19 September 2016 Principle component analysis and linear discriminant analysis of multi-spectral autofluorescence imaging data for differentiating basal cell carcinoma and healthy skin
Author Affiliations +
Abstract
In present paper, an ability to differentiate basal cell carcinoma (BCC) and healthy skin by combining multi-spectral autofluorescence imaging, principle component analysis (PCA), and linear discriminant analysis (LDA) has been demonstrated. For this purpose, the experimental setup, which includes excitation and detection branches, has been assembled. The excitation branch utilizes a mercury arc lamp equipped with a 365-nm narrow-linewidth excitation filter, a beam homogenizer, and a mechanical chopper. The detection branch employs a set of bandpass filters with the central wavelength of spectral transparency of λ = 400, 450, 500, and 550 nm, and a digital camera. The setup has been used to study three samples of freshly excised BCC. PCA and LDA have been implemented to analyze the data of multi-spectral fluorescence imaging. Observed results of this pilot study highlight the advantages of proposed imaging technique for skin cancer diagnosis.
Conference Presentation
© (2016) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Nikita V. Chernomyrdin, Kirill I. Zaytsev, Anastasiya D. Lesnichaya, Konstantin G. Kudrin, Olga P. Cherkasova, Vladimir N. Kurlov, Irina A. Shikunova, Alexei V. Perchik, Stanislav O. Yurchenko, Igor V. Reshetov, "Principle component analysis and linear discriminant analysis of multi-spectral autofluorescence imaging data for differentiating basal cell carcinoma and healthy skin", Proc. SPIE 9976, Imaging Spectrometry XXI, 99760B (19 September 2016); doi: 10.1117/12.2237607; https://doi.org/10.1117/12.2237607
PROCEEDINGS
9 PAGES + PRESENTATION

SHARE
Back to Top