A tactile information feedback processing device and method for lower limb rehabilitation training

Abstract

The present invention discloses a tactile information feedback processing device and method for lower limb rehabilitation training. The device includes: a display module, a lower limb movement acquisition module, a training evaluation module, a feedback processing module, a lower limb rehabilitation training module, and a muscle information acquisition module; the lower limb movement acquisition module is configured to acquire a set of key point spatial position matrices during the process of the user using the lower limb rehabilitation training module for human lower limb muscle strength training; the training evaluation module is configured to perform joint evaluation processing on the force value sequence and the set of key point spatial position matrices to obtain a training effect evaluation value sequence; the lower limb rehabilitation training module is configured to perform rehabilitation training on the user's lower limbs; the feedback processing module is configured to perform feedback calculation processing to obtain a feedback output value; the muscle information acquisition module is configured to acquire a set of the user's muscle information.

Description

Technical Field

[0001] The present invention relates to the fields of exoskeleton robots and intelligent control, and particularly to a tactile information feedback processing device and method for lower limb rehabilitation training. Background Art

[0002] With the aggravation of the aging degree of our society, the number of patients with lower limb motor dysfunction caused by various diseases including stroke is increasing continuously. In addition, the number of patients with nerve or limb injuries caused by work injuries, traffic accidents, diseases, etc. has also increased significantly.

[0003] For the recovery of patients with lower limb movement disorders, since they cannot walk, it is crucial to carry out rehabilitation training. The traditional rehabilitation process based on rehabilitation physicians has defects such as insufficient personnel, high costs, long rehabilitation cycles, boring processes, poor initiative, inability to accurately evaluate the rehabilitation status, and inability to give feedback.

[0004] Current various types of lower limb rehabilitation training devices mainly perform strength training on the lower limbs of users, lack the exploration and utilization of other somatosensory characteristics, and have the problem that the training scenario is single and cannot be expanded, resulting in limited training effects and efficiency. Summary of the Invention

[0005] The present invention mainly solves the problem that current various types of lower limb rehabilitation training devices mainly perform strength training on the lower limbs of users, lack the exploration and utilization of other somatosensory characteristics, and the training scenario is single and cannot be expanded, resulting in limited training effects and efficiency. The present invention discloses a tactile information feedback processing device and method for lower limb rehabilitation training.

[0006] In the first aspect of the embodiments of the present application, a tactile information feedback processing device for lower limb rehabilitation training is disclosed, including: a display module, a lower limb movement acquisition module, a training evaluation module, a feedback processing module, a lower limb rehabilitation training module, and a muscle information acquisition module;

[0007] The display module is used to display training scenario information;

[0008] The lower limb movement acquisition module is connected to the training evaluation module and is used to acquire a set of key point spatial position matrices during the process of the user performing human lower limb muscle strength training using the lower limb rehabilitation training module; the set of key point spatial position matrices includes key point spatial position matrices;

[0009] The training evaluation module is used to perform joint evaluation processing on the force value sequence and the set of key point spatial position matrices to obtain a sequence of training effect evaluation values;

[0010] The lower limb rehabilitation training module is connected to the training evaluation module and is used to perform rehabilitation training on the user's lower limbs, and collect a sequence of force values exerted by the user during the rehabilitation training;

[0011] The feedback processing module is connected to the training evaluation module and the muscle information collection module, and is used to perform feedback calculation processing on the force value sequence, the training effect evaluation value sequence, and the muscle information set to obtain a feedback output value; the feedback output value is used to adjust the tactile stimulation amount of the lower limb rehabilitation training module to the user during the user's lower limb rehabilitation training;

[0012] The muscle information collection module is used to collect the user's muscle information set; the muscle information set includes body fat percentage, total body water value, total protein value, total inorganic salt value, right lower muscle weight, left lower muscle weight, and extracellular water ratio value.

[0013] The lower limb motion collection module is used to collect real-time images of the training scene, perform human key point recognition on the real-time images of the training scene, and obtain a set of key point spatial position matrices; the set of key point spatial position matrices includes the key point spatial position matrix at each data collection moment; the row vector of the key point spatial position matrix is the position coordinate of each human lower limb key point.

[0014] The lower limb rehabilitation training module includes a fixed seat, a foot pedal, and a rotating connecting arm; the rotating connecting arm is used to connect the foot pedal to the fixed seat; a force sensor is arranged on the foot pedal, and the force sensor is used to measure a sequence of force values exerted by the user during the lower limb rehabilitation training using the lower limb rehabilitation training module; a plurality of soft needle-like structures for generating tactile stimulation are arranged on the surface of the foot pedal.

[0015] In the second aspect of the embodiments of the present application, a tactile information feedback processing method for lower limb rehabilitation training is disclosed, which is implemented by using the tactile information feedback processing device for lower limb rehabilitation training, and includes:

[0016] S1, using the display module to display training scene information; using the lower limb rehabilitation training module to perform rehabilitation training on the user's lower limbs, and collect a sequence of force values exerted by the user during the rehabilitation training;

[0017] S2, using the lower limb motion collection module to collect a set of key point spatial position matrices of the user during the lower limb rehabilitation training using the lower limb rehabilitation training module; using the muscle information collection module to collect the user's muscle information set;

[0018] S3. Use the training evaluation module to perform joint evaluation processing on the sequence of strength values and the set of key point spatial position matrices to obtain a sequence of training effect evaluation values;

[0019] S4. Use the feedback processing module to perform feedback calculation processing on the sequence of strength values, the sequence of training effect evaluation values, and the set of muscle information to obtain a feedback output value.

[0020] The joint evaluation processing of the sequence of strength values and the set of key point spatial position matrices to obtain a sequence of training effect evaluation values includes:

[0021] S31. Perform time alignment processing on the sequence of strength values and the set of key point spatial position matrices to obtain an aligned sequence of strength values and a set of key point spatial position matrices;

[0022] S32. Obtain a key point standard position vector and a standard strength value f; use the key point standard position vector as a row vector and copy it in the column direction to obtain a key point standard position matrix with the same dimension as the key point spatial position matrix;

[0023] S33. For each data acquisition moment, extract the strength value f and the key point spatial position matrix at the data acquisition moment from the aligned sequence of strength values and the set of key point spatial position matrices;

[0024] S34. Perform difference calculation processing on the key point spatial position matrix and the key point standard position matrix at the data acquisition moment to obtain a position evaluation sub-value;

[0025] S35. Perform joint difference calculation processing on the strength value and the position evaluation sub-value at the data acquisition moment to obtain the training effect evaluation value at the data acquisition moment;

[0026] The expression for the joint difference calculation processing is:

[0027]

[0028] where H2 is the training effect evaluation value at the data acquisition moment, and exp represents the exponential operation of the constant e;

[0029] S36. Use the training effect evaluation values at all data acquisition moments to construct a sequence of training effect evaluation values.

[0030] The expression for the difference calculation processing is:

[0031]

[0032] where M and N are the row dimension and column dimension of the key point spatial position matrix respectively, δ jis the evaluation sub-item value of the j-th column, A ij and B ij are the elements of the i-th row and the j-th column of the key point spatial position matrix and the key point standard position matrix respectively, H1 is the position evaluation sub-value, and ρ and τ are preset constant values.

[0033] Performing feedback calculation processing on the force value sequence, the training effect evaluation value sequence, and the muscle information set to obtain a feedback output value, including:

[0034] Performing first feedback calculation processing on the force value sequence and the training effect evaluation value sequence to obtain an intermediate feedback quantity;

[0035] Performing second feedback calculation processing on the intermediate feedback quantity, the training effect evaluation value sequence, and the muscle information set to obtain a feedback output value.

[0036] The expression of the first feedback calculation processing is:

[0037]

[0038] where H3 is the intermediate feedback quantity, N0 is the length of the training effect evaluation value sequence, H 2,i is the i-th element of the training effect evaluation value sequence, H 2,max is the maximum value element of the training effect evaluation value sequence, and f i is the i-th element of the force value sequence.

[0039] Performing second feedback calculation processing on the intermediate feedback quantity, the training effect evaluation value sequence, and the muscle information set to obtain a feedback output value, including:

[0040] Obtaining a standard muscle information set; the standard muscle information set includes the standard values of each muscle index;

[0041] Performing difference evaluation processing on the muscle information set and the standard muscle information set to obtain a muscle evaluation result value je;

[0042] Performing feedback calculation processing on the intermediate feedback quantity, the training effect evaluation value sequence, and the muscle evaluation result value to obtain a feedback output value ne.

[0043] The expression of the feedback calculation processing is:

[0044]

[0045] where L1(), L2(), and L3() respectively represent the first Laguerre polynomial, the second Laguerre polynomial, and the third Laguerre polynomial, and κ1 and κ2 respectively represent the preset first weighting coefficient and second weighting coefficient.

[0046] The beneficial effects of the present invention are as follows:

[0047] The present invention solves the problem that current various types of lower limb rehabilitation training devices mainly perform strength training on the user's lower limbs, lack the exploration and utilization of other somatosensory features, and have a single training scenario that cannot be expanded, resulting in limited training effects and efficiency.

[0048] Aiming at the problem of accurately calculating the tactile information feedback amount for lower limb rehabilitation training, the present invention realizes the comprehensive coverage of the user's training information by fusing the force value sequence, the training effect evaluation value sequence, muscle information, etc., ensuring the comprehensiveness and accuracy of the feedback result.

[0049] In the process of calculating the tactile information feedback amount for lower limb rehabilitation training, the present invention performs a first feedback calculation process on the force value sequence and the training effect evaluation value sequence to obtain an intermediate feedback amount; performs a second feedback calculation process on the intermediate feedback amount, the training effect evaluation value sequence, and the muscle information set to obtain a feedback output value. Through the two-step calculation method, the stability and convergence of the feedback result are ensured, and the divergence of the feedback result is avoided.

[0050] When evaluating the training effect, the present invention performs a joint evaluation process on the force value sequence and the key point spatial position matrix set to obtain the training effect evaluation value sequence, ensuring the comprehensiveness of the training evaluation result. The difference calculation process is performed on the key point spatial position matrix at the data acquisition moment and the key point standard position matrix to obtain a position evaluation sub-value, and a model for joint difference calculation process is established to ensure the real-time and accuracy of the training effect evaluation value. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a composition diagram of the device of the present invention;

[0052] Figure 2 It is an implementation flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0053] To better understand the content of the present invention, an embodiment is given here.

[0054] Figure 1 It is a composition diagram of the device of the present invention; Figure 2 It is an implementation flowchart of the method of the present invention.

[0055] Aiming at the problem that current various types of lower limb rehabilitation training devices mainly perform strength training on the user's lower limbs, lack the exploration and utilization of other somatosensory features, and have a single training scenario that cannot be expanded, resulting in limited training effects and efficiency, the present invention discloses a tactile information feedback processing device and method for lower limb rehabilitation training.

[0056] In the first aspect of the embodiments of the present application, a tactile information feedback processing device for lower limb rehabilitation training is disclosed, including: a display module, a lower limb movement acquisition module, a training evaluation module, a feedback processing module, a lower limb rehabilitation training module, and a muscle information acquisition module;

[0057] The display module is used to display training scenario information;

[0058] The lower limb movement acquisition module is connected to the training evaluation module and is used to acquire a set of key point spatial position matrices during the process of the user performing lower limb muscle strength training using the lower limb rehabilitation training module; the set of key point spatial position matrices includes key point spatial position matrices;

[0059] The training evaluation module is used to perform joint evaluation processing on the force value sequence and the set of key point spatial position matrices to obtain a training effect evaluation value sequence;

[0060] The lower limb rehabilitation training module is connected to the training evaluation module and is used to perform rehabilitation training on the user's lower limbs and acquire a force value sequence exerted by the user during the rehabilitation training;

[0061] The feedback processing module is connected to the training evaluation module and the muscle information acquisition module and is used to perform feedback calculation processing on the force value sequence, the training effect evaluation value sequence, and the muscle information set to obtain a feedback output value; the feedback output value is used to adjust the tactile stimulation amount of the lower limb rehabilitation training module on the user during the user's lower limb rehabilitation training;

[0062] The muscle information acquisition module is used to acquire a set of user's muscle information; the set of muscle information includes body fat percentage, total body water value, total protein value, total inorganic salt value, right lower muscle weight, left lower muscle weight, and extracellular water ratio value;

[0063] The lower limb movement acquisition module is used to acquire real-time images of the training scenario, perform human key point recognition on the real-time images of the training scenario, and obtain a set of key point spatial position matrices; the set of key point spatial position matrices includes key point spatial position matrices at each acquisition moment; the row vectors of the key point spatial position matrices are the position coordinates of each human lower limb key point;

[0064] The lower limb rehabilitation training module includes a fixed seat, a foot pedal, and a rotating connecting arm; the rotating connecting arm is used to connect the foot pedal to the fixed seat; a mechanical sensor is arranged on the foot pedal, and the mechanical sensor is used to measure a force value sequence exerted by the user during the lower limb rehabilitation training using the lower limb rehabilitation training module; a plurality of soft needle-like structures for generating tactile stimulation are arranged on the surface of the foot pedal.

[0065] The soft needle-like structure can be a needle-like structure made of rubber material.

[0066] The training evaluation module is used to perform joint evaluation processing on the sequence of force values and the set of key point spatial position matrices to obtain a sequence of training effect evaluation values, including:

[0067] The training evaluation module performs time alignment processing on the sequence of force values and the set of key point spatial position matrices to obtain an aligned sequence of force values and a set of key point spatial position matrices;

[0068] Obtain the standard position vector of the key point and the standard force value f; use the standard position vector of the key point as a row vector and copy it in the column direction to obtain a standard key point position matrix with the same dimension as the key point spatial position matrix;

[0069] For each data acquisition moment, extract the force value f and the key point spatial position matrix at the data acquisition moment from the aligned sequence of force values and the set of key point spatial position matrices;

[0070] Perform difference calculation processing on the key point spatial position matrix and the standard key point position matrix at the data acquisition moment to obtain a position evaluation sub-value;

[0071] The expression of the difference calculation processing is:

[0072]

[0073]

[0074] Among them, M and N are the row dimension and column dimension of the key point spatial position matrix respectively, δ j is the evaluation sub-item value of the j-th column, A ij and B ij are the elements of the i-th row and j-th column of the key point spatial position matrix and the standard key point position matrix respectively, H1 is the position evaluation sub-value, and ρ and τ are preset constant values;

[0075] ρ is a preset constant value, and its value can be 0.8, and τ can be 0.2.

[0076] Perform joint difference calculation processing on the force value and the position evaluation sub-value at the data acquisition moment to obtain the training effect evaluation value at the data acquisition moment;

[0077] The expression of the joint difference calculation processing is:

[0078]

[0079] Among them, H2 is the training effect evaluation value at the data collection moment, and exp represents the exponential operation of the constant e;

[0080] Using the training effect evaluation values at all data collection moments, a training effect evaluation value sequence is constructed;

[0081] The feedback processing module is used to perform feedback calculation processing on the force value sequence, the training effect evaluation value sequence, and the muscle information set to obtain a feedback output value, including:

[0082] The feedback processing module performs first feedback calculation processing on the force value sequence and the training effect evaluation value sequence to obtain an intermediate feedback quantity;

[0083] Performing second feedback calculation processing on the intermediate feedback quantity, the training effect evaluation value sequence, and the muscle information set to obtain a feedback output value.

[0084] The expression of the first feedback calculation processing is:

[0085]

[0086] Among them, H3 is the intermediate feedback quantity, N0 is the length of the training effect evaluation value sequence, H 2,i is the i-th element of the training effect evaluation value sequence, H 2,max is the maximum value element of the training effect evaluation value sequence, f i is the i-th element of the force value sequence;

[0087] Performing second feedback calculation processing on the intermediate feedback quantity, the training effect evaluation value sequence, and the muscle information set to obtain a feedback output value, including:

[0088] Obtaining a standard muscle information set; the standard muscle information set includes the standard values of each muscle index;

[0089] Performing difference evaluation processing on the muscle information set and the standard muscle information set to obtain a muscle evaluation result value je;

[0090] Performing feedback calculation processing on the intermediate feedback quantity, the training effect evaluation value sequence, and the muscle evaluation result value to obtain a feedback output value ne;

[0091] The expression of the feedback calculation processing is:

[0092]

[0093] Wherein, L1(), L2(), L3() represent a first-order Laguerre polynomial, a second-order Laguerre polynomial, and a third-order Laguerre polynomial, respectively, and κ1 and κ2 represent a preset first weighting coefficient and a second weighting coefficient, respectively;

[0094] The lower limb rehabilitation training module includes a fixed seat, a first rotating connecting arm, a second rotating connecting arm and a foot pedal which are rotatably connected in sequence; the second rotating connecting arm and the fixed seat are rotatably arranged at both ends of the first rotating connecting arm, and the foot pedal is rotatably connected to one end of the second rotating connecting arm away from the first rotating connecting arm; a driving component is used to drive the first rotating connecting arm and the second rotating connecting arm to rotate, and the driving component includes a motor mounted on the fixed seat, and a synchronous belt connected to the rotating shafts at both ends of the first rotating connecting arm, and one end of the first rotating connecting arm is fastened to the driving shaft of the motor; when the motor drives the first rotating connecting arm and the second rotating connecting arm to be linked, the foot pedal is in a horizontal movement state.

[0095] The amount of tactile stimulation determines the setting density and radius of the soft needle-like structure on the surface of the foot pedal.

[0096] The muscle information collection module can be implemented by using a human body composition analyzer.

[0097] The human body key point recognition can be implemented by using SURF feature point detection or corner point detection algorithm, or OpenPose algorithm in OpenCV.

[0098] The key points of the human lower limbs include knee joints, ankle joints, feet, etc.

[0099] The step of performing difference evaluation processing on the muscle information set and the standard muscle information set to obtain a muscle evaluation result value includes:

[0100] Performing time alignment processing on the muscle information set and the standard muscle information set to obtain a time-aligned muscle information set and a time-aligned standard muscle information set;

[0101] A muscle information matrix is ​​constructed using the time-aligned muscle information set; the row vector of the muscle information is a muscle information sequence;

[0102] Taking the time-aligned standard muscle information set as a row vector, the row vector is copied in the column direction to obtain a standard muscle matrix; the dimension of the standard muscle matrix is ​​the same as the dimension of the muscle information matrix;

[0103] Subtracting the muscle information matrix from the standard muscle matrix to obtain a difference matrix;

[0104] Calculate the eigen-mean of the difference matrix to obtain the mean vector;

[0105] The expression for the eigen-mean calculation is:

[0106]

[0107] where b p is the p-th element of the mean vector, A pq is the element at the p-th row and q-th column of the difference matrix, and n is the column dimension of the difference matrix;

[0108] Calculate the weight vector of the difference matrix to obtain the weight vector;

[0109] The expression for the weight vector calculation is:

[0110]

[0111] where v p is the p-th element of the standard muscle information set for time alignment, and f p is the p-th element of the weight vector;

[0112] Perform weighted summation on the weight vector and the mean vector to obtain the muscle evaluation result value;

[0113] The time alignment process includes:

[0114] When the acquisition time interval of the muscle information set is greater than the time interval of the standard muscle information set, perform interpolation processing on the data of adjacent muscle information sets to obtain a time-aligned muscle information set;

[0115] When the acquisition time interval of the muscle information set is less than the time interval of the standard muscle information set, perform sampling processing on the muscle information set to obtain data with the same time interval as the standard muscle information set, and use the sampled muscle information set as the time-aligned muscle information set;

[0116] In the second aspect of the embodiments of the present application, a tactile information feedback processing method for lower limb rehabilitation training is disclosed, which is implemented by using the tactile information feedback processing device for lower limb rehabilitation training, and includes:

[0117] S1, use the display module to display training scenario information; use the lower limb rehabilitation training module to perform rehabilitation training on the user's lower limbs, and collect a sequence of force values exerted by the user during the rehabilitation training;

[0118] S2. Using the lower limb motion acquisition module, collect the set of key point spatial position matrices of the user during the lower limb rehabilitation training using the lower limb rehabilitation training module; using the muscle information acquisition module, collect the set of muscle information of the user.

[0119] S3. Using the training evaluation module, jointly evaluate and process the force value sequence and the set of key point spatial position matrices to obtain a training effect evaluation value sequence.

[0120] S4. Using the feedback processing module, perform feedback calculation processing on the force value sequence, the training effect evaluation value sequence, and the set of muscle information to obtain a feedback output value.

[0121] The joint evaluation and processing of the force value sequence and the set of key point spatial position matrices to obtain a training effect evaluation value sequence includes:

[0122] Perform time alignment processing on the force value sequence and the set of key point spatial position matrices to obtain an aligned force value sequence and a set of key point spatial position matrices.

[0123] Obtain the key point standard position vector and the standard force value f; use the key point standard position vector as a row vector and copy it column-wise to obtain a key point standard position matrix with the same dimension as the key point spatial position matrix.

[0124] For each data acquisition moment, extract the force value f and the key point spatial position matrix at the data acquisition moment from the aligned force value sequence and the set of key point spatial position matrices.

[0125] Perform difference calculation processing on the key point spatial position matrix and the key point standard position matrix at the data acquisition moment to obtain a position evaluation sub-value.

[0126] The expression of the difference calculation processing is:

[0127]

[0128] where M and N are the row dimension and column dimension of the key point spatial position matrix respectively, δ j is the evaluation sub-item value of the j-th column, A ij and B ij are the elements of the i-th row and j-th column of the key point spatial position matrix and the key point standard position matrix respectively, H1 is the position evaluation sub-value, and ρ and τ are preset constant values.

[0129] ρ is a preset constant value, and its value can be 0.8, and τ can take a value of 0.2.

[0130] Perform a joint difference calculation on the force value and the position evaluation sub-value at the data acquisition moment to obtain the training effect evaluation value at the data acquisition moment;

[0131] The expression for the joint difference calculation is:

[0132]

[0133] where H2 is the training effect evaluation value at the data acquisition moment, and exp represents the exponential operation of the constant e;

[0134] Use the training effect evaluation values at all data acquisition moments to construct a training effect evaluation value sequence;

[0135] The feedback calculation on the force value sequence, the training effect evaluation value sequence, and the muscle information set to obtain a feedback output value includes:

[0136] Perform a first feedback calculation on the force value sequence and the training effect evaluation value sequence to obtain an intermediate feedback quantity;

[0137] Perform a second feedback calculation on the intermediate feedback quantity, the training effect evaluation value sequence, and the muscle information set to obtain a feedback output value.

[0138] The expression for the first feedback calculation is:

[0139]

[0140] where H3 is the intermediate feedback quantity, N0 is the length of the training effect evaluation value sequence, H 2,i is the i-th element of the training effect evaluation value sequence, H 2,max is the maximum value element of the training effect evaluation value sequence, and f i is the i-th element of the force value sequence;

[0141] The feedback calculation on the intermediate feedback quantity, the training effect evaluation value sequence, and the muscle information set to obtain a feedback output value includes:

[0142] Obtain a standard muscle information set; the standard muscle information set includes the standard values of each muscle index;

[0143] Perform a difference evaluation on the muscle information set and the standard muscle information set to obtain a muscle evaluation result value je;

[0144] Perform a feedback calculation on the intermediate feedback quantity, the training effect evaluation value sequence, and the muscle evaluation result value to obtain a feedback output value ne;

[0145] The expression of the feedback calculation process is:

[0146]

[0147] Wherein, L1(), L2(), L3() represent a first-order Laguerre polynomial, a second-order Laguerre polynomial, and a third-order Laguerre polynomial, respectively, and κ1 and κ2 represent a preset first weighting coefficient and a second weighting coefficient, respectively;

[0148] The lower limb rehabilitation training module includes a fixed seat, a first rotating connecting arm, a second rotating connecting arm and a foot pedal which are rotatably connected in sequence; the second rotating connecting arm and the fixed seat are rotatably arranged at both ends of the first rotating connecting arm, and the foot pedal is rotatably connected to one end of the second rotating connecting arm away from the first rotating connecting arm; a driving component is used to drive the first rotating connecting arm and the second rotating connecting arm to rotate, and the driving component includes a motor mounted on the fixed seat, and a synchronous belt connected to the rotating shafts at both ends of the first rotating connecting arm, and one end of the first rotating connecting arm is fastened to the driving shaft of the motor; when the motor drives the first rotating connecting arm and the second rotating connecting arm to be linked, the foot pedal is in a horizontal movement state.

[0149] The amount of tactile stimulation determines the setting density and radius of the soft needle-like structure on the surface of the foot pedal.

[0150] The muscle information collection module can be implemented by using a human body composition analyzer.

[0151] The human body key point recognition can be implemented by using SURF feature point detection or corner point detection algorithm, or OpenPose algorithm in OpenCV.

[0152] The key points of the human lower limbs include knee joints, ankle joints, feet, etc.

[0153] The step of performing difference evaluation processing on the muscle information set and the standard muscle information set to obtain a muscle evaluation result value includes:

[0154] Performing time alignment processing on the muscle information set and the standard muscle information set to obtain a time-aligned muscle information set and a time-aligned standard muscle information set;

[0155] A muscle information matrix is ​​constructed using the time-aligned muscle information set; the row vector of the muscle information is a muscle information sequence;

[0156] Taking the time-aligned standard muscle information set as a row vector, the row vector is copied in the column direction to obtain a standard muscle matrix; the dimension of the standard muscle matrix is ​​the same as the dimension of the muscle information matrix;

[0157] Subtract the muscle information matrix from the standard muscle matrix to obtain a difference matrix;

[0158] Calculate the eigen mean value of the difference matrix to obtain an eigenvector;

[0159] The expression for the eigen mean value calculation is:

[0160]

[0161] where b p is the p-th element of the eigenvector, A pq is the element at the p-th row and q-th column of the difference matrix, and n is the column dimension of the difference matrix;

[0162] Calculate the weight vector of the difference matrix to obtain a weight vector;

[0163] The expression for the weight vector calculation is:

[0164]

[0165] where v p is the p-th element of the time-aligned standard muscle information set, and f p is the p-th element of the weight vector;

[0166] Perform a weighted sum of the weight vector and the eigenvector to obtain a muscle evaluation result value;

[0167] The time alignment process includes:

[0168] When the acquisition time interval of the muscle information set is greater than the time interval of the standard muscle information set, perform interpolation processing on the data of adjacent muscle information sets to obtain a time-aligned muscle information set;

[0169] When the acquisition time interval of the muscle information set is less than the time interval of the standard muscle information set, perform sampling processing on the muscle information set to obtain data with the same time interval as the standard muscle information set, and use the sampled muscle information set as the time-aligned muscle information set.

[0170] The above are only embodiments of the present application and are not used to limit the present application. For those skilled in the art, the present application may have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A tactile information feedback processing device for lower limb rehabilitation training, characterized in that: include: Display module, lower limb motion acquisition module, training evaluation module, feedback processing module, lower limb rehabilitation training module, muscle information acquisition module; The display module is used to display training scene information; The lower limb motion acquisition module is connected to the training evaluation module and is used to acquire a key point spatial position matrix set of the user during the lower limb muscle strength training of the human body using the lower limb rehabilitation training module; the key point spatial position matrix set includes a key point spatial position matrix; The training evaluation module is used to perform joint evaluation processing on the force value sequence and the key point spatial position matrix set to obtain a training effect evaluation value sequence; The lower limb rehabilitation training module is connected to the training evaluation module and is used to perform rehabilitation training on the user's lower limbs and collect a sequence of force values ​​applied by the user during the rehabilitation training; The feedback processing module is connected to the training evaluation module and the muscle information acquisition module, and is used to perform feedback calculation processing on the strength value sequence, the training effect evaluation value sequence, and the muscle information set to obtain a feedback output value; The feedback output value is used to adjust the tactile stimulation amount of the lower limb rehabilitation training module to the user when the user is performing lower limb rehabilitation training; the tactile stimulation amount determines the setting density and radius of the soft needle-like structure on the surface of the pedal; The muscle information collection module is used to collect the user's muscle information set; the muscle information set includes body fat percentage, total body water value, total protein value, total inorganic salt value, right lower muscle weight, left lower muscle weight and extracellular water ratio value; The lower limb rehabilitation training module includes a fixed seat, a foot pedal and a rotating connecting arm; the rotating connecting arm is used to connect the foot pedal to the fixed seat; a mechanical sensor is provided on the foot pedal, and the mechanical sensor is used to measure the force value sequence applied by the user when performing lower limb rehabilitation training using the lower limb rehabilitation training module; the surface of the foot pedal is provided with a plurality of soft needle-like structures for generating tactile stimulation.

2. The tactile information feedback processing device for lower limb rehabilitation training according to claim 1, characterized in that: The lower limb motion acquisition module is used to acquire real-time images of training scenes, identify key points of the human body in the real-time images of training scenes, and obtain a set of key point spatial position matrices; the set of key point spatial position matrices includes the key point spatial position matrix at each data acquisition moment; the row vectors of the key point spatial position matrix are the position coordinates of each key point of the lower limbs of the human body.

3. A tactile information feedback processing method for lower limb rehabilitation training, characterized in that: The method is implemented by using the tactile information feedback processing device for lower limb rehabilitation training according to any one of claims 1 to 2, comprising: S1, using the display module to display training scene information; using the lower limb rehabilitation training module to perform rehabilitation training on the user's lower limbs, and collecting a sequence of force values ​​applied by the user during the rehabilitation training; S2, using the lower limb motion acquisition module to acquire a set of key point spatial position matrices of the user during the lower limb rehabilitation training process using the lower limb rehabilitation training module; using the muscle information acquisition module to acquire a set of muscle information of the user; S3, using the training evaluation module to jointly evaluate the force value sequence and the key point spatial position matrix set to obtain a training effect evaluation value sequence; S4, using the feedback processing module, performing feedback calculation processing on the strength value sequence, the training effect evaluation value sequence, and the muscle information set to obtain a feedback output value.

4. The tactile information feedback processing method for lower limb rehabilitation training according to claim 3, characterized in that: The joint evaluation process of the force value sequence and the key point spatial position matrix set to obtain the training effect evaluation value sequence includes: S31, performing time alignment processing on the force value sequence and the key point spatial position matrix set to obtain an aligned force value sequence and key point spatial position matrix set; S32, obtaining the key point standard position vector and the standard force value f0; taking the key point standard position vector as a row vector, and copying it in the column direction to obtain a key point standard position matrix with the same dimension as the key point spatial position matrix; S33, for each data collection moment, extracting the force value f and the key point spatial position matrix at the data collection moment from the aligned force value sequence and key point spatial position matrix set; S34, performing difference calculation processing on the key point spatial position matrix and the key point standard position matrix at the time of data collection to obtain a position evaluation sub-value; S35, performing joint difference calculation processing on the strength value and the position evaluation sub-value at the data collection moment to obtain a training effect evaluation value at the data collection moment; The expression for the joint difference calculation process is: Wherein, H2 is the training effect evaluation value at the time of data collection, and exp represents the exponential operation of the constant e; S36, constructing a training effect evaluation value sequence using the training effect evaluation values ​​at all data collection moments.

5. The tactile information feedback processing method for lower limb rehabilitation training according to claim 4, characterized in that: The expression for the difference calculation process is: Among them, M and N are the row dimension and column dimension of the key point spatial position matrix, respectively, and δ j is the evaluation item value of the jth column, A ij and B ij are the elements of the i-th row and j-th column of the key point spatial position matrix and the key point standard position matrix respectively, H1 is the position evaluation subvalue, ρ and τ are preset constant values.

6. The tactile information feedback processing method for lower limb rehabilitation training according to claim 5, characterized in that: The feedback calculation process is performed on the strength value sequence, the training effect evaluation value sequence, and the muscle information set to obtain a feedback output value, including: Performing a first feedback calculation process on the strength value sequence and the training effect evaluation value sequence to obtain an intermediate feedback amount; A second feedback calculation process is performed on the intermediate feedback amount, the training effect evaluation value sequence and the muscle information set to obtain a feedback output value.

7. The tactile information feedback processing method for lower limb rehabilitation training according to claim 6, characterized in that: The expression of the first feedback calculation process is: Among them, H3 is the intermediate feedback amount, N0 is the length of the training effect evaluation value sequence, H 2,i is the i-th element of the training effect evaluation value sequence, H 2,max is the maximum value element of the training effect evaluation value sequence, f i is the i-th element of the strength value sequence.

8. The tactile information feedback processing method for lower limb rehabilitation training according to claim 7, characterized in that: The performing a second feedback calculation process on the intermediate feedback amount, the training effect evaluation value sequence and the muscle information set to obtain a feedback output value includes: Obtaining a standard muscle information set; the standard muscle information set includes a standard value of each muscle index; Performing difference evaluation processing on the muscle information set and the standard muscle information set to obtain a muscle evaluation result value je; Feedback calculation processing is performed on the intermediate feedback amount, the training effect evaluation value sequence and the muscle evaluation result value to obtain a feedback output value ne.

9. The tactile information feedback processing method for lower limb rehabilitation training according to claim 8, characterized in that: The expression of the feedback calculation process is: Among them, L1(), L2(), L3() represent the first-order Laguerre polynomial, the second-order Laguerre polynomial, and the third-order Laguerre polynomial respectively, and κ1 and k2 represent the preset first weighting coefficient and the second weighting coefficient respectively.

Images