Presented at the Neonatal Society 2014 Autumn Meeting.
Green G1, Davis S1, Villarroel M2, Guazzi A2, Jorge J2, Watkinson P3, Tarassenko L2, McCormick K1
1 Neonatal Unit, John Radcliffe Hospital, Oxford University Hospitals NHS Trust
2 Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford
3 Nuffield Department of Clinical Neurosciences, University of Oxford
Background: Vital sign monitoring is important in the routine care of the preterm neonate. There have been few advances in neonatal monitoring for more than 30 years. Standard monitoring currently relies on ‘wired’ methods which can increase the need for handling and damage fragile skin. We present a novel non-contact method of monitoring, using a digital video camera positioned over the infant’s incubator which we hypothesise is as accurate as conventional monitoring and will provide entirely new information about the neonate’s physiological status.
Methods: The study received ethical approval and is funded by the Wellcome Trust/EPSRC. Preterm infants in HDU were filmed for a maximum of 4 days each using a digital video camera positioned over a specially modified incubator. Vital sign data from the standard patient monitor was collected concurrently and the camera data was analysed retrospectively. Routine care continued throughout the study. Signal analysis was used to obtain a reflectance photoplethysmogram waveform (PPG) from the raw camera data. Mathematical modelling was then used to identify frequencies corresponding to the physiological parameters of interest from the noisy PPG waveform.
Results: The first two infants were filmed for a total of 39.8 hours during which 24.9 hours of valid camera data was obtained (time periods when the infants was resting quietly inside the incubator). Estimates of heart rate were possible for 80.3% of the valid camera data time with a mean absolute error of <3bpm when compared with ECG heart rate. Data obtained from these first infants have shown that it is possible to continuously monitor heart rate and respiratory rate using this non-contact method. It is also possible to demonstrate and track changes in oxygen saturations. This is the first time that continuous monitoring of vital signs using a non-contact method has been reported in the neonatal population. These initial results which demonstrate feasibility are part of a larger study to evaluate the use of this technology in the neonatal care setting.
Conclusion: Continuous non-contact vital sign monitoring is possible in the neonatal care setting. Our method is not only able to replicate standard ‘wired’ monitoring but in the future is likely to produce novel parameters such as perfusion maps and movement indices.
Corresponding author: email@example.com
1. Villarroel M et al. Continuous non-contact vital sign monitoring in neonatal intensive care unit. Healthcare Technology Letters 2014, Vol. 1, Iss. 3, pp. 87-91. Published online ahead of print.