Raman vs. Fourier transform spectroscopy in diagnostic medicine.

To the Editor: We read the paper by Wang et al. (1) with great interest, and their findings add further support to existing evidence in the literature reporting that Fourier transform infrared (FTIR) spectroscopy is highly effective for the detection of malignant and premalignant tissues. The paper is an excellent piece of work, but we would like to make a few points. Certainly FTIR spectroscopy has a great future for use on tissue samples that contain little or no water, such as microscope sections or the partially dehydrated tissue samples used by the authors. However, hydrated or in vivo samples are a problem for FTIR spectroscopy because the water peak obscures much of the useful information in the biomolecular range. In contrast, Raman spectroscopy does not have this problem; thus, we believe that Raman has more potential, particularly in the area of in vivo diagnosis, which must be considered the Holy Grail in this field (2). This objective is entirely feasible because remote fiber optic probes are now available that allow collection of Raman spectra in vivo, for example, during a colposcopical examination (3). We agree with the authors that autofluorescence can be a problem with conventional Raman, but the use of Fourier transform Raman spectroscopy largely overcomes this (4). There are now enough results from studies using Raman spectroscopy to show that it has equal sensitivity to results obtained with FTIR spectroscopy (5). However, these are minor points, and the paper overall is outstanding. We hope that it receives the wide readership it deserves and so helps to establish FTIR and Raman spectroscopy as routine diagnostic tools in medicine.