AIDS prediction method based on incremental neural network model

Abstract

The invention relates to the technical field of medical treatment, and discloses an AIDS prediction method based on an incremental neural network model, and the method comprises the following steps: S1, collecting HIV monitoring follow-up visit dynamic data, and carrying out the preprocessing of the data, S2, carrying out the neural network integration of part of target data in different time periods, and carrying out the prediction of the target data; the HIV disease prediction method comprises the following steps of: S1, acquiring a whole target data set, S3, sequentially carrying out incremental circulation on the whole target data set and outputting an incremental data set, S4, establishing an HIV disease prediction model according to the incremental data set, S5, carrying out key setting on a prediction result and outputting the key, carrying out integrated processing on part of target data through neural network integration to obtain the whole target data set, and carrying out HIV disease prediction on the whole target data set. The whole target data set is subjected to incremental circulation processing through the incremental neural network model, the incremental data set is calculated and output based on the confidence coefficient, the incremental data set serves as a variable to be input into the HIV prediction model to obtain the prediction result, and the accuracy and efficiency of HIV attack prediction are improved.

Description

Technical Field

[0001] The present invention relates to the field of medical technology, and more specifically to an AIDS prediction method based on an incremental neural network model. Background Art

[0002] Artificial intelligence and machine learning are important components of contemporary information technology. They realize the automatic processing and analysis of data by simulating the human learning and decision-making process. In the medical field, artificial intelligence and machine learning technologies are widely used in disease prediction, diagnosis, treatment and prevention, greatly improving the efficiency and accuracy of medical services.

[0003] A neural network is a computational model that simulates the structure and function of a biological nervous system. It consists of multiple neurons and transmits and processes information by connecting weights and activation functions. In the field of machine learning, neural networks are widely used. The incremental neural network model is a special form of neural network that can update and optimize the model by adding new data without retraining the entire model. This model is particularly suitable for processing large-scale, dynamically changing data sets.

[0004] There are some limitations in the existing HIV incidence prediction technology, such as the complexity of data processing, the accuracy of the prediction model, and the security of the prediction results. Traditional prediction methods often rely on static data sets and ignore the dynamic changes of HIV infection. However, the incremental neural network model can realize the monitoring of dynamic data. In addition, due to the sensitivity of HIV data, the privacy protection of prediction results has also become an urgent problem to be solved. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides an AIDS prediction method based on an incremental neural network model to solve the problems existing in the above-mentioned background technology.

[0006] The present invention provides the following technical solution: an AIDS prediction method based on an incremental neural network model, comprising the following steps:

[0007] Step S1: Collect HIV monitoring and follow-up dynamic data and pre-process the data: obtain some HIV monitoring and follow-up dynamic data of different time periods in the database of "Basic Information System for AIDS Prevention and Control of China Center for Disease Control and Prevention", and pre-process the data to obtain some target data;

[0008] Step S2: Performing neural network integration on partial target data of different time periods: Performing integration processing on partial target data of different time periods through neural network integration to obtain the overall target data set in the database;

[0009] Step S3: performing incremental cycles on the overall target data set in sequence and outputting the incremental data set: performing incremental cycle processing on the overall target data set through the incremental neural network model, and outputting the incremental data set based on the confidence calculation;

[0010] Step S4: Establishing an HIV prediction model based on the incremental data set: inputting the incremental data set as a variable into the HIV prediction model, and the HIV prediction model outputs a prediction result;

[0011] Step S5: Setting a key for the prediction result and outputting it: Using an encryption algorithm to set a key, encrypt the prediction result on this basis, and output the encrypted prediction result to the user end to complete the HIV incidence prediction.

[0012] Preferably, when collecting HIV monitoring and follow-up dynamic data and preprocessing the data, the specific contents of obtaining HIV monitoring and follow-up dynamic data in the database and preprocessing the data to obtain the target data are: the HIV monitoring and follow-up dynamic data are sample test data of different categories, including epidemiological survey results of the tester, CD4 + T cell count, viral load and drug resistance detection, and preprocessing the data to obtain partial target data.

[0013] Preferably, the specific content of performing neural network integration on partial target data in different time periods is:

[0014] Part of the target data set obtained at different time periods is represented as S t , t=1, 2, 3, ..., T, where t represents the number of the time period for collecting partial target data, and the S t It is the dynamic data of some HIV monitoring and follow-up in different time periods in the database;

[0015] The overall target data set in the database is S, and the partial target data set S is extracted through neural network integration t In the knowledge, we can obtain the knowledge in the overall data set S.

[0016] Preferably, the steps for acquiring the knowledge in the overall target data set S are as follows:

[0017] Step S01: In the partial target data set S t In the example, the set of samples of the nth class is represented as S t (n) , the mean of the nth class of samples is u t (n) , in the overall target data set S, the set of samples of the nth class is represented as S (n) , the mean of the n-th class sample is represented by u (n) ;

[0018] Step S02: Using a partial target dataset S t The mean value u of the nth class sample in t (n) Calculate the mean u of the nth class sample in the overall target data set S (n) , the calculation formula is: where u (n) Represents the mean of the nth class of samples in the overall target data set S, m t (n) Represents the set S of the nth class samples in part of the target data set t (n) The number of samples in m (n) Represents the set S of the nth class of samples in the overall target data set (n) The number of samples in ;

[0019] Step S03: The calculated mean value u of the nth class sample in the overall target data set S is (n) As the target data of the nth class sample;

[0020] Repeat the above steps to obtain the target data of each type of sample in the overall target data set S, where S = {u (1) ,u (2) , ..., u (n)}.

[0021] Preferably, the specific contents of performing incremental cycles on the entire target data set in sequence and outputting incremental data are as follows:

[0022] Perform incremental calculation and judgment on the target data of each type of samples in the overall target data set S;

[0023] Incremental calculation is performed on the unselected target data according to the incremental calculation judgment result to form an incremental data set of the target data.

[0024] Preferably, the specific contents of performing incremental calculation and judgment on the target data of each type of samples in the overall target data set S are as follows:

[0025] Calculate the proportion of each type of samples in the overall target data set S. The calculation formula is: Where L (n) Represents the proportion of each type of samples in the overall target data set S, m (n) represents the number of samples of the nth class in the overall target data set, and m represents the total number of samples of each class in the overall target data set S;

[0026] Calculate the confidence of the target data of each type of sample in the overall target data set S. The calculation formula is: Where Z (n)Represents the confidence of the target data of each type of sample in the overall target data set S, u (n) represents the mean of the nth class samples in the overall target data set S, maxu t (n) Represents part of the target dataset S t The maximum value of the mean of the nth class samples in , Z represents the normalization parameter;

[0027] The confidence Z of the target data of each type of sample in the overall target data set S (n) Compared with the preset threshold, if the confidence level Z of the target data (n) If the confidence level Z of the target data is greater than the preset threshold, the target data is selected. (n) If the value is less than or equal to the preset threshold, it is determined that the target data is not selected, and an incremental calculation is performed on the unselected target data.

[0028] Preferably, the incremental calculation is performed on the unselected target data according to the incremental calculation judgment result to form an incremental data set of the target data, and the calculation formula of the incremental calculation is: Where Δu (n) represents the target data after incremental calculation, λ represents the incremental factor, maxu t (n) Represents part of the target dataset S t The maximum value of the mean of the n-th class samples, minu t (n) Represents part of the target dataset S t The minimum value of the mean of the nth class of samples in the algorithm is used to calculate the confidence of the target data after the incremental calculation. The algorithm is repeated until the target data is selected, and the selected target data after the incremental calculation is converted into the confidence value of the target data. (n) Replace the original target data u (n) , forming an incremental data set ΔS of the target data, ΔS = {Δu (1) , Δu (2) , ..., Δu (n)}, if the original target data u (n) If no incremental calculation is performed, Δu (n) =u (n) .

[0029] Preferably, the specific content of establishing the HIV prediction model based on the incremental data set is: inputting the incremental data set as a variable into the HIV prediction model, and the expression of the HIV prediction model is: Where Y represents the prediction result of HIV prediction model, U (n) Represents the standard value of each type of sample. If the standard value is an interval value, the middle value of the interval is taken, ω (n) Represents the weight value of each type of samples.

[0030] Preferably, the specific content of key setting and outputting the prediction result is: encrypting the prediction result of the HIV prediction model, when the prediction result of the HIV prediction model is greater than or equal to the preset risk value, the prediction result is a high risk of disease, when the prediction result of the HIV prediction model is less than the preset risk value, the prediction result is a low risk of disease, and outputting the result to the user end, the user end decrypts the prediction result of the HIV prediction model through the key to complete the prediction of HIV.

[0031] Technical effects and advantages of the present invention:

[0032] The present invention is provided with step S1: collecting HIV monitoring follow-up dynamic data and preprocessing the data, step S2: performing neural network integration on partial target data of different time periods, step S3: performing incremental cycles on the overall target data set in turn and outputting the incremental data set, step S4: establishing an HIV prediction model according to the incremental data set, and step S5: performing key setting on the prediction result and outputting it, an AIDS prediction method based on an incremental neural network model improves the accuracy and efficiency of HIV prediction;

[0033] Through neural network integration, partial target data in different time periods are integrated and processed to obtain the overall target data set in the database. The overall target data set is incrementally and cyclically processed according to the incremental neural network model, and the incremental data set is output based on confidence calculation to realize dynamic change monitoring of HIV monitoring and follow-up dynamic data, more accurately reflect the potential risk of HIV infection and thus improve the accuracy of prediction. The introduction of incremental cyclic processing and confidence calculation enables the model to be dynamically adjusted and optimized, and has good robustness to outliers and noise in the data set, thereby improving the stability and reliability of the prediction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a flowchart of an AIDS prediction method based on an incremental neural network model. DETAILED DESCRIPTION

[0035] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. In addition, the forms of the various structures described in the following embodiments are only examples. The AIDS prediction method based on the incremental neural network model involved in the present invention is not limited to the various structures described in the following embodiments. All other implementations obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0036] like Figure 1As shown, the present invention provides an AIDS prediction method based on an incremental neural network model, comprising the following steps:

[0037] Step S1: Collect HIV monitoring and follow-up dynamic data and pre-process the data: obtain some HIV monitoring and follow-up dynamic data of different time periods in the "Basic Information System for AIDS Prevention and Control of China Center for Disease Control and Prevention" database (hereinafter referred to as "database"), and pre-process the data to obtain some target data;

[0038] Step S2: Performing neural network integration on partial target data of different time periods: Performing integration processing on partial target data of different time periods through neural network integration to obtain the overall target data set in the database;

[0039] Step S3: performing incremental cycles on the overall target data set in sequence and outputting the incremental data set: performing incremental cycle processing on the overall target data set through the incremental neural network model, and outputting the incremental data set based on the confidence calculation;

[0040] Step S4: Establishing an HIV prediction model based on the incremental data set: inputting the incremental data set as a variable into the HIV prediction model, and the HIV prediction model outputs a prediction result;

[0041] Step S5: Setting a key for the prediction result and outputting it: Using an encryption algorithm to set a key, encrypt the prediction result on this basis, and output the encrypted prediction result to the user end to complete the HIV incidence prediction.

[0042] In this embodiment, it should be specifically explained that when collecting HIV monitoring and follow-up dynamic data and preprocessing the data, the specific content of obtaining the HIV monitoring and follow-up dynamic data in the database and preprocessing the data to obtain the target data is: the HIV monitoring and follow-up dynamic data is sample test data of different categories, including epidemiological survey results of the tester, CD4 + T cell count, viral load and drug resistance detection, and preprocessing the data to obtain partial target data;

[0043] The preprocessing operation includes steps such as data cleaning, data standardization and data normalization. Data cleaning is used to remove outliers and fill missing values ​​to ensure data quality; data standardization converts data into a unified dimension to facilitate subsequent processing; data normalization scales the data to the [0, 1] interval to eliminate the magnitude differences between data of different dimensions.

[0044] In this embodiment, it should be specifically explained that the specific content of performing neural network integration on part of the target data in different time periods is:

[0045] Part of the target data set obtained at different time periods is represented as S t , t=1, 2, 3, ..., T, where t represents the number of the time period for collecting partial target data, and the S t It is the dynamic data of some HIV monitoring and follow-up in different time periods in the database;

[0046] The overall target data set in the database is S, and the partial target data set S is extracted through neural network integration t In the knowledge, we can obtain the knowledge in the overall data set S.

[0047] In this embodiment, it should be specifically explained that the steps of acquiring the knowledge in the overall target data set S are as follows:

[0048] Step S01: In the partial target data set S t In the example, the set of samples of the nth class is represented as S t (n) , the mean of the nth class of samples is u t (n) , in the overall target data set S, the set of samples of the nth class is represented as S (n) , the mean of the n-th class sample is represented by u (n) , where the number of samples is the number of times data is collected for some target data in different time periods;

[0049] Step S02: Using a partial target dataset S t The mean value u of the nth class sample in t (n) Calculate the mean u of the nth class sample in the overall target data set S (n) , the calculation formula is: where u (n) Represents the mean of the nth class of samples in the overall target data set S, m t (n) Represents the set S of the nth class samples in part of the target data set t (n) The number of samples in m (n) Represents the set S of the nth class of samples in the overall target data set (n) The number of samples in ;

[0050] Step S03: The calculated mean value u of the nth class sample in the overall target data set S is (n) As the target data of the nth class sample;

[0051] Repeat the above steps to obtain the target data of each type of sample in the overall target data set S, where S = {u (1) ,u (2) , ..., u (n)}.

[0052] In this embodiment, it should be specifically explained that the specific contents of performing incremental cycles on the entire target data set in sequence and outputting incremental data are as follows:

[0053] Perform incremental calculation and judgment on the target data of each type of samples in the overall target data set S;

[0054] Incremental calculation is performed on the unselected target data according to the incremental calculation judgment result to form an incremental data set of the target data.

[0055] In this embodiment, it should be specifically explained that the specific contents of performing incremental calculation and judgment on the target data of each type of samples in the overall target data set S are as follows:

[0056] Calculate the proportion of each type of samples in the overall target data set S. The calculation formula is: Where L (n) Represents the proportion of each type of samples in the overall target data set S, m (n) represents the number of samples of the nth class in the overall target data set, and m represents the total number of samples of each class in the overall target data set S;

[0057] Calculate the confidence of the target data of each type of sample in the overall target data set S. The calculation formula is: Where Z (n) Represents the confidence of the target data of each type of sample in the overall target data set S, u (n) represents the mean of the nth class samples in the overall target data set S, maxu t (n) Represents part of the target dataset S t The maximum value of the mean of the nth class samples in , Z represents the normalization parameter;

[0058] The confidence Z of the target data of each type of sample in the overall target data set S (n) Compared with the preset threshold, if the confidence level Z of the target data (n) If the confidence level Z of the target data is greater than the preset threshold, the target data is selected. (n) If the value is less than or equal to the preset threshold, it is determined that the target data is not selected, and an incremental calculation is performed on the unselected target data.

[0059] In this embodiment, it should be specifically explained that the incremental calculation is performed on the unselected target data according to the incremental calculation judgment result to form an incremental data set of the target data. The calculation formula of the incremental calculation is: Where Δu (n) represents the target data after incremental calculation, λ represents the incremental factor, maxu t (n)Represents part of the target dataset S t The maximum value of the mean of the n-th class samples, minu t (n) Represents part of the target dataset S t The minimum value of the mean of the nth class of samples in the algorithm is used to calculate the confidence of the target data after the incremental calculation. The algorithm is repeated until the target data is selected, and the selected target data after the incremental calculation is converted into the confidence value of the target data. (n) Replace the original target data u (n) , forming an incremental data set ΔS of the target data, ΔS = {Δu (1) , Δu (2) , ..., Δu (n)}, if the original target data u (n) If no incremental calculation is performed, Δu (n) =u (n) .

[0060] In this embodiment, it should be specifically explained that the specific content of establishing the HIV prediction model based on the incremental data set is: inputting the incremental data set as a variable into the HIV prediction model, and the expression of the HIV prediction model is: Where Y represents the prediction result of HIV prediction model, U (n) Represents the standard value of each type of sample. If the standard value is an interval value, the middle value of the interval is taken, ω (n) Represents the weight value of each type of samples.

[0061] In this embodiment, it should be specifically explained that the specific content of the key setting and outputting of the prediction result is: encrypting the prediction result of the HIV prediction model, when the prediction result of the HIV prediction model is greater than or equal to the preset risk value, the prediction result is a high risk of disease, when the prediction result of the HIV prediction model is less than the preset risk value, the prediction result is a low risk of disease, and outputting the result to the user end, the user end decrypts the prediction result of the HIV prediction model through the key, and completes the prediction of HIV disease;

[0062] The encryption process steps are as follows:

[0063] Key selection: Select N = pq, where p and q are large prime numbers, and the plaintext space T 1 =Z N , ciphertext space T 2 =Z N ×Z N , the user's private key is p, q, and the public key is N;

[0064] Encryption algorithm: Divide the prediction result M into l message packets, where M = m 1 , m 2 , m3 , ..., m l , using encryption algorithm, randomly select constant k 1 and k 2 , calculate c x =((m x +p×k 1 )modN,(m x +q×k 2 )modN), x=1, 2, 3, ..., l, calculate the ciphertext C=c 1 , c 2 , c 3 , ..., c l ;

[0065] Decryption algorithm: Calculate m x mod p=(m x +q×k 2 ×mod N)×mod p,m x mod q=(m x +p×k 1 ×mod N)×mod q, and get the decryption result D(c x )=((m x modp)qq -1 +(m x modq)pp -1 )modN, output the decrypted ciphertext D=D(c 1 ), D(c 2 ), D(c 3 ), ..., D(c l ).

[0066] The difference between this embodiment and the prior art is that this embodiment is provided with step S1: collecting HIV monitoring follow-up dynamic data and preprocessing the data, step S2: performing neural network integration on partial target data of different time periods, step S3: performing incremental cycles on the overall target data set in turn and outputting the incremental data set, step S4: establishing an HIV prediction model based on the incremental data set, step S5: setting a key for the prediction result and outputting it, an AIDS prediction method based on an incremental neural network model establishes an HIV prediction model based on the incremental data set, thereby improving the accuracy and efficiency of HIV prediction;

[0067] Through neural network integration, partial target data of different time periods are integrated and processed to obtain the overall target data set in the database. The overall target data set is incrementally and cyclically processed according to the incremental neural network model, and the incremental data set is output based on confidence calculation to realize dynamic change monitoring of HIV clinical dynamic data, more accurately reflect the potential risk of HIV onset, thereby improving the accuracy of prediction. The introduction of incremental cyclic processing and confidence calculation enables the model to be dynamically adjusted and optimized, and has good robustness to outliers and noise in the data set, thereby improving the stability and reliability of the prediction results.

[0068] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0069] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An AIDS prediction method based on an incremental neural network model, characterized in that: The following steps are involved: Step S1: Collect HIV monitoring and follow-up dynamic data and pre-process the data: obtain some HIV monitoring and follow-up dynamic data of different time periods in the "Basic Information System for AIDS Prevention and Control of China Center for Disease Control and Prevention" database, and pre-process the data to obtain some target data; Step S2: Performing neural network integration on partial target data of different time periods: Performing integration processing on partial target data of different time periods through neural network integration to obtain the overall target data set in the database; Step S3: performing incremental cycles on the overall target data set in sequence and outputting the incremental data set: performing incremental cycle processing on the overall target data set through the incremental neural network model, and outputting the incremental data set based on the confidence calculation; Step S4: Establishing an HIV prediction model based on the incremental data set: inputting the incremental data set as a variable into the HIV prediction model, and the HIV prediction model outputs a prediction result; Step S5: Setting a key for the prediction result and outputting it: Using an encryption algorithm to set a key, encrypt the prediction result on this basis, and output the encrypted prediction result to the user end to complete the HIV incidence prediction.

2. The AIDS prediction method based on the incremental neural network model according to claim 1, characterized in that: When collecting HIV monitoring and follow-up dynamic data and preprocessing the data, the specific contents of obtaining HIV monitoring and follow-up dynamic data in the database and preprocessing the data to obtain the target data are as follows: the HIV monitoring and follow-up dynamic data are sample test data of different categories, including epidemiological survey results of the tester, CD4 + T cell count, viral load and drug resistance detection, and preprocessing the data to obtain partial target data.

3. The AIDS prediction method based on the incremental neural network model according to claim 1, characterized in that: The specific content of neural network integration of partial target data in different time periods is as follows: Part of the target data set obtained at different time periods is represented as S t , t=1, 2, 3, ..., T, where t represents the number of the time period for collecting partial target data, and the S t It is the dynamic data of some HIV monitoring and follow-up in different time periods in the database; The overall target data set in the database is S, and the partial target data set S is extracted through neural network integration t In the knowledge, we can obtain the knowledge in the overall data set S.

4. The AIDS prediction method based on the incremental neural network model according to claim 2, characterized in that: The steps for acquiring knowledge in the overall target data set S are as follows: Step S01: In the partial target data set S t In the example, the set of samples of the nth class is represented as S t (n) , the mean of the nth class of samples is u t (n) , in the overall target data set S, the set of samples of the nth class is represented as S (n) , the mean of the nth class of samples is represented by u (n) ; Step S02: Using a partial target dataset S t The mean value u of the nth class sample in t (n) Calculate the mean u of the nth class sample in the overall target data set S (n) , the calculation formula is: where u (n) Represents the mean of the nth class of samples in the overall target data set S, m t (n) Represents the set S of the nth class samples in part of the target data set t (n) The number of samples in m (n) Represents the set S of the nth class of samples in the overall target data set (n) The number of samples in ; Step S03: The calculated mean value u of the nth class sample in the overall target data set S is (n) As the target data of the nth class sample; Repeat the above steps to obtain the target data of each type of sample in the overall target data set S, where S = {u (1) ,u (2) , ..., u (n) }.

5. The AIDS prediction method based on the incremental neural network model according to claim 1, characterized in that: The specific contents of performing incremental cycles on the overall target data set in sequence and outputting incremental data are as follows: Perform incremental calculation and judgment on the target data of each type of samples in the overall target data set S; Incremental calculation is performed on the unselected target data according to the incremental calculation judgment result to form an incremental data set of the target data.

6. The AIDS prediction method based on the incremental neural network model according to claim 5, characterized in that: The specific contents of the incremental calculation and judgment of the target data of each type of samples in the overall target data set S are as follows: Calculate the proportion of each type of samples in the overall target data set S. The calculation formula is: Where L (n) Represents the proportion of each type of samples in the overall target data set S, m (n) represents the number of samples of the nth class in the overall target data set, and m represents the total number of samples of each class in the overall target data set S; Calculate the confidence of the target data of each type of sample in the overall target data set S. The calculation formula is: Where Z (n) Represents the confidence of the target data of each type of sample in the overall target data set S, u (n) represents the mean of the nth class samples in the overall target data set S, maxu t (n) Represents part of the target dataset S t The maximum value of the mean of the nth class samples in , Z represents the normalization parameter; The confidence Z of the target data of each type of sample in the overall target data set S (n) Compared with the preset threshold, if the confidence level Z of the target data (n) If the confidence level Z of the target data is greater than the preset threshold, the target data is selected. (n) If the value is less than or equal to the preset threshold, it is determined that the target data is not selected, and an incremental calculation is performed on the unselected target data.

7. The AIDS prediction method based on the incremental neural network model according to claim 5, characterized in that: The incremental calculation is performed on the unselected target data according to the incremental calculation judgment result to form an incremental data set of the target data. The calculation formula of the incremental calculation is: Where Δu (n) represents the target data after incremental calculation, λ represents the incremental factor, maxu t (n) Represents part of the target dataset S t The maximum value of the mean of the n-th class samples, minu t (n) Represents part of the target dataset S t The minimum value of the mean of the nth class of samples in the algorithm is used to calculate the confidence of the target data after the incremental calculation. The algorithm is repeated until the target data is selected, and the selected target data after the incremental calculation is converted into the confidence value of the target data. (n) Replace the original target data u (n) , forming an incremental data set ΔS of the target data, ΔS = {Δu (1) , Δu (2) , ..., Δu (n) }, if the original target data u (n) If no incremental calculation is performed, Δu (n) =u (n) .

8. The AIDS prediction method based on the incremental neural network model according to claim 1, characterized in that: The specific content of establishing the HIV prediction model based on the incremental data set is: inputting the incremental data set as a variable into the HIV prediction model, and the expression of the HIV prediction model is: Where Y represents the prediction result of HIV prediction model, U (n) Represents the standard value of each type of sample. If the standard value is an interval value, the middle value of the interval is taken, ω (n) Represents the weight value of each type of samples.

9. The AIDS prediction method based on the incremental neural network model according to claim 1, characterized in that: The specific content of key setting and outputting the prediction results is: encrypting the prediction results of the HIV prediction model, when the prediction results of the HIV prediction model are greater than or equal to the preset risk value, the prediction result is a high risk of disease, when the prediction results of the HIV prediction model are less than the preset risk value, the prediction result is a low risk of disease, and the result is output to the user end, the user end decrypts the prediction results of the HIV prediction model through the key, and completes the prediction of HIV.

Images