A dynamic intensity modulation device and optimization method based on dynamic penumbra model

By introducing dynamic penumbra model and direct machine parameter optimization in dynamic intensity regulation technology, the problems of long exposure time and accuracy error in dynamic intensity regulation are solved, and fast and accurate dynamic intensity regulation plan generation is achieved, which improves treatment efficiency and Accuracy.

CN115317811BActive Publication Date: 2024-11-22SUPERACCURACY SCIENCE & TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210985163.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-08-17
Publication Date
2024-11-22
Estimated Expiration
2042-08-17

AI Technical Summary

Technical Problem

The existing dynamic intensity-modulation technology has problems in radiation therapy with long exposure time, low efficiency and accuracy error caused by leaf penumbra.

Method used

Using a dynamic intensity-modulation device and optimization method based on the dynamic penumbra model, a fast and accurate dynamic intensity-modulation plan is generated by establishing a timing model of blade speed and a variable penumbra model using direct machine parameter optimization.

Benefits of technology

It has achieved shortening treatment time, improving the efficiency and accuracy of dynamic intensity adjustment, and generating a fast and accurate dynamic intensity adjustment plan.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115317811B_ABST
    Figure CN115317811B_ABST
Patent Text Reader

Abstract

The present invention discloses a dynamic intensity modulation device and optimization method based on a dynamic penumbra model. The present invention relates to the field of medical physics. The specific optimization method comprises the following steps: firstly, a blade speed timing function is used to establish a timing model of blade speed at different times under the premise of the maximum speed and maximum acceleration allowed by the blade; secondly, in the process of dynamic intensity modulation optimization, a direct machine parameter optimization method is used to quickly optimize and generate a dynamic intensity modulation plan based on the established timing model of blade speed at different times and the dynamic penumbra model. The present invention can quickly optimize and obtain a dynamic intensity modulation plan that meets clinical requirements.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of medical physics and relates to a model for dynamic intensity modulation optimization problem, and specifically, to a dynamic intensity modulation device and optimization method based on a dynamic penumbra model. Background Art

[0002] With the vigorous development of radiotherapy technology, devices and methods for radiotherapy planning have been proposed and put into use one after another, especially the dynamic intensity modulation technology which has developed rapidly in recent years.

[0003] Dynamic intensity modulation means that the leaves of the multi-leaf collimator are in motion during radiation irradiation, and each pair of corresponding leaves of the multi-leaf collimator is used to scan the target area under computer control during radiation irradiation to adjust the intensity of the radiation field. The dynamic mode relies on the adjustment of the speed of each leaf movement and the irradiation dose rate to achieve intensity adjustment. For dynamic intensity modulation, since the treatment machine is always emitting beams during the movement of the leaves during treatment, the treatment beam emission time should be shortened as much as possible in order to shorten the total treatment implementation time; the current dynamic intensity modulation has the following defects: first, the irradiation time is long and the efficiency is low; second, the penumbra of the leaves is not considered, resulting in errors in the actual accuracy. Summary of the invention

[0004] Purpose of the invention: The purpose of the present invention is to provide a dynamic intensity modulation device and optimization method based on a dynamic penumbra model, to establish a timing model based on the blade speed at different times, and to implement a direct machine parameter optimization model based on this, so as to achieve fast and accurate optimization of the dynamic intensity modulation plan.

[0005] Technical solution: A dynamic intensity modulation device based on a dynamic penumbra model described in the present invention includes a user input device, a blade speed-related timing model device, a dynamic intensity modulation optimization device and a user output device;

[0006] Further, the user input device includes basic parameter information such as the user input organ tissue, maximum blade movement speed, maximum blade acceleration, dose change rate, gantry angle and plan optimization parameters;

[0007] Furthermore, the blade speed timing model device includes using a blade speed-time function to establish a blade speed timing model at different times; specifically, during the irradiation process of the radiation source, dose adjustment is achieved based on the relative movement of the grating blades, and under the premise of the maximum speed and maximum acceleration allowed by the blades, the blade speed timing model at different times is established by considering the non-uniform movement of the blades;

[0008] Furthermore, the dynamic intensity modulation optimization device includes obtaining a partial derivative of the objective function with respect to the blade motion time through a timing model of the blade velocity, and generating a dynamic intensity modulation plan by direct machine parameter optimization based on the established timing model of the blade velocity and a variable penumbra model of the position;

[0009] Furthermore, the user output device includes information such as the blade movement time and the corresponding movement position, the blade movement speed and the blade acceleration obtained by the blade through the optimization device;

[0010] The present invention also proposes a dynamic intensity modulation optimization method based on a dynamic penumbra model, which is composed of two modules: a blade speed timing model building module and a dynamic intensity modulation optimization module;

[0011] Wherein, the blade speed-time establishment module uses machine parameters as input parameters and uses blade variable speed time function to establish a time series model of blade speed;

[0012] The dynamic intensity modulation optimization module generates a dynamic intensity modulation plan based on the established blade speed timing model and direct machine parameter optimization based on the penumbra model.

[0013] Furthermore, the blade speed-time establishment module includes:

[0014] The blade velocity-time function is used to establish a time series model of the blade velocity at different times: During the irradiation process of the radiation source, the dose adjustment is realized based on the relative movement of the grating blades. Under the premise of the maximum speed and maximum acceleration allowed by the blade, the non-uniform motion of the blade is considered to establish a time series model of the blade velocity at different times. The calculation formula is as follows:

[0015] Formula (1) represents the relationship between the blade position x and the blade acceleration a(t) and velocity v(t):

[0016] x=γ(a(t),v(t)) (1)

[0017] Wherein, a(t) represents the acceleration of the blade at time t, v(t) represents the velocity of the blade at time t, and γ represents the mapping relationship between x and the blade acceleration a(t) and velocity v(t).

[0018] Formula (2) represents the acceleration constraint of the blade at time t:

[0019] |a(t)| max (2)

[0020] Among them, a max is the maximum acceleration of the blade motion.

[0021] Formula (3) represents the velocity constraint of the blade at time t:​

[0022] |v(t)| <v max (3)

[0023] Among them, v max is the maximum speed of the blade movement.

[0024] Furthermore, the execution steps of the dynamic intensity modulation optimization module are as follows:

[0025] Step 1: The user inputs the blade parameters, builds the blade model, and uses the Monte Carlo method to simulate the penumbra parameters at different positions: discretization is performed from the initial position of the blade movement to the final position, and the Monte Carlo method is simulated for each position. In this way, the penumbra parameters at different positions can be obtained;

[0026] Step 2: According to the penumbra parameters at different positions, a position-based variable penumbra model is established. The penumbra model takes into account the scattering of beams at different positions and corrects the position deviation;

[0027] Step 3: Use the direct machine parameter optimization method of the variable penumbra model to optimize the sub-field shape; when the end condition is met, it ends, otherwise continue to use the direct machine parameter device and method of the variable penumbra model to optimize the sub-field shape;

[0028] Further, in step 2, the establishment of the position-based variable penumbra model is: for penumbra parameters at different positions, calculations are performed at different relative positions of the blade and the radiation source, and when the radiation source is blocked by the blade to different degrees, a set of different intensity curves are formed, and the variable penumbra model can be obtained by data fitting;

[0029] In step 3, the direct machine parameter optimization method uses direct machine parameter optimization to adjust the blade position, uses the partial derivative of the objective function with respect to time, and uses the conjugate gradient method to calculate the minimum value of the objective function. The objective function calculation formula is as follows:

[0030] F=F(x)=min∑ i [∑ j M xij *M j (x)-D i ] 2 (4)

[0031]

[0032] Where v(t) and a(t) are from equation (2) and (3) respectively, F is the objective function value, x is the leaf position, 1≤i≤S, 1≤j≤N, where S and N are both positive integers, S represents the total number of organ points, N represents the total number of particle beams, and Mxij is the dose contribution of the jth particle beam to the ith organ point when the leaf position is x, D i is the prescription dose for the ith organ point, M j (x) is the dose intensity distribution after the jth particle beam passes through leaf x, t is the leaf movement time, It is a function of the blade's velocity, acceleration and time at position x.

[0033] The partial derivative of the objective function F in the above formula is calculated, and the partial derivative of the objective function with respect to the grating time is obtained through the timing model of the blade velocity. The partial derivative calculation formula is as follows:

[0034]

[0035] in, From formula (1), F is the objective function value, t is the leaf movement time, x is the leaf position, D is the dose contribution matrix of the organ point, and M is the dose intensity distribution.

[0036] Direct machine parameter optimization is used to adjust the blade position, and the partial derivative of the objective function with respect to time is used to calculate the minimum value of the objective function using the conjugate gradient method to quickly and accurately optimize the dynamic intensity modulation plan.

[0037] Beneficial effects: Compared with the prior art, the present invention establishes a timing model of blade speed through blade speed-time parameters, innovatively considers a position-based variable penumbra model, and adopts direct machine parameter constraint optimization, which can quickly deliver accurate dynamic intensity modulation plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a module diagram of a dynamic intensity modulation device based on a dynamic penumbra model of the present invention;

[0039] Figure 2 It is a structural flow chart of a dynamic intensity modulation optimization method based on a dynamic penumbra model of the present invention. DETAILED DESCRIPTION

[0040] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0041] The present invention discloses a dynamic intensity modulation device based on a dynamic penumbra model, comprising an interconnected user input device, a blade speed-related timing model device, a dynamic intensity modulation optimization device, and a user output device;

[0042] The user input device includes basic parameter information such as the user input organ tissue, maximum blade movement speed, maximum blade acceleration, dose change rate, gantry angle, and plan optimization parameters;

[0043] The blade speed timing model device uses machine parameters as input parameters and uses blade variable speed time function to establish the blade speed timing model;

[0044] The dynamic intensity modulation optimization device generates a dynamic intensity modulation plan by directly optimizing machine parameters based on the established blade speed timing model.

[0045] The user output device specifically includes: information such as the blade movement time and the corresponding movement position, the blade movement speed, and the blade acceleration obtained by the blade through the optimization device;

[0046] Furthermore, the blade speed timing model device specifically includes:

[0047] The blade velocity-time function is used to establish a time series model of the blade velocity at different times: During the irradiation process of the radiation source, the dose adjustment is realized based on the relative movement of the grating blades. Under the premise of the maximum speed and maximum acceleration allowed by the blade, the non-uniform motion of the blade is considered, and the time series model of the blade velocity at different times is established. The calculation formula is as follows:

[0048] Formula (1) represents the relationship between the blade position x and the blade acceleration a(t) and velocity v(t):

[0049] x=γ(a(t),v(t)) (1)

[0050] Wherein, a(t) represents the acceleration of the blade at time t, v(t) represents the velocity of the blade at time t, and γ represents the mapping relationship between x and the blade acceleration a(t) and velocity v(t).

[0051] Formula (2) represents the acceleration constraint of the blade at time t:

[0052] |a(t)| max (2)

[0053] Among them, a max is the maximum acceleration of the blade motion.

[0054] Formula (3) represents the velocity constraint of the blade at time t:

[0055] |v(t)| <v max (3)

[0056] Among them, v max is the maximum speed of the blade movement.

[0057] ​Furthermore, the dynamic intensity modulation optimization device specifically includes: obtaining partial derivatives of blade motion time through a timing model of blade speed, and generating a dynamic intensity modulation plan by direct machine parameter optimization based on the established timing model of blade speed.

[0058] The specific steps are as follows:

[0059] Step 1: The user inputs basic parameter information such as organ tissue, maximum blade movement speed, maximum blade acceleration, dose change rate, gantry angle, and plan optimization parameters;

[0060] Step 2: A timing model of blade velocity is established at different times using a blade velocity-time function. During the irradiation process of the radiation source, dose adjustment is achieved based on the relative motion of the grating blades. Under the premise of the maximum speed and maximum acceleration allowed by the blades, the non-uniform motion of the blades is considered to establish a timing model of blade velocity at different times.

[0061] Step 3, using the direct machine parameter optimization method of the variable penumbra model to optimize the sub-field shape; when the end condition is met, then end, otherwise continue to use the direct machine parameter device and method of the variable penumbra model to optimize the sub-field shape;

[0062] Step 2 includes the following specific steps: the time partial derivative of the objective function with respect to the grating is obtained through the time series model of the blade velocity. The objective function calculation formula is as follows:

[0063] F=F(x)=min∑ i [∑ j M xij *M j (x)-D i ] 2 (4)

[0064]

[0065] Where v(t) and a(t) are from equation (2) and (3) respectively, F is the objective function value, x is the leaf position, 1≤i≤S, 1≤j≤N, where S and N are both positive integers, S represents the total number of organ points, N represents the total number of particle beams, and M xij is the dose contribution of the jth particle beam to the ith organ point when the leaf position is x, D i is the prescription dose for the ith organ point, M j (x) is the dose intensity distribution after the jth particle beam passes through leaf x, t is the leaf movement time, It is a function of the blade's velocity, acceleration and time at position x.

[0066] The partial derivative of the objective function F in the above formula is calculated, and the partial derivative of the objective function with respect to the grating time is obtained through the timing model of the blade velocity. The partial derivative calculation formula is as follows:

[0067]

[0068] in, From formula (1), F is the objective function value, t is the leaf movement time, x is the leaf position, D is the dose contribution matrix of the organ point, and M is the dose intensity distribution.

[0069] Step 3 includes the following specific steps:

[0070] Step (3.1), publishing the control point sub-field at the corresponding blade motion position;

[0071] Step (3.2), optimizing the sub-field shape based on the timing model of blade speed and machine parameter constraints, and optimizing the weights of control points;

[0072] Step (3.3), determine the location where the generation can be executed; if it can be generated, output the optimized dynamic intensity modulation plan to meet the delivered radiation field until the end;

[0073] If not, return to step (3.2) and continue to optimize the weights of the control points.

[0074] Furthermore, the user output device specifically includes: information such as the blade movement time and the corresponding movement position, the blade movement speed, and the blade acceleration obtained by the blade through the optimization device;

[0075] The specific implementation example of the present invention is Figure 2 As stated, Figure 2 The present invention is a structural flow chart of a dynamic intensity modulation optimization method based on a dynamic penumbra model according to an embodiment of the present invention. A radiotherapy plan for improving the speed and effect of dynamic intensity modulation optimization based on a timing model of blade speed includes the following steps:

[0076] Step 1: Set the radiation field, optimize parameters and organ tissue information.

[0077] Specifically, the input image is a CT image of the prostate, which contains information about organs at risk, target organs, etc. At the same time, 9 treatment fields are set, the grid size is set to 0.5 cm, and the tungsten gate is selected to be non-isocentrically symmetric; the conditional constraints of the target area and organs at risk are as follows:

[0078]

[0079]

[0080] Step 2: Pencil beam matrix intensity optimization.

[0081] The input image with set parameters is subjected to organ matrix intensity optimization.

[0082] Step 3: Divide the intensity into sub-fields.

[0083] The sub-fields are segmented according to the intensity optimization results. The number of sub-fields under each radiation field is set in the present invention. When the volume of a sub-field is smaller than a certain size, the sub-field will be discarded and the sub-fields that meet the requirements will be retained.

[0084] Step 4: Generate dynamic intensity control points

[0085] The sub-fields that meet the requirements are converted into discrete distribution control points of dynamic intensity modulation, the speed and time model of the blades are calculated, and the distribution of the sub-fields is carried out according to the distribution of the sub-fields.

[0086] Step 5: distribute the control points corresponding to the moving positions of the multi-leaf grating.

[0087] Distribute the control points one by one to the corresponding leaf positions.

[0088] Step 6: Optimize the sub-field model based on the established blade velocity time series model, variable penumbra model of position, and machine parameters.

[0089] The dynamic intensity modulation control point information is input into the dynamic intensity modulation optimization, which needs to consider direct machine parameter optimization and blade speed time model, variable penumbra model, machine parameter optimization, etc. In the present invention, some parameters are set as follows: the minimum blade speed is 3mm / s, the maximum speed does not exceed 30mm / s, and the maximum acceleration is 12mm / s^2. When the optimization objective function value F decreases by less than the set threshold, the dynamic intensity modulation optimization ends.

[0090] Step 7: Output the planned dynamic intensity modulation plan.

[0091] See the dynamic IMRT optimization scheme that incorporates a variable penumbra model based on established blade velocity timing and position.

[0092] The present invention provides a dynamic intensity modulation device and optimization method based on a dynamic penumbra model, establishes a time series model based on blade speeds at different times, and implements a direct machine parameter optimization model based on the time series model, and optimizes quickly and accurately based on an existing dynamic intensity modulation plan. Therefore, the present invention has important practical value.

Claims

1. A dynamic intensity modulation device based on a dynamic penumbra model, characterized in that: It includes a user input device, a blade speed-related timing model device, a dynamic intensity modulation optimization device and a user output device; The user input device includes user input of organ tissue, maximum blade movement speed, maximum blade acceleration, dose change rate, gantry angle and plan optimization parameters; The blade speed time series model device includes establishing a blade speed time series model using a blade speed-time function; The dynamic intensity modulation optimization device includes obtaining the partial derivative of the objective function with respect to the blade motion time through the time series model of the blade speed; The user output device includes the blade movement time and the corresponding movement position, the blade movement speed and the blade acceleration information obtained by the blade through the optimization device; It also includes a model building module and a dynamic intensity modulation optimization module; The model building module uses the blade speed-time function to build a time series model of the blade speed at different times according to the optimized parameter information: Specifically, during the irradiation process of the radiation source, the dose adjustment is realized based on the relative movement of the grating blades. Under the premise of the maximum speed and maximum acceleration allowed by the blades, the non-uniform motion of the blades is considered, and a time series model of the blade speed at different times is established. The calculation formula is as follows: Formula (1) represents the relationship between the blade position x and the blade acceleration a(t) and velocity v(t): x=γ(a(t),v(t)) (1) Where a(t) represents the acceleration of the blade at time t, v(t) represents the velocity of the blade at time t, and γ represents the mapping relationship between the blade position x and the blade acceleration a(t) and velocity v(t); Formula (2) represents the acceleration constraint of the blade at time t: |a(t)|<a max (2) Among them, a max Indicates the maximum acceleration of the blade motion; Formula (3) represents the speed constraint of the blade at time t: |v(t)|<v max (3) Among them, v max Indicates the maximum speed of blade movement; The specific execution steps of the dynamic intensity modulation optimization module are as follows: Step (1): The user inputs blade parameters, constructs a blade model, and uses the Monte Carlo method to simulate a large number of photons passing through the constructed blade model to simulate penumbra parameters at different positions; Step (2): According to the penumbra parameters at different positions, a position-based variable penumbra model is established, where the penumbra model takes into account the scattering and transmission of the beam at different positions and corrects the position deviation; Specifically, the penumbra parameters at different positions are calculated at different relative positions of the blade and the radiation source. When the radiation source is blocked by the blade to different degrees, a set of different intensity curves are formed. Through data fitting, a variable penumbra model can be obtained. Step (3): using a direct machine parameter optimization method of a variable penumbra model to optimize the sub-field shape; when the end condition is met, the process ends; otherwise, the direct machine parameter device and method of a variable penumbra model are continued to be used to optimize the sub-field shape; Specifically, the blade position is adjusted by direct machine parameter optimization, the partial derivative of the objective function with respect to time is used, and the conjugate gradient method is used to calculate the minimum value of the objective function. The objective function calculation formula is as follows: F=F(x)=min∑ i [∑ j M xij *M j (x)-D i ] 2 (4) Where v(t) and a(t) are from equation (2) and (3) respectively, F is the objective function value, x is the leaf position, 1≤i≤S, 1≤j≤N, where S and N are both positive integers, S represents the total number of organ points, N represents the total number of particle beams, and M xij is the dose contribution of the jth particle beam to the ith organ point when the leaf position is x, D i is the prescribed dose for the ith organ point, M j (x) is the dose intensity distribution after the jth particle beam passes through leaf x, t is the leaf movement time, It is a function of the blade's velocity, acceleration and time at position x; The partial derivative of the objective function F in the above formula is calculated, and the partial derivative of the objective function with respect to the grating time is obtained through the timing model of the blade velocity. The partial derivative calculation formula is as follows: in, From formula (1), F is the objective function value, t is the leaf movement time, x is the leaf position, and D is the dose contribution matrix of the organ point.