Abstract
Canonical correlation analysis, a multivariate statistical technique, was used to investigate the degree of association between tissue optical properties and spatially resolved reflectance signals. Monte Carlo modeling was employed to simulate signals corresponding to different combinations of optical properties and these data sets were fed as input to statistical analysis. The results show that it is possible to adjust the separation and angular orientation of source and detector fibers such that the effect of a particular optical property will be augmented among coincident variations in other properties. The trends observed exhibit differences when compared with a conventional univariate sensitivity analysis in which only a single property is varied whereas all other parameters of interest are kept constant.
© 2010 Optical Society of America
Full Article | PDF ArticleOSA Recommended Articles
Johannes Swartling, Jan S. Dam, and Stefan Andersson-Engels
Appl. Opt. 42(22) 4612-4620 (2003)
Alwin Kienle, Florian K. Forster, and Raimund Hibst
Opt. Lett. 26(20) 1571-1573 (2001)
Thomas J. Farrell, Michael S. Patterson, and Matthias Essenpreis
Appl. Opt. 37(10) 1958-1972 (1998)