Rheumatism immune nephropathy early warning method and system based on data analysis

Abstract

The invention discloses a rheumatism immune nephropathy early warning method and system based on data analysis, and the method comprises the steps: dividing patients suffering from rheumatism immune nephropathy into different types of diseases, obtaining the correlation coefficient between any examination data of the patients and the severity of the diseases, and obtaining the correlation coefficient between the examination data and the severity of the diseases of the patients; obtaining a final correlation coefficient of any item of examination data of the patients in accordance with the disease type, and obtaining the weight of the patients in subsequent prediction of the risk index of the unknown patients suffering from the disease type based on the difference distance of the examination data and the disease severity difference distance between the patients in the type; the method comprises the steps of obtaining a risk index estimated value of an unknown patient suffering from the type of disease in combination with examination data of a to-be-predicted patient and related data of the type of patient, and then performing more accurate and timely early warning in combination with a single risk index estimated value or multiple time sequence changes of risk indexes of the single risk index estimated value. Therefore, a more accurate medical decision is provided for the patient in time.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing, and relates to an early warning method and system for rheumatic immune nephropathy based on data analysis. Background Art

[0002] Rheumatic immune nephropathy is a kidney damage disease caused by autoimmune abnormalities, covering subtypes such as lupus nephritis, antineutrophil cytoplasmic antibody-associated nephritis, and rheumatoid arthritis-related nephropathy. Its early symptoms are latent and non-specific. By the time of diagnosis, patients often already have irreversible renal function damage or even progress to end-stage renal disease (ESRD). The traditional diagnosis mode relies on clinical symptoms and laboratory indicators (such as proteinuria and elevated serum creatinine), but these indicators are significantly abnormal only after more than 50% of the nephrons are damaged, resulting in a lag in the intervention time. Therefore, how to achieve early warning of rheumatic immune nephropathy is an urgent problem to be solved. Summary of the Invention

[0003] The purpose of the present invention is to solve the problem in the prior art that the early symptoms of rheumatic immune nephropathy are not obvious and timely intervention cannot be carried out, and to provide an early warning method and system for rheumatic immune nephropathy based on data analysis.

[0004] To achieve the above purpose, the present invention adopts the following technical solutions: An early warning method for rheumatic immune nephropathy based on data analysis, including: Obtain the medical test data of patients with rheumatic immune nephropathy, and divide the patients with rheumatic immune nephropathy into different types according to the type of immune disease; Based on any type of patients, perform normalization processing on different item inspection data in the medical test data of this type of patients respectively; Based on the normalized different item inspection data, obtain the correlation coefficient between any item of inspection data and the severity of the rheumatic immune disease of the patient; Perform a union process on the correlation coefficients of all item inspection data to obtain the final correlation coefficient between any item of inspection data and this disease type; Based on the obtained final correlation coefficient, obtain the difference distance between any two patients among the patients of this disease type; Based on the severity of the disease of the patient and the difference distance, obtain the reference weight of the patients of this disease type when estimating the risk index of an unknown patient suffering from this type of disease; the severity of the disease of the patient is obtained by performing normalization processing on the disease severity score value, and the disease severity score value is the data in the medical test data of the patient with rheumatic immune nephropathy; Based on the reference weight and the difference distance between any two patients, obtain the risk index of any unknown patient suffering from any type of rheumatic immune nephropathy disease; Judge the relationship between the risk index and the set threshold. If it is greater, it is determined that the current patient suffers from rheumatic immune nephropathy disease.

[0005] A further improvement of the present invention lies in: Furthermore, based on the normalized inspection data of different items, obtain the correlation coefficient between the inspection data of any item and the severity of the patient's rheumatic immune disease, specifically: ; Wherein, represents the correlation coefficient between the i-th inspection data of the currently analyzed patient and the severity of the patient suffering from this type of disease; represents the absolute value operation symbol; represents the corresponding value of the i-th normalized inspection data of the currently analyzed patient; represents the reference range when the i-th normalized inspection data of the currently analyzed patient is normal; represents the normalized severity of this type of disease of the currently analyzed patient; represents the median of the reference range when the i-th normalized inspection data of the currently analyzed patient is normal, that is, represents the reference value of this item of inspection data; represents the ratio of the absolute value of the difference between the corresponding value of the i-th inspection data of the currently analyzed patient and the reference value of this item of inspection data to the severity of the disease suffered by the patient. If this value is larger, the higher the relevance of the amplitude change of this item of data to this type of disease of the patient, otherwise the lower the correlation between this item of inspection data and the type of disease suffered by the patient.

[0006] Furthermore, perform a union process on the correlation coefficients of all items of inspection data to obtain the final correlation coefficient between the inspection data of any item and this type of disease, specifically: aggregate the correlation coefficients of all items of inspection data of this type of patient as the target set; and arrange the correlation coefficients of the correlation coefficient set in order, select the median after sorting of this set, calculate the absolute value of the difference between each correlation coefficient in this set and the selected median, and use it as the distance value between each correlation coefficient and this median data; based on the distance value between each correlation coefficient and the median and the numerical size of the corresponding correlation coefficient, obtain the final correlation coefficient between the inspection data of any item and this type of disease.

[0007] Furthermore, based on the distance value between each correlation coefficient and the median and the numerical size of the corresponding correlation coefficient, obtain the final correlation coefficient between the inspection data of any item and this type of disease, specifically: ; Among them, represents the final correlation coefficient between the i-th examination data and the correlation coefficient of this type of disease among the patients with the disease type being analyzed currently; represents the median after sorting the target correlation coefficient set corresponding to the i-th examination data among the patients with the disease type being analyzed currently; represents the number of elements in the target correlation coefficient set corresponding to the i-th examination data among the patients with the disease type being analyzed currently; and respectively represent the s-th correlation coefficient value after sorting the target correlation coefficient set corresponding to the i-th examination data among the patients with the disease type being analyzed currently and the distance value between this correlation coefficient and the median data; c represents a constant term to prevent the denominator from being zero.

[0008] Furthermore, based on the obtained final correlation coefficient, obtain the difference distance between any two patients among the patients of this disease type, specifically: ; Among them, represents the difference distance between patient a and patient b among the patients of the disease type being analyzed currently, represents the square root operation; n represents the number of examination data items of these two patients among the patients of the disease type being analyzed currently; represents the final correlation coefficient obtained from the i-th examination data and this disease type among the patients of the disease type being analyzed currently; and respectively represent the normalized i-th examination data values corresponding to patient a and patient b among the patients of the disease type being analyzed currently; represents the product of the final correlation coefficient obtained from the i-th examination data and this disease type among the patients of the disease type being analyzed currently and the absolute value of the difference between the i-th examination data values corresponding to patient a and patient b among the patients of the disease type being analyzed currently. If the final correlation coefficient of this examination data is larger, it indicates that the contribution of this examination data to the diagnosis and diagnosis of this disease type is higher. Then, when the difference between the examination data of these two patients is smaller, the disease symptoms of these two patients may be more similar, and the difference distance between the disease symptoms and the disease severity of these two patients may be more similar; represents the square root of the sum of the squares of the product of the absolute value of the difference between the normalized i-th examination data values corresponding to patient a and patient b among the patients of the disease type being analyzed currently and their correlation coefficients related to this disease type, that is, it represents the difference distance between these two patients for this disease type.

[0009] Further, based on the disease severity and the difference distance of the patient, obtain the patients of this disease type as the reference weight when estimating the risk index of an unknown patient suffering from this type of disease. Specifically: ; Among them, represents the reference weight of the j-th patient among the known patients of the currently analyzed disease type when estimating the risk of an unknown patient of this disease type in the future; , represent the disease severities of the j-th patient and the r-th patient among the known patients of the currently analyzed disease type; g represents the number of currently known patients of this disease type; represents the overall difference between the severity value of the j-th patient suffering from this disease type and the severity values of the other patients of this disease type among the known patients of the currently analyzed disease type; represents the sum of the difference distances between the j-th patient and each of the other patients of this disease type among the known patients of the currently analyzed disease type, that is, it represents the difference distance between the examination data of this patient and the examination data of the other patients of this disease type; because the theory of model prediction here is that the greater the difference in examination data between patients, the greater the severity of the disease of this patient, so there is a positive correlation to a certain extent; Therefore, when the value is smaller, that is, when the value is larger, it indicates that the similarity between the disease severity of this patient and the patient metric distance is higher. Then, when predicting the disease risk index in the future, the reference weight when it participates in the prediction is larger, and the accurate prediction result may be higher.

[0010] Further, based on the reference weight and the difference distance between any two patients, obtain the risk index of any unknown patient suffering from any type of rheumatic immune nephropathy disease. Specifically: ; Among them, represents the risk index of any unknown patient p suffering from this type of disease; represents the normalization function; p represents the predicted unknown patient; represents the difference distance between the examination data of the predicted patient P and the j-th patient among the current patients of this disease type; represents the sum of the product of the reference weight corresponding to each patient of this disease type when participating in the prediction and its difference distance from the examination data of the predicted patient p, divided by the sum of the reference weights corresponding to each patient of this disease type when participating in the prediction, so as to obtain the risk index of the predicted patient p suffering from this type of disease.

[0011] Further, determine the relationship between the risk index and the set threshold. If it is greater, it is determined that the current patient has a rheumatic immune nephropathy type disease, specifically: If the risk index of any type of disease in the currently predicted patient for the classified type of disease is greater than the threshold T, then it indicates that the patient may have the risk of this type of disease, or monitor the currently predicted patient in terms of time series. When it is found that the disease risk index shows an upward trend in time series, that is, its slope is greater than 0, it indicates that the patient has the tendency of the risk of this type of immune disease.

[0012] An early warning system for rheumatic immune nephropathy based on data analysis includes: A division module, which obtains the medical test data of rheumatic immune nephropathy patients and divides the rheumatic immune nephropathy patients into different types according to the type of immune disease; A normalization processing module, which performs normalization processing on different item inspection data in the medical test data of patients of any type based on the patients of any type; A first acquisition module, which obtains the correlation coefficient between any item of inspection data and the severity of the patient's rheumatic immune disease based on the normalized different item inspection data; A second acquisition module, which performs a merging process on the correlation coefficients of all item inspection data to obtain the final correlation coefficient between any item of inspection data and this type of disease; A difference distance acquisition module, which obtains the difference distance between any two patients among the patients of this type of disease based on the obtained final correlation coefficient; A reference weight acquisition module, which obtains the reference weight of the patients of this type of disease when estimating the risk index of an unknown patient suffering from this type of disease based on the severity of the patient's disease and the difference distance; the severity of the patient's disease is obtained by normalizing the disease severity score value, and the disease severity score value is the data in the medical test data of rheumatic immune nephropathy patients; A risk index acquisition module, which obtains the risk index of any unknown patient suffering from any type of rheumatic immune nephropathy disease based on the reference weight and the difference distance between any two patients; A judgment module, which judges the relationship between the risk index and the set threshold. If it is greater, it is determined that the current patient has a rheumatic immune nephropathy type disease.

[0013] Compared with the prior art, the present invention has the following beneficial effects: The present invention classifies patients with rheumatic immune nephropathy into different types of diseases. For any type of patient, the correlation coefficient between any item of examination data of the patient and the severity of the disease is obtained, and the final correlation coefficient of any item of examination data of the patients conforming to this type of disease is obtained. Based on the difference distance of the examination data and the difference distance of the disease severity between the patients of this type, the weight for predicting the risk index of an unknown patient having this type of disease is obtained. By combining the examination data of the patient to be predicted and the relevant data of the patients of this type, the predicted value of the risk index of the unknown patient having this type of disease is obtained. Then, by combining the single risk index predicted value or the change of its multiple risk indexes in time series, a more accurate and timely warning is carried out, so as to provide more accurate medical decisions for patients in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0015] Figure 1 It is a schematic flow chart of the early warning method for rheumatic immune nephropathy based on data analysis of the present invention; Figure 2 It is a schematic structural diagram of the early warning system for rheumatic immune nephropathy based on data analysis of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Usually, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0017] Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed present invention, but merely represents the selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0018] It should be noted that similar reference numerals and letters indicate similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0019] In the description of the embodiments of the present invention, it should be noted that if terms such as "upper", "lower", "horizontal", "inner", etc. are used to indicate the orientation or positional relationship, it is based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the invention product is usually placed during use. This is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, so it should not be construed as a limitation to the present invention. In addition, terms such as "first", "second", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0020] In addition, if the term "horizontal" appears, it does not mean that the component is required to be absolutely horizontal, but it can be slightly inclined. For example, "horizontal" only means that its direction is more horizontal relative to "vertical", and it does not mean that the structure must be completely horizontal, but it can be slightly inclined.

[0021] In the description of the embodiments of the present invention, it should also be noted that unless otherwise clearly specified and limited, if terms such as "set", "installed", "connected", "linked" are used, they should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0022] The following further describes the present invention in detail with reference to the drawings: See Figure 1 , the present invention discloses an early warning method for rheumatic immune nephropathy based on data analysis, including: S101, obtaining the medical test data of patients with rheumatic immune nephropathy, and classifying the patients with rheumatic immune nephropathy into different types according to the types of immune diseases; S102, based on any type of patients, performing normalization processing on different item inspection data in the medical test data of this type of patients; S103, based on the normalized different item inspection data, obtaining the correlation coefficient between any item of inspection data and the severity of the rheumatic immune disease of the patient; ; Wherein, represents the correlation coefficient between the i-th item of inspection data in the currently analyzed patient and the severity of this type of disease of the patient; represents the absolute value operation symbol; represents the corresponding value of the i-th normalized inspection data in the currently analyzed patient; represents the reference range when the i-th normalized examination data of the currently analyzed patient is normal; represents the normalized severity of this type of disease of the currently analyzed patient; represents the median of the reference range when the i-th normalized examination data of the currently analyzed patient is normal, that is, represents the reference value of this examination data; represents the ratio of the absolute value of the difference between the value corresponding to the i-th examination data of the currently analyzed patient and the reference value of this examination data to the severity of the disease suffered by this patient. If this value is larger, the relevance of the amplitude change of this item of data to this type of disease of this patient is higher, otherwise the relevance of this examination data to the type of disease suffered by this patient is lower.

[0023] S104, perform a union process on the correlation coefficients of all items of examination data to obtain the final correlation coefficient between any item of examination data and this type of disease; Collect the correlation coefficients of all items of examination data of this type of patient as the target set; and arrange the correlation coefficients of the correlation coefficient set in order, select the median of the sorted set, calculate the absolute value of the difference between each correlation coefficient in this set and the selected median, and use it as the distance value between each correlation coefficient and this median data; based on the distance value between each correlation coefficient and the median and the numerical size of the corresponding correlation coefficient, obtain the final correlation coefficient between any item of examination data and this type of disease.

[0024] The obtaining of the final correlation coefficient between any item of examination data and this type of disease based on the distance value between each correlation coefficient and the median and the numerical size of the corresponding correlation coefficient is specifically: ; where represents the final correlation coefficient between the i-th examination data and this type of disease of the currently analyzed type of disease patient; represents the median of the target correlation coefficient set sorted by the i-th examination data of the currently analyzed type of disease patient; represents the number of elements in the target correlation coefficient set corresponding to the i-th examination data of the currently analyzed type of disease patient; 、 respectively represent the s-th correlation coefficient value sorted by the i-th examination data of the currently analyzed type of disease patient in the target correlation coefficient set and the distance value between this correlation coefficient and this median data; c represents a constant term to prevent the denominator from being 0.

[0025] S105, based on the obtained final correlation coefficient, obtain the difference distance between any two patients among the patients of this type of disease; ; Among them, represents the difference distance between patient a and patient b among the patients of the currently analyzed disease type, represents the square root operation; n represents the number of data items examined for these two patients among the patients of the currently analyzed disease type; represents the final correlation coefficient obtained from the i-th examination data in the currently analyzed disease type; 、 respectively represent the normalized i-th examination data values of patient a and patient b among the patients of the currently analyzed disease type; represents the absolute value of the difference between the product of the final correlation coefficient obtained from the i-th examination data in the currently analyzed disease type and the i-th examination data values of patient a and patient b among the patients of the currently analyzed disease type. If the final correlation coefficient of this examination data is larger, it indicates that the contribution of this examination data to the diagnosis and diagnosis of this disease type is higher. Then, when the difference between the examination data of these two patients is smaller, the disease symptoms of these two patients may be more similar, and the difference distance between the disease symptoms and disease severity of these two patients may be more similar. represents the square root of the sum of the squares of the product of the absolute value of the difference between the normalized examination data values of each identical item of patient a and patient b among the patients of the currently analyzed disease type and their correlation coefficients related to this type of disease, that is, it represents the difference distance between these two patients for this type of disease.

[0026] S106. Based on the disease severity and difference distance of the patient, obtain the patients of this disease type as the reference weight when estimating the risk index of an unknown patient suffering from this type of disease in the future; the disease severity of the patient is obtained by normalizing the disease severity score value, and the disease severity score value is the data in the medical test data of patients with rheumatic immune nephropathy; ; Among them, represents the reference weight of the j-th patient among the known patients of the currently analyzed disease type when estimating the risk of an unknown patient suffering from this disease type in the future; 、 represent the disease severities of the j-th patient and the r-th patient among the known patients of the currently analyzed disease type; g represents the number of currently known patients of this disease type; represents the overall difference between the disease severity value of the j-th patient suffering from this disease type and the disease severity values of the other patients of this disease type among the known patients of the currently analyzed disease type; It represents the total difference distance between the j-th patient and each of the remaining patients among the known patients of the currently analyzed type of disease, that is, it represents the difference distance between the examination data of this patient and the examination data of the remaining patients of this disease type; because the theory predicted by the model here is that the greater the difference in examination data between patients, the greater the severity of the disease of this patient, so it shows a positive correlation to a certain extent; Therefore, when the value is smaller, that is when the value is larger, it indicates that the similarity between the severity of the disease of this patient and the patient metric distance is higher. Then, when predicting the disease risk index subsequently, the reference weight when it participates in the prediction is larger, and the accurate prediction result may be higher.

[0027] S107. Based on the reference weight and the difference distance between any two patients, obtain the risk index of any unknown patient suffering from any type of rheumatic immune nephropathy disease; ; Among them, represents the risk index of any unknown patient p suffering from this type of disease; represents the normalization function; p represents the predicted unknown patient; represents the difference distance between the examination data of the predicted patient P and the j-th patient among the current patients of this type of disease; represents the sum of the product of the reference weight corresponding to each patient of this type of disease when participating in the prediction and its difference distance from the examination data of the predicted patient p divided by the sum of the reference weights corresponding to each patient of this type of disease when participating in the prediction, so as to obtain the risk index of the predicted patient p suffering from this type of disease.

[0028] S108. Judge the relationship between the risk index and the set threshold. If it is greater, it is determined that the current patient suffers from rheumatic immune nephropathy type disease.

[0029] If the risk index of any type of disease of the currently predicted patient for the classified type of disease is greater than the threshold T, then it indicates that it may have the risk of suffering from this type of disease, or monitor the currently predicted patient in terms of time series. When it is found that its disease risk index shows an upward trend in time series, that is, its slope is greater than 0, it indicates that it has the trend of the risk of this type of immune disease.

[0030] See Figure 2 , the present invention discloses an early warning system for rheumatic immune nephropathy based on data analysis, including: A division module, the division module obtains the medical test data of rheumatic immune nephropathy patients, and divides the rheumatic immune nephropathy patients into different types according to the type of immune disease; A normalization processing module, which performs normalization processing on different item inspection data in the medical test data of patients of any type based on patients of any type; A first acquisition module, which acquires the correlation coefficient between the inspection data of any item and the severity of the patient's rheumatic immune disease based on the normalized different item inspection data; A second acquisition module, which performs a union process on the correlation coefficients of all item inspection data to obtain the final correlation coefficient between the inspection data of any item and the disease type; A difference distance acquisition module, which acquires the difference distance between any two patients among the patients of this disease type based on the obtained final correlation coefficient; A reference weight acquisition module, which acquires the reference weight of the patients of this disease type when participating in the risk index prediction of unknown patients suffering from this type of disease based on the severity of the patient's disease and the difference distance; the severity of the patient's disease is obtained by normalizing according to the disease severity score value; the disease severity score value is the data in the medical test data of patients with rheumatic immune nephropathy; A risk index acquisition module, which acquires the risk index of any unknown patient suffering from any type of rheumatic immune nephropathy disease based on the reference weight and the difference distance between any two patients; A judgment module, which judges the relationship between the risk index and the set threshold. If it is greater, it is determined that the current patient suffers from a rheumatic immune nephropathy type disease.

[0031] Embodiment: The present invention discloses an early warning method for rheumatic immune nephropathy based on data analysis, including: Step 1: Collect relevant medical data of patients using medical and other related devices.

[0032] Collect relevant data of patients through manual input and machine collection, including the basic information of patients, such as name, gender, etc., and clinically collected information, such as blood routine, urine routine, renal function index data, the disease type of patients with rheumatic immune diseases, the disease severity score value, etc.

[0033] Step 2: Divide patients with rheumatic immune nephropathy into different types according to the type of immune disease.

[0034] In order to more accurately predict a patient's rheumatic immune nephropathy, first, the examination data of patients already suffering from rheumatic immune nephropathy are collected, so as to use it as relevant reference data for subsequent risk index analysis of the incidence and type of rheumatic immune nephropathy in patients without rheumatic immune diseases. Since among the data collected from patients, the deviation of different item examination data may have different degrees of influence on their rheumatic immune nephropathy and the risk index of disease occurrence, therefore, the risk index of each item of examination data for the prediction of rheumatic immune nephropathy is also different.

[0035] Patients already suffering from rheumatic immune nephropathy are divided into different types according to their immune disease types, such as systemic lupus erythematosus (SLE) nephritis, antineutrophil cytoplasmic antibody (ANCA)-associated nephritis, etc. Then, based on any type of patient, normalization processing is performed on different item examination data in the medical test data of this type of patient.

[0036] Taking any type as an example for analysis, for example, at this time, the patients with systemic lupus erythematosus SLE nephritis type are analyzed.

[0037] Normalization processing is performed on any single item of data collected from this type of patient, such as white blood cells, red blood cells, etc., to avoid the influence of differences brought by different data dimensions.

[0038] For any item of data in this type of patient, the correlation coefficient of the severity of the rheumatic immune disease type suffered by this patient can be expressed as: ; Among them, represents the correlation coefficient between the i-th item of examination data in the currently analyzed patient and the severity of this type of disease suffered by this patient; represents the absolute value operation symbol; represents the corresponding value of the i-th normalized examination data in the currently analyzed patient; represents the reference range when the i-th normalized examination data in the currently analyzed patient is normal; represents the normalized severity of this type of disease in the currently analyzed patient; represents the median value of the reference range when the i-th normalized examination data in the currently analyzed patient is normal, that is, represents the reference value of this item of examination data; represents the ratio of the absolute value of the difference between the corresponding value of the i-th examination data in the currently analyzed patient and the reference value of this item of examination data to the severity of the disease suffered by this patient. If this value is larger, the higher the relevance of the amplitude change of this item of data to this type of disease of this patient, otherwise the lower the correlation between this item of examination data and the disease type suffered by this patient.

[0039] Step 3: Obtain the reference weight when a patient with any type of disease participates in the subsequent risk prediction of this type of disease for unknown patients.

[0040] After obtaining the correlation coefficient between any type of patient's any test item data and the disease severity, for patients of the same type, the correlation coefficient between each patient's corresponding data and the disease severity can be obtained. Theoretically speaking, the correlation coefficients of this item of data for these patients are relatively close. However, in order to avoid the uneven distribution of these correlation coefficients caused by the differences in individual attributes of a very small number of patients, etc., which may affect the accuracy of determining the final correlation coefficient.

[0041] Step 3.1, perform a merging process on the correlation coefficients of all test data items to obtain the final correlation coefficient between any test data item and this type of disease; First, take the correlation coefficients of any test item data of this type of patient as an analysis target set, and call it the target correlation coefficient set.

[0042] Then arrange this correlation coefficient set in order. Here, ascending order is taken as an example for analysis.

[0043] Then select the median of the sorted set. Since the median is not easily affected by abnormal or uneven data distributions and has a relatively high accuracy as a reference, in order to consider the overall situation of the current inspection coefficients of all patients for this item and combine its overall distribution, the overall correlation coefficient of this item of data for this type of disease can be obtained.

[0044] Therefore, further calculate the absolute value of the difference between the remaining elements in the set and the selected median, and take it as the distance value between the remaining elements and this median data.

[0045] Then, for the correlation coefficient between this test data item and this type of disease, it can be obtained by a weighted method, so as to reduce the influence caused by the particularity of individual attributes in the data collected from patients with this type of disease or the deviation in the sampling data records, and then improve the accuracy of the subsequent risk index prediction and warning for unknown patients with this type of rheumatic immune disease leading to kidney disease.

[0046] Therefore, for patients with this type of disease, the calculation formula for the final correlation coefficient between the corresponding test data item and this type of disease is as follows: ; Wherein, represents the final correlation coefficient between the i-th test data item and this type of disease among patients with the currently analyzed type of disease; represents the median after sorting the set of target correlation coefficients corresponding to the i-th examination data among patients with the currently analyzed type of disease; represents the number of elements in the set of target correlation coefficients corresponding to the i-th examination data among patients with the currently analyzed type of disease; 、 respectively represent the s-th correlation coefficient value after sorting the set of target correlation coefficients corresponding to the i-th examination data among patients with the currently analyzed type of disease and the distance value between this correlation coefficient and the median data; c represents a constant term to prevent the denominator from being zero.

[0047] Step 3.2, based on the obtained final correlation coefficients, obtain the difference distance between any two patients among patients with this type of disease.

[0048] Then, for patients with this type of disease, the difference distance between any two patients can be expressed as: ; where, represents the difference distance between patient a and patient b among patients with the currently analyzed type of disease, represents the square root operation; n represents the number of examination data items of these two patients among patients with the currently analyzed type of disease; represents the final correlation coefficient obtained from the i-th examination data and this type of disease among the currently analyzed type of disease; 、 respectively represent the normalized i-th examination data values of patient a and patient b among patients with the currently analyzed type of disease; represents the absolute value of the difference between the final correlation coefficient of the i-th examination data and the product of the i-th examination data value corresponding to patient a and patient b among patients with the currently analyzed type of disease. If the final correlation coefficient of this examination data is larger, it indicates that the contribution of this examination data to the diagnosis and diagnosis of this type of disease is higher. Then, when the difference between the examination data of these two patients is smaller, the symptoms of these two patients may be more similar, and the difference distance between the symptoms and disease severity of these two patients may be more similar. represents the square root of the sum of the squares of the product of the absolute value of the difference between the normalized i-th examination data values of patient a and patient b corresponding to each same item of examination data and their correlation coefficients related to this type of disease, that is, it represents the difference distance between these two patients for this type of disease.

[0049] Through the above, the measurement distance between any two patients under any type of disease can be obtained.

[0050] Step 3.3: Based on the severity of the patient's condition and the difference distance, obtain the patients of this disease type as the reference weight when estimating the risk index of an unknown patient suffering from this type of disease in the future.

[0051] When calculating the risk index of an unknown patient suffering from this type of disease, it is necessary to comprehensively analyze the distance and severity to obtain its weight for future reference.

[0052] Therefore, for the patients of this disease type, the similarity between the severity of their disease and the patient measurement distance is used to obtain their reference weight when participating in the risk index estimation of an unknown patient suffering from this type of disease in the future. The calculation formula is as follows: ; Among them, represents the reference weight of the j-th patient among the known patients of the disease type being analyzed currently when estimating the risk of an unknown patient of this disease type in the future; 、 represent the severity of the disease of the j-th patient and the r-th patient among the known patients of the disease type being analyzed currently; g represents the number of known patients of this disease type; represents the overall difference between the severity value of the j-th patient suffering from this disease type and the severity values of the other patients of this disease type among the known patients of the disease type being analyzed currently; represents the sum of the difference distances between the j-th patient and each of the other patients of this disease type among the known patients of the disease type being analyzed currently, that is, the difference distance between the examination data of this patient and the examination data of the other patients of this disease type; because the theory of model prediction here is that the greater the difference in examination data between patients, the greater the severity of the disease of this patient, so there is a positive correlation to a certain extent; Therefore, when the value is smaller, that is, when the value is larger, it indicates that the similarity between the severity of the disease of this patient and the patient measurement distance is higher. Then, when predicting the disease risk index in the future, the reference weight when participating in the prediction is larger, and the accurate prediction result may be higher.

[0053] Step 4: Based on the reference weight and the difference distance between any two patients, obtain the risk index of any unknown patient suffering from any type of rheumatic immune nephropathy.

[0054] After obtaining the reference weight of any disease type patient corresponding to the prediction of an unknown patient suffering from this disease, the risk index of an unknown patient corresponding to suffering from this type of disease can be expressed as: ; Among them, represents the risk index of any unknown patient p suffering from this type of disease; represents a normalization function; p represents the predicted unknown patient; represents the difference distance of the examination data between the predicted patient P and the j-th patient among the current patients with this type of disease; represents the sum of the reference weights corresponding to each patient with this type of disease multiplied by the difference distance of their examination data from the predicted patient p, divided by the sum of the reference weights corresponding to each patient with this type of disease participating in the prediction, so as to obtain the risk index of the predicted patient p suffering from this type of disease.

[0055] Step 5, give early warning and prompt for the predicted patient suffering from rheumatic immune nephropathy type diseases.

[0056] After obtaining the risk indices of different arbitrary patients suffering from different types of rheumatic immune diseases, then set a threshold T, where T is set to 0.2. If for the current predicted patient among the classified types of diseases, the risk index of any type of disease is greater than the threshold T, then it indicates that there may be a risk of suffering from this type of disease, or conduct temporal monitoring on the current predicted patient. When it is found that the risk index shows an upward trend in time series, that is, its slope is greater than 0, it indicates a trend of suffering from this type of immune disease risk, and timely intervention and treatment are required to avoid delay of the condition, thereby causing greater damage to the kidneys and the body, etc.

[0057] The terminal device provided by the embodiment of the present invention. The terminal device of this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in the above-mentioned various method embodiments. Or, when the processor executes the computer program, it implements the functions of each module / unit in the above-mentioned various device embodiments.

[0058] The computer program can be divided into one or more modules / units, and the one or more modules / units are stored in the memory and executed by the processor to complete the present invention.

[0059] The terminal device can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The terminal device may include, but is not limited to, a processor and a memory.

[0060] The processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0061] The memory can be used to store the computer program and / or module. By running or executing the computer program and / or module stored in the memory, and by invoking the data stored in the memory, the processor realizes various functions of the terminal device.

[0062] If the modules / units integrated in the terminal device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-described various method embodiments can be realized. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0063] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An early warning method for rheumatic immune nephropathy based on data analysis, characterized in that: include: Obtain medical examination data of patients with rheumatic immune nephropathy and classify patients with rheumatic immune nephropathy into different types according to the type of immune disease; Based on any type of patient, normalizing different items of examination data in the medical examination data of the type of patient respectively; Based on the normalized examination data of different items, the correlation coefficient between any examination data and the severity of the patient's rheumatic autoimmune disease is obtained; The correlation coefficients of all the examination data are collected and processed to obtain the final correlation coefficient between any examination data and the disease type; Based on the obtained final correlation coefficient, the difference distance between any two patients of the disease type is obtained; Based on the severity of the patient's disease and the difference distance, the patient of this disease type is obtained as a reference weight for the subsequent estimation of the risk index of unknown patients suffering from this type of disease; the severity of the patient's disease is normalized according to the disease severity score value to obtain the severity of the patient's disease; the disease severity score value is the data in the medical examination data of patients with rheumatic immune nephropathy; Based on the reference weight and the difference distance between any two patients, a risk index of any unknown patient suffering from any type of rheumatoid immune nephropathy is obtained; The relationship between the risk index and the set threshold is determined. If it is greater than, it is determined that the current patient suffers from rheumatoid immune nephropathy type disease.

2. The early warning method for rheumatic immune nephropathy based on data analysis according to claim 1, characterized in that: The correlation coefficient between any one of the examination data and the severity of the patient's rheumatic autoimmune disease is obtained based on the normalized examination data of different items, specifically: ； in, Indicates the correlation coefficient between the i-th examination data of the patient currently analyzed and the severity of the type of disease suffered by the patient; Indicates the absolute value operator symbol; Indicates the corresponding value of the normalized examination data of the i-th item in the currently analyzed patient; Indicates the normal reference range of the normalized examination data of the i-th item in the currently analyzed patient; Indicates the normalized severity of the type of disease in the patient currently being analyzed; It indicates the median value of the normalized examination data of the i-th item in the patient currently analyzed, that is, the reference value of the examination data; It represents the ratio of the absolute value of the difference between the corresponding value of the i-th examination data of the currently analyzed patient and the reference value of the examination data to the severity of the patient's disease. If the value is larger, the correlation between the amplitude change of the data and the type of disease of the patient is higher. Otherwise, the correlation between the examination data and the type of disease of the patient is lower.

3. The early warning method for rheumatic immune nephropathy based on data analysis according to claim 2, characterized in that: The described method of aggregating and processing the correlation coefficients of all items of examination data to obtain the final correlation coefficient between any item of examination data and the disease type is as follows: aggregating the correlation coefficients of all items of examination data of patients of this type as a target set; arranging the correlation coefficients of the correlation coefficient set in sequence, selecting the median of the sorted set, calculating the absolute value of the difference between each correlation coefficient in the set and the selected median, and using it as the distance value between each correlation coefficient and the median data; based on the distance value between each correlation coefficient and the median and the corresponding numerical value of each correlation coefficient, obtaining the final correlation coefficient between any item of examination data and the disease type.

4. The early warning method for rheumatic immune nephropathy based on data analysis according to claim 3, characterized in that: The method of obtaining the final correlation coefficient between any item of examination data and the disease type based on the distance between each correlation coefficient and the median and the corresponding numerical value of each correlation coefficient is as follows: ； in, It represents the final correlation coefficient between the i-th examination data and the type of disease in the patients with the currently analyzed type of disease; It represents the median of the sorted set of target correlation coefficients corresponding to the i-th examination data among the patients with the currently analyzed type of disease; Indicates the number of elements in the target correlation coefficient set corresponding to the i-th examination data in the patients with the disease type currently analyzed; , They respectively represent the sth correlation value of the target correlation coefficient set corresponding to the i-th examination data among patients with the type of disease currently being analyzed, and the distance value between the correlation coefficient and the median data; c represents a constant term to prevent the denominator from being zero.

5. The early warning method for rheumatic immune nephropathy based on data analysis according to claim 4, characterized in that: The method of obtaining the difference distance between any two patients of the disease type based on the obtained final correlation coefficient is as follows: ； in, Indicates the difference distance between the patients of the currently analyzed disease type, patient A and patient B, represents the square root operation; n represents the number of data items examined for these two patients among the patients with the disease type currently being analyzed; Indicates the final correlation coefficient between the i-th examination data and the disease type currently being analyzed; , They represent the normalized data values ​​of the i-th examination item corresponding to the patient of the currently analyzed disease type, patient a and patient b respectively; It means that in the currently analyzed disease type, the final correlation coefficient between the i-th examination data and the disease type is multiplied by the absolute value of the difference between the values ​​of the i-th examination data of the patients of the currently analyzed disease type, patients A and B. If the final correlation coefficient of the examination data is larger, it means that the contribution of the examination data to the diagnosis and diagnosis of the disease type is higher. Then, when the difference between the examination data of the two patients is smaller, the symptoms of the two patients may be closer, and the difference in the symptoms and severity of the two patients may be closer. It means that for patients with the type of disease currently being analyzed, the square root of the sum of the squares of the absolute value of the difference in normalized examination data values ​​for each identical item between patient A and patient B multiplied by the product of their correlation coefficients with the type of disease is the difference distance between the two patients for the type of disease.

6. The early warning method for rheumatic immune nephropathy based on data analysis according to claim 5, characterized in that: The patient with the disease type is obtained based on the severity of the patient's disease and the difference distance as a reference weight for subsequent risk index estimation of unknown patients with the disease type, specifically: ； in, Indicates that among the patients with known diseases of the type currently analyzed, the jth patient is used as a reference weight for subsequent risk estimation of unknown patients with the disease of the type being analyzed; , represents the severity of the disease of the jth patient and the rth patient among the known patients of the disease type currently analyzed; g represents the number of patients with the disease type currently known; It indicates the overall difference between the severity value of the disease type of the jth patient among the known patients of the disease type currently analyzed and the severity value of the disease type of the rest of the patients of the same type; It indicates the sum of the difference distances between the jth patient and the other patients with the same type of disease among the known patients with the disease type currently being analyzed, that is, it indicates the difference distance between the examination data of this patient and the examination data of the other patients with the same disease type; because the theory predicted by the model here is that the greater the difference in the examination data between patients, the greater the severity of the patient's disease, so it shows a positive correlation to a certain extent; Therefore, when The smaller the value, the The larger the value is, the higher the similarity between the severity of the patient's disease and the patient's measured distance is. Therefore, when predicting the disease risk index in the future, the greater the reference weight it participates in the prediction, and the more accurate the prediction result may be.

7. The early warning method for rheumatic immune nephropathy based on data analysis according to claim 6, characterized in that: The risk index of any unknown patient suffering from any rheumatic immune nephropathy type disease is obtained based on the reference weight and the difference distance between any two patients, specifically: ； in, represents the risk index of any unknown patient p suffering from this type of disease; represents the normalization function; p represents the predicted unknown patient; Represents the difference distance between the predicted patient P and the jth patient among the current patients with the same type of disease; The risk index of predicting patient p with this type of disease is obtained by multiplying the sum of the difference distances between the patient's examination data and the predicted patient p by the sum of the reference weights of each patient with this type of disease.

8. The early warning method for rheumatic immune nephropathy based on data analysis according to claim 7, characterized in that: The relationship between the risk index and the set threshold is determined to be greater than, then the current patient is determined to suffer from rheumatoid immune nephropathy type disease, specifically: If the risk index of the currently predicted patient for any type of disease among the classified types of diseases is greater than the threshold T, it means that the patient may be at risk of suffering from this type of disease. Alternatively, the currently predicted patient is monitored over time. When it is found that his disease risk index shows an upward trend over time, that is, its slope is greater than 0, it means that he has a trend of risk of this type of immune disease.

9. An early warning system for rheumatic immune nephropathy based on data analysis, characterized in that: include: A classification module, wherein the classification module obtains medical examination data of patients with rheumatic immune nephropathy and classifies the patients with rheumatic immune nephropathy into different types according to the types of immune diseases; A normalization processing module, which performs normalization processing on different items of medical examination data of any type of patient respectively; A first acquisition module, which acquires a correlation coefficient between any one of the examination data and the severity of the patient's rheumatism-immunity disease based on the normalized different examination data; A second acquisition module, wherein the second acquisition module aggregates and processes the correlation coefficients of all items of examination data to obtain a final correlation coefficient between any item of examination data and the disease type; A difference distance acquisition module, wherein the difference distance acquisition module acquires the difference distance between any two patients of the disease type based on the acquired final correlation coefficient; A reference weight acquisition module, which acquires patients of the same disease type as reference weights for subsequent estimation of the risk index of unknown patients suffering from the same disease type based on the disease severity and difference distance of the patient; the disease severity of the patient is obtained by normalizing the disease severity score value; the disease severity score value is data in the medical examination data of patients with rheumatic immune nephropathy; A risk index acquisition module, which acquires the risk index of any unknown patient suffering from any rheumatic immune nephropathy type disease based on the reference weight and the difference distance between any two patients; A judgment module is used to judge the relationship between the risk index and a set threshold value. If the relationship is greater than the threshold value, it is determined that the current patient suffers from rheumatic immune nephropathy.

Images