The Medical AI & and Cognitive Engineering (MAICE) lab aims to improve patient safety by enhancing clinician performance using advanced technologies such as machine learning and virtual and augmented reality. Their projects have been funded by the NIH, NASA, DoD, NSF, and AHRQ.
Current Projects
Development of an Enhanced Simulation Model with Biometric Cognitive Load Measurements for More Efficient En Route Care Training
Defense Health Agency (DHA)/DoD/US Air Force | 2023-2024
PI/PDs: Maj William Davis (Sant Antonio Military Medical Center), Benjamin Easter (University of Colorado School of Medicine) | Site PIs: Roger Dias (BWH/Harvard), Steven Yule (University of Edinburgh)
The goal of this project is to adapt and validate multimodal individual cognitive load measurements and nontechnical team performance assessments to provide data describing decision-making capacity and team vulnerabilities during En Route Care Training.
A Robot-Assisted Perfusion System to Improve Patient Safety in the Cardiac Operating Room
NIH/NHLBI (R01 HL157457) | 2022-2025
PI/PDs: Roger Dias (BWH/Harvard), Matthew Gombolay (Georgia Tech) | Site PIs: Marco Zenati (VA Hospital/Harvard), Julian Goldman (MGH/Harvard)
The goal of this project is to develop and gather initial validity evidence of a proof-of-concept for a Robot-Assisted Perfusion System (RAPS) that can be integrated into the cardiac surgery workflow as a non-human teammate. The RAPS will support the perfusion team in a way that perfusionists will still keep control of the perfusion system (i.e. human-in-the-loop approach), but cognitively supported by the RAPS.
An AI Coach to Enhance Surgical Teamwork in the Cardiac Operating Room
NSF (22048507) | 2022-2026
PI/PDs: Vaibhav Unhelkar (Rice University) | Co-PIs: Roger Dias (BWH/Harvard), Julie Shah (MIT), Marco Zenati (VA Hospital/Harvard)
The strength of our proposed dedicated Artificial Intelligence Coaching System (AI Coach) tasked with monitoring and managing OR team members’ cognitive states is in its real-time potential to improve teamwork, surgical performance, and clinical outcomes. Our ultimate goal is to develop machine partners that learn from the best human team members how to monitor our performance and participate in real-time to back up our weaknesses and improve human team decision-making.
Novel Assessments of Technical and Non-Technical Cardiac Surgery Quality
NIH/NHLBI (R01) | 2019-2024
PD/PIs: Donald Likosky and Fancis Pagani (Michigan Medicine), Steven Yule (University of Edinburgh) | Co-I: Roger Dias (BWH/Harvard)
This project aims to use machine learning (computer vision) to develop automated and objective metrics of technical and non-technical skills in cardiac surgery.
A Novel Cognition-Based Guidance System to Improve Surgical Safety
NIH/NHLBI (R01) | 2020-2024
PI/PD: Marco Zenati (VA Hospital/Harvard) | Site PIs: Roger Dias (BWH/Harvard), Julian Goldman (MGH), George Avrunin (UMass)
The aim of this project is to design, develop, and validate an intelligence cognitive-guidance system to support cardiac surgery teams during complex procedures, reducing human errors and improving patient safety in the operating room.
Development Of Enhanced Simulation Model with Biometric Cognitive Load Measurements for More Efficient En Route Care Training
Defense Health Agency (DHA) – DoD | 2022-2023
PI/PDs: Maj William Davis (San Antonio Military Medical Center), Benjamin Ester (University of Colorado School of Medicine) | Site PIs: Roger Dias (BWH/Harvard), Steven Yule (University of Edinburgh)
Our intent for this proposal is to adapt and validate multimodal individual cognitive load measurements and nontechnical team performance assessments to provide data describing decision-making capacity and team vulnerabilities in En Route Care Training ERCC).
Completed Projects
Strategies and Tools to Enhance Performance and Patient Safety (TeamSTEPPS)
Agency for Healthcare Research and Quality (AHRQ) | 2021-2022
PI/PDs: Alex Hannenberg (Ariadne Labs) | Co-Is: Roger Dias (BWH/Harvard) and Deborah Navedo (BWH)
The goal of this project is to update the Team STEPPS curriculum. The STRATUS team helped with simulation-based education materials.
A Robot-Assisted Perfusion System to Improve Patient Safety in the Cardiac Operating Room
NIH/NHLBI (R56 HL1574570) | 2021-2022
PI/PDs: Roger Dias (BWH/Harvard), Matthew Gombolay (Georgia Tech) | Site PIs: Marco Zenati (VA Hospital/Harvard), Julian Goldman (MGH/Harvard)
This project aims to develop and gather initial validity evidence of a proof-of-concept for a Robot-Assisted Perfusion System (RAPS) that can be integrated into the cardiac surgery workflow as a non-human teammate. The RAPS will support the perfusion team in a way that perfusionists will still keep control of the perfusion system (i.e. human-in-the-loop approach), but cognitively supported by the RAPS.
CORE-Military: A Virtual Reality Platform for Emergency Care Training and Assessment in Austere Environments
Department of Defense (DoD) Health Agency, STTR Program (W81XWH21P0042) | 2021-2021
PI/PD: Roger Dias
The long-term goal of this project is to develop and validate a high-fidelity mobile-based VR platform (CORE-Military) for frontline service members in austere environments, enabling effective just-in-time emergency care training and personalized assessment through AI-enabled learning analytics.
Mixed Reality (MR) Care-Delivery Guidance System to Support Medical Event Management on Long Duration Exploration Missions
NASA/TRISH | 2020-2022
PI/PD: Roger Dias | Science-PI: Steven Yule
This project is a collaboration between Brigham and Women’s Hospital, Harvard Medical School, and McMaster University. The goal is to develop a mixed reality platform for emergency medicine training and clinical decision support of astronauts during long duration space missions.
Emotional Well-Being Among ED Providers During the COVID-19 Pandemic
BWH Department of Emergency Medicine – Esther Khan Seed Grant | 2020-2021
PI: Roger Dias | Co-Is: Andrew Eyre and Charles Pozner
The aim of this project is to investigate the impact of the COVID-19 pandemic on ED provider well-being and validate digital biomarkers of emotional well-being using objective physiological metrics, such as heart rate variability and EEG.
Improved Teamwork to Decrease Errors and Mitigate their Consequences
CRICO | 2019-2021
PIs: Charles Pozner and Madelyn Pearson | Co-I Roger Dias
The aim of this project to increase the use of the principles of Crisis Resource Management by all providers at Brigham and Women’s Hospital as a tool to minimize communication errors that often contribute to preventable medical errors.
Simulation-Based Countermeasure Development to Mitigate Team and System Vulnerabilities During Medical Event Management on Long Duration Space Missions
NASA/HERO Omnibus | 2019-2020
PI: Steven Yule | Science-PI: Roger Dias
This project is a collaboration between Brigham and Women’s Hospital, Harvard Medical School, and McMaster University and aims to investigate the role of crew autonomy in management of medical emergencies in deep space exploration missions. Computer vision and wearable physiological sensors are used to develop objective measures of team performance.