CN117174294A - Method and system for constructing slow-resistance lung evaluation model - Google Patents
Method and system for constructing slow-resistance lung evaluation model Download PDFInfo
- Publication number
- CN117174294A CN117174294A CN202311155305.5A CN202311155305A CN117174294A CN 117174294 A CN117174294 A CN 117174294A CN 202311155305 A CN202311155305 A CN 202311155305A CN 117174294 A CN117174294 A CN 117174294A
- Authority
- CN
- China
- Prior art keywords
- carbon dioxide
- respiratory
- phase
- volumetric
- data
- 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
- 210000004072 lung Anatomy 0.000 title claims abstract description 40
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000013210 evaluation model Methods 0.000 title abstract description 10
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 claims abstract description 217
- 229910002092 carbon dioxide Inorganic materials 0.000 claims abstract description 106
- 239000001569 carbon dioxide Substances 0.000 claims abstract description 106
- 230000000241 respiratory effect Effects 0.000 claims abstract description 65
- 238000012549 training Methods 0.000 claims abstract description 34
- 230000000877 morphologic effect Effects 0.000 claims abstract description 27
- 238000003745 diagnosis Methods 0.000 claims abstract description 17
- 230000029058 respiratory gaseous exchange Effects 0.000 claims abstract description 16
- 239000013598 vector Substances 0.000 claims abstract description 6
- 108700025647 major vault Proteins 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 8
- 238000011156 evaluation Methods 0.000 claims description 7
- 208000006545 Chronic Obstructive Pulmonary Disease Diseases 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 230000000903 blocking effect Effects 0.000 claims description 4
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 230000001174 ascending effect Effects 0.000 claims description 2
- 238000007619 statistical method Methods 0.000 claims description 2
- 230000002685 pulmonary effect Effects 0.000 abstract description 8
- 230000004199 lung function Effects 0.000 description 8
- 238000009423 ventilation Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 229910002090 carbon oxide Inorganic materials 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 3
- 230000015556 catabolic process Effects 0.000 description 3
- 238000006731 degradation reaction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000017531 blood circulation Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000001771 impaired effect Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000000414 obstructive effect Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 208000024172 Cardiovascular disease Diseases 0.000 description 1
- 208000017667 Chronic Disease Diseases 0.000 description 1
- 208000014085 Chronic respiratory disease Diseases 0.000 description 1
- 206010028980 Neoplasm Diseases 0.000 description 1
- 208000011623 Obstructive Lung disease Diseases 0.000 description 1
- 208000037883 airway inflammation Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229940124630 bronchodilator Drugs 0.000 description 1
- 239000000168 bronchodilator agent Substances 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 230000001684 chronic effect Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 206010012601 diabetes mellitus Diseases 0.000 description 1
- 230000004064 dysfunction Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 208000010125 myocardial infarction Diseases 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000009325 pulmonary function Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 210000002345 respiratory system Anatomy 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The application provides a method and a system for constructing a slow-resistance lung evaluation model, comprising the following steps: acquiring historical human respiratory data to obtain a carbon dioxide concentration sequence and a respiratory flow rate sequence; marking the data containing slow pulmonary resistance in the historical human respiratory data; determining a volumetric carbon dioxide waveform map based on the carbon dioxide concentration sequence and the respiration flow rate sequence; extracting characteristics of the volumetric carbon dioxide waveform graph to obtain morphological characteristics and derivative parameters of the volumetric carbon dioxide waveform graph; constructing a training model, and training by taking morphological characteristics of a volumetric carbon dioxide waveform curve and derivative parameters thereof as feature vectors of a training set; and collecting the respiratory data of the detected human body, inputting the respiratory data into a trained model, and outputting a diagnosis result. The application can learn and train the morphological characteristics of the volumetric carbon dioxide waveform and the derivative parameters thereof by utilizing the respiratory data naturally captured by the measured human body in the calm respiratory state, and provides an alternative means for diagnosing and grading the slow-resistance lung.
Description
Technical Field
The application relates to the technical field of digital medical treatment, in particular to a method and a system for constructing a slow-resistance lung evaluation model.
Background
Chronic obstructive pulmonary disease is a chronic respiratory disease characterized primarily by airflow limitation and has become one of the world 'four-major chronic diseases' including cardiovascular diseases (e.g., heart attacks and strokes), cancer, diabetes. The prevalence rate of slow lung obstruction is high, the mortality/disability rate is high, and the life quality of patients is seriously affected. The diagnosis and grading of the slow-blocking lung is a primary link of treatment and management of the slow-blocking lung, and is important for prolonging the life of a patient and improving the quality of life.
Current clinical assessment of chronic obstructive pulmonary disease relies primarily on lung function examination and clinical symptoms. The first second forced expiratory volume (Forced Expiratory Volume in second, FEV 1) and forced vital capacity (Forced Vital Capacity, FVC) after patient inhalation of bronchodilators are two important indicators for diagnosing slow lung resistance, and for determining the severity of airflow obstruction in slow lung resistance patients. The traditional method has great limitation in infants, old people and people with serious degradation of lung function, and patients cannot successfully complete lung function examination due to self limitation, so that misdiagnosis and missed diagnosis are extremely easy to occur, and follow-up treatment and rehabilitation are influenced. Therefore, a more convenient, rapid and reliable slow-resistance lung evaluation system is needed to fill the blank of the detection object of the traditional method for detecting the slow-resistance lung.
Disclosure of Invention
The application aims to provide a method and a system for constructing a slow-resistance lung evaluation model, which are used for solving the problems of severe lung function examination and test standard, complex acquisition flow and great limitation in infants, old people and people with serious lung function degradation in the prior art.
To solve the above technical problems, according to some embodiments, the present application provides a method for constructing a slow-resistance lung evaluation model, including:
acquiring historical human respiratory data to obtain a carbon dioxide concentration sequence and a respiratory flow rate sequence; marking the data containing slow pulmonary resistance in the historical human respiratory data;
determining a volumetric carbon dioxide waveform map based on the carbon dioxide concentration sequence and the respiration flow rate sequence;
extracting characteristics of the volumetric carbon dioxide waveform graph to obtain morphological characteristics of the volumetric carbon dioxide waveform graph corresponding to the exhalation process and derivative parameters thereof;
constructing a training model, and training by taking morphological characteristics and derivative parameters of a volumetric carbon dioxide waveform curve as a training data set;
and collecting the respiratory data of the detected human body, inputting the respiratory data into a trained model, and outputting a diagnosis result.
Further, before determining the volumetric carbon dioxide waveform, the method further comprises:
denoising the respiratory flow rate sequence.
Further, before determining the volumetric carbon dioxide waveform, the method further comprises:
and interpolating the carbon dioxide concentration sequence according to the sampling rate of the respiratory flow rate sequence to obtain a reconstructed carbon dioxide concentration sequence.
Further, determining a volumetric carbon dioxide waveform map, comprising:
and fitting the reconstructed carbon oxide concentration sequence and the expired gas volume by taking the reconstructed carbon dioxide concentration sequence as a vertical axis and the expired gas volume as a horizontal axis to obtain a volumetric carbon dioxide oscillogram.
Further, before the feature extraction of the volumetric carbon dioxide waveform diagram, the method comprises:
dividing a volumetric carbon dioxide oscillogram into a phase I, a phase II and a phase III; phase I is baseline, indicating that early exhaled gas is free of carbon dioxide; phase II is the rising phase, which means that the air in alveoli is mixed with the air in the airway; stage III is plateau phase, and represents the highest concentration of pure alveolar gas carbon dioxide in the late stage of expiration.
Further, morphological features of the volumetric carbon dioxide waveform profile, comprising:
the carbon dioxide concentration value corresponding to the inflection point between adjacent phases, the volume of the next exhaled breath, the phase II slope, the phase III slope, the ratio of the phase II to the phase III slope, the included angle corresponding to the inflection point of the phase I and the phase II, the included angle corresponding to the inflection point of the phase II and the phase III, the area under the phase II curve and the area under the phase III curve.
Further, derived parameters are obtained from the morphological features by statistical analysis, the derived parameters include:
the concentration of carbon dioxide is 0% -25% -50% -75%, the volume of the expired gas and the area under the curve are corresponding to each interval, and the stage III carbon dioxide stage-stage concentration is the same; and the average value, the maximum value and the minimum value of the carbon dioxide concentration values of the inflection points in the breathing periods and the average value, the maximum value and the minimum value of the exhaled air volumes of the inflection points in the breathing periods.
Further, training includes:
and determining the respiratory airflow obstruction degree to divide a plurality of grades according to the morphological characteristics of the volumetric carbon dioxide waveform curve and the derivative parameters thereof, and calculating the probability of each grade.
Further, outputting the diagnostic result includes:
whether the detected human respiratory data contains slow lung blocking or not; if yes, outputting the level and the probability of the respiratory airflow obstruction degree.
Another aspect of the present application proposes a system for constructing a slow-blocking lung assessment model, comprising:
the acquisition module is used for acquiring the respiratory data of the slow-resistance lung human body to obtain a carbon dioxide concentration sequence and a respiratory flow rate sequence; marking the data of slow lung resistance contained in the historical human respiratory data; the device is also used for collecting the breathing data of the detected human body;
a data processing module for determining a volumetric carbon dioxide waveform map based on the carbon dioxide concentration sequence and the respiratory flow rate sequence; the method is also used for extracting the characteristics of the volumetric carbon dioxide oscillogram to obtain morphological characteristics of a carbon dioxide oscillogram corresponding to the expiration process and derivative parameters thereof;
the model training module is used for constructing a training model, taking morphological characteristics of the carbon dioxide waveform curve and derivative parameters thereof as feature vectors of a training set, taking diagnosis standards of slow lung resistance in GOLD (Global Initiative for Chronic Obstructive Lung Disease) guidelines as labels, and training data;
and the slow-resistance lung evaluation module is used for inputting the acquired respiratory data of the detected human body into the trained model and outputting a diagnosis result.
The technical scheme of the application has at least the following beneficial technical effects:
the method and the system for constructing the slow-resistance lung evaluation model can utilize the breathing data naturally captured by the tested human body in the calm breathing state, do not need complicated breathing skill guidance, and are suitable for tested persons in all age groups; the method is simple and easy, low in cost and high in sensitivity, and provides a new alternative means for diagnosing and classifying the slow-resistance lung.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the conventional technology, the drawings required for the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to the drawings without inventive effort for those skilled in the art.
FIG. 1 is a flow chart of a method of constructing a slow-blocking lung evaluation model in one embodiment of the application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the embodiments of the present application will be described in detail below with reference to the accompanying drawings. However, it will be understood by those of ordinary skill in the art that in various embodiments of the present application, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, the claimed technical solution of the present application can be realized without these technical details and various changes and modifications based on the following embodiments. The following embodiments are divided for convenience of description, and should not be construed as limiting the specific implementation of the present application, and the embodiments can be combined with each other and cited with each other without contradiction.
At present, the lung function inspection and test standard is harsh, the acquisition flow is complex, and the problems of great limitation exist in infants, old people and people with serious lung function degradation in the prior art.
In order to solve the above problems, a method for constructing a slow-resistance lung evaluation model according to an embodiment of the present application comprises the following specific steps:
synchronously acquiring respiratory data of a slow-resistance lung human body in a natural respiratory state by using a signal acquisition device, and acquiring historical human body respiratory data to obtain a carbon dioxide concentration sequence and a respiratory flow rate sequence; marking the data containing slow pulmonary resistance in the historical human respiratory data; wherein the collection device can be a differential pressure sensor and a carbon dioxide sensor; the data containing the slow-blocking lung is referred to GOLD (Global Initiative for Chronic Obstructive Lung Disease) guidelines, and the pulmonary ventilation function of the subject is evaluated based on diagnosis by a professional clinical technician and doctor, including diagnosis of the slow-blocking lung and the severity level of the slow-blocking lung.
Pretreatment of the obtained carbon dioxide concentration sequence and respiratory flow rate sequence: the respiratory flow rate sequence pretreatment includes: denoising the breathing flow rate sequence; interpolating the carbon dioxide concentration sequence according to the sampling rate of the respiratory flow rate sequence to obtain a reconstructed carbon dioxide concentration sequence; and fitting the reconstructed carbon oxide concentration sequence and the breathing flow velocity sequence by taking the reconstructed carbon oxide concentration sequence as a vertical axis and the volume of exhaled gas as a horizontal axis to obtain a volumetric carbon dioxide oscillogram.
During the acquisition process, due to the instability of the respiratory flow rate, certain Gaussian noise exists, which is unfavorable for the later feature extraction. Thus, the stop band frequency is removed and the amplitude-frequency characteristic of the signal (respiratory flow rate sequence) is maintained using a butterworth-filtered low-pass filter, the filter amplitude-frequency characteristic curve being described as follows:
where G (ω) is the gain of the filter, H (jω) is its frequency response, G 0 Gain of DC signal, w c For the filter cut-off frequency, n is the filter order.
Pretreatment of carbon dioxide concentration sequences: the sampling rate of the carbon dioxide concentration sequence is usually lower than that of the respiratory flow rate sequence, so that interpolation is needed to be carried out on the original carbon dioxide concentration sequence to ensure the smoothness of the data and facilitate the subsequent fitting with the respiratory flow rate sequence. The interpolation mode selects a cubic spline interpolation method, and the specific function is solved as follows:
(1) Input known node (x i ，y i ) (i=0, 1,., n) and satisfies the boundary condition y' 1 ＝y″ n ＝0；
(2) Calculate h j ＝x j+1 -x j (j＝1，2...，n-1)；
(3) Calculation of
(4) Calculation of
(5) Solving a system of equations using a catch-up method, wherein α 1 ＝1，γ n ＝1；
(6) Outputting the expression of each interval;
(7) And calculating interpolation to obtain a reconstructed carbon dioxide concentration sequence.
After obtaining the sequence of the reconstructed carbon dioxide concentration, drawing a volumetric carbon dioxide oscillogram:
the chronic obstructive pulmonary patient has impaired pulmonary function due to the effects of airway inflammation, and the like, gas exchange is disordered, and the content of carbon dioxide in the exhaled gas is changed compared with that of normal people. The monitoring of the volume carbon dioxide can effectively detect the noninvasive monitoring of the respiratory system function, and the gas exchange condition of the lung can be reflected by measuring the change of the concentration of carbon dioxide exhaled by a detected human body along with the volume of exhaled air, so that the lung ventilation function of the detected human body can be further characterized. In addition, the volumetric carbon dioxide diagram can be naturally captured in the calm breathing state of the detected human body, complicated breathing skill guidance is not needed, and the volumetric carbon dioxide diagram is suitable for subjects of all age groups and particularly has excellent performance in the elderly and infants.
In the embodiment, the reconstructed carbon oxide concentration sequence is taken as a vertical axis, the exhaled gas volume is taken as a horizontal axis, and the reconstructed carbon oxide concentration sequence and the exhaled gas volume are fitted to obtain a volumetric carbon dioxide oscillogram, wherein the exhaled gas volume is determined according to the denoised respiratory flow velocity sequence; and performing curve fitting on the waveform by using parameter optimization of a polynomial. The waveform function model is expressed as:
f(t，x)＝f 0 (t，x)+f 1 (t，x)+f 2 (t，x)；
where t is an argument, x is a parameter vector, x= [ x ] 1 ，x 2 ，x 3 ，x 4 ，x 5 ，x 6 ，x 7 ]The method comprises the steps of carrying out a first treatment on the surface of the Wherein,
f 0 (t,x)＝x 1
optimizing by utilizing Levenberg-Marquardt algorithmThe iterative solution is completed by the nonlinear least square method; minimizing residual functionsAnd (3) completing calculation of the carbon dioxide concentration sequence and the fitting parameters of the expired gas volume curve to obtain a volumetric carbon dioxide oscillogram.
Extracting characteristics of the volumetric carbon dioxide oscillogram;
dividing the volumetric carbon dioxide waveform map into three parts, the first phase (phase I) being the baseline, the phase being early exhalation gas, free of carbon dioxide; the second phase (II phase) is an ascending phase, the gas in alveoli is mixed with the gas in the airway, and the concentration of carbon dioxide is gradually increased; the third phase (phase III) is the plateau phase, and the concentration of pure alveolar gas carbon dioxide reaches the highest at the late stage of expiration. Abnormal changes in the volumetric carbon dioxide waveform profile during exhalation can occur when pulmonary ventilation is impaired.
The application utilizes the fitting curve of the volumetric carbon dioxide waveform to extract morphological characteristics with physiological significance, and the main morphological characteristic parameters specifically comprise: the carbon dioxide concentration value corresponding to the inflection point between adjacent phases, the volume of the next exhaled breath, the phase II slope, the phase III slope, the ratio of the phase II to the phase III slope, the included angle corresponding to the inflection point of the phase I and the phase II, the included angle corresponding to the inflection point of the phase II and the phase III, the area under the phase II curve, the area under the phase III curve and the like.
Compared with healthy people, the lung elasticity and the airway structure of the patient with slow resistance lung are changed, and the influence of the changes can be obtained from the morphological characteristics. The amount of gas from the initial expiratory volume to the end of expiration in a complete expiration is related to the expiratory flow rate and the elasticity of the lungs and can be used as a basis for the evaluation of the lung function. Specifically, phase II is a phase in which the expiratory flow rate gradually decreases, and its slope reflects the airflow dynamics at the early stage of expiration, and can generally reflect the small airway resistance change of the human body; stage III is the end-tidal phase, with the slope representing the degree of gas-air mixing at the end of the airway. It is often affected by airway branches, such as an imbalance in ventilation/blood flow ratio; the included angles at each stage can reflect the smoothness and elasticity of the airway, smaller included angles are generally associated with increased airway resistance, while larger included angles may be associated with changes in airway branching and elastic properties; the area of each phase is generally related to the gas mixing at the end of the airway and the ventilation/blood flow ratio. The morphological characteristics of the volumetric capnography proposed by the model are all based on clinical researches and have physiological meanings and clinical interpretability.
Further, the morphological features are statistically analyzed to obtain derived parameters, the derived parameters including: the concentration of carbon dioxide is 0% -25% -50% -75%, the volume of the expired gas and the area under the curve are corresponding to each interval, and the stage III carbon dioxide stage-stage concentration is the same; and the average value, the maximum value and the minimum value of the carbon dioxide concentration values of the inflection points in the breathing periods and the average value, the maximum value and the minimum value of the exhaled air volumes of the inflection points in the breathing periods.
The derived parameters provide the mathematical characteristics of the volumetric carbon dioxide waveform from a statistical level, identify abnormal patterns, and compare the differences in different individual or time step data. As a supplement to morphological features, the auxiliary model comprehensively evaluates pulmonary ventilation conditions, and performs a slow obstructive pulmonary assessment.
Training a model;
firstly, dividing a plurality of grades for the obstruction degree of the respiratory airflow according to morphological characteristics of a volumetric carbon dioxide waveform curve and derivative parameters thereof, and calculating the probability of each grade.
Constructing a training model, and training morphological characteristics and derivative parameters of the graded volumetric carbon dioxide waveform curve as feature vectors of a training set; the markers contain data of slow-blocking lungs.
The LightGBM model is selected as a training model, and has the advantages of being fast and efficient, good in expandability, high in accuracy and the like. In terms of model parameter setting, 500 trees were used, each tree having a maximum depth of 3, a learning rate of 0.1, and a minimum number of samples of leaf nodes of 20. During model training, independent tests are adopted, independent subjects are used as division basis, and feature leakage is avoided. The historical human breath data is divided into a training set and a verification set, and the training and parameter adjustment are carried out in a cross verification mode so as to prevent over fitting and under fitting. One-hot encoding whether a slow obstructive pulmonary disease is present or not and grading ventilation dysfunction; for each sample, the LightGBM classifies it to the corresponding leaf node according to the decision tree combination of the model and obtains a predictive score on each leaf node. And carrying out weighted summarization on the prediction scores according to the weights and votes of the leaf nodes, and outputting class labels of tasks to obtain a final prediction result. In the training process, early stop ping technology is used, when the loss function on the verification set is not reduced continuously for a plurality of times, the training is stopped to avoid over fitting, and finally, the slow-resistance lung diagnosis and classification model with higher accuracy rate is obtained through training.
And collecting the respiratory data of the detected human body, inputting the collected respiratory data of the detected human body into a trained model, and outputting a diagnosis result. The output diagnosis result specifically includes: judging whether the detected human respiratory data contains slow lung blocking; if yes, outputting the level and the probability of the respiratory airflow obstruction degree.
In this embodiment, the process from the respiratory data acquisition, preprocessing, feature extraction, to the processing data of the model, and the diagnostic results, grading results are optionally visualized.
Another aspect of the present application proposes a system for constructing a slow-blocking lung assessment model, comprising:
the acquisition module is used for acquiring the respiratory data of the slow-resistance lung human body to obtain a carbon dioxide concentration sequence and a respiratory flow rate sequence; marking the data of slow lung resistance contained in the historical human respiratory data; the device is also used for collecting the breathing data of the detected human body;
a data processing module for determining a volumetric carbon dioxide waveform map based on the carbon dioxide concentration sequence and the respiratory flow rate sequence; the method is also used for extracting the characteristics of the volumetric carbon dioxide oscillogram to obtain morphological characteristics of a carbon dioxide oscillogram corresponding to the expiration process and derivative parameters thereof;
the model training module is used for constructing a training model, taking morphological characteristics of the carbon dioxide waveform curve and derivative parameters thereof as feature vectors of a training set, taking diagnosis standards of slow lung resistance in GOLD (Global Initiative for Chronic Obstructive Lung Disease) guidelines as labels, and training data;
and the slow-resistance lung evaluation module is used for inputting the acquired respiratory data of the detected human body into the trained model and outputting a diagnosis result.
In this embodiment, the system for constructing the slow-resistance-lung evaluation model further includes a visualization client, the visualization client performs preliminary display on the detected human respiratory data detected by the signal acquisition module, and completes data uploading and storage, and then performs operations such as preprocessing and feature extraction on the capnography signals through the signal processing module, sends the extracted features into the trained LightGBM model to perform diagnosis evaluation on the slow-resistance-lung in sequence, and gives a slow-resistance-lung evaluation report.
In the description of the present application, the terms "one embodiment," "some embodiments," "particular embodiments," and the like, mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. In the present application, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
It is to be understood that the above-described embodiments of the present application are merely illustrative of or explanation of the principles of the present application and are in no way limiting of the application. Accordingly, any modification, equivalent replacement, improvement, etc. made without departing from the spirit and scope of the present application should be included in the scope of the present application. Furthermore, the appended claims are intended to cover all such changes and modifications that fall within the scope and boundary of the appended claims, or equivalents of such scope and boundary.
Claims (10)
1. A method for constructing a slow-blocking lung assessment model, comprising:
acquiring historical human respiratory data to obtain a carbon dioxide concentration sequence and a respiratory flow rate sequence; marking the data of slow lung resistance contained in the historical human respiratory data;
determining a volumetric carbon dioxide waveform map based on the carbon dioxide concentration sequence and the respiratory flow rate sequence;
extracting characteristics of the volumetric carbon dioxide waveform graph to obtain morphological characteristics of the volumetric carbon dioxide waveform graph corresponding to the expiration process and derivative parameters thereof;
constructing a training model, and training by taking morphological characteristics and derivative parameters of the volumetric carbon dioxide waveform curve as a training data set;
and collecting the respiratory data of the detected human body, inputting the respiratory data into a trained model, and outputting a diagnosis result.
2. The method of claim 1, further comprising, prior to said determining the volumetric carbon dioxide waveform map:
denoising the respiratory flow rate sequence.
3. The method of claim 1, further comprising, prior to said determining the volumetric carbon dioxide waveform map:
and interpolating the carbon dioxide concentration sequence according to the sampling rate of the respiratory flow velocity sequence to obtain a reconstructed carbon dioxide concentration sequence.
4. A method according to claim 3, wherein said determining a volumetric carbon dioxide waveform map comprises:
and fitting the reconstructed carbon dioxide concentration sequence and the expired gas volume by taking the reconstructed carbon dioxide concentration sequence as a vertical axis and the expired gas volume as a horizontal axis to obtain the volumetric carbon dioxide oscillogram.
5. The method of claim 1, wherein prior to feature extraction of the volumetric carbon dioxide waveform map, comprising:
dividing the volumetric carbon dioxide oscillogram into a phase I, a phase II and a phase III; phase I is a baseline, indicating that early exhaled gas is free of carbon dioxide; the phase II is an ascending phase, and represents that air in alveoli and air in an air passage are mixed; the stage III is a plateau stage, and represents the highest concentration of pure alveolar gas carbon dioxide in the late stage of expiration.
6. The method of claim 5, wherein the morphological feature of the volumetric carbon dioxide waveform profile comprises:
the carbon dioxide concentration value corresponding to the inflection point between adjacent phases, the volume of the next exhaled breath, the phase II slope, the phase III slope, the ratio of the phase II to the phase III slope, the included angle corresponding to the inflection point of the phase I and the phase II, the included angle corresponding to the inflection point of the phase II and the phase III, the area under the phase II curve and the area under the phase III curve.
7. The method of claim 6, wherein the derived parameters are obtained from a statistical analysis of the morphological features, the derived parameters comprising:
the concentration of carbon dioxide is 0% -25% -50% -75%, the volume of the expired gas and the area under the curve are corresponding to each interval, and the stage III carbon dioxide stage-stage concentration is the same; and means, maxima, and minima of the carbon dioxide concentration values at the plurality of inflection points over the plurality of respiratory cycles, and means, maxima, and minima of the exhaled breath volume at the plurality of inflection points over the plurality of respiratory cycles.
8. The method of claim 7, wherein the training comprises:
and determining the respiratory airflow blocking degree to divide a plurality of grades according to the morphological characteristics of the volumetric carbon dioxide waveform curve and the derivative parameters thereof, and calculating the probability of each grade.
9. The method of claim 8, wherein outputting the diagnostic result comprises:
whether the detected human respiratory data contains slow lung blocking or not; if yes, outputting the level and the probability of the respiratory airflow obstruction degree.
10. A system for constructing a slow-blocking lung assessment model, comprising:
the acquisition module is used for acquiring the respiratory data of the slow-resistance lung human body to obtain a carbon dioxide concentration sequence and a respiratory flow rate sequence; marking the data of slow lung resistance contained in the historical human respiratory data; the device is also used for collecting the breathing data of the detected human body;
a data processing module for determining a volumetric carbon dioxide waveform map based on the carbon dioxide concentration sequence and the respiratory flow rate sequence; the method is also used for extracting the characteristics of the volumetric carbon dioxide oscillogram to obtain morphological characteristics of a carbon dioxide oscillogram corresponding to the expiration process and derivative parameters thereof;
the model training module is used for constructing a training model, taking morphological characteristics of the carbon dioxide waveform curve and derivative parameters thereof as feature vectors of a training set, taking diagnosis standards of slow lung resistance in GOLD (GlobalInitiative for Chronic Obstructive Lung Disease) guidelines as labels, and training data;
and the slow-resistance lung evaluation module is used for inputting the acquired respiratory data of the detected human body into the trained model and outputting a diagnosis result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311155305.5A CN117174294A (en) | 2023-09-07 | 2023-09-07 | Method and system for constructing slow-resistance lung evaluation model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311155305.5A CN117174294A (en) | 2023-09-07 | 2023-09-07 | Method and system for constructing slow-resistance lung evaluation model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117174294A true CN117174294A (en) | 2023-12-05 |
Family
ID=88931500
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311155305.5A Pending CN117174294A (en) | 2023-09-07 | 2023-09-07 | Method and system for constructing slow-resistance lung evaluation model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117174294A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117672532A (en) * | 2024-01-31 | 2024-03-08 | 吉林大学 | Hospitalized patient nursing risk assessment early warning monitoring system and method |
-
2023
- 2023-09-07 CN CN202311155305.5A patent/CN117174294A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117672532A (en) * | 2024-01-31 | 2024-03-08 | 吉林大学 | Hospitalized patient nursing risk assessment early warning monitoring system and method |
| CN117672532B (en) * | 2024-01-31 | 2024-04-05 | 吉林大学 | Hospitalized patient nursing risk assessment early warning monitoring system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US5800361A (en) | Non-invasive estimation of arterial blood gases | |
| US5632281A (en) | Non-invasive estimation of arterial blood gases | |
| JP5431971B2 (en) | End-breathing gas estimation system and method | |
| CN108670200A (en) | A kind of sleep sound of snoring classification and Detection method and system based on deep learning | |
| CN102172328A (en) | Method for detecting and disciminatng breathing patterns from respiratory signals | |
| CA2464029A1 (en) | Non-invasive ventilation monitor | |
| CN110533658A (en) | Intelligent pulmonary emphysema diagnostic message processing system and method, information data processing terminal | |
| CN112842321B (en) | Lung ventilation function detection method, device and medium based on flow-volume loop diagram | |
| CN117174294A (en) | Method and system for constructing slow-resistance lung evaluation model | |
| CN111598868B (en) | Lung ultrasonic image identification method and system | |
| Doheny et al. | Estimation of respiratory rate and exhale duration using audio signals recorded by smartphone microphones | |
| CN118133209B (en) | Data processing method based on breath trainer data analysis | |
| US10426426B2 (en) | Methods and apparatus for performing dynamic respiratory classification and tracking | |
| Kazemi et al. | Frequency analysis of capnogram signals to differentiate asthmatic and non-asthmatic conditions using radial basis function neural networks | |
| Bhagya et al. | Speed of sound-based capnographic sensor with second-generation CNN for automated classification of cardiorespiratory abnormalities | |
| CN108742628A (en) | A kind of breathing detection method | |
| Faria et al. | Forced oscillation technique in the detection of smoking-induced respiratory changes | |
| Kazemi et al. | Assessment on methods used to detect asthmatic and non-asthmatic conditions using capnogram signal | |
| Pham et al. | Automated quality control of forced oscillation measurements: respiratory artifact detection with advanced feature extraction | |
| Lutfi | Vital capacity derived spirometric measurements | |
| CN116108345B (en) | Second heart sound width division detection method based on parameter estimation | |
| Karasan et al. | An enhanced mechanistic model for capnography, with application to CHF-COPD discrimination | |
| TWI837948B (en) | Method for fully detecting apnea and hypopnea events | |
| CN118121184A (en) | Pulmonary function assessment method, system and equipment based on respiratory impedance change | |
| WO2021010150A1 (en) | Respiratory function testing device, data processing method, and respiratory function testing system |
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 |