Decoding calf patterns: advancing automated health monitoring with algorithmic insights

Summary

The Dutch veal sector is the biggest veal producer in the EU. One important concern for the farmers in the current production system is the high morbidity and high mortality rates. The result is often a high use of antibiotics at veal farms. Conventional on-farm health monitoring practice is based on visual appraisal and clinical examinations performed by farmers and veterinarians, which is linked to disadvantages such as identifying sick calves late. The general advances in precision livestock farming (PLF) make it possible to start developing an automated tool for health monitoring in veal calves, to assist the conventional way of calf health monitoring.

This thesis focuses on learning about the patterns in healthy calves, which lay the foundation for the next step of detecting ‘deviations’ of patterns in individual calves, which might indicate the sickness. Furthermore, the author noticed risks linked to the development of PLF-based solutions.