ITR: An Infrastructure for Veterinary Telemedicine - Proactive Herd Health Management for Disease Prevention from Farm to Market

该项目的目标是研究和开发一个原型，技术丰富的基础设施，以支持通过与兽医健康信息网络交互的远程可穿戴设备经济地持续监测牛的健康状况。无线电台、分布式数据库和生理评估算法将在这种环境中发挥关键作用。这些系统正在提高畜牧业应对和预测疾病发病及其流行病学传播的能力，无论是来自自然事件还是恐怖事件。从这项工作中吸取的趋势分析和健康预测经验教训已应用于人类健康状况预测和非人类疾病传播。研究人员在牛群聚集点（如饲料仓和水槽）附近放置了兼容蓝牙的监测站。这些工作站上传的数据来自环境传感器、动物佩戴的具有全球定位功能的蓝牙设备，以及可穿戴/远程生物医学传感器。在将牧场汇总数据上传到区域数据库之前，本地算法会执行快速数据分析，在区域数据库中，数据与附近生产者提供的信息相关联。重要的发现会立即向地区兽医、生产者和当局公布。为了实现这一目标，研究解决了一些重要的信息技术问题，包括调度算法，该算法自适应地确定数据分析应该发生在哪里，哪些领域需要更深入的分析；具有近实时约束的多无线数据流优先排序算法以及适用于特定领域的数据挖掘技术来发现问题指标。这个跨学科项目正在解决农业和国家社区内的一个关键需求：应用研究，使动物科学行业能够对牛的疾病发病及其流行病学传播作出反应和预测。通过开发分布式软件基础设施、全面的生理监测工具集和新的处理算法（例如，用于数据存储、融合和解释），我们正在帮助改善畜牧业的金融稳定性，同时提高我们对流行病灾害的准备水平，无论是来自自然还是恐怖事件。

The goal of this project is to research and develop a prototype,technology-rich infrastructure to support economical continuous monitoringof cattle state of health via remote and wearable devices that interactwith a veterinary health information network. Wireless stations,distributed databases, and physiological assessment algorithms will play akey role in this environment. These systems are improving the ability ofthe livestock industry to react to and predict disease onset and itsepidemiological spread, whether from natural or terrorist events. Trendanalysis and health prediction lessons learned from this effort haveapplication to human state-of-health prediction and disease spread inhuman populations.The researchers are placing Bluetooth-compliant monitoring stations nearcattle congregation points, such as feed bunks and watering troughs. Thesestations upload data from environmental sensors, Bluetooth-enabled deviceswith global positioning capability worn by the animals, andwearable/remote biomedical sensors. Local algorithms perform rapid dataanalysis prior to uploading ranch summary data to regional databases,where the data is correlated with information provided by nearbyproducers. Significant findings are then immediately broadcast to regionalveterinarians, producers, and authorities. In order to accomplish this, the research addresses a number of important information technology issues, including scheduling algorithms that adaptively determine where data analysis should occur and which areas require more in-depth analysis; prioritization algorithms for multiple wireless data streams with near-real-time constraints; and area-appropriate data mining techniques to find problem indicators.The interdisciplinary project is addressing a critical need within the agriculturaland national communities: applied research that allows the animalsciences industry to react to and predict disease onset in cattle and itsepidemiological spread. Through the development of a distributed softwareinfrastructure, a comprehensive physiological monitoring toolset, and newprocessing algorithms (e.g., for data storage, fusion, andinterpretation), we are helping to improve the financial stability of thelivestock industry while simultaneously raising our level of preparednessfor epidemiological disasters, whether from natural or terrorist events.