Novel approaches to improving calf health: new technology and data analytics

保持良好的小牛健康对养殖企业的生产力和可持续性都很重要。牛健康福利组织(GB)的一份报告显示，3.6%的公牛和2.5%的小母牛在1个月龄以下死亡。牛犊健康状况不佳的影响包括牛犊损失、兽医治疗费用和劳动力成本增加。小牛的健康状况不佳也会导致晚年的生长发育较差。如果将动物作为小母牛的替代品饲养，小牛病将对原产地奶牛养殖场产生不利影响，但当小牛被出售给饲养单位时，也将影响来自乳制品的牛肉部门。及早发现疾病将有助于减少这些福利、生产力和经济影响，并减少抗生素的使用。提高动物对疾病挑战的适应能力对于减少对动物和农场的影响也很重要。该项目的目的是探索一些降低牛犊死亡率和发病率的方法。目标1.疾病的行为指标：将计算机生成的诊断与治疗策略结合起来。研究表明，在实验环境中，喂养和活动模式可用于在农民确诊前2天检测健康状况不佳(例如Bowen等人，2019年)。然而，目前尚不清楚使用这种系统是否会降低总体治疗率，也不知道农场工作人员应用这些计算机生成的诊断的准确性有多高。将使用已建立的疾病行为指标模型来创建健康警报。针对小牛治疗组的协议将定义在警报后应采取的治疗行动，并与标准农场卫生治疗做法下的对照组进行比较。还将收集环境数据、生长和摄入量数据以及牛犊历史的关键属性，并将使用先进的数据分析方法(包括机器学习)来评估模型诊断的有效性，以及对牛犊健康结果的治疗方案。有许多新兴技术可以应用于疾病的早期检测。这些包括心率变异性、血氧饱和度或肺功能的评估。试点试验将用于评估这些措施在疾病检测中的有效性和实用性。最有希望的新监测技术(S)将在一项试验中进行评估，比较患病和健康牛犊的反应。将添加额外的数据(如上所述)，并使用机器学习技术来评估这些新方法的诊断效果。目的3.营养层面对疾病挑战反应的影响。通常情况下，小牛会被给予有限的牛奶配额，以鼓励他们转向固体饲料。然而，最近的一项研究表明，处于低营养水平(PON)的牛犊对疾病挑战表现出夸大的反应(Sharon等人，2019年)。这项拟议的研究将调查不同的脑桥对牛群自然疾病挑战后免疫学和行为结果的影响。BSAS年会.Sharon等人，(2019)。《乳品科学》。102：9082-9096

Maintaining good calf health is important for both the productivity and sustainability of farming enterprises. A report by the Cattle Health and Welfare Group (GB)) showed that 3.6% of dairy bull calves and 2.5% of dairy heifer calves died under 1 month of age. The impacts of poor calf health include calf losses, costs of veterinary treatment and increased labour costs. Ill health in calves also leads to poorer growth in later life. Calf disease adversely affects the origin dairy farm if animals are kept as heifer replacements, but will also affect the beef-from-dairy sector when calves are sold to rearer units. Early detection of disease would help reduce these welfare, productivity and economic impacts and also reduce antibiotic use. Improving animal resilience to disease challenge is also important in reducing the impact on the animal and the farm. The aim of this project is to explore a number of approaches to reducing calf mortality and morbidity. Objective 1. Behavioural indicators of disease: coupling computer generated diagnoses with treatment strategies. Research has shown that in experimental settings, feeding and activity patterns can be used to detect ill health up to 2 days prior to farmer diagnosis (e.g. Bowen et al., 2019). However, it is not known whether using such systems will reduce overall treatment rates or how accurate these computer-generated diagnoses are when applied by farm staff. Health alerts will be created using established models of behavioural indicators of disease. A protocol for treatment groups of calves will define what treatment action should be taken after an alert and compared to control groups under standard farm health treatment practices. Environmental data, growth and intake data, and key attributes of the calf's history will also be gathered and advanced data analytics (including machine learning) will be used to assess the effectiveness of the model diagnosis coupled with a treatment programme on calf health outcomes.Objective 2. Novel indicators of disease. There are a number of emerging technologies that could be applied in the early detection of disease. These include assessments of heart rate variability, blood oxygen saturation or lung function. Pilot trials will be used to assess the validity and practicality of these measures in disease detection. The most promising new monitoring technology(s) will be assessed in a trial comparing responses in sick and healthy calves. Additional data (as above) will be added and machine learning techniques used to assess the efficacy of diagnosis of these novel methods. Objective 3. Effect of plane of nutrition on response to disease challenge.Calves are typically fed a restricted milk allowance to encourage them move onto solid feed. However, a recent study indicated that calves on a low plane of nutrition (PON) show exaggerated responses to disease challenge (Sharon et al., 2019). The propsed study will investigate effects of different PONs on immunological and behavioural outcomes following natural disease challenge in calves.References Bowen et al (2019). BSAS Annual Meeting.Sharon et al., (2019). J. Dairy Sci. 102:9082-9096