Chinese Medical E-ournals Database
Adv Search
中华妇幼临床医学杂志(电子版)

Chinese Journal of Obstetrics & Gynecology and Pediatrics(Electronic Edition) ›› 2024, Vol. 20 ›› Issue (01): 105 -113. doi: 10.3877/cma.j.issn.1673-5250.2024.01.014

Original Article

Analysis of early pregnancy-related influencing factors of gestational diabetes mellitus, and clinical value of building gestational diabetes mellitus prediction model based on four machine learning algorithms of glycolipid-related biochemical indexes and demographic information of pregnant women in early pregnancy

Li Li1, Mei Ma2, Xinxin Huang3, Danlin Yang1, Mian Pan1,†()   

  1. 1. Department of Obstetrics, Fujian Maternity and Child Health Hospital, College of Clinical Medicine for Obstetrics & Gynecology and Pediatrics, Fujian Medical University, Fuzhou 350001, Fujian Province, China
    2. Department of Laboratory Medicine, Fujian Maternity and Child Health Hospital, College of Clinical Medicine for Obstetrics & Gynecology and Pediatrics, Fujian Medical University, Fuzhou 350001, Fujian Province, China
    3. Department of Healthcare, Fujian Maternity and Child Health Hospital, College of Clinical Medicine for Obstetrics & Gynecology and Pediatrics, Fujian Medical University, Fuzhou 350001, Fujian Province, China
  • Received:2023-11-11 Revised:2024-01-06 Published:2024-02-01
  • Corresponding author: Mian Pan
  • Supported by:
    Natural Science Foundation of Department of Science and Technology in Fujian Province(2021J01406)
HTML ( ) PDF ( )
Objective

To investigate the early pregnancy-related influencing factors of gestational diabetes mellitus (GDM), as well as the clinical value of building GDM prediction model based on the glycolipids-related biochemical indexes in early pregnancy and demographic information using four machine learning algorithms.

Methods

A total of 6 257 pregnant women with gestational age of 10 to 13+ 6 gestational weeks who had their first prenatal examinations in Fujian Maternity and Child Health Hospital from December 2021 to December 2022 were selected for the study. The pregnant women were categorized into the GDM group (n=1 499, GDM pregnant women) and the non-GDM group (n=4 758, non-GDM pregnant women) according to whether or not they were diagnosed with GDM at 24 to 27+ 6 gestational weeks by retrospective analysis. Early pregnancy-related influencing factors on the development of GDM in pregnant women were analyzed using multivariate unconditional logistic regression analysis. Based on the biochemical indexes related to glycolipids in early pregnancy and demographic information in pregnant women (8 variables), four machine learning algorithms, namely, decision tree (DT), logistic regression (LR), random forest (RF), and extreme gradient boosting (XGB) were used to build GDM prediction models, and ten-fold cross-validation was used to assess the performance of each model, and area under curve (AUC) of the receiver operating characteristic (ROC) curve among the GDM prediction models constructed by the four algorithms were compared. The study was approved by the Ethics Committee of Fujian Maternity and Child Health Hospital (Approval No. 2021KRD018). All pregnant women had signed the informed consent forms for clinical research.

Results

①The results of multivariate unconditional logistic regression analysis showed that pregnant women with advanced age (delivery age≥35 years) (OR=1.95, 95%CI: 1.70-2.24, P<0.001), with pre-pregnancy body mass index (BMI) ≥ 18.5-24.0 kg/m2 (OR=1.32, 95%CI: 1.11-1.58, P=0.002), pre-pregnancy BMI ≥ 24.0-28.0 kg/m2 (OR=2.17, 95%CI: 1.73-2.73, P<0.001), pre-pregnancy BMI ≥28.0 kg/m2 (OR=2.53, 95%CI: 1.70-3.78, P<0.001), elevated serum apolipoprotein (Apo) B levels during early pregnancy (OR=3.06, 95%CI: 2.14-4.37, P<0.001), and increased serum fasting glucose (FPG) concentration in early pregnancy (OR=2.08, 95%CI: 1.79-2.41, P<0.001) were all independent early pregnancy-related risk factors for the development of GDM in pregnant women. ②According to the magnitude of the eigenvalue in the classification results of the 4 classifiers, the results of GDM prediction model constructed using 8 variables of maternal age, degree of education and pre-pregnancy BMI, serum levels of total cholesterol (TC), triglyceride (TG), ApoA1, ApoB, and FPG during early pregnancy showed that the AUC of the GDM prediction models built by the 4 algorithms, namely, DT, LR, RF, and XGB, were 0.645 (95%CI: 0.591-0.698), 0.699 (95%CI: 0.641-0.749), 0.672 (95%CI: 0.621-0.772), and 0.597 (95%CI: 0.553-0.663), respectively, and the AUC of the LR algorithm was greater than that of the XGB algorithm, and the difference was statistically significant (Z=2.38, P=0.017), and there was no significant difference in AUC of pairwise comparison of the rest of the algorithms (P>0.05). ③The ten-fold cross-validation results showed that the average AUC of the GDM prediction models constructed by the four algorithms, DT, LR, RF, and XGB, were 0.586±0.025, 0.661±0.020, 0.632±0.023, and 0.576±0.019, respectively.

Conclusions

Based on the biochemical indexes related to glycolipids in early pregnancy and demographic data, GDM prediction models constructed with LR and RF algorithms, which has a certain predictive value for GDM, and helps to screen the high-risk group of GDM at early stage, and to provide clinical interventions when necessary to reduce GDM-related adverse pregnancy outcome of mother and fetus.

Key words: Pregnancy trimester, first, Diabetes, gestational, Blood glucose, Blood lipids, Machine learning, Algorithms, Logistic models, Decision trees, Random forest
表1 2组早孕期孕妇一般临床资料及糖脂相关生化指标比较
组别 例数 年龄[例数(%)] 受教育程度[例数(%)]
≥35岁 <35岁 初中/中专及以下 高中/大专 大学本科 研究生及以上
GDM组 1 499 543(36.2) 956(63.8) 273(18.3) 484(32.4) 637(42.6) 102(6.8)
非GDM组 4 758 943(19.8) 3 815(80.2) 752(15.8) 1 486(31.3) 2 130(44.9) 378(8.0)
统计量   χ2＝168.49 χ2＝7.69
P值   <0.001 0.053
组别 例数 孕前BMI[例数(%)] 血清TC浓度[例数(%)]
<18.5 kg/m2 ≥18.5～24.0 kg/m2 ≥24.0～28.0 kg/m2 ≥28.0 kg/m2 <5.18 mmol/L ≥5.18 mmol/L
GDM组 1 499 182(12.1) 990(66.0) 272(18.2) 55(3.7) 1 252(83.5) 247(16.5)
非GDM组 4 758 954(20.1) 3 278(68.9) 453(9.5) 73(1.5) 4 134(86.9) 624(13.1)
统计量   χ2＝139.20 χ2＝10.48
P值   <0.001 0.001
组别 例数 血清TG浓度[例数(%)] 血清HDL-C浓度[例数(%)] 血清LDL-C浓度[例数(%)]
<1.7 mmol/L ≥1.7 mmol/L <1.04 mmol/L ≥1.04 mmol/L <3.37 mmol/L ≥3.37 mmol/L
GDM组 1 499 1 255(83.7) 244(16.3) 10(0.7) 1 489(99.3) 1 460(97.4) 39(2.6)
非GDM组 4 758 4 052(85.2) 706(14.8) 15(0.3) 4 743(99.7) 4 671(98.2) 87(1.8)
统计量   χ2＝1.72 χ2＝2.72 χ2＝3.07
P值   0.189 0.099 0.080
组别 例数 血清ApoA1水平(g/L，±s) 血清ApoB水平(g/L，±s) 血清FPG浓度(mmol/L，±s) ApoB/A1[M(Q1，Q3)]
GDM组 1 499 1.57±0.23 0.76±0.16 4.7±0.5 0.481(0.411，0.561)
非GDM组 4 758 1.55±0.23 0.72±0.17 4.6±0.4 0.458(0.393，0.535)
统计量   t＝2.79 t＝8.50 t＝12.65 Z＝6.46
P值   0.005 <0.001 <0.001 <0.010
表2 影响孕妇发生GDM的多因素非条件logistic回归分析
影响因素 B SE Wald值 P值 OR值 OR值95%CI
孕妇年龄≥35岁(vs <35岁) 0.67 0.07 97.81 <0.001 1.95 1.70～2.24
孕前BMI(vs <18.5 kg/m2)            
≥18.5～24.0 kg/m2 0.28 0.09 9.49 0.002 1.32 1.11～1.58
≥24.0～28.0 kg/m2 0.78 0.12 44.57 <0.001 2.17 1.73～2.73
≥28.0 kg/m2 0.93 0.20 20.63 <0.001 2.53 1.70～3.78
早孕期血清ApoB(连续变量) 1.12 0.18 37.50 <0.001 3.06 2.14～4.37
早孕期血清FPG(连续变量) 0.73 0.08 94.13 <0.001 2.08 1.79～2.41
常数项 －5.20 0.39 176.48 <0.001 0.01 —
图1 4种机器学习算法构建GDM预测模型的ROC-AUC比较注：GDM为妊娠期糖尿病，ROC曲线为受试者工作特征曲线，AUC为曲线下面积。DT为决策树，LR为逻辑回归，RF为随机森林，XGB为极致梯度提升
表3 4种机器学习算法构建GDM预测模型的4个性能评估指标比较
机器学习算法 准确率(%) 准确率95%CI(%) 精确率(%) 精确率95%CI(%)
DT 78.8 78.1～79.6 65.5 64.4～66.6
LR 77.2 76.9～77.6 76.5 76.1～77.0
RF 77.3 77.1～77.5 76.6 76.3～76.9
XGB 78.0 77.5～78.5 72.6 71.7～73.4
机器学习算法 召回率(%) 召回率95%CI(%) AUC AUC 95%CI
DT 74.7 73.2～76.2 0.645 0.591～0.698
LR 98.0 97.7～98.4 0.699 0.641～0.749
RF 98.0 97.6～98.4 0.672 0.621～0.772
XGB 89.1 88.0～90.2 0.597 0.553～0.663
图2 在训练集中预测GDM患者的4种机器学习算法的十折交叉验证AUC结果折线图注：GDM为妊娠期糖尿病，AUC为曲线下面积。DT为决策树，LR为逻辑回归，RF为随机森林，XGB为极致梯度提升
[1]
Kinnunen J, Nikkinen H, Keikkala E, et al. Gestational diabetes is associated with the risk of offspring′s congenital anomalies: a register-based cohort study[J]. BMC Pregnancy Childbirth, 2023, 23(1): 708. DOI: 10.1186/s12884-023-05996-6.
[2]
马雨鸿，马华姝，乔宗旭，等. 妊娠期糖尿病危险因素的结构方程模型分析[J]. 中国卫生统计，2022, 39(3): 446-449. DOI: 10.3969/j.issn.1002-3674.2022.03.029.
[3]
Akash MSH, Noureen S, Rehman K, et al. Investigating the biochemical association of gestational diabetes mellitus with dyslipidemia and hemoglobin[J]. Front Med (Lausanne), 2023, 10: 1242939. DOI: 10.3389/fmed.2023.1242939.
[4]
Tong JN, Chen YX, Guan XN, et al. Association between the cut-off value of the first trimester fasting plasma glucose level and gestational diabetes mellitus: a retrospective study from Southern China[J]. BMC Pregnancy Childbirth, 2022, 22(1): 540. DOI: 10.1186/s12884-022-04874-x.
[5]
Shen L, Wang D, Huang Y, et al. Longitudinal trends in lipid profiles during pregnancy: association with gestational diabetes mellitus and longitudinal trends in insulin indices[J]. Front Endocrinol (Lausanne), 2022, 13: 1080633. DOI: 10.3389/fendo.2022.1080633.
[6]
Kouiti M, Hernández-Muñiz C, Youlyouz-Marfak I, et al. Preventing gestational diabetes mellitus by improving healthy diet and/or physical activity during pregnancy: an umbrella review[J]. Nutrients, 2022, 14(10): 2066. DOI: 10.3390/nu14102066.
[7]
Shepherd E, Gomersall JC, Tieu J, et al. Combined diet and exercise interventions for preventing gestational diabetes mellitus[J]. Cochrane Database Syst Rev, 2017, 11(11): CD010443. DOI: 10.1002/14651858.CD010443.pub3.
[8]
Abell SK, Shorakae S, Boyle JA, et al. Role of serum biomarkers to optimise a validated clinical risk prediction tool for gestational diabetes[J]. Aust N Z J Obstet Gynaecol, 2019, 59(2): 251-257. DOI: 10.1111/ajo.12833.
[9]
中华医学会妇产科学分会产科学组，中华医学会围产医学分会，中国妇幼保健协会妊娠合并糖尿病专业委员会. 妊娠期高血糖诊治指南(2022)[第一部分][J]. 中华妇产科杂志，2022, 57(1): 3-12. DOI: 10.3760/cma.j.cn112141-20210917-00528.
[10]
杨剑锋，乔佩蕊，李永梅，等. 机器学习分类问题及算法研究综述[J]. 统计与决策，2019, 35(6): 36-40. DOI: 10.13546/j.cnki.tjyjc.2019.06.008.
[11]
Greener JG, Kandathil SM, Moffat L, et al. A guide to machine learning for biologists[J]. Nat Rev Mol Cell Biol, 2022, 23(1): 40-55. DOI: 10.1038/s41580-021-00407-0.
[12]
韩娜，刘珏，金楚瑶，等. 2013—2017年北京市通州区34 637例孕妇妊娠期糖尿病流行情况及其影响因素研究[J]. 中华疾病控制杂志，2019, 23(2): 156-161. DOI: 10.16462/j.cnki.zhjbkz.2019.02.007.
[13]
Zhu H, Zhao Z, Xu J, et al. The prevalence of gestational diabetes mellitus before and after the implementation of the universal two-child policy in China[J]. Front Endocrinol (Lausanne), 2022, 13: 960877. DOI: 10.3389/fendo.2022.960877.
[14]
Wu Q, Chen Y, Zhou M, et al. An early prediction model for gestational diabetes mellitus based on genetic variants and clinical characteristics in China[J]. Diabetol Metab Syndr, 2022, 14(1): 15. DOI: 10.1186/s13098-022-00788-y.
[15]
Harrison CL, Lombard CB, East C, et al. Risk stratification in early pregnancy for women at increased risk of gestational diabetes[J]. Diabetes Res Clin Pract, 2015, 107(1): 61-68. DOI: 10.1016/j.diabres.2014.09.006.
[16]
Zheng Y, Hou W, Xiao J, et al. Application value of predictive model based on maternal coagulation function and glycolipid metabolism indicators in early diagnosis of gestational diabetes mellitus[J]. Front Public Health, 2022, 10: 850191. DOI: 10.3389/fpubh.2022.850191.
[17]
Snyder BM, Baer RJ, Oltman SP, et al. Early pregnancy prediction of gestational diabetes mellitus risk using prenatal screening biomarkers in nulliparous women[J]. Diabetes Res Clin Pract, 2020, 163: 108139. DOI: 10.1016/j.diabres.2020.108139.
[18]
Xia J, Song Y, Rawal S, et al. Vitamin D status during pregnancy and the risk of gestational diabetes mellitus: a longitudinal study in a multiracial cohort[J]. Diabetes Obes Metab, 2019, 21(8): 1895-1905. DOI: 10.1111/dom.13748.
[19]
Corcoran SM, Achamallah N, Loughlin JO, et al. First trimester serum biomarkers to predict gestational diabetes in a high-risk cohort: striving for clinically useful thresholds[J]. Eur J Obstet Gynecol Reprod Biol, 2018, 222: 7-12. DOI: 10.1016/j.ejogrb.2017.12.051.
[20]
Thériault S, Forest JC, Massé J, et al. Validation of early risk-prediction models for gestational diabetes based on clinical characteristics[J]. Diabetes Res Clin Pract, 2014, 103(3): 419-425. DOI: 10.1016/j.diabres.2013.12.009.
[21]
Naylor CD, Sermer M, Chen E, et al. Selective screening for gestational diabetes mellitus. Toronto Trihospital Gestational Diabetes Project Investigators[J]. N Engl J Med, 1997, 337(22): 1591-1596. DOI: 10.1056/NEJM199711273372204.
[22]
Caliskan E, Kayikcioglu F, Oztürk N, et al. A population-based risk factor scoring will decrease unnecessary testing for the diagnosis of gestational diabetes mellitus[J]. Acta Obstet Gynecol Scand, 2004, 83(6): 524-530. DOI: 10.1111/j.0001-6349.2004.00389.x.
[23]
van Leeuwen M, Opmeer BC, Zweers EJK, et al. Estimating the risk of gestational diabetes mellitus: a clinical prediction model based on patient characteristics and medical history[J]. BJOG, 2010, 117(1): 69-75. DOI: 10.1111/j.1471-0528.2009.02425.x.
[24]
Teede HJ, Harrison CL, Teh WT, et al. Gestational diabetes: development of an early risk prediction tool to facilitate opportunities for prevention[J]. Aust N Z J Obstet Gynaecol, 2011, 51(6): 499-504. DOI: 10.1111/j.1479-828x.2011.01356.x.
[25]
Sweeting AN, Wong J, Appelblom H, et al. A novel early pregnancy risk prediction model for gestational diabetes mellitus[J]. Fetal Diagn Ther, 2019, 45(2): 76-84. DOI: 10.1159/000486853.
[26]
Liu R, Zhan Y, Liu X, et al. Stacking ensemble method for gestational diabetes mellitus prediction in Chinese pregnant women: a prospective cohort study[J]. J Healthc Eng, 2022, 2022: 8948082. DOI: 10.1155/2022/8948082.
[27]
Zhang Z, Yang L, Han W, et al. Machine learning prediction models for gestational diabetes mellitus: Meta-analysis[J]. J Med Internet Res, 2022, 24(3): e26634. DOI: 10.2196/26634.
[28]
Liu H, Li J, Leng J, et al. Machine learning risk score for prediction of gestational diabetes in early pregnancy in Tianjin, China[J]. Diabetes Metab Res Rev, 2021, 37(5): e3397. DOI: 10.1002/dmrr.3397.
[1] Ye Luo, Mengling Hu, Xiaofan Huang, Jinpeng Lin, Zhuman Li, Shaobai Wang. Support vector machine for classification of osteoarthritis and ligamentous injuries in knee[J]. Chinese Journal of Joint Surgery(Electronic Edition), 2024, 18(02): 351-358.
[2] Beibei Wang, Qixiu Dong, Hongyan Xi, Qingyun Yu, Lijun Zhang, Guang Shi. Analysis of influencing factors of medical abortion failure of pregnant women in early pregnancy and construction of related prediction model and its predictive value for medical abortion success[J]. Chinese Journal of Obstetrics & Gynecology and Pediatrics(Electronic Edition), 2023, 19(05): 588-594.
[3] Chao Xue, Ye Zhang, Ying Zhao, Jiancheng Han, Xiaoyan Gu, Lin Sun, Xiaowei Liu, Wei Song, Yihua He. Ultrasonographic characteristics and prognosis of fetal congenital absent pulmonary valve syndrome[J]. Chinese Journal of Obstetrics & Gynecology and Pediatrics(Electronic Edition), 2023, 19(04): 410-418.
[4] Zhengyun Hu, Jianwei Shi, Jianwei Shen, Bing Wang, Chunmiao Jiang, Chong Liu. Identification of hub genes associated with bronchopulmonary dysplasia in preterm infants based on machine learning[J]. Chinese Journal of Obstetrics & Gynecology and Pediatrics(Electronic Edition), 2023, 19(04): 446-454.
[5] Jian Mao. Pathogen distribution of community acquired pneumonia in elderly patients with diabetes and the predictive value of fasting blood glucose and glycosylated hemoglobin[J]. Chinese Journal of Experimental and Clinical Infectious Diseases(Electronic Edition), 2023, 17(06): 408-415.
[6] Shuaihua Fan, Wei Guo, Jun Guo. Current situation and prospect of application of decision tree algorithm based on machine learning in prognosis prediction of bloodstream infection[J]. Chinese Journal of Experimental and Clinical Infectious Diseases(Electronic Edition), 2023, 17(05): 289-293.
[7] Yicheng Huang, Haiqi Liang, Qihuan He, Faye Wei, Shubo Yang, Shuting Tan, Gaoqiang Zhai, Jiwen Cheng. Machine learning models assess the roles of RAS subfamily genes in immunotherapy based on bladder cancer[J]. Chinese Journal of Endourology(Electronic Edition), 2024, 18(02): 131-140.
[8] Mengshi Lu, Wei Liu, Jiawei Ma, Dandan Ji, Xuan Jia, Xinping Zhan, Liang Luo. Application progress of artificial intelligence in acute respiratory distress syndrome management[J]. Chinese Journal of Critical Care & Intensive Care Medicine(Electronic Edition), 2024, 10(01): 66-71.
[9] Xianjin Chen, Qinqin Wu, Changchun He, Qinghua Zhang. Research on constructing prognosis prediction model of intracerebral hemorrhage using multimodal medical data and machine learning[J]. Chinese Journal of Brain Diseases and Rehabilitation(Electronic Edition), 2023, 13(04): 193-198.
[10] Han Zhu, Shichao Zhuo, Di Wu, Yanan Zhu, Jiaxin Han. Correlation between preoperative plasma fibrinogen, blood lipid level, MMR expression and pathological characteristics and prognosis of colorectal cancer[J]. Chinese Journal of Digestion and Medical Imageology(Electronic Edition), 2024, 14(02): 141-145.
[11] Qinqin Zhu, Juanjuan Ci, Lu Cui, Hairong Xu, Yuxin Li, Yanbo Ding. Value of coagulation function index, blood lipid level, C-reactive protein and neutrophil-lymphocyte ratio in the activity evaluation and clinical diagnosis of Crohn's disease[J]. Chinese Journal of Digestion and Medical Imageology(Electronic Edition), 2024, 14(01): 35-40.
[12] Yan Huang, Xiaowei Liu, Chunling Yang, Ye Lan. Clinical characteristics and correlation between blood glucose metabolism and severity of acute pancreatitis with diabetes[J]. Chinese Journal of Digestion and Medical Imageology(Electronic Edition), 2023, 13(06): 439-442.
[13] Fuxiao Duan, Xinyu Wang, Shuang Sun, Zhiyu Yu, Cheng Zhang. Establishment and evaluation of a predictive model for perineural invasion in colorectal cancer patients[J]. Chinese Journal of Clinicians(Electronic Edition), 2023, 17(11): 1154-1162.
[14] Zhentian Guo, Zongming Zhang, Yue Zhao, Limin Liu, Chong Zhang, Zhuo Liu, Hui Qi, Kun Tian. Machine learning algorithms for predicting postoperative hospital stay in elderly patients with acute cholecystitis[J]. Chinese Journal of Clinicians(Electronic Edition), 2023, 17(09): 955-961.
[15] Junjie Wang, Xiaoliang Yin, Erteng Liu, Jun Lu, Peng Qi, Shen Hu, Ximeng Yang, Kunpeng Chen, Dong Zhang, Daming Wang. Predicting successful endovascular recanalization for non-acute occlusion of internal carotid artery: potential value of machine learning[J]. Chinese Journal of Cerebrovascular Diseases(Electronic Edition), 2023, 17(05): 464-470.
Viewed
Full text


Abstract

Copyright © Chinese Medical Association, Chinese Medical Multimedia Press  京ICP备14006079号-1  网络出版服务许可证：(署)网出证(京)字第075号