Health systems across the nation may soon be able to simulate patient care at the bedside, thanks to cutting-edge tools being developed by leading clinical artificial intelligence researchers at the University of Florida.
To celebrate UF AI Days on campus, UF Intelligent Clinical Care Center (IC3) and PRISMAp Director, Azra Bihorac, M.D. MS FCCM FASN, and IC3 Assistant Director of Clinical AI, Benjamin Shickel, Ph.D., gave an overview of a project to create a “digital twin” or virtual replica in the health care setting. They were joined by assistant professor in the School of Architecture, Karla Ochoa, Ph.D. and assistant professor in the Department of Environmental Engineering Sciences, Christine Angelini, Ph.D., who discussed their own digital twin projects in a panel discussion.
As part of a collaboration that also included IC3 Co-Director Parisa Rashidi, Ph.D., the center’s interdisciplinary team collaborated with industry experts from Mark III Systems and NVIDIA to create a virtual replica of an intensive care unit (ICU) room. Developed in NVIDIA Omniverse using more than 80 photos and videos from a real-world, UF Health Heart & Vascular Hospital ICU room as a reference, the twinned ICU will provide opportunities for providers to plan and simulate care strategies in the virtual world to optimize real-world health care delivery.
“Physical healthcare spaces have not kept up with the rapid pace of technological advancement,” Bihorac said. “Our ICU twin is just the first step towards our vision to create a health care metaverse for optimizing patient care, health care processes and smart hospital spaces of the future using the power of AI.”
The group intends for the replica, named the “Digital Twin of the Intelligent Hospital” to integrate real-time streaming of patient information with precise, state-of-the-art data analytics.
“Digital twins differ from traditional simulations in their ability to dynamically respond to real-world data,” Shickel said. “Synchronizing real-time patient data streams with our virtual environment will allow for high-quality monitoring and simulation tools that automatically adapt to real-world conditions.”
According to Bihorac, once completed, the “twin” may also serve as an educational platform to help streamline training and improve the patient experience through virtual care pathways.
“One day, we will be inside of this environment, using this intervention for training our care teams and planning out the logistics of patient care,” Bihorac said.