A method and system for generating sports rehabilitation strategy based on Bayesian network

Abstract

The present invention relates to the technical field of strategy generation, and specifically discloses a method and system for generating a sports rehabilitation strategy based on a Bayesian network. The method comprises: acquiring the motion state data of a detection object, obtaining the sports rehabilitation needs of the detection object by comparison, constructing a Bayesian network model, associating the sports rehabilitation needs of the detection object with the current rehabilitation state of the detection object, obtaining the rehabilitation state of the object, acquiring the rehabilitation state feedback data of the detection object, and obtaining a personalized sports rehabilitation strategy by comparison in combination with the rehabilitation state of the detection object. The present invention solves the problem of singleness and universality in the current generation of rehabilitation exercise strategies for detection objects, helps to have a more comprehensive understanding of the rehabilitation state and rehabilitation needs of the detection objects, and accurate diagnosis and rehabilitation strategy generation help to reduce errors and uncertainties in the process.

Description

Technical Field

[0001] The present invention relates to the technical field of strategy generation, and in particular to a method and system for generating a sports rehabilitation strategy based on a Bayesian network. Background Art

[0002] The method and system for generating sports rehabilitation strategies based on Bayesian networks are designed to improve the effectiveness and personalization of rehabilitation treatment. Traditional rehabilitation treatment usually develops rehabilitation plans based on expert experience and clinical guidelines, which lack personalization and accuracy. The rehabilitation process involves complex associations of multiple factors, such as the subject's condition, physiological condition, motor ability, rehabilitation training effect, etc. It is difficult for traditional methods to fully consider the relationship between these factors. By constructing a rehabilitation Bayesian network model, researchers can use multi-source data such as the subject's clinical data, motor test results, rehabilitation training records, etc. to reveal the potential relationships and influencing factors between different factors in the rehabilitation process, thereby improving the effectiveness and efficiency of rehabilitation treatment.

[0003] Nowadays, there are still some shortcomings in the research on a method and system for generating exercise rehabilitation strategies based on Bayesian networks. Specifically, the current generation of rehabilitation exercise strategies for test subjects is single and universal. The singleness leads to the generated rehabilitation exercise strategies being limited to specific rehabilitation scenarios, lacking comprehensive consideration and treatment of different rehabilitation problems, and unable to provide comprehensive rehabilitation strategies. There is less personalized consideration of the individual characteristics and needs of the test subjects, and a lack of modeling and analysis of the personalized characteristics of the test subjects, resulting in the generated rehabilitation strategies may not conform to the individual differences of the test subjects, affecting the personalization and precision of the rehabilitation effects. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a method and system for generating a motor rehabilitation strategy based on a Bayesian network, which can effectively solve the problems involved in the above-mentioned background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: The first aspect of the present invention provides a method for generating a motion rehabilitation strategy based on a Bayesian network, comprising the following steps: acquiring motion state data of the detection object, and obtaining the motion rehabilitation needs of the detection object based on the acquired motion state data of the detection object; constructing a Bayesian network model based on the compared motion rehabilitation needs of the detection object, associating the motion rehabilitation needs of the detection object with the current rehabilitation status of the detection object, and obtaining the rehabilitation status of the object; acquiring rehabilitation status feedback data of the detection object, and obtaining a personalized motion rehabilitation strategy based on the acquired rehabilitation status feedback data of the detection object and the rehabilitation status of the detection object.

[0006] As a further method, the motion state data of the detection object specifically includes the motion ability data and the posture data of the detection object, wherein the motion ability data of the detection object includes the motion reaction speed of the detection object, the joint movement angle of the detection object's part requiring rehabilitation, and the maximum duration of the detection object's motion; the posture data of the detection object includes the ratio of the distribution proportion of the detection object's weight between the left and right feet, the spinal curvature angle of the detection object, and the gait speed of the detection object.

[0007] As a further method, based on the acquired motion state data of the detection object, the motion rehabilitation needs of the detection object are obtained by comparison. The specific analysis process is: based on the acquired motion state data of the detection object, a comprehensive analysis is performed to obtain the motion state evaluation value of the detection object, and the motion state evaluation value of the detection object is used as the analysis basis for comparing the motion rehabilitation needs of the detection object; the motion state evaluation value of the detection object is compared with the motion rehabilitation needs of the detection object corresponding to the motion state evaluation values ​​of each detection object stored in the database, so as to obtain the motion rehabilitation needs of the detection object corresponding to the motion state evaluation value of the detection object.

[0008] As a further method, the motion state evaluation value of the detection object is analyzed as follows:

[0009]

[0010] Where ω is the motion state evaluation value of the detected object, ω 1 is the motor ability evaluation value of the detected object, ω 2 is the posture evaluation value of the detected object, μ 1 is the weight factor of the motor ability evaluation value of the set detection object, μ 2 is the weight factor of the posture evaluation value of the set detection object, and e is a natural constant.

[0011] As a further method, a Bayesian network model is constructed based on the motor rehabilitation needs of the detected object obtained by comparison, and the motor rehabilitation needs of the detected object and the current rehabilitation status of the detected object are associated to obtain the rehabilitation status of the object. The specific analysis process is: determine the nodes in the Bayesian network, including the age, gender, and motor rehabilitation needs of the detected object, and combine the weight factors of the age, gender, and motor rehabilitation needs of the detected object stored in the database, associate the motor rehabilitation needs of the detected object with the rehabilitation status of the detected object based on the Bayesian network, construct a Bayesian network model, and obtain the rehabilitation status of the detected object based on the Bayesian network association results.

[0012] As a further approach, the subject's recovery status is monitored, specifically including motor function, pain level, and quality of life.

[0013] As a further method, the rehabilitation status feedback data of the test subject specifically includes the average pain duration of the test subject, the number of rehabilitation training sessions of the test subject, and the total duration of the rehabilitation training of the test subject.

[0014] As a further method, based on the acquired rehabilitation status feedback data of the detection object and combined with the rehabilitation status of the detection object, a personalized exercise rehabilitation strategy is obtained by comparison. The specific analysis process is: based on the acquired rehabilitation status feedback data of the detection object, a comprehensive analysis is performed to obtain the rehabilitation status feedback characteristic value of the detection object, and the rehabilitation status feedback characteristic value of the detection object is used as the analysis basis for comparing and obtaining the personalized exercise rehabilitation strategy; the rehabilitation status feedback characteristic value of the detection object and the rehabilitation status of the detection object are set as designated labels, and the designated labels are compared with the personalized exercise rehabilitation strategies corresponding to the designated labels stored in the database to obtain the personalized exercise rehabilitation strategy corresponding to the designated labels.

[0015] As a further method, the rehabilitation status feedback characteristic value of the detection object is analyzed as follows:

[0016]

[0017] In the formula, δ is the rehabilitation status feedback characteristic value of the test subject, TT is the average pain duration of the test subject, LN is the number of rehabilitation training times of the test subject, LT is the total rehabilitation training time of the test subject, and TT is the average pain duration of the test subject. 0 The defined pain duration of the test subjects stored in the database, LN 0 is the reference rehabilitation training times of the test object stored in the database, LT 0 is the total duration of the reference rehabilitation training of the test object stored in the database, θ 1 is the compensation factor of the average pain duration of the test subject, θ 2 is the compensation factor for the number of rehabilitation trainings of the set test object, θ 3 is the compensation factor for the total duration of rehabilitation training of the set test object, and e is a natural constant.

[0018] The second aspect of the present invention provides a motion rehabilitation strategy generation system based on a Bayesian network, including a motion rehabilitation demand comparison module for a detection object, a rehabilitation status analysis module for a detection object, and a personalized motion rehabilitation strategy acquisition module, wherein: the motion rehabilitation demand comparison module for a detection object is used to obtain the motion state data of the detection object, and obtain the motion rehabilitation demand of the detection object based on the acquired motion state data of the detection object by comparison; the rehabilitation status analysis module for a detection object is used to construct a Bayesian network model based on the motion rehabilitation demand of the detection object obtained by comparison, associate the motion rehabilitation demand of the detection object with the current rehabilitation status of the detection object, and obtain the rehabilitation status of the object; the personalized motion rehabilitation strategy acquisition module is used to obtain the rehabilitation status feedback data of the detection object, and obtain a personalized motion rehabilitation strategy based on the acquired rehabilitation status feedback data of the detection object and the rehabilitation status of the detection object.

[0019] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) The present invention provides a method and system for generating a motor rehabilitation strategy based on a Bayesian network, which obtains the motor ability data of the subject and the posture data of the subject by comprehensive acquisition, and obtains the motor rehabilitation needs of the subject by comparison. It can combine the motor ability data and posture data of the subject for comprehensive analysis, and comprehensively evaluate the rehabilitation needs of the subject. By jointly analyzing the data on motor ability and posture, it can more comprehensively understand the rehabilitation status and rehabilitation needs of the subject, thereby providing a more accurate and comprehensive information basis for the generation of rehabilitation strategies, and can also accurately judge their rehabilitation needs. Accurate diagnosis and generation of rehabilitation strategies help to improve the effect of rehabilitation treatment and reduce errors and uncertainties in the process.

[0020] (2) The present invention can evaluate the past rehabilitation progress of the test subject through the rehabilitation status feedback data of the test subject, which helps to more accurately evaluate the rehabilitation effect of the test subject, so as to improve the effect and efficiency of the exercise rehabilitation strategy. The rehabilitation status feedback data can reflect the individual characteristics and needs of the test subject. By analyzing these data, the rehabilitation status of the test subject can be better understood, so as to adjust the rehabilitation plan in an individualized manner. According to the specific situation of the feedback data, a personalized exercise rehabilitation strategy that better meets the rehabilitation needs of the test subject can be formulated.

[0021] (3) The present invention obtains personalized exercise rehabilitation strategies based on the rehabilitation status feedback data of the detected object in combination with the rehabilitation status of the detected object. By comparing the rehabilitation status feedback data and the rehabilitation status, the rehabilitation needs and progress of the detected object can be more accurately understood, and personalized exercise rehabilitation strategies can be tailored for each detected object. According to the specific situation and needs of the detected object, the system can generate personalized rehabilitation plans that meet their individual characteristics and rehabilitation requirements, thereby improving the pertinence and effectiveness of rehabilitation treatment. By comparing the rehabilitation status feedback data and the rehabilitation status, the detected object can also have a clearer understanding of his or her own rehabilitation progress and effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0023] Figure 1 The figure is a schematic flow chart of the method steps of the present invention.

[0024] Figure 2 It is a schematic diagram of system module connection of the present invention.

[0025] Figure 3 An image showing how the recovery status of the detected subject is fed back with characteristic values ​​as the average pain duration of the detected subject changes. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only 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 ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0027] Reference Figure 1 As shown, the first aspect of the present invention provides a method for generating a motor rehabilitation strategy based on a Bayesian network, comprising: acquiring the motion state data of the detection object, and obtaining the motion rehabilitation needs of the detection object based on the acquired motion state data of the detection object.

[0028] Specifically, the motion state data of the detected object includes the motion ability data and posture data of the detected object, wherein the motion ability data of the detected object includes the motion reaction speed of the detected object, the joint activity angle of the detected object's part requiring rehabilitation, and the maximum duration of the detected object's exercise. The motion reaction speed of the detected object refers to the reaction speed exhibited by the detected object when performing a certain exercise. In rehabilitation treatment, monitoring and evaluating the motion reaction speed of the detected object can help medical staff understand the functional state of the detected object's neuromuscular system and the detected object's adaptability to rehabilitation training. By monitoring the motion reaction speed, the motion coordination, reaction sensitivity and motion execution ability of the detected object can be evaluated, thereby guiding The progress and direction of rehabilitation treatment, the joint motion angle of the subject's part that needs rehabilitation refers to monitoring and recording the joint motion angle of the injured or damaged part of the subject during rehabilitation treatment. By monitoring the joint motion angle, medical staff can evaluate the functional status, power balance and range of joint motion of the damaged part of the subject. Understanding the joint motion angle of the subject can help design a personalized rehabilitation plan, restore the normal range of joint motion, and enhance the stability and function of the joint. The maximum duration of the subject's exercise refers to the longest time the subject can continue to exercise during rehabilitation training or other sports activities. Measuring the maximum duration of exercise can help evaluate the subject's endurance and stamina level and understand the subject's response to exercise. Adaptability and physical condition. According to the maximum duration of the subject's exercise, medical staff can adjust the intensity and duration of rehabilitation training to help the subject recover function and improve physical fitness. The subject's posture data includes the ratio of the subject's weight distribution in the left and right feet, the subject's spinal curvature angle, and the subject's gait speed. The distribution ratio of the subject's weight in the left and right feet is used to evaluate the distribution of the subject's weight in the left and right feet when the subject is standing and walking. Under normal circumstances, the weight should be evenly distributed on both feet, but some people may have an unbalanced weight distribution problem on the left and right feet. This may be caused by incorrect posture, muscle imbalance, joint problems, etc. By measuring the weight distribution on the left and right feet By measuring the distribution ratio of the spine, we can understand whether the standing and walking posture of the test subject is normal, and whether relevant adjustments and rehabilitation training are needed. The spinal curvature angle of the test subject is used to evaluate the curvature of the spine of the test subject, which mainly refers to the physiological curvature and anatomical curvature of the spine. A normal spine should have an appropriate physiological curvature, including the curvature of the cervical spine, thoracic spine and lumbar spine. If the spinal curvature angle is abnormal, it may lead to incorrect posture, back pain, joint problems, etc. By measuring the spinal curvature angle, it can help evaluate and monitor the health of the spine and provide guidance for rehabilitation treatment. The gait speed of the test subject is used to evaluate the speed and gait characteristics of the test subject when walking. The gait speed can reflect the athletic ability and gait stability of the test subject.It is also an important indicator for evaluating the effectiveness of rehabilitation treatment. A gait speed that is too fast or too slow may indicate a certain degree of problem, such as muscle weakness, balance disorders, neurological problems, etc. By measuring gait speed, medical staff can understand the gait characteristics of the test subject, evaluate gait stability, and develop corresponding rehabilitation training plans to help the test subject improve gait safety and exercise ability.

[0029] In a specific embodiment, the movement reaction speed can reflect the functional state of the neuromuscular system of the test subject, which is crucial to the effect of rehabilitation treatment. The measurement of the joint movement angle can help understand the functional state and range of motion of the damaged part of the test subject. If the movement reaction speed of the test subject is fast, it will usually promote the improvement of the joint movement angle to a certain extent. The fast movement reaction speed helps to improve the coordination and sensitivity of movement, thereby providing better support for the joint movement angle of the damaged part. The movement reaction speed and the maximum duration of movement are both indicators of the movement ability and adaptability of the test subject. The fast movement reaction speed is usually related to better movement coordination and execution ability, which may cause the test subject to maintain a longer duration during exercise. Therefore, the faster movement reaction speed may have a positive effect on the maximum duration of the test subject's movement. The joint movement angle of the damaged part of the test subject can affect their exercise efficiency and endurance. If the joint range of motion is limited or the joint is unstable, the test subject may be in the process of progress. Feeling pain or discomfort during long-term exercise may reduce the maximum duration of exercise. By improving the range of motion and stability of joints, the endurance and exercise efficiency of the test subject may be improved, and the duration of exercise may be increased. Different weight distribution ratios between the left and right feet may affect the angle of spinal curvature. If a person habitually concentrates his weight on one foot, it may cause abnormal spinal curvature, such as deviation and scoliosis. This unbalanced weight distribution may cause additional pressure on the spine and affect the health and posture of the spine. The imbalance of weight between the left and right feet may affect the gait speed. If a person's weight is unevenly distributed when walking, it may cause an unstable gait and uneven walking speed. Such an abnormal gait may affect the gait speed, thereby affecting the walking ability and efficiency of the test subject. The curvature angle of the spine may affect the gait speed. If the spinal curvature is abnormal, it may cause incorrect or unstable body posture, which in turn affects the stability and speed of the gait. Certain spinal problems may cause abnormal gait, reduce gait speed and increase gait instability.

[0030] Specifically, based on the acquired motion state data of the detection object, the motion rehabilitation needs of the detection object are obtained by comparison. The specific analysis process is: based on the acquired motion state data of the detection object, a comprehensive analysis is performed to obtain the motion state evaluation value of the detection object, and the motion state evaluation value of the detection object is used as the analysis basis for comparing the motion rehabilitation needs of the detection object; the motion state evaluation value of the detection object is compared with the motion rehabilitation needs of the detection object corresponding to the motion state evaluation values ​​of each detection object stored in the database, so as to obtain the motion rehabilitation needs of the detection object corresponding to the motion state evaluation value of the detection object.

[0031] It needs to be explained that the above-mentioned method obtains the motor ability data of the object and the posture data of the test object by comprehensive comparison, obtains the motor rehabilitation needs of the test object, and can conduct a comprehensive analysis in combination with the motor ability data and posture data of the test object to comprehensively evaluate the rehabilitation needs of the test object. By jointly analyzing the data on motor ability and posture, a more comprehensive understanding of the rehabilitation status and rehabilitation needs of the test object can be obtained, thereby providing a more accurate and comprehensive information basis for the generation of rehabilitation strategies, and accurately judging their rehabilitation needs. Accurate diagnosis and generation of rehabilitation strategies can help improve the effectiveness of rehabilitation treatment and reduce errors and uncertainties in the process.

[0032] In a specific embodiment, by evaluating the movement state of the test subject and comparing it with the reference values ​​in the database, a more personalized and targeted rehabilitation plan can be formulated to accurately meet the rehabilitation needs of the test subject. The movement state evaluation values ​​obtained by comparison can help to promptly discover the problems and progress of the test subject, and adjust the rehabilitation plan in time to ensure the rehabilitation effect. The rehabilitation needs determined according to the movement state evaluation values ​​can carry out targeted rehabilitation training according to the symptoms and problems of the test subject, helping the test subject to improve his movement state and improve the rehabilitation effect.

[0033] Specifically, the motion state evaluation value of the detection object is analyzed as follows:

[0034]

[0035] Where ω is the motion state evaluation value of the detected object, ω 1 is the motor ability evaluation value of the detected object, ω 2 is the posture evaluation value of the detected object, μ 1 is the weight factor of the motor ability evaluation value of the set detection object, μ 2 is the weight factor of the posture evaluation value of the set detection object, and e is a natural constant.

[0036] It should be explained that the above-mentioned motion state evaluation value of the test object is calculated by the motion ability evaluation value of the test object and the posture evaluation value of the test object. The motion ability evaluation value and the posture evaluation value focus on the motor function and the body posture respectively. By combining the evaluation of these two aspects, a comprehensive evaluation of the motion state of the test object can be obtained, so that its overall motor function and posture control ability can be better understood. Motor ability and posture evaluation are commonly used evaluation methods in the field of sports rehabilitation. Combining them can provide a more comprehensive motion state evaluation, which can better identify potential movement disorders or posture problems and provide a reference for the formulation of personalized rehabilitation plans. By combining multiple evaluation indicators, the deviation of the evaluation results can be reduced and the accuracy of the evaluation can be improved. In this way, the motion state of the test object can be more reliably determined, and more accurate guidance can be provided for the treatment and rehabilitation process. The motion state evaluation value can help medical professionals better understand the specific situation of the test object, so as to formulate personalized rehabilitation plans and treatment plans. By adjusting the treatment measures in a targeted manner, the rehabilitation and motor function recovery of the test object can be more effectively promoted. By regularly evaluating the motion state, the rehabilitation progress and treatment effect of the test object can be tracked, so that the rehabilitation plan can be adjusted in time and the effectiveness of the treatment plan can be evaluated to ensure that the test object obtains the best rehabilitation results.

[0037] It should be explained that the specific analysis process of the above-mentioned test object's motor ability evaluation value is as follows:

[0038]

[0039] In the formula, ω 1 is the motor ability evaluation value of the test subject, FV is the motor reaction speed of the test subject, GJ is the joint activity angle of the test subject's part that needs rehabilitation, YC is the maximum duration of the test subject's exercise, and FV 0 is the motion reaction speed of the detected object stored in the database, GJ 0 The reference joint activity angle of the part of the object to be rehabilitated stored in the database, YC 0 is the motion reference duration of the detected object stored in the database, σ 1 is the compensation factor of the motion reaction speed of the set detection object, σ 2 is the compensation factor of the joint activity angle of the part of the test object that needs rehabilitation, σ 3 It is the compensation factor for the maximum duration of the motion of the set detection object.

[0040] It should be explained that the above-mentioned motor ability evaluation value of the test subject is calculated through the test subject's motor reaction speed, the test subject's joint movement angle of the part that needs rehabilitation, and the test subject's maximum continuous movement duration. It can comprehensively evaluate the test subject's motor ability, including flexibility, reaction ability and endurance. The motor ability evaluation value is calculated through objective test indicators, which avoids the deviation that may be caused by subjective evaluation and can provide more accurate and reliable evaluation results. Understanding the test subject's motor ability level can help rehabilitation experts better design personalized rehabilitation plans, conduct targeted training and adjustments to improve motor ability and promote rehabilitation. Through the quantified results of the motor ability evaluation value, the test subject can clearly understand his or her rehabilitation progress and goals, enhance the confidence and motivation for rehabilitation, and more actively participate in rehabilitation training. For test subjects with different motor ability levels, Rationally configure rehabilitation resources, optimize the utilization efficiency of rehabilitation resources, and improve the overall effect of rehabilitation treatment. The subject's movement reaction speed, joint activity angle of the part requiring rehabilitation, and the maximum duration of exercise are obtained through motion evaluation equipment or sensors. For example, motion capture systems, inertial measurement units, motion capture cameras, etc. can be used to measure the subject's movement reaction speed, joint activity angle, and duration. The subject's movement reaction speed, reference joint activity angle of the part requiring rehabilitation, and reference duration of exercise stored in the database are obtained through previous evaluations or tests. The compensation factors for the subject's movement reaction speed, joint activity angle of the part requiring rehabilitation, and maximum duration of exercise are set by a professional medical team based on the subject's specific circumstances and rehabilitation goals. The subject's motor ability evaluation value is used as the basis for analyzing the subject's motor rehabilitation needs by comparison.

[0041] It should be explained that the posture evaluation value of the above detection object is analyzed in the following specific process:

[0042]

[0043] In the formula, ω 2 is the posture evaluation value of the test subject, TB is the ratio of the distribution ratio of the test subject's weight between the left and right feet, JW is the spinal curvature angle of the test subject, BV is the gait speed of the test subject, and TB 0 is the ratio of the distribution ratio of the reference weight of the test object stored in the database between the left and right feet, JW 0 The defined spinal curvature angle of the test object stored in the database, BV 0 is the defined gait speed of the detected object stored in the database, σ 4 is the compensation factor for the ratio of the weight distribution of the subject between the left and right feet, σ 5 is the compensation factor of the spinal curvature angle of the set test object, σ 6is the compensation factor of the gait speed of the set detection object, and e is a natural constant.

[0044] It should be explained that the posture assessment value of the above-mentioned test subject is calculated through the ratio of the distribution ratio of the test subject's weight in the left and right feet, the test subject's spinal curvature angle, and the test subject's gait speed. The posture assessment can be quantified, so that medical professionals can objectively understand the test subject's posture state. Through digital indicators, doctors can more accurately track the progress of the test subject and formulate personalized rehabilitation plans accordingly. The distribution ratio of weight in the left and right feet can reflect the postural balance of the test subject and help to discover possible asymmetry or deviation of the test subject. The measurement of the spinal curvature angle can reveal the degree of spinal deviation and help to identify spinal-related problems. The gait speed reflects the walking ability and balance of the test subject. By regularly evaluating the posture, the medical team can discover the possible posture problems or bad habits of the test subject early, which helps to take timely intervention measures to prevent the problem from further deteriorating and improve the rehabilitation effect. The posture assessment results can provide an important reference for the formulation of rehabilitation plans. According to the evaluation results, the medical team can design targeted rehabilitation programs, including posture training, balance training, spinal adjustment, etc., to help the subjects improve their posture, alleviate symptoms, and prevent future injuries and pain. The ratio of the weight distribution between the left and right feet can be measured by a ground reaction plate, a sensor array or similar device. The sensor can measure the pressure distribution of each foot and calculate the ratio. The spinal curvature angle uses imaging technology such as X-rays to obtain images of the spine, and uses specific software tools to analyze the images and measure the curvature angle of the spine. Gait speed can be obtained in a variety of ways, including using pedometers, motion sensors or video analysis. These devices can track an individual's walking movements and calculate gait speed accordingly. The reference values ​​and boundary values ​​stored in the database are usually collected and determined in large-scale studies. The set compensation factor is set by a professional medical team based on the specific situation and rehabilitation goals of the subject. The posture assessment value of the subject is used as the basis for comparison to obtain the analysis of the subject's exercise rehabilitation needs.

[0045] Based on the motor rehabilitation needs of the detected object obtained by comparison, a Bayesian network model is constructed to associate the motor rehabilitation needs of the detected object with the current rehabilitation status of the detected object to obtain the rehabilitation status of the object.

[0046] Further, based on the motor rehabilitation needs of the detected object obtained by comparison, a Bayesian network model is constructed, and the motor rehabilitation needs of the detected object and the current rehabilitation status of the detected object are associated to obtain the rehabilitation status of the object. The specific analysis process is: determine the nodes in the Bayesian network, specifically including the age, gender, and motor rehabilitation needs of the detected object, and combine the weight factors of the age, gender, and motor rehabilitation needs of the detected object stored in the database, and associate the motor rehabilitation needs of the detected object with the rehabilitation status of the detected object based on the Bayesian network to construct a Bayesian network model. Based on the Bayesian network association results, the rehabilitation status of the detected object is obtained. First, the nodes of the Bayesian network are determined. The nodes of the Bayesian network represent various variables that need to be analyzed, including: age, gender, motor rehabilitation needs (specific needs obtained from the preliminary assessment, such as motor function to be enhanced, body parts to be improved, etc.) and rehabilitation status (including motor function recovery degree, pain level, quality of life). By defining the dependency relationship (directed edge) between each node, the causal relationship between variables is represented. For example: age and gender may directly affect multiple aspects of the rehabilitation status, and motor rehabilitation needs directly affect the improvement of the rehabilitation status. Combined with the information of the test object stored in the database (age, gender, sports rehabilitation needs), the corresponding weight factors are extracted. These weight factors help quantify the degree of influence of each variable on the rehabilitation status. For example, the rehabilitation progress of certain age groups may be faster or slower than that of other age groups, and gender may affect the choice of rehabilitation methods. Use historical rehabilitation data to learn the conditional probability tables (CPTs) of these relationships to ensure that the model can accurately reflect the actual situation. The conditional probability table lists the probability of each state of the child node given the parent node state. These expressions are based on Bayesian theorem and can reflect the dependency and causal relationship between variables. For example, if we have a node representing pain level and its parent node is rehabilitation training type, CPTs describe the probability of experiencing various pain levels under different training types. In Bayesian networks, CPTs are used for probabilistic inference, that is, predicting the state of other nodes based on the state of known nodes.

[0047] Specifically, the rehabilitation status of the test subject includes motor function, pain level, and quality of life. Motor function can be represented by a continuous value or a discrete value, reflecting the motor ability of the test subject during the rehabilitation process. The continuous value may be a numerical range, such as a percentage between 0 and 100, indicating the degree of motor function. The discrete value may be several levels, such as "excellent", "good", "average", "poor" levels, or numbers from 1 to 5 to represent different degrees. The pain level can be represented in a similar way to the motor function. The continuous value may represent the degree of pain, such as a number between 0 and 10, where 0 represents no pain and 10 represents severe pain. The discrete value may represent the degree of pain, such as "mild", "moderate", and "severe". The quality of life can be measured by various standardized questionnaires or assessment tools, such as the SF-36 health survey questionnaire, and the score can be continuous or discrete.

[0048] In a specific embodiment, based on the Bayesian network model, personalized rehabilitation needs prediction can be made for each test subject based on individual characteristics such as age and gender of the test subject, combined with the data of sports rehabilitation needs. Through model construction, sports rehabilitation needs decisions can be made more scientifically based on data and test subject characteristics, avoiding interference from subjective factors and improving the objectivity and accuracy of decision-making. The Bayesian network model can help analyze and understand the situation of the test subject more quickly, thereby more effectively formulating personalized rehabilitation plans and improving work efficiency. The model can help assess the risks of different test subjects in the rehabilitation process and take corresponding preventive measures in advance to minimize the risk of possible complications and rehabilitation failure.

[0049] It needs to be explained that the results of the above-mentioned Bayesian network model association can be used to tailor personalized rehabilitation solutions according to the motor rehabilitation needs and rehabilitation status of different test subjects, thereby improving the rehabilitation effect. According to the Bayesian network association results, the rehabilitation status of the test subjects can be timely understood, and the rehabilitation plan can be dynamically adjusted to ensure that the test subjects obtain better rehabilitation effects. Through the Bayesian network model association results, the rehabilitation trend of the test subjects can be more accurately predicted, providing a scientific basis for the medical team and guiding the subsequent rehabilitation work. By associating motor rehabilitation needs and rehabilitation status together, the rehabilitation effect of the test subjects can be more comprehensively evaluated, helping the test subjects to achieve comprehensive and effective rehabilitation.

[0050] Obtain rehabilitation status feedback data of the test subject, and obtain a personalized exercise rehabilitation strategy based on the obtained rehabilitation status feedback data of the test subject in combination with the rehabilitation status of the test subject.

[0051] Specifically, the rehabilitation status feedback data of the test subjects include the average pain duration of the test subjects, the number of rehabilitation training times of the test subjects, and the total duration of the rehabilitation training of the test subjects. The average pain duration indicates the average duration of pain felt by the test subjects each time during the rehabilitation process. By monitoring this indicator, medical staff can understand the intensity and duration of pain of the test subjects during rehabilitation training in order to adjust the rehabilitation plan and pain management measures. The number of rehabilitation training times indicates the number of times the test subjects have undergone rehabilitation training. An increase in the number of rehabilitation training times usually means that the test subjects are actively participating in the rehabilitation process, which helps to promote the improvement of rehabilitation effects. The total duration of rehabilitation training indicates the total duration of the test subjects in rehabilitation training. The total duration of rehabilitation training reflects the time and energy invested by the test subjects in the rehabilitation process, which is of great significance for evaluating the rehabilitation effect and the rehabilitation status of the test subjects.

[0052] It needs to be explained that if the average pain duration is longer, it may affect the number of rehabilitation training sessions for the test subjects. The test subjects may feel a certain degree of hindrance due to the pain, resulting in a reduction in the number of rehabilitation training sessions. The number of rehabilitation training sessions and the total duration of rehabilitation training are usually positively correlated. As the number of training sessions increases, the total duration will also increase accordingly. Long-term pain may affect the total duration of the test subjects' rehabilitation training. The test subjects may reduce the training duration due to the distress caused by the pain.

[0053] It needs to be explained that the above-mentioned rehabilitation status feedback data of the test subject can be used to evaluate the test subject's previous rehabilitation progress, which helps to more accurately evaluate the test subject's rehabilitation effect, so as to improve the effectiveness and efficiency of the exercise rehabilitation strategy. The rehabilitation status feedback data can reflect the individual characteristics and needs of the test subject. By analyzing these data, we can better understand the test subject's rehabilitation situation, so as to personalize the rehabilitation plan. According to the specific circumstances of the feedback data, we can formulate a personalized exercise rehabilitation strategy that better meets the rehabilitation needs of the test subject.

[0054] Furthermore, based on the acquired rehabilitation status feedback data of the detected object and in combination with the rehabilitation status of the detected object, a personalized exercise rehabilitation strategy is obtained by comparison. The specific analysis process is: based on the acquired rehabilitation status feedback data of the detected object, a comprehensive analysis is performed to obtain the rehabilitation status feedback characteristic value of the detected object, and the rehabilitation status feedback characteristic value of the detected object is used as the analysis basis for comparing and obtaining the personalized exercise rehabilitation strategy; the rehabilitation status feedback characteristic value of the detected object and the rehabilitation status of the detected object are set as designated labels, and the designated labels are compared with the personalized exercise rehabilitation strategies corresponding to the designated labels stored in the database to obtain the personalized exercise rehabilitation strategy corresponding to the designated labels.

[0055] It needs to be explained that the above-mentioned personalized exercise rehabilitation strategy is obtained by comparing the rehabilitation status feedback data of the test subject in combination with the rehabilitation status of the test subject. By comparing the rehabilitation status feedback data and the rehabilitation status, the rehabilitation needs and progress of the test subject can be understood more accurately, and a personalized exercise rehabilitation strategy can be tailored for each test subject. According to the specific situation and needs of the test subject, the system can generate a personalized rehabilitation plan that meets their individual characteristics and rehabilitation requirements, thereby improving the pertinence and effectiveness of rehabilitation treatment. By comparing the rehabilitation status feedback data and the rehabilitation status, it can also help the test subject to have a clearer understanding of his or her own rehabilitation progress and effects.

[0056] Furthermore, the recovery status feedback characteristic value of the detection object is analyzed in the following specific process:

[0057]

[0058] In the formula, δ is the rehabilitation status feedback characteristic value of the test subject, TT is the average pain duration of the test subject, LN is the number of rehabilitation training times of the test subject, LT is the total rehabilitation training time of the test subject, and TT is the average pain duration of the test subject. 0 The defined pain duration of the test subjects stored in the database, LN 0 is the reference rehabilitation training times of the test object stored in the database, LT 0 is the total duration of the reference rehabilitation training of the test object stored in the database, θ 1 is the compensation factor of the average pain duration of the test subject, θ 2 is the compensation factor for the number of rehabilitation trainings of the set test object, θ 3 is the compensation factor for the total duration of rehabilitation training of the set test object, and e is a natural constant.

[0059] It needs to be explained that the above-mentioned rehabilitation status feedback characteristic values ​​of the test subjects are calculated through the average pain duration of the test subjects, the number of rehabilitation training times of the test subjects, and the total duration of rehabilitation training of the test subjects. The characteristic values ​​provide objective data indicators, which can help medical personnel to evaluate the rehabilitation status of the test subjects more comprehensively and accurately. The average pain duration reflects the degree of pain of the test subjects during the rehabilitation process, and the number and total duration of rehabilitation training reflect the enthusiasm and investment of the test subjects in rehabilitation training. Based on the calculation results of these characteristic values, we can better understand the rehabilitation status of the test subjects, so as to adjust the personalized rehabilitation plan. If the pain duration is long, it may be necessary to adjust the intensity or method of rehabilitation training. If the number and duration of training are short, it may be necessary to increase the frequency or duration of training. By monitoring and analyzing these characteristic values, problems or deficiencies in the rehabilitation process can be discovered in time, and corresponding measures can be taken to make adjustments, thereby improving the rehabilitation effect. Personalized rehabilitation plans can better meet the needs of the test subjects and improve the pertinence and effectiveness of rehabilitation training. Understanding and paying attention to these rehabilitation status feedback characteristic values ​​can help It helps the test subjects to have a deeper understanding of their own rehabilitation status, enhance their participation and enthusiasm in rehabilitation training, and make the test subjects aware of their own efforts and progress, which helps to improve their self-management and self-adjustment capabilities during the rehabilitation process. Through face-to-face or online surveys, the test subjects or their guardians are asked about the duration of pain, the number of rehabilitation training sessions and the total duration. Actual observed data can be provided, and data can also be obtained from the records of hospitals, rehabilitation centers or other medical institutions. These records usually contain the medical records, rehabilitation training plans and rehabilitation progress of the test subjects, and can provide detailed information about pain duration and rehabilitation training. The defined pain duration, reference number of rehabilitation training sessions, and reference total duration of rehabilitation training of the test subjects stored in the database are determined through previous studies, clinical practice or the guidance of professional institutions, and can be used as reference standards to evaluate the rehabilitation status of the test subjects. The compensation factors for the average pain duration of the test subjects, the compensation factors for the number of rehabilitation training sessions, and the compensation factors for the total duration of rehabilitation training are set based on professional knowledge, clinical experience or research data, and are used to adjust the rehabilitation plan and progress of the test subjects.

[0060] like Figure 3As shown, it is an image of the rehabilitation state feedback characteristic value of the detection object as the average pain duration of the detection object changes, wherein the x-axis represents the average pain duration of the detection object, and the y-axis represents the rehabilitation state feedback characteristic value of the detection object, which can help intuitively understand how the average pain duration of the detection object affects the rehabilitation state feedback characteristic value of the detection object. The larger the average pain duration of the detection object, the smaller the rehabilitation state feedback characteristic value of the detection object, indicating that the feedback of the historical rehabilitation training effect is worse. As the average pain duration of the detection object increases, its influence on the rehabilitation state feedback characteristic value of the detection object gradually weakens and finally stabilizes at a lower value. The number of rehabilitation trainings of the detection object is set to 10, the total rehabilitation training time of the detection object is 100, the defined pain duration of the detection object stored in the database is 5, the number of reference rehabilitation trainings of the detection object stored in the database is 5, and the total reference rehabilitation training time of the detection object stored in the database is 50. The compensation factor of the average pain duration of the detection object is set to 0.3, the compensation factor of the number of rehabilitation trainings of the detection object is set to 0.4, and the compensation factor of the total rehabilitation training time of the detection object is set to 0.4. Only the size of the average pain duration of the detection object is changed. Example values ​​of the average pain duration of the detection object are as follows:

[0061] Table 1: Example values ​​of the average pain duration of the test subject in the test subject's recovery status feedback feature value

[0062]

[0063]

[0064] It should be noted that the greater the number of rehabilitation training times of the detected object, the greater the characteristic value of the rehabilitation status feedback of the detected object, indicating that the feedback of the historical rehabilitation training effect is better. The average pain duration of the detected object is set to 9, the total rehabilitation training time of the detected object is 100, the defined pain duration of the detected object stored in the database is 5, the reference rehabilitation training times of the detected object stored in the database is 5, the total reference rehabilitation training time of the detected object stored in the database is 50, the compensation factor of the average pain duration of the detected object is set to 0.3, the compensation factor of the number of rehabilitation training times of the detected object is set to 0.4, and the compensation factor of the total rehabilitation training time of the detected object is set to 0.4. Only the number of rehabilitation training times of the detected object is changed. Example values ​​of the number of rehabilitation training times of the detected object are as follows:

[0065] Table 2: Example values ​​of the number of rehabilitation training sessions of the test subject in the rehabilitation status feedback feature value of the test subject

[0066]

[0067] Reference Figure 2 As shown, the second aspect of the present invention provides a motion rehabilitation strategy generation system based on a Bayesian network, including a motion rehabilitation demand comparison module for a detection object, a rehabilitation status analysis module for a detection object, and a personalized motion rehabilitation strategy acquisition module, wherein: the motion rehabilitation demand comparison module for a detection object is used to obtain the motion state data of the detection object, and obtain the motion rehabilitation demand of the detection object by comparison based on the acquired motion state data of the detection object; the rehabilitation status analysis module for the detection object is used to construct a Bayesian network model based on the motion rehabilitation demand of the detection object obtained by comparison, associate the motion rehabilitation demand of the detection object with the current rehabilitation status of the detection object, and obtain the rehabilitation status of the object; the personalized motion rehabilitation strategy acquisition module is used to obtain the rehabilitation status feedback data of the detection object, and obtain a personalized motion rehabilitation strategy by comparison based on the acquired rehabilitation status feedback data of the detection object and the rehabilitation status of the detection object.

[0068] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. A method for generating a motor rehabilitation strategy based on a Bayesian network, characterized in that: The following steps are involved: Acquire the motion state data of the detection object, and obtain the motion rehabilitation needs of the detection object based on the acquired motion state data of the detection object; Based on the motor rehabilitation needs of the detected object obtained by comparison, a Bayesian network model is constructed to associate the motor rehabilitation needs of the detected object with the current rehabilitation status of the detected object to obtain the rehabilitation status of the object; Acquire rehabilitation status feedback data of the test subject, and obtain a personalized exercise rehabilitation strategy based on the acquired rehabilitation status feedback data of the test subject and the rehabilitation status of the test subject; The motion state data of the detected object is obtained based on the comparison to obtain the motion rehabilitation needs of the detected object. The specific analysis process is as follows: Based on the acquired motion state data of the detected object, a motion state evaluation value of the detected object is obtained through comprehensive analysis, and the motion state evaluation value of the detected object is used as an analysis basis for comparing and obtaining the motion rehabilitation needs of the detected object; Comparing the motion state evaluation value of the detection object with the motion rehabilitation needs of the detection object corresponding to the motion state evaluation values ​​of each detection object stored in the database, to obtain the motion rehabilitation needs of the detection object corresponding to the motion state evaluation value of the detection object; The motion state evaluation value of the detection object, the specific analysis process is: Wherein, ω is the motion state evaluation value of the detection object, ω1 is the motion ability evaluation value of the detection object, ω2 is the posture evaluation value of the detection object, μ1 is the weight factor of the motion ability evaluation value of the detection object, μ2 is the weight factor of the posture evaluation value of the detection object, and e is a natural constant; The athletic ability evaluation value of the test object is analyzed as follows: Wherein, ω1 is the motor ability evaluation value of the detected object, FV is the motor reaction speed of the detected object, GJ is the joint activity angle of the part requiring rehabilitation of the detected object, YC is the maximum duration of the movement of the detected object, FV0 is the motor reaction speed of the detected object stored in the database, GJ0 is the reference joint activity angle of the part requiring rehabilitation of the detected object stored in the database, YC0 is the reference duration of the movement of the detected object stored in the database, σ1 is the compensation factor of the motor reaction speed of the detected object, σ2 is the compensation factor of the joint activity angle of the part requiring rehabilitation of the detected object, and σ3 is the compensation factor of the maximum duration of the movement of the detected object set; The posture evaluation value of the detection object, the specific analysis process is: Wherein, ω2 is the posture evaluation value of the detection object, TB is the ratio of the distribution ratio of the detection object's weight in the left and right feet, JW is the detection object's spinal curvature angle, BV is the detection object's gait speed, TB0 is the ratio of the distribution ratio of the reference weight of the detection object in the left and right feet stored in the database, JW0 is the defined spinal curvature angle of the detection object stored in the database, BV0 is the defined gait speed of the detection object stored in the database, σ4 is the compensation factor of the set detection object's weight distribution ratio in the left and right feet, σ5 is the compensation factor of the set detection object's spinal curvature angle, σ6 is the compensation factor of the set detection object's gait speed, and e is a natural constant.

2. The method for generating a sports rehabilitation strategy based on a Bayesian network according to claim 1, characterized in that: The motion state data of the detection object specifically includes the motion ability data of the detection object and the posture data of the detection object, wherein the motion ability data of the detection object includes the motion reaction speed of the detection object, the joint movement angle of the detection object's part requiring rehabilitation, and the maximum duration of the detection object's motion; the posture data of the detection object includes the ratio of the distribution proportion of the detection object's weight between the left and right feet, the spinal curvature angle of the detection object, and the gait speed of the detection object.

3. The method for generating a motor rehabilitation strategy based on a Bayesian network according to claim 1, characterized in that: The motor rehabilitation needs of the detected object obtained based on the comparison are used to construct a Bayesian network model, and the motor rehabilitation needs of the detected object are associated with the current rehabilitation status of the detected object to obtain the rehabilitation status of the object. The specific analysis process is as follows: Determine the nodes in the Bayesian network, specifically including the age, gender, and motor rehabilitation needs of the test subject, combine the weight factors of the age, gender, and motor rehabilitation needs of the test subject stored in the database, associate the motor rehabilitation needs of the test subject with the rehabilitation status of the test subject based on the Bayesian network, build a Bayesian network model, and obtain the rehabilitation status of the test subject based on the Bayesian network association results.

4. The method for generating a sports rehabilitation strategy based on a Bayesian network according to claim 1, characterized in that: The rehabilitation status feedback data of the detected subject specifically includes the average pain duration of the detected subject, the number of rehabilitation training times of the detected subject, and the total duration of rehabilitation training of the detected subject.

5. The method for generating a motor rehabilitation strategy based on a Bayesian network according to claim 4, characterized in that: The rehabilitation status feedback data of the detected object obtained is combined with the rehabilitation status of the detected object to obtain a personalized exercise rehabilitation strategy. The specific analysis process is as follows: Based on the acquired rehabilitation status feedback data of the test subject, a comprehensive analysis is performed to obtain a rehabilitation status feedback characteristic value of the test subject, and the rehabilitation status feedback characteristic value of the test subject is used as an analysis basis for obtaining a personalized exercise rehabilitation strategy by comparison; The rehabilitation status feedback characteristic value of the detection object and the rehabilitation status of the detection object are set as a designated label, and the designated label is compared with the personalized exercise rehabilitation strategy corresponding to each designated label stored in the database to obtain the personalized exercise rehabilitation strategy corresponding to the designated label; The recovery state feedback characteristic value of the detection object is analyzed in detail as follows: Wherein, δ is the rehabilitation status feedback characteristic value of the detected object, TT is the average pain duration of the detected object, LN is the number of rehabilitation trainings of the detected object, LT is the total rehabilitation training duration of the detected object, TT0 is the defined pain duration of the detected object stored in the database, LN0 is the reference number of rehabilitation trainings of the detected object stored in the database, LT0 is the reference total rehabilitation training duration of the detected object stored in the database, θ1 is the compensation factor of the set average pain duration of the detected object, θ2 is the compensation factor of the set number of rehabilitation trainings of the detected object, θ3 is the compensation factor of the set total rehabilitation training duration of the detected object, and e is a natural constant.

6. A motor rehabilitation strategy generation system based on Bayesian network, characterized in that: A method for generating a motor rehabilitation strategy based on a Bayesian network as described in any one of claims 1 to 5, comprising a motor rehabilitation demand comparison module for a detection object, a rehabilitation status analysis module for a detection object, and a personalized motor rehabilitation strategy acquisition module, wherein: The movement rehabilitation demand comparison module of the detection object is used to obtain the movement state data of the detection object, and compare the movement rehabilitation demand of the detection object based on the obtained movement state data of the detection object; The rehabilitation status analysis module of the detected object is used to construct a Bayesian network model based on the motor rehabilitation needs of the detected object obtained by comparison, associate the motor rehabilitation needs of the detected object with the current rehabilitation status of the detected object, and obtain the rehabilitation status of the object; The personalized exercise rehabilitation strategy acquisition module is used to obtain the rehabilitation status feedback data of the detection object, and obtain a personalized exercise rehabilitation strategy based on the obtained rehabilitation status feedback data of the detection object and the rehabilitation status of the detection object.

Images