Italian Journal of Animal Science, cilt.25, sa.1, ss.321-330, 2026 (SCI-Expanded, Scopus)
Early detection of health disorders is vital for sustainable dairy production and animal welfare. This prospective observational study evaluated whether continuous monitoring of locomotor activity using neck-mounted sensors can serve as a behavioural biomarker for common health disorders in Holstein and Simmental cows during the transition period. A total of 96 cows (48 per breed) were monitored for 12 months. Digestive disorders showed the highest diagnostic accuracy in Simmental cows (AUC = 0.90), while indigestion and mastitis were reliably detected in Holsteins (AUC = 0.79 and 0.76, respectively). Reproductive and hoof disorders altered activity patterns but exhibited low sensitivity despite high specificity, indicating that locomotor monitoring alone is insufficient for comprehensive detection. Disease status significantly predicted locomotor activity (F(1,88)=14.38, p < 0.001), whereas breed differences were significant only for reproductive conditions. These findings demonstrate the potential of breed-adapted, multimodal sensor monitoring as a practical tool in precision herd health management, particularly for early detection of metabolic and digestive disorders in dairy cows.