Increased piglet survival enabled by AI technology

Piglet survival is an important trait both economically and from an animal welfare perspective. In 2022, the average piglet mortality from birth to weaning was 12%, which means that there are opportunities to reduce this further. Important factors affecting piglet survival are the sow's exterior and movement, as well as maternal behaviour in the farrowing pen both around farrowing and during lactation. These traits are low heritable and is often measured manually. This increases the likelihood of subjectivity and high costs associated with registration. This project will use artificial intelligence to record maternal ability traits automatically and objectively. This will be done with multimodal analyses with different input parameters. Models to detect these traits will be developed based on video and audio data from farrowing pens. This allows us to record new traits related to maternal ability automatically, objectively and with higher precision. This can improve the heritability and increased genetic progress for these traits and hence increase piglet survival. By including audio, we can capture traits related to communication between mother and offspring. This allows us to understand more of the vocal communication. The traits that will be investigated are selected based on literature studies that will identify maternal ability traits of the sow that increase piglet survival. These can be exterior traits that provides functional animals with good locomotion and longevity, and traits related to farrowing, such as nest-building behavior and the farrowing process. When the traits are recorded on a sufficient number of animals, genetic analyses will be performed to detect genetic variation in these traits. We will also investigate genetic correlations to other traits in the breeding goal and investigate whether they are suitable for implementation in Norsvin's breeding program, to directly improve piglet survival and animal welfare.

Animal welfare traits have been included in Norsvin’s breeding programme since 2001, and maternal ability was included in 2010. Currently, mothering ability traits, including piglet survival and exterior score, have a low heritability. In addition, certain traits are scored by employees, with risk of subjectivity and being high cost when recording at scale. This results in a low response to selection as observed in the last years. This project aims to develop multi-modal analysis methods for obtaining, objectively and automatically, phenotypes describing positive mothering abilities. By doing so, we expect to get improve specificity of the phenotypes, and thereby improve the heritability, genetic progress, and piglet survival. We will focus on measuring traits that have been identified in the literature as positively correlated with piglet survival, like gait-related traits and farrowing related traits. We will acquire data with multiple modalities to better capture the complexity of the traits: sow communication will require combining audio and video of the free-farrowing pen, whereas leg conformation will require lateral capture of the animal. By using existing and developing new analysis methods based on machine learning, we will extract high quality phenotypes that will reflect the traits of interest. Finally, we will be analysing these phenotypes against genomic data for these animals, with the final objective to include these phenotypes in our breeding programme to directly increase piglet survival and thus animal welfare.