E-Health monitoring using camera: Measurement of vital parameters in a noisy environment

We investigated contact-less solution to study micro circulation and its spatial heterogeneity by means of RGB imaging and to estimate blood oxygenation and heart rate. Oxygen saturation SpO2 and heart rate HR are indicators directly related to the cardiovascular diseases. Detecting low level of oxygen and abnormal heart rate can be an indicator of fatal heart problem and require urgent medical attention. Traditionally, measurement of HR and SpO2 is made through the probes or pulse oximeter attached to the skin surface and hence could be a potential way of spreading contagious infections. For a non-contact assessment of pulse rate, imaging photoplethysmogram iPPG has been proposed, where data is acquired through the video frames and different image analysis-based methods are used to extract required information. We detects face from the sequence of images using Viola Jones algorithm, followed by the extraction of region of interest, filtering performed in spatial domain removed artifacts, and intensity signal gives iPPG estimation by using CHROM method. The estimation of HR and SpO2 from these signals are evaluated on publicly available data set containing facial videos and reference contact photoplethysmogram. Furthermore, we also investigate the performance of these methods on data acquired under diverse illumination conditions. Relative correlation and % error gives satisfactory results to extend the method for more diverse scenario.