Towards the Digital Lung: A Deep Learning Approach to Simulating Physiological Lung Function During Ex Vivo Lung Perfusion

Purpose: Ex vivo lung perfusion (EVLP) is an established reconditioning platform that generates multi-modal physiological data from isolated human lungs and is, thus, well-suited for machine learning. Digital twins —computer simulations of physical objects—are an emerging concept in medicine; however, they are unexplored in transplant medicine. Herein, we describe the development of a deep learning approach to simulate physiological lung function by leveraging high-resolution time-series lung ventilation d ata and demonstrate its predictive utility during EVLP.
Source: The Journal of Heart and Lung Transplantation - Category: Transplant Surgery Authors: Source Type: research