[go: up one dir, main page]

WO2024124360A1 - Système et procédé de planification de trajectoire basée sur une optimisation - Google Patents

Système et procédé de planification de trajectoire basée sur une optimisation Download PDF

Info

Publication number
WO2024124360A1
WO2024124360A1 PCT/CA2023/051683 CA2023051683W WO2024124360A1 WO 2024124360 A1 WO2024124360 A1 WO 2024124360A1 CA 2023051683 W CA2023051683 W CA 2023051683W WO 2024124360 A1 WO2024124360 A1 WO 2024124360A1
Authority
WO
WIPO (PCT)
Prior art keywords
delivery
target
processor
trajectory
paths
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CA2023/051683
Other languages
English (en)
Inventor
Michael KUDLA
Deidre BATCHELAR
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Provincial Health Services Authority
Original Assignee
Provincial Health Services Authority
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Provincial Health Services Authority filed Critical Provincial Health Services Authority
Publication of WO2024124360A1 publication Critical patent/WO2024124360A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00—Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/10—Computer-aided planning, simulation or modelling of surgical operations
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B17/00—Surgical instruments, devices or methods
    • A61B17/34—Trocars; Puncturing needles
    • A61B17/3403—Needle locating or guiding means
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
    • A61N5/00—Radiation therapy
    • A61N5/10—X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy
    • A61N5/1001—X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy using radiation sources introduced into or applied onto the body; brachytherapy
    • A61N5/1014—Intracavitary radiation therapy
    • A61N5/1016—Gynaecological radiation therapy
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
    • A61N5/00—Radiation therapy
    • A61N5/10—X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy
    • A61N5/103—Treatment planning systems
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00—Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/10—Computer-aided planning, simulation or modelling of surgical operations
    • A61B2034/107—Visualisation of planned trajectories or target regions

Definitions

  • This disclosure relates generally to trajectory planning and in particular to methods and systems for planning and optimizing trajectory paths through a target volume.
  • Pathfinding is a common requirement of many industries. In particular, determining the required trajectory from a starting point to a desired target location was initially performed manually based on a user’s experience and judgement. However, such human determinations may be prone to error and inconsistent results. In particular in fields in which the target zone is a region requiring access at a plurality of locations, such human determinations of the multiple paths are time consuming to produce and may be further prone to an uneven distribution within that zone or region. Examples of such fields may include treatment of tumors, planning flight paths of drones and the like.
  • HDR-BT Image-guided vaginal and pelvic high dose rate brachytherapy
  • EBRT pelvic External Beam Radiation Therapy
  • IS Interstitial
  • IC needle trajectory planning have been primarily for vaginal and/or cervical brachytherapy applications.
  • These optimization techniques have been organized around identifying needle start and end positions and determining a trajectory path through the allowable trajectory space.
  • the optimization problem is constructed as a linear or quadratic objective function, with linear weighted costs (rather than hard constraints).
  • the objective is typically to find the shortest and straightest path while being constrained to certain curvature and intersection limits (either with other needles or other areas of interest).
  • the optimization is then solved using a technique called sequential convex optimization.
  • brachytherapy involves the use of radioactive isotopes emitting ionizing radiation which can be lethal to human cells.
  • Brachytherapy radiation sources typically come as small seed-like sources. Broadly, these sources can be divided into two categories: low dose rate (LDR) sources, which are typically low activity and low energy sources, and HDR sources, which are high energy and high activity sources.
  • LDR low dose rate
  • HDR high energy and high activity sources.
  • the HDR source is welded to a wire which is a component of a robotic device called an afterloader.
  • the afterloader stores the source in a tungsten safe and can be programmed to drive the source wire to predetermined positions (dwell positions), each for a predetermined time (dwell time), inside brachytherapy applicators.
  • BT treatment is as follows: One or multiple treatment applicators are inserted to the vagina, cervix, or into tissue directly. The applicator is secured in place, and then imaging techniques such as Computed Tomography (CT) or Magnetic Resonance Imaging (MRI) are used to confirm that it is properly positioned. The treatment target volume and applicator positioning are identified on postoperative images, and dwell time planning is performed. Once the dwell time planning is completed, the BT plan is delivered by transiting the HDR-BT radioactive source to each dwell position, for each planned dwell time.
  • CT Computed Tomography
  • MRI Magnetic Resonance Imaging
  • Brachytherapy can be used successfully for treatment of many cervical and vaginal tumours.
  • BT becomes more difficult to perform, and IS needles are often required to treat the tumour adequately. This is because high dwell times are required to treat entire large tumours from within IS applicators, which results in too-high organ doses. Needles are required, to assist in placing the sources directly within the tumour, so that the BT dose can cover the tumour while sparing organ tissue.
  • IC/IS applicator options While there are a variety of “one-size-fits-all” IC/IS applicator options available, most centers have limited different applicators at their disposal, reducing the clinical practicality of patient-specific source placement.
  • a method for determining at least one set of plurality of paths to a target zone comprising receiving at a processor, at least one dataset representing a target zone, providing an introduction region extending along an axis to the target zone, utilizing the processor, defining at least one non-intersecting target path through the introduction region and extending into the target zone and outputting a representation of the plurality of trajectory paths.
  • the delivery region may be defined by a delivery body.
  • the delivery body may include a plurality of delivery passages, each of the plurality of delivery passages extending along the delivery body from a free proximal end to a delivery exit, each delivery exit being axially aligned with one of the plurality of delivery trajectory path.
  • Each delivery passage may be sized and shaped to permit passage of a needle delivery member therethrough to the target zone along the delivery trajectory path.
  • the processor may be configured to define at least a first surface around the introduction region intersecting the target zone, defining a plurality of sites located within the target region on the first surface on the first surface and iteratively grouping the sites using centroidal veronoi tessellations into a predetermined quantity of first target locations.
  • the method may further comprise, utilizing the processor, defining at least one successive expanded surface after the first surface, defining a plurality of successive sites located within the target region on the successive expanded surface, using the first target locations, iteratively grouping the successive sites using centroidal veronoi tessellations into a predetermined quantity of successive target locations and defining a best fit line between each successive target location with its corresponding first target location.
  • the method may further comprise utilizing the processor, limiting each best fit line to a predetermined angle or constrained operating space relative to the axis of the introduction region.
  • the method may further comprise utilizing the processor extending each best fit line to a unique non-intersecting delivery passage extending along the introduction region.
  • the method may further comprise utilizing the processor, radiusing the intersection between each delivery passage and its corresponding exit to a predefined minimum radius.
  • the method may further comprise providing a plurality of sets of trajectory paths determined from a plurality of input conditions.
  • the method may further comprise receiving an input representing a defined body, scanning a region around the defined body with a reference introduction region in its desired position and generate a cloud of points around and within the defined body to define the target volume.
  • the defined body may be sized to correspond to a delivery body, the delivery body containing passages each ending aligned with at the at least one deliver trajectory path. Scanning may be performed from a group consisting of magnetic resonance imaging and computed tomography.
  • the method may further comprise utilizing the processor converting the delivery body in to a three dimensional model having passages therethrough ending in passage exits each aligned with one of the at least one delivery trajectory paths.
  • the method may further comprise manufacturing the delivery body according to the three dimensional model utilizing a computer controlled manufacturing device.
  • the computer controlled manufacturing device may comprise an additive manufacturing device.
  • a system for generating at least one set of plurality of paths to a target zone comprising a processor, a communication system operable to receive at least one image representing a target zone and a memory system having stored thereon executable instructions that, when executed by the processor cause the system to define at least one non-intersecting trajectory path through the introduction region and extending into the target zone and output a representation of the plurality of trajectory paths.
  • Figure 1 is an illustration of a plurality application paths extending through and exiting a delivery template along a plurality of trajectory paths to a target zone as determined by the method of the present disclosure.
  • Figure 2 is an illustration of a system for determining and creating an individualized delivery template having a plurality of trajectory paths extending therefrom.
  • Figure 3 is a schematic of the system of Figure 2.
  • Figure 4 is a delivery template apparatus produced by the method of the present disclosure.
  • Figure 5 is an end view of the delivery template apparatus of Figure2.
  • Figure 6a-7i are illustrations of the stages for the process for determining the plurality of delivery trajectory paths in accordance with the method of the present disclosure.
  • Figure 7a-7d are illustrations of the optimization of the centroids of a single shell for use in determining the delivery trajectories in the method of Figure 4.
  • Figure 8a-8d are illustrations of the trajectory angular optimization for use in the system of Figure 2.
  • Figure 9a-9c are illustrations of the process for forming introduction paths and radiused transitions for each of the plurality of delivery trajectory paths in the system of Figure 2.
  • Figure 10a-10b are illustration of the process for providing straight delivery paths through the delivery template in the system of Figure 2.
  • Figure 11a-11C are images of delivery paths to a tumour as compared to conventional delivery plan.
  • Figure 12 is a perspective view of a delivery template as used in the system of Figure 2.
  • Figure 13 is a flow chart of the process for determining a plurality of delivery trajectory paths in the system of Figure 2.
  • Figure14 is a flow chart of the process for optimizing the delivery trajectory paths of the System of Figure 2.
  • the target volume 6 may be any region defining a space which may include the target object 8, such as a tumour into which a plurality of non-intersecting paths are required.
  • target volumes may be wildfires, requiring multiple approaches for fire fighting vehicles, areal display volumes requiring multiple approaches for display vehicles or other autonomous swarm drones or vehicles, or cancer treatment requiring multiple injection sights for radiation or drug introduction to a tumour.
  • the present disclosure includes systems and methods for generating candidate needle trajectory sets, such as for use in gynecological high dose rate brachytherapy (HDRBT) using patient-specific cylinder templates, and auto-planned trajectory set-based preplans are evaluated for a retrospective patient cohort.
  • HDRBT high dose rate brachytherapy
  • one trajectory set from a generated library may be selected with optimized pre-plan dwell times to generate a simulated preplan.
  • a customized interstitial treatment template for patients may be constructed using rapid additive manufacturing to accurately places flexible HDRBT needles in the selected pre-planned positions.
  • the system 10 includes a processor 12 operably coupled to a network or input/output interface 20, a scanner 26 and a manufacturing device 28.
  • the system 10 includes a memory 14 that stores machine instructions that, when executed by the processor 12, cause the processor 12 to perform one or more of the operations and methods described herein.
  • the memory 14 may be of any known type including a cache memory unit for temporary local storage of instructions, data, or computer addresses.
  • the system 10 may further include display 16 for displaying one or more system outputs to a user and input device 18 for receiving and inputs from the user, such as, without limitation a mouse or a keyboard.
  • the network interface 20 may comprise a network input 22 operable to receive a scanned image or other data for use by the processor and a network output 24 operable to transmit one or more outputs from the processor 12 across a network as are commonly known.
  • the scanner may comprise any scanning device operable to scan the target object and surrounding regions.
  • the scanner may comprise a magnetic resonance imaging device, computed tomography, positron emission tomography, image scanners or the like.
  • the manufacturing device 28 may comprise a computer controlled machine, such as a mill or lathe, or an additive manufacturing device such as a 3D printer as are commonly known.
  • processor is intended to broadly encompass any type of device or combination of devices capable of performing the functions described herein, including (without limitation) other types of microprocessors, microcontrollers, other integrated circuits, other types of circuits or combinations of circuits, logic gates or gate arrays, or programmable devices of any sort, for example, either alone or in combination with other such devices located at the same location or remotely from each other. Additional types of processor(s) will be apparent to those ordinarily skilled in the art upon review of this specification, and substitution of any such other types of processor(s) is considered not to depart from the scope of the present invention as defined herein.
  • the processor 12 can be implemented as a single-chip, multiple chips and/or other electrical components including one or more integrated circuits and printed circuit boards.
  • Computer code comprising instructions for the processor(s) to carry out the various embodiments, aspects, features, etc. of the present disclosure may reside in the memory 14.
  • the code may be broken into separate routines, products, etc. to carry forth specific steps disclosed herein.
  • the processor 12 can be implemented as a single-chip, multiple chips and/or other electrical components including one or more integrated circuits and printed circuit boards.
  • the processor 12 together with a suitable operating system may operate to execute instructions in the form of computer code and produce and use data.
  • the operating system may be Windows-based, Mac-based, or Unix or Linux-based, among other suitable operating systems. Operating systems are generally well known and will not be described in further detail here.
  • Memory 14 may include various tangible, non-transitory computer-readable media including Read-Only Memory (ROM) and/or Random-Access Memory (RAM). As is well known in the art, ROM acts to transfer data and instructions uni-directionally to the processor 12, and RAM is used typically to transfer data and instructions in a bi-directional manner. In the various embodiments disclosed herein, RAM includes computer program instructions that when executed by the processor 12 cause the processor 12 to execute the program instructions described in greater detail below. More generally, the term “memory” as used herein encompasses one or more storage mediums and generally provides a place to store computer code (e.g., software and/or firmware) and data. It may comprise, for example, electronic, optical, magnetic, or any other storage or transmission device capable of providing the processor 12 with program instructions. Memory 14 may further include a floppy disk, CD- ROM, DVD, magnetic disk, memory chip, ASIC, FPGA, EEPROM, EPROM, flash memory, optical media, or any other suitable memory from which processor 12 can read instructions in computer programming languages.
  • the system 10 is adapted to receive a data file of a target object, such as, by way of non-limiting example a tumour and determine one or more target paths for delivery thereto.
  • the target paths may define delivery passages or paths of needles, radiation, medication or the like.
  • the system 10, using the manufacturing device may be adapted to rapidly manufacture a delivery body 40 as illustrated in Figures 4 and 5.
  • the delivery body 40 for use, by way of non-limiting example in treating cervical cancers may comprise an elongate body extending between proximate and distal ends, 42 and 44, respectively.
  • the elongate body includes a plurality of delivery passages 46 extending therethrough from the proximate end to positions proximate to the distal end 44.
  • the exit end 48 of the delivery passages are manufactured to align with the target paths as determined by the system as will be more fully described below.
  • the delivery body 40 may be positioned within the vagina of a patient with the exit ends 48 oriented in the desired orientations and flexible needles or other delivery members 50 passed therethrough to the target body so as to provide a specified treatment to that target body.
  • the delivery body 40 may be formed of any suitable material including plastics and the like.
  • the processor 12 may receive an initial image or scan of the target body.
  • the target body may be a tumour as shown in Figure 13 as step 202 of an exemplary method 200 of the present disclosure.
  • Such images are commonly in a Digital Imaging and Communications in Medicine (DICOM) standard although it will be appreciated that other image types may be utilized as well.
  • the target volume may be defined by the data file (which may be of a different size, and in particular, larger than the target body) and it’s relation to the delivery body is determined the use of a defined body (not shown) having the same size and shape as the intended delivery body being inserted into the patient’s vagina at step 204 and a scan taken of the treatment region by the scanner.
  • the processor may be configured to determine the relation of the target volume to the delivery or defined body.
  • the target volume 6 as illustrated in Figure 6 may conform to a high risk target volume, as is commonly known, and may be determined based on an inputted skip margin and other considerations inputted by a user. It will be appreciated that other parameters and definitions for the target volume for the trajectory paths may also be defined and selected by a user.
  • the processor 12 constructs a dense point cloud of a plurality of points from the target volume contours within the interior thereof in step 208.
  • a method for determining at least one set of trajectory paths 100 from a delivery region 102 is illustrated in Figure 6a and set out in more detail in Figure 14 as step 210 from Figure 13 and in further detail in Figure 14.
  • the delivery region 102 is meant to mean a region in space extending to the target volume 6 that may be an open or atmospheric space only as in the case for drone or other vehicle path determining, or a delivery body 40 for passing one or more needles 50 or delivery devices to the target volume as in the case for medical treatments.
  • the present method provides a means to optimize, design and build an individualized delivery body for a particular patient and their particular tumour or to apply the required treatment over their individual target volume 6.
  • the delivery region may be substantially cylindrical along an axis 45 of the delivery body.
  • the processor 12 first defines a first surface 104 around the delivery region 102 interesting at least a portion of the target volume 6 at step 212.
  • the first surface 102 may be substantially spherical around the delivery region or may be any other shape as desired by a user.
  • the surfaces may be expanded by any means, such as, by way of non-limiting example, non-uniform vector field, or by a constant or variable offset in a particular direction.
  • the processor 12 then defines plurality of points or locations 106 on the first surface 104 from above within the target volume 6. The points may be distributed across the surface in any desired pattern or selection criteria and may be arranged in ay manner, including, without limitation regularly, irregularly, randomly or any calculated set of points.
  • the processor 12 next selects a predetermined number of starting points 108 on the first surface 104 to define a starting centroids 110 in step 214.
  • the starting points 108 may be predetermined or randomly selected by the processor 12.
  • the processor then optimizes the location of each centoid110 as will be further set out below for that first shell.
  • the processor 12 is adapted to locate distribute the points 106 into a number of regions having equal areas corresponding to one of the predefined number of desired trajectories.
  • the processor 12 initially defines a plurality of points 106 across the first surface 104 within the target volume 6.
  • the processor then defines an initial set of centroids 110 at predefined, default or randomly generated positions.
  • the processor 12 determines which initial centroid 110 is closes to each point and groups those points having a common closest centroid (illustrated as groups 112, 114 and 116 in Figure 7b. Once all points 106 have been assigned to groups 112, 114 and 116, the processor then recalculates the new centroid 118, 120 and 122 for each group as illustrated in Figure 7c as shown in step 216 of Figure 14.
  • the processor then resets each of the points 106 and repeats the process shown in Figures 7a through 7c for the new centroids 118, 120 and 122, thereafter moving the centroids to their new calculated location until either a predetermined number of iterations are achieved or a specified percentage or value change in the centroid location is achieved between iterations indicating stable and consistent sized zones as illustrated in Figures 7d and 6c as determined at step 218.
  • Other mathematical analysis modelling may be optionally utilized for such uniformly spaced points on the surface, including, but not limited to, grid sampling, random sampling with density control, poisson disk sampling fibonacci lattice or golden ratio sampling, blue noise sampling or surface parameterization
  • the processor then radially extends an initial target path 124, 126 and 128 through each of the centroids in the first surface 104 and defines a subsequent surface 130.
  • the initial target 124, 126 and 128 paths are then used to define the new initial centroids on the subsequent shell and the process illustrated in Figure 7 and 6c and 6d repeated for each successive surface.
  • a best fit line 132 is applied between the corresponding centroids of each surface and that best fit line is then utilized for the next starting centroid. For each best fit or regression line 132 the process continues until no points are associated with that initial centroid within the target volume indicating that the best fit line has now exited the target volume.
  • the target paths 124, 126 and 128 are extended to intersect the next subsequent surface as illustrated in Figure 6h. Once all target paths 124, 126 and 128 have exited the target volume, the points may be thereafter removed or not considered for further analysis as shown in Figure 6i as shown in step 220 and the best fit through the points determined in step 222.
  • the first step utilized by the processor is to solve for the interstitial regions of the curved needles in the cylinder.
  • This step uses a modified Centroidal Voronoi Tesselation (CVT) algorithm referred here to as Regression-based Expanding Volume CVT (REVCVT) which attempts to evenly space needles inside the region of the HRCTV.
  • CVT Centroidal Voronoi Tesselation
  • REVCVT Regression-based Expanding Volume CVT
  • the cylinder normals are used to determine which voxels of the HRCTV are touching the surface of the cylinder.
  • CVT regions and centroids are generated on this resultant intersection surface.
  • Progressively larger cylinder shells are created and this same algorithm is applied to each shell until a shell of a certain diameter is reached that no longer intersects any part of the HRCTV.
  • the centroids of the first shell are used as seed points for the CVT on the next shell.
  • lines of best fit for matching centroids from the previous shells are generated and extrapolated to the next shell.
  • the processor may continue with each path until all paths exit the target volume. This algorithm continues until all the line segments have endpoints outside the HRCTV.
  • Each needle is extended into the cylinder an amount called the barrel length, which ensures the needle is exiting the cylinder in a straight trajectory.
  • Algorithm 1 Exemplary For Generating Interstital Needle Segment function REV CV T(Target, Cylinder, Normals, Npaths) i 0
  • CentroidNormals GetNormals(Centroids, Normals)
  • NewEndpoints Extrapolate(FinalPaths, StepSize)
  • Seeds Append(EP, Seeds) continue end if end for end while return N end function
  • the processor 12 may receive a specified maximum angle for each target path from the axis 45 of the delivery body 40 as illustrated generally in Figure 8a. It will be appreciated that the maximum angle may be inputted by a user as specified. It will further more be appreciated that such maximum angle is determined or dictated by the abilities of the delivery needle or other device expected to travel the delivery passage and trajectory path such as minimum bend radius. It will furthermore be appreciated that for with needles or the like, too great of an angle may cause binding or jamming of the needle within the path. In other embodiments, such binding will not be a consideration, however minimum turn radius may still be a limiting consideration.
  • each target path 124, 126 and 128 is extended into the delivery region by a predefined barrel length.
  • the barrel length is predefined as an input by a user as the distance within the delivery body 40 that the delivery passage 46 must extend in a straight line at the end portion 48 of the delivery passage so as to ensure stability of the delivered needle into the target volume 6 resulting in the final target paths as illustrated in Figure 8d.
  • a arc 150, 152 and 154 is appended to the interior end in step 224 of each target path as illustrated in Figure 9b the arc radius may be specified by a user as dependant upon the bend tolerances of the delivery apparatus to be passed through the delivery passages in step 226.
  • the processor checks for intersections, being actual intersects between delivery paths 160 or passages not having minimum wall thickness therebetween and varies the radius of one or more of the arcs to eliminate or reduce such intersections. It will be appreciated that methods for such intersection elimination are known.
  • additional straight delivery paths 170 may furthermore be extended through the delivery region and confirmed to be non-intersecting with the existing delivery passages according to known methods.
  • the remainder intracavitary section of the needle is constructed by appending a tangent semicircular segment with the minimum allowable radius to each segment, which directs the needle path toward the base of the cylinder. The remaining straight length of the needle to the base is then appended.
  • an intersection check is performed between each needle and between each needle and the walls of the cylinder. If there are any intersections, a python-implemented LBFGS-B optimization algorithm is used to vary the radius used to generate the IC section of each needle, in order to find a set of radii that results in no inter-needle or wall intersections.
  • the processor may form the resulting delivery region into a solid or 3D model thereby defining the delivery body 40 having the delivery passages 46 extending therethrough.
  • Such model may be provided to a user, transmitted across a network through the network output 24 or transmitted to a manufacturing device 28 to manufacture the delivery body for use in administering the desired treatment to a patient in step 230 of Figure 13.
  • DICOM files may be generated for each solution, ready for export across a network for use elsewhere.
  • Utilizing additive manufacturing devices to form the delivery body 40 based on the present methods will provide an individualized delivery device for medication and/or treatments to a patient that are not possible at the speeds of conventional planning and manufacturing techniques.
  • the use of the present methods for creating the individualized delivery body will also allow for an accuracy of placement of the desired treatments that is not currently possible using conventional hand guided needles.
  • the method was shown to generate high-quality clinical plans which are dosimetrically equivalent to manually created plans by clinical experts, in significantly less time.
  • the resultant needle trajectories generated using this technique require no adjustment in order to generate acceptable clinical plans. This allows for the creation of high-quality, robust, patient-specific templates in a short clinical time frame.
  • the resultant library of plans allows the clinician to evaluate multiple results at once, and select the best option for their patient, weighing coverage, toxicity, and interstitial needle trade-offs in step 228 of Figure 13.
  • the method is independent of any dose calculation method, meaning it can be implemented into any existing commercial treatment planning system workflow. Further investigations will extend this work to include in-the-loop dwell time optimization and Pareto-front-generation.
  • interstitial curved needle trajectories are shown as the optimization runs, which allows for a quick sanity check of the input parameters such as skip margin and step size as the optimization is running.
  • the user can stop the optimization at any time if there is a noticeable issue with the optimization as it is running. This means that the user doesn’t have to wait until the end of optimization in order to adjust a setting, which saves planning time.
  • it gives confidence to the user that their parameters are set appropriately and they can work on other tasks as the optimization completes.
  • FIG 11 images representing the delivery of radiation to a tumour utilizing the present method for generating an individualized delivery body 40 is illustrated as compared to conventional methods.
  • a conventional delivery and calculation system is utilized to delivery radiology to a patient having a tumour 180.
  • the tumour has a High-Risk Clinical Target Volume (HRCTV) 182 therearound being the desired delivery region.
  • HRCTV High-Risk Clinical Target Volume
  • the actual delivery of radiation through the one or more radiology delivery needles 186 extends to a 100% iodose region 184 which can be seen to not include the entire HRCTV 162.
  • a delivery body 40 designed using the present methods is also shown as being utilized on the same patient however, as can be seen the 100% iodose region 166 using the individualized delivery body results in a complete coverage of the HRCTV
  • the processor 12 may be adapted to generate a plurality of potential plans for selection by a user.
  • the user may input one or more parameters into the processor 12 for the processor to generate a plurality of sets of target path plans for a user to select from.
  • a user may input one or more parameters at the same time or at multiple times to cause the processor 12 to generate multiple or alternative sets of target path plans for selection by the user.
  • a total number of plans Np Nn * Ni is generated.
  • the algorithm sorts the plans based on the number of curved interstitial needles and feasibility. In particular, both feasible and unfeasible plans may be generated, as the processor may be unable to resolve inter-needle or wall intersections.
  • the sorting may be ranked, first preferring fewer intersections, then more interstitial curved needles or any other ranking criteria as desired by a user.
  • a sorted table showing which resultant plans have feasible needle trajectories (some results may have unresolved intersections or edge violations) is produced below in Table 1.
  • a further benefit utilizing the methodology and system disclosed herein is the ability to compare plan robustness and the utility of adding extra needles to a plan. For instance, 25 needle trajectory sets can be generated very quickly. If a user wishes to perform a trade-off analysis for different numbers of curved needles, the user can pick one plan for each number of curved needles (for example 3-7 curved needles) and import each plan back into the planning system. After optimizing the dwell times for each plan using the same planning goals and constraints, this allows the user explore dosimetric tradeoffs very quickly between plans with different numbers of curved needles. Thus one could evaluate the benefit of having different numbers of curved needles in the pre-planning phase and, optimizing the dose coverage while minimizing patient trauma from interstitial needles.
  • the user may input the following variables to the processor before beginning the above methodology: number of curved interstitial needles desired, or a range of curved interstitial needles, a skip margin (which may be inputted by any suitable means, including without limitation a single number, an exclusion zone represented by geometry tesselaton or a solid model) of depth closest to the cylinder which the optimization will ignore number of plans to generate for each number of curved interstitial needles, barrel length, minimum radius, intersection tolerance, and search step size.
  • number of curved interstitial needles desired or a range of curved interstitial needles
  • a skip margin which may be inputted by any suitable means, including without limitation a single number, an exclusion zone represented by geometry tesselaton or a solid model of depth closest to the cylinder which the optimization will ignore number of plans to generate for each number of curved interstitial needles, barrel length, minimum radius, intersection tolerance, and search step size.
  • the processor 12 then receives as inputs, the DICOM plan and structure files exported directly from the planning system, from which the processor extracts the High-Risk Clinical Target Volume (HRCTV) and a planar contour of a dummy cylinder or defined body, which is inserted into the patient for the patient’s preplan scan.
  • HRCTV High-Risk Clinical Target Volume
  • the system 10 may be configured to optimize the dwell time, defined as the duration of time each needle spends at a designated position along the trajectory path.
  • dwell time optimization may be performed by optimizing the dwell times according to a cost function with weights definable by the user.
  • the does is calculated and the three-dimensional dose distribution and dose volume histograms are reported to the user.
  • the does optimization is performed for each solution of needle trajectory is performed.
  • Such optimizations may be performed using a dwell time algorithm such as, by way of non-limited example, a limited memory Broyden-Fletcher-Goldfarb- Shanno (LBFGS-B) algorithm or other suitable optimization algorithm.
  • LPFGS-B limited memory Broyden-Fletcher-Goldfarb- Shanno
  • multiple dwell time optimization runs may be performed such that more than one dwell time optimization result is generated for each needle trajectory solution.
  • Such multiple results may be displayed in a plan comparison format such as a pareto front.
  • a search algorithm such as an LBFGS-B algorithm may be used to modify the needle trajectories by rotating and/or translating the needle trajectories and /or modifying the needle trajectory path points in loop until between needle trajectories using the search algorithm and performing the dwell time optimization until a stop criteria is met such as by way of non-limiting example number of iterations, time or a desired cost function value.

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Animal Behavior & Ethology (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Surgery (AREA)
  • Pathology (AREA)
  • Radiology & Medical Imaging (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Robotics (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)

Abstract

La présente invention concerne un procédé et un système permettant de déterminer au moins un ensemble de plusieurs trajectoires vers une zone cible, comprenant la réception par un processeur d'au moins une image représentant une zone cible, la fourniture d'une région d'introduction allongée s'étendant le long d'un axe et d'un rayon autour de la zone cible, l'utilisation du processeur, l'établissement d'au moins une trajectoire sans intersection à travers la région d'introduction et s'étendant jusqu'à la zone cible, et la production d'une représentation de la pluralité de trajectoires. L'établissement de la trajectoire est mis en œuvre à l'aide des premiers emplacements cibles, par le regroupement itératif des sites successifs à l'aide de tessellations de Voronoi centroidales en un nombre prédéterminé d'emplacements cibles successifs, et par la définition d'une ligne d'ajustement optimal entre chaque emplacement cible successif et le premier emplacement cible correspondant.
PCT/CA2023/051683 2022-12-15 2023-12-15 Système et procédé de planification de trajectoire basée sur une optimisation Ceased WO2024124360A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US202263432724P 2022-12-15 2022-12-15
US63/432,724 2022-12-15

Publications (1)

Publication Number Publication Date
WO2024124360A1 true WO2024124360A1 (fr) 2024-06-20

Family

ID=91484171

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CA2023/051683 Ceased WO2024124360A1 (fr) 2022-12-15 2023-12-15 Système et procédé de planification de trajectoire basée sur une optimisation

Country Status (1)

Country Link
WO (1) WO2024124360A1 (fr)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6505065B1 (en) * 1999-10-29 2003-01-07 Koninklijke Philips Electronics, N.V. Methods and apparatus for planning and executing minimally invasive procedures for in-vivo placement of objects

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6505065B1 (en) * 1999-10-29 2003-01-07 Koninklijke Philips Electronics, N.V. Methods and apparatus for planning and executing minimally invasive procedures for in-vivo placement of objects

Similar Documents

Publication Publication Date Title
JP7245352B2 (ja) 粒子を用いた回転式の放射線治療を提供する方法
US11376447B2 (en) Methods for user adaptive radiation therapy planning and systems using the same
US9044602B2 (en) Sequential optimizations for treatment planning
US6719683B2 (en) Radiotherapy treatment planning with multiple inverse planning results
US9827445B2 (en) Automatic creation and selection of dose prediction models for treatment plans
US7801270B2 (en) Treatment plan optimization method for radiation therapy
US20210339049A1 (en) Radiotherapy treatment plans using differentiable dose functions
CN107708808B (zh) 用于在放射治疗期间使用质量指标进行靶追踪的系统和方法
EP3157628B1 (fr) Système et produit programme d'ordinateur pour planification de traitement inverse
JP2004041725A (ja) 実時間放射線治療計画システム
EP3532163B1 (fr) Interface utilisateur graphique pour la planification d'un traitement itératif
WO2019090429A1 (fr) Planification de radiothérapie assortie d'une optimisation de multiples sous-ensembles cibles
CN105377368A (zh) 用于控制imrt递送复杂度的交互式基于剂量梯度的优化技术
US20060274925A1 (en) Generating a volume of interest using a dose isocontour
Yang Oncentra brachytherapy planning system
CN107998519A (zh) 一种用于imrt的分子动力学再优化算法
WO2024124360A1 (fr) Système et procédé de planification de trajectoire basée sur une optimisation
EP3875147B1 (fr) Planification de traitement
CN106794359B (zh) 剂量率调制的立体定向放射性手术
CN115006740A (zh) 治疗计划系统、放疗系统以及治疗计划制定方法
EP4664479A1 (fr) Procédé et appareil d'évaluation de plan de traitement par rayonnement
CN118698049A (zh) 一种基于强化学习的3d打印施源器针道优化方法
CN102458577A (zh) 放疗传输系统和放疗治疗计划
Poulin et al. Comparison Between HIPO and CVT as Well as DVHO and IPSA Algorithm for Catheter and Dose Optimization in HDR Brachytherapy

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23901864

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 23901864

Country of ref document: EP

Kind code of ref document: A1