Asynchronous Breathing and Patient Ventilator Interaction Assessment: A Machine Learning Approach

While the prevalence and consequences of asynchronous breathing (AB) due to suboptimal patient-ventilator interaction (PVI) is unknown, a method to investigate the impact of AB on patient’s outcome is required. For this reason, this thesis investigates the use of machine learning algorithms in assessing the quality of PVI. These models can potentially aid clinicians to evaluate the patient’s condition during mechanical ventilation treatment; thus, allowing better decision making.