Laser speckle flowgraphy

A commercially available LSFG system (LSFG-NAVI; Softcare Co., Ltd., Fukuoka, Japan) was used to measure perfusion in a custom-built model eye and in retinal vessels in vivo before and during flicker stimulation. The principles of LSFG have been described in detail previously.[26–28] Briefly, the technique is based on the observation that if an irregular surface is illuminated with a coherent light source the backscattered light gives the appearance of a consistent scatter pattern. Moving corpuscular blood components such as erythrocytes cause a distinct variation in the speckle pattern, becoming manifest in a reduction of speckle contrast and variation, caused by integration of dynamic speckle during camera exposure time. As described in more detail below, flow information is generated based on statistical analysis of the above described speckle pattern, which is recorded by a digital camera.

The LSFG device used in this study comprised a fundus camera supplied with an 830-nm diode laser and a digital charge-coupled device camera (750 x 360 pixels). The principal output parameter of LSFG, mean blur rate (MBR), is a parameter representing a measurement of relative blood flow velocity and is expressed in arbitrary units (au). During a single LSFG scan, a total of 118 images are continuously recorded at a rate of 30 frames per second over a time period of approximately 4 seconds. The device is equipped with an analysis software (LSFG Analyzer version 3.1.58; Softcare), which synchronizes and mathematically averages all 118 acquired images to produce a so-called ‘‘composite map” showing the distribution of perfusion in the ocular fundus within one cardiac cycle.

Quantification of blood flow of the LSFG system used is based on first order statistics using the standard deviation of the intensity distribution of the speckle pattern (δ). This equals to the mean intensity <I> under idealized conditions, but may be lower in reality du to inhomogeneities in the scattering tissue. For mapping blood flow space–time correlation function of the speckle intensity fluctuation is calculated. The reciprocal of speckle contrast δ/<I> is dependent on the ratio of the correlation time of the temporal fluctuations in intensity (τ_c) and exposure time (T) and

Speckle contrast is then calculated as follows:

where c is a constant dependent on exposure time. The normalized blur (NB) which is an approximation of the reciprocal of speckle contrast. In the present system velocity is calculated based on square blur rate (SBR). In case of a CCD array the SBR can be approximated as

where where I_n,m,t is the intensity detected at the pixel (n,m), is integrated over the exposure time. The bracket <> denotes average over 32 frames of the integrated light intensity[29]. The MBR calculated as a running average over 26 pixels and proportional to SBR.