Removing dynamic distortions from laser speckle flowgraphy using Eigen-decomposition and spatial filtering

Laser speckle flowgraphy (LSFG) has been widely used in the investigation of blood flows in ophthalmology. However, the dynamic changes of the ocular optics can impose artificial contrasts to the LSFG, corrupting the detection of both retinal vasculature and blood pulsation at the posterior segment of the human eye. In this study, we propose to use Eigen-decomposition method to separate the spatially and temporally varying speckle patterns from the static tissues. Spatial filtering is further applied to remove the distortion-correlated modulation of the speckle patterns. We experimentally show that with the proposed method, the integrity of blood vessels is significantly improved and the distortions in pulse waveforms can be well corrected.