Comparison of Mueller matrix differential decomposition and transformation

Abstract

Mueller matrix polarimetry (MMP) quantify the changes in polarization of light beam during propagation through the sample, as comprehended in the Mueller matrix M. However, several polarization effects being occurring simultaneously in the tissues are present in the elements of M in an intermixed form, obstructing their unique interpretation. Several decomposition schemes are used to untangle these polarization effects. Mueller matrix transformation (MMT) and Mueller matrix differential decomposition (MMDD) are famous formalisms to achieve this objective. Herein, these two formalisms (i.e., MMT and MMDD) have been compared for different tissue (n = 34) and phantom (n = 22) samples. Specifically, the two methods were quantitatively compared for two pairs of polarization parameters, namely the total depolarization (ΔT,1-b) and total retardance (R,t). The findings revealed consistent mean total depolarization values (p = 0.27 for tissue samples and p = 0.99 for phantom samples) for the two formalisms, while the difference between the mean total retardance was highly significant (p < 0.001 and p < 0.01 for tissue and phantom samples, respectively). Moreover, linear fitting of the scatter plot for the corresponding polarization parameters of the two formalisms (ΔT vs 1-b) and (R vs t) showed a good correlation for both tissue and phantom samples. These findings suggest that despite the differences between the two methods, the corresponding polarization parameters show a good correlation.