EAGER: Data Analysis for Nursing Care Assistance

美国长期患者护理成本的上升、护士的短缺以及老年人数量的增加，使得研究用于老年人/患者护理的智能系统的可能性势在必行。其中一个挑战是如何预防跌倒，跌倒往往会导致严重受伤，以及相当大的医院和病人费用。在这个探索性项目中，Pi和她的团队将专注于分析患者动作的感官数据，努力开发自动检测和预测老年患者跌倒的算法。他们的目标是很好地理解如何将多模式感觉数据与跌倒的领域知识相结合来描述跌倒前患者的行为，以确定开发包含机器学习算法的自动警报系统的可行性，以通过对潜在跌倒的警告来帮助人类护士和机器人护理人员。更广泛的影响：项目成果将为未来智能系统的开发铺平道路，以减少患者跌倒的发生率，这是一个主要的社会问题。该项目将为研究生研究人员提供丰富的跨学科培训，还将加强北卡罗来纳大学夏洛特分校的现有计划，通过深化女性和少数族裔本科生参与研究，扩大本科生(REU)对计算机和研究经验的参与。

The rising cost of long-term patient care, the shortage of nurses, and the increasing number of seniors in the United States make it imperative to investigate the possibility of intelligent systems for elder/patient care. One challenge is how to prevent falls, which often result in serious injury and considerable hospital and patient costs. In this exploratory project the PI and her team will focus on analyzing sensory data of patient actions in an effort to develop algorithms for the automatic detection and prediction of falls among elderly patients. Their goal is to gain a good understanding of how multimodal sensory data combined with domain knowledge of falls can be used to characterize pre-fall patient actions, in order to determine the feasibility of developing automatic alert systems that incorporate machine learning algorithms to assist human nurses and robotic caregivers by warning of potential falls. Broader Impacts: Project outcomes will pave the way for future development of intelligent systems to reduce the incidence of patient falls, which is a major societal concern. The project will provide a rich spectrum of interdisciplinary training for graduate student researchers, and will also strengthen UNC Charlotte's existing programs in broadening participation in computing and in research experiences for undergraduates (REU) by deepening involvement of women and minority undergraduate students in research.