Anti-unbalance lower limb rehabilitation training system and method

Abstract

The invention provides an anti-unbalance lower limb rehabilitation training system and method. The method comprises the following steps: collecting simulated training data; extracting gait training tracks of all normal people in each gait cycle from the simulation training data, and determining an imbalance compensation coefficient of the target object during lower limb rehabilitation training through each gait training track; obtaining an initial gait track of the target object, and determining difference association features of gaits of the target object in each gait cycle according to the initial gait track; track correction is carried out on a dependence path when the target object carries out gait conversion between adjacent gait cycles through the unbalance compensation coefficient and each difference correlation feature, and a self-adaptive anti-unbalance track is obtained; and the lower limb rehabilitation training equipment performs personalized anti-unbalance adjustment according to the self-adaptive anti-unbalance trajectory. By adopting the scheme of the invention, personalized anti-unbalance rehabilitation training can be realized according to the difference of different patients on different rehabilitation training requirements, so that the effectiveness of the anti-unbalance lower limb rehabilitation training equipment can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of lower limb rehabilitation training, and more specifically, to an anti-imbalance lower limb rehabilitation training system and method. Background Art

[0002] Lower limb rehabilitation training equipment is a medical device designed to help patients recover lower limb function. These devices can help patients who need lower limb rehabilitation due to surgery, injury and other problems; there are many types of lower limb rehabilitation training equipment, including: passive training equipment, active training equipment, hybrid training equipment, functional electrical stimulation equipment, balance and stability training equipment, aquatic rehabilitation equipment and virtual reality rehabilitation equipment.

[0003] Anti-imbalance lower limb rehabilitation training equipment is usually designed to help patients maintain balance during training, prevent falls, and provide a safer rehabilitation environment. Although the existing anti-imbalance lower limb rehabilitation training equipment is equipped with sensors and real-time feedback mechanisms, it can monitor the patient's center of gravity in real time and provide feedback and correction instructions to help patients maintain balance during training. However, due to the large differences in gait characteristics of different patients, the unified training model in the existing anti-imbalance lower limb rehabilitation training is difficult to meet the rehabilitation needs of different patients, resulting in the safety and training effects of patients using anti-imbalance lower limb rehabilitation training equipment for rehabilitation training. It is difficult to achieve expected results. Therefore, how to achieve personalized anti-imbalance rehabilitation training according to the differences in different rehabilitation training needs of different patients, thereby improving the effectiveness of anti-imbalance lower limb rehabilitation training equipment and accelerating the rehabilitation process has become a difficult problem faced by the industry. Summary of the invention

[0004] The present application provides an anti-imbalance lower limb rehabilitation training system and method, which can realize personalized anti-imbalance rehabilitation training according to the differences in rehabilitation training needs of different patients, thereby improving the effectiveness of the anti-imbalance lower limb rehabilitation training equipment.

[0005] In a first aspect, the present application provides an anti-unbalance adjustment method, comprising: Collect the training records of different normal people during lower limb rehabilitation simulation training, and then obtain simulation training data; Extracting gait training trajectories of all normal people in each gait cycle from the simulated training data, performing balance evaluation on gait transitions of all normal people between adjacent gait cycles through each gait training trajectory, obtaining a conversion balance cost at each gait transition, and then determining an imbalance compensation coefficient when the target object performs lower limb rehabilitation training based on all conversion balance costs; Acquire the initial gait trajectory of the target object when performing lower limb rehabilitation training, determine the trajectory constraints of the target object and all normal people in each gait cycle according to the initial gait trajectory, and then determine the difference correlation characteristics of the gait of the target object in each gait cycle through the morphological differences between the target object and different normal people and all trajectory constraints; Extracting dependent paths when the target object performs gait conversion between adjacent gait cycles from the initial gait trajectory, performing trajectory correction on each dependent path by using the imbalance compensation coefficient and each difference association feature, and obtaining an adaptive anti-imbalance trajectory; The lower limb rehabilitation training equipment of the target object performs personalized anti-imbalance adjustment according to the adaptive anti-imbalance trajectory.

[0006] In some embodiments, extracting the gait training trajectories of all normal people in each gait cycle from the simulation training data specifically includes: Acquire all gait cycles during simulated rehabilitation training; For each gait cycle, extracting gait information of each normal person in the gait cycle from the simulation training data as periodic gait information; Performing data fitting on the gait information of each normal person in the periodic gait information to obtain a trajectory curve of each normal person; The gait training trajectories of all normal people in the gait cycle are determined according to all the trajectory curves, and then the gait training trajectories of all normal people in each gait cycle are obtained.

[0007] In some embodiments, the balance evaluation is performed on the gait transitions between adjacent gait cycles of all normal people through various gait training trajectories, and the conversion balance cost at each gait transition is obtained, which specifically includes: For each gait transition, the gait transition point between adjacent gait cycles of each normal person is determined according to the gait training trajectory in the gait cycle before the gait transition; Obtaining the ground reaction force of each normal person at the gait transition point; The cost of each gait transition point is evaluated according to all ground reaction forces, and the transition balance cost of gait transition between adjacent gait cycles of all normal people is obtained, and then the transition balance cost of each gait transition is obtained.

[0008] In some embodiments, determining the imbalance compensation coefficient of the target object when performing lower limb rehabilitation training based on all conversion balance costs specifically includes: Determine the balance threshold of all normal subjects during simulated rehabilitation training; Determine multiple balance differences when the target object performs lower limb rehabilitation training according to the balance threshold and all conversion balance costs; The imbalance compensation coefficient of the target subject during lower limb rehabilitation training is determined based on all balance differences.

[0009] In some embodiments, determining the trajectory constraints of the target object and all normal people in each gait cycle according to the initial gait trajectory specifically includes: For each gait cycle, obtaining a periodic gait trajectory under the gait cycle in the initial gait trajectory; Determine the trajectory residual of the target object and the corresponding normal person in the gait cycle according to the cycle gait trajectory and the trajectory curve of each normal person in the gait training trajectory; The trajectory constraints between the target object and all normal people in the gait cycle are determined based on all trajectory residuals, and then the trajectory constraints between the target object and all normal people in each gait cycle are obtained.

[0010] In some embodiments, determining the difference-related features of the gait of the target object in each gait cycle through the morphological differences between the target object and different normal people and all trajectory constraints specifically includes: Determine the morphological differences between the target subject and different normal people; For each gait cycle, obtain the trajectory constraints of the target object and the normal person in the gait cycle; Determine the feature coupling degree between the target object and the corresponding normal person according to the trajectory constraint and the morphological difference between the target object and each normal person; Based on all the feature coupling degrees, the difference correlation features of the target object in the gait cycle are determined, and then the difference correlation features of the gait of the target object in each gait cycle are obtained.

[0011] In some embodiments, performing trajectory correction on each dependent path by using the imbalance compensation coefficient and each difference correlation feature to obtain an adaptive anti-imbalance trajectory specifically includes: For each gait cycle, obtain the dependency path of the target object between the gait cycle and the next gait cycle; Correcting the dependent path using the difference correlation feature of the target object in the next gait cycle and the imbalance compensation coefficient to obtain an anti-imbalance trajectory of the dependent path between the gait cycle and the next gait cycle; Then, an anti-imbalance trajectory that depends on the path between each gait cycle and the next gait cycle is obtained; The adaptive anti-unbalance trajectory of the target object is obtained through all anti-unbalance trajectories.

[0012] In a second aspect, the present application provides a lower limb rehabilitation training system for preventing imbalance, comprising a lower limb rehabilitation training device and a rehabilitation training control center, wherein the rehabilitation training control center comprises: The acquisition module is used to collect the training records of different normal people during lower limb rehabilitation simulation training, and then obtain simulation training data; a processing module, for extracting gait training trajectories of all normal persons in each gait cycle from the simulated training data, performing balance evaluation on gait transitions of all normal persons between adjacent gait cycles through each gait training trajectory, obtaining a conversion balance cost at each gait transition, and then determining an imbalance compensation coefficient when the target subject performs lower limb rehabilitation training based on all conversion balance costs; The processing module is also used to obtain the initial gait trajectory of the target object when performing lower limb rehabilitation training, determine the trajectory constraints of the target object and all normal people in each gait cycle according to the initial gait trajectory, and then determine the difference correlation characteristics of the gait of the target object in each gait cycle through the morphological differences between the target object and different normal people and all trajectory constraints; The processing module is further used to extract the dependent path of the target object when performing gait conversion between adjacent gait cycles from the initial gait trajectory, and perform trajectory correction on each dependent path by using the imbalance compensation coefficient and each difference association feature to obtain an adaptive anti-imbalance trajectory; The execution module is used to instruct the lower limb rehabilitation training equipment of the target object to perform personalized anti-imbalance adjustment according to the adaptive anti-imbalance trajectory.

[0013] In a third aspect, the present application provides a computer device, comprising a memory and a processor, wherein the memory stores codes, and the processor is configured to obtain the codes and execute the above-mentioned anti-imbalance adjustment method.

[0014] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned anti-imbalance adjustment method is implemented.

[0015] The technical solution provided by the embodiments disclosed in this application has the following beneficial effects: In the anti-imbalance lower limb rehabilitation training system and method provided in the present application, firstly, the lower limb rehabilitation training equipment is started, and the training records of different normal people performing lower limb rehabilitation simulation training are collected to obtain simulation training data; the gait training trajectories of all normal people in each gait cycle are extracted from the simulation training data, and the gait conversion between adjacent gait cycles of all normal people is balanced through each gait training trajectory to obtain the conversion balance cost of each gait conversion, and then the imbalance compensation coefficient of the target object when performing lower limb rehabilitation training is determined based on all the conversion balance costs; the initial gait trajectory of the target object is obtained, and then the difference correlation characteristics of the gait of the target object in each gait cycle are determined through the morphological differences between the target object and different normal people; the dependent path of the target object when performing gait conversion between adjacent gait cycles is extracted from the initial gait trajectory, and the trajectory of each dependent path is corrected through the imbalance compensation coefficient and each difference correlation characteristic to obtain an adaptive anti-imbalance trajectory; the lower limb rehabilitation training equipment of the target object performs personalized anti-imbalance adjustment according to the adaptive anti-imbalance trajectory.

[0016] It can be seen that the present application performs personalized anti-imbalance training for lower limb rehabilitation of the target object based on the adaptive anti-imbalance trajectory; firstly, by determining the imbalance compensation coefficient through each gait training trajectory, the adjustment parameter for compensating for the imbalance caused by unstable posture or uncoordinated gait of the target object during gait transition can be obtained. The determination of the imbalance compensation coefficient can help the patient better maintain balance, reduce the risk of imbalance or fall of the patient, enhance the stability of gait transition, and thus improve the effectiveness of the lower limb rehabilitation training equipment; secondly, by determining the difference correlation characteristics of the gait of the target object in each gait cycle, the key indicators describing the gait inconsistency between the initial gait trajectory of the target object and the simulated trajectory curve of a normal person can be obtained. The above steps can accurately locate the abnormal performance of the patient in the gait cycle, such as the joint angle in the gait cycle. The deviation of degree or inconsistency of step frequency can be corrected through these difference-related features, which can help patients gradually approach normal gait, thereby improving the accuracy and effectiveness of lower limb rehabilitation training; then, an adaptive anti-imbalance trajectory is obtained through trajectory correction, which can obtain a gait trajectory that can help patients avoid imbalance or falls during lower limb rehabilitation training. Trajectory correction solves the problem that a unified training model is difficult to meet the rehabilitation needs of different patients due to the large differences in gait characteristics of different patients; finally, the lower limb rehabilitation training equipment of the target object performs personalized anti-imbalance adjustment according to the adaptive anti-imbalance trajectory; in summary, based on the above scheme, personalized anti-imbalance rehabilitation training can be achieved according to the differences in different rehabilitation training needs of different patients, thereby improving the effectiveness of anti-imbalance lower limb rehabilitation training equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1is an exemplary flow chart of an anti-unbalance adjustment method according to some embodiments of the present application; Figure 2 is an exemplary flow chart of determining a gait training trajectory according to some embodiments of the present application; Figure 3 is an exemplary flow chart of determining difference association features according to some embodiments of the present application; Figure 4 is a schematic diagram of exemplary hardware and / or software of a rehabilitation training control center according to some embodiments of the present application; Figure 5 It is a structural schematic diagram of a computer device for implementing an anti-imbalance adjustment method according to some embodiments of the present application. DETAILED DESCRIPTION

[0018] The embodiment of the present application provides a lower limb rehabilitation training system and method for preventing imbalance, the core of which is to collect simulation training data; extract the gait training trajectories of all normal people in each gait cycle from the simulation training data, and determine the imbalance compensation coefficient of the target object when performing lower limb rehabilitation training through each gait training trajectory; obtain the initial gait trajectory of the target object, and determine the difference correlation characteristics of the gait of the target object in each gait cycle according to the initial gait trajectory; correct the trajectory of the dependent path of the target object when performing gait conversion between adjacent gait cycles through the imbalance compensation coefficient and each difference correlation feature, and obtain an adaptive anti-imbalance trajectory; and perform personalized anti-imbalance training on the target object according to the adaptive anti-imbalance trajectory. By adopting the scheme of the present application, personalized anti-imbalance rehabilitation training can be realized according to the differences in different rehabilitation training needs of different patients, thereby improving the effectiveness of the anti-imbalance lower limb rehabilitation training equipment.

[0019] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. Figure 1 , which is an exemplary flow chart of an anti-unbalance adjustment method according to some embodiments of the present application. The anti-unbalance adjustment method 100 mainly includes the following steps: In step 101, a lower limb rehabilitation training device is started to collect training records of different normal people performing lower limb rehabilitation simulation training, thereby obtaining simulation training data.

[0020] It should be noted that, in the present application, the simulated training data is a collection of training records of all normal people undergoing simulated rehabilitation training, and the training records contain gait information and body information of the normal people during simulated rehabilitation training. The gait information includes the normal people's walking speed, step length, joint angles and ground reaction force during walking, and the body information includes various body parameters of normal people, such as gender, age, height, weight, hip width, thigh length, knee width, calf length and foot length, among which a normal person refers to a person who has no physical injuries and can move his lower limbs normally.

[0021] In specific implementation, twenty normal female people and twenty normal male people can be selected. For each normal person, the body information of the normal person is first collected through existing measuring equipment (such as a tape measure), and then the normal person is asked to perform 20 simulated rehabilitation trainings with different gait cycles, and the gait information of the normal person is collected through an existing motion capture system (for example: NOKOV motion capture system). The collection of body information and gait information can be used as a training record of the normal person. The training records of different normal people can be obtained in the above manner.

[0022] It should be noted that, in the present application, stride length refers to the horizontal distance between the first contact and the second contact of the same side foot with the ground; gait cycle refers to a complete gait process, including the entire process from the first contact of the unilateral heel to the second contact of the same side heel; joint angles refer to the knee joint angle, hip joint angle and ankle joint angle. In the present application, only the knee joint angle is processed, and other joint angles can be processed in the same way as the knee joint angle.

[0023] In step 102, the gait training trajectories of all normal people in each gait cycle are extracted from the simulated training data, and the gait transitions of all normal people between adjacent gait cycles are evaluated for balance through each gait training trajectory to obtain the conversion balance cost of each gait transition, and then the imbalance compensation coefficient of the target object during lower limb rehabilitation training is determined based on all the conversion balance costs.

[0024] In some embodiments, reference Figure 2 , which is an exemplary flow chart of determining gait training trajectories according to some embodiments of the present application. In the present application, the gait training trajectories of all normal people in each gait cycle can be extracted from the simulated training data by the following steps: In step 1021, all gait cycles in the simulated rehabilitation training are obtained; In step 1022, for each gait cycle, gait information of each normal person in the gait cycle is extracted from the simulation training data as periodic gait information; In step 1023, data fitting is performed on the gait information of each normal person in the periodic gait information to obtain a trajectory curve of each normal person; In step 1024, the gait training trajectories of all normal persons in the gait cycle are determined according to all the trajectory curves, and then the gait training trajectories of all normal persons in each gait cycle are obtained.

[0025] It should be noted that, in the present application, periodic gait information refers to a data set of gait information fragments within different gait cycles, and the gait information fragments refer to the gait information that becomes more accurate and smooth after denoising and can truly reflect the gait characteristics of a normal person; the gait training trajectory is a collection of trajectory curves of all normal people, and the trajectory curve refers to the motion path of a normal person's joint angle changes in different gait cycles.

[0026] In the specific implementation, first, all gait cycles in the simulated rehabilitation training are obtained; secondly, for each gait cycle, the gait information of each normal person in the gait cycle is extracted from the simulated training data, and then the data with a frequency higher than 6 Hz in each gait information is removed through an existing filter (such as a low-pass filter), and the filtered data is used as the gait information fragment corresponding to the normal person, and then the collection of the gait information fragments of all normal people is used as the periodic gait information; then, a normal person is selected from all normal people as the selected normal person, and the existing data fitting model (such as a linear regression model) is used to fit the data set of the joint angles of the selected normal person in the periodic gait information into a trajectory curve, and this trajectory curve is used as the trajectory curve of the selected normal person, and the gait information of the remaining normal people is continued to be fitted with data to obtain the trajectory curve of each normal person in the gait cycle; finally, the collection of the trajectory curves of all normal people is used as the gait training trajectory in the gait cycle, and the gait training trajectory of all normal people in each gait cycle can be obtained in the above manner.

[0027] In some embodiments, the following steps may be used to perform a balance assessment on the gait transitions of all normal people between adjacent gait cycles through various gait training trajectories to obtain the transition balance cost at each gait transition: For each gait transition, the gait transition point between adjacent gait cycles of each normal person is determined according to the gait training trajectory in the gait cycle before the gait transition; Obtaining the ground reaction force of each normal person at the gait transition point; The cost of each gait transition point is evaluated according to all ground reaction forces, and the transition balance cost of gait transition between adjacent gait cycles of all normal people is obtained, and then the transition balance cost of each gait transition is obtained.

[0028] It should be noted that, in the present application, the gait transition point is the transition node where a normal person transitions from the previous gait cycle to the next gait cycle; the ground reaction force refers to the reverse force exerted by the ground on the lower limbs of a normal person when the lower limbs contact the ground; the transition balance cost is the force required for a normal person to maintain body balance when performing gait transitions between adjacent gait cycles. The greater the transition balance cost, the more difficult it is for a normal person to maintain an imbalance when performing a gait transition, and the smaller the transition balance cost, the less difficult it is for a normal person to maintain an imbalance when performing a gait transition.

[0029] In specific implementation, first, for each gait conversion, a normal person is selected from all normal persons as the selected normal person, and a trajectory curve in the gait training trajectory of the selected normal person in the gait cycle before the gait conversion is obtained, and an existing mathematical method (such as a quadratic difference method) is used to detect the inflection point in the trajectory curve, and the inflection point close to the tail end of the trajectory curve is used as the gait conversion point between adjacent gait cycles of the selected normal person, and the gait conversion points of the remaining normal persons between adjacent gait cycles are continuously determined to obtain the gait conversion point of each normal person between adjacent gait cycles; then, the ground reaction force received by each normal person at the gait conversion point is obtained in the simulated training data; Finally, a normal person is selected from all the normal persons as the selected normal person. The calf length of the selected normal person is first obtained in the simulated training data. The cosine of the knee joint angle of the selected normal person at the corresponding gait transition point and the calf length of the selected normal person can be multiplied by the ground reaction force, and the product obtained is used as the balance torque of the selected normal person when performing gait transition. The balance torque of the remaining normal persons when performing gait transition is further determined, and the balance torque of each normal person when performing gait transition is obtained. Then the average value of all balance torques is used as the conversion balance cost of all normal people when performing gait transition between adjacent gait cycles, and the conversion balance cost of each gait transition is obtained.

[0030] In some embodiments, determining the imbalance compensation coefficient of the target object when performing lower limb rehabilitation training based on all conversion balance costs can be implemented by the following steps: Determine the balance threshold of all normal subjects during simulated rehabilitation training; Determine multiple balance differences when the target object performs lower limb rehabilitation training according to the balance threshold and all conversion balance costs; The imbalance compensation coefficient of the target subject during lower limb rehabilitation training is determined based on all balance differences.

[0031] It should be noted that in the present application, the imbalance compensation coefficient is an adjustment parameter used to compensate for the imbalance caused by unstable posture or uncoordinated gait of the target object during gait transition. It can help patients better maintain balance, reduce the risk of imbalance or falling of patients, enhance the stability of gait transition, and thus improve the rehabilitation effect.

[0032] In specific implementation, first, the average value of all transition balance costs can be used as the balance threshold for normal people during simulated rehabilitation training. The balance threshold is a key parameter for evaluating the gait stability of normal people. If the transition balance cost of the patient during gait transition during lower limb rehabilitation training exceeds the balance threshold, it means that there is a high risk of imbalance at the gait transition point; then, the ratio of the difference between each transition balance cost and the balance threshold to the balance threshold is used as the balance difference of the corresponding transition balance cost, and the balance difference refers to the deviation between the balance state of a normal person and the preset balance threshold during each gait transition; finally, the sum of all balance differences is divided by the total number of balance differences, and the quotient obtained is used as the imbalance compensation coefficient for the target object during lower limb rehabilitation training.

[0033] In step 103, the initial gait trajectory of the target object during lower limb rehabilitation training is obtained, and the trajectory constraints of the target object and all normal people in each gait cycle are determined based on the initial gait trajectory, and then the difference correlation characteristics of the target object's gait in each gait cycle are determined through the morphological differences between the target object and different normal people and all trajectory constraints.

[0034] It should be noted that in the present application, the initial gait trajectory is a trajectory showing changes in joint angles used to guide the target object to perform lower limb rehabilitation training; in specific implementation, various body parameters of the target object can be obtained from the target object's diagnostic opinion, and then a gait generation model trained using the gait information and body information of all healthy subjects in the simulated training data is loaded (such as a long short-term memory network plus two fully connected layers), and the various body parameters of the target object are used as the input of the gait generation model. After executing the gait generation model, the output of the second fully connected layer of the gait generation model is used as the initial gait trajectory of the target patient during lower limb rehabilitation training.

[0035] In some embodiments, determining the trajectory constraints of the target object and all normal people in each gait cycle according to the initial gait trajectory can be achieved by using the following steps: For each gait cycle, obtaining a periodic gait trajectory under the gait cycle in the initial gait trajectory; Determine the trajectory residual of the target object and the corresponding normal person in the gait cycle according to the cycle gait trajectory and the trajectory curve of each normal person in the gait training trajectory; The trajectory constraints between the target object and all normal people in the gait cycle are determined based on all trajectory residuals, and then the trajectory constraints between the target object and all normal people in each gait cycle are obtained.

[0036] It should be noted that, in the present application, the periodic gait trajectory refers to the trajectory of the initial gait trajectory in different gait cycles; the trajectory residual refers to the degree of deviation between the initial gait trajectory of the target object and the gait training trajectory of a normal person in terms of joint angles; the trajectory constraint is a parameter that describes the gait inconsistency between the initial gait trajectory of the target object and the gait training trajectory of a normal person.

[0037] In the specific implementation, first, the initial gait trajectory of the target object is divided according to the gait cycle, and the trajectory under each gait cycle is used as the periodic gait trajectory in the corresponding gait cycle; then, for each gait cycle, the periodic gait trajectory under the gait cycle is first aligned with the trajectory curve of each normal person in the gait training trajectory according to the gait cycle, and the length of the gait cycle is uniformly divided into 100 discrete values ​​using the existing method of unifying data length (such as linear interpolation), and then a normal person is selected from all normal people as the selected normal person, and the average value of the difference between the knee joint angle of the periodic gait trajectory and the trajectory curve of the selected normal person at each discrete value is used as the trajectory residual of the target object and the selected normal person in the gait cycle, and the trajectory residual of the target object and the remaining normal people in the gait cycle is further determined to obtain the trajectory residual of the target object and each normal person in the gait cycle; finally, the set of all trajectory residuals is used as the trajectory constraint of the target object and the normal person in the gait cycle, and the trajectory constraint of the target object and the normal person in each gait cycle can be obtained through the above steps.

[0038] In some embodiments, reference Figure 3 , which is an exemplary flow chart of determining difference-related features according to some embodiments of the present application. In the present application, the difference-related features of the gait of the target object in each gait cycle are determined by the morphological differences between the target object and different normal people and all trajectory constraints, which can be implemented by the following steps: In step 1031, morphological differences between the target object and different normal persons are determined; In step 1032, for each gait cycle, trajectory constraints of the target object and all normal people in the gait cycle are obtained; In step 1033, the feature coupling degree between the target object and the corresponding normal person is determined according to the trajectory constraint and the morphological difference between the target object and each normal person; In step 1034, the difference correlation features of the target object in the gait cycle are determined based on all the feature coupling degrees, and then the difference correlation features of the gait of the target object in each gait cycle are obtained.

[0039] It should be noted that, in the present application, morphological difference refers to the degree of difference between various physical parameters of the target object and normal people; feature coupling degree is a key parameter that describes the degree of correlation between the target object and normal people in physical parameters and gait trajectory; difference correlation feature is a feature that describes the overall difference between the target object and normal people in different gait cycles.

[0040] In the specific implementation, first, for each normal person, the square root of the sum of the squares of the differences between various physical parameters of the target object and the corresponding physical parameters in the physical information of the normal person is taken, and the obtained value is used as the morphological difference between the target object and the normal person. The morphological difference between the target object and different normal people can be obtained in the above manner; secondly, for each gait cycle, the trajectory constraints of the target object and the normal person in the gait cycle are obtained; then, a normal person is selected from all the normal people as the selected normal person, and the ratio of the trajectory residual of the selected normal person in the trajectory constraint and the morphological difference between the target object and the selected normal person is used as the feature coupling degree between the target object and the selected normal person, and the feature coupling degree between the target object and the remaining normal people is further determined to obtain the feature coupling degree between the target object and each normal person; finally, the set of the feature coupling degrees between the target object and each normal person is used as the difference correlation feature of the gait of the target object in the gait cycle. The difference correlation feature of the gait of the target object in each gait cycle can be obtained through the above steps.

[0041] In step 104, dependent paths of the target object when performing gait transitions between adjacent gait cycles are extracted from the initial gait trajectory, and trajectory correction is performed on each dependent path using the imbalance compensation coefficient and each difference-related feature to obtain an adaptive anti-imbalance trajectory.

[0042] In some embodiments, extracting the dependent path of the target object when performing gait transitions between adjacent gait cycles in the initial gait trajectory can be achieved by using the following steps: Determining a gait switching point of the target object in each gait cycle according to the initial gait trajectory; The initial gait trajectory between the gait switching points in every two adjacent gait cycles is extracted as the dependent path when the target object performs gait transition between adjacent gait cycles.

[0043] It should be noted that, in the present application, the gait switching point is the key point at which the target object performs gait transition from the previous gait cycle to the next gait cycle; the dependent path refers to the motion trajectory that captures the dynamic changes of the target object's joint angles when the target object transitions from one gait cycle to the next gait cycle; in specific implementation, first, for each gait cycle, the existing mathematical method (such as the quadratic difference method) can be used to detect the inflection point of the initial gait trajectory within the gait cycle, and the inflection point close to the end of the initial gait trajectory is used as the gait switching point of the target object in the gait cycle. The gait switching point of the target object in each gait cycle can be obtained through the above steps; then, for each gait cycle, the initial gait trajectory of the target object between the gait switching point of the gait cycle and the gait switching point of the next gait cycle can be extracted as the dependent path when the target object performs gait transition between the gait cycle and the adjacent gait cycle. The dependent path when the target object performs gait transition between each gait cycle and the adjacent gait cycle can be obtained through the above steps.

[0044] It should be noted that the gait cycles in the present application are arranged adjacently in chronological order.

[0045] In some embodiments, the trajectory correction of each dependent path is performed by using the imbalance compensation coefficient and each difference correlation feature to obtain an adaptive anti-imbalance trajectory, which can be achieved by the following steps: For each gait cycle, obtain the dependency path of the target object between the gait cycle and the next gait cycle; Correcting the dependent path using the difference correlation feature of the target object in the next gait cycle and the imbalance compensation coefficient to obtain an anti-imbalance trajectory of the dependent path between the gait cycle and the next gait cycle; Then, an anti-imbalance trajectory that depends on the path between each gait cycle and the next gait cycle is obtained; The adaptive anti-unbalance trajectory of the target object is obtained through all anti-unbalance trajectories.

[0046] It should be noted that the adaptive anti-imbalance trajectory in the present application is a gait trajectory preset based on the initial gait trajectory of the target object, which can help patients maintain balance and avoid imbalance or falls during lower limb rehabilitation training; the anti-imbalance trajectory is a correction trajectory preset based on the imbalance compensation coefficient and the difference correlation characteristics of the target object when the target object performs gait conversion.

[0047] In the specific implementation, first, for each gait cycle, the dependent path of the target object between the gait cycle and the next gait cycle is obtained; secondly, a dynamic model (such as a forward dynamic model) is loaded as a trajectory correction model, and the difference correlation feature of the target object in the next gait cycle adjacent to the gait cycle is obtained, and the sum of the feature coupling degree in the difference correlation feature and the knee joint angle in the gait transfer trajectory is used as the angle limit of the knee joint in the system parameters input to the trajectory correction model, and the imbalance compensation coefficient is used as the moment of inertia in the system parameters input to the trajectory correction model. After executing the trajectory correction model, the trajectory of the knee joint in the output of the trajectory correction model is used as the anti-imbalance trajectory of the dependent path. The anti-imbalance trajectory of the dependent path between each gait cycle and the next gait cycle can be obtained through the above steps; finally, each anti-imbalance trajectory replaces the corresponding dependent path in the initial gait trajectory, and the obtained new gait trajectory is used as the adaptive anti-imbalance trajectory of the target object.

[0048] In step 105, the lower limb rehabilitation training device of the target object performs personalized anti-imbalance adjustment according to the adaptive anti-imbalance trajectory.

[0049] In specific implementation, the lower limb rehabilitation training equipment of the target object performs personalized anti-imbalance adjustment according to the adaptive anti-imbalance trajectory, which uses the adaptive anti-imbalance trajectory as the motion trajectory of the target object during lower limb rehabilitation training, thereby completing personalized anti-imbalance training for lower limb rehabilitation of the target object.

[0050] In addition, in another aspect of the present application, in some embodiments, the present application provides an anti-imbalance lower limb rehabilitation training system, the anti-imbalance lower limb rehabilitation training system includes a lower limb rehabilitation training device and a rehabilitation training control center, referring to Figure 4 , which is a schematic diagram of exemplary hardware and / or software of a rehabilitation training control center according to some embodiments of the present application, the rehabilitation training control center 400 includes: a collection module 401, a processing module 402 and an execution module 403, which are described as follows: The acquisition module 401 in this application is mainly used to collect training records of different normal people during lower limb rehabilitation simulation training, and then obtain simulation training data; Processing module 402, in the present application, is mainly used to extract gait training trajectories of all normal people in each gait cycle from the simulated training data, perform balance evaluation on the gait transitions of all normal people between adjacent gait cycles through each gait training trajectory, obtain the conversion balance cost at each gait transition, and then determine the imbalance compensation coefficient of the target object when performing lower limb rehabilitation training based on all the conversion balance costs; It should be noted that the processing module 402 in the present application is also used to obtain the initial gait trajectory of the target object when performing lower limb rehabilitation training, determine the trajectory constraints of the target object and all normal people in each gait cycle according to the initial gait trajectory, and then determine the difference correlation characteristics of the gait of the target object in each gait cycle through the morphological differences between the target object and different normal people and all trajectory constraints; In addition, it should be noted that the processing module 402 in the present application is also used to extract the dependent path of the target object when performing gait conversion between adjacent gait cycles in the initial gait trajectory, and perform trajectory correction on each dependent path through the imbalance compensation coefficient and each difference association feature to obtain an adaptive anti-imbalance trajectory; The execution module 403 in the present application is mainly used to instruct the lower limb rehabilitation training equipment of the target object to perform personalized anti-imbalance adjustment according to the adaptive anti-imbalance trajectory.

[0051] In addition, the present application also provides a computer device, which includes a memory and a processor, wherein the memory stores codes, and the processor is configured to obtain the codes and execute the above-mentioned anti-unbalance adjustment method.

[0052] In some embodiments, reference Figure 5 , which is a schematic diagram of the structure of a computer device for implementing an anti-unbalance adjustment method according to some embodiments of the present application. The anti-unbalance adjustment method in the above embodiment can be Figure 5 The computer device 500 shown in the figure is implemented, and the computer device 500 includes at least one processor 501, a communication bus 502, a memory 503 and at least one communication interface 504.

[0053] The processor 501 may be a general-purpose central processing unit (CPU) or an application specific integrated circuit (ASIC).

[0054] The communication bus 502 may be used to transmit information between the above-mentioned components.

[0055] The memory 503 may be a read only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read only memory (EEPROM), a compact disc read only memory (CD ROM) or other optical disc storage, an optical disc storage (including a compressed optical disc, a laser disc, an optical disc, a digital versatile disc, a Blu-ray disc, etc.), a magnetic disk or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The memory 503 may exist independently and be connected to the processor 501 via the communication bus 502. The memory 503 may also be integrated with the processor 501.

[0056] The memory 503 is used to store the program code for executing the solution of the present application, and the execution is controlled by the processor 501. The processor 501 is used to execute the program code stored in the memory 503. The program code may include one or more software modules. The anti-unbalance adjustment method in the above embodiment can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0057] The communication interface 504 uses any transceiver or other device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0058] In a specific implementation, as an embodiment, a computer device may include multiple processors, each of which may be a single-core (single CPU) processor or a multi-core (multi CPU) processor. The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (such as computer program instructions).

[0059] The above-mentioned computer device may be a general-purpose computer device or a special-purpose computer device. In a specific implementation, the computer device may be a desktop computer, a portable computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device or an embedded device. The embodiment of the present application does not limit the type of computer device.

[0060] In addition, the present application also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned anti-unbalance adjustment method is implemented.

[0061] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0062] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. An anti-imbalance adjustment method for anti-imbalance lower limb rehabilitation training equipment, characterized in that: The steps include: Collect the training records of different normal people during lower limb rehabilitation simulation training, and then obtain simulation training data; Extracting gait training trajectories of all normal people in each gait cycle from the simulated training data, performing balance evaluation on gait transitions of all normal people between adjacent gait cycles through each gait training trajectory, obtaining a conversion balance cost at each gait transition, and then determining an imbalance compensation coefficient when the target object performs lower limb rehabilitation training based on all conversion balance costs; Acquire the initial gait trajectory of the target object when performing lower limb rehabilitation training, determine the trajectory constraints of the target object and all normal people in each gait cycle according to the initial gait trajectory, and then determine the difference correlation characteristics of the gait of the target object in each gait cycle through the morphological differences between the target object and different normal people and all trajectory constraints; Extracting dependent paths when the target object performs gait conversion between adjacent gait cycles from the initial gait trajectory, performing trajectory correction on each dependent path by using the imbalance compensation coefficient and each difference association feature, and obtaining an adaptive anti-imbalance trajectory; The lower limb rehabilitation training equipment of the target object performs personalized anti-imbalance adjustment according to the adaptive anti-imbalance trajectory.

2. The method according to claim 1, characterized in that Extracting the gait training trajectories of all normal people in each gait cycle from the simulation training data specifically includes: Acquire all gait cycles during simulated rehabilitation training; For each gait cycle, extracting gait information of each normal person in the gait cycle from the simulation training data as periodic gait information; Performing data fitting on the gait information of each normal person in the periodic gait information to obtain a trajectory curve of each normal person; The gait training trajectories of all normal people in the gait cycle are determined according to all the trajectory curves, and then the gait training trajectories of all normal people in each gait cycle are obtained.

3. The method according to claim 1, characterized in that Through each gait training trajectory, the gait transition between adjacent gait cycles of all normal people is evaluated for balance, and the conversion balance cost at each gait transition is obtained, which specifically includes: For each gait transition, the gait transition point between adjacent gait cycles of each normal person is determined according to the gait training trajectory in the gait cycle before the gait transition; Obtaining the ground reaction force of each normal person at the gait transition point; The cost of each gait transition point is evaluated according to all ground reaction forces, and the transition balance cost of gait transition between adjacent gait cycles of all normal people is obtained, and then the transition balance cost of each gait transition is obtained.

4. The method according to claim 1, characterized in that The imbalance compensation coefficients for the target subjects during lower limb rehabilitation training are determined based on all conversion balance costs, including: Determine the balance threshold of all normal subjects during simulated rehabilitation training; Determine multiple balance differences when the target object performs lower limb rehabilitation training according to the balance threshold and all conversion balance costs; The imbalance compensation coefficient of the target subject during lower limb rehabilitation training is determined based on all balance differences.

5. The method according to claim 1, characterized in that Determining the trajectory constraints of the target object and all normal people in each gait cycle according to the initial gait trajectory specifically includes: For each gait cycle, obtaining a periodic gait trajectory under the gait cycle in the initial gait trajectory; Determine the trajectory residual of the target object and the corresponding normal person in the gait cycle according to the cycle gait trajectory and the trajectory curve of each normal person in the gait training trajectory; The trajectory constraints between the target object and all normal people in the gait cycle are determined based on all trajectory residuals, and then the trajectory constraints between the target object and all normal people in each gait cycle are obtained.

6. The method according to claim 1, characterized in that The morphological differences between the target object and different normal people and all trajectory constraints determine the difference-related features of the target object's gait in each gait cycle, including: Determine the morphological differences between the target subject and different normal people; For each gait cycle, obtain the trajectory constraints of the target object and the normal person in the gait cycle; Determine the feature coupling degree between the target object and the corresponding normal person according to the trajectory constraint and the morphological difference between the target object and each normal person; Based on all the feature coupling degrees, the difference correlation features of the target object in the gait cycle are determined, and then the difference correlation features of the gait of the target object in each gait cycle are obtained.

7. The method according to claim 1, characterized in that The trajectory of each dependent path is corrected by using the imbalance compensation coefficient and each difference correlation feature to obtain an adaptive anti-imbalance trajectory, which specifically includes: For each gait cycle, obtain the dependency path of the target object between the gait cycle and the next gait cycle; Correcting the dependent path using the difference correlation feature of the target object in the next gait cycle and the imbalance compensation coefficient to obtain an anti-imbalance trajectory of the dependent path between the gait cycle and the next gait cycle; Then, an anti-imbalance trajectory that depends on the path between each gait cycle and the next gait cycle is obtained; The adaptive anti-unbalance trajectory of the target object is obtained through all anti-unbalance trajectories.

8. A lower limb rehabilitation training system for preventing imbalance, characterized in that: It includes lower limb rehabilitation training equipment and a rehabilitation training control center, and the rehabilitation training control center includes: The acquisition module is used to collect the training records of different normal people during lower limb rehabilitation simulation training, and then obtain simulation training data; a processing module, for extracting gait training trajectories of all normal persons in each gait cycle from the simulated training data, performing balance evaluation on gait transitions of all normal persons between adjacent gait cycles through each gait training trajectory, obtaining a conversion balance cost at each gait transition, and then determining an imbalance compensation coefficient when the target subject performs lower limb rehabilitation training based on all conversion balance costs; The processing module is also used to obtain the initial gait trajectory of the target object when performing lower limb rehabilitation training, determine the trajectory constraints of the target object and all normal people in each gait cycle according to the initial gait trajectory, and then determine the difference correlation characteristics of the gait of the target object in each gait cycle through the morphological differences between the target object and different normal people and all trajectory constraints; The processing module is further used to extract the dependent path of the target object when performing gait conversion between adjacent gait cycles from the initial gait trajectory, and perform trajectory correction on each dependent path by using the imbalance compensation coefficient and each difference association feature to obtain an adaptive anti-imbalance trajectory; The execution module is used to instruct the lower limb rehabilitation training equipment of the target object to perform personalized anti-imbalance adjustment according to the adaptive anti-imbalance trajectory.

9. A computer device, characterized in that: The computer device comprises a memory and a processor, wherein the memory stores codes, and the processor is configured to obtain the codes and execute the anti-unbalance adjustment method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the anti-unbalance adjustment method according to any one of claims 1 to 7 is implemented.

Images