Machine Learning for Febrile Infants – A New Paradigm? – AAP News

  • Lauren
  • August 27, 2020
  • Comments Off on Machine Learning for Febrile Infants – A New Paradigm? – AAP News

Perhaps many people are like me in that hearing the word “machine learning” for the
first time brings forth images of Skynet from The Terminator movies or Haley Joel Osment’s character from the Steven Spielberg’s film A.I. Artificial Intelligence. However, machine learning has now become a regular part of our vernacular when it
comes to predictive modeling in many conditions. In this issue of Pediatrics(10.1542/peds.2019-4096), Ramgopal et al use machine learning methods to derive and validate a new prediction
model for risk stratification of febrile infants ≤60 days of age. Using various machine
learning approaches, the authors developed a prediction model with high sensitivity
and specificity compared with recent prediction models for febrile infants.

So, are machine learning models the new paradigm for risk stratification of febrile
infants? The results are intriguing, particularly the high specificity of the model,
but further work must be done, as explained nicely by Chamberlain et al in an accompanying
commentary (10.1542/peds.2020-012203). In addition to needing external validation in a new, diverse sample of febrile
infants, the biggest question in practice is how to use a machine learning model for
risk stratification. Unlike traditional prediction models, machine learning models
do not provide clean cutoff values for predictors that can be easily used by clinicians
at the bedside. The models use complex algorithms for risk predictions that can, at
face value, be challenging to comprehend, not unlike the plotline of the Christopher
Nolan movie Inception. However, the models are not intended to be used in a similar manner to traditional
prediction models. Machine learning models are to be integrated into electronic health
records or other types of clinical decision support to provide more accurate risk
predictions for clinicians to use at the bedside.

So, is machine learning the future for risk stratification of febrile infants? Maybe
not yet–pending further validation of the model and an assessment of how best to implement
it in practice. Nonetheless, it is exciting to see this methodology applied to febrile
infants with an eye towards the future.

Copyright © 2020 American Academy of Pediatrics