Physiological Evaluation of Visually Induced Motion Sickness Using Independent Component Analysis of Photoplethysmogram

Makoto ABE, Makoto YOSHIZAWA, Norihiro SUGITA, Akira TANAKA, Noriyasu HOMMA, Tomoyuki YAMBE, Shin-ichi NITTA
Vol. 2 (2013) p. 25-31

For continuous evaluation of the effects of visually induced motion sickness (VIMS), we previously proposed a physiological index ρmax that represents the maximum cross-correlation coefficient between blood pressure (BP) and heart rate, the frequency components of which are limited to the Mayer wave-related band. However, ρmaxrequires continuous BP measurement using a bulky measuring device that is difficult to handle. Thus, we previously proposed an easier method for obtaining ρmax without continuous measurement of BP, which uses only a finger photoplethysmogram (PPG). In this method, independent component analysis (ICA) is used to extract BP-related signals from the PPG signals. However, continuous BP measurement is needed to determine the mixing matrix used in ICA. In order to achieve practical application of the method, this study aim to verify whether each subject’s mixing matrix can be estimated based on short-term continuous BP measurement. Being able to do so would mean that ρmax can be obtained from an estimated mixing matrix without continuously measuring BP during an experiment that presents a visual image to a subject. The validity of the proposed method was assessed by experiments performed on 28 subjects watching a swaying video image. From the ex-perimental results, we verified that the proposed method is able to extract independent components related to BP to yield ρmax between heart rate and each independent component used to evaluate the effects of VIMS. This result suggests that the effects of VIMS can be evaluated using short-term continuous BP measurement before the evaluation task.