Grant: Collaborative Research and Development Grants

Pressure-sensitive mats (PSM) are being actively investigated and applied to monitoring adult patients, both in hospital and in their home. PSM measure contact pressure using a grid of sensors and produce data in the form of an image. These data have been widely used among adult patient populations to assess a variety of health indicators. They are particularly appropriate for continuous patient monitoring since they are non-invasive and non-intrusive, often being placed below the mattress. However, there is little research into the use of pressure-sensitive mats in monitoring neonates, particularly those in the neonatal intensive care unit (NICU).* Patients in the NICU are carefully monitored at all times using physiological monitors. Continuous monitoring of physiologic data assists clinicians in making diagnoses and prognoses. In particular, blood oxygen is carefully tracked and alarms are generated whenever a neonate's oxygen levels fall below set thresholds. The rate of false alarms is very high leading to ""alarm fatigue"" which, in turn, can lead to missed alarms which are incorrectly dismissed. Many of these false alarms are caused by motion-induced artifacts on the SpO2 leads. Furthermore, the current method of estimating respiration rate is known to be unreliable. * This research will investigate the potential for PSM to accurately measure respiration and to detect patient movements. Sensor fusion between PSM and physiological data may lead to improved monitoring of neonates with reduced false alarm rates. This research will first establish the suitability of PSM for patient monitoring of neonates; their low mass presents unique challenges for detection and monitoring. Bench testing will be conducted in the patient simulation facility at CHEO to establish baseline potential for PSM-based detection of movement and respiration for neonates. The collection of actual patient data will then be initiated in the CHEO NICU, where PSM data will be collected simultaneously with SpO2, ECG, and bedside observations of clinical events and interventions. These data will be used to develop novel data analytics algorithms for PSM-based estimation of respiration rates and the detection and suppression of motion-induced false alarms.*