Analysis of image-based sow activity patterns reveals several associations with piglet survival and early growth

Abstract

An activity pattern describes variations in activities over time. The objectives of this study are to automatically predict sow activity from computer vision over 11 days peripartum and estimate how sow behavior influences piglet's performance during early lactation. The analysis of video images used the convolutional neural network (CNN) YOLO for sow detection and posture classification of 21 Large White and 22 Meishan primiparous sows housed in individual farrowing pens. A longitudinal analysis and a clustering method were combined to identify groups of sows with a similar activity pattern. Traits under study are as follows: (i) the distribution of time spent daily in different postures and (ii) different activities while standing. Six postures were included along with three classes of standing activities, i.e., eating, drinking, and other, which can be in motion or not and root-pawing or not. They correspond to a postural budget and a standing-activity budget. Groups of sows with similar changes in their budget over the period (D-3 to D-1; D0 and D1–D7) were identified with the k-means clustering method. Next, behavioral traits (time spent daily in each posture, frequency of postural changes) were used as explanatory variables in the Cox proportional hazards model for survival and in the linear model for growth. Piglet survival was influenced by sow behavior on D-1 and during the period D1–D7. Piglets born from sows that were standing and doing an activity other than drinking and eating on D-1 had a 26% lower risk of dying than other piglets. Those born from sows that changed posture more frequently on D1–D7 had a 44% lower risk of dying. The number of postural changes, which illustrate sow restlessness, influenced piglet growth in the three periods. The average daily gain of piglets born from sows that were more restless on D1–D7 and that changed posture more frequently to hide their udder on D0 decreased by 22 and 45 g/d, respectively. Conversely, those born from sows that changed posture more frequently to hide their udder during the period of D1–D7 grew faster (+71 g/d) than the other piglets. Sow restlessness at different time periods influenced piglet performance.

Sow behavior on the day of farrowing: The main determinant of early piglet growth among maternal ability traits

“…In a study focused on the 24 hours around farrowing, Illmann et al 21 showed that the number of postural changes at farrowing was positively correlated with piglet growth on the rst day after birth. In the same population as in the current study, we showed that early growth (to D7) of piglets from sows that change postural more often to hide the udder on the day of farrowing is lower than that of sows that do not hide the udder 22 . While their growth is on average 250 g/d over lactation, the average weight gain of pure Meishan piglets during the rst week after birth was 110 g/d, and that of pure Large White piglets was 160 g/d.…”

“…Conversely, the two breeds do not differ in their general behavior in the days after farrowing, as described in Girardie et al 22 . The same range of variation was observed for postural and standing activity.…”

“…This study is based on a crossbreeding scheme, set up to quantify direct and maternal effects on piglet development during the suckling phase (the present study was carried out on sows in rst parity) and before, during the intrauterine phase (e.g., study 61 on the same sows in second parity). The experimental design is described in detail in Girardie et al 22 . This design included 21 primiparous LW and 22 primiparous MS sows kept in individual farrowing pens at the Le Magneraud INRAE GenESI experimental farm (doi: 10.15454/1.5572415481185847E12).…”

“…Farrowing duration was de ned as the interval between birth of the rst piglet and the last piglet in the litter (min = 30 min and max = 390). An environmental variable, the pen, was recorded:, this variable is divided into four categorical variables (A, B, C and D) 22 according to its position in the maternity room. We designated pen A as the pen that was the nearest to the weighing area in the central corridor and D as the pen that was the furthest from the weighing area.…”

“…The video recording started 3 days before farrowing (D-3) and stopped 7 days after farrowing (at D7 of lactation) for all sows. Sow behavior was extracted from the video recording using convolution neural networks 22 . The automatically registered behaviors were (i) postural: standing (ST), sitting (SI), sternal lying (SL), lateral lying (LL) and lateral lying with udder exposed (LLU); (ii) standing activity: eating, drinking and doing something else (exploring or rooting).…”