Interactive lattice radiotherapy planning method and system based on functional image

Abstract

The invention discloses an interactive lattice radiotherapy planning method and system based on a functional image, and relates to the technical field of radiotherapy, and the method comprises the steps: obtaining a radiotherapy positioning CT image of a patient and a functional metabolism image after the space registration with the radiotherapy positioning CT image; based on the radiotherapy positioning CT image and a preset safety boundary parameter, determining a candidate area of lattice layout in the tumor target area; for each candidate position in the candidate region, adaptively calculating lattice parameters for the position based on the corresponding SUV value of the position in the functional metabolism image; determining a final lattice center set from the candidate positions through an iterative screening algorithm on the basis of the calculated lattice parameters of the candidate positions and preset geometric constraints and biological constraints, and generating a lattice target region set on the basis of the final lattice centers and the lattice diameters corresponding to the final lattice centers; and outputting the lattice target region set for dose calculation of a radiotherapy planning system. Individualized dosage improvement really based on tumor internal heterogeneity is achieved.

Description

Technical Field

[0001] This invention relates to the field of radiotherapy technology, and in particular to an interactive lattice radiotherapy planning method and system based on functional imaging. Background Technology

[0002] Radiation therapy is one of the main methods of cancer treatment. For large tumors, conventional radiation therapy usually delivers a uniform dose to the gross tumor volume (GTV). However, the tumor interior is not homogeneous; different regions exhibit significant biological heterogeneity in terms of cell proliferation activity, metabolic level, blood supply, and hypoxia. Regions with high metabolic activity and active proliferation are often relatively resistant to radiation, which is a major cause of local recurrence after treatment. Therefore, a uniform dose irradiation pattern that ignores this biological heterogeneity may fail to effectively eliminate radiation-resistant subregions within the tumor, affecting treatment efficacy.

[0003] Functional imaging techniques, such as fluorodeoxyglucose positron emission tomography-computed tomography (18F-FDG PET-CT), can non-invasively reflect the metabolic activity of tumors. The quantitative indicator, Standardized Uptake Value (SUV), is closely related to the glycolytic level of tumor cells; high SUV regions typically indicate high proliferative activity and potential radioresistance. Utilizing the biometabolic information provided by PET-CT to guide the optimization of radiotherapy dose distribution has become an important research direction for improving treatment precision.

[0004] Lattice radiation therapy (LRT), also known as spatially fractionated radiation therapy (SFRT), is an emerging radiotherapy technique. Its principle involves arranging multiple discrete high-dose "lattice" points (or "peaks") within the tumor volume according to a specific spatial pattern, while the regions between these lattice points receive relatively lower doses, thus creating a highly non-uniform dose distribution throughout the tumor. This "peak-valley" dose pattern is thought to potentially induce unique radiobiological effects, helping to overcome problems such as tumor heterogeneity and hypoxia. In existing technologies, the placement of lattice points is often based on simple geometric rules, such as pre-defining the position and size of lattice points within the tumor at fixed intervals (e.g., cubic or spherical grids). While this geometric placement method is easy to implement, it relies entirely on the morphological contours of the tumor and fails to take into account the intratumoral biological functional heterogeneity information revealed by PET-CT, potentially leading to insufficient or missed dose boosting to radiation-resistant subregions.

[0005] To incorporate bioinformation into treatment planning, some attempts have been made in existing technologies, such as using high-metabolic regions as one of the optimization targets, or setting simple bioweighting factors in dot placement algorithms. However, these methods are generally limited, perhaps only optimizing the location of lattice points, or imposing biosignal-based constraints on only a single parameter (such as the number of lattice points), failing to dynamically and quantitatively map and transform continuous biometabolic information (such as SUV values) into key physical parameters that determine dose distribution (such as the individualized diameter of each lattice and the minimum spacing between lattices).

[0006] Furthermore, existing automated placement algorithms often operate as a closed "black box." Physicians input parameters before algorithm execution but find it difficult to intervene in real-time during the calculation process, and cannot quickly and intuitively assess key characteristics of the placement plan (such as lattice number, total volume, spatial distribution, and proximity to hazardous structures) after the results are generated. Due to the complexity of medical imaging and the diversity of clinical situations, fully automated algorithms may produce placement plans that do not meet clinical safety standards. For example, if a high SUV region happens to have important blood vessels or bronchi passing through it, the algorithm may still place lattice points there, posing a potential risk. Physicians, lacking effective interactive tools and real-time feedback information, find it difficult to easily correct and optimize such plans.

[0007] In summary, existing lattice radiotherapy planning technologies still have significant shortcomings in terms of how to fully utilize the bioinformation of functional imaging to achieve individualized and adaptive site placement, and how to effectively integrate the professional judgment and real-time interaction of clinicians. Summary of the Invention

[0008] This invention aims to address the technical problems of existing LRT lattice layouts that rely on geometric rules, neglect biological heterogeneity, and lack clinical physician interaction and real-time feedback.

[0009] In a first aspect, to address the aforementioned technical problems, the present invention provides an interactive lattice radiotherapy planning method based on functional imaging, comprising the following steps: S1. Acquire the patient's radiotherapy localization CT image and a functional metabolic image spatially registered with the radiotherapy localization CT image; the functional metabolic image includes SUV values ​​reflecting metabolic activity; S2. Based on the radiotherapy positioning CT image and preset safety boundary parameters, determine the candidate region for lattice layout within the tumor target area; S3. For each candidate location within the candidate region, lattice parameters are adaptively calculated for that location based on the SUV value corresponding to that location in the functional metabolic image; the lattice parameters include at least the lattice diameter, which is positively correlated with the SUV value, and the minimum inter-lattice spacing, which is negatively correlated with the SUV value. S4. Based on the calculated lattice parameters of each candidate position, as well as the preset geometric and biological constraints, an iterative screening algorithm is used to determine the final set of lattice centers from the candidate positions, and a set of lattice target regions is generated based on the final lattice centers and their corresponding lattice diameters. S5. Output the set of lattice target regions for dose calculation in the radiotherapy planning system.

[0010] Furthermore, the spatial registration in S1 specifically includes: Acquire CT sub-images acquired simultaneously with the functional metabolic images; Perform global rigid registration between the CT sub-image and the radiotherapy localization CT image; Based on the results of the rigid registration, nonlinear deformation registration is then performed to obtain the coordinate transformation relationship from the functional metabolic image space to the radiotherapy localization CT image space. The functional metabolic image is resampled using the coordinate transformation relationship to align it spatially with the radiotherapy localization CT image.

[0011] Furthermore, the determination of candidate regions in S2 includes: The tumor target area is obtained from the radiotherapy localization CT image; The tumor target area is shrunk within a designated first safety boundary to obtain a first region; Obtain at least one organ at risk and extend the organ at risk beyond a designated second safety boundary; The second region is obtained by subtracting the expanded volume of the endangered organ from the first region. The second region is uniformly shrunk inward by a specified distance to obtain the candidate region; wherein the specified distance is determined based on a predefined minimum lattice spacing.

[0012] Furthermore, the step of adaptively calculating the lattice parameters in S3 includes: Extract the SUV values ​​for all locations within the candidate region and normalize them to... The interval is used to obtain the normalized SUV value; For any of the candidate positions, the corresponding lattice diameter and minimum interlattice spacing are calculated based on its normalized SUV value using the following functional relationship:

[0013]

[0014] In the formula, Indicates the diameter of the lattice; This represents the minimum interlattice spacing; , These are the preset minimum and maximum lattice diameters, respectively; , These are the preset minimum and maximum lattice spacings, respectively; Indicates candidate position The normalized SUV value.

[0015] Furthermore, the iterative filtering algorithm in S4 is specifically as follows: All candidate locations within the candidate region are sorted in descending order according to their normalized SUV values; Initialize an empty list of selected lattices; The candidate positions are traversed sequentially. For the currently traversed candidate position, the three-dimensional spatial distance between it and each existing lattice in the selected lattice list is calculated. If the three-dimensional spatial distance satisfies the constraint conditions, the current candidate position is added as the new lattice center to the list of selected lattices.

[0016] Furthermore, the constraint condition is: three-dimensional spatial distance ≥ (biological constraint + geometric constraint); Wherein, the biological constraint equals ; The geometric constraints are equal to .

[0017] Furthermore, S4 also includes target region boundary constraints: After determining the current candidate position as the lattice center, a spherical lattice target area is generated with the lattice center as the center and the corresponding lattice diameter as the diameter. Determine whether the spherical lattice target area exceeds the preset allowable layout area; if it does, adjust the diameter of the spherical lattice target area to the preset minimum lattice diameter.

[0018] Furthermore, an interaction step is included before S2 or after S4: The system receives at least one definition of a lattice avoidance region input by the user, wherein the lattice avoidance region is used to identify a three-dimensional spatial volume in which lattice placement is prohibited. When determining the candidate region in step S2, the lattice avoidance region is excluded from the candidate region; And / or, during the iterative screening algorithm in S4, exclude candidate positions whose centers are located within the lattice avoidance region.

[0019] Furthermore, following S5, an evaluation feedback step is also included: Generate and display quantitative evaluation parameters related to the set of lattice target regions; wherein the quantitative evaluation parameters include at least one of the following: total number of lattices, total lattice volume, percentage of lattice volume to tumor target region, center coordinates of each lattice, SUV value of each lattice, and diameter of each lattice. Receive user instructions to adjust the lattice generation parameters or the lattice avoidance region based on the quantitative evaluation parameters; Based on the adjustment instructions, S3 and / or S4 are re-executed to generate the adjusted set of lattice target regions.

[0020] A second aspect of the present invention provides an interactive lattice radiotherapy planning system based on functional imaging for implementing the method, comprising: The image acquisition and preprocessing module is used to acquire the patient's radiotherapy positioning CT image and the functional metabolic image spatially registered with the radiotherapy positioning CT image; The candidate region determination module is used to determine the candidate regions of the lattice layout within the tumor target area based on the radiotherapy positioning CT image and preset safety boundary parameters. The lattice parameter adaptive calculation module is used to adaptively calculate lattice parameters for each candidate location within the candidate region based on the SUV value corresponding to that location in the functional metabolism image. The lattice parameters include at least the lattice diameter and the minimum interlattice spacing. The lattice screening and generation module is used to determine the final set of lattice centers from the candidate positions based on the calculated lattice parameters of each candidate position, as well as preset geometric constraints and biological constraints, through an iterative screening algorithm, and to generate a set of lattice target regions based on the final lattice centers and their corresponding lattice diameters. The output module is used to output the set of lattice target regions.

[0021] Furthermore, it also includes: An interactive lattice avoidance region definition module is used to receive a user-defined lattice avoidance region and provide the lattice avoidance region to the candidate region determination module and / or the lattice filtering and generation module to exclude the placement of lattices within the lattice avoidance region. The real-time feedback module is used to display the quantitative evaluation parameters of the lattice target region set on the graphical user interface after the lattice target region set is generated.

[0022] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: This invention dynamically and adaptively maps quantitative biometabolic information (SUV value) from functional imaging into key physical parameters that determine the lattice radiotherapy dose distribution, namely the lattice diameter and the minimum inter-lattice spacing, achieving truly individualized dose enhancement based on tumor intratumoral heterogeneity. At the same time, by introducing an interactive lattice avoidance region definition and real-time protocol feedback module, the physician's clinical experience and judgment are deeply integrated into the automated process, effectively avoiding the risk of automatic spotting on key anatomical structures. While improving the accuracy of treatment for radioresistant subregions, it ensures the clinical safety and operability of the protocol. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart disclosed in the present invention; Figure 2 This is the system interaction interface disclosed in the embodiments of the present invention; Figure 3 This is the real-time feedback interface for lattice parameters disclosed in the embodiments of the present invention; Figure 4 This is a lattice arrangement diagram based on functional image drawing disclosed in an embodiment of the present invention; Figure 5 This is a schematic diagram comparing a lattice arrangement diagram drawn based on a functional image and a lattice arrangement diagram drawn based on a script, as disclosed in an embodiment of the present invention. Detailed Implementation

[0025] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] This invention aims to provide an interactive lattice radiotherapy planning method and system based on functional imaging, addressing the problems of existing lattice radiotherapy planning that rely on geometric placement, ignore tumor biological heterogeneity, and lack effective clinician intervention and real-time feedback. The invention will be described in detail below with specific implementation steps.

[0027] S1. Multimodal medical image acquisition, registration and preprocessing: Acquire the patient's radiotherapy-guided CT images and functional metabolic images spatially registered with the radiotherapy-guided CT images; wherein, the functional metabolic images contain SUV values ​​reflecting metabolic activity.

[0028] The purpose of this step is to obtain spatially aligned structural and functional images that can be used for precise guidance.

[0029] First, acquire CT sub-images acquired on the same machine as the functional metabolic images. Specifically, acquire the patient's baseline 18F-FDG PET-CT image containing functional metabolic information and the radiotherapy localization CT image acquired in the patient's treatment position.

[0030] PET-CT images consist of a set of PET images (reflecting metabolic activity) and a set of CT images acquired in the same scan (hereinafter referred to as "CT sub-images"). Since CT sub-images and PET images are completed in a single scan session, they share the same DICOM coordinate system reference, thus possessing inherent spatial alignment.

[0031] To achieve accurate mapping of PET image metabolic information into the radiotherapy positioning CT space, image spatial registration is required, which includes global rigid registration and nonlinear deformation registration. Rigid registration aims to correct overall translational and rotational deviations between images caused by differences in patient positioning and bed position; nonlinear deformation registration is used to correct local deformations of organs or tissues themselves (such as those caused by respiration, intestinal peristalsis, etc.). Specifically: Rigid registration: Global rigid registration is performed using the radiotherapy localization CT image as the fixed image and the CT sub-image as the floating image. A six-degree-of-freedom (three translation parameters and three rotation parameters) rigid body transformation model is adopted. Mutual information is used as a similarity measure for optimization. Mutual information measures the statistical dependence between the gray-level distributions of two images; the mutual information value is maximized when the spatial positions of the two images are optimally aligned. The formula for calculating mutual information is as follows:

[0032] In the formula, and These represent radiotherapy localization CT images and CT sub-images, respectively. and Representing images respectively and images grayscale values ​​in; It is the joint probability distribution of the two; and Each is its own marginal probability distribution.

[0033] The simplex method is then used to iteratively optimize the transformation parameters until the optimal rigid transformation parameters that maximize mutual information are found. The simplex method is a direct search iterative method that does not require gradient information; it optimizes the objective function by dynamically adjusting the positions of the simplex vertices. In rigid registration, it efficiently searches for six-degree-of-freedom rigid body transformation parameters (three translations and three rotations), progressively approximating the optimal solution that maximizes mutual information. This algorithm avoids complex derivative calculations and is particularly suitable for optimization problems involving non-smooth similarity measures such as mutual information. Ultimately, it reliably finds the optimal transformation parameters that align the floating image with the fixed image space.

[0034] Nonlinear Deformation Registration: Based on the results of rigid registration, a B-spline free deformation model is used for local nonlinear registration. This model generates a smooth and continuous deformation field by adjusting the positions of control grid points, which can flexibly simulate complex deformations of local tissues and correct for tissue and organ deformations caused by respiration, peristalsis, etc. Normalized cross-correlation is used as a similarity measure in this stage, which has good robustness to local linear changes in grayscale. The formula for normalized cross-correlation is as follows:

[0035] In the formula, and Images and At voxel points The grayscale value at that location; and These are the average gray values ​​within local windows of the two images, respectively.

[0036] The L-BFGS-B optimization algorithm is then used to solve for the B-spline transformation parameters to obtain the accurate deformation field. L-BFGS-B is a memory-efficient quasi-Newton optimization algorithm designed specifically for large-scale optimization problems with boundary constraints. It approximates the Hessian matrix by storing the gradient information of the most recent iterations, avoiding the high cost of directly calculating and storing the complete matrix. In nonlinear deformation registration, it can efficiently optimize the B-spline control point parameters while ensuring that the transformation parameters are within a reasonable boundary range. This algorithm has a fast convergence speed and is particularly suitable for handling complex optimization problems such as B-spline deformation fields with a large number of parameters and boundary constraints.

[0037] By combining rigid transformation parameters with a nonlinear deformation field, a complete coordinate transformation from the original PET image space to the radiotherapy localization CT space is obtained. This transformation is then applied to resample the original PET image, aligning it with the radiotherapy localization CT image space, resulting in a transformed PET image that is completely consistent with the localization CT space.

[0038] In a further step, to suppress noise and preserve the main structure in the PET images, a three-dimensional Gaussian filter preprocessing is performed on the resampled PET images. Specifically, a three-dimensional Gaussian function is used to define the filter kernel, and its mathematical expression is as follows:

[0039] In the formula, These are the coordinates of the pixel relative to the center of the filter kernel; It is the standard deviation, and a key parameter of the Gaussian distribution, which determines the degree of dispersion of the distribution. The larger the value, the wider the effective range of the Gaussian kernel, and the more significant the image smoothing effect; conversely, the smaller the value, the wider the effective range of the Gaussian kernel. The smaller the value, the more concentrated the kernel, and the weaker the smoothing effect. Convolving the image with the Gaussian kernel generated by this function effectively suppresses noise while preserving the overall structural information of the image. The preprocessed PET image will be used as the input for subsequent bio-guided standardized functional imaging.

[0040] S2. Definition of clinical parameters and generation of candidate biological target areas: Based on radiotherapy localization CT images and preset safety boundary parameters, candidate regions for lattice layout are determined within the tumor target area.

[0041] This step pre-determines safety boundary parameters based on clinical experience and defines a safe area for lattice placement based on anatomical structure.

[0042] First, the physician delineates the tumor target volume (GTV) on the radiotherapy localization CT image. The system predefines or allows physicians to adjust the following key parameters, with example values ​​shown in Table 1 below: Table 1

[0043] The determination of candidate regions includes the following steps: First, the tumor target volume (GTV) is delineated on the radiotherapy localization CT image; then, the tumor target volume is recessed inward along six directions (anterior, posterior, left, right, head, and foot) to a predetermined safety boundary.

[0044] Next, identify nearby organs at risk and extend them outward along six predetermined safety boundaries.

[0045] Furthermore, by subtracting the volume of all outward-expanded organs at risk from the inward-shrinking tumor target area, a preliminary safe deployment area is obtained, denoted as GTV-Target.

[0046] Finally, to ensure that the interlattice maintains a minimum safe distance, the GTV-Target is further and uniformly moved inward by a certain distance. The final volume obtained is the candidate region for lattice layout.

[0047] S3. Normalization of Biological Metabolic Information and Adaptive Calculation of Lattice Parameters: For each candidate location within the candidate region, lattice parameters are adaptively calculated for that location based on the SUV value corresponding to that location in the functional metabolic image. The lattice parameters include at least the lattice diameter, which is positively correlated with the SUV value, and the minimum inter-lattice spacing, which is negatively correlated with the SUV value.

[0048] The core of this step is to normalize the biological metabolic information (standardized uptake value, i.e. SUV value) of the candidate region and adaptively calculate the lattice parameters for each voxel based on this.

[0049] The steps for adaptively calculating lattice parameters include: First, extract all voxels within the candidate region. The standardized intake value (SUV value) is denoted as To eliminate absolute numerical differences, the SUV value of each voxel is normalized to... The normalized SUV value is obtained by dividing the interval; the normalization formula is as follows:

[0050] In the formula, and These represent the minimum and maximum SUV values ​​of all voxels within the candidate region, respectively.

[0051] Next, for any candidate position, based on its normalized SUV value, its corresponding lattice diameter and minimum interlattice spacing are calculated using the following functional relationship:

[0052]

[0053] In the formula, Indicates the lattice diameter; Indicates the minimum inter-lattice spacing; , These are the preset minimum and maximum lattice diameters, respectively; , These are the preset minimum and maximum lattice spacings, respectively; Indicates candidate position The normalized SUV value.

[0054] This design allows for the activation of high-metabolic regions (high... It automatically generates lattices with larger diameters and smaller spacing, thereby achieving targeted enhancement of the dosage.

[0055] S4. Lattice screening and optimization generation with integrated interactive constraints: Based on the calculated lattice parameters of each candidate position, as well as the preset geometric and biological constraints, the set of final lattice centers is determined from the candidate positions through an iterative screening algorithm, and a set of lattice target areas is generated based on the final lattice centers and their corresponding lattice diameters.

[0056] This step relies on the candidate regions defined in step S2 and the adaptive lattice parameters calculated in step S3 to generate the final set of lattice target regions (GTV-LAT) through an iterative screening algorithm.

[0057] The iterative screening algorithm is as follows: S41. Select all voxels within the candidate region. According to its normalized standard intake value Sort the data in descending order. This operation aims to ensure that the algorithm prioritizes candidate sites with the highest metabolic activity and the greatest need for increased doses.

[0058] S42. Initialize an empty list of selected lattices. The first element of the self-sorted list (i.e.,...) The iteration process begins with the voxel with the highest intake value.

[0059] S43. Traverse the sorted candidate positions in order. For the currently traversed candidate position, calculate the three-dimensional spatial distance between it and each existing lattice in the selected lattice list.

[0060] S44. When the three-dimensional spatial distance satisfies the constraint condition, the current candidate position is added to the selected lattice list as the new lattice center. The constraint condition is: Three-dimensional spatial distance ≥ (biological constraints + geometric constraints) Biological constraints equal to Biological constraints ensure that the lattice spacing can be adaptively adjusted according to the biological activity of the region.

[0061] Geometric constraints equal to . This ensures that the lattice diameter can be adaptively adjusted according to the bioactivity of its region.

[0062] Further solutions also include target boundary constraints: S45. After determining the current candidate position as the lattice center, a spherical lattice target area is generated with the corresponding lattice diameter as the center.

[0063] S46. Determine whether the spherical lattice target area exceeds the preset allowable layout area; if it does, adjust the diameter of the spherical lattice target area to the preset minimum lattice diameter.

[0064] Finally, the coordinates of the center of the spherical lattice target area and the corresponding lattice diameter are updated to the "Selected Lattice" list.

[0065] In a further embodiment, an interactive step is included before step S2 or after step S4, specifically: a. Receive at least one definition of a lattice avoidance region from user input. The lattice avoidance region is used to identify the three-dimensional spatial volume where lattice placement is prohibited.

[0066] b. When determining the candidate region in step S2, exclude the lattice avoidance region from the candidate region; and / or when performing the iterative screening algorithm in S4, exclude the candidate position whose center is located within the lattice avoidance region.

[0067] S5, Interactive Feedback, Protocol Evaluation and Output: Outputs a set of lattice target regions for dose calculation in the radiotherapy planning system.

[0068] The purpose of this step is to visualize the algorithm results, provide physicians with key decision-making data and an interface for iterative optimization, thereby building a closed-loop workflow.

[0069] Specifically, it includes: S51, 3D visualization rendering: The GTV-LAT set (multiple spheres) finally generated in step S4 is rendered and displayed in the system's 3D image view (e.g., PET-CT fusion view) so that physicians can intuitively evaluate the spatial distribution of the lattice.

[0070] S52. Real-time Feedback of Scheme Parameters: The quantitative evaluation parameters of this scheme are displayed in real time in the sidebar or pop-up window, including overall information and individual information. The overall information includes the total number of crystal lattices N and the total crystal volume. Information such as the percentage of lattice volume to GTV is included. Individual information includes the ID of each lattice, the three-dimensional coordinates of its center point (IEC 61217 coordinate system), and the corresponding... Details such as values ​​and lattice diameter are listed.

[0071] S53. Physician Decision-Making and Interaction Optimization: Physicians combine visualization effects and quantitative data for evaluation. If dissatisfied, the following steps can be performed to trigger the system to re-execute step S4: 1) Adjust the shape or position of the lattice evasion region.

[0072] 2) Adjust the adaptive lattice parameters in step S3.

[0073] S54. Final Output: After the physician confirms the plan, the system will directly export the final GTV-LAT structure set (containing the geometric definitions of all lattice spheres) to the radiotherapy treatment planning system (TPS) in DICOM RT Structure Set format or other compatible formats, as the physical target input for subsequent lattice radiotherapy dose calculation and optimization.

[0074] In a further embodiment, after step S5, an evaluation feedback step is also included: First, generate and display quantitative evaluation parameters related to the lattice target region set; wherein, the quantitative evaluation parameters include at least one of the following: total number of lattices, total lattice volume, percentage of lattice volume to tumor target region, center coordinates of each lattice, SUV value of each lattice, and diameter of each lattice.

[0075] Then, it receives user instructions to adjust the lattice generation parameters or lattice avoidance regions based on the quantitative evaluation parameters.

[0076] Finally, based on the adjustment instructions, steps S3 and / or S4 are re-executed to generate the adjusted lattice target region set.

[0077] The present invention also provides an interactive lattice radiotherapy planning system based on functional imaging to implement the above method, which mainly includes an image acquisition and preprocessing module, a candidate region determination module, a lattice parameter adaptive calculation module, a lattice screening and generation module, and an output module.

[0078] The image acquisition and preprocessing module is used to acquire the patient's radiotherapy-guided CT images and functional metabolic images spatially registered with the radiotherapy-guided CT images.

[0079] The candidate region determination module is used to determine candidate regions for lattice layout within the tumor target area based on radiotherapy positioning CT images and preset safety boundary parameters.

[0080] The adaptive lattice parameter calculation module is used to adaptively calculate lattice parameters for each candidate location within a candidate region, based on the SUV value corresponding to that location in the functional metabolic image. The lattice parameters include at least the lattice diameter and the minimum inter-lattice spacing.

[0081] The lattice screening and generation module is used to determine the final set of lattice centers from the candidate positions based on the calculated lattice parameters of each candidate position, as well as the preset geometric and biological constraints, through an iterative screening algorithm, and to generate a set of lattice target regions based on the final lattice centers and their corresponding lattice diameters.

[0082] The output module is used to output the set of lattice target regions.

[0083] Optionally, the system also includes an interactive lattice avoidance region definition module and a real-time feedback module.

[0084] The interactive lattice avoidance region definition module receives user-defined lattice avoidance regions and provides them to the candidate region determination module and / or the lattice screening and generation module to exclude lattice placement within these regions. Examples include major blood vessels, bronchi, or other critical structures deemed unsuitable for placement by physicians within the tumor target area.

[0085] The real-time feedback module is used to immediately display all key information of the generated lattice scheme on the graphical user interface after the lattice target area set is generated (step S4). For example, the total number of lattices (N), the total lattice volume (N...), etc. Overall information such as lattice number (ID) and three-dimensional coordinates of the center point (IEC61217 coordinate system). Individual information such as value and lattice diameter.

[0086] To verify the effectiveness of this technical solution, in a specific example, the effect achieved by this technical solution is as follows: Figure 2-5 As shown.

[0087] Figure 2 The graphical user interface of the interactive lattice radiotherapy planning system based on functional imaging provided in this embodiment of the invention is illustrated. The interface integrates and displays the patient's PET functional images and CT anatomical images. Physicians can delineate target areas, define avoidance zones, adjust parameters, and view the three-dimensional reconstructed tumor and organ-at-risk structures in real time, demonstrating the system's interactivity.

[0088] Figure 3 This is a schematic diagram of the real-time feedback interface for lattice parameters in an embodiment of the present invention. The interface automatically pops up after the lattice generation algorithm is executed, clearly displaying overall information such as the total number of lattices, total volume, and percentage of GTV in the form of tables and statistical panels, as well as individual details such as the center coordinates, normalized SUV value, and diameter of each lattice, thus achieving transparency and evaluability of key information in the scheme.

[0089] Figure 4 A three-dimensional schematic diagram of the lattice arrangement generated using this method is presented. The circles in the figure (circle A in the figure) represent the lattice target area (GTV-LAT) adaptively calculated based on the SUV value in the PET-CT functional image. Its size and spatial arrangement are not uniform, but are matched with the metabolically active area (high SUV) of the underlying image, which intuitively reflects the individualized lattice layout effect guided by biological metabolic information.

[0090] Figure 5For comparison, the diagram visually contrasts the lattice arrangement produced by the proposed biological guidance method (circle A in the figure) with that produced by the traditional dot-matrix method based on a fixed geometric script (circle B in the figure). It can be seen that the lattice generated by the proposed method is more concentrated in high-metabolic regions, and the lattice size exhibits a gradient variation; while the lattice generated by the traditional method is uniformly distributed and of consistent size, failing to reflect the biological heterogeneity within the tumor, thus highlighting the technical advantage of the proposed method in achieving precise dose enhancement.

[0091] This scheme quantitatively transforms SUV metabolic information into two core physical parameters: lattice diameter and spacing, enabling precise, targeted, and controllable dose escalation in the resistance subregion.

[0092] Secondly, the interactive lattice avoidance region definition module gives physicians ultimate control, allowing them to manually reject the algorithm's placement of points in high-risk areas (such as large blood vessels) within the GTV, greatly improving clinical safety and solving the blind spots of automated algorithms.

[0093] Furthermore, the real-time feedback module makes the lattice generation process transparent, allowing physicians to immediately obtain key evaluation indicators such as the number and volume of lattices, facilitating rapid iteration and optimization of the plan, and realizing a closed loop between the algorithm and physician decision-making.

[0094] Clearly, this invention provides a complete technical solution that integrates algorithms, interactive interfaces, and information feedback, which can be seamlessly embedded into existing TPS workflows and has extremely high clinical translational value and feasibility.

[0095] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An interactive lattice radiotherapy planning method based on functional imaging, characterized in that, Includes the following steps: S1. Acquire the patient's radiotherapy localization CT image and a functional metabolic image spatially registered with the radiotherapy localization CT image; the functional metabolic image includes SUV values ​​reflecting metabolic activity; S2. Based on the radiotherapy positioning CT image and preset safety boundary parameters, determine the candidate region for lattice layout within the tumor target area; S3. For each candidate location within the candidate region, lattice parameters are adaptively calculated for that location based on the SUV value corresponding to that location in the functional metabolic image; the lattice parameters include at least the lattice diameter, which is positively correlated with the SUV value, and the minimum inter-lattice spacing, which is negatively correlated with the SUV value. S4. Based on the calculated lattice parameters of each candidate position, as well as the preset geometric and biological constraints, an iterative screening algorithm is used to determine the final set of lattice centers from the candidate positions, and a set of lattice target regions is generated based on the final lattice centers and their corresponding lattice diameters. S5. Output the set of lattice target regions for dose calculation in the radiotherapy planning system.

2. The interactive lattice radiotherapy planning method based on functional imaging according to claim 1, characterized in that, The spatial registration in S1 specifically includes: Acquire CT sub-images acquired simultaneously with the functional metabolic images; Perform global rigid registration between the CT sub-image and the radiotherapy localization CT image; Based on the results of the rigid registration, nonlinear deformation registration is then performed to obtain the coordinate transformation relationship from the functional metabolic image space to the radiotherapy localization CT image space. The functional metabolic image is resampled using the coordinate transformation relationship to align it spatially with the radiotherapy localization CT image.

3. The interactive lattice radiotherapy planning method based on functional imaging according to claim 1, characterized in that, The determination of candidate regions in S2 includes: The tumor target area is obtained from the radiotherapy localization CT image; The tumor target area is shrunk within a designated first safety boundary to obtain a first region; Obtain at least one organ at risk and extend the organ at risk beyond a designated second safety boundary; The second region is obtained by subtracting the expanded volume of the endangered organ from the first region. The second region is uniformly shrunk inward by a specified distance to obtain the candidate region; wherein the specified distance is determined based on a predefined minimum lattice spacing.

4. The interactive lattice radiotherapy planning method based on functional imaging according to claim 1, characterized in that, The step of adaptively calculating lattice parameters in S3 includes: Extract the SUV values ​​for all locations within the candidate region and normalize them to... The interval is used to obtain the normalized SUV value; For any of the candidate positions, the corresponding lattice diameter and minimum interlattice spacing are calculated based on its normalized SUV value using the following functional relationship: In the formula, Indicates the diameter of the lattice; This represents the minimum interlattice spacing; , These are the preset minimum and maximum lattice diameters, respectively; , These are the preset minimum and maximum lattice spacings, respectively; Indicates candidate position The normalized SUV value.

5. The interactive lattice radiotherapy planning method based on functional imaging according to claim 4, characterized in that, The iterative filtering algorithm in S4 is as follows: All candidate locations within the candidate region are sorted in descending order according to their normalized SUV values; Initialize an empty list of selected lattices; The candidate positions are traversed sequentially. For the currently traversed candidate position, the three-dimensional spatial distance between it and each existing lattice in the selected lattice list is calculated. If the three-dimensional spatial distance satisfies the constraint conditions, the current candidate position is added as the new lattice center to the list of selected lattices.

6. The interactive lattice radiotherapy planning method based on functional imaging according to claim 5, characterized in that, The constraint condition is: three-dimensional spatial distance ≥ (biological constraint + geometric constraint); Wherein, the biological constraint equals ; The geometric constraints are equal to .

7. The interactive lattice radiotherapy planning method based on functional imaging according to claim 5, characterized in that, S4 also includes target region boundary constraints: After determining the current candidate position as the lattice center, a spherical lattice target area is generated with the lattice center as the center and the corresponding lattice diameter as the diameter. Determine whether the spherical lattice target area exceeds the preset allowable deployment area; If the diameter exceeds the limit, the diameter of the spherical lattice target area will be adjusted to the preset minimum lattice diameter.

8. The interactive lattice radiotherapy planning method based on functional imaging according to claim 1, characterized in that, An interactive step is also included before S2 or after S4: The system receives at least one definition of a lattice avoidance region input by the user, wherein the lattice avoidance region is used to identify a three-dimensional spatial volume in which lattice placement is prohibited. When determining the candidate region in step S2, the lattice avoidance region is excluded from the candidate region; And / or, during the iterative screening algorithm in S4, exclude candidate positions whose centers are located within the lattice avoidance region.

9. The interactive lattice radiotherapy planning method based on functional imaging according to claim 8, characterized in that, Following S5, an evaluation feedback step is also included: Generate and display quantitative evaluation parameters related to the set of lattice target regions; wherein the quantitative evaluation parameters include at least one of the following: total number of lattices, total lattice volume, percentage of lattice volume to tumor target region, center coordinates of each lattice, SUV value of each lattice, and diameter of each lattice. Receive user instructions to adjust the lattice generation parameters or the lattice avoidance region based on the quantitative evaluation parameters; Based on the adjustment instructions, S3 and / or S4 are re-executed to generate the adjusted set of lattice target regions.

10. An interactive lattice radiotherapy planning system based on functional imaging that implements the method as described in any one of claims 1-9, characterized in that, include: The image acquisition and preprocessing module is used to acquire the patient's radiotherapy positioning CT image and the functional metabolic image spatially registered with the radiotherapy positioning CT image; The candidate region determination module is used to determine the candidate regions of the lattice layout within the tumor target area based on the radiotherapy positioning CT image and preset safety boundary parameters. The lattice parameter adaptive calculation module is used to adaptively calculate lattice parameters for each candidate location within the candidate region based on the SUV value corresponding to that location in the functional metabolism image. The lattice parameters include at least the lattice diameter and the minimum interlattice spacing. The lattice screening and generation module is used to determine the final set of lattice centers from the candidate positions based on the calculated lattice parameters of each candidate position, as well as preset geometric constraints and biological constraints, through an iterative screening algorithm, and to generate a set of lattice target regions based on the final lattice centers and their corresponding lattice diameters. The output module is used to output the set of lattice target regions.

11. The interactive lattice radiotherapy planning system based on functional imaging according to claim 10, characterized in that, Also includes: An interactive lattice avoidance region definition module is used to receive a user-defined lattice avoidance region and provide the lattice avoidance region to the candidate region determination module and / or the lattice filtering and generation module to exclude the placement of lattices within the lattice avoidance region. A real-time feedback module is used to display the quantitative evaluation parameters of the lattice target region set on a graphical user interface after the lattice target region set is generated.

Images