CN111973188A - Method for estimating respiratory mechanics parameter based on neural network - Google Patents
Method for estimating respiratory mechanics parameter based on neural network Download PDFInfo
- Publication number
- CN111973188A CN111973188A CN202010868056.4A CN202010868056A CN111973188A CN 111973188 A CN111973188 A CN 111973188A CN 202010868056 A CN202010868056 A CN 202010868056A CN 111973188 A CN111973188 A CN 111973188A
- Authority
- CN
- China
- Prior art keywords
- respiratory
- flow
- neural network
- pressure signals
- clinical
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Pathology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Physiology (AREA)
- Veterinary Medicine (AREA)
- General Engineering & Computer Science (AREA)
- Heart & Thoracic Surgery (AREA)
- Computing Systems (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Physics & Mathematics (AREA)
- Pulmonology (AREA)
- Databases & Information Systems (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Fuzzy Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention relates to a method for estimating respiratory system mechanical parameters (airway resistance coefficient and compliance) by utilizing a neural network based on gas flow and pressure signals in an artificial airway. The method comprises the following steps: 1. acquiring airway flow and pressure signals, and measuring airway resistance coefficient and compliance by a blocking method to obtain a clinical respiratory signal sample set; 2. generating a simulated respiratory signal sample set by using a respiratory mechanics model; 3. processing the clinical and simulated respiratory signal sample sets to form a training set; 4. and carrying out neural network training to obtain the estimation of the respiratory mechanics parameters. The method can estimate the resistance coefficient and the compliance of the airway according to the flow pressure signal of the airway, effectively solves the problem that the respiratory mechanics parameter is difficult to estimate under the condition of spontaneous respiration, and provides reference and help for the adjustment of a clinical treatment scheme. The method uses simulation data to assist neural network model training, greatly reduces the requirement of clinical data, and improves the feasibility of the method.
Description
Technical Field
The invention belongs to the field of physiological signal processing and characteristic parameter estimation methods, and particularly relates to a method for extracting respiratory mechanics parameters from an artificial airway pressure flow signal based on a neural network method, which is used for dynamic monitoring of the respiratory mechanics parameters of a mechanical ventilation patient and provides an important reference basis for monitoring and diagnosing and treating a disease process.
Background
Many respiratory diseases manifest as changes in respiratory mechanics parameters, such as significant expiratory flow limitation, i.e., an increase in airway resistance coefficient, in patients with chronic obstructive pulmonary disease, which is often accompanied by a decrease in pulmonary compliance. The respiratory mechanics parameter represented by airway resistance coefficient and compliance is an index capable of directly reflecting the performance of the respiratory system of a patient, and the respiratory mechanics parameter of a mechanically ventilated patient is also one of important references for setting the mechanical ventilation parameter by a doctor, so that the effective monitoring of the respiratory mechanics parameter has important significance for monitoring the lung disease condition. At present, the method for detecting the respiratory mechanics parameter mainly uses a breathing machine to perform a blocking method to measure the respiratory mechanics parameter, or performs the estimation of the respiratory mechanics parameter based on a respiratory mechanics model. The blocking method must block ventilation and cannot realize real-time monitoring, but the method for parameter estimation based on the breathing mechanics model is suitable for patients without spontaneous breathing, once the patients have spontaneous breathing and need invasive measurement of esophageal pressure instead of pleural pressure to complete estimation, the method is invasive and cannot realize real-time monitoring. The invention provides a method for estimating respiratory mechanics parameters based on a neural network, which is used for collecting airway flow pressure signals on the premise of not influencing normal ventilation, and performing breath-by-breath parameter joint estimation by means of the trained neural network, and has important significance for monitoring pulmonary disease conditions of mechanically ventilated patients.
Disclosure of Invention
The invention aims to provide a dynamic estimation method of respiratory mechanics parameters, and particularly relates to a method for acquiring airway flow and pressure signals in an artificial airway, forming a fixed-length single-cycle combined clinical sample through preprocessing and signal combination, simultaneously simulating the respiratory flow and the pressure signals by using a respiratory mechanics classical model, mainly realizing the supplement of clinical cases, preprocessing the simulated flow and the pressure signals, combining the preprocessed flow and the pressure signals to form a fixed-length single-cycle combined simulated sample, and training a neural network by using the clinical sample and the simulated sample together, so that the neural network can realize the estimation of the respiratory mechanics parameters (airway resistance coefficient and compliance). The obtained neural network can realize the estimation of respiratory mechanics parameters based on flow pressure signals acquired in real time in the artificial airway of a clinical mechanical ventilation patient, and provides meaningful reference for a clinician to monitor the state of a respiratory system disease.
The technical scheme of the invention is as follows; a method for estimating respiratory mechanics parameters based on a neural network comprises the steps of processing a clinical sample and a simulation sample to respectively obtain the clinical sample and the simulation sample with fixed-length single-cycle combination characteristics, and utilizing the processed clinical sample and the simulation sample with the fixed-length single-cycle combination to jointly realize training of an artificial neural network, so that the artificial neural network obtained after training can realize estimation of the respiratory mechanics parameters based on a single-cycle flow pressure signal.
Further, according to the method for estimating respiratory mechanics parameters based on the neural network, the clinical sample is obtained by collecting flow and pressure signals of a patient ventilated by a respirator on an artificial airway, preprocessing the signals, injecting muscle relaxation later, setting the respirator for volume control in a mechanical ventilation state of the patient without spontaneous respiration, measuring the respiratory mechanics parameters by a respirator blocking method, and marking the sample to obtain a clinical respiratory flow and pressure signal sample set.
Further, the method for estimating respiratory mechanics parameters based on neural network as described above, wherein the simulated sample is a respiratory mechanics model parameter, namely airway resistance coefficient (R) and compliance (C), randomly set in the range of extreme pathology and clinical sample shortage through the respiratory mechanics model, the ventilator action mode is selected, the simulated respiratory flow and pressure signal are generated, random noise is added to the generated flow and pressure signal, and a simulated respiratory signal sample set is obtained to supplement the problem of non-uniformity of the clinical sample, wherein the set parameter is a sample mark.
Further, the method for estimating respiratory mechanics parameters based on neural network as described above, wherein the neural network training is implemented by using a training set composed of limited clinical data and a large amount of simulation data, and the respiratory flow and the pressure signal are subjected to period identification, down-sampling, truncation, and zero padding, and the corresponding flow and pressure signals are combined to form a single period combined signal sample set which can be used for network training. For clinical or simulation signals, intercepting the flow and pressure of a respiratory cycle, performing down-sampling, wherein the final sampling rate is 15Hz, only 75 sampling points are respectively selected for the flow and pressure signals to ensure that the length of each sample is the same, if the flow or pressure signal of one respiratory cycle is less than 75 points, the flow and pressure signals are respectively complemented to 75 points, if the flow or pressure signal of one respiratory cycle exceeds 75 points, the flow and pressure signals are respectively taken from the first 75 points, and the flow and pressure signals are spliced in the sequence of the flow and pressure signals in the front and the pressure signals in the back to form a single-cycle combined signal sample. A large number of monocycle combined signal samples constitute the training set.
Further, the method for estimating respiratory mechanics parameters based on the neural network as described above, wherein the neural network is a three-layer neural network, the input layer has 150 nodes, the output layer has 2 nodes, the hidden layer has 5 nodes, 2 hidden layer nodes are used for estimating airway resistance coefficients, 3 hidden layer nodes are used for estimating compliance, the neural network is trained through a supervised training set formed by a large number of single-period combined signal samples until a stopping criterion is met, each level of nodes of the neural network is determined, and the neural network which can be used for estimating respiratory mechanics parameters is obtained after the accuracy and generalization capability of a model are verified by using clinical test samples.
The invention has the following beneficial effects: the respiratory mechanics parameter estimation method provided by the invention fully utilizes the characteristics of self-learning, better robustness and the like of the neural network, avoids the difficulty that the respiratory mechanics parameters cannot be estimated when spontaneous respiration exists, and simultaneously carries out sample supplement through simulation data, thereby greatly reducing the requirement of clinical sample data, realizing network training on the basis of a small amount of clinical experiment samples, and estimating the respiratory mechanics parameters of patients with spontaneous respiration in real time.
The neural network adopted in the invention has a three-layer structure, 150 nodes of an input layer, 5 nodes of a hidden layer and 2 nodes of an output layer, wherein 2 nodes of the hidden layer are used for estimating an airway resistance coefficient, and 3 nodes of the hidden layer are used for estimating compliance. The structure of the related neural network is obtained through experimental optimization, so that the current airway resistance coefficient and the compliance of a patient can be estimated as accurately as possible under the condition of as few nodes as possible, and an important reference is provided for the diagnosis and treatment of the respiratory system disease process.
Drawings
FIG. 1 is a flow chart of a method for estimating respiratory mechanics parameters based on a neural network;
FIG. 2 is a respiratory signal processing flow;
fig. 3 is a diagram of a neural network architecture for estimating respiratory mechanics parameters.
Detailed Description
The invention is described in further detail below with reference to specific embodiments and with reference to the following drawings.
The method for estimating respiratory mechanics parameters (airway resistance coefficient and compliance) based on the neural network is shown in fig. 1, a clinical case for ventilation by using an invasive ventilator is selected, gas flow and pressure signals (01) in an artificial airway are collected to serve as clinical respiratory signal samples, and then the current respiratory mechanics parameters (02), namely the airway resistance coefficient and the compliance, of the patient are measured by using a blocking method. The measured respiratory mechanics parameters (02) are used as sample marks, and form a clinical respiratory signal sample set (03) together with corresponding flow and pressure signals (01). In view of the problems of limited clinical cases and uneven distribution of respiratory mechanics parameters, the invention proposes a method for supplementing a clinical respiratory signal sample set (03) with a simulated respiratory signal sample set (07). The construction of the simulated respiratory signal sample set (07) is based on a respiratory mechanics model (05), a breathing mechanics parameter (04) which is randomly set in the range of extreme pathology and clinical sample shortage, namely an airway resistance coefficient and compliance, a breathing machine action mode is selected, simulated respiratory flow and pressure signals (06) are generated, random noise is added to the generated flow and pressure signals, and the simulated respiratory signal sample set (07) is obtained. The clinical respiration signal sample set (03) and the simulation respiration signal sample set (07) are combined and processed by the respiration signal to form a monocycle combined signal sample set (08) which can be used for neural network training. The monocycle combined signal sample set (08) is used for training a three-layer neural network (09), and parameters of all nodes of the three-layer fully-connected network (09) are determined, so that the neural network can be used for estimating respiratory mechanics parameters (airway resistance coefficient and compliance).
The single-period combined signal sample set (08) is obtained by combining a clinical respiration signal sample set (03) and a simulated respiration signal sample set (07) after signal processing, the specific processing process is as shown in fig. 2, the respiratory flow and pressure signals are respectively filtered, high-frequency noise is removed, period identification is carried out subsequently, the length of one respiration period is intercepted, down-sampling is carried out on the signal of each respiration period, the final sampling rate is 15Hz, and the sampling frequency ensures that the sample length is reduced while the characteristics of the respiration signals are kept as much as possible. The length of each respiratory cycle is different, in order to ensure that the length of each sample is the same, only 75 sampling points are respectively selected for flow and pressure signals, if the flow or pressure signal of one respiratory cycle is less than 75 points, the flow and pressure signals are respectively complemented to 75 points, if the flow or pressure signal of one respiratory cycle exceeds 75 points, the flow and pressure signals are respectively taken from the first 75 points, and the flow and pressure signals are spliced in the sequence of the flow signal at the front and the pressure signal at the back to form a single-cycle combined signal sample. Each sample in the single-period combined signal sample set (08) has the same length and can reflect flow and pressure signals of one respiratory period, and the corresponding respiratory mechanical parameters are marked at the same time, so that the single-period combined signal sample set can be used for training a neural network.
As shown in fig. 3, a neural network selected by the method for estimating respiratory mechanics parameters based on a neural network is a three-layer neural network (09) which is divided into an input layer, a hidden layer and an output layer, wherein the input layer has 150 nodes, the hidden layer has 5 nodes, and the output layer has 2 nodes, wherein 2 nodes of the hidden layer are used for estimating airway resistance coefficients (R), and 3 nodes of the hidden layer are used for estimating compliance (C). And taking the single-period combined signal sample set (08) as a neural network training set, carrying out neural network training until a stopping criterion is met, determining each level of nodes of the neural network, and verifying the accuracy and generalization capability of the model by using a clinical test sample to obtain the neural network for estimating respiratory mechanics parameters.
In summary, the present invention provides a method for estimating respiratory mechanics parameters based on a neural network, which realizes the estimation of respiratory mechanics parameters through a three-layer neural network with a fixed structure. The method fully considers the difficulty of estimating physiological parameters by the neural network, provides a simulation signal obtained based on classical model simulation for solving the problems of insufficient clinical signals and uneven distribution of clinical samples, and provides a sample set for neural network training by merging respiratory flow and pressure signals after a series of processing, so that the estimation of airway resistance coefficient and compliance based on the neural network becomes possible.
The method provided by the invention not only can be suitable for estimating the respiratory mechanics parameter of a patient without spontaneous respiration, but also can be suitable for estimating the airway resistance coefficient and the compliance parameter of a patient with spontaneous respiration. In the early research, the breathing flow and pressure signals of a patient with spontaneous respiration are used for network training, and the method is not influenced by the spontaneous respiration, can estimate the resistance coefficient and the compliance of an air outlet channel under the condition of not detecting the intra-pleural pressure or the esophageal pressure, is not influenced by a ventilation mode, and has important reference significance for monitoring and diagnosing the disease process of the patient with the respirator.
The above example is only one embodiment of the present invention. It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is intended to include such modifications and variations.
Claims (5)
1. A method for estimating respiratory mechanics parameters, namely an airway resistance coefficient and compliance, based on a neural network is characterized by comprising the following steps:
collecting flow and pressure signals in an artificial airway, preprocessing the signals, and measuring airway resistance coefficient and compliance by a blocking method to obtain clinical respiratory flow and pressure signal samples with parameter marks;
secondly, setting airway resistance coefficients and compliance parameters in the model by using a respiratory mechanics model, generating a simulated respiratory flow pressure signal, adding noise, and obtaining a simulated respiratory flow and pressure signal sample, wherein the set airway resistance coefficients and compliance are sample marks;
thirdly, according to the clinical and simulated flow and pressure signal samples obtained in the first step and the second step, carrying out period identification, down-sampling and truncation/zero filling, and combining the corresponding flow and pressure signals to form a single period combined signal sample;
and fourthly, training the neural network by using a large number of single-period combined signal samples obtained in the third step until a stopping criterion is met, determining parameters of each level of nodes of the neural network, and verifying the accuracy and the generalization capability of the model by using clinical test samples, wherein the network can be used for estimating respiratory mechanics parameters, namely airway resistance and compliance.
2. The method for estimating respiratory mechanics parameters based on neural networks according to claim 1, wherein the first step specifically comprises: collecting flow and pressure signals of a mechanical ventilation patient on an artificial airway, preprocessing the signals, injecting muscle relaxation later to enable the patient to enter a mechanical ventilation state without spontaneous respiration, setting a respirator for volume control, measuring respiratory mechanical parameters by a respirator blocking method, and carrying out sample marking to obtain clinical respiratory flow and pressure signal samples with marks.
3. The method for estimating respiratory mechanics parameters based on neural networks according to claim 1, wherein the second step specifically comprises: the method comprises the steps of selecting a breathing machine action mode by using a breathing mechanics model, randomly setting breathing mechanics model parameters, namely an airway resistance coefficient (R) and a compliance (C), in a parameter range of clinical sample insufficiency, generating simulated respiratory flow and pressure signals, and adding random noise to the generated flow and pressure signals to obtain simulated respiratory signal samples so as to supplement the problem of non-uniformity of the clinical samples, wherein the parameters set in the model simulation process are sample marks.
4. The method for estimating respiratory mechanics parameters based on neural networks according to claim 1, wherein the third step specifically comprises: the signal samples obtained according to the claim 2 and the claim 3 are down-sampled, the final sampling rate is 15Hz, the flow and pressure signals of one respiratory cycle are intercepted, if the flow or pressure signals of one respiratory cycle are less than 75 points, the flow and pressure signals are respectively complemented to 75 points, if the flow or pressure signals of one respiratory cycle exceed 75 points, the flow and pressure signals are respectively taken from the first 75 points, and the flow and pressure signals are spliced and combined in the sequence of the flow and pressure signals at the front to form the single-cycle combined signal sample.
5. The method for estimating respiratory mechanics parameters based on neural networks according to claim 1, wherein the fourth step specifically comprises: establishing a three-layer neural network, inputting 150 nodes of a layer, outputting 2 nodes of the layer and 5 nodes of a hidden layer, wherein 2 nodes of the hidden layer are used for estimating an airway resistance coefficient, 3 nodes of the hidden layer are used for estimating compliance, taking a large number of combined signal samples obtained in the claim 4 as a neural network training set, carrying out neural network training until a stopping criterion is met, determining parameters of each level of nodes of the neural network, and verifying the accuracy and generalization capability of a model by using clinical test samples to obtain the neural network capable of being used for estimating respiratory mechanics parameters.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010868056.4A CN111973188A (en) | 2020-08-26 | 2020-08-26 | Method for estimating respiratory mechanics parameter based on neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010868056.4A CN111973188A (en) | 2020-08-26 | 2020-08-26 | Method for estimating respiratory mechanics parameter based on neural network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111973188A true CN111973188A (en) | 2020-11-24 |
Family
ID=73443447
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010868056.4A Pending CN111973188A (en) | 2020-08-26 | 2020-08-26 | Method for estimating respiratory mechanics parameter based on neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111973188A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114185372A (en) * | 2021-11-08 | 2022-03-15 | 北京谊安医疗系统股份有限公司 | Ventilation pressure lifting rate control system and control method for breathing machine |
CN115804585A (en) * | 2023-02-08 | 2023-03-17 | 浙江大学 | Method and system for detecting high resistance of air passage based on mechanical ventilation waveform |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1421902A1 (en) * | 2002-11-20 | 2004-05-26 | Maquet Critical Care AB | Method for assessing pulmonary stress and a breathing apparatus |
CN102770070A (en) * | 2009-12-28 | 2012-11-07 | 佛罗里达大学研究基金会有限公司 | System and method for assessing real time pulmonary mechanics |
CN104135925A (en) * | 2012-02-20 | 2014-11-05 | 佛罗里达大学研究基金会有限公司 | Method and apparatus for predicting work of breathing |
CN109350063A (en) * | 2018-12-03 | 2019-02-19 | 北京航空航天大学 | A kind of breathing mechanics parameters detection device and method suitable for chronic obstructive pulmonary disease monitoring |
CN110755077A (en) * | 2019-11-18 | 2020-02-07 | 北华大学 | Lung elasticity coefficient measuring method and system based on breathing device |
-
2020
- 2020-08-26 CN CN202010868056.4A patent/CN111973188A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1421902A1 (en) * | 2002-11-20 | 2004-05-26 | Maquet Critical Care AB | Method for assessing pulmonary stress and a breathing apparatus |
CN102770070A (en) * | 2009-12-28 | 2012-11-07 | 佛罗里达大学研究基金会有限公司 | System and method for assessing real time pulmonary mechanics |
CN104135925A (en) * | 2012-02-20 | 2014-11-05 | 佛罗里达大学研究基金会有限公司 | Method and apparatus for predicting work of breathing |
CN109350063A (en) * | 2018-12-03 | 2019-02-19 | 北京航空航天大学 | A kind of breathing mechanics parameters detection device and method suitable for chronic obstructive pulmonary disease monitoring |
CN110755077A (en) * | 2019-11-18 | 2020-02-07 | 北华大学 | Lung elasticity coefficient measuring method and system based on breathing device |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114185372A (en) * | 2021-11-08 | 2022-03-15 | 北京谊安医疗系统股份有限公司 | Ventilation pressure lifting rate control system and control method for breathing machine |
CN114185372B (en) * | 2021-11-08 | 2023-09-19 | 北京谊安医疗系统股份有限公司 | Ventilation pressure lifting rate control system and control method for breathing machine |
CN115804585A (en) * | 2023-02-08 | 2023-03-17 | 浙江大学 | Method and system for detecting high resistance of air passage based on mechanical ventilation waveform |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102770069B (en) | Nitric oxide measurement method and apparatus | |
JP6564318B2 (en) | A system for evaluating real-time lung mechanics | |
EP1551293B1 (en) | System and method for the automatic detection of the expiratory flow limitation | |
JP5053083B2 (en) | Method and apparatus for non-invasive prediction of positive end expiratory pressure (PEEPi) in a patient receiving ventilator support | |
US8100836B2 (en) | Augmented RIC model of respiratory systems | |
CN108670200A (en) | A kind of sleep sound of snoring classification and Detection method and system based on deep learning | |
CN103169476B (en) | Method and device for identification and early warning of respiration wave form image | |
JP2005537068A5 (en) | ||
CN109350063B (en) | Respiratory mechanics parameter detection device and method suitable for chronic obstructive pulmonary disease monitoring | |
CN111973188A (en) | Method for estimating respiratory mechanics parameter based on neural network | |
JP2014526926A (en) | Event sequencing and method using acoustic breathing markers | |
CN101133954B (en) | Respiration frequency detecting method and apparatus | |
Raoufy et al. | Nonlinear model for estimating respiratory volume based on thoracoabdominal breathing movements | |
CN111278355A (en) | Method for early identification of recurrence of chronic obstructive pulmonary disease | |
CN103735267B (en) | A kind of device based on sound of snoring examination OSAHS | |
CN115775629B (en) | Method for assessing risk of PPCs (point-to-point) based on sub-polar motion test before operation | |
CN109659031A (en) | Lung function index prediction meanss and determining method | |
CN111840931A (en) | Double-level respiratory function monitoring and intervention equipment | |
Saatci et al. | Respiratory parameter estimation in non-invasive ventilation based on generalized Gaussian noise models | |
CN114176563B (en) | Attached wireless respiration monitoring device, computer equipment and storage medium | |
CN116153517B (en) | Recovery evaluation method and system based on lung function exercise monitoring model | |
CN115227923A (en) | Model construction method for evaluating patient airway resistance and lung compliance, terminal device and storage medium | |
Brook et al. | Correction and standardisation of lung oscillometry techniques using parameter inference: A study group report | |
Gajda et al. | Identification of the human respiratory system during experiment with negative pressure impulse excitation | |
Kirkness et al. | Developing quantitative physiological phenotypes of sleep apnea for epidemiological studies |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201124 |