...
首页> 外文期刊>Applied Animal Behaviour Science >Development of scoring systems for abnormal rising and lying down by dairy cattle, and their relationship with other welfare outcome measures
【24h】

Development of scoring systems for abnormal rising and lying down by dairy cattle, and their relationship with other welfare outcome measures

机译:奶牛异常上升和躺下的评分系统的发展及其与其他福利结果措施的关系

获取原文
获取原文并翻译 | 示例
           

摘要

The objective of this study was to develop two scoring systems to measure the extent of abnormal rising and lying-down behaviors in dairy cows, and to assess the relationship of the behaviors with other welfare outcome measures. Forty-eight lactating Holstein cows were randomly selected in a tie-stall house and enrolled over 2 replicates. Each replicate was video recorded for 24 h, once per week for 10 consecutive weeks. In weeks 1, 2, and 3 of replicate 1, all rising and lying-down events were identified from the video recordings by a single trained observer. Behaviors recorded while rising included duration of rising (sec), attempts at rising, presence (1) or absence (0) of contact with environment, backward movement on carpal joints, delayed rising, and horse rising. Behaviors recorded while lying-down were duration of intention movements (sec), duration of lying-down (sec), attempts at lying-down, presence (1) or absence (0) of contact with environment, hindquarters shifting, slipping, and dog-sitting. An event classifier was also used to categorize each rising and lying-down event as normal (0) or abnormal (1) based on the presence of at least one abnormal behavior. Averaging scores for each behavior and event classifier from 4 daytime events and 2 night-time events was sufficient to yield a strong correlation with full 24 h averages for each behavior and event classifier (r(s) >= 0.82). Three independent observers were trained to score all behaviors and the event classifier with high repeatability (>= 0.80 kappa). The observers were each assigned a portion of weeks 2, 3, 6, 8, and 10 of replicates 1 and 2 to score the daily percentage occurrence of each behavior and event classifier using the outlined regime. Lameness status and injury were also assessed live by trained observers, and lying time was automatically recorded in weeks 2, 3, 6, 8, and 10 of replicates 1 and 2. Multivariate mixed models were used to assess the association of each behavior with other measures, including of welfare. A subset of the behaviors was positively associated with injury severity and body size, and negatively associated with lying time (P <= 0.05). Both scoring systems were highly repeatable tools for assessing the extent of abnormal rising and lying-down behaviors in dairy cows, and were associated with other welfare outcomes indicative of cow comfort.
机译:本研究的目的是制定两种评分系统,以衡量奶牛异常上升和卧下行为的程度,并评估与其他福利结果措施的行为关系。在Tie-Stall房屋中随机选择四十八个哺乳期荷斯坦奶牛,并注册了2次重复。每次复制为34小时的视频,连续10周每周一次。在第1,2和3周内复制1中,通过单个训练的观察者从视频录制中识别所有上升和躺下的事件。记录的行为在上升的情况下包括上升的持续时间(秒),尝试升高,存在(1)或缺席(0)与环境的联系,落后的腕表,延迟上升和马升起。在躺下时令人记录的行为是有意运动的持续时间(SEC),躺下的持续时间(秒),躺下的尝试,存在(1)或缺席(0)与环境接触,后躯移位,滑倒和狗坐。事件分类器还用于基于至少一种异常行为的存在,将每个上升和卧本事件分类为正常(0)或异常(1)。每个行为和事件分类器的平均分数来自4个日间事件和2夜事件足以为每个行为和事件分类器的完整24小时平均产生强烈的相关性(R(s)> = 0.82)。培训三个独立的观察者,以获得所有行为和具有高重复性的活动分类器(> = 0.80 kappa)。每个观察者分配了一部分的一部分时间2,3,6,8和10的重复1和2,以使用概述的制度对每个行为和事件分类器的每日百分比发生。还通过培训的观察者评估跛足状态和损伤,并在第2,3,6,8和10周内自动记录时间,3,6,8和10的重复1和2.多变量混合模型用于评估每个行为与其他行为的关联措施,包括福利。行为的子集与损伤严重程度和体积呈正相关,与躺卧时间负相关(P <= 0.05)。评分系统都是高度可重复的工具,用于评估奶牛的异常上升和卧下行为的程度,以及与牛舒适的其他福利结果有关。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号