The goal of this research is to correlate Raman spectra with tissue pathology using principal component analysis. Principal component analysis enables us to reduce our data set into a smaller number of mathematical lineshapes, called principal components, which represent the spectral variations in the data. A linear combination of these principal components can then be used to fit individual data. Finally, logisitic regression is used to correlate fitting coefficients with disease classifications and develop a diagnostic algorithm. So far, we have developed an initial decision algorithm and are in the process of testing it on data not included in the original data set.
Showing the most recent 10 out of 178 publications