Heart rate measurement in neonatal patients using a webcamera

At present there is a clear need for non-contact monitoring of the physiological signs of the patients. The system proposed in this paper aims to measure the heart rate of neonatal infants without any direct contact with the patient. The solution proposed is based on the use of standard, low-cost and commercially available digital webcamera by which it has been possible to observe defined portions of the patient face; the sequence of such images has then been used, by a specifically developed algorithm (based on Indipendent Component Analysis), to extract the heart rate of the patients. Data collected on 7 patients demonstrate the feasibility of the measurement method proposed. Data acquired on the same patients with standard electrocardiography (ECG) has been used for comparison. Bland-Altman analysis of data show close correlation of the heart rates measured with the two approaches (correlation coefficient of 0.94) with an uncertainty of 4.5 bpm (k=1). This technique has a valuable interest for the use in clinical environment as non-contact, easily deployable and economic monitoring system, but it also shows an interesting potential for remote, home health monitoring.