Deep Learning Assessment of Bedding Conditions and Dairy Cattle Foot Health

Abstract

Introduction: The foot condition is an important criterion affecting the well-being and productivity of dairy cattle. Unfavorable bedding conditions result in increased moisture, bacteria and mechanical stress that contribute to development of hoof diseases, lameness and discomfort. Conventional methods of assessing the foot health state of cows include visual inspection. It takes considerable amount of time and cannot be used to continuously monitor the health of dairy cattle in large-scale facilities. In this paper, we propose a new deep learning-based approach for evaluating the connection between bedding environment and foot health of dairy cows. The suggested methodology included analysis of environmental conditions in the bedding area and cattle foot health indices based on images and behaviour. The obtained data were analyzed by using the principles of deep learning. The obtained results showed that the deep learning models allowed us to assess the differences in the quality of bedding and their effect on foot health of cattle. The study concluded that integrating deep learning with livestock health assessment can provide an efficient solution for precision dairy farming by supporting proactive bedding management, reducing lameness risks, and improving cattle welfare