Rapid intensity modulated radiation therapy optimization method and system

Abstract

The invention discloses a rapid intensity modulated radiation therapy optimization method and system, and the method comprises the steps: receiving imported clinical treatment data of a patient, and calculating a dose contribution matrix based on the clinical treatment data of the patient; dividing the dose contribution matrix into a plurality of matrix blocks along the organ direction to obtain an organ sub-region sequence; dividing the dose contribution matrix into a plurality of matrix blocks along the radiation field direction to obtain a radiation field sub-region sequence; partitioning a dose contribution matrix by using the organ sub-region sequence and the radiation field sub-region sequence to obtain dose contribution sub-matrix blocks; and performing intensity modulated radiation therapy optimization by using the dose contribution sub-matrix block. According to the fast intensity modulated radiation therapy optimization method and system provided by the invention, the original dose contribution matrix is converted into a series of matrix blocks before optimization, resource occupation in the optimization process is reduced, and the optimization speed is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intensity modulated radiotherapy optimization, and in particular to a rapid intensity modulated radiotherapy optimization method and system. Background Art

[0002] Intensity-modulated radiotherapy (IMRT) optimization is an advanced radiotherapy technique. Using a rotating treatment machine, tumors can be irradiated in multiple directions while simultaneously adjusting the dose and irradiation field to minimize damage to surrounding normal tissue. Compared to traditional methods, IMRT treatments offer shorter treatment times, more precise dose distribution, and improved target adaptability. However, IMRT optimization faces challenges such as complex dose distribution and high computational resource requirements. During IMRT optimization, a dose contribution matrix is ​​often used to represent the mapping between field intensity and voxel dose. However, storing this matrix requires significant storage space.

[0003] The existing Chinese invention patent application, application number 202310086294.3, entitled "A System for Optimizing Intensity-Modulated Radiotherapy Beam Angles," belongs to the field of intensity-modulated radiotherapy and discloses an intensity-modulated radiotherapy beam angle optimization system. The system compresses radiotherapy data and uses sparse reading to calculate a dose deposition matrix. The system then solves the intensity distribution matrix of the beam direction using a conjugate gradient algorithm. The system then calculates a dose distribution matrix based on the dose deposition matrix and the intensity distribution matrix. The system constrains the beam direction using a lateral suppression function and a neighboring effect function. The system then optimizes the dose distribution matrix using voxel dose conditions within the sub-field to obtain a target dose for each beam direction. The system then calculates the importance of the beam direction based on the target dose, suppression value, and neighboring effect value. The system then eliminates the least important beam direction until the number of remaining beam directions equals a predetermined number of angles. The system then converts the beam direction into a beam angle and outputs the beam angle and its corresponding target dose. Although the beam merging method employed in this invention patent application avoids optimizing unimportant beams by eliminating beam angles, it alters the normal optimization process and does not address the technical issue of the large storage space required to store the matrix.

[0004] In view of this, the present invention patent is proposed. Summary of the Invention

[0005] To solve the memory usage problem of the dose contribution matrix in intensity-modulated radiotherapy optimization, the present invention provides a rapid intensity-modulated radiotherapy optimization method and system. Specifically, the following technical solutions are adopted:

[0006] A rapid intensity-modulated radiotherapy optimization method comprising:

[0007] receiving imported patient clinical visit data, and calculating a dose contribution matrix based on the patient clinical visit data;

[0008] Divide the dose contribution matrix into multiple matrix blocks along the organ direction to obtain an organ sub-region sequence;

[0009] The dose contribution matrix is ​​divided into multiple matrix blocks along the radiation field direction to obtain a radiation field sub-area sequence;

[0010] Dividing the dose contribution matrix into blocks using the organ sub-region sequence and the field sub-region sequence to obtain dose contribution sub-matrix blocks;

[0011] The dose contribution sub-matrix block is used to perform intensity modulated radiotherapy optimization.

[0012] As an optional embodiment of the present invention, in a rapid intensity modulated radiotherapy optimization method of the present invention, the received imported patient clinical treatment data is used to calculate the dose contribution matrix based on the patient clinical treatment data. The dose contribution matrix has the form A=(a n，ijk ) N，M , where A represents the dose contribution matrix, a n，ijk represents the dose contribution of unit pencil beam n to unit voxel ijk of the patient, N represents the number of pencil beams, and M represents the number of organ voxels.

[0013] As an optional embodiment of the present invention, in a rapid intensity modulated radiotherapy optimization method of the present invention, the dose contribution matrix is ​​divided into multiple matrix blocks along the organ direction to obtain an organ sub-region sequence {R1, R2, ..., R P}, R p Index is I p The voxel structure of the organ sub-region sequence {R1, R2R..., R P} include:

[0014] Step S101, initialize the organ sub-region sequence, organ sub-region R p The characteristic f p =a N,ijk =(a 1，ijk ,...,a N，ijk ), weight w p =w ijk , voxel index I p ={ijk}, where the subregion index p corresponds to the voxel index ijk in one-to-one correspondence;

[0015] Step S102, calculate R p and R p′ Similarity

[0016] Step S103: Select the organ sub-region pair with the highest similarity. and satisfy

[0017] Step S104: and Merge into new organ subregions Characterization of novel organ subregions Weight Organ Point Index

[0018] Step S105: organ sub-region sequence {R1, R2, ..., R P}Delete organ sub-region and Adding new organ subregions

[0019] Step S106, repeating steps S102 to S105 until the number of organ sub-regions is less than the preset number P of sub-regions.

[0020] As an optional embodiment of the present invention, in a rapid intensity modulated radiotherapy optimization method of the present invention, the dose contribution matrix is ​​divided into multiple matrix blocks along the radiation field direction to obtain a radiation field sub-region sequence {E1, E2, ..., E Q}, E q By index J q The pencil-shaped bundle of E q Different pencil beams are located at the same field angle on the same arc.

[0021] As an optional embodiment of the present invention, in a rapid intensity modulated radiotherapy optimization method of the present invention, the dose contribution matrix is ​​divided into blocks using the organ sub-region sequence and the field sub-region sequence to obtain the dose contribution sub-matrix block. in

[0022] For the dose contribution sub-matrix block A p，q Do truncated singular value decomposition Among them U k =(u1, u2, ..., u k ) is an n×k matrix, V k =(v1, v2, ..., v k ) is an m×k matrix, Σ k =diag(σ1,σ2,...,σ k ) is an n×m matrix, each dose contribution sub-matrix block A p，q The truncation threshold k is obtained according to the truncation threshold search algorithm.

[0023] As an optional embodiment of the present invention, in a rapid intensity modulated radiotherapy optimization method of the present invention, each matrix block A p，q The calculation process of the truncation threshold k includes:

[0024] Step S201, obtain the remaining working memory MemFree;

[0025] Step S202, define the initial threshold threshold = threshold min > 0;

[0026] Step S203, calculate the truncated k value of each dose contribution sub - matrix block A p，q where k′ = min(m, n);

[0027] Step S204, calculate the pre - allocated memory MemSum threshold = ∑ p ∑ q (n p，q × k p，q + k p，q × mp,q, if MemSumthreshold < MemFree, terminate the iteration, otherwise increase the threshold threshold and go to Step S203.

[0028] As an optional implementation manner of the present invention, in a fast intensity - modulated radiotherapy optimization method of the present invention, the intensity - modulated radiotherapy optimization using the dose contribution sub - matrix block includes:

[0029] Using the matrix block U p，q obtained by performing truncated singular value decomposition on the dose contribution sub - matrix block A p，q，k , Σ p，q，k’ and the sub - field intensity to calculate the organ dose, and then calculate the optimized intensity - modulated radiotherapy dose. The calculation method of the intensity - modulated radiotherapy dose is

[0030] where,

[0031] Ψ q is a vector composed of elements corresponding to J q , and D p is a vector composed of elements corresponding to I p .

[0032] As an optional implementation manner of the present invention, in a fast intensity - modulated radiotherapy optimization method of the present invention, the intensity - modulated radiotherapy optimization using the dose contribution sub - matrix block includes:

[0033] Using the matrix block U p，q obtained by performing truncated singular value decomposition on the dose contribution sub - matrix block A p，q，k , Σ p，q，k’ calculate the update direction of the intensity during the optimization process

[0034]

[0034] in,

[0035]

[0036] grad(Ψ) represents the gradient of the objective function with respect to the subfield intensity, grad(D) represents the gradient of the objective function with respect to the organ dose, and grad(Ψ) q is the index J in grad(Ψ) q The vector of elements, grad(D) p is the index I in grad(D) p A vector consisting of the elements of .

[0037] As an optional embodiment of the present invention, in a rapid intensity modulated radiotherapy optimization method of the present invention, the use of the dose contribution sub-matrix block to optimize intensity modulated radiotherapy includes:

[0038] Calculate the update step length α of the field intensity;

[0039] Update the field intensity using the following method: Ψ = Ψ - α·grad(Ψ);

[0040] Determine whether the optimized field intensity meets the clinical treatment requirements. If so, end the optimization; otherwise, repeat the optimization process.

[0041] The present invention also provides a system for implementing the rapid intensity modulated radiotherapy optimization method, comprising:

[0042] a data preprocessing module, receiving imported patient clinical visit data, calculating a dose contribution matrix based on the patient clinical visit data, dividing the dose contribution matrix into a plurality of matrix blocks along the organ direction to obtain an organ sub-region sequence, dividing the dose contribution matrix into a plurality of matrix blocks along the radiation field direction to obtain a radiation sub-region sequence, and using the organ sub-region sequence and the radiation sub-region sequence to block the dose contribution matrix to obtain a dose contribution sub-matrix block;

[0043] The intensity modulated radiotherapy optimization module uses the dose contribution sub-matrix block to perform intensity modulated radiotherapy optimization.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] The present invention provides a rapid intensity-modulated radiotherapy optimization method and system. Before performing intensity-modulated radiotherapy optimization, the method and system utilize the similarity of different voxels in the same field of view of the dose contribution matrix without interfering with the normal optimization process, and creatively utilize matrix blocking and matrix compression methods to achieve compression of the dose contribution matrix, thereby reducing the resource usage of intensity-modulated optimization, improving the optimization speed, and shortening the optimization time. Therefore, the method and system have practical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A flowchart of a method for optimizing rapid intensity modulated radiotherapy according to an embodiment of the present invention. DETAILED DESCRIPTION

[0047] To make the purpose, technical solutions and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them.

[0048] Therefore, the following detailed description of the embodiments of the present invention is not intended to limit the scope of the claimed invention, but merely represents some embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort shall fall within the scope of protection of the present invention.

[0049] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features and technical solutions therein may be combined with each other.

[0050] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0051] In the description of the present invention, it should be noted that the terms "upper" and "lower" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, or the orientations or positional relationships in which the inventive product is typically placed when in use, or the orientations or positional relationships commonly understood by those skilled in the art. Such terms are intended solely to facilitate the description of the present invention and simplify the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limitations on the present invention. Furthermore, the terms "first" and "second" and the like are used solely for distinction and should not be construed as indicating or implying relative importance.

[0052] See also Figure 1 As shown, a rapid intensity modulated radiotherapy optimization method of this embodiment includes:

[0053] receiving imported patient clinical visit data, and calculating a dose contribution matrix based on the patient clinical visit data;

[0054] Divide the dose contribution matrix into multiple matrix blocks along the organ direction to obtain an organ sub-region sequence;

[0055] The dose contribution matrix is ​​divided into multiple matrix blocks along the radiation field direction to obtain a radiation field sub-area sequence;

[0056] Dividing the dose contribution matrix into blocks using the organ sub-region sequence and the field sub-region sequence to obtain dose contribution sub-matrix blocks;

[0057] The dose contribution sub-matrix block is used to perform intensity modulated radiotherapy optimization.

[0058] A rapid intensity-modulated radiotherapy optimization method of this embodiment, without interfering with the normal optimization process, utilizes the similarity of different voxels in the same field of the dose contribution matrix before performing intensity-modulated radiotherapy optimization, and creatively uses matrix blocking and matrix compression methods to achieve compression of the dose contribution matrix, thereby reducing the resource usage of intensity-modulated optimization, improving the optimization speed, and shortening the optimization time, and therefore has practical application value.

[0059] Specifically, the patient clinical consultation data of this embodiment mainly includes the following contents:

[0060] Basic information: including the patient's name, age, gender and other personal identification information, as well as the examination date, etc. This information ensures the accuracy and identifiability of the report and provides important reference for subsequent diagnosis and treatment.

[0061] Test Items: Detailed description of the types of imaging tests performed, such as X-rays, CT (computed tomography), MRI (magnetic resonance imaging), etc. These tests use different technologies to provide detailed information about the patient's internal structure, helping doctors make a diagnosis.

[0062] Imaging findings: This section describes the abnormalities or normal findings observed by the doctor in the image, including descriptions of normal anatomy and the location, size, morphology, and density of abnormal lesions. This section is the core of the radiology report and helps doctors reach a diagnosis.

[0063] Diagnostic opinion: The doctor's diagnostic conclusion based on the imaging findings, including the nature of the lesion (such as tumor, inflammation, cyst, etc.), as well as possible disease clues and preliminary diagnosis.

[0064] Recommendation section: includes suggestions for the patient's subsequent treatment or further examination to help patients understand treatment options and prognosis.

[0065] Furthermore, in a rapid intensity modulated radiotherapy optimization method of this embodiment, the received imported patient clinical treatment data is used to calculate the dose contribution matrix based on the patient clinical treatment data. The dose contribution matrix has the form A=(a n，ijk ) N，M , where A represents the dose contribution matrix, a n，ijk represents the dose contribution of unit pencil beam n to unit voxel ijk of the patient, N represents the number of pencil beams, and M represents the number of organ voxels.

[0066] In a rapid intensity modulated radiotherapy optimization method of this embodiment, the dose contribution matrix is ​​divided into multiple matrix blocks along the organ direction to obtain an organ sub-region sequence {R1, R2, ..., R P}, R p Index is I p The voxels of the organ sub-region sequence {R1, R2, ..., R P} include:

[0067] Step S101, initialize the organ sub-region sequence, organ sub-region R p The characteristic f p =a N，ijk =(a 1，ijk ,...,a N，ijk ), weight w p =w ijk , voxel index I p ={ijk}, where the subregion index p corresponds to the voxel index ijk in one-to-one correspondence;

[0068] Step S102, calculate R p and R p’ Similarity

[0069] Step S103: Select the sub-region pair with the highest similarity. and satisfy

[0070] Step S104: and Merge into new areas Characteristics of the new area Weight Organ Point Index

[0071] Step S105: organ sub-region sequence {R1, R2, ..., R P}Delete sub-area and Supplementary sub-areas

[0072] Step S106, repeating steps S102 to S105 until the number of organ sub-regions is less than the preset number P of sub-regions.

[0073] In a rapid intensity modulated radiotherapy optimization method of this embodiment, the dose contribution matrix is ​​divided into multiple matrix blocks along the radiation field direction to obtain a radiation field sub-region sequence {E1, E2, ..., E Q}, E qComposed of a pencil beam indexed by J q and different pencil beams are located at the same beam angle of the same arc. q In a fast intensity-modulated radiotherapy optimization method of this embodiment, the dose contribution matrix is partitioned by using the organ sub-region sequence and the beam sub-region sequence to obtain dose contribution sub-matrix blocks

[0074] Among them For the dose contribution sub-matrix block A

[0075] Perform truncated singular value decomposition p，q where U =(u1, u2,..., u k ) is an n×k matrix, V k =(v1, v2,..., v k ) is an m×k matrix, Σ k =diag(σ1, σ2,..., σ k ) is an n×m matrix, and the truncation threshold k of each dose contribution sub-matrix block A k is obtained according to the truncation threshold search algorithm. p，q Furthermore, in a fast intensity-modulated radiotherapy optimization method of this embodiment, the calculation process of the truncation threshold k of each matrix block A

[0076] includes: p，q Step S201, obtain the remaining working memory MemFree;

[0077] Step S202, define the initial threshold threshold = threshold

[0078] >0; min Step S203, calculate the truncation k value of each dose contribution sub-matrix block A

[0079] where k' = min(m, n); p，q Step S204, calculate the pre-allocated memory MemSum where MemSum

[0080] =∑ threshold ∑ p ∑ q (n p，q ×k p，q +k p，q ×mp,q, if MemSumthreshold < MemFree, terminate the iteration, otherwise increase the threshold threshold and go to step S203.

[0081] In a fast intensity-modulated radiotherapy optimization method of this embodiment, the intensity-modulated radiotherapy optimization using the dose contribution sub-matrix block includes:

[0082] Using the dose contribution sub-matrix block A p，q Do truncated singular value decomposition of the matrix block U p，q，k ,Σ p，q，k’ The organ dose is calculated by the intensity of the field and then the optimized intensity modulated radiation dose is calculated. The calculation method of the intensity modulated radiation dose is:

[0083] in,

[0084] Ψ q By J q The vector consisting of the elements corresponding to Ψ, D p byI p The vector corresponding to the elements of D.

[0085] In a rapid intensity modulated radiotherapy optimization method of this embodiment, the use of the dose contribution sub-matrix block to optimize intensity modulated radiotherapy includes:

[0086] Using the dose contribution sub-matrix block A p，q Do truncated singular value decomposition of the matrix block U p，q，k ,Σ p，q，k’ Calculate the update direction of strength during optimization

[0087] in,

[0088]

[0089] grad9Ψ) represents the gradient of the objective function with respect to the subfield intensity, grad(D) represents the gradient of the objective function with respect to the organ dose, and grad(Ψ) q is the index J in grad(Ψ) q The vector of elements, grad(D) p The index in grad(D) is I p A vector consisting of the elements of .

[0090] Furthermore, in a rapid intensity modulated radiotherapy optimization method of this embodiment, the use of the dose contribution sub-matrix block to optimize intensity modulated radiotherapy includes:

[0091] Calculate the update step length α of the field intensity;

[0092] Update the field intensity using the following method: Ψ = Ψ - α·grad(Ψ);

[0093] Determine whether the optimized field intensity meets the clinical treatment requirements. If so, end the optimization; otherwise, repeat the optimization process.

[0094] As an optional implementation of this embodiment, a specific example of a rapid intensity modulated radiotherapy optimization method of this embodiment includes:

[0095] This embodiment also provides a system for implementing the rapid intensity modulated radiotherapy optimization method, including:

[0096] a data preprocessing module, receiving imported patient clinical visit data, calculating a dose contribution matrix based on the patient clinical visit data, dividing the dose contribution matrix into a plurality of matrix blocks along the organ direction to obtain an organ sub-region sequence, dividing the dose contribution matrix into a plurality of matrix blocks along the radiation field direction to obtain a radiation sub-region sequence, and using the organ sub-region sequence and the radiation sub-region sequence to block the dose contribution matrix to obtain a dose contribution sub-matrix block;

[0097] The intensity modulated radiotherapy optimization module uses the dose contribution sub-matrix block to perform intensity modulated radiotherapy optimization.

[0098] In this embodiment, a rapid intensity-modulated radiotherapy optimization system is provided. Without interfering with the normal optimization process, the data preprocessing module utilizes the similarity of different voxels in the same field of the dose contribution matrix, and creatively uses matrix blocking and matrix compression methods to achieve compression of the dose contribution matrix, thereby reducing the resource usage of the intensity-modulated radiotherapy optimization module, improving the optimization speed, and shortening the optimization time. Therefore, it has practical application value.

[0099] A system for implementing the rapid intensity modulated radiotherapy optimization method in this embodiment has a specific workflow example as follows:

[0100] Data preprocessing module: import the patient's clinical treatment data and use the patient's clinical treatment data to calculate the dose contribution matrix A = (a n，ijk ) N，M , where A represents the dose contribution matrix, a n，ijk represents the dose contribution of unit pencil beam n to unit voxel ijk of the patient, N represents the number of pencil beams, and M represents the number of organ voxels.

[0101] The data preprocessing module needs to convert the dose contribution matrix into a series of small matrices. The conversion steps are as follows:

[0102] Step S1, the data preprocessing module includes an organ grid merging module, which divides the dose contribution matrix into multiple matrix blocks along the organ direction. The organ sub-region sequence corresponding to the matrix block is {R1, R2, ..., R P}, R p By index I p The voxel composition is as follows to obtain the organ sub-region sequence {R1, R2, ..., R P}

[0103] Step S101, initialize the organ sub-region sequence, organ sub-region R p The characteristic f p =a N，ijk =(a 1，ijk ,...,a N，ijk ), weight w p =w ijk , voxel index I p ={ijk}, where the subregion index p corresponds to the voxel index ijk in one-to-one correspondence;

[0104] Step S102, calculate R p and R p' Similarity

[0105] Step S103: Select the sub-region pair with the highest similarity. and satisfy

[0106] Step S104: and Merge into new areas Characteristics of the new area Weight Organ Point Index

[0107] Step S105: organ sub-region sequence {R1, R2, ..., R P}Delete sub-area and Supplementary sub-areas

[0108] Step S106, repeating steps S102 to S105 until the number of organ sub-regions is less than the preset number P of sub-regions.

[0109] Step S2, the data preprocessing module includes a field grid merging module, which divides the dose contribution matrix into multiple matrix blocks along the field direction. The field region sequence corresponding to the matrix blocks is {E1, E2, ..., E Q}, E q By index J q The pencil-shaped bundle of E q Different pencil beams are located at the same field angle on the same arc.

[0110] Step S3, the data preprocessing module includes a dose contribution matrix compression module, which uses the organ sub-region sequence {R1, R2, ..., R P}, and the beam sub-region sequence {E1, E2,..., E Q}, the dose contribution matrix A is partitioned

[0111]

[0112] Where For A p，q Perform truncated singular value decomposition Where U k =(u1, u2,..., u k ) is an n×k matrix, V k =(v1, v2,..., v k ) is an m×k matrix, Σ k =diag(σ1, σ2,..., σ k ) is an n×m matrix. The numerical value k is obtained according to the truncated threshold search algorithm, and the calculation steps are as follows:

[0113] Step S301, obtain the remaining working memory MemFree;

[0114] Step S302, define the initial threshold threshold = threshold min >O;

[0115] Step S303, calculate the truncated k value of each matrix block A p，q Where k' = minm,n;

[0116] Step S304, calculate the pre-allocated memory MemSum threshold =∑ p ∑ q (n p，q ×k p，q +k p，q ×mp,q, if MemSumthreshold < MemFree, terminate the iteration, otherwise increase the threshold threshold and go to step S303.

[0117] Intensity-modulated radiotherapy optimization module: Using the matrix blocks {U k , ∑ k , V k} in the data preprocessing module for intensity-modulated optimization. According to the input constraint information and device information, an optimization objective is constructed, and the conjugate gradient method is used to optimize the objective. The optimization steps are as follows

[0118] Step S401, initialize the beam intensity Ψ;

[0119] Step S402, calculate the optimization objective F(D(Ψ)), and the dose calculation method is ​in

[0120]

[0121] Ψ q By J q The vector consisting of the elements corresponding to Ψ, D p By J p The vector corresponding to the elements of D;

[0122] Step S403: Calculate the update direction of the sub-field intensity Ψ in

[0123]

[0124] grad(Ψ) represents the gradient of the objective function with respect to the subfield intensity, grad(D) represents the gradient of the objective function with respect to the organ dose, and grad(Ψ) q is the index J in grad(Ψ) q The vector of elements, grad(D) p The index in grad(D) is I p A vector consisting of the elements of ;

[0125] Step S405, calculating the update step length α of the field intensity;

[0126] Step S406, updating the field intensity, the updating method is Ψ=Ψ-α·grad(Ψ);

[0127] Step S407, determining whether the optimized field intensity meets the clinical treatment requirements, if so, the optimization is terminated, otherwise, steps S402 to S405 are repeated.

[0128] The above embodiments are only used to illustrate the present invention and are not intended to limit the technical solutions described in the present invention. Although this specification has described the present invention in detail with reference to the above embodiments, the present invention is not limited to the above specific implementation methods. Therefore, any modification or equivalent replacement of the present invention; and all technical solutions and improvements thereof that do not depart from the spirit and scope of the invention are included in the scope of the claims of the present invention.

Claims

1. A rapid intensity modulated radiotherapy optimization method, characterized in that: include: receiving imported clinical visit data of patients, and calculating a dose contribution matrix based on the clinical visit data of patients; Divide the dose contribution matrix into multiple matrix blocks along the organ direction to obtain an organ sub-region sequence; The dose contribution matrix is ​​divided into multiple matrix blocks along the radiation field direction to obtain a radiation field sub-area sequence; Using the organ sub-region sequence and the field sub-region sequence to divide the dose contribution matrix into blocks, to obtain dose contribution sub-matrix blocks; The dose contribution sub-matrix block is used to perform intensity modulated radiotherapy optimization.

2. A rapid intensity modulated radiotherapy optimization method according to claim 1, characterized in that: The step of receiving the imported clinical visit data of the patient and calculating the dose contribution matrix based on the clinical visit data of the patient, wherein the dose contribution matrix has the form A=(a n，ijk ) N，M , where A represents the dose contribution matrix, a n，iik represents the dose contribution of unit pencil beam n to unit voxel ijk of the patient, N represents the number of pencil beams, and M represents the number of organ voxels.

3. A rapid intensity modulated radiotherapy optimization method according to claim 2, characterized in that: The dose contribution matrix is ​​divided into a plurality of matrix blocks along the organ direction to obtain an organ sub-region sequence {R1, R2, ..., R P }, R p Index is I p The organ sub-region sequence {R1, R2, ..., R P } include: Step S101, initializing the organ sub-region sequence, the organ sub-region R p The characteristic f p =a N，ijk =(a 1，ijk , ..., a N，ijk ), weight w p =w ijk , voxel index I p ={ijk}, where the sub-region index p corresponds to the voxel index ijk one by one; Step S102, calculate R p and R p’ Similarity Step S103: Select the organ sub-region pair with the highest similarity. and satisfy Step S104: and Merge into new organ sub-regions Characterization of novel organ subregions Weight Organ Point Index Step S105: organ sub-region sequence {R1, R2, ..., R P }Delete organ sub-region and Addition of new organ sub-regions Step S106, repeating steps S102 to S105 until the number of organ sub-regions is less than the preset number P of sub-regions.

4. A rapid intensity modulated radiotherapy optimization method according to claim 3, characterized in that: The dose contribution matrix is ​​divided into a plurality of matrix blocks along the beam direction to obtain a beam sub-area sequence {E1, E2, ..., E Q }, E q By index J q The pencil-shaped beam is composed of q Different pencil beams are located at the same field angle in the same arc.

5. A rapid intensity modulated radiotherapy optimization method according to claim 4, characterized in that: The dose contribution matrix is ​​divided into blocks using the organ sub-region sequence and the field sub-region sequence to obtain dose contribution sub-matrix blocks. in For the dose contribution sub-matrix block A p，q Do truncated singular value decomposition Among them U k =(u1, u2, ..., u k ) is an n×k matrix, V k =(v1, v2, ..., v k ) is an m×k matrix, Σ k =diag(σ1,σ2,…,σ k ) is an n×m matrix, each dose contribution submatrix block A p，q The truncation threshold k is obtained according to the truncation threshold search algorithm.

6. A rapid intensity modulated radiotherapy optimization method according to claim 5, characterized in that: Each matrix block A p，q The calculation process of the truncation threshold k includes: Step S201, obtaining the remaining working memory MemFree; Step S202, define the initial threshold threshold=threshold min >0; Step S203, calculate each dose contribution sub-matrix block A p，q The cutoff k value Where k′=min(m,n); Step S204, calculate the pre-allocated memory MemSum threshold = ∑ p ∑ q (n p，q × k p，q + k p，q × mp,q. If MemSumthreshold < MemFree, terminate the iteration; otherwise, increase the threshold threshold and go to step S203.

7. A rapid intensity modulated radiotherapy optimization method according to claim 5, characterized in that: The optimization of intensity modulated radiotherapy using the dose contribution sub-matrix block includes: Using the dose contribution sub-matrix block A p，q Do truncated singular value decomposition of the matrix block U p，q，k Σ p，q，k , The organ dose is calculated by the intensity of the field and then the optimized intensity modulated radiation dose is calculated. The calculation method of the intensity modulated radiation dose is: in, Ψ q By J q The vector corresponding to the elements of Ψ, D p byI p The vector corresponding to the elements of D.

8. A rapid intensity modulated radiotherapy optimization method according to claim 7, characterized in that: The optimization of intensity modulated radiotherapy using the dose contribution sub-matrix block includes: Using the dose contribution sub-matrix block A p，q Do truncated singular value decomposition of the matrix block U p，q，k Σ p，q，k , Calculate the update direction of strength during optimization in, grad(Ψ) represents the gradient of the objective function with respect to the subfield intensity, grad(D) represents the gradient of the objective function with respect to the organ dose, and grad(Ψ) q is the index J in grad(Ψ) q The vector of elements of , grad(D) p is the index I in grad(D) p A vector consisting of the elements of .

9. A rapid intensity modulated radiotherapy optimization method according to claim 8, characterized in that: The optimization of intensity modulated radiotherapy using the dose contribution sub-matrix block includes: Calculate the field intensity update step length α; Update the field intensity, the updating method is Ψ = Ψ - α · grad (Ψ); Determine whether the optimized field intensity meets the clinical treatment requirements. If so, end the optimization; otherwise, repeat the optimization process.

10. A system for implementing the rapid intensity modulated radiotherapy optimization method according to any one of claims 1 to 9, characterized in that: include: a data preprocessing module, receiving imported clinical visit data of patients, calculating a dose contribution matrix based on the clinical visit data of patients, dividing the dose contribution matrix into a plurality of matrix blocks along the organ direction to obtain an organ sub-region sequence, dividing the dose contribution matrix into a plurality of matrix blocks along the radiation field direction to obtain a radiation field sub-region sequence, and using the organ sub-region sequence and the radiation field sub-region sequence to divide the dose contribution matrix into blocks to obtain dose contribution sub-matrix blocks; The intensity modulated radiotherapy optimization module uses the dose contribution sub-matrix block to perform intensity modulated radiotherapy optimization.

Images