A machine learning approach to estimating preterm infants survival: development of the Preterm Infants Survival Assessment (PISA) predictor

Podda, Marco; Bacciu, Davide; Micheli, Alessio; Bellù, Roberto; Placidi, Giulia; Gagliardi, Luigi

doi:10.1038/s41598-018-31920-6

Cited by 68 publications

(74 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In a more recent work, Podda et al [21] presented multiple machine learning methods for predicting neonatal mortality in the terms of the probability of survival: logistic regression, k-nearest neighbor, random forest, gradient boosting machine, support vector machine, and neural network. Our work also considers logistic regression, k-nearest neighbor, random forest, and support vector machine, but we also use five additional classifiers and evaluate our classifiers in predicting different morbidities.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Methods for Neonatal Mortality and Morbidity Classification

et al. 2020

View full text Add to dashboard Cite

Preterm birth is the leading cause of mortality in children under the age of five. In particular, low birth weight and low gestational age are associated with an increased risk of mortality. Preterm birth also increases the risks of several complications, which can increase the risk of death, or cause longterm morbidities with both individual and societal impacts. In this work, we use machine learning for prediction of neonatal mortality as well as neonatal morbidities of bronchopulmonary dysplasia, necrotizing enterocolitis, and retinopathy of prematurity, among very low birth weight infants. Our predictors include time series data and clinical variables collected at the neonatal intensive care unit of Children's Hospital, Helsinki University Hospital. We examine 9 different classifiers and present our main results in AUROC, similar to our previous studies, and in F1-score, which we propose for classifier selection in this study. We also investigate how the predictive performance of the classifiers evolves as the length of time series is increased, and examine the relative importance of different features using the random forest classifier, which we found to generally perform the best in all tasks. Our systematic study also involves different data preprocessing methods which can be used to improve classifier sensitivities. Our best classifier AUROC is 0.922 in the prediction of mortality, 0.899 in the prediction of bronchopulmonary dysplasia, 0.806 in the prediction of necrotizing enterocolitis, and 0.846 in the prediction of retinopathy of prematurity. Our best classifier F1-score is 0.493 in the prediction of mortality, 0.704 in the prediction of bronchopulmonary dysplasia, 0.215 in the prediction of necrotizing enterocolitis, and 0.368 in the prediction of retinopathy of prematurity.

show abstract

Section: Introductionmentioning

confidence: 99%

Machine Learning Methods for Neonatal Mortality and Morbidity Classification

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Progress is now being made in the use of more advanced techniques, such as machine learning, to develop such prediction tools. 35 All of the foregoing studies focused on survival over relatively short time frames and did not consider outcomes in later life among survivors. Other studies examined cognitive outcomes in survivors 14,[17][18][19] but did not always include the full spectrum of gestational age or information on non-survivors.…”

Section: Discussionmentioning

confidence: 99%

Predicting Long-Term Survival Without Major Disability for Infants Born Preterm

Bourke

Wong

Srinivasjois

et al. 2019

The Journal of Pediatrics

View full text Add to dashboard Cite

Objective To describe the long-term neurodevelopmental and cognitive outcomes for children born preterm. Study design In this retrospective cohort study, information on children born in Western Australia between 1983 and 2010 was obtained through linkage to population databases on births, deaths, and disabilities. For the purpose of this study, disability was defined as a diagnosis of intellectual disability, autism, or cerebral palsy. The Kaplan-Meier method was used to estimate the probability of disability-free survival up to age 25 years by gestational age. The effect of covariates and predicted survival was examined using parametric survival models. Results Of the 720 901 recorded live births, 12 083 children were diagnosed with disability, and 5662 died without any disability diagnosis. The estimated probability of disability-free survival to 25 years was 4.1% for those born at gestational age 22 weeks, 19.7% for those born at 23 weeks, 42.4% for those born at 24 weeks, 53.0% for those born at 25 weeks, 78.3% for those born at 28 weeks, and 97.2% for those born full term (39-41 weeks). There was substantial disparity in the predicted probability of disability-free survival for children born at all gestational ages by birth profile, with 5-year estimates of 4.9% and 10.4% among Aboriginal and Caucasian populations, respectively, born at 24-27 weeks and considered at high risk (based on low Apgar score, male sex, low sociodemographic status, and remote region of residence) and 91.2% and 93.3%, respectively, for those at low risk (ie, high Apgar score, female sex, high sociodemographic status, residence in a major city). Conclusions Apgar score, birth weight, sex, socioeconomic status, and maternal ethnicity, in addition to gestational age, have pronounced impacts on disability-free survival.

show abstract

“…Artificial intelligence, particularly Artificial Neural Networks (ANN) are robust methodologies for the forecasting of diagnosis/prognosis with high predictive accuracy, and nowadays are used to support medical decisions in NICUs (17,18). Different computational models have been developed to predict adult, pediatric and neonatal sepsis (19)(20)(21)(22)(23).…”

Section: Introductionmentioning

confidence: 99%

Neonatal Sepsis Diagnosis Decision-Making Based on Artificial Neural Networks

Helguera-Repetto

Soto-Ramirez

Villavicencio-Carrisoza

et al. 2020

Front. Pediatr.

View full text Add to dashboard Cite

Neonatal sepsis remains difficult to diagnose due to its non-specific signs and symptoms. Traditional scoring systems help to discriminate between septic or not patients, but they do not consider every single patient particularity. Thus, the purpose of this study was to develop an early-and late-onset neonatal sepsis diagnosis model, based on clinical maternal and neonatal data from electronic records, at the time of clinical suspicion. A predictive model was obtained by training and validating an artificial Neural Networks (ANN) algorithm with a balanced dataset consisting of preterm and term non-septic or septic neonates (early-and late-onset), with negative and positive culture results, respectively, using 25 maternal and neonatal features. The outcome of the model was sepsis or not. The performance measures of the model, evaluated with an independent dataset, outperformed physician's diagnosis using the same features based on traditional scoring systems, with a 93.3% sensitivity, an 80.0% specificity, a 94.4% AUROC, and a regression coefficient of 0.974 between actual and simulated results. The model also performed well-relative to the state-of-the-art methods using similar maternal/neonatal variables. The top 10 factors estimating sepsis were maternal age, cervicovaginitis and neonatal: fever, apneas, platelet counts, gender, bradypnea, band cells, catheter use, and birth weight.

show abstract

A machine learning approach to estimating preterm infants survival: development of the Preterm Infants Survival Assessment (PISA) predictor

Cited by 68 publications

References 28 publications

Machine Learning Methods for Neonatal Mortality and Morbidity Classification

Machine Learning Methods for Neonatal Mortality and Morbidity Classification

Predicting Long-Term Survival Without Major Disability for Infants Born Preterm

Neonatal Sepsis Diagnosis Decision-Making Based on Artificial Neural Networks

Contact Info

Product

Resources

About