FW-HTF: Human-Machine Teaming for Medical Decision Making

FW-HTF：用于医疗决策的人机协作

• 批准号: 1840088
• 负责人: Suchi Saria
• 金额: $ 150万
• 依托单位: Johns Hopkins University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-01-01 至 2024-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1840088&HistoricalAwards=false
• 关键词: FW; HTF; Human; Machine; Teaming

The Future of Work at the Human-Technology Frontier (FW-HTF) is one of 10 new Big Ideas for Future Investment announced by NSF. The FW-HTF cross-directorate program aims to respond to the challenges and opportunities of the changing landscape of jobs and work by supporting convergent research. This award fulfills part of that aim. Algorithmic advances in artificial intelligence are transforming human work in diverse areas including transportation, finance, national security, and medicine. Machine intelligence presents opportunities to increase human work productivity and the quality of jobs through augmenting human capabilities. Effective teaming between humans and intelligent machines similar to effective human-human teamwork has the potential to yield significant near-term gains. This project explores the challenges of human-machine teaming in medical decision making. Health care is one of the most difficult challenges that the United States is facing. The US spends $3 trillion dollars in health care each year, while medical error is the third leading cause of death. Human-machine cognitive teaming creates a new model of patient care in which providers team with intelligent cognitive assistants to enhance quality of care under time pressure, taxing workloads, and uncertainties in medical conditions. This project explores the potential for effective human-machine teaming to mitigate such challenging problems in health care.Specifically, this project seeks to understand (1) whether human-machine teaming can benefit medical decision making and decision making in other related high stakes domains; (2) the guiding principles for designing effective human-machine teams; (3) barriers that currently exist for building such teams; (4) novel solutions needed to address barriers in order to develop highly performant teams; and (5) the economic and societal impacts of the planned approach for human-machine teaming. Understanding effective human-machine teaming, including the broader implications in the workspace and in human workflows, will contribute to positive transformation of human work. In particular, it is anticipated that the outcomes of this project will result in improvements in hospital utilization and reduction of medical errors. The project integrates multiple disciplinary perspectives, including computer science, medical expertise, health policy, and decision making. The impacts of the research will extend to multiple hospitals in the Baltimore region. Furthermore, the project will engage local high school students in summer research experiences, and the outcomes of the research will be integrated into undergraduate curricula.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.

人类技术前沿工作的未来(FW-HTF)是美国国家科学基金会宣布的未来投资的十大新想法之一。FW-HTF跨部门计划旨在通过支持融合研究来应对不断变化的工作和工作格局带来的挑战和机遇。这一奖项部分实现了这一目标。人工智能在算法上的进步正在改变人类在交通、金融、国家安全和医学等不同领域的工作。机器智能提供了通过增强人类能力来提高人类工作生产率和工作质量的机会。人类和智能机器之间的有效合作类似于有效的人与人之间的合作，有可能在短期内产生显着的收益。这个项目探索了人机协作在医疗决策中的挑战。医疗保健是美国面临的最困难的挑战之一。美国每年在医疗保健上花费3万亿美元，而医疗差错是导致死亡的第三大原因。人机认知协作创建了一种新的患者护理模式，其中提供者与智能认知助手合作，以在时间压力、繁重的工作量和医疗条件的不确定性下提高护理质量。这个项目探索了有效的人机合作的潜力，以缓解卫生保健中的这种挑战性问题。具体地说，这个项目试图了解(1)人机合作是否有助于医疗决策和其他相关高风险领域的决策；(2)设计有效的人机团队的指导原则；(3)目前建立这样的团队存在的障碍；(4)需要解决障碍的新解决方案，以发展高绩效的团队；以及(5)计划中的人机合作方法的经济和社会影响。了解有效的人机协作，包括在工作空间和人类工作流程中的更广泛影响，将有助于人类工作的积极转变。特别是，预计该项目的成果将改善医院利用率，减少医疗差错。该项目融合了多个学科的视角，包括计算机科学、医疗专业知识、卫生政策和决策。这项研究的影响将延伸到巴尔的摩地区的多家医院。此外，该项目将让当地高中生参与暑期研究体验，研究结果将被整合到本科课程中。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。

项目成果

Understanding User Reliance on AI in Assisted Decision-Making

• DOI: 10.1145/3555572
• 发表时间: 2022-11
• 期刊: Proceedings of the ACM on Human-Computer Interaction
• 作者: Shiye Cao;Chien-Ming Huang

How Mock Model Training Enhances User Perceptions of AI Systems

• 发表时间: 2021-11
• 期刊: ArXiv
• 作者: Amama Mahmood;G. Ajaykumar;Chien-Ming Huang

Mitigating knowledge imbalance in AI-advised decision-making through collaborative user involvement

• DOI: 10.1016/j.ijhcs.2022.102977
• 发表时间: 2022-12-23
• 期刊: INTERNATIONAL JOURNAL OF HUMAN-COMPUTER STUDIES
• 影响因子: 5.4
• 作者: Gomez, Catalina;Unberath, Mathias;Huang, Chien-Ming
• 通讯作者: Huang, Chien-Ming

JAWS: Auditing Predictive Uncertainty Under Covariate Shift

• 发表时间: 2022-07
• 作者: Drew Prinster;Anqi Liu;S. Saria