Enhancing the safety and efficacy of mechanical ventilation in acute respiratory distress syndrome

This thesis presents a series of innovative studies aimed at improving patient outcomes and advancing the understanding of critical care physiology for patients who are being treated for Acute Respiratory Distress Syndrome (ARDS). The first projected presented involved the development and validation of a novel shared ventilation system. This system leverages adjustable pressure-limiting (APL) valve technology to independently titrate tidal volumes to two patients. This allows a greater use of lung-protective ventilation in this context, and offers a potential solution for resource-constrained settings, including pandemics and mass casualty events. The second project focused on the development and validation of a C5.0 machine learning model to successfully predict short-term mortality in critically ill patients with ARDS undergoing an initial session of prone positioning. This model, trained on a limited dataset of routine clinical parameters, demonstrated promising predictive performance, suggesting its potential utility in clinical decision-making. Such an approach could prompt a detailed review of the management of particularly high-risk patients, or the early consideration of other rescue therapies. The final project involved the expansion of a mathematical cardiorespiratory simulator to incorporate the effects of invasive ventilation in the prone position. This advanced simulation tool could be used to investigate the underlying mechanisms of prone positioning, optimise ventilator settings, and explore the potential impact of various therapeutic interventions. Collectively, these studies contribute to the advancement of critical care practice and provide valuable insights into the complex physiological responses to critical illness.

Publisher

University of Galway

Rights

Attribution-NonCommercial 4.0 International

cbn

Collections

University of Galway Theses (Research Masters Theses)