Fourier transform infrared imaging (FTIRI) is a powerful tool for analyzing biochemical changes in tumoral tissues. The head and neck region is characterized by a great variety of lesions, with different degrees of malignancy, which are often difficult to diagnose. Schneiderian papillomas are sinonasal benign neoplasms arising from the Schneiderian mucosa; they can evolve into malignant tumoral lesions (squamous cell carcinoma). In addition, they can sometimes be confused with the more common inflammatory polyps. Therefore, an early and definitive diagnosis of this pathology is mandatory. Progressing in our research on the study of oral cavity lesions, 15 sections consisting of inflammatory sinonasal polyps, benign Schneiderian papillomas, and sinonasal undifferentiated carcinomas were analyzed using FTIRI. To allow a rigorous description of these pathologies and to gain objective diagnosis, the epithelial layer and the adjacent connective tissue of each section were separately investigated by following a multivariate analysis approach. According to the nature of the lesion, interesting modifications were detected in the average spectra of the different tissue components, above all in the lipid and protein patterns. Specific band-area ratios acting as spectral markers of the different pathologies were also highlighted.