AI-guided puncture robot microwave ablation instrument precision positioning and treatment system

Abstract

This application discloses an AI-guided microwave ablation device for precise positioning and treatment, which relates to the field of data analysis technology. Based on the comprehensive analysis of the geometric feature data of the lesion corresponding to each planned puncture path, the displacement state data of the lesion during the respiratory displacement change measurement time, and the vascular distribution state data, the optimal puncture path is obtained. The system comprehensively acquires and quantifies the geometric feature complexity of the lesion, the dynamic deviation of respiratory displacement, and the fine vascular distribution state. On this basis, multi-dimensional fusion analysis is performed to achieve the analysis from the initial planning of the puncture path to the selection of the best puncture path. Finally, the optimal puncture path that comprehensively considers safety, feasibility, and stability is recommended to improve the accuracy, safety, and success rate of the microwave ablation device.

Description

Technical Field

[0001] This application falls under the field of data analysis, specifically an AI-guided puncture robot microwave ablation device for precise positioning and treatment. Background Technology

[0002] Percutaneous interventional procedures (such as biopsy, ablation, and particle implantation) are key methods in tumor diagnosis and treatment, and their success rate and safety highly depend on the precise planning and selection of the puncture path. Traditional puncture path planning methods are mainly based on static medical images, focusing on basic geometric constraints such as minimizing the straight-line distance and avoiding visible large blood vessels and vital organs. However, such methods have significant limitations when dealing with complex clinical scenarios.

[0003] The morphology (such as irregularity, lobulation, internal necrotic areas) and size of lesions directly affect the difficulty of puncture and the adequacy of treatment. Existing methods usually do not quantitatively assess the complexity of lesion geometric features, making it difficult to quantify the impact of different pathways on the ease of operation and coverage in complex multi-pathway selection.

[0004] Lesions in the chest and abdomen are significantly affected by respiratory movements, and their location can dynamically shift during the respiratory cycle. Traditional static planning cannot effectively reflect this dynamic change. Although technologies such as 4D-CT or respiratory gating can acquire motion information, there is a lack of systematic assessment methods in the path planning stage for how to quantitatively evaluate the displacement amplitude, trajectory stability (deviation), and potential risks to puncture accuracy of lesions along different preset paths.

[0005] Avoiding blood vessels is a core requirement for preventing bleeding complications. Existing methods can identify major blood vessels, but lack a refined quantitative assessment of the distribution density and location of small blood vessels around the lesion and in the area through which the path passes. In particular, the location of blood vessels can change under the influence of respiratory movements, and static blood vessel avoidance strategies are risky.

[0006] An ideal puncture path requires an optimal balance among multiple interdependent factors such as geometric accessibility, respiratory motion stability, and vascular safety. Current methods rely heavily on physicians' subjective assessment of a single or few factors, lacking an objective decision support system that can uniformly quantify and comprehensively analyze multi-dimensional data such as lesion geometric complexity, respiratory displacement deviation, and vascular distribution, and automatically recommend the optimal cascade path.

[0007] To address the problems raised in this background, this application presents an AI-guided puncture robot microwave ablation device for precise positioning and treatment. Summary of the Invention

[0008] To address the aforementioned technical shortcomings, this invention proposes an AI-guided puncture robot microwave ablation device for precise positioning and treatment. By comprehensively acquiring and quantifying the geometric complexity of the lesion, the dynamic deviation of respiratory displacement, and the fine vascular distribution, and based on this, multi-dimensional fusion analysis is performed to analyze from the initial planned path to the selection of the optimal path, and finally recommend the optimal puncture path that comprehensively considers safety, feasibility, and stability, thereby improving the accuracy, safety, and success rate of the microwave ablation device.

[0009] To achieve the above objectives, the present invention adopts the following technical solution: This application provides an AI-guided puncture robot microwave ablation device for precise positioning and treatment, comprising:

[0010] The lesion feature information acquisition module is used to acquire the geometric feature data of the lesion corresponding to each planned puncture path, the displacement state data of the lesion respiratory displacement change measurement time, and the blood vessel distribution state data.

[0011] The geometric feature complexity assessment module is used to assess the geometric feature complexity of lesions corresponding to each planned puncture path based on the lesion geometric feature data and lesion volume data.

[0012] The respiratory displacement deviation assessment module is used to assess the respiratory displacement deviation of lesions corresponding to each planned puncture path based on the state data of the respiratory displacement change measurement time of each planned puncture path.

[0013] The preferred puncture path analysis module is used to obtain each preferred puncture path based on the lesion geometric feature complexity assessment value and lesion respiratory displacement deviation assessment value corresponding to each planned puncture path.

[0014] The module for evaluating the impact of each preferred puncture path is used to evaluate the impact of each preferred puncture path based on the evaluation results of the geometric complexity of the lesion features, the evaluation results of the respiratory displacement deviation, and the vascular distribution data of each planned puncture path.

[0015] The optimal path analysis and recommendation module is used to obtain the optimal puncture path based on the impact assessment results of each preferred puncture path.

[0016] Specifically, the process of obtaining the geometric feature data of the lesion, the displacement state data of the lesion during the respiratory displacement measurement time, and the vascular distribution state data corresponding to each planned puncture path includes:

[0017] Geometric feature data of lesions corresponding to each planned puncture path were obtained by using 3D box counting method and database on 3D lesion model. The geometric feature data of lesions corresponding to each planned puncture path includes fractal dimension data and lesion volume data of lesions corresponding to each planned puncture path.

[0018] The state data of the lesion respiratory displacement change within the measurement time of each planned puncture path were obtained by collecting the 3D displacement curve of the lesion with respect to the respiratory cycle in each planned puncture path.

[0019] The vascular distribution data corresponding to each planned puncture path is obtained by processing the vascular contrast CT / MRI images. The vascular distribution data corresponding to each planned puncture path includes vascular distribution density data and vascular distribution distance data.

[0020] The acquired data is stored in a storage component for use in the analysis process.

[0021] Specifically, the process of evaluating the geometric feature complexity of lesions corresponding to each planned puncture path based on the lesion geometric feature data and lesion volume data includes: evaluating the geometric feature complexity of lesions corresponding to each planned puncture path based on the lesion fractal dimension data and lesion volume data, wherein the calculation formula for evaluating the geometric feature complexity of the lesion corresponding to the z-th planned puncture path is as follows: Where z is the number corresponding to each planned puncture path, z is any term from 1 to Y, and lg is the logarithm to the base 10. For the fractal dimension data of the lesion corresponding to the z-th planned puncture path, The maximum fractal dimension is V, where V is the lesion volume. This is a reference value for lesion volume. , These represent the fractal dimension of the lesion and the volume proportion weight, respectively. It should be noted that fractal dimension is a measure of the complexity of an object's shape; in this formula, it quantifies the complexity of the lesion boundary. The purpose of this setting is to compare the fractal characteristics of lesions in different planned puncture paths and to perform logarithmic normalization on the fractal dimension of the lesions. This helps to reduce the impact of differences in fractal dimension caused by different scales and calculation methods on the geometric complexity of lesions corresponding to different planned puncture paths. The purpose of this section is to quantify the impact of lesion volume on the geometric complexity of the lesion by comparing the lesion volume with the reference lesion volume.

[0022] Specifically, the process of evaluating the deviation of lesion respiratory displacement corresponding to each planned puncture path based on the state data during the measurement time of respiratory displacement change of lesion corresponding to each planned puncture path includes: evaluating the deviation of lesion respiratory displacement corresponding to each planned puncture path based on the state data during the measurement time of respiratory displacement change of lesion corresponding to each planned puncture path, wherein the calculation formula for the evaluation value of the respiratory displacement deviation of the z-th lesion is: ,in, The start time for measuring respiratory displacement changes at the lesion site. The measurement end time is the respiratory displacement change at the lesion site, and sin is the sine function. Let be the time-varying characteristic of the respiratory amplitude corresponding to the z-th planned puncture path at time t. , This represents the maximum displacement of the lesion under deep breathing conditions corresponding to the z-th planned puncture path. This is the respiratory attenuation coefficient. Let e ​​be the minimum displacement during the respiratory cycle corresponding to the z-th planned puncture path, where e is an exponential function, t is the time variable, and T is the respiratory cycle. Let E(t) be the lesion origin offset during inspiration corresponding to the z-th planned puncture path, E(t) be the equipment measurement deviation at time t, and dt be the time integral. For reference displacement deviation, it should be noted that in this formula... Its function is to quantify the rate at which breathing intensity changes from deep to shallow breathing, where T is the time required to complete one full breath. The phase shift is used to correct for individual differences in breathing patterns; the integral part in this formula... The system captures the continuous cumulative changes in lesion displacement corresponding to each planned puncture path during respiration, simulating the periodic movement of the lesion during respiration. The formula for assessing lesion respiratory displacement deviation is designed to adapt to the attenuation of individual respiratory intensity and individual baseline differences. The equipment error term E(t) directly compensates for the inherent deviation of the measuring equipment, avoiding underestimation of the actual lesion displacement. The equipment error term E(t) is obtained by using a discrete model of E(t). The process of constructing the discrete model of E(t) is as follows: repeatability tests are performed to simulate the respiratory process in a specific time interval, the deviation data of the equipment at different time points are recorded, multiple registration experiments are performed on the recorded results, the rate of change of error over time is quantified, a time-dependent error curve is generated, and the output result is a time series dataset, thus constructing the discrete model of E(t).

[0023] Specifically, the process of obtaining each preferred puncture path based on the lesion geometric feature complexity assessment value and lesion respiratory displacement deviation assessment value corresponding to each planned puncture path includes: weighting and summing the lesion geometric feature complexity assessment value and lesion respiratory displacement deviation assessment value corresponding to each planned puncture path to obtain the evaluation value of each preferred puncture path; comparing the obtained evaluation value of each preferred puncture path with a set preferred puncture path evaluation value threshold; if the evaluation value of a certain preferred puncture path is greater than or equal to the set preferred puncture path evaluation value threshold, then the corresponding preferred puncture path is judged as a non-preferred puncture path. If the evaluation value of a preferred puncture path is less than the set threshold for the evaluation value of a preferred puncture path, then the corresponding preferred puncture path is judged as a preferred puncture path, and so on, to obtain each preferred puncture path. It should be noted that the evaluation value of each preferred puncture path is obtained by weighting and summing the evaluation values ​​of the lesion geometric feature complexity and the lesion respiratory displacement deviation corresponding to each planned puncture path. The weighted summation method is used to comprehensively evaluate the difficulty corresponding to the lesion geometric feature complexity evaluation value and the risk corresponding to the respiratory displacement deviation evaluation value to obtain each preferred puncture path, thereby improving the accuracy of each preferred puncture path.

[0024] Specifically, the process of evaluating the impact of each preferred puncture path based on the assessment results of the geometric complexity of the lesion features of each planned puncture path, the assessment results of respiratory displacement deviation, and the vascular distribution data of each planned puncture path includes:

[0025] The geometric feature complexity assessment results of lesions in each planned puncture path, the respiratory displacement deviation assessment results, and the vascular distribution status data of each planned puncture path were obtained from the lesion geometric feature complexity assessment results, respiratory displacement deviation assessment results, and vascular distribution status data of each planned puncture path.

[0026] Based on the assessment results of the geometric feature complexity of the lesions along each preferred puncture path, the assessment results of respiratory displacement deviation, and the vascular distribution density and vascular distribution distance data of each preferred puncture path, the impact of each preferred puncture path is assessed. The formula for calculating the impact assessment value of the i-th preferred puncture path is as follows: Where i is the number corresponding to each preferred puncture path, i can be any one from 1 to N, and f is the number corresponding to each estimated measurement point of the puncture path, f can be any one from 1 to R. This represents the respiratory displacement deviation assessment value for the lesion corresponding to the i-th preferred puncture path. Let be the geometric feature complexity evaluation value of the lesion corresponding to the i-th preferred puncture path. Here, c represents the vascular distribution density data corresponding to the f-th estimated measurement point along the i-th puncture path, and c represents the reference vascular distribution density. The distance from the i-th preferred puncture path and the f-th estimated measurement point to the adjacent blood vessel is the average distance, where h is the reference distance. It should be noted that in this formula... The reason for this setting is that the risk added by the difficulty of the lesion corresponding to each preferred puncture path cannot be ignored. (Formula) Part of the method uses the average value of the deviations between the vascular distribution density and distance of all measurement points on each preferred puncture path and the reference value as a standard to measure the influence of each preferred puncture path. At the same time, it considers the influence of respiratory displacement deviation of each preferred puncture path, which improves the accuracy of the influence assessment value of each preferred puncture path. The influence of the geometric feature complexity of each preferred puncture path is partially incorporated into the impact assessment of each preferred puncture path. The higher the complexity, the greater the impact assessment value of each puncture path.

[0027] Specifically, the process of obtaining the optimal puncture path based on the impact assessment results of each preferred puncture path includes: obtaining the impact assessment results of each preferred puncture path, arranging the impact assessment values ​​of each preferred puncture path in ascending order, and taking the preferred puncture path corresponding to the impact assessment result of the first preferred puncture path as the optimal puncture path.

[0028] Compared with existing technologies, the beneficial effects of this application are: acquiring lesion geometric feature data, lesion respiratory displacement change measurement time displacement state data, and vascular distribution state data corresponding to each planned puncture path; comprehensively analyzing the lesion geometric feature data, lesion respiratory displacement change measurement time displacement state data, and vascular distribution state data corresponding to each planned puncture path to obtain the optimal puncture path; comprehensively acquiring and quantitatively evaluating the geometric feature complexity, respiratory displacement dynamic deviation, and fine vascular distribution state of the lesion; and on this basis, performing multi-dimensional fusion analysis to achieve the analysis from preliminary planning of puncture path to screening and selection of optimal puncture path, and finally recommending the optimal puncture path that comprehensively considers safety, feasibility, and stability, thereby improving the accuracy, safety, and success rate of microwave ablation device. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the overall process of the AI-guided puncture robot microwave ablation device for precise positioning and treatment in this application.

[0030] Figure 2 This is a schematic diagram of the overall framework of the AI-guided puncture robot microwave ablation device for precise positioning and treatment in this application.

[0031] Figure 3 This is a schematic diagram illustrating the process of obtaining the optimal puncture path for the AI-guided puncture robot microwave ablation device precision positioning and treatment system of this application. Detailed Implementation

[0032] To better understand this application, various aspects of this application will be described in more detail with reference to the accompanying drawings.

[0033] To address the technical problems raised in the background art, this application provides a preferred embodiment:

[0034] The specific content of this embodiment is as follows:

[0035] like Figure 1 and Figure 2 As shown, one embodiment of the present invention provides an AI-guided puncture robot microwave ablation device for precise positioning and treatment, comprising:

[0036] The lesion feature information acquisition module is used to acquire the geometric feature data of the lesion corresponding to each planned puncture path, the displacement state data of the lesion respiratory displacement change measurement time, and the blood vessel distribution state data.

[0037] In this embodiment, the specific process of obtaining the geometric feature data of the lesion, the displacement state data of the lesion during the respiratory displacement change measurement time, and the vascular distribution state data corresponding to each planned puncture path is as follows:

[0038] 3D volumetric data of the lesion were obtained through preoperative enhanced CT or high-resolution MRI. AI segmentation tools were used to delineate the lesion outline and generate a 3D model. The fractal dimension of the lesion was obtained using the 3D box counting method: the 3D space was divided into a cubic mesh with side length s, and the number of cubes Q(s) containing at least one lesion volume pixel was counted. s was gradually decreased (e.g., s = 32, 16, 8, 4, 2, 1 volume pixels), and this step was repeated. A double logarithmic curve was fitted: lg(Q(s)) ~ lg(1 / s). The formula for calculating the fractal dimension D is: ; Geometric feature data of lesions corresponding to each planned puncture path were obtained by using 3D box counting method and database on 3D lesion model. The geometric feature data of lesions corresponding to each planned puncture path includes fractal dimension data and lesion volume data of lesions corresponding to each planned puncture path.

[0039] Continuous scanning was performed while the patient was breathing freely. 3D image sequences were reconstructed according to respiratory phases. The displacement of the lesion was tracked in real time using a fast sequence. The motion trajectory was captured in real time using an electromagnetic positioning system. The CT / MRI images of different respiratory phases were aligned using an elastic matching tool. The displacement vector field of the lesion was calculated. Abdominal pressure sensor data and image timestamps were synchronized. A mapping model between respiratory phase and lesion position was established. The 3D displacement curve of the lesion with respect to the respiratory cycle in each planned puncture path was output. The state data of the lesion respiratory displacement change within the measurement time corresponding to each planned puncture path were obtained by collecting the 3D displacement curve of the lesion with respect to the respiratory cycle in each planned puncture path.

[0040] CT / MRI images of blood vessels were obtained by scanning during the arterial and venous phases. Contrast agents were used to enhance the contrast of blood vessels in the CT / MRI images. The enhanced CT / MRI images were then processed: the vessel centerlines were extracted using region growing and level set algorithms to segment the vessels; ROI buffers were set around the puncture path to calculate vessel density; and a distance transform algorithm was used to extract the vessel distribution distance values. The processed CT / MRI images were then used to obtain the vessel distribution status data corresponding to each planned puncture path, including vessel density data and vessel distribution distance data for each planned puncture path.

[0041] The acquired data is stored in a storage component for use in the analysis process.

[0042] The geometric feature complexity assessment module is used to assess the geometric feature complexity of lesions corresponding to each planned puncture path based on the lesion geometric feature data and lesion volume data.

[0043] In this embodiment, the specific process of evaluating the geometric feature complexity of the lesion corresponding to each planned puncture path based on the lesion geometric feature data and lesion volume data includes: evaluating the geometric feature complexity of the lesion corresponding to each planned puncture path based on the lesion fractal dimension data and lesion volume data, wherein the calculation formula for evaluating the geometric feature complexity of the lesion corresponding to the z-th planned puncture path is as follows: Where z is the number corresponding to each planned puncture path, z is any term from 1 to Y, and lg is the logarithm to the base 10. For the fractal dimension data of the lesion corresponding to the z-th planned puncture path, The maximum fractal dimension is V, where V is the lesion volume. This is a reference value for lesion volume. , These are the fractal dimension of the lesion and the volume proportion weight, respectively. It should be noted that the fractal dimension is a measure used to describe the complexity of the shape of an object. In this formula, the fractal dimension is used to quantify the complexity of the lesion boundary. The purpose of this setting is to quantify the complexity of the lesion boundary corresponding to each planned puncture path; the higher the value, the more complex the lesion morphology. The purpose of setting is to be the maximum value of the fractal dimension of the lesion, serving as an upper limit reference for the evaluation of the fractal dimension of the lesion, and ensuring that the evaluation results are within a reasonable range; the purpose of setting V is to directly reflect the size of the lesion and is an important indicator for evaluating the geometric complexity of the lesion. The purpose of this setting is to provide a reference value for lesion volume, which is used to standardize lesion volume and facilitate comparison between different lesions. , The purpose of this setting is to balance the influence of lesion fractal dimension and volume on the assessment of lesion geometric complexity, ensuring the comprehensiveness of the assessment results; in this formula The purpose of this setting is to compare the fractal features of lesions in different planned puncture paths and to normalize the fractal dimension of the lesions. This helps to reduce the impact of differences in fractal dimension caused by different scales and calculation methods on the geometric complexity of lesions corresponding to different planned puncture paths. In this formula... The purpose of this section is to quantify the impact of lesion volume on the geometric complexity of lesions by comparing the lesion volume with the reference lesion volume. The benefits and basis of logarithmic normalization of the fractal dimension in this formula are as follows: After logarithmic normalization, the calculated results of the fractal dimension for lesions corresponding to different planned puncture paths can compress the differences in the assessment of lesion geometric complexity into a smoother trend, facilitating the analysis and calculation of the assessment values ​​of lesion geometric complexity corresponding to different planned puncture paths. The larger the fractal dimension of the lesion, the higher the complexity of the lesion fractal shape at various scales. The fractal dimensions of traditional geometric shapes, such as line segments, planes, and solids, are 1, 2, and 3, respectively. However, the dimension of the lesion fractal shape is usually non-integer and lies between these integer dimensions. The logarithmically normalized fractal dimension can more accurately describe the non-integer dimension, thereby reflecting the complexity of the lesion geometric features corresponding to different planned puncture paths. An example is provided to illustrate this. and Settings: Reasonable and Setting parameters can improve diagnostic accuracy and better quantify the difficulty of lesion treatment under different puncture paths. It is 0.7. When the value is 0.3, it means that the fractal dimension of the lesion has a greater impact on the evaluation results than the proportion of lesion volume. This indicates that the complexity of the lesion morphology has a greater impact on the evaluation results. When evaluating each planned puncture path, more emphasis is placed on the complexity of the lesion boundary. In other words, the more complex the lesion morphology, the higher its corresponding evaluation value may be.

[0044] The respiratory displacement deviation assessment module is used to assess the respiratory displacement deviation of lesions corresponding to each planned puncture path based on the state data of the respiratory displacement change measurement time of each planned puncture path.

[0045] In this embodiment, the specific process of evaluating the deviation of lesion respiratory displacement corresponding to each planned puncture path based on the state data during the measurement time of respiratory displacement change of lesion corresponding to each planned puncture path includes: evaluating the deviation of lesion respiratory displacement corresponding to each planned puncture path based on the state data during the measurement time of respiratory displacement change of lesion corresponding to each planned puncture path, wherein the calculation formula for the evaluation value of the respiratory displacement deviation of the z-th lesion is: ,in, The start time for measuring respiratory displacement changes at the lesion site. The end time of measurement of respiratory displacement changes at the lesion. Let be the time-varying characteristic of the respiratory amplitude corresponding to the z-th planned puncture path at time t. , This represents the maximum displacement of the lesion under deep breathing conditions corresponding to the z-th planned puncture path. This is the respiratory attenuation coefficient. Let e ​​be the minimum displacement during the respiratory cycle corresponding to the z-th planned puncture path, where e is an exponential function, t is the time variable, and T is the respiratory cycle. Let E(t) be the offset of the lesion origin during inspiration corresponding to the z-th planned puncture path, E(t) be the equipment measurement deviation at time t, and dt be the time integral. For reference displacement deviation, it should be noted that in this formula... Its function is to quantify the rate at which breathing intensity changes from deep to shallow breathing, where T is the time required to complete one full breath. The phase shift is used to correct for individual differences in breathing patterns; the integral part in this formula... The system captures the continuous cumulative changes in lesion displacement corresponding to each planned puncture path during respiration, simulating the periodic movement of the lesion during respiration. This ensures that the formula for assessing lesion respiratory displacement deviation can adapt to the decay of individual respiratory intensity and individual baseline differences. The equipment error term E(t) directly compensates for the inherent deviation of the measuring equipment, avoiding underestimation of the actual lesion displacement. The equipment error term E(t) is obtained through a discrete model of E(t). The process of constructing the discrete model of E(t) is as follows: repeatability tests are performed to simulate the respiratory process in a specific time interval, the deviation data of the equipment at different time points are recorded, multiple registration experiments are conducted on the recorded results, the rate of change of error over time is quantified, a time-dependent error curve is generated, and the output result is a time series dataset, thus constructing the discrete model of E(t). The reasons for the settings of each part of this formula are: the sine function simulates the inherent rhythmicity of individual respiration, and the period parameter T and To ensure that the assessment of respiratory displacement deviation of lesions can be dynamically adjusted to adapt to the respiratory rate and behavioral patterns of different individuals, while the error term E(t) addresses the limitations of equipment in practical applications, such as noise in optical tracking, thereby enhancing the robustness of the assessment of respiratory displacement deviation of lesions.

[0046] The preferred puncture path analysis module is used to obtain each preferred puncture path based on the lesion geometric feature complexity assessment value and lesion respiratory displacement deviation assessment value corresponding to each planned puncture path.

[0047] In this embodiment, the specific process of obtaining each preferred puncture path based on the lesion geometric feature complexity assessment value and lesion respiratory displacement deviation assessment value corresponding to each planned puncture path includes: weighting and summing the lesion geometric feature complexity assessment value and lesion respiratory displacement deviation assessment value corresponding to each planned puncture path to obtain the evaluation value of each preferred puncture path; comparing the obtained evaluation value of each preferred puncture path with a set preferred puncture path evaluation value threshold; if the evaluation value of a certain preferred puncture path is greater than or equal to the set preferred puncture path evaluation value threshold, then the corresponding preferred puncture path is judged as a non-preferred puncture path. If the evaluation value of a preferred puncture path is less than the set threshold for the evaluation value of a preferred puncture path, then the corresponding preferred puncture path is judged as a preferred puncture path, and so on, to obtain each preferred puncture path. It should be noted that the evaluation value of each preferred puncture path is obtained by weighting and summing the evaluation values ​​of the lesion geometric feature complexity and the lesion respiratory displacement deviation corresponding to each planned puncture path. The weighted summation method is used to comprehensively evaluate the difficulty corresponding to the lesion geometric feature complexity evaluation value and the risk corresponding to the respiratory displacement deviation evaluation value to obtain each preferred puncture path, thereby improving the accuracy of each preferred puncture path.

[0048] The module for evaluating the impact of each preferred puncture path is used to evaluate the impact of each preferred puncture path based on the evaluation results of the geometric complexity of the lesion features, the evaluation results of the respiratory displacement deviation, and the vascular distribution data of each planned puncture path.

[0049] In this embodiment, the specific process of evaluating the impact of each preferred puncture path based on the assessment results of the geometric feature complexity of the lesion along each planned puncture path, the assessment results of respiratory displacement deviation, and the vascular distribution data of each planned puncture path includes:

[0050] The geometric feature complexity assessment results of lesions in each planned puncture path, the respiratory displacement deviation assessment results, and the vascular distribution status data of each planned puncture path were obtained from the lesion geometric feature complexity assessment results, respiratory displacement deviation assessment results, and vascular distribution status data of each planned puncture path.

[0051] Based on the assessment results of the geometric feature complexity of the lesions along each preferred puncture path, the assessment results of respiratory displacement deviation, and the vascular distribution density and vascular distribution distance data of each preferred puncture path, the impact of each preferred puncture path is assessed. The formula for calculating the impact assessment value of the i-th preferred puncture path is as follows: Where i is the number corresponding to each preferred puncture path, i can be any one from 1 to N, and f is the number corresponding to each estimated measurement point of the puncture path, f can be any one from 1 to R. This represents the respiratory displacement deviation assessment value for the lesion corresponding to the i-th preferred puncture path. Let be the geometric feature complexity evaluation value of the lesion corresponding to the i-th preferred puncture path. Here, c represents the vascular distribution density data corresponding to the f-th estimated measurement point of the i-th preferred puncture path, and c represents the reference vascular distribution density. Let h be the average distance from the f-th estimated measurement point of the i-th preferred puncture path to the adjacent blood vessel, and h be the reference distance. It should be noted that in this formula... Its purpose is to measure the risk impact of deviations caused by breathing on each preferred puncture path. Its purpose is to measure the complexity of the lesions corresponding to each preferred puncture path and to reflect the difficulty of each preferred puncture path. The reason for this setting is that the risk added by the difficulty of the lesion corresponding to each preferred puncture path cannot be ignored. and As core elements for measuring the risk level of each preferred puncture path due to vascular distribution density and distance, c and h are set to standardize the risk level of each preferred puncture path. Partially, by summing the deviations of the vascular distribution density and distance at all measurement points along each preferred puncture path from the reference values ​​and averaging the results, the influence of respiratory displacement deviation along each preferred puncture path is considered, thus improving the accuracy of the influence assessment values ​​for each preferred puncture path. This formula... The influence of the geometric complexity of each preferred puncture path is incorporated into the impact assessment of each preferred puncture path. The higher the complexity, the greater the impact assessment value of each puncture path. The advantage of this formula is that it comprehensively considers the vascular distribution density, distance, lesion displacement deviation caused by respiration, and lesion geometric complexity corresponding to the preferred puncture path, making the impact assessment of each preferred puncture path more comprehensive and improving the accuracy of the impact assessment value. For example, the setting method of c and h in this formula can be illustrated by summarizing the vascular distribution density data and vascular distribution distance in the lesion area separately and taking the average value of the vascular distribution density and vascular distribution distance in the lesion area to ensure the standardization of the impact assessment of each preferred puncture path.

[0052] The optimal path analysis and recommendation module is used to obtain the optimal puncture path based on the impact assessment results of each preferred puncture path.

[0053] like Figure 3As shown, in this embodiment, the specific process of obtaining the optimal puncture path based on the impact assessment results of each preferred puncture path includes: obtaining the impact assessment results of each preferred puncture path, arranging the impact assessment values ​​of each preferred puncture path in ascending order, and taking the preferred puncture path corresponding to the impact assessment result of the first preferred puncture path as the optimal puncture path.

[0054] Based on the above implementation, this embodiment has the following advantages over the prior art: This embodiment acquires the geometric feature data of the lesion, the displacement state data of the lesion during the respiratory displacement change measurement time, and the vascular distribution state data corresponding to each planned puncture path; it comprehensively analyzes the geometric feature data of the lesion, the displacement state data of the lesion during the respiratory displacement change measurement time, and the vascular distribution state data corresponding to each planned puncture path to obtain the optimal puncture path; it comprehensively acquires and quantitatively evaluates the geometric feature complexity of the lesion, the dynamic deviation of respiratory displacement, and the fine vascular distribution state; and on this basis, it performs multi-dimensional fusion analysis to achieve the analysis from the initial planning of puncture path to the selection of the best puncture path, and finally recommends the optimal puncture path that comprehensively considers safety, feasibility, and stability, so as to improve the accuracy, safety, and success rate of the microwave ablation device.

[0055] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments under the guidance of the present invention without departing from the spirit and scope of the present invention. All of these variations are within the protection scope of the present invention.

Claims

1. An AI-guided puncture robot microwave ablation device for precise positioning and treatment, characterized in that: include: The lesion feature information acquisition module is used to acquire the geometric feature data of the lesion corresponding to each planned puncture path, the displacement state data of the lesion respiratory displacement change measurement time, and the blood vessel distribution state data. The specific process includes: Geometric feature data of lesions corresponding to each planned puncture path are obtained through 3D model of the lesion. The geometric feature data of lesions corresponding to each planned puncture path includes fractal dimension data and lesion volume data of the lesion corresponding to each planned puncture path. The state data of the lesion respiratory displacement change within the measurement time of each planned puncture path were obtained by collecting the 3D displacement curve of the lesion with respect to the respiratory cycle in each planned puncture path. The vascular distribution data corresponding to each planned puncture path is obtained by processing the vascular contrast CT / MRI images. The vascular distribution data corresponding to each planned puncture path includes vascular distribution density data and vascular distribution distance data. The geometric feature complexity assessment module is used to assess the geometric feature complexity of lesions corresponding to each planned puncture path based on the lesion geometric feature data and lesion volume data. The specific process includes: evaluating the geometric feature complexity of the lesions corresponding to each planned puncture path based on the fractal dimension data and lesion volume data of each planned puncture path. The formula for calculating the geometric feature complexity of the lesion corresponding to the z-th planned puncture path is as follows: Where z is the number corresponding to each planned puncture path, z is any term from 1 to Y, and lg is the logarithm to the base 10. For the fractal dimension data of the lesion corresponding to the z-th planned puncture path, The maximum fractal dimension is V, where V is the lesion volume. This is a reference value for lesion volume. , These are the fractal dimension and volume percentage weights of the lesion, respectively. The respiratory displacement deviation assessment module is used to assess the respiratory displacement deviation of lesions corresponding to each planned puncture path based on the state data of the respiratory displacement change measurement time of each planned puncture path. The specific process includes: assessing the deviation of respiratory displacement of lesions corresponding to each planned puncture path based on the state data of respiratory displacement change within the measurement time corresponding to each planned puncture path. The formula for calculating the respiratory displacement deviation assessment value of the z-th lesion is as follows: ,in, The start time for measuring respiratory displacement changes at the lesion site. The end time of measurement of respiratory displacement changes at the lesion. Let sin be the sine function, and T be the respiratory cycle, to describe the time-varying characteristics of the respiratory amplitude at time t corresponding to the z-th planned puncture path. Let E(t) be the offset of the lesion origin during inspiration corresponding to the z-th planned puncture path, and E(t) be the equipment measurement deviation at time t. Let dt be the reference displacement deviation, and dt be the integral over time. The preferred puncture path analysis module is used to obtain each preferred puncture path based on the lesion geometric feature complexity assessment value and lesion respiratory displacement deviation assessment value corresponding to each planned puncture path. The module for evaluating the impact of each preferred puncture path is used to evaluate the impact of each preferred puncture path based on the evaluation results of the geometric complexity of the lesion features, the evaluation results of the respiratory displacement deviation, and the vascular distribution data of each planned puncture path. The optimal path analysis and recommendation module is used to obtain the optimal puncture path based on the impact assessment results of each preferred puncture path.

2. The AI-guided puncture robot microwave ablation device for precise positioning and treatment as described in claim 1, characterized in that, The specific process of obtaining each preferred puncture path based on the lesion geometric feature complexity assessment value and lesion respiratory displacement deviation assessment value corresponding to each planned puncture path includes: weighting and summing the lesion geometric feature complexity assessment value and lesion respiratory displacement deviation assessment value corresponding to each planned puncture path to obtain the evaluation value of each preferred puncture path; comparing the obtained evaluation value of each preferred puncture path with a set preferred puncture path evaluation value threshold; if the evaluation value of a preferred puncture path is greater than or equal to the set preferred puncture path evaluation value threshold, then the corresponding preferred puncture path is judged as a non-preferred puncture path; if the evaluation value of a preferred puncture path is less than the set preferred puncture path evaluation value threshold, then the corresponding preferred puncture path is judged as a preferred puncture path, and so on, to obtain each preferred puncture path.

3. The AI-guided puncture robot microwave ablation device for precise positioning and treatment as described in claim 2, characterized in that, The specific process of evaluating the impact of each preferred puncture path based on the assessment results of the geometric complexity of the lesion features, the assessment results of respiratory displacement deviation, and the vascular distribution data of each planned puncture path includes: The geometric feature complexity assessment results, respiratory displacement deviation assessment results, and vascular distribution status data of each preferred puncture path were obtained from the lesion geometric feature complexity assessment results, respiratory displacement deviation assessment results, and vascular distribution status data of each planned puncture path. The impact of each preferred puncture path was assessed based on the results of the evaluation of the geometric feature complexity of the lesion, the evaluation of the respiratory displacement deviation, the vascular distribution density data and the vascular distribution distance data of each preferred puncture path.

4. The AI-guided puncture robot microwave ablation device for precise positioning and treatment as described in claim 3, characterized in that, The specific process of obtaining the optimal puncture path based on the impact assessment results of each preferred puncture path includes: obtaining the impact assessment results of each preferred puncture path, arranging the impact assessment values ​​of each preferred puncture path in ascending order, and taking the preferred puncture path corresponding to the impact assessment result of the first preferred puncture path as the optimal puncture path.

5. The AI-guided puncture robot microwave ablation device for precise positioning and treatment as described in claim 1, characterized in that, The specific formula for obtaining the time-varying characteristics of the respiratory amplitude corresponding to the z-th planned puncture path is as follows: ,in, This represents the maximum displacement of the lesion under deep breathing conditions corresponding to the z-th planned puncture path. This is the respiratory attenuation coefficient. Let be the minimum displacement during the respiratory cycle corresponding to the z-th planned puncture path, e be an exponential function, and t be a time variable.

Images