CN107811635A - A kind of health status sorting technique and device based on physiology signal - Google Patents
A kind of health status sorting technique and device based on physiology signal Download PDFInfo
- Publication number
- CN107811635A CN107811635A CN201610817417.6A CN201610817417A CN107811635A CN 107811635 A CN107811635 A CN 107811635A CN 201610817417 A CN201610817417 A CN 201610817417A CN 107811635 A CN107811635 A CN 107811635A
- Authority
- CN
- China
- Prior art keywords
- frequency wavelet
- wavelet coefficient
- low
- frequency
- signal
- 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.)
- Granted
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/389—Electromyography [EMG]
-
- 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
-
- 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
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Veterinary Medicine (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The present invention relates to human physiological signal treatment technical field, more particularly to a kind of health status sorting technique and device based on physiology signal.The health status sorting technique based on physiology signal includes：Step a：Wavelet decomposition processing is carried out to the human body surface myoelectric signal of collection, obtains every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient；Step b：Complexity processing is carried out to every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient, obtains the complexity value of every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient；Step c：The sample entropy and complexity value of every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient are respectively compared, determines wavelet coefficient otherness parameter, is classified according to the wavelet coefficient otherness parameters on human health status.The present invention is simple to operate, and cost is cheap, objective effectively the health status of subject can be classified.
Description
Technical field
The present invention relates to human physiological signal treatment technical field, more particularly to a kind of health based on physiology signal
State classification method and device.
Background technology
Back pain be one group using lower back, waist sacrum pain of buttock as the syndrome of cardinal symptom, be medical science of recovery therapy, clinical medicine
With the common disease of medicine in field of sports medicine, investigation shows, back pain is to be only second to the infection of the upper respiratory tract and occupy deputy
Common disease, its lifelong prevalence rate may be up to 60%-90%, year prevalence rate be 15%-45%, be mainly between 35-50 years old.
In recent years, as scientific and technological and industrialized quickening, the continuous improvement of people's life stress, back pain illness illness age have substantially
Downward trend.Not exclusively outside middle age and old group back pain illness rate height, youth group back pain illness rate is also into increasing
Long trend, mainly based on muscle-derived back pain.And easily ill occupation is mainly driver, the crowd that bends over one's desk working, operation
Doctor, nurse and professional athlete etc., 40% people is there are about after ill can reduce recreation, and 20% people's daily life is obvious
It is limited, 5% people's ADL critical constraints.Due to the very long of back pain treatment cycle and repeatedly, cause medical expense
Also compare higher, the U.S. early in phase early 1990s is used to treating every year the expense of back pain, and to be just up to 150-500 hundred million beautiful
Member.Back pain, which has become, to be caused dysfunction, disables and delay work, increase social economical burden and influence the weight of human life quality
Want reason.
However, because people lack enough understanding and attention to back pain illness, many back pain pain patients are rising
When just there are light symptoms, often ignore treatment, do not go to hospital to see a doctor or go rehabilitation institution to carry out correct rehabilitation
There is lumbar vertebrae keyboard protrusion disease to block in training, and selects the mode such as rest and some massages to mitigate illness, if things go on like this
Under incorrect work posture and the intensity of pressure repeatability of this illness and the order of severity are constantly aggravated, ultimately resulted in
Lumbar vertebrae keyboard protrudes, and severe patient will cause dysfunction and disable.
Now clinically, the diagnosis to protrusion of lumber intervertebral disc is mainly carried out by MRI magnetic resonance imagings or CT scan
Judge, this diagnostic mode is often the diagnosis selection that pain patients selection is carried out in the case of in great pain, common to examine
Disconnected result is to have suffered from protrusion of lumber intervertebral disc disease.And this mode is the shadow of static collection pain patients privileged site
, can not dynamic realtime, the irregular part dynamic change situation for obtaining pain patients as learning figure.It is and how more early
Discovery back pain symptom, and carry out rehabilitation training or taking simple measure to be adjusted in time, avoid aching with this
The evolution of symptoms of pain patient is just particularly important into protrusion of lumber intervertebral disc.
It using bone as lever, joint is hinge that the motion of human body, which is, contraction of muscle for power, and in the domination of nervous system
It is lower to coordinate to complete.Contraction of muscle plays engine in human motion, and power is provided for human motion.Back pain illness limits
Waist muscle activity processed, causes muscle function to degenerate, and the decline of the contractility of low back muscle can directly affect lumbar spinal column
Structural stability, the damage of Facet joint and its surrounding ligaments tissue and interverbebral disc is caused, so as to cause back pain.Trunk muscles
Tension force protection backbone caused by meat contraction, backbone is acted on caused pressure, therefore tied when backbone sustains damage
When structure sexually revises, muscle systems often changes prior to other structures.Therefore, the back pain of a variety of causes is all in different journeys
It is relevant with muscle systems dysfunction on degree, particularly maintain the stable related core muscle group of lumbar vertebrae.Lumbar vertebrae core stable muscle group
As the chief component of vertebra active subsystem, play the role of for the stability and activity for maintaining vertebra important.
The long-term work of back pain pain patients, or overwork cause waist muscle that lesion occurs, so as to cause motion god
Impulsion is provided through member to slow down, ability to work declines.The change of myoarchitecture can make the electro-physiological signals of muscle in motion process
Change, gather this electro-physiological signals at present and mainly obtained by surface electromyogram signal equipment.Surface electromyogram signal is
A kind of micro- electric signal of non-stationary, its advanced 30-150ms more general than limb motion are produced, and its amplitude is in 0.01-10mV, master
To want energy to concentrate between 0-500Hz, contain abundant information, acquisition technique is ripe, and is noninvasive collection, thus by crowd
The favor of more researchers.And different frequencies can be decomposed in layer by wavelet decomposition according to wavelet decomposition theory, signal
On rate passage, characteristic that more can really in a certain frequency of reaction signal, because the signal after decomposition is in frequency content
It is more single than primary signal, and wavelet decomposition has been made smoothly to signal, thus after decomposing signal stationarity it is better than primary signal
Much.
For example, patent of invention 201510232056.4 propose it is a kind of based on the mode of wavelet function feedback by human pulse
Direct current and AC compounent separation in signal, which is by way of wavelet decomposition and reconstruct by the direct current in pulse signal and friendship
Flow component separates, but the patent is only that direct current and the AC compounent separation of pulse signal are carried out using wavelet decomposition, is only
A kind of mode for separating waveform, more in-depth study is not carried out for the signal transacting after separation.
The content of the invention
The invention provides a kind of health status sorting technique and device based on physiology signal, it is intended at least one
Determine solve one of above-mentioned technical problem of the prior art in degree.
In order to solve the above problems, the invention provides following technical scheme：
A kind of health status sorting technique based on physiology signal, including：
Step a：Wavelet decomposition processing is carried out to the human body surface myoelectric signal of collection, obtain every layer of low-frequency wavelet coefficients and
High-frequency wavelet coefficient；
Step b：Complexity processing is carried out to every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient, obtains every layer of low frequency
The complexity value of wavelet coefficient and high-frequency wavelet coefficient；
Step c：The sample entropy and complexity value of every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient are respectively compared,
Wavelet coefficient otherness parameter is determined, is classified according to the wavelet coefficient otherness parameters on human health status.
The technical scheme that the embodiment of the present invention is taken also includes：The step a also includes：Gather the table at human body pain position
Facial muscle electric signal；The surface electromyogram signal acquisition mode is：Subject is carried out continuously certain time using predetermined stance
Number, the surface electromyogram signal at human body pain position is gathered during the progress of the stance.
The technical scheme that the embodiment of the present invention is taken also includes：The step a also includes：Using Matlab2010b to adopting
The surface electromyogram signal collected carries out noise-removed filtering processing.
The technical scheme that the embodiment of the present invention is taken also includes：It is described to every layer of low frequency wavelet in the step b
Coefficient and high-frequency wavelet coefficient carry out complexity processing and specifically included：Every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient are distinguished
Two-value coarse processing is carried out, and the data after the processing of two-value coarse are carried out at complexity using Lempel_Ziv algorithms
Reason.
The technical scheme that the embodiment of the present invention is taken also includes：In the step a, the small wavelength-division of surface electromyogram signal
It is four layers to solve the number of plies；In the step c, the sample for being respectively compared every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient
This entropy and complexity value, determine that wavelet coefficient otherness parameter is specially：First compare low-frequency wavelet coefficients and high frequency wavelet system
Number carries out the sample entropy of two-value coarse before and after the processing, then compares low-frequency wavelet coefficients and high-frequency wavelet coefficient is carried out
Sample entropy and complexity value after the processing of two-value coarse, obtain wavelet coefficient otherness parameter；The wavelet coefficient difference
Property parameter is the 4th layer of low-frequency wavelet coefficients complexity value.
Another technical scheme that the embodiment of the present invention is taken is：A kind of health status classification dress based on physiology signal
Put, including：
Wavelet decomposition module：For carrying out wavelet decomposition processing to the human body surface myoelectric signal of collection, every layer of acquisition is low
Frequency wavelet coefficient and high-frequency wavelet coefficient；
Complexity processing module：For being carried out to every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient at complexity
Reason, obtains the complexity value of every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient；
Parameter calculating module：For being respectively compared the sample entropy of every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient
And complexity value, wavelet coefficient otherness parameter is determined, is entered according to the wavelet coefficient otherness parameters on human health status
Row classification.
The technical scheme that the embodiment of the present invention is taken also includes signal acquisition module, and the signal acquisition module is used to gather
The surface electromyogram signal at human body pain position；The surface electromyogram signal acquisition mode is：Subject uses predetermined standing appearance
Gesture is carried out continuously certain number, and the surface electromyogram signal at human body pain position is gathered during the progress of the stance.
The technical scheme that the embodiment of the present invention is taken also includes signal filtration module, and the signal filtration module is used to use
Matlab2010b carries out noise-removed filtering processing to the surface electromyogram signal collected.
The technical scheme that the embodiment of the present invention is taken also includes coarse processing module, and the coarse processing module is used for
Carry out two-value coarse processing respectively to every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient, the complexity processing module uses
Lempel_Ziv algorithms carry out complexity processing to the data after the processing of two-value coarse.
The technical scheme that the embodiment of the present invention is taken also includes：The wavelet decomposition module is by the small wavelength-division of surface electromyogram signal
It is four layers to solve the number of plies；The parameter calculating module determines that wavelet coefficient otherness parameter specifically includes：First compare low frequency wavelet system
Number and high-frequency wavelet coefficient carry out the sample entropy of two-value coarse before and after the processing, then compare low-frequency wavelet coefficients and high frequency is small
Wave system number carries out sample entropy and complexity value after progress two-value coarse processing, obtains wavelet coefficient otherness parameter；Institute
It is the 4th layer of low-frequency wavelet coefficients complexity value to state wavelet coefficient otherness parameter.
Relative to prior art, beneficial effect caused by the embodiment of the present invention is：The embodiment of the present invention based on human body
The health status sorting technique and device of physiological signal to the surface electromyogram signal of painful area by carrying out wavelet decomposition, extraction
Go out every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient, and two-value coarse processing is carried out to every layer of wavelet coefficient, after processing
Coefficient carry out Lempel_Ziv product complexity theory processing, extract can truly, objectively reflect pain patients and healthy person it
Between otherness complexity value, effectively the health status of subject is classified so as to objective, is easy to doctor to make more
Add correct pathological diagnosis.The present invention is noninvasive, painless, simple to operate, and cost is cheap, and real-time and visuality are good, safety and sanitation,
Pain patients are acceptant, it is easy to accomplish；Can the periodically constantly rehabilitation to pain patients in clinical practice, rehabilitation training
Effect is monitored in real time, understands the illness situation of change of pain patients in time, contributes to doctor and rehabilitation Shi Jinhang to treat and arrange
The adjustment with training method is applied, helps pain patients early recovery.
Brief description of the drawings
Fig. 1 is the flow chart of the health status sorting technique based on physiology signal of the embodiment of the present invention；
Fig. 2 be for wavelet decomposition after each layer wavelet coefficient schematic diagram；
Fig. 3 is low-frequency wavelet coefficients and high-frequency wavelet coefficient Sample Entropy and complexity curve map；
Fig. 4 (a) is healthy person low-frequency wavelet coefficients and high-frequency wavelet coefficient Sample Entropy and complexity curve map, Fig. 4 (b) are
Pain patients low-frequency wavelet coefficients and high-frequency wavelet coefficient Sample Entropy and complexity curve map；
Fig. 5 is that pain patients and healthy person surface electromyogram signal wavelet decomposition low-frequency wavelet coefficients complexity compare signal
Figure；
Fig. 6 is the structural representation of the health status sorter based on physiology signal of the embodiment of the present invention.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that specific embodiment described herein is only to explain the present invention, not
For limiting the present invention.
The health status sorting technique based on physiology signal and device of the embodiment of the present invention use Matlab softwares
Wavelet decomposition is carried out to the surface electromyogram signal of the painful area of collection, extracts every layer of low-frequency wavelet coefficients and high frequency wavelet system
Number, and two-value coarse processing is carried out to every layer of wavelet coefficient, Lempel_Ziv product complexity theories are carried out to the coefficient after processing
Processing, extracts the complexity value that can show difference between back pain pain patients and healthy person, so as to subject for pain
The health status at position is classified.In the examples below, the present invention is only carried out specific so that painful area is lower back portion as an example
Explanation, it will be understood that in other embodiments of the present invention, painful area can also include other positions, such as cervical vertebra, lumbar vertebrae or
Four limbs etc..
Referring to Fig. 1, it is the flow chart of the health status sorting technique based on physiology signal of the embodiment of the present invention.
The health status sorting technique based on physiology signal of the embodiment of the present invention comprises the following steps：
Step 100：Gather the surface electromyogram signal of multidigit subject's waist multifidus muscle；
In step 100, subject includes pain patients and healthy person respectively, and signal acquisition mode is：Subject uses
Standard gestures are uprightly stood, and then bend trunk forward as far as possible, after reaching itself bending maximum position, are returned and are stood
Position, certain number (embodiment of the present invention is preferably four times) is carried out continuously, and do the best forward and bend in upright standing-trunk-
During returning to erect position, the surface electromyogram signal of subject's waist left and right sides multifidus is gathered；In other implementations of the present invention
In example, the signal acquisition mode of surface electromyogram signal and collection position are according to the different and different of painful area.
Step 200：Noise-removed filtering processing is carried out to the multifidus muscle surface electromyographic signal collected, to filtered table
Facial muscle electric signal carries out wavelet decomposition processing, and obtains every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient；
In step 200, noise-removed filtering processing mode be using Matlab2010b carry out noise-removed filtering, the present invention its
In his embodiment, other noise-removed filtering modes can be also taken., will after the surface electromyogram signal progress noise-removed filtering processing of collection
Surface electromyogram signal wavelet decomposition is 4 layers, specifically as shown in Fig. 2 being each layer wavelet coefficient schematic diagram after wavelet decomposition, its
In, low frequency layer is cA1, cA2, cA3, cA4, and high frequency layer is cD1, cD2, cD3, cD4.
Step 300：Carry out two-value coarse processing respectively to every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient；
Step 400：Complexity processing is carried out to the data after the processing of two-value coarse using Lempel_Ziv algorithms, respectively
Obtain the complexity value of every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient；
Step 500：The sample entropy and complexity value of every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient are respectively compared, really
Determine wavelet coefficient otherness parameter；
In step 500, wavelet coefficient manner of comparison is：First compare low-frequency wavelet coefficients and high-frequency wavelet coefficient carries out two
It is worth the sample entropy of coarse before and after the processing, then compares low-frequency wavelet coefficients and high-frequency wavelet coefficient carries out carrying out two-value coarse grain
Sample entropy and complexity value after change processing, obtain surface electromyogram signal wavelet coefficient otherness parameter, as differentiation pain
Otherness parameter between patient and healthy person.
Wherein, sample entropy and complexity value are all the indexs for embodying signal complexity, but for back pain illness
Pain patients, research finds that complexity value and sample entropy have larger difference, as shown in figure 3, be low-frequency wavelet coefficients and
High-frequency wavelet coefficient Sample Entropy and complexity curve map.Wherein, top curve does not carry out two-value coarse after representing wavelet decomposition
The low-frequency wavelet coefficients and high-frequency wavelet coefficient Sample Entropy of processing；Intermediate curve is carried out at two-value coarse after representing wavelet decomposition
The low-frequency wavelet coefficients and high-frequency wavelet coefficient Sample Entropy of reason；Issue after curve represents wavelet decomposition and carry out two-value coarse processing
Low-frequency wavelet coefficients and high-frequency wavelet coefficient complexity.Comparative sample entropy and complexity curve, as a result show：Two-value is not carried out
After coarse processing and progress two-value coarse processing, be present very big difference in low-frequency wavelet coefficients sample entropy, but become substantially
Gesture is consistent, and high-frequency wavelet coefficient sample entropy is without significant difference；Carry out two-value coarse after, low-frequency wavelet coefficients Sample Entropy and
There were significant differences for complexity, and trend also has different, and high-frequency wavelet coefficient Sample Entropy and complexity are without significant difference
Property.Also referring to Fig. 4 (a) and Fig. 4 (b), Fig. 4 (a) be healthy person low-frequency wavelet coefficients and high-frequency wavelet coefficient Sample Entropy and
Complexity curve map, Fig. 4 (b) are pain patients low-frequency wavelet coefficients and high-frequency wavelet coefficient Sample Entropy and complexity curve map.
Healthier person and the Sample Entropy and complexity of pain patients wavelet coefficient, do not carry out two-value coarse processing after wavelet decomposition
Wavelet coefficient in, low frequency (cA1, cA2, cA3, cA4) wavelet coefficient and high frequency (cD1, cD2, cD3, cD4) wavelet coefficient sample
Entropy equal not assuming a marked difference property in pain patients and healthy person；After wavelet decomposition carries out two-value coarse processing
Wavelet coefficient sample entropy, each low frequency (cA1, cA2, cA3, cA4) wavelet coefficient and high frequency in pain patients and healthy person
(cD1, cD2, cD3, cD4) wavelet coefficient does not show significant group property yet；And carry out two-value coarse in wavelet decomposition
After processing, Lempel_Ziv complicated dynamic behaviours are carried out to wavelet coefficient, as a result found in pain patients and healthy person, the only the 4th
There were significant differences for layer low-frequency wavelet coefficients cA4 complexities, therefore using the 4th layer of low-frequency wavelet coefficients cA4 complexity value as wavelet systems
Number otherness parameter, classifies to the health status of subject.
Step 600：The health status at subject for pain position is classified according to wavelet coefficient otherness parameter.
In step 600, the embodiment of the present invention is made with the 4th layer of low-frequency wavelet coefficients complexity value cA4 of surface electromyogram signal
For wavelet coefficient otherness parameter, pain patients and healthy person are distinguished, specifically as shown in figure 5, being pain patients and healthy person table
Facial muscle electric signal wavelet decomposition low-frequency wavelet coefficients complexity comparison schematic diagram.As a result show, surface electromyogram signal is carried out small
After Wave Decomposition, low-frequency wavelet coefficients cA4 complexity values can reflect, significant otherness between pain patients and healthy person.Cause
This, by physiology flash-over characteristic during multifidus muscular movement in itself, is handled by wavelet decomposition, at two-value coarse
Reason and Lempel_Ziv complexities processing after the 4th layer of low-frequency wavelet coefficients complexity value, can truly, objectively reflect pain
Otherness between patient and healthy person.Due to this mode it is noninvasive, it is painless, simple to operate, can be monitored in real time, facing
Periodically constantly pain patients can be monitored in bed application, rehabilitation training, understand the illness change of pain patients in time
Situation, contribute to the adjustment of doctor and rehabilitation Shi Jinhang remedy measures and training method, help pain patients early recovery.
Referring to Fig. 6, it is that the structure of the health status sorter based on physiology signal of the embodiment of the present invention is shown
It is intended to.The health status sorter based on physiology signal of the embodiment of the present invention includes signal acquisition module, signal is filtered
Ripple module, wavelet decomposition module, coarse processing module, complexity processing module and parameter calculating module.Specifically：
Signal acquisition module：For gathering the surface electromyogram signal of multidigit subject's waist multifidus muscle；Wherein, it is tested
Person includes pain patients and healthy person respectively, and signal acquisition mode is：Subject is uprightly stood using standard gestures, then to the greatest extent may be used
Energy bends trunk forward, after reaching itself bending maximum position, returns to erect position, is carried out continuously certain number, and straight
Vertical standing-trunk is done the best forward during bending-return erect position, gathers the surface of subject's waist left and right sides multifidus
Electromyographic signal；In other embodiments of the present invention, the collection position of surface electromyogram signal is according to the different and different of painful area.
Signal filtration module：For carrying out noise-removed filtering processing to the multifidus muscle surface electromyographic signal collected；Go
Filtering process of making an uproar mode is to carry out noise-removed filtering using Matlab2010b, in other embodiments of the present invention, can also take other
Noise-removed filtering mode.
Wavelet decomposition module：For to filtered surface electromyogram signal carry out wavelet decomposition processing, and obtain every layer it is low
Frequency wavelet coefficient and high-frequency wavelet coefficient；After surface electromyogram signal carries out noise-removed filtering processing, wavelet decomposition module is by surface flesh
Electric signal wavelet decomposition is 4 layers, specifically as shown in Fig. 2 being each layer wavelet coefficient schematic diagram after wavelet decomposition, wherein, low frequency
Layer is cA1, cA2, cA3, cA4, and high frequency layer is cD1, cD2, cD3, cD4.
Coarse processing module：For carrying out two-value coarse respectively to every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient
Processing；
Complexity processing module：It is complicated for being carried out using Lempel_Ziv algorithms to the data after the processing of two-value coarse
Degree processing, respectively obtain the complexity value of every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient；
Parameter calculating module：For being respectively compared the sample entropy of every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient and answering
Miscellaneous angle value, wavelet coefficient otherness parameter is determined, the healthy shape according to wavelet coefficient otherness parameter to subject for pain position
State is classified.Wherein, wavelet coefficient manner of comparison is：First compare low-frequency wavelet coefficients and high-frequency wavelet coefficient progress two-value is thick
Sample entropy before and after granulated processed, then compares low-frequency wavelet coefficients and high-frequency wavelet coefficient carry out at two-value coarse
Sample entropy and complexity value after reason, surface electromyogram signal wavelet coefficient otherness parameter is obtained, as differentiation pain patients
Otherness parameter between healthy person.
Sample entropy and complexity value are all the indexs for embodying signal complexity, but for the pain of back pain illness
Patient, research finds that complexity value and sample entropy have larger difference, as shown in figure 3, being that low-frequency wavelet coefficients and high frequency are small
Wave system numerical example entropy and complexity curve map.Wherein, top curve does not carry out two-value coarse processing after representing wavelet decomposition
Low-frequency wavelet coefficients and high-frequency wavelet coefficient Sample Entropy；Intermediate curve carries out the low of two-value coarse processing after representing wavelet decomposition
Frequency wavelet coefficient and high-frequency wavelet coefficient Sample Entropy；Issue the low frequency that curve represents progress two-value coarse processing after wavelet decomposition
Wavelet coefficient and high-frequency wavelet coefficient complexity.Comparative sample entropy and complexity curve, as a result show：Two-value coarse is not carried out
After processing and progress two-value coarse processing, there is very big difference in low-frequency wavelet coefficients sample entropy, but basic trend is consistent,
And high-frequency wavelet coefficient sample entropy is without significant difference；After carrying out two-value coarse, low-frequency wavelet coefficients Sample Entropy and complexity
There were significant differences, and trend also has different, and high-frequency wavelet coefficient Sample Entropy and complexity are without the significance difference opposite sex.Please one
And refering to Fig. 4 (a) and Fig. 4 (b), Fig. 4 (a) is that healthy person low-frequency wavelet coefficients and high-frequency wavelet coefficient Sample Entropy and complexity are write music
Line chart, Fig. 4 (b) are pain patients low-frequency wavelet coefficients and high-frequency wavelet coefficient Sample Entropy and complexity curve map.It is healthier
Person and the Sample Entropy and complexity of pain patients wavelet coefficient, do not carry out the wavelet systems of two-value coarse processing after wavelet decomposition
In number, low frequency (cA1, cA2, cA3, cA4) wavelet coefficient and high frequency (cD1, cD2, cD3, cD4) wavelet coefficient sample entropy are aching
Equal not assuming a marked difference property in pain patient and healthy person；Wavelet systems after wavelet decomposition carries out two-value coarse processing
Numerical example entropy, in pain patients and healthy person each low frequency (cA1, cA2, cA3, cA4) wavelet coefficient and high frequency (cD1, cD2,
CD3, cD4) wavelet coefficient do not show significant group property yet；And after wavelet decomposition carries out two-value coarse processing, it is right
Wavelet coefficient carries out Lempel_Ziv complicated dynamic behaviours, as a result finds in pain patients and healthy person, only the 4th layer of low frequency is small
There were significant differences for wave system number cA4 complexities, therefore using the 4th layer of low-frequency wavelet coefficients cA4 complexity value as wavelet coefficient otherness
Parameter, the health status at subject for pain position is classified.Specifically as shown in figure 5, being pain patients and healthy person surface
Electromyographic signal wavelet decomposition low-frequency wavelet coefficients complexity comparison schematic diagram.As a result show, small echo is carried out to surface electromyogram signal
After decomposition, low-frequency wavelet coefficients complexity value cA4 can reflect, significant otherness between pain patients and healthy person.Therefore,
By physiology flash-over characteristic during multifidus muscular movement in itself, handled by wavelet decomposition, along with the processing of two-value coarse
With the 4th layer of low-frequency wavelet coefficients complexity value after the processing of Lempel_Ziv complexities, it can truly, objectively reflect that pain is suffered from
Otherness between person and healthy person.
Based on above-described embodiment, the present invention acquires 101 nurses and leaned forward the surface myoelectric of the multifidus muscle in motion
Signal, after filtering process, wavelet decomposition is carried out to surface electromyogram signal, obtains low-frequency wavelet coefficients and high frequency wavelet system
Number, and the property difference of the lower wavelet coefficient of different disposal operation is compared, obtain by two-value coarse and Lempel_Ziv
The 4th layer of low-frequency wavelet coefficients complexity value after processing, it was demonstrated that the 4th layer of low-frequency wavelet coefficients complexity value can with it is objective,
It is effective to distinguish back pain patient and health population.
The health status sorting technique based on physiology signal and device of the embodiment of the present invention pass through to painful area
Surface electromyogram signal carry out wavelet decomposition, extract every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient, and to every layer of small echo
Coefficient carries out two-value coarse processing, and Lempel_Ziv product complexity theory processing is carried out to the coefficient after processing, and extracting can be true
Real, the objectively otherness between reflection pain patients and healthy person complexity value, so as to objective effectively to subject's
Health status is classified, and is easy to doctor to make more correct pathological diagnosis.Noninvasive, painless, simple to operate, cost of the invention
Cheap, real-time and visuality are good, safety and sanitation, and pain patients are acceptant, it is easy to accomplish；In clinical practice, rehabilitation training
In periodically constantly the rehabilitation efficacy of pain patients can be monitored in real time, in time understand pain patients illness change feelings
Condition, contribute to the adjustment of doctor and rehabilitation Shi Jinhang remedy measures and training method, help pain patients early recovery.
The foregoing description of the disclosed embodiments, professional and technical personnel in the field are enable to realize or using the present invention.
A variety of modifications to these embodiments will be apparent for those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, it is of the invention
The embodiments shown herein is not intended to be limited to, and is to fit to and principles disclosed herein and features of novelty phase one
The most wide scope caused.
Claims (10)
- A kind of 1. health status sorting technique based on physiology signal, it is characterised in that including：Step a：Wavelet decomposition processing is carried out to the human body surface myoelectric signal of collection, obtains every layer of low-frequency wavelet coefficients and high frequency Wavelet coefficient；Step b：Complexity processing is carried out to every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient, obtains every layer of low frequency wavelet The complexity value of coefficient and high-frequency wavelet coefficient；Step c：The sample entropy and complexity value of every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient are respectively compared, it is determined that Wavelet coefficient otherness parameter, classified according to the wavelet coefficient otherness parameters on human health status.
- 2. the health status sorting technique according to claim 1 based on physiology signal, it is characterised in that the step Rapid a also includes：Gather the surface electromyogram signal at human body pain position；The surface electromyogram signal acquisition mode is：Subject adopts Certain number is carried out continuously with predetermined stance, human body pain position is gathered during the progress of the stance Surface electromyogram signal.
- 3. the health status sorting technique according to claim 2 based on physiology signal, it is characterised in that the step Rapid a also includes：Noise-removed filtering processing is carried out to the surface electromyogram signal collected using Matlab2010b.
- 4. the health status sorting technique according to claim 3 based on physiology signal, it is characterised in that described It is described that every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient progress complexity processing are specifically included in step b：To every layer Low-frequency wavelet coefficients and high-frequency wavelet coefficient carry out two-value coarse processing respectively, and thick to two-value using Lempel_Ziv algorithms Data after granulated processed carry out complexity processing.
- 5. the health status sorting technique according to claim 4 based on physiology signal, it is characterised in that described In step a, the surface electromyogram signal wavelet decomposition number of plies is four layers；In the step c, it is described be respectively compared it is described every layer The sample entropy and complexity value of low-frequency wavelet coefficients and high-frequency wavelet coefficient, determine that wavelet coefficient otherness parameter is specially： First compare low-frequency wavelet coefficients and high-frequency wavelet coefficient carries out the sample entropy of two-value coarse before and after the processing, then compare low frequency Wavelet coefficient and high-frequency wavelet coefficient carry out sample entropy and complexity value after progress two-value coarse processing, obtain wavelet systems Number otherness parameter；The wavelet coefficient otherness parameter is the 4th layer of low-frequency wavelet coefficients complexity value.
- A kind of 6. health status sorter based on physiology signal, it is characterised in that including：Wavelet decomposition module：For carrying out wavelet decomposition processing to the human body surface myoelectric signal of collection, it is small to obtain every layer of low frequency Wave system number and high-frequency wavelet coefficient；Complexity processing module：For carrying out complexity processing to every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient, obtain To the complexity value of every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient；Parameter calculating module：For being respectively compared the sample entropy of every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient and answering Miscellaneous angle value, wavelet coefficient otherness parameter is determined, divided according to the wavelet coefficient otherness parameters on human health status Class.
- 7. the health status sorter according to claim 6 based on physiology signal, it is characterised in that also include Signal acquisition module, the signal acquisition module are used for the surface electromyogram signal for gathering human body pain position；The surface myoelectric Signal acquisition mode is：Subject is carried out continuously certain number using predetermined stance, in the progress of the stance During gather the surface electromyogram signal at human body pain position.
- 8. the health status sorter according to claim 7 based on physiology signal, it is characterised in that also include Signal filtration module, the signal filtration module are used to remove the surface electromyogram signal collected using Matlab2010b Make an uproar filtering process.
- 9. the health status sorter according to claim 8 based on physiology signal, it is characterised in that also include Coarse processing module, the coarse processing module are used to carry out every layer of low-frequency wavelet coefficients and high-frequency wavelet coefficient respectively The processing of two-value coarse, the complexity processing module are entered using Lempel_Ziv algorithms to the data after the processing of two-value coarse The processing of row complexity.
- 10. the health status sorter according to claim 9 based on physiology signal, it is characterised in that described The surface electromyogram signal wavelet decomposition number of plies is four layers by wavelet decomposition module；The parameter calculating module determines wavelet coefficient difference Property parameter specifically includes：First compare low-frequency wavelet coefficients and high-frequency wavelet coefficient carries out the Sample Entropy of two-value coarse before and after the processing Value, then compares low-frequency wavelet coefficients and high-frequency wavelet coefficient carries out sample entropy and complexity after progress two-value coarse processing Angle value, obtain wavelet coefficient otherness parameter；The wavelet coefficient otherness parameter is the 4th layer of low-frequency wavelet coefficients complexity Value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610817417.6A CN107811635B (en) | 2016-09-12 | 2016-09-12 | Health state classification method and device based on human physiological signals |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610817417.6A CN107811635B (en) | 2016-09-12 | 2016-09-12 | Health state classification method and device based on human physiological signals |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107811635A true CN107811635A (en) | 2018-03-20 |
| CN107811635B CN107811635B (en) | 2021-01-19 |
Family
ID=61600513
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610817417.6A Active CN107811635B (en) | 2016-09-12 | 2016-09-12 | Health state classification method and device based on human physiological signals |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107811635B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109498001A (en) * | 2018-12-25 | 2019-03-22 | 深圳和而泰数据资源与云技术有限公司 | Sleep quality appraisal procedure and device |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000010455A1 (en) * | 1998-08-24 | 2000-03-02 | Emory University | Method and apparatus for predicting the onset of seizures based on features derived from signals indicative of brain activity |
| US20080108909A1 (en) * | 2006-10-02 | 2008-05-08 | The Cleveland Clinic Foundation | Prediction of clinical outcome using large array surface myoelectric potentials |
| CN102488516A (en) * | 2011-12-13 | 2012-06-13 | 湖州康普医疗器械科技有限公司 | Nonlinear electroencephalogram signal analysis method and device |
| CN102697495A (en) * | 2012-05-23 | 2012-10-03 | 杭州电子科技大学 | Second-generation wavelet electromyographic signal noise eliminating method based on ensemble empirical mode decomposition |
| CN102961203A (en) * | 2012-12-10 | 2013-03-13 | 杭州电子科技大学 | Method for identifying surface electromyography (sEMG) on basis of empirical mode decomposition (EMD) sample entropy |
| CN103345641A (en) * | 2013-07-16 | 2013-10-09 | 杭州电子科技大学 | Hand electromyographic signal motion recognition method based on wavelet entropy and support vector machine |
| CN103584872A (en) * | 2013-10-29 | 2014-02-19 | 燕山大学 | Psychological stress assessment method based on multi-physiological-parameter integration |
| WO2015150931A1 (en) * | 2014-04-03 | 2015-10-08 | Universiti Brunei Darussalam | Realtime biofeedback mechanism and data presentation for knee injury rehabilitation monitoring and a soft real time intelligent system thereof |
-
2016
- 2016-09-12 CN CN201610817417.6A patent/CN107811635B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000010455A1 (en) * | 1998-08-24 | 2000-03-02 | Emory University | Method and apparatus for predicting the onset of seizures based on features derived from signals indicative of brain activity |
| US20080108909A1 (en) * | 2006-10-02 | 2008-05-08 | The Cleveland Clinic Foundation | Prediction of clinical outcome using large array surface myoelectric potentials |
| CN102488516A (en) * | 2011-12-13 | 2012-06-13 | 湖州康普医疗器械科技有限公司 | Nonlinear electroencephalogram signal analysis method and device |
| CN102697495A (en) * | 2012-05-23 | 2012-10-03 | 杭州电子科技大学 | Second-generation wavelet electromyographic signal noise eliminating method based on ensemble empirical mode decomposition |
| CN102961203A (en) * | 2012-12-10 | 2013-03-13 | 杭州电子科技大学 | Method for identifying surface electromyography (sEMG) on basis of empirical mode decomposition (EMD) sample entropy |
| CN103345641A (en) * | 2013-07-16 | 2013-10-09 | 杭州电子科技大学 | Hand electromyographic signal motion recognition method based on wavelet entropy and support vector machine |
| CN103584872A (en) * | 2013-10-29 | 2014-02-19 | 燕山大学 | Psychological stress assessment method based on multi-physiological-parameter integration |
| WO2015150931A1 (en) * | 2014-04-03 | 2015-10-08 | Universiti Brunei Darussalam | Realtime biofeedback mechanism and data presentation for knee injury rehabilitation monitoring and a soft real time intelligent system thereof |
Non-Patent Citations (6)
| Title |
|---|
| J. ZHAO.ET AL.: "A Novel EMG Motion Pattern Classifier Based on Wavelet Transform and Nonlinearity Analysis Method", 《 2006 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND BIOMIMETICS, KUNMING, 2006》 * |
| REGER STEVEN I;SHAH AMRIK: "Classification of large array surface myoelectric potentials from subjects with and without low back pain", 《JOURNAL OF ELECTROMYOGRAPHY AND KINESIOLOGY : OFFICIAL JOURNAL OF THE INTERNATIONAL SOCIETY OF ELECTROPHYSIOLOGICAL KINESIOLOGY》 * |
| SHRAVANTI KALWA ; H.T. PATIL: "Neuromuscular Disease Classification by Wavelet Decomposition Technique", 《2015 INTERNATIONAL CONFERENCE ON COMMUNICATIONS AND SIGNAL PROCESSING (ICCSP)》 * |
| 曹玉珍: "表面肌电信号的时变AR模型参数估计及其在下腰痛评估中的应用研究", 《北京生物医学工程》 * |
| 曹铭: "意识障碍患者脑电信号的非线性动力学评价分析", 《中国优秀硕士学位论文全文数据库》 * |
| 肖毅: "基于熵理论和复杂度的肌电信号分析", 《科技导报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109498001A (en) * | 2018-12-25 | 2019-03-22 | 深圳和而泰数据资源与云技术有限公司 | Sleep quality appraisal procedure and device |
| CN109498001B (en) * | 2018-12-25 | 2021-08-31 | 深圳和而泰数据资源与云技术有限公司 | Sleep quality evaluation method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107811635B (en) | 2021-01-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rafiee et al. | Wavelet basis functions in biomedical signal processing | |
| Curt et al. | Traumatic cervical spinal cord injury: relation between somatosensory evoked potentials, neurological deficit, and hand function | |
| Hu et al. | Is the psoas a hip flexor in the active straight leg raise? | |
| Suresh et al. | Examination of motor unit control properties in stroke survivors using surface EMG decomposition: a preliminary report | |
| CN107137080A (en) | Chronic back pain patient muscle's active state determination methods and system | |
| McHugh et al. | Epidural spinal cord stimulation for motor recovery in spinal cord injury: A systematic review | |
| Moreside et al. | Temporal Patterns of the Trunk Muscles Remain Altered in a Low Back–Injured Population Despite Subjective Reports of Recovery | |
| Pantall et al. | Changes in muscle activation patterns in response to enhanced sensory input during treadmill stepping in infants born with myelomeningocele | |
| Tsang et al. | Relationship between neck acceleration and muscle activation in people with chronic neck pain: implications for functional disability | |
| Hu et al. | The impact of evoked cutaneous afferents on voluntary reaching movement in patients with Parkinson’s disease | |
| CN107811635A (en) | A kind of health status sorting technique and device based on physiology signal | |
| Vulfsons et al. | Decreased amplitude of surface electromyo-graphic recordings of muscle activation along the posterior myofascial kinematic chain in subjects with chronic nonspecific low back pain compared to healthy subjects | |
| Sandhya et al. | Single channel event related (de) synchronization (erd/ers) analysis of motor execution in stroke affected foot drop subjects | |
| Sahu et al. | Effect of various standing poses of yoga on the musculoskeletal system using EMG | |
| Andreoni et al. | Example of clinical applications of wearable monitoring systems | |
| Gao et al. | [Retracted] Study on the Mechanism of Pathological Recognition Based on Bioelectrical Impedance Spectrum to the Elbow Joint | |
| Praba et al. | Analysis of Physiological Signals During Therapeutic Yoga Practice | |
| RU2147858C1 (en) | Method for treating degenerative dystrophy diseases of cervical region of the vertebral column | |
| Melo et al. | Technological aspects of traumatic spinal cord injury rehabilitation | |
| CN202665917U (en) | Vertical human thoracolumbar spine healthcare therapeutic instrument | |
| Schilling | Leg Paralysis Can Be Treated | |
| Strazza | Advanced techniques for EMG-based assessment of muscular co-contraction during walking | |
| Peña Pitarch et al. | Preface: a special selection on biomechanical applications in medical science-Part I | |
| Talija | Biomedical Implantable Low-Power Fault-Tolerant Systems | |
| Tharu | Trans-spinal electrical stimulation for improving trunk and sitting function in tetraplegics with cervical cord injury |
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 |