Error control in algorithmic approach to step-and-shoot intensity modulated radiation therapy
The present invention provides a method comprising the following steps: (a) partitioning an intensity modulated beam into a set of sub-IMBs; and (b) partitioning the sub-IMBs into segments, wherein steps (a) and (b) introduce no machine delivery error. The present invention also provides a method comprising the following steps: (a) recursively partitioning an intensity modulated beam into plateaus; and (b) partitioning the plateaus into segments, wherein step (a) comprises determining a tradeoff between machine delivery error and the number of segments into which the plateaus will be partitioned in step (b).
1. A method comprising the following steps:
(a) recursively partitioning an intensity modulated beam into plateaus; and
(b) partitioning the plateaus into segments; and
(c) storing in a storage medium a treatment plan based on the segments of step (b), wherein step (a) comprises determining a tradeoff between machine delivery error and the number of segments into which the plateaus will be partitioned in step (b).
2. The method of claim 1 , where step (a) further comprises: partitioning the intensity modulated beam (IMB) into a sub-IMB, and wherein the method further comprises the following step:
(d) determining the tradeoff curve between the number of segments and error for the sub-IMB.
3. A method comprising the following steps:
(a) recursively partitioning an intensity modulated beam into plateaus;
(b) partitioning the plateaus into segments; and
(c) storing in a storage medium a treatment plan based on the segments of step (b), wherein step (a) comprises determining a tradeoff between machine delivery error and the number of segments into which the plateaus will be partitioned in step (b) and wherein step (b) comprises the following the steps:
(d) constructing a directed graph based on the plateaus;
(e) determining a series of minimum cost flows in the directed graph;
(f) determining a tradeoff between the number of segments and machine delivery errors based on the determined minimum cost flows, and
(g) extracting the segments from the directed graph based on the tradeoff determined in step (f) to thereby partition the plateaus.
4. The method of claim 3 , wherein step (d) comprises the following steps:
(h) constructing a directed graph based on intensity modulated beam grids of the plateaus of an intensity modulated beam, wherein vertices of the directed graph include all intersection points of the intensity modulated beam grids of the plateaus that are either part of a peak or an interior of a polygon corresponding to the intensity modulated beam, and wherein the edges of the directed graph include those edges of the plateaus whose end points are graph vertices in the directed graph;
(i) for all graph vertices corresponding to the same peak direction, adding either a source vertex or sink vertex to the directed graph;
(j) adding edges to the directed graph from the source vertex or sink vertex to respective peak vertices;
(k) assigning a unit flow capacity to each graph vertex and edge of the directed graph; and
(l) to each edge that is vertical assigning a cost that is equal to the geometric length of the vertical edge, and to each edge that is horizontal a cost of zero.
5. The method of claim 4 , wherein step (h) comprises the following steps:
(m) extracting connections between a left peak and a right peak of the polygon based on the minimum cost flow;
(n) eliminating any remaining unconnected peaks using only horizontal edges; and
(o) extracting resulting segments from the directed graph to thereby partition the plateaus.
6. The method of claim 5 , wherein for an edge involving a left peak vertex, the direction of the edge is away from the left peak.
7. The method of claim 5 , wherein for an edge involving a right peak vertex, the direction of the edge is toward the right peak.
8. The method of claim 5 , wherein the direction of a horizontal edge that does not involve any peak vertex is from left to right.
9. The method of claim 5 , wherein the direction of a vertical edge that does not involve a peak vertex is bidirectional.
10. The method of claim 5 , wherein, for all left peak vertices, a source vertex is added to the directed graph.
11. The method of claim 5 , wherein, for all right peak vertices, a sink vertex is added to the directed graph.
12. The method of claim 5 , wherein the minimum cost flow in the directed graph yields a maximum set of mutually disjoint paths.
13. A method comprising the following steps:
(a) recursively partitioning an intensity modulated beam into plateaus;
(b) partitioning the plateaus into segments; and
(c) storing in a storage medium a treatment plan based on the segments of step (b), wherein step (a) comprises determining a tradeoff between machine delivery error and the number of segments into which the plateaus will be partitioned in step (b), wherein the intensity modulated beam comprises a plurality of z-posts, and wherein step (a) comprises the following steps:
(d) choosing an intensity level d wherein, 1≦d≦I max , wherein I max is the maximum intensity level of the intensity modulated beam;
(e) selecting all z-posts having a height of at least d to form a set of selected z-posts;
(f) cutting each selected z-post whose height is ≧d at level d to form a plateau of z-posts of height d;
(g) removing the plateau of z-posts from the intensity modulated beam; and
(h) recursively partitioning any remaining portion of the intensity modulated beam until there is no remaining portion of the intensity modulated beam.
14. The method of claim 13 , wherein step (h) comprises repeating steps (d), (e), (f) and (g) on any remaining portion of the intensity modulated beam until there is no remaining portion of the intensity modulated beam.
15. The method of claim 13 , wherein d is a function of the maximum intensity level I max .
16. The method of claim 13 , wherein
d
=
⌈
I
max
2
⌉
.
17. The method of claim 13 , wherein d is based on the volume of the plateau of z-posts.
18. The method of claim 13 wherein d=2 i , where i=Int(log 2 (I max ))−1).
19. The method of claim 13 , wherein d has a value such that the volume of the plateau of z-posts is maximized.
20. The method of claim 13 , wherein d is determined by the slice method.
21. A method comprising the following steps:
(a) recursively partitioning an intensity modulated beam into plateaus;
(b) partitioning the plateaus into segments; and
(c) storing in a storage medium a treatment plan based on the segments of step (b), wherein step (a) comprises partitioning the intensity modulated beam into a sub-IMB and wherein step (a) comprises determining the tradeoff curve between the number of segments and error for the sub-IMB by the following steps:
(d) generating a canonical delivery option for each set of field opening endpoints for a plurality of intensity modulated beam columns of the sub-IMB;
(e) for every two consecutive intensity modulated beam columns, determining a series of minimum cost g-matchings between the canonical delivery options of the intensity modulated beam columns;
(f) determining the tradeoff curves based on the g-matchings; and
(g) extracting the segments for the sub-IMB based on the matched field openings of the canonical delivery options from the determined g-matchings in an increasing order of consecutive intensity modulated beam columns to thereby partition the sub-IMB.
22. The method of claim 21 , wherein in step (d) only one set of left and right field opening endpoints is generated for each intensity beam column.
23. The method of claim 21 , wherein in step (d) more than one set of left and right field opening endpoints is generated for each intensity beam column.
24. The method of claim 21 , wherein the matched field openings have no interdigitation and minimum leaf separation.
25. The method of claim 21 , wherein the matched field openings have no minimum leaf separation and no interdigitation.
26. The method of claim 21 , wherein the matched field openings have interdigitation.
27. The method of claim 21 , wherein the method further comprises choosing a number m of Steiner points for forming the sets of field opening endpoints, and wherein each unique canonical delivery option contains at most m Steiner points for each set of field opening endpoints.
28. A method comprising the following steps:
(a) recursively partitioning an intensity modulated beam into plateaus;
(b) partitioning the plateaus into segments, wherein step (a) comprises determining a tradeoff between machine delivery error and the number of segments into which the plateaus will be partitioned in step (b);
(c) linking the segments to thereby reduce the number of segments; and
(d) storing in a storage medium a treatment plan based on the linked segments of step (c).
29. The method of claim 28 , wherein step (a) further comprises: partitioning the intensity modulated beam (IMB) into a sub-IMB, and wherein the method further comprises the following step:
(e) determining the tradeoff curve between the number of segments and error for the sub-IMB.
30. A method comprising the following steps:
(a) recursively partitioning an intensity modulated beam into plateaus;
(b) partitioning the plateaus into segments, wherein step (a) comprises determining a tradeoff between machine delivery error and the number of segments into which the plateaus will be partitioned in step (b);
(c) linking the segments to thereby reduce the number of segments; and
(d) storing in a storage medium a treatment plan based on the linked segments of step (c), wherein step (c) further comprises the following steps:
(e) constructing a directed graph H based on the segments extracted;
(f) determining a minimum cost g-path cover on the directed graph H;
(g) extracting the final segments based on the g-path cover on H therefore segmenting the entire intensity modulated beam.
31. The method of claim 30 , wherein step (e) further comprises the following steps:
(h) introducing a vertex for each segment;
(i) introducing an edge between any two segments that can be delivered together; and
(j) assigning a cost to each edge that is the equal to the amount of machine delivery error introduced when delivering the two segments together.
32. The method of claim 31 , wherein in step (i), the two segments have no minimum leaf separation and no interdigitation when for a multileaf collimator.
33. The method of claim 31 , wherein in step (i), the two segments have interdigitation.
34. A method comprises the following steps:
(a) generating a canonical delivery option for each set of field opening endpoints for an intensity modulated beam;
(b) for every two consecutive intensity modulated beam columns of the intensity modulated beam, determining a series of minimum cost g-matchings between the canonical delivery options of the intensity modulated beam columns;
(c) determining the tradeoff curves based on the g-matchings;
(d) extracting the segments for the intensity modulated beam based on the matched field openings of the canonical delivery options from the determined g-matchings in an increasing order of consecutive intensity modulated beam columns to thereby partition the intensity modulate beam; and
(e) storing in a storage medium a treatment plan based on the extracted segments of step (d).
35. A method comprising the following steps:
(a) recursively partitioning an intensity modulated beam into plateaus; and
(b) partitioning the plateaus into segments; and
(c) implementing a treatment plan based on the segments of step (b), wherein step (a) comprises determining a tradeoff between machine delivery error and the number of segments into which the plateaus will be partitioned in step (b).
36. The method of claim 35 , wherein step (a) further comprises: partitioning the intensity modulated beam (IMB) into a sub-IMB, and wherein the method further comprises the following step:
(d) determining the tradeoff curve between the number of segments and error for the sub-IMB.
37. A method comprising the following steps:
(a) recursively partitioning an intensity modulated beam into plateaus;
(b) partitioning the plateaus into segments; and
(c) implementing a treatment plan based on the segments of step (b), wherein step (a) comprises determining a tradeoff between machine delivery error and the number of segments into which the plateaus will be partitioned in step (b) and wherein step (b) comprises the following the steps:
(d) constructing a directed graph based on the plateaus;
(e) determining a series of minimum cost flows in the directed graph;
(f) determining a tradeoff between the number of segments and machine delivery errors based on the determined minimum cost flows, and
(g) extracting the segments from the directed graph based on the tradeoff determined in step (f) to thereby partition the plateaus.
38. The method of claim 37 , wherein step (d) comprises the following steps:
(h) constructing a directed graph based on intensity modulated beam grids of the plateaus of an intensity modulated beam, wherein vertices of the directed graph include all intersection points of the intensity modulated beam grids of the plateaus that are either part of a peak or an interior of a polygon corresponding to the intensity modulated beam, and wherein the edges of the directed graph include those edges of the plateaus whose end points are graph vertices in the directed graph;
(i) for all graph vertices corresponding to the same peak direction, adding either a source vertex or sink vertex to the directed graph;
(j) adding edges to the directed graph from the source vertex or sink vertex to respective peak vertices;
(k) assigning a unit flow capacity to each graph vertex and edge of the directed graph; and
(l) to each edge that is vertical assigning a cost that is equal to the geometric length of the vertical edge, and to each edge that is horizontal a cost of zero.
39. The method of claim 38 , wherein step (h) comprises the following steps:
(m) extracting connections between a left peak and a right peak of the polygon based on the minimum cost flow;
(n) eliminating any remaining unconnected peaks using only horizontal edges; and
(o) extracting resulting segments from the directed graph to thereby partition the plateaus.
40. The method of claim 39 , wherein for an edge involving a left peak vertex, the direction of the edge is away from the left peak.
41. The method of claim 39 , wherein for an edge involving a right peak vertex, the direction of the edge is toward the right peak.
42. The method of claim 39 , wherein the direction of a horizontal edge that does not involve any peak vertex is from left to right.
43. The method of claim 39 , wherein the direction of a vertical edge that does not involve a peak vertex is bidirectional.
44. The method of claim 39 , wherein, for all left peak vertices, a source vertex is added to the directed graph.
45. The method of claim 39 , wherein, for all right peak vertices, a sink vertex is added to the directed graph.
46. The method of claim 39 , wherein the minimum cost flow in the directed graph yields a maximum set of mutually disjoint paths.
47. A method comprising the following steps:
(a) recursively partitioning an intensity modulated beam into plateaus;
(b) partitioning the plateaus into segments; and
(c) implementing a treatment plan based on the segments of step (b), wherein step (a) comprises determining a tradeoff between machine delivery error and the number of segments into which the plateaus will be partitioned in step (b), wherein the intensity modulated beam comprises a plurality of z-posts, and wherein step (a) comprises the following steps:
(d) choosing an intensity level d wherein, 1≦d≦I max , wherein I max is the maximum intensity level of the intensity modulated beam;
(e) selecting all z-posts having a height of at least d to form a set of selected z-posts;
(f) cutting each selected z-post whose height is ≧d at level d to form a plateau of z-posts of height d;
(g) removing the plateau of z-posts from the intensity modulated beam; and
(h) recursively partitioning any remaining portion of the intensity modulated beam until there is no remaining portion of the intensity modulated beam.
48. The method of claim 47 , wherein step (h) comprises repeating steps (d), (e), (f) and (g) on any remaining portion of the intensity modulated beam until there is no remaining portion of the intensity modulated beam.
49. The method of claim 47 , wherein d is a function of the maximum intensity level I max .
50. The method of claim 47 , wherein
d
=
⌈
I
max
2
⌉
.
51. The method of claim 47 , wherein d is based on the volume of the plateau of z-posts.
52. The method of claim 47 wherein d=2 i , where i=Int(log 2 (I max ))−1).
53. The method of claim 47 , wherein d has a value such that the volume of the plateau of z-posts is maximized.
54. The method of claim 47 , wherein d is determined by the slice method.
55. A method comprising the following steps:
(a) recursively partitioning an intensity modulated beam into plateaus;
(b) partitioning the plateaus into segments; and
(c) implementing a treatment plan based on the segments of step (b), wherein step (a) comprises partitioning the intensity modulated beam into a sub-IMB and wherein step (a) comprises determining the tradeoff curve between the number of segments and error for the sub-IMB by the following steps:
(d) generating a canonical delivery option for each set of field opening endpoints for a plurality of intensity modulated beam columns of the sub-IMB;
(e) for every two consecutive intensity modulated beam columns, determining a series of minimum cost g-matchings between the canonical delivery options of the intensity modulated beam columns;
(f) determining the tradeoff curves based on the g-matchings; and
(g) extracting the segments for the sub-IMB based on the matched field openings of the canonical delivery options from the determined g-matchings in an increasing order of consecutive intensity modulated beam columns to thereby partition the sub-IMB.
56. The method of claim 55 , wherein in step (d) only one set of left and right field opening endpoints is generated for each intensity beam column.
57. The method of claim 55 , wherein in step (d) more than one set of left and right field opening endpoints is generated for each intensity beam column.
58. The method of claim 55 , wherein the matched field openings have no interdigitation and minimum leaf separation.
59. The method of claim 55 , wherein the matched field openings have no minimum leaf separation and no interdigitation.
60. The method of claim 55 , wherein the matched field openings have interdigitation.
61. The method of claim 55 , wherein the method further comprises choosing a number m of Steiner points for forming the sets of field opening endpoints, and wherein each unique canonical delivery option contains at most m Steiner points for each set of field opening endpoints.
62. A method comprising the following steps:
(a) recursively partitioning an intensity modulated beam into plateaus;
(b) partitioning the plateaus into segments, wherein step (a) comprises determining a tradeoff between machine delivery error and the number of segments into which the plateaus will be partitioned in step (b);
(c) linking the segments to thereby reduce the number of segments; and
(d) implementing a treatment plan based on the linked segments of step (c).
63. The method of claim 62 , wherein step (a) further comprises: partitioning the intensity modulated beam (IMB) into a sub-IMB, and wherein the method further comprises the following step:
(f) determining the tradeoff curve between the number of segments and error for the sub-IMB.
64. A method comprising the following steps:
(a) recursively partitioning an intensity modulated beam into plateaus;
(b) partitioning the plateaus into segments, wherein step (a) comprises determining a tradeoff between machine delivery error and the number of segments into which the plateaus will be partitioned in step (b);
(c) linking the segments to thereby reduce the number of segments; and
(d) implementing a treatment plan based on the linked segments of step (c), wherein step (c) further comprises the following steps:
(e) constructing a directed graph H based on the segments extracted;
(f) determining a minimum cost g-path cover on the directed graph H;
(g) extracting the final segments based on the g-path cover on H therefore segmenting the entire intensity modulated beam.
65. The method of claim 64 , wherein step (e) further comprises the following steps:
(h) introducing a vertex for each segment;
(i) introducing an edge between any two segments that can be delivered together; and
(j) assigning a cost to each edge that is the equal to the amount of machine delivery error introduced when delivering the two segments together.
66. The method of claim 65 , wherein in step (i), the two segments have no minimum leaf separation and no interdigitation when for a multileaf collimator.
67. The method of claim 65 , wherein in step (i), the two segments have interdigitation.
68. A method comprises the following steps:
(a) generating a canonical delivery option for each set of field opening endpoints for an intensity modulated beam;
(b) for every two consecutive intensity modulated beam columns of the intensity modulated beam, determining a series of minimum cost g-matchings between the canonical delivery options of the intensity modulated beam columns;
(c) determining the tradeoff curves based on the g-matchings;
(d) extracting the segments for the intensity modulated beam based on the matched field openings of the canonical delivery options from the determined g-matchings in an increasing order of consecutive intensity modulated beam columns to thereby partition the intensity modulate beam; and
(e) implementing a treatment plan based on the extracted segments of step (d).