There are an estimated 45 million inpatient surgical procedures per year in the U.S., where critical incidents account for 832,500 deaths and between 1,350,000 and 7,650,000 cases of significant harm to the patient. Of these errors, teamwork failures (e.g. misunderstanding procedural instructions and not acknowledging and repeating back drug dosage levels) are a significant factor - up to 70% for medical operations such as surgery. Reducing teamwork failures requires team training using effective protocols; however, few team training opportunities on these protocols exist.

Intellectual merit: To address this deficiency in training, the project will substitute unavailable human team members with mixed reality humans (MRH). MRHs will plug into roles of unavailable human team members to facilitate training in multi-party scenarios. The research team will first develop systems capable of having MRH inhabit training environments alongside human trainees. Then, they will develop novel conversational modeling techniques to support a training experience to be conducted with any combination of human trainees and virtual human teammates. Finally, the team will conduct a set of user studies with the system to explore the effect of mixed reality humans, the impact of mixed reality humans on team dynamics, and the efficacy of team training with mixed reality humans. The result of this work will be effective self-contained, portable MRH systems that integrate into clinical training environments.

Broader impacts: The project will result in important educational tools for team training that will ultimately lower the high social and financial costs of team errors in medicine. Better trained teams make fewer and less serious mistakes, thereby improving patient outcomes. The tools and findings will be integrated into the curriculum of continuing medical education and new hire orientation at the University of Florida. While initially applied and evaluated in a clinical setting, the investigators anticipate that mixed reality human team training will be applicable to other multiparty scenarios in aviation, the military, education, and crisis response.

Agency
National Science Foundation (NSF)
Institute
Division of Information and Intelligent Systems (IIS)
Application #
1161491
Program Officer
William Bainbridge
Project Start
Project End
Budget Start
2012-07-01
Budget End
2018-06-30
Support Year
Fiscal Year
2011
Total Cost
$1,087,380
Indirect Cost
Name
University of Florida
Department
Type
DUNS #
City
Gainesville
State
FL
Country
United States
Zip Code
32611