Medical health data analysis method based on privacy protection and information security

Abstract

The invention discloses a medical health data analysis method based on privacy protection and information security, and relates to the technical field of health management, and the method comprises the steps: forming a data security management mechanism; obtaining corrected health data; analyzing and correcting the health data to form a disease distribution coefficient; predicting and forming a disease development trend based on the disease distribution coefficient; key preventive diseases are selected, and outbreak time points of the key preventive diseases are formed; analyzing to obtain a coping database of key preventive diseases; and the key preventive diseases are prevented in advance by using the data in the coping database. By correcting the abnormal value of the health data, forming a data security management mechanism and forming an outbreak time point for key disease prevention, abnormal data can be found in time, encryption can be carried out by using algorithms with different complexity degrees according to different privacy requirements of the data, excessive encryption is avoided, and the security of the health data is improved. And the more accurate outbreak time prediction can be carried out on the possible outbreak diseases.

Description

Technical Field

[0001] The present invention relates to the field of health management technology, and in particular to a medical health data analysis method based on privacy protection and information security. Background Art

[0002] With the rapid development of information technology, big data has become a national strategic resource. In the field of health care, the application of big data is becoming increasingly in-depth, bringing unprecedented changes to medical services, disease prevention, health management, etc. The collection and analysis of medical and health data provides strong support for the improvement of medical services and medical decision-making. When analyzing health data, data privacy protection and early prediction of infectious disease outbreaks are particularly important. Data leakage will damage the reputation of patients or cause financial losses to patients due to fraud. If infectious disease outbreaks are not handled properly, they are likely to cause public health problems.

[0003] Existing medical and health data analysis lacks judgment on data accuracy and disease outbreaks, and cannot prepare and respond in sufficient time in advance. At the same time, it does not adequately match data privacy protection and is prone to over-encryption of data. Over-encryption algorithms are more complex and consume more computing power. Summary of the invention

[0004] In order to solve the above technical problems, a medical and health data analysis method based on privacy protection and information security is provided. This technical solution solves the problems raised in the above background technology.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is: A medical health data analysis method based on privacy protection and information security, comprising: Form a data security management mechanism: Based on the privacy of health data, at least one data classification range is obtained. During the transmission and call of health data, the health data is encrypted and decrypted in a hierarchical manner according to the data classification range to which the health data belongs; Correcting outliers in health data to obtain corrected health data; Use data security management mechanisms to transmit and call corrected health data, analyze corrected health data, and form disease distribution coefficients; Based on the disease distribution coefficient, predict the development trend of the disease; Based on the development trend of diseases, select key preventable diseases and form outbreak time points of key preventable diseases; The analysis resulted in a response database for key preventable diseases, which includes data on disease cause discovery, medical resource allocation demand data, and drug development data; Use data from the response database to carry out early prevention of key preventable diseases.

[0006] Preferably, obtaining at least one data classification range based on the privacy of health data comprises the following steps: Health data is classified to obtain personnel information and medical data. Personnel information consists of name, address, age and contact information, and medical data consists of medical history, examination data, treatment plan and medication information. At least one hazard factor caused by the leakage of personal information and medical data based on big data; Obtain the extent of property loss and reputation loss caused by the harmful factors; Based on big data, obtain medical data leakage incidents involving reputation compensation, obtain the degree of reputation loss in medical data leakage incidents and the amount of compensation for reputation loss; The amount of compensation for reputation loss in a medical data breach incident is divided by the degree of reputation loss in the medical data breach incident to obtain a cost quantification factor for reputation loss; Obtain at least one hazard factor caused by health data leakage, and use the privacy comprehensive formula to calculate the privacy coefficient of the health data; Obtaining a value range of the privacy coefficient, evenly dividing the value range of the privacy coefficient to obtain at least one privacy interval, and using the privacy interval as a data classification range; The privacy comprehensive formula is as follows: , Among them, A is the privacy coefficient of health data, i is the subscript, and n is the number of harmful factors caused by health data leakage. is the cost quantification factor, is the reputation loss degree of the i-th hazard factor caused by health data leakage, is the property loss caused by the i-th hazard factor caused by health data leakage.

[0007] Preferably, the hierarchical encryption and decryption of the health data according to the data classification range to which the health data belongs comprises the following steps: The value of the midpoint of the data classification range is used as the feature loss cost; Based on big data, the acceptable loss cost of personnel is obtained. The acceptable loss cost is the cost that can be ignored within the economic scope of personnel. Based on the big data, at least one encryption and decryption method is obtained, and a probability of cracking the encryption and decryption method is obtained; Pair the encryption and decryption methods with the data classification range, satisfying that the probability of decryption of the encryption and decryption methods is equal to the acceptable loss cost divided by the characteristic loss cost of the paired data classification range; The health data is encrypted and decrypted using the encryption and decryption method corresponding to the data classification range to which the privacy coefficient of the health data belongs.

[0008] Preferably, the correction of abnormal values ​​of health data to obtain corrected health data comprises the following steps: For at least one inspection item involved in the inspection data, the inspection data is allocated according to the inspection item to form inspection classification data, wherein the inspection classification data corresponds to the inspection item; The inspection items with the lowest detection error rate are regarded as target inspection items, and the inspection items other than the target inspection items are regarded as non-target inspection items; Based on the historical data, a normal fluctuation range of the ratio of the historical data of the non-target examination items to the historical data of the target examination items of the same person suffering from the disease corresponding to the examination data is obtained; When the ratio of the detection classification data of the non-target inspection items to the detection classification data of the target inspection items exceeds the normal fluctuation range, the non-target inspection items shall be re-tested; The examination data is tested for matching with the medical history, treatment plan, and medication information. The matching test obtains the medical history, treatment plan, and medication information corresponding to the sample data that is consistent with the examination data, and compares it with the medical history, treatment plan, and medication information corresponding to the examination data; The medical history, treatment plan or medication information corresponding to the discrepant examination data is retrieved to obtain corrected health data.

[0009] Preferably, the analyzing and correcting health data to form a disease distribution coefficient comprises the following steps: The corrected health data is classified according to the disease category. Based on the hierarchical analysis method, the weights of the examination items in the examination data in the corrected health data are multiplied and accumulated by the weights of the examination items and the detection classification data of the examination items to obtain the severity of the corrected health data; The severity of the corrected health data in the disease category is accumulated to obtain the severity of the disease category; The severity of all disease categories is summarized as the disease distribution coefficient.

[0010] Preferably, the prediction of the disease development trend based on the disease distribution coefficient comprises the following steps: At each preset time interval, the severity of the disease category is obtained; The severity of the disease category is paired with the acquired time point and fitted to obtain a development fitting function, wherein the time point is the independent variable and the severity of the disease category is the dependent variable; The development fitting function is differentiated to obtain the first-order development derivative function, and the first-order development derivative function is differentiated to obtain the second-order development derivative function; The first-order derivative function and second-order derivative function of disease category development are summarized as the disease development trend.

[0011] Preferably, the step of selecting key preventable diseases based on the disease development trend comprises the following steps: At the initial moment and the current moment of the analysis, the proportion of time points at which the first-order derivative function of development is greater than 0 and the second-order derivative function of development is greater than 0 is counted as the proportion identification value; Uniformly select at least one sampling point in the interval (0, 1), and under the condition that the proportion recognition value is equal to the value at the sampling point, integrate the development second-order derivative function within the range formed by the initial moment and the current moment to obtain the second-order recognition value; The value at the sampling point corresponding to the second-order identification value closest to 0 is used as the discriminant value of the proportion identification value; Diseases in the disease categories corresponding to the recognition values ​​of the proportions exceeding the discrimination values ​​are taken as key preventable diseases.

[0012] Preferably, forming the outbreak time point of key preventable diseases comprises the following steps: Calculate the current growth rate of the second-order derivative function of the key preventable disease, and set the subsequent growth rates of the second-order derivative function of the development to the current growth rate; Obtain the minimum value of the second-order derivative function of the development within a time range with a preset distance from the current time as the characteristic value; The result of subtracting the characteristic value from 0 is divided by the current growth rate and superimposed with the current time to obtain the outbreak time point.

[0013] Preferably, the analysis to obtain a database of key disease prevention responses includes the following steps: Extract etiological predisposing factors from medical history of modified health data containing disease categories of key preventable diseases and summarize them into etiological discovery data; Extract the medical resources required for treatment from the treatment plans of the modified health data of the disease categories containing the key preventable diseases, and summarize them into the medical resource allocation demand data; Drug ingredients that have a therapeutic effect are extracted from the medication information of the revised health data of disease categories that include key preventable diseases and summarized as drug research and development data.

[0014] Preferably, the method of using the data in the response database to carry out early prevention of key preventable diseases comprises the following steps: Based on the data in the response database, cause suppression plans, medical configuration adjustment plans and drug development plans are formed.

[0015] Compared with the prior art, the present invention has the following beneficial effects: By correcting outliers in health data, forming a data security management mechanism, and establishing outbreak time points for key disease prevention, abnormal data can be discovered in a timely manner and the data can be retrieved to ensure data accuracy, thereby ensuring the credibility of subsequent analysis. When protecting data privacy, encryption can be performed using algorithms of varying complexity based on different data privacy requirements to avoid over-encryption, thereby controlling the computing power used for encryption. In addition, diseases can be screened to make more accurate predictions of the outbreak time of diseases that may break out, and data for prevention of diseases that may break out can be provided based on the analysis results, thereby enabling targeted prevention in advance. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flowchart of a medical health data analysis method based on privacy protection and information security of the present invention; Figure 2 A schematic diagram of a process for obtaining at least one data classification range based on the privacy of health data of the present invention; Figure 3 It is a schematic diagram of the process of hierarchical encryption and decryption of health data according to the data classification range to which the health data belongs in the present invention; Figure 4 A schematic diagram of a process of correcting abnormal values ​​in health data to obtain corrected health data according to the present invention; Figure 5 A schematic diagram of the process of analyzing and correcting health data to form a disease distribution coefficient according to the present invention; Figure 6 A schematic diagram of a process for predicting the development trend of a disease based on the disease distribution coefficient of the present invention; Figure 7 A schematic diagram of a process for selecting key preventable diseases based on the disease development trend of the present invention; Figure 8 A schematic diagram of a process for forming outbreak time points of key preventable diseases according to the present invention; Fig. 9 The present invention is a flowchart of obtaining a response database for key preventable diseases through analysis. DETAILED DESCRIPTION

[0017] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0018] Reference Figure 1 As shown, a medical health data analysis method based on privacy protection and information security includes: Form a data security management mechanism: Based on the privacy of health data, at least one data classification range is obtained. During the transmission and call of health data, the health data is encrypted and decrypted in a hierarchical manner according to the data classification range to which the health data belongs; Correcting outliers in health data to obtain corrected health data; Use data security management mechanisms to transmit and call corrected health data, analyze corrected health data, and form disease distribution coefficients; Based on the disease distribution coefficient, predict the development trend of the disease; Based on the development trend of diseases, select key preventable diseases and form outbreak time points of key preventable diseases; The analysis resulted in a response database for key preventable diseases, which includes data on disease cause discovery, medical resource allocation demand data, and drug development data; Use data from the response database to carry out early prevention of key preventable diseases.

[0019] Reference Figure 2 As shown, based on the privacy of health data, obtaining at least one data classification range includes the following steps: Health data is classified to obtain personnel information and medical data. Personnel information consists of name, address, age and contact information, and medical data consists of medical history, examination data, treatment plan and medication information. At least one hazard factor caused by the leakage of personal information and medical data based on big data; Obtain the extent of property loss and reputation loss caused by the harmful factors; Based on big data, obtain medical data leakage incidents involving reputation compensation, obtain the degree of reputation loss in medical data leakage incidents and the amount of compensation for reputation loss; The amount of compensation for reputation loss in a medical data breach incident is divided by the degree of reputation loss in the medical data breach incident to obtain a cost quantification factor for reputation loss; Obtain at least one hazard factor caused by health data leakage, and use the privacy comprehensive formula to calculate the privacy coefficient of the health data; Obtaining a value range of the privacy coefficient, evenly dividing the value range of the privacy coefficient to obtain at least one privacy interval, and using the privacy interval as a data classification range; The privacy comprehensive formula is as follows: , Among them, A is the privacy coefficient of health data, i is the subscript, and n is the number of harmful factors caused by health data leakage. is the cost quantification factor, is the reputation loss degree of the i-th hazard factor caused by health data leakage, is the property loss caused by the i-th hazard factor caused by health data leakage.

[0020] When dividing data to form a data classification range, it mainly depends on the cost caused by data leakage. When data is leaked, scammers may take advantage of the patient's desire to seek medical treatment to commit fraud, thereby causing economic losses to the patient. In addition, the disease that the patient may have is highly private. Once known by others, it will cause damage to the patient's reputation, but the loss of reputation will not be directly reflected in the economy. Therefore, in order to evaluate the loss of reputation, it is necessary to convert the reputation loss, thereby forming a cost quantification factor, and then the privacy coefficient of the health data can be calculated, and based on this, a data classification range is formed.

[0021] Reference Figure 3 As shown, according to the data classification range to which the health data belongs, hierarchical encryption and decryption of health data includes the following steps: The value of the midpoint of the data classification range is used as the feature loss cost; Based on big data, the acceptable loss cost of personnel is obtained. The acceptable loss cost is the cost that can be ignored within the economic scope of personnel. Based on the big data, at least one encryption and decryption method is obtained, and a probability of cracking the encryption and decryption method is obtained; Pair the encryption and decryption methods with the data classification range, satisfying that the probability of decryption of the encryption and decryption methods is equal to the acceptable loss cost divided by the characteristic loss cost of the paired data classification range; The health data is encrypted and decrypted using the encryption and decryption method corresponding to the data classification range to which the privacy coefficient of the health data belongs.

[0022] For example, when a person's income is 100, the acceptable cost of loss will not be higher than 10. When it occurs, it will not have much impact on the person himself. There is a certain probability of encrypted data leakage. Therefore, the actual impact is equivalent to the multiplication of the characteristic loss cost and the decryption probability. Then, using the above-mentioned encryption method, the probability-weighted loss caused by the leakage is very close to the acceptable cost of loss, and will not have a greater impact on the person. Therefore, such encryption can protect data privacy. According to common sense, the greater the probability of decryption, the greater the complexity of the encryption and decryption method, and thus the more time it takes to encrypt and decrypt. Therefore, using the above-mentioned targeted encryption and decryption methods can take into account both privacy protection and computing power control.

[0023] Reference Figure 4 As shown, correcting abnormal values ​​of health data to obtain corrected health data includes the following steps: For at least one inspection item involved in the inspection data, the inspection data is allocated according to the inspection item to form inspection classification data, wherein the inspection classification data corresponds to the inspection item; The inspection items with the lowest detection error rate are regarded as target inspection items, and the inspection items other than the target inspection items are regarded as non-target inspection items; Based on the historical data, a normal fluctuation range of the ratio of the historical data of the non-target examination items to the historical data of the target examination items of the same person suffering from the disease corresponding to the examination data is obtained; When the ratio of the detection classification data of the non-target inspection items to the detection classification data of the target inspection items exceeds the normal fluctuation range, the non-target inspection items shall be re-tested; The examination data is tested for matching with the medical history, treatment plan, and medication information. The matching test obtains the medical history, treatment plan, and medication information corresponding to the sample data that is consistent with the examination data, and compares it with the medical history, treatment plan, and medication information corresponding to the examination data; The medical history, treatment plan or medication information corresponding to the discrepant examination data is retrieved to obtain corrected health data.

[0024] Because under the same disease, there is a certain internal relationship between the detection classification data. Although there is a certain fluctuation in the proportional relationship between them, it will fluctuate within a range. However, when the fluctuation is too large, it can be judged that the data of the item is wrong. However, when obtaining the proportion, an item is required as a reference, but the data of this item may also be wrong. In order to reduce the impact of this situation, the inspection item with the lowest detection error rate is used as the target inspection item, and the probability of error will be greatly reduced.

[0025] Reference Figure 5 As shown, analyzing and correcting health data to form disease distribution coefficients includes the following steps: The corrected health data is classified according to the disease category. Based on the hierarchical analysis method, the weights of the examination items in the examination data in the corrected health data are multiplied and accumulated by the weights of the examination items and the detection classification data of the examination items to obtain the severity of the corrected health data; The severity of the corrected health data in the disease category is accumulated to obtain the severity of the disease category; The severity of all disease categories is summarized as the disease distribution coefficient.

[0026] The disease category includes a lot of corrected health data, and each corrected health data corresponds to a severity. Therefore, it is necessary to accumulate the severity of the corrected health data in the disease category. When the severity of the corrected health data is calculated, it is completed by weighted accumulation of the detection classification data. When the corrected health data is normal, the detection classification data will fluctuate at the normal value, and the weighted summary result will also fluctuate at a certain value. However, when it is abnormal, the detection classification data will deviate significantly from the normal value. As a result, the severity of the corrected health data is very different between abnormal and normal situations. Therefore, the severity of the corrected health data calculated using this method can reflect the abnormality of the corrected health data.

[0027] Reference Figure 6 As shown, based on the disease distribution coefficient, predicting the disease development trend includes the following steps: At each preset time interval, the severity of the disease category is obtained; The severity of the disease category is paired with the acquired time point and fitted to obtain a development fitting function, wherein the time point is the independent variable and the severity of the disease category is the dependent variable; The development fitting function is differentiated to obtain the first-order development derivative function, and the first-order development derivative function is differentiated to obtain the second-order development derivative function; The first-order derivative function and second-order derivative function of disease category development are summarized as the disease development trend.

[0028] Reference Figure 7 As shown, based on the disease development trend, selecting key preventable diseases includes the following steps: At the initial moment and the current moment of the analysis, the proportion of time points at which the first-order derivative function of development is greater than 0 and the second-order derivative function of development is greater than 0 is counted as the proportion identification value; Uniformly select at least one sampling point in the interval (0, 1), and under the condition that the proportion recognition value is equal to the value at the sampling point, integrate the development second-order derivative function within the range formed by the initial moment and the current moment to obtain the second-order recognition value; The value at the sampling point corresponding to the second-order identification value closest to 0 is used as the discriminant value of the proportion identification value; Diseases in the disease categories corresponding to the recognition values ​​of the proportions exceeding the discrimination values ​​are taken as key preventable diseases.

[0029] Here, the key preventable diseases are mainly identified through the development of the first-order derivative function and the development of the second-order derivative function. The time point corresponding to the proportion identification value is the time point when the development of the first-order derivative function is greater than 0 and the development of the second-order derivative function is greater than 0. The development of the first-order derivative function will continue to grow in a short period of time, which will cause the development of the fitting function to continue to grow in a short period of time, which means that the severity of the disease is increasing. For other situations, such as the time point when the development of the first-order derivative function is less than 0 and the development of the second-order derivative function is greater than 0, although the development of the first-order derivative function will grow, because it is less than 0, the development of the fitting function will not grow in a short period of time, and therefore, the severity of the disease will not increase. At the time point when the first-order derivative of development is greater than 0 and the second-order derivative of development is less than 0, since the first-order derivative of development is decreasing, the growth trend of the development fitting function in a short period of time is decreasing. Therefore, the severity of the disease will weaken. Therefore, there is no need to consider it. For the time point when the first-order derivative of development is less than 0 and the second-order derivative of development is less than 0, it is obviously not necessary to consider it. Therefore, only the proportion of time points when the first-order derivative of development is greater than 0 and the second-order derivative of development is greater than 0 is considered, and the proportion identification value is obtained. The corresponding time point of the proportion identification value provides the main disease growth point. Therefore, considering it can more accurately form the outbreak time point of the key disease prevention.

[0030] Reference Figure 8 As shown, forming the outbreak time points for key disease prevention includes the following steps: Calculate the current growth rate of the second-order derivative function of the key preventable disease, and set the subsequent growth rates of the second-order derivative function of the development to the current growth rate; Obtain the minimum value of the second-order derivative function of the development within a time range with a preset distance from the current time as the characteristic value; The result of subtracting the characteristic value from 0 is divided by the current growth rate and superimposed with the current time to obtain the outbreak time point.

[0031] Reference Fig. 9 As shown, the analysis of the database of key disease prevention responses includes the following steps: Extract etiological predisposing factors from medical history of modified health data containing disease categories of key preventable diseases and summarize them into etiological discovery data; Extract the medical resources required for treatment from the treatment plans of the modified health data of the disease categories containing the key preventable diseases, and summarize them into the medical resource allocation demand data; Drug ingredients that have a therapeutic effect are extracted from the medication information of the revised health data of disease categories that include key preventable diseases and summarized as drug research and development data.

[0032] This is to provide data assistance for subsequent prevention.

[0033] Using data from the response database to prevent priority diseases in advance includes the following steps: Based on the data in the response database, cause suppression plans, medical configuration adjustment plans and drug development plans are formed.

[0034] Furthermore, the present solution also proposes a storage medium on which a computer-readable program is stored, and when the computer-readable program is called, the above-mentioned medical and health data analysis method based on privacy protection and information security is executed.

[0035] It is understandable that the storage medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid state drive (SSD).

[0036] To sum up, the advantages of the present invention are: by correcting abnormal values ​​of health data, forming a data security management mechanism and forming outbreak time points for key disease prevention, abnormal data can be discovered in time, and the data can be retrieved to ensure the accuracy of the data, thereby ensuring the credibility of subsequent analysis. When protecting data privacy, it can be encrypted using algorithms of different complexity according to different privacy requirements of the data to avoid excessive encryption, thereby controlling the computing power used for encryption. In addition, diseases can be screened, and the outbreak time of diseases that may break out can be predicted more accurately, and data for prevention can be provided for diseases that may break out based on the analysis results, so that targeted prevention can be carried out in advance.

[0037] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A medical health data analysis method based on privacy protection and information security, characterized in that: include: Form a data security management mechanism: Based on the privacy of health data, at least one data classification range is obtained. During the transmission and call of health data, the health data is encrypted and decrypted in a hierarchical manner according to the data classification range to which the health data belongs; Correcting outliers in health data to obtain corrected health data; Use data security management mechanisms to transmit and call corrected health data, analyze corrected health data, and form disease distribution coefficients; Based on the disease distribution coefficient, predict the development trend of the disease; Based on the development trend of diseases, select key preventable diseases and form outbreak time points of key preventable diseases; The analysis resulted in a response database for key preventable diseases, which includes data on disease cause discovery, medical resource allocation demand data, and drug development data; Use data from the response database to carry out early prevention of key preventable diseases.

2. A medical health data analysis method based on privacy protection and information security according to claim 1, characterized in that: The step of obtaining at least one data classification range based on the privacy of health data comprises the following steps: Classify the health data to obtain personnel information and medical data. The personnel information consists of name, address, age and contact information, and the medical data consists of medical history, examination data, treatment plan and medication information. At least one hazard factor caused by the leakage of personal information and medical data based on big data; Obtain the extent of property loss and reputation loss caused by the harmful factors; Based on big data, obtain medical data leakage incidents involving reputation compensation, obtain the degree of reputation loss in medical data leakage incidents and the amount of compensation for reputation loss; The amount of compensation for reputation loss in a medical data breach incident is divided by the degree of reputation loss in the medical data breach incident to obtain a cost quantification factor for reputation loss; Obtain at least one hazard factor caused by health data leakage, and use the privacy comprehensive formula to calculate the privacy coefficient of the health data; Obtaining a value range of the privacy coefficient, evenly dividing the value range of the privacy coefficient to obtain at least one privacy interval, and using the privacy interval as a data classification range; The privacy comprehensive formula is as follows: , Among them, A is the privacy coefficient of health data, i is the subscript, and n is the number of harmful factors caused by health data leakage. is the cost quantification factor, is the reputation loss degree of the i-th hazard factor caused by health data leakage, is the property loss caused by the i-th hazard factor caused by health data leakage.

3. A medical health data analysis method based on privacy protection and information security according to claim 2, characterized in that: The hierarchical encryption and decryption of health data according to the data classification range to which the health data belongs includes the following steps: The value of the midpoint of the data classification range is used as the feature loss cost; Based on big data, the acceptable loss cost of personnel is obtained. The acceptable loss cost is the cost that can be ignored within the economic scope of personnel. Based on the big data, at least one encryption and decryption method is obtained, and a probability of cracking the encryption and decryption method is obtained; Pair the encryption and decryption methods with the data classification range, satisfying that the probability of decryption of the encryption and decryption methods is equal to the acceptable loss cost divided by the characteristic loss cost of the paired data classification range; The health data is encrypted and decrypted using the encryption and decryption method corresponding to the data classification range to which the privacy coefficient of the health data belongs.

4. A medical health data analysis method based on privacy protection and information security according to claim 3, characterized in that: The correction of abnormal values ​​in the health data to obtain the corrected health data comprises the following steps: For at least one inspection item involved in the inspection data, the inspection data is allocated according to the inspection item to form inspection classification data, wherein the inspection classification data corresponds to the inspection item; The inspection items with the lowest detection error rate are regarded as target inspection items, and the inspection items other than the target inspection items are regarded as non-target inspection items; Based on the historical data, a normal fluctuation range of the ratio of the historical data of the non-target examination items to the historical data of the target examination items of the same person suffering from the disease corresponding to the examination data is obtained; When the ratio of the detection classification data of the non-target inspection items to the detection classification data of the target inspection items exceeds the normal fluctuation range, the non-target inspection items shall be re-tested; The examination data is tested for matching with the medical history, treatment plan, and medication information. The matching test obtains the medical history, treatment plan, and medication information corresponding to the sample data that is consistent with the examination data, and compares it with the medical history, treatment plan, and medication information corresponding to the examination data; The medical history, treatment plan or medication information corresponding to the discrepant examination data is retrieved to obtain corrected health data.

5. A medical health data analysis method based on privacy protection and information security according to claim 4, characterized in that: The analysis and correction of health data to form disease distribution coefficients includes the following steps: The corrected health data is classified according to the disease category. Based on the hierarchical analysis method, the weights of the examination items in the examination data in the corrected health data are multiplied and accumulated by the weights of the examination items and the detection classification data of the examination items to obtain the severity of the corrected health data; The severity of the corrected health data in the disease category is accumulated to obtain the severity of the disease category; The severity of all disease categories is summarized as the disease distribution coefficient.

6. A medical health data analysis method based on privacy protection and information security according to claim 5, characterized in that: The method of predicting the disease development trend based on the disease distribution coefficient comprises the following steps: At each preset time interval, the severity of the disease category is obtained; The severity of the disease category is paired with the acquired time point and fitted to obtain a development fitting function, wherein the time point is the independent variable and the severity of the disease category is the dependent variable; The development fitting function is differentiated to obtain the first-order development derivative function, and the first-order development derivative function is differentiated to obtain the second-order development derivative function; The first-order derivative function and second-order derivative function of disease category development are summarized as the disease development trend.

7. A medical health data analysis method based on privacy protection and information security according to claim 6, characterized in that: The selection of key preventable diseases based on disease development trends includes the following steps: At the initial moment and the current moment of the analysis, the proportion of time points at which the first-order derivative function of development is greater than 0 and the second-order derivative function of development is greater than 0 is counted as the proportion identification value; Uniformly select at least one sampling point in the interval (0, 1), and under the condition that the proportion recognition value is equal to the value at the sampling point, integrate the development second-order derivative function within the range formed by the initial moment and the current moment to obtain the second-order recognition value; The value at the sampling point corresponding to the second-order identification value closest to 0 is used as the discriminant value of the proportion identification value; Diseases in the disease categories corresponding to the recognition values ​​of the proportions exceeding the discrimination values ​​are taken as key preventable diseases.

8. A medical health data analysis method based on privacy protection and information security according to claim 7, characterized in that: The forming of the outbreak time point of the key preventable disease comprises the following steps: Calculate the current growth rate of the second-order derivative function of the key preventable disease, and set the subsequent growth rates of the second-order derivative function of the development to the current growth rate; Obtain the minimum value of the second-order derivative function of the development within a time range with a preset distance from the current time as the characteristic value; The result of subtracting the characteristic value from 0 is divided by the current growth rate and superimposed with the current time to obtain the outbreak time point.

9. A medical health data analysis method based on privacy protection and information security according to claim 8, characterized in that: The analysis to obtain the key disease prevention response database includes the following steps: Extract etiological predisposing factors from medical history of modified health data containing disease categories of key preventable diseases and summarize them into etiological discovery data; Extract the medical resources required for treatment from the treatment plans of the modified health data of the disease categories containing the key preventable diseases, and summarize them into the medical resource allocation demand data; Drug ingredients that have a therapeutic effect are extracted from the medication information of the revised health data of disease categories that include key preventable diseases and summarized as drug research and development data.

10. A medical health data analysis method based on privacy protection and information security according to claim 9, characterized in that: The use of data in the response database to prevent key preventable diseases in advance includes the following steps: Based on the data in the response database, cause suppression plans, medical configuration adjustment plans and drug development plans are formed.

Images