Quantitative parameter analysis of effects of particle density on the imaging signals in OCT

The particle density in media such as the density of the red cells in blood is of great importance for many applications. It has been found that the concentration of low concentration solution could be calculated by using some simple models, which are not applicable for higher concentration cases. In this work, we report a method of deriving information about particle density from temporal statistics of optical coherence tomography (OCT) signals. An experimental observation-based method is presented. Through two groups of phantom experiments with high-concentration and low-concentration Intralipid solutions, the probability density distribution of structural information at a specific depth is obtained, and four parameters are defined to characterize the curves: intercept, peak amplitude, peak count, and full width at half maximum (FWHM). The law of these four parameters changing with concentration are obtained. In order to test if the phantom can model human skin, the probability density distribution of human skin is obtained through in vivo experiments. It is found that the four parameters of the skin are not strongly related to those of the Intralipid solution. Then, the reasons for this phenomenon are analyzed. The data analysis in this paper preliminarily shows that the probability density distribution, which has a depth dependence is also affected by the sample concentration, but the correlation between different samples is low. This conclusion has the potential to quantify sample concentrations in the future.

Automated summary

In this study, a method of deriving particle density information from temporal statistics of optical coherence tomography (OCT) signals is presented. By conducting phantom experiments with high and low concentration solutions, the probability density distribution of structural information at a specific depth is obtained, and four parameters are defined to characterize the curves. Through in vivo experiments on human skin, it is found that the four parameters of the skin are not strongly related to those of the simulated solution, and the reasons for this phenomenon are analyzed. The preliminary results of this research show that the probability density distribution, which has a depth dependence, is also affected by the sample concentration.