Aortic dissection rescue guiding method and system based on multi-dimensional data fusion

Abstract

The invention relates to the technical field of medical information, in particular to an aortic dissection rescue guidance method and system based on multi-dimensional data fusion, the method comprises the following steps: acquiring comprehensive medical data of a patient, the comprehensive medical data comprising at least one of real-time vital sign data, iconography data and medical history information; performing fusion processing on the comprehensive medical data, constructing a disease condition portrait based on a fusion data set, capturing a disease condition change trend, and obtaining a dynamic disease condition portrait of the patient; constructing a rule base based on historical data, and matching the patient's condition portraits with rules in the rule base to obtain condition features and risk indexes; a risk assessment standard is established based on the illness state characteristics and the risk indexes, the illness state severity is assessed through the risk assessment standard, possible complications are predicted, and personalized rescue suggestions and decision support are provided for medical staff.

Description

Technical Field

[0001] The present invention relates to the field of medical information technology, and in particular to an aortic dissection rescue guidance method and system based on multi-dimensional data fusion. Background Art

[0002] Aortic dissection is a critical cardiovascular disease with an acute onset, rapid progression, high mortality rate, and requires rapid medical intervention. Its mortality rate is very high, about 40% at the first onset, and the mortality rate increases by 1% every hour. In addition, the misdiagnosis rate of aortic dissection is also very high. Among patients who are finally diagnosed with aortic dissection, about one-third of patients will be misdiagnosed. Patients suspected of aortic dissection should undergo imaging examinations as soon as possible, but the imaging diagnosis of aortic dissection is relatively complicated and requires interpretation by experienced doctors, which increases the difficulty and time cost of diagnosis to a certain extent.

[0003] At present, the rescue guidance methods for aortic dissection mainly include drug therapy, surgical treatment and interventional treatment. However, the selection and effect evaluation of these treatment methods depend on accurate diagnosis and condition assessment. Therefore, improving the diagnostic accuracy and treatment efficiency of aortic dissection is the key to reducing its mortality rate and improving patient prognosis.

[0004] With the continuous advancement of medical technology, multi-dimensional data fusion technology is increasingly being used in the medical field. By integrating data from different medical imaging devices, physiological monitoring devices, and clinical information systems, a comprehensive view of patient information can be formed, which can provide doctors with a more accurate and comprehensive basis for diagnosis. Summary of the invention

[0005] In order to form a comprehensive patient information view and provide doctors with a more accurate and comprehensive diagnosis basis, the purpose of the present invention is to provide an aortic dissection rescue guidance method and system based on multi-dimensional data fusion. The technical solutions adopted are as follows:

[0006] In a first aspect, the present application discloses a method for guiding the rescue of aortic dissection based on multi-dimensional data fusion, the method comprising:

[0007] S1. Obtaining comprehensive medical data of the patient, wherein the comprehensive medical data includes at least one of real-time vital sign data, imaging data, and medical history information;

[0008] S2. Fusing the comprehensive medical data, and constructing a disease profile based on the fused data set, and capturing the disease change trend, to obtain a dynamic disease profile of the patient;

[0009] S3. Build a rule base based on historical data, and match the patient's condition profile with the rules in the rule base to obtain condition characteristics and risk indicators;

[0010] S4. Establish risk assessment standards based on disease characteristics and risk indicators, evaluate the severity of the disease and predict possible complications through the risk assessment standards, so as to provide personalized rescue suggestions and decision support to medical staff.

[0011] Furthermore, in step S2, the comprehensive medical data is fused and processed, and a disease profile is constructed based on the fused data set, and the disease change trend is captured to obtain a dynamic disease profile of the patient, including:

[0012] S21, using a principal component analysis method and a nonlinear dimensionality reduction technique to fuse the comprehensive medical data to obtain a fused data set containing multi-dimensional information;

[0013] S22. Based on the fused data set, construct a static condition portrait of the patient using augmented reality and three-dimensional modeling technology;

[0014] S23, using time series analysis and machine learning algorithms to perform in-depth mining on the comprehensive data set to capture the changing trend of the patient's condition;

[0015] S24. Update the patient's three-dimensional virtual model and condition portrait based on the disease change trend to achieve real-time update and visual display of the patient's dynamic condition portrait.

[0016] Furthermore, in step S21, the principal component analysis method is used to fuse the comprehensive medical data through nonlinear dimensionality reduction technology to obtain a comprehensive data set containing multi-dimensional information, including:

[0017] S211, preprocessing the comprehensive medical data to obtain a preprocessed data set;

[0018] S212, performing linear dimension reduction on the preprocessed data set using a principal component analysis method to obtain a principal component set;

[0019] S213, based on feature importance evaluation and data distribution characteristics, weighting and nonlinearly combining the principal component set to obtain a weighted nonlinear principal component set;

[0020] S214. Perform deep dimensionality reduction processing on the weighted nonlinear principal component set by nonlinear dimensionality reduction technology to capture the nonlinear relationship and complex structure in the data and obtain a comprehensive data set containing multi-dimensional information.

[0021] Furthermore, in step S212, the linear dimension reduction is performed on the preprocessed data set using the principal component analysis method to obtain a principal component set, including:

[0022] S2121, calculating the covariance matrix between the features in the preprocessed data set;

[0023] S2122, performing eigendecomposition on the covariance matrix to obtain multiple eigenvalues ​​and corresponding eigenvectors;

[0024] S2123, setting a k value range according to the size of the eigenvalue, and selecting a k value that minimizes the reconstruction error within the k value range as the optimal k value;

[0025] S2124, according to the optimal k value, select eigenvectors corresponding to k largest eigenvalues, and construct a principal component space based on the k eigenvectors;

[0026] S2125. Project the preprocessed data set into the principal component space to obtain a principal component set after dimensionality reduction.

[0027] Furthermore, in step S23, the comprehensive data set is deeply mined using time series analysis and machine learning algorithms to capture the changing trend of the patient's condition, including:

[0028] S231, determining the key physiological indicators of the patient based on the comprehensive data set, and processing the key physiological indicators using time series analysis technology to obtain a corresponding dynamic change trend graph;

[0029] S232, based on the moving average and the threshold setting, analyzing the abnormal fluctuation points from the dynamic change trend graph, and marking the potential health risk areas;

[0030] S233. Process abnormal fluctuation points and potential health risk areas based on machine learning algorithms to predict the changing trend of the patient's condition.

[0031] Furthermore, in step S232, based on the moving average and the threshold setting, abnormal fluctuation points are analyzed from the dynamic change trend chart, and potential health risk areas are marked, including:

[0032] S2321, smoothing the dynamic change trend graph based on a moving average line to obtain a basic trend graph from which random fluctuations are filtered out;

[0033] S2322. For each data point in the basic trend graph, calculate the distance between the data point and the moving average line to determine the deviation between the two;

[0034] S2323, setting a threshold value based on historical data analysis, and when it is determined that the deviation of a corresponding data point is greater than the threshold value, treating the corresponding data point as an abnormal fluctuation point;

[0035] S2324. Perform cluster analysis based on the abnormal fluctuation points obtained to identify and mark potential health risk areas.

[0036] Furthermore, in step S3, the rule base is constructed based on historical data, and the patient's condition profile is matched with the rules in the rule base to obtain condition characteristics and risk indicators, including:

[0037] S31. Using data mining algorithms, extract historical disease information from historical data;

[0038] S32. Generate rules for describing the correlation between historical disease information and disease characteristics, as well as risk indicators, based on data analysis technology based on statistical learning methods;

[0039] S33, importing the generated rules into the rule base to form a corresponding rule system;

[0040] S34, extracting real-time condition information from the patient's condition portrait, and performing rule-based reasoning on the real-time condition information and various rules in the rule system to obtain a target rule that meets the current patient's condition;

[0041] S35. Analyze the target rule to obtain the disease characteristics and risk indicators that are consistent with the current patient condition.

[0042] Furthermore, in step S4, the risk assessment criteria are established based on the disease characteristics and risk indicators, and the severity of the disease is assessed by the risk assessment criteria, as well as possible complications are predicted, including:

[0043] S41. Determine risk factors based on disease characteristics and risk indicators, perform quantitative scoring of each risk factor through the risk matrix method, and establish risk assessment standards based on the scoring results and expert experience, wherein the risk assessment standards integrate the diversity, dynamic changes and potential complications of the disease, and can automatically generate a personalized risk assessment report;

[0044] S42. Using the risk assessment criteria, assess the severity of the disease based on real-time updated disease data, and predict possible complications.

[0045] In the second aspect, the present application discloses an aortic dissection rescue guidance system based on multi-dimensional data fusion, the system comprising a data acquisition module, a data fusion and processing module, a rule base construction and matching module, and a condition assessment and prediction module, wherein:

[0046] The data acquisition module is used to acquire the patient's comprehensive medical data, wherein the comprehensive medical data includes at least one of real-time vital sign data, imaging data, and medical history information;

[0047] The data fusion and processing module is used to fuse the comprehensive medical data, and to construct a disease profile based on the fused data set, and to capture the disease change trend, so as to obtain a dynamic disease profile of the patient;

[0048] The rule base construction and matching module is used to construct a rule base based on historical data, and match the patient's condition profile with the rules in the rule base to obtain condition characteristics and risk indicators;

[0049] The condition assessment and prediction module is used to establish risk assessment standards based on condition characteristics and risk indicators, assess the severity of the condition through the risk assessment standards, and predict possible complications, so as to provide personalized rescue suggestions and decision support to medical staff.

[0050] In a third aspect, the present application discloses a computer-readable storage medium storing a computer program, characterized in that when the computer program is executed by a processor, the aortic dissection rescue guidance method based on multi-dimensional data fusion is implemented.

[0051] The present invention has the following beneficial effects:

[0052] 1) Using the integrated medical data after fusion processing to build a dynamic disease profile, it can capture the trend of disease changes in real time, so as to form a comprehensive patient information view, timely reflect the patient's real-time health status and potential risks, and provide medical staff with more accurate and timely patient status monitoring;

[0053] 2) By matching the rule base built based on historical data with the patient's condition profile, the disease characteristics and risk indicators can be automatically identified, and then risk assessment standards can be established. In this way, the severity of the disease can be quickly and accurately assessed, and the potential risk of complications can be predicted, thereby guiding medical staff to develop more reasonable and effective treatment plans and improve the scientificity and efficiency of medical decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1A method flow chart of a method for guiding aortic dissection rescue based on multi-dimensional data fusion provided by one embodiment of the present invention;

[0056] Figure 2 A system structure diagram of an aortic dissection rescue guidance system based on multi-dimensional data fusion provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of an aortic dissection rescue guidance method and system based on multi-dimensional data fusion proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0058] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0059] The specific scheme of the aortic dissection rescue guidance method and system based on multi-dimensional data fusion provided by the present invention is described in detail below with reference to the accompanying drawings.

[0060] See also Figure 1 , which shows a method flow chart of an aortic dissection rescue guidance method based on multi-dimensional data fusion provided by an embodiment of the present invention, the method comprising:

[0061] Step S1, obtaining comprehensive medical data of the patient, wherein the comprehensive medical data includes at least one of real-time vital sign data, imaging data, and medical history information.

[0062] Step S2, fusing the comprehensive medical data, and constructing a disease profile based on the fused data set, and capturing the disease change trend, to obtain a dynamic disease profile of the patient.

[0063] Step S3, building a rule base based on historical data, and matching the patient's condition profile with the rules in the rule base to obtain condition characteristics and risk indicators.

[0064] Step S4, establishing risk assessment criteria based on disease characteristics and risk indicators, assessing the severity of the disease and predicting possible complications through the risk assessment criteria, so as to provide personalized rescue suggestions and decision support for medical staff.

[0065] As can be seen from the above, the aortic dissection rescue guidance method based on multi-dimensional data fusion disclosed in this application uses the integrated medical data after fusion processing to construct a dynamic disease portrait, which can capture the trend of disease changes in real time, so that a comprehensive patient information view can be formed, timely reflecting the patient's real-time health status and potential risks, and providing medical staff with more accurate and timely patient status monitoring; by matching the rule base built based on historical data with the patient's disease portrait, the disease characteristics and risk indicators can be automatically identified, and then the risk assessment standard can be established. In this way, the severity of the disease can be quickly and accurately assessed, and the potential risk of complications can be predicted, thereby guiding medical staff to formulate more reasonable and effective treatment plans and improve the scientificity and efficiency of medical decision-making.

[0066] In one embodiment, in step S2, the integrated medical data is fused and processed, and a disease profile is constructed based on the fused data set, and the disease change trend is captured to obtain a dynamic disease profile of the patient, including:

[0067] Step S21, using the principal component analysis method and nonlinear dimensionality reduction technology to fuse the comprehensive medical data to obtain a fused data set containing multi-dimensional information.

[0068] Specifically, this application will preprocess the comprehensive medical data set, including data cleaning, missing value processing, outlier detection, etc., to ensure the accuracy of the data. Afterwards, the principal component analysis method is used to perform linear dimensionality reduction on the preprocessed data set to extract the main components in the data and obtain a set of principal components. Afterwards, in order to better reflect the nonlinear relationship between the key information and features in the data, this application also combines the data distribution characteristics and feature importance evaluation to weight and nonlinearly combine the obtained principal component set. Finally, in order to capture the nonlinear relationships and complex structures in the data, a nonlinear dimensionality reduction technique is introduced. By mapping the weighted nonlinearly combined data set to a low-dimensional space, while retaining the key information and nonlinear relationships in the data, a comprehensive data set containing multi-dimensional information is obtained.

[0069] Step S22, based on the fused data set, using augmented reality and three-dimensional modeling technology to construct a static condition portrait of the patient.

[0070] Specifically, this application will extract key information related to the patient's condition from the fused data set, such as the location, size, morphology, state of surrounding tissues, and possible pathological types of the lesions. Afterwards, a three-dimensional modeling software is used to construct a three-dimensional model of the lesion and its surrounding tissues based on the extracted key information. Finally, the constructed three-dimensional model is imported into the augmented reality application, and the three-dimensional model is fused with the patient's actual imaging data (such as CT scans, MRI images, etc.) through AR technology. This fusion can be presented in real time from the doctor's perspective, allowing the doctor to intuitively see the patient's static condition portrait.

[0071] Step S23, using time series analysis and machine learning algorithms to perform in-depth mining on the comprehensive data set to capture the changing trend of the patient's condition.

[0072] Specifically, this application will first determine the patient's key physiological indicators from the comprehensive data set, which can reflect the patient's health status and changes in the condition. Subsequently, these key physiological indicators are processed using time series analysis technology, and the changing patterns and trends of the indicators over time are revealed by calculating the moving average, drawing dynamic change trend charts, and other methods. After that, based on the moving average and threshold setting, the abnormal fluctuation points are analyzed from the dynamic change trend chart. These abnormal fluctuation points may indicate that the patient's health status has changed significantly or there are potential risks. At the same time, this application also combines cluster analysis methods to mark potential health risk areas, which may correspond to specific disease states or health problems. Finally, machine learning algorithms are used to process abnormal fluctuation points and potential health risk areas. Through training models, feature extraction, and pattern recognition, it is possible to learn the patient's disease change pattern and predict future disease change trends.

[0073] Step S24, updating the patient's three-dimensional virtual model and condition portrait based on the disease change trend, so as to achieve real-time updating and visual display of the patient's dynamic condition portrait.

[0074] Specifically, this application will adjust the relevant parameters in the three-dimensional virtual model based on the trend of disease changes, such as the size, shape, position and functional status of the organs, to ensure that the model can accurately reflect the patient's latest health status, and determine the key change points of the patient's health status according to the trend of disease changes. These key change points include the transformation of the disease, or the emergence of new pathological characteristics, etc. Finally, based on these key change points, the disease portrait is synchronously updated, and the disease portrait is used to reflect the changes in the patient's health status in real time, so as to provide accurate data support for the subsequent calculation of disease characteristics and risk indicators.

[0075] In the above embodiment, the comprehensive data set is deeply mined through time series analysis and machine learning algorithms to capture the trend of the patient's condition change. Based on this trend, the patient's three-dimensional virtual model and condition portrait can be updated in real time. This real-time update and visual display helps to accurately reflect the patient's latest health status and provide accurate data support for subsequent calculation of condition characteristics and risk indicators.

[0076] In one embodiment, in step S21, the principal component analysis method is used to fuse the comprehensive medical data through nonlinear dimensionality reduction technology to obtain a comprehensive data set containing multi-dimensional information, including:

[0077] Step S211, preprocessing the comprehensive medical data to obtain a preprocessed data set.

[0078] Specifically, the present application will perform data cleaning, data normalization, and feature selection processing on the comprehensive medical data, wherein the data cleaning processing includes removing duplicate data, processing missing values ​​(such as using the mean or median to fill missing values), deleting invalid data (such as data that does not conform to medical logic), etc. Data normalization processing includes, for numerical data, uniformly scaling the data to the range of 0 to 1 to eliminate the differences between different dimensions; for text-type medical history information, text preprocessing and vectorization operations are used to achieve normalization of medical history information; for image-type imaging data, pixel value normalization (corresponding to grayscale images), channel normalization (corresponding to color images), and specific physical quantity range normalization (corresponding to medical images, such as CT, MRI) are used for normalization processing. Feature selection includes screening out feature variables that have significant correlation with the analysis target from the original feature set based on correlation analysis (such as Pearson correlation coefficient).

[0079] Step S212: perform linear dimensionality reduction on the preprocessed data set using a principal component analysis method to obtain a principal component set.

[0080] Specifically, the application will calculate the covariance matrix between each feature in the preprocessed data set to understand the correlation between the features. Afterwards, the covariance matrix is ​​subjected to eigendecomposition to obtain multiple eigenvalues ​​and corresponding eigenvectors. Among them, the eigenvalue represents the importance between each principal component, and the eigenvector illustrates the direction of each principal component in the original feature space. Then, according to the size of the eigenvalue, the eigenvector corresponding to the k largest eigenvalues ​​is selected, and the principal component space is formed by these k eigenvectors. Finally, after the original data set is projected onto the principal component space, the principal component set after dimensionality reduction can be obtained. The calculation of this process can be realized by calculating the dot product between each sample (i.e., observed value) and each principal component eigenvector in the original data set. Finally, a k-dimensional data set is obtained.

[0081] In one of the embodiments, k is not a fixed value determined by preset rules or experience. This application takes into account the inherent structural characteristics of the data and the analysis purpose requirements, and determines the value of k through dynamic evaluation and optimization strategies. Specifically, after obtaining the eigenvalue and eigenvector through eigendecomposition, this application will first set a corresponding k value range according to the size of the eigenvalue. Afterwards, within this k value range, the data reconstruction error analysis method is used to evaluate the data reconstruction error under different k values, and quantify the degree of information loss in the dimensionality reduction process.

[0082] It should be noted that the application of data reconstruction error analysis will further quantify the degree of information loss during the dimensionality reduction process. For each k value, this application will project the original data set into a principal component space consisting of k principal components to obtain a data set after dimensionality reduction. Afterwards, the degree of information loss during the dimensionality reduction process is further determined by calculating the reconstruction error between the original data set and the data set after dimensionality reduction (such as using the mean square error as a quantitative indicator of the reconstruction error). By comparing the reconstruction errors under different k values, the k value that minimizes the reconstruction error is used as the optimal k value.

[0083] Step S213: Based on the feature importance evaluation and the data distribution characteristics, the principal component set is weighted and nonlinearly combined to obtain a weighted nonlinear principal component set.

[0084] Specifically, this application will determine the weight of each principal component in the nonlinear combination by using the proportional distribution method according to the eigenvalue size corresponding to each principal component. For details, please refer to the formula: Among them, w i represents the weight of the i-th principal component, t i represents the eigenvalue of the i-th principal component, and n represents the total number of principal components in the principal component set.

[0085] Furthermore, according to the data distribution characteristics, taking into account the possible nonlinear relationship and nonlinear trend between the data, the present application introduces logarithmic function, periodic kernel function, sigmoid function, and polynomial function to perform nonlinear transformation on the principal components, where:

[0086] For data with an exponential growth trend (such as the changing trend of certain physiological indicators under specific disease states, etc.), this application will apply a logarithmic function to perform nonlinear transformation processing on it.

[0087] The logarithmic function can effectively compress numbers with exponential growth trends), and this application will apply the logarithmic function to perform nonlinear transformation processing on them;

[0088] For data with periodic changes (such as the circadian rhythm of certain vital signs), this application will use a periodic kernel function for nonlinear transformation processing to capture the periodic characteristics of the data;

[0089] For data that needs to be normalized to a specific range (such as quantitative indicators of imaging data), this application will apply the sigmoid function for nonlinear transformation to map the data to the required output range;

[0090] For data that present a polynomial relationship (such as the relationship between certain physiological indicators and disease severity), this application will apply a polynomial function for nonlinear transformation to capture the complex nonlinear relationship between the data.

[0091] Furthermore, the principal components obtained after nonlinear transformation (i.e., nonlinear principal components) are weighted and combined according to their corresponding weights to obtain a weighted nonlinear principal component set, where each element in the set is the result of the original data after nonlinear transformation and weighting. For details, please refer to the formula: Among them, z i represents the i-th principal component, and f(*) represents the nonlinear mapping function.

[0092] Step S214, performing deep dimensionality reduction processing on the weighted nonlinear principal component set by nonlinear dimensionality reduction technology to capture the nonlinear relationship and complex structure in the data and obtain a comprehensive data set containing multi-dimensional information.

[0093] Specifically, the present application uses the kernel principal component analysis method to perform deep dimensionality reduction processing on the weighted nonlinear principal component set to capture the nonlinear relationships and complex structures in the data and obtain a comprehensive data set containing multi-dimensional information. Among them, the application of the kernel principal component analysis method will map the original data to a high-dimensional feature space, in which the originally linearly inseparable data will become linearly separable. Then, a linear principal component analysis is performed in this space to capture the nonlinear characteristics of the data by calculating the covariance matrix of the data and finding its largest eigenvector. These nonlinear principal components can reflect the intrinsic structure of the data in a low-dimensional space, thereby achieving dimensionality reduction.

[0094] In the above embodiment, on the one hand, based on the feature importance evaluation and data distribution characteristics, the principal component set is weighted and nonlinearly combined, so that important features can be highlighted and unimportant features can be suppressed, thereby extracting more critical information. On the other hand, the weighted nonlinear principal component set is deeply reduced in dimension through nonlinear dimensionality reduction technology, so that nonlinear relationships and complex structures in the data can be captured, and more refined and in-depth information can be extracted.

[0095] In one embodiment, in step S212, the linear dimension reduction is performed on the preprocessed data set using a principal component analysis method to obtain a principal component set, including:

[0096] Step S2121, calculating the covariance matrix between the features in the preprocessed data set.

[0097] Step S2122, performing eigendecomposition on the covariance matrix to obtain multiple eigenvalues ​​and corresponding eigenvectors.

[0098] Step S2123, setting a k value range according to the size of the eigenvalue, and selecting a k value that minimizes the reconstruction error within the k value range as the optimal k value.

[0099] Step S2124: According to the optimal k value, select the eigenvectors corresponding to the k largest eigenvalues, and construct a principal component space based on these k eigenvectors.

[0100] Step S2125, projecting the preprocessed data set into the principal component space to obtain a principal component set after dimensionality reduction.

[0101] It should be noted that based on step S2121 to step S2125, the present application adopts the classic dimensionality reduction technique of principal component analysis. By calculating the covariance matrix between the features in the preprocessed data set, the linear correlation between the features is captured, and the main components in the data are identified by the eigendecomposition of the covariance matrix. These main components are represented by eigenvectors, which constitute a new, low-dimensional feature space. When selecting the optimal k value, the present application adopts the method of data reconstruction error analysis. The application of this method not only helps us determine how many principal components to retain to minimize information loss, but also ensures that the data set after dimensionality reduction can retain the structure and information of the original data as much as possible. By comparing the reconstruction errors under different k values, it is possible to find a balance point, while ensuring data quality, to achieve effective reduction of dimensions.

[0102] In the above embodiment, by setting the k value range and calculating the reconstruction error, the k value that minimizes the reconstruction error can be automatically selected as the optimal k value. This can avoid the subjectivity and uncertainty caused by manual selection of the k value, improve the accuracy and objectivity of parameter selection, and ensure the accuracy of data dimensionality reduction.

[0103] In one embodiment, in step S23, the use of time series analysis and machine learning algorithms to perform deep mining on the comprehensive data set to capture the patient's condition change trend includes:

[0104] Step S231, determining the key physiological indicators of the patient based on the comprehensive data set, and processing the key physiological indicators using time series analysis technology to obtain a corresponding dynamic change trend graph.

[0105] Specifically, first, this application will screen out key physiological indicators directly related to the patient's health status from the comprehensive data set, such as heart rate, blood pressure, blood sugar, blood oxygen saturation, etc. These indicators can reflect the patient's physiological state and its changing trends. Then, the time series analysis technology is used to perform time series analysis on the selected key physiological indicators to capture the laws and trends of the indicators over time, and then generate the corresponding dynamic change trend graph. Among them, the dynamic change trend graph can intuitively display the fluctuations of the patient's physiological indicators, providing a basis for subsequent abnormal fluctuation point detection and disease change trend prediction.

[0106] Step S232, based on the moving average and the threshold setting, analyze the abnormal fluctuation points from the dynamic change trend chart and mark the potential health risk areas.

[0107] Specifically, the present application will first smooth the dynamic change trend graph based on the moving average to reduce random fluctuations, wherein the moving average is specifically based on the average value of the data points in the window to generate a smooth curve, which can reflect the overall trend of the data series while filtering out random fluctuations in the short term. Afterwards, for each data point in the trend graph, the corresponding deviation index is obtained based on the distance between the calculated data point and the moving average, which can reflect the degree of deviation of the data point relative to the moving average. Afterwards, the threshold is set based on the historical data analysis, and when it is determined that the deviation index of the corresponding data point is greater than the set threshold, the data point is taken as an abnormal fluctuation point and recorded. Finally, based on the recorded abnormal fluctuation points, the present application will further analyze the distribution of these points, and cluster similar abnormal fluctuation points together through cluster analysis to mark potential health risk areas.

[0108] Step S233, processing abnormal fluctuation points and potential health risk areas based on machine learning algorithms to predict the patient's condition change trend.

[0109] Specifically, this application will collect data related to abnormal fluctuation points and potential health risk areas, including the values ​​of key physiological indicators, timestamps, abnormal types and other information, and divide the training set and validation set based on these data. Taking into account the temporal and complexity of the data, this application uses a machine learning model based on the long short-term memory network LSTM, and inputs the training set into the model for training. During the training process, the model will predict the patient's condition change trend based on the characteristics of abnormal fluctuation points and potential health risk areas in historical data, and their correlation with the condition change trend, and output the confidence interval of the prediction result, where the confidence interval can reflect the uncertainty of the prediction result to provide a reference for the reliability of the prediction result.

[0110] In the above embodiment, abnormal fluctuation points and potential health risk areas are processed based on machine learning algorithms to predict the patient's condition change trend. This intelligent prediction method can provide more scientific decision support for subsequent updates of the patient's dynamic condition portrait.

[0111] In one embodiment, in step S232, based on the moving average and the threshold setting, abnormal fluctuation points are analyzed from the dynamic change trend graph, and potential health risk areas are marked, including:

[0112] Step S2321, smoothing the dynamic change trend graph based on a moving average line to obtain a basic trend graph with random fluctuations filtered out.

[0113] Specifically, the present application will first determine a suitable moving average window size, such as 5 days, 10 days or longer time periods, according to the characteristics of the data and the purpose of analysis. This window size determines the number of data points involved in the average value calculation. Afterwards, slide this window along the time axis, and calculate the average value of the data points in each window. The smooth curve formed by connecting these average values ​​is the moving average. Afterwards, by linear interpolation (such as using the linear interpolation formula y=y0+α(y1-y0) to calculate the value of the interpolation point, where y0 and y1 are the values ​​of known data points, and α is an interpolation coefficient between 0 and 1, which is determined according to the position ratio relationship between the original data points and the corresponding points on the moving average) the original data points in the dynamic change trend graph are combined with the moving average to obtain a basic trend graph that is smoother and has filtered out most of the random fluctuations.

[0114] Step S2322, for each data point in the basic trend graph, calculate the distance between the data point and the moving average line to determine the deviation between the two.

[0115] Specifically, the present application uses the vertical distance method to calculate the distance between each data point and the moving average. Specifically, for each data point in the trend graph, the present application will find its corresponding point on the moving average and calculate the vertical distance between the two points. The result obtained is the deviation between the two. It should be noted that this deviation reflects the degree of deviation of the data point relative to the overall trend.

[0116] Step S2323, setting a threshold value based on historical data analysis, and when it is determined that the deviation of the corresponding data point is greater than the threshold value, the corresponding data point is regarded as an abnormal fluctuation point.

[0117] Step S2324, performing cluster analysis based on the obtained abnormal fluctuation points to identify and mark potential health risk areas.

[0118] Specifically, the present application will perform cluster analysis based on each abnormal fluctuation point obtained to obtain multiple cluster groups, wherein each cluster group represents a potential health risk area, and the area contains similar abnormal fluctuation points.

[0119] Afterwards, in order to facilitate subsequent machine learning processing, this application will vector-type mark the potential health risk areas according to the location, size and density of data points in the cluster grouping, including at least one statistical feature of the mean, variance, maximum value, and minimum value. Among them:

[0120] The mean reflects the average level of abnormal fluctuation points within the cluster grouping, which helps to understand the overall fluctuation trend of the area;

[0121] The variance measures the degree of dispersion of abnormal fluctuation points within the cluster grouping. The larger the variance, the more dispersed the fluctuation points in the area, which may indicate a higher instability of the disease.

[0122] The maximum value records the highest value of the abnormal fluctuation point within the cluster grouping, and the minimum value records the lowest value of the abnormal fluctuation point within the cluster grouping. Together with the maximum value, they can form a comprehensive description of the fluctuation range of the disease.

[0123] Through such processing, it is possible to provide rich input features for subsequent machine learning algorithms. These features can comprehensively describe the characteristics of potential health risk areas. Machine learning algorithms can use these features to train predictive models, thereby achieving accurate predictions of patient condition change trends.

[0124] In one embodiment, in step S3, the rule base is constructed based on historical data, and the patient's condition profile is matched with the rules in the rule base to obtain condition characteristics and risk indicators, including:

[0125] Step S31, using a data mining algorithm to extract historical disease information from historical data.

[0126] Specifically, the present application uses a data mining algorithm based on association rules to mine frequent item sets and calculate confidence in historical data, measures the prevalence of frequent item sets in the data set based on their support, and uses confidence to evaluate the strength and reliability of the association between item sets (such as calculating conditional probability to quantify the strength of the association from one item set (such as a specific symptom combination) to another item set (such as a potential disease)), thereby identifying the association between key information such as medical history, symptoms, and test results, and then extracting valuable historical medical information from historical data.

[0127] Step S32, generating rules for describing the correlation between historical disease information, disease characteristics, and risk indicators according to data analysis technology based on statistical learning methods.

[0128] Specifically, this application uses correlation analysis technology to calculate the relationship between historical medical information (such as medical history, symptoms, and test results) and disease characteristics (such as disease type), as well as risk indicators (such as complication risk). By calculating the correlation coefficient (the calculation of the correlation coefficient belongs to the prior art, and this application does not limit its specific calculation formula), the strength of the association between these variables is quantified, and the size of the correlation coefficient is used to generate rules for describing the relationship between historical medical information and disease characteristics, as well as risk indicators. The general generation logic of the rules includes:

[0129] If the absolute value of the correlation coefficient is greater than a preset correlation threshold (such as 0.7), it is considered that there is a significant correlation between the historical disease information and the corresponding disease characteristics and risk indicators;

[0130] If the absolute value of the correlation coefficient is close to 0, it is considered that there is no significant correlation between the historical medical information and the corresponding medical characteristics and risk indicators.

[0131] Step S33, importing the generated rules into the rule base to form a corresponding rule system.

[0132] Specifically, the present application will classify and store the generated rules to form a corresponding structured rule base, wherein each rule is assigned a unique identifier and description information to facilitate subsequent index matching.

[0133] Step S34, extracting real-time condition information from the patient's condition portrait, and performing rule-based reasoning on the real-time condition information and the various rules in the rule system to obtain target rules that meet the current patient condition.

[0134] Step S35, parsing the target rule to obtain the disease characteristics and risk indicators that are consistent with the current patient condition.

[0135] Based on steps S34 to S35, it should be noted that the present application will compare and infer the extracted real-time medical information with each rule in the rule system one by one. This reasoning process is based on rule matching and logical judgment, aiming to find the rules that best match or are most relevant to the current patient's condition. During the reasoning process, the similarity or degree of match between the real-time medical information and the historical medical information described in the rules will be considered. Once the target rule that matches the current patient's condition is found, the condition characteristics and risk indicators that match the current patient's condition will be fed back, and the patient's current risk status will be evaluated based on the condition characteristics and risk indicators mentioned in the target rule.

[0136] In the above embodiments, on the one hand, rule generation based on statistical learning methods can quantify the strength of association between historical medical information and medical characteristics and risk indicators, thereby improving the accuracy of rule matching. On the other hand, by reasoning with the rules in the rule system based on real-time medical information, the most suitable target rules can be found for each patient's specific situation, thereby realizing personalized risk assessment and early warning, and helping doctors to more accurately judge the patient's health status and potential risks.

[0137] In one embodiment, in step S4, the risk assessment criteria are established based on the disease characteristics and risk indicators, and the severity of the disease is assessed by the risk assessment criteria, and possible complications are predicted, including:

[0138] Step S41, determine the risk factors based on the disease characteristics and risk indicators, quantitatively score each risk factor through the risk matrix method, and establish a risk assessment standard based on the scoring results and expert experience. The risk assessment standard integrates the diversity, dynamic changes and potential complications of the disease, and can automatically generate a personalized risk assessment report.

[0139] Specifically, this application will first identify and determine the risk factors related to the evaluation object based on the characteristics of the disease and risk indicators. After that, the risk matrix method is used to quantitatively score each risk factor based on the possibility and severity of the risk, that is, to assign it a specific quantitative value. Finally, based on the results of the quantitative scoring and expert experience, the diversity of the disease, the dynamic characteristics of the disease, and the potential threat of complications are comprehensively considered to establish the corresponding risk assessment standards.

[0140] Step S42, using the risk assessment criteria, assessing the severity of the disease based on the real-time updated disease data, and predicting possible complications.

[0141] Specifically, this application will input the real-time updated disease data into the risk assessment model, and calculate the patient's risk score and risk level according to the risk assessment criteria. After that, the severity of the disease is assessed by comparing the preset thresholds in combination with the risk score and risk level, and possible complications are predicted based on the correlation analysis of risk factors and the statistical laws of historical data.

[0142] Please refer to Figure 2 The present application discloses an aortic dissection rescue guidance system based on multi-dimensional data fusion, the system comprising a data acquisition module, a data fusion and processing module, a rule base construction and matching module, and a condition assessment and prediction module, wherein:

[0143] The data acquisition module is used to acquire the patient's comprehensive medical data, and the comprehensive medical data includes at least one of real-time vital sign data, imaging data, and medical history information.

[0144] The data fusion and processing module is used to fuse the comprehensive medical data, and to construct a disease profile based on the fused data set, and to capture the disease change trend, so as to obtain a dynamic disease profile of the patient.

[0145] The rule base construction and matching module is used to construct a rule base based on historical data, and match the patient's condition profile with the rules in the rule base to obtain condition characteristics and risk indicators.

[0146] The condition assessment and prediction module is used to establish risk assessment standards based on condition characteristics and risk indicators, assess the severity of the condition through the risk assessment standards, and predict possible complications, so as to provide personalized rescue suggestions and decision support to medical staff.

[0147] As can be seen from the above, the aortic dissection rescue guidance system based on multi-dimensional data fusion disclosed in this application uses the integrated medical data after fusion processing to build a dynamic disease portrait, which can capture the trend of disease changes in real time, so that a comprehensive patient information view can be formed, timely reflecting the patient's real-time health status and potential risks, and providing medical staff with more accurate and timely patient status monitoring; by matching the rule base built based on historical data with the patient's disease portrait, the disease characteristics and risk indicators can be automatically identified, and then the risk assessment standard can be established. In this way, the severity of the disease can be quickly and accurately assessed, and the potential risk of complications can be predicted, thereby guiding medical staff to formulate more reasonable and effective treatment plans and improve the scientificity and efficiency of medical decision-making.

[0148] In one of the embodiments, the data fusion and processing module is also used to use the principal component analysis method to fuse the comprehensive medical data through nonlinear dimensionality reduction technology to obtain a fused data set containing multi-dimensional information; based on the fused data set, the patient's static condition portrait is constructed using augmented reality and three-dimensional modeling technology; time series analysis and machine learning algorithms are used to deeply mine the comprehensive data set to capture the patient's condition change trend; based on the condition change trend, the patient's three-dimensional virtual model and condition portrait are updated to achieve real-time update and visualization of the patient's dynamic condition portrait.

[0149] In one embodiment, the data fusion and processing module is also used to preprocess the comprehensive medical data to obtain a preprocessed data set; use the principal component analysis method to perform linear dimensionality reduction on the preprocessed data set to obtain a principal component set; based on feature importance evaluation and data distribution characteristics, the principal component set is weighted and nonlinearly combined to obtain a weighted nonlinear principal component set; and the weighted nonlinear principal component set is deeply reduced in dimensionality by nonlinear dimensionality reduction technology to capture the nonlinear relationships and complex structures in the data to obtain a comprehensive data set containing multi-dimensional information.

[0150] In one embodiment, the data fusion and processing module is also used to calculate the covariance matrix between the features in the preprocessed data set; perform eigendecomposition on the covariance matrix to obtain multiple eigenvalues ​​and corresponding eigenvectors; set the k value range according to the size of the eigenvalue, and select the k value that minimizes the reconstruction error as the optimal k value within the k value range; according to the optimal k value, select the eigenvectors corresponding to the k largest eigenvalues, and construct a principal component space based on these k eigenvectors; project the preprocessed data set into the principal component space to obtain a reduced-dimensional principal component set.

[0151] In one of the embodiments, the data fusion and processing module is also used to determine the patient's key physiological indicators based on the comprehensive data set, and use time series analysis technology to process the key physiological indicators to obtain a corresponding dynamic change trend graph; based on the moving average and threshold setting, analyze abnormal fluctuation points from the dynamic change trend graph and mark potential health risk areas; process abnormal fluctuation points and potential health risk areas based on a machine learning algorithm to predict the patient's condition change trend.

[0152] In one of the embodiments, the data fusion and processing module is also used to smooth the dynamically changing trend graph based on a moving average to obtain a basic trend graph that filters out random fluctuations; for each data point in the basic trend graph, the distance between the data point and the moving average is calculated to determine the deviation between the two; a threshold is set based on historical data analysis, and when it is determined that the deviation of the corresponding data point is greater than the threshold, the corresponding data point is used as an abnormal fluctuation point; cluster analysis is performed based on the obtained abnormal fluctuation points to identify and mark potential health risk areas.

[0153] In one of the embodiments, the rule base construction and matching module is also used to extract historical medical information from historical data using a data mining algorithm; generate rules for describing the relationship between historical medical information and medical characteristics, as well as risk indicators based on data analysis technology based on statistical learning methods; import the generated rules into the rule base to form a corresponding rule system; extract real-time medical information from the patient's medical profile, and perform rule-based reasoning on the real-time medical information and the various rules in the rule system to obtain target rules that meet the current patient's condition; and parse the target rules to obtain medical characteristics and risk indicators that meet the current patient's condition.

[0154] In one embodiment, the disease assessment and prediction module is also used to determine risk factors based on disease characteristics and risk indicators, to quantitatively score each risk factor through the risk matrix method, and to establish risk assessment standards based on the scoring results and expert experience, wherein the risk assessment standards integrate the diversity, dynamic changes and potential complications of the disease, and can automatically generate personalized risk assessment reports; through the risk assessment standards, the severity of the disease is assessed based on real-time updated disease data, and possible complications are predicted.

[0155] Furthermore, the present application also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the aortic dissection rescue guidance method based on multi-dimensional data fusion.

[0156] As can be seen from the above, the computer-readable storage medium disclosed in this application uses the integrated medical data after fusion processing to construct a dynamic disease profile, which can capture the trend of disease changes in real time, so that a comprehensive patient information view can be formed, timely reflecting the patient's real-time health status and potential risks, and providing medical staff with more accurate and timely patient status monitoring; by matching the rule base built based on historical data with the patient's disease profile, it can automatically identify disease characteristics and risk indicators, and then establish risk assessment standards. In this way, it is possible to quickly and accurately assess the severity of the disease and predict potential complications, thereby guiding medical staff to formulate more reasonable and effective treatment plans and improve the scientificity and efficiency of medical decision-making.

[0157] Furthermore, the present application also discloses an aortic dissection rescue device based on multi-dimensional data fusion, comprising a communication interface, a memory, a communication bus and a processor, wherein the processor, the communication interface and the memory communicate with each other through the communication bus;

[0158] The memory is used to store computer programs;

[0159] The processor is used to implement the steps of the aortic dissection rescue guidance method based on multi-dimensional data fusion when executing the program stored in the memory.

[0160] As can be seen from the above, the computer-readable storage medium disclosed in this application uses the integrated medical data after fusion processing to construct a dynamic disease profile, which can capture the trend of disease changes in real time, so that a comprehensive patient information view can be formed, timely reflecting the patient's real-time health status and potential risks, and providing medical staff with more accurate and timely patient status monitoring; by matching the rule base built based on historical data with the patient's disease profile, it can automatically identify disease characteristics and risk indicators, and then establish risk assessment standards. In this way, it is possible to quickly and accurately assess the severity of the disease and predict potential complications, thereby guiding medical staff to formulate more reasonable and effective treatment plans and improve the scientificity and efficiency of medical decision-making.

[0161] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0162] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A rescue guidance method for aortic dissection based on multi-dimensional data fusion, characterized in that: The method comprises: S1. Obtaining comprehensive medical data of the patient, wherein the comprehensive medical data includes at least one of real-time vital sign data, imaging data, and medical history information; S2. Fusing the comprehensive medical data, and constructing a disease profile based on the fused data set, and capturing the disease change trend, to obtain a dynamic disease profile of the patient; S3. Build a rule base based on historical data, and match the patient's condition profile with the rules in the rule base to obtain condition characteristics and risk indicators; S4. Establish risk assessment standards based on disease characteristics and risk indicators, evaluate the severity of the disease and predict possible complications through the risk assessment standards, so as to provide personalized rescue suggestions and decision support to medical staff.

2. The method according to claim 1, characterized in that: In step S2, the comprehensive medical data is fused and processed, and a disease profile is constructed based on the fused data set, and the disease change trend is captured to obtain a dynamic disease profile of the patient, including: S21, using a principal component analysis method and a nonlinear dimensionality reduction technique to fuse the comprehensive medical data to obtain a fused data set containing multi-dimensional information; S22. Based on the fused data set, construct a static condition portrait of the patient using augmented reality and three-dimensional modeling technology; S23, using time series analysis and machine learning algorithms to perform in-depth mining on the comprehensive data set to capture the changing trend of the patient's condition; S24. Update the patient's three-dimensional virtual model and condition portrait based on the disease change trend to achieve real-time update and visual display of the patient's dynamic condition portrait.

3. The method according to claim 2, characterized in that In step S21, the principal component analysis method is used to fuse the comprehensive medical data through nonlinear dimensionality reduction technology to obtain a comprehensive data set containing multi-dimensional information, including: S211, preprocessing the comprehensive medical data to obtain a preprocessed data set; S212, performing linear dimension reduction on the preprocessed data set using a principal component analysis method to obtain a principal component set; S213, based on feature importance evaluation and data distribution characteristics, weighting and nonlinearly combining the principal component set to obtain a weighted nonlinear principal component set; S214. Perform deep dimensionality reduction processing on the weighted nonlinear principal component set by nonlinear dimensionality reduction technology to capture the nonlinear relationship and complex structure in the data and obtain a comprehensive data set containing multi-dimensional information.

4. The method according to claim 3, characterized in that In step S212, the linear dimension reduction is performed on the preprocessed data set using the principal component analysis method to obtain a principal component set, including: S2121, calculating the covariance matrix between the features in the preprocessed data set; S2122, performing eigendecomposition on the covariance matrix to obtain multiple eigenvalues ​​and corresponding eigenvectors; S2123, setting a k value range according to the size of the eigenvalue, and selecting a k value that minimizes the reconstruction error within the k value range as the optimal k value; S2124, according to the optimal k value, select eigenvectors corresponding to k largest eigenvalues, and construct a principal component space based on the k eigenvectors; S2125. Project the preprocessed data set into the principal component space to obtain a principal component set after dimensionality reduction.

5. The method according to claim 2, characterized in that: In step S23, the comprehensive data set is deeply mined using time series analysis and machine learning algorithms to capture the patient's condition change trend, including: S231, determining the key physiological indicators of the patient based on the comprehensive data set, and processing the key physiological indicators using time series analysis technology to obtain a corresponding dynamic change trend graph; S232, based on the moving average and the threshold setting, analyzing the abnormal fluctuation points from the dynamic change trend graph, and marking the potential health risk areas; S233. Process abnormal fluctuation points and potential health risk areas based on machine learning algorithms to predict the changing trend of the patient's condition.

6. The method according to claim 5, characterized in that In step S232, based on the moving average and the threshold setting, abnormal fluctuation points are analyzed from the dynamic change trend chart, and potential health risk areas are marked, including: S2321, smoothing the dynamic change trend graph based on a moving average line to obtain a basic trend graph from which random fluctuations are filtered out; S2322. For each data point in the basic trend graph, calculate the distance between the data point and the moving average line to determine the deviation between the two; S2323, setting a threshold value based on historical data analysis, and when it is determined that the deviation of a corresponding data point is greater than the threshold value, treating the corresponding data point as an abnormal fluctuation point; S2324. Perform cluster analysis based on the abnormal fluctuation points obtained to identify and mark potential health risk areas.

7. The method according to claim 1, characterized in that In step S3, the rule base is constructed based on historical data, and the patient's condition profile is matched with the rules in the rule base to obtain condition characteristics and risk indicators, including: S31. Using data mining algorithms, extract historical disease information from historical data; S32. Generate rules for describing the correlation between historical disease information and disease characteristics, as well as risk indicators, based on data analysis technology based on statistical learning methods; S33, importing the generated rules into the rule base to form a corresponding rule system; S34, extracting real-time condition information from the patient's condition portrait, and performing rule-based reasoning on the real-time condition information and various rules in the rule system to obtain a target rule that meets the current patient's condition; S35. Analyze the target rule to obtain the disease characteristics and risk indicators that are consistent with the current patient condition.

8. The method according to claim 1, characterized in that In step S4, the risk assessment criteria are established based on the disease characteristics and risk indicators, and the severity of the disease is assessed by the risk assessment criteria, as well as possible complications are predicted, including: S41. Determine risk factors based on disease characteristics and risk indicators, perform quantitative scoring of each risk factor through the risk matrix method, and establish risk assessment standards based on the scoring results and expert experience, wherein the risk assessment standards integrate the diversity, dynamic changes and potential complications of the disease, and can automatically generate a personalized risk assessment report; S42. Using the risk assessment criteria, assess the severity of the disease based on real-time updated disease data, and predict possible complications.

9. An aortic dissection rescue guidance system based on multi-dimensional data fusion, characterized in that: The system includes a data acquisition module, a data fusion and processing module, a rule base construction and matching module, and a disease assessment and prediction module, wherein: The data acquisition module is used to acquire the patient's comprehensive medical data, wherein the comprehensive medical data includes at least one of real-time vital sign data, imaging data, and medical history information; The data fusion and processing module is used to fuse the comprehensive medical data, and to construct a disease profile based on the fused data set, and to capture the disease change trend, so as to obtain a dynamic disease profile of the patient; The rule base construction and matching module is used to construct a rule base based on historical data, and match the patient's condition profile with the rules in the rule base to obtain condition characteristics and risk indicators; The condition assessment and prediction module is used to establish risk assessment standards based on condition characteristics and risk indicators, assess the severity of the condition through the risk assessment standards, and predict possible complications, so as to provide personalized rescue suggestions and decision support to medical staff.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, the aortic dissection rescue guidance method based on multi-dimensional data fusion is implemented.

Images