Medical data security access control method and system based on cloud network side-end collaboration

Abstract

The invention discloses a medical data security access control method and system based on cloud network side-end cooperation. The method and system are used for solving the problems of privacy disclosure, data abuse, access control failure and the like in cross-institution medical data cooperation. The method provided by the invention comprises the steps that the edge device collects medical data of a user; performing preprocessing and homomorphic encryption on the medical data on the edge computing node, and uploading the encrypted medical data to a cloud platform; when a use request of a calculation participant for target medical data of a target user is received, obtaining a smart contract; checking whether the authority of the calculation participant accords with an authorization rule or not through the smart contract; if yes, activating an SMPC protocol, and generating a random seed through a VRF or a chaotic mapping function; performing encryption calculation by using a random seed to obtain an encrypted calculation result; and when the calculation participant requests to obtain the calculation result, decrypting and authorizing to access the calculation result according to the verification result of the random seed.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and system for secure access control of medical data based on cloud-network-edge-end collaboration. Background Art

[0002] Medical data contains a large amount of personal health information, including patients' medical records, diagnosis results, treatment plans, drug use history, genetic data, etc. This information not only involves personal privacy, but is also closely related to the life and health of patients. Once data leakage occurs, it will not only infringe on patients' privacy, but may also cause identity theft, financial fraud, social discrimination and other problems. In serious cases, it may even affect patients' treatment results and trust.

[0003] In the current medical field, with the continuous improvement of informatization, cross-institutional sharing and multi-party collaboration of medical data have become the norm. In the existing medical data sharing and processing scenarios, medical data is usually transmitted from local storage devices (such as hospital internal servers) to other institutions (such as research institutions) or central servers for processing in plain text or simple encrypted form. For example, the data is encrypted using a common symmetric encryption algorithm (such as AES) and then transmitted over the network. In terms of storage, data is stored in a single server or cloud storage, and access rights are managed through access control lists (ACLs).

[0004] However, in medical research, in order to conduct cross-institutional medical research or clinical trials, it is necessary to combine the data of all parties for large-scale analysis and modeling. In this process, human errors or malicious authorizations are prone to occur. Even if authorization authentication and encryption technology are used during the transmission process, the possibility of malicious access or tampering of data cannot be ruled out, which can easily lead to privacy leakage and data abuse, and it is impossible to achieve safe and effective medical data access control. Summary of the invention

[0005] The present application provides a method and system for secure access control of medical data based on cloud-network-edge collaboration, which is used to solve problems such as privacy leakage, data abuse, and access control failure in cross-institutional medical data cooperation.

[0006] The first aspect of the present application provides a method for secure access control of medical data based on cloud-network-edge-end collaboration, including: Collecting medical data of users through edge devices, wherein the edge devices include personal wearable devices and medical instruments and equipment; Preprocessing and homomorphically encrypting the medical data on the edge computing node, uploading the encrypted medical data to a cloud platform, and managing the modification and access to the medical data through blockchain technology; When receiving a request from a computing participant to use target medical data of a target user, obtaining a smart contract deployed by the target user on the blockchain, wherein the smart contract contains authorization rules related to user authorization, doctor role authentication, and medical institution compliance; Checking whether the authority of the computing participant complies with the authorization rules through the smart contract; If yes, the SMPC protocol is activated and a random seed is generated through VRF or chaos mapping function; Inputting the target medical data into a calculation process, performing encryption calculation using the random seed, and obtaining an encrypted calculation result; When the computing participant requests to obtain the computing result, the computing result is decrypted and authorized to be accessed according to the verification result of the random seed.

[0007] Optionally, the preprocessing and homomorphic encryption of the medical data on the edge computing node includes: Preprocessing the medical data on the edge computing node, wherein the preprocessing includes standardization processing, denoising processing, and abnormality correction processing; When the medical data is imaging data, a wavelet transform based on Daubechies wavelet or Symlet wavelet is applied to the medical image to decompose the medical image into different wavelet frequency components, and homomorphic encryption is performed on each of the wavelet frequency components; When the medical data is non-imaging data, the medical data is converted from multi-dimensional data into a vector form of fixed dimension, and the medical data in vector form is homomorphically encrypted using the CKKS homomorphic encryption algorithm. SIMD technology is used for parallel computing during the homomorphic encryption process.

[0008] Optionally, before preprocessing and homomorphically encrypting the medical data on the edge computing node, the method further includes: Based on the basic information of the user and the meta-information of the medical data, generating a data identifier of the medical data through zero-knowledge proof; The step of uploading the encrypted medical data to the cloud platform includes: The encrypted medical data and the data identifier are uploaded to the cloud platform.

[0009] Optionally, the activating the SMPC protocol and generating a random seed through a VRF or a chaotic mapping function includes: Activate the SMPC protocol to determine the public key information of the computing participant and the target data identifier corresponding to the target medical data; The target medical data, the public key information and the target data identifier are taken as input, and a random seed is generated through a VRF or chaotic mapping function.

[0010] Optionally, the step of taking the target medical data, the public key information and the target data identifier as input and generating a random seed through VRF includes: For each computing participant, a VRF output value is calculated based on the target medical data, the public key information and the timestamp; Performing hash processing on the public key information of all computing participants and the target data identifier to obtain a public key hash value; The public key hash value, the VRF output value and the initial random seed are combined to generate a random seed through VRF.

[0011] Optionally, the step of taking the target medical data, the public key information and the target data identifier as input and generating a random seed through a chaotic mapping function includes: converting the target medical data, the public key information and the target data identifier into an input vector; Selecting an initial state value of a chaotic mapping function in the input vector, and performing iterative calculation using a Logistic chaotic mapping function to obtain a generated sequence; The values ​​in the generated sequence are selected according to a preset rule to form a random seed.

[0012] Optionally, before preprocessing and homomorphically encrypting the medical data on the edge computing node, the method further includes: Determine whether the user's identity information is legal through decentralized identity verification on the edge computing node; If so, the medical data is preprocessed and homomorphically encrypted on the edge computing node.

[0013] The second aspect of the present application provides a system for secure access control of medical data based on cloud-network-edge-end collaboration, including: A collection unit, used to collect the user's medical data through an edge device, wherein the edge device includes a personal wearable device and a medical instrument device; An encryption unit, used to pre-process and homomorphically encrypt the medical data on the edge computing node, and upload the encrypted medical data to a cloud platform, where the cloud platform is used to manage the modification and access to the medical data through blockchain technology; An acquisition unit, configured to, when receiving a request from a computing participant to use target medical data of a target user, acquire a smart contract deployed by the target user on a blockchain, wherein the smart contract contains authorization rules related to user authorization, doctor role authentication, and compliance of a medical institution; A checking unit, configured to check whether the authority of the computing participant complies with the authorization rule through the smart contract; A generating unit, configured to activate the SMPC protocol and generate a random seed through a VRF or a chaotic mapping function when the inspection of the inspection unit passes; A computing unit, used for inputting the target medical data into a computing process, performing encryption computing using the random seed, and obtaining an encrypted computing result; The authorization unit is used to decrypt and authorize access to the calculation result according to the verification result of the random seed when the calculation participant requests to obtain the calculation result.

[0014] A third aspect of the present application provides a device for secure access control of medical data based on cloud-network-edge-end collaboration, the device comprising: Processor, memory, input-output unit, and bus; The processor is connected to the memory, the input and output unit, and the bus; The memory stores a program, and the processor calls the program to execute the first aspect and any optional method of medical data security access control based on cloud-network-edge-end collaboration in the first aspect.

[0015] The fourth aspect of the present application provides a computer-readable storage medium, on which a program is stored. When the program is executed on a computer, the method for secure access control of medical data based on cloud-network-edge-end collaboration according to the first aspect and any optional item in the first aspect is executed.

[0016] It can be seen from the above technical solutions that this application has the following advantages: First, medical data is collected through edge devices, and edge computing nodes immediately perform preprocessing and homomorphic encryption, reducing the risk of data being maliciously acquired in the initial stage. The processed data is then uploaded to the cloud platform, and the modification and access of the data are managed with the help of blockchain technology. The tamper-proof nature of the blockchain ensures the authenticity and traceability of data operation records. When the computing participant makes a request to use the target medical data, the permission check is performed according to the authorization rules in the smart contract on the blockchain. The smart contract comprehensively considers multiple factors such as user authorization, doctor role authentication, and medical institution compliance, so as to accurately judge the permissions of the computing participant and effectively avoid human errors or malicious authorization. If the permissions meet the requirements, the SMPC protocol is activated and a random seed is generated through VRF or chaotic mapping function. The random seed further enhances the confidentiality and security of the data in the subsequent encryption calculation process, and the privacy will not be leaked even if the data is in the calculation process. Finally, only when the computing participant requests to obtain the result and passes the random seed verification, the authorization access calculation result is decrypted to achieve safe and effective medical data access control, which fundamentally solves the problems of privacy leakage, data abuse, and access control failure in traditional methods in cross-institutional medical data cooperation. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solution in the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A flowchart of an embodiment of a method for secure access control of medical data based on cloud-network-edge-end collaboration provided in this application; Figure 2 A flowchart of another embodiment of the method for secure access control of medical data based on cloud-network-edge-end collaboration provided in this application; Figure 3 A schematic diagram of the structure of an embodiment of a system for secure access control of medical data based on cloud-network-edge-end collaboration provided in this application; Figure 4 A schematic diagram of the physical structure of an embodiment of a device for secure access control of medical data based on cloud-network-edge-end collaboration provided in this application. DETAILED DESCRIPTION

[0019] The present application provides a method and system for secure access control of medical data based on cloud-network-edge collaboration, which is used to solve problems such as privacy leakage, data abuse, and access control failure in cross-institutional medical data cooperation.

[0020] See also Figure 1 , Figure 1 An embodiment of a method for secure access control of medical data based on cloud-network-edge-end collaboration provided in this application includes: 101. Collect users’ medical data through edge devices, including personal wearable devices and medical instruments and equipment; A single type of medical data can often only reflect one aspect of a patient's health status. Modern medicine increasingly emphasizes multi-dimensional, comprehensive diagnosis and treatment methods, and the patient's health status changes dynamically. Traditional medical data collection methods mainly rely on patients to go to medical institutions for regular checkups, which can only obtain data at discrete time points. However, the development process of many diseases is continuous. For example, there may be gradual changes in physiological parameters such as heart rate and blood pressure before a heart attack. Real-time or regular collection of users' medical data through edge devices can provide more comprehensive support for users' medical decisions.

[0021] Among them, personal wearable devices refer to electronic devices that can be worn directly on the human body and can monitor human physiological parameters, movement status or environmental information in real time, such as smart bracelets, smart watches, etc. Medical instruments and equipment refer to professional equipment used in medical institutions for medical processes such as disease diagnosis, treatment, monitoring or rehabilitation. These devices are based on different physical principles and technical means and can generate various types of medical data, such as imaging data (X-ray, CT, MRI and other images), electrophysiological data (electrocardiogram, electroencephalogram, etc.), biochemical test data (blood, urine and other test results), etc.

[0022] Wearable devices and medical instruments can collect users' medical data in real time, perform preliminary processing close to the data source, reduce latency and shorten the time it takes to transmit data to the cloud. Medical data is distributed in different scenarios, and edge devices can make data collection more decentralized, avoiding single system bottlenecks and reducing dependence on central servers.

[0023] 102. Preprocess and homomorphically encrypt medical data on edge computing nodes, and upload the encrypted medical data to the cloud platform, which is used to manage the modification and access of medical data through blockchain technology; As a computing facility close to the data source or user end, the edge computing node will integrate the data received from personal wearable devices and medical instruments and equipment. The edge computing node can be a small server, gateway device, or a dedicated edge computing box deployed inside a medical institution, and the specific details are not limited here. On the edge computing node, the collected medical data is preprocessed and homomorphically encrypted. Through homomorphic encryption, the medical data remains encrypted throughout the entire processing process, and the original data will not be leaked even during the calculation and transmission process, thereby effectively protecting patient privacy.

[0024] After preprocessing and encryption, the medical data is uploaded to the cloud platform. As the core part of data storage and management, the cloud platform combines with blockchain technology to record each data upload, modification and access operation. All operations on medical data (such as data upload, permission change, data modification, etc.) will be recorded on the blockchain to ensure the data is tamper-proof, transparent and traceable.

[0025] 103. When receiving a computing participant's request to use the target medical data of the target user, obtain a smart contract deployed by the target user on the blockchain, wherein the smart contract contains authorization rules related to user authorization, doctor role authentication, and medical institution compliance; In the scenario of medical data sharing and collaboration, with the increase in the number of participants and the increase in data sensitivity, it is crucial to ensure the legal and compliant use of data. Traditional methods based on manual review and simple permission settings are difficult to meet the complex and dynamically changing medical data access needs. On the one hand, the access needs of different users (such as doctors, researchers, etc.) to medical data vary depending on their roles; on the other hand, when medical institutions cooperate with each other or with external research institutions, it is necessary to strictly control the compliance of data outflow and use. As an automated execution contract based on blockchain, smart contracts have the characteristics of being tamper-proof and automatically executed, and can achieve refined and automated management of medical data access rights.

[0026] Based on this, when the receiving computing participant makes a request to use the target user's target medical data, the cloud platform will first locate the smart contract deployed by the target user on the blockchain. This smart contract was created and deployed on the blockchain when the target user initially connected his medical data to the data management system. Specifically, the cloud platform can identify the target user's unique identifier (such as patient ID, etc.), find the corresponding smart contract address in the blockchain ledger, and then obtain the complete smart contract content.

[0027] The smart contract contains a series of detailed and rigorous authorization rules, including but not limited to user authorization, doctor role authentication, and medical institution compliance. For example, in terms of user authorization, it will be clearly stipulated which parts of the target user's medical data can be authorized to others for use, and under what conditions (such as a specific time period, a specific research project, etc.) the authorization can be made. In the doctor role authentication section, the access rights of doctors in different departments and at different levels to various types of medical data can be listed in detail. For example, the attending physician of the cardiology department can access the electrocardiogram, cardiac ultrasound and other related data of the patients he is responsible for, but for some special genetic testing data, only doctors with relevant genetic research qualifications may have access. For the compliance of medical institutions, it will be made clear that only other medical institutions or research institutions that have a cooperative relationship with the target user's medical institution and are within the scope of cooperation can obtain the corresponding access rights. In addition, the specific matters of cooperation (such as the name of the cooperation project, start and end time, etc.) will be defined to ensure that the use of data complies with the cooperation agreement.

[0028] It should be noted that computing participants specifically refer to individuals, organizations or systems that participate in the operation of target medical data in medical data processing, analysis or collaboration scenarios. These participants usually have different roles and purposes, such as doctors, researchers, data analysis departments of medical institutions, external scientific research institutions, etc. Computing participants need to access and use target medical data for the purpose of diagnosis and treatment, medical research, statistical analysis of medical data, etc.

[0029] 104. Check whether the permissions of computing participants comply with authorization rules through smart contracts; After obtaining the smart contract, the cloud platform will automatically extract the authorization rules and compare and verify the relevant information of the computing participants (such as identity, organization, role, etc.) with these authorization rules. The verification process is automatically executed based on the code logic without human intervention. For example, if the computing participant is a doctor, the system will first confirm the authenticity of his doctor's identity (by checking with the doctor registration information database of the medical institution, etc.), and then determine whether he has the right to access the target medical data based on his department, level, and the access rights of the doctor in this role specified in the smart contract. During the verification process, hash processing and encryption comparison of various data and information can also be combined to ensure the accuracy and security of the information.

[0030] When it is determined that the authority of the computing participant meets the user's authorization rules, the subsequent step 105 is executed. When it is determined that the authority of the computing participant does not meet the user's authorization rules, the computing participant's access request to the target medical data is rejected, and the authority non-compliance is recorded in detail for subsequent audit and analysis.

[0031] 105. Activate the SMPC protocol and generate a random seed through VRF or chaos mapping function; When the smart contract checks determine that the permissions of the computing participants meet the user's authorization rules, the system automatically triggers the activation of the SMPC protocol. The SMPC protocol implements privacy protection through a series of complex cryptographic techniques and algorithms. In the collaborative computing process, the private data of each participant is encrypted, and a specific computing method is used so that the final correct computing result can only be obtained when all participants participate in the calculation and execute according to the protocol. In this process, no party can obtain the private data of other participants alone.

[0032] In the SMPC protocol, in order to protect the data privacy of all parties and the security of the calculation results, it is necessary to generate a random seed, which will be used for encryption calculation and result verification. Specifically, the random seed generation method includes VRF or chaotic mapping function, where VRF generates verifiable random numbers by using encryption keys, and the participants can prove that the generated random numbers are generated by predetermined rules without the possibility of tampering. The chaotic mapping function uses the unpredictability of the chaotic system to generate complex random seeds. Chaotic mapping is highly sensitive, that is, a small change in the initial conditions will lead to a huge difference in the output results, ensuring the unpredictability of random numbers. By using VRF or chaotic mapping functions to generate random seeds, the calculation participants can verify whether the generated random seeds are fair and just, and no party can manipulate the random seed generation process.

[0033] 106. Input the target medical data into the calculation process, use the random seed to perform encryption calculation, and obtain the encrypted calculation result; First, the target medical data is preprocessed as necessary to meet the format requirements of subsequent encryption calculations. Then the random seed is integrated into the operation process of the encryption algorithm. Taking the symmetric encryption algorithm as an example, assuming that the AES algorithm is used, when performing encryption operations, in addition to using the target medical data as input, the random seed will also be involved in the encryption operation in some way (such as as part of the key expansion, or combined with the original key, etc.). Due to the randomness of the random seed during each calculation, even for the same target medical data, different encryption results will be obtained, greatly increasing the confidentiality of the data.

[0034] After the above encryption calculation process, the final calculation result is in encrypted form. The encrypted calculation result is an encrypted value, vector, matrix or other data structure. The specific form depends on the original type of the target medical data and the calculation task and encryption algorithm used. It should be noted that for some complex calculation tasks, multiple iterative calculations may be required. In each iteration, it is necessary to ensure that the random seed continues to play a role to maintain the randomness and confidentiality of the encryption calculation. For example, when performing machine learning model training calculations based on medical data, it may be necessary to update the model parameters multiple times. Each iteration requires re-encrypting the data based on the random seed to ensure the privacy of the data throughout the training process.

[0035] 107. When a computing participant requests to obtain the computing result, the computing result is decrypted and authorized to be accessed based on the verification result of the random seed.

[0036] After the calculation process is completed, the calculation participants can obtain the final calculation results by making a request. At this time, the calculation results are still encrypted and must be legally authorized to be decrypted and accessed. When requesting the calculation results, the system needs to check whether the participant has the authority to obtain the calculation results according to the pre-set smart contract rules. After that, the random seed used for the encryption process is recalculated using the verification information provided by the participant (such as public key, data identifier, etc.). The cloud platform ensures the integrity of the calculation and the reliability of the results by verifying whether the generated random seed is consistent with the random seed used in the calculation process. If the verification is successful, it means that the calculation process has not been tampered with and the decryption process can continue.

[0037] When the random seed is verified, the system will decrypt the encrypted calculation results according to the preset key or decryption protocol. The decryption process requires the keys of each participant (such as partial keys generated by the SMPC protocol), or a specific decryption algorithm to restore the plaintext data. The decrypted results will be authorized to the legitimate calculation participants according to the authorization rules in the smart contract. For example, only requesters who meet the "doctor identity authentication" and "data compliance" rules can access the results. If it is used for clinical research, it may only be authorized to relevant research institutions and compliance personnel.

[0038] In practical applications, this application is particularly suitable for remote medical diagnosis scenarios and medical research cooperation scenarios, as illustrated below: In telemedicine, patients may continuously collect health data through personal wearable devices (such as smart bracelets, home blood pressure monitors, etc.), or undergo examinations at primary medical institutions through medical instruments and equipment (such as blood glucose meters, electrocardiographs, etc.). The medical data collected by these edge devices will first be processed at the edge computing node. For example, a patient with diabetes uses a blood glucose meter to measure blood glucose data at home. After the edge computing node confirms the legitimacy of the patient's identity through decentralized identity authentication, it pre-processes and homomorphically encrypts the blood glucose data and then uploads it to the cloud platform. When a specialist initiates a request to use the patient's target medical data (here, blood glucose data), the cloud platform obtains the smart contract deployed by the patient on the blockchain. The contract stipulates that only doctors who are authorized by the patient and have the corresponding professional qualification certification (doctor role certification) and the compliance of the medical institution can access the data. If the doctor meets the authorization rules, the SMPC protocol is activated and a random seed is generated. The encrypted blood glucose data is input into the calculation process, and the encrypted calculation results such as the analysis of the patient's blood glucose fluctuation trend are obtained by using the random seed encryption calculation. When a doctor requests the results, the data is decrypted and authorized based on the verification result of the random seed. The doctor can then develop an accurate remote treatment plan for the patient based on the data. During this process, the patient's medical data is always under secure access control to prevent data leakage and illegal use.

[0039] In the medical research cooperation scenario, multiple medical institutions or research institutions may cooperate to conduct a study on the treatment effect of a certain disease. Each institution collects a variety of medical data of patients through edge devices, performs homomorphic encryption at the edge computing node, and then uploads it to the cloud platform. When the research team, as a computing participant, requests to use the target medical data of the target patient, the cloud platform checks the authority of the research team based on the blockchain smart contract, such as whether the patient's research authorization has been obtained, whether the professional qualifications of the research team members meet the requirements, and whether the research institution is compliant. If the authority meets the requirements, a random seed is generated through a series of complex operations for encryption calculation to obtain the encrypted calculation results required for the research (such as the comparative analysis results of various indicators of patients under different treatment plans). Finally, the research team obtains the decrypted calculation results under the verification conditions. This can not only ensure the safe sharing of patient medical data in multi-institutional scientific research cooperation, but also enable scientific research work to proceed smoothly and promote the development of medical research.

[0040] In this embodiment, first, medical data is collected through edge devices, and edge computing nodes are immediately preprocessed and homomorphically encrypted, reducing the risk of data being maliciously acquired in the initial stage. The processed data is then uploaded to the cloud platform, and the modification and access of the data are managed with the help of blockchain technology. The tamper-proof nature of the blockchain ensures the authenticity and traceability of the data operation records. When the computing participant makes a request to use the target medical data, the authority check is performed according to the authorization rules in the smart contract on the blockchain. The smart contract comprehensively considers multiple factors such as user authorization, doctor role authentication, and medical institution compliance, so as to accurately judge the authority of the computing participant, effectively avoiding the occurrence of human errors or malicious authorization. If the authority meets the requirements, the SMPC protocol is activated and a random seed is generated through VRF or chaotic mapping function. The random seed further enhances the confidentiality and security of the data in the subsequent encryption calculation process, and the data will not leak privacy even during the calculation process. Finally, only when the computing participant requests to obtain the result and passes the random seed verification, the authorization access calculation result is decrypted to achieve safe and effective medical data access control, fundamentally solving the problems of privacy leakage, data abuse, and access control failure in traditional methods in cross-institutional medical data cooperation.

[0041] The following is a detailed description of the method for secure access control of medical data based on cloud-network-edge-end collaboration provided by this application. Figure 2 , Figure 2 Another embodiment of the method for secure access control of medical data based on cloud-network-edge-end collaboration provided by the present application includes: 201. Collect user medical data through edge devices, including personal wearable devices and medical instruments and equipment; In this embodiment, step 201 is similar to step 101 in the aforementioned embodiment and will not be described again here.

[0042] 202. Determine whether the user's identity information is legal through decentralized identity authentication on the edge computing node; In this embodiment, before the edge computing node processes the user's medical data, it is first necessary to determine whether the user's identity information is legal through decentralized identity authentication. Specifically, the user needs to register on the cloud platform and submit personal information (such as identity proof, health data, medical insurance information, etc.). After encryption, this information is generated on a decentralized identity management platform, such as a blockchain-based identity system, to generate a digital identity. The digital identity can contain a public key, identity authentication credentials, and other necessary metadata. Each user has a unique digital identity on the decentralized identity platform. The user holds a private key corresponding to the digital identity, which can prove his identity. The private key is stored on the user's device and managed by the user himself to ensure that the private key will not be leaked.

[0043] Authentication through edge computing nodes can authenticate users before data is uploaded, preventing illegal or unauthorized users from accessing the system and reducing the risk of identity theft or unauthorized users uploading incorrect data. In addition, decentralized identity authentication ensures the privacy and control of user identities, further protecting the privacy and security of medical data. Subsequent steps will only be executed when the user's identity information is confirmed to be legitimate.

[0044] 203. Generate a data identifier for medical data through zero-knowledge proof based on the user's basic information and the meta-information of medical data; After confirming that the user's identity information is legal, collect the user's basic information and the meta-information of the medical data. The user's basic information may include name, age, gender, ID number, etc., which can distinguish and locate the user to a certain extent. The meta-information of medical data covers the data type (such as electrocardiogram data, blood test data, etc.), collection time, collection equipment, etc., which helps to further clarify the specific characteristics and source of each piece of medical data. The user's basic information and the meta-information of the medical data are used as input, and a data identifier is generated according to the rules and processes of the zero-knowledge proof algorithm (such as zk-SNARKs or zk-STARKs). The data identifier generated by zero-knowledge proof is an encrypted identifier that can uniquely identify a specific piece of medical data. The data identifier is associated with the meta-information of the medical data, but does not directly disclose any privacy content. Only with the user's authorization can the relevant party obtain more information through the identifier.

[0045] Afterwards, the generated data identifier is stored in association with the corresponding medical data, so that in subsequent data processing, query, verification and other operations, the data identifier can be used to quickly and accurately locate and identify the relevant medical data, while ensuring that the privacy of the medical data is not leaked.

[0046] 204. Preprocess the medical data on the edge computing node, including standardization, denoising and anomaly correction; Since medical data collected from various sources often have problems such as inconsistent formats, noise, and possible outliers, these problems will affect the subsequent accurate analysis, calculation, and sharing of medical data, so it is necessary to pre-process the medical data on the edge computing node. Specific pre-processing operations include but are not limited to standardization, denoising, and anomaly correction. After pre-processing, the format of medical data will be more unified and concise, and obvious abnormal data can be excluded, thus providing a reliable foundation for subsequent medical data analysis, calculation, sharing, and other applications.

[0047] After the medical data is processed and before homomorphic encryption is performed, the medical data needs to be divided into imaging data and non-imaging data. Different encryption methods are used for different types of medical data to improve encryption efficiency while ensuring the encryption effect. This is based on the fact that imaging data and non-imaging data are very different in structure. Imaging data is two-dimensional or three-dimensional image data, containing a large amount of pixel information, and the data volume is usually large and has spatial correlation; non-imaging data may be multi-dimensional numerical or text data, with a relatively simple data structure, but may contain a variety of different types of data (such as numerical values, text, etc.). This structural difference leads to different applicability of encryption methods. In addition, in the subsequent medical data processing process, the computational requirements for imaging data and non-imaging data are also different. For example, for imaging data, it may be necessary to perform image processing operations such as filtering and edge detection in an encrypted state; for non-imaging data, it may involve more calculations such as statistical analysis and machine learning model training. These different computational requirements require different encryption methods so that the corresponding calculations can be effectively performed after encryption.

[0048] 205. When the medical data is imaging data, a wavelet transform based on Daubechies wavelet or Symlet wavelet is applied to the medical image to decompose the medical image into different wavelet frequency components, and homomorphic encryption is performed on each wavelet frequency component; Image data mainly refers to image data generated by medical imaging equipment (such as X-ray machines, CT scans, MRI, ultrasound equipment, etc.). For example, medical images (such as CT scan images, MRI images, X-rays, ultrasound images, etc.) are all image data. Image data is usually a high-dimensional, complex image file, and special attention should be paid to image features and data dimensions when processing.

[0049] In this embodiment, when the medical data is imaging data, a wavelet transform method based on Daubechies wavelet or Symlet wavelet is first selected, and the selected wavelet transform is applied to the medical image to decompose it into different wavelet frequency components. Taking a two-dimensional medical image as an example, the entire medical image is decomposed into different wavelet frequency components by taking each row and each column of the image as input for wavelet transform operation, and each frequency band represents different details of the image (such as high-frequency components contain edge details of the image, and low-frequency components contain general contour information). Thereafter, each component is homomorphically encrypted separately. For imaging data, a homomorphic encryption algorithm that supports encrypted calculations can be used, such as the Paillier homomorphic encryption algorithm or the BFV homomorphic encryption algorithm. These encryption algorithms can support addition and multiplication operations on encrypted data to ensure privacy protection during the calculation process. Through homomorphic encryption, the specific content of imaging data will not be exposed, and calculations can be performed in an encrypted state (for example, image segmentation, feature extraction, etc.).

[0050] 206. When the medical data is non-imaging data, the medical data is converted from multi-dimensional data to a fixed-dimensional vector form, and the CKKS homomorphic encryption algorithm is used to homomorphically encrypt the vector medical data. SIMD technology is used for parallel computing during the homomorphic encryption process; Non-imaging data: including structured data such as basic patient information, medical history, test results, sensor data, etc., usually presented in digital, text, time series and other formats. For example, blood sugar levels, electrocardiogram (ECG) data, medical records, laboratory test reports, etc. are non-imaging data. Therefore, in this embodiment, when the medical data is non-imaging data, it must first be converted from multidimensional data to a vector form of fixed dimension. This process requires determining the appropriate vector dimension based on the specific characteristics of the data and subsequent computing requirements. On the one hand, it is necessary to ensure that the vector can completely contain all important data elements that need to be processed, and no key information should be omitted. On the other hand, the appropriate vector dimension should be determined based on the type of calculation that may be performed later (such as statistical analysis, machine learning model training, etc.), to avoid excessive increase in computational complexity due to too high a dimension, or too low a dimension to meet the computing requirements.

[0051] For example, when training a machine learning model based on medical data, if you need to consider multiple physiological parameters and their relationships, you may need to set a relatively high vector dimension to accommodate enough information. However, if you are only doing simple statistical analysis, such as calculating the mean, standard deviation, etc., you may only need a relatively low-dimensional vector to meet your needs. For example, for patient data containing multiple physiological parameters (such as heart rate, blood pressure, blood oxygen, etc.), these parameters may be arranged in a certain order to form a vector of fixed dimension. Assuming that the four physiological parameters of heart rate, blood pressure, blood oxygen, and body temperature are selected, they can be arranged in sequence into a four-dimensional vector, such as [heart rate value, blood pressure value, blood oxygen value, body temperature value].

[0052] After that, the medical data converted into vector form is homomorphically encrypted using the CKKS homomorphic encryption algorithm. The CKKS homomorphic encryption algorithm is mainly used to process real or complex vector data, allowing specific algebraic operations on encrypted data. The decrypted result is the same as the result of the same operation on the plaintext. It should be noted that in the process of homomorphic encryption for non-imaging data, SIMD technology is required for parallel computing. SIMD technology can package multiple data elements in one instruction for processing, thereby improving computing efficiency. For example, for the above-mentioned medical data converted into vector form, SIMD technology can be used to process multiple elements in the vector (such as different physiological parameter values) at the same time, instead of processing them one by one, which can greatly shorten the calculation time of homomorphic encryption.

[0053] Steps 205 and 206 perform homomorphic encryption on the medical data to ensure the privacy of the data. Different encryption methods are used for imaging data and non-imaging data, which can adapt to the characteristics of different types of data, thereby improving encryption efficiency and computing performance.

[0054] 207. Upload the encrypted medical data and data identifiers to the cloud platform, which is used to manage the modification and access of medical data through blockchain technology; The encrypted medical data and data identifiers are uploaded to the cloud platform, and the modification and access of medical data are managed using blockchain technology, thus achieving the security, reliability and compliance of medical data during storage, sharing and use. The uploaded medical data and data identifiers will be stored in each node on the blockchain, and a corresponding indexing mechanism can be established to enable quick and accurate query and location of specific medical data. For example, data identifiers, user information, timestamps, etc. can be used as index conditions to facilitate the rapid search for target medical data when data needs to be accessed or modified later.

[0055] It is known that the data identifier is generated based on the user's basic information and medical data metadata through zero-knowledge proof and is unique. In the case where a large amount of complex and diverse medical data is stored on the cloud platform, each piece of medical data can be quickly and accurately identified through the data identifier. In the process of generating data identifiers through zero-knowledge proof, the specific content of the medical data will not be leaked. In the subsequent data query, sharing and other operations after the data identifier is uploaded to the cloud platform, other users or systems can only use the data identifier to determine whether it is the required data without having to access the actual content of the medical data. In the cloud platform, an environment where multiple users and multiple institutions share data, it can effectively prevent the unnecessary exposure of sensitive medical information.

[0056] 208. When receiving a request from a computing participant to use the target medical data of a target user, obtain a smart contract deployed by the target user on the blockchain, wherein the smart contract contains authorization rules related to user authorization, doctor role authentication, and medical institution compliance; 209. Check whether the permissions of computing participants comply with authorization rules through smart contracts; In this embodiment, step 208-step 209 are similar to step 103-step 104 of the aforementioned embodiment, and are not described again here.

[0057] 210. Activate the SMPC protocol to determine the public key information of the computing participant and the target data identifier corresponding to the target medical data; When the relevant preconditions are met, the cloud platform automatically activates the SMPC protocol. After receiving the activation notification, each computing participant performs initialization settings according to the requirements of the SMPC protocol based on its own computing environment and the data it has. For example, prepare a local data copy for collaborative computing and configure relevant computing parameters, such as encryption keys, computing algorithms, etc.

[0058] In the SMPC protocol, each computing participant usually uses a public key to encrypt his or her own data, while the private key is held only by the party itself. The public key information is the identifier of the computing party, ensuring the security of the data during the computing process. Therefore, for each computing participant, the cloud platform needs to extract its public key information, which usually includes the public key itself and some metadata related to the public key, such as the owner of the public key, the generation time of the public key, etc. By determining the public key information of each computing participant, it is ensured that each participant uses a legitimate key to encrypt and decrypt data during the computing process, avoiding malicious computing parties from performing malicious operations by forging public keys. In addition, the cloud platform also needs to determine the target data identifier corresponding to the target medical data. The identifier of the target medical data ensures the uniqueness, traceability and verifiability of the data, so that each participant in the multi-party computing can accurately confirm the target data in which he or she participates, avoiding the risk of confusion or data tampering, and ensuring that the correctness of the data involved can be tracked and verified throughout the SMPC protocol.

[0059] 211. Taking the target medical data, public key information and target data identifier as input, a random seed is generated through a VRF or chaotic mapping function; In this embodiment, the target medical data, public key information and target data identifier are used as inputs, and a random seed is generated through a VRF or chaotic mapping function. The random seed is used to control the encryption process, ensure the unpredictability and tamperability of the calculation process, and encrypt the data. In a multi-party computing scenario, the generated random seed can ensure the fairness of the calculation process and prevent the calculation results from being manipulated or tampered by a single party. By using the target medical data itself and the target data identifier associated with it as input, it can be ensured that each specific data set will generate a unique random seed, and even the same computing task will generate different random seeds on different data sets, thereby increasing privacy protection. In addition, the target medical data itself contains sensitive patient information, and the target data identifier may also indirectly reveal some data-related information. By combining them with the random seed generation process, the randomness of the random seed can be used to hide or confuse this information in subsequent encryption calculations. In addition, the public key information of the participants will also be used as one of the inputs to ensure that the participants will not tamper with the random seed generation process on their own. Each computing participant can only use his or her own private key to generate the corresponding public key, and the public key is public. Any party can verify the authenticity of the generated random seed, so that the random seed has a clear chain of evidence in the generation and verification process, and the difficulty of data tampering is greatly increased.

[0060] Specifically, depending on the specific application scenario and security requirements, choose to use a verifiable random function (VRF) or a chaotic mapping function to generate random seeds. In practical applications, VRF is more suitable for scenarios that require fairness and transparency, especially when multiple parties work together and need to verify the random number generation process. The chaotic mapping function is more suitable for scenarios with high requirements for speed and computing efficiency, especially in applications with large amounts of data or that require fast responses. The following are explained separately: 1. VRF: A1. For each computing participant, the VRF output value is calculated based on the target medical data, public key information and timestamp; In the process of selecting VRF to generate random seeds, for each computing participant, the target medical data, the public key information of the computing participant and the timestamp must be accurately obtained first. The target medical data is the core medical data content around which this calculation revolves. The public key information contains the public key itself and related metadata. The timestamp is used to record the time point of the current operation. It can provide a reference for the time dimension of the calculation process to ensure the timeliness and sequence of the calculation. The target medical data, public key information and timestamp are used as input parameters of the verifiable random function (VRF), and a series of complex cryptographic operations such as hash operations and modulo operations are performed according to the specific calculation rules of VRF. These operations process the input information in a nonlinear way, so as to calculate the corresponding VRF output value for each computing participant.

[0061] A2. Hash the public key information and target data identifier of all computing participants to obtain a public key hash value; The public key information and target data identifier of all computing participants are hashed. The hash function converts the input information into a hash value of fixed length. This hash value is one-way, that is, it is difficult to infer the original input information from the hash value, and the probability of different input information generating the same hash value is extremely low, which is collision-resistant. The public key hash value obtained by hashing is a consensus identifier generated by combining the information of all computing participants, which is used to ensure that the generated random seed is not only related to a single participant, but also takes into account the consensus of the entire multi-party computation.

[0062] A3. Combine the public key hash value, VRF output value and the initial random seed to generate a random seed through VRF.

[0063] The public key hash value, VRF output value and initial random seed are combined to generate random seeds through VRF, further increasing the complexity of generating random seeds and ensuring their unpredictability and security. For example, they can be combined together through operations such as addition and multiplication (depending on the algorithm settings of VRF), and after further calculation and processing, a random seed is finally generated through VRF. It should be noted that the initial random seed, as the initial condition of the calculation, can be a preset value, or an external random source (such as timestamp, external event, etc.) can be introduced to increase diversity.

[0064] By combining multiple factors (public key, VRF output, initial random seed) to generate random seeds, the security of the calculation can be enhanced. This combination makes the generated random seeds less susceptible to control or prediction by a single factor. In addition, different information sources and encryption algorithms are mixed in the process, and the generated random seeds are difficult to be manipulated individually, ensuring privacy during the calculation process. The random seeds generated in this way are not only random, but VRF also provides verifiability. Any participant can verify the generation process and correctness of the random seeds, avoiding the possibility of human tampering.

[0065] 2. Chaotic mapping function: B1, converting the target medical data, public key information and target data identifier into an input vector; In the process of selecting the chaotic mapping function to generate random seeds, the target medical data, public key information and target data identifier are first integrated and transformed to form an input vector. Specifically, for the target medical data, if it is numerical data (such as physiological index values, etc.), it can be normalized so that its value range meets the requirements of the chaotic mapping function, and it can usually be converted into a value in the interval [0, 1]. For character data (such as part of the character content in the public key information, the target data identifier, etc.), operations such as encoding conversion can be performed to convert it into a form suitable for numerical calculation. After these processes, the processed target medical data, public key information and target data identifier are arranged and combined in a certain order into an input vector. For example, the target medical data part can be placed first, and then the public key information and target data identifier parts can be placed in sequence to form a complete vector form.

[0066] B2. Select the initial state value of the chaotic mapping function in the input vector, use the Logistic chaotic mapping function to perform iterative calculation, and obtain the generated sequence; In the constructed input vector, the initial state value of the chaotic mapping function needs to be selected. The selection method can be carried out according to the preset rules. A common selection method is to select the first element of the input vector as the initial state value. For example, if the input vector is arranged in the above order, the first value (after normalization) or the first character (after encoding conversion) of the target medical data part can be used as the initial state value. It is also possible to use a random selection method to randomly select one from all elements of the input vector as the initial state value. Regardless of which method is used, it is necessary to ensure that the selected initial state value can accurately reflect the characteristics of the input vector, and can generate a generation sequence with good randomness based on this in subsequent iterative calculations.

[0067] The selected Logistic chaos mapping function is used for iterative calculation. The form of the Logistic chaos mapping function is: ,in represents the state value of the nth iteration, is the control parameter, and usually takes a value between [3.57, 4] to show good chaotic characteristics. Substitute the selected initial state value into the Logistic chaotic mapping function as , and then perform multiple iterations according to the function formula. For example, from Start by calculating , , …and a series of state values, and these state values ​​form a generation sequence.

[0068] B3. Select values ​​in the generated sequence according to preset rules to form a random seed.

[0069] The random seed is formed by selecting values ​​from the generated sequence according to a preset rule. The preset rule can be set in a variety of ways, which are not limited here. For example, a value can be selected at a certain interval, such as selecting one every 3 values, and the selected values ​​are combined together to form a random seed. Or, several values ​​are selected according to a certain probability, such as selecting a value in the generated sequence with a probability of 0.5, and the selected values ​​are combined to form a random seed.

[0070] 212. Input the target medical data into the calculation process, use the random seed to perform encryption calculation, and obtain the encrypted calculation result; 213. When a computing participant requests to obtain the computing result, the computing result is decrypted and authorized to be accessed based on the verification result of the random seed.

[0071] In this embodiment, step 212-step 213 are similar to step 106-step 107 of the aforementioned embodiment, and are not described again here.

[0072] The following is a detailed description of the system of the method for secure access control of medical data based on cloud-network-edge-end collaboration provided by this application. Figure 3 , Figure 3 An embodiment of a system for a method for secure access control of medical data based on cloud-network-edge-end collaboration provided in this application includes: A collecting unit 301 is used to collect the user's medical data through edge devices, where the edge devices include personal wearable devices and medical instruments and equipment; An encryption unit 302 is used to pre-process and homomorphically encrypt the medical data on the edge computing node, and upload the encrypted medical data to the cloud platform, which is used to manage the modification and access of the medical data through blockchain technology; The acquisition unit 303 is used to acquire the smart contract deployed by the target user on the blockchain when receiving a request from a computing participant to use the target medical data of the target user, wherein the smart contract contains authorization rules related to user authorization, doctor role authentication, and medical institution compliance; A checking unit 304 is used to check whether the authority of the computing participant complies with the authorization rules through the smart contract; The generating unit 305 is used to activate the SMPC protocol and generate a random seed through a VRF or a chaotic mapping function when the inspection of the inspection unit passes; The calculation unit 306 is used to input the target medical data into the calculation process, perform encryption calculation using a random seed, and obtain an encrypted calculation result; The authorization unit 307 is used to decrypt and authorize access to the calculation result according to the verification result of the random seed when the calculation participant requests to obtain the calculation result.

[0073] Optionally, the encryption unit 302 is specifically used for: Preprocess the medical data on the edge computing node, including standardization, denoising, and anomaly correction. When the medical data is imaging data, wavelet transform based on Daubechies wavelet or Symlet wavelet is applied to the medical image to decompose the medical image into different wavelet frequency components, and homomorphic encryption is performed on each wavelet frequency component; When the medical data is non-imaging data, the medical data is converted from multi-dimensional data to a vector form of fixed dimension, and the CKKS homomorphic encryption algorithm is used to homomorphically encrypt the medical data in vector form. SIMD technology is used for parallel computing during the homomorphic encryption process.

[0074] Optionally, the encryption unit 302 is further configured to: Based on the user's basic information and the meta-information of the medical data, a data identifier for the medical data is generated through zero-knowledge proof; Upload the encrypted medical data and data identifiers to the cloud platform.

[0075] Optionally, the generating unit 305 is specifically configured to: Activate the SMPC protocol to determine the public key information of the computing participants and the target data identifier corresponding to the target medical data; The target medical data, public key information and target data identifier are taken as input, and a random seed is generated through a VRF or chaotic mapping function.

[0076] Optionally, the generating unit 305 is further configured to: For each computing participant, the VRF output value is calculated based on the target medical data, public key information and timestamp; Hash the public key information and target data identifier of all computing participants to obtain a public key hash value; Combine the public key hash value, VRF output value and the initial random seed to generate a random seed through VRF.

[0077] Optionally, the generating unit 305 is further configured to: converting target medical data, public key information, and target data identifier into an input vector; Select the initial state value of the chaotic mapping function in the input vector, use the Logistic chaotic mapping function to perform iterative calculation, and obtain the generated sequence; The values ​​in the generated sequence are selected according to preset rules to form a random seed.

[0078] Optionally, the system further includes: A verification unit 308, used to determine whether the user's identity information is legal through decentralized identity verification on the edge computing node; The encryption unit 302 is specifically used to pre-process and homomorphically encrypt the medical data on the edge computing node when the verification result of the verification unit 308 is legal.

[0079] In this embodiment, the functions of each unit are the same as those described above. Figure 1 or Figure 2 The steps in the method embodiment shown correspond to each other and will not be repeated here.

[0080] This application also provides, see Figure 4 , Figure 4 An embodiment of a device for a method for secure access control of medical data based on cloud-network-edge-end collaboration provided in this application includes: Processor 401, memory 402, input and output unit 403, bus 404; The processor 401 is connected to the memory 402, the input and output unit 403 and the bus 404; The memory 402 stores a program, and the processor 401 calls the program to execute any of the above methods for medical data security access control based on cloud-network-edge-end collaboration.

[0081] The present application also relates to a computer-readable storage medium, on which a program is stored. When the program runs on a computer, the computer executes any of the above methods for medical data security access control based on cloud-network-edge-end collaboration.

[0082] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0083] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0084] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0085] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0086] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), disk or optical disk, etc. Various media that can store program codes.

Claims

1. A method for secure access control of medical data based on cloud-network-edge-end collaboration, characterized in that: The method comprises: Collecting medical data of users through edge devices, wherein the edge devices include personal wearable devices and medical instruments and equipment; Preprocessing and homomorphically encrypting the medical data on the edge computing node, uploading the encrypted medical data to a cloud platform, and managing the modification and access to the medical data through blockchain technology; When receiving a request from a computing participant to use target medical data of a target user, obtaining a smart contract deployed by the target user on the blockchain, wherein the smart contract contains authorization rules related to user authorization, doctor role authentication, and medical institution compliance; Checking whether the authority of the computing participant complies with the authorization rules through the smart contract; If yes, the SMPC protocol is activated and a random seed is generated through VRF or chaos mapping function; Inputting the target medical data into a calculation process, performing encryption calculation using the random seed, and obtaining an encrypted calculation result; When the computing participant requests to obtain the computing result, the computing result is decrypted and authorized to be accessed according to the verification result of the random seed.

2. The method according to claim 1, characterized in that The preprocessing and homomorphic encryption of the medical data on the edge computing node includes: Preprocessing the medical data on the edge computing node, wherein the preprocessing includes standardization processing, denoising processing, and abnormality correction processing; When the medical data is imaging data, a wavelet transform based on Daubechies wavelet or Symlet wavelet is applied to the medical image to decompose the medical image into different wavelet frequency components, and homomorphic encryption is performed on each of the wavelet frequency components; When the medical data is non-imaging data, the medical data is converted from multi-dimensional data into a vector form of fixed dimension, and the medical data in vector form is homomorphically encrypted using the CKKS homomorphic encryption algorithm. SIMD technology is used for parallel computing during the homomorphic encryption process.

3. The method according to claim 1, characterized in that Before preprocessing and homomorphically encrypting the medical data on the edge computing node, the method further includes: Based on the basic information of the user and the meta-information of the medical data, generating a data identifier of the medical data through zero-knowledge proof; The step of uploading the encrypted medical data to the cloud platform includes: The encrypted medical data and the data identifier are uploaded to the cloud platform.

4. The method according to claim 3, characterized in that The activating SMPC protocol and generating a random seed through a VRF or chaos mapping function include: Activate the SMPC protocol to determine the public key information of the computing participant and the target data identifier corresponding to the target medical data; The target medical data, the public key information and the target data identifier are taken as input, and a random seed is generated through a VRF or chaotic mapping function.

5. The method according to claim 4, characterized in that The step of taking the target medical data, the public key information and the target data identifier as input and generating a random seed through VRF includes: For each computing participant, a VRF output value is calculated based on the target medical data, the public key information and the timestamp; Performing hash processing on the public key information of all computing participants and the target data identifier to obtain a public key hash value; The public key hash value, the VRF output value and the initial random seed are combined to generate a random seed through VRF.

6. The method according to claim 4, characterized in that The step of taking the target medical data, the public key information and the target data identifier as input and generating a random seed through a chaotic mapping function comprises: converting the target medical data, the public key information and the target data identifier into an input vector; Selecting an initial state value of a chaotic mapping function in the input vector, and performing iterative calculation using a Logistic chaotic mapping function to obtain a generated sequence; The values ​​in the generated sequence are selected according to a preset rule to form a random seed.

7. The method according to any one of claims 1 to 6, characterized in that Before the medical data is preprocessed and homomorphically encrypted on the edge computing node, it includes: Determine whether the user's identity information is legal through decentralized identity verification on the edge computing node; If so, the medical data is preprocessed and homomorphically encrypted on the edge computing node.

8. A system for secure access control of medical data based on cloud-network-edge-end collaboration, characterized in that: The system comprises: A collection unit, used to collect the user's medical data through an edge device, wherein the edge device includes a personal wearable device and a medical instrument device; An encryption unit, used to pre-process and homomorphically encrypt the medical data on the edge computing node, and upload the encrypted medical data to a cloud platform, where the cloud platform is used to manage the modification and access to the medical data through blockchain technology; An acquisition unit, configured to, when receiving a request from a computing participant to use target medical data of a target user, acquire a smart contract deployed by the target user on a blockchain, wherein the smart contract contains authorization rules related to user authorization, doctor role authentication, and compliance of a medical institution; A checking unit, configured to check whether the authority of the computing participant complies with the authorization rule through the smart contract; A generating unit, configured to activate the SMPC protocol and generate a random seed through a VRF or a chaotic mapping function when the inspection of the inspection unit passes; A computing unit, used for inputting the target medical data into a computing process, performing encryption computing using the random seed, and obtaining an encrypted computing result; The authorization unit is used to decrypt and authorize access to the calculation result according to the verification result of the random seed when the calculation participant requests to obtain the calculation result.

9. A device for secure access control of medical data based on cloud-network-edge-end collaboration, characterized in that: The device comprises: Processor, memory, input-output unit, and bus; The processor is connected to the memory, the input and output unit, and the bus; The memory stores a program, and the processor calls the program to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, and when the program is executed on a computer, the method according to any one of claims 1 to 7 is performed.

Images