CN106539587A - A kind of fall risk assessment and monitoring system and appraisal procedure based on sensor of doing more physical exercises - Google Patents
A kind of fall risk assessment and monitoring system and appraisal procedure based on sensor of doing more physical exercises Download PDFInfo
- Publication number
- CN106539587A CN106539587A CN201611123591.7A CN201611123591A CN106539587A CN 106539587 A CN106539587 A CN 106539587A CN 201611123591 A CN201611123591 A CN 201611123591A CN 106539587 A CN106539587 A CN 106539587A
- Authority
- CN
- China
- Prior art keywords
- module
- angle
- acceleration
- gait
- data analysis
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/1116—Determining posture transitions
- A61B5/1117—Fall detection
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/6813—Specially adapted to be attached to a specific body part
- A61B5/6823—Trunk, e.g., chest, back, abdomen, hip
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/6813—Specially adapted to be attached to a specific body part
- A61B5/6829—Foot or ankle
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/683—Means for maintaining contact with the body
- A61B5/6831—Straps, bands or harnesses
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7253—Details of waveform analysis characterised by using transforms
- A61B5/7257—Details of waveform analysis characterised by using transforms using Fourier transforms
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Veterinary Medicine (AREA)
- Public Health (AREA)
- Surgery (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Physiology (AREA)
- Artificial Intelligence (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Mathematical Physics (AREA)
- Dentistry (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The present invention proposes a kind of fall risk assessment based on three nine axis movement sensors and monitoring system and appraisal procedure, it is possible to achieve tumble state-detection, fall risk assessment, falling protective device control.System includes：Information acquisition module, wireless communication module and data analysis module.39 axis movement sensors, two of which are mainly used to be placed on two shoe linings, one is placed on waist, obtains the initial data of acceleration, angular speed, angle by three sensors, carries out data signal transmission using bluetooth module, be transferred to data analysis terminal.By analysis of the motion sensor to various data-signals, detection is fallen down state and combines Tinetti evaluation and test table criterion evaluation risk of falls, feeds back to tester.Invention additionally discloses a kind of gait evaluation method, mainly by the multi-signal type such as acceleration, angular speed, angle, is modeled to human body, gait feature is extracted, is estimated with reference to Tinetti.
Description
Technical field
The invention belongs to medical treatment ＆ health, signal analysis calculating and machine learning techniques field, more particularly to it is a kind of based on many
The fall risk assessment of motion sensor and monitoring system and appraisal procedure.
Background technology
In today's society, with economic continuous development, the problem of an aging population is increasingly sharpened, and its elderly's is strong
Health safety problem also becomes the focus of people's growing interest.Wherein, write in sanitation Ministry tissue in 2007《Chinese injury prevention
Report》In, tumble is the first cause that old man is subject to unexpected injury.After generation of falling, on the one hand it is tumble institute band itself
Carry out human injury, be on the other hand fall after succoured without timely safeguard procedures and timely the secondary wound for causing
Evil.This has a strong impact on the normal life ability and the state of mind of the elderly.
At present, some companies and research and development institution are also dedicated to based on sensor, for the anti-tumble of old man in terms of carry out
Research, such as 2012, " tumble automatic help mobile phone " Ai Fulai A03 that Shenzhen Ai Fulai Science and Technology Ltd.s release could
Emergency is automatically positioned and is detected after Falls Among Old People；2015, " tumble autoalarm and its a report were invented by Shenzhen University
Alarm method ", can detect the transducing signal of human body tumble and be sent to specified unit.But, current research more stops
How to avoid in the aspect of secondary damage after staying in tumble, energy is concentrated on and protects old from source by little researcher
People, it is to avoid once injure.
Meanwhile, in terms of the risk assessment fallen.The patent of Publication No. CN202086479 describes a kind of human body less
Analysis method, its by plantar pressure sensor gather characteristics of human body, but system complex, it is not portable, three-dimensional ginseng can not be reflected
Number, function are limited；Publication number CN20120209092's describes a kind of body gait evaluating system and method, main by adopting
Collection three-dimensional acceleration signal carries out gait analysis to human body, but is only the integrality that acceleration signal can not calculate human body, essence
Exactness is restricted.
The content of the invention
For above-mentioned deficiency, the present invention proposes a kind of fall risk assessment based on sensor of doing more physical exercises and monitoring system
And appraisal procedure, the present invention can realize tumble state-detection, fall risk assessment, solve traditional fall monitoring system pre-
On the other hand anti-tumble and the deficiency in terms of tumble is protected, it is bulky that integrated device also solves conventional detection devices,
The low problem of detection efficiency and accuracy rate.The apparatus function is complete, and integrated level is high, and control is simple, is easy to combine with other designs
Using.
The technical scheme adopted in the present invention is as follows：A kind of fall risk assessment based on multiple nine axis movement sensors and
Monitoring system, the system include information acquisition module, wireless transport module and data analysis module,
Described information acquisition module to measure and process both feet and waist nine axle motor messages, including acceleration,
Angular speed, angle, the nine axle motor messages after the wireless communication module is by process are sent out by the wireless communication module
Give the data analysis module；
The wireless communication module is to carry out the data between the signal acquisition module and the data analysis module
Exchange；
The data analysis module includes signal processing module, fall monitoring module and risk evaluation module；Wherein,
Described signal processing module is carried out to human body for being filtered denoising and frequency-domain analysis to the signal for gathering
Modeling, extracts body gait feature；
The fall monitoring module is weighted process for the body gait feature for extracting signal processing module, obtains
Overall target, judges whether to fall down；
The risk evaluation module carries out neural metwork training for the body gait feature for extracting signal processing module,
Comprehensive evaluation result is obtained, tester or doctor is fed back to.
Further, each information collection node includes nine axis movement sensors and its peripheral circuit and wireless receiving and dispatching
Module；Wherein,
Nine axis movement sensor is used for acceleration, angular speed and the angle initial data for measuring evaluated position；
The peripheral circuit includes power-supplying circuit and hardware signal processing circuit, and it is right that the power-supplying circuit is used for
Nine axis movement sensors are powered, and the hardware signal processing circuit is for nine axles that obtain to nine axis movement sensor measurements
Motor message carries out Dynamic Kalman Filtering；
The radio receiving transmitting module passes through described wireless for the nine axle motor messages for gathering described information acquisition node
Communication module is sent to the data analysis module, and is sent by the wireless communication module reception data analysis module
Feedback and control instruction.
Further, signal processing module is filtered denoising and coordinate system transformation to the signal for gathering.It is specific as follows：
First the signal to gathering carries out Fast Fourier Transform (FFT), and is filtered on this basis, filters off high-frequency noise interference；Then
Coordinate system change is carried out by attitude angle to acceleration, angular speed, obtain human motion absolute acceleration under world coordinates with
Angular speed.
Further, signal processing module carries out mathematical modeling to human body, specific as follows：By the trunk of human body, thigh, little
Leg is equivalent to inverted pendulum respectively, by kinematics analysis, leg length is calculated with step-length, is that revolute is fast by angle with human synovial
Calculating speed is spent, speed is obtained with integrated acceleration, foot is obtained with rate integrating and the relative position of waist, the relative position of combination
The overall attitude that limbs are analyzed with leg length is put, so as to the motion state to whole human body is modeled, overall motion number is obtained in that
According to.
Further, the body gait feature includes：When step-length, foot height, sufficient angle, cadence, leg speed, lower limb standing phase
Between ratio, body centre's acceleration, the limiting value of lower limb trunk angle, centre of body weight vertical direction displacement and in the short time
Exercise data dominant frequency and amplitude in the section time.
Further, fall monitoring module is by human body absolute acceleration, body and ground-angle, centre of body weight in the short time
Vertical direction displacement is weighted, and obtains overall target, judges whether to fall down, and weighting weight passes through neutral net
The gradient descent method of model is constantly modified.
Further, risk evaluation module sets up neural network model, is divided into 3 layers, one layer include step-length, step width, foot it is high,
Sufficient angle, cadence, leg speed, the limiting value of the ratio, body centre's acceleration and lower limb trunk angle of lower limb standing phase time；Two
Layer include starting stage normality, mop floor with the presence or absence of single pin, biped gait symmetry, paces continuity, trunk stability with
And muscle enabling capabilities；Three layers of output includes that gait is healthy, gait is unhealthy and various gait illnesss；Described neutral net
Model has parameter (W, b)=(W(1), b(1), W(2), b(2)), wherein W(i)For parameter composition is linked between i-th layer to i+1 layer
Matrix, matrixes of the b (i) for the bias term composition of i+1 layer；The activation value of ground floor is input, and last layer activation value passes through chain
Connect the input value that parameter weighting calculates next layer, input value is calculated next layer by activation primitive sigmoid functions and swashs
Value living, transmission are exported；
Training process parameters weighting attenuation parameter λ and learning rate α is first set in modeling process, by cost function
Partial derivative is asked to be iterated to parameter is linked between each layer.
A kind of method being estimated using above-mentioned system on human body gait, it is characterised in that comprise the following steps：
Wearable nine axis movement sensor is worn on waist and biped information collection node by step 1. respectively, enters rower
It is fixed, the three-dimensional motion acceleration of collection human motion, angular speed, angle, and carry out hardware signal processing；
The nine axle motor messages that described information acquisition node is gathered by radio receiving transmitting module by step 2. are by the nothing
Line communication module is sent to the data analysis module, and receives the data analysis module by the wireless communication module
The feedback come and control instruction.
Data analysis module described in step 3. calculates human body motion feature to be included：Step-length, step width, foot are high, sufficient angle, step
Frequently, leg speed, the ratio of lower limb standing phase time, body centre's acceleration, the limiting value of lower limb trunk angle, body weight in the short time
The displacement of heart vertical direction and a period of time interior exercise data dominant frequency and amplitude.
Step 4. integrative medicine using Tinetti evaluation and test tables standard is quantified；The human body motion feature is led to
Cross neural network model to be reflected in medical science judgment criteria, evaluate starting stage normality, mop floor with the presence or absence of single pin, biped step
State symmetry, paces continuity, trunk stability and muscle enabling capabilities.
Step 5. is calculated to the gait assessment criteria scores listed by Tinetti evaluation and test tables by neural network model,
Obtain fall risk coefficient.
The invention has the beneficial effects as follows：Motion sensor small volume, it is easy to carry, strong adaptability is applicable to all trouble
Person and most of environment；Man-machine harmony, substantially reduces user's burden；Add angular speed to gather with rate signal, can calculate
For the more multi objective analyzed, evaluate, recognition accuracy is high, the model of neural metwork training is more accurate；Using wearable
Sensor and real time computation system, can realize real-time protection, from damaging that source greatly reduces that the elderly is subject to；Low cost,
Can reduce detecting and protect cost.
Description of the drawings
The present invention is further described with case study on implementation below in conjunction with the accompanying drawings.
Fig. 1 is to drop to risk assessment and monitoring system composition schematic diagram；
Fig. 2 is to drop to risk assessment and monitoring analysis process figure；
Specific embodiment
To make the object, technical solutions and advantages of the present invention become more apparent, below in conjunction with specific embodiment, and reference
Accompanying drawing, the present invention is described in more detail.
Fig. 1 is the structural representation of body gait evaluating system of the present invention, as shown in figure 1, the body gait of the present invention is commented
Examining system includes：Information acquisition module, wireless communication module and data analysis module.
Wherein, described information acquisition module is to measure and process nine axle motor messages of both feet and waist, including angle
Speed, angular speed, angle, receive the instruction that the data analysis terminal sends by the wireless communication module, and will process
Nine axle motor messages afterwards are sent to the data analysis terminal by the wireless communication module；
Described information acquisition module includes three information collection nodes, respectively positioned at biped and waist；Each information is adopted
Collection node all includes nine axis movement sensors and its peripheral circuit and radio receiving transmitting module；Wherein, the nine axles motion-sensing
Device is MPU9250, including accelerometer and gyroscope, for measuring the originals such as acceleration, angular speed and the angle at evaluated position
Beginning data；The peripheral circuit includes power-supplying circuit and hardware signal processing circuit, and it is right that the power-supplying circuit is used for
Accelerometer and gyroscope are powered, using sheet lithium battery；The hardware signal processing circuit is for nine axles motion biography
The nine axle motor messages that sensor measurement is obtained carry out Dynamic Kalman Filtering, filter noise；
The radio receiving transmitting module passes through described wireless for the nine axle motor messages for gathering described information acquisition node
Communication module is sent to the data analysis module, and is sent by the wireless communication module reception data analysis terminal
Feedback and control instruction, wireless communication module uses bluetooth communication or wifi communication modules.
The signal processing module is filtered denoising and coordinate system transformation to the signal for gathering.It is specific as follows：It is first right
The spectrum distribution that Fast Fourier Transform (FFT) is carried out from motor message, signal is obtained of collection, filters off high fdrequency component, filters off height with this
Frequency noise jamming；Then attitude of the human body under world coordinate system is obtained by the attitude angle measured by sensor, by triangle
Computing carries out coordinate system transformation to acceleration, angular speed, obtains human motion absolute acceleration and angular speed under world coordinates.
The signal processing module carries out mathematical modeling to human body, specific as follows：By the trunk of human body, thigh, shank point
Inverted pendulum is not equivalent to, by kinematics analysis, step-length is obtained with the quadratic integral of acceleration, leg is estimated by step-length long；
Leg is calculated with the amplitude of centre of body weight displacement and the limiting value at foot gesture angle simultaneously long, mutually corrected；With model of human ankle
It is that revolute passes through angular speed calculation speed, the integral operation of waist is corrected with this, counting accuracy is improved；With acceleration one
Secondary integration obtains speed；Relative displacement between limbs is obtained with rate integrating, so as to obtain the relative position of foot and waist；It is logical
Cross the movement calculation method of rational mechanics, it is known that relative position, leg length and attitude angle, analyze the mass motion of human body, to whole
The motion state modeling of human body, is obtained in that overall exercise data.
Described integral operation, by integral process to the amendment of speed improving integral operation accuracy；Amendment
Mode include judgement to foot movement speed zero point, by the angular speed calculation angle modification waist speed of ankle-joint revolute
Degree.
The body gait feature includes：Step-length, foot are high, sufficient angle, cadence, leg speed, the ratio of lower limb standing phase time, body
Body central acceleration, the limiting value of lower limb trunk angle, in the short time displacement of centre of body weight vertical direction and a period of time in
Exercise data dominant frequency and amplitude.Step-length is as obtained by foot horizontal direction acceleration quadratic integral computing；Sufficient high pass foot
Obtained by vertical direction acceleration quadratic integral；Sufficient angle carries out plus-minus by the attitude angle that sensor is collected and obtains；Cadence
For the inverse of cycle time average after substep；Leg speed is step-length divided by cycle time；Displacement of the ratio of lower limb standing phase for waist
In the ratio of ascent stage and the time of decline stage；Centre of body weight has three-dimensional acceleration, and the acceleration around different axles leads to
Cross three-dimensional initial data to be synthesized；The limiting value of lower limb trunk angle is obtained by upper body attitude angle limiting value plus-minus；Short time
Interior centre of body weight vertical direction is displaced through obtained by the vertical direction acceleration quadratic integral of waist；Exercise data in a period of time
Exercise data of the dominant frequency with amplitude for some time carries out Fourier transformation and obtains.
Centre of body weight is vertical by human body absolute acceleration, body and ground-angle, in the short time for the fall monitoring module
Direction displacement is weighted, and obtains overall target, judges whether to fall down, and weighting weight passes through gradient descent method not
Disconnected to be modified, amendment concrete mode is design cost function, is carried out come the parameter to weighting by local derviation is sought to cost function
Iteration.
The risk evaluation module sets up neural network model, is divided into 3 layers, and one layer includes that step-length, step width, foot height, foot are pressed from both sides
Angle, cadence, leg speed, the limiting value of the ratio, body centre's acceleration and lower limb trunk angle of lower limb standing phase time；Two layers of bag
Include starting stage normality, mop floor with the presence or absence of single pin, biped gait symmetry, paces continuity, trunk stability and flesh
Meat enabling capabilities；Three layers of output includes that gait is healthy, gait is unhealthy and various gait illnesss；Described neural network model
There are parameter (W, b)=(W(1), b(1), W(2), b(2)), wherein W(i)For the matrix of parameter composition is linked between i-th layer to i+1 layer,
Matrixes of the b (i) for the bias term composition of i+1 layer；The activation value of ground floor is input, and last layer activation value is by linking parameter
The input value that next layer of weighted calculation, input value are calculated next layer of activation value by activation primitive sigmoid functions, pass
Pass and exported；
Training process parameters weighting attenuation parameter λ and learning rate α is first set in modeling process, by cost function
Partial derivative is asked to be iterated to parameter is linked between each layer.
The method that the body gait is estimated, comprises the following steps：
Wearable nine axis movement sensor is worn on waist and biped information collection node by step 1. respectively, enters rower
It is fixed, the three-dimensional motion acceleration of collection human motion, angular speed, angle, and carry out hardware signal processing；
The nine axle motor messages that described information acquisition node is gathered by radio receiving transmitting module by step 2. are by the nothing
Line communication module is sent to the data analysis module, and receives the data analysis module by the wireless communication module
The feedback come and control instruction.
Data analysis module described in step 3. calculates human body motion feature to be included：Step-length, step width, foot are high, sufficient angle, step
Frequently, leg speed, the ratio of lower limb standing phase time, body centre's acceleration, the limiting value of lower limb trunk angle, body weight in the short time
The displacement of heart vertical direction and a period of time interior exercise data dominant frequency and amplitude.
Step 4. integrative medicine using Tinetti evaluation and test tables standard is quantified；The human body motion feature is led to
Cross neural network model to be reflected in medical science judgment criteria, evaluate starting stage normality, mop floor with the presence or absence of single pin, biped step
State symmetry, paces continuity, trunk stability and muscle enabling capabilities.
Step 5. is calculated to the gait assessment criteria scores listed by Tinetti evaluation and test tables by neural network model,
Obtain fall risk coefficient.
The system uses a kind of Human Body Gait Analysis flow process, comprises the following steps：
Step 1. is separately fixed at evaluated object three information collection nodes with elastic band in user using front
Both feet and waist, it is ensured which is not fall off relative to the position fixed.
Step 2. turns on the power equipment, and waiting system starts, communication connection, and system prompt is upright, waits to be calibrated completing.
When new user uses, manually determined carries out a fc-specific test FC to step 3., user according to system prompt,
Start to walk from inactive state, start straight walking a period of time forward, after the completion of waiting system points out the test collection, system can root
Feature extraction is carried out according to the data of collection, calculates that leg is long, it is according to the average proportions of organization of human body, estimation ankle-joint height, big
Leg leg length, centre of body weight height.
Under step 4. risk assessment pattern, testee starts to walk from inactive state, starts walking, and data acquisition module is to original
Beginning data are acquired process, and are communicated by radio receiving transmitting module, the data analysis after certain hour of walking
Module sends evaluation and test termination instruction, and collection is finished.The gait data that described data analysis end-on is received is analyzed, instead
Feedback primary Calculation result.Training can be asked whether simultaneously, after the correct sample correspondence result for obtaining extraneous input, by institute
The alternative manner for stating neural network model carries out the training of the sample to model.
Under step 5. fall monitoring pattern, the exercise data of sensor Real-time Collection human body is calculated using fall detection in real time
Method carries out tumble state trend inspection；When the generation of tumble trend is detected, falling protective device (such as explosion type is controlled rapidly
Air bag), before actual injury of falling occurs, human body is protected by protector.
Particular embodiments described above, has been carried out to the purpose of the present invention, technical scheme and beneficial effect further in detail
Describe bright, the be should be understood that specific embodiment that the foregoing is only the present invention in detail, be not limited to the present invention, it is all
Within the spirit and principles in the present invention, any modification, equivalent substitution and improvements done etc., should be included in the guarantor of the present invention
Within the scope of shield.
Claims (8)
1. a kind of fall risk assessment and monitoring system based on multiple nine axis movement sensors, the system include information gathering mould
Block, wireless transport module and data analysis module, it is characterised in that
Described information acquisition module to measure and process both feet and waist nine axle motor messages, including acceleration, angle speed
Degree, angle, the nine axle motor messages after the wireless communication module is by process are sent to by the wireless communication module
The data analysis module；
The wireless communication module is to carry out the data exchange between the signal acquisition module and the data analysis module；
The data analysis module includes signal processing module, fall monitoring module and risk evaluation module；Wherein,
Described signal processing module is built to human body for being filtered denoising and frequency-domain analysis to the signal for gathering
Mould, extracts body gait feature；
The fall monitoring module is weighted process for the body gait feature for extracting signal processing module, obtains synthesis
Index, judges whether to fall down；
The risk evaluation module carries out neural metwork training for the body gait feature for extracting signal processing module, obtains
Comprehensive evaluation result, feeds back to tester or doctor.
2. system according to claim 1, it is characterised in that each information collection node includes nine axis movement sensors
And its peripheral circuit and radio receiving transmitting module；Wherein,
Nine axis movement sensor is used for acceleration, angular speed and the angle initial data for measuring evaluated position；
The peripheral circuit includes power-supplying circuit and hardware signal processing circuit, and the power-supplying circuit is for nine axles
Motion sensor is powered, and the hardware signal processing circuit is moved for nine axles obtained to nine axis movement sensor measurements
Signal carries out Dynamic Kalman Filtering；
The radio receiving transmitting module is for the nine axle motor messages that gather described information acquisition node by the radio communication
Module is sent to the data analysis module, and by the wireless communication module receive that the data analysis module sends it is anti-
Feedback and control instruction.
3. system according to claim 1, it is characterised in that signal processing module is filtered denoising to the signal for gathering
And coordinate system transformation.It is specific as follows：First the signal to gathering carries out Fast Fourier Transform (FFT), and is filtered on this basis
Ripple, filters off high-frequency noise interference；Then coordinate system change is carried out by attitude angle to acceleration, angular speed, obtains world coordinates
Under human motion absolute acceleration and angular speed.
4. system according to claim 1, it is characterised in that signal processing module carries out mathematical modeling to human body, specifically
It is as follows：The trunk of human body, thigh, shank are equivalent to into inverted pendulum respectively, by kinematics analysis, calculated with step-length leg length,
Speed is obtained as revolute with human synovial by angular speed calculation speed, with integrated acceleration, foot is obtained with rate integrating
The overall attitude of limbs is analyzed with the relative position of waist, with reference to relative position and leg length, so as to the motion shape to whole human body
State is modeled, and is obtained in that overall exercise data.
5. system according to claim 1, it is characterised in that the body gait feature includes：Step-length, foot are high, foot is pressed from both sides
Angle, cadence, leg speed, the ratio of lower limb standing phase time, body centre's acceleration, the limiting value of lower limb trunk angle, in the short time
The displacement of centre of body weight vertical direction and a period of time interior exercise data dominant frequency and amplitude.
6. system according to claim 1, it is characterised in that fall monitoring module by human body absolute acceleration, body with
Ground-angle, in the short time displacement of centre of body weight vertical direction be weighted, obtain overall target, judge whether to occur
Fall down, weight weight and be constantly modified by the gradient descent method of neural network model.
7. system according to claim 1, it is characterised in that risk evaluation module sets up neural network model, is divided into 3
Layer, one layer includes high step-length, step width, foot, sufficient angle, cadence, leg speed, the ratio of lower limb standing phase time, body centre's acceleration
And the limiting value of lower limb trunk angle；Two layers include starting stage normality, mop floor with the presence or absence of single pin, biped gait it is symmetrical
Property, paces continuity, trunk stability and muscle enabling capabilities；Three layers of output include that gait is healthy, gait is unhealthy and
Various gait illnesss；Described neural network model has parameter (W, b)=(W(1), b(1), W(2), b(2)), wherein W(i)For i-th layer
To the matrix that parameter composition is linked between i+1 layer, matrixes of the b (i) for the bias term composition of i+1 layer；The activation of ground floor
It is worth to be input into, last layer activation value calculates next layer of input value by linking parameter weighting, input value passes through activation primitive
Sigmoid functions are calculated next layer of activation value, and transmission is exported；
Training process parameters weighting attenuation parameter λ and learning rate α is first set in modeling process, by cost function
Partial derivative is asked to be iterated to parameter is linked between each layer.
8. a kind of usage right requires the method that the system on human body gait described in 1 is estimated, it is characterised in that including following
Step：
Wearable nine axis movement sensor is worn on waist and biped information collection node by step 1. respectively, is demarcated,
The three-dimensional motion acceleration of collection human motion, angular speed, angle, and carry out hardware signal processing；
Step 2. passes through the channel radio by the nine axle motor messages that described information acquisition node is gathered by radio receiving transmitting module
Letter module is sent to the data analysis module, and receives what the data analysis module was sent by the wireless communication module
Feedback and control instruction.
Data analysis module described in step 3. calculates human body motion feature to be included：Step-length, step width, foot are high, sufficient angle, cadence, step
Speed, the ratio of lower limb standing phase time, body centre's acceleration, the limiting value of lower limb trunk angle, in the short time centre of body weight erect
Nogata exercise data dominant frequency and amplitude to displacement and in a period of time.
Step 4. integrative medicine using Tinetti evaluation and test tables standard is quantified；By the human body motion feature by god
Jing network models are reflected in medical science judgment criteria, evaluate starting stage normality, mop floor with the presence or absence of single pin, biped gait pair
Title property, paces continuity, trunk stability and muscle enabling capabilities.
Step 5. is calculated to the gait assessment criteria scores listed by Tinetti evaluation and test tables by neural network model, is obtained
Fall risk coefficient.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611123591.7A CN106539587A (en) | 2016-12-08 | 2016-12-08 | A kind of fall risk assessment and monitoring system and appraisal procedure based on sensor of doing more physical exercises |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611123591.7A CN106539587A (en) | 2016-12-08 | 2016-12-08 | A kind of fall risk assessment and monitoring system and appraisal procedure based on sensor of doing more physical exercises |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106539587A true CN106539587A (en) | 2017-03-29 |
Family
ID=58396946
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611123591.7A Pending CN106539587A (en) | 2016-12-08 | 2016-12-08 | A kind of fall risk assessment and monitoring system and appraisal procedure based on sensor of doing more physical exercises |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106539587A (en) |
Cited By (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107019501A (en) * | 2017-05-05 | 2017-08-08 | 山东师范大学 | Detection method and system are fallen down based on genetic algorithm and the long-range of probabilistic neural network |
| CN107616798A (en) * | 2017-09-19 | 2018-01-23 | 北京工业大学 | A kind of gait asymmetry detection method based on acceleration of gravity |
| CN108309232A (en) * | 2017-12-21 | 2018-07-24 | 北京城市系统工程研究中心 | A kind of vest type attitude stability device for fast detecting and detection method |
| CN108338791A (en) * | 2018-02-09 | 2018-07-31 | 张立海 | The detection device and detection method of unstable motion data |
| WO2018223562A1 (en) * | 2017-06-09 | 2018-12-13 | Midea Group Co., Ltd. | System and method for care support at home |
| CN109793644A (en) * | 2017-11-17 | 2019-05-24 | 丰田自动车株式会社 | Gait evaluation device, gait training system and gait evaluation method |
| CN109805936A (en) * | 2019-01-18 | 2019-05-28 | 深圳大学 | Falling over of human body detection system based on ground vibration signal |
| CN109991896A (en) * | 2017-12-29 | 2019-07-09 | 深圳市优必选科技有限公司 | The method, apparatus and storage device of robot tumble anticipation |
| CN110151190A (en) * | 2019-05-23 | 2019-08-23 | 西南科技大学 | A kind of orthopaedics postoperative rehabilitation monitoring method and system |
| CN110245744A (en) * | 2019-05-29 | 2019-09-17 | 金华诺普视信息技术研究所有限公司 | It is a kind of that detection method is fallen down based on multilayer perceptron |
| CN110367996A (en) * | 2019-08-30 | 2019-10-25 | 方磊 | A kind of method and electronic equipment for assessing human body fall risk |
| CN110659677A (en) * | 2019-09-10 | 2020-01-07 | 电子科技大学 | Human body falling detection method based on movable sensor combination equipment |
| CN110765925A (en) * | 2019-10-18 | 2020-02-07 | 河南大学 | Carrier detection and gait recognition method based on improved twin neural network |
| CN111263940A (en) * | 2017-10-24 | 2020-06-09 | 亚萨合莱有限公司 | Wearable device with multiple biometrics |
| CN111330213A (en) * | 2020-03-24 | 2020-06-26 | 孝感峰创智能科技有限公司 | Anti-falling pre-judging method and system for VR (virtual reality) treadmill and readable storage medium |
| CN111631719A (en) * | 2020-05-21 | 2020-09-08 | 北京城市系统工程研究中心 | Method for predicting falling risk of old people |
| CN111931568A (en) * | 2020-06-30 | 2020-11-13 | 青岛科技大学 | Human body falling detection method and system based on enhanced learning |
| CN112137846A (en) * | 2019-06-28 | 2020-12-29 | 丰田自动车株式会社 | Learning system, walking training system, method, program, and learning completion model |
| CN112137837A (en) * | 2019-06-27 | 2020-12-29 | 丰田自动车株式会社 | Learning system, walking training system, method, program, and learning completion model |
| CN112137839A (en) * | 2019-06-27 | 2020-12-29 | 丰田自动车株式会社 | Learning system, walking training system, method, program, and learned model |
| CN112137838A (en) * | 2019-06-27 | 2020-12-29 | 丰田自动车株式会社 | Learning device, walking training system, method, program, and learning completion model |
| CN112386249A (en) * | 2019-08-14 | 2021-02-23 | 中移(苏州)软件技术有限公司 | Fall detection method and device, equipment and storage medium |
| CN112438724A (en) * | 2019-08-29 | 2021-03-05 | 松下电器（美国）知识产权公司 | Fall risk assessment method, fall risk assessment device, and recording medium |
| CN112494034A (en) * | 2020-11-30 | 2021-03-16 | 重庆优乃特医疗器械有限责任公司 | Data processing and analyzing system and method based on 3D posture detection and analysis |
| CN112784662A (en) * | 2018-12-30 | 2021-05-11 | 奥瞳系统科技有限公司 | Video-based fall risk evaluation system |
| CN112929332A (en) * | 2021-01-19 | 2021-06-08 | 北京睿芯高通量科技有限公司 | Multi-axis object motion detection system and detection method |
| CN115530774A (en) * | 2021-06-30 | 2022-12-30 | 荣耀终端有限公司 | Epilepsy detection method and device |
| CN117084671A (en) * | 2023-10-19 | 2023-11-21 | 首都医科大学宣武医院 | Motion evaluation system based on gyroscope signals |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103505219A (en) * | 2012-06-20 | 2014-01-15 | 中国科学院电子学研究所 | System and method for human body gait evaluation |
| JP2014208257A (en) * | 2014-06-11 | 2014-11-06 | 国立大学法人東北大学 | Gait analysis system |
| CN104729507A (en) * | 2015-04-13 | 2015-06-24 | 大连理工大学 | Gait recognition method based on inertial sensor |
| CN105512635A (en) * | 2015-12-15 | 2016-04-20 | 鲁东大学 | Category attribute fused deep network underground target identification method and system |
| CN105590408A (en) * | 2016-02-06 | 2016-05-18 | 高强 | Human body falling detection method and protection device |
| CN105877757A (en) * | 2016-03-30 | 2016-08-24 | 哈尔滨理工大学 | Multi-sensor integrated human motion posture capturing and recognizing device |
-
2016
- 2016-12-08 CN CN201611123591.7A patent/CN106539587A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103505219A (en) * | 2012-06-20 | 2014-01-15 | 中国科学院电子学研究所 | System and method for human body gait evaluation |
| JP2014208257A (en) * | 2014-06-11 | 2014-11-06 | 国立大学法人東北大学 | Gait analysis system |
| CN104729507A (en) * | 2015-04-13 | 2015-06-24 | 大连理工大学 | Gait recognition method based on inertial sensor |
| CN105512635A (en) * | 2015-12-15 | 2016-04-20 | 鲁东大学 | Category attribute fused deep network underground target identification method and system |
| CN105590408A (en) * | 2016-02-06 | 2016-05-18 | 高强 | Human body falling detection method and protection device |
| CN105877757A (en) * | 2016-03-30 | 2016-08-24 | 哈尔滨理工大学 | Multi-sensor integrated human motion posture capturing and recognizing device |
Cited By (45)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107019501A (en) * | 2017-05-05 | 2017-08-08 | 山东师范大学 | Detection method and system are fallen down based on genetic algorithm and the long-range of probabilistic neural network |
| US10438136B2 (en) | 2017-06-09 | 2019-10-08 | Midea Group Co., Ltd. | System and method for care support at home |
| WO2018223562A1 (en) * | 2017-06-09 | 2018-12-13 | Midea Group Co., Ltd. | System and method for care support at home |
| CN107616798A (en) * | 2017-09-19 | 2018-01-23 | 北京工业大学 | A kind of gait asymmetry detection method based on acceleration of gravity |
| CN107616798B (en) * | 2017-09-19 | 2020-06-19 | 北京工业大学 | Gait asymmetry detection method based on gravity acceleration |
| CN111263940A (en) * | 2017-10-24 | 2020-06-09 | 亚萨合莱有限公司 | Wearable device with multiple biometrics |
| CN111263940B (en) * | 2017-10-24 | 2023-04-14 | 亚萨合莱有限公司 | Wearable device with multiple biometrics |
| CN109793644B (en) * | 2017-11-17 | 2022-01-04 | 丰田自动车株式会社 | Gait evaluation device, gait training system, and gait evaluation method |
| CN109793644A (en) * | 2017-11-17 | 2019-05-24 | 丰田自动车株式会社 | Gait evaluation device, gait training system and gait evaluation method |
| CN108309232A (en) * | 2017-12-21 | 2018-07-24 | 北京城市系统工程研究中心 | A kind of vest type attitude stability device for fast detecting and detection method |
| CN109991896A (en) * | 2017-12-29 | 2019-07-09 | 深圳市优必选科技有限公司 | The method, apparatus and storage device of robot tumble anticipation |
| CN109991896B (en) * | 2017-12-29 | 2020-08-25 | 深圳市优必选科技有限公司 | Robot falling prediction method and device and storage device |
| CN108338791A (en) * | 2018-02-09 | 2018-07-31 | 张立海 | The detection device and detection method of unstable motion data |
| CN112784662A (en) * | 2018-12-30 | 2021-05-11 | 奥瞳系统科技有限公司 | Video-based fall risk evaluation system |
| CN109805936A (en) * | 2019-01-18 | 2019-05-28 | 深圳大学 | Falling over of human body detection system based on ground vibration signal |
| CN109805936B (en) * | 2019-01-18 | 2021-08-13 | 深圳大学 | Human body tumbling detection system based on ground vibration signal |
| CN110151190A (en) * | 2019-05-23 | 2019-08-23 | 西南科技大学 | A kind of orthopaedics postoperative rehabilitation monitoring method and system |
| CN110245744A (en) * | 2019-05-29 | 2019-09-17 | 金华诺普视信息技术研究所有限公司 | It is a kind of that detection method is fallen down based on multilayer perceptron |
| CN112137839A (en) * | 2019-06-27 | 2020-12-29 | 丰田自动车株式会社 | Learning system, walking training system, method, program, and learned model |
| US11839465B2 (en) | 2019-06-27 | 2023-12-12 | Toyota Jidosha Kabushiki Kaisha | Learning system, walking training system, method, program, and trained model |
| CN112137837A (en) * | 2019-06-27 | 2020-12-29 | 丰田自动车株式会社 | Learning system, walking training system, method, program, and learning completion model |
| US11944427B2 (en) | 2019-06-27 | 2024-04-02 | Toyota Jidosha Kabushiki Kaisha | Learning system, walking training system, method, program and trained model |
| CN112137838A (en) * | 2019-06-27 | 2020-12-29 | 丰田自动车株式会社 | Learning device, walking training system, method, program, and learning completion model |
| CN112137837B (en) * | 2019-06-27 | 2023-03-10 | 丰田自动车株式会社 | Learning system, walking training system, method, program, and learning completion model |
| US11937918B2 (en) | 2019-06-27 | 2024-03-26 | Toyota Jidosha Kabushiki Kaisha | Learning system, walking training system, method, program, and trained model |
| CN112137839B (en) * | 2019-06-27 | 2023-08-01 | 丰田自动车株式会社 | Learning system, walking training method, program, and learning-completed model |
| CN112137846A (en) * | 2019-06-28 | 2020-12-29 | 丰田自动车株式会社 | Learning system, walking training system, method, program, and learning completion model |
| CN112137846B (en) * | 2019-06-28 | 2023-07-04 | 丰田自动车株式会社 | Learning system, walking training system, method, program, and learning-completed model |
| CN112386249B (en) * | 2019-08-14 | 2023-09-19 | 中移(苏州)软件技术有限公司 | Fall detection method and device, equipment and storage medium |
| CN112386249A (en) * | 2019-08-14 | 2021-02-23 | 中移(苏州)软件技术有限公司 | Fall detection method and device, equipment and storage medium |
| CN112438724A (en) * | 2019-08-29 | 2021-03-05 | 松下电器（美国）知识产权公司 | Fall risk assessment method, fall risk assessment device, and recording medium |
| CN110367996A (en) * | 2019-08-30 | 2019-10-25 | 方磊 | A kind of method and electronic equipment for assessing human body fall risk |
| CN110659677A (en) * | 2019-09-10 | 2020-01-07 | 电子科技大学 | Human body falling detection method based on movable sensor combination equipment |
| CN110765925A (en) * | 2019-10-18 | 2020-02-07 | 河南大学 | Carrier detection and gait recognition method based on improved twin neural network |
| CN110765925B (en) * | 2019-10-18 | 2023-05-09 | 河南大学 | Method for detecting carrying object and identifying gait based on improved twin neural network |
| CN111330213A (en) * | 2020-03-24 | 2020-06-26 | 孝感峰创智能科技有限公司 | Anti-falling pre-judging method and system for VR (virtual reality) treadmill and readable storage medium |
| CN111631719B (en) * | 2020-05-21 | 2023-08-11 | 北京城市系统工程研究中心 | Method for predicting fall risk of old people |
| CN111631719A (en) * | 2020-05-21 | 2020-09-08 | 北京城市系统工程研究中心 | Method for predicting falling risk of old people |
| CN111931568A (en) * | 2020-06-30 | 2020-11-13 | 青岛科技大学 | Human body falling detection method and system based on enhanced learning |
| CN112494034A (en) * | 2020-11-30 | 2021-03-16 | 重庆优乃特医疗器械有限责任公司 | Data processing and analyzing system and method based on 3D posture detection and analysis |
| CN112929332A (en) * | 2021-01-19 | 2021-06-08 | 北京睿芯高通量科技有限公司 | Multi-axis object motion detection system and detection method |
| CN115530774A (en) * | 2021-06-30 | 2022-12-30 | 荣耀终端有限公司 | Epilepsy detection method and device |
| CN115530774B (en) * | 2021-06-30 | 2024-03-26 | 荣耀终端有限公司 | Epilepsy detection method and device |
| CN117084671A (en) * | 2023-10-19 | 2023-11-21 | 首都医科大学宣武医院 | Motion evaluation system based on gyroscope signals |
| CN117084671B (en) * | 2023-10-19 | 2024-04-02 | 首都医科大学宣武医院 | Motion evaluation system based on gyroscope signals |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106539587A (en) | A kind of fall risk assessment and monitoring system and appraisal procedure based on sensor of doing more physical exercises | |
| CN106166071B (en) | A kind of acquisition method and equipment of gait parameter | |
| CN110021398B (en) | Gait analysis and training method and system | |
| CN102824177B (en) | Three-dimensional human body gait quantitative analysis system and method | |
| WO2019232899A1 (en) | Comprehensive evaluation system and method for physical fitness and muscular strength | |
| JP5005055B2 (en) | Monitoring system and method for muscle strength and exercise / physical ability of limbs | |
| CN104159644A (en) | Apparatus and method for analysing a golf swing | |
| CN109480857B (en) | Device and method for detecting frozen gait of Parkinson disease patient | |
| KR101421122B1 (en) | Prediction Method of Muscle Fatigue and Prediction System of Muscle Fatigue | |
| KR20190014641A (en) | System and method for Gait analysis | |
| CN104656112B (en) | Based on surface electromyogram signal and the used personal localization method and devices combined of MEMS | |
| Chiu et al. | Pelvic kinematic method for determining vertical jump height | |
| CN102930133A (en) | Integrated rapid balance and gait assessment system and method | |
| CN106730766A (en) | A kind of evaluation method of the adult motor's load based on heart rate data | |
| CN107961523A (en) | Human body training system and intelligent body-building system based on heart rate detection | |
| CN109567812A (en) | Gait analysis system based on Intelligent insole | |
| JP2020151267A (en) | Body motion estimation system for person or the like | |
| Liu et al. | Deep learning based ground reaction force estimation for stair walking using kinematic data | |
| CN108325201A (en) | A kind of cadence detection method and equipment based on Multi-source Information Fusion | |
| CN111178338A (en) | Method for establishing database and standardized model in gait analysis system | |
| CN108079521A (en) | A kind of running test method and system | |
| CN110197727A (en) | Upper limb modeling method and motion function assessment system based on artificial neural network | |
| JP2017148287A (en) | Evaluation method of stumbling risk | |
| CN106913342A (en) | A kind of gait test system and method based on fiber grating and pressure sensor | |
| CN114010453A (en) | Feedback wearing system based on motion signal and preprocessing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170329 |
|
| RJ01 | Rejection of invention patent application after publication |