Files
Abstract
Mastitis remains a significant challenge for maximizing dairy cow health and well-being and reducing antibiotic usage. Therefore, investigation of novel mammary health parameters (MHP) for improving mastitis treatment plans is necessary. Our objectives were to: a) evaluate the relationship between differential somatic cell counts (DSCC) and antibiotic success during subclinical mastitis, b) determine DSCC thresholds associated with cure after antibiotic treatment, and b) evaluate how DSCC performs as a potential MHP in selective dry cow therapy (SDCT) programs modeled using machine learning. Results found differences in DSCC between quarters that cured an intramammary infection (IMI) compared to quarters that failed to cure following antibiotic administration. Total leukocyte counts were the optimal MHP to use in setting DSCC thresholds for maximizing IMI cure rates. Conversely, the integration of DSCC into machine learning models for detection of mastitis at the time of dry off did not improve the models’ classification metrics. In addition, the combination of these MHP did not improve the models’ classification metrics compared to each alone.