WELL-CALF: optimising accuracy for commercial adoption

奶牛群中的小牛犊是奶牛场和奶牛业的基础动物，但有几种疾病是地方病。牛呼吸道疾病(BRD)影响了11%的小牛，是导致10个月龄牛生产性能不佳(和死亡)的主要原因，导致英国农民每年损失约80 Mpa(兽医治疗，降低终身生产力，死亡率；Zoetis，2022)。热应激影响牛犊的免疫状态，使其更容易患病(Vet Sci，2020)。整合牛犊一生中的健康和生产数据对该行业来说是一个巨大的挑战，因此目前在不事先了解农场或动物的疾病或生产状况的情况下做出管理决策。CATL用于健康监测的自动化监测解决方案改善了结果，但仍有进一步优化的潜力。Well-calf项目的最初目标是(I)开发和实施一个整合整个动物一生的健康和生产参数的数据平台，(Ii)为犊牛饲养单位开发一个创新的农业工程系统，将新的多传感器平台与性能数据集成在一起来监测犊牛的健康状况，以及(Iii)利用先进的数据分析来对个体的健康状况进行分类，以促进快速、个性化的治疗。对于这个项目，我们的目标是通过完善牛呼吸道疾病的预测模型来扩展和改进Well-calf项目，使用先进的机器学习技术。

Calves from dairy herds are the foundation animals for dairy farms and the dairy-beef industry, but several diseases are endemic. Bovine respiratory disease (BRD) affects 11% of all calves and is the leading cause of poor performance (and mortality) in cattle <10 months of age, costing UK farmers ~£80Mp.a.(veterinary treatments, reduced lifetime productivity, mortality;Zoetis,2022). Thermal stress impacts the immune status and leaves calves more susceptible to illness(Vet Sci,2020).Integrating health and production data across a calf's life represents a significant challenge to the industry, thus management decisions are currently made without any prior knowledge of a farm or animal's disease or production status. CATL's automated monitoring solutions for health surveillance improves results but further potential for optimisation exists.The initial aims of the WELL-CALF project were to (i) develop and implement a data platform integrating health and production parameters across animal lifetime, (ii) to develop an innovative agri-engineering system for calf-rearer units integrating a novel multi-sensor platform with performance data to monitor calf health status, and (iii) to utilise advanced data analytics to classify health conditions in individuals, to facilitate rapid, individually-tailored treatment.For this project, we aim to extend and improve on the WELL-CALF project by refining prediction models for bovine respiratory disease, using advanced machine learning techniques.