A guide needle path navigation method and system based on electromagnetic sensing

Abstract

This invention discloses a guide needle path navigation method and system based on electromagnetic sensing, comprising: if the guide needle position deviates from the correction path by more than a preset safety threshold in a dynamic image sequence, then the deviation area is detected by an image processing algorithm, a deviation vector is generated and mapped to a three-dimensional coordinate space to determine the precise adjustment angle and distance of the guide needle; based on the deviation vector and the correction path coordinate sequence, the rotation angle and advance distance required for guide needle adjustment are calculated using an inverse kinematics algorithm, a control signal is generated, and transmitted to the guide needle driving device through an electromagnetic feedback system to obtain an automatic adjustment command for the guide needle; the adjusted guide needle position and attitude data are obtained from the execution result of the guide needle driving device, and the positioning accuracy is calculated by comparing it with the correction path coordinate sequence using an error analysis algorithm to determine whether the guide needle has reached the sub-millimeter level accuracy requirement.

Description

Technical Field

[0001] This invention belongs to the field of surgical navigation and visualization technology, and particularly relates to a guide needle path navigation method and system based on electromagnetic sensing. Background Technology

[0002] In instrument positioning in the knee joint region, existing technologies have significant limitations: traditional positioning devices rely on human experience to determine instrument position, lacking precise path planning and real-time guidance, leading to increased operational complexity and error risk. Traditional methods manually adjust instrument orientation and rely on external observation channels to avoid damage to critical structures; this not only increases operational complexity but also affects positioning accuracy due to the lack of real-time position feedback. The core technical challenges are: first, the inability to pre-generate optimal instrument movement paths, making it difficult to dynamically track the instrument's three-dimensional position and orientation during operation; second, the lack of real-time visualization guidance, requiring additional auxiliary devices to observe the path, further increasing operational complexity; and finally, the lack of an automatic correction mechanism during positioning, making it difficult to quickly correct deviations. These technical problems limit the accuracy and efficiency of instrument positioning. Therefore, integrating electromagnetic tracking, intelligent path planning, and real-time visualization technologies to achieve precise dynamic guidance of the instrument path has become a key issue in improving the performance of positioning systems. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention proposes a guide needle path navigation method and system based on electromagnetic sensing, thereby resolving the issues present in the prior art.

[0004] In a first aspect, to achieve the above objectives, the present invention provides a guide needle path navigation method based on electromagnetic sensing, comprising the following steps:

[0005] Three-dimensional structural information of the knee joint is obtained from the patient's preoperative MRI image data, and a three-dimensional anatomical model including the distribution of blood vessels and nerves is generated.

[0006] For the aforementioned three-dimensional anatomical model, an artificial intelligence path planning algorithm is used to generate a three-dimensional coordinate sequence of the optimal guide needle puncture path;

[0007] A miniature electromagnetic transmitter is integrated at the tip of the guide needle, and the three-dimensional position coordinates and attitude angle of the guide needle are captured in real time through an externally arranged electromagnetic receiver array;

[0008] If the deviation between the real-time movement trajectory of the guide needle and the planned path exceeds a preset threshold, a dynamic path calibration algorithm is used to generate a corrected path coordinate sequence.

[0009] Based on the real-time motion trajectory and the corrected path coordinate sequence, a dynamic image sequence containing guide needle movement guidance information is generated using virtual reality rendering technology.

[0010] If the position of the guide pin deviates from the correction path by more than a preset safety threshold in the dynamic image sequence, an image processing algorithm is used to generate a deviation vector and map it to a three-dimensional coordinate space to determine the adjustment angle and distance of the guide pin.

[0011] Based on the deviation vector and the coordinate sequence of the correction path, an inverse kinematics algorithm is used to generate a control signal to adjust the direction of the guide needle's movement.

[0012] Verify whether the adjusted guide pin positioning accuracy meets the sub-millimeter level requirement. If it does, update the navigation image sequence in the virtual reality interface to complete the closed-loop control.

[0013] Optionally, the process of generating the three-dimensional anatomical model includes:

[0014] Separate the tibia, femur, and posterior cruciate ligament remnant regions from MRI imaging data;

[0015] Bone regions are labeled using a threshold segmentation algorithm;

[0016] A region growing algorithm is used to expand from the boundary of the skeletal region and mark the vascular and neural regions.

[0017] Optionally, the process of generating the optimal guide needle puncture path includes:

[0018] Analyze the geometric constraints of the tibial positioning point and the femoral junction area;

[0019] By combining vascular and nerve distribution data, a node coordinate sequence is generated by avoiding high-risk areas using the A* path planning algorithm;

[0020] The node coordinate sequence is smoothed using cubic spline interpolation.

[0021] Optionally, the process of real-time capture of the three-dimensional position coordinates and attitude angle of the guide pin includes:

[0022] The relative distance between the tip of the guide pin and the electromagnetic receiver array is calculated using a triangulation algorithm based on the electromagnetic field signal strength.

[0023] The real-time three-dimensional coordinates of the guide pin are determined based on the geometric layout of the receiver array.

[0024] Optionally, the implementation process of the dynamic path calibration algorithm includes:

[0025] The coordinate sequence of the real-time trajectory and the planned path is fitted using the least squares method;

[0026] Calculate the angle between the current direction of the guide needle's travel and the direction of the correction path;

[0027] If the included angle exceeds a preset angle threshold, direction adjustment parameters are generated and guide needle motion control parameters are updated.

[0028] Optionally, the process of verifying the positioning accuracy of the guide pin includes:

[0029] Compare the adjusted guide pin position data with the calibration path coordinate sequence;

[0030] The average deviation distance is calculated using an error analysis algorithm.

[0031] If the average deviation distance is less than the sub-millimeter threshold, the positioning accuracy is deemed to meet the standard.

[0032] Secondly, the present invention also provides a guide needle path navigation system based on electromagnetic sensing, for implementing a guide needle path navigation method based on electromagnetic sensing, the system comprising:

[0033] The three-dimensional structural information acquisition module is used to acquire three-dimensional structural information of the knee joint from the patient's preoperative MRI image data and generate a three-dimensional anatomical model including the distribution of blood vessels and nerves.

[0034] An artificial intelligence path planning module is used to generate a three-dimensional coordinate sequence of the optimal guide needle puncture path for the three-dimensional anatomical model.

[0035] The electromagnetic positioning module is used to capture the three-dimensional position coordinates and attitude angle of the guide needle in real time through a miniature electromagnetic transmitter integrated at the front end of the guide needle and an externally arranged electromagnetic receiver array.

[0036] The dynamic path calibration module is used to generate a corrected path coordinate sequence when the deviation between the real-time movement trajectory of the guide needle and the planned path exceeds a preset threshold.

[0037] The virtual reality navigation module is used to generate a dynamic image sequence containing guide needle movement guidance information based on the real-time motion trajectory and the coordinate sequence of the correction path.

[0038] The deviation detection module is used to generate a deviation vector and map it to a three-dimensional coordinate space when the position of the guide pin deviates from the correction path by more than a preset safety threshold in a dynamic image sequence.

[0039] The guide needle adjustment control module is used to generate control signals based on the deviation vector and the coordinate sequence of the correction path to adjust the direction of the guide needle's movement;

[0040] The positioning accuracy verification module is used to verify whether the adjusted guide pin positioning accuracy meets the sub-millimeter level requirements.

[0041] The navigation image update module is used to update the navigation image sequence in the virtual reality interface when the positioning accuracy meets the standard, thus completing closed-loop control.

[0042] Optionally, the three-dimensional structural information acquisition module includes:

[0043] Image segmentation unit is used to separate the tibia, femur and posterior cruciate ligament remnant regions from MRI image data;

[0044] Skeletal labeling unit, used to label skeletal regions using a threshold segmentation algorithm;

[0045] The vascular and neural labeling unit is used to expand and label vascular and neural regions from the skeletal region boundary using a region growing algorithm.

[0046] Thirdly, the present invention also provides a computer terminal device, comprising:

[0047] One or more processors;

[0048] A memory, coupled to the processor, for storing one or more programs;

[0049] When the one or more programs are executed by the one or more processors, the one or more processors implement a guide needle path navigation method based on electromagnetic sensing.

[0050] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a guide needle path navigation method based on electromagnetic sensing.

[0051] Compared with the prior art, the present invention has the following advantages and technical effects:

[0052] This invention provides a guide needle path navigation method and system based on electromagnetic sensing, used to assist in the precise positioning of medical devices in the knee joint region. The system analyzes the patient's preoperative MRI images to generate a three-dimensional anatomical model including the distribution of blood vessels and nerves, and uses an intelligent algorithm to plan the movement path of the guide needle. During positioning, the spatial position of the guide needle is captured in real time using electromagnetic positioning technology, and combined with a dynamic path calibration algorithm and virtual reality display to generate navigation guidance information. When the guide needle deviates from the preset path, the system automatically calculates adjustment parameters and performs precise adjustments through electromagnetic feedback, achieving sub-millimeter-level positioning accuracy. This invention significantly improves the accuracy of device positioning and operational safety by integrating electromagnetic sensing, dynamic calibration, and virtual reality technologies. Attached Figure Description

[0053] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0054] Figure 1 This is a flowchart of a guide needle path navigation method based on electromagnetic sensing, according to an embodiment of the present invention.

[0055] Figure 2 This is a simplified schematic diagram of part of the processing logic of the guide needle path navigation method based on electromagnetic sensing according to an embodiment of the present invention;

[0056] Figure 3 This is a simplified schematic diagram of part of the processing logic of the guide needle path navigation method based on electromagnetic sensing according to an embodiment of the present invention;

[0057] Figure 4 This is a schematic diagram of the electromagnetic sensing-based needle path navigation system according to an embodiment of the present invention. Detailed Implementation

[0058] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0059] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0060] Example 1

[0061] like Figures 1 to 3 As shown, this embodiment provides a guide needle path navigation method based on electromagnetic sensing, including:

[0062] Three-dimensional structural information of the knee joint is obtained from the patient's preoperative MRI image data, and a three-dimensional anatomical model including the distribution of blood vessels and nerves is generated.

[0063] For the aforementioned three-dimensional anatomical model, an artificial intelligence path planning algorithm is used to generate a three-dimensional coordinate sequence of the optimal guide needle puncture path;

[0064] A miniature electromagnetic transmitter is integrated at the tip of the guide needle, and the three-dimensional position coordinates and attitude angle of the guide needle are captured in real time through an externally arranged electromagnetic receiver array;

[0065] If the deviation between the real-time movement trajectory of the guide needle and the planned path exceeds a preset threshold, a dynamic path calibration algorithm is used to generate a corrected path coordinate sequence.

[0066] Based on the real-time motion trajectory and the corrected path coordinate sequence, a dynamic image sequence containing guide needle movement guidance information is generated using virtual reality rendering technology.

[0067] If the position of the guide pin deviates from the correction path by more than a preset safety threshold in the dynamic image sequence, an image processing algorithm is used to generate a deviation vector and map it to a three-dimensional coordinate space to determine the adjustment angle and distance of the guide pin.

[0068] Based on the deviation vector and the coordinate sequence of the correction path, an inverse kinematics algorithm is used to generate a control signal to adjust the direction of the guide needle's movement.

[0069] Verify whether the adjusted guide pin positioning accuracy meets the sub-millimeter level requirement. If it does, update the navigation image sequence in the virtual reality interface to complete the closed-loop control.

[0070] Specifically, the three-dimensional structural information of the knee joint is obtained from the patient's preoperative MRI image data. The tibia, femur and posterior cruciate ligament stump regions are separated using image segmentation algorithms to generate a three-dimensional anatomical model containing the distribution of blood vessels and nerves, and to determine the initial surgical planning area.

[0071] For the three-dimensional anatomical model, an artificial intelligence path planning algorithm is used to generate the optimal guide needle puncture path by analyzing the geometric constraints of the tibial positioning point and the femoral connection area, combined with the vascular and nerve distribution data, and obtain the three-dimensional coordinate sequence of the planned path.

[0072] A miniature electromagnetic transmitter is integrated at the front end of the guide needle, and an electromagnetic receiver array is arranged externally. The position coordinates and attitude angle of the guide needle in three-dimensional space are captured in real time through the triangulation algorithm of electromagnetic field signal intensity, so as to obtain the real-time motion trajectory data of the guide needle.

[0073] If the deviation between the real-time motion trajectory data of the guide needle and the three-dimensional coordinate sequence of the planned path exceeds a preset threshold, a dynamic path calibration algorithm is adopted. The real-time trajectory and the planned path are fitted by the least squares method to generate a corrected path coordinate sequence and determine the adjusted direction of the guide needle.

[0074] Based on the real-time motion trajectory data of the guide needle and the corrected path coordinate sequence, virtual reality rendering technology is used to overlay and display the three-dimensional anatomical model, the planned path and the current position of the guide needle, generating a dynamic image sequence containing a stereoscopic visual navigation interface, and obtaining intuitive guide needle movement guidance information.

[0075] If the position of the guide pin deviates from the correction path by more than a preset safety threshold in the dynamic image sequence, the deviation area is detected by the image processing algorithm, a deviation vector is generated and mapped to the three-dimensional coordinate space, and the precise adjustment angle and distance of the guide pin are determined.

[0076] Based on the deviation vector and the coordinate sequence of the correction path, the inverse kinematics algorithm is used to calculate the rotation angle and advance distance required for the guide needle adjustment, generate a control signal, and transmit it to the guide needle drive device through the electromagnetic feedback system to obtain the automatic adjustment command for the guide needle;

[0077] The adjusted position and attitude data of the guide pin are obtained from the execution results of the guide pin drive device. By comparing them with the coordinate sequence of the correction path, the positioning accuracy is calculated using an error analysis algorithm to determine whether the guide pin meets the requirements of sub-millimeter accuracy.

[0078] If the positioning accuracy meets the preset threshold, the superimposed display of the guide needle position and the three-dimensional anatomical model is updated through the virtual reality interface to generate the final navigation image sequence, determine the real-time guidance data of the surgical path, and complete the closed-loop control of the entire guide needle positioning process.

[0079] As one implementation method in this embodiment, the process of generating the three-dimensional anatomical model includes:

[0080] Separate the tibia, femur, and posterior cruciate ligament remnant regions from MRI imaging data;

[0081] Bone regions are labeled using a threshold segmentation algorithm;

[0082] A region growing algorithm is used to expand from the boundary of the skeletal region and mark the vascular and neural regions.

[0083] Specifically, in step S101, three-dimensional structural information of the knee joint is obtained from the patient's preoperative MRI image data. The tibia, femur and posterior cruciate ligament stump regions are separated using an image segmentation algorithm to generate a three-dimensional anatomical model containing the distribution of blood vessels and nerves, and the initial surgical planning area is determined.

[0084] Raw 3D structural data is obtained from MRI images. A stereomicroscopy algorithm is used to reconstruct the 3D structure, initially locating the tibia, femur, and posterior cruciate ligament (PCL) remnant regions to obtain a 3D structural dataset containing bones and soft tissues. Based on this dataset, a threshold segmentation algorithm is used. If a pixel's grayscale value is greater than a preset bone threshold, it is marked as a bone region, resulting in a separated bone region dataset. Using this bone region dataset, a region growing algorithm is used to expand outwards from the bone region boundaries. If a pixel's grayscale value meets a preset vascular threshold, it is marked as a vascular region; if it meets a preset neural threshold, it is marked as a neural region, generating a 3D anatomical model containing vascular and neural distributions. Based on this 3D anatomical model, a stereoscopic projection algorithm is used to project the tibia, femur, and PCL remnant regions, obtaining a transformed projection dataset. The initial surgical planning area is determined using this projection dataset.

[0085] More specifically, when acquiring raw 3D structural data from MRI images, T2-weighted imaging technology can be used to scan the knee joint region, generating a high-resolution 3D stereoscopic dataset with a resolution of 0.5mm × 0.5mm × 1mm. Stereoscopic microscopy algorithms, through voxel interpolation and surface reconstruction, generate a continuous 3D structure including the tibia, femur, and posterior cruciate ligament remnants. This method preserves the boundary details of soft tissue and bone, facilitating subsequent precise localization.

[0086] In one possible implementation, the initial localization of the tibia, femur, and posterior cruciate ligament remnant regions can be achieved by combining region labeling and geometric feature extraction. The tibia, due to its tubular bone structure, can be located through gradient analysis along its long axis; the femur, due to its ball-and-socket joint characteristics, can be identified through curvature analysis; and the posterior cruciate ligament remnant, due to its fibrous structure, can be initially labeled through texture features. This localization method effectively distinguishes different anatomical regions, improving the accuracy of subsequent segmentation. When processing 3D structural datasets using a threshold segmentation algorithm, a bone grayscale threshold of 2000 HU (Hounsfield units) can be set; pixels exceeding this value are labeled as bone regions.

[0087] It should be noted that the threshold needs to be calibrated according to the specific equipment and scanning parameters to avoid misidentifying soft tissue as bone. The separated bone region dataset can clearly show the contours of the tibia and femur, providing a basis for subsequent blood vessel and nerve segmentation. Based on the bone region dataset, when the region growing algorithm expands outward from the bone boundary, the grayscale threshold for blood vessels can be set to 800-1200 HU, and the grayscale threshold for nerves to 300-600 HU. Growing starts from the outer boundary of the tibia. If a pixel has a grayscale of 900 HU, it is marked as a blood vessel; if the grayscale is 400 HU, it is marked as a nerve. This method uses grayscale differences to achieve accurate classification. The generated three-dimensional anatomical model containing the distribution of blood vessels and nerves can intuitively show the anatomical relationships of the knee joint, which is helpful for surgical planning.

[0088] In one embodiment, when performing projection transformations on the tibia, femur, and posterior cruciate ligament (PCL) remnant region, the stereoscopic projection algorithm can employ orthogonal projection to generate a two-dimensional projection dataset. For the PCL remnant, projections can be made from both anteroposterior and lateral directions, preserving its spatial location information. The transformed projection dataset highlights the relative positions of key anatomical structures, reducing the complexity of surgical planning.

[0089] When determining the initial surgical planning area using projection datasets, anatomical landmarks and lesion locations can be considered. If a tear exists at the posterior cruciate ligament stump, the coordinates of the tear site can be marked in the projection dataset, and the surgical incision and repair path can be planned in conjunction with the projected contours of the tibia and femur. This method improves surgical precision, reduces intraoperative adjustment time, and significantly enhances repair outcomes. The overall workflow of the above technical solution ensures seamless integration from imaging data to surgical planning. Especially in posterior cruciate ligament reconstruction surgery, precise information on vascular and nerve distribution can help avoid intraoperative damage and improve postoperative recovery.

[0090] As one implementation method in this embodiment, the process of generating the optimal guide needle puncture path includes:

[0091] Analyze the geometric constraints of the tibial positioning point and the femoral junction area;

[0092] By combining vascular and nerve distribution data, a node coordinate sequence is generated by avoiding high-risk areas using the A* path planning algorithm;

[0093] The node coordinate sequence is smoothed using cubic spline interpolation.

[0094] Specifically, in step S102, for the three-dimensional anatomical model, an artificial intelligence path planning algorithm is used to analyze the geometric constraints of the tibial positioning point and the femoral connection area, and combine the vascular and nerve distribution data to generate the optimal guide needle puncture path, thereby obtaining the three-dimensional coordinate sequence of the planned path.

[0095] More specifically, coordinate data of the tibia positioning point and the femoral connection region are obtained from a three-dimensional anatomical model. This coordinate data includes the three-dimensional coordinates of the tibia positioning point and the boundary point coordinates of the femoral connection region. Using spatial geometric calculation methods, the distance between the two regions is calculated for the three-dimensional coordinates of the tibia positioning point and the boundary point coordinates of the femoral connection region. If the distance is greater than a preset threshold, it is marked as a valid connection region, resulting in a coordinate dataset of the valid connection region. Based on the coordinate dataset of the valid connection region, combined with the coordinate data of blood vessel and nerve distribution, a region segmentation algorithm is used. For the coordinate points of the blood vessel and nerve distribution, if the distance between the coordinate point and the boundary of the valid connection region is less than a preset threshold, it is marked as a high-risk region, resulting in a coordinate dataset of the high-risk region. Using the coordinate dataset of the high-risk region, an A-path planning algorithm is used to search for a path from the tibia positioning point to the femoral connection region. Constraints are set to avoid the high-risk region, and the optimal path is calculated, resulting in a sequence of node coordinates for the optimal path. Based on the node coordinate sequence of the optimal path, a cubic spline interpolation method is used to smooth the node coordinate sequence to generate a continuous guide needle puncture path, thus obtaining the three-dimensional coordinate sequence of the guide needle puncture path.

[0096] For example, when obtaining coordinate data of the tibial positioning point and the femoral junction region from a three-dimensional anatomical model, a voxel labeling method can be used. The tibial positioning point is typically selected as the center of the tibial plateau. By scanning the flat area of ​​the proximal tibia in the three-dimensional model, the point with the highest voxel density is determined as the center coordinate, for example, (50, 30, 20). The boundary points of the femoral junction region can be obtained through curvature analysis, identifying the edge points of the articular surface at the distal end of the femur to form a boundary point set, such as (60, 35, 25), (62, 34, 24), etc. This method can accurately capture anatomical features, facilitating subsequent calculations.

[0097] In one possible implementation, the distance between the tibial positioning point and the femoral junction can be calculated using the Euclidean distance formula. Assuming the coordinates of the tibial positioning point are (50, 30, 20) and a boundary point of the femoral junction is (60, 35, 25), the distance between the two points is calculated using spatial geometry. If a preset threshold of 15 mm is set, and the calculated result is greater than 15 mm, then the region to which that boundary point belongs is marked as a valid junction. This selection method ensures the anatomical rationality of the junction region, providing a reliable basis for path planning.

[0098] Specifically, distance field analysis can be used to label high-risk areas based on the coordinate dataset of the effective connectivity region, combined with vascular and nerve distribution data. Assuming the vascular coordinates are (58, 33, 23), the nerve coordinates are (59, 34, 24), the boundary point of the effective connectivity region is (60, 35, 25), and the preset distance threshold is 2 mm, if the distance between a vascular point and the boundary point is less than 2 mm, it is marked as a high-risk area. This method can effectively identify areas close to key anatomical structures, reducing surgical risks.

[0099] When using the A* path planning algorithm to search for the optimal path from the tibial landmark to the femoral junction, high-risk areas can be designated as no-entry zones. In the path search from (50,30,20) to (60,35,25), the algorithm prioritizes nodes that avoid high-risk areas, such as detouring to (55,32,22). This path planning method ensures the safety of the guide needle puncture path while maintaining optimal path length.

[0100] In one embodiment, when performing cubic spline interpolation on the node coordinate sequence of the optimal path, the node coordinate sequence, such as (50,30,20), (55,32,22), or (60,35,25), can be input into the interpolation model to generate a smooth guide needle puncture path. The interpolated path coordinate points are continuous and have no acute angle changes; for example, a new coordinate point (52,31,21) is generated. This smoothing process ensures the stability of the guide needle during puncture and improves the smoothness of the surgical procedure.

[0101] Understandably, each step of the above method revolves around the precise localization of the knee joint's anatomical structures and the planning of a safe path. Through the accurate acquisition of coordinate data, the reasonable screening of distance calculations, the detailed marking of high-risk areas, and the optimization and smoothing of the path, the entire process forms a rigorous logical chain, providing reliable technical support for surgical planning.

[0102] As one implementation method in this embodiment, the process of real-time capturing the three-dimensional position coordinates and attitude angle of the guide needle includes:

[0103] The relative distance between the tip of the guide pin and the electromagnetic receiver array is calculated using a triangulation algorithm based on the electromagnetic field signal strength.

[0104] The real-time three-dimensional coordinates of the guide pin are determined based on the geometric layout of the receiver array.

[0105] Specifically, in step S103, a miniature electromagnetic transmitter is integrated at the front end of the guide needle, and an electromagnetic receiver array is arranged externally. The position coordinates and attitude angle of the guide needle in three-dimensional space are captured in real time through a triangulation algorithm based on the electromagnetic field signal strength, so as to obtain the real-time motion trajectory data of the guide needle.

[0106] More specifically, electromagnetic signal data is acquired from a miniature electromagnetic transmitter at the tip of the guide pin, and valid signals with field strength values ​​greater than a preset threshold are marked to obtain a field strength dataset of the valid signals. The strength of the valid signals is measured by multiple electromagnetic receivers in the receiving array, and the relative distance between the receivers and the tip of the guide pin is calculated to obtain a distance dataset. Based on the distance dataset, a triangulation algorithm is used to calculate the position coordinates of the tip of the guide pin in three-dimensional space, taking into account the geometric layout of the receiving array, to obtain the position coordinate dataset.

[0107] For example, when acquiring electromagnetic signal data from the miniature electromagnetic transmitter at the tip of the guide pin, valid signals can be filtered by setting a field strength threshold. The electromagnetic transmitter works by generating electromagnetic waves of a specific frequency, with signal strength attenuating with distance. Assuming a threshold of 0.5 mT, signals below this value are considered noise. In practice, the signal emitted by the tip of the guide pin is captured by a receiving array, and signals with field strength values ​​such as 0.8 mT and 0.6 mT are recorded as valid signals, forming a field strength dataset. This filtering method ensures data reliability and facilitates subsequent positioning calculations.

[0108] In one possible implementation, multiple electromagnetic receivers in a receiving array measure the effective signal strength and calculate the relative distance to the tip of the guide pin. The receivers are typically arranged at known coordinates in three-dimensional space, such as four receivers located at (0,0,0), (10,0,0), (0,10,0), and (0,0,10). Using a signal strength attenuation model, the distance can be estimated; a field strength of 0.8 mT corresponds to a distance of 5 mm, and 0.6 mT corresponds to 6 mm. This method utilizes the propagation characteristics of electromagnetic waves to ensure the accuracy of the distance dataset, providing a foundation for positioning.

[0109] Specifically, when calculating the three-dimensional coordinates of the guide pin's tip using the triangulation algorithm, the geometric layout of the receiving array must be considered. Triangulation is based on multi-point ranging, using receiver coordinates and distance datasets to deduce the transmitter's spatial position. Assuming the distance dataset is 5mm, 6mm, 5.5mm, and 6.2mm, combined with the receiver coordinates, the guide pin's tip coordinates can be determined as (4,3,2). To improve accuracy, a weighted average method can be introduced, prioritizing receiver data with higher signal strength. This method effectively addresses signal interference and improves the stability of the positioning results.

[0110] Triangulation algorithms can further improve accuracy when dealing with complex anatomical environments through iterative optimization. For example, in knee surgery, the guide needle needs to avoid bone and soft tissue, and the receiving array may be interfered with by metal implants. Iterative algorithms can generate a more reliable coordinate dataset through multiple calculations, eliminating abnormal distance data, such as optimizing from an initial estimate of (4.1, 3.2, 2.1) to (4, 3, 2). This approach adapts to the dynamic environment during surgery, ensuring real-time tracking of the guide needle position. The position coordinate dataset can be used to guide real-time adjustments of the guide needle. Assuming the target surgical point is (5, 4, 3) and the current guide needle tip coordinates are (4, 3, 2), the system can prompt adjustments to the direction and distance. This real-time feedback mechanism helps surgeons precisely control the guide needle path, improving the smoothness of the surgical procedure.

[0111] Understandably, the above method revolves closely around electromagnetic signal processing and three-dimensional positioning. From signal selection to distance calculation and triangulation, each step prioritizes data accuracy, forming a complete logical chain. This method is highly applicable in knee joint guide pin positioning and can effectively support surgical navigation needs.

[0112] As one implementation method in this embodiment, the dynamic path calibration algorithm includes the following steps:

[0113] The coordinate sequence of the real-time trajectory and the planned path is fitted using the least squares method;

[0114] Calculate the angle between the current direction of the guide needle's travel and the direction of the correction path;

[0115] If the included angle exceeds a preset angle threshold, direction adjustment parameters are generated and guide needle motion control parameters are updated.

[0116] Specifically, in step S104, if the deviation between the real-time motion trajectory data of the guide needle and the three-dimensional coordinate sequence of the planned path exceeds a preset threshold, a dynamic path calibration algorithm is adopted. The real-time trajectory and the planned path are fitted by the least squares method to generate a corrected path coordinate sequence and determine the adjusted direction of the guide needle.

[0117] More specifically, the real-time motion trajectory data of the guide pin is acquired to obtain a three-dimensional coordinate sequence of the real-time trajectory. For a preset planned path three-dimensional coordinate sequence, the Euclidean distance between the real-time trajectory three-dimensional coordinate sequence and the planned path three-dimensional coordinate sequence is calculated. If the Euclidean distance is greater than a preset deviation threshold, the trajectory segment dataset to be calibrated is determined. The real-time trajectory and the planned path in the trajectory segment dataset are fitted using the least squares method to obtain a three-dimensional coordinate dataset of the correction path. Based on the three-dimensional coordinate dataset of the correction path, the angle between the current travel direction of the guide pin and the direction of the correction path is calculated. If the angle is greater than a preset angle threshold, direction adjustment parameters are generated to obtain an adjusted travel direction dataset. Using the travel direction dataset, the motion control parameters of the guide pin are updated to obtain an updated real-time trajectory three-dimensional coordinate sequence.

[0118] For example, when acquiring real-time motion trajectory data of the guide needle, the displacement signal of the guide needle tip can be captured by an externally deployed sensor array to form a three-dimensional coordinate sequence. The sensor array typically employs optical or electromagnetic positioning technology to record the position of the guide needle in the surgical space in real time. In knee surgery, the coordinates of the guide needle tip are recorded every 0.1 seconds, forming sequences such as (2,3,4) and (2.1,3.2,4.1). This high-frequency sampling ensures the continuity of the trajectory data, providing a reliable basis for subsequent deviation analysis.

[0119] In one possible implementation, calculating the Euclidean distance between the real-time trajectory and the planned path requires pre-setting the coordinate sequence of the planned path, such as (2,3,5) or (2.2,3.3,5.2). The Euclidean distance is obtained by comparing the difference between the real-time coordinates and the planned coordinates. The distance between the real-time point (2.1,3.2,4.1) and the planned point (2.2,3.3,5.2) is approximately 1.1 mm. If the preset deviation threshold is 0.5 mm, this distance exceeds the limit, indicating a trajectory deviation, and that segment of the trajectory needs to be calibrated. This method can quickly identify deviation segments, facilitating precise adjustments.

[0120] Specifically, when fitting the deviation trajectory segment using the least squares method, the coordinates of the real-time trajectory can be fitted to the planned path. The real-time trajectory segment includes points (2.1, 3.2, 4.1) and (2.3, 3.4, 4.3), while the planned path segment is (2.2, 3.3, 5.2) and (2.4, 3.5, 5.4). The least squares method generates a smooth correction path, with output coordinates such as (2.15, 3.25, 4.8). This fitting method smooths trajectory fluctuations and generates correction data that better fits the planned path.

[0121] When calculating the angle between the guide needle's travel direction and the correction path direction, the direction vector can be derived based on the coordinate sequence of the correction path. The correction path runs from (2.15, 3.25, 4.8) to (2.4, 3.5, 5.0), with a direction vector of (0.25, 0.25, 0.2). If the current direction vector of the guide needle is (0.2, 0.3, 0.1), the angle is approximately 15 degrees. If the preset angle threshold is 10 degrees, the direction needs to be adjusted. This vector analysis method is intuitive and efficient, ensuring the guide needle travels in the correct direction.

[0122] In one embodiment, when generating direction adjustment parameters, the adjustment angle and step size can be calculated based on the included angle. A 15-degree included angle can be decomposed into adjustments in the horizontal and vertical directions, generating parameters such as "rotate 8 degrees to the right, adjust 5 degrees upwards". These parameters form a travel direction dataset, which is directly used to update the motion control of the guide needle. This parameterized adjustment method enables fine-grained control of the guide needle movement. After updating the guide needle motion control parameters, a new trajectory coordinate sequence can be generated through a real-time feedback mechanism. After adjustment, the guide needle moves from (2.3, 3.4, 4.3) to (2.35, 3.45, 4.7), which is closer to the planned path. This closed-loop control method can continuously optimize the trajectory, ensuring that the guide needle accurately conforms to the surgical requirements.

[0123] It should be noted that the above method is data-driven at its core, forming a complete logical chain from trajectory acquisition to direction adjustment, and is suitable for high-precision scenarios such as knee surgery. Each step aims for real-time performance and accuracy, supporting the continuity and reliability of surgical navigation.

[0124] As one implementation method in this embodiment, in step S105, based on the real-time motion trajectory data of the guide needle and the corrected path coordinate sequence, virtual reality rendering technology is used to overlay and display the three-dimensional anatomical model, the planned path and the current position of the guide needle, generating a dynamic image sequence containing a stereoscopic visual navigation interface, and obtaining intuitive guide needle movement guidance information.

[0125] Specifically, the process involves acquiring the real-time trajectory 3D coordinate sequence of the guide needle, the correction path 3D coordinate sequence, and pre-established 3D anatomical model data. Using data overlay technology, the real-time trajectory 3D coordinate sequence, the correction path 3D coordinate sequence, and the 3D anatomical model data are aligned in a virtual reality environment to obtain an initial 3D rendering dataset. Virtual reality rendering technology is used to process the initial 3D rendering dataset with lighting and texture mapping to generate a rendered image sequence. If the frame rate of the rendered image sequence is lower than a preset threshold, interpolation algorithms are used to supplement intermediate frames, resulting in a dynamic image sequence. Based on the dynamic image sequence, the relative positional deviation between the current position of the guide needle and the correction path 3D coordinate sequence is calculated to obtain deviation data. Visual feedback technology is used to overlay the deviation data and the dynamic image sequence onto the navigation interface to generate a navigation guidance dataset. If the deviation data is greater than a preset threshold, a highlighted prompt is generated to confirm the dynamic navigation guidance information.

[0126] For example, when acquiring the real-time three-dimensional coordinate sequence of the guide needle trajectory, the position data of the guide needle in the surgical space can be captured by an electromagnetic positioning system. In knee surgery, the coordinates of the guide needle tip are recorded every 0.05 seconds, forming sequences such as (1.5, 2.8, 3.6) and (1.6, 2.9, 3.7). The three-dimensional coordinate sequence of the correction path can be generated by a path planning algorithm, such as (1.7, 3.0, 3.8) and (1.8, 3.1, 3.9). The three-dimensional anatomical model data is generated based on preoperative CT scans, including the spatial structure of the knee joint bones and soft tissues. This multi-source data acquisition method ensures the accuracy of subsequent data overlay.

[0127] In one possible implementation, data overlay technology aligns real-time trajectories, correction paths, and 3D anatomical models within a virtual reality environment.

[0128] A coordinate transformation method based on feature points can be used to unify the coordinates of the three components. For example, anatomical landmarks of the femur and tibia in the knee joint can be selected as references, and coordinate alignment can be achieved through matrix transformation to generate an initial 3D rendering dataset. This method can ensure spatial consistency across different data sources, providing a reliable foundation for rendering.

[0129] Furthermore, when processing the initial 3D rendering dataset, virtual reality rendering technology can use the Phong lighting model to simulate lighting effects and add bone and soft tissue surface details to the anatomical model through texture mapping.

[0130] The guide needle trajectory is rendered as a red line, the correction path as a green line, and the anatomical model is displayed as a semi-transparent skeletal structure. If the rendering frame rate is lower than a preset threshold, such as 60 frames per second, intermediate frames can be generated using a linear interpolation algorithm. Transitional coordinates (1.7, 3.0, 3.9) are inserted between the coordinates (1.6, 2.9, 3.7) and (1.8, 3.1, 3.9) of two frames to create a smooth dynamic image sequence. This processing improves visual continuity.

[0131] Understandably, when calculating the relative positional deviation between the current position of the guide pin and the correction path, the distance between the real-time coordinates and the coordinates of the correction path can be compared. The current position of the guide pin is (1.6, 2.9, 3.7), and the correction path point is (1.7, 3.0, 3.8), with a deviation of approximately 0.17 mm. If the preset threshold is 0.1 mm, the deviation exceeds the limit. This deviation analysis provides a quantitative basis for navigation.

[0132] In one embodiment, visual feedback technology overlays deviation data onto the navigation interface. In the virtual reality headset, the deviation data is displayed as yellow numbers next to the guide needle trajectory, while a dynamic image sequence updates the relative position of the guide needle and the anatomical model in real time. If the deviation is greater than 0.1 mm, a highlighted prompt is generated, such as a flashing red arrow indicating the correction direction. This intuitive feedback allows the surgeon to quickly adjust the guide needle position.

[0133] It should be noted that the generation of the navigation guidance dataset relies on the fusion of deviation data and dynamic image sequences. The navigation interface displays that the guide needle should move 0.1mm to the right and adjust downwards by 0.05mm, while highlighting the correction path direction. This closed-loop guidance method improves the accuracy and real-time performance of surgical navigation.

[0134] As one implementation method in this embodiment, in step S106, if the position of the guide needle in the dynamic image sequence deviates from the correction path by more than a preset safety threshold, the deviation area is detected by the image processing algorithm, a deviation vector is generated and mapped to the three-dimensional coordinate space, and the precise adjustment angle and distance of the guide needle are determined.

[0135] The system acquires real-time 3D coordinate data of the guide pin and the 3D coordinate sequence of the correction path from a dynamic image sequence. It then uses an image processing algorithm to segment the dynamic image sequence, identifying the deviation region between the current position of the guide pin and the correction path, generating a first deviation dataset. If the deviation value in the first deviation dataset exceeds a preset safety threshold, a vector calculation method is used to extract the deviation direction and distance from the first deviation dataset, generating first deviation vector data. Using a coordinate mapping algorithm, the first deviation vector data is converted to a 3D coordinate space. Combined with the 3D coordinate sequence of the correction path, the adjustment angle and adjustment distance of the guide pin are calculated to obtain a first adjustment parameter dataset. Based on the first adjustment parameter dataset, an image rendering algorithm is used to process the navigation interface, superimposing the adjustment angle and adjustment distance onto the dynamic image sequence to generate a real-time feedback navigation guidance image sequence.

[0136] For example, when acquiring real-time three-dimensional coordinate data of the guide needle, an optical positioning system can capture the movement trajectory of the guide needle in the surgical space. In knee surgery, the coordinates of the guide needle tip are recorded every 0.1 seconds, forming sequences such as (2.0, 3.5, 4.0) and (2.1, 3.6, 4.1). The three-dimensional coordinate sequence of the correction path is generated based on preoperative planning, such as (2.2, 3.7, 4.2) and (2.3, 3.8, 4.3). The optical positioning system tracks the marked points on the guide needle using an infrared camera to ensure coordinate accuracy. This method provides a reliable data foundation for subsequent deviation analysis.

[0137] In one possible implementation, when the image processing algorithm performs region segmentation on a dynamic image sequence, it can use edge detection methods to identify the contours of the guide pin and the correction path. The Canny edge detection algorithm is used to extract the guide pin's position region, generating a binary image, which is then compared with a preset region of the correction path to identify the deviation region. Assuming the current position of the guide pin is (2.1, 3.6, 4.1) and the correction path point is (2.2, 3.7, 4.2), the deviation region is represented by the set of pixels where the guide pin deviates from the path. This method facilitates the generation of the first deviation dataset.

[0138] It should be noted that if the deviation value in the first deviation dataset exceeds the preset safety threshold, such as 0.15mm, the deviation vector calculation can be triggered.

[0139] Specifically, the vector calculation method extracts the deviation direction and distance by comparing the guide pin coordinates with the correction path coordinates. The deviation direction between the guide pin coordinates (2.1, 3.6, 4.1) and the path point (2.2, 3.7, 4.2) is along the positive X, Y, and Z axes, with a distance of approximately 0.17 mm, generating the first deviation vector data. This quantitative analysis provides a basis for adjusting the parameter calculations.

[0140] As one implementation method, when the coordinate mapping algorithm transforms the first deviation vector data to a three-dimensional coordinate space, it can use an affine transformation method. Through matrix operations, the deviation vector is mapped to the correction path coordinate system. The calculation requires the guide pin to be adjusted by 0.1mm along the X-axis and 0.1mm along the Y-axis, with a rotation angle of 5 degrees, generating the first adjustment parameter dataset. This method ensures the spatial consistency of the adjustment parameters.

[0141] In one embodiment, the image rendering algorithm can use texture overlay technology to visualize adjustment parameters when processing the navigation interface. In the navigation interface, the adjustment angle is displayed as a blue arrow, and the adjustment distance is marked with green numbers, overlaid on a dynamic image sequence. The rendering process maintains 60 frames per second to ensure the smoothness of the navigation guidance image sequence. This intuitive presentation allows the operator to quickly understand the adjustment requirements.

[0142] Understandably, the real-time feedback navigation guidance image sequence dynamically updates and adjusts parameters, forming a closed-loop guidance mechanism. After the guide pin is adjusted, the new coordinate sequence is compared with the correction path, and the deviation is reduced to 0.05mm, with a green confirmation mark displayed on the interface. This mechanism improves the real-time performance and accuracy of surgical navigation.

[0143] As one implementation method in this embodiment, in step S107, based on the deviation vector and the coordinate sequence of the correction path, the inverse kinematics algorithm is used to calculate the rotation angle and advance distance required for the guide needle adjustment, generate a control signal, and transmit it to the guide needle drive device through the electromagnetic feedback system to obtain the automatic adjustment command for the guide needle.

[0144] Specifically, a dynamic image sequence is acquired, and an image processing algorithm is used to segment the dynamic image sequence to obtain real-time 3D coordinate data of the guide pin and a 3D coordinate sequence of the correction path. Based on the real-time 3D coordinate data of the guide pin and the 3D coordinate sequence of the correction path, the deviation is calculated to generate a first deviation dataset. If the deviation value in the first deviation dataset exceeds a preset safety threshold, the deviation direction and distance are extracted using a vector calculation method to obtain first deviation vector data. Using an inverse kinematics algorithm, the rotation angle and advance distance of the guide pin are calculated based on the first deviation vector data and the 3D coordinate sequence of the correction path to generate a first adjustment parameter dataset. Based on the first adjustment parameter dataset, a control signal is generated and transmitted to the drive device through an electromagnetic feedback system to obtain an automatic adjustment command for the guide pin. Based on the automatic adjustment command, the movement trajectory of the guide pin is adjusted to generate corrected 3D coordinate data of the guide pin.

[0145] For example, when acquiring a dynamic image sequence, the movement of the guide needle in the surgical space can be captured by a high-speed infrared camera system to form a high-resolution dynamic image sequence.

[0146] It should be noted that this system typically combines optical marking technology, with specific reflective marks attached to the surface of the guide needle. The camera records the mark positions at a frequency of 100 frames per second, ensuring that the captured image sequence has high temporal resolution. In knee replacement surgery, every tiny displacement of the guide needle can be accurately recorded, forming a continuous stream of image data, providing a reliable basis for subsequent processing.

[0147] In one possible implementation, when performing region segmentation on a dynamic image sequence, the image processing algorithm can employ a deep learning-based segmentation method.

[0148] Specifically, a pre-trained convolutional neural network model is used to identify the feature regions of the guide needle and the correction path. The network model, trained with preoperative labeled data, can distinguish the metallic outline of the guide needle from the boundary of the surrounding soft tissue. The guide needle region is segmented into a set of white pixels, and the correction path region into a set of green pixels. The real-time 3D coordinates of the guide needle (e.g., 2.3, 3.8, 4.4) and the coordinates of the correction path (e.g., 2.4, 3.9, 4.5) are output. This method improves the robustness of the segmentation.

[0149] Understandably, when calculating the deviation, a first deviation dataset is generated by comparing the real-time coordinates of the guide pin with the coordinates of the correction path. The difference between the guide pin coordinates 2.3, 3.8, 4.4 and the path coordinates 2.4, 3.9, 4.5 indicates that the deviation direction is in the positive X, Y, Z axis direction, and the deviation distance is approximately 0.2 mm. If the preset safety threshold is 0.15 mm, subsequent processing is triggered. This axis-by-axis comparison method ensures the comprehensiveness of the deviation analysis.

[0150] In one embodiment, when generating the first deviation vector data, a vector projection method can be used to extract the deviation direction and distance. The deviation vector is constructed from the coordinate difference, with the direction pointing towards the correction path and the distance approximated by Euclidean distance. The deviation vector indicates that the guide needle needs to move 0.1 mm in the positive X-axis direction and 0.1 mm in the positive Y-axis direction. This quantification method provides a clear basis for the calculation of adjustment parameters.

[0151] Specifically, when using inverse kinematics algorithms to calculate the rotation angle and advance distance of the guide needle, iterative optimization can be performed based on the joint parameters of the robotic arm. The algorithm analyzes the angle between the deviation vector and the correction path, determining that the guide needle needs to rotate 6 degrees and advance 0.15 mm, generating the first adjustment parameter dataset. This method ensures that the adjustment commands are consistent with the geometric constraints of the surgical space. When generating control signals, the adjustment parameters can be converted into electrical signals for the electromagnetic feedback system via a digital-to-analog converter. Specifically, the signals drive the servo motors of the robotic arm to adjust the guide needle's posture.

[0152] In one embodiment, the electromagnetic feedback system transmits signals with a millisecond delay to ensure real-time command. After receiving the command to rotate 6 degrees, the guide pin quickly adjusts to the target angle.

[0153] Understandably, after adjusting the guide pin's trajectory, the correction effect can be verified by re-collecting coordinates. The adjusted guide pin coordinates become 2.4, 3.9, 4.5, consistent with the correction path, indicating successful trajectory correction. This closed-loop verification mechanism improves the reliability of navigation.

[0154] As one implementation method in this embodiment, the process of verifying the positioning accuracy of the guide pin includes:

[0155] Compare the adjusted guide pin position data with the calibration path coordinate sequence;

[0156] The average deviation distance is calculated using an error analysis algorithm.

[0157] If the average deviation distance is less than the sub-millimeter threshold, the positioning accuracy is deemed to meet the standard.

[0158] Specifically, in step S108, the adjusted position and attitude data of the guide pin are obtained from the execution result of the guide pin driving device. By comparing it with the coordinate sequence of the correction path, the positioning accuracy is calculated using an error analysis algorithm to determine whether the guide pin meets the requirements of sub-millimeter accuracy.

[0159] More specifically, the adjusted guide pin position and attitude data are obtained from the execution results of the guide pin driving device. A data parsing algorithm is used to extract the three-dimensional coordinates and orientation angles of the guide pin, resulting in a first position dataset and a first attitude dataset. Based on the first position dataset and the first attitude dataset, a data comparison is performed with a preset correction path coordinate sequence. A vector difference calculation method is used to calculate the deviation between the guide pin position and attitude, resulting in a first deviation dataset. For the first deviation dataset, a least squares algorithm is used for error analysis to calculate the average deviation distance between the guide pin position and the correction path, obtaining a positioning accuracy value. If the positioning accuracy value exceeds a preset sub-millimeter threshold, an accuracy failure signal is generated; if the positioning accuracy value is within the threshold range, an accuracy compliance signal is generated. The positioning state of the guide pin is determined based on the accuracy compliance signal or the accuracy failure signal.

[0160] For example, when obtaining the execution result of the guide pin driving device, the real-time position and attitude data of the guide pin can be collected by a high-precision sensor system.

[0161] Specifically, the sensors include a laser rangefinder and a gyroscope, used to capture the three-dimensional coordinates and orientation angle of the guide needle, respectively. In knee surgery, the laser rangefinder records the position of the guide needle tip at a frequency of 200 times per second, generating coordinate data such as 2.5, 3.7, 4.3; the gyroscope records the tilt angle of the guide needle relative to the Z-axis, such as 5 degrees. This high-frequency acquisition method ensures the real-time nature and accuracy of the data, providing reliable input for subsequent analysis.

[0162] In one possible implementation, the data parsing algorithm can use filtering techniques to extract valid coordinates and angles.

[0163] As one implementation method, the Kalman filter algorithm is used to process sensor noise and smooth the three-dimensional coordinates and orientation angles of the guide pin. The original coordinates 2.5, 3.7, 4.3 are optimized to 2.51, 3.72, 4.29 after filtering, and the angle is corrected from 5 degrees to 4.8 degrees. This processing method improves the accuracy of the first position dataset and the first attitude dataset, laying the foundation for deviation analysis.

[0164] It should be noted that when comparing with the calibration path coordinate sequence, data alignment can be performed using a point-by-point matching method.

[0165] Specifically, the calibration path coordinates, such as 2.6, 3.8, 4.4, are compared with the guide needle coordinates, 2.51, 3.72, 4.29, axis by axis, and the vector difference is calculated. The deviation is 0.09 for the X-axis, 0.08 for the Y-axis, and 0.11 for the Z-axis. This axis-by-axis analysis method ensures the comprehensiveness of the deviation calculation, and the generated first deviation dataset clearly reflects the direction of the guide needle deviation.

[0166] In one embodiment, when analyzing the first deviation dataset using the least squares method, the average deviation distance can be calculated by fitting the deviation point set. Multiple sets of collected deviation data show that the average distance between the guide pin position and the correction path is 0.12 mm. If the preset sub-millimeter threshold is 0.1 mm, a positioning accuracy value of 0.12 mm triggers an accuracy failure signal. This analysis method quantifies the positioning error, providing a basis for subsequent adjustments.

[0167] When generating accuracy signals, the analysis results can be converted into inputs for the control system via a signal encoder. A signal indicating insufficient accuracy may suggest that the guide pin needs further fine-tuning, while a signal indicating satisfactory accuracy confirms that the current positioning status is reliable.

[0168] Specifically, the accuracy-compliant signal can be directly transmitted to the navigation system, instructing the guide pin to continue along the current path. This signal generation method improves the system's automation level.

[0169] Understandably, when determining the positioning status of the guide pin, trend analysis can be performed by combining historical deviation data.

[0170] As one implementation method, if the deviation shown by three consecutive accuracy-compliant signals is within 0.05 mm, it confirms that the guide pin is stably on the target path. If the accuracy-uncompliant signal appears repeatedly, it indicates a possible problem with the robotic arm calibration. In a stable state, the guide pin coordinates continuously match the calibration path, indicating that the positioning system is operating well. This trend analysis enhances the robustness of the state determination and provides reliable assurance for surgical navigation.

[0171] As one implementation method in this embodiment, in step S109, if the positioning accuracy meets the preset threshold, the superimposed display of the guide needle position and the three-dimensional anatomical model is updated through the virtual reality interface to generate the final navigation image sequence, determine the real-time guidance data of the surgical path, and complete the closed-loop control of the entire process of guide needle positioning.

[0172] Specifically, guide needle position data and 3D anatomical model data are acquired from a virtual reality interface. A coordinate transformation algorithm is used to map the guide needle position data to the reference coordinate system of the 3D anatomical model, resulting in a first overlay dataset. Based on the first overlay dataset, an image rendering algorithm is used to generate navigation images containing the guide needle position and the 3D anatomical structure, resulting in a first image sequence. For the first image sequence and preset surgical path data, a vector matching algorithm is used to calculate the deviation between the guide needle position and the surgical path data, resulting in first deviation data. It is determined whether the first deviation data is within a preset threshold range, resulting in first path guidance data. If the first path guidance data meets the preset threshold, a closed-loop control algorithm is used to update the motion parameters of the guide needle, resulting in real-time guidance data. Based on the real-time guidance data, the final surgical navigation path is determined.

[0173] For example, obtaining guide needle position data and three-dimensional anatomical model data from a virtual reality interface can be achieved through a head-mounted display device and a motion tracking system.

[0174] Specifically, the head-mounted display has a built-in high-resolution camera that captures optical marks on the guide pin, generating positional data such as coordinates 3.2, 4.1, and 5.0. The 3D anatomical model data is generated from preoperative CT scans, including the skeletal and soft tissue structures of the knee joint. This approach ensures the real-time nature of the data source and the accuracy of the anatomical structure, providing a foundation for subsequent mapping.

[0175] In one possible implementation, a coordinate transformation algorithm is used to map the guide needle position data to the reference coordinate system of the three-dimensional anatomical model.

[0176] It should be noted that the coordinate transformation is based on the principle of rigid body transformation. The guide pin coordinates are aligned with the model coordinates through a rotation matrix and a translation vector. The guide pin coordinates 3.2, 4.1, 5.0 are transformed to 3.25, 4.08, 4.95, forming the first overlay dataset. This mapping method ensures spatial consistency between the two sets of data, facilitating subsequent image generation.

[0177] Specifically, the image rendering algorithm can be based on the principle of stereomicroscopy to transform the first overlay dataset into a navigation image.

[0178] In one embodiment, the rendering algorithm utilizes GPU acceleration to generate a real-time image sequence containing the guide needle position and knee joint structures. The guide needle tip is marked in red, and the bones are rendered in gray transparent, generating a first image sequence. This visualization method intuitively presents the relative position of the guide needle and anatomical structures, facilitating intraoperative navigation.

[0179] As one implementation method, a vector matching algorithm is used to calculate the deviation between the guide needle position and the preset surgical path. The surgical path data is 3.3, 4.2, 5.1, and the guide needle positions are 3.25, 4.08, 4.95. After comparing along each axis, the deviations are 0.05, 0.12, and 0.15, forming the first deviation data. This point-by-point analysis method clearly quantifies the direction of deviation, providing a basis for path guidance.

[0180] In one embodiment, when determining whether the first deviation data is within a threshold range, the preset threshold is 0.1 mm. Deviations of 0.05, 0.12, and 0.15 indicate that the Y-axis and Z-axis exceed the threshold, generating first path guidance data and prompting an adjustment of the guide needle direction. This determination method improves the accuracy of navigation.

[0181] Understandably, the closed-loop control algorithm updates the guide needle motion parameters based on deviation data. Based on deviations of 0.12 and 0.15, the algorithm adjusts the robotic arm's rotation angle and propulsion speed, generating real-time guidance data. This closed-loop mechanism ensures the guide needle dynamically follows the target path. When determining the final surgical navigation path, the real-time guidance data is compared with the surgical path data. If the deviation remains below 0.1 mm, the path is considered reliable. This approach, through multi-dimensional data verification, ensures the stability of the navigation path and provides reliable guidance for the surgery.

[0182] Based on this, this invention provides a guide needle path navigation method based on electromagnetic sensing to assist in the precise positioning of medical devices in the knee joint region. The system analyzes the patient's preoperative MRI images to generate a three-dimensional anatomical model including the distribution of blood vessels and nerves, and uses an intelligent algorithm to plan the movement path of the guide needle. During positioning, the spatial position of the guide needle is captured in real time using electromagnetic positioning technology, and combined with a dynamic path calibration algorithm and virtual reality display to generate navigation guidance information. When the guide needle deviates from the preset path, the system automatically calculates adjustment parameters and performs precise adjustments through electromagnetic feedback, achieving sub-millimeter-level positioning accuracy. This invention significantly improves the accuracy of device positioning and operational safety by integrating electromagnetic sensing, dynamic calibration, and virtual reality technologies.

[0183] Example 2

[0184] In this embodiment, a computer terminal device is provided, including:

[0185] One or more processors;

[0186] A memory, coupled to the processor, for storing one or more programs;

[0187] When the one or more programs are executed by the one or more processors, the one or more processors implement the methods in the above embodiments.

[0188] In this embodiment, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the methods described in the above embodiments.

[0189] In this embodiment, an electronic device is also provided, including a memory and a processor. The memory stores a computer program, and the processor is configured to run the computer program to perform the methods described in the above embodiments.

[0190] The aforementioned program can run on a processor or be stored in memory (or a computer-readable medium). Computer-readable media include both permanent and non-permanent, removable and non-removable media, and information storage can be achieved by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0191] These computer programs may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes can be implemented using different modules, and different steps can be implemented using different modules.

[0192] This embodiment provides such a device or system. The system, referred to as an electromagnetic sensing-based needle path navigation system, includes:

[0193] The three-dimensional structural information acquisition module is used to acquire three-dimensional structural information of the knee joint from the patient's preoperative MRI image data and generate a three-dimensional anatomical model including the distribution of blood vessels and nerves.

[0194] An artificial intelligence path planning module is used to generate a three-dimensional coordinate sequence of the optimal guide needle puncture path for the three-dimensional anatomical model.

[0195] The electromagnetic positioning module is used to capture the three-dimensional position coordinates and attitude angle of the guide needle in real time through a miniature electromagnetic transmitter integrated at the front end of the guide needle and an externally arranged electromagnetic receiver array.

[0196] The dynamic path calibration module is used to generate a corrected path coordinate sequence when the deviation between the real-time movement trajectory of the guide needle and the planned path exceeds a preset threshold.

[0197] The virtual reality navigation module is used to generate a dynamic image sequence containing guide needle movement guidance information based on the real-time motion trajectory and the coordinate sequence of the correction path.

[0198] The deviation detection module is used to generate a deviation vector and map it to a three-dimensional coordinate space when the position of the guide pin deviates from the correction path by more than a preset safety threshold in a dynamic image sequence.

[0199] The guide needle adjustment control module is used to generate control signals based on the deviation vector and the coordinate sequence of the correction path to adjust the direction of the guide needle's movement;

[0200] The positioning accuracy verification module is used to verify whether the adjusted guide pin positioning accuracy meets the sub-millimeter level requirements.

[0201] The navigation image update module is used to update the navigation image sequence in the virtual reality interface when the positioning accuracy meets the standard, thus completing closed-loop control.

[0202] As one implementation method in this embodiment, the three-dimensional structural information acquisition module includes:

[0203] Image segmentation unit is used to separate the tibia, femur and posterior cruciate ligament remnant regions from MRI image data;

[0204] Skeletal labeling unit, used to label skeletal regions using a threshold segmentation algorithm;

[0205] The vascular and neural labeling unit is used to expand and label vascular and neural regions from the skeletal region boundary using a region growing algorithm.

[0206] As one implementation method in this embodiment, the artificial intelligence path planning module includes:

[0207] The geometric constraint analysis unit is used to analyze the geometric constraints of the tibial positioning point and the femoral connection area.

[0208] The path avoidance unit is used to generate a node coordinate sequence by combining vascular and nerve distribution data and using the A* path planning algorithm to avoid high-risk areas.

[0209] The path smoothing unit is used to smooth the node coordinate sequence using cubic spline interpolation.

[0210] As one implementation method in this embodiment, the electromagnetic positioning module includes:

[0211] The signal strength calculation unit is used to calculate the relative distance between the tip of the guide pin and the electromagnetic receiver array using a triangulation algorithm based on the electromagnetic field signal strength.

[0212] The coordinate resolution unit is used to determine the real-time three-dimensional coordinates of the guide pin based on the geometric layout of the receiver array.

[0213] As one implementation method in this embodiment, the dynamic path calibration module includes:

[0214] The trajectory fitting unit is used to fit the coordinate sequence of the real-time trajectory and the planned path using the least squares method.

[0215] The direction deviation calculation unit is used to calculate the angle between the current travel direction of the guide needle and the correction path direction;

[0216] The parameter generation unit is used to generate direction adjustment parameters and update guide needle motion control parameters when the included angle exceeds a preset angle threshold.

[0217] As one implementation method in this embodiment, the positioning accuracy verification module includes:

[0218] The data comparison unit is used to compare the adjusted guide pin position data with the correction path coordinate sequence.

[0219] The error calculation unit is used to calculate the average deviation distance through an error analysis algorithm.

[0220] The determination unit is used to determine that the positioning accuracy meets the standard when the average deviation distance is less than the sub-millimeter threshold.

[0221] The system or apparatus is used to implement the functions of the methods in the above embodiments. Each module in the system or apparatus corresponds to each step in the method, as has been described in the method and will not be repeated here.

[0222] The above implementation method solves the problem of guide needle path navigation based on electromagnetic sensing in related technologies, thereby ensuring that the problems existing in the prior art are resolved.

[0223] Example 3

[0224] Based on the same general inventive concept, this invention also provides a needle path navigation system based on electromagnetic sensing. The needle path navigation system based on electromagnetic sensing provided by this invention will be described below. The needle path navigation system based on electromagnetic sensing described below can be referred to in correspondence with the needle path navigation method based on electromagnetic sensing described above. Figure 4 As shown, the system includes:

[0225] The three-dimensional structural information acquisition module is used to obtain the three-dimensional structural information of the knee joint from the patient's preoperative MRI image data. It uses image segmentation algorithms to separate the tibia, femur and posterior cruciate ligament stump regions, generate a three-dimensional anatomical model including the distribution of blood vessels and nerves, and determine the initial surgical planning area.

[0226] The AI ​​path planning module is used to generate the optimal guide needle puncture path for a three-dimensional anatomical model by analyzing the geometric constraints of the tibial positioning point and the femoral connection area, combined with vascular and nerve distribution data, and obtaining the three-dimensional coordinate sequence of the planned path.

[0227] The electromagnetic positioning module is used to integrate a miniature electromagnetic transmitter at the front end of the guide needle and arrange an electromagnetic receiver array externally. Through the triangulation positioning algorithm of electromagnetic field signal strength, it captures the position coordinates and attitude angle of the guide needle in three-dimensional space in real time and obtains the real-time motion trajectory data of the guide needle.

[0228] The dynamic path calibration module is used to generate a corrected path coordinate sequence and determine the adjusted guide needle travel direction if the deviation between the real-time motion trajectory data of the guide needle and the three-dimensional coordinate sequence of the planned path exceeds a preset threshold.

[0229] The virtual reality navigation module is used to overlay and display the three-dimensional anatomical model, the planned path and the current position of the guide needle based on the real-time motion trajectory data of the guide needle and the corrected path coordinate sequence using virtual reality rendering technology. This generates a dynamic image sequence containing a stereoscopic visual navigation interface, providing intuitive guide needle movement guidance information.

[0230] The deviation detection module is used to detect the deviation area through image processing algorithms if the position of the guide needle in the dynamic image sequence deviates from the correction path by more than a preset safety threshold, generate a deviation vector and map it to a three-dimensional coordinate space, and determine the precise adjustment angle and distance of the guide needle.

[0231] The guide needle adjustment control module is used to calculate the required rotation angle and advance distance for guide needle adjustment based on the deviation vector and the coordinate sequence of the correction path using an inverse kinematics algorithm, generate a control signal, and transmit it to the guide needle drive device through an electromagnetic feedback system to obtain the automatic adjustment command for the guide needle;

[0232] The positioning accuracy verification module is used to obtain the adjusted guide pin position and attitude data from the execution results of the guide pin drive device. By comparing it with the calibration path coordinate sequence, the positioning accuracy is calculated using an error analysis algorithm to determine whether the guide pin meets the requirements of sub-millimeter accuracy.

[0233] The navigation image update module is used to update the superimposed display of the guide needle position and the three-dimensional anatomical model through the virtual reality interface if the positioning accuracy meets the preset threshold, generate the final navigation image sequence, determine the real-time guidance data of the surgical path, and complete the closed-loop control of the entire process of guide needle positioning.

[0234] It should be understood that the guide needle path navigation system based on electromagnetic sensing provided in the embodiments of the present invention possesses all the advantages of the guide needle path navigation method based on electromagnetic sensing provided in the above embodiments.

[0235] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included 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 guide needle path navigation system based on electromagnetic sensing, characterized in that, The system includes: The three-dimensional structural information acquisition module is used to acquire three-dimensional structural information of the knee joint from the patient's preoperative MRI image data and generate a three-dimensional anatomical model including the distribution of blood vessels and nerves. An artificial intelligence path planning module is used to generate a three-dimensional coordinate sequence of the optimal guide needle puncture path for the three-dimensional anatomical model. The electromagnetic positioning module is used to capture the three-dimensional position coordinates and attitude angle of the guide needle in real time through a miniature electromagnetic transmitter integrated at the front end of the guide needle and an externally arranged electromagnetic receiver array. The dynamic path calibration module is used to generate a corrected path coordinate sequence when the deviation between the real-time movement trajectory of the guide needle and the planned path exceeds a preset threshold. The virtual reality navigation module is used to generate a dynamic image sequence containing guide needle movement guidance information based on the real-time motion trajectory and the coordinate sequence of the correction path. The deviation detection module is used to generate a deviation vector and map it to a three-dimensional coordinate space when the position of the guide pin deviates from the correction path by more than a preset safety threshold in a dynamic image sequence. The guide needle adjustment control module is used to generate control signals based on the deviation vector and the coordinate sequence of the correction path to adjust the direction of the guide needle's movement; The positioning accuracy verification module is used to verify whether the adjusted guide pin positioning accuracy meets the sub-millimeter level requirements. The navigation image update module is used to update the navigation image sequence in the virtual reality interface when the positioning accuracy meets the standard, thus completing closed-loop control.

2. The system according to claim 1, characterized in that, The three-dimensional structural information acquisition module includes: Image segmentation unit is used to separate the tibia, femur and posterior cruciate ligament remnant regions from MRI image data; Skeletal labeling unit, used to label skeletal regions using a threshold segmentation algorithm; The vascular and neural labeling unit is used to expand and label vascular and neural regions from the skeletal region boundary using a region growing algorithm.

3. The system according to claim 1, characterized in that, The artificial intelligence path planning module includes: The geometric constraint analysis unit is used to analyze the geometric constraints of the tibial positioning point and the femoral connection area. The path avoidance unit is used to generate a node coordinate sequence by combining vascular and nerve distribution data and using the A* path planning algorithm to avoid high-risk areas. The path smoothing unit is used to smooth the node coordinate sequence using cubic spline interpolation.

4. The system according to claim 1, characterized in that, The electromagnetic positioning module includes: The signal strength calculation unit is used to calculate the relative distance between the tip of the guide pin and the electromagnetic receiver array using a triangulation algorithm based on the electromagnetic field signal strength. The coordinate resolution unit is used to determine the real-time three-dimensional coordinates of the guide pin based on the geometric layout of the receiver array.

5. The system according to claim 1, characterized in that, The dynamic path calibration module includes: The trajectory fitting unit is used to fit the coordinate sequence of the real-time trajectory and the planned path using the least squares method. The direction deviation calculation unit is used to calculate the angle between the current travel direction of the guide needle and the correction path direction; The parameter generation unit is used to generate direction adjustment parameters and update guide needle motion control parameters when the included angle exceeds a preset angle threshold.

6. The system according to claim 1, characterized in that, The positioning accuracy verification module includes: The data comparison unit is used to compare the adjusted guide pin position data with the correction path coordinate sequence. The error calculation unit is used to calculate the average deviation distance through an error analysis algorithm. The determination unit is used to determine that the positioning accuracy meets the standard when the average deviation distance is less than the sub-millimeter threshold.

7. A computer terminal device, characterized in that, include: One or more processors; A memory, coupled to the processor, for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the electromagnetic sensing-based needle path navigation system as described in any one of claims 1-6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the guide needle path navigation system based on electromagnetic sensing as described in any one of claims 1-6.

Images