Tumor ablation decision-making method and system driven by CT big data

Abstract

The invention discloses a tumor ablation decision-making method and system driven by CT big data. The method comprises the following steps: acquiring CT image data of a patient, preprocessing, and extracting to obtain an anatomical structure feature identifier set; calculating possible puncture path connecting lines to obtain a candidate path space set; adopting a discrete optimization algorithm to obtain an optimal solution of the puncture path; managing a needle point and a needle entry point of the ablation needle, and controlling an intraoperative needle passage to accord with a planned path; calculating through a heat conduction equation, generating and updating an ablation thermal field model, and dynamically displaying ablation thermal field distribution; and calculating an ablation index, and adjusting an ablation parameter or a puncture path. The system comprises a data set extraction module, a puncture path connecting line calculation module, an optimal space set module, a puncture path optimal solution calculation module, a needle passage control module, a thermal field distribution display module and an adjustment module. The precision and safety of tumor ablation can be greatly improved, excessive ablation or incomplete ablation is avoided, and damage to normal tissue is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of tumor ablation, and in particular to a tumor ablation decision method and system driven by CT big data. Background Art

[0002] Microwave ablation and radiofrequency ablation are important methods for treating liver cancer and other tumors in modern medicine. Microwave ablation destroys tumor cells through the heat energy generated by high-frequency electromagnetic waves and is often used to treat liver tumors. Radiofrequency ablation guides radiofrequency waves to the target area, quickly heats and causes local tissue necrosis. These two methods have become one of the main options for liver cancer treatment due to their less trauma, quick recovery and fewer complications.

[0003] However, current ablation surgeries still have some technical deficiencies, especially in the planning, execution, and monitoring stages of the surgery. Before surgery, doctors usually rely on CT images for empirical planning, develop ablation plans, and predict the range of ablation; during surgery, puncture operations are guided by real-time images (such as CT or ultrasound), but the determination of the puncture path and ablation range is still highly dependent on the doctor's experience. For large tumors or difficult-to-ablate areas, this experience-based approach often leads to incomplete ablation or excessive ablation range, posing potential risks to patients.

[0004] The main shortcomings of the existing technology include:

[0005] (1) Lack of quantitative planning methods. In existing radiofrequency ablation and microwave ablation procedures, the preoperative ablation plan is usually roughly estimated based on CT images. There is a lack of accurate quantitative planning tools. Doctors can usually only determine the insertion point and puncture depth of the ablation needle based on experience, but there is no scientific calculation method to accurately plan the ablation range. This may result in an excessively large ablation range, damage to normal tissues, or incomplete ablation, and failure to completely eliminate the tumor.

[0006] (2) Reliance on physician experience. Existing ablation systems rely mainly on physician experience to arrange the needle tract during surgery. Physicians need to monitor CT or ultrasound images in real time. However, the accuracy of ablation needle placement is limited by human factors such as physician experience and spatial imagination. Therefore, ablation needles often leave part of the tumor due to inaccurate operation, increasing the risk of recurrence.

[0007] (3) Lack of real-time dynamic monitoring. The existing system lacks real-time ablation index and ablation size monitoring, and cannot provide real-time feedback on the effectiveness of the ablation area. Physicians cannot obtain the thermal field distribution, ablation effect, and coverage of the ablation area during the ablation process in real time, which affects the accuracy of the ablation process and the treatment effect.

[0008] (4) Overly simplistic coverage targets. Currently, most ablation systems only focus on 100% coverage of the ablation range, ignoring the problem of over-ablation rate. In some treatments for large tumors, complete ablation may not be the best choice. Excessive ablation may not only cause damage to normal tissues, but also affect the patient's postoperative recovery. Summary of the invention

[0009] The purpose of the present invention is to provide a CT big data driven tumor ablation decision method and system to solve at least one of the above technical problems, which can greatly improve the accuracy and safety of tumor ablation, avoid excessive or incomplete ablation, and reduce damage to normal tissues.

[0010] The embodiment of the present invention is achieved as follows:

[0011] A tumor ablation decision-making method driven by CT big data, comprising:

[0012] The CT image data of the patient is obtained, preprocessed, and the image segmentation and feature extraction technology is used to extract the anatomical structure feature identification set from the CT image data.

[0013] For the extracted anatomical structure feature identification set, possible puncture path connection lines are calculated to obtain a candidate path space set.

[0014] Set the constraints of the puncture path and screen out the optimal candidate path space set.

[0015] A discrete optimization algorithm is used to calculate the ablation target position and obtain the optimal solution for the puncture path.

[0016] Manage the needle tip and entry point of the ablation needle, and control the needle path during surgery to ensure it complies with the planned path.

[0017] According to the real-time ablation parameters, the ablation thermal field model is calculated, generated and updated through the heat conduction equation, and the ablation thermal field distribution is dynamically displayed.

[0018] Calculate ablation indicators, and adjust ablation parameters or puncture paths according to the ablation indicators.

[0019] In a preferred embodiment of the present invention, in the above-mentioned CT big data driven tumor ablation decision method, the step of obtaining the patient's CT image data, performing preprocessing, and using image segmentation and feature extraction technology to extract an anatomical structure feature identification set from the CT image data includes:

[0020] The CT scanner scans the lesion area of ​​the patient and generates two-dimensional slice data in a specified medical image format.

[0021] The two-dimensional slice data is standardized to obtain CT image data.

[0022] Organs, tumors, blood vessels and other organ structures are segmented from the CT image data using a threshold-based segmentation method to generate a corresponding anatomical structure model.

[0023] The polygon clipper algorithm in the visualization development toolkit is used to clip the CT image data according to the set bounding box to obtain the target organ and the organ structure related to the puncture path.

[0024] The bounding box is set by the boundaries of the target organ and the actual length of the ablation tool.

[0025] According to the cropped CT image data, various types of related anatomical structure feature identification sets are generated.

[0026] Its technical effects are: by acquiring the patient's lesion area through CT scanning and generating two-dimensional slice data, it can provide accurate anatomical structure information, including the location, size, shape of the tumor and its relationship with surrounding tissues, ensuring the consistency and reliability of the data; using a threshold-based segmentation method to separate important anatomical structures such as tumors, organs, and blood vessels from CT images, accurately calibrating the morphology and position of the tumor, and providing accurate tumor boundary information, it can help the system accurately identify the lesion area and avoid affecting normal tissues. By segmenting different anatomical structures, the system can generate an independent model for each important area (such as tumors, organs, blood vessels, etc.), which is convenient for subsequent ablation path planning and dynamic monitoring during surgery; through image segmentation, feature extraction and cropping, a set of anatomical structure feature identifiers is finally generated, including the precise geometric morphology and positional relationship of the tumor and its surrounding tissues, providing detailed spatial and structural data for subsequent puncture path planning, ablation thermal field simulation, ablation effect evaluation, etc.

[0027] In a preferred embodiment of the present invention, in the above-mentioned CT big data driven tumor ablation decision method, the anatomical structure feature identification set includes:

[0028] The surface tissue position is extracted from the surface tissue model of the cropped CT image data, representing the spatial position of the surface tissue and serving as the starting point of the puncture path.

[0029] The internal organ position is extracted from the internal organ model of the cropped CT image data, represents the spatial position of the internal organ, and is used to determine the shortest distance between the puncture path and the organ.

[0030] The tumor target position is extracted from the tumor region model of the cropped CT image data, represents the target position of the tumor, and is used to identify the position of the ablation target.

[0031] The ablation coverage area is generated by extending the tumor target position and adding a safety boundary, representing the ablation area, and is used to evaluate whether the ablation area effectively covers the tumor.

[0032] The healthy tissue region is generated by expanding the ablation coverage area and selecting the center point of the external voxel, representing the healthy tissue region outside the ablation region, and is used to determine whether the ablation region causes damage to the healthy tissue.

[0033] Its technical effects are as follows: extraction of surface tissue positions ensures that the needle track can start from the appropriate position; extraction of internal organ positions allows the system to determine the shortest distance between the puncture path and important organs (such as the liver, blood vessels, etc.), thereby optimizing the puncture path; extraction of tumor target positions is the core of ablation surgery planning, and accurate identification of tumor positions can ensure the accuracy of ablation targets; generation of ablation coverage areas is based on the expansion of tumor target positions and the addition of safety boundaries, ensuring that the ablation range completely covers the tumor area and prevents omissions or insufficient ablation; extraction and expansion of healthy tissue areas help determine whether the ablation area will cause damage to healthy tissues. By extracting and managing these key anatomical structural features, the tumor ablation process can be accurately planned before and during surgery, providing precise puncture path planning, reducing accidental injury to normal tissues, optimizing tumor positioning, ensuring that the ablation range covers the entire tumor, protecting healthy tissues, and preventing damage caused by excessive ablation.

[0034] In a preferred embodiment of the present invention, in the above-mentioned CT big data driven tumor ablation decision method, the extracted anatomical structure feature identification set, calculating possible puncture path connection lines, and obtaining a candidate path space set includes:

[0035] All connecting lines between the superficial tissue locations and the tumor target locations are calculated.

[0036] In combination with the internal organ position, puncture needle depth and puncture needle length, taking into account the safety of the path, the shortest distance and the puncture angle, all possible puncture path connection lines are calculated to obtain a candidate path space set.

[0037] Its technical effect is: by combining the position of surface tissue, tumor target position and internal organ position, possible puncture path connection lines are calculated to ensure that the path selection avoids direct collision or contact with key organs during puncture as much as possible; the selection of the shortest path maximizes puncture efficiency, reduces patient discomfort and trauma, reduces the time and difficulty of doctor's operation, and combines the depth and length of the puncture needle to optimize the path so that the ablation needle can accurately reach the tumor target position in the shortest possible path, thereby improving the treatment effect.

[0038] In a preferred embodiment of the present invention, in the above-mentioned CT big data driven tumor ablation decision method, the constraint conditions of setting the puncture path and screening the optimal candidate path space set include:

[0039] The depth of the puncture path is set not to exceed the maximum puncture length of the puncture needle.

[0040] The angle between the puncture path and the tumor boundary is set to be no less than a preset value.

[0041] The shortest distance between the puncture path and the internal organ position is set to be no less than a preset value. If the puncture path must pass through normal organ tissue before reaching the tumor target position, the thickness of the normal organ tissue is no less than a preset value.

[0042] The optimal candidate path space set is screened.

[0043] Its technical effects are: setting the depth of the puncture path not to exceed the maximum puncture length of the puncture needle can effectively prevent the puncture needle from penetrating too deep and reduce damage to the patient's tissue; by setting the angle between the puncture path and the tumor boundary to be no less than the preset value, it prevents the puncture path from being too parallel or too perpendicular to the tumor boundary, causing the puncture needle to be unable to accurately reach the tumor target position; setting the shortest distance between the puncture path and the internal organs to be no less than the preset value effectively avoids direct contact between the puncture path and key organs (such as blood vessels, liver, etc.), and reduces the risk of damage to normal organs during operation. By setting the constraints of the puncture path and screening the optimal candidate path space set, the safety of the puncture path is improved, damage to normal tissues and key organs is reduced, tumor targeting accuracy is ensured, the ablation effect is improved, intraoperative complications are reduced, the risk of accidental injury to healthy tissues and key organs is reduced, the intelligence and automation of path planning are improved, human factors and misjudgment are reduced, healthy tissues are protected, and the ablation area is ensured to effectively cover the tumor.

[0044] In a preferred embodiment of the present invention, in the above-mentioned CT big data driven tumor ablation decision method, the management of the needle tip point and the needle entry point of the ablation needle and the control of the needle path during the operation to conform to the planned path include:

[0045] Initial positioning of the ablation needle Among them, P tip,0 is the initial needle tip position of the ablation needle, P entry,0 is the initial insertion point of the ablation needle, d0 is the initial effective length of the ablation needle, is the initial direction unit vector of the ablation needle.

[0046] Monitor the actual position of the ablation needle in real time, compare it with the planned path, and calculate the error of the needle path Among them, E needle(t) is the needle path error at time t, which indicates the deviation between the actual position of the needle path and the planned path, in meters, P needle (t) is the needle track position measured in real time, C path (t) is the expected position of the planned path at time t, is the unit vector of the needle track direction measured in real time, is the direction unit vector of the planned path at time t, and λ1 is the weight coefficient used to balance the position error and direction error.

[0047] If the needle path error E needle If (t) exceeds the set tolerance range, the feedback control algorithm is used to calculate the offset that needs to be adjusted and perform needle path correction. The calculation formula is δP(t) = K adjust ·E needle (t), δP(t) is the adjustment offset of the needle track at time t, K adjust To adjust the gain matrix, E represents the ratio and direction of needle path correction. needle (t) is the needle track error.

[0048] Update the needle entry point and needle tip position of the ablation needle to obtain a new needle track configuration Among them, P tip,new (t) is the updated needle tip position, P entry,new (t) is the updated needle entry point position, d(t) is the effective length of the current ablation needle, The unit vector of the current needle track direction.

[0049] Update the needle path so that the target position of the needle path covers the ablation coverage area. The update formula of the target position is P target =C target (t) = C path (t)+δC(t), where P target is the target position of the ablation needle, indicating the ablation coverage area where the tumor is located, C target (t) is the target position path, δC(t) is the path offset, which indicates the adjustment of the planned path.

[0050] Its technical effects are: by initially positioning the ablation needle and monitoring the position of the needle track in real time, the needle tip position and the needle entry point can be ensured to be accurate; by calculating the error of the needle track and using the feedback control algorithm to correct it, the deviation of the ablation needle in actual operation can be effectively corrected; by updating the needle track path, the target position of the ablation needle can cover the ablation coverage area where the tumor is located, ensuring that the entire tumor area can be effectively ablated; the needle track correction process can effectively avoid complications caused by path errors, such as puncture errors, accidental injury to surrounding tissues, etc. The application of the feedback control algorithm can correct the needle track position deviation in a timely manner, avoid sudden accidents during surgery, and improve the safety of patients.

[0051] In a preferred embodiment of the present invention, in the above-mentioned CT big data driven tumor ablation decision method, the management of the needle tip point and the needle entry point of the ablation needle and the control of the needle path during the operation to conform to the planned path also include:

[0052] A constrained optimization algorithm is used to ensure that the movement of the needle tip of the ablation needle does not exceed the safety range of the ablation coverage area. The objective function is: Among them, E needle (t) is the error function, Ω is the constraint area, which represents the effective space in which the needle track of the ablation needle can move, and is the safe distance between the ablation coverage area and the healthy tissue area.

[0053] Its technical effects are: through the constraint optimization algorithm, it is ensured that the needle tip of the ablation needle does not exceed the set safety area during movement. According to the design of the objective function, the path of the ablation needle is dynamically adjusted to ensure that the needle tip always moves within the safe distance range; by defining the error function, it is ensured that the path of the ablation needle tip meets the requirements of the planned path, while avoiding dangerous areas and preventing the needle path from deviating from the ablation target area; combined with real-time monitoring data, the ablation needle path is dynamically calculated and corrected. If it is found that the ablation needle is approaching or about to cross the safety boundary, the needle path is adjusted in real time according to the set constraints, thereby effectively avoiding deviation from the established path.

[0054] In a preferred embodiment of the present invention, in the above-mentioned CT big data driven tumor ablation decision method, the calculation, generation and updating of the ablation thermal field model by the heat conduction equation according to the real-time ablation parameters and the dynamic display of the ablation thermal field distribution include:

[0055] Real-time ablation parameters include ablation power P ablation , ablation needle length L needle , ablation needle position (x, y, z) and prescription temperature T prescription .

[0056] Establish a heat conduction equation to describe the temperature change and thermal field distribution in the tumor area Among them, ρ is the density of the medium, c is the specific heat capacity of the medium, T is the temperature, t is the time, k is the thermal conductivity, and Q is the heat source term.

[0057] According to the heat conduction equation, combined with real-time ablation parameters and the patient's anatomical information, the heat conduction equation is solved by numerical calculation method to obtain the distribution model of the ablation thermal field. Among them, T ambient is the ambient temperature, P ablation (i) is the power of the i-th ablation needle, r is the radial distance from the ablation needle, σ i is the standard deviation of ablation heat source diffusion, and N is the number of ablation needles.

[0058] According to the real-time monitoring data, the ablation thermal field model T(x,y,z,t new )=T(x,y,z,t old )+ΔT(x,y,z,Δt), where ΔT(x,y,z,Δt) is the temperature change within the time step Δt.

[0059] The calculated ablation thermal field model is visualized to display the ablation effect of the tumor area in real time.

[0060] Its technical effects are: through the combination of heat conduction equations and real-time ablation parameters, the ablation thermal field model is updated in real time, and temperature changes are captured and fed back in real time to ensure that the ablation process is accurately executed and avoid overheating or insufficient temperature problems; through the dynamic display of thermal field distribution, doctors can monitor the thermal field changes in real time, and immediately adjust the ablation power or needle position when necessary, so as to optimize the ablation effect and avoid excessive damage to healthy tissue; through the heat conduction equation model, the temperature distribution during the ablation process is accurately simulated, especially the thermal field changes in the tumor area. The real-time update and visual display of the ablation thermal field model help to timely determine whether the ablation needle is accurately positioned in the tumor target area, whether there is heat leakage affecting healthy tissue, improve safety, and avoid damage to healthy tissue; the ablation power, needle position or other parameters can be adjusted more efficiently, thereby improving treatment efficiency and reducing treatment time.

[0061] In a preferred embodiment of the present invention, in the above-mentioned CT big data driven tumor ablation decision method, the calculating of ablation indicators and adjusting ablation parameters or puncture paths according to the ablation indicators include:

[0062] Calculating coverage It is used to indicate the proportion of the ablation coverage area covering the tumor target position, wherein TA is the ablation coverage area, A is the coverage range of the tumor target position, and |·| is the area or volume of the region.

[0063] The non-ablation rate U=(1-C)×100% was calculated.

[0064] Calculated ablation rate It is used to indicate the proportion of the ablation coverage area beyond the tumor targeting position and affecting normal tissue, wherein NT is the healthy tissue area beyond the tumor targeting position.

[0065] An objective function is set to minimize the sum of the non-ablation rate U and the over-ablation rate O.

[0066] A constraint condition is set that the coverage rate C is not less than a preset value.

[0067] The real-time ablation parameters are adjusted using a discrete optimization algorithm to optimize the path so as to minimize the objective function while satisfying the constraint conditions.

[0068] According to the optimization result, the real-time ablation parameters are updated, and the puncture path is adjusted based on the optimal solution.

[0069] Its technical effects are: by calculating the coverage rate, non-ablation rate and over-ablation rate, the overlap between the ablation area and the tumor target position can be accurately quantified to ensure that the ablation area effectively covers the tumor target area; through the discrete optimization algorithm, the ablation path is optimized in real time according to the objective function of minimizing the sum of the non-ablation rate and the over-ablation rate.

[0070] A CT big data driven tumor ablation decision system, comprising:

[0071] The data set extraction module is used to obtain the patient's CT image data, perform preprocessing, and use image segmentation and feature extraction technology to extract an anatomical structure feature identification set from the CT image data.

[0072] The puncture path connection line calculation module is used to calculate possible puncture path connection lines for the extracted anatomical structure feature identification set to obtain a candidate path space set.

[0073] The optimal space set module is used to set the constraints of the puncture path and screen the optimal candidate path space set.

[0074] The puncture path optimal solution calculation module is used to calculate the ablation target position using a discrete optimization algorithm to obtain the optimal solution of the puncture path.

[0075] The needle track control module is used to manage the needle tip and entry point of the ablation needle and control the needle track during the operation to conform to the planned path.

[0076] The thermal field distribution display module is used to calculate, generate and update the ablation thermal field model according to the real-time ablation parameters through the heat conduction equation, and dynamically display the ablation thermal field distribution.

[0077] The adjustment module is used to calculate ablation indicators and adjust ablation parameters or puncture paths according to the ablation indicators.

[0078] The beneficial effects of the embodiments of the present invention are:

[0079] The present invention obtains the patient's lesion area through CT scanning and generates two-dimensional slice data, and then uses image segmentation and feature extraction technology to accurately segment anatomical structures such as tumors, organs, and blood vessels after standardized processing. By segmenting important anatomical structures such as tumors, organs, and blood vessels, the system can accurately obtain the relationship between the morphology and position of the tumor and the surrounding tissues; accurately calibrate the boundaries and positions of the tumor, effectively avoid accidental injury to normal tissues during surgery, and provide high-precision data for subsequent path planning and dynamic monitoring; based on an accurate anatomical structure model, it provides accurate spatial and geometric data for subsequent puncture path planning, thermal field simulation, and ablation effect evaluation.

[0080] The present invention calculates possible puncture path connection lines and selects the best candidate path based on constraints such as depth, angle, and safety. The puncture path is optimized by combining the spatial information of the tumor target position, surface tissue, and internal organs to reduce damage to key organs during the puncture process; the shortest path and appropriate angle are selected so that the ablation needle can quickly and accurately reach the tumor target position, reducing trauma and discomfort; and constraints such as the shortest distance between the puncture path and the tumor boundary and internal organs are set to effectively reduce the risk to healthy tissues during the puncture process.

[0081] The present invention monitors the needle tip and entry point of the ablation needle in real time, and uses a feedback control algorithm to adjust the path to ensure that the needle path during the operation always conforms to the planned path. The needle path error is monitored in real time, and the feedback control algorithm can correct the deviation of the needle path in time to ensure that the ablation needle accurately reaches the tumor target area; by dynamically adjusting the needle path, puncture errors or accidental injury to surrounding healthy tissues caused by path errors are avoided, and the safety of treatment is improved; it ensures that the ablation needle can effectively cover the entire tumor area, optimize the ablation effect, and avoid excessive or insufficient ablation.

[0082] The present invention calculates the ablation thermal field distribution through real-time ablation parameters and heat conduction equations, and updates the ablation thermal field model in real time. By combining the heat conduction equation with real-time parameters, the ablation power and needle position are dynamically adjusted to ensure that the temperature change during the ablation process meets the preset standards and avoid overheating or insufficient temperature; the ablation thermal field distribution is dynamically displayed to understand the ablation situation of the tumor area in real time, and adjust the treatment plan as needed to optimize the treatment effect; by monitoring the distribution of the ablation thermal field, it is ensured that the ablation process will not affect healthy tissues, effectively avoiding damage caused by excessive ablation;

[0083] By calculating the coverage rate, non-ablation rate and over-ablation rate, the ablation path and parameters are adjusted using a discrete optimization algorithm. By calculating the coverage rate, non-ablation rate and over-ablation rate, it is ensured that the ablation area can completely cover the tumor target position while avoiding excessive damage to healthy tissue; the discrete optimization algorithm is used to minimize the sum of non-ablation and over-ablation, ensuring that the ablation path and parameters are dynamically adjusted during the treatment process to achieve the best treatment effect; through precise control of the ablation coverage area, damage to healthy tissue is avoided, intraoperative complications are reduced, and safety is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0085] Figure 1 This is a flow chart of the CT big data driven tumor ablation decision-making method of the present invention. DETAILED DESCRIPTION

[0086] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0087] Please refer to Figure 1 The first embodiment of the present invention provides a tumor ablation decision-making method driven by CT big data, including: a tumor ablation decision-making method driven by CT big data, which includes: obtaining the patient's CT image data, performing preprocessing, and using image segmentation and feature extraction technology to extract a set of anatomical structure feature identifiers from the CT image data; calculating possible puncture path connection lines for the extracted anatomical structure feature identifier set to obtain a candidate path space set; setting puncture path constraint conditions to screen and obtain an optimal candidate path space set; using a discrete optimization algorithm to calculate the ablation target position to obtain the optimal solution for the puncture path; managing the needle tip and needle entry point of the ablation needle to control the needle track during the operation to conform to the planned path; calculating, generating and updating the ablation thermal field model according to the real-time ablation parameters through the heat conduction equation, and dynamically displaying the ablation thermal field distribution; calculating ablation indicators, and adjusting the ablation parameters or puncture path according to the ablation indicators.

[0088] In a preferred embodiment of the present invention, in the above-mentioned CT big data driven tumor ablation decision method, the CT image data of the patient is obtained, preprocessed, and the image segmentation and feature extraction technology is used to extract the anatomical structure feature identification set from the CT image data, including: scanning the patient's lesion area with a CT scanner to generate two-dimensional slice data in a specified medical image format; standardizing the two-dimensional slice data to obtain CT image data; using a threshold-based segmentation method to segment organs, tumors, blood vessels and other organ structures from the CT image data to generate corresponding anatomical structure models; using the polygon clipper algorithm in the visualization development toolkit to clip the CT image data according to the set bounding box to obtain the target organ and the organ structure related to the puncture path; the bounding box is set by the boundary of the target organ and the actual length of the ablation tool; and generating various types of related anatomical structure feature identification sets according to the clipped CT image data.

[0089] Its technical effects are: by acquiring the patient's lesion area through CT scanning and generating two-dimensional slice data, it can provide accurate anatomical structure information, including the location, size, shape of the tumor and its relationship with surrounding tissues, ensuring the consistency and reliability of the data; using a threshold-based segmentation method to separate important anatomical structures such as tumors, organs, and blood vessels from CT images, accurately calibrating the morphology and position of the tumor, and providing accurate tumor boundary information, it can help the system accurately identify the lesion area and avoid affecting normal tissues. By segmenting different anatomical structures, the system can generate an independent model for each important area (such as tumors, organs, blood vessels, etc.), which is convenient for subsequent ablation path planning and dynamic monitoring during surgery; through image segmentation, feature extraction and cropping, a set of anatomical structure feature identifiers is finally generated, including the precise geometric morphology and positional relationship of the tumor and its surrounding tissues, providing detailed spatial and structural data for subsequent puncture path planning, ablation thermal field simulation, ablation effect evaluation, etc.

[0090] In a preferred embodiment of the present invention, in the above-mentioned CT big data driven tumor ablation decision method, the anatomical structure feature identification set includes: surface tissue position, extracted from the surface tissue model of the cropped CT image data, representing the spatial position of the surface tissue, serving as the starting point of the puncture path; internal organ position, extracted from the internal organ model of the cropped CT image data, representing the spatial position of the internal organ, used to determine the shortest distance between the puncture path and the organ; tumor target position, extracted from the tumor region model of the cropped CT image data, representing the target position of the tumor, used to identify the location of the ablation target; ablation coverage area, generated by extending the tumor target position and adding a safety boundary, representing the ablation area, used to evaluate whether the ablation area effectively covers the tumor; healthy tissue area, generated by extending the ablation coverage area and selecting the center point of the external voxel, representing the healthy tissue area outside the ablation area, used to determine whether the ablation area causes damage to the healthy tissue.

[0091] Its technical effects are as follows: extraction of surface tissue positions ensures that the needle track can start from the appropriate position; extraction of internal organ positions allows the system to determine the shortest distance between the puncture path and important organs (such as the liver, blood vessels, etc.), thereby optimizing the puncture path; extraction of tumor target positions is the core of ablation surgery planning, and accurate identification of tumor positions can ensure the accuracy of ablation targets; generation of ablation coverage areas is based on the expansion of tumor target positions and the addition of safety boundaries, ensuring that the ablation range completely covers the tumor area and prevents omissions or insufficient ablation; extraction and expansion of healthy tissue areas help determine whether the ablation area will cause damage to healthy tissues. By extracting and managing these key anatomical structural features, the tumor ablation process can be accurately planned before and during surgery, providing precise puncture path planning, reducing accidental injury to normal tissues, optimizing tumor positioning, ensuring that the ablation range covers the entire tumor, protecting healthy tissues, and preventing damage caused by excessive ablation.

[0092] In a preferred embodiment of the present invention, in the above-mentioned CT big data driven tumor ablation decision method, the extracted anatomical structure feature identification set is used to calculate possible puncture path connection lines to obtain a candidate path space set, including: calculating all connection lines between the surface tissue position and the tumor target position; combining the internal organ position, puncture needle depth and puncture needle length, considering the safety of the path, the shortest distance and the puncture angle, calculating all possible puncture path connection lines to obtain a candidate path space set.

[0093] Its technical effect is: by combining the position of surface tissue, tumor target position and internal organ position, possible puncture path connection lines are calculated to ensure that the path selection avoids direct collision or contact with key organs during puncture as much as possible; the selection of the shortest path maximizes puncture efficiency, reduces patient discomfort and trauma, reduces the time and difficulty of doctor's operation, and combines the depth and length of the puncture needle to optimize the path so that the ablation needle can accurately reach the tumor target position in the shortest possible path, thereby improving the treatment effect.

[0094] In a preferred embodiment of the present invention, in the above-mentioned CT big data driven tumor ablation decision method, the constraints of the puncture path are set, and the optimal candidate path space set is screened to obtain the following: the depth of the puncture path is set not to exceed the maximum puncture length of the puncture needle; the angle between the puncture path and the tumor boundary is set to be not less than a preset value; the shortest distance between the puncture path and the internal organ position is set to be not less than a preset value, and if the puncture path must pass through normal organ tissue before reaching the tumor target position, the thickness of the normal organ tissue is not less than a preset value; and the optimal candidate path space set is screened to obtain.

[0095] Its technical effects are: setting the depth of the puncture path not to exceed the maximum puncture length of the puncture needle can effectively prevent the puncture needle from penetrating too deep and reduce damage to the patient's tissue; by setting the angle between the puncture path and the tumor boundary to be no less than the preset value, it prevents the puncture path from being too parallel or too perpendicular to the tumor boundary, causing the puncture needle to be unable to accurately reach the tumor target position; setting the shortest distance between the puncture path and the internal organs to be no less than the preset value effectively avoids direct contact between the puncture path and key organs (such as blood vessels, liver, etc.), and reduces the risk of damage to normal organs during operation. By setting the constraints of the puncture path and screening the optimal candidate path space set, the safety of the puncture path is improved, damage to normal tissues and key organs is reduced, tumor targeting accuracy is ensured, the ablation effect is improved, intraoperative complications are reduced, the risk of accidental injury to healthy tissues and key organs is reduced, the intelligence and automation of path planning are improved, human factors and misjudgment are reduced, healthy tissues are protected, and the ablation area is ensured to effectively cover the tumor.

[0096] In a preferred embodiment of the present invention, in the above-mentioned CT big data driven tumor ablation decision method, the management of the needle tip and the needle entry point of the ablation needle and the control of the needle path during the operation to conform to the planned path include: initial positioning of the ablation needle Among them, P tip,0 is the initial needle tip position of the ablation needle, P entry,0 is the initial insertion point of the ablation needle, d0 is the initial effective length of the ablation needle, is the initial direction unit vector of the ablation needle; the actual position of the ablation needle is monitored in real time, and compared with the planned path to calculate the error of the needle path Among them, E needle (t) is the needle path error at time t, which indicates the deviation between the actual position of the needle path and the planned path, in meters, P needle (t) is the needle track position measured in real time, C path (t) is the expected position of the planned path at time t, is the unit vector of the needle track direction measured in real time, is the direction unit vector of the planned path at time t, λ1 is the weight coefficient used to balance the position error and direction error; if the needle path error E needle If (t) exceeds the set tolerance range, the feedback control algorithm is used to calculate the offset that needs to be adjusted and perform needle path correction. The calculation formula is δP(t) = K adjust ·E needle (t), δP(t) is the adjustment offset of the needle track at time t, K adjust To adjust the gain matrix, E represents the ratio and direction of needle path correction. needle (t) is the needle track error; update the needle entry point position and needle tip position of the ablation needle to obtain a new needle track configuration Among them, P tip,new (t) is the updated needle tip position, P entry,new (t) is the updated needle entry point position, d(t) is the effective length of the current ablation needle, is the direction unit vector of the current needle track; the needle track path is updated so that the target position of the needle track covers the ablation coverage area. The update formula of the target position is P target =C target (t) = C path (t)+δC(t), where P target is the target position of the ablation needle, indicating the ablation coverage area where the tumor is located, C target (t) is the target position path, δC(t) is the path offset, which indicates the adjustment of the planned path.

[0097] Its technical effects are: by initially positioning the ablation needle and monitoring the position of the needle track in real time, the needle tip position and the needle entry point can be ensured to be accurate; by calculating the error of the needle track and using the feedback control algorithm to correct it, the deviation of the ablation needle in actual operation can be effectively corrected; by updating the needle track path, the target position of the ablation needle can cover the ablation coverage area where the tumor is located, ensuring that the entire tumor area can be effectively ablated; the needle track correction process can effectively avoid complications caused by path errors, such as puncture errors, accidental injury to surrounding tissues, etc. The application of the feedback control algorithm can correct the needle track position deviation in a timely manner, avoid sudden accidents during surgery, and improve the safety of patients.

[0098] In a preferred embodiment of the present invention, in the above-mentioned CT big data driven tumor ablation decision method, the management of the needle tip and the needle entry point of the ablation needle and the control of the needle path during the operation to conform to the planned path also include: using a constrained optimization algorithm to ensure that the movement of the needle tip of the ablation needle does not exceed the safety range of the ablation coverage area, and the objective function is Among them, E needle (t) is the error function, Ω is the constraint area, which represents the effective space in which the needle track of the ablation needle can move, and is the safe distance between the ablation coverage area and the healthy tissue area.

[0099] Its technical effects are: through the constraint optimization algorithm, it is ensured that the needle tip of the ablation needle does not exceed the set safety area during movement. According to the design of the objective function, the path of the ablation needle is dynamically adjusted to ensure that the needle tip always moves within the safe distance range; by defining the error function, it is ensured that the path of the ablation needle tip meets the requirements of the planned path, while avoiding dangerous areas and preventing the needle path from deviating from the ablation target area; combined with real-time monitoring data, the ablation needle path is dynamically calculated and corrected. If it is found that the ablation needle is approaching or about to cross the safety boundary, the needle path is adjusted in real time according to the set constraints, thereby effectively avoiding deviation from the established path.

[0100] In a preferred embodiment of the present invention, in the above-mentioned CT big data driven tumor ablation decision method, the ablation thermal field model is calculated, generated and updated through the heat conduction equation according to the real-time ablation parameters, and the ablation thermal field distribution is dynamically displayed, including: the real-time ablation parameters include the ablation power P ablation , ablation needle length L needle , ablation needle position (x, y, z) and prescription temperature T prescription ; Establish a heat conduction equation to describe the temperature change and thermal field distribution in the tumor area Where ρ is the density of the medium, in kg / m 3, c is the specific heat capacity of the medium, in J / (kg·K), T is the temperature, in K, t is the time, in s, k is the thermal conductivity, in W / (m·K), and Q is the heat source term, describing the heat released by the ablation needle, in W / m 3 According to the heat conduction equation, combined with the real-time ablation parameters and the patient's anatomical information, the heat conduction equation is solved by numerical calculation method to obtain the distribution model of the ablation thermal field. Among them, T ambient is the ambient temperature, P ablation (i) is the power of the i-th ablation needle, r is the radial distance from the ablation needle, σ i is the standard deviation of ablation heat source diffusion, N is the number of ablation needles; according to real-time monitoring data, the ablation thermal field model T(x, y, z, t new )=T(x,y,z,t old )+ΔT(x,y,z,Δt), where ΔT(x,y,z,Δt) is the temperature change within the time step Δt, and the unit is K; the calculated ablation thermal field model is visualized to display the ablation effect of the tumor area in real time.

[0101] Its technical effects are: through the combination of heat conduction equations and real-time ablation parameters, the ablation thermal field model is updated in real time, and temperature changes are captured and fed back in real time to ensure that the ablation process is accurately executed and avoid overheating or insufficient temperature problems; through the dynamic display of thermal field distribution, doctors can monitor the thermal field changes in real time, and immediately adjust the ablation power or needle position when necessary, so as to optimize the ablation effect and avoid excessive damage to healthy tissue; through the heat conduction equation model, the temperature distribution during the ablation process is accurately simulated, especially the thermal field changes in the tumor area. The real-time update and visual display of the ablation thermal field model help to timely determine whether the ablation needle is accurately positioned in the tumor target area, whether there is heat leakage affecting healthy tissue, improve safety, and avoid damage to healthy tissue; the ablation power, needle position or other parameters can be adjusted more efficiently, thereby improving treatment efficiency and reducing treatment time.

[0102] In a preferred embodiment of the present invention, in the above-mentioned CT big data driven tumor ablation decision method, the calculation of ablation indicators and the adjustment of ablation parameters or puncture paths according to the ablation indicators include: calculating coverage It is used to indicate the proportion of the ablation coverage area covering the tumor target position, where TA is the ablation coverage area, A is the coverage range of the tumor target position, and |·| is the area or volume of the region; calculate the non-ablation rate U = (1-C) × 100%; calculate the over-ablation rate Used to represent the proportion of the ablation coverage area that exceeds the tumor targeting position and affects normal tissue, wherein NT is the healthy tissue area beyond the tumor targeting position; setting an objective function that minimizes the sum of the non-ablation rate U and the over-ablation rate O; setting a constraint condition that the coverage rate C is not less than a preset value; using a discrete optimization algorithm to adjust the real-time ablation parameters and optimize the path so that the objective function is minimized while satisfying the constraint conditions; updating the real-time ablation parameters according to the optimization results, and adjusting the puncture path based on the optimal solution.

[0103] Its technical effects are: by calculating the coverage rate, non-ablation rate and over-ablation rate, the overlap between the ablation area and the tumor target position can be accurately quantified to ensure that the ablation area effectively covers the tumor target area; through the discrete optimization algorithm, the ablation path is optimized in real time according to the objective function of minimizing the sum of the non-ablation rate and the over-ablation rate.

[0104] The second embodiment of the present invention provides a tumor ablation decision system driven by CT big data, which includes: a data set extraction module, which is used to obtain the patient's CT image data, perform preprocessing, and use image segmentation and feature extraction technology to extract an anatomical structure feature identification set from the CT image data; a puncture path connection line calculation module, which is used to calculate possible puncture path connection lines for the extracted anatomical structure feature identification set to obtain a candidate path space set; an optimal space set module, which is used to set the constraint conditions of the puncture path and screen the optimal candidate path space set; a puncture path optimal solution calculation module, which is used to use a discrete optimization algorithm to calculate the ablation target position and obtain the optimal solution of the puncture path; a needle track control module, which is used to manage the needle tip and entry point of the ablation needle and control the needle track during the operation to conform to the planned path; a thermal field distribution display module, which is used to calculate, generate and update the ablation thermal field model according to the real-time ablation parameters through the heat conduction equation, and dynamically display the ablation thermal field distribution; an adjustment module, which is used to calculate the ablation index and adjust the ablation parameter or puncture path according to the ablation index.

[0105] The computer program product of the CT big data driven tumor ablation decision method and device provided in the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the method in the previous method embodiment. The specific implementation can be found in the method embodiment, which will not be repeated here.

[0106] Specifically, the storage medium can be a general storage medium, such as a mobile disk, a hard disk, etc. When the computer program on the storage medium is run, it can execute the above-mentioned CT big data-driven tumor ablation decision method, thereby greatly improving the accuracy and safety of tumor ablation, avoiding the problem of excessive ablation or incomplete ablation, and reducing damage to normal tissues.

[0107] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0108] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A tumor ablation decision-making method driven by CT big data, characterized in that: include: Acquire CT image data of the patient, perform preprocessing, and use image segmentation and feature extraction technology to extract a set of anatomical structure feature identifiers from the CT image data; For the extracted anatomical structure feature identification set, possible puncture path connection lines are calculated to obtain a candidate path space set; Set the constraints of the puncture path and screen out the optimal candidate path space set; A discrete optimization algorithm is used to calculate the ablation target position and obtain the optimal solution for the puncture path; Manage the needle tip and entry point of the ablation needle, and control the needle path during the operation to conform to the planned path; According to the real-time ablation parameters, the ablation thermal field model is calculated, generated and updated through the heat conduction equation, and the ablation thermal field distribution is dynamically displayed; Calculate ablation indicators, and adjust ablation parameters or puncture paths according to the ablation indicators.

2. The CT big data driven tumor ablation decision method according to claim 1, characterized in that: The step of obtaining CT image data of the patient, performing preprocessing, and extracting an anatomical structure feature identification set from the CT image data using image segmentation and feature extraction technology includes: The CT scanner scans the patient's lesion area and generates two-dimensional slice data in a specified medical image format; Performing standardization processing on the two-dimensional slice data to obtain CT image data; Segmenting organs, tumors, blood vessels and other organ structures from the CT image data using a threshold-based segmentation method to generate a corresponding anatomical structure model; Using a polygon clipper algorithm in a visualization development toolkit, clipping the CT image data according to a set bounding box to obtain a target organ and an organ structure related to a puncture path; The bounding box is set by the boundary of the target organ and the actual length of the ablation tool; According to the cropped CT image data, various types of related anatomical structure feature identification sets are generated.

3. The CT big data driven tumor ablation decision method according to claim 2, characterized in that: The anatomical structure feature identification set includes: The surface tissue position is extracted from the surface tissue model of the cropped CT image data, representing the spatial position of the surface tissue and serving as the starting point of the puncture path; The internal organ position is extracted from the internal organ model of the cropped CT image data, represents the spatial position of the internal organ, and is used to determine the shortest distance between the puncture path and the organ; A tumor target position, extracted from the tumor region model of the cropped CT image data, represents the target position of the tumor and is used to identify the position of the ablation target; an ablation coverage area, which is generated by extending the tumor target position and adding a safety boundary, and represents the ablation area, and is used to evaluate whether the ablation area effectively covers the tumor; The healthy tissue region is generated by expanding the ablation coverage area and selecting the center point of the external voxel, representing the healthy tissue region outside the ablation region, and is used to determine whether the ablation region causes damage to the healthy tissue.

4. The CT big data driven tumor ablation decision method according to claim 3, characterized in that: The extracted anatomical structure feature identification set is used to calculate possible puncture path connection lines to obtain a candidate path space set, including: calculating all connection lines between the surface tissue position and the tumor target position; In combination with the internal organ position, puncture needle depth and puncture needle length, taking into account the safety of the path, the shortest distance and the puncture angle, all possible puncture path connection lines are calculated to obtain a candidate path space set.

5. The CT big data driven tumor ablation decision method according to claim 3, characterized in that: The constraint conditions of the puncture path are set, and the optimal candidate path space set is screened and obtained, including: The depth of the puncture path is set not to exceed the maximum puncture length of the puncture needle; Setting the angle between the puncture path and the tumor boundary to be no less than a preset value; The shortest distance between the puncture path and the internal organ position is set to be no less than a preset value, and if the puncture path must pass through normal organ tissue before reaching the tumor target position, the thickness of the normal organ tissue is no less than a preset value; The optimal candidate path space set is screened.

6. The CT big data driven tumor ablation decision method according to claim 3, characterized in that: Managing the needle tip and needle entry point of the ablation needle and controlling the needle path to conform to the planned path during the operation includes: Initial positioning of the ablation needle Among them, P tip,0 is the initial needle tip position of the ablation needle, P entry,0 is the initial insertion point of the ablation needle, d0 is the initial effective length of the ablation needle, is the initial direction unit vector of the ablation needle; Monitor the actual position of the ablation needle in real time, compare it with the planned path, and calculate the error of the needle path Among them, E needle (t) is the needle path error at time t, which indicates the deviation between the actual position of the needle path and the planned path, in meters, P needle (t) is the needle track position measured in real time, C path (t) is the expected position of the planned path at time t, is the unit vector of the needle track direction measured in real time, is the direction unit vector of the planned path at time t, λ1 is the weight coefficient used to balance the position error and direction error; If the needle path error E needle If (t) exceeds the set tolerance range, the feedback control algorithm is used to calculate the offset that needs to be adjusted and perform needle path correction. The calculation formula is δP(t) = K adjust ·E needle (t), δP(t) is the adjustment offset of the needle track at time t, K adjust To adjust the gain matrix, E represents the ratio and direction of needle path correction. needle (t) is the needle track error; Update the needle entry point and needle tip position of the ablation needle to obtain a new needle track configuration Among them, P tip,new (t) is the updated needle tip position, P entry,new (t) is the updated needle entry point position, d(t) is the effective length of the current ablation needle, is the unit vector of the current needle track direction; Update the needle path so that the target position of the needle path covers the ablation coverage area. The update formula of the target position is P target =C target (t) = C path (t)+δC(t), where P target is the target position of the ablation needle, indicating the ablation coverage area where the tumor is located, C target (t) is the target position path, δC(t) is the path offset, which indicates the adjustment of the planned path.

7. The CT big data driven tumor ablation decision method according to claim 6, characterized in that: The managing the needle tip and the needle entry point of the ablation needle and controlling the needle path to conform to the planned path during the operation also includes: A constrained optimization algorithm is used to ensure that the movement of the needle tip of the ablation needle does not exceed the safety range of the ablation coverage area. The objective function is: Among them, E needle (t) is the error function, Ω is the constraint area, which represents the effective space in which the needle track of the ablation needle can move, and is the safe distance between the ablation coverage area and the healthy tissue area.

8. The CT big data driven tumor ablation decision method according to claim 1, characterized in that: The method of calculating, generating and updating the ablation thermal field model according to the real-time ablation parameters through the heat conduction equation and dynamically displaying the ablation thermal field distribution includes: Real-time ablation parameters include ablation power P ablation , ablation needle length L needle , ablation needle position (x, y, z) and prescription temperature T prescription ; Establish a heat conduction equation to describe the temperature change and thermal field distribution in the tumor area Among them, ρ is the density of the medium, c is the specific heat capacity of the medium, T is the temperature, t is the time, k is the thermal conductivity, and Q is the heat source term; According to the heat conduction equation, combined with real-time ablation parameters and the patient's anatomical information, the heat conduction equation is solved by numerical calculation method to obtain the distribution model of the ablation thermal field. Among them, T ambient is the ambient temperature, P ablation (i) is the power of the i-th ablation needle, r is the radial distance from the ablation needle, σ i is the standard deviation of ablation heat source diffusion, N is the number of ablation needles; According to the real-time monitoring data, the ablation thermal field model T(x,y,z,t new )=T(x,y,z,t old )+ΔT(x,y,z,Δt), where ΔT(x,y,z,Δt) is the temperature change within the time step Δt; The calculated ablation thermal field model is visualized to display the ablation effect of the tumor area in real time.

9. The CT big data driven tumor ablation decision method according to claim 8, characterized in that: The calculating of the ablation index and adjusting the ablation parameter or puncture path according to the ablation index includes: Calculating coverage It is used to indicate the proportion of the ablation coverage area covering the tumor target position, wherein TA is the ablation coverage area, A is the coverage range of the tumor target position, and |·| is the area or volume of the region; Calculate the non-ablation rate U = (1-C) × 100%; Calculated ablation rate It is used to indicate the proportion of the ablation coverage area beyond the tumor target position and affecting normal tissue, wherein NT is the healthy tissue area beyond the tumor target position; Setting an objective function that minimizes the sum of the non-ablation rate U and the over-ablation rate O; Setting a constraint condition that the coverage rate C is not less than a preset value; Using a discrete optimization algorithm to adjust the real-time ablation parameters and optimize the path so that the objective function is minimized while satisfying the constraint conditions; According to the optimization result, the real-time ablation parameters are updated, and the puncture path is adjusted based on the optimal solution.

10. A tumor ablation decision system driven by CT big data, characterized in that: include: A data set extraction module is used to obtain the patient's CT image data, perform preprocessing, and use image segmentation and feature extraction technology to extract an anatomical structure feature identification set from the CT image data; A puncture path connection line calculation module is used to calculate possible puncture path connection lines for the extracted anatomical structure feature identification set to obtain a candidate path space set; The optimal space set module is used to set the constraints of the puncture path and screen the optimal candidate path space set; The puncture path optimal solution calculation module is used to calculate the ablation target position using a discrete optimization algorithm to obtain the optimal solution of the puncture path; The needle track control module is used to manage the needle tip and entry point of the ablation needle and control the needle track to conform to the planned path during the operation; A thermal field distribution display module is used to calculate, generate and update the ablation thermal field model according to the real-time ablation parameters through the heat conduction equation, and dynamically display the ablation thermal field distribution; The adjustment module is used to calculate ablation indicators and adjust ablation parameters or puncture paths according to the ablation indicators.

Images