Physical examination user privacy protection method, device and system

Abstract

The application discloses a physical examination user privacy protection method, device and system, a corresponding model is established according to participant data, a prediction result is obtained according to a requirement, a complete quantitative evaluation system is established for uncertainty of the model prediction result, a trained model is verified and parameter-adjusted under new data based on a trusted third party, and the trusted third party returns the adjusted model and parameters to each party based on data attributes of each party after training through processed data attributes, so that the same result is output by the calculation model under the condition that private data of each participant (including a service party and each party of an internet business platform) does not go out of a domain or under the condition that the private data of each participant is not leaked.

Description

[0001] This application is a divisional application of Chinese invention patent application filed on August 5, 2021, with application number CN202110893693.1 and invention title "Method, Apparatus and System for Protecting User Privacy during Physical Examination". Technical Field

[0002] This invention relates to the field of computer information security technology, and in particular to a method, apparatus and system for protecting the privacy of users undergoing physical examinations. Background Technology

[0003] With rapid economic development and the widespread adoption of the internet and mobile internet, it has become a trend to collect data based on user behavior patterns and build models to predict user behavior preferences. However, the security of personal user data has increasingly become a focus of attention for all sectors of society. As a result, countries around the world are introducing data security regulations to protect users' personal privacy.

[0004] Secure multi-party computation addresses the problem of protecting privacy in collaborative computation among a group of distrustful participants. "Multi-party" often refers to strongly related parties in an application scenario. Its main characteristic is that the computation is completed without the original privacy data being accessible to all participants, while ensuring the accuracy of the computation and the credibility of the output results.

[0005] Quantitative assessment of uncertainty measures the uncertainty of a calculation using quantitative methods, thereby avoiding the information gaps caused by providing only the raw calculation results.

[0006] Transfer learning is a machine learning method in which a pre-trained model is reapplied to a different task. It is mainly used to address the problems that may arise when the training data and the application data are significantly inconsistent. Summary of the Invention

[0007] To address the aforementioned shortcomings, the technical problem this invention aims to solve is how to fully utilize the advantages of big data and modeling to accurately assess a patient's condition and provide a precise treatment plan, while protecting the patient's privacy.

[0008] To address the aforementioned deficiencies, the present invention aims to provide a method, system, electronic device, computer storage medium, and program product for protecting the privacy of users undergoing physical examinations.

[0009] According to one aspect of the embodiments of this specification, a method for protecting the privacy of users undergoing physical examinations is provided. This method involves joint modeling and training by at least one service provider and at least one internet business platform. Each service provider and internet business platform performs preliminary modeling based on its own collected data, establishes a model for each type of business based on historical business information and historical big data, obtains prediction results according to requirements, establishes a complete quantitative evaluation system for the uncertainty of the model prediction results, and verifies and tunes the trained model on new data using a trusted third party to reduce the uncertainty of prediction progress caused by differences in the training models of each party. Each party provides processed data attributes to the trusted third party, which then trains the model based on these attributes and returns the tuned model and parameters to each party. Each party calculates and returns prediction results to the trusted third party based on the new model, parameters, and its own data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0010] Preferably, the service provider and the internet business platform conduct preliminary modeling based on the data they have collected, and then build models to model big data of different business types, different groups of people and different detection schemes.

[0011] Preferably, the uncertainty system to be evaluated includes, but is not limited to, uncertainty in model prediction accuracy, uncertainty in model generalization ability, and uncertainty in data integrity.

[0012] Preferably, the data from at least one service provider has different dimensions, including but not limited to user personal information, query history information, relevant business data, and preliminary response plan data.

[0013] Preferably, the trusted third party has a trusted data storage space, including storage for data sent by all parties and independent model computation space.

[0014] Preferably, a trusted third party sends the model and exchanges modeling parameters to at least one service provider and an internet business platform.

[0015] According to another aspect of the embodiments of this specification, a method for protecting the privacy of users undergoing physical examinations is provided, applied to an internet business platform, comprising:

[0016] After receiving a joint modeling request from at least one service provider, the system performs preliminary modeling based on its collected data. It establishes models for each business based on historical business information and historical big data, obtains prediction results according to requirements, and establishes a complete quantitative evaluation system for the uncertainty of the model prediction results. Based on a trusted third party, the trained model is validated and its parameters are tuned on new data. The system provides the processed data attributes to the trusted third party. The trusted third party trains the model based on the data attributes of all parties and returns the tuned model and parameters to the Internet business platform. The Internet business platform calculates and returns the prediction results to the trusted third party based on the new model and parameters and the collected data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0017] Preferably, each party on the internet business platform conducts preliminary modeling based on the data they have collected, and then builds models to model big data of different business types, different groups of people and different detection schemes.

[0018] Preferably, the uncertainty system to be evaluated includes, but is not limited to, uncertainty in model prediction accuracy, uncertainty in model generalization ability, and uncertainty in data integrity.

[0019] According to another aspect of the embodiments of this specification, a method for protecting the privacy of users undergoing physical examinations is provided, applied to a trusted third-party platform, comprising: upon receiving a request for joint modeling training from at least one service provider and at least one internet business platform, establishing a training model, and receiving parameters from the service provider and internet business platform respectively after preliminary modeling based on their own collected data; verifying and adjusting parameters based on new data sent by each party; receiving processed data attributes provided by each party, training based on the data attributes of each party, and returning the optimized model and parameters to each party; each party calculating and returning prediction results to the trusted third party based on the new model and parameters and their respective data; if the result consistency condition is met, the process ends; otherwise, the process iterates until the result consistency condition is met.

[0020] Preferably, the third-party trusted platform has a trusted data storage space, including storage for data sent by all parties and independent model computing space.

[0021] Preferably, a trusted third-party platform sends the model and exchanges modeling parameters with at least one service provider and an internet business platform.

[0022] According to another aspect of the embodiments of this specification, a health checkup user privacy protection system is provided, comprising: at least one service provider, at least one internet business platform, and a trusted third-party platform, wherein,

[0023] The service provider performs preliminary modeling based on its collected data, establishes models for each business based on historical business information and historical big data, obtains prediction results according to requirements, establishes a complete quantitative evaluation system for the uncertainty of the model prediction results, and verifies and tunes the trained model on new data based on the trusted third party, provides the processed data attributes to the trusted third party, the trusted third party trains the model based on the data attributes of all parties and returns the tuned model and parameters to the Internet business platform, the Internet business platform calculates based on the new model and parameters and the collected data and returns the prediction results to the trusted third party, if the result consistency condition is met, the process ends; if not, the process iterates until the result consistency condition is met.

[0024] After receiving the joint modeling request from the service provider, the Internet business platform performs preliminary modeling based on its collected data, establishes models for each business based on historical business information and historical big data, obtains prediction results according to requirements, establishes a complete quantitative evaluation system for the uncertainty of the model prediction results, verifies and tunes the trained model on new data based on the trusted third party, provides the processed data attributes to the trusted third party, the trusted third party trains the model based on the data attributes of all parties and returns the tuned model and parameters to the Internet business platform, the Internet business platform calculates based on the new model and parameters and the collected data and returns the prediction results to the trusted third party, if the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0025] After receiving a request from the service provider and the internet business platform for joint modeling and training, the third-party trusted platform establishes a training model and receives parameters from the service provider and the internet business platform after preliminary modeling based on their respective collected data. It then verifies and tunes the parameters based on new data sent by each party, receives processed data attributes provided by each party, trains the model based on these attributes, and returns the tuned model and parameters to each party. Each party calculates and returns prediction results to the trusted third party based on the new model, parameters, and their own data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0026] Preferably, the uncertainty system to be evaluated includes, but is not limited to, uncertainty in model prediction accuracy, uncertainty in model generalization ability, and uncertainty in data integrity.

[0027] According to another aspect of the embodiments of this specification, a computer-readable storage medium is provided that stores a computer program / instructions thereon, which, when executed by a processor, performs the following steps:

[0028] The service provider and the internet business platform each conduct preliminary modeling based on their collected data, establish models for each business based on historical business information and historical big data, obtain prediction results according to requirements, and establish a complete quantitative evaluation system for the uncertainty of the model prediction results. Based on a trusted third party, the trained model is verified and its parameters are adjusted on new data to reduce the uncertainty of prediction progress caused by the differences in the training models of each party. Each party provides the processed data attributes to the trusted third party, which trains the model based on the data attributes of each party and returns the optimized model and parameters to each party. Each party calculates and returns the prediction results to the trusted third party based on the new model and parameters and its own data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0029] According to another aspect of the embodiments of this specification, a computer program product is provided, including a computer program / instructions that, when executed by a processor, perform the following steps:

[0030] The service provider and the internet business platform each conduct preliminary modeling based on their collected data, establish models for each business based on historical business information and historical big data, obtain prediction results according to requirements, and establish a complete quantitative evaluation system for the uncertainty of the model prediction results. Based on a trusted third party, the trained model is verified and its parameters are adjusted on new data to reduce the uncertainty of prediction progress caused by the differences in the training models of each party. Each party provides the processed data attributes to the trusted third party, which trains the model based on the data attributes of each party and returns the optimized model and parameters to each party. Each party calculates and returns the prediction results to the trusted third party based on the new model and parameters and its own data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0031] According to another aspect of the embodiments of this specification, an electronic device is provided, comprising:

[0032] Processor; and

[0033] A memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the following operations:

[0034] After receiving a joint modeling request from at least one service provider, the system performs preliminary modeling based on its collected data. It establishes models for each business based on historical business information and historical big data, obtains prediction results according to requirements, and establishes a complete quantitative evaluation system for the uncertainty of the model prediction results. Based on a trusted third party, the trained model is validated and its parameters are tuned on new data. The system provides the processed data attributes to the trusted third party. The trusted third party trains the model based on the data attributes of all parties and returns the tuned model and parameters to the Internet business platform. The Internet business platform calculates and returns the prediction results to the trusted third party based on the new model and parameters and the collected data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0035] According to another aspect of the embodiments of this specification, a computer-readable storage medium is provided that stores a computer program / instructions thereon, which, when executed by a processor, performs the following steps:

[0036] After receiving a joint modeling request from at least one service provider, the system performs preliminary modeling based on its collected data. It establishes models for each business based on historical business information and historical big data, obtains prediction results according to requirements, and establishes a complete quantitative evaluation system for the uncertainty of the model prediction results. Based on a trusted third party, the trained model is validated and its parameters are tuned on new data. The system provides the processed data attributes to the trusted third party. The trusted third party trains the model based on the data attributes of all parties and returns the tuned model and parameters to the Internet business platform. The Internet business platform calculates and returns the prediction results to the trusted third party based on the new model and parameters and the collected data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0037] According to another aspect of the embodiments of this specification, a computer program product is provided, including a computer program / instructions that, when executed by a processor, perform the following steps:

[0038] After receiving a joint modeling request from at least one service provider, the system performs preliminary modeling based on its collected data. It establishes models for each business based on historical business information and historical big data, obtains prediction results according to requirements, and establishes a complete quantitative evaluation system for the uncertainty of the model prediction results. Based on a trusted third party, the trained model is validated and its parameters are tuned on new data. The system provides the processed data attributes to the trusted third party. The trusted third party trains the model based on the data attributes of all parties and returns the tuned model and parameters to the Internet business platform. The Internet business platform calculates and returns the prediction results to the trusted third party based on the new model and parameters and the collected data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0039] According to another aspect of the embodiments of this specification, an electronic device is provided, comprising:

[0040] Processor; and

[0041] A memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the following operations:

[0042] Upon receiving a request for joint modeling and training from at least one service provider and at least one internet business platform, a training model is established. The model is then received by the service provider and the internet business platform based on their respective collected data and the parameters after initial modeling. The model is then validated and adjusted based on new data sent by each party. The model is trained based on the processed data attributes provided by each party and the optimized model and parameters are returned to each party. Each party calculates and returns the prediction results to a trusted third party based on the new model and parameters and their own data. If the result consistency condition is met, the process ends. If not, the process iterates until the result consistency condition is met.

[0043] According to another aspect of the embodiments of this specification, a computer-readable storage medium is provided that stores a computer program / instructions thereon, which, when executed by a processor, performs the following steps:

[0044] Upon receiving a request for joint modeling and training from at least one service provider and at least one internet business platform, a training model is established. The model is then received by the service provider and the internet business platform based on their respective collected data and the parameters after initial modeling. The model is then validated and adjusted based on new data sent by each party. The model is trained based on the processed data attributes provided by each party and the optimized model and parameters are returned to each party. Each party calculates and returns the prediction results to a trusted third party based on the new model and parameters and their own data. If the result consistency condition is met, the process ends. If not, the process iterates until the result consistency condition is met.

[0045] According to another aspect of the embodiments of this specification, a computer program product is provided, including a computer program / instructions that, when executed by a processor, perform the following steps:

[0046] Upon receiving a request for joint modeling and training from at least one service provider and at least one internet business platform, a training model is established. The model is then received by the service provider and the internet business platform based on their respective collected data and the parameters after initial modeling. The model is then validated and adjusted based on new data sent by each party. The model is trained based on the processed data attributes provided by each party and the optimized model and parameters are returned to each party. Each party calculates and returns the prediction results to a trusted third party based on the new model and parameters and their own data. If the result consistency condition is met, the process ends. If not, the process iterates until the result consistency condition is met.

[0047] This invention, by applying a privacy-protected multi-party secure computational basis and combining it with the concept of transfer learning, enables multiple service providers and internet business platforms that store patient information to jointly model and collaboratively infer data without sharing their own. This allows for the prediction of big data models for related businesses, the summarization of disease progression under various data sets, and the establishment of a quantitative assessment system for uncertainty to reasonably measure the uncertainty of prediction results. Consequently, it enables reasonable prediction of related outcomes and business development trends, providing decision-makers with a scientific reference. Attached Figure Description

[0048] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments of the present invention will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This diagram illustrates a structural schematic of an embodiment of the method for protecting user privacy during physical examinations according to the present invention.

[0050] Figure 2 A schematic diagram of another embodiment of the method for protecting user privacy during physical examinations according to the present invention is shown;

[0051] Figure 3 A schematic diagram of another embodiment of the method for protecting user privacy during physical examinations according to the present invention is shown;

[0052] Figure 4 This diagram illustrates a flowchart of an embodiment of the method for protecting user privacy during physical examinations according to the present invention.

[0053] Figure 5 This diagram illustrates another embodiment of the method for protecting user privacy during physical examinations according to the present invention.

[0054] Figure 6 A schematic flowchart of another embodiment of the method for protecting user privacy during physical examinations according to the present invention is shown. Detailed Implementation

[0055] The features and exemplary embodiments of various aspects of the present invention will now be described in detail. To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only configured to explain the present invention and are not configured to limit the present invention. For those skilled in the art, the present invention can be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present invention by illustrating examples of the invention.

[0056] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0057] like Figure 1 As shown in this specification, one embodiment provides a method for protecting the privacy of users undergoing physical examinations. This method involves joint modeling and training by at least one service provider 11 and at least one internet business platform 12. Each service provider 11 and internet business platform 12 performs preliminary modeling based on its own collected data, establishing models for each type of business based on historical business information and historical big data. Prediction results are obtained according to requirements. A complete quantitative evaluation system is established to assess the uncertainty of the model prediction results. The trained model is validated and its parameters are adjusted using new data based on a trusted third party (or a trusted third-party platform) 13, reducing the uncertainty in prediction progress caused by differences in the training models of each party. Each party provides processed data attributes to the trusted third party 13. The trusted third party 13 trains the model based on the data attributes of each party and returns the optimized model and parameters to each party. Each party calculates and returns prediction results to the trusted third party 13 based on the new model and parameters and its own data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0058] In some embodiments, the service provider and the internet business platform conduct preliminary modeling based on the data they have collected, and model the big data of different business types, different groups of people and different detection schemes by establishing the model.

[0059] It is worth noting that the uncertainty system that needs to be evaluated includes, but is not limited to, uncertainty in model prediction accuracy, uncertainty in model generalization ability, and uncertainty in data integrity.

[0060] In some embodiments, the data from at least one service provider has different dimensions, including but not limited to user personal information, query history information, relevant business data, and preliminary response plan data.

[0061] In some embodiments, a trusted third party has a trusted data storage space, including a space for storing data sent by all parties and an independent model computation space.

[0062] In some embodiments, a trusted third party sends models and exchanges modeling parameters to at least one service provider and an internet business platform.

[0063] In a specific example, the aforementioned service provider can be an external partner organization, such as a chain medical examination center. These medical examination centers are service organizations with medical qualifications. Some are affiliated units of hospitals, while others are professional chain medical examination institutions. Common examples include iKang Healthcare Group, Meinian Onehealth Healthcare, and Ciming Healthcare.

[0064] In a specific example, an internet business platform can be a comprehensive internet health and medical platform, such as AliHealth and JD Health; it can also be a platform specializing in medical and health consultations, such as Chunyu Doctor and DXY; or it can be a specialized medical platform, such as Good Mood Mobile Medical, a professional mobile medical platform focusing on the central nervous system.

[0065] like Figure 2 As shown, taking an external partner organization as an example, one embodiment of this specification provides a method for protecting the privacy of physical examination users. This method involves joint modeling and training between at least one physical examination institution and at least one internet business platform. Each institution and platform performs preliminary modeling based on its own collected data, establishing models for each business based on historical business information and big data. Prediction results are obtained according to requirements. A complete quantitative evaluation system is established to assess the uncertainty of the model's prediction results. A trusted third party verifies and adjusts the trained model on new data, reducing the uncertainty in prediction progress caused by differences in the training models of each party. Each party provides processed data attributes to the trusted third party. The trusted third party trains the model based on these attributes and returns the optimized model and parameters to each party. Each party calculates and returns prediction results to the trusted third party based on the new model, parameters, and their own data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0066] like Figure 3As shown, in a specific example, a medical examination institution, a service provider, and an internet business platform can be participating parties. The service provider can act as a trusted third-party platform or as a participating party. As a participating party, one embodiment of this specification provides a method for protecting the privacy of medical examination users. This method involves joint modeling and training by at least one medical examination institution, at least one service provider, and at least one internet business platform. Each party performs preliminary modeling based on its collected data, establishes a model for each type of business based on historical business information and historical big data, obtains prediction results according to requirements, establishes a complete quantitative evaluation system for the uncertainty of the model prediction results, and verifies and adjusts the trained model on new data based on a trusted third party to reduce the uncertainty of prediction progress caused by differences in the training models of each party. Each party provides processed data attributes to the trusted third party, which then trains the model based on the data attributes and returns the optimized model and parameters to each party. Each party calculates and returns prediction results to the trusted third party based on the new model and parameters and its own data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0067] In embodiments of the present invention, the trusted third-party platform can be a national medical regulatory agency, such as local medical insurance centers or pharmaceutical supervision and management departments.

[0068] In a specific example, to enhance data privacy protection and improve oversight of data and modeling activities by all parties, the service provider, internet business platform, and trusted third-party platform act as nodes in the consortium blockchain. Based on the blockchain (consortium blockchain), each participant participates as a node, and data verification, data attributes, and modeling parameters are hashed, encrypted, and stored on the blockchain, generating corresponding digital certificates as unique identifiers. In the event of medical malpractice disputes later, the actions of all parties in the entire behavioral chain can be located and confirmed using trusted means, providing a foundation for clarifying responsibilities.

[0069] like Figure 4 As shown, from the perspective of an internet business platform, an embodiment of a method for protecting the privacy of users undergoing physical examinations is provided, including the following steps:

[0070] S201. After receiving a joint modeling request from at least one service provider, perform preliminary modeling based on the data collected by itself;

[0071] S202. Build a model for each business based on historical business information and historical big data, and obtain prediction results based on demand;

[0072] S203. Establish a complete quantitative evaluation system for the uncertainty of model prediction results; verify and tune the trained model on new data based on a trusted third party, and provide the processed data attributes to the trusted third party.

[0073] S204. A trusted third party trains the model and parameters based on the data attributes of each party and then returns the optimized model and parameters to the Internet business platform.

[0074] S205. The Internet business platform calculates and returns the prediction results to a trusted third party based on the new model and parameters and the collected data.

[0075] S206. If the result consistency condition is met, the process ends; otherwise, the iteration continues until the result consistency condition is met.

[0076] In some embodiments, each party of the Internet business platform performs preliminary modeling based on the data it has collected, and models big data of different business types, different groups of people and different detection schemes by establishing models.

[0077] It is particularly important to note that the uncertainty system that needs to be evaluated includes, but is not limited to, uncertainty in the model's prediction accuracy, uncertainty in the model's generalization ability, and uncertainty in data integrity.

[0078] like Figure 5 As shown, from the perspective of a trusted third-party platform, an embodiment of a method for protecting the privacy of users undergoing physical examinations is provided, including the following steps:

[0079] S301. Upon receiving a request for joint modeling training from at least one service provider and at least one Internet business platform, establish a training model.

[0080] S302. Receive the parameters after preliminary modeling by the service provider and the Internet business platform based on the data they have collected;

[0081] S303. Verify and adjust parameters based on new data sent by all parties;

[0082] S304. After receiving the processed data attributes provided by each party, the model and parameters are trained based on the data attributes of each party and then returned to each party.

[0083] S305. Each party shall calculate and return the prediction results to a trusted third party based on the new model, parameters, and their respective data.

[0084] S306. If the result consistency condition is met, the process ends; otherwise, the iteration continues until the result consistency condition is met.

[0085] In some embodiments, the third-party trusted platform has a trusted data storage space, including a space for storing data sent by all parties and an independent model computation space.

[0086] In some embodiments, a third-party trusted platform sends models and exchanges modeling parameters to at least one service provider and an internet business platform.

[0087] like Figure 6 As shown, from the perspective of a medical examination institution, an embodiment of a method for protecting the privacy of medical examination users is provided, including the following steps:

[0088] S401. Upon receiving a joint modeling request from at least one service provider and / or at least one Internet business platform, perform preliminary modeling based on the data collected by itself.

[0089] S402. Build a model for each business based on historical business information and historical big data, and obtain prediction results based on demand;

[0090] S403. Establish a complete quantitative evaluation system for the uncertainty of model prediction results; verify and tune the trained model on new data based on a trusted third party, and provide the processed data attributes to the trusted third party.

[0091] S404. A trusted third party trains the model and parameters based on the data attributes of all parties and then returns the optimized model and parameters to the medical examination institution.

[0092] S405. The medical examination institution calculates and returns the prediction results to a trusted third party based on the new model and parameters and the collected data.

[0093] S406. If the result consistency condition is met, the process ends; otherwise, the iteration continues until the result consistency condition is met.

[0094] In some embodiments, medical examination institutions conduct preliminary modeling based on the data they have collected, and then build models to model big data of different business types, different populations and different testing schemes.

[0095] It is worth noting that the uncertainty system that needs to be evaluated includes, but is not limited to, uncertainty in model prediction accuracy, uncertainty in model generalization ability, and uncertainty in data integrity.

[0096] According to another embodiment, a health checkup user privacy protection system is also provided, comprising: at least one service provider, at least one internet business platform, and a trusted third-party platform, wherein,

[0097] The service provider performs preliminary modeling based on its collected data, establishes models for each business based on historical business information and historical big data, obtains prediction results according to requirements, establishes a complete quantitative evaluation system for the uncertainty of the model prediction results, and verifies and tunes the trained model on new data based on the trusted third party, provides the processed data attributes to the trusted third party, the trusted third party trains the model based on the data attributes of all parties and returns the tuned model and parameters to the Internet business platform, the Internet business platform calculates based on the new model and parameters and the collected data and returns the prediction results to the trusted third party, if the result consistency condition is met, the process ends; if not, the process iterates until the result consistency condition is met.

[0098] After receiving the joint modeling request from the service provider, the Internet business platform performs preliminary modeling based on its collected data. It establishes models for each business based on historical business information and historical big data, obtains prediction results according to requirements, and establishes a complete quantitative evaluation system for the uncertainty of the model prediction results. Based on the trusted third party, it verifies and tunes the trained model under new data and provides the trusted third party with the processed data attributes. The trusted third party trains the model based on the data attributes of all parties and returns the tuned model and parameters to the Internet business platform. The Internet business platform calculates and returns the prediction results to the trusted third party based on the new model and parameters and the collected data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0099] After receiving a request from the service provider and the internet business platform for joint modeling and training, the third-party trusted platform establishes a training model and receives parameters from the service provider and the internet business platform after preliminary modeling based on their respective collected data. It then verifies and tunes the parameters based on new data sent by each party, receives the processed data attributes provided by each party, trains the model based on the data attributes of each party, and returns the tuned model and parameters to each party. Each party calculates and returns the prediction results to the trusted third party based on the new model and parameters and their own data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0100] In some embodiments, the uncertainty system to be evaluated includes, but is not limited to, uncertainty in model prediction accuracy, uncertainty in model generalization ability, and uncertainty in data integrity.

[0101] According to another embodiment, a computer-readable storage medium is also provided, on which a computer program / instructions are stored, which, when executed by a processor, perform the following steps:

[0102] The service provider and the internet business platform each conduct preliminary modeling based on their collected data. They establish models for each business based on historical business information and big data, obtain prediction results according to requirements, and establish a complete quantitative evaluation system for the uncertainty of the model prediction results. Based on a trusted third party, the trained model is validated and its parameters are adjusted on new data to reduce the uncertainty of prediction progress caused by the differences in the training models of each party. Each party provides the processed data attributes to the trusted third party. The trusted third party trains the model based on the data attributes of each party and returns the optimized model and parameters to each party. Each party calculates and returns the prediction results to the trusted third party based on the new model and parameters and their own data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0103] According to another embodiment, a computer program product is also provided, including a computer program / instructions that, when executed by a processor, perform the following steps:

[0104] The service provider and the internet business platform each conduct preliminary modeling based on their collected data. They establish models for each business based on historical business information and big data, obtain prediction results according to requirements, and establish a complete quantitative evaluation system for the uncertainty of the model prediction results. Based on a trusted third party, the trained model is validated and its parameters are adjusted on new data to reduce the uncertainty of prediction progress caused by the differences in the training models of each party. Each party provides the processed data attributes to the trusted third party. The trusted third party trains the model based on the data attributes of each party and returns the optimized model and parameters to each party. Each party calculates and returns the prediction results to the trusted third party based on the new model and parameters and their own data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0105] According to another aspect of the embodiments of this specification, an electronic device is also provided, comprising:

[0106] Processor; and

[0107] A memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the following operations:

[0108] After receiving a joint modeling request from at least one service provider, the system performs preliminary modeling based on its collected data. It establishes models for each business based on historical business information and historical big data, obtains prediction results according to requirements, and establishes a complete quantitative evaluation system for the uncertainty of the model prediction results. Based on a trusted third party, the trained model is validated and its parameters are tuned on new data. The system provides the processed data attributes to the trusted third party. The trusted third party trains the model based on the data attributes of all parties and returns the tuned model and parameters to the Internet business platform. The Internet business platform calculates and returns the prediction results to the trusted third party based on the new model and parameters and the collected data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0109] According to another aspect of the embodiments of this specification, a computer-readable storage medium is also provided, on which a computer program / instructions are stored, which, when executed by a processor, perform the following steps:

[0110] After receiving a joint modeling request from at least one service provider, the system performs preliminary modeling based on its collected data. It establishes models for each business based on historical business information and historical big data, obtains prediction results according to requirements, and establishes a complete quantitative evaluation system for the uncertainty of the model prediction results. Based on a trusted third party, the trained model is validated and its parameters are tuned on new data. The system provides the processed data attributes to the trusted third party. The trusted third party trains the model based on the data attributes of all parties and returns the tuned model and parameters to the Internet business platform. The Internet business platform calculates and returns the prediction results to the trusted third party based on the new model and parameters and the collected data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0111] According to another aspect of the embodiments of this specification, a computer program product is provided, including a computer program / instructions that, when executed by a processor, perform the following steps:

[0112] After receiving a joint modeling request from at least one service provider, the system performs preliminary modeling based on its collected data. It establishes models for each business based on historical business information and historical big data, obtains prediction results according to requirements, and establishes a complete quantitative evaluation system for the uncertainty of the model prediction results. Based on a trusted third party, the trained model is validated and its parameters are tuned on new data. The system provides the processed data attributes to the trusted third party. The trusted third party trains the model based on the data attributes of all parties and returns the tuned model and parameters to the Internet business platform. The Internet business platform calculates and returns the prediction results to the trusted third party based on the new model and parameters and the collected data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

[0113] According to another aspect of the embodiments of this specification, an electronic device is provided, comprising:

[0114] Processor; and

[0115] A memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the following operations:

[0116] Upon receiving a request for joint modeling and training from at least one service provider and at least one internet business platform, a training model is established. The model is then received by the service provider and the internet business platform based on their respective collected data and the parameters after initial modeling. The model is then validated and adjusted based on new data sent by each party. The model is trained based on the processed data attributes provided by each party and the optimized model and parameters are returned to each party. Each party calculates and returns the prediction results to a trusted third party based on the new model and parameters and their own data. If the result consistency condition is met, the process ends. If not, the process iterates until the result consistency condition is met.

[0117] According to another aspect of the embodiments of this specification, a computer-readable storage medium is provided that stores a computer program / instructions thereon, which, when executed by a processor, performs the following steps:

[0118] Upon receiving a request for joint modeling and training from at least one service provider and at least one internet business platform, a training model is established. The model is then received by the service provider and the internet business platform based on their respective collected data and the parameters after initial modeling. The model is then validated and adjusted based on new data sent by each party. The model is trained based on the processed data attributes provided by each party and the optimized model and parameters are returned to each party. Each party calculates and returns the prediction results to a trusted third party based on the new model and parameters and their own data. If the result consistency condition is met, the process ends. If not, the process iterates until the result consistency condition is met.

[0119] According to another aspect of the embodiments of this specification, a computer program product is provided, including a computer program / instructions that, when executed by a processor, perform the following steps:

[0120] Upon receiving a request for joint modeling and training from at least one service provider and at least one internet business platform, a training model is established. The model is then received by the service provider and the internet business platform based on their respective collected data and the parameters after initial modeling. The model is then validated and adjusted based on new data sent by each party. The model is trained based on the processed data attributes provided by each party and the optimized model and parameters are returned to each party. Each party calculates and returns the prediction results to a trusted third party based on the new model and parameters and their own data. If the result consistency condition is met, the process ends. If not, the process iterates until the result consistency condition is met.

[0121] The multi-scheme secure computation framework provided by the embodiments of the present invention can ensure that the private data of each participant (including service providers and Internet business platforms) does not leave the domain, or in other words, that the private data of each participant is not leaked, so that the computation model outputs the same result.

[0122] This invention utilizes a privacy-protected multi-party secure computational basis combined with the concept of transfer learning to enable multiple service providers and internet business platforms storing patient information to jointly model and collaboratively infer data without sharing their own. This allows for the prediction of big data models for related businesses, the summarization of disease progression under various data sets, and the establishment of a quantitative assessment system for uncertainty to reasonably measure the uncertainty of prediction results. Consequently, it enables reasonable prediction of business outcomes and development trends, providing decision-makers with a scientific reference.

[0123] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0124] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0125] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0126] This application can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0127] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0128] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0129] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0130] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0131] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0132] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0133] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0134] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for protecting the privacy of users undergoing physical examinations, comprising joint modeling and training by at least one chain physical examination institution with medical qualifications and at least one internet business platform. Each of the physical examination institution and the internet business platform performs preliminary modeling based on its own collected data, establishes a model for each type of business based on historical business information and historical big data, obtains prediction results according to requirements, establishes a complete quantitative evaluation system for the uncertainty of the model prediction results, verifies and adjusts the trained model on new data based on a trusted third party, reduces the uncertainty of prediction accuracy caused by differences in the training models of each party, provides the trusted third party with processed data attributes, the trusted third party trains based on the data attributes of each party and returns the optimized model and parameters to each party, each party calculates and returns the prediction results to the trusted third party based on the new model and parameters and its own data, and if the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

2. According to the method for protecting the privacy of physical examination users as described in claim 1, the physical examination institution and the Internet business platform shall perform preliminary modeling based on the data they have collected, and model the big data of different business types, different groups of people and different physical examination plans by establishing the model.

3. The method for protecting the privacy of physical examination users according to claim 1 or 2, the uncertainty system to be evaluated includes uncertainty in model prediction accuracy, uncertainty in model generalization ability, and uncertainty in data integrity.

4. The method for protecting the privacy of physical examination users according to claim 1, wherein the data of the physical examination institution has different dimensions, including user personal information, query history information, relevant business data, and preliminary response plan data.

5. The method for protecting the privacy of physical examination users according to claim 1, wherein the trusted third party has a trusted data storage space, including a space for storing data sent by all parties and an independent model computation space.

6. In the method for protecting the privacy of physical examination users according to claim 5, the trusted third party sends the model and exchanges modeling parameters with the physical examination institution and the Internet business platform.

7. A method for protecting the privacy of users undergoing physical examinations, applied to a trusted third party, comprising: Upon receiving a request for joint modeling training from at least one chain medical examination institution and at least one internet business platform, a training model is established. The model receives parameters from the medical examination institution and the internet business platform after preliminary modeling based on their respective collected data. The model is then validated and its parameters are adjusted based on new data sent by each party. The model is trained based on the processed data attributes provided by each party and then returned to each party with the optimized model and parameters. Each party calculates and returns the prediction results to a trusted third party based on the new model, parameters, and their own data. If the result consistency condition is met, the process ends. If not, the process iterates until the result consistency condition is met.

8. The method according to claim 7, wherein the trusted third party has a trusted data storage space, including a space for storing data sent by all parties and an independent model computation space.

9. The method according to claim 7, wherein the trusted third party sends the model and exchanges modeling parameters to at least one medical examination institution and an Internet business platform.

10. A user privacy protection system for physical examinations, comprising: At least one chain medical examination institution, at least one internet business platform, and a trusted third party, wherein, The medical examination institution conducts preliminary modeling based on its own collected data, establishes models for each type of business based on historical business information and historical big data, obtains prediction results according to needs, establishes a complete quantitative evaluation system for the uncertainty of the model prediction results, and verifies and adjusts the trained model on new data based on the trusted third party, and provides the processed data attributes to the trusted third party. After receiving the joint modeling request from the medical examination institution, the Internet business platform performs preliminary modeling based on the data it has collected, establishes models for each type of business based on historical business information and historical big data, obtains prediction results according to requirements, establishes a complete quantitative evaluation system for the uncertainty of the model prediction results, and verifies and adjusts the trained model on new data based on the trusted third party, and provides the processed data attributes to the trusted third party. After receiving the request for joint modeling and training from the medical examination institution and the internet business platform, the trusted third party establishes a training model and receives the parameters of the preliminary modeling performed by the medical examination institution and the internet business platform based on their respective collected data. The third party verifies and adjusts the parameters based on the new data sent by each party, receives the processed data attributes provided by each party, trains based on the data attributes of each party, and returns the optimized model and parameters to each party. Each party calculates and returns the prediction results to the trusted third party based on the new model and parameters and their own data. If the result consistency condition is met, the process ends; otherwise, it iterates until the result consistency condition is met.

Images