Repeated pregnancy loss risk early warning method and system based on big data

Abstract

The invention belongs to the technical field of pregnancy risk early warning, and discloses a repeated pregnancy loss risk early warning method and system based on big data, and the method comprises the steps: collecting basic feature data and physiological index data of a research object, marking a normal pregnancy or repeated pregnancy loss mark according to the pregnancy condition, and forming a big database; classifying the data according to basic feature data, and analyzing the association between the physiological indexes and the repeated pregnancy loss risk to form a reference database; obtaining basic feature data of a to-be-evaluated object, and matching target data under the corresponding classification from the reference database; after the physiological indexes of the to-be-evaluated object are obtained, the risk degree is evaluated through comparison with the target data, and finally a risk early warning result is generated. Through an individualized layered evaluation strategy and vectorization analysis of multi-dimensional physiological indexes, the problem of inaccurate early warning caused by adoption of a unified standard in the prior art is solved, and the accuracy of risk early warning is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of pregnancy risk warning technology, and more specifically, to a method and system for early warning of recurrent pregnancy loss risk based on big data. Background Art

[0002] Recurrent pregnancy loss (RPL), defined as two or more consecutive miscarriages, is a complex and challenging clinical problem in obstetrics and gynecology. With the postponement of childbearing age and changes in environmental factors, the incidence of RPL is on the rise, placing a significant psychological and financial burden on patients' families.

[0003] Existing RPL risk assessment methods often use a single threshold judgment or a simple scoring system, resulting in limitations in the degree of quantification. These methods usually compare various physiological indicators with fixed reference ranges, lack in-depth analysis of the relationship between indicators, and it is difficult to objectively evaluate the interaction between various indicators and their combined effects, resulting in a lack of accuracy and reliability in risk assessment results. In addition, the current assessment system lacks an effective individualized risk stratification mechanism. Women with different characteristics have significant individual differences in their physiological baseline status and potential risk factors, but current assessment methods often use relatively unified reference standards and fail to conduct refined classification assessments based on individual characteristics, which further reduces the accuracy of risk warnings. Summary of the Invention

[0004] In order to overcome the problem of low accuracy of risk warning in the existing technology, the present invention proposes a recurrent pregnancy loss risk warning method and system based on big data to solve the above problem.

[0005] The present invention provides the following technical solutions:

[0006] A big data-based early warning method for recurrent pregnancy loss risk, comprising:

[0007] Collect a large number of basic characteristics and physiological index data of research subjects, and mark the physiological index data with normal pregnancy mark or recurrent pregnancy loss mark according to their pregnancy status to establish a large database;

[0008] The physiological indicator data in the large database are classified according to the basic characteristic information. By analyzing the classified physiological indicator data, the correlation data between the physiological indicators in each category and the risk of recurrent pregnancy loss are determined to obtain a reference database;

[0009] Obtain basic characteristic data of the subject to be evaluated. Based on the basic characteristic data, match the physiological indicators and recurrent pregnancy loss risk association data under the corresponding classification from the reference database and record them as target data;

[0010] Obtaining physiological indicators of the subject to be evaluated, and assessing the risk level based on the physiological indicators of the subject to be evaluated and target data;

[0011] Based on the risk level, a recurrent pregnancy loss risk warning result is generated.

[0012] Preferably, the basic characteristic data include age, BMI index and number of pregnancies; the physiological index data include hormone level data, immune index data and coagulation function index data.

[0013] Preferably, the classification of the physiological indicator data in the large database according to basic characteristic information includes:

[0014] The data is divided into several age interval groups according to the preset age threshold; the data is divided into several BMI interval groups according to the preset BMI index threshold; the data is divided into several pregnancy number groups according to the preset pregnancy number threshold;

[0015] The complete classification was formed by the intersection of age interval groups, BMI interval groups and number of pregnancies groups.

[0016] Preferably, analyzing the classified physiological indicator data to determine the association data between the physiological indicators in each category and the risk of recurrent pregnancy loss includes:

[0017] For each physiological indicator data under each category, a hormone level vector is established based on the hormone level data, an immune indicator vector is established based on the immune indicator data, and a coagulation function vector is established based on the coagulation function indicator data. The hormone level vector, immune indicator vector, and coagulation function vector are sequentially combined to obtain a physiological indicator vector;

[0018] Based on the labels of the physiological indicator data, the hormone level vector, immune indicator vector, coagulation function vector, and physiological indicator vector are marked with corresponding normal pregnancy markers or recurrent pregnancy loss markers;

[0019] According to the hormone level vector, immune index vector, coagulation function vector and physiological index vector of normal pregnancy markers, the corresponding hormone standard vector, immune standard vector, coagulation standard vector and physiological standard vector are obtained;

[0020] According to the hormone level vector, immune index vector, coagulation function vector and physiological index vector of recurrent pregnancy loss markers, the corresponding hormone risk vector, immune risk vector, coagulation risk vector and physiological risk vector are obtained;

[0021] The hormone standard vector, immune standard vector, coagulation standard vector, physiological standard vector, hormone risk vector, immune risk vector, coagulation risk vector and physiological risk vector were combined into the data on the association between physiological indicators and the risk of recurrent pregnancy loss.

[0022] Preferably, establishing the hormone level vector based on the hormone level data includes: normalizing the data of estrogen level, progesterone level, human chorionic gonadotropin level, luteinizing hormone level and follicle-stimulating hormone level, and combining the normalized data in order into the hormone level vector;

[0023] The step of establishing an immune index vector based on the immune index data includes normalizing the data of antiphospholipid antibody content, anticardiolipin antibody content, NK cell activity value, Th1 to Th2 ratio, and cytokine level value, and sequentially combining the normalized data into an immune index vector;

[0024] The method of establishing a coagulation function vector based on the coagulation function index data includes normalizing the data of prothrombin time value, activated partial thromboplastin time value, D-dimer content and fibrinogen content, and sequentially combining the normalized data into a coagulation function vector.

[0025] Preferably, the methods for obtaining the hormone standard vector, immune standard vector, coagulation standard vector and physiological standard vector include:

[0026] Calculate the center vectors of the hormone level vector, immune index vector, coagulation function vector and physiological index vector with normal pregnancy markers respectively as the corresponding hormone standard vector, immune standard vector, coagulation standard vector and physiological standard vector;

[0027] Methods for obtaining the hormone risk vector, immune risk vector, coagulation risk vector, and physiological risk vector include:

[0028] Calculate the center vectors of hormone level vectors, immune index vectors, coagulation function vectors, and physiological index vectors with recurrent pregnancy loss markers as the corresponding hormone risk vectors, immune risk vectors, coagulation risk vectors, and physiological risk vectors respectively;

[0029] The center vector is the vector in the vector set that has the smallest sum of squared distances to all other vectors.

[0030] Preferably, the risk assessment based on the physiological indicators and target data of the subject to be assessed includes:

[0031] According to the physiological indicators of the subject to be evaluated, a hormone level vector, an immune indicator vector, a coagulation function vector and a physiological indicator vector are established for the subject to be evaluated;

[0032] Calculate the distance between each vector of the object to be evaluated and the corresponding standard vector and risk vector in the target data respectively;

[0033] The hormone level risk index, immune indicator risk index, coagulation function risk index and comprehensive physiological risk index were obtained by normalizing the ratio of the distance between each vector and the risk vector to the distance between each vector and the standard vector.

[0034] The risk level is obtained through weighted calculation based on various risk indices.

[0035] Preferably, the risk level is obtained by weighted calculation based on various risk indices, including:

[0036] Calculate the distance value between the hormone standard vector and the hormone risk vector, the distance value between the immune standard vector and the immune risk vector, the distance value between the coagulation standard vector and the coagulation risk vector, and the distance value between the physiological standard vector and the physiological risk vector;

[0037] Based on the above distance values, weight coefficients are assigned to each risk index according to the rule that the larger the distance value, the larger the weight coefficient, and the sum of the weight coefficients is 1;

[0038] The risk level is obtained by weighting and summing up each risk index according to the assigned weight coefficient.

[0039] Preferably, generating a recurrent pregnancy loss risk warning result according to the risk level includes:

[0040] Set warning thresholds α and β, where α<β;

[0041] When the overall risk of recurrent pregnancy loss is less than α, a low-risk warning result is generated;

[0042] When the overall risk of recurrent pregnancy loss is greater than or equal to α and less than β, a moderate risk warning result is generated;

[0043] When the overall risk of recurrent pregnancy loss is greater than or equal to β, a high-risk warning result is generated.

[0044] The present invention also provides a recurrent pregnancy loss risk warning system based on big data, which is used to implement a recurrent pregnancy loss risk warning method based on big data, comprising:

[0045] The data collection module is used to collect a large number of basic characteristics and physiological index data of research subjects, and mark the physiological index data with normal pregnancy marks or recurrent pregnancy loss marks according to their pregnancy status to establish a large database;

[0046] The data analysis module is used to classify the physiological indicator data in the large database according to the basic characteristic information, and by analyzing the classified physiological indicator data, determine the correlation data between the physiological indicators in each category and the risk of recurrent pregnancy loss, and obtain a reference database;

[0047] The feature matching module is used to obtain the basic feature information of the subject to be evaluated, and based on the basic feature information, match the physiological indicators under the corresponding classification with the recurrent pregnancy loss risk association data from the reference database and record them as target data;

[0048] A risk assessment module is used to obtain physiological indicators of the subject to be assessed and assess the risk level based on the physiological indicators of the subject to be assessed and the target data;

[0049] The early warning output module is used to generate recurrent pregnancy loss risk early warning results based on the risk level.

[0050] The present invention provides a method and system for early warning of recurrent pregnancy loss risk based on big data, which has the following beneficial effects:

[0051] By establishing a large database containing basic characteristic data and multidimensional physiological indicators, and classifying them according to pregnancy outcomes, this method achieves comprehensive collection and systematic management of physiological indicators related to recurrent pregnancy loss. This large-sample data-based analysis method overcomes the limitations of traditional single threshold judgments and provides a more reliable data foundation for risk assessment.

[0052] By categorizing data based on age, BMI, and number of pregnancies, a reference database tailored to the characteristics of different populations was established. This personalized, stratified assessment strategy addresses the inaccurate early warning issues associated with the use of uniform standards in existing technologies, making risk assessments more tailored to individual circumstances.

[0053] By constructing hormone level vectors, immune indicator vectors, and coagulation function vectors and combining them into a comprehensive physiological indicator vector, this method enables integrated analysis of multidimensional physiological indicators. This vectorization method considers the independent influence of each indicator and the interactions between them, overcoming the shortcomings of existing technologies that lack consideration of indicator correlations. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 Schematic diagram of a process of a recurrent pregnancy loss risk early warning method based on big data of the present invention;

[0055] Figure 2 This is a module schematic diagram of a recurrent pregnancy loss risk early warning system based on big data of the present invention. DETAILED DESCRIPTION

[0056] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0057] Example 1

[0058] See also Figure 1 In this embodiment, a method for early warning of recurrent pregnancy loss risk based on big data includes:

[0059] S1. Collect basic characteristics and physiological index data of a large number of research subjects, and mark the physiological index data with normal pregnancy label or recurrent pregnancy loss label according to their pregnancy status to establish a large database;

[0060] Basic characteristic data include age, BMI index and number of pregnancies; physiological indicator data include hormone level data, immune indicator data and coagulation function indicator data.

[0061] In this embodiment, data can be obtained by extracting electronic medical record data from the hospital obstetrics and gynecology outpatient system. For each research subject, their age (accurate to years), BMI index (calculated by weight and height, accurate to one decimal place) and number of pregnancies are recorded as basic characteristic data. Then, the physiological index data of the research subject are collected, including hormone level data, immune index data and coagulation function index data. Various data can be collected in a conventional manner. For example, hormone level data can be collected by collecting venous blood samples on the 3rd to 5th day of the menstrual cycle, and the level values ​​of estrogen, progesterone, human chorionic gonadotropin, luteinizing hormone and follicle-stimulating hormone are measured using chemiluminescence immunoassay.

[0062] Next, physiological indicator data were tagged based on the subject's pregnancy status. The patient's pregnancy history and outcome were determined by reviewing the patient's medical records and follow-up records. For subjects whose previous pregnancies were successful and who had not experienced spontaneous abortion or other adverse pregnancy outcomes, their physiological indicator data were tagged as normal pregnancy. For subjects with a history of two or more consecutive spontaneous abortions (≤20 weeks of gestation), their physiological indicator data were tagged as recurrent pregnancy loss. Finally, all collected basic characteristic data, physiological indicator data, and their corresponding tags were integrated into a unified database to form a large database.

[0063] S2. Classifying the physiological indicator data in the large database according to basic characteristic information, analyzing the classified physiological indicator data, determining the correlation data between the physiological indicators in each category and the risk of recurrent pregnancy loss, and obtaining a reference database;

[0064] The physiological indicator data in the large database are classified according to basic characteristic information, including:

[0065] The data is divided into several age interval groups according to the preset age threshold; the data is divided into several BMI interval groups according to the preset BMI index threshold; the data is divided into several pregnancy number groups according to the preset pregnancy number threshold;

[0066] The complete classification was formed by the intersection of age interval groups, BMI interval groups and number of pregnancies groups.

[0067] By analyzing the classified physiological index data, the data related to the physiological indexes in each category and the risk of recurrent pregnancy loss were determined to include:

[0068] For each physiological indicator data under each category, a hormone level vector is established based on the hormone level data, an immune indicator vector is established based on the immune indicator data, and a coagulation function vector is established based on the coagulation function indicator data. The hormone level vector, immune indicator vector, and coagulation function vector are sequentially combined to obtain a physiological indicator vector;

[0069] Based on the labels of the physiological indicator data, the hormone level vector, immune indicator vector, coagulation function vector, and physiological indicator vector are marked with corresponding normal pregnancy markers or recurrent pregnancy loss markers;

[0070] According to the hormone level vector, immune index vector, coagulation function vector and physiological index vector of normal pregnancy markers, the corresponding hormone standard vector, immune standard vector, coagulation standard vector and physiological standard vector are obtained;

[0071] According to the hormone level vector, immune index vector, coagulation function vector and physiological index vector of recurrent pregnancy loss markers, the corresponding hormone risk vector, immune risk vector, coagulation risk vector and physiological risk vector are obtained;

[0072] The hormone standard vector, immune standard vector, coagulation standard vector, physiological standard vector, hormone risk vector, immune risk vector, coagulation risk vector and physiological risk vector were combined into the data on the association between physiological indicators and the risk of recurrent pregnancy loss.

[0073] Establishing a hormone level vector based on hormone level data includes: normalizing the data of estrogen level, progesterone level, human chorionic gonadotropin level, luteinizing hormone level and follicle-stimulating hormone level, and sequentially combining the normalized data into a hormone level vector;

[0074] Establishing an immune index vector based on immune index data includes: normalizing the data of antiphospholipid antibody content, anticardiolipin antibody content, NK cell activity value, Th1 (helper T lymphocyte 1) to Th2 (helper T lymphocyte 2) ratio and cytokine level value, and sequentially combining the normalized data into an immune index vector;

[0075] Establishing a coagulation function vector based on coagulation function index data includes: normalizing the data of prothrombin time value, activated partial thromboplastin time value, D-dimer content and fibrinogen content, and combining the normalized data into a coagulation function vector in sequence.

[0076] Methods for obtaining hormone standard vectors, immune standard vectors, coagulation standard vectors, and physiological standard vectors include:

[0077] Calculate the center vectors of the hormone level vector, immune index vector, coagulation function vector and physiological index vector with normal pregnancy markers respectively as the corresponding hormone standard vector, immune standard vector, coagulation standard vector and physiological standard vector;

[0078] Methods for obtaining hormone risk vectors, immune risk vectors, coagulation risk vectors, and physiological risk vectors include:

[0079] Calculate the center vectors of hormone level vectors, immune index vectors, coagulation function vectors, and physiological index vectors with recurrent pregnancy loss markers as the corresponding hormone risk vectors, immune risk vectors, coagulation risk vectors, and physiological risk vectors respectively;

[0080] The center vector is the vector in the vector set that has the smallest sum of squared distances to all other vectors.

[0081] In this embodiment, a preset age threshold is first set to divide the data into several age interval groups. Specifically, the age ranges can be divided into five categories: under 25, 26-30, 31-35, 36-40, and 41-45. Then, a preset BMI threshold is set to divide the data into several BMI interval groups. For example, the BMI can be divided into four intervals: less than 18.5, greater than or equal to 18.5 and less than 24, greater than or equal to 24 and less than 28, and greater than or equal to 28. Next, a preset number of pregnancies threshold is set to divide the data into several number of pregnancies groups. For example, the number of pregnancies can be divided into three groups: less than or equal to 1, 2, and greater than 2. Finally, a complete classification is formed by the intersection of the age interval groups, BMI interval groups, and number of pregnancies groups. For example, those under 25, with a BMI less than 18.5 and 1 or less pregnancy, constitute one complete classification; those under 25, with a BMI less than 18.5 and 2 pregnancies, constitute another complete classification, and so on.

[0082] Next, by analyzing the classified physiological indicator data, we determined the association between each category's physiological indicator and the risk of recurrent pregnancy loss. For each category's physiological indicator data, we first constructed a hormone level vector based on the hormone level data. Specifically, we normalized the estrogen, progesterone, human chorionic gonadotropin, luteinizing hormone, and follicle-stimulating hormone data and sequentially combined the normalized data into a hormone level vector. Normalization can be performed using the maximum-minimum normalization method, which converts the raw data to the [0,1] interval. Similarly, we obtained the immune indicator vector and the coagulation function vector. Finally, we sequentially combined the hormone level vector, immune indicator vector, and coagulation function vector to obtain the physiological indicator vector.

[0083] Then, based on the labels of the physiological indicator data, the hormone level vector, immune indicator vector, coagulation function vector, and physiological indicator vector are marked with corresponding normal pregnancy markers or recurrent pregnancy loss markers. By querying the tag information in the large database, the tag type corresponding to each vector can be determined.

[0084] Next, based on the hormone level vectors, immune index vectors, coagulation function vectors, and physiological index vectors that are markers of normal pregnancy, the corresponding hormone standard vectors, immune standard vectors, coagulation standard vectors, and physiological standard vectors are obtained. Specifically, the center vectors of the hormone level vectors, immune index vectors, coagulation function vectors, and physiological index vectors that are markers of normal pregnancy are calculated as the corresponding hormone standard vectors, immune standard vectors, coagulation standard vectors, and physiological standard vectors. The center vector is the vector in the vector set that has the smallest sum of squared distances to all other vectors.

[0085] Similarly, based on the hormone level vector, immune index vector, coagulation function vector, and physiological index vector of the recurrent pregnancy loss marker, the corresponding hormone risk vector, immune risk vector, coagulation risk vector, and physiological risk vector are obtained. The specific method is to calculate the center vector of the hormone level vector, immune index vector, coagulation function vector, and physiological index vector with the recurrent pregnancy loss marker respectively as the corresponding hormone risk vector, immune risk vector, coagulation risk vector, and physiological risk vector.

[0086] It should be noted that the vector distance can be calculated using common distance calculation methods such as Euclidean distance, Manhattan distance, and cosine similarity. In this embodiment, the Euclidean distance calculation method can be used, that is, the square root of the sum of the squares of the differences between the corresponding elements of the two vectors.

[0087] Finally, the hormonal standard vector, immune standard vector, coagulation standard vector, physiological standard vector, hormonal risk vector, immune risk vector, coagulation risk vector, and physiological risk vector were combined to form a reference database of data correlating physiological indicators with the risk of recurrent pregnancy loss. The standard vectors and risk vectors generated in this way can effectively measure the physiological differences between recurrent pregnancy loss and normal pregnancies. The standard vectors represent the central tendency of physiological indicators in the normal pregnancy population, while the risk vectors represent the central tendency of physiological indicators in the recurrent pregnancy loss population. The differences between the two vectors reflect the systematic differences in physiological indicators among populations with different pregnancy outcomes, providing a quantitative basis for identifying high-risk individuals.

[0088] S3. Obtain basic characteristic data of the subject to be evaluated. Based on the basic characteristic data, match the physiological indicators of the corresponding categories with the data associated with recurrent pregnancy loss risk from the reference database and record them as target data;

[0089] In this example, basic profile data for the subject to be assessed, including age, BMI, and number of pregnancies, is obtained. Based on this profile, the subject's classification is determined, such as under 25 years old, BMI less than 18.5, and one or fewer pregnancies. Next, a reference database is searched for data associated with the physiological indicators and recurrent pregnancy loss risk corresponding to this classification. The resulting data is recorded as target data for subsequent risk assessment.

[0090] S4. Obtaining physiological indicators of the subject to be evaluated, and assessing the risk level based on the physiological indicators of the subject to be evaluated and the target data;

[0091] The risk level is assessed based on the subject's physiological indicators and target data, including:

[0092] According to the physiological indicators of the subject to be evaluated, a hormone level vector, an immune indicator vector, a coagulation function vector and a physiological indicator vector are established for the subject to be evaluated;

[0093] Calculate the distance between each vector of the object to be evaluated and the corresponding standard vector and risk vector in the target data respectively;

[0094] The hormone level risk index, immune indicator risk index, coagulation function risk index and comprehensive physiological risk index were obtained by normalizing the ratio of the distance between each vector and the risk vector to the distance between each vector and the standard vector.

[0095] The risk level is obtained through weighted calculation based on various risk indices.

[0096] According to various risk indices, the risk levels are calculated through weighted calculations, including:

[0097] Calculate the distance value between the hormone standard vector and the hormone risk vector, the distance value between the immune standard vector and the immune risk vector, the distance value between the coagulation standard vector and the coagulation risk vector, and the distance value between the physiological standard vector and the physiological risk vector;

[0098] Based on the above distance values, weight coefficients are assigned to each risk index according to the rule that the larger the distance value, the larger the weight coefficient, and the sum of the weight coefficients is 1;

[0099] The risk level is obtained by weighting and summing up each risk index according to the assigned weight coefficient.

[0100] In this embodiment, after obtaining the subject's physiological indicators, the risk level is assessed based on the subject's physiological indicators and target data. First, based on the subject's physiological indicators, a hormone level vector, an immune indicator vector, a coagulation function vector, and a physiological indicator vector are established. The specific steps are the same as those previously described, namely, each physiological indicator data is normalized and sequentially combined into corresponding vectors.

[0101] Next, the distances between each vector of the object to be evaluated and the corresponding standard vector and risk vector in the target data are calculated. Similarly, the distances between the immune index vector, coagulation function vector, and physiological index vector and the corresponding standard vector and risk vector are calculated.

[0102] Since the closer the physiological indicator vector to the standard vector is to the subject being assessed, the closer their physiological state is to that of a normal pregnancy group; and the closer the distance to the risk vector is to the risk vector, the closer their physiological state is to that of a recurrent pregnancy loss group, the closer the distance is to the risk vector. Therefore, by calculating the ratio of these two distances and performing appropriate transformations, we can derive an index reflecting the degree of risk. In this way, we can obtain the hormone level risk index, the immune indicator risk index, the coagulation function risk index, and the comprehensive physiological risk index.

[0103] Finally, the risk level was determined through weighted calculation based on each risk index. First, the distances between the hormone standard vector and the hormone risk vector, the distance between the immune standard vector and the immune risk vector, the distance between the coagulation standard vector and the coagulation risk vector, and the distance between the physiological standard vector and the physiological risk vector were calculated. These distances reflect the degree of difference between the normal pregnancy group and the recurrent pregnancy loss group in each physiological indicator dimension.

[0104] Since a greater distance between the standard vector and the risk vector means that the physiological indicator has a stronger discriminatory ability in distinguishing between the two pregnancy states, it should play a greater role in risk assessment. Therefore, based on the above distance value, a weight coefficient is assigned to each risk index according to the rule that the larger the distance value, the larger the weight coefficient, and the sum of the weight coefficients is 1.

[0105] Through this method, the calculated risk level can not only comprehensively consider the impact of various physiological indicators on the risk of recurrent pregnancy loss, but also adaptively adjust the weight of each indicator in risk assessment according to its distinguishing ability, thereby improving the accuracy and pertinence of risk warning.

[0106] S5. Generate recurrent pregnancy loss risk warning results based on the risk level.

[0107] Based on the risk level, the following recurrent pregnancy loss risk warning results are generated:

[0108] Set warning thresholds α and β, where α<β;

[0109] When the overall risk of recurrent pregnancy loss is less than α, a low-risk warning result is generated;

[0110] When the overall risk of recurrent pregnancy loss is greater than or equal to α and less than β, a moderate risk warning result is generated;

[0111] When the overall risk of recurrent pregnancy loss is greater than or equal to β, a high-risk warning result is generated.

[0112] In this embodiment, first, warning thresholds α and β need to be set, where α < β. These two thresholds can be determined based on the standardized risk index described above. For example, when each risk index is standardized to be between [0, 1], threshold α can be directly set to 0.4 and threshold β to 0.7. Alternatively, these thresholds can be set through statistical analysis of historical data.

[0113] Next, the calculated overall risk level of recurrent pregnancy loss is compared with the preset threshold to generate a corresponding warning result: when the overall risk level of recurrent pregnancy loss is less than α, it indicates that the physiological state of the subject to be evaluated is closer to the normal pregnancy group and the risk is lower, so a low-risk warning result is generated; when the overall risk level of recurrent pregnancy loss is greater than or equal to α and less than β, it indicates that the physiological state of the subject to be evaluated is between the normal pregnancy group and the recurrent pregnancy loss group, and there is a certain risk, so a medium-risk warning result is generated; when the overall risk level of recurrent pregnancy loss is greater than or equal to β, it indicates that the physiological state of the subject to be evaluated is closer to the recurrent pregnancy loss group and the risk is higher, so a high-risk warning result is generated.

[0114] Through this three-level warning mechanism, more refined risk warnings can be provided according to the degree of risk, thereby better preventing the occurrence of recurrent pregnancy loss.

[0115] Example 2

[0116] See also Figure 2 The present invention provides a recurrent pregnancy loss risk early warning system based on big data, which is used to implement a recurrent pregnancy loss risk early warning method based on big data, including:

[0117] The data collection module is used to collect a large number of basic characteristics and physiological index data of research subjects, and mark the physiological index data with normal pregnancy marks or recurrent pregnancy loss marks according to their pregnancy status to establish a large database;

[0118] The data analysis module is used to classify the physiological indicator data in the large database according to the basic characteristic information, and by analyzing the classified physiological indicator data, determine the correlation data between the physiological indicators in each category and the risk of recurrent pregnancy loss, and obtain a reference database;

[0119] The feature matching module is used to obtain the basic feature information of the subject to be evaluated, and based on the basic feature information, match the physiological indicators under the corresponding classification with the recurrent pregnancy loss risk association data from the reference database and record them as target data;

[0120] A risk assessment module is used to obtain physiological indicators of the subject to be assessed and assess the risk level based on the physiological indicators of the subject to be assessed and the target data;

[0121] The early warning output module is used to generate recurrent pregnancy loss risk early warning results based on the risk level.

[0122] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only one type. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0123] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed by the present invention, which should be covered by the scope of protection of the present invention.

[0124] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for early warning of recurrent pregnancy loss risk based on big data, characterized in that: include: Collect a large number of basic characteristics and physiological index data of research subjects, and mark the physiological index data with normal pregnancy mark or recurrent pregnancy loss mark according to their pregnancy status to establish a large database; The physiological indicator data in the large database are classified according to the basic characteristic information. By analyzing the classified physiological indicator data, the correlation data between the physiological indicators in each category and the risk of recurrent pregnancy loss are determined to obtain a reference database; Obtain basic characteristic data of the subject to be evaluated. Based on the basic characteristic data, match the physiological indicators and recurrent pregnancy loss risk association data under the corresponding classification from the reference database and record them as target data; Obtaining physiological indicators of the subject to be evaluated, and assessing the risk level based on the physiological indicators of the subject to be evaluated and the target data; Based on the risk level, a recurrent pregnancy loss risk warning result is generated.

2. The method for early warning of recurrent pregnancy loss risk based on big data according to claim 1, characterized in that: The basic characteristic data include age, BMI index and number of pregnancies; the physiological index data include hormone level data, immune index data and coagulation function index data.

3. The method for early warning of recurrent pregnancy loss risk based on big data according to claim 2, characterized in that: The classification of the physiological indicator data in the large database according to basic characteristic information includes: The data is divided into several age interval groups according to the preset age threshold; the data is divided into several BMI interval groups according to the preset BMI index threshold; the data is divided into several pregnancy number groups according to the preset pregnancy number threshold; The complete classification was formed by the intersection of age interval groups, BMI interval groups and number of pregnancies groups.

4. The method for early warning of recurrent pregnancy loss risk based on big data according to claim 3, characterized in that: The analysis of the classified physiological indicator data to determine the correlation data between the physiological indicators in each category and the risk of recurrent pregnancy loss includes: For each physiological indicator data under each category, a hormone level vector is established based on the hormone level data, an immune indicator vector is established based on the immune indicator data, and a coagulation function vector is established based on the coagulation function indicator data. The hormone level vector, immune indicator vector, and coagulation function vector are sequentially combined to obtain a physiological indicator vector; Based on the labels of the physiological indicator data, the hormone level vector, immune indicator vector, coagulation function vector, and physiological indicator vector are marked with corresponding normal pregnancy labels or recurrent pregnancy loss labels; According to the hormone level vector, immune index vector, coagulation function vector and physiological index vector of normal pregnancy markers, the corresponding hormone standard vector, immune standard vector, coagulation standard vector and physiological standard vector are obtained; According to the hormone level vector, immune index vector, coagulation function vector and physiological index vector of recurrent pregnancy loss markers, the corresponding hormone risk vector, immune risk vector, coagulation risk vector and physiological risk vector are obtained; The hormone standard vector, immune standard vector, coagulation standard vector, physiological standard vector, hormone risk vector, immune risk vector, coagulation risk vector and physiological risk vector were combined into the data on the association between physiological indicators and the risk of recurrent pregnancy loss.

5. The method for early warning of recurrent pregnancy loss risk based on big data according to claim 4, characterized in that: The step of establishing the hormone level vector based on the hormone level data includes normalizing the data of estrogen level, progesterone level, human chorionic gonadotropin level, luteinizing hormone level, and follicle-stimulating hormone level, and sequentially combining the normalized data into the hormone level vector; The step of establishing an immune index vector based on the immune index data includes normalizing the data of antiphospholipid antibody content, anticardiolipin antibody content, NK cell activity value, Th1 to Th2 ratio, and cytokine level value, and sequentially combining the normalized data into an immune index vector; The method of establishing a coagulation function vector based on the coagulation function index data includes normalizing the data of prothrombin time value, activated partial thromboplastin time value, D-dimer content and fibrinogen content, and sequentially combining the normalized data into a coagulation function vector.

6. The method for early warning of recurrent pregnancy loss risk based on big data according to claim 5, characterized in that: Methods for obtaining the hormone standard vector, immune standard vector, coagulation standard vector, and physiological standard vector include: Calculate the center vectors of the hormone level vector, immune index vector, coagulation function vector and physiological index vector with normal pregnancy markers respectively as the corresponding hormone standard vector, immune standard vector, coagulation standard vector and physiological standard vector; Methods for obtaining the hormone risk vector, immune risk vector, coagulation risk vector, and physiological risk vector include: Calculate the center vectors of hormone level vectors, immune index vectors, coagulation function vectors, and physiological index vectors with recurrent pregnancy loss markers as the corresponding hormone risk vectors, immune risk vectors, coagulation risk vectors, and physiological risk vectors respectively; The center vector is the vector in the vector set that has the smallest sum of squared distances to all other vectors.

7. The method for early warning of recurrent pregnancy loss risk based on big data according to claim 6, characterized in that: The risk assessment based on the physiological indicators and target data of the subject to be assessed includes: According to the physiological indicators of the subject to be evaluated, a hormone level vector, an immune indicator vector, a coagulation function vector and a physiological indicator vector are established for the subject to be evaluated; Calculate the distance between each vector of the object to be evaluated and the corresponding standard vector and risk vector in the target data respectively; The hormone level risk index, immune indicator risk index, coagulation function risk index and comprehensive physiological risk index were obtained by normalizing the ratio of the distance between each vector and the risk vector to the distance between each vector and the standard vector. The risk level is obtained through weighted calculation based on various risk indices.

8. The method for early warning of recurrent pregnancy loss risk based on big data according to claim 7, characterized in that: The risk levels obtained through weighted calculation based on various risk indices include: Calculate the distance value between the hormone standard vector and the hormone risk vector, the distance value between the immune standard vector and the immune risk vector, the distance value between the coagulation standard vector and the coagulation risk vector, and the distance value between the physiological standard vector and the physiological risk vector; Based on the above distance values, weight coefficients are assigned to each risk index according to the rule that the larger the distance value, the larger the weight coefficient, and the sum of the weight coefficients is 1; The risk level is obtained by weighting and summing up each risk index according to the assigned weight coefficient.

9. The method for early warning of recurrent pregnancy loss risk based on big data according to claim 8, characterized in that: The recurrent pregnancy loss risk warning results generated based on the risk level include: Set warning thresholds α and β, where α<β; When the overall risk of recurrent pregnancy loss is less than α, a low-risk warning result is generated; When the overall risk of recurrent pregnancy loss is greater than or equal to α and less than β, a moderate risk warning result is generated; When the overall risk of recurrent pregnancy loss is greater than or equal to β, a high-risk warning result is generated.

10. A recurrent pregnancy loss risk warning system based on big data, used to implement the recurrent pregnancy loss risk warning method based on big data according to any one of claims 1 to 9, characterized in that: include: The data collection module is used to collect a large number of basic characteristics and physiological index data of research subjects, and mark the physiological index data with normal pregnancy marks or recurrent pregnancy loss marks according to their pregnancy status to establish a large database; The data analysis module is used to classify the physiological indicator data in the large database according to the basic characteristic information, and by analyzing the classified physiological indicator data, determine the correlation data between the physiological indicators in each category and the risk of recurrent pregnancy loss, and obtain a reference database; The feature matching module is used to obtain the basic feature information of the subject to be evaluated, and based on the basic feature information, match the physiological indicators under the corresponding classification with the recurrent pregnancy loss risk association data from the reference database and record them as target data; A risk assessment module is used to obtain physiological indicators of the subject to be assessed and assess the risk level based on the physiological indicators of the subject to be assessed and the target data; The early warning output module is used to generate recurrent pregnancy loss risk early warning results based on the risk level.

Images