Method, device, medium and computer program product for predicting childbirth enormous risk of gestational diabetes mellitus women

Abstract

The invention discloses a risk prediction method, device, medium and computer program product for childbirth of a woman suffering from gestational diabetes mellitus. The prediction method comprises the following steps: acquiring examination result data of a form index and a biochemical index of the woman suffering from gestational diabetes mellitus in the middle pregnancy period; and processing the obtained data to obtain a data set, analyzing the data set through a prediction model, and predicting the risk of delivery of giant children by the gestational diabetes mellitus women. According to the prediction method, the selected indexes are reasonable, comprehensive and easy to collect, the insulin resistance related index TyG-BMI is innovatively introduced, the pertinence is high, and the prediction accuracy is high; meanwhile, the prediction risk is visually displayed, the influence degree of each risk factor can be visually displayed, clinical application of obstetricians and communication between the obstetricians and patients are facilitated, and great popularization value is achieved.

Description

Technical Field

[0001] The present invention relates to perinatal medicine, and in particular to a method, device, medium and computer program product for predicting the risk of macrosomia in women with gestational diabetes mellitus. Background Art

[0002] Macrosomia (birth weight ≥4000g) is a major public health issue affecting maternal and infant health. Its occurrence is significantly associated with short- and long-term risks such as neonatal asphyxia, birth trauma, and maternal complications. In clinical practice, the prediction of macrosomia using conventional abdominal ultrasound is often affected by factors such as fetal position, amniotic fluid volume, and operator experience, resulting in limited predictive value. Consequently, medical professionals have been committed to developing relevant risk prediction models in recent years. However, current models often rely on late-pregnancy biometric parameters or umbilical cord blood metabolomics testing, which have drawbacks such as a late prediction window (mainly after 37 weeks of gestation), high testing costs, or difficulty in obtaining samples. Furthermore, existing models generally lack visual risk assessment tools, limiting their widespread application in primary healthcare institutions.

[0003] Among the many risk factors for macrosomia, gestational diabetes mellitus (GDM) has been shown to be an independent risk factor, significantly increasing the incidence of macrosomia compared with healthy pregnant women. Accurately predicting the risk of macrosomia in pregnant women with GDM is crucial for early intervention and improving maternal and fetal outcomes. However, specific prediction tools for this high-risk population are lacking. Summary of the Invention

[0004] Purpose of the invention: The purpose of the present invention is to provide a method for predicting the risk of macrosomia in women with gestational diabetes mellitus. The second purpose is to provide equipment, media and computer program products for use in this method.

[0005] Technical solution: The method for predicting the risk of macrosomia in women with gestational diabetes mellitus according to the present invention comprises the following steps:

[0006] (1) Collect data, including mid-pregnancy morphological indicators and blood biochemical index examination results of women with gestational diabetes;

[0007] (2) Process the data to obtain a data set;

[0008] (3) Analyzing the data set using a prediction model to predict the risk of macrosomia in women with gestational diabetes, wherein the prediction model is:

[0009] P=1÷{1+exp[-(-16.0208-0.073513×X1-0.064942×X2-0.724375×X3+

[0010] 0.007517×X4+0.088095×X5+0.016533×X6)]};

[0011] Where P is the probability of giving birth to a macrosomia in women with gestational diabetes, X1 is the free thyroxine value, X2 is the serum total protein value, X3 is the ratio of alanine aminotransferase to aspartate aminotransferase, X4 is the lactate dehydrogenase value, X5 is the height value, and X6 is the product of the triglyceride-glucose index and body mass index.

[0012] Preferably, in the method for predicting the risk of macrosomia in women with gestational diabetes,

[0013] When the obtained probability is less than 0.1, it is predicted that women with gestational diabetes have a low risk of delivering a macrosomia;

[0014] When the obtained probability is 0.1-0.6, it is predicted that women with gestational diabetes are at medium risk of giving birth to a macrosomia;

[0015] When the obtained probability is greater than 0.6, it is predicted that women with gestational diabetes are at high risk of giving birth to a macrosomia.

[0016] Preferably, the second trimester in step 1 is 20-24 weeks of pregnancy.

[0017] Preferably, the morphological indicators in step 1 include height and body mass index, and the biochemical indicators include free thyroxine, serum total protein, alanine aminotransferase to aspartate aminotransferase ratio, lactate dehydrogenase, and triglyceride-glucose index.

[0018] Preferably, the step of processing data in step 2 includes:

[0019] (21) The collected data were converted to units: free thyroxine units were converted to pmol / L, serum total protein units were converted to g / L, lactate dehydrogenase units were converted to U / L, and height units were converted to cm;

[0020] (22) Remove the data units and keep only the numerical values.

[0021] Preferably, after step 3, the data set is scored, and the total score is calculated, and a visual image is generated by combining the probability obtained by the prediction model.

[0022] Preferably, in the data set, the scoring formula for each data item includes:

[0023] (a) Free thyroxine value scoring formula: P FT4 =-2×X1+52;

[0024] (b) Score formula for serum total protein value: P TP =-1.7769×X2+142.27;

[0025] (c) Scoring formula for the ratio of alanine aminotransferase to aspartate aminotransferase: P ALT / AST =-20×X3+60;

[0026] (d) Lactate dehydrogenase value scoring formula: P LDH =0.207×X4-10.417;

[0027] (e) Height score formula: P h =2.4333×X5-340.89;

[0028] (f) The scoring formula for the product of triglyceride-glucose index and body mass index is: P TyG-BMI =0.4563×X6-64.073;

[0029] Among them, P FT4 is the score of free thyroxine value, X1 is the free thyroxine value; P TP is the score assigned to the serum total protein value, X2 is the serum total protein value; P ALT / AST is the score of the ratio of alanine aminotransferase to aspartate aminotransferase, X3 is the ratio of alanine aminotransferase to aspartate aminotransferase; P LDH is the score of lactate dehydrogenase value, X4 is the lactate dehydrogenase value; P h is the score assigned to the height value, X5 is the height value; P TyG-BMI It is the score of the product of triglyceride-glucose index and body mass index, X6 is the product of triglyceride-glucose index and body mass index. If the calculated score is less than 0, 0 points will be given; if the calculated score is greater than 100, 100 points will be given.

[0030] The computer device of the present invention includes a memory, a processor, and a computer program stored in the memory, and the processor implements the above-mentioned prediction method when executing the computer program.

[0031] The computer-readable storage medium of the present invention stores a computer program thereon, and the computer program implements the aforementioned prediction method when executed by a processor.

[0032] The computer program product of the present invention includes a computer program, and when the computer program is executed by a processor, the prediction method is implemented.

[0033] Beneficial effects: Compared with the existing technology, the present invention has the following significant advantages: 1. The prediction method is applicable to pregnant women with gestational diabetes, with strong pertinence and high prediction accuracy; 2. The selected prediction indicators are reasonable and comprehensive, easy to collect, and the innovative introduction of insulin resistance-related indicators TyG-BMI further improves the accuracy of prediction; 3. The visual presentation of predicted risks can intuitively show the degree of influence of various risk factors, facilitate obstetricians' clinical application and communication with patients, and has great promotion value. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 Results of screening potential predictor variables for Lasso regression analysis;

[0035] Figure 2 The ROC curves of the risk prediction model for macrosomia in women with gestational diabetes mellitus in the training and validation sets were used.

[0036] Figure 3 To provide calibration curves for the risk prediction model for macrosomia in women with gestational diabetes mellitus in the training and validation sets;

[0037] Figure 4 To develop clinical decision curves for the macrosomia risk prediction model for women with gestational diabetes in training and validation sets;

[0038] Figure 5 Visualization of the results of the method for predicting the risk of macrosomia in women with gestational diabetes applied to pregnant woman 5;

[0039] Figure 6 Visualization of the results of the method for predicting the risk of macrosomia in women with gestational diabetes mellitus applied to pregnant women 9;

[0040] Figure 7 The present invention is a schematic diagram of the structure of a device that can realize a method for predicting the birth of a macrosomia in women with gestational diabetes. DETAILED DESCRIPTION

[0041] The technical solution of the present invention is further described below.

[0042] Example 1: Establishment of a risk prediction model for macrosomia in women with gestational diabetes mellitus (GDM)

[0043] (1) Summarize the information of 8619 pregnant women admitted to the hospital from 2017 to 2022 and conduct data analysis and screening;

[0044] Exclusion criteria were as follows: twin or multiple pregnancy, stillbirth; history of diabetes, liver disease, kidney disease, hypertension, coronary heart disease, connective tissue disease, polycystic ovary syndrome, etc. before pregnancy; no prenatal examination record in this hospital;

[0045] Inclusion criteria were as follows: complete basic information; age ranged from 18 to 45 years; maternal gestational age ≥ 37 weeks; single live birth without malformation; complete maternal basic information, mid-pregnancy examination information, biochemical indicators (including thyroid-stimulating hormone, free thyroxine, anti-thyroid peroxidase antibody, total protein, albumin, glucose, total cholesterol, triglycerides, alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase, and lactate dehydrogenase), pregnancy complications or comorbidities, maternal and fetal pregnancy outcomes, and blood glucose values at three time points during the oral glucose tolerance test (OGTT);

[0046] The determination of macrosomia was based on international standards: newborns with a birth weight ≥ 4000 g. Finally, a database of 5550 pregnant women with gestational diabetes was established.

[0047] (2) Data processing and variable screening. The case database was randomly divided into training sets (n training =3884) and validation set (Validation Set, n validation =1666), among which the incidence of macrosomia in the training set and validation set were 7.96% (309 / 3884) and 8.58% (143 / 1666), respectively;

[0048] (3) Pregnant women in the training set were divided into two groups (GDM-normal group and GDM-macrosomia group) according to the birth weight of their newborns. Statistical analysis was performed using SPSS 27.0. Normally distributed quantitative data were expressed as mean ± standard deviation (x ± s), and the two groups were compared using the independent sample t test. Non-normally distributed quantitative data were expressed as median and interquartile range (P25, P75), and the two groups were compared using the Mann-Whitney test. The differences in maternal factors, mid-pregnancy biochemical indicators, especially insulin resistance indicators, and other risk factors were compared between the GDM-normal group and the GDM-macrosomia group. The results showed that the height, pre-pregnancy weight and pre-pregnancy BMI, mid-pregnancy weight and BMI, biochemical indicators (including TG, FBG and LDH), TyG-BMI and neonatal birth weight of the GDM-macrosomia group were significantly higher than those of the GDM-normal group (P<0.05), while the TP, ALT, AST, FT4 levels and ALT / AST ratio in the GDM-macrosomia group were significantly lower than those in the GDM-normal group (P<0.05). See Table 1 for details.

[0049] Table 1 Comparison of clinical data between macrosomia and normal infants in the training set

[0050]

[0051]

[0052] (4) In the training set, the Lasso regression method was used to further screen potential predictors of macrosomia in pregnant women with gestational diabetes mellitus by using the glmnet package of R software (v4.4.2). The results were as follows: Figure 1 As shown, there are 7 variables with non-zero coefficients, including FBG, FT4, TP, ALT / AST, LDH, height, and TyG-BMI.

[0053] (5) A multivariate logistic regression analysis was performed using the risk factors identified by the univariate logistic regression analysis in step 3 and the Lasso regression analysis in step 4 as independent variables and the occurrence of macrosomia as the dependent variable. The glm{stats} package of R software (v4.4.2) was used. The results are shown in Table 2. Finally, six independent risk factors for macrosomia in pregnant women with gestational diabetes were identified, namely, FT4, TP, ALT / AST, LDH, height, and TyG-BMI.

[0054] Table 2 Multivariate Logistic regression analysis of factors influencing the delivery of macrosomia by pregnant women with GDM

[0055]

[0056] Furthermore, the following risk prediction model for women with gestational diabetes giving birth to a macrosomia can be derived from the above table:

[0057] P=1÷{1+exp[-(-16.0208-0.073513×X1-0.064942×X2-0.724375×X3+

[0058] 0.007517×X4+0.088095×X5+0.016533×X6)]};

[0059] Example 2: Evaluation of a risk prediction model for macrosomia in women with gestational diabetes mellitus

[0060] The pROC and ggROC packages in R software (v4.4.2) were used to analyze the discrimination of the prediction model, plot the receiver operating characteristic curve (ROC), and calculate the area under the curve (AUC). The calibrate and val.prob functions in the rms package in R software (v4.4.2) were used to plot the calibration curve of the prediction algorithm, perform calibration tests, and calculate the mean absolute error (MAE). The decision curve analysis (DCA) was drawn using the rmda package in R software (v4.4.2) to evaluate the practicality and application value of the prediction model.

[0061] The ROC curves of the risk prediction model for macrosomia in women with gestational diabetes mellitus in the training set and validation set are as follows: Figure 2 As shown in the figure, the AUC in the ROC curves of the training set and the validation set are 0.714 and 0.722 respectively, indicating that the performance on the training set and the validation set is relatively consistent, with good discrimination performance and no obvious overfitting or underfitting.

[0062] The calibration curves of the risk prediction model for macrosomia in women with gestational diabetes mellitus in the training set and validation set are as follows: Figure 3 As shown in the figure, there is consistency between the predicted probability and the actual probability of the prediction model on the training set and the validation set, and the prediction model is more accurate in the medium and high probability areas of the training set and the validation set.

[0063] The clinical decision curves of the risk prediction model for macrosomia in women with gestational diabetes in the training set and validation set are as follows: Figure 4 As shown, the prediction model shows positive net benefits in both data sets, that is, the model's predictions are valuable and beneficial.

[0064] Example 3: Method for predicting the probability of macrosomia in women with gestational diabetes mellitus (GDM) during pregnancy

[0065] Pregnant women who were hospitalized for delivery between July and September 2022 and whose discharge diagnosis included "gestational diabetes" were collected for further evaluation. The data for each pregnant woman were processed as follows:

[0066] (1) Data collection included morphological indicators of women with gestational diabetes during the second trimester (22 weeks of gestation), including height and body mass index; and biochemical indicators, including free thyroxine (FT4), serum total protein (TP), alanine aminotransferase to aspartate aminotransferase ratio (ALT / AST), lactate dehydrogenase (LDH), and triglyceride-glucose index (TyG-BMI);

[0067] (2) Convert the obtained data into units, where free thyroxine is converted to pmol / L, serum total protein is converted to g / L, lactate dehydrogenase is converted to U / L, and height is converted to cm; after conversion, remove the data units and retain only the numerical values; obtain the data set;

[0068] (3) Analyzing the data set using a prediction model to predict the risk of macrosomia in women with gestational diabetes, wherein the prediction model is:

[0069] P=1÷{1+exp[-(-16.0208-0.073513×X1-0.064942×X2-0.724375×X3+

[0070] 0.007517×X4+0.088095×X5+0.016533×X6)]};

[0071] Where P is the probability of giving birth to a macrosomia in women with gestational diabetes, X1 is the free thyroxine value, X2 is the serum total protein value, X3 is the ratio of alanine aminotransferase to aspartate aminotransferase, X4 is the lactate dehydrogenase value, X5 is the height value, and X6 is the product of the triglyceride-glucose index and body mass index.

[0072] (4) Assign points to the data set obtained in step 2, calculate the total score, and generate a visual image based on the probability obtained by the prediction model. The scoring formula for each data item is as follows:

[0073] Free thyroxine value scoring formula: P FT4 =-2×X1+52;

[0074] The scoring formula for serum total protein value is: P TP =-1.7769×X2+142.27;

[0075] The scoring formula for the ratio of alanine aminotransferase to aspartate aminotransferase is: P ALT / AST =-20×X3+60;

[0076] Lactate dehydrogenase value scoring formula: P LDH =0.207×X4-10.417;

[0077] Height value assignment formula: P h =2.4333×X5-340.89;

[0078] The scoring formula for the product of triglyceride-glucose index and body mass index is: P TyG-BMI =0.4563×X6-64.073;

[0079] Among them, P FT4 is the score of free thyroxine value, X1 is the free thyroxine value; P TP is the score assigned to the serum total protein value, X2 is the serum total protein value; P ALT / AST is the score of the ratio of alanine aminotransferase to aspartate aminotransferase, X3 is the ratio of alanine aminotransferase to aspartate aminotransferase; P LDH is the score of lactate dehydrogenase value, X4 is the lactate dehydrogenase value; P h is the score assigned to the height value, X5 is the height value; P TyG-BMI X is the score of the product of triglyceride-glucose index and body mass index, and X6 is the product of triglyceride-glucose index and body mass index. If the calculated score is less than 0, 0 points will be given; if the calculated score is greater than 100, 100 points will be given.

[0080] The data of a total of 10 pregnant women are as follows:

[0081] Pregnant case 1, the height of the pregnant woman was 162 cm, FT4 12.70 pmol / L, TP 65.1 g / L, ALT 5.6 U / L, AST 12.3 U / L, LDH 202.1 U / L, TyG-BMI 273.49, gestational age 39 weeks + 4, and the birth weight of the newborn was 4450 g;

[0082] Pregnant case 2, the height of the pregnant woman was 168 cm, FT4 14.62 pmol / L, TP 62.8 g / L, ALT 7.2 U / L, AST 13.4 U / L, LDH 259.2 U / L, TyG-BMI 248.28, gestational age 40 weeks + 2, and the birth weight of the newborn was 4080 g;

[0083] Pregnant case 3, the maternal height was 165 cm, FT4 12.55 pmol / L, TP 59.1 g / L, ALT 6.6 U / L, AST 10.9 U / L, LDH 138.0 U / L, TyG-BMI 277.40, gestational age 39 weeks + 5, and the birth weight of the newborn was 4390 g;

[0084] Pregnant case 4, the height of the pregnant woman was 173 cm, FT4 12.23 pmol / L, TP 64.8 g / L, ALT 13.9 U / L, AST 16.8 U / L, LDH 191.6 U / L, TyG-BMI 255.48, gestational period 40 weeks + 2, and the birth weight of the newborn was 4940 g;

[0085] Pregnant case 5, the height of the pregnant woman was 165 cm, FT4 11.35 pmol / L, TP 70.8 g / L, ALT 6.5 U / L, AST40.2 U / L, LDH 609.7 U / L, TyG-BMI 213.00, gestational age 39 weeks + 2, and the birth weight of the newborn was 4070 g;

[0086] Pregnant case 6, maternal height 160cm, FT4 11.90pmol / L, TP 63.4g / L, ALT 22.2U / L, AST15.5U / L, LDH 138.1U / L, TyG-BMI 195.04, gestational age 39 weeks + 5, newborn weight 2600g;

[0087] Pregnant case 7, maternal height 168cm, FT4 13.95pmol / L, TP 73.6g / L, ALT 30.7U / L, AST22.4U / L, LDH 143.8U / L, TyG-BMI 146.14, gestational age 40 weeks + 6, newborn weight 3190g;

[0088] Pregnant case 8, maternal height 153cm, FT4 14.50pmol / L, TP 64.3g / L, ALT 9.9U / L, AST18.4U / L, LDH 162.0U / L, TyG-BMI 181.16, gestational age 39 weeks + 5, newborn weight 3310g;

[0089] Pregnant case 9, maternal height 158cm, FT4 12.52pmol / L, TP 64.3g / L, ALT 17.7U / L, AST21.3U / L, LDH 151.6U / L, TyG-BMI 189.24, gestational age 40 weeks + 4, newborn weight 3090g;

[0090] Pregnant case 10, the pregnant woman was 158 cm tall, FT4 13.46 pmol / L, TP 62.3 g / L, ALT 11.8 U / L, AST 15.9 U / L, LDH 132.6 U / L, TyG-BMI 157.00, gestational period 41 weeks + 2, and the newborn weighed 3020 g.

[0091] The predicted probability and score of the risk of delivering a macrosomia for all pregnant women, as well as the actual delivery situation, are shown in Table 3.

[0092] Table 3 Predicted probability, score, and actual delivery situation

[0093]

[0094] Among them, the visualization images of the risk prediction results of macrosomia for pregnant women 5 and 9 are as follows: Figure 5 、 6 As shown, combined with the results shown in Table 1 above, it can be seen that the method for predicting macrosomia in women with gestational diabetes mellitus (GDM) using the model obtained in Example 1 is highly targeted and has high prediction accuracy; and because the predicted risk is presented visually, the influence of each risk factor can be intuitively displayed, which is convenient for obstetricians in clinical application and communication with patients, and has great promotion value.

[0095] Example 4: Device, computer-readable storage medium, and computer program product capable of implementing a method for predicting macrosomia in women with gestational diabetes mellitus (GDM)

[0096] Figure 7 The present invention provides a schematic diagram of the physical structure of an electronic device, which may include a processor 610, a memory 620, a communication interface 630, and a communication bus 640. The processor 610, the memory 620, and the communication interface 630 communicate with each other via the communication bus 640. The processor 610 may invoke logic instructions stored in the memory 620 to execute a method for predicting fetal macrosomia in women with gestational diabetes mellitus (GDM).

[0097] In addition, the logical instructions in the above-mentioned memory 620 can be implemented in the form of software functional units and when sold or used as an independent product, can be stored in a computer-readable storage medium, including: USB flash drive, mobile hard disk, read-only memory, random access memory, magnetic disk or optical disk, etc., various media that can store program codes.

[0098] On the other hand, the present invention also provides a computer program product, which includes a computer program. When executed by a processor, the computer can execute the above-mentioned method for predicting the birth of a large baby by women with gestational diabetes mellitus (GDM).

Claims

1. A method for predicting the risk of macrosomia in women with gestational diabetes mellitus, characterized by the following steps: include: (1) Collect data, including mid-pregnancy morphological indicators and blood biochemical index examination results of women with gestational diabetes; (2) Process the data to obtain a data set; (3) Analyzing the data set using a prediction model to predict the risk of macrosomia in women with gestational diabetes, wherein the prediction model is: P=1÷{1+exp[-(-16.0208-0.073513×X1-0.064942×X2-0.724375×X3+ 0.007517×X4+0.088095×X5+0.016533×X6)]}; Where P is the probability of giving birth to a macrosomia in women with gestational diabetes, X1 is the free thyroxine value, X2 is the serum total protein value, X3 is the ratio of alanine aminotransferase to aspartate aminotransferase, X4 is the lactate dehydrogenase value, X5 is the height value, and X6 is the product of the triglyceride-glucose index and body mass index.

2. The method for predicting the risk of macrosomia in women with gestational diabetes mellitus according to claim 1, wherein: When the obtained probability is less than 0.1, it is predicted that women with gestational diabetes have a low risk of delivering a macrosomia; When the obtained probability is 0.1-0.6, it is predicted that women with gestational diabetes are at medium risk of giving birth to a macrosomia; When the obtained probability is greater than 0.6, it is predicted that women with gestational diabetes are at high risk of giving birth to a macrosomia.

3. The method for predicting the risk of macrosomia in women with gestational diabetes mellitus according to claim 1, wherein: The second trimester described in step 1 is 20-24 weeks of pregnancy.

4. The method for predicting the risk of macrosomia in women with gestational diabetes mellitus according to claim 1, wherein: The morphological indicators in step 1 include height and body mass index, and the biochemical indicators include free thyroxine, serum total protein, alanine aminotransferase to aspartate aminotransferase ratio, lactate dehydrogenase, and triglyceride-glucose index.

5. The method for predicting the risk of macrosomia in women with gestational diabetes mellitus according to claim 1, wherein: The steps of processing data in step 2 include: (21) The collected data were converted to units: free thyroxine units were converted to pmol / L, serum total protein units were converted to g / L, lactate dehydrogenase units were converted to U / L, and height units were converted to cm; (22) Remove the data units and keep only the numerical values.

6. The method for predicting the risk of macrosomia in women with gestational diabetes mellitus according to claim 1, wherein: After step 3, the data set is scored and the total score is calculated. Combined with the probability obtained by the prediction model, a visualization image is generated.

7. The method for predicting the risk of macrosomia in women with gestational diabetes mellitus according to claim 6, wherein: In the data set, the scoring formula for each data item includes: (a) Free thyroxine value scoring formula: P FT4 =-2×X1+52; (b) Score formula for serum total protein value: P TP =-1.7769×X2+142.27; (c) Scoring formula for the ratio of alanine aminotransferase to aspartate aminotransferase: P ALT / AST =-20×X3+60; (d) Lactate dehydrogenase value scoring formula: P LDH =0.207×X4-10.417; (e) Height score formula: P h =2.4333×X5-340.89; (f) The scoring formula for the product of triglyceride-glucose index and body mass index is: P TyG-BMI =0.4563×X6-64.073; Among them, P FT4 is the score of free thyroxine value, X1 is the free thyroxine value; P TP is the score of serum total protein value, X2 is the serum total protein value; P ALT / AST is the score of the ratio of alanine aminotransferase to aspartate aminotransferase, X3 is the ratio of alanine aminotransferase to aspartate aminotransferase; P LDH is the score of lactate dehydrogenase value, X4 is the lactate dehydrogenase value; P h is the score assigned to the height value, X5 is the height value; P TyG-BMI It is the score of the product of triglyceride-glucose index and body mass index, X6 is the product of triglyceride-glucose index and body mass index. If the calculated score is less than 0, 0 points will be given; if the calculated score is greater than 100, 100 points will be given.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory, wherein: When the processor executes the computer program, the prediction method according to any one of claims 1 to 7 is 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 prediction method according to any one of claims 1 to 7 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the prediction method according to any one of claims 1 to 7 is implemented.

Images