0


流感样疾病数据的地方到全球分析

Local-to-global analysis of influenza-like-illness data
课程网址: http://videolectures.net/sikdd2019_costa_analysis_of_influenza/  
主讲教师: João Pita Costa
开课单位: Jožef Stefan研究所人工智能实验室
开课时间: 2019-11-14
课程语种: 英语
中文简介:
在这个社会互联的时代,对适当、强大和有效的疫情情报工具的需求日益增加。流感等全球卫生倡议可能在公共卫生的未来发挥核心作用。本文介绍了对流感网络倡议的贡献,描述了基于Elasticsearch的本地中心及其数据源的新监测系统。通常情况下,国家公共卫生机构优先考虑探索内部生成的数据,因此无法在全球流感平台中解决。该平台可供没有编程专业知识的卫生专业人员使用,以鼓励和增强他们在繁忙的内部IT部门中的独立性,并进一步提高他们自己研究的有效性。然后,可以考虑将最有意义的数据可视化模块集成到完整的Influenzanet平台中,以服务于整个网络,从而在全球层面进行合作。通过这种方法,我们还表明了一个积极的中心在针对自己的优先事项开展调查方面的重要性。在这方面,作为一个例子,我们还描述了使用葡萄牙ILI季节(2005年至2013年)将最先进方法应用于本地数据集的新结果。本研究基于Streamstory方法的应用。它旨在展示这种多用途方法在以下方面的潜力:(i)确定数据驱动的ILI季节;(ii)将ILI发生率与天气数据的维度相关联;和(iii)比较四种不同ILI定义中的发病率。
课程简介: The need for appropriate, robust and efficient epidemic intelligence tools is increasing in this age of a connected society. Global health initiatives, such as Influenzanet, potentially have a central role in the future of Public Health. This paper presents the contributions to the Influenzanet initiative, describing a new monitoring system for local hubs and their data sources, based on Elasticsearch. It is often the case that the exploration of internally generated data is prioritised by national public health institutions, and therefore cannot be addressed in the global Influenzanet platform. This platform can be used by health professionals without programming expertise to encourage and enhance their independence from busy in-house IT departments and further contribute to the effectiveness of their own research. The most meaningful data visualization modules can then be considered for integration into the full Influenzanet platform that will serve the complete network, thus collaborating at a global level. With this approach we also show the importance that an active hub in carrying out its own investigations towards its own priorities. In that regard and as an example, we also describe new results on the application of state-of-the-art approaches to a local data set, using the Portuguese ILI seasons between 2005 and 2013. This study is based on the application of the Streamstory approach. It aims to show the potential of this versatile approach in: (i) identifying data-driven ILI seasons; (ii) relating the ILI incidence to the dimensions of weather data; and (iii) comparing the incidence throughout four different ILI definitions.
关 键 词: 流感样疾病数据; 数据挖掘; 地方到全球分析
课程来源: 视频讲座网
数据采集: 2022-09-14:cyh
最后编审: 2022-09-19:cyh
阅读次数: 29