Rehabilitation training monitoring system after orthopedic surgery

Abstract

The invention provides an orthopedic post-operation rehabilitation training monitoring system, which comprises a first determination module used for determining a rehabilitation training plan for a patient; the rehabilitation training plan comprises a first-stage training plan and a second-stage training plan; the transmission module is used for acquiring training images and motion sensor data of the patient and transmitting the training images and the motion sensor data to the recognition module when the patient executes the first-stage training plan; the recognition module is used for recognizing the training images and the motion sensor data and determining training features; and the display module is used for matching the training features with preset training features in the first-stage training plan, determining a training progress according to a matching result and displaying the training progress. On-line monitoring and data tracking are achieved, the training features of the patient during execution of the rehabilitation training plan are accurately recognized, then the training progress is accurately determined, and the rehabilitation condition of the patient can be comprehensively known.

Description

Technical Field

[0001] The present invention relates to the field of monitoring technology, and in particular to a post-orthopedic surgery rehabilitation training monitoring system. Background Art

[0002] Rehabilitation training refers to the patient's own exercise, massage, muscle stretching, and joint movement to restore normal limb movement in addition to medical treatment for bone and joint injuries. The purpose of rehabilitation functional exercise is to restore the inherent functions of the joints of the trunk and limbs, and prevent various changes that are not conducive to joint movement due to bleeding, such as joint contracture, ligament shortening, muscle stiffness, synovial adhesion, etc. Even if there are hematomas, edema, and ossification around the joints, the scars can be loosened and softened through activities, and no longer affect joint movement. However, there are still shortcomings in the existing monitoring technology. There is a lack of online monitoring and data tracking in the monitoring process, and the training characteristics of the patient in the implementation of the rehabilitation training plan cannot be accurately identified, and the training progress cannot be accurately determined, which is not conducive to a comprehensive understanding of the patient's rehabilitation status, and it is impossible to make reasonable supervision adjustments based on the performance of the patient's rehabilitation joints in the rehabilitation training. At the same time, it is impossible to adjust the rehabilitation training decisions based on the rehabilitation status of the current rehabilitation cycle, which is not conducive to the patient's stable recovery. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems in the above-mentioned technologies to a certain extent. To this end, the purpose of the present invention is to propose a post-orthopedic rehabilitation training monitoring system to achieve online monitoring and data tracking, accurately identify the training characteristics of patients in the implementation of rehabilitation training plans, and then accurately determine the training progress, so as to fully understand the patient's rehabilitation status, make rational supervision adjustments based on the performance of the patient's rehabilitation joints in rehabilitation training, and make rehabilitation training decision adjustments based on the rehabilitation status of the current rehabilitation cycle, so as to facilitate the patient's stable recovery.

[0004] To achieve the above objectives, the embodiment of the present invention provides a post-orthopedic surgery rehabilitation training monitoring system, comprising:

[0005] A first determination module is used to determine a rehabilitation training plan for the patient; the rehabilitation training plan includes a first-stage training plan and a second-stage training plan;

[0006] A transmission module, used to obtain the patient's training images and motion sensor data and transmit them to the recognition module when the patient performs the first stage training plan;

[0007] A recognition module, used to recognize training images and motion sensor data and determine training features;

[0008] The display module is used to match the training characteristics with the preset training characteristics in the first-stage training plan, determine the training progress according to the matching results and display it.

[0009] According to some embodiments of the present invention, the first determining module includes:

[0010] The acquisition module is used to obtain the patient's basic information and surgical information;

[0011] The matching module is used to match the basic information and surgical information in the preset database and determine the rehabilitation training plan based on the matching results.

[0012] According to some embodiments of the present invention, it further includes: a transmission verification module, which is used to calculate the connection state value between the transmission module and the recognition module during the transmission period when the transmission module transmits the patient's training image and motion sensor data to the recognition module;

[0013] a n =2t(δ 2 n +μ 2 n ) / λ n

[0014]

[0015] Wherein, P is the connection status value between the transmission module and the identification module in the transmission period t; a n is the patient's movement trajectory; δ 2 n is the variance, μ 2 n is the mean, V is the maximum moving speed of the patient; λ n is the patient's moving position; e is a natural constant; R is the communication radius of the transmission module; D is the distance between the transmission module and the recognition module; θ is the patient's moving direction, which obeys the uniform distribution of (0,2π);

[0016] The connection state value between the transmission module and the identification module in the transmission period is compared with a preset connection state value, and when it is determined that the connection state value is less than the preset connection state value, a warning message of data loss is generated.

[0017] According to some embodiments of the present invention, the identification module includes:

[0018] The second determination module is used to input the training image into a pre-trained human body recognition model to determine the key points of the face and the key points of the limbs;

[0019] A third determination module is used to determine a facial image according to facial key points, and determine facial features of the patient according to the facial image;

[0020] A fourth determination module is used to determine a limb image according to limb key points, and determine a patient's behavioral characteristics according to the limb image;

[0021] a fifth determination module, configured to determine a motion intensity characteristic of the patient based on the motion sensor data;

[0022] The training features include facial features, behavioral features and motion intensity features.

[0023] According to some embodiments of the present invention, the third determination module is used to determine geometric features, texture features and shape features of the face in the facial image based on a face recognition algorithm as facial features.

[0024] According to some embodiments of the present invention, the fourth determining module includes:

[0025] The first generation module is used to perform edge detection on the limb image, determine edge points, and generate initial contour lines;

[0026] Enhancement modules for:

[0027] Perform enhancement processing on the initial contour line to obtain the target contour line;

[0028] The patient's behavioral characteristics are determined based on the target contour lines through a posture estimation algorithm.

[0029] According to some embodiments of the present invention, the enhancement module includes:

[0030] The initial contour lines are enhanced based on a recurrent neural network;

[0031]

[0032] Where W(k) is the discrete signal of the initial contour line before enhancement processing; W N (k) is the enhanced signal of the initial contour line obtained after N iterations in the recurrent neural network, that is, the target contour line; M is the initial contour line before enhancement processing; M is the number of pixels on the initial contour line; T N-1 (k+i) is the weight at the i-th pixel at the N-1th iteration in the recurrent neural network, T N-1 (k+i)>0;W N-1 (k+i) is the enhanced signal at the i-th pixel at the N-1th iteration in the recurrent neural network.

[0033] According to some embodiments of the present invention, the fifth determining module includes:

[0034] A second generating module is used to determine a first inertial feature of the patient in a static state, and to generate a first antisymmetric matrix according to the first inertial feature;

[0035] a third generating module, configured to determine a second inertial characteristic of the patient according to the motion sensor data, and generate a second antisymmetric matrix according to the second inertial characteristic;

[0036] The calculation module is used to calculate the patient's motion intensity characteristics according to the first antisymmetric matrix and the second antisymmetric matrix.

[0037] According to some embodiments of the present invention, the method further includes: a positioning module, configured to determine and display the patient's positioning data.

[0038] According to some embodiments of the present invention, it further includes: a fourth generating module, which is used to generate prompt information for executing the second stage training plan when it is determined that the training progress is greater than a preset progress threshold.

[0039] The present invention proposes a post-orthopedic surgery rehabilitation training monitoring system to achieve online monitoring and data tracking, accurately identify the training characteristics of patients in executing rehabilitation training plans, and then accurately determine the training progress, so as to fully understand the patient's rehabilitation status, make rational supervision adjustments based on the patient's performance in rehabilitation training of the rehabilitation joints, and at the same time, make rehabilitation training decision adjustments based on the rehabilitation status of the current rehabilitation cycle, so as to facilitate the patient's stable recovery.

[0040] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0041] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0043] Figure 1 is a block diagram of a post-orthopedic surgery rehabilitation training monitoring system according to an embodiment of the present invention;

[0044] Figure 2 is a block diagram of a first determination module according to an embodiment of the present invention;

[0045] Figure 3 is a block diagram of a recognition module according to one embodiment of the present invention. DETAILED DESCRIPTION

[0046] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0047] like Figure 1 As shown, the embodiment of the present invention provides a post-orthopedic surgery rehabilitation training monitoring system, comprising:

[0048] A first determination module is used to determine a rehabilitation training plan for the patient; the rehabilitation training plan includes a first-stage training plan and a second-stage training plan;

[0049] A transmission module, used to obtain the patient's training images and motion sensor data and transmit them to the recognition module when the patient performs the first stage training plan;

[0050] A recognition module, used to recognize training images and motion sensor data and determine training features;

[0051] The display module is used to match the training characteristics with the preset training characteristics in the first-stage training plan, determine the training progress according to the matching results and display it.

[0052] The working principle of the above technical solution: the training intensity of the second-stage training plan is greater than the training intensity of the first-stage training plan. The training image is an image of the patient's human body movement during training. The motion sensor data includes acceleration, angular velocity, etc. The recognition module recognizes the training image and motion sensor data to determine the training features; the training features include the patient's posture, movement amplitude, speed, and strength. Based on the recognition module, it is convenient to accurately determine the training features by combining images and sensors. The training features extracted by the recognition module are matched with the preset training features in the first-stage training plan. According to the matching results, the patient's training progress is displayed, including completion, accuracy, areas for improvement, etc.

[0053] The beneficial effects of the above technical solution are: realizing online monitoring and data tracking, accurately identifying the training characteristics of patients in executing rehabilitation training plans, and then accurately determining the training progress, so as to fully understand the patient's rehabilitation status, and make rational supervision adjustments based on the performance of the patient's rehabilitation joints in rehabilitation training. At the same time, rehabilitation training decision adjustments are made according to the rehabilitation status of the current rehabilitation cycle, so as to facilitate the patient's stable recovery.

[0054] like Figure 2 As shown, according to some embodiments of the present invention, the first determining module includes:

[0055] The acquisition module is used to obtain the patient's basic information and surgical information;

[0056] The matching module is used to match the basic information and surgical information in the preset database and determine the rehabilitation training plan based on the matching results.

[0057] The working principle of the above technical solution: Basic information includes the patient's age, gender, height, weight, health status, past medical history, etc. The surgical information covers the type of surgery, surgical site, surgical date, post-operative recovery, etc. The main task of the matching module is to match the basic information and surgical information collected by the acquisition module in the preset database to determine the most suitable rehabilitation training plan for the patient. The preset database contains a large number of rehabilitation training plans for different types of surgeries and basic patient information. These plans are based on medical research and clinical experience and are designed to help patients recover quickly and safely. Rehabilitation training plans include specific training items, training frequency, training intensity, etc.

[0058] The beneficial effects of the above technical solution are as follows: through the collaboration of the acquisition module and the matching module, the first determination module can develop a rehabilitation training plan that is both in line with the patient's physical condition and targeted at their surgical characteristics, helping to ensure that the patient can undergo rehabilitation training in a scientific and reasonable manner, thereby improving the rehabilitation effect.

[0059] According to some embodiments of the present invention, it further includes: a transmission verification module, which is used to calculate the connection state value between the transmission module and the recognition module during the transmission period when the transmission module transmits the patient's training image and motion sensor data to the recognition module;

[0060] a n =2t(δ 2 n +μ 2 n ) / λ n

[0061]

[0062] Wherein, P is the connection status value between the transmission module and the identification module in the transmission period t; a n is the patient's movement trajectory; δ 2 n is the variance, μ 2 n is the mean, V is the maximum movement speed of the patient; λ n is the patient's moving position; e is a natural constant; R is the communication radius of the transmission module; D is the distance between the transmission module and the recognition module; θ is the patient's moving direction, which obeys the uniform distribution of (0,2π);

[0063] The connection state value between the transmission module and the identification module in the transmission period is compared with a preset connection state value, and when it is determined that the connection state value is less than the preset connection state value, a warning message of data loss is generated.

[0064] The working principle of the above technical solution is: the transmission verification module calculates and monitors the connection status value between the transmission module and the identification module during the data transmission process. This value reflects the stability and reliability of data transmission. The connection status value P is calculated, which is an integral expression that takes into account multiple factors, including the patient's movement trajectory, variance, mean, maximum movement speed, movement position, communication radius, distance between modules, and movement direction. n ,The impact of the patient’s motion trajectory on the connection state is reflected by the variance and mean,,which are calculated based on the patient’s maximum movement speed. 2 n and μ 2 n It reflects the stability and central tendency of patient movement. The two values ​​R and D determine the physical limits of data transmission. θ indicates that the patient may move in any direction. The calculated connection status value P is compared with the preset connection status value. If P is less than the preset value, it indicates that there may be a risk of instability or loss during data transmission, and the system will generate a data loss warning message. When the connection status value is lower than the preset value, the transmission verification module will trigger the warning mechanism and send a data loss warning message to the user or system administrator. This helps to detect and solve problems in a timely manner and ensure the integrity and accuracy of the data.

[0065] Beneficial effects of the above technical solution: The introduction of the transmission verification module provides important data transmission guarantee for the post-orthopedic rehabilitation training monitoring system. By calculating and monitoring the connection status value, the system can promptly discover and solve problems that may exist in the data transmission process, ensuring the accuracy and completeness of the patient training data.

[0066] like Figure 3 As shown, according to some embodiments of the present invention, the identification module includes:

[0067] The second determination module is used to input the training image into a pre-trained human body recognition model to determine the key points of the face and the key points of the limbs;

[0068] A third determination module is used to determine a facial image according to facial key points, and determine facial features of the patient according to the facial image;

[0069] A fourth determination module is used to determine a limb image according to limb key points, and determine a patient's behavioral characteristics according to the limb image;

[0070] a fifth determination module, configured to determine a motion intensity characteristic of the patient based on the motion sensor data;

[0071] The training features include facial features, behavioral features and motion intensity features.

[0072] The working principle of the above technical solution is as follows: The second determination module is responsible for inputting the training image into the pre-trained human body recognition model. This model can identify and locate the key points of the human body in the image, including the key points of the face and the key points of the limbs. The human body recognition model is a model based on deep learning, which is trained on a large amount of human image data to learn to identify and locate the key parts of the human body. The third determination module extracts the facial image from the training image according to the facial key point information provided by the second determination module to determine the facial features of the patient. The fourth determination module extracts the limb image from the training image according to the limb key point information. Then, it analyzes these images to determine the patient's behavioral characteristics, including the patient's posture, movement, range of motion, etc. They reflect the patient's actual behavior when performing rehabilitation training. The fifth determination module receives the motion sensor data from the transmission module to determine the patient's motion intensity characteristics. The motion sensor data includes information such as acceleration and angular velocity, providing real-time feedback on the patient's motion status. The motion intensity characteristics include the patient's movement speed, strength, endurance, etc. They help to evaluate the patient's training intensity and recovery.

[0073] Beneficial effects of the above technical solution: Through the collaboration of the second, third, fourth and fifth determination modules, the recognition module can determine the patient's training features, including facial features, behavioral features and exercise intensity features. These features are used to match the preset training features in the rehabilitation training plan formulated by the first determination module to evaluate the patient's training progress and effect.

[0074] According to some embodiments of the present invention, the third determination module is used to determine geometric features, texture features and shape features of the face in the facial image based on a face recognition algorithm as facial features.

[0075] The working principle of the above technical solution: Geometric features mainly focus on the relative position and proportional relationship of key points on the face. For example, the relative positions of eyes, nose, mouth and other parts, as well as the distance and angle between them. Texture features mainly focus on the texture and color information of facial skin. For example, skin smoothness, color distribution, spots, wrinkles, etc. These features help capture subtle changes in the face, such as scars left after surgery or skin improvements during recovery. Shape features mainly focus on the overall shape and contour of the face. For example, the width, length, shape of the chin, etc. of the face. These features are important for distinguishing the face shapes of different individuals. Determine the facial performance information of the patient during training based on facial features.

[0076] The beneficial effect of the above technical solution is that through the application of the third determination module, the system can accurately extract and determine the facial features of the patient.

[0077] According to some embodiments of the present invention, the fourth determining module includes:

[0078] The first generation module is used to perform edge detection on the limb image, determine edge points, and generate initial contour lines;

[0079] Enhancement modules for:

[0080] Perform enhancement processing on the initial contour line to obtain the target contour line;

[0081] The patient's behavioral characteristics are determined based on the target contour lines through a posture estimation algorithm.

[0082] The working principle of the above technical solution is as follows: the first generation module receives the limb image provided by the second determination module and performs edge detection on it. It is used to identify edge points in the image, which are the boundaries between different areas in the image. Various edge detection algorithms are used, such as Canny edge detection, Sobel operator, Prewitt operator, etc. These algorithms can identify edge points in the image and generate initial contour lines. The output of the first generation module is the initial contour lines, which represent the edges of different parts in the limb image. The enhancement module receives the initial contour lines generated by the first generation module and performs enhancement processing on them. The purpose of the enhancement processing is to make the contour lines clearer and more accurate so that the subsequent posture estimation algorithm can more accurately identify the patient's behavioral characteristics. The enhancement processing can reduce the jitter and noise of the contour lines, making them smoother and more continuous. After the enhancement processing, the enhancement module uses the posture estimation algorithm to determine the patient's behavioral characteristics. The posture estimation algorithm is an algorithm based on machine learning or deep learning technology, which can extract key posture information from the contour lines, such as joint positions, angles, etc. The output of the enhancement module is the patient's behavioral characteristics, which include joint angles, limb positions, motion trajectories, etc. They reflect the actual behavior of the patient when performing rehabilitation training.

[0083] Beneficial effects of the above technical solution: Through the collaboration of the first generation module and the enhancement module, the fourth determination module can accurately extract and determine the patient's behavioral characteristics from the limb image. These characteristics not only help to evaluate the patient's training effect, but can also be used to generate personalized training reports or suggestions, providing patients with more comprehensive and personalized rehabilitation training services.

[0084] According to some embodiments of the present invention, the enhancement module includes:

[0085] The initial contour lines are enhanced based on a recurrent neural network;

[0086]

[0087] Where W(k) is the discrete signal of the initial contour line before enhancement processing; W N (k) is the enhanced signal of the initial contour line obtained after N iterations in the recurrent neural network, that is, the target contour line; S is the initial contour line before enhancement processing; M is the number of pixels on the initial contour line; T N-1 (k+i) is the weight at the i-th pixel at the N-1th iteration in the recurrent neural network, T N-1 (k+i)>0;W N-1 (k+i) is the enhanced signal at the i-th pixel at the N-1th iteration in the recurrent neural network.

[0088] The working principle of the above technical solution is as follows: the enhancement module receives the initial contour line generated by the first generation module and applies a recurrent neural network to enhance it. Through multiple iterations, the module can gradually optimize the contour line to make it clearer and more accurate, so that the subsequent posture estimation algorithm can more accurately identify the patient's behavioral characteristics. The discrete signal W(k) of the initial contour line, where k represents the pixel index on the contour line. In each iteration, the discrete signal of the current contour line is processed using a recurrent neural network. Through multiple iterations, the enhanced signal of the contour line is gradually optimized.

[0089] The beneficial effect of the above technical solution is that by using a recurrent neural network to enhance the initial contour lines, the enhancement module can gradually optimize the accuracy and clarity of the contour lines. This processing method not only improves the recognition accuracy of the posture estimation algorithm, but also provides more reliable data support for subsequent rehabilitation training monitoring.

[0090] According to some embodiments of the present invention, the fifth determining module includes:

[0091] A second generating module is used to determine a first inertial feature of the patient in a static state, and to generate a first antisymmetric matrix according to the first inertial feature;

[0092] a third generating module, configured to determine a second inertial characteristic of the patient according to the motion sensor data, and generate a second antisymmetric matrix according to the second inertial characteristic;

[0093] The calculation module is used to calculate the patient's motion intensity characteristics according to the first antisymmetric matrix and the second antisymmetric matrix.

[0094] The working principle of the above technical solution: the static state can be the patient's resting state before rehabilitation training, or the state of remaining motionless at a specific point in time. The first inertial feature includes the patient's acceleration, angular velocity, etc., which reflects the patient's physical state in a static state. According to the first inertial feature, the second generation module generates a first antisymmetric matrix. The antisymmetric matrix is ​​a special matrix whose elements satisfy A[i][j]=-A[j][i]. This structure helps to capture the interactions and relationships between various parts of the body. The third generation module is used to determine the patient's second inertial feature based on the motion sensor data, which represents the patient's physical state during exercise. The patient's motion intensity characteristics are calculated based on the first antisymmetric matrix and the second antisymmetric matrix, including: determining the difference between the two antisymmetric matrices based on matrix subtraction;

[0095] ΔA=A2-A1

[0096] Among them, A1 is the first antisymmetric matrix; A2 is the second antisymmetric matrix; ΔA is the difference matrix between the second antisymmetric matrix and the first antisymmetric matrix.

[0097] Perform eigenvalue decomposition on the difference matrix ΔA. Eigenvalue decomposition is a mathematical method used to decompose a matrix into its eigenvalues ​​and eigenvectors. The size of the eigenvalue can reflect the "strength" or "importance" of the matrix in certain directions. Let the eigenvalues ​​of ΔA be {λ1,λ2,...,λn}, where n is the dimension of the matrix (i.e., the number of physical features). One or more exercise intensity indicators can be calculated based on the eigenvalues. These indicators include: Total eigenvalue: the sum of all eigenvalues, reflecting the change in overall exercise intensity. Maximum eigenvalue: reflects the most important exercise change. The variance of the eigenvalue is used to measure the uniformity of the distribution of exercise intensity. Determine the exercise intensity characteristics based on the exercise intensity indicators.

[0098] The beneficial effects of the above technical solution are: it is easy to accurately determine the patient's exercise intensity characteristics, and then it is easy to judge the training progress.

[0099] According to some embodiments of the present invention, the method further includes: a positioning module, configured to determine and display the patient's positioning data.

[0100] The working principle and beneficial effects of the above technical solution: The positioning module can capture the patient's location information in real time, ensuring that medical staff can understand the patient's dynamics at any time.

[0101] According to some embodiments of the present invention, it further includes: a fourth generating module, which is used to generate prompt information for executing the second stage training plan when it is determined that the training progress is greater than a preset progress threshold.

[0102] The working principle and beneficial effects of the above technical solution: The fourth generation module can continuously monitor the patient's rehabilitation training progress and set a preset progress threshold. When the patient's training progress exceeds this threshold, it means that the patient has completed the training goal of the first stage and is ready to enter the next stage of training. Once it is determined that the patient's training progress is greater than the preset progress threshold, the fourth generation module will automatically generate a prompt message to notify the patient and / or medical staff that they can start the second stage training plan. Through intelligent monitoring and prompts, the fourth generation module can help patients and medical staff to grasp the training progress more accurately, so as to adjust the training plan in time and improve the training efficiency. After the patient reaches the preset progress threshold, the system can automatically generate a second stage training plan that is more in line with the patient's current physical condition and training needs, and realize personalized training. The fourth generation module plays an important role in the post-orthopedic rehabilitation training monitoring system. It can not only intelligently monitor the patient's training progress, but also provide timely prompt information at critical moments to help patients and medical staff better conduct rehabilitation training.

[0103] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A post-orthopedic surgery rehabilitation training monitoring system, characterized in that: include: A first determination module is used to determine a rehabilitation training plan for the patient; The rehabilitation training plan includes a first-stage training plan and a second-stage training plan; A transmission module, used to obtain the patient's training images and motion sensor data and transmit them to the recognition module when the patient performs the first stage training plan; A recognition module, used to recognize training images and motion sensor data and determine training features; The display module is used to match the training characteristics with the preset training characteristics in the first-stage training plan, determine the training progress according to the matching results and display it.

2. The post-orthopedic surgery rehabilitation training monitoring system according to claim 1, characterized in that: The first determining module includes: The acquisition module is used to obtain the patient's basic information and surgical information; The matching module is used to match the basic information and surgical information in the preset database and determine the rehabilitation training plan based on the matching results.

3. The post-orthopedic surgery rehabilitation training monitoring system according to claim 1, characterized in that: Also includes: A transmission verification module, used to calculate the connection status value between the transmission module and the recognition module during the transmission period when the transmission module transmits the patient's training image and motion sensor data to the recognition module; a n =2t(δ 2 n +m 2 n ) / min n Wherein, P is the connection status value between the transmission module and the identification module in the transmission period t; a n is the patient's movement trajectory; δ 2 n is the variance, μ 2 n is the mean, V is the maximum moving speed of the patient; λ n is the patient's moving position; e is a natural constant; R is the communication radius of the transmission module; D is the distance between the transmission module and the recognition module; θ is the patient's moving direction, which obeys the uniform distribution of (0,2π); The connection state value between the transmission module and the identification module in the transmission period is compared with a preset connection state value, and when it is determined that the connection state value is less than the preset connection state value, a warning message of data loss is generated.

4. The post-orthopedic surgery rehabilitation training monitoring system according to claim 1, characterized in that: The identification module comprises: The second determination module is used to input the training image into a pre-trained human body recognition model to determine the key points of the face and the key points of the limbs; A third determination module is used to determine a facial image according to facial key points, and determine facial features of the patient according to the facial image; A fourth determination module is used to determine a limb image according to limb key points, and determine a patient's behavioral characteristics according to the limb image; a fifth determination module, configured to determine a patient's motion intensity characteristic based on the motion sensor data; The training features include facial features, behavioral features and motion intensity features.

5. The post-orthopedic surgery rehabilitation training monitoring system according to claim 4, characterized in that: The third determination module is used to determine the geometric features, texture features and shape features of the face in the face image based on the face recognition algorithm as the face features.

6. The post-orthopedic surgery rehabilitation training monitoring system according to claim 4, characterized in that: The fourth determining module includes: The first generation module is used to perform edge detection on the limb image, determine edge points, and generate initial contour lines; Enhancement modules for: Perform enhancement processing on the initial contour line to obtain the target contour line; The patient's behavioral characteristics are determined based on the target contour lines through a posture estimation algorithm.

7. The post-orthopedic surgery rehabilitation training monitoring system according to claim 6, characterized in that: The enhancement module comprises: The initial contour lines are enhanced based on a recurrent neural network; Where W(k) is the discrete signal of the initial contour line before enhancement processing; W N (k) is the enhanced signal of the initial contour line obtained after N iterations in the recurrent neural network, that is, the target contour line; S is the initial contour line before enhancement processing; M is the number of pixels on the initial contour line; T N-1 (k+i) is the weight at the i-th pixel at the N-1th iteration in the recurrent neural network, T N-1 (k+i)>0;W N-1 (k+i) is the enhanced signal at the i-th pixel at the N-1th iteration in the recurrent neural network.

8. The post-orthopedic surgery rehabilitation training monitoring system according to claim 4, characterized in that: The fifth determining module comprises: A second generating module is used to determine a first inertial feature of the patient in a static state, and to generate a first antisymmetric matrix according to the first inertial feature; a third generating module, configured to determine a second inertial characteristic of the patient according to the motion sensor data, and generate a second antisymmetric matrix according to the second inertial characteristic; The calculation module is used to calculate the patient's motion intensity characteristics according to the first antisymmetric matrix and the second antisymmetric matrix.

9. The post-orthopedic surgery rehabilitation training monitoring system according to claim 1, characterized in that: Also includes: The positioning module is used to determine the patient's positioning data and display it.

10. The post-orthopedic surgery rehabilitation training monitoring system according to claim 1, characterized in that: Also includes: The fourth generating module is used to generate prompt information for executing the second stage training plan when it is determined that the training progress is greater than a preset progress threshold.

Images