Spinal pedicle screw automatic planning method, system, device and storage medium

Abstract

The present application relates to a method, system, device and storage medium for automatic planning of spinal pedicle screws, wherein the method comprises: based on the three-dimensional model of the single vertebra of the target spine, segmenting to obtain a target structure model; according to the positional relationship between the target structure model and the screw, roughly planning the screw path direction to obtain a rough planning result; based on the rough planning result, optimizing the preset optimization target between the screw and the target structure model to obtain a fine planning result. Through the progressive multivariate conditional spinal pedicle screw automatic planning method in this application, the screw planning result is more in line with clinical standards, solving the problem that the single quantitative index in the related technology leads to poor reliability of the final planning result and difficulty in meeting clinical requirements.

Description

Technical Field

[0001] The present application relates to the field of medical technology, and in particular to a method, system, device and storage medium for automatic planning of spinal pedicle screws. Background Art

[0002] Spinal pedicle screw placement has become a common surgical procedure for stabilizing the spine and is widely used to treat vertebral fractures, scoliosis, herniated discs and other diseases. This procedure generally involves inserting screws into the pedicles of multiple vertebrae and then fixing the screw heads on the fusion hardware to form a good biomechanical structure to stabilize the spine. Due to the complex and diverse morphology of the vertebrae and the presence of important tissues and organs such as the spinal cord, nerve roots and blood vessels around the pedicles, traditional manual screw placement requires a lot of time and effort. Therefore, the application of automatic pedicle screw planning methods can greatly improve the efficiency of the operation.

[0003] Due to the complex vertebral morphology, the biomechanical structure is usually difficult to quantify, resulting in the inability to form a unified quantitative standard for spinal pedicle screw planning. Therefore, existing methods usually perform automatic screw planning by quantifying the screw placement effect, such as bone size, tightening strength, and various angle constraints. For example, based on the shape characteristics of the lumbar vertebrae, the pedicles are segmented, and then the optimization algorithm is used to maximize the amount of bone that the screw passes through to perform lumbar screw placement planning. However, these quantitative indicators are relatively simple and insufficient to cover the actual clinical situation, resulting in poor reliability of the final planning results and difficulty in meeting clinical requirements.

[0004] There is currently no effective solution to the problem that related technologies have a single quantitative indicator, which leads to poor reliability of the final planning results and difficulty in meeting clinical requirements. Summary of the invention

[0005] In this embodiment, a spinal pedicle screw automatic planning method, system, device and storage medium are provided to solve the problem in related technologies that the quantitative index is single, resulting in poor reliability of the final planning result and difficulty in meeting clinical requirements.

[0006] In a first aspect, a method for automatic planning of spinal pedicle screws is provided in this embodiment, comprising:

[0007] Based on the three-dimensional model of the single vertebra of the target spine, the target structure model is segmented and obtained;

[0008] According to the positional relationship between the target structure model and the screw, a rough planning is performed on the screw path direction to obtain a rough planning result;

[0009] Based on the rough planning result, the preset optimization target between the screw and the target structure model is optimized to obtain a fine planning result.

[0010] In some embodiments, segmenting the target structure model based on the three-dimensional model of a single vertebra of the subject's spine includes:

[0011] Based on the 3D model of the single vertebra of the target spine, 3D point cloud data is obtained and the target structure is labeled;

[0012] A point cloud segmentation model is obtained through training, and the three-dimensional point cloud data is segmented according to the annotations to obtain the target structure model.

[0013] In some of the embodiments, the rough planning of the screw path direction is performed according to the positional relationship between the target structure model and the screw to obtain the rough planning result, including:

[0014] Performing cylinder fitting based on the three-dimensional model of the pedicle in the target structure model, calculating the central axis and cross-sectional diameter of the fitted cylinder, and using them as the initial path direction and screw diameter for screw placement to obtain an initial planning result;

[0015] By limiting the angle between the upper end plate and the initial path direction of the screw in the target structure model, the initial path direction of the screw is roughly planned to obtain the rough planning result.

[0016] In some of the embodiments, the rough planning of the screw path direction is performed according to the positional relationship between the target structure model and the screw to obtain the rough planning result, including:

[0017] Based on the constraint condition of the angle between the upper end plate and the initial path direction of the screw in the target structure model, cylindrical fitting is performed on the pedicle three-dimensional model;

[0018] The central axis and cross-sectional diameter of the fitting cylinder are calculated and used as the path direction and screw diameter for screw insertion to obtain the rough planning result.

[0019] In some embodiments, the optimizing the preset optimization target between the screw and the target structure model based on the rough planning result to obtain the fine planning result includes:

[0020] Obtaining an adjustment range of the screw according to the rough planning result;

[0021] Taking the distance between the screw and the upper end plate in the target structural model and the adjustment range as optimization constraints, taking at least the degree of cohesion between the screw and the target structural model as the preset optimization target, and establishing a precise planning optimization model;

[0022] Based on the precise planning optimization model, a precise planning result is obtained.

[0023] In some embodiments, obtaining the adjustment range of the screw based on the rough planning result includes:

[0024] Obtaining a screw entry point and a screw tail point according to the screw direction in the rough planning result;

[0025] Establishing a first intersection region with the rear of the outer surface of the vertebral trunk in the target structure model according to the screw entry point, and using the first intersection region as a starting point coordinate set of the screw path;

[0026] Establishing a second intersection region in front of the outer surface of the vertebral trunk according to the screw tail point, and using the second intersection region as a set of end point coordinates of the screw path;

[0027] Based on the starting point coordinate set and the ending point coordinate set, an adjustment range of the screw path direction is obtained.

[0028] In some embodiments, taking the cohesion degree between the screw and the target structural model as the preset optimization target includes:

[0029] The screw path direction, screw radius, and screw length together indicate the degree of cohesion;

[0030] The screw path direction is a vector formed by the coordinates of the starting point and the coordinates of the ending point;

[0031] The screw radius is the minimum distance from the outer surface of the pedicle to the screw path direction;

[0032] The screw length is the Euclidean distance between the starting point coordinates and the ending point coordinates.

[0033] In a second aspect, in this embodiment, a spinal pedicle screw automatic planning system is provided, comprising: a segmentation module, a coarse planning module and a fine planning module;

[0034] The segmentation module is used to segment and obtain a target structure model based on the three-dimensional model of a single vertebra of the object spine;

[0035] The rough planning module is used to perform rough planning on the screw path direction according to the positional relationship between the target structure model and the screw to obtain a rough planning result;

[0036] The fine planning module is used to optimize the preset optimization target between the screw and the target structure model based on the rough planning result to obtain a fine planning result.

[0037] In a third aspect, a computer device is provided in this embodiment, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the automatic planning method for spinal pedicle screws described in the first aspect is implemented.

[0038] In a fourth aspect, in this embodiment, a storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the automatic planning method for spinal pedicle screws described in the first aspect is implemented.

[0039] Compared with the related art, the present embodiment provides a method, system, device and storage medium for automatic planning of pedicle screws of the spine. The target structure model is obtained by segmenting based on the three-dimensional model of a single vertebra of the target spine; the screw path direction is roughly planned by limiting the positional relationship between the target structure model and the screw to obtain a rough planning result; based on the rough planning result, the preset optimization target between the screw and the target structure model is optimized to obtain a fine planning result, which solves the problem in the related art that the quantitative index is single, resulting in poor reliability of the final planning result and difficulty in meeting clinical requirements, and realizes a progressive multivariate conditional automatic planning method for pedicle screws of the spine, so that the screw planning result is more in line with clinical standards.

[0040] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0042] Figure 1 The hardware structure block diagram of the terminal of the spinal pedicle screw automatic planning method in this embodiment;

[0043] Figure 2 is a flow chart of the automatic planning method for spinal pedicle screws in this embodiment;

[0044] Figure 3 Schematic diagram of the segmentation effect of the pedicle, the upper end plate and the lower end plate in this embodiment;

[0045] Figure 4 is a schematic diagram of the angle between the line connecting the centroids of the left pedicle and the right pedicle and the direction of the initial path of the screw in this embodiment;

[0046] Figure 5 Schematic diagram of the angle between the upper end plate and the initial path direction of the screw in this embodiment;

[0047] Figure 6 Schematic diagram of the rough planning result in this embodiment;

[0048] Figure 7-Figure 9 These are schematic diagrams of the precise planning results for the cervical spine, thoracic spine, and lumbar spine;

[0049] Fig.10 A flowchart of a spinal pedicle screw automatic planning method in a preferred embodiment;

[0050] Fig.11 It is a structural block diagram of a spinal pedicle screw automatic planning system in one embodiment.

[0051] In the figure: 102, processor; 104, memory; 106, transmission device; 108, input and output device; 10, segmentation module; 20, rough planning module; 30, fine planning module. DETAILED DESCRIPTION

[0052] In order to more clearly understand the purpose, technical solutions and advantages of the present application, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments.

[0053] Unless otherwise defined, the technical terms or scientific terms involved in this application shall have the general meaning understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the", "these" and the like in this application do not represent quantitative restrictions, and they can be singular or plural. The terms "include", "comprise", "have" and any variants thereof involved in this application are intended to cover non-exclusive inclusions; for example, a process, method and system, product or device comprising a series of steps or modules (units) is not limited to the listed steps or modules (units), but may include unlisted steps or modules (units), or may include other steps or modules (units) inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether directly or indirectly. The "multiple" involved in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, "A and / or B" can mean: A exists alone, A and B exist at the same time, and B exists alone. Usually, the character " / " indicates that the objects associated with each other are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.

[0054] The method embodiment provided in this embodiment can be executed in a terminal, a computer or a similar computing device. For example, running on a terminal, Figure 1 : is a hardware structure block diagram of a terminal of the spinal pedicle screw automatic planning method of this embodiment. Figure 1 As shown, the terminal may include one or more ( Figure 1 Only one is shown in the figure) processor 102 and memory 104 for storing data, wherein processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA. The above terminal may also include a transmission device 106 and an input and output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0055] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the automatic planning method for spinal pedicle screws in the present embodiment. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0056] The transmission device 106 is used to receive or send data via a network. The above network includes a wireless network provided by the communication provider of the terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (Radio Frequency, referred to as RF) module, which is used to communicate with the Internet wirelessly.

[0057] Spinal pedicle screw placement has become a common surgical procedure for stabilizing the spine and is widely used to treat vertebral fractures, scoliosis, herniated discs and other diseases. This procedure generally involves inserting screws into the pedicles of multiple vertebrae and then fixing the screw heads on the fusion hardware to form a good biomechanical structure to stabilize the spine. Due to the complex and diverse morphology of the vertebrae and the presence of important tissues and organs such as the spinal cord, nerve roots and blood vessels around the pedicles, traditional manual screw placement requires a lot of time and effort. Therefore, the application of automatic pedicle screw planning methods can greatly improve the efficiency of the operation.

[0058] The existing automatic planning methods for pedicle screws can be roughly divided into two types: one is to use the segmentation results of the pedicles to establish the quantitative indicators of screw planning and optimize the parameters to achieve the best; the other is to use machine learning methods to learn the rules based on the screw planning data marked by the doctor, and directly output the planning results. The optimal pedicle screw placement plan should be to form the best biomechanical structure of the spinal pedicle after the screw is placed, but this optimal biomechanical structure is usually difficult to quantify through a single quantitative indicator, and due to the doctor's planning preferences, the spinal pedicle screw planning cannot form a unified quantitative standard. In order to solve the above problems, a spinal pedicle screw automatic planning method with progressive multivariate quantitative conditions is provided in the following embodiments, so that the screw planning results are more in line with clinical standards.

[0059] In this embodiment, a method for automatic planning of spinal pedicle screws is provided. Figure 2 is a flow chart of the method of this embodiment, such as Figure 2 As shown, the method comprises the following steps:

[0060] Step S210: segmenting the target structure model based on the three-dimensional model of the single vertebra of the target spine.

[0061] Specifically, the single spinal vertebra of the object is usually converted into three-dimensional point cloud data, and then a point cloud segmentation model is obtained based on convolutional neural network training. The target structure is annotated, and then the three-dimensional point cloud data of the single spinal vertebra is segmented using the point cloud segmentation model to obtain a target structure model, wherein the target structure includes but is not limited to the left pedicle, the right pedicle, the upper end plate, and the lower end plate.

[0062] Step S220, performing rough planning on the screw path direction according to the positional relationship between the target structure model and the screw, and obtaining a rough planning result.

[0063] Specifically, the purpose of rough planning is to use some relatively abstract clinical rules to quickly obtain a rough planning result, narrow the scope for the next step of fine planning, and make the implementation of fine planning more stable and rapid.

[0064] In this step, the cylinder equation can be established by fitting the cylinder in the pedicle area to obtain the initial path direction of the screw, and then the positional relationship between the target structure model and the screw is restricted to perform rough planning on the screw path direction. The positional relationship between the target structure model and the screw can be the angle between the upper end plate and the initial path direction of the screw, or the angle between the line connecting the center of mass of the left pedicle and the right pedicle and the initial path direction of the screw. Further, the initial path direction of the screw can be translated in a certain direction by an angle, etc., according to the implementer's preference, so as to obtain a better rough planning result.

[0065] Step S230: Based on the rough planning result, the preset optimization target between the screw and the target structure model is optimized to obtain a fine planning result.

[0066] Specifically, clinically, it is generally required that the screws keep a minimum safe distance from the outer surface of the vertebra while being close to the upper end plate, and that the screw diameter and screw length are as large as possible under the constraints to form cohesion as much as possible, that is, the screw tip is close to the center of the vertebra. Due to the morphological characteristics of the vertebra, when the screw meets other clinical conditions, the length and cohesion of the screw are positively correlated. The above-mentioned preset optimization target at least includes the cohesion between the screw and the target structure model. When the cohesion between the screw and the target structure model is used as the preset optimization target, optimization parameters such as screw diameter, screw length and screw path direction are included. These three optimization parameters are often mutually constrained. The maximum diameter and maximum length of the screw are also limited by the direction of the screw path. Among them, the screw path direction can be determined by the starting point and the end point, and the screw length and screw diameter can also be calculated by the starting point, the end point and the target structure model obtained by segmentation.

[0067] Based on the rough planning results, the adjustment range of the starting and ending points of the screw path direction is obtained, and then combined with at least the degree of cohesion between the screw and the target structure model as the preset optimization goal, some clinical requirements can be converted into multiple quantitative indicators and constraints, and the screw path direction, screw diameter and screw length can be optimized at the same time to obtain better fine planning results.

[0068] The above steps obtain the target structure by segmenting from the three-dimensional model of a single vertebra of the spine, and then perform some rough rule corrections based on the target structure and the direction of the screw path to obtain a rough planning result. Finally, based on the rough planning result, further optimization is performed under fine planning conditions such as the safety distance between the screw and the outer surface of the vertebra, the cohesion of the screw, etc. to obtain a fine planning result. This embodiment provides an optimization strategy using progressive rules, defines more diversified and effective quantitative indicators for screw placement, and makes the screw planning results more in line with clinical standards, thereby solving the problem in the prior art that the quantitative indicators are single, resulting in poor reliability of the final planning results and difficulty in meeting clinical requirements.

[0069] Furthermore, the above-mentioned progressive optimization rules from coarse planning to fine planning have strong versatility. They are not only applicable to the pedicles of a single spinal vertebra in the above embodiments, but can also be applied to the automatic planning of pedicle screws of vertebrae such as the cervical, thoracic and lumbar vertebrae by adaptively adjusting the above-mentioned coarse rules and fine rules according to the specific vertebral structure.

[0070] In some embodiments, the segmentation of the target structure model based on the three-dimensional model of a single vertebra of the subject's spine includes:

[0071] Based on the three-dimensional model of a single vertebra of the object spine, three-dimensional point cloud data is obtained, and the target structure is labeled; a point cloud segmentation model is obtained through training, and the three-dimensional point cloud data is segmented according to the annotation to obtain a target structure model.

[0072] Specifically, the single vertebral model of the target spine is converted into three-dimensional point cloud data, and then the target structure is annotated. The point cloud segmentation model is obtained based on the convolutional neural network training by using the component segmentation method, and the three-dimensional point cloud data is input into the trained convolutional neural network model to obtain the segmented target structure accordingly. Among them, the target structure can be the upper end plate, the lower end plate, and the pedicle. Figure 3 Schematic diagram of the segmentation effect of the pedicle, the upper end plate and the lower end plate in this embodiment. Figure 3 As shown, Figure 3 Two views (left and right) of a single vertebral three-dimensional model of the spine are provided, as well as schematic diagrams of the upper end plate, lower end plate and pedicle segmented in the two views.

[0073] Preferably, a point cloud segmentation model can be obtained by training a PointNet++ neural network. PointNet++ is a hierarchical neural network that iteratively uses PointNet on nested partitions of an input point set, and then uses the distance in the metric space and the growth of the context scale to learn local features. In addition, since the density of point cloud data collected at different locations is different, multi-scale features can be adaptively combined.

[0074] In some of the embodiments, the process of performing rough planning on the screw path direction according to the positional relationship between the target structure model and the screw to obtain the rough planning result includes:

[0075] Based on the three-dimensional model of the pedicle in the target structure model, a cylinder is fitted, and the central axis and cross-sectional diameter of the fitted cylinder are calculated and used as the initial path direction and screw diameter for screw placement to obtain the initial planning result;

[0076] By limiting the angle between the upper end plate and the initial path direction of the screw in the target structure model, the initial path direction of the screw is roughly planned to obtain the rough planning result.

[0077] Specifically, firstly, a cylindrical fitting is performed on the pedicle area to establish a cylindrical equation. Specifically, based on the pedicle structure obtained by segmentation in the above embodiment, the maximum inscribed cylinder of the pedicle cross section is found by the RANSAC method, and then the central axis and cross-sectional circle diameter of the cylinder are calculated according to the established cylindrical equation. In this embodiment, the screw is approximately equivalent to a cylinder, the diameter of the screw is the cross-sectional diameter of the cylinder, and the initial path direction of the screw insertion is the direction of the central axis of the cylinder, and the initial planning result is obtained.

[0078] Due to the complexity and diversity of the vertebral morphology of the spine, if the screw is placed only along the central axis of the pedicle according to the initial planning results, sometimes the planning results will be poor, especially when the angle between the central axis of the pedicle and the upper endplate of the vertebra is too large, the screw placed along the central axis of the pedicle may break through the upper endplate bone cortex, so it is necessary to make a rough correction based on some rough rules, among which the rough rules include the angle between the upper endplate and the initial path direction of the screw, or Figure 4 The schematic diagram of the angle θ between the center of mass line of the left pedicle and the right pedicle and the initial path direction of the screw can further be the implementer's preference of translating the initial path direction of the screw in a certain direction by an angle, so there is no over-reliance on rules and parameters.

[0079] In this embodiment, Figure 5 is a schematic diagram of the angle β between the upper end plate and the initial path direction of the screw in this embodiment, as shown in Figure 5 As shown in the figure, the initial path direction is roughly corrected by limiting the angle between the upper end plate and the initial path direction of the screw. Specifically, the angle can be calculated by calculating the upper end plate fitting plane and the initial path direction. Among them, the point set formed by the upper end plate is obtained according to the segmentation result of the upper end plate, and a plane is fitted according to these point sets so that the point sets are distributed on the left and right of the plane. The least squares fitting method is used to minimize the sum of the distances from the point set to the plane, or a small threshold is determined by the RANSAC method, such as 1mm and 0.5mm, etc., to find a plane so that the number of points whose distances from the upper end plate point set to this plane do not exceed the set threshold is the largest, so as to obtain the upper end plate fitting plane.

[0080] Furthermore, the angle β of the aforementioned included angle ranges from 0 to 5 degrees, and can be specifically 1 degree, 2 degrees, 3 degrees or 4 degrees, etc., and is preferably 3 degrees.

[0081] In another embodiment, another process for coarsely planning the direction of a screw path is provided, comprising:

[0082] Based on the constraint condition of the angle between the upper end plate and the initial path direction of the screw in the target structure model, a cylindrical fitting was performed on the pedicle three-dimensional model; the central axis and cross-sectional circle diameter of the fitted cylinder were calculated and used as the path direction and screw diameter for screw insertion to obtain a rough planning result.

[0083] Specifically, the correction condition of the rough rule is added to the cylindrical fitting process of the pedicle three-dimensional model, so that the cylindrical fitting must be performed under the condition that the screw passes through the pedicle and does not break the rough rule. Therefore, based on the restriction condition of the angle between the upper end plate and the initial path direction of the screw, the schematic diagram of the angle is as follows Figure 5 As shown, the limiting condition is the angle range given in the above embodiment.

[0084] After obtaining the fitted cylinder, the central axis and cross-sectional diameter of the fitted cylinder are calculated and used as the path direction and screw diameter for screw insertion, respectively, which is the rough planning result.

[0085] In the above two embodiments, the screw path direction is roughly planned based on the angle restrictions in some rough rules, including first obtaining the initial path direction through cylinder fitting and then performing rough planning, or directly adding rough planning restrictions during cylinder fitting, so as to directly or indirectly obtain the path direction of the screw after rough planning. Furthermore, according to some relatively abstract clinical rules, a rough planning result can be quickly obtained, which can narrow the scope for the next step of fine planning and make fine planning more stable and faster.

[0086] Figure 6 is a schematic diagram of the rough planning result in this embodiment, such as Figure 6 As shown, after rough planning in the above two embodiments, the vertebral surface is reconstructed based on the three-dimensional model of the single vertebra to obtain a rough planning result, wherein the intersection of the screw path direction and the vertebral surface is the initial entry point and the screw tail point of the screw, and the entry point and the screw tail point constitute the screw path direction in the rough planning result.

[0087] In some embodiments, the above-mentioned optimization of the preset optimization target between the screw and the target structure model based on the rough planning result to obtain the fine planning result includes the following steps:

[0088] Step S310, obtaining the adjustment range of the screw according to the rough planning result.

[0089] Specifically, fine planning is to determine the adjustable range of the pedicle screw based on the result of rough planning, and further optimize the path direction and diameter of the screw under fine rules. The adjustment range of the screw is determined by the following steps:

[0090] Step S311, obtaining the screw entry point and screw tail point according to the screw direction in the rough planning result.

[0091] like Figure 6 The entry point and the tail point of the screw shown in the rough planning result can be regarded as the starting point and the end point of the screw path direction, thereby determining the screw path direction.

[0092] Step S312: A first intersection region with the rear of the outer surface of the vertebral trunk in the target structure model is established according to the screw entry point, and the first intersection region is used as a starting point coordinate set of the screw path.

[0093] Step S313, establishing a second intersection region with the front of the outer surface of the vertebral trunk according to the screw tail point, and using the second intersection region as the end point coordinate set of the screw path.

[0094] In the above steps S312 and S313, the starting point set and the ending point set of the screw path direction are first determined based on the rough planning results. Specifically, the nail entry point in the rough planning results is taken as the center, and a sphere is established with the first radius a. The first intersection area of ​​the sphere and the outer surface of the vertebral body trunk is used as the starting point coordinate set. In addition, the nail tail point in the rough planning result is taken as the center, and a sphere is established with the second radius b. The second intersection area in front of the outer surface of the vertebral body trunk is used as the ending point coordinate set. Among them, the first radius a and the second radius b control the range of fine planning adjustment. If a and b are larger, the adjustment range is larger. In addition, the larger the adjustment range, the larger the optimized search space, the greater the possibility of obtaining the best planning result, and the optimized search time will also be increased. Since the size of the vertebral body is limited, there is no need for an excessively large adjustment range. The value range of a and b is 0mm-5mm, preferably 3mm.

[0095] Step S314, based on the starting point coordinate set and the ending point coordinate set, obtain the adjustment range of the screw path direction.

[0096] According to the starting point coordinate set and the ending point coordinate set determined in the above steps, the adjustment range of the starting point and the ending point in the optimization is obtained accordingly. Since the screw path direction is determined by the starting point and the ending point, the adjustment range of the screw path direction is obtained.

[0097] Through the above steps S311-S314, the adjustment range of the starting point and the end point of the screw path direction is obtained from the rough planning result, and the adjustment range is used as a constraint condition in the fine planning optimization to search for the best planning result in the adjustment range.

[0098] Step S320, taking the distance between the screw and the upper end plate in the target structure model and the adjustment range as optimization constraints, and taking at least the degree of cohesion between the screw and the target structure model as a preset optimization target, to establish a fine planning optimization model.

[0099] Specifically, the adjustment range obtained in step S310 is used as an optimization constraint. In addition, since it is generally clinically required that the screws maintain a minimum safety distance while being close to the upper end plate, the bone density of the upper end plate is large, and the screws can obtain the maximum holding force when close to the upper end plate to improve the mechanical structure of the planning result, constraints are added to the precise planning optimization model to limit the distance between the screws and the upper end plate, and a corresponding safety distance must be reserved to prevent the screws from breaking through the vertebral surface.

[0100] Furthermore, the above-mentioned preset optimization target also includes screw length and screw radius. When the degree of cohesion between the screw and the target structure model, the screw length and the screw radius are used as the preset optimization targets, the screw diameter and the screw length are as large as possible under the constraint conditions to form cohesion as much as possible, that is, the screw tip is close to the center of the vertebral body. It includes optimization parameters such as screw diameter, screw length and screw path direction. These three optimization parameters are often mutually restricted. The maximum diameter and maximum length of the screw are also limited by the direction of the screw path. Among them, the screw length can be calculated by the starting point and the end point, and the screw diameter is obtained by the starting point and the end point to obtain the linear equation of the screw axis, and the minimum distance from the point set formed by the pedicle to the screw axis is calculated through the target structure model of pedicle segmentation.

[0101] Specifically, the screw path direction, screw radius and screw length together indicate the degree of cohesion; the screw path direction is a vector formed by the starting point coordinates and the ending point coordinates; the screw radius is the minimum distance from the outer surface of the pedicle to the screw path direction; and the screw length is the Euclidean distance between the starting point coordinates and the ending point coordinates.

[0102] According to the above optimization constraints and optimization objectives, the fine planning optimization model is as follows:

[0103]

[0104] stt 1 ≤d(x)-r(x)≤t 2 .

[0105] Among them, x is the vector formed by the coordinates of the starting point and the ending point in the screw path direction, that is, the optimization object; l(x) is the screw length, that is, the Euclidean distance between the starting point and the ending point; r(x) is the screw radius, that is, the minimum distance from the outer surface of the pedicle to the central axis of the screw; r 0 is the screw radius obtained by rough planning; ɑ and β are constants, α≥2 (ɑ is preferably 3, but not limited to 3) is used to control the weight of the screw cohesion degree, and the larger the α value is, the higher the cohesion requirement for the screw is; β≥1 (β is preferably 2, but not limited to 2) is used to control the weight of the screw radius, and the larger the value is, the higher the radius requirement for the screw is.

[0106] The above optimization model also includes two optimization constraints, where Ω represents the adjustment range of the starting point and the end point of the screw path direction, which is specifically obtained through step S310 in the above embodiment, and a constraint condition is also added to limit the distance between the screw and the upper end plate and leave a corresponding safety distance to prevent the screw from breaking through the vertebral surface, d(x) is the shortest distance from the screw centerline to the upper end plate, and t 1 and t 2 They are the upper and lower bounds of the restricted safety distance respectively.

[0107] By establishing a refined planning optimization model according to constraints and optimization objectives in this step, it is possible to further use multiple and effective quantitative indicators of screw placement under refined rules on the basis of the rough planning results, so that the screw planning results after refined planning are more in line with clinical standards.

[0108] Step S330, obtaining a refined planning result based on the refined planning optimization model.

[0109] Specifically, the optimization may be performed using, but not limited to, an optimization algorithm based on an evolutionary algorithm. The effective diameter and length of the screw are adjusted according to the specification limits of the actual screw, wherein the diameter of the screw is generally an integer multiple of 0.5 mm, and the length of the screw is an integer multiple of 5 mm. Figures 7 to 9 The diagrams in the middle are the schematic diagrams of the precise planning results of the cervical spine, thoracic spine, and lumbar spine.

[0110] Through the fine planning optimization model established in this embodiment, various clinical requirements can be defined as multiple and effective quantitative indicators based on the rough planning results, so as to construct a progressive optimization strategy, so that the screw planning results are more in line with clinical standards, without relying on a large number of experienced doctors to mark the planning data, and avoiding related uncertain risks.

[0111] The present embodiment is described and illustrated below through preferred embodiments.

[0112] Fig.10 FIG. 1 is a flow chart of the automatic planning method for spinal pedicle screws of the preferred embodiment. Fig.10 As shown, the method comprises the following steps:

[0113] Step S100, obtaining three-dimensional point cloud data based on the three-dimensional model of a single vertebra of the object spine, and annotating the target structure; obtaining a point cloud segmentation model through training, segmenting the three-dimensional point cloud data according to the annotation, and obtaining a target structure model.

[0114] The target structure model may be a left pedicle, a right pedicle, an upper end plate, a lower end plate, etc.

[0115] Step S101, based on the constraint condition of the angle between the upper end plate and the initial path direction of the screw, a cylindrical fitting is performed on the pedicle three-dimensional model.

[0116] Step S102, the central axis and cross-sectional diameter of the fitting cylinder are calculated and used as the path direction and screw diameter for screw insertion to obtain a rough planning result.

[0117] Step S103, obtaining the screw entry point and screw tail point according to the screw direction in the rough planning result.

[0118] Step S104, a first intersection area with the rear of the outer surface of the vertebral trunk in the target structure model is established according to the nail entry point, and the first intersection area is used as the starting point coordinate set of the screw path; a second intersection area with the front of the outer surface of the vertebral trunk is established according to the nail tail point, and the second intersection area is used as the ending point coordinate set of the screw path.

[0119] Step S105, obtaining the adjustment range of the screw path direction based on the starting point coordinate set and the ending point coordinate set.

[0120] Step S106, the screw path direction, screw radius and screw length are used together to indicate the degree of cohesion, the distance between the screw and the upper end plate in the target structure model and the adjustment range are used as optimization constraints, the cohesion degree between the screw and the target structure model, the screw length and the screw radius are used as optimization targets, and a precise planning optimization model is established.

[0121] Step S107, obtaining a refined planning result based on the refined planning optimization model.

[0122] It should be noted that the steps shown in the above process or the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0123] In the present embodiment, a spinal pedicle screw automatic planning system is also provided, and the system is used to implement the above-mentioned embodiment and preferred implementation mode, and the description that has been made will not be repeated. The terms "module", "unit", "subunit" etc. used below can implement the combination of software and / or hardware of the predetermined function. Although the system described in the following embodiments is preferably implemented with software, the implementation of hardware, or the combination of software and hardware is also possible and conceived.

[0124] Fig.11 : is a structural block diagram of the automatic planning system for spinal pedicle screws of this embodiment. Fig.11 As shown, the system includes: a segmentation module 10, a rough planning module 20 and a fine planning module 30;

[0125] The segmentation module 10 is used to segment and obtain a target structure model based on the three-dimensional model of a single vertebra of the object spine.

[0126] The rough planning module 20 is used to perform rough planning on the screw path direction according to the positional relationship between the target structure model and the screw to obtain a rough planning result.

[0127] The fine planning module 30 is used to optimize the preset optimization target between the screw and the target structure model based on the rough planning result to obtain the fine planning result.

[0128] Through the system provided in this embodiment, an optimization strategy using progressive rules is provided, which defines more diversified and effective quantitative indicators for screw placement, so that the screw planning results are more in line with clinical standards, thereby solving the problem in the prior art that the quantitative indicators are single, resulting in poor reliability of the final planning results and difficulty in meeting clinical requirements.

[0129] Furthermore, the above-mentioned progressive optimization rules from coarse planning to fine planning have strong versatility. They are not only applicable to the pedicles of a single spinal vertebra in the above embodiments, but can also be applied to the automatic planning of pedicle screws of vertebrae such as the cervical, thoracic and lumbar vertebrae by adaptively adjusting the above-mentioned coarse rules and fine rules according to the specific vertebral structure.

[0130] It should be noted that the above modules can be functional modules or program modules, and can be implemented by software or hardware. For modules implemented by hardware, the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0131] In this embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0132] Optionally, the computer device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0133] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementation modes, and will not be repeated in this embodiment.

[0134] In addition, in combination with the automatic planning method for spinal pedicle screws provided in the above embodiments, a storage medium can also be provided in this embodiment to implement the method. The storage medium stores a computer program; when the computer program is executed by a processor, any automatic planning method for spinal pedicle screws in the above embodiments is implemented.

[0135] It should be understood that the specific embodiments described herein are only used to explain the application, rather than to limit it. Based on the embodiments provided in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the protection scope of this application.

[0136] Obviously, the drawings are only some examples or embodiments of the present application. For ordinary technicians in the field, the present application can also be applied to other similar situations based on these drawings without creative work. In addition, it is understandable that although the work done in this development process may be complicated and lengthy, for ordinary technicians in the field, certain changes in design, manufacturing or production based on the technical content disclosed in this application are only conventional technical means and should not be regarded as insufficient content disclosed in this application.

[0137] The term "embodiment" in this application refers to a specific feature, structure or characteristic described in conjunction with the embodiment that can be included in at least one embodiment of the present application. The appearance of this phrase in various locations in the specification does not necessarily mean the same embodiment, nor does it mean that it is mutually exclusive with other embodiments and is independent or optional. It is clearly or implicitly understood by those of ordinary skill in the art that the embodiments described in this application can be combined with other embodiments without conflict.

[0138] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of patent protection. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the scope of protection of the present application. Therefore, the scope of protection of the present application shall be subject to the attached claims.

Claims

1. A method for automatic planning of spinal pedicle screws, characterized in that: include: Based on the three-dimensional model of the single vertebra of the target spine, the target structure model is segmented and obtained; According to the positional relationship between the target structure model and the screw, a rough planning is performed on the screw path direction to obtain a rough planning result; Based on the rough planning result, the preset optimization target between the screw and the target structure model is optimized to obtain a fine planning result; These include: Obtaining a screw entry point and a screw tail point according to the screw direction in the rough planning result; Establishing a first intersection region with the rear of the outer surface of the vertebral trunk in the target structure model according to the screw entry point, and using the first intersection region as a starting point coordinate set of the screw path; Establishing a second intersection region in front of the outer surface of the vertebral trunk according to the screw tail point, and using the second intersection region as a set of end point coordinates of the screw path; Based on the starting point coordinate set and the ending point coordinate set, obtaining an adjustment range of the screw path direction; Taking the distance between the screw and the upper end plate in the target structural model and the adjustment range as optimization constraints, taking at least the degree of cohesion between the screw and the target structural model as the preset optimization target, and establishing a precise planning optimization model; Based on the precise planning optimization model, a precise planning result is obtained.

2. The method for automatic planning of spinal pedicle screws according to claim 1, characterized in that: The method of segmenting a target structure model based on a three-dimensional model of a single vertebra of the target spine includes: Based on the 3D model of the single vertebra of the target spine, 3D point cloud data is obtained and the target structure is labeled; A point cloud segmentation model is obtained through training, and the three-dimensional point cloud data is segmented according to the annotations to obtain the target structure model.

3. The method for automatic planning of spinal pedicle screws according to claim 1, characterized in that: The step of performing rough planning on the screw path direction according to the positional relationship between the target structure model and the screw to obtain a rough planning result includes: Performing cylinder fitting based on the three-dimensional model of the pedicle in the target structure model, calculating the central axis and cross-sectional diameter of the fitted cylinder, and using them as the initial path direction and screw diameter for screw placement to obtain an initial planning result; By limiting the angle between the upper end plate and the initial path direction of the screw in the target structure model, the initial path direction of the screw is roughly planned to obtain the rough planning result.

4. The method for automatic planning of spinal pedicle screws according to claim 1, characterized in that: The step of performing rough planning on the screw path direction according to the positional relationship between the target structure model and the screw to obtain a rough planning result includes: Based on the constraint condition of the angle between the upper end plate and the initial path direction of the screw in the target structure model, cylindrical fitting is performed on the pedicle three-dimensional model; The central axis and cross-sectional diameter of the fitting cylinder are calculated and used as the path direction and screw diameter for screw insertion to obtain the rough planning result.

5. The method for automatic planning of spinal pedicle screws according to claim 1, characterized in that: The method of taking the cohesion degree between the screw and the target structural model as the preset optimization target includes: The screw path direction, screw radius, and screw length together indicate the degree of cohesion; The screw path direction is a vector formed by the coordinates of the starting point and the coordinates of the ending point; The screw radius is the minimum distance from the outer surface of the pedicle to the screw path direction; The screw length is the Euclidean distance between the starting point coordinates and the ending point coordinates.

6. A spinal pedicle screw automatic planning system, characterized in that: include: Segmentation module, rough planning module and fine planning module; The segmentation module is used to segment and obtain a target structure model based on the three-dimensional model of a single vertebra of the object spine; The rough planning module is used to perform rough planning on the screw path direction according to the positional relationship between the target structure model and the screw to obtain a rough planning result; The fine planning module is used to optimize the preset optimization target between the screw and the target structure model based on the rough planning result to obtain a fine planning result; These include: Obtaining a screw entry point and a screw tail point according to the screw direction in the rough planning result; Establishing a first intersection region with the rear of the outer surface of the vertebral trunk in the target structure model according to the screw entry point, and using the first intersection region as a starting point coordinate set of the screw path; Establishing a second intersection region in front of the outer surface of the vertebral trunk according to the screw tail point, and using the second intersection region as a set of end point coordinates of the screw path; Based on the starting point coordinate set and the ending point coordinate set, obtaining an adjustment range of the screw path direction; Taking the distance between the screw and the upper end plate in the target structural model and the adjustment range as optimization constraints, taking at least the degree of cohesion between the screw and the target structural model as the preset optimization target, and establishing a precise planning optimization model; Based on the precise planning optimization model, a precise planning result is obtained.

7. A computer device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to execute the automatic spinal pedicle screw planning method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the automatic spinal pedicle screw planning method according to any one of claims 1 to 5 are implemented.

Images