ABSTRACT
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383
has been applied to many di—erent areas of studies such as lipidomics, drug delivery, medical imaging, and biomass energy production, etc. as will be discussed in later chapters (Wang et al., 2005, 2011; Bégin et al., 2009; Kim et al., 2010; Saar et al., 2010b; Pezacki et al., 2011). While some of these problems could also be studied with confocal Raman microscopy, its slow imaging speed is a major deterrent to its widespread use, especially for live cells or samples that change with time. However, confocal Raman microscopy has a distinctive advantage over current narrowband implementations of coherent Raman microscopy in that it provides complete and reliable Raman spectra. is is important for analyzing complex systems such as biological cells because in most cases, multiple components have major overlapping Raman contributions (Puppels et al., 1990; Feofanov et al., 2000; Anita, 2003; Ellis and Goodacre, 2006; Yamakoshi et al., 2011). In order to extract quantitative information of any particular component, at least acquisition at several Raman bands or a portion of the full spectrum, followed by multivariate data analysis is required (Enejder et al., 2005; Haka et al., 2005). One approach to overcome the con–icting requirement of imaging speed and information content is to combine Raman spectroscopy with a fast imaging technique (Caspers et al., 2003; Slipchenko et al., 2009). Spectroscopy acquisition is only performed at certain points of interest in the images. With very limited sampling, these techniques could not provide chemical maps at high imaging speeds.
