Joint injury prevention method, device, medium and product for rehabilitation training

Abstract

The invention discloses a joint injury prevention method and device for rehabilitation training, a medium and a product. The method comprises the following steps: in response to a training trigger signal for a predetermined joint part of a target object, controlling a predetermined terminal to display a training video corresponding to a rehabilitation scheme, and obtaining a predetermined first feature threshold combination for the rehabilitation scheme; in the process that the target object follows the training video to conduct rehabilitation training, actual motion data for the preset joint part are obtained through a sensor in the training robot; determining an actual feature data combination corresponding to the actual motion data; and if the actual feature data combination exceeds the first feature threshold combination, stopping playing the training video, and outputting a passive training signal to the training robot, so that the training robot is converted from the current non-passive training state to the passive training state. According to the technical scheme provided by the embodiment of the invention, the problem of joint injury of the target object in the rehabilitation training process can be avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of rehabilitation training, and in particular to a joint injury prevention method, device, medium and product for rehabilitation training. BACKGROUND

[0002] Stroke, also known as apoplexy, is a series of symptoms caused by the rupture of cerebral blood vessels or thrombosis, which leads to the interruption of blood supply to the brain. The incidence of stroke is increasing year by year, which brings great burden and pressure to patients and families. However, through stroke rehabilitation training, patients can achieve a certain degree of rehabilitation, improve the quality of life, and reduce related complications, which makes the importance of rehabilitation training increasingly valued by people.

[0003] However, during the rehabilitation training process, due to non-standard movements and unequal force, the overall training effect will be greatly reduced. Therefore, it is necessary to provide a rehabilitation training method to improve the overall effect of rehabilitation training. SUMMARY

[0004] The present application provides a joint injury prevention method, device, medium and product for rehabilitation training to solve the problem that the existing rehabilitation training method cannot guarantee to improve the training effect.

[0005] According to an aspect of the present application, a joint injury prevention method for rehabilitation training is provided, which is suitable for a target object wearing a training robot, comprising:

[0006] In response to a training trigger signal for a predetermined joint position of the target object, a predetermined terminal displays a training video corresponding to a rehabilitation scheme, and a first feature threshold combination for the rehabilitation scheme is obtained in advance;

[0007] During the rehabilitation training of the target object following the training video, the actual motion data for the predetermined joint position is obtained through the sensor in the training robot;

[0008] The actual feature data combination corresponding to the actual motion data is determined;

[0009] If the actual feature data combination exceeds the first feature threshold combination, the training video is stopped playing, and a passive training signal is output to the training robot, so that the training robot is converted from the current non-passive training state to the passive training state.

[0010] According to another aspect of the present application, a joint injury prevention device for rehabilitation training is provided, which is suitable for a target object wearing a training robot, comprising:

[0011] In response to a training trigger signal for a predetermined joint position of a target object, a terminal device displays a training video corresponding to a rehabilitation scheme, and acquires a first feature threshold combination for the rehabilitation scheme;

[0012] During rehabilitation training of the target object following the training video, actual motion data for the predetermined joint position is acquired by a sensor in the training robot;

[0013] An actual feature data combination corresponding to the actual motion data is determined;

[0014] If the actual feature data combination exceeds the first feature threshold combination, the training video is stopped, and a passive training signal is output to the training robot to switch the training robot from a current non-passive training state to a passive training state.

[0015] According to another aspect of the present application, an electronic device is provided, the electronic device comprising:

[0016] at least one processor; and

[0017] a memory connected to the at least one processor in communication; wherein

[0018] the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the joint injury prevention method for rehabilitation training according to any one of the embodiments of the present application.

[0019] According to another aspect of the present application, a computer readable storage medium is provided, the computer readable storage medium stores computer instructions for enabling a processor to implement the joint injury prevention method for rehabilitation training according to any one of the embodiments of the present application when executed by the processor.

[0020] According to another aspect of the present application, a computer program product is provided, the computer program product comprises a computer program, and the computer program implements the joint injury prevention method for rehabilitation training according to any one of the embodiments when executed by a processor.

[0021] The technical scheme provided by the embodiment of the present application, since the first feature threshold combination corresponds to the predetermined joint part of the target object, that is, it is an individualized threshold combination; during the participation of the robot in the rehabilitation training of the target object, one or more sensors built-in the robot will collect actual motion data of the predetermined joint part in real time, therefore, the size relationship between the real-time feature combination corresponding to the actual motion data and the first feature threshold combination can determine whether the predetermined part of the target object has a risk, and in the case that there is a risk, stop playing the training, and control the training robot to switch from the non-passive training state to the passive motion state, so as to protect the predetermined joint part of the target object from the physical level during the end of the training, and achieve the technical effect of specifying an individualized rehabilitation training scheme for the target object.

[0022] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0024] Figure 1 is a flowchart of a joint injury prevention method for rehabilitation training provided according to the embodiment of the present application;

[0025] Figure 2 is another flowchart of a joint injury prevention method for rehabilitation training provided according to the embodiment of the present application;

[0026] Figure 3A is a structural schematic diagram of a joint injury prevention device for rehabilitation training provided according to the embodiment of the present application;

[0027] Figure 3B is another structural schematic diagram of a joint injury prevention device for rehabilitation training provided according to the embodiment of the present application;

[0028] Figure 4 is a structural schematic diagram of an electronic device for implementing the joint injury prevention method for rehabilitation training according to the embodiment of the present application. DETAILED DESCRIPTION

[0029] In the following, the technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of embodiments of the present application, rather than all embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work should fall within the protection scope of the present application.

[0030] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0031] Figure 1 The flowchart of the joint injury prevention method for rehabilitation training provided by the embodiment of the present application can be applied to the case where the active training of the user is changed to passive training when it is found that the target object has a higher possibility of joint injury in the process of analyzing the rehabilitation training data of the target object in real time. The method can be executed by a joint injury prevention device for rehabilitation training, which can be realized in the form of hardware and / or software, and can be configured in an electronic device. As shown in the figure, the method comprises: Figure 1

[0032] S110, in response to a training trigger signal for a predetermined joint part of a target object, controlling a predetermined terminal to display a training video corresponding to a rehabilitation scheme, and obtaining a first feature threshold combination predetermined for the rehabilitation scheme.

[0033] The target object is a patient in need of rehabilitation training.

[0034] ​The predetermined joint part is a designated joint part, such as a shoulder joint, a knee joint, an elbow joint, etc. During the rehabilitation training process, the target object may have a risk of joint injury due to inaccurate posture, for example, for the shoulder joint, repetitive overhead motion (such as throwing, weight lifting) will cause the supraspinatus and infraspinatus tendons to be overloaded, and compensatory motion (such as shrugging shoulders) will increase tendon friction; for the elbow joint, overextension of the elbow joint (pushing action) during rehabilitation training will cause joint capsule or ligament injury. For the elbow joint, overextension of the elbow joint (pushing action) may cause joint capsule or ligament injury, and pressure concentrated on the lateral side of the palm may indicate compensatory hyperextension; for the knee joint, the knee joint bears high load when the knee joint is flexed less than 30 degrees (such as landing after jumping, sudden stop), and high repetitive load (such as climbing stairs at too fast a speed).

[0035] The predetermined terminal can be a general display terminal or a virtual reality device.

[0036] The rehabilitation training of the target object needs to be performed according to the rehabilitation plan formulated by the doctor. Specifically, the doctor formulates the rehabilitation plan according to the patient's file information, such as basic physical characteristics, disease history, pain history, contraindications, maximum range of motion of the predetermined joint part, etc.

[0037] The first feature threshold combination is a feature data combination corresponding to the rehabilitation plan. The first feature threshold combination corresponding to different joint parts is different.

[0038] The training video corresponds to the rehabilitation plan, which is a video for guiding the user to perform rehabilitation training. Specifically, the training action corresponding to the rehabilitation plan, such as walking, squatting, lifting legs, etc. Both of them need to be pre-stored in a designated location.

[0039] In one embodiment, the action displayed in the training video is a standard action corresponding to the action identification of the rehabilitation plan. This embodiment is suitable for the action display in the training video being a real person, and the displayed action being natural and real.

[0040] In one embodiment, the attribute data of the action displayed in the training video is consistent with the attribute data of the action corresponding to the rehabilitation plan, such as the identification of the action, the amplitude of the action, the direction of the action, etc. This embodiment is suitable for the case where the training video required based on the Ai technology and the rehabilitation plan is generated. This embodiment can intuitively display the content of the rehabilitation plan.

[0041] S120, in the process that the target object follows the training video to perform rehabilitation training, the actual motion data for the predetermined joint part is acquired through the sensor in the training robot.

[0042] The actual motion data includes joint motion angle and motion speed. Taking the shoulder joint as an example, the joint motion angle includes forward flexion, abduction, external rotation, internal rotation, etc.; taking the elbow joint as an example, the joint motion angle includes flexion, extension, pronation, supination; taking the knee joint as an example, the joint motion angle includes flexion, extension.

[0043] For the shoulder joint, the abduction angle and the external rotation speed of the shoulder joint are collected by the inertial sensor, and the activation imbalance time of the supraspinatus muscle and the deltoid muscle is collected by the electromyography sensor. For the elbow joint, the inertial sensor located at the distal end of the upper arm and the proximal end of the forearm collects the elbow joint flexion angle, angular velocity and acceleration in real time, and the flexible pressure sensor located on the palm contact surface and the back of the hand collects the hand pressure distribution (palm / dorsal pressure ratio); for the knee joint, the inertial sensor is used to collect the knee joint flexion angle, angular velocity and acceleration, and the inertial sensor is worn at the distal end of the thigh and the proximal end of the tibia; the plantar pressure distribution and the pressure center trajectory are obtained by the plantar pressure pad; the electromyography sensor is configured on the surface of the muscle to collect the electromyography data of the quadriceps femoris (rectus femoris, vastus lateralis) and the activation intensity and cooperativity data of the hamstrings.

[0044] S130, determining the actual feature data combination corresponding to the actual motion data.

[0045] After the actual motion data is determined, the required features are extracted from the actual motion data to obtain the actual feature data combination.

[0046] For example, for the elbow joint, a fourth-order Butterworth low-pass filter (cutoff frequency 10 Hz) is used to eliminate high-frequency noise, and the elbow joint flexion angle is calculated based on the quaternion, with an error of 2%, and the formula is as follows:

[0047] Where q0 is the real part, corresponding to the cosine half angle of the rotation angle, i.e. θ is the angle of rotation around the axis, q1 is the coefficient of the imaginary part i, corresponding to the product of the component of the rotation axis in the x direction and the sine half angle, i.e. n x is the component of the rotation axis unit vector on the x axis.

[0048] The palm and dorsal pressure ratio is:

[0049] N is not zero, P palm is the palm pressure, P dorsum is the dorsal pressure, and ∈ is a regularization term to avoid a small amount of denominator being zero, defined in the formula as N is a non-zero parameter, ensuring that ∈>0.

[0050] For the knee joint, a fourth-order Butterworth low-pass filter (cutoff frequency 10 Hz) was used to remove high-frequency noise, and the knee flexion angle was calculated based on the quaternion solution with an error of 2%. The standard deviation of the foot pressure center trajectory and the forefoot / hindfoot pressure ratio were calculated.

[0051] For the shoulder joint, the actual feature data combination includes shoulder abduction angle, external rotation speed, and supraspinatus and deltoid activation imbalance data. For the elbow joint, the actual feature data combination includes at least one of elbow hyperextension angle, hand back pressure ratio, and angular velocity peak. For the knee joint, the actual feature data combination includes flexion angle, vertical ground reaction force peak, vastus lateralis / vastus medialis activation ratio, and angular velocity.

[0052] S140, if the actual feature data combination exceeds the first feature threshold combination, stop playing the training video, and output a passive training signal to the training robot to make the training robot switch from the current non-passive training state to the passive training state.

[0053] For example, if the shoulder abduction is greater than or equal to the first abduction threshold, and the feature threshold is 1-3 minutes of cumulative abduction, it is determined that the shoulder joint of the target object has a first injury risk. For example, if the elbow hyperextension angle is greater than or equal to the first elbow hyperextension angle threshold, or the hand back pressure ratio is less than 0.3, it is determined that the target object's elbow joint has a first injury risk. For example, if the knee flexion angle is less than the first knee flexion angle threshold, and the angular velocity is greater than the first knee angular velocity threshold, it is determined that the target object's knee joint has a first injury risk.

[0054] Regardless of the joint, once it is determined that it has a first risk, the training video is stopped, and a passive training signal is output to the training robot to make the training robot switch from the current active training state to the passive training state, or from the current non-assisted state to the passive training state, to physically protect the joint of the target object.

[0055] The technical scheme provided by the embodiment of the present application, since the first feature threshold combination corresponds to the predetermined joint part of the target object, that is, it is a personalized threshold combination; during the participation of the robot in the rehabilitation training of the target object, one or more sensors built-in the robot collect actual motion data of the predetermined joint part in real time, therefore, the size relationship between the real-time feature combination corresponding to the actual motion data and the first feature threshold combination can determine whether the predetermined part of the target object has a risk, and in the case of a risk, the training is stopped, and the robot is controlled to switch from the non-passive training state to the passive motion state, so as to protect the predetermined joint part of the target object from the physical level during the end of the training, and the technical effect of specifying a personalized rehabilitation training scheme for the target object is achieved.

[0056] On the basis of the foregoing embodiment, while the first feature threshold combination for the rehabilitation scheme is acquired, a second feature threshold combination and / or a third feature threshold combination for the rehabilitation scheme are also acquired; if the actual feature data combination exceeds the second feature data combination, the training task difficulty is reduced, and an increased resistance signal is output to the training robot, so that the training robot increases the resistance to the predetermined joint part according to the increased resistance signal; if the actual feature data combination exceeds the third feature threshold combination, prompt information is output; wherein each feature threshold in the second feature threshold combination is smaller than the corresponding feature threshold in the first feature threshold combination, and each feature threshold in the third feature threshold combination is smaller than the corresponding feature threshold in the second feature threshold combination.

[0057] Specifically, the second feature threshold combination is a feature threshold combination corresponding to a secondary risk, if the actual feature data combination exceeds the second feature threshold combination, it is determined that the predetermined joint part of the target object has a secondary damage risk, and the training task difficulty is reduced, and an increased resistance signal is output to the training robot, so that the training robot increases the resistance to the predetermined joint part according to the increased resistance signal; the third feature threshold combination is a feature threshold combination corresponding to a tertiary risk, if the actual feature data combination exceeds the third feature threshold combination, it is determined that the predetermined joint part of the target object has a tertiary damage risk, and prompt information is output, such as at least one of voice prompt information, character prompt information or pattern prompt information. Since each feature threshold in the second feature threshold combination is smaller than the corresponding feature threshold in the first feature threshold combination, and each feature threshold in the third feature threshold combination is smaller than the corresponding feature threshold in the second feature threshold combination, the risk degree of the first risk is higher than that of the secondary risk, and the risk degree of the secondary risk is higher than that of the tertiary risk. Through the risk control of different levels, the embodiment achieves the technical effects of early detection and early prevention for low risk, and timely detection and timely intervention for medium risk.

[0058] The third-level risk prompt information is "pay attention to the action trajectory" for the shoulder joint, and is "reduce squatting depth", knee pad micro-vibration, etc. for the knee joint. The first-level risk. For example, after the training robot is converted from the active mode to the passive mode after the training is paused, a correction video is played to make the target object's action more standard and safer.

[0059] Figure 2 Another flowchart of the joint injury prevention method for rehabilitation training provided by the embodiment of the present application adds the determination of the target state probability of the predetermined joint position on the basis of the above-mentioned embodiment. As shown in the figure, the method comprises the following steps. Figure 2

[0060] S210, in response to the training trigger signal for the target object's predetermined joint position, the predetermined terminal displays the training video corresponding to the rehabilitation scheme, and the first feature threshold combination for the rehabilitation scheme is obtained in advance.

[0061] S220, in the process of the target object following the training video for rehabilitation training, the actual motion data for the predetermined joint position is obtained through the sensor in the training robot.

[0062] S230, the actual feature data combination corresponding to the actual motion data is determined.

[0063] S240, if the actual feature data combination exceeds the first feature threshold combination, the training video is stopped playing, and the passive training signal is output to the training robot to convert the training robot from the current non-passive training state to the passive training state.

[0064] S250, the historical feature data combination corresponding to the actual motion data in the first historical time range is determined.

[0065] The first historical time range is set as a configurable item, and the user can set it according to the user's age, the predetermined joint position, and other factors. For example, for the same joint position, the target object A is a child, and the target object B is a young adult, the first historical time range of the target object A should be smaller than the second historical time range of the target object B. For two target objects with similar physical conditions, the first historical time range for the elbow joint can be configured to be greater than the first historical time range for the knee joint, for example, the first historical time range for the elbow joint is 5 seconds, and the first historical time range for the knee joint is 3 seconds.

[0066] ​The historical feature data combination corresponding to different parts is different. For example, in the case that the predetermined joint part is a shoulder joint, the historical feature data combination includes the abduction angle cumulative time of the shoulder joint, the supraspinatus muscle integral electromyography value fatigue index, and the scapula stability score; in the case that the predetermined joint part is an elbow joint, the historical feature data combination includes the elbow joint hyperextension angle, the peak value of the angular velocity, the standard deviation of the acceleration, the hand back pressure proportion, the pressure center trajectory deviation variance, the interval time of continuous hyperextension events, and the hyperextension duration proportion; in the case that the predetermined joint part is a knee joint, the historical feature data combination includes kinematic feature data, kinetic feature data, electromyography feature data, and time sequence feature data, the kinematic feature data includes the peak value of the flexion angle, the standard deviation of the angular velocity, and the number of acceleration mutations; the kinetic feature data includes the peak value of the vertical ground reaction force and the pressure center deviation; the electromyography feature data includes the quadriceps-hamstring muscle synergistic contraction rate, the lateral vastus muscle / medial vastus muscle activation ratio, the time sequence feature data includes the interval time of continuous high-load actions and the cumulative time of the flexion angle, and the high-load action is an action with a load data greater than or equal to a predetermined load.

[0067] S260, input the historical feature data combination into the pre-trained joint state prediction model to obtain the probability of the occurrence of the target state of the predetermined joint part within a future predetermined time length.

[0068] The joint state can be set according to the requirements of the predetermined joint part, such as injury risk, hyperextension risk of the elbow joint, etc.

[0069] In one embodiment, in the case that the predetermined joint part is a shoulder joint, the joint state prediction model is an XGBoost classification model; in the case that the predetermined joint part is an elbow joint or a knee joint, the joint state prediction model is a long short-term memory network model.

[0070] The long short-term memory network model can be a unidirectional long short-term memory network model or a bidirectional long short-term memory network model. For example, the unidirectional long short-term memory network model includes 2 layers of networks, a regularization layer, and a full connection layer, wherein each layer of network includes 64 units, in the regularization layer, the probability of each neural unit being discarded is 20%, and the output dimension of the full connection layer is 1 (single neuron) using a Sigmoid activation function (output value is compressed to 0-1).

[0071] Optionally, the predetermined joint part is a shoulder joint, and the output result is the probability of the occurrence of a rotator cuff injury within 1 minute in the future; the predetermined joint part is an elbow joint, and the output result is the probability of the occurrence of hyperextension within 5 seconds in the future. The predetermined joint part is a knee joint, and the output result is the probability of the occurrence of an anterior cruciate ligament and / or meniscus injury within a week in the future.

[0072] S270, outputting the warning information in the case that the probability meets the warning condition.

[0073] The warning condition comprises one or more levels of warning sub-conditions, and each warning sub-condition corresponds to different warning information.

[0074] In an embodiment, the warning sub-conditions of each level are configured as modifiable items, and the user can set them according to the age and physical condition of the target object.

[0075] The warning information can be that there is a small risk of injury at a predetermined joint position, a large risk of injury at a predetermined joint position, or a very large risk of injury at a predetermined joint position.

[0076] The technical solution provided by the embodiments of the present application not only performs real-time early warning, but also analyzes the actual feature data combination in the first predetermined time range based on the pre-trained joint state prediction model, determines the probability of the occurrence of the target state of the predetermined joint in the future target time period, and determines the warning information according to the probability, thereby achieving the technical efficiency of predicting the future state of the predetermined joint position based on long-term data, and achieving the technical effect of rehabilitation training based on the combination of real-time state and future state, which can significantly improve the effectiveness and scientificity of rehabilitation training.

[0077] On the basis of the foregoing embodiments, whether the predetermined risk threshold needs to be adjusted is determined according to the size relationship between the actual feature data combination in the target time period and the predetermined risk threshold and the probability; if the predetermined risk threshold needs to be adjusted, the predetermined risk threshold is adjusted.

[0078] For example, if the probability is greater than the predetermined risk threshold, but the actual feature data combination does not exceed the second feature threshold combination or the third feature threshold combination, the second feature threshold combination is reduced according to a predetermined percentage reduction strategy. Specifically, if the probability is greater than the predetermined risk threshold, it means that the target object has the problem of excessive force in the current training mode; however, the actual feature data combination does not exceed the second feature threshold combination or the third feature threshold combination in the real-time early warning judgment, that is, the target object does not have the problem of excessive force in the current training mode; since the first historical time range corresponds to a longer time dimension, it can better reflect the load condition of the joint, and therefore the probability output by the pre-trained joint state model is more reliable, and it is determined that the current second feature threshold combination or third feature threshold combination is set too high, and therefore each threshold feature data in the second feature threshold combination or third feature threshold combination is reduced according to the predetermined threshold adjustment strategy, so that the second feature threshold combination or third feature threshold combination is more matched to the physical condition of the target object.

[0079] On the basis of the foregoing embodiment, the sample required for training the joint state prediction model includes a label annotated by a rehabilitation doctor. The sample is actual collected data to which Gaussian noise is added and time stretching is performed. The noise can be Gaussian noise, and the time stretching can be a speed variation of 10% positive and negative. The loss function used by the joint state prediction model can be focalLoss to solve the class imbalance problem, wherein alpha is used to balance the weights of positive and negative samples, and can be 0.25; gamma is used to adjust the difficulty sample weight, and can be 2; the learning rate of the model can use the default value, for example, 0.001. It should be noted that the training of the model can be completed by using an existing training platform, and the present embodiment is not limited in detail.

[0080] Figure 3A A structural diagram of a joint injury prevention device for rehabilitation training provided by an embodiment of the present application is shown in FIG. 1. As shown in the figure, the device includes: Figure 3A

[0081] The response module 31 is configured to respond to a training trigger signal for a predetermined joint part of a target object, control a predetermined terminal to display a training video corresponding to a rehabilitation scheme, and obtain a first feature threshold combination predetermined for the rehabilitation scheme.

[0082] The data acquisition module 32 is configured to acquire actual motion data for the predetermined joint part through a sensor in the training robot during rehabilitation training of the target object following the training video.

[0083] The feature determination module 33 is configured to determine an actual feature data combination corresponding to the actual motion data.

[0084] The control module 34 is configured to stop playing the training video and output a passive training signal to the training robot if the actual feature data combination exceeds the first feature threshold combination, so that the training robot is switched from a current non-passive training state to a passive training state.

[0085] In one embodiment, the response module 31 is further configured to obtain a second feature threshold combination and / or a third feature threshold combination predetermined for the rehabilitation scheme.

[0086] ​The control module 34 is further configured to: if the actual feature data combination exceeds the second feature data combination, reduce the training task difficulty and output a resistance increase signal to the training robot, so that the training robot increases the resistance to the predetermined joint position according to the resistance increase signal; and if the actual feature data combination exceeds the third feature threshold combination, output a prompt information; wherein each feature threshold in the second feature threshold combination is smaller than the corresponding feature threshold in the first feature threshold combination, and each feature threshold in the third feature threshold combination is smaller than the corresponding feature threshold in the second feature threshold combination.

[0087] In one embodiment, as shown in FIG. 1, the device further comprises a prediction module 35, which comprises: Figure 3B

[0088] a feature unit configured to determine a historical feature data combination corresponding to the actual motion data in a first historical time range;

[0089] a probability unit configured to input the historical feature data combination into a pre-trained joint state prediction model to obtain a probability of the predetermined joint position being in a target state in a future predetermined time period;

[0090] a warning unit configured to output a warning information if the probability meets a warning condition.

[0091] In one embodiment, the prediction module 35 is further configured to:

[0092] determine whether the predetermined risk threshold needs to be adjusted according to the size relationship between the actual feature data combination and the predetermined risk threshold in a target time period and the probability;

[0093] if the predetermined risk threshold needs to be adjusted, adjust the predetermined risk threshold.

[0094] In one embodiment, when the predetermined joint position is a shoulder joint, the historical feature data combination comprises an abduction angle cumulative time of the shoulder joint, an upper trapezius integrated electromyography value fatigue index, and a scapula stability score.

[0095] when the predetermined joint position is an elbow joint, the historical feature data combination comprises an elbow joint hyperextension angle, an angular velocity peak value, an acceleration standard deviation, a hand back pressure proportion, a pressure center trajectory deviation variance, a continuous hyperextension event interval time, and a hyperextension duration proportion.

[0096] ​In the case that the predetermined joint part is a knee joint, the historical feature data combination includes kinematics feature data, dynamics feature data, electromyography feature data and timing feature data, the kinematics feature data includes a flexion angle peak value, an angular velocity standard deviation and an acceleration mutation number; the dynamics feature data includes a vertical ground reaction force peak value and a center of pressure offset, the electromyography feature data includes a quadriceps-hamstring muscle synergistic contraction rate, a lateral vastus muscle / medial vastus muscle activation ratio, and the timing feature data includes a continuous high-load action interval time and a flexion angle cumulative time, the high-load action is an action with a load data greater than or equal to a predetermined load.

[0097] In one embodiment, in the case that the predetermined joint part is a shoulder joint, the joint state prediction model is an XGBoost classification model.

[0098] In the case that the predetermined joint part is an elbow joint or a knee joint, the joint state prediction model is a long short-term memory network model.

[0099] The technical scheme provided by the embodiment of the present application, since the first feature threshold combination corresponds to the predetermined joint part of the target object, that is, it is a personalized threshold combination; during the participation of the robot in the rehabilitation training of the target object, one or more sensors built-in the robot will collect actual motion data of the predetermined joint part in real time, therefore, the size relationship between the real-time feature combination corresponding to the actual motion data and the first feature threshold combination can determine whether there is a risk in the predetermined part of the target object, and in the case that there is a risk, the training is stopped, and the training robot is controlled to be converted from a non-passive training state to a passive motion state, so as to protect the predetermined joint part of the target object from a physical level during the end of the training, and the technical effect of specifying a personalized rehabilitation training scheme for the target object is achieved.

[0100] The joint injury prevention device for rehabilitation training provided by the embodiment of the present application can execute the joint injury prevention method for rehabilitation training provided by any embodiment of the present application, has the corresponding function modules and beneficial effects of the execution method.

[0101] Figure 4A structural diagram of an electronic device 10 that can be used to implement embodiments of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smartphones, wearable devices (e.g., headsets, glasses, watches, etc.), and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present application described and / or claimed in this document.

[0102] As shown, Figure 4 The electronic device 10 includes at least one processor 11, and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., communicatively connected to the at least one processor 11, where the memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer programs stored in the read-only memory (ROM) 12 or loaded into the random access memory (RAM) 13 from the storage unit 18. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0103] Various components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc., an output unit 17, such as various types of displays, speakers, etc., a storage unit 18, such as a magnetic disk, an optical disk, etc., and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0104] The processor 11 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs various methods and processes described above, such as the joint injury prevention method for rehabilitation training.

[0105] In some embodiments, the joint injury prevention method for rehabilitation training can be implemented as a computer program tangibly embodied in a computer readable storage medium, e.g., storage unit 18. In some embodiments, parts or all of the computer program can be loaded and / or installed onto electronic device 10 via, e.g., ROM 12 and / or communication unit 19. When the computer program is loaded onto RAM 13 and executed by processor 11, one or more steps of the joint injury prevention method for rehabilitation training described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the joint injury prevention method for rehabilitation training by other means, e.g., with the aid of firmware.

[0106] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (PLD), a computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0107] Computer programs used to implement the methods of the application can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the computer program, when executed by the processor, implements the functions / acts specified in the flowcharts and / or block diagrams. The computer program can be executed entirely on a machine, partially on a machine, partially on a machine as part of a standalone software package, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0108] In the context of the present application, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. A computer-readable storage medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of a machine-readable storage medium will include one or more lines of a program of instructions in a transitory signal, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0109] To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0110] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0111] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. Servers can be cloud servers, also known as cloud computing servers or cloud hosts, which are a host product in the cloud computing service system to solve the defects of great management difficulty and weak business scalability in traditional physical hosts and VPS services.

[0112] The embodiment of the present application also provides a computer program product comprising a computer program which, when executed by a processor, implements the joint injury prevention method for rehabilitation training provided by any embodiment of the present application.

[0113] The computer program product can be written in one or more programming languages or combinations of languages including object-oriented languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0114] It should be understood that the various forms of flow shown above can be re-ordered, added to, or deleted from without departing from the scope of the present application. For example, the steps recited in the present application can be performed in parallel, in series, or in a different order, and the present application is not limited in this regard.

[0115] The specific embodiments described above are not intended to limit the scope of the present application. Those skilled in the art will understand that various modifications, combinations, sub-combinations, and alternatives can be made to the specific embodiments without departing from the spirit and principles of the present application. Any further modifications, equivalents, and / or alternatives come within the scope of the present application as recited by the claims.

Claims

1. A joint injury prevention method for rehabilitation training, characterized by, An object adapted to wear a training robot, comprising: in response to a training trigger signal for a predetermined joint position of the target object, controlling a predetermined terminal to display a training video corresponding to a rehabilitation scheme, and obtaining a first feature threshold combination of the rehabilitation scheme; in the process of the target object following the training video for rehabilitation training, obtaining actual motion data for the predetermined joint position through a sensor in the training robot; determining an actual feature data combination corresponding to the actual motion data; if the actual feature data combination exceeds the first feature threshold combination, stopping playing the training video, and outputting a passive training signal to the training robot to make the training robot switch from a current non-passive training state to a passive training state.

2. The method of claim 1, wherein, the first feature threshold combination of the rehabilitation scheme, also includes: obtaining a second feature threshold combination and / or a third feature threshold combination of the rehabilitation scheme; after determining the actual feature data combination corresponding to the actual motion data, also includes: if the actual feature data combination exceeds the second feature data combination, reducing the difficulty of the training task, and outputting a resistance increase signal to the training robot to make the training robot increase the resistance to the predetermined joint position according to the resistance increase signal; if the actual feature data combination exceeds the third feature threshold combination, outputting prompt information; wherein each feature threshold in the second feature threshold combination is less than the corresponding feature threshold in the first feature threshold combination, and each feature threshold in the third feature threshold combination is less than the corresponding feature threshold in the second feature threshold combination.

3. The method of claim 1, wherein, also includes: determining a historical feature data combination corresponding to the actual motion data in a first historical time range; inputting the historical feature data combination into a pre-trained joint state prediction model to obtain a probability of the predetermined joint position appearing in a target state in a future predetermined time length; if the probability meets an alert condition, outputting alert information.

4. The method of claim 3, wherein, the predetermined joint position is an elbow joint, and the output of the alert information when the probability meets the alert condition includes: if the probability is greater than a first risk probability threshold, reducing the training difficulty of the training content corresponding to the training video, and outputting a resistance increase signal to the training robot to make the training robot increase the resistance to the predetermined joint position according to the resistance increase signal.

5. The method of claim 4, wherein, after outputting the alert information when the probability meets the alert condition, also includes: determining whether the predetermined risk threshold needs to be adjusted according to the size relationship between the actual feature data combination and the predetermined risk threshold in a target time period and the probability; if the predetermined risk threshold needs to be adjusted, adjusting the predetermined risk threshold.

6. The method of claim 3, wherein in the case that the predetermined joint position is a shoulder joint, the historical feature data combination includes the cumulative time of the abduction angle of the shoulder joint, the fatigue index of the integrated electromyogram of the supraspinatus muscle, and the scapula stability score. In the case that the predetermined joint site is an elbow joint, the historical feature data combination includes an elbow hyperextension angle, an angular velocity peak value, an acceleration standard deviation, a hand back pressure proportion, a pressure center trajectory deviation variance, a continuous hyperextension event interval time, and a hyperextension duration proportion. In the case that the predetermined joint site is a knee joint, the historical feature data combination includes kinematic feature data, kinetic feature data, electromyographic feature data, and timing feature data, the kinematic feature data includes a flexion angle peak value, an angular velocity standard deviation, and an acceleration mutation number; the kinetic feature data includes a vertical ground reaction force peak value and a pressure center deviation amount, the electromyographic feature data includes a quadriceps-hamstring muscle synergistic contraction rate, a vastus lateralis / vastus medialis activation ratio, and timing feature data includes a continuous high-load motion interval time and a flexion angle cumulative time, the high-load motion being a motion with a load data greater than or equal to a predetermined load.

7. The method of claim 3, wherein, In the case that the predetermined joint site is a shoulder joint, the joint state prediction model is an XGBoost classification model. In the case that the predetermined joint site is an elbow joint or a knee joint, the joint state prediction model is a long short-term memory network model.

8. An electronic device, comprising: The electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the joint injury prevention method for rehabilitation training according to any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions for causing the processor to execute the joint injury prevention method for rehabilitation training according to any one of claims 1-7 when executed.

10. A computer program product, characterised in that, The computer program product includes a computer program that, when executed by a processor, implements the joint injury prevention method for rehabilitation training according to any one of claims 1-7.

Images