Prediction of Neurodevelopmental Outcomes in Very Preterm Infants: Comparing Machine Learning Methods to Logistic Regression

Purpose: Is machine learning (ML) superior to the traditionally used logistic regression (LR) in prediction of neurodevelopmental outcomes in preterm infants? Objectives: To develop and internally validate a ML model to predict neurodevelopmental impairment (NDI) in very preterm infants (<31 week...

Full description

Saved in:
Bibliographic Details
Main Authors: Jehier Afifi, Tahani Ahmad, Alessandro Guida, Michael John Vincer, Samuel Alan Stewart
Format: Article
Language:English
Published: MDPI AG 2024-12-01
Series:Children
Subjects:
Online Access:https://www.mdpi.com/2227-9067/11/12/1512
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846105270616326144
author Jehier Afifi
Tahani Ahmad
Alessandro Guida
Michael John Vincer
Samuel Alan Stewart
author_facet Jehier Afifi
Tahani Ahmad
Alessandro Guida
Michael John Vincer
Samuel Alan Stewart
author_sort Jehier Afifi
collection DOAJ
description Purpose: Is machine learning (ML) superior to the traditionally used logistic regression (LR) in prediction of neurodevelopmental outcomes in preterm infants? Objectives: To develop and internally validate a ML model to predict neurodevelopmental impairment (NDI) in very preterm infants (<31 weeks) at 36 months corrected age, using clinical predictors. Methods: A retrospective cohort of very preterm infants (23<sup>0–</sup>30<sup>6</sup> weeks) born between January 2004 and December 2016 in Nova Scotia, Canada. Survivors with neurodevelopmental assessment at 36 months corrected age were included. The study sample was randomly split (80:20) into a development and testing datasets. We compared four methods: LR, elastic net (EN), random forest ensemble (RF) and gradient boosting (XGB), in relation to discrimination (AUC), calibration, and diagnostic properties. Results: Of 811 eligible infants, 663 were included (mean gestational age 28 weeks, mean birth weight 1137 g and 52% male). Of those, 195 (29%) developed NDI and 468 (71%) did not. On internal validation using the testing dataset, all four models provided good discrimination of NDI with comparable AUC. RF was superior to the other three methods with a higher AUC (0.79 vs. 0.74, 0.74, and 0.73 for XGB, EN and LR, respectively), but all models have overlapped CIs. Conclusions: In this population-based cohort of very preterm infants, RF was superior to conventional LR in prediction of NDI at 3 years corrected age. Accurate prediction of preterm infants at risk of NDI enables early referrals for intervention programs and resources allocation toward those who are most likely to benefit.
format Article
id doaj-art-3b87ace4c27b461aa1cd8c7555d3a638
institution Kabale University
issn 2227-9067
language English
publishDate 2024-12-01
publisher MDPI AG
record_format Article
series Children
spelling doaj-art-3b87ace4c27b461aa1cd8c7555d3a6382024-12-27T14:17:56ZengMDPI AGChildren2227-90672024-12-011112151210.3390/children11121512Prediction of Neurodevelopmental Outcomes in Very Preterm Infants: Comparing Machine Learning Methods to Logistic RegressionJehier Afifi0Tahani Ahmad1Alessandro Guida2Michael John Vincer3Samuel Alan Stewart4Division of Neonatal Perinatal Medicine, Department of Pediatrics, Dalhousie University, Halifax, NS B3K 6R8, CanadaDepartment of Diagnostic Imaging, Dalhousie University, Halifax, NS B3K 6R8, CanadaDepartment of Diagnostic Imaging, Dalhousie University, Halifax, NS B3K 6R8, CanadaDivision of Neonatal Perinatal Medicine, Department of Pediatrics, Dalhousie University, Halifax, NS B3K 6R8, CanadaDepartment of Community Health & Epidemiology, Dalhousie University, Halifax, NS B3H 1V7, CanadaPurpose: Is machine learning (ML) superior to the traditionally used logistic regression (LR) in prediction of neurodevelopmental outcomes in preterm infants? Objectives: To develop and internally validate a ML model to predict neurodevelopmental impairment (NDI) in very preterm infants (<31 weeks) at 36 months corrected age, using clinical predictors. Methods: A retrospective cohort of very preterm infants (23<sup>0–</sup>30<sup>6</sup> weeks) born between January 2004 and December 2016 in Nova Scotia, Canada. Survivors with neurodevelopmental assessment at 36 months corrected age were included. The study sample was randomly split (80:20) into a development and testing datasets. We compared four methods: LR, elastic net (EN), random forest ensemble (RF) and gradient boosting (XGB), in relation to discrimination (AUC), calibration, and diagnostic properties. Results: Of 811 eligible infants, 663 were included (mean gestational age 28 weeks, mean birth weight 1137 g and 52% male). Of those, 195 (29%) developed NDI and 468 (71%) did not. On internal validation using the testing dataset, all four models provided good discrimination of NDI with comparable AUC. RF was superior to the other three methods with a higher AUC (0.79 vs. 0.74, 0.74, and 0.73 for XGB, EN and LR, respectively), but all models have overlapped CIs. Conclusions: In this population-based cohort of very preterm infants, RF was superior to conventional LR in prediction of NDI at 3 years corrected age. Accurate prediction of preterm infants at risk of NDI enables early referrals for intervention programs and resources allocation toward those who are most likely to benefit.https://www.mdpi.com/2227-9067/11/12/1512machine learningpredictive modelingpreterm infantsneurodevelopment
spellingShingle Jehier Afifi
Tahani Ahmad
Alessandro Guida
Michael John Vincer
Samuel Alan Stewart
Prediction of Neurodevelopmental Outcomes in Very Preterm Infants: Comparing Machine Learning Methods to Logistic Regression
Children
machine learning
predictive modeling
preterm infants
neurodevelopment
title Prediction of Neurodevelopmental Outcomes in Very Preterm Infants: Comparing Machine Learning Methods to Logistic Regression
title_full Prediction of Neurodevelopmental Outcomes in Very Preterm Infants: Comparing Machine Learning Methods to Logistic Regression
title_fullStr Prediction of Neurodevelopmental Outcomes in Very Preterm Infants: Comparing Machine Learning Methods to Logistic Regression
title_full_unstemmed Prediction of Neurodevelopmental Outcomes in Very Preterm Infants: Comparing Machine Learning Methods to Logistic Regression
title_short Prediction of Neurodevelopmental Outcomes in Very Preterm Infants: Comparing Machine Learning Methods to Logistic Regression
title_sort prediction of neurodevelopmental outcomes in very preterm infants comparing machine learning methods to logistic regression
topic machine learning
predictive modeling
preterm infants
neurodevelopment
url https://www.mdpi.com/2227-9067/11/12/1512
work_keys_str_mv AT jehierafifi predictionofneurodevelopmentaloutcomesinverypreterminfantscomparingmachinelearningmethodstologisticregression
AT tahaniahmad predictionofneurodevelopmentaloutcomesinverypreterminfantscomparingmachinelearningmethodstologisticregression
AT alessandroguida predictionofneurodevelopmentaloutcomesinverypreterminfantscomparingmachinelearningmethodstologisticregression
AT michaeljohnvincer predictionofneurodevelopmentaloutcomesinverypreterminfantscomparingmachinelearningmethodstologisticregression
AT samuelalanstewart predictionofneurodevelopmentaloutcomesinverypreterminfantscomparingmachinelearningmethodstologisticregression