Analysis of human skin tissue by Raman microspectroscopy: Dealing with the background

• Published in: Vibrational spectroscopy 2012-07, Vol.61, p.124-132
• Main Authors: Bonnier, F., Ali, S.M., Knief, P., Lambkin, H., Flynn, K., McDonagh, V., Healy, C., Lee, T.C., Lyng, F.M., Byrne, H.J.
• Format: Article
• Language: English
• Subjects: Collection; Evolution; Fluorescence; Human tissue analysis; Lasers; Photodamage; Raman background; Raman spectra; Raman spectroscopy; Samples; Scattering; Spectra; Statistical analysis; Statistical methods
• ISSN: 0924-2031; 1873-3697
• DOI: 10.1016/j.vibspec.2012.03.009

Raman microspectroscopy is widely used for molecular characterisation of tissue samples. Nevertheless, when working in vitro on tissue sections, the presence of a broad background to the spectra remains problematic and its removal requires advanced methods for pre-processing of the data. To date, research efforts have been primarily devoted to development of techniques of statistical analysis to extract the relevant information contained in the spectra. However, few attempts have been made to understand the origin of the background and to improve the protocols used for the collection of Raman spectra that could lead to the reduction or elimination of the background. It has been demonstrated that measurement at 785nm in water immersion significantly reduces the Raman background of both pure biochemical components and tissue sections, associating the background at 785nm with a scattering phenomenon rather than fluorescence. It is however of interest to probe the dependence of the observed background and any time evolution normally associated with photobleaching of fluorophores, under dry and immersed conditions, on the source wavelength. Using 785nm or 660nm as source, extended exposure of dried skin tissue sections to the laser results in a time dependent reduction of the background present in the Raman spectra. When working in water immersion, the overall background as well as the evolution over time is greatly reduced and the background is seen to stabilise after ∼20s exposure. Using 532nm or 473nm as source for the examination of dried tissue sections, visible photodamage of the sample limits the laser power usable for the collection of spectra to 5mW. Immersion of the tissue sections protects against photodamage and laser powers of up to 30mW can be used without any visible damage. Under these conditions, the background is significantly reduced and good quality Raman spectra can be recorded. By adapting the protocol usually used for the collection of Raman spectra, this study clearly demonstrates that other approaches rather than mathematical manipulation of the data can be used to deal with the intrinsic background commonly observable. Notably, the dependence of the background and its time evolution under prolonged exposure on sample environment potentially sheds light on its origin as due to sample morphology (scattering) rather than chemical content (fluorescence). Overall, the study demonstrates that, in addition to reduced background, the photostab