Three-dimensional path planning method and computer program product for cardiac intervention surgery robot

Abstract

The application discloses a kind of cardiac intervention surgery robot three-dimensional path planning method and computer program product, according to real-time environment detection and image information in operation, real-time dynamic adjustment and update the three-dimensional model;According to the three-dimensional model and the real-time position of surgical robot, the optimal three-dimensional path of surgical robot is calculated by improved genetic algorithm;Fitness function F preset by genetic algorithm considers path length, obstacle avoidance ability, surgical safety, in different stages of operation, by adjusting the value of consideration factor weight to emphasize different optimization goals.The application calculates the optimal three-dimensional path of surgical robot by genetic algorithm considering path length, obstacle avoidance ability, surgical safety, realizes intelligent, accurate planning of surgical path, and has intraoperative real-time feedback and dynamic adjustment capability, improves the accuracy and safety of operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of surgical robot assistance, and more particularly, to a three-dimensional path planning method and computer program product for a cardiac intervention surgery robot. BACKGROUND

[0002] As an important branch of modern medical technology, cardiac valve intervention surgery aims to repair or replace heart valves through minimally invasive methods, reduce surgical trauma, and improve patient quality of life. The complex anatomical structure of the heart valve region includes multiple valves and the surrounding blood vessels, myocardium, pericardium, and other tissues. These structures are intertwined, forming a highly delicate and dynamic environment. Therefore, in order to improve surgical precision and reduce surgical risk, finding an optimal surgical path in the three-dimensional space of the heart valve is of great significance to cardiac valve intervention surgery.

[0003] Traditional path planning for cardiac valve intervention surgery mainly relies on the experience of doctors and medical image data. These methods can guide the selection of surgical paths to some extent, but they have obvious limitations: 1. They are highly dependent on the judgment of doctors, have certain subjectivity and uncertainty, and have high labor costs; 2. Due to the high safety requirements and the need to avoid postoperative complications of cardiac surgery, there is still a lack of appropriate algorithm models to implement more operational path planning; 3. The current path planning algorithm lacks intelligence and adaptability, and cannot be flexibly adjusted and optimized according to the specific situation of the patient and the surgical requirements.

[0004] Therefore, it is a technical problem to be solved to develop a more efficient and safe, highly adaptive three-dimensional path planning method for a cardiac intervention surgery robot. SUMMARY

[0005] In view of the above defects of the prior art, the present application provides a three-dimensional path planning method and computer program product for a cardiac intervention surgery robot, which calculates the optimal three-dimensional path of the surgical robot by considering the path length, obstacle avoidance ability, and surgical safety using a genetic algorithm, realizes intelligent and precise planning of the surgical path, and has the ability of real-time feedback and dynamic adjustment during surgery, thereby improving the accuracy and safety of the surgery.

[0006] To achieve the above-mentioned purpose, on the one hand, the present application provides a three-dimensional path planning method for a cardiac intervention surgery robot, characterized in that it comprises the following steps:

[0007] reading the three-dimensional structure medical image of the patient's heart and constructing a three-dimensional model of the patient's heart accordingly; dynamically adjusting and updating the three-dimensional model in real time according to real-time environmental detection and image information during surgery;

[0008] According to the three-dimensional model and the real-time position of the surgical robot, the optimal three-dimensional path of the surgical robot is calculated by an improved genetic algorithm; the fitness function F preset by the genetic algorithm considers path length F L , obstacle avoidance capability F o , and surgical safety F S , and is expressed as:

[0009] F=w L ·F L +w O ·F O +w S ·F S ;

[0010] wherein the weights w L +w O +w S =1,

[0011] In the formula, x i , y i represent the coordinates in the horizontal direction, z i is the height coordinate, d i is the distance from each point p i on the path to the nearest obstacle, and ε is a very small positive number; R(P) is the output of a risk assessment model based on machine learning;

[0012] In different stages of the surgery, different optimization objectives are emphasized by adjusting the values of w L , w O , and w S .

[0013] When designing the fitness function of the genetic algorithm in the method, multiple key factors of the surgical path are fully considered, such as path length, obstacle avoidance capability, and surgical safety, wherein the obstacle avoidance capability is evaluated by calculating the distance from each point on the path to the nearest obstacle, and the surgical safety is quantified in combination with a machine learning risk assessment model R(P). The weights of the factors are dynamically adjusted according to the surgical stage, ensuring that the planned path not only meets the physical feasibility but also takes into account the surgical safety. In this way, the path planning not only considers the physical feasibility, but also takes into account the safety and success rate of the surgery.

[0014] Further, since the respective evaluation criteria of F L , F O , and F S are quite different, in order to make each function have the same sensitivity to the total fitness function, the path length F L , the obstacle avoidance capability F o , and the surgical safety F SStandardization is performed, and the standardization formula is:

[0015]

[0016] where P i represents a path; F k (k=L, O, S) is standardized as The fitness function is:

[0017]

[0018] Further, constructing a three-dimensional model of the patient's heart includes the following sub-steps:

[0019] Step S1.1, sequentially using median filtering, bilateral filtering and Gaussian filtering to process the image, effectively removing noise in the three-dimensional structure medical image of the patient's heart, and improving the image quality;

[0020] Step S1.2, using a deep learning model to segment the image processed in step S1.1, identifying and marking the heart valve, blood vessels and tissue structure;

[0021] Step S1.3, three-dimensional reconstruction of the structures including the heart valve and blood vessels in the image, and fusion of the heart valve segmentation module, blood vessel feature module and robot segmentation module using registration technology, to obtain a three-dimensional model integrating multiple information.

[0022] Further, the deep model of step S1.2 is a U-Net-based deep model with an encoder-decoder structure, which can fuse low-level and high-level features, is suitable for medical image segmentation, and improves the segmentation accuracy. On the basis of U-Net, other CNN (Convolutional Neural Network) structures (such as VGG (Visual Geometry Group), ResNet (Residual Network), etc.) can also be combined to enhance the feature extraction capability.

[0023] Further, in step S1.3, Taubin smoothing and detail enhancement are used to finely process the three-dimensional model during three-dimensional reconstruction.

[0024] Further, calculating the optimal three-dimensional path of the surgical robot by genetic algorithm includes the following sub-steps:

[0025] Step S2.1, initialize the population and randomly generate a certain number of individuals (i.e. potential paths) as the initial population; determine the solution space, each individual is composed of a series of genes (path point coordinates), representing a path from the starting position to the target;

[0026] Step S2.2, calculate the fitness of each individual according to the preset fitness function;

[0027] Step S2.3, select individuals with high fitness as parents;

[0028] Step S2.4, randomly select two parent individuals from the selected individuals, perform cross operation to generate new child individuals, and form a new population;

[0029] Step S2.5, judge whether the new population reaches the termination condition (such as reaching the preset iteration number or finding a solution that meets the specific condition); if yes, end the algorithm and output the optimal three-dimensional path; if not, repeat steps 2.2-2.4.

[0030] Further, the roulette method is used to select parents in step S2.3: first, calculate the probability Pi of each individual being inherited into the next generation population; then calculate the cumulative probability qi of each individual, which represents the sum of the selection probabilities of the first i individuals; then generate a uniformly distributed pseudo-random number r in the interval [0, 1], which is used to simulate the pointer position in the roulette; traverse the cumulative probability array q to find the first k value that satisfies q[k-1]<r≤q[k], then select the kth individual as the parent; in order to avoid the loss of excellent individuals, an elite retention strategy is introduced in the selection process, and individuals with high fitness are directly retained to the next generation.

[0031] Further, in step S2.4, when selecting parents for crossing, a multi-point crossing method is used. This method can more flexibly combine the path characteristics of different parent individuals.

[0032] Further, the implementation steps of the risk assessment model based on machine learning to predict possible complications or risks under different paths are as follows:

[0033] Extract features related to the safety of the surgical path from the three-dimensional model; the features include the distance from each point on the path to key anatomical structures (such as blood vessel walls, heart valves, etc.), the curvature of the path, whether the path passes through high-risk areas (such as severely calcified valve areas), and the relative position of the path to known complication occurrence areas;

[0034] Collect historical surgery data, including successful and failed surgery cases, and the safety assessment results of the surgical path in each case as label data;

[0035] Use the extracted features and label data to train any supervised learning model including logistic regression, random forest, gradient boosting tree (GBDT), neural network, and optimize the parameters of the supervised learning model to obtain a risk assessment model based on machine learning, whose output is R(P).

[0036] In another aspect, the present application provides a computer program product, characterized in that when the computer program product is run on a computer or device, the computer or device is caused to perform the three-dimensional path planning method for a cardiac intervention surgery robot as described above.

[0037] Compared with the prior art, the present application has the following technical effects:

[0038] (1) In the present application, the fitness function is designed by fully considering multiple key factors of the surgical path, such as path length, obstacle avoidance ability, and surgical safety. By dynamically adjusting the weights of each factor according to the specific situation, the present application can flexibly respond to the needs of different surgical scenarios. In addition, the introduction of a risk assessment model based on machine learning further improves the safety and accuracy of path planning, ensuring that the surgical path is both efficient and safe.

[0039] (2) The present application uses a genetic algorithm for path planning of a cardiac intervention surgery robot. This method automatically searches and optimizes the path by simulating the natural evolution process, making the path selection more objective and accurate, significantly improving the calculation speed and efficiency of path search, effectively reducing human error, and greatly saving labor costs and time.

[0040] (3) In the present application, when constructing the three-dimensional model of the heart, information such as the heart valve segmentation module and the blood vessel feature module is fused. This comprehensive information fusion strategy makes the constructed three-dimensional model more detailed and accurate, and can more realistically reflect the actual structure of the patient's heart, providing a solid foundation for subsequent path planning.

[0041] (4) The present application closely integrates path planning with real-time environment detection and image information, achieving real-time feedback and dynamic adjustment during the surgery. During the surgery, the system can dynamically update the three-dimensional model according to the real-time detected environmental changes (such as blood vessel deformation, instrument deviation, etc.), and accordingly optimize the surgical path in real time. This real-time feedback and dynamic adjustment mechanism ensures the accuracy and safety of the surgical process, reducing the risk of surgery. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 The flowchart of the path planning method for a cardiac intervention surgery robot in an embodiment of the present application;

[0043] Figure 2 The flowchart of the genetic algorithm in an embodiment of the present application;

[0044] Figure 3 The schematic diagram of the solution space of the genetic algorithm in an embodiment of the present application. DETAILED DESCRIPTION

[0045] The application will be further described below in conjunction with the accompanying drawings and specific embodiments, but not as a limitation of the application.

[0046] In the following detailed description, numerous specific details are set forth in order to provide a more thorough understanding of the application. However, it will be apparent to one skilled in the art that the known algorithms and models (such as U-Net deep model, VGG model, ResNet model, Taubin smoothing algorithm, supervised learning model) do not show detailed processes to avoid obscuring the main idea of the application.

[0047] In addition, the sequence of actions, steps, etc. in the devices and methods shown in the claims, specification and drawings can be implemented in any order as long as there is no specific sequence limitation and the output of the previous process is not used in the subsequent process.

[0048] Embodiments

[0049] This embodiment first introduces a three-dimensional path planning method for a cardiac interventional surgery robot, referring to Figure 1 , including the following steps:

[0050] Step S1, read the three-dimensional medical image of the patient's heart, process and construct the three-dimensional model of the patient's heart.

[0051] More specifically, step S1 includes the following sub-steps:

[0052] Step S1.1, remove noise points in the image using filtering technology: first perform median filtering, for each pixel point in the image, take a neighborhood (such as 3x3, 5x5, etc.) centered on it, sort the pixel values in the neighborhood and take the median as the new value of the pixel, the purpose is to remove the salt and pepper noise in the image.

[0053] Then, bilateral filtering is performed on the image after median filtering, considering the spatial proximity and pixel value similarity of the pixels, and the two are weighted and averaged. The formula is:

[0054]

[0055] Where g(x,y) is the output pixel value, f(k,l) is the pixel value in the neighborhood S, and w(i,j,k,l) is the weight, which depends on the spatial proximity d(i,j,k,l) and the pixel value similarity r(i,j,k,l):

[0056]

[0057] Finally, Gaussian filtering is performed, and the image is convolved with a Gaussian function as the convolution kernel. The calculation formula is as follows:

[0058]

[0059] where G(x, y) is a Gaussian kernel, and σ is the standard deviation, which determines the smoothing degree of the filter.

[0060] Step S1.2, segment the image processed in step S1.1 using a deep learning model to identify and label the heart valve, blood vessels, and tissue structures. The model used is the U-Net model, which performs well in terms of small data volume requirements and fast training speed. In addition, this embodiment introduces VGG or ResNet convolutional blocks to enhance feature extraction capabilities by embedding these structure blocks in the encoder of U-Net.

[0061] Step S1.3, use volume rendering (Marching Cubes algorithm) to reconstruct the heart valve, blood vessels, and other structures of the image in three dimensions. Use image registration technology to align the images of different modalities (such as the heart valve segmentation module, the blood vessel feature module, and the robot segmentation module) to the same coordinate system, and then fuse the segmentation results of each modality. Apply the Taubin smoothing algorithm to smooth the three-dimensional model, and use edge sharpening, texture enhancement, and other techniques to improve the visual effect and accuracy of the three-dimensional model.

[0062] Step S2, according to the three-dimensional model and the real-time position of the surgical robot, calculate the optimal three-dimensional path of the surgical robot through the improved genetic algorithm. The flowchart of the genetic algorithm is shown in Figure 2 , which includes:

[0063] Step S2.1, initialize the population and randomly generate a certain number of individuals (i.e., potential paths) as the initial population. For example, the population size is Q, and the crossover probability is P c . Before solving the problem, it is important to encode the solution space and determine the gene encoding method to establish a one-to-one mapping relationship between the problem and the genetic algorithm. The actual encoding method usually uses real numbers or floating-point numbers within a certain range to represent the solution space, which can directly decode the solution space, reduce the amount of calculation, and improve the search efficiency of the algorithm. Therefore, this embodiment uses the actual encoding method as the gene encoding. The gene encoding method is shown in Figure 3 , where X n , Y n represent the coordinates in the horizontal direction, and Z n is the height. This set of genes represents a path from the starting position to the target. Determine the solution space, and each individual is composed of a series of genes (path point coordinates) representing a path from the starting position to the target.

[0064] Step S2.2, calculate the fitness of each individual according to the preset fitness function. The design of the fitness function is carried out by using a penalty function, and the core of three-dimensional path planning is to design a path that meets the constraint conditions, achieves the shortest and safest path. We consider the path length, obstacle avoidance ability and surgical safety as the factors for the fitness function.

[0065] The path length is:

[0066]

[0067] The obstacle avoidance ability is:

[0068]

[0069] Let each point p on the path be i The distance from the nearest obstacle is d i , and ε is a very small positive number to avoid division by zero. The reciprocal of the distance is used here because the larger the distance (i.e., the farther away from the obstacle), the smaller the obstacle avoidance ability evaluation value, which is contrary to the direction of maximizing fitness usually pursued in genetic algorithms, so the reciprocal is taken.

[0070] The surgical safety calculation formula is:

[0071]

[0072] Let the output of the machine learning-based risk assessment model be R(P), where P represents the path. Similarly, the reciprocal form is also used here to ensure the consistency of the fitness function.

[0073] The machine learning-based risk assessment model is implemented by the following steps:

[0074] Collect historical surgery data, including successful and failed surgery cases, and the safety assessment results of the surgery path in each case, which can be scored by experienced doctors or determined according to postoperative complications.

[0075] Secondly, select a supervised learning model, as it involves predicting unknown data based on known data, so it is suitable to use supervised learning models such as logistic regression, random forest, gradient boosting tree (GBDT), neural network, etc. The neural network model is used in this embodiment.

[0076] Finally, use data to train the machine learning model, adjust the model parameters, and optimize the model performance.

[0077] For a new surgical path P, its relevant features are extracted, including the distance from each point on the path to the heart valve, the curvature of the path, whether the path passes through a high-risk area (such as a severely calcified valve area), and the relative position of the path to the area where known complications occur, etc. These features are input into the trained machine learning model to obtain the risk assessment value R(P) of the path P.

[0078] Let the weights of path length, obstacle avoidance ability, and surgical safety be w L , w O , and w S , respectively, and these weights can be dynamically adjusted according to the surgical stage, then the comprehensive fitness function F can be represented as:

[0079] F = w L ·F L + w O ·F O + w S ·F S

[0080] wherein

[0081] w L + w O + w S = 1

[0082] In different stages of surgery, different optimization objectives can be emphasized by adjusting the values of w L , w O , and w S . In the early stage of surgery, when the robot needs to quickly reach the target area, the optimization of path length is particularly important, at this time, the value of w L should be increased to find the shortest path. However, when approaching the surgical target, if the path length conflicts with other factors (obstacle avoidance and safety), the weight of w L should be appropriately reduced to ensure the safety and obstacle avoidance ability of the path. During the surgery, if the robot encounters complex or unknown anatomical structures, the weight of w O should be increased to ensure that the robot path planning can bypass all potential obstacles and avoid unnecessary damage to the patient. In the critical stage of surgery, such as approaching important blood vessels or tissues, the weight of w S should be increased to ensure that the surgical instrument can safely pass through complex areas.

[0083] Since the respective standards of F L , F O , and F S are very different, in order to make each function have the same sensitivity to the total fitness function, we standardize the above three functions. The standardization formula is as follows:

[0084]

[0085] where P i represents the path.

[0086] After normalization, F k (k = L, O, S) becomes So the fitness function becomes:

[0087]

[0088] Step S2.3, select individuals with high fitness as parents. Exemplarily, the present embodiment uses a roulette method as the selection strategy, which has a certain randomness, which helps the algorithm to jump out of the local optimal solution. The basic idea is that the probability of each individual being selected is proportional to the size of its fitness, and its implementation process can be summarized in the following steps:

[0089] First, calculate the probability P i of each individual being inherited into the next generation population, which can be directly obtained by dividing the fitness of the individual by the total fitness of the population:

[0090]

[0091] Then calculate the cumulative probability q i of each individual, which represents the sum of the selection probabilities of the first i individuals:

[0092]

[0093] Next, generate a pseudo-random number r uniformly distributed in the interval [0, 1], which is used to simulate the pointer position in the roulette. Traverse the cumulative probability array q, find the first k value that satisfies q[k-1] < r ≤ q[k], then select the kth individual as the parent. This process is repeated until a sufficient number of parent individuals are selected.

[0094] In addition, in this process, in order to avoid the loss of excellent individuals, an elite preservation strategy can be introduced in the selection process, that is, the individual with the highest fitness is directly reserved to the next generation.

[0095] Step S2.4, randomly select two parent individuals from the selected individuals, perform crossover operation to generate new offspring individuals, and form a new population. Exemplarily, multiple-point crossover is used to cross the parents, which increases the diversity of the population and helps to search a wider solution space by selecting multiple crossover points on the three-dimensional path and exchanging part of the path between these points. The following is a detailed description of this process:

[0096] First, two parent individuals are selected from the current population according to the fitness function.

[0097] Then, a plurality of crossover points are randomly selected on the three-dimensional path. The number of crossover points can be determined according to the complexity of the path and the specific requirements of the problem. In the heart valve intervention surgery, 2-3 crossover points are appropriate. Ensure that the crossover points on the two parent paths are arranged in the same order for subsequent path exchange.

[0098] Next, the parent paths are divided into a plurality of sub-segments at each crossover point. For each parent, from the first crossover point to the last crossover point, the sub-segments are alternately selected for exchange. For example, the first sub-segment of parent A is exchanged with the first sub-segment of parent B, the second sub-segment of parent A is exchanged with the second sub-segment of parent B, and so on. Finally, the exchanged sub-segments are reconnected to form two new child paths.

[0099] Step S2.5, determine whether the new population reaches the termination condition, which determines when the algorithm stops iteration and outputs the optimal solution. If satisfied, end the algorithm and output the optimal three-dimensional path, if not, repeat steps 2.2-2.4. More specifically, first, set the termination condition, one is to reach the preset iteration number, before the algorithm starts, set a maximum iteration number (T), when the algorithm iteration number reaches this value, the algorithm stops;

[0100] Two is to judge whether a solution space meets a certain condition (such as the shortest path, the safest), if satisfied, this solution is considered good enough, the algorithm can be terminated in advance.

[0101] For the first termination condition, an iteration counter (t) can be maintained in the algorithm, and the value of the counter (t = t + 1) is updated at the end of each iteration.

[0102] For the second termination condition, the present application selects to use the path as the judgment condition, sets a suitable threshold L0, compares the path length L of each solution with L0 after each iteration, if L≤L0, the termination condition is satisfied.

[0103] Step S3, according to the actual situation in the operation process (such as blood vessel deformation, robot deviation, etc.), through real-time environment detection and image information, dynamically update the three-dimensional model, and then repeat step S2.

[0104] If the above cardiac intervention surgery robot three-dimensional path planning method is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Therefore, the technical solution essentially or the part that contributes to the prior art or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions to make an electronic device (which can be a personal computer, a server, or a network device) execute all or part of the steps of the method described in various embodiments of the application. The aforementioned storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program code storage media.

[0105] In summary, the present application provides a cardiac intervention surgery robot three-dimensional path planning method and computer program product, which dynamically adjusts and updates the three-dimensional model in real time according to real-time intraoperative environment detection and image information; according to the three-dimensional model and the real-time position of the surgical robot, the optimal three-dimensional path of the surgical robot is calculated through an improved genetic algorithm; the fitness function F preset by the genetic algorithm considers the path length F L , obstacle avoidance ability F o , and surgical safety F S . In different stages of the operation, the values of the weight of the considered factors are adjusted to emphasize different optimization objectives. The present application calculates the optimal three-dimensional path of the surgical robot through a genetic algorithm considering the path length, obstacle avoidance ability, and surgical safety, realizes intelligent and precise planning of the surgical path, and has real-time feedback and dynamic adjustment ability during the operation, thereby improving the accuracy and safety of the operation.

[0106] Those skilled in the art should understand that those skilled in the art can realize variations in combination with the prior art and the above embodiments, which are not described here. Such variations do not affect the essential content of the present application and are not described here.

[0107] The preferred embodiments of the present application have been described. It is to be understood that the application is not limited to the specific embodiments described above, and that devices and structures not specifically described but which are understood to be implemented in a conventional manner in the art are to be understood as such; any person skilled in the art, without departing from the technical solution of the present application, can make many possible changes and modifications to the technical solution of the present application disclosed above, or modify it into equivalent embodiments, which do not affect the essential content of the present application. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application, without departing from the technical solution of the present application, shall still fall within the scope of protection of the technical solution of the present application.

[0108] Those skilled in the art can appreciate that the units of the examples described in combination with the present embodiment, i.e. the algorithm steps, can be implemented in electronic hardware or in combination of computer software and electronic hardware. Whether the functions are implemented in hardware or software manner depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to implement the described functions for each specific application, but such implementation shall not be considered as beyond the scope of the present application.

Claims

1. A method for three-dimensional path planning for a cardiac intervention robot, characterized in that, The method comprises the following steps: reading a three-dimensional structure medical image of a patient's heart and constructing a three-dimensional model of the patient's heart based on the image; dynamically adjusting and updating the three-dimensional model in real time according to real-time environmental detection and image information during surgery; According to the three-dimensional model and the real-time position of the surgical robot, the optimal three-dimensional path of the surgical robot is calculated by an improved genetic algorithm; the fitness function F preset by the genetic algorithm considers path length F L , obstacle avoidance capability F o , and surgical safety F S , and is expressed as: ； wherein the weights , , , , where x i , y i denote the coordinates in the horizontal direction, z i is the height coordinate, d i is the distance of each point p i on the path to the nearest obstacle, is a small positive number; R(P) is the output of a machine learning based risk assessment model; At different stages of the surgery, different optimization objectives are emphasized by adjusting the values of w L , w O and w S ​ The implementation steps of the risk assessment model based on machine learning to predict possible complications or risks under different paths are as follows: extracting features related to the safety of the surgical path from the three-dimensional model; the features include the distance from each point on the path to the key anatomical structure, the curvature of the path, whether the path passes through a high-risk area, and the relative position of the path to the known complication occurrence area; collecting historical surgery data, including successful and failed surgery cases, and the safety evaluation results of the surgical path in each case as label data; training any supervised learning model including logistic regression, random forest, gradient boosting tree, and neural network using the extracted features and label data, and obtaining the risk assessment model based on machine learning after optimizing the parameters of the supervised learning model, with the output being R(P). 2.The cardiac intervention surgery robot three-dimensional path planning method according to claim 1, characterized in that, Path length F L Obstacle avoidance capability F o Surgical safety F S Standardization, the standardization formula is: where P i represents a path; normalized to where k = L, O, S, the fitness function is: .

3. The cardiac intervention surgery robot three-dimensional path planning method according to claim 1 or 2, characterized in that, The construction of the three-dimensional model of the patient's heart comprises the following sub-steps: Step S1.1, sequentially using median filtering, bilateral filtering and Gaussian filtering to process the image; Step S1.2, using a deep learning model to segment the image processed in step S1.1, identifying and marking the heart valve, blood vessels and tissue structure; Step S1.3, three-dimensional reconstruction of the structures including heart valve and blood vessels in the image, and fusion of the heart valve segmentation module, blood vessel feature module and robot segmentation module using registration technology to obtain a three-dimensional model integrating multiple information.

4. The cardiac intervention procedure robot three-dimensional path planning method of claim 3, wherein, The deep model of step S1.2 is a U-Net-based deep model with an encoder-decoder structure that can fuse low-level and high-level features. 5.The cardiac intervention surgery robot three-dimensional path planning method of claim 3, wherein, In step S1.3, Taubin smoothing and detail enhancement are used for fine processing of the three-dimensional model during three-dimensional reconstruction. 6.The cardiac intervention surgery robot three-dimensional path planning method according to claim 1 or 2, characterized in that, The optimal three-dimensional path of the surgical robot calculated by the genetic algorithm comprises the following sub-steps: Step S2.1, initialize the population and randomly generate a certain number of individuals as the initial population, with the individual being a potential path; determine the solution space, each individual is composed of a series of genes, and the gene is a path point coordinate representing a path from the starting position to the target; Step S2.2, calculate the fitness of each individual according to the preset fitness function; Step S2.3, select individuals with high fitness as parents; Step S2.4, randomly select two parent individuals from the selected individuals and perform crossover operation to generate new child individuals, forming a new population; Step S2.5, determine whether the new population meets the termination condition; if yes, end the algorithm and output the optimal three-dimensional path, otherwise repeat steps 2.2-2.

4.

7. The cardiac intervention procedure robot three-dimensional path planning method of claim 6, wherein, The roulette method is used to select the parent in the step S2.3: firstly, the probability Pi of each individual being inherited into the next generation population is calculated; then the cumulative probability qi of each individual is calculated, which represents the sum of the selection probabilities of the first i individuals; then a pseudo-random number r with uniform distribution in the interval [0, 1] is generated, which is used to simulate the pointer position in the roulette; the cumulative probability array q is traversed to find the first k value satisfying q[k-1] < r≤ q[k], and then the kth individual is selected as the parent; in the selection process, the elite reservation strategy is introduced, and the individual with high fitness is directly reserved to the next generation. 8.The cardiac intervention procedure robot three-dimensional path planning method of claim 6, wherein, In the step S2.4, the multi-point crossover is used when the parent is selected for crossover.

9. A computer program product, characterised in that, When the computer program product runs on the computer or the device, the computer or the device is caused to execute the three-dimensional path planning method of the cardiac intervention surgery robot as claimed in any one of claims 1 to 8.

Images