Bonestroo, John
- Institutionen för kliniska vetenskaper, Sveriges lantbruksuniversitet
- DeLaval International AB
- Wageningen University & Research
Konferensartikel2022Vetenskapligt granskad
Bonestroo, J.; van, der, Voort, M.; Hogeveen, H.; Emanuelson, U.; Klaas, I.C.; Fall, N.
Knowing which mastitis cases are at risk of becoming chronic as early as possible during a (subclinical) episode would be helpful in limiting transmission of chronic mastitis, and unnecessary culling. The Online Cell Counter (OCC) enables the collection of data on Somatic Cell Count (SCC) at each milking. The aim of this study was to develop a forecasting model of mastitis chronicity after the initial increase in SCC and to examine predictive performance of such a model. We used sensor data from 14 European and North American dairy farms with an automatic milking system (AMS) and an OCC (DeLaval International AB). Chronicity was defined as the lack of a structural decrease below 200,000 SCC/ml in 50 days after the day at which the prediction was performed. This prediction was performed using OCC data from 30 days prior to the day where the forecast was made. The label (i.e. to-be-predicted status) indicates whether the cow would recover or turn chronic. A random forest classification model was trained on data from seven randomly selected farms and the data of the remaining seven farms were used to estimate the predictive performance. These results were compared with a default approach that approximated how farmers would diagnose chronicity with monthly SCC data. On average, the model outperformed the default approach on all farms based on accuracy, Matthew’s Correlation Coefficient, sensitivity, and specificity. This study shows that it is possible to predict the mastitis chronicity status with high accuracy using past SCC data from the OCC.
automatic milking system; chronic mastitis; udder inflammation
Titel: Precision Livestock Farming '22 papers presented at the 10th European Conference on Precision Livestock Farming
ISBN: 9788396536006Utgivare: Organising Committee of the 10th European Conference on Precision Livestock Farming (ECPLF), University of Veterinary Medicine Vienna
10th European Conference on Precision Livestock Farming, ECPLF 2022
Husdjursvetenskap
Klinisk vetenskap
https://res.slu.se/id/publ/129706