Prediction method and system for clinical prognosis of preeclampsia patient

Abstract

The invention discloses a method and a system for predicting clinical prognosis of preeclampsia patients. The method comprises the steps that pregnancy parameters, clinical enzyme concentration and blood coagulation parameters of a target patient are obtained, a prediction system for clinical prognosis of the preeclampsia patient is constructed, and the prediction system comprises a premature preeclampsia prediction model, a severe preeclampsia prediction model and a pregnancy duration prediction model; inputting the pregnancy parameter, the clinical enzyme concentration and the blood coagulation parameter into a pre-constructed pre-eclampsia patient clinical prognosis prediction system to obtain a pre-eclampsia clinical prognosis result of the target patient; the premature preeclampsia prediction model is used for outputting the probability of occurrence of the premature preeclampsia of the patient, the severe preeclampsia prediction model is used for outputting the probability of occurrence of the severe preeclampsia of the patient, the pregnancy duration prediction model is used for outputting the pregnancy duration of the patient. The method has excellent prediction capability, is simple and rapid in prediction, reduces the incidence rate of preeclampsia and complications thereof, and has a very wide application prospect.

Description

Technical Field

[0001] The present invention belongs to the field of medicine, and in particular relates to a method and system for predicting the clinical prognosis of patients with preeclampsia. Background Art

[0002] Preeclampsia is one of the most serious pregnancy-related disorders worldwide and usually occurs after 20 weeks of gestation.

[0003] There is heterogeneity in the clinical prognosis of different types of preeclampsia. Compared with patients with late-onset preeclampsia (≥34 weeks of gestation), patients with early-onset preeclampsia (<34 weeks of gestation) usually show more serious clinical complications. In addition, if not treated promptly and effectively, patients with preeclampsia are also at risk of rapidly deteriorating and progressing to severe preeclampsia, including eclampsia, stroke, HELLP (hemolysis, elevated liver enzymes, and low platelets) syndrome, placental abruption, renal failure, and pulmonary edema. In addition, for newborns born from patients with early-onset preeclampsia or severe preeclampsia, the risk of premature birth, perinatal death, neurodevelopmental delay, and later cardiovascular and metabolic diseases is also greatly increased.

[0004] Currently, timely delivery of the fetus is the only clear and effective treatment for preeclampsia. Therefore, in clinical management, early identification of preeclampsia, especially early-onset preeclampsia and severe preeclampsia, and prediction of pregnancy duration are crucial to reducing perinatal adverse events in pregnant women and fetuses. Summary of the invention

[0005] In order to solve the problems existing in the background technology, the purpose of the present invention is to provide a method and system for predicting the clinical prognosis of patients with preeclampsia.

[0006] The technical solution adopted by the present invention is:

[0007] 1. A method for predicting the clinical prognosis of a patient with preeclampsia, comprising the following steps:

[0008] Step S1, obtaining pregnancy parameters, clinical enzyme concentrations and coagulation parameters of a target patient;

[0009] Step S2, constructing a prediction system for the clinical prognosis of patients with preeclampsia, wherein the prediction system includes an early-onset preeclampsia prediction model, a severe preeclampsia prediction model, and a pregnancy duration prediction model;

[0010] Step S3: inputting the pregnancy parameters, clinical enzyme concentrations and coagulation parameters obtained in step S1 into a pre-constructed prediction system for the clinical prognosis of preeclampsia patients to obtain the clinical prognosis result of preeclampsia for the target patient.

[0011] The patient's pregnancy parameters include the patient's number of pregnancies, miscarriage history and twin pregnancy; the patient's clinical enzyme content includes the patient's lactate dehydrogenase concentration and aspartate aminotransferase concentration; the patient's coagulation parameters include the patient's albumin mass concentration, thrombin time and mean platelet volume.

[0012] The early-onset preeclampsia prediction model is used to obtain the probability value Y1 of the patient developing early-onset preeclampsia. The input of the early-onset preeclampsia prediction model includes the number of pregnancies, albumin mass concentration and lactate dehydrogenase concentration of the patient. The probability value Y1 of the patient developing early-onset preeclampsia is obtained by processing according to the following formula:

[0013] Y1=(-9.44×10 -7 ×X1 3 +0.000280513×X1 2 -0.01511536×X1+0.206303505)×100%;

[0015] X1=16.67347×K1-4.166666667×C al +0.279349446×C LDH +160.985377632;

[0016] Among them, X1 represents the coefficient of early-onset preeclampsia; K1 represents the coefficient of pregnancy times, K1∈{0,1}, when the patient's pregnancy times are ≥2 times, K1 takes 1, otherwise K1 takes 0; C al Indicates albumin mass concentration, with the unit of albumin mass concentration being g / L; C LDH Indicates the lactate dehydrogenase concentration in U / L.

[0017] The early-onset preeclampsia prediction model has a built-in corresponding nomogram 1, nomogram 1 is as follows Figure 1 shown.

[0018] The severe preeclampsia prediction model is used to obtain a probability value Y2 of a patient developing severe preeclampsia. The input of the severe preeclampsia prediction model includes the patient's abortion history, thrombin time, aspartate aminotransferase concentration, and lactate dehydrogenase concentration. The probability value Y2 of a patient developing severe preeclampsia is obtained by processing according to the following formula:

[0019] Y2=(-1.751×10 -6 ×X2 3 +0.000191573×X2 2 +0.01074591×X2-0.163494649)

[0020] ×100%;

[0021] X2=7.35331×K2+6.721470194×Tt+0.666666667×C AST +0.171525005×

[0022] C LDH -104.461848027;

[0023] Among them, X2 represents the incidence coefficient of severe preeclampsia; K2 represents the coefficient of miscarriage history, K2∈{0,1}, if the patient's miscarriage history is "yes", K2 takes 1, otherwise K2 takes 0; Tt represents the thrombin time, and the thrombin time is in seconds; C AST Indicates the concentration of aspartate aminotransferase, the unit of aspartate aminotransferase concentration is U / L; C LDH Indicates the concentration of lactate dehydrogenase.

[0024] The prediction model for severe preeclampsia has a built-in corresponding nomogram 2, nomogram 2 is as follows Figure 2 shown.

[0025] The gestational duration prediction model is used to obtain the gestational duration of the patient. The input of the gestational duration prediction model includes the patient's miscarriage history, twin pregnancy, mean platelet volume, thrombin time, albumin mass concentration and lactate dehydrogenase concentration. The gestational duration of the patient is obtained according to the following formula:

[0026] The probability value Y26 of the patient giving birth in the 26th week is as follows: Y26 = (3.28 × 10 -7 ×X3 3 -0.000302356×X3 2 +0.081379813×X3-5.900267484)×100%;

[0027] The probability value Y28 of the patient giving birth in the 28th week is as follows: Y28 = (5.67 × 10 -7 ×X3 3 -0.000440897×X3 2 +0.103844391×X3-6.816429124)×100%;

[0028] The probability value Y30 of the patient giving birth in the 30th week is as follows: Y30 = (5.67 × 10 -7 ×X3 3 -0.000393535×X3 2 +0.080616007×X3-4.255086261)×100%;

[0029] The probability value Y32 of the patient giving birth in the 32nd week is as follows: Y32 = (5.67 × 10 -7 ×X3 3 -0.000326823×X3 2 +0.052370854×X3-1.664983279)×100%;

[0030] The probability value Y34 of the patient giving birth in the 34th week is as follows: Y34 = (5.67 × 10 -7 ×X3 3 -0.000267548×X3 2 +0.031664189×X3-0.21317268)×100%;

[0031] The probability value Y36 of the patient giving birth in the 36th week is as follows: Y36 = (5.67 × 10 -7 ×X3 3 -0.000201787×X3 2 +0.013523963×X3+0.643735124)×100%;

[0032] The probability value Y38 of the patient giving birth in the 38th week is as follows: Y38 = (2.5 × 10 -7 ×X3 3 -7.6191×10 -5 ×X3 2 -0.002937768×X3+0.94484719)×100%;

[0033] X3=21.95633×K2+31.43655×K3+4.888174999×Vm+11.046264528×

[0034] Tt-3.27916811×C al +0.333333333×C LDH -63.280978471;

[0035] Among them, X3 represents the coefficient of pregnancy duration; K2 represents the coefficient of miscarriage history, K2∈{0,1}, if the patient's miscarriage history is "yes", K2 takes 1, otherwise K2 takes 0; K3 represents the coefficient of twin pregnancy, K3∈{1,2}, if the patient's twin pregnancy is "yes", K3 takes 2, otherwise K3 takes 1; Vm represents the mean platelet volume, the mean platelet volume is in fL; Tt represents the thrombin time; C al Indicates albumin mass concentration; C LDH Indicates the concentration of lactate dehydrogenase.

[0036] The gestational duration prediction model has a built-in corresponding nomogram 3, nomogram 3 is as follows Figure 3 The above seven time points are commonly used clinically for measuring the duration of pregnancy.

[0037] The clinical prognosis of preeclampsia of the target patient includes the probability of the patient developing early-onset preeclampsia, the probability of the patient developing severe preeclampsia, and the duration of the patient's pregnancy.

[0038] 2. A prediction system for clinical prognosis of preeclampsia patients for implementing the above method:

[0039] Includes an early-onset preeclampsia prediction model, which is used to output the probability of a patient developing early-onset preeclampsia;

[0040] It includes a severe preeclampsia prediction model, which is used to output the probability of a patient developing severe preeclampsia;

[0041] Includes a pregnancy duration prediction model to output the patient's pregnancy duration.

[0042] The early-onset preeclampsia prediction model and the severe preeclampsia prediction model are both designed to predict different subtypes of preeclampsia, while the pregnancy duration prediction model focuses on predicting the duration of pregnancy. Although their prediction targets are different, the three models can complement each other in clinical applications and jointly provide decisions for pregnancy management, aiming to improve the prognosis of pregnant women and perinatal infants by identifying high-risk pregnant women early so that preventive and therapeutic measures can be taken in time.

[0043] The early-onset preeclampsia prediction model, severe preeclampsia prediction model and pregnancy duration prediction model of the present invention all include three modules, namely, input module, calculation module and output module. In the system of the present invention, the input module transmits patient information to the calculation module, and the input module of the early-onset preeclampsia prediction model includes the number of pregnancies, albumin mass concentration and lactate dehydrogenase concentration; the input module of the severe preeclampsia prediction model includes abortion history, thrombin time, aspartate aminotransferase concentration and lactate dehydrogenase concentration; the input module of the pregnancy duration prediction model includes abortion history, twin pregnancy, mean platelet volume, thrombin time, albumin concentration and lactate dehydrogenase concentration. The built-in nomogram nomogram shows a good C index and area under the ROC curve. The corresponding nomogram built into the calculation module can obtain the probability of early-onset preeclampsia, the probability of severe preeclampsia and the probability of delivery in patients. The output module can output the corresponding probability value.

[0044] The beneficial effects of the present invention are:

[0045] 1. The patient information required by the prediction system of the present invention has the advantage of being simple and easy to obtain, and the system of the present invention has excellent prediction capabilities.

[0046] 2. The present invention has good stability and efficacy, can easily and effectively predict the clinical prognosis of patients with preeclampsia, and has good application prospects.

[0047] 3. The prediction system of the present invention can provide clinicians with support for personalized management plans. By accurately assessing patient risks, it can help doctors develop targeted monitoring and intervention measures, thereby optimizing patient treatment plans, reducing the incidence of preeclampsia and its complications, and improving maternal and infant safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a nomogram for predicting early-onset preeclampsia in patients;

[0049] Figure 2 It is a nomogram for predicting severe preeclampsia in patients;

[0050] Figure 3 It is a nomogram for predicting the duration of pregnancy in patients with preeclampsia;

[0051] Figure 4 is the ROC curve of nomogram 1;

[0052] Figure 5 This is the ROC curve of nomogram 2;

[0053] Figure 6 This is the ROC curve of nomogram 3;

[0054] Figure 7 is the calibration curve graph of nomogram 1;

[0055] Figure 8 is the calibration curve graph of nomogram 2;

[0056] Fig. 9 This is the calibration curve of nomogram 3. DETAILED DESCRIPTION

[0057] The present invention is described in detail below in conjunction with specific implementation cases. The following implementation cases will help those skilled in the art to further understand the present invention, but will not limit the present invention in any form.

[0058] A specific embodiment of the present invention comprises the following steps:

[0059] Step S1, obtaining pregnancy parameters, clinical enzyme concentrations and coagulation parameters of a target patient;

[0060] Step S2, constructing a prediction system for the clinical prognosis of patients with preeclampsia, the prediction system including an early-onset preeclampsia prediction model, a severe preeclampsia prediction model and a pregnancy duration prediction model;

[0061] Step S3: input the pregnancy parameters, clinical enzyme concentrations and coagulation parameters obtained in step S1 into a pre-constructed prediction system for the clinical prognosis of preeclampsia patients to obtain the clinical prognosis results of preeclampsia for the target patient.

[0062] The patient's pregnancy parameters included the patient's number of pregnancies, miscarriage history, and twin pregnancy; the patient's clinical enzyme content included the patient's lactate dehydrogenase concentration and aspartate aminotransferase concentration; the patient's coagulation parameters included the patient's albumin mass concentration, thrombin time, and mean platelet volume.

[0063] The early-onset preeclampsia prediction model is used to obtain the probability value Y1 of the patient developing early-onset preeclampsia. The input of the early-onset preeclampsia prediction model includes the patient's number of pregnancies, albumin mass concentration, and lactate dehydrogenase concentration. The probability value Y1 of the patient developing early-onset preeclampsia is obtained according to the following formula:

[0064] Y1=(-9.44×10 -7 ×X1 3 +0.000280513×X1 2 -0.01511536×X1+0.206303505)×100%;

[0066] X1=16.67347×K1-4.166666667×C al +0.279349446×C LDH +160.985377632;

[0067] Among them, X1 represents the coefficient of early-onset preeclampsia; K1 represents the coefficient of pregnancy times, K1∈{0,1}, when the patient's pregnancy times are ≥2 times, K1 takes 1, otherwise K1 takes 0; C al Indicates albumin mass concentration, with the unit of albumin mass concentration being g / L; C LDH Indicates the lactate dehydrogenase concentration in U / L.

[0068] The early-onset preeclampsia prediction model has a built-in corresponding nomogram 1, nomogram 1 is as follows Figure 1 shown.

[0069] The severe preeclampsia prediction model is used to obtain the probability value Y2 of severe preeclampsia in patients. The input of the severe preeclampsia prediction model includes the patient's abortion history, thrombin time, aspartate aminotransferase concentration, and lactate dehydrogenase concentration. The probability value Y2 of severe preeclampsia in patients is obtained according to the following formula:

[0070] Y2=(-1.751×10 -6 ×X2 3 +0.000191573×X2 2 +0.01074591×X2-0.163494649)×100%;

[0072] X2=7.35331×K2+6.721470194×Tt+0.666666667×C AST +0.171525005×

[0073] C LDH -104.461848027;

[0074] Among them, X2 represents the incidence coefficient of severe preeclampsia; K2 represents the coefficient of miscarriage history, K2∈{0,1}, if the patient's miscarriage history is "yes", K2 takes 1, otherwise K2 takes 0; Tt represents the thrombin time, and the thrombin time is in seconds; C AST Indicates the concentration of aspartate aminotransferase, the unit of aspartate aminotransferase concentration is U / L; C LDH Indicates the concentration of lactate dehydrogenase.

[0075] The prediction model for severe preeclampsia has a built-in corresponding nomogram 2, nomogram 2 is as follows Figure 2 shown.

[0076] The pregnancy duration prediction model is used to obtain the patient's pregnancy duration. The input of the pregnancy duration prediction model includes the patient's miscarriage history, twin pregnancy, mean platelet volume, thrombin time, albumin mass concentration and lactate dehydrogenase concentration. The patient's pregnancy duration is obtained according to the following formula:

[0077] The probability value Y26 of the patient giving birth in the 26th week is as follows: Y26 = (3.28 × 10 -7 ×X3 3 -0.000302356×X3 2 +0.081379813×X3-5.900267484)×100%;

[0078] The probability value Y28 of the patient giving birth in the 28th week is as follows: Y28 = (5.67 × 10 -7 ×X3 3-0.000440897×X3 2 +0.103844391×X3-6.816429124)×100%;

[0079] The probability value Y30 of the patient giving birth in the 30th week is as follows: Y30 = (5.67 × 10 -7 ×X3 3 -0.000393535×X3 2 +0.080616007×X3-4.255086261)×100%;

[0080] The probability value Y32 of the patient giving birth in the 32nd week is as follows: Y32 = (5.67 × 10 -7 ×X3 3 -0.000326823×X3 2 +0.052370854×X3-1.664983279)×100%;

[0081] The probability value Y34 of the patient giving birth in the 34th week is as follows: Y34 = (5.67 × 10 -7 ×X3 3 -0.000267548×X3 2 +0.031664189×X3-0.21317268)×100%;

[0082] The probability value Y36 of the patient giving birth in the 36th week is as follows: Y36 = (5.67 × 10 -7 ×X3 3 -0.000201787×X3 2 +0.013523963×X3+0.643735124)×100%;

[0083] The probability value Y38 of the patient giving birth in the 38th week is as follows: Y38 = (2.5 × 10 -7 ×X3 3 -7.6191×10 -5 ×X3 2 -0.002937768×X3+0.94484719)×100%;

[0084] X3=21.95633×K2+31.43655×K3+4.888174999×Vm+11.046264528×

[0085] Tt-3.27916811×C al +0.333333333×C LDH -63.280978471;

[0086] Among them, X3 represents the coefficient of pregnancy duration; K2 represents the coefficient of miscarriage history, K2∈{0,1}, if the patient's miscarriage history is "yes", K2 takes 1, otherwise K2 takes 0; K3 represents the coefficient of twin pregnancy, K3∈{1,2}, if the patient's twin pregnancy is "yes", K3 takes 2, otherwise K3 takes 1; Vm represents the mean platelet volume, the mean platelet volume is in fL; Tt represents the thrombin time; C al Indicates albumin mass concentration; C LDH Indicates the concentration of lactate dehydrogenase.

[0087] The gestational duration prediction model has a built-in corresponding nomogram 3, nomogram 3 is as follows Figure 3 shown.

[0088] The clinical prognostic outcomes of preeclampsia for the target patients include the probability of the patient developing early-onset preeclampsia, the probability of the patient developing severe preeclampsia, and the duration of the patient's pregnancy.

[0089] The prediction system for the clinical prognosis of patients with preeclampsia of the present invention includes: an early-onset preeclampsia prediction model for outputting the probability of a patient developing early-onset preeclampsia; a severe preeclampsia prediction model for outputting the probability of a patient developing severe preeclampsia; and a pregnancy duration prediction model for outputting the pregnancy duration of the patient.

[0090] The specific steps of the first embodiment of the present invention are as follows:

[0091] Step S1, preliminary preparation: obtaining the target patient's blood test information and pregnancy parameters, the blood test information including clinical enzyme concentration and coagulation parameters;

[0092] Basic information: pregnancy number coefficient (pregnancy number ≥ 2 times is recorded as 1, 1 time is recorded as 0), miscarriage history coefficient (miscarriage history is recorded as 1, no record is recorded as 0), twin pregnancy coefficient (twin pregnancy is recorded as 2, no record is recorded as 1)

[0093] Blood test information: Blood test information was collected within one week before patients started treatment.

[0094] The main information obtained includes: albumin mass concentration (in g / L), lactate dehydrogenase concentration (in U / L), thrombin time (in seconds), aspartate aminotransferase concentration (in U / L), and mean platelet volume (in fL).

[0095] Step S2, constructing a prediction system for the clinical prognosis of patients with preeclampsia, the prediction system including an early-onset preeclampsia prediction model, a severe preeclampsia prediction model and a pregnancy duration prediction model;

[0096] Step S3: input the blood test information and pregnancy parameters obtained in step S1 into a pre-constructed prediction system for the clinical prognosis of preeclampsia patients to obtain the clinical prognosis result of preeclampsia for the target patient.

[0097] Step S3 is as follows: the user inputs the number of pregnancies, albumin concentration and lactate dehydrogenase concentration in the input module of the early-onset preeclampsia prediction model, and the calculation module can substitute the above information into nomogram 1 for calculation, and obtain the probability of the patient having early-onset preeclampsia, which is presented to the user through the output module. The user inputs the history of miscarriage, thrombin time, aspartate aminotransferase concentration and lactate dehydrogenase concentration in the input module of the severe preeclampsia prediction model, and the calculation module can substitute the above information into nomogram 2 for calculation, and obtain the probability of the patient having severe preeclampsia, which is presented to the user through the output module. The user inputs the history of miscarriage, twin pregnancy, mean platelet volume, thrombin time, albumin concentration and lactate dehydrogenase concentration in the input module of the pregnancy duration prediction model, and the calculation module can substitute the above information into nomogram 3 for calculation, and obtain the probability of the patient giving birth in each week, which is presented to the user through the output module.

[0098] The prediction model of the present invention is the key technical feature of the prediction system for the clinical prognosis of patients with preeclampsia. The prediction effect of the model of the present invention will be further described below.

[0099] Verification of the effect of the model of the present invention:

[0100] Inclusion criteria: This part includes patients with preeclampsia diagnosed in the Run Run Shaw Hospital affiliated to Zhejiang University School of Medicine from 2017 to 2022.

[0101] The main inclusion criteria are as follows: meeting the diagnostic criteria for preeclampsia in the Diagnosis and Treatment of Hypertension and Preeclampsia in Pregnancy: Chinese Clinical Practice Guidelines (2020), mainly including systolic blood pressure higher than 140 mmHg and / or diastolic blood pressure higher than 90 mmHg after 20 weeks of pregnancy, accompanied by the following symptoms: (1) urine protein ≥ 0.3 g / 24 hours, or urine protein / creatinine ratio ≥ 0.3; (2) dysfunction of important organs such as heart function, liver function, kidney function and hematopoietic function.

[0102] The main exclusion criteria were as follows: (1) patients with previous hypertension and immune dysfunction; (2) patients without complete maternal and infant records. This study was approved by the Ethics Committee of the Run Run Shaw Hospital, Zhejiang University School of Medicine.

[0103] Research subjects: A total of 361 patients with preeclampsia were included in this study. The 361 research subjects were divided into 2 groups: Group 1 with 253 cases was the training group, and Group 2 with 108 cases was the validation group.

[0104] Clinical prognosis analysis:

[0105] In group 1, multivariate logistic regression model analysis showed that compared with patients with late-onset preeclampsia, patients with early-onset preeclampsia were more likely to have multiple pregnancies (odds ratio OR = 3.244, P (P value) = 0.005), low albumin (OR = 0.745, P < 0.001) and high lactate dehydrogenase (OR = 1.020, P < 0.001); multivariate logistic regression model analysis showed that compared with patients with non-severe preeclampsia, patients with severe preeclampsia were more likely to have a history of miscarriage (OR = 2.057, P = 0.017), prolonged thrombin time (OR = 1.934, P < 0.001), high asparagus root growth factor (OR = 2.06, P < 0.001), and high uric acid (OR = 1.90, P < 0.001). Multivariate Cox risk model analysis showed that history of miscarriage (HR=1.822, P<0.001), twin pregnancy (HR=2.233, P<0.001), high mean platelet volume (HR=1.134, P=0.014), prolonged thrombin time (HR=1.351, P<0.001), high lactate dehydrogenase (HR=1.009, P<0.001) and low albumin (HR=0.912, P<0.001) were significantly associated with shorter gestational duration.

[0106] The above results indicate that these indicators can specifically predict the clinical prognosis of patients with preeclampsia.

[0107] Model building and validation:

[0108] 1) Model building

[0109] like Figure 1 Based on the three indicators of number of pregnancies, albumin and lactate dehydrogenase, nomogram 1 was established to predict the probability of early-onset preeclampsia in patients; Figure 2 Based on the four indicators of miscarriage history, thrombin time, aspartate aminotransferase and lactate dehydrogenase, nomogram 2 was established to predict the probability of severe preeclampsia in patients; Figure 3 Based on six indicators including history of miscarriage, twin pregnancy, mean platelet volume, thrombin time, albumin and lactate dehydrogenase, nomogram 3 was established to predict the probability of delivery at 26, 28, 30, 32, 34, 36 and 38 weeks in patients with preeclampsia.

[0110] 2) Model evaluation

[0111] like Figure 4 As shown in a, the area under the ROC curve of nomogram 1 for predicting the occurrence of early-onset preeclampsia in patients is 0.843; Figure 5As shown in a, the area under the ROC curve of nomogram 2 for predicting the occurrence of severe preeclampsia in patients was 0.814; Figure 6 As shown in a, the areas under the ROC curve of nomogram 3 for predicting delivery at 26, 28, 30, 32, 34, 36, and 38 weeks in patients with preeclampsia were 0.852, 0.892, 0.929, 0.900, 0.831, 0.854, and 0.815, respectively.

[0112] In order to check whether the model prediction results are consistent with the actual results, three nomogram calibration curves are drawn in group 1, respectively. Figure 7 a. Figure 8 a and Fig. 9 a- Fig. 9 As shown in Figure 6, the predicted results of the three models are in good agreement with the actual results.

[0113] 3) Model Validation

[0114] A total of 108 patients in group 2 were used to externally validate the performance of the three nomograms.

[0115] In group 2, if Figure 4 As shown in b, the area under the ROC curve of nomogram 1 for predicting the occurrence of early-onset preeclampsia in patients is 0.828; Figure 5 As shown in b, the area under the ROC curve of nomogram 2 for predicting the occurrence of severe preeclampsia in patients was 0.823; Figure 6 As shown in b, the areas under the ROC curve of nomogram 3 for predicting delivery at 26, 28, 30, 32, 34, 36, and 38 weeks in patients with preeclampsia were 0.935, 0.800, 0.866, 0.857, 0.832, 0.800, and 0.851, respectively.

[0116] In order to check whether the model prediction results are consistent with the actual results, three nomogram calibration curves were drawn in group 2, respectively. Figure 7 b. Figure 8 b. and Fig. 9 h- Fig. 9 As shown in Figure 2, the results show that the predicted results of the three models are in good agreement with the actual results.

[0117] In the specific implementation, a simple web calculator for calculating the probability of clinical prognosis was made based on the calculation formula of the three nomograms. After inputting the numerical value of the variable, the corresponding probability value can be automatically output. It can help clinicians evaluate the probability of different clinical prognoses when patients develop preeclampsia, so as to formulate different personalized management measures.

[0118] The system for predicting the clinical prognosis of patients with preeclampsia of the present invention has good stability and efficiency, can easily and effectively predict the clinical prognosis of patients with preeclampsia, and has good application prospects.

[0119] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting the clinical prognosis of patients with preeclampsia, characterized in that: The following steps are involved: Step S1, obtaining pregnancy parameters, clinical enzyme concentrations and coagulation parameters of a target patient; Step S2, constructing a prediction system for the clinical prognosis of patients with preeclampsia, wherein the prediction system includes an early-onset preeclampsia prediction model, a severe preeclampsia prediction model, and a pregnancy duration prediction model; Step S3: inputting the pregnancy parameters, clinical enzyme concentrations and coagulation parameters obtained in step S1 into a pre-constructed prediction system for the clinical prognosis of preeclampsia patients to obtain the clinical prognosis result of preeclampsia for the target patient.

2. The method for predicting clinical prognosis of a patient with preeclampsia according to claim 1, characterized in that: The patient's pregnancy parameters include the patient's number of pregnancies, miscarriage history and twin pregnancy; the patient's clinical enzyme content includes the patient's lactate dehydrogenase concentration and aspartate aminotransferase concentration; the patient's coagulation parameters include the patient's albumin mass concentration, thrombin time and mean platelet volume.

3. The method for predicting clinical prognosis of a patient with preeclampsia according to claim 2, characterized in that: The early-onset preeclampsia prediction model is used to obtain the probability value Y1 of the patient developing early-onset preeclampsia. The input of the early-onset preeclampsia prediction model includes the number of pregnancies, albumin mass concentration and lactate dehydrogenase concentration of the patient. The probability value Y1 of the patient developing early-onset preeclampsia is obtained by processing according to the following formula: Y1=(-9.44×10 -7 ×X1 3 +0.000280513×X1 2 -0.01511536×X1+0.206303505)×100%; X1=16.67347×K1-4.166666667×C al +0.279349446×C LDH +160.985377632; Among them, X1 represents the coefficient of early-onset preeclampsia; K1 represents the coefficient of pregnancy times, K1∈{0,1}, when the patient's pregnancy times are ≥2 times, K1 takes 1, otherwise K1 takes 0; C al Indicates albumin mass concentration; C LDH Indicates the concentration of lactate dehydrogenase.

4. The method for predicting clinical prognosis of a patient with preeclampsia according to claim 2, characterized in that: The severe preeclampsia prediction model is used to obtain a probability value Y2 of a patient developing severe preeclampsia. The input of the severe preeclampsia prediction model includes the patient's abortion history, thrombin time, aspartate aminotransferase concentration, and lactate dehydrogenase concentration. The probability value Y2 of a patient developing severe preeclampsia is obtained by processing according to the following formula: Y2=(-1.751×10 -6 ×X2 3 +0.000191573×X2 2 +0.01074591×X2-0.163494649)×100%; <h2 style=";text-align:left;direction:ltr">X2 = 7.35331×K2 + 6.721470194×Tt + 0.666666667×C<h2 style=";text-align:left;direction:ltr"> AST <h2 style=";text-align:left;direction:ltr"> +0.171525005× C LDH -104.461848027； Among them, X2 represents the incidence coefficient of severe preeclampsia; K2 represents the coefficient of miscarriage history, K2∈{0,1}, if the patient's miscarriage history is "yes", K2 takes 1, otherwise K2 takes 0; Tt represents the thrombin time; C AST represents the concentration of aspartate aminotransferase; C LDH Indicates the concentration of lactate dehydrogenase.

5. The method for predicting clinical prognosis of a patient with preeclampsia according to claim 2, characterized in that: The gestational duration prediction model is used to obtain the gestational duration of the patient. The input of the gestational duration prediction model includes the patient's miscarriage history, twin pregnancy, mean platelet volume, thrombin time, albumin mass concentration and lactate dehydrogenase concentration. The gestational duration of the patient is obtained according to the following formula: The probability value Y26 of the patient giving birth in the 26th week is as follows: Y26 = (3.28 × 10 -7 ×X3 3 -0.000302356×X3 2 +0.081379813×X3-5.900267484)×100%; The probability value Y28 of the patient giving birth in the 28th week is as follows: Y28 = (5.67 × 10 -7 ×X3 3 -0.000440897×X3 2 +0.103844391×X3-6.816429124)×100%; The probability value Y30 of the patient giving birth in the 30th week is as follows: Y30 = (5.67 × 10 -7 ×X3 3 -0.000393535×X3 2 +0.080616007×X3-4.255086261)×100%; The probability value Y32 of the patient giving birth in the 32nd week is as follows: Y32 = (5.67 × 10 -7 ×X3 3 -0.000326823×X3 2 +0.052370854×X3-1.664983279)×100%; The probability value Y34 of the patient giving birth in the 34th week is as follows: Y34 = (5.67 × 10 -7 ×X3 3 -0.000267548×X3 2 +0.031664189×X3-0.21317268)×100%; The probability value Y36 of the patient giving birth in the 36th week is as follows: Y36 = (5.67 × 10 -7 ×X3 3 -0.000201787×X3 2 +0.013523963×X3+0.643735124)×100%; The probability value Y38 of the patient giving birth in the 38th week is as follows: Y38 = (2.5 × 10 -7 ×X3 3 -7.6191×10 -5 ×X3 2 -0.002937768×X3+0.94484719)×100%; X3=21.95633×K2+31.43655×K3+4.888174999×Vm+11.046264528× Tt-3.27916811×C al +0.333333333×C LDH -63.280978471; Among them, X3 represents the coefficient of pregnancy duration; K2 represents the coefficient of miscarriage history, K2∈{0,1}, if the patient's miscarriage history is "yes", K2 takes 1, otherwise K2 takes 0; K3 represents the coefficient of twin pregnancy, K3∈{1,2}, if the patient's twin pregnancy is "yes", K3 takes 2, otherwise K3 takes 1; Vm represents the mean platelet volume; Tt represents the thrombin time; C al Indicates albumin mass concentration; C LDH Indicates the concentration of lactate dehydrogenase.

6. The method for predicting clinical prognosis of a patient with preeclampsia according to claim 1, characterized in that: The clinical prognosis of preeclampsia in the target patient includes the probability of the patient developing early-onset preeclampsia, the probability of the patient developing severe preeclampsia, and the duration of the patient's pregnancy.

7. A system for predicting clinical prognosis of preeclampsia patients for implementing the method according to any one of claims 1 to 6, characterized in that: Includes an early-onset preeclampsia prediction model, which is used to output the probability of a patient developing early-onset preeclampsia; It includes a severe preeclampsia prediction model, which is used to output the probability of a patient developing severe preeclampsia; Includes a pregnancy duration prediction model to output the patient's pregnancy duration.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images