Seminar Series

Over-distension Prediction via Hysteresis Loop Analysis and Patient-specific Basis Functions in a Virtual Patient Model


Qianhui Sun


Department of Mechanical Engineering, University of Canterbury

Time & Place

Thu, 10 Mar 2022 14:20:00 NZDT in E6, Engineering Core


In clinical care, mechanical ventilation (MV) is a key treatment to keep patients alveoli open and improve oxygenation. Staircase recruitment maneuvers (RMs)
followed with positive-end-expiratory-pressure (PEEP) is one common protective MV approach for acute respiratory distress syndrome (ARDS) and respiratory
failure patients. However, during this procedure, excessive or insufficient support (delivered airway pressure or tidal volume) caused by suboptimal PEEP settings
can lead to ventilator induced lung injury (VILI), increasing morbidity and mortality. Meanwhile, significant intra- and inter- patient variability during MV treatment
further increases the difficulty in optimising clinical care. Currently, an optimal  PEEP level for patients is commonly defined as the best compromise between
recruitment and over-distension. However, no effective standardised method exists for clinicians to determine the optimal patient-specific PEEP, leading to uncertainty and increased risk in care.
In this presentation, a predictive lung mechanical model will be introduced. It is able to detect and monitor distension at bedside, particularly continuously and non-invasively, with a high accuracy to predict the likelihood of its occurrence at higher PEEP levels. Hence, the proposed model is able to guide personalized clinical care (MV settings) with minimal harm.

Supervisor: Distinguished Professor Geoff Chase