Mechanical ventilation method of respiratory system based on double-exponential model and system thereof

Abstract

A mechanical ventilation method of a respiratory system based on a double-exponential model and a system thereof are provided. The method includes: S101 acquiring collected data of a respiratory circu

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of mechanical ventilation, in particular to a mechanical ventilation method of a respiratory system based on a double-exponential model and a system thereof. BACKGROUND

[0002] In the study of mechanical ventilation of a respiratory system, lung tissues have a certain step response characteristics, and especially the step characteristics are very obvious in volume-controlled ventilation mode. It is difficult for the classical lumped parameter model based on differential equations to describe and reflect the characteristics well. Therefore, it is urgent to provide a new parameter model to deeply understand the characteristics of the respiratory system and reduce the ventilation injury resulted from a ventilator. SUMMARY

[0003] The purpose of the present disclosure is to provide a mechanical ventilation method of a respiratory system based on a double-exponential model and a system thereof, which can improve the accuracy and adaptability of mechanical ventilation control of the respiratory system.

[0004] In order to achieve the above objectives, the present disclosure provides the following scheme.

[0005] A mechanical ventilation method of a respiratory system based on a double-exponential model is provided, including:

[0006] acquiring collected data of a respiratory circuit in any ventilation mode, wherein the collected data includes a pressure value, a gas flow rate and a respiratory volume;

[0007] obtaining the double-exponential model with the collected data as input and control parameters of the respiratory system as output according to the collected data, wherein the control parameters include respiratory system elasticity, airway resistance and positive end expiratory pressure;

[0008] acquiring real-time collected data, and determining the control parameters by using the double-exponential model; and

[0009] carrying out mechanical ventilation control of the respiratory system according to the control parameters.

[0010] In some embodiments, the double-exponential model is: P(t)=Ey(t)+RV(t)+0^'^ •

[0011] where P(t) is the pressure at time t; V'(t) is a first derivative of the respiratory volume V(t) at time t, which represents the gas flow rate; E is the respiratory system elasticity, R is the airway resistance, Po is the positive end expiratory pressure; {xi, X2, ..., xm} are exponential parameters, and {ai, a2, ..., am} are constant coefficients corresponding to double-exponential terms.

[0012] In some embodiments, the obtaining the double-exponential model with the collected data as input and control parameters of the respiratory system as output according to the collected data includes:

[0013] using a particle swarm optimization to perform machine learning on the doubleexponential model by using the collected data.

[0014] In some embodiments, subsequent to obtaining the double-exponential model with the collected data as input and control parameters of the respiratory system as output according to the collected data, the method further includes:

[0015] evaluating the double-exponential model by using a sum of squared residuals and a root mean square error.

[0016] A mechanical ventilation system of a respiratory system based on a double-exponential model is provided, including:

[0017] a collected data acquiring module, configured to acquire collected data of a respiratory circuit in any ventilation mode, wherein the collected data includes a pressure value, a gas flow rate and a respiratory volume;

[0018] a double-exponential model determining module, configured to obtain the doubleexponential model with the collected data as input and control parameters of the respiratory system as output according to the collected data, wherein the control parameters include respiratory system elasticity, airway resistance and positive end expiratory pressure;

[0019] a control parameter determining module, configured to acquire real-time collected data, and determine the control parameters by using the double-exponential model; and

[0020] a mechanical ventilation control module, configured to carry out mechanical ventilation control of the respiratory system according to the control parameters.

[0021] According to the specific embodiment provided by the present disclosure, the present disclosure provides the following technical effects.

[0022] The present disclosure provides a mechanical ventilation method of a respiratory system based on a double-exponential model and a system thereof. The method includes obtaining a double-exponential model with the collected data as input and control parameters of the respiratory system as output according to the collected data, and then determining the control parameters by using the double-exponential model; reflecting the step response characteristics of the respiratory system by the double-exponential model, which is easier to obtain a better fitting degree than the classical respiratory mechanics (single-compartment) model, so as to improve the accuracy and adaptability of mechanical ventilation control of the respiratory system. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to explain the embodiments of the present disclosure or the technical schemes in the prior art more clearly, the drawings that need to be used in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present disclosure. For those skilled in the art, other drawings can be obtained according to these drawings without creative labor.

[0024] FIG. 1 is a schematic flow diagram of a mechanical ventilation method of a respiratory system based on a double-exponential model according to the present disclosure.

[0025] FIG. 2 is a schematic diagram showing a comparison among the measured data and the simulated data of a single-compartment model and a double-exponential model. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] The technical schemes in the embodiments of the present disclosure will be clearly and completely described with reference to the drawings in the embodiments of the present disclosure hereinafter. Obviously, the described embodiments are only some embodiments of the present disclosure, rather than all of the embodiments. Based on the embodiment of the present disclosure, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present disclosure.

[0027] The purpose of the present disclosure is to provide a mechanical ventilation method of a respiratory system based on a double-exponential model and a system thereof, which can improve the accuracy and adaptability of mechanical ventilation control of the respiratory system.

[0028] In order to make the above objects, features and advantages of the present disclosure more obvious and understandable, the present disclosure will be explained in further detail with reference to the drawings and detailed description hereinafter.

[0029] As shown in FIG. 1, a mechanical ventilation method of a respiratory system based on a double-exponential model according to the present disclosure includes steps S101 -S104

[0030] In step S101, collected data of a respiratory circuit in any ventilation mode is acquired, where the collected data includes a pressure value, a gas flow rate and a respiratory volume.

[0031] In step SI02, the double-exponential model with the collected data as input and control parameters of the respiratory system as output is obtained according to the collected data, where the control parameters include respiratory system elasticity, airway resistance and positive end expiratory pressure.

[0032] The double-exponential model is: P(t) = EV(t) + RV'(t) +          +                        + Po

[0033] where P(t) is the pressure at time t; V(t) is a first derivative of the respiratory volume V(t) at time t, which represents the gas flow rate; E is the respiratory system elasticity, R is the airway resistance, Po is the positive end expiratory pressure; {xi, X2, ..., xm} are exponential parameters, and {ai, a2, ..., am] are constant coefficients corresponding to double-exponential terms.

[0034] In practical application, the double-exponential model can be simplified to reflect the step response characteristics of the respiratory system using a few double-exponential terms. Generally, there are several double-exponential terms corresponding to several step jumps in the respiratory volume, such as: P(t) = EV(t) + RV'(t) + ae^E^t) + PQ Or P(t) = EV(t) + RV'(0 + + a2eex^V(t) + Po.

[0035] For the model with m double-exponential terms (plus 3 non-double-exponential terms, amounting to m+3 terms in total), the pressure value, the flow rate value and the accumulated gas volume value (the respiratory volume) of the respiratory circuit are collected in real time. The values of the double-exponential terms are obtained according to the number m of doubleexponential terms of the predefined model and the corresponding form, and a set of specified exponential parameters{ai, a2, ..., am] and {xi, X2, ..., xm}, which are sorted into data in the corresponding matrix form as the input of the algorithm. That is, when xm is specified, the value of the double-exponential term ( / ) can be calculated correspondingly. If the values of the respiratory circuit are collected at n time points, the input data includes n-dimensional air pressure vector P and data matrix X with n rows and m+3 columns: k v; vi ... c i E, F2 ••• V? 1 V v' F’ Vm 1 V N rN yN AJ

[0036] Column Vm in the above formula (the second column from the right) is the value except the constant coefficient obtained by combining the double-exponential term, and the same is true for other corresponding double-exponential terms.

[0037] For the model with m double-index terms, after a set of initial index parameters {ai, a2, ..., am} and {xi, X2, ..., xm} are specified, the fitting results of various coefficients can be obtained by matrix calculation according to the following expression. A = [XtXY'XtP

[0038] The coefficient vectors are as follows:

[0039] The calculation method proposed in the above formula is derived from the extreme point where the partial derivative for each coefficient is 0 when the sum of squared residuals (SSR) is taken as the minimum value. By this step, the parameters of the specified double-exponential model have been obtained, and the exponential parameters can be further optimized.

[0040] After the coefficients of the specified double-exponential model are obtained, combined with the measured data, the SSR and the root mean square error (RMSE) of the corresponding model can be calculated by model simulation. In order to optimize the most suitable series of exponential parameters {xi, X2, ..., xm} based on data, the particle swarm optimization is selected among various machine learning algorithms in consideration of computational efficiency. In genetic algorithms, simulated annealing and other similar algorithms, although the particle swarm optimization may fall into the local optimal solution, the particle swarm optimization converges faster and pays more attention to timeliness in practical application.

[0041] In the iterative optimization process, the parameters to be optimized are a series of exponential parameters {xi, X2, ..., xm}. Each time a set of index parameters is specified, all the coefficient parameters of the double-exponential model can be obtained based on the set of exponential parameters, and in turn the SSR / RMSE can be obtained. The parameters can be 5 continuously updated following the change trend of the SSR or the RMSE. The optimization goal of the double-exponential model is to minimize the SSR or the RMSE. The double-exponential model with a better state than the initial state can be obtained after iterative calculation.

[0042] In step S103, real-time collected data is acquired, and the control parameters are determined by using the double-exponential model.

[0043] In step SI04, mechanical ventilation control of the respiratory system is carried out according to the control parameters.

[0044] Description will be carried out by the following specific embodiments.

[0045] In respiratory mechanics modeling, the classical single-compartment model is: P = EV + RV' + P0 (1)

[0046] Based on such classical model, the example of a single double-exponential model is supplemented for showing the calculation results, and the parameters of the following model are estimated according to the present disclosure: P = EV + RV+aee V(t) + Pf} (2)

[0047] The parameter estimation is based on the measured data of animal experiments (measured by Drager Savina 300 ventilator). The data of the pressure p (in unit of mbar), the flow rate v' (in unit of L / s) and the respiratory volume v (in unit of L) every 10ms are recorded as one row in chronological order (for the data, see caselVCACl.txt, and the data within 150s is intercepted).

[0048] Based on the example python+matlab code main.m (*.py and *.m are python and matlab subfunctions called by the main code, and *.slx is the model file), the calculation results of three model parameters after reading the measured data are shown in Table 1. Table 1 Type of models Estimated parameters and index results Single-compartment model E=27.4861, R=6.1912, Po=3.4O41, RMSE=7.7975, run_time=0.3069(s) Double-exponential model E=25.7535, R=6.2159, a=9.8121E-8, x=9.5379, Po=3.45O4, RMSE=6.3248, run_time=32.9677(s)

[0049] The RMSE in Table 1 is converted into the unit of milliliter, and run time stands for running time. As can be seen from Table 1, the method proposed in the present disclosure is practical and effective, and the double-exponential model has a better fitting effect and a lower RMSE (root mean square error) value. Further, the measured data and the flow V fitting of Formula (1) and Formula (2) are plotted, as shown in FIG. 2 (the result obtained with 10 particles over 10 iterations).

[0050] Corresponding to the method provided in the above embodiments, the present disclosure further provides a mechanical ventilation system of a respiratory system based on a doubleexponential model, including a collected data acquiring module, a double-exponential model determining module, a control parameter determining module, and a mechanical ventilation control module.

[0051] The collected data acquiring module is configured to acquire collected data of a respiratory circuit in any ventilation mode, where the collected data includes a pressure value, a gas flow rate and a respiratory volume.

[0052] The double-exponential model determining module is configured to obtain a doubleexponential model with the collected data as input and control parameters of the respiratory system as output according to the collected data, where the control parameters include respiratory system elasticity, airway resistance and positive end expiratory pressure.

[0053] The control parameter determining module is configured to acquire real-time collected data, and determine the control parameters by using the double-exponential model.

[0054] The mechanical ventilation control module is configured to carry out mechanical ventilation control of the respiratory system according to the control parameters.

[0055] In this specification, various embodiments are described in a progressive way. The differences between each embodiment and other embodiments are highlighted, and the same and similar parts of various embodiments can be referred to each other. Since the system provided in the embodiment corresponds to the method provided in the embodiment, the system is described simply. Refer to the description of the method for the relevant points.

[0056] In the present disclosure, specific examples are applied to illustrate the principle and implementation of the present disclosure, and the explanations of the above embodiments are only used to help understand the method and core ideas of the present disclosure. At the same time, according to the idea of the present disclosure, there will be some changes in the specific implementation and application scope for those skilled in the art. To sum up, the contents of the specification should not be construed as limiting the present disclosure.

Claims

1. A mechanical ventilation system of a respiratory system based on a double-exponential model, comprising:a collected data acquiring module, configured to acquire collected data of a respiratory circuit in any ventilation mode, wherein the collected data comprises a pressure value, a gas flow rate and a respiratory volume;a double-exponential model determining module, configured to obtain the doubleexponential model with the collected data as input and control parameters of the respiratory system as output according to the collected data, wherein the control parameters comprise respiratory system elasticity, airway resistance and positive end expiratory pressure;a control parameter determining module, configured to acquire real-time collected data, and determine the control parameters by using the double-exponential model; anda mechanical ventilation control module, configured to carry out mechanical ventilation control of the respiratory system according to the control parameters,wherein the double-exponential model is:P(t) = EV^ + RV\t) +           +            ■ ■amea-f',nV(t) + Pa;wherein P(t) is pressure at time t; V(t) is a first derivative of the respiratory volume V(t) at the time t, which represents the gas flow rate; E is the respiratory system elasticity, R is the airway resistance, Po is the positive end expiratory pressure; {xi, X2, ..., xm} are exponential parameters, and {ai, a2, ..., am} are constant coefficients corresponding to double-exponential terms.

2. The mechanical ventilation system according to claim 1, wherein the double-exponential model determining module comprises:using a particle swarm optimization to perform machine learning on the double-exponential model by using the collected data.

3. The mechanical ventilation system according to claim 1, wherein further comprises:evaluating the double-exponential model by using a sum of squared residuals and a root mean square error.

Images