Emergency language communication management method and system based on disaster environment analysis

Abstract

The application relates to the technical field of emergency medical management, and discloses an emergency language communication management method and system based on disaster environment analysis; the method comprises the following steps: acquiring pre-seismic and post-seismic detection data of each registered hearing aid in a historical earthquake event, seismic physical parameters and hearing aid equipment parameters, calculating a real failure state label and extracting a seismic influence feature vector, and constructing a hearing aid failure prediction model; acquiring real-time seismic physical parameters, inputting the model in combination with the equipment parameters of each hearing aid in the current region, and obtaining a failure label prediction value; setting a risk level threshold according to the distribution of historical data, counting the number of people wearing hearing aids in different risk levels, and matching and outputting an emergency language communication measure suggestion list suitable for a hospital emergency scene; the application realizes accurate evaluation of communication risks of people wearing hearing aids under an earthquake event and scientific allocation of emergency resources.

Description

Technical Field

[0001] This application relates to the field of emergency medical management technology, and in particular to emergency language communication management methods and systems based on disaster environment analysis. Background Technology

[0002] When natural disasters such as earthquakes occur, they are often accompanied by a large number of casualties. In a short period of time, a large number of injured people are rushed to the emergency department of the hospital for treatment. In such disaster scenarios, the efficiency of information communication is directly related to the treatment effect of the injured. People who wear hearing aids are a group that needs special attention. This is because the electromagnetic waves and vibrations generated by earthquakes may interfere with or damage hearing aids, causing wearers to have difficulty hearing medical advice, need to communicate repeatedly, or be unable to communicate at all when they are in the hospital. This not only affects the treatment efficiency of the wearers themselves, but may also cause anxiety due to communication barriers, further exacerbating the pressure on order at the emergency room.

[0003] Currently, hospital emergency departments rely primarily on the personal experience of management personnel for emergency response during earthquakes. The general approach involves assessing potential communication problems based on limited disaster information before the injured arrive, and then allocating translators, communication support staff, and written communication equipment based on experience. However, this reliance on personal experience lacks in-depth analysis of the correlation between earthquake physical parameters and hearing aid malfunctions. It fails to accurately predict the specific impact of different earthquake characteristics on hearing aid wearers, resulting in often blind and delayed emergency resource allocation. Responses are frequently reactive after communication problems occur, making it difficult to accurately deploy language communication resources before the injured arrive.

[0004] Therefore, how to quantitatively assess the communication risks of hearing aid wearers during earthquakes and accurately allocate emergency language communication resources based on the assessment results has become a pressing technical problem to be solved in this field. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, this application provides an emergency language communication management method and system based on disaster environment analysis, aiming to solve the problems in existing technologies where hospital emergency departments have difficulty predicting the communication risks of hearing aid wearers and cannot accurately allocate emergency language communication resources during earthquake disasters.

[0006] To achieve the above objectives, this application adopts the following technical solution: Firstly, this application provides an emergency language communication management method based on disaster environment analysis, including the following steps: S1. Obtain earthquake event data, hearing aid test data, hearing aid device parameters, and user data for the current region; S2. Based on the earthquake event data, hearing aid test data, hearing aid device parameters and wearer data of the current region, extract the earthquake impact feature vector and real fault status label corresponding to the valid hearing aid samples in the historical earthquake events of the current region. S3. Based on the earthquake impact feature vector and real fault state label corresponding to the valid hearing aid samples in the current region's historical earthquake events, construct a hearing aid fault prediction model; S4. Obtain real-time earthquake event data and hearing aid device parameter data for the current region, and combine them with the hearing aid fault prediction model to predict the fault status of the hearing aid under real-time earthquake events. S5. Based on the predicted real-time earthquake event hearing aid malfunction status, combined with the current regional hearing aid wearer data, assess the communication barriers of hearing aid wearers under the real-time earthquake event, and output a list of emergency language communication measures recommendations based on the assessment results.

[0007] According to the above technical solution, the steps for obtaining earthquake event data, hearing aid test data, hearing aid device parameters, and user data for the current region include: Step S11: Obtain earthquake event data for the current region. Specifically, extract relevant data for each historical earthquake event and real-time earthquake event in the current region from the data records of the earthquake monitoring department, including geomagnetic wave intensity, geomagnetic wave frequency, and earthquake vibration amplitude. Step S12: Obtain hearing aid testing data for the current region. Specifically, extract the total harmonic distortion (THD) value of each hearing aid in the most recent testing record before each historical earthquake event from the after-sales service records of local hearing aid retailers, hearing aid repair records of the Disabled Persons' Federation, and ENT visit records of hospitals. Also extract the THD value of the hearing aid when it was sent for repair within three months after the earthquake event. Step S13: Obtain the parameters of each registered hearing aid device and the total number of corresponding wearers from the population statistics department and the registration information of the Disabled Persons' Federation. The device parameters specifically include the center frequency of the hearing aid's sensitive frequency band, the half-width of the hearing aid's sensitive frequency band, and the hearing aid's shock resistance level.

[0008] According to the above technical solution, based on current regional earthquake event data, hearing aid testing data, hearing aid device parameters, and user data, the steps for extracting earthquake impact feature vectors and true fault status labels corresponding to valid hearing aid samples from historical earthquake events in the current region include: Step S21: From the hearing aid detection data of the current region, obtain the total harmonic distortion (THD) detection value of each hearing aid before and after each historical earthquake event, and screen out hearing aids with both pre-earthquake and post-earthquake detection records as valid samples; at the same time, since THD is the core indicator for measuring the degree of sound distortion of hearing aids, its value change can objectively reflect whether the hearing aid has noise and signal attenuation related faults due to earthquakes, subtract the pre-earthquake THD value from the post-earthquake THD value to obtain the true fault status label of each valid sample in each historical earthquake event; Step S22: For each valid hearing aid sample, obtain the geomagnetic wave frequency and earthquake vibration amplitude from the corresponding historical earthquake event data, and obtain the center frequency of the hearing aid sensitive frequency band, the half width of the hearing aid sensitive frequency band, and the earthquake resistance level of the hearing aid from the hearing aid device parameter data. In this way, the hearing aid sensitive frequency band matching degree and the hearing aid vibration impact coefficient are calculated and combined to form the earthquake influence feature vector of each valid hearing aid sample.

[0009] According to the above technical solution, the steps for constructing a hearing aid fault prediction model based on the earthquake impact feature vector and actual fault state label corresponding to valid hearing aid samples in historical earthquake events in the current region include: Step S31: Based on the earthquake impact feature vector and real fault state label of effective hearing aid samples in historical earthquake events, the gradient boosting regression tree algorithm is adopted. By initializing the prediction function and setting the various hyperparameters required for model iterative training, the basic framework and training rules of the individual hearing aid fault prediction model are determined, providing the initial state and constraints for subsequent iterative optimization of model parameters. Step S32: Based on the determined hearing aid fault prediction model algorithm, hyperparameters, and initial prediction function, multiple decision trees are gradually constructed through multiple rounds of iterative training. In each iteration, the negative gradient between the fault label prediction value output by the current hearing aid fault prediction model and the actual fault state label is calculated as the fitting target. The new decision tree is trained to learn the prediction bias correction law reflected by the negative gradient, and the new decision tree is added to the model in a learning rate weighted manner, so that the prediction ability of the hearing aid fault prediction model continuously improves with the increase of iteration rounds. Finally, all decision trees are weighted and combined to construct the hearing aid fault prediction model.

[0010] According to the above technical solution, the steps for obtaining real-time earthquake event data and hearing aid device parameter data in the current region, and combining them with a hearing aid fault prediction model, to predict the fault status of hearing aids under real-time earthquake events include: Step S41: Obtain the seismic physical parameters of the real-time earthquake event in the current region and the device parameters of each registered hearing aid according to the method of extracting the earthquake impact feature vector in S2, calculate the hearing aid sensitive frequency band matching degree and hearing aid vibration impact coefficient of each registered hearing aid in the current region, and combine them to form the earthquake impact feature vector of each registered hearing aid in the current region. Step S42: Input the earthquake impact feature vector of each registered hearing aid in the current region into the hearing aid fault prediction model. The model is calculated sequentially by 150 decision trees and finally outputs the fault label prediction value corresponding to each registered hearing aid. This value represents the degree of change in total harmonic distortion that each registered hearing aid may experience after a real-time earthquake event.

[0011] According to the above technical solution, based on the predicted hearing aid malfunction status under real-time earthquake events, combined with current hearing aid wearer data in the region, the steps for assessing the communication barriers of hearing aid wearers under real-time earthquake events, and matching and outputting a list of emergency language communication measures recommendations based on the assessment results, include: Step S51: Based on the distribution of the true fault status labels of valid hearing aid samples in historical data, set a risk level threshold, and combine the predicted fault label value of each registered hearing aid under real-time earthquake events with the current hearing aid wearer data in the region to count the number of hearing aid wearers at different risk levels. Step S52: Based on the number of hearing aid wearers at different risk levels, generate a list of emergency language communication measures recommendations for hospital emergency scenarios.

[0012] Secondly, this application also provides an emergency language communication management system based on disaster environment analysis, including: The data acquisition module is used to acquire earthquake event data, hearing aid test data, hearing aid device parameters, and user data for the current region. The feature extraction module is used to extract earthquake impact feature vectors and real fault status labels corresponding to valid hearing aid samples in historical earthquake events in the current region, based on earthquake event data, hearing aid test data, hearing aid device parameters and wearer data in the current region. The model building module is used to build a hearing aid failure prediction model based on the earthquake impact feature vector and the real failure state label corresponding to the valid hearing aid samples in the current region's historical earthquake events. The fault prediction module is used to acquire real-time earthquake event data and hearing aid device parameter data in the current area, and combine them with the hearing aid fault prediction model to predict the fault status of the hearing aid under real-time earthquake events. The measures output module is used to assess the communication barriers of hearing aid wearers under real-time earthquake events based on the predicted hearing aid malfunction status and the current hearing aid wearer data in the region, and to output a list of emergency language communication measures recommendations based on the assessment results.

[0013] Thirdly, this application provides an electronic device comprising a processor and a memory, wherein the memory stores a computer program that can be called by the processor, and the processor executes an emergency language communication management method based on disaster environment analysis by calling the computer program stored in the memory.

[0014] Fourthly, this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform an emergency language communication management method based on disaster environment analysis.

[0015] Compared with the prior art, this application has the following advantages and beneficial effects: This application addresses the communication support needs of hearing aid wearers, a special group affected by earthquakes. By constructing a mapping relationship between earthquake physical parameters and the degree of hearing aid malfunction, it transforms the traditional experience-based emergency communication resource allocation into a precise configuration based on individualized risk quantification. This enables hospital emergency departments to quickly grasp the overall risk status of hearing aid wearers after an earthquake and deploy corresponding language communication support measures in advance according to the number of people at different risk levels. It effectively solves the problems of difficulty in predicting the impact of earthquakes on hearing aid wearers and the blind and delayed allocation of emergency resources in existing technologies, significantly improving the language communication support capabilities and emergency response efficiency of hospital emergency departments for special groups in disaster scenarios. Attached Figure Description

[0016] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is an overall flowchart of the emergency language communication management method based on disaster environment analysis provided in the embodiments of this application; Figure 2 This is a data acquisition flowchart provided in an embodiment of this application; Figure 3 This is a flowchart of the feature vector extraction process provided in an embodiment of this application; Figure 4 This is a flowchart of the hearing aid fault prediction model construction provided in the embodiments of this application; Figure 5 This is a flowchart of real-time hearing aid fault status analysis provided in the embodiments of this application; Figure 6This is a flowchart illustrating the output of an emergency language communication measures suggestion list provided in this application embodiment. Detailed Implementation

[0017] The technical solution of this application will be further described in detail below with reference to the accompanying drawings and specific embodiments, so as to enable those skilled in the art to understand and implement it; it should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0018] Please see Figure 1 , Figure 1 This is an overall flowchart of the emergency language communication management method based on disaster environment analysis provided in the embodiments of this application, which specifically includes the following steps: S1. Obtain earthquake event data, hearing aid test data, hearing aid device parameters, and user data for the current region.

[0019] Please see Figure 2 , Figure 2 The complete technical process for data acquisition in the embodiments of this application is illustrated, and the specific steps are as follows: Step S11: Obtain earthquake event data for the current region. Specifically, extract relevant data for each historical earthquake event and real-time earthquake event in the current region from the data records of the earthquake monitoring department, including geomagnetic wave intensity, geomagnetic wave frequency, and earthquake vibration amplitude. Step S12: Obtain hearing aid testing data for the current region. Specifically, extract the total harmonic distortion (THD) value of each hearing aid in the most recent testing record before each historical earthquake event from the after-sales service records of local hearing aid retailers, hearing aid repair records of the Disabled Persons' Federation, and ENT visit records of hospitals. Also extract the THD value of the hearing aid when it was sent for repair within three months after the earthquake event. Step S13: Obtain the parameters of each registered hearing aid device and the total number of corresponding wearers from the population statistics department and the registration information of the Disabled Persons' Federation. The device parameters specifically include the center frequency of the hearing aid sensitive frequency band, the half-width of the hearing aid sensitive frequency band, and the shock resistance level of the hearing aid. The data acquisition process in this embodiment integrates earthquake event data from earthquake monitoring departments, test records from hearing aid retailers and medical institutions, and information on hearing aid wearers from population statistics departments. This fully covers multi-source data, including earthquake physical parameters, individual hearing aid test data, equipment parameters, and population characteristics, providing a comprehensive and traceable data foundation for the subsequent construction of individual hearing aid fault prediction models.

[0020] S2. Based on the earthquake event data, hearing aid test data, hearing aid device parameters, and wearer data of the current region, extract the earthquake impact feature vector and the actual fault status label corresponding to the valid hearing aid samples in the historical earthquake events of the current region.

[0021] Please see Figure 3 , Figure 3 This is a flowchart of the feature vector extraction process provided in an embodiment of this application. The specific steps are as follows: Step S21: From the hearing aid detection data of the current region, obtain the total harmonic distortion (THD) detection value of each hearing aid before and after each historical earthquake event, and screen out hearing aids with both pre-earthquake and post-earthquake detection records as valid samples; at the same time, since THD is the core indicator for measuring the degree of sound distortion of hearing aids, its value change can objectively reflect whether the hearing aid has noise and signal attenuation related faults due to earthquakes, subtract the pre-earthquake THD value from the post-earthquake THD value to obtain the true fault status label of each valid sample in each historical earthquake event; Step S22: For each valid hearing aid sample, obtain the geomagnetic wave frequency and earthquake amplitude from the corresponding historical earthquake event data, and obtain the center frequency, half-width, and seismic resistance level of the hearing aid's sensitive frequency band from the hearing aid device parameter data. This allows for the calculation of the hearing aid's sensitive frequency band matching degree and vibration impact coefficient, which are then combined to form the earthquake impact feature vector for each valid hearing aid sample. The specific calculation process is as follows: First, the internal circuitry of hearing aids has a sensitive range for electromagnetic waves of specific frequencies. When the frequency of earthquake magnetic waves falls within this range, it can easily cause resonance interference. Therefore, the proximity of the magnetic wave frequency to the center frequency of the hearing aid's sensitive frequency band is used to quantify this interference risk, i.e., the hearing aid's sensitive frequency band matching degree. When the difference between the magnetic wave frequency and the center frequency of the hearing aid's sensitive frequency band is less than half the width of the hearing aid's sensitive frequency band, it indicates that the magnetic wave frequency falls within the sensitive frequency band, and the hearing aid's sensitive frequency band matching degree is equal to one minus this difference divided by the half width of the hearing aid's sensitive frequency band. When the difference between the magnetic wave frequency and the center frequency of the hearing aid's sensitive frequency band is greater than or equal to the half width of the hearing aid's sensitive frequency band, it indicates that the magnetic wave frequency does not fall within the sensitive frequency band, and the hearing aid's sensitive frequency band matching degree is zero. Secondly, earthquake vibration amplitude is an indicator of the intensity of earthquake mechanical action, while the earthquake resistance level of hearing aids reflects the equipment's ability to withstand vibration. The ratio of the two can quantify the mechanical impact risk of earthquakes on hearing aids, i.e., the hearing aid vibration impact coefficient. By combining the hearing aid sensitive frequency band matching degree and the hearing aid vibration impact coefficient, the earthquake impact feature vector of each effective hearing aid sample is obtained. The feature vector extraction process in this embodiment combines earthquake physical parameters with individual hearing aid device parameters to quantify the actual impact of earthquakes on each hearing aid from two dimensions: frequency matching and vibration impact. This transforms the originally abstract earthquake effect into a calculable index with clear physical meaning. At the same time, based on the detection data of total harmonic distortion before and after the earthquake, the model calculates the real fault state label, enabling the model to learn the mapping relationship between earthquake parameters and the degree of fault of individual hearing aids. This provides a scientific and quantifiable data foundation for the subsequent construction of individualized fault prediction models.

[0022] S3. Based on the earthquake impact feature vector and the actual fault state label corresponding to the valid hearing aid samples in the current region's historical earthquake events, construct a hearing aid fault prediction model.

[0023] Please see Figure 4 , Figure 4 This is a flowchart of the hearing aid fault prediction model construction provided in this application embodiment. The specific steps are as follows: Step S31: Based on the earthquake impact feature vectors and true fault state labels of valid hearing aid samples from historical earthquake events, the gradient boosting regression tree algorithm is used to determine the basic framework and training rules of the individual hearing aid fault prediction model by initializing the prediction function and setting the hyperparameters required for iterative model training. This provides the initial state and constraints for subsequent iterative optimization of model parameters. The specific steps are as follows: First, considering that the hearing aid's sensitive frequency band matching degree reflects the proximity between the geomagnetic wave frequency and the sensitive frequency band of the hearing aid's internal circuitry, and the hearing aid's vibration shock coefficient reflects the relative relationship between the earthquake vibration amplitude and the hearing aid's earthquake resistance level, the influence of these two characteristics on the hearing aid's total harmonic distortion (THD) may have a complex nonlinear synergistic effect. For example, when the hearing aid's sensitive frequency band matching degree is high and the hearing aid's vibration shock coefficient is large, the resonance interference caused by the geomagnetic wave and the mechanical stress caused by the vibration may work together to cause nonlinear distortion in the hearing aid's internal components, making the increase in THD much greater than the sum of the effects of the two factors alone. This interaction is difficult to characterize with a linear model. At the same time, different effective hearing aid samples are independent of each other and there is no temporal dependency. Therefore, there is no need to use temporal models such as recurrent neural networks. Gradient boosting regression tree, as a tree-based ensemble learning method, can automatically capture the nonlinear interaction between features. It is not sensitive to feature scale and can output feature importance after training. It is easy to understand the influence of hearing aid sensitive frequency band matching degree and hearing aid vibration impact coefficient on the total harmonic distortion change. It has good interpretability. Therefore, the gradient boosting regression tree algorithm is used to construct a hearing aid fault prediction model. Secondly, considering that the model needs to strike a balance between fitting ability and computational efficiency, the number of weak learners, i.e., the number of decision trees, is set to 150. Since a smaller learning rate can effectively prevent the model from overfitting, but at the same time requires more weak learners to achieve the same fitting effect, the learning rate is set to 0.08. To avoid the model becoming too complex and learning noise in the training data, the growth depth of each decision tree needs to be limited. Considering that the influencing features only have two dimensions, excessively deep trees are prone to overfitting, so the maximum depth of the decision tree is set to 4. To enhance the model's generalization ability, 70% of the samples are randomly selected in each iteration to train the current decision tree, i.e., the subsampling ratio is set to 0.7. Since the hearing aid fault prediction model predicts discrete values ​​such as the net change in total harmonic distortion, the least squares loss function suitable for regression tasks is adopted as the loss function. Next, given that a reasonable global estimate of the fault state is the average level of the true fault state labels of all valid hearing aid samples without considering the hearing aid sensitive band matching degree and hearing aid vibration shock coefficient of any specific valid hearing aid samples, the initial prediction function is set as a constant function, specifically the arithmetic mean of the true fault state labels of all valid hearing aid samples: ; In the formula, This is the initial prediction function for the hearing aid failure prediction model, which is the basic estimate of the failure label prediction value for a valid hearing aid sample without considering the seismic influence feature vector of the specific valid hearing aid sample. For the first The actual fault status label corresponding to each valid hearing aid sample; The total number of valid hearing aid samples. ; The initialization formula of the hearing aid fault prediction model in this embodiment provides a reasonable global learning starting point for the model of fault state. This starting point represents the overall average level of the true fault state labels of all effective hearing aid samples. This means that all subsequent learning based on earthquake impact feature vectors is clearly aimed at explaining and predicting the deviation of the predicted fault label value of a specific effective hearing aid sample from this historical baseline, thereby ensuring that the model focuses on the changes in fault risk caused by the differences in earthquake physical parameters and individual hearing aid parameters. Taking 2000 valid hearing aid samples collected from a certain region as an example, the true fault status labels of these valid hearing aid samples include different values ​​such as -2%, 3%, and 5%. After calculating the average value of all valid hearing aid samples, the initial prediction function value is 1.5%. Thus, before learning any relationship between earthquake impact feature vectors and true fault status labels, the hearing aid fault prediction model outputs 1.5% as the initial predicted fault label value for any input earthquake impact feature vector. In the subsequent 150 iterations of training, the hearing aid fault prediction model will analyze the correlation between the two dimensions of hearing aid sensitive frequency band matching degree and hearing aid vibration impact coefficient in the earthquake impact feature vector and the true fault status label. It will gradually learn that when the hearing aid sensitive frequency band matching degree is higher than the average level, the fault label prediction value should be adjusted upward; when the hearing aid vibration impact coefficient exceeds a certain threshold, the fault label prediction value should increase significantly, thereby achieving accurate prediction of fault label prediction values ​​under different combinations of earthquake impact feature vectors. Step S32: Based on the determined hearing aid fault prediction model algorithm, hyperparameters, and initial prediction function, multiple decision trees are gradually constructed through multiple rounds of iterative training. In each iteration, the negative gradient between the predicted fault label value output by the current hearing aid fault prediction model and the actual fault state label is calculated as the fitting target. The new decision tree is trained to learn the prediction bias correction law reflected by this negative gradient, and the new decision tree is added to the model in a learning rate weighted manner, so that the prediction ability of the hearing aid fault prediction model continuously improves with the increase of iteration rounds. Finally, all decision trees are weighted and combined to construct the hearing aid fault prediction model. The specific steps are as follows: First, based on the preset maximum number of iterations of 150, the iterative training process of the hearing aid fault prediction model is initiated. Let the current iteration round be... ,initialization ; In the In each iteration, the following steps are performed: A1. For each valid hearing aid sample, calculate the residual between the fault label prediction value output by the hearing aid fault prediction model in the current iteration and the actual fault state label of the valid hearing aid sample. Use this residual as the negative gradient to be fitted in this iteration. In the formula, For the first The number of valid hearing aid samples in the first The negative gradient vector of each iteration represents the deviation between the predicted fault label value output by the current hearing aid fault prediction model and the actual fault state label. For the front The hearing aid failure prediction model obtained after the first iteration is used for the second iteration. Fault label prediction values ​​output by a valid hearing aid sample; For the first Earthquake impact feature vector of a valid hearing aid sample; Taking the 7th valid hearing aid sample recorded in a certain region as an example, the true fault status label of this valid hearing aid sample is 5%, which means that the total harmonic distortion of this hearing aid increased by five percentage points after the earthquake event. When the 5th iteration is completed, the previous 4 iterations have been completed. The current hearing aid fault prediction model predicts the fault label value of this valid hearing aid sample as 3%. The negative gradient of the 5th iteration is 2%. The meaning of this negative gradient is that the current hearing aid fault prediction model predicts the fault label value of this valid hearing aid sample as 2 percentage points lower. The decision tree to be built in the 5th iteration needs to correct this prediction deviation so that the fault label prediction value output by the model is closer to the true fault status label. A2. Using the earthquake impact feature vector of each valid hearing aid sample as input and the negative gradient value calculated in A1 as the fitting target, train a decision tree. The training process of this decision tree aims to group valid hearing aid samples with similar combinations of earthquake impact feature vectors into the same leaf node, and to make the negative gradient values ​​of valid hearing aid samples within the same leaf node as close as possible, thereby learning the correction rules for the deviation of fault label prediction values ​​under different combinations of earthquake impact feature vectors. The specific training process is as follows: The decision tree is built starting from the root node, which contains all valid hearing aid samples for training. For the node currently being built, a dimension is randomly selected from the two dimensions of the earthquake impact feature vector as a segmentation feature. The values ​​of all valid hearing aid samples contained in the current node on the selected feature dimension are obtained, the minimum and maximum values ​​of the feature are determined, and a segmentation value is randomly generated within the interval formed by the minimum and maximum values. Next, all valid hearing aid samples in the current node that satisfy the condition that the feature value is less than the segmentation value are assigned to the left child node, and all valid hearing aid samples that satisfy the condition that the feature value is greater than or equal to the segmentation value are assigned to the right child node. Then, the above segmentation process is recursively performed on the left and right child nodes until any one of the following three conditions is met: the current node reaches the preset maximum depth 4, the current node contains only one valid hearing aid sample, or all valid hearing aid samples in the current node have the same value in both dimensions of the earthquake influence feature vector. At this point, further segmentation of the node is stopped. Once all nodes have stopped splitting, the decision tree is complete. The output value of each leaf node is the average of the negative gradient values ​​of all valid hearing aid samples contained in that node. ; In the formula, For the first The output function of each decision tree, for any input earthquake influence feature vector. Starting from the root node, the algorithm traverses downwards based on the segmentation features and values ​​of each node, eventually landing on a leaf node. The output value of that leaf node is the result. The output; The set of valid hearing aid samples falling into this leaf node is a group of valid hearing aid samples with similar combinations of earthquake influence feature vectors. This represents the number of valid hearing aid samples in this leaf node. Taking the fifth decision tree constructed in the fifth iteration as an example, the root node contains all 2000 valid hearing aid samples. At the root node, the hearing aid sensitive frequency band matching degree from the earthquake influence feature vector is randomly selected as the segmentation feature. The hearing aid sensitive frequency band matching degree of the 2000 valid hearing aid samples ranges from 0.2 to 0.9. A segmentation value of 0.5 is randomly generated within the range of 0.2 to 0.9. 800 valid hearing aid samples with a sensitive frequency band matching degree less than 0.5 are assigned to the left child node, and 1200 valid hearing aid samples with a matching degree greater than or equal to 0.5 are assigned to the right child node. Then, the left and right child nodes are recursively segmented until all nodes meet the termination condition, ultimately generating a complete decision tree. When the seventh valid hearing aid… The hearing aid sample with a sensitive frequency band matching degree of 0.45, which is less than 0.5, is assigned to a path containing the left child node and eventually falls into a leaf node containing 50 valid hearing aid samples. The negative gradient values ​​of these 50 valid hearing aid samples in the 5th iteration are 2%, 1%, 1.5%, etc. The output value of this leaf node is the average of these 50 negative gradient values. Assuming that the calculated output value is 1.8%, for all valid hearing aid samples assigned to this leaf node, that is, those valid hearing aid samples with a sensitive frequency band matching degree of less than 0.5 in the earthquake impact feature vector and also segmented into the same path in another dimension, the correction given by the 5th decision tree is an increase of 1.8% in the fault label prediction value. A3, the first Each decision tree is added to the current model using a learning rate weighted approach, resulting in an updated hearing aid malfunction prediction model: In the formula, For any input earthquake impact feature vector, the updated hearing aid fault prediction function outputs the updated fault label prediction value. For the front The hearing aid fault prediction function obtained after round of iteration; The learning rate; Taking the 5th iteration as an example, when m=5, the 5th decision tree is added to the model obtained from the first 4 iterations. For the 7th valid hearing aid sample, the predicted fault label value before the 5th iteration is 3%, and the output of the 5th decision tree for this valid hearing aid sample is 1.8%. Therefore, the updated predicted fault label value is 3% + 0.08 × 1.8% = 3.144%. The updated predicted fault label value is adjusted from 3% to 3.144%, which is closer to the actual 5% of the valid hearing aid sample. This shows that this iteration has a positive correction effect on the prediction of the valid hearing aid sample, making the predicted value closer to the actual fault state label. Then, increment the current iteration number by one and repeat steps A1 to A3 to build the 2nd, 3rd, and up to the 150th decision tree in sequence. In each iteration, the newly built decision tree attempts to fit the residual between the fault label prediction value output by the hearing aid fault prediction model in the previous iteration and the actual fault state label. As the number of iterations increases, the fault label prediction value output by the hearing aid fault prediction model gradually approaches the actual fault state label, and the prediction error of the model on the training set gradually decreases. Finally, the 150 decision trees obtained from the training are saved together to construct the hearing aid failure prediction model. For each registered hearing aid in any current earthquake event, its earthquake impact feature vector is extracted, input into the hearing aid failure prediction model, and then processed by the 150 decision trees in sequence. Finally, the corresponding fault label prediction value is output. This value represents the degree of change in total harmonic distortion that the hearing aid failure prediction model predicts may occur after the current earthquake event. The hearing aid fault prediction model construction process in this embodiment incorporates the earthquake impact feature vector of each valid hearing aid sample and the corresponding real fault state label into a unified supervised learning framework. The final output fault label prediction value provides a scientific and quantifiable data basis for subsequent assessment of the communication impairment of registered hearing aid wearers under the current earthquake event. This enables the allocation of emergency language communication resources to be based on an accurate grasp of the degree of individual hearing aid faults, effectively improving the efficiency of emergency language support and the scientific nature of resource allocation for hearing aid wearers in disaster scenarios.

[0024] S4. Obtain real-time earthquake event data and hearing aid device parameter data for the current region, and combine them with the hearing aid fault prediction model to predict the fault status of the hearing aid under real-time earthquake events.

[0025] Please see Figure 5 , Figure 5 This is a flowchart of real-time hearing aid fault status analysis provided in an embodiment of this application. The specific steps are as follows: Step S41: Obtain the seismic physical parameters of the real-time earthquake event in the current region and the device parameters of each registered hearing aid according to the method of extracting the earthquake impact feature vector in S2, calculate the hearing aid sensitive frequency band matching degree and hearing aid vibration impact coefficient of each registered hearing aid in the current region, and combine them to form the earthquake impact feature vector of each registered hearing aid in the current region. Step S42: Input the earthquake impact feature vector of each registered hearing aid in the current region into the hearing aid fault prediction model. The model is calculated sequentially by 150 decision trees and finally outputs the fault label prediction value corresponding to each registered hearing aid. This value represents the degree of change in total harmonic distortion that each registered hearing aid may experience after a real-time earthquake event. The real-time hearing aid fault status analysis process in this embodiment enables rapid prediction of the fault status of each registered hearing aid after a real-time earthquake event. It transforms abstract earthquake physical parameters into specific individual hearing aid fault label prediction values, providing direct data support for subsequent assessment of communication barriers among hearing aid wearers. This allows the allocation of emergency language communication resources to be based on a precise understanding of the individual fault degree of each hearing aid, effectively improving the pre-disaster prediction capability.

[0026] S5. Based on the predicted real-time earthquake event hearing aid malfunction status, combined with the current regional hearing aid wearer data, assess the communication barriers of hearing aid wearers under the real-time earthquake event, and output a list of emergency language communication measures recommendations based on the assessment results.

[0027] Please see Figure 6 , Figure 6 This is a flowchart illustrating the output of an emergency language communication measures suggestion list provided in this application embodiment. The specific steps are as follows: Step S51: Based on the distribution of the true fault status labels of valid hearing aid samples in historical data, set a risk level threshold, and combine the predicted fault label values ​​of each registered hearing aid under real-time earthquake events with the current hearing aid wearer data in the region to count the number of hearing aid wearers at different risk levels; the specific steps are as follows: First, the true fault status labels of all valid hearing aid samples from all historical earthquake events are arranged in ascending order, and the 30th percentile is taken as the low-risk threshold, and the 70th percentile is taken as the medium-risk threshold. Secondly, based on the predicted fault label value of each registered hearing aid in the current region, the predicted fault label value of each registered hearing aid is compared with the low-risk threshold and the medium-risk threshold to determine the risk level: if the predicted fault label value is less than the low-risk threshold, the wearer of the hearing aid is determined to be at the normal risk level; if the predicted fault label value is greater than or equal to the low-risk threshold and less than the medium-risk threshold, the wearer of the hearing aid is determined to be at the attention risk level; if the predicted fault label value is greater than or equal to the medium-risk threshold, the wearer of the hearing aid is determined to be at the dangerous risk level. Finally, based on the current data on hearing aid wearers in the region, the number of hearing aid wearers in the region at the levels of general risk, concern risk, and dangerous risk is calculated and denoted as follows: and Simultaneously calculate the proportion of people with normal risk, high-risk, and dangerous risk levels out of the total number of hearing aid wearers. and ; Step S52: Based on the number of hearing aid wearers at different risk levels, generate a list of recommended emergency language communication measures for hospital emergency scenarios; the specific steps are as follows: First, based on the proportion of people at regular risk Percentage of people paying attention to risks Percentage of people at risk To determine the risk characteristics of this earthquake event: like If the percentage is greater than 30%, the earthquake event is judged to be a dangerous-dominant risk type; if Less than or equal to 30% and If the risk rate is greater than 40%, the earthquake event is classified as a risk of primary concern; if... Less than or equal to 30%, Less than or equal to 40% and If the risk rate is less than 30%, the earthquake event is judged to have an equilibrium risk characteristic; if... If the probability is greater than 70%, the earthquake event is determined to be a conventional-dominant risk event; otherwise, it is determined to be a mixed-type risk event. Secondly, based on different risk characteristics, a corresponding list of emergency language communication measures is provided: If the risk is determined to be of a dangerous nature, the emergency language communication measures recommended list includes: increasing the number of sign language interpreters, equipping writing boards and electronic text communication devices, setting up communication assistance points in various areas of the emergency department, and establishing a rapid distribution channel for emergency contact cards; If the risk is identified as a concern-driven risk, the emergency language communication measures recommended list includes: equipping triage desks, waiting areas, and consultation rooms with written medical orders and large-font reminder boards; increasing the number of communication support staff; and setting up slow-speed communication windows. If the risk is determined to be of a conventional, dominant type, the emergency language communication measures recommendation list includes: maintaining the existing communication process and providing routine triage services. If the risk is determined to be of a balanced type, the emergency language communication measures recommended list includes: providing written medical orders and large-font notice boards, increasing communication support personnel, setting up a slow-speed communication window, and preparing sign language interpreters; If the risk is determined to be of a mixed nature, the emergency language communication measures recommendation list includes a comprehensive solution that incorporates all of the above measures. Finally, all the matched emergency language communication measures suggestions are compiled into a complete list of emergency language communication measures suggestions, which is then output to the relevant department managers to ensure that all resources are accurately allocated. The emergency language communication measures suggestion list output process in this embodiment realizes a quantitative assessment of the overall risk characteristics of hearing aid wearers under real-time earthquake events. This enables limited communication support resources to be accurately allocated based on the actual risk characteristics of the earthquake event, effectively improving the resource allocation efficiency and accuracy of emergency language communication in hospital emergency departments under disaster scenarios.

[0028] This application provides an emergency language communication management system based on disaster environment analysis, including: The data acquisition module is used to acquire earthquake event data, hearing aid test data, hearing aid device parameters, and user data for the current region. The feature extraction module is used to extract earthquake impact feature vectors and real fault status labels corresponding to valid hearing aid samples in historical earthquake events in the current region, based on earthquake event data, hearing aid test data, hearing aid device parameters and wearer data in the current region. The model building module is used to build a hearing aid failure prediction model based on the earthquake impact feature vector and the real failure state label corresponding to the valid hearing aid samples in the current region's historical earthquake events. The fault prediction module is used to acquire real-time earthquake event data and hearing aid device parameter data in the current area, and combine them with the hearing aid fault prediction model to predict the fault status of the hearing aid under real-time earthquake events. The measures output module is used to assess the communication barriers of hearing aid wearers under real-time earthquake events based on the predicted hearing aid malfunction status and the current hearing aid wearer data in the region, and to output a list of emergency language communication measures recommendations based on the assessment results.

[0029] This application provides an electronic device, including a memory, a processor, and a communication bus; the memory and the processor are connected via the communication bus; the memory stores an emergency language communication management method based on disaster environment analysis that can be loaded and executed by the processor as provided in the above embodiments.

[0030] The memory can be used to store instructions, programs, code, code sets, or instruction sets; the memory may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function, and instructions for implementing the emergency language communication management method based on disaster environment analysis provided in the above embodiments, etc.; the data storage area may store data involved in the emergency language communication management method based on disaster environment analysis provided in the above embodiments, etc.

[0031] The processor may include one or more processing cores; the processor executes or runs instructions, programs, code sets or instruction sets stored in memory, calls data stored in memory, and performs various functions and processes data in this application; the processor may be at least one of the following: Application-Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), Central Processing Unit (CPU), controller, microcontroller and microprocessor; it is understood that for different devices, the electronic device used to implement the above processor functions may also be other, and the embodiments of this application do not specifically limit it.

[0032] A communication bus may include a path for transmitting information between the aforementioned components; the communication bus may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc.; the communication bus may be divided into address bus, data bus, control bus, etc.

[0033] This application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed as described in the above embodiments: an emergency language communication management method based on disaster environment analysis.

[0034] In this embodiment, a computer-readable storage medium can be a tangible device that holds and stores instructions used by an instruction execution device; a computer-readable storage medium can be, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination thereof; specifically, a computer-readable storage medium can be a portable computer disk, a hard disk, a USB flash drive, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital multifunction disc (DVD), a memory stick, a floppy disk, an optical disk, a magnetic disk, a mechanical encoding device, or any combination thereof.

[0035] The terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0036] The above description is merely a preferred embodiment of this application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of this application is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the foregoing application concept; for example, technical solutions formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions applied in this application.

Claims

1. An emergency language communication management method based on disaster environment analysis, characterized in that, Includes the following steps: S1. Obtain earthquake event data, hearing aid test data, hearing aid device parameters, and user data for the current region; S2. Based on the earthquake event data, hearing aid test data, hearing aid device parameters and wearer data of the current region, extract the earthquake impact feature vector and real fault status label corresponding to the valid hearing aid samples in the historical earthquake events of the current region. S3. Based on the earthquake impact feature vector and real fault state label corresponding to the valid hearing aid samples in the current region's historical earthquake events, construct a hearing aid fault prediction model; S4. Obtain real-time earthquake event data and hearing aid device parameter data for the current region, and combine them with the hearing aid fault prediction model to predict the fault status of the hearing aid under real-time earthquake events. S5. Based on the predicted real-time earthquake event hearing aid malfunction status, combined with the current regional hearing aid wearer data, assess the communication barriers of hearing aid wearers under the real-time earthquake event, and output a list of emergency language communication measures recommendations based on the assessment results.

2. The emergency language communication management method based on disaster environment analysis according to claim 1, characterized in that, The acquisition of current regional earthquake event data, hearing aid test data, hearing aid device parameters, and user data includes the following specific content: To obtain earthquake event data for the current region, specifically, data related to each historical and real-time earthquake event in the current region is extracted from the data records of earthquake monitoring departments, including geomagnetic wave intensity, geomagnetic wave frequency, and earthquake vibration amplitude. To obtain hearing aid testing data for the current region, specifically, extract the total harmonic distortion (THD) value of each hearing aid in the most recent test record before each historical earthquake event from the after-sales service records of local hearing aid retailers, hearing aid repair records from the Disabled Persons' Federation, and ENT visit records of hospitals. Also extract the THD value of the hearing aid when it was sent for repair within three months after the earthquake event. Obtain the parameters of each registered hearing aid device and the total number of corresponding wearers in the current region from the population statistics department and the registration information of the Disabled Persons' Federation. The device parameters specifically include the center frequency of the hearing aid's sensitive frequency band, the half-width of the hearing aid's sensitive frequency band, and the hearing aid's shock resistance level.

3. The emergency language communication management method based on disaster environment analysis according to claim 2, characterized in that, Based on current regional earthquake event data, hearing aid testing data, hearing aid device parameters, and user data, the process extracts earthquake impact feature vectors and actual fault status labels corresponding to valid hearing aid samples from historical earthquake events in the current region. This includes the following specific content: From the current hearing aid testing data, the total harmonic distortion (THD) values ​​of each hearing aid before and after each historical earthquake event are obtained. Hearing aids with both pre-earthquake and post-earthquake testing records are selected as valid samples. At the same time, since THD is the core indicator for measuring the degree of sound distortion of hearing aids, its value change can objectively reflect whether the hearing aid has noise and signal attenuation-related faults due to earthquakes. Therefore, the THD value before the earthquake is subtracted from the post-earthquake THD value to obtain the true fault status label of each valid sample in each historical earthquake event. For each valid hearing aid sample, the geomagnetic wave frequency and earthquake amplitude are obtained from the corresponding historical earthquake event data, and the center frequency, half-width, and seismic resistance level of the hearing aid sensitive frequency band are obtained from the hearing aid device parameter data. Thus, the hearing aid sensitive frequency band matching degree and the hearing aid vibration impact coefficient are calculated and combined to form the earthquake impact feature vector of each valid hearing aid sample.

4. The emergency language communication management method based on disaster environment analysis according to claim 3, characterized in that, The hearing aid fault prediction model is constructed based on the earthquake impact feature vector and the actual fault state label corresponding to valid hearing aid samples in the current region's historical earthquake events, including the following specific contents: Based on the earthquake impact feature vector and real fault state label of valid hearing aid samples in historical earthquake events, the gradient boosting regression tree algorithm is adopted. By initializing the prediction function and setting the various hyperparameters required for model iterative training, the basic framework and training rules of the individual hearing aid fault prediction model are determined, providing the initial state and constraints for subsequent iterative optimization of model parameters. Based on a defined hearing aid fault prediction model algorithm, hyperparameters, and initial prediction function, multiple decision trees are gradually constructed through multiple rounds of iterative training. In each iteration, the negative gradient between the predicted fault label value output by the current hearing aid fault prediction model and the actual fault state label is calculated as the fitting target. The new decision tree is trained to learn the prediction bias correction law reflected by the negative gradient, and the new decision tree is added to the model in a learning rate weighted manner. This allows the predictive ability of the hearing aid fault prediction model to continuously improve with the increase of iteration rounds. Finally, all decision trees are weighted and combined to construct the hearing aid fault prediction model.

5. The emergency language communication management method based on disaster environment analysis according to claim 4, characterized in that, The process of acquiring real-time earthquake event data and hearing aid device parameter data for the current region, and combining this with a hearing aid fault prediction model to predict the fault status of the hearing aid under real-time earthquake events, includes the following specific content: The seismic physical parameters of the real-time earthquake event in the current region and the device parameters of each registered hearing aid are obtained according to the method of extracting the earthquake impact feature vector in S2. The hearing aid sensitive frequency band matching degree and hearing aid vibration shock coefficient of each registered hearing aid in the current region are calculated and combined to form the earthquake impact feature vector of each registered hearing aid in the current region. The earthquake impact feature vector of each registered hearing aid in the current region is input into the hearing aid fault prediction model. The model is calculated sequentially through 150 decision trees and finally outputs the fault label prediction value corresponding to each registered hearing aid. This value represents the degree of change in total harmonic distortion that each registered hearing aid may experience after a real-time earthquake event.

6. The emergency language communication management method based on disaster environment analysis according to claim 5, characterized in that, The method involves assessing the communication barriers faced by hearing aid wearers during real-time earthquakes based on predicted hearing aid malfunction status and local hearing aid user data. Based on the assessment results, a list of recommended emergency language communication measures is generated, including the following specific details: Based on the distribution of real fault status labels of valid hearing aid samples in historical data, risk level thresholds are set. Combined with the predicted fault label values ​​of each registered hearing aid under real-time earthquake events and the current hearing aid wearer data in the region, the number of hearing aid wearers at different risk levels is counted. Based on the number of hearing aid wearers at different risk levels, a list of recommended emergency language communication measures for hospital emergency scenarios is generated.

7. An emergency language communication management system based on disaster environment analysis, implemented based on the emergency language communication management method based on disaster environment analysis as described in any one of claims 1-6, characterized in that, The system includes: The data acquisition module is used to acquire earthquake event data, hearing aid test data, hearing aid device parameters, and user data for the current region. The feature extraction module is used to extract earthquake impact feature vectors and real fault status labels corresponding to valid hearing aid samples in historical earthquake events in the current region, based on earthquake event data, hearing aid test data, hearing aid device parameters and wearer data in the current region. The model building module is used to build a hearing aid failure prediction model based on the earthquake impact feature vector and the real failure state label corresponding to the valid hearing aid samples in the current region's historical earthquake events. The fault prediction module is used to acquire real-time earthquake event data and hearing aid device parameter data in the current area, and combine them with the hearing aid fault prediction model to predict the fault status of the hearing aid under real-time earthquake events. The measures output module is used to assess the communication barriers of hearing aid wearers under real-time earthquake events based on the predicted hearing aid malfunction status and the current hearing aid wearer data in the region, and to output a list of emergency language communication measures recommendations based on the assessment results.

8. An electronic device comprising a processor and a memory, characterized in that, The memory stores a computer program that can be called by a processor; the processor executes the emergency language communication management method based on disaster environment analysis as described in any one of claims 1-6 by calling the computer program stored in the memory.

9. A computer-readable storage medium storing instructions, characterized in that, When the instructions are executed on a computer, the computer performs the emergency language communication management method based on disaster environment analysis as described in any one of claims 1-6.

Images