Optical biopsy of skin tumors

A real-time, non-invasive, image processing based optical biopsy method is investigated. Three most frequently occurring skin tumors, basal cell carcinoma (BCC), squamous cell carcinoma (SCC), and malignant melanoma (MM) are studied along with normal moles. Images are obtained using a charge-coupled device (CCD) camera with eight narrow band filters ranging from 450 nm to 800 nm at 50 nm interval. Principal components analysis (PCA) is used to extract the main color features. BCC, and SCC can be differentiated from the MM and normal moles using PCA. Texture information is also considered to further classify different tumor types. A spatial gray-level co-occurrence matrix is used to obtain tumor texture roughness. MM and normal moles were differentiated by it. BCC and SCC are further classified using Fisher´s linear discriminant analysis (LDA)