Upper limb rehabilitation training method and system based on human body motion model

Abstract

The invention discloses an upper limb rehabilitation training method and system based on a human body motion model, and relates to the technical field of rehabilitation training. Constructing a human body motion model and outputting simulation data; calculating a physiological index coefficient, and further calculating the exercise intensity of each training action; calculating a motion mode complexity index of each training action so as to obtain and calculate the training duration of each training action; and calculating a motion deviation value, and further calculating a comprehensive dynamic deviation index. Presetting a threshold value and judging whether the rehabilitation training plan is suitable for the current state of the patient; whether a rehabilitation training plan needs to be changed; according to the method, the suitability of the rehabilitation training plan can be evaluated in real time, and the training scheme is dynamically adjusted; the training is always fit with the condition of the patient, and the rehabilitation efficiency and success rate are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of rehabilitation training, and specifically to an upper limb rehabilitation training method and system based on a human motion model. Background Art

[0002] With the acceleration of the modern life rhythm and the advent of an aging society, the number of patients with upper limb injuries and functional impairments is increasing day by day. As an important part of the human body for daily activities and work, the recovery of the upper limb function is crucial for the quality of life of patients and their return to society. However, traditional upper limb rehabilitation training faces many challenges and is difficult to formulate accurate rehabilitation programs according to the individual physiological structure and rehabilitation progress of patients, resulting in uneven rehabilitation effects and unable to meet the needs of personalized rehabilitation.

[0003] To solve the above problems, existing upper limb rehabilitation training technologies have introduced the concept of personalization to a certain extent. Some programs evaluate the basic physical information of patients and simple injury conditions, and initially formulate rehabilitation training plans. At the same time, data such as joint movement angles and muscle electrical signals during the upper limb movement of patients are collected using joint angle sensors, electromyography sensors, etc., to monitor the effect of rehabilitation training and make limited adjustments to the training plan.

[0004] Although the existing technologies have made certain progress, there are still obvious deficiencies. On the one hand, relying only on basic physical information and partial movement data, the adaptability of the training plan cannot be evaluated. On the other hand, in the formulation and adjustment of the rehabilitation training plan, the existing systems lack dynamic and in-depth analysis of the individual differences and rehabilitation progress of patients. The formulated rehabilitation training plans often lack flexibility and pertinence, and are difficult to adapt to the changing needs of patients at different rehabilitation stages. Summary of the Invention

[0005] (I) Technical Problems to be Solved In view of the deficiencies of the existing technologies, the present invention provides an upper limb rehabilitation training method and system based on a human motion model. By collecting multiple data of patients; constructing a human motion model and outputting simulation data; calculating physiological index coefficients, further calculating the exercise intensity of each training action; calculating the exercise mode complexity index of each training action, and then obtaining the training duration of each training action; calculating the motion deviation value, and then calculating the comprehensive dynamic deviation index. Presetting a threshold and judging whether the rehabilitation training plan is suitable for the current state of the patient; calculating the comprehensive evaluation index of the rehabilitation progress, presetting a threshold and judging whether the rehabilitation progress of the patient reaches the expectation, solving the problems of low flexibility and low efficiency of traditional rehabilitation training.

[0006] (II) Technical Solutions To achieve the above objectives, the present invention is realized through the following technical solutions: An upper limb rehabilitation training method based on a human body movement model, including: Collect the personal information data, upper limb bone structure data, muscle strength distribution data, and joint range of motion data of the patient; Take the upper limb bone structure data, muscle strength distribution data, and joint range of motion data as input data to construct a human body movement model; import the patient's rehabilitation training plan and conduct simulations, and output the upper limb movement simulation data, muscle strength change simulation data, and joint movement parameter simulation data as output data through the human body movement model; According to the upper limb movement simulation data, muscle strength change simulation data, and joint movement parameter simulation data, calculate the physiological index coefficient P; according to the personal information data and the physiological index coefficient P, calculate the exercise intensity E of each training action; calculate the exercise mode complexity index CM of each training action based on the exercise intensity E of each training action; calculate the training duration t of each training action according to the exercise intensity E and the exercise mode complexity index CM of each training action; According to the training duration t of each training action, muscle strength change simulation data, and joint movement parameter simulation data, calculate the movement deviation value CH; calculate the comprehensive dynamic deviation index ST according to the movement deviation value CH, muscle strength distribution data, and joint range of motion data; preset the comprehensive dynamic deviation index threshold, compare the comprehensive dynamic deviation index ST with the comprehensive dynamic deviation index threshold, and judge whether the rehabilitation training plan is suitable for the patient's current state and whether the rehabilitation training plan needs to be replaced according to the comparison result.

[0007] In the preferred scheme of the above upper limb rehabilitation training method based on a human body movement model: The method for calculating the physiological index coefficient P is: The upper limb movement simulation data includes the average value LA of the upper limb bone length of the patient; The muscle strength change simulation data includes the average value FA of the upper limb muscle strength of the patient; The joint movement parameter simulation data includes the average value PA of the upper limb joint movement angle range of the patient; According to the average value LA of the upper limb bone length, the average value FA of the upper limb muscle strength, and the average value PA of the upper limb joint movement angle range, calculate the physiological index coefficient P, and the formula relied on is: ; Among them, LR is the average value of the upper limb bone length of normal people under the same conditions; FR is the average value of the upper limb joint movement angle range of normal people under the same conditions; PR is the average value of the upper limb joint movement angle range of normal people under the same conditions; α1 is The weight coefficient of, with a value of 0.1 to 0.3; α2 is The weight coefficient has a value range of 0.3 to 0.4; α3 is The weight coefficient has a value range of 0.3 to 0.5; and α1 + α2 + α3 = 1.

[0008] In the preferred embodiment of the upper limb rehabilitation training method based on the human motion model described above: The method for calculating the exercise intensity E of each rehabilitation training action is as follows: Personal information data includes age AG and gender GG; Based on age AG, gender GG, and the physiological index coefficient P, calculate the exercise intensity E of each rehabilitation training action. The calculation formula is: ; where B is the basic exercise intensity of each rehabilitation training action; M is the adaptation coefficient of this rehabilitation training action in the human motion model; II is the upper limb injury severity index.

[0009] In the preferred embodiment of the upper limb rehabilitation training method based on the human motion model described above: The method for calculating the exercise pattern complexity index CM of each rehabilitation training action is as follows: Based on the exercise intensity E of each rehabilitation training action, calculate the exercise pattern complexity index CM of each rehabilitation training action. The formula used is: ; where F i is the basic complexity of the i-th exercise pattern, i is the serial number of different exercise patterns, with a value range of [1, n]; n is the total number of exercise patterns, taking a positive integer value; RE is the achievable probability of this exercise pattern in the human motion model; E j is the exercise intensity of the j-th training action, j is the serial number of the training action, with a value range of [1, m]; m is the total number of training actions, taking a positive integer value.

[0010] In the preferred embodiment of the upper limb rehabilitation training method based on the human motion model described above: The method for calculating the training duration t of each rehabilitation training action is as follows: Based on the exercise intensity E and the exercise pattern complexity index CM of each rehabilitation training action, calculate the training duration t of each rehabilitation training action. The formula used is: ; where DS is the basic training duration of each rehabilitation training action; CM j is the exercise pattern complexity index of the j-th rehabilitation training action; T is the time limit of the preset rehabilitation goal.

[0011] In the preferred embodiment of the upper limb rehabilitation training method based on the human motion model described above: The method for calculating the motion deviation value CH is as follows: The muscle strength change simulation data also includes the muscle electrical signal intensity value H k ; The joint movement parameter simulation data also includes the joint angle value A k ; According to the training duration t of each rehabilitation training action, the muscle electrical signal intensity value H k and the joint angle value A k , calculate the motion deviation value CH, and the formula is: ; where H k is the muscle electrical signal intensity value collected at the k-th time point; k is the time point serial number, and the value range is [1, N]; N is the total number of collected time points, and the value is a positive integer; HE k is the expected muscle electrical signal intensity value of the human motion model under the corresponding training action at the k-th time point; A k is the joint angle value collected at the k-th time point; AE k is the joint angle value of the human motion model under the corresponding training action at the k-th time point; T k is the motion trajectory speed value collected at the k-th time point; TE k is the motion trajectory speed value of the human motion model under the corresponding training action at the k-th time point; ω1 is 's weight coefficient, and the value range is 0.2 to 0.4; ω2 is 's weight coefficient, and the value range is 0.3 to 0.5; ω3 is 's weight coefficient, and the value range is 0.1 to 0.5; and ω1 + ω2 + ω3 = 1.

[0012] In the preferred scheme of the above upper limb rehabilitation training method based on the human motion model: the method for calculating the comprehensive dynamic deviation index ST is: According to the muscle electrical signal intensity value H k , the joint angle value A k and the motion deviation value CH, calculate the comprehensive dynamic deviation index ST, and the formula is: ; where is the muscle electrical signal intensity value collected at the (k + 1)-th time point; is the joint angle value collected at the (k + 1)-th time point; is the motion trajectory speed value collected at the (k + 1)-th time point.

[0013] In the preferred scheme of the above upper limb rehabilitation training method based on the human motion model: the method for judging whether the rehabilitation training plan is suitable for the patient's current state is: Preset the comprehensive dynamic deviation index threshold STth; When the comprehensive dynamic deviation index ST ≤ the comprehensive dynamic deviation index threshold STth, it is determined that the rehabilitation training plan is suitable for the patient's current state, and the training continues according to the rehabilitation training plan; When the comprehensive dynamic deviation index ST > the comprehensive dynamic deviation index threshold STth, it is determined that the rehabilitation training plan is not suitable for the patient's current state, and the rehabilitation training plan is replaced.

[0014] In the preferred embodiment of the above-mentioned upper limb rehabilitation training method based on the human motion model: The method for determining whether the patient's rehabilitation progress has reached the expectation is as follows: Based on the comprehensive dynamic deviation index ST, calculate the comprehensive rehabilitation progress evaluation index PN, and the calculation formula is: ; where DF is the duration of rehabilitation training; DFmax is the preset maximum rehabilitation training time; EW is the energy consumption value during rehabilitation training; EWmax is the preset maximum energy consumption value; PQ is the percentage increase in the upper limb strength of the patient; MQ is the percentage increase in the movement range of the patient's upper limb; φ1 is the weight coefficient of, with a value range of 0.1 - 0.4; φ2 is the weight coefficient of, with a value range of 0.05 - 0.3; φ3 is the weight coefficient of, with a value range of 0.2 - 0.5; φ4 is the weight coefficient of the percentage increase in the upper limb strength PQ of the patient, with a value range of 0.2 - 0.4; φ5 is the weight coefficient of the percentage increase in the movement range MQ of the patient's upper limb, with a value range of 0.05 - 0.1; and φ1 + φ2 + φ3 + φ4 + φ5 = 1; Preset the comprehensive rehabilitation progress evaluation index threshold PNth; When the comprehensive rehabilitation progress evaluation index PN < the comprehensive rehabilitation progress evaluation index threshold PNth, it is determined that the patient's rehabilitation progress has not reached the expectation, and the rehabilitation training plan is adjusted; When the comprehensive rehabilitation progress evaluation index PN ≥ the comprehensive rehabilitation progress evaluation index threshold PNth, it is determined that the patient's rehabilitation progress has reached the expectation.

[0015] In the preferred embodiment of the above-mentioned upper limb rehabilitation training method based on the human motion model: If the rehabilitation training plan is suitable for the patient's current state, select this rehabilitation training plan for rehabilitation training; and regularly evaluate the rehabilitation training plan.

[0016] The present invention also discloses an upper limb rehabilitation training system based on the human motion model, which is used to implement the above-mentioned upper limb rehabilitation training method based on the human motion model, and includes: A data acquisition module, which is used to collect the patient's personal information data, upper limb bone structure data, muscle strength distribution data, and joint range of motion data; A model construction module, which is used to take the upper limb bone structure data, muscle strength distribution data, and joint range of motion data as input data to construct a human motion model; import the patient's rehabilitation training plan and conduct simulations, and output the upper limb motion simulation data, muscle strength change simulation data, and joint activity parameter simulation data as output data through the human motion model; A training plan generation module, which is used to calculate the physiological index coefficient P according to the upper limb motion simulation data, muscle strength change simulation data, and joint activity parameter simulation data; calculate the exercise intensity E of each training action according to the personal information data and the physiological index coefficient P; calculate the exercise mode complexity index CM of each training action based on the exercise intensity E of each training action; calculate the training duration t of each training action according to the exercise intensity E and the exercise mode complexity index CM of each training action; A training module, which is used to calculate the motion deviation value CH according to the training duration t of each training action, the muscle strength change simulation data, and the joint activity parameter simulation data; calculate the comprehensive dynamic deviation index ST according to the motion deviation value CH, the muscle strength distribution data, and the joint range of motion data; preset the comprehensive dynamic deviation index threshold, compare the comprehensive dynamic deviation index ST with the comprehensive dynamic deviation index threshold, and judge whether the rehabilitation training plan is suitable for the patient's current state and whether the rehabilitation training plan needs to be replaced according to the comparison result.

[0017] (III) Beneficial effects The present invention provides an upper limb rehabilitation training method and system based on a human motion model, which has the following beneficial effects: (1) Collecting the patient's personal information data, upper limb bone structure data, muscle strength distribution data, and joint range of motion data can comprehensively and accurately obtain the patient's physical condition information, provide a detailed and reliable data basis for subsequent model construction, and ensure the pertinence and effectiveness of rehabilitation training.

[0018] (2) Using the upper limb-related data as input data to construct a human motion model, importing the training plan for simulation, and outputting multiple simulation data. This operation can pre-enact the rehabilitation training process, intuitively present the effects of training, help discover potential problems in advance, optimize the training plan, and improve the scientificity and rationality of rehabilitation training.

[0019] (3) Calculating the exercise intensity, exercise mode complexity index, and training duration of training actions. The accurate determination of these parameters can reasonably arrange the training intensity and difficulty according to the patient's individual situation and rehabilitation needs, ensure that the training can effectively promote rehabilitation without causing excessive burden to the patient, and improve the safety and effectiveness of rehabilitation training.

[0020] (4) Calculate the motion deviation value and the comprehensive dynamic deviation index, and compare them with the preset threshold to determine whether the training plan is appropriate. This process can monitor the training effect in real time, timely detect the compatibility problem between the training plan and the actual state of the patient, so as to adjust the training plan in time, ensure that the rehabilitation training always fits the current state of the patient, and improve the rehabilitation efficiency. Brief Description of the Drawings

[0021] Figure 1 It is a schematic diagram of the steps of the upper limb rehabilitation training method based on the human motion model of the present invention. Detailed Description of the Invention

[0022] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0023] Please refer to Figure 1 , the present invention provides an upper limb rehabilitation training method based on a human motion model, including: Step 1: Collect the patient's personal information data, upper limb bone structure data, muscle strength distribution data, and joint range of motion data.

[0024] Comprehensive Step 1: Comprehensively collect data such as the patient's personal information, upper limb bone structure, muscle strength distribution, and joint range of motion, providing detailed and accurate basic information for subsequent rehabilitation training, helping to deeply understand the patient's individual conditions, making the rehabilitation plan more targeted, and avoiding the blindness of training caused by information deficiency.

[0025] Step 2: Use the upper limb bone structure data, muscle strength distribution data, and joint range of motion data as input data to construct a human motion model; import the patient's rehabilitation training plan and simulate it, and output the upper limb motion simulation data, muscle strength change simulation data, and joint movement parameter simulation data as output data through the human motion model.

[0026] Step 201: Use the LSTM long short-term memory network algorithm to train a large amount of normal human upper limb motion data to learn the patterns and rules of normal upper limb motion. Then input the patient's upper limb bone structure data, muscle strength distribution data, and joint range of motion data into the trained model to perform personalized adjustment on the model so that it can accurately simulate the upper limb motion state of the patient.

[0027] Step 202: Import the patient's current rehabilitation training plan into the constructed human motion model, activate the simulation function of the model, and simulate the upper limb movement process of the patient when performing the rehabilitation training plan according to the parameters of the training actions and the mechanical relationship of the human motion model. During the simulation process, the model will calculate and output the upper limb movement simulation data, muscle strength change simulation data, and joint activity parameter simulation data in real time.

[0028] Combined steps 201 to 202: Constructing a human motion model and simulating the training plan using upper limb related data can intuitively present the expected effect of the rehabilitation training, evaluate the feasibility and potential risks of different training plans in advance, provide a scientific basis for optimizing the training plan, reduce the trial-and-error cost in actual training, and improve the scientificity and rationality of the training.

[0029] Step 3: Calculate the physiological index coefficient P according to the upper limb movement simulation data, muscle strength change simulation data, and joint activity parameter simulation data; calculate the exercise intensity E of each training action according to the personal information data and the physiological index coefficient P; calculate the exercise pattern complexity index CM of each training action based on the exercise intensity E of each training action; calculate the training duration t of each training action according to the exercise intensity E and the exercise pattern complexity index CM of each training action.

[0030] It should be noted that before performing the formula calculation on the upper limb movement simulation data, muscle strength change simulation data, and joint activity parameter simulation data, data preprocessing such as normalization preprocessing needs to be carried out to eliminate the dimension of each parameter and facilitate the calculation and analysis of subsequent formulas.

[0031] Step 301: Calculate the physiological index coefficient P, and the specific method is as follows: The upper limb movement simulation data includes the average value LA of the patient's upper limb bone length.

[0032] It should be noted that by using CT or MRI medical imaging equipment to perform tomographic scans on the patient's upper limb, three-dimensional image data is obtained, and then imported into Mimics medical image processing software. Through the image segmentation function, the upper limb bones are separated from the overall image to obtain a three-dimensional bone model. Use the measurement tool in the software to measure the lengths of the main bones such as the humerus, ulna, and radius respectively. Record the measured bone length data of the humerus, ulna, radius, etc. Then calculate the average value of these bone lengths to obtain the average value LA of the upper limb bone length.

[0033] The muscle strength change simulation data includes the average value FA of the patient's upper limb muscle strength.

[0034] It should be noted that electromyography sensors are attached to the surface of the patient's upper limb muscle groups to collect muscle electrical signals. The signal-strength conversion algorithm is used to convert the electrical signal intensity into muscle strength values, and the strength of different muscle groups is recorded respectively. After multiple measurements, the strength values of each muscle group are summarized and their average value is calculated, so as to obtain the average value of upper limb muscle strength FA.

[0035] The simulated data of joint movement parameters includes the average value of the upper limb joint movement angle range PA of the patient.

[0036] It should be noted that joint angle sensors are fixed at each joint of the patient's upper limb, such as the shoulder joint, elbow joint, and wrist joint. The patient is asked to perform movements such as maximum flexion and extension, rotation of each joint of the upper limb, and the sensor collects the joint movement angle data in real time. After multiple repeated measurements, the movement angle range of each joint is calculated, that is, the difference between the maximum angle and the minimum angle. Finally, the movement angle range values of each joint are summarized and averaged to obtain the average value of the upper limb joint movement angle range PA.

[0037] According to the average value of upper limb bone length LA, the average value of upper limb muscle strength FA, and the average value of upper limb joint movement angle range PA, the physiological index coefficient P is calculated. The formula is as follows: ; where LR is the average value of the upper limb bone length of normal people under the same conditions; FR is the average value of the upper limb joint movement angle range of normal people under the same conditions; PR is the average value of the upper limb joint movement angle range of normal people under the same conditions; α1 is the weight coefficient of, with a value range of 0.1 - 0.3; α2 is the weight coefficient of, with a value range of 0.3 - 0.4; α3 is the weight coefficient of, with a value range of 0.3 - 0.5; and α1 + α2 + α3 = 1.

[0038] It should be noted that by querying the NLM database or the Chinese Biomedical Literature Database, etc., the average value of the upper limb bone length LR of normal people, the average value of the upper limb joint movement angle range FR of normal people, and the average value of the upper limb joint movement angle range PR of normal people under the same conditions are found.

[0039] It should be noted that This item indicates that if this ratio is close to 1, it means that the upper limb bone length of the patient is similar to that of normal people; if the ratio deviates greatly from 1, it means that there is a difference between the patient's bone length and the normal situation. This ratio being close to 1 means that the muscle strength of the patient is equivalent to that of normal people; a ratio less than 1 indicates that the patient's muscle strength is weak, and a ratio greater than 1 indicates that the muscle strength is strong. The ratio reflects the difference between the patient's joint range of motion and the normal level, and can be used to judge whether the flexibility and motor function of the patient's joints are normal. Multiply these three parts by their respective weight coefficients and then add them together, and then multiply by 100% to obtain the physiological index coefficient. Multiplying by 100% is to convert the result into a percentage form, which is convenient for intuitively understanding and comparing the state of the patient's upper limb physiological function relative to that of a normal person.

[0040] Step 302: Calculate the exercise intensity E of each rehabilitation training action. The specific method is as follows: Personal information data includes age AG and gender GG.

[0041] It should be noted that by querying the patient's medical record file, age AG and gender GG are obtained therefrom.

[0042] According to age AG, gender GG and physiological index coefficient P, calculate the exercise intensity E of each rehabilitation training action. The calculation formula is: ; Among them, B is the basic exercise intensity of each rehabilitation training action; M is the adaptation coefficient of this rehabilitation training action in the human motion model; II is the upper limb injury severity index.

[0043] It should be noted that to obtain the basic exercise intensity B of each rehabilitation training action, by searching the rehabilitation training action standard database, the corresponding basic exercise intensity index is preset in this database. By matching and comparing the specific rehabilitation training action with the actions in the database, its basic exercise intensity B is determined.

[0044] To obtain the adaptation coefficient M of this rehabilitation training action in the human motion model, each rehabilitation training action is input into the constructed human motion model for simulation. The model simulates the action execution process according to the individual physiological data of the patient's upper limb bone structure, muscle strength distribution, joint range of motion, etc. By analyzing the force conditions of each joint and the degree of fit between the muscle force generation mode and the patient's actual physiological state during action execution, the adaptation coefficient M is calculated.

[0045] To obtain the upper limb injury severity index II, obtain the actual area AZ of the upper limb injury, the total area ATA of the upper limb, and the injury depth DV, and calculate the upper limb injury severity index II therefrom. The calculation formula is: ; Among them, Dmax is the maximum possible injury depth of the upper limb.

[0046] It should be noted that B is the basic exercise intensity of each rehabilitation training action and is an inherent attribute of the action itself. M reflects the adaptability of the action to the upper limb motor ability of the patient. The higher the adaptability, the higher the initial exercise intensity. The degree of upper limb injury is considered. The more severe the injury, the lower the initial exercise intensity. The comprehensive health index of the patient is combined. The better the health condition, the higher the initial exercise intensity. The influence of gender is considered, and the exercise intensity is finely adjusted. The age factor is considered. The older the age, the relatively lower the initial exercise intensity. By multiplying the above-mentioned multiple parts, the exercise intensity E of each rehabilitation training action is obtained.

[0047] Step 303: Calculate the motion pattern complexity index CM of each rehabilitation training action. The specific method is as follows: Based on the exercise intensity E of each rehabilitation training action, calculate the motion pattern complexity index CM of each rehabilitation training action. The formula is as follows: ; Among them, F i is the basic complexity of the i-th motion pattern, i is the serial number of different motion patterns, and the value range is [1, n]; n is the total number of motion patterns, and the value is a positive integer; RE is the achievable probability of this motion pattern in the human motion model; E j is the exercise intensity of the j-th training action, j is the serial number of the training action, and the value range is [1, m]; m is the total number of training actions, and the value is a positive integer.

[0048] It should be noted that to obtain the basic complexity F i of the i-th motion pattern, count the number of upper limb joints NO j involved in the i-th motion pattern and the number of muscle groups NM participating in the motion, and calculate the basic complexity F i of the i-th motion pattern based on this. The calculation formula is: ; Among them, is the weight coefficient of the number of upper limb joints NO j involved in the i-th motion pattern, and the value range is 0.5 - 0.7; is the weight coefficient of the number of muscle groups NM participating in the motion, and the value range is 0.4 - 0.6; and .

[0049] Obtain the achievable probability RE of the motion pattern in the human motion model by calculating the ratio of the number of times the model meets the motion pattern execution conditions during the simulation process to the total number of simulations to determine the achievable probability RE. For example, during the simulation of the execution of a certain motion pattern 1000 times, if the model can successfully execute the motion pattern as expected 800 times, then the achievable probability RE of this motion pattern in the human motion model of this patient is 0.8.

[0050] It should be noted that the basic complexity F of the motion pattern i reflects the complexity of the i-th motion pattern itself. The achievable probability RE reflects the feasibility of a specific motion pattern in the human motion model. This term represents the proportion of the exercise intensity of the j-th training action in the total exercise intensity of all training actions. This means that for training actions with a greater exercise intensity, a higher weight will be assigned to the motion pattern adopted when calculating the motion pattern complexity index. The formula combines the basic complexity, achievable probability of different motion patterns, and the weight of the exercise intensity of training actions to obtain the motion pattern complexity index CM of each rehabilitation training action.

[0051] Step 304: Calculate the training duration t of each rehabilitation training action. The specific method is as follows: Calculate the training duration t of each rehabilitation training action according to the exercise intensity E and the motion pattern complexity index CM of each rehabilitation training action. The formula used is: ; where DS is the basic training duration of each rehabilitation training action; CM j is the motion pattern complexity index of the j-th rehabilitation training action; T is the time limit of the preset rehabilitation goal.

[0052] It should be noted that to obtain the basic training duration DS of each rehabilitation training action, collect the rehabilitation training data of a large number of similar patients, analyze the actual training duration of each rehabilitation training action when achieving similar rehabilitation goals. Through the statistics and analysis of these data, calculate the average training duration of each rehabilitation training action, and use this average value as the basic training duration DS of each rehabilitation training action.

[0053] Obtain the time limit T of the preset rehabilitation goal. Obtain the shortest rehabilitation time Tmin for similar upper limb injuries in the best recovery situation and the longest rehabilitation time Tmax for similar upper limb injuries in the general and poor recovery situations, and calculate the time limit T of the preset rehabilitation goal based on this. The calculation formula is: ; where r is the adjustment coefficient of the upper limb injury severity index II.

[0054] It should be noted that DS is the basic training duration for each rehabilitation training action. By comprehensively considering the initial exercise intensity and the complexity of the exercise pattern of the training action, the relative importance of this action among all training actions is determined. The time limit of the rehabilitation goal is considered. As the rehabilitation time approaches, the training duration will be appropriately adjusted. By multiplying these three parts, the training duration t of each rehabilitation training action is obtained.

[0055] Comprehensive steps 301 to 304: Calculate the exercise intensity, exercise pattern complexity index, and training duration of the training action, and the training content can be accurately customized according to the specific situation of the patient, ensuring that the training intensity and difficulty are adapted to the patient's rehabilitation stage and physical ability, which can not only effectively promote rehabilitation but also prevent injuries caused by overtraining, improving the safety and effectiveness of training.

[0056] Step four: According to the training duration t of each training action, the simulated data of muscle strength change, and the simulated data of joint movement parameters, calculate the movement deviation value CH; according to the movement deviation value CH, the muscle strength distribution data, and the joint movement range data, calculate the comprehensive dynamic deviation index ST; preset the threshold of the comprehensive dynamic deviation index, compare the comprehensive dynamic deviation index ST with the threshold of the comprehensive dynamic deviation index, and judge whether the rehabilitation training plan is suitable for the patient's current state and whether the rehabilitation training plan needs to be replaced according to the comparison result.

[0057] Step 401: Calculate the movement deviation value CH. The specific method is as follows: The simulated data of muscle strength change also includes the muscle electrical signal intensity value H k .

[0058] It should be noted that the muscle electrical signal is collected by using a surface electromyography sensor placed on the skin to obtain the muscle electrical signal intensity value H k .

[0059] The simulated data of joint movement parameters also includes the joint angle value A k .

[0060] It should be noted that the joint angle value A is obtained by directly measuring the joint angle value with an electronic goniometer sensor k .

[0061] According to the training duration t of each rehabilitation training action, the muscle electrical signal intensity value H k and the joint angle value A k , calculate the movement deviation value CH. The formula is as follows: ; where H kis the muscle electrical signal intensity value collected at the k-th time point; k is the time point serial number, with a value range of [1, N]; N is the total number of collected time points, and N is a positive integer; HE k is the expected muscle electrical signal intensity value of the human motion model under the corresponding training action at the k-th time point; A k is the joint angle value collected at the k-th time point; AE k is the joint angle value of the human motion model under the corresponding training action at the k-th time point; T k is the motion trajectory speed value collected at the k-th time point; TE k is the motion trajectory speed value of the human motion model under the corresponding training action at the k-th time point; ω1 is the weight coefficient of, with a value range of 0.2 to 0.4; ω2 is the weight coefficient of, with a value range of 0.3 to 0.5; ω3 is the weight coefficient of, with a value range of 0.1 to 0.5; and ω1 + ω2 + ω3 = 1.

[0062] It should be noted that to obtain the expected muscle electrical signal intensity value HE k of the human motion model under the corresponding training action at the k-th time point, human data such as the upper limb bone structure and muscle strength distribution of the patient are input into the trained human motion model. For the k-th rehabilitation training action, the model simulates the expected muscle electrical signal intensity value HE k corresponding to the k-th time point according to the learned normal mode and the patient's individual situation.

[0063] To obtain the joint angle value AE k of the human motion model under the corresponding training action at the k-th time point, data such as the patient's bone structure and joint range of motion are incorporated into the human motion model. Based on the standard action model and combined with the patient's individual situation, the model adjusts and predicts the joint angle to obtain the expected joint angle value AE k corresponding to the k-th time point under the training action.

[0064] To obtain the motion trajectory speed value TE k of the human motion model under the corresponding training action at the k-th time point, motion parameters such as the patient's limb length and joint mobility are input into the human motion model. The model simulates the motion trajectory speed value TE k corresponding to the k-th time point under the training action according to the established speed-time relationship model and the patient's individual parameters.

[0065] It should be noted that this item is the muscle electrical signal intensity value collected by the patient at each time point k The expected muscle electrical signal intensity value at the same time point as the human body movement model under the corresponding training action A comparison is made. By calculating the sum of the squares of the differences, the deviation degree between the actual value and the expected value is highlighted. Similarly, the sum of the squares of the differences is used to measure the difference between the actual situation of the joint angle and the expectation, reflecting whether the patient's joint movement conforms to the expected pattern. This part of the calculation reflects the deviation between the patient's movement speed and the expectation, which is of great significance for evaluating the coordination and accuracy of the movement. Multiply the above three comparison items by their respective weight coefficients and then add them together, and then multiply by the training duration adjustment factor , to obtain the movement deviation value CH.

[0066] Step 402: Calculate the comprehensive dynamic deviation index ST, and the specific method is as follows: According to the muscle electrical signal intensity value H k , the joint angle value A k and the movement deviation value CH, calculate the comprehensive dynamic deviation index ST, and the formula is: ; wherein, is the muscle electrical signal intensity value collected at the k + 1th time point; is the joint angle value collected at the k + 1th time point; is the movement trajectory speed value collected at the k + 1th time point.

[0067] It should be noted that is to normalize the movement deviation value to obtain a relatively stable and convenient basis for comparison of the deviation amount. The sum of the change amounts of the muscle electrical signal intensity values at adjacent time points is calculated. The change of the muscle electrical signal intensity can reflect the dynamic change of muscle force. The sum of the change amounts of the joint angle values at adjacent time points is calculated. The dynamic change of the joint angle is crucial for evaluating the coordination and accuracy of joint movement. The sum of the change amounts of the movement trajectory speed values at adjacent time points is calculated. The change of the movement trajectory speed reflects the speed stability and coordination of the patient's movement. Add the above three sums of change amounts at adjacent time points, and then divide by the sum of the muscle electrical signal intensity values, joint angle values and movement trajectory speed values at all time points , to obtain a relative change ratio. This ratio reflects the relative importance degree of the dynamic changes of the muscle electrical signal intensity, joint angle and movement trajectory speed relative to the overall data during the entire rehabilitation training process. Add 1 to this ratio and then compare it with By multiplying them, the basic amount of motion deviation and the dynamic changes of various factors can be comprehensively considered, and finally the comprehensive dynamic deviation index ST can be obtained.

[0068] Step 402: Determine whether the rehabilitation training plan is suitable for the patient's current state. The specific method is as follows: Preset the comprehensive dynamic deviation index threshold STth; by collecting data of multiple patients with good upper limb rehabilitation training effects and smooth rehabilitation processes, including the muscle electrical signal intensity value H k , joint angle value A k and motion deviation value CH, calculate the comprehensive dynamic deviation index ST of each patient, further calculate the average value of the comprehensive dynamic deviation index STavg, and use the average value of the comprehensive dynamic deviation index STavg as the comprehensive dynamic deviation index threshold STth.

[0069] When the comprehensive dynamic deviation index ST ≤ the comprehensive dynamic deviation index threshold STth, it is determined that the rehabilitation training plan is suitable for the patient's current state, and continue to train according to the rehabilitation training plan; When the comprehensive dynamic deviation index ST > the comprehensive dynamic deviation index threshold STth, it is determined that the rehabilitation training plan is not suitable for the patient's current state, and replace the rehabilitation training plan.

[0070] It should be noted that if the rehabilitation training plan is suitable for the patient's current state, select this rehabilitation training plan for rehabilitation training; and regularly evaluate the rehabilitation training plan.

[0071] Combining Step 401 and Step 402: Calculate the motion deviation value and the comprehensive dynamic deviation index, compare with the preset threshold to determine whether the training plan is appropriate, can monitor the training effect in real time, timely discover the deviation between the training plan and the actual state of the patient, quickly respond and adjust the training plan, ensure that the training always fits the current rehabilitation needs of the patient, and improve the rehabilitation efficiency and quality.

[0072] On the other hand, the present invention also discloses an upper limb rehabilitation training system based on a human motion model for implementing the above-mentioned upper limb rehabilitation training method based on a human motion model, including: A data acquisition module for collecting the patient's personal information data, upper limb bone structure data, muscle strength distribution data, and joint range of motion data; A model construction module for using the upper limb bone structure data, muscle strength distribution data, and joint range of motion data as input data to construct a human motion model; importing the patient's rehabilitation training plan and performing simulations, and outputting the upper limb motion simulation data, muscle strength change simulation data, and joint activity parameter simulation data as output data through the human motion model; A training plan generation module, configured to calculate a physiological index coefficient P according to upper limb movement simulation data, muscle strength change simulation data, and joint activity parameter simulation data; calculate the exercise intensity E of each training action according to personal information data and the physiological index coefficient P; calculate the exercise mode complexity index CM of each training action based on the exercise intensity E of each training action; calculate the training duration t of each training action according to the exercise intensity E and the exercise mode complexity index CM of each training action. A training module, configured to calculate a motion deviation value CH according to the training duration t of each training action, muscle strength change simulation data, and joint activity parameter simulation data; calculate a comprehensive dynamic deviation index ST according to the motion deviation value CH, muscle strength distribution data, and joint activity range data; preset a comprehensive dynamic deviation index threshold, compare the comprehensive dynamic deviation index ST with the comprehensive dynamic deviation index threshold, and judge whether the rehabilitation training plan is suitable for the patient's current state and whether the rehabilitation training plan needs to be replaced according to the comparison result.

[0073] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those of ordinary skill in the art will realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution.

[0074] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0075] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application.

Claims

1. An upper limb rehabilitation training method based on a human body motion model, characterized in that: Including: Collecting the patient's personal information data, upper limb bone structure data, muscle strength distribution data, and joint range of motion data; Using the upper limb bone structure data, muscle strength distribution data, and joint range of motion data as input data to construct a human motion model; importing the patient's rehabilitation training plan and conducting simulations, and outputting upper limb motion simulation data, muscle strength change simulation data, and joint activity parameter simulation data as output data through the human motion model; Calculating the physiological index coefficient P based on the upper limb motion simulation data, muscle strength change simulation data, and joint activity parameter simulation data; Calculating the exercise intensity E of each training action based on the personal information data and the physiological index coefficient P; Calculating the exercise pattern complexity index CM of each training action based on the exercise intensity E of each training action; calculating the training duration t of each training action based on the exercise intensity E and the exercise pattern complexity index CM of each training action; Calculating the motion deviation value CH based on the training duration t of each training action, the muscle strength change simulation data, and the joint activity parameter simulation data; calculating the comprehensive dynamic deviation index ST based on the motion deviation value CH, the muscle strength distribution data, and the joint range of motion data; presetting a comprehensive dynamic deviation index threshold, comparing the comprehensive dynamic deviation index ST with the comprehensive dynamic deviation index threshold, and judging whether the rehabilitation training plan is suitable for the patient's current state and whether the rehabilitation training plan needs to be replaced according to the comparison result.

2. The upper limb rehabilitation training method based on the human body motion model according to claim 1, wherein: The method for calculating the physiological index coefficient P is: The upper limb motion simulation data includes the average upper limb bone length LA of the patient; The muscle strength change simulation data includes the average upper limb muscle strength FA of the patient; The joint activity parameter simulation data includes the average upper limb joint activity angle range PA of the patient; Calculating the physiological index coefficient P based on the average upper limb bone length LA, the average upper limb muscle strength FA, and the average upper limb joint activity angle range PA, and the formula used is: ； Among them, LR is the average value of the upper limb bone length of normal people under the same conditions; FR is the average value of the range of upper limb joint movement angles of normal people under the same conditions; PR is the average value of the range of upper limb joint movement angles of normal people under the same conditions; α1 is The weight coefficient of, with a value range of 0.1 to 0.3; α2 is The weight coefficient of, with a value range of 0.3 to 0.4; α3 is The weight coefficient of, with a value range of 0.3 to 0.5; and α1 + α2 + α3 = 1.

3. The upper limb rehabilitation training method based on a human body motion model according to claim 2, characterized in that: The method for calculating the exercise intensity E of each rehabilitation training action is: The personal information data includes age AG and gender GG; Calculating the exercise intensity E of each rehabilitation training action based on age AG, gender GG, and the physiological index coefficient P, and the calculation formula is: ； Where B is the basic exercise intensity of each rehabilitation training action; M is the adaptation coefficient of this rehabilitation training action in the human motion model; II is the upper limb injury severity index.

4. The upper limb rehabilitation training method based on the human body motion model according to claim 3, characterized in that: The method for calculating the exercise pattern complexity index CM of each rehabilitation training action is: Calculating the exercise pattern complexity index CM of each rehabilitation training action based on the exercise intensity E of each rehabilitation training action, and the formula used is: ； Among them, F i is the basic complexity of the i-th motion pattern, where i is the serial number of different motion patterns, taking values in [1, n]; n is the total number of motion patterns, taking positive integer values; RE is the achievable probability of this motion pattern in the human motion model; E j is the exercise intensity of the j-th training action, where j is the serial number of the training action, taking values in [1, m]; m is the total number of training actions, taking positive integer values.

5. The upper limb rehabilitation training method based on a human body motion model according to claim 4, characterized in that: The method for calculating the training duration t of each rehabilitation training action is: Calculating the training duration t of each rehabilitation training action based on the exercise intensity E and the exercise pattern complexity index CM of each rehabilitation training action, and the formula used is: ； Among them, DS is the basic training duration of each rehabilitation training action; CM j is the movement pattern complexity index of the j-th rehabilitation training action; T is the time limit of the preset rehabilitation goal.

6. The upper limb rehabilitation training method based on the human body motion model according to claim 5, wherein: The method for calculating the motion deviation value CH is: The muscle strength change simulation data also includes the muscle electrical signal intensity value H k ; The simulated data of joint movement parameters also includes the joint angle value A k ; According to the training duration t of each rehabilitation training action, the muscle electrical signal intensity value H k and the joint angle value A k , calculate the motion deviation value CH, and the formula is as follows: ； Among them, H k is the muscle electrical signal intensity value collected at the k-th time point; k is the time point serial number, with a value range of [1, N]; N is the total number of collected time points, and N is a positive integer; HE k is the expected muscle electrical signal intensity value of the human motion model under the corresponding training action at the k-th time point; A k is the joint angle value collected at the k-th time point; AE k is the joint angle value of the human motion model under the corresponding training action at the k-th time point; T k is the motion trajectory speed value collected at the k-th time point; TE k is the motion trajectory speed value of the human body motion model corresponding to the training action at the k-th time point; ω1 is weight coefficient of, with a value range of 0.2 to 0.4; ω2 is weight coefficient of, with a value range of 0.3 to 0.5; ω3 is weight coefficient of, with a value range of 0.1 to 0.5; and ω1 + ω2 + ω3 = 1.

7. The upper limb rehabilitation training method based on the human body motion model according to claim 6, characterized in that: The method for calculating the comprehensive dynamic deviation index ST is: According to the muscle electrical signal intensity value H k and the joint angle value A k and the motion deviation value CH, calculate the comprehensive dynamic deviation index ST, and the formula is as follows: ； Among them, is the muscle electrical signal intensity value collected at the (k + 1)-th time point; is the joint angle value collected at the (k + 1)-th time point; is the motion trajectory speed value collected at the (k + 1)-th time point.

8. The upper limb rehabilitation training method based on a human body motion model according to claim 7, wherein: The method for judging whether the rehabilitation training plan is suitable for the patient's current state is: Presetting a comprehensive dynamic deviation index threshold STth; When the comprehensive dynamic deviation index ST ≤ the comprehensive dynamic deviation index threshold STth, it is determined that the rehabilitation training plan is suitable for the patient's current state, and the training continues according to the rehabilitation training plan; When the comprehensive dynamic deviation index ST > the comprehensive dynamic deviation index threshold STth, it is determined that the rehabilitation training plan is not suitable for the patient's current state, and the rehabilitation training plan is replaced.

9. The upper limb rehabilitation training method based on the human body movement model according to claim 8, characterized in that: If the rehabilitation training plan is suitable for the patient's current state, select this rehabilitation training plan for rehabilitation training; and regularly evaluate the rehabilitation training plan.

10. An upper limb rehabilitation training system based on a human motion model, characterized in that: A data acquisition module for collecting the patient's personal information data, upper limb bone structure data, muscle strength distribution data, and joint range of motion data; A model construction module for using the upper limb bone structure data, muscle strength distribution data, and joint range of motion data as input data to construct a human motion model; importing the patient's rehabilitation training plan and simulating it, and outputting upper limb motion simulation data, muscle strength change simulation data, and joint activity parameter simulation data as output data through the human motion model; A training plan generation module for calculating the physiological index coefficient P according to the upper limb motion simulation data, muscle strength change simulation data, and joint activity parameter simulation data; Calculating the exercise intensity E of each training action according to the personal information data and the physiological index coefficient P; Calculating the exercise pattern complexity index CM of each training action based on the exercise intensity E of each training action; calculating the training duration t of each training action according to the exercise intensity E and the exercise pattern complexity index CM of each training action; A training module for calculating the motion deviation value CH according to the training duration t of each training action, the muscle strength change simulation data, and the joint activity parameter simulation data; calculating the comprehensive dynamic deviation index ST according to the motion deviation value CH, the muscle strength distribution data, and the joint range of motion data; presetting the comprehensive dynamic deviation index threshold, comparing the comprehensive dynamic deviation index ST with the comprehensive dynamic deviation index threshold, and judging whether the rehabilitation training plan is suitable for the patient's current state and whether the rehabilitation training plan needs to be replaced according to the comparison result.

Images