Block chain-based physical examination data management method and system

Abstract

The invention relates to the technical field of data management, in particular to a block chain-based physical examination data management method and system, and the method comprises the following steps: analyzing the dates, events, gender and age of a plurality of physical examination data on a block chain based on the physical examination data of a user on the chain, classifying the physical examination data, adding an index tag, and storing the classified physical examination data in a database; and generating a physical examination data index set. According to the invention, by classifying the physical examination data and adding the index tags, the organization and retrieval efficiency of the data is improved, the intention of a querier is identified, the query processing response speed is improved, behavior pattern identification and parameter comparison are carried out on the access behaviors of a plurality of visitors, and the security and the abnormal behavior detection capability of the system are improved; according to the method, access authority is matched for various visitors, the security of access management is improved, data privacy is protected, the compliance of data processing is improved, physical examination data is classified and archived and uploaded to a block chain, data storage and management are optimized, and the utilization efficiency and traceability of the data are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to a method and system for managing health examination data based on a blockchain. Background Art

[0002] The technical field of data management aims to effectively store, retrieve, update, and protect data. By organizing, storing, managing, and protecting data, it ensures the quality, availability, and security of data throughout its life cycle, including data storage design, data security, data integration, data migration, data access optimization, using database management systems, data warehouses, and data lakes to process and maintain data. By controlling data access and implementing data protection procedures, it improves data compliance and privacy protection, optimizes data quality and access efficiency, helps enterprises and organizations reduce costs, supports business decisions, and enhances competitiveness.

[0003] Among them, the method for managing health examination data based on a blockchain aims to use blockchain technology to improve the storage, access, and sharing of health examination data. By the decentralized characteristics of the blockchain, it ensures the immutability and transparency of data, enhances data security and privacy protection, uses the blockchain to encrypt and distribute the storage of health examination data, avoids the risks of single-point failures and data leaks, and by creating a trusted data access mechanism, optimizes the secure access and sharing of health examination information by authorized users and institutions, improving the efficiency and reliability of health data processing.

[0004] Traditional health examination data management technologies have insufficient data storage and retrieval efficiency capabilities, lack of detailed classification and effective indexing of health examination data, resulting in slow speeds when retrieving target information in batch data. In the face of complex access control requirements, permission allocation is not flexible enough, unable to effectively handle different user roles and access scenarios, leading to data leakage risks. When processing health data, there is a lack of analysis of access behaviors and detection of abnormal behaviors, unable to timely detect and prevent security threats, increasing the risk of the system being attacked and data being tampered with. The measures for data desensitization and privacy protection are not perfect enough, and when sharing sensitive health data, it cannot fully protect user privacy. Summary of the Invention

[0005] The purpose of the present invention is to solve the drawbacks existing in the prior art, and to propose a method and system for managing health examination data based on a blockchain.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions. A method for managing health examination data based on a blockchain includes the following steps:

[0007] S1: Analyze the dates, events, genders, and ages of multiple physical examination data on the blockchain based on the on-chain user physical examination data, classify the physical examination data, add index tags, and generate a physical examination data index set;

[0008] S2: Based on the physical examination data index set, analyze the query keywords in combination with the physical examination data query records, identify the queryer's intention, and match the query keywords with the physical examination data to obtain a data fuzzy query tag;

[0009] S3: Based on the data fuzzy query tag, analyze the visitor's login time, data access frequency, and access type, analyze the access behaviors of multiple visitors, and generate an abnormal access behavior record through behavior pattern recognition and parameter comparison;

[0010] S4: Based on the abnormal access behavior record, analyze the visitor's user role, behavior record, and access log, match access permissions for multiple visitors, and generate access permission allocation parameters;

[0011] S5: Based on the access permission allocation parameters, analyze the privacy information in the physical examination data and perform desensitization processing, set access permission rules according to the privacy protection level, and generate privacy-protected physical examination data;

[0012] S6: Based on the privacy-protected physical examination data, analyze the physical examination items and physical examination values in the physical examination results, archive the physical examination data, and upload it to the blockchain to generate a physical examination information classification record.

[0013] As a further solution of the present invention, the physical examination data index set includes an examinee information index, a physical examination date index, and a physical examination event index. The data fuzzy query tag includes a keyword matching tag, a query intention analysis tag, and an associated physical examination parameter tag. The abnormal access behavior record includes an abnormal access time record, a frequent access record, and an unauthorized access record. The access permission allocation parameters include a doctor access level, a nurse access level, and an administrative staff access level. The privacy-protected physical examination data includes identity information desensitization data, access restriction data, and encryption measure data. The physical examination information classification record includes a routine examination classification result, a special pathological analysis classification result, and an emergency event classification information.

[0014] As a further solution of the present invention, the steps of analyzing the dates, events, genders, and ages of multiple physical examination data on the blockchain based on the on-chain user physical examination data, classifying the physical examination data, and adding index tags to generate a physical examination data index set are specifically as follows:

[0015] S101: Based on the on-chain user physical examination data, analyze the dates and events of multiple physical examination data, analyze the occurrence time of the event and the recording date of the physical examination data, classify the physical examination data, and generate date event classification information;

[0016] S102: Based on the date event classification information, by analyzing the gender and age of the examinees in multiple physical examination data, classify the physical examination data by gender and age to generate gender-age classification data;

[0017] S103: Based on the gender-age classification data, match index tags for the physical examination data, and generate a physical examination data index set by recording classification and tag information.

[0018] As a further solution of the present invention, the steps of analyzing query keywords based on the physical examination data index set, combining with the physical examination data query record, identifying the queryer's intention and matching the query keywords with the physical examination data to obtain a data fuzzy query tag are specifically as follows:

[0019] S201: Based on the physical examination data index set, combine with the physical examination data query record, analyze and identify query keywords, including the occurrence frequency and relevance of query keywords, to generate a query keyword analysis result;

[0020] S202: Based on the query keyword analysis result, analyze the query statement and keywords, identify the query intention through the relevance and frequency of keywords, and generate query intention identification information;

[0021] S203: Based on the query intention identification information, by matching the query keywords with the physical examination data, construct the association information between the query keywords and the query intention to obtain a data fuzzy query tag.

[0022] As a further solution of the present invention, the steps of analyzing the visitor's login time, data access frequency, and access type based on the data fuzzy query tag, analyzing the access behaviors of multiple visitors, and generating an abnormal access behavior record through behavior pattern recognition and parameter comparison are specifically as follows:

[0023] S301: Based on the data fuzzy query tag, analyze the visitor's login time, data access frequency, and access type in the access record to generate an access log analysis result;

[0024] S302: Based on the access log analysis result, by analyzing the access behaviors of multiple visitors, identify multiple access patterns and mark abnormal access behaviors to generate an access pattern analysis result;

[0025] S303: Based on the access pattern analysis result, identify the abnormal behaviors of multiple visitors by analyzing the behavior patterns, compare the historical access parameters, and use the support vector machine algorithm to analyze the persistence and consistency of the abnormal behaviors to generate an abnormal access behavior record.

[0026] As a further solution of the present invention, the support vector machine algorithm, according to the formula:

[0027] y = sign(w1x1 + w2x2 + w3x3 + w4x4 + b)

[0028] identifies abnormal behaviors, where y is the identification result, w1 is the weight coefficient of the access time interval, x1 is the access time interval, w2 is the weight coefficient of the data access volume, x2 is the data access volume, w3 is the weight coefficient of the access source IP, x3 is the access source IP, w4 is the weight coefficient of the access device type, x4 is the access device type, b is the bias term, and sign is the sign function.

[0029] As a further solution of the present invention, based on the abnormal access behavior records, analyzing the visitor user roles, behavior records, and access logs, and matching access permissions for multiple visitors, the steps of generating access permission allocation parameters are specifically as follows:

[0030] S401: Based on the abnormal access behavior records, analyze the user roles and behavior information in the abnormal behaviors, identify the types of visitors, including doctors, nurses, and administrative staff, and generate role behavior analysis data;

[0031] S402: Based on the role behavior analysis data, combined with the matching access needs and security requirements of the visitor roles, set multiple levels of access permissions, match the data access needs of multiple roles, and obtain role permission matching information;

[0032] S403: Based on the role permission matching information, combined with the behavior patterns and access requirements of multiple visitor roles, construct access permission rules at multiple levels, including data types, access times, and editing permissions, and generate access permission allocation parameters.

[0033] As a further solution of the present invention, based on the access permission allocation parameters, analyze the privacy information in the physical examination data and perform desensitization processing, and set access permission rules according to the privacy protection level, the steps of generating privacy protection physical examination data are specifically as follows:

[0034] S501: Based on the access permission allocation parameters, analyze and identify the privacy information in the physical examination data, including the name, address, and phone number of the examinee, and generate a privacy information identification result;

[0035] S502: Based on the privacy information identification result, perform desensitization processing on the physical examination privacy information, and generate a privacy desensitization processing result by replacing the privacy information with an anonymous identifier;

[0036] S503: Based on the privacy desensitization processing result, set access permission rules according to the privacy protection level, and combine with the privacy protection level information in the physical examination data to generate privacy-protected physical examination data.

[0037] As a further solution of the present invention, based on the privacy-protected physical examination data, analyze the physical examination items and physical examination values in the physical examination results, file the physical examination data and upload it to the blockchain, and the steps of generating the classified record of physical examination information are specifically as follows:

[0038] S601: Based on the privacy-protected physical examination data, analyze the physical examination items in the health physical examination data, identify the physical examination item names and classify them according to the item categories to generate physical examination item classification data;

[0039] S602: Based on the physical examination item classification data, analyze the physical examination values in the physical examination results, classify and record the physical examination values according to the categories to generate physical examination value differentiation information;

[0040] S603: Based on the physical examination value differentiation information, establish multiple types of archive units for the physical examination data according to the physical examination items, values and types, and number and classify the archive units to generate the classified record of physical examination information.

[0041] A health physical examination data management system based on blockchain, the health physical examination data management system based on blockchain is used to execute the above-mentioned health physical examination data management method based on blockchain, and the system includes:

[0042] The information processing and classification module classifies the date, event, gender and age information in the data based on the on-chain user physical examination data, matches index tags for various information, and generates data query index information;

[0043] The query word analysis module uses the data query index information to analyze the physical examination data query record, identifies the search intention of the querier through judgment and selection logic, matches the query keywords with the physical examination data, and generates a query intention recognition result;

[0044] The access log analysis module analyzes and records the login time, data access frequency and access type of the visitor based on the query intention recognition result, identifies abnormal access behaviors, and generates visitor record analysis data;

[0045] The user permission configuration module uses the visitor record analysis data to analyze the user roles and behavior records of various visitors, matches access permissions for various visitors, and generates a role access level matching result;

[0046] Based on the role access level matching result, the data archiving and processing module analyzes and identifies the privacy information in the physical examination data, performs desensitization processing on the privacy information, classifies and archives the physical examination results according to the type of physical examination items, and generates a classified record of physical examination information.

[0047] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0048] In the present invention, by classifying the physical examination data and adding index tags, the organization and retrieval efficiency of the data are improved. By combining the identification of query keywords and the analysis of the queryer's intention, the query keywords are matched with the physical examination data, enhancing the accuracy and response speed of query processing. By analyzing the login time, data access frequency, and access type of visitors, the access behaviors of multiple visitors are recognized in terms of behavior patterns and parameter comparison, enhancing the system's security and abnormal behavior detection capabilities. By matching access permissions for various visitors, generating access permission allocation parameters, providing more secure permission control and access management, setting access permission rules according to the privacy protection level, generating privacy-protected physical examination data, protecting data privacy and improving the compliance of data processing. By archiving the physical examination data according to the physical examination items, values, and types and uploading them to the blockchain, the data storage and management are optimized, improving the utilization efficiency and traceability of the data, and enhancing the efficiency, security, and privacy protection capabilities of physical examination data management. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 is a schematic diagram of the working process of the present invention;

[0050] Figure 2 is a detailed flowchart of S1 of the present invention;

[0051] Figure 3 is a detailed flowchart of S2 of the present invention;

[0052] Figure 4 is a detailed flowchart of S3 of the present invention;

[0053] Figure 5 is a detailed flowchart of S4 of the present invention;

[0054] Figure 6 is a detailed flowchart of S5 of the present invention;

[0055] Figure 7 is a detailed flowchart of S6 of the present invention;

[0056] Figure 8 is a system flowchart of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0057] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, 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 used to explain the present invention and are not used to limit the present invention.

[0058] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention. In addition, in the description of the present invention, the meaning of "a plurality of" is two or more unless otherwise specifically defined.

[0059] Embodiment 1

[0060] Please refer to Figure 1 , the present invention provides a technical solution, a method for managing health examination data based on blockchain, including the following steps:

[0061] S1: Based on the on-chain user health examination data, analyze the date, event, gender and age of multiple health examination data on the blockchain, classify the health examination data and add index tags to generate a health examination data index set;

[0062] S2: Based on the health examination data index set, analyze the query keywords in combination with the health examination data query records, identify the queryer's intention and match the query keywords with the health examination data to obtain a data fuzzy query tag;

[0063] S3: Based on the data fuzzy query tag, analyze the visitor's login time, data access frequency, and access type, analyze the access behaviors of multiple visitors, and generate abnormal access behavior records through behavior pattern recognition and parameter comparison;

[0064] S4: Based on the abnormal access behavior records, analyze the visitor's user role, behavior records and access logs, match access permissions for multiple visitors, and generate access permission allocation parameters;

[0065] S5: Based on the access permission allocation parameters, analyze the privacy information in the health examination data and perform desensitization processing, set access permission rules according to the privacy protection level, and generate privacy-protected health examination data;

[0066] S6: Based on the privacy-protected health examination data, analyze the examination items and examination values in the examination results, archive the health examination data and upload it to the blockchain to generate a health examination information classification record.

[0067] The physical examination data index set includes the index of examinee information, the index of physical examination date, and the index of physical examination events. The data fuzzy query tags include the keyword matching tag, the query intention analysis tag, and the associated physical examination parameter tag. The abnormal access behavior records include the abnormal access time record, the frequent access record, and the unauthorized access record. The access permission allocation parameters include the doctor access level, the nurse access level, and the administrative staff access level. The privacy-protected physical examination data includes the desensitized data of identity information, the access restriction data, and the encryption measure data. The physical examination information classification record includes the classification results of routine examinations, the classification results of special pathological analyses, and the classification information of emergency events.

[0068] Please refer to Figure 2 , based on the on-chain physical examination data of users, analyze the dates, events, genders, and ages of multiple physical examination data on the blockchain, classify the physical examination data, and add index tags. The specific steps for generating the physical examination data index set are as follows:

[0069] S101: Based on the on-chain physical examination data of users, analyze the dates and events of multiple physical examination data, analyze the occurrence time of events and the recording dates of physical examination data, classify the physical examination data, and the specific process for generating the date-event classification information is as follows;

[0070] In the sub-step S101, based on the on-chain physical examination data of users, analyze the dates and events of multiple physical examination data, collect the dates, times, and specific events of physical examinations in the physical examination records of each user, including blood pressure measurement and blood glucose detection, calculate the occurrence frequency and time interval of each event, create time series data including all events, set a time window to normalize the data, use the hierarchical clustering method to classify the events, establish classification criteria by calculating the occurrence frequency and interval of events, generate the average time interval and standard deviation of each type of event, and set the classification threshold. The formula is where C i is the average time interval of the i-th type of event, i is the event classification number, n is the total number of events, T ij is the time interval of the j-th event in the i-th category, j is the number of the event in the classification, classify the physical examination data, and generate the date-event classification information.

[0071] S102: Based on the date-event classification information, by analyzing the genders and ages of examinees in multiple physical examination data, classify the physical examination data by gender and age. The specific process for generating the gender-age classification data is as follows;

[0072] In sub-step S102, based on the date event classification information, by analyzing the gender and age of the examinees in multiple physical examination data, the basic information of the examinees, including gender, date of birth and age, is extracted. The examinees are divided into two groups: male and female according to gender, and then each group is further subdivided by age group, including 18 - 30 years old, 31 - 45 years old, 46 - 60 years old and over 60 years old. The distribution of physical examination data for each age group is calculated, and the chi-square test method is used to analyze the correlation between gender, age and physical examination data. By analyzing the frequency of physical examination events in each age group and gender group, a classification standard is established. The formula is where P(A|B) is the probability of event A occurring when event B occurs, A is the target physical examination event, B is the gender and age classification of the examinee, P(A∩B) is the probability of events A and B occurring simultaneously, and P(B) is the probability of event B occurring. The physical examination data is classified by gender and age to generate gender-age classification data.

[0073] S103: Based on the gender-age classification data, the process of matching index tags for the physical examination data and generating the physical examination data index set by recording classification and tag information is as follows;

[0074] In sub-step S103, based on the gender-age classification data, index tags are matched for the physical examination data. A unique index tag is assigned to each gender and age group, including M18 - 30 representing the male group aged 18 - 30, F31 - 45 representing the female group aged 31 - 45. The data for each group is numbered and marked. By establishing a data index table, the classification and corresponding tag information of each physical examination data are recorded, and the occurrence frequency of each type of tag is calculated. The weighted average method is used to calculate the weight of the physical examination data. The formula is where, W i is the weight of the i-th type of data, i is the data classification number, w j is the weight coefficient of the j-th tag, j is the tag number, f ij is the frequency of the j-th tag in the i-th type of data, and m is the total number of tags, generating the physical examination data index set.

[0075] Please refer to Figure 3 , based on the physical examination data index set, combined with the physical examination data query record, the steps of analyzing the query keyword, identifying the query intention and matching the query keyword with the physical examination data to obtain the data fuzzy query tag are as follows:

[0076] S201: Based on the physical examination data index set, combined with the physical examination data query record, analyze and identify the query keyword, including the occurrence frequency and relevance of the query keyword, and the process of generating the query keyword analysis result is as follows;

[0077] In sub-step S201, based on the physical examination data index set, combined with the physical examination data query records, collect the query records, extract the keywords in multiple queries, count the occurrence frequencies of multiple keywords, calculate the relevance of multiple keywords, and use the co-occurrence matrix method to quantify the relevance between keywords. The formula is where R ij is the relevance between keyword i and keyword j, i is the number of the first keyword, j is the number of the second keyword, and C ij is the co-occurrence times of keyword i and keyword j, C i is the total occurrence times of keyword i, C h is the total occurrence times of keyword j. Analyze and identify the query keywords to generate the analysis result of the query keywords.

[0078] S202: Based on the analysis result of the query keywords, analyze the query statement and keywords, and identify the query intention through the relevance and frequency of the keywords. The specific process of generating the query intention recognition information is as follows;

[0079] In sub-step S202, based on the analysis result of the query keywords, analyze the query statement and keywords, extract multiple keywords in the query statement, use the keyword relevance and frequency to construct the relevance matrix of multiple keywords, and adopt the term frequency-inverse document frequency method to evaluate the importance of multiple keywords. The formula is where TFIDF i is the importance of keyword i, i is the number of the target keyword, TF i is the term frequency of keyword i in the query, N is the total number of query records, and DF i is the number of query records including keyword i. Identify the query intention through the relevance and frequency of the keywords to generate the query intention recognition information.

[0080] S203: Based on the query intention recognition information, by matching the query keywords with the physical examination data, construct the association information between the query keywords and the query intention to obtain the process of the data fuzzy query label as follows;

[0081] In sub-step S203, based on the query intention recognition information, by matching the query keywords with the physical examination data, extract multiple keywords in the physical examination data index set, calculate the matching degree between multiple query keywords and the physical examination data keywords. The formula is where M ij is the matching degree between query keyword i and physical examination data keyword j, i is the number of the query keyword, j is the number of the physical examination data keyword, Q i is the word vector of query keyword i, D j is the word vector of physical examination data keyword j, |Q i| is the vector length of the query keyword i, | D j | is the vector length of the physical examination data keyword j, construct the association information between the query keyword and the query intent, and obtain the data fuzzy query label.

[0082] Please refer to Figure 4 , based on the data fuzzy query label, analyze the visitor login time, data access frequency, and access type, and the steps for analyzing the access behaviors of multiple visitors and generating abnormal access behavior records through behavior pattern recognition and parameter comparison are as follows:

[0083] S301: Based on the data fuzzy query label, analyze the visitor login time, data access frequency, and access type in the access record, and the process of generating the access log analysis result is as follows;

[0084] In the S301 sub-step, based on the data fuzzy query label, analyze the visitor login time, data access frequency, and access type in the access record, collect the login time data of multiple visitors, calculate the number of logins and login duration of multiple visitors per day, record the data access frequency, including the number of accesses per day and the duration of each access, classify the access type, including browsing, downloading, and uploading, count the access frequency of each type, and use the time series analysis method to analyze the distribution of the login time. The formula is Among them, F i is the access frequency of the i-th access type, i is the access type number, V i is the number of accesses of the i-th access type, T i is the total access time, and generate the access log analysis result.

[0085] S302: Based on the access log analysis result, identify multiple access patterns and mark abnormal access behaviors by analyzing the access behaviors of multiple visitors. The process of generating the access pattern analysis result is as follows;

[0086] In the S302 sub-step, based on the access log analysis result, by analyzing the access behaviors of multiple visitors, classify the access behavior records by visitor, calculate the average access frequency and access duration of multiple visitors, use the clustering analysis method to identify different access patterns, and count the occurrence frequency of each access pattern. The formula is Among them, M k is the average behavior value of the k-th access pattern, k is the access pattern number, n k is the number of visitors of the k-th access pattern, B ik is the behavior record of the i-th visitor in the k-th access pattern, mark the abnormal access behavior, and generate the access pattern analysis result.

[0087] S303: Based on the analysis results of the access pattern, by analyzing the behavior pattern to identify the abnormal behaviors of multiple visitors, comparing the historical access parameters, and using the support vector machine algorithm to analyze the persistence and consistency of the abnormal behaviors, the process of generating the abnormal access behavior record is specifically as follows;

[0088] In sub-step S303, based on the analysis results of the access pattern, by analyzing the behavior pattern to identify the abnormal behaviors of multiple visitors, extracting the behavior patterns of multiple visitors, calculating the occurrence frequency and duration of the abnormal behaviors, comparing the historical access parameters, analyzing the persistence and consistency of the abnormal behaviors, and generating the abnormal access behavior record.

[0089] The support vector machine algorithm, according to the formula:

[0090] y = sign(w1x1 + w2x2 + w3x3 + w4x4 + b)

[0091] Identifying the abnormal access behavior, where y is the identification result, 1 represents normal behavior, -1 represents abnormal behavior, w1 is the weight coefficient of the access time interval, x1 is the access time interval, w2 is the weight coefficient of the data access volume, x2 is the data access volume, w3 is the weight coefficient of the access source IP, x3 is the access source IP, w4 is the weight coefficient of the access device type, x4 is the access device type, b is the bias term, and sign is the sign function used to determine the sign of the result.

[0092] The specific execution process of the formula is as follows:

[0093] Collect the access time interval, data access volume, access source IP, and access device type data, initialize the weights w1, w2, w3, and w4 to random values, use the training data for iterative optimization, calculate the loss function of each iteration, update the weights through the gradient descent method until the loss function converges, set the initial bias term b to 0, combine the training data and the optimized weights, adjust the bias term b, substitute the normalized parameters for calculation, identify the abnormal behavior, and analyze the persistence and consistency of the abnormal behavior.

[0094] Please refer to Figure 5 , based on the abnormal access behavior record, analyzing the user roles, behavior records, and access logs of the visitors, and the steps of matching access permissions for multiple visitors and generating access permission allocation parameters are specifically as follows:

[0095] S401: Based on the abnormal access behavior record, analyzing the user roles and behavior information in the abnormal behavior, identifying the types of visitors, including doctors, nurses, and administrative staff, and the process of generating the role behavior analysis data is specifically as follows;

[0096] In sub-step S401, based on the abnormal access behavior records, analyze the user roles and behavior information in the abnormal behavior, collect the user role data in multiple abnormal access records, count the abnormal access frequencies of multiple roles, calculate the distribution of abnormal behavior types of multiple roles, and the formula is where B i is the behavior abnormality rate of the i-th type of role, i is the user role number, and A ij is the number of times of the j-th abnormal behavior of the i-th type of role, and U i is the total number of users of the i-th type of role. Identify the types of visitors, including doctors, nurses, and administrative staff, and generate role behavior analysis data.

[0097] S402: Based on the role behavior analysis data, combined with the matching access needs and security requirements of the visitor role, set multiple levels of access permissions, and the process of matching the data access needs of multiple roles to obtain role permission matching information is as follows;

[0098] In sub-step S402, based on the role behavior analysis data, combined with the matching access needs and security requirements of the visitor role, collect and analyze the access behavior logs of multiple roles in different time and scenarios. Through statistical analysis, identify the access frequencies, access types, and access objects of multiple roles, and use the role requirement matching formula to set permissions for each role. The formula is where R i is the access permission level of role i, w j is the security weight, x ij is the access requirement score of role i for resource j, j is the resource index, numbered from 1 to n, and calculate the role permission matching information.

[0099] S403: Based on the role permission matching information, combined with the behavior patterns and access needs of multiple visitor roles, construct multiple levels of access permission rules, including data types, access times, and editing permissions. The process of generating access permission allocation parameters is as follows;

[0100] In sub-step S403, based on the role permission matching information, combined with the behavior patterns and access needs of multiple visitor roles, analyze the data processing needs and access time periods of multiple roles, formulate corresponding access permission rules, including data types, access times, and editing permissions, and use the access permission rule formula to classify and manage each type of data. The formula is where P k represents the access permission level of data type k, v m represents the access weight of time period m, y mk represents the access requirement level for data type k within time period m, m is the time period index, numbered from 1 to p, and k is the data type index, and generate access permission allocation parameters.

[0101] Please refer to Figure 6 , based on the access permission allocation parameters, analyze the privacy information in the physical examination data and perform desensitization processing. Set the access permission rules according to the privacy protection level. The specific steps for generating the privacy-protected physical examination data are as follows:

[0102] S501: Based on the access permission allocation parameters, analyze and identify the privacy information in the physical examination data, including the name, address, and phone number of the examinee. The specific process for generating the privacy information identification result is as follows;

[0103] In the sub-step of S501, based on the access permission allocation parameters, analyze and identify the privacy information in the physical examination data, extract multiple sensitive fields in the physical examination data, including the name, address, and phone number, perform pattern matching and keyword identification, count the occurrence frequency of each type of privacy information, and calculate the distribution of the target field in the differential dataset. The formula is where R i is the recognition rate of the i-th type of privacy information, i is the privacy information type number, I ih is the number of occurrences of the i-th type of privacy information in the j-th dataset, D i is the total number of datasets including the i-th type of privacy information. Analyze and identify the privacy information in the physical examination data to generate the privacy information identification result.

[0104] S502: Based on the privacy information identification result, perform desensitization processing on the physical examination privacy information. The specific process for generating the privacy desensitization processing result is as follows;

[0105] In the sub-step of S502, based on the privacy information identification result, perform desensitization processing on the physical examination privacy information, mark the identified privacy information fields, select the matching anonymous identifier, and replace multiple fields, including replacing the name with a unique code, replacing the address with a blurred geographical information, and replacing the phone number with a randomly generated number segment. Calculate the desensitization coverage rate. The formula is where C i is the desensitization coverage rate of the i-th type of privacy information, i is the privacy information type number, S ij is the number of desensitization times of the i-th type of privacy information in the j-th dataset, P i is the total number of occurrences of the i-th type of privacy information. Generate the privacy desensitization processing result by replacing the privacy information with an anonymous identifier.

[0106] S503: Based on the privacy desensitization processing result, set the access permission rules according to the privacy protection level, and combine the privacy protection level information in the physical examination data. The specific process for generating the privacy-protected physical examination data is as follows;

[0107] In sub-step S503, based on the privacy desensitization processing result, access permission rules are set according to the privacy protection level. A privacy protection level is assigned to each type of privacy information, including high, medium, and low. Corresponding access permissions are set according to the differentiation level, restricting access permissions to information with a high protection level and allowing access to information with medium and low protection levels under certain conditions. The access control ratios for multiple levels are calculated. The formula is where P i is the access control ratio of the i-th type of privacy information, i is the privacy information type number, A ij is the number of accesses of the i-th type of privacy information in the j-th dataset, G i is the total protection level score of the i-th type of privacy information. Combining the privacy protection level information in the physical examination data, privacy protection physical examination data is generated.

[0108] Please refer to Figure 7 , based on the privacy protection physical examination data, analyze the physical examination items and physical examination values in the physical examination results, file the physical examination data and upload it to the blockchain. The steps for generating the physical examination information classification record are specifically as follows:

[0109] S601: Based on the privacy protection physical examination data, analyze the physical examination items in the health physical examination data, identify the physical examination item names and classify them according to the item categories. The process for generating the physical examination item classification data is specifically as follows;

[0110] In sub-step S601, based on the privacy protection physical examination data, analyze the physical examination items in the health physical examination data, extract the item names from multiple physical examination records, use keyword matching and natural language processing methods to identify the names and categories of multiple items, and count the occurrence frequencies of multiple items. The formula is where C i is the occurrence frequency of the i-th type of item, i is the item number, N i is the total occurrence times of the i-th type of item, T is the total number of items. Classify according to the item categories to generate the physical examination item classification data.

[0111] S602: Based on the physical examination item classification data, analyze the physical examination values in the physical examination results, classify and record the physical examination values according to the categories. The process for generating the physical examination value differentiation information is specifically as follows;

[0112] In sub-step S602, based on the physical examination item classification data, analyze the physical examination values in the physical examination results, extract the numerical data corresponding to multiple physical examination items, calculate the distribution of multiple values in the differentiation categories, and use statistical analysis methods to calculate the average value and standard deviation of the values. The formula is where M i is the average physical examination value of the i-th type of item, i is the item number, n i is the number of physical examination records of the i-th type of item, Vij is the j-th physical examination value of the i-th type of project. Classify and record the physical examination values according to the category to generate physical examination value differentiation information.

[0113] S603: Based on the physical examination value differentiation information, establish multiple types of archiving units for the physical examination data according to the physical examination items, values, and types, and number and classify the archiving units to generate the process of classifying and recording the physical examination information specifically as follows;

[0114] In the sub-step of S603, based on the physical examination value differentiation information, establish multiple types of archiving units for the physical examination data according to the physical examination items, values, and types, subdivide the numerical data of multiple physical examination items by type, set the numbering rules and classification criteria for the archiving units, and calculate the storage capacity and data volume of multiple archiving units. The formula is where, S i is the storage capacity of the i-th type of archiving unit, i is the archiving unit number, D ij is the j-th data volume of the i-th type of archiving unit, C i is the total number of classifications of the i-th type of archiving unit, and generate the classified record of the physical examination information.

[0115] Please refer to Figure 8 , a blockchain-based health examination data management system. The blockchain-based health examination data management system is used to execute the above-mentioned blockchain-based health examination data management method. The system includes:

[0116] The information processing and classification module classifies and processes the date, event, gender, and age information in the data based on the on-chain user physical examination data, matches index tags for various information, and generates data query index information;

[0117] The query word analysis module uses the data query index information to analyze the physical examination data query records, identifies the search intention of the queryer through judgment and selection logic, and matches the query keywords with the physical examination data to generate a query intention recognition result;

[0118] The access log analysis module analyzes and records the login time, data access frequency, and access type of the visitor based on the query intention recognition result, identifies abnormal access behaviors, and generates visitor record analysis data;

[0119] The user permission configuration module uses the visitor record analysis data to analyze the user roles and behavior records of various visitors, matches access permissions for various visitors, and generates a role access level matching result;

[0120] The data archiving and processing module analyzes and identifies the privacy information in the physical examination data based on the role access level matching result, performs desensitization processing on the privacy information, classifies and archives the physical examination results according to the physical examination item types, and generates the classified record of the physical examination information.

[0121] The above are only the preferred embodiments of the present invention, and do not limit the present invention in other forms. Any person skilled in the art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A health examination data management method based on blockchain, characterized in that: The following steps are involved: Based on the physical examination data of users on the chain, the date, event, gender and age of multiple physical examination data on the blockchain are analyzed, the physical examination data are classified and index tags are added to generate a physical examination data index set; Based on the physical examination data index set, the query keywords are analyzed in combination with the physical examination data query records, the queryer's intention is identified and the query keywords are matched with the physical examination data to obtain a data fuzzy query tag; Based on the data fuzzy query tag, analyze the visitor's login time, data access frequency, and access type, analyze the access behavior of multiple visitors, and generate abnormal access behavior records through behavior pattern recognition and parameter comparison; Based on the abnormal access behavior records, analyze the visitor's user role, behavior records and access logs, match access rights for multiple visitors, and generate access rights allocation parameters; Based on the access permission allocation parameters, the privacy information in the physical examination data is analyzed and desensitized, and access permission rules are set according to the privacy protection level to generate privacy-protected physical examination data; Based on the privacy-protected physical examination data, the physical examination items and physical examination values ​​in the physical examination results are analyzed, the physical examination data are archived and uploaded to the blockchain, and a classified record of the physical examination information is generated.

2. The blockchain-based health examination data management method according to claim 1 is characterized in that: The physical examination data index set includes a physical examination subject information index, a physical examination date index and a physical examination event index; the data fuzzy query tag includes a keyword matching tag, a query intent analysis tag and an associated physical examination parameter tag; the abnormal access behavior record includes an abnormal access time record, a frequent access record and an unauthorized access record; the access permission allocation parameters include a doctor access level, a nurse access level and an administrative staff access level; the privacy protection physical examination data includes identity information desensitized data, access restriction data and encryption measures data; the physical examination information classification record includes routine examination classification results, special pathology analysis classification results and emergency event classification information.

3. The blockchain-based health examination data management method according to claim 1 is characterized in that: Based on the physical examination data of users on the chain, the date, event, gender and age of multiple physical examination data on the blockchain are analyzed, the physical examination data are classified and index tags are added. The specific steps for generating the physical examination data index set are as follows: Based on the physical examination data of users on the chain, analyze the dates and events of multiple physical examination data, analyze the occurrence time of the event and the recording date of the physical examination data, classify the physical examination data, and generate date event classification information; Based on the date event classification information, by analyzing the gender and age of the examinee in the plurality of physical examination data, the physical examination data are classified by gender and age to generate gender and age classification data; Based on the gender and age classification data, index labels are matched to the physical examination data, and a physical examination data index set is generated by recording the classification and label information.

4. The blockchain-based health examination data management method according to claim 1 is characterized in that: Based on the physical examination data index set, the query keywords are analyzed in combination with the physical examination data query records, the queryer's intention is identified and the query keywords are matched with the physical examination data to obtain the data fuzzy query label. Specifically, the steps are as follows: Based on the physical examination data index set, combined with the physical examination data query record, analyzing and identifying query keywords, including the frequency of occurrence and relevance of the query keywords, and generating query keyword analysis results; Based on the query keyword analysis result, the query statement and keywords are analyzed, the query intent is identified through the relevance and frequency of the keywords, and query intent identification information is generated; Based on the query intent recognition information, by matching the query keywords with the physical examination data, the association information between the query keywords and the query intent is constructed to obtain a data fuzzy query label.

5. The blockchain-based health examination data management method according to claim 1 is characterized in that: Based on the data fuzzy query tag, analyze the visitor's login time, data access frequency, and access type, analyze the access behavior of multiple visitors, and generate abnormal access behavior records through behavior pattern recognition and parameter comparison. Specifically, the steps are as follows: Based on the data fuzzy query tag, analyze the visitor's login time, data access frequency, and access type in the access record to generate an access log analysis result; Based on the access log analysis results, by analyzing the access behaviors of multiple visitors, identifying multiple access patterns and marking abnormal access behaviors, an access pattern analysis result is generated; Based on the access pattern analysis results, the abnormal behaviors of multiple visitors are identified by analyzing the behavior patterns, and the historical access parameters are compared. The support vector machine algorithm is used to analyze the persistence and consistency of the abnormal behaviors, and the abnormal access behavior records are generated.

6. The blockchain-based health examination data management method according to claim 5 is characterized in that: The support vector machine algorithm is based on the formula: y=sign(w1x1+w2x2+w3x3+w4x4+b) Identify abnormal behavior, where y is the identification result, w1 is the weight coefficient of the access time interval, x1 is the access time interval, w2 is the weight coefficient of the data access volume, x2 is the data access volume, w3 is the weight coefficient of the access source IP, x3 is the access source IP, w4 is the weight coefficient of the access device type, x4 is the access device type, b is the bias term, and sign is the sign function.

7. The blockchain-based health examination data management method according to claim 1 is characterized in that: Based on the abnormal access behavior records, the steps of analyzing the visitor user role, behavior records and access logs, matching access rights for multiple visitors, and generating access rights allocation parameters are as follows: Based on the abnormal access behavior records, analyze the user roles and behavior information in the abnormal behavior, identify the types of visitors, including doctors, nurses, and administrative staff, and generate role behavior analysis data; Based on the role behavior analysis data, combined with the matching access needs and security requirements of the visitor roles, multiple levels of access rights are set to match the data access needs of multiple roles, and role permission matching information is obtained; Based on the role permission matching information, combined with the behavior patterns and access requirements of multiple visitor roles, multiple levels of access permission rules are constructed, including data type, access time, and editing permission, and access permission allocation parameters are generated.

8. The blockchain-based health examination data management method according to claim 1 is characterized in that: Based on the access permission allocation parameters, the privacy information in the physical examination data is analyzed and desensitized, and the access permission rules are set according to the privacy protection level. The steps of generating privacy protection physical examination data are specifically as follows: Based on the access rights allocation parameters, analyzing and identifying the private information in the physical examination data, including the name, address, and telephone number of the examinee, and generating a private information identification result; Based on the privacy information identification result, the physical examination privacy information is desensitized, and the privacy desensitization result is generated by replacing the privacy information with an anonymous identifier; Based on the privacy desensitization processing result, access permission rules are set according to the privacy protection level, and privacy protection physical examination data is generated in combination with the privacy protection level information in the physical examination data.

9. The blockchain-based health examination data management method according to claim 1 is characterized in that: Based on the privacy-protected physical examination data, the physical examination items and physical examination values ​​in the physical examination results are analyzed, the physical examination data is archived and uploaded to the blockchain, and the steps of generating a classified record of the physical examination information are specifically as follows: Based on the privacy-protected physical examination data, analyzing the physical examination items in the health physical examination data, identifying the names of the physical examination items and classifying them according to the item categories, and generating physical examination item classification data; Based on the physical examination item classification data, analyzing the physical examination values ​​in the physical examination results, classifying and recording the physical examination values ​​according to categories, and generating physical examination value distinction information; Based on the physical examination numerical differentiation information, multiple types of archive units are established for the physical examination data according to the physical examination items, numerical values ​​and types, and the archive units are numbered and classified to generate a classified record of the physical examination information.

10. A health examination data management system based on blockchain, characterized in that: According to the blockchain-based health examination data management method according to any one of claims 1 to 9, the system comprises: The information processing and classification module classifies the date, event, gender and age information in the data based on the physical examination data of users on the chain, matches index tags for various information, and generates data query index information; The query word analysis module uses the data query index information to analyze the physical examination data query record, identifies the search intent of the inquirer through judgment and selection logic, matches the query keywords with the physical examination data, and generates a query intent recognition result; The access log analysis module analyzes and records the visitor's login time, data access frequency and access type based on the query intention recognition result, identifies abnormal access behavior, and generates visitor record analysis data; The user rights configuration module uses the visitor record analysis data to analyze the user roles and behavior records of various visitors, matches access rights for various visitors, and generates role access level matching results; The data archiving processing module analyzes and identifies the privacy information in the physical examination data based on the role access level matching result, desensitizes the privacy information, classifies and archives the physical examination results according to the type of physical examination items, and generates a classification record of the physical examination information.

Images