AN ORTHOGONAL DOUBLE-LAYER GRATING DYNAMIC INTENSITY MODULATION SEGMENTATION METHOD BASED ON QUADRANT

Abstract

The invention discloses an orthogonal dual-layer grating dynamic intensity modulation segmentation method based on quadrant, specifically include the following steps: S1: the fluence distribution under each beam is calculated through the radiation treatment planning system; S2: use orthogonal double-layer collimator for fluence segmentation; S3: divide the quadrant, divide the field surrounded by the upper, lower, left and right leaves into at least two quadrants, to obtain the fluence distribution and the corresponding leaf sequence of each quadrant; S4: perform regional planning of the fluence in each quadrant to obtain multiple different regions and determine the segmentation mode of different regions; S5: for any quadrant, use two mutually orthogonal leaf groups for segmentation. The present invention completes the dynamic segmentation of any shape target area and multiple target areas through the mutual cooperative movement of the upper and lower layers of orthogonal leaves, realizes the dynamic segmentation of the upper and lower layers of the orthogonal dual-layer collimator from two directions, avoids the end surface perspective between the leaves, and improves the segmentation efficiency.

Claims

1. An orthogonal dual-layer grating dynamic intensity modulation segmentation method based on quadrant including the following steps: S1: the fluence distribution under each radiation field is calculated by the treatment planning system, it can be expressed as a fluence F(x,y) within the radiation field range in the isocenter plane; S2: use orthogonal double-layer collimator for fluence segmentation; S3: divide the quadrant, divide the field surrounded by the upper, lower, left and right leaves into at least two quadrants, to obtain the fluence distribution and the corresponding leaf sequence of each quadrant; S4: perform regional planning of the fluence in each quadrant to obtain multiple different regions and determine the segmentation mode of different regions; S5: for any quadrant, use two mutually orthogonal leaf groups for segmentation.

2. The method according to claim 1, wherein the step S2 includes that rays are projected under the isocenter plane coordinate system S-XY, and the upper and lower leaves are located at four directions of the isocenter plane respectively.

3. The method according to claim 2, wherein, for the step S3, the maximum point or minimum point in the target area can be a segmentation center.

4. The method according to claim 2, wherein for the step S3, if there is a concave fluence distribution or an annular target area fluence distribution, the quadrant division takes the minimum point as the segmentation center; if there are multiple concave areas, the center of the concave target area is located on the intersection line of the two quadrants.

5. The method according to claim 2, wherein for the step S3, after completing the quadrant segmentation, each quadrant corresponds to a ray fluence within one region of the beam range, and corresponds to at least one pair of mutually orthogonal leaf groups.

6. The method according to claim 1, wherein the step S4 specifically comprises the following: the surroundings of any peak or trough in the fluence can be divided into four regions A1, A2, A3, A3 and A4 according to the value of partial derivatives; when ∂F/∂x<=0, and ∂F/∂y<=0, the region is region A1; when ∂F/∂x>0, and ∂F/∂y<=0, the region is region A2; when ∂F/∂x<=0, and ∂F/∂y>0, the region is region A3; when ∂F/∂x>0, and ∂F/∂y>0, the region is region A4.

7. The method according to claim 1, wherein the step S5 specifically comprises the following steps: S5.1: determine the initial position of the leaves; S5.2: establish the leaves movement trajectory function; S5.3: apply the time-delay differential equation to solve the leaf motion trajectory function.

8. The method according to claim 7, wherein the step S5.1 specifically includes the following: two orthogonal leaf groups, defining one group as an active leaf, the other group as a passive leaf, the active leaf moving from the beam edge to the beam center, and the passive leaf retreating from the beam center to the beam edge; the initial position can be determined as the active leaf at the edge of the field and the passive leaf at the quadrant junction.

9. The method according to claim 8, wherein the step S5.2 specifically comprises the following: under the premise of a given accelerator dose rate and the blade moving in one direction, a leaf trajectory is described by a corresponding relationship between leaf position and fluence size, given the accelerator dose rate and the leaves moving along one direction; the leaf end position is used as the leaf position, the relationship between the leaf position and its corresponding fluence size is a monotonic function in the absence of other leaves blocking.

10. The method according to claim 9, wherein the step S5.3 specifically comprises the following: knowing the starting position of each leaf position, the fluence function of the next moment by the fluence function of the previous position.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0039] FIG. 1 shows a three-dimensional view of the strong ray fluence of a concave target region provided by the invention embodiment.

[0040] FIG. 2(a) and FIG. 2(b) show the position distribution of orthogonal dual-layer collimator under the beam frame system provided by the invention embodiment;

[0041] FIG. 2(a) is a schematic diagram of the structure of the rays and the dual-layer collimator;

[0042] FIG. 2(b) shows the position relationship between the upper leaves, the lower leaves and the isocenter plane.

[0043] FIG. 3 is a schematic diagram of the quadrant division and the leaf sequence assignment of the fluence provided by the invention embodiment.

[0044] FIG. 4(a), FIG. 4(b), and FIG. 4(c) show a diagram of three typical quadrant segmentation modes provided by the invention embodiment;

[0045] FIG. 4(a) is a diagram of the first quadrant segmentation mode;

[0046] FIG. 4(b) is a diagram of the second quadrant segmentation mode;

[0047] FIG. 4(c) is a diagram of the third quadrant segmentation mode;

[0048] FIG. 5 is a schematic determination of the initial position of the active (passive) leaf of the first quadrant provided by the invention embodiment.

[0049] FIG. 6 is a graph showing the functional relationship between the position of the right leaves in the first quadrant and the size of the fluence value provided by the invention embodiment.

[0050] FIG. 7 is a graph showing the functional relationship between the position of the passive leaf and the magnitude of the fluence value when the front leaf in the first quadrant is used as the passive leaf according to an embodiment of the invention.

[0051] FIG. 8 is a schematic diagram of the leaf trajectory function provided by the inventive embodiment.

[0052] FIG. 9 is a schematic diagram of the area division around a peak or trough provided by the invention embodiment.

[0053] FIG. 10 is a schematic diagram of the area division within the first quadrant provided by the invention embodiment.

[0054] FIG. 11 is a 3D view of the fluence distribution in the first quadrant provided by the invention embodiment.

[0055] FIG. 12 is a schematic diagram of the initial position in the first quadrant after the region division provided by the invention embodiment.

[0056] FIG. 13(a) and FIG. 13(b) show a motion track curve of an active leaf or a passive leaf in the first quadrant provided by the invention embodiment;

[0057] FIG. 13(a) is the motion trajectory diagram of the right leaves;

[0058] FIG. 13(b) is the motion trajectory diagram of the front leaves.

[0059] FIG. 14 is a fluence three-dimensional view segmented by the leaves trajectory provided by the invention embodiment.

[0060] Among them: 1—peak or trough; 2—valley center.

DETAILED DESCRIPTION OF THE EMBODIMENTS

[0061] The preferred embodiments of the present invention are described in detail below in connection with the accompanying drawings.

[0062] To achieve the object of the present invention, in some embodiments of an orthogonal dual-layer grating dynamic intensity modulation segmentation method based on quadrant, an orthogonal dual-layer grating dynamic intensity modulation segmentation method based on quadrant, comprising the following steps of:

[0063] S1: the fluence distribution under each radiation field is calculated by the treatment planning system, it can be expressed as a fluence F(x,y) within the radiation field range in the isocenter plane;

[0064] S2: use orthogonal double-layer collimator for fluence segmentation;

[0065] S3: divide the quadrant, divide the field surrounded by the upper, lower, left and right leaves into at least two quadrants, to obtain the fluence distribution and the corresponding leaf sequence of each quadrant;

[0066] S4: perform regional planning of the fluence in each quadrant to obtain multiple different regions and determine the segmentation mode of different regions;

[0067] S5: for any quadrant, use two mutually orthogonal leaf groups for segmentation.

[0068] In step S3, as shown in FIG. 3, each quadrant corresponds to two different sets of leaf sequences, and the adjacent quadrant combination may also form a separate quadrant that may correspond to three different leaves; the quadrant division shall try to ensure the total time for each quadrant.

[0069] In order to further optimize the implementation effect of the present invention, in other embodiments, the other features are the same, except that the step S2 includes that rays are projected under the isocenter plane coordinate system S-XY, and the upper and lower leaves are located at four directions of the isocenter plane respectively.

[0070] In order to further optimize the implementation effect of the present invention, in other embodiments, the other feature techniques are the same, except in that for the step S3, the maximum point or minimum point in the target area can be a segmentation center.

[0071] In order to further optimize the implementation effect of the present invention, in other embodiments, the other feature techniques are the same, as shown in FIG. 4, for step S3, if there is a concave fluence distribution or an annular target area fluence distribution, the quadrant division takes the minimum point as the segmentation center;

[0072] if there are multiple concave areas, the center of the concave target area is located on the intersection line of the two quadrants.

[0073] Specifically, the “◯” in FIG. 4 represents the valley center 2.

[0074] In order to further optimize the implementation effect of the present invention, in other embodiments, the remaining features are the same, except in that, for step S3, after completing the quadrant segmentation, each quadrant corresponds to a ray fluence within one region of the beam range, and corresponds to at least one pair of mutually orthogonal leaf groups.

[0075] In order to further optimize the implementation effect of the present invention, in other embodiments, the other features are the same, in that step S4 specifically includes the following: as shown in FIG. 9, the surroundings of any peak or trough 1 in the fluence can be divided into four regions A1, A2, A3, A3 and A4 according to the value of partial derivatives;

[0076] when ∂F/∂x<=0, and ∂F/∂y<=0, the region is region A1;

[0077] when ∂F/∂x>0, and ∂F/∂y<=0, the region is region A2;

[0078] when ∂F/∂x<=0, and ∂F/∂y>0, the region is region A3;

[0079] when ∂F/∂x>0, and ∂F/∂y>0, the region is region A4.

[0080] In this embodiment, in order to improve the segmentation efficiency, different regions are adopted for different regions in a quadrant. As shown in FIG. 10, in the first quadrant, regions A1 and A3 can be independently divided in right leaves, anterior leaves and A2 can be used for independent segmentation in region A1, and two groups of leaves are required for orthogonal coordination segmentation in region A4.

[0081] To further optimize the implementation effect of the present invention, in other embodiments, the remaining characteristic techniques are the same, except that step S5 specifically includes the following steps:

[0082] S5.1: determine the initial position of the leaves;

[0083] S5.2: establish the leaves movement trajectory function;

[0084] S5.3: apply the time-delay differential equation to solve the leaf motion trajectory function.

[0085] Further, step S5.1 includes the following: two orthogonal leaf groups, defining one group as an active leaf, the other group as a passive leaf, the active leaf moving from the beam edge to the beam center, and the passive leaf retreating from the beam center to the beam edge;

[0086] the initial position can be determined as the active leaf at the edge of the field and the passive leaf at the quadrant junction.

[0087] It is worth noting that it is necessary to avoid that a pair of oppositely arranged leaves are the passive leaf when defining the passive leaf. As shown in FIG. 5, it is an initial position situation of the first quadrant leaves, the right leaves is acted as a set of active leaf and the left leaves is acted as a set of passive leaf.

[0088] Further, step S5.2 includes the following: under the premise of a given accelerator dose rate and the blade moving in one direction, a leaf trajectory is described by a corresponding relationship between leaf position and fluence size, given the accelerator dose rate and the leaves moving along one direction;

[0089] the leaf end position is used as the leaf position, the relationship between the leaf position and its corresponding fluence size is a monotonic function in the absence of other leaves blocking. In this embodiment, given the accelerator dose rate and the leaves moving in one direction, since the fluence value is proportional to the time, a leaf trajectory is described by a corresponding relationship between leaf position and fluence size, given the accelerator dose rate and the leaves moving along one direction. Taking the leaf end position as the leaf position, it can be found that in the absence of other blades blocking, the relationship between the leaf position and its corresponding fluence size is always a monotonic function. As shown in FIG. 6, taking any right leaves of the first quadrant as an example, the relationship between its position and its corresponding fluence is shown in the figure, denoted as g2 (x, yi), where x is the corresponding end position of the right leaves at a certain time, and yi is the corresponding ordinate of the right leaves under the coordinate system S. Similarly, the relationship between the passive leaf position and its corresponding fluence size is also a monotonic function. As shown in FIG. 7, taking the front leaves in the first quadrant as an example, when the front leaves is used as passive leaf, the front leaves retreat from the initial position 0 to the positive direction of the Y axis under the coordinate system S, and the independent shielding amount of the front blade at each position is h1 (xi,y)=T−g1(xi,y), where T is the total exposure, g1(xi,y) is the fluence corresponding to the position of the front leaves, y is the end face position corresponding to the front leaves at a certain time, and xi is the abscissa corresponding to the front leaves under the coordinate system S.

[0090] Further, step S5.3 includes the following: knowing the starting position of each leaf position, the fluence function of the next moment by the fluence function of the previous position.

[0091] As shown in FIG. 8, taking the right leaves (active leaf) at any point in the first quadrant as an example, the time-lag differential equation is established for the fluence function g.sub.2(x, y) Solution, is as follows:

[00001]$\{\begin{array}{c}{g}_2\left(x,y\right)=\max \left(g_{21},g_{22},g_{23}\right)\\ g_{21}\left(x,y\right)=g_2\left(x+\Delta x,y\right)+k_{\min }.Math.\Delta x\\ g_{22}\left(x,y\right)=g_2\left(x+\Delta x,y\right)-\frac{\partial F}{\partial x}.Math.\Delta x\\ g_{23}\left(x,y\right)=g_2\left(x,y-\Delta y\right)+\frac{\partial F}{\partial y}.Math.\Delta y+k_{\min }.Math.\Delta y\end{array}$

[0092] Specifically, Δx and Δy are the amount of change of the leaf position at two adjacent moments under the coordinate system S, ∂F/∂x and ∂F/∂y are the partial derivative of the fluence along x and y, and k.sub.min is the maximum speed limit of leaf corresponding to the minimum fluence incremental efficiency, g.sub.21(x,y), g.sub.22(x,y) and g.sub.23(x,y) are the fluence function values obtained by searching along the x-axis direction, the derivative along the flux surface direction, and the y-axis direction, respectively.

[0093] The above multiple embodiments can be realized in cross-over and parallel ways.

[0094] The present invention completes the dynamic segmentation of any shape target area (concave target area, annular target area, etc.) and multiple target areas through the mutual cooperative movement of the upper and lower layers of orthogonal leaves, realizes the dynamic segmentation of the upper and lower layers of the orthogonal dual-layer collimator from two directions, avoids the end surface perspective between the leaves, and improves the segmentation efficiency.

[0095] To illustrate the specific embodiment process of the present invention, an annular target area of a multi-connected region is described.

[0096] The specific process is as follows:

[0097] S1: the radiation fluence of a circular target area with a multiconnected area imported from the treatment planning system can be expressed as a fluence F (x, y) within the isocenter plane, as shown in FIG. 1;

[0098] S2: use orthogonal dual-layer collimator for fluence segmentation, as shown in FIG. 2(a), ray through the projection of the upper and lower two collimators in the isocenter plane coordinate system S-XY, and the upper and lower leaves are located at four directions of the isocenter plane respectively, as shown in FIG. 2(b), the upper leaves are located at the front and rear ends of the isocenter plane, the lower leaves are located at the left and right ends of the isocenter plane;

[0099] S3: divide the quadrant to obtain the fluence distribution of the four quadrants and the corresponding leaf sequence. FIG. 11 is the fluence distribution of the first quadrant;

[0100] S4: perform regional planning of the fluence in each quadrant, obtain all regions A1, A2, A3, A4, and determine the same time. As shown in FIG. 10, taking the first quadrant as an example, the region A1 in the first quadrant is independently divided by the right leaves, the region A3 in the first quadrant is independently divided by the right leaves; the region A2 in the first quadrant is independently divided by the front leaves. And the region A4 in the first quadrant is divided by orthogonal mating of all right leaves and front leaves S5.1: after the regions are divided, determine the initial position of the active leaf and the passive leaf. In FIG. 12, the initial position of all the leaves in the first quadrant is shown. In the overlapping region A4, the front leaves in the region A4 is set as the passive leaf, and the initial positions of all the passive leaf in the region A4 are located at the junction of the quadrants, and the retraction motion is performed.

[0101] S5.2: establish the leaf trajectory function, as shown in FIG. 13, the trajectory of two groups of orthogonal leaves in the first quadrant overlapping area. FIG. 13(a) shows the movement of the right leaves and FIG. 13(b) shows the movement of the front leaves.

[0102] S5.3: the segmented fluence F′(x, y) is calculated by the leaf trajectory without considering the influence of penumbra, and its 3D distribution is shown in FIG. 14. The relative errors are compared using the vector two norm and the initial fluence diagram as follows:

TABLE-US-00001 segNum 50 100 250 500 1000 10000 Σ||F′ − 0.0416 0.0206 0.0084 0.0042 0.0021 2.05E−04 F||.sub.2/ΣF

[0103] Wherein, segNum refers to the equal fraction of the fluence segmentation, which is inversely proportional to the segmentation step. The larger segNum is, the finer the segmentation of the fluence is.

[0104] The beneficial achievements of the present invention compared to the prior art are:

[0105] 1. improve conformal degree: it can realize the dynamic segmentation of various complex target areas, such as concave target area and annular target area;

[0106] 2. improve the segmentation accuracy: the orthogonal leaf pairs are used to cooperate with the movement, through the spatial interleaving, to avoid the leaves closed position collision, to solve the problem of the end surface perspective between the lesf pairs;

[0107] 3. realize the intensity segmentation of multi-target areas: four groups of leaves with different directions can be divided into at least two, up to four quadrants, and the multi-target area problems within four can be divided simultaneously;

[0108] 4. Segmentation efficiency is high: through the direction derivative characteristics of the intensity distribution surface, a peak or trough area is divided into four parts, and different leaf segmentation methods are adopted to improve the segmentation efficiency;

[0109] 5. It can track the dynamic target area from two directions: a pair of orthogonal leaves is used to segment the target area, and the dynamic target area can be tracked by moving in two directions.

[0110] For the preferred embodiment of the invention, it should be noted that for ordinary persons skilled in the art, several deformation and improvements may be made without departing from the creative idea of the invention, which fall within the protection scope of the invention.