Click on “Download PDF” for the PDF version or on the title for the HTML version.
If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.
Activity analysis to detect lameness in pigs with a UHF-RFID system
Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.orgCitation: 10th International Livestock Environment Symposium (ILES X) .(doi:10.13031/iles.18-068)
Authors: Anita Kapun, Felix Adrion, Eva Gallmann
Keywords: Behavior, fattening pigs, lameness, monitoring, UHF RFID
Abstract. Analyzing animals‘ behavior can lead to conclusions on their health status. The objectives of this paper are 1) to monitor the behavior of growing-finishing pigs throughout the fattening period with an ultra-high frequency radio frequency identification system (UHF-RFID) and 2) to detect lameness of individual animals with a calculated measure of activity, so-called “virtual walking distances”. The RFID system consisted of UHF readers and antennas at specific hotspots in the pen, as well as UHF transponder ear tags for each pig and a software for data processing. For this study, 300 pigs were monitored with the RFID system over three fattening periods. The pigs were visually examined for their health condition twice a week. The “virtual walking distances” were calculated from the visiting order of the different hotspots and the distances between them. Because of the high variability of the daily virtual walking distance between different pigs, this measure has to be regarded individually for each pig to detect a beginning or ongoing lameness. For this purpose, a linear moving window regression model was tested on the data of the 300 pigs in this study. Lameness events, which were associated with a major difference between predicted and actual value, were detected by the model for 32% of the actual lame pigs. In other lameness cases, a drop of the graph in general was visible. Unfortunately, about 45% of the pigs with a severe lameness were not detectable due to a lack of change in virtual walking distance. In the next step, several adjustments and additions to the model will be tested to improve the detection rate of lameness events.(Download PDF) (Export to EndNotes)