A secure storage and sharing method for medical data based on anomaly detection
By building an identity authentication center, permission matching center and distributed anomaly detection model in medical data storage and sharing, integrating security problems at all levels of blockchain, the multi-level security risks of blockchain in medical data storage and sharing are solved. The problem has achieved efficient and comprehensive security protection and system stability of medical data.
Patent Information
- Application Number
- CN202210900919.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-07-28
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2042-07-28
AI Technical Summary
Blockchain has multiple levels of security risks in the process of medical data storage and sharing, including unauthorized access, data leakage, access denial, etc. The existing technology fails to systematically integrate security issues at all levels, resulting in the security of medical data. Sex is difficult to guarantee.
Build a secure storage and sharing method of medical data based on abnormal detection, collect system operation data through identity authentication center, permission matching center and distributed abnormal detection model, perform abnormal behavior detection, and limit permissions on abnormal nodes to integrate blockchain For security issues at all levels, build a complete distributed exception detection architecture.
It realizes efficient and comprehensive security protection for the medical data blockchain system, promptly detects and handles abnormal operation behaviors, avoids data leakage, and ensures the stability and security of system operation.
Smart Images

Figure CN115396155B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of medical service information technology, and in particular to a method for secure storage and sharing of medical data based on anomaly detection. Background Art
[0002] With the development of the information industry, the informatization of medical services has become an important part of my country's medical reform, among which the gradual integration of medical business applications and basic network platforms has become a new direction for the development of informatization. Blockchain technology has technical characteristics such as efficient sharing, interactive trust, tamper-proof and traceable. It can play an important role in the informatization flow of high-quality diagnosis and treatment resources, improve the doctor-patient trust system, build a new cross-medical institution sharing and cooperation model, strengthen the supervision and regulation of the medical service system, and solve the current pain points in the field of medical service informatization technology.
[0003] Applying blockchain to distributed anomaly detection system is an innovative technology. Blockchain has efficient information storage, sharing and management mechanisms, and is integrated into new technological and industrial innovations. However, in recent years, blockchain ecological security accidents have occurred frequently. Its core mechanism and technical architecture are mainly in the application layer, extension layer, protocol layer and storage layer, and different layers face different security risks. In the storage layer, interactive data storage is usually carried out by combining distributed databases. In this process, the blockchain network database infrastructure may be subject to unauthorized illegal access and intrusion, data security accidents caused by physical operation failures of database servers, traditional network attack threats against the underlying Internet, and security issues such as data loss and leakage caused by misoperation and poor management. In the protocol layer, blockchain network construction and security assurance are achieved through P2P network structure and data encryption mechanism, but there are still security threats including traffic attacks and malicious nodes. In the extension layer, the development of smart contracts may have security vulnerabilities and backdoors, resulting in problems such as access denial and unauthorized access in the blockchain system. In the application layer, due to the diverse and frequent data interactions between peer nodes of blockchain business, private key theft, account theft, and operating environment vulnerabilities may occur.
[0004] To sum up, the problem with the existing blockchain security technology is that it does not systematically integrate the security issues of the above-mentioned blockchain layers. The present invention regards it as a whole as a blockchain operation security issue, and provides efficient and comprehensive medical data storage and shared security protection by collecting and detecting abnormal behavior of distributed and time-series system operation data generated during the operation process. Summary of the invention
[0005] The purpose of the present invention is to provide a method for secure storage and sharing of medical data based on anomaly detection, which divides medical services into three data objects: data owners, data collectors, and data applicants. After the collectors and applicants perform identity authentication and authority matching with the owners, the medical data content is stored or shared, the system operation data of the entire interaction process is collected, and distributed abnormal behavior detection is performed to ensure the security of system operation.
[0006] The purpose of the present invention can be achieved through the following technical solutions:
[0007] A method for secure storage and sharing of medical data based on anomaly detection, the method comprising the following steps:
[0008] Step 1: Build a medical service blockchain network, identity authentication center, authority matching center and distributed anomaly detection model;
[0009] Step 2: The medical data collector and the applicant send a request to the identity authentication center;
[0010] Step 3: Authenticate the requester and match the authenticated requester with the requested object.
[0011] Step 4: The requester interacts with the medical service;
[0012] Step 5: The system runs data collection and prediction, returns anomaly detection results, and restricts permissions for abnormal nodes.
[0013] As a further solution of the present invention: the step 1 comprises:
[0014] (1) The distributed anomaly detection model selects data monitoring nodes in the medical data blockchain network to collect system operation data and obtains the original system data X = (x 1 ,x 2 ,...,x n );
[0015] (2) Calculate the correlation coefficient matrix R and the corresponding eigenvalue λ of the original system data X i With the eigenvector μ i ;
[0016] (3) Sort the eigenvalues from small to large λ min ≤...≤λ max ;
[0017] (4) Calculate the contribution rate of the first j eigenvalues Eliminate alpha j ≤20% of the eigenvalues corresponding to the data, the processed system data Y = (y 1 ,y 2,...,y n );
[0018] (5) Input data Y into LSTM neural network training and optimization to obtain anomaly detection model M LSTM .
[0019] As a further solution of the present invention: the step three comprises:
[0020] (1) The identity authentication center generates the data requester's digital certificate DC based on the national secret SM2 digital signature algorithm r Determine identity;
[0021] (2) The authority matching center encodes the ID of the data requester r 、Data owner code ID o and medical service type T s Determine the authority matching relationship between medical service objects and medical data M p :
[0022]
[0023] (4) If the permission matching relationship M p The result is N p Then reject the data exchange request; if the result is Y p Then according to the data requester's encoding ID r , Data owner code ID o and medical service type T s Calculate the authority matching data content C of medical service objects and medical data o :
[0024]
[0025] Among them C pt Indicates that if the requested data is full-authority data, it will be directly obtained; C dm Indicates that if the requested data is not fully authorized data, it will be obtained after being desensitized.
[0026] As a further solution of the present invention: the step five comprises:
[0027] (1) The system’s operating data at m time points X now =(x n1 ,x n2 ,...,x nm ) Input the anomaly detection model M in step 1 LSTM , get the system's future operation forecast data Y at x time points predicted =(y p1 ,y p2 ,...,ypx );
[0028] (2) Using the actual value Y of the operating data real =(y r1 ,y r2 ,...,y rx ) Determine the system operation data statistics value Y statistics =(y s1 ,y s2 ,...,y sx ) and abnormal detection return value V return :
[0029]
[0030]
[0031] where y sn ,y rn and pn Respectively represent the statistical value, actual value and predicted value at the nth time point;
[0032] (3) For the return value of R abnormal Update the medical data blockchain nodes and limit their permissions.
[0033] Beneficial effects of the present invention:
[0034] (1) The present invention uses a medical data security storage and sharing method based on anomaly detection in a distributed anomaly detection system to timely detect and process abnormal operating behaviors in a medical data blockchain system, thereby ensuring the accuracy and stability of system operation and avoiding medical data leakage caused by blockchain operation loopholes or physical failures;
[0035] (2) The present invention systematically integrates the security issues in the design of each level of the blockchain, regards it as a whole as the security issue of the blockchain operation state, builds a complete distributed anomaly detection architecture, collects and detects abnormal behavior of the distributed and time series operation data generated during the operation of the system, submits the detection results to the authority matching center for security response, and provides efficient and comprehensive medical data storage and sharing security protection;
[0036] (3) The present invention divides medical services into three data objects: data owners, data collectors, and data applicants. After the collectors and applicants perform identity authentication and authority matching with the owners, they store or share the medical data content; the system operation environment monitoring tool is used to collect the system operation data in the entire interaction process in a distributed manner, and the anomaly detection model is submitted for security detection, to achieve abnormal behavior detection for security risks of the blockchain's underlying Internet and blockchain host, and to ensure the stable and secure operation of the medical data blockchain system. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] The present invention will be further described below in conjunction with the accompanying drawings.
[0038] Figure 1 is a flow chart of the present invention;
[0039] Figure 2 It is a schematic diagram of the application of the medical data security storage method based on anomaly detection of the present invention;
[0040] Figure 3 It is a schematic diagram of the application of the medical data security sharing method based on anomaly detection of the present invention. DETAILED DESCRIPTION
[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0042] The present invention divides medical services into three data objects: data owners, data collectors, and data applicants. After the collectors and applicants perform identity authentication and authority matching with the owners, they store or share the medical data content. The system operation environment monitoring tool is used to collect the system operation data in the entire interaction process in a distributed manner, and the anomaly detection model is submitted for security detection, so as to realize the abnormal behavior detection of the security risks of the blockchain's underlying Internet and blockchain host, and ensure the stable and secure operation of the medical data blockchain system.
[0043] See also Figure 1 As shown, the method for secure storage and sharing of medical data based on abnormality detection of the present invention comprises the following steps:
[0044] S101: Build a medical service blockchain network, identity authentication center, authority matching center and distributed anomaly detection model;
[0045] S102: The medical data collector and the applicant send a request to the identity authentication center;
[0046] S103: Authenticate the identity of the requester and perform authority matching between the authenticated requester and the request object;
[0047] S104: The requester performs medical service interaction;
[0048] S105: The system runs data collection and prediction, returns anomaly detection results, and restricts permissions for abnormal nodes.
[0049] See also Figure 2 As shown, the application scenario of the present invention is the secure storage process of medical data. The network includes two data objects, namely, medical data collectors and medical data owners, an identity authentication center, an authority matching center, a medical database, and a distributed anomaly detection module. The medical data collector needs to submit a collection application on the blockchain and obtain approval from the medical data owner before data collection can be performed. Before uploading the medical data, it needs to be approved by the identity authentication center and the authority matching center, and the data must be encrypted and uploaded to the medical database. The entire system operation process collects operation data through the system monitoring tool and inputs it into the anomaly detection module for processing to ensure the security of the medical data storage process.
[0050] See also Figure 3 As shown, the application scenario of the present invention is the secure sharing process of medical data. The network includes two data objects, namely, medical data applicants and medical data owners, an identity authentication center, an authority matching center, a medical database, and a distributed anomaly detection module;
[0051] Medical data applicants can submit permission applications and usage applications on the blockchain. Permission applications can enable applicants to have more data scenario applications under the authorization of the data owner. Usage applications need to be approved by the identity authentication center and the permission matching center. The medical database decrypts and desensitizes the application data and submits it to the data applicant for use. The entire system operation process collects operation data through system monitoring tools and inputs it into the anomaly detection module for processing to ensure the security of the medical data storage process.
[0052] Specifically, the distributed anomaly detection system includes applicants, a medical data blockchain, and a medical database, and the applicants include medical data collectors, medical data applicants, and medical data owners;
[0053] Medical data collectors need to submit a collection application on the medical data blockchain and obtain approval from the medical data owner before they can collect data;
[0054] Before uploading medical data, medical data collectors must pass authentication through the identity authentication center and the authority matching center, encrypt the data, and then upload it to the medical database;
[0055] Medical data applicants submit permission applications and usage applications on the medical data blockchain;
[0056] Medical data applicants are authenticated by the identity authentication center and the authority matching center. The medical database decrypts and desensitizes the application data and then submits the data for use by the medical data applicants.
[0057] The medical data owner authorizes and approves the applications of medical data collectors and medical data applicants.
[0058] In a specific implementation, the method for secure storage and sharing of medical data based on anomaly detection specifically includes the following steps:
[0059] Step 1: Build a medical service blockchain network, identity authentication center, authority matching center and distributed anomaly detection model:
[0060] 1.1) Build a medical service blockchain network based on the Hyperledger Fabric platform, an identity authentication center based on the national secret SM2 digital signature algorithm, and an authority matching center;
[0061] 1.2) Distributed anomaly detection model selects data monitoring nodes in the medical data blockchain network to collect system operation data and obtains the original system data X = (x 1 ,x 2 ,...,x n );
[0062] 1.3) Calculate the correlation coefficient matrix R and the corresponding eigenvalue λ of the original system data X i With the eigenvector μ i ;
[0063] 1.4) Sort the eigenvalues from small to large λ min ≤...≤λ max ;
[0064] 1.5) Calculate the contribution rate of the first j eigenvalues Eliminate alpha j ≤20% of the eigenvalues corresponding to the data, the processed system data Y = (y 1 ,y 2 ,...,y n );
[0065] 1.6) Input data Y into LSTM neural network training and optimization to obtain anomaly detection model M LSTM .
[0066] Step 2: The medical data collector and the applicant send a request to the identity authentication center CA, R upload →CA;
[0067] Step 3: The requester performs identity authentication, and matches the authenticated requester with the request object for permissions:
[0068] 3.1) The identity authentication center generates the data requester's digital certificate DC based on the national secret SM2 digital signature algorithm r Determine identity;
[0069] 3.2) The authority matching center encodes the ID of the data requester r, Data owner code ID o and medical service type T s Determine the authority matching relationship between medical service objects and medical data M p :
[0070]
[0071] 3.3) If the permission matching relation M p The result is N p Then reject the data exchange request; if the result is Y p Then according to the data requester's encoding ID r , Data owner code ID o and medical service type T s Calculate the authority matching data content C of medical service objects and medical data o :
[0072]
[0073] Among them C pt Indicates that if the requested data is full-authority data, it will be directly obtained; C dm Indicates that if the requested data is not fully authorized data, it will be obtained after being desensitized.
[0074] Step 4: Requester interacts with medical services: upload medical data upload →DB; Medical data download DB→Data download .
[0075] Step 5: The system runs data collection and prediction, returns anomaly detection results, and restricts permissions for abnormal nodes:
[0076] 5.1) The system's operating data at m time points X now =(x n1 ,x n2 ,...,x nm ) Input the anomaly detection model M in step 1 LSTM , get the system's future operation forecast data Y at x time points predicted =(y p1 ,y p2 ,...,y px );
[0077] 5.2) Using the actual value Y of the operating data real =(y r1 ,y r2 ,...,y rx ) Determine the system operation data statistics value Y statistics =(y s1 ,ys2 ,...,y sx ) and abnormal detection return value V return :
[0078]
[0079]
[0080] where y sn ,y rn and pn Respectively represent the statistical value, actual value and predicted value at the nth time point;
[0081] 5.3) For the return value of R abnormal Update the medical data blockchain nodes and limit their permissions.
[0082] The core point of the present invention is: by using the medical data security storage and sharing method based on anomaly detection in the distributed anomaly detection system, the abnormal operation behavior in the medical data blockchain system is detected and processed in time, the accuracy and stability of the system operation are guaranteed, and the medical data leakage caused by blockchain operation loopholes or physical failures is avoided. The present invention systematically integrates the security issues in the blockchain design at all levels, regards it as a blockchain operation security issue, and constructs a complete distributed anomaly detection architecture. By collecting and detecting abnormal behavior of the distributed and time series operation data generated during the system operation, the detection results are submitted to the authority matching center for security response, providing efficient and comprehensive medical data storage and sharing security protection;
[0083] The core point of the present invention is: medical services are divided into three data objects: data owners, data collectors, and data applicants. After the collectors and applicants perform identity authentication and authority matching with the owners, the medical data content is stored or shared; the system operation environment monitoring tool is used to collect the system operation data in the entire interaction process in a distributed manner, and an anomaly detection model is submitted for security detection, to achieve abnormal behavior detection for security risks of the blockchain's underlying Internet and blockchain host, and to ensure the stable and secure operation of the medical data blockchain system.
[0084] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.
Claims
1. A method for secure storage and sharing of medical data based on anomaly detection, characterized in that: The secure storage and sharing method comprises the following steps: Step 1: Build a medical service blockchain network, identity authentication center, authority matching center and distributed anomaly detection model; Step 2: The medical data collector and the applicant send a request to the identity authentication center; Step 3: Authenticate the requester and match the authenticated requester with the requested object. Step 4: The requester interacts with the medical service; Step 5: The system runs data collection and prediction, returns anomaly detection results, and restricts permissions for abnormal nodes; The step five comprises: (1) The system m Operation data at each time point Input the anomaly detection model in step 1 , get the system future x Operation forecast data at each time point ; (2) Using actual operating data Determine system operation data statistics and anomaly detection return value : ; ; in , , Respectively represent the statistical value, actual value and predicted value at the nth time point; (3) For the return value Update the medical data blockchain nodes and limit their permissions.
2. The method for secure storage and sharing of medical data based on anomaly detection according to claim 1 is characterized in that: The step one comprises: (1) The distributed anomaly detection model selects data monitoring nodes in the medical data blockchain network to collect system operation data and obtain the original system data ; (2) Calculate the correlation coefficient matrix of the original system data X R and the corresponding eigenvalues With the eigenvector ; (3) Sort the eigenvalues from small to large ; (4) Before calculation j Contribution rate of item eigenvalue , remove The data corresponding to the eigenvalues of some parts are processed to obtain the system data ; (5) Input data Y into LSTM neural network training and optimization to obtain anomaly detection model .
3. The method for secure storage and sharing of medical data based on anomaly detection according to claim 1 is characterized in that: The step three comprises: (1) The identity authentication center generates the data requester's digital certificate based on the national secret SM2 digital signature algorithm Determine identity; (2) The authority matching center codes the data requester , Data owner code and type of medical services Determine the authority matching relationship between medical service objects and medical data ; ; (3) If the permissions match the relationship The result is The data interaction request is rejected; if the result is Then according to the data requester code , Data owner code and type of medical services Calculate the authority matching data content of medical service objects and medical data ; ; in Indicates that if the requested data is full-authority data, it will be directly obtained; Indicates that if the requested data is not fully authorized data, it will be obtained after being desensitized.
