Predicting somatic cell counts in an automatic milking rotary using conductivity and generalised additive models

Abstract

This study investigated if conductivity sensor data from an automatic milking rotary (AMR) could be used for explaining composite somatic cell count (CMSCC). By predicting somatic cell counts for individual cows, monitoring of udder health can be improved by reducing both labour and costs for the farmer. During a period of eight weeks, milk samples from 380 Holstein- Friesian cows milked in an AMR were analysed once weekly for composite somatic cell count (CMSCC). Regularly recorded sensor data at quarter and composite level from the herd management system was stored and used as input to generalised additive models (GAM) used in the analyses. Several new conductivity variables were created combining quarter conductivity. Furthermore, past period values, i.e. lags of up to seven days (14 milking sessions), were added to the variables. A Multivariable GAM was fitted in order to compare the importance of the potential predictor variables as well as model performance. The variance in conductivity between quarters, one milk session before the CMSCC sample event, was the most important variable. Conductivity variables from combined quarters are suggested to be important in describing CMSCC. We conclude that using data from only six milk sessions before the CMSCC sample event is sufficient to obtain a relatively high degree of explanation (R2adj = 0.78). To evaluate the practical applicability of these results, we recommend investigating whether it is possible to predict CMSCC using GAM.

Keywords

Automatic milking; Quarter conductivity; Somatic cell count; Udder health

Published in

Title: Precision Livestock Farming '17 : Papers Presented at the 8th European Conference on Precision Livestock Farming : Nantes, France 12-14 September ‘17
Publisher: Organising Committee of the 9th European Conference on Precision Livestock Farming (ECPLF), Teagasc, Animal and Grassland Research and Innovation Centre

Conference

9th European Conference on Precision Livestock Farming, ECPLF 2019, Cork, Ireland, 260819 - 290819

SLU Authors

• Emanuelson, Ulf

  • Department of Clinical Sciences, Swedish University of Agricultural Sciences
    

UKÄ Subject classification

Animal and Dairy Science

Publication identifier

• ISBN: 9781841706542

Permanent link to this page (URI)

https://res.slu.se/id/publ/129911