Prediction and detection of apnoea in preterm infants
1 Jan 2020 - 31 Dec 2023
People funded by this grant:
Dargaville P,
Gale T,
Salmon B,
Turner P
Project Total: $823,982; We propose a 3 year study with the over-arching goal of developing tools to improve the management of apnoea in preterm infants on non-invasive respiratory support. Our hypotheses are that pattern recognition and machine-learning software can be used to predict impending apnoeic events with a high degree of certainty, and that a combination of innovative monitoring devices measuring torso movement and acoustic transmission of airway noises can improve the detection of central and obstructive apnoeic events.
Background
Modern respiratory management for preterm infants involves protracted use of continuous positive airway pressure (CPAP). This less-invasive form of respiratory support has advantages over intubation and ventilation, but relies on adequate respiratory drive and airway patency at a time when both of these are prone to failure. Central and obstructive apnoeic events are well-recognized causes of hypoxemic and/or bradycardic episodes in preterm infants, which may have lasting neurodevelopmental consequences. At present central pauses in respiration can be difficult to reliably identify in infants on CPAP, and most obstructive episodes go undetected. Moreover, there is no way of knowing when an infant is about to stop breathing.
Research proposal
Over a 3 year period we will assemble a unique compendium of physiological recordings in preterm infants <30 weeks gestation on CPAP using multimodal monitoring equipment. In our first aim we will apply sophisticated machine-learning techniques to the physiological data to develop a predictive tool to identify a pre-apnoeic state in real-time. In our second aim we will determine whether a) respiratory pauses can be reliably identified using a unique contactless respiratory monitoring system, and b) whether obstructive apnoea can be detected in preterm infants on CPAP with acoustic sensors. Funded by: National Health & Medical Research CouncilNational Health & Medical Research Council - Grant-Ideas ($1,647,964); University of Tasmania.