A novel accelerometry approach combining information on classified behaviors and quantified physical activity for assessing health status of cattle: a preliminary study

Abstract

Evaluation of animal behavior provides information on health and well-being in animals. In this preliminary study, we investigated the effectiveness of an accelerometry approach to evaluate changes in health status of cattle by combining physical activity quantified by dynamic body acceleration (DBA) and qualitative behavioral classification data. Although feeding low vitamin A (VA) diets to fattening cattle is a popular technique to enhance marbling of beef in Japan, the effects of VA restriction on behavior of fattening cattle have been unclear. Therefore, the effect of VA restriction on behavioral change in feedlot steers was assessed by the approach as an example of evaluation of the change in health status. Ten Japanese Black fattening steers were divided into VA restricted (RES) and supplemented (SUP) groups (five in each group). The RES steers were fed lower VA diets than SUP from 11 to 20 months of age. The measurements of blood VA concentrations, accelerations and classified behaviors of the steers were conducted at 18 and 24 months of age. An accelerometer was fitted dorsally to each steer and three axial accelerations were measured at 10 Hz. The raw acceleration data was processed to calculate vectorial DBA (VeDBA) per second. Behavior of the steers was recorded for 8.5 h on one day and visually classified per second as lying, standing, walking or feeding. As the results, the blood VA concentration at 18 months of age was lower in RES than SUP (P < 0.05) (53.2 and 95.6 IU/dL, respectively) but the concentration was similar between the groups at 24 months of age. All VA concentrations were over the recommended minimum value and none of the steers showed clinical manifestations of VA deficiency, indicating that VA restriction was controlled as expected. The durations of classified behaviors and the overall mean VeDBA were similar between the groups. However, the VeDBA for walking was higher in RES than SUP at both ages (P < 0.05). The results suggested that animals under VA restriction might slightly alter their gait as a stress response at 18 months of age and a residual of this might be detected even after the increase of blood VA concentration at 24 months of age; small changes in behavior that could not be detectable by visual classification were quantified via accelerometry. In the future, a combination of automatic behavioral classification and quantified physical activity such as DBA will be useful for feedback on animal health and well-being.