Detection of changes in the circadian rhythm of cattle in relation to disease, stress, and reproductive events

Disease and stress can disrupt the circadian rhythm of activity in animals. Sensor technologies automatically detect variations daily activity, but it remains difficult to exactly when disruption starts. Here we report a mathematical Fourier-Based Approximation with Thresholding (FBAT) method designed changes cows whatever cause change (typically disease, stress, oestrus). We used data from an indoor positioning system that provides time per hour spent by each cow resting, alleys, or eating. calculated hourly level attributing weight activity. considered 36-h series Fourier transform model during first last 24 h these series. then compared Euclidian distance between two models against given threshold above which had changed. tested on four datasets (giving cumulative total ~120000 cow*days) included disease episodes (acidosis, lameness, mastitis other infectious diseases), reproductive events (oestrus calving) external stimuli animals (e.g. relocation). The obtained over 80% recall normal days detected 95% abnormal rhythms due health events. FBAT could be implemented precision livestock farming monitoring tools alert caretakers individual needing specific care. also has potential anomalies humans guide healthcare intervention wild disturbances. anticipate chronobiological studies apply help relate