A Machine Learning and Optimization Framework for the Early Diagnosis of Bovine Respiratory Disease

Bovine Respiratory Disease (BRD) is an infection of the respiratory tract that leading reason for antimicrobial use in dairy calves and represents 22% calf mortalities. The costs effects BRD can severely damage a farm’s economy, since raising one largest economic investments, diagnosing requires intensive specialized labor hard to find. Precision technologies based on Internet Things (IoT), such as automatic feeders, scales, accelerometers, help detect behavioral changes before outward clinical signs BRD. Such early detection enables treatment, thus faster recovery, with less long term effects. In this paper, we propose framework diagnosis, its detection, identification persistency status using precision IoT technologies. We adopt machine learning model paired cost-sensitive feature selection problem called Cost Optimization Worth (COW). COW maximizes prediction accuracy given budget constraint. show COWis NP-Hard, efficient heuristic polynomial complexity Cost-Aware Learning Feature (CALF). validate our methodology real dataset collected from 159 during preweaning period. Results approach outperforms recent state-of-the-art solution. Numerically, achieve 88% labeling sick healthy calves, 70% are predicted 4 days prior 80% detected within first five sickness.