Minimally invasive surgery planning method and system based on image navigation

Abstract

The invention discloses a minimally invasive surgery planning method and system based on image navigation, and belongs to the technical field of medical information. The method specifically comprises the steps that a three-dimensional model of a patient is obtained through CT scanning; performing structure analysis on the three-dimensional model, performing grid division on the organ region, calculating gray value difference among grids, identifying a lesion grid, calculating the shortest distance from the three-dimensional distance from any in-vitro candidate point to the dangerous region to the geometric center of the lesion region according to the three-dimensional coordinates of the skin region and the lesion grid, and calculating the three-dimensional distance from any in-vitro candidate point to the dangerous region; and calculating an optimal score of any in-vitro candidate point based on the three-dimensional Euclidean distance, selecting an optimal puncture point, generating an in-vivo and in-vitro candidate point set through orthogonal projection and vertical extension, repeating the process until any puncture point reaches the lesion area, connecting all puncture points to obtain a final puncture path, and completing the in-vivo and in-vitro candidate point set. The optimal ablation path is automatically obtained through repeated iterative optimization, and the liver tumor ablation efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of medical information technology, in particular to a minimally invasive surgery planning method and system based on image navigation. BACKGROUND

[0002] Liver tumor thermal ablation therapy (including radiofrequency ablation and microwave ablation) has become the third radical method after surgical resection and liver transplantation. It is under the guidance of medical imaging to insert an ablation needle (radiofrequency electrode / microwave antenna) into the tumor as a heat source to heat the tumor tissue to a high temperature, causing the tumor to coagulative necrosis in situ.

[0003] In the treatment of liver cancer, radiofrequency ablation of liver tumor is commonly used to ablate the tumor target area. In order to improve the accuracy of ablation and ensure the treatment efficiency, the tumor target area of the patient needs to be positioned and analyzed, so as to design the puncture path in advance. At present, the puncture path planning is based on the ablation frequency, needle entry point and puncture depth determined by the treating doctor according to the patient's examination image and personal experience, which is easy to cause the ablation range to be too large, causing irreversible damage to the patient's body. SUMMARY

[0004] The purpose of the present application is to provide a minimally invasive surgery planning method and system based on image navigation, which solves the following technical problems: In the treatment of liver cancer, radiofrequency ablation of liver tumor is commonly used to ablate the tumor target area. In order to improve the accuracy of ablation and ensure the treatment efficiency, the tumor target area of the patient needs to be positioned and analyzed, so as to design the puncture path in advance. At present, the puncture path planning is based on the ablation frequency, needle entry point and puncture depth determined by the treating doctor according to the patient's examination image and personal experience, which is easy to cause the ablation range to be too large, causing irreversible damage to the patient's body.

[0005] The purpose of the present application can be achieved by the following technical solutions: A minimally invasive surgery planning method based on image navigation, comprising the following steps: S1, continuously scanning the patient's scanning area by dynamic CT technology to obtain a plurality of detection images, and constructing a three-dimensional model according to the detection images; S2, performing structural analysis on the three-dimensional model based on a medical image segmentation algorithm to obtain skin, organs, blood vessels and rib regions, marking the blood vessel and rib regions as dangerous regions, performing grid division on the organ region, and if the gray value of any grid region is within a preset gray threshold range, marking the grid as a lesion grid, and marking the region composed of all lesion grids as a lesion region; S3, constructing an extracorporeal candidate point set according to a skin region, calculating a shortest distance DF1 of each extracorporeal candidate point to a dangerous region by using an Euclidean distance transformation algorithm; analyzing a vertex coordinate of a lesion region and calculating a geometric center coordinate of the lesion region by using a centroid method to obtain a shortest distance DF2 of each extracorporeal candidate point to the geometric center; calculating a preferred score of each extracorporeal candidate point according to DF1 and DF2, and selecting an extracorporeal candidate point with the highest preferred score as an initial puncture point; S4, based on three-dimensional coordinates of any extracorporeal candidate point set, performing orthogonal projection on any extracorporeal candidate point to obtain a corresponding body surface projection point and extending a preset vertical interval distance d in a direction of a normal vector of the projection point into the body to generate an intracorporeal candidate point set having a spatial corresponding relationship with the extracorporeal candidate point, repeating the above process for any intracorporeal candidate point to obtain a second puncture point; repeating the above process until a first puncture point reaches a lesion region, and connecting all puncture points to obtain a final puncture path.

[0006] As a further scheme of the present application, in the S1, the specific construction process of the three-dimensional model is: The detection image is registered by using a rigid registration algorithm based on feature point matching, and the registered detection image is stacked and reconstructed in a preset structure order, isotropic interpolation processing is performed to form a three-dimensional volume data set in a unified coordinate system, and a three-dimensional view with depth information is generated by simulating an optical penetration path by using a ray casting method, and an isosurface of a specific gray value is extracted by simultaneously combining threshold segmentation and a Marching Cubes algorithm, so that the three-dimensional model is constructed.

[0007] As a further scheme of the present application, in the S3, the calculation process of the shortest distance is: A spatial acceleration grid is constructed for any dangerous region to obtain all face patches of the dangerous region, the shortest distance of the extracorporeal candidate point to each face patch is calculated, the minimum value of all face patch distances is selected as a to-be-determined distance value, the to-be-determined distance values corresponding to all dangerous regions are obtained, and the minimum value in the to-be-determined distance values is selected as the shortest distance.

[0008] As a further scheme of the present application, in the S3, the specific calculation process of the preferred score is: ; Wherein, a and b are preset unit coefficients, DF1i is the shortest distance of the i th extracorporeal candidate point to the dangerous region, MaxDF1 is the maximum value of the shortest distances of all extracorporeal candidate points in the skin region to the dangerous region, DF2i is the shortest distance of the i th extracorporeal candidate point to the geometric center, and MaxDF2 is the maximum value of the shortest distances of all extracorporeal candidate points in the skin region to the geometric center.

[0009] As a further scheme of the present application, in the S2, the setting process of the preset gray threshold is: Obtaining CT image samples of several liver tumors, manually labeling the tumor regions on the CT image samples and obtaining the gray values of the tumor regions, obtaining the maximum and minimum values of all gray values, and obtaining a gray threshold range according to the maximum and minimum values.

[0010] As a further scheme of the present application: in the S3, if the preferred scores of two or more extracorporeal candidate points are equal and are both maximum values, then both are marked as initial puncture points, and corresponding puncture paths are generated respectively.

[0011] As a further scheme of the present application: further comprising connecting the initial puncture point and the second puncture point to obtain a path direction vector v, calculating an included angle θ between the path direction vector v and a skin region normal vector n, if the included angle θ is greater than or equal to a preset included angle threshold, then the second puncture point is removed, and a second puncture point is selected again according to the preferred score and a corresponding puncture path is generated.

[0012] As a further scheme of the present application: in the S4, further comprising when there are several puncture paths, obtaining the lengths of the puncture paths, and selecting the puncture path with the shortest length as the final puncture path.

[0013] A minimally invasive surgery planning system based on image navigation, comprising: A data acquisition module for continuously scanning a patient scanning area by dynamic CT technology to obtain a plurality of detection images, and constructing a three-dimensional model according to the detection images; A structure analysis module for performing structural analysis on the three-dimensional model based on a medical image segmentation algorithm to obtain skin, organ, blood vessel and rib regions, marking the blood vessel and rib regions as dangerous regions, performing grid division on the organ region, and if the gray value of any grid region is within a preset gray threshold range, marking the grid as a lesion grid, and marking the region composed of all lesion grids as a lesion region; An initial determination module for constructing an extracorporeal candidate point set according to the skin region, calculating the shortest distance DF1 from each extracorporeal candidate point to the dangerous region using an Euclidean distance transformation algorithm, analyzing the vertex coordinates of the lesion region and calculating the geometric center coordinates of the lesion region using a centroid method to obtain the shortest distance DF2 from each extracorporeal candidate point to the geometric center, calculating the preferred score of each extracorporeal candidate point according to DF1 and DF2, and selecting the extracorporeal candidate point with the highest preferred score as the initial puncture point; The path planning module is used for implementing orthogonal projection on any extracorporeal candidate point based on three-dimensional coordinates of any extracorporeal candidate point set to obtain a corresponding body surface projection point and extending a preset vertical interval distance d in the direction of the normal vector of the projection point to generate an intracorporeal candidate point set having a spatial corresponding relationship with the extracorporeal candidate point, repeating the above process for any intracorporeal candidate point to obtain a second puncture point; and repeating the above process until a first puncture point reaches a lesion area, and then connecting all the puncture points to obtain a final puncture path.

[0014] The beneficial effects of the present application are: The present application first generates a three-dimensional model in the patient's body by dynamic CT scanning, analyzes the skin, organs, blood vessels and ribs and other regions by medical image segmentation technology, calibrates the dangerous area to avoid damage to the key tissues during the operation, divides the organ area into grids and analyzes the gray scale, automatically identifies the lesion area combined with the preset threshold, reduces the subjective error of manual interpretation, calculates the shortest distance from the extracorporeal candidate point to the dangerous area and the center of the lesion, which can be understood as the shortest distance from the extracorporeal candidate point to the dangerous area measuring the spatial distance between the extracorporeal candidate point and the dangerous area, avoiding key structures such as blood vessels and ribs, and the smaller the spatial distance, the greater the actual risk; the shortest distance from the extracorporeal candidate point to the center of the lesion reflects the proximity of the extracorporeal candidate point to the center of the lesion, and the smaller the spatial distance, the closer the puncture point to the target area, the smaller the distance means the higher the positioning accuracy, which is more conducive to accurately reaching the core of the lesion, and the optimal score is calculated according to the spatial distance, the best initial puncture point is dynamically selected based on the optimal score, and the intracorporeal candidate point is generated by orthogonal projection and normal vector extension, and the puncture path is gradually iterated to ensure that the path avoids the dangerous area and directly reaches the lesion target. The present application automatically obtains the best ablation path through repeated iteration and optimization, and improves the ablation efficiency of liver tumors. BRIEF DESCRIPTION OF DRAWINGS

[0015] The present application will be further described below with reference to the accompanying drawings.

[0016] Figure 1 It is a minimally invasive surgery planning method flowchart based on image navigation. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0018] Please refer to Figure 1 The present application is a minimally invasive surgery planning method based on image navigation, which comprises the following steps: S1, a plurality of detection images are obtained by continuously scanning a patient scanning area through a dynamic CT technique, and a three-dimensional model is constructed according to the detection images; S2, structure analysis is performed on the three-dimensional model based on a medical image segmentation algorithm, a skin region, an organ region, a blood vessel region and a rib region are obtained, the blood vessel region and the rib region are marked as a dangerous region, the organ region is divided into grids, if the gray value of any grid region is within a preset gray threshold range, the grid is marked as a lesion grid, and a region composed of all the lesion grids is marked as a lesion region; S3, an extracorporeal candidate point set is constructed according to the skin region, the shortest distance DF1 of each extracorporeal candidate point to the dangerous region is calculated using an Euclidean distance transformation algorithm, the vertex coordinates of the lesion region are analyzed and the geometric center coordinates of the lesion region are calculated using a centroid method, and the shortest distance DF2 of each extracorporeal candidate point to the geometric center is obtained, the preferred score of each extracorporeal candidate point is calculated according to DF1 and DF2, and the extracorporeal candidate point with the highest preferred score is selected as an initial puncture point; S4, based on the three-dimensional coordinates of any extracorporeal candidate point set, orthogonal projection is performed on any extracorporeal candidate point to obtain a corresponding body surface projection point, and the body surface projection point is extended into the body along the normal vector direction of the projection point by a preset vertical interval distance d, thereby generating an intracorporeal candidate point set having a spatial corresponding relationship with the extracorporeal candidate point, the above process is repeated for any intracorporeal candidate point, and a second puncture point is obtained, and the above process is repeated until the first puncture point reaches the lesion region, then all the puncture points are connected, and a final puncture path is obtained.

[0019] The application first generates a three-dimensional model in the patient's body by dynamic CT scanning, analyzes the skin, organ, blood vessel and rib regions through medical image segmentation technology, marks the dangerous region to avoid damaging key tissues during the operation, divides the organ region into grids and analyzes the gray value, automatically identifies the lesion region in combination with the preset threshold, reduces the subjective error of manual interpretation, calculates the shortest distance of the extracorporeal candidate point to the dangerous region and the lesion center, which can be understood as the shortest distance of the extracorporeal candidate point to the dangerous region measuring the spatial distance between the extracorporeal candidate point and the dangerous region, avoiding key structures such as blood vessels and ribs, the smaller the spatial distance, the greater the actual risk; the shortest distance of the extracorporeal candidate point to the lesion center reflects the proximity of the extracorporeal candidate point to the lesion center, the smaller the spatial distance, the higher the positioning accuracy, which is beneficial to accurately reaching the lesion core, and the preferred score is calculated according to the spatial distance, the best initial puncture point is dynamically selected based on the preferred score, the intracorporeal candidate point is generated by orthogonal projection and normal vector extension, and the puncture path is gradually iterated to ensure that the path avoids the dangerous region and directly reaches the lesion target, the application automatically obtains the best ablation path through repeated iteration and optimization, and the liver tumor ablation efficiency is improved.

[0020] As a preferred embodiment of the present application, in the S1, the specific construction process of the three-dimensional model is: The rigid registration algorithm based on feature point matching is used for detecting image registration, and the registered detection image is stacked and reconstructed according to the preset structure sequence, and a three-dimensional body data set of a unified coordinate system is formed through isotropic interpolation processing, and a three-dimensional view with depth information is generated by simulating the optical penetration path through the ray casting method, and the isosurface of a specific gray value is extracted through the threshold segmentation and Marching Cubes algorithm, thereby constructing a three-dimensional model.

[0021] The rigid registration algorithm based on feature point matching is used for detecting image registration, and the registered detection image is stacked and reconstructed according to the preset structure sequence, and a three-dimensional body data set of a unified coordinate system is formed through isotropic interpolation processing, and a three-dimensional view with depth information is generated by simulating the optical penetration path through the ray casting method, and the isosurface of a specific gray value is extracted through the threshold segmentation and Marching Cubes algorithm, thereby constructing a three-dimensional model.

[0022] As a preferred embodiment of the present application, in the S3, the calculation process of the shortest distance is: The space acceleration grid is constructed for any dangerous area to obtain all face patches of the dangerous area, the shortest distance from the extracorporeal candidate point to each face patch is calculated, the minimum value of the distances of all face patches is selected as the undetermined distance value, the undetermined distance values corresponding to all dangerous areas are obtained, and the minimum value in the undetermined distance values is selected as the shortest distance.

[0023] It can be understood that it is cumbersome and computationally intensive to directly calculate the distance of a point to a complex three-dimensional dangerous region, and by constructing a spatial acceleration grid, the dangerous region can be discretized into multiple patches, the complex three-dimensional distance calculation is converted into the distance calculation of a point to each patch, the calculation complexity and amount are greatly reduced, the calculation efficiency is improved, the shortest distance of each patch to the extracorporeal candidate point is calculated and the minimum value is taken as the distance value, which can ensure that the distance calculation result corresponding to each dangerous region accurately reflects the real shortest distance of the point to the region, and then the minimum value is selected from all the distance values of the dangerous regions as the final shortest distance DF1, which can ensure that the distance value is the safest distance of the extracorporeal candidate point to all dangerous regions.

[0024] As a preferred embodiment of the present application, in the S3, the specific calculation process of the preferred score is: ; Wherein, a and b are preset unit coefficients, DF1i is the shortest distance of the i th extracorporeal candidate point to the dangerous region, MaxDF1 is the maximum value of the shortest distance of all extracorporeal candidate points in the skin region to the dangerous region, DF2i is the shortest distance of the i th extracorporeal candidate point to the geometric center, and MaxDF2 is the maximum value of the shortest distance of all extracorporeal candidate points in the skin region to the geometric center.

[0025] As a preferred embodiment of the present application, in the S2, the setting process of the preset gray scale threshold is: Obtain a plurality of CT image samples of liver tumors, manually label the tumor regions on the CT image samples and obtain the gray scale values of the tumor regions, obtain the maximum and minimum values of all gray scale values, and obtain the gray scale threshold range according to the maximum and minimum values.

[0026] As a preferred embodiment of the present application, in the S3, if the preferred scores of two or more extracorporeal candidate points are equal and are all maximum values, they are all marked as initial puncture points, and corresponding puncture paths are generated respectively.

[0027] When there are multiple in-vitro candidate points with equal optimal scores and all being the maximum, these points are all marked as initial puncture points and corresponding puncture paths are generated respectively, because equal optimal scores mean that these candidate points are at the same optimal level in terms of comprehensive performance of avoiding dangerous areas (reflected by the shortest distance DF1) and approaching the geometric center of the lesion area (reflected by the shortest distance DF2), and a single standard cannot further distinguish their advantages and disadvantages. At this time, retaining all equally optimal initial puncture points and generating multiple puncture paths can avoid missing more flexible or more suitable potential paths due to the hard selection of a unique puncture point. The advantage of this is that it provides multiple alternatives for surgical planning, increasing the flexibility and fault tolerance of path selection.

[0028] As a preferred embodiment of the present application, it further includes connecting the initial puncture point and the second puncture point to obtain a path direction vector v, calculating the included angle θ between the path direction vector v and the skin area normal vector n, and if the included angle θ is greater than or equal to a preset included angle threshold, the second puncture point is removed, and a second puncture point is selected again according to the optimal score and a corresponding puncture path is generated.

[0029] If the included angle θ is too large, it will cause the puncture direction to be too inclined, increasing the risk of the path deviating from the predetermined trajectory during the puncture process, and even making it difficult to accurately enter the target area in the body or damage the tissues along the way due to poor puncture angle. By setting a preset included angle threshold to screen the path direction, a reasonable angle between the puncture path and the skin surface can be ensured, making the puncture operation more in line with the mechanical properties of human anatomy and the convenience of surgical operation. By angle constraint, the rationality and operability of the puncture path are improved, avoiding puncture difficulties or increased risks caused by abnormal angles, ensuring the feasibility of the puncture path from the direction, and ensuring that the direction of each puncture path meets the basic angle requirements of surgical operation. By strictly controlling the angle rationality of the puncture direction, the intraoperative adjustment or operation errors caused by improper path direction are reduced, and the entire puncture path not only considers safety and efficiency in terms of spatial distance.

[0030] As a preferred embodiment of the present application, in S3, when there are several puncture paths, the lengths of the puncture paths are obtained, and the puncture path with the shortest length is selected as the final puncture path.

[0031] By comparing the lengths of the paths and selecting the shortest path, the efficiency of the puncture path can be optimized, and unnecessary tissue penetration and operation steps can be reduced. On the basis of ensuring the safety and feasibility of the path, the minimally invasive and economic nature of the surgery is further improved. By quantifying the length of the path as a key indicator, the optimal solution is selected from multiple compliant paths, so that the surgical planning has the advantage of operation efficiency in addition to safety and accuracy. By shortening the puncture path, the probability of complications such as bleeding and infection during the operation is reduced.

[0032] A minimally invasive surgery planning system based on image navigation, comprising: A data acquisition module is configured to continuously scan a patient scanning area by dynamic CT technology to obtain a plurality of detection images, and to construct a three-dimensional model according to the detection images; A structure analysis module is configured to perform structural analysis on the three-dimensional model based on a medical image segmentation algorithm to obtain a skin region, an organ region, a blood vessel region and a rib region, to mark the blood vessel region and the rib region as a dangerous region, to perform grid division on the organ region, to mark a grid as a lesion grid if a gray value of any grid region is within a preset gray threshold range, and to mark a region composed of all lesion grids as a lesion region; An initial determination module is configured to construct an extracorporeal candidate point set according to the skin region, to calculate a shortest distance DF1 from each extracorporeal candidate point to the dangerous region using an Euclidean distance transformation algorithm, to analyze vertex coordinates of the lesion region and calculate a geometric center coordinate of the lesion region using a centroid method to obtain a shortest distance DF2 from each extracorporeal candidate point to the geometric center, and to calculate a preferred score of each extracorporeal candidate point according to DF1 and DF2, and to select an extracorporeal candidate point with the highest preferred score as an initial puncture point; A path planning module is configured to perform orthogonal projection on any extracorporeal candidate point set based on three-dimensional coordinates of the extracorporeal candidate point set to obtain a corresponding body surface projection point, to extend a preset vertical interval distance d in a direction of a normal vector of the projection point into the body, to generate an intracorporeal candidate point set having a spatial corresponding relationship with the extracorporeal candidate point, to repeat the above process for any intracorporeal candidate point to obtain a second puncture point, and to repeat the above process until a first puncture point reaches the lesion region, to connect all puncture points to obtain a final puncture path.

[0033] The above describes one embodiment of the present application in detail, but the content described is only a preferred embodiment of the present application and cannot be considered as limiting the scope of the present application. Any equivalent changes and improvements made according to the scope of the present application should still be within the scope of the present application.

Claims

1. A minimally invasive surgery planning method based on image navigation, characterized in that: The following steps are involved: S1, continuously scanning a patient's scanning area using dynamic CT technology to obtain a plurality of detection images, and constructing a three-dimensional model based on the detection images; S2: Perform structural analysis of the 3D model based on a medical image segmentation algorithm to obtain skin, organ, blood vessel, and rib regions. The blood vessel and rib regions are marked as risk areas, and the organ regions are divided into grids. If the grayscale value of any grid region is within a preset grayscale threshold, the grid is marked as a lesion grid, and the area consisting of all lesion grids is marked as a lesion area. S3: Construct an in vitro candidate point set based on the skin area and use the Euclidean distance transformation algorithm to calculate the shortest distance DF1 from each in vitro candidate point to the dangerous area. Analyze the vertex coordinates of the lesion area and use the centroid method to calculate the coordinates of the geometric center of the lesion area to obtain the shortest distance DF2 from each in vitro candidate point to the geometric center. Calculate the preference score of each in vitro candidate point based on DF1 and DF2, and select the in vitro candidate point with the highest preference score as the initial puncture point. S4, based on the three-dimensional coordinates of any in vitro candidate point set, perform orthogonal projection on any in vitro candidate point to obtain the corresponding surface projection point and extend the preset vertical interval d into the body along the normal vector direction of the projection point to generate an in vivo candidate point set that has a spatial correspondence with the in vitro candidate point. Repeat the above process for any in vivo candidate point to obtain the second puncture point; repeat the above process until the first puncture point reaches the lesion area, then connect all puncture points to obtain the final puncture path.

2. The minimally invasive surgery planning method based on image navigation according to claim 1, characterized in that: In S1, the specific construction process of the three-dimensional model is as follows: A rigid registration algorithm based on feature point matching is used to align the detection images. The registered detection images are stacked and reconstructed in a preset structural order. Isotropic interpolation is performed to form a three-dimensional volume dataset in a unified coordinate system. The optical penetration path is simulated by ray casting to generate a three-dimensional view with depth information. Threshold segmentation and the Marching Cubes algorithm are combined to extract isosurfaces of specific grayscale values ​​to construct a three-dimensional model.

3. The minimally invasive surgery planning method based on image navigation according to claim 1, characterized in that: In S3, the shortest distance calculation process is: Perform spatial acceleration mesh construction on any dangerous area to obtain all the facets of the dangerous area, calculate the shortest distance from the in vitro candidate point to each facet, select the minimum value of all facet distances as the pending distance value, obtain the pending distance values ​​corresponding to all dangerous areas, and select the minimum value among the pending distance values ​​as the shortest distance.

4. The minimally invasive surgery planning method based on image navigation according to claim 1, characterized in that: In S3, the specific calculation process of the preferred score is: ； Among them, a and b are preset unit coefficients, DF1i is the shortest distance from the i-th external candidate point to the dangerous area, MaxDF1 is the maximum value of the shortest distances from all external candidate points in the skin area to the dangerous area, DF2i is the shortest distance from the i-th external candidate point to the geometric center, and MaxDF2 is the maximum value of the shortest distances from all external candidate points in the skin area to the geometric center.

5. The minimally invasive surgery planning method based on image navigation according to claim 1, characterized in that: In S2, the process of setting the preset grayscale threshold is as follows: Several CT image samples of liver tumors are obtained, the tumor area on the CT image samples is manually annotated, the grayscale value of the tumor area is obtained, the maximum and minimum values ​​of all grayscale values ​​are obtained, and the grayscale threshold range is obtained according to the maximum and minimum values.

6. The minimally invasive surgery planning method based on image navigation according to claim 1, characterized in that: In the above-mentioned S3, if there are two or more in vitro candidate points whose preference scores are equal and both are maximum values, they are all marked as initial puncture points, and corresponding puncture paths are generated for each of them.

7. The minimally invasive surgery planning method based on image navigation according to claim 6, characterized in that: It also includes connecting the initial puncture point and the second puncture point to obtain a path direction vector v, calculating the angle θ between the path direction vector v and the skin area normal vector n, and if the angle θ is greater than or equal to a preset angle threshold, eliminating the second puncture point, and reselecting the second puncture point based on the preferred score and generating a corresponding puncture path.

8. The method for minimally invasive surgery planning based on image navigation according to claim 1, characterized in that: The step S4 further includes obtaining the length of each puncture path when there are multiple puncture paths, and selecting the puncture path with the shortest length as the final puncture path.

9. A minimally invasive surgery planning system based on image navigation, characterized in that: include: A data acquisition module is used to continuously scan the patient's scanning area using dynamic CT technology to obtain a number of detection images, and to construct a three-dimensional model based on the detection images; The structural analysis module is used to perform structural analysis on the 3D model based on a medical image segmentation algorithm to obtain the skin, organ, blood vessel and rib regions. The blood vessel and rib regions are marked as risk areas, and the organ regions are divided into grids. If the grayscale value of any grid area is within a preset grayscale threshold, the grid is marked as a lesion grid, and the area consisting of all lesion grids is marked as a lesion area. The initial determination module is used to construct an in vitro candidate point set based on the skin area, calculate the shortest distance DF1 from each in vitro candidate point to the dangerous area using the Euclidean distance transformation algorithm; analyze the vertex coordinates of the lesion area and calculate the coordinates of the geometric center of the lesion area using the centroid method to obtain the shortest distance DF2 from each in vitro candidate point to the geometric center; calculate the preferred score of each in vitro candidate point based on DF1 and DF2, and select the in vitro candidate point with the highest preferred score as the initial puncture point; The path planning module is used to perform orthogonal projection on any external candidate point based on the three-dimensional coordinates of the external candidate point set to obtain the corresponding surface projection point and extend the preset vertical interval distance d into the body along the normal vector direction of the projection point to generate an internal candidate point set that has a spatial correspondence with the external candidate point. The above process is repeated for any internal candidate point to obtain the second puncture point; the above process is repeated until the first puncture point reaches the lesion area, and all puncture points are connected to obtain the final puncture path.

Images