University of Veterinary Medicine Vienna - Research portal

Diagrammed Link to Homepage University of Veterinary Medicine, Vienna

Selected Publication:

Publication type: Journal Article
Document type: Full Paper

Year: 2019

Author(s): Krieger, S; Oczak, M; Lidauer, L; Berger, A; Kickinger, F; Öhlschuster, M; Auer, W; Drillich, M; Iwersen, M

Title: An ear-attached accelerometer as an on-farm device to predict the onset of calving in dairy cows.

Source: Biosystems Engineering 2019; 184: 190-199

Authors Vetmeduni Vienna:

Drillich Marc
Iwersen Michael
Krieger Stefanie
Oczak Maciej

Vetmed Research Units
University Clinic for Ruminants, Clinical Unit of Herd Management in ruminants
Institute of Animal Welfare Science

Project(s): Evaluation of an ear tag based accelerometer for detecting bovine respiratory disease in dairy calves

The objective of the present study of dairy cows was to analyse data collected by an ear-tag containing an accelerometer, prepartum and during birth, (1) to determine activity, rumination and lying time of the dams, (2) to use data to develop an algorithm to predict calving and (3) to test the performance of this algorithm. Four weeks before the expected day of calving, we fixed an ear-tag, which contained a tri-axial accelerometer, to the dam's right ear. In total, 894 calving data sets were eligible for the final analyses. In total, six input variables based on accelerometer data were used to develop an algorithm for calving prediction; rumination time, activity, and lying time of individual animals and in relation to the remaining animals in the group. To predict the impending calving on the basis of these input variables, transfer function models were used. The performance of the algorithm predicting parturition was determined at the time periods of -72, -48, -24, -12, -6, -3 and -1 h before the expulsion of the calf. The highest balanced accuracy was achieved at -1 h, with 74%. The highest sensitivity was achieved at -1 h (54%), whereas the lowest was at -72 h (19%). Our findings indicate that it is possible to predict parturition in dairy cows based on input variables from acceleration data. Further studies should test and improve the algorithm under different farming and management systems. (C) 2019 IAgrE. Published by Elsevier Ltd. All rights reserved.

© University of Veterinary Medicine ViennaHelp and Downloads