A method, device and equipment for evaluating knee joint motion function

Abstract

This invention discloses a method, device, and equipment for assessing knee joint motor function, relating to the field of smart healthcare technology. The method includes: acquiring a motion video of a user performing a preset movement; obtaining the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint based on the motion video; obtaining the three-dimensional motion posture of the target knee joint based on the two-dimensional pixel coordinates and bone segment orientation features; and obtaining the assessment result of the target knee joint's motor function based on the three-dimensional motion posture. This invention achieves low-cost, portable, automated, and standardized quantification of three-dimensional motor function, comprehensively improving the objectivity, scientific rigor, and repeatability of rehabilitation follow-up and functional assessment results. It provides real-time feedback and guidance for patients' daily rehabilitation training, effectively enhancing patient compliance with rehabilitation.

Description

Technical Field

[0001] This invention relates to the field of smart medical technology, and in particular to a method, device, and equipment for assessing knee joint movement function. Background Technology

[0002] In recent years, with the significant increase in the incidence of sports injuries and degenerative osteoarthritis, knee joint diseases have become a major public health issue affecting human quality of life. Clinical medicine has placed more stringent demands on the accurate assessment and scientific measurement of knee joint function, especially in the rehabilitation processes following knee arthroscopy, anterior and posterior cruciate ligament reconstruction, meniscus repair, and fracture surgery. Objective and consistent data recording and methods that facilitate long-term patient follow-up management have become important requirements.

[0003] However, traditional clinical subjective assessment methods rely primarily on the physician's individual clinical experience and the observation and scoring of the patient's subjective performance. These traditional methods include the physician visually assessing the patient's gait, squatting, and stair climbing, as well as manually measuring the patient's knee flexion and extension range using a goniometer. These methods suffer from unstable results and significant subjective bias. On the other hand, while existing three-dimensional motion capture technology offers high accuracy, it is expensive, requires complex equipment, and has strict site limitations, making it unsuitable for primary healthcare institutions and home rehabilitation environments. Summary of the Invention

[0004] In view of the above, the present invention aims to provide a method, apparatus and device for evaluating knee joint movement function, so as to solve the aforementioned technical problems.

[0005] The technical solution adopted in this invention is as follows:

[0006] This invention provides a method for assessing knee joint mobility, including:

[0007] Acquire motion videos of the user performing preset actions;

[0008] Based on the motion video, the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint are obtained;

[0009] Based on the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint, the three-dimensional motion posture of the target knee joint is obtained.

[0010] Based on the three-dimensional motion posture, the evaluation results of the target knee joint's motor function are obtained.

[0011] Optionally, based on the motion video, the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint are obtained, including:

[0012] The motion video is decomposed into a sequence of video frames;

[0013] The number of target knee joints is extracted from the motion video;

[0014] Based on the video frame sequence and the number of bone segments contained in the target knee joint, the two-dimensional pixel coordinates of the target knee joint are obtained;

[0015] Based on the two-dimensional pixel coordinates of the target knee joint, the bone segment direction vector of the target knee joint in the pixel plane is obtained;

[0016] The bone segment orientation features are obtained based on the bone segment orientation vector.

[0017] Optionally, based on the two-dimensional pixel coordinates of the target knee joint, the bone segment orientation vector of the target knee joint in the pixel plane is obtained, including:

[0018] Based on the two-dimensional pixel coordinates of the starting position and the ending position of each bone segment in the target knee joint, the bone segment direction vectors in the pixel plane of the target knee joint are obtained.

[0019] Optionally, based on the bone segment orientation vector, the bone segment orientation features are obtained, including:

[0020] Based on the bone segment direction vector, the planar direction angle of the bone segment is obtained;

[0021] The planar orientation angle is feature-encoded to obtain the bone segment orientation feature.

[0022] Optionally, based on the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint, the three-dimensional motion posture of the target knee joint is obtained, including:

[0023] The three-dimensional motion posture of the target knee joint includes the three-dimensional coordinate tensor of the target knee joint and the rotational posture of each bone segment;

[0024] The first feature tensor is obtained based on the two-dimensional pixel coordinates.

[0025] Based on the directional characteristics of the bone segments, a second feature tensor is obtained;

[0026] The first feature tensor and the second feature tensor are fused by a preset fusion weight to obtain a fused feature tensor;

[0027] The fused feature tensor is mapped to the target knee joint to obtain the three-dimensional coordinate tensor of the target knee joint;

[0028] Quaternion modeling is performed on the three-dimensional coordinate tensor to obtain a quaternion sequence of bone segment rotational postures.

[0029] Optionally, based on the three-dimensional motion posture, an evaluation result of the target knee joint's motor function is obtained, including:

[0030] The flexion angle of the target knee joint is obtained based on the angle between the three-dimensional coordinate tensor and the three-dimensional spatial vector of the preset reference point.

[0031] Based on the periodic changes in the flexion angle, the symmetry data of the leg is obtained;

[0032] Based on the quaternion sequence, the rotation amplitude of the bone segment is calculated to obtain knee joint rotational stability data;

[0033] Based on the flexion angle, symmetry data, and knee joint rotational stability data, the assessment results of the target knee joint's motor function are obtained.

[0034] Optionally, based on the periodic changes in the flexion angle, symmetry data of the leg is obtained, including:

[0035] Gait cycle data are obtained based on the periodic changes in the buckling angle;

[0036] Based on the gait cycle data, the symmetry data of the legs is obtained.

[0037] The present invention also provides a device for assessing knee joint movement function, comprising:

[0038] The acquisition module is used to acquire motion videos of the user performing a target action;

[0039] The processing module is used to obtain the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint based on the motion video; to obtain the three-dimensional motion posture of the target knee joint based on the two-dimensional pixel coordinates and bone segment orientation features; and to obtain the evaluation result of the motion function of the target knee joint based on the three-dimensional motion posture.

[0040] The present invention also provides a computing device, comprising: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method described above.

[0041] The present invention also provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method described above.

[0042] The above-described solution of the present invention has at least the following beneficial effects:

[0043] The above-described solution of the present invention acquires a motion video of a user performing a preset action; based on the motion video, it obtains the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint; based on the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint, it obtains the three-dimensional motion posture of the target knee joint; and based on the three-dimensional motion posture, it obtains the assessment result of the target knee joint's motor function. The solution of the present invention achieves low-cost, portable, automated, and standardized quantification of three-dimensional motor function, comprehensively improving the objectivity, scientific rigor, and repeatability of rehabilitation follow-up and functional assessment results, providing real-time feedback and guidance for patients' daily rehabilitation training, and effectively enhancing patient compliance with rehabilitation. Attached Figure Description

[0044] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described below with reference to the accompanying drawings, wherein:

[0045] Figure 1 A flowchart of a method for evaluating knee joint movement function provided in an embodiment of the present invention.

[0046] Figure 2 This is a schematic diagram of a knee joint motion function assessment device provided in an embodiment of the present invention. Detailed Implementation

[0047] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0048] This invention proposes an embodiment of a method for assessing knee joint motor function, specifically, as follows: Figure 1 As shown, it includes:

[0049] Step 11: Obtain motion videos of the user performing preset actions.

[0050] In this embodiment, users can record videos of standardized movements such as walking in a straight line or squatting in hospitals, communities, or homes using video recording devices such as smartphones or cameras. The videos must cover at least three complete movement cycles, such as gait or squatting.

[0051] Step 12: Based on the motion video, obtain the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint.

[0052] In this embodiment, after acquiring motion video, the motion video is decomposed into a sequence of video frames, and the image clarity, joint visibility, and motion integrity are detected in real time. If problems such as occlusion, out-of-focus, or missing the target are found, the user is prompted to re-acquire the video to ensure that high-quality and standardized raw input is provided for subsequent analysis.

[0053] Using 2D human keypoint detection algorithms such as OpenPose, the video frame sequence was identified to determine the 2D pixel coordinates, number of joints, and number of bone segments in each frame of the major lower limb joints, including the hip, knee, and ankle. The 2D pixel coordinates were then converted into tensor representations.

[0054] ;

[0055] X= ;

[0056] in, X represents the two-dimensional pixel coordinates; T represents the number of video frames; J represents the number of joints; and t represents the current video frame. ;j represents the current joint, j .

[0057] Furthermore, for the two-dimensional pixel coordinates of a bone segment of a joint in any video frame, through mapping... , where represent the start and end points of each bone segment, and B is the number of bone segments;

[0058] Specifically, for any bone segment The coordinates of the starting point of the bone segment are:

[0059] ;

[0060] The coordinates of the end point of the bone segment are:

[0061] ;

[0062] The planar direction vector of bone segment b in frame t is defined as:

[0063] ;

[0064] The planar rotation angle of bone segment b in frame t is defined as follows:

[0065] ;

[0066] And the plane rotation angle is encoded into a direction feature vector using cosine-sine coding:

[0067] ;

[0068] Under the above directional feature vector representation, using As a characteristic of bone segment orientation, it avoids angularity. The periodicity of the periodicity leads to discontinuity problems.

[0069] Step 13: Based on the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint, obtain the three-dimensional motion posture of the target knee joint.

[0070] In this embodiment, the 3D pose reconstruction of the target knee joint is performed using 2D pixel coordinates and bone segment orientation features. This embodiment is implemented based on the 3D pose reconstruction algorithm of the MotionAGFormerSports architecture. This 3D pose reconstruction algorithm adopts a two-stream deep neural network structure, integrating the Transformer spatiotemporal branch and the GCN skeleton topology branch. The Transformer branch, through a spatiotemporal multi-head self-attention mechanism, can capture long-distance global dependencies in the spatial and temporal dimensions, effectively modeling the overall rhythm, periodic changes, and dynamic relationships between keyframes during the movement process. The GCN branch constructs a skeleton graph based on the human anatomical structure, performs local feature aggregation and modeling on spatially adjacent nodes (such as parent-child joints, left-right symmetrical joints), explores the anatomical coordination and dynamic constraints between joints, and utilizes left-right symmetrical sampling to improve the model's generalization performance for unseen movement types.

[0071] Inputting the two-dimensional pixel coordinates and bone segment orientation features into the three-dimensional pose reconstruction algorithm, the Transformer spatiotemporal branch outputs the spatiotemporal feature tensor as follows:

[0072] ;

[0073] The skeleton topology tensor output by the GCN skeleton topology branch is:

[0074] ;

[0075] in, , These represent the channel dimensions of the output features of the two branches.

[0076] To achieve adaptive fusion of features from the two branches, learnable fusion weights are introduced:

[0077] ;

[0078] ;

[0079] The overall characteristics after fusion are denoted as:

[0080] ;

[0081] Based on this, the three-dimensional coordinate tensors of each joint in each frame are obtained through subsequent prediction head (regression network) mapping:

[0082] ;

[0083] in, For each frame, the three-dimensional coordinate tensor of each joint; , representing the predicted 3D spatial coordinates of the j-th joint in the t-th frame.

[0084] Meanwhile, the three-dimensional spatial orientation of the bone segments is modeled using quaternions, and the rotational attitude of the b-th bone segment in frame t is represented by a unit quaternion as follows:

[0085] , ;

[0086] in, The unit quaternion represents the rotational orientation of the bone segment. , , , These are elements of a quaternion.

[0087] Next, convert the quaternion sequence of all bone segments into tensor form:

[0088] .

[0089] Step 14: Based on the three-dimensional motion posture, obtain the evaluation result of the target knee joint's motion function.

[0090] In this embodiment, based on the aforementioned three-dimensional motion posture reconstruction results and bone segment quaternion sequences, a quantitative medical analysis of the knee joint functional status is automatically performed.

[0091] After 3D pose reconstruction, the knee flexion angle is calculated for each video frame based on the positions of the hip, knee, and ankle. The calculation is performed using the formula for the angle between 3D spatial vectors.

[0092] ;

[0093] in, This refers to the flexion angle of the knee joint. Let be the three-dimensional spatial coordinate vector of the hip joint of the target leg in frame t. Let be the three-dimensional spatial coordinate vector of the knee joint of the target leg in frame t. Let be the three-dimensional spatial coordinate vector of the ankle joint of the target leg in frame t.

[0094] All of these originate from the three-dimensional coordinate tensors obtained in step 13 through the three-dimensional pose reconstruction algorithm. t represents the video frame number, and j represents the joint number. This represents the three-dimensional spatial coordinates of the j-th joint in frame t.

[0095] When calculating the knee flexion angle: vector , representing the femoral direction vector (knee → hip), vector , representing the tibial direction vector (knee → ankle) and the knee flexion angle. The angle between the two three-dimensional vectors mentioned above reflects the spatial angle between the femoral axis and the tibial axis, and is used to quantify the knee joint's flexion-extension range of motion. When the knee joint flexion increases, it indicates an increase in the degree of knee flexion; when... When it approaches 180°, it means that it is close to being fully extended.

[0096] By continuously calculating the flexion angle for each frame, a complete flexion-extension curve is output, and the range of motion, such as the maximum flexion angle and the maximum extension angle, is automatically calculated, providing a quantitative basis for rehabilitation activity.

[0097] Furthermore, by analyzing the periodic changes in the buckling angle over time, the gait cycle can be automatically detected. And the symmetry of the left and right legs. The gait cycle here... The time interval between two adjacent steps, in seconds, is determined by the frequency of knee flexion angles in both legs.

[0098] By analyzing gait cycles Get the number to get the step frequency :

[0099] ;

[0100] The gait cycles of the left and right legs are recorded separately as follows: and The corresponding step frequency is and .

[0101] By combining statistics such as stride length and phase difference of the left and right legs, we can further construct left and right gait symmetry indices (such as cycle ratio, absolute value of phase difference, etc.) to quantitatively assess gait symmetry and abnormalities.

[0102] By statistically comparing the cycles, phase differences, and stride lengths of the left and right legs, gait abnormalities, hemiplegia, or compensatory phenomena can be identified, and the symmetry of the left and right gait can be quantitatively output.

[0103] Furthermore, based on the bone segment quaternion sequence, the spatial rotation of the tibia relative to the femur is extracted, the rotation amplitude and standard deviation are calculated, and knee joint rotational stability data are obtained.

[0104] N video frames are acquired within a detection period. The relative rotation angle of the tibia with respect to the femur at the k-th moment is denoted as:

[0105] ;

[0106] The sample mean of this sequence is:

[0107] ;

[0108] The corresponding standard deviation of rotational stability (degree of fluctuation in rotational amplitude) is defined as follows:

[0109] ;

[0110] in, For rotational stability index, The relative rotation angle between the tibia and femur along the selected axis is calculated using a sequence of bone segment quaternions.

[0111] The larger the value, the greater the fluctuation in the rotation angle, which means the worse the joint rotational stability, suggesting possible abnormalities in joint function or neuromuscular control.

[0112] The tibia-femur relative rotation angle along the selected axis, obtained by converting bone segment quaternion sequences, is described in the following process:

[0113] Define the postural quaternions of the femur and tibia:

[0114] Let the unit pose quaternion of the femur at time k be:

[0115] ;

[0116] The unit posture quaternion of the tibia is:

[0117] ;

[0118] The above quaternions are all unit quaternions, satisfying .

[0119] Furthermore, calculate the relative rotation quaternion of the tibia relative to the femur:

[0120] The relative rotation quaternion of the tibia relative to the femur is defined as:

[0121] ;

[0122] in, To represent quaternion multiplication, It is the inverse quaternion of the femoral posture quaternion.

[0123] Since the inverse of a unit quaternion is equal to its conjugate:

[0124] ;

[0125] From this, we can obtain the relative rotation quaternion:

[0126] ;

[0127] Converting relative quaternions to rotation axis-angle representation, any unit quaternion can be represented as:

[0128] q= ;

[0129] in, The total rotation angle is... The unit rotation axis vector.

[0130] Therefore, we can conclude that The direction of the rotation axis is: .

[0131] Furthermore, extract the rotational component along a specified axis (such as an internal or external rotation axis):

[0132] In the local coordinate system of the knee joint, define the following: Z-axis as the internal / external rotation axis, X-axis as the flexion / extension axis, and Y-axis as the internal / external rotation axis. Let the unit internal / external rotation axis vector be... The rotation angle of the tibia relative to the femur in this axis is defined as:

[0133] ;

[0134] Wherein, "·" represents the vector dot product, indicating the projection component of the total rotation angle onto the specified axis.

[0135] In another implementation, the relative quaternions can be converted to Euler angles (e.g., ZXY order), and the corresponding axial angles can be directly taken, for example:

[0136] If the Z-axis is used as the internal / external rotation axis, then:

[0137] ;

[0138] This angle is the internal and external rotation angle of the tibia relative to the femur.

[0139] In summary, the knee joint's motor function was comprehensively evaluated based on flexion angle, symmetry data, and knee joint rotational stability data.

[0140] Finally, a structured assessment report is generated based on all quantitative analysis results, including core indicators such as flexion angle-time curves, gait cycles, and rotational stability, presented in charts and text. The report supports automatic archiving and remote follow-up within the hospital system, enabling doctors to quickly understand rehabilitation progress, identify functional abnormalities, and assist in developing precise and scientific rehabilitation plans, significantly improving the intelligence and informatization level of rehabilitation management.

[0141] The knee joint motor function assessment method in the above embodiments of the present invention achieves low-cost, portable, automated and standardized three-dimensional motor function quantification, comprehensively improves the objectivity, scientificity and repeatability of rehabilitation follow-up and functional assessment results, provides real-time feedback and guidance for patients' daily rehabilitation training, and effectively enhances patients' compliance with rehabilitation.

[0142] like Figure 2 As shown, the present invention also provides a knee joint movement function assessment device 20, comprising:

[0143] Acquisition module 21 is used to acquire motion videos of the user performing a target action;

[0144] The processing module 22 is used to obtain the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint based on the motion video; to obtain the three-dimensional motion posture of the target knee joint based on the two-dimensional pixel coordinates and bone segment orientation features; and to obtain the evaluation result of the motion function of the target knee joint based on the three-dimensional motion posture.

[0145] Optionally, based on the motion video, the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint are obtained, including:

[0146] The motion video is decomposed into a sequence of video frames;

[0147] The number of target knee joints is extracted from the motion video;

[0148] Based on the video frame sequence and the number of bone segments contained in the target knee joint, the two-dimensional pixel coordinates of the target knee joint are obtained;

[0149] Based on the two-dimensional pixel coordinates of the target knee joint, the bone segment direction vector of the target knee joint in the pixel plane is obtained;

[0150] The bone segment orientation features are obtained based on the bone segment orientation vector.

[0151] Optionally, based on the two-dimensional pixel coordinates of the target knee joint, the bone segment orientation vector of the target knee joint in the pixel plane is obtained, including:

[0152] Based on the two-dimensional pixel coordinates of the starting position and the ending position of each bone segment in the target knee joint, the bone segment direction vectors in the pixel plane of the target knee joint are obtained.

[0153] Optionally, based on the bone segment orientation vector, the bone segment orientation features are obtained, including:

[0154] Based on the bone segment direction vector, the planar direction angle of the bone segment is obtained;

[0155] The planar orientation angle is feature-encoded to obtain the bone segment orientation feature.

[0156] Optionally, based on the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint, the three-dimensional motion posture of the target knee joint is obtained, including:

[0157] The three-dimensional motion posture of the target knee joint includes the three-dimensional coordinate tensor of the target knee joint and the rotational posture of each bone segment;

[0158] The first feature tensor is obtained based on the two-dimensional pixel coordinates.

[0159] Based on the directional characteristics of the bone segments, a second feature tensor is obtained;

[0160] The first feature tensor and the second feature tensor are fused by a preset fusion weight to obtain a fused feature tensor;

[0161] The fused feature tensor is mapped to the target knee joint to obtain the three-dimensional coordinate tensor of the target knee joint;

[0162] Quaternion modeling is performed on the three-dimensional coordinate tensor to obtain a quaternion sequence of bone segment rotational postures.

[0163] Optionally, based on the three-dimensional motion posture, an evaluation result of the target knee joint's motor function is obtained, including:

[0164] The flexion angle of the target knee joint is obtained based on the angle between the three-dimensional coordinate tensor and the three-dimensional spatial vector of the preset reference point.

[0165] Based on the periodic changes in the flexion angle, the symmetry data of the leg is obtained;

[0166] Based on the quaternion sequence, the rotation amplitude of the bone segment is calculated to obtain knee joint rotational stability data;

[0167] Based on the flexion angle, symmetry data, and knee joint rotational stability data, the assessment results of the target knee joint's motor function are obtained.

[0168] Optionally, based on the periodic changes in the flexion angle, symmetry data of the leg is obtained, including:

[0169] Gait cycle data are obtained based on the periodic changes in the buckling angle;

[0170] Based on the gait cycle data, the symmetry data of the legs is obtained.

[0171] It should be noted that this device is the same as the method described above. All implementations in the above method embodiments are applicable to the embodiments of this device and can achieve the same technical effect.

[0172] An embodiment of the present invention also provides a computing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method described in the above embodiments. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0173] In this embodiment of the invention, a computer-readable storage medium is also provided, storing instructions that, when executed on a computer, cause the computer to perform the method described in the above embodiments. All implementations of the methods described in the above embodiments are applicable to this embodiment and can achieve the same technical effect.

[0174] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0175] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0176] In the embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0177] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0178] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0179] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0180] Furthermore, it should be noted that in the apparatus and method of the present invention, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent solutions of the present invention. Moreover, the steps performing the above series of processes can naturally be executed in the order described, but are not necessarily required to be executed in chronological order; some steps can be executed in parallel or independently of each other. Those skilled in the art will understand that all or any step or component of the method and apparatus of the present invention can be implemented in any computing device (including processors, storage media, etc.) or network of computing devices, in hardware, firmware, software, or a combination thereof. This is something that those skilled in the art can achieve by using their basic programming skills after reading the description of the present invention.

[0181] Therefore, the object of the present invention can also be achieved by running a program or a set of programs on any computing device. The computing device can be a known general-purpose device. Therefore, the object of the present invention can also be achieved simply by providing a program product containing program code implementing the method or apparatus. That is, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any known storage medium or any storage medium developed in the future. It should also be noted that in the apparatus and method of the present invention, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent to the present invention. Furthermore, the steps performing the above series of processes can naturally be performed in the order described, but are not necessarily required to be performed in chronological order. Some steps can be performed in parallel or independently of each other.

[0182] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for assessing knee joint motor function, characterized in that, include: Acquire motion videos of the user performing preset actions; Based on the motion video, the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint are obtained; Based on the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint, the three-dimensional motion posture of the target knee joint is obtained. Based on the three-dimensional motion posture, the evaluation results of the target knee joint's motor function are obtained.

2. The method for assessing knee joint function according to claim 1, characterized in that, Based on the motion video, the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint are obtained, including: The motion video is decomposed into a sequence of video frames; The number of target knee joints is extracted from the motion video; Based on the video frame sequence and the number of bone segments contained in the target knee joint, the two-dimensional pixel coordinates of the target knee joint are obtained; Based on the two-dimensional pixel coordinates of the target knee joint, the bone segment direction vector of the target knee joint in the pixel plane is obtained; The bone segment orientation features are obtained based on the bone segment orientation vector.

3. The method for assessing knee joint function according to claim 2, characterized in that, Based on the two-dimensional pixel coordinates of the target knee joint, the bone segment orientation vector of the target knee joint in the pixel plane is obtained, including: Based on the two-dimensional pixel coordinates of the starting position and the ending position of each bone segment in the target knee joint, the bone segment direction vectors in the pixel plane of the target knee joint are obtained.

4. The method for assessing knee joint function according to claim 2, characterized in that, Based on the bone segment orientation vector, the bone segment orientation features are obtained, including: Based on the bone segment direction vector, the planar direction angle of the bone segment is obtained; The planar orientation angle is feature-encoded to obtain the bone segment orientation feature.

5. The method for assessing knee joint function according to claim 1, characterized in that, Based on the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint, the three-dimensional motion posture of the target knee joint is obtained, including: The three-dimensional motion posture of the target knee joint includes the three-dimensional coordinate tensor of the target knee joint and the rotational posture of each bone segment; The first feature tensor is obtained based on the two-dimensional pixel coordinates. Based on the directional characteristics of the bone segments, a second feature tensor is obtained; The first feature tensor and the second feature tensor are fused by a preset fusion weight to obtain a fused feature tensor; The fused feature tensor is mapped to the target knee joint to obtain the three-dimensional coordinate tensor of the target knee joint; Quaternion modeling is performed on the three-dimensional coordinate tensor to obtain a quaternion sequence of bone segment rotational postures.

6. The method for assessing knee joint function according to claim 5, characterized in that, Based on the three-dimensional motion posture, the assessment results of the target knee joint's motor function are obtained, including: The flexion angle of the target knee joint is obtained based on the angle between the three-dimensional coordinate tensor and the three-dimensional spatial vector of the preset reference point. Based on the periodic changes in the flexion angle, the symmetry data of the leg is obtained; Based on the quaternion sequence, the rotation amplitude of the bone segment is calculated to obtain knee joint rotational stability data; Based on the flexion angle, symmetry data, and knee joint rotational stability data, the assessment results of the target knee joint's motor function are obtained.

7. The method for assessing knee joint function according to claim 6, characterized in that, Based on the periodic changes in the flexion angle, the symmetry data of the leg is obtained, including: Gait cycle data are obtained based on the periodic changes in the buckling angle; Based on the gait cycle data, the symmetry data of the legs is obtained.

8. A device for assessing knee joint movement function, characterized in that, include: The acquisition module is used to acquire motion videos of the user performing a target action; The processing module is used to obtain the two-dimensional pixel coordinates and bone segment orientation features of the target knee joint based on the motion video; to obtain the three-dimensional motion posture of the target knee joint based on the two-dimensional pixel coordinates and bone segment orientation features; and to obtain the evaluation result of the motion function of the target knee joint based on the three-dimensional motion posture.

9. A computing device, characterized in that, include: A processor, a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The system stores instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1 to 7.

Images