A kind of early warning method and related equipment for acute altitude sickness

Abstract

The present invention provides an early warning method for acute plateau reaction and related equipment, which are applied to the field of data processing technology. The present application obtains target user information and a training data set; processes the target user information to generate attribute information of the target user; pre-processes the training data set to generate a training data set with target feature data, wherein the target feature data is used to characterize physiological factors affecting the user's acute plateau reaction; obtains an initial acute plateau reaction early warning model that matches the target feature data; trains the initial acute plateau reaction early warning model based on the training data set with the target feature data to generate a target acute plateau reaction early warning model; processes the total triiodothyronine, free triiodothyronine and free thyroxine of the target user based on the target acute plateau reaction early warning model to generate early warning information.

Description

Technical Field

[0001] The invention relates to the technical field of data processing, and in particular to an early warning method for acute altitude sickness and related equipment. Background Art

[0002] According to statistics, there are approximately 81.6 million people living in high altitude areas above 2,500 meters in the world, and more than 100 million people who have lived in low altitude areas for a long time enter high altitude areas for work or tourism every year. Rapidly entering high altitude areas and poor adaptation may lead to AMS, which often occurs within the first 3 days of entering the plateau area. Headache is the most important symptom of AMS, and it is usually accompanied by fatigue, dizziness, poor appetite, nausea and sleep disorders. The above symptoms usually resolve on their own in 3-10 days. However, it is worth noting that nearly 1% of the AMS population may progress to life-threatening high-altitude pulmonary edema and high-altitude cerebral edema.

[0003] With the development of tourism, facility construction, military and other industries, people's demand for entering the plateau has greatly increased, and it is of great significance to effectively predict the incidence of AMS. The occurrence of AMS is mainly related to the altitude and ascent speed of the area visited. Studies have also found that age, gender, altitude of the original residence and obesity are also important factors affecting the incidence of AMS. In addition, AMS is also closely related to personal susceptibility. The pathogenesis of AMS has not been fully elucidated. Previous studies have found that sympathetic nerve overactivation, endothelial dysfunction and inflammatory response may be involved in the pathogenesis of acute mountain sickness. However, many studies have only compared the symptoms or incidence of AMS in various situations. At present, domestic and foreign research still cannot meet the actual needs of early warning of AMS, early prevention or screening of susceptible populations. Therefore, a new AMS incidence prediction strategy is urgently needed to screen susceptible populations for acute mountain sickness after entering the plateau.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] The purpose of this application is to provide an early warning method and related equipment for acute altitude sickness, which at least overcomes the problems of the prior art to a certain extent, by collecting physiological data of target users, extracting and processing features of the data to identify abnormal physiological features and their co-occurrence frequencies, and then generating correlation information. The target early warning model is used to process the TT3, FT3 and FT4 of the target user, generate risk warning values, and provide personalized AMS risk assessment for users in high altitude areas.

[0006] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by the practice of the present invention.

[0007] According to one aspect of the present application, there is provided an early warning method for acute mountain sickness, comprising: obtaining target user information and a training data set, wherein the target user information comprises user physiological parameter information; processing the target user information to generate attribute information of the target user, wherein the attribute information of the target user comprises total triiodothyronine, free triiodothyronine and free thyroxine of the target user; preprocessing the training data set to generate a training data set with target feature data, wherein the target feature data is used to characterize physiological factors affecting the user's acute mountain sickness; obtaining an initial acute mountain sickness early warning model matching the target feature data, wherein the initial acute mountain sickness early warning model is set based on the target user information received within a preset time period; training the initial acute mountain sickness early warning model based on the training data set with the target feature data to generate a target acute mountain sickness early warning model; processing the total triiodothyronine, free triiodothyronine and free thyroxine of the target user based on the target acute mountain sickness early warning model to generate early warning information.

[0008] In one embodiment of the present application, the target user information is processed to generate attribute information of the target user, including: performing feature extraction processing on the target user information to generate a feature data set; processing the feature data set to obtain abnormal physiological characteristics and correlation co-occurrence frequencies; processing the abnormal physiological characteristics and the correlation co-occurrence frequencies to generate correlation information, wherein the correlation information is used to characterize the correlation between physiological characteristics and acute altitude sickness; processing the feature data set based on the correlation information to generate total triiodothyronine, free triiodothyronine and free thyroxine of the target user.

[0009] In one embodiment of the present application, the target user information is processed to generate the attribute information of the target user, and further includes: the method includes a calculation formula for calculating the correlation co-occurrence frequency, and the calculation formula is:

[0010]

[0011] in, represents the number of times physiological features i and j co-occur within a specific window size, represents the frequency of occurrence of physiological characteristic i in the total number of physiological parameters, Represents the frequency of occurrence of physiological feature j in the total number of physiological parameters.

[0012] In one embodiment of the present application, the target user information is processed to generate the attribute information of the target user, and further includes: the method includes a calculation formula for calculating the correlation information, and the calculation formula is:

[0013]

[0014]

[0015] in, is the co-occurrence probability, is the cosine similarity, is the Jaccard similarity.

[0016] In one embodiment of the present application, the training data set is preprocessed to generate a training data set with target feature data, including: extracting features from the training data set to determine an original feature library; dividing each feature data set according to the original feature library to generate a training data set and a verification data set; using a classifier to predict each verification data set divided from the original feature library to determine a prediction result; using a preset algorithm to train each training data set divided from the original feature library to obtain a verification data set class prediction result; generating target feature data based on the prediction result and the verification data set class prediction result, wherein the target feature data is used to characterize the physiological factors that affect the user's acute altitude sickness.

[0017] In one embodiment of the present application, the initial acute mountain sickness warning model is trained based on the training data set with target feature data to generate a target acute mountain sickness warning model, including: extracting multiple data groups from the training set, wherein each data group contains a preset number of data samples, wherein at least one data sample includes identification information; training the initial acute mountain sickness warning model based on the data samples in the multiple data groups to generate a trained acute mountain sickness warning model; processing the trained acute mountain sickness warning model based on the verification data set to generate a verification result; if the data sample containing the identification information in the verification result indicates that the physiological factors affecting the user's acute mountain sickness are in an abnormal state, then the trained acute mountain sickness warning model is used as the target acute mountain sickness warning model.

[0018] In one embodiment of the present application, the total triiodothyronine, free triiodothyronine and free thyroxine of the target user are processed based on the target acute altitude sickness warning model to generate warning information, including: the total triiodothyronine, free triiodothyronine and free thyroxine of the target user are processed based on the target acute altitude sickness warning model to generate a risk warning value; the risk warning value is processed based on a preset risk threshold to generate warning information; the method includes a calculation formula for calculating the risk warning value, and the calculation formula is:

[0019] P = -12.176+3.402 TT3+27.560 FT3 / FT4;

[0020] Among them, P is the risk warning value, TT3 is total triiodothyronine, FT3 is free triiodothyronine, and FT4 is free thyroxine.

[0021] Another aspect of the present application is an early warning device for acute mountain sickness, characterized in that it includes: an acquisition module, used to acquire target user information and a training data set, wherein the target user information includes user physiological parameter information; acquire an initial acute mountain sickness early warning model that matches the target characteristic data, wherein the initial acute mountain sickness early warning model is set based on the target user information received within a preset time period; a processing module, used to process the target user information to generate attribute information of the target user, wherein the attribute information of the target user includes total triiodothyronine, free triiodothyronine and free thyroxine of the target user; pre-process the training data set to generate a training data set with target characteristic data, wherein the target characteristic data is used to characterize physiological factors affecting the user's acute mountain sickness; train the initial acute mountain sickness early warning model based on the training data set with the target characteristic data to generate a target acute mountain sickness early warning model; process the total triiodothyronine, free triiodothyronine and free thyroxine of the target user based on the target acute mountain sickness early warning model to generate early warning information.

[0022] According to another aspect of the present application, an electronic device is characterized in that it includes: a first processor; and a memory for storing executable instructions of the first processor; wherein the first processor is configured to execute the above-mentioned early warning method for acute altitude sickness by executing the executable instructions.

[0023] According to another aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a second processor, the above-mentioned early warning method for acute altitude sickness is implemented.

[0024] According to another aspect of the present application, a computer program product is provided, including a computer program, characterized in that when the computer program is executed by a third processor, the above-mentioned early warning method for acute altitude sickness is implemented.

[0025] The present application provides an early warning method and related equipment for acute altitude sickness. The server collects the physiological data of the target user, including total triiodothyronine (TT3), free triiodothyronine (FT3) and free thyroxine (FT4). These data are feature extracted and processed to identify abnormal physiological features and their co-occurrence frequencies, thereby generating correlation information. This information is used to preprocess the training data set, divide the original feature library, and train the initial AMS early warning model accordingly. During the model training process, the classifier is used to predict the validation data set, and the training data set is trained by a preset algorithm to generate a trained early warning model. The target early warning model is used to process the target user's TT3, FT3 and FT4 to generate a risk warning value. Combined with the statistical analysis of physiological parameters, a personalized AMS risk assessment is provided for users in high altitude areas. Through timely early warning information, users can take appropriate preventive measures to reduce the risk of AMS.

[0026] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A flow chart showing an early warning method for acute mountain sickness provided by an embodiment of the present application;

[0028] Figure 2 A schematic structural diagram of an early warning device for acute altitude sickness provided by an embodiment of the present application is shown;

[0029] Figure 3 A schematic diagram of the structure of an electronic device provided by an embodiment of the present application is shown;

[0030] Figure 4 A schematic diagram of a storage medium provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0031] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0032] Combine the following Figure 1 To describe the early warning method for acute mountain sickness according to the exemplary embodiment of the present application. It should be noted that the following application scenarios are only shown for the purpose of facilitating the understanding of the spirit and principle of the present application, and the embodiments of the present application are not limited in this respect. On the contrary, the embodiments of the present application can be applied to any applicable scenario.

[0033] In one embodiment, the present application also proposes an early warning method and related equipment for acute altitude sickness. Figure 1 The following schematically shows a flow chart of an early warning method for acute mountain sickness according to an embodiment of the present application. Figure 1 As shown, the method is applied to a server, comprising:

[0034] S101, obtaining target user information and training data set.

[0035] In one embodiment, the target user information includes user physiological parameter information, and height and weight are measured using standard methods. Body mass index (BMI) is calculated as weight (kg) divided by height squared (m2). Systolic and diastolic blood pressures are collected twice in a row by an Omron electronic blood pressure monitor HEM-7430 when the participant is in a quiet and relaxed sitting position, and the average is taken as the final result. White blood cells (WBC), hemoglobin (HGB), and platelets (PLT) are measured using an automatic blood cell analyzer Sysmex XN. Thyroid hormones including total triiodothyronine (TT3), total thyroid hormone (TT4), free triiodothyronine (FT3), free thyroid hormone (FT4), and thyroid stimulating hormone (TSH) are detected using a chemiluminescence analyzer ADVIA Centaur XPTSystem. The Beckman Coulter automatic biochemical analyzer was used to detect serum creatinine (Cr), total cholesterol (TC), triglyceride (TG), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), alanine aminotransferase (ALT), aspartate aminotransferase (AST), total protein (TP), albumin (ALB), and high-sensitivity C-reactive protein (USCRP) levels.

[0036] In another embodiment, the measurement of TT3 levels and FT3 / FT4 ratio in low altitude areas is usually performed through blood tests. The following are the general steps and methods for measuring these indicators:

[0037] Collection of blood samples: It is usually necessary to draw venous blood from the subject, with the blood volume generally being 2-3 ml. No anticoagulant is used to separate the serum for measurement.

[0038] Measurement of TT3: The determination of total triiodothyronine (TT3) can be performed by radioimmunoassay (RIA), enzyme-linked immunosorbent assay (ELISA), chemiluminescent immunoassay, etc. These methods can detect the total amount of T3 in serum, including the part bound to plasma proteins and the free part.

[0039] Measurement of FT3 and FT4: Free triiodothyronine (FT3) and free thyroxine (FT4) are usually measured by immunoassays that specifically measure levels of free hormones that are not bound to plasma proteins.

[0040] Calculation of FT3 / FT4 ratio: After obtaining the measured values ​​of FT3 and FT4, their ratio can be calculated by simple division.

[0041] In addition, before the test, the person being tested needs to avoid eating iodine-rich foods, such as kelp and seaweed, as these foods may affect thyroid hormone levels. In addition, certain medications may affect the test results, so these medications may need to be discontinued before the test.

[0042] Reference value: The reference value of serum TT3 level in normal adults is 1.54-3.08nmol / L, the reference value of FT3 is 3.6±0.6ng / L, and the reference value of FT4 is 13.1±2.9ng / L. These reference values ​​may vary depending on the laboratory and detection method.

[0043] The results of TT3, FT3, and FT4 measurements can reflect the functional status of the thyroid gland. For example, these indicators may be elevated in hyperthyroidism and decreased in hypothyroidism. The FT3 / FT4 ratio may provide additional information when evaluating thyroid function and diagnosing certain diseases.

[0044] S102: Process the target user information to generate attribute information of the target user.

[0045] In one embodiment, feature extraction processing is performed on target user information to generate a feature data set, the feature data set is processed to obtain abnormal physiological features and correlation co-occurrence frequencies, the abnormal physiological features and correlation co-occurrence frequencies are processed to generate correlation information, wherein the correlation information is used to characterize the correlation between physiological features and acute altitude sickness, the feature data set is processed based on the correlation information to generate total triiodothyronine, free triiodothyronine and free thyroxine of the target user, and the attribute information of the target user includes the total triiodothyronine, free triiodothyronine and free thyroxine of the target user.

[0046] Specifically, various physiological characteristics are extracted from the health data of the target user, such as age, gender, BMI, blood pressure, heart rate, blood indicators (WBC, HGB, PLT), biochemical indicators (transaminase, creatinine, urea, lipid profile), thyroid hormone (TT3, FT3, FT4, TT4, TSH), etc. The extracted features are organized into a data set, with each row representing a user and each column representing a physiological characteristic. The normal range is determined for each physiological characteristic, and those characteristics that exceed the normal range are marked as abnormal.

[0047] Count the number of times each abnormal feature appears in the data set and calculate the co-occurrence frequency between abnormal features. Use statistical methods (such as Pearson correlation coefficient, Spearman rank correlation coefficient, or chi-square test) to evaluate the association between abnormal physiological features and AMS, and construct a correlation matrix containing the correlation measures between each physiological feature and AMS.

[0048] Using the information in the correlation matrix, weight or adjust the feature dataset to reflect the strength of each feature's association with AMS. Use machine learning algorithms (such as logistic regression, random forest, or neural network) to predict the risk of users developing AMS, and generate prediction values ​​or classification results for TT3, FT3, and FT4 for each target user based on the processed feature dataset and correlation information. Verify the accuracy of the prediction model on an independent test set, and evaluate the performance of the model using methods such as cross-validation and AUC-ROC curves. Interpret the model results, determine which physiological features are strong predictors of AMS, and apply the model to actual clinical or research scenarios to predict and prevent AMS.

[0049] In another embodiment, the method includes a calculation formula for calculating the correlation co-occurrence frequency, and the calculation formula is:

[0050]

[0051] in, represents the number of times physiological features i and j co-occur within a specific window size, represents the frequency of occurrence of physiological characteristic i in the total number of physiological parameters, Represents the frequency of occurrence of physiological feature j in the total number of physiological parameters.

[0052] Co-occurrence frequency calculations can help applicants identify which physiological traits are more likely to occur together under specific conditions (such as high altitude environments), which may indicate the risk of a certain health condition or disease. By analyzing these co-occurrence frequencies, researchers can better understand the relationship between different physiological traits and how they work together to affect an individual's adaptation to specific environmental conditions (such as high altitude).

[0053] In another implementation, the method includes a calculation formula for calculating the correlation information, and the calculation formula is:

[0054]

[0055]

[0056] in, is the co-occurrence probability, is the cosine similarity, is the Jaccard similarity.

[0057] Specifically, It is a comprehensive indicator used to measure the correlation between two physiological characteristics i and j. is the weight coefficient, which is used to adjust the relative importance of different correlation measures in the comprehensive index. is the co-occurrence frequency between features i and j; is the co-occurrence probability, which represents the probability of feature i and j co-occurring relative to the sum of the probabilities of feature i co-occurring with all other features; is the cosine similarity, which is the cosine similarity between features i and j. It measures the angle between two feature vectors and is used to evaluate their directional similarity rather than their magnitude. is the Jaccard similarity, which is the Jaccard similarity between features i and j, and it measures the size of the intersection of two feature sets relative to the size of their union.

[0058] These metrics comprehensively consider the frequency, probability, directional similarity, and set similarity of co-occurrence between features, providing a comprehensive perspective for evaluating the correlation between features. The weight coefficients \( w_1, w_2, w_3, w_4 \) allow the importance of different metrics to be adjusted according to the specific situation, making the calculation of correlation information more flexible and adaptable. This comprehensive calculation method can be applied to various data analysis scenarios, including but not limited to medical research, health monitoring, disease prediction, etc., and helps to identify and understand patterns and associations in complex data sets.

[0059] S103, preprocessing the training data set to generate a training data set with target feature data.

[0060] In one implementation, feature extraction is performed on a training data set to determine an original feature library, each feature data set is divided according to the original feature library, a training data set and a verification data set are generated, and a classifier is used to predict each verification data set divided from the original feature library to determine a prediction result. A preset algorithm is used to divide each training data set in the original feature library for training to obtain a verification data set class prediction result, and target feature data is generated based on the prediction result and the verification data set class prediction result, and the target feature data is used to characterize the physiological factors that affect the user's acute altitude sickness.

[0061] Identify and select those features that best characterize the user's acute mountain sickness (AMS) risk from a large number of physiological features. Specifically, extract physiological features that may affect AMS from the target user's health data. These features constitute the original feature library, which is divided into a training dataset and a validation dataset. The training dataset is used to build and train the model, and the validation dataset is used to evaluate the performance and generalization ability of the model.

[0062] Use classifiers (such as logistic regression, support vector machine, random forest, neural network, etc.) to predict the validation data set and determine the prediction results, which are the model's predictions on whether the samples in the validation data set will develop into AMS. Use the preset algorithm (the same classifier as the prediction stage) to train on the training data set to obtain a trained model that can predict the category of the validation data set. Based on the prediction results and the prediction results of the validation data set, select the most influential features and generate target feature data, which is a selected feature subset used to characterize the physiological factors that affect the user's AMS.

[0063] Use feature selection techniques (such as recursive feature elimination, model-based feature selection, etc.) to evaluate the contribution of each feature in the original feature library to predicting AMS, select those features that contribute most to the model's prediction performance, and optimize the model by adjusting model parameters, using different algorithms, or applying feature engineering. The purpose of optimization is to improve the accuracy and reliability of the model's AMS prediction.

[0064] Interpret the target feature data, understand why these specific physiological factors are associated with AMS risk, and apply the optimized model and target feature data to actual scenarios, such as AMS risk assessment for users in high altitude areas. Specifically, in the risk assessment model of acute mountain sickness (AMS), the target feature data refers to those physiological features that are identified as significantly associated with the risk of AMS. These features are selected from the original feature library through a data-driven approach, and they can effectively help the model predict the likelihood of an individual developing AMS in a high-altitude environment. The target feature data usually include those physiological indicators that are most important for AMS risk prediction, such as red blood cell count, heart rate, blood pressure, certain biochemical indicators (such as lactate level), and thyroid hormone levels in the blood. The correlation between these features and AMS risk may be based on physiological principles. For example, a higher red blood cell count may be associated with better oxygen carrying capacity, which may reduce the risk of AMS; while changes in certain biochemical indicators may indicate the body's adaptability to high-altitude environments. Through statistical analysis, such as correlation analysis, regression analysis, or feature selection algorithms in machine learning, it can be determined which features are most statistically significantly associated with AMS risk.

[0065] The optimized model and target feature data can be used to assess the AMS risk of users in actual high-altitude environments. For example, before mountaineering, hiking or other high-altitude activities, the AMS risk of participants can be assessed by measuring these key physiological characteristics. Based on the results of the risk assessment, appropriate preventive measures can be taken, such as adjusting the ascent speed, using auxiliary oxygen or taking medication in advance. For individuals with different physiological characteristics, personalized health advice and adjusted activity plans can be provided to reduce the risk of AMS.

[0066] Through methods such as cross-validation and hyperparameter tuning, the performance of the model can be optimized to ensure its accuracy and robustness in practical applications. Feature engineering is used to improve the expression of features, such as data transformation, feature construction, or dimensionality reduction techniques to enhance the predictive ability of the model. In this way, target feature data and optimized models can serve as powerful tools to help predict and manage health risks in high-altitude environments, thereby improving safety and activity success.

[0067] S104, obtaining an initial acute altitude sickness warning model that matches the target characteristic data.

[0068] In one embodiment, an initial acute mountain sickness (AMS) warning model matching the target feature data is obtained. The initial acute mountain sickness warning model is set based on the target user information received within a preset time period. Specifically, the main purpose of the model is to predict the risk of an individual developing AMS within a specific time, and the health data of the target user is collected within a preset time period, which should include physiological characteristics related to the risk of AMS. Data cleaning is a key step in data analysis and machine learning projects, which directly affects the quality and predictive performance of the model. Specifically, check the duplicate rows in the data set and delete them to avoid bias in the analysis, correct or delete erroneous data caused by input errors, data entry errors, or data transmission errors, and ensure that all data columns are of the correct data type, such as date format, numeric type, etc. If the number of missing values ​​is small, the records containing missing values ​​can be directly deleted. For numerical features, the missing values ​​can be filled with the mean, median, or mode; for categorical features, the mode or a specific category can be filled. Use statistical methods (such as regression interpolation, k-nearest neighbor interpolation) or machine learning algorithms (such as decision trees, random forests) to predict missing values ​​and build a model to predict missing values. This method is usually used when the number of missing values ​​is large.

[0069] Use statistical methods (such as standard deviation, box plots, Z scores) or machine learning algorithms (such as DBSCAN, IsolationForest) to identify outliers. If the outliers are caused by errors, these records can be deleted directly. If possible, correct the outliers to a reasonable range. In some cases, outliers may represent important information, such as fraud detection, and these values ​​should be retained. Convert the data to a distribution with zero mean and unit variance. The dimensions and numerical ranges of different features may vary greatly. Normalization ensures that all features have the same importance in the model and avoids the model being biased towards features with larger numerical ranges.

[0070] Select the features most relevant to the target variable to reduce the complexity of the model and the risk of overfitting, create new features or transform existing features to improve the performance of the model. Before cleaning the data, you need to fully understand the business context of the data and the data generation process. When dealing with missing values ​​and outliers, you should act with caution to avoid introducing biases. Record the steps of data cleaning and preprocessing to facilitate the reproducibility of the model and subsequent audits, ensure the quality of the data set, and lay a solid foundation for building an accurate and reliable AMS early warning model. Choose an appropriate statistical or machine learning model based on the nature of the problem and the characteristics of the data. Common models include logistic regression, decision trees, random forests, support vector machines, neural networks, etc.

[0071] Use the target feature data to train the selected model to build the initial AMS warning model, and use the validation data set to evaluate the model's predictive performance. Common evaluation indicators include accuracy, recall, precision, F1 score, and area under the ROC curve (AUC). Adjust the model parameters based on the validation results, perform feature selection or engineering to optimize the model performance, and deploy the trained initial AMS warning model to actual applications, such as medical monitoring systems in high-altitude areas. Continuously monitor the performance of the model in actual applications, and update and adjust it based on new data and feedback.

[0072] During the implementation process, the setting of the initial AMS warning model includes the following features: the model needs to consider the time sensitivity of the data, because the user's physiological state may change over time, and the model can be dynamically updated based on the newly received user information to adapt to the changes in the user's state. In addition, the model contains personalized elements and can make customized predictions based on the user's historical data and physiological characteristics, thereby building an effective initial AMS warning model to provide health risk warnings for people active in high-altitude environments.

[0073] S105, training the initial acute mountain sickness warning model based on the training data set with target feature data to generate a target acute mountain sickness warning model.

[0074] In one embodiment, multiple data sets are extracted from a training set, wherein each data set contains a preset number of data samples, wherein at least one data sample includes identification information, an initial acute mountain sickness warning model is trained based on the data samples in the multiple data sets to generate a trained acute mountain sickness warning model, and the trained acute mountain sickness warning model is processed based on a verification data set to generate a verification result. If the data sample containing identification information in the verification result indicates that the physiological factors affecting the user's acute mountain sickness are in an abnormal state, the trained acute mountain sickness warning model is used as the target acute mountain sickness warning model.

[0075] Specifically, multiple data sets are extracted from the training set, each data set contains a certain number of data samples, and it is ensured that at least one data sample in each data set contains identification information, which may refer to abnormal states of physiological factors related to AMS. The initial AMS early warning model is trained using the data samples in the multiple data sets. Through the training process, the model learns the patterns and relationships in the data, so that it can predict the AMS risk. After the training is completed, the initial model is updated to a trained AMS early warning model, which contains the parameters learned from the training data.

[0076] The trained AMS warning model is evaluated using an independent validation dataset to generate validation results that demonstrate the model's predictive power for samples in the validation dataset. The validation results are analyzed with special attention paid to data samples that contain identification information. If the validation results show that data samples containing identification information (i.e., users whose physiological factors are in an abnormal state) are correctly identified as AMS high risk by the model, then the performance of the model is considered satisfactory. If the validation results meet expectations, i.e., the model can effectively identify AMS risks, then the trained AMS warning model becomes the target acute mountain sickness warning model.

[0077] In addition, it is necessary to ensure that the training and validation datasets are representative of the target population so that the model can be generalized to new users and avoid model overfitting, which may require regularization techniques or simplified models, and to select appropriate performance indicators to evaluate the model, such as sensitivity, specificity, accuracy, etc. Even after the model is confirmed as the target model, its performance in actual operation needs to be continuously monitored and adjusted as needed. Ensure that the AMS warning model is not only effective in theory, but also can accurately predict the AMS risk of individuals in practical applications, thereby providing better health protection for participants in activities in high altitude areas.

[0078] S106, processing the total triiodothyronine, free triiodothyronine and free thyroxine of the target user based on the target acute altitude sickness warning model to generate warning information.

[0079] In one embodiment, the total triiodothyronine, free triiodothyronine and free thyroxine of the target user are processed based on the target acute altitude sickness warning model to generate a risk warning value, and the risk warning value is processed based on a preset risk threshold to generate warning information.

[0080] First, the physiological data of the target user is collected, including the levels of total triiodothyronine (TT3), free triiodothyronine (FT3), and free thyroxine (FT4). The target AMS early warning model applies a given calculation formula to calculate the risk warning value of each user. The specific formula is as follows:

[0081] P = -12.176+3.402 TT3+27.560 FT3 / FT4;

[0082] Among them, P is the risk warning value, TT3 is total triiodothyronine, FT3 is free triiodothyronine, FT4 is free thyroxine, and TT3, FT3 and FT4 are the actual measured values ​​of the user.

[0083] As an example, suppose a target user has the following physiological measurements:

[0084] TT3 = 1.0 ug / L, FT3 = 5.0 pmol / L, FT4 = 15 pmol / L.

[0085] According to the formula, calculate the risk warning value (P):

[0086] .

[0087] The calculated risk warning value (P) is evaluated based on the preset risk threshold. For example, if the threshold is set to 0.61: If P> 0.61, the user is considered to have a high risk of AMS, and a warning message is generated to alert the user and medical professionals; if P If the value is 0.61, the user is considered to be at low risk and may not need additional precautions. Based on the generated warning information, notifications can be sent to users to advise them to take appropriate precautions, such as slowing down the ascent rate, using supplementary oxygen, or taking medication. The user's physiological data and AMS symptoms are continuously monitored so that the warning model and preventive measures can be adjusted in a timely manner, which helps to better manage and mitigate the risk of AMS in high-altitude environments.

[0088] This method is an acute mountain sickness (AMS) early warning system that aims to predict the risk of AMS by analyzing the user's physiological parameters. First, the system collects the target user's physiological data, including total triiodothyronine (TT3), free triiodothyronine (FT3), and free thyroxine (FT4). These data are feature extracted and processed to identify abnormal physiological features and their co-occurrence frequencies, thereby generating correlation information.

[0089] The correlation information is further processed through calculation formulas, including co-occurrence probability, cosine similarity, and Jaccard similarity, to evaluate the association between physiological features and AMS. Then, the training data set is preprocessed using this information to divide the original feature library, and the initial AMS warning model is trained based on it. During the model training process, the validation data set is predicted using the classifier, and the training data set is trained using a preset algorithm to generate a trained warning model.

[0090] Finally, the target warning model is used to process the target user's TT3, FT3, and FT4 to generate risk warning values. These warning values ​​are compared with the preset risk thresholds to generate the final warning information. If the performance of the warning model on the validation data set meets specific conditions, that is, it can accurately identify the abnormal state of physiological factors that affect the user's AMS risk, the model will be confirmed as the target acute mountain sickness warning model.

[0091] The core of this method is to provide personalized AMS risk assessment for users in high altitude areas through machine learning technology combined with statistical analysis of physiological parameters. With timely warning information, users can take appropriate preventive measures to reduce the risk of AMS.

[0092] In one embodiment, if Figure 2 As shown, the present application also provides an early warning device for acute altitude sickness, comprising:

[0093] The acquisition module 201 is used to acquire target user information and a training data set, wherein the target user information includes user physiological parameter information; acquire an initial acute plateau reaction warning model that matches the target characteristic data, wherein the initial acute plateau reaction warning model is set based on the target user information received within a preset time period;

[0094] Processing module 202 is used to process the target user information to generate attribute information of the target user, wherein the attribute information of the target user includes total triiodothyronine, free triiodothyronine and free thyroxine of the target user; preprocess the training data set to generate a training data set with target feature data, wherein the target feature data is used to characterize the physiological factors affecting the user's acute mountainous plateau reaction; train the initial acute mountainous plateau reaction warning model based on the training data set with target feature data to generate a target acute mountainous plateau reaction warning model; process the total triiodothyronine, free triiodothyronine and free thyroxine of the target user based on the target acute mountainous plateau reaction warning model to generate warning information.

[0095] The present application embodiment provides an electronic device, such as Figure 3 As shown, the electronic device 3 includes a first processor 300, a memory 301, a bus 302 and a communication interface 303. The first processor 300, the communication interface 303 and the memory 301 are connected via the bus 302; the memory 301 stores a computer program that can be run on the first processor 300, and when the first processor 300 runs the computer program, the early warning method for acute altitude sickness provided in any of the aforementioned embodiments of the present application is executed.

[0096] The memory 301 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 303 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0097] The bus 302 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory 301 is used to store a program, and the first processor 300 executes the program after receiving an execution instruction. The early warning method for acute mountain sickness disclosed in any implementation of the aforementioned embodiment of the present application may be applied to the first processor 300, or implemented by the first processor 300.

[0098] The first processor 300 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the first processor 300. The above-mentioned first processor 300 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a readily available programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be embodied as a hardware decoding processor to execute, or a combination of hardware and software modules in the decoding processor to execute. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 301, and the first processor 300 reads the information in the memory 301 and completes the steps of the above method in combination with its hardware.

[0099] The electronic device provided in the above-mentioned embodiments of the present application and the early warning method for acute altitude sickness provided in the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0100] The present application embodiment provides a computer-readable storage medium, such as Figure 4 As shown, the computer-readable storage medium 401 stores a computer program, and when the computer program is read and executed by the second processor 402, the early warning method for acute altitude sickness as described above is implemented.

[0101] The technical solution of the embodiment of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling an electronic device (which may be an air conditioner, a refrigeration device, a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the method described in the embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

[0102] The computer-readable storage medium provided in the above-mentioned embodiments of the present application and the early warning method for acute altitude sickness provided in the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0103] An embodiment of the present application provides a computer program product, including a computer program, wherein the computer program is executed by a third processor to implement the method described above.

[0104] The computer program product provided in the above-mentioned embodiments of the present application and the early warning method for acute altitude sickness provided in the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0105] It should be noted that, in this application, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or still includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0106] Each embodiment in the present application is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the early warning method, electronic device, electronic device, and readable storage medium embodiment for evaluating acute mountain sickness, since they are basically similar to the early warning method embodiment for acute mountain sickness described above, the description is relatively simple, and the relevant parts can be referred to the partial description of the early warning method embodiment for acute mountain sickness described above.

[0107] Although the present application is disclosed as above, the present application is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present application. Therefore, the protection scope of the present application shall be subject to the scope defined by the claims.

Claims

1. A method for early warning of acute altitude sickness, characterized in that: include: Acquire target user information and a training data set, wherein the target user information includes user physiological parameter information; Processing the target user information to generate attribute information of the target user, wherein the attribute information of the target user includes total triiodothyronine, free triiodothyronine and free thyroxine of the target user; Preprocessing the training data set to generate a training data set with target feature data, wherein the target feature data is used to characterize physiological factors that affect the user's acute altitude sickness; Acquire an initial acute plateau reaction warning model that matches the target characteristic data, wherein the initial acute plateau reaction warning model is set based on target user information received within a preset time period; Training the initial acute mountain sickness warning model based on the training data set with target feature data to generate a target acute mountain sickness warning model; Based on the target acute altitude sickness warning model, the total triiodothyronine, free triiodothyronine and free thyroxine of the target user are processed to generate warning information; Processing the target user information to generate attribute information of the target user, including: performing feature extraction processing on the target user information to generate a feature data set; processing the feature data set to obtain abnormal physiological characteristics and correlation co-occurrence frequencies; processing the abnormal physiological characteristics and the correlation co-occurrence frequencies to generate correlation information, wherein the correlation information is used to characterize the correlation between physiological characteristics and acute altitude sickness; processing the feature data set based on the correlation information to generate total triiodothyronine, free triiodothyronine and free thyroxine of the target user; The method includes a calculation formula for calculating the correlation co-occurrence frequency, and the calculation formula is: ； ； ； in, represents the number of times physiological features i and j co-occur within a specific window size, represents the frequency of occurrence of physiological characteristic i in the total number of physiological parameters, represents the frequency of occurrence of j physiological feature in the total number of physiological parameters; The method includes a calculation formula for calculating correlation information, and the calculation formula is: ； ； ； ； in, is the co-occurrence probability, is the cosine similarity, is the Jaccard similarity; The total triiodothyronine, free triiodothyronine and free thyroxine of the target user are processed based on the target acute altitude sickness warning model to generate warning information, including: based on the target acute altitude sickness warning model, the total triiodothyronine, free triiodothyronine and free thyroxine of the target user are processed to generate a risk warning value; based on a preset risk threshold, the risk warning value is processed to generate warning information; the method includes a calculation formula for calculating the risk warning value, and the calculation formula is: P = -12.176 + 3.402 TT3+27.560 FT3 / FT4; where P is the risk warning value, TT3 is total triiodothyronine, FT3 is free triiodothyronine, and FT4 is free thyroxine.

2. The method according to claim 1, characterized in that Preprocessing the training data set to generate a training data set with target feature data includes: Extracting features from the training data set to determine an original feature library; Divide each feature data set according to the original feature library to generate a training data set and a verification data set; Use the classifier to divide the original feature library into various verification data sets for prediction and determine the prediction results; Use the preset algorithm to divide each training data set in the original feature library for training, and obtain the prediction results of the validation data set class; According to the prediction results and the prediction results of the verification data set class, target feature data is generated, and the target feature data is used to characterize the physiological factors that affect the user's acute altitude sickness.

3. The method according to claim 2, characterized in that The initial acute mountain sickness warning model is trained based on the training data set with target feature data to generate a target acute mountain sickness warning model, including: Extracting a plurality of data groups from the training data set, wherein each data group includes a preset number of data samples, wherein at least one data sample includes identification information; Training the initial acute mountain sickness warning model based on data samples in the plurality of data groups to generate a trained acute mountain sickness warning model; Processing the trained acute altitude sickness warning model based on the verification data set to generate a verification result; If the data sample containing identification information in the verification result indicates that the physiological factors affecting the user's acute mountain sickness are in an abnormal state, the trained acute mountain sickness warning model is used as the target acute mountain sickness warning model.

4. An early warning device for acute altitude sickness, characterized in that: The device comprises: An acquisition module is used to acquire target user information and a training data set, wherein the target user information includes user physiological parameter information; acquire an initial acute plateau reaction warning model that matches the target characteristic data, wherein the initial acute plateau reaction warning model is set based on the target user information received within a preset time period; A processing module is used to process the target user information to generate attribute information of the target user, wherein the attribute information of the target user includes total triiodothyronine, free triiodothyronine and free thyroxine of the target user; preprocess the training data set to generate a training data set with target feature data, wherein the target feature data is used to characterize physiological factors that affect the user's acute mountain sickness; train the initial acute mountain sickness warning model based on the training data set with target feature data to generate a target acute mountain sickness warning model; process the total triiodothyronine, free triiodothyronine and free thyroxine of the target user based on the target acute mountain sickness warning model to generate warning information; Processing the target user information to generate attribute information of the target user, including: performing feature extraction processing on the target user information to generate a feature data set; processing the feature data set to obtain abnormal physiological characteristics and correlation co-occurrence frequencies; processing the abnormal physiological characteristics and the correlation co-occurrence frequencies to generate correlation information, wherein the correlation information is used to characterize the correlation between physiological characteristics and acute altitude sickness; processing the feature data set based on the correlation information to generate total triiodothyronine, free triiodothyronine and free thyroxine of the target user; The calculation formula for calculating the correlation co-occurrence frequency is: ； ； ； in, represents the number of times physiological features i and j co-occur within a specific window size, represents the frequency of occurrence of physiological characteristic i in the total number of physiological parameters, represents the frequency of occurrence of j physiological feature in the total number of physiological parameters; The calculation formula used to calculate the correlation information is: ； ； ； ； in, is the co-occurrence probability, is the cosine similarity, is the Jaccard similarity; The total triiodothyronine, free triiodothyronine and free thyroxine of the target user are processed based on the target acute altitude sickness warning model to generate warning information, including: the total triiodothyronine, free triiodothyronine and free thyroxine of the target user are processed based on the target acute altitude sickness warning model to generate a risk warning value; the risk warning value is processed based on a preset risk threshold to generate warning information; the calculation formula for calculating the risk warning value is: P = -12.176 + 3.402 TT3+27.560 FT3 / FT4; where P is the risk warning value, TT3 is total triiodothyronine, FT3 is free triiodothyronine, and FT4 is free thyroxine.

5. An electronic device, characterized in that: include: a first processor; and a memory for storing executable instructions of the first processor; Wherein, the first processor is configured to execute the early warning method for acute mountain sickness according to any one of claims 1 to 3 by executing the executable instructions.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the second processor, the early warning method for acute mountain sickness described in any one of claims 1 to 3 is implemented.

Images