CN107854127B - Motion state detection method and device - Google Patents
Motion state detection method and device Download PDFInfo
- Publication number
- CN107854127B CN107854127B CN201711034365.6A CN201711034365A CN107854127B CN 107854127 B CN107854127 B CN 107854127B CN 201711034365 A CN201711034365 A CN 201711034365A CN 107854127 B CN107854127 B CN 107854127B
- Authority
- CN
- China
- Prior art keywords
- data
- state
- sub
- test channel
- motion
- 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.)
- Active
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
- A61B5/1455—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue using optical sensors, e.g. spectral photometrical oximeters
- A61B5/14551—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue using optical sensors, e.g. spectral photometrical oximeters for measuring blood gases
- A61B5/14553—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue using optical sensors, e.g. spectral photometrical oximeters for measuring blood gases specially adapted for cerebral tissue
-
- 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
-
- 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
- A61B5/1118—Determining activity level
-
- 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
- A61B5/6803—Head-worn items, e.g. helmets, masks, headphones or goggles
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/15—Biometric patterns based on physiological signals, e.g. heartbeat, blood flow
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Veterinary Medicine (AREA)
- Surgery (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Molecular Biology (AREA)
- Medical Informatics (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Physiology (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Dentistry (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Neurology (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Optics & Photonics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a method and a device for detecting a motion state, wherein the method comprises the following steps: detecting brain hemoglobin data of a tested object through N test channels in the process that the tested object executes a movement task; decomposing the brain hemoglobin data of each test channel into M sub-frequency bands aiming at the brain hemoglobin data of each test channel, wherein M is a positive integer; reconstructing the cerebral hemoglobin data of each sub-channel of each test channel to obtain a corresponding time sequence signal, and calculating the change rate of the time sequence signal to be used as data to be processed; quantizing the data, counting the quantization result of the brain hemoglobin data, and determining the corresponding feature vector of each sub-channel of each test channel based on the quantization result; and integrating the feature vectors with intersection in the spatial distribution under each sub-frequency band, and determining the feature vectors corresponding to the optimal recognition rate.
Description
Technical Field
The invention relates to a motion detection technology, in particular to a method and a device for detecting a motion state based on brain hemoglobin information.
Background
The aging problem of the population is a prominent problem in the current society, and the aging problem causes the body movement function of the old to be obviously reduced, particularly the walking ability of the lower limbs. Similarly, patients with Spinal Cord Injury (SCI) often have sensory and motor dysfunction in their limbs. In addition, the number of patients with motor dysfunction in the lower limbs due to traffic accidents, industrial injuries, accidental injuries and other diseases is increasing. All the patients need timely rehabilitation and walking-aid training. Moreover, research shows that active participation in consciousness has more obvious rehabilitation effect than passive training in the rehabilitation training process of patients. Therefore, the research of the walking aid equipment with adjustable initiative is developed, efficient rehabilitation training is provided for the people with motor dysfunction, the people with motor dysfunction are assisted to recover the ability of independent walking, and the walking aid has important social significance.
In recent years, the identification of the movement state based on myogenic electrical signals or movement information of the body has been greatly developed, but for some patients with critical dysfunction, myogenic information or movement information of the body acquired through myogenic signals and biomechanical signals is weak and signals are extremely difficult to acquire.
Disclosure of Invention
In order to solve the above technical problem, embodiments of the present invention provide a method and an apparatus for detecting a motion state, and a computer storage medium.
The method for detecting the motion state provided by the embodiment of the invention comprises the following steps:
in the process that a tested object executes a movement task, detecting brain hemoglobin data of the tested object through N testing channels, wherein N is a positive integer, and the brain hemoglobin data comprises: a difference of total oxyhemoglobin data, oxyhemoglobin data and deoxyhemoglobin data;
decomposing the brain hemoglobin data of each test channel into M sub-frequency bands aiming at the brain hemoglobin data of each test channel, wherein M is a positive integer;
reconstructing the cerebral hemoglobin data of each sub-channel of each test channel to obtain a corresponding time sequence signal, and calculating the change rate of the time sequence signal to be used as data to be processed;
quantizing the data, counting the quantization result of the brain hemoglobin data, and determining the corresponding feature vector of each sub-channel of each test channel based on the quantization result;
and integrating the feature vectors with intersection in the spatial distribution under each sub-frequency band, and determining the feature vectors corresponding to the optimal recognition rate.
In the embodiment of the present invention, the process of executing the movement task by the object to be tested includes:
the method comprises the following steps that a tested object sequentially moves in a first state, a second state, a third state and a fourth state according to a path with a preset length;
wherein the motion of the first state is: a first step length and a first step speed; the motion of the second state is: a first step length and a second step speed; the motion of the third state is: second step length and first step speed movement; the motion of the fourth state is: and (3) moving at a second step length and a third step speed, wherein the first step length is smaller than the second step length, the first step speed is smaller than the second step speed, and the second step speed is smaller than the third step speed.
In the embodiment of the invention, the tested object keeps a rest with a preset time before executing the movement of each state.
In an embodiment of the present invention, before decomposing the cerebral hemoglobin data into M sub-bands, the method further includes:
and filtering the brain hemoglobin data of each test channel, and normalizing the brain hemoglobin data after filtering.
In an embodiment of the present invention, the decomposing the cerebral hemoglobin data into M sub-bands includes:
decomposing the brain hemoglobin data by adopting a wavelet packet decomposition method to obtain the brain hemoglobin data corresponding to the following sub-frequency bands: 0 to 0.03Hz, 0.03 to 0.06Hz, 0.06 to 0.09Hz, 0.09 to 0.12Hz, 0.12 to 0.15Hz, 0.15 to 0.18 Hz.
In this embodiment of the present invention, the quantizing the data includes:
the data are classified into the following three types according to the numerical value: 1. 0, -1.
In this embodiment of the present invention, the determining the corresponding feature vector of each sub-channel of each test channel based on the quantization result includes:
counting the quantization result of each tested object;
when the probability that the quantization results of the target sub-bands of the target test channel are the same is greater than or equal to a preset threshold value, setting the feature vector corresponding to the target sub-band of the target test channel to be 1;
and when the probability that the quantization results of the target sub-bands of the target test channel are the same is smaller than a preset threshold value, setting the feature vector corresponding to the target sub-band of the target test channel to be 0.
In the embodiment of the present invention, the determining the feature vector corresponding to the optimal recognition rate includes:
and performing permutation and combination on all the feature vectors, calculating the recognition rate of each permutation and combination, and determining the feature vector corresponding to the optimal recognition rate.
The detection device of the motion state provided by the embodiment of the invention comprises:
the brain hemoglobin data testing unit is used for detecting the brain hemoglobin data of a tested object through N testing channels in the process that the tested object executes a movement task, wherein N is a positive integer, and the brain hemoglobin data comprises: a difference of total oxyhemoglobin data, oxyhemoglobin data and deoxyhemoglobin data;
the decomposition unit is used for decomposing the cerebral hemoglobin data of each test channel into M sub-frequency bands, wherein M is a positive integer;
the change rate calculation unit is used for reconstructing the cerebral hemoglobin data of each sub-channel of each test channel to obtain a corresponding time sequence signal, and calculating the change rate of the time sequence signal to be used as data to be processed;
the characteristic vector calculation unit is used for quantizing the data, counting the quantization result of the brain hemoglobin data and determining the corresponding characteristic vector of each sub-channel of each test channel based on the quantization result;
and the result determining unit is used for integrating the feature vectors with intersection in the spatial distribution under each sub-frequency band and determining the feature vectors corresponding to the optimal recognition rate.
In the embodiment of the present invention, the process of executing the movement task by the object to be tested includes:
the method comprises the following steps that a tested object sequentially moves in a first state, a second state, a third state and a fourth state according to a path with a preset length;
wherein the motion of the first state is: a first step length and a first step speed; the motion of the second state is: a first step length and a second step speed; the motion of the third state is: second step length and first step speed movement; the motion of the fourth state is: and (3) moving at a second step length and a third step speed, wherein the first step length is smaller than the second step length, the first step speed is smaller than the second step speed, and the second step speed is smaller than the third step speed.
In the embodiment of the invention, the tested object keeps a rest with a preset time before executing the movement of each state.
In the embodiment of the present invention, the apparatus further includes:
and the preprocessing unit is used for filtering the brain hemoglobin data of each test channel and normalizing the filtered brain hemoglobin data.
In an embodiment of the present invention, the decomposition unit is specifically configured to decompose the brain hemoglobin data by using a wavelet packet decomposition method, so as to obtain the brain hemoglobin data corresponding to the following sub-bands: 0 to 0.03Hz, 0.03 to 0.06Hz, 0.06 to 0.09Hz, 0.09 to 0.12Hz, 0.12 to 0.15Hz, 0.15 to 0.18 Hz.
In this embodiment of the present invention, the feature vector calculating unit is further configured to divide the data into the following three types according to the numerical values: 1. 0, -1.
In the embodiment of the invention, the feature vector calculation unit is further configured to count a quantization result of each tested object; when the probability that the quantization results of the target sub-bands of the target test channel are the same is greater than or equal to a preset threshold value, setting the feature vector corresponding to the target sub-band of the target test channel to be 1; and when the probability that the quantization results of the target sub-bands of the target test channel are the same is smaller than a preset threshold value, setting the feature vector corresponding to the target sub-band of the target test channel to be 0.
In the embodiment of the present invention, the result determining unit is further configured to perform permutation and combination on all the feature vectors, calculate the recognition rate of each permutation and combination, and determine the feature vector corresponding to the optimal recognition rate.
The computer storage medium provided by the embodiment of the invention stores computer executable instructions, and the computer executable instructions are executed by the processor to realize the motion state detection method.
According to the technical scheme of the embodiment of the invention, in the process that a tested object executes a movement task, the brain hemoglobin data of the tested object is detected through N test channels, wherein N is a positive integer, and the brain hemoglobin data comprises: a difference of total oxyhemoglobin data, oxyhemoglobin data and deoxyhemoglobin data; decomposing the brain hemoglobin data of each test channel into M sub-frequency bands aiming at the brain hemoglobin data of each test channel, wherein M is a positive integer; reconstructing the cerebral hemoglobin data of each sub-channel of each test channel to obtain a corresponding time sequence signal, and calculating the change rate of the time sequence signal to be used as data to be processed; quantizing the data, counting the quantization result of the brain hemoglobin data, and determining the corresponding feature vector of each sub-channel of each test channel based on the quantization result; and integrating the feature vectors with intersection in the spatial distribution under each sub-frequency band, and determining the feature vectors corresponding to the optimal recognition rate. By adopting the technical scheme of the embodiment of the invention, 1) the test experiment for acquiring the brain information by applying the fNIRS technology is simple and convenient, has low requirement on the external environment and does not have any negative effect on the tested object. The tested person walks in the natural environment in the whole test process, and the obtained motion state identification result is more favorable for providing a spontaneous motion control instruction for rehabilitation/walking aid equipment; the brain cortex biological information is obtained under the natural scene of cognitive activities, and the practical value of walking state identification is increased. 2) The walking state is identified based on the rate of change of the cerebral cortical hemoglobin concentration, and the accuracy of the identification can be improved using total oxyhemoglobin and the difference (relative change) of oxyhemoglobin and deoxyhemoglobin as study parameters. 3) By using a wavelet packet transformation method, the blood oxygen characteristics are researched in a frequency division mode, and the spatial distribution characteristics of the frequency division mode are integrated through a position division quantization and statistics method, so that the multidimensional blood oxygen characteristics are facilitated, the influence of individual difference is eliminated, and the identification accuracy is improved.
Drawings
FIG. 1 is a schematic flow chart of a method for detecting a motion state according to an embodiment of the present invention;
FIG. 2 is a diagram of the brain cortex movement-related region and test channel distribution according to an embodiment of the present invention;
FIG. 3 is a timing diagram of a motion experiment according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of statistical eigenvectors of total oxyhemoglobin according to an embodiment of the invention;
FIG. 5 is a schematic illustration of a statistical eigenvector of the difference between oxygenated hemoglobin and deoxygenated hemoglobin according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of a motion state detection device according to an embodiment of the present invention.
Detailed Description
So that the manner in which the features and aspects of the embodiments of the present invention can be understood in detail, a more particular description of the embodiments of the invention, briefly summarized above, may be had by reference to the embodiments, some of which are illustrated in the appended drawings.
The identification of the movement state based on the brain information has a distinct advantage over the identification of the movement state based on the myogenic electrical signals of the body or the movement information. Brain information acquisition techniques for the movement state mainly include Electroencephalogram (EEG) and Near Infrared spectral brain function imaging (fNIRS). Among them, the EEG technique has very high requirements for the test environment, and requires visual stimulation during the test or requires high-intensity training at the early stage of the test. And the fNIRS technology supports continuous testing in a natural environment and movement testing in a tested spontaneous movement state. Therefore, the embodiment of the invention mainly applies the fNIRS technology to test the brain information to research and identify the spontaneous walking consciousness, and lays a foundation for realizing an intelligent rehabilitation medical auxiliary means based on the brain-computer interface technology.
The technical scheme of the embodiment of the invention provides a method for detecting the motion state based on brain hemoglobin information, wherein the motion state at least comprises a walking step length and a walking speed.
During specific implementation, the fNIRS technology is applied to record the cerebral cortical hemoglobin information of a tested object (simply referred to as a tested object) in different motion states in real time, the tested object autonomously controls all motions in the test process, and the tracking measurement of the brain information is realized in a natural environment without external stimulation. The method mainly identifies small step length or middle step length and slow speed or medium speed (the middle speed or middle step length refers to the normal step speed and step length of the patient) aiming at patients without or with weak movement ability, and lays an important theoretical foundation for realizing the autonomous consciousness control of rehabilitation training equipment or intelligent walking aid equipment.
Fig. 1 is a schematic flow chart of a method for detecting a motion state according to an embodiment of the present invention, and as shown in fig. 1, the method for detecting a motion state includes the following steps:
step 101: in the process that a tested object executes a movement task, detecting brain hemoglobin data of the tested object through N testing channels, wherein N is a positive integer, and the brain hemoglobin data comprises: total oxyhemoglobin data, difference of oxyhemoglobin data and deoxyhemoglobin data.
In the embodiment of the present invention, the process of executing the movement task by the object to be tested includes:
the method comprises the following steps that a tested object sequentially moves in a first state, a second state, a third state and a fourth state according to a path with a preset length;
wherein the motion of the first state is: a first step length and a first step speed; the motion of the second state is: a first step length and a second step speed; the motion of the third state is: second step length and first step speed movement; the motion of the fourth state is: and (3) moving at a second step length and a third step speed, wherein the first step length is smaller than the second step length, the first step speed is smaller than the second step speed, and the second step speed is smaller than the third step speed.
Wherein the subject is kept at rest for a preset time period before performing the motion of each state.
For example: the test was conducted for two different walking speeds (e.g., slow and medium walking, or medium and high walking) in the case of small steps (about 9 steps) and medium steps (about 6 to 7 steps) within a fixed length of 4.4m (the maximum movable range of the experimental device data transmission line). Here, all step sizes and pace provide only one relative concept, the size of which is controlled by the test himself.
In the embodiment of the invention, the brain hemoglobin information recorded in the motion starting period is detected so as to: total oxyhemoglobin data, and the difference between the oxyhemoglobin data and the deoxyhemoglobin data, are the main analytical parameters.
Step 102: decomposing the cerebral hemoglobin data of each test channel into M sub-frequency bands aiming at the cerebral hemoglobin data of each test channel, wherein M is a positive integer.
In an embodiment of the present invention, before decomposing the cerebral hemoglobin data into M sub-bands, the method further includes:
and filtering the brain hemoglobin data of each test channel, and normalizing the brain hemoglobin data after filtering.
In an embodiment of the present invention, the decomposing the cerebral hemoglobin data into M sub-bands includes:
decomposing the brain hemoglobin data by adopting a wavelet packet decomposition method to obtain the brain hemoglobin data corresponding to the following sub-frequency bands: 0 to 0.03Hz, 0.03 to 0.06Hz, 0.06 to 0.09Hz, 0.09 to 0.12Hz, 0.12 to 0.15Hz, 0.15 to 0.18 Hz.
Step 103: and reconstructing the cerebral hemoglobin data of each sub-channel of each test channel to obtain a corresponding time sequence signal, and calculating the change rate of the time sequence signal to be used as data to be processed.
Step 104: and quantizing the data, counting the quantization result of the brain hemoglobin data, and determining the corresponding feature vector of each sub-channel of each test channel based on the quantization result.
In this embodiment of the present invention, the quantizing the data includes: the data are classified into the following three types according to the numerical value: 1. 0, -1.
In this embodiment of the present invention, the determining the corresponding feature vector of each sub-channel of each test channel based on the quantization result includes:
counting the quantization result of each tested object;
when the probability that the quantization results of the target sub-bands of the target test channel are the same is greater than or equal to a preset threshold value, setting the feature vector corresponding to the target sub-band of the target test channel to be 1;
and when the probability that the quantization results of the target sub-bands of the target test channel are the same is smaller than a preset threshold value, setting the feature vector corresponding to the target sub-band of the target test channel to be 0.
Step 105: and integrating the feature vectors with intersection in the spatial distribution under each sub-frequency band, and determining the feature vectors corresponding to the optimal recognition rate.
In the embodiment of the present invention, the determining the feature vector corresponding to the optimal recognition rate includes:
and (3) using a libsvm algorithm to perform permutation and combination on all the feature vectors, calculating the recognition rate of each permutation and combination, and determining the feature vector corresponding to the optimal recognition rate.
The technical solution of the embodiments of the present invention is further described in detail with reference to specific application examples.
1. Application scenarios: the slow and medium-speed walking tasks in small and medium-speed states are tried to be carried out within 4.4m of a fixed length. The specific step length and the step speed are controlled by the testee; in the whole experiment process, near-infrared brain imaging equipment FORIE-3000 is used for collecting the hemoglobin information of the tested cerebral cortex, and the sampling period is 0.13 second. The layout of the headset is shown in fig. 2, a test channel is formed between each pair of transmitting probes and receiving probes, and the main test range includes a Prefrontal cortex (PFC) region, an eye movement (FEC) region, an auxiliary motion (SMA) region, and a Premotor (PMC) region. Where the numbers between the probes represent the test channels and Cz is the center point of the entire brain.
The specific process of the experiment is as follows: before the task is started, the tested object is kept in a resting state for about 1 minute, then walking tasks are developed according to the established sequence of the experiment, after each walking task is finished, the object is rested in place for about 30 seconds and then returns to the original place, after the rest is continued for about 30 seconds, the next movement is started, and when all four gaits are finished, the walking is repeated. The starting and stopping time of the task is controlled by the testee, the whole experimental process is controlled by the testee, but the length of the rest time of the testee is informed before the experiment (and the rest time cannot be controlled by the number). Each time a trial begins and ends, the laboratory operator will MARK (MARK) the data for real-time testing. Experimental timing diagrams are shown in fig. 3, where SL represents small step slow speed, SM represents small step medium speed, ML represents medium step slow speed, and MM represents medium step high speed, and fig. 3 schematically shows timing diagrams of 4 different exercise states, where R represents rest and B represents backward.
2. Collecting cerebral cortex hemoglobin information recorded in a motion starting period, taking total oxyhemoglobin data and a difference value of the oxyhemoglobin data and deoxyhemoglobin data as two analysis parameters to extract characteristic vectors of the respective parameters, and specifically comprising the following steps:
firstly, in order to ensure the real-time property of analysis data and judge the motion consciousness of a tested object in time, a rest section before motion is intercepted as a research object so as to ensure that different motion states are judged according to the consciousness of the tested object before a task starts, and the main frequency range of each motion state is determined by using a power spectrum density method.
Secondly, filtering the data of each test channel in order to eliminate the null shift phenomenon of the signal, and keeping the data at the baseline position.
And thirdly, because individual differences exist among all the tested channels, the test channel obviously activated in 22 channels is conveniently found out, and the maximum and minimum normalization processing is carried out on the filtered data.
And fourthly, performing wavelet packet decomposition on the data processed in the previous step, reconstructing 6 frequency bands (0-0.03 Hz, 0.03-0.06 Hz, 0.06-0.09 Hz, 0.09-0.12 Hz, 0.12-0.15 Hz and 0.15-0.18 Hz) to obtain time sequence signals of the corresponding frequency bands, further calculating the change rate of each time sequence signal, intercepting 8 points before the Mark on a time domain as a research object (namely data in 1s before the start of the motion), and researching the characteristic difference of the time period before the start of each motion state.
Fifthly, dividing the blood oxygen parameters of 6 sub-frequency bands of 22 channels into three categories according to the magnitude of the values (the proportion of front and back 30%), and respectively representing the three categories by numbers 1, 0 and-1. Further, each tested data is counted, and when the quantization value of a certain frequency sub-band of a certain channel is the same and reaches 78% probability, the corresponding value is set to 1, otherwise, the corresponding value is set to 0, as shown in fig. 4 and 5.
Fig. 4 is a schematic diagram of statistical eigenvectors of total oxyhemoglobin according to an embodiment of the present invention, where 4 graphs in fig. 4 respectively represent eigenvectors of total oxyhemoglobin statistically obtained in different motion states, where a horizontal axis represents a corresponding test channel and a corresponding test region, and a vertical axis represents 6 sub-bands.
Fig. 5 is a schematic diagram of statistical feature vectors of differences between oxygenated hemoglobin and deoxygenated hemoglobin according to an embodiment of the present invention, where 4 graphs in fig. 5 respectively represent feature vectors of oxygenated and deoxygenated differences statistically obtained in different motion states, where a horizontal axis represents a corresponding test channel and a corresponding test area, and a vertical axis represents 6 sub-bands.
Sixthly, aiming at the channel information of each frequency band, combining the adjacent channel characteristics in space, and using a libsvm algorithm to arrange and combine all the characteristic vectors to calculate the optimal recognition rate.
3. And (3) recognition results:
in order to better help patients with different requirements to carry out rehabilitation training, the 4 states of the research are divided into 2 combinations according to the severity of the illness state of the patients. The combination is as follows: SL, SM and ML, for subjects who did not have walking ability in the early stage or had particularly weak walking ability; combining two: SL, SM, ML, MM, for patients who have some walking ability but cannot walk independently.
And aiming at each sub-frequency band, integrating total oxygen and a channel adjacent to the difference value between the level oxygen and the deoxidation in the space as a feature vector, and confirming the feature vector capable of obtaining the highest recognition rate by using a libsvm algorithm carried by Matlab software. Finally, the walking state of the other 4 tested bits is verified, and the identification rates of the first combination and the second combination under the step size and the pace speed state of the 4 tested bits are 83.33% and 68.75%.
Fig. 6 is a schematic structural diagram of a motion state detection device according to an embodiment of the present invention, and as shown in fig. 6, the device includes:
the brain hemoglobin data testing unit 601 is configured to detect brain hemoglobin data of a subject to be tested through N testing channels in a process that the subject to be tested performs a motor task, where N is a positive integer, where the brain hemoglobin data includes: a difference of total oxyhemoglobin data, oxyhemoglobin data and deoxyhemoglobin data;
the decomposition unit 602 is configured to decompose, for the cerebral hemoglobin data of each test channel, the cerebral hemoglobin data into M sub-bands, where M is a positive integer;
a change rate calculation unit 603, configured to reconstruct, for the brain hemoglobin data of each sub-channel of each test channel, the brain hemoglobin data to obtain a corresponding time sequence signal, and calculate a change rate of the time sequence signal as data to be processed;
a feature vector calculation unit 604, configured to quantize the data, count a quantization result of the brain hemoglobin data, and determine a corresponding feature vector of each sub-channel of each test channel based on the quantization result;
the result determining unit 605 is configured to integrate the feature vectors with intersection in the spatial distribution under each sub-band, and determine a feature vector corresponding to the optimal recognition rate.
In the embodiment of the present invention, the process of executing the movement task by the object to be tested includes:
the method comprises the following steps that a tested object sequentially moves in a first state, a second state, a third state and a fourth state according to a path with a preset length;
wherein the motion of the first state is: a first step length and a first step speed; the motion of the second state is: a first step length and a second step speed; the motion of the third state is: second step length and first step speed movement; the motion of the fourth state is: and (3) moving at a second step length and a third step speed, wherein the first step length is smaller than the second step length, the first step speed is smaller than the second step speed, and the second step speed is smaller than the third step speed.
In the embodiment of the invention, the tested object keeps a rest with a preset time before executing the movement of each state.
In the embodiment of the present invention, the apparatus further includes:
and the preprocessing unit 606 is configured to perform filtering processing on the brain hemoglobin data of each test channel, and perform normalization processing on the brain hemoglobin data after the filtering processing.
In an embodiment of the present invention, the decomposition unit 602 is specifically configured to decompose the cerebral hemoglobin data by using a wavelet packet decomposition method, so as to obtain the cerebral hemoglobin data corresponding to the following sub-bands: 0 to 0.03Hz, 0.03 to 0.06Hz, 0.06 to 0.09Hz, 0.09 to 0.12Hz, 0.12 to 0.15Hz, 0.15 to 0.18 Hz.
In this embodiment of the present invention, the feature vector calculating unit 604 is further configured to divide the data into the following three types according to the numerical values: 1. 0, -1.
In this embodiment of the present invention, the feature vector calculating unit 604 is further configured to count a quantization result of each tested object; when the probability that the quantization results of the target sub-bands of the target test channel are the same is greater than or equal to a preset threshold value, setting the feature vector corresponding to the target sub-band of the target test channel to be 1; and when the probability that the quantization results of the target sub-bands of the target test channel are the same is smaller than a preset threshold value, setting the feature vector corresponding to the target sub-band of the target test channel to be 0.
In this embodiment of the present invention, the result determining unit 605 is further configured to perform permutation and combination on all the feature vectors, calculate the recognition rate of each permutation and combination, and determine the feature vector corresponding to the optimal recognition rate.
It should be understood by those skilled in the art that the implementation functions of each unit in the motion state detection device shown in fig. 6 can be understood by referring to the related description of the motion state detection method. The functions of the units in the motion state detection apparatus shown in fig. 6 may be implemented by a program running on a processor, or may be implemented by specific logic circuits.
The device according to the embodiment of the present invention may also be stored in a computer-readable storage medium if it is implemented in the form of a software functional module and sold or used as an independent product. Based on such understanding, the technical solutions of the embodiments of the present invention may be essentially implemented or a part contributing to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read Only Memory (ROM), a magnetic disk, or an optical disk. Thus, embodiments of the invention are not limited to any specific combination of hardware and software.
Accordingly, the embodiment of the present invention further provides a computer storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, the method for detecting a motion state according to the embodiment of the present invention is implemented.
The technical schemes described in the embodiments of the present invention can be combined arbitrarily without conflict.
In the embodiments provided in the present invention, it should be understood that the disclosed method and intelligent device may be implemented in other ways. The above-described device embodiments are merely illustrative, for example, the division of the unit is only a logical functional division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the coupling, direct coupling or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or units may be electrical, mechanical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, all the functional units in the embodiments of the present invention may be integrated into one second processing unit, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention.
Claims (10)
1. A method for detecting a motion state, the method comprising:
in the process that a tested object executes a movement task, detecting brain hemoglobin data of the tested object through N testing channels, wherein N is a positive integer, and the brain hemoglobin data comprises: a difference of total oxyhemoglobin data, oxyhemoglobin data and deoxyhemoglobin data;
filtering the brain hemoglobin data of each test channel, and normalizing the brain hemoglobin data after filtering;
decomposing the brain hemoglobin data of each test channel into M sub-frequency bands aiming at the brain hemoglobin data of each test channel, wherein M is a positive integer;
reconstructing the cerebral hemoglobin data of each sub-channel of each test channel to obtain a corresponding time sequence signal, and calculating the change rate of the time sequence signal to be used as data to be processed;
quantizing the data to be processed, counting the quantization result of the brain hemoglobin data, and determining the corresponding characteristic value of each sub-channel of each test channel based on the quantization result;
and integrating the feature vectors with intersection in the spatial distribution under each sub-frequency band, and determining the feature vectors corresponding to the optimal recognition rate.
2. The method for detecting a motion state according to claim 1, wherein the process of executing the motion task by the object to be tested comprises:
the method comprises the following steps that a tested object sequentially moves in a first state, a second state, a third state and a fourth state according to a path with a preset length;
wherein the motion of the first state is: a first step length and a first step speed; the motion of the second state is: a first step length and a second step speed; the motion of the third state is: second step length and first step speed movement; the motion of the fourth state is: and (3) moving at a second step length and a third step speed, wherein the first step length is smaller than the second step length, the first step speed is smaller than the second step speed, and the second step speed is smaller than the third step speed.
3. The method for detecting a motion state according to claim 1, wherein the quantizing the data to be processed includes:
dividing the data to be processed into the following three types according to the numerical value: 1. 0, -1.
4. The method according to claim 1, wherein the determining the corresponding feature value of each sub-channel of each test channel based on the quantization result comprises:
counting the quantization result of each tested object;
when the probability that the quantization results of the target sub-bands of the target test channel are the same is greater than or equal to a preset threshold value, setting a characteristic value corresponding to the target sub-band of the target test channel to be 1;
and when the probability that the quantization results of the target sub-bands of the target test channel are the same is smaller than a preset threshold value, setting the characteristic value corresponding to the target sub-band of the target test channel to be 0.
5. The method for detecting motion state according to claim 1, wherein the determining the feature vector corresponding to the optimal recognition rate includes:
and performing permutation and combination on all the feature vectors, calculating the recognition rate of each permutation and combination, and determining the feature vector corresponding to the optimal recognition rate.
6. An apparatus for detecting a state of motion, the apparatus comprising:
the brain hemoglobin data testing unit is used for detecting the brain hemoglobin data of a tested object through N testing channels in the process that the tested object executes a movement task, wherein N is a positive integer, and the brain hemoglobin data comprises: a difference of total oxyhemoglobin data, oxyhemoglobin data and deoxyhemoglobin data;
the preprocessing unit is used for filtering the brain hemoglobin data of each test channel and normalizing the filtered brain hemoglobin data;
the decomposition unit is used for decomposing the cerebral hemoglobin data of each test channel into M sub-frequency bands, wherein M is a positive integer;
the change rate calculation unit is used for reconstructing the cerebral hemoglobin data of each sub-channel of each test channel to obtain a corresponding time sequence signal, and calculating the change rate of the time sequence signal to be used as data to be processed;
the characteristic vector calculation unit is used for quantizing the data to be processed, counting the quantization result of the brain hemoglobin data, and determining the corresponding characteristic vector of each sub-channel of each test channel based on the quantization result;
and the result determining unit is used for integrating the feature vectors with intersection in the spatial distribution under each sub-frequency band and determining the feature vectors corresponding to the optimal recognition rate.
7. The apparatus for detecting motion state according to claim 6, wherein the process of executing the motion task by the object under test includes:
the method comprises the following steps that a tested object sequentially moves in a first state, a second state, a third state and a fourth state according to a path with a preset length;
wherein the motion of the first state is: a first step length and a first step speed; the motion of the second state is: a first step length and a second step speed; the motion of the third state is: second step length and first step speed movement; the motion of the fourth state is: and (3) moving at a second step length and a third step speed, wherein the first step length is smaller than the second step length, the first step speed is smaller than the second step speed, and the second step speed is smaller than the third step speed.
8. The apparatus according to claim 6, wherein the feature vector calculating unit is further configured to classify the data to be processed into the following three categories according to the magnitude of the data: 1. 0, -1.
9. The apparatus according to claim 6, wherein the feature vector calculating unit is further configured to count a quantization result of each bit of the object under test; when the probability that the quantization results of the target sub-bands of the target test channel are the same is greater than or equal to a preset threshold value, setting a characteristic value corresponding to the target sub-band of the target test channel to be 1; and when the probability that the quantization results of the target sub-bands of the target test channel are the same is smaller than a preset threshold value, setting the characteristic value corresponding to the target sub-band of the target test channel to be 0.
10. The apparatus according to claim 6, wherein the result determining unit is further configured to perform permutation and combination on all the feature vectors, calculate the recognition rate of each permutation and combination, and determine the feature vector corresponding to the optimal recognition rate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711034365.6A CN107854127B (en) | 2017-10-30 | 2017-10-30 | Motion state detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711034365.6A CN107854127B (en) | 2017-10-30 | 2017-10-30 | Motion state detection method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107854127A CN107854127A (en) | 2018-03-30 |
| CN107854127B true CN107854127B (en) | 2021-04-09 |
Family
ID=61696478
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711034365.6A Active CN107854127B (en) | 2017-10-30 | 2017-10-30 | Motion state detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107854127B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109243569A (en) * | 2018-06-29 | 2019-01-18 | 苏州大学 | The recognition methods of spontaneous phantom object displacement orientation is tested based on brain information identification |
| CN109044365A (en) * | 2018-07-02 | 2018-12-21 | 苏州大学 | The recognition methods of two dimensional motion state based on brain hemoglobin information |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104207790A (en) * | 2014-09-22 | 2014-12-17 | 青岛蓝图文化传播有限公司市南分公司 | Bracelet for detecting blood oxygen and behavior state |
| US9526946B1 (en) * | 2008-08-29 | 2016-12-27 | Gary Zets | Enhanced system and method for vibrotactile guided therapy |
| CN106419850A (en) * | 2016-11-03 | 2017-02-22 | 国家康复辅具研究中心 | Dynamic brain function detection method and system based on near infrared spectrum and blood pressure information |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008157832A (en) * | 2006-12-26 | 2008-07-10 | National Institute Of Advanced Industrial & Technology | Optical measuring instrument for living body |
| US20080177163A1 (en) * | 2007-01-19 | 2008-07-24 | O2 Medtech, Inc. | Volumetric image formation from optical scans of biological tissue with multiple applications including deep brain oxygenation level monitoring |
| JP5167368B2 (en) * | 2008-10-29 | 2013-03-21 | トヨタ自動車株式会社 | MOBILE BODY CONTROL DEVICE AND MOBILE BODY CONTROL METHOD |
| ES2705526T3 (en) * | 2012-09-11 | 2019-03-25 | Life Corp Sa | Wearable communication platform |
| CN104027072B (en) * | 2014-06-04 | 2016-02-10 | 陕西科技大学 | A kind of scan-type vein blood vessel synchronous imaging and indicating device and method |
| CN106917510A (en) * | 2015-12-25 | 2017-07-04 | 世铠精密股份有限公司 | Screw-type Rebar bonding device and its forming method |
| CN106725520A (en) * | 2016-12-31 | 2017-05-31 | 中国科学院苏州生物医学工程技术研究所 | The signal processing method system of brain blood oxygen detection |
| CN106901749A (en) * | 2017-01-06 | 2017-06-30 | 苏州大学 | A kind of extracting method of cortex hemoglobin information representation locomitivity parameter |
-
2017
- 2017-10-30 CN CN201711034365.6A patent/CN107854127B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9526946B1 (en) * | 2008-08-29 | 2016-12-27 | Gary Zets | Enhanced system and method for vibrotactile guided therapy |
| CN104207790A (en) * | 2014-09-22 | 2014-12-17 | 青岛蓝图文化传播有限公司市南分公司 | Bracelet for detecting blood oxygen and behavior state |
| CN106419850A (en) * | 2016-11-03 | 2017-02-22 | 国家康复辅具研究中心 | Dynamic brain function detection method and system based on near infrared spectrum and blood pressure information |
Non-Patent Citations (1)
| Title |
|---|
| "To classify two-dimensional motion state of step length and walking speed by applying cerebral hemoglobin information";Hedian Jin等;《2017 10th International Conference on Human System Interactions》;20170717;第216-222页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107854127A (en) | 2018-03-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Gao et al. | Deep learninig of EEG signals for emotion recognition | |
| Niazi et al. | Detection of movement intention from single-trial movement-related cortical potentials | |
| Hosseini et al. | Emotional stress recognition system for affective computing based on bio-signals | |
| US20120226185A1 (en) | Readiness potential-based brain-computer interface device and method | |
| Costa et al. | Decoding the attentional demands of gait through EEG gamma band features | |
| Hosni et al. | Classification of EEG signals using different feature extraction techniques for mental-task BCI | |
| EP1934677A1 (en) | Method and system for detecting and classifying facial muscle movements | |
| Bozhkov et al. | EEG-based subject independent affective computing models | |
| CN109247917A (en) | A kind of spatial hearing induces P300 EEG signal identification method and device | |
| Pun et al. | Brain-computer interaction research at the Computer Vision and Multimedia Laboratory, University of Geneva | |
| Awan et al. | Effective classification of EEG signals using K-nearest neighbor algorithm | |
| CN107854127B (en) | Motion state detection method and device | |
| Shiman et al. | Towards decoding of functional movements from the same limb using EEG | |
| Ramanand et al. | Complexity quantification of dense array EEG using sample entropy analysis | |
| Hasan et al. | Preliminary study on real-time prediction of gait acceleration intention from volition-associated EEG patterns | |
| Yang et al. | Non-contact early warning of shaking palsy | |
| Abhishek et al. | Stress analysis of a computer game player using electroencephalogram | |
| Vettivel et al. | System for detecting student attention pertaining and alerting | |
| Planelles et al. | First steps in the development of an EEG-based system to detect intention of gait initiation | |
| Sulaiman et al. | Interpretation of human thought using EEG signals and LabVIEW | |
| Ming et al. | Multiscale entropy analysis of attention ralated EEG based on motor imaginary potential | |
| Plechawska-Wojcik et al. | Construction of neural nets in brain-computer interface for robot arm steering | |
| Kalafatovich et al. | Subject-independent object classification based on convolutional neural network from EEG signals | |
| CN109087693B (en) | Method for judging turning movement intention based on brain information | |
| Ming et al. | Electroencephalograph (EEG) signal processing method of motor imaginary potential for attention level classification |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220113 Address after: 215500 No.156 Xinyang Avenue, Xinzhuang Town, Changshu City, Suzhou City, Jiangsu Province Patentee after: Suzhou bairuixin Intelligent Technology Co.,Ltd. Address before: Room 1105, engineering building, North Campus of Jiangsu University, 178 Ganjiang East Road, Gusu District, Suzhou City, Jiangsu Province Patentee before: SOOCHOW University |