CN100508885C - Fault-tolerant method and device in respiratory mechanics monitoring system - Google Patents
Fault-tolerant method and device in respiratory mechanics monitoring system Download PDFInfo
- Publication number
- CN100508885C CN100508885C CNB200610062226XA CN200610062226A CN100508885C CN 100508885 C CN100508885 C CN 100508885C CN B200610062226X A CNB200610062226X A CN B200610062226XA CN 200610062226 A CN200610062226 A CN 200610062226A CN 100508885 C CN100508885 C CN 100508885C
- Authority
- CN
- China
- Prior art keywords
- value
- parameter
- breathing
- phase
- respiratory waveform
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The present invention discloses a fault tolerance method and a device in the respiratory mechanics monitoring system, which is used for excluding the abnormal parameters in the derived parameters after the calculation step of the parameters; the present invention can temporarily save the calculated derived parameters of N consecutive respiratory cycles in the manner of first-in and first-out, and compare the parameter value P0 of the current cycle with the former parameter value Psurp plus 1 which has the non-abnormal fluctuation, if the absolute value of P0 minus Psurp plus 1 is less than the absolute value of one third of Psurp plus 1, then the parameter value P0 of the current cycle is replaced by the calculated arithmetic mean value of the derived parameter of the recent N cycles; if the absolute value of P0 minus Psurp plus 1 is more than the absolute value of one third of Psurp plus 1, then the parameter value P0 of the current cycle is replaced by the calculated average value of the remaining points in the recent N adjacent points which remove all the abnormal fluctuation points, and the number of the consecutive abnormal fluctuation points plus 1; when the number of the consecutive abnormal fluctuation points is more than 3, the parameter value P0 of the current cycle is replaced by the arithmetic mean value of the consecutive N derived parameter points. The present invention excludes the adverse effects on the feedback caused by signal abnormal sudden changes through the selective average of the derived parameters, so as to ensure the correctness of derived parameters.
Description
[technical field]
The present invention relates to a kind of medical respiratory mechanics monitoring system, relate in particular to method and apparatus fault-tolerant in a kind of respiratory mechanics monitoring system.
[background technology]
In medical monitoring, breathing mechanics module can directly be measured the flow of human body respiration air-flow and the pressure in the gas circuit, and depict the real-time respiratory waveform of two basic parameters, utilize these two periodic Changing Patterns of parameter to calculate the derived parameter of 18 clinical uses.Breathing mechanics module obtains after these parameters the numerical value that mode with serial communication sends these parameters and removes to control anesthetic machine, respirator or monitor.Because the respiratory mechanics monitor parameter has directly been controlled feedback effect to the ventilation of anesthetic machine, respirator, if these parameters are wrong, may cause anesthetic machine, respirator or monitor malfunction, thereby cause the severe disorder of ventilating, even jeopardize patient's life.So in respiratory mechanics monitoring system, the phase identification of respiratory waveform cycle and inspiratory phase and expiratory phase is particularly important, because breathing cycle and phase identification mistake will output error derived parameter, for example when inspiratory phase, export the parameter of non-inspiratory phase, or the parameter of output inspiratory phase when inspiratory phase not, these all can cause anesthetic machine, respirator or the control of monitor mistake.
What breathing cycle in the existing respiratory mechanics monitoring system and phase identification were used usually is great-leap-forward method for waveform identification at zero point, and this is a kind of reliable method for waveform identification that rises recently.This recognition methods will be divided into two phases 1 and 2 and critical points 3 breathing cycle.By the judgement of flow number being changed the numerical value of breathing phases marking variable, thereby judge the different phase (as shown in Figure 1) of breathing cycle and even one-period.Illustrate the method for this waveform recognition below: the cur_point_sign that is designated that establishes the present flow rate sampled point, the positive and negative pre_point_sign that is designated of previous sampled point, when being changed to 1, the sign of sampled point represents that present flow rate for just, represents that when the sign of sampled point is changed to-1 present flow rate is negative; Stage of living in breathing cycle is wave_counter, and wherein wave_counter equals " 1 " expression and breathing cycle is in inspiratory phase; Wave_counter equals " 2 " and shows that the breathing cycle is in expiratory phase; Wave_counter equals " 3 " expression breathing cycle end point, and it is unidentified to respiratory waveform that wave_counter equals " 0 " expression.Then the flow process of breathing identification as shown in Figure 2.
Flow value according to real-time identifies the breathing cycle.If the value of flow, is then thought the flow symbol greater than going beyond value zero point for just, cur_point_sign puts 1; If the value of flow, is then thought the flow symbol less than going beyond value minus zero point for negative, cur_point_sign puts-1.The key point symbol equals the product of current point symbol and last point symbol, if negative then waveform enumerator adds 1; The symbol of current point identification cur_point_sign is paid the symbol of last point identification pre_point_sign.Because " 1 " point should be air-breathing starting point,, it is put 0 so equal 1 and flow (illustrates that breathing is in expiration phase) when going beyond value less than the negative zero point when the waveform enumerator.Like this, wave_counter=1 when the breathing cycle is in inspiratory phase; Wave_counter=2 when being in expiratory phase; The wave_counter=3 when breathing cycle finishes, the inspiratory phase that enters following one-period simultaneously.When wave_counter=0, the mechanics of breathing detection system is not exported respiration parameter, and when wave_counter=1, output is the parameter of inspiratory phase, when wave_counter=2, output be the parameter of expiratory phase.
Present respiratory mechanics monitor mostly adopts this breathing recognition methods, but can there be various interference in actual applications, for example mechanical noise interference, valve performance interference, null offset and paroxysmal abnormality breathing etc. all can make respiratory waveform unusual, cause the respiratory waveform recognition failures.
Mechanical noise comprises that promptly high frequency (comparing with respiratory frequency, the down together) oscillator signal that the physical impacts of extraneous flow sensor is introduced also comprises the high frequency spurious signal that the outside noise interference in air flow forms.
Valve performance disturbs and appears at usually under the mechanical ventilation situation of anesthetic machine and respirator.Since inlet valve be not closed completely (investigation find most inlet valves on the market light or heavy have this feature), when expiratory phase changed, some high frequency oscillation ripples (as shown in Figure 3) can appear to mandatory ventilation in flow waveform usually from inspiratory phase.Observe in the experiment and find that the amplitude of these Sassers is more considerable, can reach 3～5LPM usually.This has introduced sizable risk to waveform recognition.
Null offset is the problem that common differential amplifier circuit is difficult to avoid.Do not have fault tolerant mechanism if zero point is fluid, waveform recognition mechanism can the wrong drift such be judged as the breathing cycle.
In addition, because patient's cough or sneeze in the sudden change that does not have also can cause under the situation of fault tolerant mechanism monitoring parameter, cause the calculation of parameter mistake, even the waveform recognition mistake.
But this zero point, the fault tolerant mechanism of great-leap-forward method for waveform identification was undesirable, can not effectively solve these interference problems.
In order to ensure normally using, therefore set up a series of fault tolerant mechanisms at this recognition methods and suitable clinical demand, original method is improved quite urgent.
[summary of the invention]
Main purpose of the present invention is exactly in order to address the above problem, and a kind of fault-tolerant method and apparatus that is used for respiratory mechanics monitoring system is provided, and eliminates unusual derived parameter, thereby avoids causing the mistake control of anesthetic machine, respirator or monitor.
Secondary objective of the present invention is exactly in order to address the above problem, a kind of fault-tolerant method and apparatus that is used for respiratory mechanics monitoring system is provided, eliminate the high frequency breath signal composition that valve closing is not sternly introduced, get rid of unusual inspiratory phase and expiratory phase, improve the accuracy rate of identification of breathing cycle, inspiratory phase and expiratory phase, guarantee the correctness of derived parameter, thereby avoid causing the mistake control of anesthetic machine, respirator or monitor.
Another purpose of the present invention is exactly in order to address the above problem, a kind of fault-tolerant method and apparatus that is used for respiratory mechanics monitoring system is provided, the error that elimination is caused because of mechanical noise interference, valve performance interference, null offset and paroxysmal abnormality breathing etc., further improve the accuracy rate of identification of breathing cycle, inspiratory phase and expiratory phase, guarantee the correctness of derived parameter, thereby avoid causing the mistake control of anesthetic machine, respirator or monitor.
For achieving the above object, the present invention proposes the fault-tolerance approach in a kind of respiratory mechanics monitoring system, may further comprise the steps:
The respiratory waveform identification step: according to the end point of inspiratory phase, expiratory phase and the breathing cycle in respiratory waveform identification of breathing cycle, and the breathing phases sign is set to represent the respective identification of inspiratory phase, expiratory phase or breathing cycle end point;
Calculation of parameter step: the inspiratory phase and the parameter in expiratory phase stage of calculating the breathing cycle;
Unusual parameter is got rid of step: the unusual parameter in the parameter that derives after the calculation of parameter step is got rid of; Described unusual parameter is got rid of step and be may further comprise the steps:
A1, with the temporary N that calculates of the mode of FIFO the derived parameter of breathing cycle, wherein N continuously=3;
The parameter value P0 of B1, comparison current period and the parameter value P of last non-unusual fluctuations
Surp+1If, | P0-P
Surp+1|＜| 1/3P
Surp+1|, then the parameter value P0 of current period is a normal point, the arithmetic mean of instantaneous value of this derived parameter in nearest N cycle calculating is replaced the parameter value P0 of current period; If | P0-P
Surp+1| ≧ | 1/3P
Surp+1|, execution in step D1 then;
D1, then the parameter value P0 of current period is the unusual fluctuations points, the parameter value P0 that the meansigma methods of removing all the other points behind all unusual fluctuations points in nearest N the consecutive points that calculate is replaced current period, and continuous unusual fluctuations are counted add 1, wherein surp is that continuous unusual fluctuations are counted, then execution in step E1;
E1, whether judge continuous unusual fluctuations number of spots, if then the arithmetic mean of instantaneous value of these successive N derived parameter points is replaced the parameter value P0 of current period greater than 3.
N in the wherein said steps A 1 is preferably greater than or equals 3, is less than or equal to 8.
Further improvement of the present invention is: also be included in the phase place determining step of carrying out in the breathing cycle identifying, described phase place determining step may further comprise the steps:
The duration of A2, detection inspiratory phase;
B2, the duration and the predefined air-breathing ultimate value of inspiratory phase compared, if the duration of inspiratory phase is more than or equal to air-breathing ultimate value, the inspiratory phase sign that then is provided with in the respiratory waveform identification step is constant, if the duration of inspiratory phase is less than air-breathing ultimate value, execution in step C2 then;
C2, this waveform is thought disturbing wave, and the inspiratory phase that is provided with in respiratory waveform identification step sign forces to be set to unidentified to the respiratory waveform sign;
The duration of D2, detection expiratory phase;
E2, the duration and the predefined expiration ultimate value of expiratory phase compared, if the duration of expiratory phase is more than or equal to the expiration ultimate value, the breathing phases sign that then is provided with in the respiratory waveform identification step is constant, if the duration of expiratory phase is less than the expiration ultimate value, execution in step F2 then;
The expiratory phase sign that is provided with in F2, the respiratory waveform identification step forces to be set to the sign of inspiratory phase.
Further comprising the steps of after step C2: when respiratory waveform reenters next inspiratory phase, with this inspiratory phase in the stage parameters calculated cover all parameters of calculating in the described disturbing wave duration.
Described phase place determining step can further include following steps:
G2, when the breathing cycle finishes, judge whether the ratio of inspiratory phase duration and expiratory phase duration exceeds predefined breathing ratio range, if exceed, all parameters of calculating in then should the breathing cycle are cancelled.
Of the present invention further the improvement is: described respiratory waveform identification step may further comprise the steps:
The flow value of A3, detection respiratory air flow;
B3, when flow value during greater than positive threshold value, put the present flow rate sampled point be designated on the occasion of, when flow value during, put the present flow rate sampled point and be designated negative value less than minus threshold value, when flow value is between the positive negative threshold value, keep present flow rate sampled point sign constant;
C3, judge whether the present flow rate sampled point sign and the product of previous traffic sampling point identification are negative value, when described product on the occasion of the time then finish; When described product is negative value, execution in step D3 then;
D3, breathing phases sign is promoted according to the order of inspiratory phase sign, expiratory phase sign, breathing cycle end point sign, and breathing phases be designated the breathing cycle during end point sign breathing phases sign be set to the inspiratory phase sign.
Of the present invention further the improvement is: in described respiratory waveform identification step, according to time sequencing sample three numerical value x1, x2, the x3 of real-time respiratory waveform, judge the residing stage of respiratory waveform according to the comparative result of three numerical value:
As x1〉x2〉during x3, this respiratory waveform is in trailing edge;
When x1＜x2＜x3, this respiratory waveform is in rising edge;
As x2〉x1 and x2〉during x3, this respiratory waveform is in peak value;
When x2＜x1 and x2＜x3, this respiratory waveform is in valley.
For achieving the above object, the present invention also proposes the fault-tolerant device in a kind of respiratory mechanics monitoring system, comprise: the respiratory waveform identification module, be used for end point, and the breathing phases sign is set to represent the respective identification of inspiratory phase, expiratory phase or breathing cycle end point according to inspiratory phase, expiratory phase and the breathing cycle in respiratory waveform identification of breathing cycle; Parameter calculating module is used to calculate the parameter and the output parameter of respiratory waveform respective stage; The unusual parameter of getting rid of with the output that is used for responding respiratory waveform identification module and parameter calculating module, with the unusual parameter of derived parameter is got rid of module, described unusual parameter is got rid of module and is comprised: pushup storage, be used for the temporary N that calculates the derived parameter of breathing cycle, wherein a N continuously 〉=3; First comparing unit is used for the parameter value P0 of current period and the parameter value P of last non-unusual fluctuations
Surp+1Compare; First processing unit is used for | P0-P
Surp+1|＜| 1/3P
Surp+1| the time, the arithmetic mean of instantaneous value of this derived parameter in nearest N cycle calculating is replaced the parameter value P0 of current period; | P0-P
Surp+1| ≧ | 1/3P
Surp+1| the time, the meansigma methods of removing all the other points behind all unusual fluctuations points in nearest N the consecutive points that calculate is replaced the parameter value P0 of current period, and continuous unusual fluctuations are counted add 1, wherein surp is that continuous unusual fluctuations are counted; Whether second processing unit is used to judge continuous unusual fluctuations number of spots greater than 3, if then the arithmetic mean of instantaneous value of these successive N derived parameter points is replaced the parameter value P0 of current period.
Further improvement of the present invention is: also comprise the phase place judge module, described phase place judge module comprises: the time detecting unit is used to detect the duration of inspiratory phase and expiratory phase; The time judging unit is used for the duration of inspiratory phase and expiratory phase is compared with predefined air-breathing ultimate value and expiration ultimate value respectively; The 3rd processing unit, be used for setting according to the judged result control breathing phase identification of time judging unit, during less than air-breathing ultimate value, this waveform is thought disturbing wave, and the breathing phases sign is set at the duration of inspiratory phase unidentified to the respiratory waveform sign; During more than or equal to air-breathing ultimate value, then the control breathing phase identification is constant at the duration of inspiratory phase; During less than the expiration ultimate value, then the breathing phases sign forces to be set to the inspiratory phase sign at the duration of expiratory phase; During more than or equal to the expiration ultimate value, then the control breathing phase identification is constant at the duration of expiratory phase; Breathe the ratio judging unit, be used for when the breathing cycle finishes, judge whether the ratio of inspiratory phase duration and expiratory phase duration exceeds predefined breathing ratio range, if exceed, then all parameters that should calculate in the breathing cycle are cancelled; Described parameter calculating module also responds the breathing phases sign of respiratory waveform identification module output, when respiratory waveform reenters next inspiratory phase, with this inspiratory phase in the stage parameters calculated cover all parameters of calculating in the described disturbing wave duration.
Of the present invention further the improvement is: also comprise the waveform locking module, described waveform locking module comprises: second sampling unit is used for according to time sequencing sample three numerical value x1, x2, the x3 of real-time respiratory waveform; Second comparing unit is used for the size of three numerical value of comparison; The waveform judging unit is used for judging the residing stage of respiratory waveform according to the comparative result of three numerical value: as x1〉x2〉during x3, this respiratory waveform is in trailing edge; When x1＜x2＜x3, this respiratory waveform is in rising edge; As x2〉x1 and x2〉during x3, this respiratory waveform is in peak value; When x2＜x1 and x2＜x3, this respiratory waveform is in valley.
The invention has the beneficial effects as follows: 1) average by derived parameter being carried out selectivity, got rid of the adverse effect that the abnormal signal sudden change causes feedback, guarantee the correctness of derived parameter.2) in the process of respiratory waveform identification, measuring range specification according to the respiratory mechanics monitor module, go out the minimum time of inspiratory phase and expiratory phase in conjunction with the physiology of respiration feature calculation, on this basis interference waveform is filtered, and further regulation sucks the ultimate value of exhalation ratio, interference waveform is filtered, effectively got rid of the interference waveform that does not sternly cause because of valve closing, avoided the parameter monitoring Problem-Error that sternly do not cause because of valve closing, improve the identification of breathing cycle, the accuracy rate of inspiratory phase and expiratory phase, guarantee the correctness of derived parameter, thereby eliminate the false judgment that these unusual waveforms cause parameter monitoring and lead to errors feedback or error diagnosis are avoided patient's health even life security are threatened.3) in zero point great-leap-forward respiratory waveform identifying, by setting positive negative threshold value and locking waveform, effectively eliminate faint interfering signal or null offset by the problem that respiratory waveform is discerned of being used as of mistake, further improved the accuracy rate of identification of breathing cycle, inspiratory phase and expiratory phase.
Feature of the present invention and advantage will be elaborated in conjunction with the accompanying drawings by embodiment.
[description of drawings]
Fig. 1 is the respiratory waveform diagram;
Fig. 2 be in the prior art zero point great-leap-forward waveform recognition process chart;
The flow waveform noise diagram that Fig. 3 introduces for inlet valve is not closed completely;
Fig. 4 is the respiratory mechanics monitoring system block diagram;
Fig. 5 is the unusual parameter exclusive method flow chart of the another kind of embodiment of the present invention;
Fig. 6 is the time period fault tolerant approach process chart of an embodiment of the present invention;
Fig. 7 is the method for waveform identification sketch map of establishing positive negative threshold value of the another kind of embodiment of the present invention;
Fig. 8 is the waveform recognition flow chart of establishing positive negative threshold value of the another kind of embodiment of the present invention;
Fig. 9 is the fault-tolerant device block diagram of an embodiment of the present invention.
[specific embodiment]
Specific embodiment one, as shown in Figure 4, the key component of respiratory mechanics monitoring system comprises sensor device part, analog amplify circuit part, AD collecting part, Wave data analysis and data processing section and waveform and parameter display part, if the respiratory mechanics monitor module is used for anesthetic machine or respirator, the parameter after the processing also is used for the feedback to ventilation control usually.Method for waveform identification of the present invention and device all occur in Wave data analysis and data processing section.
Present embodiment adopts unusual parameter exclusive method to carry out fault-tolerant processing by the parameter to the output of mechanics of breathing detection system, with the correctness of the cycle parameter that ensures respiration.
Its particular flow sheet as shown in Figure 5.
The feedback that makes the mistake for fear of fluctuating widely of derived parameter and to control mechanism, after all derived parameter collections or calculating, these data are admitted to a data treatment mechanism and carry out unusual exclusive method fault-tolerant processing.This mechanism has been opened up length in memorizer be RAM (random access memory) memory space of 13 (13 derived parameters) x8 (maximum 8 continuous sampling points) x32bits (data length), the continuous sampling that is used to handle derived parameter is counted and can be selected, but must be 3-8, minimum 3 is to be resumed when occurring 3 continuously because of abnormity point to be normal point, and maximum 8 is an empirical value.This mechanism is with derived parameter P0～Pn-1 of principle N the breathing cycle of record of FIFO (FIFO), and N is preferably 3-8.Come the number of times of records abnormal fluctuation point with variable Surp.Elder generation is with this parameter value P of parameter current value P0 and last non-unusual fluctuations
Surp+1Compare, for example during Surp=2, the P3 of parameter current value P0 and its front relatively because P1, P2 are the unusual fluctuations points.This parameter value P in parameter current value P0 and last non-unusual fluctuations
Surp+1The absolute value of difference less than | 1/3P
Surp+1| situation under, think that unusual fluctuations do not appear in this parameter, then make Surp=0; | P0-P
Surp+1| difference more than or equal to | 1/3P
Surp+1| situation under, think that tentatively P0 belongs to the unusual fluctuations point, change excessive isolated point and be considered to the unusual fluctuations point, for can not main points, this point participate in arithmetic average, only calculates the arithmetic mean of instantaneous value of all the other parameters behind all unusual fluctuations points of removing in N the derived parameter wherein, the meansigma methods of P1～Pn-1 for example, and this meansigma methods replaced P0, and unusual fluctuations are counted add 1, even Surp=Surp+1.When unusual fluctuations point occurring continuous 3 times, think that these three points are normal fluctuation, thereby but recover at these 3 for main points.If the abnormity point that occurs is less than 3 continuously, these points are finally regarded as the unusual fluctuations point so.The parameter values of these abnormity point is replaced by the parameter values of a nearest non-unusual fluctuations point before this point, and then carries out the arithmetic average of N continuous point, and this meansigma methods is replaced P0.P0 is transferred to host computer, the respirator of back, anesthetic machine etc. are carried out feedback control, owing to guaranteed the correct of parameter current value P0, thus the error feedback of respirator, anesthetic machine etc. avoided to the back, thus avoided the malfunction of respirator, anesthetic machine.
Present embodiment has been dispeled the catastrophe point of unusual (isolating), has kept normal catastrophe point, thereby has guaranteed to be transferred to the accuracy of the parameter of host computer.
Specific embodiment two, on the basis of specific embodiment one, for further eliminating the high frequency breath signal composition (period of waves is less than the waveform of the breathing cycle measuring range lowest limit) that valve closing is not sternly introduced, at foundation respiratory waveform state of living in (wave_counter) when each derived parameter calculates or gathers to the breathing cycle, increase to inspiratory duration with to expiratory duration and IE judgement than (be breathing time than).For example, for inspiratory duration less than 160ms, expiratory duration is less than the waveform of 200ms, and IE is than the breathing cycle outside 4:1～1:10, it is thought interfering signal without exception and do not carry out refreshing and sending of derived parameter, relevant because of this method with the time air-breathing and that exhale, can be described as the time period method.
Its handling process may further comprise the steps as shown in Figure 6:
In the flow process of calculation of parameter, be provided with the time reference of time reference amount as the breathing cycle.Wave_counter transmits (for example wave_counter incremental steps from Fig. 2) from the waveform recognition step, the another reference amount as calculation of parameter has identified residing state of breathing cycle.Wherein " 1 " expression breathing cycle is in inspiratory phase; " 2 " show that the breathing cycle is in expiratory phase; " 3 " expression breathing cycle end point, this constantly respiratory mechanics monitor module with a series of activities such as execution cycle calculation of parameter, parameter transmission.
In step 10, when wave_counter=1 (thinking that promptly respiratory waveform enters inspiratory phase), pick up counting, and execution in step 11, calculate all parameters in the inspiratory phase, comprise volume integral, inspiratory duration, inspiration peak flow, surge pressure etc.
In step 12, (wave_counter from 1 changes 2 into time) reads time value when air-breathing end, and execution in step 13 then;
In step 13, inspiratory duration (being the duration of inspiratory phase) and predefined air-breathing ultimate value are compared, if the duration of inspiratory phase is less than air-breathing ultimate value, then execution in step 20, this waveform is thought disturbing wave, and wave_counter forces to be set to 0 (being unidentified to respiratory waveform), and when respiratory waveform entered new inspiratory phase again, all calculation of parameter of carrying out in the disturbing wave were capped.If the inspiratory phase that the duration of inspiratory phase more than or equal to air-breathing ultimate value, then is provided with in respiratory waveform identification step sign is constant, execution in step 14 then;
In step 14, calculate all parameters in the expiratory phase;
In step 15, (when wave_counter equals 3) reads time value when the breathing cycle finishes, and execution in step 16 then;
In step 16, expiratory duration (being the duration of expiratory phase) and predefined expiration ultimate value are compared, if expiratory duration, illustrates that the expiratory phase that is recognized is a disturbing wave less than the expiration ultimate value, inspiratory phase reality does not finish.Force this moment wave_counter to be set to 1, turn to step 10 then, proceed the calculation of parameter in the inspiratory phase, up to entering new expiratory phase.If the expiratory phase that the duration of expiratory phase more than or equal to the expiration ultimate value, then is provided with in respiratory waveform identification step sign is constant, execution in step 17 then;
In step 17, carry out breathing cycle calculation of parameter and processing, execution in step 18 then;
In step 18, breathing cycle is when finishing, obtain the ratio (IE ratio) of air-breathing and expiratory duration, judge whether the IE ratio exceeds predefined breathing ratio range, if exceed, be that the IE ratio exceeds monitoring range (exceeding physiological range already), illustrate that the monitoring of this breathing cycle does not have reference value, its reason may be to connect again or other unpredictable incidents after the air flue pipeline disconnects.This moment the data that monitored in this monitoring periods are cancelled (promptly not sending), and wave_counter forces to be set to 0, turn to step 20, wait for new inspiratory phase to host computer.If do not exceed the breathing ratio range, then execution in step 19, send parameter to host computer.
Present embodiment has effectively been got rid of the interference waveform that does not sternly cause because of valve closing.
Specific embodiment three, present embodiment are to continue improved plan on the basis of specific embodiment one, two, great-leap-forward recognition method at zero point is adopted in the respiratory waveform recognition methods, just increase the setting of positive negative threshold value, as shown in Figure 7, the flow value that monitors also compares with positive negative threshold value, rather than compares with " 0 " in the picture prior art.The respiratory waveform identification process of present embodiment may further comprise the steps as shown in Figure 8:
In step 30, detect the flow value of respiratory air flow, execution in step 31 then;
In step 31,, put cur_point_sign=1 when flow value during greater than positive threshold value, when flow value during, put cur_point_sign=-1, when flow value is between the positive negative threshold value less than minus threshold value, keep present flow rate sampled point sign constant, execution in step 32 then;
In step 32, judge whether the product of cur_point_sign and pre_point_sign is-1, when being+1, then finishes described product this time identification; When described product was-1, then execution in step 33;
In step 33, breathing phases is identified wave_counter add 1, execution in step 34 then;
In step 34, give pre_point_sign with the value of cur_point_sign, execution in step 35 then;
In step 35, judge whether wave_counter=1 and flow＜0 sets up, if execution in step 36 then; If not execution in step 37 then;
In step 36, wave_counter forces to be set to 0, finishes this time identification then;
In step 37, judge whether it is wave_counter=3, if then execution in step 38, wave_counter is set to 1, if not then finishing this time identification.
Present embodiment has effectively been got rid of faint interfering signal and null offset.
Specific embodiment four, present embodiment adopt rising, trailing edge lock method further to improve the identification of breathing cycle.
When parameter monitoring, the real time data of flow and pressure respectively is provided with the buffer memory x1 of 3 data length, x2 and x3 are in each size of this three point value relatively constantly.These three points are exactly the flow waveform parameter of gathering in real time, and sample frequency can be 100Hz, that is to say every 0.01 second to gather a point, and x1, x2, x3 are three continuity points.
As x1〉x2〉during x3, this parameter waveform is in trailing edge;
When x1＜x2＜x3, this parameter waveform is in rising edge;
As x2〉x1 and x2〉during x3, this parameter waveform is in peak value;
When x2＜x1 and x2＜x3, this parameter waveform is in valley.
Utilize this fault-tolerance approach accurately pressure and flow waveform (pressure and flow waveform and volume waveform general designation respiratory waveform) residing stage to be locked, breathes in each residing stage in the moment thereby further grasped.For waveform recognition provides further information, can strengthen the accuracy of waveform recognition according to this information, unusual waveforms is differentiated out.
Be illustrated in figure 9 as the preferred embodiment that forms according to said method, comprise respiratory waveform identification module, parameter calculating module and unusual parameter eliminating module.The respiratory waveform identification module is used for the end point according to the inspiratory phase in respiratory waveform identification of breathing cycle, expiratory phase and breathing cycle, and the breathing phases sign is set to represent the respective identification of inspiratory phase, expiratory phase or breathing cycle end point; Parameter calculating module is used to calculate the parameter and the output parameter of respiratory waveform respective stage; Unusual parameter is got rid of the output of module responds respiratory waveform identification module and parameter calculating module, and the unusual parameter in the derived parameter is got rid of.Unusual parameter is got rid of module and is comprised pushup storage, first comparing unit, first processing unit and second processing unit.Pushup storage links to each other with parameter calculating module, the derived parameter of the N that temporary parameter calculating module calculates a continuous breathing cycle, wherein N 〉=3; First comparing unit is used for the parameter value P0 of current period and the parameter value P of last non-unusual fluctuations
Surp+1Compare; First processing unit is used for | P0-P
Surp+1|＜| 1/3P
Surp+1| the time, the meansigma methods of N derived parameter calculating is replaced the parameter value P0 of current period; | P0-P
Surp+1| ≧ | 1/3P
Surp+1| the time, the meansigma methods of removing all the other points behind all unusual fluctuations points in N the derived parameter is replaced the parameter value P0 of current period, and the continuous unusual fluctuations surp that counts is added 1; Second processing unit is used to judge that whether continuous unusual fluctuations count greater than 3, if then the arithmetic mean of instantaneous value of this successive N derived parameter point is replaced the parameter value P0 of current period.Thereby the parameter value P0 that has avoided current period is a catastrophe point, with parameter value P0 the respirator of back, anesthetic machine etc. is carried out feedback control again.
Can further include the phase place judge module, the phase place judge module comprises time detecting unit, time judging unit, the 3rd processing unit and breathes the ratio judging unit.The time detecting unit is used to detect the duration of inspiratory phase and expiratory phase; The time judging unit is used for the duration of inspiratory phase and expiratory phase is compared with predefined air-breathing ultimate value and expiration ultimate value respectively; The 3rd processing unit is used for the setting according to the judged result control breathing phase identification of time judging unit, at the duration of inspiratory phase during less than air-breathing ultimate value, this waveform is thought disturbing wave, and the breathing phases sign is set to unidentified to the respiratory waveform sign; During more than or equal to air-breathing ultimate value, then the control breathing phase identification is constant at the duration of inspiratory phase; During less than the expiration ultimate value, then the breathing phases sign forces to be set to the inspiratory phase sign at the duration of expiratory phase; During more than or equal to the expiration ultimate value, then the control breathing phase identification is constant at the duration of expiratory phase; Breathe the ratio judging unit and be used for when the breathing cycle finishes, judge whether the ratio of inspiratory phase duration and expiratory phase duration exceeds predefined breathing ratio range, if exceed, then all parameters that should calculate in the breathing cycle are cancelled; Parameter calculating module also responds the breathing phases sign of respiratory waveform identification module output, when respiratory waveform reenters next inspiratory phase, with this inspiratory phase in the stage parameters calculated cover all parameters of calculating in the described disturbing wave duration.
Can further include the waveform locking module, specifically comprise second sampling unit, second comparing unit and waveform judging unit.Second sampling unit is used for according to time sequencing sample three numerical value x1, x2, the x3 of real-time respiratory waveform; Second comparing unit is used for the size of three numerical value of comparison; The waveform judging unit is used for judging the residing stage of respiratory waveform according to the comparative result of three numerical value:
As x1〉x2〉during x3, this respiratory waveform is in trailing edge;
When x1＜x2＜x3, this respiratory waveform is in rising edge;
As x2〉x1 and x2〉during x3, this respiratory waveform is in peak value;
When x2＜x1 and x2＜x3, this respiratory waveform is in valley.
In sum, the present invention is directed to the fault-tolerant defect of insufficient of great-leap-forward recognition method at zero point, a series of specific aim fault tolerant mechanisms have been formulated, eliminate false judgment that these unusual waveforms cause parameter monitoring and lead to errors feedback or error diagnosis with lower cost, and then avoided patient's health even life security to be on the hazard.
Claims (11)
1. the fault-tolerant device in the respiratory mechanics monitoring system comprises:
The respiratory waveform identification module is used for the end point according to inspiratory phase, expiratory phase and the breathing cycle in respiratory waveform identification of breathing cycle, and the breathing phases sign is set to represent the respective identification of inspiratory phase, expiratory phase or breathing cycle end point;
Parameter calculating module is used to calculate the parameter and the output parameter of respiratory waveform respective stage; It is characterized in that: the unusual parameter of also comprise the output that is used for responding respiratory waveform identification module and parameter calculating module, the unusual parameter of derived parameter being got rid of is got rid of module, and described unusual parameter is got rid of module and comprised:
Pushup storage is used for the temporary N that calculates the derived parameter of breathing cycle, wherein a N continuously 〉=3;
First comparing unit is used for the parameter value P0 of current period and the parameter value P of last non-unusual fluctuations
Surp+1Compare;
First processing unit is used for | P0-P
Surp+1|＜| 1/3P
Surp+1| the time, the arithmetic mean of instantaneous value of this derived parameter in nearest N cycle calculating is replaced the parameter value P0 of current period; | P0-P
Surp+1| ≧ | 1/3P
Surp+1| the time, the meansigma methods of removing all the other points behind all unusual fluctuations points in nearest N the consecutive points that calculate is replaced the parameter value P0 of current period, and continuous unusual fluctuations are counted add 1, wherein surp is that continuous unusual fluctuations are counted;
Whether second processing unit is used to judge continuous unusual fluctuations number of spots greater than 3, if then the arithmetic mean of instantaneous value of these successive N derived parameter points is replaced the parameter value P0 of current period.
2. fault-tolerant device as claimed in claim 1 is characterized in that: also comprise the phase place judge module, described phase place judge module comprises:
The time detecting unit is used to detect the duration of inspiratory phase and expiratory phase;
The time judging unit is used for the duration of inspiratory phase and expiratory phase is compared with predefined air-breathing ultimate value and expiration ultimate value respectively;
The 3rd processing unit, be used for setting according to the judged result control breathing phase identification of time judging unit, during less than air-breathing ultimate value, this waveform is thought disturbing wave, and the breathing phases sign is set at the duration of inspiratory phase unidentified to the respiratory waveform sign; More than or equal to air-breathing ultimate value, then the control breathing phase identification is constant at the duration of inspiratory phase; During less than the expiration ultimate value, then the breathing phases sign forces to be set to the inspiratory phase sign at the duration of expiratory phase; During more than or equal to the expiration ultimate value, then the control breathing phase identification is constant at the duration of expiratory phase;
Breathe the ratio judging unit, be used for when the breathing cycle finishes, judge whether the ratio of inspiratory phase duration and expiratory phase duration exceeds predefined breathing ratio range, if exceed, then all parameters that should calculate in the breathing cycle are cancelled;
Described parameter calculating module also responds the breathing phases sign of respiratory waveform identification module output, when respiratory waveform reenters next inspiratory phase, with this inspiratory phase in the stage parameters calculated cover all parameters of calculating in the described disturbing wave duration.
3. fault-tolerant device as claimed in claim 1 or 2 is characterized in that: also comprise the waveform locking module, described waveform locking module comprises:
Second sampling unit is used for according to time sequencing sample three numerical value x1, x2, the x3 of real-time respiratory waveform;
Second comparing unit is used for the size of three numerical value of comparison;
The waveform judging unit is used for judging the residing stage of respiratory waveform according to the comparative result of three numerical value: as x1〉x2〉during x3, this respiratory waveform is in trailing edge; When x1＜x2＜x3, this respiratory waveform is in rising edge; As x2〉x1 and x2〉during x3, this respiratory waveform is in peak value; When x2＜x1 and x2＜x3, this respiratory waveform is in valley.
4. the fault-tolerance approach in the respiratory mechanics monitoring system is characterized in that may further comprise the steps:
The respiratory waveform identification step: according to the end point of inspiratory phase, expiratory phase and the breathing cycle in respiratory waveform identification of breathing cycle, and the breathing phases sign is set to represent the respective identification of inspiratory phase, expiratory phase or breathing cycle end point;
Calculation of parameter step: the inspiratory phase and the parameter in expiratory phase stage of calculating the breathing cycle;
Unusual parameter is got rid of step: the unusual parameter in the derived parameter after the calculation of parameter step is got rid of; Described unusual parameter is got rid of step and be may further comprise the steps:
A1, with the temporary N that calculates of the mode of FIFO the derived parameter of breathing cycle, wherein N continuously=3;
The parameter value P0 of B1, comparison current period and the parameter value P of last non-unusual fluctuations
Surp+1If, | P0-P
Surp+1|＜| 1/3P
Surp+1|, then the parameter value P0 of current period is a normal point, the arithmetic mean of instantaneous value of this derived parameter in nearest N cycle calculating is replaced the parameter value P0 of current period; If | P0-P
Surp+1| ≧ | 1/3P
Surp+1|, execution in step D1 then;
D1, then the parameter value P0 of current period is the unusual fluctuations points, the parameter value P0 that the meansigma methods of removing all the other points behind all unusual fluctuations points in nearest N the consecutive points that calculate is replaced current period, and continuous unusual fluctuations are counted add 1, wherein surp is that continuous unusual fluctuations are counted, then execution in step E1;
E1, whether judge continuous unusual fluctuations number of spots, if then the arithmetic mean of instantaneous value of these successive N derived parameter points is replaced the parameter value P0 of current period greater than 3.
5. fault-tolerance approach as claimed in claim 4 is characterized in that: the N in the described steps A 1 is less than or equal to 8.
6. fault-tolerance approach as claimed in claim 4 is characterized in that also being included in the phase place determining step of carrying out in the breathing cycle identifying, and described phase place determining step may further comprise the steps:
The duration of A2, detection inspiratory phase;
B2, the duration and the predefined air-breathing ultimate value of inspiratory phase compared, if the duration of inspiratory phase is more than or equal to air-breathing ultimate value, the inspiratory phase sign that then is provided with in the respiratory waveform identification step is constant, if the duration of inspiratory phase is less than air-breathing ultimate value, execution in step C2 then;
C2, this waveform is thought disturbing wave, and the inspiratory phase that is provided with in respiratory waveform identification step sign forces to be set to unidentified to the respiratory waveform sign;
The duration of D2, detection expiratory phase;
E2, the duration and the predefined expiration ultimate value of expiratory phase compared, if the duration of expiratory phase is more than or equal to the expiration ultimate value, the breathing phases sign that then is provided with in the respiratory waveform identification step is constant, if the duration of expiratory phase is less than the expiration ultimate value, execution in step F2 then;
The expiratory phase sign that is provided with in F2, the respiratory waveform identification step forces to be set to the sign of inspiratory phase.
7. fault-tolerance approach as claimed in claim 6, it is characterized in that: further comprising the steps of after step C2: when respiratory waveform reenters next inspiratory phase, with this inspiratory phase in the stage parameters calculated cover all parameters of calculating in the described disturbing wave duration.
8. fault-tolerance approach as claimed in claim 7 is characterized in that: described phase place determining step is further comprising the steps of:
G2, when the breathing cycle finishes, judge whether the ratio of inspiratory phase duration and expiratory phase duration exceeds predefined breathing ratio range, if exceed, all parameters of calculating in then should the breathing cycle are cancelled.
9. as each described fault-tolerance approach in the claim 4 to 8, it is characterized in that: described respiratory waveform identification step may further comprise the steps:
The flow value of A3, detection respiratory air flow;
B3, when flow value during greater than positive threshold value, put the present flow rate sampled point be designated on the occasion of, when flow value during, put the present flow rate sampled point and be designated negative value less than minus threshold value, when flow value is between the positive negative threshold value, keep present flow rate sampled point sign constant;
C3, judge whether the present flow rate sampled point sign and the product of previous traffic sampling point identification are negative value, when described product on the occasion of the time then finish; When described product is negative value, execution in step D3 then:
D3, breathing phases sign is promoted according to the order of inspiratory phase sign, expiratory phase sign, breathing cycle end point sign, and breathing phases be designated the breathing cycle during end point sign breathing phases sign be set to the inspiratory phase sign.
10. as each described fault-tolerance approach in the claim 4 to 8, it is characterized in that: in described respiratory waveform identification step, according to time sequencing sample three numerical value x1, x2, the x3 of real-time respiratory waveform, judge the residing stage of respiratory waveform according to the comparative result of three numerical value:
As x1〉x2〉during x3, this respiratory waveform is in trailing edge;
When x1＜x2＜x3, this respiratory waveform is in rising edge;
As x2〉x1 and x2〉during x3, this respiratory waveform is in peak value;
When x2＜x1 and x2＜x3, this respiratory waveform is in valley.
11. fault-tolerance approach as claimed in claim 9, it is characterized in that: in described respiratory waveform identification step, according to time sequencing sample three numerical value x1, x2, the x3 of real-time respiratory waveform, judge the residing stage of respiratory waveform according to the comparative result of three numerical value:
As x1〉x2〉during x3, this respiratory waveform is in trailing edge;
When x1＜x2＜x3, this respiratory waveform is in rising edge;
As x2〉x1 and x2〉during x3, this respiratory waveform is in peak value;
When x2＜x1 and x2＜x3, this respiratory waveform is in valley.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB200610062226XA CN100508885C (en) | 2006-08-18 | 2006-08-18 | Fault-tolerant method and device in respiratory mechanics monitoring system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB200610062226XA CN100508885C (en) | 2006-08-18 | 2006-08-18 | Fault-tolerant method and device in respiratory mechanics monitoring system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101125084A CN101125084A (en) | 2008-02-20 |
| CN100508885C true CN100508885C (en) | 2009-07-08 |
Family
ID=39093283
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB200610062226XA Expired - Fee Related CN100508885C (en) | 2006-08-18 | 2006-08-18 | Fault-tolerant method and device in respiratory mechanics monitoring system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100508885C (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012080920A1 (en) | 2010-12-17 | 2012-06-21 | Koninklijke Philips Electronics N.V. | System and method for determining one or more breathing parameters of a subject |
| CN104535827A (en) * | 2015-01-12 | 2015-04-22 | 航天科工深圳（集团）有限公司 | Defective point removing method and system used in AD sampling |
| CN107847184B (en) * | 2015-06-30 | 2021-01-22 | 皇家飞利浦有限公司 | Control device for a medical aerosol delivery device |
| CN107990952B (en) * | 2017-11-28 | 2020-01-31 | 重庆智慧水务有限公司 | Method for processing instantaneous flow of flow meter and flow meter |
| CN113116336A (en) * | 2021-03-22 | 2021-07-16 | 深圳市安保科技有限公司 | Respiration detection method and device, and computer storage medium |
| CN115970109B (en) * | 2023-03-17 | 2023-08-04 | 苏州鱼跃医疗科技有限公司 | Respiratory ventilation prediction preprocessing method, ventilator, controller and storage medium |
-
2006
- 2006-08-18 CN CNB200610062226XA patent/CN100508885C/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN101125084A (en) | 2008-02-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2677927B1 (en) | Respiration monitoring method and system | |
| CN100508885C (en) | Fault-tolerant method and device in respiratory mechanics monitoring system | |
| EP2945537B1 (en) | Detection of sleep apnea using respiratory signals | |
| CN101087559B (en) | Method for detecting and discriminating breathing patterns from respiratory signals | |
| CN102343118B (en) | Method and apparatus for improved flow limitation detection of obstructive sleep apnea | |
| CN102458245B (en) | Discrimination of cheyne -stokes breathing patterns by use of oximetry signals | |
| CN101815465B (en) | Electrocardiogram derived apnoea/hypopnea index | |
| US20090247837A1 (en) | System And Method For Diagnosing Sleep Apnea | |
| CN101563028A (en) | Patient monitoring system and method | |
| CN102266616B (en) | Method and device for monitoring blockage of breathing equipment | |
| CN111563451B (en) | Mechanical ventilation ineffective inhalation effort identification method based on multi-scale wavelet characteristics | |
| CN107252313A (en) | The monitoring method and system of a kind of safe driving, automobile, readable storage medium storing program for executing | |
| CN100508884C (en) | Fault-tolerant method and device in respiratory mechanics monitoring system | |
| WO2011148766A1 (en) | Method and system for reliable inspiration-to-expiration ratio extraction from acoustic physiological signal | |
| CN111919242A (en) | System and method for processing multiple signals | |
| JP5544365B2 (en) | Improvements in multi-parameter monitoring or improvements related to multi-parameter monitoring | |
| CN111227792B (en) | Apnea detection method and system, electronic device and storage medium | |
| Avcı et al. | Sleep apnea detection using wavelet analysis of ECG derived respiratory signal | |
| CN103892797A (en) | Signal processing method and device for sleep structure analysis | |
| CN109876262B (en) | Automatic detection method for effusion in breathing machine pipeline based on wavelet | |
| CN115414026A (en) | Automatic breath segmentation method and system based on flow velocity waveform | |
| CN108742628A (en) | A kind of breathing detection method | |
| Avcı et al. | Comparison of the ANN based classification accuracy for real time sleep apnea detection methods | |
| Guul et al. | Portable prescreening system for sleep apnea | |
| Beatty et al. | Shape-only identification of breathing system failure |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090708 Termination date: 20160818 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |