Medical information sharing method based on medical big data

Abstract

The invention relates to the field of medical care informatics, in particular to a medical information sharing method based on medical big data. The method comprises the following steps: acquiring medical data, and performing priority scoring on the medical data to obtain the priority of the medical data; based on the priority of the medical data, performing privacy protection processing on the medical data to obtain data after privacy protection processing; a dynamic authority control mechanism is introduced, the access authority of the medical data is distributed, a multi-party interaction verification mechanism is introduced, the access of the medical data is verified, and meanwhile, a self-adaptive data flow fault tolerance and prediction mechanism is introduced to recover and compensate the medical data. The problem that an existing medical data sharing and using method cannot dynamically adjust the access permission of medical data, and the data sharing efficiency is affected is solved. The security of information sharing is influenced due to lack of dynamic disturbance and noise injection mechanisms; and an effective fault-tolerant and data recovery mechanism is lacked, so that the data coherence and the sharing effectiveness are influenced.

Description

Technical Field

[0001] The present invention relates to the field of medical care informatics, and in particular to a medical information sharing method based on medical big data. Background Art

[0002] With the rapid development of medical information technology and big data technology, the generation and application of medical data have become more and more extensive and complex. Medical data comes from various sources, including electronic medical records (EMR), diagnostic records, images, laboratory reports, etc. from hospitals, clinics, pharmacies and other medical institutions, as well as health information submitted by patients, such as physiological data recorded by mobile devices. These medical data provide important references for personalized treatment, clinical decision support, public health monitoring and medical research, and play a vital role in improving the quality of medical services, optimizing diagnosis and treatment processes, and promoting the progress of medical research.

[0003] However, there are many challenges in sharing and using medical data: First, since medical data is highly sensitive and private, how to ensure that patient privacy is not leaked while sharing becomes a core issue; as the demand for data access by various medical institutions and related parties increases, the importance of privacy protection is also increasing, and it is necessary to design a mechanism that can dynamically adjust the intensity of privacy protection to meet the needs of different scenarios and data types; secondly, the diversity and complexity of medical data leads to large differences in data formats between different systems and institutions. Although some international standards (such as HL7, FHIR, etc.) have standardized the format of medical data, in actual operation, how to ensure data compatibility across institutions and platforms is still an important technical problem.

[0004] In summary, the existing methods of sharing and using medical data have the following technical problems: the inability to dynamically adjust the access rights to medical data according to the security level and priority of the data, which can easily lead to the leakage or over-protection of sensitive information, thereby affecting the efficiency of data sharing; the lack of dynamic perturbation and noise injection mechanisms can easily lead to information interception and tampering, and the inability to track and verify each data transmission throughout the process, which can easily lead to unauthorized access or data tampering, affecting the security of information sharing; the lack of effective fault-tolerant and data recovery mechanisms, and the inability to adaptively compensate for lost or incomplete data, resulting in interrupted or incomplete data transmission, affecting the consistency of data and the effectiveness of sharing. Summary of the invention

[0005] The present invention provides a medical information sharing method based on medical big data to solve the problems that the sharing and use methods of medical data cannot dynamically adjust the access rights of medical data according to the security level and priority of the data, which easily leads to the leakage or over-protection of sensitive information, thereby affecting the data sharing efficiency; lack of dynamic disturbance and noise injection mechanism, which easily leads to information interception and tampering, and cannot track and verify each data transmission throughout the process, which easily leads to unauthorized access or data tampering, affecting the security of information sharing; lack of effective fault tolerance and data recovery mechanism, and cannot adaptively compensate for lost or incomplete data, resulting in interrupted or incomplete data transmission, affecting the consistency of data and the effectiveness of sharing.

[0006] A medical information sharing method based on medical big data of the present invention specifically includes the following technical solutions:

[0007] A medical information sharing method based on medical big data comprises the following steps:

[0008] S1. Collect medical data and perform priority scoring on the medical data to obtain the priority of the medical data; perform privacy protection processing on the medical data based on the priority of the medical data to obtain privacy-protected data;

[0009] S2. Based on the data processed with privacy protection, a dynamic permission control mechanism is introduced to allocate access rights to medical data. Based on the access rights to medical data, access to medical data is verified, and an adaptive data stream fault tolerance and prediction mechanism is introduced to recover and compensate for medical data.

[0010] Preferably, the S1 specifically includes:

[0011] Medical data is processed in layers, and a priority scoring formula is introduced to perform priority scoring on the medical data to obtain the priority of the medical data; the specific formula is as follows:

[0012]

[0013] Among them, P i (t) represents the i-th type of medical data d i Priority at time t; R ij (t) is the medical data of category i in the jth scenario i The importance of dynamic weight; m is the number of scenes; S ij is the i-th type of medical data d in the j-th scenario i Privacy sensitivity of T ij (t) is the i-th type of medical data d i Time sensitivity in the jth scene; parameter α 1j , α2j , α 3j is the coefficient used to adjust the weight of each variable.

[0014] Preferably, the S1 specifically includes:

[0015] Based on the priority of medical data, a multi-level privacy protection algorithm is introduced to obtain privacy-protected data.

[0016] Preferably, the S1 specifically includes:

[0017] The multi-level privacy protection algorithm performs privacy protection on medical data by introducing dynamic perturbation and noise injection mechanisms. The specific formula is as follows:

[0018]

[0019] in, is the i-th type of medical data d i Data obtained after privacy protection processing; Δf k is the global sensitivity of the k-th dimension medical data; ∈ k It’s the privacy budget; is the medical data dimension; G k (τ) is the Gaussian noise function, τ is a random variable; δ il is the i-th type of medical data d i The time perturbation coefficient of the lth time perturbation dimension in T l is the time scale of the lth time perturbation cycle; M is the time perturbation dimension; η i is a parameter to adjust the noise intensity; i is the time decay coefficient; t is the time.

[0020] Preferably, the S2 specifically includes:

[0021] The dynamic permission control mechanism combines the user's access request permission and the static security level of the data after privacy protection processing to obtain the access permission of the medical data. The specific formula is as follows:

[0022]

[0023] Among them, A iu (t) is the data of user u after privacy protection processing of the i-th category at time t access rights; θ is the adjustment parameter for controlling access rights; Qi(t) is the data after privacy protection processing for the i-th category The frequency of access requests; C u (t) is the access permission level of user u at time t; L i is the data after the privacy protection processing of the i-th category Static safety level; V u is the authentication level of user u; is the activation function, based on the data after the privacy protection of the i-th category and the i-th category medical data d i The priority P at time t i (t), calculates whether user u can access the privacy-preserving data of the i-th category at time t; μ is the coefficient of the time-related perturbation; Ω u is the time dynamic adjustment coefficient; ρ is the adjustment parameter of the activation function.

[0024] Preferably, the S2 specifically includes:

[0025] Based on the access rights of medical data, a multi-party interactive verification mechanism is introduced to verify the access to medical data and obtain the access verification result of the medical data.

[0026] Preferably, the S2 specifically includes:

[0027] When the access verification result of medical data V ij When (t) = 1, the data after privacy protection at the current time point is used as the medical data at the future time; when the access verification result V ij When (t)<1, an adaptive data stream fault tolerance and prediction mechanism is introduced to predict medical data in the future and repair and compensate the medical data.

[0028] Preferably, the S2 specifically includes:

[0029] The specific formula of the adaptive data flow fault tolerance and prediction mechanism is as follows:

[0030]

[0031] in, is the data after privacy protection processing of the i-th category at time t+1 The predicted value represents the medical data in the future; φ is the data of the i-th category after privacy protection processing at time t. The weight coefficient of is the data after privacy protection processing of the i-th category at time t-1; ψ is the weight coefficient of the data change rate; △t is the time difference; is the accumulated value of data from time 0 to time t; v is the weight coefficient of the noise compensation term; G(Δf i ) is based on the data after the privacy protection processing of the i-th category The global sensitivity Δf i Noise compensation function.

[0032] The beneficial effects of the technical solution of the present invention are:

[0033] 1. Build a hierarchical structure for medical data and assign priorities. Medical data in different scenarios can automatically adjust the sharing order according to their importance and urgency, ensuring that high-priority critical medical data can be transmitted and shared first in emergency situations. This not only improves the efficiency of data sharing, but also ensures that medical workers can quickly obtain the required key information, shortening the time for medical decision-making;

[0034] 2. Based on the global sensitivity and priority of medical data, a multi-level privacy protection algorithm is designed, and a dynamic perturbation and noise injection mechanism is introduced to provide a higher level of protection for sensitive medical data. The privacy protection mechanism is dynamically adjusted according to the priority of medical data to ensure that patient privacy is protected to the maximum extent without affecting medical efficiency, and to prevent medical data from being tampered with or unauthorized access during transmission. The dynamically adjusted privacy protection mechanism can adapt to the needs of different types of medical data and provide more stringent protection for sensitive data.

[0035] 3. The dynamic permission control mechanism of the present invention automatically allocates access rights to the privacy-protected data according to the user's access request permission level, the static security level of the privacy-protected data, and the frequency of access requests; the introduction of activation functions and time fluctuation terms enables permission control to not only adapt to different medical scenarios (for example, allowing doctors to quickly access patient data in an emergency), but also dynamically adjust over time to prevent the rigidity and abuse of access rights; the dynamic permission control mechanism greatly enhances the security of information sharing while ensuring the legal access to medical data;

[0036] 4. Ensure the integrity, legitimacy and transparency of medical data during transmission by real-time tracking and verification of each data transmission; the multi-party interactive verification mechanism prevents medical data from being tampered with or accessed without authorization during transmission by verifying the user's digital signature and the data hash value after privacy protection processing; each data sharing operation can be accurately tracked, and the security and legitimacy of medical data can be guaranteed, thereby ensuring the traceability of the entire data sharing process;

[0037] 5. The present invention introduces an adaptive data stream fault tolerance and prediction mechanism to adaptively compensate for anomalies (such as loss, tampering, delay, etc.) in the data transmission process, and predict future medical data based on the data processed with privacy protection, to ensure that medical data can still be fully restored in the event of network fluctuations or system errors. Even in the event of data transmission failure, compensation and prediction can still be performed through historical data processed with privacy protection and data processed with privacy protection at the current moment, to ensure the continuity and integrity of shared data. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a flow chart of a medical information sharing method based on medical big data described in the present invention. DETAILED DESCRIPTION

[0039] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all 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.

[0040] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0041] The following is a detailed description of a specific scheme of a medical information sharing method based on medical big data provided by the present invention in conjunction with the accompanying drawings.

[0042] Refer to the attached Figure 1 , which shows a flow chart of a medical information sharing method based on medical big data provided by an embodiment of the present invention, the method comprising the following steps:

[0043] S1. Collect medical data and perform priority scoring on the medical data to obtain the priority of the medical data; perform privacy protection processing on the medical data based on the priority of the medical data to obtain privacy-protected data;

[0044] Medical data comes from various sources, not only from different medical institutions (such as hospitals, clinics, pharmacies, etc.), but also from patients' voluntary submission; the collected medical data includes electronic medical record data, diagnostic record data, drug prescription data, imaging data, laboratory report data, and health data submitted by patients themselves (such as physiological data recorded by mobile devices).

[0045] All medical data needs to be collected through standardized interfaces and follow internationally recognized medical data format standards, such as HL7, FHIR, etc. This not only ensures the compatibility of data formats between different data sources, but also lays the foundation for subsequent multi-layer data processing and dynamic permission control.

[0046] The specific collection process is as follows: set different types of medical data D = {d1, d2, ..., d n}, n represents the type of medical data. The medical data collected by each data source includes basic information (such as patient identity, date of birth, etc.), clinical information (such as diagnosis, medication history) and highly sensitive information (such as genomic data, psychological assessment).

[0047] Each type of medical data is processed in layers according to actual clinical scenarios and needs, and the data priority is adjusted dynamically. In order to accurately describe the data priority at each level, a priority scoring formula is introduced to score the priority of medical data and obtain the priority of medical data. The specific formula is as follows:

[0048]

[0049] Among them, P i (t) represents the i-th type of medical data d i Priority at time t; R ij (t) is the medical data of category i in the jth scenario i The importance dynamic weight of m is the number of scenarios, including emergency medical scenarios, routine outpatient scenarios, etc.; S ij is the i-th type of medical data d in the j-th scenario i The privacy sensitivity of T affects whether it is transmitted first; ij (t) is the i-th type of medical data d i The time sensitivity in the jth scenario reflects the urgency of decision-making for this type of medical data over time; parameter α 1j , α 2j , α 3j It is a coefficient used to adjust the weight of each variable and determine the impact of each factor in the overall score.

[0050] Medical data at different levels have different sensitivities. In order to protect sensitive medical information, a privacy protection mechanism is introduced to perform privacy protection according to the priority of medical data. A multi-level privacy protection algorithm is designed in the privacy protection mechanism to protect patient privacy by introducing dynamic perturbations and noise injection mechanisms. The specific formula of the multi-level privacy protection algorithm is as follows:

[0051]

[0052] in, is the i-th type of medical data d i The data obtained after privacy protection processing, that is, the data after privacy protection processing; Δf k is the global sensitivity of the k-th dimension medical data, which is used to measure the impact of this type of medical data on the risk of privacy leakage; ∈ k is the privacy budget of the k-th dimension medical data, which is used to control the intensity of noise injection; is the medical data dimension; G k (τ) is the Gaussian noise function used to analyze the i-th medical data d i Introducing noise, τ is a random variable; δ il is the i-th type of medical data d i The time perturbation coefficient of the lth time perturbation dimension in T l is the time scale of the lth time perturbation cycle; M is the time perturbation dimension; η i is a parameter to adjust the noise intensity; i is the time decay coefficient, which is used to control the decay of data privacy protection over time; t is time; It is a time decay term, which means that the privacy protection of medical data will gradually decay over time, thus balancing sharing efficiency and privacy protection. Multi-level protection is performed on sensitive information during data transmission and sharing to ensure that even if the medical data is intercepted during transmission, the patient's real information cannot be restored.

[0053] S2. Based on the data processed with privacy protection, a dynamic permission control mechanism is introduced to allocate access rights to medical data. Based on the access rights to medical data, access to medical data is verified, and an adaptive data stream fault tolerance and prediction mechanism is introduced to recover and compensate for medical data.

[0054] In order to ensure the safe sharing of medical data among different medical institutions, a dynamic permission control mechanism is designed. The access rights to medical data are determined by the user's access request rights and the static security level of the data after privacy protection processing. The formula is as follows:

[0055]

[0056] Among them, A iu (t) is the data of user u after privacy protection processing of the i-th category at time t access rights; θ is the adjustment parameter for controlling access rights; Q i (t) is the data after privacy protection processing for category i The frequency of access requests; C u (t) is the access permission level of user u at time t; L i is the data after the privacy protection processing of the i-th category Static safety level; V u is the authentication level of user u; is the activation function, based on the data after the privacy protection of the i-th category and the i-th category medical data d i The priority P at time t i(t), calculates whether user u can access the privacy-preserving data of category i at time t; μ is the coefficient of time-related perturbation, which affects the amplitude of dynamic changes in access rights; Ω u is the time dynamic adjustment coefficient; ρ is the adjustment parameter of the activation function, which is used to adjust the response speed or sensitivity of the activation function. In order to prevent the access control from being too rigid, the time fluctuation term is introduced Ensure that access rights are adjusted over time. Through the dynamic permission control mechanism, ensure that medical data can be accessed in compliance with the user's access request permissions and the static security level of the data after privacy protection processing, which not only ensures information sharing, but also strengthens the security of medical data.

[0057] When the data access request passes the permission control, the access process of medical data is tracked and verified throughout to ensure the integrity and transparency of medical data transmission. A multi-party interactive verification mechanism is designed. The formula is as follows:

[0058]

[0059] Among them, V ij (t) represents the data after user u processes the privacy protection of the i-th category at time t The access verification result is the access verification result of the medical data; ζ is the adjustment coefficient in the verification calculation; is the data after the privacy protection processing of the i-th category The hash value is used to verify the integrity of the data after privacy protection processing; S u is the digital signature of user u, ensuring the legitimacy of the user's identity; κ is the adjustment coefficient of the log record in the access verification result; R i is the data after the privacy protection processing of the i-th category The importance weight of T iu is a time-related parameter, representing the data after the i-th privacy protection processing The sharing frequency among user u; Indicates XOR operation. Verify the consistency between the user's access request permissions and the hash value of the data after privacy protection processing to prevent data tampering and unauthorized access. Ensure that every access operation can be accurately tracked and verified throughout the data sharing process to ensure the security, transparency and integrity of medical data.

[0060] Medical data that has been controlled and verified may be lost or delayed during transmission due to network conditions or system errors. For this reason, an adaptive data stream fault tolerance and prediction mechanism is introduced to ensure rapid recovery and compensation when data is lost. The adaptive data stream fault tolerance and prediction mechanism predicts medical data in the future based on the data processed after privacy protection. The formula is as follows:

[0061]

[0062] in, is the data after privacy protection processing of the i-th category at time t+1 The predicted value is the medical data in the future; φ is the data of the i-th category after privacy protection at time t. (i.e. the weight coefficient of current data); is the data after the privacy protection processing of the i-th category at time t-1 (i.e., the data at the previous moment); ψ is the weight coefficient of the data change rate; △t is the time difference; is the accumulated value of data from time 0 to time t; ν is the weight coefficient of the noise compensation term; G(Δf i ) is based on the data after the privacy protection processing of the i-th category The global sensitivity Δf i Noise compensation function; Δf i is the data after the privacy protection processing of the i-th category The global sensitivity of reflects the potential risk of privacy leakage of data after privacy protection processing. It ensures that in the event of data loss or transmission error, data recovery can be performed adaptively to ensure the integrity and consistency of medical data.

[0063] If the access verification result of medical data is V ij (t) = 1, indicating that the data transmission is complete and no tampering or loss has occurred. At this time, the adaptive data flow fault tolerance and prediction mechanism will not be activated, but the data processed by the privacy protection of the i-th category will be directly used. Data at the current time point t As medical data at future time t+1;

[0064] If the access verification result of medical data is V ij If (t) < 1, it means that there are problems in the data transmission process (such as data loss, tampering, incomplete transmission, etc.). At this time, data compensation will be performed based on the data at the previous moment after the privacy protection processing and the historical data (that is, the privacy protection processing data before the current moment) to predict the medical data at the future time t+1.

[0065] The final implementation process of medical information sharing is as follows:

[0066] First, medical data is processed in layers according to actual clinical scenarios and needs, and priorities are dynamically assigned;

[0067] Subsequently, the medical data is processed into privacy-protected data under the action of the privacy protection mechanism to ensure that privacy is not leaked during the sharing process;

[0068] Then, the medical data that can be shared is dynamically controlled according to the user's access request permissions and the static security level of the data after privacy protection processing; each shared medical data transmission is verified to ensure the legitimacy, integrity and transparency of the medical data sharing behavior; if errors or data loss occur during the transmission process, the medical data is repaired and compensated through the adaptive data stream fault tolerance and prediction mechanism to ensure the accuracy and integrity of the shared data. In summary, a medical information sharing method based on medical big data is completed.

[0069] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0070] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0071] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A medical information sharing method based on medical big data, characterized in that: The following steps are involved: S1. Collect medical data and perform priority scoring on the medical data to obtain the priority of the medical data; perform privacy protection processing on the medical data based on the priority of the medical data to obtain privacy-protected data; S2. Based on the data processed with privacy protection, a dynamic permission control mechanism is introduced to allocate access rights to medical data. Based on the access rights to medical data, access to medical data is verified, and an adaptive data stream fault tolerance and prediction mechanism is introduced to recover and compensate for medical data.

2. A medical information sharing method based on medical big data according to claim 1, characterized in that: The S1 specifically includes: The medical data is processed in layers, and a priority scoring formula is introduced to score the priority of the medical data to obtain the priority of the medical data; the specific formula is as follows: Among them, P i (t) represents the i-th type of medical data d i Priority at time t; R ij (t) is the medical data of the i-th category in the j-th scenario i The importance of dynamic weight; m is the number of scenes; S ij is the i-th type of medical data d in the j-th scenario i Privacy sensitivity of T ij (t) is the i-th type of medical data d i Time sensitivity in the jth scene; parameter α 1j , α 2j , α 3j is the coefficient used to adjust the weight of each variable.

3. A medical information sharing method based on medical big data according to claim 2, characterized in that: The S1 specifically includes: Based on the priority of medical data, a multi-level privacy protection algorithm is introduced to obtain privacy-protected data.

4. A medical information sharing method based on medical big data according to claim 3, characterized in that: The S1 specifically includes: The multi-level privacy protection algorithm performs privacy protection on medical data by introducing dynamic perturbation and noise injection mechanisms. The specific formula is as follows: in, is the i-th type of medical data d i Data obtained after privacy protection processing; Δf k is the global sensitivity of the k-th dimension medical data; ∈ k It’s the privacy budget; is the medical data dimension; G k (τ) is the Gaussian noise function, τ is a random variable; δ il is the i-th type of medical data d i The time perturbation coefficient of the lth time perturbation dimension in T l is the time scale of the lth time perturbation cycle; M is the time perturbation dimension; η i is a parameter to adjust the noise intensity; i is the time decay coefficient; t is the time.

5. The medical information sharing method based on medical big data according to claim 1 is characterized in that: The S2 specifically includes: The dynamic permission control mechanism combines the user's access request permission and the static security level of the data after privacy protection processing to obtain the access permission of the medical data. The specific formula is as follows: Among them, A iu (t) is the data of user u after privacy protection processing of the i-th category at time t access rights; θ is the adjustment parameter for controlling access rights; Q i (t) is the data after privacy protection processing for category i The frequency of access requests; C u (t) is the access permission level of user u at time t; L i is the data after the privacy protection processing of the i-th category Static safety level; V u is the authentication level of user u; is the activation function, based on the data after the privacy protection of the i-th category and the i-th category medical data d i The priority P at time t i (t), calculates whether user u can access the privacy-preserving data of the i-th category at time t; μ is the coefficient of the time-related perturbation; Ω u is the time dynamic adjustment coefficient; ρ is the adjustment parameter of the activation function.

6. A medical information sharing method based on medical big data according to claim 5, characterized in that: The S2 specifically includes: Based on the access rights of medical data, a multi-party interactive verification mechanism is introduced to verify the access to medical data and obtain the access verification result of the medical data.

7. A medical information sharing method based on medical big data according to claim 6, characterized in that: The S2 specifically includes: When the access verification result of medical data V ij When (t) = 1, the data after privacy protection at the current time point is used as the medical data at the future time; when the access verification result V ij When (t) < 1, an adaptive data stream fault tolerance and prediction mechanism is introduced to predict medical data in the future and repair and compensate the medical data.

8. A medical information sharing method based on medical big data according to claim 7, characterized in that: The S2 specifically includes: The specific formula of the adaptive data flow fault tolerance and prediction mechanism is as follows: in, is the data after privacy protection processing of the i-th category at time t+1 The predicted value represents the medical data in the future; φ is the data of the i-th category after privacy protection processing at time t. The weight coefficient of is the data after privacy protection processing of the i-th category at time t-1; ψ is the weight coefficient of the data change rate; Δt is the time difference; is the accumulated value of data from time 0 to time t; v is the weight coefficient of the noise compensation term; G(Δf i ) is based on the data after the privacy protection processing of the i-th category The global sensitivity Δf i Noise compensation function.

Images