RESUMO
BACKGROUND AND OBJECTIVE: The purpose of this study is to isolate pure biochemical compounds' eigenspectra and to classify skin cancer tumors. STUDY DESIGN/MATERIALS AND METHODS: Fourier transform infrared fiberoptic evanescent wave (FTIR-FEW) spectra, in the middle infrared (MIR) region, of human normal skin tissue and cancer tumors were analyzed using chemical factor analysis. RESULTS: Eigenspectra of biochemical species were isolated and some of the eigenspectra have been preliminarily identified as due to protein peptide bond and lipid carbonyl vibrations. Cluster analysis was used for classification and good agreement with prior pathological classifications, specifically for normal skin tissue and melanoma tumors, has been found. However the cluster analysis suggests substantial variability in basaloma tumor biochemical characteristics. In addition this study has demonstrated that chemical factor analysis can be carried out directly on raw data to extract biochemical component eigenspectra and classify skin states. Most importantly, it has been demonstrated that the combination of FTIR-FEW technique and chemical factor analysis has potential as a clinical diagnostic tool.
