Training method and system based on three-dimensional attitude evaluation

Abstract

The invention relates to the technical field of rehabilitation training, in particular to a training method and system based on three-dimensional posture evaluation, and the method comprises the steps: collecting training data of a user on a static and dynamic balance plate; the standing posture instability of the user in the static state is calculated, wherein the standing posture instability reflects the uniformity of pressure distribution of the user in four quadrants on the static balance plate; the overall trajectory length of the user on the center of gravity of the dynamic plate and the overall stability index of the dynamic plate are used. The static and dynamic balance plates are used for simultaneously collecting pressure distribution and IMU attitude data, the static standing attitude instability coefficient and the dynamic instability coefficient of the user are quantified respectively, the composite instability index is obtained through normalization and weighting calculation, the comprehensive ability of the user in the two aspects of static attitude control and dynamic balance reaction is visually reflected, and the user experience is improved. And the accuracy of user balance capability evaluation is improved, and a personalized training target is accurately formulated.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of rehabilitation training. More particularly, the present application relates to a three-dimensional posture evaluation training method and system. BACKGROUND

[0002] Static and dynamic balance boards mainly measure and evaluate the posture stability of users in static and dynamic environments through pressure sensors and inertial measurement units (IMUs) and other sensors. The basic working principle is: in static mode, the balance board is fixed, the user stands on the board, and the body center of gravity is judged by monitoring the pressure distribution of each area of the sole; in dynamic mode, the balance board can freely tilt or swing in multiple directions, and the balance keeping ability of the user under external disturbance is analyzed by means of the board posture change and human body reaction data. This kind of device has been widely used in rehabilitation training, sports performance evaluation and fall risk screening fields, and can provide certain objective data support.

[0003] However, the existing technology still has certain deficiencies in comprehensive evaluation of user balance ability. On the one hand, most methods only evaluate static posture stability or dynamic response ability separately, lack of comprehensive quantitative indicators combining the two, leading to one-sided training scheme, which cannot accurately reflect the overall balance of the user. On the other hand, some systems only rely on single indicators such as board tilt angle or pressure center trajectory in dynamic evaluation, ignoring important information such as trunk angular velocity reflecting human core reaction speed and amplitude, making it difficult to fully reflect the real adjustment ability of the user under sudden disturbance. These deficiencies will lead to low accuracy of training prescription, which may cause the training difficulty to be unmatched with the actual ability of the user, affecting the rehabilitation effect and safety. SUMMARY

[0004] The present application provides a three-dimensional posture evaluation training method and system, aiming to solve the problem that it is difficult to fully reflect the real adjustment ability of the user under sudden disturbance in related technologies, which may cause the training difficulty to be unmatched with the actual ability of the user, and then affect the rehabilitation effect.

[0005] In a first aspect, the present application provides a three-dimensional posture-based training method, comprising: collecting training data of a user on a static and dynamic balance board; calculating standing posture instability of the user in a static state, wherein the standing posture instability reflects uniformity of pressure distribution of the user in four quadrants on the static balance board; calculating a dynamic instability coefficient of the user based on a total trajectory length of a center of gravity of the user on the dynamic balance board and a total stability index of the dynamic balance board, and performing weighted summation of the dynamic instability coefficient of the user and the standing posture instability of the user in the static state to obtain a composite instability index of the user; and determining a training target of the user according to the size of the composite instability index of the user. By fusing the static posture instability and the dynamic instability coefficient, a unified composite instability index is constructed, comprehensive quantitative evaluation of balance ability is realized, and personalized training targets can be accurately formulated according to the index results, which not only improves the scientificity and pertinence of rehabilitation training, but also effectively identifies high-risk groups and improves rehabilitation effect and safety.

[0006] Further, the method for obtaining the four quadrants comprises: dividing an upper surface area of the static balance board into four quadrants, and the four quadrants correspond to left front, right front, left rear and right rear respectively.

[0007] Further, the calculation of the standing posture instability of the user in the static state comprises: locking the dynamic balance board in the static and dynamic balance board; collecting pressure values of the user standing on the static and dynamic balance board for a period of time; respectively counting pressure values in the four quadrants, and calculating the standing posture instability of the user in the static state according to differences between the pressure values in the four quadrants.

[0008] Further, the calculation of the standing posture instability of the user in the static state comprises: locking the dynamic balance board in the static and dynamic balance board; collecting pressure values of the user standing on the static and dynamic balance board for a period of time; respectively counting pressure values in the four quadrants, and calculating the standing posture instability of the user in the static state according to differences between the pressure values in the four quadrants. ; in the formula, represents the standing posture instability of the user in the static state for a preset period of time, represents an average pressure value collected in the left front quadrant of the static balance board for a preset period of time, represents an average pressure value collected in the right front quadrant of the static balance board for a preset period of time, represents an average pressure value collected in the left rear quadrant of the static balance board for a preset period of time, represents an average pressure value collected in the right rear quadrant of the static balance board for a preset period of time. By collecting average pressure values of the four quadrants of the static balance board and calculating the proportion of the front and rear force difference in the total pressure, the degree of center of gravity front and rear offset of the user in the static standing state can be quickly, intuitively and stably reflected.

[0009] Further, the calculation formula of the composite instability index of the user is: ; in the formula, represents the composite instability index of the user, representing the standing posture instability of the user, representing the dynamic instability coefficient of the user, representing the weight corresponding to the standing posture instability of the user, representing the weight corresponding to the standing posture instability of the user.

[0010] Further, the dynamic instability coefficient of the user is calculated, and the calculation formula is: In the formula, representing the dynamic instability coefficient of the user, representing the peak value of the sagittal plane angular velocity mutation monitored by the chest IMU in a period of time, representing the overall stability index of the dynamic board. By calculating the dynamic instability coefficient and combining the peak value of the sagittal plane angular velocity mutation and the stability index, the performance of the user in dynamic balance can be accurately reflected.

[0011] Further, according to the size of the composite instability index of the user, the training target of the user is determined, including: according to the size of the composite instability index of the user, dividing into three training intervals, and one training interval corresponds to one training target.

[0012] Further, the three training intervals are divided, including: dividing greater than a first threshold value into a high-risk interval, dividing greater than or equal to a second threshold value and less than the first threshold value into a medium-risk interval, and dividing less than or equal to the second threshold value into a low-risk interval.

[0013] Further, the experience value of the first threshold value is 0.7, and the experience value of the second threshold value is 0.4. The second aspect of the present application also provides a three-dimensional posture evaluation training system based on the three-dimensional posture evaluation training method, including a processor and a memory, the memory stores a computer program, and the processor executes the computer program to realize the three-dimensional posture evaluation training method.

[0014] Beneficial effects: the static and dynamic balance board is used to simultaneously collect pressure distribution and IMU posture data, the static standing posture instability and the dynamic instability coefficient of the user are quantified respectively, the composite instability index is calculated through normalization and weighting, the comprehensive ability of the user in static posture control and dynamic balance reaction is directly reflected, and the precision of the balance ability evaluation of the user is improved, and the personalized training target is accurately formulated. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 is a flowchart schematically showing determination of a training target according to an embodiment of the present application. DETAILED DESCRIPTION

[0016] The specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0017] As shown in Figure 1 S101: collecting data.

[0018] Specifically, the patient stands on the static and dynamic balance plate, wherein the static balance plate is divided into four quadrants, respectively corresponding to the pressure distribution of the left front ( ), right front ( ), left rear ( ) and right rear ( ) four areas. And the static and dynamic balance plate is provided with a pressure sensor and an IMU gyroscope, which is used to collect the pressure value of the static and dynamic balance plate when the user stands on the static and dynamic balance plate.

[0019] S102: calculating the standing posture instability of the user in static state.

[0020] Specifically, the dynamic balance plate in the static and dynamic balance plate is locked, the user stands on the static and dynamic balance plate, and then the pressure value of the user standing on the static balance plate within a preset time period is collected, and the standing posture instability of the user in static state is calculated according to the pressure distribution of the user standing on the static balance plate. The standing posture instability and the uniformity of the pressure distribution in the four quadrants are related. The more uniform the pressure distribution in the four quadrants is, the higher the standing posture instability of the user in static state is, and vice versa.

[0021] As described above, a calculation formula for calculating the standing posture instability of the user in static state is provided, and the formula is: . In the formula, represents the standing posture instability of the user in static state within a preset time period, represents the average pressure value collected in the left front quadrant of the static balance plate within a preset time period, represents the average pressure value collected in the right front quadrant of the static balance plate within a preset time period, represents the average pressure value collected in the left rear quadrant of the static balance plate within a preset time period, represents the average pressure value collected in the right rear quadrant of the static balance plate within a preset time period.

[0022] It can be understood that when the patient's body center of gravity is perfectly centered, the pressure distribution of the patient in the left-right direction is theoretically balanced, that is, the value of (Q1+Q3) is approximately equal to the value of (Q2+Q4), at this time the numerator of the formula tends to 0, and the calculated standing posture instability value also tends to 0, which indicates that the patient's static posture is very stable. On the contrary, if the patient's body has obvious sway or deviates to one side, the total pressure on one side will be significantly greater than that on the other side, causing the value of the numerator to increase, and the value of the calculated standing posture instability will also increase, thereby directly reflecting the instability degree of the static posture. The value of the standing posture instability will be used as a key indicator reflecting the initial state of the patient.

[0023] S103: Calculate the dynamic instability coefficient of the user.

[0024] After completing the static evaluation, the system will suddenly release the lock of the dynamic plate, so that it can be freely moved, and the data within a period of time after the dynamic plate is released is counted, which can be 3 seconds or 5 seconds, etc. The purpose of this step is to evaluate the patient's rapid response and adjustment ability from the expected stable state to the unexpected dynamic disturbance state, and in order to comprehensively evaluate the patient's dynamic adjustment ability and potential fall risk in this transition process, a dynamic instability coefficient is constructed. The basis of its construction is that an efficient balance response not only requires a small body core (trunk) sway amplitude, but also requires a stable and effective adjustment trajectory of the foot support surface (dynamic plate).

[0025] According to the above, a calculation formula for calculating the dynamic instability coefficient of the user is provided, and the formula is: . In the formula, represents the dynamic instability coefficient of the user, represents the peak value of the sudden change of the angular velocity in the sagittal plane monitored by the chest IMU within a period of time, represents the overall stability index of the dynamic plate, which is calculated by the IMU data built in the dynamic plate, and reflects the posture stability degree of the plate within a period of time, represents the overall trajectory length of the center of gravity of the dynamic plate, which reflects the moving distance of the center of pressure of the patient's foot to maintain balance. Wherein, the period of time is a self-defined time, which can be 10 seconds or 20 seconds, etc. The peak value refers to the maximum instantaneous change value of the angular velocity during the movement, which usually appears when the balance is disturbed, the start / stop action is started, or the posture is suddenly adjusted. For example, in the balance test, when the platform is suddenly tilted or the body is subjected to external force, the body will quickly adjust the posture in the sagittal plane, causing an obvious peak in the angular velocity curve, and the value of the peak is the peak value.

[0026] For calculating the overall stability index , a calculation formula is provided, and the calculation formula is: . In the formula, represents the overall stability index of the dynamic board in a period of time, is the standard deviation of the Euler angles (or the attitude angles converted from quaternions) of the board surface in a period of time; epsilon is added to avoid division by zero, and the empirical value of epsilon is 1. The Euler angles are actually used to describe the attitude (orientation and inclination) of the dynamic balance board in three-dimensional space by using three angles (roll angle: the angle of inclination of the board surface to the left and right; pitch angle: the angle of inclination of the board surface to the front and back; yaw angle: the angle of rotation of the board surface in the horizontal plane), and the angle data can be obtained by the gyroscope on the static and dynamic balance board.

[0027] It can be seen from the formula that the value of the dynamic instability coefficient of the user is positively correlated with the peak value of the sudden change of the chest angular velocity , and is negatively correlated with the product of the stability index of the dynamic board and the center of gravity trajectory . When a patient has poor dynamic adjustment ability, the trunk will often produce violent and large amplitude shaking to maintain balance, resulting in a significant increase in the value of . At the same time, the adjustment strategy of the foot bottom may be inefficient, which is manifested in that the board surface shakes violently (the value of is low) and the center of gravity moves in a long and disordered trajectory (the value of is large), and these factors together cause the value of the dynamic instability coefficient of the user to be low.

[0028] It should be noted that the lower the value of the dynamic instability coefficient, the more serious the static posture imbalance (the standing posture instability is high), or the poorer the dynamic adjustment ability, the higher the risk of falling, and the balance training needs to be strengthened. The higher the value of the dynamic instability coefficient, the lower the standing posture instability, and the more efficient the dynamic adjustment, the better the balance ability, and the higher difficulty training can be performed.

[0029] S104: Calculate the composite instability index of the user.

[0030] Specifically, the dynamic instability coefficient of the user obtained in the above step and the standing posture instability of the user are both normalized, then the dynamic instability coefficient of the user and the standing posture instability of the user are weighted and summed, and the value after the weighted sum is taken as the composite instability index of the user.

[0031] As described above, the calculation formula of the composite instability index is provided, and the calculation formula is: . In the formula, represents the composite instability index of the user, represents the standing posture instability of the user, represents the dynamic instability coefficient of the user, a weight corresponding to the standing posture instability of the user, a weight corresponding to the standing posture instability of the user.

[0032] The greater the value of the composite instability index of the user, the higher the comprehensive instability of the user in static posture control and dynamic adjustment ability, the greater the risk of falling, and the training prescription should be reduced in difficulty and increased in balance ability training. The smaller the value of the composite instability index of the user, the better the balance ability of the user, and the training intensity can be gradually increased.

[0033] For the weight, The weight corresponding to the static instability reflects the influence degree of the static posture instability on the composite instability index in the comprehensive evaluation, The greater the value of the weight, the more important the influence of the static posture, that is, even if the dynamic performance is good, as long as the static performance is poor, the composite instability index will be pulled up. The weight corresponding to the dynamic instability reflects the influence degree of the dynamic instability on the composite instability index in the comprehensive evaluation, The greater the value of the weight, the more important the dynamic adjustment ability in the composite instability index, that is, poor dynamic response will strongly pull up the composite instability index.

[0034] S105: Determine the training target according to the composite instability index of the user.

[0035] Specifically, according to the size of the composite instability index of the user, three training intervals are divided. For example: greater than the first threshold value is divided into a high-risk interval, greater than or equal to the second threshold value and less than the first threshold value is divided into a medium-risk interval, and less than or equal to the second threshold value is divided into a low-risk interval. The empirical value of the first threshold value is 0.7, and the empirical value of the second threshold value is 0.4. Then, according to the training interval corresponding to the composite instability index of the current user, the training target in the training interval is obtained, so that the user performs training. The higher the training interval, the simpler the difficulty of training.

[0036] For example, if the composite instability index of the current user falls into the high risk interval, it indicates that the static ability and dynamic ability of the current user are both unstable, and the static training target is to maintain a standing posture for 30-60s. The dynamic training target is low-intensity dynamic disturbance (small amplitude and slow speed). If the composite instability index of the current user falls into the medium risk interval, it indicates that the static ability and dynamic ability of the current user are both unstable, and the static training target is a medium-difficulty static task (standing with eyes closed, narrowing the standing distance). The dynamic training target is to introduce multi-directional disturbance. If the composite instability index of the current user falls into the low risk interval, it indicates that the static ability and dynamic ability of the current user are both unstable, and the static training target is a high-difficulty static task (single leg, soft pad, head rotation). The dynamic training target is high-frequency and large-amplitude disturbance.

[0037] The application further provides a three-dimensional posture evaluation training system. The system comprises a processor and a memory, and the memory stores computer program instructions which, when executed by the processor, implement the three-dimensional posture evaluation training method according to the first aspect of the application.

[0038] The system further comprises a communication bus and a communication interface and other components well known to those skilled in the art, the settings and functions of which are known in the art, and thus will not be described here.

[0039] In the present application, the aforementioned memory can be any tangible medium containing or storing a program which can be used by or in conjunction with an instruction execution system, device or apparatus. For example, the computer readable storage medium can be any suitable magnetic storage medium or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random-Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC) and the like, or any other medium that can be used to store desired information and can be accessed by an application, module or both. Any such computer storage medium can be part of the device or accessible or connectable to the device. Any application or module described in the present application can be implemented using computer readable / executable instructions stored or otherwise held by such computer readable medium.

[0040] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation to the scope of the application. It should be noted that for those skilled in the art, without departing from the concept of the present application, several modifications and improvements can be made, which are all within the protection scope of the present application.

Claims

1. A training method and system based on three-dimensional pose evaluation, characterized in that, include: Collect training data from users on static and dynamic balance boards; The user's standing posture instability under static conditions is calculated, where the standing posture instability reflects the uniformity of the pressure distribution of the user in the four quadrants of the static balance plate. Based on the overall trajectory length of the user's center of gravity on the dynamic board and the overall stability index of the dynamic board, the user's dynamic instability coefficient is calculated, and the user's dynamic instability coefficient and the user's standing posture instability under static conditions are weighted and summed to obtain the user's composite instability index. The user's training objective is determined based on the magnitude of the user's composite instability index.

2. The training method and system based on three-dimensional pose evaluation according to claim 1, characterized in that, The methods for obtaining the four quadrants include: The upper surface area of ​​the static balance plate is divided into four quadrants, which correspond to the four regions of left front, right front, left rear and right rear, respectively.

3. The training method and system based on three-dimensional pose evaluation according to claim 1, characterized in that, The instability of a user's standing posture in a static state is calculated, including: Lock the dynamic balance plate in the static-dynamic balance plate; Collect pressure values ​​of a user standing on a static and dynamic balance plate over a period of time; The pressure values ​​in each of the four quadrants are counted separately, and the user's static standing posture instability is calculated based on the differences between the pressure values ​​in the four quadrants.

4. The training method and system based on three-dimensional pose evaluation according to claim 3, characterized in that, The instability of a user's standing posture in a static state is calculated using the following formula: ； In the formula, This indicates the instability of the user's standing posture during a preset time period while in a static state. This represents the average pressure value collected in the left front quadrant of the static balance plate over a preset time period. This represents the average pressure value collected in the right front quadrant of the static balance plate over a preset time period. This represents the average pressure value collected in the left rear quadrant of the static balance plate over a preset time period. This represents the average pressure value collected in the right rear quadrant of the static balance plate over a preset time period.

5. The training method and system based on three-dimensional pose evaluation according to claim 1, characterized in that, The formula for calculating the user's composite instability index is: ； In the formula, This represents the user's composite instability index. This indicates that the user's standing posture is unstable. This represents the user's dynamic instability coefficient. This represents the weight corresponding to the instability of the user's standing posture. This represents the weight corresponding to the instability of the user's standing posture.

6. The training method and system based on three-dimensional pose evaluation according to claim 1, characterized in that, The user's dynamic instability coefficient is calculated using the following formula: In the formula, This represents the user's dynamic instability coefficient. This indicates the peak value of the sagittal angular velocity abruptly detected by the chest IMU over a period of time. This represents the overall stability index of the dynamic board.

7. The training method and system based on three-dimensional pose evaluation according to claim 1, characterized in that, Based on the magnitude of the user's composite instability index, determine the user's training objectives, including: Based on the magnitude of the user's composite instability index, the training is divided into three intervals, with each interval corresponding to a training objective.

8. The training method and system based on three-dimensional pose evaluation according to claim 1, characterized in that, The training is divided into three intervals, including: Those exceeding the first threshold are classified as high-risk, those exceeding or equal to the second threshold but less than the first threshold are classified as medium-risk, and those less than or equal to the second threshold are classified as low-risk.

9. The three-dimensional pose evaluation training method and system according to claim 8, characterized in that, The empirical value of the first threshold is 0.7, and the empirical value of the second threshold is 0.

4.

10. A three-dimensional pose evaluation training system, comprising a processor and a memory, characterized in that, The memory stores a computer program, and the processor executes the computer program to implement the three-dimensional pose evaluation training method and system as described in any one of claims 1-9.

Images