CN108926814B - A Personalized Human Body Balance Training System - Google Patents

Abstract

The invention discloses a personalized human body balance training system, which consists of (1) a motion platform module (2), a signal acquisition module, (3) a signal preprocessing module, (4) an analysis module and (5) a personalized task generation module based on data driving; the method is characterized in that: the individual balance adaptability index is obtained by collecting, preprocessing and analyzing data of the human body moving along with the motion platform under 9 inclination modes of the motion platform, and then the individualized training prescription based on data driving is automatically generated. The invention can make up the blank that the personalized training prescription can not be generated in a self-adaptive manner in the current human body balance system.

Description

A Personalized Human Body Balance Training System

technical field

The invention relates to a balance training system, in particular to a human body balance training system that can be based on data-driven personalized prescriptions.

Background technique

Balance training is a task-specific training that reduces muscle tension, stabilizes blood circulation, prevents osteoporosis, improves gastrointestinal excretion, prevents joint stiffness, and improves breathing by training patients' sense of dynamic and static balance.

Balance training devices usually use two independent systems: a balance trainer and a balance measuring instrument. When using the balance measuring instrument, the user stands on the instrument and measures the balance ability of the user through a sensory interactive test that measures the patient's sense of balance and stability limit, and provides a rough semi-quantitative basis for the user's balance training. .

However, even if the semi-quantitative evaluation is known, the current balance training devices on the market can basically only provide fixed training tasks or specific mode training tasks. In other words, these devices can only provide single or statistical experience-based training tasks. However, these training devices cannot generate individualized training tasks from the results of measuring balance, and cannot meet the needs of different individuals: if the training intensity is not enough, the training effect will be greatly reduced; if the training intensity is too high, it will cause excessive harm to the user. healthy. Therefore, how to solve the personalized balance training is a major problem in the design of the current balance training device.

SUMMARY OF THE INVENTION

The purpose of the present invention is to solve the problem of automatically generating training tasks, thereby making up for the blank of the current balance training device that cannot generate personalized balance training tasks.

The present invention is realized in this way, a personalized human body balance training system is composed of a motion platform module (1), a signal acquisition module (2), a signal preprocessing module (3), an analysis module (4), and a data-driven Personalized task generation module (5) composition

In the motion platform module (1) of the method adopted in the embodiment of the present invention, the motion platform module further includes: a mechanical device that can estimate movement according to a preset in the motion platform module, the individual can stand or sit on the device, and the motion platform module pre-stores 9 81 different combined motion modes in total

Further, the preset trajectories include: left and right sine curves, front and rear sine curves, "8"-shaped curves, straight lines before and after random noise, straight lines before and after random noise, "m" shaped trace curves, elliptic curves and Arbitrary closed curve trajectory; the human body can stand or sit on the motion platform to keep the body stable; preset 81 combined motion modes, including 9 tilt modes, 3 amplitude intensity modes and 3 speed modes; among which 9 tilt modes The modes include: motion platform horizontal mode, motion platform tilt forward mode, motion platform tilt backward mode, motion platform tilt left mode, motion platform tilt right mode, motion platform tilt forward left mode, motion platform tilt left back mode , the motion platform is tilted forward to the right, and the motion platform is tilted backward to the right; 3 amplitude intensity modes include: small amplitude mode, medium amplitude mode, large amplitude mode; 3 speed modes include: slow mode, medium speed mode, fast mode model;

The embodiment of the present invention adopts the signal acquisition module (2) in the method, and further includes: during the movement of the human body with the movement platform, the movement platform, the left and right wrists of the human body, the left and right ankles of the human body, the neck, and the waist and abdomen are arranged in 7 regions. Acceleration sensor, forming 7-way time series S _i (i=1,2,3...,7);

Further, each acceleration signal starts to collect signals synchronously, the collection time is 1 minute to 15 minutes, and the sampling speed is 100Hz to 300Hz;

In the signal preprocessing module (3) of the method adopted in the embodiment of the present invention, the module further includes: the module is responsible for reconstructing the phase space of the seven acceleration signals per minute and completing the de-noising processing, so as to obtain the seven signals of the movement of the human body with the motion platform. Seven high-dimensional phase space attractors A _i (i=1,2,3...7) corresponding to the road acceleration signal

Further, the phase space reconstruction adopts the Takens phase space reconstruction method, wherein the embedded dimension m and the delay time τ are determined by the correlation dimension and mutual information method;

Further, the denoising process adopts the local manifold projection method or the principal component analysis based on the symplectic geometry of the phase space to complete the phase space denoising.

In the data analysis module (4) of the method adopted in the embodiment of the present invention, the module further includes: the module obtains the balance adaptability index of the human body under 9 tilt modes of the motion platform;

Further, the real-time coupling index refers to, in the p-th tilt mode, obtained from the trace of the coupling matrix C _p (n) of the human body and the motion platform module, further, the coupling matrix C _p (n)={ C _ijp (n)} is composed of one time series S _ip (n) (i=1,2,...,7) to other time series S _jp (n) (j=1,2,...,7) between It is composed of the similarity coefficient C _ijp (n);

The embodiment of the present invention adopts the data-driven task generation module (5) in the method, and further includes: establishing a historical database of the balance adaptability index of the individual on the exercise platform, and being able to calculate the balance adaptability of the individual based on the balance adaptability of the exercise platform under 9 tilt modes of the exercise platform. The index automatically gives personalized training prescriptions;

Further, the database collects the corresponding data of 7 regions of the exercise platform, the left and right wrists of the human body, the left and right ankles of the human body, the neck, and the waist and abdomen under the preset combined exercise mode of the exercise platform for people aged 3 to 85. 7-way acceleration data, and there are balance adaptability indices of different individuals under 9 different tilt angles. After each training, the balance adaptability indices b _p (n) (p= 1,2,…,9) store and update the database;

Further, the crowd balance adaptability index refers to: evenly dividing the balance adaptability index of different tilt angles of the platform into 9 grades;

Further, different motion modes refer to: the levels from level 1 to level 9 correspond to 9 different motion range and speed combination mode tasks of the motion platform: level 1 is a slow-speed small-amplitude mode task, and level 2 is a medium-speed small-amplitude mode task , Level 3 is fast small amplitude mode task, level 4 is slow medium amplitude mode task, level 5 is medium speed medium amplitude mode task, level 6 is fast medium amplitude mode task, level 7 is slow large amplitude mode task, 8 Level 9 is for medium-speed large-amplitude mode tasks, and level 9 is for fast-amplitude mode tasks;

Further, the amplitude is determined by the amplitude angle, and its size represents the angle between the maximum deviation position of the motion platform and the initial static position; the amplitude angle of the small amplitude mode is 5°-10°, and the amplitude angle range of the medium amplitude mode is 10°. -20°, the amplitude angle range of the large amplitude mode is 20°-30°;

Further, the speed is determined by the number of cycles completed per second, the range of the slow mode speed is to complete 1-5 cycles per second, the range of the medium speed mode speed is to complete 5-10 cycles per second, the fast mode The range of task speed is to complete 10-15 cycles of motion per second;

Further, the personalized training scheme is determined according to the individual history records, and the proportion w _p (n+1) (p=1, 2, . . . , 9) of the training tasks inclined in 9 directions is selected according to The individual's last balance adaptive ability index b _p (n) (p=1,2,...,9) generated;

Further, the proportion of training tasks inclined in each direction in the next training is:

The proportion of each oblique direction in the initial training is 1/9;

Further, when the proportion of a certain direction w _p < 0.1, the movement speed of the inclined direction is increased by one level on the original basis; when the proportion of a certain direction w _p < 0.05, the movement amplitude of the inclined direction Add a level on the original basis; by comparing with the balance adaptability index in the existing recorded population in the database, determine the initial motion amplitude and initial motion speed in different tilt directions that match the individual;

The present invention has the following beneficial effects:

This patent first uses the nonlinear dynamic method to obtain the coupling degree between the human body and the motion platform, and then obtains the overall balance energy index of the human body and the balance ability index of different parts of the human body under different motion states. The personalized balanced training method is obtained according to these indices, and the personalized training task based on data-driven is the innovation of the present invention.

The system is expected to be used in the balance training of sports athletes and rehabilitation patients. On the basis of ensuring individual safety, it can adaptively add and subtract balance training tasks, thereby significantly increasing the efficiency of human balance training.

Description of drawings

FIG. 1 is a schematic diagram of the overall system structure of the present invention.

Detailed ways

The present invention will be described below through specific embodiments, but the present invention is not limited thereto.

First, the user first enters his own information into the database, such as height, weight, age, current physical condition, etc.

Before you start training, take the measuring device and place it in a specific area. The signal acquisition module is placed on the motion platform, the left and right wrists of the human body, the left and right ankles, the neck, and the waist and abdomen of the human body. The device is a 6-axis acceleration sensor. The acquisition time is 1 minute to 15 minutes. The sampling speed is 100Hz to 300Hz, and then form 7-way time series Si ( _i =1,2,3...7);

During use, the human body can stand or sit on the motion platform to keep the body stable.

The motion platform has a variety of preset trajectories and preset 81 combined motion modes.

The preset trajectories include: left and right sine curves, front and back sine curves, "8"-shaped curves, front and rear straight lines with random noise, front and rear straight lines with random noise, "m" glyph traces, elliptic curves and arbitrary closed curve traces;

The preset 81 combined motion modes include: 9 tilt modes, 3 amplitude intensity modes and 3 speed modes; the 9 tilt modes include: motion platform horizontal mode, motion platform tilt forward mode, motion platform tilt backward Mode, Motion Platform Tilt Left Mode, Motion Platform Tilt Right Mode, Motion Platform Left Forward Tilt Mode, Motion Platform Left Backward Tilt Mode, Motion Platform Right Forward Tilt Mode, Motion Platform Right Backward Tilt Mode; 3 Amplitude Intensities Modes include: small amplitude mode, medium amplitude mode, large amplitude mode; 3 speed modes include: slow mode, medium speed mode, fast mode;

The amplitude is determined by the amplitude angle, which represents the angle between the maximum deviation position of the motion platform and the initial static position; the amplitude angle of the small amplitude mode is 5°-10°, and the amplitude angle range of the medium amplitude mode is 10°-20° , the amplitude angle range of the large amplitude mode is 20°-30°;

The speed is determined by the number of cycles completed per second. The speed range of slow mode is 1-5 cycles per second, the speed range of medium speed is 5-10 cycles per second, and the task speed of fast mode is The range is to complete 10-15 cycles of motion per second;

If the user uses it for the first time, it is necessary to select the trajectory of the motion platform, and the ratio of the 9 tilt modes is 1/9; the amplitude intensity mode is: small amplitude mode, and the speed mode is slow mode, and the result of this time is selected as the selection. Reference for platform motion modes.

The signal preprocessing module will use the Takens phase space reconstruction method, in which the embedding dimension m and the delay time τ are determined by the correlation dimension and mutual information method; and the local manifold projection method or principal component analysis based on phase space symplectic geometry is used to complete Phase space denoising.

The signal analysis module calculates the balance adaptability index b _p (n) (p=1,2,…,9) of the human body corresponding to the movement of the human body with the motion platform module under the 9 tilt modes of the exercise platform and the overall balance adaptability of the human body Index B(n).

b _p (n) (p=1,2,...,9) is obtained from the trace of the coupling matrix C _p (n) of the human body and the motion platform module, where the coupling matrix C _p (n)={C _ijp (n )} is the similarity coefficient C _ijp between one time series S _ip (n) (i=1,2,...,7) to other time series S _jp (n) (j=1,2,...,7) (n) constituted;

The phase-space coupling index C _ij is specifically realized by:

1. Local manifold-preserving structure of phase space A: In this embodiment, the prediction method is obtained by using the local manifold-preserving structure method. First, local _{linearization} is used for all points in phase space A. The 3 points closest to the Euclidean distance represent:

Among them, W _ip is the weight coefficient of the phase point x _ip in the neighborhood point group of the phase point x _i :

d _ip is the Euclidean distance between the out-of-phase points x _i and x _ip , and d _i1 is the minimum value of the Euclidean distance between the out-of-phase points x _i and x _ip

2. Obtain the predicted phase space A _ij : The prediction method is to apply any phase point xi in a certain phase space A _i to other phase spaces A _j ( _i ≠j) according to its own local manifold feature structure, Obtain the corresponding predicted phase points:

Traverse all the phase points in Ai, and all the predicted phase points x _j form the predicted phase space A _ij of A _i to A _j .

3. Obtain the coupling index C _ij of phase space A _j to phase space A _i

The time series S _ij corresponding to the predicted phase space A _ij is subjected to pairwise correlation analysis with the time series S _j of the acceleration signal corresponding to the original reconstructed phase space A _j , and the obtained correlation coefficient C _ij is used as the phase space A _j for the phase space A _i coupling degree.

When the user uses the training system for the first time, the initial personalized training task needs to be determined according to the first training, and this needs to compare the overall balance adaptive ability index B of the individual with the population in the database.

The database collects the corresponding 7-way acceleration data of the 7 regions of the motion platform, the left and right wrists of the human body, the left and right ankles, the neck, and the waist and abdomen of the 3-85-year-old people under the preset combined motion mode of the motion platform. There are different individuals in the balance adaptability index under 9 different inclination angles; according to the overall balance adaptability index of the crowd, the balance adaptability index of different inclination angles of the platform is evenly divided into 9 grades;

The exercise mode is also divided into 9 levels, which are used as the initial personalized training prescription: the levels from level 1 to level 9 correspond to 9 different motion range and speed combination mode tasks on the exercise platform: level 1 is the slow and small amplitude mode task, Level 2 is a medium-speed small-amplitude mode task, level 3 is a fast small-amplitude mode task, level 4 is a slow-speed medium-amplitude mode task, level 5 is a medium-speed medium-amplitude mode task, level 6 is a fast medium-amplitude mode task, and level 7 It is a slow-amplitude mode task, level 8 is a medium-speed large-scale mode task, and level 9 is a fast-amplitude mode task;

In each subsequent training session, the personalized training regimen is determined based on the individual's historical records, and the proportion of training tasks tilted in the 9 directions w _p (n+1) (p = 1, 2, . . . , 9) The selection is generated according to the individual's last balanced fitness index b _p (n) (p=1,2,...,9);

Further, the proportion of training tasks inclined in each direction in the next training is:

The proportion of each inclination direction in the initial training of the initial training prescription is 1/9;

Further, when the proportion of a certain direction w _p < 0.1, the movement speed of the inclined direction is increased by one level on the original basis; when the proportion of a certain direction w _p < 0.05, the movement amplitude of the inclined direction Add one level to the original.

Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art, without departing from the scope of the technical solution of the present invention, can make many possible changes and modifications to the technical solution of the present invention by using the methods and technical contents disclosed above, or modify it into an equivalent implementation of equivalent changes. example. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solutions of the present invention still fall within the protection scope of the technical solutions of the present invention.

Claims (4)

1. The utility model provides a personalized human balance training system, comprises motion platform module (1), signal acquisition module (2), signal preprocessing module (3), analysis module (4) and individualized task generating module (5) based on data drive, its characterized in that:

the motion platform module (1) is a mechanical device which can move according to a preset track, an individual can stand or sit on the mechanical device, and the motion platform module prestores 9 inclination modes, 3 motion speeds and 3 amplitude intensities, wherein the total number of the motion modes is 81 different combination motion modes;

the motion platform module and four limbs and core areas of the human body are provided with acceleration sensors, the signal acquisition module (2) is responsible for recording 7 corresponding paths of acceleration signals of 7 areas of the motion platform, the left and right wrists of the human body, the left and right ankles and the neck of the human body and the waist and abdomen in real time in the process that the human body keeps balance along with the motion of the motion platform module, and a 7-path time sequence is formed

S0_i(n)(i＝1,2,3,4,5,6,7)；

The signal preprocessing module (3) is responsible for performing phase space reconstruction on the 7 paths of acceleration signals per minute and completing denoising processing to obtain a denoised 7-path time sequence S_i(n)(i＝1,2,3,4,5,6,7)；

The data analysis module (4) is responsible for respectively obtaining the balance adaptability indexes of the human body in 9 inclination modes of the motion platform, bp (n) (p is 1,2,3,4,5,6,7,8 and 9) and the overall balance adaptability index B of the human body; the index is obtained from the trace of a coupling matrix C p between the human body and the motion platform module, where the coupling matrix C p (n) { Cij p (n) } is composed of a similarity coefficient Cij p (n) between a path of time sequence Si p (n) (i ═ 1,2,3,4,5,6,7) and other time sequence Sj p (n) (j ═ 1,2,3,4,5,6,7), where n denotes the number of times of training, p denotes different tilt modes of the motion platform, and i and j denote two paths of signals;

the personalized task generation module (5) based on data driving is responsible for enabling the individual to move on the motion platformAutomatically giving an individualized training prescription based on the balance adaptability index in 9 inclination modes, wherein a balance adaptability index historical database of an individual on a motion platform is established; the database collects corresponding 7-path acceleration data of 7 areas of a motion platform, the left wrist, the right wrist, the left ankle, the neck and the waist and abdomen of a person in a combined motion mode preset by the motion platform for people in the age range of 3 to 85 years, and balance adaptability indexes of different individuals under 9 different inclination angles are stored; evenly dividing different inclination angle balance adaptability indexes of the platform into 9 grades according to the crowd overall balance adaptability index B, and taking the grades as the reference of an initial personalized training task; the grades from 1 grade to 9 grades correspond to 9 different motion amplitude and speed combined mode tasks of the motion platform: level 1 is a slow-speed small-amplitude mode task, level 2 is a medium-speed small-amplitude mode task, level 3 is a fast small-amplitude mode task, level 4 is a slow-speed medium-amplitude mode task, level 5 is a medium-speed medium-amplitude mode task, level 6 is a fast medium-amplitude mode task, level 7 is a slow large-amplitude mode task, level 8 is a medium-speed large-amplitude mode task, and level 9 is a fast large-amplitude mode task; the amplitude is determined by an amplitude angle, and the amplitude represents the angle between the maximum deviation position and the initial rest position of the motion platform; the amplitude angle of the small amplitude mode is 5-10 degrees, the amplitude angle range of the medium amplitude mode is 10-20 degrees, and the amplitude angle range of the large amplitude mode is 20-30 degrees; the range of the slow mode speed is 1-5 periodic movements per second, the range of the medium mode speed is 5-10 periodic movements per second, and the range of the fast mode task speed is 10-15 periodic movements per second; after each training, the balance adaptability index b of the individual in 9 inclination modes of the motion platform_p(n) (p ═ 1,2,3,4,5,6,7,8,9) into and updating the database; the personalized training scheme is determined according to the individual historical records and is used for training the proportion w of training tasks inclined in 9 directions_p(n +1) (p ═ 1,2,3,4,5,6,7,8,9) is selected based on the individual's last balance adaptability index b_p(n) (p is 1,2,3,4,5,6,7,8,9), and the proportion of each training task with a slant direction appearing in the next training is:

wherein the ratio of each tilt direction occurring in the primary training is 1/9;

when the ratio w of a specific direction_p<When 0.1, the moving speed of the inclined direction is increased by one grade on the original basis; when the ratio w of a specific direction_p<When 0.05, the motion amplitude of the inclined direction is increased by one grade on the original basis; and comparing the data with the balance adaptability indexes in the recorded population in the database to determine the initial motion amplitude and the initial motion speed of the individual in different inclination directions.

2. The system of claim 1, wherein the motion platform module comprises: the platform can do preset track motion; the preset trajectory includes: left and right sinusoidal curves, front and rear sinusoidal curves, "8" -shaped curves, front and rear straight lines with random noise, "meter" -shaped trace curves, elliptic curves and any closed curve tracks; the human body can stand or sit on the motion platform to keep the body stable; 81 combined motion modes are preset, wherein the motion modes comprise 9 tilting modes, 3 amplitude intensity modes and 3 speed modes; wherein the 9 tilt modes include: a motion platform horizontal mode, a motion platform forward tilt mode, a motion platform backward tilt mode, a motion platform left tilt mode, a motion platform right tilt mode, a motion platform left front tilt mode, a motion platform left back tilt mode, a motion platform right front tilt mode, a motion platform right back tilt mode; the 3 amplitude intensity modes include: a small-amplitude mode, a medium-amplitude mode and a large-amplitude mode; the 3 speed modes include: slow mode, medium mode, fast mode.

3. The system according to claim 1, wherein the signal acquisition module comprises: in the process that the human body moves along with the motion platform, 7 paths of corresponding acceleration signals of 7 areas of the motion platform, the left wrist, the right wrist, the left ankle, the right ankle, the neck and the waist and abdomen of the human body are obtained; each acceleration signal synchronously starts to collect signals, the collection time is 1 minute to 15 minutes, and the sampling speed is 100Hz to 300 Hz.

4. The system according to claim 1, wherein the signal preprocessing module comprises: adopting a Takens phase space reconstruction method, wherein the embedding dimension m and the delay time tau are determined by a correlation dimension and mutual information method; and the phase space denoising is finished by adopting a local manifold projection method or principal component analysis based on the phase space octave geometry.

Images