Quantitative evaluation method for respiratory training

Abstract

The invention discloses a respiratory training quantitative evaluation method, which comprises the following steps of: acquiring physiological parameter characteristics related to heart rate, respiration and brain function of a human body through a human body physiological data acquisition platform, and performing dimensionality reduction, dimensionless processing and weighted summation on the characteristics to obtain a respiratory training quantitative evaluation result. The method comprises the following specific steps: firstly, collecting human heart rate, respiration and brain function related physiological features through a physiological parameter collection platform, then calculating feature classification weight of each dimension through random forest and cross validation and removing redundant feature dimensions, and then removing dimensions of features in a feature subset through a dimensionless method; and finally, performing weighted summation on the feature data through the feature weight to obtain a respiratory training quantitative evaluation result. Quantitative evaluation is carried out on respiratory training, so that people can be helped to formulate more scientific respiratory training strategies and methods.

Description

Technical Field

[0001] The invention relates to the technical field of signal processing, and in particular to a quantitative evaluation method for breathing training. Background Art

[0002] Breathing is a chemical process in which cells of organisms decompose organic matter through mitochondrial oxidation and generate energy. Breathing is closely related to the physiological health of the human body. Insufficient breathing will lead to insufficient oxygen content in the body, resulting in lack of nutrients in cells, and may also cause anxiety, depression and other problems. Breathing training can restore and enhance the physiological and psychological functions of the human body through controlled breathing exercises. Breathing training is also a widely used intervention measure to help individuals cope with fatigue and stress. In addition, in the process of learning wind instruments, breathing training can improve the performer's skills and endurance, and significantly improve the performer's performance such as pitch and timbre. However, improper breathing training methods and intensity can also cause harm to the human body. Therefore, it is necessary to evaluate breathing training to guide the designation and implementation of strategies.

[0003] At present, the mainstream evaluation method for respiratory training is to evaluate through vital capacity and respiratory muscle related indicators. In this method, the respiratory muscle measurement process requires the use of professional medical instruments, which is relatively complicated to implement. In addition, the current indicators of respiratory training are mainly based on respiratory related functions, lacking multi-dimensional and comprehensive information. In addition, the interpretation of current respiratory training indicators is relatively complicated, and people cannot intuitively obtain information about the intensity of respiratory training from the test data. Summary of the invention

[0004] The purpose of the present invention is to solve the defects of the prior art that data collection is complex and indicators are difficult to interpret, a human physiological parameter collection platform is constructed to realize the collection of human physiological signals, random forest and cross-validation are used to reduce the dimension of features and obtain the importance weights of feature subsets, dimensionlessization is used to remove the signal dimension, and the importance weights of feature subsets are used to perform weighted summation on each signal to obtain a quantitative scoring result of respiratory training, thereby realizing low-cost data collection and objective and perceptible evaluation results; reflecting the physiological state of the human body through the characteristics of breathing, vital capacity, and brain function, and enhancing the accuracy of respiratory training evaluation; by reducing the dimension of high-dimensional human physiological characteristics and extracting importance weights, the reduced-dimensional features are compressed to the same dimension, a feature subset with strong correlation with respiratory training is obtained and the feature dimension is removed, the quantitative scoring of respiratory training is realized, and the quantitative evaluation of respiratory training is effectively realized.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions: A quantitative evaluation method for breathing training is provided, which collects multi-dimensional physiological parameter features of a human body through a human physiological parameter collection platform, reduces the dimension of the collected features and removes redundant feature dimensions, removes the dimension of each feature in the feature subset, and obtains an accurate quantitative score of breathing training. The evaluation method comprises the following steps: S1. Collect multi-dimensional human physiological parameter characteristics through the human physiological parameter collection platform; S2, obtain feature subsets and their weights through random forest and cross validation; S3, removing the dimension of each feature in the feature subset by dimensionless transformation; S4. Obtain the quantitative evaluation result of breathing training by weighted summation of feature weights.

[0006] Further, in step S1, the human physiological parameter acquisition platform is composed of a control module, a heart rate signal acquisition module, a respiratory signal acquisition module and a brain function signal acquisition module, wherein the control module guides the test process, processes and analyzes the collected raw data and scores them, the heart rate signal acquisition module acquires the raw data of the heart rate signal, the respiratory signal acquisition module acquires the raw data related to the vital capacity signal through the vital capacity acquisition unit, acquires the raw data related to the expiratory airflow signal through the expiratory airflow acquisition unit, combines the raw data of the vital capacity and the expiratory airflow signal to form complete raw data of the respiratory signal, and the brain function signal acquisition module acquires the original brain function signal; The process of using the control module to guide the parameter collection process and drive the heart rate signal collection module, the breathing signal collection module, and the brain function signal collection module to collect the original parameters of multi-dimensional human physiological parameters, and processing and analyzing the original parameters to obtain the multi-dimensional physiological parameter characteristics of the human body is as follows: S101, the control module guides the user to complete the heart rate original signal acquisition process, the heart rate signal acquisition module collects face video through the video recording device, and the control module processes the face video through the RPPG algorithm to extract heart rate related features. The steps are as follows: S101.1. In an environment with sufficient light, the control module prompts the user to keep a steady heart rate and keep the body still. At the same time, the heart rate signal acquisition module uses a video recording device to collect face videos in the background. After the face video is collected, the control module completes subsequent signal processing and analysis. S101.2, dividing the collected face video into frames to obtain a face video frame sequence, so as to facilitate subsequent tasks such as face recognition and face feature point detection; S101.3, performing face recognition and face feature point detection on the face video image sequence, and locating the region of interest according to the face feature points; S101.4, extracting the color channel values ​​of the pixels in the region of interest as the original pulse wave signal; S101.5, performing modal decomposition on the original pulse wave signal to obtain multiple intrinsic mode function (IMF) components, performing fast Fourier transform on each IMF component, calculating the frequency corresponding to the maximum peak in the spectrum diagram, and classifying each component into a high-frequency component and a low-frequency component according to a certain frequency threshold; S101.6, performing denoising processing on the selected high-frequency IMF component by wavelet threshold filtering to remove noise in the high-frequency signal; S101.7. Remove the baseline drift of the selected low-frequency IMF and remove the low-frequency components obtained through modal decomposition. Offsets in the signal due to instrument drift, environmental changes, or other low-frequency factors; S101.8, superimposing the high-frequency IMF component processed in step S103.6 and the low-frequency IMF component processed in step S103.7 to reconstruct the denoised signal; S101.9, performing bandpass filtering on the denoised signal to enhance the signal-to-noise ratio of the signal, and obtaining a final pulse wave signal of the region of interest; S101.10, extracting heart rate related features from the final pulse wave signal by using time domain features; S102, the control module guides the user to complete the breathing-related parameter collection process, the breathing signal collection module drives the vital capacity collection unit to collect the user's original vital capacity signal, drives the airflow signal collection unit to collect the user's original expiratory airflow signal, and the control module obtains the vital capacity characteristics and expiratory airflow characteristics after processing and analysis, and combines them to form a complete breathing-related characteristic. The specific steps are as follows: S102.1. The control module prompts the user to take a deep breath first while being relaxed and calm, and then use the vital capacity collection unit to quickly exhale to the limit, and the vital capacity collection unit collects and records the expiratory airflow signal when the user exhales quickly; S102.2. The control module processes and analyzes the expiratory airflow signal when the user exhales rapidly to obtain the vital capacity characteristics of the user; S102.3. The control module prompts the user to take a deep breath first in a relaxed and calm state, and then use the airflow signal acquisition unit to breathe slowly, steadily, and for a long time until the limit is reached. The airflow signal acquisition unit acquires and records the exhalation airflow signal when the user exhales steadily for a long time; S102.4. The control module processes and analyzes the exhalation airflow signal when the user exhales steadily for a long time to obtain the exhalation airflow characteristics of the user; S102.5, the control module summarizes the vital capacity and expiratory airflow related characteristics to obtain complete breathing characteristics; S103, the control module guides the user to complete the memory experiment, the brain function acquisition module displays symbols on the display device, the user memorizes and inputs the memorized symbols through the input device, and the control module calculates the user's memory accuracy as the brain function feature. The specific steps are as follows: S103.1. The control module prompts the user to look at the display device in a good mental state, and the display device sequentially displays any number of characteristic symbols such as numbers, characters, and graphics, and the user memorizes these symbols and the corresponding sequence; S103.2. After a certain period of time, these symbols disappear; S103.3. The user inputs the memorized symbols and corresponding sequences through an input device within a certain time range; S103.4, repeat steps S103.1-S103.3 for a certain number of rounds; S103.5, the control module calculates the accuracy of the characters input by the user, and uses the accuracy as the user's brain function feature; S104, after the control module completes the processing of the original data related to heart rate, breathing, and brain function, it analyzes and obtains the characteristics of each dimension, and obtains and displays the quantitative scoring results of breathing training through subsequent processing; Furthermore, the process of obtaining the feature subset and its weight in step S2 is as follows: S201, in step S1, the evaluation features of multiple physiological parameters are obtained, and feature subsets are constructed by random extraction of these features, and random extraction is performed multiple times to obtain feature subsets with different numbers of features; S202. Evaluate and score each feature subset. The evaluation indicators can be common feature evaluation indicators such as precision and accuracy. Precision indicates the proportion of positive samples that are actually positive. Recall indicates the proportion of all samples that are actually positive that are correctly predicted to be positive. In terms of scoring, f1-weighted is the most commonly used feature subset scoring indicator. f1-weighted comprehensively considers factors such as category frequency. The final score is proportional to the importance of the feature subset. After completion, the corresponding relationship between the number of features and the score in different feature subsets is obtained; S203, calculating the number of features with the highest score based on the corresponding relationship between the number of features and the score in step S202; S204, processing the features, by removing different features and calculating the scores, calculating the importance weight of each feature from the change in the scores, removing the features with the lowest weight, iterating multiple times until the number of features is the number of features with the highest score in step S203, reducing the influence of irrelevant features, preventing the dimensionality disaster, and finally obtaining the feature subset with the strongest correlation; S205, by removing different features and calculating the scoring scores, the importance weight of each feature in the feature subset obtained in step S204 is calculated according to the change of the scores; Furthermore, the process of removing the dimension of each feature in the feature subset by dimensionless transformation in step S3 is as follows: S301, for different feature dimensions in different feature subsets in step S2, analyze the corresponding feature distribution characteristics, transform the feature direction, and offset the feature value by the reference value, transform each feature into a value range of positive numbers and positively correlated with the physiological state of the human body, such as the heart rate value is negatively correlated with the physiological state of the human body, and obtain the transformed feature by taking its negative value and adding an offset value; S302, through the relationship between the difference between the feature value and the feature extreme value, compress the value range of all features so that the value range of all features is between [0,1], and obtain dimensionless and comparable features of the same dimension; Furthermore, the process of obtaining the quantitative evaluation result of breathing training by weighted summation of feature weights in step S4 is as follows: S401: Compare the importance weight percentages of each parameter feature and the corresponding parameter feature dimension obtained in step S205 Multiply them to obtain the weighted value of each parameter; S402, adding weighted values ​​of each parameter to obtain a quantitative evaluation score of the breathing training, where the score ranges from [0, 1]; S403. The human physiological parameter acquisition platform displays the quantitative evaluation score of the breathing training.

[0007] Compared with the prior art, the present invention has the following advantages and beneficial effects: (1) The present invention collects the original data related to human heart rate, respiration, and brain function through a low-cost, easy-to-use heart rate, exhalation, and brain function physiological parameter collection module, simplifies the collection process of relevant parameters in the respiratory training evaluation process, and reduces the demand for complex medical instruments.

[0008] (2) The present invention simplifies the process of multi-dimensional physiological parameter collection through the guidance of the control module, and at the same time improves the accuracy of the test results. With the guidance of the control module, the user can easily complete the entire test process independently without additional guidance and teaching, which reduces the complexity of the entire system and solves the problem of complex physiological parameter collection process.

[0009] (3) The present invention combines cross-validation with feature selection to calculate the importance of multi-dimensional physiological parameter features, and based on the number of optimal feature subsets obtained by cross-validation, removes the physiological parameter features with the lowest importance one by one through feature elimination, thereby removing redundant feature data dimensions and improving computational efficiency and accuracy.

[0010] (4) The present invention removes the dimension of the feature by dimensionless transformation and performs weighted summation based on the importance weights of each feature to obtain a quantitative scoring result of breathing training. The quantitative scoring can help people establish an accurate and intuitive understanding of the intensity of breathing training and help people make more scientific decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 is a flow chart of a quantitative evaluation method for breathing training in an embodiment of the present invention; Figure 2 is an overall structural diagram of a human physiological signal acquisition platform in an embodiment of the present invention; Figure 3 is a green channel waveform diagram of the original pulse wave signal in an embodiment of the present invention; Figure 4 is a schematic diagram of the result of peak value detection of a filtered signal in an embodiment of the present invention; Figure 5 is a schematic diagram of a signal showing a change in the exhaled air pressure relative to the atmospheric pressure when a human body blows steadily in an embodiment of the present invention; Figure 6 It is the result of truncating the portion of the air pressure signal in the time domain where the signal value is 0 in the embodiment of the present invention; Figure 7 This is a waveform diagram of the air pressure signal after removing the unstable parts at the beginning and end stages in the embodiment of the present invention; Figure 8 is a waveform diagram of an air flow velocity signal in an embodiment of the present invention; Fig. 9 It is a schematic diagram of the corresponding relationship between the number of features in different feature subsets and the f1-weighted score in an embodiment of the present invention. DETAILED DESCRIPTION

[0013] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present application.

[0014] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments.

[0015] Example This embodiment discloses a quantitative evaluation method for breathing training, such as Figure 1 As shown, the specific steps are as follows: S1. Collect multi-dimensional physiological parameter characteristics of the human body through the human physiological parameter acquisition platform. The human physiological parameter acquisition platform consists of a control module combined with a heart rate, breathing, and brain function original signal acquisition module, which guides users to collect original data and analyze to obtain multi-dimensional physiological parameter characteristics of the human body. The specific steps are as follows: S101, the control module is a computer program that guides the entire test process. At the same time, it processes and analyzes the original data collected by the heart rate, breathing, and brain function signal acquisition modules to obtain corresponding features. The heart rate signal acquisition module collects face video through a computer camera. After completion, the control module analyzes and obtains heart rate-related features. In the respiratory signal acquisition module, the vital capacity acquisition unit collects the air pressure signal when the human body exhales with all its strength through an air pressure sensor, and the exhalation airflow acquisition unit collects the air pressure signal when the human body exhales steadily through an air pressure sensor. The control module calculates the airflow velocity signal through the relationship formula between airflow and air pressure, and then obtains vital capacity characteristics and airflow-related characteristics. The brain function signal acquisition module collects the original signal of human memory ability through the memory UI test program, and the control module analyzes the memory accuracy to obtain the human brain function characteristics. After obtaining the multi-dimensional physiological characteristics of the human body, the control module performs comprehensive processing and analysis to obtain a quantitative score for breathing training. Figure 2 This is the overall structure diagram of the human physiological signal acquisition platform; S102, the control module prompts the user to look directly at the computer screen, and the user starts the test process after being ready; S103, the screen displays a 2-minute countdown, and at the same time the heart rate signal acquisition module drives the computer camera background to 1920×1080 resolution 30FPS to collect user face video signals; S104, after the screen countdown ends, the control module records the face video signal and obtains the human heart rate related features through the RPPG algorithm. The specific steps are as follows: S104.1. Frame the collected face video according to the FPS to obtain a face video frame sequence of 30 frames per second; S104.2. Perform face recognition and facial feature point detection on the face video images one by one through the machine learning face recognition library mediapipe, obtain 68 facial key points of the face, and then locate the region of interest based on the facial feature points; S104.3, extract the green pixel color channel value of the region of interest, and calculate its average value as the signal value, Figure 3 This is the green channel signal waveform of the original pulse wave signal; S104.4, performing empirical mode decomposition on the preprocessed pulse wave signal to obtain five intrinsic mode function components, performing fast Fourier transform on each IMF component to calculate the frequency corresponding to the maximum peak, and using a certain frequency threshold as a classification to divide the IMF component into four high-frequency components and one fundamental frequency component; S104.5, performing wavelet denoising on the selected four high-frequency IMF components, the wavelet basis function in the wavelet threshold denoising is the "db9" wavelet, the wavelet decomposition layer number is 3, the threshold function is the global hard threshold function, and the threshold is 0.1 times the maximum value of the signal; S104.6. Performing baseline fitting on the selected low-frequency IMF through quadratic polynomial regression model operation and removing baseline drift to obtain a signal after drift removal; S104.7, linearly add the four high-frequency IMF components processed in step S104.5 and the one low-frequency IMF component processed in step S104.6 to obtain a pure denoised signal; S104.8, processing the reconstructed pulse wave signal using a Butterworth bandpass filter, with the filter order set to 5 and the filter bandpass frequency set to 0.6-3.1 Hz; S104.9. Detect the peaks in the bandpass filtered signal using a peak detection algorithm. After obtaining the peak value of the signal, the heart rate can be calculated by counting the number of peaks within one minute. The instantaneous heart rate and heart rate variability and other related human physiological parameter characteristics can be calculated by the time interval between adjacent peaks. Figure 4 A schematic diagram showing the results of peak detection of the filtered signal; S105, the control module instructs the user to collect the user's breathing-related signals through the breathing signal collection module, and the specific steps are as follows: S105.1. After the control module drives the air pressure sensor in the vital capacity collection unit to start collecting data for a few seconds, it instructs the user to relax and take a deep breath, then blow quickly towards the air pressure sensor, and stop after blowing to the limit. After a few seconds of stopping, the control module stops recording the signal value of the air pressure sensor; S105.2, the control module calculates the change value of the air pressure in the air pressure signal compared to the atmospheric pressure; S105.3. The equation of state for an ideal gas is: Where P represents pressure, V represents gas volume, represents the number of moles of gas, represents the ideal gas constant and T represents the temperature.

[0016] When blowing, the calculation formula of gas pressure and blowing air flow rate is: in Indicates the gas pressure. Indicates the density of the gas (unit: kg / m³), Indicates the flow rate of gas (unit: m / s).

[0017] The control module obtains the exhalation airflow signal when the user blows with full force through the relationship formula between airflow and air pressure, and then calculates the user's vital capacity characteristics by integrating the airflow velocity over time and multiplying it by the exhalation cross-section; S105.4. The control module drives the air pressure sensor in the exhalation airflow signal collection unit to start collecting data for a few seconds, and then instructs the user to relax the body and take a deep breath, and then blow air steadily and slowly toward the air pressure sensor, and stop blowing after the air reaches the limit. After stopping for a few seconds, the control module stops recording the signal value of the air pressure sensor; S105.5, the control module calculates the change value of the air pressure in the air pressure signal compared to the atmospheric pressure, Figure 5 This is a schematic diagram of the change signal of the exhaled air pressure relative to the atmospheric pressure when the human body blows steadily; S105.6. Calculate the time domain value of the start and end stage signal value of the air pressure signal change value when it is 0, perform time domain truncation on the signal, and remove the signal period without exhalation at the beginning and end. Figure 6 It is the result of truncating the part of the pressure signal with a signal value of 0 in the time domain; S105.7. Cut off 5% of the time length of the signal at the beginning and end of the signal time domain, remove the unstable part at the beginning and end of the signal acquisition, and retain the stable part of the signal. Figure 7 The waveform diagram of the air pressure signal after removing the unstable parts at the beginning and end stages; S105.8. The control module obtains the airflow velocity signal when the user blows slowly and steadily through the relationship between airflow and air pressure. Figure 8 is the waveform diagram of air flow velocity signal; S105.9. The control module analyzes the airflow velocity signal when the user blows slowly and steadily and obtains airflow-related characteristics; S106, the control module guides the user to perform a memory experiment through the memory UI test program, the specific process is as follows: S106.1. The memory UI test program displays a 20-digit random number on the computer screen, and the screen displays a countdown for 20 seconds, and the user memorizes the number; S106.2. After the countdown is completed, the number disappears and an input box is displayed. At the same time, the countdown starts again for 20 seconds. The user then copies the number according to his memory. S106.3. When the countdown ends, the user input ends and cannot be input again, and one round of experiment ends; S106.4. Experimental steps S106.1-S106.3 are repeated for 3 rounds, and the memory UI test program records the random numbers displayed in each round and the corresponding input numbers of the user; S106.5. The control module calculates the average memory accuracy of the user in three rounds of experiments as an evaluation feature of the brain function index.

[0018] S2. Remove redundant data dimensions through random forest and cross-validation to obtain feature subsets and their weights. The specific steps are as follows: S201, in step S1, the evaluation features of multiple physiological parameters are obtained, and feature subsets are constructed by random extraction of these features, and random extraction is performed multiple times to obtain feature subsets with different numbers of features; S202, evaluate and score each feature subset, using common feature evaluation indicators such as precision and accuracy, and using f1-weighted as the feature subset scoring indicator. After completion, the corresponding relationship between the number of features and the scoring scores in different feature subsets is obtained. Fig. 9 The corresponding relationship between the number of features in different feature subsets and the f1-weighted score is shown; S203, from the corresponding relationship between the number of features and the scoring scores in step S202, it is calculated that the highest scoring score is when the number of features is 8, and the score is also high when the number of features is 4-5; S204, processing the features, removing different features and calculating the scoring scores, calculating the importance weight of each feature from the change in the scores, removing the feature with the lowest weight, iterating multiple times until the number of features is 8, and obtaining the feature parameter set at this time, which is 8 parameters including average airflow velocity, airflow skewness, root mean square velocity, average RR interval, maximum vital capacity, coefficient of variation of airflow velocity, heart beats per minute, and memory accuracy. Considering that the scoring score when the number of features is 5 is close to that when the number is 8, and the airflow velocity-related data in the feature parameter set may have a strong cross-correlation, the average airflow velocity, average RR interval, maximum vital capacity, heart beats per minute, and memory accuracy are finally selected as evaluation indicators; S205, by removing different features and calculating the scoring scores, the importance weights of each feature in the feature subset in step S204 are calculated from the changes in the scores as follows: Table 1. Importance weight table of each feature S3. Remove the dimension of each feature in the feature subset by dimensionless transformation: S301. For different feature dimensions in different feature subsets in step S2, the corresponding dimensionless conversion methods shown in the following table are used for processing: Table 2. Dimensionless table After dimensionless processing, the value range of each dimension feature data is compressed to the interval [0,1], and all are positively correlated with breathing training; S4. Obtain the quantitative evaluation results of breathing training by weighted summation of feature weights: S401, multiplying each parameter feature by the importance weight percentage of the corresponding parameter feature dimension to obtain a weighted value of each parameter; S402, adding weighted values ​​of each parameter to obtain a quantitative evaluation score of the breathing training; S403, the control module displays the quantitative evaluation score of the breathing training. Through the evaluation score, the user can accurately understand the intensity of the current breathing training, confirm whether the breathing training is reasonable, and formulate a more scientific breathing training execution strategy.

[0019] In summary, in view of the problem that physiological parameter detection is complex and costly, the present invention proposes for the first time to characterize and calculate respiratory training by using simple and low-cost heart rate, breathing, and brain function related features, thereby eliminating the complex testing process of respiratory muscle related parameters. The physiological parameter detection process is simplified by the human physiological parameter acquisition platform, and fast and low-cost data acquisition is achieved. By introducing heart rate related features and brain function related features, and adding expiratory airflow related features to traditional vital capacity features, the comprehensiveness and scientificity of the evaluation indicators are enhanced. In view of the problem that the interpretation of respiratory training indicators is complex and is not conducive to the large-scale promotion of respiratory training, the user's respiratory training intensity is quantitatively evaluated by using physiological parameters. The quantitative standard objectively and accurately reflects the physiological state of the human body, can accurately evaluate the intensity of respiratory training, and provide objective, real and perceptible evaluation results, which is conducive to users making more scientific adjustments during respiratory training.

[0020] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

[0021] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A quantitative evaluation method for breathing training, characterized in that: The respiratory training quantitative assessment method comprises the following steps: S1. Collect multi-dimensional human physiological parameter characteristics through the human physiological parameter collection platform; S2, obtain feature subsets and their feature weights through random forest and cross validation; S3, removing the dimension of each human physiological parameter feature in the feature subset by dimensionless transformation; S4. Perform weighted summation of the human physiological parameter features processed in step S3 by using feature weights to obtain a quantitative evaluation result of the breathing training.

2. The quantitative evaluation method for respiratory training according to claim 1, characterized in that: In step S1, the human physiological parameter acquisition platform is composed of a control module, a heart rate signal acquisition module, a respiratory signal acquisition module and a brain function signal acquisition module. The control module guides the parameter acquisition process and drives the heart rate signal acquisition module, the respiratory signal acquisition module and the brain function signal acquisition module to acquire multi-dimensional original parameters of human physiological parameters, and processes and analyzes the original parameters to obtain multi-dimensional physiological parameter characteristics of the human body.

3. The quantitative evaluation method for respiratory training according to claim 2, characterized in that: The specific process of step S1 collecting multi-dimensional human physiological parameter characteristics through the human physiological parameter collection platform is as follows: S101, the control module guides the user to complete the heart rate original signal acquisition process, the heart rate signal acquisition module collects face video through the video recording device, and the control module processes the face video to extract heart rate related features; S102, the control module guides the user to complete the breathing-related parameter collection process, the breathing signal collection module drives the vital capacity collection unit to collect the user's original vital capacity signal, drives the airflow signal collection unit to collect the user's original expiratory airflow signal, and the control module obtains the vital capacity characteristics and expiratory airflow characteristics after processing and analysis, and combines them to form a complete breathing-related characteristic; S103, the control module guides the user to complete the memory experiment, the brain function acquisition module displays symbols on the display device, the user memorizes and inputs the memorized symbols through the input device, and the control module calculates the user's memory accuracy as the brain function feature; S104, after the control module completes the processing of the original data related to heart rate, breathing, and brain function, it analyzes and obtains the characteristics of each dimension, and obtains and displays the quantitative scoring results of the breathing training through subsequent processing.

4. The quantitative evaluation method for respiratory training according to claim 1, characterized in that: The process of obtaining feature subsets and their weights through random forest and cross validation in step S2 is as follows: S201, constructing feature subsets of the above-mentioned heart rate-related features, breathing-related features, and brain function features by random sampling; S202, evaluate and score each feature subset, and the final score is proportional to the importance of the feature subset. After completion, the corresponding relationship between the number of features and the score in different feature subsets is obtained; S203, calculating the number of features when the score is the highest based on the corresponding relationship between the number of features and the score; S204, processing the features, removing different features and calculating the scoring scores, calculating the importance weight of each feature from the change in the score, removing the feature with the lowest weight, iterating multiple times and finally obtaining the feature subset with the strongest correlation; S205 , by removing different features and calculating the scoring scores, the importance weight of each feature in the feature subset obtained in step S204 is calculated based on the change in the scores.

5. The quantitative evaluation method for respiratory training according to claim 1, characterized in that: The process of removing the dimension of each feature in the feature subset by dimensionless transformation in step S3 is as follows: S301, for different feature dimensions, analyzing the corresponding value distribution characteristics, by transforming the feature direction and offsetting the feature value by the reference value, transforming each feature into a positive number with a value range, and the positive number is positively correlated with the physiological state of the human body; S302: Compress the value ranges of all features so that the value ranges of all features are between [0, 1].

6. The quantitative evaluation method for respiratory training according to claim 4, characterized in that: The process of obtaining the quantitative score of breathing training in step S4 is as follows: S401, multiplying each human physiological parameter feature by the human physiological parameter feature importance weight obtained in step S205 to obtain a weighted value of each human physiological parameter feature; S402, adding up the weighted values ​​of the physiological parameter characteristics of each human body to obtain a quantitative evaluation score of the breathing training, where the score ranges from [0, 1]; S403. The human physiological parameter acquisition platform displays the quantitative evaluation score of the breathing training.

Images