Hyperspectral imaging of biological targets: The difference a high resolution spectral dimension and multivariate analysis can make

Abstract

Hyperspectral imaging coupled with multivariate data analysis is a powerful new tool for understanding complex biological and biomedical samples. The advantages and drawbacks of adding a spectral dimension and multivariate data analysis to optical microscopy for biological interrogation will be demonstrated with three applications - DNA microarrays, live cell imaging, and in-situ hybridization labeled tissue. These data are selected to present the type of impact hyperspectral imaging can have in biomedical science. Images are acquired using our state-of-the-art hyperspectral imaging system and multivariate data analysis is used to extract pure component spectra and corresponding independent concentration maps of all fluorescent species. In most cases the data analysis algorithms are successful with little or no information given a priori and generate images that are free of the influences of spectral crosstalk, cellular autofluorescence, and other background emissions that often plague traditional fluorescence microscopy.