September 19, 2025

Machine learning-based prediction of subclinical mastitis in large-scale dairy herds using a locally established somatic cell count threshold

View Full Paper

Ask AI

Bookmark

Authors

Discussion

This analysis identifies a somatic cell count threshold and uses machine learning to predict subclinical mastitis incidence.

The study establishes a somatic cell count threshold of 353,000 cells/mL to identify subclinical mastitis.
Model 1, using CatBoost, achieved a 74.5% accuracy and an AUC of 0.801, indicating strong predictive performance.
Milk conductivity was identified as the most influential predictor for subclinical mastitis across different models.
In Model 2, overall performance decreased, but logistic regression still maintained an accuracy of 67.7%.

Rathnayake et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark