Ablation applicator mode matched with grid template

Abstract

A pattern matching method (142) of an applicator to a mesh for calculating a mesh-matched insertion pose for an applicator insertion pattern used in mesh-based ablation, the method comprising, at a computing device (14): receiving mesh and applicator information (144); a computational iteration is performed on an applicator pattern along a plurality of apertures in a grid, where the computational iteration includes advancing an applicator pattern object through an entire grid object, where the applicator pattern is defined by a plurality of vertices corresponding to a plurality of applicator entry trajectories and an ablation trajectory (146), where the plurality of vertices correspond to the plurality of applicator entry trajectories, and where the plurality of vertices correspond to the plurality of applicator entry trajectories. Each vertex corresponds to an applicator entry point of a corresponding applicator entry trajectory, where the ablation trajectory is parallel to and between the plurality of applicator entry trajectories; and storing pattern position information for corresponding applicator patterns that match the apertures in the grid during the iteration, where the pattern position information includes an entry trajectory that includes one or more applicator pattern vertex positions and a grid plane normal vector (148).

Description

Technical Field

[0001] This invention relates generally to treatment planning, and more particularly to combined ablation treatment planning. Background Technology

[0002] In interventional oncology, percutaneous ablation (including both thermal and non-thermal ablation techniques) is an interventional cancer treatment option. Thermal ablation can be delivered using various ablation modalities, such as radiofrequency ablation (RFA), microwave ablation (MWA), high-intensity focused ultrasound (HIFU), local laser ablation (FLA), and cryoablation. Non-thermal ablation techniques include irreversible electroporation (IRE), which uses non-thermal energy to create permanent nanopores in the cell membrane, thereby disrupting cellular homeostasis and ultimately causing cell damage within an applied electric field.

[0003] In clinical practice, these ablation procedures involve placing one or more ablation agents within or near the target area with image guidance. Typically, physicians place these needle-like agents while carefully reviewing real-time ultrasound or interventional radiology images (CT / MR / CBCT), determining the expected ablation location based on manufacturer-provided information, clinical trial results, and personal experience. Agent placement is often determined based on the tumor's location and size, the device manufacturer's specifications, and the physician's experience. To minimize unnecessary damage to nearby healthy tissue, the ablation zone should be positioned as close as possible to the tumor's contour.

[0004] Currently available ablation planning only supports regular ellipsoidal ablation zones centered on the axis of the needle applicator. For some commercial devices, manufacturer data also specifies simpler (e.g., ellipsoidal) composite ablation zones, including ablation achieved by simultaneously ablating multiple applicators positioned in a fixed pattern relative to each other. Planning such composite ablations is a complex process, typically requiring applicators to be positioned in a fixed pattern in three-dimensional space while still achieving complete tumor coverage and avoiding surrounding critical structures. Therefore, clinical users (and automated methods) face a challenging planning problem with extremely high degrees of freedom and a very large solution space. Subsequently, clinical users face at least one challenge: how to accurately place the applicator to achieve the planned ablation. An additional challenge arises when using a guided mesh template for ablation delivery, as the applicator pattern needs to match the mesh apertures for proper operation. For clinical users, manually determining possible applicator placements is a lengthy and arduous process, especially when dealing with the complexity of multi-vertex insertion patterns. Summary of the Invention

[0005] The present invention provides a pattern matching method for the applicator and the mesh according to claim 1.

[0006] One objective of this invention is to improve existing systems for planning composite ablation zones in ablation therapy, thereby improving the positioning of the applicator within a mesh. To better address these issues, in a first aspect of the invention, a method is provided for calculating a mesh-matched insertion pose of an applicator insertion pattern used in mesh-based ablation. The method includes, at a computing device: receiving mesh and applicator information; performing computational iterations of the applicator pattern along multiple holes in the mesh, wherein the computational iterations include advancing an applicator pattern object through the entire mesh object. The applicator pattern is defined by multiple vertices corresponding to multiple applicator entry trajectories and ablation trajectories. Each vertex corresponds to the applicator entry point of the corresponding applicator entry trajectory. The ablation trajectories are parallel to and between the multiple applicator entry trajectories. In particular, the ablation trajectories are parallel to the direction of the corresponding applicator. The method further includes storing pattern position information of the corresponding applicator pattern that matches the holes in the mesh during the iterations. The pattern position information includes the entry trajectory containing the applicator pattern vertex positions and mesh plane normals.

[0007] In other words, at the pattern location that matches the mesh, the corresponding entry trajectory is stored. For example, this can be stored in a list of entry trajectories. Such entry trajectories for a single matching pattern location include, or are defined by, the positions of multiple practitioner pattern vertices and the mesh plane normal vector. For example, if multiple pattern locations that match the mesh are identified during the iteration, these pattern locations are stored as pattern location information, for example, in a list of entry trajectories, where each entry in the list represents a pattern location that matches the mesh. In other words, such a list of entry trajectories can be considered as a set of entry trajectories.

[0008] By using the vertex position and the mesh plane normal vector as components in defining the entry trajectory, the entry trajectory can be defined based on the trajectory origin (vertex position) and the trajectory direction (mesh plane normal vector), respectively. This allows for the calculation of the mesh pattern insertion pose, where "insertion" and "entry" are interchangeable terms. In other words, the pattern position information includes information about the entry trajectory, which includes the vertex position (representing the trajectory origin) and the mesh plane normal vector (representing the trajectory direction).

[0009] By iterating over the entire grid, multiple grid-matching insertion poses (if any) can be determined for the delegated agent insertion pattern in the ablation zone selected for tumor treatment.

[0010] In one embodiment, the computational iteration includes iteratively advancing the treater pattern by advancing it one hole in a first direction and then one hole in an orthogonal second direction until all possible treater patterns have been tried in both the first and second directions. In this search method, it is assumed that the relative coordinates of the treaters present in a pattern are adapted to the grid spacing along the row and column directions.

[0011] In one embodiment, the computational iteration includes, for each hole in the mesh: successively anchoring each vertex of the vertices of the treater pattern to each hole in the mesh; for each treater pattern position relative to the anchored vertex, determining whether all other vertices of the treater pattern match the corresponding hole in the mesh, wherein storing the pattern position information includes: storing the pattern position information for each of the treater pattern positions corresponding to the anchored vertex: at each of the treater pattern positions corresponding to the anchored vertex, the other vertices match the corresponding hole in the mesh; and rejecting the treater pattern position for the given anchor position if at least one vertex of the vertices does not match a mesh hole at the given anchor position. In this search method, no other assumptions are strictly imposed on the geometry of the mesh and the treater pattern compared to the first method.

[0012] In one embodiment, the method further includes rotating the treater pattern by a defined angular step, and repeating the anchoring, determining, storing, and rejecting processes sequentially. Matching via step rotation enables a more thorough mesh hole matching search than when the rotation is constrained to ninety degrees.

[0013] In one embodiment, the treater pattern comprises two vertices, and for a given hole that coincides with one of the two vertices, the computational iteration comprises: determining the distance between the two vertices; searching for all holes at a distance from the given hole, wherein storing the pattern position information comprises: storing pattern position information for all treater patterns located at the distance for holes where the two vertices match the mesh; and repeating the determination and the search for successive holes. This search method is a modified version of the aforementioned prior method, wherein a continuous (rather than angular step) treater pattern is implicitly performed.

[0014] In one embodiment, the treater pattern comprises two or more vertices, and for a given hole that coincides with one of the two vertices, the computational iteration comprises: determining the distance between the two vertices; searching for all holes that are at the distance from the given hole and defining each pair of vertices that match the hole along the search distance as a set; for each set, determining the remaining treater pattern vertices, wherein storing the pattern position information comprises: storing the pattern position information for each set if the remaining vertices of the treater pattern in each set also match the holes in the mesh; and repeating the determination, the search, and the determination for successive holes. This search method is similar to the previous method, applying implicitly performed continuous (rather than angular step) treater patterns, but is applicable to treater patterns with more than two vertices.

[0015] The present invention also provides a pattern matching system for an applicator and a mesh. Specifically, an applicator-mesh pattern matching module is provided, which is used to calculate the mesh-matching insertion posture of the applicator insertion pattern used in mesh-based ablation. This applicator-mesh pattern matching module includes a computing device configured to: Receive grid and applicator information; The applicator pattern is computed iteratively along multiple holes in the mesh, wherein the computational iteration includes advancing the applicator pattern object through the entire mesh object, wherein the applicator pattern is defined by multiple vertices corresponding to multiple applicator entry trajectories and ablation trajectories, wherein each vertex corresponds to the applicator entry point of the corresponding applicator entry trajectory, wherein the ablation trajectories are parallel to the multiple applicator entry trajectories and lie between the multiple applicator entry trajectories; and The pattern position information of the corresponding treater pattern that matches the hole in the mesh during the iteration is stored, wherein the pattern position information includes an entry trajectory containing the vertex position of the treater pattern and the mesh plane normal vector.

[0016] These and other aspects of the invention will become apparent and elucidated with reference to one or more embodiments described below. Attached Figure Description

[0017] Many aspects of the invention can be better understood by referring to the following schematic drawings. The components in the drawings are not necessarily drawn to scale, but rather the emphasis is on clearly illustrating the principles of the invention. Furthermore, in the drawings, the same reference numerals refer to corresponding parts throughout several views.

[0018] Figures 1A to 1CThis is an example representation of a composite ablation zone according to certain embodiments of the pattern matching method of the applicator and the mesh according to embodiments of the present invention.

[0019] Figure 2 This is a schematic diagram of an embodiment of an example ablation therapy system according to an embodiment of the present invention, which facilitates planning (including applicator placement planning using a pattern matching method of applicator and grid) and guidance for the treatment of tumor masses or lesions.

[0020] Figure 3 The illustration is a schematic diagram of an administration pattern comprising a set or group of three administration devices arranged in a triangular pattern according to an embodiment of the present invention, wherein the centroid of the triangular pattern serves as the origin of the composite ablation entry trajectory.

[0021] Figures 4A to 4C This is a schematic diagram illustrating example composite ablation zones for dual-applicator linear mode used alone and dual-applicator linear mode used in combination with additional applicator sets.

[0022] Figure 5 (a) and Figure 5 (b) is a schematic diagram illustrating regular and irregular rectangular grids that can be used by a pattern matching method of the applicator and the grid according to an embodiment of the present invention.

[0023] Figure 6 (a) to Figure 6 (d) is a schematic diagram illustrating example applicator patterns that match and do not match the holes in the grid.

[0024] Figure 7 (a) to Figure 7 (f) is a schematic diagram of an embodiment of a pattern matching method for a first applicator and a grid according to an embodiment of the present invention, wherein the pattern matching method for a first applicator and a grid advances the applicator hole by hole according to the x-direction and / or y-direction displacement for a dual applicator linear pattern in a grid usage scenario.

[0025] Figure 8 (a) to Figure 8 (d) is a schematic diagram of an embodiment of a pattern matching method for a second applicator and a mesh according to an embodiment of the present invention. This example second applicator and mesh pattern matching method causes the applicator to advance along the mesh and, for each hole, anchors one vertex and iterates multiple applicator patterns around the anchored hole to determine the matching of unanchored vertices with mesh holes.

[0026] Figure 9 A to Figure 9B is a schematic diagram illustrating examples of unmatched and matched applicator patterns for an example second applicator and grid pattern matching method according to an embodiment of the present invention.

[0027] Figure 10 (a) to Figure 10 (d) is a schematic diagram illustrating, according to an embodiment of the present invention, the use of rotation on the currently processed mesh hole to process multiple pattern positions and orientations on the mesh plane for the pattern matching method of the example second applicator and the mesh.

[0028] Figure 11 (a) to Figure 11 (d) is an illustrative example of a pattern matching method between a second applicator and a grid, and a schematic diagram illustrating the ignoring or omitting of redundant placement patterns anchored on different grid holes, according to an embodiment of the present invention.

[0029] Figure 12 This is a flowchart illustrating an embodiment of a pattern matching method for a second applicator and a grid according to an embodiment of the present invention.

[0030] Figure 13 (a) to Figure 13 (f) is a schematic diagram of an embodiment of a pattern matching method for a third applicator and a grid according to an embodiment of the present invention, which modifies the pattern matching method for a second applicator and a grid by using a continuous applicator pattern matching process.

[0031] Figure 14 (a) to Figure 14 (c) is a schematic diagram illustrating unmixed composite ablation using different applicators and grid patterns according to an embodiment of the present invention.

[0032] Figure 15 This is a flowchart illustrating an embodiment of a pattern matching method for a treatment device and a mesh according to an embodiment of the present invention. Detailed Implementation

[0033] A cancer diagnosis can be terrifying for patients. Beyond the anxiety of whether they can overcome cancer, patients may worry about the treatment options and side effects. Historically, surgery has been an open procedure involving large incisions, allowing doctors to directly observe the tumor while attempting to remove it. Thanks to technological advancements, a new procedure has emerged that allows access to cancerous tissue through smaller incisions—a procedure known as minimally invasive surgery. This means patients may experience less pain and go home sooner, and may also mean lower risks and shorter recovery times.

[0034] Because these procedures are less invasive than open surgery, the doctor will not be able to directly observe what's inside the patient's body. He or she can use image guidance (relying on, for example, magnetic resonance imaging (MRI), ultrasound, or other medical imaging) to determine the location of the tumor within the patient's body. For instance, in an ablation procedure, the doctor might insert a needle with a microlaser at its tip into the body, use image guidance to navigate the laser to the location of the tumor, and then use the laser to burn the cancerous tissue. Sometimes multiple needles may be used simultaneously to deliver the treatment, and in some cases, the placement of these needles may be guided by a grid. Given the complexity of the planning, software can be used to create applicator placement plans to attempt to burn cancerous tissue while protecting healthy tissue in ways that are impossible for humans.

[0035] This document discloses certain embodiments of methods and associated systems or devices for pattern matching of (ablation) applicators to meshes (hereinafter referred to as applicator-mesh pattern matching methods), which are used to find multiple (e.g., all) possible fixed applicator pattern locations that match the mesh template apertures. For example, when delivering ablation treatment using a guided mesh template (e.g., a 13×13 brachytherapy mesh), the planning algorithm should be able to explore many (if not all) possible applicator pattern locations that match the mesh geometry (e.g., identifying a set of placement patterns where all applicator locations are (precisely) fitted to the mesh apertures). Certain embodiments of applicator-mesh pattern matching methods typically include three methods for efficiently determining a set of composite ablation applicator pattern locations that match the mesh geometry and can be used as inputs for treatment planning selection (e.g., optimization) and its subsequent execution.

[0036] Incidentally, existing applicator pattern placement schemes for collaborative grids rely on manual placement (e.g., determined by clinicians). Finding possible applicator pattern placement locations along the grid is a lengthy and tedious process, particularly complex for applicator patterns with more than two applicators. For example, not all applicator pattern locations match the corresponding holes in the grid, and in some cases, the positioning and orientation of the applicator pattern may result in the applicator entry point being outside the grid. In contrast, certain embodiments of applicator-grid pattern matching methods can automatically determine applicator patterns that match grid holes, thereby shortening the applicator pattern placement planning phase and reducing some of the burden on clinicians.

[0037] Having summarized certain features of the applicator-mesh pattern matching method of this disclosure, a detailed description of the applicator-mesh pattern matching method will now be provided with reference to the accompanying drawings. While the applicator-mesh pattern matching method will be described in conjunction with these drawings, it is not intended to limit the matching method to the one or more embodiments disclosed herein. Furthermore, although specific details of one or more embodiments are specified or described in the description, these specific details are not necessarily integral to each embodiment, and the various advantages stated are not necessarily all related to a single embodiment. Rather, this document is intended to cover alternatives, modifications, and equivalents within the principles and scope of this disclosure as defined by the appended claims. For example, two or more embodiments may be interchanged or combined in any combination. Moreover, it should be understood in the context of this disclosure that the claims are not necessarily limited to the specific embodiments listed in the description.

[0038] While this disclosure focuses on the automatic placement of a set of applicators (e.g., including one or more applicators corresponding to an ablation trajectory) within a grid at all possible locations (including orientations), the context of ablation resulting from placement and selection according to an ablation treatment plan is briefly described below. Existing ablation planning methods are typically based on manufacturer data regarding the expected ellipsoidal ablation zone achieved using various combinations of power and duration with the ablation device. In the case of composite ablation, manufacturer data regarding the expected ellipsoidal composite ablation zone uses combinations of applicator spacing, power, and duration. Such composite ablation zones are strictly based on a fixed pattern of applicators, where each applicator is treated as an independent entity. Existing publications also use a simple binary union of ellipsoids to describe or represent composite ablation zones, but such techniques are not suitable for microwave (MW) devices and irreversible electroporation (IRE) devices that deliver complex ablation zones with non-ellipsoidal shapes. In the market, various probe models (the terms "probe" and "application device" are used interchangeably in this document) come with datasheets indicating the expected geometry of each single ablation zone delivered by a given probe configuration when used individually, and the expected geometry of the expected composite ablation zone delivered by probe configurations with different geometries. In some devices, the expected composite ablation zone has an ellipsoidal shape; however, in other devices (e.g., devices used for MWA or IRE treatments), the composite ablation zone may have a rather complex shape, resulting from the mixing of multiple ellipsoids delivered by each probe, as shown below. Figures 1A to 1C Examples of complex composite ablation zones 10A to 10C are shown in the figure.

[0039] In the following description, the ablation resulting from the execution of planned applicator placement (which involves pattern matching methods for applicators and meshes, and selection (e.g., optimization) methods) may take the form of: a regular ablation zone shape centered along the applicator axis (e.g., ellipsoid, cylinder, etc.), and / or based on composite ablation formed by multiple applicators positioned in a fixed pattern relative to each other or configured by the user. Furthermore, for composite ablation resulting from the aforementioned planning, the calculation of the ablation zone may be based on in vivo or in vitro experimental results, a binary union of ellipsoids, or a mixture of implicit functions to represent the composite ablation zone.

[0040] It is worth noting that the reference to composite ablation zones in this document refers to the expected total ablation zone shape delivered by the set or group of executed applicators. While this disclosure focuses on the planning phase, specifically grid-based applicator positioning to ultimately achieve composite ablation, applicator-grid pattern matching methods can also be used to plan (and ultimately execute) non-composite ablation zones.

[0041] It is worth noting that, in describing certain embodiments of the treater-mesh pattern matching method, references to the positioning or placement of the treater pattern along the mesh refer to the positioning (including orientation, such as via step rotation or continuous rotation) of the computational objects of the mesh and the treater pattern (e.g., in object-oriented programming or logical space). Furthermore, references to the treater pattern (object) are intended to include single or multiple treaters arranged to deliver ablation trajectories, the treater arrangement being defined by treater trajector trajector vertices or vertices relative to the ablation trajectory and aligned or matched with holes along the mesh.

[0042] Figure 2The diagram illustrates an ablation therapy system 12 that facilitates the planning and guidance of one or more ablation protocols to direct the treatment of target areas (e.g., tumor masses or lesions, including any remaining tissue) within a subject, and also includes a treater pattern localization function using a pattern-matching method between the treater and a grid. It is noteworthy that the ablation therapy system 12 described herein is responsible for treatment based on planning and guidance functions; however, in some embodiments, planning, guidance, and / or treatment may be implemented as independent systems communicating or cooperating with each other. Successful treatment of large tumors is achieved by planning the ablation probe placement to ensure that no part of the tumor is left untreated and by accurately executing the plan. Generally, the ablation therapy system 12 provides automated and interactive planning of ablation therapy, guidance of treater placement, and planning-guided ablation therapy treatment. The ablation plan ensures coverage of most (or possibly all) of the tumor area and reports the number of ablations involved in ablation treatment using a specific probe or group of probes(s). The ablation therapy system 12 also utilizes selection techniques to minimize the number of ablations. Because the procedure is quantitative, it can be performed using a robot and / or guided by registered images (e.g., by quantitatively tracking the ablation probe).

[0043] In the depicted embodiments, the functionality of the ablation therapy system 12 is implemented as a system embodied by software and hardware components located in the same location, embodied as a computing device 14 (which may include medical devices) and multiple subsystems, wherein the computing device 14 is operatively connected to the multiple subsystems as described below. In some embodiments, the computing device functionality may be embedded in one of the subsystems, or one or more of the subsystem functionalities described herein may be integrated into a smaller number of devices. It should be understood that in some embodiments, the functionality of the ablation therapy system 12 may be implemented via multiple computing devices that may be networked in similar locations or located in different locations (possibly geographically distant) and connected via one or more networks. The computing device 14 includes one or more processors 16 (e.g., 16A, ..., 16N), one or more input / output interfaces 18, and memory 20, etc., coupled to one or more data buses (e.g., data bus 22).

[0044] One or more processors 16 may be embodied as custom processors or commercial processors, including single-core or multi-core central processing units (CPUs), tensor processing units (TPUs), graphics processing units (GPUs), vector processing units (VPUs), or as an auxiliary processor among multiple processors, semiconductor-based microprocessors (in the form of microchips), macroprocessors, one or more application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), multiple appropriately configured digital logic gates, and / or other existing electrical configurations (including discrete elements that can coordinate the overall operation of computing device 14 individually or in various combinations).

[0045] I / O interface 18 includes hardware and / or software for providing one or more interfaces to various subsystems, including one or more user interfaces 24, imaging subsystem 26, and ablation subsystem 28. I / O interface 18 may also include additional functionality, including a communication interface for network-based communication. For example, I / O interface 18 may include wired modems and / or cellular modems, and / or establish communication with other devices or systems via Ethernet connections, hybrid / fiber-coaxial (HFC), copper cables (e.g., Digital Subscriber Line (DSL), Asymmetric DSL, etc.) using one or more communication protocols (e.g., TCP / IP, UDP, etc.). Generally, the I / O interface 18 that cooperates with the communication module (not shown) includes suitable hardware to enable information communication via the Public Switched Telephone Network (PSTN), Ordinary Old-Style Telephone Service (POTS), Integrated Services Digital Network (ISDN), Ethernet, fiber optic, DSL / ADSL, Wi-Fi, cellular networks (e.g., 3G, 4G, 5G, Global System for Mobile Communications (GSM), General Packet Radio Service (GPRS), etc.), Bluetooth, Near Field Communication (NFC), Zigbee, etc., using TCP / IP, UDP, HTTP, DSL protocols.

[0046] One or more user interfaces 24 may include a keyboard, scroll wheel, mouse, microphone, immersive head-mounted device, one or more display devices, etc., which enable a user to input and / or output, or to input and / or output to a user, and / or to provide visual information to a user. In some embodiments, one or more user interfaces 24 may cooperate with associated software to enable augmented reality or virtual reality. When a display device is included, one or more user interfaces 24 enable the display of segmented anatomical structures (including one or more tumors), virtual (e.g., simulated) and actual placement of the ablation device, identification of the ablation zone and its coverage, and virtual bounding boxes (e.g., mesh objects). In some embodiments, one or more user interfaces 24 may be directly coupled to a data bus 22.

[0047] Imaging subsystem 26 includes one or more imaging subsystems and / or image storage subsystems for enabling visualization of the tumor, virtual or physical applicator placement, and / or single or combined ablation zones during preoperative and intraoperative imaging (e.g., during actual applicator placement and / or ablation treatment). Imaging subsystem 26 may include ultrasound imaging (e.g., real-time 2D / 3D), fluoroscopy (real-time), magnetic resonance imaging (e.g., MR, real-time or static 2D / 3D), computed tomography (e.g., CT (3D static)), cone-beam computed tomography (e.g., CBCT (3D static)), single-photon emission computed tomography (SPECT), and / or positron emission tomography (PET). In some embodiments, images may be retrieved from a picture archiving and communication system (PACS) or any other suitable imaging component or delivery system.

[0048] The ablation subsystem 28 may include any of a variety of ablation modalities, including radiofrequency ablation (RFA), microwave ablation (MWA), high-intensity focused ultrasound (HIFU), local laser ablation (FLA), irreversible electroporation (IRE), and cryoablation. In some embodiments, software in memory 20 may be used to interact with the ablation subsystem 28 to adjust settings (e.g., probe distance, power, duration), track the positioning of the applicator, and / or control the actual placement and activation of the applicator in the case of robot-based ablation therapy.

[0049] Memory 20 may include any one or any combination of volatile memory elements (e.g., random access memory RAM, such as DRAM and SRAM) and non-volatile memory elements (e.g., ROM, flash memory, solid-state memory, EPROM, EEPROM, hard disk, magnetic tape, CDROM, etc.). Memory 20 may store a native operating system, one or more native applications, an emulation system, or emulation applications for various operating systems and / or emulation hardware platforms, any one of the emulation operating systems and / or emulation hardware platforms, or emulation applications for the emulation operating system. In some embodiments, a separate storage device (STOR DEV) may be coupled to data bus 22, or as one or more devices connected to a network via I / O interface 18 and one or more networks. The storage device may be embodied as persistent memory (e.g., optical, magnetic, and / or semiconductor memory and its associated drives). In some embodiments, the storage device or memory 20 may store manufacturer data associated with various agents and agent modes.

[0050] exist Figure 2 In the depicted embodiments, memory 20 includes an operating system 30 (OS) (e.g., LINUX, macOS, Windows, etc.) and ablation therapy software 32. In one embodiment, the ablation therapy software 32 includes multiple modules (e.g., executable code) co-hosted on computing device 14, but in some embodiments, these modules may be distributed across various systems or subsystems on one or more networks. The ablation therapy software 32 includes a graphical user interface (GUI) module 34, a segmentation module 36, and an ablation module 38. The ablation therapy software 32 also includes a debugging module 40 and a planning module 44. In one embodiment, the planning module 44 includes an agent-mesh pattern matching (AGPM) module 42 and a selection module 46. The debugging module 40 may include additional functionality (e.g., implicit functions or algorithms for mixing composite ablation zones, ellipsoidal union algorithms) or manufacturer data regarding the expected composite (and single) ablation. In some embodiments, such information may be stored in other locations (e.g., in the ablation module 38). The ablation therapy software 32 also includes a guidance module 48, which guides the placement of the applicator via graphics superimposed on an image, in cooperation with one or more user interfaces 24, imaging subsystem 26, ablation subsystem 28, and other modules. It is noteworthy that the functionality of the modules in the ablation therapy software 32 can be shared. In some embodiments, fewer or additional modules may be present. For example, the functionality of some modules may be combined, or additional functions may be implemented but not shown (e.g., communication modules, security modules, etc.).

[0051] This section briefly explains the function of one or more of the various modules, with further explanation below. The GUI module 34 provides the reconstruction and rendering of image data received from the imaging subsystem 26 on one or more user interfaces 24 (e.g., a display device). Objects (e.g., lesions, organs, vitals, and / or healthy anatomy (or generally, areas not intended for ablation)) can be automatically segmented using existing algorithms, or manually segmented via the segmentation module 36 using drawing tools along an axis. The segmentation module 36 generates descriptions of volumetric regions associated with these specific objects according to existing methods. The GUI module 34 may also provide visualization of the ablation zone, the applicator pattern, and the applicator pattern placement for application to one or more target areas (whether virtual placement (e.g., preoperative) or actual placement (during intraoperative implementation)).

[0052] The ablation module 38 works in conjunction with the GUI module 34 to provide options for displaying user-selectable applicator and / or applicator area. For example, the ablation module 38 may optionally (e.g., for additional or alternative schemes stored in the debugging module 40 or elsewhere) store manufacturer data regarding the intended ellipsoidal ablation area based on different settings of the ablation subsystem 28 (e.g., power, duration, probe spacing, etc.) and / or predefined applicator patterns (e.g., single probe, dual probe, triple probe, etc.) from manufacturer data. In some embodiments, the ablation module 38 may utilize electromagnetic sensors or other types of sensors on or near the applicator to track the positioning of the applicator in the ablation subsystem 28, and / or use image analysis for tracking. Based on these settings, the ablation module 38 controls the applicator during the ablation process, and in some embodiments, the ablation module 38 may work in conjunction with robot controls to control the placement and activation of the applicator.

[0053] The debugging module 40, which works in conjunction with the GUI module 34, enables the user to configure the applicator and / or ablation zone, and / or select predefined applicator patterns. The debugging module 40 may also include implicit function modules (not shown) that can be used to support non-ellipsoidal composite ablation zones formed by mixing ellipsoidal zones based on multiple applicators (one applicator pattern) from a set of applicators. The debugging module 40 can be used in a meshed or meshless environment. The debugging module 40 can provide fixed positioning (e.g., in one embodiment, fixed positioning as specified by manufacturer data obtained through the ablation module 38) or be configured by the user (e.g., a clinical user) via the GUI module 34.

[0054] Planning module 44 provides automated and / or interactive planning for single and combined ablation procedures. For example, based on segmented anatomical regions, planning module 44 utilizes commissioning data to pre-calculate multiple (e.g., all possible) probe or applicator configurations, allowing selection of a preferred approach or plan based on these configurations. For grid-based planning, planning module 44 uses applicator-grid pattern matching module 42, which includes algorithms for calculating iterative applicator pattern objects within a grid object to determine possible locations (including orientations) where the applicator pattern matches a hole in the grid. In one embodiment, applicator-grid pattern matching module 42 includes three different algorithms for searching and / or determining (e.g., all) possible pattern locations that match holes in the grid, which are provided to selection module 46 for selection and final treatment. In one embodiment, these algorithms include: a first matching algorithm, wherein matching is performed by shifting the treater pattern along all rows and columns of the grid; a second matching algorithm, which includes matching using discrete local point set matching; and a third matching algorithm, which includes matching using continuous local point set matching. Further descriptions of these algorithms / methods are set forth below.

[0055] The planning module 44 also includes a selection module 46 that analyzes information associated with the localization and ablation zone of (one or more) tumors and applicator patterns, and defines a set of oriented ablation sites. In one embodiment, the selection module 46 includes existing optimization features that identify the minimum (or fewer) number of ablations required to cover a tumor region (e.g., tumor or tumor plus margin). In some embodiments, the selection module 46 identifies oriented ablation sites that avoid healthy tissue (e.g., minimize or reduce the risk of collateral damage). In some embodiments, additional object volumes are segmented to represent critical areas of tissue or bone that should not be ablated, and the selection module 46 avoids or reduces the risk of ablation in these areas while attempting to generate the minimum number of ablations or minimize collateral damage. However, in some cases, the selection module 46 may generate unablated areas, in which case an alert will be issued to the user, and these areas can be displayed on (one or more) user interfaces 24.

[0056] The guidance module 48 provides interactive user guidance for placing the treatment device according to the planned location and selection results, as described below.

[0057] The following describes a general description of example operation of the ablation therapy software 32. After tumor and at-risk organ segmentation is completed via the segmentation module 36, a debugging module 40 is used to enable debugging under the guidance or instruction of a user (e.g., a clinical expert) via the GUI module 34. For example, during debugging, the clinical expert decides which probes and probe settings to use. Probe settings may include probe type, geometric entry pattern (e.g., single probe, linear arrangement of two probes, linear or triangular, patterned arrangement of three probes, square or rectangular, patterned arrangement of four probes, etc.), and the expected ablation zone (e.g., ellipsoid or complex hybrid shape). In one embodiment, the ablation therapy software 32 may provide a debugging GUI (e.g., via the GUI module 34) for interaction with a clinical expert to handle complex hybrid shapes, where hybrid parameters can be manually tested or, in some cases, automatically calculated if the user loads mesh or binary volume data representing the expected composite ablation zone according to manufacturer data.

[0058] Once commissioning is complete, the operation of the ablation therapy software 32 will continue to be performed by the planning module 44. In a mesh scenario, the applicator-mesh pattern matching module 42 can use one of the matching methods to iterate over a commissioned applicator pattern (object) across the entire mesh (object) and pass this information to the selection module 46. In one embodiment, the selection module 46 uses the information passed by the applicator-mesh pattern matching module 42, along with other commissioning data, to pre-calculate possible probe configurations and select the preferred scheme or plan for treating the target area. For example, the selection module 46 can determine which mixed (or in some embodiments, unmixed and / or single) ablation areas should be selected or excluded. As an illustrative example, if the selection module 46 includes optimization capabilities, its iterative solver can select a plan from multiple (in some embodiments, all) possible pre-calculated probe configurations (e.g., single probes and / or groups of probes delivering simple ellipsoidal and / or mixed or unmixed ablation areas, respectively), such as a single probe delivering ellipsoidal ablation at one location in the target area and a group of triangular probes delivering composite mixed ablation at another location in the target area.

[0059] In one embodiment, selection module 46 includes a set-coverage, greedy iterative optimizer for selecting a set of preferred probe groups (e.g., triangular patterns or other geometric patterns, including squares, rectangles, points, etc.) to completely cover the tumor (target area) while avoiding nearby organs of risk. It is noteworthy that in some embodiments, the selected preferred set may be an optimized or optimal set, or it may be a set that, while not optimal, is sufficient to meet specific criteria. For example, example criteria may include: the number of iterations or computation time in the selection process, expected treatment time, the number, size, and / or maturity of tumors, the assessed or predicted risk of malignancy, confidence in this assessment or prediction, the subject and / or the subject's family history, and other factors established by the medical community and / or clinical experts. One or more of these factors may be measured and / or weighted in some way relative to one or more of the other factors, which may result in outcomes that are not as optimal as the optimization function could achieve, but are preferred based on established criteria. In some embodiments, the preferred approach may include more subjectivity in considering the outcome compared to an optimization function, and / or in some embodiments, an optimization function may be used, but constrained by computational resources and / or computation time.

[0060] To further explain, particularly when using a grid template, the selection module 46 utilizes the possible pattern placements determined by the applicator-grid pattern matching module 42 to select (e.g., in some embodiments, selection can be achieved via an optimization algorithm) an applicator pattern (and thus a set of applicators) for treatment / treatment. After the planning phase, the operation continues, with the actual placement of the probe guided by the guidance module 48, followed by the actual execution of the ablation treatment.

[0061] It is worth noting that the memory 20 and the storage device may be referred to herein as non-transient computer-readable storage media, etc.

[0062] The execution of the ablation therapy software 32 can be carried out by one or more processors under the management and / or control of the operating system 30.

[0063] When certain embodiments of computing device 14 are implemented at least in part by software (including firmware), it should be noted that ablation therapy software 32 (and its corresponding components) can be stored on various non-transient computer-readable (storage) media for use with or connection to various computer-related systems or methods. As used herein, a computer-readable medium may include an electronic, magnetic, optical, or other physical device or apparatus that may include or store computer processes (e.g., executable code or instructions) for or connection to a computer-related system or method. The software may be embedded in various computer-readable media for use with or connection to an instruction execution system, apparatus, or device (e.g., a computer-based system, a processor-included system, or other system capable of obtaining and executing instructions from and from an instruction execution system, apparatus, or device).

[0064] When certain embodiments of computing device 14 are implemented at least in part by hardware, such functionality may be implemented using any or a combination of the following techniques, all of which are already present in the prior art: discrete logic circuit(s) having logic gates for implementing logic functions on data signals, application-specific integrated circuit (ASIC) having appropriately combined logic gates, programmable gate array(s) (one or more), field-programmable gate array (FPGA), TPU, GPU and / or other accelerators / coprocessors, etc.

[0065] Those skilled in the art should understand in the context of this disclosure that the example computing device 14 is merely an illustration of one embodiment, and that some embodiments of the computing device may include fewer or additional components, and / or be compatible with... Figure 2 The associated functionalities of the various components depicted herein may be combined, or in some embodiments further distributed in additional modules or computing devices. It should be understood that certain well-known components of the computer system are omitted herein to avoid obscuring the more relevant features of computing device 14.

[0066] Before delving into the details of certain embodiments of the applicator-mesh pattern matching module 42, let us briefly describe some requirements that need to be met when delivering the composite ablation zone as described in the supplier's datasheet. The definition or representation of the composite ablation zone (also known as commissioning) specifies the position and orientation of the applicator (or equivalent probe) relative to the ablation zone, as well as the dimensions of the ablation zone. This commissioning can be performed as a fixed (e.g., performed by the system manufacturer) or configurable (e.g., performed by the system's end user) operation.

[0067] When configured by the end user, the system is equipped with a graphical user interface (GUI) (e.g., via...). Figure 2The GUI module 34 allows specifying commissioning information, including the relative positioning of the applicator and the composite ablation zone (e.g., tip offset, spacing, etc.) and / or ablation zone dimensions (e.g., the major and minor axes of the ellipsoid). The relative positioning of the applicator and the ablation zone can be represented relative to the center of the ablation zone or the tip position of any associated applicator. Furthermore, the GUI enables the user to virtually position a selected single probe or a group of probes during the commissioning phase and provides real-time feedback on the resulting ablation zone (e.g., composite or other types).

[0068] Users can access the system through a GUI (e.g., via a GUI such as...) Figure 2 The GUI module 34 shown, in conjunction with one or more user interfaces 24, specifies the composite ablation zone. This interface provides a CAD-like drawing environment, enabling the relative positioning of the applicator and the composite ablation zone, or allowing selection from predefined placement patterns. For these predefined patterns, a limited set of parameters is sufficient to achieve the relative positioning of the applicator and the composite ablation zone. These predefined placement patterns reproduce information provided by the ablation device manufacturer and include, but are not limited to: applicator patterns including: a straight line with an applicator spacing of 's' (2 applicators), a triangle with a side length of 's' (3 applicators), a square with a side length of 's' (4 applicators), a trapezoid specified by length and angle (4 applicators), etc. When the user selects one of these patterns and provides the relevant parameter values ​​(one or more), the system calculates the relative positioning of the applicator with respect to the center of the ablation zone. For example, refer to... Figure 3 The diagram shows a set of three applicators arranged in a triangular pattern 50. The three vertices V1, V2, and V3 represent the respective applicator entry points (corresponding to entry trajectories), and the triangular pattern 50 includes a composite ablation zone entry trajectory 52 (or simply ablation trajectory) corresponding to the centroid of the triangle. The parameter s represents the length of each side of the triangle.

[0069] For example, for an equilateral triangle insertion pattern with side length s, the origin of the ablation entry trajectory can be placed at (x,y) = (0,0), and the origins of the three applicator entry trajector trajectories (i.e., triangle vertices V1, V2, and V3) can be obtained as follows: Formula 1 Formula 2 Formula 3 As described above, the ablation therapy system 12 ( Figure 2 It can support not only ellipsoidal composite ablation zones, but also non-ellipsoidal composite ablation zones (such as ablation zones implemented via IRE), by providing a debugging GUI (e.g., a GUI drawn on a display device, such as...). Figure 2As shown, for example, in cooperation with the debugging module 40 via the GUI module 34 and the user interface(s) 24, the interface is capable of describing the composite ablation zone as a combination of multiple ellipsoids and includes additional blending parameters (e.g., via the implicit function module in the debugging module 40 ( Figure 2 ). Blending the individual ellipsoids into a composite ablation zone can be based on the mathematical concept of an implicit function, which is provided by an algorithm in the implicit function module of the debugging module 40. An implicit function F(P) representing a surface specifies that if F(P) = 0, then the position P lies on the surface. If the position is inside or outside the surface, then F(P) < 0 or F(P) > 0, respectively. For an ellipsoidal surface centered at the centroid Pc = (xc, yc, zc), its implicit function is defined as: Equation 4 To compute the composite ablation, the implicit functions of Ns individual ablation zones can be combined via another implicit function: Equation 5 Here, the blending function can be defined in various ways, including but not limited to: Equation 6 Equation 7 where b, a, R, and n are blending parameters, and these two blending functions are used to enforce a smooth transition between two basic surfaces (e.g., two ellipsoids). The exponential blending function of Equation 6 includes the parameter b, where b > 0 is a parameter for controlling the steepness of the blending curve (e.g., when b = 1, it includes two separate ellipses, when b = 0.5, it includes a slight blend with a shape somewhat like an hourglass, and when b = 0.3, it includes a more extensive blend with an overall shape that is more elongated and slightly narrowed in the middle region). Equation 7 includes a distance-based blending function that is true when x < R and otherwise g(x) = 0. In Equation 7, a ∈ [0, 1], n N*, where R>0 is the mixing radius. Here, the mixing effect disappears when x>R; parameter n adjusts the mixing function by step size, and parameter a is fine-tuned between steps. The values ​​of one or more mixing parameters can be set directly by user-provided numerical values, or indirectly by automatically adjusting the mixing parameters to simulate the composite ablation shape specified by the user-provided grid. For example, the user can change one or more mixing parameter values, resulting in a mixing shape that can be calculated in real time, and the debugging module 40 can cooperate with the GUI module 34 to provide visual feedback on the new mixing shape. In some embodiments, the calculated shape can be plotted next to the shape proposed by the manufacturer. In some embodiments, a quality metric (e.g., Dice coefficient) can be provided to show the degree of overlap between the current mixing shape and the reference shape. Regarding the grid-based numerical determination method, when the ablation device manufacturer calculates the expected composite ablation shape (e.g., by modeling the probe and anatomical structure using bio-heat transfer), these calculated composite ablation zones can be provided as a grid or binary mask. In this case, mixing function parameters can be inferred using, for example, I-square optimization methods or other methods that minimize or reduce the error between the calculated grid and the manufacturer-provided reference ground truth grid.

[0070] As described above, the ablation therapy system 12 includes an automatic planning component (e.g., via...). Figure 2 The planning module 44 supports users (e.g., clinical experts) in establishing ablation plans. Ablation plan selection typically involves, preferably (e.g., based on one or more user criteria), or in some embodiments, optimally selecting a set of applicators to deliver single / regular and / or composite ablation zones, thereby ablating the target area (e.g., tumor plus optional margin) as well as possible (e.g., at 100% volume) while avoiding nearby organs of risk and normal tissue. The automated planning functionality of the planning module 44 can be implemented to support the planning of composite ablation zones, as well as the planning of regular ablation zones implemented using a single applicator in grid-based and freehand ablation procedures. When a composite ablation zone is tuned and used for a given treatment, multiple applicators are inserted into the patient according to a defined placement pattern and simultaneously activated to achieve the composite ablation zone within the target area. The delivered composite ablation zones are distributed along an ablation trajectory located between the various entry trajectories used for applicator insertion. Typically, this so-called composite ablation trajectory passes through the center of the composite ablation zone, which coincides with the centroid of the corresponding applicator. The direction of this composite ablation trajectory is parallel to the direction of the corresponding applicator.

[0071] Figures 4A to 4BExamples of ablation shapes for the dual-applicator linear (placement) mode are shown, determined for both meshed and unmesh environments. Applicator patterns and corresponding dimensions may be referenced and stored in the device datasheet, for example, by ablation module 38 ( Figure 2 Storage. In this specific example, refer to... Figure 4A The diagram shows an applicator set 54A (also referred to herein as an applicator pattern) including applicators 56A and 56B, which define the vertices or applicator entry trajector of the ablation trajectory 58 relative to this particular set 54A. The applicator spacing dimension d may be appropriate for a given set of applicators; in one example, it may be 1.0 cm. Figure 4B This displays an applicator set (or group) 54B, including applicators 56A and 56B, which define the vertices or applicator entry trajector of the ablation trajectory 58 relative to this particular set 54B. It is worth noting that... Figure 4A and Figure 4B The modes described in the document deliver two different composite ablation zones 60 and 62 for two different device settings (e.g., power and on-time or duration at a given probe spacing size). For example, the ablation size of composite ablation zone 60 might be based on a higher power level provided via applicator assembly 54A, with a value of dimension d, for example, 1.0 cm, compared to the ablation size of composite ablation zone 62 produced by applicator assembly 54B. In other words, applicator assemblies 54A and 54B might have similar dimensions but be used at different power and / or duration levels to provide different composite ablation shapes / sizes.

[0072] Figure 4C The composite ablation zone resulting from the setting differences is displayed, as well as the selection module 46 ( Figure 2 The example combination of composite ablation zones 60 and 62 is selected to plan the ablation treatment procedure. That is, in addition to the individual composite ablation zones 60 and 62, selection module 46 can also combine additional applicator groups using the same and / or different placement patterns to provide broader coverage, such as the combined composite ablation zones 64 and 66 shown. While in Figures 4A to 4C The diagram depicts a combination of simple composite ablation zones 60 and 62 (e.g., elliptical in shape), but the ellipse was chosen as the shape for illustrative purposes only. It should be understood that the composite ablation zone group can be ellipsoidal or non-ellipsoidal (e.g., a mixed shape), for example... Figures 1A to 1C As shown. In some embodiments, when selecting a preferred plan covering the tumor region (based on the selection of the calibrated probe), the selection module 46 may take into account geometric information about the probe group(s) and the composite ablation zone.

[0073] Having described an overview of ablation treatment planning, attention now turns specifically to grid-based planning. When delivering ablation treatments using a guided grid template (e.g., a 13×13 brachytherapy grid), the planning algorithm should explore all possible applicator pattern locations that match the grid geometry (e.g., identifying a set of placement patterns where all applicator locations precisely fit the grid apertures). However, the planning algorithm should also account for the fact that the grid may have various forms. For example, refer to... Figure 5 (a) and Figure 5 (b) shows a 2D regular rectangular grid 68 and a 2D irregular grid 70, respectively. Generally speaking, except for things like... Figure 5 (a) In addition to the regular pattern grid like grid 68, which has repeating, fixed, and regular hole spacing in the x-direction Sx and y-direction Sy, some grids may exhibit an irregular layout, with their grid holes sparsely positioned in space and without fixed spacing in the x and / or y directions, such as... Figure 5 The grid 70 in (b) is shown. These differences in grid types may affect the choice of the method for matching the applied treater pattern with the grid.

[0074] A precise match between the debugged applicator pattern and the selected mesh template is a crucial prerequisite for the actual execution of the final preferred or optimized treatment plan. In other words, if the vertices of the applicator pattern do not match the mesh hole spacing, the applicator pattern cannot be used for treatment. Figure 6 (a) to Figure 6 (d) This helps to illustrate the iterative process of achieving matches and not achieving matches. In this example, a rectangular treater pattern 74 is iterated on a mesh 72 (also referred to herein as a mesh template) (e.g., advanced and processed according to different positions / orientations). That is, the treater pattern 74 is positioned so that one of its vertices matches (e.g., anchors) the currently processed mesh hole, and it is determined whether the other vertices of the rectangular treater pattern 74 (which do not match the currently processed mesh hole and are therefore unanchored) match the mesh hole at each pattern position. Figure 6 (a) to Figure 6 In (b), Figure 6 (a) and Figure 6 (b) The first two positions of the rectangular applicator pattern 74 failed to achieve a match between the unanchored vertices and the mesh holes, therefore these applicator patterns cannot be used for disposal planning selection (and ultimately executed). Instead, Figure 6 (c) to Figure 6As shown in (d), the rectangular applicator pattern 74 has unanchored vertices at both pattern locations that perfectly match the mesh holes, thus these applicator patterns can be selected as potential entry points for disposal planning. Indeed, due to the inherent complexity of the shapes and sizes of multiple applicator patterns, as well as mesh resolution and constraints (e.g., in some mesh templates, such as Civco14GA, some column holes are never used and should be ignored during disposal planning), new sets of features are needed for subsequent planning selections (e.g., selection using ensemble coverage optimization algorithms). Certain embodiments of applicator-mesh pattern matching methods are designed to address one or more of these challenges. For example, given a set of debugged devices, applicator-mesh pattern matching methods can sample all possible applicator pattern locations that match mesh holes. Ultimately, the set of all possible sampled applicator patterns (e.g., entry points) is provided as input for subsequent disposal planning selection operations.

[0075] To determine the set of all applicator patterns that match a given mesh template geometry, certain embodiments of the applicator-mesh pattern matching method can take any of three variant forms, which can generally be described as brute-force methods, including: a first method, matching by translating applicator patterns along all rows and columns of the mesh; a second method, matching by discrete local point set matching; and a third method, matching by continuous local point set matching. As explained below, the main difference between the first and second methods lies in how the applicator patterns and mesh holes are represented and processed during mutual matching.

[0076] Generally, in mesh scenarios, sampling into a trajectory group is strongly constrained by the regular mesh geometry and the mesh aperture sampling values. For example, an equilateral triangle pattern cannot be delivered via mesh templates supported by some current systems because an equilateral triangle has a 60-degree angle, while the diagonal between apertures in a square mesh is 45 degrees. Conversely, if the relative distance between the applicators is a multiple of the mesh aperture spacing, a group of 2, 3, or 4 applicators with linear / segment patterns can be delivered via a mesh template. Figure 7 (a) to Figure 7 (f) An embodiment of a first example applicator and mesh pattern matching method (hereinafter referred to as the first method) 76 is illustrated, which shows a simple and direct way to determine the placement pattern position that matches the mesh or mesh template's mesh holes, wherein the linear pattern matches the mesh hole spacing size. For example, refer to Figure 7 (a) to Figure 7 (f) shows an example entry trajectory (and corresponding ablation trajectory) based on x-direction and / or y-direction displacement, for a dual-applicator linear pattern in a mesh usage scenario. Figure 7 (a) To highlight the features of this figure, these features also apply to... Figure 7(b) to Figure 7 The other mesh illustration in (f) shows a mesh or mesh template 78 sampled from an applicator pattern 80 comprising a pair of linearly arranged applicators. Applicator pattern 80 in this example includes: linearly arranged applicator vertices 82A, 82B corresponding to their respective applicator entry trajectories, and corresponding ablation trajectories 84, which form a linear pattern. Applicator pattern 80 processes (e.g., samples) a given row at a given xy position in mesh 78 to derive an ablation configuration area, and then iterates (e.g., shifts column-by-column), for example, in the x-direction (as indicated by shift symbol 86). This shift 86 may be 0.5 cm in some cases, or adopt any mesh aperture spacing used, or in some embodiments a multiple of the mesh aperture spacing. Applicator pattern 80 is further sampled at this advance position ( Figure 7 (b)), and the process is repeated at each advancement point in the line, such as Figure 7 (c) shows. This process is repeated for consecutive rows, as indicated by row advancement symbol 88. In general, for the first method 76, the applicator pattern 80 is initially placed at the beginning of the first row, parallel to the row axis, and iteratively shifted according to the grid hole column spacing until the end of the row is reached. Then, the same pattern is placed at the beginning of the second row and iteratively shifted until the end of the grid in that row is reached, and so on, until the last row is processed. In some embodiments, the iterative shift continuing in the second row may begin from the end of the row, and then when advancing to the third row, the iteration of the third row begins from the beginning of the third row, and so on, each consecutive row starting from its beginning or end according to some alternating or zigzag, continuous row advancement pattern. In some embodiments, the iteration may begin from the bottom row (at either end of the grid) and proceed upwards to the top left corner of the grid.

[0077] Then, the first method 76 includes orthogonally rotating the applicator pattern 80 (90 degrees) and continuing the process in a similar manner, i.e., processing the same applicator pattern 80 along all grid hole columns by moving the pattern according to the grid hole row spacing, in conjunction with the above. Figure 7 (d) to Figure 7 (f) The method described further is similar. It is worth noting that in some embodiments, method 76 may be performed in a different (e.g., reverse) order, for example, first according to... Figure 7 (d) to Figure 7 (f), rotate, then according to Figure 7 (a) to Figure 7 (c) Continue.

[0078] refer to Figure 7 (d) to Figure 7In (f), the grid template 78, the applicator pattern 80 is sampled and shifted, as indicated by shift symbol 90. This shift can also be the grid aperture spacing in an orthogonal direction (e.g., the y-direction here) (or, in some embodiments, a multiple thereof), using a similar iterative approach. Similarly, as indicated by column advance symbol 92, the sampling process continues against columns. It is noteworthy that in some embodiments, shifts 86 and 90 can be performed in different orders (e.g., reverse, alternating, etc.), and / or rows and / or columns can be skipped. The first method 76 automatically calculates the applicator pattern 80 matching the grid spacing. Subsequently, single, and / or mixed and / or unmixed composite ablation zones corresponding to these patterns are selected (e.g., using an optimization algorithm). More specifically, at each processed pattern location matching the grid, the corresponding entry trajectory (e.g., a pair of all applicator pattern vertex positions and grid plane normals) is stored in a general list of entry trajectories and provided as input to the selection module 46. Figure 2 ).

[0079] It is worth noting that the first method 76 assumes that the relative coordinates of the applicators present in a pattern are perfectly adapted to the grid spacing in both the row and column directions. If the pattern geometry does not match the grid spacing in one or both directions, certain pattern positions on the grid will be ignored (e.g., omitted from storage), resulting in a partially (or even completely) empty list of entry trajectories. Furthermore, it is worth noting that rotation in the first method 76 is restricted to ninety degrees. Additionally, the first method 76 assumes that the grid geometry is regular, for example... Figure 5 (a) shows a two-dimensional regular rectangular grid 68.

[0080] The following describes a second example of a pattern matching method between the applicator and the mesh (hereinafter referred to as the second method). The second method is more general than the first method 76 because it defines the mesh as a set of two-dimensional hole coordinates P: Where N is the total number of mesh holes (e.g., for a brachytherapy mesh with 13×13 holes, N = 169). Similarly, the placement pattern is defined as the set V of 2D relative coordinates of the pattern vertices with respect to the pattern center: Where L represents the total number of vertices in the treater pattern (e.g., for the triangle pattern, L=3, with three treaters placed along the three vertices accordingly). Compared to the first method 76, in the second method, other assumptions regarding the mesh and treater pattern geometry are no longer (strictly) maintained. Reference Figure 8 (a) to Figure 8(d) shows a brief description of the second method (denoted as second method 94a), using a two-dimensional regular rectangular mesh 96 and a treater pattern 98 in the form of a rectangular pattern with four vertices. In this example, one vertex (top left corner) of the treater pattern 98 is anchored at... Figure 8 In (a), on the currently processed mesh hole 100, anchored vertices are displayed as solid black dots, while unanchored vertices are displayed as solid white squares. Figure 8 In (b), the applicator mode 98 is shifted to the left so that another top vertex (top right corner) now matches the currently processed mesh hole 100 and becomes the anchor vertex. Figure 8 In (c), Figure 8 (a) The applied vertex pattern 98 shown is shifted upwards so that another vertex (bottom left) matches the currently processed mesh hole 100, which now becomes the anchor vertex. Then... Figure 8 In (d), Figure 8 (c) shows that the applicator pattern 98 is translated to the left so that the lower right vertex matches the currently processed mesh hole 100, becoming the anchor vertex. It is worth noting that this example does not assume a specific translation order in all cases, and in some embodiments, the translations may be performed in a different order to successively anchor each vertex of the applicator pattern 98 to the currently processed mesh hole 100. Figure 8 (a) to Figure 8 Each pattern position / orientation shown in (d) is accepted and stored because all unanchored pattern vertices at each position match a mesh hole. In the second method 94a of the example, in general, a given administrator pattern is iteratively anchored to each point P in set P via each vertex Vi in its set V. j In one embodiment, if and only if all L pattern vertices existing in V also exist in set P (i.e., (All mode vertices are also mesh holes), an anchored at point P. j Only the appropriate healer pattern will be accepted and collected into a suitable pattern list. Figure 9 A shows two examples where the positioning of the applicator pattern 98 on mesh 96 is rejected because not all vertices at every depicted pattern placement location match mesh holes. On the other hand, Figure 9 B shows two examples where the location of the treater pattern 98 on mesh 96 is accepted and stored (collected) because all vertices of each depicted pattern placement location match the mesh holes.

[0081] As part of the matching process, the second method 94a can optionally be extended to include step angle rotation. (See reference) Figure 10 (a) to Figure 10(d) shows an example of second method 94b, illustrating step angle rotation. The figure shows mesh 96, generator pattern 98, and the currently processed mesh hole 100, with each vertex of generator pattern 98 being iteratively anchored. In this example, with... Figure 8 (a) Similarly, one vertex of the applicator pattern 98 (the top-left vertex when viewed in a horizontal orientation parallel to the longest axis of the applicator pattern 98) is in Figure 10 In (a), it is anchored to the currently processed mesh hole 100, but rotated by a defined angular step. In other words, for the currently processed mesh hole 100 and each vertex, it can be configured as described above for... Figure 8 (a) to Figure 8 (d) Describes the translation of the treater pattern 98, and a defined rotation to perform the matching. Figure 10 In (b), for a given angular rotation, the applicator pattern 98 is translated upwards and to the right, so that the lower left vertex (also viewed from the perspective when the applicator pattern 98 is in a horizontal position) now matches the currently processed mesh hole 100 and now becomes the anchor vertex. Figure 10 (c) to Figure 10 In (d), for a given angular rotation, the applicator pattern 98 is translated such that the bottom right vertex becomes the anchor vertex, and the top right vertex (using the same horizontal orientation as the pattern described above) becomes the anchor vertex. It is worth noting that in this example, no pattern will be accepted because unanchored vertices do not match the mesh holes. It is also worth noting that the order of translation may differ from... Figure 10 (a) to Figure 10 (d) The order depicted in the example is different. In one embodiment, the treater pattern 98 is rigidly rotated by different angular steps γ to verify and match all possible discrete rotation pattern positions and orientations on the mesh plane. The angular step size can be pre-configured or configured by the user (e.g., via a configuration menu). Smaller step sizes result in slower iterative matching searches but may achieve higher matching accuracy (conversely, larger step sizes may result in faster performance and lower relative accuracy). Pattern rotation can be achieved by simply multiplying the coordinates of all pattern vertices in set V by a basic 2D rotation matrix: It is worth noting that an embodiment of the second method, which includes the common functions of the second methods 94a and 94b, may be referred to as the second method 94.

[0082] In some embodiments, when matching the pattern vertices in V with the mesh hole positions in P, the algorithm of the second method can provide and use a certain distance tolerance (epsilon or This feature allows for the acceptance and collection of patterns that are not precisely located on the grid holes, but whose distance is still within a given tolerance. This characteristic can be beneficial in clinical scenarios where clinicians accept some inaccuracies when the applicator is placed via the grid. To further explain, tolerance... The aim is to ensure that even when the distance between the mesh hole and the pattern vertex is less than a given epsilon... In certain situations, the insertion pattern can be selected. This tolerance provides greater freedom in the matching search. The returned insertion position may not be precisely executed (e.g., the pattern of the applicator ultimately located via the mesh holes may differ slightly from what is shown on the device datasheet). However, using tolerance means that clinicians will assume responsibility for allowing minor deviations from the datasheet information, by adding a certain amount of... Tolerance (e.g., in x millimeters).

[0083] In some embodiments, to avoid redundant calculations, the iterative matching search process of the second method ignores already located patterns that have been anchored and processed at the current and previous points in P. For example, refer to Figure 11 (a) to Figure 11 (d) The figure shows practitioner mode 98. Although the mesh holes corresponding to the anchor vertices of practitioner mode 98 have changed, the same mesh holes are matched during the search. For example, in Figure 11 In (a), the processed mesh hole 100a is located at the center of mesh 96, and the top left vertex of the applicator mode is anchored here. Figure 11 In (b), the processed mesh hole 100 is located in Figure 11 (a) The upper right vertex of the applicator pattern 98 is not anchored, but it matches the same mesh hole as the vertex of applicator pattern 98. Therefore, in some embodiments, the second method skips or ignores (e.g., omits) the collection or storage. Figure 11 (b) Treatment device mode 98. Similarly, Figure 11 (c) and Figure 11 (d) shows that the anchor vertex is located at Figure 11 (a) The positions of the unanchored lower left and unanchored lower right vertices in the treatment mode 98, therefore Figure 11 (c) to Figure 11 These treater patterns 98 in (d) are skipped or ignored during storage because they are related to Figure 11 (b) is the same as the treatment mode 98, targeting Figure 11 The treatment mode 98 in (a) is redundant.

[0084] It is worth noting that in some embodiments of the second method 94, in certain cases, when treatment is performed using a 2D rigid regular mesh (e.g., a brachytherapy mesh template) with fixed spacings Sx and Sy along the mesh columns and rows, the matching algorithm of the second method can utilize this additional geometric information to filter out treatment patterns that can never fit the mesh geometry. For example, a pattern-relative-mesh geometry check can be performed by calculating all relative distances of all vertex pairs in set V along the x and y mesh plane axes. If all calculated distances are equal to (or multiples of) the corresponding Sx and Sy spacings, the current treatment pattern may match a mesh hole and can be further processed to retrieve its (all) possible feasible locations on the mesh; otherwise, the currently processed treatment pattern may be directly ignored (not stored).

[0085] Now for reference Figure 12 , Figure 12 This is a flowchart illustrating one embodiment of Example Second Method 94. Method 94 receives input in step 102. For example, Method 94 receives multiple objects, including a mesh, a treater pattern, and a rotation step. These objects can be received from local or remote storage, for example, via a GET command. In step 104, Method 94 receives the mesh's geometry data from the mesh object. Again, this can be achieved via a GET command. While steps 102 and 104 are described using the GET command, other existing mechanisms and / or programming languages ​​can also be used to receive information and / or perform similar functions. Method 94 also includes initializing a list of all mesh patterns in step 106. For example, this is the initialization of a list of matching insertion patterns found in the iterative process. In step 108, Method 94 sets the angle rotation γ to zero.

[0086] In step 110, method 94 rotates the applicator pattern by an angle (e.g., zero angle rotation in the first loop). In step 112, method 94 determines whether the applicator pattern matches the mesh geometry. For example, in step 112, method 94 performs a pattern-relative-mesh geometry check, which is performed by calculating all relative distances along the x and y mesh plane axes for all vertex pairs in set V. If all calculated distances are equal to (or, in some embodiments, multiples of) the corresponding Sx and Sy spacings, the current pattern may match a mesh hole and can be further processed to retrieve its multiple or all possible feasible locations on the mesh (otherwise, the currently processed pattern can be ignored). In other words, if the determination in step 112 is yes, method 94 proceeds to step 114, in which the method calculates a list of all mesh positioning patterns. As described above, in one embodiment, method 94 matches the applicator pattern on all mesh holes and adds all matching patterns to a list. It is worth noting that... Figure 12The flowchart shows an outer loop that iteratively increases the angle rotation γ and adds newly discovered patterns to the return list. The flowchart also shows an inner loop (e.g., within step 114) where method 94 calculates a total list of patterns by iteratively translating and anchoring the currently rotated applied pattern across all mesh holes (similar to the process performed by the first method 76). In other words, the outer loop increases the angle rotation γ by a step size, while the inner loop (in step 114) translates this rotated pattern across all mesh holes.

[0087] In step 116, method 94 adds the list of current pattern positions (matching mesh holes) from step 114 to the list of all mesh patterns. In step 118, method 94 determines whether the angular rotation is less than 359 degrees. It is worth noting that step 118 may also be reached if step 112 determines it to be negative (e.g., no). If the angular rotation is less than 359 degrees, the method proceeds to step 120, where the angular rotation is adjusted by a rotation step (the rotation step used to rotate the insertion pattern, in degrees). In one embodiment, the rotation step is user-defined. From there, the loop continues (e.g., in step 110). If the angular rotation is not less than 359 degrees in step 118, the method proceeds to step 122, where a list of all matching mesh-relative applicator patterns is returned. Similar to what is done with the first method 76, the returned list (e.g., a list of all applicator patterns matching the mesh) is provided as input to selection module 46. Figure 2 ).

[0088] It is worth noting that, similar to the `get` command, the `return` command used here is merely illustrative of a function within a flow using the example programming language, and in some embodiments, similar functionality can be performed using other existing mechanisms (e.g., using other programming languages). In some embodiments, certain steps may be combined or otherwise modified to perform similar functionality.

[0089] Having described the first method 76 and the second method 94, attention now turns to a third example of a treater-mesh pattern matching method (hereinafter referred to as the third method), which uses continuous local point set matching to perform the matching. The third method can be viewed as a modification of the second method, where continuous treater pattern rotation is implicitly performed within the (brute-force) iterative mesh pattern matching search process. That is, the angular step size in γ is not required to verify and match all possible rotational pattern positions and orientations on the mesh plane. Similar to the second method, for the third method, the mesh is also defined as a set of 2D hole coordinates P, and the treater pattern consists of a set V of 2D relative coordinates of the 2D pattern vertices relative to the pattern center position. In this third iterative matching algorithm, for all mesh points P in the set P... j Perform iterations. In each j-th iteration, the first administrator mode vertex V1 is positioned at the currently processed grid point P. j Above (e.g.) Figure 13 (As shown further), the algorithm then continues based on the following different conditions.

[0090] If L=1 (e.g., the treater pattern includes only a single vertex, or a single treater delivers a single non-composite ablation trajectory or region), then in V1=P j The current treater pattern at the location is collected into a list of selected mesh-matching treater patterns, and the algorithm continues to the next iteration and moves on to the next mesh point in set P. In the simple case where the total number of treater pattern vertices (i.e., L) equals 1, the third method translates one point over all mesh holes. This is an extreme case where only non-composite ablation shapes (e.g., standard ellipsoidal ablation zones) are used to cover standard treatment delivery.

[0091] For the case where L>2, refer to the third method (in Figure 13 (a) to Figure 13 (f) is denoted as 124), where Figure 13 (a) The steps of method 124 (a) are illustrated. Figure 13 (b) illustrates step (b) of method 124, while Figure 13 (c) to Figure 13 (f) Step (c) for illustrated method 124. For the case of L=2, the first administrator mode vertex V1 is placed at the currently processed grid point P. j Then (step (a)) Figure 13 (a) Calculate the distance d between mode vertices V1 and V2. 12 (As shown in the top diagram between steps (a) and (b),) and perform the operation on all items related to P in set P. j Distance d 12The search for grid points. Here, the subset of grid points found is defined as S. j (For example, see step (b)) Figure 13 (b), where subset S j This includes grid points at specified distances along the dashed circle (in this example, four grid points at positions 12, 3, 6, and 9 on the dashed circle). If the cardinality of the set is |S j |=0 (that is, no match for P was found in set P) j Distance d 12 If the grid points are not specified, then the current j-th iteration stops, and the algorithm proceeds to the next iteration. However, if |S... j If |=ns>0, then all combinations of ns types of healer modes (where V1 is located at P) j V2 is alternately positioned in set S j Each grid point in the set P is collected into a list of selected grid matching patterns. Finally, the algorithm continues to the next iteration and moves on to the next grid point in the set P. In fact, the third method for L=2 is to search at all angles in continuous space.

[0092] If L>2 and |S j If |=ns>0, then by positioning V1 at grid point P j And alternately position V2 in set S j Each of the s grid points in the set is used to determine the location of the potential treater pattern in the set ns (see step (b)). Figure 13 (b)). For each of these s pairs of pattern vertices in the grid positions (V1, V2), all the remaining ones can be easily retrieved using the pattern relative coordinates in set V. The position of each agent pattern vertex (see...) Figure 13 (c) to Figure 13 (f)). If and only if the remainder When all pattern vertices also exist in set P (i.e., (All pattern vertices are also mesh holes), and each of these ns located treater patterns will be included in the list of selected mesh matching patterns. Finally, the algorithm continues to the next iteration and moves on to the next mesh point in set P.

[0093] To explain further, Figure 13 (a) to Figure 13 (f) A rectangular applicator pattern (shown in the lower diagram between steps (a) and (b)) is used to illustrate the iterative algorithm of the third method for the case L>2. Figure 13 In (a), step (a) of the iterative algorithm is shown, where the first pattern vertex V1 is placed at the currently processed grid point P.j Location. Figure 13 (b) shows the steps of the iterative algorithm, where the subset S j It consists of 4 grid points, which are related to the processed grid point P. j The distance is d 12 =|V1-V2|. In Figure 13 (c) to Figure 13 (f) shows step (c) of the iterative algorithm, where V1 is positioned at grid point P. j (Step (a)), and alternately position V2 in set S. j For each of the four grid points in set P (step (b)), the location set of four potential remedy patterns is retrieved and displayed; for each of these four pairs of grid locations (V1, V2), the locations of the remaining two remedy pattern vertices V3 and V4 (white squares) can be easily retrieved using the pattern relative coordinates in set P. The remaining two remedy pattern vertices V3 and V4 are retrieved if and only if they also exist in set P (i.e., ...). Each of these four located treater patterns (all pattern vertices are also mesh holes) will be included in the list of selected mesh matching patterns.

[0094] It is worth noting that some of the features described above for the second method also apply to the third method. For example, for the third method, when matching the pattern vertices in V with the mesh hole positions in P, the algorithm can provide and use distance tolerance. This allows for the simultaneous acceptance and collection of patterns that are not precisely located on the mesh holes, but whose distance is still less than a given tolerance. Similarly, to avoid any redundant computation in the iterative matching search process, all computed treater patterns that have been processed in previous iterations or whose vertices are partially outside the mesh plane's view will be ignored.

[0095] Finally, the list of all selected treater patterns matching the mesh is combined with the mesh plane normal vector to populate the list of all entering trajectory groups, which is used as the selection module 46. Figure 2The input is then described. To further explain, once the mesh is located by a clinical expert, it can be perfectly represented in 3D space using a local coordinate system, where the xy-axis of the 3D guiding mesh plane is typically parallel to the two longer sides of the mesh, while the z-axis is orthogonal to the xy-plane and parallel to the third shorter side of the mesh (i.e., the mesh plane normal). The direction of the z-axis also typically points from the outside towards the inside of the subject's body (i.e., towards the tumor region of interest). In one embodiment, the algorithm automatically calculates the positions of a set of applicator patterns that are matching mesh holes. These mesh holes are simply 3D points on the mesh xy-plane. To represent a "trajectory" in 3D space, a point (i.e., the trajectory origin) and a direction vector are needed. Therefore, a list of all entering trajectory groups is obtained by adding the mesh plane normal (i.e., the trajectory direction vector) to the entry point (i.e., the trajectory origin) of the automatically calculated applicator pattern.

[0096] Although primarily described within the context of composite ablation, it should be understood that all three methods (first, second, and third) also support a special case: a standard, single-application virtual ablation zone delivered via a single applicator. In this case, for example, for the second and third methods, the pattern vector V comprises only a single element, and the iterative algorithm returns as a result a list of entry trajectories orthogonal to the mesh plane and originating from all mesh apertures.

[0097] Now for reference Figure 14 (a) to Figure 14 (c) illustrates examples of composite ablation zones generated via linear and square applicator patterns and using a third-method mesh pattern matching approach. Composite zones generated from multiple applicators inserted via selected insertion patterns may have different shapes. While it should be understood that composite zones can be simulated using mixed shapes, Figure 14 (a) to Figure 14 (c) In the simulation example depicted, a simple ellipsoid (unmixed, e.g., via the implicit function described above) is used here for ease of illustration. Figure 14 In (a), applicator pattern 126 comprises two linearly arranged applicators inserted into a 13×13 regular brachytherapy grid 128. Applicator pattern 126 delivers a simulated composite (ellipsoidal) ablation zone 130. Figure 14 In (b), the applicator pattern 132 comprises four applicators arranged in a rectangular pattern, inserted into a grid 134 consisting of 21×21 holes with the middle rows and columns removed. The applicator pattern 132 delivers a simulated composite ablation zone 136. Similarly, in Figure 14 In (c), the applicator mode 138 includes four applicators, which are (compared to) Figure 14(b) Different methods were used to insert into mesh 134 and deliver simulated composite ablation zones 140 as shown. Each of these simulations was the result of mesh pattern matching, i.e., the third method identified a set of all placement patterns that would fit all applicator positions to the mesh.

[0098] In view of the above description, it should be understood that an embodiment of a method for calculating the pattern matching of the agent and the mesh for agent insertion patterns in mesh-based ablation (e.g., in one embodiment using...) Figure 2 The computing device 14), which generally describes three methods, in Figure 15 This is referred to as method 142. Method 142 includes: receiving mesh and applicator information (144); performing computational iterations on applicator patterns along multiple holes in the mesh, wherein the computational iterations include advancing applicator pattern objects through the entire mesh object, wherein the applicator pattern is defined by one or more vertices corresponding to one or more applicator entry trajectories and an ablation trajectory (146); and storing pattern position information of corresponding applicator patterns that match the holes in the mesh, wherein the pattern position information includes entry trajectories containing the positions of one or more applicator pattern vertices and mesh plane normals (148).

[0099] Although the invention has been illustrated and described in detail in the accompanying drawings and the foregoing description, such illustrations and descriptions should be considered illustrative or exemplary, not restrictive; the invention is not limited to the disclosed embodiments. Other variations of the disclosed embodiments can be understood and implemented by those skilled in the art upon studying the drawings, the disclosure, and the appended claims in practicing the claimed invention. It is noteworthy that various combinations of the disclosed embodiments can be used; therefore, reference to an embodiment or an example is not intended to exclude features of that embodiment from use in combination with features of other embodiments. In the claims, the word "comprising" does not exclude other elements or steps, and the words "a" or "an" do not exclude multiple. A single processor or other unit can implement the functions of several items recited in the claims. The fact that certain measures are recited in dissimilar dependent claims does not in itself imply that combinations of these measures cannot be advantageously used. Computer processes can be stored / distributed on suitable media, such as optical or solid-state media provided as part of or with other hardware, but can also be distributed in other forms. No reference numerals in the claims should be construed as limiting the scope.

Claims

1. A method (142) of ablation device and grid pattern matching for computing grid-matched insertion poses of an ablation device insertion pattern used in grid-based ablation, the method comprising: at a computing device (14): receiving grid and ablation device information (144); computing iterations of an ablation device pattern along a plurality of holes in a grid, wherein a computing iteration comprises advancing an ablation device pattern object through an entire grid object, wherein the ablation device pattern is defined by a plurality of vertices corresponding to a plurality of ablation device entry trajectories and an ablation trajectory (146), wherein each vertex corresponds to an ablation device entry point of a corresponding ablation device entry trajectory, wherein the ablation trajectory is parallel to and between the plurality of ablation device entry trajectories; and storing pattern position information of a corresponding ablation device pattern that matches the holes in the grid during the iterations, wherein the pattern position information comprises entry trajectories comprising ablation device pattern vertex positions and grid plane normal vectors (148).

2. The method according to the preceding claim, wherein, The computing iterations comprise iteratively advancing the ablation device pattern by advancing the ablation device pattern one hole in a first direction and subsequently one hole in an orthogonal second direction until all possible ablation device patterns have been tried in the first and second directions (76).

3. The method according to any of the preceding claims, wherein, The first direction is parallel to a row or column.

4. The method according to any of the preceding claims, wherein, The computing iterations comprise, for each hole of the grid: successively anchoring each of the vertices of the ablation device pattern to the each hole of the grid; determining, for each ablation device pattern position relative to the anchored vertex, whether all other vertices of the ablation device pattern match respective holes of the grid, wherein storing the pattern position information comprises storing the pattern position information for each of the ablation device pattern positions corresponding to the anchored vertex for which the other vertices match the respective holes of the grid at each of the ablation device pattern positions corresponding to the anchored vertex; and rejecting ablation device pattern positions for a given anchoring position if at least one of the vertices does not match a grid hole at the given anchoring position (94).

5. The method of any of the preceding claims, further comprising: rotating the ablation device pattern by a defined angle step and repeating the successive anchoring, determining, storing, and rejecting.

6. The method of any of the preceding claims, further comprising: applying a distance tolerance in determining the match between the vertices and the holes.

7. The method of any of the preceding claims, further comprising: omitting redundant pattern position information from storage or pattern position information for ablation device patterns in which at least one of the vertices is outside the grid.

8. The method according to any of the preceding claims, wherein, The ablation device pattern comprises two vertices and, for a given hole coinciding with one of the two vertices, the computing iterations comprise: determining a distance between the two vertices; searching for all holes at the distance from the given hole, wherein storing the pattern position information comprises storing pattern position information for all stent patterns of holes at the distance where the two vertices match the grid; and repeating the determining and the searching for successive holes (124).

9. The method according to any of the preceding claims, wherein, The stent pattern comprises more than two vertices, and for a given hole coinciding with one of the two vertices, the computation iteration comprises: determining a distance between the two vertices; searching for all holes at the distance from the given hole and defining each pair of vertices matching the holes along the search distance as a set; for each set, determining remaining stent pattern vertices, wherein storing the pattern position information comprises storing pattern position information for each set if the remaining vertices of the stent pattern of each set also match holes of the grid; and repeating the determining, the searching, and the determining for successive holes (124).

10. The method of any of the preceding claims, further comprising: Applying a distance tolerance in determining the match between the vertices and the holes.

11. The method of any of the preceding claims, further comprising: Omitting from storage redundant pattern position information or pattern position information for stent patterns where at least one of the vertices is outside the grid.

12. The method of any of the preceding claims, further comprising: Providing the stent patterns and the pattern position information to a planning algorithm to select stent patterns for use in an ablation procedure.

13. The method of any of the preceding claims, wherein, For a single vertex, the pattern position information comprises a list of entry trajectories corresponding to all holes of the grid.

14. A computing device (14) for implementing any of the preceding method claims.

15. A non-transitory computer-readable storage medium (20) comprising instructions that, when executed by one or more processors, cause the one or more processors to implement a method according to any of the preceding claims.

Images