A swallowing effect prediction method and system based on swallowing training instrument

Abstract

The present invention provides a swallowing effect prediction method and system based on a swallowing trainer. The method includes acquiring swallowing training data transmitted by the swallowing trainer, preprocessing the swallowing training data to obtain preprocessed data; performing Fourier transform and data amplification on the preprocessed data to obtain amplified data; performing high-frequency and low-frequency decomposition and denoising processing on the amplified data to obtain denoised data; using an improved particle swarm optimization algorithm and a density clustering algorithm to identify abnormal data on the denoised data to obtain target data; acquiring template training data, inputting the template training data into a preset prediction model for training, inputting the target data into the trained preset prediction model for effect prediction, and outputting a swallowing effect prediction result. The present invention can retain the characteristics of the original data and effectively eliminate low-quality data, thereby improving the final prediction accuracy of the model.

Description

Technical Field

[0001] The present invention belongs to the technical field of effect prediction, and in particular relates to a swallowing effect prediction method and system based on a swallowing training device. Background Art

[0002] A swallowing trainer is a medical device specifically designed to help patients improve and restore their swallowing function. It typically consists of a piezoresistive pressure sensor, a laryngeal muscle trainer, and a controller. The piezoresistive pressure sensor is placed in the patient's throat to detect the strength of the laryngeal muscles and generate an electrical signal representing this strength. The laryngeal muscle trainer is placed in the patient's throat during swallowing training to provide targeted training for the laryngeal muscles. The controller is responsible for receiving the electrical signals and controlling the laryngeal muscle trainer's operation after receiving swallowing training instructions. The controller also uploads the electrical signals to a host computer for display and analysis by the treating physician.

[0003] For the general identification of swallowing training effects, it is usually judged based on the doctor's experience, which has great limitations. In the existing technology, there is also the use of prediction models to assist in the prediction of swallowing effects. However, in the actual prediction process, the data transmitted by the sensor generally includes the patient's current swallowing function data, the data of the laryngeal muscles during swallowing training, the electromyographic signal data, etc. For the original data, due to the influence of the environment, there will be a lot of noise in the collected signal data, which will affect the accuracy of the swallowing effect prediction. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention provides a swallowing effect prediction method and system based on a swallowing training instrument, which are used to solve the technical problems in the prior art.

[0005] In one aspect, the present invention provides the following technical solution: a swallowing effect prediction method based on a swallowing training device, comprising:

[0006] Acquiring swallowing training data transmitted by a swallowing training device, and preprocessing the swallowing training data to obtain preprocessed data;

[0007] Performing Fourier transform and data amplification on the preprocessed data to obtain amplified data;

[0008] performing high-frequency and low-frequency decomposition and denoising processing on the amplified data to obtain denoised data;

[0009] An improved particle swarm optimization algorithm and a density clustering algorithm are used to identify abnormal data on the denoised data to obtain target data;

[0010] Acquire template training data, input the template training data into a preset prediction model for training, input the target data into the trained preset prediction model for effect prediction, and output a swallowing effect prediction result.

[0011] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention first obtains the swallowing training data transmitted by the swallowing training device, and pre-processes the swallowing training data to obtain pre-processed data; then performs Fourier transform and data amplification on the pre-processed data to obtain amplified data; then performs high-frequency and low-frequency decomposition and denoising processing on the amplified data to obtain denoised data; then uses the improved particle swarm optimization algorithm and density clustering algorithm to identify abnormal data on the denoised data to obtain target data; finally, obtains template training data, inputs the template training data into the preset prediction model for training, and inputs the target data into the trained preset prediction model for effect prediction to output the swallowing effect prediction result. The present invention avoids the problem of too little sample data affecting the model prediction accuracy by amplifying, denoising and identifying abnormalities in the data. It can effectively remove the noise in the data while retaining the characteristics of the original data. Finally, it can effectively eliminate low-quality data, thereby improving the final prediction accuracy of the model.

[0012] Preferably, the step of performing Fourier transform and data amplification on the preprocessed data to obtain amplified data includes:

[0013] Perform fast Fourier transform on the preprocessed data, and add Zero points to obtain the transformed data :

[0014] ;

[0015] Where, is the frequency resolution, is the sampling rate, is the number of sampling points used for fast Fourier transform;

[0016] The transformation data is converted to obtain the real number part and the imaginary part :

[0017] ;

[0018] Where, is the imaginary number symbol;

[0019] Add random perturbations to the real part and the imaginary part , to obtain the first perturbation part With the second perturbation part :

[0020] ; ; ;

[0021] Where, 、 Respectively represent the lower and upper frequency limits of the transformed data, Indicates the number of groups of transformed data;

[0022] The first perturbation part With the second perturbation part performing combination to obtain combined data, performing fast inverse Fourier transform on the combined data to obtain restored data, and adding first several data of the restored data to the preprocessed data to obtain intermediate amplified data;

[0023] Based on the intermediate amplified data, the real number part With the imaginary part Confirm amplification data.

[0024] Preferably, the intermediate amplified data, the real number part With the imaginary part The steps to determine amplification data include:

[0025] Based on the real part With the imaginary part Calculate data phase :

[0026] ;

[0027] Replace the data phase of several random sets of transformed data with Random values ​​between to get the adjusted phase ;

[0028] Based on the adjusted phase Determine adjustment data :

[0029] ;

[0030] Where, is the fast inverse Fourier transform;

[0031] The first several data of the adjusted data are added to the intermediate amplified data to obtain the amplified data.

[0032] Preferably, the step of performing high-frequency and low-frequency decomposition and denoising processing on the amplified data to obtain denoised data includes:

[0033] The amplification data is decomposed into several Components and residual components;

[0034] Set the contribution function of the amplification data:

[0035] ;

[0036] Where, Represents a function variable, To amplify the data, To solve the interval, For time, is the function value;

[0037] Determine the distribution characteristics of the contribution function on the coordinate axis and determine the solution interval in the function value set as the final interval ;

[0038] Based on the final interval Calculate each of the Correlation of components :

[0039] ;

[0040] Where, Indicates the indivual Quantity;

[0041] Denoised data is determined based on the degree of association.

[0042] Preferably, the step of determining the denoised data based on the correlation degree includes:

[0043] Set the first correlation threshold and the second correlation threshold, and select The components are stored in a reconstructed component set, wherein the first correlation threshold is greater than the second correlation threshold;

[0044] The correlation degree is not less than the second correlation threshold and less than the first correlation threshold. The components are stored in the denoising component set;

[0045] Determine the mean square error , and the denoising component set is calculated based on the mean square error The components are denoised and smoothed to obtain the processed Quantity;

[0046] ;

[0047] Where, are the polynomial coefficients, is a power of the polynomial, In the denoising component set The number of servings, is the first indivual Quantity;

[0048] The processing The components, the reconstructed component set and the residual component are combined and the signal is reconstructed to obtain denoised data.

[0049] Preferably, the step of using an improved particle swarm optimization algorithm and a density clustering algorithm to identify abnormal data on the denoised data to obtain target data includes:

[0050] The neighborhood radius in the density clustering algorithm and the minimum number of points in the neighborhood As two optimization targets of the improved particle swarm optimization algorithm, the neighborhood radius and the minimum number of points in the neighborhood Corresponding to the velocity and position of the particle respectively;

[0051] Determining the initial position and initial velocity of each particle in the improved particle swarm optimization algorithm, and performing initial clustering on the denoised data based on the initial position and the initial velocity using a density clustering algorithm;

[0052] Calculate the fitness of each particle in the clustering process :

[0053] ;

[0054] Where, Indicates the denoised data The average distance between a sample and other samples in its cluster, Indicates the denoised data The average distance between a sample and other samples in the cluster;

[0055] Iteratively update the inertia weight of each particle :

[0056] ;

[0057] Where, Indicates the current iteration number, represents the total number of iterations, 、 Represent the starting weight and final weight respectively;

[0058] Iteratively update the first learning factor and the second learning factor :

[0059] ; ;

[0060] Where, Respectively represent the starting value and final value of the first learning factor, Respectively represent the starting value and final value of the second learning factor;

[0061] According to the fitness, the updated inertia weight , the updated first learning factor With the updated second learning factor Determine target data.

[0062] Preferably, the inertia weight after the fitness is updated , the updated first learning factor With the updated second learning factor The steps to determine target data include:

[0063] According to the fitness, inertia weight , the first learning factor and the second learning factor Iteratively update the velocity and position of each particle:

[0064] ;

[0065] ;

[0066] Where, Respectively represent 、 The velocity of the particle after iterations, Respectively represent 、 The position of the particle after iterations, Represent the first random number and the second random number respectively, They represent the individual optimal solution and the group optimal solution respectively;

[0067] Repeat the iterative process until the iteration stop condition is met and output the updated neighborhood radius obtained by the final iteration and the minimum number of points included in the updated neighborhood;

[0068] Based on the updated neighborhood radius and the minimum number of points included in the updated neighborhood, the denoised data is density clustered using a density clustering algorithm to obtain target data.

[0069] In a second aspect, the present invention provides the following technical solution: a swallowing effect prediction system based on a swallowing training device, the system comprising:

[0070] a processing module, configured to obtain swallowing training data transmitted by the swallowing training device and preprocess the swallowing training data to obtain preprocessed data;

[0071] an amplification module, configured to perform Fourier transform and data amplification on the preprocessed data to obtain amplified data;

[0072] a denoising module, configured to perform high-frequency and low-frequency decomposition and denoising processing on the amplified data to obtain denoised data;

[0073] An identification module is used to identify abnormal data on the denoised data using an improved particle swarm optimization algorithm and a density clustering algorithm to obtain target data;

[0074] The prediction module is used to obtain template training data, input the template training data into a preset prediction model for training, input the target data into the trained preset prediction model for effect prediction, and output a swallowing effect prediction result.

[0075] In a third aspect, the present invention provides the following technical solution: a computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor; when the processor executes the computer program, the method for predicting swallowing effects based on a swallowing training device as described above is implemented.

[0076] In a fourth aspect, the present invention provides the following technical solution: a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the swallowing effect prediction method based on a swallowing training device as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0078] Figure 1 This is a flow chart of a swallowing effect prediction method based on a swallowing training device provided in Example 1 of the present invention;

[0079] Figure 2 This is a structural block diagram of a swallowing effect prediction system based on a swallowing training device provided in the second embodiment of the present invention;

[0080] Figure 3A schematic diagram of the hardware structure of a computer provided in another embodiment of the present invention.

[0081] The embodiments of the present invention will be further described below with reference to the accompanying drawings. DETAILED DESCRIPTION

[0082] The following describes embodiments of the present invention in detail, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the embodiments of the present invention, and should not be construed as limiting the present invention.

[0083] Example 1

[0084] In the first embodiment of the present invention, Figure 1 As shown, a swallowing effect prediction method based on a swallowing training device includes:

[0085] S1. Obtaining swallowing training data transmitted by a swallowing training device, and preprocessing the swallowing training data to obtain preprocessed data;

[0086] Among them, the swallowing training data here includes current swallowing function data, laryngeal muscle data during swallowing training, electromyographic signal data, etc., and the preprocessing process here includes format conversion, smoothing processing, etc., and the above preprocessing process is a common step in the existing technology, and the swallowing training device here is a commonly used instrument in the existing technology, so it will not be repeated here.

[0087] S2. Performing Fourier transform and data amplification on the preprocessed data to obtain amplified data;

[0088] Wherein, the step S2 includes:

[0089] S21, performing fast Fourier transform on the pre-processed data, and adding Zero points to obtain the transformed data :

[0090] ;

[0091] Where, is the frequency resolution, is the sampling rate, is the number of sampling points used for fast Fourier transform;

[0092] Specifically, the discrete time domain data of each channel can be converted into discrete frequency domain data through a fast Fourier transform operation.

[0093] S22, converting the transformed data to obtain the real number part and the imaginary part :

[0094] ;

[0095] Where, Is the imaginary number symbol.

[0096] S23, adding random perturbations to the real part and the imaginary part , to obtain the first perturbation part With the second perturbation part :

[0097] ; ; ;

[0098] Where, 、 Respectively represent the lower and upper frequency limits of the transformed data, Indicates the number of groups of transformed data;

[0099] Specifically, the purpose of adding random perturbations here is to mask the noise. By adding several zero points in the process of fast Fourier transform, the 、 Become original At the same time, due to the symmetry of the node center in the frequency domain data, the real and imaginary parts of The data can be equivalent to the first data.

[0100] S24, the first disturbance part With the second perturbation part performing combination to obtain combined data, performing fast inverse Fourier transform on the combined data to obtain restored data, and adding first several data of the restored data to the preprocessed data to obtain intermediate amplified data;

[0101] S25, based on the intermediate amplified data, the real number part With the imaginary part Determine amplification data;

[0102] Wherein, the step S25 includes:

[0103] S251, based on the real number part With the imaginary part Calculate data phase :

[0104] .

[0105] S252, replace the data phase of several random groups of transformed data with Random values ​​between to get the adjusted phase ;

[0106] Specifically, for each set of data, there is a data phase, so several sets of data are randomly selected and selected. The original phase can be replaced by a random value between , and the adjusted phase can be obtained, while the original phase is used for the one that is not replaced.

[0107] S253, based on the phase adjustment Determine adjustment data :

[0108] ;

[0109] Where, is the fast inverse Fourier transform.

[0110] S254: Add the first several data of the adjusted data to the intermediate amplified data to obtain amplified data.

[0111] S3, performing high-frequency and low-frequency decomposition and denoising processing on the amplified data to obtain denoised data;

[0112] Wherein, the step S3 includes:

[0113] S31, using a preset algorithm to decompose the amplification data into several Components and residual components;

[0114] Specifically, the preset algorithm here is the adaptive noise complete set empirical mode decomposition algorithm.

[0115] S32, setting the contribution function of the amplification data:

[0116] ;

[0117] Where, Represents a function variable, To amplify the data, To solve the interval, For time, is the function value.

[0118] S33, determining the distribution characteristics of the contribution function on the coordinate axis and determining the solution interval in the function value concentration as the final interval ;

[0119] Specifically, the contribution function can reflect the contribution of data at different time scales. At the same time, according to the contribution function, the maximum value of the contribution function is distributed near 0 according to the random characteristics of the noise, and the remaining function values ​​​​except 0 are a slowly decaying process and will not decay to a very small value. Therefore, for its function value, it is concentrated near 0. Therefore, this application sets the final interval to .

[0120] S34, based on the final interval Calculate each of the Correlation of components :

[0121] ;

[0122] Where, Indicates the indivual Quantity.

[0123] S35, determining denoised data based on the correlation degree;

[0124] Wherein, the step S35 includes:

[0125] S351, set a first correlation threshold and a second correlation threshold, and set the correlation not less than the first correlation threshold The components are stored in a reconstructed component set, wherein the first correlation threshold is greater than the second correlation threshold;

[0126] Specifically, the first correlation threshold here is 0.001, and the second correlation threshold is 0.0005. By setting the first correlation threshold, components related to noise and components related to signal features can be classified, and the purpose of setting the second correlation threshold is to eliminate components with a large amount of noise.

[0127] S352, the correlation degree is not less than the second correlation threshold and less than the first correlation threshold The components are stored in the denoising component set.

[0128] S353, determine the mean square error , and the denoising component set is calculated based on the mean square error The components are denoised and smoothed to obtain the processed Quantity;

[0129] ;

[0130] Where, are the polynomial coefficients, is a power of the polynomial, In the denoising component set The number of servings, is the first indivual Quantity;

[0131] Specifically, for the denoising component set, the sequence number of the middle component is 0, that is, the above denoising process is based on The square difference of the component centered is the smallest, that is, the denoising process is completed by the polynomial least squares method.

[0132] S354, the processing The components, the reconstructed component set and the residual component are combined and the signal is reconstructed to obtain denoised data.

[0133] S4. Using an improved particle swarm optimization algorithm and a density clustering algorithm to identify abnormal data on the denoised data to obtain target data;

[0134] Wherein, the step S4 includes:

[0135] S41. The neighborhood radius in the density clustering algorithm and the minimum number of points in the neighborhood As two optimization targets of the improved particle swarm optimization algorithm, the neighborhood radius and the minimum number of points in the neighborhood Corresponding to the velocity and position of the particle respectively;

[0136] Specifically, the density clustering algorithm here is a clustering algorithm commonly used in the prior art. The two most important parameters in the density clustering algorithm are replaced by the speed and position of the particles in the improved particle swarm optimization algorithm. That is, in the process of density clustering, the improved particle swarm optimization algorithm is used to iteratively update the parameters for optimization.

[0137] S42, determining the initial position and initial velocity of each particle in the improved particle swarm optimization algorithm, and performing initial clustering on the denoised data according to the initial position and the initial velocity using a density clustering algorithm.

[0138] S43. Calculate the fitness of each particle in the clustering process :

[0139] ;

[0140] Where, Indicates the denoised data The average distance between a sample and other samples in its cluster, Indicates the denoised data The average distance between a sample and other samples in the cluster;

[0141] Specifically, the purpose of calculating the fitness is to ensure that the position of the particle is optimized towards the maximum fitness value.

[0142] S44. Iteratively update the inertia weight of each particle :

[0143] ;

[0144] Where, Indicates the current iteration number, represents the total number of iterations, 、 Represent the starting weight and final weight respectively;

[0145] Specifically, the purpose of updating the weights is to have greater global optimization ability in the early stage of the algorithm operation and greater local convergence ability in the later stage.

[0146] S45. Iteratively update the first learning factor and the second learning factor :

[0147] ; ;

[0148] Where, Respectively represent the starting value and final value of the first learning factor, Respectively represent the starting value and final value of the second learning factor;

[0149] Specifically, the first learning factor and the second learning factor here are asymmetric learning factors. In the early stage, a larger first learning factor and a smaller second learning factor are used to make particles focus more on the individual optimal solution and reduce the influence of the cluster. This allows particles to cover the entire search space as much as possible. As the number of iterations increases, the first learning factor decreases linearly and the second learning factor increases linearly, thereby enhancing the particle's ability to converge to the global optimal point.

[0150] S46, according to the fitness, the updated inertia weight , the updated first learning factor With the updated second learning factor Determine target data;

[0151] Wherein, the step S46 includes:

[0152] S461, according to the fitness, inertia weight , the first learning factor and the second learning factor Iteratively update the velocity and position of each particle:

[0153] ;

[0154] ;

[0155] Where, Respectively represent 、 The velocity of the particle after iterations, Respectively represent 、 The position of the particle after iterations, Represent the first random number and the second random number respectively, They represent the individual optimal solution and the group optimal solution respectively;

[0156] The first random number and the second random number are specifically random numbers between 0 and 1.

[0157] S462, repeat the iterative process until the iteration stop condition is met and output the updated neighborhood radius obtained by the final iteration and the minimum number of included points in the updated neighborhood.

[0158] S463, performing density clustering on the denoised data based on the updated neighborhood radius and the minimum number of points in the updated neighborhood using a density clustering algorithm to obtain target data;

[0159] Specifically, in the process of density clustering, several clusters will be obtained, and there will be some outliers outside the clusters. These outliers are abnormal data. Only the data in the clusters are retained to obtain the target data.

[0160] S5. Acquire template training data, input the template training data into a preset prediction model for training, input the target data into the trained preset prediction model for effect prediction, and output a swallowing effect prediction result;

[0161] Specifically, the preset prediction model here is specifically a BP neural network model in the prior art. By inputting the target data into the trained model, the final swallowing effect prediction result can be output.

[0162] The swallowing effect prediction method based on the swallowing training instrument provided in the first embodiment of the present invention first obtains the swallowing training data transmitted by the swallowing training instrument, and pre-processes the swallowing training data to obtain pre-processed data; then performs Fourier transform and data amplification on the pre-processed data to obtain amplified data; then performs high-frequency and low-frequency decomposition and denoising processing on the amplified data to obtain denoised data; then uses the improved particle swarm optimization algorithm and density clustering algorithm to identify abnormal data on the denoised data to obtain target data; finally, obtains template training data, inputs the template training data into a preset prediction model for training, and inputs the target data into the trained preset prediction model for effect prediction to output a swallowing effect prediction result. The present invention avoids the problem of too little sample data affecting the model prediction accuracy by amplifying, denoising and identifying abnormalities in the data, can effectively remove the noise in the data while retaining the characteristics of the original data, and finally effectively eliminates low-quality data, thereby improving the final prediction accuracy of the model.

[0163] Example 2

[0164] like Figure 2 As shown, in a second embodiment of the present invention, a swallowing effect prediction system based on a swallowing training device is provided, and the system includes:

[0165] Processing module 1, used for acquiring swallowing training data transmitted by the swallowing training device, and preprocessing the swallowing training data to obtain preprocessed data;

[0166] an amplification module 2, configured to perform Fourier transform and data amplification on the preprocessed data to obtain amplified data;

[0167] Denoising module 3, used for performing high-frequency and low-frequency decomposition and denoising processing on the amplified data to obtain denoised data;

[0168] Identification module 4, used to identify abnormal data on the denoised data using an improved particle swarm optimization algorithm and a density clustering algorithm to obtain target data;

[0169] Prediction module 5 is used to obtain template training data, input the template training data into a preset prediction model for training, input the target data into the trained preset prediction model for effect prediction, and output a swallowing effect prediction result;

[0170] The amplification module 2 includes:

[0171] The transformation submodule is used to perform fast Fourier transform on the pre-processed data and add Zero points to obtain the transformed data :

[0172] ;

[0173] Where, is the frequency resolution, is the sampling rate, is the number of sampling points used for fast Fourier transform;

[0174] The conversion submodule is used to convert the transformation data into a real number part. and the imaginary part :

[0175] ;

[0176] Where, is the imaginary number symbol;

[0177] Add a submodule for adding random perturbations to the real part and the imaginary part , to obtain the first perturbation part With the second perturbation part :

[0178] ; ; ;

[0179] Where, 、 Respectively represent the lower and upper frequency limits of the transformed data, Indicates the number of groups of transformed data;

[0180] The intermediate amplicon module is used to convert the first perturbation part With the second perturbation part performing combination to obtain combined data, performing fast inverse Fourier transform on the combined data to obtain restored data, and adding first several data of the restored data to the preprocessed data to obtain intermediate amplified data;

[0181] The final amplicon module is configured to: With the imaginary part Confirm amplification data.

[0182] The final amplicon module comprises:

[0183] Phase unit for the real part based on the With the imaginary part Calculate data phase :

[0184] ;

[0185] The replacement unit is used to replace the data phase of several random sets of transformed data with Random values ​​between to get the adjusted phase ;

[0186] An adjusting unit for adjusting the phase based on the Determine adjustment data :

[0187] ;

[0188] Where, is the fast inverse Fourier transform;

[0189] The amplification unit is configured to add the first several data of the adjustment data to the intermediate amplified data to obtain the amplified data.

[0190] The denoising module 3 includes:

[0191] The decomposition submodule is used to decompose the amplification data into several Components and residual components;

[0192] The contribution submodule is used to set the contribution function of the amplification data:

[0193] ;

[0194] Where, Represents a function variable, To amplify the data, To solve the interval, For time, is the function value;

[0195] The interval submodule is used to determine the distribution characteristics of the contribution function on the coordinate axis and determine the solution interval in the function value set as the final interval ;

[0196] An association submodule for Calculate each of the Correlation of components :

[0197] ;

[0198] Where, Indicates the indivual Quantity;

[0199] The denoising submodule is configured to determine denoised data based on the correlation degree.

[0200] The denoising submodule includes:

[0201] The dividing unit is used to set a first correlation threshold and a second correlation threshold, and divide the The components are stored in a reconstructed component set, wherein the first correlation threshold is greater than the second correlation threshold;

[0202] The denoising component unit is used to convert the correlation degree not less than the second correlation threshold and less than the first correlation threshold The components are stored in the denoising component set;

[0203] Smoothing unit, used to determine the mean square error , and the denoising component set is calculated based on the mean square error The components are denoised and smoothed to obtain the processed Quantity;

[0204] ;

[0205] Where, are the polynomial coefficients, is a power of the polynomial, In the denoising component set The number of servings, is the first indivual Quantity;

[0206] A reconstruction unit for processing The components, the reconstructed component set and the residual component are combined and the signal is reconstructed to obtain denoised data.

[0207] The identification module 4 includes:

[0208] The submodule to be optimized is used to convert the neighborhood radius in the density clustering algorithm and the minimum number of points in the neighborhood As two optimization targets of the improved particle swarm optimization algorithm, the neighborhood radius and the minimum number of points in the neighborhood Corresponding to the velocity and position of the particle respectively;

[0209] An initial submodule, used to determine the initial position and initial velocity of each particle in the improved particle swarm optimization algorithm, and perform initial clustering on the denoised data based on the initial position and the initial velocity using a density clustering algorithm;

[0210] Fitness submodule, used to calculate the fitness of each particle in the clustering process :

[0211] ;

[0212] Where, Indicates the denoised data The average distance between a sample and other samples in its cluster, Indicates the denoised data The average distance between a sample and other samples in the cluster;

[0213] The first update submodule is used to iteratively update the inertia weight of each particle :

[0214] ;

[0215] Where, Indicates the current iteration number, represents the total number of iterations, 、 Represent the starting weight and final weight respectively;

[0216] The second update submodule is used to iteratively update the first learning factor and the second learning factor :

[0217] ; ;

[0218] Where, Respectively represent the starting value and final value of the first learning factor, Respectively represent the starting value and final value of the second learning factor;

[0219] Output submodule, used to output the inertia weight according to the fitness and the updated , the updated first learning factor With the updated second learning factor Determine target data.

[0220] The output submodule includes:

[0221] Update unit, for updating the fitness, inertia weight , the first learning factor and the second learning factor Iteratively update the velocity and position of each particle:

[0222] ;

[0223] ;

[0224] Where, Respectively represent 、 The velocity of the particle after iterations, Respectively represent 、 The position of the particle after iterations, Represent the first random number and the second random number respectively, They represent the individual optimal solution and the group optimal solution respectively;

[0225] A parameter output unit is used to repeat the iterative process until the iteration stop condition is met and output the updated neighborhood radius obtained by the final iteration and the minimum number of points included in the updated neighborhood;

[0226] A clustering unit is used to perform density clustering on the denoised data based on the updated neighborhood radius and the minimum number of points included in the updated neighborhood by using a density clustering algorithm to obtain target data.

[0227] In other embodiments of the present invention, the embodiments of the present invention provide the following technical solutions: a computer comprising a memory 102, a processor 101, and a computer program stored on the memory 102 and executable on the processor 101; when the processor 101 executes the computer program, the swallowing effect prediction method based on the swallowing training instrument as described above is implemented.

[0228] Specifically, the processor 101 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured as one or more integrated circuits for implementing the embodiments of the present invention.

[0229] Memory 102 may include a large-capacity memory for data or instructions. By way of example, and not limitation, memory 102 may include a hard disk drive (HDD), a floppy disk drive, a solid-state drive (SSD), flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 102 may include removable or non-removable (or fixed) media. Where appropriate, memory 102 may be internal or external to the data processing device. In certain embodiments, memory 102 is non-volatile memory. In certain embodiments, memory 102 includes read-only memory (ROM) and random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM) or a flash memory (FLASH), or a combination of two or more of these. Under appropriate circumstances, the RAM can be a static random access memory (SRAM) or a dynamic random access memory (DRAM), where the DRAM can be a fast page mode dynamic random access memory (FPMDRAM), an extended data output dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0230] The memory 102 may be used to store or cache various data files that need to be processed and / or used for communication, as well as possible computer program instructions executed by the processor 101 .

[0231] The processor 101 reads and executes the computer program instructions stored in the memory 102 to implement the above-mentioned swallowing effect prediction method based on the swallowing training device.

[0232] In some embodiments, the computer may further include a communication interface 103 and a bus 100. Figure 3 As shown, the processor 101 , the memory 102 , and the communication interface 103 are connected via a bus 100 and communicate with each other.

[0233] The communication interface 103 is used to implement communication between the various modules, devices, units, and / or equipment in the embodiments of the present invention. The communication interface 103 can also implement data communication with other components such as external devices, image / data acquisition equipment, databases, external storage, and image / data processing workstations.

[0234] Bus 100 includes hardware, software, or both, and couples components of a computer device to each other. Bus 100 includes, but is not limited to, at least one of the following: a data bus, an address bus, a control bus, an expansion bus, and a local bus. By way of example and not limitation, bus 100 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses, or a combination of two or more of these. Bus 100 may include one or more buses, where appropriate. Although embodiments of the present invention describe and illustrate a particular bus, the present invention contemplates any suitable bus or interconnect.

[0235] The computer can execute the swallowing effect prediction method based on the swallowing training instrument of the present invention based on the swallowing training instrument based on the acquired swallowing effect prediction system, thereby realizing the swallowing effect prediction based on the swallowing training instrument.

[0236] In some further embodiments of the present invention, in combination with the above-mentioned swallowing effect prediction method based on the swallowing training device, the embodiments of the present invention provide the following technical solutions: a storage medium having a computer program stored thereon, and when the computer program is executed by the processor, the above-mentioned swallowing effect prediction method based on the swallowing training device is implemented.

[0237] Those skilled in the art will appreciate that the logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device), or in conjunction with such instruction execution system, apparatus, or device. For purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by an instruction execution system, apparatus, or device, or in conjunction with such instruction execution system, apparatus, or device.

[0238] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0239] It should be understood that various components of the present invention may be implemented using hardware, software, firmware, or a combination thereof. In the aforementioned embodiments, multiple steps or methods may be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one or a combination of the following technologies known in the art may be used: a discrete logic circuit having logic gate circuits for implementing logic functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0240] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0241] The above-described embodiments merely represent several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent. It should be noted that a person of ordinary skill in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, and these variations and improvements fall within the scope of protection of the present invention. Therefore, the scope of protection of the patent for this invention shall be determined by the appended claims.

Claims

1. A swallowing effect prediction method based on a swallowing training device, characterized in that: include: Acquiring swallowing training data transmitted by a swallowing training device, and preprocessing the swallowing training data to obtain preprocessed data; Performing Fourier transform and data amplification on the preprocessed data to obtain amplified data; performing high-frequency and low-frequency decomposition and denoising processing on the amplified data to obtain denoised data; An improved particle swarm optimization algorithm and a density clustering algorithm are used to identify abnormal data on the denoised data to obtain target data; Acquire template training data, input the template training data into a preset prediction model for training, input the target data into the trained preset prediction model for effect prediction, and output a swallowing effect prediction result; The step of performing Fourier transform and data amplification on the pre-processed data to obtain amplified data includes: Perform fast Fourier transform on the preprocessed data, and add Zero points to obtain the transformed data : ； Where, is the frequency resolution, is the sampling rate, is the number of sampling points used for fast Fourier transform; The transformation data is converted to obtain the real number part and the imaginary part : ； Where, is the imaginary number symbol; Add random perturbations to the real part and the imaginary part , to obtain the first perturbation part With the second perturbation part : ； ； ； Where, 、 Respectively represent the lower and upper frequency limits of the transformed data, Indicates the number of groups of transformed data; The first perturbation part With the second perturbation part performing combination to obtain combined data, performing fast inverse Fourier transform on the combined data to obtain restored data, and adding first several data of the restored data to the preprocessed data to obtain intermediate amplified data; Based on the intermediate amplified data, the real number part With the imaginary part Determine amplification data; Based on the intermediate amplified data, the real number part With the imaginary part The steps to determine amplification data include: Based on the real part With the imaginary part Calculate data phase : ； Replace the data phase of several random sets of transformed data with Random values ​​between to get the adjusted phase ; Based on the adjusted phase Determine adjustment data : ； Where, is the fast inverse Fourier transform; Adding first several data of the adjusted data to the intermediate amplified data to obtain amplified data; The step of performing high-frequency and low-frequency decomposition and denoising processing on the amplified data to obtain denoised data includes: The amplification data is decomposed into several Components and residual components; Set the contribution function of the amplification data: ； Where, Represents a function variable, To amplify the data, To solve the interval, For time, is the function value; Determine the distribution characteristics of the contribution function on the coordinate axis and determine the solution interval in the function value set as the final interval ; Based on the final interval Calculate each of the Correlation of components : ； Where, Indicates the indivual Quantity; determining denoised data based on the degree of association; The step of determining the denoised data based on the correlation degree comprises: Set the first correlation threshold and the second correlation threshold, and select The components are stored in a reconstructed component set, wherein the first correlation threshold is greater than the second correlation threshold; The correlation degree is not less than the second correlation threshold and less than the first correlation threshold. The components are stored in the denoising component set; Determine the mean square error , and the denoising component set is calculated based on the mean square error The components are denoised and smoothed to obtain the processed Quantity; ； Where, are the polynomial coefficients, is a power of the polynomial, In the denoising component set The number of servings, is the first indivual Quantity; The processing The components, the reconstructed component set and the residual component are combined and the signal is reconstructed to obtain denoised data.

2. The swallowing effect prediction method based on the swallowing training instrument according to claim 1, characterized in that: The step of using the improved particle swarm optimization algorithm and the density clustering algorithm to identify abnormal data on the denoised data to obtain target data includes: The neighborhood radius in the density clustering algorithm and the minimum number of points in the neighborhood As two optimization targets of the improved particle swarm optimization algorithm, the neighborhood radius and the minimum number of points in the neighborhood Corresponding to the velocity and position of the particle respectively; Determining the initial position and initial velocity of each particle in the improved particle swarm optimization algorithm, and performing initial clustering on the denoised data based on the initial position and the initial velocity using a density clustering algorithm; Calculate the fitness of each particle in the clustering process : ； Where, Indicates the denoised data The average distance between a sample and other samples in its cluster, Indicates the denoised data The average distance between a sample and other samples in the cluster; Iteratively update the inertia weight of each particle : ； Where, Indicates the current iteration number, represents the total number of iterations, 、 Represent the starting weight and final weight respectively; Iteratively update the first learning factor and the second learning factor : ； ； Where, Respectively represent the starting value and final value of the first learning factor, Respectively represent the starting value and final value of the second learning factor; According to the fitness, the updated inertia weight , the updated first learning factor With the updated second learning factor Determine target data.

3. The swallowing effect prediction method based on the swallowing training instrument according to claim 2, characterized in that: According to the fitness, the updated inertia weight , the updated first learning factor With the updated second learning factor The steps to determine target data include: According to the fitness, inertia weight , the first learning factor and the second learning factor Iteratively update the velocity and position of each particle: ； ； Where, Respectively represent 、 The velocity of the particle after iterations, Respectively represent 、 The position of the particle after iterations, Represent the first random number and the second random number respectively, They represent the individual optimal solution and the group optimal solution respectively; Repeat the iterative process until the iteration stop condition is met and output the updated neighborhood radius obtained by the final iteration and the minimum number of points included in the updated neighborhood; Based on the updated neighborhood radius and the minimum number of points included in the updated neighborhood, the denoised data is density clustered using a density clustering algorithm to obtain target data.

4. A swallowing effect prediction system based on a swallowing training device, wherein the system adopts the swallowing effect prediction system based on a swallowing training device according to claim 1, characterized in that: The system comprises: a processing module, configured to obtain swallowing training data transmitted by the swallowing training device and preprocess the swallowing training data to obtain preprocessed data; an amplification module, configured to perform Fourier transform and data amplification on the preprocessed data to obtain amplified data; a denoising module, configured to perform high-frequency and low-frequency decomposition and denoising processing on the amplified data to obtain denoised data; An identification module is used to identify abnormal data on the denoised data using an improved particle swarm optimization algorithm and a density clustering algorithm to obtain target data; The prediction module is used to obtain template training data, input the template training data into a preset prediction model for training, input the target data into the trained preset prediction model for effect prediction, and output a swallowing effect prediction result.

5. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the swallowing effect prediction method based on the swallowing training instrument according to any one of claims 1 to 3 is implemented.

6. A storage medium, characterized in that The storage medium stores a computer program, and when the computer program is executed by the processor, the method for predicting swallowing effect based on the swallowing training device according to any one of claims 1 to 3 is implemented.

Images