Interventional instrument path planning method applied to cerebral vascular intervention and related product

Abstract

The application discloses an intervention instrument path planning method applied to cerebrovascular intervention and related products. The intervention instrument path planning method applied to cerebrovascular intervention comprises the following steps: acquiring a cerebrovascular graph network of a target patient and a starting node and a terminal node of intervention of an intervention instrument; the cerebrovascular graph network is composed of nodes and edges between the nodes; discretizing the intervention instrument to be intervened to obtain an intervention instrument discrete model; placing the intervention instrument discrete model at the starting node on the cerebrovascular graph network, and then performing path planning on the intervention instrument on the cerebrovascular graph network by using a preset path search algorithm until the terminal node is reached, so that a target path reachable by the intervention instrument is obtained; in the path planning process, the deformation of the intervention instrument discrete model is simulated by using a preset simulation algorithm, and physical reachability of the intervention instrument is judged based on a simulation result. By implementing the application, the authenticity and reliability of path planning are improved.

Description

Technical Field

[0001] This application generally relates to the field of medical image processing technology. More specifically, this application relates to an interventional device path planning method and related products used in cerebrovascular interventions. Background Technology

[0002] Interventional cerebrovascular surgery is a minimally invasive neurosurgical or neurological procedure specifically designed to treat lesions in the blood vessels of the brain and neck. During this procedure, surgeons insert interventional devices (such as microcatheters or stents) through the femoral or radial artery into the cerebral blood vessels and guide them along the complex and winding pathways to the lesion site. Therefore, the choice of interventional device pathway directly impacts the success rate of the procedure and the risk of complications.

[0003] Currently, interventional device path planning is based on the geometry of blood vessels to find the shortest path. However, the deformation of the interventional device itself is not considered during the path planning process, resulting in paths that are physically unreachable in actual operation, leading to low reliability of the path planning.

[0004] In view of this, there is an urgent need to provide a method and related products for interventional device path planning in cerebrovascular intervention to improve the accuracy and reliability of path planning. Summary of the Invention

[0005] In order to at least solve one or more of the technical problems mentioned above, this application proposes an interventional device path planning method and related products for cerebrovascular intervention in several aspects.

[0006] In a first aspect, this application provides a method for path planning of interventional devices applied to cerebrovascular intervention, comprising: acquiring a cerebrovascular network of a target patient and a starting node and a termination node for interventional device intervention; wherein the cerebrovascular network consists of nodes and edges between nodes; discretizing the interventional device to be intervened to obtain a discrete model of the interventional device; placing the discrete model of the interventional device on the cerebrovascular network after the starting node, and using a preset path search algorithm to plan the path of the interventional device on the cerebrovascular network until the termination node is reached, thereby obtaining a target path reachable by the interventional device; wherein, during the path planning process, a preset simulation algorithm is used to simulate the deformation of the discrete model of the interventional device, and the physical reachability of the interventional device is determined based on the simulation results.

[0007] In some embodiments, acquiring the brain vascular network of the target patient includes: acquiring a three-dimensional vascular image of the target patient's brain; performing a vascular mask extraction operation on the three-dimensional vascular image of the brain to obtain a vascular mask; performing a vascular centerline extraction operation on the vascular mask to obtain the vascular centerline of each blood vessel on the vascular mask; and constructing the brain vascular network based on the vascular centerline.

[0008] In some embodiments, simulating the deformation of the discrete model of the interventional device using a preset simulation algorithm during path planning includes: when the preset path search algorithm determines the next node the discrete model of the interventional device will move from the current node to on the cerebral vascular network, obtaining the predicted positions of each particle when the discrete model of the interventional device moves to the next node; the next node is one of the adjacent nodes of the current node; when the predicted position does not meet the set constraints, calculating the correction amount of the discrete model of the interventional device based on the predicted position, the set constraints, and the positions of each particle on the discrete model of the interventional device; wherein the set constraints include at least one of the following: the discrete model of the interventional device cannot penetrate the blood vessel wall, the bending angle of the discrete model of the interventional device is less than or equal to a set angle threshold, and the discrete model of the interventional device is continuous; correcting the discrete model of the interventional device based on the correction amount to obtain the simulation result.

[0009] In some embodiments, determining the next node to which the discrete model of the interventional device can move from the current node on the cerebral vascular map network using the preset path search algorithm includes: calculating the total generation value of the discrete model of the interventional device moving to each candidate node on the cerebral vascular map network; wherein the candidate node is all the adjacent nodes that the discrete model of the interventional device can reach from the current node on the cerebral vascular map network; and determining the next node based on the total generation value of the discrete model of the interventional device moving to each candidate node on the cerebral vascular map network.

[0010] In some embodiments, the determination of the physical accessibility of the interventional device based on the simulation results of the discrete model of the interventional device includes: when the simulation results do not meet any of the set constraints, determining that the physical accessibility determination result is physically unreachable; otherwise, determining that the physical accessibility determination result is physically reachable.

[0011] In some embodiments, the method further includes: when the physical reachability judgment result is physically reachable, after pruning other candidate nodes, controlling the interventional device discrete model to move from the current node to the next node, and taking the next node as the new current node, returning to execute the step of determining the next node to which the interventional device discrete model moves from the current node on the cerebral vascular network using the preset path search algorithm; when the physical reachability judgment result is physically inaccessible, after pruning the next node, determining a new next node at the current node based on the total cost of the interventional device discrete model moving to each candidate node on the cerebral vascular network, until a next node indicating physical reachability is determined.

[0012] In some embodiments, the preset path search algorithm is A The algorithm is one of the following: Dijkstra's algorithm, RRT algorithm; and / or the preset simulation algorithm is one of the following: PBD simulation algorithm, XPBD simulation algorithm, mass-spring simulation algorithm.

[0013] In a second aspect, this application provides an interventional device path planning device for cerebrovascular intervention, comprising: a data acquisition module for acquiring the cerebrovascular network of a target patient and the start and end nodes of the interventional device intervention; a discretization module for discretizing the interventional device to be intervened into a discrete model of the interventional device; and a path planning module for placing the discrete model of the interventional device on the cerebrovascular network after the start node, and using a preset path search algorithm to plan the path of the interventional device on the cerebrovascular network until the end node is reached, thereby obtaining a target path reachable by the interventional device; wherein, during the path planning process, a preset simulation algorithm is used to simulate the deformation of the discrete model of the interventional device, and the physical reachability of the interventional device is determined based on the simulation results.

[0014] In a third aspect, this application provides an electronic device comprising: a processor configured to execute program instructions; and a memory configured to store the program instructions, which, when loaded and executed by the processor, cause the processor to perform an interventional device path planning method for cerebrovascular intervention as described in the first aspect or any of the embodiments of the first aspect.

[0015] In a fourth aspect, this application provides a computer-readable storage medium storing program instructions that, when loaded and executed by a processor, cause the processor to perform an interventional device path planning method for cerebrovascular intervention as described in the first aspect or any of the embodiments of the first aspect.

[0016] The interventional device path planning method and related products applied to cerebrovascular interventions, as described above, introduce a discrete model of the interventional device during the path planning process. A physical simulation algorithm (i.e., a preset simulation algorithm) is used to simulate the deformation of the discrete model. Based on the simulation results, the physical accessibility of the interventional device is determined. By coupling path planning and physical accessibility assessment during path planning, the realism and reliability of the path planning are improved. This method is applicable to preoperative planning and intraoperative assistance for cerebrovascular interventional surgeries such as cerebral aneurysms, cerebral vascular stenosis, and ischemic stroke. Furthermore, the physical accessibility of the path is determined during the path planning process, rather than afterward, which also improves the efficiency of path planning. Attached Figure Description

[0017] The above and other objects, features, and advantages of exemplary embodiments of this application will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the drawings, several embodiments of this application are illustrated by way of example and not limitation, and the same or corresponding reference numerals denote the same or corresponding parts, wherein:

[0018] Figure 1 An exemplary flowchart of an interventional device path planning method for cerebrovascular intervention, according to some embodiments of this application, is shown; Figure 2 Exemplary structural block diagrams of interventional device path planning devices for cerebrovascular interventions according to some embodiments of this application are shown; Figure 3 An exemplary structural block diagram of an electronic device according to some embodiments of this application is shown. Detailed Implementation

[0019] 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, not all, of the embodiments of this application. 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.

[0020] It should be understood that the terms "comprising" and "including" used in the specification and claims of this application indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0021] It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application. As used in this specification and claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this specification and claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations.

[0022] As used in this specification and claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if [described condition or event] is detected" may be interpreted, depending on the context, as "once determined," "in response to determination," "once [described condition or event] is detected," or "in response to detection of [described condition or event]."

[0023] The specific embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0024] Exemplary application scenarios Interventional cerebrovascular surgery is a minimally invasive neurosurgical or neurological procedure specifically designed to treat lesions in the blood vessels of the brain and neck. During this procedure, surgeons insert interventional devices (such as microcatheters or stents) through the femoral or radial artery into the cerebral blood vessels and guide them along the complex and winding pathways to the lesion site. Therefore, the choice of interventional device pathway directly impacts the success rate of the procedure and the risk of complications.

[0025] Currently, interventional device pathway planning is based on the geometry of blood vessels to search for the shortest path or the path with the least curvature. Many of these searched paths are geometrically reachable but physically inaccessible. Therefore, after obtaining the path, it is necessary to determine its physical reachability.

[0026] However, the physical characteristics and deformation of the interventional device itself were not considered during the path planning process, resulting in paths that were unreachable in actual operation, leading to low reliability of the path planning. Furthermore, performing physical reachability assessment only after a path has been found reduces the efficiency of path planning.

[0027] In view of this, there is an urgent need to provide a method and related products for interventional device path planning in cerebrovascular intervention, so as to achieve rapid path planning while improving the accuracy and reliability of path planning.

[0028] Figure 1An exemplary flowchart of an interventional device path planning method 100 for cerebrovascular intervention, according to some embodiments of this application, is shown. It is understood that the above-described interventional device path planning method 100 for cerebrovascular intervention can be executed by any suitable device with data processing capabilities, such as, but not limited to, terminal devices, processors, and servers.

[0029] like Figure 1 The aforementioned interventional device path planning method 100 applied to cerebrovascular intervention includes: Step S110: Obtaining the cerebrovascular network of the target patient and the starting and ending nodes of the interventional device intervention; wherein, the cerebrovascular network consists of nodes and edges between nodes; Step S120: Discretizing the interventional device to be intervened to obtain a discrete model of the interventional device; Step S130: Placing the discrete model of the interventional device on the cerebrovascular network after the starting node, and using a preset path search algorithm to plan the path of the interventional device on the cerebrovascular network until the ending node is reached, thereby obtaining the target path reachable by the interventional device; wherein, during the path planning process, a preset simulation algorithm is used to simulate the deformation of the discrete model of the interventional device, and the physical reachability of the interventional device is judged based on the simulation results.

[0030] For example, the target patient in step S110 above refers to a patient with cerebrovascular lesions (e.g., cerebral aneurysm, cerebral vascular stenosis, cerebral thrombosis, etc.). The cerebral vascular network refers to a topological structure that models the cerebral blood vessels of the target patient. In this embodiment, the cerebral vascular network consists of nodes and edges between nodes. Nodes can be bifurcation points of blood vessels, sampling points on the center line of blood vessels, etc. The line segment connecting two adjacent nodes is called an edge.

[0031] Specifically, the aforementioned cerebral vascular network can be used Let V represent the set of nodes and E represent the set of edges.

[0032] In this embodiment, the starting point of the interventional device intervention in step S110 is the entry point of the interventional device into the cerebral blood vessel, which is the starting point of the path planning; the ending point is the location of the lesion in the cerebral blood vessel, which is the ending point of the path planning. The lesion location can be, for example, the location of an aneurysm, the distal end of a stenotic segment, the site of thrombosis, etc., and this embodiment does not specifically limit it.

[0033] In this embodiment, the aforementioned cerebral vascular network can be obtained by modeling the acquired cerebral vascular images of the target patient. The specific process can be found in the examples described below, and will not be repeated here. The start and end nodes can be obtained by receiving selection information on the cerebral vascular network structure, i.e., the start and end nodes are manually selected on the cerebral vascular network structure.

[0034] For example, the interventional device in step S120 above is an interventional device used in cerebrovascular interventional surgery to reach the lesion and deliver embolization materials or drugs, such as microcatheters, stents, etc. In this embodiment of the application, the interventional device is discretized into a physical model, that is, the continuous interventional device is modeled as an ordered, discrete chain of particles (i.e., the discrete model of the interventional device). The motion state (position, velocity, etc.) of each particle can be controlled by a preset simulation algorithm to facilitate subsequent simulation.

[0035] This application uses microcatheters as an example of interventional devices. The discrete model of the interventional device (i.e., the microcatheter model) can be denoted as: , This is a discrete model of a microcatheter. The particles represent the discrete model of the microcatheter; N is the number of particles. In this embodiment, the microcatheter model can be divided into N particles, i.e. L is the length of the microcatheter. This represents the length of each particle.

[0036] Of course, conventional mass-spring models, continuous rod models, etc., can also be used to discretize interventional devices in the embodiments of this application. The embodiments of this application do not specifically limit the discretization method of interventional device models, and can be selected according to the actual situation.

[0037] For example, there can be many kinds of preset path search algorithms in step S130 above, such as A Algorithms, Dijkstra's algorithm, Rapid Expanding Random Tree (RRT) algorithm, etc., are not specifically limited to the preset path search algorithm in this application embodiment, and can be selected according to the actual situation.

[0038] Among them, the above A The algorithm is a directional shortest path search algorithm. Compared to ordinary shortest path search algorithms, it can be guided by a cost function to make A... The algorithm has better performance; therefore, the preferred embodiment of this application is A. The algorithm, and the following examples will also use A. The algorithm will be used as an example for description.

[0039] For example, there can be many kinds of preset simulation algorithms in step S130 above, such as position-based dynamics (PBD) simulation algorithm, force-based simulation algorithm, impulse-based simulation algorithm, etc. This application embodiment does not specifically limit this, and can be selected according to the actual situation.

[0040] The aforementioned PBD simulation algorithm is an extremely important physical simulation method in the fields of computer graphics and real-time simulation. Compared with force-based simulation and impulse-based simulation, PBD simulation has higher stability, computational speed, and ease of use, which can meet the real-time requirements of clinical practice and is widely used in medical surgical simulation. Therefore, the PBD simulation algorithm is preferred in the embodiments of this application, and the following embodiments are also described using the PBD simulation algorithm as an example.

[0041] Traditional A In the algorithm, nodes only contain location information, that is... The optimal path is found in the cerebral vascular network. In this embodiment, A... The algorithm was improved by incorporating the form of interventional instruments into the nodes, i.e. This expands the state space. Where X represents the set of particle positions in the discrete model of the interventional device, and x represents each particle, which is used to describe the current spatial form of the interventional device.

[0042] Therefore, path planning for interventional devices depends not only on the node locations (i.e., the geometry of the blood vessels) but also on the actual physical state of the interventional device. The specific path planning method is as follows: In this embodiment of the application, when performing path planning for the interventional device, the discrete model of the interventional device can be placed at the starting node of the cerebral vascular network. Specifically, the front end of the discrete model of the interventional device (e.g., the first two particles) is placed at the starting node, and the other particles of the discrete model of the interventional device are initialized along the structure of the discrete model of the interventional device.

[0043] Then, a preset path search algorithm is used to plan the path of the interventional device on the cerebral vascular network. Specifically, based on the cerebral vascular network, the path is explored step by step from the starting node to the ending node to obtain an optimal path (i.e., the target path). Specific implementation details can be found in the examples described below, and will not be repeated here.

[0044] In this embodiment, during the path planning process, a preset simulation algorithm can be used to simulate the deformation of the discrete model of the interventional device, and the physical accessibility of the interventional device can be determined based on the simulation results. Specifically, for each node the discrete interventional model advances, the preset simulation algorithm simulates the deformation of the discrete interventional device model under this advancement (e.g., bending, collision, etc.). Then, based on the simulation results, the physical accessibility of the interventional device is determined to see if the node the discrete interventional model is advancing towards is physically reachable. If physically reachable, the advancement continues; if physically unreachable, a new node is selected for simulation. This process is repeated until the front end of the interventional device model successfully reaches the terminal node, thus obtaining the target path.

[0045] This application's embodiments introduce a discrete model of the interventional device during the path planning process and use a physical simulation algorithm (i.e., a preset simulation algorithm) to simulate the deformation of the discrete model. Based on the simulation results, the physical accessibility of the interventional device is determined. By coupling path planning and physical accessibility assessment during path planning, the realism and reliability of path planning are improved. This approach is applicable to preoperative planning and intraoperative assistance for cerebrovascular interventional surgeries such as cerebral aneurysms, cerebral vascular stenosis, and ischemic stroke. Furthermore, by determining the physical accessibility of the path during the path planning process, rather than verifying it afterward, the efficiency of path planning is also improved.

[0046] As an optional embodiment of this application, obtaining a brain vascular network of a target patient includes: obtaining a three-dimensional vascular image of the target patient's brain; performing a vascular mask extraction operation on the three-dimensional vascular image of the brain to obtain a vascular mask; performing a vascular centerline extraction operation on the vascular mask to obtain the vascular centerline of each blood vessel on the vascular mask; and constructing a brain vascular network based on the vascular centerline.

[0047] For example, the aforementioned three-dimensional vascular images of the brain can be three-dimensional volume data such as computed tomography angiography (CTA) images, digital subtraction angiography (DSA) images, and magnetic resonance angiography (MRA) images. These can be acquired in real time using corresponding medical equipment or retrieved from a medical database. This application does not specifically limit the three-dimensional vascular images of the brain or the methods for acquiring them; these can be determined according to actual circumstances.

[0048] In this embodiment, after acquiring a three-dimensional brain vascular image, a vascular mask extraction operation is performed on the brain vascular image to extract the blood vessels from the brain vascular image, thus obtaining the vascular mask. Specifically, the vascular mask can be extracted using a deep learning model (e.g., 3D-Unet) or a region growing algorithm.

[0049] The specific process of extracting blood vessel masks using deep learning models and region growing algorithms is a conventional technique, which can be found in relevant literature and will not be elaborated here.

[0050] For example, the aforementioned vessel centerline refers to the axis connecting the centers of the vessel cross-sections. In this embodiment, after obtaining the vessel mask, a vessel centerline extraction operation is performed on the vessel mask to obtain the vessel centerlines of each vessel on the vessel mask. Specifically, the vessel centerlines can be extracted using 3Dskeletonization or a minimum path algorithm.

[0051] As for the specific methods for extracting the centerline of blood vessels using 3Dskeletonization or the minimum path algorithm, these are conventional techniques, and can be found in the descriptions in relevant materials, so they will not be elaborated here.

[0052] Then, a cerebral vascular network is constructed based on the vascular centerline. Specifically, the vascular centerline is sampled at fixed intervals to obtain a series of sampling points. These sampling points, as well as endpoints and bifurcation points, are used as nodes in the cerebral vascular network. The centerline segment between two adjacent nodes is used as the edge of the cerebral vascular network, thus obtaining the cerebral vascular network.

[0053] In this embodiment of the application, after obtaining the cerebral vascular network, the attribute information of each edge can also be calculated. This attribute information may include, but is not limited to, the edge length (i.e.,...). ), vessel radius (r) and curvature (i.e. ).

[0054] The curvature can be calculated using the following formula:

[0055] The above Represents the curvature at the i-th node; , , and represent the three-dimensional coordinates of the i-th, i-1-th, and i+1-th nodes on the central line of the blood vessel, respectively.

[0056] The embodiments of this application construct a cerebral vascular network using clinical standard images, which can be adapted to real surgical scenarios; and automatically extracts and models blood vessels, reducing manual intervention and improving modeling efficiency and accuracy; and the cerebral vascular network simplifies the topological structure of blood vessels in images, making subsequent path search and simulation easier.

[0057] As an optional embodiment of this application, the simulation of the deformation of the discrete model of the interventional device using a preset simulation algorithm during the path planning process in step S130 includes: when the preset path search algorithm determines the next node to which the discrete model of the interventional device will move from the current node on the cerebral vascular network, obtaining the predicted position of each particle when the discrete model of the interventional device moves to the next node; when the predicted position does not meet the set constraint conditions, calculating the correction amount of the discrete model of the interventional device based on the predicted position, the set constraint conditions, and the position of each particle on the discrete model of the interventional device; wherein, the set constraint conditions include at least one of the following: the discrete model of the interventional device cannot penetrate the blood vessel wall, the bending angle of the discrete model of the interventional device is less than or equal to the set angle threshold, and the discrete model of the interventional device is continuous; the discrete model of the interventional device is corrected based on the correction amount to obtain the simulation result.

[0058] For example, the aforementioned current node (denoted as...) ) refers to the current node in the discrete model of the interventional device; the next node (denoted as ) ) refers to using A The algorithm selects the optimal candidate node to move forward. This next node is adjacent to the current node. How the next node is determined is illustrated in the following example, and will not be repeated here.

[0059] In this embodiment of the application, after determining the next node, the discrete model of the interventional device is processed from the current node (i.e., based on a preset simulation algorithm) Proceed to the next node (i.e.) The deformation of the device is simulated to obtain the shape of the discrete model of the interventional device when it moves to the next node (i.e., the simulation result).

[0060] For example, the predicted position mentioned above is a temporary, uncorrected particle position obtained solely from the discrete model of the interventional device propelled by external force, without considering any constraints (e.g., length constraints, bending constraints, collision constraints, etc.).

[0061] In this embodiment of the application, after obtaining the predicted position of each particle, a judgment is made on whether the predicted position needs to be corrected based on the set constraints. For details, please refer to the following embodiment.

[0062] The constraints set above are predefined and may include at least one of the following: collision constraints, bending constraints, interventional device continuity constraints, etc., and each constraint has a corresponding constraint function.

[0063] The collision constraint refers to the requirement that the discrete model of the interventional device cannot penetrate the blood vessel wall, which can be specifically defined using a distance field.

[0064] in, Let i represent the distance field from the i-th particle to the blood vessel wall, when When the value is less than 0, it is considered that the discrete model of the interventional device has penetrated the blood vessel wall and needs to be corrected. Indicates the surface of the blood vessel wall; Let i represent the i-th particle.

[0065] Bending constraint refers to ensuring that the bending angle of the discrete model of the interventional device is less than or equal to a set angle threshold, preventing excessive bending of the interventional device that could lead to breakage or damage to blood vessels. In this embodiment, the bending angle of the discrete model of the interventional device is defined as: ,in, These represent the (i-1), ith, and (i+1)th particles. This represents the angle formed by the (i-1)th, ith, and (i+1)th particles. It needs to be less than or equal to (That is, setting an angle threshold, the maximum angle at which the interventional device is allowed to bend). When They concluded that the bending angle of the discrete model of the interventional device was too large and needed to be corrected.

[0066] The continuity constraint of interventional devices refers to the continuity of the discrete model of the interventional device, that is, the distance between particles is constant, ensuring the integrity of the interventional device's shape. In the embodiments of this application, the continuity of the discrete model of the interventional device is defined as: Where C is a constraint function. When C is not equal to 0, the discrete model of the interventional device is considered to be discontinuous, with breaks or stacking, and needs to be corrected.

[0067] For each defined constraint, if it is determined that the discrete model of the interventional device needs to be modified based on that constraint, the modification amount under that constraint is calculated. Specifically, the modification amount can be calculated using the following formula:

[0068] in, This represents the position correction amount for the i-th particle; Represents the constraint function; , This represents the mass of the i-th particle; This represents the predicted position of the i-th particle; This represents the gradient of the constraint function at the i-th particle; Let represent the mass of the j-th particle.

[0069] Then, the discrete model of the interventional device is corrected based on the correction amount. Specifically, the discrete model of the interventional device can be corrected using the following formula:

[0070]

[0071] in, Indicates the corrected position of the particle; The particle's position at the previous moment is represented by v; the particle's velocity is represented by v. This represents the simulation step size, which is a fixed value. denoted by ; p represents the predicted position.

[0072] The embodiments of this application ensure stability by implicitly calculating the speed of position difference.

[0073] Specifically, for continuous constraints on interventional devices, the correction amount for continuous constraints on interventional devices can be:

[0074] Where C represents the continuous constraint function of the interventional device.

[0075] For bending constraints, the bending constraint function can be expressed as:

[0076]

[0077] in , , These represent the positions of three adjacent particles. This represents the angle formed by the (i-1)th, ith, and (i+1)th particles. This indicates that an angle threshold is set. When... At that time, it was considered that the discrete model of the interventional device was excessively bent and needed to be corrected. The bending constraint correction amount obtained based on the bending constraint function can be expressed as:

[0078] in, This indicates the amount of bending constraint correction.

[0079] For collision constraints, the collision constraint function can be expressed as:

[0080] Let i represent the distance field from the i-th particle to the blood vessel wall, when When the particle penetrates the blood vessel wall, the collision constraint correction can be:

[0081] in, This represents the external normal vector of the blood vessel wall corresponding to the particle's location. This indicates the depth to which the particle penetrates the blood vessel wall. This represents the collision constraint correction amount.

[0082] In this embodiment of the application, after correcting the discrete model of the interventional device based on the correction amount, the simulation results of the discrete model of the interventional device are obtained.

[0083] In this embodiment, to reduce computational load and improve computational efficiency, only the front-end region of the discrete model of the interventional device can be simulated locally. Specifically, the first K particles of the discrete model of the interventional device are selected as the front-end region for simulation. The front-end region can be denoted as... , where K is the window size for the local simulation.

[0084] After the simulation, the physical accessibility of the interventional device is determined based on the simulation results, and the physical accessibility determination result is obtained. Specifically, if the simulation result does not meet any of the above-mentioned constraints, the physical accessibility determination result is determined to be physically inaccessible; otherwise, the physical accessibility determination result is determined to be physically accessible.

[0085] In this embodiment of the application, if the simulation result does not meet any of the above-mentioned set constraints, that is, if a particle penetrates the blood vessel wall (i.e., and / or the bending angle exceeds the set angle threshold (i.e., ...) ) and / or, there is a discontinuity in the interventional device (i.e. If the simulation results meet all the above-mentioned constraints, the physical accessibility judgment result is determined to be physically unreachable; otherwise, if the simulation results meet all the above-mentioned constraints, the physical accessibility judgment result is determined to be physically reachable.

[0086] When the physical reachability judgment result indicates that the device is physically reachable, the discrete model of the interventional device is controlled to move from the current node to the next node after pruning other candidate nodes. This embodiment of the application reduces the invalid search space by pruning the remaining paths in advance.

[0087] Then, the next node reached is taken as the new current node, and the process is repeated to determine the next node to be reached from the current node using the preset path search algorithm. This process is repeated until the termination node is reached, thus obtaining the target path that the interventional device can reach.

[0088] If the physical reachability assessment result is "physically unreachable," the next node is pruned and no longer participates in the subsequent search. Then, a new next node is determined from the other neighboring nodes of the current node to continue the simulation and physical reachability assessment until a next node that is physically reachable is determined.

[0089] The following describes how to use a preset path search algorithm to determine the next node that the discrete model of the interventional device can reach from the current node on the cerebral vascular network. Specifically, the total value of the discrete model of the interventional device reaching each candidate node on the cerebral vascular network is calculated. The candidate nodes are all the adjacent nodes that the discrete model of the interventional device can reach from the current node on the cerebral vascular network. The next node is determined based on the total value of the discrete model of the interventional device reaching each candidate node on the cerebral vascular network.

[0090] In this embodiment, besides the termination node, each current node may have at least one next adjacent node (denoted as a candidate node) that it can reach. When there are multiple candidate nodes, the total cost of the interventional device discrete model advancing to each candidate node on the cerebral vascular network can be calculated separately; the next node is determined based on the total cost of the interventional device discrete model advancing to each candidate node on the cerebral vascular network.

[0091] Specifically, the total agency value mentioned above is an indicator used in the preset path search algorithm to measure the quality of a path. It can include the actual cost of paths already traversed, or the estimated cost of untraversed paths. In this embodiment, the total agency value is calculated using the following formula:

[0092] in, This indicates the value of the general agent; the smaller the value, the better the distribution path. This represents the true cost of the discrete model of interventional devices from the starting node to the current node; This represents the estimated cost of the discrete model of interventional devices from the current node to the termination node.

[0093] Based on the introduction of the aforementioned constraints, the calculation of the total agency value can also incorporate related terms of the constraints, specifically:

[0094]

[0095] in, , , , All of these represent cost weighting coefficients, which are fixed values; Indicates the position of the termination node; H indicates the degree of bending exceeding the limit; H indicates the length of the path already traveled.

[0096] After calculating the total generation value of the discrete model of the interventional device on the cerebral vascular network to each candidate node, the next node is determined based on each total generation value. For example, the candidate node corresponding to the minimum total generation value is taken as the next node.

[0097] Based on this, when the physical reachability judgment result is that the physical reachability is not physical, a new next node is determined from the other adjacent nodes of the current node to continue the simulation and physical reachability judgment. For example, the candidate node corresponding to the second smallest total value can be taken as the next node.

[0098] Figure 2 An exemplary structural block diagram of an interventional device path planning apparatus 200 for cerebrovascular intervention, according to some embodiments of this application, is shown.

[0099] like Figure 2 As shown, the interventional device path planning device 200 applied to cerebrovascular intervention includes: a data acquisition module 210, used to acquire the cerebrovascular network of the target patient and the start and end nodes of the interventional device intervention; a discretization module 220, used to discretize the interventional device to be intervened, to obtain a discrete model of the interventional device; and a path planning module 230, used to place the discrete model of the interventional device on the cerebrovascular network after the start node, and use a preset path search algorithm to plan the path of the interventional device on the cerebrovascular network until the end node is reached, to obtain the target path that the interventional device can reach; wherein, during the path planning process, a preset simulation algorithm is used to simulate the deformation of the discrete model of the interventional device, and the physical reachability of the interventional device is judged based on the simulation results.

[0100] As an optional embodiment of this application, the data acquisition module 210 is specifically used for: acquiring three-dimensional vascular images of the target patient's brain; performing vascular mask extraction on the three-dimensional vascular images of the brain to obtain a vascular mask; performing vascular centerline extraction on the vascular mask to obtain the vascular centerlines of each blood vessel on the vascular mask; and constructing a brain vascular network based on the vascular centerlines.

[0101] As an optional embodiment of this application, the path planning module 230 is specifically used for: when determining the next node to which the discrete model of the interventional device will move from the current node on the cerebral vascular network using a preset path search algorithm, obtaining the predicted position of each particle when the discrete model of the interventional device moves to the next node; the next node is one of the adjacent nodes of the current node; when the predicted position does not meet the set constraint conditions, calculating the correction amount of the discrete model of the interventional device based on the predicted position, the set constraint conditions, and the position of each particle on the discrete model of the interventional device; wherein, the set constraint conditions include at least one of the following: the discrete model of the interventional device cannot penetrate the blood vessel wall, the bending angle of the discrete model of the interventional device is less than or equal to the set angle threshold, and the discrete model of the interventional device is continuous; correcting the discrete model of the interventional device based on the correction amount to obtain the simulation results.

[0102] As an optional embodiment of this application, the path planning module 230 described above determines the next node to which the discrete interventional device model can move from the current node on the cerebral vascular map network using a preset path search algorithm, including: calculating the total cost of the discrete interventional device model moving to each candidate node on the cerebral vascular map network; wherein, the candidate nodes are all the adjacent nodes that the discrete interventional device model can reach from the current node on the cerebral vascular map network; and determining the next node based on the total cost of the discrete interventional device model moving to each candidate node on the cerebral vascular map network.

[0103] As an optional embodiment of this application, the physical accessibility judgment of the interventional device based on the simulation results of the discrete model of the interventional device in the path planning module 230 includes: when the simulation result does not meet any of the set constraints, determining that the physical accessibility judgment result is physically unreachable; otherwise, determining that the physical accessibility judgment result is physically reachable.

[0104] As an optional embodiment of this application, the interventional device path planning device 200 applied to cerebrovascular intervention further includes: a first loop module, used to prune other candidate nodes and control the interventional device discrete model to move from the current node to the next node when the physical reachability judgment result is physically reachable, and take the next node as the new current node, and return to execute the step of determining the next node to which the interventional device discrete model moves from the current node on the cerebrovascular network using a preset path search algorithm; a second loop module, used to prune the next node when the physical reachability judgment result is physically inaccessible, and determine the new next node based on the total cost of the interventional device discrete model moving to each candidate node on the cerebrovascular network at the current node, until a physical reachability judgment result indicates that the next node is physically reachable.

[0105] As an optional embodiment of this application, the preset path search algorithm is A. One of the following algorithms: algorithm, Dijkstra algorithm, RRT algorithm; and / or, the preset simulation algorithm is one of the following: PBD simulation algorithm, XPBD simulation algorithm, mass-spring simulation algorithm.

[0106] For specific implementation methods and beneficial effects, please refer to the above embodiment of the interventional device path planning method 100 applied to cerebrovascular intervention, which will not be repeated here.

[0107] Correspondingly, embodiments of this application also provide Figure 2 The hardware structure diagram of the device shown is as follows: Figure 3 As shown, the electronic device 300 can be a device for implementing the above-described interventional device path planning method 100 applied to cerebrovascular intervention. For example... Figure 3 As shown, the electronic device 300 includes a processor 310 and a memory 320. The memory 320 is configured to store program instructions; the processor 310 is configured to load and execute the program instructions stored in the memory 320 to implement the embodiment of the interventional device path planning method 100 applied to cerebrovascular intervention as shown above.

[0108] As one embodiment, memory 320 can be any electronic, magnetic, optical, or other physical storage device that can contain or store information such as program instructions, data, etc. For example, memory 320 can be volatile memory, non-volatile memory, or similar storage media. Specifically, memory 320 can be RAM (Random Access Memory), flash memory, storage drives (such as hard disk drives), solid-state drives, any type of storage disk (such as optical discs, DVDs, etc.), or similar storage media, or combinations thereof.

[0109] This concludes the process. Figure 3 Description of the electronic device shown.

[0110] While numerous embodiments of this application have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will arise for those skilled in the art without departing from the spirit and intent of this application. It should be understood that various alternatives to the embodiments of this application described herein may be employed in the practice of this application. The appended claims are intended to define the scope of protection of this application and therefore cover equivalents or alternatives within the scope of these claims.

Claims

1. A method for interventional device path planning applied to cerebrovascular intervention, characterized in that, include: Obtain the cerebral vascular network of the target patient, as well as the start and end nodes of the interventional device intervention; wherein, the cerebral vascular network consists of nodes and edges between nodes; The interventional device to be intervened is discretized to obtain a discrete model of the interventional device. After placing the discrete model of the interventional device on the cerebral vascular network at the starting node, a preset path search algorithm is used to plan the path of the interventional device on the cerebral vascular network until the termination node is reached, thus obtaining the target path that the interventional device can reach. In the path planning process, a preset simulation algorithm is used to simulate the deformation of the discrete model of the interventional device, and the physical reachability of the interventional device is judged based on the simulation results. The step of simulating the deformation of the discrete model of the interventional device using a preset simulation algorithm during the path planning process includes: When the preset path search algorithm is used to determine the next node that the discrete model of the interventional device will move to from the current node on the cerebral vascular network, the predicted position of each particle when the discrete model of the interventional device moves to the next node is obtained; the next node is one of the adjacent nodes of the current node; When the predicted position does not meet the set constraints, the correction amount of the discrete model of the interventional device is calculated based on the predicted position, the set constraints, and the positions of each particle on the discrete model of the interventional device; wherein, the set constraints include at least one of the following: the discrete model of the interventional device cannot penetrate the blood vessel wall, the bending angle of the discrete model of the interventional device is less than or equal to the set angle threshold, and the discrete model of the interventional device is continuous. The discrete model of the interventional device is corrected based on the correction amount to obtain simulation results; The step of using the preset path search algorithm to determine the next node for the discrete model of the interventional device to move from the current node on the cerebral vascular network includes: Calculate the total generation value of the discrete model of the interventional device as it progresses to each candidate node on the cerebral vascular network; wherein, the candidate node is all the adjacent nodes that the discrete model of the interventional device can reach from the current node on the cerebral vascular network. The next node is determined based on the total scalar value of each candidate node advancing to the cerebral vascular network using the discrete model of the interventional device. The step of determining the physical accessibility of the interventional device based on the simulation results of the discrete model of the interventional device includes: If the simulation result does not meet any of the set constraints, the physical reachability judgment result is determined to be physically unreachable; Conversely, if the physical accessibility determination result is determined to be physically accessible; The method further includes: When the physical reachability judgment result is physical reachable, after pruning other candidate nodes, the interventional device discrete model is controlled to move from the current node to the next node, and the next node is taken as the new current node. Then, the process returns to execute the step of using the preset path search algorithm to determine the next node that the interventional device discrete model moves from the current node to on the cerebral vascular network. When the physical reachability judgment result is physically unreachable, after pruning the next node, a new next node is determined based on the total value of each candidate node in the cerebral vascular network at the current node, according to the discrete model of the interventional device, until a next node that is physically reachable is determined according to the physical reachability judgment result.

2. The method according to claim 1, characterized in that, The method for acquiring the cerebral vascular network of the target patient includes: Obtain three-dimensional vascular images of the brain of the target patient; A vascular mask extraction operation is performed on the three-dimensional vascular image of the brain to obtain a vascular mask; Perform a blood vessel centerline extraction operation on the blood vessel mask to obtain the blood vessel centerlines of each blood vessel on the blood vessel mask; The brain vascular network is constructed based on the central line of the blood vessels.

3. The method according to claim 1, characterized in that, The preset path search algorithm is A. The algorithm is one of the following: Dijkstra's algorithm, RRT algorithm; and / or the preset simulation algorithm is one of the following: PBD simulation algorithm, XPBD simulation algorithm, mass-spring simulation algorithm.

4. An interventional device path planning device for cerebrovascular intervention, characterized in that, include: The data acquisition module is used to acquire the cerebral vascular network of the target patient, as well as the start and end nodes of the interventional device intervention. The discrete module is used to discretize the interventional device to be intervened, and obtain a discrete model of the interventional device. The path planning module is used to place the discrete model of the interventional device on the cerebral vascular network after the starting node, and use a preset path search algorithm to plan the path of the interventional device on the cerebral vascular network until the termination node is reached, so as to obtain the target path that the interventional device can reach; wherein, during the path planning process, a preset simulation algorithm is used to simulate the deformation of the discrete model of the interventional device, and the physical reachability of the interventional device is judged based on the simulation results. Specifically, the path planning module is used for: When the preset path search algorithm is used to determine the next node that the discrete model of the interventional device will move to from the current node on the cerebral vascular network, the predicted position of each particle when the discrete model of the interventional device moves to the next node is obtained; the next node is one of the adjacent nodes of the current node; When the predicted position does not meet the set constraints, the correction amount of the discrete model of the interventional device is calculated based on the predicted position, the set constraints, and the positions of each particle on the discrete model of the interventional device; wherein, the set constraints include at least one of the following: the discrete model of the interventional device cannot penetrate the blood vessel wall, the bending angle of the discrete model of the interventional device is less than or equal to the set angle threshold, and the discrete model of the interventional device is continuous. The discrete model of the interventional device is corrected based on the correction amount to obtain simulation results; The step of using the preset path search algorithm to determine the next node for the discrete model of the interventional device to move from the current node on the cerebral vascular network includes: Calculate the total generation value of the discrete model of the interventional device as it progresses to each candidate node on the cerebral vascular network; wherein, the candidate node is all the adjacent nodes that the discrete model of the interventional device can reach from the current node on the cerebral vascular network. The next node is determined based on the total scalar value of each candidate node advancing to the cerebral vascular network using the discrete model of the interventional device. The step of determining the physical accessibility of the interventional device based on the simulation results of the discrete model of the interventional device includes: If the simulation result does not meet any of the set constraints, the physical reachability judgment result is determined to be physically unreachable; Conversely, if the physical accessibility determination result is determined to be physically accessible; The device further includes: The first loop module is used to prune other candidate nodes when the physical reachability judgment result is physically reachable, and then control the discrete model of the interventional device to move from the current node to the next node, and take the next node as the new current node, and return to execute the step of using the preset path search algorithm to determine the next node to which the discrete model of the interventional device moves from the current node on the cerebral vascular network. The second loop module is used to prune the next node when the physical reachability judgment result is physically unreachable, and then determine the new next node based on the total value of each candidate node in the cerebral vascular network based on the discrete model of the interventional device, until the next node that is physically reachable is determined by the physical reachability judgment result.

5. An electronic device, characterized in that, include: A processor, configured to execute program instructions; as well as A memory configured to store the program instructions, which, when loaded and executed by the processor, cause the processor to perform the interventional device path planning method for cerebrovascular intervention as described in any one of claims 1-3.

6. A computer-readable storage medium storing program instructions, characterized in that, When the program instructions are loaded and executed by the processor, the processor performs the interventional device path planning method for cerebrovascular intervention as described in any one of claims 1-3.

Images