CN105361885A - Respiratory waveform identification method and system - Google Patents

Respiratory waveform identification method and system Download PDF

Info

Publication number
CN105361885A
CN105361885A CN201410424270.5A CN201410424270A CN105361885A CN 105361885 A CN105361885 A CN 105361885A CN 201410424270 A CN201410424270 A CN 201410424270A CN 105361885 A CN105361885 A CN 105361885A
Authority
CN
China
Prior art keywords
waveform
respiratory
sample point
air flow
amplitude
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
Application number
CN201410424270.5A
Other languages
Chinese (zh)
Other versions
CN105361885B (en
Inventor
林康平
林耿弘
詹皓羽
黄圣澄
施博锺
尤景良
李达龙
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wellell Inc
Original Assignee
Apex Medical Corp
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Apex Medical Corp filed Critical Apex Medical Corp
Priority to CN201410424270.5A priority Critical patent/CN105361885B/en
Publication of CN105361885A publication Critical patent/CN105361885A/en
Application granted granted Critical
Publication of CN105361885B publication Critical patent/CN105361885B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

The invention discloses a respiratory waveform identification method. The method includes the steps of a, detecting a respiratory air flow in a respiratory cycle; b, measuring the amplitude of the respiratory air flow and the time of the respiratory cycle; c, by multiple sampling points, determining an inhaling waveform and an exhaling waveform according to the amplitude and the time; d, normalizing the amplitude and the time of one of the inhaling waveform and the exhaling waveform to build a normalized waveform; accumulating a difference value between a reference waveform and the normalized waveform so as to calculate a flow coefficient which determines that the normalized waveform can be identified as a normal respiration state and an abnormal respiration state. The method has the advantages that a curve (or a weighted curve or a standard waveform) is used to fit the inhaling waveform or the exhaling waveform so as to calculate the difference value, and the normal respiration state and the abnormal respiration state are identified by accumulating the difference value. The invention further provides a respiratory waveform identification system.

Description

Respiratory waveform discrimination method and system thereof
Technical field
The present invention relates to breathing state identification field, particularly by respiratory waveform discrimination method at least partially and the system thereof of at least one of them matching respiratory air flow of curve, weighted curve and reference waveform.
Background technology
Usually, in the process that patient breathes voluntarily, when obtaining required oxygen concentration (calling abnormal breathing state in the following text), then this patient can maintain the respiratory movement of needed by human body by respirator, to supply this oxygen concentration.
This abnormal breathing state normally occurs in the process of this patient sleeps, usually this abnormal breathing state can divide into again transient respiratory apnea (apnea), shallow breathing (hypopnea) or shallow slow breathing (flowlimitation) further, and DIYU hereafter describes in detail respectively.
The definition of this transient respiratory apnea is in this patient sleeps's process, because Apnea causes pulmonary to exceed several tens of seconds without air circulation, and causes the phenomenon that in this blood samples of patients, blood oxygen concentration reduces.Suffer from this patient of this Sleep Apnea disease, sleep quality can be affected because of sleep apnea, and have the hidden danger such as life is abnormal, initiation cardiovascular disease, heart disease or cerebral hemorrhage.
This shallow breathing is defined as the situation that adult causes gas flow to reduce because generation breathing shoals in sleep procedure.Except aforesaid situation, this shallow breathing also likely simultaneously concurrent Oxygen saturation decline (oxygendesaturation), and reach the situation of of short duration awakening (arousal) of more than tens of second.For example, in sleep procedure, if the respiratory movement of the throughput of this adult and thorax abdomen reduces by 50% ~ 70% than normal situation, and the Oxygen saturation of this adult's blood at least reduces 4%, then meet the definition of this shallow breathing.
This shallow slow breathing is defined as this patient and is partly obstructed because of respiratory tract, causes the throughput of air-flow lower than normal flow.In detail, in sleep procedure, due to factors such as inharmonious, the pharyngeal soft tissue bulky of upper respiratory tract muscle group, tonsil hypertrophy or hypertrophies of tonsils, cause this patient in sleep procedure, situation upper respiratory tract partial blockage occurring repeatedly or all blocks, and this situation may allow this patient awaken momently.
Shallow breathing or shallow slow breathing can be determined whether by Apnea-shallow breathing index (apnea-hypopneaindex, AHI).Wherein, the summation number of times of generation Apnea per hour and shallow breathing in this Apnea-shallow breathing exponential representation sleep.If AHI is more than or equal to 15 times/hour or AHI is more than or equal to 5 times/hour, then represent that this patient drowsy, (or claim snoring) (snore) that snore may occur by day, witnesses to breathe and interrupt, and because of short durationly choking/the suck-back symptom of awakening without a break.
It should be noted that above snore (or claiming this snoring) is defined as this air-flow through narrow upper respiratory tract, produce a kind of symptom of the sound because vibrations occur for the soft tissue of throat, soft palate or uvula.
Aforementioned mentioned respirator solves the most effective mode of abnormal breathing state at present.This patient by this respirator gas without interruption, to reach the effect of supplying oxygen concentration.
The kind of respirator divides into positive pressure type respirator and negative-pressure type respirator further according to the form of pressure supply.
Illustrate for positive pressure type respirator.Positive pressure type respirator, by detecting the respiratory air flow of this patient, assists this patient to carry out complete respiratory movement to judge to be in due course.
The detection method of respiratory air flow can divide into embedded type detection technique and non-embedded type detection technique.Embedded type detection technique is that detecting instrument implant into body is inner, such as cardiac ventricles, pulmonary or larynx etc.; Non-embedded type detection technique is the respiratory air flow directly utilizing positive pressure respiration machine testing patient.The detection technique of the latter, compared to the former detection technique, owing to not needing implant into body inner, therefore can detect the respiratory air flow of patient rapidly and easily, be unlikely and allow this patient produce uncomfortable sensation.
But, though non-embedded type detection technique can detect patient easily, but traditional respirator still existing defects, namely traditional respirator also cannot distinguish the order of severity in abnormal breathing state effectively and accurately, this problem will cause this respirator in the process to patient's persistent pressure, likely suppress the breathing of this patient, and cause the defect of more serious respiratory disorder.
Therefore, how to allow respirator accurately identify this order of severity and suitably can export applicable gas pressure, be important and need the problem of solution badly.
In view of this, the present invention proposes a kind of respiratory waveform discrimination method and its system, can accurately identify eupnea state and improper breathing state in a respiratory air flow.
Summary of the invention
The first object of the present invention is to provide a kind of respiratory waveform discrimination method, by amplitude and the time of a regular respiratory air flow in real time, allow the respiratory air flow after normalization can calculate accurate discharge coefficient by an algorithm, and from this discharge coefficient, identify that this respiratory air flow belongs to eupnea state and improper breathing state, wherein this improper breathing state comprises transient respiratory apnea (apnea), shallow breathing (hypopnea) or shallow slow breathing (flowlimitation).
The second object of the present invention is according to above-mentioned respiratory waveform discrimination method, extract the inspiratory waveform in this respiratory air flow or an expiration waveform, to reduce the time of this algorithm calculation, and this respiratory waveform discrimination method can by the state of single inspiratory waveform or this respiratory air flow of expiration waveform recognition.
The third object of the present invention is according to above-mentioned respiratory waveform discrimination method, utilize curve-fitting method, utilize first order linear curve, multi order linear curve, weighted curve and reference waveform matching this inspiratory waveform normalized maybe this expiration waveform, to judge the order of severity of this improper breathing state.
The fourth object of the present invention is according to above-mentioned respiratory waveform discrimination method, in fit procedure, redistributes for sample point specific in this respiratory air flow, to judge the order of severity of this improper breathing state according to weight.
The fifth object of the present invention is according to above-mentioned respiratory waveform discrimination method, select and normalized this inspiratory waveform waveform that maybe this expiration waveform is closest in multiple reference waveform, by calculating the error of two waveforms, to judge the order of severity of this improper breathing state.
The sixth object of the present invention is according to above-mentioned respiratory waveform discrimination method, select a part of waveform of this inspiratory waveform normalized maybe this expiration waveform, and calculate the difference (standard deviation of such as absolute error) of one of them of this portion waveshape and these reference waveforms, to judge the order of severity of this improper breathing state.
The seventh object of the present invention is to provide a kind of respiratory waveform identification system, by detecting a respiratory air flow, and from this respiratory air flow, identifies eupnea state and improper breathing state.
Described respiratory waveform identification system has at least one pneumatic sensor, a processing unit and a display unit.Pneumatic sensor can carry out all work relating to detecting step herein.Processing unit is in order to carry out all measurements, sampling, decision waveform, cumulative, matching, normalization, calculated flow rate coefficient herein, to determine any step relating to analytic method of breathing state, setting flow threshold values, comparison, judgement degree of congestion etc.Display unit is in order to show breathing state.
For reaching above-mentioned object and other object, the invention provides a kind of respiratory waveform discrimination method, for identifying eupnea state and improper breathing state in a respiratory air flow, this respiratory waveform discrimination method comprises (a) and detects this respiratory air flow in a breathing cycle; B () is measured the amplitude of this respiratory air flow and is measured time of this breathing cycle, wherein the length of this time is at least enough to rising waveform or falling waveform that this respiratory air flow detected; C () multiple sample point is according to this amplitude and this Time dependent one inspiratory waveform and an expiration waveform; D this amplitude of one of them of () this inspiratory waveform of normalization and this expiration waveform and this time, to set up regular waveform; And (e) cumulative difference between reference waveform and this regular waveform, to calculate discharge coefficient, this discharge coefficient determines that this regular waveform is identified as this eupnea state and this improper breathing state.
For reaching above-mentioned object and other object, the invention provides a kind of respiratory waveform discrimination method, for identifying eupnea state and improper breathing state in a respiratory air flow, this respiratory waveform discrimination method comprises (a1) and detects a respiratory air flow in this breathing cycle; (b1) measure the amplitude of this respiratory air flow and measure time of this breathing cycle, wherein the length of this time is at least enough to rising waveform or falling waveform that this respiratory air flow detected; (c1) multiple sample point is according to this amplitude and this Time dependent one inspiratory waveform and an expiration waveform; (d1) this inspiratory waveform of normalization and this amplitude of one of them of this expiration waveform and this time, to set up regular waveform; (e1) this regular waveform of a linear equation matching is utilized; (f1) difference between this linear equation and this regular waveform is calculated by weighted least squares, and this difference calculates discharge coefficient by a weighting function again, this discharge coefficient determines that this regular waveform is identified as this eupnea state and this improper breathing state.
For reaching above-mentioned object and other object, the invention provides a kind of respiratory waveform discrimination method, for identifying eupnea state and improper breathing state in a respiratory air flow, this respiratory waveform discrimination method comprises (a2) and detects this respiratory air flow in a breathing cycle; (b2) measure the amplitude of this respiratory air flow and measure time of this breathing cycle, wherein the length of this time is at least enough to rising waveform or falling waveform that this respiratory air flow detected; (c2) multiple sample point according to this amplitude and this time to determine an inspiratory waveform and an expiration waveform; (d2) this inspiratory waveform of normalization and this amplitude of one of them of this expiration waveform and this time, to set up regular waveform; (e2) calculate multiple reference waveform and this regular waveform, so that one of them of these reference waveforms is set as reference waveform, and this reference waveform in these reference waveforms closest to this regular waveform; (f2) cumulative difference between reference waveform and this regular waveform, to calculate discharge coefficient, this discharge coefficient determines that this regular waveform is identified as this eupnea state and this improper breathing state.
For reaching above-mentioned object and other object, the invention provides a kind of respiratory waveform identification system, for identifying eupnea state and improper breathing state in a respiratory air flow, this respiratory waveform identification system comprises pneumatic sensor, processing unit and display unit.This pneumatic sensor is the changes in flow rate detecting this respiratory air flow within the time.Wherein, the length of this time is at least enough to rising waveform or falling waveform that this respiratory air flow detected.This processing unit is connected to this pneumatic sensor.This processing unit receives the changes in flow rate signal of this respiratory air flow, and by the amplitude of this respiratory air flow of Algorithm Analysis and the time of measuring this breathing cycle to calculate discharge coefficient.According to this discharge coefficient, this algorithm determines that this respiratory air flow belongs to this eupnea state or this improper breathing state.This display unit is connected to this processing unit.This display unit shows this eupnea state or this improper breathing state.Wherein, this algorithm samples this amplitude and this time by multiple sample point, this amplitude and this Time dependent one inspiratory waveform and an expiration waveform, and this amplitude of one of them of this algorithm this inspiratory waveform of normalization and this expiration waveform and this time, to set up regular waveform.This algorithm by calculating the difference between reference waveform and this regular waveform, or calculates the difference between a linear equation and this regular waveform, and obtains this discharge coefficient.
Generally speaking, respiratory waveform discrimination method of the present invention and system thereof, utilize the waveform of multiple curve fitting one respiratory air flow, and calculate difference therebetween.By this mathematic interpolation flow coefficient, this discharge coefficient determines that this respiratory air flow belongs to this eupnea state or this improper breathing state.This discharge coefficient, except can determining different breathing states, more can distinguish the order of severity of this improper breathing state further, such as transient respiratory apnea, shallow breathing or shallow slow breathing.
It should be noted that, in this shallow breathing or this shallow slow breathing, method of the present invention and system thereof, also can calculate out such as AHI index etc. according to this discharge coefficient further, to judge to belong to obstructive sleep apnea symptom or the symptom of snore (or claiming snoring).
Compared to prior art, the invention solves respirator in prior art and effectively cannot judge the order of severity of improper breathing state, and cause this respirator in the process to patient's persistent pressure, the breathing of this patient may be suppressed and cause the defect of more serious respiratory disorder.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of the respiratory waveform discrimination method of the embodiment of the present invention.
Fig. 2 (a) is the oscillogram of eupnea state.
Fig. 2 (b) is the oscillogram of improper breathing state.
Fig. 3 is the oscillogram of waveform after amplitude normalization with time normalization of the improper breathing state of key diagram 2 (b).
Fig. 4 is the oscillogram of reference waveform S (i).
Fig. 5 is the cell schematics of the respiratory waveform identification system of the embodiment of the present invention.
Reference numeral:
10 respiratory waveform identification systems
12 pneumatic sensors
14 processing units
16 display units
BS respiratory air flow
A amplitude
T time
FC discharge coefficient
NBS eupnea state
The improper breathing state of ANBS
Detailed description of the invention
For fully understanding the object of the present invention, feature and effect, by following specific embodiment, and graphic appended by coordinating, the present invention is described in detail, illustrates as rear:
Fig. 1 is the schematic flow sheet of the respiratory waveform discrimination method of the embodiment of the present invention.As shown in fig. 1, this respiratory waveform discrimination method in the waveform of a respiratory air flow, can identify eupnea state and improper breathing state.
Fig. 2 (a) is the oscillogram of eupnea state.Fig. 2 (a) is according to patient in aspiratory action and exhale action process, the oscillogram of gas flow change in respiratory tract; Wherein, x-axis direction express time T and y-axis direction represent amplitude A.In the waveform of time 0 to t1, this waveform can represent the variations in flow of aspiratory action.In the waveform of time t1 to t2, this waveform can represent the variations in flow of exhale action.In this eupnea state, the variations in flow of this aspiratory action is substantially equal to the variations in flow of this exhale action.
Fig. 2 (b) is the oscillogram of non-eupnea state.In this improper breathing state, because of the impact of being blocked in various degree in this respiratory tract, the variations in flow of this aspiratory action is caused to be not equal to the variations in flow of this exhale action.
Get back to Fig. 1, the step of this respiratory waveform discrimination method originates in step S11, detects this respiratory air flow in a breathing cycle.For Fig. 2 (b), this breathing cycle has been defined as this aspiratory action and this exhale action time once, and namely in figure, this time 0 to t2 represents a cycle.
Step S12 is the amplitude measuring this respiratory air flow, and measures the time of this breathing cycle.In Fig. 2 (b), the scope of this amplitude between A1 to-A1, and this time be 0 to t2.
Step S13 is that multiple sample point determines an inspiratory waveform (between time 0 to t1) and an expiration waveform (between time t1 to t2) according to this amplitude A and this time T.For example, by the sampling mode of low frequency (approximate number Hz), in this amplitude A to the coordinate axes of time T producing corresponding sample point, to depict this waveform.
Step S14 is this amplitude A of one of them and this time T of this inspiratory waveform of normalization and this expiration waveform, to set up regular waveform.This regular waveform in the lump with reference to figure 3, can which show and extract this inspiratory waveform.In figure 3, because the amplitude of this inspiratory waveform is through normalized calculating, make the size of this amplitude A be changed into 1 by original A1 by regular, and the length of this time T is changed into t0 by original this time t1 by regular.
Step S15 is the difference (such as this difference is least absolute error value and/or least squares error value) between cumulative reference waveform and this regular waveform, to calculate discharge coefficient, this discharge coefficient determines that this regular waveform can be identified as this eupnea state or this improper breathing state.This reference waveform can comprise several forms, is presented below respectively.
A) order-1 linear equation formula
This reference waveform meets order-1 linear equation formula a 1i+a 2, the Mathematical representation of this discharge coefficient is wherein, C1 (i) is this order-1 linear equation formula, a 1with a 2for constant, i is each sample point, and x is the starting point of each sample point, and y is the terminating point of each sample point, and F (i) is this amplitude A of this regular waveform at sample point i.
B) Second Order Linear Differential Equation formula
This reference waveform meets Second Order Linear Differential Equation formula a 1i 2+ a 2i+a 3, the Mathematical representation of this discharge coefficient is wherein C2 (i) is this Second Order Linear Differential Equation formula, a 1, a 2with a 3for constant, i is each sample point, and x is the starting point of each sample point, and y is the terminating point of each sample point, and F (i) is this amplitude A of this regular waveform at sample point i.
Because this Second Order Linear Differential Equation formula is curve, compared to order-1 linear equation formula, the fitting effect of this Second Order Linear Differential Equation formula is better.
C) three rank linear equations
This reference waveform meets three rank linear equation a 1i 3+ a 2i 2+ a 3i+a 4, the Mathematical representation of this discharge coefficient is wherein C3 (i) is this three rank linear equation, a 1, a 2, a 3with a 4for constant, i is each sample point, and x is the starting point of each sample point, and y is the terminating point of each sample point, and F (i) is this amplitude A of this regular waveform at sample point i.
Compared to the fitting effect of Second Order Linear Differential Equation formula, the advantage that this three rank linear equation can provide this Second Order Linear Differential Equation formula effectively to offset about judgment curves further.
D) three rank weighted linear equations
This reference waveform meets three rank weighted linear equation a 1i 3+ a 2i 2+ a 3i+a 4, the Mathematical representation of this discharge coefficient is wherein W (i) is defined as follows:
Wherein, C4 (i) is this three rank linear equation, a 1, a 2, a 3with a 4for constant, i is each sample point, and x is the starting point of each sample point, and y is the terminating point of each sample point, and W (i) is weighting function, and A, B and C are weighted value, and F (i) is this amplitude A of this regular waveform at sample point i.
Compared to the fitting effect of three rank linear equations, this three rank weighted linear equation can be weighted for the numerical value of such as starting point x or terminating point y further, with to this three rank weighted linear equation depict curve endpoint value adjust, with reduce this endpoint value cause error in calculating.
In the present embodiment, this three rank weighted linear equation only lists this starting point and this terminating point is that example illustrates, but this starting point be in fact not limited in the present embodiment or this terminating point.In other words, the sample position of weighting, except this starting point aforesaid or this terminating point, more can increase other specific sample point, or replaced by other sample point.
For example, in another embodiment, the scope of this weighted value A can be set between 50 to 100, the range set of this weighted value B between 200 to 400 and this weighted value C be 1.
E) reference waveform
This reference waveform meets one of them of multiple reference waveform S (i), in the lump can with reference to figure 4, and it is the oscillogram of reference waveform S (i).In the diagram, reference waveform S (i) can be exemplified as waveform S (i) under three eupnea states 1, S (i) 2with S (i) 3.The Mathematical representation of this discharge coefficient is:
Σ i = x y ( Ki - μ ) 2 L ; Wherein
Ki = Σ i = x y R × | F ( i ) - S ( i ) | ;
μ = 1 L ΣKi ;
R = ΣS ( i ) ΣF ( i ) ;
Wherein, i is each sample point, x is the starting point of each sample point, y is the terminating point of each sample point, L is the time span between this starting point and this terminating point, and Ki is the absolute error of one of them of this regular waveform and these reference waveforms, and μ is the meansigma methods of this absolute error, R is this regular waveform and the area under the curve of one of them ratio of reference waveform, and F (i) is this amplitude A of this regular waveform at sample point i.
For example, in another embodiment, this starting point is set as 20% of sample point, and this terminating point is set as 80% of sample point.
At this reference waveform above-mentioned a) in embodiment e), discharge coefficient can set multiple flow threshold values further.By these flow threshold values of this discharge coefficient comparison, the degree of congestion of respiratory tract can be judged from this discharge coefficient, such as this degree of congestion, according to this respiratory air flow of this respiratory tract, can be divided into again respiratory smooth, slight respiratory obstruction, severe respiratory obstruction further or snore.
Though it should be noted that this above-mentioned reference waveform is exemplified as order-1 linear equation formula to three rank linear equation; But in fact, this reference waveform easily extensible of the present invention is to the embodiment higher than three rank linear equations (under be called multi order linear equation).In the equational embodiment of multi order linear, this regular waveform of the curve waveform matching formed by this multi order linear equation, and calculate the difference between this two waveform, to obtain this discharge coefficient.This discharge coefficient can determine that this regular waveform is identified as this eupnea state and this improper breathing state.Similarly, also a weighting function can be imported in addition in this multi order linear equation, to reduce the error of calculation of endpoint value.
It should be noted that to judge more accurately, this discharge coefficient that this multi order linear equation calculates out also by comparison these flow threshold values aforesaid, to judge this degree of congestion of this respiratory tract further.
Fig. 5 is the cell schematics of the respiratory waveform identification system of the embodiment of the present invention.In Figure 5, this respiratory waveform identification system 10 can identify eupnea N state BS and improper breathing state ANBS in respiratory air flow BS.
This respiratory waveform identification system comprises pneumatic sensor 12, processing unit 14 and display unit 16.
Within the time, this pneumatic sensor 12 detects the changes in flow rate of this respiratory air flow BS.The length of this time is at least enough to rising waveform or falling waveform that this respiratory air flow BS detected.
This processing unit 14 connects this pneumatic sensor 12.This processing unit 14 receives the changes in flow rate signal of this respiratory air flow BS, and measures the time T of this breathing cycle by the amplitude A of this respiratory air flow of Algorithm Analysis BS, and then calculates this discharge coefficient FC.Moreover according to this discharge coefficient FC, this algorithm determines that this respiratory air flow belongs to this eupnea N state BS or this improper breathing state ANBS.
For example, this algorithm samples this amplitude A and this time T by multiple sample point.This amplitude A and this time T determine an inspiratory waveform and an expiration waveform.This amplitude A of one of them of this algorithm this inspiratory waveform of normalization and this expiration waveform and this time T, to set up regular waveform.This algorithm passes through the difference of calculating one reference waveform and this regular waveform, or calculates the difference between a linear equation and this regular waveform, to draw this discharge coefficient FC.
In another embodiment, this reference waveform is selected from one of them of multiple reference waveform, and by this reference waveform of selecting closest to this regular waveform.
This display unit 16 connects this processing unit 14.This display unit 16 shows this eupnea N state BS or this improper breathing state ANBS.
In an other embodiment, this respiratory waveform identification system 10 also comprises storage element (not shown).This storage element connects this processing unit 12, to store this discharge coefficient FC.
The present invention is open with preferred embodiment hereinbefore, but has the knack of the technology person and it should be understood that this embodiment only for describing the present invention, and should not be read as and limit the scope of the invention.It should be noted, all changes with this embodiment equivalence and displacement, all should be covered by category of the present invention.Therefore, protection scope of the present invention is when being as the criterion with defining of claims.

Claims (17)

1. a respiratory waveform discrimination method, can identify eupnea state and improper breathing state in the waveform of respiratory air flow, this respiratory waveform discrimination method comprises:
Described respiratory air flow is detected within the breathing cycle;
Measure the amplitude of described respiratory air flow and measure time of described breathing cycle, wherein the length of this time is at least enough to rising waveform or falling waveform that described respiratory air flow detected;
Multiple sample point is according to described amplitude and described Time dependent one inspiratory waveform and an expiration waveform;
The described amplitude of one of them of the described inspiratory waveform of normalization and expiration waveform and described time, to set up regular waveform; And
Cumulative difference between reference waveform and described regular waveform, to calculate discharge coefficient, this discharge coefficient determines that described regular waveform is identified as eupnea state and improper breathing state.
2. respiratory waveform discrimination method according to claim 1, wherein said reference waveform meets order-1 linear equation formula, and the Mathematical representation of this order-1 linear equation formula C1 (i) is:
A 1i+a 2; And
The Mathematical representation of described discharge coefficient is:
Σ i = x y | C 1 ( i ) - F ( i ) | ;
Wherein a 1with a 2for constant, i is described sample point, and x is the starting point of described sample point, and y is the terminating point of described sample point, and F (i) is the described amplitude of the described regular waveform at sample point i.
3. respiratory waveform discrimination method according to claim 1, wherein said reference waveform meets Second Order Linear Differential Equation formula, and the Mathematical representation of this Second Order Linear Differential Equation formula C2 (i) is:
A 1i 2+ a 2i+a 3; And
The Mathematical representation of described discharge coefficient is:
Σ i = x y | C 2 ( i ) - F ( i ) | ;
Wherein a 1, a 2with a 3for constant, i is described sample point, and x is the starting point of described sample point, and y is the terminating point of described sample point, and F (i) is the described amplitude of the described regular waveform at sample point i.
4. respiratory waveform discrimination method according to claim 1, wherein said reference waveform meets three rank linear equations, and the Mathematical representation on these three rank linear equation C3 (i) is:
A 1i 3+ a 2i 2+ a 3i+a 4; And
The Mathematical representation of described discharge coefficient is:
Σ i = x y | C 3 ( i ) - F ( i ) | ;
Wherein a 1, a 2, a 3with a 4for constant, i is described sample point, and x is the starting point of described sample point, and y is the terminating point of described sample point, and F (i) is the described amplitude of the described regular waveform at sample point.
5. respiratory waveform discrimination method according to claim 1, wherein said reference waveform meets three rank weighted linear equations, and the Mathematical representation on these three rank weighted linear equation C4 (i) is:
A 1i 3+ a 2i 2+ a 3i+a 4; And
The Mathematical representation of described discharge coefficient is:
Σ i = x y W ( i ) × | C 4 ( i ) - F ( i ) | ; Wherein
Wherein a 1, a 2, a 3with a 4for constant, i is described sample point, and x is the starting point of described sample point, y is the terminating point of described sample point, W (i) is weighting function, and A, B and C are weighted value, and F (i) is the described amplitude of the described regular waveform at sample point i.
6. respiratory waveform discrimination method according to claim 5, the range set of wherein said weighted value A between 50 to 100, the range set of described weighted value B between 200 to 400 and described weighted value C be 1.
7. the respiratory waveform discrimination method according to any one of claim 2 to 5, also comprises:
Set multiple flow threshold values; And
With flow threshold values described in described discharge coefficient comparison, to judge the degree of congestion of respiratory tract from described discharge coefficient.
8. respiratory waveform discrimination method according to claim 7, wherein according to the described respiratory air flow of described respiratory tract, described degree of congestion is divided into respiratory smooth, slight respiratory obstruction, severe respiratory obstruction or is snored by described flow threshold values.
9. respiratory waveform discrimination method according to claim 1, wherein said reference waveform meets one of them of multiple reference waveform S (i), and the Mathematical representation of described discharge coefficient is:
Σ i = x y ( Ki - μ ) 2 L ; Wherein
Ki = Σ i = x y R × | F ( i ) - S ( i ) | ;
μ = 1 L ΣKi ;
R = ΣS ( i ) ΣF ( i ) ;
Wherein i is described sample point, x is the starting point of described sample point, y is the terminating point of described sample point, L is the time span between described starting point and described terminating point, Ki is the absolute error of one of them of described regular waveform and described reference waveform, μ is the meansigma methods of described absolute error, and R is described regular waveform and the area under the curve of one of them ratio of described reference waveform, and F (i) is the described amplitude amount of the described regular waveform at sample point i.
10. respiratory waveform discrimination method according to claim 9, is wherein set as 20% of described sample point by described starting point, and described terminating point is set as 80% of described sample point.
11. 1 kinds of respiratory waveform discrimination methods, for identifying eupnea state and improper breathing state in respiratory air flow, this respiratory waveform discrimination method comprises:
Respiratory air flow is detected in a breathing cycle;
Measure the amplitude of described respiratory air flow and measure time of described breathing cycle, wherein the length of this time is at least enough to rising waveform or falling waveform that described respiratory air flow detected;
Multiple sample point is according to described amplitude and described Time dependent one inspiratory waveform and an expiration waveform;
The described amplitude of one of them of the described inspiratory waveform of normalization and described expiration waveform and described time, to set up regular waveform;
Utilize regular waveform described in a linear equation matching; And
The difference between described linear equation and described regular waveform is calculated by weighted least squares, and this difference calculates discharge coefficient according to a weighting function, this discharge coefficient determines that described regular waveform is identified as eupnea state and improper breathing state.
12. 1 kinds of respiratory waveform discrimination methods, for identifying eupnea state and improper breathing state in respiratory air flow, this respiratory waveform discrimination method comprises:
Described respiratory air flow is detected in a breathing cycle;
Measure the amplitude of described respiratory air flow and measure time of described breathing cycle, wherein the length of this time is at least enough to rising waveform or falling waveform that described respiratory air flow detected;
Multiple sample point is according to described amplitude and described Time dependent one inspiratory waveform and an expiration waveform;
The described amplitude of one of them of the described inspiratory waveform of normalization and described expiration waveform and described time, to set up regular waveform;
Calculate multiple reference waveform and described regular waveform, so that one of them of described reference waveform is set as reference waveform, wherein this reference waveform among described reference waveform closest to described regular waveform; And
Cumulative difference between reference waveform and described regular waveform, to calculate discharge coefficient, this discharge coefficient determines that described regular waveform is identified as eupnea state and improper breathing state.
13. respiratory waveform discrimination methods according to claim 1,11 or 12, wherein said difference be least absolute error value and least squares error value at least one of them.
14. respiratory waveform discrimination methods according to claim 1,11 or 12, wherein said improper breathing state is transient respiratory apnea, shallow breathing or shallow slow breathing.
15. 1 kinds of respiratory waveform identification systems, for identifying eupnea state and improper breathing state in respiratory air flow, this respiratory waveform identification system comprises:
Pneumatic sensor, it detects the changes in flow rate of described respiratory air flow within the time, and wherein the length of this time is at least enough to rising waveform or falling waveform that described respiratory air flow detected;
Processing unit, it is connected to described pneumatic sensor, described processing unit receives the changes in flow rate signal of described respiratory air flow, and by the amplitude of respiratory air flow described in an Algorithm Analysis and the time of measuring the described breathing cycle to calculate discharge coefficient, according to this discharge coefficient, described algorithm determines that described respiratory air flow belongs to eupnea state or improper breathing state; And
Display unit, it is connected to described processing unit, and described display unit shows described eupnea state or described improper breathing state;
Wherein said algorithm samples described amplitude and described time by multiple sample point, described amplitude and described Time dependent one inspiratory waveform and an expiration waveform, and the described amplitude of one of them of the described algorithm described inspiratory waveform of normalization and described expiration waveform and described time, to set up regular waveform, described algorithm is by the difference between calculating one reference waveform and described regular waveform, or the difference calculated between a linear equation and described regular waveform, to obtain described discharge coefficient.
16. respiratory waveform identification systems according to claim 15, also comprise storage element, it is connected to described processing unit, and described storage element stores described discharge coefficient.
17. respiratory waveform identification systems according to claim 15, wherein said reference waveform is selected from one of them of multiple reference waveform, and described reference waveform in described reference waveform closest to described regular waveform.
CN201410424270.5A 2014-08-26 2014-08-26 Respiratory waveform identification system Active CN105361885B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410424270.5A CN105361885B (en) 2014-08-26 2014-08-26 Respiratory waveform identification system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410424270.5A CN105361885B (en) 2014-08-26 2014-08-26 Respiratory waveform identification system

Publications (2)

Publication Number Publication Date
CN105361885A true CN105361885A (en) 2016-03-02
CN105361885B CN105361885B (en) 2018-12-11

Family

ID=55364874

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410424270.5A Active CN105361885B (en) 2014-08-26 2014-08-26 Respiratory waveform identification system

Country Status (1)

Country Link
CN (1) CN105361885B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108186018A (en) * 2017-11-23 2018-06-22 苏州朗开信通信息技术有限公司 A kind of breath data processing method and processing device
CN110237374A (en) * 2019-05-22 2019-09-17 东南大学 A kind of device and its application method generating given inspiratory waveform
CN113419569A (en) * 2020-06-18 2021-09-21 湖北工业大学 Frequency-adjustable cigarette smoke flow control method and device
CN114376559A (en) * 2022-01-18 2022-04-22 高昌生医股份有限公司 Respiration baseline tracking acceleration method

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07222732A (en) * 1994-02-10 1995-08-22 Tadamichi Meguro Overall evaluating method and device of respiratory tract hypersensitivity and respiratory tract obturation
CN101394885A (en) * 2006-03-06 2009-03-25 雷斯梅德有限公司 Method and apparatus for improved flow limitation detection of obstructive sleep apnea
CN102481127A (en) * 2009-08-13 2012-05-30 帝人制药株式会社 Device for calculating respiratory waveform information and medical device using respiratory waveform information
CN103169476A (en) * 2013-03-14 2013-06-26 中山大学 Method and device for identification and early warning of respiration wave form image
WO2013138071A1 (en) * 2012-03-15 2013-09-19 Board Of Trustees Of Michigan State University Food intake monitoring system using apnea detection in breathing signals
CN103961105A (en) * 2014-05-22 2014-08-06 深圳市帝迈生物技术有限公司 Method and system for performing snore recognition and strength output and breathing machine

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07222732A (en) * 1994-02-10 1995-08-22 Tadamichi Meguro Overall evaluating method and device of respiratory tract hypersensitivity and respiratory tract obturation
CN101394885A (en) * 2006-03-06 2009-03-25 雷斯梅德有限公司 Method and apparatus for improved flow limitation detection of obstructive sleep apnea
CN102481127A (en) * 2009-08-13 2012-05-30 帝人制药株式会社 Device for calculating respiratory waveform information and medical device using respiratory waveform information
WO2013138071A1 (en) * 2012-03-15 2013-09-19 Board Of Trustees Of Michigan State University Food intake monitoring system using apnea detection in breathing signals
CN103169476A (en) * 2013-03-14 2013-06-26 中山大学 Method and device for identification and early warning of respiration wave form image
CN103961105A (en) * 2014-05-22 2014-08-06 深圳市帝迈生物技术有限公司 Method and system for performing snore recognition and strength output and breathing machine

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108186018A (en) * 2017-11-23 2018-06-22 苏州朗开信通信息技术有限公司 A kind of breath data processing method and processing device
CN110237374A (en) * 2019-05-22 2019-09-17 东南大学 A kind of device and its application method generating given inspiratory waveform
CN110237374B (en) * 2019-05-22 2022-03-08 东南大学 Device for generating given inspiratory waveform and using method thereof
CN113419569A (en) * 2020-06-18 2021-09-21 湖北工业大学 Frequency-adjustable cigarette smoke flow control method and device
CN114376559A (en) * 2022-01-18 2022-04-22 高昌生医股份有限公司 Respiration baseline tracking acceleration method
CN114376559B (en) * 2022-01-18 2023-09-19 高昌生医股份有限公司 Respiratory datum line tracking acceleration method

Also Published As

Publication number Publication date
CN105361885B (en) 2018-12-11

Similar Documents

Publication Publication Date Title
US10278639B2 (en) Method and system for sleep detection
JP6574648B2 (en) Automatic control for flow restriction detection
JP6564318B2 (en) A system for evaluating real-time lung mechanics
US7276031B2 (en) System and method for classifying patient's breathing using artificial neural network
AU2009233596B2 (en) Method and Device for Carrying Out a Signal-Processing Viewing of a Measurement Signal that is Correlated to the Respiratory Activity of an Individual
EP3821930B1 (en) Methods and devices with leak detection
AU683550B2 (en) Apparatus for detecting and correcting airflow limitation
CN103445781B (en) device for judging breathing state
US20060249149A1 (en) Method and arrangement for the tiration of physiological measuring signals in conjunction with the observation of a patient in terms of sleep-related respiratory problems
CN103961105B (en) Snoring is identified and the method and system of intensity output and a kind of respirator
EP3062682B1 (en) Apparatus and method for detecting health deterioration
JP5593480B2 (en) Apnea / hypopnea determination device
CN105361885A (en) Respiratory waveform identification method and system
US20160081631A1 (en) Respiratory waveform recognition method and system
TW201544074A (en) Breathing waveform recognition method and system thereof
EP2989979B1 (en) Respiratory waveform recognition method and system
EP3200686B1 (en) Device for assessing respiratory data in a monitored subject
Johnson et al. Flow limitation/obstruction with recovery breath (FLOW) event for improved scoring of mild obstructive sleep apnea without electroencephalography
Staykov et al. Nasal Pressure Derived Airflow Limitation and Ventilation Measurements are Resilient to Reduced Signal Quality
AU2012225239A1 (en) Apparatus for quantifying respiratory and inspiratory airflow
JP2007260127A (en) Respiration measuring instrument, and sleeping condition discrimination system
AU2014218399B2 (en) Respiratory Waveform Recognition Method and System
Perez-Macias et al. Time characteristics of prolonged partial obstruction periods using an Emfit mattress
Montserrat et al. Effect of using the flow or the volume signals on the measurement of nonapneic respiratory events
WO2022133942A1 (en) Medical ventilation device and ventilation monitoring method

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant