Machine-Learning Approach for Risk Estimation and Risk Prediction of the Effect of Climate on Bovine Respiratory Disease

Bovine respiratory disease (BRD) is a major cause of illness and death in cattle; however, its global extent distribution remain unclear. As climate change continues to impact the environment, it important understand environmental factors contributing BRD’s emergence re-emergence. In this study, we used machine-learning models remotely sensed data at 2.5 min (21 km2) resolution layers estimate risk BRD predict potential future distribution. We analysed 13,431 cases from 1727 cities worldwide between 2005 2021 using two models, maximum entropy (MaxEnt) Boosted Regression Trees (BRT), geographical globally with varying model parameters. Different re-sampling regimes were visualise measure various sources uncertainty prediction performance. The best-fitting was assessed based on area under receiver operator curve (AUC-ROC), positive predictive power Cohen’s Kappa. found that BRT had better compared MaxEnt. Our findings showed favourable habitats for occurrence associated mean annual temperature, precipitation coldest quarter, diurnal range minimum temperature month. Similarly, not limited currently known suitable regions Europe west central Africa but extends other areas, such as Russia, China Australia. This study highlights need surveillance early detection systems prevent spread across borders. also underscore importance bio-security livestock sector interventions, policy-making farmer education, address animal diseases emergencies new areas.