In the MSM Consortium, computational modeling provides a platform or framework to integrate knowledge and data across multiple spatial and/or temporal scales, systematically revealing what is known and unknown in a system and stimulating the creation of testable hypotheses. This approach enables the integration of large disparate data sets and the application of modeling and simulation to predict emergent dynamics. In one paradigm, we ask models to reflect certain aspects of the real world in some dictated context and then draw conclusions from the simulations generated by these models. The role of modeling and simulation is finished before the technology is deployed to end-users in this approach. In another paradigm, modeling and simulation might be used to augment the utility of a technology via an integrated feedback loop. Digital twins are one realization of this second paradigm.
A digital twin is a digital replica of a living or non-living physical entity, such as a manufacturing process, medical device, piece of medical equipment, and even a person. Utilizing sensor data, digital twins combine simulation and analytics to gain insight into present and future operational states of each physical twin. The resulting digital twin can be wrapped with artificial intelligence, combined with ensembles of similar twins, or used in tandem with other predictive tools to analyze and diagnose operational states and to optimize performance under real-world operating conditions. This enables companies to make predictions about future performance, improve operations and productivity, and reduce the risk of unplanned downtime. Essential to the digital twin concept is a two-way connection between the physical asset and its virtual representation through the Internet of Things (IoT). This connection ensures that the virtual twin remains a faithful representation of the physical asset and provides a platform to predict the physical assets immediate future.
The healthcare industry is currently being disrupted by digital twin technology, where digital twins can represent diverse elements of the treatment process, ranging from medical devices to patients to healthcare delivery systems and other aspects of patient care. Tailoring treatment options based on the response of each individual patient is expected to be one of the biggest benefits. Another is the ability to detect and warn of an impending health issue before it occurs. Digital twin technology may also transform how treatments are deployed, unifying existing monitoring technologies into an integrated platform that can rapidly diagnose an individual’s disease state and then evaluate treatment options based on knowledge of not only characteristics of the various therapeutic options, but also estimates of the patients current and future pathological condition. Therefore, digital twins will not only result in faster, safer, and more efficient healthcare delivery to patients, but also improve our definition and image of a healthy patient.
October 24, 2019 9:00-9:15am: Digital Twin Keynote: Mark Palmer, Medtronic
October 25, 2019 8:45 - 9:00am: Integration Keynote:Gunnar Cedersund, Linköping University
W Andrew Pruett and Marc Horner (2019) Digital Twins in Healthcare: An Overview, abstract submitted to the 2019 BMES-FDAFrontiers in Medical Devices Conference
Rusty Irving, GE Global Research - 2019 BMES-FDA Frontiers in Medical Devices Conference, presentation
Breakout Session charge:
- current state of the art models for Digital Twins
- ML-MSM opportunities for new approaches to Digital Twins
- challenges ML-MSM modelers can address
For questions regarding the Digital Twin Overview, please contact:
Leonardo Angelone, Program Officer, National Institute on Drug Abuse (NIDA), NIH
Back to 2019 ML-MSM main page