Evaluation of a novel, non-contact camera-based photoplethysmography heart rate monitoring system for infants requiring neonatal resuscitation in the delivery room
Thursday, February 13, 2020
4:00 PM – 4:15 PM
Location: Max Bell - MB251
Introduction: Heart rate (HR) monitoring is critical for decision-making and assessing the newborn’s status during resuscitation. Electrocardiography (ECG) is the current gold standard for HR monitoring in the delivery room. However, using electrodes increase the risk of serious skin damage, injury, and pain, particularly for premature infants with delicate skin. The aim of this study was to evaluate the accuracy and feasibility of a novel non-contact camera-based photoplethysmography (cPPG) system as an alternative for HR monitoring in the delivery room for the first time.
Methods: Low birth-weight, preterm, newborn infants (n=40, <1500 g, <36 weeks gestation) were recruited from the Royal Alexandra Hospital delivery rooms. A novel cPPG system was designed for the delivery room and utilized to collect HR data for these infants, alongside ECG. CPPG utilizes colour changes resulting from changes in blood vessel volume during the cardiac cycle to determine HR.A Bland-Altman analysis was conducted to determine the level of agreement between cPPG and ECG HR measurements.
Results: Newborn infants included in this study had a mean (SD) birth weight of 1225 (475) g and gestational age of 28.6 (2.6) weeks. Of these, 16/40 infants were male and 37/40 received antenatal steroids. The Bland-Altman analysis showed a mean difference (95% levels of agreement) of 0.38 (-7.99 to 8.74) beats/minute between cPPG HR and ECG. Motion artefacts and frequency harmonics occasionally resulted in signal dropout and/or low signal reliability.
Conclusion: Our study demonstrated non-contact cPPG HR monitoring is feasible for the delivery room and had a similar accuracy to ECG for HR monitoring. However, further refinement and clinical trials are warranted to enhance and evaluate this technology before its clinical implementation.