Environmental and social determinants of health enhance machine learning models for pneumonia readmission - Takeaways - MDSpire
Coming Soon: Introducing MDSpire News. Learn more
Conexiant’s news site is now MDSpire News. Learn more

Incorporating Environmental and Social Factors Improves Machine Learning Predictions for Pneumonia Readmissions

  • By

  • Jack A. Cummins

  • Feifan Liu

  • June 15, 2026

Share

  • 1

    Choi et al. incorporated residential greenness, measured by NDVI, into machine learning models predicting pneumonia readmissions.

  • 2

    NDVI was included among the final 21 predictors after a rigorous three-step feature selection process involving 22,600 patients.

  • 3

    Ablation analysis is suggested to evaluate the performance gain from NDVI integration in predictive models.

  • 4

    NDVI serves as a proxy for social determinants of health, highlighting disparities in residential greenness across socioeconomic contexts.

  • 5

    Future research should explore NDVI's impact on model performance across sociodemographic subgroups to address equity in health predictions.

Original Source(s)

Environmental and social determinants of health enhance machine learning models for pneumonia readmission

  • by Jack A. Cummins, Feifan Liu

  • June 15, 2026

Related Content

Application of deep learning methods in the classification of normal and pneumonia lung images

Ensitrelvir Reduced COVID After Household Exposure

In SCORPIO-PEP, symptomatic COVID-19 occurred in 3% vs 9% of household contacts by day 10.

Cough Audio Recognition for Early Detection of Respiratory Diseases: Algorithm Development and Validation Study