Abstract
Machine learning-based cell classifiers use cell images to automate cell-type discrimination, which is increasingly becoming beneficial in biological studies and biomedical applications. Brightfield or fluorescence images are generally employed as the classifier input variables. We propose to use Raman spectral images and a method to extract features from these spatial patterns and explore the value of this information for cell discrimination. Raman images provide information regarding distribution of chemical compounds of the considered biological entity. Since each spectral wavelength can be used to reconstruct the distribution of a given compound, spectral images provide multiple channels of information, each representing a different pattern, in contrast to brightfield and fluorescence images. Using a dataset of single living cells, we demonstrate that the spatial information can be ranked by a Fisher discriminant score, and that the top-ranked features can accurately classify cell types. This method is compared with the conventional Raman spectral analysis. We also propose to combine the information from whole spectral analyses and selected spatial features and show that this yields higher classification accuracy. This method provides the basis for a novel and systematic analysis of cell-type investigation using Raman spectral imaging, which may benefit several studies and biomedical applications.
Highlights
Fundamental research and applications in biological and biomedical fields increasingly rely on automated laboratory systems to perform cytological profiling
We propose a new method to classify cells based on the features extracted from the spectral images of molecular components in single living cells
Hepa[1,2,3,4,5,6] (Hepa) is a cell line derived from mouse hepatoma, neuro2a (N2a) is a neuroblastoma cell line, and mesenchymal stem cells (MSC) are primary cells
Summary
Fundamental research and applications in biological and biomedical fields increasingly rely on automated laboratory systems to perform cytological profiling. Previous studies have primarily used brightfield (transmission) images or fluorescence images of subcellular structures to extract mathematical features[5,6]. A hyperspectral image allows one to reconstruct an image of a given molecular compound (i.e., wavenumber, or spectral band), giving a spatial pattern of its distribution within the cell (Fig. 1II). We propose a novel, comprehensive method to classify living cells based on the mathematical patterns extracted from Raman hyperspectral images of single-cells (Fig. 1). Using a dataset of hyperspectral images from three mouse cells lines, we demonstrate that the accuracy and robustness of the classification can increase when using an image pattern rather than an average spectrum representing the cell. The current study provides supporting evidence that our methodology can benefit the analysis of hyperspectral images in biological and biomedical studies
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.