WO2016132279A1

WO2016132279A1 - Enhancement of simultaneous estimation of respiratory parameters via superimposed pressure signal

Info

Publication number: WO2016132279A1
Application number: PCT/IB2016/050807
Authority: WO
Inventors: Dong Wang; Francesco VICARIO; Nikolaos KARAMOLEGKOS; Antonio Albanese; Nicolas Wadih Chbat
Original assignee: Koninklijke Philips N.V.
Priority date: 2015-02-18
Filing date: 2016-02-16
Publication date: 2016-08-25

Abstract

A ventilator system includes a ventilator (10) configured to deliver an air flow at positive pressure to a ventilated patient (12). A signal generator (50) is configured to generate a non-therapeutic pressure signal having amplitude less than the maximum positive pressure of the air flow delivered to the ventilated patient and having the highest frequency component of at least 0.5 Hz. A signal combiner (52) is configured to superimpose the non-therapeutic pressure signal on the air flow delivered to the ventilated patient. A ventilator monitor (44) is programmed to estimate respiratory system' s resistance R and compliance C and respiratory muscle pressure generated by the ventilated patient based on pressure and air flow measurements. The non-therapeutic pressure signal is effective to impart a signal component with frequencies beyond normal respiratory signal frequency range to the measured pressure and air flow, which improves the robustness of the estimation task.

Description

Enhancement Of Simultaneous Estimation Of Respiratory Parameters Via Superimposed Pressure Signal

The following relates to the respiratory therapy arts, respiratory monitoring arts, and related arts.

Various types of respiratory therapy employ a mechanical ventilator. In passive patient therapy, the patient is unable to breathe, and the ventilator operates in a pressure control mode in which the ventilator performs the entire work of breathing (WoB). In active patient therapy, the patient can perform some of the necessary work but cannot meet the respiratory demands independently. Thus, ventilator operates in a pressure support mode to provide sufficient pressure to overcome any deficiency in the patient' s ability to breathe. Volume control modes of ventilator operation are also known, in which flow rate or volume is the controlled parameter, rather than controlling pressure (although pressure limit settings may also be applied to guard against pulmonary barotrauma), and are mainly used for passive patient therapy.

In determining ventilator settings and subsequent monitoring of a mechanically ventilated patient, it may be advantageous to measure various respiratory parameters. In the case of pressure support mode ventilation (PSV), assessment of the patient's work of breathing, which is the clinical parameter commonly used to infer the patient's effort per breath, is facilitated by evaluating the respiratory muscle pressure, P_mus(t), over the breathing cycle. More specifically, the WoB is computed by integrating P_mus(t) over the inhaled volume. For passive patient ventilation, it may be advantageous to verify that P_mus~0 throughout the breathing cycle (indicating no appreciable WoB is being provided by the patient). Respiratory parameters such as respiratory resistance (R) and compliance (C) may also be of interest, or may need to be determined in order to assess other parameters.

Estimating P_mus(t) in support modalities of mechanical ventilation enables the ventilator to be set such that the patient and ventilator share the mechanical work performed on the respiratory system. Quantitative assessment of P_mus(t) can be used to select the appropriate level of ventilation support in order to prevent both atrophy and fatigue of patient's respiratory muscles. The respiratory muscle pressure P_mus (t) is typically assessed by measuring the esophageal pressure (P_es) via insertion of a balloon-tipped catheter in the patient's esophagus. The measured P_es(t) is assumed to be a good proxy for the pleural pressure (P_pi) and can be used, in conjunction with an estimate of chest wall compliance C_cw, to compute the WoB via the so-called Campbell diagram or, equivalently, via explicit computation of P_mus(t) and then of WoB.

Estimates of respiratory R and C are important per se, as they provide quantitative information to the physician about the mechanical properties of the patient's respiratory system and can be used to diagnose respiratory diseases and to select the appropriate ventilation modalities and therapeutic paths. Moreover, R and C can also be used to estimate P_mus(t) as a non-invasive alternative to the use of the esophageal catheter. Assuming R and C are known, P_mus(t) is suitably calculated by the following equation (known as the Equation of Motion of the Lungs):

V(t)

p_y(t) = RV(t) + ~ + P_mus(t) + P_Q (1)

where P_y(t) is the pressure measured at the Y-piece of the ventilator (also known as pressure at the mouth of the patient), V( ) is the flow of air into and out of the patient respiratory system (measured again at the Y-piece), V( ) is the net volume of air delivered to the patient (measured by integrating the flow signal V(t) over time), and P₀ is a constant term to account for the pressure at the end of expiration.

In the case of a passive patient expending no breathing effort, it follows that P_mus(t) = 0 throughout the breathing cycle and Equation (1) reduces to P_y(t) = RV(t) +

+ P_Q. For the passive patient, P_y(t), V(t), and V(t) waveforms are fully determined by the selected ventilator settings and directly measureable, so that it is straightforward to generate a sufficient data set to determine R and C. By contrast, in the case of an active patient who is providing some WoB, the value of P_mus(t) varies with time over the breath cycle, and Equation (1) is not easily solved.

For active patients, Equation (1) has generally been applied to non-invasively estimate P_mus(t) using a two-step approach, where R and C are estimated first and then Equation (1) is applied to compute P_mus(t) using the estimated values of R and C. Estimation of R and C may be performed by applying the flow -interrupter technique (also called End Inspiratory Pause, EIP). However, the flow -interrupter technique has the disadvantage of interfering with the ventilation pattern supplied to the patient. Moreover, the patient's respiratory muscles ought to be fully relaxed during the EIP maneuver in order for the computation of R and C to be valid, which may not always be the case. Another difficulty is that the values for R and C assessed via the EIP maneuver may be different from the R and C values attained during the ventilation pattern for which P_mus(t) is to be determined. The EIP maneuver is performed in a specific ventilation mode (Volume Assisted Control, VAC) and the resulting R and C values might not be representative of the corresponding values that determine the dynamics of the lung mechanics under other ventilation modes, such as PSV, potentially leading to error in the subsequently computed P_mus(t)-

Another approach for estimating R and C in the case of an active patient is to apply least-squares fitting of Equation (1) to flow and pressure measurements under specific conditions for which the term P_mus (t) is assumed to be zero. Some conditions for which P_mus(t) could be assumed to be close to zero include: (1) periods of patient paralysis while Continuous Mandatory Ventilation (CMV) is applied; (2) periods of high Pressure Support Ventilation (PSV) levels; (3) specific portions of every pressure-supported breath that extend both during the inhalation and the exhalation phases; and (4) exhalation portions of pressure- supported breaths, where the flow signal satisfies specific conditions that are indicative of the absence of patient inspiratory efforts. However, Conditions (1) and (2) are undesirable clinical states that cannot be properly induced as an expedient for measuring R and C. The assumption of P_mUs(t)~0 for Condition (3) is questionable, especially during the inhalation phase. Condition (4) provides only a limited amount of data for the least squares fitting procedure. In sum, it has been difficult to attain a clinically useful period of sufficient time duration for which /³ _mus(t)~0 is reliably achieved in an active patient in order to estimate R and C.

The following provides a new and improved system and method which overcome these problems and others.

In accordance with one aspect, a medical ventilator system comprises: a ventilator configured to deliver a therapeutic air flow at positive pressure to a ventilated patient; a signal generator configured to generate a non-therapeutic pressure signal having amplitude less than the maximum positive pressure of the ventilatory pattern delivered to the patient and having a frequency component of at least 0.5 Hz (and more preferably at least 1 Hz, and still more preferably at least 5 Hz); and a signal combiner configured to superimpose the non-therapeutic pressure signal on the therapeutic one, thus altering the air flow delivered to the ventilated patient. The non-therapeutic pressure signal may, for example, be is a sinusoidal signal, a square wave signal, a triangle or saw-tooth wave signal, or a non-cyclical chirp or frequency sweeping signal. In some embodiments the ventilator includes a ventilator compressor configured to deliver the air flow to the ventilated patient; and a ventilator controller comprising a microprocessor programmed to generate and send a control signal to the ventilator compressor in order to deliver pressure support ventilation to the ventilated patient. In such embodiments the signal combiner may be configured to superimpose the non-therapeutic pressure signal on the PSV control signal that is sent to the ventilator compressor, and the signal generator and signal combiner may in narrower embodiments comprise said ventilator controller programmed to generate and superimpose the non-therapeutic pressure signal on the PSV control signal generated by the ventilator controller.

In accordance with another aspect, a non-transitory storage medium is provided, which stores instructions readable and executable by one or more microprocessors of a medical ventilator to command the medical ventilator to perform a method comprising: generating a non-therapeutic pressure signal; and superimposing the non-therapeutic pressure signal on air flow delivered by the medical ventilator to a ventilated patient. The method may further include operating the medical ventilator to deliver the air flow to the ventilated patient under pressure support ventilation (PSV), and the non-therapeutic pressure signal may be superimposed on a PSV control signal that is sent to a ventilator compressor of the medical ventilator. The superimposed non-therapeutic pressure signal is effective to impart signal components to patient pressure and airflow waveforms provided by the medical ventilator. The method may further comprise computing respiratory muscle pressure generated by the ventilated patient during respiration based on at least the patient pressure and airflow measurements acquired by the medical ventilator.

In accordance with another aspect, a medical ventilation method consists of delivering ventilation to a ventilated patient using a medical ventilator, and superimposing a cyclical or non-cyclical non-therapeutic pressure signal with a frequency component higher than the patient respiration rate on the ventilation delivered to the ventilated patient. In some embodiments, the method comprises: delivering pressure support ventilation (PSV) to the ventilated patient; measuring pressure at a Y-piece or T-piece coupled to the ventilated patient using a pressure sensor; measuring air flow at the Y-piece or T-piece coupled to the ventilated patient using a flowmeter; and computing respiratory muscle pressure generated by the ventilated patient during respiration based at least on the measured pressure and air flow. Computing the respiratory muscle pressure may comprise solving a system of equations formed by applying an equation of motion of the lungs to the pressure and air flow measured at a plurality of successive times while the superimposed cyclical non-therapeutic pressure signal has a cycle frequency that is effective to make the data matrix of the system of equations well-conditioned.

One advantage resides in providing reduced noise in respiratory data analysis of a ventilated patient without impacting the ventilation therapy.

Another advantage resides in more accurate estimation of respiratory parameters such as respiratory system' s resistance R and compliance C, respiratory muscle pressure P_mus(t), and Work of Breathing (WoB), especially for (but not limited to) the case of a patient providing some WoB such that P_mus (t) varies over the breath cycle.

Further advantages of the present invention will be appreciated to those of ordinary skill in the art upon reading and understand the following detailed description.

The invention may take form in various components and arrangements of components, and in various steps and arrangements of steps. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention.

FIGURE 1 diagrammatic ally shows a ventilation system.

FIGURE 2 diagrammatically shows a data analysis algorithm disclosed herein which simultaneously estimates multiple respiration parameters by approximating the respiratory muscle pressure P_mus(t) by a low-order polynomial function.

FIGURE 3 plots simulated respiration waveforms over about three breaths, with sensitivity of the parameter matrix plotted in the lowermost plot of FIGURE 3.

FIGURE 4 plots simulated respiration waveforms over about three breaths with a small amplitude, high frequency pressure signal AP(t) superimposed on the ventilator- applied pressure, with sensitivity of the parameter matrix plotted in the lowermost plot of FIGURE 4.

FIGURE 5 plots simulated values for respiratory system's resistance R and compliance C, and respiratory muscle pressure P_mus(t) for the same simulation shown in FIGURE 3, along with values for these parameters estimated from the data of FIGURE 3 using the approach described with reference to FIGURE 2.

FIGURE 6 plots simulated values for respiratory system's resistance R and compliance C, and respiratory muscle pressure P_mus(t) for the same simulation shown in FIGURE 4, along with values for these parameters estimated from the data of FIGURE 4 using the approach described with reference to FIGURE 2. With reference to FIGURE 1, a medical ventilator system includes a medical ventilator 10 that delivers air flow at positive pressure to a patient 12 via an inlet air hose 14. Exhaled air returns to the ventilator 10 via an exhalation air hose 16. A Y-piece 20 of the ventilator system serves to couple air from the discharge end of the inlet air hose 14 to the patient during inhalation and serves to couple exhaled air from the patient into the exhalation air hose 16 during exhalation. Note that the Y-piece is sometimes referred to by other nomenclatures, such as a T-piece 20. Not shown in FIGURE 1 are numerous other ancillary components that may be provided depending upon the respiratory therapy being received by the patient 12. Such ancillary components may include, by way of illustration: an oxygen bottle or other medical-grade oxygen source for delivering a controlled level of oxygen to the air flow (usually controlled by the Fraction of Inspired Oxygen (Fi0₂) ventilator parameter set by the physician or other medical personnel); a humidifier plumbed into the inlet line 14; a nasogastric tube to provide the patient 12 with nourishment; and so forth. The ventilator 10 includes a user interface including, in the illustrative example, a touch- sensitive display component 22 via which the physician, respiratory therapist, or other medical personnel can configure ventilator operation and monitor measured physiological signals and operating parameters of the ventilator 10. Additionally or alternatively, the user interface may include physical user input controls (buttons, dials, switches, et cetera), a keyboard, a mouse, audible alarm device(s), indicator light(s), or so forth.

With continuing reference to FIGURE 1, in an upper portion some additional salient aspects of the ventilator system are diagrammatically illustrated in a block diagram format, including the ventilator 10 represented as a simplified block diagram, and the Y-piece 20 as a diagrammatic box with operative connections indicated by connecting arrows. In the illustrative example, the ventilator 10 is operating in a pressure support ventilation (PSV) mode as implemented by a controller 30. PSV is an appropriate ventilation mode for an active patient who is capable for expending at least some Work of Breathing (WoB) - that is, whose diaphragm and other chest muscles are acting to at least assist in operating the lungs to perform breathing. In the PSV mode, the pressure provided by the ventilator 10 via the inlet air hose 14 operates together with the patient's WoB to perform the breathing. More generally, the controller 30 may implement various ventilation modes depending on the patient's condition and the therapy to be delivered. For example, in the case of a passive patient who is providing no WoB, the controller 30 may operate the ventilator 10 in a Pressure Control Ventilation (PCV) mode. (Note that in some classification schemes PSV is considered a type of PCV mode, since in both PCV and PSV the pressure applied by the ventilator 10 is a controlled parameter). Volume control ventilation modes are also sometimes used, although pressure limit settings may also be applied in volume control ventilation to guard against pulmonary barotrauma. In general, the ventilation controller 30 is implemented as a microprocessor with ancillary electronics such as read-only memory (ROM), electronically erasable read only memory (EEPROM), flash memory, or another non-volatile memory component storing software or firmware executed by the microprocessor, random access memory (RAM) chip(s) to provide working memory, and so forth. If EEPROM, flash memory, or other updatable memory is used to store the software or firmware, then capabilities of the ventilator 10 can advantageously be updated (within the limits of its hardware components) by updating the software or firmware.

The PSV controller 30 outputs a desired pressure control signal as a function of time, which is used to control a ventilator compressor 32 (e.g. a pneumatic pump, turbopump, or so forth) that generates air flow at the controlled positive pressure that is applied to the Y-piece 20 via the inlet air hose 14. Depending upon the respiratory therapy to be provided, an oxygen regulator 34 may add a controlled fraction of oxygen to the air flow to achieve a Fraction of Inspired Oxygen (Fi0₂) set by the physician, respiratory specialist, or other medical personnel who sets the configuration of the ventilator 10 for the patient 12. The pressure of the air flow may vary during the breathing cycle to provide pressure-driven or pressure-assisted inhalation and to reduce pressure to facilitate exhalation.

The ventilator system typically further includes physiological monitoring sensors, such as an illustrative pressure sensor 40 and an illustrative flowmeter 42. The pressure sensor 40 measures the pressure at the Y-piece 20 (also known as pressure at the mouth of the patient), which is denoted here as P_y(t). The flowmeter 42 measures the air flow rate into and out of the Y-piece 20, denoted herein as V( ). The flowmeter 42 also directly or indirectly provides the net volume of air delivered to the patient, denoted herein as V( ), which may be directly measured or may be derived by integrating the flow rate V(t) over time. These measured values P_y(t), V(t), V(t), optionally along with other information such as the ventilator settings (e.g. Fi0₂, the pressure profile delivered by the PSV control, et cetera) may be variously used by a ventilator monitor 44 to efficacy of the mechanical ventilation, to detect any deterioration of the state of the patient 12, to detect any malfunction of the ventilator 10, or so forth. As with the ventilator controller 30, the ventilator monitor 44 is implemented as a microprocessor with ancillary electronics, and may be updateable by updating the software or firmware. In some embodiments, the ventilator controller 30 and the ventilator monitor 44 may be implemented by a common microprocessor, and the controller and monitor functions may be integrated at various levels - for example, it is contemplated to provide feedback-based ventilation control based on the measured values P_y(t), V( ), V(t) or parameters derived therefrom.

Of salient interest here is assessment of the Work of Breathing (WoB), or of its derivative, the respiratory muscle pressure P_mus(t)- In general, WoB can be computed by integrating P_mus(t) over the inhaled volume. In approaches disclosed herein, the assessment leverages the Equation of Motion of the Lungs given in Equation (1) herein, and hence the respiratory system's resistance R and compliance C are also salient parameters of interest. Equation (1) is evaluated with respect to a dataset of N data points measured over one or more breath cycles. Formally, the problem can be stated as follows:

Y = ΧΘ (2) where:

V = [ Py(X) P_y(2) ... P_y(N) ]^T Pressure at Y-piece at times 1, . . . ,N

V = [ V(l) V(2) ... V(N) ]^T Flow rate at times Ι, . , . ,Ν

V = [ V(l) V 2 ... V(N) ]^T Net air volume at times Ι , . , . ,Ν

Θ = [ R 1/C PmusiX) Pmus(2 - Pmus W ]^T Parameters to be determined and matrix X is an (N + 2) X N matrix given by X = [ V V /_w], where I_N is an N X N identity matrix. By solving the system of equations Y = ΧΘ for the parameter vector Θ, the resistance R, compliance C, and respiratory muscle pressure P_mus i) can be obtained. However, the system of equations represented by Equation (2) has more unknowns (N+2 unknowns) than equations (N equations), and hence is an underdetermined problem that cannot be solved because it has an infinite number of solutions, only one of which is the true "physical" solution.

In addition to being underdetermined, the inventors have found that the set of equations represented by matrix Equation (2) is very sensitive to measurement noise, unknown disturbances and unmodeled dynamics. Problematically, the noise is on the same time scale as the variations in the measured signals P_y(t), V t), V t) and in the estimated respiratory muscle pressure Pmus(t - Thus, even if the underdetermined nature of the simultaneous estimation problem is somehow overcome, the resulting parameter values tend to be noisy and hence of limited clinical value.

With continuing reference to FIGURE 1, it is disclosed herein to counteract the effect of noise by superposing a relatively high-frequency and small-amplitude pressure signal, denoted as AP(t) herein, generated by a signal generator 50, onto the normal pressure profile supplied by the ventilator 10. As illustrated in FIGURE 1, this can be done by adding a small- amplitude sinusoidal AP(t) to the controlled pressure signal output by the controller 30 using a signal combiner 52 prior to its input to the ventilator compressor 32. The amplitude of AP(t) is preferably chosen to be low enough to not appreciably impact the therapeutic value of the PSV signal output by the controller 30. That is, the superimposed cyclical pressure signal should be a non-therapeutic pressure signal that does not contribute to the ventilation therapy delivered to the ventilated patient 12, and also does not adversely affect the ventilation therapy delivered to the ventilated patient 12. The frequency of AP(t) is preferably high enough to be significantly higher than the breath frequency (e.g. typically a few breaths per minute corresponding to a frequency of, e.g., about 0.2 Hz for 5-sec breaths).

To illustrate, matrix Equation (2) is solved in an illustrate approach disclosed herein as follows. The simultaneous estimation of the R, C and P_mus(t) characterizing one breath (made of, without loss of generality, N recorded time samples) by Equation (2) is an underdetermined problem, since it requires the computation of N + 2 unknowns (N values for the N time samples of P_mus(t), plus an additional unknown for R, and an additional unknown for C) from N equations corresponding to the N time samples. However, it is recognized herein that the N equations are not independent. Rather, it can be expected that the value of P_mus (t) for neighboring samples should be continuous, because P_mus(t) should vary relatively slowly over time. As disclosed herein, P_mus(t) is approximated locally (that is, over a small number of samples s < N) by a n^th-order polynomial function suitably written as P_mus(t) = a₀ + a_xt +— V a_ntⁿ. This approximation is used to construct a least squares (LS) problem over a time window of s samples (where s < N) in which the unknowns are R, C, and a₀, a_n. By keeping n + 3 < s (and in some embodiments n « s), the underdeterminacy is overcome. The local approximation of P_mus(t) by a polynomial is supported by the physiological intuition that P_mus(t) is a smooth signal, with no abrupt discontinuities, and should vary relatively slowly over time.

The resulting LS problem to be solved is fully determined (and preferably overdetermined), but the inventors have found that it still tends to be sensitive to noise and/or disturbances in the measurements. The reason lies with the fact that the flow and volume signals (in particular the latter) can be approximated by a polynomial as well, being smooth functions of time. More generally, noise in the measured signals P_y(t), V(t), V(t) typically has comparable time-domain characteristics to the time-domain characteristics of P_y(t),

V( ), V(t), and P_mus(t), so that the fitted polynomial representation of P_mus(t) can erroneously fit to a noise component. The noise problem is fundamental to any analysis of the breathing cycle due to the many sources of noises such as air flow-induced vibration or other motion of the Y-piece 20, vibrations from the compressor 32, cycling (opening and closing or other actuation) of the oxygen regulator 34, and so forth. As disclosed herein, such noise is suitably addressed by superposition of a small- amplitude and relatively-high frequency pressure signal AP(t) onto the PSV or other ventilation pressure profile normally supplied by the ventilator 10. The small-amplitude, high frequency superimposed pressure variation AP(t) imparts a corresponding small-amplitude, high frequency variation in the pressure Py(t) at the Y-piece 20 and in the air flow V t (and, to a lesser extent, to the net volume V( ), although here the integration may partly remove the impact of AP(t)). This makes the measured signals of interest different in kind from the noise, leading to more robust extraction of the respiratory muscle pressure P_mus(t) from these signals. Thus, the LS solution (or other analysis of the breath cycle measurements) is made robust against noise and disturbances.

With reference now to FIGURE 2, the illustrative approach for overcoming the underdeterminancy of matrix Equation (2) is described in additional detail. The estimation of R, C, and P_mus (t) at each time 1, . . . ,N is obtained by solving a LS problem over a window of length s. For the usual case in which s < N, the window slides forward in time (that is, the window of width s is applied to successive increments of width s in the time series of samples 1,..,N). In real-time patient monitoring, this can be done as a sliding window - as each successive group of s samples are acquired, the fitting is performed so as to provide a real-time simultaneous estimate for R, C, P_mus(t)- The windows of width s can be non-overlapping; alternatively, it is contemplated for the neighboring windows of width s to overlap, which can provide a smoothing effect. FIGURE 2 illustrates the case in which the polynomial approximation of P_mus(t) over the time window of width s is of order n = 2, that is, a polynomial: P_mus(t) = a₀ + a_xt + a₂t². The matrix Equation (2) solved for the window of width s has the same form as Equation (2), but the parameters vector is different. To distinguish, the parameter vector is written as φ (rather than the parameter vector Θ of Equation (2)), the matrix X is replaced by a matrix χ, and the set of equations becomes: Υ = χφ (2a) where:

Y = [ Py(X P_y{2) ... P_y(s) ] For window of s samples

V = [ V{1) V(2 - V(s) ]^T For window of s samples

V = [ V{1) V(Z ... V(s) ]^T For window of s samples

φ = [ R 1/C a_{Q x} ... a_n ]^T Reduced to n + 3 unknowns and matrix χ is an s X (n + 3) matrix given

In the above notation, the first sample in the window of width s is designated without loss of generality as sample t=l, so that the last sample in the window is designated as sample t = s. Matrix Equation (2a) thus represents a set of s equations with n + 3 unknowns, and is overdetermined so long as s > (n + 3) . More typically, s » n. For example, in one illustrative example n = 2 (quadratic approximation for P_mus (t)), the sampling rate is 100 Hz, and the window is 0.6 sec long corresponding to s = 60.

Assuming an overdetermined set of equations, the matrix Equation (2a) can be solved in the least squares sense according to:

Φ = (.χ^ΓχΥ¹χ^ΓΥ (3)

Alternatively, an iterative least squares approximation approach such as gradient descent or Levenberg-Marquardt can be used to solve Equation (3) for the parameters φ.

The illustrative approach employs a polynomial approximation of order n of P_mus(t) over the time window of width s. The order n is chosen to be n > 2. Choosing a higher order provides the polynomial approximation with greater flexibility to represent changes in P_mus(t) over the time window of width s; however, it also adds additional parameters (the total number of parameters is n + 3) which makes the least squares fitting less robust. It is expected that n = 2, n = 3, or n = 4 will be sufficient in most instances, although n > 4 is also contemplated. Moreover, it will be appreciated that the approach can be generalized to approximating P_mus(t) over the time window of width s by any continuous function that is smooth over the window of width s (i.e. that is fully differentiable over the window of width s). Other contemplated continuous and smooth approximation functions include spline functions, e.g. cubic spline functions.

The sensitivity of the solution to noise and/or disturbances present in the measured data can be quantified by the condition number of the data matrix (X or χ, depending on whether Equation (2) or Equation (2a) is considered). The condition number of a matrix can be interpreted as a (worst-case) amplification gain of errors in the data matrix entries. Although replacing X by χ as just described reduces the number of unknowns in order to make the problem tractable, noise can still be an issue.

To illustrate, with reference to FIGURE 3 the normal interaction between the ventilator and a patient is emulated using a computer-simulated Lung Emulator. This normal interaction gives rise to waveforms of flow and volume, plotted in FIGURE 3, that make the data matrix ill-conditioned. Hence, the parameters estimated via Equation (3), are sensitive to noise or error in the measured data. In FIGURE 3, the sensitivity of the data matrix is plotted in the lowermost plot of FIGURE 3. Note in this plot that the sensitivity ordinate ranges [0 , 200,000] .

As previously described, this noise can be counteracted by superimposition of the low amplitude, high frequency component AP(t) .

To illustrate, FIGURE 4 shows how the superposition of a sinusoidal signal AP(t) of small- amplitude (1 cmH₂0 in this example) and relatively high-frequency (5 Hz in this example) significantly reduces the sensitivity (condition number) of the data matrix, thus improving the robustness against noise. Note that in FIGURE 4 the lowermost sensitivity plot has a sensitivity ordinate range of only [0 , 5000] .

Without being limited to any particular theory of operation, it is believed that the superposed signal AP(t) introduces variations in the flow and volume signals which cannot be described by the polynomial structure chosen for P_mus(t)- This enhances the algorithm robustness, making it less sensitive to measurement noise, unknown disturbances and unmodeled dynamics. More generally, the superposed signal AP(t) is believed to introduce variations in the measured signals, e.g. P_y(t), V(t), P_mus (t), which are different in kind from the variations due to typical sources of noise or disturbance, which enhances the robustness of any data analysis performed on these signals.

With reference to FIGURES 5 and 6, as further illustration the robustness of the estimates obtained from the data of FIGURE 3 with noise added is shown in FIGURE 5, while the robustness of the estimates obtained from the data of FIGURE 4 with noise added is shown in FIGURE 6. In other words, FIGURE 5 shows the estimation robustness when no small amplitude, high frequency pressure signal AP(t) is superimposed, while FIGURE 6 shows the estimation robustness when the pressure signal AP(t) is superimposed. The reduction in estimation error observed in FIGURE 6 compared with FIGURE 5 is readily apparent.

The superposition of the pressure signal AP(t) is designed to improve the estimation process while not interfering with the patient respiration. The magnitude of AP(t) (e.g. 1 cmH₂0 in illustrative FIGURES 4 and 6) is small enough that the ventilation therapy is not affected. Yet the superimposed pressure signal AP(t) provides a significant benefit in terms of noise robustness (or equivalently, a decrease in sensitivity) in the simultaneous assessment of R, C, and P_mus(t)-

In some embodiments the signal generator 50 and signal combiner 52 are implemented in software or firmware as part of the software or firmware of the ventilator controller 30. In such embodiments, the components 50, 52 can be implemented in the factory state, or as a retrofit to an existing ventilator by an appropriate control software or firmware update. Such software or firmware, either in the factory state or as a software or firmware update, is suitably provided in the form of a non-transitory storage medium storing instructions readable and executable by the microprocessor of the controller 30 to cause the ventilator 10 to perform ventilation of the ventilated patient 12 including the superimposing the cyclical signal AP(t) onto the positive pressure of the air flow delivered to the patient. The non-transitory storage medium may, for example, comprise a flash memory, optical disk, or other storage medium directly loaded into or physically connected with the ventilator 10, or may comprise a RAID or other storage medium accessed via a network server in which case the ventilator 10 connects with the network (e.g. the Internet or a hospital network) in order to download the software or firmware from the network server.

In other embodiments, it is contemplated for the signal generator 50 and signal combiner 52 to be components separate from the ventilator controller 30. For example, the signal generator 50 could be a voltage-controlled oscillator (VCO) or other oscillator circuit outputting the cyclical signal AP(t), and the signal combiner 52 could be an operational amplifier (op-amp)-based signal combiner circuit or other signal combiner circuit implemented in hardware.

The superimposed signal AP(t) should contain sufficiently high frequencies to make the data matrix X or χ well-conditioned. At the same time, such frequencies should be within the bandwidth of the ventilator hardware and the patient's respiratory system. While AP(t) has been described herein as sinusoidal in illustrative examples, more generally the superimposed signal AP(t) can have a shape other than sinusoidal (e.g. a square wave). In general the superimposed high frequency signal AP(t) has a frequency component, e.g. the frequency of a sinusoid in the case of a sinusoidal signal, or the inverse of the square wave period in the case of a square wave signal, or the inverse of the triangle wave period in the case of a saw-tooth or other triangular signal, or so forth. The cycle frequency (or another frequency component of the superimposed high frequency signal AP(t)) should be higher than the respiration rate (i.e. the frequency of the breathing cycle) so that AP(t) (and its impact on other signals such as flow rate V(t)) can be distinguished from respiration-related signal variations. Normal adult respiration rate is on the order of 12-20 breaths/minute (0.20-0.34 Hz); thus, in some embodiments the cycle frequency is at least 0.5 Hz, and more preferably at least 1 Hz. In the illustrative embodiment of FIGURES 4 and 6 the cycle frequency is 5 Hz. On the other hand, infants can have respiration rate of as high as 60 breaths/minute (1 Hz), so for an infant ventilator the superimposed signal AP(t) preferably has a cycle frequency of at least 3 Hz, and more preferably at least 5 Hz. Since ventilators are sometimes not infant- or adult- specific, in some embodiments a cycle frequency sufficient for infants is preferable. However, the superimposed signal AP(t) can also be a non-cyclical signal with significant high frequency components, such chirp signal or frequency sweep signal.

The amplitude of the superimposed signal AP(t) should be lower than the maximum positive pressure of the air flow delivered to the ventilated patient by the ventilator 10. More preferably, the superimposed signal AP(t) should be substantially lower than this maximum positive pressure, i.e. sufficiently lower that the superimposed signal AP(t) does not interfere with the therapeutic ventilation of the patient. The superimposed signal AP(t) is a non-therapeutic pressure signal that does not contribute, positively or negatively, to the ventilation therapy delivered to the patient. The invention has been described with reference to the preferred embodiments. Modifications and alterations may occur to others upon reading and understanding the preceding detailed description. It is intended that the invention be construed as including all such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.

Claims

CLAIMS:

1. A medical ventilator system comprising:

a ventilator (10) configured to deliver an air flow at positive pressure to a ventilated patient (12);

a signal generator (50) configured to generate a non-therapeutic pressure signal having amplitude less than the maximum positive pressure of the air flow delivered to the ventilated patient and having a frequency component of at least 0.5 Hz; and

a signal combiner (52) configured to superimpose the non-therapeutic pressure signal on pressure waveform delivered to the ventilated patient.

2. The medical ventilator system of claim 1 wherein the non-therapeutic pressure signal has a frequency component of at least 1 Hz.

3. The medical ventilator system of claim 1 wherein the non-therapeutic pressure signal is cyclical with a cycle frequency of at least 1 Hz.

4. The medical ventilator system of any one of claims 1-4 wherein the non-therapeutic pressure signal is a sinusoidal signal, a square wave signal, a triangle or sawtooth wave signal or a non-cyclical chirp or frequency sweeping signal.

5. The medical ventilator system of any one of claims 1-4 wherein the ventilator (10) includes:

a ventilator compressor (32) configured to deliver the air flow to the ventilated patient; and

a ventilator controller (30) comprising a microprocessor programmed to generate and send a control signal to the ventilator compressor in order to deliver pressure support ventilation (PSV) to the ventilated patient.

6. The medical ventilator system of claim 5 wherein the signal combiner (52) is configured to superimpose the non-therapeutic pressure signal on the PSV control signal that is sent to the ventilator compressor (32).

7. The medical ventilator system of claim 6 wherein the signal generator (50) and the signal combiner (52) comprise said ventilator controller (30) programmed to generate and superimpose the non-therapeutic pressure signal on the PSV control signal generated by the ventilator controller.

8. The medical ventilator system of claim 6 wherein:

the signal generator (50) comprises an oscillator circuit; and

the signal combiner (52) comprises a signal combiner circuit.

9. The medical ventilator system of any one of claims 1-8 further comprising: an inlet air hose (14) conveying the air flow from the ventilator (10) to the ventilated patient (12); and

a Y-piece or T-piece (20) connected with the discharge end of the inlet air hose to convey the air flow from the inlet air hose into the ventilated patient.

10. The medical ventilator system of any one of claims 1-9 further comprising:

a pressure sensor (40) configured to measure pressure of air inspired by or expired from ventilated patient (12);

a flowmeter (42) configured to measure air flow into or out of the ventilated patient; and

a ventilator monitor (44) comprising a microprocessor programmed to process information including at least the pressure measured by the pressure sensor and the air flow measured by the flowmeter to generate information about ventilation of the ventilated patient;

wherein the non-therapeutic pressure signal superimposed on the therapeutic respiratory pattern delivered to the ventilated patient is detectable in the pressure measured by the pressure sensor and is detectable in the air flow measured by the flowmeter.

11. The medical ventilator system of claim 10 wherein the ventilator monitor (44) is programmed to estimate respiratory system's resistance R and compliance C and respiratory muscle pressure generated by the ventilated patient based on at least the pressure measured by the pressure sensor (40) and the air flow measured by the flowmeter (42).

12. A non-transitory storage medium storing instructions readable and executable by one or more microprocessors of a medical ventilator (10) to cause the medical ventilator to perform a method comprising:

generating a non-therapeutic pressure signal; and

superimposing the non-therapeutic pressure signal on air flow delivered by the medical ventilator to a ventilated patient (12).

13. The non-transitory storage medium of claim 12 wherein the non-therapeutic pressure signal has a frequency component of at least 1 Hz.

14. The non-transitory storage medium of claim 12 wherein the non-therapeutic pressure signal is cyclical with has a cycle frequency of at least 5 Hz.

15. The non-transitory storage medium of any one of claims 12- 14 wherein method further comprises:

operating the medical ventilator (10) to deliver the air flow to the ventilated patient under pressure support ventilation (PSV).

16. The non-transitory storage medium of claim 15 wherein the superimposing comprises:

superimposing the non-therapeutic pressure signal on a PSV control signal that is sent to a ventilator compressor (32) of the medical ventilator (10).

17. The non-transitory storage medium of any one of claims 12- 16 wherein the superimposed non-therapeutic pressure signal is effective to impart a signal component to patient pressure and airflow measurements acquired by the medical ventilator (10), and the method further comprises: estimating respiratory system' s resistance R and compliance C and respiratory muscle pressure produced by the ventilated patient during respiration based on at least the patient pressure and airflow measurements acquired by the medical ventilator.

18. A medical ventilation method comprising:

delivering ventilation to a ventilated patient (12) using a medical ventilator

(10); and

superimposing a cyclical or non-cyclical non-therapeutic pressure signal with a frequency component higher than the patient respiration rate on the ventilation delivered to the ventilated patient.

19. The medical ventilation method of claim 18 wherein the delivering comprises delivering pressure support ventilation (PSV) to the ventilated patient (12), and the method further comprises:

measuring pressure at a Y-piece or T-piece (20) coupled to the ventilated patient (12) using a pressure sensor (40);

measuring air flow at the Y-piece or T-piece (20) coupled to the ventilated patient (12) using a flowmeter (42); and

computing respiratory muscle pressure produced by the ventilated patient during respiration based at least on the measured pressure and air flow.

20. The medical ventilation method of claim 19 wherein:

the computing comprises solving a system of equations formed by applying an equation of motion of the lungs to the pressure and air flow measured at a plurality of successive times, and

the superimposing comprises superimposing said cyclical or non-cyclical non-therapeutic pressure signal having a frequency component that is effective to make a data matrix of the system of equations well-conditioned.