A space-time puncture path planning system and method based on straight line constraint

Abstract

The application discloses a kind of space-time puncture path planning system and method based on linear constraint, it is related to puncture path optimization technical field.The method collects CT image data of patient chest and abdomen under different respiratory phase, constructs displacement vector field and organ movement heat map based on CT image data;Based on preset skin needle entry point, tumor target and organ movement heat map, determine candidate intermediate point in chest and abdominal organ region, and generate key point set;Based on the connection rule, the skin needle entry point in the key point set, candidate intermediate point and tumor target are connected to form at least one path edge;According to organ movement heat map, the comprehensive weight of path edge is calculated, and the weighted directed graph is constructed;Path search is carried out in the weighted directed graph, and candidate path set is generated;According to the cost of each candidate path calculated according to the comprehensive weight, the final puncture path is selected, which improves the intraoperative puncture precision and reduces the risk of complications.

Description

Technical Field

[0001] This application relates to the field of puncture path optimization technology, specifically a spatiotemporal puncture path planning system and method based on linear constraints. Background Technology

[0002] Percutaneous interventional surgery is a key technique for tumor diagnosis and treatment, with its core challenge lying in organ deformation and displacement caused by respiratory motion. Current puncture path planning methods primarily rely on static CT images to generate optimal paths, achieving safe puncture by avoiding dangerous areas such as bones and major blood vessels. However, this method has an inherent flaw—lack of respiratory motion compensation. The static path cannot adapt to the dynamic displacement of organs during surgery, resulting in actual puncture errors as high as 3-5 mm, especially in respiratory-sensitive organs such as the lungs and liver. The root cause of this lack of compensation is that current techniques simplify respiratory motion into a linear model, ignoring the nonlinear characteristics of the inspiratory / expiratory process, leading to a phase recognition deviation greater than 1.5 phase units.

[0003] The lack of respiratory movement compensation leads to the failure of pathway planning. During the execution of static pathways, organ displacement causes the needle tip to deviate from the preset trajectory, requiring repeated adjustments to the puncture angle, which increases the operation time by an average of 15 minutes. Furthermore, unforeseen rapid displacement areas are prone to causing blood vessel rupture or organ damage, increasing the risk of complications. Summary of the Invention

[0004] The purpose of this application is to provide a spatiotemporal puncture path planning system and method based on linear constraints to solve the problems mentioned in the background art.

[0005] In a first aspect, one embodiment of this application provides a puncture path planning method based on linear constraints. The method includes: acquiring CT image data of the patient's chest and abdomen at different respiratory phases; constructing a displacement vector field and an organ motion heatmap based on the CT image data, whereby the displacement vector field reflects organ motion information and the organ motion heatmap represents the average motion velocity of each point within the organ region; determining candidate intermediate points within the chest and abdominal organ region based on preset skin insertion points, tumor target points, and the organ motion heatmap, and generating a key point set; connecting the skin insertion points, candidate intermediate points, and tumor target points in the key point set according to preset connection rules to form at least one path edge; calculating the comprehensive weight of the path edge based on the organ motion heatmap, and constructing a weighted directed graph based on the path edge and the comprehensive weight; performing path search in the weighted directed graph to generate a candidate path set; and then... The cost of each candidate path is calculated based on a comprehensive weight, and the final puncture path is selected from the candidate path set based on the cost. This includes the following steps: According to the weighted directed graph, single-segment paths that do not pass through candidate intermediate points and double-segment paths that pass through candidate intermediate points are added to the candidate path set; the cost of a single-segment path is calculated based on the comprehensive weight; the cost from the starting point of the path edge to the candidate intermediate point and the cost from the candidate intermediate point to the ending point of the path edge are calculated based on the comprehensive weight, and the cost from the starting point to the candidate intermediate point and the cost from the candidate intermediate point to the ending point are added together to obtain the cost of the double-segment path; the additional penalty term for the double-segment path is calculated based on the phase difference, and combined with the cost of the double-segment path, the final cost of the double-segment path is obtained; the costs of the single-segment paths and the final costs of the double-segment paths are sorted in ascending order, constraints are set, and the path with the minimum cost from the candidate path set that satisfies the constraints is selected as the final puncture path.

[0006] In conjunction with the first aspect, in some implementations of the first aspect, constructing a displacement vector field and an organ motion heatmap based on CT image data includes: obtaining a phase-displacement mapping table and three-dimensional volume data based on CT image data; calculating a displacement vector field based on the phase-displacement mapping table and three-dimensional volume data; calculating the average motion velocity of each voxel based on the displacement vector field, and using the average motion velocity as a heatmap value to generate an organ motion heatmap.

[0007] In conjunction with the first aspect, in certain implementations of the first aspect, a displacement vector field is calculated based on the phase-displacement mapping table and three-dimensional volume data. Based on the displacement vector field, the average motion velocity of each voxel is calculated, and the average motion velocity is used as a heatmap value to generate an organ motion heatmap. This includes: setting the end-expiratory phase as the reference phase based on the phase-displacement mapping table and three-dimensional volume data, calculating the displacement vector field of other respiratory phases relative to the reference phase; calculating the instantaneous velocity of each voxel at different respiratory phases based on the displacement vector field, calculating the average motion velocity of each voxel based on the instantaneous velocity; and generating an organ motion heatmap using the average motion velocity as a heatmap value.

[0008] In conjunction with the first aspect, in certain implementations of the first aspect, based on preset skin needle insertion points, tumor target points, and organ motion heatmaps, candidate intermediate points are determined within the thoracic and abdominal organ region, and a key point set is generated. This includes: verifying that the skin needle insertion point is located above the body surface contour or within a first preset distance from the body surface contour, and / or verifying that the tumor target point is located within the tumor segmentation boundary and within a second preset distance from the tumor segmentation boundary; calculating the main puncture direction based on the skin needle insertion point and the tumor target point; calculating the sampling plane based on the skin needle insertion point and the main puncture direction; using the intersection of the sampling plane and the thoracic and abdominal organs as candidate intermediate points; and generating a key point set based on the candidate intermediate points and the displacement vector field.

[0009] In conjunction with the first aspect, in some implementations of the first aspect, the comprehensive weight of the path edges is calculated based on the organ motion heatmap, and a weighted directed graph is constructed based on the path edges and the comprehensive weight. This includes: calculating the length term of the path edge based on the starting and ending coordinates of the path edge; discretizing the path edge based on the heatmap value of the organ motion heatmap and calculating the risk term of the path edge; calculating the comprehensive weight of the path edge based on the length term and the risk term; performing safety filtering on the path edges, removing path edges that pass through preset dangerous organs, and / or path edges whose angle with the baseline path exceeds a preset angle threshold; and constructing a weighted directed graph using the filtered path edges and the comprehensive weight.

[0010] In conjunction with the first aspect, in some implementations of the first aspect, a phase-displacement mapping table and three-dimensional volume data are obtained based on CT image data, including: dividing the CT image data of the patient's chest and abdomen into respiratory phases and recording the phase displacement of the respiratory process; assigning phase numbers to the respiratory phases to obtain a phase-displacement mapping table; and performing filtered back projection reconstruction on each phase based on the CT image data to obtain three-dimensional volume data.

[0011] Secondly, this application provides a puncture path planning system based on linear constraints, comprising: an acquisition and construction module for acquiring CT image data of the patient's chest and abdomen at different respiratory phases, constructing a displacement vector field and an organ motion heatmap based on the CT image data, wherein the displacement vector field reflects organ motion information and the organ motion heatmap represents the average motion velocity of each point within the organ region; a key point generation module for determining candidate intermediate points within the chest and abdomen organ region based on preset skin needle insertion points, tumor target points, and organ motion heatmaps, and generating a key point set; a path edge generation module for connecting the skin needle insertion points, candidate intermediate points, and tumor target points in the key point set to form at least one path edge based on preset connection rules; a weighted directed graph construction module for calculating the comprehensive weight of the path edges based on the organ motion heatmap, and constructing a weighted directed graph based on the path edges and comprehensive weights; and a path planning module for performing path planning on the weighted directed graph. The path search generates a candidate path set. The path generation module calculates the cost of each candidate path based on a comprehensive weight, and selects the final puncture path from the candidate path set based on the cost. This includes the following steps: Based on the weighted directed graph, single-segment paths that do not pass through candidate intermediate points and double-segment paths that do pass through candidate intermediate points are added to the candidate path set; the cost of a single-segment path is calculated based on the comprehensive weight; the cost from the starting point of the path edge to the candidate intermediate point and the cost from the candidate intermediate point to the ending point of the path edge are calculated based on the comprehensive weight, and the cost from the starting point to the candidate intermediate point and the cost from the candidate intermediate point to the ending point are added together to obtain the cost of the double-segment path; an additional penalty term for the double-segment path is calculated based on the phase difference, and combined with the cost of the double-segment path, the final cost of the double-segment path is obtained; the costs of the single-segment paths and the final costs of the double-segment paths are sorted in ascending order, constraints are set, and the path with the minimum cost from the candidate path set that satisfies the constraints is selected as the final puncture path.

[0012] Thirdly, this application provides an electronic device, including: a processor; and a memory storing computer program instructions, which, when executed by the processor, implement the steps in the puncture path planning method based on linear constraints mentioned in the first aspect.

[0013] Fourthly, this application provides a computer-readable storage medium storing computer program instructions, which, when executed by a processor, cause the processor to perform the steps in the puncture path planning method based on linear constraints mentioned in the first aspect.

[0014] The puncture path planning method based on linear constraints provided in this application solves the problem of path failure in the respiratory motion scenario by establishing a spatiotemporal model of respiratory motion and quantifying organ displacement risk, combined with real-time respiratory phase synchronization navigation. It realizes dynamic risk-aware puncture path planning, which significantly improves intraoperative puncture accuracy and reduces the risk of complications while ensuring path safety. Attached Figure Description

[0015] Figure 1 The diagram shown is a flowchart of a puncture path planning method based on linear constraints provided in an exemplary embodiment of this application.

[0016] Figure 2 The diagram shown is a flowchart of a puncture path planning method based on linear constraints provided in another exemplary embodiment of this application.

[0017] Figure 3 The diagram shown is a flowchart of a puncture path planning method based on linear constraints provided in another exemplary embodiment of this application.

[0018] Figure 4 The diagram shown is a flowchart of a puncture path planning method based on linear constraints provided in another exemplary embodiment of this application.

[0019] Figure 5 The diagram shown is a flowchart of a puncture path planning method based on linear constraints provided in another exemplary embodiment of this application.

[0020] Figure 6 The diagram shown is a flowchart of a puncture path planning method based on linear constraints provided in another exemplary embodiment of this application.

[0021] Figure 7 The diagram shown is a flowchart of a puncture path planning method based on linear constraints provided in another exemplary embodiment of this application.

[0022] Figure 8 The diagram shown is a structural schematic of a puncture path planning system based on linear constraints provided in an exemplary embodiment of this application.

[0023] Figure 9 The diagram shown is a structural schematic of an electronic device provided in an exemplary embodiment of this application. Detailed Implementation

[0024] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0025] The following is combined Figures 1 to 7 This application provides a detailed description of a puncture path planning method based on linear constraints.

[0026] Figure 1 The diagram shown is a flowchart illustrating a puncture path planning method based on linear constraints provided in an exemplary embodiment of this application. Figure 1 As shown in the embodiments of this application, the puncture path planning method based on linear constraints includes the following steps.

[0027] Step 100: Collect CT image data of the patient's chest and abdomen at different respiratory phases, and construct a displacement vector field and organ motion thermogram based on the CT image data.

[0028] The displacement vector field reflects organ motion information, and the organ motion heatmap represents the average motion velocity of each point within the organ region.

[0029] Step 102: Based on the preset skin needle entry point, tumor target point and the organ motion heat map, determine candidate intermediate points in the thoracic and abdominal organ region and generate a key point set.

[0030] For example, the skin entry point refers to the puncture start position designated by the doctor on the patient's body surface, used to jointly define the main puncture direction vector with the tumor target and only allowed as the starting point of the path.

[0031] For example, a tumor target point refers to the puncture target location specified by the doctor inside the tumor. The effectiveness of the positioning needs to be ensured through the shortest distance verification. It is used to form the main axis of the path with the skin needle entry point and participate in the calculation of the endpoint weight of single / double path.

[0032] Step 104: Based on the preset connection rules, at least one path edge is formed by connecting the skin needle entry point, candidate intermediate point and tumor target point in the key point set.

[0033] Specifically, based on the key point set and preset connection rules, the skin needle entry point, candidate intermediate point and tumor target point are connected to obtain path edges. Based on the heatmap value, the comprehensive weight of the path edges is calculated, and path edges with risks are eliminated to obtain a weighted directed graph.

[0034] For example, the connection rule refers to the topological constraint conditions for establishing directed edges between nodes in the path graph, including three levels of restrictions: in-phase constraint, puncture sequence constraint, and prohibition rule, which are used to avoid path breakage due to respiratory phase misalignment and to comply with the physical constraints of unidirectional advancement of puncture instruments.

[0035] For example, a path edge is a directed straight line segment connecting two key points, carrying a comprehensive weight value. The weight is calculated by weighting the length and risk terms, and is used to assess organ movement risk through discretized path points and to drive the optimal path search algorithm based on the weight value.

[0036] Step 106: Calculate the comprehensive weight of the path edges based on the organ motion heatmap, and construct a weighted directed graph based on the path edges and comprehensive weights.

[0037] For example, a weighted directed graph is a data structure consisting of a set of nodes (key points), a set of directed edges (path edges), and edge weights, used to store all candidate solutions for single / double-segment paths and to accelerate the dynamic projection of intraoperative paths through spatial indexing.

[0038] Step 108: Perform path search in the weighted directed graph to generate a candidate path set.

[0039] Step 110: Calculate the cost of each candidate path based on the comprehensive weight, and select the final puncture path from the candidate path set based on the cost.

[0040] Specifically, based on the weighted directed graph, single-segment paths that do not pass through candidate intermediate points and double-segment paths that pass through candidate intermediate points are added to the candidate path set. Based on the comprehensive weight, the cost of single-segment paths and double-segment paths is calculated, and the path with the minimum cost is selected as the final puncture path to complete the puncture path planning.

[0041] For example, a single-segment path refers to a straight puncture path without a midpoint, defined as a single side from the skin needle entry point to the tumor target point, used as an optimized solution for simple scenarios with small organ displacement and unobstructed areas.

[0042] For example, a dual-path refers to two path edges connected by a midpoint. As an obstacle avoidance scheme in complex scenarios, it is used to avoid large blood vessels or high-speed areas by passing through the midpoint. The penalty term ensures that the two punctures are completed in the same respiratory phase.

[0043] The puncture path planning method based on linear constraints provided in this application solves the problem of path failure in the respiratory motion scenario by establishing a spatiotemporal model of respiratory motion and quantifying organ displacement risk, combined with real-time respiratory phase synchronization navigation. It realizes dynamic risk-aware puncture path planning, which significantly improves intraoperative puncture accuracy and reduces the risk of complications while ensuring path safety.

[0044] Figure 2 The diagram shown is a flowchart illustrating a puncture path planning method based on straight-line constraints provided in another exemplary embodiment of this application. Figure 2 As shown in the embodiments of this application, the puncture path planning method based on linear constraints constructs a displacement vector field and an organ motion heatmap based on CT image data, including:

[0045] Step 200: Based on the CT image data, obtain the phase-displacement mapping table and three-dimensional volume data.

[0046] For example, CT image data refers to the projection signals directly acquired by the CT scanner detector without image reconstruction. The acquisition time of each projection data is marked by a timestamp, bound to the respiratory displacement signal, and converted into three-dimensional volume data by a filtered back projection reconstruction algorithm.

[0047] For example, a phase-displacement mapping table is a quantitative correspondence table that associates discrete respiratory phase numbers with the spatial displacement of chest wall marker points. The average height displacement of the marker points under a specific respiratory phase is used to represent the spatial displacement.

[0048] Step 202: Calculate the displacement vector field based on the phase-displacement mapping table and the three-dimensional volume data.

[0049] For example, three-dimensional volume data refers to a digital three-dimensional matrix reconstructed by filtered back projection, whose basic unit is a voxel. It is used for non-rigid registration with reference to the volume data of the reference phase to generate a displacement vector field, calculate instantaneous velocity based on voxel displacement, and construct organ motion heatmap.

[0050] For example, the displacement vector field refers to a mathematical model describing the three-dimensional spatial position changes of an organ during the respiratory cycle. It is a three-dimensional vector matrix generated by a non-rigid registration algorithm, where each vector represents the displacement of a voxel from the reference phase to the target phase.

[0051] Step 204: Calculate the average motion velocity of each voxel based on the displacement vector field, and use the average motion velocity as the heat map value to generate an organ motion heat map.

[0052] For example, respiratory phase refers to the time state marker after discretizing the continuous respiratory cycle, which is used to quantify the dynamic position of organs during respiratory movement. Phase numbers are assigned by the displacement of chest wall marker points to generate a phase-annotated CT dataset.

[0053] For example, a voxel is the smallest unit of three-dimensional volume data, containing spatial coordinates and CT values. Each vector in the displacement vector field corresponds to a voxel, and the organ motion heatmap value is calculated based on the average velocity of the voxels.

[0054] For example, organ motion thermogram refers to a three-dimensional probability distribution model generated based on eight sets of displacement fields during the respiratory cycle, a three-dimensional matrix aligned with the CT image space, where each voxel stores a heat map value to quantify the average displacement risk of the corresponding anatomical location during respiratory motion.

[0055] Figure 3 The diagram shown is a flowchart illustrating a puncture path planning method based on straight-line constraints provided in another exemplary embodiment of this application. Figure 3 As shown in the embodiment of this application, the puncture path planning method based on linear constraints calculates the displacement vector field according to the phase-displacement mapping table and three-dimensional volume data, calculates the average motion velocity of each voxel according to the displacement vector field, and uses the average motion velocity as the heat map value to generate an organ motion heat map, including the following steps.

[0056] Step 300: Based on the phase-displacement mapping table and three-dimensional volume data, set the end-expiratory phase as the reference phase, and calculate the displacement vector field of other respiratory phases relative to the reference phase.

[0057] Specifically, the input is the registered 3D volume data {phase0:R} CT [0], phase 1: R CT [1], ..., phase 7: R CT [7]}. Input the phase-displacement mapping table {phase 0 = this period's minimum displacement, ..., phase 4 = this period's maximum displacement, ..., phase 7 = this period's minimum displacement}. Check the spatial consistency of the 3D volume data after registration of the 8 phases (the matrix size is 512×512×300). Verify the continuity of phase numbering by checking whether phases 0 to 7 are a complete sequence. Using phase 4 as the reference phase marker, output the verified 3D volume data. Based on the verified 3D volume data and the phase-displacement mapping table, perform the following for each non-reference phase (phase≠4): use the CT image of the current phase as a floating image, use the reference phase CT (phase4) as the reference image, and register using the Demons algorithm to generate a displacement vector field.

[0058] The formula for the displacement vector field is: D(x, y, z, t) = [dx, dy, dz];

[0059] Where (x, y, z) represent voxel coordinates; dx, dy, dz are the displacement components (in mm) of the voxel from the reference phase to phase t; t represents the target phase numbered {0, 1, 2, 3, 4, 5, 6, 7}, corresponding to 8 phases.

[0060] It should be understood that the organ displacement is minimal at the end of expiration (phase 4). min Since the anatomical structure is most stable, the end of expiration (phase 4) is chosen as the benchmark.

[0061] Step 302: Based on the displacement vector field, calculate the instantaneous velocity of each voxel at different breathing phases, and calculate the average velocity of each voxel based on the instantaneous velocity.

[0062] Specifically, based on the displacement vector field D(x, y, z, t), a zero displacement field is generated for phase 4 itself through an interpolation compensation algorithm: D(x, y, z, 4) = [0, 0, 0].

[0063] Based on the displacement vector field D(x, y, z, t), 7 sets of displacement fields are generated for the non-reference phase. Combined with the zero displacement field, 8 sets of displacement fields are obtained. The displacement field data structure is [x, y, z] → [dx, dy, dz].

[0064] Step 304: Use the average motion velocity as the heat map value to generate an organ motion heat map.

[0065] Specifically, based on the eight sets of displacement vector fields, the instantaneous velocity V(x, y, z, t) is calculated for each voxel coordinate (x, y, z) at different phases t:

[0066] V(x, y, z, t) = sqrt(dx) 2 + dy 2 + dz 2 ) / Δt;

[0067] Where Δt = 0.5 seconds, Δt represents a fixed time interval; sqrt(dx 2 + dy 2 + dz 2 ) represents the magnitude of the displacement vector.

[0068] Based on the 8 sets of displacement fields and instantaneous velocities V(x, y, z, t), calculate the average velocity V_avg(x, y, z) for each voxel coordinate (x, y, z):

[0069] V_avg(x, y, z) = (1 / 8) * Σ V(x, y, z, t);

[0070] Where Σ represents the summation of instantaneous velocities over the eight phases.

[0071] Based on the average motion velocity V_avg(x, y, z), an organ motion heatmap is created as a matrix with dimensions of 512×512×300. Each voxel includes a heatmap value, heatmap value = V_avg(x, y, z), where the heatmap value ranges from [0, 10 mm / s].

[0072] For example, organ motion thermogram refers to a three-dimensional probability distribution model generated based on eight sets of displacement fields during the respiratory cycle, a three-dimensional matrix aligned with the CT image space, where each voxel stores a heat map value to quantify the average displacement risk of the corresponding anatomical location during respiratory motion.

[0073] Based on the organ motion heatmap, a 3×3×3 convolution kernel is used to filter the heatmap to eliminate local noise caused by registration.

[0074] Based on the organ motion heatmap, the baseline phase organ segmentation results are loaded, and the average motion velocity V_avg(x, y, z) of non-organ regions is forced to zero. Organs include the lungs, liver, and large blood vessels.

[0075] The organ motion heatmap is linearly scaled to the [0, 10] range to obtain the optimized organ motion heatmap.

[0076] Heatmap value = 10 * (original value - minimum heatmap value) / (maximum heatmap value - minimum heatmap value);

[0077] Among them, the maximum and minimum values ​​of the heatmap are the calculated maximum and minimum average motion speeds.

[0078] The puncture path planning method based on linear constraints provided in this application solves the problems of organ deformation caused by respiratory motion, the risk of tissue dynamic displacement ignored by traditional path planning, and the interference of heat map noise and non-organ regions by non-rigid registration, calculating average movement velocity, Gaussian smoothing, organ mask constraint and normalization, and accurately generating voxel-level displacement vector field.

[0079] Figure 4 The diagram shown is a flowchart illustrating a puncture path planning method based on straight-line constraints provided in another exemplary embodiment of this application. Figure 1 This application extends from the embodiments shown. Figure 4 The illustrated embodiment will be described in detail below. Figure 4 The illustrated embodiments and Figure 1The differences between the embodiments shown are not repeated here, and the similarities are not repeated here.

[0080] like Figure 4 As shown in the embodiments of this application, the puncture path planning method based on linear constraints determines candidate intermediate points in the thoracic and abdominal organ regions based on preset skin needle entry points, tumor target points, and organ motion heat maps, and generates a set of key points, including the following steps.

[0081] Step 400: Verify that the skin needle entry point is located above the body surface contour or within a first preset distance from the body surface contour, and / or verify that the tumor target point is located within the tumor segmentation boundary and is greater than a second preset distance from the tumor segmentation boundary.

[0082] It should be understood that the first preset distance ranges from [1, 3 mm]. The second preset distance ranges from [2, 5 mm]. The first and second preset distances can be selected according to the actual situation.

[0083] In one embodiment, a skin needle entry point and a tumor target point can be obtained. If the skin needle entry point is more than 2 mm away from the body surface contour, the skin needle entry point is reselected. If the tumor target point is not within the tumor segmentation area or the tumor target point is within the tumor segmentation area but is less than 2 mm away from the tumor segmentation boundary, the tumor target point is reselected.

[0084] Specifically, this involves obtaining the optimized organ motion heatmap and inputting the skin needle insertion point S specified by the doctor. x s , y s , z s ), Tumor target T = ( x t , y t , z t ) and reference phase CT images.

[0085] If the distance between the skin entry point S and the body surface contour is greater than 2mm, it is determined that the entry point is not within the body surface contour, and a new skin entry point is selected.

[0086] If the tumor target is not within the tumor segmentation area, or if the tumor target is within the tumor segmentation area but is less than 2 mm from the tumor segmentation boundary, then the location coordinates of the tumor target do not meet the puncture path requirements, and a new tumor target is selected.

[0087] Step 402: Calculate the main puncture direction based on the skin entry point and the tumor target point.

[0088] Specifically, based on the skin entry point S = ( xs , y s , z s ) and tumor target T = ( x t , y t , z t ), Calculate the main puncture direction:

[0089] ;

[0090] Where ||TS|| represents the Euclidean distance from the skin needle insertion point S to the tumor target point T.

[0091] Step 404: Calculate the sampling plane based on the skin entry point and the main puncture direction.

[0092] For example, the sampling plane refers to a set of equidistant planes perpendicular to the puncture direction, with a default plane spacing of 10mm, used for candidate intermediate point search under anatomical constraints and for organ boundary identification in planar CT slices using the Canny algorithm.

[0093] Specifically, based on the skin entry point S = ( x s , y s , z s ) and the main puncture direction, calculate the sampling plane:

[0094] ;

[0095] in, P k Indicates the first k Reference points of each sampling plane, k It is the plane index, with values ​​{ k = 1, 2, ...,N};d plane The plane spacing is set to 10mm in this embodiment.

[0096] For example, k Reference points P k The plane in which the reference points are located is the sampling plane. Therefore, by calculating multiple reference points, the sampling plane is obtained.

[0097] According to the skin entry point S = ( x s , y s , z s) and tumor target T = ( x t , y t , z t ), calculate the total number of planes N, and obtain N planes perpendicular to the puncture direction.

[0098] ;

[0099] Among them, the function floor () indicates the rounding down operation.

[0100] Step 406: The intersection of the sampling plane and the thoracic and abdominal organs is taken as the candidate intermediate point.

[0101] For example, the candidate intermediate point is the intersection of the sampling plane perpendicular to the puncture direction and the organ boundary, serving as the turning anchor point of the two-segment path to provide multiple alternative path branches.

[0102] Step 408: Generate a set of key points based on the candidate intermediate points and the displacement vector field.

[0103] Specifically, based on the sampling plane and the reference phase CT image, the Canny edge detection algorithm is used to extract the organ contours in the CT image for each plane, calculate the organ boundary equations between the plane and the organs (lung, liver, large blood vessels), and solve for the intersection point X of the boundary:

[0104] ;

[0105] If the line connecting the intersection point X to the skin needle insertion point S and the tumor target point T is within the interval [0, 5 mm], and the heat map value of the intersection point X is within the interval [0, 7 mm / s], then the intersection point X is added to the candidate intermediate point set.

[0106] After the traversal is complete, the set of candidate intermediate points is obtained. M = { M 1, M 2, ..., M k},in, M k This represents the coordinates of the k-th candidate intermediate point.

[0107] Based on the displacement vector field D(x, y, z, t) and the candidate intermediate point set M = { M 1, M 2, ..., M kFor each coordinate point P(x, y, z) and phase t, calculate the compensated coordinates at phase t: Pt = (x0, y0, z0) + D(x, y, z, t); where (x0, y0, z0) represents the spatial coordinates of the reference phase (t=4); and D(x, y, z, t) represents the displacement vector field.

[0108] For example, the displacement vector field D(x, y, z, t) is used to filter intermediate points with small phase fluctuations.

[0109] Based on the compensated coordinates Pt, the following four-dimensional keypoint set is generated:

[0110] ;

[0111] Among them, S t M is the skin insertion point at phase t; k,t It is the k-th candidate intermediate point at phase t; T t It is the tumor target at phase t.

[0112] Filter the four-dimensional keypoint set; if the skin needle entry point S... t To candidate intermediate point M k,t If there are coordinate points on the line segment with a CT value greater than 300HU (passing through the skeleton), they are removed from the four-dimensional keypoint set.

[0113] If the candidate intermediate point M k,t To the tumor target T t If a line segment exceeds or does not reach the tumor area, it will be removed from the four-dimensional keypoints.

[0114] Based on the four-dimensional keypoint set, calculate the candidate intermediate point M. k,t Standard deviation of positional fluctuation σ k :

[0115] ;

[0116] in, μ k M represents k At the average position of the 8 phases μ k The calculation formula is .

[0117] Set organ thresholds based on organ type {lung: threshold 5mm, liver: threshold 3mm, default: 2mm}. For different organ regions, the standard deviation of the candidate midpoint's fluctuation is considered. σ k If the value is greater than 0 and less than the organ threshold, then the intermediate point is retained.

[0118] The puncture path planning method based on linear constraints provided in this application solves the problems of insufficient degrees of freedom in manually planned paths and intermediate points located in high-risk or invalid areas by generating the puncture main direction plane and filtering by combining heat map values ​​and distances. It provides a candidate path space with anatomical structure constraints and filters safe nodes with low movement speeds.

[0119] Figure 5 The diagram shown is a flowchart illustrating a puncture path planning method based on straight-line constraints provided in another exemplary embodiment of this application. Figure 1 This application extends from the embodiments shown. Figure 5 The illustrated embodiment will be described in detail below. Figure 5 The illustrated embodiments and Figure 1 The differences between the embodiments shown are not repeated here, and the similarities are not repeated here.

[0120] like Figure 5 As shown in the embodiment of this application, the puncture path planning method based on linear constraints calculates the comprehensive weight of the path edges according to the organ motion heatmap, and constructs a weighted directed graph based on the path edges and the comprehensive weight, including the following steps.

[0121] Step 500: Calculate the length of the path edge based on the starting and ending coordinates of the path edge.

[0122] Specifically, according to the connection rules, the coordinates of the starting node of the connection are set to A = ( x a , y a , z a The coordinates of the endpoint node are B = ( x b , y b , z b ), calculate the length L(A, B) of the path edge AB:

[0123] ;

[0124] Step 502: Based on the heatmap values ​​of the organ motion heatmap, the path edges are discretized, and the risk terms of the path edges are calculated.

[0125] Specifically, the path edge AB is discretized into 100 points p. i {( x × i / 100), ( y × i / 100), ( z × iCalculate the risk term R(A, B) using the formula: / 100)}.

[0126] ;

[0127] Wherein, V(p) i () is point p i The motion heatmap value at the location.

[0128] Step 504: Calculate the comprehensive weight of the path edges based on the length and risk terms.

[0129] Specifically, the overall weight W(A,B) is calculated based on the length term L(A,B) and the risk term R(A,B): W(A,B) = 0.7 × L(A,B) + 0.3 × R(A,B).

[0130] Step 506: Perform safety filtering on the path edges, and remove path edges that pass through preset dangerous organs, and / or path edges whose angle with the reference path exceeds a preset angle threshold.

[0131] It should be understood that the baseline path refers to the path from the starting point to the candidate intermediate point in a two-segment path. The preset angle threshold can be determined according to the actual situation, and this application does not impose specific limitations.

[0132] In one embodiment, the path edges and the comprehensive weights are used to form a path edge set. If any point of the path edge exceeds a preset value range, or the CT value of any point of the path edge is greater than a preset threshold, or any point of the path edge passes through any organ, or the angle between adjacent paths of the path edge is greater than an angle threshold, then the path edge is deleted from the path edge set.

[0133] Specifically, the coordinates of path edge AB and the comprehensive weight W(A, B) are used to form a weighted set of path edges (A, B, W).

[0134] If there exists a point V(p) on the path edge AB i If ) ∈ [7mm, +∞), then delete the path edge from the path edge set (A, B, W).

[0135] If there exists a point V(p) on the path edge AB i If the CT value of a path edge is greater than 300HU, then the path edge is deleted from the path edge set (A, B, W).

[0136] If there exists a point V(p) on the path edge AB i If a path passes through any organ, then that path edge is removed from the path edge set (A, B, W).

[0137] If for S t →M k,t →T t Path, angle between adjacent paths If the path edge is removed from the path edge set (A, B, W), then that path edge will be deleted.

[0138] For example, the angle θ between adjacent paths is calculated in the reference phase (phase4) coordinate system.

[0139] Step 508: Construct a weighted directed graph using the retained path edges after filtering and the overall weights.

[0140] For each node in the filtered path edge set, create a weighted directed graph. If any node exists within the preset radius threshold range of the current node, add it to the weighted directed graph.

[0141] Specifically, the filtered set of path edges (A, B, W) is organized, and a weighted directed graph is created for each node. If any node exists within a radius of 5mm of the current node, it is called a neighbor node and added to the weighted directed graph. The storage format is: node ID: [(neighbor ID, weight), ...].

[0142] The puncture path planning method based on linear constraints provided in this application solves the problems of path evaluation relying solely on geometric distance and path crossing infeasible areas by jointly calculating path length and risk score and real-time filtering of bone / blood vessel / thermograph thresholds, thus ensuring the spatiotemporal continuity and clinical feasibility of the path.

[0143] Figure 6 The diagram shown is a flowchart illustrating a puncture path planning method based on straight-line constraints provided in another exemplary embodiment of this application. Figure 1 This application extends from the embodiments shown. Figure 6 The illustrated embodiment will be described in detail below. Figure 6 The illustrated embodiments and Figure 1 The differences between the embodiments shown are not repeated here, and the similarities are not repeated here.

[0144] like Figure 6 As shown in the embodiments of this application, the puncture path planning method based on straight line constraints calculates the cost of each candidate path according to the comprehensive weight, and selects the final puncture path from the candidate path set based on the cost, including the following steps.

[0145] Step 600: Based on the weighted directed graph, add single-segment paths that do not pass through candidate intermediate points and double-segment paths that pass through candidate intermediate points to the candidate path set.

[0146] Specifically, based on the weighted directed graph, the path is divided into two segments with the same phase, starting from the starting point S. t →Midpoint M k,t →End point T t The starting point S of a single path segment in phase t →End point Tt For each phase t ∈ [0, 7], add all S t →T t Connection and S t →M k,t →T t Connect and record the node sequence of each path to generate a candidate path set {(x, y, z, t, V), ...}, which includes the node sequence (x, y, z) and phase information (t, V).

[0147] Step 602: Calculate the cost of a single path segment based on the comprehensive weight.

[0148] Specifically, according to the comprehensive weight calculation formula W(A, B), the starting point S is... t To the finish line T t Single-segment path cost: C single = W(S t T t ).

[0149] Step 604: Based on the comprehensive weight, calculate the cost from the starting point of the path edge to the candidate intermediate point and the cost from the candidate intermediate point to the ending point of the path edge. Add the cost from the starting point to the candidate intermediate point and the cost from the candidate intermediate point to the ending point to obtain the cost of the two-segment path.

[0150] Specifically, according to the comprehensive weight calculation formula W(A, B), the starting point S is... t To the midpoint M k,t Midpoint M k,t To the finish line T t Calculate the cost of a two-segment path: C double = W(S t M k,t ) + W(M k,t T t ).

[0151] Calculate the additional penalty term for a two-segment path: P switch = 0.1 × |t M - t S |, where |t M - t S | represents the phase difference between time M and time S.

[0152] It should be understood that the puncture process of the two-segment path can be carried out at different phases, for example, at time M, when the phase is t. M The puncture path starts from the path origin S. t →Midpoint M k,t At time S, the position is t S The puncture path starts from the midpoint M. k,t→End point T t .

[0153] Step 606: Calculate the additional penalty term for the two-segment path based on the phase difference, and combine it with the cost of the two-segment path to obtain the final cost of the two-segment path.

[0154] Specifically, the final cost of the two-segment path is calculated based on the cost of the two-segment path and the additional penalty term: C total =C double + P switch Based on the candidate path set {(x, y, z, t, V), ...}, the cost of a single path segment, and the final cost of a two-segment path, a costed path set {(x, y, z, t, V, ...} is generated. total / C single ), ...}.

[0155] Step 608: Sort the cost of the single-segment path and the final cost of the double-segment path in ascending order, set constraints, and select the path with the minimum cost from the candidate path set that satisfies the constraints as the final puncture path.

[0156] Specifically, the cost C of the entire candidate path set single and C total Sort the candidate paths in ascending order, define constraints, and consider the set of candidate paths with costs that satisfy the constraints {(x, y, z, t, V, C}. total / C single The path with the lowest cost among (), ..., is selected as the final puncture path, where the constraints are:

[0157] .

[0158] For example, the constraint condition represents selecting path edges from the candidate path set that satisfy the condition that the angle between adjacent paths is less than or equal to 30° and the heat map value of the path edge is less than or equal to 7 mm / s.

[0159] For a two-segment path, calculate the optimal midpoint position: Where δ represents the angle fine-tuning threshold, and in this embodiment, the default δ = ± 2mm is set; the function is optimized by adjusting the angle fine-tuning threshold δ. The value of is used to find the optimal midpoint position.

[0160] Colors are assigned based on the risk item R(A, B): if R(A, B) = 0, the path is green; if R(A, B) ∈ (0, 5], the path is yellow; and if R(A, B) ∈ (5, 7], the path is red. Motion heatmap slices are extracted along the final puncture path, and the points closest to the hazardous organs are marked to generate a safety profile.

[0161] The puncture path planning method based on linear constraints provided in this application solves the problems of joint optimization of multi-path structures and the limitation of the optimal path by local optimal solutions by enumerating the cost of single / double-segment paths and global search of the minimum cost path. It covers the cost space of all-phase candidate paths and outputs the spatiotemporally optimal puncture path.

[0162] Figure 7 The diagram shown is a flowchart illustrating a puncture path planning method based on straight-line constraints provided in another exemplary embodiment of this application. Figure 2 This application extends from the embodiments shown. Figure 7 The illustrated embodiment will be described in detail below. Figure 7 The illustrated embodiments and Figure 2 The differences between the embodiments shown are not repeated here, and the similarities are not repeated here.

[0163] like Figure 7 As shown in the embodiment of this application, the puncture path planning method based on linear constraints obtains a phase-displacement mapping table and three-dimensional volume data based on CT image data, including the following steps.

[0164] Step 700: Divide the patient's chest and abdominal CT image data into respiratory phases and record the phase shift during the respiratory process.

[0165] Specifically, the process involves inputting physiological signals from the patient's free breathing in the chest and abdomen; inputting raw detection data from a 256-slice spiral computed tomography (CT) scanner; attaching three marker points (including reflective beads) to the patient's chest wall; and recording the three-dimensional coordinates of the marker points at a frequency of 30Hz using infrared optical tracking to generate a respiratory displacement curve: disp(t) = average height change of the marker points. A CT scan is triggered when the respiratory displacement reaches a preset phase point, with each phase scan lasting 0.5 seconds, for a total of 8 phases acquired. The preset phase points refer to the 0%, 12.5%, ..., 87.5%, and 100% cyclic positions of a respiratory process.

[0166] By fixing a positioning coordinate system (including visible markers) on the CT bed, the rigid transformation matrix between the three-dimensional coordinates of infrared optical tracking and the positioning coordinates of the CT bed is calculated to obtain time-stamped CT image data.

[0167] For example, the positioning coordinate system refers to establishing a coordinate system with the center point of the CT bed as the origin, the horizontal direction of the CT bed as the X-axis, the vertical direction of the CT bed as the Y-axis, and the vertical direction of the CT bed as the Z-axis.

[0168] For example, CT image data includes eight phases and timestamps.

[0169] Based on the CT image data, the respiratory displacement-time series is output as: {[t0, disp(0)], [t1, disp(1)], ..., [t7, disp(7)]}. A 4×4 spatial transformation matrix is ​​obtained based on the CT image data and the respiratory displacement-time series.

[0170] For example, the spatial transformation matrix refers to the transformation of the three-dimensional coordinates of infrared optical tracking into the positioning coordinates of the CT bed.

[0171] Step 702: Number the breathing phases to obtain a phase-displacement mapping table.

[0172] Specifically, the input respiratory displacement-time series is {[t0, disp(0)], [t1, disp(1)], ..., [t7, disp(7)]}.

[0173] Identify continuous respiratory cycles, collect the peak of the current respiratory cycle to the peak of the next respiratory cycle, and if the fluctuation amplitude is less than the amplitude threshold, discard irregular respiratory cycles and divide each cycle into 8 phase intervals; wherein, in this embodiment, the amplitude threshold is set to 2mm.

[0174] Assign a phase number (0 to 7) to each CT scan moment: Phase number = round(7 × (current displacement - minimum displacement in this cycle) / (displacement amplitude in this cycle)). Where the phase number is an integer from 0 to 7, representing eight equal parts of the respiratory cycle; the function round() represents rounding to the nearest integer; the current displacement represents the average height of the marker point at time t; the minimum displacement in this cycle represents the lowest position of the marker point in the current respiratory cycle; the displacement amplitude in this cycle = maximum displacement in this cycle - minimum displacement in this cycle.

[0175] Based on the phase number, the phase-annotated CT dataset is obtained: {CT_phase0, CT_phase1, ..., CT_phase7}. Based on the phase number, the phase-displacement mapping table is obtained: {phase0 = minimum displacement of this period, ..., phase4 = maximum displacement of this period, ..., phase7 = minimum displacement of this period}.

[0176] Step 704: Based on the CT image data, perform filtered backprojection reconstruction on each phase to obtain three-dimensional volume data.

[0177] Specifically, based on CT image data, filtered backprojection reconstruction is performed on each phase data to generate 8 sets of 3D volume data R. CT [ x (a 512×512×300 matrix), where, x It represents {0, 1, 2, 3, 4, 5, 6, 7}, corresponding to 8 phases.

[0178] For example, filtered back projection reconstruction refers to applying a specific filter to CT image data to suppress high-frequency noise and sharpen edges, then projecting the filtered CT image data back along the ray path into the image space, and superimposing it to generate a cross-sectional image. In this application, it is used to convert CT image data into three-dimensional volume data, providing an artifact-free image basis for subsequent respiratory phase segmentation and displacement field calculation.

[0179] For example, three-dimensional volume data refers to a digital three-dimensional matrix composed of cross-sectional images, where each basic unit is called a voxel, and each voxel includes positioning coordinates and CT values.

[0180] Using end-expiratory phase (phase 4) as the baseline, non-rigid registration is performed on phase data other than phase 4 using the Demons algorithm, outputting registered 3D volume data. Based on the registered 3D volume data, a spatially aligned 4D-CT dataset is output: {phase0:R} CT [0], phase 1: R CT [1], ..., phase 7: R CT [7]}.

[0181] For example, non-rigid registration is used to align all phases to the same spatial coordinate system.

[0182] The puncture path planning method based on linear constraints provided in this application solves the problems of spatiotemporal asynchrony between intraoperative respiratory signals and CT scans, phase division errors caused by irregular breathing, and inconsistencies in the spatial coordinate systems of multi-phase CT images by binding respiratory marker points and CT-triggered scanning, normalizing respiratory cycles and nonlinear phase allocation, and filtering back projection reconstruction and cross-phase non-rigid registration. It achieves millisecond-level precision in synchronous acquisition of multimodal data and generation of an 8-phase respiratory model that strictly corresponds to the physiological state.

[0183] In another embodiment, the puncture path planning method based on linear constraints provided in this application includes the following steps: based on preset connection rules, at least one path edge is formed by connecting the skin needle entry point, candidate intermediate point and tumor target point in the key point set.

[0184] Based on the organ motion heatmap, the heatmap value of each node in the keypoint set is linearly fitted.

[0185] Specifically, based on the optimized four-dimensional keypoint set and the optimized organ motion heatmap, the heatmap value V corresponding to the straight line segment where each node is located is calculated by linear fitting, and then marked in the optimized four-dimensional keypoint set, i.e., the skin needle entry point S. t = (x, y, z, t, V), candidate intermediate point M k,t= (x, y, z, t, V), tumor target T t = (x, y, z, t, V).

[0186] For example, a node refers to a candidate intermediate point, a skin needle insertion point, and a tumor target point in the optimized four-dimensional keypoint set.

[0187] Based on the key point set and preset connection rules, the skin needle entry point, candidate intermediate point and tumor target point are connected to obtain the path edge. The connection rules include in-phase connection rules, topological constraint rules and prohibition connection rules.

[0188] Specifically, define the connection rules:

[0189] 1. Same-phase connection rule: Only nodes with the same phase t are allowed to connect to each other.

[0190] 2. Topological constraint rules: Connections must conform to the puncture sequence: skin entry point S t → Candidate intermediate point M k,t →Tumor target T t Or starting point S t →End point T t Candidate intermediate points cannot be connected by skipping steps.

[0191] 3. Prohibited Connection Rules: Cross-phase connections, candidate intermediate point interconnections, reverse connections, and using tumor targets as skin injection points are prohibited.

[0192] Based on the connection rules, a five-dimensional keypoint matrix that conforms to the connection rules is generated. The five-dimensional keypoint matrix includes three-dimensional coordinates (x, y, z), phase t, and heatmap values ​​V of the path edges.

[0193] The puncture path planning method based on linear constraints provided in this application solves the risk of cross-phase path breakage or reverse puncture by using in-phase connections and topological constraint rules, ensuring the spatiotemporal continuity and clinical feasibility of the path.

[0194] In another embodiment, after completing the puncture path planning, the method further includes:

[0195] Based on the coordinates of the chest wall marker (Marker) now The system includes a respiratory phase-displacement mapping table, organ motion thermograms, and the final puncture path, and calculates the displacement d of the marker point in real time. current = ||Marker now - Marker ref ||;Among them, Marker ref These are the coordinates of the reference position at the end of exhalation. If the displacement of the marker point is d... current < d maxIf the marker point is currently in the inhalation phase, calculate the phase time t of the marker point. real = (d current - d min ) / α. Where, d max Indicates the maximum phase displacement; d min α represents the minimum phase displacement; α represents the average slope of the intake section.

[0196] If the displacement of the marker point is d current ≥ d max If the marker point is currently in the platform segment, calculate the marker point phase time t. real :

[0197] ;

[0198] Here, γ represents the proportion of time the plateau segment takes up in the entire respiratory cycle.

[0199] If the displacement of the marker point is d min ≤ d current < d max If the marker is currently in the expiratory phase, calculate the phase time t of the marker. real :

[0200] ;

[0201] Where β represents the average slope of the expiratory phase.

[0202] Based on the phase t of the marked point real The final real-time phase t is obtained as t = round(7 × t real / t total ); where round() is the floor function; t total The duration of a complete respiratory cycle is expressed by the following formula:

[0203] .

[0204] For example, the plateau segment refers to the period of time during a respiratory cycle when the maximum chest wall displacement is maintained between the inspiratory and expiratory segments.

[0205] For a single path segment, extract the complete path S of phase t. t →T t For a two-segment path, extract path S. t →M k,t and path M k,t →T t Align and position coordinates according to the spatial transformation matrix: P nav = T × P CT .

[0206] Where T represents a 4×4 spatial transformation matrix, used to assist in transforming the CT coordinate system to the optical navigation coordinate system. The current path segment is displayed as a solid red line in the navigation interface, while adjacent phase paths are displayed as dashed gray lines.

[0207] For a single-segment path, during the preparation puncture phase, the state machine control center issues a guiding command: complete the full puncture at phase t. For a two-segment path, path S... t →M k,t The stage state machine control issues a guiding command: puncture to the midpoint at phase t. For path M, which is a two-segment path... k,t →T t The phased state machine controls and issues a guiding command: puncture to the tumor target site in the same phase t. The electromagnetic sensor tracks the position coordinates of the puncture needle tip (tip), and if ||tip - M... k,t If || ∈ [0, 2mm], then the puncture needle tip reaches the midpoint, and a text guidance command is output. For the current path segment A→B, interpolate the heat map value along the current path segment in real time. If V(p) is detected... i If the value is within the range of (5, 7 mm / s], the navigation interface will display a flashing red dot, and the puncture procedure will be stopped immediately.

[0208] Calculate the perpendicular distance d between the puncture needle tip and the preset path. prep = ||(tip - A) × d AB ||。 Where, d AB This is the direction vector of the current path segment. If the perpendicular distance d between the puncture needle tip and the preset path... prep If the deviation is greater than 3mm, the puncture operation is stopped and the needle is retracted for re-puncture. If the electromagnetic sensor detects a deviation of greater than 5mm between the needle tip and the preset path, a re-path planning command is triggered, and a path correction command or re-path planning request is output.

[0209] Figure 8 The diagram shown is a structural schematic of a puncture path planning system based on linear constraints provided in an exemplary embodiment of this application. Figure 8 As shown, the puncture path planning system based on straight line constraints provided in this application includes: a data acquisition and construction module 800, a key point generation module 802, a path edge generation module 804, a weighted directed graph construction module 806, a path planning module 808, and a path generation module 810.

[0210] The acquisition and construction module 800 is used to acquire CT image data of the patient's chest and abdomen at different respiratory phases, and construct a displacement vector field and organ motion heatmap based on the CT image data. The displacement vector field reflects organ motion information, and the organ motion heatmap represents the average motion velocity of each point in the organ region. The key point generation module 802 is used to determine candidate intermediate points in the chest and abdomen organ region based on preset skin needle insertion points, tumor target points, and organ motion heatmaps, and generate a key point set. The path edge generation module 804 is used to connect the skin needle insertion points, candidate intermediate points, and tumor target points in the key point set to form at least one path edge based on preset connection rules. The weighted directed graph construction module 806 is used to calculate the comprehensive weight of the path edges based on the organ motion heatmap, and construct a weighted directed graph based on the path edges and comprehensive weights. The path planning module 808 is used to perform path search in the weighted directed graph and generate a candidate path set. The path generation module 810 is used to calculate the cost of each candidate path based on the comprehensive weights, and select the final puncture path from the candidate path set based on the cost.

[0211] It should be understood that the operation and functions of the relevant modules mentioned in the puncture path planning system based on linear constraints can be referenced above. Figures 1 to 7 The proposed puncture path planning method based on straight-line constraints will not be elaborated upon here to avoid repetition.

[0212] Figure 9 The diagram shown is a structural schematic of an electronic device provided in an exemplary embodiment of this application. Figure 9 As shown, the electronic device 90 includes one or more processors 901 and memory 902.

[0213] The processor 901 may be a central processing unit (CPU) or other form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 90 to perform desired functions.

[0214] The memory 902 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 901 may execute the program instructions to implement the path edge costs, comprehensive weights, and / or other desired functions of the various embodiments of this application described above. Various contents such as keypoint sets and displacement vector fields may also be stored in the computer-readable storage medium.

[0215] In one example, the electronic device 90 may also include an input device 903 and an output device 904, which are interconnected via a bus system and / or other forms of connection mechanism (not shown).

[0216] The input device 903 may include, for example, a keyboard, a mouse, etc.

[0217] The output device 904 can output various information to the outside, including the puncture path. The output device 904 may include, for example, a display, a speaker, a printer, and a communication network and its connected remote output devices, etc.

[0218] Of course, for the sake of simplicity, Figure 9 Only some of the components of the electronic device 90 relevant to this application are shown in this illustration; components such as buses, input / output interfaces, etc., are omitted. In addition, the electronic device 90 may include any other suitable components depending on the specific application.

[0219] In addition to the methods and devices described above, embodiments of this application may also be computer program products, which include computer program instructions that, when executed by a processor, cause the processor to perform the steps in the linear constraint-based puncture path planning method according to various embodiments of this application described above.

[0220] Computer program products can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this application. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0221] Furthermore, embodiments of this application may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps of the puncture path planning method based on linear constraints according to various embodiments of this application described above.

[0222] Computer-readable storage media may take the form of any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may, for example, include, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0223] The basic principles of this application have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the application to the necessity of employing the aforementioned specific details for implementation.

[0224] The block diagrams of devices, apparatuses, devices, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0225] It should also be noted that in the apparatus, equipment, and methods of this application, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of this application.

[0226] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this application. Therefore, this application is not intended to be limited to the aspects shown herein, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0227] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this application to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A straight-line constraint based puncture path planning method, characterized in that, The method comprises the following steps: Collecting CT image data of a patient's chest and abdomen at different respiratory phases, constructing a displacement vector field and an organ motion heat map based on the CT image data, the displacement vector field reflecting organ motion information, and the organ motion heat map representing the average motion speed of each point in an organ region; Based on a preset skin needle entry point, a tumor target point, and the organ motion heat map, determining a candidate intermediate point in the chest and abdominal organ region, and generating a key point set; Based on a preset connection rule, connecting the skin needle entry point, the candidate intermediate point, and the tumor target point in the key point set to form at least one path edge; According to the organ motion heat map, calculating the comprehensive weight of the path edge, and constructing a weighted directed graph according to the path edge and the comprehensive weight; Searching for a path in the weighted directed graph to generate a candidate path set; According to the comprehensive weight, calculating the cost of each candidate path, and selecting a final puncture path from the candidate path set based on the cost, comprising the following steps: According to the weighted directed graph, adding a single-segment path not passing through the candidate intermediate point and a double-segment path passing through the candidate intermediate point to the candidate path set; According to the comprehensive weight, calculating the cost of the single-segment path; According to the comprehensive weight, calculating the cost from the start point of the path edge to the candidate intermediate point and the cost from the candidate intermediate point to the end point of the path edge, adding the cost from the start point to the candidate intermediate point to the cost from the candidate intermediate point to the end point to obtain the cost of the double-segment path; According to the phase difference, calculating an additional penalty term of the double-segment path, combining the cost of the double-segment path to obtain the final cost of the double-segment path; Arranging the cost of the single-segment path and the final cost of the double-segment path in ascending order, setting a constraint condition, and selecting the path with the minimum cost from the candidate path set that meets the constraint condition as the final puncture path.

2. The straight-line constraint-based puncture path planning method of claim 1, wherein, The method comprises the following steps: Based on the CT image data, obtaining a phase-displacement mapping table and three-dimensional body data; According to the phase-displacement mapping table and the three-dimensional body data, calculating the displacement vector field; According to the displacement vector field, calculating the average motion speed of each voxel, taking the average motion speed as a heat map value, and generating the organ motion heat map.

3. The straight-line constraint-based puncture path planning method of claim 2, wherein, The method comprises the following steps: According to the phase-displacement mapping table and the three-dimensional body data, setting the phase at the end of expiration as a reference phase, and calculating the displacement vector field of other respiratory phases relative to the reference phase; Based on the displacement vector field, calculating the instantaneous speed of each voxel at different respiratory phases, and calculating the average motion speed of each voxel according to the instantaneous speed; Taking the average motion speed as the heat map value, and generating the organ motion heat map.

4. The straight-line constraint-based puncture path planning method according to any one of claims 1-3, characterized in that, The candidate intermediate points are determined in the chest and abdominal organ region based on the preset skin needle entry point, tumor target point, and the organ motion heat map, and a key point set is generated, including: Verify that the skin needle entry point is located on the body surface contour or within a first preset distance from the body surface contour, and / or verify that the tumor target point is located within the tumor segmentation boundary and is greater than a second preset distance from the tumor segmentation boundary; Calculate the main direction of puncture according to the skin needle entry point and the tumor target point; Calculate the sampling plane according to the skin needle entry point and the main direction of puncture; The intersection of the sampling plane and the chest and abdominal organ is taken as the candidate intermediate point; Generate the key point set according to the candidate intermediate point and the displacement vector field.

5. The straight-line constraint-based puncture path planning method according to any one of claims 1-3, wherein, The comprehensive weight of the path edge is calculated according to the organ motion heat map, and a weighted directed graph is constructed according to the path edge and the comprehensive weight, including: According to the starting point coordinates and end point coordinates of the path edge, the length term of the path edge is calculated; The path edge is discretized according to the heat map value of the organ motion heat map, and the risk term of the path edge is calculated; According to the length term and the risk term, the comprehensive weight of the path edge is calculated; The safety of the path edge is filtered, and the path edge passing through the preset dangerous organ and / or the path edge with an angle exceeding the preset angle threshold with the reference path is removed; The path edge and the comprehensive weight retained after filtering are used to construct the weighted directed graph.

6. The straight-line constraint-based puncture path planning method of claim 2, wherein, The phase-displacement mapping table and the three-dimensional body data are obtained based on the CT image data, including: Divide the CT image data of the patient's chest and abdomen into respiratory phases, and record the phase displacement of the respiratory process; Phase numbering is performed on the respiratory phases to obtain the phase-displacement mapping table; According to the CT image data, filtered back projection reconstruction is performed on each phase to obtain the three-dimensional body data.

7. A straight-line constraint based puncture path planning system, characterized by, It includes: The acquisition and construction module is used to acquire CT image data of the patient's chest and abdomen at different respiratory phases, and to construct a displacement vector field and an organ motion heat map based on the CT image data, wherein the displacement vector field reflects the organ motion information, and the organ motion heat map represents the average motion speed of each point in the organ region; The key point generation module is used to determine candidate intermediate points in the chest and abdominal organ region based on a preset skin needle entry point, a tumor target point, and the organ motion heat map, and to generate a key point set; The path edge generation module is used to connect the skin needle entry point, the candidate intermediate point, and the tumor target point in the key point set to form at least one path edge based on a preset connection rule; The weighted directed graph construction module is used to calculate the comprehensive weight of the path edge according to the organ motion heat map, and to construct a weighted directed graph according to the path edge and the comprehensive weight; The path planning module is used to search for paths in the weighted directed graph to generate a candidate path set; The path generation module is used to calculate the cost of each candidate path according to the comprehensive weight, and to select a final puncture path from the candidate path set based on the cost, including the following steps: According to the weighted directed graph, a single-section path not passing through the candidate intermediate point and a double-section path passing through the candidate intermediate point are added into a candidate path set; According to the comprehensive weight, a cost of the single-section path is calculated; According to the comprehensive weight, a cost from a start point of the path edge to the candidate intermediate point and a cost from the candidate intermediate point to an end point of the path edge are added to obtain a cost of the double-section path; According to the phase difference, an additional penalty term of the double-section path is calculated, and a final cost of the double-section path is obtained by combining the cost of the double-section path; The cost of the single-section path and the final cost of the double-section path are arranged in ascending order, and a path with a minimum cost in the candidate path set satisfying a constraint condition is selected as the final puncture path.

8. An electronic device, comprising: Comprise: a processor; and a memory, wherein computer program instructions are stored in the memory, and the computer program instructions, when executed by the processor, cause the processor to perform the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, Computer program instructions are stored on a computer readable storage medium, and the computer program instructions, when executed by the processor, cause the processor to perform the steps of the method according to any one of claims 1 to 6.

Images