Automatic lameness detection in sows using the Sow Stance Information System (SowSIS), a pilot study

Petra Briene, Olga Szczodry, Pieterjan De Geest, Annelies Van Nuffel, Jürgen Vangeyte, Bart Ampe, Sam Millet, Frank Tuyttens, Jarissa Maselyne

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureC3: Congres abstractpeer review


Lameness is a very prevalent problem in breeding sows, which often goes undetected for long periods of time. This can have severe consequences for animal welfare and has impact on the productive performance of sows. Lameness in sows is often hard to detect as sows are not observed while walking on a daily basis. Automatic detection of lameness in sows could help pig farmers to recognize the problem sooner, allowing them to act and thus possibly preventing the problem from getting worse. The SowSIS consists of 4 force plates, providing output for each leg separately. It is built into a Nedap electronic sow feeder (ESF) to allow continuous data collection without interfering with the sows. Several stance information variables can be extracted from the data. In a previous study comparing single measures of 6 sound and lame sows, weight asymmetry, number of weight shifts and number of kicks were the most promising variables to detect lameness extracted from the SowSIS data. The SowSIS has since been improved by using multiple load cell-mounting for higher accuracy per plate. A pilot test study is currently ongoing where data is continuously collected by 4 separate SowSIS in group-housing for a period of 74 days during gestation. Feeding visits are extracted from the data and linked to the ESF data to allow identification of individual sows. All sows are visually scored for gait twice a week using a 150mm continuous tagged visual analogue scale (tVAS) to determine whether they are lame or not. Case studies of lame and non-lame sows will be selected from the data of 2 production groups containing 16 and 18 sows. These data will be analysed and compared to the visual gait scores to determine the ability of the SowSIS to correctly detect lame sows and to test which of the pre-determined variables are most useful to detect lameness in different cases. The dataset is unique as it provides insight into the development and the course of lameness over a longer period of time in individual breeding sows. The results from several case studies will be presented at the conference.
Oorspronkelijke taalEngels
TitelBook of Abstracts of the 69th Annual Meeting of the European Federation of Animal Science : Book of abstracts No. 24 (2018)
UitgeverijWageningen Academic Publishers
ISBN van geprinte versie978-90-8686-323-5
ISBN van elektronische versie978-90-8686-871-1
PublicatiestatusGepubliceerd - 31-aug.-2018
Evenement 69th Annual Meeting of the European
Federation of Animal Science
- Valamar Resort, Dubrovnik, Kroatië
Duur: 27-aug.-201831-aug.-2018
Congresnummer: 69


Bekijk de onderzoeksthema's van 'Automatic lameness detection in sows using the Sow Stance Information System (SowSIS), a pilot study'. Samen vormen ze een unieke vingerafdruk.

Dit citeren