CN115631532A - A training action classification and completion scoring method in a rehabilitation training system - Google Patents

Abstract

The invention discloses a training action classification and completion scoring method in a rehabilitation training system, which is applied to a doctor-patient interaction system for rehabilitation training. The method includes: step 1, patients follow the guidance instructions and standard training actions preset in the system Guide the video to complete rehabilitation training actions, and collect action data information; Step 2, classify the training actions collected in Step 1 according to the preset training action classification criteria, and use different completion scoring methods for different categories; Step 3 , according to the training action collected in step 1, extract appropriate action features, and combine the completion scoring method corresponding to step 2 to calculate the completion score of the current training action. The present invention can obtain a training score that is more in line with the rehabilitation training situation, and the score can help the rehabilitation doctor understand the patient's rehabilitation training situation more intuitively; at the same time, it is also convenient for the patient to know whether his own training action is standard, so as to correct it in time.

Description

A training action classification and completion scoring method in a rehabilitation training system

technical field

The invention relates to the field of smart medical care, in particular to a method for classifying training actions and scoring completion in a rehabilitation training system.

Background technique

As the aging population continues to intensify, the prevalence of cardiovascular and cerebrovascular diseases in social groups is also gradually increasing. At the same time, the number of accidental disabilities caused by frequent traffic accidents is also increasing. In the process of treating paralyzed and disabled patients, the rehabilitation training of patients is particularly important, and a large amount of medical resources need to be invested. Faced with the above situation, there is an urgent need for a system that can automatically guide patients to complete rehabilitation training.

The traditional rehabilitation training program includes a set of systematic movements such as arm joint extension, elbow joint rotation, limb stretching, horse stance squat, lunge leg press, etc. The above-mentioned set of movements is formulated by the rehabilitation physician and guides the patient to follow the plan every day Quantitative completion; the standardization of rehabilitation training movements needs to be judged by professional rehabilitation physicians. Due to the variety of training movements and the extremely professional and subjective evaluation of the completion of training movements, a large number of professional medical staff are required for rehabilitation training. Patients provide training guidance and training effect evaluation; in the face of the current shortage of medical staff, there is an urgent need for a method that can automatically guide patients' rehabilitation training and score their rehabilitation training actions.

Due to the wide variety of rehabilitation exercises and the different evaluation standards for different rehabilitation exercises, and the scoring of rehabilitation training is highly subjective and professional, it requires a lot of experienced rehabilitation experts to give accurate training scores. The existing calculation methods for the completion score of rehabilitation training actions do not consider the complex situations in practical applications to classify rehabilitation training actions, and use the same or similar methods to calculate training scores for different types of actions, so it is impossible to obtain reasonable and Accurate scoring of exercise completion. The Chinese invention patent with the publication number CN108346457A discloses a rehabilitation training evaluation method, device and system. The program obtains the comparison between the patient's body movements and the preset standard body movements, and obtains the conformity between the current body movements and the standard body movements. degree, and the rehabilitation training score of the current limb is generated from the degree of conformity. Although this solution can handle the rehabilitation training score of the patient, it is only suitable for the calculation of the score calculation of a single joint or limb rehabilitation exercise, and is not suitable for the calculation of the score calculation of the rehabilitation training movement of multiple joints or limbs. It is suitable for the calculation of rehabilitation training scores for continuous state training actions. The Chinese invention patent with the publication number CN113707268A discloses a rehabilitation training evaluation method and system. The program analyzes the movement of the video and compares the patient's training action with the preset movement trajectory to obtain the standard degree of the patient's training action. , so as to obtain the patient's rehabilitation training score. Although this solution can solve the calculation problem of rehabilitation training scores for single-joint training actions and multi-joint combined training actions, it still cannot solve the problem of calculating rehabilitation training scores for continuous state training actions. It can be seen that there is still a lack of methods for systematic analysis of rehabilitation training movements and reasonable and accurate evaluation of the completion of rehabilitation training movements.

To this end, we propose a computer technology to classify existing rehabilitation training actions, and use a more targeted method to score the completion of rehabilitation actions for different action categories.

Contents of the invention

The purpose of the present invention is to solve the defects of the prior art, and propose a training action classification and completion scoring method in a rehabilitation training system. The training score is combined with the rehabilitation training plan to judge the degree of rehabilitation.

In order to achieve the above-mentioned purpose, the realization technical scheme of the present invention is:

A training action classification and completion scoring method in a rehabilitation training system, comprising the following steps:

Step 1: Collect the action data information of the patient when performing rehabilitation training actions according to the guiding instructions preset in the system and the standard training action guiding video;

Step 2: According to the preset training action classification criteria, classify the action data information collected in step 1, and select different completion scoring methods for different classifications;

Step 3: According to the action data information collected in step 1, extract the action features, and combine the completion scoring method corresponding to step 2 to calculate the completion score of the current training action.

Preferably, the specific method of collecting motion data in step 1 is: collecting the patient's rehabilitation training motion video, and using the bone point recognition algorithm to obtain the patient's bone point coordinate information during the training process as motion data.

Preferably, the preset training action classification criteria specifically include: single-joint training action, multi-joint combination training action, continuous state training action, wherein, single-joint training action refers to the movement of a single joint or limb from the starting position to the ending position multi-joint combination training action refers to the process action that is composed of multiple joints or limbs, moving from the start position to the end position; continuous state training action refers to the movement of multiple joints or limbs of the body to a certain After the state, and maintain a certain period of state action.

Preferably, the extracted action features include: angles between joints, normalized distances between bone points, completion time of training actions, and duration of action states.

Preferably, the single-joint training action is used to calculate and score the range of motion and time of a single joint. The specific calculation method for the score is: detect the minimum and maximum values of the range of motion of a single joint during the entire exercise process, and based on the difference between the two It is worth scoring the patient's training range of motion combined with the time to complete the motion. The specific scoring formula is as follows:

Among them, F _single is the score of a single joint training action; Φ is the amplitude value of a single joint or limb movement during rehabilitation training, Φ _s is the standard value of a single joint or limb movement amplitude during rehabilitation training; T is the patient's completed The action time of a single joint or limb training process, T _s is the standard time to complete the action of a single joint or limb training process; A ₁ is the weight value of the impact of the range of motion on the rehabilitation training score; A ₂ is the action completion time on the rehabilitation training score Affects the weight value.

Preferably, the multi-joint combined training action adopts the coordination of multiple joints, and the range of motion of each joint should be considered comprehensively when calculating the comprehensive score according to the range of motion of each joint. The specific calculation method is: detecting the range of motion of each joint during the entire exercise process According to the difference between the minimum and maximum values, the range of motion of each joint in the patient's training action is obtained according to the difference between the two, and different weights are assigned to the range of motion of each joint, combined with the time to complete the action for scoring. The specific scoring formula is: :

为康复训练动作过程中第i个关节或肢体动作的单独评分，并且1<i<＝n，n是当前多关节组合训练动作涉及到关节或者肢体动作的数目；Φ_i为康复训练动作过程中第i个关节或肢体动作的幅度值；Φ_si为康复训练动作过程中第i个关节或肢体动作的标准幅度值；T_i为病患完成第i个关节或肢体动作的时间；T_si为完成第i个关节或肢体动作的标准时间；

为第i个关节或肢体动作幅度对康复训练评分的影响权重值；

为第i个关节或肢体动作完成时间对康复训练评分的影响权重值；B_i为第i个关节或肢体动作评分对整体评分的影响权重值。Among them, F _multi is the score of multi-joint combination training action;

It is the individual score of the i-th joint or body movement during the rehabilitation training action, and 1<i<=n, n is the number of joints or body movements involved in the current multi-joint combined training action; Φ _i is the rehabilitation training action process The amplitude value of the i-th joint or limb movement; Φ _si is the standard amplitude value of the i-th joint or limb movement during rehabilitation training; T _i is the time for the patient to complete the i-th joint or limb movement; T _si is The standard time to complete the i-th joint or body movement;

It is the weight value of the impact of the i-th joint or limb movement range on the rehabilitation training score;

B i is the influence weight value of the ith joint or body movement completion time on the rehabilitation training score; B _i is the influence weight value of the i joint or body movement score on the overall score.

Preferably, the continuous state training action should be qualified when it reaches the standard action state and maintains it for a certain period of time. The specific calculation method is: detect the range of motion of a single or multiple joints during the entire exercise process, and select a stable peak value As the range of action state of one or more joints in the patient's training action; and this value is combined with the duration of the action state for scoring. The specific scoring formula is:

Among them, F _state is the score of the continuous state training action; k is the action state matching coefficient of the continuous state training action; T is the action state duration of the continuous state training action; T _s is the standard action state of the continuous state training action Duration; sim(Q, Q _s ) is the similarity between the action state of the continuous state training action and the standard action state. When the similarity of the action state is greater than the similarity threshold, the cumulative action state duration is started; TH is the persistence The action state similarity threshold of the state training action; m is the number of action state features involved in the current continuous state training action judged by expert experience; Aj is the _jth action state feature in the continuous state training action. The influence weight value of the state similarity, and 1<=j<=m; Q is the state feature vector composed of all j action state features involved in the continuous state training action, where δ _j is the jth in the training action action state features; Q _s is the standard state feature vector composed of all j standard action state features involved in the continuous state training action, where δs _j is the jth standard action state feature in the training action.

Beneficial effects of the present invention:

The present invention classifies different rehabilitation training actions in advance, and uses a more targeted scoring method to score different types of actions. The scoring is more accurate and more consistent with the subjective judgment of medical staff.

The present invention can obtain a more accurate and more intuitive and reasonable scoring result when it is applied to the calculation of rehabilitation training action scores; medical staff can analyze the rehabilitation training situation of patients more accurately through this score; at the same time, patients can also be more In order to intuitively understand the completion of each rehabilitation exercise, rather than a general score, it is convenient for patients to improve their rehabilitation exercises.

Description of drawings

The features and advantages of the present invention will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the invention in any way. In the accompanying drawings:

Fig. 1 is the overall system block diagram of the specific embodiment of the present invention;

Fig. 2 is the flow chart of using the patient personnel terminal in the specific embodiment of the present invention;

Fig. 3 is the information schematic diagram of the virtual human skeleton point of the specific embodiment of the present invention;

Fig. 4 is a process diagram of the lifting action of the right arm in a specific embodiment of the present invention;

Fig. 5 is a diagram of the action process of the right leg lunge and leg press in a specific embodiment of the present invention;

Fig. 6 is an action process diagram of squatting in a horse stance in a specific embodiment of the present invention;

The numbers in the figure are: 1 represents the nose bone point, 2 represents the right shoulder bone point, 3 represents the left shoulder bone point, 4 represents the right elbow bone point, 5 represents the left elbow bone point, 6 represents the right wrist bone point, 7 represents the left wrist bone point, 8 represents the right hip bone point, 9 represents the left hip bone point, 10 represents the right knee bone point, 11 represents the left knee bone point, 12 represents the right ankle bone point, and 13 represents the left ankle bone point.

Detailed ways

The following description and drawings illustrate specific embodiments of the invention sufficiently to enable those skilled in the art to practice them.

It should be clear that the described embodiments are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

When the following description refers to the accompanying drawings, the same numerals in different drawings indicate the same or similar elements unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the present invention. Rather, they are merely examples of systems and methods consistent with aspects of the invention as recited in the appended claims.

In the description of the present invention, it should be understood that the terms "first", "second" and so on are only used for descriptive purposes, and should not be understood to indicate or imply relative importance. Those of ordinary skill in the art can understand the specific meanings of the above terms in the present invention in specific situations. In addition, in the description of the present invention, unless otherwise specified, "plurality" means two or more. "And/or" describes the association relationship of associated objects, indicating that there may be three types of relationships, for example, A and/or B may indicate: A exists alone, A and B exist simultaneously, and B exists independently. The character "/" generally indicates that the contextual objects are an "or" relationship.

Referring to Fig. 1, the specific embodiment of the present invention provides an interactive system for rehabilitation action training, including a training plan setting module, a training result query module, a training plan acquisition module, a training action data acquisition module, and a training action classification scoring module , training result query upload module, medical personnel terminal, patient personnel terminal, system administrator terminal and server system,

The medical personnel terminal is used by rehabilitation physicians and other personnel. The rehabilitation physician can select rehabilitation training actions for patients and formulate rehabilitation training plans on this terminal. At the same time, rehabilitation physicians can view the patient's rehabilitation training result data on this terminal. The terminal includes training plan settings module and training result query module,

The training plan setting module is used to formulate a suitable rehabilitation training plan for patients. The specific implementation method is: the rehabilitation physician first selects the appropriate rehabilitation training action according to the patient's injury, illness, disability, etc., and then determines the daily completion times and training time of the training action according to the patient's physical condition, and finally the entire training plan sent to patients through the server system,

The training result query module is used to facilitate the rehabilitation physician to view the results of the patient's rehabilitation action training. This module allows rehabilitation physicians to view the scores of patients completing different rehabilitation actions, and provide training suggestions or modify training plans for patients according to the training results of patients.

The patient personnel terminal is used by rehabilitation training patients or their family members. The specific use process is shown in Figure 2. Patients and their family members can view the patient's rehabilitation training plan on the terminal, and complete the planned rehabilitation training actions according to the guidance. The built-in scoring system of the terminal will score according to the completion of the rehabilitation training action, and display the rehabilitation training action score on the patient terminal for viewing. At the same time, the rehabilitation training action score will be saved and uploaded to the medical staff terminal for viewing by the rehabilitation doctor patient training. The terminal includes a training plan acquisition module, a training action classification and scoring module, and a training result query module reporting module.

The training plan obtaining module is used to obtain and view the rehabilitation training plan formulated by the rehabilitation physician. The rehabilitation training plan contains several rehabilitation training actions. Patients can view the training actions in the rehabilitation training plan through the training plan acquisition module, as well as the number of times the actions are completed, the completion standards of the actions, and the training guidance of the rehabilitation physician.

The training action data acquisition module is used to collect the patient's rehabilitation training action data information. In this embodiment, the camera equipment is used to collect the action video of the patient, and then the coordinate information of each bone point of the patient during the rehabilitation action process is obtained through a human bone point recognition algorithm such as BlazePose. The patient's rehabilitation training action data information is obtained by processing the coordinate information of the bone points.

It should be noted that what is used in this embodiment is the BlazePose human bone point recognition algorithm, which can identify 33 bone points in total, but in practical applications, some useless bone point coordinate information can be discarded according to the situation. The present invention The bone point information used is shown in Figure 3. However, in practical applications, the above human skeleton point recognition algorithm may also be OpenPose, etc., and the human skeleton point recognition algorithm is not limited in this embodiment.

The training action classification and scoring module is used to classify and score the rehabilitation training actions of patients. Due to the variety and complexity of rehabilitation training movements, a single scoring method cannot accurately and effectively evaluate the completion and accuracy of the patient's training movements. Therefore, it is necessary to pre-classify different rehabilitation training movements, and then use different to obtain a more accurate training completion score.

Rehabilitation training movements can be roughly divided into three categories: single-joint training movements, multi-joint combination training movements, and continuous state training movements. Targeted calculation methods for different types of training movements can effectively evaluate the accuracy of training movements and degree of completion.

Single-joint training movements, including trunk joint flexion, extension and rotation, have no fixed initial state or fixed end state, and only need to combine the range of motion of a single joint during rehabilitation training and the time to complete the movement to calculate and score.

The scoring formula is as follows:

Among them, F _single is the score of a single joint training action; Φ is the amplitude value of a single joint or limb movement during rehabilitation training, Φ _s is the standard value of a single joint or limb movement amplitude during rehabilitation training; T is the patient's completed The action time of a single joint or limb training process, T _s is the standard time to complete the action of a single joint or limb training process; A ₁ is the weight value of the impact of the range of motion on the rehabilitation training score; A ₂ is the action completion time on the rehabilitation training score Affects the weight value; the judgment standard of joint or limb movement range should adopt different judgment features according to different joints or limb movements.

The following takes the lifting action of the right arm joint as an example to explain the scoring calculation process in detail. The lifting action process of the right arm joint is shown in Figure 4. From (a) initial state through (b) intermediate state movement to (c) end state, the whole During the process, the right arm is gradually raised, and the right arm joint is trained.

According to the guidance of the patient terminal, the patient completes the lifting action of the right arm joint with the right side of the body facing the camera equipment, and obtains the bone point coordinate data of the patient during the action process through the training action data acquisition module.

For the lifting action of the right arm joint, the coordinate information of three bone points needs to be collected, which are as follows:

The abscissa of the right hip bone point and the ordinate of the right hip bone point, hereinafter referred to as RHx and RHy;

The abscissa of the right shoulder bone point and the ordinate of the right shoulder bone point are referred to as RSx and RSy below;

The abscissa of the right elbow bone point and the ordinate of the right elbow bone point are referred to as REx and REy below;

The formula for calculating the action angle W of the right arm joint is:

Calculate the angle of the right arm joint during the lifting action of the entire right arm joint according to the above formula, and then process to obtain the maximum value W _max and minimum value W _min of the right arm joint angle during the action process and the time T to complete the action, which is the same as the preset During the standard action process, the amplitude value W _s of the right arm joint angle change is compared with the standard action completion time T _s to calculate the score. The impact of the action range on the rehabilitation training score can be changed by the preset weight value A ₁ ; the action completion The influence of time _on the rehabilitation training score can be changed through the preset weight value A2.

The specific scoring formula is:

The above-mentioned score calculation process of the right elbow joint rotation action is only an embodiment of the action score calculation process of the single joint training process. The movements of the single-joint training process not only include the rotation of the right elbow joint, and the factors affecting the scoring include not only the angle of the joints, but also the normalized distance of the bone joints can be used for scoring some movements.

Multi-joint combination training actions, including lunge presses, double-arm stretching and other actions, such actions have neither a fixed initial state nor a fixed end state, and the coordination of multiple joints is required in the rehabilitation training process, so for this The range of motion of each joint should be considered comprehensively when calculating the score of such actions.

The scoring formula is as follows:

为康复训练动作过程中第i个关节或肢体动作的单独评分，并且1<i<＝n，n是当前多关节组合训练动作涉及到关节或者肢体动作的数目；Φ_i为康复训练动作过程中第i个关节或肢体动作的幅度值；Φ_si为康复训练动作过程中第i个关节或肢体动作的标准幅度值；T_i为病患完成第i个关节或肢体动作的时间；T_si为完成第i个关节或肢体动作的标准时间；

为第i个关节或肢体动作幅度对康复训练评分的影响权重值；

为第i个关节或肢体动作完成时间对康复训练评分的影响权重值；B_i为第i个关节或肢体动作评分对整体评分的影响权重值；关节或肢体动作幅度的判断标准应根据不同的关节或肢体动作采用不同的判断特征。Among them, F _multi is the score of multi-joint combination training action;

It is the individual score of the i-th joint or body movement during the rehabilitation training action, and 1<i<=n, n is the number of joints or body movements involved in the current multi-joint combined training action; Φ _i is the rehabilitation training action process The amplitude value of the i-th joint or limb movement; Φ _si is the standard amplitude value of the i-th joint or limb movement during rehabilitation training; T _i is the time for the patient to complete the i-th joint or limb movement; T _si is The standard time to complete the i-th joint or body movement;

It is the weight value of the impact of the i-th joint or limb movement range on the rehabilitation training score;

B i is the impact weight value of the i-th joint or body movement completion time on the rehabilitation training score; B _i is the influence weight value of the i-th joint or body movement score on the overall score; the judgment standard of the joint or body movement range should be based on different Different judgment features are used for joint or body movements.

The following takes the right leg lunge and leg press as an example to explain the scoring calculation process in detail. The score of right leg lunge and leg press is based on expert experience and needs to be judged using two joints or body movement characteristics, which are the bending angle of the right knee joint and the normalized distance between the left and right ankles, so the value of n in the above formula is 2. The right leg lunge leg press action process is shown in Figure 5, from (a) initial state through (b) intermediate state movement to (c) end state, during the whole process, press the right knee joint with both hands and squat continuously, and at the same time, the left foot and With the right feet separated by a certain distance, the right knee joint and hip joint are trained.

According to the guidance of the patient's personnel terminal, the patient completes the right leg lunge and leg press action with the right side of the body facing the camera equipment, and obtains the bone point coordinate data of the patient during the action process through the training action data acquisition module.

For the right leg lunge and leg press action, it is necessary to collect the coordinate information of four bone points, which are as follows:

The abscissa of the right hip bone point and the ordinate of the right hip bone point, hereinafter referred to as RHx and RHy;

The abscissa of the right knee bone point and the ordinate of the right knee bone point, hereinafter referred to as RKx and RKy;

The abscissa of the right ankle bone point and the ordinate of the right ankle bone point, hereinafter referred to as RAx and RAy;

The abscissa of the left ankle bone point and the ordinate of the left ankle bone point, hereinafter referred to as LAx and LAy;

The formula for calculating the action angle W of the knee joint of the right leg is:

The formula for calculating the normalized distance D of the left and right ankle joints is:

来改变，右膝关节动作完成时间对康复训练评分的影响可以通过预设的权重值

来改变。同时对于弓步压腿动作来说，完成训练动作时左右脚踝关节的归一化距离D和标准动作中左右脚踝关节的归一化距离D_s对康复训练效果也有影响；左右脚踝关节的归一化距离对康复训练评分的影响可以通过预设的权重值

来改变，左右脚的动作完成时间对康复训练评分的影响可以通过预设的权重值

来改变，右膝关节动作完成度评分对整体评分的影响可以通过预设的权重值B₁来改变，左右脚的距离评分对整体评分的影响可以通过预设的权重值B₂来改变,具体评分计算公式为：According to the above formula, the angle of the knee joint of the right leg during the entire lunge leg press action is calculated, and then processed to obtain the maximum value W _max and the minimum value W _min of the knee joint angle of the right leg during the action process and the time T for completing the action of the right knee joint _1. Comparing the amplitude value W _s of the knee joint angle change of the right leg during the preset standard action process and the standard right knee joint action completion time T _s1 to calculate the score; the influence of the right knee joint angle on the rehabilitation training score can be obtained by predicting set the weight value

To change, the influence of the completion time of the right knee joint action on the rehabilitation score can be changed by the preset weight value

to change. At the same time, for the lunge press action, the normalized distance D of the left and right ankle joints when completing the training action and the normalized distance D _s of the left and right ankle joints in the standard action also have an impact on the rehabilitation training effect; the normalized distance of the left and right ankle joints The impact of distance on rehabilitation training score can be adjusted by the preset weight value

To change, the impact of the completion time of the left and right feet on the score of rehabilitation training can be adjusted by the preset weight value

To change, the impact of the right knee joint movement completion score on the overall score can be changed through the preset weight value B ₁ , and the impact of the left and right foot distance scores on the overall score can be changed through the preset weight value B ₂ . The scoring formula is:

The score calculation process of the above-mentioned lunge leg press action is only an embodiment of the action score calculation process of the multi-joint combination training process. The multi-joint combination training process does not only include lunge and leg press movements, and the factors affecting the combination score do not only include joint angles and joint normalized distances.

Sustained state training movements include horse stance squats, raising hands and standing. Patients with such movements only need to reach a certain movement state and maintain it for a certain period of time to be considered qualified.

The scoring formula is as follows:

Among them, F _state is the score of the continuous state training action; k is the action state matching coefficient of the continuous state training action; T is the action state duration of the continuous state training action; T _s is the standard action state of the continuous state training action Duration; sim(Q, Q _s ) is the similarity between the action state of the continuous state training action and the standard action state. When the similarity of the action state is greater than the similarity threshold, the cumulative action state duration is started; TH is the persistence The action state similarity threshold of the state training action; m is the number of action state features involved in the current continuous state training action judged by expert experience; Aj is the _jth action state feature in the continuous state training action. The influence weight value of the state similarity, and 1<=j<=m; Q is the state feature vector composed of all j action state features involved in the continuous state training action, where δ _j is the jth in the training action action state features; Q _s is the standard state feature vector composed of all j standard action state features involved in the continuous state training action, where δs _j is the jth standard action state feature in the training action; different continuous Sexual state training actions need to choose different action state characteristics according to expert experience.

The following takes the horse stance squat as an example to explain the scoring calculation process in detail. According to expert experience, the completion score of the horse stance squat needs to be judged using three joints or body movement characteristics, which are the bending angle of the right knee joint and the left knee joint. The bending angle of the upper limb and the vertical angle of the upper limb, so the value of m in the above formula is 3. The process of squatting in a horse stance is shown in Figure 6. From (a) the initial state through (b) the intermediate state to (c) the end state, the knees are constantly bent during the whole process, while the upper body is kept relatively vertical to the ground, and the knees Joints are trained.

According to the guidance of the patient personnel terminal, the patient faces the camera equipment with the right side of the body and completes the squat movement of the horse stance, and obtains the bone point coordinate data of the patient during the movement process through the training movement data acquisition module. For horse stance squat action, it is necessary to collect the coordinate information of three bone points, which are as follows:

The abscissa of the right hip bone point and the ordinate of the right hip bone point, hereinafter referred to as RHx and RHy;

The abscissa of the right knee bone point and the ordinate of the right knee bone point, hereinafter referred to as RKx and RKy;

The abscissa of the right ankle bone point and the ordinate of the right ankle bone point, hereinafter referred to as RAx and RAy;

The abscissa of the left hip bone point and the ordinate of the left hip bone point, hereinafter referred to as LHx and LHy;

The abscissa of the left knee bone point and the ordinate of the left knee bone point, hereinafter referred to as LKx and LKy;

The abscissa of the left ankle bone point and the ordinate of the left ankle bone point, hereinafter referred to as LAx and LAy;

The abscissa of the right shoulder bone point and the ordinate of the right shoulder bone point are referred to as RSx and RSy below;

The horse stance squat movement needs to bend the knee joint to a certain angle and keep the upper limbs upright. If this state is satisfied and lasts for a period of time, it will be judged as a full score, and there is no need to judge the process in the middle of the movement.

The formula for calculating the action angle W ₁ of the knee joint of the right leg is:

The formula for calculating the action angle W ₂ of the knee joint of the left leg is:

The formula for calculating the vertical angle W ₃ of the upper limbs is:

The specific scoring formula is:

Among them, F ₃ is the score of the horse stance squat action; k is the action state matching coefficient of the horse stance squat action; T is the action state duration of the horse stance squat action; T _s is the standard action state of the horse stance squat action Duration; sim(Q, Q _s ) is the similarity between the action state of the horse stance squat action and the standard action state. When the similarity of the action state is greater than the similarity threshold, the duration of the action state will be accumulated; TH is the horse stance The action state similarity threshold of the squat action; Q is the state feature vector composed of the action state features of the right knee joint, left knee joint, and upper limbs in the horse stance squat action, where W ₁ , W ₂ , and W ₃ are horse The action state characteristics of the right knee joint, left knee joint and upper limb in the step squat action; Q _s is the standard state feature vector composed of the action state characteristics of the right knee joint, left knee joint and upper limb in the horse step squat action, where W _s1 , W _s2 , and W _s3 are the standard action state characteristics of the right knee joint, left knee joint, and upper limb in the horse stance squat, respectively; A ₁ , A ₂ , and A ₃ are the right knee joint in the horse stance squat. , the weight value of the impact of the action state of the left knee joint and upper limb on the similarity of the overall action state.

The above-mentioned score calculation process of the horse stance squat action is only an embodiment of the score calculation of the state training action, and the state training action does not only include the horse stance squat action.

The training result query and reporting module is used for patients to query their own rehabilitation training scores. If the score is too low, the patient can choose to complete the rehabilitation training action again to make the action more standard and obtain better rehabilitation training effect; if the patient If you are satisfied with the score, you can choose to save the score of the rehabilitation training action and the data of the rehabilitation training action through the server system and upload it to the terminal of the medical staff, so that the rehabilitation doctor can give advice on rehabilitation training or modify the rehabilitation training plan.

The server system is used to complete the transmission and storage of interactive data among medical personnel terminals, patient personnel terminals and system management personnel terminals.

The system administrator terminal is used by the management personnel of the rehabilitation training interactive system. Through this terminal, the management personnel can complete the modification and deletion of the rehabilitation training plan set by the rehabilitation physician and the rehabilitation training action scores uploaded by the patients. At the same time, the management personnel can realize User registration and logout of medical personnel terminals and patient personnel terminals.

The technical features of the above examples can be combined arbitrarily. To make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, it should be considered It is the range described in this manual.

Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations all fall into the scope of the accompanying claims. within the limited range.

Claims (7)

1. training action classification and degree of completion scoring method in a rehabilitation training system, is characterized in that, comprises the steps: Step 1: Collect the action data information of the patient when performing rehabilitation training actions according to the guiding instructions preset in the system and the standard training action guiding video; Step 2: According to the preset training action classification criteria, classify the action data information collected in step 1, and select different completion scoring methods for different classifications; Step 3: According to the action data information collected in step 1, extract the action features, and combine the completion scoring method corresponding to step 2 to calculate the completion score of the current training action. 2. The training action classification and completion scoring method in the rehabilitation training system as claimed in claim 1 is characterized in that, the specific method of step 1 collecting action data is: collecting the patient's rehabilitation training action video, and using the bone point recognition algorithm , to obtain the patient's bone point coordinate information during the training process as the action data. 3. The training action classification and completion scoring method in the rehabilitation training system according to claim 1, wherein the preset training action classification criteria specifically include: single joint training action, multi-joint combined training action, continuous state Training actions, among them, single-joint training actions refer to the procedural actions of a single joint or limb moving from the start position to the end position; multi-joint combination training actions refer to the combination of multiple joints or limbs, moving from the start position to the end position The process of action; continuous state training action refers to the movement of multiple joints or limbs of the body to a certain state and maintain the state action for a certain period of time. 4. The training action classification and completion scoring method in the rehabilitation training system as claimed in claim 1, wherein the extracted action features include: angles between joints, normalized distances between bone points, training actions The completion time and the duration of the action state. 5. training action classification and completion scoring method in the rehabilitation training system as claimed in claim 3, is characterized in that, the specific scoring formula of single-joint training action is as follows:

Among them, F _single is the score of a single joint training action; Φ is the amplitude value of a single joint or limb movement during rehabilitation training, Φ _s is the standard value of a single joint or limb movement amplitude during rehabilitation training; T is the patient's completed The movement time of a single joint or limb training process, Ts is the standard time to complete the movement of a single joint or limb training process; A ₁ is the weight value of the influence of the range of motion on the rehabilitation training score; A ₂ is the influence of the movement completion time on the rehabilitation training score Weights. 6. training action classification and completion scoring method in the rehabilitation training system as claimed in claim 3, is characterized in that, the concrete scoring formula of multi-joint combination training action is:

Among them, F _multi is the score of multi-joint combination training action;

It is the individual score of the i-th joint or body movement during the rehabilitation training action, and 1<i<=n, n is the number of joints or body movements involved in the current multi-joint combined training action; Φ _i is the rehabilitation training action process The amplitude value of the i-th joint or limb movement; Φ _si is the standard amplitude value of the i-th joint or limb movement during rehabilitation training; T _i is the time for the patient to complete the i-th joint or limb movement; T _si is The standard time to complete the i-th joint or body movement;

It is the weight value of the impact of the i-th joint or limb movement range on the rehabilitation training score;

B i is the influence weight value of the ith joint or body movement completion time on the rehabilitation training score; B _i is the influence weight value of the i joint or body movement score on the overall score. 7. training action classification and completion scoring method in the rehabilitation training system as claimed in claim 3, is characterized in that, the concrete scoring formula of continuous state training action is:

Among them, F _state is the score of the continuous state training action; k is the action state matching coefficient of the continuous state training action; T is the action state duration of the continuous state training action; T _s is the standard action state of the continuous state training action Duration; sim(Q, Q _s ) is the similarity between the action state of the continuous state training action and the standard action state. When the similarity of the action state is greater than the similarity threshold, the cumulative action state duration is started; TH is the persistence The action state similarity threshold of the state training action; m is the number of action state features involved in the current continuous state training action judged by expert experience; Aj is the _jth action state feature in the continuous state training action. The influence weight value of the state similarity, and 1<=j<=m; Q is the state feature vector composed of all j action state features involved in the continuous state training action, where δ _j is the jth in the training action action state features; Q _s is the standard state feature vector composed of all j standard action state features involved in the continuous state training action, where δ _sj is the jth standard action state feature in the training action.

Images