Message ferrying-based secure communication method and device for mobile terminal of internet hospital

Abstract

The present application relates to the field of medical information technology, and especially relates to an internet hospital mobile terminal secure communication method and device based on message ferrying, the steps of the method comprising: obtaining a medical data interaction request initiated by a mobile terminal, performing data interaction request cluster classification according to the semantic correlation degree of the medical data interaction request; obtaining patient medical record data corresponding to the data interaction request cluster and performing illness evolution path analysis, determining a data pre-caching strategy according to the illness evolution path and evolution confidence; performing feature analysis on the data interaction request cluster and the ferrying node respectively and determining a target ferrying node through feature matching; and executing the data pre-caching strategy through the target ferrying node and performing message ferrying. The present application optimizes the data pre-caching strategy of the ferrying node, realizes safe, efficient and low-delay data interaction oriented to a medical scene, and effectively improves the data transmission efficiency of the internet hospital mobile terminal.

Description

Technical Field

[0001] This invention relates to the field of medical information technology, and in particular to a secure communication method and device for mobile terminals of Internet hospitals based on message transfer. Background Technology

[0002] With the rapid development of internet-based healthcare services, mobile-based internet hospitals have become an important form of extending medical resources. Users are increasingly using mobile devices to conduct remote consultations, obtain prescriptions, and search for treatment suggestions.

[0003] Traditional medical data exchange methods typically employ a centralized server architecture, which is prone to increased latency due to network congestion or single points of failure. Furthermore, medical data is highly private and complex. Therefore, introducing a message transfer mechanism can effectively alleviate the communication bottlenecks and privacy risks associated with mobile terminals directly accessing the central node of the medical system.

[0004] However, the data pre-caching strategy of existing message ferrying mechanisms is usually statically configured or preloaded based on historical average access frequency. It lacks specific optimization for medical scenarios and fails to effectively consider the semantic features of medical data interaction requests, the evolution trend of individual patients' conditions, and the time sensitivity of medical data itself, resulting in a decrease in medical data cache hit rate and an increase in response latency. Summary of the Invention

[0005] To overcome the defects and shortcomings of existing technologies, this invention provides a secure communication method and device for mobile terminals of Internet hospitals based on message ferrying. By optimizing the data pre-caching strategy of ferry nodes, it achieves secure, efficient and low-latency data interaction for medical scenarios, effectively improving the data transmission efficiency of mobile terminals of Internet hospitals.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] In a first aspect, the present invention provides a secure communication method for mobile terminals in an internet hospital based on message ferrying, comprising: S100, acquiring medical data interaction requests initiated by the mobile terminal, and classifying data interaction request clusters according to the semantic relevance of the medical data interaction requests; S200, acquiring patient medical record data corresponding to the data interaction request clusters and performing disease evolution path analysis, and determining a data pre-caching strategy based on the disease evolution path and evolution confidence; S300, performing feature analysis on the data interaction request clusters and ferrying nodes respectively, and determining the target ferrying node through feature matching; S400, executing the data pre-caching strategy and performing message ferrying through the target ferrying node.

[0008] Furthermore, the specific steps for classifying data interaction request clusters in step S100 include:

[0009] S110. Semantic parsing of each medical data interaction request is performed using a natural language processing model, and semantic feature vectors of each medical data interaction request are extracted.

[0010] S120. Determine the semantic correlation degree between each medical data interaction request based on the semantic feature vector. The semantic correlation degree is the cosine similarity between the semantic feature vectors of the medical data interaction requests.

[0011] S130. Medical data interaction requests with a semantic relevance greater than a preset semantic relevance threshold are divided into data interaction request clusters.

[0012] Furthermore, the specific steps for determining the data pre-caching strategy in step S200 include:

[0013] S210. Obtain the patient's medical record data corresponding to the data interaction request cluster;

[0014] S220. Construct a tree-like path diagram of the patient's disease evolution based on the patient's medical record data. The nodes in the disease evolution tree-like path diagram represent the disease evolution state, and the edges represent the disease evolution direction.

[0015] S230. Determine the evolution confidence of each disease evolution path in the disease evolution tree diagram and take the disease evolution path with the highest evolution confidence as the target disease evolution path.

[0016] S240. Use the disease evolution data corresponding to the target disease evolution path as pre-cached data and determine the cache refresh time window of the pre-cached data.

[0017] Furthermore, the specific steps for determining the evolutionary confidence level in step S230 include:

[0018] S231. Extract the disease evolution paths from the disease evolution tree diagram and calculate the joint transition probability of all adjacent nodes on each disease evolution path:

[0019] ;

[0020] in, For the first The joint transition probability of all adjacent nodes on the disease evolution path. For the node Transfer to node The probability, Indicates the first The number of nodes in each disease evolution path;

[0021] S232. The joint transition probability of all adjacent nodes on the disease evolution path is used as the evolution confidence of each disease evolution path.

[0022] Furthermore, the specific steps for determining the cache refresh time window for the pre-cached data in step S240 include:

[0023] S241. Obtain historical medical record data and filter out path-fitting cases, construct a path-fitting case dataset. Path-fitting cases are those in historical medical record data whose path similarity between the disease evolution path and the target disease evolution path is greater than a preset path similarity threshold.

[0024] S242. Extract the disease evolution cycle of path-adapted cases from the path-adapted case dataset and take the weighted average of the disease evolution cycles of all path-adapted cases as the path evolution cycle. The weight is the ratio of the path similarity corresponding to the path-adapted case to the sum of the path similarities.

[0025] S243. Extract the change characteristics of each data type in the disease evolution through the path-adapted case dataset, and use the Pearson correlation coefficient between the numerical change of each data type and the change of disease status as the correlation degree of disease evolution.

[0026] S244. Obtain pre-cached data, determine the validity period of each data type in the pre-cached data, and determine the data value period by combining the correlation with the disease evolution: ,in, For data types The data value cycle For data types The corresponding data validity period, Represents data type The correlation between the disease progression;

[0027] S245. The minimum value between the path evolution cycle of the disease evolution path and the data value cycle of the pre-cached data is used as the cache refresh time window for the pre-cached data.

[0028] Further, the determination of the target ferry node in step S300 includes:

[0029] S310. Extract the request features of the data interaction request cluster and the node features of the ferry node and perform vector modeling to obtain the request feature vector and the node feature vector.

[0030] S320. The weighted cosine similarity between the request feature vector and the node feature vector is used as the feature matching degree between the data interaction request cluster and the ferry node.

[0031] S330. Select the ferry node with the highest feature matching degree as the target ferry node for the data interaction request cluster.

[0032] Secondly, the present invention provides a secure communication device for a mobile terminal of an internet hospital based on message transfer, comprising:

[0033] The request cluster classification module is used to obtain medical data interaction requests initiated by mobile terminals and classify the data interaction request clusters according to the semantic relevance of the medical data interaction requests.

[0034] The pre-caching strategy determination module is connected to the request cluster classification module. It is used to obtain patient medical record data corresponding to the data interaction request cluster and perform disease evolution path analysis. Based on the disease evolution path and evolution confidence, it determines the data pre-caching strategy.

[0035] The target node determination module, connected to the pre-caching strategy determination module, is used to perform feature analysis on the data interaction request cluster and the ferry node respectively, and determine the target ferry node by feature matching.

[0036] The pre-caching strategy execution module is connected to the target node determination module and is used to execute the data pre-caching strategy and perform message ferrying through the target ferry node.

[0037] Thirdly, the present invention provides an electronic device, comprising: a processor and a memory, wherein the memory stores a computer program that can be called by the processor, and the processor executes a secure communication method for a mobile terminal of an Internet hospital based on message ferry by calling the computer program stored in the memory.

[0038] Fourthly, the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform a secure communication method for a mobile terminal of an Internet hospital based on message transfer.

[0039] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0040] This invention introduces a data pre-caching strategy driven by semantic correlation-based data interaction request cluster classification, disease evolution path analysis, and intelligent ferry node selection mechanism based on node feature matching. This enables an efficient, secure, and low-latency data communication method for medical scenarios. By dynamically adjusting the cached content and refresh timing based on changes in individual patient conditions, it effectively improves the data transmission efficiency in time-sensitive medical scenarios. Attached Figure Description

[0041] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0042] Figure 1 This is a flowchart illustrating the secure communication method for mobile terminals of internet hospitals based on message transfer provided in an embodiment of the present invention.

[0043] Figure 2 This is a flowchart illustrating step S100 provided in an embodiment of the present invention;

[0044] Figure 3 This is a flowchart illustrating step S200 provided in an embodiment of the present invention;

[0045] Figure 4 This is a flowchart illustrating step S240 provided in an embodiment of the present invention;

[0046] Figure 5 This is a flowchart illustrating step S300 provided in an embodiment of the present invention;

[0047] Figure 6 This is a schematic diagram of the structure of the secure communication device for mobile terminals of Internet hospitals based on message transfer provided in an embodiment of the present invention;

[0048] Figure 7 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0049] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0050] Please see Figure 1 , Figure 1 This is a schematic diagram of the overall process of the secure communication method for mobile terminals of Internet hospitals based on message transfer provided in this embodiment of the invention, which specifically includes the following steps:

[0051] S100: Obtain medical data interaction requests initiated by the mobile terminal, and classify the data interaction request clusters according to the semantic relevance of the medical data interaction requests;

[0052] By semantically categorizing medical data interaction requests, request clusters with high semantic relevance and similar data needs are constructed. This allows subsequent disease evolution path analysis, data pre-caching strategy formulation, and ferry node selection to be centrally processed and optimized around these request clusters, thereby improving system processing efficiency, data hit rate, and response accuracy. Please refer to [link to relevant documentation]. Figure 2 , Figure 2 This is a flowchart illustrating step S100 provided in an embodiment of the present invention, including:

[0053] S110. Semantically analyze each medical data interaction request using a natural language processing model and extract the semantic feature vector of each medical data interaction request. Perform a structured semantic expression on the natural language medical requests submitted by the user and convert each request into a computable feature vector. The natural language processing model can be any one of Med-BERT, BioBERT and ClinicalBERT models.

[0054] S120. Determine the semantic correlation between each medical data interaction request based on the semantic feature vector. The semantic correlation is the cosine similarity between the semantic feature vectors of the medical data interaction requests. By calculating the cosine similarity, requests with similar meanings but different expressions can be identified, avoiding the semantic fragmentation problem caused by differences in expression, ensuring that semantically similar requests can be correctly classified, and providing support for the unified formulation of caching and communication strategies for highly related requests.

[0055] S130. Medical data interaction requests with semantic relevance greater than a preset semantic relevance threshold are divided into data interaction request clusters. By aggregating requests for the same type of disease, treatment stage, or medical behavior, subsequent disease evolution path reasoning and data cache scheduling become more targeted and have batch processing advantages.

[0056] S200: Obtain patient medical record data corresponding to the data interaction request cluster and perform disease evolution path analysis; determine data pre-caching strategy based on disease evolution path and evolution confidence.

[0057] In mobile internet healthcare scenarios, the timeliness of medical responses is crucial. Traditional "on-demand" data access models struggle to meet these demands. Disease evolution path analysis can predict impending changes in disease status and pre-cachise relevant medical data, shifting from a "passive response" to a "proactive supply" approach. This improves the timeliness and accuracy of data acquisition while reducing system communication latency and load. Please refer to [link to relevant documentation]. Figure 3 , Figure 3 This is a flowchart illustrating step S200 provided in an embodiment of the present invention, including:

[0058] S210. Obtain the patient's medical record data corresponding to the data interaction request cluster;

[0059] S220. Construct a disease evolution tree path diagram based on the patient's medical record data. The nodes in the disease evolution tree path diagram represent the disease evolution state, and the edges represent the disease evolution direction. By constructing the disease evolution tree path diagram, the stages and branches of disease development can be effectively revealed, providing a data foundation for evolution confidence assessment and pre-caching decision-making. The methods for constructing the patient's disease evolution tree path diagram include path modeling based on clinical rules, state transition analysis based on medical record sequence mining (e.g., Markov model), disease evolution modeling based on deep learning models (e.g., LSTM model or Transformer model), and common path extraction methods based on graph mining and path clustering.

[0060] S230. Determine the evolution confidence of each disease evolution path in the disease evolution tree path diagram and take the disease evolution path with the highest evolution confidence as the target disease evolution path. In multi-path evolution, only some disease evolution paths appear frequently or are logically reasonable in clinical practice. Therefore, it is necessary to select the most representative disease evolution path as the decision basis for the pre-caching strategy.

[0061] The joint transition probability is used to quantify the credibility and probability of occurrence of each path in actual clinical evolution. Using the joint transition probability as the evolutionary confidence of each disease evolution path reflects the statistical support strength for the disease evolving along a certain path. This allows for the selection of the most representative and predictive target disease evolution path from multiple paths. The specific steps in step S230 for determining the evolutionary confidence include:

[0062] S231. Extract the disease evolution paths from the disease evolution tree diagram and calculate the joint transition probability of all adjacent nodes on each disease evolution path:

[0063] ;

[0064] in, For the first The joint transition probability of all adjacent nodes on the disease evolution path. For the node Transfer to node The probability of, where, node and nodes These are two adjacent nodes on the disease progression path, which can be obtained through analysis and statistical analysis of a large number of historical cases using historical medical record data. Indicates the first The number of nodes in each disease evolution path Indicates starting from the initial node Continuous evolution to the tail node The joint metastasis probability is the product of all metastasis probabilities in the disease evolution path.

[0065] S232. The joint transition probability of all adjacent nodes on the disease evolution path is used as the evolution confidence of each disease evolution path.

[0066] S240. Use the disease evolution data corresponding to the target disease evolution path as pre-cached data and determine the cache refresh time window of the pre-cached data. The disease evolution data includes diagnostic data, examination and test data, treatment plan data and medical behavior data.

[0067] In data pre-caching strategies, the cache refresh time window controls the effective duration and update frequency of pre-cached data. Its core function is to balance data freshness with the efficiency of cache resource utilization. By setting a reasonable refresh window, it can be ensured that cached data still has timeliness and clinical reference value when a user makes a request, avoiding cache invalidation or redundant storage space occupation. Especially in medical scenarios, the sensitivity of data varies depending on the patient's condition. Establishing a precise refresh cycle is crucial for ensuring the accuracy of diagnostic data, improving response speed, and reducing system load. Please refer to [link to relevant documentation]. Figure 4 , Figure 4 This is a flowchart illustrating step S240 provided in an embodiment of the present invention, including:

[0068] S241. Obtain historical medical record data and filter out path-fitting cases, construct a path-fitting case dataset. Path-fitting cases are those in historical medical record data whose path similarity to the target path is greater than a preset path similarity threshold. Path-fitting cases can provide experience on data changes in the evolution of similar diseases, serving as an important reference for predicting the evolution rhythm of the target path, ensuring that subsequent analysis is based on representative real disease course data, and improving the accuracy and reliability of the caching strategy.

[0069] S242. Extract the disease evolution cycle of path-adapted cases from the path-adapted case dataset and use the weighted average of the disease evolution cycles of all path-adapted cases as the path evolution cycle. The weight is the ratio of the path similarity corresponding to the path-adapted case to the sum of the path similarities. The disease evolution cycle reflects the speed of the patient's state change under the path and is one of the core bases for judging the cache refresh time window. By transforming historical medical record data into dynamic update basis, the pre-caching strategy can be closer to the actual development rhythm of the disease.

[0070] S243. Extract the change characteristics of each data type in the disease evolution through the path-adapted case dataset, and use the Pearson correlation coefficient between the numerical change of each data type and the change of disease status as the disease evolution correlation degree. The disease evolution correlation degree is used to quantify the sensitivity and correlation of various medical data in the disease evolution.

[0071] S244. Obtain pre-cached data, determine the validity period of each data type in the pre-cached data, and determine the data value period by combining the correlation with the disease evolution: ,in, For data types The data value cycle For data types The corresponding data validity period can be obtained through medical clinical guidelines or through statistical analysis of historical medical record data. The steps for obtaining validity period based on statistical analysis of historical medical record data include: performing a time span analysis on the same data type in historical cases, and using the average time from the generation of that data type to its first update as the data validity period. Represents data type The correlation between the disease progression and the disease progression. For data types The absolute value of the correlation between disease progression and disease evolution is used. The sign of the correlation only indicates the directional relationship between data changes and disease progression. Therefore, the absolute value is used to represent the strength of the correlation to disease progression to avoid directional interference with the logic. The '1' indicates that the data validity period is used as a benchmark to ensure that when... At that time, the data value cycle is still the data validity period;

[0072] S245. The minimum value between the path evolution cycle of the disease evolution path and the data value cycle of the pre-cached data is used as the cache refresh time window of the pre-cached data. By comprehensively considering the disease development rhythm and data validity, a refresh boundary that will not cause cache failure due to data aging or disease changes is found, ensuring a balance between the stability and timeliness of the caching strategy.

[0073] S300: Perform feature analysis on the data interaction request cluster and the ferry node respectively, and determine the target ferry node through feature matching;

[0074] Because different relay nodes vary in network bandwidth, computing power, cache status, and geographical location, intelligently matching the most suitable data relay node based on request characteristics will significantly improve the system's load balancing capabilities and reduce communication latency. Please refer to [link to relevant documentation]. Figure 5 , Figure 5 This is a flowchart illustrating step S300 provided in an embodiment of the present invention, including:

[0075] S310. Extract the request features of the data interaction request cluster and the node features of the ferry node and perform vector modeling to obtain the request feature vector and node feature vector, providing a quantitative basis for accurate matching between complex features.

[0076] S320. The weighted cosine similarity between the request feature vector and the node feature vector is used as the feature matching degree between the data interaction request cluster and the ferry node. Since different request features have different importance in matching (such as being more sensitive to urgency), weighting can highlight the influence of key factors and make the matching closer to actual needs.

[0077] S330. The ferry node with the highest feature matching degree is used as the target ferry node of the data interaction request cluster. The target ferry node is used to undertake pre-cached data and subsequent data transmission tasks, effectively reducing the waiting time and failure rate in the data scheduling process, and improving the stability of the communication link and the quality of service perception.

[0078] S400 executes the data pre-caching strategy and performs message ferrying through the target ferry node.

[0079] In this embodiment of the invention, the determination of parameters such as weighted weights, preset semantic relevance thresholds, and preset path similarity thresholds can be achieved by: constructing a dataset by acquiring medical data interaction requests, patient medical record data, and historical medical record data; substituting these data into the dataset to calculate feature matching degree, semantic relevance, and path similarity; simultaneously acquiring expert judgments on the feature matching degree between the data interaction request cluster and the ferry node, the semantic relevance between medical data interaction requests, and the path similarity between the disease evolution path in historical medical record data and the target disease evolution path; importing the calculated feature matching degree, semantic relevance, path similarity, and judgment results into fitting software; and outputting the weighted weights, preset semantic relevance thresholds, and preset path similarity thresholds that meet the maximum judgment accuracy.

[0080] Please see Figure 6 , Figure 6 This is a schematic diagram of the structure of a secure communication device for a mobile terminal of an internet hospital based on message transfer, provided in an embodiment of the present invention, including:

[0081] The request cluster classification module 210 is used to obtain medical data interaction requests initiated by the mobile terminal and classify the data interaction request clusters according to the semantic relevance of the medical data interaction requests.

[0082] The pre-caching strategy determination module 220 is connected to the request cluster classification module 210 and is used to obtain patient medical record data corresponding to the data interaction request cluster and perform disease evolution path analysis, and determine the data pre-caching strategy based on the disease evolution path and evolution confidence.

[0083] The target node determination module 230 is connected to the pre-caching strategy determination module 220 and is used to perform feature analysis on the data interaction request cluster and the ferry node respectively and determine the target ferry node by feature matching.

[0084] The pre-caching strategy execution module 240 is connected to the target node determination module 230 and is used to execute the data pre-caching strategy and perform message ferrying through the target ferry node.

[0085] In this embodiment of the invention, the request cluster classification module 210 is used to obtain medical data interaction requests initiated by the mobile terminal, and to classify the data interaction request clusters according to the semantic relevance of the medical data interaction requests, including:

[0086] The semantics of each medical data interaction request are analyzed using a natural language processing model, and the semantic feature vectors of each medical data interaction request are extracted.

[0087] The semantic correlation between each medical data interaction request is determined based on the semantic feature vectors. The semantic correlation is the cosine similarity between the semantic feature vectors of the medical data interaction requests.

[0088] Medical data interaction requests with a semantic relevance greater than a preset semantic relevance threshold are classified into data interaction request clusters.

[0089] In this embodiment of the invention, the pre-caching strategy determination module 220 is used to acquire patient medical record data corresponding to the data interaction request cluster and perform disease evolution path analysis, and determine a data pre-caching strategy based on the disease evolution path and evolution confidence, including:

[0090] Retrieve patient medical record data corresponding to the data interaction request cluster;

[0091] Based on the patient's medical record data, a tree-like path diagram of the patient's disease evolution is constructed. The nodes in the disease evolution tree-like path diagram represent the disease evolution state, and the edges represent the disease evolution direction.

[0092] The evolutionary confidence of each disease evolution path in the disease evolution tree diagram is determined, and the disease evolution path with the highest evolutionary confidence is selected as the target disease evolution path. The specific steps for determining the evolutionary confidence include:

[0093] Extract the disease evolution paths from the disease evolution tree path graph and calculate the joint transition probability of all adjacent nodes on each disease evolution path;

[0094] The joint transition probability of all adjacent nodes on the disease evolution path is used as the evolution confidence of each disease evolution path;

[0095] The disease evolution data corresponding to the target disease evolution path is used as pre-cached data, and the cache refresh time window of the pre-cached data is determined, including:

[0096] Historical medical record data is obtained and path-fitting cases are selected. A path-fitting case dataset is constructed. Path-fitting cases are those in historical medical record data whose path similarity to the path of disease evolution is greater than a preset path similarity threshold.

[0097] The disease evolution cycle of path-adapted cases is extracted from the path-adapted case dataset, and the weighted mean of the disease evolution cycles of all path-adapted cases is taken as the path evolution cycle. The weight is the ratio of the path similarity corresponding to the path-adapted case to the sum of the path similarities.

[0098] By extracting the change characteristics of each data type in the disease evolution from the path-adapted case dataset, the Pearson correlation coefficient between the numerical change of each data type and the change of disease status is used as the correlation degree of disease evolution.

[0099] Obtain pre-cached data to determine the validity period of each data type in the pre-cached data and combine it with the correlation of disease evolution to determine the data value cycle;

[0100] The minimum value between the path evolution cycle of the disease progression path and the data value cycle of the pre-cached data is used as the cache refresh time window for the pre-cached data.

[0101] In this embodiment of the invention, the target node determination module 230 is used to perform feature analysis on the data interaction request cluster and the ferry node respectively, and determine the target ferry node through feature matching, including:

[0102] Extract the request features of the data interaction request cluster and the node features of the ferry node, and perform vector modeling to obtain the request feature vector and the node feature vector.

[0103] The weighted cosine similarity between the request feature vector and the node feature vector is used as the feature matching degree between the data interaction request cluster and the ferry node.

[0104] The ferry node with the highest feature matching degree is selected as the target ferry node for the data interaction request cluster.

[0105] The parameters and steps for implementing the corresponding functions of each unit module in the secure communication device for mobile terminals of Internet hospitals based on message transfer of the present invention can be referred to the parameters and steps in the embodiments of the secure communication method for mobile terminals of Internet hospitals based on message transfer, and will not be repeated here.

[0106] Please refer to Figure 7The present invention also provides an electronic device 300, including a memory 310, a processor 320, and a communication bus 330; the memory 310 and the processor 320 are connected via the communication bus 330. The memory 310 stores a secure communication method for a mobile terminal of an Internet hospital based on message transfer, which can be loaded by the processor 320 and executed as provided in the above embodiments.

[0107] The memory 310 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 310 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for at least one function, and instructions for implementing the message-ferry-based secure communication method for mobile terminals of internet hospitals provided in the above embodiments. The data storage area may store data involved in the message-ferry-based secure communication method for mobile terminals of internet hospitals provided in the above embodiments.

[0108] Processor 320 may include one or more processing cores. Processor 320 executes instructions, programs, code sets, or instruction sets stored in memory 310, and calls data stored in memory 310 to perform various functions and process data according to the present invention. Processor 320 may be at least one of the following: Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), Central Processing Unit (CPU), Controller, Microcontroller, and Microprocessor. It is understood that, for different devices, the electronic devices used to implement the functions of processor 320 may also be other types, and the embodiments of the present invention do not specifically limit this.

[0109] The communication bus 330 may include a path for transmitting information between the aforementioned components. The communication bus 330 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 330 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7The symbol is represented by a single double arrow, but this does not mean that there is only one bus or one type of bus.

[0110] This invention provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed as described in the above embodiments for a secure communication method for mobile terminals of internet hospitals based on message ferrying.

[0111] In this embodiment of the invention, the computer-readable storage medium can be a tangible device that holds and stores instructions used by an instruction execution device. The computer-readable storage medium can be, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination thereof. Specifically, the computer-readable storage medium can be a portable computer disk, a hard disk, a USB flash drive, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), lectern random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory stick, floppy disk, optical disk, magnetic disk, mechanical encoding device, or any combination thereof.

[0112] The terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0113] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this application is not limited to the technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions claimed in this invention.

Claims

1. A message ferrying-based Internet hospital mobile terminal secure communication method, characterized in that, Comprise: S100, acquire the medical data interaction request initiated by the mobile terminal, and classify the data interaction request cluster according to the semantic correlation degree between each medical data interaction request; S200, acquire the patient medical record data corresponding to the data interaction request cluster and perform path analysis of the disease evolution, determine the data pre-caching strategy according to the path of the disease evolution and the evolution confidence; S300, respectively analyze the features of the data interaction request cluster and the transfer node, and determine the target transfer node through the feature matching based on the weighted cosine similarity; S400, execute the data pre-caching strategy through the target transfer node and perform message transfer.

2. The message ferrying based secure communication method for internet hospital mobile terminal according to claim 1, characterized in that, The specific steps of classifying the data interaction request cluster in step S100 include: S110, perform semantic analysis on each medical data interaction request through a natural language processing model and extract the semantic feature vector of each medical data interaction request; S120, determine the semantic correlation degree between each medical data interaction request according to the semantic feature vector, and the semantic correlation degree is the cosine similarity between the semantic feature vectors of the medical data interaction requests; S130, divide the medical data interaction requests with a semantic correlation degree greater than a preset semantic correlation degree threshold into a data interaction request cluster.

3. The method of claim 1, wherein the method further comprises: The specific steps of determining the data pre-caching strategy in step S200 include: S210, acquire the patient medical record data corresponding to the data interaction request cluster; S220, construct a patient disease evolution tree path graph according to the patient medical record data, wherein the nodes in the disease evolution tree path graph represent the disease evolution state, and the edges represent the disease evolution direction; S230, determine the evolution confidence of each disease evolution path in the disease evolution tree path graph and take the disease evolution path with the highest evolution confidence as the target disease evolution path; S240, take the disease evolution data corresponding to the target disease evolution path as the pre-caching data and determine the cache refresh time window of the pre-caching data.

4. The method according to claim 3, wherein, The specific steps of determining the evolution confidence in step S230 include: S231, extract the disease evolution paths in the disease evolution tree path graph and calculate the joint transition probability of all adjacent nodes on each disease evolution path; ； in, For the first The joint transition probability of all adjacent nodes on the disease evolution path. For the node Transfer to node The probability, Indicates the first The number of nodes in each disease progression path; S232, take the joint transition probability of all adjacent nodes on the disease evolution path as the evolution confidence of each disease evolution path.

5. The method of claim 3, wherein the method further comprises: The specific steps of determining the cache refresh time window of the pre-caching data in step S240 include: S241, acquire historical medical record data and filter out path adaptation cases, construct a path adaptation case data set, and the path adaptation case is a case in the historical medical record data whose path similarity between the disease evolution path and the target disease evolution path is greater than a preset path similarity threshold; S242, extract the disease evolution period of the path adaptation case through the path adaptation case data set, and take the weighted mean of the disease evolution periods of all path adaptation cases as the path evolution period, and the weight is the ratio of the path similarity corresponding to the path adaptation case to the sum of the path similarities; S243, extract the change characteristics of each data type in the evolution of the disease through the path adaptation case data set, and take the Pearson correlation coefficient between the numerical change of each data type and the disease state change as the disease evolution correlation degree; S244, acquiring pre-cached data determines the data validity period corresponding to each data type in the pre-cached data and determines the data value period in combination with the illness evolution correlation degree: wherein, is the data value period of the data type , is the data validity period corresponding to the data type , represents the illness evolution correlation degree of the data type . S245, take the minimum value in the path evolution period of the disease evolution path and the data value period of the pre-cached data as the cache refresh time window of the pre-cached data.

6. The method of claim 1, wherein the method further comprises: The determination of the target ferry node in step S300 includes: S310, extract the request features of the data interaction request cluster and the node features of the ferry node and perform vector modeling to obtain the request feature vector and the node feature vector; S320, take the weighted cosine similarity of the request feature vector and the node feature vector as the feature matching degree of the data interaction request cluster and the ferry node; S330, take the ferry node with the highest feature matching degree as the target ferry node of the data interaction request cluster.

7. An Internet hospital mobile terminal secure communication device based on message ferrying, used to implement the Internet hospital mobile terminal secure communication method based on message ferrying in any one of claims 1-6, characterized in that, The device includes: A request cluster classification module configured to obtain medical data interaction requests initiated by mobile terminals, and to classify data interaction request clusters according to the semantic correlation degrees between the medical data interaction requests; A pre-caching strategy determination module connected to the request cluster classification module, configured to obtain patient medical record data corresponding to the data interaction request clusters and to analyze disease evolution paths, and to determine a data pre-caching strategy according to the disease evolution paths and evolution confidence degrees; A target node determination module connected to the pre-caching strategy determination module, configured to analyze the features of the data interaction request clusters and the ferry nodes respectively, and to determine a target ferry node through feature matching based on weighted cosine similarity; A pre-caching strategy execution module connected to the target node determination module, configured to execute the data pre-caching strategy through the target ferry node and to perform message ferrying.

8. An electronic device comprising: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; characterized in that the processor executes a message ferrying-based internet hospital mobile terminal secure communication method according to any one of claims 1-6 by calling the computer program stored in the memory.

9. A computer-readable storage medium, characterized in that, An instruction is stored, and when the instruction is run on a computer, the computer executes a message ferrying-based internet hospital mobile terminal secure communication method according to any one of claims 1-6.

Images