Efficient Blind Spectral Unmixing of FluorescentlyLabeled Samples Using Multi-Layer Non-Negative MatrixFactorization

Abstract: The ample variety of labeling dyes and staining methods available in fluorescence microscopy has enabled biologists toadvance in the understanding of living organisms at cellular and molecular level. When two or more fluorescent dyes areused in the same preparation, or one dye is used in the presence of autofluorescence, the separation of the fluorescentemissions can become problematic. Various approaches have been recently proposed to solve this problem. Among them,blind non-negative matrix factorization is gaining interest since it requires little assumptions about the spectra and concentration of the fluorochromes. In this paper, we propose a novel algorithm for blind spectral separation that addressessome of the shortcomings of existing solutions: namely, their dependency on the initialization and their slow convergence.We apply this new algorithm to two relevant problems in fluorescence microscopy: autofluorescence elimination andspectral unmixing of multi-labeled samples. Our results show that our new algorithm performs well when compared withthe state-of-the-art approaches for a much faster implementation.

dye; fluorochrome; algorithm; article; autofluorescence; controlled study; fluorescence; fluorescence microscopy; non negative matrix factorization; spectral unmixing algorithm; staining; statistical model; algorithms; female; fluorescent dyes; humans; image processing; computer-assisted; male; microscopy; models; theoretical; neoplasms; staining; labeling

Efficient Blind Spectral Unmixing of FluorescentlyLabeled Samples Using Multi-Layer Non-Negative MatrixFactorization Articles