Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
May 9, 2026Journal of Animal Science

Data-driven Forecasting of Nursery Piglet Mortality in Commercial Swine Systems

View Full Paper
Ask AI
Bookmark
Share

Authors

MCMateus de Castro Duarte CardosoTTThinh TienJSJackson C Sterle

Discussion

Loading...

Member takes

Overview

Randomized trial develops a predictive framework for nursery piglet mortality risk in commercial swine systems, suggesting improved intervention strategies.

Key Points

  • This research aims to create a predictive model for nursery piglet mortality using historical production data.
  • Analyzed 306 variables from multiple swine farms to gather production metrics.
  • Utilized machine learning algorithms, Extreme Gradient Boosting and Extra Trees, and blended their outputs.
  • Evaluated model performance on a hold-out test set using correlation, accuracy, and other metrics.
  • The ensemble model achieved a Pearson correlation of 0.672 and accuracy of 0.823 on the independent test set.
  • Sensitivity was 0.806 and negative predictive value reached 0.884, indicating effective risk identification.
  • Isotonic regression calibration improved prediction reliability without compromising correlation strength.

Cite This Study

Cardoso et al. (2026) studied this question.

synapsesocial.com/papers/69fecfcdb9154b0b82876bd1https://doi.org/10.1093/jas/skag107.010
View Full Paper
Ask AI
Bookmark
Share