Quantitative Chemical Imaging and Unsupervised Analysis Using Hyperspectral Coherent Anti-Stokes Raman Scattering Microscopy

Abstract

In this work, we report a method to acquire and analyze hyperspectral coherent anti-Stokes Raman scattering (CARS) microscopy images of organic materials and biological samples resulting in an unbiased quantitative chemical analysis. The method employs singular value decomposition on the square root of the CARS intensity, providing an automatic determination of the components above noise, which are retained. Complex CARS susceptibility spectra, which are linear in the chemical composition, are retrieved from the CARS intensity spectra using the causality of the susceptibility by two methods, and their performance is evaluated by comparison with Raman spectra. We use non-negative matrix factorization applied to the imaginary part and the nonresonant real part of the susceptibility with an additional concentration constraint to obtain absolute susceptibility spectra of independently varying chemical components and their absolute concentration. We demonstrate the ability of the method to provide quantitative chemical analysis on known lipid mixtures. We then show the relevance of the method by imaging lipid-rich stem-cell-derived mouse adipocytes as well as differentiated embryonic stem cells with a low density of lipids. We retrieve and visualize the most significant chemical components with spectra given by water, lipid, and proteins segmenting the image into the cell surrounding, lipid droplets, cytosol, and the nucleus, and we reveal the chemical structure of the cells, with details visualized by the projection of the chemical contrast into a few relevant channels.

Denoising Stimulated Raman Spectroscopic Images by Total Variation Minimization

“…SRS spectroscopic images can be decomposed into chemical maps of major components by methods including linear unmixing 8 , multivariate curve resolution (MCR) 17 , independent component analysis 18 and spectral phasor analysis 40 . For images suffering from low SNR, these methods might lead to false spectral results.…”

Coherent Raman Scattering Microscopy in Biology and Medicine

“…Tuning the time delay between the two chirped pulses can substantially change the overlapping difference in frequency, exciting different Raman shifts. Spectral focusing has been utilized by many groups for hyperspectral CARS imaging (66–69). The third method is to perform the Fourier transform on a time-resolved CARS trace (70, 71).…”

In Situ and In Vivo Molecular Analysis by Coherent Raman Scattering Microscopy

“…SVD is widely used for noise reduction in spectroscopic images (32, 34–36). This method first factorizes the data matrix D into three matrix factors, D = USV T , where the unitary matrix U corresponds to an array of spectral vectors; S is a diagonal matrix composed of singular values; and V corresponds to an array of spatial vectors.…”

“…Along with the efforts to improve instrumentation, various chemometrics approaches have been adopted to decompose the three-dimensional data ( x-y -Ω) in a hyperspectral CRS image into chemical maps (21, 27, 32, 33). The spectral fidelity and image quality were improved by algorithms that remove the noise in spatial and spectral domains (34–37).…”