FW-HTF-RM: Enhancing Future Work of Nursing Professionals through Collaborative Human-Robot Interfaces

这项人类-技术前沿的未来工作(FW-HTF)研究项目推进了护理行业的愿景，在那里协作式人机接口(Chris)可以提高护士的工作效率并减少工作压力。这个项目中的接口广泛地定义为使用先进的传感器、移动计算和显示设备在人和智能机器(如机器人)之间的任何链接。在这个项目中，INTERCES将智能地“适应”，在未来的医疗环境中为护士和患者提供身体和认知方面的帮助。该项目将追求四个目标。该团队将制定护理任务的分类，以确定哪些任务可以合理地委托给智能机器人。然后，该团队将比较两种新型Chris在机器人辅助行走任务(即防止摔倒)中促进稳定和有效的人-机器人共享控制的能力。他们还将调查CHRI推荐系统，该系统协调患者坐着的任务，如生命体征监测和在几个护士和移动机械手之间提取物品。这里的目标是确定每组患者和护士的最佳机器人助手数量。最后，该团队将评估该技术对护士、患者和医疗机构的社会和经济影响。该项目将推动科学进步，促进国民健康，为设计未来的护理助理机器人提供蓝图，为适应机器人的医疗设施设计提供信息，并将智能机器人助理的使用指导推进到正式的护理教育、护士培训和资格认证中。该项目的其他潜在好处包括在路易斯维尔大学开发机器人学和机器学习方面的教学计划，让护理本科生参与这项研究，以及通过俄克拉荷马州立大学卫生系统创新中心扩展到农村初级保健诊所和医院环境。该项目包括四个目标：行为观察、文档审查、任务清单和危急事件将被分析以制定任务、技能和背景分类，以确定可以分配给智能机器人护士助理的护理任务；团队将开发两个Chris，然后将他们在辅助行走任务中的表现与人类护理人员的能力进行比较。这些接口利用神经网络和通用算法，根据用户的心理生理和触觉信号进行调整，旨在允许新手护士和患者操作带有可穿戴传感器的机器人，以防止摔倒。然后，该团队将使用协作过滤、混合推荐和机器学习技术来增强实体Chris的信息能力。这里的目标是开发一种智能推荐系统，该系统将提高机器人助手的部署效率，这些助手能够执行患者坐着的任务，如生命体征监测和物品提取。该项目中的接口将由大约150名专家和新手用户、护生和模拟患者进行评估，以促进对哪些类型的任务更好地分配给人，哪些可以在护理场景中委托给机器人的理解。最后，该团队将通过O*Net数据库对该技术对护理成本的影响和未来医疗行业的技能需求进行经济分析。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

This Future of Work at the Human-Technology Frontier (FW-HTF) research project advances a vision for the profession of nursing where collaborative human-robot interfaces (CHRIs) can enhance nurse productivity and reduce on-the-job stress. Interfaces in this project are broadly defined as any link between humans and intelligent machines such as robots, using advanced sensors, mobile computing, and display devices. In this project, interfaces will intelligently “adapt” to provide both physical and cognitive assistance to nurses and patients in future healthcare environments. The project will pursue four objectives. The team will develop a taxonomy of nursing tasks to determine those that can be justifiably delegated to intelligent robots. The team will then compare the ability of two novel CHRIs to facilitate stable and effective shared human-robot control of a robot-assisted walking task (i.e., fall prevention). They will also investigate a CHRI recommender system that coordinates patient sitting tasks such as vital signs monitor and item fetching among several nurses and mobile manipulators. The goal here is to determine the optimal number of robotic assistants per group of patients and nurses. Finally, the team will evaluate the social and economic impact of the technology on nurses, patients, and healthcare facilities. The project will promote the progress of science and advance the national health by providing a blueprint for engineering future nursing assistant robots, for informing healthcare facility design to accommodate the robots, and for advancing instruction on the use of intelligent robotic assistants into formal nursing education, nurse training, and credentialing. Other potential benefits of the project include the development of instructional programs in robotics and machine learning at the University of Louisville, involvement of undergraduate nursing students in this research, and outreach to rural primary care clinics and hospital settings through the Center for Health Systems Innovation at the Oklahoma State University.This project includes four objectives: Behavioral observation, documentation reviews, task inventories and critical incidents will be analyzed to develop a task, skill, and context taxonomy to identify nursing tasks that can be assigned to intelligent robotic nurse assistants; The team will develop two CHRIs and then compare their performance in an assisted walking task with the abilities of human nursing staff. These interfaces utilize neural networks and generic algorithms and adjust to psycho-physiological and tactile signals from users and are designed to allow novice nurses and patients to operate robots with wearable sensors for prevention of falls. The team will then enhance the informational capabilities of the physical CHRIs using collaborative filtering, hybrid recommendation and machine learning techniques. The goal here is to develop an intelligent recommender system that will promote efficiency in the deployment of robotic assistants capable of performing patient sitting tasks such as vital signs monitoring and item fetching. The interfaces in this project will be evaluated by approximately 150 expert and novice users, nursing students and simulated patients to advance understanding of which types of tasks are better assigned to people and which may be delegated to robots in nursing scenarios. Finally, the team will perform economic analyses of the impact of the technology on nursing costs and the skilling needs for future healthcare industry through O*NET databases.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.