Leveraging nonlinear relationships and interactions to improve 30-day pneumonia readmission machine learning models

Eric M Mortensen, Nkiruka Atuegwu, Shane J Sacco · 2026

Read Leveraging nonlinear relationships and interactions to improve 30-day pneumonia readmission machine learning models by Eric M Mortensen, Nkiruka Atuegwu, Shane J Sacco (2026) 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/plos-pone-0349804.txt.

Description

Eric M Mortensen, Nkiruka Atuegwu, and Shane J Sacco tested whether fancy learning models beat plain regression at predicting return to hospital within thirty days after pneumonia. Records covered just over twenty seven hundred adults at one academic centre from 2018 to 2024, with full vital sign values kept as continuous inputs rather than cut into bands.

About ten percent returned within thirty days. Gradient boosting and neural nets scored about the same as plain regression on both accuracy scales. Drug abuse history and a kidney lab value emerged as underused predictors. Extra model complexity bought nothing here; better data and follow up design may matter more than smarter maths.

Catalog record

CREATOREric M Mortensen, Nkiruka Atuegwu, Shane J Sacco
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE33529
LICENSECC_BY
RIGHTSCC BY
SOURCEPLOS

Source: https://doi.org/10.1371/journal.pone.0349804

Browse the full library · About the Tower of Babel