We report a novel analytical approach to identify individual components of a cell’s endogenous fluorescence, recorded by spectrally-resolved time-correlated single photon counting (TCSPC). Time-resolved area-normalized emission spectroscopy (TRANES) and principal component analysis (PCA) were applied to estimate the number of spectral components after metabolic modulation of cardiac cells following excitation with a 375 nm picosecond laser. Linear unmixing of TCSPC data spectrally decomposed individual components in living cells, while using characteristics of endogenously fluorescing molecules in solvents as a reference spectral database. Our data demonstrate the presence of three individual components, corresponding to the nicotinamide adenine dinucleotide (phosphate) (NAD(P)H) in organic and inorganic solvents and to the residual flavoprotein fluorescence. The presented analytical approach offers a new alternative for the spectral separation of multi-wavelength fluorescence lifetime spectroscopy data to the conventional analysis, and opens a new possibility for the use of pattern recognition for fast resolution of components in 2D fluorescence lifetime microscopy images.