Children antinuclear antibody positive immune thrombocytopenia patient database establishment method and device

Abstract

The invention discloses a child antinuclear antibody positive immune thrombocytopenia patient database establishment method and device. The child antinuclear antibody positive immune thrombocytopenia patient database establishment method comprises the following steps: acquiring a child patient basic screening data set and child patient subdivision clinical data provided by multiple centers; acquiring a structured full-dimensional data set of the child patient; carrying out data standardization and unified coding processing on the data so as to form a child patient standardization data set; performing permission setting distribution on each piece of data in the child patient standardized data set to form an encrypted security data set; and constructing a child antinuclear antibody positive immune thrombocytopenia patient database according to the encrypted security data set. According to the database established by the method disclosed by the invention, immune characteristics such as antinuclear antibody spectrum, anti-ENA spectrum, complement, lymphocyte subpopulation and the like as well as platelet level, clinical manifestation and treatment reaction are comprehensively recorded, so that a doctor can pay attention to immunology abnormality of a patient earlier during recording.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of database establishment, in particular to a database establishment method and device for children with anti-nuclear antibody positive immune thrombocytopenia. BACKGROUND

[0002] Primary immune thrombocytopenia (ITP) is an acquired autoimmune and hemorrhagic disease. The disease has obvious heterogeneity, and there are differences in clinical manifestations and prognosis. 75-90% of children with the disease can be effectively treated by initial first-line treatment and can be relieved within one year, but part of the children will eventually progress to chronic refractory ITP even after standard treatment. Previous studies have shown that 13%-65% of children with ITP have anti-nuclear antibodies (ANA), and children with ANA are more likely to progress to chronic refractory ITP or other autoimmune diseases, and have a poorer prognosis. Moreover, the efficacy of first-line treatments such as human immunoglobulin (IVIG) and glucocorticoids, and second-line treatment drugs such as TPO receptor agonists (TPO-RA) and rituximab for these children is significantly different from that for children with ANA-negative.

[0003] Previous studies by our team have shown that children with ANA-positive ITP have a significantly increased risk of developing autoimmune diseases such as SLE, which is 7.17 times that of children with ANA-negative ITP, and the clinical manifestations of SLE resulting from the former are different from those of SLE directly. In terms of treatment, TPO-RA, which is the preferred second-line treatment, has been shown in a small sample of single-center retrospective studies to have not achieved good results in ANA-positive ITP patients. Rituximab, on the other hand, may have better efficacy in ANA-negative patients. The significant difference in drug efficacy also suggests that ANA-positive ITP should be distinguished from ordinary ITP patients, indicating that there are unique characteristics in terms of treatment options and efficacy. We speculate that ANA-positive ITP may be a more worthy of attention subclass of ITP.

[0004] Therefore, it is urgent to establish a multi-center database of ANA-positive ITP children in China to in-depth analyze the clinical characteristics and long-term outcomes of ANA-positive ITP children, explore the intrinsic immunopathogenesis of the disease, and establish a precise diagnosis and stratified treatment system guided by molecular markers. In addition, we innovatively include lymphocyte subsets, complement, and anti-ENA spectrum into the database in order to provide data support for exploring the clinical characteristics of ANA-positive ITP children.

[0005] Therefore, it is desirable to have a technical solution to overcome or at least alleviate at least one of the aforementioned deficiencies of the prior art.

[0006] SUMMARY

[0007] The present application aims to provide a database establishment method for children with anti-nuclear antibody positive immune thrombocytopenia to overcome or at least alleviate at least one of the above-mentioned defects of the prior art.

[0008] To achieve the above-mentioned purpose, the present application provides a database establishment method for children with anti-nuclear antibody positive immune thrombocytopenia, which comprises:

[0009] Obtaining a basic screening data set of the children and subdivided clinical data of the children provided by multiple centers;

[0010] According to the basic screening data set of the children and the subdivided clinical data of the children provided by multiple centers, a structured full-dimension data set of the children is formed by entering;

[0011] The data in the structured full-dimension data set of the children is standardized and uniformly coded, thereby forming a standardized data set of the children;

[0012] The permissions of each data in the standardized data set of the children are set and distributed, thereby forming an encrypted secure data set;

[0013] According to the encrypted secure data set, a database for children with anti-nuclear antibody positive immune thrombocytopenia is constructed.

[0014] Optionally, before the database for children with anti-nuclear antibody positive immune thrombocytopenia is constructed according to the encrypted secure data set, the database establishment method for children with anti-nuclear antibody positive immune thrombocytopenia further comprises:

[0015] Obtaining preset clinical empirical parameters;

[0016] According to the preset clinical empirical parameters and the encrypted secure data set, a dynamic risk stratification-treatment response prediction report is established for each child in the encrypted secure data set.

[0017] Optionally, the preset clinical empirical parameters include core variable data, and the core variable data includes the follow-up ANA titer, the relative baseline change of lymphocyte subgroup B cell level, the platelet count fluctuation coefficient, the ICR bleeding classification, and the response rate of the previous treatment scheme;

[0018] The dynamic risk stratification-treatment response prediction report established for each child in the encrypted secure data set according to the preset clinical empirical parameters and the encrypted secure data set comprises:

[0019] The core variable data is standardized to obtain standardized core variable data;

[0020] According to the standardized core variable data, a dynamic risk score and individualized treatment response prediction information are obtained.

[0021] Optionally, the standardization of the core variable data to obtain the standardized core variable data comprises:

[0022] The standardization of the follow-up ANA titer of each patient is performed by the following method:

[0023] According to the follow-up data of the ANA-positive ITP patient, the follow-up ANA titer is graded, and the grading comprises three grades, namely negative, low titer and high titer, wherein the titer ≤ 1:80 is low titer, and the titer ≥ 1:160 is high titer.

[0024] The follow-up ANA titer is standardized according to the grading, so as to obtain the standardized ANA titer.

[0025] Optionally, the standardization of the relative baseline change of the B cell level of the lymphocyte subpopulation of each patient is performed by the following method:

[0026] An absolute value correction coefficient is generated according to the detection value of the B cell level of the lymphocyte subpopulation of the patient.

[0027] The relative baseline change of the B cell level of the lymphocyte subpopulation is standardized according to the following formula:

[0028]

[0029] Wherein, S ΔB is the relative baseline change of the B cell level of the lymphocyte subpopulation; ΔB adjust is the corrected relative baseline change rate of the B cell level of the lymphocyte subpopulation.

[0030] Optionally, the platelet count fluctuation coefficient is obtained by the following formula:

[0031] All platelet count detection values within the last one month of the patient are extracted.

[0032] The mean value of all platelet count detection values is calculated by an exponentially weighted moving average method.

[0033] According to the mean value of all platelet count detection values and all platelet count detection values within the last one month of the patient, a time-weighted fluctuation standard deviation is obtained to obtain a time-weighted fluctuation coefficient.

[0034] The platelet count fluctuation coefficient is standardized by the following formula:

[0035]

[0036] wherein S CV-PLT represents the standardized value of platelet fluctuation coefficient; CV PLT_ewm is the time-weighted fluctuation coefficient of platelet count.

[0037] Optionally, the standardization of the ICR bleeding grade of each child is performed by the following method:

[0038] The specific bleeding manifestations are obtained from the follow-up records, including skin bleeding (such as petechia, ecchymosis, purpura), mucosal bleeding (such as epistaxis, gingival bleeding, oral mucosal bleeding, ocular conjunctival bleeding), organ bleeding (such as gastrointestinal bleeding, respiratory tract bleeding, urinary tract bleeding), central nervous system bleeding;

[0039] The basic assignment and site weight are set for different bleeding manifestations respectively:

[0040] Skin bleeding: basic assignment 1.0, site weight 1.0 (low risk);

[0041] Mucosal bleeding: basic assignment 2.0, wherein the site weight of epistaxis and gingival bleeding is 1.5, and the site weight of oral mucosal bleeding and ocular conjunctival bleeding is 1.8 (moderate risk);

[0042] Organ bleeding: basic assignment 3.0, wherein the site weight of gastrointestinal bleeding and urinary tract bleeding is 3.0, and the site weight of respiratory tract bleeding is 3.5 (high risk);

[0043] Central nervous system bleeding: basic assignment 4.0, site weight 5.0 (extremely high risk);

[0044] When a child has multiple bleeding manifestations at the same time, the bleeding manifestation corresponding to the highest weighted grade is taken for calculation; if there is no bleeding symptom, it is determined as skin bleeding (no bleeding risk);

[0045] The weighted grade G is calculated according to the basic assignment and site weight: Bleed_weight = basic assignment × site weight.

[0046] The standardization of the ICR bleeding grade is performed by the following formula:

[0047]

[0048] wherein S Bleed is the standardized ICR bleeding grade; 20.0 is the highest weighted grade (grade IV), and 1.0 is the lowest weighted grade (grade I); G Bleed_weight is the weighted grade.

[0049] Optionally, the response rate of the previous treatment regimen is obtained by the following method:

[0050] We extracted the follow-up records of all previous treatment regimens for the children and screened the platelet counts at multiple key time points. According to the ITP treatment guidelines, the efficacy at each key time point (1, 3, 6, 12, 18, 24 months) was determined as the response rate of the previous treatment regimen. The response rate of the previous treatment regimen included remission, partial remission, and no effect. Remission and partial remission were considered as effective times.

[0051] Obtain sustained relief time;

[0052] A weighted number of effective counts is generated based on the number of effective counts and the duration of relief.

[0053] The weighted response rate is generated based on the weighted number of valid responses and the total number of quality solutions.

[0054] Optionally, the dynamic risk score is obtained using the following formula:

[0055] RS Dynamic =(α×S ANA +β×S ΔB +γ×S CV-PLT +δ×S Bleed +∈×S RTx )×F comorbidity ×F family ;

[0056] Among them, RS Dynamic For dynamic risk scoring; S ANA For standardized ANA titers; S ΔB This represents the change in the level of the B lymphocyte subset relative to baseline; S CV-PLT S is the standardized platelet count fluctuation coefficient; Bleed Standardized ICR bleeding grading; S RTx The standardized weighted response rate is denoted as α; α is the weighting coefficient, representing the weight of the standardized ANA titer value, reflecting the influence of ANA titer on risk, with a value of 0.32; β is the weighting coefficient, representing the weight of the standardized value of the change in B-cell lymphocyte subset levels relative to baseline, with a value of 0.28; γ is the weighting coefficient, representing the weight of the standardized platelet variability coefficient value, reflecting the influence of platelet stability on risk, with a value of 0.21; δ is the weighting coefficient, representing the weight of the standardized ICR bleeding grade value, reflecting the influence of bleeding risk on overall risk, with a value of 0.12; ε is the weighting coefficient, representing the weight of the standardized treatment response rate value, reflecting the influence of previous treatment effects on current risk, with a value of 0.07; F comorbidity The positive factor, referring to the comorbidity symptom correction factor, reflects the impact of whether the child has other autoimmune-related symptoms on the risk; F familyThe correction factor, specifically the family history correction factor, reflects the impact of whether a child has a family history of autoimmune diseases on the risk. It takes a value of 0.90 or 1.0 and is based on family clustering studies.

[0057] The individualized treatment response prediction information is calculated using the following formula:

[0058]

[0059] Among them, P RTX To predict the 12-month remission probability of rituximab treatment based on changes in immune markers; P0 is the baseline probability parameter, referring to the baseline remission probability of ANA-positive ITP children using rituximab; k is the influence coefficient of B cell change rate; ΔB adjust F represents the corrected rate of change of B cells relative to baseline. age is the age correction factor; m is the negative correction coefficient for ANA titer; TANA is the standardized ANA titer; F ΔB F is a correction factor for the relative baseline change in B cell subset levels of lymphocytes; Tx-history This is a factor for correcting previous treatment history.

[0060] This application also provides a device for establishing a database of children with positive antinuclear antibody-mediated immune thrombocytopenic purpura (ITP). The device includes:

[0061] The basic information acquisition module is used to acquire the basic screening dataset of the children and the detailed clinical data of the children provided by multiple centers.

[0062] The structured full-dimensional dataset acquisition module is used to form a structured full-dimensional dataset of patients based on the basic screening dataset of patients and the detailed clinical data of patients provided by multiple centers.

[0063] The standardized dataset acquisition module for patients is used to standardize and uniformly encode the data in the structured full-dimensional dataset of patients, thereby forming a standardized dataset for patients.

[0064] An encryption module is used to set and assign permissions to each piece of data in the standardized dataset of sick children, thereby forming an encrypted secure dataset.

[0065] A database creation module is used to construct a database of children with antinuclear antibody-positive immune thrombocytopenic purpura based on an encrypted secure dataset.

[0066] The database of children with antinuclear antibody-positive immune thrombocytopenic purpura (ANTP) established by the method described in this application comprehensively records immune characteristics such as antinuclear antibody profile, anti-ENA profile, complement, and lymphocyte subsets, along with platelet levels, clinical manifestations, and treatment responses. This allows physicians to detect immunological abnormalities earlier during data entry and helps predict future disease outcomes. Attached Figure Description

[0067] Figure 1 This is a flowchart illustrating a method for establishing a database of children with positive antinuclear antibody immune thrombocytopenic purpura according to an embodiment of this application. Detailed Implementation

[0068] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be described in more detail below with reference to the accompanying drawings. In the drawings, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The described embodiments are some, but not all, embodiments of this application. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. The embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0069] In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting the scope of protection of this application.

[0070] like Figure 1 The methods for establishing the database of children with positive antinuclear antibody immune thrombocytopenic purpura (ITP) included:

[0071] Obtain basic screening datasets of children with the disease and detailed clinical data of children with the disease provided by multiple centers;

[0072] A structured, multi-dimensional dataset of patients was created by inputting basic screening data and detailed clinical data from multiple centers.

[0073] Data standardization and unified coding were performed on the structured full-dimensional dataset of sick children to form a standardized dataset of sick children;

[0074] Permissions are set and assigned to each data point in the standardized dataset of sick children, thereby forming an encrypted and secure dataset;

[0075] A database of children with positive antinuclear antibody immune thrombocytopenic purpura was constructed based on a secure, encrypted dataset.

[0076] In this embodiment, the basic screening dataset for the children can be obtained in the following way:

[0077] Obtain original data of ITP patients from multiple centers (including patient medical record number, unique ID, ITP diagnosis record, ANA test report, and basic medical information);

[0078] The pre-defined inclusion criteria (meeting the diagnostic criteria for ITP in children + having complete ANA test records during the course of the disease) and exclusion criteria (comorbid other malignant hematological diseases, data missing rate > 20%).

[0079] Using a dual-core criterion of "ITP diagnostic record + ANA test record", the study screened patient data that met the inclusion criteria through a keyword matching algorithm (identifying diagnostic terms such as "immune thrombocytopenic purpura", "ITP", "antinuclear antibody positive", "ANA(+)") and data integrity verification (confirming that key information such as ANA titer and test date is not missing).

[0080] For patient data from the same center, a dual comparison algorithm of "medical record number + unique ID" is used to automatically identify and remove duplicate patient records (such as the same patient visiting multiple times but creating duplicate entries), and suspected duplicate data is marked "awaiting manual review" to ensure the uniqueness of patient data from a single center.

[0081] In this embodiment, the basic screening dataset for patients includes the patient's unique identifier, ITP diagnosis results, and basic ANA test information. There are no duplicate records, and the data missing rate is <5%.

[0082] In this embodiment, the following method can be used to form a structured, multi-dimensional dataset of patients based on the basic screening dataset and the detailed clinical data of patients provided by multiple centers:

[0083] The data was entered into modules based on the "patient's entire life cycle," and the data was structured accordingly.

[0084] General data entry module: Collect and enter the demographic data of the children, including gender, age (accurate to month), ethnicity, date of birth (format uniformly "YYYY-MM-DD"), place of residence (province / city / county level), family history of autoimmune diseases (presence and specific disease name);

[0085] Baseline clinical data entry: Enter the clinical data (clinical manifestations, laboratory indicators) of the child when ITP was first clearly diagnosed;

[0086] Follow-up clinical data module entry: Enter data on changes in the child's condition or medication adjustments according to "time of each visit", with a focus on updating platelet count, bleeding grade, new / adjusted symptoms, and re-examined immune indicators;

[0087] Treatment plan and response module entry: Differentiate between first-line / second-line / other treatment plans, and record specific medication information, treatment response at each time point, and adverse drug reactions;

[0088] Disease outcome module entry: Enter the child's status at the last follow-up, including disease remission / continuation / progression and reasons for loss to follow-up.

[0089] In this embodiment, the structured full-dimensional dataset of the child includes general information, baseline clinical data, follow-up clinical data, treatment plan and response data, and disease outcome data.

[0090] In this embodiment, the data standardization and unified encoding of the structured full-dimensional dataset of the children with illnesses, thereby forming a standardized dataset of the children with illnesses, can be achieved by the following method:

[0091] Standardize the format: unify the format for representing date, numerical, and text data (e.g., date "YYYY-MM-DD", platelet count notation "×10"). 9 / L”);

[0092] Coding standardization: Mapping information such as disease diagnosis, bleeding classification, treatment plan, and immune indicators into industry standard codes to ensure cross-center data compatibility.

[0093] In this embodiment, permissions are set and assigned for each piece of data in the standardized dataset of sick children, thereby forming an encrypted secure dataset, including:

[0094] Real-name account creation: Generate a unique real-name account for each participating researcher (or other personnel who can access the database) (researchers can bind their center and position), and set a strong password policy (including uppercase and lowercase letters, numbers, and special symbols, and change it regularly);

[0095] Hierarchical permission allocation:

[0096] Data entry permissions: Researchers are only allowed to log in and enter / modify patient data under their own account and center, and view / retrieve patient information entered by themselves. They cannot access unauthorized data from other researchers or other centers, and do not have batch export permissions.

[0097] Administrator privileges: Allows center administrators to log in and view data of all children in the center, or the general administrator to view all data of multiple centers. Supports searching by keywords (such as "ANA titer ≥1:320" and "rituximab treatment") and variables (such as age range and treatment time period). Data in Excel format within a specified range can be exported (the exporter, export time, and purpose are indicated when exporting).

[0098] Data encryption and operation logs: The AES-256 algorithm is used to encrypt and store the sensitive information of the children (such as date of birth and place of residence). The operation logs of all accounts are recorded in real time (including login time, data entry / modification / retrieval / export operations, IP address). Abnormal operations (such as consecutive failed logins, batch retrieval) trigger alarms.

[0099] In this embodiment, the database of children with positive antinuclear antibody immune thrombocytopenic purpura (ITP) can be constructed based on the encrypted secure dataset using the following method:

[0100] Using the "unique identifier of the patient" as the core index, a structured database is constructed according to the following five core modules:

[0101] General Information Module: Stores encrypted information such as the child's gender, age, ethnicity, date of birth (after anonymization), family history, and affiliated center;

[0102] Baseline clinical module: Stores clinical manifestations (bleeding grade, non-bleeding symptoms) and laboratory indicators (complete blood count, immune indicators, bone marrow results) at the time of initial diagnosis by coding;

[0103] Follow-up tracking module: Stores the date of each follow-up visit, changes in clinical manifestations, and re-examination indicators (platelet count, ANA titer, complement level, lymphocyte subsets) in chronological order;

[0104] Treatment management module: Stores treatment plans (code + specific medication information), treatment response (platelet level, bleeding score, B cell count at each time point), and adverse drug reactions;

[0105] Disease outcome module: Stores the disease status (remission / progression / loss to follow-up) at the last follow-up visit, and the diagnosis of progressive disease (code + disease name).

[0106] It also includes data retrieval functionality (supporting keyword / variable combination filtering), Excel export functionality (export by module / time range), and a data update interface (allowing data entry personnel to supplement follow-up data).

[0107] In this embodiment, the clinical data section includes:

[0108] 1. Clinical manifestations:

[0109] 1) Date of onset or follow-up, height and weight;

[0110] 2) Bleeding-related clinical manifestations, including skin bleeding, mucosal bleeding, organ bleeding, and central nervous system bleeding, and ICR bleeding grade assessment based on the actual bleeding situation; for example, specific bleeding manifestations are obtained from follow-up records, including skin bleeding (such as petechiae, ecchymosis, purpura), mucosal bleeding (such as nosebleeds, gingival bleeding, oral mucosal bleeding, conjunctival bleeding), organ bleeding (such as gastrointestinal bleeding, respiratory bleeding, urinary tract bleeding), and central nervous system bleeding;

[0111] 3) Non-bleeding-related clinical manifestations, including joint symptoms, rash, dry mouth and eyes, oral ulcers, hair loss, unexplained fever, lymphadenopathy, hepatosplenomegaly, etc., which may indicate the prognosis of ITP patients, and other clinical manifestations can be filled in the form.

[0112] 2. Laboratory indicators: Complete blood count; Immune indicators: including antinuclear antibody spectrum (including anti-dsDNA antibody), anti-ENA antibody spectrum, antiphospholipid antibody spectrum and lupus anticoagulant, antiplatelet antibody, immunoglobulin, complement, lymphocyte subsets, erythrocyte sedimentation rate, CRP, RF, Coombs test, etc.; peripheral blood smear, bone marrow smear and biopsy results.

[0113] 3. Treatment plan and treatment response:

[0114] 1) Treatment regimens include first-line therapy: human immunoglobulin (IVIG), glucocorticoids; second-line therapy: TPO, TPO receptor agonists (TPO-RA), rituximab, hydroxychloroquine, mycophenolate mofetil, cyclosporine, rapamycin and other immunomodulators and immunosuppressants, splenectomy; other drug treatments; all include specific dosage, date of administration, duration of administration and other information.

[0115] 2) Treatment response includes: platelet levels and bleeding score at 1, 3, 6, 12, 18, and 24 months after medication; for rituximab, it also includes B-cell count and ANA results.

[0116] 3) Adverse drug reactions: Clinical manifestations that are suspected to be related to adverse drug reactions in children treated with rituximab, TPO-RA or rapamycin during the treatment process.

[0117] In this embodiment, before constructing the database of children with antinuclear antibody-positive immune thrombocytopenic purpura (ANTBCP) based on the encrypted secure dataset, the method for establishing the database of children with ANTBCP further includes:

[0118] Obtain preset clinical evidence parameters;

[0119] Based on preset clinical empirical parameters and encrypted secure datasets, a dynamic risk stratification-treatment response prediction report is generated for each child in the encrypted secure dataset.

[0120] In this embodiment, the preset clinical empirical parameters are historical data, mainly used to determine the coefficients of each parameter. For example, based on the follow-up data of 1200 ANA-positive ITP children, the following weights were fitted: ANA titer weight α = 0.32, lymphocyte subset B cell detection value weight β = 0.25, platelet fluctuation coefficient weight γ = 0.21, bleeding grade weight δ = 0.12, treatment response rate weight ε = 0.07; the baseline probability P0 for predicting rituximab treatment response was 0.45, the influence coefficient of B cell change rate k = 1.8, and the ANA titer correction coefficient m = 0.6.

[0121] In this embodiment, the preset clinical empirical parameters include core variable data, which include ANA titer, B cell subset detection value, platelet count fluctuation coefficient, ICR bleeding grade, and response rate of previous treatment regimens at follow-up.

[0122] The process of establishing a dynamic risk stratification-treatment response prediction report for each child in the encrypted secure dataset based on preset clinical empirical parameters and encrypted secure datasets includes:

[0123] Standardize the core variable data to obtain standardized core variable data;

[0124] Dynamic risk scores and individualized treatment response prediction information are obtained based on standardized core variable data.

[0125] In this embodiment, the standardization process for the core variable data to obtain standardized core variable data includes:

[0126] The ANA titer at each child's follow-up was standardized using the following method:

[0127] The ANA titer at the time of follow-up was graded based on the follow-up data of ANA-positive ITP children. The graded range included three levels: negative, low titer, and high titer. The titer ≤ 1:80 was considered low titer, and the titer ≥ 1:160 was considered high titer.

[0128] The ANA titer during the follow-up period is standardized based on the grade amplitude to obtain the standardized ANA titer.

[0129] Specifically, the current follow-up ANA test results are extracted from the "encrypted secure dataset" and recorded as the original titer values ​​(e.g., "1:80", "1:160", "1:320", "1:640", "1:1280"). Invalid records such as "not tested" and "suspected positive" are removed. If there are multiple ANA tests within the previous 3 months of the current follow-up, the highest titer value is used (clinical evidence shows that a higher titer is more indicative of disease activity risk). If there are no test records within 3 months, the period is extended to 6 months. If there are still no records, the result is marked as "to be supplemented" and the latest test results are added after verification by the data entry person.

[0130] Based on pre-defined clinical empirical parameters (such as follow-up data from 1200 ANA-positive ITP children mentioned above), ANA titer is positively correlated with the risk of disease progression to SLE (the risk increases 3.2 times when the titer is ≥1:320 compared to 1:80). Therefore, the original titer is assigned a rank T. ANAraw Negative titers are assigned a value of 1, low titers are assigned a value of 2 (low risk), and high titers are assigned a value of 3 (low to medium risk).

[0131] Assign the original titer level the value T ANAraw Mapping to the [0,1] interval ensures that the risk differences between different titers can be quantified and compared. The lowest assigned value is 1 (negative), and the highest is 3 (1:1280). The standardized formula is as follows:

[0132]

[0133] Among them, S ANA The standardized ANA titer; T ANAraw Assign a value to the original titer level.

[0134] In this embodiment, the standardization of the changes in the B-cell subset levels of lymphocytes relative to baseline for each child was performed using the following method:

[0135] An absolute numerical correction coefficient is generated based on the B-cell subset detection value of the child's lymphocytes;

[0136] The changes in the B-cell subset of lymphocytes relative to baseline were standardized using the following formula:

[0137]

[0138] Among them, S ΔB This represents the change in the level of the B lymphocyte subset relative to baseline; ΔB adjust This represents the rate of change of the B-cell subset of lymphocytes relative to baseline after correction.

[0139] In this embodiment, the platelet count fluctuation coefficient is obtained by the following formula:

[0140] Extract all platelet count values ​​from the child's previous month;

[0141] The mean of all platelet counts was calculated using an exponentially weighted moving average.

[0142] The time-weighted fluctuation coefficient was obtained by using the mean of all platelet counts and all platelet counts of the child in the previous month, through the time-weighted standard deviation of fluctuation.

[0143] For example, extract all platelet count values ​​(PLT1, PLT2, ..., PLTn, unit: ×10) within one month prior to the current follow-up. 9 / L), requiring n≥3 tests (if n<3, extend to 24 months to ensure sufficient sample size to reflect fluctuation trends); outliers are removed using the modified Z-score method: first calculate the median and median absolute deviation (MAD) of all PLTs, then calculate Zi = (0.6745 × (PLTi - median)) / MAD for each PLT. If Zi > 3.5, it is considered an outlier (e.g., PLT = 5 × 10). 9 / L (which may be a temporary value during the acute bleeding phase), if n<2 after removal, the mean PLT fluctuation of children with the same treatment regimen (such as all using TPO-RA) in the database is used instead.

[0144] In clinical practice, recent fluctuations in platelet counts have a greater impact on current risk (fluctuations in the past 5 days reflect the current condition better than fluctuations in 1 month ago). Therefore, an exponentially weighted moving average (EWMA) is used to calculate the mean, giving more weight to recent data.

[0145] Let t be the time interval between the i-th test and the current follow-up. i (Unit: days), Weight:

[0146] ω i =e^(-0.1×t) i (0.1 is the attenuation coefficient for clinical fit, t) i The smaller ω i The larger, such as t i When ω = 1 i ≈0.905, t i =6 o'clock ω i ≈0.549).

[0147] In this embodiment, the time-weighted average formula is as follows:

[0148]

[0149] Among them, PLT ewm PLT is the time-weighted moving average of platelet counts; n is the number of valid platelet count tests within the month prior to the current follow-up (extended to 3 months if the number of tests is less than 3), requiring n≥3 (if n<2 after removing outliers, the mean platelet fluctuation of children with the same treatment regimen in the database should be used instead to ensure that the sample size is sufficient to reflect the fluctuation trend); PLT i The value of the i-th platelet count; ω i The time weight of the i-th platelet count value;

[0150] Using time-weighted standard deviation instead of ordinary standard deviation more accurately reflects recent volatility:

[0151] The formula for time-weighted standard deviation is:

[0152]

[0153] Where σPLT_ewm is the time-weighted standard deviation of platelet count; PLT ewm PLT is the time-weighted moving average of platelet counts. ewm PLT is the time-weighted moving average of platelet counts. i The value of the i-th platelet count; ω i The time weight of the i-th platelet count value;

[0154] The formula for the time-weighted fluctuation coefficient (reflecting the stability of platelet counts; the larger the coefficient, the less stable the platelet count) is:

[0155]

[0156] Among them, CV PLT_ewm σPLT_ewm is the time-weighted fluctuation coefficient of platelet count; σPLT_ewm is the time-weighted standard deviation of platelet count; PLT ewm Time-weighted moving average of platelet count;

[0157] In this embodiment, the smaller the fluctuation coefficient, the more stable the condition and the lower the risk; therefore, the CV is... PLT_ewm Mapped to the [0,1] interval. Based on clinical data, CV PLT_ewm When the disease rate is ≤20%, the patient's condition is stable, and the CV (cardiogram) is low. PLT_ewm When the percentage is ≥70%, the condition is extremely unstable. The standardized formula is:

[0158]

[0159] Among them, S CV-PLT The standardized value representing the platelet variability coefficient; CVPLT_ewm The time-weighted fluctuation coefficient for platelet count.

[0160] In this embodiment, the standardization of the ICR bleeding grade for each child is performed using the following method:

[0161] Different weights were assigned to different ICR bleeding grades;

[0162] In this embodiment, the current bleeding symptoms are extracted from the follow-up records, and the original grade is determined according to the International Consensus (ICR) bleeding grading standard: Grade I (skin bleeding only, such as petechiae, ecchymosis), Grade II (mucosal bleeding, such as nosebleeds, gingival bleeding), Grade III (organ bleeding, such as gastrointestinal bleeding, hematuria), Grade IV (central nervous system bleeding); if multiple bleeding symptoms exist at the same time (such as skin ecchymosis + nosebleed), the highest grade is taken (Grade II in this case); if there are no bleeding symptoms, it is determined as Grade I (no bleeding risk).

[0163] The clinical risks differ significantly among different bleeding sites (central nervous system bleeding has a high mortality rate, while skin bleeding has a low risk); therefore, site weighting is added to the original grading.

[0164] Grade I (Skin Bleeding): Base value 1.0, location weight 1.0 (low risk), weighted grade = 1.0 × 1.0 = 1.0;

[0165] Grade II (mucosal bleeding): Base score 2.0, location weight 1.5 (medium risk), weighted grade = 2.0 × 1.5 = 3.0;

[0166] Grade III (Organ Bleeding): Base score 3.0, site weight 3.0 (high risk), weighted grade = 3.0 × 3.0 = 9.0;

[0167] Grade IV (Central Hemorrhage): Base score 4.0, site weight 5.0 (extremely high risk), weighted grade = 4.0 × 5.0 = 20.0;

[0168] The weighted classification is denoted as G. Bleed_weight .

[0169] In this embodiment, the higher the bleeding grade, the higher the risk, and the lower the standardized value should be. Therefore, G... Bleed_weight Mapping to the [0,1] interval, the formula is as follows:

[0170]

[0171] Among them, S Bleed This represents the standardized ICR bleeding grade; 20.0 is the highest weighted grade (Grade IV), and 1.0 is the lowest weighted grade (Grade I); G Bleed_weight The weighted classification is used.

[0172] For example, Grade II bleeding (nosebleed) is calculated as follows:

[0173] G Bleed_weight =3.0; S Bleed = (20-3) / 19≈0.89, corresponding to a low risk of bleeding.

[0174] In this embodiment, the response rate of the previous treatment plan is obtained in the following way:

[0175] Retrieve follow-up records of all previous treatment regimens (such as IVIG, glucocorticoids, rituximab, TPO-RA) for the child, and screen for platelet counts at two key time points: 1 month and 3 months post-treatment (short-term / intermediate efficacy assessment points); assess efficacy according to ITP treatment guidelines: remission (PLT ≥ 100 × 10⁻⁶). 9 / L and no bleeding symptoms), partial relief (50×10 9 / L≤PLT<100×10 9 / L and bleeding improved), ineffective (PLT < 50 × 10 9 / L or no improvement in bleeding), with relief and partial relief considered as effective number of times;

[0176] Obtain sustained relief time;

[0177] In this embodiment, short-term remission (e.g., 1-month remission) and long-term remission (e.g., 3-month remission) have different clinical significance, with long-term remission indicating better treatment efficacy. Therefore, a duration-of-response weighting (W) is incorporated. duration ):

[0178] Duration of remission is defined as "the time from the first attainment of remission after treatment to the first occurrence of non-remission". If the duration of remission is ≥6 months, W duration =1.5 (long-term effective, highest weight); 3-6 months, W duration =1.2 (effective in the medium term); 1-3 months, W duration =1.0 (short-term effective); <1 month, W duration =0.5 (temporarily effective, lowest weight);

[0179] A weighted effective count is generated based on the number of effective responses and the duration of remission; in this embodiment, the weighted effective count = Σ(effective response of each treatment regimen × corresponding W) duration );

[0180] The weighted response rate is generated based on the weighted number of effective responses and the total number of quality solutions, using the following formula:

[0181] The denominator "total number of treatment plans × 1.5" is due to the maximum Wduration =1.5, ensuring R Tx_weight ≤1.

[0182] Because of R Tx_weight In the interval [0,1], the standardized value is directly equal to the weighted response rate, as shown in the formula:

[0183] S RTx =min(1,R) Tx_weight ); where S RTx R is the standard value of the weighted response rate. Tx_weight This is the weighted response rate.

[0184] In this embodiment, the dynamic risk score is obtained using the following formula:

[0185] RS Dynamic =(α×S ANA +β×S ΔB +γ×S CV-PLT +δ×S Bleed +∈×S RTx )×F comorbidity ×F family ;

[0186] Among them, RS Dynamic For dynamic risk scoring; S ANA For standardized ANA titers; S ΔB This represents the change in the level of the B lymphocyte subset relative to baseline; S CV-PLT S is the standardized platelet count fluctuation coefficient; Bleed Standardized ICR bleeding grading; S RTx α is the standardized weighted response rate; α is the weighting coefficient, referring to the weight of the standardized ANA titer value (S_ANA), reflecting the degree of influence of ANA titer on risk, with a value of 0.32. β is the weighting coefficient, referring to the weight of the standardized value of the change in the B lymphocyte subset level relative to baseline, with a value of 0.28.

[0187] γ is the weighting coefficient, which refers to the weight of the standardized value of the platelet volatility coefficient, reflecting the degree of influence of platelet stability on risk, and is set to 0.21.

[0188] δ is the weighting coefficient, which refers to the weight of the standardized value of ICR bleeding grade, reflecting the degree of impact of bleeding risk on overall risk, and has a value of 0.12.

[0189] ∈ is the weighting coefficient, which refers to the weight of the standardized value of the treatment response rate, reflecting the degree of influence of past treatment effects on current risk, and has a value of 0.07.

[0190] F comorbidityThe positive factor refers to the comorbidity symptom correction factor, which reflects the impact of whether the child has other autoimmune-related symptoms (such as rash, joint pain, dry mouth and eyes) on the risk. The value is 0.85 (≥1 comorbidity symptom, risk increased, final score needs to be reduced) or 1.0 (no comorbidity symptom, no correction), based on clinical evidence (children with comorbidity symptoms have a 23% increased risk of progression).

[0191] F family The correction factor, specifically the family history correction factor, reflects the impact of a child's family history of autoimmune diseases (such as parents / siblings having SLE or rheumatoid arthritis) on risk. It is set at 0.90 (with family history, risk increases, lowering the final score) or 1.0 (no family history, no correction), based on family clustering studies (children with a family history have an 18% increased risk of progression).

[0192] In this embodiment, the individualized treatment response prediction information is calculated using the following formula:

[0193]

[0194] Among them, P RTX To predict the 12-month remission probability of rituximab treatment based on changes in immune markers; P0 is the baseline probability parameter, referring to the baseline remission probability of ANA-positive ITP children using rituximab; k is the influence coefficient of B cell change rate; ΔB adjust F represents the corrected rate of change of B cells relative to baseline. age is the age correction factor; m is the negative correction coefficient for ANA titer; TANA is the standardized ANA titer; F ΔB F is a correction factor for the relative baseline change in B cell subset levels of lymphocytes; Tx-history This is a factor for correcting previous treatment history.

[0195] Other technical solutions may also include setting personalized treatment response prediction information for other types of drugs, such as TPO receptor agonists and glucocorticoids, which are not within the scope of protection of this application.

[0196] In existing technologies, traditional clinical assessments for ANA-positive ITP patients often rely on qualitative descriptions such as "high / low ANA titer" and "low / high platelet count," which are easily influenced by differences in physician experience. This application incorporates five core indicators: ANA titer, changes in B-cell levels relative to baseline, platelet fluctuations, bleeding grade, and treatment response. These indicators cover the entire spectrum of "immune activation, disease stability, and treatment efficacy," avoiding misjudgments based on a single indicator (e.g., looking only at normal platelet counts while ignoring elevated ANA titers and decreased complement levels may lead to a missed assessment of high progression risk).

[0197] Based on the weighting coefficients fitted from the follow-up data of 1200 children (α=0.32, β=0.28, etc.), each standardized indicator is transformed into a dynamic risk score in the range of [0,1]. The risk level is divided by clear thresholds of "low (≥0.8), medium (0.5-0.8), and high (<0.5)", allowing doctors to intuitively see the predicted score and thus judge the risk.

[0198] And output the specific probability of treatment relief (e.g., P). RTX =0.68), and gives clear clinical decision thresholds (≥0.65 is recommended to use, <0.4 is recommended to switch to another drug). Doctors do not need to rely on "experience inference" and can directly formulate a plan based on probability (e.g., if the predicted probability of rituximab for a certain child is 0.72, it is preferred; if it is 0.35, then TPO-RA is used instead).

[0199] The output "Dynamic Risk Stratification Report" and "Treatment Response Prediction Report" can be directly embedded into the doctor's workstation to provide a basis for follow-up visits (such as triggering a "follow-up visit within 2 weeks" warning for high-risk children), reducing trial and error costs;

[0200] The risk score and treatment prediction probability add "risk label" and "efficacy label" to the database. Subsequent research can directly conduct stratified analysis based on these labels (such as comparing the difference in SLE progression rate between children with "high risk RS_Dynamic < 0.5" and "low risk RS_Dynamic ≥ 0.8", or analyzing the common clinical characteristics of children with "P_RTX ≥ 0.65"), without having to reorganize the original data, which greatly shortens the research preparation time.

[0201] Standardized variables (e.g., S) ANA ) and residual analysis of prediction formulas (such as actual remission rate and P) RTX The difference between the two values ​​can be used to discover new biomarkers (such as the discovery of "a certain type of lymphocyte subset and P"). RTX The indicator “significantly correlated with residuals” suggests that it can serve as a supplementary predictor of rituximab efficacy, providing direction for further optimization of precision medicine.

[0202] This application also provides a device for establishing a database of children with antinuclear antibody-positive immune thrombocytopenic purpura (ANTP). The device includes a basic information acquisition module, a structured full-dimensional dataset acquisition module, a standardized dataset acquisition module for patients, an encryption module, and a database establishment module.

[0203] The basic information acquisition module is used to acquire basic screening datasets of children and detailed clinical data of children provided by multiple centers;

[0204] The structured full-dimensional dataset acquisition module is used to form a structured full-dimensional dataset of patients based on the basic screening dataset of patients and the detailed clinical data of patients provided by multiple centers.

[0205] The standardized dataset acquisition module for sick children is used to standardize and uniformly encode the data in the structured full-dimensional dataset of sick children, thereby forming a standardized dataset for sick children.

[0206] The encryption module is used to set and assign permissions for each piece of data in the standardized dataset of sick children, thereby forming an encrypted secure dataset.

[0207] The database creation module is used to build a database of children with antinuclear antibody-positive immune thrombocytopenic purpura based on an encrypted secure dataset.

[0208] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and not to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for establishing a database of children with positive antinuclear antibody-mediated immune thrombocytopenic purpura, characterized in that, The method for establishing the database of children with positive antinuclear antibody immune thrombocytopenic purpura includes: Obtain basic screening datasets of children with the disease and detailed clinical data of children with the disease provided by multiple centers; A structured, multi-dimensional dataset of patients was created by inputting basic screening data and detailed clinical data from multiple centers. Data standardization and unified coding were performed on the structured full-dimensional dataset of sick children to form a standardized dataset of sick children; Permissions are set and assigned to each data point in the standardized dataset of sick children, thereby forming an encrypted and secure dataset; A database of children with positive antinuclear antibody immune thrombocytopenic purpura was constructed based on a secure, encrypted dataset.

2. The method for establishing a database of children with positive antinuclear antibody immune thrombocytopenic purpura as described in claim 1, characterized in that, Before constructing the database of children with antinuclear antibody-positive immune thrombocytopenic purpura (ANTAPP) based on the encrypted secure dataset, the method for establishing the database of children with ANTAPP further includes: Obtain preset clinical evidence parameters; Based on preset clinical empirical parameters and encrypted secure datasets, a dynamic risk stratification-treatment response prediction report is generated for each child in the encrypted secure dataset.

3. The method for establishing a database of children with positive antinuclear antibody immune thrombocytopenic purpura as described in claim 2, characterized in that, The preset clinical empirical parameters include core variable data, which include ANA titer at follow-up, changes in B-cell lymphocyte subsets relative to baseline, platelet count fluctuation coefficient, ICR bleeding grade, and response rate of previous treatment regimens. The process of establishing a dynamic risk stratification-treatment response prediction report for each child in the encrypted secure dataset based on preset clinical empirical parameters and encrypted secure datasets includes: Standardize the core variable data to obtain standardized core variable data; Dynamic risk scores and individualized treatment response prediction information are obtained based on standardized core variable data.

4. The method for establishing a database of children with positive antinuclear antibody immune thrombocytopenic purpura as described in claim 3, characterized in that, The standardization process for core variable data, thereby obtaining standardized core variable data, includes: The ANA titer at each child's follow-up was standardized using the following method: The ANA titer at the time of follow-up was graded based on the follow-up data of ANA-positive ITP children. The graded range included three levels: negative, low titer, and high titer. The titer ≤ 1:80 was considered low titer, and the titer ≥ 1:160 was considered high titer. The ANA titer during the follow-up period is standardized based on the grade amplitude to obtain the standardized ANA titer.

5. The method for establishing a database of children with positive antinuclear antibody immune thrombocytopenic purpura as described in claim 4, characterized in that, The standardization of the changes in B-cell levels relative to baseline for each child's lymphocyte subset was performed using the following method: An absolute numerical correction coefficient is generated based on the B-cell subset detection value of the child's lymphocytes; The changes in the B-cell subset of lymphocytes relative to baseline were standardized using the following formula: Among them, S ΔB This represents the change in the level of the B lymphocyte subset relative to baseline; ΔB adjust This represents the rate of change of the B-cell subset of lymphocytes relative to baseline after correction.

6. The method for establishing a database of children with positive antinuclear antibody immune thrombocytopenic purpura as described in claim 5, characterized in that, The platelet count fluctuation coefficient is obtained using the following formula: Extract all platelet count values ​​from the child's previous month; The mean of all platelet counts was calculated using an exponentially weighted moving average. The time-weighted fluctuation coefficient was obtained by using the mean of all platelet counts and all platelet counts of the child in the previous month, through the time-weighted standard deviation of fluctuation. The platelet count fluctuation coefficient is standardized using the following formula: Among them, S CV-PLT The standardized value representing the platelet variability coefficient; CV PLT_ewm The time-weighted fluctuation coefficient for platelet count.

7. The method for establishing a database of children with positive antinuclear antibody immune thrombocytopenic purpura as described in claim 6, characterized in that, The standardized processing of the ICR bleeding grade for each child was performed using the following method: Specific bleeding manifestations were obtained from follow-up records, including skin bleeding, mucosal bleeding, organ bleeding, and central nervous system bleeding. Set base values ​​and site weights for different bleeding manifestations: Skin bleeding: Base value 1.0, location weight 1.0; Mucosal bleeding: Base value 2.0, with a weight of 1.5 for the location of nasal bleeding and gingival bleeding, and a weight of 1.8 for the location of oral mucosal bleeding and conjunctival bleeding; Organ bleeding: Base value 3.0, with a weight of 3.0 for gastrointestinal bleeding and urinary tract bleeding and a weight of 3.5 for respiratory bleeding. Central nervous system hemorrhage: base value 4.0, location weight 5.0; When a child has multiple bleeding manifestations, the highest weighted classification corresponding to the bleeding manifestation is used for calculation; if there are no bleeding symptoms, it is determined to be skin bleeding. Weighted classification is calculated based on the base values ​​and location weights: G Bleed_weight = Base value × Part weight; The following formula was used to standardize the ICR bleeding grading: Among them, S Bleed This represents the standardized ICR bleeding grade; 20.0 is the highest weighted grade (Grade IV), and 1.0 is the lowest weighted grade (Grade I); G Bleed_weight The weighted classification is used.

8. The method for establishing a database of children with positive antinuclear antibody immune thrombocytopenic purpura as described in claim 7, characterized in that, The response rate of the previous treatment regimen was obtained in the following manner: We extracted all previous treatment regimens and their post-treatment follow-up records, and screened platelet counts at multiple key time points. We determined the efficacy of each key time point as the response rate of previous treatment regimens according to the ITP treatment guidelines. The response rate of previous treatment regimens included remission, partial remission, and no response. Remission and partial remission were considered as effective times. Obtain sustained relief time; A weighted number of effective counts is generated based on the number of effective counts and the duration of relief. The weighted response rate is generated based on the weighted number of valid responses and the total number of quality solutions.

9. The method for establishing a database of children with positive antinuclear antibody immune thrombocytopenic purpura as described in claim 8, characterized in that, The dynamic risk score is obtained using the following formula: RS Dynamic =(α×S ANA +β×S ΔB +γ×S CV-PLT +δ×S Bleed +∈×S RTx )×F comorbidity ×F family ; Among them, RS Dynamic For dynamic risk scoring; S ANA For standardized ANA titers; S ΔB This represents the change in the level of the B lymphocyte subset relative to baseline; S CV-PLT S is the standardized platelet count fluctuation coefficient; Bleed Standardized ICR bleeding grading; S RTx The standardized weighted response rate; α is the weighting factor, representing the weight of the standardized ANA titer value, reflecting the degree of influence of ANA titer on risk, with a value of 0.32; β is the weighting factor, representing the weight of the standardized value of the change in B-cell lymphocyte subset levels relative to baseline; γ is the weighting factor, representing the weight of the standardized value of platelet variability coefficient; δ is the weighting factor, representing the weight of the standardized value of ICR bleeding grade; ε is the weighting factor, representing the weight of the standardized value of treatment response rate; F comorbidity The positive factor, referring to the comorbidity symptom correction factor, reflects the impact of whether the child has other autoimmune-related symptoms on the risk; F family The correction factor refers to the family history correction factor; The individualized treatment response prediction information is calculated using the following formula: Among them, P RTX To predict the 12-month remission probability of rituximab treatment based on changes in immune markers; P0 is the baseline probability parameter, referring to the baseline remission probability of ANA-positive ITP children using rituximab; k is the influence coefficient of B cell change rate; ΔB adjust F represents the corrected rate of change of B cells relative to baseline; age is the age correction factor; m is the negative correction coefficient for ANA titer; TANA is the standardized ANA titer; F ΔB F is a correction factor for the relative baseline change in B cell subset levels of lymphocytes; Tx-history This is a factor for correcting previous treatment history.

10. A device for establishing a database of children with positive antinuclear antibody immune thrombocytopenic purpura, characterized in that, The device for establishing a database of children with positive antinuclear antibody immune thrombocytopenic purpura includes: The basic information acquisition module is used to acquire the basic screening dataset of the children and the detailed clinical data of the children provided by multiple centers. The structured full-dimensional dataset acquisition module is used to form a structured full-dimensional dataset of patients based on the basic screening dataset of patients and the detailed clinical data of patients provided by multiple centers. The standardized dataset acquisition module for patients is used to standardize and uniformly encode the data in the structured full-dimensional dataset of patients, thereby forming a standardized dataset for patients. An encryption module is used to set and assign permissions to each piece of data in the standardized dataset of sick children, thereby forming an encrypted secure dataset. A database creation module is used to construct a database of children with antinuclear antibody-positive immune thrombocytopenic purpura based on an encrypted secure dataset.

Images