WELL-CALF: optimising accuracy for commercial adoption

奶牛群的小牛是奶牛场和奶牛业的基础动物，但有几种疾病是地方性的。牛呼吸道疾病（BRD）影响了11%的小牛，是导致10个月以下牛表现不佳（和死亡）的主要原因，每年给英国农民造成约80万英镑的损失。（兽医治疗，降低终身生产力，死亡率；Zoetis,2022）。热应激会影响免疫状态，使小牛更容易生病（Vet Sci,2020）。整合小牛整个生命周期的健康和生产数据对该行业来说是一个重大挑战，因此目前的管理决策是在没有事先了解农场或动物疾病或生产状况的情况下做出的。宁德时代用于健康监测的自动化监测解决方案改善了结果，但存在进一步优化的潜力。WELL-CALF项目的最初目标是(i)开发和实施一个数据平台，整合动物整个生命周期的健康和生产参数；（ii）为小牛饲养单位开发一个创新的农业工程系统，集成一个具有性能数据的新型多传感器平台，以监测小牛的健康状况；（iii）利用先进的数据分析对个体的健康状况进行分类，以促进快速、个性化的治疗。在这个项目中，我们的目标是通过使用先进的机器学习技术，改进牛呼吸道疾病的预测模型，从而扩展和改进WELL-CALF项目。

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.