Thermal microchip sensors can automate body temperature measurements. The best site of implantation is still unknown, and the accuracy and precision of body temperature predictions based on microchip data need to be investigated. The aim of this study was to investigate the best site for microchip implant for monitoring body temperature in dairy calves. Seventeen calves were used (32.2±5.2kg of body weight) and the microchips were implanted four days after birth. The microchips were implanted at navel, ear and tail base (subcutaneous), neck (cleidocephalicus) and internal face of leg (gracilis) (intramuscular). Rectal temperature (RT, °C), obtained with a clinical thermometer, was considered as core temperature. Air temperature (AT), relative humidity (RH) and the temperature and humidity index (THI) were evaluated at the same time of rectal and microchip temperature measurements over 56 days. The range of AT, RH and THI was 7.6-34.4°C, 17.5-99.0% and 50.6 to 91.5. The average for rectum, ear, neck, tail, leg, and navel were 38.7; 36.9; 38.0; 37.0, 37.8 and 37.0°C. The intramuscular implantations had closest values to RT. The correlations between RT and ear, neck, tail, leg, and navel temperatures were 0.56, 0.60, 0.60, 0.53 e 0.48. The RT prediction based on microchip data had precision (rc) ranged between 0.49 and 0.60 and accuracy (Cb) between 0.79 and 0.88. The inclusion of AT, RH and THI as predictive variables in models decrease the mean absolute error (23%) and increase the precision (21.3%) and accuracy (10.2%). The Concordance Correlation Coefficient and root-mean-square error for equations using tail or neck microchips were 0.68 and 0.67, and 0.29 and 0.28°C, respectively. The tail base is a promising site for microchip implantation to predict rectal temperature. The inclusion of air temperature as a predictive variable in the models is recommended.
Read full abstract