Joint soft tissue balancing method, device, electronic device and storage medium

Abstract

The application provides a joint soft tissue balancing method and device, electronic equipment and storage medium, relates to the technical field of medical instruments, and comprises the following steps: fusing joint internal pressure data and joint pose data, establishing the correlation between joint contact force and joint space attitude; based on the joint gap displacement data collected during the operation of the patient and the corresponding joint contact force response data, a soft tissue stiffness model is constructed; based on the soft tissue stiffness model, the joint gap data, and the joint gap change amount, the joint contact force prediction value of the candidate prosthesis specification combination is determined; an optimization function is constructed, each specification combination of the candidate prosthesis is traversed, and the specification combination corresponding to the optimal value of the optimization function is taken as a recommended configuration scheme. The method and device provided by the application construct an intelligent closed-loop solution of "measurement-modeling-prediction-optimization" driven by data as a whole, and realize accurate, efficient and personalized joint soft tissue balancing.

Description

Technical Field

[0001] This application relates to the field of medical device technology, and in particular to a method, apparatus, electronic device and storage medium for balancing soft tissues of a joint. Background Technology

[0002] In shoulder replacement surgery, achieving ideal soft tissue balance is one of the key factors determining the success of the surgery. A well-balanced joint ensures postoperative joint stability, provides maximum pain-free range of motion, and helps prolong the long-term survival rate of the prosthesis.

[0003] The relevant techniques primarily rely on the surgeon's clinical experience and subjective intuition to achieve joint soft tissue balance. The surgeon passively moves the joint during surgery, judging the appropriate soft tissue tension by feel. This method is highly dependent on the surgeon's personal experience, lacks objective and quantifiable standards, and makes it difficult to guarantee the consistency and repeatability of results.

[0004] Therefore, how to achieve precise, efficient, and personalized joint soft tissue balance has become a technical problem that the industry urgently needs to solve. Summary of the Invention

[0005] This application provides a method, apparatus, electronic device, and storage medium for balancing joint soft tissues, addressing the technical problem of how to achieve precise, efficient, and personalized balancing of joint soft tissues.

[0006] This application provides a method for balancing soft tissues in a joint, including: By integrating intra-articular pressure data collected from intelligent prosthesis trial models and joint pose data collected from surgical navigation systems, a correlation between joint contact force and joint spatial posture is established. Based on the joint space displacement data and corresponding joint contact force response data collected during the patient's operation, a soft tissue stiffness model of the patient is constructed. Based on the soft tissue stiffness model, joint clearance data under different joint spatial postures, and the change in joint clearance, the predicted joint contact force values ​​of the candidate prosthesis specification combination are determined within the full range of motion of the joint; the joint clearance data under different joint spatial postures are determined based on the correlation; the change in joint clearance is caused by the change in geometric parameters in the candidate prosthesis specification combination. An optimization function is constructed based on the correlation and the predicted joint contact force. The various specification combinations of the candidate prostheses are traversed, and the specification combination corresponding to the optimal value of the optimization function is taken as the recommended configuration scheme of the candidate prosthesis.

[0007] In some embodiments, establishing the correlation between joint contact force and joint spatial posture by fusing intra-articular pressure data collected from intelligent prosthesis trial models and joint pose data collected from surgical navigation systems includes: The joint pose data collected by the surgical navigation system is time-interpolated to synchronize the joint pose data with the intra-articular pressure data. The position coordinates of the load center are mapped from the intelligent prosthesis trial model coordinate system to the glenoid coordinate system to obtain the motion trajectory of the load center in the glenoid coordinate system. Based on the time-synchronized joint pose data, the intra-joint pressure data, and the motion trajectory of the load center, a four-dimensional tensor for characterizing the correlation is constructed on the discretized joint angle grid. The four-dimensional tensor is used to store the average contact force, contact force standard deviation, shear force vector, and number of data sampling points within each angular interval.

[0008] In some embodiments, the time interpolation of the joint pose data acquired by the surgical navigation system includes: The rotational quaternions in the joint pose data are interpolated using quaternion spherical linear interpolation; Linear interpolation is used to interpolate the three-dimensional translation vector in the joint pose data.

[0009] In some embodiments, constructing the patient's soft tissue stiffness model based on intraoperative joint space displacement data and corresponding joint contact force response data includes: Axial traction test was performed on the patient under a specific joint posture to obtain the patient's joint space displacement data and joint contact force response data; The patient's joint space displacement data and joint contact force response data were nonlinearly fitted to construct a soft tissue stiffness model.

[0010] In some embodiments, determining the predicted joint contact force values ​​of candidate prosthesis combinations across the entire joint range of motion based on the soft tissue stiffness model, joint space data under different joint spatial postures, and joint space variation includes: Determine the geometric parameters that need to be adjusted in the candidate prosthesis specification combination; the geometric parameters include at least one of the following: liner thickness, eccentricity size, and eccentricity rotation angle; Based on the changes in the geometric parameters, positive kinematics and geometric reconstruction calculations are performed to obtain the changes in joint space under different joint spatial postures. By substituting the joint space data and joint space variation under different joint spatial postures into the soft tissue stiffness model, the predicted joint contact force values ​​of the candidate prosthesis specification combination are obtained within the full range of joint motion.

[0011] In some embodiments, constructing the optimization function based on the correlation and the predicted joint contact force includes: The optimization function is constructed based on at least one of the full-radius tension consistency constraint function, the joint center stability constraint function, and the peak pressure penalty function. The full-radius tension consistency constraint function is used to constrain the predicted joint contact force value to remain within the ideal range throughout the entire joint range of motion; the joint center stability constraint function is used to constrain the load center trajectory obtained based on the correlation relationship to be close to the geometric center of the glenoid cavity; and the peak pressure penalty function is used to penalize the predicted joint contact force value from reaching its maximum value.

[0012] In some embodiments, traversing the various size combinations of the candidate prostheses includes: A grid search is performed within the search space comprised of the various size combinations of the candidate prostheses.

[0013] This application provides a joint soft tissue balancing device, comprising: The measurement module is used to integrate intra-articular pressure data collected by the intelligent prosthesis trial model and joint pose data collected by the surgical navigation system to establish the correlation between joint contact force and joint spatial posture. The modeling module is used to construct a soft tissue stiffness model of the patient based on the joint space displacement data and corresponding joint contact force response data collected during the operation. The prediction module is used to determine the predicted joint contact force values ​​of candidate prosthesis specifications within the full range of motion of the joint based on the soft tissue stiffness model, joint space data under different joint spatial postures, and the change in joint space; the joint space data under different joint spatial postures is determined based on the correlation relationship; the change in joint space is caused by the change in geometric parameters in the candidate prosthesis specifications. An optimization module is used to construct an optimization function based on the correlation and the predicted joint contact force value, traverse all specification combinations of the candidate prostheses, and take the specification combination corresponding to the optimal value of the optimization function as the recommended configuration scheme of the candidate prosthesis.

[0014] This application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the joint soft tissue balancing method.

[0015] This application provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the joint soft tissue balancing method.

[0016] The joint soft tissue balancing method, device, electronic device, and storage medium provided in this application fundamentally solve the problem of separation between pressure values ​​and spatial posture information in related technologies by establishing the correlation between joint contact force and joint spatial posture, providing a clear and accurate targeting basis for subsequent prosthesis adjustment; by constructing a soft tissue stiffness model of the patient, the prediction is made more consistent with the patient's actual physiological condition, enabling the assessment and planning of soft tissue balancing to address individual differences in the patient's soft tissue; by virtually predicting the joint contact force of candidate prosthesis specifications across the entire range of motion, the problem of soft tissue edema and prolonged operation time caused by repeated mold changes during surgery is solved; by constructing an optimization function and iteratively searching to recommend the optimal solution, the scientific nature, accuracy, and final joint function effect of surgical decisions are maximized; and a comprehensive intelligent closed-loop solution based on a patient-individualized model and driven by data, namely "measurement-modeling-prediction-optimization," is constructed, achieving accurate, efficient, and personalized joint soft tissue balancing, which helps improve postoperative function and long-term outcomes for patients. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating the joint soft tissue balancing method provided in this application.

[0020] Figure 2 This is a schematic diagram of the joint soft tissue balancing device provided in this application.

[0021] Figure 3 This is a schematic diagram of the joint soft tissue balance system provided in this application.

[0022] Figure 4 This is a schematic diagram of the structure of the electronic device provided in this application. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0024] It should be noted that the terms "first," "second," etc., used in this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps, units, or modules is not necessarily limited to those explicitly listed, but may include other steps, units, or modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0025] The joint soft tissue balancing methods in related technologies have the following shortcomings: (1) Decoupling of "force" and "position" information leads to ambiguous diagnosis: The intelligent trial molding method in related technologies can only provide real-time pressure values, but cannot accurately correlate these pressure values ​​with specific joint spatial postures (such as abduction angle and rotation angle). Doctors cannot know in which motion phase "excessive tension" or "relaxation" occurs, resulting in a lack of precise targeting basis when adjusting the prosthesis planning.

[0026] (2) Relying on the "trial and error method" is inefficient and prone to iatrogenic injury: When a joint is found to be too loose or too tight, doctors usually need to repeatedly change solid pads of different thicknesses for testing. This method not only significantly prolongs the operation time and increases the risk of infection, but may also cause soft tissue edema and damage due to repeated insertion and removal of trial molds, or cause excessive joint contraction due to experience-based judgment errors, affecting postoperative mobility.

[0027] (3) Ignoring soft tissue nonlinearity and individual patient differences leads to inaccurate planning: Related techniques often simplify the soft tissues around the joint as linear springs or use a uniform pressure threshold for assessment. However, the compliance (stiffness) of soft tissues such as the joint capsule, rotator cuff, and deltoid muscle varies significantly among different patients, and their mechanical behavior itself has nonlinear characteristics. Using a single assessment standard cannot achieve true personalized balance, which may lead to postoperative joint instability, abnormal wear, or limited mobility.

[0028] In order to address the shortcomings of related technologies,Figure 1 This is a flowchart illustrating the joint soft tissue balancing method provided in this application, as shown below. Figure 1 As shown, the method includes steps 110, 120, 130 and 140.

[0029] Step 110: Integrate the intra-articular pressure data collected by the intelligent prosthesis trial model and the joint pose data collected by the surgical navigation system to establish the correlation between joint contact force and joint spatial posture.

[0030] Specifically, the joint soft tissue balancing method provided in this application is executed by a joint soft tissue balancing device or system. This device can be implemented in software, such as a joint soft tissue balancing program; or it can be a device that executes the joint soft tissue balancing method, such as a terminal, computer, or server.

[0031] In shoulder replacement surgery, surgeons temporarily implant a smart prosthesis mold integrating pressure sensors into the patient's joint. This smart prosthesis mold can collect intra-articular pressure data in real time. Intra-articular pressure data here is a general term for mechanical data, which may specifically include, but is not limited to: the total contact force amplitude of the joint contact surface (e.g., in Newtons (N) or pounds (lbs), the distribution of contact pressure on the prosthesis sensor plane, and the two-dimensional or three-dimensional coordinates of the center of load (CoL) calculated from the pressure distribution. The pressure sensing device can be a piezoresistive, capacitive, or piezoelectric sensor array.

[0032] Surgical navigation systems (such as optical or electromagnetic navigation systems) track and acquire joint pose data in real time using tracking markers fixed to the patient's bones (such as the scapula) and surgical instruments / prostheses (such as the humeral prosthesis). Joint pose data here is a general term for kinematic data, used to describe the relative position and orientation between two or more components within the joint. Specifically, it can be represented as the humeral prosthesis coordinate system relative to the glenoid (or scapular) coordinate system. The homogeneous transformation matrix contains the three-dimensional rotation and three-dimensional translation information between the two.

[0033] Because the sampling devices and sampling frequencies in intelligent prosthesis trial models and surgical navigation systems are usually different (e.g., pressure sensors use 100Hz, navigation systems use 60Hz, where Hz is the unit of frequency, Hertz), and their data reside in their respective independent coordinate systems, fusion is a necessary prerequisite for establishing a correlation. The fusion process can include both time synchronization and spatial synchronization.

[0034] (1) Time synchronization: The aim is to match each pressure data point with a precise pose data at the same time. This can be achieved by time interpolating the data stream with a low sampling frequency (such as navigation data) to generate data that is aligned with the timestamps of the high-frequency data stream (such as pressure data).

[0035] (2) Spatial synchronization: The aim is to unify data from different coordinate systems into the same global coordinate system, such as the glenoid coordinate system. This requires establishing a transformation matrix between the sensor coordinate system, the prosthesis coordinate system and the glenoid coordinate system through pre-calibrated or known geometric relationships, so that the load center position measured in the local coordinate system can be mapped to the global joint coordinate system in real time.

[0036] After achieving spatiotemporal synchronization, the system guides the surgeon to perform passive joint movements within a certain range during surgery (e.g., flexion, extension, internal / external rotation, abduction, etc.). During this process, the system continuously records the synchronized data stream. To establish the correlation between joint contact force and joint spatial posture, the system can discretize the continuous motion. For example, the joint's range of motion (e.g., a spherical space defined by abduction / adduction angles and external / internal rotation angles) can be divided into several angular grid intervals. Then, the system assigns the massive amount of collected data points to the corresponding angular grids based on their respective joint postures and calculates the statistical values ​​of key mechanical indicators within each grid, such as average contact force, standard deviation of force, and average load center position. The resulting database or multidimensional data structure, mapping joint spatial posture (angular intervals) to a set of mechanical indicators, constitutes the correlation defined in this application. This correlation provides the surgeon with an intuitive force-position map or tensor table, clearly indicating at which specific angle of movement a mechanical abnormality occurs.

[0037] Step 120: Based on the joint space displacement data and corresponding joint contact force response data collected during the operation, construct the patient's soft tissue stiffness model.

[0038] Specifically, during surgery, in a predetermined, representative joint posture (e.g., 30 degrees of abduction, neutral rotation), the system guides the surgeon through a standardized testing procedure, such as a distraction test. In this test, the surgeon applies a gentle, continuously varying force in a specific direction (e.g., the humeral shaft axis), causing a slight separation of the articular surfaces. During this process, the surgical navigation system precisely measures and records the resulting joint space displacement data (i.e., the change in distance between the articular surfaces), while pressure sensors within the intelligent prosthesis simultaneously record the corresponding joint contact force response data (i.e., the change in contact force). Through this process, the system can obtain one or more sets of discrete, corresponding data points (displacement and force values).

[0039] After obtaining these measured data points, the system uses them to construct a soft tissue stiffness model for the patient. This construction refers to a process of model identification or parameter fitting. The system can pre-define a mathematical function form that describes the mechanical behavior of soft tissue, such as a nonlinear function. Then, a numerical optimization algorithm (such as the least squares method) is used to perform curve fitting on the collected (displacement, force) data points, thereby calculating the undetermined parameters in the mathematical function model. Once these parameters are determined, this mathematical function with patient-specific parameters constitutes the patient's soft tissue stiffness model. This model digitally expresses the comprehensive stiffness characteristics of the patient's periarticular soft tissues (including joint capsules, ligaments, muscles, etc.), i.e., "how much force is required to stretch a certain distance."

[0040] Step 130: Based on the soft tissue stiffness model, joint space data under different joint spatial postures, and joint space variation, determine the predicted joint contact force value of the candidate prosthesis specification combination in the full range of motion; the joint space data under different joint spatial postures is determined based on the correlation relationship; the joint space variation is caused by the change of geometric parameters in the candidate prosthesis specification combination.

[0041] Specifically, the system needs to determine candidate prosthesis size combinations. This refers to the set of implant components with different geometric parameters that the surgeon can choose from during surgery. These geometric parameters may include, for example, the thickness of the glenoid liner (e.g., +0mm, +3mm, +6mm, where mm is the unit of length), the offset of the humeral head or tray (e.g., 0mm, 1.5mm, 3.5mm), and the rotational placement angle of the offset (e.g., 0-360 degrees). Any combination of these parameters constitutes a candidate size combination.

[0042] For each candidate specification combination, the system needs to predict its impact on joint contact forces. This prediction process is as follows: (1) Determining the change in joint space: The system has a built-in geometric calculation module. When the user selects a candidate specification combination that is different from the current physical model on the software interface (for example, changing the liner thickness from +0mm to +3mm), the module can accurately calculate, based on the geometric model of the implant and the kinematics of the joint, how much the change in this geometric parameter will cause in each posture of the entire range of motion (ROM). This change in space directly caused by the change in geometric parameters is the change in joint space.

[0043] (2) Obtain joint space data under different joint spatial postures: This refers to the joint space value under different postures in the baseline state before changing the prosthesis specifications. This data can be measured synchronously by the surgical navigation system and stored in the association database along with the posture information during the first step of establishing the association.

[0044] (3) Predicted contact force: The system adds the joint space data of the baseline state to the calculated change in joint space to obtain the total joint space (or total soft tissue elongation) in each posture after adopting the new candidate specification. Then, this new total elongation is used as input and substituted into the constructed patient-specific soft tissue stiffness model to calculate the corresponding predicted joint contact force value.

[0045] By repeating the above calculations for all postures across the entire range of motion of the joint, the system can generate a complete predictive biomechanics curve covering the entire range of motion for each candidate prosthesis specification combination.

[0046] Step 140: Construct an optimization function based on the correlation and predicted joint contact force values, traverse all specification combinations of candidate prostheses, and take the specification combination corresponding to the optimal value of the optimization function as the recommended configuration scheme of the candidate prosthesis.

[0047] Specifically, the system first needs to construct an optimization function (also known as a cost function or objective function). This function is a mathematical expression whose purpose is to quantify the "goodness" or "badness" of a candidate solution into a single score. The function is constructed based on clinical expectations of ideal joint balance. For example, an ideal joint should have moderate and uniform tension throughout its range of motion, while remaining stable and free from dislocation or edge impingement. Therefore, the optimization function can comprehensively consider multiple objectives, such as: (1) Based on the predicted joint contact force: assess whether the predicted contact force curve falls within the ideal mechanical window set by the doctor (e.g., 10-25 lbs). If it exceeds this window, a penalty will be imposed.

[0048] (2) Based on correlation: For example, using the load center trajectory data in the established correlation database, evaluate whether the load center can be better maintained in the central region of the glenoid cavity under new mechanical conditions to ensure joint stability.

[0049] After constructing the optimization function, the system will iterate through all possible combinations of candidate prostheses. This means that the system will automatically and systematically predict the joint contact forces for each possible combination and calculate the corresponding optimization function value for each combination. This iteration can be achieved using global search algorithms such as grid search.

[0050] Finally, the system sorts the optimized function values ​​of all combinations and outputs one or more specification combinations with the best function values ​​as recommended configuration schemes to the doctor. The output results can be very intuitive, for example: "Recommended scheme A: Replace with +6mm liner and rotate the humeral tray offset backward by 30 degrees. Predicted effect: Average tension increases to 15lbs, and the load center returns to the central green zone." The joint soft tissue balancing method provided in this application fundamentally solves the problem of separating pressure values ​​and spatial posture information in related technologies by establishing the correlation between joint contact force and joint spatial posture, providing a clear and accurate targeting basis for subsequent prosthesis adjustment. By constructing a soft tissue stiffness model of the patient, the prediction is made more consistent with the patient's actual physiological condition, enabling the assessment and planning of soft tissue balancing to address individual differences in the patient's soft tissue. By virtually predicting the joint contact force of candidate prosthesis specifications across the entire range of motion, the problem of soft tissue edema and prolonged operation time caused by repeated mold changes during surgery is solved. By constructing an optimization function and iteratively searching for and recommending the optimal solution, the scientific nature, accuracy, and final joint function effect of surgical decisions are maximized. Overall, an intelligent closed-loop solution based on a patient-individualized model and driven by data, consisting of "measurement-modeling-prediction-optimization," is constructed, achieving accurate, efficient, and personalized joint soft tissue balancing, which helps improve the patient's postoperative function and long-term outcomes.

[0051] It should be noted that each implementation method of this application can be freely combined, rearranged, or executed individually, and does not need to rely on or depend on a fixed execution order.

[0052] In some embodiments, the intra-articular pressure data collected by the intelligent prosthesis trial model and the joint pose data collected by the surgical navigation system are integrated to establish the correlation between joint contact force and joint spatial posture, including: Time interpolation is performed on the joint pose data collected by the surgical navigation system to synchronize the joint pose data with the intra-articular pressure data in time. Map the position coordinates of the load center from the intelligent prosthesis trial model coordinate system to the glenoid coordinate system to obtain the motion trajectory of the load center in the glenoid coordinate system. Based on the time-synchronized joint pose data and intra-joint pressure data, as well as the motion trajectory of the load center, a four-dimensional tensor for characterizing the correlation is constructed on the discretized joint angle mesh. The four-dimensional tensor is used to store the average contact force, contact force standard deviation, shear force vector, and number of data sampling points within each angular interval.

[0053] Specifically, the pressure sensing device inside the intelligent prosthesis mold and the external surgical navigation system typically operate at different sampling frequencies. For example, the sampling frequency of the pressure sensing device... It could be 100Hz, while the sampling frequency of the surgical navigation system... It could be 60Hz. To associate each pressure data point with a precisely time-corresponding joint pose, this timestamp mismatch must be addressed.

[0054] Intelligent prosthesis trial model outputs mechanical vector .

[0055] in, for The amplitude of the joint contact force measured by the intelligent trial model at any time; Represent The three-dimensional coordinates of the load center in the intelligent test mold sensor coordinate system.

[0056] The surgical navigation system outputs a transformation matrix in the humeral prosthesis coordinate system. Relative to the glenoid coordinate system of Homogeneous transformation matrix It includes rotation and translation information.

[0057] The load center (CoL) coordinates measured by the pressure sensor are in its own sensor local coordinate system. However, from a clinical diagnostic perspective, doctors are more concerned with the position of the point of force application relative to the patient's anatomical structures (especially the glenoid fossa). Therefore, spatial coordinate mapping is necessary.

[0058] This mapping process requires the transformation matrix provided by the surgical navigation system and pre-calibrated equipment geometry. The specific transformation process can be represented by the following formula: .

[0059] in, for The load center at any moment is in the glenoid coordinate system. The three-dimensional spatial coordinates are used for subsequent eccentricity and stability assessment; for Humeral prosthesis coordinate system measured by the surgical navigation system at all times Relative to the glenoid coordinate system of The homogeneous transformation matrix contains rotation and translation information; Coordinate system for intelligent prosthesis trial model Relative to the humeral prosthesis coordinate system The transformation matrix is ​​invariant after the trial mold is installed in place; The load center is in the intelligent prosthesis test model coordinate system. The three-dimensional homogeneous coordinates in the model.

[0060] When the surgeon passively moves the patient's joints during surgery, the system continuously performs the above calculations, obtaining a series of... The points constitute the trajectory of the load center in the glenoid coordinate system. This trajectory is crucial for assessing the dynamic stability of the joint, such as whether the trajectory remains within the safe central area of ​​the glenoid.

[0061] To transform massive, continuous data streams into a structured database that is easy to query and analyze, embodiments of this application discretize the continuous motion space of a joint. Specifically, the joint's continuous motion space can be discretized by two main degrees of freedom (e.g., the abduction / adduction angles of the shoulder joint). and external / internal rotation angle The spherical space defined by ) is divided into small angular grid intervals (e.g., outward extension 30 degrees ± 2 degrees, outward rotation 10 degrees ± 2 degrees).

[0062] Then, the system iterates through all data points that have undergone spatiotemporal synchronization and places each data point into the corresponding angle grid based on the joint angle of each data point. For all data points falling within the same grid interval, the system calculates the statistical values ​​of their key mechanical parameters and stores these statistical values ​​in a multidimensional data structure corresponding to that grid interval.

[0063] In this embodiment of the application, the data structure is defined as a four-dimensional tensor. The four dimensions here refer to the storage of four key mechanical and quality control information dimensions at each two-dimensional angular coordinate point.

[0064] The average contact force represents the average value of all collected contact forces within a specific grid interval (e.g., 30 degrees ± 2 degrees outward extension, 10 degrees ± 2 degrees outward rotation).

[0065] The standard deviation of contact force represents the standard deviation of contact force within a grid interval, used to assess the mechanical consistency and data reliability when doctors test at this angle.

[0066] It is a shear force vector, obtained by decomposing the contact force in the direction parallel to the glenoid surface, and is used to assess the risk of prosthesis slippage.

[0067] This represents the total number of data sampling points that fall within the grid interval of this angle. If this value is less than a set threshold, the data at that angle is considered unreliable, and the system must prompt the doctor to re-adjust the angle.

[0068] The joint soft tissue balancing method provided in this application synchronously collects mechanical data from intelligent prosthesis trial models and kinematic data from surgical navigation systems. Through time interpolation and spatial coordinate transformation, it precisely correlates pressure values, pressure center positions, and specific joint postures (such as abduction and rotation angles). Subsequently, the continuous motion is discretized into an angular grid, constructing a four-dimensional tensor to systematically store key mechanical information such as average contact force and pressure center distribution within each posture range. This establishes a "force-position" correlation database (i.e., a four-dimensional tensor), creating a digital atlas that comprehensively, quantitatively, and intuitively reflects the complete mechanical properties of the joint under different postures.

[0069] In some embodiments, time interpolation is performed on the joint pose data acquired by the surgical navigation system, including: Quaternion spherical linear interpolation is used to interpolate the rotational quaternions in the joint pose data; Linear interpolation is used to interpolate the translation vectors in the joint pose data.

[0070] Specifically, the embodiments of this application employ a time interpolation method. Specifically, for any given time interpolation... The system continuously collects pressure data points from pressure sensors and identifies the two nearest adjacent pose data points on the timeline, both collected by the surgical navigation system (located at...). and At that moment, among them Then, the system calculates using an interpolation algorithm. Precise pose at any given moment.

[0071] Considering that joint pose data includes both rotation and translation, different interpolation functions can be used to ensure the accuracy and smoothness of interpolation: (1) For the rotating part, rotational quaternions are usually used to avoid gimbal lock-up. In this embodiment, the rotational quaternion is interpolated using the Slerp algorithm, which can be expressed by the formula: .

[0072] in, for The rotation quaternion obtained after time-interpolation; for Rotational quaternions at time; for Rotational quaternions at time; The time interpolation weighting coefficients can be expressed as: , .

[0073] This algorithm ensures that the angular velocity is uniform during the interpolation process and that the path is the shortest arc on the sphere, thus truly reflecting the smooth rotation of the joint.

[0074] (2) For the translation part, i.e. the three-dimensional translation vector, the linear interpolation (Lerp) algorithm can be used for processing, which can be expressed by the formula: .

[0075] in, for The three-dimensional translation vector obtained after time-interpolation; for The three-dimensional translation vector at time; for The three-dimensional translation vector at time step.

[0076] Through this time synchronization step, the system obtains one-to-one pressure-pose data pairs under a unified time reference.

[0077] The joint soft tissue balancing method provided in this application uses different, optimal interpolation algorithms for the rotation and translation components in the pose data, ensuring that the final time-synchronized pose data can approximate the actual motion trajectory of the joint within the sampling gap to the greatest extent possible, thereby improving the underlying data quality and accuracy of the "force-position" correlation database.

[0078] In some embodiments, a soft tissue stiffness model of the patient is constructed based on intraoperative joint space displacement data and corresponding joint contact force response data, including: Axial traction tests were performed on patients under specific joint postures to obtain joint space displacement data and joint contact force response data. Nonlinear fitting was performed on the patient's joint space displacement data and joint contact force response data to construct a soft tissue stiffness model.

[0079] Specifically, to ensure the repeatability and clinical relevance of the test, traction testing is typically performed in one or more pre-defined specific joint postures. For example, in shoulder surgery, a stable and functionally important posture with 30° abduction and neutral rotation can be selected. Choosing a specific posture isolates other variables, allowing the measured mechanical response to more purely reflect the stiffness of soft tissue under specific tension.

[0080] In this selected specific joint position, the surgeon, guided by the system, applies a gentle, continuous traction force along the axis of the humerus. This force causes a slight separation of the humeral head from the glenoid cavity, thereby stretching the soft tissues around the joint (such as the joint capsule, ligaments, etc.).

[0081] Throughout the traction process, different sensors within the system work together to collect two sets of key data in real time and synchronously: (1) Joint space displacement data Precise measurements are taken by the surgical navigation system. By tracking markers fixed to the humerus and scapula, the navigation system can calculate the change in distance between the articular surfaces, i.e., the elongation of the soft tissue, in real time with sub-millimeter precision.

[0082] (2) Joint contact force response data The pressure is measured synchronously by a pressure sensor within the intelligent prosthesis mold. As the joint gap is widened, the soft tissue tension gradually decreases, and the contact force decreases accordingly. The sensor records this dynamic change in force.

[0083] The data acquisition process is continuous. A single traction-release operation can generate a series (dozens or even hundreds) of corresponding data points (displacement values). Force value ), Indicates the sequence number of the data point.

[0084] After obtaining the measured data points, the next step is to transform them into a mathematical model with predictive capabilities.

[0085] The mechanical behavior of biological soft tissues exhibits significant nonlinear characteristics—that is, when stretched, their resistance to deformation (stiffness) increases exponentially, showing a "the more it is stretched, the stiffer it becomes" trend. Therefore, this application employs nonlinear fitting to more realistically capture this physical characteristic. The system pre-defines a function model capable of describing this nonlinear behavior. In a specific embodiment, an exponential function model can be used: .

[0086] in, The elongation of soft tissue is The tension generated during time; The basic tension coefficient of the tissue; The two parameters are nonlinear stiffening coefficients; they are undetermined model parameters that together define the shape of the stiffness curve specific to this patient.

[0087] The system employs a numerical optimization algorithm to fit a series of collected data points to solve for parameters A and B. In a specific embodiment, the Levenberg-Marquardt algorithm can be used. This algorithm is an efficient iterative algorithm commonly used to solve nonlinear least squares problems. It can find a set of optimal values ​​for A and B that minimizes the overall error (sum of squared residuals) between the theoretical force value calculated by the above exponential function model and all measured force values.

[0088] After fitting, the resulting function equation with definite parameters A and B constitutes the patient's soft tissue stiffness model.

[0089] The joint soft tissue balancing method provided in this application obtains mechanical response data reflecting the patient's true physiological condition through standardized intraoperative traction testing. Then, through nonlinear fitting, these discrete and noisy data points are transformed into an accurate and continuous mathematical model. A high-fidelity personalized stiffness model is created for each patient, which greatly improves the accuracy of the entire technical solution.

[0090] In some embodiments, based on a soft tissue stiffness model, joint space data under different joint spatial postures, and joint space variation, the predicted joint contact force values ​​for candidate prosthesis size combinations across the entire joint range of motion are determined, including: Determine the geometric parameters that need to be adjusted in the candidate prosthesis size combination; the geometric parameters include at least one of the following: liner thickness, eccentricity size, and eccentricity rotation angle; Based on the changes in geometric parameters, positive kinematics and geometric reconstruction calculations are performed to obtain the changes in joint space under different joint spatial postures. By substituting the joint space data and joint space variation under different joint spatial postures into the soft tissue stiffness model, the predicted joint contact force values ​​of candidate prosthesis specifications are obtained for the entire joint range of motion.

[0091] Specifically, before performing virtual predictions, the system needs to define adjustable variables, namely the geometric parameters in the candidate prosthesis specification combinations. These parameters represent the specifications of the implant components that surgeons can select and replace during actual surgery.

[0092] In one specific embodiment, these adjustable geometric parameters can be parameterized as a vector. .

[0093] The liner thickness refers to the thickness of the polyethylene liner on the glenoid side. This is a critical parameter that directly affects the joint's tightness. For example, available specifications can be increments relative to the current trial mold, such as +0mm, +3mm, and +6mm.

[0094] The offset refers to the distance the center of rotation of the humeral head or humeral tray is offset from its geometric center. By selecting components with different offset sizes (such as 0mm, 1.5mm, 3.5mm), the center of rotation of the joint can be changed, thereby adjusting the soft tissue tension in a specific direction.

[0095] The eccentricity rotation angle refers to the placement angle of the eccentric direction when using components with eccentricity. This angle is typically adjustable from 0 to 360 degrees and is used to finely control soft tissue tension during specific movement phases (such as flexion or extension).

[0096] The physician or system can virtually adjust at least one of the above parameters to create a combination of candidate prosthesis specifications. For example, changing only the liner thickness, or changing both the liner thickness and the eccentricity rotation angle simultaneously.

[0097] Once a candidate prosthesis specification combination is selected (i.e., geometric parameter vector), from Become The system needs to calculate the direct impact of this change on joint geometry. This is accomplished through the system's built-in computational engine, which includes a forward kinematic model of the joint and a geometric model of the implant.

[0098] Therefore, the change in joint space caused by changes in geometric parameters This can be obtained by subtracting the calculation results from the old and new configurations: .

[0099] in, This is a function for calculating forward kinematics and geometric reconstruction. The input to this function is a vector of geometric parameters representing the prosthesis configuration (prosthesis specification combination). The output is in a specific joint pose. Below, the amount of change in joint space caused by changes in prosthesis geometric parameters. .

[0100] This calculation applies to all postures across the entire range of joint motion. This process is performed to obtain a curve or surface that covers the entire range of gap variation.

[0101] At any joint posture Below, the total elongation of soft tissue after adopting the new prosthesis specifications. It is equal to its elongation under the reference condition. (i.e., joint space data under different joint spatial postures, which can be measured and stored when establishing correlations) and the amount of joint space change caused by changes in geometric parameters. The sum can be expressed by the formula: .

[0102] The system calculates the new total elongation and substitutes it into the patient's soft tissue stiffness model. From this, the stiffness at that posture can be calculated. Below are the predicted joint contact forces after adopting the new prosthesis specifications. This can be expressed as a formula: .

[0103] The system repeats this calculation for all discrete attitude points within the entire range of motion (ROM) of the joint, and finally generates a complete map of the predicted joint contact forces of the candidate prosthesis specification combination within the entire range of motion of the joint.

[0104] The joint soft tissue balancing method provided in this application embodiment is based on the patient's personalized stiffness model. The adjustable geometric parameters of the prosthesis are virtually modified in the software. The changes in joint space caused by the parameter changes are predicted through geometric calculations and mapped onto the stiffness model. The predicted contact force under the new configuration in the full range of motion of the joint is directly calculated, realizing "virtual trial molding". This replaces physical replacement and completely replaces the inefficient, high-risk and invasive physical trial and error process in traditional surgery. It not only greatly shortens the surgical decision time and reduces the surgical risk, but also enables doctors to explore far more prosthesis combination options than physical trial and error.

[0105] In some embodiments, an optimization function is constructed based on the correlation and predicted joint contact force values, including: An optimization function is constructed based on at least one of the full-radius tension consistency constraint function, joint center stability constraint function, and peak pressure penalty function. Among them, the full-radius tension consistency constraint function is used to constrain the predicted joint contact force value to remain within the ideal range throughout the entire joint range of motion; the joint center stability constraint function is used to constrain the load center trajectory obtained based on the correlation relationship to be close to the geometric center of the glenoid; and the peak pressure penalty function is used to penalize the predicted joint contact force value from reaching its maximum value.

[0106] Specifically, the optimization function in the embodiments of this application For, its input is a vector of geometric parameters of the candidate prosthesis specification combination. The output is a scalar score. This optimization function can be constructed by mathematically modeling at least one or more key clinical objectives designed to comprehensively evaluate the overall performance of a prosthesis configuration.

[0107] The optimization function can be composed of a weighted combination of the following three sub-functions: (1) Full-radian tension consistency constraint function .

[0108] The goal of this function is to ensure that the soft tissue tension of the joint is neither too tight nor too loose throughout its entire range of motion (ROM), but is always maintained within a clinically ideal range.

[0109] First, define the ideal range of joint contact force. This window can be preset by the system or customized by the surgeon according to the patient's condition. This represents the minimum joint contact force under ideal conditions. This represents the maximum value of the joint contact force under ideal conditions. For example, it can be set from 10 pounds to 25 pounds.

[0110] The function then evaluates the entire range of motion of the joint. Within the three-dimensional angular envelope space covered when establishing the association, the predicted value of the joint contact force is... The degree of deviation from this ideal range. This function can be constructed as an integral, imposing a quadratic penalty on all predictive power values ​​that deviate from the ideal range. The formula can be expressed as: .

[0111] The meaning of this formula is: when predictive power... When the error falls within this range, the error cost is 0; if it exceeds this range, a quadratic penalty is incurred.

[0112] (2) Joint center stability constraint function .

[0113] The goal of this function is to ensure good stability of the joint during dynamic activities, that is, the point of force application (load center) should be as close as possible to the geometric center of the glenoid to avoid "seesaw effect", edge loading or subdislocation.

[0114] This function utilizes the trajectory of the load center in the glenoid coordinate system, calculated and stored through spatial mapping during the establishment of the association. The goal is to minimize the Euclidean distance from the load center to the geometric center of the glenoid (which can be set as the origin of the coordinate system).

[0115] This function can be constructed as the integral of the square of the load center eccentricity over the entire range of motion of the joint. The formula can be expressed as: .

[0116] in, Indicates posture Below, a prosthesis is used. When, it is the square of the distance from the center of the load to the center of the glenoid fossa.

[0117] The smaller the integral value, the more concentrated the stress point is in the center of the prosthesis during the entire activity, and the lower the probability of the prosthesis experiencing a "seesaw effect" or edge wear.

[0118] (3) Peak pressure penalty function .

[0119] The goal of this function is to prevent localized, excessively high contact pressures in any joint orientation, as excessively high peak pressures are one of the main causes of premature wear and failure of the liner.

[0120] This function can be directly defined as the maximum value of the predicted contact force curve. The formula can be expressed as: .

[0121] During the optimization process, the system will tend to select... Choose a smaller value option to avoid configurations that, while having adequate tension most of the time, can produce harmful pressure spikes at certain extreme angles.

[0122] Ultimately, the optimization function J(x) can be constructed as a weighted sum of one or more of the above subfunctions: .

[0123] in, , , These are the weight coefficients of each sub-function, which can be adjusted according to the focus of the surgery. When optimizing the system, the goal is to find a prosthesis configuration that minimizes the value of the overall optimization function J(x).

[0124] The joint soft tissue balancing method provided in this application provides a clear, objective, and comprehensive evaluation standard for subsequent automated search by constructing an optimization function that integrates three core clinical objectives: tension consistency, joint stability, and peak pressure control.

[0125] In some embodiments, iterating through various combinations of candidate prostheses includes: A grid search is performed within the search space comprised of various combinations of candidate prostheses.

[0126] Specifically, in actual clinical applications, the specifications of implants (such as joint prostheses) are not continuously variable, but rather provided in a discrete form. For example, the thickness of the polyethylene liner may only have a limited number of options (e.g., +0mm, +3mm, +6mm), and the eccentricity is also fixed in a few increments (e.g., 0mm, 1.5mm, 3.5mm). Therefore, all these possible combinations of discrete parameters together constitute a discrete, multi-dimensional search space.

[0127] Considering the discrete nature of the search space, this embodiment preferably employs a grid search algorithm to perform the traversal. The specific implementation process is as follows: (1) Define the search grid: The system first defines the search space grid based on the available implant database. Each dimension of this grid corresponds to an adjustable geometric parameter (such as liner thickness, eccentricity size, eccentricity rotation angle), and the nodes in that dimension are all available discrete specification values ​​of that parameter.

[0128] (2) Perform exhaustive traversal: The grid search algorithm systematically and without omissions visits every node in the search space in an exhaustive manner. Specifically, the system will traverse all possible values ​​of the first parameter through multiple nested loops, and within each value, it will traverse all possible values ​​of the second parameter, and so on, until all possible combinations of specifications are generated.

[0129] (3) Evaluate each node: For each specification combination traversed (i.e. each node in the mesh), the system will perform a virtual prediction process to calculate the mechanical performance of the combination in the full range of motion of the joints; then, the prediction results will be substituted into the optimization function to calculate the comprehensive score of the combination.

[0130] (4) Determining the global optimum: The system records each specification combination and its corresponding optimization function score. After traversing all nodes, the system compares all recorded scores and finds the specification combination corresponding to the node that makes the optimization function value optimal (e.g., minimum or maximum). This combination is the global optimum among all available options and is output as the final recommended solution.

[0131] The joint soft tissue balancing method provided in this application uses grid search to ensure that the recommended solution found is the theoretically global optimal solution within all available specifications, thereby greatly improving the reliability and scientific nature of the decision-making.

[0132] The apparatus provided in the embodiments of this application is described below. The apparatus described below can be referred to in correspondence with the method described above.

[0133] Figure 2This is a schematic diagram of the joint soft tissue balancing device provided in this application, as shown below. Figure 2 As shown, the device includes: The measurement module 210 is used to integrate the intra-articular pressure data collected by the intelligent prosthesis trial model and the joint posture data collected by the surgical navigation system to establish the correlation between joint contact force and joint spatial posture. Modeling module 220 is used to construct a soft tissue stiffness model of the patient based on the joint space displacement data and corresponding joint contact force response data collected during the operation. The prediction module 230 is used to determine the predicted joint contact force values ​​of candidate prosthesis specifications in the full range of motion of the joint based on the soft tissue stiffness model, joint space data under different joint spatial postures, and the change in joint space; the joint space data under different joint spatial postures is determined based on the correlation relationship; the change in joint space is caused by the change in geometric parameters in the candidate prosthesis specifications. The optimization module 240 is used to construct an optimization function based on the correlation and predicted joint contact force values, traverse the various specification combinations of candidate prostheses, and take the specification combination corresponding to the optimal value of the optimization function as the recommended configuration scheme of the candidate prosthesis.

[0134] The joint soft tissue balancing device provided in this application fundamentally solves the problem of separating pressure values ​​and spatial posture information in related technologies by establishing a correlation between joint contact force and joint spatial posture, providing a clear and accurate targeting basis for subsequent prosthesis adjustment. By constructing a soft tissue stiffness model of the patient, the prediction is made more consistent with the patient's actual physiological condition, enabling the assessment and planning of soft tissue balancing to address individual differences in the patient's soft tissue. By virtually predicting the joint contact force of candidate prosthesis specifications across the entire range of motion, the problem of soft tissue edema and prolonged operation time caused by repeated mold changes during surgery is solved. By constructing an optimization function and iteratively searching to recommend the optimal solution, the scientific nature, accuracy, and final joint function effect of surgical decisions are maximized. Overall, an intelligent closed-loop solution based on a patient-individualized model and driven by data, consisting of "measurement-modeling-prediction-optimization," is constructed, achieving accurate, efficient, and personalized joint soft tissue balancing, which helps improve the patient's postoperative function and long-term outcomes.

[0135] Figure 3 This is a schematic diagram of the joint soft tissue balance system provided in this application, as shown below. Figure 3 As shown, the system includes an input layer 310, a model layer 320, a prediction layer 330, and an output layer 340.

[0136] The input layer is used to acquire intraoperative joint space displacement data and corresponding joint contact force response data collected from the patient.

[0137] The model layer is used to perform nonlinear fitting on the joint space displacement data and the corresponding joint contact force response data to obtain the patient's soft tissue stiffness model.

[0138] The prediction layer is used to determine the predicted joint contact force values ​​of candidate prosthesis specification combinations within the full range of motion of the joint, based on the soft tissue stiffness model, joint space data under different joint spatial postures, and joint space variation. An optimization function is constructed based on the correlation and the predicted joint contact force values. The optimization function iterates through each specification combination of candidate prostheses, and the specification combination corresponding to the optimal value of the optimization function is used as the recommended configuration scheme of the candidate prosthesis.

[0139] Among them, the joint space data under different joint spatial postures are determined based on the correlation; the change in joint space is caused by the change in geometric parameters in the candidate prosthesis specification combination; the correlation is established based on the intra-articular pressure data collected by the intelligent prosthesis trial model and the joint posture data collected by the surgical navigation system.

[0140] The output layer is used to output recommended configuration schemes.

[0141] The joint soft tissue balance system provided in this application embodiment achieves precise and efficient soft tissue balance adjustment, which helps to achieve a more ideal joint biomechanical environment, thereby improving postoperative joint stability, reducing the risk of complications such as dislocation and abnormal wear, and improving the postoperative functional recovery effect of patients.

[0142] Figure 4 This is a schematic diagram of the structure of the electronic device provided in this application, such as... Figure 4 As shown, the electronic device may include: a processor 410, a communications interface 420, a memory 430, and a communications bus 440, wherein the processor, communications interface, and memory communicate with each other via the communications bus. The processor can invoke logical commands stored in the memory to execute the methods described in the above embodiments, for example: By integrating intra-articular pressure data collected from intelligent prosthesis trial models and joint pose data collected from the surgical navigation system, a correlation between joint contact force and joint spatial posture is established. Based on intraoperative joint space displacement data and corresponding joint contact force response data collected from the patient, a soft tissue stiffness model of the patient is constructed. Based on the soft tissue stiffness model, joint space data under different joint spatial postures, and joint space variation, the predicted joint contact force values ​​of candidate prosthesis specifications are determined for the entire range of motion. The joint space data under different joint spatial postures are determined based on the correlation. The joint space variation is caused by changes in geometric parameters in the candidate prosthesis specifications. Based on the correlation and predicted joint contact force values, an optimization function is constructed, and iterates through all specifications of the candidate prostheses. The specification combination corresponding to the optimal value of the optimization function is used as the recommended configuration scheme of the candidate prosthesis.

[0143] Furthermore, the logical commands in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or part 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 commands 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 application. 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.

[0144] The processor in the electronic device provided in this application embodiment can call logical instructions in the memory to implement the above method. Its specific implementation method is the same as the aforementioned method implementation method and can achieve the same beneficial effect, which will not be repeated here.

[0145] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the methods provided in the above embodiments.

[0146] The specific implementation method is the same as the aforementioned method implementation method and can achieve the same beneficial effects, so it will not be repeated here.

[0147] This application provides a computer program product, including a computer program that, when executed by a processor, implements the method described above.

[0148] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0149] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0150] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications 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 this application.

Claims

1. A method for balancing soft tissues in a joint, characterized in that, include: By integrating intra-articular pressure data collected from intelligent prosthesis trial models and joint pose data collected from surgical navigation systems, a correlation between joint contact force and joint spatial posture is established. Based on the joint space displacement data and corresponding joint contact force response data collected during the patient's operation, a soft tissue stiffness model of the patient is constructed. Based on the soft tissue stiffness model, joint space data under different joint spatial postures, and joint space variation, the predicted joint contact force values ​​of candidate prosthesis specifications are determined within the full range of motion of the joint; the joint space data under different joint spatial postures are determined based on the correlation. The change in joint space is caused by the change in geometric parameters in the candidate prosthesis specification combination; An optimization function is constructed based on the correlation and the predicted joint contact force. The various specification combinations of the candidate prostheses are traversed, and the specification combination corresponding to the optimal value of the optimization function is taken as the recommended configuration scheme of the candidate prosthesis.

2. The joint soft tissue balancing method according to claim 1, characterized in that, The method integrates intra-articular pressure data collected from the intelligent prosthesis trial model and joint pose data collected from the surgical navigation system to establish a correlation between joint contact force and joint spatial posture, including: The joint pose data collected by the surgical navigation system is time-interpolated to synchronize the joint pose data with the intra-articular pressure data. The position coordinates of the load center are mapped from the intelligent prosthesis trial model coordinate system to the glenoid coordinate system to obtain the motion trajectory of the load center in the glenoid coordinate system. Based on the time-synchronized joint pose data, the intra-joint pressure data, and the motion trajectory of the load center, a four-dimensional tensor for characterizing the correlation is constructed on the discretized joint angle grid. The four-dimensional tensor is used to store the average contact force, contact force standard deviation, shear force vector, and number of data sampling points within each angular interval.

3. The joint soft tissue balancing method according to claim 2, characterized in that, The step of interpolating the joint pose data collected by the surgical navigation system over time includes: The rotational quaternions in the joint pose data are interpolated using quaternion spherical linear interpolation; Linear interpolation is used to interpolate the three-dimensional translation vector in the joint pose data.

4. The joint soft tissue balancing method according to claim 1, characterized in that, The soft tissue stiffness model of the patient is constructed based on the joint space displacement data and corresponding joint contact force response data collected during the operation, including: Axial traction test was performed on the patient under a specific joint posture to obtain the patient's joint space displacement data and joint contact force response data; The patient's joint space displacement data and joint contact force response data were nonlinearly fitted to construct a soft tissue stiffness model.

5. The joint soft tissue balancing method according to claim 1, characterized in that, The step of determining the predicted joint contact force values ​​of candidate prosthesis combinations across the entire range of motion, based on the soft tissue stiffness model, joint space data under different joint spatial postures, and joint space variation, includes: Determine the geometric parameters that need to be adjusted in the candidate prosthesis specification combination; the geometric parameters include at least one of the following: liner thickness, eccentricity size, and eccentricity rotation angle; Based on the changes in the geometric parameters, positive kinematics and geometric reconstruction calculations are performed to obtain the changes in joint space under different joint spatial postures. By substituting the joint space data and joint space variation under different joint spatial postures into the soft tissue stiffness model, the predicted joint contact force values ​​of the candidate prosthesis specification combination are obtained within the full range of joint motion.

6. The method for balancing soft tissues of a joint according to claim 1, characterized in that, The construction of the optimization function based on the correlation and the predicted joint contact force includes: The optimization function is constructed based on at least one of the full-radius tension consistency constraint function, the joint center stability constraint function, and the peak pressure penalty function. The full-radius tension consistency constraint function is used to constrain the predicted joint contact force value to remain within the ideal range throughout the entire joint range of motion; the joint center stability constraint function is used to constrain the load center trajectory obtained based on the correlation relationship to be close to the geometric center of the glenoid cavity; and the peak pressure penalty function is used to penalize the predicted joint contact force value from reaching its maximum value.

7. The method for balancing soft tissues of a joint according to claim 1, characterized in that, The process of traversing the various size combinations of the candidate prostheses includes: A grid search is performed within the search space comprised of the various size combinations of the candidate prostheses.

8. A joint soft tissue balancing device, characterized in that, include: The measurement module is used to integrate intra-articular pressure data collected by the intelligent prosthesis trial model and joint pose data collected by the surgical navigation system to establish the correlation between joint contact force and joint spatial posture. The modeling module is used to construct a soft tissue stiffness model of the patient based on the joint space displacement data and corresponding joint contact force response data collected during the operation. The prediction module is used to determine the predicted joint contact force values ​​of candidate prosthesis specifications within the full range of motion of the joint, based on the soft tissue stiffness model, joint space data under different joint spatial postures, and joint space variation; the joint space data under different joint spatial postures are determined based on the correlation relationship. The change in joint space is caused by the change in geometric parameters in the candidate prosthesis specification combination; An optimization module is used to construct an optimization function based on the correlation and the predicted joint contact force value, traverse all specification combinations of the candidate prostheses, and take the specification combination corresponding to the optimal value of the optimization function as the recommended configuration scheme of the candidate prosthesis.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the joint soft tissue balancing method according to any one of claims 1 to 7.

10. A non-transitory 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 joint soft tissue balancing method according to any one of claims 1 to 7.

Images