A method and system for recommending gold marker implantation locations

Abstract

This application provides a method and system for recommending gold marker implantation sites, relating to the field of image-guided radiotherapy technology. The method includes acquiring a patient's CT image data; determining several gold marker implantation points based on the CT image data; combining the gold marker implantation points according to preset aggregation rules to construct several gold marker point sets; evaluating the three-dimensional geometric stability index and the two-dimensional observable stability index of each gold marker point set; combining the preset evaluation rules and the geometric stability index and observable stability index corresponding to each gold marker point set to select the optimal gold marker point set from all gold marker point sets; using the gold marker implantation points in the optimal gold marker point set as recommended gold marker implantation sites; and proactively recommending the optimal implantation point by quantitatively evaluating the geometric stability and observable stability of the gold marker based on CT images before implantation, thereby preventing distribution degradation problems, avoiding rework after implantation, and ensuring tracking accuracy and treatment efficiency from the source.

Description

Technical Field

[0001] This invention relates to the field of image-guided radiotherapy technology, and in particular to a method and system for recommending the implantation site of a gold nanoparticle. Background Technology

[0002] In the field of image-guided radiotherapy, gold markers, due to their high contrast and resolution, are often implanted in patients as alternative reference points for tumor location, used to establish spatial localization and real-time tracking models of tumors. Related gold marker tracking technology is widely used for the localization and real-time monitoring of soft tissue tumors, especially in areas such as the liver, pancreas, lungs, and abdominal lymph nodes where tumors are difficult to identify directly through imaging.

[0003] By implanting multiple gold markers inside or around the tumor, the radiotherapy system can establish a three-dimensional spatial relationship between the gold markers before treatment and acquire the spatial position of the gold markers in real time during treatment. Based on the spatial relationship between the gold markers and the tumor in the planned CT scan, the system can track the real-time position of the tumor, thereby achieving high-precision radiotherapy.

[0004] However, the following obvious problems still exist in the existing implantation and application of gold standard: (1) In clinical practice, the implantation location of gold markers is mainly selected by interventional physicians based on experience, and there is a lack of quantitative planning methods based on individual patient imaging data. This empirical implantation method is prone to causing unreasonable spatial distribution of gold markers, such as multiple gold markers located on the same straight line or approximately coplanar, which affects the establishment of subsequent tracking models.

[0005] (2) Radiotherapy systems typically treat multiple gold markers as a rigid body structure, calculating the tumor location by measuring the overall translation and rotation of the gold markers. If the geometric distribution of the gold marker point set in space degenerates (e.g., collinearity or overly concentrated distribution), the stability of the rigid body pose calculation decreases significantly, potentially leading to increased tracking errors or even tracking failure. In the CyberKnife system, gold marker tracking is not based on direct three-dimensional imaging, but rather relies on two-dimensional projection observations of digital radiography (DR) images with a fixed angle (approximately ±45°), and the three-dimensional pose of the target is calculated by inverse projection matching. Therefore, the "trackability" of the spatial distribution of gold markers depends not only on their geometric relationship in three-dimensional space, but also on their imaging geometric characteristics under the aforementioned dual-DR projection system.

[0006] (3) In the existing process, whether the gold marker meets the tracking requirements is often only discovered after implantation, during the planning CT scan and treatment plan establishment. Once it is confirmed that the gold marker geometry is not suitable for tracking, treatment often needs to be delayed, the gold marker re-implanted, or a non-tracking approach is used, which reduces treatment efficiency.

[0007] (4) Even if some systems can assess the quality of gold markers during the treatment planning stage, they cannot provide reverse guidance for the selection of gold marker implantation sites.

[0008] This paper proposes a method, system, computer device, computer-readable storage medium, and computer program product for recommending gold label implantation sites. Summary of the Invention

[0009] This specification provides a recommended method, system, computer device, computer-readable storage medium, and computer program product for gold nanoparticle implantation, to overcome the problems of existing gold nanoparticle implantation processes, such as reliance on experience, lack of quantitative planning, and failure to track after implantation, which affect the accuracy and safety of radiotherapy.

[0010] This application provides a method for recommending gold nanoparticle implantation sites, which employs the following technical solution: Acquire the patient's CT image data; Based on the CT image data, several gold implantation sites were determined. The gold marker implantation points are combined according to preset aggregation rules to construct several gold marker point sets; For each set of gold standard points, evaluate its three-dimensional geometric stability index and its two-dimensional observable stability index. Combining the preset evaluation rules, and the geometric stability index and the observable stability index corresponding to each set of gold points, the optimal set of gold points is selected from all the sets of gold points. The gold implantation points in the optimal set of gold implantation points are used as the recommended gold implantation locations. Optionally, for each of the gold standard point sets, the evaluation of its three-dimensional geometric stability index and its two-dimensional observable stability index includes: Based on the three-dimensional spatial distribution of the gold standard point set, the geometric stability index is analyzed; specifically, the three-dimensional spatial distribution matrix of the gold standard point set is decomposed into eigenvalues; and the geometric stability index is calculated based on the discrete eigenvalues ​​obtained from the decomposition. The observable stability index is analyzed based on the projection of the gold standard point set onto multiple preset imaging planes. Specifically, the gold standard point set is projected onto multiple preset imaging planes; eigenvalue decomposition is performed on the two-dimensional spatial distribution matrix corresponding to each imaging plane; and the observable stability index is calculated based on the multiple sets of projection eigenvalues ​​obtained from the decomposition.

[0011] Optionally, determining several gold implantation sites based on the CT image data includes: Locate the geometric center of the tumor from the CT image data; The original region is constructed in the CT image data based on the geometric center of the tumor. The candidate implantation region is divided from the original region based on anatomical structure information; Sampling is performed in the candidate implantation area according to a preset sampling method to obtain multiple gold-labeled implantation points.

[0012] Optionally, the analysis of the geometric stability index based on the three-dimensional spatial distribution of the gold standard point set further includes: Construct the geometric center of the gold standard point set; The three-dimensional spatial distribution matrix is ​​constructed based on the offset vector of each gold implantation point in the gold implantation point set relative to the geometric center of the point set.

[0013] Optionally, calculating the geometric stability index based on the discrete eigenvalues ​​obtained from the decomposition includes: The ratio of the smallest discrete eigenvalue to the largest discrete eigenvalue is used as the geometric stability index.

[0014] Optionally, each set of projected feature values ​​includes two projected feature values; The calculation of the observable stability index based on multiple sets of projected eigenvalues ​​obtained from the decomposition includes: For each set of projected eigenvalues, the ratio of the smallest projected eigenvalue to the largest projected eigenvalue is used as the observable stability score. The minimum observable stability score is used as the observable stability index.

[0015] Optionally, the step of combining preset evaluation rules and the geometric stability index and observable stability index corresponding to each set of gold points to select the optimal set of gold points includes: Based on the relationship between the geometric stability index and the preset first threshold, it is determined whether the gold standard point set meets the geometric stability requirements; Based on the relationship between the observable stability index and the preset second threshold, it is determined whether the gold standard point set meets the observable stability requirements. If the set of gold markers simultaneously satisfies both the geometric stability requirement and the observable stability requirement, then the set of gold markers is considered a candidate set of gold markers. The degree of stability is calculated based on the geometric stability index and the observable stability index; Select one or more of the candidate gold marker sets with the highest stability as the optimal gold marker set.

[0016] This application provides a recommendation system for gold nanoparticle implantation sites, which employs the following technical solution: The acquisition module is used to acquire the patient's CT image data; The positioning module determines several gold implantation points based on the CT image data. The combination module is used to combine the gold label implantation points according to preset aggregation rules to construct several gold label point sets; The evaluation module is used to evaluate the three-dimensional geometric stability index and the two-dimensional observable stability index for each of the gold standard point sets. The filtering module is used to combine preset evaluation rules and the geometric stability index and the observable stability index corresponding to each set of gold points to filter out the optimal set of gold points from all the sets of gold points. The recommendation module is used to select the gold implantation points from the optimal set of gold implantation points as recommended gold implantation locations. Optionally, the evaluation module includes: The geometric stability analysis submodule is used to analyze the geometric stability index based on the three-dimensional spatial distribution of the gold standard point set; specifically, it performs eigenvalue decomposition on the three-dimensional spatial distribution matrix of the gold standard point set; and calculates the geometric stability index based on the discrete eigenvalues ​​obtained from the decomposition. The observable stability analysis submodule is used to analyze the observable stability index based on the projection of the gold standard point set onto multiple preset imaging planes; specifically, the gold standard point set is projected onto multiple preset imaging planes; eigenvalue decomposition is performed on the two-dimensional spatial distribution matrix corresponding to each imaging plane; and the observable stability index is calculated based on the multiple sets of projection eigenvalues ​​obtained from the decomposition.

[0017] Optionally, the positioning module includes: The central positioning submodule is used to locate the geometric center of the tumor from the CT image data; A region construction submodule is used to construct an original region in the CT image data based on the geometric center of the tumor. The region optimization submodule is used to divide the candidate implantation region from the original region based on anatomical structure information; The sampling submodule is used to sample the candidate implantation area according to a preset sampling method to obtain multiple gold label implantation points.

[0018] Optionally, the analysis of the geometric stability index based on the three-dimensional spatial distribution of the gold standard point set further includes: Construct the geometric center of the gold standard point set; The three-dimensional spatial distribution matrix is ​​constructed based on the offset vector of each gold implantation point in the gold implantation point set relative to the geometric center of the point set.

[0019] Optionally, calculating the geometric stability index based on the discrete eigenvalues ​​obtained from the decomposition includes: The ratio of the smallest discrete eigenvalue to the largest discrete eigenvalue is used as the geometric stability index.

[0020] Optionally, each set of projected feature values ​​includes two projected feature values; The calculation of the observable stability index based on multiple sets of projected eigenvalues ​​obtained from the decomposition includes: For each set of projected eigenvalues, the ratio of the smallest projected eigenvalue to the largest projected eigenvalue is used as the observable stability score. The minimum observable stability score is used as the observable stability index.

[0021] Optionally, the step of combining preset evaluation rules and the geometric stability index and observable stability index corresponding to each set of gold points to select the optimal set of gold points includes: Based on the relationship between the geometric stability index and the preset first threshold, it is determined whether the gold standard point set meets the geometric stability requirements; Based on the relationship between the observable stability index and the preset second threshold, it is determined whether the gold standard point set meets the observable stability requirements. If the set of gold markers simultaneously satisfies both the geometric stability requirement and the observable stability requirement, then the set of gold markers is considered a candidate set of gold markers. The degree of stability is calculated based on the geometric stability index and the observable stability index; Select one or more of the candidate gold marker sets with the highest stability as the optimal gold marker set.

[0022] This specification also provides a computer device, wherein the computer device includes: Processor; and, A memory that stores computer-executable instructions, which, when executed, cause the processor to perform any of the methods described above.

[0023] This specification also provides a computer-readable storage medium that stores one or more programs / instructions that, when executed by a processor, implement any of the methods described above.

[0024] This specification also provides a computer program product, wherein the computer program product includes: a computer program / instruction, which, when executed by a processor, implements any of the methods described above.

[0025] This invention, prior to gold marker implantation, treats multiple gold markers as a rigid body structure based on the patient's 3D CT images and the spatial location of the tumor. It quantitatively analyzes the spatial geometric characteristics of candidate gold marker implantation points, calculates the geometric stability of different gold marker point sets in 3D space, and thus optimizes the gold marker implantation location. It automatically or semi-automatically calculates the optimal combination of multiple gold marker implantation locations in 3D space, providing clinicians or interventional physicians with a scientific and intuitive implantation reference before gold marker implantation, improving the efficiency of gold marker tracking model establishment. Furthermore, the selected gold marker point set possesses good rigid body geometric stability while meeting anatomical safety constraints, thereby improving the success rate and stability of gold marker tracking model establishment in subsequent image-guided radiotherapy. Attached Figure Description

[0026] Figure 1 A flowchart illustrating a method for recommending gold implantation sites provided in the embodiments of this specification; Figure 2 This is a schematic diagram of a gold implantation site recommendation system provided in an embodiment of this specification; Figure 3 A schematic diagram of the structure of a computer device provided in the embodiments of this specification; Figure 4 This is a schematic diagram of a computer-readable storage medium provided for an embodiment of this specification. Detailed Implementation

[0027] The following description is intended to disclose the present invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art. The basic principles of the invention defined in the following description can be applied to other embodiments, modifications, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the invention.

[0028] Exemplary embodiments of the invention will now be described more fully with reference to the accompanying drawings. While conforming to the inventive concept, the features, structures, characteristics, or other details described in a particular embodiment may be combined in one or more other embodiments in a suitable manner.

[0029] The terms “and / or” or “and / or” include all combinations of any one or more of the listed items.

[0030] If the technical solution of this application involves personal information, the product using this technical solution has clearly informed the user of the personal information processing rules and obtained the user's voluntary consent before processing the personal information. If the technical solution of this application involves sensitive personal information, the product using this technical solution has obtained the user's separate consent before processing the sensitive personal information, and also meets the requirement of "express consent". For example, at personal information collection devices such as cameras, clear and prominent signs are set up to inform users that they have entered the scope of personal information collection and that personal information will be collected. If an individual voluntarily enters the collection scope, it is deemed that they have agreed to the collection of their personal information; or on the personal information processing device, with clear signs / information informing users of the personal information processing rules, authorization is obtained from the individual through pop-up information or by asking the individual to upload their personal information; wherein, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the types of personal information processed.

[0031] Figure 1 A flowchart illustrating a method for recommending gold implantation sites provided in embodiments of this specification includes: S1 acquires the patient's CT image data; S2 determines several gold implantation points based on the CT image data; S3 combines the gold label implantation points according to preset aggregation rules to construct several gold label point sets; S4 evaluates the three-dimensional geometric stability index and the two-dimensional observable stability index for each set of gold reference points. S5 combines the preset evaluation rules with the geometric stability index and the observable stability index corresponding to each set of gold points to select the optimal set of gold points from all the sets of gold points. S6 selects the gold implantation points from the optimal set of gold implantation points as the recommended gold implantation locations. This invention is mainly used for preoperative planning and optimization of gold marker implantation position in image-guided radiotherapy (IGRT), and is especially suitable for radiotherapy scenarios that require real-time tracking of gold markers, such as high-precision motion management radiotherapy systems like CyberKnife and helical tomotherapy systems.

[0032] This invention does not depend on a specific tumor site or a single gold standard quantity, and it is applicable to various tumor scenarios such as the liver, pancreas, and lungs where respiration or deformation may occur.

[0033] In actual clinical applications, the implantation position of gold markers usually depends on the operator's experience. Due to the lack of a unified and quantitative planning method, problems such as collinearity of gold markers, spatial distribution degradation, and geometric instability are prone to occur, which can lead to the failure of subsequent tracking model establishment or insufficient tracking accuracy, thereby affecting the accuracy and safety of radiotherapy.

[0034] Addressing the significant shortcomings of existing technologies in gold marker implantation planning, this invention provides a method for recommending gold marker implantation locations. It abstracts the gold marker tracking problem into a rigid body geometric stability problem. By quantitatively calculating the spatial distribution characteristics and geometric stability indices of the gold marker point set, and under the premise of satisfying anatomical structure safety and interventional needle insertion pathway constraints, it can quantitatively calculate and optimize the gold marker implantation location based on the patient's CT image data and tumor location before gold marker implantation, automatically or semi-automatically recommending the optimal implantation location for the gold marker combination. This method does not rely on the operator's personal experience and can effectively reduce problems such as gold marker collinearity and excessively concentrated spatial distribution, thereby improving the reliability and stability of gold marker tracking in image-guided radiotherapy, as well as the overall effectiveness of radiotherapy. Specifically, this invention includes: S1 acquires the patient's CT image data; The patient was scanned using a CT simulation localization scanning device to acquire three-dimensional CT data containing the tumor target area and surrounding tissues. And establish the patient's CT coordinate system.

[0035] The three-dimensional CT data is preprocessed. In one embodiment of this specification, spatial calibration and normalization are performed on the three-dimensional CT data to obtain CT image data. And unified to the patient's CT coordinate system.

[0036] S2 determines several gold implantation points based on the CT image data; Among them, gold label refers to a metal Fiducial Marker.

[0037] S21 locates the geometric center of the tumor from the CT image data; In one embodiment of this specification, based on CT image data The tumor target area outline is delineated manually or automatically to determine the tumor voxel set. Among them, tumor voxel collection It is the set of coordinates of all tumor voxels.

[0038] Calculating the geometric center of the tumor based on tumor voxel set Specifically, The coordinates of each tumor voxel are as follows: .

[0039] In one embodiment of this specification, all voxels are traversed to collect tumor voxels. The coordinates are: the average of the x-coordinates of all tumor voxels is taken as the first x-coordinate, the average of the y-coordinates of all tumor voxels is taken as the first y-coordinate, and the average of the z-coordinates of all tumor voxels is taken as the first z-coordinate; the geometric center of the tumor is determined based on the first x-coordinate, the first y-coordinate, and the first z-coordinate.

[0040] S22 constructs an original region in the CT image data based on the geometric center of the tumor; With the geometric center of the tumor For reference, in CT image data The original region that meets the preset regional conditions is defined.

[0041] The preset region conditions include: tumor voxels. Geometric center of tumor The Euclidean distance is greater than or equal to the minimum area distance. And less than or equal to the maximum area distance ,Right now, .

[0042] That is, the original region is all regions located "with the tumor's geometric center" Center of the sphere, inner radius is Outer radius is A collection of tumor voxels within a spherical shell.

[0043] S23 Divide the candidate implantation region from the original region based on anatomical structure information; Based on anatomical information, blood vessels, bile ducts, airways, and non-punctureable areas are excluded from the original region to obtain the final candidate implantation area. .

[0044] In one embodiment of this specification, in response to a user’s manual region selection, a plurality of feasible candidate implantation regions are pre-defined around the target area.

[0045] S24 samples the candidate implantation area according to a preset sampling method to obtain multiple gold-labeled implantation points.

[0046] Based on the candidate implantation area, perform regular or random sampling in the CT coordinate system to extract several gold-labeled implantation points. .

[0047] Summarize all gold standard implantation points Construct the set of original points .

[0048] Any two gold implantation points in the original point set Satisfying the minimum distance constraint: any two gold implantation points The distance is greater than or equal to the minimum interval distance; that is, , .

[0049] After acquiring and preprocessing the patient's three-dimensional CT data, this invention determines the three-dimensional spatial location of the tumor. Based on this, it identifies candidate implantation areas suitable for gold standard implantation, thereby providing a reliable imaging basis for subsequent gold standard location calculation and facilitating preoperative planning of gold standard implantation location based on three-dimensional CT data.

[0050] S3 combines the gold label implantation points according to preset aggregation rules to construct several gold label point sets; S31 combines the gold label implantation points according to preset aggregation rules to construct several gold label point sets; From the set of original points Select at least three gold-labeled implantation sites to construct a gold-labeled site set: that is, , .

[0051] Preferably, there are 3 or 4. In one embodiment of this specification, the gold marker implantation points are arbitrarily combined based on preset aggregation rules (selecting 3 or 4 gold markers) to generate multiple gold marker point sets for subsequent stability evaluation.

[0052] S32 treats each set of gold reference points as an independent three-dimensional rigid body structure.

[0053] In one embodiment of this specification, it is assumed that there are currently six selectable gold marker implantation sites, denoted as A, B, C, D, E, and F. In actual treatment, it is not necessarily necessary to use all six gold markers simultaneously for tracking. Instead, the tracking system focuses more on whether the relative distribution of the selected gold markers in three-dimensional space is stable and whether it is conducive to accurately determining the spatial pose changes of the tumor.

[0054] Therefore, the core objective of this invention is not "to use as many gold markers as possible," but rather to select a set of gold marker points with optimal stability from these candidates, typically three or four. Specifically, during the calculation process, all possible sets of gold marker points are enumerated from the six points A to F. For example, if three gold markers are selected, multiple sets of three-point combinations will be generated (e.g., A, B, C; A, C, F; B, D, E; ...); if four gold markers are selected, multiple sets of four-point combinations will be generated.

[0055] S4 evaluates the three-dimensional geometric stability index and the two-dimensional observable stability index for each set of gold reference points. This invention performs stability calculations on three-dimensional rigid body structures based on rigid body geometric properties, and evaluates observable stability indices by introducing dual-DR projection geometric constraints of the wave-sweep.

[0056] S41 Analyzes the geometric stability index based on the three-dimensional spatial distribution of the gold standard point set; S411 Constructs the geometric center of the gold standard point set; For each set of reference points, its geometric center is calculated to eliminate the overall positional offset of the point set and the influence of overall translation. Specifically, the geometric center of the point set... .

[0057] In one embodiment of this specification, for the gold-labeled point set Traverse all gold label implantation points and collect gold label implantation points. coordinates ; implant all gold standard points The average of the x-coordinates is used as the second x-coordinate. All gold standard implantation points The average of the y-coordinates is used as the second y-coordinate. All gold standard implantation points The average value of the z-coordinates is used as the second z-coordinate. Determine the geometric center of the point set based on the second x-coordinate, second y-coordinate, and second z-coordinate. .

[0058] S412 constructs the three-dimensional spatial distribution matrix based on the offset vector of each gold implantation point in the gold implantation point set relative to the geometric center of the point set.

[0059] The offset vector is determined based on the difference between the gold implantation point and the geometric center of the point set. ,Right now, .

[0060] Summarize all offset vectors to construct a three-dimensional spatial distribution matrix. The three-dimensional spatial distribution matrix... .

[0061] To elaborate, the three-dimensional spatial distribution matrix .

[0062] This invention captures the dispersion and correlation of the gold standard point set in each direction by calculating the offset vector of all points relative to the center and constructing the covariance matrix (three-dimensional spatial distribution matrix M).

[0063] S413 performs eigenvalue decomposition on the three-dimensional spatial distribution matrix of the gold standard point set; The three-dimensional spatial distribution matrix is ​​decomposed into eigenvalues ​​to obtain discrete eigenvalues ​​reflecting the main expansion direction and the degradation direction of the gold standard point set in three-dimensional space.

[0064] In one embodiment of this specification, the characteristic equation is solved. Three discrete eigenvalues ​​were obtained. .in It is a 3×3 identity matrix.

[0065] The geometric stability of the gold standard point set is evaluated by determining whether it is approximately collinear or near-planar based on the magnitude of the discrete eigenvalues, which is the geometric stability of the three-dimensional rigid body structure.

[0066] S414 calculates the geometric stability index based on the discrete eigenvalues ​​obtained from the decomposition; For any set of gold reference points, this invention does not concern itself with their absolute positions, but rather with their relative distribution in space, i.e. whether these points are: too close to a straight line, or roughly distributed in the same plane, or fully spread out in all three spatial directions.

[0067] Specifically, this process can be likened to this: if all the gold markers are almost lined up in a straight line, it will be difficult to distinguish the true pose changes when the tumor rotates slightly; if the gold markers are distributed in three-dimensional space, the spatial changes of the tumor can be identified more stably and reliably, regardless of whether the tumor translates or rotates.

[0068] Based on this, the present invention first classifies discrete eigenvalues ​​according to their magnitude relationship, and obtains the first discrete eigenvalue for each. Second discrete eigenvalue With the third discrete eigenvalue ;and .

[0069] in, Characterizes the degree of discreteness of the gold standard point set in the main expansion direction; Characterizes the degree of discreteness of the gold standard point set in the secondary expansion direction; This characterizes the degree of dispersion of the gold standard point set along the third orthogonal direction (degeneracy direction). When When the value is close to 0, it indicates that the implantation points of the gold-labeled points in the gold-labeled point set are approximately coplanar; when... When the value is close to 0, it indicates that the implantation points of the gold markers in the gold marker set are approximately collinear.

[0070] Preferably, the ratio of the smallest discrete eigenvalue to the largest discrete eigenvalue is used as the geometric stability index. .Right now, .

[0071] This invention abstracts the geometric structure formed by all gold marker implantation points in the gold marker set in three-dimensional space into a rigid body structure. By quantitatively analyzing the spatial distribution characteristics of the gold marker set, calculating its geometric stability index, and selecting gold marker sets that have good distribution, are non-collinear and non-coplanar in three-dimensional space, the stability of rigid body attitude calculation is improved by optimizing the gold marker set based on the rigid body geometric stability.

[0072] S42 analyzes the observable stability index based on the projection of the gold standard point set onto multiple preset imaging planes; To avoid subjective judgment, this invention converts the spatial distribution characteristics of each gold marker set into a stability index to measure whether the gold marker set is "stable" in three-dimensional space. In simple terms, the observable stability index reflects whether the gold marker set's expansion is balanced across the three spatial directions and whether there is a significant "collapse direction." If the spatial expansion of a gold marker set in a certain direction is extremely small, it indicates that the gold marker set is not sensitive to pose changes in that direction, and its stability evaluation will be correspondingly reduced.

[0073] Therefore, after completing the spatial geometric stability optimization based on the rigid body structure, the geometric constraints of the unique dual DR imaging system of CyberKnife also need to be taken into consideration.

[0074] S421 projects the gold dot set onto multiple preset imaging planes; Considering that the CyberKnife system uses a two-path digital ray (DR) imaging method with a fixed included angle (approximately ±45°) for real-time tracking, the three-dimensional pose calculation depends on the observability of the gold standard points under the aforementioned dual-projection system. This invention defines the unit vectors of the two DR imaging directions as follows: the first DR imaging direction... Second DR imaging direction The two DR imaging directions are fixed, known parameters in the device coordinate system, corresponding to approximately +45° and 10° respectively. Imaging geometry at 45°.

[0075] The gold implantation points in the gold-labeled point set are projected onto two DR imaging planes respectively to obtain the corresponding two-dimensional projection point set; wherein, the two-dimensional projection point set . .

[0076] In one embodiment of this specification, for each gold implantation point in the gold implantation point set Along the first DR imaging direction respectively Orthogonal projection is performed to obtain a two-dimensional projection point set: for each gold implantation point in the gold implantation point set... Along the second DR imaging direction respectively Perform orthogonal projection to obtain another two-dimensional set of projection points.

[0077] S422 constructs a two-dimensional spatial distribution matrix for each set of projection points; For each two-dimensional projection point set, calculate its two-dimensional geometric center; based on the distances of each point in the two-dimensional projection point set relative to the two-dimensional geometric center, construct a two-dimensional spatial distribution matrix. The two-dimensional spatial distribution matrix... .

[0078] S424 performs eigenvalue decomposition on the two-dimensional spatial distribution matrix corresponding to each imaging plane; The two-dimensional spatial distribution matrix is ​​decomposed into eigenvalues ​​to obtain projection features that reflect the distribution of the gold standard point set in the projection plane.

[0079] In one embodiment of this specification, the characteristic equation is solved. This yields a set of projection feature values. Each set of projection feature values ​​includes two projection feature values. . It is a 2×2 identity matrix.

[0080] S425 calculates the observable stability index based on the multiple sets of projected eigenvalues ​​obtained from the decomposition.

[0081] Each two-dimensional projection distribution matrix These correspond to two projected eigenvalues.

[0082] The projection eigenvalues ​​are classified according to their magnitude relationship, and the first projection eigenvalue is obtained from each classification. With the second projection eigenvalue ;and .

[0083] When in any DR projection, the following conditions are met If the gold marker set is found to be approximately collinear in the DR projection direction of approximately ±45°, then the gold marker set is considered to have a risk of observability degradation under the CyberKnife tracking system, which can easily lead to a decrease in the observability of the CyberKnife's three-dimensional pose inverse calculation.

[0084] Therefore, for each set of projected eigenvalues, the ratio of the smallest projected eigenvalue to the largest projected eigenvalue is used as the observable stability score; the smallest observable stability score is used as the observable stability index. That is, the observable stability index under DR projection. .

[0085] This invention not only evaluates the rigid geometric distribution characteristics of gold markers in three-dimensional space, but also introduces observability constraints under the CyberKnife dual-DR projection system. This effectively identifies gold marker point sets that appear reasonable in three-dimensional space but pose a risk of degradation in actual DR projection. Based on DR projection eigenvalues ​​and projection collinearity risk assessment, the robustness and safety of CyberKnife real-time tracking are improved, avoiding tracking failures due to projection collinearity or depth-direction degradation.

[0086] Moreover, by assessing the geometric stability and projection observability of the gold marker set before or during the treatment planning stage, this invention can identify gold marker schemes with risks of collinearity, near-plane, or DR projection degradation in advance. This reduces / avoids the risk of tracking system establishment failure, treatment interruption, or forced switching of schemes due to unreasonable spatial distribution of gold markers, thereby reducing the need for repeated implantation or changes in treatment schemes and improving the reliability of the overall treatment process.

[0087] S5 combines the preset evaluation rules with the geometric stability index and the observable stability index corresponding to each set of gold points to select the optimal set of gold points from all the sets of gold points. S51 determines whether the gold standard point set meets the geometric stability requirements based on the relationship between the geometric stability index and the preset first threshold. When geometric stability index Greater than or equal to the preset first threshold At that time, it was determined that the gold standard point set met the geometric stability requirements.

[0088] S52 determines whether the gold standard point set meets the observable stability requirements based on the relationship between the observable stability index and the preset second threshold. When observable stability indicators Greater than or equal to the preset second threshold At that time, it was determined that the gold standard point set met the observable stability requirements.

[0089] S53 If the set of gold markers simultaneously satisfies both the geometric stability requirement and the observable stability requirement, then the set of gold markers is identified as a candidate set of gold markers. In other words, if and only if the following conditions are met simultaneously: and When the gold standard point set is determined to satisfy both the three-dimensional rigid body geometric stability requirements and the observability requirements under the dual DR projection geometry of the waveguide, it can be used for real-time tracking.

[0090] At this point, only the set of gold reference points that simultaneously satisfies the geometric stability requirements of a three-dimensional rigid body and the observability stability requirements of the dual DR projection of the waveguide is retained as a candidate set of gold reference points, and the rest of the set of gold reference points is discarded.

[0091] This invention establishes a clear quantitative judgment standard through eigenvalues ​​(discrete eigenvalues ​​and projected eigenvalues) and stability indices (geometric stability indices corresponding to 3D and observable stability indices corresponding to DR projection). By constructing a 3D spatial distribution matrix of the gold target point set and combining it with the dual-path ±45° digital ray (DR) projection geometry of the CyberKnife system, the spatial stability of the gold target point set is quantitatively evaluated. Compared with traditional methods that rely solely on empirical rules (such as avoiding collinearity and maximizing dispersion), this invention provides an objective and quantitative gold target availability judgment standard, which can accurately determine whether the gold target point set has good tracking conditions from both mathematical and equipment geometric perspectives. It reduces human error, improves the consistency and repeatability of gold target selection decisions, and significantly improves the stability and accuracy of the gold target tracking model.

[0092] S54 calculates the degree of stability based on the geometric stability index and the observable stability index; In one embodiment of this specification, the stability level = Geometric stability index The observable stability index.

[0093] in, The influence weight of the geometric stability index; The influence weight of the observable stability index.

[0094] S55 selects one or more of the candidate gold marker sets with the highest stability as the optimal gold marker set.

[0095] S6 selects the gold implantation points from the optimal set of gold implantation points as the recommended gold implantation locations. This invention repeats the above evaluation process for all gold marker implantation sites and ranks them according to their stability. Finally, from the set of gold marker sites that meet all stability requirements, one or more sets of gold marker sites with the highest stability (i.e., the best stability index) are selected as the optimal gold marker set. Based on the optimal gold marker set, the recommended number of gold markers, the three-dimensional spatial coordinates of each gold marker, and / or multiple optional gold marker implantation schemes are output. In response to the operator's selection and / or modification of the optimal gold marker set, the final gold marker implantation location is determined and the operation is performed.

[0096] The method for recommending gold marker implantation locations based on quantitative calculation provided by this invention can automatically calculate and screen a set of gold markers with high spatial geometric stability and good observability in both DR projections during the gold marker implantation planning or treatment planning stages. This reduces repeated discussions before treatment, rework during treatment, or changes in treatment plans. Furthermore, the gold marker implantation decision no longer relies entirely on the operator's personal experience, thereby reducing the workload of the clinical and physical teams, improving the overall efficiency of the CyberKnife treatment process, and maintaining consistency in implantation quality among different operators.

[0097] It is important to emphasize that this recommendation does not rely on the superiority or inferiority of a single gold marker position, but rather on the geometric stability of the gold marker point set as a whole in three-dimensional space, and the observable stability of the actively constrained three-dimensional rigid body point set under the ±45° dual DR imaging geometry of the CyberKnife. Therefore, it is more beneficial for subsequent real-time tracking and pose determination.

[0098] Meanwhile, this invention can be extended to the combined evaluation of different numbers of gold markers and can be integrated with existing CyberKnife systems or image-guided workflows, enhancing CyberKnife's ability to identify tumor translation and rotation, helping to improve the accuracy of tumor target localization, reduce dose deviation caused by tracking errors, thereby improving treatment safety, and has good prospects for engineering applications.

[0099] Figure 2 This is a schematic diagram of a gold implantation site recommendation system provided in an embodiment of this specification. The system includes: The acquisition module 210 is used to acquire the patient's CT image data; The positioning module 220 determines several gold implantation points based on the CT image data. The combination module 230 is used to combine the gold label implantation points according to a preset aggregation rule to construct a number of gold label point sets; Evaluation module 240 is used to evaluate the three-dimensional geometric stability index and the two-dimensional observable stability index for each of the gold standard point sets. The screening module 250 is used to select the optimal set of gold points from all the gold point sets by combining the preset evaluation rules and the geometric stability index and the observable stability index corresponding to each set of gold points. The recommendation module 260 is used to select the gold implantation points in the optimal set of gold implantation points as recommended gold implantation locations. Optionally, the evaluation module 240 includes: The geometric stability analysis submodule is used to analyze the geometric stability index based on the three-dimensional spatial distribution of the gold standard point set; specifically, it performs eigenvalue decomposition on the three-dimensional spatial distribution matrix of the gold standard point set; and calculates the geometric stability index based on the discrete eigenvalues ​​obtained from the decomposition. The observable stability analysis submodule is used to analyze the observable stability index based on the projection of the gold standard point set onto multiple preset imaging planes; specifically, the gold standard point set is projected onto multiple preset imaging planes; eigenvalue decomposition is performed on the two-dimensional spatial distribution matrix corresponding to each imaging plane; and the observable stability index is calculated based on the multiple sets of projection eigenvalues ​​obtained from the decomposition.

[0100] Optionally, the positioning module 220 includes: The central positioning submodule is used to locate the geometric center of the tumor from the CT image data; A region construction submodule is used to construct an original region in the CT image data based on the geometric center of the tumor. The region optimization submodule is used to divide the candidate implantation region from the original region based on anatomical structure information; The sampling submodule is used to sample the candidate implantation area according to a preset sampling method to obtain multiple gold label implantation points.

[0101] Optionally, the analysis of the geometric stability index based on the three-dimensional spatial distribution of the gold standard point set further includes: Construct the geometric center of the gold standard point set; The three-dimensional spatial distribution matrix is ​​constructed based on the offset vector of each gold implantation point in the gold implantation point set relative to the geometric center of the point set.

[0102] Optionally, calculating the geometric stability index based on the discrete eigenvalues ​​obtained from the decomposition includes: The ratio of the smallest discrete eigenvalue to the largest discrete eigenvalue is used as the geometric stability index.

[0103] Optionally, each set of projected feature values ​​includes two projected feature values; The calculation of the observable stability index based on multiple sets of projected eigenvalues ​​obtained from the decomposition includes: For each set of projected eigenvalues, the ratio of the smallest projected eigenvalue to the largest projected eigenvalue is used as the observable stability score. The minimum observable stability score is used as the observable stability index.

[0104] Optionally, the step of combining preset evaluation rules and the geometric stability index and observable stability index corresponding to each set of gold points to select the optimal set of gold points includes: Based on the relationship between the geometric stability index and the preset first threshold, it is determined whether the gold standard point set meets the geometric stability requirements; Based on the relationship between the observable stability index and the preset second threshold, it is determined whether the gold standard point set meets the observable stability requirements. If the set of gold markers simultaneously satisfies both the geometric stability requirement and the observable stability requirement, then the set of gold markers is considered a candidate set of gold markers. The degree of stability is calculated based on the geometric stability index and the observable stability index; Select one or more of the candidate gold marker sets with the highest stability as the optimal gold marker set.

[0105] The functions of the system in this embodiment have been described in the above method embodiments. Therefore, for any parts not detailed in this embodiment, please refer to the relevant descriptions in the foregoing embodiments, which will not be repeated here.

[0106] Figure 3 This is a schematic diagram of the structure of a computer device provided in an embodiment of this specification. The computer device includes a memory 301 and a processor 302. The memory 301 is used to store computer-executable instructions. When the computer-executable instructions are executed by the processor 302, they can implement the steps of the above-described method embodiment.

[0107] Figure 4 This is a schematic diagram of the structure of a computer-readable storage medium 400 provided in an embodiment of this specification. The computer-readable storage medium 400 stores one or more computer programs, which, when executed by a processor, can implement the steps of the above-described method embodiments.

[0108] This specification also provides a computer program product, including a computer program / computer executable instructions, which, when executed by a processor, can implement the steps of the above-described method embodiments.

[0109] Those skilled in the art will understand that all or part of the processes in the above method embodiments can be implemented by a computer program instructing related hardware. When the computer program is executed, it may include the processes of the embodiments of the above methods.

[0110] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for recommending the implantation site of a gold nanoparticle, characterized in that, include: Acquire the patient's CT image data; Based on the CT image data, several gold implantation sites were determined. The gold marker implantation points are combined according to preset aggregation rules to construct several gold marker point sets; For each set of gold standard points, evaluate its three-dimensional geometric stability index and its two-dimensional observable stability index. Combining the preset evaluation rules, and the geometric stability index and the observable stability index corresponding to each set of gold points, the optimal set of gold points is selected from all the sets of gold points. The gold implantation points in the optimal set of gold implantation points are used as the recommended gold implantation locations. Specifically, for each set of gold reference points, the evaluation of its three-dimensional geometric stability index and its two-dimensional observable stability index includes: Based on the three-dimensional spatial distribution of the gold standard point set, the geometric stability index is analyzed; specifically, the three-dimensional spatial distribution matrix of the gold standard point set is decomposed into eigenvalues; and the geometric stability index is calculated based on the discrete eigenvalues ​​obtained from the decomposition. The observable stability index is analyzed based on the projection of the gold standard point set onto multiple preset imaging planes. Specifically, the gold standard point set is projected onto multiple preset imaging planes; eigenvalue decomposition is performed on the two-dimensional spatial distribution matrix corresponding to each imaging plane; and the observable stability index is calculated based on the multiple sets of projection eigenvalues ​​obtained from the decomposition.

2. The method for recommending the implantation site of the gold marker as described in claim 1, characterized in that, The determination of several gold implantation sites based on the CT image data includes: Locate the geometric center of the tumor from the CT image data; The original region is constructed in the CT image data based on the geometric center of the tumor. Candidate implantation areas are divided from the original region based on anatomical information; Sampling is performed in the candidate implantation area according to a preset sampling method to obtain multiple gold-labeled implantation points.

3. The method for recommending the implantation site of the gold marker as described in claim 1, characterized in that, The analysis of the geometric stability index based on the three-dimensional spatial distribution of the gold standard point set further includes: Construct the geometric center of the gold standard point set; The three-dimensional spatial distribution matrix is ​​constructed based on the offset vector of each gold implantation point in the gold implantation point set relative to the geometric center of the point set.

4. The method for recommending the implantation site of the gold marker as described in claim 1, characterized in that, The calculation of the geometric stability index based on the discrete eigenvalues ​​obtained from the decomposition includes: The ratio of the smallest discrete eigenvalue to the largest discrete eigenvalue is used as the geometric stability index.

5. The method for recommending the implantation site of the gold marker as described in claim 1, characterized in that, Each set of projected feature values ​​includes two projected feature values; The calculation of the observable stability index based on multiple sets of projected eigenvalues ​​obtained from the decomposition includes: For each set of projected eigenvalues, the ratio of the smallest projected eigenvalue to the largest projected eigenvalue is used as the observable stability score. The minimum observable stability score is used as the observable stability index.

6. The method for recommending the implantation site of the gold marker as described in claim 1, characterized in that, The step of combining preset evaluation rules and the geometric stability index and observable stability index corresponding to each set of gold points to select the optimal set of gold points from all the sets of gold points includes: Based on the relationship between the geometric stability index and the preset first threshold, it is determined whether the gold standard point set meets the geometric stability requirements; Based on the relationship between the observable stability index and the preset second threshold, it is determined whether the gold standard point set meets the observable stability requirements. If the set of gold markers simultaneously satisfies both the geometric stability requirement and the observable stability requirement, then the set of gold markers is considered a candidate set of gold markers. The degree of stability is calculated based on the geometric stability index and the observable stability index; Select one or more of the candidate gold marker sets with the highest stability as the optimal gold marker set.

7. A recommendation system for gold label implantation sites, characterized in that, include: The acquisition module is used to acquire the patient's CT image data; The positioning module determines several gold implantation points based on the CT image data. The combination module is used to combine the gold label implantation points according to preset aggregation rules to construct several gold label point sets; The evaluation module is used to evaluate the three-dimensional geometric stability index and the two-dimensional observable stability index for each of the gold standard point sets. The filtering module is used to combine preset evaluation rules and the geometric stability index and the observable stability index corresponding to each set of gold points to filter out the optimal set of gold points from all the sets of gold points. The recommendation module is used to select the gold implantation points from the optimal set of gold implantation points as recommended gold implantation locations. The evaluation module includes: The geometric stability analysis submodule is used to analyze the geometric stability index based on the three-dimensional spatial distribution of the gold standard point set; specifically, it performs eigenvalue decomposition on the three-dimensional spatial distribution matrix of the gold standard point set; and calculates the geometric stability index based on the discrete eigenvalues ​​obtained from the decomposition. The observable stability analysis submodule is used to analyze the observable stability index based on the projection of the gold standard point set onto multiple preset imaging planes; specifically, the gold standard point set is projected onto multiple preset imaging planes; eigenvalue decomposition is performed on the two-dimensional spatial distribution matrix corresponding to each imaging plane; and the observable stability index is calculated based on the multiple sets of projection eigenvalues ​​obtained from the decomposition.

8. A computer device, characterized in that, The computer device includes: Processor; and, A memory storing computer-executable instructions, which, when executed, cause the processor to perform the method of any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores one or more programs / instructions, which, when executed by a processor, implement the method as described in any one of claims 1-6.

10. A computer program product, characterized in that, Includes a computer program / instruction, which, when executed by a processor, implements the method as described in any one of claims 1-6.

Images