Mouth-nose breathing data acquisition and analysis method and system

Abstract

The invention discloses a method and a system for collecting and analyzing oral-nasal respiration data, relates to the technical field of medical monitoring, and aims to solve the problem that self-adaptive threshold adjustment based on a traditional image detection algorithm cannot be suitable for continuous and repeated graphic modes. Dividing a plurality of partial images based on image edge information, collecting mouth and nose breathing data variation rates, breathing frequency change intensity and texture complexity of the partial images, establishing a data analysis model, obtaining an image evaluation coefficient, comparing the image evaluation coefficient with an image threshold, and counting the number and proportion of important partial images; through the network state information obtained through network detection and the proportion of important parts in the image, the transmission duration is determined through the fuzzy Bayesian neural network, after the transmission duration is compared with the preset transmission duration, the transmission image is adjusted according to the comparison result, the limitation of a traditional image detection algorithm is solved, and meanwhile network resource consumption is reduced.

Description

Technical Field

[0001] The present invention relates to the field of medical monitoring technology, and more specifically, to a method and system for collecting and analyzing oral and nasal breathing data. Background Art

[0002] In recent years, with the development of medical technology and the increasing demand for health monitoring, oral and nasal respiratory data collection and analysis systems have been widely used in clinical diagnosis, health monitoring, exercise physiology and other fields. The existing technology can collect basic respiratory data such as respiratory rate, airflow, gas composition, etc., and has the function of storing and transmitting data, and can transmit data to PC or cloud server via wired or wireless means.

[0003] The prior art has the following deficiencies:

[0004] At present, wireless transmission mainly relies on network bandwidth to analyze the image data collected by the oral and nasal tubes and transmit it to the PC or cloud server. However, there are also factors such as network delay and signal interference. The threshold of the existing image detection algorithm is usually based on the Otsu algorithm to obtain the threshold for threshold segmentation to solve the limitation of manually setting high and low thresholds in the traditional algorithm. However, the oral and nasal breathing images form repeated graphic patterns in the spectrum diagram, and their subtle changes cannot be detected, which leads to errors in data analysis and excessive consumption of network resources. Therefore, a method and system for collecting and analyzing oral and nasal breathing data are proposed.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method and system for collecting and analyzing oral and nasal respiratory data, which solves the problems raised in the above-mentioned background technology by using different product inspection methods.

[0007] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a method for collecting and analyzing oral and nasal breathing data, comprising S1: reducing noise interference by graying and filtering the oral and nasal breathing image, extracting features from the image using a texture analysis method, dividing the image into multiple partial images according to image edge information, collecting partial images, obtaining the oral and nasal breathing data variability rate, respiratory frequency change intensity and texture complexity through an image edge detection algorithm, and collecting network status information through network monitoring;

[0008] S2: Obtain the variability rate of oral and nasal breathing data, the intensity of respiratory frequency change, and texture complexity of some images, establish a data analysis model, and obtain the image evaluation coefficient;

[0009] S3: Obtaining the image evaluation coefficient and comparing it with the image threshold to obtain the comparison result, and counting the number of important parts of the image and their proportion according to each comparison result and the number of comparison results;

[0010] S4; obtaining network status information and the important proportion of some images in the image, and obtaining the time domain upload bandwidth and time domain delay duration through data processing;

[0011] S5: According to the time domain upload bandwidth, the time domain delay duration and the importance ratio of some images in the image, the transmission duration is determined using a fuzzy Bayesian neural network, and compared with the preset transmission duration. The transmitted image is readjusted according to the comparison result to obtain an adjustment result.

[0012] In a preferred embodiment, specifically, the color image is converted into a grayscale image, the image is smoothed using a Gaussian filter G to reduce noise, and the smoothed image is divided into multiple partial images using a Canny edge detection algorithm, and the specific steps are:

[0013] Step A1: Apply the Canny edge detection algorithm to detect edges in the smoothed image;

[0014] Step A2: Detect contours from edge detection results;

[0015] Step A3: For each contour, create a binary mask;

[0016] Step A4: Apply the mask to the original image to extract the corresponding part of the image.

[0017] In a preferred embodiment, the original image is divided into a plurality of partial images, wherein c represents the cth partial image;

[0018] By collecting the inhalation data below or above the high threshold and the exhalation data below or above the low threshold, the number of occurrences is obtained, and the ratio of the number of occurrences to the total number of breaths is used to obtain the variability rate of oral and nasal breathing data Ar. c ;

[0019] By determining the healthy respiratory cycle interval, the respiratory frequencies in some images that are greater than or less than the healthy respiratory cycle interval are collected, and the absolute value of the difference between the maximum and minimum values ​​of the respiratory cycle interval and the maximum and minimum values ​​of the healthy respiratory cycle interval are subtracted to obtain the relative deviation of the single respiratory frequency, and the difference is added to obtain the respiratory frequency change intensity Dv c ;

[0020] The texture complexity Te is calculated by extracting texture features through the gray level co-occurrence matrix c .

[0021] In a preferred embodiment, the variability rate of oral and nasal breathing data, the intensity of respiratory frequency change and texture complexity are obtained, and a data analysis model is established to obtain the image evaluation coefficient Pt c , based on the following formula:

[0022]

[0023] Where Pt c is the image evaluation coefficient, as well as are the preset proportional coefficients of texture complexity, variability of oral and nasal breathing data, and intensity of breathing frequency change, respectively, and as well as Both are greater than 0.

[0024] In a preferred embodiment, an image evaluation coefficient is obtained, and the image evaluation coefficient is compared and analyzed with a continuously iterated image threshold;

[0025] If the image evaluation coefficient is greater than or equal to the image threshold, the part of the image is marked as an important image and an important signal is generated;

[0026] If the image evaluation coefficient is less than the image threshold, the portion of the image is marked as a non-important image and an end signal is generated;

[0027] The image evaluation coefficient calculation will be performed multiple times to obtain the image evaluation coefficient values ​​of multiple partial images in one image and two groups of comparison results of the multiple partial images; based on the comparison results, the image evaluation coefficient values ​​and their proportions of the partial images that are statistically evaluated as important signals are obtained.

[0028] In a preferred embodiment, the network status information includes the time domain upload bandwidth and the time domain delay duration:

[0029] By dividing the time domain into different sizes and marking the time domain as i, and dividing n time domains, the time domain upload bandwidth B (T i );

[0030] After dividing the time domain, use the ping command or other network measurement tools to measure the delay and obtain the time domain delay length D (T i );

[0031] The importance ratio Ze of partial images in the image is obtained by the ratio of the total number of partial images in the image to the important images.

[0032] In a preferred embodiment, a fuzzy Bayesian neural network is used to determine the transmission duration, the time domain upload bandwidth, the time domain delay duration, and the important proportion of the partial image in the image are defined as input nodes in the input layer, and the transmission duration is defined as the output layer;

[0033] Formulate a fuzzy Bayesian neural network to describe the impact of time domain upload bandwidth, time domain delay duration, and the importance ratio of some images in the image on the transmission duration;

[0034] According to the fuzzy Bayesian neural network, the transmission duration is determined by the time domain upload bandwidth, the time domain delay duration and the importance ratio of some images in the image.

[0035] In a preferred embodiment, after the transmission duration is determined, it is compared with the preset transmission duration. If it is less than the preset transmission duration, important images in the image are transmitted to the PC. Otherwise, an alarm is issued and the number of resources required for one transmission is adjusted.

[0036] An oral and nasal breathing data acquisition and analysis system includes an acquisition module, a processing module, an image analysis module, a resource allocation module and an output terminal;

[0037] The acquisition module is used to acquire images, divide them into multiple partial images according to the image edge information, acquire partial images, obtain the mouth and nose breathing data variation rate, breathing frequency change intensity and texture complexity through the image edge detection algorithm, and send them to the processing module, and collect network status information through network monitoring and send it to the resource allocation module;

[0038] The processing module is used to obtain the variability rate of some oral and nasal breathing data, the intensity of respiratory frequency change and texture complexity, establish a data analysis model, obtain the image evaluation coefficient, and send it to the image analysis module;

[0039] The image analysis module is used to obtain the image evaluation coefficient and compare it with the image threshold to obtain the comparison result. According to each comparison result and the number of comparison results, the number and proportion of important parts of the image are counted and sent to the resource allocation module.

[0040] The resource allocation module is used to obtain network status information and the important proportion of some images in the image, obtain the transmission time through the fuzzy Bayesian neural network, and compare it with the preset transmission time. According to the comparison result, the transmission image is readjusted and the adjustment result is sent to the output end;

[0041] The output end is used to receive the adjustment result and report it.

[0042] Technical effects and advantages of the present invention:

[0043] 1. The present invention divides an image into multiple partial images according to image edge information, collects texture complexity, oral and nasal breathing data variation rate and respiratory frequency change intensity of the multiple partial images, obtains image evaluation coefficients by establishing a data analysis model, and compares the coefficients with image thresholds to obtain comparison results. The number and proportion of important partial images in the image are counted according to the comparison results and the number of comparison results, which solves the limitations of traditional image detection algorithms in oral and nasal breathing data image detection, reduces network resources occupied by images while maintaining image quality, and improves the overall efficiency of image transmission.

[0044] 2. The present invention obtains network status information and the important proportion of some images in the image, establishes a fuzzy Bayesian algorithm through the time domain upload bandwidth, time domain delay duration and the important proportion of some images in the image to determine the transmission duration, and compares and analyzes with the preset transmission duration to obtain the transmission result, and adjusts the number of resources required for one transmission, dynamically adjusts the resources, reduces the impact of network congestion on transmission, and ensures the timely transmission and processing of key image information. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 The present invention is a method flow chart of a method for collecting and analyzing oral and nasal breathing data.

[0046] Figure 2 This is a module schematic diagram of an oral and nasal breathing data acquisition and analysis system of the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] Example 1

[0049] Oral and nasal breathing data is regular, and the time interval between each inhalation and exhalation forms a repeated graphic pattern in the spectrum diagram. The existing image edge detection segments the areas with obvious frequency intensity changes, but for repeated waveform images, it is not possible to detect the subtle degree of abnormality. Therefore, it is necessary to further limit the image edge detection algorithm to determine the important proportion of oral and nasal breathing image data.

[0050] See also Figure 1 , a method for collecting and analyzing oral and nasal breathing data, the specific operation process is as follows:

[0051] S1: Reduce noise interference by graying and filtering the oral and nasal breathing images, use texture analysis methods to extract features from the images, divide the images into multiple partial images based on image edge information, collect partial images, and obtain the oral and nasal breathing data variability rate, respiratory frequency change intensity, and texture complexity through image edge detection algorithms. At the same time, collect network status information through network monitoring.

[0052] Among them, image edge information includes the variability rate of oral and nasal breathing data, the intensity of respiratory frequency change, and texture complexity.

[0053] Among them, the image edge detection algorithm is used to identify areas in the image where the gray level changes sharply, which correspond to the boundaries of objects in the image.

[0054] Specifically, the color image is converted into a grayscale image, and the Gaussian filter G is used to smooth the image and reduce noise. The formula is expressed as: Among them, I blur (x, y) is the pixel value of the smoothed image at position (x, y), I gra y(x+u, y+v) is the pixel value of the grayscale image at position (x+u, y+v), G(u, k) is the weight value of the Gaussian filter at offset (u, k), u, k is the offset range of the filter kernel, usually the filter radius.

[0055] The smoothed image is divided into multiple partial images using the Canny edge detection algorithm. The specific steps are as follows:

[0056] Step A1: Apply the Canny edge detection algorithm to detect the edges in the smoothed image. The formula is as follows: E(x, y) = Canny(I blur (x,y),T low , T high ), where T low is the low threshold, used for edge connections; T high is a high threshold, used for strong edge detection;

[0057] Step A2: Detect contours from edge detection results: C = findContours(E), where C is a set of detected contours;

[0058] Step A3: For each contour C∈C, create a binary mask M c :

[0059]

[0060] Among them, M c (x, y) is the mask value of the pixel at (x, y), which is 255 inside the contour and 0 outside the contour;

[0061] Step A4: Set the mask M c Applied to the original image I to extract the corresponding part of the image I c ;

[0062]

[0063] Among them, I c (x, y) is the value of the extracted part of the image at the pixel point, M c When (x, y) = 255, I c (x, y) = I(x, y), otherwise I c (x, y)=0.

[0064] Through the above steps, the original image is divided into a plurality of partial images, wherein c represents the cth partial image;

[0065] Among them, the low threshold and the high threshold in step A1 are further limited. Since the oral and nasal breathing image data is a relatively repetitive waveform image, the peak and valley values ​​that are repeated the most times in the waveform image are analyzed and defined as the inhalation and exhalation stages. At the same time, the high threshold is defined according to the peak height and the low threshold is defined according to the valley depth; therefore, the high threshold and the low threshold of the Canny edge detection algorithm change with the change of the user.

[0066] The logic for obtaining the variability rate of oral and nasal breathing data is to collect the inhalation data below or above the high threshold and the exhalation data below or above the low threshold, obtain their occurrence times, and compare them with the total number of breaths to obtain the variability rate of oral and nasal breathing data Ar c .

[0067] The logic of obtaining the intensity of respiratory frequency change is to determine the healthy respiratory cycle interval, collect the respiratory frequencies in some images that are greater than or less than the healthy respiratory cycle interval, and subtract the absolute value of the maximum and minimum values ​​of the respiratory cycle interval from the absolute value of the maximum and minimum values ​​of the healthy respiratory cycle interval to obtain the relative deviation of the single respiratory frequency, and add them to obtain the respiratory frequency change intensity Dv c .

[0068] It should be noted that the healthy breathing cycle range is based on a cross-sectional study mentioned in "BMC Public Health", which includes the relationship between breathing patterns during spontaneous breathing and physical health. The parameters of the breathing pattern mentioned include respiratory rate (RR), which refers to the number of breaths per minute.

[0069] Among them, if the respiratory frequency is lower than the minimum value of the healthy respiratory cycle interval, the respiratory frequency is subtracted from the minimum value of the healthy respiratory cycle interval to indicate the degree of shortness of breath; if the respiratory frequency is higher than the maximum value of the healthy respiratory cycle interval, the respiratory frequency is subtracted from the maximum value of the healthy respiratory cycle interval to indicate the degree of bradypnea.

[0070] The logic of obtaining texture complexity is to extract texture features from GLCM to calculate texture complexity, and calculate image contrast, entropy value and energy according to the formula.

[0071] It should be noted that GLCM is the gray level co-occurrence matrix. Specifically, GLCM describes the frequency distribution of the co-occurrence of pixel gray values ​​at a given distance and direction, thereby reflecting the texture characteristics of the image; each element P(i, j) represents the frequency of the co-occurrence of a pixel with a gray value of i and a pixel with a gray value of j at a specified distance d and direction θ. The size of the matrix is ​​L×L, where L is the gray level of the image;

[0072] Among them, the contrast formula is: Describes the degree of local variation of gray levels in an image;

[0073] Among them, the entropy value formula is: Measures the randomness of the image grayscale distribution;

[0074] Among them, the energy formula is: Describes the uniformity of the grayscale of the image;

[0075] Specifically, when the contrast and entropy values ​​are higher, the texture randomness and complexity are higher. Conversely, when the energy is higher, the texture uniformity is higher and the texture complexity is lower.

[0076] Specifically, the texture complexity calculation formula is obtained by weighted average calculation, and the formula is as follows:

[0077] Te c =α1*Co c +α2*Et c -α3*En c

[0078] In the formula, Te c is the texture complexity, α1, α2 and α3 are preset proportional coefficients of contrast, entropy and energy, and α1, α2 and α3 are all greater than 0.

[0079] S2: Obtain the variability rate of oral and nasal breathing data, the intensity of respiratory frequency change, and texture complexity of some images, establish a data analysis model, and obtain the image evaluation coefficient;

[0080] Obtain the variability rate of oral and nasal breathing data, the intensity of respiratory frequency change, and texture complexity, establish a data analysis model, and obtain the image evaluation coefficient Pt c , based on the following formula:

[0081]

[0082] Where Pt c is the image evaluation coefficient, as well as are the preset proportional coefficients of texture complexity, variability of oral and nasal breathing data, and intensity of breathing frequency change, respectively, and as well as Both are greater than 0.

[0083] Among them, texture complexity, variability of oral and nasal breathing data, and intensity of respiratory frequency change are all data-based expressions of the image evaluation coefficient value and the current priority of partial image transmission;

[0084] It can be seen from the above formula that when the texture complexity, the variability rate of oral and nasal breathing data, and the intensity of respiratory frequency change are higher, it means that when part of the image is transmitted earlier, the image evaluation coefficient is larger, and vice versa, when part of the image is transmitted later, the image evaluation coefficient is smaller;

[0085] S3: Obtaining the image evaluation coefficient and comparing it with the image threshold to obtain the comparison result, and counting the number of important parts of the image and their proportion according to each comparison result and the number of comparison results;

[0086] The logic for obtaining the image threshold is to collect a set of distributed samples from a historical part of the image transmission database, divide the data set into a training set and an evaluation set, set the evaluation indicators and clustering algorithm, and in each round of cross-validation, train the model on the training set and evaluate the model performance on the evaluation set. Then, adjust the image threshold based on the performance of the validation set. Therefore, the image threshold is constantly iteratively updated.

[0087] In the present invention, the clustering algorithm is a type of unsupervised learning algorithm, which is used to standardize all image feature data to ensure that each feature is compared on the same scale, randomly select k initial centroids, and assign each partial image to the cluster where the nearest centroid is located according to the Euclidean distance. The new centroid of each cluster is calculated as the centroid of the next round of iteration, and the iteration is repeated until the centroid no longer changes or the change amplitude is less than the set threshold, and the final image threshold is determined.

[0088] Obtaining an image evaluation coefficient, and comparing and analyzing the image evaluation coefficient with the iterative image threshold;

[0089] If the image evaluation coefficient is greater than or equal to the image threshold, the part of the image is marked as an important image and an important signal is generated;

[0090] If the image evaluation coefficient is less than the image threshold, the portion of the image is marked as a non-important image and an end signal is generated;

[0091] The image evaluation coefficient calculation is performed multiple times to obtain the image evaluation coefficient values ​​of multiple partial images in one image and two groups of comparison results of the multiple partial images; based on the comparison results, the image evaluation coefficient values ​​and their proportions of the partial images evaluated as important signals are statistically analyzed;

[0092] The present invention divides an image into multiple partial images according to image edge information, collects texture complexity, oral and nasal breathing data variation rate and respiratory frequency variation intensity of the multiple partial images, obtains image evaluation coefficient by establishing a data analysis model, and compares it with the image threshold to obtain a comparison result, and counts the number of important partial images in the image and their proportion according to each comparison result and the number of comparison results, thereby solving the limitations of traditional image detection algorithms in oral and nasal breathing data image detection, reducing network resources occupied by images while maintaining image quality, and improving the overall efficiency of image transmission.

[0093] Example 2

[0094] In Example 1 of the present invention, an example is given of dividing an image into multiple partial images based on image edge information, collecting texture complexity, oral and nasal breathing data variation rate and respiratory frequency change intensity of the multiple partial images, obtaining an image evaluation coefficient by establishing a data analysis model, and comparing and analyzing the image evaluation coefficient with an image threshold to obtain a comparison result, and calculating the number of important partial images in the image and their numerical values ​​according to each comparison result and the number of comparison results; however, in Example 1, the image is only divided into multiple partial images, and the image importance is obtained by the importance ratio of the partial images. Obviously, although the image can be transmitted according to the importance, it cannot transmit all important images when it is applied to the scenario of high immediacy when network resources are scarce; in view of the above problems, Example 2 of the present invention is further refined;

[0095] S4: Obtain network status information and the important proportion of some images in the image, and obtain the time domain upload bandwidth and time domain delay duration through data processing;

[0096] The network status information includes the time domain upload bandwidth and the time domain delay duration;

[0097] The logic of obtaining time domain upload bandwidth is to divide the time domain into different sizes. For the traffic or the peak period of network usage, the time domain is planned, and the peak period and the off-peak period are marked as i, and n time domains are divided according to the formula: Among them, Data Size (T i ) is the amount of data transmitted in the i-th time domain, Time(T i ) is the duration of the i-th time domain, B(T i ) is the time domain upload bandwidth;

[0098] It should be noted that the bandwidth usage pattern is analyzed based on the formula: If B(T i )>μ b , then it is the peak period, if B(T i )≤μ b , then it is the off-peak period; the number of time domain divisions is set by the experimenters according to the specific implementation situation and is not limited here;

[0099] The logic for obtaining the time domain delay is to divide the time domain and use the ping command or other network measurement tools to measure the delay according to the formula Among them, D(T i ) represents the time period T i The time domain delay length within For time period T i The jth measured delay value within N i For time period T i The number of measurements within

[0100] The importance ratio of partial images within the image is obtained by the ratio of the total number of partial images within the image to the important images.

[0101] S5: According to the time domain upload bandwidth, the time domain delay duration and the important proportion of the partial image in the image, the transmission duration is determined using a fuzzy Bayesian neural network, and compared with the preset transmission duration. The transmitted image is readjusted according to the comparison result to obtain an adjustment result;

[0102] Fuzzy Bayesian neural network is a neural network model that combines fuzzy logic and Bayesian reasoning. Its network structure mainly includes input layer, hidden layer and output layer.

[0103] The input layer contains the feature vectors of the input data's time domain upload bandwidth, time domain delay duration, and the important proportion of the partial image in the image. Each feature vector corresponds to an input node of the network.

[0104] From the above, we can see that the time domain upload bandwidth, the time domain delay duration, and the important proportion of the partial image in the image are all single features for evaluating the transmission duration, so the input layer contains three input nodes;

[0105] The output layer generates the final output of the network, and each output node corresponds to the transmission time T evaluated by the network:

[0106] Below, the training method of the hidden layer is described in detail:

[0107] (1) By collecting historical data, the hidden layer is trained by continuously inputting the time domain upload bandwidth, time domain delay duration, the important proportion of the image part in the image, and the output transmission duration;

[0108] (2) Through the gradient descent method, the model parameters are initialized first, and the weight coefficients β1, β2, and β3 are set to input the time domain upload bandwidth, time domain delay duration, and the important proportion of the partial image in the image to obtain: β1*B(T i ),β2*D(T i ), β3*Ze; for each subsequent iteration, the gradient of the loss function with respect to the parameters needs to be calculated. The gradient shows the rate of change of the loss function at the current parameter value and the direction in which the parameters should be adjusted to reduce the loss; usually, the mean square error loss function is used as the loss function, and its mathematical expression is:

[0109]

[0110] Where J(θ) is the mean square error loss function, θ represents the parameters of the model, N is the number of samples, and y i is the true value of the i-th sample, is the value of the i-th sample predicted by the model; the sample refers to the drying time of the area, so as to calculate the loss; the gradient of the loss function with respect to the model parameters is calculated by the back propagation algorithm, which calculates the gradient layer by layer through the chain rule and propagates the gradient from the output layer to the input layer; now it is known that the output is T and the loss function is J(θ), then the gradient of the loss function with respect to the model parameter θ can be expressed as:

[0111]

[0112] (3) Update the parameters of the neural network based on the calculated gradient. The formula for updating the parameters is usually:

[0113] θ new =θ old - learning rate * gradient;

[0114] Among them, θ old is the current parameter value, the learning rate is a pre-set hyperparameter, and the gradient is calculated using the back-propagation algorithm described above;

[0115] The process of presetting the learning rate is usually based on historical data experience or trying different values. Generally speaking, the choice of learning rate depends on many factors, including the characteristics of the data set, the complexity of the model, the choice of optimization algorithm, and the convergence speed during training. Usually, the learning rate of the gradient descent method ranges from 0.1, 0.01, 0.001, etc.

[0116] It should be noted that a larger learning rate may cause the parameter update to be too large, making the optimization process unstable or even unable to converge; while a smaller learning rate may cause the optimization process to converge too slowly; during the training process, the learning rate usually needs to be adjusted to improve the performance and convergence speed of the model. A common strategy is to use the learning rate decay technique, that is, gradually reduce the learning rate as the training progresses, so as to search the parameter space more carefully when approaching the optimal solution;

[0117] The sizes of the training set and the validation set are divided. Generally speaking, the training set contains enough samples to ensure that the model has enough data to learn the patterns and features of the task, while the validation set is used to evaluate the generalization ability of the model and adjust the learning rate. Usually, the division is carried out in chronological order, that is, the earlier data is used as the training set and the newer data is used as the validation set. This can better simulate the generalization ability of the model for future data in actual applications. Of course, this embodiment only illustrates the detailed analysis of a certain method, and does not represent the uniqueness of this method, which will not be repeated here.

[0118] (4) Repeat the above steps (1) to (3) for iteration. By setting a tolerance parameter, the training is stopped when the training performance does not improve in several consecutive iterations, which can avoid overfitting and save resources.

[0119] When the transmission time is determined, it is compared with the preset transmission time. If it is less than the preset transmission time, the important images in the image are transmitted to the PC. Otherwise, an alarm is issued to adjust the number of resources required for one transmission.

[0120] It should be noted that the important images in the transmitted image, that is, the important partial images, need to mobilize and occupy different amounts of resources. The time domain upload bandwidth and time domain delay duration can reflect the amount of resources remaining in the current time domain, and the fuzzy Bayesian algorithm is used to determine the transmission duration;

[0121] The present invention obtains network status information and the important proportion of some images in the image, establishes a fuzzy Bayesian algorithm through the time domain upload bandwidth, time domain delay duration and the important proportion of some images in the image to determine the transmission duration, and compares and analyzes with the preset transmission duration to obtain the transmission result, and adjusts the number of resources required for one transmission, dynamically adjusts the resources, reduces the impact of network congestion on transmission, and ensures the timely transmission and processing of key image information.

[0122] Example 3

[0123] See also Figure 2 , an oral and nasal breathing data acquisition and analysis system, comprising an acquisition module, a processing module, an image analysis module, a resource allocation module and an output terminal;

[0124] The acquisition module is used to acquire images, divide them into multiple partial images according to the image edge information, acquire partial images, obtain the mouth and nose breathing data variation rate, breathing frequency change intensity and texture complexity through the image edge detection algorithm, and send them to the processing module, and collect network status information through network monitoring and send it to the resource allocation module;

[0125] The processing module is used to obtain the variability rate of some oral and nasal breathing data, the intensity of respiratory frequency change and texture complexity, establish a data analysis model, obtain the image evaluation coefficient, and send it to the image analysis module;

[0126] The image analysis module is used to obtain the image evaluation coefficient and compare it with the image threshold to obtain the comparison result. According to each comparison result and the number of comparison results, the number and proportion of important parts of the image are counted and sent to the resource allocation module.

[0127] The resource allocation module is used to obtain network status information and the important proportion of some images in the image, obtain the transmission time through the fuzzy Bayesian neural network, and compare it with the preset transmission time. According to the comparison result, the transmission image is readjusted and the adjustment result is sent to the output end;

[0128] The output end is used to receive the adjustment result and report it.

[0129] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0130] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0131] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0132] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0133] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0134] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0135] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0136] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0137] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0138] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for collecting and analyzing oral and nasal breathing data, characterized in that: include: S1: Reduce noise interference by graying and filtering the oral and nasal breathing images, extract features from the images using texture analysis methods, divide the images into multiple partial images based on image edge information, collect partial images, and obtain the oral and nasal breathing data variation rate, respiratory frequency change intensity, and texture complexity through image edge detection algorithms. At the same time, collect network status information through network monitoring; S2: Obtain the variability rate of oral and nasal breathing data, the intensity of respiratory frequency change, and texture complexity of some images, establish a data analysis model, and obtain the image evaluation coefficient; S3: Obtaining the image evaluation coefficient and comparing it with the image threshold to obtain the comparison result, and counting the number of important parts of the image and their proportion according to each comparison result and the number of comparison results; S4: Obtain network status information and the important proportion of some images in the image, and obtain the time domain upload bandwidth and time domain delay duration through data processing; S5: According to the time domain upload bandwidth, the time domain delay duration and the importance ratio of some images in the image, the transmission duration is determined using a fuzzy Bayesian neural network, and compared with the preset transmission duration. The transmitted image is readjusted according to the comparison result to obtain an adjustment result.

2. The method for collecting and analyzing oral and nasal breathing data according to claim 1, characterized in that: Specifically, the color image is converted into a grayscale image, the image is smoothed using a Gaussian filter G to reduce noise, and the smoothed image is divided into multiple partial images using the Canny edge detection algorithm. The specific steps are as follows: Step A1: Apply the Canny edge detection algorithm to detect edges in the smoothed image; Step A2: Detect contours from edge detection results; Step A3: For each contour, create a binary mask; Step A4: Apply the mask to the original image to extract the corresponding part of the image.

3. The method for collecting and analyzing oral and nasal breathing data according to claim 2, characterized in that: Divide the original image into multiple partial images, where c represents the cth partial image; By collecting the inhalation data below or above the high threshold and the exhalation data below or above the low threshold, the number of occurrences is obtained, and the ratio of the number of occurrences to the total number of breaths is used to obtain the variability rate of oral and nasal breathing data Ar. c ; By determining the healthy respiratory cycle interval, the respiratory frequencies in some images that are greater than or less than the healthy respiratory cycle interval are collected, and the absolute value of the difference between the maximum and minimum values ​​of the respiratory cycle interval and the maximum and minimum values ​​of the healthy respiratory cycle interval are subtracted to obtain the relative deviation of the single respiratory frequency, and the difference is added to obtain the respiratory frequency change intensity Dv c ; The texture complexity Te is calculated by extracting texture features through the gray level co-occurrence matrix c .

4. The method for collecting and analyzing oral and nasal breathing data according to claim 3, characterized in that: Obtain the variability rate of oral and nasal breathing data, the intensity of respiratory frequency change, and texture complexity, establish a data analysis model, and obtain the image evaluation coefficient Pt c , based on the following formula: Where Pt c is the image evaluation coefficient, as well as are the preset proportional coefficients of texture complexity, variability of oral and nasal breathing data, and intensity of breathing frequency change, respectively, and as well as Both are greater than 0.

5. The method for collecting and analyzing oral and nasal breathing data according to claim 4, characterized in that: Obtaining an image evaluation coefficient, and comparing and analyzing the image evaluation coefficient with the iterative image threshold; If the image evaluation coefficient is greater than or equal to the image threshold, the part of the image is marked as an important image and an important signal is generated; If the image evaluation coefficient is less than the image threshold, the portion of the image is marked as a non-important image and an end signal is generated; The image evaluation coefficient calculation will be performed multiple times to obtain the image evaluation coefficient values ​​of multiple partial images in one image and two groups of comparison results of the multiple partial images; based on the comparison results, the image evaluation coefficient values ​​and their proportions of the partial images that are statistically evaluated as important signals are obtained.

6. The method for collecting and analyzing oral and nasal breathing data according to claim 1, characterized in that: Network status information includes time domain upload bandwidth and time domain delay duration; By dividing the time domain into different sizes and marking the time domain as i, and dividing n time domains, the time domain upload bandwidth B (T i ); After dividing the time domain, use the ping command or other network measurement tools to measure the delay and obtain the time domain delay length D (T i ); The importance ratio Ze of partial images in the image is obtained by the ratio of the total number of partial images in the image to the important images.

7. The method for collecting and analyzing oral and nasal breathing data according to claim 6, characterized in that: The fuzzy Bayesian neural network is used to determine the transmission time, and the time domain upload bandwidth, time domain delay time, and the important proportion of the partial image in the image are defined as the input nodes in the input layer, and the transmission time is defined as the output layer; Formulate a fuzzy Bayesian neural network to describe the impact of time domain upload bandwidth, time domain delay duration, and the importance ratio of some images in the image on the transmission duration; According to the fuzzy Bayesian neural network, the transmission duration is determined by the time domain upload bandwidth, the time domain delay duration and the importance ratio of some images in the image.

8. The method for collecting and analyzing oral and nasal breathing data according to claim 7, characterized in that: After the transmission time is determined, it is compared with the preset transmission time. If it is less than the preset transmission time, the important images in the image will be transmitted to the PC. Otherwise, an alarm will be issued and the number of resources required for one transmission will be adjusted.

9. An oronasal breathing data collection and analysis system, used to implement the oronasal breathing data collection and analysis method according to any one of claims 1 to 8, characterized in that: It includes an acquisition module, a processing module, an image analysis module, a resource allocation module and an output terminal; The acquisition module is used to acquire images, divide them into multiple partial images according to the image edge information, acquire partial images, obtain the mouth and nose breathing data variation rate, breathing frequency change intensity and texture complexity through the image edge detection algorithm, and send them to the processing module, and collect network status information through network monitoring and send it to the resource allocation module; The processing module is used to obtain the variability rate of some oral and nasal breathing data, the intensity of respiratory frequency change and texture complexity, establish a data analysis model, obtain the image evaluation coefficient, and send it to the image analysis module; The image analysis module is used to obtain the image evaluation coefficient and compare it with the image threshold to obtain the comparison result. According to each comparison result and the number of comparison results, the number and proportion of important parts of the image are counted and sent to the resource allocation module. The resource allocation module is used to obtain network status information and the important proportion of some images in the image, obtain the transmission time through the fuzzy Bayesian neural network, and compare it with the preset transmission time. According to the comparison result, the transmission image is readjusted and the adjustment result is sent to the output end; The output end is used to receive the adjustment result and report it.

Images