A CKD high-risk factor screening system based on big data

Abstract

The present invention belongs to the field of medical health information technology, and provides a CKD high-risk factor screening system based on big data. The CKD medical record information of multiple CKD patients in a historical period is extracted from a hospital system through ETL technology, and analyzed to obtain high-frequency factors. The analysis is based on the high-frequency factors to obtain a preliminary screening high-risk factor table. The preliminary screening high-risk factor tables in multiple consecutive historical periods are analyzed and processed, and the ranking changes of high-frequency factors in multiple preliminary screening high-risk factor tables and the frequencies of occurrence in multiple consecutive historical periods are analyzed. The high-frequency factors are further screened to obtain sensitive factors, and the sensitive factors are summarized to obtain a sensitive high-risk factor table, so as to screen the factors of CKD patients layer by layer and determine the high-risk screening conditions, which is conducive to tracking and comparing all patients in the hospital according to the medical record information of the hospital patients, and improves the statistical efficiency of CKD patients.

Description

Technical Field

[0001] The present invention belongs to the field of medical health information technology, and specifically is a CKD high-risk factor screening system based on big data. Background Art

[0002] With the improvement of people's living standards, the prevalence of CKD (chronic kidney disease) has been on the rise. We now screen the medical information of historical CKD patients to obtain a high-risk screening table, and diagnose patients based on the high-risk screening table. The purpose is not only to improve the diagnostic efficiency of hospital doctors, but also to improve the screening accuracy of CKD patients, which will help hospitals to conduct targeted screening and management of patients.

[0003] In the prior art, it mainly relies on the doctor's experience and the patient's clinical symptoms. Due to the lack of systematic data analysis, it is often difficult to accurately identify all CKD high-risk factors, resulting in low accuracy of screening results. Therefore, this application uses ETL technology to extract CKD medical record information of multiple CKD patients in the historical period from the hospital system, and analyzes and screens them layer by layer to determine high-risk screening conditions, which is conducive to tracking and comparing all patients in the hospital based on the medical record information of the hospital patients, thereby improving the statistical efficiency of CKD patients. In addition, by self-checking the sensitive factors in the sensitive high-risk factor table with the disease factors of multiple CKD historical patients, determining the factors to be optimized, and analyzing and processing the factors to be optimized, a high-risk screening table is obtained, which not only improves the diagnostic efficiency of hospital doctors, but also improves the screening accuracy of CKD patients, which helps hospitals to conduct targeted screening and management of patients.

[0004] To this end, the present invention provides a CKD high-risk factor screening system based on big data. Summary of the invention

[0005] In order to make up for the deficiencies of the prior art, at least one technical problem raised in the background technology is solved.

[0006] The technical solution adopted by the present invention to solve the technical problem is: a CKD high-risk factor screening system based on big data, comprising:

[0007] Preliminary screening module: The CKD medical records of multiple patients with a history of CKD are extracted from the hospital system through ETL technology, and the CKD medical records are analyzed and processed to obtain high-frequency factors, which are then sorted and summarized into a preliminary screening high-risk factor table;

[0008] Sensitive screening module: Based on the initial screening high-risk factor table, the initial screening high-risk factor table in multiple continuous historical periods is analyzed and processed to obtain high-frequency factor data, and the high-frequency factor data is analyzed and processed to obtain the high-risk stable value, which is compared with the high-risk stable threshold to obtain the sensitive high-risk factor table;

[0009] Self-examination matching module: self-examination and screening of the sensitive factors in the sensitive high-risk factor table and the CKD medical records of multiple patients with a history of CKD to obtain self-examination screening data, analyze and process the self-examination screening data, output the self-examination matching degree, and compare it with the self-examination matching degree threshold to obtain the factors to be optimized;

[0010] Matching optimization module: Based on the factors to be optimized, the factors to be optimized are screened and analyzed to obtain non-optimized factors and optimized factors, and then analyzed and processed based on the optimized factors to obtain a high-risk screening table, which is helpful for the screening management of CKD patients.

[0011] As a further solution of the present invention, the high-frequency factors are obtained by:

[0012] The CKD medical records of multiple historical CKD patients were compared for overlap of disease factors. The process is as follows:

[0013] Integrate the CKD disease factors of multiple CKD patients to obtain a disease factor set;

[0014] In the patient factor set, repeated CKD disease factors were marked as high-frequency factors.

[0015] As a further solution of the present invention, the method for obtaining the initial screening high-risk factor table is:

[0016] The number of repeated high-frequency factors was counted, and the ratio of the number of repeated high-frequency factors to the total number of CKD disease factors in all CKD patients was calculated to obtain the high-frequency ranking value;

[0017] Sort the high-frequency factors in the disease factor set in descending order according to the high-frequency ranking value to obtain a preliminary screening high-risk factor table.

[0018] As a further solution of the present invention, the high-frequency factor data is obtained by:

[0019] Integrate the high-risk factor tables of multiple consecutive historical periods into a continuous analysis set R={ , , ,......, }, where the elements in the continuous analysis set R are sorted according to the time series of multiple continuous historical periods, It is represented as the initial screening high-risk factor table of the nth historical period, and n is represented as the total number of initial screening high-risk factor tables of consecutive historical periods;

[0020] Perform matrix processing on all elements in the continuous analysis set R corresponding to the high-frequency factors in the initial screening high-risk factor table to obtain the continuous analysis matrix , where j represents the total number of high-frequency factors in the initial screening high-risk factor table corresponding to the elements in the continuous analysis set R;

[0021] In the continuous analysis matrix, the same high-frequency factors are extracted as target analysis factors;

[0022] Obtain the ranking of any target analysis factor in different initial screening high-risk factor tables, make mutual differences, take the absolute value, and obtain the ranking difference;

[0023] The ranking differences of the target analysis elements are summed and averaged to obtain the ranking change value, which is marked as ;

[0024] Get the number of target analysis factors in the continuous analysis matrix, calculate the ratio of the number of target analysis factors in the continuous analysis matrix to the total number of high-frequency factors in the continuous analysis matrix, get the high-frequency number value, and mark it as .

[0025] As a further solution of the present invention, the high-risk stability value is obtained by:

[0026] Change the ranking value With high frequency value Substituting into the formula: , calculate the high-risk stability value ,in, , Expressed as a preset scaling factor.

[0027] As a further solution of the present invention, the method for obtaining the sensitive high-risk factor table is:

[0028] The high-risk stability value is compared with the high-risk stability threshold, and the process is as follows:

[0029] If the high-risk stability value is greater than or equal to the high-risk stability threshold, a high-risk stability signal is generated, and the high-frequency factor corresponding to the generation of the high-risk stability signal is marked as a sensitive factor;

[0030] If the high-risk stability value is less than the high-risk stability threshold, a high-risk instability signal is generated;

[0031] All sensitive factors are sorted from large to small according to their corresponding high-risk stable values, and summarized and integrated to obtain a sensitive high-risk factor table.

[0032] As a further solution of the present invention, the self-examination screening data is obtained by:

[0033] Extract sensitive factors from the sensitive high-risk factor table, and compare the sensitive factors with the CKD disease factors of multiple patients with a history of CKD. If the CKD disease factors of the CKD patient with a history of CKD have sensitive factors, the CKD patient with the sensitive factors will be marked as a sensitive overlap patient, and the CKD disease factors that overlap with the sensitive factors will be marked as sensitive overlap factors.

[0034] The number of sensitive overlapping patients is counted, and the ratio of the number of sensitive overlapping patients to the total number of multiple CKD history patients is calculated to obtain the patient matching value;

[0035] The number of sensitive overlapping factors in the CKD disease factors of sensitive overlapping patients was counted, and the ratio of the number of sensitive overlapping factors in the CKD disease factors of sensitive overlapping patients to the total number of CKD disease factors of sensitive overlapping patients was calculated to obtain the unit matching degree;

[0036] The unit matching degrees corresponding to all sensitive overlapping patients are added and averaged to obtain the matching degree value.

[0037] As a further solution of the present invention, the method for obtaining the factors to be optimized is:

[0038] The patient matching value is multiplied by the matching degree value to obtain the self-test matching degree;

[0039] The self-check matching degree is compared with the self-check matching degree threshold. If the self-check matching degree is less than the self-check matching degree threshold, a factor optimization signal is generated, and the sensitive factor corresponding to the factor optimization signal is marked as a factor to be optimized.

[0040] As a further solution of the present invention, the non-optimization factors and the optimization factors are obtained in the following manner:

[0041] Subtract the self-check matching degree corresponding to the factor to be optimized from the self-check matching degree threshold, take the absolute value, and obtain the self-check matching difference;

[0042] Calculate the ratio of the high-risk stability value corresponding to the factor to be optimized to the self-check matching difference to obtain the factor optimization value;

[0043] The factor optimization value is compared with the preset factor optimization threshold. The process is as follows:

[0044] If the factor optimization value is greater than or equal to the factor optimization threshold, the factor to be optimized is marked as a non-optimized factor;

[0045] If the factor optimization value is less than the factor optimization threshold, the factor to be optimized is marked as an optimization factor.

[0046] As a further solution of the present invention, the high-risk screening table is obtained by:

[0047] Compare the factor optimization values ​​corresponding to all optimization factors, and delete the optimization factors one by one from small to large until all the optimization factors are deleted;

[0048] The non-optimized factors are sorted from small to large according to the corresponding self-test matching differences, and are added back into the sensitive high-risk factor table to obtain a high-risk screening table.

[0049] The beneficial effects of the present invention are as follows:

[0050] (1) The present invention extracts CKD medical record information of multiple CKD patients in a historical period from a hospital system through ETL technology, and analyzes it to obtain high-frequency factors. By comparing the number of each high-frequency factor, the high-frequency factors are summarized from large to small to obtain a preliminary screening high-risk factor table. The preliminary screening high-risk factor table in multiple consecutive historical periods is analyzed and processed, and the ranking changes of high-frequency factors in multiple preliminary screening high-risk factor tables and the frequencies of occurrence in multiple consecutive historical periods are analyzed. The high-frequency factors are further screened to obtain sensitive factors, and the sensitive factors are summarized to obtain a sensitive high-risk factor table, thereby screening the factors of CKD patients layer by layer and determining high-risk screening conditions, which is conducive to tracking and comparing all patients in the hospital according to the medical record information of the hospital patients, thereby improving the statistical efficiency of CKD patients;

[0051] (2) The present invention determines the factors to be optimized by self-checking the sensitive factors in the sensitive high-risk factor table with the disease factors of multiple CKD historical patients, and performs analysis and processing based on the optimized factors to calculate the self-check matching difference, which reflects the degree of matching deviation between the factors to be optimized and the CKD disease factors, and calculates the ratio of the high-risk stable value corresponding to the factors to be optimized to the self-check matching difference to obtain the factor optimization value, which comprehensively reflects the degree of matching deviation between the factors to be optimized and the CKD disease factors, as well as the stability in multiple continuous historical cycles, thereby improving the screening accuracy of CKD patients in hospitals and facilitating hospitals to conduct targeted screening and management of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The present invention will be further described below in conjunction with the accompanying drawings.

[0053] Figure 1 It is a system module diagram of a CKD high-risk factor screening system based on big data of the present invention;

[0054] Figure 2 It is a step flow chart of a CKD high-risk factor screening system based on big data of the present invention. DETAILED DESCRIPTION

[0055] In order to make the technical means, creative factors, objectives and effects achieved by the present invention easy to understand, the present invention is further described below in conjunction with specific implementation methods.

[0056] Example 1

[0057] like Figure 1 - Figure 2 As shown, a CKD high-risk factor screening system based on big data according to an embodiment of the present invention includes:

[0058] Preliminary screening module: The CKD medical record information of multiple patients with a history of CKD is extracted from the hospital system through ETL technology. The CKD medical record information includes CKD disease factors. The CKD medical record information is analyzed and processed to obtain high-frequency factors, which are sorted and summarized into a preliminary screening high-risk factor table;

[0059] In some embodiments, the CKD medical records of multiple historical CKD patients are compared for overlap of disease factors, and the process is as follows:

[0060] Integrate the CKD disease factors of multiple CKD patients to obtain a disease factor set;

[0061] It should be noted that the disease factor set includes CKD disease factors of multiple CKD patients, and there are repeated CKD disease factors (CKD disease factors common to two or more CKD patients);

[0062] In the patient factor set, repeated CKD disease factors are marked as high-frequency factors, and the number of repeated high-frequency factors is counted. The ratio of the number of repeated high-frequency factors to the total number of CKD disease factors of all CKD patients is calculated to obtain a high-frequency ranking value;

[0063] Sort the high-frequency factors in the disease factor set in descending order according to the high-frequency ranking value to obtain a preliminary screening high-risk factor table;

[0064] Sensitive screening module: Based on the initial screening high-risk factor table, the initial screening high-risk factor table in multiple continuous historical periods is analyzed and processed to obtain high-frequency factor data, where the high-frequency factor data includes ranking change values ​​and high-frequency quantity values. The high-frequency factor data is analyzed and processed to obtain a high-risk stable value, and compared with the high-risk stable threshold. If the high-risk stable value is greater than or equal to the high-risk stable threshold, a high-risk stable signal is generated to obtain sensitive factors, which are summarized and integrated to obtain a sensitive high-risk factor table;

[0065] In some embodiments, the initial screening high-risk factor tables of multiple consecutive historical periods are integrated into a continuous analysis set R={ , , ,......, }, where the elements in the continuous analysis set R are sorted according to the time series of multiple continuous historical periods, It is represented as the initial screening high-risk factor table of the nth historical period, and n is represented as the total number of initial screening high-risk factor tables of consecutive historical periods;

[0066] It should be noted that there is an initial high-risk factor table within a historical period;

[0067] Perform matrix processing on all elements in the continuous analysis set R corresponding to the high-frequency factors in the initial screening high-risk factor table to obtain the continuous analysis matrix , where j represents the total number of high-frequency factors in the initial screening high-risk factor table corresponding to the elements in the continuous analysis set R;

[0068] For example, Expressed as Corresponding to the high-frequency factors ranked first in the initial screening high-risk factor table, Expressed as The high-frequency factors ranked first in the corresponding initial screening high-risk factor table;

[0069] Exemplarily, in the continuous analysis matrix, the same high-frequency factors are extracted as target analysis factors;

[0070] Obtain the ranking of any target analysis factor in different initial screening high-risk factor tables, make mutual differences, take the absolute value, and obtain the ranking difference;

[0071] The ranking differences of the target analysis elements are summed and averaged to obtain the ranking change value, which is marked as ;

[0072] Get the number of target analysis factors in the continuous analysis matrix, calculate the ratio of the number of target analysis factors in the continuous analysis matrix to the total number of high-frequency factors in the continuous analysis matrix, get the high-frequency number value, and mark it as ;

[0073] Change the ranking value With high frequency value Substituting into the formula: , calculate the high-risk stability value ,in, , is represented by a preset scaling factor, and The value is 3.264. The value is 6.325;

[0074] It can be understood that the high-risk stability value means: measuring the stability of the target analysis factor in multiple consecutive historical periods, and reflecting the ranking change of the target analysis factor in different initial screening high-risk factor tables through the ranking change value. If the ranking change is smaller, it means that the stability of the target analysis factor in multiple consecutive historical periods is stronger, otherwise, the stability is weaker. The high-frequency data value reflects the frequency of the target analysis factor in multiple consecutive historical periods, that is, the prevalence and importance in multiple initial screening high-risk factor tables. If the frequency is higher, it means that the target analysis factor has been identified as a high-risk factor many times, and the stability is stronger, otherwise, the stability is weaker.

[0075] Compare the high-risk stability value with the high-risk stability threshold, the process is as follows:

[0076] If the high-risk stability value is greater than or equal to the high-risk stability threshold, the ranking change of the target analysis factor in multiple consecutive historical cycles is small, and the target analysis factor appears frequently in multiple initial screening high-risk factor tables, a high-risk stability signal is generated, and the high-frequency factor corresponding to the generation of the high-risk stability signal is marked as a sensitive factor;

[0077] If the high-risk stability value is less than the high-risk stability threshold, the ranking of the target analysis factor has changed greatly in multiple consecutive historical cycles, and the target analysis factor appears less frequently in multiple initial screening high-risk factor tables, generating a high-risk unstable signal;

[0078] All sensitive factors are sorted from large to small according to their corresponding high-risk stable values, and a sensitive high-risk factor table is obtained by summarizing and integrating them;

[0079] The specific implementation scheme of the embodiment of the present invention is: extracting CKD medical record information of multiple CKD patients in the historical period from the hospital system through ETL technology, and analyzing to obtain high-frequency factors, by comparing the number of each high-frequency factor, summarizing the high-frequency factors in a descending order to obtain a preliminary screening high-risk factor table, analyzing and processing the preliminary screening high-risk factor table in multiple consecutive historical periods, analyzing the ranking changes of high-frequency factors in multiple preliminary screening high-risk factor tables, and the frequency of occurrence in multiple consecutive historical periods, further screening the high-frequency factors to obtain sensitive factors, and summarizing the sensitive factors to obtain a sensitive high-risk factor table, thereby screening the factors of CKD patients layer by layer and determining high-risk screening conditions, which is conducive to tracking and comparing all patients in the hospital according to the medical record information of the hospital patients, and improving the statistical efficiency of CKD patients.

[0080] Example 2

[0081] like Figure 1 - Figure 2 As shown, based on Example 1, a CKD high-risk factor screening system based on big data according to an embodiment of the present invention includes:

[0082] Self-examination matching module: self-examination and screening of the sensitive factors in the sensitive high-risk factor table with the CKD medical records of multiple CKD patients to obtain self-examination screening data, wherein the self-examination screening data includes patient matching values ​​and matching degree values, and the patient matching values ​​and matching degree values ​​are analyzed and processed to output the self-examination matching degree, which is compared with the self-examination matching degree threshold. If the self-examination matching degree is less than the self-examination matching degree threshold, a factor optimization signal is generated, and the sensitive factor corresponding to the factor optimization signal is marked as a factor to be optimized;

[0083] In some embodiments, sensitive factors in the sensitive high-risk factor table are extracted, and the sensitive factors are compared with CKD disease factors of multiple patients with a history of CKD, and the process is as follows:

[0084] If the CKD disease factors of the CKD history patient have sensitive factors, the CKD history patient with the sensitive factors will be marked as a sensitive overlap patient, and the CKD disease factors that overlap with the sensitive factors will be marked as sensitive overlap factors;

[0085] If the CKD patient factor of the CKD history patient does not have sensitive factors, the CKD history patient without sensitive factors is marked as a sensitive non-overlapping patient;

[0086] It should be noted that the sensitive factors are extracted in the order in which they are arranged in the sensitive high-risk factor table (in the order from small to large high-risk stable values);

[0087] The number of sensitive overlapping patients is counted, and the ratio of the number of sensitive overlapping patients to the total number of multiple CKD history patients is calculated to obtain the patient matching value;

[0088] The number of sensitive overlapping factors in the CKD disease factors of sensitive overlapping patients was counted, and the ratio of the number of sensitive overlapping factors in the CKD disease factors of sensitive overlapping patients to the total number of CKD disease factors of sensitive overlapping patients was calculated to obtain the unit matching degree;

[0089] The unit matching degrees corresponding to all sensitive overlapping patients are added and averaged to obtain the matching degree value;

[0090] The patient matching value is multiplied by the matching degree value to obtain the self-test matching degree;

[0091] It can be understood that the meaning of the self-test matching degree is to evaluate the overlap between the sensitive factors in the sensitive high-risk factor table and the CKD patient factors of multiple CKD history patients, through the prevalence of sensitive factors in CKD history patients, and the matching degree between sensitive factors and CKD disease factors. Specifically, if this value is larger, it means that the sensitive factor exists in more CKD patients, and the matching degree between the sensitive factor and the CKD disease factors is higher. On the contrary, if this value is smaller, it means that the sensitive factor exists in smaller CKD patients, and the matching degree between the sensitive factor and the CKD patient factors is lower.

[0092] The self-check match is compared with the self-check match threshold, and the process is as follows:

[0093] If the self-test matching degree is greater than or equal to the self-test matching degree threshold, it means that the sensitive factor exists in more CKD patients, and the matching degree between the sensitive factor and the CKD disease factor is high, a self-test matching high signal is generated, and the sensitive factor corresponding to the self-test matching high signal is marked as a non-optimized factor;

[0094] If the self-test matching degree is less than the self-test matching degree threshold, it means that the sensitive factor exists in more CKD patients, and the matching degree between the sensitive factor and the CKD disease factor is low, a factor optimization signal is generated, and the sensitive factor corresponding to the factor optimization signal is marked as a factor to be optimized;

[0095] Matching optimization module: Based on the factors to be optimized, the factors to be optimized are screened and analyzed to obtain non-optimized factors and optimized factors, and then the non-optimized factors and optimized factors are analyzed and processed to obtain a high-risk screening table, which is helpful for the screening management of CKD patients;

[0096] Specifically, based on the factor optimization signal, the self-check matching degree corresponding to the factor to be optimized is subtracted from the self-check matching degree threshold, and the absolute value is taken to obtain the self-check matching difference;

[0097] Calculate the ratio of the high-risk stability value corresponding to the factor to be optimized to the self-check matching difference to obtain the factor optimization value;

[0098] The factor optimization value is compared with the preset factor optimization threshold. The process is as follows:

[0099] If the factor optimization value is greater than or equal to the factor optimization threshold, the factor to be optimized is marked as a non-optimized factor;

[0100] If the factor optimization value is less than the factor optimization threshold, the factor to be optimized is marked as an optimization factor and the optimization operation is performed. The process is as follows:

[0101] Compare the factor optimization values ​​corresponding to all optimization factors, and delete the optimization factors one by one from small to large until all the optimization factors are deleted;

[0102] The non-optimized factors are sorted from small to large according to the corresponding self-test matching differences, and are added back to the sensitive high-risk factor table to obtain a high-risk screening table;

[0103] The specific implementation scheme of the embodiment of the present invention is: by self-checking the sensitive factors in the sensitive high-risk factor table with the disease factors of multiple CKD historical patients, the factors to be optimized are determined, and analysis and processing are performed based on the optimized factors, and the self-check matching difference is calculated to reflect the degree of matching deviation between the factors to be optimized and the CKD disease factors, and the high-risk stable value corresponding to the factors to be optimized is calculated with the ratio of the self-check matching difference to obtain the factor optimization value, which comprehensively reflects the degree of matching deviation between the factors to be optimized and the CKD disease factors, as well as the stability in multiple continuous historical cycles, thereby improving the hospital's screening accuracy for CKD patients and helping the hospital to conduct targeted screening and management of patients.

[0104] The CKD medical record information of patients with a history of CKD involved in this application has been obtained with full consent and authorization, and the collection, use and processing of relevant information complies with the relevant laws, regulations and standards of relevant countries and regions.

[0105] The above shows and describes the basic principles, main factors and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A CKD high-risk factor screening system based on big data, characterized by: include: Preliminary screening module: The CKD medical records of multiple patients with a history of CKD are extracted from the hospital system through ETL technology, and the CKD medical records are analyzed and processed to obtain high-frequency factors, which are then sorted and summarized into a preliminary screening high-risk factor table; Sensitive screening module: Based on the initial screening high-risk factor table, the initial screening high-risk factor table in multiple continuous historical periods is analyzed and processed to obtain high-frequency factor data, and the high-frequency factor data is analyzed and processed to obtain the high-risk stable value, which is compared with the high-risk stable threshold to obtain the sensitive high-risk factor table; Self-examination matching module: self-examination and screening of the sensitive factors in the sensitive high-risk factor table and the CKD medical records of multiple patients with a history of CKD to obtain self-examination screening data, analyze and process the self-examination screening data, output the self-examination matching degree, and compare it with the self-examination matching degree threshold to obtain the factors to be optimized; The self-examination screening data is obtained by: Extract sensitive factors from the sensitive high-risk factor table, and compare the sensitive factors with the CKD disease factors of multiple patients with a history of CKD. If the CKD disease factors of the CKD patient with a history of CKD have sensitive factors, the CKD patient with the sensitive factors will be marked as a sensitive overlap patient, and the CKD disease factors that overlap with the sensitive factors will be marked as sensitive overlap factors. The number of sensitive overlapping patients is counted, and the ratio of the number of sensitive overlapping patients to the total number of multiple CKD history patients is calculated to obtain the patient matching value; The number of sensitive overlapping factors in the CKD disease factors of sensitive overlapping patients was counted, and the ratio of the number of sensitive overlapping factors in the CKD disease factors of sensitive overlapping patients to the total number of CKD disease factors of sensitive overlapping patients was calculated to obtain the unit matching degree; The unit matching degrees corresponding to all sensitive overlapping patients are added and averaged to obtain the matching degree value; Matching optimization module: Based on the factors to be optimized, the factors to be optimized are screened and analyzed to obtain non-optimized factors and optimized factors, and then analyzed and processed based on the optimized factors to obtain a high-risk screening table, which is helpful for the screening management of CKD patients.

2. The CKD high-risk factor screening system based on big data according to claim 1, characterized in that: The high-frequency factors are obtained as follows: The CKD medical records of multiple historical CKD patients were compared for overlap of disease factors. The process is as follows: Integrate the CKD disease factors of multiple CKD patients to obtain a disease factor set; In the patient factor set, repeated CKD disease factors were marked as high-frequency factors.

3. The CKD high-risk factor screening system based on big data according to claim 2, characterized in that: The method of obtaining the initial screening high-risk factor table is: The number of repeated high-frequency factors was counted, and the ratio of the number of repeated high-frequency factors to the total number of CKD disease factors in all CKD patients was calculated to obtain the high-frequency ranking value; Sort the high-frequency factors in the disease factor set in descending order according to the high-frequency ranking value to obtain a preliminary screening high-risk factor table.

4. The CKD high-risk factor screening system based on big data according to claim 1, characterized in that: The method of obtaining high-frequency factor data is as follows: Integrate the high-risk factor tables of multiple consecutive historical periods into a continuous analysis set R={ , , ,......, }, where the elements in the continuous analysis set R are sorted according to the time series of multiple continuous historical periods, It is represented as the initial screening high-risk factor table of the nth historical period, and n is represented as the total number of initial screening high-risk factor tables of consecutive historical periods; Perform matrix processing on all elements in the continuous analysis set R corresponding to the high-frequency factors in the initial screening high-risk factor table to obtain the continuous analysis matrix , where j represents the total number of high-frequency factors in the initial screening high-risk factor table corresponding to the elements in the continuous analysis set R; In the continuous analysis matrix, the same high-frequency factors are extracted as target analysis factors; Obtain the ranking of any target analysis factor in different initial screening high-risk factor tables, make mutual differences, take the absolute value, and obtain the ranking difference; The ranking differences of the target analysis elements are summed and averaged to obtain the ranking change value, which is marked as ; Get the number of target analysis factors in the continuous analysis matrix, calculate the ratio of the number of target analysis factors in the continuous analysis matrix to the total number of high-frequency factors in the continuous analysis matrix, get the high-frequency number value, and mark it as .

5. The CKD high-risk factor screening system based on big data according to claim 4 is characterized in that: The high-risk stability value is obtained as follows: Change the ranking value With high frequency value Substituting into the formula: , calculate the high-risk stability value ,in, , Expressed as a preset scaling factor.

6. The CKD high-risk factor screening system based on big data according to claim 5, characterized in that: The method of obtaining the sensitive high-risk factor table is as follows: The high-risk stability value is compared with the high-risk stability threshold, and the process is as follows: If the high-risk stability value is greater than or equal to the high-risk stability threshold, a high-risk stability signal is generated, and the high-frequency factor corresponding to the generation of the high-risk stability signal is marked as a sensitive factor; If the high-risk stability value is less than the high-risk stability threshold, a high-risk instability signal is generated; All sensitive factors are sorted from large to small according to their corresponding high-risk stable values, and summarized and integrated to obtain a sensitive high-risk factor table.

7. The CKD high-risk factor screening system based on big data according to claim 1, characterized in that: The method for obtaining the factors to be optimized is: The patient matching value is multiplied by the matching degree value to obtain the self-test matching degree; The self-check matching degree is compared with the self-check matching degree threshold. If the self-check matching degree is less than the self-check matching degree threshold, a factor optimization signal is generated, and the sensitive factor corresponding to the factor optimization signal is marked as a factor to be optimized.

8. The CKD high-risk factor screening system based on big data according to claim 1, characterized in that: The non-optimization factors and optimization factors are obtained as follows: Subtract the self-check matching degree corresponding to the factor to be optimized from the self-check matching degree threshold, take the absolute value, and obtain the self-check matching difference; Calculate the ratio of the high-risk stability value corresponding to the factor to be optimized to the self-check matching difference to obtain the factor optimization value; The factor optimization value is compared with the preset factor optimization threshold. The process is as follows: If the factor optimization value is greater than or equal to the factor optimization threshold, the factor to be optimized is marked as a non-optimized factor; If the factor optimization value is less than the factor optimization threshold, the factor to be optimized is marked as an optimization factor.

9. The CKD high-risk factor screening system based on big data according to claim 8, characterized in that: The high-risk screening form can be obtained as follows: Compare the factor optimization values ​​corresponding to all optimization factors, and delete the optimization factors one by one from small to large until all the optimization factors are deleted; The non-optimized factors are sorted from small to large according to the corresponding self-test matching differences, and are added back into the sensitive high-risk factor table to obtain a high-risk screening table.

Images