Literature DB >> 34104876

External validation of machine learning models including newborn metabolomic markers for postnatal gestational age estimation in East and South-East Asian infants.

Steven Hawken1,2, Malia S Q Murphy1, Robin Ducharme1, A Brianne Bota1, Lindsay A Wilson1, Wei Cheng1, Ma-Am Joy Tumulak3, Maria Melanie Liberty Alcausin3, Ma Elouisa Reyes3, Wenjuan Qiu4, Beth K Potter2, Julian Little2, Mark Walker1,5, Lin Zhang6,7, Carmencita Padilla8,9, Pranesh Chakraborty10,11, Kumanan Wilson1,12,13.   

Abstract

Background: Postnatal gestational age (GA) algorithms derived from newborn metabolic profiles have emerged as a novel method of acquiring population-level preterm birth estimates in low resource settings. To date, model development and validation have been carried out in North American settings. Validation outside of these settings is warranted.  
Methods:  This was a retrospective database study using data from newborn screening programs in Canada, the Philippines and China. ELASTICNET machine learning models were developed to estimate GA in a cohort of infants from Canada using sex, birth weight and metabolomic markers from newborn heel prick blood samples. Final models were internally validated in an independent sample of Canadian infants, and externally validated in infant cohorts from the Philippines and China. 
Results:  Cohorts included 39,666 infants from Canada, 82,909 from the Philippines and 4,448 from China.  For the full model including sex, birth weight and metabolomic markers, GA estimates were within ±5 days of ultrasound values in the Canadian internal validation (mean absolute error (MAE) 0.71, 95% CI: 0.71, 0.72), and within ±6 days of ultrasound GA in both the Filipino (0.90 (0.90, 0.91)) and Chinese cohorts (0.89 (0.86, 0.92)). Despite the decreased accuracy in external settings, our models incorporating metabolomic markers performed better than the baseline model, which relied on sex and birth weight alone. In preterm and growth-restricted infants, the accuracy of metabolomic models was markedly higher than the baseline model. Conclusions: Accuracy of metabolic GA algorithms was attenuated when applied in external settings.  Models including metabolomic markers demonstrated higher accuracy than models using sex and birth weight alone. As innovators look to take this work to scale, further investigation of modeling and data normalization techniques will be needed to improve robustness and generalizability of metabolomic GA estimates in low resource settings, where this could have the most clinical utility. Copyright:
© 2021 Hawken S et al.

Entities:  

Keywords:  biological modelling; gestational age; newborn screening; preterm birth

Year:  2021        PMID: 34104876      PMCID: PMC8160452.2          DOI: 10.12688/gatesopenres.13131.2

Source DB:  PubMed          Journal:  Gates Open Res        ISSN: 2572-4754


  19 in total

Review 1.  Sample normalization methods in quantitative metabolomics.

Authors:  Yiman Wu; Liang Li
Journal:  J Chromatogr A       Date:  2015-12-10       Impact factor: 4.759

2.  Accurate prediction of gestational age using newborn screening analyte data.

Authors:  Kumanan Wilson; Steven Hawken; Beth K Potter; Pranesh Chakraborty; Mark Walker; Robin Ducharme; Julian Little
Journal:  Am J Obstet Gynecol       Date:  2015-10-28       Impact factor: 8.661

3.  Birth weight curves tailored to maternal world region.

Authors:  Joel G Ray; Michael Sgro; Muhammad M Mamdani; Richard H Glazier; Alan Bocking; Robert Hilliard; Marcelo L Urquia
Journal:  J Obstet Gynaecol Can       Date:  2012-02

Review 4.  Diagnostic Accuracy of Neonatal Assessment for Gestational Age Determination: A Systematic Review.

Authors:  Anne Cc Lee; Pratik Panchal; Lian Folger; Hilary Whelan; Rachel Whelan; Bernard Rosner; Hannah Blencowe; Joy E Lawn
Journal:  Pediatrics       Date:  2017-11-17       Impact factor: 7.124

5.  Clinical and environmental influences on metabolic biomarkers collected for newborn screening.

Authors:  Kelli K Ryckman; Stanton L Berberich; Oleg A Shchelochkov; Daniel E Cook; Jeffrey C Murray
Journal:  Clin Biochem       Date:  2012-09-23       Impact factor: 3.281

6.  International standards for newborn weight, length, and head circumference by gestational age and sex: the Newborn Cross-Sectional Study of the INTERGROWTH-21st Project.

Authors:  José Villar; Leila Cheikh Ismail; Cesar G Victora; Eric O Ohuma; Enrico Bertino; Doug G Altman; Ann Lambert; Aris T Papageorghiou; Maria Carvalho; Yasmin A Jaffer; Michael G Gravett; Manorama Purwar; Ihunnaya O Frederick; Alison J Noble; Ruyan Pang; Fernando C Barros; Cameron Chumlea; Zulfiqar A Bhutta; Stephen H Kennedy
Journal:  Lancet       Date:  2014-09-06       Impact factor: 79.321

7.  Centering, scaling, and transformations: improving the biological information content of metabolomics data.

Authors:  Robert A van den Berg; Huub C J Hoefsloot; Johan A Westerhuis; Age K Smilde; Mariët J van der Werf
Journal:  BMC Genomics       Date:  2006-06-08       Impact factor: 3.969

8.  Gestational dating by metabolic profile at birth: a California cohort study.

Authors:  Laura L Jelliffe-Pawlowski; Mary E Norton; Rebecca J Baer; Nicole Santos; George W Rutherford
Journal:  Am J Obstet Gynecol       Date:  2015-12-11       Impact factor: 8.661

9.  Global, regional, and national causes of under-5 mortality in 2000-15: an updated systematic analysis with implications for the Sustainable Development Goals.

Authors:  Li Liu; Shefali Oza; Dan Hogan; Yue Chu; Jamie Perin; Jun Zhu; Joy E Lawn; Simon Cousens; Colin Mathers; Robert E Black
Journal:  Lancet       Date:  2016-11-11       Impact factor: 79.321

10.  Predicting gestational age using neonatal metabolic markers.

Authors:  Kelli K Ryckman; Stanton L Berberich; John M Dagle
Journal:  Am J Obstet Gynecol       Date:  2015-12-02       Impact factor: 8.661

View more
  1 in total

1.  Socioeconomic disparities in adverse birth outcomes in the Philippines.

Authors:  Ryan C V Lintao; Erlidia F Llamas-Clark; Ourlad Alzeus G Tantengco
Journal:  Lancet Reg Health West Pac       Date:  2022-04-11
  1 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.