Medical data conversion method and system

Abstract

The invention relates to the technical field of medical data processing, in particular to a medical data conversion method and system, and the method comprises the steps: analyzing the relevance between medical data, and determining a data group with an incidence relation; a collaborative conversion rule is set for the data sets with the incidence relation, one data in the data sets is set to be converted through an elliptic equation, the other data in the data sets is set to be mapped through spherical coordinates, and the elliptic eccentricity of conversion of one data is inversely proportional to the spherical radius of conversion of the other data; and acquiring medical data to be converted, performing collaborative geometric conversion operation on the medical data to be converted according to a collaborative conversion rule set in the collaborative conversion rule base and the allocated shared timestamp offset, and introducing a noise disturbance factor in the conversion process. According to the method, the collaborative conversion rule base is designed by utilizing the relevance among the medical data, so that the relevance data are mutually constrained in the conversion process to form a conversion chain, and the complexity of reverse cracking is increased.

Description

Technical Field

[0001] This application relates to the technical field of medical data processing, and in particular to a medical data conversion method and system. Background Technology

[0002] In the medical field, multi-parameter joint monitoring scenarios (such as ICU multi-sign monitoring) generate a large amount of medical data. These data are closely correlated, for example, heart rate and blood oxygen saturation, blood glucose and insulin levels. Current encryption and transformation methods for medical data mostly focus on single transformation methods, lacking consideration for the correlations between data. When the transformation method for a single data point is compromised, the associated data is highly susceptible to leakage, seriously threatening the security of medical data. Summary of the Invention

[0003] In view of the technical problems mentioned above, this application proposes a medical data conversion method and system.

[0004] Firstly, this application proposes a medical data conversion method, comprising the following steps: Analyze the correlations between medical data to identify data groups with related relationships; To establish collaborative transformation rules for the data groups with related relationships, the rules specifically include: Assign a unique identifier to each data group; One set of data is transformed using an elliptic equation, while another set of data is transformed using spherical coordinate mapping, and the eccentricity of the ellipse transformed by the first set of data is inversely proportional to the radius of the sphere transformed by the other set of data. Assign shared timestamp offsets to associated data in the data group; The medical data to be converted is obtained, and a collaborative geometric transformation operation is performed on the medical data to be converted according to the collaborative transformation rules set in the collaborative transformation rule base and the allocated shared timestamp offset, and a noise disturbance factor is introduced during the transformation process.

[0005] By adopting the above technical solution and leveraging the correlation between medical data, a collaborative conversion rule base is designed to make the related data mutually constrain each other during the conversion process, forming a conversion chain, which increases the complexity of reverse engineering and effectively improves the overall data obfuscation strength.

[0006] Preferably, the shared timestamp offset is periodically adjusted based on data transmission frequency and association strength, including: The initial offset is calculated based on the initial acquisition time of the data in the data set and the average time difference of historical collaborative conversion. Real-time monitoring of data transmission frequency and quantification of correlation strength; determination of adjustment period based on data transmission frequency; and calculation of adjustment range based on correlation strength score. Based on data sensitivity, pseudo-random perturbations are introduced into the adjustment range to enhance security; Preferably, when the shared timestamp offset reaches the adjustment period, the conversion process of the data group is temporarily frozen, a new offset is calculated and synchronized to the timestamp generation module of all data in the data group, and the conversion process continues after unlocking.

[0007] Preferably, the introduction of noise disturbance factors during the conversion process includes: Perturbation factors are generated for the ellipse transformation results and the spherical mapping results based on data sensitivity.

[0008] Preferably, the parameters of the spherical coordinate mapping include the spherical radius, polar angle, and azimuth angle. The polar angle and azimuth angle are dynamically determined by a pseudo-random number generation algorithm, and the pseudo-random number generation algorithm is associated with the correlation coefficient of medical data in the data set.

[0009] Preferably, the historical conversion records of the associated data are retrieved, and the rationality of the current conversion parameters is verified using the historical conversion records.

[0010] Preferably, the collaborative geometric transformation effect is quantitatively evaluated by calculating the confusion, correlation retention, and recovery accuracy among the transformed medical data, and the collaborative transformation rule base is optimized based on the evaluation results.

[0011] Preferably, the data sets with correlations include heart rate and blood oxygen saturation data sets, blood glucose and insulin level data sets, blood pressure and heart rate variability data sets, and respiratory rate and blood oxygen saturation fluctuation data sets.

[0012] Preferably, weakly correlated medical data sets are obtained, and a hierarchical transformation strategy is adopted. The main data in the weakly correlated data sets are transformed using elliptical equations or spherical coordinate mapping, while the correlated data are transformed using simple geometric transformations.

[0013] Secondly, this application also proposes a medical data conversion system, the system comprising: The associated data group acquisition module is configured to analyze the correlation between medical data and identify data groups with related relationships. The collaborative transformation rule setting module configures collaborative transformation rules for the data groups with related relationships, specifically including: Assign a unique identifier to each data group; One set of data is transformed using an elliptic equation, while another set of data is transformed using spherical coordinate mapping, and the eccentricity of the ellipse transformed by the first set of data is inversely proportional to the radius of the sphere transformed by the other set of data. Assign shared timestamp offsets to associated data in the data group; The conversion module is configured to acquire medical data to be converted, perform collaborative geometric conversion operations on the medical data to be converted according to the collaborative conversion rules set in the collaborative conversion rule base and the allocated shared timestamp offset, and introduce noise disturbance factors during the conversion process.

[0014] The beneficial effects of this invention are as follows: It utilizes the correlation between medical data to design a collaborative conversion rule base, enabling related data to mutually constrain each other during the conversion process, forming a conversion chain. This increases the complexity of reverse engineering and effectively enhances the overall data obfuscation strength. By allocating shared timestamp offsets to strongly correlated data groups and employing a dynamic update mechanism, it ensures the synchronization of collaborative conversion, guarantees the accuracy and consistency of data conversion, and further improves security. The introduction of a noise disturbance factor based on data sensitivity during the conversion process enhances the obfuscation effect without affecting data correlation. The rule base supports dynamic expansion, facilitating the automatic generation of conversion rules when adding new medical data types, improving the adaptability and scalability of the mechanism. Secondary verification of conversion parameters avoids abnormal conversions and ensures conversion reliability. It supports multiple related data groups and allows user customization, expanding the application scope of the mechanism. The conversion effect evaluation module can quantitatively evaluate the conversion effect and provide feedback to optimize the rule base, continuously improving the mechanism. A tiered conversion strategy is adopted for weakly correlated data groups, balancing conversion efficiency and security. It avoids the problem of related data leakage after a single data conversion is cracked, making it suitable for multi-parameter joint monitoring scenarios (such as ICU multi-sign monitoring), providing strong protection for medical data security. Attached Figure Description

[0015] The accompanying drawings are included to provide a further understanding of the embodiments and are incorporated in and constitute a part of this specification. The drawings illustrate embodiments and, together with the description, serve to explain the principles of this application. Other embodiments and many anticipated advantages of these embodiments will be readily recognized as they become better understood through reference to the following detailed description. Elements in the drawings are not necessarily to scale. The same reference numerals refer to corresponding similar parts.

[0016] Figure 1 This is a flowchart of a medical data conversion method according to this application.

[0017] Figure 2 This is a schematic diagram of a specific embodiment of the medical data conversion method applicable to this application.

[0018] Figure 3 This is a schematic diagram of the generation of collaborative conversion rules in one embodiment of this application.

[0019] Figure 4 This is a schematic diagram of a collaborative conversion method in one embodiment of this application.

[0020] Figure 5 This is a schematic diagram of collaborative conversion rule base feedback optimization in one embodiment of this application.

[0021] Figure 6 This is a schematic diagram of the module structure of a medical data conversion system in one embodiment of this application. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to represent selected embodiments of the invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0024] Figure 1 A flowchart of a medical data conversion method according to this application is shown. Figure 2 A schematic diagram of a specific embodiment of the medical data conversion method applicable to this application is shown in conjunction with reference to the reference. Figure 1 and Figure 2 The method specifically includes the following steps: S1: Analyze the correlation between medical data and identify data groups with correlation relationships; In one specific embodiment, the data sets with correlations include heart rate and blood oxygen saturation data sets, blood glucose and insulin level data sets, blood pressure and heart rate variability data sets, and respiratory rate and blood oxygen saturation fluctuation data sets.

[0025] S2: Set collaborative transformation rules for data groups with related relationships, specifically including: Assign a unique identifier to each data group; Set one data point in the data set to be transformed using an elliptic equation, while another data point is transformed using spherical coordinate mapping, and the eccentricity of the ellipse transformed by the first data point is inversely proportional to the radius of the sphere transformed by the second data point; Assign shared timestamp offsets to associated data in the data group; In one specific embodiment, the shared timestamp offset is periodically adjusted based on data transmission frequency and association strength, including: The initial offset is calculated based on the initial acquisition time of the data in the data set and the average time difference of historical collaborative conversion. Real-time monitoring of data transmission frequency and quantification of correlation strength; determination of adjustment period based on data transmission frequency; and calculation of adjustment range based on correlation strength score. Based on data sensitivity, pseudo-random perturbations are introduced into the adjustment range to enhance security. In one specific embodiment, when the shared timestamp offset reaches the adjustment period, the conversion process of the data group is temporarily frozen, a new offset is calculated and synchronized to the timestamp generation module of all data in the data group, and the conversion process continues after the timestamp is unlocked.

[0026] In one specific embodiment, the parameters of the spherical coordinate mapping include the spherical radius, polar angle, and azimuth angle. The polar angle and azimuth angle are dynamically determined by a pseudo-random number generation algorithm, and the pseudo-random number generation algorithm is associated with the correlation coefficient of medical data in the data set.

[0027] Retrieve historical transformation records of related data and use these records to verify the rationality of the current transformation parameters.

[0028] S3: Obtain the medical data to be transformed, and perform a collaborative geometric transformation operation on the medical data to be transformed according to the collaborative transformation rules set in the collaborative transformation rule base and the allocated shared timestamp offset, while introducing noise perturbation factors during the transformation process. In one specific embodiment, introducing noise perturbation factors during the conversion process includes: Perturbation factors are generated for the ellipse transformation results and the spherical mapping results based on data sensitivity.

[0029] In one specific embodiment, the above method further includes the steps of: quantitatively evaluating the collaborative geometric transformation effect by calculating the confusion degree, correlation retention degree and recovery accuracy among the transformed medical data, and optimizing the collaborative transformation rule base based on the evaluation results.

[0030] In one specific embodiment, the above method further includes the steps of: obtaining a weakly correlated data group of medical data, and adopting a hierarchical transformation strategy to transform the main data in the weakly correlated data group using elliptical equations or spherical coordinate mapping, while the correlated data is transformed using simple geometric transformations.

[0031] In a specific embodiment, the medical data conversion method of this application will be described in detail below: First, construct a collaborative conversion rule base: It analyzes the correlation between medical data and identifies data groups with correlations, such as heart rate and blood oxygen saturation data groups, blood glucose and insulin level data groups, etc. It can also include blood pressure and heart rate variability data groups, respiratory rate and blood oxygen saturation fluctuation data groups, etc., and supports users to add custom correlated data groups.

[0032] For each data set, collaborative transformation rules are defined. When a data point in a data set is transformed using an elliptic equation, the related data in that set automatically match the complementary parameters of the spherical coordinate mapping, and the transformation parameters are dynamically adjusted based on the correlation coefficient between the two. In the collaborative transformation rules, when a data point is transformed using an elliptic equation and the related data uses spherical coordinate mapping, the eccentricity of the ellipse transformed from the data point is inversely proportional to the radius of the sphere transformed from the related data. A secondary verification mechanism for the transformation parameters is also included, verifying the rationality of the current parameters through the historical transformation records of the related data to avoid abnormal transformations.

[0033] Meanwhile, the rule base supports dynamic expansion, automatically generating new collaborative conversion rules based on newly added medical data types and their relationships, eliminating the need for manual recoding. (See reference...) Figure 3 .

[0034] Assign a shared timestamp offset to strongly correlated data groups: For strongly correlated data groups, a shared timestamp offset is allocated to ensure the synchronization of the collaborative transformation. The shared timestamp offset employs a dynamic update mechanism, periodically adjusted based on data transmission frequency and correlation strength, further enhancing the dynamism and security of the transformation. The specific steps are as follows: 1. Definition and Identification of Strongly Related Data Groups Based on the identified associated data groups (such as heart rate and blood oxygen saturation, blood glucose and insulin levels, etc.) in the collaborative transformation rule base, the strongly associated data groups are identified through quantitative screening using the correlation coefficient (correlation coefficient ≥ 0.8). Assign a unique identifier (such as GroupID) to each group, and record the correspondence and transformation logic between the main data (such as ellipse equation transformation data) and related data (such as spherical coordinate mapping data) within the group.

[0035] 2. Initial shared timestamp offset setting Based on the initial acquisition time of the master data within the group, and combined with the average time difference of historical collaborative conversion (such as the average synchronous acquisition time difference of heart rate and blood oxygen saturation), the initial shared offset Δt0 is calculated. Ensure that the initial timestamp benchmark for all data within the group is consistent (e.g., the timestamp for heart rate data is t, and the timestamp for blood oxygen data is t+Δt0).

[0036] 3. Real-time monitoring of data transmission frequency and correlation strength Transmission frequency monitoring: Data is divided into high frequency (≥60 times / minute, such as real-time heart rate), medium frequency (30-59 times / minute, such as respiratory rate), and low frequency (<30 times / minute, such as blood glucose) according to the number of transmissions per unit time, and the frequency fluctuation value ΔF is recorded; Association strength quantification: The association strength score S is calculated by the consistency of collaborative transformation parameters (such as the inverse deviation between the eccentricity e of the ellipse and the radius r of the sphere) and the temporal synchronization deviation (the difference in timestamps after data transformation within the group). (S∈[0,1], S≥0.8 indicates strong association).

[0037] 4. Calculation of adjustment period and amplitude Adjustment period T: High frequency data T=30 seconds, medium frequency T=5 minutes, low frequency T=30 minutes; Adjustment magnitude ΔΔt: When S≥0.9 (extremely strong correlation), ΔΔt=±(1-3ms) plus ±0.5ms pseudo-random perturbation; when 0.8≤S<0.9 (strong correlation), ΔΔt=±(3-5ms) plus ±1ms perturbation (perturbation factor is positively correlated with data sensitivity).

[0038] In one specific embodiment, the data sensitivity is obtained using a three-dimensional quantification method: 1) Business understanding dimension Business process analysis: Analyze the collection principles, frequency basis, and clinical scenarios of the target medical data, and combine the normal fluctuation range of the data under different physiological states to initially classify it according to "the smaller the fluctuation range, the higher the diagnostic relevance, and the higher the sensitivity"; Indicator Relationship Analysis: Clarify the inherent logic between target data and related data. If the following formula is satisfied, it is judged as high sensitivity: 2) Data Feature Analysis Dimensions Statistical characteristic calculation: Collect ≥1000 historical samples and calculate the mean ( ),variance( ), standard deviation ( ), determine the normal distribution range ( If the following formula is met and clinical intervention is required, it is considered high sensitivity; if only routine recording is required, it is considered low sensitivity: Trend analysis: Plot a time series curve for ≥30 days. If it meets the following formula, it is judged as high sensitivity; if it only reflects changes in the physiological cycle, it is judged as low sensitivity: 3) Historical data comparison dimensions Individual comparison: Establish an individual history database of 6 months or more, calculate the bias rate. If the bias rate is greater than or equal to 10% and requires physician evaluation, it is considered high sensitivity; if the bias rate is less than 5% and there is no need for intervention, it is considered low sensitivity. Population comparison: Construct a database of patients with similar characteristics. If a patient meets the following formula and is positively correlated with the risk of complications, the sensitivity is considered high; otherwise, it is considered low. Perturbation factor and data sensitivity correlation scheme 1) High-sensitivity data rules Factor range: Strong association group association strength: ; Strong association group association strength ; A pseudo-random algorithm for assigning values ​​to business metrics.

[0039] Disturbance type: Composite disturbance a. Multiplication perturbation: ( These are the coordinate values ​​after the co-geometric transformation. (Result after multiplication perturbation) b. 50% probability additive perturbation: ( For the final perturbation result, K is a magnitude coefficient set according to clinical precision, such as K=1 for heart rate and K=0.1 for blood pressure. 2) Low-sensitivity data rules Factor range: Correlation strength ; Correlation strength ; The values ​​are selected using a simplified pseudo-random algorithm.

[0040] Perturbation method: Multiplicative perturbation only ( These are the coordinate values ​​after the co-geometric transformation. (This is the result after multiplication perturbation).

[0041] 5. Offset update and synchronization When the adjustment period T is reached, the data group conversion process is temporarily frozen (≤10ms), and the new offset Δt is calculated. n ₊1=Δt n +ΔΔt; The new offset is synchronized to the timestamp generation module of all data in the group, and the unlocking process ensures that the time base is consistent.

[0042] 6. Verify the adjustment effect Synchronization verification: If the timestamp deviation of the data within the group after adjustment is ≤20ms, the verification is successful; otherwise, it will backtrack to the original offset, and the next cycle will be shortened to 1 / 2. Security verification: Analyze the pattern of the adjustment range in the last 10 adjustments. If a fixed pattern exists, increase the disturbance range to prevent it from being cracked.

[0043] Perform co-geometric transformation: Obtain the medical data to be converted and determine the data group to which the data belongs.

[0044] Based on the transformation rules for the corresponding data group in the collaborative transformation rule base and the allocated shared timestamp offset, a collaborative geometric transformation operation is performed on the medical data. A noise perturbation factor is introduced during the transformation process. This factor is dynamically generated based on data sensitivity to enhance the obfuscation effect without affecting data correlation.

[0045] 1. Medical Data Acquisition and Grouping Collect medical data to be converted (such as heart rate H and blood oxygen saturation S), and determine the strong correlation group (such as heart rate-blood oxygen group) of the data by matching data identifiers with the collaborative conversion rule base.

[0046] 2. Determining conversion parameters Ellipse equation parameters: For the master data (such as heart rate H), the major semi-axis a and minor semi-axis b of the ellipse are adaptively adjusted according to its fluctuation range (if the heart rate fluctuation is large, the difference between a and b increases), and the eccentricity is calculated. ; Spherical coordinate parameters: For correlated data (such as blood oxygen S), the spherical radius r is determined according to the collaborative rules (inversely proportional to e, r=k / e, k is a proportionality constant), and the polar angle θ and azimuth angle φ are generated by a pseudo-random number algorithm associated with the correlation coefficient.

[0047] In a specific embodiment, the association logic between the pseudo-random number algorithm and the association coefficient S (data set association strength) is as follows: Assuming the correlation strength S of the heart rate-blood oxygen data set is 0.95 (extremely strong correlation), the polar angle is generated using the "correlation strength weighted linear congruential method". (range [0, ]) and azimuth (range [0,2) The specific steps are as follows: (1) Pseudo-random number algorithm selection: linear congruential method The basic formula is: in: Let be the nth random number, a be the multiplier, c be the increment, and m be the modulus (all positive integers).

[0048] (2) Weighted adjustment of algorithm parameters by correlation strength S By incorporating the correlation strength S into the calculation of a and c, random number generation is bound to the correlation strength: Substitute S=0.95 into the calculated parameters: (3) Generate polar angle θ and azimuth angle Initialize seed : Retrieve the last 6 digits of the current data collection timestamp (e.g., if the timestamp is 1699999999999). =999999mod65536=999999-15 65536 = 999999 - 983040 = 16959); Generate the first random number : ; Mapped to polar angle θ: ; Generate a second random number : ; Mapped to azimuth : .

[0049] (4) Verification of the influence of correlation strength on random numbers If the correlation strength decreases to S=0.8 (strong correlation), recalculate the parameters: ; ; Same seed =16959 below, , mapping , compared with S=0.95 The significant differences demonstrate that the correlation strength can effectively control the random number generation results and ensure that the parameters match the data correlation.

[0050] Table 1 is an example of parameter types and core association logic: Table 1 Parameter type Calculation results Core Relational Logic ellipse semi-major axis a 45 Adaptive adjustment based on heart rate fluctuation range The minor axis b of the ellipse 31.5 The greater the fluctuation, the greater the difference from 'a'. ellipse eccentricity e ≈0.714 Calculated from a and b, reflecting the data dispersion spherical radius r ≈7.00 It is inversely proportional to e (r=k / e), where k is the precision constant. Polar angle θ ≈44.3° (when S=0.95) The pseudo-random number algorithm parameters (a, c) are adjusted by S weighting. Azimuth φ ≈279.9° (when S=0.95) Random number mapping range matching angle physical meaning This example clearly demonstrates the calculation of elliptical parameters for master data and spherical parameters for associated data. Furthermore, the binding of the pseudo-random number algorithm with the association coefficient ensures the correlation and security of parameters during collaborative transformation.

[0051] 3. Secondary verification of conversion parameters Call the historical conversion records to verify the rationality of the current parameters (such as whether the deviation of the inverse relationship between r and e is ≤5%, and whether the random generation of θ and φ conforms to the group association logic). If the verification fails, the parameters will be recalculated (e.g., the values ​​of a, b, or k will be corrected based on the historical average).

[0052] 4. Introduce noise disturbance factor The perturbation factor is dynamically generated based on the data sensitivity (e.g., heart rate sensitivity is higher than that of ordinary vital signs data), and the method of generating the perturbation factor is the same as that mentioned above. Apply small perturbations (e.g., x1'=x1×(1±perturbation factor)) to the elliptic transformation result (x1,y1) and the spherical mapping result (x2,y2,z2).

[0053] 5. Perform the conversion based on the shared timestamp offset. The conversion results of primary data and related data are appended with a dynamically adjusted shared timestamp offset (e.g., H conversion timestamp is t+Δt, S conversion timestamp is synchronized to t+Δt). Output the complete data after collaborative transformation (including perturbed coordinates and synchronization timestamps).

[0054] The expression for the transformation of the elliptic equation is as follows: ; Where a and b are the major and minor semi-axes of the ellipse, and the eccentricity of the ellipse is... The values ​​of a and b can be adaptively adjusted according to the fluctuation range of the data, making the elliptic equation more closely match the characteristics of different types of medical data.

[0055] The expression for spherical coordinate mapping is: Where r is the radius of the sphere. Polar angle, This is the azimuth angle. and The number is dynamically determined using a pseudo-random number generation algorithm, and the algorithm is correlated with the correlation coefficient of the data set. (Refer to...) Figure 4 .

[0056] Conversion performance evaluation and optimization: It also includes a conversion effect evaluation module, which quantitatively evaluates the collaborative geometric conversion effect by calculating the confusion, correlation retention, and recovery accuracy of the converted data, and optimizes the collaborative conversion rule base based on the evaluation results. (Refer to...) Figure 5 .

[0057] 1. Calculation of evaluation indicators Confusion level: The degree of difference between the transformed data and the original data is calculated using the information entropy algorithm (the higher the entropy value, the better the confusion effect). The information entropy value represents the degree of difference between the transformed data and the original data. The higher the value, the greater the confusion (difference) between the two sets of data, and the better the confusion effect of the transformation mechanism. n represents the total number of possible difference states in the comparison between the transformed data and the original data. This represents the i-th specific difference state in the comparison of two sets of data. Let represent the probability of the i-th difference state occurring, with a value in the range [0,1], and the sum of the probabilities of all difference states satisfies . . This represents a base-2 logarithmic operation, making the unit of information entropy a bit, which conforms to the common standard for information quantification in information theory.

[0058] Relevance retention: The consistency between the correlation parameters of the data within the group after conversion (such as the inverse relationship between e and r, timestamp synchronization deviation) and the data before conversion (≥90% is acceptable). Recovery accuracy: The data is restored using the inverse transformation algorithm, and the error rate between the restored value and the original value is calculated (≤5% is acceptable).

[0059] 2. Judgment of Evaluation Results If the confusion level is greater than or equal to the preset threshold (e.g., information entropy ≥ 7), the correlation retention is greater than or equal to 90%, and the recovery accuracy is greater than or equal to 95%, it is judged as "qualified"; If any indicator fails to meet the standard, it is judged as "unqualified" and the optimization process is triggered.

[0060] 3. Optimize feedback and rule base updates For "unqualified" results, identify the problematic steps: If the confusion level is insufficient, expand the adjustment range of the elliptic parameters a and b (e.g., expand the original a∈[5,10] to [3,12]) or increase the upper limit of the noise perturbation factor; If the correlation retention is low, adjust the ratio constant k of r and e (e.g., refit the value of k based on historical qualified data). If the recovery accuracy is low, optimize the inversion logic of θ and φ in the inverse transformation algorithm (such as introducing historical random number generation rules to assist in the recovery).

[0061] The optimized parameters are updated to the collaborative transformation rule base and applied to the next round of transformation.

[0062] 4. Iterative verification Re-execute the collaborative transformation using the updated rule base, repeating steps 1-3 until the evaluation result is "qualified"; Record the changes in metrics before and after each optimization to form an optimization log (used as a reference for generating rules for subsequent new data groups).

[0063] Hierarchical transformation of weakly correlated data groups: For weakly correlated data sets, a hierarchical transformation strategy is adopted, that is, the master data is transformed by elliptic equation or spherical coordinate mapping, while the correlated data is transformed by simple geometric transformations (such as translation and rotation) to balance transformation efficiency and security.

[0064] In another specific embodiment, taking heart rate and blood oxygen saturation data in ICU multi-sign monitoring as an example, the collaborative geometric transformation mechanism based on data correlation of the present invention will be specifically explained: Build a collaborative conversion rule base: Analysis confirmed that heart rate data and blood oxygen saturation data were a strongly correlated dataset.

[0065] When heart rate data is converted using an elliptic equation, the blood oxygen saturation data is automatically matched with complementary parameters for spherical coordinate mapping, and the eccentricity of the ellipse converted from heart rate data is inversely proportional to the radius of the sphere converted from blood oxygen saturation data. A secondary verification mechanism for the conversion parameters is also implemented, validating the rationality of the current parameters through historical conversion records of both methods.

[0066] The rule base supports dynamic expansion. If other medical data related to heart rate is added in the future, new collaborative conversion rules can be automatically generated.

[0067] Allocate shared timestamp offset: Assign a shared timestamp offset to the strongly correlated dataset of heart rate and blood oxygen saturation. This offset is adjusted hourly based on the data transmission frequency and correlation strength between the two.

[0068] Perform co-geometric transformation: Obtain heart rate data H and blood oxygen saturation data S at a specific moment.

[0069] The heart rate data H is transformed using an elliptic equation. Based on the fluctuation range of the heart rate data, the major semi-axis a=6 and the minor semi-axis b=4 of the ellipse are adaptively adjusted. The eccentricity of the ellipse is then calculated. .

[0070] According to the co-transformation rule, the spherical radius *r* of the transformed blood oxygen saturation data *S* is inversely proportional to the eccentricity *e* of the transformed heart rate data, i.e. (k is a proportionality constant, here we take k=5), then .

[0071] In spherical coordinate mapping and The number is dynamically determined using a pseudo-random number generation algorithm that correlates with the heart rate and blood oxygen saturation data sets. It is assumed that the generated numbers... , .

[0072] During the conversion process, a noise perturbation factor based on the sensitivity of heart rate and blood oxygen saturation data is introduced, with a assumed factor value of 0.02, to slightly perturb the converted data.

[0073] Combined with shared timestamp offset This completes the spherical coordinate mapping transformation of the blood oxygen saturation data S.

[0074] Conversion performance evaluation and optimization: After the conversion is complete, the conversion effect evaluation module calculates the confusion level, correlation retention, and recovery accuracy of the converted data. If the evaluation results show insufficient confusion level, the results are fed back to the collaborative conversion rule base to optimize the parameter adjustment range of the ellipse equation or the calculation ratio of the sphere radius.

[0075] Example of processing weakly related data groups: If there are other data that are weakly correlated with the heart rate data, such as a specific respiratory data segment, a hierarchical transformation strategy is adopted. The heart rate master data is still transformed using the elliptic equation, while the respiratory data segment is transformed by translation with a translation amount of 2 units to balance transformation efficiency and safety.

[0076] Further reference Figure 6 As an implementation of the above-described method, this application provides an embodiment of a medical data conversion system, which is similar to... Figure 1Corresponding to the method embodiments shown, the system can be specifically applied to various electronic devices.

[0077] refer to Figure 6 A medical data conversion system, comprising: The associated data group acquisition module 101 is configured to analyze the correlation between medical data and identify data groups with related relationships. The collaborative transformation rule setting module 102 is configured to set collaborative transformation rules for the data groups with related relationships, specifically including: Assign a unique identifier to each data group; One set of data is transformed using an elliptic equation, while another set of data is transformed using spherical coordinate mapping, and the eccentricity of the ellipse transformed by the first set of data is inversely proportional to the radius of the sphere transformed by the other set of data. Assign shared timestamp offsets to associated data in the data group; The conversion module 103 is configured to acquire the medical data to be converted, perform a collaborative geometric conversion operation on the medical data to be converted according to the collaborative conversion rules set in the collaborative conversion rule base and the allocated shared timestamp offset, and introduce a noise disturbance factor during the conversion process.

[0078] This application leverages the correlation between medical data to design a collaborative conversion rule base. This system creates mutual constraints among related data during conversion, forming a conversion chain that increases the complexity of reverse engineering and effectively enhances the overall data obfuscation strength. By assigning shared timestamp offsets to strongly correlated data groups and employing a dynamic update mechanism, the synchronization of collaborative conversion is ensured, guaranteeing the accuracy and consistency of data conversion and further improving security. A noise perturbation factor based on data sensitivity is introduced during the conversion process, enhancing the obfuscation effect without affecting data correlation. The rule base supports dynamic expansion, facilitating the automatic generation of conversion rules when adding new medical data types, improving the mechanism's adaptability and scalability. Secondary verification of conversion parameters prevents abnormal conversions and ensures conversion reliability. Support for multiple correlated data groups and user-defined rules expand the mechanism's application scope. The conversion effect evaluation module can quantitatively evaluate the conversion effect and provide feedback to optimize the rule base, continuously improving the mechanism. A tiered conversion strategy is adopted for weakly correlated data groups, balancing conversion efficiency and security. This avoids the problem of related data leakage after a single data conversion is compromised, making it suitable for multi-parameter joint monitoring scenarios (such as ICU multi-vital sign monitoring), providing strong protection for medical data security.

[0079] In another aspect, this application also provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the following... Figure 1 The method shown.

[0080] It should be noted that the computer-readable storage medium described in this application can be a computer-readable signal medium, a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable storage medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable storage medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0081] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0082] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0083] The specific embodiments of this application have been described above, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0084] In the description of this application, it should be understood that the terms "upper," "lower," "inner," "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are used only for the convenience of describing this application and for simplification, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The simple fact that certain measures are recited in mutually different dependent claims does not indicate that combinations of these measures cannot be used for improvement. Any reference signs in the claims should not be construed as limiting the scope.

Claims

1. A method for converting medical data, characterized in that, The method includes the following steps: Analyze the correlations between medical data to identify data groups with related relationships; To establish collaborative transformation rules for the data groups with related relationships, the rules specifically include: Assign a unique identifier to each data group; One set of data is transformed using an elliptic equation, while another set of data is transformed using spherical coordinate mapping, and the eccentricity of the ellipse transformed by the first set of data is inversely proportional to the radius of the sphere transformed by the other set of data. Assign shared timestamp offsets to the associated data in the data group; The medical data to be converted is obtained, and a collaborative geometric transformation operation is performed on the medical data to be converted according to the collaborative transformation rules set in the collaborative transformation rule base and the allocated shared timestamp offset, and a noise disturbance factor is introduced during the transformation process.

2. The medical data conversion method according to claim 1, characterized in that, It also includes the following steps: The shared timestamp offset is periodically adjusted based on data transmission frequency and association strength, including: The initial offset is calculated based on the initial acquisition time of the data in the data set and the average time difference of historical collaborative conversion. Real-time monitoring of data transmission frequency and quantification of correlation strength; determination of adjustment period based on data transmission frequency; and calculation of adjustment range based on correlation strength score. Based on data sensitivity, pseudo-random perturbations are introduced into the adjustment range to enhance security.

3. The medical data conversion method according to claim 2, characterized in that, It also includes the following steps: When the shared timestamp offset reaches the adjustment period, the conversion process of the data group is temporarily frozen, a new offset is calculated and synchronized to the timestamp generation module of all data in the data group, and the conversion process continues after the offset is unlocked.

4. The medical data conversion method according to claim 1, characterized in that, The noise disturbance factor introduced during the conversion process includes: Perturbation factors are generated for the ellipse transformation results and the spherical mapping results based on data sensitivity.

5. The medical data conversion method according to claim 1, characterized in that: The parameters of the spherical coordinate mapping include the spherical radius, polar angle, and azimuth angle. The polar angle and azimuth angle are dynamically determined by a pseudo-random number generation algorithm, and the pseudo-random number generation algorithm is associated with the correlation coefficient of medical data in the data set.

6. The medical data conversion method according to claim 1, characterized in that, It also includes the following steps: The historical conversion records of the associated data are retrieved, and the rationality of the current conversion parameters is verified using the historical conversion records. If the verification fails, the current conversion parameters are corrected based on the historical average.

7. The medical data conversion method according to claim 1, characterized in that, It also includes the following steps: The collaborative geometric transformation effect is quantitatively evaluated by calculating the confusion, correlation retention, and recovery accuracy among the transformed medical data, and the collaborative transformation rule base is optimized based on the evaluation results.

8. The medical data conversion method according to claim 1, characterized in that: The data sets with correlations include heart rate and blood oxygen saturation data sets, blood glucose and insulin level data sets, blood pressure and heart rate variability data sets, and respiratory rate and blood oxygen saturation fluctuation data sets.

9. A medical data conversion method according to claim 1, characterized in that, It also includes the following steps: Weakly correlated data groups of medical data are obtained, and a hierarchical transformation strategy is adopted. The main data in the weakly correlated data groups are transformed by elliptic equations or spherical coordinate mapping, while the correlated data are transformed by simple geometric transformation.

10. A medical data conversion system, characterized in that, The system includes: The associated data group acquisition module is configured to analyze the correlation between medical data and identify data groups with related relationships. The collaborative transformation rule setting module configures collaborative transformation rules for the data groups with related relationships, specifically including: Assign a unique identifier to each data group; One set of data is transformed using an elliptic equation, while another set of data is transformed using spherical coordinate mapping, and the eccentricity of the ellipse transformed by the first set of data is inversely proportional to the radius of the sphere transformed by the other set of data. Assign shared timestamp offsets to associated data in the data group; The conversion module is configured to acquire medical data to be converted, perform collaborative geometric conversion operations on the medical data to be converted according to the collaborative conversion rules set in the collaborative conversion rule base and the allocated shared timestamp offset, and introduce noise disturbance factors during the conversion process.

Images