Discovery and Visualization of New Information from Clinical Reports in the EHR

描述（由申请人提供）：电子健康记录（EHR）系统通过减少处方和计算机化订购中的冗余来改善患者护理，但矛盾的是，也会产生其他类型的信息冗余，导致信息过载。这为临床医生提供安全有效的护理提出了挑战，特别是对于需要在漫长的病史中综合许多临床要素的复杂患者。我们假设可以通过改进自动检测新信息的方法、促进实践中新信息的可视化和优化电子病历临床记录界面来支持提供者使用临床记录。虽然人们对支持循证医学很感兴趣，但很少有人关注如何帮助临床医生为个体患者导航和合成越来越多的电子数据。非结构化叙事文本是现代电子病历的重要组成部分。文本允许临床医生以一种容易被他人理解的方式传达复杂和微妙的信息。虽然分析一组病人的笔记可能是艰巨的，但它对于做出诊断和治疗决定是必要的。目前，这一过程受到许多因素的阻碍，包括这些文本中大量冗余信息，文档数量增加，用户界面（UI）设计不理想，以及与患者互动的时间有限。我们迫切需要

DESCRIPTION (provided by applicant): Electronic Health Record (EHR) systems improve patient care by reducing redundancy in prescribing and computerized ordering but paradoxically also generate other types of information redundancy that lead to information overload. This presents a challenge for clinicians in providing safe and effective care especially with complex patients requiring synthesis of many clinical elements across a lengthy medical history. We hypothesize that provider usage of clinical notes can be supported through refinement of automated methods to detect new information, facilitation of new information visualization in practice, and EHR clinical note interface optimization. While there is much interest in supporting evidence-based medicine, little attention has been given to assisting clinicians in navigating and synthesizing growing amounts of electronic data for individual patients. Unstructured narrative text is an important part of modern EHRs. Text allows clinicians to communicate complex and nuanced information in a manner that is easily comprehended by others. While analyzing a collection of patient's notes can be formidable, it is necessary for making diagnostic and therapeutic decisions. Currently, this process is hindered by many factors, including large amounts of redundant information in these texts, increasing numbers of documents, suboptimal user interface (UI) design, and limited time to interact with patients. There is a critical need to optimize the use of EHR clinical notes for providers, which we propose to address in three aims: 1) Refine computational methods to identify new information in clinical notes, 2) Assess the effect of visualizing new information in clinical notes in an inpatient hospitalist setting, and 3) Discover elements of a rationally designed EHR graphical UI to facilitate clinical document usage in practice. Successful accomplishment of these aims will lay a foundation to make clinicians more efficient, improve decision-making, decrease cognitive load, and potentially increase clinician satisfaction associated with using clinical documents in EHR systems.