Swallowing monitoring system and control method thereof
By perceived correlation of hyoid bone and laryngeal movement data of patients with swallowing dysphagia and combined with cough characteristics, heterogeneous association fusion of multi-source sensor data is achieved, solving the problem of delayed dynamic monitoring of cough information during swallowing, and improving diagnosis and Accuracy of treatment.
Patent Information
- Application Number
- CN202410995093.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-24
- Publication Date
- 2024-11-01
AI Technical Summary
In the existing swallowing monitoring technology, the heterogeneous association fusion problem of multi-source sensor data leads to delayed dynamic monitoring of choking information during swallowing, and the noise complexity is high, affecting the accuracy of diagnosis and treatment.
By perceived correlation of hyoid and laryngeal motion data, perceived correlation and interactive entropy are determined, and the cough characteristics and sensor correlation relationship are combined to achieve dynamic monitoring and evaluation of choking information during swallowing.
It reduces the delay of multimodal association fusion, provides a more comprehensive evaluation of swallowing function, improves diagnosis and treatment accuracy, and enhances understanding of the status of patients with swallowing dysphagia and monitoring of choking.
Smart Images

Figure CN118873099A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of swallowing monitoring, and more specifically, to a swallowing monitoring system and a control method thereof. Background Art
[0002] Swallowing monitoring is a technical means of monitoring the patient's swallowing behavior through a monitoring system, which is used to record and analyze the patient's individual swallowing behavior. This monitoring is usually achieved through various means such as sound, movement or image, and aims to evaluate and manage the normality or abnormality of swallowing function. Swallowing monitoring is of great significance for diagnosing various swallowing disorders (such as dysphagia caused by neuromuscular diseases) and formulating personalized rehabilitation treatments. By monitoring and analyzing key parameters in the swallowing process, it can help medical professionals understand the patient's swallowing function status more accurately and intervene and manage accordingly.
[0003] In the swallowing monitoring process of patients with dysphagia, the main monitoring methods are bedside evaluation and instrument evaluation. Instrument evaluation, such as video fluoroscopic swallowing study (VFSS) or fiber optic endoscopic evaluation of swallowing (FEES), but the radiation problem in VFSS, the lack of professional talents for FEES operation, the specificity of the equipment, the time and economic cost of evaluation limit its widespread use. By monitoring different parts of the patient through multi-source sensors and mining the monitored data, swallowing information such as the number of swallows and the number of choking and coughing during swallowing can be obtained in time. However, since the monitoring data attributes of multi-source sensors are multimodal, there is a complex nonlinear separable space composed of noise, which leads to a complex conversion process, resulting in excessive delay during multimodal fusion. Using multi-source sensor data for heterogeneous correlation fusion can effectively solve the above problems. Therefore, how to perform heterogeneous correlation fusion on multi-source sensor data in the swallowing monitoring process to realize dynamic monitoring of choking and coughing information when patients swallow has become a difficult problem faced by the industry. Summary of the invention
[0004] The present application provides a swallowing monitoring system and a control method thereof, which can perform heterogeneous correlation fusion on multi-source sensor data during the swallowing monitoring process to realize dynamic monitoring of the patient's choking information during swallowing.
[0005] In a first aspect, the present application provides a control method for a swallowing monitoring system, comprising the following steps: Based on a multi-source sensor group, the movement information of the hyoid bone and larynx of patients with dysphagia is monitored, and the hyoid bone movement data and laryngeal movement data are obtained; Performing swallowing perception association on the hyoid bone movement data and the laryngeal movement data to obtain hyoid bone-larynx perception association information, and then determining the perception association degree of the hyoid bone and larynx associated movement of the dysphagia patient from the perception association information; Determine the interaction entropy of each associated node when the hyoid bone and larynx of the dysphagia patient are in associated motion according to the hyoid bone movement data and the laryngeal movement data, determine the mutual correlation coefficient of each associated node during the swallowing process of the dysphagia patient according to all the interaction entropies and the perception correlation degree, and then perform perception fusion on the swallowing associated data of all associated nodes in the perception association information through all the mutual correlation coefficients to obtain the first fusion feature of the dysphagia patient when swallowing; Acoustic information of cough of dysphagia patients is monitored based on the multi-source sensor group, and cough data is obtained, cough features of dysphagia patients are extracted from the cough data, and second fusion features of choking cough of dysphagia patients during swallowing are determined based on the cough features, the first fusion features, and the correlation between the sensors in the multi-source sensor group; The swallowing information of the dysphagia patient is determined according to the second fusion feature and a preset swallowing assessment standard.
[0006] Preferably, performing swallowing perception association on the hyoid bone movement data and the laryngeal movement data to obtain hyoid bone-larynx perception association information specifically includes: Dynamically sensing the hyoid bone movement data to obtain a plurality of hyoid bone dynamic data segments; Dynamically sensing the laryngeal motion data to obtain a plurality of laryngeal dynamic data segments; Performing swallowing association on all hyoid bone dynamic data segments and all laryngeal dynamic data segments to obtain a plurality of associated hyoid bone-laryngeal dynamic data segments; The sensory association information of the hyoid bone and the larynx is determined through all the hyoid bone and the larynx dynamic data segments.
[0007] Preferably, determining the perceived correlation degree of the associated movement of the hyoid bone and larynx of a dysphagia patient from the perceived correlation information specifically includes: Acquire all associated nodes of the hyoid bone and larynx of the dysphagia patient during associated movement from the perceptual associated information; Determining the hyoid-laryngeal motion ratio of each associated node based on the perceived associated information; The perceptual correlation of the hyoid and laryngeal movements in dysphagic patients was determined using all hyoid-laryngeal motion ratios.
[0008] Preferably, determining the mutual correlation coefficients of the associated nodes in the swallowing process of the dysphagia patient by all the interaction entropies and the perceived correlations specifically includes: Determine the mutual correlation between adjacent associated nodes through all interaction entropies; The mutual correlation coefficients of the various correlation nodes in the swallowing process of the dysphagia patient are determined according to all the mutual correlations and the perceptual correlation.
[0009] Preferably, the swallowing-related data of all associated nodes in the perception-related information are subjected to perception fusion through all the mutual correlation coefficients to obtain the first fusion feature of swallowing of the dysphagia patient, which specifically includes: Acquire swallowing-related data of each associated node from the perception-related information, and then determine the perception variance of each associated node; Determine the fusion weight coefficient of the corresponding associated node through all the mutual correlation coefficients and the perception variance of each associated node; The swallowing-related data of each associated node are fused based on all fusion weight coefficients to obtain the first fusion feature of the swallowing of the dysphagia patient.
[0010] Preferably, determining the second fusion feature of the choking cough of the dysphagia patient during swallowing based on the cough feature, the first fusion feature and the correlation between the sensors in the multi-source sensor group specifically includes: Determine the association relationship between each sensor in the multi-source sensor group; Performing correlation analysis on the cough feature and the first fusion feature to obtain correlation features of choking and coughing during swallowing in patients with dysphagia; The associated features are associated and fused according to the associated relationship to obtain a second fused feature of choking during swallowing of a dysphagia patient.
[0011] Preferably, determining the swallowing information of the dysphagia patient according to the second fusion feature and the preset swallowing assessment standard specifically includes: determining a plurality of swallowing feature parameters according to the second fusion feature and a preset swallowing assessment standard; The swallowing information of patients with dysphagia is determined through all swallowing characteristic parameters.
[0012] In a second aspect, the present application provides a swallowing monitoring system, the swallowing monitoring system comprising a swallowing monitoring control unit, the swallowing monitoring control unit comprising: An acquisition module, used for monitoring the movement information of the hyoid bone and larynx of a dysphagia patient based on a multi-source sensor group, and then acquiring hyoid bone movement data and laryngeal movement data; A processing module, configured to perform swallowing perception association on the hyoid bone movement data and the laryngeal movement data to obtain hyoid bone-larynx perception association information, and further determine the perception association degree of the hyoid bone and larynx of the dysphagia patient during the associated movement from the perception association information; The processing module is further used to determine the interaction entropy of each associated node when the hyoid bone and larynx of the dysphagia patient move in association according to the hyoid bone movement data and the laryngeal movement data, determine the mutual correlation coefficient of each associated node during the swallowing process of the dysphagia patient according to all the interaction entropies and the perception correlation degree, and then perform perception fusion on the swallowing association data of all associated nodes in the perception association information through all the mutual correlation coefficients to obtain a first fusion feature when the dysphagia patient swallows; The processing module is further used to monitor the acoustic information of the dysphagia patient's cough based on the multi-source sensor group, thereby obtaining cough data, extracting the cough feature of the dysphagia patient from the cough data, and then determining the second fusion feature of the choking cough of the dysphagia patient during swallowing based on the cough feature, the first fusion feature and the correlation relationship between the sensors in the multi-source sensor group; An execution module is used to determine swallowing information of a dysphagia patient according to the second fusion feature and a preset swallowing assessment standard.
[0013] In a third aspect, the present application provides a computer device, comprising a memory and a processor, wherein the memory stores codes, and the processor is configured to obtain the codes and execute the above-mentioned control method of the swallowing monitoring system.
[0014] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the control method of the swallowing monitoring system described above is implemented.
[0015] The technical solution provided by the embodiments disclosed in this application has the following beneficial effects: In an embodiment of the present application, the movement information of the hyoid bone and larynx of a dysphagia patient is monitored based on a multi-source sensor group, thereby obtaining hyoid bone movement data and laryngeal movement data; swallowing perception association is performed on the hyoid bone movement data and the laryngeal movement data to obtain hyoid bone-larynx perception association information, and then the perception association degree of the hyoid bone and larynx of the dysphagia patient during the associated movement is determined from the perception association information; the interaction entropy of each associated node during the associated movement of the hyoid bone and larynx of the dysphagia patient is determined based on the hyoid bone movement data and the laryngeal movement data, and the interaction entropy of each associated node during the swallowing process of the dysphagia patient is determined from all the interaction entropies and the perception association degree. Correlation coefficient, and then perform perceptual fusion on the swallowing-related data of all associated nodes in the perceptual correlation information through all the mutual correlation coefficients to obtain the first fusion feature of the swallowing of the dysphagia patient; based on the multi-source sensor group, monitor the acoustic information of the cough of the dysphagia patient, and then obtain the cough data, extract the cough feature of the dysphagia patient from the cough data, and then determine the second fusion feature of the choking cough of the dysphagia patient when swallowing according to the cough feature, the first fusion feature and the correlation relationship between each sensor in the multi-source sensor group; determine the swallowing information of the dysphagia patient's swallowing according to the second fusion feature and the preset swallowing evaluation standard.
[0016] It can be seen that the present application determines the swallowing information of swallowing patients with dysphagia based on the second fusion feature and the preset swallowing assessment standard; wherein, the determination of the perceptual correlation can quantify the degree of associated movement of the hyoid bone and the larynx during the swallowing process, providing a basis for subsequent analysis, and the associated fusion between heterogeneous data enables the data of different sensors to work together to form more comprehensive swallowing movement data, thereby improving the understanding of the swallowing state of patients with dysphagia, and facilitating the assessment of patients' swallowing disorders and the formulation of treatment plans; secondly, in the process of determining the first fusion feature, by calculating the interaction entropy of each associated node when the hyoid bone and the larynx move, and combining the perceptual correlation, the mutual correlation coefficient of each associated node can be obtained, and the first fusion feature formed by perceptual fusion of the swallowing associated data in the perceptual correlation information through all the mutual correlation coefficients can reduce the complexity of the nonlinear separable space constituted by the noise, thereby reducing the delay during multimodal association fusion, and through the The first fusion feature reflects the patient's overall swallowing function during the swallowing process, and enables the interactive characteristics of multi-source sensor data at different nodes to be comprehensively evaluated, thereby constructing a comprehensive swallowing function evaluation model. Through this fusion of multi-dimensional data, the patient's swallowing ability can be more accurately evaluated, and more accurate diagnosis and treatment recommendations can be provided; then, the second fusion feature combines cough data, swallowing movement data and the correlation information between sensors, reflecting the overall characteristics of the patient's choking during swallowing, which helps to comprehensively monitor the patient's choking during swallowing, and then dynamically monitor the patient's choking information during swallowing. It can evaluate the patient's swallowing safety and effectiveness in real time, and provide clinicians with more detailed swallowing disorder information to optimize treatment plans and improve patients' quality of life; in summary, based on the above scheme, the multi-source sensor data in the swallowing monitoring process can be heterogeneously associated and fused to achieve dynamic monitoring of the patient's choking information during swallowing. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 is an exemplary flow chart of a control method of a swallowing monitoring system according to some embodiments of the present application; Figure 2 is a schematic diagram of a process for determining a first fusion feature according to some embodiments of the present application; Figure 3 is a schematic diagram of a process for determining a second fusion feature according to some embodiments of the present application; Figure 4 is a schematic diagram of exemplary software modules of a swallowing monitoring control unit according to some embodiments of the present application; Figure 5 It is a structural schematic diagram of a computer device for implementing a control method of a swallowing monitoring system according to some embodiments of the present application. DETAILED DESCRIPTION
[0018] The core of this application is to determine the perceptual correlation by performing swallowing perception association on the hyoid bone movement data and the laryngeal movement data; determine the interaction entropy of the associated nodes when the hyoid bone and the larynx are associated, determine the mutual correlation coefficients of each associated node in the swallowing process by all the interaction entropies and the perceptual correlation, perform perceptual fusion on the swallowing associated data of all associated nodes in the perceptual association information by all the mutual correlation coefficients, and obtain the first fusion feature of swallowing of patients with dysphagia; determine the second fusion feature of the patient's choking cough when swallowing by the extracted cough feature, the first fusion feature and the correlation relationship between each sensor in the multi-source sensor group; determine the swallowing information of the patient's swallowing based on the second fusion feature. Using the scheme of this application, the multi-source sensor data in the swallowing monitoring process can be heterogeneously associated and fused to achieve dynamic monitoring of the patient's choking cough information when swallowing.
[0019] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. Figure 1 , which is an exemplary flow chart of a control method of a swallowing monitoring system according to some embodiments of the present application. The control method 100 of the swallowing monitoring system mainly includes the following steps: In step 101, the movement information of the hyoid bone and the larynx of the dysphagia patient is monitored based on a multi-source sensor group, thereby obtaining hyoid bone movement data and laryngeal movement data.
[0020] In specific implementation, monitoring the movement information of the hyoid bone and larynx of patients with dysphagia based on a multi-source sensor group, and then obtaining the hyoid bone movement data and the laryngeal movement data can be achieved in the following manner, namely: setting the sampling frequency of each sensor in the multi-source sensor group, collecting the movement information of the hyoid bone and larynx of patients with dysphagia through the multi-source sensor group with the set sampling frequency, and storing the collected movement information in a data storage unit, and further obtaining the hyoid bone movement data and the laryngeal movement data collected by the multi-source sensor group within a specified time period from the data storage unit, wherein the specified time period refers to the time period corresponding to the past hour to the current moment, and in other embodiments, the specified time period can also be set to other time periods, which are not limited here.
[0021] It should be noted that the multi-source sensor group in the present application refers to a unit composed of multiple sensors of different types. In the present embodiment, the types of sensors in the multi-source sensor group include acceleration sensors, ultrasonic sensors and microphones. The acceleration sensor is used to detect the movement and vibration of the hyoid bone and larynx, the ultrasonic sensor is used to obtain the distance change between the hyoid bone and the larynx, and the microphone is used to collect acoustic data of coughing, wherein the sampling frequency can be set in the range of 22-32kHz; it should also be noted that the motion information in the present application includes: hyoid bone displacement, hyoid bone movement time, laryngeal displacement and laryngeal movement time; in addition, the hyoid bone movement data and laryngeal movement data are both numerical data.
[0022] It should also be noted that the hyoid bone movement data in the present application includes static data segments and dynamic data segments, wherein the static data segment refers to the data segment monitored when the hyoid bone is stationary, and the dynamic data segment refers to the data segment monitored when the hyoid bone is moving; similarly, the laryngeal movement data also includes static data segments and dynamic data segments, wherein the static data segment refers to the data segment monitored when the larynx is stationary, and the dynamic data segment refers to the data segment monitored when the larynx is moving.
[0023] In step 102, swallowing perception association is performed on the hyoid bone movement data and the laryngeal movement data to obtain hyoid bone-larynx perception association information, and then the perception association degree of the hyoid bone and larynx of the dysphagia patient during the associated movement is determined based on the perception association information.
[0024] In some embodiments, the swallowing perception association is performed on the hyoid bone movement data and the laryngeal movement data to obtain the hyoid bone-larynx perception association information, which can be achieved by using the following steps: Dynamically sensing the hyoid bone movement data to obtain a plurality of hyoid bone dynamic data segments; Dynamically sensing the laryngeal motion data to obtain a plurality of laryngeal dynamic data segments; Performing swallowing association on all hyoid bone dynamic data segments and all laryngeal dynamic data segments to obtain a plurality of associated hyoid bone-laryngeal dynamic data segments; The sensory association information of the hyoid bone and the larynx is determined through all the hyoid bone and the larynx dynamic data segments.
[0025] In specific implementation, the dynamic perception of the hyoid movement data to obtain multiple hyoid dynamic data segments can be achieved in the following manner, namely: initialize a dynamic perception model, use the hyoid movement data as the input data of the dynamic perception model, and then use each data segment output by the dynamic perception model as the hyoid dynamic data segment; in addition, the dynamic perception of the laryngeal movement data to obtain multiple laryngeal dynamic data segments can be achieved in the following manner, namely: initialize another dynamic perception model, use the laryngeal movement data as the input data of the dynamic perception model, and then use each data segment output by the dynamic perception model as the laryngeal dynamic data segment.
[0026] It should be noted that the dynamic perception model in the present application is a dynamic data segment used to detect hyoid movement or laryngeal movement. The dynamic perception model is a detection model obtained by preprocessing, feature extraction, model training and verification of a large amount of hyoid movement data or laryngeal movement data. The algorithm structure in the model training process can adopt a convolutional neural network. In other embodiments, other existing algorithm structures can also be used, which will not be repeated here.
[0027] In specific implementation, swallowing association is performed on all hyoid bone dynamic data segments and all laryngeal dynamic data segments to obtain multiple associated hyoid bone-laryngeal dynamic data segments, which can be achieved in the following manner, namely: obtaining the timestamps of all laryngeal dynamic data segments, selecting a hyoid bone dynamic data segment, obtaining the timestamp of the hyoid bone dynamic data segment, subtracting the timestamps of all laryngeal dynamic data segments from the timestamp of the hyoid bone dynamic data segment, and comparing all the differences to obtain the minimum difference, and then comparing the minimum difference with a preset time difference threshold. When the minimum difference is less than the preset time difference threshold, the laryngeal dynamic data segment corresponding to the minimum difference is used as the associated data segment of the hyoid bone dynamic data segment, and the array composed of the associated data segment and the hyoid bone dynamic data segment is used as the associated hyoid bone-laryngeal dynamic data segment. The above steps are repeated to obtain the remaining hyoid bone-laryngeal dynamic data segments.
[0028] It should be noted that the time difference threshold in the present application is an indicator for measuring the maximum time difference of the associated movement of the hyoid bone and the larynx. The time difference threshold can be set based on a large amount of experimental data on the associated movement of the hyoid bone and the larynx. In other embodiments, the time difference threshold can also be preset in other ways, which is not limited here.
[0029] In specific implementation, determining the perceptual association information of the hyoid bone and the larynx through all the hyoid bone and the larynx dynamic data segments can be achieved in the following manner, namely: taking the set consisting of all the hyoid bone and the larynx dynamic data segments as the perceptual association information of the hyoid bone and the larynx. It should be noted that the perceptual association information in the present application includes the information of the hyoid bone and the larynx associated movement. In this embodiment, by analyzing the information of the hyoid bone and the larynx associated movement, effective judgment support can be provided for the patient's swallowing behavior.
[0030] It should be noted that the swallowing perception association in the present application refers to the process of quantifying the mutual correlation between the hyoid bone and the larynx during the swallowing process. By performing correlation analysis on the hyoid bone movement and the laryngeal movement, the mutual correlation between the hyoid bone and the larynx during the swallowing process can be quantified. This correlation reflects the interaction between the hyoid bone and the larynx during the swallowing process, and the degree of coordination between them in completing the swallowing action together.
[0031] In some embodiments, determining the perceived correlation degree of the associated movement of the hyoid bone and the larynx of a dysphagia patient from the perceived correlation information can be achieved by the following steps: Acquire all associated nodes of the hyoid bone and larynx of the dysphagia patient during associated movement from the perceptual associated information; Determining the hyoid-laryngeal motion ratio of each associated node based on the perceived associated information; The perceptual correlation of the hyoid and laryngeal movements in dysphagic patients was determined using all hyoid-laryngeal motion ratios.
[0032] It should be noted that the associated nodes in the present application refer to the key nodes of the movement of the hyoid bone and larynx during swallowing. These nodes contain position, time and movement information. The information contained in each associated node helps to describe the dynamic behavior of the hyoid bone and larynx during swallowing. In the present application, the moment when swallowing starts, the moment when the hyoid bone reaches the maximum displacement and the moment when the larynx rises the maximum can be used as associated nodes. In specific implementation, all associated nodes of the associated movement of the hyoid bone and larynx of patients with dysphagia are obtained from the perceived associated information. The following method can be used, namely: all hyoid-larynx dynamic data segments are obtained from the perceived associated information, and the associated nodes of the associated movement of the hyoid bone and larynx of patients with dysphagia are identified from each hyoid-larynx dynamic data segment through a pre-trained node recognition model; it should also be noted that the pre-trained node recognition model in the present application realizes the recognition of key nodes by detecting abnormal data in the hyoid-larynx dynamic data segment. In other embodiments, the node recognition model can also realize the recognition of key nodes based on big data training, which is not limited here.
[0033] In specific implementation, determining the hyoid bone-laryngeal motion ratio of each associated node based on the perceived association information can be achieved in the following manner, namely: obtaining the hyoid bone displacement data and the laryngeal displacement data at each associated node from the perceived association information, and then taking the ratio of the hyoid bone displacement data to the laryngeal displacement data as the hyoid bone-laryngeal motion ratio of the corresponding associated node.
[0034] It should be noted that the hyoid-laryngeal motion ratio in the present application is an indicator for measuring the relative movement degree of the hyoid bone displacement and the laryngeal displacement, which reflects the degree of coordination between the hyoid bone and the larynx during swallowing. The larger the hyoid-laryngeal motion ratio, the better the coordination between the hyoid bone and the larynx during swallowing, and the smaller the hyoid-laryngeal motion ratio, the worse the coordination between the hyoid bone and the larynx during swallowing.
[0035] In specific implementation, determining the perceived correlation of the hyoid bone and larynx in the dysphagia patient's associated movements through all hyoid bone-laryngeal movement ratios can be achieved in the following manner, namely: firstly calculating the mean value and standard deviation of all hyoid bone-laryngeal movement ratios, and then using the ratio of the mean value to the standard deviation as the perceived correlation of the hyoid bone and larynx in the dysphagia patient's associated movements.
[0036] It should be noted that the perceived correlation in the present application is an indicator to measure the degree of correlation between the hyoid bone and the larynx during the swallowing process. It reflects the degree of closeness of the hyoid bone and the larynx in terms of movement coordination and synchronization. The larger the perceived correlation, the closer the hyoid bone and the larynx are in terms of movement coordination and synchronization. The smaller the perceived correlation, the sparser the hyoid bone and the larynx are in terms of movement coordination and synchronization. Among them, coordination describes the degree of coordination between the movements of the hyoid bone and the larynx, that is, whether the movement of one part (the hyoid bone) can effectively coordinate with the movement of another part (the larynx).
[0037] In step 103, the interaction entropy of each associated node when the hyoid bone and larynx of the dysphagia patient move in association is determined based on the hyoid movement data and the laryngeal movement data, and the mutual correlation coefficient of each associated node during the swallowing process of the dysphagia patient is determined based on all the interaction entropies and the perceptual correlation degree. Then, the swallowing associated data of all associated nodes in the perceptual association information are perceptually fused through all the mutual correlation coefficients to obtain the first fusion feature of the dysphagia patient when swallowing.
[0038] In some embodiments, determining the interaction entropy of each associated node when the hyoid bone and larynx of a dysphagia patient move in association with each other according to the hyoid bone movement data and the laryngeal movement data can be implemented by the following steps: Obtain all the associated nodes when the hyoid bone and larynx are in associated motion in patients with dysphagia; Determine the marginal probability distribution of the hyoid bone of each associated node through the hyoid bone movement data; Determine the marginal probability distribution of each associated node larynx by using the larynx motion data; Determine a joint probability distribution of each associated node using the hyoid bone motion data and the laryngeal motion data; The interaction entropy of each associated node during the associated movement of the hyoid and larynx in patients with dysphagia was determined based on all joint probability distributions, the marginal probability distributions of all associated nodes of the hyoid, and the marginal probability distributions of all associated nodes of the larynx.
[0039] In the specific implementation, first, select an associated node, intercept the hyoid node data segment at the associated node from the hyoid movement data, and intercept the laryngeal node data segment at the associated node from the laryngeal movement data; secondly, calculate the joint probability distribution of the associated node by the probability of co-occurrence of data points in the hyoid node data segment and the laryngeal node data segment; then, calculate the marginal probability distribution of the hyoid bone by the probability of occurrence of data points in the hyoid node data segment, and calculate the marginal probability distribution of the larynx by the probability of occurrence of data points in the laryngeal node data segment; finally, substitute the joint probability distribution, the marginal probability distribution of the hyoid bone and the marginal probability distribution of the larynx into the information entropy calculation formula to obtain the interactive entropy of the associated node; repeat the above steps to obtain the interactive entropy of the remaining associated nodes.
[0040] It should be noted that the hyoid node data segment refers to a data segment composed of hyoid movement data of a predetermined length at the associated node, and the laryngeal node data segment refers to a data segment composed of laryngeal movement data of a predetermined length at the associated node, wherein the predetermined length is usually a length consisting of 100 data points. In other embodiments, it can also be set to other lengths, which are not limited here.
[0041] It should also be noted that the interaction entropy in the present application is an indicator for measuring the uncertainty of the associated motion of the hyoid bone and larynx at the associated node. The larger the interaction entropy, the greater the uncertainty of the associated motion of the hyoid bone and larynx at the associated node. The smaller the interaction entropy, the smaller the uncertainty of the associated motion of the hyoid bone and larynx at the associated node.
[0042] In some embodiments, determining the correlation coefficients of the associated nodes in the swallowing process of the dysphagia patient from all the interaction entropies and the perceived correlation can be implemented by the following steps: Determine the mutual correlation between adjacent associated nodes through all interaction entropies; The mutual correlation coefficients of the various correlation nodes in the swallowing process of the dysphagia patient are determined according to all the mutual correlations and the perceptual correlation.
[0043] In specific implementation, determining the mutual correlation between adjacent associated nodes through all the interaction entropies can be implemented in the following manner, namely: selecting an associated node as a target associated node, selecting an associated node adjacent to the target associated node as a first designated associated node, obtaining the interaction entropy corresponding to the target associated node, obtaining the interaction entropy corresponding to the first designated associated node, taking the ratio between the interaction entropy corresponding to the first designated associated node and the interaction entropy corresponding to the target associated node as the mutual correlation between the target associated node and the first designated associated node, then selecting another associated node adjacent to the target associated node as a second designated associated node, and then obtaining the mutual correlation between the target associated node and the second designated associated node, and then taking the average of the mutual correlation between the target associated node and the first designated associated node and the mutual correlation between the target associated node and the second designated associated node as the mutual correlation between the target associated node and the adjacent associated nodes, and repeating the above steps to obtain the mutual correlation between the remaining associated nodes and the adjacent associated nodes.
[0044] In specific implementation, determining the mutual correlation coefficient of each associated node in the swallowing process of a dysphagia patient based on all the mutual correlations and the perceptual correlation can be achieved in the following manner, namely: selecting a mutual correlation, adjusting the mutual correlation by the perceptual correlation to obtain the mutual correlation coefficient of the associated node corresponding to the mutual correlation, repeating the above steps to obtain the mutual correlation coefficient of the associated nodes corresponding to the remaining mutual correlations. As a preferred embodiment, adjusting the mutual correlation by the perceptual correlation to obtain the mutual correlation coefficient of the associated node corresponding to the mutual correlation can be achieved in the following manner, namely: the natural exponent of the opposite of the perceptual correlation can be used as an adjustment coefficient, and then the product of the adjustment coefficient and the mutual correlation can be used as the mutual correlation coefficient of the associated node corresponding to the mutual correlation. It should be noted that the perceptual correlation can be used as an adjustment coefficient to correct the deviation calculated based on the monitoring data to ensure that the final mutual correlation coefficient is more in line with the actual clinical situation, thereby improving the accuracy of diagnosis.
[0045] It should be noted that the mutual correlation coefficient in the present application is an indicator to measure the mutual correlation between adjacent nodes in the swallowing process. The larger the mutual correlation coefficient is, the greater the mutual correlation between adjacent nodes in the swallowing process is. The smaller the mutual correlation coefficient is, the smaller the mutual correlation between adjacent nodes in the swallowing process is. The mutual correlation coefficient can be used to identify the relationship between key nodes in the swallowing process, thereby discovering potential swallowing behavior disorders.
[0046] In some embodiments, reference Figure 2As shown, this figure is a schematic diagram of the process of determining the first fusion feature in some embodiments of the present application. In this embodiment, the swallowing-related data of all associated nodes in the perception-related information are sensed and fused through all the mutual correlation coefficients to obtain the first fusion feature of the swallowing of the dysphagia patient, which can be achieved by the following steps: In step 1031, swallowing associated data of each associated node is obtained from the perception associated information, and then the perception variance of each associated node is determined; In step 1032, the fusion weight coefficient of the corresponding associated node is determined by all the mutual correlation coefficients and the perceptual variance of each associated node; In step 1033, the swallowing-related data of each associated node are fused based on all fusion weight coefficients to obtain a first fusion feature of the swallowing of the dysphagia patient.
[0047] It should be noted that the swallowing-related data in the present application refers to the hyoid-laryngeal data of the hyoid bone and larynx when the patient swallows with associated movements. In the present embodiment, an array consisting of all hyoid node data segments and laryngeal node data segments in the above embodiments can be used as swallowing-related data. In specific implementation, the swallowing-related data of each associated node is obtained from the perceptual association information, and then the perceptual variance of each associated node is determined. The following method can be used, namely: select an associated node, and obtain the swallowing-related data of the associated node from the perceptual association information, wherein the swallowing-related data includes a hyoid node data segment and a laryngeal node data segment, and then the variances of the hyoid node data segment and the laryngeal node data segment are calculated respectively, and then the average value of the variance of the hyoid node data segment and the variance of the laryngeal node data segment is used as the perceptual variance of the associated node, and the above steps are repeated to obtain the perceptual variances of the remaining associated nodes.
[0048] It should also be noted that the perceptual variance in this application refers to the variance of swallowing-related data when the patient's hyoid bone and laryngeal associated perceptual movements are associated when swallowing. The perceptual variance can reflect the degree of discreteness of swallowing-related data, and then the stability and consistency of swallowing-related data can be evaluated.
[0049] In specific implementation, determining the fusion weight coefficient of the corresponding associated node through all the mutual correlation coefficients and the perceptual variance of each associated node can be implemented in the following manner, namely: selecting an associated node, obtaining the mutual correlation coefficient of the associated node and the perceptual variance of the associated node, and then adjusting the mutual correlation coefficient of the associated node through the perceptual variance of the associated node to obtain the fusion weight coefficient of the associated node, repeating the above steps to obtain the fusion weight coefficients of the remaining associated nodes. As a preferred embodiment, adjusting the mutual correlation coefficient of the associated node through the perceptual variance of the associated node to obtain the fusion weight coefficient of the associated node can be implemented in the following manner, namely: the natural exponent of the opposite number of the perceptual variance can be used as an adjustment coefficient, and then the product of the adjustment coefficient and the mutual correlation coefficient is used as the fusion weight coefficient of the associated node.
[0050] In specific implementation, the swallowing-related data of each associated node are fused based on all fusion weight coefficients to obtain the first fusion feature of the swallowing of patients with dysphagia, which can be achieved in the following manner, namely: first, the hyoid bone-laryngeal movement association features of the corresponding associated nodes are extracted from each swallowing-related data respectively, and the hyoid bone-laryngeal movement association features include: hyoid bone-laryngeal movement amplitude and timestamp; then, the hyoid bone-laryngeal movement amplitude in each hyoid bone-laryngeal movement association feature is weighted by all fusion weight coefficients to obtain the weighted feature of each associated node, and the weighted feature is a feature composed of the weighted value of the hyoid bone-laryngeal movement amplitude in the hyoid bone-laryngeal movement association feature and the timestamp; finally, all weighted features are arranged (in chronological order) as a vector according to the corresponding timestamps, and the arranged vector is used as the first fusion feature of the swallowing of patients with dysphagia.
[0051] It should be noted that the present application obtains a first fusion feature that reflects the comprehensive movement state during swallowing by fusing data of multiple related nodes (motion data of the hyoid bone and larynx). The first fusion feature can reflect the comprehensive movement characteristics of each related node of the hyoid bone and larynx during swallowing, and is a fusion indicator for evaluating swallowing function. The first fusion feature can provide timely feedback on the number of swallows and swallowing time nodes of patients with dysphagia within a specified time period.
[0052] In step 104, the acoustic information of the cough of the dysphagia patient is monitored based on the multi-source sensor group to obtain cough data, and the cough characteristics of the dysphagia patient are extracted from the cough data. Then, the second fusion feature of the choking cough of the dysphagia patient during swallowing is determined based on the cough characteristics, the first fusion feature and the correlation between the sensors in the multi-source sensor group.
[0053] In specific implementation, the acoustic information of the cough of the dysphagia patient is monitored based on the multi-source sensor group, and then the cough data is obtained in the following manner, namely: the acoustic information of the cough of the dysphagia patient is collected by the multi-source sensor group with a sampling frequency set, and the collected acoustic information is stored in a data storage unit, and the cough data collected by the multi-source sensor group within a specified time period is further obtained from the data storage unit, wherein the sampling frequency can range from 22kHz to 32kHz, and the specified time period refers to the time period corresponding to the past hour to the current moment. In other embodiments, the specified time period can also be set to other time periods, which are not limited here. In addition, the cough data is numerical data.
[0054] In some embodiments, extracting cough features of dysphagia patients from the cough data may be achieved by the following steps: Segmenting the cough data to obtain a plurality of cough data segments; Determine the time-frequency correlation feature of the corresponding cough data segment according to the time domain feature and frequency domain feature of each cough data segment; The cough characteristics of patients with dysphagia were determined by all time-frequency correlation features.
[0055] In specific implementation, the cough data is processed in segments to obtain multiple cough data segments, which can be implemented in the following manner, namely, the cough data can be detected by a signal peak detection method in the prior art to obtain multiple peak points, and then the cough data is processed in segments based on all the peak points to obtain multiple cough data segments; the time-frequency correlation features of the corresponding cough data segments are determined according to the time domain features and frequency domain features of each cough data segment, which can be implemented in the following manner, namely, a cough data segment is selected, the time domain features and frequency domain features of the cough data segment are obtained, and then an array (time domain features, frequency domain features) composed of the time domain features and the frequency domain features is used as the time-frequency correlation features of the cough data segment, and the above steps are repeated to obtain the time-frequency correlation features of the remaining cough data segments.
[0056] It should be noted that the time domain features in the present application include: kurtosis and standard deviation, and the frequency domain features include: spectrum peak and spectrum energy. In other embodiments, the time domain features and the frequency domain features also include other features, which are not limited here.
[0057] In specific implementation, determining the cough characteristics of patients with dysphagia through all time-frequency correlation features can be achieved in the following manner, namely: initializing a cough detection model, using all time-frequency correlation features as initialization parameters of the cough detection model, so that the cough detection model outputs the cough characteristics of patients with dysphagia; it should be noted that the cough detection model in the present application refers to a detection model for cough judgment by extracting key features from time-frequency information, and the cough detection model can be pre-trained with a large amount of cough data, which will not be elaborated here.
[0058] It should be noted that the cough feature in the present application is a vector reflecting the cough information of a patient with dysphagia, which includes the timestamp and cough amplitude of the patient with dysphagia when coughing. The cough feature can be used to clearly identify the number of times a patient with dysphagia coughs within a specified time period and the intensity of each cough.
[0059] In some embodiments, reference Figure 3 As shown, this figure is a schematic diagram of the process of determining the second fusion feature in some embodiments of the present application. In this embodiment, the second fusion feature of the choking cough of the dysphagia patient during swallowing can be determined by the association relationship between the cough feature, the first fusion feature and each sensor in the multi-source sensor group, which can be implemented by the following steps: In step 1041, the association relationship between the sensors in the multi-source sensor group is determined; In step 1042, correlation analysis is performed on the cough feature and the first fusion feature to obtain correlation features of choking and coughing during swallowing of the dysphagia patient; In step 1043, the associated features are associated and fused according to the associated relationship to obtain a second fused feature of choking during swallowing of the dysphagia patient.
[0060] In specific implementation, determining the correlation relationship between each sensor in the multi-source sensor group can be achieved in the following manner, namely: obtaining data collected by each sensor in the multi-source sensor group, using statistical methods (such as Pearson correlation coefficient) to perform correlation analysis on the data collected by each sensor, and obtaining the correlation coefficient between each sensor, and then using the correlation coefficient to reflect the correlation relationship between each sensor; it should be noted that the correlation relationship in this application refers to the mutual dependence of the data collected by each sensor, and this relationship can be used to quantify the degree of correlation between the data obtained by different sensors in time or characteristics.
[0061] In a specific implementation, the cough feature and the first fusion feature are correlated and analyzed to obtain the correlation feature of choking cough during swallowing in patients with dysphagia. The correlation analysis can be implemented in the following manner, namely: according to the timestamp information in the cough feature and the timestamp information in the first fusion feature, a timestamp is obtained from the cough feature, and the timestamp is used as the designated timestamp. The timestamp closest to the designated timestamp is obtained from the first fusion feature and used as the target timestamp. The difference between the target timestamp and the designated timestamp is compared with a preset threshold. When the difference is less than the preset threshold, the timestamp corresponding to the designated timestamp is compared. The node is used as the time node of choking and coughing when swallowing in patients with dysphagia, and the array composed of the cough amplitude corresponding to the specified timestamp and the hyoid bone-laryngeal movement amplitude corresponding to the target timestamp (cough amplitude, hyoid bone-laryngeal movement amplitude) is used as the choking and coughing associated feature of the time node corresponding to the timestamp, and the above steps are repeated to obtain the choking and coughing associated features of the time nodes corresponding to the remaining timestamps, and then the vector composed of all the choking and coughing associated features is used as the associated feature of choking and coughing when swallowing in patients with dysphagia. It should be noted that the preset threshold reflects the length of the delay time of choking and coughing in patients with dysphagia when swallowing. The preset threshold can be set according to a large amount of medical observation data and is not limited here.
[0062] In specific implementation, the association features are associated and fused by the association relationship to obtain the second fused feature of choking when swallowing for patients with dysphagia. This can be achieved in the following way, namely: the product of the association relationship and the association feature can be used as the second fused feature of choking when swallowing for patients with dysphagia. It should be noted that in the present application, the association features are associated and fused according to the degree of association between the sensors, which can reduce the system error caused by the sensitivity difference of different sensors, thereby improving the detection accuracy of choking when swallowing for patients with dysphagia. In addition, in the present application, the compensated choking association feature is used as the compensated choking association feature.
[0063] It should be noted that the second fusion feature in the present application refers to a fusion feature that is used to describe the choking and coughing situation that occurs during swallowing by fusing the hyoid-laryngeal movement characteristics and the coughing characteristics. The second fusion feature can be used to assess the risk of choking and coughing in patients with dysphagia during swallowing. The second fusion feature can provide timely feedback on the number of times and duration of choking and coughing in patients with dysphagia during swallowing, thereby providing more accurate treatment monitoring information for the treated patients.
[0064] In step 105, swallowing information of the dysphagia patient is determined according to the second fusion feature and a preset swallowing assessment standard.
[0065] In some embodiments, determining the swallowing information of the dysphagia patient according to the second fusion feature and the preset swallowing assessment standard can be achieved by the following steps: determining a plurality of swallowing feature parameters according to the second fusion feature and a preset swallowing assessment standard; The swallowing information of patients with dysphagia is determined through all swallowing characteristic parameters.
[0066] In specific implementation, determining multiple swallowing feature parameters according to the second fusion feature and the preset swallowing evaluation standard can be implemented in the following manner, namely: obtaining a choking and coughing associated threshold from the preset swallowing evaluation standard, obtaining all the compensating choking and coughing associated features in the second fusion feature, and comparing all the compensating choking and coughing associated features with the choking and coughing associated threshold; for each compensating choking and coughing associated feature, when the compensating choking and coughing associated feature is greater than the choking and coughing associated threshold, using the compensating choking and coughing associated feature as a swallowing feature, and obtaining the timestamp corresponding to the swallowing feature, and then using the array consisting of the swallowing feature and the timestamp corresponding to the swallowing feature as a swallowing feature parameter, thereby obtaining multiple swallowing feature parameters; determining the swallowing information of a dysphagia patient through all the swallowing feature parameters can be implemented in the following manner, namely: obtaining the total number of all swallowing feature parameters, using the total number of all swallowing feature parameters as the number of times the dysphagia patient chokes when swallowing, and using the timestamp in each swallowing feature parameter as the time node when the dysphagia patient chokes when swallowing.
[0067] It should be noted that the preset swallowing assessment standard in the present application is an assessment standard for detecting choking during swallowing in patients with dysphagia, and can be set based on a large amount of expert experience. In other embodiments, other methods can also be used for setting, which are not limited here. In addition, it should also be noted that the present application can monitor the swallowing information of patients with dysphagia and the information of choking during swallowing through a multi-source sensor group.
[0068] On the other hand, in some embodiments, the present application provides a swallowing monitoring system, the swallowing monitoring system comprising a swallowing monitoring control unit, referring to Figure 4 , which is a schematic diagram of exemplary software modules of a swallowing monitoring control unit according to some embodiments of the present application, the swallowing monitoring control unit 400 includes: an acquisition module 401, a processing module 402 and an execution module 403, which are described as follows: Acquisition module 401, in the present application, acquisition module 401 is mainly used to monitor the movement information of the hyoid bone and larynx of patients with dysphagia based on a multi-source sensor group, and then acquire hyoid bone movement data and laryngeal movement data; Processing module 402, in the present application, the processing module 402 is used to perform swallowing perception association on the hyoid bone movement data and the laryngeal movement data to obtain hyoid bone-larynx perception association information, and then determine the perception association degree of the hyoid bone and larynx associated movement of the dysphagia patient from the perception association information; The processing module 402 in the present application is also used to determine the interaction entropy of each associated node when the hyoid bone and larynx of the dysphagia patient are in associated motion according to the hyoid bone motion data and the laryngeal motion data, determine the mutual correlation coefficient of each associated node during the swallowing process of the dysphagia patient by all the interaction entropies and the perceptual correlation degree, and then perform perceptual fusion on the swallowing associated data of all associated nodes in the perceptual association information by all the mutual correlation coefficients to obtain the first fusion feature of the dysphagia patient when swallowing; The processing module 402 in the present application is also used to monitor the acoustic information of the cough of the dysphagia patient based on the multi-source sensor group, thereby obtaining cough data, extracting the cough feature of the dysphagia patient from the cough data, and then determining the second fusion feature of the choking cough of the dysphagia patient during swallowing based on the cough feature, the first fusion feature and the correlation relationship between each sensor in the multi-source sensor group; The execution module 403 in the present application is mainly used to determine the swallowing information of the swallowing disorder patient according to the second fusion feature and the preset swallowing assessment standard.
[0069] In addition, the present application also provides a computer device, which includes a memory and a processor, wherein the memory stores codes, and the processor is configured to obtain the codes and execute the above-mentioned control method of the swallowing monitoring system.
[0070] In some embodiments, reference Figure 5 , which is a schematic diagram of the structure of a computer device for implementing a control method of a swallowing monitoring system according to some embodiments of the present application. The control method of the swallowing monitoring system in the above embodiment can be Figure 5 The computer device 500 shown in the figure is implemented, and the computer device 500 includes at least one processor 501, a communication bus 502, a memory 503 and at least one communication interface 504.
[0071] The processor 501 may be a general-purpose central processing unit (CPU) or an application specific integrated circuit (ASIC).
[0072] The communication bus 502 may be used to transmit information between the above-mentioned components.
[0073] The memory 503 may be a read only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read only memory (EEPROM), a compact disc read only memory (CD ROM) or other optical disc storage, an optical disc storage (including a compressed optical disc, a laser disc, an optical disc, a digital versatile disc, a Blu-ray disc, etc.), a magnetic disk or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The memory 503 may exist independently and be connected to the processor 501 via the communication bus 502. The memory 503 may also be integrated with the processor 501.
[0074] The memory 503 is used to store the program code for executing the solution of the present application, and the execution is controlled by the processor 501. The processor 501 is used to execute the program code stored in the memory 503. The program code may include one or more software modules. The control method of the swallowing monitoring system in the above embodiment can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.
[0075] The communication interface 504 uses any transceiver or other device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.
[0076] In a specific implementation, as an embodiment, a computer device may include multiple processors, each of which may be a single-core (single CPU) processor or a multi-core (multi CPU) processor. The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (such as computer program instructions).
[0077] The above-mentioned computer device may be a general-purpose computer device or a special-purpose computer device. In a specific implementation, the computer device may be a desktop computer, a portable computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device or an embedded device. The embodiment of the present application does not limit the type of computer device.
[0078] In addition, the present application also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the control method of the swallowing monitoring system described above is implemented.
[0079] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.
[0080] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.
Claims
1. A method for controlling a swallowing monitoring system, characterized in that: The steps include: Based on a multi-source sensor group, the movement information of the hyoid bone and larynx of patients with dysphagia is monitored, and the hyoid bone movement data and laryngeal movement data are obtained; Performing swallowing perception association on the hyoid bone movement data and the laryngeal movement data to obtain hyoid bone-larynx perception association information, and then determining the perception association degree of the hyoid bone and larynx associated movement of the dysphagia patient from the perception association information; Determine the interaction entropy of each associated node when the hyoid bone and larynx of the dysphagia patient are in associated motion according to the hyoid bone movement data and the laryngeal movement data, determine the mutual correlation coefficient of each associated node during the swallowing process of the dysphagia patient according to all the interaction entropies and the perception correlation degree, and then perform perception fusion on the swallowing associated data of all associated nodes in the perception association information through all the mutual correlation coefficients to obtain the first fusion feature of the dysphagia patient when swallowing; Acoustic information of cough of dysphagia patients is monitored based on the multi-source sensor group, and cough data is obtained, cough features of dysphagia patients are extracted from the cough data, and second fusion features of choking cough of dysphagia patients during swallowing are determined based on the cough features, the first fusion features, and the correlation between the sensors in the multi-source sensor group; The swallowing information of the dysphagia patient is determined according to the second fusion feature and a preset swallowing assessment standard.
2. The method according to claim 1, characterized in that Performing swallowing perception association on the hyoid bone movement data and the laryngeal movement data to obtain hyoid bone-larynx perception association information specifically includes: Dynamically sensing the hyoid bone movement data to obtain a plurality of hyoid bone dynamic data segments; Dynamically sensing the laryngeal motion data to obtain a plurality of laryngeal dynamic data segments; Performing swallowing association on all hyoid bone dynamic data segments and all laryngeal dynamic data segments to obtain a plurality of associated hyoid bone-laryngeal dynamic data segments; The sensory association information of the hyoid bone and the larynx is determined through all the hyoid bone and the larynx dynamic data segments.
3. The method according to claim 1, characterized in that Determining the perception correlation degree of the hyoid bone and larynx of the dysphagia patient when the hyoid bone and larynx are in associated motion based on the perception correlation information specifically includes: Acquire all associated nodes of the hyoid bone and larynx of the dysphagia patient during associated movement from the perceptual associated information; Determining the hyoid-laryngeal motion ratio of each associated node based on the perceived associated information; The perceptual correlation of the hyoid and laryngeal movements in dysphagic patients was determined using all hyoid-laryngeal motion ratios.
4. The method according to claim 1, characterized in that Determining the mutual correlation coefficients of each associated node in the swallowing process of a dysphagia patient by using all the interaction entropies and the perceived correlation degree specifically includes: Determine the mutual correlation between adjacent associated nodes through all interaction entropies; The mutual correlation coefficients of the various correlation nodes in the swallowing process of the dysphagia patient are determined according to all the mutual correlations and the perceptual correlation.
5. The method according to claim 1, characterized in that The swallowing-related data of all associated nodes in the perception-related information are subjected to perception fusion through all the mutual correlation coefficients to obtain the first fusion feature of swallowing of the dysphagia patient, which specifically includes: Acquire swallowing-related data of each associated node from the perception-related information, and then determine the perception variance of each associated node; Determine the fusion weight coefficient of the corresponding associated node through all the mutual correlation coefficients and the perception variance of each associated node; The swallowing-related data of each associated node are fused based on all fusion weight coefficients to obtain the first fusion feature of the swallowing of the dysphagia patient.
6. The method according to claim 1, characterized in that Determining the second fusion feature of choking and coughing during swallowing of a dysphagia patient based on the cough feature, the first fusion feature, and the correlation between the sensors in the multi-source sensor group specifically includes: Determine the association relationship between each sensor in the multi-source sensor group; Performing correlation analysis on the cough feature and the first fusion feature to obtain correlation features of choking and coughing during swallowing in patients with dysphagia; The associated features are associated and fused according to the associated relationship to obtain a second fused feature of choking during swallowing of a dysphagia patient.
7. The method according to claim 1, characterized in that Determining the swallowing information of the dysphagia patient based on the second fusion feature and the preset swallowing assessment standard specifically includes: determining a plurality of swallowing feature parameters according to the second fusion feature and a preset swallowing assessment standard; The swallowing information of patients with dysphagia is determined through all swallowing characteristic parameters.
8. A swallowing monitoring system, comprising a swallowing monitoring control unit, characterized in that: The swallowing monitoring control unit comprises: An acquisition module, used for monitoring the movement information of the hyoid bone and larynx of a dysphagia patient based on a multi-source sensor group, and then acquiring hyoid bone movement data and laryngeal movement data; A processing module, configured to perform swallowing perception association on the hyoid bone movement data and the laryngeal movement data to obtain hyoid bone-larynx perception association information, and further determine the perception association degree of the hyoid bone and larynx of the dysphagia patient during the associated movement from the perception association information; The processing module is further used to determine the interaction entropy of each associated node when the hyoid bone and larynx of the dysphagia patient move in association according to the hyoid bone movement data and the laryngeal movement data, determine the mutual correlation coefficient of each associated node during the swallowing process of the dysphagia patient according to all the interaction entropies and the perception correlation degree, and then perform perception fusion on the swallowing association data of all associated nodes in the perception association information through all the mutual correlation coefficients to obtain a first fusion feature when the dysphagia patient swallows; The processing module is further used to monitor the acoustic information of the dysphagia patient's cough based on the multi-source sensor group, thereby obtaining cough data, extracting the cough feature of the dysphagia patient from the cough data, and then determining the second fusion feature of the choking cough of the dysphagia patient during swallowing based on the cough feature, the first fusion feature and the correlation relationship between the sensors in the multi-source sensor group; An execution module is used to determine swallowing information of a dysphagia patient according to the second fusion feature and a preset swallowing assessment standard.
9. A computer device, comprising a memory and a processor, wherein the memory stores a code, characterized in that: The processor is configured to obtain the code and execute the control method of the swallowing monitoring system according to any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the control method of the swallowing monitoring system according to any one of claims 1 to 7 is implemented.
Citation Information
Cited By
Maxillofacial muscle group function dynamic capturing method for patients with dysphagia
CN121512462A
Method for dynamic capture of maxillofacial muscle function for patients with swallowing disorders
CN121512462B
