delivery parameters suitable for use by the radiation delivery apparatus for treatment of the subject.

62. A method according to claim 61 wherein determining one or more radiation delivery parameters based on the updated achievable dose estimate comprises performing an iterative optimization process.

63. A method according to claim 61 wherein performing the iterative optimization process comprises optimizing the one or more radiation delivery parameters which will provide an optimized dose distribution estimate similar within an acceptable range to the updated achievable dose estimate.

64. A method according to claim 62 wherein performing the iterative optimization process comprises estimating an optimized dose distribution over a three-dimensional range of voxels resulting from an optimization beam having a spatially varying two-dimensional optimization intensity distribution characterized by a two-dimensional array of optimization beamlets wherein each optimization beamlet is associated with a corresponding optimization intensity value, the estimating comprising:

associating an optimization ray line with each optimization beamlet, the optimization ray line representing a projection of the optimization beamlet into space;

convolving the two-dimensional optimization intensity distribution with a two-dimensional optimization dose estimate kernel to obtain a two-dimensional optimization convolved intensity distribution, the two-dimensional optimization convolved intensity distribution comprising an optimization convolved intensity value for each optimization beamlet; and

for each optimization beamlet in the two-dimensional array of optimization beamlets;

identifying voxels in the three-dimensional range of voxels that are intersected by the optimization ray line associated with the optimization beamlet; and

adding an optimization dose contribution to the intersected voxels, the added optimization dose contribution based on the optimization convolved intensity value of the optimization beamlet.

66. A computer program product carrying instructions embodied in a non-transitory computer-readable medium, the instructions when executed by a suitable processor cause the processor to perform a method for permitting manipulation of an achievable dose distribution estimate deliverable by a radiation delivery apparatus for proposed treatment of a subject, the achievable dose distribution estimate defined over a three-dimensional range of voxels with a dose value for each voxel, the method comprising the method of claim 1.

3-6. -6. (canceled)

15-19. -19. (canceled)

24-28. -28. (canceled)

32. (canceled)

37-38. -38. (canceled)

42. (canceled)

44. (canceled)

47-48. -48. (canceled)

58-59. -59. (canceled)

65. A system for permitting manipulation of an achievable dose distribution estimate deliverable by a radiation delivery apparatus for proposed treatment of a subject, the achievable dose distribution estimate defined over a three-dimensional range of voxels with a dose value for each voxel, the system comprising a controller configured to:

determine a dose modification voxel for which it is desired to modify the dose value and a corresponding magnitude of desired dose modification;

for each of a plurality of beams;

characterize the beam as a two-dimensional array of beamlets, wherein each beamlet is associated with a corresponding intensity value and a ray line representing a projection of the beamlet into space; and

identify one or more dose-change beamlets which have associated ray lines that intersect the dose modification voxel;

modify the intensity values of the identified dose-change beamlets based on the magnitude of the desired dose modification, wherein the controller is configured to modify the intensity value of the identified dose change beamlets by, for each of the identified dose-change beamlets, modifying the intensity value of the identified dose-change beamlet by a corresponding amount, the corresponding amount based on the magnitude of the desired dose modification; and

update the achievable dose distribution estimate to account for the modified intensity values of the identified dose-change beamlets.

67. A method for estimating a dose distribution over a three-dimensional range of voxels proposed to contain a subject resulting from one or more beams, each beam comprising a spatially varying two-dimensional intensity distribution characterized by a two-dimensional array of beamlets wherein each beamlet is associated with a corresponding intensity value, the method comprising, for each of the one or more beams:

associating a ray line with each beamlet, the ray line representing a projection of the beamlet into space;

convolving the two-dimensional intensity distribution with a two-dimensional dose estimate kernel to obtain a two-dimensional convolved intensity distribution, the two-dimensional convolved intensity distribution comprising a convolved intensity value for each beamlet; and

for each beamlet in the two-dimensional array of beamlets;

identifying voxels in the three-dimensional range of voxels that are intersected by the ray line associated with the beamlet; and

adding a dose contribution to the intersected voxels, the added dose contribution based on the convolved intensity value of the beamlet.

83. A computer program product carrying instructions embodied in a non-transitory computer-readable medium, the instructions when executed by a suitable processor cause the processor to perform a method for estimating a dose distribution over a three-dimensional range of voxels proposed to contain a subject resulting from one or more beams, each beam comprising a spatially varying two-dimensional intensity distribution characterized by a two-dimensional array of beamlets wherein each beamlet is associated with a corresponding intensity value, the method comprising the method of claim 67.

68-81. -81. (canceled)

82. A system for estimating a dose distribution over a three-dimensional range of voxels proposed to contain a subject resulting from one or more beams, each beam comprising a spatially varying two-dimensional intensity distribution characterized by a two-dimensional array of beamlets wherein each beamlet is associated with a corresponding intensity value, the system comprising a controller configured to, for each of the one or more beams:

associate a ray line with each beamlet, the ray line representing a projection of the beamlet into space;

convolve the two-dimensional intensity distribution with a two-dimensional dose estimate kernel to obtain a two-dimensional convolved intensity distribution, the two-dimensional convolved intensity distribution comprising a convolved intensity value for each beamlet; and

for each beamlet in the two-dimensional array of beamlets;

identify voxels in the three-dimensional range of voxels that are intersected by the ray line associated with the beamlet; and

add a dose contribution to the intersected voxels, the added dose contribution based on the convolved intensity value of the beamlet.

84-88. -88. (canceled)