中山大学附属第七医院超声科,广东 深圳 518107
陈心雨,第一作者,研究方向:产前超声诊断,E-mail: chenxy796@mail2.sysu.edu.cn
纸质出版日期:2024-07-20,
收稿日期:2024-05-07,
录用日期:2024-06-15
扫 描 看 全 文
陈心雨,朱云晓.结合临床与超声影像多模态数据构建小于胎龄儿预测模型[J].中山大学学报(医学科学版),2024,45(04):637-648.
CHEN Xinyu,ZHU Yunxiao.Constructing Prediction Models for Small for Gestational Age Based on Multimodal Clinical and Ultrasonographic Data[J].Journal of Sun Yat-sen University(Medical Sciences),2024,45(04):637-648.
陈心雨,朱云晓.结合临床与超声影像多模态数据构建小于胎龄儿预测模型[J].中山大学学报(医学科学版),2024,45(04):637-648. DOI: 10.13471/j.cnki.j.sun.yat-sen.univ(med.sci).20240617.001.
CHEN Xinyu,ZHU Yunxiao.Constructing Prediction Models for Small for Gestational Age Based on Multimodal Clinical and Ultrasonographic Data[J].Journal of Sun Yat-sen University(Medical Sciences),2024,45(04):637-648. DOI: 10.13471/j.cnki.j.sun.yat-sen.univ(med.sci).20240617.001.
目的
2
探索结合妊娠早期和妊娠中期的临床数据与超声影像多模态数据对小于胎龄儿(SGA)的预测价值,构建和内部验证基于多种机器学习算法的SGA 预测模型。
方法
2
本研究回顾性纳入单胎妊娠孕妇1 307例,根据INTERGROWTH-21st胎儿生长标准诊断SGA并采集包括临床一般资料、生化检验数据及产前超声筛查数据的多模态数据。轻度梯度增强算法(XGBoost)用于计算变量重要性,七种机器学习算法用于预测模型的构建与内部验证,受试者工作特性曲线下面积(AUC)作为衡量预测效能的主要指标,结合10%假阳性率下的灵敏度进行模型间比较。
结果
2
基于临床一般资料与生化检验数据构建的最优预测模型AUC为0.70 , 95%CI (0.609, 0.791),灵敏度为0.38 , 95%CI (0.236, 0.519)。基于产前超声筛查数据构建的最优预测模型效能优于前者,其AUC为0.77 , 95%CI (0.687, 0.858),灵敏度为0.62 , 95%CI (0.457, 0.743)。两个数据集合并组成多模态临床数据集,其最佳预测模型效能进一步提升,AUC为0.91 , 95%CI (0.851, 0.972),灵敏度为0.88 , 95%CI (0.745, 0.947),且模型校准显示拟合优度佳。
结论
2
本研究采用机器学习算法充分地探索了妊娠早期结合妊娠中期的不同类型临床数据对SGA的预测价值,证明了多模态临床数据用于SGA的筛查的绝对优势,为孕妇个体化管理提供准确且有效的参考依据。
Objective
2
To explore the predictive value of multimodal clinical and ultrasonographic data in first- and second-trimester for small for gestational age (SGA), so as to build and internally validate SGA prediction models based on multiple machine learning algorithms.
Methods
2
This retrospective study enrolled 1,307 pregnant women with singleton pregnancies, diagnosed SGA according to INTERGROWTH-21st fetal growth criteria, and collected multimodal clinical data including general clinical information, biochemical test data, and prenatal ultrasound screening data. Extreme gradient boosting (XGBoost) algorithm was used to calculate the importance of variables. Seven machine learning algorithms were used to construct and internally verify the prediction models. The area under the receiver operating characteristic curve (AUC) was used as the main indicator to measure the prediction performance and used to compare predictive performance between models with the sensitivity at a 10% false positive rate.
Results
2
The optimal prediction model built based on general clinical information and biochemical test data had an AUC of 0.70, 95%CI (0.609, 0.791) and a sensitivity of 0.38, 95%CI (0.236, 0.519). The optimal prediction model based on prenatal ultrasound screening data was better than the former, with an AUC of 0.77, 95%CI (0.687, 0.858) and a sensitivity of 0.62, 95%CI (0.457, 0.743). The two data sets were combined to form the multimodal clinical dataset, and the performance of the best prediction model was further improved with an AUC of 0.91, 95%CI (0.851, 0.972) and a sensitivity of 0.88, 95%CI (0.745, 0.947), and the model calibration showed good goodness of fit.
Conclusion
2
By using machine learning algorithms to fully explore the predictive value of different types of clinical data for SGA in first- and second-trimester, this study proves the absolute advantages of multimodal clinical data for SGA screening, and provides an accurate and effective reference for personalized management of pregnant women.
小于胎龄儿预测模型机器学习妊娠早期妊娠中期
small for gestational ageprediction modelmachine learningfirst-trimestersecond-trimester
Lees CC, Romero R, Stampalija T, et al. Clinical Opinion: The diagnosis and management of suspected fetal growth restriction: an evidence-based approach[J]. Am J Obstet Gynecol, 2022, 226(3): 366-378.
段涛, 杨慧霞, 胡娅莉, 等. 胎儿生长受限专家共识(2019版)[J]. 中国产前诊断杂志(电子版), 2019, 11(4): 78-98.
Duan T, Yang HX, Hu YL, et al. Expert consensus on fetal growth restriction (2019 edition)[J]. Chin J Prenat Diagn (Electronic Version), 2019, 11(4): 78-98.
Francis A, Hugh O, Gardosi J. Customized vs INTERGROWTH-21st standards for the assessment of birthweight and stillbirth risk at term[J]. Am J Obstet Gynecol, 2018, 218(2S): S692-S699.
Katz J, Lee AC, Kozuki N, et al. Mortality risk in preterm and small-for-gestational-age infants in low-income and middle-income countries: a pooled country analysis[J]. Lancet (London, England), 2013, 382(9890): 417-425.
Adanikin A, Lawlor DA, Pell JP, et al. Association of birthweight centiles and early childhood development of singleton infants born from 37 weeks of gestation in Scotland: A population-based cohort study[J]. PLoS Medicine, 2022, 19(10): e1004108.
张妙莲, 李婳婳, 潘焯仪, 等. 限制性胎盘嵌合体的妊娠与分娩结局[J]. 广东医学, 2023, 44(10): 1297-1303.
Zhang ML, Li HH, Pan ZY, et al. Pregnancy and delivery outcomes in restrictive placental mosaicism[J]. Guang Dong Med J, 2023, 44(10): 1297-1303.
Maguolo A, Olivieri F, Zusi C, et al. The risk of metabolic derangements is higher in children and adolescents with overweight or obesity born small for gestational age[J]. Nutr Metab Cardiovasc Dis, 2021, 31(6): 1903-1910.
Cauzzo C, Chiavaroli V, Di Valerio S, et al. Birth size, growth trajectory and later cardio-metabolic risk[J]. Front Endocrinol (Lausanne), 2023, 14: 1187261.
Xiao H, Chen H, Chen X, et al. Comprehensive assessment of the genetic characteristics of small for gestational age newborns in NICU: from dignosis of genetic disorders to prediction of prognosis[J]. Genome Med, 2023, 15(1): 112.
吴薇, 罗小平. 小于胎龄儿的终生管理[J]. 重庆医科大学学报, 2022, 47(3): 256-258.
Wu W, Luo XP. Lifetime management of small for gestational age[J]. J Chongqing Med Univ, 2022, 47(3): 256-258.
Whelan R, Schaeffer L, Olson I, et al. Measurement of symphysis fundal height for gestational age estimation in low-to-middle-income countries: a systematic review and meta-analysis[J]. PloS One, 2022, 17(8): e0272718.
Ruchob R, Rutherford JN, Bell AF. A systematic review of placental biomarkers predicting small-for-gestational-age neonates[J]. Biol Res Nurs, 2018, 20(3): 272-283.
Leite DFB, Morillon AC, Melo Júnior EF, et al. Examining the predictive accuracy of metabolomics for small-for-gestational-age babies: a systematic review[J]. BMJ Open, 2019, 9(8): e031238.
马骏楠, 谢敏, 孙倩, 等. 胎儿生长受限分类与管理研究进展[J]. 实用妇产科杂志, 2021, 37(8): 595-598.
Ma JN, Xie M, Sun Q, et al. Research progress on the classification and management of fetal growth restriction [J]. J Pract Obstet Gynecol, 2021, 37(8): 595-598.
Crovetto F, Crispi F, Casas R, et al. Effects of Mediterranean diet or mindfulness-based stress reduction on prevention of small-for-gestational age birth weights in newborns born to at-risk pregnant individuals: The IMPACT BCN randomized clinical trial[J]. JAMA, 2021, 326(21): 2150-2160.
Stirnemann J, Villar J, Salomon LJ, et al. International estimated fetal weight standards of the INTERGROWTH-21st Project[J]. Ultrasound Obstet Gynecol, 2017, 49(4): 478-486.
Zhao J, Yuan Y, Tao J, et al. Which fetal growth charts should be used? A retrospective observational study in China[J]. Chin Med J (Engl), 2022, 135(16): 1969-1977.
Poon LCY, Zymeri NA, Zamprakou A, et al. Protocol for measurement of mean arterial pressure at 11-13 weeks' gestation[J]. Fetal Diagnosis and Therapy, 2012, 31(1): 42-48.
Bhide A, Acharya G, Baschat A, et al. ISUOG Practice Guidelines (updated): use of Doppler velocimetry in obstetrics[J]. Ultrasound In Obstet Gynecol, 2021, 58(2): 331-339.
中华医学会超声医学分会妇产超声学组, 国家卫生健康委妇幼司全国产前诊断专家组医学影像组. 超声产前筛查指南[J]. 中华超声影像学杂志, 2022, (1): 1-12.
Obstetrics and Gynecology Group of the Ultrasound Medicine Branch of the Chinese Medical Association, Medical Imaging Group of the National Prenatal Diagnosis Expert Group of the Department of Maternity and Infants of the National Health Commission. Ultrasound prenatal screening guidelines [J]. Chin J Ultrasonogr, 2022, (1): 1-12.
Mcdonald GC. Ridge regression[J]. Wiley Interdisciplinary Reviews: computational statistics, 2009, 1(1): 93-100.
Lavalley MP. Logistic regression[J]. Circulation, 2008, 117(18): 2395-2399.
JaKSuykens, Vandewalle J. Least squares support vector machine classifiers[J]. Neural Processing Letters, 1999, 9(3): 293-300.
Breiman L. Random forests[J]. Machine Learning, 2001, 45(1): 5-32.
Chen T, Guestrin C. XGBoost: A scalable tree boosting system, F, 2016 [C]. Association for Computing Machinery.
Gardner MW, Dorling SR. Artificial neural networks (the multilayer perceptron) - a review of applications in the atmospheric sciences[Z]. BritainGreat; PERGAMON PRESSINC. 1998: 2627-2636.
Derakhshan A, Peeters RP, Taylor PN, et al. Association of maternal thyroid function with birthweight: a systematic review and individual-participant data meta-analysis[J]. Lancet Diabetes Endocrinol, 2020, 8(6): 501-510.
Vrijkotte TGM, Hrudey EJ, Twickler MB. Early maternal thyroid function during gestation is associated with fetal growth, particularly in male newborns[J]. J Clin Endocrinol Metab, 2017, 102(3): 1059-1066.
León G, Murcia M, Rebagliato M, et al. Maternal thyroid dysfunction during gestation, preterm delivery, and birthweight. The infancia y medio ambiente cohort, spain[J]. Paediatr Perinat Epidemiol, 2015, 29(2): 113-122.
Kingdom JC, Audette MC, Hobson SR, et al. A placenta clinic approach to the diagnosis and management of fetal growth restriction[J]. Am J Obstet Gynecol, 2018, 218(2s): S803-s817.
Clapp MA, Berry M, Shook LL, et al. Low fetal fraction and birth weight in women with negative first-trimester cell-free DNA screening[J]. Am J Perinatol, 2020, 37(1): 86-91.
Shook LL, Clapp MA, Roberts PS, et al. High fetal fraction on first trimester cell-free dna aneuploidy screening and adverse pregnancy outcomes[J]. Am J Perinatol, 2020, 37(1): 8-13.
Chappell LC, Gurung V, Seed PT, et al. Ursodeoxycholic acid versus placebo, and early term delivery versus expectant management, in women with intrahepatic cholestasis of pregnancy: semifactorial randomised clinical trial[J]. BMJ (Clinical Researched), 2012, 344: e3799.
Yeary RA. Embryotoxicity of bilirubin[J]. Am J Obstet Gynecol, 1977, 127(5): 497-498.
Song F, Chen Y, Chen L, et al. Association of elevated maternal serum total bile acids with low birth weight and intrauterine fetal growth restriction[J]. JAMA Netw Open, 2021, 4(7): e2117409.
邱玉梅, 廖利琼, 刘洋, 等. 维生素D在妊娠期肝内胆汁淤积症患者血清和脐带血的水平及其与总胆汁酸的相关性研究[J]. 新医学, 2023, 54(11): 815-820.
Qiu YM, Liao LQ, Liu Y, et al. Vitamin D levels in serum and umbilical cord blood and its correlation with total bile acids in patients with intrahepatic cholestasis of pregnancy[J]. J New Med, 2023, 54(11): 815-820.
Xu Y, Ni M, Zhang Q, et al. Correlation between crown-rump length in the first trimester of pregnancy and neonatal outcomes[J]. BMC Pediatr, 2022, 22(1): 386.
Rossavik IK, Deter RL, Hadlock FP. Mathematical modeling of fetal growth. Ⅳ. Evaluation of trunk growth using the abdominal profile area[J]. J Clin Ultrasound, 1987, 15(1): 31-35.
Palacios J, Rodríguez S, Rodríguez JI. Intra-uterine long bone growth in small-for-gestational-age infants[J]. Eur J Pediatr, 1992, 151(4): 304-307.
Meyer WJ, Gauthier D, Ramakrishnan V, et al. Ultrasonographic detection of abnormal fetal growth with the gestational age-independent, transverse cerebellar diameter/abdominal circumference ratio[J]. Am J Obstet Gynecol, 1994, 171(4): 1057-1063.
Hiersch L, Melamed N. Fetal growth velocity and body proportion in the assessment of growth[J]. Am J Obstet Gynecol, 2018, 218(2S): S700-S711.
Saw SN, Biswas A, Mattar CNZ, et al. Machine learning improves early prediction of small-for-gestational-age births and reveals nuchal fold thickness as unexpected predictor[J]. Prenat Diagn, 2021, 41(4): 505-516.
Sterkenburg AS, Hoffmann A, Reichel J, et al. Nuchal skinfold thickness: a novel parameter for assessment of body composition in childhood craniopharyngioma[J]. J Clin Endocrinol Metab, 2016, 101(12): 4922-4930.
王音, 贾佳, 高宁宁, 等. 胎儿小脑延髓池宽度发育特点及其性别差异[J]. 实用临床医药杂志, 2023, 27(2): 117-121.
Wang Y, Jia J, Gao NN, et al. Developmental characteristics of width of cerebellomedullary cistern in fetus and its differences in gender[J]. J Clin Med Pract, 2023, 27(2): 117-121.
Papastefanou I, Wright D, Syngelaki A, et al. Competing-risks model for prediction of small-for-gestational-age neonate from biophysical and biochemical markers at 11-13 weeks' gestation[J]. Ultrasound Obstet Gynecol, 2021, 57(1): 52-61.
Papastefanou I, Nowacka U, Syngelaki A, et al. Competing risks model for prediction of small-for-gestational-age neonates from biophysical markers at 19 to 24 weeks' gestation[J]. Am J Obstet Gynecol, 2021, 225(5): 530.e1-530.e19.
Sotiriadis A, Figueras F, Eleftheriades M, et al. First-trimester and combined first- and second-trimester prediction of small-for-gestational age and late fetal growth restriction[J]. Ultrasound Obstet Gynecol, 2019, 53(1): 55-61.
Shouval R, Bondi O, Mishan H, et al. Application of machine learning algorithms for clinical predictive modeling: a data-mining approach in SCT[J]. Bone Marrow Transplantation, 2014, 49(3): 332-337.
Wu Q, Nasoz F, Jung J, et al. Machine learning approaches for fracture risk assessment: a comparative analysis of genomic and phenotypic data in 5130 older men[J]. Calcif Tissue Int, 2020, 107(4): 353-361.
Deo RC. Machine learning in medicine[J]. Circulation, 2015, 132(20): 1920-1930.
0
浏览量
2
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构