INTEAM+ Objective Measurements for Team Performance
INTEAM+: A revolution in healthcare education through VR and data-driven feedback for more effective teamwork?
Duration: May 2023 – April 2024
Status: Completed
Educational Level: Tertiary Level
Topic: Data Science for Education, Digital Tools
Keywords: Data Science, Process Data, VR
Initial Situation
Teamwork in emergency situations is crucial. Learning how to work together, for example between nurses and doctors in simulation training, is complex. Feedback, typically from video observations or measurement instruments, is often subjective and not available in real time.
Previous Project
In the previous project INTEAM, efforts focused on fostering interprofessional collaboration between medical and nursing students already during their training. Through joint team training using VR simulations, key competencies such as structured communication, emergency management, and mutual role understanding were to be strengthened, thereby laying the foundation for more effective collaboration in their future professional practice.
Objectives
With INTEAM+, we aim to explore the use of objective metrics for assessing team performance in medical settings. Parameters such as heart rate (ECG), electrodermal activity (EDA), and eye tracking are evaluated for their potential to capture and assess teamwork, coordination, and team leadership.
Method
Team training in virtual reality was conducted and recorded. We are now focusing on analyzing the physiological data from the video recordings. This phase aims to identify fundamental patterns in team communication and dynamics. These findings will not only improve our VR training methods, but also serve as a basis for future projects that examine these patterns in greater detail. Our goal is to use this deeper understanding to make teamwork in healthcare more effective and thus lay the groundwork for further research and development.
Results
We introduce a data-driven framework for the evaluation of team dynamics in emergency medical team training, that enables debriefing by leveraging the use of wearable devices that capture physiological data and motion data during training. The system processes raw signals in real time across multiple timescales to detect team dynamics. A Machine Learning (ML) model then analyzes the processed data, identifying synchronization patterns linked related to team coordination and effectiveness. The system operates on a continual ML paradigm, enabling self-monitoring, self-training, and self-adaptation as new data is collected. Results show a 90% accuracy between the COACT index and subjective team performance ratings, validating the potential of the proposed framework as an objective, AI-driven team assessment tool for emergency.
Implemented Translation
Further follow-up projects are designed to investigate the use of the developed measurements in various acute medical settings and to ensure their practical applicability in the training of medical professionals. This implementation will take place at BFH as a project partner and in the medical studies program at the University of Bern. We hope that our findings will lead to an optimization of interprofessional training for healthcare professions in Switzerland.
Project Lead
Project Collaborators
