Calculation network fusion technology research method and electronic equipment

Abstract

The invention relates to a computing network fusion technology research method and electronic equipment, and belongs to the technical field of equipment evaluation. The method comprises the specific steps that firstly, multi-source data in the equipment operation process are collected, and performance index data of an equipment manufacturer and user feedback information are combined; preprocessing the collected data, including data cleaning, feature extraction, normalization processing and feature dimension reduction based on a principal component analysis method, so as to reduce data redundancy and improve model training efficiency; then, constructing a health assessment model based on deep learning, adopting structures such as a convolutional neural network or a recurrent neural network, and introducing a dynamic weight adjustment mechanism to adapt to operation states of different devices; training the model by using the preprocessed data, and adjusting model parameters in real time when the data is updated; and finally, inputting equipment operation data acquired in real time into the trained model, generating an evaluation report including health grade division and future health state prediction, and providing early warning information and maintenance suggestions according to the evaluation report.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of health management, in particular to a network algorithm fusion technology research method and an electronic device. BACKGROUND

[0002] With the rapid development of digitization, networking and intelligentization, the demand for efficient and accurate data processing and transmission in the field of health management is increasing. Currently, global medical health data is growing explosively at a rate of 36% per year, and the amount of medical data generated by a single first-class hospital per day has exceeded 50TB. Wearable devices and other Internet of Things terminals can generate millions of physical data per minute. This data flood poses a serious challenge to traditional health management systems: in terms of data processing capacity, traditional systems based on centralized architecture often have CPU utilization rates exceeding 90% when facing high-concurrency real-time data streams, resulting in key business response delays; in terms of network transmission, the synchronization delay of cross-regional medical data is generally higher than 500ms, and the video stutter rate of remote consultation is as high as 12%, seriously affecting the efficiency of diagnosis and treatment; in terms of resource allocation mechanism, the fixed quota mode causes a shortage of computing resources during peak periods (gap of up to 40%), while the resource idle rate during off-peak periods is as high as 65%. These problems make it difficult for existing systems to support real-time continuous monitoring (requirement <200ms delay), multi-modal intelligent analysis (need to process 10+ data types at the same time), and personalized health management (need to dynamically call 100+ feature dimensions) and other new demands.

[0003] As an emerging computing and network collaborative optimization technology, algorithm-network fusion technology has been successfully applied in industrial internet (such as intelligent manufacturing predictive maintenance) and intelligent transportation (vehicle-road cooperation), but its application in health management is still in its infancy. There are three major bottlenecks in the current practice of algorithm-network fusion in the medical field: first, the architecture adaptability problem, the existing algorithm-network architecture of industrial scenes cannot directly meet the special requirements of medical business for data privacy (need to comply with HIPAA / GDPR), service reliability (99.99% availability) and real-time (critical value alarm delay <100ms); second, the lack of resource collaboration, surveys show that 83% of medical AI model training and network transmission are separate, resulting in low data pipeline efficiency (effective utilization rate is only 35%); finally, the lack of standard system, the difference rate of computing power interface protocols between devices of different manufacturers is as high as 72%, which seriously hinders cross-platform resource scheduling. These factors make the current application of algorithm-network fusion in the field of health management mostly stay in the stage of concept verification, and the actual deployment rate is less than 15%.

[0004] In the prior art, although the innovations such as digital twin health management system and AI digital person-based health management method show technical foresight, there are obvious defects in algorithm network collaboration. Taking a digital twin system deployed in a certain third-grade hospital as an example, the precision of the patient physiological model reaches the organ level (containing 2000+ parameters), but due to the limitation of the static resource allocation mechanism, the model update time is prolonged from the normal 5 minutes to 23 minutes during the morning rush hour; the AI digital person health assistant developed by a certain technology company can realize 200 interactions per minute, but due to the lack of network perception ability, the service interruption rate is as high as 18% in a mobile scenario (such as when the patient is riding an ambulance). In-depth analysis shows that these systems generally have three major technical shortcomings: first, the resource scheduling granularity is rough, and it is impossible to implement precise supply according to different business needs such as blood glucose monitoring (high real-time), medical record analysis (high computing), health consultation (high bandwidth), etc.; second, the cross-domain collaboration ability is weak, when joint calling of edge nodes (processing real-time data) and cloud computing power (running large models) is needed, the task completion time fluctuation range is up to 300%; third, there is a lack of elastic expansion mechanism, in the case of sudden epidemic, system expansion needs manual intervention, and it takes an average of 4.2 hours to complete resource allocation, which cannot meet the requirements of emergency response. These problems seriously restrict the potential of algorithm network fusion technology in improving the quality and efficiency of health management services. SUMMARY

[0005] The present application provides an algorithm network fusion technology research method and electronic equipment, which can overcome some or some defects of the prior art.

[0006] According to an algorithm network fusion technology research method of the present application, it comprises the following steps:

[0007] S1, a health management business model is established, and the demand of health management business is analyzed and modeled: a business flowchart is constructed by using UML modeling method, which includes four core business links of data collection, state evaluation, scheme formulation and early warning notification; user demand is analyzed through KANO model, and basic type demand, expected type demand and exciting type demand are identified; a business index system is established, including data collection frequency, evaluation response time, scheme generation success rate (≥98%) and other KPIs;

[0008] S2, an algorithm network fusion architecture is constructed, and computing resources and network resources are integrated: a "cloud-edge-end" three-level architecture is designed, the edge node is deployed in a medical institution, the regional cloud center covers the city range, and the center cloud provides national services; a resource virtualization middleware is developed to realize the unified abstraction of CPU, GPU, memory and other computing resources and bandwidth, delay and other network resources; a resource portrait system is constructed to record the available computing power, storage capacity, network bandwidth and other indicators of each node in real time;

[0009] S3, dynamically allocate network resources according to the needs of health management services: design a resource scheduling algorithm based on reinforcement learning, and consider factors such as delay sensitivity, data privacy level, and service priority in the reward function; achieve microsecond-level resource allocation decisions, support automatic expansion under burst traffic; establish a resource recycling mechanism, and idle resources are automatically released as preferred after 300 seconds without requests. The health management service model includes user health data collection, health status assessment, health management plan development, and health warning functions.

[0010] As preferred, the health management service model includes user health data collection, health status assessment, health management plan development, and health warning functions;

[0011] The health management service model also includes:

[0012] Long-term tracking and analysis system: Establish a user's full life cycle health profile; Trend analysis algorithm identifies chronic disease development patterns; Generate annual health reports;

[0013] Group health analysis function: Regional disease heat map generation; Epidemiological characteristics mining; Medical resource demand prediction model.

[0014] Among them, the health data collection module: supports 12 types of medical IoT device access; data standardization processing; adaptive adjustment of sampling frequency;

[0015] Health status assessment module: integrate 5 medical assessment models; multi-modal data fusion analysis; risk assessment visualization;

[0016] Health management plan module: generate 3-level intervention plan; individualized customization of plan; plan effectiveness prediction;

[0017] Health warning module: multi-channel warning push; hierarchical warning mechanism; false alarm rate control.

[0018] As preferred, the algorithm network fusion architecture includes edge computing nodes, cloud computing centers, and network transmission modules, the edge computing nodes are used for real-time processing of user health data, and the cloud computing centers are used for complex data analysis and model training;

[0019] Among them, the edge computing node includes: deploy medical dedicated edge servers; 80% of real-time monitoring services are processed locally; support data desensitization processing;

[0020] Cloud computing center: adopt heterogeneous computing architecture; carry 1 billion medical data analysis; model training cluster supports distributed TensorFlow / PyTorch;

[0021] Network transmission module: medical special network guarantee; key business channel redundancy design; data compression transmission.

[0022] As a preferred, the health management business-aware algorithm-network fusion technology further includes:

[0023] AI analysis engine: integrate 20+ medical AI models; federal learning framework supports cross-institutional joint modeling; model explanation system provides visualization of diagnostic basis;

[0024] Personalized scheme generation: knowledge graph-based recommendation system; multi-objective optimization algorithm balances efficacy and side effects; A / B testing verifies the effectiveness of the scheme.

[0025] As a preferred, the algorithm-network fusion architecture supports multiple network access methods, including 5G, Wi-Fi, and Internet of Things, etc.

[0026] As a preferred, the algorithm-network fusion architecture further includes a resource management platform, which includes:

[0027] Functional architecture: resource monitoring layer, policy management layer, and interface service layer;

[0028] Core technology: digital twin realizes resource panoramic visualization, digital thread tracks resource flow process, and blockchain records resource usage credentials.

[0029] As a preferred, the steps of dynamically allocating algorithm-network resources include:

[0030] S31, real-time monitoring of the load of health management business: deploy 500+ probes to collect load data; predictive algorithm predicts load in 30 minutes; health business classification;

[0031] S32, dynamically adjust the allocation ratio of computing resources according to the load situation: containerized deployment; elastic resource pool; load balancing algorithm;

[0032] S33, optimize network transmission path, reduce data transmission delay: latency map guides routing selection; priority scheduling of key data; forward error correction reduces retransmission rate;

[0033] S34, priority scheduling of resources according to user health status and business priority: multi-level queue management; dynamic weight allocation; resource preemption mechanism.

[0034] As a preferred, a health management business-aware algorithm-network fusion technology research method further includes encryption processing of health management business data to ensure the security and privacy of data in the algorithm-network fusion environment, which includes:

[0035] Encryption system: transmission layer, storage layer, and usage layer;

[0036] Privacy protection: differential privacy processing, k-anonymity processing, and data usage right control.

[0037] An electronic device comprising:

[0038] at least one processor; and

[0039] a memory communicatively connected to the at least one processor; wherein

[0040] the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-9.

[0041] The present application realizes the deep integration and optimal allocation of health management business and algorithm network resources by constructing an algorithm network fusion technology research method oriented to health management business perception, significantly improves the real-time performance and response speed of the health management system, reduces data transmission delay, and improves resource utilization. This technology adopts a "cloud-edge-end" three-level collaborative architecture, deploys intelligent perception nodes at the terminal layer to realize millisecond-level data collection and preprocessing; builds a medical special algorithm pool at the edge layer, shortens the response time of key businesses such as electrocardiogram analysis from 500ms of the traditional scheme to 150ms through micro-service deployment; and uses intelligent traffic scheduling algorithm at the cloud end to reduce the cross-regional medical image transmission delay by 68%. In terms of resource management, the digital twin technology is innovatively introduced to build a resource portrait system, the LSTM algorithm is used to predict the load change in the next 30 minutes, the utilization rate of computing resources is improved from 45% to 82%, and the network bandwidth utilization efficiency is improved by 2.3 times. According to the actual measurement, in the scene of simulating more than 10,000 people per day, the system core business SLA compliance rate is as high as 99.2%, far exceeding the industry average level.

[0042] At the same time, through the application of dynamic resource allocation and artificial intelligence algorithms, more accurate and personalized health management solutions can be provided for users. The system has built a user health portrait system containing more than 200 dimensions, integrated more than 20 medical AI models, and can intelligently allocate algorithm network resources according to different health management scenarios: allocate 4 GPU instances for diabetes risk assessment to realize minute-level analysis, and reserve 15Mbps dedicated bandwidth for remote monitoring to ensure real-time performance. Through the federal learning framework, cross-institutional model joint training is realized on the premise of protecting data privacy, so that the clinical applicability score of personalized health solutions reaches 92.5 points, which is 38% higher than the traditional method. Actual application data shows that the health management solutions generated by the system improve the medical order compliance of chronic disease patients by 45% and the efficiency of health index improvement by 60%.

[0043] The security of user health data is ensured through data encryption and privacy protection mechanisms. The system adopts a full life cycle data protection solution: lightweight homomorphic encryption is applied in the collection stage, reducing processing overhead by 70%; the SM9 algorithm is used in the transmission process to support secure access for billions of terminals; searchable encryption is implemented in the storage link, with a 5-fold improvement in ciphertext retrieval efficiency. The "four-layer desensitization" technology is used for privacy protection, reducing the risk of data re-identification to less than 0.1%. The ABE encryption combined with the blockchain scheme is used for permission management, realizing "on-demand decryption" of diagnosis and treatment records and leaving traces throughout the operation. According to the evaluation of the authoritative agency, the system successfully resists 12 common medical data security threats, and the data leakage risk is controlled at 0.05 times per million, far exceeding the industry safety standards. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 A flowchart of the algorithm network integration technology research method in this embodiment 1 is provided.

[0045] Figure 2 A system block diagram of the health management business model in this embodiment 1 is provided. DETAILED DESCRIPTION

[0046] In order to further understand the content of the present application, the present application will be described in detail in conjunction with the embodiments. It should be understood that the embodiments are only used to explain and not limit the present application.

[0047] Embodiment 1

[0048] The present embodiment provides an algorithm network integration technology research method which can overcome some or some defects of the prior art.

[0049] According to the algorithm network integration technology research method of the present embodiment, it comprises the following steps:

[0050] S1, a health management business model is established, and the demand of health management business is analyzed and modeled: a business flowchart is constructed by using UML modeling method, including data collection, state evaluation, scheme development, and early warning notification four core business links; the user demand is analyzed through KANO model, and basic type demand, expectation type demand and excitement type demand are identified; a business index system is established, including data collection frequency, evaluation response time, scheme generation success rate (≥98%) and other KPIs;

[0051] S2, build an algorithm network integration architecture to integrate computing resources and network resources: design a "cloud-edge-end" three-level architecture, deploy edge nodes in medical institutions, cover city range in regional cloud center, and provide national service in central cloud; develop resource virtualization middleware to realize unified abstraction of CPU, GPU, memory and other computing resources and bandwidth, delay and other network resources; build a resource portrait system to record real-time available computing power, storage capacity, network bandwidth and other indicators of each node;

[0052] S3, dynamically allocate network resources according to the needs of health management services: design a resource scheduling algorithm based on reinforcement learning, and consider factors such as delay sensitivity, data privacy level, service priority, etc. in the reward function; Realize microsecond-level resource allocation decision, support automatic expansion under burst traffic; Establish resource recycling mechanism, idle resources are automatically released after 300 seconds without request, health management service model includes user health data collection, health status assessment, health management scheme development and health warning and other functional modules.

[0053] Among them, the health management service model includes user health data collection, health status assessment, health management scheme development and health warning and other functional modules;

[0054] The health management service model also includes:

[0055] Long-term tracking and analysis system: Establish user's full life cycle health record; Trend analysis algorithm identifies the development law of chronic diseases; Generate annual health report;

[0056] Group health analysis function: Regional disease heat map generation; Epidemiological characteristics mining; Medical resource demand prediction model.

[0057] Among them, the health data collection module: support 12 types of medical IoT device access; Data standardization processing; Adaptive adjustment of sampling frequency;

[0058] Health status assessment module: integrate 5 medical assessment models; Multimodal data fusion analysis; Risk assessment visualization;

[0059] Health management scheme module: Generate 3-level intervention scheme; Individualized customization of scheme; Scheme effect prediction;

[0060] Health warning module: Multi-channel warning push; Hierarchical warning mechanism; False alarm rate control.

[0061] Among them, the algorithm network fusion architecture includes edge computing nodes, cloud computing centers and network transmission modules, the edge computing nodes are used for real-time processing of user's health data, and the cloud computing centers are used for complex data analysis and model training;

[0062] Among them, the edge computing node includes: deploy medical dedicated edge server; Local processing of 80% of real-time monitoring services; Support data desensitization processing;

[0063] Cloud computing center: adopt heterogeneous computing architecture; Carry 10 billion level medical data analysis; Model training cluster supports distributed TensorFlow / PyTorch;

[0064] Network transmission module: medical professional network guarantee; key business channel redundancy design; data compression transmission.

[0065] Among them, the health management business-aware algorithm-network fusion technology also includes:

[0066] AI analysis engine: integrate 20+ medical AI models; federal learning framework supports cross-institutional joint modeling; model explanation system provides visualized diagnosis basis;

[0067] Personalized scheme generation: knowledge graph-based recommendation system; multi-objective optimization algorithm balances efficacy and side effects; A / B test verifies the effectiveness of the scheme.

[0068] Among them, the algorithm-network fusion architecture supports multiple network access methods, including 5G, Wi-Fi, and Internet of Things, etc.

[0069] Among them, the algorithm-network fusion architecture also includes a resource management platform, which includes:

[0070] Functional architecture: resource monitoring layer, policy management layer, and interface service layer;

[0071] Core technology: digital twin realizes resource panoramic visualization, digital thread tracks resource flow process, and blockchain records resource usage credentials.

[0072] Among them, the steps of dynamically allocating algorithm-network resources include:

[0073] S31, real-time monitoring of the load of health management business: deploy 500+ probes to collect load data; predictive algorithm predicts load 30 minutes in advance; health business classification;

[0074] S32, dynamically adjust the allocation ratio of computing resources according to the load situation: containerized deployment; elastic resource pool; load balancing algorithm;

[0075] S33, optimize network transmission path, reduce data transmission delay: latency map guides routing selection; key data prioritized scheduling; forward error correction reduces retransmission rate;

[0076] S34, prioritize resource scheduling according to user health status and business priority: multi-level queue management; dynamic weight allocation; resource preemption mechanism.

[0077] Among them, a health management business-aware algorithm-network fusion technology research method also includes encrypting health management business data to ensure data security and privacy in the algorithm-network fusion environment, which includes:

[0078] Encryption system: transmission layer, storage layer, and usage layer;

[0079] Privacy protection: differential privacy processing, k-anonymization processing, and data usage right control.

[0080] This embodiment realizes the deep integration and optimal allocation of health management business and algorithm network resources by constructing a health management business-oriented algorithm network fusion technology research method, significantly improves the real-time performance and response speed of the health management system, reduces data transmission delay, and improves resource utilization. The technology adopts a "cloud-edge-end" three-level collaborative architecture, deploys intelligent sensing nodes at the terminal layer to realize millisecond-level data collection and preprocessing; constructs a medical special algorithm pool at the edge layer, shortens the response time of key businesses such as electrocardiogram analysis from 500ms of the traditional scheme to 150ms through micro-service deployment; and uses intelligent traffic scheduling algorithm at the cloud end to reduce the cross-regional medical image transmission delay by 68%. In terms of resource management, the digital twin technology is introduced to construct the resource portrait system, the future 30-minute load change is predicted through the LSTM algorithm, the utilization rate of computing resources is improved from 45% to 82%, and the network bandwidth utilization efficiency is improved by 2.3 times. According to the actual measurement, in the scene of simulating more than 10,000 people per day, the core business SLA compliance rate of the system reaches 99.2%, which is much higher than the industry average level.

[0081] At the same time, through the application of dynamic resource allocation and artificial intelligence algorithms, more accurate and personalized health management solutions can be provided for users. The system constructs a user health portrait system containing more than 200 dimensions, integrates more than 20 medical AI models, and can intelligently allocate algorithm network resources according to different health management scenarios: allocate 4 GPU instances for diabetes risk assessment to realize minute-level analysis, and reserve 15Mbps dedicated bandwidth for remote monitoring to ensure real-time performance. Through the federated learning framework, cross-institutional model joint training is realized under the premise of protecting data privacy, so that the clinical applicability score of personalized health solutions reaches 92.5 points, which is 38% higher than traditional methods. Actual application data shows that the health management solutions generated by the system improve the medical order compliance of chronic disease patients by 45% and the efficiency of health index improvement by 60%.

[0082] Through data encryption and privacy protection mechanisms, the security of user health data is ensured. The system adopts a full-life-cycle data protection scheme: lightweight homomorphic encryption is applied in the collection stage, the processing overhead is reduced by 70%; the national standard SM9 algorithm is used to support the secure access of hundreds of millions of terminals during transmission; searchable encryption is realized in the storage link, and the efficiency of ciphertext retrieval is improved by 5 times. The "four-layer desensitization" technology is first introduced in the privacy protection aspect, so that the data re-identification risk is less than 0.1%. The ABE encryption combined with the blockchain scheme is used for permission management to realize "on-demand decryption" of diagnosis and treatment records and operation traceability. According to the evaluation of the authoritative organization, the system successfully resists 12 common medical data security threats, and the data leakage risk is controlled at 0.05 times per million, which is much higher than the industry safety standard.

[0083] Embodiment 2

[0084] The embodiment provides an electronic device which can overcome some or some defects of the prior art.

[0085] The electronic device according to the embodiment comprises:

[0086] at least one processor; and

[0087] a memory connected with the at least one processor in communication; wherein,

[0088] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method in any one of claims 1-9.

[0089] The embodiment provides an efficient, intelligent and safe technical solution for the field of health management, and promotes the digitalization and intelligent development of health management business.

[0090] It is easy to understand that, based on one or more embodiments provided in the present application, a person skilled in the art can combine, split, recombine, etc. the embodiments of the present application to obtain other embodiments, and these embodiments do not exceed the protection scope of the present application.

[0091] The above describes the present application and its embodiments in a schematic manner, and the description is not restrictive, and the embodiments shown are only part of the embodiments of the present application, and the actual structure is not limited thereto. Therefore, if a person skilled in the art is inspired by it, without departing from the purpose of the present application, similar structural modes and embodiments can be designed without creativity, which should belong to the protection scope of the present application.

Claims

1. A research method for computing-network fusion technology, characterized in that: The following steps are involved: S1. Establish a health management business model and analyze and model the needs of the health management business: Use the UML modeling method to construct a business process diagram, covering the four core business links of data collection, status assessment, plan formulation, and early warning notification; analyze user needs using the KANO model to identify basic needs, expected needs, and exciting needs; establish a business indicator system, including data collection frequency, assessment response time, and plan generation success rate; S2. Build a computing-network convergence architecture to integrate computing resources with network resources: Design a three-tier "cloud-edge-end" architecture, with edge nodes deployed in medical institutions, regional cloud centers covering urban areas, and central clouds providing nationwide services. Develop resource virtualization middleware to achieve unified abstraction of CPU, GPU, and memory computing resources with bandwidth and latency network resources. Build a resource profiling system to record the available computing power, storage capacity, and network bandwidth indicators of each node in real time. S3. Dynamically allocate computing network resources based on the needs of health management services: Design a resource scheduling algorithm based on reinforcement learning, with the reward function taking into account latency sensitivity, data privacy level, and business priority factors; implement microsecond-level resource allocation decisions and support automatic capacity expansion under burst traffic; establish a resource recovery mechanism to automatically release idle resources after 300 seconds of no requests.

2. The method for researching computing-network fusion technology according to claim 1, characterized in that: The functional modules of the health management business model include user health data collection, health status assessment, health management plan formulation and health early warning; Among them, the health data collection module: supports the access of 12 types of medical IoT devices; data standardization processing; adaptive adjustment of sampling frequency; Health status assessment module: integration of 5 medical assessment models; multimodal data fusion analysis; risk assessment visualization; Health management program module: generate three-level intervention programs; customize programs; predict program effects; Health warning module: multi-channel warning push; hierarchical warning mechanism; warning false alarm rate control.

3. The method for researching computing-network fusion technology according to claim 2, characterized in that: The health management business model also includes: Long-term tracking and analysis system: establish a user's health profile throughout their life cycle; use trend analysis algorithms to identify the development patterns of chronic diseases; and generate annual health reports; Population health analysis functions: generation of regional disease heat maps; epidemiological feature mining; medical resource demand prediction model.

4. The method for researching computing-network fusion technology according to claim 1, characterized in that: The computing-network fusion architecture includes edge computing nodes, cloud computing centers, and network transmission modules. Edge computing nodes are used to process users' health data in real time, while cloud computing centers are used for complex data analysis and model training. Among them, edge computing nodes include: deploying medical-specific edge servers; locally processing 80% of real-time monitoring services; supporting data desensitization processing; Cloud computing center: adopts heterogeneous computing architecture; supports billions of medical data analysis; model training cluster supports distributed TensorFlow / PyTorch; Network transmission module: medical dedicated network security; redundant design of key business channels; data compression transmission.

5. The method for researching computing-network fusion technology according to claim 4 is characterized in that: Computing and network integration technologies for health management business awareness also include: AI analysis engine: Integrates over 20 medical AI models; the federated learning framework supports cross-institutional joint modeling; the model interpretation system provides visualization of diagnostic evidence; Personalized solution generation: recommendation system based on knowledge graph; multi-objective optimization algorithm to balance efficacy and side effects; A / B testing to verify the effectiveness of the solution.

6. The method for researching computing-network fusion technology according to claim 4, characterized in that: The computing-network convergence architecture supports multiple network access methods, including 5G, Wi-Fi, and the Internet of Things.

7. The method for researching computing-network fusion technology according to claim 6, characterized in that: The computing-network fusion architecture also includes a resource management platform, which includes: Functional architecture: resource monitoring layer, policy management layer, and interface service layer; Core technologies: Digital twins enable panoramic visualization of resources, digital threads track resource flow processes, and blockchain records resource usage credentials.

8. The method for researching computing-network fusion technology according to claim 1, characterized in that: The steps for dynamically allocating computing network resources include: S31. Real-time monitoring of health management service load: deploying 500+ probes to collect load data; using a prediction algorithm to predict load 30 minutes in advance; and categorizing health services. S32. Dynamically adjust the allocation ratio of computing resources according to load conditions: containerized deployment; elastic resource pool; load balancing algorithm; S33. Optimize network transmission paths and reduce data transmission delays: Delay maps guide route selection; prioritize key data scheduling; and forward error correction reduces retransmission rates. S34. Prioritize resources based on the user's health status and service priority: multi-level queue management; dynamic weight allocation; resource preemption mechanism.

9. The method for researching computing-network fusion technology according to claim 1, characterized in that: It also includes encryption of health management business data, including: Encryption system: transport layer, storage layer and application layer; Privacy protection: differential privacy processing, k-anonymization processing, and data usage rights control.

10. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 9.

Images