Bansal Shweta, Chowell Gerardo, Simonsen Lone, Vespignani Alessandro, Viboud Cécile
Fogarty International Center, National Institutes of Health, Bethesda, Maryland.
Department of Biology, Georgetown University, Washington D.C.
J Infect Dis. 2016 Dec 1;214(suppl_4):S375-S379. doi: 10.1093/infdis/jiw400.
We devote a special issue of the Journal of Infectious Diseases to review the recent advances of big data in strengthening disease surveillance, monitoring medical adverse events, informing transmission models, and tracking patient sentiments and mobility. We consider a broad definition of big data for public health, one encompassing patient information gathered from high-volume electronic health records and participatory surveillance systems, as well as mining of digital traces such as social media, Internet searches, and cell-phone logs. We introduce nine independent contributions to this special issue and highlight several cross-cutting areas that require further research, including representativeness, biases, volatility, and validation, and the need for robust statistical and hypotheses-driven analyses. Overall, we are optimistic that the big-data revolution will vastly improve the granularity and timeliness of available epidemiological information, with hybrid systems augmenting rather than supplanting traditional surveillance systems, and better prospects for accurate infectious diseases models and forecasts.
我们在《传染病杂志》特刊中回顾大数据在加强疾病监测、监测医疗不良事件、为传播模型提供信息以及追踪患者情绪和流动性方面的最新进展。我们对公共卫生领域的大数据采用广义定义，涵盖从大量电子健康记录和参与性监测系统收集的患者信息，以及对社交媒体、互联网搜索和手机日志等数字痕迹的挖掘。我们介绍了本期特刊的九篇独立投稿，并强调了几个需要进一步研究的交叉领域，包括代表性、偏差、波动性和验证，以及进行稳健的统计和假设驱动分析的必要性。总体而言，我们乐观地认为，大数据革命将极大地提高现有流行病学信息的粒度和及时性，混合系统将增强而非取代传统监测系统，并且准确的传染病模型和预测的前景更好。
