Abstract
Selective Plane Illumination Microscopy (SPIM) has become an emerging technology since its first application for 3D in-vivo imaging of the development of a living organism. An extensive number of works have been published, improving both the speed of acquisition and the resolution of the systems. Furthermore, multispectral imaging allows the effective separation of overlapping signals associated with different fluorophores from the spectrum over the whole field-of-view of the analyzed sample. To eliminate the need of using fluorescent dyes, this technique can also be applied to autofluorescence imaging. However, the effective separation of the overlapped spectra in autofluorescence imaging necessitates the use of mathematical tools. In this work, we explore the application of a method based on Principal Component Analysis (PCA) that enables tissue characterization upon spectral autofluorescence data without the use of fluorophores. Thus, enabling the separation of different tissue types in fixed and living samples with no need of staining techniques. Two procedures are described for acquiring spectral data, including a single excitation based method and a multi-excitation scanning approach. In both cases, we demonstrate the effective separation of various tissue types based on their unique autofluorescence spectra.