Validation of radio frequency identification with a current transducer to quantify the use of an automatic grooming brush in pre-weaned dairy calves

The objective of this study was to develop and validate an automated recording system of grooming brushes used in dairy calves aged 5 to 6 weeks. Holstein calves (n = 8) housed at the University of Kentucky were fitted with an ultra-high frequency (UHF) radio frequency identification (RFID) ear tag. The calf automatic grooming brush was isolated in a corner of the pen. Current transducers were used to confirm and log brush activation and motion. Two video cameras continuously recorded brush use to create a true value for brush use by each calf during the experimental period. All RFID recordings of brush use alone and in combination with the transducer meter system were regressed against video recordings. After development and adjustment of the system, we predicted the regression of all RFID readings and video observations to be moderate due to false positive readings, and to improve with inclusion of a transducer meter. For daily brush time (min), the coefficient of determination for RFID readings against video observations was weak (r²= 0.26), and for daily brush visits, the correlation coefficient for the RFID reading system against video observations was moderate (r²=0.35). However, incorporation of a transducer meter data to eliminate false positive RFID readings resulted in a strong correlation coefficient for daily brush time (min r²=0.66). These results suggest that automatic grooming brush use in dairy calves has the potential to be automatically recorded using the UHF RFID system in combination with a transducer meter system. Future projects should include other variables and improve the UHF RFID recordings to avoid false negatives.