Abstract

A novel classification method based on support vector machine (SVM) technique is investigated to discriminate the cancerous tissue from normal tissue with light induced autofluorescence signals. The autofluorescence spectra were measured in vivo from 85 nasopharyngeal carcinoma lesions and 131 normal tissue sites from 59 subjects during routine nasal endoscopy. It was found that the SVM based algorithms were able to achieve diagnostic accuracy with 94% sensitivity and over 96% specificity. In comparison with the previously developed algorithms based on principal component analysis, we found that the SVM algorithm produced better accuracy in diagnosing the cancerous tissue.

© 2003 Optical Society of America