Private data desensitization method, system and equipment based on medical data

Abstract

The invention discloses a privacy data desensitization method, system and device based on medical data, and the method comprises the steps: carrying out the full scanning of a header of a database, and obtaining a field name contained in the database; mapping the field name with a preset information tag in a desensitization tag library to obtain a data asset directory; when the doctor-seeing data is called by a system, according to the field names corresponding to the field data in the called doctor-seeing data, obtaining the corresponding information labels of the field names corresponding to the field data through the data asset directory, and judging whether the field data corresponding to each information label needs to be subjected to desensitization processing or not according to a preset method; and when the field data corresponding to the information tag needs to be desensitized, selecting a desensitization rule in a desensitization rule library corresponding to the information tag according to a preset rule, desensitizing the field data corresponding to the information tag according to the desensitization rule, and returning the field data to a system.

Description

Technical Field

[0001] The present application relates to the field of data security technology, and in particular to a privacy data desensitization method system and device based on medical data. Background Art

[0002] The health and medical big data centers currently established in various places are mainly used to build health and medical big data platforms, gather medical data from various hospitals, and form various thematic data through governance and integration of such data to provide public services. With the aggregation of hospital data, personal medical data and health and medical conditions are integrated into the data center. The massive amount of health and medical data managed by the data center can easily form all-round personal medical data. This type of data has extremely high scientific research value and economic value, and is also personal privacy data. If strict control is not implemented, the security of medical privacy data cannot be guaranteed. Strict data security control methods are required to ensure the legality of data use.

[0003] At present, health and medical data desensitization mainly relies on traditional technical means such as data encryption, anonymization, and perturbation. These technologies have reduced the risk of data leakage to a certain extent, but there are still many deficiencies in practical applications. For example, they often lack the ability to identify and classify data fields, and it is difficult to perform refined desensitization according to the actual use scenarios of the data and the different permissions of the visitors.

[0004] Therefore, a privacy data desensitization method based on medical data is needed, which uses data desensitization technology to perform fine-grained desensitization on the patient's medical data, while ensuring the security of the patient's information and ensuring that the desensitized data can still meet scientific research, clinical and other needs. Summary of the invention

[0005] The embodiments of the present application provide a privacy data desensitization method system and device based on medical data, which solves the current lack of refined desensitization of patients' medical data, and can ensure that the desensitized data can still meet the technical problems of scientific research, clinical and other needs while ensuring the security of patients' information.

[0006] In the first aspect, an embodiment of the present application provides a privacy data desensitization method based on medical data, comprising: performing a full scan on the table header of the database to obtain the field names contained in the database; mapping the field names with the information tags preset in the desensitization tag library to obtain a data asset directory; wherein the desensitization tag library contains a number of information tags, and each information tag corresponds to a desensitization rule library; when the medical data is called by the system, according to the field names corresponding to the field data in the called medical data, the corresponding information tags corresponding to the field names of the field data are obtained through the data asset directory, and it is determined according to the preset method whether the field data corresponding to each information tag needs to be desensitized; when the field data corresponding to the information tag needs to be desensitized, a desensitization rule is selected in the desensitization rule library corresponding to the information tag according to the preset rule, and the field data corresponding to the information tag is desensitized according to the desensitization rule, and then returned to the system.

[0007] Furthermore, a data asset directory is obtained, which specifically includes: scanning and identifying field names through an NLP algorithm to obtain corresponding information; matching the obtained information with preset information tags according to semantics, and forming a mapping relationship between the field names and the matched information tags; and saving the mapping relationship between the field names and the information tags as a data asset directory.

[0008] Furthermore, information labels are classified into strong labels and weak labels, and a desensitized weak label combination library is set, the desensitized weak label combination library includes several weak label combinations, and each combination includes several weak labels.

[0009] Furthermore, the preset method specifically includes: classifying the information tags corresponding to the field names of each field data in the called medical data; for information tags classified as strong labels: determining whether the field data corresponding to the information tag needs to be desensitized; for information tags classified as weak labels: determining whether there is a weak label combination in the desensitized weak label combination library among all information tags classified as weak labels; if so, determining whether the field data corresponding to the label combination needs to be desensitized; if not, not desensitizing the field data corresponding to the information tags classified as weak labels.

[0010] Furthermore, each desensitization rule library contains a number of desensitization rules of different levels for the information tag; the preset rules specifically include: selecting a desensitization rule of a corresponding level in the desensitization rule library of the information tag according to the caller's authority.

[0011] Furthermore, it also includes: incrementally scanning the database according to a preset time threshold, matching the field names of the newly added tables in the database with preset information tags, and storing the mapping relationship between the field names and the matched information tags in the data asset directory.

[0012] Furthermore, it also includes: when the field name of the newly added table does not match the preset information tag, the newly added field name is added as a new information tag to the desensitizing tag library, and the corresponding information tag classification and desensitizing rule library are set for the new information tag; when the new information tag is classified as a weak tag, the corresponding weak tag combination is added to the desensitizing weak tag combination library.

[0013] In the second aspect, the embodiment of the present application also provides a privacy data desensitization system based on medical data, which is used to implement any one of the steps of the above method. The system includes: a scanning module, which performs a full scan on the table header of the database to obtain the field names contained in the database; a mapping module, which maps the field names with the preset information tags in the desensitization label library to obtain a data asset directory; wherein the desensitization label library contains a number of information tags, and each information tag corresponds to a desensitization rule library; a judgment module, when the medical data is called by the system, the judgment module obtains the corresponding information tag corresponding to the field name of the field data through the data asset directory according to the field name corresponding to each field data in the called medical data, and judges whether the field data corresponding to each information tag needs to be desensitized according to the preset method; the desensitization module, when the word corresponding to the information tag When segment data needs to be desensitized, the desensitizing module selects a desensitizing rule from the desensitizing rule library corresponding to the information tag according to the preset rules, desensitizes the field data corresponding to the information tag according to the desensitizing rule, and returns it to the system; the incremental scanning module, the incremental scanning module performs an incremental scan on the database according to the preset time threshold, matches the field name of the newly added table in the database with the preset information tag, and stores the mapping relationship between the field name and the matched information tag in the data asset directory; the update module, the update module is used to add the newly added field name as a new information tag to the desensitizing tag library when the field name of the newly added table does not match the preset information tag, and sets the corresponding information tag classification and desensitizing rule library for the new information tag; when the new information tag is classified as a weak tag, add the corresponding weak tag combination to the desensitizing weak tag combination library.

[0014] Furthermore, information labels are classified into strong labels and weak labels, and a desensitized weak label combination library is set, the desensitized weak label combination library contains several weak label combinations, and each combination contains several weak labels; the preset method specifically includes: classifying the information labels corresponding to the field names of each field data in the called medical data; for information labels classified as strong labels: judging whether the field data corresponding to the information label needs to be desensitized; for information labels classified as weak labels: judging whether there is a weak label combination in the desensitized weak label combination library among all information labels classified as weak labels, if so, judging that the field data corresponding to the label combination needs to be desensitized; if not, not desensitizing the field data corresponding to the information labels classified as weak labels.

[0015] In the third aspect, an embodiment of the present application also provides a privacy data desensitizing device based on medical data, the device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can: implement any one step of the above method.

[0016] The embodiment of the present application provides a privacy data desensitization method for health and medical big data. By scanning database fields and using the NLP algorithm to map field names with preset information tags, a data asset directory is automatically generated, which significantly improves the automated management and matching efficiency of desensitization rules. According to the information tags corresponding to the field names, strong tags and weak tags are classified and processed to further refine the desensitization control of privacy data, implement differentiated desensitization strategies for data of different sensitivities, and enhance the flexibility and security of privacy data protection. In addition, a permission control mechanism is adopted, which can select desensitization rules of corresponding levels according to the caller's permissions, thereby ensuring a balance between data availability and privacy protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings: Figure 1 A method flow chart of a privacy data desensitization method based on medical data provided in an embodiment of the present application; Figure 2 A flow chart of a method for obtaining a data asset catalog provided in an embodiment of the present application; Figure 3 A method flow chart of a preset method provided in an embodiment of the present application; Figure 4 A schematic diagram of the internal structure of a privacy data desensitization system based on medical data provided in an embodiment of the present application.

[0018] Figure 5 A schematic diagram of the internal structure of a privacy data desensitizing device based on medical data provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0020] The technical solution proposed in the embodiments of the present application is described in detail below with reference to the accompanying drawings.

[0021] Figure 1 A flowchart of a privacy data desensitization method based on medical data provided in an embodiment of the present application. Figure 1 As shown, the privacy data desensitization method based on medical data provided in the embodiment of the present application specifically includes the following steps: Step S01: perform a full scan on the table header of the database to obtain the field names contained in the database.

[0022] In one embodiment of the present application, the method also includes: performing an incremental scan of the database according to a preset time threshold, matching the field names of the newly added tables in the database with preset information tags, and storing the mapping relationship between the field names and the matched information tags in the data asset directory.

[0023] Step S02: Map the field name with the information tag preset in the desensitizing tag library to obtain a data asset directory.

[0024] The desensitizing tag library contains several information tags, and each information tag corresponds to a desensitizing rule library.

[0025] Optionally, the desensitization rules preset in the desensitization rule library may include: for the patient's name, address and illness, a desensitization rule is used to truncate the patient's name field data, address field data and the patient's illness field data; for age, a desensitization rule is used to calculate the mean of the age data and then set a random value to replace the mean attachment; for ID card number and mobile phone number, a desensitization rule is used to replace the ID card number data and mobile phone number data with key values ​​in the dictionary; for fingerprint data, face data and medical image data, a desensitization rule is used to blur the image; for the date of consultation, a desensitization rule is used to offset the date of consultation data by a preset offset.

[0026] Figure 2 A flow chart of a method for obtaining a data asset directory provided in an embodiment of the present application, such as Figure 2 As shown, the data asset catalog includes: Through the NLP algorithm, the field names are scanned and identified to obtain the corresponding information.

[0027] The obtained information is matched with the preset information tag according to semantics, and a mapping relationship is formed between the field name and the matched information tag.

[0028] The mapping relationship between the field name and the information tag is saved as a data asset directory.

[0029] It should be noted that NLP algorithms are existing technologies that use technologies such as word segmentation, part-of-speech tagging, and named entity recognition in natural language processing to analyze and identify field name information in the database. Through these technologies, key information in field names can be accurately extracted, and based on semantic similarity calculation or semantic matching algorithms, the identified field name information is semantically matched with preset information tags to generate a mapping relationship between field names and information tags, thereby effectively building a data asset catalog and realizing automated desensitization of privacy data.

[0030] Step S03, when the medical data is called by the system, according to the field name corresponding to each field data in the called medical data, the corresponding information label of the field name corresponding to the field data is obtained through the data asset directory, and it is determined according to the preset method whether the field data corresponding to each information label needs to be desensitized.

[0031] In one embodiment of the present application, information tags are classified into strong tags and weak tags, and a desensitized weak tag combination library is set, wherein the desensitized weak tag combination library includes several weak tag combinations, and each combination includes several weak tags.

[0032] Specifically, a strong tag is a single information tag that directly identifies an individual and an information tag related to disease information, including at least: the patient's name, ID number, mobile phone number, address, fingerprint data, face data, and the patient's illness; among them, the patient's illness includes infectious diseases and sensitive diseases; infectious diseases include Class A infectious diseases and Class B infectious diseases defined in the "Law of the People's Republic of China on the Prevention and Control of Infectious Diseases (Revised)", and sensitive diseases are preset non-communicable diseases that affect personal safety and reputation. In other words, this application mainly desensitizes diseases that may affect the privacy and reputation of patients, protecting the privacy and safety of patients.

[0033] Weak labels are information labels that can only be combined to identify an individual, such as the patient's medical insurance coverage + disease symptoms + gender + height + age.

[0034] Figure 3 A method flow chart of the preset method provided in the embodiment of the present application, such as Figure 3 As shown, the preset method specifically includes: Classify the information labels corresponding to the field names of the data fields in the called medical consultation data.

[0035] For informational tags classified as strong tags: It is determined that the field data corresponding to the information tag needs to be desensitized.

[0036] For information labels classified as weak labels: It is determined whether there is a weak label combination in the desensitized weak label combination library among all the information labels classified as weak labels. If so, it is determined that each field data corresponding to the label combination needs to be desensitized.

[0037] If it does not exist, the field data corresponding to the information label classified as a weak label will not be desensitized.

[0038] That is to say, the embodiment of the present application further refines the desensitization control of privacy data by classifying and processing strong labels and weak labels according to the information labels corresponding to the field names, implements differentiated desensitization strategies for data of different sensitivities, and enhances the flexibility and security of privacy data protection.

[0039] In one embodiment of the present application, during the incremental scanning process, when the field name of the newly added table does not have a matching preset information tag, the newly added field name is added to the desensitizing tag library as a new information tag, and a corresponding information tag classification and desensitizing rule library are set for the new information tag.

[0040] When the new information label is classified as a weak label, the corresponding weak label combination is added to the desensitized weak label combination library.

[0041] That is to say, this application sets up the function of updating and adding information tags. When new field names appear, the information tags in the desensitizing tag library are updated, so that this application can perform continuous optimization and improve the flexibility and adaptability of the system.

[0042] Step S04: When the field data corresponding to the information tag needs to be desensitized, a desensitization rule is selected from the desensitization rule library corresponding to the information tag according to a preset rule, and the field data corresponding to the information tag is desensitized according to the desensitization rule and then returned to the system.

[0043] In one embodiment of the present application, each desensitization rule library contains a number of desensitization rules of different levels for the information tag.

[0044] Specifically, the preset rules include: According to the caller's authority, a desensitization rule of a corresponding level is selected from the desensitization rule library of the information tag.

[0045] In one embodiment of the present application, user rights include attending physicians, patients, other doctors and other patients, so a number of desensitization rules are set for the desensitization rule library whose information label is ID card number, including: Corresponding to the desensitization rule 1 of the attending physician, the last four digits of the ID number are desensitized using a replacement algorithm.

[0046] According to desensitization rule 2 for the patient, no desensitization treatment is performed.

[0047] Corresponding to the desensitization rule 3 of other doctors, the first six digits and the last four digits of the ID number are desensitized using a replacement algorithm.

[0048] Corresponding to desensitization rule 4 for other patients, a replacement algorithm is used to desensitize all digits of the ID card number.

[0049] That is to say, the embodiment of the present application adopts a permission control mechanism, which can select the corresponding level of desensitization rules according to the caller's permissions, thereby ensuring a balance between data availability and privacy protection.

[0050] In one embodiment, a static desensitization process is also included, including: batch reading a number of field data in the database according to the number of user requirements, matching corresponding information tags through the data asset directory, and selecting preset desensitization rules in the desensitization rule library corresponding to the information tags to perform desensitization processing on the number of field data, and outputting them as a demand desensitized data set.

[0051] It should be noted that the data desensitization function usually only reads the original data, and then modifies or replaces the data according to the preset desensitization rules, and finally generates a desensitized data copy. This desensitized data copy will be saved to the preset target location, and the original data will not be changed during the desensitization process. Therefore, from the perspective of data modification, data desensitization will not affect the original data called.

[0052] The above is an embodiment of the method proposed in this application. Based on the same inventive concept, the embodiment of this application also provides a system, whose structure is as follows: Figure 4 shown.

[0053] The present application embodiment provides a privacy data desensitization system based on medical data, the system comprising: The scanning module performs a full scan on the table header of the database to obtain the field names contained in the database.

[0054] A mapping module maps the field name with a preset information tag in a desensitizing tag library to obtain a data asset directory; wherein the desensitizing tag library contains a plurality of information tags, and each information tag corresponds to a desensitizing rule library.

[0055] The judgment module, when the medical data is called by the system, obtains the corresponding information tag of the field name corresponding to the field data through the data asset directory according to the field name corresponding to each field data in the called medical data, and judges whether the field data corresponding to each information tag needs to be desensitized according to the preset method.

[0056] In one embodiment of the present application, information tags are classified into strong tags and weak tags, and a desensitized weak tag combination library is set, wherein the desensitized weak tag combination library includes several weak tag combinations, and each combination includes several weak tags.

[0057] Specifically, the preset method includes: Classify the information labels corresponding to the field names of the data fields in the called medical consultation data.

[0058] For informational tags classified as strong tags: It is determined that the field data corresponding to the information tag needs to be desensitized.

[0059] For information labels classified as weak labels: It is determined whether there is a weak label combination in the desensitized weak label combination library among all the information labels classified as weak labels. If so, it is determined that each field data corresponding to the label combination needs to be desensitized.

[0060] If it does not exist, the field data corresponding to the information label classified as a weak label will not be desensitized.

[0061] Desensitizing module: when the field data corresponding to the information tag needs to be desensitized, the desensitizing module selects a desensitizing rule from the desensitizing rule library corresponding to the information tag according to preset rules, desensitizes the field data corresponding to the information tag according to the desensitizing rule, and returns it to the system.

[0062] An incremental scanning module performs incremental scanning on the database according to a preset time threshold, matches the field names of the newly added tables in the database with preset information tags, and stores the mapping relationship between the field names and the matched information tags in the data asset directory.

[0063] An update module is used to add the newly added field name as a new information tag to the desensitizing tag library when the field name of the newly added table does not match the preset information tag, and set the corresponding information tag classification and desensitizing rule library for the new information tag; when the new information tag is classified as a weak tag, add the corresponding weak tag combination to the desensitizing weak tag combination library.

[0064] The above is a system embodiment proposed in this application. Based on the same inventive concept, this application embodiment also provides a device, whose structure is as follows Figure 5 shown.

[0065] Figure 5 A schematic diagram of the internal structure of a privacy data desensitization device based on medical data provided in an embodiment of the present application. Figure 5 As shown, the device includes: at least one processor 201; and, a memory 202 communicatively connected to the at least one processor; The memory 202 stores instructions that can be executed by at least one processor, and the instructions are executed by at least one processor 201 to enable at least one processor 201 to: Any step of the above method can be implemented.

[0066] Each embodiment in this application is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the IoT device and medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0067] The system and device provided in the embodiments of the present application correspond one-to-one to the method. Therefore, the system and device also have similar beneficial technical effects to the corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the system and device will not be repeated here.

[0068] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0069] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0070] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0071] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0072] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0073] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0074] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules 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 technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0075] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0076] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A privacy data desensitization method based on medical data, characterized in that: The method comprises: Scan the entire database header to obtain the field names contained in the database; Mapping the field name with the information tag preset in the desensitizing tag library to obtain a data asset directory; wherein the desensitizing tag library contains a plurality of information tags, and each information tag corresponds to a desensitizing rule library; When the medical data is called by the system, according to the field name corresponding to each field data in the called medical data, the corresponding information tag of the field name corresponding to the field data is obtained through the data asset directory, and it is determined according to the preset method whether the field data corresponding to each information tag needs to be desensitized; When the field data corresponding to the information tag needs to be desensitized, a desensitization rule is selected from the desensitization rule library corresponding to the information tag according to the preset rule, and the field data corresponding to the information tag is desensitized according to the desensitization rule and then returned to the system.

2. According to claim 1, a privacy data desensitization method based on medical data is characterized in that: Map the field names with preset information tags to obtain a data asset directory, which specifically includes: Through the NLP algorithm, the field names are scanned and identified to obtain the corresponding information; Matching the obtained information with the preset information tags according to semantics, and forming a mapping relationship between the field name and the matched information tag; The mapping relationship between the field name and the information tag is saved as a data asset directory.

3. A privacy data desensitization method based on medical data according to claim 1, characterized in that: Information labels are classified into strong labels and weak labels, and a desensitized weak label combination library is set, wherein the desensitized weak label combination library includes several weak label combinations, and each combination includes several weak labels.

4. A privacy data desensitization method based on medical data according to claim 3, characterized in that: The preset method specifically includes: Classify the information labels corresponding to the field names of the data fields in the called medical consultation data; For informational tags classified as strong tags: Determine whether the field data corresponding to the information tag needs to be desensitized; For information labels classified as weak labels: Determine whether there is a weak label combination in the desensitized weak label combination library among all information labels classified as weak labels. If so, determine whether desensitization processing needs to be performed on each field data corresponding to the label combination; If it does not exist, the field data corresponding to the information label classified as a weak label will not be desensitized.

5. The method for desensitizing privacy data based on medical data according to claim 1, characterized in that: Each desensitization rule base contains several desensitization rules of different levels for the information tag; The preset rules specifically include: selecting a desensitization rule of a corresponding level from a desensitization rule library of the information tag according to the authority of the caller.

6. The method for desensitizing privacy data based on medical data according to claim 1, characterized in that: Also includes: The database is incrementally scanned according to a preset time threshold, the field names of the newly added tables in the database are matched with preset information tags, and the mapping relationship between the field names and the matched information tags is stored in the data asset directory.

7. A privacy data desensitization method based on medical data according to claim 3 or claim 7, characterized in that: Also includes: When the field name of the newly added table does not match the preset information tag, the newly added field name is added as a new information tag to the desensitization tag library, and the corresponding information tag classification and desensitization rule library are set for the new information tag; When the new information label is classified as a weak label, the corresponding weak label combination is added to the desensitized weak label combination library.

8. A privacy data desensitization system based on medical data, used to implement the method described in any one of claims 1 to 7, characterized in that: The system comprises: A scanning module, wherein the scanning module performs a full scan on the table header of the database to obtain the field names contained in the database; A mapping module, wherein the mapping module maps the field name with an information tag preset in a desensitizing tag library to obtain a data asset directory; wherein the desensitizing tag library contains a plurality of information tags, and each information tag corresponds to a desensitizing rule library; A judgment module, when the medical data is called by the system, the judgment module obtains the corresponding information tag of the field name corresponding to the field data through the data asset directory according to the field name corresponding to each field data in the called medical data, and judges whether the field data corresponding to each information tag needs to be desensitized according to a preset method; A desensitization module, when the field data corresponding to the information tag needs to be desensitized, selects a desensitization rule from a desensitization rule library corresponding to the information tag according to a preset rule, desensitizes the field data corresponding to the information tag according to the desensitization rule, and returns it to the system; An incremental scanning module performs incremental scanning on the database according to a preset time threshold, matches the field names of the newly added tables in the database with preset information tags, and stores the mapping relationship between the field names and the matched information tags in the data asset directory. An update module is used to add the newly added field name as a new information tag to the desensitizing tag library when the field name of the newly added table does not match the preset information tag, and set the corresponding information tag classification and desensitizing rule library for the new information tag; when the new information tag is classified as a weak tag, add the corresponding weak tag combination to the desensitizing weak tag combination library.

9. A privacy data desensitization system based on medical data according to claim 8, characterized in that: The information labels are classified into strong labels and weak labels, and a desensitized weak label combination library is set, wherein the desensitized weak label combination library includes a plurality of weak label combinations, and each combination includes a plurality of weak labels; The preset method specifically includes: Classify the information labels corresponding to the field names of the data fields in the called medical consultation data; For informational tags classified as strong tags: Determine whether the field data corresponding to the information tag needs to be desensitized; For information labels classified as weak labels: Determine whether there is a weak label combination in the desensitized weak label combination library among all information labels classified as weak labels. If so, determine whether desensitization processing needs to be performed on each field data corresponding to the label combination; If it does not exist, the field data corresponding to the information label classified as a weak label will not be desensitized.

10. A privacy data desensitization device based on medical data, characterized in that: The device comprises: at least one processor; and, a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: Implement the steps of the method according to any one of claims 1 to 7.

Images