CN109350071A - A kind of flexible gait monitoring method calculated based on artificial intelligence - Google Patents
A kind of flexible gait monitoring method calculated based on artificial intelligence Download PDFInfo
- Publication number
- CN109350071A CN109350071A CN201811325766.1A CN201811325766A CN109350071A CN 109350071 A CN109350071 A CN 109350071A CN 201811325766 A CN201811325766 A CN 201811325766A CN 109350071 A CN109350071 A CN 109350071A
- Authority
- CN
- China
- Prior art keywords
- training
- human
- knee
- neural network
- artificial intelligence
- 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
- A61B5/112—Gait analysis
-
- 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
- A61B5/6802—Sensor mounted on worn items
-
- 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
- A61B5/6828—Leg
-
- 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
- A61B5/6832—Means for maintaining contact with the body using adhesives
- A61B5/6833—Adhesive patches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/30—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Public Health (AREA)
- Molecular Biology (AREA)
- Veterinary Medicine (AREA)
- Surgery (AREA)
- Heart & Thoracic Surgery (AREA)
- Pathology (AREA)
- Animal Behavior & Ethology (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Physical Education & Sports Medicine (AREA)
- Epidemiology (AREA)
- Mathematical Physics (AREA)
- Primary Health Care (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Physiology (AREA)
- Dentistry (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a kind of flexible gait monitoring methods calculated based on artificial intelligence, its main feature is that can be in the neural network run on FPGA using human knee when walking and vola multiple groups pressure data training one, then the neural network that the multiple groups pressure data acquired in real time inputs completion training is compared and is identified, realize the intellectual monitoring of human figure health, the training of the neural network is that the artificial intelligence learning system that the knee of sensor array acquisition and plantar pressure signal are accessed three-layer network framework after ADC module processing learns, the neural network input feature vector comparison module for completing training is compared and identifies with the pressure signal inputted in real time, realize the intellectual monitoring of human figure health.The present invention has the accuracy rate of prediction high compared with prior art, and structure is simple, easy to use, realizes effective intellectual monitoring of human figure health, and for human figure and health status carries out reliable analysis and research provides reliable, effective data and supports.
Description
Technical field
The present invention relates to artificial intelligence learning art field, especially a kind of energy artificial intelligence running on the FPGA is calculated
The flexible gait monitoring method learnt.
Background technique
Gait is the behavioural characteristic of mankind's walking, is related to muscle and the joint association of foot, ankle, knee, hip, trunk, neck, shoulder and arm
With movement, by being analyzed gait and being studied, human figure and human health status can be predicted.Good step
Row habit facilitates people and keeps physically and mentally healthy, and nonstandard walking for years can be aggravated to triceps and heel string
Pressure stimulation, bring harm to body.Along with the rapid development of wearable technology, gait monitoring device is realized small-sized
Change, wearableization, by formed in teen-age athletic posture, patient it is postoperative in terms of play a significant role, make to move
The coverage rate of sufficient state rectificative training greatly promotes.
Application No. is CN201610519761.7, patent name is the acquisition method and equipment of a kind of gait parameter, open
A kind of acquisition method of gait data, comprising: obtain the voice signal curve of left and right foot；It is extracted according to peak detection algorithm
Characterize that node contacts to earth the peak position of sound enough and not a node is contacted to earth the peak position of sound enough in the voice signal curve out, so
The step pitch Lsd=V of each step is calculated afterwardsSound( t2-t1)；Wherein, VSoundFor the speed that sound is propagated in air, t1And t2Point
Be not be fixed on different feet gait data acquisition device collect with primary single foot contact to earth issued sound at the time of,
In, t2It contacts to earth enough time of sound to collect not a node, t1It contacts to earth time of sound to collect node, is walked by acquisition enough
Road voice signal often walks to the day for human beings monitoring.
Gait can be subdivided into eight stages, respectively initially land the phase, support the reaction phase, and midpoint supports the phase, support the later period,
Early period is swung, early stage is swung, swing mid-term and swings the later period.In different gait phases, human knee is different with foot form, just
Begin the phase knee in the wrong 0 that landso, heel contacts to earth；Reaction phase knee is supported to bend 0o-20o, sole lands, and center of gravity gradually moves to foot center, vola
In parallel；Midpoint supports phase knee to bend 20o-0o, foot central supported, heel is gradually liftoff；Later period knee is supported to bend 0o, heel is liftoff, sole
It lands；It swings knee early period and bends 0o-40o, foot is gradually liftoff；It swings early stage knee and bends degree by 40oBecome 60o, foot is liftoff；Pendulum
Dynamic mid-term knee bends degree by 60 oBecome 30o, foot is liftoff；It swings latter stage knee and bends 30 o-0o, foot is liftoff, this eight gait ranks
Human knee and foot have different forms in section, and the pressure of generation is also different.Initially the phase knee that lands does not stress, heel by
Power, sole do not stress；Support reaction phase knee by no pressure to by smaller pressure, heel stress, sole stress；Midpoint supports the phase
Knee by by smaller pressure to not stressing, heel stress, sole stress；Support later period knee does not stress, and heel does not stress, sole
Stress；Knee early period is swung by not stressing to by smaller pressure, heel does not stress, and sole does not stress；Swing early stage knee by by
Smaller pressure is to by larger pressure, and heel does not stress, and sole does not stress；Swing mid-term knee by by larger pressure to by smaller pressure
Power, heel do not stress, and sole does not stress；It swings later period knee not stressing, heel does not stress, and sole does not stress.
The prior art pays close attention to influence of the foot to gait mostly, does not consider influence of the knee to gait, so as to gait
Analysis and research poor reliability, largely effect on the prediction to human figure and human health status.
Summary of the invention
The purpose of the present invention is a kind of flexible steps calculated based on artificial intelligence for designing in view of the deficiencies of the prior art
State monitoring method, human knee and plantar pressure signal, multiple groups pressure data when acquiring walking using pliable pressure sensor array
Artificial intelligence learning system is transported to after circuit module is handled, and the data of acquisition are carried out with the study and training of neural network, benefit
Human locomotion each stage can be identified with training successful neural network, realize effective intelligence prison of human figure health
It surveys, for human figure and health status carries out reliable analysis and research provides reliable, effective data and supports.
The object of the present invention is achieved like this: a kind of flexible gait monitoring method calculated based on artificial intelligence, special
Using human knee when walking and vola multiple groups pressure data one energy neural network running on the FPGA of training, then point is
The neural network that the multiple groups pressure data acquired in real time inputs completion training is compared and is identified, realizes human figure health
Intellectual monitoring, the training of the neural network carries out in the steps below:
Step 1: pliable pressure sensor array is acquired walking when human knee and vola multiple groups pressure signal input ADC turn
Change the mold the conversion that block carries out modulus signal.
Step 2: inputting above-mentioned multiple groups digital pressure signal by input layer, hidden layer neuron and output layer
The artificial intelligence learning system of neuron three-layer network framework is learnt, and is transported to its learning outcome by output layer neuron
Training parameter comparison unit.
Step 3: meeting the training parameter of setting after the above-mentioned trained parameter comparison unit of learning outcome compares, then training
Terminate；If above-mentioned learning outcome is not able to satisfy the training parameter of setting, input layer nerve is returned after being corrected the adjustment of weight
Member, and hidden layer neuron and output layer neuron are again introduced by input layer, so carry out repeatedly circulation study
Training, until learning outcome meets the training parameter of setting.
Step 4: neural network and real-time digital pressure signal input feature vector comparison module that will complete training, are counted
It calculates and identifies, realize the intellectual monitoring of human figure health.
The pliable pressure sensor array is 64 × 64 sensor arrays, and is attached at knee and vola, and acquisition initially lands
Phase, the midpoint support phase, in the support later period, swings early period, swings early stage, swinging mid-term and swing the later period eight steps the support reaction phase
The pressure data in state stage.
The ADC conversion module is concatenated by gating circuit with analog to digital conversion circuit, and the gating circuit uses
CD4067 chip.
The present invention has the accuracy rate of prediction high compared with prior art, and structure is simple, easy to use, realizes human figure
Effective intellectual monitoring of health, for human figure and health status carries out reliable analysis and research provides reliable, effective data
It supports.
Detailed description of the invention
Fig. 1 is flow chart of the present invention；
Fig. 2 is gait phase and knee and foot loading process figure；
Fig. 3 is implementation example figure.
Specific embodiment
Refering to attached drawing 1, human knee and vola multiple groups pressure data one neural network of training when the present invention utilizes walking,
Then the neural network that the multiple groups pressure data acquired in real time inputs completion training is compared and is identified, realize human figure
The intellectual monitoring of health, the training of the neural network carry out in the steps below:
Step 1: pliable pressure sensor array 1 is acquired walking when human knee and vola multiple groups pressure signal input ADC turn
The conversion that block 2 carries out modulus signal is changed the mold, the ADC conversion module 2 is by the gating circuit 21 and analog to digital conversion circuit 22 that concatenate
Composition；The gating circuit uses CD4067 chip.
Step 2: by the input of above-mentioned multiple groups digital pressure signal by input layer 31, hidden layer neuron 32 and defeated
The artificial intelligence learning system 3 of 33 three-layer network framework of layer neuron is learnt out, and is learned by output layer neuron 33
Practise result output.
Step 3: above-mentioned learning outcome is inputted into training parameter comparison unit 34, if meeting the training parameter requirement of setting,
Then training terminates；If above-mentioned learning outcome does not meet the training parameter requirement of setting, each layer mind is corrected by amendment weight 35
Input layer 31 is again returned to after the weight of member, and hidden layer neuron is again inputted by input layer 31
32, and learnt by hidden layer neuron 32 into output layer neuron 33, until learning outcome meets training parameter comparison
The training parameter set in unit 34.
Step 4: the neural network for completing training and real-time digital pressure signal input feature vector comparison module 4 are counted
It calculates and identifies, realize the intellectual monitoring of human figure health.
Refering to attached drawing 2, when people's walking, gait can be segmented are as follows: initially land the phase (heel stress, sole and knee not by
Power) → support reaction phase (heel, knee and ball of foot stress) → midpoint support phase (heel, ball of foot and knee stress)
→ support later period (knee and heel do not stress, ball of foot stress) → swings early period (knee stress, heel and sole do not stress)
→ swinging early stage (knee stress, foot and sole do not stress) → swing mid-term (knee stress, foot and sole do not stress) →
Swing eight gait phases of later period (knee stress, foot and sole do not stress).Eight kinds of gaits are pressed caused by foot and knee
Power distribution is different, and knee is stretched along with bending, and vola impetus changes between heel and sole, and the bending of knee leads to knee
The pressure value that the pliable pressure sensor 1 at place is experienced changes.When heel is put forth effort in vola, pliable pressure sensor array 1
Heel member stress it is significant, when sole is put forth effort in vola, the sole part stress of pliable pressure sensor array 1 is significant.It attaches
The signal that pliable pressure sensor array 1 in knee and vola acquires converts analog signals into digital letter by ADC conversion module 2
It is delivered to artificial intelligence learning system 3 after number, neural network is trained based on multiple groups pressure data, training is successfully neural
Network can identify the human locomotion stage.
By following specific embodiments, the present invention is described in further detail.
Embodiment 1
Refering to attached drawing 3, the pliable pressure sensor array 1 be 64 × 64 sensor arrays, be attached at subject 6 knee and
Vola, the acquisition initial phase that lands, the midpoint support phase, in the support later period, swing early period, swing early stage, swing mid-term the support reaction phase
With the pressure data for swinging eight gait phases of later period.The sensor array line number that pliable pressure sensor array 1 uses is 64 rows, column
Number is 64 column, and multiple sensing units work at the same time, and acquires the pressure value of foot and knee.Pliable pressure sensor array 1 will acquire
To pressure signal transport to ADC conversion module 2, the conversion of gated circuit 21 and analog to digital conversion circuit 22, by subject 6 eight
Input artificial intelligence learning system 3 is learnt after the analog signal of walking phase data is converted to digital signal.Each walking
Stage acquires 3600 groups of data respectively, and the training parameter being arranged in training parameter unit 34 is respectively as follows: trained minimal error:
0.001；Allow the maximum step number of training: 5000 steps；Learning rate: 0.05；A training result is shown at interval of 10 steps.Training knot
Fruit meets the parameter request being arranged in training parameter unit 34, then training terminates；If above-mentioned learning outcome does not meet the instruction of setting
Practice parameter request, then again returns to input layer 31 after amendment weight 35 corrects the weight of each layer neuron, and by
Input layer 31 is again inputted into hidden layer neuron 32, and by hidden layer neuron 32 enter output layer neuron 33 into
Row study, until learning outcome meets the training parameter set in training parameter comparison unit 34.It will complete the nerve net of training
Network and real-time digital pressure signal input feature vector comparison module 4.When 6 walking again of subject, pliable pressure sensor array 1 will
The knee and foot pressure data acquired in real time is inputted by feature comparison module 4 and carries out human-step in trained neural network
The intellectual monitoring of human figure health is realized in the identification of row order section.
Above only the present invention is further illustrated, and not to limit this patent, all is equivalence enforcement of the present invention,
It is intended to be limited solely by within the scope of the claims of this patent.
Claims (3)
1. a kind of flexible gait monitoring method calculated based on artificial intelligence, it is characterised in that human knee and foot when utilizing walking
Bottom multiple groups pressure data trains an energy neural network running on the FPGA, then that the multiple groups pressure data acquired in real time is defeated
The neural network for entering to complete training is compared and identifies, realizes the intellectual monitoring of human figure health, the neural network
Training carries out in the steps below:
Step 1: pliable pressure sensor array is acquired walking when human knee and vola multiple groups pressure signal input ADC turn
Change the mold the conversion that block carries out modulus signal；
Step 2: by the input of above-mentioned multiple groups digital pressure signal by input layer, hidden layer neuron and output layer nerve
The artificial intelligence learning system of first three-layer network framework is learnt, and its learning outcome is transported to training by output layer neuron
Parameter comparison unit；
Step 3: meeting the training parameter of setting, then training knot after the above-mentioned trained parameter comparison unit of learning outcome compares
Beam；If above-mentioned learning outcome is not able to satisfy the training parameter of setting, input layer is returned after being corrected the adjustment of weight,
And hidden layer neuron and output layer neuron are again introduced by input layer, so carry out the instruction of repeatedly circulation study
Practice, until learning outcome meets the training parameter of setting；
Step 4: will complete training neural network and real-time digital pressure signal input feature vector comparison module, carry out calculate and
The intellectual monitoring of human figure health is realized in identification.
2. the flexible gait monitoring method calculated according to claim 1 based on artificial intelligence, it is characterised in that the flexibility
Pressure sensing array is 64 × 64 sensor arrays, and is attached at knee and vola, acquisition initially land the phase, the support reaction phase, in
Point support phase, support later period swing early period, swing early stage, swing mid-term and swing the pressure data of eight gait phases of later period.
3. the flexible gait monitoring method calculated according to claim 1 based on artificial intelligence, it is characterised in that the ADC turns
Mold changing block is made of the gating circuit and analog to digital conversion circuit concatenated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811325766.1A CN109350071A (en) | 2018-11-08 | 2018-11-08 | A kind of flexible gait monitoring method calculated based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811325766.1A CN109350071A (en) | 2018-11-08 | 2018-11-08 | A kind of flexible gait monitoring method calculated based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109350071A true CN109350071A (en) | 2019-02-19 |
Family
ID=65344451
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811325766.1A Pending CN109350071A (en) | 2018-11-08 | 2018-11-08 | A kind of flexible gait monitoring method calculated based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109350071A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110638455A (en) * | 2019-09-26 | 2020-01-03 | 京东方科技集团股份有限公司 | Server, system, device and medium for evaluating user rehabilitation status |
| CN110893100A (en) * | 2019-12-16 | 2020-03-20 | 广东轻工职业技术学院 | Device and method for monitoring posture change based on plantar pressure sensor |
| CN112137834A (en) * | 2019-06-27 | 2020-12-29 | 丰田自动车株式会社 | Learning system, rehabilitation support system, method, program, and learning completion model |
| CN113520375A (en) * | 2021-07-21 | 2021-10-22 | 深圳大学 | Gait phase dividing method, device, storage medium and system |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101807245A (en) * | 2010-03-02 | 2010-08-18 | 天津大学 | Artificial neural network-based multi-source gait feature extraction and identification method |
| CN105631195A (en) * | 2015-12-18 | 2016-06-01 | 合肥工业大学 | Wearable multi-information fusion gait analysis system and method thereof |
| CN106821680A (en) * | 2017-02-27 | 2017-06-13 | 浙江工业大学 | A kind of upper limb healing ectoskeleton control method based on lower limb gait |
| CN106853284A (en) * | 2016-12-30 | 2017-06-16 | 苏州能斯达电子科技有限公司 | A kind of Intelligent worn device for monitoring foot motion |
| CN107480651A (en) * | 2017-08-25 | 2017-12-15 | 清华大学深圳研究生院 | Abnormal gait detection method and abnormal gait detecting system |
| CN107536613A (en) * | 2016-06-29 | 2018-01-05 | 深圳光启合众科技有限公司 | Robot and its human body lower limbs Gait Recognition apparatus and method |
| CN107622260A (en) * | 2017-10-26 | 2018-01-23 | 杭州电子科技大学 | Lower limb gait phase identification method based on multi-source bio signal |
| CA3036279A1 (en) * | 2016-09-08 | 2018-03-15 | Trexo Robotics Inc. | Mobile weight-bearing powered orthosis device |
| CN108091380A (en) * | 2017-11-30 | 2018-05-29 | 中科院合肥技术创新工程院 | Teenager's basic exercise ability training system and method based on multi-sensor fusion |
-
2018
- 2018-11-08 CN CN201811325766.1A patent/CN109350071A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101807245A (en) * | 2010-03-02 | 2010-08-18 | 天津大学 | Artificial neural network-based multi-source gait feature extraction and identification method |
| CN105631195A (en) * | 2015-12-18 | 2016-06-01 | 合肥工业大学 | Wearable multi-information fusion gait analysis system and method thereof |
| CN107536613A (en) * | 2016-06-29 | 2018-01-05 | 深圳光启合众科技有限公司 | Robot and its human body lower limbs Gait Recognition apparatus and method |
| CA3036279A1 (en) * | 2016-09-08 | 2018-03-15 | Trexo Robotics Inc. | Mobile weight-bearing powered orthosis device |
| CN106853284A (en) * | 2016-12-30 | 2017-06-16 | 苏州能斯达电子科技有限公司 | A kind of Intelligent worn device for monitoring foot motion |
| CN106821680A (en) * | 2017-02-27 | 2017-06-13 | 浙江工业大学 | A kind of upper limb healing ectoskeleton control method based on lower limb gait |
| CN107480651A (en) * | 2017-08-25 | 2017-12-15 | 清华大学深圳研究生院 | Abnormal gait detection method and abnormal gait detecting system |
| CN107622260A (en) * | 2017-10-26 | 2018-01-23 | 杭州电子科技大学 | Lower limb gait phase identification method based on multi-source bio signal |
| CN108091380A (en) * | 2017-11-30 | 2018-05-29 | 中科院合肥技术创新工程院 | Teenager's basic exercise ability training system and method based on multi-sensor fusion |
Non-Patent Citations (3)
| Title |
|---|
| 樊瑜波，张明: "《生物力学研究前沿系列 康复工程生物力学》", 31 December 2017 * |
| 袁娜: "基于概率神经网络的步态识别", 《中国优秀硕士学位论文全文数据库》 * |
| 颉梦宁，李风雷: "《运动人体科学实验指导》", 31 December 2015 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112137834A (en) * | 2019-06-27 | 2020-12-29 | 丰田自动车株式会社 | Learning system, rehabilitation support system, method, program, and learning completion model |
| CN110638455A (en) * | 2019-09-26 | 2020-01-03 | 京东方科技集团股份有限公司 | Server, system, device and medium for evaluating user rehabilitation status |
| CN110638455B (en) * | 2019-09-26 | 2022-06-14 | 京东方科技集团股份有限公司 | Server, system, device and medium for evaluating user rehabilitation status |
| CN110893100A (en) * | 2019-12-16 | 2020-03-20 | 广东轻工职业技术学院 | Device and method for monitoring posture change based on plantar pressure sensor |
| CN113520375A (en) * | 2021-07-21 | 2021-10-22 | 深圳大学 | Gait phase dividing method, device, storage medium and system |
| CN113520375B (en) * | 2021-07-21 | 2023-08-01 | 深圳大学 | Gait phase dividing method, device, storage medium and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109350071A (en) | A kind of flexible gait monitoring method calculated based on artificial intelligence | |
| CN108683724A (en) | A kind of intelligence children's safety and gait health monitoring system | |
| CN110236550B (en) | Human gait prediction device based on multi-mode deep learning | |
| Wang et al. | Human Gait Recognition System Based on Support Vector Machine Algorithm and Using Wearable Sensors. | |
| CN108363959B (en) | Table tennis or badminton sport motion identification method | |
| CN112766142A (en) | Plantar pressure image processing method, plantar pressure image identification method and gait analysis system | |
| Hossain et al. | Deepbbwae-net: A cnn-rnn based deep superlearner for estimating lower extremity sagittal plane joint kinematics using shoe-mounted imu sensors in daily living | |
| CN104656112B (en) | Based on surface electromyogram signal and the used personal localization method and devices combined of MEMS | |
| CN111950383B (en) | Joint angle-based rhythmic movement collaborative analysis method | |
| Zhang et al. | Real-time gait phase recognition based on time domain features of multi-MEMS inertial sensors | |
| Kang et al. | Subject-independent continuous locomotion mode classification for robotic hip exoskeleton applications | |
| CN109567812A (en) | Gait analysis system based on Intelligent insole | |
| CN109431510A (en) | A kind of flexible gait monitoring device calculated based on artificial intelligence | |
| Hollinger et al. | The influence of gait phase on predicting lower-limb joint angles | |
| CN210077662U (en) | Flexible gait monitoring device | |
| Jiang et al. | Deep learning algorithm based wearable device for basketball stance recognition in basketball | |
| Cao et al. | Research on Human Sports Rehabilitation Design Based on Object‐Oriented Technology | |
| Na et al. | Deep domain adaptation, pseudo-labeling, and shallow network for accurate and fast gait prediction of unlabeled datasets | |
| Yen et al. | Feature Fusion-Based Deep Learning Network to Recognize Table Tennis Actions. | |
| Lueken et al. | Using synthesized imu data to train a long-short term memory-based neural network for unobtrusive gait analysis with a sparse sensor setup | |
| Li et al. | 3d knee and hip angle estimation with reduced wearable imus via transfer learning during yoga, golf, swimming, badminton, and dance | |
| Cornish et al. | Hip contact forces can be predicted with a neural network using only synthesised key points and electromyography in people with hip osteoarthritis | |
| Sivakumar et al. | Estimation of joint angle from ground reaction force in human gait | |
| Liang et al. | Deep-learning model for the prediction of lower-limb joint moments using single inertial measurement unit during different locomotive activities | |
| CN114533043A (en) | Method for automatically evaluating limb dysfunction of hemiplegic patient |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190219 |
|
| WD01 | Invention patent application deemed withdrawn after publication |