CN106539586B - A kind of respiratory rate calculation method and device - Google Patents
A kind of respiratory rate calculation method and device Download PDFInfo
- Publication number
- CN106539586B CN106539586B CN201610976238.7A CN201610976238A CN106539586B CN 106539586 B CN106539586 B CN 106539586B CN 201610976238 A CN201610976238 A CN 201610976238A CN 106539586 B CN106539586 B CN 106539586B
- Authority
- CN
- China
- Prior art keywords
- signal
- respiratory rate
- parameter
- parameter signal
- breath
- 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
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
- A61B5/0816—Measuring devices for examining respiratory frequency
-
- 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
- A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Pathology (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Veterinary Medicine (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Heart & Thoracic Surgery (AREA)
- Physiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Pulmonology (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Cardiology (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
A kind of respiratory rate calculation method and device, this method comprises: obtaining electrocardiosignal and being pre-processed, the extracting parameter signal from the electrocardiosignal, the parameter signal includes training parameter signal and test parameter signal；Sef-adapting filter is adjusted by training parameter signal, the test is handled with the sef-adapting filter adjusted to obtain the first breath signal with parameter signal, and calculate the first respiratory rate according to first breath signal；Neural network training model is constructed by training parameter signal, by the neural network training model from the test with obtaining the second breath signal in parameter signal, and the second respiratory rate is calculated according to second breath signal；Final respiratory rate is obtained by the first respiratory rate described in weighted calculation and second respiratory rate.The present invention extracts breath signal by sef-adapting filter and neural network from electrocardiosignal, obtains accurate respiratory rate, is suitable for daily monitoring.
Description
Technical field
The present invention relates to field of ECG signal processing, more particularly to a kind of respiratory rate calculation method and device.
Background technique
Method currently used for calculating respiratory rate mainly has: impedance volumetric method, with high-frequency constant current source measurement chest impedance
Variation is to extract respiration information；Sensor method uses temperature, pressure, humidity and gas flow transducer as nostril sensor；Capacitor
Method causes capacitance to generate corresponding variation when breathing；Breath sound method, by picking up breath sound identification of breathing；Ultrasonic method, benefit
Doppler phenomenon is generated with ultrasonic wave, detects respiratory rate.Not only need to increase signal acquisition component using these methods, and
And the shadow noon by movement and environment, be not suitable for daily monitoring.
A large amount of clinical datas show that respiratory movement can cause the variation of electrocardiogram.By electrocardiogram, we can observe that
By the change of the caused ecg wave form peak-to-peak value of chest exercise and cardiac position variation within the respiratory cycle.This is because breathing
In period, the heart electric axis rotation in description heart electric wave main propagation direction causes QRS complex form to be changed.From electrocardio
The method that breath signal (ECG-DerivedRespiration, EDR) is extracted in signal does not need sensor special and hardware mould
Block detects breath signal, it is only necessary to obtain electrocardiosignal with ECG monitor, avoid above two detection method to human body
Constraint, makes it possible dynamic breathing detection.
However in the prior art, the technology of breath signal and immature is extracted from electrocardiosignal, there are still some problems.Example
Such as by electrocardiosignal extract breath signal during due to electrocardio, the non-linear of breath signal, randomness and non-stationary
Characteristic, the problem of causing the loss and cross jamming of breath signal, keep the respiratory rate being calculated inaccurate, and the Shandong of system
Stick is poor.
Summary of the invention
In view of the above situation, it is necessary to which the problem that inaccuracy is calculated for respiratory rate in the prior art provides one kind and is based on
The method and device of electrocardiosignal calculating respiratory rate.
The present invention provides a kind of respiratory rate calculation methods, comprising:
It obtains electrocardiosignal and is pre-processed, the extracting parameter signal from the electrocardiosignal, the parameter signal packet
Include training parameter signal and test parameter signal；
Sef-adapting filter is adjusted by training parameter signal, with the sef-adapting filter adjusted to institute
It states test to be handled to obtain the first breath signal with parameter signal, and the first breathing is calculated according to first breath signal
Rate；
By the training with parameter signal construct neural network training model, by the neural network training model from
The test calculates the second respiratory rate according to second breath signal with obtaining the second breath signal in parameter signal；
Final respiratory rate is obtained by the first respiratory rate described in weighted calculation and second respiratory rate.
The above method, wherein it is described by the training parameter signal adjust sef-adapting filter the step of include:
The breath signal estimated is handled with parameter signal to the training by sef-adapting filter；
The difference for the breath signal that the breath signal and impedance method estimated described in calculating obtain, and adjust the adaptive filter
The parameter of wave device makes the difference in default range.
The above method, wherein the step of sef-adapting filter handles the parameter signal include:
Fuzzy reasoning is carried out to the parameter signal according to preset fuzzy rule, after obtaining the first of the fuzzy rule
Part；
Function expansion is carried out to the parameter signal by orthogonal basis function, obtains the second consequent of the fuzzy rule；
According to first consequent and the second consequent, the output signal of the sef-adapting filter is determined.
The above method, wherein it is described that fuzzy reasoning is carried out to the parameter signal according to preset fuzzy rule, obtain institute
The step of stating the first consequent of fuzzy rule include:
The parameter signal is calculated in the degree of membership of the fuzzy subset of each neuron node of sef-adapting filter；
The excitation density of every fuzzy rule is calculated according to the degree of membership；
The excitation density is normalized, the first consequent of the fuzzy rule is obtained.
The above method, wherein the calculation formula of the degree of membership are as follows:
Wherein,
In above-mentioned formula,Indicate degree of membership of the input signal vector in j-th of neuron node, l=1,2 ..., r table
Show the number of input signal variable, j=1,2 ..., n indicate the number of subordinating degree function, μlj(xl) indicate first of reference input letter
Number variable xlIn the subordinating degree function of j-th of neuron node, cljIndicate first of reference-input signal in j-th of neuron section
The center of the subordinating degree function of point,Indicate the width of the subordinating degree function of j-th of neuron node.
The above method, wherein the calculation formula of the excitation density are as follows:
Wherein,For the excitation density of j-th strip fuzzy rule, cljIndicate first of input signal in j-th of neuron section
The center of the subordinating degree function of point,Indicate the width of the subordinating degree function of j-th of neuron node.
The above method, wherein the calculation formula of first consequent are as follows:
Wherein,For the excitation density of j-th strip fuzzy rule,For the first consequent of j-th strip fuzzy rule, n is mould
Paste the quantity of rule.
The above method, wherein the orthogonal basis function are as follows:
The calculation formula of second consequent are as follows:
Wherein, Chm(xl) be output vector in first of element m-th of Chebyshev's orthogonal polynomial, T be transposition grasp
Make, M is the number of the orthogonal basis function, wjFor the second consequent of j-th strip fuzzy rule, α1j,α2j,...,αMjFor j-th strip mould
The consequent parameter set of rule is pasted, M is the number of the orthogonal basis function.
The above method, wherein it is described according to first consequent and the second consequent, determine the sef-adapting filter output
Signal the step of include:
The signal of the sef-adapting filter output is determined according to the following formula:
Y is the signal of sef-adapting filter output, OjFor the first consequent of j-th strip fuzzy rule, wjFor j-th strip mould
Paste the second consequent of rule.
The above method, wherein it is described by the training with parameter signal construct neural network training model the step of it
Before further include, by the parameter signal carry out dimensionality reduction, obtain principal component electrocardiosignal；
Described the step of constructing neural network training model by the training parameter signal includes being used with the training
Input sample of the corresponding principal component electrocardiosignal of parameter signal as neural network, the reference breath signal obtained with impedance method
As the training objective of the neural network, neural network training model is constructed；
It is described by the neural network training model from the test with obtaining the second breath signal in parameter signal
Step includes that the test is input to the neural network training model with the corresponding principal component electrocardiosignal of parameter signal
In, obtain final breath signal.
The above method, wherein it is described by the parameter signal carry out dimensionality reduction the step of include:
Parameter signal progress significance test is obtained into significant characteristics signal；
Dimensionality reduction is carried out to the significant characteristics signal by principal component analysis, obtains principal component electrocardiosignal.
The above method, wherein described that parameter signal progress significance test is obtained into the step of significant characteristics signal
Suddenly include:
The parameter signal is standardized, standard cardioelectric signal is obtained；
The standard cardioelectric signal is subjected to variance analysis and F is examined, obtains significant characteristics signal.
The above method, wherein the neural network use dynamic BP algorithm, the dynamic BP algorithm by factor of momentum into
Row weighed value adjusting, the calculation formula of weighed value adjusting are as follows:
W (k+1)=w (k)+Δ w (k+1)
Wherein, α indicates that the learning rate of network, η indicate factor of momentum,It indicates that kth time is reversed to pass
The error partial differential passed, w (k) indicate the threshold value or weight of kth time back transfer, and E (k) indicates the error of kth time back transfer
Summation.
The above method, wherein the neural network determines the number of the number of hidden nodes, the trial and error procedure packet by trial and error procedure
Include step:
First hidden node number is arranged near 1+i/2, gradually increases Hidden nodes to 2*i+1, it is bent to form error performance
Line；
The quantity of hidden node is determined by analytical error performance curve, wherein i is input layer number.
The above method, wherein the acquisition electrocardiosignal is simultaneously pre-processed, and extracting parameter is believed from the electrocardiosignal
Number the step of include:
It obtains electrocardiosignal and carries out power frequency filtering, obtain filtering signal；
The RR interval series signal and RW amplitude sequence signal in the filtering signal are extracted by threshold method.
The above method, wherein described to be obtained finally by the first respiratory rate described in weighted calculation and second respiratory rate
Respiratory rate the step of include:
Calculate separately the weight factor μ 1 and μ 2 of first breath signal and the second breath signal, calculation formula are as follows:
μ 1=(P1,1-P2,1)/P1,1
μ 2=(P1,2-P2,2)/P1,2
Final respiratory rate is calculated according to the weight factor μ 1 and μ 2 and first respiratory rate and the second respiratory rate, is counted
Calculation formula is R=μ 1*R1+ μ 2*R2,
Wherein, P1,1For the Power Spectrum Distribution of the first breath signal, P1,2For the Power Spectrum Distribution of the second breath signal, P2,1
For the Power Spectrum Distribution of the interference of the first breath signal, P2,2For the Power Spectrum Distribution of the interference of the second breath signal, R1 first
Respiratory rate is breathed, R2 is the second respiratory rate.
The present invention also provides a kind of respiratory rate computing devices, comprising:
ECG's data compression module obtains electrocardiosignal and is pre-processed, and extracting parameter is believed from the electrocardiosignal
Number, the parameter signal includes training parameter signal and test parameter signal；
First respiratory rate computing device, for adjusting sef-adapting filter by the training parameter signal and to adjust
The sef-adapting filter afterwards handles the test with parameter signal to obtain the first breath signal, and according to described
One breath signal calculates the first respiratory rate；
Second respiratory rate computing device, for constructing neural network training model by training parameter signal and leading to
The neural network training model is crossed from the test with obtaining the second breath signal in parameter signal, and is exhaled according to described second
It inhales signal and calculates the second respiratory rate；
Weighted calculation device, it is final for being obtained by the first respiratory rate described in weighted calculation and second respiratory rate
Respiratory rate.
Above-mentioned apparatus, wherein the first respiratory rate computing device includes:
It estimates breath signal and obtains module, for being carried out to the training with parameter signal correspondence by sef-adapting filter
Handle the breath signal estimated；
Module is adjusted, for calculating the breath signal estimated and with reference to the difference of breath signal, and is adjusted described adaptive
The parameter of filter, makes the difference in default range, the sef-adapting filter after being adjusted.
Above-mentioned apparatus, wherein the second respiratory rate computing device includes:
Dimension-reduction treatment module obtains principal component electrocardiosignal for the parameter signal to be carried out dimensionality reduction；
Training module, for using the corresponding principal component electrocardiosignal of parameter signal as the defeated of neural network using the training
Enter sample, using the reference breath signal that impedance method obtains as the training objective of the neural network training neural network；
Second breath signal obtains module, for inputting the corresponding principal component electrocardiosignal of the test parameter signal
In neural network after to the training, the second breath signal is obtained.
The present invention is obtained from electrocardiosignal respectively by adjusting sef-adapting filter and building neural network training model
First respiratory rate and the second respiratory rate, then by the first respiratory rate of weighted calculation and the second respiratory rate, it is more accurate to obtain
Respiratory rate.And dedicated sensor and hardware module detection breath signal are not needed, it is only necessary to be obtained with ECG monitor
Electrocardiosignal, avoid the constraint to human body, be suitable for daily monitoring, can real-time monitoring subject respiratory state.
Detailed description of the invention
Fig. 1 is the flow chart for the respiratory rate calculation method that first embodiment of the invention provides；
Fig. 2 is electrocardiosignal figure；
Fig. 3 is the filtered electrocardiosignal figure of power frequency；
Fig. 4 is the flow chart for the respiratory rate calculation method that second embodiment of the invention provides；
Fig. 5 is the flow chart of sef-adapting filter processing parameter signal；
Fig. 6 is the functional block diagram of sef-adapting filter；
Fig. 7 is the flow chart for the respiratory rate calculation method that third embodiment of the invention provides；
Fig. 8 is the structural block diagram for the respiratory rate computing device that fourth embodiment of the invention provides；
Fig. 9 is the structure of the first respiratory rate computing device in the respiratory rate computing device that fifth embodiment of the invention provides
Block diagram；
Figure 10 is the knot of the second respiratory rate computing device in the respiratory rate computing device that sixth embodiment of the invention provides
Structure block diagram.
Specific embodiment
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached
The embodiment of figure description is exemplary, and for explaining only the invention, and is not considered as limiting the invention.
Referring to following description and drawings, it will be clear that these and other aspects of the embodiment of the present invention.In these descriptions
In attached drawing, some particular implementations in the embodiment of the present invention are specifically disclosed, to indicate to implement implementation of the invention
Some modes of the principle of example, but it is to be understood that the scope of embodiments of the invention is not limited.On the contrary, of the invention
Embodiment includes all changes, modification and the equivalent fallen within the scope of the spirit and intension of attached claims.
Referring to Fig. 1, for the respiratory rate calculation method in bright first implementation of this law, including step S11~S14.
Step S11 obtains electrocardiosignal and is pre-processed, the extracting parameter signal from the electrocardiosignal, the ginseng
Number signal includes training parameter signal and test parameter signal.The present invention is obtained original by augmented unipolar limb lead
Electrocardiosignal, since original electro-cardiologic signals include a large amount of Hz noise, needs as shown in Fig. 2, being original electrocardiosignal figure
50Hz notch filter is carried out, filters out Hz noise, the filtered electrocardiosignal of power frequency is as shown in Figure 3.Then it is mentioned by threshold method
Take the filtered RR interval series signal of power frequency and RW amplitude sequence signal, the as described parameter signal.In the present embodiment, for
Parameter adjustment and training neural network of the parameter signal a part that the ECG signal processing of acquisition extracts for filter, separately
A part is for extracting final breath signal in the neural network after being input to sef-adapting filter adjusted and training.
Step S12 adjusts sef-adapting filter by training parameter signal, with the adaptive filter adjusted
Wave device is handled to obtain the first breath signal with parameter signal to the test, and calculates the according to first breath signal
One respiratory rate.
Above-mentioned steps obtain the degree of purity requirement for meeting breath signal by the adaptive adjustment of sef-adapting filter
The model of sef-adapting filter, and the sef-adapting filter by adjusting after demodulates the breath signal in electrocardiosignal
Come, then calculates the breath signal reconciled and come out and obtain respiratory rate, as the first respiratory rate.
Step S13 constructs neural network training model by training parameter signal, and passes through the neural network
Training pattern, with the second breath signal is obtained in parameter signal, calculates second according to second breath signal and exhales from the test
Suction rate.In the step, by constructing neural network training model, breath signal is obtained from electrocardiosignal, and calculate acquisition
Breath signal obtains respiratory rate, and the respiratory rate which obtains is the second respiratory rate.
Step S14 obtains final respiratory rate by the first respiratory rate described in weighted calculation and second respiratory rate.
In above-mentioned steps S14, pass through the Power Spectrum Distribution and interference for analyzing the first breath signal and second signal first
Power Spectrum Distribution obtains the weight factor μ 1 and μ 2 of reflection signal quality quality, calculation formula are as follows:
μ 1=(P1,1-P2,1)/P1,1
μ 2=(P1,2-P2,2)/P1,2
Wherein, P1,1For the Power Spectrum Distribution of the first breath signal, P1,2For the Power Spectrum Distribution of the second breath signal, P2,1
For the Power Spectrum Distribution of the interference of the first breath signal, P2,2For the Power Spectrum Distribution of the interference of the second breath signal；
Final respiratory rate is calculated further according to the weight factor μ 1 and μ 2 and first respiratory rate and the second respiratory rate,
Calculation formula is R=μ 1*R1+ μ 2*R2, wherein R1 is the first breathing respiratory rate, and R2 is the second respiratory rate.Wherein, respiratory rate R1
With the calculation method of R2 are as follows: find the wave crest or trough of the first breath signal and the second breath signal；Calculate the wave crest or
The period of trough is simultaneously scaled respiratory rate according to sampling rate conversion, and the first respiratory rate and the second respiratory rate can be obtained.
Further, the first respiratory rate or the second respiratory rate that can also will be unsatisfactory for quality requirement by the way that baseline threshold is arranged
It deletes, that is, when the first respiratory rate R1 is lower than baseline threshold, with the second respiratory rate R2 for final respiratory rate, when the second breathing
When rate R2 is lower than baseline threshold, with the first respiratory rate R1 for final respiratory rate, as the first respiratory rate R1 and the second respiratory rate R2
When greater than baseline threshold, two respiratory rates are merged, weighted sum obtains final respiratory rate.
In order to obtain accurate breath signal from electrocardiosignal and accurate respiratory rate be calculated, the present embodiment passes through
Sef-adapting filter and building neural network training model are adjusted, the first respiratory rate and second is obtained from electrocardiosignal respectively and exhales
Suction rate, then by the first respiratory rate of weighted calculation and the second respiratory rate, accurate respiratory rate can be obtained.It does not need dedicated
Sensor and hardware module detect breath signal, it is only necessary to ECG monitor obtain electrocardiosignal, avoid to human body
Constraint, be suitable for daily monitoring, can real-time monitoring subject respiratory state.
Referring to Fig. 4, a kind of respiratory rate calculation method provided for second embodiment of the invention.The present embodiment is real first
It is further to step S12 to be improved on the basis of applying example.It is improved in that step S12, passes through the trained ginseng
Number signal adjusts sef-adapting filter, is handled with parameter signal with the sef-adapting filter adjusted the test
Obtain the first breath signal, and according to the process that first breath signal calculates the first respiratory rate specifically include step S21~
S23。
Step S21 is handled with parameter signal the training breathing letter estimated by sef-adapting filter
Number.Further, in step S21, sef-adapting filter includes such as Fig. 5 with the process that parameter signal is handled to the training
Shown step S211~S213.
Step S211 carries out fuzzy reasoning with parameter signal to the training according to preset fuzzy rule, obtains described
First consequent of fuzzy rule.In step S211, the first consequent for obtaining the fuzzy rule specifically includes step a~c.
Step a calculates the fuzzy subset of each neuron node of the trained parameter signal in sef-adapting filter
Degree of membership.
In view of signal has certain delay during acquisition from generating electrode, training is carried out with parameter signal n (k)
Fuzzy reasoning is carried out after delay process again.Tapped delay line is accessed after n (k), artefact signal obtains r dimension by r-1 delay
Output vector X (k)=[x1(k),x2(k),...,xr(k)]T.R dimension output vector enters the first layer of sef-adapting filter,
Each neuron node of this layer is a subordinating degree function, and can choose Gaussian function is subordinating degree function, specific as follows:
The calculation formula of the degree of membership are as follows:
Wherein,
In above-mentioned formula,Indicate degree of membership of the input signal vector in j-th of neuron node, l=1,2 ..., r table
Show the number of input signal variable, j=1,2 ..., n indicate the number of subordinating degree function, μlj(xl) indicate that first of input signal becomes
Measure subordinating degree function of the x in j-th of neuron node, cljIndicate first of reference-input signal in the person in servitude of j-th of neuron node
The center of category degree function,Indicate the width of the subordinating degree function of j-th of neuron node.
Step b calculates the excitation density of every fuzzy rule according to the degree of membership.The excitation density calculation formula
Are as follows:
Wherein,For the excitation density of j-th strip fuzzy rule, cljIndicate first of input signal in j-th of neuron section
The center of the subordinating degree function of point,Indicate the width of the subordinating degree function of j-th of neuron node.
The excitation density is normalized in step c, obtains the first consequent of the fuzzy rule.Described
One consequent are as follows:
Wherein,For the excitation density of j-th strip fuzzy rule,For the first consequent of j-th strip fuzzy rule, n is mould
Paste the quantity of rule.
Step S212 carries out function expansion with parameter signal to the training by orthogonal basis function, obtains described fuzzy
Second consequent of rule.
The present embodiment carries out the function expansion of parameter signal, function chain neural network by function chain neural network (FLNN)
Orthogonal basis using Chebyshev's orthogonal polynomial ((ChebyshevOrthogonalPolynomials, COP):
Ch0(x)=1
Ch1(x)=x
Ch2(x)=2x2-1
...
Chm+1(x)=2xChm(x)-Chm-1(x)
The basic function T such as formula of FLNN:
Wherein, Chm(xl) be output vector in first of element m-th of Chebyshev's orthogonal polynomial, T be transposition grasp
Make, M is the number of the orthogonal basis function.In practical situations, function expansion can also be carried out using other methods.By letter
Low-dimensional, can be expanded to higher dimensional space by number extension, be realized non-linear.
FLNN exports the second consequent of the fuzzy rule:
Wherein, wjFor the second consequent of j-th strip fuzzy rule, α1j,α2j,...,αMjJoin for the consequent of j-th strip fuzzy rule
Manifold, M are the number of the orthogonal basis function.
Function chain neural network is applied in sef-adapting filter by the present embodiment, will be former defeated by one group of orthogonal basis function
Enter vector and carry out dimension extension, linear dimensions is extended to non-linear, obtains the second consequent of fuzzy rule, it is adaptive to enhance
The Nonlinear Processing ability of filter.
Step S213 determines the output signal of the sef-adapting filter, i.e., according to first consequent and the second consequent
The breath signal estimated.The formula of the signal of the sef-adapting filter output is determined according to first consequent and the second consequent
Are as follows:
Wherein, y is the signal of the adaptive sef-adapting filter output, OjFor the first consequent of j-th strip fuzzy rule,
wjFor the second consequent of j-th strip fuzzy rule.
Step S22, calculates the difference of the breath signal of the breath signal estimated and impedance method acquisition, and adjusts described adaptive
The parameter for answering filter, makes the difference in default range, the sef-adapting filter after being adjusted.In the step, resistance
The reference breath signal that anti-method obtains refers to the respiration information for measuring the variation of chest impedance by high-frequency constant current source to obtain.
The functional block diagram of sef-adapting filter is as shown in fig. 6, extract the RR interval series signal of pretreated electrocardiosignal
It is input in sef-adapting filter with RW amplitude sequence signal n (k).The parameter signal is handled by sef-adapting filter to obtain
The breath signal y (k) estimated.The difference of the breath signal q (k) obtained by computing impedance method and the breath signal estimated, i.e. q
(k)-y (k) obtains deviation e (k), and the deviation is the smaller the better, and deviation is smaller, indicates the breathing letter of sef-adapting filter output
It is number purer.Sef-adapting filter is constantly readjusted, and reaches deviation e (k) in default range, to meet breath signal
Degree of purity requirement, thus the sef-adapting filter after being adjusted.Breathing can be obtained by the impedance that impedance method detects human body
Signal adjusts the parameter of sef-adapting filter using the breath signal that impedance method obtains as the breath signal of reference.
Step S23 is handled to obtain to the test by adjusting the sef-adapting filter after parameter with parameter signal
First breath signal, and the first respiratory rate is calculated according to first breath signal.Sef-adapting filter pair after adjusting parameter
The process that the parameter signal is handled can refer to the process that sef-adapting filter handles training with parameter signal, i.e. step
S211~S213 and step S22~S23.
The present embodiment passes through adaptive using the breath signal that impedance method obtains as breath signal, sef-adapting filter is referred to
Adjustment makes the breath signal estimated approach with reference to breath signal, obtains the adaptive filter for the degree of purity requirement for meeting breath signal
The model of wave device.Sef-adapting filter by adjusting after demodulates the respiratory rate signal on electrocardiosignal come to get arriving
First breath signal, and calculate the first respiratory rate of the first breath signal.
The fuzzy reasoning of the present embodiment fusion function chain neural network and sef-adapting filter extracts breathing from electrocardio wave
Signal, fully considered electrocardio, breath signal it is non-linear, the characteristics such as randomness and non-stationary are reduced as far as useful exhale
The loss of signal is inhaled, the robustness of system is more preferable.And believed by the electrocardio that Fuzzy Nonlinear handles input adaptive filter
Number, the cross jamming problem being able to solve during signal extraction.It is by the first respiratory rate of weighted calculation and the second respiratory rate
Accurate respiratory rate can be obtained, realize the monitoring to subject's Respiratory behavior.
Referring to Fig. 7, for the respiratory rate calculation method in third embodiment of the invention.The present embodiment is in first embodiment
On the basis of, it is further to step S13 to be improved.It is improved in that step S13, passes through the training parameter signal
Construct neural network training model, and by the neural network training model from the test with obtaining second in parameter signal
Breath signal specifically includes step S31~S35 according to the process that second breath signal calculates the second respiratory rate.
Parameter signal progress significance test is obtained significant characteristics signal by step S31.
In another embodiment of the invention, described that significant characteristics are obtained to parameter signal progress significance test
The process of signal includes step S311~S312.
Step S311 is standardized the parameter signal, obtains standard cardioelectric signal.In order to avoid not same amount
The difference of guiding principle data is more advantageous to analysis using unified dimension after standardization.
Standardization formula are as follows:
Wherein, X'ijIt is the new data after standardization；Mj、SjRespectively indicate the arithmetic mean of instantaneous value and standard of initial data j column
(inclined) is poor, and n is sample size.
The standard cardioelectric signal is carried out variance analysis (Analysis of variance, abbreviation by step S312
ANOVA) and F examines (homogeneity test of variance), obtains significant characteristics signal.
Assuming that breath signal Y and each electrocardiosignal X of input meet Y=X β+ε, X and Y are done linearly based on this hypothesis
Recurrence regression analysis, available variance table, as shown in table 1.
1 the results of analysis of variance table of table
Wherein, SSR is match value and desired quadratic sum, and SSE is the quadratic sum of initial value and match value, SSTO be initial value with
Desired quadratic sum.According to variance analysis as a result, significant indexes F=MSR/MSE can be calculated.Setting conspicuousness refers to
Mark the threshold value Fmin of F.The n for choosing F > Fmin in input feature vector X ties up index, i.e. significant characteristics signal, as principal component analysis
Input.
Step S32 carries out dimensionality reduction to the significant characteristics signal by principal component analysis, obtains principal component electrocardio letter
Number.
Above-mentioned steps S31's and S32 is the process for carrying out to tie up processing to parameter signal, high, significant to obtain contribution rate
The high principal component electrocardiosignal of property.
Step S33 constructs neural network.The hidden layer of the neural network is arranged 1, and the number of nodes of hidden layer is true according to trial and error procedure
It is fixed.First hidden node number is arranged near 1+i/2, gradually increases Hidden nodes to 2*i+1, and is continued growing until not receiving
It holds back, obtains error performance curve, most suitable Hidden nodes are determined by analytical error performance curve, wherein i is input layer
Interstitial content, input layer number are also characterized the dimension of sample, that is, having several features just has several input nodes.The nerve net
The learning rate of network is determined as 0.02~0.2 by trial and error procedure.
In the present embodiment, the learning algorithm sampling column Wen Baige-Ma Kuaerte (Levenberg- of neural network
Marquart, LM) algorithm, algorithm fast convergence rate when weight is less, and convenient for using MatLab (matrix&
Laboratory, matrix labotstory) programming realization.Since the present invention passes through, ANOVA and F is examined and principal component analysis is to electrocardio
Signal carries out dimensionality reduction, and weight is less, is more suitable for the learning algorithm.
In other embodiments of the invention, the learning algorithm of neural network can also use Momentum BP Algorithm, that is, pass through
Factor of momentum carries out weighed value adjusting, the calculation formula of Momentum BP Algorithm weighed value adjusting are as follows:
W (k+1)=w (k)+Δ w (k+1) (2)
Wherein, α indicates that the learning rate of network, η indicate factor of momentum η ∈ (0,1),Indicate kth
The error partial differential of secondary back transfer, w (k) indicate the threshold value or weight of kth time back transfer, and E (k) indicates that kth time is reversed and passes
The sum of the deviations passed.
Weighed value adjusting amplitude in the algorithm next time depends on the adjustment effect of last weight, and adjustment amount is generally along same
One partial differential direction decreases or increases.When last adjustment amplitude is too big, then two formula opposite signs of front and back；When last adjustment amount
When smaller, two formula of front and back (1), (2) symbol is identical.The more general BP algorithm good in convergence effect of Momentum BP Algorithm, convergence time is short,
Extraction effect is more preferable.
Step S34 uses the corresponding principal component electrocardiosignal of parameter signal as the input sample of neural network using the training
This obtains neural network using the reference breath signal that impedance method obtains as the training objective of neural network training neural network
Training pattern.
The test is input to neural metwork training mould with the corresponding principal component electrocardiosignal of parameter signal by step S35
In type, the second breath signal is obtained, and the second respiratory rate is calculated according to second breath signal.
Parameter signal in the present invention is the RR interval series signal and RW amplitude sequence signal of electrocardiosignal, with the parameter
Input sample space of the signal as neural network.It is 12 leads that standard leads electrocardiosignal more, calculates separately the RR of each lead
After interval series signal and RW amplitude sequence signal, the characteristic value for needing to be input to neural network reaches 24, and each lead it
Between it is interrelated, cause the dimension of input sample larger and containing linearly related input item, therefore the present invention uses conspicuousness
It examines and obtains the significant characteristics signal that contribution rate is high, conspicuousness is high, then significant characteristics are believed using Principal Component Analysis
Number carry out dimensionality reduction, obtain principal component electrocardiosignal.Principal component analysis and significance analysis combine, and can effectively realize dimensionality reduction
Denoising, improves the convergence precision and extraction efficiency of neural network.
The present embodiment passes through the neural network training model accurately, quickly by building neural network training model
Breath signal is extracted from electrocardiosignal in ground.Accurate breathing can be obtained by the first respiratory rate of weighted calculation and the second respiratory rate
Rate realizes the monitoring to subject's Respiratory behavior.
Referring to Fig. 8, providing a kind of respiratory rate computing device for fourth embodiment of the invention, comprising: at electrocardiosignal
Manage module, the first respiratory rate computing device, the second respiratory rate computing device, weighted calculation device.
ECG's data compression module is for obtaining electrocardiosignal and being pre-processed, the extracting parameter from the electrocardiosignal
Signal, the parameter signal include training parameter signal and test parameter signal；
First respiratory rate computing device is used to extract the first breath signal from electrocardiosignal by sef-adapting filter, and
The first respiratory rate is calculated according to the first breath signal.
Second respiratory rate computing device is used to construct neural network training model by training parameter signal and lead to
The neural network training model is crossed from the test with obtaining the second breath signal in parameter signal, and is exhaled according to described second
It inhales signal and calculates the second respiratory rate.
Weighted calculation device, it is final for being obtained by the first respiratory rate described in weighted calculation and second respiratory rate
Respiratory rate.
The present embodiment is obtained from electrocardiosignal respectively by adjusting sef-adapting filter and building neural network training model
The first respiratory rate and the second respiratory rate are taken, then by the first respiratory rate of weighted calculation and the second respiratory rate, can be obtained compared with subject to
True respiratory rate.Dedicated sensor and hardware module detection breath signal are not needed, it is only necessary to be obtained with ECG monitor
Electrocardiosignal avoids the constraint to human body, be suitable for daily monitoring, can real-time monitoring subject respiratory state.
As shown in figure 9, a kind of improvement as fourth embodiment of the invention, in fifth embodiment of the invention, described first
Respiratory rate computing device specifically includes:
It estimates breath signal and obtains module, for being carried out to the training with parameter signal correspondence by sef-adapting filter
Handle the breath signal estimated；
Module is adjusted, calculates the breath signal estimated and the difference with reference to breath signal, and adjust the adaptive-filtering
The parameter of device, makes the difference in default range, the sef-adapting filter after being adjusted；
First breath signal obtains module, for the sef-adapting filter by adjusting after to the test parameter
Signal is handled to obtain the first breath signal；
First respiratory rate computational submodule, for calculating the first respiratory rate according to first breath signal.
In above-mentioned first respiratory rate computing device, the sef-adapting filter includes:
Fuzzy reasoning module obtains institute for carrying out fuzzy reasoning to the parameter signal according to preset fuzzy rule
State the first consequent of fuzzy rule；
Function expansion module obtains the mould for carrying out function expansion to the parameter signal by orthogonal basis function
Paste the second consequent of rule；
Signal determining module, for determining the defeated of the sef-adapting filter according to first consequent and the second consequent
Signal out.
In the present embodiment, the parameter for adjusting sef-adapting filter with parameter by training obtains meeting the pure of breath signal
The sef-adapting filter that cleanliness requires, and the sef-adapting filter by adjusting after extracts first from test parameter signal and exhales
Signal is inhaled, the first respiratory rate is calculated according to the first breath signal.
Referring to Fig. 10, a kind of improvement as fourth embodiment of the invention, in sixth embodiment of the invention, described second
Respiratory rate computing device specifically includes:
Dimension-reduction treatment module obtains principal component electrocardiosignal for the parameter signal to be carried out dimensionality reduction；
Training module, for using the corresponding principal component electrocardiosignal of parameter signal as the defeated of neural network using the training
Enter sample, using the reference breath signal that impedance method obtains as the training objective of the neural network training neural network；
Second breath signal obtains module, for inputting the corresponding principal component electrocardiosignal of the test parameter signal
In neural network after to the training, the second breath signal is obtained；
Second respiratory rate computational submodule, for calculating the second respiratory rate according to second breath signal.
Further, above-mentioned dimension-reduction treatment module includes:
Significance test module for carrying out significance test to the parameter signal, and exports significant characteristics signal；
Principal component analysis module obtains principal component electrocardiosignal for carrying out dimensionality reduction to the significant characteristics signal.
Further, above-mentioned significance test unit includes:
Standardization unit obtains standard cardioelectric signal for being standardized to the parameter signal；
The standard cardioelectric signal is carried out variance analysis and F is examined by significant characteristics signal extraction unit, is extracted significant
Property characteristic signal.
The present embodiment obtains the significant characteristics signal that contribution rate is high, conspicuousness is high using significance test, then uses
Principal Component Analysis carries out dimensionality reduction to significant characteristics signal, obtains principal component electrocardiosignal.By principal component analysis and significantly
Property the method that combines of analysis, dimensionality reduction denoising can be effectively realized, improve the convergence precision and extraction effect of neural network.With
Training uses the corresponding principal component electrocardiosignal of parameter signal as the input sample of neural network, the breathing letter obtained with impedance method
Training objective number as neural network, training neural network, constructs neural network training model.By test parameter signal pair
The principal component electrocardiosignal answered is input in neural network training model, can obtain the second breath signal, and then can calculate
Second respiratory rate.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms are not
Centainly refer to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be any
One or more embodiment or examples in can be combined in any suitable manner.
Although an embodiment of the present invention has been shown and described, it will be understood by those skilled in the art that: not
A variety of change, modification, replacement and modification can be carried out to these embodiments in the case where being detached from the principle of the present invention and objective, this
The range of invention is defined by the claims and their equivalents.
Claims (19)
1. a kind of respiratory rate calculation method characterized by comprising
It obtains electrocardiosignal and is pre-processed, the extracting parameter signal from the electrocardiosignal, the parameter signal includes instruction
Practice parameter signal and test parameter signal；
Sef-adapting filter is adjusted by training parameter signal, with the sef-adapting filter adjusted to the survey
Parameter signal on probation is handled to obtain the first breath signal, and calculates the first respiratory rate according to first breath signal；
Neural network training model is constructed by training parameter signal, by the neural network training model from described
Test calculates the second respiratory rate according to second breath signal with obtaining the second breath signal in parameter signal；
Final respiratory rate is obtained by the first respiratory rate described in weighted calculation and second respiratory rate.
2. the method as described in claim 1, which is characterized in that described to adjust adaptive filter by training parameter signal
The step of wave device includes:
The breath signal estimated is handled with parameter signal to the training by sef-adapting filter；
The difference for the breath signal that the breath signal and impedance method estimated described in calculating obtain, and adjust the sef-adapting filter
Parameter, make the difference in default range.
3. method according to claim 2, which is characterized in that the sef-adapting filter handles the parameter signal
The step of include:
Fuzzy reasoning is carried out to the parameter signal according to preset fuzzy rule, obtains the first consequent of the fuzzy rule；
Function expansion is carried out to the parameter signal by orthogonal basis function, obtains the second consequent of the fuzzy rule；
According to first consequent and the second consequent, the output signal of the sef-adapting filter is determined.
4. method as claimed in claim 3, which is characterized in that it is described according to preset fuzzy rule to the parameter signal into
Row fuzzy reasoning, the step of obtaining the first consequent of the fuzzy rule include:
The parameter signal is calculated in the degree of membership of the fuzzy subset of each neuron node of sef-adapting filter；
The excitation density of every fuzzy rule is calculated according to the degree of membership；
The excitation density is normalized, the first consequent of the fuzzy rule is obtained.
5. method as claimed in claim 4, which is characterized in that the calculation formula of the degree of membership are as follows:
Wherein,
In above-mentioned formula,Indicate degree of membership of the input signal vector in j-th of neuron node, l=1,2 ..., r indicate defeated
Enter the number of signal variable, j=1,2 ..., n indicate the number of subordinating degree function, μlj(xl) indicate that first of reference-input signal becomes
Measure xlIn the subordinating degree function of j-th of neuron node, cljIndicate first of reference-input signal in j-th neuron node
The center of subordinating degree function,Indicate the width of the subordinating degree function of j-th of neuron node.
6. method as claimed in claim 4, which is characterized in that the calculation formula of the excitation density are as follows:
Wherein,For the excitation density of j-th strip fuzzy rule, cljIndicate first of input signal in j-th neuron node
The center of subordinating degree function,Indicate the width of the subordinating degree function of j-th of neuron node.
7. method as claimed in claim 4, which is characterized in that the calculation formula of first consequent are as follows:
Wherein,For the excitation density of j-th strip fuzzy rule,For the first consequent of j-th strip fuzzy rule, n is fuzzy rule
Quantity then.
8. method as claimed in claim 3, which is characterized in that the orthogonal basis function are as follows:
The calculation formula of second consequent are as follows:
Wherein, Chm(xl) be first of element in output vector m-th of Chebyshev's orthogonal polynomial, T is that transposition operates, and M is
The number of the orthogonal basis function, wjFor the second consequent of j-th strip fuzzy rule, α1j,α2j,...,αMjFor j-th strip fuzzy rule
Consequent parameter set, M be the orthogonal basis function number.
9. method as claimed in claim 3, which is characterized in that it is described according to first consequent and the second consequent, determine institute
State sef-adapting filter output signal the step of include:
The signal of the sef-adapting filter output is determined according to the following formula:
Y is the signal of sef-adapting filter output, OjFor the first consequent of j-th strip fuzzy rule, wjFor the fuzzy rule of j-th strip
The second consequent then.
10. the method as described in claim 1, which is characterized in that described to construct nerve net by training parameter signal
It further include that the parameter signal is subjected to dimensionality reduction, obtains principal component electrocardiosignal before the step of network training pattern；
Described the step of constructing neural network training model by the training parameter signal, includes, with the training parameter
Input sample of the corresponding principal component electrocardiosignal of signal as neural network, using the reference breath signal that impedance method obtains as
The training objective of the neural network constructs neural network training model；
It is described by the neural network training model from the test with the second breath signal is obtained in parameter signal the step of
Including the test is input in the neural network training model with the corresponding principal component electrocardiosignal of parameter signal, is obtained
Take final breath signal.
11. method as claimed in claim 10, which is characterized in that the step of parameter signal is carried out dimensionality reduction packet
It includes:
Parameter signal progress significance test is obtained into significant characteristics signal；
Dimensionality reduction is carried out to the significant characteristics signal by principal component analysis, obtains principal component electrocardiosignal.
12. method as claimed in claim 11, which is characterized in that described to obtain parameter signal progress significance test
The step of significant characteristics signal includes:
The parameter signal is standardized, standard cardioelectric signal is obtained；
The standard cardioelectric signal is subjected to variance analysis and F is examined, obtains significant characteristics signal.
13. method as claimed in claim 10, which is characterized in that the neural network uses dynamic BP algorithm, the dynamic
BP algorithm carries out weighed value adjusting, the calculation formula of weighed value adjusting by factor of momentum are as follows:
W (k+1)=w (k)+Δ w (k+1)
Wherein, α indicates that the learning rate of network, η indicate factor of momentum,Indicate kth time back transfer
Error partial differential, w (k) indicate the threshold value or weight of kth time back transfer, and E (k) indicates the sum of the deviations of kth time back transfer.
14. method as claimed in claim 10, which is characterized in that the neural network determines the number of hidden nodes by trial and error procedure
Number, the trial and error procedure comprising steps of
First hidden node number is arranged near 1+i/2, gradually increases Hidden nodes to 2*i+1, forms error performance curve；
The quantity of hidden node is determined by analytical error performance curve, wherein i is input layer number.
15. the method as described in claim 1, which is characterized in that the acquisition electrocardiosignal is simultaneously pre-processed, from the heart
The step of extracting parameter signal, includes: in electric signal
It obtains electrocardiosignal and carries out power frequency filtering, obtain filtering signal；
The RR interval series signal and RW amplitude sequence signal in the filtering signal are extracted by threshold method.
16. the method as described in claim 1, which is characterized in that described by the first respiratory rate described in weighted calculation and described
Second respiratory rate obtains the step of final respiratory rate and includes:
Calculate separately the weight factor μ 1 and μ 2 of first breath signal and the second breath signal, calculation formula are as follows:
μ 1=(P1,1-P2,1)/P1,1
μ 2=(P1,2-P2,2)/P1,2
Final respiratory rate is calculated according to the weight factor μ 1 and μ 2 and first respiratory rate and the second respiratory rate, is calculated public
Formula is R=μ 1*R1+ μ 2*R2,
Wherein, P1,1For the Power Spectrum Distribution of the first breath signal, P1,2For the Power Spectrum Distribution of the second breath signal, P2,1It is
The Power Spectrum Distribution of the interference of one breath signal, P2,2For the Power Spectrum Distribution of the interference of the second breath signal, R1 is the first breathing
Respiratory rate, R2 are the second respiratory rate.
17. a kind of respiratory rate computing device characterized by comprising
ECG's data compression module obtains electrocardiosignal and is pre-processed, the extracting parameter signal from the electrocardiosignal, institute
Stating parameter signal includes training parameter signal and test parameter signal；
First respiratory rate computing device, for adjusting sef-adapting filter and by training parameter signal with adjusted
The sef-adapting filter handles the test with parameter signal to obtain the first breath signal, and exhales according to described first
It inhales signal and calculates the first respiratory rate；
Second respiratory rate computing device, for constructing neural network training model by the training parameter signal and passing through institute
Neural network training model is stated from the test with obtaining the second breath signal in parameter signal, and is believed according to second breathing
Number calculate the second respiratory rate；
Weighted calculation device, for obtaining final breathing by the first respiratory rate described in weighted calculation and second respiratory rate
Rate.
18. device as claimed in claim 17, which is characterized in that the first respiratory rate computing device includes:
It estimates breath signal and obtains module, for being handled with parameter signal correspondence by sef-adapting filter the training
The breath signal estimated；
Module is adjusted, for calculating the breath signal estimated and with reference to the difference of breath signal, and adjusts the adaptive-filtering
The parameter of device, makes the difference in default range, the sef-adapting filter after being adjusted.
19. device as claimed in claim 17, which is characterized in that the second respiratory rate computing device includes:
Dimension-reduction treatment module obtains principal component electrocardiosignal for the parameter signal to be carried out dimensionality reduction；
Training module, for using the corresponding principal component electrocardiosignal of parameter signal as the input sample of neural network using the training
This, using the reference breath signal that impedance method obtains as the training objective of the neural network training neural network；
Second breath signal obtains module, for the test to be input to instruction with the corresponding principal component electrocardiosignal of parameter signal
In the neural network after white silk, the second breath signal is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610976238.7A CN106539586B (en) | 2016-11-07 | 2016-11-07 | A kind of respiratory rate calculation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610976238.7A CN106539586B (en) | 2016-11-07 | 2016-11-07 | A kind of respiratory rate calculation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106539586A CN106539586A (en) | 2017-03-29 |
| CN106539586B true CN106539586B (en) | 2019-07-16 |
Family
ID=58394366
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610976238.7A Active CN106539586B (en) | 2016-11-07 | 2016-11-07 | A kind of respiratory rate calculation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106539586B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109464147A (en) * | 2019-01-09 | 2019-03-15 | 浙江强脑科技有限公司 | Breath signal detection method, device and computer readable storage medium |
| CN114027822B (en) * | 2021-04-19 | 2022-11-25 | 北京超思电子技术有限责任公司 | Respiration rate measuring method and device based on PPG signal |
| CN113769217B (en) * | 2021-09-17 | 2023-07-07 | 北华大学 | Respiratory system tidal volume control method, device and equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101528126A (en) * | 2006-12-21 | 2009-09-09 | 弗雷森纽斯医疗护理德国有限责任公司 | Method and device for the determination of breath frequency |
| CN104398252A (en) * | 2014-11-05 | 2015-03-11 | 深圳先进技术研究院 | Electrocardiogram signal processing method and device |
| CN104720808A (en) * | 2013-12-20 | 2015-06-24 | 中国移动通信集团公司 | Human sleep respiration detection method and device |
| CN105722457A (en) * | 2013-11-20 | 2016-06-29 | 通用电气公司 | Method and system for determining respiration rate |
| CN105769122A (en) * | 2016-03-16 | 2016-07-20 | 中国科学院电子学研究所 | Method for measuring sleep apnea hyponea index |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110066008A1 (en) * | 2009-09-14 | 2011-03-17 | Matt Banet | Body-worn monitor for measuring respiration rate |
-
2016
- 2016-11-07 CN CN201610976238.7A patent/CN106539586B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101528126A (en) * | 2006-12-21 | 2009-09-09 | 弗雷森纽斯医疗护理德国有限责任公司 | Method and device for the determination of breath frequency |
| CN105722457A (en) * | 2013-11-20 | 2016-06-29 | 通用电气公司 | Method and system for determining respiration rate |
| CN104720808A (en) * | 2013-12-20 | 2015-06-24 | 中国移动通信集团公司 | Human sleep respiration detection method and device |
| CN104398252A (en) * | 2014-11-05 | 2015-03-11 | 深圳先进技术研究院 | Electrocardiogram signal processing method and device |
| CN105769122A (en) * | 2016-03-16 | 2016-07-20 | 中国科学院电子学研究所 | Method for measuring sleep apnea hyponea index |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106539586A (en) | 2017-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106108889B (en) | Electrocardiogram classification method based on deep learning algorithm | |
| Sarkaleh et al. | Classification of ECG arrhythmias using discrete wavelet transform and neural networks | |
| CN108388912B (en) | Sleep staging method based on multi-sensor feature optimization algorithm | |
| CN106388825B (en) | A kind of respiratory rate extracting method and device | |
| CN106539586B (en) | A kind of respiratory rate calculation method and device | |
| CN108416367B (en) | Sleep staging method based on multi-sensor data decision-level fusion | |
| Sankur et al. | Comparison of AR-based algorithms for respiratory sounds classification | |
| CN106580324B (en) | A kind of breath signal extracting method and device | |
| CN108498089A (en) | A kind of noninvasive continuous BP measurement method based on deep neural network | |
| Leung et al. | Classification of heart sounds using time-frequency method and artificial neural networks | |
| CN107736894A (en) | A kind of electrocardiosignal Emotion identification method based on deep learning | |
| CN116705337B (en) | Health data acquisition and intelligent analysis method | |
| CN109745033A (en) | Dynamic electrocardiogram method for evaluating quality based on time-frequency two-dimensional image and machine learning | |
| CN110327036A (en) | The method of breath signal and respiratory rate is extracted from wearable ECG | |
| CN107422841A (en) | A kind of man-machine interaction method based on contactless Emotion identification | |
| CN106510716B (en) | A kind of method and device calculating respiratory rate based on electrocardiosignal | |
| CN111887858A (en) | Ballistocardiogram signal heart rate estimation method based on cross-modal mapping | |
| CN108681685A (en) | A kind of body work intension recognizing method based on human body surface myoelectric signal | |
| Dar et al. | Spectral features and optimal hierarchical attention networks for pulmonary abnormality detection from the respiratory sound signals | |
| CN106901694A (en) | A kind of respiratory rate extracting method and device | |
| CN114403847B (en) | Respiration state detection method and system based on correlation of abdominal and lung data | |
| CN109934179B (en) | Human body action recognition method based on automatic feature selection and integrated learning algorithm | |
| CN106388824B (en) | A kind of respiratory rate extracting method and device | |
| CN111973188A (en) | Method for estimating respiratory mechanics parameter based on neural network | |
| CN108665066A (en) | Portable respirator flow curve scaling method based on BP neural network |
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 |