Comparison of Estrus Parameters in Nulliparous Heifers by Two Automated Activity Monitoring Systems

Abstract Text: The aim of this study was to compare a commercial (Heatime, SCR Engineers, Israel) and a research based (IceTag, IceRobotics, Scotland) activity monitoring system on their ability to measure estrus episodes in nulliparous heifers (n=57; 119 estrus episodes) starting at 12 mo old. Only heifers detected by Heatime system were evaluated, therefore only the sensitivity of the systems could be measured for accuracy. Ultrasound scanning to describe ovarian structures and blood sampling for estradiol were performed at each estrus episode and later used to determine heat detection precision. Secondary signs of estrus (mucus, uterine tonus, and visual mounting and standing) were also recorded. Heifers were housed in dynamic groups of 24 animals in a free stall barn from May 2012 to August 2013. Data was analyzed using ANOVA and Pearson correlations using proc GLM, CORR and REG of SAS. Sensitivity of Heatime system was 84.7% (94/111), whereas IceTag had 98.7% (74/75) sensitivity. Estrus episodes lasted 12.7±5.6 h on Heatime and 15.0±3.9 h on IceTag and were highly correlated between them (r=0.60; P<0.01). Mean time difference of estrus initiation and end was 3.5±4.3 h and 2.9±4.9 h (IceTag as reference). Peak activity was also positively correlated (r=0.62; P<0.01) between systems and was 76.6±19.9 index value and 4.6±1.7 times increase related to baseline on Heatime and IceTag, respectively. Duration and peak activity were highly correlated for Heatime (r=0.63; P<0.01), but not significant when using the IceTags (r=0.13; P=0.26). Diameter of preovulatory follicle (15.7±2.6 mm; mean±SD) had no correlation with duration or peak activity as measured by both sensors. However, concentrations of estradiol in plasma were correlated with both duration (r=0.47) and peak activity (r=0.36; P<0.01) measured only by the Heatime system. Number of baseline steps/h measured by the IceTag system had a strong negative correlation (P<0.01) with peak activity in both sensors (r=-0.37 and -0.70). Heifers with more than one secondary sign of estrus had great peak activity and duration (P<0.05) on the Heatime system, but not on the IceTag sensor. Secondary signs were not affected by follicle diameter and concentration of estradiol. Results indicate that both activity monitoring systems identified estrus with high sensitivity and that major measurements (i.e. episodes, duration, peak/intensity, initiation of estrus) were mostly correlated. However, some variables such as concentration of estradiol and secondary signs of estrus should be analyzed independently for research purposes.

Keywords: automated monitoring system, dairy cows, estrus detection