Adsa Logo White Adsa Title White

Evaluation and characterization of estrus alerts from an ear-attached accelerometer-based automated estrus detection system.

J. O. Giordano

Events

06-23-2020

Abstract:

241
Evaluation and characterization of estrus alerts from an ear-attached accelerometer-based automated estrus detection system.
E. M. Schilkowsky, G. E. Granados, E. M. Sitko, M. Masello, M. M. Perez, J. O. Giordano*. Cornell University Ithaca, NY.

Our objectives were to evaluate an ear-attached automated estrus detection system (AEDS; Smartbow; Zoetis Inc.) that monitored physical activity (PHA) and rumination time (RUM) for detection of estrus and characterize AEDS estrus alert features. Lactating Holstein cows (n = 216) had their estrous cycle synchronized at 50 � 3 DIM or 18 � 3 d after AI. For 7 d after induction of luteolysis with PGF (d 0) visual observation of estrus behavior (30 min 2 times per day) and data from a pressure-activated tail-head sensor (HeatWatch) were used as reference test (RT) for estrus. Concomitantly, estrus alerts and their features were collected from the AEDS. Progesterone levels confirmed luteal regression and ultrasonography timing of ovulation. Performance metrics for the AEDS were obtained with PROC FREQ of SAS using only estrus detected by the RT or detection of estrus by the RT and ovulation for cows with a discrepancy between the RT and AEDS. Based on the RT, 87% presented estrus and ovulated, 1% presented estrus and did not ovulate, 6% did not present estrus but ovulated, and 6% did not present estrus nor ovulation. Sensitivity (Se), specificity (Sp), negative predictive value (NPV), positive predictive value (PPV), and balanced accuracy (BA) were 92 (88—96), 69 (51—87), 53 (36—70), 96 (93—99), and 80% (70—92), respectively when only the RT was used. After correcting for ovulation for cows with a discrepancy between the RT and AEDS, Se, Sp, NPV, PPV, and BA (mean and 95% CI) were 93 (88—96), 92 (62—100), 44 (32—57), 99 (96—100), and 93% (75—98), respectively. The mean � SD and range for the interval from induction of luteolysis to estrus alerts, estrus alert duration, and the onset of estrus alerts to ovulation interval were: 72.2 � 18.1 (30—160), 13.5 � 3.8 (2—28), and 23.8 � 7.1 (4—44) h. Cows spent 8.5 � 18.1 (2—28) min per hour in high active time (measure of estrus intensity). We concluded that an ear-attached AEDS that monitored PHA and RUM was effective at detecting cows in estrus and generated few false positive alerts when accounting for ovulation.

Keywords: ear-attached sensor, estrus, cow.

Biography: NA