The non-invasive reading of body temperature and pulse rate by means of imaging techniques facilitates the regular (and even constant) monitoring of livestock. We are conducting a series of tests on the dairy herd of The Research Institute for Farm Animal Biology (FBN) at Dummerstorf, in the course of which we collect image data. Hybrid methods based on established analysis procedures and innovative AI approaches are being developed for the reliable interpretation of this type of vital data. These methods are intended to identify health problems such as metabolic and respiratory diseases, fever, cardiovascular problems, stress and pain at an early stage. The development work also includes determining the appropriate quality range of the hardware employed in order to establish from the outset the right balance between optimal measurement accuracy and affordability for the market. In the long term, such data can be stored in animal medical records and used in the further development of predictive models for early disease detection.
In order to drive forward the acceptance and adoption of the technology, it will be made available through established B2B partners who integrate our image-based analytics into their hardware that is widely and commonly used in the livestock sector.