Near-infrared autofluorescence spectroscopy for in vivo diagnosis of cervical precancer

The purpose of this study is to explore the feasibility of utilizing near-infrared (NIR) autofluorescence spectroscopy for in vivo diagnosis of precancer (i.e., dysplasia) in the cervix. A rapid NIR spectroscopy system in combination with a fiber-optic probe was developed for the in vivo NIR fluorescence measurements under the 785 nm laser excitation. Multivariate statistical techniques including principal component analysis (PCA) and linear discriminant analysis (LDA) were employed to develop the diagnostic algorithms for spectra classification. Classification result obtained from the PCA-LDA model based on tissue NIR autofluorescence data yielded a diagnostic sensitivity of 84.8% and specificity of 85.1% for discrimination of precancer from normal cervical sites. The results demonstrate that NIR autofluorescence technique has the capacity for the noninvasive, in vivo diagnosis of precancer in the cervix.