F-V curve processing method and device based on wavelet analysis, electronic equipment and computer storage medium

Abstract

The invention provides an F-V curve processing method and device based on wavelet analysis, electronic equipment and a computer storage medium. The F-V curve processing method comprises the following steps: S1, wavelet transformation: selecting a wavelet basis function, and performing wavelet decomposition on an F-V curve to obtain a plurality of wavelet coefficients; s2, feature extraction: extracting features of related signals from the wavelet coefficients; the features comprise wavelet energy features reflecting an abnormal state of the F-V curve, wavelet statistical features reflecting energy distribution and fluctuation conditions of the F-V curve, and wavelet entropy features reflecting nonlinear changes of the F-V curve; s3, feature selection: adopting a principal component analysis method to screen out wavelet coefficients capable of analyzing the F-V curve and having significant features; and S4, F-V curve reconstruction: carrying out wavelet reconstruction by using the screened wavelet coefficients to obtain a reconstructed F-V curve. According to the processing method, the F-V curve is decomposed and reconstructed on the basis of the wavelet analysis principle, noise can be accurately removed, and effective characteristics of the F-V curve can be reserved.

Description

Technical Field

[0001] The present invention relates to the technical field of physiological signal analysis, and more specifically, to a FV curve processing method, device, electronic device and computer storage medium based on wavelet analysis. Background Art

[0002] Pulmonary function testing is an important diagnostic tool for evaluating respiratory diseases. It measures lung capacity and airflow rate to diagnose and monitor various respiratory diseases such as asthma and chronic obstructive pulmonary disease (COPD).

[0003] The flow volume curve (FV curve for short) is an important tool for evaluating pulmonary ventilation function. The FV curve mainly reflects the changes in flow and volume, and is suitable for evaluating airway obstruction and lung volume problems, and can provide an important reference for clinical diagnosis and treatment. However, pulmonary ventilation function data based on spirometer instruments usually have the following characteristics: periodicity; individual differences; detection noise; dynamic variability.

[0004] The FV curve is easily disturbed by tiny signal noise, which increases the risk of misdiagnosis and missed diagnosis. Signal processing methods are generally used to optimize the FV curve. However, traditional signal processing methods are prone to introduce additional errors in the decomposition process, resulting in inaccurate detection results. These traditional signal processing methods lack sufficient robustness when processing complex signals and are easily affected by noise, resulting in unstable decomposition results. When faced with actual biomedical signals, modal mixing problems are prone to occur, making it difficult to clearly distinguish different modal functions.

[0005] Therefore, it is of great significance to provide a processing method that can accurately remove noise and retain the effective features of the FV curve. Summary of the invention

[0006] The object of the present invention is to provide a processing method capable of accurately removing noise and retaining effective features of FV curve.

[0007] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0008] In a first aspect, the present invention provides a FV curve processing method based on wavelet analysis, comprising the following steps:

[0009] S1. Wavelet transform:

[0010] Select wavelet basis function, perform wavelet decomposition on FV curve, and obtain multiple wavelet coefficients;

[0011] S2. Feature extraction:

[0012] Extracting the features of the relevant signal from the wavelet coefficients; the features include the wavelet energy features reflecting the abnormal state of the FV curve, the wavelet statistical features reflecting the energy distribution and fluctuation of the FV curve, and the wavelet entropy features reflecting the nonlinear changes of the FV curve;

[0013] S3. Feature selection:

[0014] The principal component analysis method was used to screen out the wavelet coefficients with significant characteristics for FV curve analysis;

[0015] S4.FV curve reconstruction:

[0016] The selected wavelet coefficients are used for wavelet reconstruction to obtain the reconstructed FV curve.

[0017] As one implementation manner, the wavelet basis function is a Daubechies wavelet.

[0018] As one implementation, the wavelet energy feature is obtained by performing a square sum operation on the wavelet coefficients. By performing a square sum operation on the wavelet coefficients, the energy distribution at each scale is obtained, and then the energy features of the signal at different frequencies are analyzed. The wavelet energy feature can effectively reflect the abnormal state of lung function.

[0019] As one implementation method, the wavelet entropy feature is the entropy value of the wavelet coefficient. The wavelet entropy feature is used to measure the complexity and uncertainty of the signal. The wavelet entropy feature can capture the nonlinear changes in the FV curve, which is helpful for subsequent classification or prediction tasks.

[0020] As one implementation method, the wavelet statistical feature is to calculate the statistics of wavelet coefficients of each scale; the statistics include but are not limited to mean, standard deviation, maximum value, and minimum value. The wavelet statistical feature reflects the energy distribution and fluctuation of the signal in different frequency bands.

[0021] Optionally, before wavelet transform, signal preprocessing is also included: the preprocessing includes processing the FV curve data by threshold setting and interquartile range (IQR) method, and then processing by signal segmentation and cubic spline interpolation method. Threshold setting and interquartile range method are used to eliminate noise, and signal segmentation and cubic spline interpolation method are used to optimize the data continuity and smoothness of the pulmonary ventilation function test curve.

[0022] In a second aspect, another object of the present invention is to provide an FV curve recognition algorithm model, wherein the recognition algorithm model includes the FV curve processing method based on wavelet analysis.

[0023] The FV curve processing method based on wavelet analysis can be used to process the training set, verification set and test set of the recognition algorithm model.

[0024] Using the FV curve processing method to optimize the FV curve before the algorithm model is identified can improve the recognition efficiency and accuracy of the algorithm model.

[0025] In a third aspect, another object of the present invention is to provide a device for a FV curve processing method based on wavelet analysis, comprising:

[0026] Input unit, used to obtain FV curve;

[0027] A processing unit, used for processing the FV curve, wherein the processing includes wavelet transformation, feature extraction, and feature selection;

[0028] The output unit is used to perform wavelet reconstruction according to the screened wavelet coefficients and output the reconstructed FV curve.

[0029] In a fourth aspect, the present invention further provides an electronic device. The electronic device comprises at least one memory and at least one processor; the at least one memory is coupled to the at least one processor, the at least one memory is used to store a computer program, the at least one processor is used to call the computer program, the computer program comprises instructions, and when the instructions are executed by the at least one processor, the electronic device executes the FV curve processing method based on wavelet analysis described in the first aspect.

[0030] In a fifth aspect, the present invention further provides a computer storage medium, wherein the computer storage medium includes computer instructions, and when the computer instructions are executed on an electronic device, the electronic device executes the FV curve processing method based on wavelet analysis described in the first aspect.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] The FV curve processing method based on wavelet analysis of the present invention decomposes and reconstructs the FV curve based on the wavelet analysis principle and in combination with the characteristics of the FV curve, and is a processing method that can accurately remove noise and retain the effective characteristics of the FV curve. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a flow chart of the FV curve processing method based on wavelet analysis.

[0034] Figure 2 This is a flow chart of the FV curve processing method (including preprocessing) based on wavelet analysis.

[0035] Figure 3 It is the functional module block diagram of the FV curve processing method based on wavelet analysis.

[0036] Figure 4 This is a specific structural block diagram of an electronic device. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0038] like Figure 1 and 2 As shown, a FV curve processing method based on wavelet analysis includes the following steps:

[0039] S0. Preprocessing:

[0040] In order to improve the accuracy and credibility of the extracted FV curve data, a comprehensive data preprocessing method was adopted.

[0041] First, the different colors in the FV curve are extracted by adjusting the image color. Then, the curves of different colors are accurately identified and labeled using the visual review method, successfully identifying abnormal data points caused by noise, which appear as irregular spikes and mutations. Based on the professional knowledge of pulmonary function testing, a threshold setting method is used to set a specific threshold T and evaluate each data point xi to identify and eliminate noise data points, as shown in the following formula:

[0042] ∣x i -μ∣>T·σ (1)

[0043] Among them, xi represents each data point; μ represents the mean value of the data set; σ represents the standard deviation of the data set.

[0044] Secondly, the interquartile range (IQR) is used as a statistical tool to define the normal fluctuation range of the data for lung function data without abnormal data points, effectively excluding data points that deviate from this range. IQR provides a robust outlier detection mechanism, significantly reducing the impact of extreme values ​​on the integrity of data analysis. The specific operation steps include: first sorting the FV curve data, then calculating the first quartile (Q1) and the third quartile (Q3) of the data set, and determining the IQR based on this, as shown in formula (2). This value reveals the data distribution between Q1 and Q3, that is, the middle 50% of the data range. Using IQR, those abnormal data points (xi) that may interfere with the overall analysis of the FV curve are effectively identified and excluded, as shown in formula (3).

[0045] IQR=Q3-Q1 (2)

[0046] Outlier definition:

[0047]

[0048] Among them, Q1 is the first quartile of the data, indicating that 25% of all the data are less than or equal to this value; Q3 is the third quartile of the data, indicating that 75% of all the data are less than or equal to this value.

[0049] The interquartile range method can accurately eliminate abnormal data points caused by test errors, patients' atypical breathing patterns, or equipment measurement deviations, thereby ensuring the high accuracy and scientificity of FV curve analysis.

[0050] Again, signal segmentation and cubic spline interpolation methods were used to improve the reliability and accuracy of the data. The implementation formula is as follows.

[0051] S i (x) = a i +b i (xx i )+c i (xx i ) 2 +d i (xx i ) 3 (4)

[0052] Where S i (x) is the spline function of the interval containing x, x i is a node or control point, and a i 、b i 、c i、 d i are the coefficients of the spline function.

[0053] By using signal segmentation technology, the data is accurately distinguished according to the physiological stage characteristics revealed by the FV curve, and each stage of the breathing process is analyzed in a targeted manner, effectively identifying and correcting the data deviation caused by the patient's irregular breathing pattern. Furthermore, by applying cubic spline interpolation technology, a smooth transition is achieved at the missing data point, outliers are corrected, and the continuity and physiological relevance of the data are significantly improved, making the processed FV curve smoother while maintaining its physiological significance.

[0054] S1. Wavelet transform:

[0055] Select wavelet basis function, perform wavelet decomposition on FV curve, and obtain multiple wavelet coefficients;

[0056] This embodiment selects Daubechies wavelet as the wavelet basis function.

[0057] Wavelet transform replaces the infinite-length trigonometric function basis with a finite-length wavelet basis that decays. Wavelet transform mainly achieves multi-scale refinement through scaling and translation, highlights the details of the problem to be processed, and effectively extracts local information. The transformation formula is as follows:

[0058]

[0059] Among them, parameter a locates the frequency, and parameter b locates the time. Since parameters a and b are both continuously changing, they need to be discretized using multi-resolution analysis (MRA):

[0060]

[0061] The meaning of each symbol in the above formula is:

[0062] Φ cwt (a, b) represents the coefficients of the continuous wavelet transform, which is the coefficient of the signal f(t) and the wavelet function ψ a,b (t) is the inner product of

[0063] <f,ψ a,b > represents the signal f(t) and the wavelet function ψ a,b (t), which is the convolution of the two.

[0064] a represents the scale parameter, which controls the expansion and contraction of the wavelet function and affects the frequency resolution. Smaller a values ​​correspond to higher frequencies, and larger a values ​​correspond to lower frequencies.

[0065] b represents the translation parameter, which controls the translation of the wavelet function and affects the time resolution.

[0066] ψ(t) represents the mother wavelet function and is the basis for constructing the wavelet basis.

[0067] ψ* (t) represents the complex conjugate of the mother wavelet function.

[0068] ψ a,b (t) represents the wavelet function at scale a and position b, which is obtained by scaling and translating the mother wavelet function.

[0069] m represents the discretized scale parameter, which corresponds to log2(a) in the continuous wavelet transform.

[0070] n represents the discretized translation parameter, corresponding to b in continuous wavelet transform.

[0071] ψ m,n (k) represents the wavelet function at discrete scale m and position n.

[0072] It represents the space of square integrable functions, that is, the integral of the square of a function over the real number line is finite.

[0073] DWT(m,n) represents the coefficients of discrete wavelet transform, which is the coefficient of the signal f(k) and the discrete wavelet function ψ m,n (k) is the inner product of

[0074] f(k) represents the discretized signal.

[0075] This implementation uses the Daubechies wavelet, and its adjustment function can be written as follows:

[0076]

[0077] At the same time, the wavelet function can also be used as a linear combination of adjustment functions:

[0078]

[0079] The meaning of each symbol in the above formula is:

[0080] φ(x) represents the adjustment function, which is one of the basic functions for constructing the wavelet basis and is used to represent the low-frequency part of the signal.

[0081] represents the second-order recurrence relation of the adjustment function, where a k is the coefficient of the adjustment function, and N is the support length of the adjustment function.

[0082] a k represents the coefficients of the adjustment function, which are real numbers and form a finite length sequence, namely a0,...,a N-1 .

[0083] ψ(x) represents the wavelet function, which is another basic function for constructing the wavelet basis and is used to represent the high-frequency part of the signal.

[0084] represents the second-order recurrence relation of the wavelet function, where b k are the coefficients of the wavelet function, and M is the support length of the wavelet function.

[0085] b k The coefficients of the wavelet function are also real numbers and form a finite length sequence called wavelet coefficients, i.e. b0,...,b N-1 .

[0086] N and M represent the support lengths of the adjustment function and the wavelet function, respectively. For the Daubechies wavelet, N and M are usually equal.

[0087] x represents the independent variable, which can be time, space or other continuous variables.

[0088] S2. Feature extraction:

[0089] Extracting the features of the relevant signal from the wavelet coefficients; the features include the wavelet energy features reflecting the abnormal state of the FV curve, the wavelet statistical features reflecting the energy distribution and fluctuation of the FV curve, and the wavelet entropy features reflecting the nonlinear changes of the FV curve;

[0090] S3. Feature selection:

[0091] The principal component analysis method is used to screen out the wavelet coefficients that can analyze the most significant characteristics of the FV curve;

[0092] The steps of principal component analysis method are as follows:

[0093] 1) Feature collection:

[0094] The features are collected from the extracted FV curve data and constructed into a matrix X, where each row represents a sample and each column represents a feature.

[0095] Assume that the dimension of matrix X is m×n, where m is the number of samples and n is the number of features.

[0096] 2) Data Center:

[0097] Center each feature in the matrix X, that is, subtract the mean of each feature to make the mean of the data zero: X centered =X-mean(X).

[0098] 3) Calculate the covariance matrix:

[0099] Calculate the covariance matrix Σ of the centered data, whose dimension is n×n:

[0100]

[0101] 4) Eigenvalue decomposition:

[0102] Perform eigenvalue decomposition on the covariance matrix Σ to obtain eigenvalues ​​and eigenvectors. The eigenvalues ​​represent the variance of each principal component, and the eigenvectors represent the direction of the principal component.

[0103] 5) Select the main components:

[0104] Sort the eigenvectors by their eigenvalues ​​and select the principal components corresponding to the first k eigenvectors with the largest eigenvalues. These principal components correspond to the directions with the largest and most significant variance in the data.

[0105] Typically, by selecting the first few principal components whose cumulative variance reaches a certain threshold (such as 90%), it is ensured that the main information is retained while reducing the data dimension.

[0106] 6) Projection to principal component space:

[0107] Project the original data onto the selected k principal component directions to form a reduced-dimensional data set: X PCA =X centered W, where W is the matrix containing the first k eigenvectors.

[0108] 7) Interpretation of results:

[0109] Projected data X PCA It is a set of significant features of the FV curve. Further analysis is performed based on these principal component features to screen out the wavelet coefficients.

[0110] S4.FV curve reconstruction:

[0111] The wavelet coefficients with significant features are selected for wavelet reconstruction to obtain the reconstructed FV curve.

[0112] This embodiment also provides an FV curve recognition algorithm model, and the recognition algorithm model includes the above-mentioned FV curve processing method based on wavelet analysis.

[0113] The FV curve processing method based on wavelet analysis can be used to process the training set, validation set and test set of the recognition algorithm model to improve the recognition efficiency and accuracy of the recognition algorithm model.

[0114] This embodiment also provides a device for the FV curve processing method based on wavelet analysis.

[0115] The following describes a device for executing the above-mentioned FV curve processing method based on wavelet analysis.

[0116] Figure 2 The specific structural block diagram of the device provided in this embodiment is shown, and the device of the FV curve processing method based on wavelet analysis includes:

[0117] An input unit 10, used for obtaining a FV curve;

[0118] A processing unit 20, used for processing the FV curve, wherein the processing includes wavelet transformation, feature extraction, and feature selection;

[0119] The output unit 30 is used to perform wavelet reconstruction according to the screened wavelet coefficients and output the reconstructed FV curve.

[0120] The device provided in this embodiment and the FV curve processing method provided in this application belong to the same concept. The specific implementation process is detailed in the full text of the specification and will not be repeated here.

[0121] This embodiment also provides an electronic device. The electronic device that can execute the above FV curve processing method is described below.

[0122] Figure 3 The specific structural block diagram of the electronic device 100 provided by the FV curve processing method of this embodiment is shown, and the electronic device 100 includes: at least one memory 101 and at least one processor 102; the at least one memory 101 is coupled to the at least one processor 102, the at least one memory 101 is used to store a computer program, and the at least one processor 102 is used to call the computer program, and the computer program includes instructions. When the instructions are executed by the at least one processor, the electronic device executes the above-mentioned FV curve processing method.

[0123] This embodiment also provides a computer storage medium. The computer storage medium containing the above FV curve processing method is described below.

[0124] A computer storage medium includes computer instructions. When the computer instructions are executed on an electronic device, the electronic device executes the above-mentioned FV curve processing method.

[0125] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A FV curve processing method based on wavelet analysis, characterized in that: The following steps are involved: S1. Wavelet transform: Select wavelet basis function, perform wavelet decomposition on FV curve, and obtain multiple wavelet coefficients; S2. Feature extraction: Extracting the features of the relevant signal from the wavelet coefficients; the features include the wavelet energy features reflecting the abnormal state of the FV curve, the wavelet statistical features reflecting the energy distribution and fluctuation of the FV curve, and the wavelet entropy features reflecting the nonlinear changes of the FV curve; S3. Feature selection: The principal component analysis method was used to screen out the wavelet coefficients with significant characteristics for FV curve analysis; S4.FV curve reconstruction: The selected wavelet coefficients are used for wavelet reconstruction to obtain the reconstructed FV curve.

2. The FV curve processing method based on wavelet analysis according to claim 1 is characterized in that: The wavelet basis function is the Daubechies wavelet.

3. The FV curve processing method based on wavelet analysis according to claim 1 is characterized in that: The wavelet energy feature is obtained by performing a square sum operation on the wavelet coefficients.

4. The FV curve processing method based on wavelet analysis according to claim 1 is characterized in that: The wavelet entropy feature is the entropy value of the wavelet coefficients.

5. The FV curve processing method based on wavelet analysis according to claim 1 is characterized in that: The wavelet statistical features are statistics for calculating wavelet coefficients of each scale; the statistics include mean, standard deviation, maximum value, and minimum value.

6. The abnormality identification method for pulmonary ventilation function test according to claim 1, characterized in that: Before wavelet transformation, signal preprocessing is also included: the preprocessing includes processing the FV curve data using threshold setting and interquartile range method, and then processing using signal segmentation and cubic spline interpolation method.

7. An FV curve recognition algorithm model, characterized in that: The identification algorithm model includes the FV curve processing method based on wavelet analysis as described in claims 1 to 6.

8. A device for processing FV curves based on wavelet analysis, characterized in that: include: Input unit, used to obtain FV curve; A processing unit, used for processing the FV curve, wherein the processing includes wavelet transformation, feature extraction, and feature selection; The output unit is used to perform wavelet reconstruction according to the screened wavelet coefficients and output the reconstructed FV curve.

9. An electronic device, characterized in that: comprising at least one memory and at least one processor; The at least one memory is coupled to the at least one processor, the at least one memory is used to store a computer program, the at least one processor is used to call the computer program, the computer program includes instructions, when the instructions are executed by the at least one processor, the electronic device executes the FV curve processing method based on wavelet analysis as described in any one of claims 1 to 6.

10. A computer storage medium, characterized in that: The method comprises computer instructions, which, when executed on an electronic device, enable the electronic device to execute the FV curve processing method based on wavelet analysis as claimed in any one of claims 1 to 6.

Images