Continuous non-contact vital sign monitoring in neonatal intensive care unit

Abstract

Current technologies to allow continuous monitoring of vital signs in pre-term infants in the hospital require adhesive electrodes or sensors to be in direct contact with the patient. These can cause stress, pain, and also damage the fragile skin of the infants. It has been established previously that the colour and volume changes in superficial blood vessels during the cardiac cycle can be measured using a digital video camera and ambient light, making it possible to obtain estimates of heart rate or breathing rate. Most of the papers in the literature on non-contact vital sign monitoring report results on adult healthy human volunteers in controlled environments for short periods of time. The authors' current clinical study involves the continuous monitoring of pre-term infants, for at least four consecutive days each, in the high-dependency care area of the Neonatal Intensive Care Unit (NICU) at the John Radcliffe Hospital in Oxford. The authors have further developed their video-based, non-contact monitoring methods to obtain continuous estimates of heart rate, respiratory rate and oxygen saturation for infants nursed in incubators. In this Letter, it is shown that continuous estimates of these three parameters can be computed with an accuracy which is clinically useful. During stable sections with minimal infant motion, the mean absolute error between the camera-derived estimates of heart rate and the reference value derived from the ECG is similar to the mean absolute error between the ECG-derived value and the heart rate value from a pulse oximeter. Continuous non-contact vital sign monitoring in the NICU using ambient light is feasible, and the authors have shown that clinically important events such as a bradycardia accompanied by a major desaturation can be identified with their algorithms for processing the video signal.

Keywords: ECG-derived value; NICU; O2; adhesive electrodes; adhesive sensors; adult healthy human volunteers; ambient light; biochemistry; biomedical equipment; biomedical optical imaging; blood vessels; bradycardia identification; breathing rate estimation; camera-derived estimates; cardiac cycle; cardiology; clinical letter; continuous heart rate estimation; continuous noncontact vital sign monitoring; continuous oxygen saturation estimation; continuous pre-term infant monitoring; continuous respiratory rate estimation; continuous vital sign monitoring technologies; digital video camera; direct patient contact; electrocardiogram; estimation theory; feature extraction; high-dependency care area; hospital; incubators; infant nursing; infant pain; infant skin damage; infant stress; major desaturation; mean absolute error; medical image processing; minimal infant motion; neonatal intensive care unit; oximetry; oxygen; paediatrics; patient care; patient monitoring; pneumodynamics; pulse oximeter; reference heart rate value; stable sections; superficial blood vessel colour change measurement; superficial blood vessel volume change measurement; time 4 day; video cameras; video recording; video signal processing algorithms; video-based noncontact monitoring methods.

Figure 1

Monitoring equipment showing a mannequin inside the study incubator in the Oxford NICU Digital video camera and camera arm are clearly visible

Figure 2

Image from video camera of infant resting quietly in the incubator

Figure 3

Region of interest extraction for a 15-second video section a Mean of the light intensity of pixels from the red, green and blue channels of a 150 × 150 region of interest b Reflectance PPG signal derived from ICA of the three channels

Figure 4

Estimates for heart rate and respiratory rate: a Heart rate estimates derived from the reflectance PPG signal extracted from the three colour channels of the video camera using ICA are shown in red; the ECG-derived estimates from the Philips monitor (reference values) are shown in black b Respiratory rate estimates derived from the same sections of video camera data are shown in red; the EIP-derived estimates (reference values) are shown in black

Figure 5

Estimates of oxygen saturation and pulse-oximeter estimates Estimates of oxygen saturation derived from the red and blue channels of the video camera are shown in red; the pulse-oximeter estimates from the Philips monitor (reference values) are shown in black DC_R/DC_B represents the ratio of the DC values in the red and blue channels

Figure 6

Inter-method agreement (percentage of estimates within a given number of beats/minute) for pairwise comparison of the three methods of deriving heart rate estimates during the 20 min segment analysed in Figs. 4 and 5