Sample entropy analysis of laser speckle fluctuations to suppress motion artifact on blood flow monitoring

Abstract

Laser speckle imaging is a common technique to monitor blood flow. The fluctuations in speckle intensity can be related to the blood flow by calculating the speckle contrast, the ratio between the standard deviation of speckle fluctuations and the average intensity. However, this simple statistic calculation is easily affected by motion artifacts. In this study, we applied sample entropy analysis instead of calculating standard deviations of the speckle fluctuations. Similar to the traditional method, sample entropy-based speckle contrast increases linearly with flow rate but was shown to be more immune to sudden movements during an upper arm occlusion test.