A light-visual-guided puncture navigation method and system

Abstract

This invention discloses a vision-guided puncture navigation method and system. The method includes: defining a hyperbolic optical projection surface equation in a non-uniform optical field model; performing multimodal optical marker tracking to obtain an optimized hyperbolic optical projection surface equation; constructing a path functional under multiple constraints in the non-uniform optical field model based on a preset puncture target and real-time obstacle information; solving the path functional using an improved gradient descent algorithm to generate a puncture path with optimal characteristics; and constructing a feedback correction quantity to correct the puncture projection based on the error between the actual needle insertion trajectory and the planned puncture path. The system includes an optical field model construction module, a multimodal optical marker tracking module, a path functional construction module, a puncture path calculation module, and a puncture projection correction module. This invention can construct an accurate optical field model, combined with an optimized path planning algorithm and adaptive feedback, to achieve high-precision and stable puncture navigation.

Description

Technical Field

[0001] This invention relates to the field of puncture navigation technology, and in particular to a puncture navigation method and system guided by optical vision. Background Technology

[0002] Puncture navigation technology has wide applications in precision operations (such as industrial manufacturing, minimally invasive procedures, and robot guidance). Its core objective is to optimize puncture paths in complex environments, improve puncture accuracy, and reduce the risk of misoperation due to path deviations. Traditional puncture navigation methods mainly rely on image guidance, mechanical control, and sensor feedback. However, these methods still face many challenges in path optimization, real-time feedback, and environmental adaptability, limiting their widespread application.

[0003] Current puncture navigation methods can be mainly classified into the following categories:

[0004] (1) Image-guided puncture navigation uses CT, MRI, or ultrasound images to image the target area and calculates the optimal puncture path based on preoperative planning. The main advantage of this method is that it provides high-resolution anatomical information, but its real-time performance is poor and there are data registration errors during image reconstruction. In addition, the imaging equipment is large and not suitable for portable or small-space operation scenarios.

[0005] (2) Puncture navigation based on robot force feedback control combines force sensors and intelligent control algorithms to optimize the puncture trajectory by adjusting the puncture force and needle depth in real time. This type of method improves path accuracy, but it is highly dependent on the hardware of the equipment, has a high cost, and suffers from error accumulation in force feedback in non-rigid media.

[0006] (3) Puncture navigation based on optical tracking obtains the relative position between the device and the target through markers and optical sensors, and realizes path planning and real-time adjustment. This method has the characteristics of non-contact and strong real-time performance, but it still faces technical bottlenecks in terms of ambient light interference, marker occlusion and optical field distortion compensation.

[0007] The existing image-guided methods suffer from coordinate registration errors during data acquisition and processing, especially during multimodal fusion, where insufficient image distortion and spatial transformation accuracy lead to puncture path deviations. Traditional path planning methods typically employ geometric models or heuristic algorithms, lacking adaptability to complex environments and struggling to calculate optimal paths under multiple constraints. Furthermore, the lack of real-time feedback mechanisms for dynamic environmental changes results in lag in path adjustments, impacting operational stability. Due to the non-uniform distribution of the light field and interference from ambient light, the tracking accuracy of optical markers is limited, particularly in optical projection calculations on curved or irregular surfaces, where significant errors occur, affecting the stability and reliability of puncture navigation. Most existing puncture navigation systems perform punctures based on preset paths, but lack intraoperative error feedback and adaptive correction mechanisms, causing the puncture path to deviate from the planned trajectory during actual operation, affecting the final operational accuracy. Summary of the Invention

[0008] In view of this, the purpose of this invention is to provide a puncture navigation method and system guided by optical vision, which can overcome the shortcomings of the prior art in terms of puncture accuracy, environmental adaptability and real-time feedback by optimizing the optical vision-guided path planning and real-time correction mechanism, and provide a more efficient and accurate navigation method for the field of precision operation.

[0009] The embodiments of the present invention are implemented as follows:

[0010] A light-vision guided puncture navigation method, comprising:

[0011] Using the position data of optical markers in the device coordinate system, a non-uniform optical field model is constructed, and the hypercurvature optical projection surface equation describing the spatial light field distribution is defined in the non-uniform optical field model.

[0012] By introducing filter response constraints into the hypercurvature optical projection surface equation and performing multimodal optical marker tracking, an optimized hypercurvature optical projection surface equation is obtained.

[0013] Based on the preset puncture target and the real-time acquired obstacle information, a path functional under multiple constraints is constructed in the non-uniform optical field model.

[0014] The improved gradient descent algorithm is used to solve the path functional to generate a puncture path with optimal characteristics.

[0015] Based on the error between the actual needle insertion trajectory and the planned puncture path, a feedback correction amount is constructed to correct the puncture projection.

[0016] In a preferred embodiment of the present invention, in the above-described optical vision-guided puncture navigation method, the step of constructing a non-uniform optical field model using the position data of optical markers in the device coordinate system, and defining the hypercurvature optical projection surface equation describing the spatial light field distribution in the non-uniform optical field model, includes:

[0017] Set the equipment coordinate system ,in, For equipment center, The axis is perpendicular to the equipment surface and consistent with the puncture direction. The axis is parallel to the device surface, defining the horizontal distribution of the optical field.

[0018] The position data of the optical markers are acquired in the device coordinate system. ,in, In three-dimensional space coordinates, As the reflection intensity weight, , where n is the number of optical markers distributed on the surface of the device, and k is the number of markers.

[0019] By employing a Gaussian weighting function, discrete optical marker data is converted into a continuous light field, thus constructing a non-uniform optical field model describing the light field intensity at a spatial point (x,y,z). ,in, For depth-adaptive Gaussian kernel width, , For adjustment coefficients, This represents the maximum detection depth.

[0020] Adjusting the light field distribution along the depth direction using the hyperbolic tangent function ,in, It is the hyperbolic tangent function. Using curvature adjustment parameters, the equation of the hypercurvature optical projection surface is obtained. .

[0021] Its technical advantages are as follows: Firstly, it uses a Gaussian weighted function to convert discrete optical marker data into a continuous light field, avoiding interpolation errors caused by sparse marker points in traditional methods. This results in a smoother and more continuous light field intensity distribution, improving the accuracy of optical projection surface modeling. Secondly, it employs a depth-adaptive Gaussian kernel width, enabling the light field model to automatically adjust its resolution based on the target depth. This provides finer optical guidance in shallower regions while improving stability and reducing signal attenuation in deeper regions. Thirdly, it uses a hyperbolic tangent function to adjust the light field distribution in the depth direction. Through the nonlinear characteristics of the tanh function, the optical projection achieves higher resolution near the device's center while transitioning smoothly away from the center, reducing distortion and improving adaptability to complex human anatomy. Finally, by constructing a hyperbolic optical projection surface equation, it can more accurately describe the distribution characteristics of optical projection in a non-uniform optical field, providing precise optical gradient information for subsequent puncture path optimization.

[0022] In a preferred embodiment of the present invention, in the above-described optical vision-guided puncture navigation method, the step of introducing a filter response constraint into the hyperbolic optical projection surface equation to perform multimodal optical marker tracking and obtain an optimized hyperbolic optical projection surface equation includes:

[0023] Introducing filter response constraints into the equation of the hypercurvature optical projection surface to perform multimodal optical marker tracking. .

[0024] in, This is the correlation filter response diagram. This indicates the degree of matching between the input image and the filter template, expressed by the inverse Fourier transform. Calculated.

[0025] For filtering template The conjugate of complex numbers, As a filter template, , For the Fourier transform of the input image, For the conjugate Fourier transform of the target template, This is the regularization coefficient.

[0026] This represents the maximum matching value of the correlation filter.

[0027] is the L2 norm of the filter template, used to normalize the response intensity.

[0028] The optimized hypercurvature optical projection surface equation is obtained. .

[0029] Its technical advantages are as follows: By calculating the matching degree between the input image and the filter template through correlation filtering, a filter response map is obtained, which can accurately identify the position of optical markers in complex environments and improve the recognition accuracy of optical markers; Fourier transform and inverse transform are used to calculate the matching degree, so that marker tracking not only relies on spatial domain information, but also combines frequency domain features, reducing noise interference and improving the stability and robustness of optical markers; It is not only applicable to single-modal optical markers, but also supports multi-modal fusion, and can stably track optical markers under different lighting, tissue depth, and environmental change conditions; Through the optimized hypercurvature optical projection surface equation, the information of multiple optical markers is fused in a unified projection coordinate system, improving adaptability to complex surgical environments and ensuring the stability of optical guidance; The method of calculating the marker matching degree by correlation filtering has lower computational complexity than traditional template matching or deep learning-based methods, and is suitable for real-time puncture navigation; Due to the use of Fourier transform to accelerate the calculation, large-scale image matching can be completed in a short time, improving the response speed of optical marker tracking, enabling the intraoperative navigation system to quickly adjust the puncture path and improve surgical efficiency.

[0030] In a preferred embodiment of the present invention, in the above-described optical vision-guided puncture navigation method, the step of constructing a path functional under multiple constraints in the non-uniform optical field model based on a preset puncture target and real-time acquired obstacle information includes:

[0031] Based on the optimized hypercurvature optical projection surface equation Calculate the gradient field of the spatial light field to obtain the gradient vector field of the optical field. ,in, Reflecting the change in optical gradient in the X direction, Reflects the change in optical gradient in the Y direction. It reflects the change in optical gradient in the Z direction.

[0032] Based on the preset puncture target and real-time obstacle information, an optical gradient guidance item is set. Path curvature constraint Obstacle avoidance items and time dynamic velocity constraints ,in, The weighting factor for the optical gradient constraint. For path curvature, As the curvature control weighting factor, The distance from the current path to the nearest obstacle. For obstacle avoidance weighting factors, For time-dynamic weighting function, Let Γ be the instantaneous velocity of the puncture path at time t. This is the preset maximum permissible puncture speed.

[0033] The path functional is constructed. .

[0034] Its technical advantages are as follows: the optical gradient guidance term guides the puncture path along the upward direction of the optical gradient, directing the puncture needle toward the optimal navigation path and improving navigation accuracy. Since the optical field gradient information comes from the optimized hyperbolic optical projection surface equation, path planning can be adaptively adjusted based on the dynamic optical environment, ensuring that the puncture path always faces the target area and reducing deviations caused by tissue deformation or optical signal interference. The path curvature constraint term controls the smoothness of the path, preventing unreasonable sharp turns and making the puncture trajectory more physiologically consistent, thus improving the feasibility and safety of the surgical procedure. By adjusting the curvature control weighting factor, it can be flexibly adjusted according to surgical needs. The flexibility of the puncture path makes it suitable for different types of puncture tasks; the obstacle avoidance parameter keeps the puncture path away from obstacles, using the dynamic distance from the path to the nearest obstacle as a constraint index, making obstacle avoidance control more intelligent and able to adapt to changes in the intraoperative environment in real time, such as the patient's slight movements or tissue deformation; the time dynamic speed constraint parameter controls the puncture speed, ensuring the dynamic stability of the path and avoiding tissue damage or path tracking errors caused by excessive speed. The puncture speed is adjusted by a time dynamic weight function, so that the puncture path can adaptively adjust the puncture rate when entering different tissue areas, such as accelerating when entering soft tissue and decelerating when approaching the target area, thereby improving the stability and safety of puncture.

[0035] In a preferred embodiment of the present invention, in the above-described optical vision-guided puncture navigation method, the step of using an improved gradient descent algorithm to solve the path functional and generate a puncture path with optimal characteristics includes:

[0036] A path update formula is established using an improved gradient descent algorithm. , This is the current puncture path. For learning rate, For the path gradient term, For anatomical structure matching items, The weights are used to control the strength of the anatomical matching constraints.

[0037] Output the optimized path direction to obtain the puncture path with optimal characteristics. .

[0038] Its technical advantages are as follows: it adopts a gradient descent strategy to optimize the path, guides the path update through the path gradient term, and gradually optimizes the puncture path toward the global optimum, thereby improving the computational efficiency of path planning; it dynamically adjusts the optimization step size through an improved learning rate, with a larger step size accelerating the convergence speed in the early stage of path optimization, and a smaller step size avoiding oscillation when the path is close to the optimum, thereby improving the fine control capability of the path.

[0039] In a preferred embodiment of the present invention, in the above-described optically guided puncture navigation method, the path gradient term The calculation formula is:

[0040] ,in, This is the second-order gradient of the light field, used for path adjustment.

[0041] In a preferred embodiment of the present invention, in the above-described optically guided puncture navigation method, the anatomical structure matching item... The calculation formula is:

[0042] .

[0043] in, This is the CT image feature vector corresponding to the current puncture path.

[0044] This represents the baseline feature vector of the preoperative MRI image.

[0045] is the L2 norm of the CT image feature vector.

[0046] is the L2 norm of the baseline feature vector of the preoperative MRI image.

[0047] In a preferred embodiment of the present invention, in the above-described optical vision-guided puncture navigation method, the step of constructing a feedback correction amount and correcting the puncture projection based on the error between the actual needle insertion trajectory and the planned puncture path includes:

[0048] Real-time acquisition of needle insertion trajectory The calculated and planned puncture path Overall trajectory error between ,in, This represents the total path length.

[0049] Based on the gradient of the overall trajectory error, and combining Gaussian smoothing filtering and light field gradient information, the error feedback correction amount is calculated. ,in, Let i be the actual location of the i-th sampling point. The gradient of the error with respect to the point. It is a Gaussian smoothing filter kernel. To adjust the step size based on feedback, To preset the puncture direction, Indicates the direction of projection error.

[0050] Using the error feedback correction amount Adjusting intraoperative projection by using a spatial transformation network to calculate dynamic projection correction distribution. ,in, This represents the original optical projection distribution. For spatial transformation networks, For STN transform parameters, This is an additional deformation compensation item.

[0051] The distribution is corrected using the dynamic projection. The optically guided puncture projection is updated, and the gradient vector field of the optical field is recalculated based on the corrected puncture projection. This yields an optimized puncture path.

[0052] Its technical advantages are as follows: During the procedure, due to tissue elasticity, operational errors, or external interference, the needle insertion trajectory may deviate from the original planned path. By constructing the overall trajectory error, the deviation between the needle insertion trajectory and the planned path can be quantified, providing accurate data for subsequent correction. By using error gradient information, the local error trend can be calculated, making the correction process more precise rather than a global linear adjustment, thereby reducing the accumulation of local errors. By introducing a Gaussian smoothing filter kernel to process the error gradient data, high-frequency noise is removed, making the error feedback correction amount smoother and avoiding path instability caused by drastic adjustments. By adjusting the feedback step size to control the error correction intensity, the path adjustment is made more stable, avoiding over-correction that could lead to new errors. Using a spatial transformation network, the dynamic projection correction distribution is calculated using the error feedback correction amount, and the optical projection is dynamically adjusted. This allows for real-time adjustment of the optical projection, ensuring that the projection information is always aligned with the target path, thus improving the accuracy of intraoperative guidance.

[0053] In a preferred embodiment of the present invention, in the above-described optical vision-guided puncture navigation method, the STN transformation parameters... The update formula is ,in, These are the initial affine transformation parameters. This is the learning rate.

[0054] The additional deformation compensation item The calculation formula is ,in, The optical projection intensity of the current organization pixel. This indicates the location of the corresponding point in the preoperative image. Used to calculate the effect of intraoperative tissue deformation on optical projection and to compensate for nonlinear distortion.

[0055] A light-visual-guided puncture navigation system, comprising:

[0056] The optical field model construction module is used to construct a non-uniform optical field model using the position data of optical markers in the device coordinate system. In the non-uniform optical field model, the equation of the hypercurvature optical projection surface describing the spatial light field distribution is defined.

[0057] The multimodal optical marker tracking module is used to introduce filter response constraints into the hyperbolic optical projection surface equation to perform multimodal optical marker tracking and obtain an optimized hyperbolic optical projection surface equation.

[0058] The path functional construction module is used to construct a path functional under multiple constraints in the non-uniform optical field model based on the preset puncture target and real-time acquired obstacle information.

[0059] The puncture path calculation module is used to solve the path functional using an improved gradient descent algorithm to generate a puncture path with optimal characteristics.

[0060] The puncture projection correction module is used to construct a feedback correction amount based on the error between the actual needle insertion trajectory and the planned puncture path, and to correct the puncture projection.

[0061] The beneficial effects of the embodiments of the present invention are:

[0062] This invention utilizes optical marker position data to construct a non-uniform optical field model and employs a hyperbolic optical projection surface equation, enabling optical projection to more accurately describe the spatial light field distribution and improve adaptability to complex tissue structures. It uses a Gaussian weighting function combined with a hyperbolic tangent function to adjust the light field distribution in the depth direction, allowing the optical field to dynamically adapt to different tissue depths and improving the model's applicability. Furthermore, by modeling the non-uniformity of the light field, it enhances the resolution of optical signals in the target area and reduces the impact of light scattering on navigation accuracy.

[0063] This invention introduces a filter response constraint and uses correlation filtering to perform multimodal optical marker tracking, effectively enhancing the detection stability of marker points and reducing the false recognition rate. Fourier transform is used to calculate the filter response map, making marker tracking more adaptable to changes in illumination, partial occlusion, and tissue deformation, thus improving the stability of the navigation system. The inverse Fourier transform is used to calculate the target matching degree, and the response intensity is normalized by the L2 norm, improving the computational efficiency of the algorithm and enabling real-time navigation.

[0064] This invention ensures that the puncture path meets clinical needs by constructing a path functional that includes optical gradient guidance, path curvature constraints, obstacle avoidance constraints, and time-dynamic velocity constraints. The path functional includes an obstacle avoidance term, which can dynamically adjust the path to avoid important tissue structures such as blood vessels and bones, thereby improving puncture safety. The introduction of a time-dynamic velocity constraint term automatically adjusts the puncture speed according to the puncture depth and tissue characteristics, reducing tissue damage and improving the success rate of the operation.

[0065] This invention utilizes an improved gradient descent algorithm to solve the path functional and dynamically adjusts the puncture trajectory through a path update formula to achieve optimal path planning. It introduces an anatomical structure matching term and matches the feature vectors of CT images with the baseline feature vectors of MRI images to ensure that the puncture path accurately matches the patient's anatomical structure, thereby improving surgical accuracy. During the path optimization process, deep learning methods are combined to adaptively adjust the anatomical features, thereby improving the robustness of path optimization.

[0066] This invention calculates feedback correction based on the error between the actual needle insertion trajectory and the planned path, and combines Gaussian smoothing filtering and light field gradient information to dynamically correct the puncture projection and reduce error accumulation. It uses STN to calculate the dynamic projection correction distribution and adaptively adjusts the optical guidance projection to improve the accuracy of intraoperative navigation. It also corrects the influence of intraoperative tissue deformation on optical projection through an additional deformation compensation term, reducing deviations caused by minor patient movements or tissue deformation and improving puncture accuracy. Attached Figure Description

[0067] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0068] Figure 1 This is a flowchart of the optical vision-guided puncture navigation method of the present invention. Detailed Implementation

[0069] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0070] Please refer to Figure 1The first embodiment of the present invention provides a puncture navigation method guided by optical vision, comprising: constructing a non-uniform optical field model using position data of optical markers in a device coordinate system; defining a hyperbolic optical projection surface equation describing the spatial light field distribution in the non-uniform optical field model; introducing filter response constraints into the hyperbolic optical projection surface equation to perform multi-modal optical marker tracking and obtain an optimized hyperbolic optical projection surface equation; constructing a path functional under multiple constraints in the non-uniform optical field model based on a preset puncture target and real-time acquired obstacle information; solving the path functional using an improved gradient descent algorithm to generate a puncture path with optimal characteristics; and constructing a feedback correction amount based on the error between the actual needle insertion trajectory and the planned puncture path to correct the puncture projection.

[0071] In a preferred embodiment of the present invention, in the above-described optical vision-guided puncture navigation method, the step of constructing a non-uniform optical field model using the position data of optical markers in the device coordinate system, and defining the hypercurvature optical projection surface equation describing the spatial light field distribution in the non-uniform optical field model, includes: setting the device coordinate system. ,in, For equipment center, The axis is perpendicular to the equipment surface and consistent with the puncture direction. The axis is parallel to the device surface, defining the horizontal distribution of the optical field; position data of the optical markers are acquired in the device coordinate system. ,in, In three-dimensional space coordinates, As the reflection intensity weight, Here, n represents the number of optical markers distributed on the device surface, and k represents the number of markers. A Gaussian weighting function is used to convert the discrete optical marker data into a continuous light field, thus constructing a non-uniform optical field model describing the light field intensity at a spatial point (x, y, z). ,in, For depth-adaptive Gaussian kernel width, , For adjustment coefficients, To achieve the maximum detection depth, a hyperbolic tangent function is used to adjust the light field distribution along the depth direction. ,in, It is the hyperbolic tangent function. Using curvature adjustment parameters, the equation of the hypercurvature optical projection surface is obtained. .

[0072] Its technical advantages are as follows: Firstly, it uses a Gaussian weighted function to convert discrete optical marker data into a continuous light field, avoiding interpolation errors caused by sparse marker points in traditional methods. This results in a smoother and more continuous light field intensity distribution, improving the accuracy of optical projection surface modeling. Secondly, it employs a depth-adaptive Gaussian kernel width, enabling the light field model to automatically adjust its resolution based on the target depth. This provides finer optical guidance in shallower regions while improving stability and reducing signal attenuation in deeper regions. Thirdly, it uses a hyperbolic tangent function to adjust the light field distribution in the depth direction. Through the nonlinear characteristics of the tanh function, the optical projection achieves higher resolution near the device's center while transitioning smoothly away from the center, reducing distortion and improving adaptability to complex human anatomy. Finally, by constructing a hyperbolic optical projection surface equation, it can more accurately describe the distribution characteristics of optical projection in a non-uniform optical field, providing precise optical gradient information for subsequent puncture path optimization.

[0073] In a preferred embodiment of the present invention, in the above-described optical vision-guided puncture navigation method, the step of introducing a filter response constraint into the hyperbolic optical projection surface equation to perform multimodal optical marker tracking and obtain an optimized hyperbolic optical projection surface equation includes: introducing a filter response constraint into the hyperbolic optical projection surface equation to perform multimodal optical marker tracking. ;in, This is the correlation filter response diagram. This indicates the degree of matching between the input image and the filter template, expressed by the inverse Fourier transform. Calculated; For filtering template The conjugate of complex numbers, As a filter template, , For the Fourier transform of the input image, For the conjugate Fourier transform of the target template, The regularization coefficient is used. This represents the maximum matching value of the correlation filter; The L2 norm of the filter template is used to normalize the response intensity; the optimized hypercurvature optical projection surface equation is obtained. .

[0074] Its technical advantages are as follows: By calculating the matching degree between the input image and the filter template through correlation filtering, a filter response map is obtained, which can accurately identify the position of optical markers in complex environments and improve the recognition accuracy of optical markers; Fourier transform and inverse transform are used to calculate the matching degree, so that marker tracking not only relies on spatial domain information, but also combines frequency domain features, reducing noise interference and improving the stability and robustness of optical markers; It is not only applicable to single-modal optical markers, but also supports multi-modal fusion, and can stably track optical markers under different lighting, tissue depth, and environmental change conditions; Through the optimized hypercurvature optical projection surface equation, the information of multiple optical markers is fused in a unified projection coordinate system, improving adaptability to complex surgical environments and ensuring the stability of optical guidance; The method of calculating the marker matching degree by correlation filtering has lower computational complexity than traditional template matching or deep learning-based methods, and is suitable for real-time puncture navigation; Due to the use of Fourier transform to accelerate the calculation, large-scale image matching can be completed in a short time, improving the response speed of optical marker tracking, enabling the intraoperative navigation system to quickly adjust the puncture path and improve surgical efficiency.

[0075] In a preferred embodiment of the present invention, in the above-described optical vision-guided puncture navigation method, the step of constructing a path functional under multiple constraints in the non-uniform optical field model based on a preset puncture target and real-time acquired obstacle information includes: based on the optimized hypercurvature optical projection surface equation. Calculate the gradient field of the spatial light field to obtain the gradient vector field of the optical field. ,in, Reflecting the change in optical gradient in the X direction, Reflects the change in optical gradient in the Y direction. It reflects the change in optical gradient in the Z direction; based on the preset puncture target and real-time acquired obstacle information, it sets optical gradient guidance terms. Path curvature constraint Obstacle avoidance items and time dynamic velocity constraints ,in, The weighting factor for the optical gradient constraint. For path curvature, As the curvature control weighting factor, The distance from the current path to the nearest obstacle. For obstacle avoidance weighting factors, For time-dynamic weighting function, Let Γ be the instantaneous velocity of the puncture path at time t. The maximum permissible puncture speed is preset; wherein, the optical gradient guidance term is used to ensure that the path follows the upward direction of the optical gradient, improving navigation accuracy; the path curvature constraint term is used to control the smoothness of the path and prevent excessive bending; the obstacle avoidance term is used to ensure that the path stays away from obstacles, improving safety; the time dynamic velocity constraint term is used to control the puncture speed and ensure the path is stable; the path functional is constructed. .

[0076] Its technical advantages are as follows: the optical gradient guidance term guides the puncture path along the upward direction of the optical gradient, directing the puncture needle toward the optimal navigation path and improving navigation accuracy. Since the optical field gradient information comes from the optimized hyperbolic optical projection surface equation, path planning can be adaptively adjusted based on the dynamic optical environment, ensuring that the puncture path always faces the target area and reducing deviations caused by tissue deformation or optical signal interference. The path curvature constraint term controls the smoothness of the path, preventing unreasonable sharp turns and making the puncture trajectory more physiologically consistent, thus improving the feasibility and safety of the surgical procedure. By adjusting the curvature control weighting factor, it can be flexibly adjusted according to surgical needs. The flexibility of the puncture path makes it suitable for different types of puncture tasks; the obstacle avoidance parameter keeps the puncture path away from obstacles, using the dynamic distance from the path to the nearest obstacle as a constraint index, making obstacle avoidance control more intelligent and able to adapt to changes in the intraoperative environment in real time, such as the patient's slight movements or tissue deformation; the time dynamic speed constraint parameter controls the puncture speed, ensuring the dynamic stability of the path and avoiding tissue damage or path tracking errors caused by excessive speed. The puncture speed is adjusted by a time dynamic weight function, so that the puncture path can adaptively adjust the puncture rate when entering different tissue areas, such as accelerating when entering soft tissue and decelerating when approaching the target area, thereby improving the stability and safety of puncture.

[0077] In a preferred embodiment of the present invention, in the above-described optical vision-guided puncture navigation method, the step of solving the path functional using an improved gradient descent algorithm to generate a puncture path with optimal characteristics includes: establishing a path update formula using the improved gradient descent algorithm. , This is the current puncture path. The learning rate is used to control the optimization step size. This is the path gradient term, representing the optimization direction of the current path. This is an anatomical structure matching option used to calculate the degree of matching between the current path and preoperative images, avoiding damage to important tissues. The weights for controlling the anatomical matching constraint strength are used to output the optimized path direction, resulting in a puncture path with optimal characteristics. .

[0078] Its technical advantages are as follows: it adopts a gradient descent strategy to optimize the path, guides the path update through the path gradient term, and gradually optimizes the puncture path toward the global optimum, thereby improving the computational efficiency of path planning; it dynamically adjusts the optimization step size through an improved learning rate, with a larger step size accelerating the convergence speed in the early stage of path optimization, and a smaller step size avoiding oscillation when the path is close to the optimum, thereby improving the fine control capability of the path.

[0079] In a preferred embodiment of the present invention, in the above-described optically guided puncture navigation method, the path gradient term The calculation formula is: ,in, This is the second-order gradient of the light field, used for path adjustment.

[0080] In a preferred embodiment of the present invention, in the above-described optically guided puncture navigation method, the anatomical structure matching item... The calculation formula is: ;in, This is the CT image feature vector corresponding to the current puncture path; The baseline feature vector of the preoperative MRI image; The L2 norm of the CT image feature vector; is the L2 norm of the baseline feature vector of the preoperative MRI image.

[0081] In a preferred embodiment of the present invention, in the above-described optical vision-guided puncture navigation method, the step of constructing a feedback correction amount and correcting the puncture projection based on the error between the actual needle insertion trajectory and the planned puncture path includes: acquiring the needle insertion trajectory in real time. The calculated and planned puncture path Overall trajectory error between ,in, The total path length is given; based on the gradient of the overall trajectory error, combined with Gaussian smoothing filtering and light field gradient information, the error feedback correction amount is calculated. ,in, Let i be the actual location of the i-th sampling point. The gradient of the error with respect to a point is used to represent the local trend of error change at that point. It is a Gaussian smoothing filter kernel used to remove noise. To adjust the step size based on feedback, To preset the puncture direction, Indicates the direction of projection error; uses the error feedback correction amount Adjusting intraoperative projection by using a spatial transformation network to calculate dynamic projection correction distribution. ,in, This represents the original optical projection distribution. For spatial transformation networks, For STN transform parameters, Additional deformation compensation term; distribution corrected using the dynamic projection. The optically guided puncture projection is updated, and the gradient vector field of the optical field is recalculated based on the corrected puncture projection. This yields an optimized puncture path.

[0082] Its technical advantages are as follows: During the procedure, due to tissue elasticity, operational errors, or external interference, the needle insertion trajectory may deviate from the original planned path. By constructing the overall trajectory error, the deviation between the needle insertion trajectory and the planned path can be quantified, providing accurate data for subsequent correction. By using error gradient information, the local error trend can be calculated, making the correction process more precise rather than a global linear adjustment, thereby reducing the accumulation of local errors. By introducing a Gaussian smoothing filter kernel to process the error gradient data, high-frequency noise is removed, making the error feedback correction amount smoother and avoiding path instability caused by drastic adjustments. By adjusting the feedback step size to control the error correction intensity, the path adjustment is made more stable, avoiding over-correction that could lead to new errors. Using a spatial transformation network, the dynamic projection correction distribution is calculated using the error feedback correction amount, and the optical projection is dynamically adjusted. This allows for real-time adjustment of the optical projection, ensuring that the projection information is always aligned with the target path, thus improving the accuracy of intraoperative guidance.

[0083] In a preferred embodiment of the present invention, in the above-described optical vision-guided puncture navigation method, the STN transformation parameters... The update formula is ,in, These are the initial affine transformation parameters. The learning rate; the additional deformation compensation term The calculation formula is ,in, The optical projection intensity of the current organization pixel. This indicates the location of the corresponding point in the preoperative image. Used to calculate the effect of intraoperative tissue deformation on optical projection and to compensate for nonlinear distortion.

[0084] A second embodiment of the present invention provides a puncture navigation system guided by optical vision, comprising: an optical field model construction module, used to construct a non-uniform optical field model using position data of optical markers in the device coordinate system, and defining a hyperbolic optical projection surface equation describing the spatial light field distribution in the non-uniform optical field model; a multi-modal optical marker tracking module, used to introduce filter response constraints into the hyperbolic optical projection surface equation to perform multi-modal optical marker tracking and obtain an optimized hyperbolic optical projection surface equation; a path functional construction module, used to construct a path functional under multiple constraints in the non-uniform optical field model based on a preset puncture target and real-time acquired obstacle information; a puncture path calculation module, used to solve the path functional using an improved gradient descent algorithm to generate a puncture path with optimal characteristics; and a puncture projection correction module, used to construct a feedback correction amount based on the error between the actual needle insertion trajectory and the planned puncture path to correct the puncture projection.

[0085] The computer program product of the optical vision-guided puncture navigation method and apparatus 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 methods in the preceding method embodiments. For specific implementation, please refer to the method embodiments, which will not be repeated here.

[0086] Specifically, the storage medium can be a general-purpose storage medium, such as a portable disk or hard disk. When the computer program on the storage medium is run, it can execute the above-mentioned optical vision-guided puncture navigation method, thereby constructing a more accurate optical field model and combining it with optimized path planning algorithms and adaptive feedback to achieve higher precision and more stable puncture navigation.

[0087] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0088] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A puncture navigation method guided by optical vision, characterized in that, include: Using the position data of optical markers in the device coordinate system, a non-uniform optical field model is constructed, and the hypercurvature optical projection surface equation describing the spatial light field distribution is defined in the non-uniform optical field model. By introducing filter response constraints into the hyperbolic optical projection surface equation and performing multimodal optical marker tracking, an optimized hyperbolic optical projection surface equation is obtained. ; Based on the preset puncture target and real-time acquired obstacle information, a path functional under multiple constraints is constructed in the non-uniform optical field model, specifically including: based on the optimized hypercurvature optical projection surface equation. Calculate the gradient field of the spatial light field to obtain the gradient vector field of the optical field. ,in, Reflecting the change in optical gradient in the X direction, Reflects the change in optical gradient in the Y direction. It reflects the change in optical gradient in the Z direction; based on the preset puncture target and real-time acquired obstacle information, it sets optical gradient guidance terms. Path curvature constraint Obstacle avoidance items and time dynamic velocity constraints ,in, The weighting factor for the optical gradient constraint. For path curvature, As the curvature control weighting factor, The distance from the current path to the nearest obstacle. For obstacle avoidance weighting factors, For time-dynamic weighting function, Let Γ be the instantaneous velocity of the puncture path at time t. To preset the maximum allowable puncture speed, a path functional is constructed. ; The improved gradient descent algorithm is used to solve the path functional to generate a puncture path with optimal characteristics; Based on the error between the actual needle insertion trajectory and the planned puncture path, a feedback correction amount is constructed to correct the puncture projection.

2. The optical vision-guided puncture navigation method according to claim 1, characterized in that, The method of constructing a non-uniform optical field model using the position data of optical markers in the device coordinate system, and defining the hypercurvature optical projection surface equations describing the spatial light field distribution in the non-uniform optical field model, includes: Set the equipment coordinate system ,in, For equipment center, The axis is perpendicular to the equipment surface and consistent with the puncture direction. The axis is parallel to the device surface, defining the horizontal distribution of the optical field; The position data of the optical markers are acquired in the device coordinate system. ,in, In three-dimensional space coordinates, As the reflection intensity weight, n is the number of optical markers distributed on the surface of the device, and k is the number of markers; By employing a Gaussian weighting function, discrete optical marker data is converted into a continuous light field, thus constructing a non-uniform optical field model describing the light field intensity at a spatial point (x,y,z). ,in, For depth-adaptive Gaussian kernel width, , For adjustment coefficients, Maximum detection depth; Adjusting the light field distribution along the depth direction using the hyperbolic tangent function ,in, It is the hyperbolic tangent function. Using curvature adjustment parameters, the equation of the hypercurvature optical projection surface is obtained. .

3. The optical vision-guided puncture navigation method according to claim 2, characterized in that, The process of introducing filter response constraints into the hyperbolic optical projection surface equation and performing multimodal optical marker tracking to obtain the optimized hyperbolic optical projection surface equation includes: Introducing filter response constraints into the equation of the hypercurvature optical projection surface to perform multimodal optical marker tracking. ; in, This is the correlation filter response diagram. This indicates the degree of matching between the input image and the filter template, expressed by the inverse Fourier transform. Calculated; For filtering template The conjugate of complex numbers, As a filter template, , For the Fourier transform of the input image, The conjugate Fourier transform of the target template. The regularization coefficient is used. This represents the maximum matching value of the correlation filter; The L2 norm of the filter template is used to normalize the response intensity; The optimized hypercurvature optical projection surface equation is obtained. .

4. The optical vision-guided puncture navigation method according to claim 3, characterized in that, The step of using an improved gradient descent algorithm to solve the path functional and generate a puncture path with optimal characteristics includes: A path update formula is established using an improved gradient descent algorithm. , This is the current puncture path. For learning rate, For the path gradient term, For anatomical structure matching items, Weights are used to control the strength of anatomical matching constraints; Output the optimized path direction to obtain the puncture path with optimal characteristics. .

5. The optical vision-guided puncture navigation method according to claim 4, characterized in that, The path gradient term The calculation formula is: ,in, This is the second-order gradient of the light field, used for path adjustment.

6. The optical vision-guided puncture navigation method according to claim 4, characterized in that, Anatomical matching items The calculation formula is: ； in, This is the CT image feature vector corresponding to the current puncture path; The baseline feature vector of the preoperative MRI image; The L2 norm of the CT image feature vector; is the L2 norm of the baseline feature vector of the preoperative MRI image.

7. The optical vision-guided puncture navigation method according to claim 4, characterized in that, The step of constructing a feedback correction amount based on the error between the actual needle insertion trajectory and the planned puncture path, and correcting the puncture projection, includes: Real-time acquisition of needle insertion trajectory The calculated and planned puncture path Overall trajectory error between ,in, This represents the total path length. Based on the gradient of the overall trajectory error, and combining Gaussian smoothing filtering and light field gradient information, the error feedback correction amount is calculated. ,in, Let i be the actual location of the i-th sampling point. The gradient of the error with respect to the point. It is a Gaussian smoothing filter kernel. To adjust the step size based on feedback, To preset the puncture direction, Indicates the direction of projection error; Using the error feedback correction amount Adjusting intraoperative projection by using a spatial transformation network to calculate dynamic projection correction distribution. ,in, This represents the original optical projection distribution. For spatial transformation networks, For STN transform parameters, This is an additional deformation compensation item; The distribution is corrected using the dynamic projection. The optically guided puncture projection is updated, and the gradient vector field of the optical field is recalculated based on the corrected puncture projection. This yields an optimized puncture path.

8. The optical vision-guided puncture navigation method according to claim 7, characterized in that, The STN transformation parameters The update formula is ,in, These are the initial affine transformation parameters. The learning rate; The additional deformation compensation item The calculation formula is: ,in, The optical projection intensity of the current organization's pixel. This indicates the location of the corresponding point in the preoperative image. Used to calculate the effect of intraoperative tissue deformation on optical projection and to compensate for nonlinear distortion.

9. A light-vision guided puncture navigation system, characterized in that, include: The optical field model construction module is used to construct a non-uniform optical field model using the position data of optical markers in the device coordinate system. In the non-uniform optical field model, the hypercurvature optical projection surface equation describing the spatial light field distribution is defined. The multimodal optical marker tracking module is used to introduce filter response constraints into the equation of the hypercurvature optical projection surface, perform multimodal optical marker tracking, and obtain the optimized hypercurvature optical projection surface equation. ; The path functional construction module is used to construct a path functional under multiple constraints in the non-uniform optical field model based on a preset puncture target and real-time acquired obstacle information. Specifically, it includes: based on the optimized hypercurvature optical projection surface equation... Calculate the gradient field of the spatial light field to obtain the gradient vector field of the optical field. ,in, Reflecting the change in optical gradient in the X direction, Reflects the change in optical gradient in the Y direction. It reflects the change in optical gradient in the Z direction; based on the preset puncture target and real-time acquired obstacle information, it sets optical gradient guidance terms. Path curvature constraint Obstacle avoidance items and time dynamic velocity constraints ,in, The weighting factor for the optical gradient constraint. For path curvature, As the curvature control weighting factor, The distance from the current path to the nearest obstacle. For obstacle avoidance weighting factors, For time-dynamic weighting function, Let Γ be the instantaneous velocity of the puncture path at time t. To preset the maximum allowable puncture speed, a path functional is constructed. ; The puncture path calculation module is used to solve the path functional using an improved gradient descent algorithm to generate a puncture path with optimal characteristics. The puncture projection correction module is used to construct a feedback correction amount based on the error between the actual needle insertion trajectory and the planned puncture path, and to correct the puncture projection.

Images