In this study, polarization second harmonic generation (SHG) imaging is used and data analysis is developed to gain contrast and to discriminate with pixel resolution, in the same image, SHG source architectures. We use mammalian tissue in which both skeletal muscle and fibrilar collagen can be found. The images are fitted point by point using an algorithm based on a biophysical model, where the coefficient of determination is utilized as a filtering mechanism. For the whole image we retrieve

more »

... r every pixel, the effective orientation, θ e , of the SHG active structures. As a result a new image is formed which its contrast depends on the values of θ e . Collagen presented in the forward direction for a predefined region of interest (ROI), peak distribution of angles θ e centered in the region of ~45°, while muscle in the region of ~65º. Consequently, collagen and muscle are represented in different colors in the same image. Thus, here we show that it is possible to gain contrast and to discriminate between collagen and muscle without the use of any exogenous labeling or any co-localization with fluorescence imaging.