Prediction of Neurodevelopmental Outcomes in Very Preterm Infants: Comparing Machine Learning Methods to Logistic Regression
Jehier Afifi, Tahani Ahmad, Alessandro Guida, Michael John Vincer, Samuel Alan Stewart · 2024
Read Prediction of Neurodevelopmental Outcomes in Very Preterm Infants: Comparing Machine Learning Methods to Logistic Regression by Jehier Afifi, Tahani Ahmad, Alessandro Guida, Michael John Vincer, Samuel Alan Stewart (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11674291.txt.
Description
Jehier Afifi and four co-authors test whether machine learning beats the standard statistical tool at predicting later development in babies born very early. Their 2024 study follows infants born before 31 weeks in Nova Scotia, Canada, between 2004 and 2016. Of 663 children checked at age three, 195 had a developmental impairment.
The team trained four models on clinical facts from the newborn period and held back a fifth of the cases for testing. All four models sorted children fairly well. A random forest did best, with an accuracy score of 0.79 against 0.73 for the standard method, though the ranges overlapped. Early flags matter because therapy works best when it starts young. The authors back careful use of the forest model for referral planning.
Catalog record
| CREATOR | Jehier Afifi, Tahani Ahmad, Alessandro Guida, Michael John Vincer, Samuel Alan Stewart |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 23 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |