Hierarchical medical calling method and device

Abstract

The present application discloses a hierarchical medical call method and device, which relates to the field of intelligent analysis of medical data. It first obtains the patient's current vital signs and uploads them to the blockchain network, creates blockchain sequence data based on them, compares them with the preset vital signs standard data to obtain a comparison result, and then determines the patient's condition level based on the comparison result, and finally generates and issues a corresponding call request based on the condition level. In this way, targeted medical services can be effectively provided by more accurately judging the patient's condition level.

Description

Technical Field

[0001] The present application relates to the field of intelligent analysis of medical data, and more specifically, to a hierarchical medical call method and device. Background Art

[0002] In the current medical scenario, the timeliness of responding to patient calls is related to life safety. In the past, medical call devices were mainly one-touch buttons, which immediately sent a call signal to medical staff after being pressed. Although the signal can be sent, in actual operation, this single call method lacks flexibility and is difficult to handle differently according to different levels of urgency.

[0003] In this regard, the invention with publication number CN116646066A proposes a hierarchical medical call method, system, storage medium and electronic device, which first obtains the patient's vital signs data and digitizes it and uploads it to the blockchain network; then generates a new block containing these indicators to ensure the consistency and security of the data. Subsequently, the system compares the new data with the preset standards to evaluate the level of the disease. Finally, a call request is automatically generated based on the severity of the disease, thereby improving the response efficiency and quality of medical services.

[0004] This invention directly compares each current vital sign data with the preset standard data, and determines the disease level based on whether it exceeds the set threshold. On the one hand, vital signs are a complex organic whole, and the various indicators are interrelated and influence each other. This method ignores the relationship between different vital sign indicators and fails to fully evaluate the patient's overall health status. On the other hand, the judgment method based on fixed thresholds is difficult to adapt to individual differences. The "normal" range of some patients may exceed the standard value, but it does not mean that there is a health problem, which will lead to unnecessary alarms or delayed treatment.

[0005] Therefore, an optimized hierarchical medical calling scheme is desired. Summary of the invention

[0006] In view of the shortcomings of the prior art, the present application provides a hierarchical medical calling method and device.

[0007] According to one aspect of the present application, a hierarchical medical call method is provided, which includes: obtaining a patient's current vital sign indicators and transmitting the current vital sign indicators to a blockchain network; creating corresponding blockchain sequence data based on the current vital sign indicators; comparing the blockchain sequence data with preset vital sign standard data to obtain a comparison result, and determining the patient's condition level based on the comparison result; generating and issuing a corresponding call request based on the patient's condition level; wherein comparing the blockchain sequence data with preset vital sign standard data to obtain a comparison result, and determining the patient's condition level based on the comparison result, includes:

[0008] Extracting current vital sign indicators from the blockchain sequence data;

[0009] respectively performing embedding coding on each data item in the current vital sign indicator and each data item in the vital sign standard data to obtain a set of embedding coding vectors of vital sign indicator data items and a set of embedding coding vectors of vital sign standard data items;

[0010] Performing inter-indicator context association coding of feature jump degree on the set of embedding coding vectors of the vital sign indicator data items and the set of embedding coding vectors of the vital sign standard data items respectively to obtain context association coding vectors of the vital sign indicator data items and context association coding vectors of the vital sign standard data items;

[0011] Obtaining the comparison result based on the vital sign indicator data item context-associated coding vector and the vital sign standard data item context-associated coding vector;

[0012] Based on the comparison result, the patient's condition grade is determined.

[0013] Furthermore, the preset vital signs standard data is stored in the blockchain network.

[0014] Furthermore, each data item in the current vital sign indicator is embedded and encoded to obtain a set of embedded coding vectors of the vital sign indicator data items, including: using a vital sign indicator embedding encoder based on a multi-layer perceptron to embed and encode each data item in the current vital sign indicator to obtain a set of embedded coding vectors of the vital sign indicator data items.

[0015] Furthermore, each data item in the vital signs standard data is embedded and encoded to obtain a set of embedded coding vectors of the vital signs standard data items, including: using the multi-layer perceptron-based vital signs indicator embedding encoder to embed and encode each data item in the vital signs standard data to obtain a set of embedded coding vectors of the vital signs standard data items.

[0016] Further, the set of the vital sign indicator data item embedded coding vectors and the set of the vital sign standard data item embedded coding vectors are respectively subjected to inter-indicator context association coding of feature jump degree to obtain the vital sign indicator data item context association coding vector and the vital sign standard data item context association coding vector, including:

[0017] Calculating the message transfer significance factor of each vital sign indicator data item embedding code vector in the set of vital sign indicator data item embedding code vectors to obtain a set of vital sign indicator data item message transfer significance factors;

[0018] Performing gated significant weight conversion on the set of vital sign indicator data item message transmission significant factors to obtain a set of vital sign indicator data item message transmission significant weights;

[0019] The weighted sum of the set of vital sign indicator data item embedding coding vectors is calculated based on the set of vital sign indicator data item message delivery significance weights to obtain the vital sign indicator data item context-related coding vector.

[0020] Further, calculating the message transmission significance factor of each vital sign indicator data item embedding code vector in the set of vital sign indicator data item embedding code vectors to obtain a set of vital sign indicator data item message transmission significance factors includes:

[0021] Calculating the characteristic jump degree of each vital sign indicator data item embedded coding vector in the set of vital sign indicator data item embedded coding vectors to obtain a set of vital sign indicator data item characteristic jump degrees;

[0022] Calculating the message passing space span of each vital sign indicator data item embedding coding vector in the set of vital sign indicator data item embedding coding vectors to obtain a set of vital sign indicator data item message passing space spans;

[0023] Based on the characteristic jump degree and message transmission space span of each vital sign indicator data item embedded coding vector in the set of vital sign indicator data item embedded coding vectors, the message transmission significance factor of each vital sign indicator data item embedded coding vector is calculated to obtain the set of vital sign indicator data item message transmission significance factors.

[0024] Furthermore, based on the context-associated coding vector of the vital sign indicator data item and the context-associated coding vector of the vital sign standard data item, the comparison result is obtained, including: calculating the positional difference between the context-associated coding vector of the vital sign indicator data item and the context-associated coding vector of the vital sign standard data item to obtain a vital sign fine-grained comparison semantic coding vector as the comparison result.

[0025] Furthermore, based on the comparison result, the patient's condition level is determined, including: inputting the vital sign fine-grained comparison semantic encoding vector into a classifier-based condition level labeling module to obtain the patient's condition level.

[0026] According to another aspect of the present application, a hierarchical medical call device is provided, which includes: a patient data acquisition and transmission module, which is used to acquire the patient's current vital sign indicators and transmit the current vital sign indicators to the blockchain network; a blockchain sequence data creation module, which is used to create corresponding blockchain sequence data based on the current vital sign indicators; a disease level analysis module, which is used to compare the blockchain sequence data with preset vital sign standard data to obtain a comparison result, and determine the patient's disease level based on the comparison result; a call request issuance module, which is used to generate and issue a corresponding call request based on the patient's disease level; wherein the disease level analysis module includes:

[0027] A current vital sign index extraction unit, used to extract current vital sign indicators from the blockchain sequence data;

[0028] A vital sign indicator embedding coding unit, used to embed and code each data item in the current vital sign indicator and each data item in the vital sign standard data to obtain a set of vital sign indicator data item embedding coding vectors and a set of vital sign standard data item embedding coding vectors;

[0029] A vital sign indicator context association coding unit, used to perform inter-indicator context association coding of feature jump degree on the set of vital sign indicator data item embedded coding vectors and the set of vital sign standard data item embedded coding vectors, respectively, to obtain vital sign indicator data item context association coding vectors and vital sign standard data item context association coding vectors;

[0030] A comparison result generating unit, configured to obtain the comparison result based on the context-associated coding vector of the vital sign indicator data item and the context-associated coding vector of the vital sign standard data item;

[0031] The condition level result generating unit is used to determine the condition level of the patient based on the comparison result.

[0032] Furthermore, the comparison result generating unit is used to calculate the positional difference between the context-associated coding vector of the vital sign indicator data item and the context-associated coding vector of the vital sign standard data item to obtain a vital sign fine-grained comparison semantic coding vector as the comparison result.

[0033] This application has significant technical effects due to the adoption of the above technical solutions:

[0034] The hierarchical medical call method and device provided by the present application first obtains the patient's current vital signs and uploads them to the blockchain network, and creates blockchain sequence data based on them, and then compares them with the preset vital signs standard data to obtain a comparison result, and then determines the patient's condition level based on the comparison result, and finally generates and issues a corresponding call request according to the condition level. In this way, by more accurately judging the patient's condition level, targeted medical services can be effectively provided. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0036] Figure 1 Flow chart of a hierarchical medical call method according to an embodiment of the present application.

[0037] Figure 2 Flow chart of step S130 in the hierarchical medical call method according to an embodiment of the present application.

[0038] Figure 3 Flow chart of step S133 in the hierarchical medical call method according to an embodiment of the present application.

[0039] Figure 4 4 is a block diagram of a hierarchical medical call device according to an embodiment of the present application. DETAILED DESCRIPTION

[0040] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0041] It should be noted that in this application, all actions to obtain signals, information or data are carried out in compliance with the relevant data protection laws and policies of the country where the user is located and with the authorization of the owner of the corresponding device.

[0042] In the current medical scenario, the timeliness of patient call response is crucial to life safety. Traditional medical call devices are one-touch buttons. Although they can send signals, they lack flexibility in operation and cannot be differentiated according to different levels of urgency. In this regard, the invention with publication number CN116646066A proposes a hierarchical medical call method, system, storage medium and electronic device, which first obtains the patient's vital signs data and uploads it to the blockchain network after digitization; then generates a new block containing these indicators to ensure the consistency and security of the data. Subsequently, the system compares the new data with the preset standard to evaluate the disease level. Finally, a call request is automatically generated according to the severity of the disease, thereby improving the response efficiency and the quality of medical services. However, the invention directly compares the current vital signs data with the preset standard data and determines the disease level according to whether it exceeds the threshold. There are defects: first, the relationship between the various indicators of the vital signs is ignored, and the overall health status of the patient cannot be fully evaluated; second, the judgment based on the fixed threshold is difficult to adapt to individual differences, which is easy to cause unnecessary alarms or delay treatment.

[0043] Based on the above invention, the present application proposes a hierarchical medical calling method. Figure 1 FIG. 1 is a flow chart of a hierarchical medical call method according to an embodiment of the present application. Figure 1 As shown, according to the hierarchical medical call method of the embodiment of the present application, it includes: S110, obtaining the patient's current vital signs indicators, and transmitting the current vital signs indicators to the blockchain network; S120, creating corresponding blockchain sequence data based on the current vital signs indicators; S130, comparing the blockchain sequence data with preset vital signs standard data to obtain a comparison result, and determining the patient's condition level based on the comparison result; S140, generating and issuing a corresponding call request based on the patient's condition level.

[0044] In step S110, the patient's current vital signs indicators are obtained, and the current vital signs indicators are transmitted to the blockchain network. It should be understood that the patient's current vital signs indicators are multifaceted, including but not limited to body temperature, blood pressure, respiratory rate and other data. These are basic vital signs indicators that can intuitively reflect the basic physical condition. Specifically, blood pressure data can reflect the stress state of the patient's cardiovascular system. Too high or too low may indicate disease risk; changes in respiratory rate and depth can indicate lung or other related diseases; body temperature data can reflect whether the body has inflammation, infection and other conditions. In addition, according to the specific conditions and monitoring needs of different patients, vital signs indicators may also include blood oxygen saturation, blood sugar value and other indicators. Blood oxygen saturation can measure the ability of blood to carry oxygen and can determine whether the patient is hypoxic; blood sugar value is crucial for diabetic patients or patients with abnormal blood sugar fluctuations, and can reflect the body's metabolic state. In general, obtaining the patient's current vital signs indicators can fully and real-time understand the patient's physical condition, and thus provide an important basis for determining the patient's condition level. In order to better manage and utilize the acquired patient's current vital signs indicator data, the current vital signs indicator can be transmitted to the blockchain network. The blockchain network uses decentralized distributed ledger technology, and the data is stored on multiple nodes, avoiding the risk of data loss caused by a single server failure. At the same time, the blockchain network encrypts the data through encryption technology to ensure that only authorized users can access it, greatly improving the security of the data.

[0045] The following is a detailed description of a specific implementation process of "obtaining the patient's current vital signs indicators and transmitting the current vital signs indicators to the blockchain network":

[0046] The first is the collection of vital signs, which is the foundation and starting point of the entire program. In various departments of the hospital, from general wards to intensive care units, a variety of professional medical equipment plays a key role. Taking body temperature measurement as an example, electronic thermometers can quickly and accurately sense changes in human body temperature and convert them into digital signals with their high-precision temperature sensors. Electronic sphygmomanometers use the oscillometric method or Korotkoff sound method to accurately measure systolic and diastolic blood pressure, which can provide important data reflecting the patient's cardiovascular health status. In more complex scenarios, multi-parameter monitors can continuously monitor multiple key vital signs such as heart rate, respiratory rate, and blood oxygen saturation at the same time. These devices not only have highly sensitive sensors, but also use advanced signal processing technology to ensure that the collected data is stable and reliable.

[0047] In order to realize automatic data transmission and improve data collection efficiency and accuracy, IoT technology is widely used. By embedding IoT modules in medical devices, these devices can be connected to the hospital's internal network and automatically and in real time transmitted to the hospital's information management system, greatly reducing the errors that may be caused by manual operation, while also improving the timeliness of data, allowing medical staff to obtain the latest health information of patients in the first place.

[0048] However, the raw data collected from medical devices is often not perfect. It may be mixed with various noises, which may be caused by the electronic interference of the device itself, the influence of environmental factors, or the loss of signal transmission. At the same time, there may be outliers, such as erroneous readings caused by equipment failure or data generated by sudden extreme physiological conditions of patients. In order to ensure the accuracy and reliability of subsequent data analysis, these raw data must be preprocessed. Data cleaning is one of the important links in preprocessing. It uses specific algorithms and rules to identify and remove erroneous or invalid data. For example, set a reasonable data range and mark and remove blood pressure or heart rate values ​​that are obviously beyond the normal range.

[0049] Data standardization is also an indispensable step. Due to differences in the manufacturers, models and measurement principles of different medical devices, the collected data may vary in format and unit. For example, some devices measure blood oxygen saturation as a percentage, while others may present it as a decimal; for blood pressure values, the unit labels of different devices may also be different. Through data standardization, these data can be unified into a standard format and unit, which is convenient for subsequent centralized analysis and processing. In terms of ensuring data security, encryption technology plays a vital role. Using advanced encryption algorithms, such as symmetric encryption algorithms or asymmetric encryption algorithms, vital sign indicator data is encrypted, so that even if the data is stolen during transmission, the thief cannot easily obtain the effective information, thereby protecting the patient's privacy and the security of medical data.

[0050] After completing data preprocessing, the next step is to connect the data to the blockchain network. Medical institutions face a variety of choices in this process, and need to select the appropriate blockchain platform based on their actual needs and technical strength. The public chain is highly open and decentralized, and is suitable for some medical research scenarios that require high data openness; the private chain focuses more on data privacy protection and internal management, and is suitable for data management within a single medical institution; while the alliance chain can achieve collaboration and data sharing among multiple medical institutions while ensuring data security and privacy. For most medical institutions, considering data privacy, security, and the need for collaboration with other medical institutions, the alliance chain is often a more ideal choice.

[0051] After selecting the blockchain platform, medical institutions need to deploy blockchain nodes internally. This involves the installation and configuration of a series of software and hardware facilities. In terms of hardware, high-performance servers are required to meet the computing power and storage capacity requirements of blockchain nodes. At the same time, stable network connections need to be guaranteed to ensure that nodes can communicate data efficiently with other nodes. In terms of software, the corresponding blockchain client software and smart contracts developed according to actual needs need to be installed. These smart contracts can define data access rights, storage rules, and data update mechanisms, etc., providing a more flexible and intelligent way to manage data. After the deployment is completed, these nodes are connected to the selected blockchain network so that data can be transmitted and shared in this secure and reliable network environment.

[0052] In the data transmission stage, interfaces and protocols play a key role. Specialized data interfaces and communication protocols can ensure that data can be stably and securely transmitted from medical devices or hospital information systems to blockchain nodes. RESTful API is a commonly used data interface that is simple and flexible and can easily realize data interaction between different systems. The HTTPS protocol provides encryption and security verification mechanisms for data transmission to prevent data from being stolen or tampered with during transmission. In order to further improve transmission efficiency and reduce network bandwidth pressure, data compression technology is also widely used. Through lossless compression algorithms, the volume of data is compressed without affecting data accuracy, thereby speeding up data transmission and ensuring that data can reach blockchain nodes in a timely and accurate manner.

[0053] After the data is successfully transmitted to the blockchain network, it enters the storage and management stage. The blockchain uses distributed ledger technology to store data in a specific structure on each node. Each data block contains a wealth of information, including key information such as timestamps and data hash values ​​in addition to the patient's vital signs. The timestamp can accurately record the time when the data is generated, providing an important basis for data traceability; the data hash value is generated through a specific hash algorithm, which is like the "fingerprint" of the data, with uniqueness and irreversibility. As long as there is any slight change in the data, the hash value will change accordingly, thereby ensuring that the data cannot be tampered with. In order to facilitate the query and use of data, an index mechanism will also be established. Medical staff or managers can quickly locate and obtain the corresponding vital signs data based on patient ID, time and other information, improving the efficiency of data utilization.

[0054] In terms of data management, access rights management is of utmost importance. In order to protect the privacy of patients, medical institutions have formulated strict access rights management policies. Only authorized medical staff, managers, etc. can access and use this data. The authorization process is usually based on the division of roles and permissions. For example, doctors can view and modify the vital signs data of their patients, while nurses can only view part of the data. At the same time, technologies such as multi-factor authentication are also used to further enhance the security of access and ensure that data is only used within the authorized scope.

[0055] Through the above implementation process, it is possible to obtain the patient's current vital signs indicators and transmit the current vital signs indicators to the blockchain network.

[0056] In step S120, corresponding blockchain sequence data is created based on the current vital sign indicators. It should be understood that blockchain technology has the characteristics of decentralization, immutability and traceability. The current vital sign indicators are created as blockchain sequence data, and these characteristics can be used to ensure the security and integrity of the data. In medical scenarios, the patient's vital sign data contains important privacy information. If stored in a traditional database, there is a risk of data tampering or leakage. The blockchain sequence data is stored on multiple nodes, each node stores a complete copy of the data, and the modification of the data needs to be approved by the majority of nodes, which greatly reduces the possibility of illegal tampering of the data and effectively protects the privacy of patients.

[0057] In step S130, the blockchain sequence data and the preset vital signs standard data are compared to obtain a comparison result, and based on the comparison result, the patient's condition level is determined. Specifically, in an embodiment of the present application, the preset vital signs standard data is stored in the blockchain network. It is worth noting that these preset vital signs standards are derived based on medical research and practical experience through experiments and other methods. Corresponding standards can be set according to different diseases and detection purposes, such as normal body temperature range, normal blood pressure range, normal heart rate range, etc.

[0058] Accordingly, in comparing the blockchain sequence data with the preset vital sign standard data to obtain a comparison result, and determining the patient's condition level based on the comparison result, the technical concept of the present application is to extract the current vital sign indicators from the blockchain sequence data, use the data analysis and comparison method based on deep learning to embed and encode each data item in the current vital sign indicators, and embed and encode each data item in the vital sign standard data, and then, respectively, perform inter-indicator context association of feature jump degree on the embedded features of each vital sign indicator data item and the embedded features of each vital sign standard data item, so as to obtain the comparison result based on the difference between the context association features of the vital sign indicator data item and the context association features of the vital sign standard data item, and intelligently judge the patient's condition level based on the comparison result. The present application can capture the complex relationship between vital sign indicators, not only consider the changes of a single indicator, but also analyze the interaction between multiple indicators, and provide a more comprehensive and detailed condition assessment. At the same time, it can dynamically adjust the "normal" range based on contextual associations to meet the needs of patients with special physical conditions and chronic diseases, thereby reducing fixed threshold misjudgments and improving assessment accuracy.

[0059] Specifically, Figure 2 FIG. 1 is a flowchart of step S130 in the hierarchical medical call method according to an embodiment of the present application. Figure 2 As shown, the step S130 includes: S131, extracting the current vital sign indicators from the blockchain sequence data; S132, respectively embedding and coding each data item in the current vital sign indicators and each data item in the vital sign standard data to obtain a set of vital sign indicator data item embedding coding vectors and a set of vital sign standard data item embedding coding vectors; S133, respectively performing inter-indicator context association coding of feature jump degree on the set of vital sign indicator data item embedding coding vectors and the set of vital sign standard data item embedding coding vectors to obtain vital sign indicator data item context association coding vectors and vital sign standard data item context association coding vectors; S134, obtaining the comparison result based on the vital sign indicator data item context association coding vectors and the vital sign standard data item context association coding vectors; S135, determining the patient's condition level based on the comparison result.

[0060] In step S131, the current vital sign indicators are extracted from the blockchain sequence data. It should be understood that the blockchain sequence data contains a large amount of information, including auxiliary information such as timestamps, data source identifiers, and hash values ​​in addition to vital sign indicators. Extracting the current vital sign indicators can accurately locate the key data used to determine the patient's condition level, and avoid irrelevant information interfering with subsequent comparisons and analyses. Taking the patient's blood pressure data as an example, in the blockchain sequence data, the blood pressure data will be stored with information such as measurement time and measurement equipment. When making a judgment on the level of the condition, the core is the blood pressure indicator itself. Extracting it can make subsequent comparisons and analyses more focused, and improve the accuracy and efficiency of judgment.

[0061] In step S132, each data item in the current vital sign indicator and each data item in the vital sign standard data are embedded and encoded respectively to obtain a set of embedded encoding vectors of vital sign indicator data items and a set of embedded encoding vectors of vital sign standard data items.

[0062] Specifically, in an embodiment of the present application, embedding and encoding each data item in the current vital sign indicator to obtain a set of embedding and encoding vectors of the vital sign indicator data items includes: using a vital sign indicator embedding encoder based on a multi-layer perceptron to embed and encode each data item in the current vital sign indicator to obtain a set of embedding and encoding vectors of the vital sign indicator data items. It should be understood that the current vital sign indicator usually contains a plurality of different data items, such as body temperature, heart rate, blood pressure, respiratory rate, blood oxygen saturation, etc. These data constitute a high-dimensional feature space, and different vital sign indicators may have different dimensions and ranges. Therefore, in order to convert data of different dimensions into a unified vector representation so that the computer can analyze and process it, the present application embeds and encodes each data item in the current vital sign indicator to map the high-dimensional vital sign data to a low-dimensional vector space, thereby obtaining a set of embedding and encoding vectors of the vital sign indicator data items. In particular, in a specific embodiment of the present application, each data item in the current vital sign indicator can be embedded and encoded by using a vital sign indicator embedding encoder based on a multi-layer perceptron to take each vital sign indicator as a dimension of an input vector, thereby obtaining a set of embedding encoding vectors of the vital sign indicator data items. A person skilled in the art should know that a multi-layer perceptron is a feedforward neural network, which consists of an input layer, one or more hidden layers, and an output layer. The neurons in the input layer receive raw data, such as vital sign indicator data items (body temperature, heart rate, blood pressure, etc.). The hidden layer performs a nonlinear transformation on the input data through multiple neurons, and each neuron receives the output of the neurons in the previous layer, and generates an output through weighted summation and an activation function (such as ReLU-corrected linear unit, Sigmoid function, etc.). The output layer outputs the final result according to the task requirements, for example, outputting an embedded encoding vector in an embedded coding task, so that the specific features of each data item can be learned and the potential value of each indicator for disease assessment can be mined.

[0063] Specifically, in an embodiment of the present application, embedding and encoding each data item in the vital sign standard data to obtain a set of embedding and encoding vectors of the vital sign standard data items includes: using the multi-layer perceptron-based vital sign indicator embedding encoder to embed and encode each data item in the vital sign standard data to obtain a set of embedding and encoding vectors of the vital sign standard data items. It should be understood that each data item in the vital sign standard data reflects a typical pattern in a healthy or diseased state, and different vital sign indicators may have different units and ranges. Therefore, in order to convert all data into a vector form in a unified format for subsequent processing and comparison, in the technical solution of the present application, embedding and encoding each data item in the vital sign standard data to obtain a set of embedding and encoding vectors of the vital sign standard data items. In particular, in a specific embodiment of the present application, the multi-layer perceptron-based vital sign indicator embedding encoder is used to embed and encode each data item in the vital sign standard data to effectively encode each data item in the standard data into a vector form using a multi-layer perceptron to obtain a set of embedding and encoding vectors of the vital sign standard data items.

[0064] In step S133, the set of vital sign indicator data item embedded coding vectors and the set of vital sign standard data item embedded coding vectors are respectively subjected to inter-indicator context association coding of feature jump degree to obtain vital sign indicator data item context association coding vectors and vital sign standard data item context association coding vectors. In particular, the processing process of the set of vital sign indicator data item embedded coding vectors is taken as an example for specific description. Specifically, Figure 3 FIG. 1 is a flowchart of step S133 in the hierarchical medical call method according to an embodiment of the present application. Figure 3 As shown, the step S133 includes: S1331, calculating the message transmission significance factor of each vital sign indicator data item embedded coding vector in the set of vital sign indicator data item embedded coding vectors to obtain a set of vital sign indicator data item message transmission significance factors; S1332, performing gated significance weight conversion on the set of vital sign indicator data item message transmission significance factors to obtain a set of vital sign indicator data item message transmission significance weights; S1333, using the set of vital sign indicator data item message transmission significance weights, calculating the weighted sum of the set of vital sign indicator data item embedded coding vectors to obtain the vital sign indicator data item context-associated coding vector.

[0065] It should be understood that although simple embedded coding can extract the features of data, it may not be able to fully explore the deep contextual associations between indicators. There are complex physiological and pathological associations between vital sign indicators, and these associations may change in different situations (such as different stages of the disease, different treatment stages, etc.). Therefore, in order to more deeply explore the causal relationship or coordinated change relationship between indicators in a specific situation, the present application uses the contextual association coding mechanism between indicators of feature jump degree to process the set of embedded coding vectors of the vital sign indicator data items and the set of embedded coding vectors of the vital sign standard data items respectively to obtain the contextual association coding vectors of the vital sign indicator data items and the contextual association coding vectors of the vital sign standard data items. That is, this mechanism, with the help of a unique information transmission mechanism and an evaluation of feature jumpiness, can effectively integrate the contents of each data item, and then fully display the mutual influence between the spans of different data items, and provide accurate data support for the subsequent determination of the severity of the disease.

[0066] Specifically, in the embodiment of the present application, the step S1331 includes: calculating the characteristic jump degree of each vital sign indicator data item embedded coding vector in the set of vital sign indicator data item embedded coding vectors to obtain a set of vital sign indicator data item characteristic jump degrees. The process can be expressed as:

[0067] ;

[0068] ;

[0069] ;

[0070] in, is a set of embedding coding vectors of the vital sign indicator data items, , , and are respectively the first, second, and third in the set of embedding coding vectors of the vital sign indicator data items. and The vital sign indicator data items are embedded in the coding vector, yes Middle The eigenvalues ​​at the positions, yes The number of eigenvalues ​​in , and They are the weighted average eigenvalues ​​of the vital sign indicator data items. and The weighted average eigenvalue of each vital sign indicator data item, It is the first in the set of characteristic jump degrees of vital sign indicator data items. The characteristic jump degree of each vital sign indicator data item;

[0071] The message passing space span of each vital sign indicator data item embedding coding vector in the set of vital sign indicator data item embedding coding vectors is calculated to obtain a set of vital sign indicator data item message passing space spans. This process can be expressed as:

[0072] ;

[0073] in, express arrive The number of message delivery times, is the first in the set of message transmission space spans of vital sign indicator data items The message transmission space span of each vital sign indicator data item;

[0074] Based on the characteristic jump degree and message transmission space span of each vital sign indicator data item embedding coding vector in the set of vital sign indicator data item embedding coding vectors, the message transmission significance factor of each vital sign indicator data item embedding coding vector is calculated to obtain the set of vital sign indicator data item message transmission significance factors. This process can be expressed as:

[0075] ;

[0076] in, It is the first in the set of characteristic jump degrees of vital sign indicator data items. The characteristic jump degree of each vital sign indicator data item, and is the modulation parameter, is the first in the set of message transmission space spans of vital sign indicator data items The message transmission space span of vital sign indicator data items, It is the first in the set of significant factors of message transmission of vital sign indicator data items. The message transmission significance factor of each vital sign indicator data item.

[0077] It should be understood that the various indicators of a patient's vital signs do not exist in isolation, but are complexly interrelated. In certain disease states, changes in one indicator may trigger a chain reaction of other indicators. By calculating the characteristic jump degree of each vital sign indicator data item embedded in the encoding vector, it helps to reveal these potential associations and the mutation characteristics between indicators, which helps the model to have a more comprehensive understanding of the patient's condition, and thus provides a strong basis for the classification and diagnosis of the condition.

[0078] Accordingly, considering that the relationship between the patient's vital signs indicators is complex, it is not limited to simple direct associations. Traditional analysis based on adjacency relationships only focuses on the connection between adjacent indicators, ignoring broader potential associations. Based on this, the present application can break through this limitation by calculating the message transmission space span of each vital sign indicator data item embedded in the coding vector, and consider the connectivity between vital sign indicators from a more macro perspective. For example, when analyzing the vital signs of patients with cardiovascular disease, there may be indirect long-distance dependencies between blood pressure, heart rate and blood oxygen saturation. By calculating the message transmission space span, these hidden association paths can be discovered, rather than just being limited to the influence between directly adjacent indicators. Moreover, the patient's vital signs change dynamically. During the treatment process, by continuously calculating the message transmission space span, the changes in the relationship between vital sign indicators can be discovered in a timely manner, and the disease classification can be dynamically adjusted.

[0079] It should be understood that human vital signs are a complex system, and each indicator has different significance for disease judgment. The characteristic jump degree of the embedding coding vector of each vital sign indicator data message transmission space span reflects the degree of change of the vital sign indicator, while the message transmission space span of each vital sign indicator data item embedding coding vector reflects the correlation range and path of the vital sign indicator. By combining the information of these two data to calculate the message transmission significance factor, the vital sign indicators that play a key role in disease judgment can be accurately located. In other words, disease assessment and grading rely on comprehensive and accurate analysis of vital signs. The calculated message transmission significance factor can assign importance scores to each vital sign indicator, comprehensively reflecting the role of the indicator in information dissemination, which helps to more accurately assess the severity of the disease, reasonably divide the disease level, and avoid misjudgment of the disease due to incomplete indicator analysis.

[0080] Then, the set of vital sign indicator data item message transmission significance factors is subjected to gated significance weight conversion to obtain a set of vital sign indicator data item message transmission significance weights. The above process can be expressed as:

[0081] ;

[0082] in, It is the first in the set of significant factors of message transmission of vital sign indicator data items. The message transmission significance factor of each vital sign indicator data item, is the normalization function, is a masking operation, is a predetermined threshold, is the first in the set of message transmission significance weights of vital sign indicator data items The message delivery significance weight of each vital sign indicator data item.

[0083] It should be understood that the set of message transmission significance factors of vital sign indicator data items contains a lot of information about the importance of each vital sign indicator in information dissemination, but not all information has the same value in judging the condition. Different vital sign indicators contribute to different degrees to the judgment of the condition at different stages of the condition and in individual cases. Through the gated significant weight conversion, the concepts of forget gate and input gate in recurrent neural network can be borrowed to selectively emphasize or suppress certain information. That is, after obtaining the set of message transmission significance weights of vital sign indicator data items through gated significant weight conversion, the model can accurately adjust the information flow according to these weights. When analyzing vital sign data, higher weights are given to information that is closely related to the current condition judgment, so that it can be more fully reflected in subsequent calculations; while for less relevant information, its weight is reduced to reduce its interference with the final result.

[0084] Finally, the weighted sum of the set of vital sign indicator data item embedding coding vectors is calculated based on the set of vital sign indicator data item message transmission significance weights to obtain the vital sign indicator data item context-related coding vector. The above process can be expressed as:

[0085] ;

[0086] in, is the first in the set of embedding coding vectors of the vital sign indicator data item The vital sign indicator data items are embedded in the coding vector, is the first in the set of message transmission significance weights of vital sign indicator data items The message transmission significance weight of each vital sign indicator data item is is the number of vectors in the set of encoding vectors into which the vital sign indicator data items are embedded, is the context-associated coding vector of the vital sign indicator data item.

[0087] It should be understood that the set of embedded coding vectors of vital sign indicator data items contains rich information about the patient's vital signs, and each vector represents the characteristics of different indicators. However, this information is relatively scattered, and it is difficult to fully grasp the condition by analyzing it separately. By calculating the weighted sum with the set of significant weights of vital sign indicator data item message transmission, multi-source information can be integrated. For example, the embedded coding vectors of different indicators such as heart rate, blood pressure, and body temperature are weighted and summed according to their respective significant weights, and the importance of each indicator in information dissemination is comprehensively considered. The obtained context-related coding vector of the vital sign indicator data item is a "comprehensive condition portrait" that can comprehensively reflect the patient's overall health status and provide a more complete basis for condition classification.

[0088] In step S134, the comparison result is obtained based on the context-associated coding vector of the vital sign indicator data item and the context-associated coding vector of the vital sign standard data item. Specifically, in the embodiment of the present application, the step S134 includes: calculating the position difference between the context-associated coding vector of the vital sign indicator data item and the context-associated coding vector of the vital sign standard data item to obtain the vital sign fine-grained comparison semantic coding vector as the comparison result. It should be understood that each position in the context-associated coding vector of the vital sign indicator data item and the context-associated coding vector of the vital sign standard data item may represent a different vital sign feature dimension. For example, a certain position may represent the correlation feature of heart rate and blood pressure, and another position may represent the correlation feature of body temperature and respiratory rate). Therefore, in order to accurately capture the difference between the current vital sign and the standard vital sign in each specific feature dimension, in the technical solution of the present application, the position difference between the context-associated coding vector of the vital sign indicator data item and the context-associated coding vector of the vital sign standard data item is calculated to obtain the vital sign fine-grained comparison semantic coding vector as the comparison result. That is, this fine-grained analysis helps to discover some subtle but potentially critical changes in the diagnosis of the disease. For example, in the early stages of some diseases, the overall vital signs may appear to be close to the standard, but there may be slight differences in some specific associated feature dimensions. Differentiation by location can highlight these local differences, thereby more accurately identifying changes in the patient's condition.

[0089] In step S135, the patient's condition level is determined based on the comparison result. Specifically, in an embodiment of the present application, step S135 includes: inputting the vital sign fine-grained comparison semantic coding vector into a classifier-based condition level labeling module to obtain the patient's condition level. That is, the classifier is good at dividing the input data into different categories. The vital sign fine-grained comparison semantic coding vector contains the difference information between the current vital signs and the standard vital signs in each fine-grained feature dimension, which provides rich and targeted features for the classification of the condition level. The classifier can learn based on these features and identify different patterns, thereby corresponding them to corresponding condition level categories. For example, a decision tree classifier can classify the vital signs into primary, secondary or tertiary condition levels according to different feature dimensions in the fine-grained comparison semantic coding vector through a series of judgment conditions, thereby accurately assessing the severity of the condition. In a specific embodiment of the present application, the vital sign fine-grained contrast semantic coding vector is input into a classifier-based condition level labeling module to obtain the patient's condition level, including: using the fully connected layer of the classifier to fully connect the vital sign fine-grained contrast semantic coding vector to obtain a vital sign fine-grained contrast semantic fully connected coding feature vector; inputting the vital sign fine-grained contrast semantic fully connected coding feature vector into the Softmax classification function of the classifier to obtain the patient's condition level.

[0090] In particular, since the set of embedding coding vectors of the vital sign indicator data items and the set of embedding coding vectors of the vital sign standard data items respectively represent the embedded coding features of each data item in the current vital sign indicator and each data item in the vital sign standard data, when performing contextual association coding between indicators of feature jump degree, the data embedding coding representation in the local sample space will have a micro jump degree-macro sequence message transmission deviation relative to the global sample space, and will be introduced into the differential feature space between the contextual association coding vector of the vital sign indicator data item and the contextual association coding vector of the vital sign standard data item, thereby causing a dynamic deviation in the mapping of features to class target probabilities when the vital sign fine-grained contrast semantic coding vector is input into the classifier-based disease level labeling module, thereby reducing the accuracy of the obtained patient disease level.

[0091] Preferably, inputting the vital sign fine-grained contrast semantic coding vector into a classifier-based disease condition labeling module to obtain the disease condition grade of the patient comprises:

[0092] Determine the label probability value corresponding to each disease level label obtained by inputting the vital sign fine-grained contrast semantic encoding vector into the disease level labeling module based on the classifier , and calculate the probability value of each label The square root of the sum of the squares of is used to obtain the semantic probability value of the fine-grained comparison of vital signs. The process can be expressed as:

[0093] ;

[0094] in, Represents the label probability value corresponding to each disease level label, Indicates the total number of disease level labels, Indicates the semantic probability value of fine-grained comparison of vital signs;

[0095] The feature mean of the vital sign fine-grained contrast semantic coding vector is multiplied by the vital sign fine-grained contrast semantic probability value to obtain the vital sign fine-grained contrast semantic statistical field value. This process can be expressed as:

[0096] ;

[0097] in, Represents the semantic probability value of fine-grained comparison of vital signs, represents the feature mean of the fine-grained contrast semantic encoding vector of the vital sign, Representing the fine-grained comparison semantic statistical field value of the vital signs;

[0098] Subtract one from the vital sign fine-grained comparison semantic statistical field value and divide it by the vital sign fine-grained comparison semantic statistical field value to obtain the vital sign fine-grained comparison semantic partial probability value. This process can be expressed as:

[0099] ;

[0100] in, represents the semantic statistical field value of the fine-grained comparison of the vital signs, represents the probability value of the fine-grained comparison semantics of the vital sign;

[0101] Calculate the fine-grained contrast semantic encoding vector of the vital sign with the power function of the fine-grained contrast semantic partial probability value of the vital sign as the exponent , and multiplied by the probability value of the fine-grained comparison semantics of the vital signs to obtain the micro-representation vector of the fine-grained comparison semantics of the vital signs. The process can be expressed as:

[0102] ;

[0103] in, It means point multiplication by position. represents the fine-grained contrast semantic encoding vector of the vital signs, represents the probability value of the fine-grained comparison semantics of the vital signs, Representation calculation by is the exponential power function, A fine-grained contrast semantic micro-representation vector representing the vital sign;

[0104] After multiplying the fine-grained contrast semantic encoding vector of the vital signs by the fine-grained contrast semantic partial probability value of the vital signs, an exponential function with a natural constant as the base is calculated to obtain the fine-grained contrast semantic macro mapping vector of the vital signs. This process can be expressed as:

[0105] ;

[0106] in, It means point multiplication by position. represents the fine-grained contrast semantic encoding vector of the vital signs, represents the probability value of the fine-grained comparison semantics of the vital signs, represents an exponential function with the natural constant e as the base, A macroscopic mapping vector representing the fine-grained contrast semantics of the vital signs;

[0107] After calculating the base 2 logarithm of the vital sign fine-grained contrast semantic micro-representation vector, the vector is weighted and summed with the vital sign fine-grained contrast semantic macro-mapping vector to obtain an optimized vital sign fine-grained contrast semantic encoding vector. This process can be expressed as:

[0108] ;

[0109] in, and They represent point-by-point multiplication and point-by-point addition, respectively. and represents the weighted hyperparameter, Represents the fine-grained contrast semantic micro-representation vector of the vital signs, represents the logarithmic function value with base 2, represents the macroscopic mapping vector of the fine-grained contrast semantics of the vital signs, A vector representing the optimized fine-grained contrast semantic encoding of vital signs;

[0110] The optimized vital sign fine-grained contrast semantic encoding vector is input into the classifier-based condition level labeling module to obtain the patient's condition level.

[0111] That is, the partial low-order derivatives of the statistical distribution field corresponding to the fine-grained contrast semantic coding vector of the vital signs are used as non-overlapping macro-feature representation behavior patches of the fine-grained contrast semantic coding vector of the vital signs, so as to organize the space of different macro-behavior patches under the non-isotropic backbone structure of the fine-grained contrast semantic coding vector of the vital signs, so as to strengthen the dynamic sensitivity of the long-series micro-complex information distribution of the fine-grained contrast semantic coding vector of the vital signs to the macro-representation behavior of class probability, thereby promoting the iterative dynamic consistency of the class target between the classification target and the extracted features during the feature space-class probability mapping, so as to improve the accuracy of the patient's condition level obtained by inputting the fine-grained contrast semantic coding vector of the vital signs into the classifier-based condition level labeling module.

[0112] In summary, step S130 is explained clearly, which uses a data analysis and comparison method based on deep learning to embed and encode each data item in the current vital sign indicator, and embeds and encodes each data item in the vital sign standard data, and then, respectively, performs inter-indicator context association of feature jump degree for the embedded features of each vital sign indicator data item and the embedded features of each vital sign standard data item, so as to obtain the comparison result based on the difference between the context association features of the vital sign indicator data item and the context association features of the vital sign standard data item, and intelligently judge the patient's condition level based on the comparison result. In this way, by capturing the complex relationship between the vital sign indicators, a more comprehensive and detailed condition assessment can be achieved. At the same time, the "normal" range can be dynamically adjusted in combination with context association to meet the needs of patients with special physiques and chronic diseases, thereby reducing fixed threshold misjudgments and improving assessment accuracy.

[0113] In step S140, based on the patient's condition level, a corresponding call request is generated and issued. It should be understood that the condition level is a quantitative assessment of the patient's health status, and different levels correspond to different urgency and medical needs. Issuing a call request according to the condition level can ensure that medical staff quickly know the patient's urgency and specific needs, and achieve accurate response. For example, a level 1 condition means that the situation is critical. At this time, the level 1 call request issued will contain detailed patient information, high-level condition identification and emergency call type, allowing medical staff to quickly prepare emergency equipment and rush to the scene to strive for precious rescue time; level 2 and level 3 conditions are handled in turn according to the degree of urgency, ensuring that patients with different conditions can receive timely and appropriate medical services. Moreover, clear and definite call requests can reduce the time for medical staff to judge the condition and prepare for work, which is conducive to improving the efficiency of medical services. That is, after receiving the call, the medical staff will plan the treatment plan in advance based on the condition level and related information in the request, and can start treatment on site to avoid delays caused by unclear information.

[0114] In summary, the hierarchical medical call method based on the embodiment of the present application is explained, which first obtains the patient's current vital signs and uploads them to the blockchain network, and creates blockchain sequence data based on them, and then compares them with the preset vital signs standard data to obtain a comparison result, and then determines the patient's condition level based on the comparison result, and finally generates and issues a corresponding call request according to the condition level. In this way, by more accurately judging the patient's condition level, targeted medical services can be effectively provided.

[0115] Figure 4 FIG. 1 is a block diagram of a hierarchical medical call device according to an embodiment of the present application. Figure 4 As shown, according to the embodiment of the present application, the hierarchical medical call device 100 includes: a patient data acquisition and transmission module 110, which is used to acquire the patient's current vital sign indicators and transmit the current vital sign indicators to the blockchain network; a blockchain sequence data creation module 120, which is used to create corresponding blockchain sequence data based on the current vital sign indicators; a disease level analysis module 130, which is used to compare the blockchain sequence data with preset vital sign standard data to obtain a comparison result, and determine the patient's disease level based on the comparison result; a call request issuance module 140, which is used to generate and issue a corresponding call request based on the patient's disease level; wherein the disease level analysis module 130 includes: a current vital sign indicator extraction unit, which is used to extract the current vital sign indicators from the blockchain sequence data; a vital sign indicator embedding coding unit, which is used to respectively Embedding and encoding are performed on each data item in the current vital sign indicator and each data item in the vital sign standard data to obtain a set of embedded coding vectors of vital sign indicator data items and a set of embedded coding vectors of vital sign standard data items; a vital sign indicator context association coding unit is used to perform inter-indicator context association coding of feature jump degree on the set of embedded coding vectors of vital sign indicator data items and the set of embedded coding vectors of vital sign standard data items respectively to obtain context association coding vectors of vital sign indicator data items and context association coding vectors of vital sign standard data items; a comparison result generating unit is used to obtain the comparison result based on the context association coding vectors of vital sign indicator data items and the context association coding vectors of vital sign standard data items; a condition level result generating unit is used to determine the condition level of the patient based on the comparison result.

[0116] Here, those skilled in the art will appreciate that the specific functions and operations of the various units and modules in the hierarchical medical call device 100 have been described in detail above. Figures 1 to 3 The hierarchical medical calling method has been described in detail in the description of the hierarchical medical calling method, and therefore, its repeated description will be omitted.

[0117] In summary, the hierarchical medical call device 100 based on the embodiment of the present application is explained, which first obtains the patient's current vital sign indicators and uploads them to the blockchain network, and creates blockchain sequence data based on them, and then compares them with the preset vital sign standard data to obtain a comparison result, and then determines the patient's condition level based on the comparison result, and finally generates and issues a corresponding call request according to the condition level. In this way, by more accurately judging the patient's condition level, targeted medical services can be effectively provided.

Claims

1. A hierarchical medical call approach, including: Obtaining the patient's current vital signs indicators and transmitting the current vital signs indicators to the blockchain network; Based on the current vital sign indicators, create corresponding blockchain sequence data; Comparing the blockchain sequence data with preset vital sign standard data to obtain a comparison result, and determining the patient's condition level based on the comparison result; Based on the patient's condition level, a corresponding call request is generated and issued; characterized in that the blockchain sequence data is compared with preset vital sign standard data to obtain a comparison result, and based on the comparison result, the patient's condition level is determined, including: Extracting current vital sign indicators from the blockchain sequence data; respectively performing embedding coding on each data item in the current vital sign indicator and each data item in the vital sign standard data to obtain a set of embedding coding vectors of vital sign indicator data items and a set of embedding coding vectors of vital sign standard data items; Performing inter-indicator context association coding of feature jump degree on the set of embedding coding vectors of the vital sign indicator data items and the set of embedding coding vectors of the vital sign standard data items respectively to obtain context association coding vectors of the vital sign indicator data items and context association coding vectors of the vital sign standard data items; Obtaining the comparison result based on the vital sign indicator data item context-associated coding vector and the vital sign standard data item context-associated coding vector; Determining the patient's condition grade based on the comparison result; The method of performing inter-indicator context association coding of feature jump degree on the set of embedded coding vectors of the vital sign indicator data items and the set of embedded coding vectors of the vital sign standard data items respectively to obtain the vital sign indicator data item context association coding vector and the vital sign standard data item context association coding vector comprises: Calculating the message transfer significance factor of each vital sign indicator data item embedding code vector in the set of vital sign indicator data item embedding code vectors to obtain a set of vital sign indicator data item message transfer significance factors; Performing gated significant weight conversion on the set of vital sign indicator data item message transmission significant factors to obtain a set of vital sign indicator data item message transmission significant weights; The weighted sum of the set of vital sign indicator data item embedding coding vectors is calculated based on the set of vital sign indicator data item message delivery significance weights to obtain the vital sign indicator data item context-related coding vector.

2. The hierarchical medical calling method according to claim 1, characterized in that: The preset vital sign standard data is stored in the blockchain network.

3. The hierarchical medical calling method according to claim 2, characterized in that: Embedding and encoding each data item in the current vital sign indicator to obtain a set of embedded coding vectors of the vital sign indicator data items, including: using a vital sign indicator embedding encoder based on a multi-layer perceptron to embedding and encoding each data item in the current vital sign indicator to obtain a set of embedded coding vectors of the vital sign indicator data items.

4. The hierarchical medical calling method according to claim 3, characterized in that: Embedding and encoding each data item in the vital signs standard data to obtain a set of embedded coding vectors of the vital signs standard data items, including: using the multi-layer perceptron-based vital signs indicator embedding encoder to embed and encode each data item in the vital signs standard data to obtain a set of embedded coding vectors of the vital signs standard data items.

5. The hierarchical medical calling method according to claim 4, characterized in that: Calculating the message transfer significance factor of each vital sign indicator data item embedding code vector in the set of vital sign indicator data item embedding code vectors to obtain a set of vital sign indicator data item message transfer significance factors, including: Calculating the characteristic jump degree of each vital sign indicator data item embedded coding vector in the set of vital sign indicator data item embedded coding vectors to obtain a set of vital sign indicator data item characteristic jump degrees; Calculating the message passing space span of each vital sign indicator data item embedding coding vector in the set of vital sign indicator data item embedding coding vectors to obtain a set of vital sign indicator data item message passing space spans; Based on the characteristic jump degree and message transmission space span of each vital sign indicator data item embedded coding vector in the set of vital sign indicator data item embedded coding vectors, the message transmission significance factor of each vital sign indicator data item embedded coding vector is calculated to obtain the set of vital sign indicator data item message transmission significance factors.

6. The hierarchical medical calling method according to claim 5, characterized in that: Based on the context-associated coding vector of the vital sign indicator data item and the context-associated coding vector of the vital sign standard data item, the comparison result is obtained, including: calculating the positional difference between the context-associated coding vector of the vital sign indicator data item and the context-associated coding vector of the vital sign standard data item to obtain a vital sign fine-grained comparison semantic coding vector as the comparison result.

7. The hierarchical medical calling method according to claim 6, characterized in that: Based on the comparison result, the patient's condition level is determined, including: inputting the vital sign fine-grained comparison semantic encoding vector into a classifier-based condition level labeling module to obtain the patient's condition level.

8. A hierarchical medical calling device, used to execute the hierarchical medical calling method according to claim 1, comprising: A patient data acquisition and transmission module is used to acquire the patient's current vital signs and transmit the current vital signs to the blockchain network; A blockchain sequence data creation module, used to create corresponding blockchain sequence data based on the current vital sign indicators; A disease level analysis module is used to compare the blockchain sequence data with preset vital sign standard data to obtain a comparison result, and determine the patient's disease level based on the comparison result; a call request issuing module is used to generate and issue a corresponding call request based on the patient's disease level; characterized in that the disease level analysis module includes: A current vital sign index extraction unit, used to extract current vital sign indicators from the blockchain sequence data; A vital sign indicator embedding coding unit, used to embed and code each data item in the current vital sign indicator and each data item in the vital sign standard data to obtain a set of vital sign indicator data item embedding coding vectors and a set of vital sign standard data item embedding coding vectors; A vital sign indicator context association coding unit, used to perform inter-indicator context association coding of feature jump degree on the set of vital sign indicator data item embedded coding vectors and the set of vital sign standard data item embedded coding vectors, respectively, to obtain vital sign indicator data item context association coding vectors and vital sign standard data item context association coding vectors; A comparison result generating unit, configured to obtain the comparison result based on the context-associated coding vector of the vital sign indicator data item and the context-associated coding vector of the vital sign standard data item; The condition level result generating unit is used to determine the condition level of the patient based on the comparison result.

9. The hierarchical medical call device according to claim 8, characterized in that: The comparison result generating unit is used to calculate the position difference between the context-associated coding vector of the vital sign indicator data item and the context-associated coding vector of the vital sign standard data item to obtain a vital sign fine-grained comparison semantic coding vector as the comparison result.

Images