Using Acceleration Data to Automatically Detect the Onset of Farrowing in Sows

Abstract

The aim of the present study was to automatically predict the onset of farrowing in crate-confined sows. (1) Background: Automatic tools are appropriate to support animal surveillance under practical farming conditions. (2) Methods: In three batches, sows in one farrowing compartment of the Futterkamp research farm were equipped with an ear sensor to sample acceleration. As a reference video, recordings of the sows were used. A classical CUSUM chart using different acceleration indices of various distribution characteristics with several scenarios were compared. (3) Results: The increase of activity mainly due to nest building behavior before the onset of farrowing could be detected with the sow individual CUSUM chart. The best performance required a statistical distribution characteristic that represented fluctuations in the signal (for example, 1st variation) combined with a transformation of this parameter by cumulating differences in the signal within certain time periods from one day to another. With this transformed signal, farrowing sows could reliably be detected. For 100% or 85% of the sows, an alarm was given within 48 or 12 h before the onset of farrowing. (4) Conclusions: Acceleration measurements in the ear of a sow are suitable for detecting the onset of farrowing in individually housed sows in commercial farrowing crates.

Dynamics of Sows’ Activity Housed in Farrowing Pens with Possibility of Temporary Crating might Indicate the Time When Sows Should be Confined in a Crate before the Onset of Farrowing

“…Measurement of the activity of sows with accelerometers installed in collars indicated, similarly to other sensor technologies, an increase of activity 16 to 20 h before the onset of farrowing [19]. When an accelerometer was installed inside an ear tag, the increase in activity was detected 48 h before the onset of farrowing [17].…”

“…Contrary to Manteuffel et al [23] and Pastell et al [14] we decided not to use accuracy, sensitivity, and specificity as measures of performance of the algorithm that was developed in this research. Instead, we decided to mainly evaluate the performance of the algorithm on the basis of the duration between the time when an alarm was generated and the onset of farrowing, similar to Traulsen et al [17] and on the basis of the associated distribution of alarms. In our opinion, accuracy, sensitivity, and specificity as measures of algorithm performance are difficult to interpret when it comes to farrowing prediction, especially when different authors use various definitions of "true positive" alarms [13,14,23] and, accordingly, there is no consensus at what time before farrowing "true positive" alarm occurs.…”

“…Then, changes in dynamics of the activity of sows were detected as an increased and decreased trend in acceleration data. Such an individual modeling approach, in which models adapt to the activity of individuals, was also applied for prediction of farrowing in research by Pastell et al [14], Traulsen et al [17], and Manteuffel et al [23], whereas in research by Oczak et al [13], Oliviero et al [16], and Cornou and Kristensen [30] group level modeling was applied. Individual modeling of sows' activity for farrowing prediction has an advantage, because the performance of such models depends on similarity in dynamics of sows' activity and is independent from baseline activity level of the animals.…”

“…Infrared photocells and force sensors have been installed in farrowing crates and as the farrowing approached, the measured activity of sows increased slightly, then, more than doubled the day before farrowing [15,16]. Accelerometers were tested for measurement of activity of sows with sensors installed in collars and in tags [17][18][19]. Measurement of the activity of sows with accelerometers installed in collars indicated, similarly to other sensor technologies, an increase of activity 16 to 20 h before the onset of farrowing [19].…”

“…Automated detection of increases in sow activity with the use of sensor technology makes possible the prediction of the onset of farrowing. This could be useful in practical conditions to shorten surveillance intervals by farm staff, and the pen could be prepared for an optimal farrowing (e.g., providing nest-building material and activating a heating source) [17]. However, for selecting the optimal period of temporary crating of sows, there is a need for a reliable method to detect when an individual sow starts and finishes nest-building behaviour.…”