An inertial measurement-based total knee arthroplasty assisted navigation method and system

Abstract

The present application relates to a kind of based on inertial measurement total knee arthroplasty auxiliary navigation method and system, solve the existing TKA technical scheme There is high learning cost, operation is complex, it is easily interfered by soft tissue, and sensor installation relies on artificial experience Problem, by setting inertial sensor in target femur, in stationary state Static data of acceleration and angular velocity are collected to calibrate sensor coordinate system, so that it is less than calibration threshold with world coordinate system Deviation;Subsequently, with hip joint as the ball center Functional movement is carried out and dynamic data of acceleration, angular velocity and attitude quaternion are collected in real time;Based on rigid body kinematics model, observation equation is constructed, and hip joint rotation center position vector and acceleration bias term are determined jointly, and regression algorithm is used to solve its theoretical position in combination with physiological range constraint;Finally, according to the theoretical position and knee center, femoral mechanical axis is determined, and the deviation of varus angle and flexion angle of osteotomy guide plate is calculated in real time, to guide operator to adjust to target range.

Description

Technical Field

[0001] This invention relates to the technical field of femoral positioning using acceleration, and more particularly to an inertial measurement-based auxiliary navigation method and system for total knee arthroplasty. Background Technology

[0002] Total knee arthroplasty (TKA) is an effective treatment for end-stage knee joint disease. In TKA, the prosthesis's alignment and osteotomy angle directly affect postoperative lower limb alignment, prosthesis wear, and long-term survival. Traditional femoral positioning methods primarily rely on mechanical guides such as intramedullary or extramedullary rods. These methods are highly invasive, increase the risk of bleeding and infection, and are heavily dependent on the surgeon's experience.

[0003] To improve osteotomy accuracy, clinical solutions such as optical navigation, electromagnetic navigation, and robot-assisted systems have emerged. However, while these systems offer high accuracy, they suffer from drawbacks including high equipment cost, large space requirements, the need for unobstructed lines of sight, and high learning costs. To address this, existing technologies also employ accelerometer / inertial navigation-based assisted positioning solutions. For example, US Patent US9706948B2 uses a six-degree-of-freedom inertial chip, processing measurement data through integration, quaternion, and Kalman filtering algorithms to track the position and orientation of bones and instruments. Chinese Patent CN112272537B mounts a specific anatomical mapping instrument to a specific site with a single known location and orientation, and uses a miniature IMU (Inertial Measurement Unit) and instrument IMU for navigation.

[0004] However, the existing technology still has the following shortcomings: (1) Existing accelerometer / inertial navigation-based solutions require multi-sensor collaboration, complex calibration procedures, or point registration steps, which are cumbersome and have a long learning curve. For example, the CN112272537B solution requires the use of at least two IMUs and a Patient-Specific Anatomical Mapper (PAM). (2) The estimation of the hip joint rotation center and force line is easily affected by soft tissue artifacts (STA) such as muscle swaying, loosening of binding, and non-rigid body motion, which leads to outlier interference and navigation drift. (3) If the sensor installation direction changes, the axis needs to be manually specified or additional calibration is required, which can easily introduce human error. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide an inertial measurement-based assisted navigation method and system for total knee arthroplasty, which at least solves the technical problems of existing navigation schemes for total knee arthroplasty, such as high learning costs, complex operation, susceptibility to soft tissue artifacts, and reliance on human experience for sensor installation direction.

[0006] In a first aspect, embodiments of the present invention provide an inertial measurement-based assisted navigation method for total knee arthroplasty, comprising:

[0007] An inertial sensor is installed on the target femur, and when the target femur is in a stationary state, the static acceleration data and static angular velocity data output by the inertial sensor are collected.

[0008] The sensor coordinate system of the inertial sensor is calibrated based on the static acceleration data and the static angular velocity data, so that the deviation between the sensor coordinate system and the world coordinate system is less than the calibration threshold.

[0009] Functional motion is performed with the hip joint of the target femur as the center of the ball, and the acceleration dynamic data, angular velocity dynamic data and attitude quaternion data output by the inertial sensor are collected in real time during the functional motion.

[0010] An observation equation is constructed based on a rigid body kinematics model. The observation equation is used to process the acceleration dynamic data, the angular velocity dynamic data, and the attitude quaternion data to determine the position vector and acceleration bias term of the hip joint rotation center in the sensor coordinate system. The theoretical position of the hip joint rotation center is then solved using a regression algorithm and physiological range constraints.

[0011] Based on the theoretical position and the position of the knee joint center, the femoral mechanical axis is determined, and the varus / valgus angle and flexion / extension angle of the osteotomy guide plate relative to the femoral mechanical axis are calculated in real time and the corresponding angle deviation is generated to guide the operator to adjust the osteotomy guide plate to the target angle range.

[0012] Because only a single inertial sensor needs to be placed on the target femur, eliminating the need for multi-sensor collaboration, optical / electromagnetic tracking equipment, or anatomical mapping instruments, equipment costs and space requirements are significantly reduced. Static calibration automatically aligns the sensor coordinate system with the world coordinate system, eliminating tedious steps such as manually specifying axes or registering points, thus shortening the learning curve. Furthermore, the observation equations constructed based on a rigid body kinematics model, by jointly solving the hip joint rotation center position vector and acceleration bias term, fundamentally reduce error accumulation, providing a reliable basis for subsequent osteotomy guide angle feedback.

[0013] As an optional implementation, the construction of the observation equations based on the rigid body kinematics model includes:

[0014] During the functional movement, the target femur and its inertial sensor are defined as rigid bodies undergoing spherical motion around the rotation center of the hip joint;

[0015] Based on the principles of rigid body kinematics, an expression for the theoretical linear acceleration of the inertial sensor in the world coordinate system is constructed. The theoretical linear acceleration is equal to the vector sum of the centripetal acceleration and the tangential acceleration. After the attitude quaternion data is transformed from the sensor coordinate system to the world coordinate system, the gravitational acceleration vector in the world coordinate system and the acceleration bias term are superimposed.

[0016] The vector sum of the centripetal acceleration and the tangential acceleration is determined by the dynamic angular velocity data, the angular acceleration data, and the unknown position vector pointing from the inertial sensor to the center of rotation of the hip joint; the acceleration bias term is determined by the error of the inertial sensor itself, and the angular acceleration data is obtained by differentiating the dynamic angular velocity data.

[0017] Based on the fact that the theoretical linear acceleration is equal to the linear acceleration actually measured by the inertial sensor after being converted to the world coordinate system, the observation equation is constructed with the unknown position vector and the acceleration bias term as variables to be solved.

[0018] The embodiments of the present invention explicitly unify centripetal acceleration, tangential acceleration, gravitational acceleration, and acceleration bias terms under a rigid body kinematics framework, which to a certain extent ensures the physical rationality of the calculation of the hip joint rotation center.

[0019] As an optional implementation, the step of processing the acceleration dynamic data, the angular velocity dynamic data, and the attitude quaternion data using the observation equation to determine the position vector and acceleration bias term of the hip joint rotation center in the sensor coordinate system includes:

[0020] The unknown position vector is constructed as a three-dimensional position vector in the sensor coordinate system, and the three-dimensional position vector and the three components of the acceleration bias term in the world coordinate system are jointly determined as a six-dimensional parameter vector; wherein, the three-dimensional position vector represents the relative positional relationship from the inertial sensor to the hip joint rotation center;

[0021] The six-dimensional parameter vector is used as the variable to be solved in the observation equation for joint solution, so as to reduce the cumulative error introduced by solving the position vector and the acceleration bias term separately.

[0022] This invention employs a six-dimensional parameter vector to jointly determine the variables to be solved in the observation equation, simultaneously solving for the position vector and acceleration bias term. Compared to solving them separately, joint prediction effectively eliminates the crosstalk effects of inertial sensor errors such as temperature drift and zero bias on the estimation of the hip joint rotation center position, thus significantly improving the accuracy and stability of hip joint rotation center positioning.

[0023] As an optional implementation, the step of using a regression algorithm and physiological range constraints to solve for the theoretical position of the hip joint rotation center includes:

[0024] The functional motion process includes multiple sampling moments. For each sampling moment, the theoretically predicted acceleration is calculated based on the estimated value of the six-dimensional parameter vector in the current iteration round, and the acceleration residual between the theoretically predicted acceleration and the measured linear acceleration is calculated. The measured linear acceleration is the linear acceleration actually measured by the inertial sensor after being converted to the world coordinate system.

[0025] The weight coefficient of the sampling moment in the iteration is dynamically allocated according to the absolute value of the acceleration residual, including: when the absolute value is less than or equal to a preset weight threshold, a first weight is assigned to the sampling moment; when the absolute value is greater than the preset weight threshold, a second weight is assigned to the sampling moment; wherein, the first weight is the full weight, and the second weight is determined by dividing the preset weight threshold by the absolute value of the acceleration residual;

[0026] By dynamically allocating the weight coefficients of each sampling time in the current iteration, a diagonal weight matrix is ​​constructed. In the current iteration, the estimated value of the six-dimensional parameter vector is updated based on the weight matrix using the weighted least squares method, and the process proceeds to the next iteration until the preset convergence condition is met.

[0027] This invention employs a dynamically weighted iterative least squares method based on acceleration residuals to automatically assign lower weights to outlier sampling points caused by soft tissue artifacts such as muscle swaying and loosening of bindings. This further enhances robustness against non-ideal motion disturbances during total knee arthroplasty.

[0028] As an optional implementation, the step of using a regression algorithm and physiological range constraints to solve for the theoretical position of the hip joint rotation center further includes:

[0029] In the objective function of the iterative optimization solution, a regularization constraint term based on Bayesian prior information is added for the component of the three-dimensional position vector in the Z-axis direction of the sensor coordinate system.

[0030] The regularization constraint term causes the solution to converge toward the prior mean, thus compensating for the insufficient observability of the Z-axis direction in the functional motion due to the motion trajectory being concentrated on the horizontal plane.

[0031] This invention addresses the issue of weak observability of the Z-axis direction in functional circular motion by introducing a soft constraint on the Z-axis, namely Bayesian prior regularization. This avoids the divergence or severe deviation from physiological reality of the Z-axis component caused by relying solely on motion data, ensuring the spatial stability and physiological consistency of the three-dimensional hip joint rotation center calculation results.

[0032] As an optional implementation, the physiological range constraint specifically includes:

[0033] After obtaining the position vector of the hip joint rotation center in the sensor coordinate system, the magnitude of the position vector is calculated.

[0034] The model length is compared with a preset physiological range, which is determined based on the statistical characteristics of adult femur anatomy, the height of the subject, preoperative imaging measurement data and / or statistical morphological models.

[0035] If the modulus exceeds the physiological range, the current solution result is determined to be abnormal.

[0036] This invention incorporates physiological range constraints to verify the anatomical rationality of the solution results. When severe binding slippage or irregular movement leads to abnormal solution results, the system can automatically identify and determine these anomalies, preventing erroneous navigation parameters from being used for subsequent osteotomy guidance and providing crucial safety redundancy.

[0037] As an optional implementation, the process of the functional motion also includes real-time monitoring of the data quality of the inertial sensor;

[0038] The real-time monitoring of the data quality of the inertial sensor includes: real-time monitoring of the signal connection strength indication value between the inertial sensor and the processing device; and real-time analysis of the waveform characteristics of the acceleration dynamic data and the angular velocity dynamic data to identify whether there are sudden acceleration changes or abnormal angular velocity drifts caused by loose binding.

[0039] When any one or more of the following conditions are detected, the operator is guided to check the stationary state of the inertial sensor or re-execute the functional motion:

[0040] The signal connection strength indicator value is lower than a preset strength threshold;

[0041] The acceleration abrupt change exceeds a preset acceleration jump threshold;

[0042] The root mean square of the motion amplitude of the dynamic angular velocity data is lower than a preset motion amplitude threshold.

[0043] This invention establishes a closed-loop monitoring system for data quality throughout the entire process. It monitors signal connection strength and acceleration / angular velocity waveform anomalies in real time, and proactively prompts and guides the operator to adjust their operation when anomalies are detected. This improves the usability and safety of femoral navigation and effectively solves the problem of data anomalies caused by loose sensor bindings, irregular movements, etc., and the lack of automatic detection.

[0044] As an optional implementation, guiding the operator to adjust the osteotomy guide plate to the target angle range includes:

[0045] A preset color coding scheme is used to hierarchically and visually encode the angular deviations, displaying the current inversion / exversion angles and flexion / extension angles to the operator in real time in a visual format; the color coding scheme includes:

[0046] When the absolute value of the deviation between the inward or outward angle and the target angle is less than or equal to the first deviation threshold, it is displayed as the first color area;

[0047] When the absolute value of the deviation is greater than the first deviation threshold and less than or equal to the second deviation threshold, it is displayed as the second color area;

[0048] When the absolute value of the deviation is greater than the second deviation threshold and less than or equal to the third deviation threshold, it is displayed in the third color area;

[0049] When the absolute value of the deviation is greater than the third deviation threshold, it is displayed in the fourth color area;

[0050] The colors of the first color region, the second color region, the third color region, and the fourth color region are all different, and the first color region corresponds to the osteotomy guide plate being adjusted to the target angle range.

[0051] This invention reduces the cognitive burden on operators by setting a multi-level color coding scheme, that is, by mapping continuous angular deviations to intuitive color areas (such as green / blue / orange / red), making the adjustment of osteotomy guide plates faster and more intuitive, and effectively improving execution efficiency and operating experience.

[0052] Secondly, embodiments of the present invention provide an inertial measurement-based auxiliary navigation system for total knee arthroplasty, comprising:

[0053] An inertial sensor is mounted on the target femur. The inertial sensor is used to collect static acceleration data and static angular velocity data when the target femur is at rest, and to collect dynamic acceleration data and dynamic angular velocity data in real time during the functional motion. The inertial sensor includes a data output module.

[0054] A positioning and navigation terminal includes a processor, a memory, a data receiving module, and a human-computer interaction device; wherein the data receiving module is communicatively connected to the data output module, and the data receiving module is used to receive the static acceleration data, the static angular velocity data, the dynamic acceleration data, and the dynamic angular velocity data; the processor executes any one of the embodiments in the first aspect described above; the human-computer interaction device is used to guide the operator to adjust the osteotomy guide plate to the target angle range.

[0055] The inertial measurement-based auxiliary navigation system for total knee arthroplasty provided in this invention has the characteristics of high integration and strong portability. It can reduce the requirements of complex optical positioning equipment or surgical robot systems in terms of cost, installation conditions, space occupation and learning cost, and reduce the impact of experience differences caused by traditional operation methods that rely on surgeon experience.

[0056] Thirdly, embodiments of the present invention provide a readable storage medium having a program stored thereon, which, when executed by a processor, implements any of the embodiments of the first aspect described above.

[0057] The embodiments of the present invention solidify the above-mentioned inertial measurement-based assisted navigation method for total knee arthroplasty in the form of a software program, so as to facilitate deployment and updates on general-purpose processing devices, reduce hardware customization costs, and improve the flexibility of the technical solution in promotion. Attached Figure Description

[0058] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments of the present invention will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, and these are all within the protection scope of the present invention.

[0059] Figure 1 This is a flowchart illustrating the inertial measurement-based assisted navigation method for total knee arthroplasty in an embodiment of the present invention.

[0060] Figure 2 This is a geometric schematic diagram of the hip joint rotation center in an embodiment of the present invention;

[0061] Figure 3 This is a schematic diagram of the structure of the inertial measurement-based total knee arthroplasty auxiliary navigation system in one embodiment of the present invention. Detailed Implementation

[0062] The features and exemplary embodiments of various aspects of the present invention will now be described in detail. To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only configured to explain the present invention and are not configured to limit the present invention. For those skilled in the art, the present invention can be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present invention by illustrating examples of the invention.

[0063] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0064] In a first aspect, embodiments of the present invention provide an inertial measurement-based assisted navigation method for total knee arthroplasty, as described in the following example. Figure 1 As shown, the method may include the following steps S101 to S105:

[0065] Step S101: Install an inertial sensor on the target femur, and collect static acceleration data and static angular velocity data output by the inertial sensor when the target femur is stationary.

[0066] The inertial sensor may include a triaxial accelerometer and a triaxial gyroscope, and can be fixed to the stable fixation area corresponding to the target femur by a special or general fixation device; the fixation device includes, but is not limited to, medical double-sided tape, special bandage, clamping fixation or adhesive fixation.

[0067] Step S102: Based on the static acceleration data and static angular velocity data, calibrate the sensor coordinate system of the inertial sensor so that the deviation between the sensor coordinate system and the world coordinate system is less than the calibration threshold.

[0068] When the target femur is at rest, the resultant acceleration vector measured by the inertial sensor theoretically only contains gravitational acceleration. Therefore, the direction of the resultant acceleration vector measured by the inertial sensor at this time is the direction of gravity in the world coordinate system. Based on this gravity vector, the relative rotation relationship between the sensor coordinate system and the world coordinate system can be automatically determined. When the deviation between the sensor coordinate system and the world coordinate system is less than the calibration threshold, the calibration of the sensor coordinate system can be considered complete. The calibration threshold can be preset according to actual application requirements and sensor performance.

[0069] Step S103: Perform functional motion with the hip joint of the target femur as the center of the ball, and collect dynamic acceleration data, dynamic angular velocity data and attitude quaternion data output by the inertial sensor in real time during the functional motion.

[0070] The inertial sensor integrates a data output module that can acquire static acceleration data, static angular velocity data, dynamic acceleration data, dynamic angular velocity data, and attitude quaternion data at a preset sampling frequency, and send this data to a processing device that has established a pre-existing communication connection with the inertial sensor. The processing device can be a smartphone, tablet, or computer, and contains a processor, memory, and a data receiving module. The processing device and the inertial sensor can be connected via wired or wireless communication, including methods such as Bluetooth and WiFi.

[0071] As an optional implementation, the data quality of the inertial sensor can also be monitored in real time during functional motion. Specifically, the signal connection strength indicator value between the inertial sensor and the processing device is monitored in real time; the waveform characteristics of dynamic acceleration and dynamic angular velocity data are analyzed in real time to identify whether there are sudden acceleration changes or abnormal angular velocity drifts caused by loose binding; when any one or more of the following conditions are detected, the operator is guided to check the fixed status of the inertial sensor or re-execute the functional motion:

[0072] The signal strength indicator value is lower than the preset strength threshold;

[0073] The acceleration abruptly exceeds the preset acceleration jump threshold;

[0074] The root mean square of the motion amplitude of the dynamic angular velocity data is lower than the preset motion amplitude threshold.

[0075] By monitoring signal connection strength and acceleration / angular velocity waveform anomalies in real time, and proactively prompting operators to adjust their operations when anomalies are detected, the usability and safety of femoral navigation are improved. This effectively solves the problem of data anomalies and lack of automatic detection caused by loose sensor bindings, irregular movements, and other issues.

[0076] Step S104: Construct observation equations based on rigid body kinematics model, process dynamic acceleration data, dynamic angular velocity data and attitude quaternion data using observation equations, determine the position vector and acceleration bias term of the hip joint rotation center in the sensor coordinate system, and solve the theoretical position of the hip joint rotation center using regression algorithm and physiological range constraints.

[0077] Specifically, the construction of observation equations based on rigid body kinematics models can include the following steps A1 to A3:

[0078] A1: During functional movement, the target femur and its inertial sensors are defined as rigid bodies undergoing spherical motion around the center of rotation of the hip joint;

[0079] A2: Based on the principles of rigid body kinematics, construct an expression for the theoretical linear acceleration of the inertial sensor in the world coordinate system;

[0080] The theoretical linear acceleration is equal to the vector sum of the centripetal and tangential accelerations, which is transformed from the sensor coordinate system to the world coordinate system using attitude quaternion data, and then superimposed with the gravitational acceleration vector and acceleration bias term in the world coordinate system. The vector sum of the centripetal and tangential accelerations is determined by the dynamic angular velocity data, the angular acceleration data, and the unknown position vector pointing from the inertial sensor to the center of hip joint rotation; the acceleration bias term is determined by the inertial sensor's own error, and the angular acceleration data is obtained by differentiating the dynamic angular velocity data.

[0081] A3: Based on the theoretical linear acceleration being equal to the linear acceleration actually measured by the inertial sensor after conversion to the world coordinate system, an observation equation is constructed with the unknown position vector and acceleration bias term as variables to be solved.

[0082] By unifying centripetal acceleration, tangential acceleration, gravitational acceleration, and acceleration bias terms within a rigid body kinematics framework, the physical rationality of the hip joint rotation center calculation is ensured to a certain extent.

[0083] Specifically, determining the position vector and acceleration offset term of the hip joint rotation center in the sensor coordinate system can include the following steps B1 to B2:

[0084] B1: Construct the unknown position vector into a three-dimensional position vector in the sensor coordinate system, and determine the three-dimensional position vector and the three components of the acceleration bias term in the world coordinate system as a six-dimensional parameter vector.

[0085] The three-dimensional position vector represents the relative positional relationship from the inertial sensor to the center of rotation of the hip joint.

[0086] B2: The six-dimensional parameter vector is used as the variable to be solved in the observation equation for joint solution, so as to reduce the cumulative error introduced by solving the position vector and acceleration bias term separately.

[0087] The six-dimensional parameter vector joint observation equation is used to solve for the variables to be solved, and the position vector and acceleration bias term are solved simultaneously. Compared with solving them separately, the joint prediction can effectively eliminate the crosstalk effect of errors such as temperature drift and zero bias of the inertial sensor on the estimation of the hip joint rotation center position, and effectively improve the accuracy and stability of the hip joint rotation center positioning.

[0088] In the solution process, a regression algorithm and physiological range constraints are used to solve for the theoretical position of the hip joint rotation center, which may include the following steps C1 to C3:

[0089] C1: During the functional motion process, there are multiple sampling moments. For each sampling moment, the theoretically predicted acceleration is calculated based on the estimated value of the six-dimensional parameter vector of the current iteration round, and the acceleration residual between the theoretically predicted acceleration and the measured linear acceleration is calculated.

[0090] Among them, the measured linear acceleration is the linear acceleration actually measured by the inertial sensor after being converted to the world coordinate system.

[0091] C2: Dynamically allocate the weight coefficient of this sampling moment in the iteration based on the absolute value of the acceleration residual;

[0092] Step C2 specifically includes: when the absolute value is less than or equal to the preset weight threshold, assigning a first weight to the sampling time; when the absolute value is greater than the preset weight threshold, assigning a second weight to the sampling time; wherein, the first weight is the full weight, and the second weight is determined based on the preset weight threshold divided by the absolute value of the acceleration residual.

[0093] C3: By dynamically allocating the weight coefficients of each sampling time in the current iteration, a diagonal weight matrix is ​​constructed, and the estimated value of the six-dimensional parameter vector is updated based on the weight matrix using the weighted least squares method in the current iteration, and then the next iteration is entered until the preset convergence condition is met.

[0094] A dynamic weighted iterative least squares method based on acceleration residuals automatically assigns lower weights to outlier sampling points caused by soft tissue artifacts such as muscle swaying and loosening of binding. This further enhances the robustness to non-ideal motion disturbances in total knee arthroplasty.

[0095] As an alternative implementation, in the process of solving the theoretical position of the hip joint rotation center using regression algorithms and physiological range constraints, a regularization constraint term based on Bayesian prior information can be added to the component of the three-dimensional position vector in the Z-axis direction of the sensor coordinate system. The regularization constraint term promotes the convergence of the solution results toward the prior mean, so as to compensate for the insufficient observability of the Z-axis direction caused by the concentration of the motion trajectory in the horizontal plane during functional motion.

[0096] By introducing soft constraints on the Z-axis, namely Bayesian prior regularization, the problem of weak observability of the Z-axis direction in functional circular motion can be effectively solved. This avoids the divergence of the Z-axis components or serious deviation from physiological reality caused by relying solely on motion data, and ensures the spatial stability and physiological consistency of the solution results of the three-dimensional hip joint rotation center.

[0097] In practice, physiological range constraints specifically include:

[0098] After obtaining the position vector of the hip joint rotation center in the sensor coordinate system, the magnitude of this position vector is calculated.

[0099] The model length is compared with a preset physiological range, which is determined based on the statistical characteristics of adult femur anatomy, the height of the subject, preoperative imaging measurement data and / or statistical morphological models.

[0100] If the modulus exceeds the physiological range, the current solution result is considered abnormal.

[0101] Physiological range constraints are used to verify the anatomical rationality of the solution results. When severe binding slippage or irregular movement leads to abnormal solution results, it can be automatically identified and judged as abnormal, preventing incorrect navigation parameters from being used for subsequent osteotomy guidance, thus playing an important role in safety redundancy.

[0102] Step S105: Determine the femoral mechanical axis based on the theoretical position and the position of the knee joint center. Calculate the varus / valgus angle and flexion / extension angle of the osteotomy guide plate relative to the femoral mechanical axis in real time and generate the corresponding angle deviation to guide the operator to adjust the osteotomy guide plate to the target angle range.

[0103] To facilitate the operator's adjustment of the osteotomy guide plate to the target angle range, a preset color-coding scheme can be used to hierarchically and visually encode angle deviations, thereby displaying the current inversion / valgus angle and flexion-extension angle to the operator in real time in a visual format. Specifically, the color-coding scheme may include:

[0104] When the absolute value of the deviation between the inward or outward turning angle and the target angle is less than or equal to the first deviation threshold, it is displayed in the first color area;

[0105] When the absolute value of the deviation is greater than the first deviation threshold and less than or equal to the second deviation threshold, it is displayed in the second color area;

[0106] When the absolute value of the deviation is greater than the second deviation threshold and less than or equal to the third deviation threshold, it is displayed in the third color area;

[0107] When the absolute value of the deviation is greater than the third deviation threshold, it is displayed in the fourth color area;

[0108] Among them, the colors of the first color area, the second color area, the third color area and the fourth color area are different from each other, and the first color area corresponds to the osteotomy guide plate being adjusted to the target angle range.

[0109] By setting up a multi-level color coding scheme, which maps continuous angular deviation values ​​to intuitive multiple color areas, the cognitive burden of reading values ​​for the operator is reduced, making the adjustment of the osteotomy guide plate faster and more intuitive, effectively improving execution efficiency and operating experience.

[0110] The following detailed description uses femoral osteotomy navigation in right total knee arthroplasty as an example to illustrate the specific implementation of the present invention. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0111] First, the device is connected and information is entered. The operator opens the navigation application installed on the processing device, which can be a smartphone or tablet, containing a processor, memory, display screen, and wireless communication module. In the application interface, the operator enters the basic information of the subject to be tested, so as to associate it with the subsequently generated femoral localization results.

[0112] Subsequently, the navigation application initiates a wireless communication scanning function to search for available miniature IMUs, i.e., miniature inertial sensors, in the surrounding area. The miniature IMU used in this embodiment may include a three-axis accelerometer and a three-axis gyroscope (or magnetometer). Its size and weight can be configured according to the actual clinical application scenario. It can be fixed to the stable fixation area corresponding to the target femur using dedicated or general-purpose fixation devices; the fixation devices include medical double-sided tape, dedicated straps, clamping fixation devices, and / or adhesive fixation devices. The miniature IMU integrates a data output module, which can transmit raw acceleration data, angular velocity data, and attitude quaternion data calculated by its built-in algorithm to the processing device via a wireless link at a preset sampling frequency.

[0113] After the application scans for the miniature IMU, it establishes a connection and binds it via a wireless transmission protocol, displaying the current connection status and signal strength indicator on the application interface. If the connection is confirmed to be normal and no signal strength indicator value is detected below a preset strength threshold, and / or acceleration abrupt changes exceed a preset acceleration jump threshold, and / or the root mean square of the motion amplitude of the angular velocity dynamic data is below a preset motion amplitude threshold, the application proceeds to the subsequent calibration and registration process. The preset strength threshold can be set according to the bandwidth required for data transmission to ensure timely data transmission; the preset acceleration jump threshold and preset motion amplitude threshold can be set according to the actual situation of functional motion to distinguish between functional and non-functional motion.

[0114] After completing the device connection, the static calibration phase begins. The operator clicks the "Start Registration" button on the application interface, at which point the target femur needs to remain absolutely still for approximately three seconds. During this static period, the miniature IMU continuously sends static acceleration and angular velocity data to the processing device, with the sampling frequency preferably set to 100 Hz. Upon receiving this data, the processor on the processing device first executes a static detection algorithm, calculating the amplitude and variance of the angular velocity data to determine if the target femur is truly at rest. If both the amplitude and variance of the detected angular velocity data are greater than preset values, significant movement is determined, and the application will prompt the operator to remain still and re-acquire data; otherwise, the static state is confirmed. After confirming the static state is valid, the processor uses the gravitational acceleration vector measured by the miniature IMU to perform a gravity self-check and align it with the sensor coordinate system.

[0115] Specifically, the resultant vector measured by a miniature IMU in a static state theoretically contains only gravitational acceleration; therefore, the direction of this vector is the direction of gravity in the world coordinate system. For example... Figure 2 As shown, the processor automatically determines the sensor coordinate system (X) based on this gravity vector. s ,Y s Z s ) and world coordinate system (X) w ,Y w Z w The system establishes the relative rotational relationship between the sensor coordinate system and the world coordinate system, and aligns the initial coordinate system. During this process, the operator does not need to manually specify the X, Y, and Z coordinate axes of the miniature IMU; the system can adaptively identify the installation orientation of the miniature IMU. Simultaneously, the processor calculates the average value of the gyroscope and acceleration data acquired during the static period as the initial zero-bias error, which is subtracted in subsequent dynamic data processing. At this point, the initial deviation between the sensor coordinate system and the world coordinate system is reduced to within the preset calibration threshold, and the static calibration step is complete.

[0116] After the static calibration is complete, the application interface will prompt the operator to "shake your thigh" to guide the dynamic data acquisition of functional movements. For example... Figure 2 As shown, the operator uses one hand to stabilize the distal end of the target femur to maintain relative stability between the miniature IMU and the target femur, while using the other hand to hold the subject's lower leg and perform small-amplitude circular or swinging movements with the hip joint of the target femur as the center. The recommended duration of this functional movement is eight to twelve seconds; in this embodiment, ten seconds is used.

[0117] During functional motion, the miniature IMU acquires and outputs dynamic acceleration data, dynamic angular velocity data, and attitude quaternion data in real time, maintaining a data output frequency of 100 Hz. The data receiving module of the processing device receives this data, which is then synchronously stored and analyzed in real time by the processor. During the execution of functional motion, the processor simultaneously initiates a data quality monitoring mechanism, which includes monitoring of the following multiple dimensions:

[0118] First, continuously monitor the signal connection strength indicator value between the miniature IMU and the processing device. When the indicator value is continuously lower than the preset strength threshold, such as -80dBmw for more than two seconds, the application will issue a prompt, suggesting that the operator move closer to the device or check the antenna environment.

[0119] Second, the waveform characteristics of the acceleration dynamic data are analyzed in real time. If the acceleration jump between adjacent sampling points exceeds the preset acceleration jump threshold, such as three times the acceleration of gravity, it is determined that the sensor may have collided or become loose, and the corresponding prompt information is triggered.

[0120] Third, the root mean square of the motion amplitude of the dynamic angular velocity data is calculated. If this value is lower than a preset motion amplitude threshold, such as 0.5 rad / s, the motion amplitude is considered too small and may not be sufficient to generate effective observation information. This will guide the operator to adjust the motion amplitude or re-execute the functional motion. This data quality monitoring mechanism provides a closed-loop guarantee for the reliability of subsequent calculation results.

[0121] After functional motion acquisition is completed, the processor will proceed to the hip joint rotation center calculation stage. First, the raw acceleration and angular velocity dynamic data are preprocessed, including using an SG (Savitzky-Golay) smoothing filter to remove high-frequency noise. The filter window width can be set to eleven sampling points, and the polynomial order is three. Simultaneously, the data is segmented into static and dynamic segments based on the angular velocity amplitude, retaining only the valid dynamic data segments for subsequent modeling. Then, the observation equations are constructed based on the rigid body kinematics model.

[0122] During functional motion, the target femur and the miniature IMU fixed to its surface can be considered as a rigid body undergoing spherical motion around the rotation center of the hip joint. According to the principles of rigid body kinematics, the theoretical linear acceleration experienced by the miniature IMU in the world coordinate system consists of the vector sum of centripetal acceleration and tangential acceleration. This vector sum is transformed from the sensor coordinate system to the world coordinate system via the rotation matrix corresponding to the attitude quaternion data, and then superimposed with the gravitational acceleration vector and acceleration bias term in the world coordinate system.

[0123] In one alternative implementation, the observation equation can be expressed with reference to the following function:

[0124] a w (t)=R(t)·(ω(t)×(ω(t)×r b )+α(t)×r b )+g w +c w

[0125] In the formula, a w (t) represents the linear acceleration in the world coordinate system at time t, R(t) is the rotation matrix calculated from the attitude quaternion, (ω(t)) is the angular velocity vector, α(t) is the angular acceleration vector obtained by numerical differentiation of the angular velocity, and r b Let g be the unknown position vector pointing from the miniature IMU to the center of hip joint rotation in the sensor coordinate system. w Let c be the vector of gravitational acceleration. w This is the acceleration bias term caused by the temperature drift and zero bias of the micro IMU itself. By setting the theoretical linear acceleration equal to the linear acceleration actually measured by the micro IMU and transformed to the world coordinate system, we can construct the observation equation with the unknown position vector and acceleration bias term as the variables to be solved.

[0126] To accurately solve the above observation equations, the processor employs a joint prediction strategy, constructing a six-dimensional parameter vector by combining the three-dimensional position vector in the sensor coordinate system (i.e., the unknown position vector) with the three components of the acceleration bias term in the world coordinate system, i.e., x=[r x ,r y ,r z ,c x ,c y ,c zSubstituting the data at each sampling time into the observation equation yields a system of linear equations about a six-dimensional parameter vector. Due to the unavoidable introduction of non-ideal factors such as muscle swaying, binding micro-motion, and soft tissue artifacts during functional motion, the observed values ​​at some sampling points will significantly deviate from the ideal rigid body motion model, forming outliers. Directly using ordinary least squares to solve this problem will result in significant deviations in the solution. Therefore, this invention employs an objective function based on the Huber loss function and uses an iterative weighted least squares method to robustly optimize and solve this objective function. The specific iterative process is as follows:

[0127] First, the initial estimated value of the six-dimensional parameter vector is calculated using ridge regression to enhance numerical stability. Then, an iterative loop is entered. In each iteration, for each sampling time, the theoretical predicted acceleration at that time is calculated based on the estimated value of the current six-dimensional parameter vector, and the acceleration residual between it and the measured linear acceleration is calculated. The weight coefficient of that sampling time in the current iteration is dynamically assigned according to the magnitude of the absolute value of the acceleration residual: when the absolute value of the residual is less than or equal to a preset weight threshold (e.g., 1.345), the sampling time is assigned the first weight, i.e., a full weight of 1; when the absolute value of the residual is greater than the preset weight threshold, the sampling time is assigned the second weight, the value of which is equal to the preset weight threshold divided by the absolute value of the acceleration residual.

[0128] The weights at all sampling times form a diagonal weight matrix. Then, in the current iteration, the predicted value of the six-dimensional parameter vector is updated using weighted least squares based on this weight matrix, and the process proceeds to the next iteration. The iteration terminates when the change in the six-dimensional parameter vector is less than the convergence tolerance or the number of iterations reaches a preset upper limit, satisfying the preset convergence condition. This weighting mechanism automatically reduces the contribution of outliers to the final solution, thereby further suppressing the interference of soft tissue artifacts on the localization of the hip joint rotation center.

[0129] Furthermore, since the trajectories of functional motions are mainly concentrated in the horizontal plane, the motion excitation of the micro-IMU in the Z-axis direction perpendicular to the motion plane is relatively weak, resulting in insufficient observability of the Z-axis component of the position vector in the observation equation. To stabilize the three-dimensional solution results, this invention adds a regularization constraint term based on Bayesian prior information, namely the Z-axis soft constraint, to the above iterative optimization objective function for the component of the three-dimensional position vector in the Z-axis direction of the sensor coordinate system. This regularization constraint term promotes the solution results to converge moderately towards the prior mean, avoiding significant deviations of the Z-axis component from physiological reality due to data noise.

[0130] After the iterative convergence yields the final solution for the six-dimensional parameter vector, the processor further performs physiological range constraint verification. Specifically, it calculates the magnitude of the obtained position vector rb and compares it with a preset physiological range interval. This physiological range interval can be pre-set or individually adjusted based on adult femoral anatomical statistical characteristics, the height of the subject, preoperative imaging measurement data, and / or statistical morphological models. If the calculated magnitude is within this interval, the current solution is considered valid; if the magnitude exceeds this interval, the solution is considered abnormal, possibly due to severe slippage of the binding or improper movement. In this case, the application will trigger an error message, guiding the operator to check the sensor fixation status and re-execute the functional motion acquisition and calculation process.

[0131] After successfully obtaining the effective theoretical position of the hip joint rotation center in the sensor coordinate system, and combining it with the pre-known or measured knee joint center position, the processor can uniquely determine the spatial orientation of the femoral mechanical axis. Subsequently, the application automatically enters the real-time navigation interface. In this interface, the processor calculates the varus / valgus angle and flexion / extension angle of the osteotomy guide relative to the femoral mechanical axis in real time at an update rate of no less than 100 Hz, and generates the corresponding angular deviations.

[0132] To intuitively guide the operator in adjusting the osteotomy guide to the target angle range, the application employs a tiered color-coded visualization scheme. Specifically, when the absolute value of the deviation between the varus / valgus angle or flexion / extension angle and the target angle is less than or equal to the first deviation threshold (e.g., one degree), the dashboard area or angle value on the interface is displayed in the first color (e.g., green), indicating that the osteotomy guide is accurately positioned. When the absolute value of the deviation is greater than the first deviation threshold but less than or equal to the second deviation threshold (e.g., two degrees), it is displayed in the second color (e.g., blue), indicating that slight adjustments are needed. When the absolute value of the deviation is greater than the second deviation threshold but less than or equal to the third deviation threshold (e.g., three degrees), it is displayed in the third color (e.g., orange), indicating that a larger adjustment is needed. When the absolute value of the deviation is greater than the third deviation threshold, it is displayed in the fourth color (e.g., red), indicating that the current deviation is too large and repositioning is required. Based on the color changes and numerical feedback on the screen, the operator holds the osteotomy guide and makes fine adjustments to its posture until the angle deviation decreases and enters the green area. At this point, the operator can confirm the position of the osteotomy guide and perform subsequent osteotomy operations.

[0133] Of course, voice prompts can also be used to guide the operator in adjusting the osteotomy guide to the target angle range. When the absolute value of the deviation between the varus / valgus angle or flexion / extension angle and the target angle is less than or equal to the first deviation threshold (e.g., one degree), the voice prompt "Ready" or "Normal" indicates that the osteotomy guide is accurately positioned. When the absolute value of the deviation is greater than the first deviation threshold but less than or equal to the second deviation threshold (e.g., two degrees), the voice prompt "Level 1 deviation," "Slight deviation," or "Abnormal" indicates that a slight adjustment is needed. When the absolute value of the deviation is greater than the second deviation threshold but less than or equal to the third deviation threshold (e.g., three degrees), the voice prompt "Level 2 deviation," "Large deviation," or "Abnormal" indicates that a larger adjustment is needed. When the absolute value of the deviation is greater than the third deviation threshold, the voice prompt "Level 3 deviation," "Excessive deviation," or "Abnormal" indicates that the current deviation is too large and repositioning is required.

[0134] After navigation is complete, the operator clicks the "Record Data" or "End" button in the application. The system will automatically save all key parameters from the navigation process, including the subject's information, target femoral side, the three-dimensional coordinates of the hip joint rotation center in the sensor coordinate system, femoral mechanical axis parameters, the final confirmed varus / valgus angle and flexion / extension angle, and a timestamped operation record. Using this data stored in memory, the application automatically generates a femoral localization result, which can be archived, printed, or shared wirelessly.

[0135] In other alternative embodiments of the present invention, certain specific technical details in the above embodiments can be adjusted or replaced accordingly. For example, in addition to Bluetooth Low Energy, the wireless communication method between the micro IMU and the processing device can be replaced with Wi-Fi, Universal Serial Bus wired connection, or dedicated radio frequency link. The corresponding device connection management module only needs to adapt to the corresponding communication protocol without affecting the implementation of the core algorithm.

[0136] Regarding the number of sensors, although this invention only requires a single micro IMU to complete femoral side navigation, in application scenarios that require simultaneous evaluation of tibial osteotomy or real-time joint kinematics, a second micro IMU can be added to the tibial side. In this case, the robust solution algorithm for the femoral hip joint rotation center in this invention can still be directly reused as a core module.

[0137] In terms of attitude calculation, the attitude quaternion output by the miniature IMU can be provided by its built-in attitude fusion algorithm. This algorithm can use different schemes such as complementary filtering, extended Kalman filtering or unscented Kalman filtering. It can also introduce magnetometer data for heading assistance. However, when there is significant magnetic interference in the instrument, the magnetometer can be turned off to maintain data accuracy.

[0138] Regarding physiological range constraints, the physiological range interval can be further individualized based on the height of the person being tested, preoperative imaging measurement data, or statistical morphological model, thereby further improving the stability of the solution under weak excitation exercise conditions.

[0139] In terms of the display method of the navigation interface, in addition to presenting it in the form of a color-coded numerical dashboard on the processing device screen, the angle deviation information can also be projected onto augmented reality glasses or directly onto the surgical field area through a micro projector. The specific threshold range of color coding can also be customized according to different usage needs.

[0140] In summary, this invention provides a detailed description of an inertial measurement-based assisted navigation method for total knee arthroplasty through the above embodiments. It fully covers the entire process from device connection, static adaptive calibration, functional motion acquisition and quality monitoring, robust joint estimation of hip joint rotation center, to real-time osteotomy angle navigation guidance and result generation, and provides a variety of feasible alternative implementation methods.

[0141] Secondly, based on the same inventive concept, embodiments of the present invention provide an inertial measurement-based auxiliary navigation system for total knee arthroplasty, please refer to... Figure 3 As shown, the system may include:

[0142] An inertial sensor 201 is mounted on the target femur. The inertial sensor 201 is used to collect static acceleration data and static angular velocity data when the target femur is at rest, and to collect dynamic acceleration data and dynamic angular velocity data in real time during functional motion. The inertial sensor 201 also includes a data output module.

[0143] The positioning and navigation terminal 202 includes a processor, a memory, a data receiving module, and a human-machine interaction device; wherein, the data receiving module is communicatively connected to the data output module, and the data receiving module is used to receive static acceleration data, static angular velocity data, dynamic acceleration data, and dynamic angular velocity data; the processor executes any one of the embodiments in the first aspect above; the human-machine interaction device is used to guide the operator to adjust the osteotomy guide plate to the target angle range.

[0144] Human-computer interaction devices may include displays, speakers, etc. The inertial measurement-based auxiliary navigation system for total knee arthroplasty provided in this embodiment of the invention features high integration and portability. It can reduce the requirements for cost, installation conditions, space occupation, and learning costs of complex optical positioning equipment or surgical robot systems, and reduce the impact of experience differences caused by traditional operation methods that rely on surgeon experience.

[0145] Thirdly, based on the same inventive concept, embodiments of the present invention provide a readable storage medium having a program stored thereon, which, when executed by a processor, implements any of the embodiments described in the first aspect above.

[0146] The embodiments of the present invention solidify the above-mentioned inertial measurement-based assisted navigation method for total knee arthroplasty in the form of a software program, so as to facilitate deployment and updates on general-purpose processing devices, reduce hardware customization costs, and improve the flexibility of the technical solution in promotion.

[0147] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.

[0148] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0149] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0150] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0151] These computer program instructions 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 of the function specified in one or more boxes.

[0152] It should also be noted that the exemplary embodiments mentioned in this invention describe methods or systems based on a series of steps or apparatus. However, this invention is not limited to the order of the steps described above; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0153] The above are merely specific embodiments of the present invention. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the protection scope of the present invention.

Claims

1. A navigation-assisted method for total knee arthroplasty based on inertial measurement, characterized in that, include: An inertial sensor is installed on the target femur, and when the target femur is in a stationary state, the static acceleration data and static angular velocity data output by the inertial sensor are collected. The sensor coordinate system of the inertial sensor is calibrated based on the static acceleration data and the static angular velocity data, so that the deviation between the sensor coordinate system and the world coordinate system is less than the calibration threshold. Functional motion is performed with the hip joint of the target femur as the center of the ball, and the acceleration dynamic data, angular velocity dynamic data and attitude quaternion data output by the inertial sensor are collected in real time during the functional motion. An observation equation is constructed based on a rigid body kinematics model. The observation equation is used to process the acceleration dynamic data, the angular velocity dynamic data, and the attitude quaternion data to determine the position vector and acceleration bias term of the hip joint rotation center in the sensor coordinate system. The theoretical position of the hip joint rotation center is then solved using a regression algorithm and physiological range constraints. Based on the theoretical position and the position of the knee joint center, the femoral mechanical axis is determined, and the varus / valgus angle and flexion / extension angle of the osteotomy guide plate relative to the femoral mechanical axis are calculated in real time and the corresponding angle deviation is generated to guide the operator to adjust the osteotomy guide plate to the target angle range.

2. The inertial measurement-based assisted navigation method for total knee arthroplasty according to claim 1, characterized in that, The observation equations constructed based on the rigid body kinematics model include: During the functional movement, the target femur and its inertial sensor are defined as rigid bodies undergoing spherical motion around the rotation center of the hip joint; Based on the principles of rigid body kinematics, an expression for the theoretical linear acceleration of the inertial sensor in the world coordinate system is constructed. The theoretical linear acceleration is equal to the vector sum of the centripetal acceleration and the tangential acceleration. After the attitude quaternion data is transformed from the sensor coordinate system to the world coordinate system, the gravitational acceleration vector in the world coordinate system and the acceleration bias term are superimposed. The vector sum of the centripetal acceleration and the tangential acceleration is determined by the dynamic angular velocity data, the angular acceleration data, and the unknown position vector pointing from the inertial sensor to the center of rotation of the hip joint; the acceleration bias term is determined by the error of the inertial sensor itself, and the angular acceleration data is obtained by differentiating the dynamic angular velocity data. Based on the fact that the theoretical linear acceleration is equal to the linear acceleration actually measured by the inertial sensor after being converted to the world coordinate system, the observation equation is constructed with the unknown position vector and the acceleration bias term as variables to be solved.

3. The inertial measurement-based assisted navigation method for total knee arthroplasty according to claim 2, characterized in that, The process of using the observation equation to process the acceleration dynamic data, the angular velocity dynamic data, and the attitude quaternion data to determine the position vector and acceleration bias term of the hip joint rotation center in the sensor coordinate system includes: The unknown position vector is constructed as a three-dimensional position vector in the sensor coordinate system, and the three-dimensional position vector and the three components of the acceleration bias term in the world coordinate system are jointly determined as a six-dimensional parameter vector; wherein, the three-dimensional position vector represents the relative positional relationship from the inertial sensor to the hip joint rotation center; The six-dimensional parameter vector is used as the variable to be solved in the observation equation for joint solution, so as to reduce the cumulative error introduced by solving the position vector and the acceleration bias term separately.

4. The inertial measurement-based assisted navigation method for total knee arthroplasty according to claim 3, characterized in that, The method of solving the theoretical position of the hip joint rotation center using regression algorithms and physiological range constraints includes: The functional motion process includes multiple sampling moments. For each sampling moment, the theoretically predicted acceleration is calculated based on the estimated value of the six-dimensional parameter vector in the current iteration round, and the acceleration residual between the theoretically predicted acceleration and the measured linear acceleration is calculated. The measured linear acceleration is the linear acceleration actually measured by the inertial sensor after being converted to the world coordinate system. The weight coefficient of the sampling moment in the iteration is dynamically allocated according to the absolute value of the acceleration residual, including: when the absolute value is less than or equal to a preset weight threshold, a first weight is assigned to the sampling moment; when the absolute value is greater than the preset weight threshold, a second weight is assigned to the sampling moment; wherein, the first weight is the full weight, and the second weight is determined by dividing the preset weight threshold by the absolute value of the acceleration residual; By dynamically allocating the weight coefficients of each sampling time in the current iteration, a diagonal weight matrix is ​​constructed. In the current iteration, the estimated value of the six-dimensional parameter vector is updated based on the weight matrix using the weighted least squares method, and the process proceeds to the next iteration until the preset convergence condition is met.

5. The inertial measurement-based assisted navigation method for total knee arthroplasty according to claim 4, characterized in that, The method of solving the theoretical position of the hip joint rotation center using regression algorithms and physiological range constraints also includes: In the objective function of the iterative optimization solution, a regularization constraint term based on Bayesian prior information is added for the component of the three-dimensional position vector in the Z-axis direction of the sensor coordinate system. The regularization constraint term causes the solution to converge toward the prior mean, thus compensating for the insufficient observability of the Z-axis direction in the functional motion due to the motion trajectory being concentrated on the horizontal plane.

6. The inertial measurement-based assisted navigation method for total knee arthroplasty according to claim 1, characterized in that, The physiological range constraints specifically include: After obtaining the position vector of the hip joint rotation center in the sensor coordinate system, the magnitude of the position vector is calculated. The model length is compared with a preset physiological range, which is determined based on the statistical characteristics of adult femur anatomy, the height of the subject, preoperative imaging measurement data and / or statistical morphological models. If the modulus exceeds the physiological range, the current solution result is determined to be abnormal.

7. The inertial measurement-based assisted navigation method for total knee arthroplasty according to claim 1, characterized in that, During the functional motion, the data quality of the inertial sensor is also monitored in real time. The real-time monitoring of the data quality of the inertial sensor includes: real-time monitoring of the signal connection strength indication value between the inertial sensor and the processing device; and real-time analysis of the waveform characteristics of the acceleration dynamic data and the angular velocity dynamic data to identify whether there are sudden acceleration changes or abnormal angular velocity drifts caused by loose binding. When any one or more of the following conditions are detected, the operator is guided to check the stationary state of the inertial sensor or re-execute the functional motion: The signal connection strength indicator value is lower than a preset strength threshold; The acceleration abrupt change exceeds a preset acceleration jump threshold; The root mean square of the motion amplitude of the dynamic angular velocity data is lower than a preset motion amplitude threshold.

8. The inertial measurement-based assisted navigation method for total knee arthroplasty according to claim 1, characterized in that, The step of guiding the operator to adjust the osteotomy guide plate to the target angle range includes: A preset color coding scheme is used to hierarchically and visually encode the angular deviations, displaying the current inversion / exversion angles and flexion / extension angles to the operator in real time in a visual format; the color coding scheme includes: When the absolute value of the deviation between the inward / outward turning angle and the target angle is less than or equal to the first deviation threshold, it is displayed as the first color area; When the absolute value of the deviation is greater than the first deviation threshold and less than or equal to the second deviation threshold, it is displayed as the second color area; When the absolute value of the deviation is greater than the second deviation threshold and less than or equal to the third deviation threshold, it is displayed in the third color area; When the absolute value of the deviation is greater than the third deviation threshold, it is displayed in the fourth color area; The colors of the first color region, the second color region, the third color region, and the fourth color region are all different, and the first color region corresponds to the osteotomy guide plate being adjusted to the target angle range.

9. An inertial measurement-based auxiliary navigation system for total knee arthroplasty, characterized in that, include: An inertial sensor is mounted on the target femur. The inertial sensor is used to collect static acceleration data and static angular velocity data when the target femur is at rest, and to collect dynamic acceleration data and dynamic angular velocity data in real time during the functional motion. The inertial sensor includes a data output module. A positioning and navigation terminal includes a processor, a memory, a data receiving module, and a human-computer interaction device; wherein, the data receiving module is communicatively connected to the data output module, and the data receiving module is used to receive the static acceleration data, the static angular velocity data, the dynamic acceleration data, and the dynamic angular velocity data; The processor executes the method as described in any one of claims 1 to 8; the human-computer interaction device is used to guide the operator to adjust the osteotomy guide plate to the target angle range.

10. A readable storage medium having a program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 8.

Images