A health record automatic updating system

By using differential privacy algorithms and event-driven update mechanisms in the health record management system, the problem of data protection and update lag is solved, real-time updates and data integrity verification are achieved, and data security and processing efficiency are improved.

CN118538354BActive Publication Date: 2024-11-22JIANGSU RUIBO INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411010802.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-26
Publication Date
2024-11-22
Estimated Expiration
2044-07-26

AI Technical Summary

Technical Problem

The existing health record management system has insufficient dynamics and adaptability of data protection, and is unable to effectively deal with new risks in the ever-changing data environment, resulting in limited privacy protection capabilities, lagging data updates, unable to promptly reflect the latest information, and lack of Data integrity verification affects the accuracy of data analysis.

Method used

Differential privacy algorithm is used to add random noise to sensitive information, real-time update of data is achieved through dynamic update rules and event-driven update mechanisms, and structural and semantic consistency verification is carried out to identify data abnormal patterns, ensuring data integrity and security.

Benefits of technology

It significantly enhances the confidentiality and response time of data, improves the efficiency and accuracy of data processing, improves the dynamic adaptability and comprehensive protection capabilities of data, and ensures the high quality and reliability of healthy records.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118538354B_ABST
    Figure CN118538354B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of health record management, and specifically to an automatic health record update system, which includes a privacy protection update module, a dynamic update rule module, an event-driven update module, and a record integrity check module. In the present invention, by introducing a differential privacy algorithm, random noise is added to sensitive health data, which significantly enhances the confidentiality of the data and controls the risk of information leakage, effectively protects the security of health information, and meets the demand for high security standards in health data processing. At the same time, the data update process is automatically triggered, so that the response time of the system is greatly shortened, and high efficiency of data processing is achieved, ensuring the real-time and accuracy of data updates. Combined with structural and semantic consistency checks and automatic identification of abnormal patterns in data, the accuracy and reliability of data are optimized, the overall data quality of health records is improved, and the dynamic adaptability of data processing and comprehensive data protection are significantly improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of health record management, and in particular to a health record automatic updating system. Background Art

[0002] The field of health record management technology involves the use of information technology to create, collect, maintain, transmit and store health information of individuals or groups, usually including electronic health record (EHR) systems, personal health record (PHR) systems and related data processing and analysis tools. The main purpose is to improve the efficiency and quality of medical services, support the decision-making process by providing real-time and accurate health information, and ensure the security and privacy protection of this information.

[0003] Among them, the health record automatic update system mainly refers to an automated technical system designed to update and maintain health records without manual intervention. Its main purpose is to ensure the timeliness and accuracy of health records, reduce the problem of inaccurate health information caused by human errors, and improve the efficiency of medical services. Through automated data processing, health information can be updated in real time to ensure the high quality and effectiveness of data.

[0004] Although existing technologies support extensive processing of health information, including creation, collection, maintenance, and storage, they often have obvious shortcomings in the dynamism and adaptability of data protection. In particular, in terms of privacy protection measures, traditional systems cannot effectively respond to new risks in the ever-changing data environment, resulting in limited privacy protection capabilities. In addition, existing technologies lack effective real-time data monitoring and immediate response mechanisms, resulting in delayed data updates and failure to reflect the latest health information in a timely manner. In terms of data integrity verification, existing systems generally do not support consistency checks during the update process, resulting in errors in data analysis and affecting the use of subsequent data. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a health record automatic updating system.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: A health record automatic update system includes:

[0007] The privacy protection update module receives health data, uses a differential privacy algorithm to add random noise to sensitive information, controls the risk of information leakage by calculating the relationship between the amount of noise and data sensitivity, and generates noise-enhanced data;

[0008] The dynamic update rule module extracts metadata definitions in the records based on the noise enhancement data, analyzes and matches update requirements and rules, ensures that the update operation complies with metadata specifications by adjusting data fields and structures, and generates structure adjustment records;

[0009] The event-driven update module monitors key data change events. If data changes are detected, the data update process is triggered based on the structure adjustment record, the data response mechanism is activated, the data is updated in real time, and an event response update record is generated;

[0010] The record integrity verification module updates the record based on the event response, performs structural and semantic consistency verification of the health record, detects data anomalies, identifies abnormal patterns in the data, and performs data quality analysis and integrity verification to obtain a health data integrity verification record.

[0011] As a further solution of the present invention, the step of acquiring the noise enhancement data is:

[0012] Receive health data, define data sensitivity, and set sensitivity index based on data attributes and access frequency ,

[0013]

[0014] in, represents the attribute weight, Represents the data access frequency, Represents the attribute sensitivity level, and obtains the data sensitivity set;

[0015] Generate random noise according to the data sensitivity set and calculate the noise amount ,

[0016]

[0017] in, is a small constant that ensures the stability of the calculation. Improve the protection of highly sensitive data. Used to adjust the noise level and generate noise parameter sets;

[0018] Combining the data sensitivity set with the noise parameter set, noise enhancement data is obtained through superposition operation,

[0019]

[0020] in, represents the original data point, is the amount of noise, is the sensitivity, and the absolute value is used to ensure that the noise remains positive, resulting in noise-enhanced data.

[0021] As a further solution of the present invention, the steps of analyzing and matching the requirements and rules are as follows:

[0022] Analyze the modification requirements from the noise-enhanced data and calculate the intensity of the requirements based on the change frequency and business impact of the data fields ,

[0023]

[0024] in, is the number of field changes, is the business importance of the field, is the field length, is the last modified time, Represents the effect of time decay and obtains the demand intensity ;

[0025] Quantitatively evaluate metadata rules and assign rule fitness ,

[0026]

[0027] in, is the scope of application of the rule, It is the strictness of the rules. is the rule implementation time, and the rule adaptability is obtained ;

[0028] Combined with the demand intensity and rule fitness Calculate the matching index ,

[0029]

[0030] in, and Represent the values ​​of demand intensity and rule adaptability respectively, and the demand intensity and rule fitness It is concluded that the matching index is obtained by normalizing the score to ensure standardization. .

[0031] As a further solution of the present invention, the steps of obtaining the structure adjustment record are:

[0032] According to the matching index Determine the tuning parameters,

[0033]

[0034] in, and It is the adjustment sensitivity and smoothing factor, which is used to adjust the sensitivity of the adjustment parameter and obtain the adjustment parameter ;

[0035] Use the tuning parameters and rule fitness Design data structure adjustment plan,

[0036]

[0037] in, is the fitness score, which is determined by the rule fitness It is calculated that is the phase angle related to the business logic. The cosine function is used to increase the adaptability of the adjustment plan to obtain the adjustment plan ;

[0038] The adjustment plan Applied to existing data structures to generate structural adjustment records ,

[0039]

[0040] in, Represents the original data structure, Indicates the data structure merge operation, ensuring that the adjustment record is consistent with the metadata specification and obtaining the structure adjustment record .

[0041] As a further solution of the present invention, the steps of obtaining the event response update record are:

[0042] Deploy data listeners to continuously monitor changes in key data fields and calculate data changes ,

[0043]

[0044] in, and Respectively, in time and The data value of Indicates the monitoring weight of the corresponding field, which is allocated according to the importance of the field. is the change sensitivity index, which determines the sensitivity to data changes and obtains the data change amount ;

[0045] Set the preset threshold to , based on the change , determine whether to trigger an update,

[0046]

[0047] Use the activation function to adjust the sensitivity of the update. Exceed When it gradually tends to 1, it means starting the update process; the starting condition is obtained;

[0048] When the update process is triggered, record the update process.

[0049]

[0050] in, Indicates the cumulative degree of change during the data update process. is the data change indicator of the time variable, To adjust the function, it is expressed as the degree of data response related to time. To monitor the time window, the integral represents the total amount of changes recorded, and the event response update record is obtained.

[0051] As a further solution of the present invention, the steps of verifying the structural and semantic consistency of the health record are:

[0052] Extract key data sets from the incident response update record,

[0053]

[0054] in, represents the selection weight of each record, Data items representing health records, constructing a weighted dataset ;

[0055] For the weighted dataset Perform structural consistency analysis and compare with predefined health record models,

[0056]

[0057] in, is the consistency evaluation function, yes The record structure in is the corresponding model structure, and the structural consistency score is evaluated. ;

[0058] According to the structural consistency score , perform semantic verification to confirm the semantic conformity of the data,

[0059]

[0060] in, is the semantic sensitivity enhancement coefficient, adjusting Enhance the response to deviation and obtain semantic consistency evaluation results .

[0061] As a further solution of the present invention, the step of identifying abnormal patterns in the data is:

[0062] For the weighted dataset , perform anomaly detection,

[0063]

[0064] in, is the weight, is the monitoring value, is the expected value. By increasing the cube root, the sensitivity of anomaly detection is improved to obtain the preliminary anomaly score. ;

[0065] Based on the , assess the significance of the anomaly,

[0066]

[0067] The abnormality score is corrected by the logic function, the recognition sensitivity of small value abnormalities is optimized, and the quantified abnormality degree is obtained. ;

[0068] According to the quantified abnormality , perform pattern analysis to identify unusual patterns in the data,

[0069]

[0070] is the significance percentage of the abnormal pattern, and the output range is adjusted using the hyperbolic tangent function to obtain the identified abnormal pattern.

[0071] As a further solution of the present invention, the steps for obtaining the health data integrity verification record are:

[0072] According to the significance percentage of the abnormal pattern , assess data integrity,

[0073]

[0074] in, Indicates the data integrity percentage, from Calculate the preliminary data integrity results ;

[0075] Based on the preliminary data integrity results , calculate the data quality score,

[0076]

[0077] in, is the quality score factor, which is used to adjust the sensitivity of the quality score and obtain the data quality score ;

[0078] Based on the data quality score , create health data integrity verification records,

[0079]

[0080] in, Verify records for completeness, including data quality scoring.

[0081] Compared with the prior art, the advantages and positive effects of the present invention are:

[0082] In the present invention, by introducing a differential privacy algorithm, random noise is added to sensitive health data, which significantly enhances the confidentiality of the data and controls the risk of information leakage, effectively protects the security of health information, and meets the demand for high security standards in health data processing. At the same time, the data update process is automatically triggered, so that the response time of the system is greatly shortened, and high efficiency of data processing is achieved, ensuring the real-time and accuracy of data updates. Combined with structural and semantic consistency checks and automatic identification of abnormal patterns in data, the accuracy and reliability of the data are optimized, the overall data quality of health records is improved, and the dynamic adaptability of data processing and comprehensive data protection are significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 is a system flow chart of the present invention;

[0084] Figure 2 This is a flow chart for obtaining noise enhancement data of the present invention;

[0085] Figure 3 Update the flow chart for analysis and matching of the requirements and rules of the present invention;

[0086] Figure 4 A flow chart for obtaining records of structural adjustment of the present invention;

[0087] Figure 5 A flowchart for obtaining an event response update record of the present invention;

[0088] Figure 6 A flowchart for checking the structural and semantic consistency of the health record of the present invention;

[0089] Figure 7 is a flow chart for identifying abnormal patterns in data of the present invention;

[0090] Figure 8 This is a flow chart for obtaining health data integrity verification records of the present invention. DETAILED DESCRIPTION

[0091] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with 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 intended to limit the present invention.

[0092] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are 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 cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined. Embodiment 1

[0093] See also Figure 1 , a health record automatic updating system comprising:

[0094] The privacy protection update module receives health data, uses a differential privacy algorithm to add random noise to sensitive information, controls the risk of information leakage by calculating the relationship between the amount of noise and data sensitivity, and generates noise-enhanced data;

[0095] The dynamic update rule module extracts metadata definitions from records based on noise enhancement data, analyzes and matches update requirements and rules, and ensures that update operations comply with metadata specifications by adjusting data fields and structures, generating structure adjustment records;

[0096] The event-driven update module monitors key data change events. If data changes are detected, the data update process is triggered based on the structure adjustment record, the data response mechanism is activated, the data is updated in real time, and the event response update record is generated;

[0097] The record integrity verification module updates records based on event responses, performs structural and semantic consistency checks on health records, detects data anomalies, identifies abnormal patterns in the data, and performs data quality analysis and integrity verification to obtain health data integrity verification records.

[0098] Noise enhancement data includes noise level, sensitivity score and confidentiality parameters. Structure adjustment records include field matching details, data structure adjustment results and metadata compliance standards. Event response update records include trigger time point, data change log and response mechanism status. Health data integrity verification records include structure verification results, semantic matching results and anomaly detection records.

[0099] See also Figure 2 , the steps to obtain noise enhancement data are:

[0100] Receive health data, define data sensitivity, and set sensitivity index based on data attributes and access frequency ,

[0101]

[0102] in, represents the attribute weight, Represents the data access frequency, Represents the attribute sensitivity level, and obtains the data sensitivity set;

[0103] Generate random noise based on data sensitivity and calculate the amount of noise ,

[0104]

[0105] in, is a small constant that ensures the stability of the calculation. Improve the protection of highly sensitive data. Used to adjust the noise level and generate noise parameter sets;

[0106] Combining the data sensitivity set with the noise parameter set, the noise enhanced data is obtained through superposition operation.

[0107]

[0108] in, represents the original data point, is the amount of noise, is the sensitivity, and the absolute value is used to ensure that the noise remains positive, resulting in noise-enhanced data.

[0109] Calculate data sensitivity:

[0110] : Attribute weight. Assume that in this scenario, the weight of a data attribute (such as personal identification information) is 2;

[0111] : Data access frequency, assuming that this type of data is accessed 10 times a day;

[0112] : Attribute sensitivity level, assuming that the sensitivity level of identification information is 5;

[0113] If there is only one sensitive information attribute, that is =1, then:

[0114]

[0115] Indicates that the total sensitivity of the given sensitive information is 100.

[0116] Calculate the amount of noise:

[0117] : The data sensitivity obtained from the previous step is 100;

[0118] : A small constant to ensure calculation stability, assumed to be 0.01;

[0119] but:

[0120]

[0121]

[0122] get =0.4604, which is the amount of noise,

[0123]

[0124] Get noise-enhanced data:

[0125] : Original data point, assuming the original value is 50;

[0126] : The amount of noise obtained from the previous step is 0.4604;

[0127] : Data sensitivity, which is 100;

[0128] but,

[0129]

[0130] It shows that the data point after adding noise is 96.04.

[0131] See also Figure 3 , the analysis and matching update steps of requirements and rules are:

[0132] Analyze modification requirements from noise-enhanced data and calculate the intensity of requirements based on the frequency of changes in data fields and business impact ,

[0133]

[0134] in, is the number of field changes, is the business importance of the field, is the field length, is the last modified time, Represents the effect of time decay and obtains the demand intensity ;

[0135] Quantitatively evaluate metadata rules and assign rule fitness ,

[0136]

[0137] in, is the scope of application of the rule, It is the strictness of the rules. is the rule implementation time, and the rule adaptability is obtained ;

[0138] Combined with demand intensity and rule fitness Calculate the matching index ,

[0139]

[0140] in, and Represent the values ​​of demand intensity and rule adaptability respectively, and the demand intensity and rule fitness It is concluded that the matching index is obtained by normalizing the score to ensure standardization. .

[0141] Calculate demand intensity:

[0142] : The number of field changes, assumed to be the number of changes in the last year;

[0143] : The business importance of the field, usually ranging from 1 (lowest) to 10 (highest);

[0144] : Field length, that is, the number of characters in the data field;

[0145] : The time since the last modification, in months;

[0146] Set the number of field changes , indicating that a data field has been modified 3 times in the past year

[0147] Set the business importance of the field , indicating that this data field is of medium importance to the business;

[0148] Set field length , indicating that the data field contains 10 characters;

[0149] Set last modified time , indicating that 6 months have passed since the last modification;

[0150] Calculate exponential decay factor ,exist hour, ( is a constant 2.71);

[0151] The intensity of demand ,

[0152]

[0153] Calculation rule fitness:

[0154] : The scope of application of the rule, ranging from 1 (local rule) to 10 (company-level rule);

[0155] : the strictness of the rule, ranging from 1 (lenient) to 10 (very strict);

[0156] : The implementation time of the rule, in years;

[0157] Set the scope of application of the rule , indicating higher-level corporate rules;

[0158] Setting the strictness of the rules , indicating that this rule is quite strict;

[0159] Set the time for the rule to take effect , indicating that the rule has been in effect for 2 years;

[0160] Calculate the logarithmic adjustment factor for , ;

[0161] Rule fitness ,

[0162]

[0163] Calculate the matching index:

[0164] : Demand intensity, which represents the quantitative value of the update demand of each field, according to to conclude;

[0165] : Rule fitness, which represents the quantitative value of the fitness of each rule. to conclude;

[0166] Assumptions =2,

[0167] (Demand intensity)

[0168] (Another rule fitness)

[0169] but,

[0170]

[0171]

[0172] The final calculation is

[0173]

[0174] The matching index is .

[0175] See also Figure 4 , the steps to obtain the structural adjustment record are:

[0176] According to the matching index Determine the tuning parameters,

[0177]

[0178] in, and It is the adjustment sensitivity and smoothing factor, which is used to adjust the sensitivity of the adjustment parameter and obtain the adjustment parameter ;

[0179] Using Tuning Parameters and rule fitness Design data structure adjustment plan,

[0180]

[0181] in, is the fitness score, which is determined by the rule fitness It is calculated that is the phase angle related to the business logic. The cosine function is used to increase the adaptability of the adjustment plan to obtain the adjustment plan ;

[0182] Will adjust the plan Applied to existing data structures to generate structural adjustment records ,

[0183]

[0184] in, Represents the original data structure, Indicates the data structure merge operation, ensuring that the adjustment record is consistent with the metadata specification and obtaining the structure adjustment record .

[0185] Calculate the adjustment parameters:

[0186] set up , adjust the sensitivity base coefficient;

[0187] set up , smoothing factor;

[0188] set up , the matching index calculated previously;

[0189] but,

[0190]

[0191] Design adjustment plan:

[0192] , rule fitness score;

[0193] , phase angle, adjusted according to business logic needs;

[0194] but,

[0195]

[0196]

[0197] Generate structural adjustment records:

[0198] If the base value of the original data structure =100;

[0199] but,

[0200]

[0201] Indicates that the adjusted plan The new data structure after application is .

[0202] See also Figure 5 , the steps to obtain the event response update record are:

[0203] Deploy data listeners to continuously monitor changes in key data fields and calculate data changes ,

[0204]

[0205] in, and Respectively, in time and The data value of Indicates the monitoring weight of the corresponding field, which is allocated according to the importance of the field. is the change sensitivity index, which determines the sensitivity to data changes and obtains the data change amount ;

[0206] Set the preset threshold to , based on the change , determine whether to trigger an update,

[0207]

[0208] Use the activation function to adjust the sensitivity of the update. Exceed When it gradually tends to 1, it means starting the update process; the starting condition is obtained;

[0209] When the update process is triggered, record the update process.

[0210]

[0211] in, Indicates the cumulative degree of change during the data update process. is the data change indicator of the time variable, To adjust the function, it is expressed as the degree of data response related to time. To monitor the time window, the integral represents the total amount of changes recorded, and the event response update record is obtained.

[0212] Calculate the data change:

[0213] Assume that three fields are monitored, .

[0214] In time and The data values ​​are

[0215]

[0216]

[0217]

[0218] Pick To enhance the response to changes,

[0219]

[0220] Start condition determination:

[0221] Assumption threshold ;

[0222] Using the previously calculated ;

[0223] Calculation start conditions:

[0224] Start conditions

[0225] Start conditions

[0226] Start conditions

[0227] Start conditions

[0228] Indicates that at a given threshold and data changes Under the condition of , the update condition is fully satisfied, almost close to the maximum value ,because Part of the calculation is Significantly greater than When it tends to 1, it means that the conditions for triggering the data update process are fully met, so the data update process will be activated in response to the detected key data changes.

[0229] Incident Response Update Log:

[0230] Assumptions Hour, (constant), (No adjustment);

[0231] but ,

[0232]

[0233] E=54.5 means that within 2 hours, the system has processed a cumulative change equivalent to 54.5, which quantifies the activity or cumulative degree of change in the data update process, helps understand the dynamic changes of data within the set time window, and then obtains event response update records.

[0234] See also Figure 6 , the steps for checking the structural and semantic consistency of health records are:

[0235] Extract key data sets from incident response update records,

[0236]

[0237] in, represents the selection weight of each record, Data items representing health records, constructing a weighted dataset ;

[0238] For the weighted dataset Perform structural consistency analysis and compare with predefined health record models,

[0239]

[0240] in, is the consistency evaluation function, yes The record structure in is the corresponding model structure, and the structural consistency score is evaluated. ;

[0241] According to the structural consistency score , perform semantic verification to confirm the semantic conformity of the data,

[0242]

[0243] in, is the semantic sensitivity enhancement coefficient, adjusting Enhance the response to deviation and obtain semantic consistency evaluation results .

[0244] Extract the key data sets:

[0245] : The selection weight of each record, assumed to be a score of the importance of the record, ranging from 1 to 10;

[0246] : Data items in health records, such as temperature or blood pressure values;

[0247] : The key data set is the sum of health records adjusted by weight;

[0248] Assume there are three health records:

[0249] Record 1: Body temperature 37.5℃, importance score 8;

[0250] Record 2: blood pressure 120 / 80 mmHg, importance score 7;

[0251] Record 3: heart rate 75bpm, importance score 6;

[0252] but,

[0253]

[0254] Evaluate data structure consistency:

[0255] : Structural matching function, here it is assumed that it returns a value between 0 (complete mismatch) and 1 (complete match);

[0256] The structure of the actual record;

[0257] : Model structure;

[0258] : The total number of records is 3;

[0259] Assumptions returns 0.8 for all three records ( ),but,

[0260]

[0261] Semantic consistency evaluation:

[0262] : Semantic sensitivity enhancement coefficient, assuming it is 5 to increase the sensitivity of the function, then,

[0263]

[0264] Values ​​close to 1 indicate high semantic consistency, indicating that the data has good consistency between the structure and the expected model, meaning that most of the data records are consistent with the predefined health data model structure and semantics.

[0265] See also Figure 7 , the steps to identify abnormal patterns in the data are:

[0266] For the weighted dataset , perform anomaly detection,

[0267]

[0268] in, is the weight, is the monitoring value, is the expected value. By increasing the cube root, the sensitivity of anomaly detection is improved to obtain the preliminary anomaly score. ;

[0269] based on , assess the significance of the anomaly,

[0270]

[0271] The abnormality score is corrected by the logic function, the recognition sensitivity of small value abnormalities is optimized, and the quantified abnormality degree is obtained. ;

[0272] According to the quantified abnormality , perform pattern analysis to identify unusual patterns in the data,

[0273]

[0274] is the significance percentage of the abnormal pattern, and the output range is adjusted using the hyperbolic tangent function to obtain the identified abnormal pattern.

[0275] Implementing anomaly detection:

[0276] : Weight factor, assumed to be 1;

[0277] : Observed values, such as 37.5℃, 120mmHg, 75bpm;

[0278] : Expected value, assuming 36.5℃, 110mmHg, 70bpm;

[0279] but,

[0280]

[0281] Quantification of abnormality:

[0282]

[0283] A high Q value indicates significant abnormal behavior in the data.

[0284] Abnormal pattern recognition:

[0285]

[0286] M=100 means that all analyzed data points are considered abnormal and no data point is considered normal, indicating that there are major problems in the data collection, processing or monitoring system and the entire process of data collection and processing needs to be re-evaluated.

[0287] See also Figure 8 , the steps to obtain the health data integrity verification record are:

[0288] Percentage of significance based on abnormal patterns , assess data integrity,

[0289]

[0290] in, Indicates the data integrity percentage, from Calculate the preliminary data integrity results ;

[0291] Based on preliminary data integrity results , calculate the data quality score,

[0292]

[0293] in, is the quality score factor, which is used to adjust the sensitivity of the quality score and obtain the data quality score ;

[0294] Based on data quality score , create health data integrity verification records,

[0295]

[0296] in, Verify records for completeness, including data quality scores, and ensure results of data quality analyses are clearly documented.

[0297] Data integrity analysis:

[0298] Calculated from the previous step =100, then,

[0299]

[0300] It means that the data integrity is 0%, which means that all data are identified as abnormal.

[0301] Calculate data quality score:

[0302]

[0303] is the data integrity percentage, which is calculated as 0. is the quality scoring factor, assumed to be 0.8, in order to adjust the final quality score to reflect the overall low quality of the data even if anomaly detection shows high anomalies.

[0304]

[0305] because is 0, so the overall quality score will also be 0.

[0306] Integrity Verification Record Generation:

[0307]

[0308] A string will be generated stating that the data has a quality score of 0, indicating that the data completeness and quality are very low and may require further review or corrective action.

[0309] Assuming that a set of patient health data is processed, including heart rate, blood pressure and body temperature, the percentage of abnormal patterns A value of 100 means that all data are marked as abnormal. Through a series of calculations, we can draw the following conclusions:

[0310] Data Integrity It is O%, which means no data is normal;

[0311] Due to very low completeness, the data quality score It is also 0, indicating that the data is unreliable;

[0312] Integrity Verification Records The data quality issues were pointed out, providing a basis for subsequent data processing and decision-making.

[0313] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A health record automatic updating system, characterized in that: The system comprises: The privacy protection update module receives health data, uses a differential privacy algorithm to add random noise to sensitive information, controls the risk of information leakage by calculating the relationship between the amount of noise and data sensitivity, and generates noise-enhanced data; The dynamic update rule module extracts metadata definitions in the records based on the noise enhancement data, analyzes and matches update requirements and rules, ensures that the update operation complies with metadata specifications by adjusting data fields and structures, and generates structure adjustment records; The event-driven update module monitors key data change events. If data changes are detected, the data update process is triggered based on the structure adjustment record, the data response mechanism is activated, the data is updated in real time, and an event response update record is generated; The record integrity verification module updates the record based on the event response, performs structural and semantic consistency verification of the health record, detects data anomalies, identifies abnormal patterns in the data, and performs data quality analysis and integrity verification to obtain a health data integrity verification record; The steps for obtaining the noise enhancement data are as follows: Receive health data, define data sensitivity, and set sensitivity index based on data attributes and access frequency , ; in, represents the attribute weight, Represents the data access frequency, Represents the attribute sensitivity level, and obtains the data sensitivity set; Generate random noise according to the data sensitivity set and calculate the noise amount , ; in, is a small constant that ensures the stability of the calculation. Improve the protection of highly sensitive data. Used to adjust the noise level and generate noise parameter sets; Combining the data sensitivity set with the noise parameter set, noise enhancement data is obtained through superposition operation, ; in, represents the original data point, is the amount of noise, is the sensitivity, using the absolute value to ensure that the noise remains positive and obtain noise-enhanced data; The steps of analyzing and matching the requirements and rules are as follows: Analyze the modification requirements from the noise-enhanced data and calculate the intensity of the requirements based on the change frequency and business impact of the data fields , ; in, is the number of field changes, is the business importance of the field, is the field length, is the last modified time, Represents the effect of time decay and obtains the demand intensity ; Quantitatively evaluate metadata rules and assign rule fitness , ; in, is the scope of application of the rule, It is the strictness of the rules. is the rule implementation time, and the rule adaptability is obtained ; Combined with the demand intensity and rule fitness Calculate the matching index , ; in, and Represent the values ​​of demand intensity and rule adaptability respectively, and the demand intensity and rule fitness It is concluded that the matching index is obtained by normalizing the score to ensure standardization. ; The steps for obtaining the structural adjustment record are: According to the matching index Determine the tuning parameters, ; in, and It is the adjustment sensitivity and smoothing factor, which is used to adjust the sensitivity of the adjustment parameter and obtain the adjustment parameter ; Use the tuning parameters and rule fitness Design data structure adjustment plan, ; in, is the fitness score, which is determined by the rule fitness It is calculated that is the phase angle related to the business logic. The cosine function is used to increase the adaptability of the adjustment plan to obtain the adjustment plan ; The adjustment plan Applied to existing data structures to generate structural adjustment records , ; in, Represents the original data structure, Indicates the data structure merge operation, ensuring that the adjustment record is consistent with the metadata specification and obtaining the structure adjustment record ; The steps for obtaining the event response update record are: Deploy data listeners to continuously monitor changes in key data fields and calculate data changes , ; in, and Respectively, in time and The data value of Indicates the monitoring weight of the corresponding field, which is allocated according to the importance of the field. is the change sensitivity index, which determines the sensitivity to data changes and obtains the data change amount ; Set the preset threshold to , based on the change , determine whether to trigger an update, ; Use the activation function to adjust the sensitivity of the update. Exceed When it gradually tends to 1, it means starting the update process; the starting condition is obtained; When the update process is triggered, record the update process. ; in, Indicates the cumulative degree of change during the data update process. is the data change indicator of the time variable, To adjust the function, it is expressed as the degree of data response related to time. To monitor the time window, the integral represents the total amount of changes recorded, and the event response update record is obtained; The steps for verifying the structural and semantic consistency of the health record are: Extract key data sets from the incident response update record, ; in, represents the selection weight of each record, Data items representing health records, constructing a weighted dataset ; For the weighted dataset Perform structural consistency analysis and compare with predefined health record models, ; in, is the consistency evaluation function, yes The record structure in is the corresponding model structure, and the structural consistency score is evaluated. ; According to the structural consistency score , perform semantic verification to confirm the semantic conformity of the data, ; in, is the semantic sensitivity enhancement coefficient, adjusting Enhance the response to deviation and obtain semantic consistency evaluation results ; The steps for identifying abnormal patterns in the data are: For the weighted dataset , perform anomaly detection, ; in, is the weight, is the monitoring value, is the expected value. By increasing the cube root, the sensitivity of anomaly detection is improved to obtain the preliminary anomaly score. ; Based on the , assess the significance of the anomaly, ; The abnormality score is corrected by the logic function, the recognition sensitivity of small value abnormalities is optimized, and the quantified abnormality degree is obtained. ; According to the quantified abnormality , perform pattern analysis to identify unusual patterns in the data, ; is the significance percentage of the abnormal pattern, and the output range is adjusted using the hyperbolic tangent function to obtain the identified abnormal pattern; The steps for obtaining the health data integrity verification record are: According to the significance percentage of the abnormal pattern , assess data integrity, ; in, Indicates the data integrity percentage, from Calculate the preliminary data integrity results ; Based on the preliminary data integrity results , calculate the data quality score, ; in, is the quality score factor, which is used to adjust the sensitivity of the quality score and obtain the data quality score ; Based on the data quality score , create health data integrity verification records, ; in, Verify records for completeness, including data quality scores.