Increasing the Accuracy of Blood Hematocrit Measurement by Triplicate Wavelength Photoplethysmography Method

The importance of continuous monitoring and rapid and accurate informing of changes in blood components (e.g., percentage of hematocrit [HCT]), especially in acute conditions, has motivated researchers to use non-invasive measurement methods. Therefore, this study aimed to evaluate the association between blood HCT and photoplethysmographic signal and the possibility of increasing the accuracy of its measurement by optical method at three specific wavelengths. In this study, photoplethysmographic signals were recorded at three different wavelengths for 25 male and female subjects (mean age of 37.56±10.40 years), who referred to the laboratory to assess their blood HCT percentage. We extracted the peak value of the signal after calibrating a special probe with a standard pulse oximetry system and applying the necessary preprocesses on the received signal. Eventually, in addition to the estimation of HCT level, we assessed the level of correlation between the optical method results and laboratory data applying Pearson’s, Kendall’s, and Spearman’s correlation coefficients. Comparison of HCT measurement by the optical method and reference levels measured by standard laboratory technique in 25 subjects showed a mean error of 0.6±0.25%. In addition, evaluation of the relationship between data of the percentage of laboratory HCT with mean of 43.43±3.43 and data obtained by photoplethysmographic signals with mean of 43.31±3.27 by three Pearson’s, Kendall’s, and Spearman’s methods demonstrated that the data of the two techniques had a significant correlation of 0.949, 0.860, and 0.955, respectively (P<0.01). Given the high correlation of characteristics of the photoplethysmographic signal at three wavelengths with blood HCT and level of accuracy of our findings, the proposed method could be exploited for accurate, clean and cost-effective monitoring of HCT level.