CN107961447B

CN107961447B - Method and device for obtaining radiotherapy plan

Info

Publication number: CN107961447B
Application number: CN201610915912.0A
Authority: CN
Inventors: 姚鹏
Original assignee: Neusoft Medical Systems Co Ltd
Current assignee: Neusoft Medical Systems Co Ltd
Priority date: 2016-10-20
Filing date: 2016-10-20
Publication date: 2020-03-24
Anticipated expiration: 2036-10-20
Also published as: CN107961447A

Abstract

The application provides a method and a device for obtaining a radiotherapy plan, wherein the method comprises the following steps: obtaining a leaf flux map according to dose constraints; dividing the leaf flux map into segments; and obtaining a radiotherapy plan by using the sub-fields. Through the technical scheme of this application, can obtain the radiotherapy plan based on the leaf flux map, rather than obtaining the radiotherapy plan based on pencil beam flux map, like this, even MLC includes the leaf of different width, when using above-mentioned radiotherapy plan, can effectively use MLC to shelter from the ray of shining to normal organ. The method for obtaining the radiotherapy plan can reduce the burden of doctors, the radiotherapy plan is more rational and reliable, the treatment quality is greatly improved, the tumor dose can be greatly increased, the dose of normal organs is reduced, the tumor control rate is improved, and the incidence rate of complications of the normal organs is reduced. Moreover, the method can ensure that the leaf flux map and the MLC field are completely corresponding in size, improve the controllability of the algorithm and reduce errors.

Description

Method and device for obtaining radiotherapy plan

Technical Field

The application relates to the technical field of medical treatment, in particular to a method and a device for obtaining a radiotherapy plan.

Background

Tumors are generally more vital than normal organs, and if the radiation can cause damage to the tumor, the damage to the normal organs is greater, so that when a radiotherapy plan is made, the tumor needs to receive as much dose as possible, and at the same time, the normal organs receive as little dose as possible. Currently, this can be achieved by IMRT (Intensity-modulated radiation therapy), which may also be referred to as inverse Intensity modulated radiation therapy or conformal Intensity modulated radiation therapy) algorithms.

In the IMRT algorithm, the radiation to the normal organ is usually blocked by using an MLC (Multi Leaf Collimator, also called Multi-Leaf grating, Multi-Leaf diaphragm, etc.) so that the tumor receives as much dose as possible while the normal organ receives as little dose as possible. Wherein the MLC may comprise a large number of leaves through which radiation that is incident on the normal organ is blocked.

The leaves are basic units forming the MLC, are made of heavy metal materials and are in a strip shape, the length of each leaf is determined by the maximum field to be formed, and the width of each leaf is several millimeters to several centimeters. If the MLC includes leaves with different widths, such as 6 mm width leaves and 3 mm width leaves, the MLC cannot be effectively used to block the rays irradiated to the normal organs when a radiotherapy plan is made by using the conventional IMRT algorithm.

Disclosure of Invention

The application provides a method for obtaining a radiotherapy plan, which comprises the following steps:

obtaining a leaf flux map according to dose constraints;

dividing the leaf flux map into segments;

and obtaining a radiotherapy plan by using the sub-fields.

The present application provides an apparatus for obtaining a radiotherapy plan, the apparatus comprising:

a first obtaining module for obtaining a leaf flux map according to dose constraints;

the segmentation module is used for segmenting the leaf flux map into sub fields;

and the second obtaining module is used for obtaining the radiotherapy plan by utilizing the sub-fields.

Based on the technical scheme, in the embodiment of the application, the radiotherapy plan can be obtained based on the leaf flux map instead of the pencil beam flux map, so that even if the MLC comprises the leaves with different widths, when the radiotherapy plan is used, the MLC can be effectively used for shielding rays irradiated to a normal organ. The method for obtaining the radiotherapy plan can reduce the burden of doctors, the radiotherapy plan is more rational and reliable, the treatment quality is greatly improved, the tumor dose can be greatly increased, the dose of normal organs is reduced, the tumor control rate is improved, and the incidence rate of complications of the normal organs is reduced. Moreover, the method can ensure that the leaf flux map and the MLC field are completely corresponding in size, improve the controllability of the algorithm and reduce errors.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments of the present application or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present application, and other drawings can be obtained by those skilled in the art according to the drawings.

Figure 1 is a flow chart of a method of obtaining a radiotherapy plan in one embodiment of the present application;

FIG. 2 is a schematic illustration of detection points in one embodiment of the present application;

FIG. 3 is a schematic view of a leaf of an MLC in an embodiment of the present application;

figure 4 is a flow chart of a method of obtaining a radiotherapy plan in another embodiment of the present application;

figure 5 is a flow chart of a method of obtaining a radiotherapy plan in another embodiment of the present application;

figure 6 is a flow chart of a method of obtaining a radiotherapy plan in another embodiment of the present application;

FIG. 7 is a hardware block diagram of a medical device in one embodiment of the present application;

fig. 8 is a block diagram of an apparatus for acquiring a radiotherapy plan according to an embodiment of the present application.

Detailed Description

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this application and the claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein is meant to encompass any and all possible combinations of one or more of the associated listed items.

It is to be understood that although the terms first, second, third, etc. may be used herein to describe various information, such information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present application. Depending on the context, moreover, the word "if" as used may be interpreted as "at … …" or "when … …" or "in response to a determination".

The embodiment of the present application provides a method for obtaining a radiotherapy plan, which may be applied to a medical device, where the medical device may obtain a radiotherapy plan by using an IMRT algorithm, output the radiotherapy plan to an accelerator of the medical device, and treat a sample (such as a patient) with the radiotherapy plan by the accelerator.

In one example, in order to obtain a radiotherapy plan satisfying a doctor, an IMRT algorithm may be used to obtain the radiotherapy plan such that a dose distribution in the radiotherapy plan is as consistent as possible with a dose distribution desired by the doctor, so that after the radiotherapy plan is output to an accelerator of a medical apparatus, when the accelerator treats a subject using the radiotherapy plan, an accurate radiation dose may be delivered to a target region (e.g., a tumor region, etc.) of the subject, so that the tumor receives as much dose as possible while a normal organ receives as little dose as possible.

In one example, referring to fig. 1, a flow chart for a medical device employing an IMRT algorithm to obtain a radiotherapy plan may be applied to the medical device, and the method may include the steps of:

pencil beam (also referred to as pencil beam kernel) weights are obtained from the dose constraints, step 101.

In one example, a pencil beam weight of 1 may be initially configured, the actual dose at the detection point is obtained based on the pencil beam weight of 1, and the dose constraint of 1 is obtained based on the actual dose at the detection point. The pencil beam weight 1 is adjusted by a preset algorithm (such as conjugate gradient method, simulated annealing, genetic algorithm, linear programming, etc.) to obtain an adjusted pencil beam weight 2. The actual dose at the checkpoints is obtained from the pencil beam weights 2 and the dose constraints 2 are obtained from the actual dose at the checkpoints. If the dose constraint 2 is smaller than the dose constraint 1, the pencil beam weight 2 is adjusted by a preset algorithm to obtain an adjusted pencil beam weight 3. If the dose constraint 2 is not smaller than the dose constraint 1, the pencil beam weight 1 is adjusted by using a preset algorithm to obtain an adjusted pencil beam weight 3. The actual dose at the checkpoints is obtained from the pencil beam weights 3 and the dose constraints 3 are obtained from the actual dose at the checkpoints. And so on, until the dose constraint satisfies the constraint, the last pencil beam weight (e.g., pencil beam weight 9 corresponding to last dose constraint 9 when dose constraint 10 satisfies the constraint) is taken as the result of step 101, or when the number of iterations reaches a preset threshold, the current pencil beam weight (e.g., pencil beam weight 10 corresponding to currently obtained dose constraint 10) is taken as the result of step 101.

In the above process, the detection point is a detection point in the target region or the region where the key organ is located. One or more detection points may be divided for one target region of a subject, or one or more detection points may be divided for a region where a key organ is located. As shown in fig. 2, in order to divide a target area into 4 detection points, a region where a key organ is located is divided into 2 detection points. In practical application, the number of the detection points can be more, and the number is not limited in the application. In fig. 2, the shape of the detecting point is a rectangle as an example, in practical application, the shape of the detecting point may be any shape, such as a circle, a diamond, a trapezoid, and other arbitrary irregular shapes, and the shape is not limited in this application.

In the above process, for the process of "adjusting pencil beam weight by using a preset algorithm", different preset algorithms have different adjustment modes, and the adjustment process of pencil beam weight is described below by combining a conjugate gradient method and simulated annealing, and for other algorithms, the adjustment process is not described again.

When the pencil beam weight is adjusted by the conjugate gradient method, the pencil beam weight may be adjusted by an adjustment step, for example, the adjusted pencil beam weight is the pencil beam weight before adjustment + the adjustment step α_kAdjusting parameter p_kWherein the step size α is adjusted_kAnd adjusting the parameter p_kMay be configured based on practical experience and in one example, step size α is adjusted_kAnd adjusting the parameter p_kMay be related to pencil beam weights before adjustment.

For example, the adjustment parameter p is obtained by the following formula_k：

In this formula, H_kThe matrix is a pre-configured matrix, and is not limited thereto. x is the number of_kTo adjust the pencil beam weights before. f is a preconfigured function, which is not limited. f (x)_k) Is directed to x_kFunction value of, i.e. x_kAs a variable of the function f, a function value f (x) is obtained_k)。

Representing a vector function, each component being a partial derivative of the function f.

For example, the adjustment step α is obtained by the following formula_k：α_k＝arg min f(x_k+α*p_k). In this formula, argmin f represents a variable value when the function f is minimized, and f is a function arranged in advance, and the adjustment parameter p can be obtained as described above_kThe function f used is different. x is the number of_kFor pencil weight before adjustment α is a pre-configured value, which is an empirical value, p_kFor the above-mentioned adjustment parameter p_k。

When adjusting the pencil beam weight using the simulated annealing algorithm, the pencil beam weight may be adjusted using a random number, for example, the adjusted pencil beam weight is the pencil beam weight before adjustment + the random number. The random number is generated based on a random distribution function, and the random number generated by the random distribution function may be different each time the pencil beam weight is adjusted, but the random number generated by the random distribution function may be located in an interval.

For the process of "obtaining the actual dose of the detected spot according to the pencil beam weight", the actual dose D of the detected spot can be obtained by using equation 1_i. In practical applications, the actual dosage at the detection point can be obtained by other methods, and the obtaining method is not limited. Wherein, in formula 1, D_iIs the actual dose at the i-th spot, N₁Indicating the number of pencil beams, N₂Indicating the number of detection points, B_ikIs the dose contribution of the kth pencil beam to the ith detection spot, w_kIs the kth pencil beam weight (i.e., dose). As can be seen from equation 1, the actual dose D at each checkpoint_iIt may be the sum of the dose contributions of all pencil beams to the detection spot.

In equation 1, in order to obtain the actual dose D_iNeed to know the pencil powerWeight w_kAnd dose contribution B_ik. For pencil beam weight w_kThe pencil beam weight w_kIs a continuously adjusted number, each time a new pencil weight w is obtained_kThe pencil beam weight w obtained at present can then be used_kSubstituting into formula 1 to obtain new actual dosage D_i. Thus, for equation 1, pencil beam weight w_kMay be a known value.

For "learned dose contribution B_ik"the pencil beams are sized first, and after the pencil beams are sized, the dose contribution of each pencil beam to the i-th detection point can be obtained. Wherein, the irradiation field of the medical device (such as an accelerator) can be divided into "squares" according to the same size, each "square" represents a pencil beam, and the size of the square is the size of the pencil beam. The dimensions of the pencil beam may be configured according to practical experience, for example: 3 mm by 3 mm, 5 mm by 5 mm, etc. In the subsequent process of the embodiment of the present application, the former number represents the length of the size, and the latter number represents the width of the size.

For the process of "obtaining the dose contribution of each pencil beam to the i-th detection point", the farther the pencil beam is from the detection point when each pencil beam irradiates the detection point, the smaller the dose contribution of the pencil beam to the detection point, and the closer the pencil beam is to the detection point, the larger the dose contribution of the pencil beam to the detection point. Furthermore, a larger size of a pencil beam indicates a larger dose of the pencil beam, a larger dose contribution of the pencil beam to the detection spot, and a smaller size of a pencil beam indicates a smaller dose of the pencil beam, a smaller dose contribution of the pencil beam to the detection spot. Based on the above principle, the dose contribution of each pencil beam to the ith detection point can be obtained, and the detailed algorithm is not described herein again.

In summary, the dose contribution B of the kth pencil beam to the ith detection spot is obtained_ikAnd kth pencil beam weight w_kLater, since the values of N1 and N2 are known, the actual dose D can be obtained_i。

For the process of "obtaining dose constraints from actual doses at the detection points", the dose constraints can be obtained using equation 2. In practical applications, the dose constraints may also be obtained in other ways, which are not limited. In formula 2, N_TIs the number of spots detected, N, in the target area (e.g., tumor area, etc.)_OARIs the number of detection points in the region where the key organ is located (such as the tumor peripheral organ). D_iIs the actual dose at the ith checkpoint,

is the target dose for the target area,

is the target dose for the region where the critical organ is located.

In equation 2, the target dose of the target region

And target dose to the region of the critical organ

Can be configured according to practical experience. For example, a doctor may configure the target dose of the target region in the body to be examined and the target dose of the region where the key organ is located according to the actual condition of the body to be examined, and the configuration process of the target dose is not described in detail. Wherein, the target dose of the target area is larger than that of the area where the key organ is located.

In formula 1 and formula 2, D_iIs the actual dose of the ith detection point, which is the detection point in the target area or the area where the key organ is located. In equation 1, the spots in all regions are labeled uniformly, so that the actual doses at 4 spots in the target region can be respectively denoted as D as shown in fig. 2₁，D₂，D₃，D₄And the actual doses of the 2 detection points in the region of the key organ are respectively recorded as D₅，D₆. In the formula 2, the index is given to the detected spots of each region individually, and therefore, for the first functional expression of the formula 2, which is a functional expression for the target region, the actual doses at the 4 detected spots in the target region can be respectively expressed as D₁，D₂，D₃，D₄The actual dose at these 4 detection points is the actual dose D given in equation 1₁，D₂，D₃，D₄. For the second function of equation 2, which is a function for the region of the critical organ, the actual doses at 2 detection points in the region of the critical organ can be respectively recorded as D₁，D₂The actual dose at these 2 detection points is the actual dose D given in equation 1₅，D₆。

In equation 2, the sum of the functions of only two regions (one target region and one region where a key organ is located) is shown, and in practical applications, the dose constraint f is the sum of the functions of all regions, i.e. the function containing multiple target regions and the function containing multiple regions where multiple key organs are located. The function of each target region is shown in the first function of formula 2, and the function of the region where each key organ is located is shown in the second function of formula 2. Since each spot is located at a target region or a region where a critical organ is located, D is used to obtain a function of each target region_iIs the actual dose at the i < th > spot in the target volume. In obtaining a function of the region in which each key organ is located, D_iIs the actual dose at the ith checkpoint in the region of the critical organ.

In equation 2, the actual dose if a spot within the target volume is closer to the target dose for that target volume

The smaller the value of the first function. In the formula 2, the first and second groups,

based on this, e.g.The actual dosage of the check point in the region of the fruit key organ is not more than

The smaller the value of the second function. Thus, if the target volume is close to the target dose

The larger the number of detection points, and the critical organ is in the area not larger than the target dose

The larger the number of spots detected, the smaller the dose constraint f.

In one example, the optimization objective is that the dose constraint f be minimal, thus, if the dose constraint f is 0, it is stated that all checkpoints within the target volume and all checkpoints within the region where the critical organ is located satisfy the dose constraint. To minimize the dose constraint f, the actual dose D at each spot in the target volume is determined_iCan be administered to the target area

The same, and for each detection point in the region of the critical organ, the actual dose D of the detection point_iMay be no greater than the target dose for the region in which the critical organ is located

In summary, in terms of the process of "adjusting the pencil beam weight by using the preset algorithm, obtaining the actual dose of the detection point according to the adjusted pencil beam weight, and obtaining the dose constraint according to the actual dose of the detection point", the final optimization objective is: the dose constraint f is minimal. To achieve this optimization goal, the pencil beam weights are adjusted continuously so that each spot in the target volume has its actual dose D_iCan be administered to the target area

Thus, the above "dose constraints satisfy the constraints" may mean: the dose constraint is minimal, e.g., 0. Of course, the dose constraint will generally not be 0 during the course of constantly adjusting pencil beam weights. Based on this, if the current dose constraint is not less than the last dose constraint and the current dose constraint is already less than the preset threshold (which may be a value close to 0), the dose constraint may be considered to satisfy the constraint condition, and the pencil beam weight corresponding to the last dose constraint is taken as the result of step 101.

In one example, after each pencil beam weight is derived, as w₁、w₂、…w_N1Etc. these pencil beam weights can be combined together to form a pencil beam flux map (also called a flux matrix), which is a matrix of these pencil beam weights. Furthermore, the size of the pencil beam represents the degree of dispersion of the pencil beam flux map, the smaller the size of the pencil beam, the higher the degree of dispersion of the pencil beam flux map, the more accurate the calculation, the larger the calculation, and the slower the speed. Conversely, the larger the size of the pencil beam, the lower the degree of dispersion of the pencil beam flux map. For example, assuming that the dimensions of the pencil beams are 5 mm by 5 mm, for a field of 10 cm by 10 cm, 20 x 20 pencil beams can be separated, and the pencil beam flux map corresponding to the 20 x 20 pencil beams is also a 20 x 20 matrix.

Step 102, the pencil beam flux map formed by the pencil beam weights is divided into sub-fields (also called sub-field matrix), and the sub-fields are the shapes of the radiation fields to be formed by the MLC, namely the radiotherapy plan finally obtained.

As shown below, to illustrate an example of the segmentation of the pencil beam flux map into subfields, to the left of the equal sign is the pencil beam flux map formed by the k pencil beam weights, and to the right of the equal sign is the segmented subfields. Of course, the following is only an example of segmenting the sub-fields, and other segmentation methods may also be adopted, which are not limited to this segmentation method. In addition, only one pencil beam flux map of 4 × 4 is given below, and the pencil beam flux map in practical application is more complicated, but the processing is similar to that of the pencil beam flux map of 4 × 4, and the description is omitted later.

In one example, for the process of "splitting the pencil beam flux map formed by pencil beam weights into subfields", the following principle may be followed: 1. the values in each sub-field divided are 0 and 1, and no other values are included. 2. The regions formed by 1 in each subfield are to be as contiguous as possible (to ensure dose accuracy). 3. The pencil beam flux map is segmented into equal sized subfields. 4. The number of segments is as small as possible (the radiation duration can be reduced). 5. The sum of the factors (i.e., the hop counts "2, 1", etc.) of all sub-fields is as small as possible (the radiation duration and the wear of the medical device can be reduced). 6. The number of 1 s in each subfield is as large as possible (dose accuracy can be guaranteed). Of course, the above procedures are only a few examples of the following principles, and in practical applications, there may be other following principles, which are not limited thereto.

In the divided sub-fields, the number of divided sub-fields represents the number of sub-fields in a certain direction, 1 in each sub-field indicates that the sub-field is not shielded by a leaf of an MLC (representing that a target area is irradiated with radiation), and 0 in each sub-field indicates that the sub-field is shielded by a leaf of an MLC (representing that a target area is not irradiated with radiation). The factor (e.g., 2, 1, etc.) of each subfield is the hop count of the medical device, which represents the time that is not or is not occluded by the leaf, e.g., hop count 1 means that the time is 1 millisecond that is not or is occluded by the leaf, hop count 2 means that the time is 2 milliseconds that is not or is occluded by the leaf, etc.

In the above-described divided sub-fields, each number (1 or 0) corresponds to the size of the pencil beam, and assuming that the size of the pencil beam is 5 mm × 5 mm, each number in the sub-field represents a size of 5 mm × 5 mm. Thus, for 0110 in the first row of the first subfield, 1 position of 0 needs to be shielded by the left blade, that is, the shielding area of the blade is 5 mm × 5 mm, and 1 position of 0 needs to be shielded by the right blade, that is, the shielding area of the blade is 5 mm × 5 mm. For 0010 in the second row of the first subfield, 2 positions of 0 need to be shielded by the left leaf, that is, the shielding area of the leaf is 10 mm × 5 mm, and 1 position of 0 need to be shielded by the right leaf, that is, the shielding area of the leaf is 5 mm × 5 mm. For 1111 of the third row of the second subfield, no occlusion is needed. For 0000 in the fourth row of the second subfield, 4 positions of 0 are shielded by one blade on the left or right, that is, the shielding area of the blade is 20 mm × 5 mm, and the shielding at other positions is similar, which is not described herein again.

To achieve the above function, see fig. 3, which is a schematic view of leaves of an MLC, two leaves are typically deployed in one plane of the MLC, one leaf on the left and one leaf on the right. The MLC leaves are long, and the length of the leaf is determined by the maximum field, for example, the field of 4 x 4, the longest leaf needs to block 4 pencil beams, and the length of 1 pencil beam is 5 mm, so the length of the leaf can be greater than 20 mm. Moreover, the blade is movable in the length direction, if 1 pencil beam needs to be shielded, the blade moves by 5 mm, so that 1 pencil beam is shielded exactly in the length direction; if 2 pencil beams need to be blocked, the blade moves by 10 mm, so that exactly 2 pencil beams are blocked in the length direction; and so on. The width of the leaf may be the same as the width of the pencil beam, e.g. 5 mm, if the width of the leaf is 5 mm, the leaf will block exactly 1 pencil beam in the width direction.

In summary, since the length and width of the leaves of the MLC are both determined (depending on the type of MLC), the length and width of the leaves need to be referenced when selecting the dimensions of the pencil beam. The length of the pencil beams can be any value as the length of the blade is long and can move in the length direction, and the blade can shield any number of pencil beams in the length direction. Therefore, it is sufficient that the width of the pencil beam is the same as the width of the blade. Since the pencil beam is square, the length of the pencil beam may be the same as the width of the pencil beam, both being the width of the leaf. In this way, the size of the pencil beam can be selected.

In one example, if all leaves in the MLC are the same width, e.g. all leaves are 5 mm wide, the size of the pencil beam may be 5 mm by 5 mm. However, if the width of all leaves in the MLC is not the same, as if the MLC includes 5 mm wide leaves and 1 cm wide leaves, how is the size of the pencil beam selected? If the pencil beam size is 5 mm by 5 mm, a 1 cm wide blade will block 2 pencil beams simultaneously in the width direction, while in practice the blade only needs to block 1 pencil beam, resulting in a blocking error. For example, for 0011 in the third row of the first subfield, 2 positions of 0 are required to be shielded by the left blade, that is, the shielding area of the blade is 10 mm × 5 mm, but since the blade simultaneously shields 2 pencil beams in the width direction, the shielding area of the blade is 10 mm × 10 mm, that is, the positions of the third row 2 of 0 and the positions of the fourth row 2 of 1 are simultaneously shielded, and the positions of the fourth row 2 of 1 are not actually required to be shielded. If the size of the pencil beam is 1 cm × 1 cm, the 5 mm wide blade cannot shield 1 pencil beam in the width direction, and actually, the blade can only shield half pencil beams, which obviously cannot meet the shielding requirement, and then shielding errors are caused.

In view of the above findings, the present application provides a way to size pencil beams and solves the above problems by using a leaf flux map to segment the subfields without using a pencil beam flux map.

In the embodiment of the present application, the following strategy may be adopted to determine the width of the pencil beam. If the MLC includes leaves of multiple widths (which may generally be referred to as the projected width at the isocenter), each width of a leaf divided by the width of the pencil beam is an integer. Also, it may be determined that the length of the pencil beam is the same as the width of the pencil beam.

According to the above strategy, if the MLC comprises 5 mm wide leaves and 1 cm wide leaves, then 5 mm can be determined as the width of the pencil beam, and therefore the size of the pencil beam is 5 mm x 5 mm. A 5 mm wide leaf may be represented using 1 pencil beam and a 1 cm wide leaf may be represented using 2 pencil beams. For another example, if the MLC includes leaves of 6 mm width, leaves of 4.5 mm width, and leaves of 3 mm width, then 1.5 mm may be determined as the width of the pencil beam, and thus the size of the pencil beam is 1.5 mm x 1.5 mm. A leaf blade of 3 mm width may be represented using 2 pencil beams, a leaf blade of 4.5 mm width using 3 pencil beams, and a leaf blade of 6 mm width using 4 pencil beams.

It is to be noted that, in determining the width of the pencil beam, on the basis of satisfying that "each width of the leaves divided by the width of the pencil beam is an integer", the width of the pencil beam may also be made as large as possible to reduce the amount of calculation. Thus, if the MLC includes 5 mm wide leaves and 1 cm wide leaves, then 5 mm may be determined as the width of the pencil beam, rather than 2.5 mm or 1 mm as the width of the pencil beam.

In practical applications, if the width of the blade is 5.9 mm, 4.6 mm, 2.9 mm, respectively, 1.5 mm may also be determined as the width of the pencil beam, and therefore, the size of the pencil beam is 1.5 mm, 2 pencil beams may be used to approximately represent a blade with a width of 2.9 mm, 3 pencil beams may be used to approximately represent a blade with a width of 4.6 mm, and 4 pencil beams may be used to approximately represent a blade with a width of 5.9 mm.

It should be noted that, in determining the width of the pencil beam, if the width of the leaves is 5.9 mm, 4.6 mm, 2.9 mm, etc., if the above-mentioned policy "each width of the leaves divided by the width of the pencil beam is an integer" is satisfied, the width of the pencil beam should be 0.1 mm, but if the width of the pencil beam is 0.1 mm, the amount of calculation is very large, and 0.1 mm is not suitable as the width of the pencil beam. At this point 1.5 mm may be chosen as the width of the pencil beam, although a division of 5.9 mm, 4.6 mm, 2.9 mm etc. by 1.5 mm is not an integer, but the result of a division of 5.9 mm, 4.6 mm, 2.9 mm etc. by 1.5 mm is already as close to an integer as possible.

After selecting the size of the pencil beam in the above manner, the following problems still remain: "if the size of the pencil-beam kernel is 5 mm by 5 mm, a 1 cm wide blade will block 2 pencil beams simultaneously in the width direction, while in practice the blade only needs to block 1 pencil beam, resulting in a blocking error". Based on this, the sub-field can be segmented using the leaf flux map instead of the pencil beam flux map, and therefore, the method for obtaining the radiotherapy plan proposed in the embodiment of the present application may include the following steps:

step 1, obtaining a leaf flux map according to dose constraints;

step 2, dividing the leaf flux map into segments;

and 3, obtaining a radiotherapy plan by using the sub-fields.

In one example, the method for obtaining the radiotherapy plan may be applied to a medical device, and after the medical device obtains the radiotherapy plan as described above, the radiotherapy plan may be output to an accelerator (e.g., a linear accelerator) of the medical device, and the accelerator forms a radiation field during radiotherapy according to the radiotherapy plan, so as to treat the object. Specifically, after the accelerator forms the radiation field in the radiotherapy process according to the radiotherapy plan, the accelerator can emit accurate radiation dose to the target area of the body to be detected by using the radiotherapy plan, so that the tumor receives the dose as much as possible, and meanwhile, the normal organ receives the dose as little as possible.

In one example, the process for "obtaining a leaf flux map according to dose constraints (i.e., step 1)" may include, but is not limited to, the following: firstly, obtaining leaf weight according to dose constraint; and obtaining a blade flux map by using the blade weight. Obtaining pencil beam weight according to dose constraint; obtaining a pencil beam flux map using the pencil beam weights; a flux map of the vanes is obtained using the pencil beam flux map.

The two modes are described in detail below with reference to two specific examples. The flow shown in fig. 4 is an implementation flow for the first embodiment. The flow shown in fig. 5 is an implementation flow for the second embodiment.

Referring to fig. 4, to obtain a flow chart of a radiotherapy plan, the method may include the following steps.

Leaf weights, instead of pencil beam weights, are obtained from dose constraints, step 401.

In one example, for the process of "obtaining leaf weights according to dose constraints", it may include: determining the weight of the blade to be optimized; obtaining the actual dosage of a detection point according to the weight of the current blade to be optimized; obtaining a dose constraint based on the actual dose; if the currently obtained dose constraint meets the constraint condition, or the iteration times reach a preset threshold value, determining the weight of the last blade to be optimized as the obtained blade weight, or determining the weight of the current blade to be optimized as the obtained blade weight; otherwise, the weight of the current blade to be optimized is adjusted, the adjusted weight of the blade is determined as the weight of the current blade to be optimized again, and the process of obtaining the actual dose of the detection point according to the weight of the current blade to be optimized is executed.

For example, a leaf weight of 1 may be initially configured, and the actual dose at a detection point is obtained according to the leaf weight of 1, and the dose constraint of 1 is obtained according to the actual dose at the detection point. And adjusting the weight 1 of the blade by adopting a preset algorithm to obtain the adjusted weight 2 of the blade. The actual doses at the check points are obtained from the leaf weights 2 and the dose constraints 2 are obtained from the actual doses at the check points. If the dose constraint 2 is smaller than the dose constraint 1, the leaf weight 2 is adjusted by adopting a preset algorithm to obtain an adjusted leaf weight 3. If the dose constraint 2 is not smaller than the dose constraint 1, the leaf weight 1 is adjusted by adopting a preset algorithm to obtain an adjusted leaf weight 3. The actual doses at the detection points are obtained from the leaf weights 3 and the dose constraints 3 are obtained from the actual doses at the detection points. And so on, until the dose constraint meets the constraint condition, taking the previous leaf weight (for example, when the dose constraint 10 meets the constraint condition, the leaf weight 9 corresponding to the previous dose constraint 9) as the result of step 401, or when the iteration number reaches a preset threshold, taking the current leaf weight (for example, the leaf weight 10 corresponding to the currently obtained dose constraint 10) as the result of step 401.

In the above process, for the process of "adjusting the leaf weight by using a preset algorithm", the preset algorithm may be a conjugate gradient method, simulated annealing, a genetic algorithm, linear programming, or the like, and the adjustment process refers to the above adjustment process of the pencil beam weight, and only the pencil beam weight is replaced by the leaf weight.

For the process of obtaining the actual dose of the detection point according to the weight of the leaf, the actual dose D of the detection point can be obtained by formula 3_i. In practical applications, the actual dosage at the detection point can be obtained by other methods, and the obtaining method is not limited. In formula 3, D_iIs the actual dose at the i-th spot, N₃May represent the number of blades, N₂Indicating the number of detection points, B_ilCan be the dose contribution of the ith leaf to the ith checkpoint, w_lMay be the ith leaf weight (i.e., dose). As can be seen from equation 3, the actual dose D at each checkpoint_iIt may be the sum of the dose contributions of all the leaves to the spot.

In equation 3, in order to obtain the actual dose D_iThe blade weight w needs to be known_lAnd dose contribution B_il. For the weight w of the blade_lThe weight w of the blade_lIs a continuously adjusted value, and each time a new blade weight w is obtained_lThe leaf weight w may then be determined_lSubstituting into equation 3 to obtain a new actual dose D_i. Thus, for equation 3, the blade weight w_lMay be a known value.

For "learned dose contribution B_il"in one example, in determining the ruler of the pencil beamAfter each leaf, there are 1 or more pencil beams per leaf, for example, 1 leaf for a pencil beam when the width of the leaf is 5 mm and the size of the pencil beam is 5 mm by 5 mm. When the width of the leaf is 1 cm and the pencil beam size is 5 mm by 5 mm, the leaf corresponds to 2 pencil beams. Based on this, if the dose contribution of the pencil beam to the ith detection point can be known, the dose contribution of the blade to the ith detection point can be known. For example, when the leaf corresponds to 1 pencil beam, the dose contribution of the leaf to the i-th detection point may be the same as the dose contribution of the pencil beam to the i-th detection point. When the leaf corresponds to 2 pencil beams, the dose contribution of the leaf to the i-th detection point may be 2 times the dose contribution of the pencil beam to the i-th detection point. Of course, the 2-fold is only an example, and the relationship of the 2-fold is not necessarily the one in practical application, and may vary according to practical situations as long as the following relationship is satisfied: the larger the dose contribution of the pencil beam to the ith detection point is, the larger the dose contribution of the blade to the ith detection point is; when the number of pencil beams corresponding to the vane is more, the dose contribution of the vane to the ith detection point is also larger, for example, when the vane 1 corresponds to 1 pencil beam and the vane 2 corresponds to 2 pencil beams, the dose contribution of the vane 2 to the ith detection point is larger than that of the vane 1 to the ith detection point.

In summary, to learn the dose contribution B_ilThe dose contribution of the pencil beam to the ith detection spot may be known first. For the process of "learning the dose contribution of the pencil beams to the i-th detection point", the dose contribution of each pencil beam to the i-th detection point is obtained after the dimensions of the pencil beam are determined. For example, the further away a pencil beam is from a respective detection point when each pencil beam impinges on that detection point, the smaller the dose contribution of that pencil beam to that detection point, and the closer the pencil beam is to that detection point, the greater the dose contribution of that pencil beam to that detection point. Furthermore, a larger size of a pencil beam indicates a larger dose of the pencil beam, a larger dose contribution of the pencil beam to the detection spot, and a smaller size of a pencil beam indicates a smaller dose of the pencil beam, a smaller dose contribution of the pencil beam to the detection spot. Based on the above principleThe dose contribution of each pencil beam to the ith detection point can be derived, and the specific algorithm is not described herein.

Therefore, after the dose contribution of the pencil beam to the ith detection point is obtained, the dose contribution of the vane to the ith detection point can be obtained based on the corresponding relation between the vane and the pencil beam. In one example, the widths of the leaves are different, such as 1 cm wide leaf, 5 mm wide leaf, etc., so the correspondence between the leaves and the pencil beam is different, and the dose contribution of the leaves to the ith detection point is different.

In summary, the process of "obtaining the actual dose of the detection point according to the weight of the leaf" may include: determining the width of the leaf (e.g. the width of the ith leaf), and the size of the pencil beam; determining the corresponding relation between the leaf and the pencil beam according to the width of the leaf and the size of the pencil beam; using the correspondence and the dose contribution of a single pencil beam to a checkpoint, the dose contribution of the leaf to the checkpoint (e.g. dose contribution B of the ith leaf to the ith checkpoint) is determined_il) (ii) a Since the values of N3 and N2 are known, the dose contribution of the lobe to the detection point, and the lobe weight (the ith lobe weight w) can be used_l) Based on the above formula 3, the actual dose D at the check point can be obtained_i。

In one example, for the process of "obtaining dose constraints from actual doses at the checkpoints", equation 2 can be used to obtain the dose constraints. For the processing of formula 2, see the content of formula 2 in step 101, which is not described herein again. Moreover, the weight of each leaf can be obtained by continuously optimizing the leaf weight so that the dose constraint satisfies the constraint condition, or the number of iterations reaches a preset threshold, which is similar to the process of obtaining the weight of each pencil beam in step 101 and will not be repeated herein.

In step 402, a blade flux map (also called a blade flux matrix) is obtained by using the blade weights.

In one example, after each leaf weight is obtained, as w₁、w₂、…w_N3Etc. can be prepared byThe leaf weights are combined together to form a leaf flux map, which is a matrix of the leaf weights. Moreover, the size of the blade represents the degree of dispersion of the blade flux map, and the smaller the size of the blade, the higher the degree of dispersion of the blade flux map, the more accurate the calculation, the larger the calculation amount, and the slower the speed. Conversely, the larger the vane size, the lower the degree of variation in the vane flux map. For example, the vane size is 5 mm by 5 mm, and for a field of 10 cm by 10 cm, the corresponding vane flux map is a 20 by 20 matrix.

In step 403, the leaf flux map is divided into sub-fields (sub-field matrix), and the sub-fields are used to obtain a radiotherapy plan. Wherein, the sub-field is the shape of the radiation field that the MLC needs to be put into, that is, the radiotherapy plan finally obtained.

In one example, for the sub-fields into which the leaf flux map is divided, that is, the field shape to which the MLC needs to be laid, a first value (e.g. 1) in the sub-field indicates that there is no occlusion by the leaf, and a second value (e.g. 0) in the sub-field indicates that there is occlusion by the leaf. Moreover, by dividing the leaf flux map into sub-fields instead of dividing the pencil beam flux map into sub-fields, occlusion errors can be avoided, and specific reasons are described in detail later.

In one example, for the process of "dividing the leaf flux map formed by the leaf weights into segments", the following principle can be followed: 1. the values in each sub-field divided are 0 and 1, and no other values are included. 2. The regions formed by 1 in each subfield are to be as contiguous as possible (to ensure dose accuracy). 3. The leaf flux map was divided into segments of the same size. 4. The number of segments is as small as possible (the radiation duration can be reduced). 5. The sum of the factors (i.e., the hop counts "2, 1", etc.) of all sub-fields is as small as possible (the radiation duration and the wear of the medical device can be reduced). 6. The number of 1 s in each subfield is as large as possible (dose accuracy can be guaranteed). Of course, the above procedures are only a few examples of the following principles, and in practical applications, there may be other following principles, which are not limited.

As shown below, an example of the division of the leaf flux map into segments is shown, with the leaf flux map formed by l leaf weights on the left of the equal sign and the divided segments on the right of the equal sign. Of course, the following is only an example of segmenting the sub-fields, and other segmentation methods may be adopted, which is not limited to this. In addition, only one 3 × 3 vane flux map is given below, and the actually applied vane flux map is more complicated, the processing is similar to that of the 3 × 3 vane flux map, and the 3 × 3 vane flux map is described as an example.

In the divided sub-fields, the number of divided sub-fields represents the number of sub-fields in a certain direction, 1 in each sub-field indicates that the sub-field is not shielded by a leaf of an MLC (representing that a target area is irradiated with radiation), and 0 in each sub-field indicates that the sub-field is shielded by a leaf of an MLC (representing that a target area is not irradiated with radiation). The factor (e.g., 2, 1, etc.) of each subfield is the hop count of the medical device, which represents the time that is not or is not occluded by the leaf, e.g., hop count 1 means that the time is 1 millisecond that is not or is occluded by the leaf, hop count 2 means that the time is 2 milliseconds that is not or is occluded by the leaf, etc. In practical application, the time corresponding to the hop count 1 is not necessarily 1 millisecond, and the time corresponding to the hop count 2 is not necessarily 2 milliseconds, but the ratio of the time corresponding to the hop count 1 to the time corresponding to the hop count 2 is 1:2, and the application is not limited to the time corresponding to the hop count.

In the above-described divided sub-fields, each number (1 or 0) corresponds to the width of the leaf, not the width of the pencil beam. Further, although the width of the leaf may be different, the leaf flux map is formed based on the leaf weight, and therefore, the sub-fields into which the leaf flux map is divided are also formed based on the leaf weight. Thus, each number in the subpicture represents the width of the leaf at the corresponding position. Taking the dimensions of the pencil beam as 1.5 mm by 1.5 mm as an example, for example, 100 of the first row of the first subfield, assuming that the leaf at the position corresponding to the first subfield is 6 mm wide, each number in the subfield represents 1.5 mm by 6 mm; assuming that the leaf in the corresponding position of the first subfield is 3 mm wide, each number in the subfield represents 1.5 mm by 3 mm. And so on. Wherein each number represents a width (e.g., 6 mm, 3 mm, etc.) that is the same as the width of the corresponding blade. And each number represents a length (e.g., 1.5 mm) that is the same as the length of a pencil (i.e., the width of a pencil).

Thus, for 100 in the first row of the first subfield, 2 positions of 0 can be blocked by the right blade, and assuming that the blade is 6 mm wide, the blocked area of the blade is 3 mm by 6 mm. For 011 in the second row of the first subfield, 1 position 0 needs to be blocked by the left blade, and the blocking area of the blade is 1.5 mm by 6 mm assuming that the blade is 6 mm wide. For the 000 in the third row of the second subfield, the left or right blade blocks 3 positions of 0, and assuming that the blade is 3 mm wide, the blocking area of the blade is 4.5 mm by 3 mm, and the blocking in other positions is similar, and will not be described again.

To achieve the above function, see fig. 3, which is a schematic view of leaves of an MLC, two leaves are typically deployed in one plane of the MLC, one leaf on the left and one leaf on the right. The leaves of the MLC are in a long strip shape, the length of each leaf is determined by the maximum field, and the leaves can move in the length direction and can shield the position of 0 in the sub-field in the length direction. The width of the leaf is the same as the width of the 0/1 position in the sub-field, for example, when the width of the 0/1 position in the sub-field is 3 mm, the width of the leaf is also 3 mm, so that the position of 0 in the sub-field is blocked in the width direction. Therefore, the leaf can well block the position of 0 in the sub-field.

After the above method is adopted, even if the widths of all leaves in the MLC are different, for example, the MLC includes leaves with a width of 6 mm, leaves with a width of 4.5 mm, and leaves with a width of 3 mm, the size of the pencil beam can be selected to be 1.5 mm by 1.5 mm, and the sub-fields are divided by using the leaf flux map, so as to avoid occlusion errors. For example, for 100 of the first row of the first subfield, 2 positions of 0 are blocked by the right blade, and assuming that the blade is 6 mm wide, the blocked area of the blade is 3 mm by 6 mm. Moreover, one 0 in the subfield represents 1.5 mm 6 mm, and 20 s represent 3 mm 6 mm, so the occlusion is correct.

Referring to fig. 5, to obtain a flow chart of a radiotherapy plan, the method may include the following steps.

Pencil beam weights are obtained from the dose constraints, step 501.

A pencil flux map is obtained using the pencil weights, step 502.

For step 501 to step 502, the processing procedure is referred to as step 101, and is not described herein again.

Using the pencil beam flux map, a bucket flux map is obtained, step 503.

In one example, for the process of "obtaining a leaf flux map using a pencil beam flux map", it may include: determining the width of the leaf and the size of the pencil beam; determining the corresponding relation between the leaf and the pencil beam according to the width of the leaf and the size of the pencil beam; and obtaining a blade flux map by utilizing the corresponding relation and the pencil beam flux map.

In one example, for a leaf flux map combined with leaf weights, a pencil beam flux map combined with pencil beam weights is associated, i.e., leaf to pencil beam correspondence. As shown in table 1, an example of a pencil beam flux map, the corresponding vane flux map is shown in table 2.

TABLE 1

0.493	0.317	0.399	0.713	0.218	0.914	0.827	0.066	0.304	0.658	0.180	0.388
												0.493	0.317	0.399	0.713	0.218	0.914	0.827	0.066	0.304	0.658	0.180	0.388
0.134	0.089	0.427	0.419	0.712	0.851	0.397	0.067	0.313	0.599	0.450	0.386
												0.134	0.089	0.427	0.419	0.712	0.851	0.397	0.067	0.313	0.599	0.450	0.386
0.134	0.089	0.427	0.419	0.712	0.851	0.397	0.067	0.313	0.599	0.450	0.386
												0.103	0.773	0.200	0.043	0.342	0.125	0.638	0.685	0.711	0.195	0.260	0.965
0.103	0.773	0.200	0.043	0.342	0.125	0.638	0.685	0.711	0.195	0.260	0.965
												0.048	0.562	0.566	0.482	0.735	0.463	0.657	0.841	0.348	0.241	0.870	0.993
0.048	0.562	0.566	0.482	0.735	0.463	0.657	0.841	0.348	0.241	0.870	0.993
												0.048	0.562	0.566	0.482	0.735	0.463	0.657	0.841	0.348	0.241	0.870	0.993
0.769	0.422	0.791	0.983	0.616	0.502	0.952	0.559	0.481	0.535	0.931	0.637
												0.769	0.422	0.791	0.983	0.616	0.502	0.952	0.559	0.481	0.535	0.931	0.637

TABLE 2

0.493	0.317	0.399	0.713	0.218	0.914	0.827	0.066	0.304	0.658	0.180	0.388
												0.134	0.089	0.427	0.419	0.712	0.851	0.397	0.067	0.313	0.599	0.450	0.386
0.103	0.773	0.200	0.043	0.342	0.125	0.638	0.685	0.711	0.195	0.260	0.965
												0.048	0.562	0.566	0.482	0.735	0.463	0.657	0.841	0.348	0.241	0.870	0.993
0.769	0.422	0.791	0.983	0.616	0.502	0.952	0.559	0.481	0.535	0.931	0.637

As can be seen from tables 1 and 2, the first lobe corresponds to two pencil beams (i.e., the lobe-to-pencil beam correspondence), and the lobe flux map for the first lobe is the first row of table 2, which corresponds to the first and second rows of table 1, i.e., the first and second rows of the pencil beam flux map. The second lobe corresponds to three pencil beams (i.e., lobe to pencil beam correspondence), and the lobe flux map for the second lobe is the second row of table 2, which corresponds to the third, fourth, and fifth rows of table 1, i.e., the third, fourth, and fifth rows of the pencil beam flux map. And so on. Therefore, based on the correspondence of the leaves to the pencil beams, it can be known that several pencil beams correspond to one leaf, e.g., two or three pencil beams correspond to one leaf, and based on the pencil beam flux maps corresponding to these several pencil beams, the leaf flux map of the leaf is determined.

In one example, the pencil beam flux map shown in table 1 is only an example, and in practical applications, assuming that the first vane corresponds to two pencil beams, i.e. the first and second rows of the pencil beam flux map correspond to the first row of the vane flux map, the values of the first and second rows of the pencil beam flux map may not be the same, and therefore, the values of the first and second rows of the pencil beam flux map may be added and divided by 2 to obtain the value of the first row of the vane flux map. Similarly, the numerical values of the third row, the fourth row, and the fifth row of the pencil beam flux map may be added, and the addition result may be divided by 3 to obtain the numerical values of the second row of the vane flux map. By analogy, a bucket flux map is determined based on the pencil beam flux map.

Step 504, the leaf flux map is divided into sub-fields (sub-field matrix), and the sub-fields are used to obtain a radiotherapy plan. Wherein, the sub-field is the shape of the radiation field that the MLC needs to be put into, that is, the radiotherapy plan finally obtained.

For step 504, the processing procedure is referred to as step 403, and is not described herein again.

In view of the problem caused by different widths of all leaves in the MLC, another method for selecting the dimensions of the pencil beam is also proposed in the embodiment of the present invention to solve the problem caused by different widths.

For pencil beam size selection, the width of all leaves in the MLC can be determined and the width of the pencil beam determined individually on a per width basis, so that the width of the pencil beam is the same as the width of the leaf, and the length of all pencil beams needs to be the same, e.g. the length of all pencil beams is the same as the width of the smallest pencil beam, so that the size of the pencil beam can be obtained. For example, if the MLC includes 5 mm wide leaves and 1 cm wide leaves, the pencil beam may be sized 5 mm x 5 mm for 5 mm wide leaves and 5 mm x 1 cm for 1 cm wide leaves. If the MLC includes leaves of 6 mm width, leaves of 4.5 mm width, leaves of 3 mm width, the pencil beam may be sized 3 mm x 6 mm for leaves of 6 mm width, 3 mm x 4.5 mm for leaves of 4.5 mm width, and 3 mm x 3 mm for leaves of 3 mm width.

Based on this selection of the pencil beam size, a procedure for obtaining a radiotherapy plan is shown in fig. 6.

Pencil beam weights are obtained from the dose constraints, step 601.

A pencil flux map is obtained using the pencil weights, step 602.

Steps

601 and 602 are similar to the processing procedure of step 101.

Steps

601, 602 differ from step 101 in that: the size of each pencil beam is the same in step 101, and may be different in

steps

601, 602, and the width of each pencil beam is the same as the width of the leaf at the corresponding position.

Step 603, the pencil beam flux map is divided into sub-fields (also called sub-field matrix), and the sub-fields are the shapes of the radiation fields to be formed by the MLC, that is, the radiotherapy plan finally obtained.

Wherein, for the process of dividing the pencil beam flux map into subfields, see step 102, step 603 differs from step 102 in that: in step 102, each number (1 or 0) in the divided sub-fields corresponds to the size of the pencil beam, and the size of each pencil beam is the same, e.g. 5 mm by 5 mm, so that each number in the sub-fields represents 5 mm by 5 mm. In step 603, the dimensions of each pencil beam may be different, with the dimensions of the pencil beam being 3 mm 6 mm for a pencil beam corresponding to a 6 mm width leaf, so that the number in the subfield corresponding to that pencil beam represents 3 mm 6 mm. For a pencil beam corresponding to a 3 mm wide leaf, the dimensions of the pencil beam are 3 mm x 3 mm, so the number in the subfield corresponding to the pencil beam represents 3 mm x 3 mm.

In summary, in step 603, each number (1 or 0) in the divided sub-fields corresponds to the width of the leaf and also corresponds to the width of the pencil beam, i.e. the width of the pencil beam is the same as the width of the leaf. For example, for 100 of the sub-fields, assuming that the leaf in its corresponding position is 6 mm wide, the dimensions of the pencil beam are 3 mm x 6 mm, each number in the sub-field representing 3 mm x 6 mm; assuming that the leaf in its corresponding position is 3 mm wide, the dimensions of the pencil beam are 3 mm by 3 mm, and each number in the subpicture represents 3 mm by 3 mm. And so on.

After the method is adopted, even if the widths of all the leaves in the MLC are different, such as the MLC comprises 5 mm-wide leaves and 1 cm-wide leaves, the shading error can be avoided.

The embodiment of the present application also provides an apparatus for obtaining a radiotherapy plan, which may be applied to a medical device, and fig. 7 is a block diagram illustrating a medical device 700 according to an exemplary embodiment. Referring to fig. 7, a medical device 700 may include a processing component 701 that further includes one or more processors and memory resources, represented by memory 702, for storing instructions, such as applications, executable by the processing component 701. The application programs stored in memory 702 may include one or more modules that each correspond to a set of instructions. Wherein, the radiotherapy plan obtaining device may be located in the memory 702, and the processing component 701 may execute the radiotherapy plan obtaining method of the embodiment of the present application through the radiotherapy plan obtaining device to obtain a radiotherapy plan and output the radiotherapy plan to the accelerator.

The medical device 700 may further comprise a power supply component 703, the power supply component 703 being configured to perform power management of the medical device 700. A wired or wireless network interface 704 configured to connect the medical device 700 to a network, and an input-output (I/O) interface 705.

The radiotherapy plan obtaining apparatus according to the embodiment of the present application may be logically divided into a plurality of modules, for example, as shown in fig. 8, and the apparatus may include: a first obtaining module 81, a segmentation module 82, a second obtaining module 83; wherein:

a first obtaining module 81 for obtaining a leaf flux map according to dose constraints;

a segmentation module 82 for segmenting the leaf flux map into segments;

a second obtaining module 83 for obtaining a radiotherapy plan using the sub-fields.

The first obtaining module 81 is specifically configured to obtain leaf weights according to dose constraints in a process of obtaining a leaf flux map according to dose constraints; obtaining a blade flux map by using the blade weight; or,

obtaining pencil beam weights based on dose constraints; obtaining a pencil beam flux map using the pencil beam weights; and obtaining a bucket flux map by using the pencil beam flux map.

The first obtaining module 81 is specifically configured to determine leaf weights to be optimized in a process of obtaining leaf weights according to dose constraints; obtaining the actual dosage of a detection point according to the weight of the current blade to be optimized; obtaining a dose constraint based on the actual dose; if the currently obtained dose constraint meets the constraint condition, or the iteration times reach a preset threshold value, determining the weight of the last blade to be optimized as the obtained blade weight, or determining the weight of the current blade to be optimized as the obtained blade weight; otherwise, the weight of the current blade to be optimized is adjusted, the adjusted weight of the blade is determined as the weight of the current blade to be optimized again, and the actual dosage of the detection point is obtained according to the weight of the current blade to be optimized.

The first obtaining module 81 is specifically configured to determine the width of the leaf and the size of the pencil beam in the process of obtaining the actual dose of the detection point according to the weight of the leaf to be optimized currently; determining the corresponding relation between the leaf and the pencil beam according to the width of the leaf and the size of the pencil beam; determining the dose contribution of the leaf to a detection point by using the corresponding relation and the dose contribution of the single pencil beam to the detection point; and obtaining the actual dose of the detection point by using the dose contribution of the blade to the detection point and the weight of the blade to be optimized currently.

The first obtaining module 81 is specifically configured to determine the width of the leaf and the size of the pencil beam in the process of obtaining the leaf flux map by using the pencil beam flux map; determining the corresponding relation between the leaf and the pencil beam according to the width of the leaf and the size of the pencil beam; and obtaining a blade flux map by using the corresponding relation and the pencil beam flux map.

The first obtaining module 81 is specifically configured to determine the width of the pencil beam in the process of determining the size of the pencil beam according to the following strategy: if the multi-leaf collimator comprises leaves of various widths, each width of the leaves divided by the width of the pencil beam is an integer; the length of a pencil beam is determined to be the same as the width of the pencil beam.

The modules of the device can be integrated into a whole or can be separately deployed. The modules can be combined into one module, and can also be further split into a plurality of sub-modules.

The disclosure of the present application is only a few specific embodiments, but the present application is not limited to these, and any variations that can be made by those skilled in the art are intended to fall within the scope of the present application.

Claims

1. A method of obtaining a radiotherapy plan, the method comprising:

obtaining pencil beam weights based on dose constraints;

obtaining a pencil beam flux map using the pencil beam weights;

obtaining a flux map of the leaf blade by using the pencil beam flux map;

dividing the leaf flux map into segments;

and obtaining a radiotherapy plan by using the sub-fields.

2. The method of claim 1,

the process for obtaining the flux map of the vane by using the pencil beam flux map specifically comprises the following steps:

determining the width of the leaf, and the size of the pencil beam;

determining the corresponding relation between the leaf and the pencil beam according to the width of the leaf and the size of the pencil beam;

and obtaining a blade flux map by using the corresponding relation and the pencil beam flux map.

3. The method of claim 2,

the process of determining the size of the pencil beam specifically includes:

the width of the pencil beam is determined according to the following strategy: if the multi-leaf collimator comprises leaves of various widths, each width of the leaves divided by the width of the pencil beam is an integer;

the length of a pencil beam is determined to be the same as the width of the pencil beam.

4. An apparatus for obtaining a radiotherapy plan, the apparatus comprising:

a first obtaining module for obtaining pencil beam weights in accordance with dose constraints; obtaining a pencil beam flux map using the pencil beam weights; obtaining a flux map of the leaf blade by using the pencil beam flux map;

5. The apparatus of claim 4,

the first obtaining module is specifically used for determining the width of the leaf and the size of the pencil beam in the process of obtaining the leaf flux map by using the pencil beam flux map; determining the corresponding relation between the leaf and the pencil beam according to the width of the leaf and the size of the pencil beam; and obtaining a blade flux map by using the corresponding relation and the pencil beam flux map.