TEATIME invited 41 experts from 13 countries to the COST headquarter in Brussels for scientific exchange and intense discussions. In recent years, we have observed that application of home cage monitoring is growing rapidly. There are now a variety of tools and applications that can already be used to track animal activity and even analyze their behavior. It became clear early on, in fact when writing the application for TEATIME, that home cage monitoring carries a huge potential for improving animal welfare.
Many of the invited researchers were already part of TEATIME, but a whole range of external experts were also involved. This was very beneficial, not only because it allows us to spread our message further, but also because it helps us to think outside the box and incorporate these perspectives into our current and future endeavors.
The overarching theme of the meeting was animal welfare and how home cage monitoring can be used to measure and improve it. We heard talks about a variety of biomarkers that can be used to automatically assess animal welfare. Beyond complex behaviors that sometimes may still require manual annotation, some of the existing systems can already measure body weight, temperature changes, respiratory rate, and urination to name just a few. But even the more complex behaviors and bodily signs included in traditional score and care sheets are subject to automatized assessment using deep learning-based solutions developed for data from video surveillance. Interestingly, we were taught that you can actually learn a lot from years of experience in automated behavioral observation of larger animals such as pigs and cattle.
The second day of the meeting was dedicated to data. Even if the last sentence says it all, here are a few teasers: Our data must not only be collected but also named, sorted, stored, managed, provided with metadata and preferably handled according to the FAIR principles (https://www.go-fair.org/fair-principles). If we can really get this right and manage to mine all the data treasures from the data swamp, then we truly would have a lot more time for a real tea time.