Machine learning approaches to predict and detect early-onset of digital dermatitis in dairy cows using sensor data.

Magana J, Gavojdian D, Menahem Y, Lazebnik T, Zamansky A, Adams-Progar A. Machine learning approaches to predict and detect early-onset of digital dermatitis in dairy cows using sensor data. Front Vet Sci. 2023;10. doi:10.3389/fvets.2023.1295430.

This study investigated the use of machine learning algorithms to detect and predict early-onset digital dermatitis (DD) in dairy cows based on sensor data. Using the Tree-Based Pipeline Optimization Tool (TPOT), the detection model achieved 79% accuracy on the day clinical signs appeared, while a combined K-means and TPOT model predicted DD two days prior to symptoms with 64% accuracy. These models could facilitate real-time automated monitoring of DD in conventional dairy environments. The results indicate that behavioral changes in cows can be leveraged for early warning systems to improve herd health management.