Managing Large Complex Data Streams/Outpatient Practice

描述（由申请人提供）： 便携式医疗设备和电子通信的进步使得能够远程监测患有许多慢性病的患者，包括糖尿病、高血压、哮喘、心力衰竭和慢性抗凝。因此，临床医生每天很快就会被数百个电子结果和消息淹没。临床医生将不再以流水线的方式工作，而将变得更像调度员或空中交通管制员，同时电子监控许多过程。临床医生将不再简单地问，“X夫人怎么样？他们还会问计算机：“在我的2000名病人中，今天哪些需要我的关注？”“无论是临床医生，还是电子病历（EMR）系统，都没有为这种变化做好准备。该项目的目标是导出一组设计原则，在特定系统的背景下进行演示和评估，帮助未来的系统开发人员（包括我们自己）构建用于管理大型复杂数据流的工具，以确保准确，高效和及时检测临床相关模式，并且具有认知管理的使用模式。具体目标是：1）开发可用于后续开发和评估研究的葡萄糖和血压远程监测数据的标记数据集; 2）分析来自糖尿病教育和远程医疗信息学（IDEATel）项目的手指针刺葡萄糖和血压数据的统计特征; 3）开发一组候选警报机制以自动识别临床相关模式; 4）开发用于临床医生审查远程监测数据的交互模型; 5）根据该模型，开发用于呈现数据的原型用户界面;以及，6）评估界面和警报策略的性能。这些目标的实现将代表的初始步骤，就构建和评估的原型软件，大型复杂的数据流和医疗保健提供者之间进行调解。

DESCRIPTION (provided by applicant): Advances in portable medical devices and in electronic communications are enabling the remote monitoring of patients with many chronic conditions, including diabetes, hypertension, asthma, heart failure and chronic anticoagulation. As a result, clinicians will soon be inundated by hundreds of electronics results and messages every day. The clinician will no longer function in an assembly-line fashion, but will become more like a dispatcher or air-traffic controller, electronically monitoring many processes simultaneously. Clinicians will no longer ask simply, "How is Mrs. X?" They will also ask the computer "Among my 2000 patients, which ones need my attention today?" Neither clinicians, nor electronic medical records (EMR) systems, are prepared for this change. The goal of this project is to derive a set of design principles, demonstrated and evaluated in the context of specific systems, that helps future system developers (including ourselves) to construct tools for the management of large and complex data streams in a way that assure accurate, efficient, and timely detection of clinically relevant patterns, and that has a mode of use that is cognitively manageable. The specific aims are to: 1) develop a tagged dataset of glucose and blood pressure remote monitoring data that can be used for later development and evaluation studies; 2) analyze the statistical characteristics of the fingerstick glucose and blood pressure data from the Informatics for Diabetes Education and Telemedicine (IDEATel) project; 3) develop a set of candidate alerting mechanisms to automate the identification of clinically relevant patterns; 4) develop an interaction model for clinician review of remote monitoring data; 5) from this model, develop prototype user interfaces for presenting the data; and, 6) evaluate the performance of the interfaces and alerting strategies. The accomplishment of these aims will represent the initial steps with respect to the constructing and evaluating of prototype software to mediate between large complex data streams and health care providers.