Noninvasive polarization-based technique for hematocrit monitoring

Abstract

Subject of study. Polarization images corresponding to the 2D distributions of polarization state parameters for light scattered by biological specimens with different hematocrit levels were obtained by shining laser light in a given polarization state of blood-containing tissue through the nail bed of the left fifth finger. Aim of work. The aim of this work is to determine whether active video polarimetry is suitable for noninvasive real-time monitoring of blood hematocrit and to develop data-based criteria to identify hematocrit levels corresponding to various body states of human test subjects. Method. Physical models of changes in the parameters of polarized radiation during interaction with a biological specimen are used. A prototype optical design was developed for a Stokes video polarimeter capable of recording the irradiance distribution within a single analysis field on the sensitive area of a photodetector array. The digital images were processed using the Stokes–Mueller matrix formalism. A comparative statistical analysis of the 3D distributions for various parameters describing the polarization state (degree of polarization and the azimuth and ellipticity of the polarization ellipse) was performed to determine whether it is possible to detect differences in the hematocrit levels of human blood. Main results. A new approach to noninvasive hematocrit monitoring based on optical polarization is proposed. We obtained 3D distributions and histograms of light polarization state parameters for two human test subjects known to have different hematocrits. Statistical analysis revealed that significant differences in the second–fourth order statistical moments of the light polarization parameters are the most relevant diagnostic indicators for comparative analysis of biological tissues having different blood hematocrit levels. Practical significance. With further development, the proposed method can be applied to both medical diagnosis and emergency treatment. Timely detection of significant deviations from the normal hematocrit based on automated calculation and analysis of the data parameters identified will enable continuous hematocrit monitoring during surgical operations and effective real-time adjustment of blood viscosity.