Analysis and interpretation of infrared microscopic maps : visualization and classification of skin components by digital staining and multivariate analysis

Four analytical methods have been applied to infrared microscopic maps ofhuman skin which contained basal cell carcinoma tumors\u2014namely, point spectroscopy,grayscale functional group mapping, digital staining, and fuzzy C-means (FCM) clusteranalysis. Spectroscopic interpretation using point spectroscopy allowed discriminationbetween normal and tumor-bearing skin components. Functional group mapping pro-vided information on the distribution of lipids, proteins, nucleic acids, and collagenwithin skin sections. While functional group mapping only allowed examination of onematerial (or a ratio of two materials) at a time, a new approach termed digital stainingallowed visualization of the distribution of up to three materials at one time. Digitalstaining thus gave a more detailed understanding of the tissue section. Finally, appli-cation of FCM cluster analysis, a nonsubjective, unsupervised classi\ufb01cation methodol-ogy, allowed us to group spectra into \ufb01ve distinct clusters, which correlated to distincttissue components in skin. Tumor-bearing skin was clearly separated from normal skin.When different analytical methods are used in conjunction, a unique understanding oftissues can be obtained. Our results demonstrate that infrared microscopy has potentialfor the pathological diagnosis of skin tumors.Peer reviewed: YesNRC publication: Ye