Preliminary Analisys on the Effect of Skin Temperature on Photoplethysmographic Signal

Abstract

Photoplethysmographic sensors represent a promising wearable technology as combines its advantages in costs and weight with a wide variety of cardiovascular parameters which can be detected. Despite their potential, the accuracy of heart rate and blood oxygenation measures are affected by several influencing parameters, including the skin temperature. Thus, a multiwavelength PPG sensor was tested in a cohort of 20 subjects to assess the influence of the external temperature in the measurements of heart rate. The performance of the PPG sensor was tested for 5 different temperatures (i.e. 15, 20, 25, 30 and 35°C) at two different measurement sites (at the tip of the finger and at the wrist). Data were subsequently analysed in Matlab and compared with those of a reference ECG-based chest-band. Preliminary results showed that the accuracy of the PPG sensor increases with temperature; specifically, the blue and green wavelengths showed the best overall performances. Red light proved to be the most sensitive to temperature variations and showed the worst overall performance. Although results should be confirmed on a larger cohort of subjects, this study could contribute with a useful indication to enhance the accuracy of PPG-based wearable devices.