JP2926392B2 - Detector - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

TECHNICAL FIELD The present invention relates to the detection device of the gas flow,
In particular, the present invention relates to a device used for a treatment device for treating respiratory arrest that occurs during sleep (hereinafter, sleep apnea syndrome).

[0002]

BACKGROUND OF THE INVENTION Sleep apnea, especially closed sleep apnea, is estimated to affect 4-9% of the population and accidentally obstructs the upper airway during sleep. Things. This is clinically daytime drowsiness, cardiac arrhythmias,
It causes pulmonary arterial hypertension, congestive heart disease, sensory disturbance and the like.

[0003] The pathological details of upper airway obstruction in patients with sleep apnea syndrome have not been determined and are generally due to anatomical or functional abnormalities of the upper airway due to increased air resistance. It is believed that. Such anomalies may be due to suction during inhalation, tongue pulling on the pharyngeal wall due to gravity, and insufficient muscle strength of the upper airway dilator to narrow the upper airway. It has also been hypothesized that the mechanism that responds to the known association of obesity with sleep apnea syndrome is that tissue is excessively soft in the front and sides of the neck, which applies sufficient pressure on the internal structures that narrow the airway. ing.

[0004] Treatment for sleep apnea syndrome includes surgery, uvulo-pharyngoplasty, obesity surgery and chin shaping. Another example of surgery for the treatment of sleep apnea is tracheostomy.
Even if these treatments are helped, there is a high risk of becoming morbid after surgery. Pharmacological treatment is generally undesirable, especially for patients heavier than mild respiratory arrest. Further, the side effects are periodic.
Practitioners are looking for non-aggressive methods of treating sleep apnea, with a high success rate and easy patient cooperation, including, for example, dieting and regular dieting when obesity is involved.

[0005] As a recent treatment method for sleep apnea syndrome, continuous positive airway pressure (CPAP), which continuously keeps the patient's airway open during sleep, has been proposed.
e) is used. For example, U.S. Pat. No. 4,655,213 and Australian Patent AU-B-83901 / 82 show a therapy based on continuous positive airway pressure applied in a patient's airway.

[0006] US Patent No. 4,773,411 shows a ventilated treatment method and apparatus. In this method, a substantially constant boosted airway pressure is provided and periodically reduced for a short period of time to a pressure no lower than atmospheric pressure.

[0007] Publications on the application of CPAP in the treatment of sleep apnea are as follows.

[0008] 1. Lindsay, DA, Issa FG, and Sullivan
CE "Mechanisms of Sleep De-saturation in Chroni
c Airflow Limitation Studied with Nasal Continuous
Positive Airway Pressure (CPAP) ", Am Rev Respir Di
s, 1982; 125: p.112. Sanders MH, Moore SE, Eveslage J. "CPAP via na
sal mask: A treat-ment for occlusive sleep apnea ",
Chest, 1983; 83: pp. 144-145. Sullivan CE, Berthon-Jones M, Issa FG. "Remiss
ion of severe obesity-hypoventilation syndrome aft
er short-term treatment during sleep withcontinuou
s positive airway pressure ", Am Rev Respir Dis, 19
83; 128: pp.177-181. Sullivan CE, Issa FG, Berthon-Jones M, Eveslag
e. "Reversal ofobstructive sleep apnea by continuo
us positive airway pressure appliedthrough the nar
es ", Lancet, 1981; 1: pp. 862-865. 5. Sullivan CE, Berthon-Jones M. Issa FG." Treatm
ent of obstructiveapnea with continuous positive a
irway pressure applied through thenose ", Am Rev Re
spir Dis, 1982; 125: p. 107. Annual Meeting Abstra
cts. Rapoport DM, Sorkin B, Garay SM, Goldring RM.
"Reversal of the 'Pickwickian Syndrome' by long-te
rm use of nocturnal nasal-airwaypressure ", N Engl
J. Med, 1982; 307: pp. 931-933. Sanders MH, Holzer BC, Pennock BE. "The effect
of nasal CPAP onvarious sleep apnea patterns ", Ch
est, 1983; 84: p. 336. Presented at the Annual Mee
ting of the American College of Chest Physicians, C
hicago, IL, October 1983.

[0009]

Although CPAP is very effective and recognized by many, especially many patients with sleep apnea are not resistant to CPAP. For this reason,
Other viable non-aggressive treatments have been studied.

Accordingly, it is an object of the present invention to provide a new and improved method and apparatus for treating respiratory arrest during sleep.

[0011]

SUMMARY OF THE INVENTION In the present invention, a pressure that is varied relative to atmospheric pressure is provided in the upper airway of a patient,
Develops airway dilators to relieve upper airway closure during sleep.

In an embodiment of the present invention, positive pressure is alternately delivered to the patient's airway at relatively high and low pressure levels. Thereby, the pressure delivered to dilate the patient's airway is an alternating large and small dilation force. The high and low positive pressure applications are initiated by the patient's natural breathing, with the high pressure provided during inhalation and the low pressure provided during exhalation.

In the treatment device according to the present invention, a predetermined and adjustable pressure is supplied to the patient's airway through an airflow generator, a variable pressure controller and a flow converter. The flow transducer generates an output signal that is used to create a signal proportional to the instantaneous flow to the patient. The instantaneous flow signal is provided to a low pass filter, which passes only a signal indicative of the average flow over time. The average flow signal is a value that indicates a positive flow such that the system has a minimum discharge from the patient circuit (ie, a small exhaust from around the respirator attached to the patient). The average flow signal indicates such emissions. Because the sum of all other flows is
It becomes zero quantitatively in a long time. That is, the suction flow rate must be equal to the discharge flow rate for a long time. That is, over a long period of time, the amount of inhaled air is equal to the volume of discharged gas.

Both the instantaneous flow rate signal and the average flow rate signal are supplied to a suction / discharge determination circuit. The decision circuit is, in its simplest form, a comparator, which compares the two signals and generates a drive signal to the pressure controller. Generally, when the instantaneous flow signal exceeds the average flow, the patient inhales and the drive signal provided to the pressure controller sets the pressure controller to supply air to the patient's airway at a predetermined pressure increase. . Similarly, if the instantaneous flow is less than the average flow, the patient will exhale,
The decision circuit generates a drive signal such that the pressure controller provides a relatively low pressure to the patient's airway. In this way, the patient's airway is kept open by alternating high and low pressures during inhalation and exhalation.

As noted above, many sleep apnea patients are not resistant to standard CPAP. Especially patient 2
5% are not CPAP effective due to the attendant inconvenience. CPAP
Uses the same pressure during inhalation and exhalation. Pressurization during both modes of breathing is difficult with the sensitivity of the exhaled and inflated chest. However, the inhalation and exhalation air resistance in the airway increases during the sleep period from the beginning of respiratory arrest, and the airway flow resistance is lower during inhalation than during inhalation. The bi-level CPAP treatment of the present invention is sufficient to keep the larynx open during exhalation, even if the pressure during exhalation is not as high as needed to release the larynx during inhalation. In addition, some patients have increased upper airway resistance during inhalation. In the present invention, a boost is provided only during inhalation, eliminating the need for a global (inhalation and exhalation) increase in airway pressure. The relatively low pressure provided during dispensing is often near or equal to atmospheric pressure. The low pressure delivered to the airways during insufflation is usually CPAP
Alleviates some of the discomfort conditions that correspond to, thereby increasing the patient's applicability.

In the conventional CPAP treatment, a pressure of about 20 cmH2O is required, and the nasal CPAP is exposed to an unnecessarily high discharge pressure. The present invention was able to independently supply high inspiratory airway pressures for low exhaled airway pressures and failed CPAP therapy.
% Of patients, and can be applied to the remaining 75% of patients.

The switching between high and low pressure is controlled by the patient's spontaneous breathing, which allows the patient to handle breathing rate and volume independently. In the present invention, the system discharge is automatically compensated, and the airflow of the system can be made smaller than that of the conventional system by wearing the nasal mask. Furthermore, in addition to automatic discharge compensation, low pressure airway pressure on the patient, improved safety, etc. are obtained.

[0018]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, an embodiment of the present invention will be described with reference to the drawings.

Referring to FIG. 1, a respiratory arrest treatment apparatus 10 according to a first embodiment of the present invention is shown in a functional block diagram. The treatment device 10 operates in a novel way. That is, respiratory gas, such as air, is applied to the patient 12 at a relatively low and high pressure (ie, equal to or higher than atmospheric pressure) for treatment of a condition known as respiratory arrest during sleep.
And, they are given alternately.

The treatment device 10 includes an airflow generator 14 (eg, a blower) that receives breathing gas from other suitable sources, such as a pressure bottle 16 or atmospheric pressure. Airflow generator 1
The gas flow from 4 is provided through a transmission tube 20 to a respiratory device such as a mask 22 attached to the patient 12. The mask 22 is preferably a nose mask or a full face mask as shown in the figure. Na can be used instead of a mask.
Anything that performs an interface operation between the patient and the sal cannulae, endotracheal or respiratory gas source may be used.

The mask 22 has a suitable exhaust port 24 for exhausting breathing gas during breathing. The exhaust port 24 is continuously open. The pressure controller 26 controls the airflow generation unit 1
It is arranged linearly in the transmission tube 20 between the mask 4 and the mask 22, and controls the pressure of the airflow in the transmission tube 20, that is, in the airway of the patient 12. For example, the outlet 24 has a cross-sectional basin that allows for a continuous evacuation of about 15 liters / minute. Exhaust port 2
The flow through 4 is one component, but a key component of the total system discharge, which is an important parameter of system operation.
In other embodiments described below, a non-rebreathing valve is used in place of the continuous opening 24.

The pressure controller 26 controls the pressure of the respiratory gas in the transmission tube 20, and thus in the patient's airway. The pressure controller 26 is preferably located downstream of the airflow generator 14 and may be in the form of a movable valve that provides a flow path open to the atmosphere through a restricted opening. The movable valve is adjustable so that all flow rates maintain a constant pressure drop at the opening, thereby maintaining a constant pressure in the transmission tube 20.

A suitable flow converter 28 is provided alongside the transmission line 20, preferably downstream of the pressure controller 26.
The output signal 29 of the flow converter 28 is provided to a flow signal conditioner 30 which generates a signal proportional to the instantaneous flow of breathing gas to the patient in the transmission tube 20.

The airflow generator 14 is not necessarily an absolute replacement component. For example, there is a blower. The blower creates a pressure head in the transmission tube 20 and, as described above, creates an amount of airflow to maintain the pressure head in the presence of the patient's respiratory cycle, the vent 24 and the activation of the pressure controller 26. . Thus, when the patient exhales, the peak exhaust flow value from the lungs is
Far more than the flow capacity. As a result, the exhaust gas flows from inside the transmission pipe 20 through the flow rate converter 28 to the pressure controller 26.
Backflow to In this manner, the instantaneous flow signal from transducer 28 varies widely from a relatively large positive flow (ie, toward the patient) to a relatively large negative flow (ie, from the patient).

The instantaneous flow signal from the signal adjusting circuit 30 is
As indicated by reference numeral 32, the signal is supplied to a decision circuit 34 such as a comparison circuit, and is also supplied to a low-pass filter 38 as indicated by reference numeral 36. Low-pass filter 38
Has a cut-off frequency low enough to remove most of the changes in the signal caused by normal breathing from the instantaneous flow input signal. The low pass filter 38 has a sufficiently long time constant so that spurious signals, abnormal flow patterns, and peak instantaneous flow valves do not dramatically affect the average flow of the system. That is, the time constant of the low-pass filter 38 is
It is chosen to be long enough to respond slowly to the instantaneous flow signal input. Thus, the steepest flow input signal that has a short impact on the average flow of the system in a short time will have a smaller impact over a longer time. This is because such a steep flow signal element is canceled in a long time. For example, instantaneous flow peak values alternate with relatively large positive and negative flow values corresponding to the peak inhalation and exhalation volumes performed by the patient during normal spontaneous breathing. Thus, the low-pass filter 3
The output of 8 is a signal proportional to the average flow rate in the system. Since the inhalation and exhalation volumes are actually offset, as described above, this signal is a positive flow corresponding to the average system exhaust (including the flow from exhaust 24).

The average flow signal from low pass filter 38 is provided to decision circuit 34, as indicated by reference numeral 40. In decision circuit 34, the instantaneous flow signal is continuously compared to the average flow signal of the system. Decision circuit 34
Controls the pressure controller 26 as the drive signal 42. The pressure value of the breathing gas in the transmission tube 20 is associated with the natural respiration of the patient 12 as described below.

When the patient begins to inhale, the instantaneous flow signal becomes a positive value higher than the positive average flow signal value. The decision circuit 34 detects this increase as the start of inhalation of the patient.
The output signal from the decision circuit 34 is supplied to the pressure controller 26. In response, the pressure controller 26 controls the transmission line 2
A high pressure gas flow is delivered into the airflow to increase the pressure in the patient 12 airway. This is the high pressure value of a bi-level CPAP system and is referred to as IPAP (Inhalation Positive Airway Pressure). During the inhalation period, the transmission tube 20
The flow rate in the inside increases to the maximum value, and then decreases with the end of inhalation.

At the beginning of exhalation, there is no airflow to the patient's lungs. As a result, the instantaneous flow signal is smaller than the average flow signal, which is a relatively constant positive flow value.
The determination circuit 34 detects this state as the start of the discharge, and supplies a drive signal to the pressure controller 26. The pressure controller 26 supplies a low-pressure gas flow into the transmission pipe 20 accordingly. This low pressure is the low level pressure value in a dual level CPAP system and is referred to as EPAP (positive exhaust airway pressure). As noted above, the EPAP pressure range includes atmospheric pressure. When the patient begins spontaneous inhalation again, the instantaneous flow signal increases again and exceeds the average flow signal. The decision circuit again provides a drive signal to the pressure controller 26 to reapply the IPAP pressure.

The above operation requires a regular comparison of at least the input signals 32 and 40 in the decision circuit 34. Although this and other actions are described as continuous, they are continuous (i.e., without interruption) and periodic (i.e., remote).
Both are included.

As mentioned above, the system 10 includes the exhaust 24
Has a specific control displacement, which ensures that the average displacement signal is at least small and positive. During inhalation, the flow rate detected at the flow transducer is the sum of the exhaust volume through exhaust outlet 24 and other system output from downstream of transducer 28 and the inhaled volume in patient 12 airway. Thus, during inhalation, the instantaneous flow signal adjusted by the adjustment circuit 30 accurately reflects the suction flow above the average flow signal. At the time of exhalation, the flow rate in the transmission pipe 20 far exceeds the flow capacity of the exhaust port 24 and flows back from the patient's lungs as a discharge flow. Therefore, the exhaust air flows back through the transmission pipe 20 through the converter 28 toward the pressure controller 26. Since the pressure transducer 26 operates to maintain the set pressure, it operates in response to flow from both the patient 12 and the airflow generator 14 and opens to an output port sufficient to accommodate the additional flow rate. . Thus, the pressure determined by the operation of the decision circuit 34 can be maintained.

In both the inhalation and exhalation cycles, the gas pressure in the transmission tube 20 can create pressure in the patient's airway to create an open airway, thereby reducing airway constriction.

In practice, it is desirable to slightly offset the switching level in the decision circuit 34 with respect to the average flow signal. This prevents the system from prematurely switching from low pressure exhaust mode to high pressure suction mode. That is, a switching decision point offset in the positive direction from the average flow of the system is set so as not to switch to the system IPAP mode until the patient actually performs a predetermined minimum amount of sufficiently natural inhalation. I have. This indicates that the beginning of inhalation is completely natural and not due to an artificial increase in airway pressure.
Even in IPAP mode, ensure that the transition to low-pressure EPAP mode is ensured before air flow to the patient's lungs reaches zero (ie, switching to EPAP occurs shortly before the patient stops inhaling). 2), a similar switching set point offset is provided. This does not cause excessive initial resistance to the patient's natural exhalation.

As is apparent from the above description, a new treatment for respiratory arrest during sleep has been proposed, according to which the airway pressure of the patient does not interfere with the patient's natural breathing at all. It is maintained at a positive high pressure during inhalation and at a relatively low pressure during exhalation. The described device applies a positive pressure to the flow of breathing gas to the patient, and a flow of air that alternates between high and low pressure in the patient's trajectory in response to the patient's natural inhalation and exhalation, during sleep. Treat patients with respiratory arrest. The described system can be used with a pressure support system such as the proportional airway pressure system shown in US Pat. No. 5,107,830.

To control pressure, the flow of breathing gas to the patient is detected and continuously processed to obtain a signal proportional to the instantaneous breathing gas flow in the system. The instantaneous flow signal is processed to remove changes due to normal patient breathing and other causes, producing a signal proportional to the average gas flow of the system. The average flow signal is continuously compared with the instantaneous flow signal to remove the relationship between the patient's natural breathing and the average flow of the system. When the instantaneous flow exceeds the average flow, the patient inhales and the gas pressure on the patient is accordingly determined to be the selected positive pressure, providing a corresponding pressure in the patient's airway. When the comparison of the instantaneous flow signal and the average flow signal indicates that the patient is exhaling, that is, when the instantaneous flow signal is less than or equal to the average flow, a corresponding low pressure is provided in the patient's airway. As such, the pressure of the respiratory gas to the patient is adjusted to the selected low pressure.

In a second embodiment of the present invention, shown in FIGS. 2 and 3, low pass filter 38 is used to reduce the expected emissions including low pass filters and other components, as shown in FIG. The computer has been swapped. Other elements of FIG. 2 are similar to the system shown in FIG. Therefore, the same elements are denoted by the same reference numerals, and the description in the embodiment of FIG. 1 is generally applicable to FIG.

As will be described later, the predicted emission amount computer 5
By using the zero function, it is possible to adjust the reference signal provided to the decision circuit 34 on a breath-by-breath basis, as well as simply providing the average flow of the system over time. In order to distinguish this new reference signal from the average flow rate, in the following description it will be referred to as the predicted emission signal or simply the predicted emission signal.

As mentioned above, the average flow reference signal of the system varies very slowly due to the long time constant of the low pass filter 38. This operational feature has been consciously employed to avoid disturbances to the reference signal due to abnormal instantaneous flow signal inputs, such as irregular breathing patterns. When it was possible to minimize the impact of the anomalies on the average flow reference signal, the average flow would change, albeit very small, corresponding to small increases and disturbances. Due to the long time constant of the low-pass filter, such a change in the reference signal is long lasting.

Furthermore, even small changes in the reference signal,
It has a very significant effect on the triggering of the system.
For example, when the inspiratory flow starts to go positive, the system
Mode, so small changes in the reference signal
Causes relatively large changes in respiratory activity required to trigger the system into IPAP mode. In many cases, changes in the reference signal are so large that the patient cannot trigger the system with normal respiratory activity. For example, if the system is turned on prior to placing the mask on the patient's face, the initial free air flow from the unmounted mask will result in a very large positive value for the initial average flow of the system. If such a value exceeds the maximum inspiratory flow in the patient's natural breathing,
The system cannot trigger between IPAP and EPAP modes. After the patient wears the mask to lower the reference signal to a value closer to the actual system output during operation, the instantaneous flow signal is greater than the average flow signal until the decision circuit has at least enough normal breathing cycles. Is not detected. As mentioned above, the low-pass filter takes some time, during which the patient can breathe naturally against a uniform positive pressure. This is the traditional CPAP
And is not intended to limit the invention.

In addition to the embodiment based on a reference signal derived from the predicted exhalation for each breath controlled entirely by the patient's natural breath, there are two more modes of operation.
One is spontaneous / timed operation, in which the system automatically triggers into the IPAP mode long enough to begin inhaling a patient after the onset of exhalation, without having to exhale in a defined period of time. To compose this, a timer is provided, which is reset when the inhalation cycle is naturally triggered or by the timer itself. In this way, only the start of inhalation is triggered by the timer. In this mode, the other part of the operating cycle
It is controlled by the patient's natural breathing and system circuitry.

Another mode is based on the instantaneous operation of the system, rather than the patient's natural respiratory activity, and responds to a timed cycle in harmony with the patient's natural breathing.

Referring to FIG. 3, the predicted discharge computer 50 includes a low pass filter 3 in addition to circuitry operative to correct for the predicted discharge signal based on each patient breath analysis.
8 ′. Circuits that quickly adjust the expected discharge flow signal after a large change in system flow due to the blower already operating before the mask is worn on the patient or a large discharge in which the system has already been started or stopped. Is provided.

The low-pass filter 38 'has a data storage function whose operation will be described later.

The low-pass filter 38 'operates in the same manner as described with reference to FIG. It provides a long-term average flow rate that is equivalent to the fixed emissions of the system, including the flow capacity of the outlet 24. This long term averaging, in the embodiment of FIG. 3, serves to adjust the predicted emission reference signal only when the flow conditions of the system change very slowly.

For breath-by-breath analysis and adjustment of the reference signal, the differential amplifier 25 receives the instantaneous flow signal as shown at 54 and predicts from the low-pass filter 38 'as shown at 56. Receive the emission signal.

The output of the differential amplifier 52 is the difference between the instantaneous flow rate and the predicted emissions patent, ie, the predicted instantaneous patient flow rate. This is apparent when considering that the instantaneous flow is the sum of the flow to the patient and the actual system drain. The predicted patient flow signal from differential amplifier 52 is provided to flow integrator 60, as indicated by reference numeral 58. The flow integrator 60 integrates the predicted patient flow at the beginning and end of each breath upon triggering on the IPAP. Another input to integrator 60 is the IPAP / EPAP status signal, as indicated by reference numeral 62. IP
The AP / EPAP status signal is a drive signal supplied to the pressure controller 26. That is, it is a signal indicating the pressure state as between IPAP / EPAP. The status signal is used to indicate the start and end of each breath for integration by the integrator 60 on a breath-by-breath basis.

The predicted emissions signal from low pass filter 38 'is equal to the true system emissions. If the patient's inhalation and exhaust volume is the same as the respiration (ie, if the patient's total positive flow is equal to the total negative flow), the integrated value calculated by the integrator 60 will be zero, and adjustment of the predicted exhaust volume Is not required. If the integrated value by the integrator 60 is not zero, the integrator 6
The integrated value in the output from 0 is supplied to a sample and hold circuit 66 as indicated by reference numeral 64. Of course,
Even if the integral value is zero, the output signal is supplied to the circuit 66, but does not adjust the predicted emission signal.

The non-zero integrated value from circuit 66, as indicated by reference numeral 68, is also applied to circuit 38 'in response to the IPAP / EPAP status signal to circuit 66, indicated by reference numeral 70, for each patient breath. To be supplied. The supply of the non-zero integral value to the circuit 38 'is adjusted so that the predicted flow signal is proportional to the integral value and, if the other conditions are the same, reduce the integral value so that it becomes zero at the next breath. To do that.

In this system, if the patient's total respiratory cycle value is zero and the system output changes, the integrator circuit corrects the output by adjusting to the expected output within about 10 breaths of the patient.

The integrator circuit 60 adjusts the predicted output during the patient's breathing cycle in response to the non-zero value. Non-zero patient respiratory values are not exceptional. For example, a patient may inhale a little more than a few breathing cycles of exhalation,
Thereafter, discharge is performed deeply and sufficiently. In this case, the integration circuit
Adjust projected emissions even when true system emissions change. However, since the reference signal correction is only about one-tenth of the total correction required in one breath, the reference signal does not change noticeably in only one or two breaths. In this way, the integration circuit is responsive to both system output and changes in the patient's normal breathing. Integrator circuits are usually
It is active, for example, when the patient is breathing fast.

Discharge circuit 74 calculates other data elements used to predict system emissions, as described below. The circuit 74 monitors the slope of the instantaneous flow waveform. If the slope value is near zero at the time of dispense (shown as state signal input 76), it indicates that the flow rate does not change. If the slope of the instantaneous flow waveform is
If it is smaller than one-half, it indicates that the discharge is completed and the total flow rate at that time is the discharge amount. However, if the predicted flow is non-zero, one component of the instantaneous flow signal must be patient flow.

If such a condition is obtained, the circuit slowly adjusts the predicted discharge and moves it so that the predicted patient flow becomes zero, and the instantaneous flow of the patient at the discharge end becomes the desired value.
Adjustments to the predicted emissions are obtained as an output of circuit 74, as indicated by reference numeral 80. If this control mechanism works,
The correction performance of the integration circuit 60 for each breath can be limited to the breath.

The output of the circuit 74 is a time constant control signal to the low pass filter 38 ', which temporarily modifies the time constant so that there is sufficient time for the instantaneous discharge to approach the instantaneous flow signal. shorten. Reducing the time constant of the low-pass filter is
The goal is to speed up the low pass filter output (system average) to approach the instantaneous flow signal input.

Another component of the predicted discharge flow control is a gloss error detector 82. The detector operates when the predicted patient flow indicated by reference numeral 84 does not show zero for about 5 seconds. Such conditions are common, for example, when the airflow generator 14 starts operating before the mask is worn on the patient. This part of the system is used to quickly stabilize operation when large changes in emissions occur.

According to the above description, the low-pass filter 38 'works so that the instantaneous flow signal becomes an output corresponding to the system flow value which is the system discharge. The reason system discharge is system flow is that patient inhalation and exhalation over time will have a positive zero flow. As described above, the system average value is a predicted value of the discharge flow rate.

The differential amplifier 52 processes the instantaneous flow signal and the instantaneous discharge flow signal to produce a predicted patient flow signal.
This is integrated, and the non-zero value of the integration is fed back to circuit 38 'to adjust the expected exhaled flow signal on a breath-by-breath basis. Integrator 60 is reset by an IPAP / EPAP status signal via line 62.

Two circuits are provided for nullifying the integration circuit, and the discharge end detector 74 generates an output for adjusting the time constant of the low-pass filter 38 ', as indicated by reference numeral 86. , The integrator 60 is reset. The gross error detector 82 processes the predicted patient flow and adjusts the predicted discharge flow under the conditions described above. The output of circuit 82 is reference
6 is used as the integrator reset signal. Integrator 60 is reset with each breath of the patient if it is disabled by circuit 74 or 82 during one breath of the patient. Therefore, a multi-reset function of the integrator 60 is required.

In operation, the system is used in natural trigger mode, natural / timed mode, or simply timed mode. In the natural mode, the decision circuit 34 continuously compares the instantaneous flow rate with the predicted discharge flow rate. The system is EPAP
In the mode, the instantaneous flow rate is about 40 CC /
It remains as it is until just over a second. If a conversion occurs, decision circuit 34 triggers the system into IPAP mode for 150 milliseconds. Thereafter, the system is typically in IPAP mode, increasing the flow rate to the patient during inhalation due to the patient's natural activity and maintaining the increased IPAP pressure.

After conversion to IPAP mode with each breath, a temporary offset is added to the predicted exhalation flow reference signal. The offset is proportional to the integral of the predicted patient flow at the beginning of the inhalation breath, and thus gradually increases over time in proportion to the patient's inhalation flow. Thus, flow levels above the expected discharge flow required to hold the system in IPAP mode during inhalation will decrease over time from the beginning of inhalation, in proportion to the inhalation flow. This increases the instantaneous flow reference signal that must be reduced to trigger the EPAP mode as the inhalation cycle lengthens. For example, if the patient inhales at a constant value of 500 CC / s until the end of inhalation, the conversion to EPAP will result in a patient flow rate of about 1
When it drops to 67 CC / sec, when it drops to 333 CC / sec after 2 seconds, when it drops to 500 CC / sec after 3 seconds ...
Occur. If the patient's inhalation is 250 CC / sec, the triggers are 1, 3, and 3 sec IPAP, 83, 167 and 2
Occurs at 50 cc / sec.

As described above, when the threshold value of the EPAP trigger matches the suction flow rate, the following advantages are obtained. First,
It is easy to end the inhalation cycle with increasing time. Next, when the system is evacuated to increase the instantaneous flow rate enough to trigger the system into IPAP mode, the system automatically turns on the EP after 3.0 seconds regardless of the patient's breathing.
Return to AP mode. This activates the emission correction circuit (ie, integrator 60) based on the capacity with each conversion to IPAP mode. In this way, if the ejection occurs suddenly, during a few breaths it will go from natural action to an automatic trigger. The circuit is not locked in IPAP mode.

Returning to the EPAP mode, the trigger threshold is about 5
It is kept above the expected discharge flow rate for 00 milliseconds, thereby keeping the system stable in EPAP mode without any conversion as the flow rate changes. After 500 ms, the trigger threshold offset is reset to zero and awaits the next inhalation activity.

The normal state of the circuit remains in EPAP mode until inhalation activity is performed by the patient. Automatic correction and adjustment to the reference signal is effective to keep the system locked in IPAP mode, while preventing automatic triggering. It is very sensitive to adjustments that change the inhalation activity and exhalation state and inhalation pattern.

In the natural / timed mode, the system operates exactly as described for natural operation, except that a minimum respiration rate can be selected in addition to natural operation. If the patient does not perform an inhalation activity within the allotted time, the system will automatically trigger into IPAP mode for 200 milliseconds. The starting pressure increased by 200 milliseconds provides sufficient time for the patient's automatic flow to exceed the reference signal to begin patient inhalation and operate in the aforementioned natural mode for the remainder of the cycle. The respiratory-per-minute timer is set so that the IP
Reset to AP.

In the timed mode of operation, the trigger between IPAP and EPAP is controlled by a timer, and breath-per-minute control is used to select the desired breathing rate, for example, from 3-30 breaths / minute. If possible, the chosen respiration rate should be coordinated with the patient's natural respiration rate. In partial IPAP control, it is used to set the portion of each breathing cycle used to the IPAP mode. For example, breathing control every minute is 10 breaths /
If set to minutes (6 seconds / breath) and partial IPAP control is 33%, the airflow generator will have 2 seconds IPAP and 4 seconds EPAP in each breath cycle.

FIG. 4 shows a control panel for controlling the above-mentioned system, and a function for setting the three operation modes of the above-mentioned natural / timed and timed operation modes.
A selector switch is included. Control switches for the natural mode include IPAP and EPAP pressure adjustment switches 90 and 92. This is IPAP and EPAP respectively
Used to establish pressure levels.

In the natural / timed mode, the control switches 90 and 92 are used to set the IPAP and EPAP levels, and the minute rate switch 94 sets the minute rate to the minimum desired rate. Used for In the timed mode, control switches 90, 92 and 94 are active, and partial IAP control switch 96 is used to establish the time percentage of each breath used in the IPAP mode.

Indicators by light emitting diodes 96, 98 and 100 are used to indicate whether the system is in the IPAP or EPAP state, and in natural / timed mode, to indicate whether the system is in natural or timed mode.

In another embodiment of the present invention, the start of inhalation and discharge is detected by referring to the flow rate waveform of the patient as shown in FIG. The comparison between the instantaneous flow rate and the above average flow rate is sufficient as a method for determining whether the patient is in the inhalation state or the exhalation state. However, other methods of determining the instantaneous flow rate can be used. These can be used independently or in conjunction with the average flow rate.

FIG. 5 shows a typical patient flow rate waveform. In this figure, the suction is shown as a positive flow above the basic line B, and the discharge is shown as a negative flow below the basic line B. The flow waveform is sampled at discrete time intervals. The current sample value is compared to that obtained earlier. This method provides a higher sensitivity of the patient's respiratory movements, as described above, so that the error in the predicted system leakage is less.

[0069] Typically, the predicted inhalation and exhalation flow rate waveforms begin after a few hundred milliseconds in each of the inhalation and exhalation phases and change slowly over a period of time to approximately the end of each phase. The current sample value of the flow rate waveform is taken periodically. The current sample value is repeatedly compared with the previous sample value. During inhalation, if the magnitude of the current sample value is less than some fractional value of the comparison sample value, the inhalation phase has been terminated. This state is used as a trigger to change to the desired discharge phase pressure. The same operation is used during the dispense period, thereby providing a trigger condition for a change to suction pressure.

FIG. 6 shows a flow for repeatedly sampling the patient flow rate waveform at time interval Δt as shown at 110 and a previous sample value such as the flow at time t-4 as shown at 112 and the current time. FIG. 11 is a flowchart of an appropriate algorithm showing a flow of repeated comparison with a value at a time. Of course, the time increase between successive samplings is small enough that comparison with the values in the fourth previous sample covers a reasonably small portion of the flow rate waveform. For example, as shown in FIG. 5, the flow value at time t, the previous four sample values at time t-4, and the time interval Δt describe a suitable proportional relationship between Δt and the time period of patient respiration. ing. However, this is only an example. In fact, electronics such as those described above can achieve this algorithm using smaller Δt intervals. In that case, other parameters of the algorithm such as the comparison sample and / or the proportionality between the compared samples are adjusted.

Referring to the flow diagram of FIG. 6, if the patient's respiratory condition is inhalation (114), the current flow sample value is compared to the previous sample value at t-4. If the current flow sample value is, for example, 75% of the comparison flow value
If less (116), the system triggers a state change to the dispense phase (118). If the current flow sample value is not less than 75% of the comparison value (12
0), no change in respiratory status is triggered.

If the respiratory condition is not inhalation and the current flow sample value is less than 75% of the selected comparison value (12
2), the discharge phase is completed rather than the suction phase,
The system triggers to change the state of breathing to inhalation (124). If it does not match the specified flow conditions, no change in the respiratory phase is triggered (1
26).

The routine of FIG. 6 is repeated continuously, as indicated at 128, and the system checks each time the current flow sample value is compared to the fourth (or other suitable) previous sample value. It is determined whether one breathing phase should trigger from another.

Of course, the embodiments described with reference to FIGS. 5 and 6 do not trigger breathing changes within the scope of the invention as described herein. Rather, FIG.
And the embodiment of FIG. 6 simply detects the change between inhalation and exhalation in the instantaneous breathing of the patient and triggers the system to deliver the specified high and low airway pressures as described above.

A common problem based on insufficient ventilator trigger sensitivity is that the ventilator cannot be triggered to IPAP upon patient inhalation, and the ventilator cannot be triggered off or EPAP at the end of patient inhalation. The patient's respiratory behavior primarily relates to the inhalation part of the respiratory cycle. This is because typically only the suction part of the cycle is active. The exhalation part of the respiratory cycle is generally passive. Another feature of the breathing cycle is that the exhalation flow rate typically approaches zero and is momentarily held to zero before the next inhalation phase begins. Thus, the stable state of the ventilator triggered by the patient is the exhalation state, which is herein EPAP.

While the above described device provides a sensitive trigger between IPAP and EPAP modes, the trigger can be further improved. The basic purpose of improving trigger sensitivity is to reduce the features of the system that can automatically trigger from EPAP to IPAP in the absence of patient inhalation, and to increase system sensitivity at the end of patient inhalation It is.
These are not contradictory objectives. Decreasing the automatic trigger sensitivity does not necessarily reduce the trigger sensitivity for the patient.

In general, for a patient whose ventilator is operating normally, the ideal ventilator trigger sensitivity is the change in state during the patient's breathing (between inhalation and exhalation) and the corresponding state change in the ventilator. Can be represented by close synchronization between Many conventional pressure support vents that trigger based on a particular flow or pressure threshold can approach such close synchronization only under certain limited conditions.

In the above-described system where the trigger is based on a particular level of predicted patient flow rate, close synchronization between system state changes and patient respiratory state changes is readily obtained. This is especially true in such patients, since the flow rate reversal of the patient without discharge limitation is sufficient to accommodate changes in the patient's breathing behavior. However, flow rate reversal cannot accurately accommodate changes in the respiratory behavior of all patients. For example, in patients experiencing dyspnea or using a pressure-supported ventilator, there is no close correspondence desired between patient activity and reversal of gas supply rate. An improvement of the synchronized trigger based on the flow rate allows for at least three different breathing cycles. The first cycle is the patient respiratory cycle. If the patient performs any action, an inhalation action is included. The patient's ejection is active or active.

The second cycle is the ventilator breathing cycle, the cycle created by the ventilator. This cycle has an inhalation phase and an exhalation phase, but the ventilator may or may not be synchronized with the corresponding phase of the patient breathing cycle.

The third cycle is a respiratory flow cycle indicated by the direction of flow to and from the patient's airways. Flow flows from the ventilator to the patient during inhalation and from the patient during exhalation. The suction / discharge change is indicated by the zero crossing of the flow between positive and negative flow.

If the patient's respiratory drive is active, the respiratory flow cycle is synchronized with the patient's respiratory cycle with the help of a ventilator. An important purpose of the trigger described here is to synchronize the cycle of the ventilator with the cycle of the patient. This allows the flow cycle to correspond to the patient's inhalation action. Due to the flow trigger, in addition to the effects of the patient's own movement, the pressure from the ventilator affects this flow. This is because the flow zero crossings do not necessarily represent a change in patient motion.

As mentioned above, flow triggering has certain effects independent of some other modes of ventilator triggering. For example, in the case of a ventilator that uses a pressure change for the trigger, when the patient performs an inhalation operation, the lung volume begins to increase. This change in lung volume results in a pressure drop and is detected by a ventilator trigger system to indicate respiratory gas delivery in a conventional closed circuit ventilator. The flow trigger eliminates the need for a pressure drop in the breathing circuit. The trigger algorithm still requires an increase in lung volume, but since the pressure in the ventilator does not drop, the patient can achieve a volume change with less energy during the inhalation operation.

Furthermore, in an open circuit where the exhaled gas flows out of the system, it is difficult to generate the desired pressure drop due to the pressure based on the trigger due to the small inhalation of the patient. That is, in an open circuit, the magnitude of patient movement required to produce a very small pressure drop must be considered. Thus, when used with an open circuit, it is very difficult to achieve sufficient sensitivity with a pressure-based trigger system to obtain a reliable trigger. Generally, open circuits are simpler than those having separate exhaust valves and exhaust pipes, are easier to use, and have lower costs. For this reason or for other reasons, it is better than a closed circuit.

In what appears to be presently good for the above-described flow rate trigger system, the flow rate signal will have a threshold equal to 40 CC / s continuously for 35 milliseconds to trigger the system to IPAP. Must exceed. If the flow rate signal falls below the threshold during a specified time interval, the threshold timer is reset to zero. These thresholds can be triggered by a minimum patient change of only 1.4 CC. This is a greatly improved sensitivity over other conventional pressure trigger vents.

However, in practice, such a minimum capacitance change becomes an overly sensitive trigger and is not practical. The noise in the flow rate signal is that the patient's flow rate is 40 CC
If so, the threshold timer is continuously reset during that period. Therefore, for a consistent trigger of the system described above, the minimum volume change is about 3
Although it is CC, this value is also a very sensitive trigger level. However, this pressure signal noise also appears in conventional pressure trigger vents, causing trigger sensitivity abnormalities. Because of the above and other operations, the theoretical maximum sensitivity of such a conventional ventilator must be better than the sensitivity actually achievable during operation.

As an example of an operation that adversely affects sensitivity in a conventional ventilator, when the ventilator response time is slow, the time when the trigger time requirement is satisfied and the gas supply pressure begin to substantially increase. The time lag between the time of day has an adverse effect on the sensitivity. During the time delay, the patient is forced to continue the inhalation operation, thereby performing much more than is necessary to trigger the ventilator. More action is needed if the ventilator responds faster. Another example occurs during a fast breathing period if the patient initiates an inhalation operation before the exhalation flow rate drops to zero. In this case, there is a further delay until the flow reverses and it is time for the discharge valve to close.

There is also a problem with triggering EPAP on discharge. A typical prior art pressure support ventilator triggers off at the end of inhalation based on, for example, patient flow to decrease the flow threshold or timing from a timer. The flow threshold for triggering on dispense can be a fixed value or a fraction of the peak flow, for example, 25% of the peak flow. One example of a problem affecting triggering of discharge in conventional ventilators is the use of moderately heavy COPD.
(Chronic Obstructive Pulmonary Disease) Patients have relatively high inhalation system resistance and possibly low inhalation system elastance.
This results in a relatively long time constant for the patient. The time constant is the time it takes for the discharge flow rate to decay to 1/4 of the peak flow rate or to a suitably selected partial value of the peak flow rate. The long time constant of a COPD patient means that the flow rate decays slowly during the inhalation period.

The only way the patient can trigger the ventilator to dispense is to increase the dispense action. That is, the patient must aggressively reduce the flow rate by respiratory muscle movement to trigger an exhalation. In addition, due to the patient's long time constant, there is considerable exhalation when the patient's inhalation cycle begins, which causes
The trigger to the air is delayed, and the sensitivity of the ventilation device is deteriorated.

To increase the sensitivity to patient movement in an improved embodiment of the present invention, the system includes:
Rather than monitoring only a single state, as represented by the description above with respect to component 82, there is provided a device that continuously monitors multiple states. In an improved embodiment, the inhalation time is limited to no longer than about 3 seconds under normal conditions, and the normal delivery flow rate is required to be approximately zero about 5 seconds after the start of delivery. ing.

To use the first of these conditions, the system monitors the predicted patient flow and is required to zero its value at least approximately once every 5 seconds. To monitor this condition, an output from a comparator that compares the patient's flow rate to zero is needed. The comparator outputs one output when the patient flow is greater than zero (eg,
A bistable device that provides the other output (eg, a logical “0” value) when the patient flow is less than or equal to zero. If the comparator output is true or false for more than about 5 seconds continuously, it generates a signal that triggers the system to EPAP.

The second condition is to monitor the IPAP / EPAP status to see if the EPAP is more than about 5 seconds.
If so, the capacity is kept at zero until there is another valid trigger for IPAP. This feature prevents the displacement of the volume signal from zero when apnea. After 5 seconds in EPAP, the capacity is normally zero.

Further improvements in trigger sensitivity require further consideration of flow rate levels to determine an appropriate trigger threshold. In particular, altering the shape or slope of the patient's flow rate waveform is used to better synchronize the ventilation system with the patient's breathing motion. FIG.
And with reference to FIG.

A representative flow rate waveform of a patient's breathing in a relaxed state is shown in FIG. The simplest approach to synchronizing the ventilator trigger to the patient's instantaneous breath is to use the zero flow point 129 as a reference for the trigger between EPAP and IPAP. However,
This approach does not address potential issues. As mentioned above, firstly, the flow rate reversal does not necessarily correspond to changes in the instantaneous breathing behavior of all patients. Secondly,
The flow rate signal is not necessarily uniform. Particularly in connection with this problem, small flow changes corresponding to noise in the electronic signal occur due to random fluctuations in the system and airway pressure caused by flow disturbances and random fluctuations in the pressure control device. The additional flow rate noise is due to small oscillations in the respiratory gas flow rate due to changes in blood volume in the patient's thoracic cavity that occur with the heartbeat. Third, the pressure support is in an all-or-nothing mode of ventilation, and the respiratory system of all patients has an elastance element, which results in improper IPAP.
A trigger to the patient results in the generation of a volume of respiratory gas if the patient's airway is not completely obstructed. In this way, a zero flow or flow reversal for the trigger often results in a failure of synchronization between the patient's inhalation action and the supply of breathing gas from the ventilator.

Referring to FIG. 7, there is shown a flow rate versus time trajectory for a respiratory flow rate of a person who instantaneously breathes. Time progresses from left to right along the horizontal axis. The flow rate changes on the vertical axis around the zero-valued base line. Of course, in the following description with reference to FIG. 7, the illustrated flow rate versus time trajectory is an analog value of the instantaneous breathing by the patient, and may be a mechanical, electronic or other means of detecting patient flow as a function of time. Obtained from detectors or sensors by means.

In FIG. 7, when the inhalation operation of the patient starts or ends, a relatively sharp disconnection or inclination change appears. As described above, the inclination of the flow rate trajectory greatly varies from patient to patient depending on the resistance value and the elastance value of the patient, but for a certain patient, it is relatively constant between inhalation and discharge. A relatively steep increase in the flow rate slope when the inhalation operation begins and a corresponding relatively steep decrease at the end of the inhalation operation can be used to identify IPAP and EPAP trigger points. The abrupt change in the flow rate slope at the beginning and end of inhalation corresponds to the beginning and end of the respiratory movement, so that the flow rate slope or shape may be a function of patient movement rather than a relatively fixed flow threshold. IP based on changes in
Used to trigger the system between the AP state and the EPAP state.

The basic algorithm of operation of the improved trigger system will now be described. The input to the system can be any point on the flow trajectory, such as point 13 in FIG.
The current flow rate at zero. The flow values are differentiated to obtain the corresponding current slope 132 of the flow function at time T. Of course, the slope corresponds to the instantaneous rate of change of the flow rate.

The slope 132 is measured by a time factor Δt and is then added to the current flow value. This allows
The subsequent time t based on the current slope 132 of the flow trajectory
A predicted flow rate 134 at + Δt is obtained. As already mentioned, the magnitude of the time interval Δt is determined by the selected scale factor at which the slope 132 was measured. The resulting predicted flow rate is assumed to be a uniform change in flow rate over the period At. In FIG. 7, the time interval Δt
The size of is exaggerated and, for simplicity,
The end of inhalation and the beginning of inhalation are different. In practice, Δt is, for example, approximately 300 milliseconds.

The accuracy of the predicted flow rate 134 in the near future is greater than the estimated uniform slope 132 by the time interval Δ
It depends on the changing slope of the flow trajectory during t. To the extent that the flow trajectory deviates from the slope 132 during the time interval Δt, the actual flow value 136 at the end of the time interval Δt
Deviates from the predicted flow rate value 134. Thus, the effectiveness of this algorithm depends on the transition from the slope 132 of the actual flow trajectory during a time interval Δt, but this transition is small as long as the patient does not undergo a sudden change in respiratory activity.

In order to further change the predicted flow rate 134,
The offset factor is added and the offset predicted value 1
38 are obtained. The offset is in the negative direction during inhalation. Thus, the predicted flow rate value at time t + Δt is
Except for the case where the flow trajectory changes in the direction of decreasing the flow rate, it is smaller than the actual flow rate value. The change in decreasing flow rate occurs at the end of inhalation when the patient stops moving or moves in the opposite direction.

The above description refers to the trigger at the end of inhalation into EPAP. As shown at 140 in FIG.
In exactly the same way, IPAP can be triggered by predicting the near future flow rate. That is, the differentiation of the instantaneous flow rate 139 and the application of the offset factor make the predicted flow rate 1
42 is obtained. Since the flow rate increases rapidly,
The offset factor in this case is a positive offset. In this case, for example, as shown as 144,
If the actual flow rate increases sufficiently beyond the predicted value 142 during the time period Δt, the system will trigger an IPAP. The actual flow rate after the predetermined time interval Δt trajectory is equal to the time T
At any point in the flow cycle that is significantly different from the predicted and offset flow value based on the flow trajectory at, based on whether the actual flow changes from the predicted flow to a positive or negative direction, the system will Triggered by EPAP.

The flow rate prediction based on the current flow rate trajectory slope and the comparison between the predicted flow rate and the actual flow rate are repeated at high speed, and the flow comparison based on the actual flow rate versus the predicted flow rate is continuous or almost continuous. A flow is obtained. For example, in an analog system the process is continuous, while in a digital system the process repeats every 10 milliseconds or faster. In the trajectory obtained from the predicted flow point, the actual flow rate does not significantly deviate from the predicted rate, so that the flow comparison of the actual flow rate to the predicted flow rate does not lead to a trigger. This only happens in the case of a suction / discharge change. Because it is based on flow prediction, the triggering method does not require a change in the patient's inhalation to achieve a trigger synchronized with the patient's instantaneous breathing. of course,
The described flow prediction differentiation technique is only one example of a suitable flow rate waveform triggering algorithm.

An analog circuit implementing the improved trigger system described above is shown in FIG. Input signal 1
46 estimates the patient flow, but may also be the total flow. According to another aspect of the present invention, as described above, the use of the estimated flow rate allows for further improved triggering if the system includes unknown leak elements. Input signal 14
6 is differentiated in an inverting differentiator 148. The group consisting of diodes, switches and operational amplifiers is shown as 150. These form a diode of switchable polarity to clip large positive displacements of the flow trajectory at the start of inhalation and large negative displacements at the start of discharge. These transitions are usually large, especially when ventilator pressure changes occur. If the large transition is not clipped, it will adversely affect the trigger circuit operation early in each respiratory phase.

The differentiated input signal is supplied to the feedback resistor 152
And the input capacitor 154. These are the delay times in the delay portion of the circuit, 300
It has a time constant equal to milliseconds. Series input resistor 1
56 limits high frequency noise outside the respiratory frequency band. The differentiated input signal is subtracted from the flow signal by the difference amplifier 172 because the output of the transfer circuit is inverted and a sum is required.

The required delay is created using two fifth-order switched-capacitor Bessel low-pass filters 158. Bessel filters have a linear phase shift for frequencies that have the effect of a time delay. Since the time delay depends on the order of the filter as well as the frequency response, the two filter sections need to provide a sufficiently high cutoff frequency for a 300 ms time delay. The 300 ms delay is determined experimentally. This value is appropriate when the time constant of the respiratory muscle activity is 50 to 100 milliseconds. Thus, a delay greater than 50-100 milliseconds is necessary to capture flow changes caused by changes in respiratory muscle activity.

The 555 type timer chip 160 has 1 KH
Set as z oscillator. R at the filter input
Along C-coupling 174, element 160 controls the cut-off frequency of switched capacitor filter 158. The comparator 162 with hysteresis, together with the input resistor 176 and the feedback resistors 166 and 164, changes its state depending on the difference between the input flow rate signal and the processed flow rate signal signal. There is a fixed hysteresis during the suction phase by the resistor 166. The hysteresis characteristics of the described circuit provide a negative offset during inhalation and a positive offset during exhalation.

During the discharge phase, the hysteresis is initially greater than during the inhalation, with the resistor 164 and the diode 1
78 depending on the magnitude of the flow signal N set.
An inverting amplifier 18 is provided to saturate when the flow signal is greater than 0.5 volts, which corresponds to a flow rate of 30 liters / minute.
0 has a gain factor of 10. Therefore, the hysteresis is doubled during the first part of the dispense phase with the 100K resistor 164 in parallel with the 82K resistor 166. In contrast, during inhalation, the 82K resistor 166 operates alone. That is, during discharge, the output of inverter 180 is positive, so that diode 178 is forward biased and resistor 164 in addition to the supply by resistor 176.
Supply current through.

Once the flow signal drops below 0.5 volts, the hysteresis decreases linearly with the flow rate, as during inhalation. This feature prevents premature triggering of the inhalation and gradually increases the sensitivity of the system towards the end of the discharge. As the flow signal is provided through resistor 164, the hysteresis is somewhat dependent on the flow signal. The described elements (82K feedback resistor 166,
With a 1,5K input resistor 156) and a +5 volt system power supply (not shown), the hysteresis is about 90 millivolts at zero flow. This corresponds to a trigger offset of 90 CC / s on the instantaneous flow signal.

FIG. 9 shows the operation of the flow rate locus waveform trigger circuit. The thick line 168 shows the flow rate signal at the-input, and the thin line 1
Reference numeral 70 denotes a flow signal at the + input to the system comparator 162. Line 170 shows that the hysteresis decreases as the flow rate approaches zero during the dispense period.

The trigger algorithm just described is referred to as a shape trigger (ie, in response to a change in the shape or slope of the flow trajectory), as distinguished from the flow trigger previously described. FIG. 10 (a) and FIG.
0 (b) illustrates two ventilator trigger algorithms used in simulating a naturally breathing patient with a large respiration time constant. FIG. 10 (a)
And in each of FIG. 10 (b), the top trace shows the estimated flow rate, the middle trace shows the patient's respiratory movement, and the bottom trace shows the ventilator generated by the mask facing the patient's airway. Shows pressure. The letter I indicates the patient's inhalation activity in each respiratory cycle.

FIG. 10 (a) illustrates the triggering problem described above and known to occur in actual patients.
The ventilator is not triggered in coincidence with the patient's inhalation motion, as shown as a discrepancy between patient motion and estimated flow and pressure at the mask. In contrast, FIG. 10 (b) shows a similarly simulated ventilator response for a patient using the trigger algorithm of the shape described above, without changing other settings. A comparison between FIG. 10 (a) and FIG. 10 (b) shows that the synchronization of the pressures was improved by natural patient movement.

The simulated patient from which FIGS. 10 (a) and 10 (b) are based shows a large discharge flow at the beginning of the inhalation operation. From FIG. 10 (b), it can be seen that the shape trigger algorithm properly triggers at the beginning and end of each suction operation, even when the suction operation begins while there is still a discharge flow. In this way,
The shape trigger algorithm can trigger the ventilator synchronously with patient activity.

FIGS. 11 (a) and 11 (b) are similar to FIGS. 10 (a) and 10 (b), respectively, except that the simulated patient has a shorter respiratory system time constant. I have. FIG. 11 (a) shows from top to bottom,
Fig. 7 shows estimated patient flow, patient breathing activity, and pressure at the mask for a trigger algorithm. FIG. 11B shows the same parameters as those for the flow rate shape trigger algorithm. From FIG. 11 (a),
It can be seen that even a substantial decrease in flow when the patient stops working does not trigger into EPAP until the decrease in flow is below 25% of peak flow. In this way, the end of inhalation is not synchronized with the end of the patient's inhalation operation. In contrast, the shape trigger algorithm of FIG. 11 (b) produces a trigger that is solidly synchronized at the end of the patient's inhalation motion for the same patient.

As can be seen from the above description, the ventilation device
The trigger can be based on a change in the shape of the flow signal corresponding to a change in the patient's breathing motion. This trigger algorithm solves the onerous problem that limits the use of conventional trigger algorithms, such as standard pressure support triggers. While the flow signal shape trigger algorithm described above operates reliably with system leak estimation, the shape trigger algorithm does not respond to system leak calculations.

The shape trigger algorithm described above uses IPAP
It works to trigger both into and out of EPAP, but it works best for triggering into EPAP, since usually larger and more abrupt changes in patient breathing occur at the end than at the beginning of inhalation.

Further, the shape trigger algorithm described above can be used in parallel with the flow trigger algorithm described above to provide a dual trigger system. This allows the secondary trigger algorithm to:
It can function when the main algorithm fails. This type of dual trigger system allows for continuous breath-by-breath operation, i.e. the secondary trigger algorithm triggers the ventilator at the desired trigger point without the primary trigger algorithm triggering I do.

Further improvements in the leakage compensation technique can be realized by the present invention. As mentioned in the description of leakage compensation, the described algorithm depends on the following two requirements. (1) the patient's inhalation and exhalation volumes are the same (the patient's remaining volume just before the start of inhalation is the same for each breath, and the inhalation and exhalation volumes for individual breaths are also the same); ( 2) When the patient inhales, the total flow is greater than the leakage flow, and when the patient exhales, the total flow is less than the leakage flow.

In the above description of the leak, only the total flow is actually measured, even if the total circuit flow of the patient consists of two components, the patient flow and the leak flow. further,
In the above description of the leak, the leak is not estimated as a function of pressure. Also, the average leak is calculated by integrating the total flow. Since the patient's inhalation and exhalation volumes are substantially the same, the average flow rate over a complete breathing cycle is equal to the average leak per cycle.

Further, the leak element itself has two elements. That is, there are known leaks of the discharge portion, and one or more unknown inadvertent leaks, such as leaks through the mask, leaks at the connection to the tube, and the like. Unknown or inadvertent leaks are extremely difficult to determine accurately as a function of both pressure and time. Therefore, the value is estimated in a range in which the leak factor needs to be determined, but the leak can be obtained without separating the leak into a known leak and an unknown leak by an important leak management improvement.

Adjust the leak estimate so that the average of the leaks is the same as the average of the actual leaks over the cumulative respiratory cycle, whether or not all leaks can be characterized as a function of time or pressure. The purpose is to do. Of course, since the flow component of a patient with an average total flow is zero, the average of the actual leak is obtained as the average of the total flow.

Assuming that leakage is a function of pressure, such as leakage from WHISPER SWIVEL ™, leakage at EPAP is
Can be assumed with an average value corresponding to the pressure between EPAP and IPAP which is smaller than that at. In other words, for leaks that are a function of pressure, the leak rate at low pressure is less than the leak rate at high pressure. This property of leakage flow can be used in algorithms more than those that compensate for system leakage.

In all the systems described, the purpose of determining a leak is to determine the patient flow in an open system. Closed systems are ideally leak-free, but in practice there are inadvertent leaks. However,
In an open system, the gas flow from the system is an important part of the total system flow, depending on whether there is a regulated leak at the discharge point or an inadvertent leak from the tubing connections and seals. Thus, the leak in the open system patient circuit can be calculated using the total flow detected by the pneumotach. The parameter referred to as the raw flow rate is the gas flow source (eg,
Blower). As such, raw flow includes both patient flow and system leaks. The purpose of the described leak management algorithm is to separate the raw flow into patient flow and leak components.

It is not necessary to directly measure the patient flow or the leak flow of the raw flow, but as a result the results of the leak calculation should be considered as estimates. One mode of leak management has been described above with reference to FIGS. The following algorithm description is another approach to leak management.

A pressure-independent leak is determined when inhalation of the patient begins. This point of the patient's breathing cycle is referred to as the breath trigger and is used as a reference point for comparing the patient's inhalation and exhalation volumes. By using the beginning of inhalation as the respiration trigger point, the leak calculation is performed using the patient's starting lung volume at approximately the same level for each calculation.

The respiration trigger reference point is determined by satisfying the following two conditions. (1) the raw flow rate is greater than the average leak rate; and (2) the patient is ready for inhalation. For the first of these conditions, a comparison between the raw flow rate and the average leak rate indicates whether the patient is inhaling or exhaling. This is because, as noted above, the nature of the raw flow element is that the raw flow exceeds the system leak during patient inhalation and is less than the system leak during patient exhalation.

The second condition for determining whether the patient is ready for inhalation is determined with reference to one of the following two conditions. (1) The patient discharges a predetermined amount of the discharge volume, for example, 1/4 of the discharge volume, or (2) the patient discharges for a predetermined time, for example, 300 ms or more. The first alternative to these conditions is
The patient's exhalation volume is at least an important part of the inhalation volume. For the second condition, the raw flow rate during discharge is less than the estimated average leak rate,
Discharge times greater than or equal to 0 milliseconds can be detected by comparing the unprocessed flow rate with the estimated average leak rate. If the unprocessed flow rate is less than 300 milliseconds below the average leak rate, it is consistent with that condition. If either of conditions 1 and 2 just described is obtained, the patient is ready for inhalation.
Thus, if any of the conditions are met and the raw flow rate is greater than the average leak, the condition of the respiratory trigger is satisfied and the described condition can be used to trigger the ventilator in the respiratory cycle.

The raw flow rate greater than the average leak rate to satisfy one respiratory trigger condition, and the raw flow rate less than the average leak rate as one parameter that can satisfy the second respiratory trigger condition. The apparent contradiction of what is needed is explained next. A condition in which the raw flow rate is less than the average leak rate indicates that the patient has initiated an inhalation cycle. That is, the patient knows that the patient has performed an initial inhalation operation by having the raw flow rate exceed the average system leak. However, as described above, the raw flow rate crosses the average leak rate signal due to spurious noise in the raw flow signal, so that a raw flow rate greater than the average leak indicates the start of a patient inhalation operation. Not necessarily. In fact, the noise in the raw flow signal causes multiple crosses between the raw flow signal and the average leakage signal. In most practical situations, the flow will cross zero based on the actual transition from inhalation to exhalation, but a particularly troublesome problem is the presence of noise that causes instantaneous transitions.

To avoid multiple triggers at these points of the signal cross that do not indicate the initial operation of the patient's inhalation, an inhalation readiness is further employed. In this state, if only one of the two conditions, indicating that the patient's breathing was in the last-discharge state, is satisfied, the raw flow rate greater than the average leak condition is enabled. In the second of these situations, a 300 millisecond period in which the raw flow rate is less than the average leak corresponds to a patient ejection period. In this way, the time component plays a role in the described breath trigger algorithm. The raw flow rate less than the average leak provides an inhalation ready output. Changing the raw flow rate to a value greater than the average leak indicates the detection of a patient's inhalation activity. In this way, the breath trigger triggers a steady inhalation cycle in synchrony with the natural patient's breathing.

Upon the occurrence of a breath trigger, the inhalation ready parameter is reset and waits for one of the above two conditions to provide for inhalation ready validation to be satisfied. The breath triggering process is repeated in response to the patient's next inhalation activity appearing once the raw flow exceeds the average system leak.

The time interval between one breath breath trigger and the next breath breath trigger is obtained and the raw flow signal is integrated over the time interval to find the raw or full volume of each breath. The raw volume is divided by the time interval between respiration triggers, for example, between successive respiration trigger pairs. This gives a recent time rate of leakage.

The patient's breathing from one breath trigger to the next, as represented by the integration of the raw flow signal, rises from zero volume to a maximum value consisting of the inspiratory volume and the leak volume during inhalation. . As the breathing cycle continues through exhalation, successive accumulations of negative flow progressively reduce the raw volume parameter as the exhaled volume is subtracted. In this way, at the next breath trigger,
The raw volume parameter from the previous breath is equal only to the leak volume and the patient's remaining volume change. However, the latter value is usually assumed to be zero.

If the only raw volume parameter is actually the leak volume, then the time division of the period of the breath will result in an estimate of the leak rate of the most recent breath. By averaging the recent leak rates over time, the average leak rate is determined and a more stable signal is obtained. This reduces the effects of interbreathing volume changes due to interbreathing changes in the patient's remaining volume. The number of respiratory cycles over which recent leaks are averaged to determine the average leak rate is one based on the balance between signal stability and fast error correction.
To any value between infinity. Long term averaging provides high stability but slows error correction. Averaging over a short period of time is less stable but provides faster error correction.

The average leak rate for which the calculation method was described is first used in an algorithm in one of the states for initializing a breath trigger. The parameters used in initializing the breath trigger are continuously updated and used to satisfy the breath trigger condition. Continued breathing triggers are used to update these parameters sequentially.

An approach to leak analysis involves the use of leak elements that are or are assumed to be a function of patient circuit pressure. This type of leak analysis is described, for example, in U.S. Pat.
It is effective in ventilation management such as the proportional support ventilation device shown in No. 830. In a proportional assist ventilator, it is important to know both the average leakage rate and the instantaneous leakage of the system. However, proportional assist ventilators include pressure variations that depend on the level of ventilatory assistance momentarily required by the patient. Since at least some components of the system leak can be a function of pressure, these factors will change somewhat in synchronization with changes in patient ventilation needs.

Other algorithms for leak analysis that can be used for proportional assist ventilators and other systems deemed appropriate are:
As described above, it is based on the requirement that the patient flow be equal to the raw flow rate less than any known leak element or any pressure dependent leak element. The known leak may be a leak having a known flow characteristic, such as any leak through WHISPER SWIVEL ™ or a plateau-type discharge valve. Known leaks are not necessarily fixed leaks, but can be a function of pressure. Since leaks of known properties can be combined with leaks of unknown properties and are assumed to be a function of pressure, the following analysis can be performed on leaks of known properties.

A pressure-dependent leak, a total leak or a component of a total leak, can be calculated as a function of system pressure from analysis of the raw flow signal. The function in which the pressure dependent leak element is related to pressure is the function for the hole that determines the pressure drop between the pressurized system and the low pressure environment, and the flow through this hole is the pressure difference across the hole multiplied by a constant k. It is proportional to the square root of. This constant is determined by the mechanical characteristics of the hole itself, for example, size, surface smoothness, and the like.

To find a pressure dependent leak, it is only necessary to characterize one hypothetical hole to represent all pressure dependent leak sources. To find a hypothetical hole, the breath trigger defined above is used to mark the beginning and end of the patient's breathing cycle. The average patient circuit pressure is measured for each breath. The average leak rate as calculated above is multiplied by the time of the previous breath to determine a pressure dependent leak volume for each breath. If the pressure dependent leak element is separated from the known leak element, the known leak element is also separated from the average leak rate before the average leak rate is used to determine the pressure dependent leak capacity.

The pressure-dependent leak volume is divided by the root mean square of the pressure of the previous breath to obtain the corresponding hole characteristic k. In the next breath, the pressure dependent leak is obtained by multiplying the calculated hole characteristic k by the instantaneous pressure. Thus, the pressure dependent leak rate is a function of the pressure through the breath. This is an important parameter in ventilation management such as the proportional support ventilation device described above. In addition, it is important for therapists to have reliable information about system leaks. Thus, pressure-dependent leak rates are useful in characterizing leaks in various venting devices.

[0138] For any of the last-mentioned leak calculations, it was assumed that the respiration trigger did not occur in a timely manner due to the inability of the raw flow signal to cross the average leak rate signal during patient breathing. In such a case, a reproduction routine is desired to cope with this. If the system leak changes, the raw flow signal increases as the leak increases, and the raw flow signal decreases as the leak decreases. It is possible to make this increase or decrease sufficiently large that the raw flow signal never crosses the average leakage rate. If this occurs, no respiration trigger occurs and no new or updated leak calculations are performed. To recover, the algorithm needs to start leaking again. This returns the calculated average leak to the line and allows the system to function. The start of the regeneration algorithm depends on whether any one of the four known physiological cases occurs. That is, (1) ejection continues for 5 seconds or more. (2) Inhalation continues for 5 seconds or more. (3) The suction volume is 5 liters or more. (4) The discharge capacity is -1 liter or less.

The ventilator must include an element that monitors the operation of detecting any one occurrence of these cases. When one such case is detected, the leak detection can be determined by adding the average leak rate for the raw flow signal by gradually adding the percentage of the difference between the raw flow rate and the average leak rate to the average leak rate. It is started by changing. Desirably, the percentage of added difference should be chosen to have a time constant of approximately one second.

Since the average leak rate gradually approaches the unprocessed flow rate value by gradually increasing the average leak rate in accordance with the regeneration algorithm, the increasing average leak rate eventually becomes one of the trigger algorithms described above. And the repetition of the leak calculation resumes with each breath cycle.

From the leak management algorithm described above, the algorithm is based on a comparison of the estimated patient flow to zero and calculates only unknown leaks as a function of pressure. In a manner similar to the leak management algorithm described above, a respiratory indicator or trigger is used as a marker to begin inhaling a patient. A respiratory trigger occurs when there are two states. That is, (1) the estimated patient flow is greater than zero. (2) The patient is ready for inhalation.
The inhalation preparation completion state is satisfied in the following cases. (01) The patient discharged a predetermined partial amount (for example, 25%).
(2) The estimated patient flow is less than zero. (3) The suction volume is greater than a constant represented by zero or a small positive value.

In the algorithm described above, the situation that requires the suction volume to be less than a certain percentage of the discharge volume and greater than a small positive value is not incompatible since these measurements are taken at different times. The inhalation readiness condition is satisfied by certain parameters including the detection that the estimated patient flow is less than zero. Thereafter, when the estimated patient flow becomes positive, a breath trigger can occur. This is because an estimated patient flow greater than zero indicates an initial patient inhalation operation.

Raw flow is integrated from one breath trigger to the next, and the square root of pressure is also integrated.
Then, at the next breath trigger, the unknown hole is calculated by dividing the value obtained from the raw flow integration by the value obtained from the square root integration of the pressure, and the characteristic k of the unknown hole is obtained. Can be As described above, the unknown hole is a hypothetical leak source that should account for the nature of the leak from all pressure dependent leak sources.

Once the unknown hole properties are calculated,
Multiplied by the square root of the patient circuit pressure at any time during the next breath, an instantaneous unknown leak calculation is made. The estimated patient flow is obtained by subtracting the instantaneous unknown leak from the raw flow.

The constant k is calculated in substantially the same manner in both algorithms that handle system leaks as gas flows through the hypothetical hole of characteristic k. The formula is different in the two ways, but the result is the same. First, one of the calculation formulas is represented by Formula 2.

[0146]

(Equation 2) In one form of the invention, leakage is calculated as a function of pressure. Therefore, the equation for calculating the constant k can be generally expressed as in equation (1).

As described above, the plateau type valve and the WHISPE
If known leak factors are included in the system, such as a leak due to the use of R SWIVEL ™, the calculations may be changed. This subtracts the known factor from the raw flow before integrating the raw flow to determine the unknown hole and before using the raw flow to determine the estimated patient flow. This results in higher accuracy. In this case, the division by the sum of the raw flow rate minus the known leak and the square root of the patient circuit pressure provides a parameter k characterizing the unknown hole corresponding to the unknown leak only. The hole characteristic k is multiplied by the square root of the patient circuit pressure during the next breathing period to obtain a measure of the unknown leak factor.
Thereafter, the estimated patient flow at any time is obtained by subtracting the known leak and the calculated unknown leak from the raw flow signal.

The modified algorithm also includes a regeneration routine for substantially the same reasons as described above. The breath trigger in this modified algorithm depends on the estimated patient flow crossing zero during patient breathing. As the leak value changes, both the raw flow and the estimated patient flow increase as the leak increases and decrease as the leak decreases. It is possible that this increase or decrease is large enough so that the estimated patient flow never crosses the zero value. If this occurs, no respiration trigger will occur and no new or updated leak calculations will be performed until the regeneration routine runs.

To correct this situation, a regeneration algorithm can be used to detect when the leakage calculation cannot be controlled according to the following two restricted parameters: (1) The inhalation volume is larger than a physiological value, for example, 4.5 liters. (2) Inhalation volume is less than non-physiological values, such as -1 liter. When any of these non-physiological events occur, the calculated leak becomes uncontrollable and the system leak must be obtained by artificially introducing a respiratory trigger. The system requires a detection device to detect certain restricted inspiratory volume values and to provide a breath trigger reference point. Restart the system operation in substantially the same way as described above with reference to the other algorithms and perform a new leak calculation. The data used in these successive calculations includes the data after the change in the amount of leakage so that the estimation of the unknown leakage is improved. Even if the new unknown leak value is fraudulent, subsequent iteration calculations are based on the data after the change in leak amount. In this way, the calculation of the unknown leak estimate is continued to approach a reliable value early.

FIG. 12 illustrates the leak management algorithm in operation, intentionally introducing a very large leak. FIG.
In 2, the four trajectories are, from top to bottom, pressure, estimated patient flow, estimated inhalation volume, and estimated leak (plus known and unknown leaks). The X axis is the passage of time, going from left to right.

As can be seen from FIG. 12, in the first three breathing cycles, the algorithm can determine the leak as a function of pressure. At the start of each inhalation, the unknown hole characteristic k is calculated. Points 1 and 2 on the estimated leak trajectory indicate the transition between one unknown hole characteristic k and the next unknown unknown hole characteristic. Point 2 on the estimated leak trajectory indicates that the point between the old and new hole characteristics is smoother than point 1. Point 2 indicates that there is a transition between unknown hole characteristics of substantially the same value.

To test the algorithm's response to leaks, very large leaks have been introduced, as indicated by point 0 on the estimated patient flow trajectory. Predictively,
The estimated patient flow and the estimated inhalation volume increase very rapidly and scale out. At time T,
The estimated inhalation volume reaches 4.5 liters and a respiration trigger occurs. In the obtained calculation, in the estimated leak portion in FIG. 12, a leak having an extremely large value is estimated as the scale-out occurs. The estimated patient flow obtained is approximately 50 liters /
It is a minute, but it should actually be zero. As indicated by point 3, the inhalation volume continues to increase rapidly, eventually reaching 4,
Generate another breath trigger when the 5 liter limit is reached. Large system leaks are removed at time T ',
This stops the increase in the suction capacity. The leak removed from the system causes the estimated patient flow and estimated inspiratory volume to drop sharply into scale. In this way,
When the inhalation volume reaches the regeneration limit of -1 liter, another breathing trigger occurs, leading to a new leak calculation, resulting in a smaller estimated leak than the previous leak calculation. This allows
The estimated patient flow returns within the scale. The new leak calculation is better than the previous one, but not perfect. With the estimated patient flow based on this new unknown leak hole characteristic, the volume again drops below the -1 liter limit, as shown at point 4, causing another breathing trigger. Once again a new unknown leak hole calculation is performed and used for the unknown leak calculation to get very good values. This returns the system to the normal state of operation. In this way, the system is also reliable for regeneration from very large changes in leakage. Smaller changes in the amount of leakage, which usually occur, can be easily dealt with by this algorithm by ordinary calculation or regeneration calculation.

Of course, the above-described embodiment operates in a good mode, but it is obvious to those skilled in the art that other embodiments and modifications can be implemented within the scope of the present invention. Therefore, the present invention is broad and limited only by the claims.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of a first embodiment of the present invention that operates according to the method of the present invention.

FIG. 2 is a functional block diagram of another embodiment of the present invention.

FIG. 3 is a functional block diagram of a predictive ejection computer in the embodiment shown in FIG. 2;

FIG. 4 is a plan view of the control panel of the present embodiment.

FIG. 5 is a diagram showing a trajectory of a patient flow rate versus time in accordance with another embodiment of the present invention.

FIG. 6 is a flowchart of the embodiment shown in FIG. 5;

FIG. 7 is a diagram illustrating a trajectory of a patient flow rate versus time for explaining another embodiment of the present invention.

FIG. 8 is a circuit diagram of a circuit according to another embodiment of the present invention.

FIG. 9 is a diagram showing a flow rate versus time trajectory showing processed and unprocessed flow rate signals.

FIG. 10 is a diagram illustrating an operation of a control algorithm according to still another embodiment of the present invention.

FIG. 11 is a diagram illustrating the operation of another control algorithm according to still another embodiment of the present invention.

FIG. 12 is a diagram illustrating the operation of still another control algorithm according to the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Therapeutic apparatus 12 Patient 14 Air flow generator 16 Pressure bottle 20 Transmission pipe 22 Mask 24 Exhaust port 26 Pressure controller 30 Flow rate signal adjustment circuit 34 Decision circuit 38 Low-pass filter

──────────────────────────────────────────────────続 き Continuation of the front page (73) Patentee 595170524 1001 Murry Ridge Road, Murrysville, Penn sylvania 15668-8550, US A. (58) Field surveyed (Int. Cl. ⁶ , DB name) A61M 16/00 370 A61M 16/00 311 A61M 16/00 345

Claims (8)

(57) [Claims] 1. An apparatus for detecting a change in a patient's respiratory condition between inhalation and exhalation, comprising: means for monitoring the patient's respiratory gas flow; Means for determining patient respiratory gas flow characteristics;
Means for comparing a preselected pair of said respiratory gas flows;
Means for identifying a change in patient breathing between inhalation and exhalation force from said comparing operation, wherein each of said preselected pair of breathing gas flows comprises a patient breathing gas flow at a predetermined time. In the detection device including the predicted value of and the actual respiratory gas flow rate of the patient at the predetermined time, differentiating the known patient respiratory gas flow rate and determining the change thereof,
A detection device comprising means for processing the transition and the known patient respiratory gas flow to determine the predicted value of the patient respiratory gas flow. 2. The detection apparatus according to claim 1, wherein said processing means further comprises means for measuring said displacement using a measurement factor to obtain the measured displacement. 3. An apparatus for detecting a change in a patient's respiratory state between inhalation and exhalation, comprising: means for monitoring the patient's respiratory gas flow; Means for determining gas flow characteristics; means for comparing a preselected pair of said respiratory gas flows; and means for identifying a change in patient breathing between inhalation and exhalation force from said comparison operation. Wherein each of the preselected pair of respiratory gas flows includes a predicted value of the patient respiratory gas flow at a predetermined time and an actual respiratory gas flow of the patient at the predetermined time. Detection device. 4. The detection device according to claim 3, wherein the predicted value of the patient respiratory gas flow is determined from an actual patient respiratory gas flow at a time before the selected time. apparatus. 5. The apparatus of claim 2, wherein the processing means further comprises means for correcting the measured transition using an offset factor to determine the predicted value of patient respiratory gas flow. A detection device characterized by the above-mentioned. 6. The detection device according to claim 2, wherein the measurement factor includes a time factor. 7. The detection device according to claim 6, wherein the time factor is determined, at least in part, by a time span between the selected time and a time prior to the selected time. A detection device characterized by the above-mentioned. 8. The detection device according to claim 7, wherein said processing means further comprises means for correcting said measured transition using an offset factor to determine said predicted value of patient respiratory gas flow. A detection device characterized by the above-mentioned.