A trustworthy speckle contrast calculation is fundamental in many applications, such as “laser speckle contrast Imaging” (LSCI), which is a non-invasive technique commonly employed to estimate relative blood speed. In LSCI, the local contrast of a speckle image is calculated using spatial, temporal analysis or a combination of both. In this work, we focus on the spatial analysis. To calculate the local spatial contrast, typically, a sliding window of 5x5 pixels is used to calculate the standard deviation (σ<sub>s</sub>) and the mean intensity (<I><sub>s</sub>) of those 5x5 pixels and the calculated contrast K<sub>S</sub>=σ<sub>s</sub>/(<I><sub>s</sub>) is assigned to the central pixel of the sliding window. In this work, we show that this experimental procedure to calculate the local speckle contrast does not match the corresponding spatial theoretical model and we propose an alternative method that considers correlations of the central pixel with the other ones. We have found a better agreement of the contrast measurement with our numerical calculation.
Speckle contrast analysis had been used for multiples purposes, for instance, laser speckle contrast imaging (LSCI) has been used to estimate the relative blood flow speed in a non-invasive way. The speckle contrast can be calculated using a spatial or temporal algorithm or a combination of both. Our work focuses into the contrast temporal algorithm. A contrast image calculated with the temporal contrast algorithm requires a sequence of L equal-sized frames. The contrast images are performed pixel by pixel, however, the experimental contrast calculation does not match with the current temporal theoretical model especially when the exposure time T is smaller than the correlation time τ<sub>c</sub>. In this work, we propose to correlate neighboring pixels along the temporal axis to improve the contrast calculation. The contrast measurements using our proposal provide a better agreement than the current models.
The aim of this study was to compare the effectiveness of Rose Bengal (RB) and Methylene Blue (MB) as photosensitizers (PS) in Photodynamic Inactivation (PDI) on planktonic cultures of <i>Candida albicans</i>, a well-known opportunistic pathogen. RB and MB at concentrations ranging from 0.5 to 60 μM and fluences of 10, 30, 45 and 60 J/cm<sup>2</sup> were tested. The light sources consist of an array of 12 led diodes with 30 mW of optical power each; 490-540 nm (green light) to activate RB and 600 -650 nm (red light) to activate MB. We first optimize the <i>in vitro</i> PDI technique using a single light dose and the optimum PS concentration. The novelty of our approach consist in reducing further the PS concentration than the optimum obtained with a single light exposure and using smaller light fluence doses by using repetitive light exposures (two to three times). MB and RB were tested for repetitive exposures at concentrations ranging from 0.1 to 10 μM, with fluences of 3 to 20 J/cm<sup>2</sup>, doses well below than those reported previously. All experiments were done in triplicate with the corresponding controls; cells without treatment, light control and dark toxicity control. RB-PDI and MB-PDI significantly reduced the number of CFU/mL when compared to the control groups. The results showed that RB was more effective than MB for <i>C. albicans</i> inactivation. Thus, we show that is possible to reduce significantly the amount of PS and light fluence requirements using repetitive light doses of PDI <i>in vitro</i>.