Knee joint traction rehabilitation method based on binocular vision

Abstract

The invention relates to the technical field of medical rehabilitation, and particularly discloses a knee joint traction rehabilitation method based on binocular vision, and the method comprises the steps: fixing knee joint rehabilitation equipment, adjusting the height and angle of a support module, and installing and calibrating a binocular vision module at the top of a support; the legs of a patient are placed on a traction platform, the height and angle of the platform are adjusted, three-dimensional coordinates of knee joints are calibrated through a binocular vision module, initial reference data are recorded, a traction force generation module is adjusted to set initial traction force, and traction force data are monitored in real time; according to the captured three-dimensional coordinates of the knee joint calibration points, the knee joint angle, displacement and the total length of the motion trail are dynamically calculated, a dynamic mechanical model is established, and the size and direction of traction force and parameters of knee joint rehabilitation equipment are adjusted in real time through a PI D control algorithm. The device is safe and convenient, effectively improves the rehabilitation effect, has the characteristics of economy and high efficiency, and can better meet the rehabilitation treatment requirements of patients.

Description

Technical Field

[0001] The present invention relates to the field of medical rehabilitation technology, and more specifically, to a binocular vision-based knee joint traction rehabilitation method. Background Art

[0002] Knee osteoarthritis is a common degenerative joint disease, which is mainly manifested by the degeneration of articular cartilage, the reduction of joint space and the formation of bone spurs, which eventually lead to symptoms such as joint pain, swelling, stiffness and limited function. The disease is common in the middle-aged and elderly population and is one of the main causes of movement disorders, seriously affecting the quality of life of patients. Symptoms can be relieved by knee traction technology. The existing public document 1 (Optimum kinematic design of a planarcable-driven parallel robot with wrench-closure gait trajectory [J]. Mechanism and Machine Theory, 2016, 99: 1-18.) has developed a rope traction knee rehabilitation robot, which uses four ropes to traction the patient's calf and the continuous movement of the treadmill to achieve gait trajectory training of the lower limbs. However, since the four ropes of the robot act on the calf, the driving force of the rope is increased, which is not conducive to the safety of rehabilitation training. At the same time, the patient needs to perform rehabilitation training on a treadmill, which requires strong synchronization between the rope traction and the treadmill, otherwise it will also be not conducive to the safety of rehabilitation training. The existing open document 2 (Gait master: A versatile locomotion interface for uneven virtual terrain [C] / / Proceedings IEEE Virtual Reality 2001. IEEE, 2001: 131-137.) has developed a pedal device including two sets supported by three hydraulic cylinders, each set of which has three degrees of freedom and can complete various forms of motion of the human foot, but the cost of the device is high and the subsequent maintenance cost is large. Therefore, it is urgent to develop a safe, convenient, economical and efficient knee joint traction rehabilitation method to better meet the needs of patients for rehabilitation treatment. Summary of the invention

[0003] In order to overcome the above-mentioned defects of the prior art, the present invention provides a knee joint traction rehabilitation method based on binocular vision, which is achieved at low cost by fixing knee joint rehabilitation equipment, adjusting bracket and platform parameters, attaching reflective markers and calibrating binocular vision modules. At the same time, the three-dimensional coordinates of the knee joint are calibrated, and the calibration point data is captured in real time by binocular vision. The relevant parameters of the knee joint are calculated, and a dynamic mechanical model is established in combination with the PID control algorithm. The traction force, direction and parameters of the knee joint rehabilitation equipment are adjusted to ensure safety and comfort, so as to solve the problems raised in the above background technology.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A binocular vision-based knee joint traction rehabilitation method comprises the following steps:.

[0006] Step S1, fix the knee joint rehabilitation device, adjust the height and angle of the bracket module, and install and calibrate the binocular vision module on the top of the bracket;

[0007] Step S2, placing the patient's legs on a traction platform and adjusting the platform height and angle, calibrating the three-dimensional coordinates of the knee joint through a binocular vision module, adjusting the traction force generation module to set the initial traction force and monitoring the traction force data in real time;

[0008] Step S3, dynamically calculating the knee joint angle, displacement and total length of the motion trajectory, establishing a dynamic mechanical model and using a PID control algorithm to adjust the traction force, direction and parameters of the knee joint rehabilitation equipment in real time;

[0009] Step S4, quantifying the patient's range of motion and activity by analyzing the data curves of the knee joint angle, traction force and motion trajectory;

[0010] In step S3, the PID control algorithm is used to optimize the traction parameters to meet the following optimization objectives: in, is the traction force at the next moment of the current moment t, is the traction force at the current moment t, e(t) is the traction force error, K p is the proportional control coefficient, K d is the differential control coefficient, K i is the integral control coefficient.

[0011] As a further solution of the present invention, in step S1, the knee traction rehabilitation device is placed on a stable ground to ensure that the base module of the knee traction rehabilitation device is firmly fixed. The base module adopts an aluminum alloy frame design, has the characteristics of lightness and durability, and is equipped with a rubber anti-skid pad at the bottom to prevent the knee traction rehabilitation device from displacement or shaking during the traction process. At the same time, in the selection of the installation position of the knee traction rehabilitation device, it should be ensured that the patient can lie in front of the knee traction rehabilitation device conveniently, and the height and angle of the bracket module of the knee traction rehabilitation device are adjusted to match the patient's knee physiological position. The bracket module has a built-in slide rail system and a locking device, which can provide a height adjustment range from 15 to 23 inches and an angle adjustment function from 24.5° to 34.9°. A binocular vision module is installed on the top of the bracket module so that its field of view covers the patient's knee joint area without obvious obstruction. The binocular vision module consists of a pair of high-resolution cameras and a calibration calculation unit. The position of the camera is kept at a distance of 50 to 70 cm from the patient's knee joint, which not only ensures that a clear image of the calibration point is captured, but also avoids image distortion due to too close distance. After the camera is installed, the binocular vision module is calibrated through the calibration plate, and the image data of the calibration plate under multiple viewing angles is obtained in turn, and the intrinsic and extrinsic parameters of the camera are calculated using the visual calibration algorithm. The intrinsic parameters include focal length, optical center position and distortion coefficient, and the extrinsic parameters include rotation matrix and baseline distance. During the calibration process, it should be ensured that the calibration plate is clearly visible in each frame of the image, and the calibration plate plane is kept at a certain angle with the camera, and the angle range is 15° to 30°.

[0012] Attach reflective markers to the patient's knee joint area to complete the visual calibration point setting. The reflective markers are made of lightweight, soft and easy-to-adhere materials, which can clearly image in the camera field of view without affecting the patient's comfort. Attaching the markers to the upper edge of the patella, the lower edge of the patella and the middle of the calf of the patient can not only accurately calibrate the three-dimensional coordinates of the knee joint, but also facilitate the capture of the dynamic movement trajectory of the knee joint.

[0013] As a further solution of the present invention, in step S2, when the patient lies in front of the knee traction rehabilitation device, the legs should be placed on the traction platform so that the knee joint is in the traction center position of the knee traction rehabilitation device. Through the built-in slide rail and height adjustment system of the bracket module, the platform height can be adjusted to between 15 and 23 inches, and the angle can be accurately adjusted within the range of 24.5° to 34.9°. The operator should select the appropriate height and angle settings based on the actual physiological bending of the patient's knee joint to ensure that the traction force can be applied along the biomechanical axis of the knee joint, thereby optimizing the treatment effect.

[0014] After completing the platform adjustment, the operator will use a fixing device to fix the patient's legs. The fixing device includes three leg straps, which are used to fix the upper part, lower part and middle part of the calf. The leg straps are made of highly elastic material, which can provide a good fixation effect without causing pressure or discomfort to the patient's legs. In addition, the selection of the fixed point must be consistent with the traction direction to ensure that the traction force can accurately act on the biomechanical axis of the knee joint. Next, the operator starts the binocular vision module to perform an initial calibration for the knee joint. After the binocular camera in the binocular vision module captures the image of the calibration point, the three-dimensional coordinates of the knee joint calibration point are calculated through a three-dimensional reconstruction algorithm and recorded as the initial reference data.

[0015] After the calibration is completed, the operator starts to set the initial parameters of the traction force generation module. The traction force module consists of an air pump and a pulley system. The traction force is controlled by adjusting the air pump pressure. The initial traction force is set in the range of 10N to 20N. The built-in pressure sensor of the traction force generation module will display the current traction force data in real time.

[0016] As a further solution of the present invention, in step S3, after the patient's leg fixation and knee traction rehabilitation equipment are initialized, the binocular camera starts the real-time capture function, collects the pixel position data of the key calibration points on the knee joint (such as the upper edge of the patella, the lower edge of the patella, and the middle of the calf) through the camera, and synchronously transmits it to the intelligent control module for processing. The two-dimensional pixel point data collected by the camera is reconstructed by the binocular three-dimensional reconstruction algorithm to generate the three-dimensional spatial coordinates of the calibration points. By analyzing the changes in the three-dimensional coordinates of the calibration points, the knee joint angle, displacement and motion trajectory are dynamically calculated. The calculation formula for the knee joint angle is: Where θ is the bending angle of the knee joint, is the vector in the direction of the thigh, is the vector in the direction of the calf, express The module length, express The calculation formula of knee joint displacement is: Where D is the displacement of the knee joint calibration point, (x 2 ,y 2 ,z 2 ) is the initial three-dimensional coordinate of the knee joint calibration point, (x 2 ′ ,y 2 ′ ,z 2 ′ ) is the real-time three-dimensional coordinate of the knee joint mark; the calculation formula of the motion trajectory of the knee joint is: Where L is the total length of the knee joint motion trajectory, (x2，i ,y 2，i , z 2，i ) is the three-dimensional coordinate of the knee joint calibration point at the i-th time point, and n is the total number of sampling points during the movement.

[0017] The intelligent control module integrates the calculated knee joint angle, displacement, total length of the motion trajectory and traction force data to establish a dynamic mechanical model, and uses the PID control algorithm to finely adjust the traction parameters of the knee joint traction rehabilitation device to improve the rehabilitation effect while ensuring the safety and comfort of the patient. The dynamic mechanical model includes the following steps:

[0018] Step Z1: During traction, the knee joint must meet the mechanical balance, and the calculation formula is: F T ·cosα=F J +F R , where F T is the magnitude of the traction force, α is the angle between the traction force and the axis of the knee joint, F J is the reaction force in the knee joint, F R Resistance created by soft tissue;

[0019] Step Z2: The displacement, angle and motion trajectory of the knee joint satisfy the following relationship: Among them, ΔX is the total displacement of the knee joint calibration point in the X direction, is the instantaneous velocity of the calibration point, Δθ is the change in the knee joint angle, θ(t) is the angle of the knee joint at the current moment, and θ 0 is the angle of the knee joint during initial calibration, and L is the total length of the knee joint motion trajectory.

[0020] As a further solution of the present invention, in step S4, the collected time series data is analyzed to quantify the range of motion and activity of the patient's knee joint. For example: by analyzing the time curve of the change in the knee joint angle, the flexibility of flexion and extension and whether the range of motion has increased can be evaluated; by analyzing the time curve of the change in the knee joint angle, the flexibility of flexion and extension and whether the range of motion has increased can be evaluated. In order to facilitate the understanding and application of the data, the analysis results are presented in a visual manner. The intelligent control module generates real-time charts, such as the time change curve of the knee joint angle, the relationship diagram between traction force and displacement, and the three-dimensional path diagram of the motion trajectory. The chart intuitively displays the dynamic change trend and treatment effect of the patient's knee joint, so that the operator and the patient can understand the rehabilitation progress at a glance.

[0021] The technical effects and advantages of a knee joint traction rehabilitation system of the present invention: the present invention can accurately calibrate the three-dimensional coordinates of the knee joint by fixing the knee joint rehabilitation equipment, adjusting the parameters of the bracket and the platform, attaching reflective markers and calibrating the binocular vision module; during the traction process, the calibration point data is captured in real time by binocular vision, the knee joint related parameters are calculated, and a dynamic mechanical model is established in combination with the PID control algorithm, and the traction force size, direction and knee joint rehabilitation equipment parameters can be adjusted to ensure safety and comfort; by analyzing the quantitative data and displaying it in a visual chart, the treatment effect and rehabilitation progress can be intuitively presented. The present invention is not only safe and convenient, and effectively improves the rehabilitation effect, but also has the characteristics of economic efficiency, and can better meet the rehabilitation treatment needs of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The present invention is a flow chart of a binocular vision-based knee joint traction rehabilitation method.

[0023] Figure 2 This is a schematic diagram of a rope-traction lower limb rehabilitation robot in the prior art.

[0024] Figure 3 The figure is a schematic diagram of a knee joint active gait training rehabilitation robot in the prior art.

[0025] Figure 4 It is a structural schematic diagram of the knee joint traction rehabilitation device of the present invention.

[0026] Figure 5 This is the force free body diagram of the knee joint traction rehabilitation device of the present invention.

[0027] Figure 6 This is a schematic structural diagram of the three-side adjustable design of the knee joint traction rehabilitation device of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] Example 1

[0030] See also Figure 1 As shown in the flowchart, the embodiment of the present invention provides a knee joint traction rehabilitation method based on binocular vision, which includes the following steps:

[0031] A knee joint traction rehabilitation method based on binocular vision comprises the following steps:

[0032] Step S1, fix the knee joint rehabilitation device, adjust the height and angle of the bracket module, and install and calibrate the binocular vision module on the top of the bracket;

[0033] Step S2, placing the patient's legs on the traction platform and adjusting the platform height and angle, calibrating the three-dimensional coordinates of the knee joint through the binocular vision module, recording the initial reference data, adjusting the traction force generation module to set the initial traction force and monitoring the traction force data in real time;

[0034] Step S3, dynamically calculating the knee joint angle, displacement and total length of the motion trajectory according to the captured three-dimensional coordinates of the knee joint calibration points, establishing a dynamic mechanical model and using a PID control algorithm to adjust the traction force, direction and parameters of the knee joint rehabilitation equipment in real time;

[0035] Step S4, quantifying the patient's range of motion and activity by analyzing the data curves of the knee joint angle, traction force and motion trajectory.

[0036] In this embodiment, see Figure 2 The schematic diagram shown is a rope-traction knee joint rehabilitation robot in the prior art. The robot uses four ropes to pull the patient's calf and the continuous movement of the treadmill to achieve gait trajectory training of the lower limbs. However, since the four ropes of the robot act on the calf, the driving force of the ropes increases, which is not conducive to the safety of rehabilitation training. At the same time, the patient needs to perform rehabilitation training on a treadmill, which requires strong synchronization between the rope traction and the treadmill, otherwise it will also be not conducive to the safety of rehabilitation training. Figure 3 The schematic diagram shown is a pedal device developed in the prior art, which includes two sets of pedal devices supported by three hydraulic cylinders. Each set of devices has three degrees of freedom and can complete various forms of movement of the human foot. However, the cost of the device is high and the subsequent maintenance cost is large.

[0037] Further, in step S1, if Figure 4As shown in the structural schematic diagram, the knee traction rehabilitation device is placed on a stable ground to ensure that the base module of the knee traction rehabilitation device is firmly fixed. The base module adopts an aluminum alloy frame design, which is lightweight and durable. The bottom is equipped with a rubber anti-skid pad to prevent the knee traction rehabilitation device from displacement or shaking during the traction process. At the same time, in the selection of the installation position of the knee traction rehabilitation device, it should be ensured that the patient can lie down in front of the knee traction rehabilitation device conveniently, and adjust the height and angle of the bracket module of the knee traction rehabilitation device to match the patient's knee physiological position. The bracket module has a built-in slide rail system and a locking device, which can provide a height adjustment range from 15 to 23 inches and an angle adjustment function from 24.5° to 34.9°. A binocular vision module is installed on the top of the bracket module so that its field of view covers the patient's knee joint area without obvious obstruction. The binocular vision module consists of a pair of high-resolution cameras and a calibration calculation unit. The position of the camera is kept at a distance of 50 to 70 cm from the patient's knee joint, which not only ensures that a clear image of the calibration point is captured, but also avoids image distortion due to too close distance. After the camera is installed, the binocular vision module is calibrated through the calibration plate, and the image data of the calibration plate under multiple viewing angles is obtained in turn, and the intrinsic and extrinsic parameters of the camera are calculated using the visual calibration algorithm. The intrinsic parameters include focal length, optical center position and distortion coefficient, and the extrinsic parameters include rotation matrix and baseline distance. During the calibration process, it should be ensured that the calibration plate is clearly visible in each frame of the image, and the calibration plate plane is kept at a certain angle with the camera, and the angle range is 15° to 30°.

[0038] Attach reflective markers to the patient's knee joint area to complete the visual calibration point setting. The reflective markers are made of lightweight, soft and easy-to-adhere materials, which can clearly image in the camera field of view without affecting the patient's comfort. Attaching the markers to the upper edge of the patella, the lower edge of the patella and the middle of the calf of the patient can not only accurately calibrate the three-dimensional coordinates of the knee joint, but also facilitate the capture of the dynamic movement trajectory of the knee joint.

[0039] In this embodiment, in order to further explain the mechanical phenomena occurring in the legs, as Figure 5 The free body diagram shown models the leg at the tilt angle of the knee traction rehabilitation device, F T is the actual traction force applied to the leg by the knee traction rehabilitation device. The average leg length from the femoral condyle of the knee to the foot is 0.061 times the total body weight. Assuming the patient's weight is 70 kg, when calculating the total weight of the calf, W = 70 × 0.061 = 4.27N. Add the forces on the x-axis and y-axis coordinates to determine the values ​​of all variables:

[0040] ∑F x =F T sin30°+FC -Ncos30°-F M cos30°-F f cos30°=0;

[0041] ∑F y =-F T cos30°-W-Nsin30°+F M sin30°+F f sin30°=0;

[0042] Where: F C is the contact force at the tibiofemoral joint, but when traction is applied, the contact force ideally becomes zero, resulting in the following force equation in the x direction. M is the force exerted by the leg muscles, tendons and ligaments at the knee. Since the leg is in an “open” position and should be relaxed, the force F M can also become zero, simplifying the above equation to the following form:

[0043] ∑F x =F T sin30°-Ncos30°-F f cos30°=0;

[0044] ∑F y =-F T cos30°-W-Nsin30°+F f sin30°=0;

[0045] Knowing all the forces, the device can be manufactured with the forces in mind. The length of the leg rest (hypotenuses) is designed to adapt to the average human leg length. The length from the knee to the sole of the foot is 0.285 times the height. Assuming that the average person is about 1.72 meters tall, the length of the leg rest is adapted to a leg length of 0.4902 meters. Figure 6 As shown in the structural diagram, the three-side adjustment of the knee traction rehabilitation equipment involves the ability to simultaneously adjust the vertical and horizontal components, and the leg rest is capable of sliding on its own to increase or decrease the length.

[0046] Furthermore, in step S2, when the patient lies in front of the knee traction rehabilitation device, the legs should be placed on the traction platform so that the knee joint is in the traction center position of the knee traction rehabilitation device. Through the built-in slide rail and height adjustment system of the bracket module, the platform height can be adjusted to between 15 and 23 inches, and the angle can be accurately adjusted within the range of 24.5° to 34.9°. The operator should select the appropriate height and angle settings based on the actual physiological bending of the patient's knee joint to ensure that the traction force can be applied along the biomechanical axis of the knee joint, thereby optimizing the treatment effect.

[0047] After completing the platform adjustment, the operator will use a fixing device to fix the patient's legs. The fixing device includes three leg straps, which are used to fix the upper part, lower part and middle part of the calf respectively. The leg straps are made of highly elastic material, which can provide a good fixation effect without causing pressure or discomfort to the patient's legs. In addition, the selection of the fixing point must be consistent with the traction direction to ensure that the traction force can accurately act on the biomechanical axis of the knee joint. Next, the operator starts the binocular vision module to perform an initial calibration for the knee joint. After the binocular camera in the binocular vision module captures the image of the calibration point, the three-dimensional coordinates of the knee joint calibration point are calculated through a three-dimensional reconstruction algorithm and recorded as the initial reference data. The initial reference data is shown in the following table:

[0048] Table 1 Initial reference data

[0049]

[0050]

[0051]

[0052] After the calibration is completed, the operator starts to set the initial parameters of the traction force generation module. The traction force module consists of an air pump and a pulley system. The traction force is controlled by adjusting the air pump pressure. The initial traction force is set in the range of 10N to 20N. The built-in pressure sensor of the traction force generation module will display the current traction force data in real time.

[0053] Further, in step S3, after the patient's leg fixation and knee traction rehabilitation equipment are initialized, the binocular camera starts the real-time capture function, collects the pixel position data of the key calibration points on the knee joint (such as the upper edge of the patella, the lower edge of the patella, and the middle of the calf) through the camera, and synchronously transmits it to the intelligent control module for processing. The two-dimensional pixel point data collected by the camera is reconstructed by the binocular three-dimensional reconstruction algorithm to generate the three-dimensional spatial coordinates of the calibration points. By analyzing the changes in the three-dimensional coordinates of the calibration points, the knee joint angle, displacement and motion trajectory are dynamically calculated, and the calculation formula of the knee joint angle is: Where θ is the bending angle of the knee joint, is the vector in the direction of the thigh, is the vector in the direction of the calf, express The module length, express The calculation formula of knee joint displacement is: Where D is the displacement of the knee joint calibration point, (x 2 ,y 2 , z 2) is the initial three-dimensional coordinate of the knee joint calibration point, (x′ 2 , y′ 2 , z′ 2 ) is the real-time three-dimensional coordinate of the knee joint mark; the calculation formula of the motion trajectory of the knee joint is: Where L is the total length of the knee joint motion trajectory, (x 2，i ,y 2，i , z 2，i ) is the three-dimensional coordinate of the knee joint calibration point at the i-th time point, and n is the total number of sampling points during the movement.

[0054] The intelligent control module integrates the calculated knee joint angle, displacement, total length of the motion trajectory and traction force data to establish a dynamic mechanical model, and uses the PID control algorithm to finely adjust the traction parameters of the knee joint traction rehabilitation device to improve the rehabilitation effect while ensuring the safety and comfort of the patient. The dynamic mechanical model includes the following steps:

[0055] Step Z1: During traction, the knee joint must meet the mechanical balance, and the calculation formula is: F T ·cosα=F J +F R , where F T is the magnitude of the traction force, α is the angle between the traction force and the axis of the knee joint, F J is the reaction force in the knee joint, F R Resistance created by soft tissue;

[0056] Step Z2: The displacement, angle and motion trajectory of the knee joint satisfy the following relationship: Among them, ΔX is the total displacement of the knee joint calibration point in the X direction, is the instantaneous velocity of the calibration point, Δθ is the change in the knee joint angle, θ(t) is the angle of the knee joint at the current moment, and θ 0 is the angle of the knee joint during initial calibration, and L is the total length of the knee joint motion trajectory;

[0057] Step Z3, using the PID control algorithm, the traction parameters must meet the following optimization objectives: in, is the traction force at the next moment, is the traction force at the current moment, e(t) is the traction force error, K p is the proportional control coefficient, K d is the differential control coefficient, K i is the integral control coefficient. If the displacement ΔX of the knee joint deviates from the target range, by adjusting F Tand α compensates for the displacement error. If the motion trajectory of the knee joint calibration point shows that the traction direction deviates from the knee joint axis (i.e., α≠0°), the intelligent control module corrects the traction direction by adjusting the angle and height of the traction platform to ensure that the traction force acts perpendicularly on the knee joint axis.

[0058] Furthermore, in step S4, the collected time series data is analyzed to quantify the range of motion and mobility of the patient's knee joint. For example: by analyzing the time curve of the change in the knee joint angle, the flexibility of flexion and extension and whether the range of motion has increased can be evaluated; by analyzing the time curve of the change in the knee joint angle, the flexibility of flexion and extension and whether the range of motion has increased can be evaluated. In order to facilitate the understanding and application of the data, the analysis results are presented in a visual manner. The intelligent control module generates real-time charts, such as the time change curve of the knee joint angle, the relationship diagram between traction force and displacement, and the three-dimensional path diagram of the motion trajectory. The chart intuitively displays the dynamic change trend and treatment effect of the patient's knee joint, so that the operator and the patient can understand the rehabilitation progress at a glance.

[0059] The present invention can accurately calibrate the three-dimensional coordinates of the knee joint by fixing the knee joint rehabilitation equipment, adjusting the parameters of the bracket and platform, attaching reflective markers and calibrating the binocular vision module; during the traction process, the binocular vision is used to capture the calibration point data in real time, calculate the relevant parameters of the knee joint, and establish a dynamic mechanical model in combination with the PID control algorithm, which can adjust the traction force, direction and parameters of the knee joint rehabilitation equipment to ensure safety and comfort; by analyzing the quantitative data and displaying it in a visual chart, the treatment effect and rehabilitation progress can be intuitively presented. The present invention is not only safe and convenient, effectively improves the rehabilitation effect, but also has the characteristics of economic efficiency, and can better meet the rehabilitation treatment needs of patients.

[0060] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0061] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A knee joint traction rehabilitation method based on binocular vision, characterized in that: The steps include: Step S1, fix the knee joint rehabilitation device, adjust the height and angle of the bracket module, and install and calibrate the binocular vision module on the top of the bracket; Step S2, placing the patient's legs on a traction platform and adjusting the platform height and angle, calibrating the three-dimensional coordinates of the knee joint through a binocular vision module, adjusting the traction force generation module to set the initial traction force and monitoring the traction force data in real time; Step S3, dynamically calculating the knee joint angle, displacement and total length of the motion trajectory, establishing a dynamic mechanical model and using a PID control algorithm to adjust the traction force, direction and parameters of the knee joint rehabilitation equipment in real time; Step S4, quantifying the patient's range of motion and activity by analyzing the data curves of the knee joint angle, traction force and motion trajectory; In step S3, the PID control algorithm is used to optimize the traction parameters to meet the following optimization objectives: in, is the traction force at the next moment of the current moment t, is the traction force at the current moment t, e(t) is the traction force error, K p is the proportional control coefficient, K d is the differential control coefficient, and K is the integral control coefficient.

2. A binocular vision-based knee joint traction rehabilitation method according to claim 1, characterized in that In step S3, a dynamic mechanical model is established, including the following specific steps: Step Z1: During the traction process, the knee joint meets the mechanical balance, and the calculation formula is: F T ·cosα=F J +F R , where F T is the magnitude of the traction force, α is the angle between the traction force and the axis of the knee joint, F J is the reaction force in the knee joint, F R Resistance created by soft tissue; Step Z2: The displacement, angle and motion trajectory of the knee joint satisfy the following relationship: Δθ=θ(t)-θ0, Among them, ΔX is the total displacement of the knee joint calibration point in the X direction, is the instantaneous velocity of the calibration point, Δθ is the change in the knee joint angle, θ(t) is the angle of the knee joint at the current moment, θ0 is the angle of the knee joint during the initial calibration, and L is the total length of the knee joint motion trajectory.

3. A binocular vision-based knee joint traction rehabilitation method according to claim 1, characterized in that In step S3, the calculation formula of the knee joint angle is: Where θ is the bending angle of the knee joint, is the vector in the direction of the thigh, is the vector in the direction of the calf, express The module length, express The calculation formula of knee joint displacement is: Where D is the displacement of the knee joint calibration point, (x2, y2, z2) is the initial three-dimensional coordinate of the knee joint calibration point, (x2 ′ ,y2 ′ ,z2 ′ ) is the real-time three-dimensional coordinate of the knee joint mark; the calculation formula of the motion trajectory of the knee joint is: Where L is the total length of the knee joint motion trajectory, (x 2,i ,y 2,i ,z 2,i ) is the three-dimensional coordinate of the knee joint calibration point at the i-th time point, and n is the total number of sampling points during the movement.

4. The binocular vision-based knee joint traction rehabilitation method according to claim 1, characterized in that In step S1, the knee traction rehabilitation equipment is placed on a flat ground to ensure that the base module is firmly fixed. The base module adopts an aluminum alloy frame design and is equipped with a rubber anti-slip pad on the bottom; the bracket module has a built-in slide rail system and a locking device, with a height adjustment range of 15 to 23 inches and an angle adjustment range of 24.5° to 34.9°; the binocular vision module consists of a pair of cameras and a calibration calculation unit, and the camera position is 50 to 70 cm away from the patient's knee joint.

5. The binocular vision-based knee joint traction rehabilitation method according to claim 1, characterized in that: In step S1, the binocular vision module is calibrated through a calibration plate, image data of the calibration plate is obtained, and the intrinsic and extrinsic parameters of the camera are calculated. The intrinsic parameters include focal length, optical center position and distortion coefficient, and the extrinsic parameters include rotation matrix and baseline distance. During the calibration process, the angle between the calibration plate plane and the camera is 15° to 30°; reflective markers are attached to three positions of the patient's upper edge of the patella, lower edge of the patella and middle of the calf.

6. The binocular vision-based knee joint traction rehabilitation method according to claim 1, characterized in that: In step S2, the operator uses a fixation device to fix the patient's legs, and the fixation device includes three leg straps, which are used to fix the upper and lower parts of the knee joint and the middle part of the calf respectively; the initial traction force is set in the range of 10N to 20N.

7. The binocular vision-based knee joint traction rehabilitation method according to claim 1, characterized in that: In step S4, the key calibration points on the knee joint collected by the camera are transmitted to the intelligent control module, and the intelligent control module generates real-time charts including a knee joint angle time change curve, a traction force and displacement relationship diagram, and a three-dimensional path diagram of the motion trajectory.

8. The binocular vision-based knee joint traction rehabilitation method according to claim 1, characterized in that: The knee joint traction rehabilitation device can transmit the patient's rehabilitation data to a designated remote terminal for remote evaluation and guidance by professional doctors.

Images