Lameness is a common problem in breeding sows, which often goes undetected for longperiods with severe consequences for animal welfare and farm productivity. Automaticlameness detection could help pig farmers to recognise and treat lameness sooner. TheSowSIS (Sow Stance Information System) is a device consisting of four force plates andproviding non-invasive force measurements per leg of the sow. In this study, the SowSISwas built into electronic sow feeders and validated for lameness detection in group-housedgestating sows. Data was automatically collected for 71 sows. Visual gait scoring wasperformed twice a week using a 150-mm tagged visual analogue scale to determine thesows’ lameness status. Only data from 32 gait scoring days were included, adding up to 674sow days. A sow was classified as lame using>60 mm as the cut-off value for the visual gaitscores. Stance variables were calculated from the SowSIS data per sow per day. First, amultivariable linear mixed model was used to detect lameness, using stance variables withsignificant influence on the gait score. The model’s performance was 78.5% sensitivity,81.4% specificity, 80.7% accuracy, 57.4% lame predictive value and 92.2% non-lame pre-dictive value. Second, five types of classification models were tested to determine the lameleg on a sub-dataset. The random forest model could predict the lame leg correctly 90% ofthe time. The SowSIS shows great promise as an on-farm lameness detection system, as itallows continuous non-invasive data collection in a practical setting.
|Status||Gepubliceerd - 1-apr-2021|